From 3f4a0b917ce72ef47e438d354c433eb645218e87 Mon Sep 17 00:00:00 2001 From: john stultz Date: Tue, 17 Oct 2006 00:09:32 -0700 Subject: [PATCH] i386 Time: Avoid PIT SMP lockups Avoid possible PIT livelock issues seen on SMP systems (and reported by Andi), by not allowing it as a clocksource on SMP boxes. However, since the PIT may no longer be present, we have to properly handle the cases where SMP systems have TSC skew and fall back from the TSC. Since the PIT isn't there, it would "fall back" to the TSC again. So this changes the jiffies rating to 1, and the TSC-bad rating value to 0. Thus you will get the following behavior priority on i386 systems: tsc [if present & stable] hpet [if present] cyclone [if present] acpi_pm [if present] pit [if UP] jiffies Rather then the current more complicated: tsc [if present & stable] hpet [if present] cyclone [if present] acpi_pm [if present] pit [if cpus < 4] tsc [if present & unstable] jiffies Signed-off-by: John Stultz Cc: Andi Kleen Cc: Ingo Molnar Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/i8253.c | 2 +- arch/i386/kernel/tsc.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/i8253.c b/arch/i386/kernel/i8253.c index 477b24daff5..9a0060b92e3 100644 --- a/arch/i386/kernel/i8253.c +++ b/arch/i386/kernel/i8253.c @@ -109,7 +109,7 @@ static struct clocksource clocksource_pit = { static int __init init_pit_clocksource(void) { - if (num_possible_cpus() > 4) /* PIT does not scale! */ + if (num_possible_cpus() > 1) /* PIT does not scale! */ return 0; clocksource_pit.mult = clocksource_hz2mult(CLOCK_TICK_RATE, 20); diff --git a/arch/i386/kernel/tsc.c b/arch/i386/kernel/tsc.c index b8fa0a8b2e4..fbc95828cd7 100644 --- a/arch/i386/kernel/tsc.c +++ b/arch/i386/kernel/tsc.c @@ -349,8 +349,8 @@ static int tsc_update_callback(void) int change = 0; /* check to see if we should switch to the safe clocksource: */ - if (clocksource_tsc.rating != 50 && check_tsc_unstable()) { - clocksource_tsc.rating = 50; + if (clocksource_tsc.rating != 0 && check_tsc_unstable()) { + clocksource_tsc.rating = 0; clocksource_reselect(); change = 1; } @@ -461,7 +461,7 @@ static int __init init_tsc_clocksource(void) clocksource_tsc.shift); /* lower the rating if we already know its unstable: */ if (check_tsc_unstable()) - clocksource_tsc.rating = 50; + clocksource_tsc.rating = 0; init_timer(&verify_tsc_freq_timer); verify_tsc_freq_timer.function = verify_tsc_freq; -- cgit v1.2.3-70-g09d2 From a460e745e8f9c75a0525ff94154a0629f9d3e05d Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 17 Oct 2006 00:10:03 -0700 Subject: [PATCH] genirq: clean up irq-flow-type naming Introduce desc->name and eliminate the handle_irq_name() hack. Add set_irq_chip_and_handler_name() to set the flow type and name at once. Signed-off-by: Ingo Molnar Acked-by: Thomas Gleixner Cc: "Eric W. Biederman" Cc: Matthew Wilcox Cc: Kyle McMartin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/i8259.c | 7 ++++--- arch/i386/kernel/io_apic.c | 17 ++++++++++------- arch/i386/kernel/irq.c | 2 +- arch/x86_64/kernel/i8259.c | 7 ++++--- arch/x86_64/kernel/io_apic.c | 15 ++++++++------- arch/x86_64/kernel/irq.c | 2 +- include/linux/irq.h | 22 ++++++++++------------ kernel/irq/chip.c | 33 +++++++++------------------------ 8 files changed, 47 insertions(+), 58 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/i8259.c b/arch/i386/kernel/i8259.c index d53eafb6daa..62996cd1708 100644 --- a/arch/i386/kernel/i8259.c +++ b/arch/i386/kernel/i8259.c @@ -113,7 +113,8 @@ void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); io_apic_irqs &= ~(1<name); - seq_printf(p, "-%s", handle_irq_name(irq_desc[i].handle_irq)); + seq_printf(p, "-%-8s", irq_desc[i].name); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c index 0612a33bb89..c4ef801b765 100644 --- a/arch/x86_64/kernel/i8259.c +++ b/arch/x86_64/kernel/i8259.c @@ -178,7 +178,8 @@ void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); io_apic_irqs &= ~(1<name); - seq_printf(p, "-%s", handle_irq_name(irq_desc[i].handle_irq)); + seq_printf(p, "-%-8s", irq_desc[i].name); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) diff --git a/include/linux/irq.h b/include/linux/irq.h index c64f3cc7e87..775f5a7da49 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -141,6 +141,7 @@ struct irq_chip { * @pending_mask: pending rebalanced interrupts * @dir: /proc/irq/ procfs entry * @affinity_entry: /proc/irq/smp_affinity procfs entry on SMP + * @name: flow handler name for /proc/interrupts output * * Pad this out to 32 bytes for cache and indexing reasons. */ @@ -165,8 +166,9 @@ struct irq_desc { cpumask_t pending_mask; #endif #ifdef CONFIG_PROC_FS - struct proc_dir_entry *dir; + struct proc_dir_entry *dir; #endif + const char *name; } ____cacheline_aligned; extern struct irq_desc irq_desc[NR_IRQS]; @@ -271,12 +273,6 @@ extern void fastcall handle_simple_irq(unsigned int irq, struct irq_desc *desc); extern void fastcall handle_percpu_irq(unsigned int irq, struct irq_desc *desc); extern void fastcall handle_bad_irq(unsigned int irq, struct irq_desc *desc); -/* - * Get a descriptive string for the highlevel handler, for - * /proc/interrupts output: - */ -extern const char *handle_irq_name(irq_flow_handler_t handle); - /* * Monolithic do_IRQ implementation. * (is an explicit fastcall, because i386 4KSTACKS calls it from assembly) @@ -326,10 +322,12 @@ extern struct irq_chip no_irq_chip; extern struct irq_chip dummy_irq_chip; extern void -set_irq_chip_and_handler(unsigned int irq, struct irq_chip *chip, - irq_flow_handler_t handle); +set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, + irq_flow_handler_t handle, const char *name); + extern void -__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained); +__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, + const char *name); /* * Set a highlevel flow handler for a given IRQ: @@ -337,7 +335,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained); static inline void set_irq_handler(unsigned int irq, irq_flow_handler_t handle) { - __set_irq_handler(irq, handle, 0); + __set_irq_handler(irq, handle, 0, NULL); } /* @@ -349,7 +347,7 @@ static inline void set_irq_chained_handler(unsigned int irq, irq_flow_handler_t handle) { - __set_irq_handler(irq, handle, 1); + __set_irq_handler(irq, handle, 1, NULL); } /* Handle dynamic irq creation and destruction */ diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index 11c99697acf..2d0dc3efe81 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -499,7 +499,8 @@ handle_percpu_irq(unsigned int irq, struct irq_desc *desc) #endif /* CONFIG_SMP */ void -__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained) +__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, + const char *name) { struct irq_desc *desc; unsigned long flags; @@ -540,6 +541,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained) desc->depth = 1; } desc->handle_irq = handle; + desc->name = name; if (handle != handle_bad_irq && is_chained) { desc->status &= ~IRQ_DISABLED; @@ -555,30 +557,13 @@ set_irq_chip_and_handler(unsigned int irq, struct irq_chip *chip, irq_flow_handler_t handle) { set_irq_chip(irq, chip); - __set_irq_handler(irq, handle, 0); + __set_irq_handler(irq, handle, 0, NULL); } -/* - * Get a descriptive string for the highlevel handler, for - * /proc/interrupts output: - */ -const char * -handle_irq_name(irq_flow_handler_t handle) +void +set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, + irq_flow_handler_t handle, const char *name) { - if (handle == handle_level_irq) - return "level "; - if (handle == handle_fasteoi_irq) - return "fasteoi"; - if (handle == handle_edge_irq) - return "edge "; - if (handle == handle_simple_irq) - return "simple "; -#ifdef CONFIG_SMP - if (handle == handle_percpu_irq) - return "percpu "; -#endif - if (handle == handle_bad_irq) - return "bad "; - - return NULL; + set_irq_chip(irq, chip); + __set_irq_handler(irq, handle, 0, name); } -- cgit v1.2.3-70-g09d2 From 3864c4894a7f4c03d69a90082a5bb0ab10e437ab Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 17 Oct 2006 00:10:26 -0700 Subject: [PATCH] lockdep: annotate i386 apm Lockdep doesn't like to enable interrupts when they are enabled already. BUG: warning at kernel/lockdep.c:1814/trace_hardirqs_on() (Not tainted) [] show_trace_log_lvl+0x58/0x16a [] show_trace+0xd/0x10 [] dump_stack+0x19/0x1b [] trace_hardirqs_on+0xa2/0x11e [] apm_bios_call_simple+0xcd/0xfd [] apm+0x92/0x5b1 [] kernel_thread_helper+0x5/0xb DWARF2 unwinder stuck at kernel_thread_helper+0x5/0xb Leftover inexact backtrace: [] show_trace+0xd/0x10 [] dump_stack+0x19/0x1b [] trace_hardirqs_on+0xa2/0x11e [] apm_bios_call_simple+0xcd/0xfd [] apm+0x92/0x5b1 [] kernel_thread_helper+0x5/0xb Signed-off-by: Peter Zijlstra Cc: Stephen Rothwell Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/apm.c | 37 +++++++++++++++++++++++++++---------- 1 file changed, 27 insertions(+), 10 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c index b42f2d914af..2af65858d32 100644 --- a/arch/i386/kernel/apm.c +++ b/arch/i386/kernel/apm.c @@ -540,11 +540,30 @@ static inline void apm_restore_cpus(cpumask_t mask) * Also, we KNOW that for the non error case of apm_bios_call, there * is no useful data returned in the low order 8 bits of eax. */ -#define APM_DO_CLI \ - if (apm_info.allow_ints) \ - local_irq_enable(); \ - else \ + +static inline unsigned long __apm_irq_save(void) +{ + unsigned long flags; + local_save_flags(flags); + if (apm_info.allow_ints) { + if (irqs_disabled_flags(flags)) + local_irq_enable(); + } else + local_irq_disable(); + + return flags; +} + +#define apm_irq_save(flags) \ + do { flags = __apm_irq_save(); } while (0) + +static inline void apm_irq_restore(unsigned long flags) +{ + if (irqs_disabled_flags(flags)) local_irq_disable(); + else if (irqs_disabled()) + local_irq_enable(); +} #ifdef APM_ZERO_SEGS # define APM_DECL_SEGS \ @@ -596,12 +615,11 @@ static u8 apm_bios_call(u32 func, u32 ebx_in, u32 ecx_in, save_desc_40 = gdt[0x40 / 8]; gdt[0x40 / 8] = bad_bios_desc; - local_save_flags(flags); - APM_DO_CLI; + apm_irq_save(flags); APM_DO_SAVE_SEGS; apm_bios_call_asm(func, ebx_in, ecx_in, eax, ebx, ecx, edx, esi); APM_DO_RESTORE_SEGS; - local_irq_restore(flags); + apm_irq_restore(flags); gdt[0x40 / 8] = save_desc_40; put_cpu(); apm_restore_cpus(cpus); @@ -640,12 +658,11 @@ static u8 apm_bios_call_simple(u32 func, u32 ebx_in, u32 ecx_in, u32 *eax) save_desc_40 = gdt[0x40 / 8]; gdt[0x40 / 8] = bad_bios_desc; - local_save_flags(flags); - APM_DO_CLI; + apm_irq_save(flags); APM_DO_SAVE_SEGS; error = apm_bios_call_simple_asm(func, ebx_in, ecx_in, eax); APM_DO_RESTORE_SEGS; - local_irq_restore(flags); + apm_irq_restore(flags); gdt[0x40 / 8] = save_desc_40; put_cpu(); apm_restore_cpus(cpus); -- cgit v1.2.3-70-g09d2 From b5e4efe7e061ff52ac97b9fa45acca529d8daeea Mon Sep 17 00:00:00 2001 From: "eiichiro.oiwa.nm@hitachi.com" Date: Thu, 28 Sep 2006 13:55:47 +0900 Subject: PCI: Turn pci_fixup_video into generic for embedded VGA pci_fixup_video turns into generic code because there are many platforms need this fixup for embedded VGA as well as x86. The Video BIOS integrates into System BIOS on a machine has embedded VGA although embedded VGA generally don't have PCI ROM. As a result, embedded VGA need the way that the sysfs rom points to the Video BIOS of System RAM (0xC0000). PCI-to-PCI Bridge Architecture specification describes the condition whether or not PCI ROM forwards VGA compatible memory address. fixup_video suits this specification. Although the Video ROM generally implements in x86 code regardless of platform, some application such as X Window System can run this code by dosemu86. Therefore, pci_fixup_video should turn into generic code. Signed-off-by: Eiichiro Oiwa Acked-by: Alan Cox Acked-by: Jesse Barnes Signed-off-by: Greg Kroah-Hartman --- arch/i386/pci/fixup.c | 45 --------------------------------------------- drivers/pci/quirks.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ drivers/pci/rom.c | 5 ++++- 3 files changed, 49 insertions(+), 46 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/pci/fixup.c b/arch/i386/pci/fixup.c index b60d7e8689e..908b410f4c9 100644 --- a/arch/i386/pci/fixup.c +++ b/arch/i386/pci/fixup.c @@ -342,51 +342,6 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PB1, pcie_r DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PC, pcie_rootport_aspm_quirk ); DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PC1, pcie_rootport_aspm_quirk ); -/* - * Fixup to mark boot BIOS video selected by BIOS before it changes - * - * From information provided by "Jon Smirl" - * - * The standard boot ROM sequence for an x86 machine uses the BIOS - * to select an initial video card for boot display. This boot video - * card will have it's BIOS copied to C0000 in system RAM. - * IORESOURCE_ROM_SHADOW is used to associate the boot video - * card with this copy. On laptops this copy has to be used since - * the main ROM may be compressed or combined with another image. - * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW - * is marked here since the boot video device will be the only enabled - * video device at this point. - */ - -static void __devinit pci_fixup_video(struct pci_dev *pdev) -{ - struct pci_dev *bridge; - struct pci_bus *bus; - u16 config; - - if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) - return; - - /* Is VGA routed to us? */ - bus = pdev->bus; - while (bus) { - bridge = bus->self; - if (bridge) { - pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, - &config); - if (!(config & PCI_BRIDGE_CTL_VGA)) - return; - } - bus = bus->parent; - } - pci_read_config_word(pdev, PCI_COMMAND, &config); - if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { - pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; - printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_video); - /* * Some Toshiba laptops need extra code to enable their TI TSB43AB22/A. * diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 1d2ccda946f..371ab8821f1 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1619,6 +1619,51 @@ static void __devinit fixup_rev1_53c810(struct pci_dev* dev) } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NCR, PCI_DEVICE_ID_NCR_53C810, fixup_rev1_53c810); +/* + * Fixup to mark boot BIOS video selected by BIOS before it changes + * + * From information provided by "Jon Smirl" + * + * The standard boot ROM sequence for an x86 machine uses the BIOS + * to select an initial video card for boot display. This boot video + * card will have it's BIOS copied to C0000 in system RAM. + * IORESOURCE_ROM_SHADOW is used to associate the boot video + * card with this copy. On laptops this copy has to be used since + * the main ROM may be compressed or combined with another image. + * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW + * is marked here since the boot video device will be the only enabled + * video device at this point. + */ + +static void __devinit fixup_video(struct pci_dev *pdev) +{ + struct pci_dev *bridge; + struct pci_bus *bus; + u16 config; + + if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) + return; + + /* Is VGA routed to us? */ + bus = pdev->bus; + while (bus) { + bridge = bus->self; + if (bridge) { + pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, + &config); + if (!(config & PCI_BRIDGE_CTL_VGA)) + return; + } + bus = bus->parent; + } + pci_read_config_word(pdev, PCI_COMMAND, &config); + if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { + pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; + printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, fixup_video); + static void pci_do_fixups(struct pci_dev *dev, struct pci_fixup *f, struct pci_fixup *end) { diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index f5ee7ce16fa..43e4a49f2cc 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -71,7 +71,10 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) void __iomem *image; int last_image; - /* IORESOURCE_ROM_SHADOW only set on x86 */ + /* + * IORESOURCE_ROM_SHADOW set if the VGA enable bit of the Bridge Control + * register is set for embedded VGA. + */ if (res->flags & IORESOURCE_ROM_SHADOW) { /* primary video rom always starts here */ start = (loff_t)0xC0000; -- cgit v1.2.3-70-g09d2 From 6b4b78fed47e7380dfe9280b154e8b9bfcd4c86c Mon Sep 17 00:00:00 2001 From: Matt Domsch Date: Fri, 29 Sep 2006 15:23:23 -0500 Subject: PCI: optionally sort device lists breadth-first Problem: New Dell PowerEdge servers have 2 embedded ethernet ports, which are labeled NIC1 and NIC2 on the chassis, in the BIOS setup screens, and in the printed documentation. Assuming no other add-in ethernet ports in the system, Linux 2.4 kernels name these eth0 and eth1 respectively. Many people have come to expect this naming. Linux 2.6 kernels name these eth1 and eth0 respectively (backwards from expectations). I also have reports that various Sun and HP servers have similar behavior. Root cause: Linux 2.4 kernels walk the pci_devices list, which happens to be sorted in breadth-first order (or pcbios_find_device order on i386, which most often is breadth-first also). 2.6 kernels have both the pci_devices list and the pci_bus_type.klist_devices list, the latter is what is walked at driver load time to match the pci_id tables; this klist happens to be in depth-first order. On systems where, for physical routing reasons, NIC1 appears on a lower bus number than NIC2, but NIC2's bridge is discovered first in the depth-first ordering, NIC2 will be discovered before NIC1. If the list were sorted breadth-first, NIC1 would be discovered before NIC2. A PowerEdge 1955 system has the following topology which easily exhibits the difference between depth-first and breadth-first device lists. -[0000:00]-+-00.0 Intel Corporation 5000P Chipset Memory Controller Hub +-02.0-[0000:03-08]--+-00.0-[0000:04-07]--+-00.0-[0000:05-06]----00.0-[0000:06]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC2, 2.4 kernel name eth1, 2.6 kernel name eth0) +-1c.0-[0000:01-02]----00.0-[0000:02]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC1, 2.4 kernel name eth0, 2.6 kernel name eth1) Other factors, such as device driver load order and the presence of PCI slots at various points in the bus hierarchy further complicate this problem; I'm not trying to solve those here, just restore the device order, and thus basic behavior, that 2.4 kernels had. Solution: The solution can come in multiple steps. Suggested fix #1: kernel Patch below optionally sorts the two device lists into breadth-first ordering to maintain compatibility with 2.4 kernels. It adds two new command line options: pci=bfsort pci=nobfsort to force the sort order, or not, as you wish. It also adds DMI checks for the specific Dell systems which exhibit "backwards" ordering, to make them "right". Suggested fix #2: udev rules from userland Many people also have the expectation that embedded NICs are always discovered before add-in NICs (which this patch does not try to do). Using the PCI IRQ Routing Table provided by system BIOS, it's easy to determine which PCI devices are embedded, or if add-in, which PCI slot they're in. I'm working on a tool that would allow udev to name ethernet devices in ascending embedded, slot 1 .. slot N order, subsort by PCI bus/dev/fn breadth-first. It'll be possible to use it independent of udev as well for those distributions that don't use udev in their installers. Suggested fix #3: system board routing rules One can constrain the system board layout to put NIC1 ahead of NIC2 regardless of breadth-first or depth-first discovery order. This adds a significant level of complexity to board routing, and may not be possible in all instances (witness the above systems from several major manufacturers). I don't want to encourage this particular train of thought too far, at the expense of not doing #1 or #2 above. Feedback appreciated. Patch tested on a Dell PowerEdge 1955 blade with 2.6.18. You'll also note I took some liberty and temporarily break the klist abstraction to simplify and speed up the sort algorithm. I think that's both safe and appropriate in this instance. Signed-off-by: Matt Domsch Signed-off-by: Greg Kroah-Hartman --- Documentation/kernel-parameters.txt | 5 ++ arch/i386/pci/common.c | 59 +++++++++++++++++++++++- arch/i386/pci/pci.h | 7 +++ drivers/pci/probe.c | 92 +++++++++++++++++++++++++++++++++++++ include/linux/pci.h | 1 + 5 files changed, 162 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index ff571f9298e..dd00fd556a6 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1231,6 +1231,11 @@ and is between 256 and 4096 characters. It is defined in the file machine check when some devices' config space is read. But various workarounds are disabled and some IOMMU drivers will not work. + bfsort Sort PCI devices into breadth-first order. + This sorting is done to get a device + order compatible with older (<= 2.4) kernels. + nobfsort Don't sort PCI devices into breadth-first order. + pcmv= [HW,PCMCIA] BadgePAD 4 pd. [PARIDE] diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c index 68bce194e68..6d5ace845e4 100644 --- a/arch/i386/pci/common.c +++ b/arch/i386/pci/common.c @@ -20,6 +20,7 @@ unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 | PCI_PROBE_MMCONF; +int pci_bf_sort; int pci_routeirq; int pcibios_last_bus = -1; unsigned long pirq_table_addr; @@ -117,6 +118,20 @@ void __devinit pcibios_fixup_bus(struct pci_bus *b) pci_read_bridge_bases(b); } +/* + * Only use DMI information to set this if nothing was passed + * on the kernel command line (which was parsed earlier). + */ + +static int __devinit set_bf_sort(struct dmi_system_id *d) +{ + if (pci_bf_sort == pci_bf_sort_default) { + pci_bf_sort = pci_dmi_bf; + printk(KERN_INFO "PCI: %s detected, enabling pci=bfsort.\n", d->ident); + } + return 0; +} + /* * Enable renumbering of PCI bus# ranges to reach all PCI busses (Cardbus) */ @@ -130,11 +145,11 @@ static int __devinit assign_all_busses(struct dmi_system_id *d) } #endif +static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { +#ifdef __i386__ /* * Laptops which need pci=assign-busses to see Cardbus cards */ -static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { -#ifdef __i386__ { .callback = assign_all_busses, .ident = "Samsung X20 Laptop", @@ -144,6 +159,38 @@ static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { }, }, #endif /* __i386__ */ + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 1950", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1950"), + }, + }, + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 1955", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1955"), + }, + }, + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 2900", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2900"), + }, + }, + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 2950", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2950"), + }, + }, {} }; @@ -189,6 +236,8 @@ static int __init pcibios_init(void) pcibios_resource_survey(); + if (pci_bf_sort >= pci_force_bf) + pci_sort_breadthfirst(); #ifdef CONFIG_PCI_BIOS if ((pci_probe & PCI_BIOS_SORT) && !(pci_probe & PCI_NO_SORT)) pcibios_sort(); @@ -203,6 +252,12 @@ char * __devinit pcibios_setup(char *str) if (!strcmp(str, "off")) { pci_probe = 0; return NULL; + } else if (!strcmp(str, "bfsort")) { + pci_bf_sort = pci_force_bf; + return NULL; + } else if (!strcmp(str, "nobfsort")) { + pci_bf_sort = pci_force_nobf; + return NULL; } #ifdef CONFIG_PCI_BIOS else if (!strcmp(str, "bios")) { diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h index 1814f74569c..ad065cebd7b 100644 --- a/arch/i386/pci/pci.h +++ b/arch/i386/pci/pci.h @@ -30,6 +30,13 @@ extern unsigned int pci_probe; extern unsigned long pirq_table_addr; +enum pci_bf_sort_state { + pci_bf_sort_default, + pci_force_nobf, + pci_force_bf, + pci_dmi_bf, +}; + /* pci-i386.c */ extern unsigned int pcibios_max_latency; diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index a3b0a5eb505..e159d660449 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -1067,3 +1067,95 @@ EXPORT_SYMBOL(pci_scan_bridge); EXPORT_SYMBOL(pci_scan_single_device); EXPORT_SYMBOL_GPL(pci_scan_child_bus); #endif + +static int __init pci_sort_bf_cmp(const struct pci_dev *a, const struct pci_dev *b) +{ + if (pci_domain_nr(a->bus) < pci_domain_nr(b->bus)) return -1; + else if (pci_domain_nr(a->bus) > pci_domain_nr(b->bus)) return 1; + + if (a->bus->number < b->bus->number) return -1; + else if (a->bus->number > b->bus->number) return 1; + + if (a->devfn < b->devfn) return -1; + else if (a->devfn > b->devfn) return 1; + + return 0; +} + +/* + * Yes, this forcably breaks the klist abstraction temporarily. It + * just wants to sort the klist, not change reference counts and + * take/drop locks rapidly in the process. It does all this while + * holding the lock for the list, so objects can't otherwise be + * added/removed while we're swizzling. + */ +static void __init pci_insertion_sort_klist(struct pci_dev *a, struct list_head *list) +{ + struct list_head *pos; + struct klist_node *n; + struct device *dev; + struct pci_dev *b; + + list_for_each(pos, list) { + n = container_of(pos, struct klist_node, n_node); + dev = container_of(n, struct device, knode_bus); + b = to_pci_dev(dev); + if (pci_sort_bf_cmp(a, b) <= 0) { + list_move_tail(&a->dev.knode_bus.n_node, &b->dev.knode_bus.n_node); + return; + } + } + list_move_tail(&a->dev.knode_bus.n_node, list); +} + +static void __init pci_sort_breadthfirst_klist(void) +{ + LIST_HEAD(sorted_devices); + struct list_head *pos, *tmp; + struct klist_node *n; + struct device *dev; + struct pci_dev *pdev; + + spin_lock(&pci_bus_type.klist_devices.k_lock); + list_for_each_safe(pos, tmp, &pci_bus_type.klist_devices.k_list) { + n = container_of(pos, struct klist_node, n_node); + dev = container_of(n, struct device, knode_bus); + pdev = to_pci_dev(dev); + pci_insertion_sort_klist(pdev, &sorted_devices); + } + list_splice(&sorted_devices, &pci_bus_type.klist_devices.k_list); + spin_unlock(&pci_bus_type.klist_devices.k_lock); +} + +static void __init pci_insertion_sort_devices(struct pci_dev *a, struct list_head *list) +{ + struct pci_dev *b; + + list_for_each_entry(b, list, global_list) { + if (pci_sort_bf_cmp(a, b) <= 0) { + list_move_tail(&a->global_list, &b->global_list); + return; + } + } + list_move_tail(&a->global_list, list); +} + +static void __init pci_sort_breadthfirst_devices(void) +{ + LIST_HEAD(sorted_devices); + struct pci_dev *dev, *tmp; + + down_write(&pci_bus_sem); + list_for_each_entry_safe(dev, tmp, &pci_devices, global_list) { + pci_insertion_sort_devices(dev, &sorted_devices); + } + list_splice(&sorted_devices, &pci_devices); + up_write(&pci_bus_sem); +} + +void __init pci_sort_breadthfirst(void) +{ + pci_sort_breadthfirst_devices(); + pci_sort_breadthfirst_klist(); +} + diff --git a/include/linux/pci.h b/include/linux/pci.h index 09bf88fc80c..4689e2a699c 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -443,6 +443,7 @@ extern void pci_remove_bus(struct pci_bus *b); extern void pci_remove_bus_device(struct pci_dev *dev); extern void pci_stop_bus_device(struct pci_dev *dev); void pci_setup_cardbus(struct pci_bus *bus); +extern void pci_sort_breadthfirst(void); /* Generic PCI functions exported to card drivers */ -- cgit v1.2.3-70-g09d2 From 3fcfab16c5b86eaa3db3a9a31adba550c5b67141 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 19 Oct 2006 23:28:16 -0700 Subject: [PATCH] separate bdi congestion functions from queue congestion functions Separate out the concept of "queue congestion" from "backing-dev congestion". Congestion is a backing-dev concept, not a queue concept. The blk_* congestion functions are retained, as wrappers around the core backing-dev congestion functions. This proper layering is needed so that NFS can cleanly use the congestion functions, and so that CONFIG_BLOCK=n actually links. Cc: "Thomas Maier" Cc: "Jens Axboe" Cc: Trond Myklebust Cc: David Howells Cc: Peter Osterlund Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/lib/usercopy.c | 3 +- block/ll_rw_blk.c | 71 --------------------------------------------- drivers/md/dm-crypt.c | 3 +- fs/fat/file.c | 3 +- fs/nfs/write.c | 4 ++- fs/reiserfs/journal.c | 3 +- fs/xfs/linux-2.6/kmem.c | 5 ++-- fs/xfs/linux-2.6/xfs_buf.c | 3 +- include/linux/backing-dev.h | 7 +++++ include/linux/blkdev.h | 24 ++++++++++++--- include/linux/writeback.h | 1 - mm/Makefile | 3 +- mm/backing-dev.c | 69 +++++++++++++++++++++++++++++++++++++++++++ mm/page-writeback.c | 17 +++-------- mm/page_alloc.c | 5 ++-- mm/shmem.c | 3 +- mm/vmscan.c | 6 ++-- 17 files changed, 126 insertions(+), 104 deletions(-) create mode 100644 mm/backing-dev.c (limited to 'arch/i386') diff --git a/arch/i386/lib/usercopy.c b/arch/i386/lib/usercopy.c index 258df6b4d7d..d22cfc9d656 100644 --- a/arch/i386/lib/usercopy.c +++ b/arch/i386/lib/usercopy.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include @@ -741,7 +742,7 @@ survive: if (retval == -ENOMEM && is_init(current)) { up_read(¤t->mm->mmap_sem); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto survive; } diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 132a858ce2c..136066583c6 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -56,11 +56,6 @@ static kmem_cache_t *requestq_cachep; */ static kmem_cache_t *iocontext_cachep; -static wait_queue_head_t congestion_wqh[2] = { - __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]), - __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1]) - }; - /* * Controlling structure to kblockd */ @@ -112,37 +107,6 @@ static void blk_queue_congestion_threshold(struct request_queue *q) q->nr_congestion_off = nr; } -/* - * A queue has just exitted congestion. Note this in the global counter of - * congested queues, and wake up anyone who was waiting for requests to be - * put back. - */ -void blk_clear_queue_congested(request_queue_t *q, int rw) -{ - enum bdi_state bit; - wait_queue_head_t *wqh = &congestion_wqh[rw]; - - bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; - clear_bit(bit, &q->backing_dev_info.state); - smp_mb__after_clear_bit(); - if (waitqueue_active(wqh)) - wake_up(wqh); -} -EXPORT_SYMBOL(blk_clear_queue_congested); - -/* - * A queue has just entered congestion. Flag that in the queue's VM-visible - * state flags and increment the global gounter of congested queues. - */ -void blk_set_queue_congested(request_queue_t *q, int rw) -{ - enum bdi_state bit; - - bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; - set_bit(bit, &q->backing_dev_info.state); -} -EXPORT_SYMBOL(blk_set_queue_congested); - /** * blk_get_backing_dev_info - get the address of a queue's backing_dev_info * @bdev: device @@ -2755,41 +2719,6 @@ void blk_end_sync_rq(struct request *rq, int error) } EXPORT_SYMBOL(blk_end_sync_rq); -/** - * blk_congestion_wait - wait for a queue to become uncongested - * @rw: READ or WRITE - * @timeout: timeout in jiffies - * - * Waits for up to @timeout jiffies for a queue (any queue) to exit congestion. - * If no queues are congested then just wait for the next request to be - * returned. - */ -long blk_congestion_wait(int rw, long timeout) -{ - long ret; - DEFINE_WAIT(wait); - wait_queue_head_t *wqh = &congestion_wqh[rw]; - - prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); - ret = io_schedule_timeout(timeout); - finish_wait(wqh, &wait); - return ret; -} - -EXPORT_SYMBOL(blk_congestion_wait); - -/** - * blk_congestion_end - wake up sleepers on a congestion queue - * @rw: READ or WRITE - */ -void blk_congestion_end(int rw) -{ - wait_queue_head_t *wqh = &congestion_wqh[rw]; - - if (waitqueue_active(wqh)) - wake_up(wqh); -} - /* * Has to be called with the request spinlock acquired */ diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 655d816760e..a625576fdee 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -602,7 +603,7 @@ static void process_write(struct crypt_io *io) /* out of memory -> run queues */ if (remaining) - blk_congestion_wait(bio_data_dir(clone), HZ/100); + congestion_wait(bio_data_dir(clone), HZ/100); } } diff --git a/fs/fat/file.c b/fs/fat/file.c index f4b8f8b3fbd..8337451e789 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -13,6 +13,7 @@ #include #include #include +#include #include int fat_generic_ioctl(struct inode *inode, struct file *filp, @@ -118,7 +119,7 @@ static int fat_file_release(struct inode *inode, struct file *filp) if ((filp->f_mode & FMODE_WRITE) && MSDOS_SB(inode->i_sb)->options.flush) { fat_flush_inodes(inode->i_sb, inode, NULL); - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } return 0; } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f6675d2c386..ca92ac36fe9 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -57,6 +57,8 @@ #include #include #include +#include + #include #include @@ -395,7 +397,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) out: clear_bit(BDI_write_congested, &bdi->state); wake_up_all(&nfs_write_congestion); - writeback_congestion_end(); + congestion_end(WRITE); return err; } diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index ad8cbc49883..85ce2326830 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -53,6 +53,7 @@ #include #include #include +#include /* gets a struct reiserfs_journal_list * from a list head */ #define JOURNAL_LIST_ENTRY(h) (list_entry((h), struct reiserfs_journal_list, \ @@ -970,7 +971,7 @@ int reiserfs_async_progress_wait(struct super_block *s) DEFINE_WAIT(wait); struct reiserfs_journal *j = SB_JOURNAL(s); if (atomic_read(&j->j_async_throttle)) - blk_congestion_wait(WRITE, HZ / 10); + congestion_wait(WRITE, HZ / 10); return 0; } diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c index d5973758981..004baf60061 100644 --- a/fs/xfs/linux-2.6/kmem.c +++ b/fs/xfs/linux-2.6/kmem.c @@ -21,6 +21,7 @@ #include #include #include +#include #include "time.h" #include "kmem.h" @@ -53,7 +54,7 @@ kmem_alloc(size_t size, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __FUNCTION__, lflags); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } while (1); } @@ -131,7 +132,7 @@ kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __FUNCTION__, lflags); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } while (1); } diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 9bbadafdcb0..db5f5a3608c 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "xfs_linux.h" STATIC kmem_zone_t *xfs_buf_zone; @@ -395,7 +396,7 @@ _xfs_buf_lookup_pages( XFS_STATS_INC(xb_page_retries); xfsbufd_wakeup(0, gfp_mask); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto retry; } diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index f7a1390d67f..7011d625559 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -10,6 +10,8 @@ #include +struct page; + /* * Bits in backing_dev_info.state */ @@ -88,6 +90,11 @@ static inline int bdi_rw_congested(struct backing_dev_info *bdi) (1 << BDI_write_congested)); } +void clear_bdi_congested(struct backing_dev_info *bdi, int rw); +void set_bdi_congested(struct backing_dev_info *bdi, int rw); +long congestion_wait(int rw, long timeout); +void congestion_end(int rw); + #define bdi_cap_writeback_dirty(bdi) \ (!((bdi)->capabilities & BDI_CAP_NO_WRITEBACK)) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 9575e3a5ff2..7bfcde2d557 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -651,8 +651,26 @@ extern void blk_recount_segments(request_queue_t *, struct bio *); extern int scsi_cmd_ioctl(struct file *, struct gendisk *, unsigned int, void __user *); extern int sg_scsi_ioctl(struct file *, struct request_queue *, struct gendisk *, struct scsi_ioctl_command __user *); -extern void blk_clear_queue_congested(request_queue_t *q, int rw); -extern void blk_set_queue_congested(request_queue_t *q, int rw); + +/* + * A queue has just exitted congestion. Note this in the global counter of + * congested queues, and wake up anyone who was waiting for requests to be + * put back. + */ +static inline void blk_clear_queue_congested(request_queue_t *q, int rw) +{ + clear_bdi_congested(&q->backing_dev_info, rw); +} + +/* + * A queue has just entered congestion. Flag that in the queue's VM-visible + * state flags and increment the global gounter of congested queues. + */ +static inline void blk_set_queue_congested(request_queue_t *q, int rw) +{ + set_bdi_congested(&q->backing_dev_info, rw); +} + extern void blk_start_queue(request_queue_t *q); extern void blk_stop_queue(request_queue_t *q); extern void blk_sync_queue(struct request_queue *q); @@ -767,10 +785,8 @@ extern int blk_queue_init_tags(request_queue_t *, int, struct blk_queue_tag *); extern void blk_queue_free_tags(request_queue_t *); extern int blk_queue_resize_tags(request_queue_t *, int); extern void blk_queue_invalidate_tags(request_queue_t *); -extern long blk_congestion_wait(int rw, long timeout); extern struct blk_queue_tag *blk_init_tags(int); extern void blk_free_tags(struct blk_queue_tag *); -extern void blk_congestion_end(int rw); static inline struct request *blk_map_queue_find_tag(struct blk_queue_tag *bqt, int tag) diff --git a/include/linux/writeback.h b/include/linux/writeback.h index a341c803286..fc35e6bdfb9 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -85,7 +85,6 @@ int wakeup_pdflush(long nr_pages); void laptop_io_completion(void); void laptop_sync_completion(void); void throttle_vm_writeout(void); -void writeback_congestion_end(void); /* These are exported to sysctl. */ extern int dirty_background_ratio; diff --git a/mm/Makefile b/mm/Makefile index 12b3a4eee88..f3c077eb0b8 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -10,7 +10,8 @@ mmu-$(CONFIG_MMU) := fremap.o highmem.o madvise.o memory.o mincore.o \ obj-y := bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \ page_alloc.o page-writeback.o pdflush.o \ readahead.o swap.o truncate.o vmscan.o \ - prio_tree.o util.o mmzone.o vmstat.o $(mmu-y) + prio_tree.o util.o mmzone.o vmstat.o backing-dev.o \ + $(mmu-y) ifeq ($(CONFIG_MMU)$(CONFIG_BLOCK),yy) obj-y += bounce.o diff --git a/mm/backing-dev.c b/mm/backing-dev.c new file mode 100644 index 00000000000..f50a2811f9d --- /dev/null +++ b/mm/backing-dev.c @@ -0,0 +1,69 @@ + +#include +#include +#include +#include +#include + +static wait_queue_head_t congestion_wqh[2] = { + __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]), + __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1]) + }; + + +void clear_bdi_congested(struct backing_dev_info *bdi, int rw) +{ + enum bdi_state bit; + wait_queue_head_t *wqh = &congestion_wqh[rw]; + + bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; + clear_bit(bit, &bdi->state); + smp_mb__after_clear_bit(); + if (waitqueue_active(wqh)) + wake_up(wqh); +} +EXPORT_SYMBOL(clear_bdi_congested); + +void set_bdi_congested(struct backing_dev_info *bdi, int rw) +{ + enum bdi_state bit; + + bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; + set_bit(bit, &bdi->state); +} +EXPORT_SYMBOL(set_bdi_congested); + +/** + * congestion_wait - wait for a backing_dev to become uncongested + * @rw: READ or WRITE + * @timeout: timeout in jiffies + * + * Waits for up to @timeout jiffies for a backing_dev (any backing_dev) to exit + * write congestion. If no backing_devs are congested then just wait for the + * next write to be completed. + */ +long congestion_wait(int rw, long timeout) +{ + long ret; + DEFINE_WAIT(wait); + wait_queue_head_t *wqh = &congestion_wqh[rw]; + + prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); + ret = io_schedule_timeout(timeout); + finish_wait(wqh, &wait); + return ret; +} +EXPORT_SYMBOL(congestion_wait); + +/** + * congestion_end - wake up sleepers on a congested backing_dev_info + * @rw: READ or WRITE + */ +void congestion_end(int rw) +{ + wait_queue_head_t *wqh = &congestion_wqh[rw]; + + if (waitqueue_active(wqh)) + wake_up(wqh); +} +EXPORT_SYMBOL(congestion_end); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index a0f33905744..8d9b19f239c 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -222,7 +222,7 @@ static void balance_dirty_pages(struct address_space *mapping) if (pages_written >= write_chunk) break; /* We've done our duty */ } - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } if (nr_reclaimable + global_page_state(NR_WRITEBACK) @@ -314,7 +314,7 @@ void throttle_vm_writeout(void) if (global_page_state(NR_UNSTABLE_NFS) + global_page_state(NR_WRITEBACK) <= dirty_thresh) break; - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } } @@ -351,7 +351,7 @@ static void background_writeout(unsigned long _min_pages) min_pages -= MAX_WRITEBACK_PAGES - wbc.nr_to_write; if (wbc.nr_to_write > 0 || wbc.pages_skipped > 0) { /* Wrote less than expected */ - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); if (!wbc.encountered_congestion) break; } @@ -422,7 +422,7 @@ static void wb_kupdate(unsigned long arg) writeback_inodes(&wbc); if (wbc.nr_to_write > 0) { if (wbc.encountered_congestion) - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); else break; /* All the old data is written */ } @@ -955,15 +955,6 @@ int test_set_page_writeback(struct page *page) } EXPORT_SYMBOL(test_set_page_writeback); -/* - * Wakes up tasks that are being throttled due to writeback congestion - */ -void writeback_congestion_end(void) -{ - blk_congestion_end(WRITE); -} -EXPORT_SYMBOL(writeback_congestion_end); - /* * Return true if any of the pages in the mapping are marged with the * passed tag. diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 40db96a655d..afee38f04d8 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include @@ -1050,7 +1051,7 @@ nofail_alloc: if (page) goto got_pg; if (gfp_mask & __GFP_NOFAIL) { - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto nofail_alloc; } } @@ -1113,7 +1114,7 @@ rebalance: do_retry = 1; } if (do_retry) { - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto rebalance; } diff --git a/mm/shmem.c b/mm/shmem.c index b378f66cf2f..4959535fc14 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -48,6 +48,7 @@ #include #include #include +#include #include #include @@ -1131,7 +1132,7 @@ repeat: page_cache_release(swappage); if (error == -ENOMEM) { /* let kswapd refresh zone for GFP_ATOMICs */ - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } goto repeat; } diff --git a/mm/vmscan.c b/mm/vmscan.c index af73c14f9d8..f05527bf792 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1059,7 +1059,7 @@ unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask) /* Take a nap, wait for some writeback to complete */ if (sc.nr_scanned && priority < DEF_PRIORITY - 2) - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } /* top priority shrink_caches still had more to do? don't OOM, then */ if (!sc.all_unreclaimable) @@ -1214,7 +1214,7 @@ scan: * another pass across the zones. */ if (total_scanned && priority < DEF_PRIORITY - 2) - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); /* * We do this so kswapd doesn't build up large priorities for @@ -1458,7 +1458,7 @@ unsigned long shrink_all_memory(unsigned long nr_pages) goto out; if (sc.nr_scanned && prio < DEF_PRIORITY - 2) - blk_congestion_wait(WRITE, HZ / 10); + congestion_wait(WRITE, HZ / 10); } lru_pages = 0; -- cgit v1.2.3-70-g09d2 From e51959faa61278c762389802faf8ba1a40676628 Mon Sep 17 00:00:00 2001 From: Zachary Amsden Date: Thu, 19 Oct 2006 23:29:04 -0700 Subject: [PATCH] Fix potential interrupts during alternative patching Interrupts must be disabled during alternative instruction patching. On systems with high timer IRQ rates, or when running in an emulator, timing differences can result in random kernel panics because of running partially patched instructions. This doesn't yet fix NMIs, which requires extricating the patch code from the late bug checking and is logically separate (and also less likely to cause problems). Signed-off-by: Zachary Amsden Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/alternative.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c index 28ab8064976..583c238e17f 100644 --- a/arch/i386/kernel/alternative.c +++ b/arch/i386/kernel/alternative.c @@ -344,6 +344,7 @@ void alternatives_smp_switch(int smp) void __init alternative_instructions(void) { + unsigned long flags; if (no_replacement) { printk(KERN_INFO "(SMP-)alternatives turned off\n"); free_init_pages("SMP alternatives", @@ -351,6 +352,8 @@ void __init alternative_instructions(void) (unsigned long)__smp_alt_end); return; } + + local_irq_save(flags); apply_alternatives(__alt_instructions, __alt_instructions_end); /* switch to patch-once-at-boottime-only mode and free the @@ -386,4 +389,5 @@ void __init alternative_instructions(void) alternatives_smp_switch(0); } #endif + local_irq_restore(flags); } -- cgit v1.2.3-70-g09d2 From 13892de19eb9007ea47430c701bdbf69df71d883 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] i386: Update defconfig Signed-off-by: Andi Kleen --- arch/i386/defconfig | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/defconfig b/arch/i386/defconfig index 60c0c02574f..97aacd6bd7d 100644 --- a/arch/i386/defconfig +++ b/arch/i386/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-rc1 -# Thu Oct 5 13:04:53 2006 +# Linux kernel version: 2.6.19-rc2-git4 +# Sat Oct 21 03:38:56 2006 # CONFIG_X86_32=y CONFIG_GENERIC_TIME=y @@ -380,8 +380,8 @@ CONFIG_INET6_XFRM_MODE_TRANSPORT=y CONFIG_INET6_XFRM_MODE_TUNNEL=y # CONFIG_INET6_XFRM_MODE_BEET is not set # CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set +CONFIG_IPV6_SIT=y # CONFIG_IPV6_TUNNEL is not set -# CONFIG_IPV6_SUBTREES is not set # CONFIG_IPV6_MULTIPLE_TABLES is not set # CONFIG_NETWORK_SECMARK is not set # CONFIG_NETFILTER is not set @@ -482,6 +482,13 @@ CONFIG_BLK_DEV_INITRD=y # CONFIG_CDROM_PKTCDVD is not set # CONFIG_ATA_OVER_ETH is not set +# +# Misc devices +# +# CONFIG_IBM_ASM is not set +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + # # ATA/ATAPI/MFM/RLL support # @@ -1024,6 +1031,7 @@ CONFIG_HANGCHECK_TIMER=y # # Dallas's 1-wire bus # +# CONFIG_W1 is not set # # Hardware Monitoring support @@ -1031,12 +1039,6 @@ CONFIG_HANGCHECK_TIMER=y # CONFIG_HWMON is not set # CONFIG_HWMON_VID is not set -# -# Misc devices -# -# CONFIG_IBM_ASM is not set -# CONFIG_TIFM_CORE is not set - # # Multimedia devices # @@ -1169,7 +1171,6 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_ATI_REMOTE2 is not set # CONFIG_USB_KEYSPAN_REMOTE is not set # CONFIG_USB_APPLETOUCH is not set -# CONFIG_USB_TRANCEVIBRATOR is not set # # USB Imaging devices @@ -1215,6 +1216,7 @@ CONFIG_USB_MON=y # CONFIG_USB_APPLEDISPLAY is not set # CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_LD is not set +# CONFIG_USB_TRANCEVIBRATOR is not set # CONFIG_USB_TEST is not set # @@ -1284,6 +1286,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y CONFIG_EXT3_FS_POSIX_ACL=y # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -1307,6 +1310,7 @@ CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set CONFIG_AUTOFS4_FS=y # CONFIG_FUSE_FS is not set +CONFIG_GENERIC_ACL=y # # CD-ROM/DVD Filesystems @@ -1384,7 +1388,6 @@ CONFIG_SUNRPC=y # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set # CONFIG_9P_FS is not set -CONFIG_GENERIC_ACL=y # # Partition Types @@ -1436,10 +1439,6 @@ CONFIG_NLS_ISO8859_15=y # CONFIG_NLS_KOI8_U is not set CONFIG_NLS_UTF8=y -# -# Distributed Lock Manager -# - # # Instrumentation Support # @@ -1480,6 +1479,7 @@ CONFIG_DEBUG_BUGVERBOSE=y CONFIG_UNWIND_INFO=y CONFIG_STACK_UNWIND=y # CONFIG_FORCED_INLINING is not set +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_LKDTM is not set CONFIG_EARLY_PRINTK=y -- cgit v1.2.3-70-g09d2 From da8604cc2d6cd4cbde873c7ba308365e402ac7bf Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] i386: fix .cfi_signal_frame copy-n-paste error This was copied, pasted but not edited. Cc: Andi Kleen Cc: Jan Beulich Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 7cc0b189b82..2d9d756fdd7 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -51,8 +51,8 @@ cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) # is .cfi_signal_frame supported too? -cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) -AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) +cflags-y += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) +AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) CFLAGS += $(cflags-y) -- cgit v1.2.3-70-g09d2 From cdfce1f5714fec7b24715f569b2fee1607350a6d Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86: Use -maccumulate-outgoing-args This avoids some problems with gcc 4.x and earlier generating invalid unwind information. In 4.1 the option is default when unwind information is enabled. And it seems to generate smaller code too, so it's probably a good thing on its own. With gcc 4.0: i386: 4683198 902112 480868 6066178 5c9002 vmlinux (before) 4449895 902112 480868 5832875 5900ab vmlinux (after) x86-64: 4939761 1449584 648216 7037561 6b6279 vmlinux (before) 4854193 1449584 648216 6951993 6a1439 vmlinux (after) On 4.1 it shouldn't make much difference because it is default when unwind is enabled anyways. Suggested by Michael Matz and Jan Beulich Cc: jbeulich@novell.com Signed-off-by: Andi Kleen --- arch/i386/Makefile | 4 ++++ arch/x86_64/Makefile | 4 ++++ 2 files changed, 8 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 2d9d756fdd7..0677908dfa0 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -42,6 +42,10 @@ cflags-$(CONFIG_REGPARM) += -mregparm=3 # temporary until string.h is fixed cflags-y += -ffreestanding +# this works around some issues with generating unwind tables in older gccs +# newer gccs do it by default +cflags-y += -maccumulate-outgoing-args + # Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use # a lot more stack due to the lack of sharing of stacklots: CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then echo $(call cc-option,-fno-unit-at-a-time); fi ;) diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index 1c0f18d4f88..13972148058 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -54,6 +54,10 @@ endif cflags-y += $(call cc-option,-funit-at-a-time) # prevent gcc from generating any FP code by mistake cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,) +# this works around some issues with generating unwind tables in older gccs +# newer gccs do it by default +cflags-y += -maccumulate-outgoing-args + # do binutils support CFI? cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) -- cgit v1.2.3-70-g09d2 From 26fd5e084e470dbe8edc6f726fc918e89b9f988c Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] i386: Fix fake return address The fake return address was being set to __KERNEL_PDA, rather than 0. Push it earlier while %eax still equals 0. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: Andrew Morton --- arch/i386/kernel/head.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index be9d883c62c..ca31f18d277 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -317,7 +317,7 @@ is386: movl $2,%ecx # set MP movl %eax,%gs lldt %ax cld # gcc2 wants the direction flag cleared at all times - pushl %eax # fake return address + pushl $0 # fake return address for unwinder #ifdef CONFIG_SMP movb ready, %cl movb $1, ready -- cgit v1.2.3-70-g09d2 From a1bae67243512ca30ceda48e3e24e25b543f8ab7 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] i386: Disable nmi watchdog on all ThinkPads Even newer Thinkpads have bugs in SMM code that causes hangs with NMI watchdog. Signed-off-by: Andi Kleen --- arch/i386/kernel/nmi.c | 10 +++++----- drivers/firmware/dmi_scan.c | 20 ++++++++++++++++++++ include/linux/dmi.h | 2 ++ 3 files changed, 27 insertions(+), 5 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index 3e8e3adb048..eaafe233a5d 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -219,11 +219,11 @@ static int __init check_nmi_watchdog(void) int cpu; /* Enable NMI watchdog for newer systems. - Actually it should be safe for most systems before 2004 too except - for some IBM systems that corrupt registers when NMI happens - during SMM. Unfortunately we don't have more exact information - on these and use this coarse check. */ - if (nmi_watchdog == NMI_DEFAULT && dmi_get_year(DMI_BIOS_DATE) >= 2004) + Probably safe on most older systems too, but let's be careful. + IBM ThinkPads use INT10 inside SMM and that allows early NMI inside SMM + which hangs the system. Disable watchdog for all thinkpads */ + if (nmi_watchdog == NMI_DEFAULT && dmi_get_year(DMI_BIOS_DATE) >= 2004 && + !dmi_name_in_vendors("ThinkPad")) nmi_watchdog = NMI_LOCAL_APIC; if ((nmi_watchdog == NMI_NONE) || (nmi_watchdog == NMI_DEFAULT)) diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c index b8b596d5778..37deee6c0c1 100644 --- a/drivers/firmware/dmi_scan.c +++ b/drivers/firmware/dmi_scan.c @@ -326,6 +326,26 @@ char *dmi_get_system_info(int field) } EXPORT_SYMBOL(dmi_get_system_info); + +/** + * dmi_name_in_vendors - Check if string is anywhere in the DMI vendor information. + * @str: Case sensitive Name + */ +int dmi_name_in_vendors(char *str) +{ + static int fields[] = { DMI_BIOS_VENDOR, DMI_BIOS_VERSION, DMI_SYS_VENDOR, + DMI_PRODUCT_NAME, DMI_PRODUCT_VERSION, DMI_BOARD_VENDOR, + DMI_BOARD_NAME, DMI_BOARD_VERSION, DMI_NONE }; + int i; + for (i = 0; fields[i] != DMI_NONE; i++) { + int f = fields[i]; + if (dmi_ident[f] && strstr(dmi_ident[f], str)) + return 1; + } + return 0; +} +EXPORT_SYMBOL(dmi_name_in_vendors); + /** * dmi_find_device - find onboard device by type/name * @type: device type or %DMI_DEV_TYPE_ANY to match all device types diff --git a/include/linux/dmi.h b/include/linux/dmi.h index 38dc403be70..904bf3d2d90 100644 --- a/include/linux/dmi.h +++ b/include/linux/dmi.h @@ -69,6 +69,7 @@ extern struct dmi_device * dmi_find_device(int type, const char *name, struct dmi_device *from); extern void dmi_scan_machine(void); extern int dmi_get_year(int field); +extern int dmi_name_in_vendors(char *str); #else @@ -77,6 +78,7 @@ static inline char * dmi_get_system_info(int field) { return NULL; } static inline struct dmi_device * dmi_find_device(int type, const char *name, struct dmi_device *from) { return NULL; } static inline int dmi_get_year(int year) { return 0; } +static inline int dmi_name_in_vendors(char *s) { return 0; } #endif -- cgit v1.2.3-70-g09d2 From 8cf2c51927bbeefafc25193d01b91f9ed3806e96 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86: Revert new unwind kernel stack termination Jan convinced me that it was unnecessary because the assembly stubs do this already on the stack. Cc: jbeulich@novell.com Signed-off-by: Andi Kleen --- arch/i386/kernel/process.c | 6 +----- arch/x86_64/kernel/entry.S | 5 ----- 2 files changed, 1 insertion(+), 10 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 57d375900af..1e1fa3e391a 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -336,7 +336,6 @@ extern void kernel_thread_helper(void); int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) { struct pt_regs regs; - int err; memset(®s, 0, sizeof(regs)); @@ -351,10 +350,7 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) regs.eflags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2; /* Ok, create the new process.. */ - err = do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL); - if (err == 0) /* terminate kernel stack */ - task_pt_regs(current)->eip = 0; - return err; + return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL); } EXPORT_SYMBOL(kernel_thread); diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index e3eddde3d3b..7d401b00d82 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -980,11 +980,6 @@ ENTRY(kernel_thread) call do_fork movq %rax,RAX(%rsp) xorl %edi,%edi - test %rax,%rax - jnz 1f - /* terminate stack in child */ - movq %rdi,RIP(%rsp) -1: /* * It isn't worth to check for reschedule here, -- cgit v1.2.3-70-g09d2 From 6b5c76b8e2ff204fa8d7201acce461188873bf2b Mon Sep 17 00:00:00 2001 From: Eiichiro Oiwa Date: Mon, 23 Oct 2006 15:14:07 +0900 Subject: PCI: fix pci_fixup_video as it blows up on sparc64 This reverts much of the original pci_fixup_video change and makes it work for all arches that need it. fixed, and tested on x86, x86_64 and IA64 dig. Signed-off-by: Eiichiro Oiwa Acked-by: David Miller Signed-off-by: Greg Kroah-Hartman --- arch/i386/pci/fixup.c | 55 ++++++++++++++++++++++++++++++++++++++++ arch/ia64/pci/Makefile | 2 +- arch/ia64/pci/fixup.c | 69 ++++++++++++++++++++++++++++++++++++++++++++++++++ drivers/pci/quirks.c | 46 --------------------------------- drivers/pci/rom.c | 5 ++-- 5 files changed, 128 insertions(+), 49 deletions(-) create mode 100644 arch/ia64/pci/fixup.c (limited to 'arch/i386') diff --git a/arch/i386/pci/fixup.c b/arch/i386/pci/fixup.c index 908b410f4c9..c1949ff38d6 100644 --- a/arch/i386/pci/fixup.c +++ b/arch/i386/pci/fixup.c @@ -342,6 +342,61 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PB1, pcie_r DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PC, pcie_rootport_aspm_quirk ); DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PC1, pcie_rootport_aspm_quirk ); +/* + * Fixup to mark boot BIOS video selected by BIOS before it changes + * + * From information provided by "Jon Smirl" + * + * The standard boot ROM sequence for an x86 machine uses the BIOS + * to select an initial video card for boot display. This boot video + * card will have it's BIOS copied to C0000 in system RAM. + * IORESOURCE_ROM_SHADOW is used to associate the boot video + * card with this copy. On laptops this copy has to be used since + * the main ROM may be compressed or combined with another image. + * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW + * is marked here since the boot video device will be the only enabled + * video device at this point. + */ + +static void __devinit pci_fixup_video(struct pci_dev *pdev) +{ + struct pci_dev *bridge; + struct pci_bus *bus; + u16 config; + + if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) + return; + + /* Is VGA routed to us? */ + bus = pdev->bus; + while (bus) { + bridge = bus->self; + + /* + * From information provided by + * "David Miller" + * The bridge control register is valid for PCI header + * type BRIDGE, or CARDBUS. Host to PCI controllers use + * PCI header type NORMAL. + */ + if (bridge + &&((bridge->hdr_type == PCI_HEADER_TYPE_BRIDGE) + ||(bridge->hdr_type == PCI_HEADER_TYPE_CARDBUS))) { + pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, + &config); + if (!(config & PCI_BRIDGE_CTL_VGA)) + return; + } + bus = bus->parent; + } + pci_read_config_word(pdev, PCI_COMMAND, &config); + if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { + pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; + printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_video); + /* * Some Toshiba laptops need extra code to enable their TI TSB43AB22/A. * diff --git a/arch/ia64/pci/Makefile b/arch/ia64/pci/Makefile index e66889e6922..fb14dc520d2 100644 --- a/arch/ia64/pci/Makefile +++ b/arch/ia64/pci/Makefile @@ -1,4 +1,4 @@ # # Makefile for the ia64-specific parts of the pci bus # -obj-y := pci.o +obj-y := pci.o fixup.o diff --git a/arch/ia64/pci/fixup.c b/arch/ia64/pci/fixup.c new file mode 100644 index 00000000000..245dc1fedc2 --- /dev/null +++ b/arch/ia64/pci/fixup.c @@ -0,0 +1,69 @@ +/* + * Exceptions for specific devices. Usually work-arounds for fatal design flaws. + * Derived from fixup.c of i386 tree. + */ + +#include +#include + +#include + +/* + * Fixup to mark boot BIOS video selected by BIOS before it changes + * + * From information provided by "Jon Smirl" + * + * The standard boot ROM sequence for an x86 machine uses the BIOS + * to select an initial video card for boot display. This boot video + * card will have it's BIOS copied to C0000 in system RAM. + * IORESOURCE_ROM_SHADOW is used to associate the boot video + * card with this copy. On laptops this copy has to be used since + * the main ROM may be compressed or combined with another image. + * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW + * is marked here since the boot video device will be the only enabled + * video device at this point. + */ + +static void __devinit pci_fixup_video(struct pci_dev *pdev) +{ + struct pci_dev *bridge; + struct pci_bus *bus; + u16 config; + + if ((strcmp(platform_name, "dig") != 0) + && (strcmp(platform_name, "hpzx1") != 0)) + return; + /* Maybe, this machine supports legacy memory map. */ + + if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) + return; + + /* Is VGA routed to us? */ + bus = pdev->bus; + while (bus) { + bridge = bus->self; + + /* + * From information provided by + * "David Miller" + * The bridge control register is valid for PCI header + * type BRIDGE, or CARDBUS. Host to PCI controllers use + * PCI header type NORMAL. + */ + if (bridge + &&((bridge->hdr_type == PCI_HEADER_TYPE_BRIDGE) + ||(bridge->hdr_type == PCI_HEADER_TYPE_CARDBUS))) { + pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, + &config); + if (!(config & PCI_BRIDGE_CTL_VGA)) + return; + } + bus = bus->parent; + } + pci_read_config_word(pdev, PCI_COMMAND, &config); + if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { + pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; + printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_video); diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index e8a7f1b1b2b..687ab4a0c6c 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1619,52 +1619,6 @@ static void __devinit fixup_rev1_53c810(struct pci_dev* dev) } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NCR, PCI_DEVICE_ID_NCR_53C810, fixup_rev1_53c810); -/* - * Fixup to mark boot BIOS video selected by BIOS before it changes - * - * From information provided by "Jon Smirl" - * - * The standard boot ROM sequence for an x86 machine uses the BIOS - * to select an initial video card for boot display. This boot video - * card will have it's BIOS copied to C0000 in system RAM. - * IORESOURCE_ROM_SHADOW is used to associate the boot video - * card with this copy. On laptops this copy has to be used since - * the main ROM may be compressed or combined with another image. - * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW - * is marked here since the boot video device will be the only enabled - * video device at this point. - */ - -static void __devinit fixup_video(struct pci_dev *pdev) -{ - struct pci_dev *bridge; - struct pci_bus *bus; - u16 config; - - if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) - return; - - /* Is VGA routed to us? */ - bus = pdev->bus; - while (bus) { - bridge = bus->self; - if (bridge) { - pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, - &config); - if (!(config & PCI_BRIDGE_CTL_VGA)) - return; - } - bus = bus->parent; - } - pci_read_config_word(pdev, PCI_COMMAND, &config); - if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { - pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; - printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, fixup_video); - - static void pci_do_fixups(struct pci_dev *dev, struct pci_fixup *f, struct pci_fixup *end) { while (f < end) { diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index 43e4a49f2cc..e1dcefc69bb 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -72,8 +72,9 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) int last_image; /* - * IORESOURCE_ROM_SHADOW set if the VGA enable bit of the Bridge Control - * register is set for embedded VGA. + * IORESOURCE_ROM_SHADOW set on x86, x86_64 and IA64 supports legacy + * memory map if the VGA enable bit of the Bridge Control register is + * set for embedded VGA. */ if (res->flags & IORESOURCE_ROM_SHADOW) { /* primary video rom always starts here */ -- cgit v1.2.3-70-g09d2 From 61ce1efe6e40233663d27ab8ac9ba9710eebcaad Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 27 Oct 2006 11:41:44 -0700 Subject: [PATCH] vmlinux.lds: consolidate initcall sections Add a vmlinux.lds.h helper macro for defining the eight-level initcall table, teach all the architectures to use it. This is a prerequisite for a patch which performs initcall synchronisation for multithreaded-probing. Cc: Greg KH Signed-off-by: Andrew Morton [ Added AVR32 as well ] Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds --- arch/alpha/kernel/vmlinux.lds.S | 8 +------- arch/arm/kernel/vmlinux.lds.S | 8 +------- arch/avr32/kernel/vmlinux.lds.c | 8 +------- arch/frv/kernel/vmlinux.lds.S | 8 +------- arch/h8300/kernel/vmlinux.lds.S | 8 +------- arch/i386/kernel/vmlinux.lds.S | 8 +------- arch/ia64/kernel/vmlinux.lds.S | 8 +------- arch/m32r/kernel/vmlinux.lds.S | 8 +------- arch/m68knommu/kernel/vmlinux.lds.S | 8 +------- arch/mips/kernel/vmlinux.lds.S | 8 +------- arch/parisc/kernel/vmlinux.lds.S | 8 +------- arch/powerpc/kernel/vmlinux.lds.S | 8 +------- arch/ppc/kernel/vmlinux.lds.S | 8 +------- arch/s390/kernel/vmlinux.lds.S | 8 +------- arch/sh/kernel/vmlinux.lds.S | 8 +------- arch/sh64/kernel/vmlinux.lds.S | 8 +------- arch/sparc/kernel/vmlinux.lds.S | 8 +------- arch/sparc64/kernel/vmlinux.lds.S | 8 +------- arch/v850/kernel/vmlinux.lds.S | 8 +------- arch/x86_64/kernel/vmlinux.lds.S | 8 +------- arch/xtensa/kernel/vmlinux.lds.S | 8 +------- include/asm-generic/vmlinux.lds.h | 10 ++++++++++ 22 files changed, 31 insertions(+), 147 deletions(-) (limited to 'arch/i386') diff --git a/arch/alpha/kernel/vmlinux.lds.S b/arch/alpha/kernel/vmlinux.lds.S index 71470e9d93b..76bf071e376 100644 --- a/arch/alpha/kernel/vmlinux.lds.S +++ b/arch/alpha/kernel/vmlinux.lds.S @@ -48,13 +48,7 @@ SECTIONS . = ALIGN(8); __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 3ca574ee277..a8fa75ea07a 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -45,13 +45,7 @@ SECTIONS *(.early_param.init) __early_end = .; __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; __con_initcall_start = .; *(.con_initcall.init) diff --git a/arch/avr32/kernel/vmlinux.lds.c b/arch/avr32/kernel/vmlinux.lds.c index cdd627c6b7d..5c4424e362b 100644 --- a/arch/avr32/kernel/vmlinux.lds.c +++ b/arch/avr32/kernel/vmlinux.lds.c @@ -38,13 +38,7 @@ SECTIONS __setup_end = .; . = ALIGN(4); __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; __con_initcall_start = .; *(.con_initcall.init) diff --git a/arch/frv/kernel/vmlinux.lds.S b/arch/frv/kernel/vmlinux.lds.S index f474534ba78..9c1fb12367f 100644 --- a/arch/frv/kernel/vmlinux.lds.S +++ b/arch/frv/kernel/vmlinux.lds.S @@ -44,13 +44,7 @@ SECTIONS __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/h8300/kernel/vmlinux.lds.S b/arch/h8300/kernel/vmlinux.lds.S index 6406c388f88..756325dd480 100644 --- a/arch/h8300/kernel/vmlinux.lds.S +++ b/arch/h8300/kernel/vmlinux.lds.S @@ -118,13 +118,7 @@ SECTIONS . = ALIGN(0x4) ; ___setup_end = .; ___initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS ___initcall_end = .; ___con_initcall_start = .; *(.con_initcall.init) diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 1e7ac1c44dd..adc1f232afe 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -126,13 +126,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S index b3b2e389d6b..d6083a0936f 100644 --- a/arch/ia64/kernel/vmlinux.lds.S +++ b/arch/ia64/kernel/vmlinux.lds.S @@ -128,13 +128,7 @@ SECTIONS .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; } diff --git a/arch/m32r/kernel/vmlinux.lds.S b/arch/m32r/kernel/vmlinux.lds.S index 13c7bb698e3..358b9cee2c6 100644 --- a/arch/m32r/kernel/vmlinux.lds.S +++ b/arch/m32r/kernel/vmlinux.lds.S @@ -83,13 +83,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/m68knommu/kernel/vmlinux.lds.S b/arch/m68knommu/kernel/vmlinux.lds.S index ccd2ceb05cf..58afa8be604 100644 --- a/arch/m68knommu/kernel/vmlinux.lds.S +++ b/arch/m68knommu/kernel/vmlinux.lds.S @@ -140,13 +140,7 @@ SECTIONS { *(.init.setup) __setup_end = .; __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; __con_initcall_start = .; *(.con_initcall.init) diff --git a/arch/mips/kernel/vmlinux.lds.S b/arch/mips/kernel/vmlinux.lds.S index 0bb9cd88945..25ed3337ce3 100644 --- a/arch/mips/kernel/vmlinux.lds.S +++ b/arch/mips/kernel/vmlinux.lds.S @@ -91,13 +91,7 @@ SECTIONS __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; diff --git a/arch/parisc/kernel/vmlinux.lds.S b/arch/parisc/kernel/vmlinux.lds.S index b3677fc8eef..7b943b45f7c 100644 --- a/arch/parisc/kernel/vmlinux.lds.S +++ b/arch/parisc/kernel/vmlinux.lds.S @@ -153,13 +153,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S index cb0e8d46c3e..e8342d86753 100644 --- a/arch/powerpc/kernel/vmlinux.lds.S +++ b/arch/powerpc/kernel/vmlinux.lds.S @@ -108,13 +108,7 @@ SECTIONS .initcall.init : { __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; } diff --git a/arch/ppc/kernel/vmlinux.lds.S b/arch/ppc/kernel/vmlinux.lds.S index 095fd332332..16e8661e1fe 100644 --- a/arch/ppc/kernel/vmlinux.lds.S +++ b/arch/ppc/kernel/vmlinux.lds.S @@ -115,13 +115,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index af9e69a0301..fe0f2e97ba7 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -83,13 +83,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S index 5eb93091818..77b4026d568 100644 --- a/arch/sh/kernel/vmlinux.lds.S +++ b/arch/sh/kernel/vmlinux.lds.S @@ -76,13 +76,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/sh64/kernel/vmlinux.lds.S b/arch/sh64/kernel/vmlinux.lds.S index a8fcc3a7158..95c4d753e35 100644 --- a/arch/sh64/kernel/vmlinux.lds.S +++ b/arch/sh64/kernel/vmlinux.lds.S @@ -108,13 +108,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : C_PHYS(.initcall.init) { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index 1dd78c84888..5cc5ff7f882 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -49,13 +49,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/sparc64/kernel/vmlinux.lds.S b/arch/sparc64/kernel/vmlinux.lds.S index b097379a49a..bd9de8c2a2a 100644 --- a/arch/sparc64/kernel/vmlinux.lds.S +++ b/arch/sparc64/kernel/vmlinux.lds.S @@ -57,13 +57,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/v850/kernel/vmlinux.lds.S b/arch/v850/kernel/vmlinux.lds.S index 63399219cd9..88d087f527c 100644 --- a/arch/v850/kernel/vmlinux.lds.S +++ b/arch/v850/kernel/vmlinux.lds.S @@ -140,13 +140,7 @@ ___setup_end = . ; \ ___initcall_start = . ; \ *(.initcall.init) \ - *(.initcall1.init) \ - *(.initcall2.init) \ - *(.initcall3.init) \ - *(.initcall4.init) \ - *(.initcall5.init) \ - *(.initcall6.init) \ - *(.initcall7.init) \ + INITCALLS \ . = ALIGN (4) ; \ ___initcall_end = . ; \ ___con_initcall_start = .; \ diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index 1283614c9b2..edb24aa714b 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -175,13 +175,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/xtensa/kernel/vmlinux.lds.S b/arch/xtensa/kernel/vmlinux.lds.S index ab6cdbd5eb6..cfe75f52872 100644 --- a/arch/xtensa/kernel/vmlinux.lds.S +++ b/arch/xtensa/kernel/vmlinux.lds.S @@ -184,13 +184,7 @@ SECTIONS __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 9d0d11c180d..e3e83bcaf71 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -213,3 +213,13 @@ #define NOTES \ .notes : { *(.note.*) } :note + +#define INITCALLS \ + *(.initcall1.init) \ + *(.initcall2.init) \ + *(.initcall3.init) \ + *(.initcall4.init) \ + *(.initcall5.init) \ + *(.initcall6.init) \ + *(.initcall7.init) + -- cgit v1.2.3-70-g09d2 From ae74589cb381cc2838107f92d4e0e1f178c6825d Mon Sep 17 00:00:00 2001 From: "bibo,mao" Date: Sat, 28 Oct 2006 10:38:29 -0700 Subject: [PATCH] fix efi_memory_present_wrapper() efi_memory_present_wrapper() parameter start/end is physical address, but function memory_present parameter is PFN, this patch converts physical address to PFN. Signed-off-by: bibo, mao Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/setup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 519e63c3c13..141041dde74 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -846,7 +846,7 @@ efi_find_max_pfn(unsigned long start, unsigned long end, void *arg) static int __init efi_memory_present_wrapper(unsigned long start, unsigned long end, void *arg) { - memory_present(0, start, end); + memory_present(0, PFN_UP(start), PFN_DOWN(end)); return 0; } -- cgit v1.2.3-70-g09d2 From 08d892f11aae7125fe078cf93ec5cf6af288c5e7 Mon Sep 17 00:00:00 2001 From: Andrey Panin Date: Sat, 28 Oct 2006 10:38:35 -0700 Subject: [PATCH] visws build fix Fix this: > Subject : CONFIG_X86_VISWS=3Dy, CONFIG_SMP=3Dn compile error > References : http://lkml.org/lkml/2006/10/7/51 > Submitter : Jesper Juhl > Caused-By : David Howells > commit 7d12e780e003f93433d49ce78cfedf4b4c52adc5 > Status : unknown Via undescribed means. Signed-off-by: Andrey Panin Cc: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mach-visws/visws_apic.c | 7 ++--- include/asm-i386/mach-visws/do_timer.h | 53 --------------------------------- include/asm-i386/mach-visws/mach_apic.h | 5 ++++ 3 files changed, 8 insertions(+), 57 deletions(-) delete mode 100644 include/asm-i386/mach-visws/do_timer.h (limited to 'arch/i386') diff --git a/arch/i386/mach-visws/visws_apic.c b/arch/i386/mach-visws/visws_apic.c index 07097ed4889..38c2b13124d 100644 --- a/arch/i386/mach-visws/visws_apic.c +++ b/arch/i386/mach-visws/visws_apic.c @@ -122,7 +122,7 @@ static void end_cobalt_irq(unsigned int irq) spin_unlock_irqrestore(&cobalt_lock, flags); } -static struct hw_interrupt_type cobalt_irq_type = { +static struct irq_chip cobalt_irq_type = { .typename = "Cobalt-APIC", .startup = startup_cobalt_irq, .shutdown = disable_cobalt_irq, @@ -159,7 +159,7 @@ static void end_piix4_master_irq(unsigned int irq) spin_unlock_irqrestore(&cobalt_lock, flags); } -static struct hw_interrupt_type piix4_master_irq_type = { +static struct irq_chip piix4_master_irq_type = { .typename = "PIIX4-master", .startup = startup_piix4_master_irq, .ack = ack_cobalt_irq, @@ -167,9 +167,8 @@ static struct hw_interrupt_type piix4_master_irq_type = { }; -static struct hw_interrupt_type piix4_virtual_irq_type = { +static struct irq_chip piix4_virtual_irq_type = { .typename = "PIIX4-virtual", - .startup = startup_8259A_irq, .shutdown = disable_8259A_irq, .enable = enable_8259A_irq, .disable = disable_8259A_irq, diff --git a/include/asm-i386/mach-visws/do_timer.h b/include/asm-i386/mach-visws/do_timer.h deleted file mode 100644 index 21cd696d4d0..00000000000 --- a/include/asm-i386/mach-visws/do_timer.h +++ /dev/null @@ -1,53 +0,0 @@ -/* defines for inline arch setup functions */ - -#include -#include -#include "cobalt.h" - -static inline void do_timer_interrupt_hook(void) -{ - /* Clear the interrupt */ - co_cpu_write(CO_CPU_STAT,co_cpu_read(CO_CPU_STAT) & ~CO_STAT_TIMEINTR); - - do_timer(1); -#ifndef CONFIG_SMP - update_process_times(user_mode_vm(irq_regs)); -#endif -/* - * In the SMP case we use the local APIC timer interrupt to do the - * profiling, except when we simulate SMP mode on a uniprocessor - * system, in that case we have to call the local interrupt handler. - */ -#ifndef CONFIG_X86_LOCAL_APIC - profile_tick(CPU_PROFILING); -#else - if (!using_apic_timer) - smp_local_timer_interrupt(); -#endif -} - -static inline int do_timer_overflow(int count) -{ - int i; - - spin_lock(&i8259A_lock); - /* - * This is tricky when I/O APICs are used; - * see do_timer_interrupt(). - */ - i = inb(0x20); - spin_unlock(&i8259A_lock); - - /* assumption about timer being IRQ0 */ - if (i & 0x01) { - /* - * We cannot detect lost timer interrupts ... - * well, that's why we call them lost, don't we? :) - * [hmm, on the Pentium and Alpha we can ... sort of] - */ - count -= LATCH; - } else { - printk("do_slow_gettimeoffset(): hardware timer problem?\n"); - } - return count; -} diff --git a/include/asm-i386/mach-visws/mach_apic.h b/include/asm-i386/mach-visws/mach_apic.h index de438c7147a..18afe6b6fc4 100644 --- a/include/asm-i386/mach-visws/mach_apic.h +++ b/include/asm-i386/mach-visws/mach_apic.h @@ -51,6 +51,11 @@ static inline void clustered_apic_check(void) { } +static inline int apicid_to_node(int logical_apicid) +{ + return 0; +} + /* Mapping from cpu number to logical apicid */ static inline int cpu_to_logical_apicid(int cpu) { -- cgit v1.2.3-70-g09d2 From 3f4b23e9833f0816b57d07401eac5e184b627fd7 Mon Sep 17 00:00:00 2001 From: Kristian Mueller Date: Sun, 29 Oct 2006 22:46:46 -0800 Subject: [PATCH] APM: URL of APM 1.2 specs has changed APM BIOS Interface Secification can now be found at http://www.microsoft.com/whdc/archive/amp_12.mspx Signed-off-by: Kristian Mueller Acked-by: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/apm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c index 2af65858d32..a60358fe9a4 100644 --- a/arch/i386/kernel/apm.c +++ b/arch/i386/kernel/apm.c @@ -198,7 +198,7 @@ * (APM) BIOS Interface Specification, Revision 1.2, February 1996. * * [This document is available from Microsoft at: - * http://www.microsoft.com/hwdev/busbios/amp_12.htm] + * http://www.microsoft.com/whdc/archive/amp_12.mspx] */ #include -- cgit v1.2.3-70-g09d2 From 130fe05dbc0114609cfef9815c0c5580b42decfa Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 1 Nov 2006 09:11:00 -0800 Subject: i386: clean up io-apic accesses This is preparation for fixing the ordering of the accesses that got broken by the commit cf4c6a2f27f5db810b69dcb1da7f194489e8ff88 when factoring out the "common" io apic routing entry accesses. Move the accessor function (that were only used by io_apic.c) out of a header file, and use proper memory-mapped accesses rather than making up our own "volatile" pointers. Signed-off-by: Linus Torvalds --- arch/i386/kernel/io_apic.c | 40 ++++++++++++++++++++++++++++++++++++++++ include/asm-i386/io_apic.h | 29 +---------------------------- 2 files changed, 41 insertions(+), 28 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 350192d6ab9..eb10bd5da64 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -91,6 +91,46 @@ static struct irq_pin_list { int apic, pin, next; } irq_2_pin[PIN_MAP_SIZE]; +struct io_apic { + unsigned int index; + unsigned int unused[3]; + unsigned int data; +}; + +static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) +{ + return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) + + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK); +} + +static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + return readl(&io_apic->data); +} + +static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + writel(value, &io_apic->data); +} + +/* + * Re-write a value: to be used for read-modify-write + * cycles where the read already set up the index register. + * + * Older SiS APIC requires we rewrite the index register + */ +static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value) +{ + volatile struct io_apic *io_apic = io_apic_base(apic); + if (sis_apic_bug) + writel(reg, &io_apic->index); + writel(value, &io_apic->data); +} + union entry_union { struct { u32 w1, w2; }; struct IO_APIC_route_entry entry; diff --git a/include/asm-i386/io_apic.h b/include/asm-i386/io_apic.h index 276ea7e8144..059a9ff28b4 100644 --- a/include/asm-i386/io_apic.h +++ b/include/asm-i386/io_apic.h @@ -12,10 +12,6 @@ #ifdef CONFIG_X86_IO_APIC -#define IO_APIC_BASE(idx) \ - ((volatile int *)(__fix_to_virt(FIX_IO_APIC_BASE_0 + idx) \ - + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK))) - /* * The structure of the IO-APIC: */ @@ -119,31 +115,8 @@ extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* non-0 if default (table-less) MP configuration */ extern int mpc_default_type; -static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) -{ - *IO_APIC_BASE(apic) = reg; - return *(IO_APIC_BASE(apic)+4); -} - -static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) -{ - *IO_APIC_BASE(apic) = reg; - *(IO_APIC_BASE(apic)+4) = value; -} - -/* - * Re-write a value: to be used for read-modify-write - * cycles where the read already set up the index register. - * - * Older SiS APIC requires we rewrite the index regiser - */ +/* Older SiS APIC requires we rewrite the index register */ extern int sis_apic_bug; -static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value) -{ - if (sis_apic_bug) - *IO_APIC_BASE(apic) = reg; - *(IO_APIC_BASE(apic)+4) = value; -} /* 1 if "noapic" boot option passed */ extern int skip_ioapic_setup; -- cgit v1.2.3-70-g09d2 From f9dadfa71bc594df09044da61d1c72701121d802 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 1 Nov 2006 10:05:35 -0800 Subject: i386: write IO APIC irq routing entries in correct order Since the "mask" bit is in the low word, when we write a new entry, we need to write the high word first, before we potentially unmask it. The exception is when we actually want to mask the interrupt, in which case we want to write the low word first to make sure that the high word doesn't change while the interrupt routing is still active. Signed-off-by: Linus Torvalds --- arch/i386/kernel/io_apic.c | 26 +++++++++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index eb10bd5da64..507983c513c 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -147,11 +147,33 @@ static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) return eu.entry; } +/* + * When we write a new IO APIC routing entry, we need to write the high + * word first! If the mask bit in the low word is clear, we will enable + * the interrupt, and we need to make sure the entry is fully populated + * before that happens. + */ static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { unsigned long flags; union entry_union eu; eu.entry = e; + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(apic, 0x11 + 2*pin, eu.w2); + io_apic_write(apic, 0x10 + 2*pin, eu.w1); + spin_unlock_irqrestore(&ioapic_lock, flags); +} + +/* + * When we mask an IO APIC routing entry, we need to write the low + * word first, in order to set the mask bit before we change the + * high bits! + */ +static void ioapic_mask_entry(int apic, int pin) +{ + unsigned long flags; + union entry_union eu = { .entry.mask = 1 }; + spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic, 0x10 + 2*pin, eu.w1); io_apic_write(apic, 0x11 + 2*pin, eu.w2); @@ -274,9 +296,7 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) /* * Disable it in the IO-APIC irq-routing table: */ - memset(&entry, 0, sizeof(entry)); - entry.mask = 1; - ioapic_write_entry(apic, pin, entry); + ioapic_mask_entry(apic, pin); } static void clear_IO_APIC (void) -- cgit v1.2.3-70-g09d2 From 6851ecc6e2fa4a01449a0fec9f4abd9aec43afde Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 2 Nov 2006 23:02:24 -0800 Subject: PCI: Revert "PCI: i386/x86_84: disable PCI resource decode on device disable" This reverts commit 53e4d30dd666d7f83598957ee4a415eefb47c9a6. It was found that it caused unneeded problems (see http://bugzilla.kernel.org/show_bug.cgi?id=7082 for details of one such issue. Signed-off-by: Greg Kroah-Hartman --- arch/i386/pci/common.c | 1 - arch/i386/pci/i386.c | 9 --------- arch/i386/pci/pci.h | 1 - 3 files changed, 11 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c index 6d5ace845e4..cdfcf971098 100644 --- a/arch/i386/pci/common.c +++ b/arch/i386/pci/common.c @@ -343,7 +343,6 @@ int pcibios_enable_device(struct pci_dev *dev, int mask) void pcibios_disable_device (struct pci_dev *dev) { - pcibios_disable_resources(dev); if (pcibios_disable_irq) pcibios_disable_irq(dev); } diff --git a/arch/i386/pci/i386.c b/arch/i386/pci/i386.c index 10154a2cac6..98580292f0d 100644 --- a/arch/i386/pci/i386.c +++ b/arch/i386/pci/i386.c @@ -242,15 +242,6 @@ int pcibios_enable_resources(struct pci_dev *dev, int mask) return 0; } -void pcibios_disable_resources(struct pci_dev *dev) -{ - u16 cmd; - - pci_read_config_word(dev, PCI_COMMAND, &cmd); - cmd &= ~(PCI_COMMAND_IO | PCI_COMMAND_MEMORY); - pci_write_config_word(dev, PCI_COMMAND, cmd); -} - /* * If we set up a device for bus mastering, we need to check the latency * timer as certain crappy BIOSes forget to set it properly. diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h index ad065cebd7b..a0a25180b61 100644 --- a/arch/i386/pci/pci.h +++ b/arch/i386/pci/pci.h @@ -43,7 +43,6 @@ extern unsigned int pcibios_max_latency; void pcibios_resource_survey(void); int pcibios_enable_resources(struct pci_dev *, int); -void pcibios_disable_resources(struct pci_dev *); /* pci-pc.c */ -- cgit v1.2.3-70-g09d2 From 90d53909443b3986569b38ef145f09ea2359af75 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 2 Nov 2006 22:07:18 -0800 Subject: [PATCH] acpi_noirq section fix WARNING: vmlinux - Section mismatch: reference to .init.data:acpi_noirq from .text between 'pcibios_penalize_isa_irq' (at offset 0xc026ffa1) and 'pirq_serverworks_get' Acked-by: "Brown, Len" Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/acpi/boot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c index ab974ff9707..22e4c466e5a 100644 --- a/arch/i386/kernel/acpi/boot.c +++ b/arch/i386/kernel/acpi/boot.c @@ -70,7 +70,7 @@ static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return #define PREFIX "ACPI: " -int acpi_noirq __initdata; /* skip ACPI IRQ initialization */ +int acpi_noirq; /* skip ACPI IRQ initialization */ int acpi_pci_disabled __initdata; /* skip ACPI PCI scan and IRQ initialization */ int acpi_ht __initdata = 1; /* enable HT */ -- cgit v1.2.3-70-g09d2 From de8e7c12430a73654ae3cedbc45428d56c6b777b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 8 Nov 2006 10:09:28 -0800 Subject: Revert "[PATCH] i386: Add MMCFG resources to i386 too" This reverts commit de09bddb9d6f96785be470c832b881e6d72d589f. It tried to reserve the MMCONFIG mmio memory ranges, but since the MMCONFIG information is broken and often bogus (which is why we don't dare use it most of the time _anyway_), it does more harm than good. Cc: Jeff Chua Cc: Adrian Bunk Cc: Andi Kleen Signed-off-by: Linus Torvalds --- arch/i386/pci/mmconfig.c | 35 ----------------------------------- 1 file changed, 35 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c index d0c3da3aa2a..c6b6d9bbc45 100644 --- a/arch/i386/pci/mmconfig.c +++ b/arch/i386/pci/mmconfig.c @@ -154,38 +154,6 @@ static struct pci_raw_ops pci_mmcfg = { .write = pci_mmcfg_write, }; - -static __init void pci_mmcfg_insert_resources(void) -{ -#define PCI_MMCFG_RESOURCE_NAME_LEN 19 - int i; - struct resource *res; - char *names; - unsigned num_buses; - - res = kcalloc(PCI_MMCFG_RESOURCE_NAME_LEN + sizeof(*res), - pci_mmcfg_config_num, GFP_KERNEL); - - if (!res) { - printk(KERN_ERR "PCI: Unable to allocate MMCONFIG resources\n"); - return; - } - - names = (void *)&res[pci_mmcfg_config_num]; - for (i = 0; i < pci_mmcfg_config_num; i++, res++) { - num_buses = pci_mmcfg_config[i].end_bus_number - - pci_mmcfg_config[i].start_bus_number + 1; - res->name = names; - snprintf(names, PCI_MMCFG_RESOURCE_NAME_LEN, "PCI MMCONFIG %u", - pci_mmcfg_config[i].pci_segment_group_number); - res->start = pci_mmcfg_config[i].base_address; - res->end = res->start + (num_buses << 20) - 1; - res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; - insert_resource(&iomem_resource, res); - names += PCI_MMCFG_RESOURCE_NAME_LEN; - } -} - /* K8 systems have some devices (typically in the builtin northbridge) that are only accessible using type1 Normally this can be expressed in the MCFG by not listing them @@ -222,8 +190,6 @@ static __init void unreachable_devices(void) } } - - void __init pci_mmcfg_init(int type) { if ((pci_probe & PCI_PROBE_MMCONF) == 0) @@ -251,5 +217,4 @@ void __init pci_mmcfg_init(int type) pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF; unreachable_devices(); - pci_mmcfg_insert_resources(); } -- cgit v1.2.3-70-g09d2 From d654c673d6394bc26e159b1057b357371b4ce1dc Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Wed, 8 Nov 2006 17:44:37 -0800 Subject: [PATCH] Regression in 2.6.19-rc microcode driver If the microcode driver is built in (rather than module) there are some, ehm, interesting effects happening due to the new "call out to userspace" behavior that is introduced.. and which runs too early. The result is a boot hang; which is really nasty. The patch below is a minimally safe patch to fix this regression for 2.6.19 by just not requesting actual microcode updates during early boot. (That is a good idea in general anyway) The "real" fix is a lot more complex given the entire cpu hotplug scenario (during cpu hotplug you normally need to load the microcode as well); but the interactions for that are just really messy at this point; this fix at least makes it work and avoids a full detangle of hotplug. Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/microcode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c index c4d0291b519..23f5984d065 100644 --- a/arch/i386/kernel/microcode.c +++ b/arch/i386/kernel/microcode.c @@ -577,7 +577,7 @@ static void microcode_init_cpu(int cpu) set_cpus_allowed(current, cpumask_of_cpu(cpu)); mutex_lock(µcode_mutex); collect_cpu_info(cpu); - if (uci->valid) + if (uci->valid && system_state == SYSTEM_RUNNING) cpu_request_microcode(cpu); mutex_unlock(µcode_mutex); set_cpus_allowed(current, old); -- cgit v1.2.3-70-g09d2 From c06cb8b1c4d25e5b4d7a2d7c2462619de1e0dbc4 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Wed, 8 Nov 2006 17:44:41 -0800 Subject: [PATCH] i386: Force data segment to be 4K aligned o Currently there is no specific alignment restriction in linker script and in some cases it can be placed non 4K aligned addresses. This fails kexec which checks that segment to be loaded is page aligned. o I guess, it does not harm data segment to be 4K aligned. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/vmlinux.lds.S | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index adc1f232afe..c6f84a0322b 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -51,6 +51,7 @@ SECTIONS __tracedata_end = .; /* writeable */ + . = ALIGN(4096); .data : AT(ADDR(.data) - LOAD_OFFSET) { /* Data */ *(.data) CONSTRUCTORS -- cgit v1.2.3-70-g09d2 From 8bdc052eccdc7893d075d3f1f7103594a458c8c4 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Wed, 8 Nov 2006 17:44:49 -0800 Subject: [PATCH] kretprobe: fix kretprobe-booster to save regs and set status There are two bugs in the kretprobe-booster. 1) It doesn't make room for gs registers. 2) It doesn't change status of the current kprobe. This status will effect the fault handling. This patch fixes these bugs and, additionally, saves skipped registers for compatibility with the original kretprobe. Signed-off-by: Masami Hiramatsu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/kprobes.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index d98e44b16fe..fc79e1e859c 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -361,8 +361,11 @@ no_kprobe: asm volatile ( ".global kretprobe_trampoline\n" "kretprobe_trampoline: \n" " pushf\n" - /* skip cs, eip, orig_eax, es, ds */ - " subl $20, %esp\n" + /* skip cs, eip, orig_eax */ + " subl $12, %esp\n" + " pushl %gs\n" + " pushl %ds\n" + " pushl %es\n" " pushl %eax\n" " pushl %ebp\n" " pushl %edi\n" @@ -373,10 +376,10 @@ no_kprobe: " movl %esp, %eax\n" " call trampoline_handler\n" /* move eflags to cs */ - " movl 48(%esp), %edx\n" - " movl %edx, 44(%esp)\n" + " movl 52(%esp), %edx\n" + " movl %edx, 48(%esp)\n" /* save true return address on eflags */ - " movl %eax, 48(%esp)\n" + " movl %eax, 52(%esp)\n" " popl %ebx\n" " popl %ecx\n" " popl %edx\n" @@ -384,8 +387,8 @@ no_kprobe: " popl %edi\n" " popl %ebp\n" " popl %eax\n" - /* skip eip, orig_eax, es, ds */ - " addl $16, %esp\n" + /* skip eip, orig_eax, es, ds, gs */ + " addl $20, %esp\n" " popf\n" " ret\n"); } @@ -404,6 +407,10 @@ fastcall void *__kprobes trampoline_handler(struct pt_regs *regs) INIT_HLIST_HEAD(&empty_rp); spin_lock_irqsave(&kretprobe_lock, flags); head = kretprobe_inst_table_head(current); + /* fixup registers */ + regs->xcs = __KERNEL_CS; + regs->eip = trampoline_address; + regs->orig_eax = 0xffffffff; /* * It is possible to have multiple instances associated with a given @@ -425,6 +432,7 @@ fastcall void *__kprobes trampoline_handler(struct pt_regs *regs) if (ri->rp && ri->rp->handler){ __get_cpu_var(current_kprobe) = &ri->rp->kp; + get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE; ri->rp->handler(ri, regs); __get_cpu_var(current_kprobe) = NULL; } -- cgit v1.2.3-70-g09d2 From ec68307cc5a8dc499e48693843bb42f6b6028458 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 8 Nov 2006 17:44:57 -0800 Subject: [PATCH] htirq: refactor so we only have one function that writes to the chip This refactoring actually optimizes the code a little by caching the value that we think the device is programmed with instead of reading it back from the hardware. Which simplifies the code a little and should speed things up a bit. This patch introduces the concept of a ht_irq_msg and modifies the architecture read/write routines to update this code. There is a minor consistency fix here as well as x86_64 forgot to initialize the htirq as masked. Signed-off-by: Eric W. Biederman Cc: Andi Kleen Acked-by: Bryan O'Sullivan Cc: Cc: Roland Dreier Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/io_apic.c | 26 ++++++++-------- arch/x86_64/kernel/io_apic.c | 31 +++++++++---------- drivers/pci/htirq.c | 72 +++++++++++++++----------------------------- include/linux/htirq.h | 11 ++++--- 4 files changed, 58 insertions(+), 82 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 507983c513c..ad84bc2802a 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -2624,18 +2624,16 @@ void arch_teardown_msi_irq(unsigned int irq) static void target_ht_irq(unsigned int irq, unsigned int dest) { - u32 low, high; - low = read_ht_irq_low(irq); - high = read_ht_irq_high(irq); + struct ht_irq_msg msg; + fetch_ht_irq_msg(irq, &msg); - low &= ~(HT_IRQ_LOW_DEST_ID_MASK); - high &= ~(HT_IRQ_HIGH_DEST_ID_MASK); + msg.address_lo &= ~(HT_IRQ_LOW_DEST_ID_MASK); + msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); - low |= HT_IRQ_LOW_DEST_ID(dest); - high |= HT_IRQ_HIGH_DEST_ID(dest); + msg.address_lo |= HT_IRQ_LOW_DEST_ID(dest); + msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); } static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) @@ -2673,7 +2671,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) vector = assign_irq_vector(irq); if (vector >= 0) { - u32 low, high; + struct ht_irq_msg msg; unsigned dest; cpumask_t tmp; @@ -2681,9 +2679,10 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) cpu_set(vector >> 8, tmp); dest = cpu_mask_to_apicid(tmp); - high = HT_IRQ_HIGH_DEST_ID(dest); + msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); - low = HT_IRQ_LOW_BASE | + msg.address_lo = + HT_IRQ_LOW_BASE | HT_IRQ_LOW_DEST_ID(dest) | HT_IRQ_LOW_VECTOR(vector) | ((INT_DEST_MODE == 0) ? @@ -2695,8 +2694,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) HT_IRQ_LOW_MT_ARBITRATED) | HT_IRQ_LOW_IRQ_MASKED; - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); set_irq_chip_and_handler_name(irq, &ht_irq_chip, handle_edge_irq, "edge"); diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 3b8f9c68ad3..41bfc49301a 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -1955,18 +1955,16 @@ void arch_teardown_msi_irq(unsigned int irq) static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) { - u32 low, high; - low = read_ht_irq_low(irq); - high = read_ht_irq_high(irq); + struct ht_irq_msg msg; + fetch_ht_irq_msg(irq, &msg); - low &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK); - high &= ~(HT_IRQ_HIGH_DEST_ID_MASK); + msg.address_lo &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK); + msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); - low |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest); - high |= HT_IRQ_HIGH_DEST_ID(dest); + msg.address_lo |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest); + msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); } static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) @@ -1987,7 +1985,7 @@ static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) dest = cpu_mask_to_apicid(tmp); - target_ht_irq(irq, dest, vector & 0xff); + target_ht_irq(irq, dest, vector); set_native_irq_info(irq, mask); } #endif @@ -2010,14 +2008,15 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) vector = assign_irq_vector(irq, TARGET_CPUS, &tmp); if (vector >= 0) { - u32 low, high; + struct ht_irq_msg msg; unsigned dest; dest = cpu_mask_to_apicid(tmp); - high = HT_IRQ_HIGH_DEST_ID(dest); + msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); - low = HT_IRQ_LOW_BASE | + msg.address_lo = + HT_IRQ_LOW_BASE | HT_IRQ_LOW_DEST_ID(dest) | HT_IRQ_LOW_VECTOR(vector) | ((INT_DEST_MODE == 0) ? @@ -2026,10 +2025,10 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) HT_IRQ_LOW_RQEOI_EDGE | ((INT_DELIVERY_MODE != dest_LowestPrio) ? HT_IRQ_LOW_MT_FIXED : - HT_IRQ_LOW_MT_ARBITRATED); + HT_IRQ_LOW_MT_ARBITRATED) | + HT_IRQ_LOW_IRQ_MASKED; - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); set_irq_chip_and_handler_name(irq, &ht_irq_chip, handle_edge_irq, "edge"); diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c index 0e27f2404a8..e346fe31f97 100644 --- a/drivers/pci/htirq.c +++ b/drivers/pci/htirq.c @@ -27,82 +27,55 @@ struct ht_irq_cfg { struct pci_dev *dev; unsigned pos; unsigned idx; + struct ht_irq_msg msg; }; -void write_ht_irq_low(unsigned int irq, u32 data) -{ - struct ht_irq_cfg *cfg = get_irq_data(irq); - unsigned long flags; - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); -} - -void write_ht_irq_high(unsigned int irq, u32 data) -{ - struct ht_irq_cfg *cfg = get_irq_data(irq); - unsigned long flags; - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); -} -u32 read_ht_irq_low(unsigned int irq) +void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg) { struct ht_irq_cfg *cfg = get_irq_data(irq); unsigned long flags; - u32 data; spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); + if (cfg->msg.address_lo != msg->address_lo) { + pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); + pci_write_config_dword(cfg->dev, cfg->pos + 4, msg->address_lo); + } + if (cfg->msg.address_hi != msg->address_hi) { + pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); + pci_write_config_dword(cfg->dev, cfg->pos + 4, msg->address_hi); + } spin_unlock_irqrestore(&ht_irq_lock, flags); - return data; + cfg->msg = *msg; } -u32 read_ht_irq_high(unsigned int irq) +void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg) { struct ht_irq_cfg *cfg = get_irq_data(irq); - unsigned long flags; - u32 data; - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); - spin_unlock_irqrestore(&ht_irq_lock, flags); - return data; + *msg = cfg->msg; } void mask_ht_irq(unsigned int irq) { struct ht_irq_cfg *cfg; - unsigned long flags; - u32 data; + struct ht_irq_msg msg; cfg = get_irq_data(irq); - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); - data |= 1; - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); + msg = cfg->msg; + msg.address_lo |= 1; + write_ht_irq_msg(irq, &msg); } void unmask_ht_irq(unsigned int irq) { struct ht_irq_cfg *cfg; - unsigned long flags; - u32 data; + struct ht_irq_msg msg; cfg = get_irq_data(irq); - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); - data &= ~1; - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); + msg = cfg->msg; + msg.address_lo &= ~1; + write_ht_irq_msg(irq, &msg); } /** @@ -152,6 +125,9 @@ int ht_create_irq(struct pci_dev *dev, int idx) cfg->dev = dev; cfg->pos = pos; cfg->idx = 0x10 + (idx * 2); + /* Initialize msg to a value that will never match the first write. */ + cfg->msg.address_lo = 0xffffffff; + cfg->msg.address_hi = 0xffffffff; irq = create_irq(); if (irq < 0) { diff --git a/include/linux/htirq.h b/include/linux/htirq.h index 1f15ce279a2..108f0d91e11 100644 --- a/include/linux/htirq.h +++ b/include/linux/htirq.h @@ -1,11 +1,14 @@ #ifndef LINUX_HTIRQ_H #define LINUX_HTIRQ_H +struct ht_irq_msg { + u32 address_lo; /* low 32 bits of the ht irq message */ + u32 address_hi; /* high 32 bits of the it irq message */ +}; + /* Helper functions.. */ -void write_ht_irq_low(unsigned int irq, u32 data); -void write_ht_irq_high(unsigned int irq, u32 data); -u32 read_ht_irq_low(unsigned int irq); -u32 read_ht_irq_high(unsigned int irq); +void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg); +void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg); void mask_ht_irq(unsigned int irq); void unmask_ht_irq(unsigned int irq); -- cgit v1.2.3-70-g09d2 From fa18f477d0987c011cce047a7c3cd1284f547a14 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86: Add acpi_user_timer_override option for Asus boards Timer overrides are normally disabled on Nvidia board because they are commonly wrong, except on new ones with HPET support. Unfortunately there are quite some Asus boards around that don't have HPET, but need a timer override. We don't know yet how to handle this transparently, but at least add a command line option to force the timer override and let them boot. Cc: len.brown@intel.com Signed-off-by: Andi Kleen --- Documentation/kernel-parameters.txt | 4 ++++ arch/i386/kernel/acpi/boot.c | 8 ++++++++ arch/i386/kernel/acpi/earlyquirk.c | 8 +++++++- arch/x86_64/kernel/early-quirks.c | 8 ++++++++ include/asm-i386/acpi.h | 1 + include/asm-x86_64/acpi.h | 1 + 6 files changed, 29 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index dd00fd556a6..67473849f20 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -164,6 +164,10 @@ and is between 256 and 4096 characters. It is defined in the file acpi_skip_timer_override [HW,ACPI] Recognize and ignore IRQ0/pin2 Interrupt Override. For broken nForce2 BIOS resulting in XT-PIC timer. + acpi_use_timer_override [HW,ACPI} + Use timer override. For some broken Nvidia NF5 boards + that require a timer override, but don't have + HPET acpi_dbg_layer= [HW,ACPI] Format: diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c index 22e4c466e5a..d12fb97a533 100644 --- a/arch/i386/kernel/acpi/boot.c +++ b/arch/i386/kernel/acpi/boot.c @@ -82,6 +82,7 @@ EXPORT_SYMBOL(acpi_strict); acpi_interrupt_flags acpi_sci_flags __initdata; int acpi_sci_override_gsi __initdata; int acpi_skip_timer_override __initdata; +int acpi_use_timer_override __initdata; #ifdef CONFIG_X86_LOCAL_APIC static u64 acpi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE; @@ -1300,6 +1301,13 @@ static int __init parse_acpi_skip_timer_override(char *arg) return 0; } early_param("acpi_skip_timer_override", parse_acpi_skip_timer_override); + +static int __init parse_acpi_use_timer_override(char *arg) +{ + acpi_use_timer_override = 1; + return 0; +} +early_param("acpi_use_timer_override", parse_acpi_use_timer_override); #endif /* CONFIG_X86_IO_APIC */ static int __init setup_acpi_sci(char *s) diff --git a/arch/i386/kernel/acpi/earlyquirk.c b/arch/i386/kernel/acpi/earlyquirk.c index fe799b11ac0..c9841692bb7 100644 --- a/arch/i386/kernel/acpi/earlyquirk.c +++ b/arch/i386/kernel/acpi/earlyquirk.c @@ -27,11 +27,17 @@ static int __init check_bridge(int vendor, int device) #ifdef CONFIG_ACPI /* According to Nvidia all timer overrides are bogus unless HPET is enabled. */ - if (vendor == PCI_VENDOR_ID_NVIDIA) { + if (!acpi_use_timer_override && vendor == PCI_VENDOR_ID_NVIDIA) { nvidia_hpet_detected = 0; acpi_table_parse(ACPI_HPET, nvidia_hpet_check); if (nvidia_hpet_detected == 0) { acpi_skip_timer_override = 1; + printk(KERN_INFO "Nvidia board " + "detected. Ignoring ACPI " + "timer override.\n"); + printk(KERN_INFO "If you got timer trouble " + "try acpi_use_timer_override\n"); + } } #endif diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c index 2b1245d8625..68273bff58c 100644 --- a/arch/x86_64/kernel/early-quirks.c +++ b/arch/x86_64/kernel/early-quirks.c @@ -45,7 +45,13 @@ static void nvidia_bugs(void) /* * All timer overrides on Nvidia are * wrong unless HPET is enabled. + * Unfortunately that's not true on many Asus boards. + * We don't know yet how to detect this automatically, but + * at least allow a command line override. */ + if (acpi_use_timer_override) + return; + nvidia_hpet_detected = 0; acpi_table_parse(ACPI_HPET, nvidia_hpet_check); if (nvidia_hpet_detected == 0) { @@ -53,6 +59,8 @@ static void nvidia_bugs(void) printk(KERN_INFO "Nvidia board " "detected. Ignoring ACPI " "timer override.\n"); + printk(KERN_INFO "If you got timer trouble " + "try acpi_use_timer_override\n"); } #endif /* RED-PEN skip them on mptables too? */ diff --git a/include/asm-i386/acpi.h b/include/asm-i386/acpi.h index 6016632d032..c80b3a94511 100644 --- a/include/asm-i386/acpi.h +++ b/include/asm-i386/acpi.h @@ -132,6 +132,7 @@ extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq); #ifdef CONFIG_X86_IO_APIC extern int acpi_skip_timer_override; +extern int acpi_use_timer_override; #endif static inline void acpi_noirq_set(void) { acpi_noirq = 1; } diff --git a/include/asm-x86_64/acpi.h b/include/asm-x86_64/acpi.h index ed59aa4c6ff..9d1916e59c0 100644 --- a/include/asm-x86_64/acpi.h +++ b/include/asm-x86_64/acpi.h @@ -163,6 +163,7 @@ extern u8 x86_acpiid_to_apicid[]; #define ARCH_HAS_POWER_INIT 1 extern int acpi_skip_timer_override; +extern int acpi_use_timer_override; #endif /*__KERNEL__*/ -- cgit v1.2.3-70-g09d2 From f3ac84324fd949f671e6cf5620add5de02498386 Mon Sep 17 00:00:00 2001 From: Daniel Ritz Date: Tue, 14 Nov 2006 02:03:25 -0800 Subject: [PATCH] fix via586 irq routing for pirq 5 Fix interrupt routing for via 586 bridges. pirq can be 5 which needs to be mapped to INTD. But currently the access functions can handle only pirq 1-4. this is similar to the other via chipsets where pirq 4 and 5 are both mapped to INTD. Fixes bugzilla #7490 Cc: Daniel Paschka Cc: Adrian Bunk Signed-off-by: Daniel Ritz Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/pci/irq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c index dbc4aae9195..69163998ade 100644 --- a/arch/i386/pci/irq.c +++ b/arch/i386/pci/irq.c @@ -255,13 +255,13 @@ static int pirq_via_set(struct pci_dev *router, struct pci_dev *dev, int pirq, i */ static int pirq_via586_get(struct pci_dev *router, struct pci_dev *dev, int pirq) { - static const unsigned int pirqmap[4] = { 3, 2, 5, 1 }; + static const unsigned int pirqmap[5] = { 3, 2, 5, 1, 1 }; return read_config_nybble(router, 0x55, pirqmap[pirq-1]); } static int pirq_via586_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq) { - static const unsigned int pirqmap[4] = { 3, 2, 5, 1 }; + static const unsigned int pirqmap[5] = { 3, 2, 5, 1, 1 }; write_config_nybble(router, 0x55, pirqmap[pirq-1], irq); return 1; } -- cgit v1.2.3-70-g09d2 From 45c99533252ef2297f37c5fdd672a3e0eb566870 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Tue, 14 Nov 2006 10:52:12 -0700 Subject: [PATCH] Use delayed disable mode of ioapic edge triggered interrupts Komuro reports that ISA interrupts do not work after a disable_irq(), causing some PCMCIA drivers to not work, with messages like eth0: Asix AX88190: io 0x300, irq 3, hw_addr xx:xx:xx:xx:xx:xx eth0: found link beat eth0: autonegotiation complete: 100baseT-FD selected eth0: interrupt(s) dropped! eth0: interrupt(s) dropped! eth0: interrupt(s) dropped! ... Linus Torvalds said: "Now, edge-triggered interrupts are a _lot_ harder to mask, because the Intel APIC is an unbelievable piece of sh*t, and has the edge-detect logic _before_ the mask logic, so if a edge happens _while_ the device is masked, you'll never ever see the edge ever again (unmasking will not cause a new edge, so you simply lost the interrupt). So when you "mask" an edge-triggered IRQ, you can't really mask it at all, because if you did that, you'd lose it forever if the IRQ comes in while you masked it. Instead, we're supposed to leave it active, and set a flag, and IF the IRQ comes in, we just remember it, and mask it at that point instead, and then on unmasking, we have to replay it by sending a self-IPI." This trivial patch solves the problem. Signed-off-by: Eric W. Biederman Cc: Ingo Molnar Acked-by: Komuro Signed-off-by: Linus Torvalds --- arch/i386/kernel/io_apic.c | 4 +++- arch/x86_64/kernel/io_apic.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index ad84bc2802a..3b7a63e0ed1 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -1287,9 +1287,11 @@ static void ioapic_register_intr(int irq, int vector, unsigned long trigger) trigger == IOAPIC_LEVEL) set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_fasteoi_irq, "fasteoi"); - else + else { + irq_desc[irq].status |= IRQ_DELAYED_DISABLE; set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_edge_irq, "edge"); + } set_intr_gate(vector, interrupt[irq]); } diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 41bfc49301a..14654e68241 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -790,9 +790,11 @@ static void ioapic_register_intr(int irq, int vector, unsigned long trigger) trigger == IOAPIC_LEVEL) set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_fasteoi_irq, "fasteoi"); - else + else { + irq_desc[irq].status |= IRQ_DELAYED_DISABLE; set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_edge_irq, "edge"); + } } static void __init setup_IO_APIC_irqs(void) -- cgit v1.2.3-70-g09d2 From dc1829a4c378d793fb3b95d56135d89a0d7ff72a Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 17 Nov 2006 14:26:18 +0100 Subject: [PATCH] i386/x86_64: ACPI cpu_idle_wait() fix The scheduler on Andreas Friedrich's hyperthreading system stopped working properly: the scheduler would never move tasks to another CPU! The lask known working kernel was 2.6.8. After a couple of attempts to corner the bug, the following smoking gun was found: BIOS reported wrong ACPI idfor the processor CPU#1: set_cpus_allowed(), swapper:1, 3 -> 2 [] show_trace_log_lvl+0x34/0x4a [] show_trace+0x2c/0x2e [] dump_stack+0x2b/0x2d [] set_cpus_allowed+0x52/0xec [] cpu_idle_wait+0x2e/0x100 [] acpi_processor_power_exit+0x45/0x58 [] acpi_processor_remove+0x46/0xea [] acpi_start_single_object+0x47/0x54 [] acpi_bus_register_driver+0xa4/0xd3 [] acpi_processor_init+0x57/0x77 [] init+0x146/0x2fd [] kernel_thread_helper+0x7/0x10 a quick look at cpu_idle_wait() shows how broken that code is on i386: it changes the init task's affinity map but never restores it ... and because all userspace tasks get forked by init, they all inherited that single-CPU affinity mask. x86_64 cloned this bug too. Signed-off-by: Ingo Molnar Cc: Andreas Friedrich Cc: Wolfgang Erig Cc: Andrew Morton Cc: Adrian Bunk Signed-off-by: Linus Torvalds --- arch/i386/kernel/process.c | 4 +++- arch/x86_64/kernel/process.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 1e1fa3e391a..dd53c58f64f 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -205,7 +205,7 @@ void cpu_idle(void) void cpu_idle_wait(void) { unsigned int cpu, this_cpu = get_cpu(); - cpumask_t map; + cpumask_t map, tmp = current->cpus_allowed; set_cpus_allowed(current, cpumask_of_cpu(this_cpu)); put_cpu(); @@ -227,6 +227,8 @@ void cpu_idle_wait(void) } cpus_and(map, map, cpu_online_map); } while (!cpus_empty(map)); + + set_cpus_allowed(current, tmp); } EXPORT_SYMBOL_GPL(cpu_idle_wait); diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index f6226055d53..7451a4c43c1 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -144,7 +144,7 @@ static void poll_idle (void) void cpu_idle_wait(void) { unsigned int cpu, this_cpu = get_cpu(); - cpumask_t map; + cpumask_t map, tmp = current->cpus_allowed; set_cpus_allowed(current, cpumask_of_cpu(this_cpu)); put_cpu(); @@ -167,6 +167,8 @@ void cpu_idle_wait(void) } cpus_and(map, map, cpu_online_map); } while (!cpus_empty(map)); + + set_cpus_allowed(current, tmp); } EXPORT_SYMBOL_GPL(cpu_idle_wait); -- cgit v1.2.3-70-g09d2 From 808dbbb6bb61173bf52946a28f99089d2efa4c55 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 17 Nov 2006 11:14:56 -0800 Subject: x86: be more careful when walking back the frame pointer chain When showing the stack backtrace, make sure that we never accept not only an unchanging frame pointer, but also a frame pointer that moves back down the stack frame. It must always grow up (toward older stack frames). I doubt this has triggered, but a subtly corrupt stack with extremely unlucky contents could cause us to loop forever on a bogus endless frame pointer chain. This review was triggered by much worse problems happening in some of the other stack unwinding code. Signed-off-by: Linus Torvalds --- arch/i386/kernel/traps.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 00489b706d2..fe9c5e8e7e6 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -129,15 +129,19 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo, #ifdef CONFIG_FRAME_POINTER while (valid_stack_ptr(tinfo, (void *)ebp)) { + unsigned long new_ebp; addr = *(unsigned long *)(ebp + 4); ops->address(data, addr); /* * break out of recursive entries (such as - * end_of_stack_stop_unwind_function): + * end_of_stack_stop_unwind_function). Also, + * we can never allow a frame pointer to + * move downwards! */ - if (ebp == *(unsigned long *)ebp) + new_ebp = *(unsigned long *)ebp; + if (new_ebp <= ebp) break; - ebp = *(unsigned long *)ebp; + ebp = new_ebp; } #else while (valid_stack_ptr(tinfo, stack)) { -- cgit v1.2.3-70-g09d2 From c4028958b6ecad064b1a6303a6a5906d4fe48d73 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 22 Nov 2006 14:57:56 +0000 Subject: WorkStruct: make allyesconfig Fix up for make allyesconfig. Signed-Off-By: David Howells --- arch/i386/kernel/cpu/mcheck/non-fatal.c | 6 +- arch/i386/kernel/smpboot.c | 11 +- arch/i386/kernel/tsc.c | 4 +- arch/powerpc/platforms/pseries/eeh_event.c | 6 +- drivers/atm/idt77252.c | 9 +- drivers/block/aoe/aoe.h | 2 +- drivers/block/aoe/aoecmd.c | 4 +- drivers/block/aoe/aoedev.c | 2 +- drivers/block/paride/pd.c | 8 +- drivers/block/paride/pseudo.h | 10 +- drivers/block/sx8.c | 7 +- drivers/block/ub.c | 8 +- drivers/bluetooth/bcm203x.c | 7 +- drivers/char/cyclades.c | 9 +- drivers/char/drm/via_dmablit.c | 6 +- drivers/char/epca.c | 8 +- drivers/char/esp.c | 14 +- drivers/char/genrtc.c | 4 +- drivers/char/hvsi.c | 16 +- drivers/char/ip2/i2lib.c | 12 +- drivers/char/ip2/ip2main.c | 23 +-- drivers/char/isicom.c | 12 +- drivers/char/moxa.c | 8 +- drivers/char/mxser.c | 9 +- drivers/char/pcmcia/synclink_cs.c | 8 +- drivers/char/sonypi.c | 4 +- drivers/char/specialix.c | 14 +- drivers/char/synclink.c | 9 +- drivers/char/synclink_gt.c | 10 +- drivers/char/synclinkmp.c | 8 +- drivers/char/tpm/tpm.c | 6 +- drivers/connector/cn_queue.c | 8 +- drivers/connector/connector.c | 31 ++-- drivers/cpufreq/cpufreq_conservative.c | 7 +- drivers/cpufreq/cpufreq_ondemand.c | 28 +-- drivers/i2c/chips/ds1374.c | 12 +- drivers/ieee1394/hosts.c | 9 +- drivers/ieee1394/hosts.h | 2 +- drivers/ieee1394/sbp2.c | 28 +-- drivers/ieee1394/sbp2.h | 2 +- drivers/infiniband/core/addr.c | 6 +- drivers/infiniband/core/cache.c | 7 +- drivers/infiniband/core/cm.c | 19 +- drivers/infiniband/core/cma.c | 10 +- drivers/infiniband/core/iwcm.c | 7 +- drivers/infiniband/core/mad.c | 25 +-- drivers/infiniband/core/mad_priv.h | 2 +- drivers/infiniband/core/mad_rmpp.c | 18 +- drivers/infiniband/core/sa_query.c | 10 +- drivers/infiniband/core/uverbs_mem.c | 7 +- drivers/infiniband/hw/ipath/ipath_user_pages.c | 7 +- drivers/infiniband/hw/mthca/mthca_catas.c | 4 +- drivers/infiniband/ulp/ipoib/ipoib.h | 16 +- drivers/infiniband/ulp/ipoib/ipoib_ib.c | 25 +-- drivers/infiniband/ulp/ipoib/ipoib_main.c | 10 +- drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 22 ++- drivers/infiniband/ulp/iser/iser_verbs.c | 10 +- drivers/infiniband/ulp/srp/ib_srp.c | 7 +- drivers/input/keyboard/lkkbd.c | 6 +- drivers/input/keyboard/sunkbd.c | 6 +- drivers/input/mouse/psmouse-base.c | 7 +- drivers/isdn/act2000/capi.c | 4 +- drivers/isdn/act2000/capi.h | 2 +- drivers/isdn/act2000/module.c | 18 +- drivers/isdn/capi/kcapi.c | 14 +- drivers/isdn/hisax/amd7930_fn.c | 7 +- drivers/isdn/hisax/config.c | 9 +- drivers/isdn/hisax/hfc4s8s_l1.c | 5 +- drivers/isdn/hisax/hfc_2bds0.c | 9 +- drivers/isdn/hisax/hfc_pci.c | 6 +- drivers/isdn/hisax/hfc_sx.c | 6 +- drivers/isdn/hisax/icc.c | 6 +- drivers/isdn/hisax/isac.c | 6 +- drivers/isdn/hisax/isar.c | 6 +- drivers/isdn/hisax/isdnl1.c | 6 +- drivers/isdn/hisax/w6692.c | 6 +- drivers/isdn/i4l/isdn_net.c | 6 +- drivers/isdn/pcbit/drv.c | 4 +- drivers/isdn/pcbit/layer2.c | 6 +- drivers/isdn/pcbit/pcbit.h | 2 + drivers/macintosh/smu.c | 4 +- drivers/md/dm-crypt.c | 8 +- drivers/md/dm-mpath.c | 18 +- drivers/md/dm-raid1.c | 4 +- drivers/md/dm-snap.c | 9 +- drivers/md/kcopyd.c | 4 +- drivers/media/dvb/b2c2/flexcop-pci.c | 9 +- drivers/media/dvb/cinergyT2/cinergyT2.c | 18 +- drivers/media/dvb/dvb-core/dvb_net.c | 19 +- drivers/media/dvb/dvb-usb/dvb-usb-remote.c | 7 +- drivers/media/dvb/dvb-usb/dvb-usb.h | 2 +- drivers/media/video/cpia_pp.c | 20 ++- drivers/media/video/cx88/cx88-input.c | 6 +- drivers/media/video/ir-kbd-i2c.c | 6 +- drivers/media/video/pvrusb2/pvrusb2-context.c | 13 +- drivers/media/video/saa6588.c | 6 +- drivers/media/video/saa7134/saa7134-empress.c | 9 +- drivers/message/fusion/mptfc.c | 14 +- drivers/message/fusion/mptlan.c | 29 +-- drivers/message/fusion/mptsas.c | 25 +-- drivers/message/fusion/mptspi.c | 14 +- drivers/message/i2o/driver.c | 2 +- drivers/message/i2o/exec-osm.c | 13 +- drivers/message/i2o/i2o_block.c | 15 +- drivers/message/i2o/i2o_block.h | 2 +- drivers/misc/tifm_7xx1.c | 18 +- drivers/mmc/mmc.c | 14 +- drivers/mmc/mmc.h | 2 +- drivers/mmc/mmc_sysfs.c | 10 +- drivers/mmc/tifm_sd.c | 28 +-- drivers/net/8139too.c | 26 +-- drivers/net/bnx2.c | 6 +- drivers/net/cassini.c | 6 +- drivers/net/chelsio/common.h | 2 +- drivers/net/chelsio/cxgb2.c | 16 +- drivers/net/e100.c | 8 +- drivers/net/ehea/ehea_main.c | 9 +- drivers/net/hamradio/baycom_epp.c | 14 +- drivers/net/irda/mcs7780.c | 6 +- drivers/net/irda/sir-dev.h | 2 +- drivers/net/irda/sir_dev.c | 8 +- drivers/net/iseries_veth.c | 12 +- drivers/net/ixgb/ixgb_main.c | 10 +- drivers/net/myri10ge/myri10ge.c | 7 +- drivers/net/ns83820.c | 10 +- drivers/net/pcmcia/xirc2ps_cs.c | 12 +- drivers/net/phy/phy.c | 9 +- drivers/net/plip.c | 38 ++-- drivers/net/qla3xxx.c | 20 ++- drivers/net/qla3xxx.h | 4 +- drivers/net/r8169.c | 23 ++- drivers/net/s2io.c | 16 +- drivers/net/s2io.h | 2 +- drivers/net/sis190.c | 13 +- drivers/net/skge.c | 15 +- drivers/net/skge.h | 2 +- drivers/net/spider_net.c | 9 +- drivers/net/sungem.c | 6 +- drivers/net/tg3.c | 6 +- drivers/net/tlan.c | 23 ++- drivers/net/tlan.h | 1 + drivers/net/tulip/21142.c | 7 +- drivers/net/tulip/timer.c | 7 +- drivers/net/tulip/tulip.h | 7 +- drivers/net/tulip/tulip_core.c | 3 +- drivers/net/wan/pc300_tty.c | 23 +-- drivers/net/wireless/bcm43xx/bcm43xx.h | 2 +- drivers/net/wireless/bcm43xx/bcm43xx_main.c | 20 ++- drivers/net/wireless/hostap/hostap.h | 2 +- drivers/net/wireless/hostap/hostap_ap.c | 19 +- drivers/net/wireless/hostap/hostap_hw.c | 21 +-- drivers/net/wireless/hostap/hostap_info.c | 6 +- drivers/net/wireless/hostap/hostap_main.c | 8 +- drivers/net/wireless/ipw2100.c | 47 ++--- drivers/net/wireless/ipw2100.h | 10 +- drivers/net/wireless/ipw2200.c | 227 +++++++++++++----------- drivers/net/wireless/ipw2200.h | 16 +- drivers/net/wireless/orinoco.c | 28 +-- drivers/net/wireless/prism54/isl_ioctl.c | 8 +- drivers/net/wireless/prism54/isl_ioctl.h | 4 +- drivers/net/wireless/prism54/islpci_dev.c | 5 +- drivers/net/wireless/prism54/islpci_eth.c | 4 +- drivers/net/wireless/prism54/islpci_eth.h | 2 +- drivers/net/wireless/prism54/islpci_mgt.c | 2 +- drivers/net/wireless/zd1211rw/zd_mac.c | 7 +- drivers/net/wireless/zd1211rw/zd_mac.h | 2 +- drivers/oprofile/cpu_buffer.c | 9 +- drivers/oprofile/cpu_buffer.h | 2 +- drivers/pci/hotplug/shpchp.h | 4 +- drivers/pci/hotplug/shpchp_core.c | 2 +- drivers/pci/hotplug/shpchp_ctrl.c | 19 +- drivers/pcmcia/ds.c | 7 +- drivers/rtc/rtc-dev.c | 7 +- drivers/scsi/NCR5380.c | 11 +- drivers/scsi/NCR5380.h | 4 +- drivers/scsi/aha152x.c | 4 +- drivers/scsi/imm.c | 12 +- drivers/scsi/ipr.c | 9 +- drivers/scsi/libiscsi.c | 7 +- drivers/scsi/libsas/sas_discover.c | 22 ++- drivers/scsi/libsas/sas_event.c | 14 +- drivers/scsi/libsas/sas_init.c | 6 +- drivers/scsi/libsas/sas_internal.h | 12 +- drivers/scsi/libsas/sas_phy.c | 45 +++-- drivers/scsi/libsas/sas_port.c | 30 ++-- drivers/scsi/ppa.c | 12 +- drivers/scsi/qla4xxx/ql4_os.c | 7 +- drivers/scsi/scsi_transport_fc.c | 60 ++++--- drivers/scsi/scsi_transport_iscsi.c | 8 +- drivers/scsi/scsi_transport_spi.c | 7 +- drivers/spi/spi_bitbang.c | 7 +- drivers/usb/atm/cxacru.c | 12 +- drivers/usb/atm/speedtch.c | 15 +- drivers/usb/atm/ueagle-atm.c | 6 +- drivers/usb/class/cdc-acm.c | 6 +- drivers/usb/core/hub.c | 18 +- drivers/usb/core/hub.h | 2 +- drivers/usb/core/message.c | 7 +- drivers/usb/core/usb.c | 9 +- drivers/usb/gadget/ether.c | 6 +- drivers/usb/host/u132-hcd.c | 62 +++---- drivers/usb/input/hid-core.c | 7 +- drivers/usb/misc/ftdi-elan.c | 86 +++------ drivers/usb/misc/phidgetkit.c | 21 ++- drivers/usb/misc/phidgetmotorcontrol.c | 11 +- drivers/usb/net/kaweth.c | 9 +- drivers/usb/net/pegasus.c | 6 +- drivers/usb/net/pegasus.h | 2 +- drivers/usb/net/usbnet.c | 7 +- drivers/usb/serial/aircable.c | 13 +- drivers/usb/serial/digi_acceleport.c | 14 +- drivers/usb/serial/ftdi_sio.c | 19 +- drivers/usb/serial/keyspan_pda.c | 22 ++- drivers/usb/serial/usb-serial.c | 7 +- drivers/usb/serial/whiteheat.c | 15 +- drivers/video/console/fbcon.c | 6 +- fs/9p/mux.c | 16 +- fs/gfs2/glock.c | 8 +- fs/ncpfs/inode.c | 8 +- fs/ncpfs/sock.c | 20 ++- fs/nfsd/nfs4state.c | 7 +- fs/ocfs2/alloc.c | 9 +- fs/ocfs2/cluster/heartbeat.c | 10 +- fs/ocfs2/cluster/quorum.c | 4 +- fs/ocfs2/cluster/tcp.c | 78 ++++---- fs/ocfs2/cluster/tcp_internal.h | 8 +- fs/ocfs2/dlm/dlmcommon.h | 2 +- fs/ocfs2/dlm/dlmdomain.c | 2 +- fs/ocfs2/dlm/dlmrecovery.c | 5 +- fs/ocfs2/dlm/userdlm.c | 10 +- fs/ocfs2/journal.c | 7 +- fs/ocfs2/journal.h | 2 +- fs/ocfs2/ocfs2.h | 2 +- fs/ocfs2/super.c | 2 +- fs/reiserfs/journal.c | 12 +- fs/xfs/linux-2.6/xfs_aops.c | 21 ++- fs/xfs/linux-2.6/xfs_buf.c | 9 +- include/linux/connector.h | 4 +- include/linux/i2o.h | 2 +- include/linux/mmc/host.h | 2 +- include/linux/ncp_fs_sb.h | 8 +- include/linux/reiserfs_fs_sb.h | 3 +- include/linux/relay.h | 2 +- include/linux/usb.h | 2 +- include/net/ieee80211softmac.h | 4 +- include/net/sctp/structs.h | 2 +- include/scsi/libsas.h | 23 ++- include/scsi/scsi_transport_fc.h | 4 +- include/scsi/scsi_transport_iscsi.h | 2 +- include/sound/ac97_codec.h | 2 +- include/sound/ak4114.h | 2 +- kernel/relay.c | 10 +- mm/swap.c | 4 +- net/atm/lec.c | 9 +- net/atm/lec.h | 2 +- net/bluetooth/hci_sysfs.c | 12 +- net/bridge/br_if.c | 10 +- net/bridge/br_private.h | 2 +- net/core/netpoll.c | 4 +- net/dccp/minisocks.c | 3 +- net/ieee80211/softmac/ieee80211softmac_assoc.c | 18 +- net/ieee80211/softmac/ieee80211softmac_auth.c | 23 +-- net/ieee80211/softmac/ieee80211softmac_event.c | 12 +- net/ieee80211/softmac/ieee80211softmac_module.c | 4 +- net/ieee80211/softmac/ieee80211softmac_priv.h | 13 +- net/ieee80211/softmac/ieee80211softmac_scan.c | 13 +- net/ieee80211/softmac/ieee80211softmac_wx.c | 6 +- net/ipv4/ipvs/ip_vs_ctl.c | 6 +- net/irda/ircomm/ircomm_tty.c | 11 +- net/sctp/associola.c | 11 +- net/sctp/endpointola.c | 10 +- net/sctp/inqueue.c | 9 +- net/xfrm/xfrm_policy.c | 8 +- net/xfrm/xfrm_state.c | 8 +- sound/aoa/aoa-gpio.h | 2 +- sound/aoa/core/snd-aoa-gpio-feature.c | 16 +- sound/aoa/core/snd-aoa-gpio-pmf.c | 16 +- sound/i2c/other/ak4114.c | 8 +- sound/pci/ac97/ac97_codec.c | 7 +- sound/pci/hda/hda_codec.c | 10 +- sound/pci/hda/hda_local.h | 1 + sound/ppc/tumbler.c | 8 +- 282 files changed, 1771 insertions(+), 1450 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/mcheck/non-fatal.c b/arch/i386/kernel/cpu/mcheck/non-fatal.c index 1f9153ae5b0..6b5d3518a1c 100644 --- a/arch/i386/kernel/cpu/mcheck/non-fatal.c +++ b/arch/i386/kernel/cpu/mcheck/non-fatal.c @@ -51,10 +51,10 @@ static void mce_checkregs (void *info) } } -static void mce_work_fn(void *data); -static DECLARE_WORK(mce_work, mce_work_fn, NULL); +static void mce_work_fn(struct work_struct *work); +static DECLARE_DELAYED_WORK(mce_work, mce_work_fn); -static void mce_work_fn(void *data) +static void mce_work_fn(struct work_struct *work) { on_each_cpu(mce_checkregs, NULL, 1, 1); schedule_delayed_work(&mce_work, MCE_RATE); diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 4bb8b77cd65..02a9b66b6ac 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -1049,13 +1049,15 @@ void cpu_exit_clear(void) struct warm_boot_cpu_info { struct completion *complete; + struct work_struct task; int apicid; int cpu; }; -static void __cpuinit do_warm_boot_cpu(void *p) +static void __cpuinit do_warm_boot_cpu(struct work_struct *work) { - struct warm_boot_cpu_info *info = p; + struct warm_boot_cpu_info *info = + container_of(work, struct warm_boot_cpu_info, task); do_boot_cpu(info->apicid, info->cpu); complete(info->complete); } @@ -1064,7 +1066,6 @@ static int __cpuinit __smp_prepare_cpu(int cpu) { DECLARE_COMPLETION_ONSTACK(done); struct warm_boot_cpu_info info; - struct work_struct task; int apicid, ret; struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); @@ -1089,7 +1090,7 @@ static int __cpuinit __smp_prepare_cpu(int cpu) info.complete = &done; info.apicid = apicid; info.cpu = cpu; - INIT_WORK(&task, do_warm_boot_cpu, &info); + INIT_WORK(&info.task, do_warm_boot_cpu); tsc_sync_disabled = 1; @@ -1097,7 +1098,7 @@ static int __cpuinit __smp_prepare_cpu(int cpu) clone_pgd_range(swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS, KERNEL_PGD_PTRS); flush_tlb_all(); - schedule_work(&task); + schedule_work(&info.task); wait_for_completion(&done); tsc_sync_disabled = 0; diff --git a/arch/i386/kernel/tsc.c b/arch/i386/kernel/tsc.c index fbc95828cd7..9810c8c9075 100644 --- a/arch/i386/kernel/tsc.c +++ b/arch/i386/kernel/tsc.c @@ -217,7 +217,7 @@ static unsigned int cpufreq_delayed_issched = 0; static unsigned int cpufreq_init = 0; static struct work_struct cpufreq_delayed_get_work; -static void handle_cpufreq_delayed_get(void *v) +static void handle_cpufreq_delayed_get(struct work_struct *work) { unsigned int cpu; @@ -306,7 +306,7 @@ static int __init cpufreq_tsc(void) { int ret; - INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get, NULL); + INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get); ret = cpufreq_register_notifier(&time_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER); if (!ret) diff --git a/arch/powerpc/platforms/pseries/eeh_event.c b/arch/powerpc/platforms/pseries/eeh_event.c index 13707745131..49037edf7d3 100644 --- a/arch/powerpc/platforms/pseries/eeh_event.c +++ b/arch/powerpc/platforms/pseries/eeh_event.c @@ -37,8 +37,8 @@ /* EEH event workqueue setup. */ static DEFINE_SPINLOCK(eeh_eventlist_lock); LIST_HEAD(eeh_eventlist); -static void eeh_thread_launcher(void *); -DECLARE_WORK(eeh_event_wq, eeh_thread_launcher, NULL); +static void eeh_thread_launcher(struct work_struct *); +DECLARE_WORK(eeh_event_wq, eeh_thread_launcher); /* Serialize reset sequences for a given pci device */ DEFINE_MUTEX(eeh_event_mutex); @@ -103,7 +103,7 @@ static int eeh_event_handler(void * dummy) * eeh_thread_launcher * @dummy - unused */ -static void eeh_thread_launcher(void *dummy) +static void eeh_thread_launcher(struct work_struct *dummy) { if (kernel_thread(eeh_event_handler, NULL, CLONE_KERNEL) < 0) printk(KERN_ERR "Failed to start EEH daemon\n"); diff --git a/drivers/atm/idt77252.c b/drivers/atm/idt77252.c index 87b17c33b3f..f4078612194 100644 --- a/drivers/atm/idt77252.c +++ b/drivers/atm/idt77252.c @@ -135,7 +135,7 @@ static int idt77252_change_qos(struct atm_vcc *vcc, struct atm_qos *qos, int flags); static int idt77252_proc_read(struct atm_dev *dev, loff_t * pos, char *page); -static void idt77252_softint(void *dev_id); +static void idt77252_softint(struct work_struct *work); static struct atmdev_ops idt77252_ops = @@ -2866,9 +2866,10 @@ out: } static void -idt77252_softint(void *dev_id) +idt77252_softint(struct work_struct *work) { - struct idt77252_dev *card = dev_id; + struct idt77252_dev *card = + container_of(work, struct idt77252_dev, tqueue); u32 stat; int done; @@ -3697,7 +3698,7 @@ idt77252_init_one(struct pci_dev *pcidev, const struct pci_device_id *id) card->pcidev = pcidev; sprintf(card->name, "idt77252-%d", card->index); - INIT_WORK(&card->tqueue, idt77252_softint, (void *)card); + INIT_WORK(&card->tqueue, idt77252_softint); membase = pci_resource_start(pcidev, 1); srambase = pci_resource_start(pcidev, 2); diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index 6d111228cfa..2308e83e5f3 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -159,7 +159,7 @@ void aoecmd_work(struct aoedev *d); void aoecmd_cfg(ushort aoemajor, unsigned char aoeminor); void aoecmd_ata_rsp(struct sk_buff *); void aoecmd_cfg_rsp(struct sk_buff *); -void aoecmd_sleepwork(void *vp); +void aoecmd_sleepwork(struct work_struct *); struct sk_buff *new_skb(ulong); int aoedev_init(void); diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 8a13b1af8ba..97f7f535f41 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -408,9 +408,9 @@ rexmit_timer(ulong vp) /* this function performs work that has been deferred until sleeping is OK */ void -aoecmd_sleepwork(void *vp) +aoecmd_sleepwork(struct work_struct *work) { - struct aoedev *d = (struct aoedev *) vp; + struct aoedev *d = container_of(work, struct aoedev, work); if (d->flags & DEVFL_GDALLOC) aoeblk_gdalloc(d); diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index 6125921bbec..05a97197c91 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -88,7 +88,7 @@ aoedev_newdev(ulong nframes) kfree(d); return NULL; } - INIT_WORK(&d->work, aoecmd_sleepwork, d); + INIT_WORK(&d->work, aoecmd_sleepwork); spin_lock_init(&d->lock); init_timer(&d->timer); d->timer.data = (ulong) d; diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c index 40a11e56797..9d9bff23f42 100644 --- a/drivers/block/paride/pd.c +++ b/drivers/block/paride/pd.c @@ -352,19 +352,19 @@ static enum action (*phase)(void); static void run_fsm(void); -static void ps_tq_int( void *data); +static void ps_tq_int(struct work_struct *work); -static DECLARE_WORK(fsm_tq, ps_tq_int, NULL); +static DECLARE_DELAYED_WORK(fsm_tq, ps_tq_int); static void schedule_fsm(void) { if (!nice) - schedule_work(&fsm_tq); + schedule_delayed_work(&fsm_tq, 0); else schedule_delayed_work(&fsm_tq, nice-1); } -static void ps_tq_int(void *data) +static void ps_tq_int(struct work_struct *work) { run_fsm(); } diff --git a/drivers/block/paride/pseudo.h b/drivers/block/paride/pseudo.h index 932342d7a8e..bc370329414 100644 --- a/drivers/block/paride/pseudo.h +++ b/drivers/block/paride/pseudo.h @@ -35,7 +35,7 @@ #include #include -static void ps_tq_int( void *data); +static void ps_tq_int(struct work_struct *work); static void (* ps_continuation)(void); static int (* ps_ready)(void); @@ -45,7 +45,7 @@ static int ps_nice = 0; static DEFINE_SPINLOCK(ps_spinlock __attribute__((unused))); -static DECLARE_WORK(ps_tq, ps_tq_int, NULL); +static DECLARE_DELAYED_WORK(ps_tq, ps_tq_int); static void ps_set_intr(void (*continuation)(void), int (*ready)(void), @@ -63,14 +63,14 @@ static void ps_set_intr(void (*continuation)(void), if (!ps_tq_active) { ps_tq_active = 1; if (!ps_nice) - schedule_work(&ps_tq); + schedule_delayed_work(&ps_tq, 0); else schedule_delayed_work(&ps_tq, ps_nice-1); } spin_unlock_irqrestore(&ps_spinlock,flags); } -static void ps_tq_int(void *data) +static void ps_tq_int(struct work_struct *work) { void (*con)(void); unsigned long flags; @@ -92,7 +92,7 @@ static void ps_tq_int(void *data) } ps_tq_active = 1; if (!ps_nice) - schedule_work(&ps_tq); + schedule_delayed_work(&ps_tq, 0); else schedule_delayed_work(&ps_tq, ps_nice-1); spin_unlock_irqrestore(&ps_spinlock,flags); diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index 47d6975268f..54509eb3391 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -1244,9 +1244,10 @@ out: return IRQ_RETVAL(handled); } -static void carm_fsm_task (void *_data) +static void carm_fsm_task (struct work_struct *work) { - struct carm_host *host = _data; + struct carm_host *host = + container_of(work, struct carm_host, fsm_task); unsigned long flags; unsigned int state; int rc, i, next_dev; @@ -1619,7 +1620,7 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) host->pdev = pdev; host->flags = pci_dac ? FL_DAC : 0; spin_lock_init(&host->lock); - INIT_WORK(&host->fsm_task, carm_fsm_task, host); + INIT_WORK(&host->fsm_task, carm_fsm_task); init_completion(&host->probe_comp); for (i = 0; i < ARRAY_SIZE(host->req); i++) diff --git a/drivers/block/ub.c b/drivers/block/ub.c index 0d5c73f0726..2098eff91e1 100644 --- a/drivers/block/ub.c +++ b/drivers/block/ub.c @@ -376,7 +376,7 @@ static int ub_submit_clear_stall(struct ub_dev *sc, struct ub_scsi_cmd *cmd, int stalled_pipe); static void ub_top_sense_done(struct ub_dev *sc, struct ub_scsi_cmd *scmd); static void ub_reset_enter(struct ub_dev *sc, int try); -static void ub_reset_task(void *arg); +static void ub_reset_task(struct work_struct *work); static int ub_sync_tur(struct ub_dev *sc, struct ub_lun *lun); static int ub_sync_read_cap(struct ub_dev *sc, struct ub_lun *lun, struct ub_capacity *ret); @@ -1558,9 +1558,9 @@ static void ub_reset_enter(struct ub_dev *sc, int try) schedule_work(&sc->reset_work); } -static void ub_reset_task(void *arg) +static void ub_reset_task(struct work_struct *work) { - struct ub_dev *sc = arg; + struct ub_dev *sc = container_of(work, struct ub_dev, reset_work); unsigned long flags; struct list_head *p; struct ub_lun *lun; @@ -2179,7 +2179,7 @@ static int ub_probe(struct usb_interface *intf, usb_init_urb(&sc->work_urb); tasklet_init(&sc->tasklet, ub_scsi_action, (unsigned long)sc); atomic_set(&sc->poison, 0); - INIT_WORK(&sc->reset_work, ub_reset_task, sc); + INIT_WORK(&sc->reset_work, ub_reset_task); init_waitqueue_head(&sc->reset_wait); init_timer(&sc->work_timer); diff --git a/drivers/bluetooth/bcm203x.c b/drivers/bluetooth/bcm203x.c index 516751754aa..9256985cbe3 100644 --- a/drivers/bluetooth/bcm203x.c +++ b/drivers/bluetooth/bcm203x.c @@ -157,9 +157,10 @@ static void bcm203x_complete(struct urb *urb) } } -static void bcm203x_work(void *user_data) +static void bcm203x_work(struct work_struct *work) { - struct bcm203x_data *data = user_data; + struct bcm203x_data *data = + container_of(work, struct bcm203x_data, work); if (usb_submit_urb(data->urb, GFP_ATOMIC) < 0) BT_ERR("Can't submit URB"); @@ -246,7 +247,7 @@ static int bcm203x_probe(struct usb_interface *intf, const struct usb_device_id release_firmware(firmware); - INIT_WORK(&data->work, bcm203x_work, (void *) data); + INIT_WORK(&data->work, bcm203x_work); usb_set_intfdata(intf, data); diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c index e608dadece2..acb2de5e3a9 100644 --- a/drivers/char/cyclades.c +++ b/drivers/char/cyclades.c @@ -926,9 +926,10 @@ cy_sched_event(struct cyclades_port *info, int event) * had to poll every port to see if that port needed servicing. */ static void -do_softint(void *private_) +do_softint(struct work_struct *work) { - struct cyclades_port *info = (struct cyclades_port *) private_; + struct cyclades_port *info = + container_of(work, struct cyclades_port, tqueue); struct tty_struct *tty; tty = info->tty; @@ -5328,7 +5329,7 @@ cy_init(void) info->blocked_open = 0; info->default_threshold = 0; info->default_timeout = 0; - INIT_WORK(&info->tqueue, do_softint, info); + INIT_WORK(&info->tqueue, do_softint); init_waitqueue_head(&info->open_wait); init_waitqueue_head(&info->close_wait); init_waitqueue_head(&info->shutdown_wait); @@ -5403,7 +5404,7 @@ cy_init(void) info->blocked_open = 0; info->default_threshold = 0; info->default_timeout = 0; - INIT_WORK(&info->tqueue, do_softint, info); + INIT_WORK(&info->tqueue, do_softint); init_waitqueue_head(&info->open_wait); init_waitqueue_head(&info->close_wait); init_waitqueue_head(&info->shutdown_wait); diff --git a/drivers/char/drm/via_dmablit.c b/drivers/char/drm/via_dmablit.c index 60c1695db30..806f9ce5f47 100644 --- a/drivers/char/drm/via_dmablit.c +++ b/drivers/char/drm/via_dmablit.c @@ -500,9 +500,9 @@ via_dmablit_timer(unsigned long data) static void -via_dmablit_workqueue(void *data) +via_dmablit_workqueue(struct work_struct *work) { - drm_via_blitq_t *blitq = (drm_via_blitq_t *) data; + drm_via_blitq_t *blitq = container_of(work, drm_via_blitq_t, wq); drm_device_t *dev = blitq->dev; unsigned long irqsave; drm_via_sg_info_t *cur_sg; @@ -571,7 +571,7 @@ via_init_dmablit(drm_device_t *dev) DRM_INIT_WAITQUEUE(blitq->blit_queue + j); } DRM_INIT_WAITQUEUE(&blitq->busy_queue); - INIT_WORK(&blitq->wq, via_dmablit_workqueue, blitq); + INIT_WORK(&blitq->wq, via_dmablit_workqueue); init_timer(&blitq->poll_timer); blitq->poll_timer.function = &via_dmablit_timer; blitq->poll_timer.data = (unsigned long) blitq; diff --git a/drivers/char/epca.c b/drivers/char/epca.c index 706733c0b36..7c71eb77980 100644 --- a/drivers/char/epca.c +++ b/drivers/char/epca.c @@ -200,7 +200,7 @@ static int pc_ioctl(struct tty_struct *, struct file *, static int info_ioctl(struct tty_struct *, struct file *, unsigned int, unsigned long); static void pc_set_termios(struct tty_struct *, struct termios *); -static void do_softint(void *); +static void do_softint(struct work_struct *work); static void pc_stop(struct tty_struct *); static void pc_start(struct tty_struct *); static void pc_throttle(struct tty_struct * tty); @@ -1505,7 +1505,7 @@ static void post_fep_init(unsigned int crd) ch->brdchan = bc; ch->mailbox = gd; - INIT_WORK(&ch->tqueue, do_softint, ch); + INIT_WORK(&ch->tqueue, do_softint); ch->board = &boards[crd]; spin_lock_irqsave(&epca_lock, flags); @@ -2566,9 +2566,9 @@ static void pc_set_termios(struct tty_struct *tty, struct termios *old_termios) /* --------------------- Begin do_softint ----------------------- */ -static void do_softint(void *private_) +static void do_softint(struct work_struct *work) { /* Begin do_softint */ - struct channel *ch = (struct channel *) private_; + struct channel *ch = container_of(work, struct channel, tqueue); /* Called in response to a modem change event */ if (ch && ch->magic == EPCA_MAGIC) { /* Begin EPCA_MAGIC */ struct tty_struct *tty = ch->tty; diff --git a/drivers/char/esp.c b/drivers/char/esp.c index 15a4ea89632..93b55196251 100644 --- a/drivers/char/esp.c +++ b/drivers/char/esp.c @@ -723,9 +723,10 @@ static irqreturn_t rs_interrupt_single(int irq, void *dev_id) * ------------------------------------------------------------------- */ -static void do_softint(void *private_) +static void do_softint(struct work_struct *work) { - struct esp_struct *info = (struct esp_struct *) private_; + struct esp_struct *info = + container_of(work, struct esp_struct, tqueue); struct tty_struct *tty; tty = info->tty; @@ -746,9 +747,10 @@ static void do_softint(void *private_) * do_serial_hangup() -> tty->hangup() -> esp_hangup() * */ -static void do_serial_hangup(void *private_) +static void do_serial_hangup(struct work_struct *work) { - struct esp_struct *info = (struct esp_struct *) private_; + struct esp_struct *info = + container_of(work, struct esp_struct, tqueue_hangup); struct tty_struct *tty; tty = info->tty; @@ -2501,8 +2503,8 @@ static int __init espserial_init(void) info->magic = ESP_MAGIC; info->close_delay = 5*HZ/10; info->closing_wait = 30*HZ; - INIT_WORK(&info->tqueue, do_softint, info); - INIT_WORK(&info->tqueue_hangup, do_serial_hangup, info); + INIT_WORK(&info->tqueue, do_softint); + INIT_WORK(&info->tqueue_hangup, do_serial_hangup); info->config.rx_timeout = rx_timeout; info->config.flow_on = flow_on; info->config.flow_off = flow_off; diff --git a/drivers/char/genrtc.c b/drivers/char/genrtc.c index 817dc409ac2..23b25ada65e 100644 --- a/drivers/char/genrtc.c +++ b/drivers/char/genrtc.c @@ -102,7 +102,7 @@ static void gen_rtc_interrupt(unsigned long arg); * Routine to poll RTC seconds field for change as often as possible, * after first RTC_UIE use timer to reduce polling */ -static void genrtc_troutine(void *data) +static void genrtc_troutine(struct work_struct *work) { unsigned int tmp = get_rtc_ss(); @@ -255,7 +255,7 @@ static inline int gen_set_rtc_irq_bit(unsigned char bit) irq_active = 1; stop_rtc_timers = 0; lostint = 0; - INIT_WORK(&genrtc_task, genrtc_troutine, NULL); + INIT_WORK(&genrtc_task, genrtc_troutine); oldsecs = get_rtc_ss(); init_timer(&timer_task); diff --git a/drivers/char/hvsi.c b/drivers/char/hvsi.c index 2cf63e7305a..82a41d5b4ed 100644 --- a/drivers/char/hvsi.c +++ b/drivers/char/hvsi.c @@ -69,7 +69,7 @@ #define __ALIGNED__ __attribute__((__aligned__(sizeof(long)))) struct hvsi_struct { - struct work_struct writer; + struct delayed_work writer; struct work_struct handshaker; wait_queue_head_t emptyq; /* woken when outbuf is emptied */ wait_queue_head_t stateq; /* woken when HVSI state changes */ @@ -744,9 +744,10 @@ static int hvsi_handshake(struct hvsi_struct *hp) return 0; } -static void hvsi_handshaker(void *arg) +static void hvsi_handshaker(struct work_struct *work) { - struct hvsi_struct *hp = (struct hvsi_struct *)arg; + struct hvsi_struct *hp = + container_of(work, struct hvsi_struct, handshaker); if (hvsi_handshake(hp) >= 0) return; @@ -951,9 +952,10 @@ static void hvsi_push(struct hvsi_struct *hp) } /* hvsi_write_worker will keep rescheduling itself until outbuf is empty */ -static void hvsi_write_worker(void *arg) +static void hvsi_write_worker(struct work_struct *work) { - struct hvsi_struct *hp = (struct hvsi_struct *)arg; + struct hvsi_struct *hp = + container_of(work, struct hvsi_struct, writer.work); unsigned long flags; #ifdef DEBUG static long start_j = 0; @@ -1287,8 +1289,8 @@ static int __init hvsi_console_init(void) } hp = &hvsi_ports[hvsi_count]; - INIT_WORK(&hp->writer, hvsi_write_worker, hp); - INIT_WORK(&hp->handshaker, hvsi_handshaker, hp); + INIT_DELAYED_WORK(&hp->writer, hvsi_write_worker); + INIT_WORK(&hp->handshaker, hvsi_handshaker); init_waitqueue_head(&hp->emptyq); init_waitqueue_head(&hp->stateq); spin_lock_init(&hp->lock); diff --git a/drivers/char/ip2/i2lib.c b/drivers/char/ip2/i2lib.c index 54d93f0345e..c213fdbdb2b 100644 --- a/drivers/char/ip2/i2lib.c +++ b/drivers/char/ip2/i2lib.c @@ -84,8 +84,8 @@ static void iiSendPendingMail(i2eBordStrPtr); static void serviceOutgoingFifo(i2eBordStrPtr); // Functions defined in ip2.c as part of interrupt handling -static void do_input(void *); -static void do_status(void *); +static void do_input(struct work_struct *); +static void do_status(struct work_struct *); //*************** //* Debug Data * @@ -331,8 +331,8 @@ i2InitChannels ( i2eBordStrPtr pB, int nChannels, i2ChanStrPtr pCh) pCh->ClosingWaitTime = 30*HZ; // Initialize task queue objects - INIT_WORK(&pCh->tqueue_input, do_input, pCh); - INIT_WORK(&pCh->tqueue_status, do_status, pCh); + INIT_WORK(&pCh->tqueue_input, do_input); + INIT_WORK(&pCh->tqueue_status, do_status); #ifdef IP2DEBUG_TRACE pCh->trace = ip2trace; @@ -1573,7 +1573,7 @@ i2StripFifo(i2eBordStrPtr pB) #ifdef USE_IQ schedule_work(&pCh->tqueue_input); #else - do_input(pCh); + do_input(&pCh->tqueue_input); #endif // Note we do not need to maintain any flow-control credits at this @@ -1810,7 +1810,7 @@ i2StripFifo(i2eBordStrPtr pB) #ifdef USE_IQ schedule_work(&pCh->tqueue_status); #else - do_status(pCh); + do_status(&pCh->tqueue_status); #endif } } diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index a3f32d46d2f..cda2459c1d6 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -189,12 +189,12 @@ static int ip2_tiocmset(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear); static void set_irq(int, int); -static void ip2_interrupt_bh(i2eBordStrPtr pB); +static void ip2_interrupt_bh(struct work_struct *work); static irqreturn_t ip2_interrupt(int irq, void *dev_id); static void ip2_poll(unsigned long arg); static inline void service_all_boards(void); -static void do_input(void *p); -static void do_status(void *p); +static void do_input(struct work_struct *); +static void do_status(struct work_struct *); static void ip2_wait_until_sent(PTTY,int); @@ -918,7 +918,7 @@ ip2_init_board( int boardnum ) pCh++; } ex_exit: - INIT_WORK(&pB->tqueue_interrupt, (void(*)(void*)) ip2_interrupt_bh, pB); + INIT_WORK(&pB->tqueue_interrupt, ip2_interrupt_bh); return; err_release_region: @@ -1125,8 +1125,8 @@ service_all_boards(void) /******************************************************************************/ -/* Function: ip2_interrupt_bh(pB) */ -/* Parameters: pB - pointer to the board structure */ +/* Function: ip2_interrupt_bh(work) */ +/* Parameters: work - pointer to the board structure */ /* Returns: Nothing */ /* */ /* Description: */ @@ -1135,8 +1135,9 @@ service_all_boards(void) /* */ /******************************************************************************/ static void -ip2_interrupt_bh(i2eBordStrPtr pB) +ip2_interrupt_bh(struct work_struct *work) { + i2eBordStrPtr pB = container_of(work, i2eBordStr, tqueue_interrupt); // pB better well be set or we have a problem! We can only get // here from the IMMEDIATE queue. Here, we process the boards. // Checking pB doesn't cost much and it saves us from the sanity checkers. @@ -1245,9 +1246,9 @@ ip2_poll(unsigned long arg) ip2trace (ITRC_NO_PORT, ITRC_INTR, ITRC_RETURN, 0 ); } -static void do_input(void *p) +static void do_input(struct work_struct *work) { - i2ChanStrPtr pCh = p; + i2ChanStrPtr pCh = container_of(work, i2ChanStr, tqueue_input); unsigned long flags; ip2trace(CHANN, ITRC_INPUT, 21, 0 ); @@ -1279,9 +1280,9 @@ static inline void isig(int sig, struct tty_struct *tty, int flush) } } -static void do_status(void *p) +static void do_status(struct work_struct *work) { - i2ChanStrPtr pCh = p; + i2ChanStrPtr pCh = container_of(work, i2ChanStr, tqueue_status); int status; status = i2GetStatus( pCh, (I2_BRK|I2_PAR|I2_FRA|I2_OVR) ); diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c index 58c955e390b..1637c1d9a4b 100644 --- a/drivers/char/isicom.c +++ b/drivers/char/isicom.c @@ -530,9 +530,9 @@ sched_again: /* Interrupt handlers */ -static void isicom_bottomhalf(void *data) +static void isicom_bottomhalf(struct work_struct *work) { - struct isi_port *port = (struct isi_port *) data; + struct isi_port *port = container_of(work, struct isi_port, bh_tqueue); struct tty_struct *tty = port->tty; if (!tty) @@ -1474,9 +1474,9 @@ static void isicom_start(struct tty_struct *tty) } /* hangup et all */ -static void do_isicom_hangup(void *data) +static void do_isicom_hangup(struct work_struct *work) { - struct isi_port *port = data; + struct isi_port *port = container_of(work, struct isi_port, hangup_tq); struct tty_struct *tty; tty = port->tty; @@ -1966,8 +1966,8 @@ static int __devinit isicom_setup(void) port->channel = channel; port->close_delay = 50 * HZ/100; port->closing_wait = 3000 * HZ/100; - INIT_WORK(&port->hangup_tq, do_isicom_hangup, port); - INIT_WORK(&port->bh_tqueue, isicom_bottomhalf, port); + INIT_WORK(&port->hangup_tq, do_isicom_hangup); + INIT_WORK(&port->bh_tqueue, isicom_bottomhalf); port->status = 0; init_waitqueue_head(&port->open_wait); init_waitqueue_head(&port->close_wait); diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c index 96cb1f07332..2d025a9fd14 100644 --- a/drivers/char/moxa.c +++ b/drivers/char/moxa.c @@ -222,7 +222,7 @@ static struct semaphore moxaBuffSem; /* * static functions: */ -static void do_moxa_softint(void *); +static void do_moxa_softint(struct work_struct *); static int moxa_open(struct tty_struct *, struct file *); static void moxa_close(struct tty_struct *, struct file *); static int moxa_write(struct tty_struct *, const unsigned char *, int); @@ -363,7 +363,7 @@ static int __init moxa_init(void) for (i = 0, ch = moxaChannels; i < MAX_PORTS; i++, ch++) { ch->type = PORT_16550A; ch->port = i; - INIT_WORK(&ch->tqueue, do_moxa_softint, ch); + INIT_WORK(&ch->tqueue, do_moxa_softint); ch->tty = NULL; ch->close_delay = 5 * HZ / 10; ch->closing_wait = 30 * HZ; @@ -509,9 +509,9 @@ static void __exit moxa_exit(void) module_init(moxa_init); module_exit(moxa_exit); -static void do_moxa_softint(void *private_) +static void do_moxa_softint(struct work_struct *work) { - struct moxa_str *ch = (struct moxa_str *) private_; + struct moxa_str *ch = container_of(work, struct moxa_str, tqueue); struct tty_struct *tty; if (ch && (tty = ch->tty)) { diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c index 048d91142c1..5ed2486b758 100644 --- a/drivers/char/mxser.c +++ b/drivers/char/mxser.c @@ -389,7 +389,7 @@ static int mxser_init(void); /* static void mxser_poll(unsigned long); */ static int mxser_get_ISA_conf(int, struct mxser_hwconf *); static int mxser_get_PCI_conf(int, int, int, struct mxser_hwconf *); -static void mxser_do_softint(void *); +static void mxser_do_softint(struct work_struct *); static int mxser_open(struct tty_struct *, struct file *); static void mxser_close(struct tty_struct *, struct file *); static int mxser_write(struct tty_struct *, const unsigned char *, int); @@ -590,7 +590,7 @@ static int mxser_initbrd(int board, struct mxser_hwconf *hwconf) info->custom_divisor = hwconf->baud_base[i] * 16; info->close_delay = 5 * HZ / 10; info->closing_wait = 30 * HZ; - INIT_WORK(&info->tqueue, mxser_do_softint, info); + INIT_WORK(&info->tqueue, mxser_do_softint); info->normal_termios = mxvar_sdriver->init_termios; init_waitqueue_head(&info->open_wait); init_waitqueue_head(&info->close_wait); @@ -917,9 +917,10 @@ static int mxser_init(void) return 0; } -static void mxser_do_softint(void *private_) +static void mxser_do_softint(struct work_struct *work) { - struct mxser_struct *info = private_; + struct mxser_struct *info = + container_of(work, struct mxser_struct, tqueue); struct tty_struct *tty; tty = info->tty; diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c index 1a0bc30b79d..e4d950072b5 100644 --- a/drivers/char/pcmcia/synclink_cs.c +++ b/drivers/char/pcmcia/synclink_cs.c @@ -421,7 +421,7 @@ static irqreturn_t mgslpc_isr(int irq, void *dev_id); /* * Bottom half interrupt handlers */ -static void bh_handler(void* Context); +static void bh_handler(struct work_struct *work); static void bh_transmit(MGSLPC_INFO *info); static void bh_status(MGSLPC_INFO *info); @@ -547,7 +547,7 @@ static int mgslpc_probe(struct pcmcia_device *link) memset(info, 0, sizeof(MGSLPC_INFO)); info->magic = MGSLPC_MAGIC; - INIT_WORK(&info->task, bh_handler, info); + INIT_WORK(&info->task, bh_handler); info->max_frame_size = 4096; info->close_delay = 5*HZ/10; info->closing_wait = 30*HZ; @@ -842,9 +842,9 @@ static int bh_action(MGSLPC_INFO *info) return rc; } -static void bh_handler(void* Context) +static void bh_handler(struct work_struct *work) { - MGSLPC_INFO *info = (MGSLPC_INFO*)Context; + MGSLPC_INFO *info = container_of(work, MGSLPC_INFO, task); int action; if (!info) diff --git a/drivers/char/sonypi.c b/drivers/char/sonypi.c index c084149153d..fc87070f186 100644 --- a/drivers/char/sonypi.c +++ b/drivers/char/sonypi.c @@ -765,7 +765,7 @@ static void sonypi_setbluetoothpower(u8 state) sonypi_device.bluetooth_power = state; } -static void input_keyrelease(void *data) +static void input_keyrelease(struct work_struct *work) { struct sonypi_keypress kp; @@ -1412,7 +1412,7 @@ static int __devinit sonypi_probe(struct platform_device *dev) goto err_inpdev_unregister; } - INIT_WORK(&sonypi_device.input_work, input_keyrelease, NULL); + INIT_WORK(&sonypi_device.input_work, input_keyrelease); } sonypi_enable(0); diff --git a/drivers/char/specialix.c b/drivers/char/specialix.c index 7e1bd9562c2..99137ab66b6 100644 --- a/drivers/char/specialix.c +++ b/drivers/char/specialix.c @@ -2261,9 +2261,10 @@ static void sx_start(struct tty_struct * tty) * do_sx_hangup() -> tty->hangup() -> sx_hangup() * */ -static void do_sx_hangup(void *private_) +static void do_sx_hangup(struct work_struct *work) { - struct specialix_port *port = (struct specialix_port *) private_; + struct specialix_port *port = + container_of(work, struct specialix_port, tqueue_hangup); struct tty_struct *tty; func_enter(); @@ -2336,9 +2337,10 @@ static void sx_set_termios(struct tty_struct * tty, struct termios * old_termios } -static void do_softint(void *private_) +static void do_softint(struct work_struct *work) { - struct specialix_port *port = (struct specialix_port *) private_; + struct specialix_port *port = + container_of(work, struct specialix_port, tqueue); struct tty_struct *tty; func_enter(); @@ -2411,8 +2413,8 @@ static int sx_init_drivers(void) memset(sx_port, 0, sizeof(sx_port)); for (i = 0; i < SX_NPORT * SX_NBOARD; i++) { sx_port[i].magic = SPECIALIX_MAGIC; - INIT_WORK(&sx_port[i].tqueue, do_softint, &sx_port[i]); - INIT_WORK(&sx_port[i].tqueue_hangup, do_sx_hangup, &sx_port[i]); + INIT_WORK(&sx_port[i].tqueue, do_softint); + INIT_WORK(&sx_port[i].tqueue_hangup, do_sx_hangup); sx_port[i].close_delay = 50 * HZ/100; sx_port[i].closing_wait = 3000 * HZ/100; init_waitqueue_head(&sx_port[i].open_wait); diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c index 06784adcc35..147c30da81e 100644 --- a/drivers/char/synclink.c +++ b/drivers/char/synclink.c @@ -802,7 +802,7 @@ static int save_tx_buffer_request(struct mgsl_struct *info,const char *Buffer, u /* * Bottom half interrupt handlers */ -static void mgsl_bh_handler(void* Context); +static void mgsl_bh_handler(struct work_struct *work); static void mgsl_bh_receive(struct mgsl_struct *info); static void mgsl_bh_transmit(struct mgsl_struct *info); static void mgsl_bh_status(struct mgsl_struct *info); @@ -1071,9 +1071,10 @@ static int mgsl_bh_action(struct mgsl_struct *info) /* * Perform bottom half processing of work items queued by ISR. */ -static void mgsl_bh_handler(void* Context) +static void mgsl_bh_handler(struct work_struct *work) { - struct mgsl_struct *info = (struct mgsl_struct*)Context; + struct mgsl_struct *info = + container_of(work, struct mgsl_struct, task); int action; if (!info) @@ -4337,7 +4338,7 @@ static struct mgsl_struct* mgsl_allocate_device(void) } else { memset(info, 0, sizeof(struct mgsl_struct)); info->magic = MGSL_MAGIC; - INIT_WORK(&info->task, mgsl_bh_handler, info); + INIT_WORK(&info->task, mgsl_bh_handler); info->max_frame_size = 4096; info->close_delay = 5*HZ/10; info->closing_wait = 30*HZ; diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c index d4334c79f8d..07f34d43dc7 100644 --- a/drivers/char/synclink_gt.c +++ b/drivers/char/synclink_gt.c @@ -485,7 +485,7 @@ static void enable_loopback(struct slgt_info *info); static void set_rate(struct slgt_info *info, u32 data_rate); static int bh_action(struct slgt_info *info); -static void bh_handler(void* context); +static void bh_handler(struct work_struct *work); static void bh_transmit(struct slgt_info *info); static void isr_serial(struct slgt_info *info); static void isr_rdma(struct slgt_info *info); @@ -1878,9 +1878,9 @@ static int bh_action(struct slgt_info *info) /* * perform bottom half processing */ -static void bh_handler(void* context) +static void bh_handler(struct work_struct *work) { - struct slgt_info *info = context; + struct slgt_info *info = container_of(work, struct slgt_info, task); int action; if (!info) @@ -3326,7 +3326,7 @@ static struct slgt_info *alloc_dev(int adapter_num, int port_num, struct pci_dev } else { memset(info, 0, sizeof(struct slgt_info)); info->magic = MGSL_MAGIC; - INIT_WORK(&info->task, bh_handler, info); + INIT_WORK(&info->task, bh_handler); info->max_frame_size = 4096; info->raw_rx_size = DMABUFSIZE; info->close_delay = 5*HZ/10; @@ -4799,6 +4799,6 @@ static void rx_timeout(unsigned long context) spin_lock_irqsave(&info->lock, flags); info->pending_bh |= BH_RECEIVE; spin_unlock_irqrestore(&info->lock, flags); - bh_handler(info); + bh_handler(&info->task); } diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c index 3e932b68137..13a57245cf2 100644 --- a/drivers/char/synclinkmp.c +++ b/drivers/char/synclinkmp.c @@ -602,7 +602,7 @@ static void enable_loopback(SLMP_INFO *info, int enable); static void set_rate(SLMP_INFO *info, u32 data_rate); static int bh_action(SLMP_INFO *info); -static void bh_handler(void* Context); +static void bh_handler(struct work_struct *work); static void bh_receive(SLMP_INFO *info); static void bh_transmit(SLMP_INFO *info); static void bh_status(SLMP_INFO *info); @@ -2063,9 +2063,9 @@ int bh_action(SLMP_INFO *info) /* Perform bottom half processing of work items queued by ISR. */ -void bh_handler(void* Context) +void bh_handler(struct work_struct *work) { - SLMP_INFO *info = (SLMP_INFO*)Context; + SLMP_INFO *info = container_of(work, SLMP_INFO, task); int action; if (!info) @@ -3805,7 +3805,7 @@ static SLMP_INFO *alloc_dev(int adapter_num, int port_num, struct pci_dev *pdev) } else { memset(info, 0, sizeof(SLMP_INFO)); info->magic = MGSL_MAGIC; - INIT_WORK(&info->task, bh_handler, info); + INIT_WORK(&info->task, bh_handler); info->max_frame_size = 4096; info->close_delay = 5*HZ/10; info->closing_wait = 30*HZ; diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c index 6ad2d3bb945..36f91a65527 100644 --- a/drivers/char/tpm/tpm.c +++ b/drivers/char/tpm/tpm.c @@ -325,9 +325,9 @@ static void user_reader_timeout(unsigned long ptr) schedule_work(&chip->work); } -static void timeout_work(void *ptr) +static void timeout_work(struct work_struct *work) { - struct tpm_chip *chip = ptr; + struct tpm_chip *chip = container_of(work, struct tpm_chip, work); down(&chip->buffer_mutex); atomic_set(&chip->data_pending, 0); @@ -1105,7 +1105,7 @@ struct tpm_chip *tpm_register_hardware(struct device *dev, const struct tpm_vend init_MUTEX(&chip->tpm_mutex); INIT_LIST_HEAD(&chip->list); - INIT_WORK(&chip->work, timeout_work, chip); + INIT_WORK(&chip->work, timeout_work); init_timer(&chip->user_read_timer); chip->user_read_timer.function = user_reader_timeout; diff --git a/drivers/connector/cn_queue.c b/drivers/connector/cn_queue.c index 05f8ce2cfb4..b418b16e910 100644 --- a/drivers/connector/cn_queue.c +++ b/drivers/connector/cn_queue.c @@ -31,9 +31,11 @@ #include #include -void cn_queue_wrapper(void *data) +void cn_queue_wrapper(struct work_struct *work) { - struct cn_callback_data *d = data; + struct cn_callback_entry *cbq = + container_of(work, struct cn_callback_entry, work.work); + struct cn_callback_data *d = &cbq->data; d->callback(d->callback_priv); @@ -57,7 +59,7 @@ static struct cn_callback_entry *cn_queue_alloc_callback_entry(char *name, struc memcpy(&cbq->id.id, id, sizeof(struct cb_id)); cbq->data.callback = callback; - INIT_WORK(&cbq->work, &cn_queue_wrapper, &cbq->data); + INIT_DELAYED_WORK(&cbq->work, &cn_queue_wrapper); return cbq; } diff --git a/drivers/connector/connector.c b/drivers/connector/connector.c index b49bacfd8de..5e7cd45d10e 100644 --- a/drivers/connector/connector.c +++ b/drivers/connector/connector.c @@ -135,40 +135,39 @@ static int cn_call_callback(struct cn_msg *msg, void (*destruct_data)(void *), v spin_lock_bh(&dev->cbdev->queue_lock); list_for_each_entry(__cbq, &dev->cbdev->queue_list, callback_entry) { if (cn_cb_equal(&__cbq->id.id, &msg->id)) { - if (likely(!test_bit(0, &__cbq->work.pending) && + if (likely(!test_bit(WORK_STRUCT_PENDING, + &__cbq->work.work.management) && __cbq->data.ddata == NULL)) { __cbq->data.callback_priv = msg; __cbq->data.ddata = data; __cbq->data.destruct_data = destruct_data; - if (queue_work(dev->cbdev->cn_queue, - &__cbq->work)) + if (queue_delayed_work( + dev->cbdev->cn_queue, + &__cbq->work, 0)) err = 0; } else { - struct work_struct *w; struct cn_callback_data *d; - w = kzalloc(sizeof(*w) + sizeof(*d), GFP_ATOMIC); - if (w) { - d = (struct cn_callback_data *)(w+1); - + __cbq = kzalloc(sizeof(*__cbq), GFP_ATOMIC); + if (__cbq) { + d = &__cbq->data; d->callback_priv = msg; d->callback = __cbq->data.callback; d->ddata = data; d->destruct_data = destruct_data; - d->free = w; + d->free = __cbq; - INIT_LIST_HEAD(&w->entry); - w->pending = 0; - w->func = &cn_queue_wrapper; - w->data = d; - init_timer(&w->timer); + INIT_DELAYED_WORK(&__cbq->work, + &cn_queue_wrapper); - if (queue_work(dev->cbdev->cn_queue, w)) + if (queue_delayed_work( + dev->cbdev->cn_queue, + &__cbq->work, 0)) err = 0; else { - kfree(w); + kfree(__cbq); err = -EINVAL; } } else diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c index c4c578defab..5ef5ede5b88 100644 --- a/drivers/cpufreq/cpufreq_conservative.c +++ b/drivers/cpufreq/cpufreq_conservative.c @@ -59,7 +59,7 @@ static unsigned int def_sampling_rate; #define MAX_SAMPLING_DOWN_FACTOR (10) #define TRANSITION_LATENCY_LIMIT (10 * 1000) -static void do_dbs_timer(void *data); +static void do_dbs_timer(struct work_struct *work); struct cpu_dbs_info_s { struct cpufreq_policy *cur_policy; @@ -82,7 +82,7 @@ static unsigned int dbs_enable; /* number of CPUs using this policy */ * is recursive for the same process. -Venki */ static DEFINE_MUTEX (dbs_mutex); -static DECLARE_WORK (dbs_work, do_dbs_timer, NULL); +static DECLARE_DELAYED_WORK(dbs_work, do_dbs_timer); struct dbs_tuners { unsigned int sampling_rate; @@ -420,7 +420,7 @@ static void dbs_check_cpu(int cpu) } } -static void do_dbs_timer(void *data) +static void do_dbs_timer(struct work_struct *work) { int i; lock_cpu_hotplug(); @@ -435,7 +435,6 @@ static void do_dbs_timer(void *data) static inline void dbs_timer_init(void) { - INIT_WORK(&dbs_work, do_dbs_timer, NULL); schedule_delayed_work(&dbs_work, usecs_to_jiffies(dbs_tuners_ins.sampling_rate)); return; diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index bf8aa45d4f0..e1cc5113c2a 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -47,13 +47,17 @@ static unsigned int def_sampling_rate; #define DEF_SAMPLING_RATE_LATENCY_MULTIPLIER (1000) #define TRANSITION_LATENCY_LIMIT (10 * 1000) -static void do_dbs_timer(void *data); +static void do_dbs_timer(struct work_struct *work); + +/* Sampling types */ +enum dbs_sample {DBS_NORMAL_SAMPLE, DBS_SUB_SAMPLE}; struct cpu_dbs_info_s { cputime64_t prev_cpu_idle; cputime64_t prev_cpu_wall; struct cpufreq_policy *cur_policy; - struct work_struct work; + struct delayed_work work; + enum dbs_sample sample_type; unsigned int enable; struct cpufreq_frequency_table *freq_table; unsigned int freq_lo; @@ -407,30 +411,31 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info) } } -/* Sampling types */ -enum {DBS_NORMAL_SAMPLE, DBS_SUB_SAMPLE}; - -static void do_dbs_timer(void *data) +static void do_dbs_timer(struct work_struct *work) { unsigned int cpu = smp_processor_id(); struct cpu_dbs_info_s *dbs_info = &per_cpu(cpu_dbs_info, cpu); + enum dbs_sample sample_type = dbs_info->sample_type; /* We want all CPUs to do sampling nearly on same jiffy */ int delay = usecs_to_jiffies(dbs_tuners_ins.sampling_rate); + + /* Permit rescheduling of this work item */ + work_release(work); + delay -= jiffies % delay; if (!dbs_info->enable) return; /* Common NORMAL_SAMPLE setup */ - INIT_WORK(&dbs_info->work, do_dbs_timer, (void *)DBS_NORMAL_SAMPLE); + dbs_info->sample_type = DBS_NORMAL_SAMPLE; if (!dbs_tuners_ins.powersave_bias || - (unsigned long) data == DBS_NORMAL_SAMPLE) { + sample_type == DBS_NORMAL_SAMPLE) { lock_cpu_hotplug(); dbs_check_cpu(dbs_info); unlock_cpu_hotplug(); if (dbs_info->freq_lo) { /* Setup timer for SUB_SAMPLE */ - INIT_WORK(&dbs_info->work, do_dbs_timer, - (void *)DBS_SUB_SAMPLE); + dbs_info->sample_type = DBS_SUB_SAMPLE; delay = dbs_info->freq_hi_jiffies; } } else { @@ -449,7 +454,8 @@ static inline void dbs_timer_init(unsigned int cpu) delay -= jiffies % delay; ondemand_powersave_bias_init(); - INIT_WORK(&dbs_info->work, do_dbs_timer, NULL); + INIT_DELAYED_WORK_NAR(&dbs_info->work, do_dbs_timer); + dbs_info->sample_type = DBS_NORMAL_SAMPLE; queue_delayed_work_on(cpu, kondemand_wq, &dbs_info->work, delay); } diff --git a/drivers/i2c/chips/ds1374.c b/drivers/i2c/chips/ds1374.c index 4630f1969a0..15edf40828b 100644 --- a/drivers/i2c/chips/ds1374.c +++ b/drivers/i2c/chips/ds1374.c @@ -140,12 +140,14 @@ ulong ds1374_get_rtc_time(void) return t1; } -static void ds1374_set_work(void *arg) +static ulong new_time; + +static void ds1374_set_work(struct work_struct *work) { ulong t1, t2; int limit = 10; /* arbitrary retry limit */ - t1 = *(ulong *) arg; + t1 = new_time; mutex_lock(&ds1374_mutex); @@ -167,11 +169,9 @@ static void ds1374_set_work(void *arg) "can't confirm time set from rtc chip\n"); } -static ulong new_time; - static struct workqueue_struct *ds1374_workqueue; -static DECLARE_WORK(ds1374_work, ds1374_set_work, &new_time); +static DECLARE_WORK(ds1374_work, ds1374_set_work); int ds1374_set_rtc_time(ulong nowtime) { @@ -180,7 +180,7 @@ int ds1374_set_rtc_time(ulong nowtime) if (in_interrupt()) queue_work(ds1374_workqueue, &ds1374_work); else - ds1374_set_work(&new_time); + ds1374_set_work(NULL); return 0; } diff --git a/drivers/ieee1394/hosts.c b/drivers/ieee1394/hosts.c index d90a3a1898c..8f4378a1631 100644 --- a/drivers/ieee1394/hosts.c +++ b/drivers/ieee1394/hosts.c @@ -31,9 +31,10 @@ #include "config_roms.h" -static void delayed_reset_bus(void * __reset_info) +static void delayed_reset_bus(struct work_struct *work) { - struct hpsb_host *host = (struct hpsb_host*)__reset_info; + struct hpsb_host *host = + container_of(work, struct hpsb_host, delayed_reset.work); int generation = host->csr.generation + 1; /* The generation field rolls over to 2 rather than 0 per IEEE @@ -145,7 +146,7 @@ struct hpsb_host *hpsb_alloc_host(struct hpsb_host_driver *drv, size_t extra, atomic_set(&h->generation, 0); - INIT_WORK(&h->delayed_reset, delayed_reset_bus, h); + INIT_DELAYED_WORK(&h->delayed_reset, delayed_reset_bus); init_timer(&h->timeout); h->timeout.data = (unsigned long) h; @@ -234,7 +235,7 @@ int hpsb_update_config_rom_image(struct hpsb_host *host) * Config ROM in the near future. */ reset_delay = HZ; - PREPARE_WORK(&host->delayed_reset, delayed_reset_bus, host); + PREPARE_DELAYED_WORK(&host->delayed_reset, delayed_reset_bus); schedule_delayed_work(&host->delayed_reset, reset_delay); return 0; diff --git a/drivers/ieee1394/hosts.h b/drivers/ieee1394/hosts.h index bc6dbfadb89..d553e38c954 100644 --- a/drivers/ieee1394/hosts.h +++ b/drivers/ieee1394/hosts.h @@ -62,7 +62,7 @@ struct hpsb_host { struct class_device class_dev; int update_config_rom; - struct work_struct delayed_reset; + struct delayed_work delayed_reset; unsigned int config_roms; struct list_head addr_space; diff --git a/drivers/ieee1394/sbp2.c b/drivers/ieee1394/sbp2.c index 6986ac18828..cd156d4e779 100644 --- a/drivers/ieee1394/sbp2.c +++ b/drivers/ieee1394/sbp2.c @@ -493,20 +493,25 @@ static void sbp2util_notify_fetch_agent(struct scsi_id_instance_data *scsi_id, scsi_unblock_requests(scsi_id->scsi_host); } -static void sbp2util_write_orb_pointer(void *p) +static void sbp2util_write_orb_pointer(struct work_struct *work) { + struct scsi_id_instance_data *scsi_id = + container_of(work, struct scsi_id_instance_data, + protocol_work.work); quadlet_t data[2]; - data[0] = ORB_SET_NODE_ID( - ((struct scsi_id_instance_data *)p)->hi->host->node_id); - data[1] = ((struct scsi_id_instance_data *)p)->last_orb_dma; + data[0] = ORB_SET_NODE_ID(scsi_id->hi->host->node_id); + data[1] = scsi_id->last_orb_dma; sbp2util_cpu_to_be32_buffer(data, 8); - sbp2util_notify_fetch_agent(p, SBP2_ORB_POINTER_OFFSET, data, 8); + sbp2util_notify_fetch_agent(scsi_id, SBP2_ORB_POINTER_OFFSET, data, 8); } -static void sbp2util_write_doorbell(void *p) +static void sbp2util_write_doorbell(struct work_struct *work) { - sbp2util_notify_fetch_agent(p, SBP2_DOORBELL_OFFSET, NULL, 4); + struct scsi_id_instance_data *scsi_id = + container_of(work, struct scsi_id_instance_data, + protocol_work.work); + sbp2util_notify_fetch_agent(scsi_id, SBP2_DOORBELL_OFFSET, NULL, 4); } /* @@ -843,7 +848,7 @@ static struct scsi_id_instance_data *sbp2_alloc_device(struct unit_directory *ud INIT_LIST_HEAD(&scsi_id->scsi_list); spin_lock_init(&scsi_id->sbp2_command_orb_lock); atomic_set(&scsi_id->state, SBP2LU_STATE_RUNNING); - INIT_WORK(&scsi_id->protocol_work, NULL, NULL); + INIT_DELAYED_WORK(&scsi_id->protocol_work, NULL); ud->device.driver_data = scsi_id; @@ -2047,11 +2052,10 @@ static void sbp2_link_orb_command(struct scsi_id_instance_data *scsi_id, * We do not accept new commands until the job is over. */ scsi_block_requests(scsi_id->scsi_host); - PREPARE_WORK(&scsi_id->protocol_work, + PREPARE_DELAYED_WORK(&scsi_id->protocol_work, last_orb ? sbp2util_write_doorbell: - sbp2util_write_orb_pointer, - scsi_id); - schedule_work(&scsi_id->protocol_work); + sbp2util_write_orb_pointer); + schedule_delayed_work(&scsi_id->protocol_work, 0); } } diff --git a/drivers/ieee1394/sbp2.h b/drivers/ieee1394/sbp2.h index abbe48e646c..1b16d6b9cf1 100644 --- a/drivers/ieee1394/sbp2.h +++ b/drivers/ieee1394/sbp2.h @@ -348,7 +348,7 @@ struct scsi_id_instance_data { unsigned workarounds; atomic_t state; - struct work_struct protocol_work; + struct delayed_work protocol_work; }; /* For use in scsi_id_instance_data.state */ diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index e11187ecc93..84b2f5cb372 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -55,11 +55,11 @@ struct addr_req { int status; }; -static void process_req(void *data); +static void process_req(struct work_struct *work); static DEFINE_MUTEX(lock); static LIST_HEAD(req_list); -static DECLARE_WORK(work, process_req, NULL); +static DECLARE_DELAYED_WORK(work, process_req); static struct workqueue_struct *addr_wq; void rdma_addr_register_client(struct rdma_addr_client *client) @@ -215,7 +215,7 @@ out: return ret; } -static void process_req(void *data) +static void process_req(struct work_struct *work) { struct addr_req *req, *temp_req; struct sockaddr_in *src_in, *dst_in; diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c index 20e9f64e67a..98272fbbfb3 100644 --- a/drivers/infiniband/core/cache.c +++ b/drivers/infiniband/core/cache.c @@ -285,9 +285,10 @@ err: kfree(tprops); } -static void ib_cache_task(void *work_ptr) +static void ib_cache_task(struct work_struct *_work) { - struct ib_update_work *work = work_ptr; + struct ib_update_work *work = + container_of(_work, struct ib_update_work, work); ib_cache_update(work->device, work->port_num); kfree(work); @@ -306,7 +307,7 @@ static void ib_cache_event(struct ib_event_handler *handler, event->event == IB_EVENT_CLIENT_REREGISTER) { work = kmalloc(sizeof *work, GFP_ATOMIC); if (work) { - INIT_WORK(&work->work, ib_cache_task, work); + INIT_WORK(&work->work, ib_cache_task); work->device = event->device; work->port_num = event->element.port_num; schedule_work(&work->work); diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index 25b1018a476..e1990f531d0 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -101,7 +101,7 @@ struct cm_av { }; struct cm_work { - struct work_struct work; + struct delayed_work work; struct list_head list; struct cm_port *port; struct ib_mad_recv_wc *mad_recv_wc; /* Received MADs */ @@ -161,7 +161,7 @@ struct cm_id_private { atomic_t work_count; }; -static void cm_work_handler(void *data); +static void cm_work_handler(struct work_struct *work); static inline void cm_deref_id(struct cm_id_private *cm_id_priv) { @@ -669,8 +669,7 @@ static struct cm_timewait_info * cm_create_timewait_info(__be32 local_id) return ERR_PTR(-ENOMEM); timewait_info->work.local_id = local_id; - INIT_WORK(&timewait_info->work.work, cm_work_handler, - &timewait_info->work); + INIT_DELAYED_WORK(&timewait_info->work.work, cm_work_handler); timewait_info->work.cm_event.event = IB_CM_TIMEWAIT_EXIT; return timewait_info; } @@ -2987,9 +2986,9 @@ static void cm_send_handler(struct ib_mad_agent *mad_agent, } } -static void cm_work_handler(void *data) +static void cm_work_handler(struct work_struct *_work) { - struct cm_work *work = data; + struct cm_work *work = container_of(_work, struct cm_work, work.work); int ret; switch (work->cm_event.event) { @@ -3079,12 +3078,12 @@ int ib_cm_establish(struct ib_cm_id *cm_id) * we need to find the cm_id once we're in the context of the * worker thread, rather than holding a reference on it. */ - INIT_WORK(&work->work, cm_work_handler, work); + INIT_DELAYED_WORK(&work->work, cm_work_handler); work->local_id = cm_id->local_id; work->remote_id = cm_id->remote_id; work->mad_recv_wc = NULL; work->cm_event.event = IB_CM_USER_ESTABLISHED; - queue_work(cm.wq, &work->work); + queue_delayed_work(cm.wq, &work->work, 0); out: return ret; } @@ -3146,11 +3145,11 @@ static void cm_recv_handler(struct ib_mad_agent *mad_agent, return; } - INIT_WORK(&work->work, cm_work_handler, work); + INIT_DELAYED_WORK(&work->work, cm_work_handler); work->cm_event.event = event; work->mad_recv_wc = mad_recv_wc; work->port = (struct cm_port *)mad_agent->context; - queue_work(cm.wq, &work->work); + queue_delayed_work(cm.wq, &work->work, 0); } static int cm_init_qp_init_attr(struct cm_id_private *cm_id_priv, diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 845090b0859..189f73f3f72 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -1341,9 +1341,9 @@ static int cma_query_ib_route(struct rdma_id_private *id_priv, int timeout_ms, return (id_priv->query_id < 0) ? id_priv->query_id : 0; } -static void cma_work_handler(void *data) +static void cma_work_handler(struct work_struct *_work) { - struct cma_work *work = data; + struct cma_work *work = container_of(_work, struct cma_work, work); struct rdma_id_private *id_priv = work->id; int destroy = 0; @@ -1374,7 +1374,7 @@ static int cma_resolve_ib_route(struct rdma_id_private *id_priv, int timeout_ms) return -ENOMEM; work->id = id_priv; - INIT_WORK(&work->work, cma_work_handler, work); + INIT_WORK(&work->work, cma_work_handler); work->old_state = CMA_ROUTE_QUERY; work->new_state = CMA_ROUTE_RESOLVED; work->event.event = RDMA_CM_EVENT_ROUTE_RESOLVED; @@ -1431,7 +1431,7 @@ static int cma_resolve_iw_route(struct rdma_id_private *id_priv, int timeout_ms) return -ENOMEM; work->id = id_priv; - INIT_WORK(&work->work, cma_work_handler, work); + INIT_WORK(&work->work, cma_work_handler); work->old_state = CMA_ROUTE_QUERY; work->new_state = CMA_ROUTE_RESOLVED; work->event.event = RDMA_CM_EVENT_ROUTE_RESOLVED; @@ -1585,7 +1585,7 @@ static int cma_resolve_loopback(struct rdma_id_private *id_priv) } work->id = id_priv; - INIT_WORK(&work->work, cma_work_handler, work); + INIT_WORK(&work->work, cma_work_handler); work->old_state = CMA_ADDR_QUERY; work->new_state = CMA_ADDR_RESOLVED; work->event.event = RDMA_CM_EVENT_ADDR_RESOLVED; diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c index c3fb304a4e8..9bfa785252d 100644 --- a/drivers/infiniband/core/iwcm.c +++ b/drivers/infiniband/core/iwcm.c @@ -828,9 +828,10 @@ static int process_event(struct iwcm_id_private *cm_id_priv, * thread asleep on the destroy_comp list vs. an object destroyed * here synchronously when the last reference is removed. */ -static void cm_work_handler(void *arg) +static void cm_work_handler(struct work_struct *_work) { - struct iwcm_work *work = arg, lwork; + struct iwcm_work lwork, *work = + container_of(_work, struct iwcm_work, work); struct iwcm_id_private *cm_id_priv = work->cm_id; unsigned long flags; int empty; @@ -899,7 +900,7 @@ static int cm_event_handler(struct iw_cm_id *cm_id, goto out; } - INIT_WORK(&work->work, cm_work_handler, work); + INIT_WORK(&work->work, cm_work_handler); work->cm_id = cm_id_priv; work->event = *iw_event; diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index a72bcea46ff..5a54ac35e96 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -65,8 +65,8 @@ static struct ib_mad_agent_private *find_mad_agent( static int ib_mad_post_receive_mads(struct ib_mad_qp_info *qp_info, struct ib_mad_private *mad); static void cancel_mads(struct ib_mad_agent_private *mad_agent_priv); -static void timeout_sends(void *data); -static void local_completions(void *data); +static void timeout_sends(struct work_struct *work); +static void local_completions(struct work_struct *work); static int add_nonoui_reg_req(struct ib_mad_reg_req *mad_reg_req, struct ib_mad_agent_private *agent_priv, u8 mgmt_class); @@ -356,10 +356,9 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, INIT_LIST_HEAD(&mad_agent_priv->wait_list); INIT_LIST_HEAD(&mad_agent_priv->done_list); INIT_LIST_HEAD(&mad_agent_priv->rmpp_list); - INIT_WORK(&mad_agent_priv->timed_work, timeout_sends, mad_agent_priv); + INIT_DELAYED_WORK(&mad_agent_priv->timed_work, timeout_sends); INIT_LIST_HEAD(&mad_agent_priv->local_list); - INIT_WORK(&mad_agent_priv->local_work, local_completions, - mad_agent_priv); + INIT_WORK(&mad_agent_priv->local_work, local_completions); atomic_set(&mad_agent_priv->refcount, 1); init_completion(&mad_agent_priv->comp); @@ -2198,12 +2197,12 @@ static void mad_error_handler(struct ib_mad_port_private *port_priv, /* * IB MAD completion callback */ -static void ib_mad_completion_handler(void *data) +static void ib_mad_completion_handler(struct work_struct *work) { struct ib_mad_port_private *port_priv; struct ib_wc wc; - port_priv = (struct ib_mad_port_private *)data; + port_priv = container_of(work, struct ib_mad_port_private, work); ib_req_notify_cq(port_priv->cq, IB_CQ_NEXT_COMP); while (ib_poll_cq(port_priv->cq, 1, &wc) == 1) { @@ -2324,7 +2323,7 @@ void ib_cancel_mad(struct ib_mad_agent *mad_agent, } EXPORT_SYMBOL(ib_cancel_mad); -static void local_completions(void *data) +static void local_completions(struct work_struct *work) { struct ib_mad_agent_private *mad_agent_priv; struct ib_mad_local_private *local; @@ -2334,7 +2333,8 @@ static void local_completions(void *data) struct ib_wc wc; struct ib_mad_send_wc mad_send_wc; - mad_agent_priv = (struct ib_mad_agent_private *)data; + mad_agent_priv = + container_of(work, struct ib_mad_agent_private, local_work); spin_lock_irqsave(&mad_agent_priv->lock, flags); while (!list_empty(&mad_agent_priv->local_list)) { @@ -2434,14 +2434,15 @@ static int retry_send(struct ib_mad_send_wr_private *mad_send_wr) return ret; } -static void timeout_sends(void *data) +static void timeout_sends(struct work_struct *work) { struct ib_mad_agent_private *mad_agent_priv; struct ib_mad_send_wr_private *mad_send_wr; struct ib_mad_send_wc mad_send_wc; unsigned long flags, delay; - mad_agent_priv = (struct ib_mad_agent_private *)data; + mad_agent_priv = container_of(work, struct ib_mad_agent_private, + timed_work.work); mad_send_wc.vendor_err = 0; spin_lock_irqsave(&mad_agent_priv->lock, flags); @@ -2799,7 +2800,7 @@ static int ib_mad_port_open(struct ib_device *device, ret = -ENOMEM; goto error8; } - INIT_WORK(&port_priv->work, ib_mad_completion_handler, port_priv); + INIT_WORK(&port_priv->work, ib_mad_completion_handler); spin_lock_irqsave(&ib_mad_port_list_lock, flags); list_add_tail(&port_priv->port_list, &ib_mad_port_list); diff --git a/drivers/infiniband/core/mad_priv.h b/drivers/infiniband/core/mad_priv.h index d06b59083f6..d5548e73e06 100644 --- a/drivers/infiniband/core/mad_priv.h +++ b/drivers/infiniband/core/mad_priv.h @@ -102,7 +102,7 @@ struct ib_mad_agent_private { struct list_head send_list; struct list_head wait_list; struct list_head done_list; - struct work_struct timed_work; + struct delayed_work timed_work; unsigned long timeout; struct list_head local_list; struct work_struct local_work; diff --git a/drivers/infiniband/core/mad_rmpp.c b/drivers/infiniband/core/mad_rmpp.c index 1ef79d015a1..3663fd7022b 100644 --- a/drivers/infiniband/core/mad_rmpp.c +++ b/drivers/infiniband/core/mad_rmpp.c @@ -45,8 +45,8 @@ enum rmpp_state { struct mad_rmpp_recv { struct ib_mad_agent_private *agent; struct list_head list; - struct work_struct timeout_work; - struct work_struct cleanup_work; + struct delayed_work timeout_work; + struct delayed_work cleanup_work; struct completion comp; enum rmpp_state state; spinlock_t lock; @@ -233,9 +233,10 @@ static void nack_recv(struct ib_mad_agent_private *agent, } } -static void recv_timeout_handler(void *data) +static void recv_timeout_handler(struct work_struct *work) { - struct mad_rmpp_recv *rmpp_recv = data; + struct mad_rmpp_recv *rmpp_recv = + container_of(work, struct mad_rmpp_recv, timeout_work.work); struct ib_mad_recv_wc *rmpp_wc; unsigned long flags; @@ -254,9 +255,10 @@ static void recv_timeout_handler(void *data) ib_free_recv_mad(rmpp_wc); } -static void recv_cleanup_handler(void *data) +static void recv_cleanup_handler(struct work_struct *work) { - struct mad_rmpp_recv *rmpp_recv = data; + struct mad_rmpp_recv *rmpp_recv = + container_of(work, struct mad_rmpp_recv, cleanup_work.work); unsigned long flags; spin_lock_irqsave(&rmpp_recv->agent->lock, flags); @@ -285,8 +287,8 @@ create_rmpp_recv(struct ib_mad_agent_private *agent, rmpp_recv->agent = agent; init_completion(&rmpp_recv->comp); - INIT_WORK(&rmpp_recv->timeout_work, recv_timeout_handler, rmpp_recv); - INIT_WORK(&rmpp_recv->cleanup_work, recv_cleanup_handler, rmpp_recv); + INIT_DELAYED_WORK(&rmpp_recv->timeout_work, recv_timeout_handler); + INIT_DELAYED_WORK(&rmpp_recv->cleanup_work, recv_cleanup_handler); spin_lock_init(&rmpp_recv->lock); rmpp_recv->state = RMPP_STATE_ACTIVE; atomic_set(&rmpp_recv->refcount, 1); diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c index 1706d3c7e95..e45afba7534 100644 --- a/drivers/infiniband/core/sa_query.c +++ b/drivers/infiniband/core/sa_query.c @@ -360,9 +360,10 @@ static void free_sm_ah(struct kref *kref) kfree(sm_ah); } -static void update_sm_ah(void *port_ptr) +static void update_sm_ah(struct work_struct *work) { - struct ib_sa_port *port = port_ptr; + struct ib_sa_port *port = + container_of(work, struct ib_sa_port, update_task); struct ib_sa_sm_ah *new_ah, *old_ah; struct ib_port_attr port_attr; struct ib_ah_attr ah_attr; @@ -992,8 +993,7 @@ static void ib_sa_add_one(struct ib_device *device) if (IS_ERR(sa_dev->port[i].agent)) goto err; - INIT_WORK(&sa_dev->port[i].update_task, - update_sm_ah, &sa_dev->port[i]); + INIT_WORK(&sa_dev->port[i].update_task, update_sm_ah); } ib_set_client_data(device, &sa_client, sa_dev); @@ -1010,7 +1010,7 @@ static void ib_sa_add_one(struct ib_device *device) goto err; for (i = 0; i <= e - s; ++i) - update_sm_ah(&sa_dev->port[i]); + update_sm_ah(&sa_dev->port[i].update_task); return; diff --git a/drivers/infiniband/core/uverbs_mem.c b/drivers/infiniband/core/uverbs_mem.c index efe147dbeb4..db12cc0841d 100644 --- a/drivers/infiniband/core/uverbs_mem.c +++ b/drivers/infiniband/core/uverbs_mem.c @@ -179,9 +179,10 @@ void ib_umem_release(struct ib_device *dev, struct ib_umem *umem) up_write(¤t->mm->mmap_sem); } -static void ib_umem_account(void *work_ptr) +static void ib_umem_account(struct work_struct *_work) { - struct ib_umem_account_work *work = work_ptr; + struct ib_umem_account_work *work = + container_of(_work, struct ib_umem_account_work, work); down_write(&work->mm->mmap_sem); work->mm->locked_vm -= work->diff; @@ -216,7 +217,7 @@ void ib_umem_release_on_close(struct ib_device *dev, struct ib_umem *umem) return; } - INIT_WORK(&work->work, ib_umem_account, work); + INIT_WORK(&work->work, ib_umem_account); work->mm = mm; work->diff = PAGE_ALIGN(umem->length + umem->offset) >> PAGE_SHIFT; diff --git a/drivers/infiniband/hw/ipath/ipath_user_pages.c b/drivers/infiniband/hw/ipath/ipath_user_pages.c index 413754b1d8a..8536aeb96af 100644 --- a/drivers/infiniband/hw/ipath/ipath_user_pages.c +++ b/drivers/infiniband/hw/ipath/ipath_user_pages.c @@ -214,9 +214,10 @@ struct ipath_user_pages_work { unsigned long num_pages; }; -static void user_pages_account(void *ptr) +static void user_pages_account(struct work_struct *_work) { - struct ipath_user_pages_work *work = ptr; + struct ipath_user_pages_work *work = + container_of(_work, struct ipath_user_pages_work, work); down_write(&work->mm->mmap_sem); work->mm->locked_vm -= work->num_pages; @@ -242,7 +243,7 @@ void ipath_release_user_pages_on_close(struct page **p, size_t num_pages) goto bail; - INIT_WORK(&work->work, user_pages_account, work); + INIT_WORK(&work->work, user_pages_account); work->mm = mm; work->num_pages = num_pages; diff --git a/drivers/infiniband/hw/mthca/mthca_catas.c b/drivers/infiniband/hw/mthca/mthca_catas.c index cd044ea2dfa..e948158a28d 100644 --- a/drivers/infiniband/hw/mthca/mthca_catas.c +++ b/drivers/infiniband/hw/mthca/mthca_catas.c @@ -57,7 +57,7 @@ static int catas_reset_disable; module_param_named(catas_reset_disable, catas_reset_disable, int, 0644); MODULE_PARM_DESC(catas_reset_disable, "disable reset on catastrophic event if nonzero"); -static void catas_reset(void *work_ptr) +static void catas_reset(struct work_struct *work) { struct mthca_dev *dev, *tmpdev; LIST_HEAD(tlist); @@ -203,7 +203,7 @@ void mthca_stop_catas_poll(struct mthca_dev *dev) int __init mthca_catas_init(void) { - INIT_WORK(&catas_work, catas_reset, NULL); + INIT_WORK(&catas_work, catas_reset); catas_wq = create_singlethread_workqueue("mthca_catas"); if (!catas_wq) diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index 0b8a79d53a0..3b2ee0eba05 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -136,11 +136,11 @@ struct ipoib_dev_priv { struct list_head multicast_list; struct rb_root multicast_tree; - struct work_struct pkey_task; - struct work_struct mcast_task; + struct delayed_work pkey_task; + struct delayed_work mcast_task; struct work_struct flush_task; struct work_struct restart_task; - struct work_struct ah_reap_task; + struct delayed_work ah_reap_task; struct ib_device *ca; u8 port; @@ -254,13 +254,13 @@ int ipoib_add_pkey_attr(struct net_device *dev); void ipoib_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_ah *address, u32 qpn); -void ipoib_reap_ah(void *dev_ptr); +void ipoib_reap_ah(struct work_struct *work); void ipoib_flush_paths(struct net_device *dev); struct ipoib_dev_priv *ipoib_intf_alloc(const char *format); int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port); -void ipoib_ib_dev_flush(void *dev); +void ipoib_ib_dev_flush(struct work_struct *work); void ipoib_ib_dev_cleanup(struct net_device *dev); int ipoib_ib_dev_open(struct net_device *dev); @@ -271,10 +271,10 @@ int ipoib_ib_dev_stop(struct net_device *dev); int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port); void ipoib_dev_cleanup(struct net_device *dev); -void ipoib_mcast_join_task(void *dev_ptr); +void ipoib_mcast_join_task(struct work_struct *work); void ipoib_mcast_send(struct net_device *dev, void *mgid, struct sk_buff *skb); -void ipoib_mcast_restart_task(void *dev_ptr); +void ipoib_mcast_restart_task(struct work_struct *work); int ipoib_mcast_start_thread(struct net_device *dev); int ipoib_mcast_stop_thread(struct net_device *dev, int flush); @@ -312,7 +312,7 @@ void ipoib_event(struct ib_event_handler *handler, int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey); int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey); -void ipoib_pkey_poll(void *dev); +void ipoib_pkey_poll(struct work_struct *work); int ipoib_pkey_dev_delay_open(struct net_device *dev); #ifdef CONFIG_INFINIBAND_IPOIB_DEBUG diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index 8bf5e9ec7c9..f10fba5d326 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -400,10 +400,11 @@ static void __ipoib_reap_ah(struct net_device *dev) spin_unlock_irq(&priv->tx_lock); } -void ipoib_reap_ah(void *dev_ptr) +void ipoib_reap_ah(struct work_struct *work) { - struct net_device *dev = dev_ptr; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = + container_of(work, struct ipoib_dev_priv, ah_reap_task.work); + struct net_device *dev = priv->dev; __ipoib_reap_ah(dev); @@ -613,10 +614,11 @@ int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port) return 0; } -void ipoib_ib_dev_flush(void *_dev) +void ipoib_ib_dev_flush(struct work_struct *work) { - struct net_device *dev = (struct net_device *)_dev; - struct ipoib_dev_priv *priv = netdev_priv(dev), *cpriv; + struct ipoib_dev_priv *cpriv, *priv = + container_of(work, struct ipoib_dev_priv, flush_task); + struct net_device *dev = priv->dev; if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags) ) { ipoib_dbg(priv, "Not flushing - IPOIB_FLAG_INITIALIZED not set.\n"); @@ -638,14 +640,14 @@ void ipoib_ib_dev_flush(void *_dev) */ if (test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) { ipoib_ib_dev_up(dev); - ipoib_mcast_restart_task(dev); + ipoib_mcast_restart_task(&priv->restart_task); } mutex_lock(&priv->vlan_mutex); /* Flush any child interfaces too */ list_for_each_entry(cpriv, &priv->child_intfs, list) - ipoib_ib_dev_flush(cpriv->dev); + ipoib_ib_dev_flush(&cpriv->flush_task); mutex_unlock(&priv->vlan_mutex); } @@ -672,10 +674,11 @@ void ipoib_ib_dev_cleanup(struct net_device *dev) * change async notification is available. */ -void ipoib_pkey_poll(void *dev_ptr) +void ipoib_pkey_poll(struct work_struct *work) { - struct net_device *dev = dev_ptr; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = + container_of(work, struct ipoib_dev_priv, pkey_task.work); + struct net_device *dev = priv->dev; ipoib_pkey_dev_check_presence(dev); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 85522daeb94..71114a8c12b 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -933,11 +933,11 @@ static void ipoib_setup(struct net_device *dev) INIT_LIST_HEAD(&priv->dead_ahs); INIT_LIST_HEAD(&priv->multicast_list); - INIT_WORK(&priv->pkey_task, ipoib_pkey_poll, priv->dev); - INIT_WORK(&priv->mcast_task, ipoib_mcast_join_task, priv->dev); - INIT_WORK(&priv->flush_task, ipoib_ib_dev_flush, priv->dev); - INIT_WORK(&priv->restart_task, ipoib_mcast_restart_task, priv->dev); - INIT_WORK(&priv->ah_reap_task, ipoib_reap_ah, priv->dev); + INIT_DELAYED_WORK(&priv->pkey_task, ipoib_pkey_poll); + INIT_DELAYED_WORK(&priv->mcast_task, ipoib_mcast_join_task); + INIT_WORK(&priv->flush_task, ipoib_ib_dev_flush); + INIT_WORK(&priv->restart_task, ipoib_mcast_restart_task); + INIT_DELAYED_WORK(&priv->ah_reap_task, ipoib_reap_ah); } struct ipoib_dev_priv *ipoib_intf_alloc(const char *name) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 3faa1820f0e..f0a4fac1a21 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -399,7 +399,8 @@ static void ipoib_mcast_join_complete(int status, mcast->backoff = 1; mutex_lock(&mcast_mutex); if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) - queue_work(ipoib_workqueue, &priv->mcast_task); + queue_delayed_work(ipoib_workqueue, + &priv->mcast_task, 0); mutex_unlock(&mcast_mutex); complete(&mcast->done); return; @@ -435,7 +436,8 @@ static void ipoib_mcast_join_complete(int status, if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) { if (status == -ETIMEDOUT) - queue_work(ipoib_workqueue, &priv->mcast_task); + queue_delayed_work(ipoib_workqueue, &priv->mcast_task, + 0); else queue_delayed_work(ipoib_workqueue, &priv->mcast_task, mcast->backoff * HZ); @@ -517,10 +519,11 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast, mcast->query_id = ret; } -void ipoib_mcast_join_task(void *dev_ptr) +void ipoib_mcast_join_task(struct work_struct *work) { - struct net_device *dev = dev_ptr; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = + container_of(work, struct ipoib_dev_priv, mcast_task.work); + struct net_device *dev = priv->dev; if (!test_bit(IPOIB_MCAST_RUN, &priv->flags)) return; @@ -610,7 +613,7 @@ int ipoib_mcast_start_thread(struct net_device *dev) mutex_lock(&mcast_mutex); if (!test_and_set_bit(IPOIB_MCAST_RUN, &priv->flags)) - queue_work(ipoib_workqueue, &priv->mcast_task); + queue_delayed_work(ipoib_workqueue, &priv->mcast_task, 0); mutex_unlock(&mcast_mutex); spin_lock_irq(&priv->lock); @@ -818,10 +821,11 @@ void ipoib_mcast_dev_flush(struct net_device *dev) } } -void ipoib_mcast_restart_task(void *dev_ptr) +void ipoib_mcast_restart_task(struct work_struct *work) { - struct net_device *dev = dev_ptr; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = + container_of(work, struct ipoib_dev_priv, restart_task); + struct net_device *dev = priv->dev; struct dev_mc_list *mclist; struct ipoib_mcast *mcast, *tmcast; LIST_HEAD(remove_list); diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index 18a00003499..693b7700289 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -48,7 +48,7 @@ static void iser_cq_tasklet_fn(unsigned long data); static void iser_cq_callback(struct ib_cq *cq, void *cq_context); -static void iser_comp_error_worker(void *data); +static void iser_comp_error_worker(struct work_struct *work); static void iser_cq_event_callback(struct ib_event *cause, void *context) { @@ -480,8 +480,7 @@ int iser_conn_init(struct iser_conn **ibconn) init_waitqueue_head(&ib_conn->wait); atomic_set(&ib_conn->post_recv_buf_count, 0); atomic_set(&ib_conn->post_send_buf_count, 0); - INIT_WORK(&ib_conn->comperror_work, iser_comp_error_worker, - ib_conn); + INIT_WORK(&ib_conn->comperror_work, iser_comp_error_worker); INIT_LIST_HEAD(&ib_conn->conn_list); spin_lock_init(&ib_conn->lock); @@ -754,9 +753,10 @@ int iser_post_send(struct iser_desc *tx_desc) return ret_val; } -static void iser_comp_error_worker(void *data) +static void iser_comp_error_worker(struct work_struct *work) { - struct iser_conn *ib_conn = data; + struct iser_conn *ib_conn = + container_of(work, struct iser_conn, comperror_work); /* getting here when the state is UP means that the conn is being * * terminated asynchronously from the iSCSI layer's perspective. */ diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c index 4b09147f438..214f66195af 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.c +++ b/drivers/infiniband/ulp/srp/ib_srp.c @@ -390,9 +390,10 @@ static void srp_disconnect_target(struct srp_target_port *target) wait_for_completion(&target->done); } -static void srp_remove_work(void *target_ptr) +static void srp_remove_work(struct work_struct *work) { - struct srp_target_port *target = target_ptr; + struct srp_target_port *target = + container_of(work, struct srp_target_port, work); spin_lock_irq(target->scsi_host->host_lock); if (target->state != SRP_TARGET_DEAD) { @@ -575,7 +576,7 @@ err: spin_lock_irq(target->scsi_host->host_lock); if (target->state == SRP_TARGET_CONNECTING) { target->state = SRP_TARGET_DEAD; - INIT_WORK(&target->work, srp_remove_work, target); + INIT_WORK(&target->work, srp_remove_work); schedule_work(&target->work); } spin_unlock_irq(target->scsi_host->host_lock); diff --git a/drivers/input/keyboard/lkkbd.c b/drivers/input/keyboard/lkkbd.c index 979b93e33da..b7f049b45b6 100644 --- a/drivers/input/keyboard/lkkbd.c +++ b/drivers/input/keyboard/lkkbd.c @@ -572,9 +572,9 @@ lkkbd_event (struct input_dev *dev, unsigned int type, unsigned int code, * were in. */ static void -lkkbd_reinit (void *data) +lkkbd_reinit (struct work_struct *work) { - struct lkkbd *lk = data; + struct lkkbd *lk = container_of(work, struct lkkbd, tq); int division; unsigned char leds_on = 0; unsigned char leds_off = 0; @@ -651,7 +651,7 @@ lkkbd_connect (struct serio *serio, struct serio_driver *drv) lk->serio = serio; lk->dev = input_dev; - INIT_WORK (&lk->tq, lkkbd_reinit, lk); + INIT_WORK (&lk->tq, lkkbd_reinit); lk->bell_volume = bell_volume; lk->keyclick_volume = keyclick_volume; lk->ctrlclick_volume = ctrlclick_volume; diff --git a/drivers/input/keyboard/sunkbd.c b/drivers/input/keyboard/sunkbd.c index cac4781103c..6cd887c5eb0 100644 --- a/drivers/input/keyboard/sunkbd.c +++ b/drivers/input/keyboard/sunkbd.c @@ -208,9 +208,9 @@ static int sunkbd_initialize(struct sunkbd *sunkbd) * were in. */ -static void sunkbd_reinit(void *data) +static void sunkbd_reinit(struct work_struct *work) { - struct sunkbd *sunkbd = data; + struct sunkbd *sunkbd = container_of(work, struct sunkbd, tq); wait_event_interruptible_timeout(sunkbd->wait, sunkbd->reset >= 0, HZ); @@ -248,7 +248,7 @@ static int sunkbd_connect(struct serio *serio, struct serio_driver *drv) sunkbd->serio = serio; sunkbd->dev = input_dev; init_waitqueue_head(&sunkbd->wait); - INIT_WORK(&sunkbd->tq, sunkbd_reinit, sunkbd); + INIT_WORK(&sunkbd->tq, sunkbd_reinit); snprintf(sunkbd->phys, sizeof(sunkbd->phys), "%s/input0", serio->phys); serio_set_drvdata(serio, sunkbd); diff --git a/drivers/input/mouse/psmouse-base.c b/drivers/input/mouse/psmouse-base.c index 6f9b2c7cc9c..52bb2226ce2 100644 --- a/drivers/input/mouse/psmouse-base.c +++ b/drivers/input/mouse/psmouse-base.c @@ -888,9 +888,10 @@ static int psmouse_poll(struct psmouse *psmouse) * psmouse_resync() attempts to re-validate current protocol. */ -static void psmouse_resync(void *p) +static void psmouse_resync(struct work_struct *work) { - struct psmouse *psmouse = p, *parent = NULL; + struct psmouse *parent = NULL, *psmouse = + container_of(work, struct psmouse, resync_work); struct serio *serio = psmouse->ps2dev.serio; psmouse_ret_t rc = PSMOUSE_GOOD_DATA; int failed = 0, enabled = 0; @@ -1121,7 +1122,7 @@ static int psmouse_connect(struct serio *serio, struct serio_driver *drv) goto out; ps2_init(&psmouse->ps2dev, serio); - INIT_WORK(&psmouse->resync_work, psmouse_resync, psmouse); + INIT_WORK(&psmouse->resync_work, psmouse_resync); psmouse->dev = input_dev; snprintf(psmouse->phys, sizeof(psmouse->phys), "%s/input0", serio->phys); diff --git a/drivers/isdn/act2000/capi.c b/drivers/isdn/act2000/capi.c index 6ae6eb32211..946c38cf6f8 100644 --- a/drivers/isdn/act2000/capi.c +++ b/drivers/isdn/act2000/capi.c @@ -627,8 +627,10 @@ handle_ack(act2000_card *card, act2000_chan *chan, __u8 blocknr) { } void -actcapi_dispatch(act2000_card *card) +actcapi_dispatch(struct work_struct *work) { + struct act2000_card *card = + container_of(work, struct act2000_card, rcv_tq); struct sk_buff *skb; actcapi_msg *msg; __u16 ccmd; diff --git a/drivers/isdn/act2000/capi.h b/drivers/isdn/act2000/capi.h index 49f453c53c6..e55f6a931f6 100644 --- a/drivers/isdn/act2000/capi.h +++ b/drivers/isdn/act2000/capi.h @@ -356,7 +356,7 @@ extern int actcapi_connect_req(act2000_card *, act2000_chan *, char *, char, int extern void actcapi_select_b2_protocol_req(act2000_card *, act2000_chan *); extern void actcapi_disconnect_b3_req(act2000_card *, act2000_chan *); extern void actcapi_connect_resp(act2000_card *, act2000_chan *, __u8); -extern void actcapi_dispatch(act2000_card *); +extern void actcapi_dispatch(struct work_struct *); #ifdef DEBUG_MSG extern void actcapi_debug_msg(struct sk_buff *skb, int); #else diff --git a/drivers/isdn/act2000/module.c b/drivers/isdn/act2000/module.c index d89dcde4ead..90593e2ef87 100644 --- a/drivers/isdn/act2000/module.c +++ b/drivers/isdn/act2000/module.c @@ -192,8 +192,11 @@ act2000_set_msn(act2000_card *card, char *eazmsn) } static void -act2000_transmit(struct act2000_card *card) +act2000_transmit(struct work_struct *work) { + struct act2000_card *card = + container_of(work, struct act2000_card, snd_tq); + switch (card->bus) { case ACT2000_BUS_ISA: act2000_isa_send(card); @@ -207,8 +210,11 @@ act2000_transmit(struct act2000_card *card) } static void -act2000_receive(struct act2000_card *card) +act2000_receive(struct work_struct *work) { + struct act2000_card *card = + container_of(work, struct act2000_card, poll_tq); + switch (card->bus) { case ACT2000_BUS_ISA: act2000_isa_receive(card); @@ -227,7 +233,7 @@ act2000_poll(unsigned long data) act2000_card * card = (act2000_card *)data; unsigned long flags; - act2000_receive(card); + act2000_receive(&card->poll_tq); spin_lock_irqsave(&card->lock, flags); mod_timer(&card->ptimer, jiffies+3); spin_unlock_irqrestore(&card->lock, flags); @@ -578,9 +584,9 @@ act2000_alloccard(int bus, int port, int irq, char *id) skb_queue_head_init(&card->sndq); skb_queue_head_init(&card->rcvq); skb_queue_head_init(&card->ackq); - INIT_WORK(&card->snd_tq, (void *) (void *) act2000_transmit, card); - INIT_WORK(&card->rcv_tq, (void *) (void *) actcapi_dispatch, card); - INIT_WORK(&card->poll_tq, (void *) (void *) act2000_receive, card); + INIT_WORK(&card->snd_tq, act2000_transmit); + INIT_WORK(&card->rcv_tq, actcapi_dispatch); + INIT_WORK(&card->poll_tq, act2000_receive); init_timer(&card->ptimer); card->interface.owner = THIS_MODULE; card->interface.channels = ACT2000_BCH; diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c index 8c4fcb9027b..783a2552631 100644 --- a/drivers/isdn/capi/kcapi.c +++ b/drivers/isdn/capi/kcapi.c @@ -208,9 +208,10 @@ static void notify_down(u32 contr) } } -static void notify_handler(void *data) +static void notify_handler(struct work_struct *work) { - struct capi_notifier *np = data; + struct capi_notifier *np = + container_of(work, struct capi_notifier, work); switch (np->cmd) { case KCI_CONTRUP: @@ -235,7 +236,7 @@ static int notify_push(unsigned int cmd, u32 controller, u16 applid, u32 ncci) if (!np) return -ENOMEM; - INIT_WORK(&np->work, notify_handler, np); + INIT_WORK(&np->work, notify_handler); np->cmd = cmd; np->controller = controller; np->applid = applid; @@ -248,10 +249,11 @@ static int notify_push(unsigned int cmd, u32 controller, u16 applid, u32 ncci) /* -------- Receiver ------------------------------------------ */ -static void recv_handler(void *_ap) +static void recv_handler(struct work_struct *work) { struct sk_buff *skb; - struct capi20_appl *ap = (struct capi20_appl *) _ap; + struct capi20_appl *ap = + container_of(work, struct capi20_appl, recv_work); if ((!ap) || (ap->release_in_progress)) return; @@ -527,7 +529,7 @@ u16 capi20_register(struct capi20_appl *ap) ap->callback = NULL; init_MUTEX(&ap->recv_sem); skb_queue_head_init(&ap->recv_queue); - INIT_WORK(&ap->recv_work, recv_handler, (void *)ap); + INIT_WORK(&ap->recv_work, recv_handler); ap->release_in_progress = 0; write_unlock_irqrestore(&application_lock, flags); diff --git a/drivers/isdn/hisax/amd7930_fn.c b/drivers/isdn/hisax/amd7930_fn.c index bec59010bc6..3b19caeba25 100644 --- a/drivers/isdn/hisax/amd7930_fn.c +++ b/drivers/isdn/hisax/amd7930_fn.c @@ -232,9 +232,10 @@ Amd7930_new_ph(struct IsdnCardState *cs) static void -Amd7930_bh(struct IsdnCardState *cs) +Amd7930_bh(struct work_struct *work) { - + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *stptr; if (!cs) @@ -789,7 +790,7 @@ Amd7930_init(struct IsdnCardState *cs) void __devinit setup_Amd7930(struct IsdnCardState *cs) { - INIT_WORK(&cs->tqueue, (void *)(void *) Amd7930_bh, cs); + INIT_WORK(&cs->tqueue, Amd7930_bh); cs->dbusytimer.function = (void *) dbusy_timer_handler; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); diff --git a/drivers/isdn/hisax/config.c b/drivers/isdn/hisax/config.c index 785b08554fc..cede72cdbb3 100644 --- a/drivers/isdn/hisax/config.c +++ b/drivers/isdn/hisax/config.c @@ -1137,7 +1137,6 @@ static int checkcard(int cardnr, char *id, int *busy_flag, struct module *lockow cs->tx_skb = NULL; cs->tx_cnt = 0; cs->event = 0; - cs->tqueue.data = cs; skb_queue_head_init(&cs->rq); skb_queue_head_init(&cs->sq); @@ -1554,7 +1553,7 @@ static void hisax_b_l2l1(struct PStack *st, int pr, void *arg); static int hisax_cardmsg(struct IsdnCardState *cs, int mt, void *arg); static int hisax_bc_setstack(struct PStack *st, struct BCState *bcs); static void hisax_bc_close(struct BCState *bcs); -static void hisax_bh(struct IsdnCardState *cs); +static void hisax_bh(struct work_struct *work); static void EChannel_proc_rcv(struct hisax_d_if *d_if); int hisax_register(struct hisax_d_if *hisax_d_if, struct hisax_b_if *b_if[], @@ -1586,7 +1585,7 @@ int hisax_register(struct hisax_d_if *hisax_d_if, struct hisax_b_if *b_if[], hisax_d_if->cs = cs; cs->hw.hisax_d_if = hisax_d_if; cs->cardmsg = hisax_cardmsg; - INIT_WORK(&cs->tqueue, (void *)(void *)hisax_bh, cs); + INIT_WORK(&cs->tqueue, hisax_bh); cs->channel[0].d_st->l2.l2l1 = hisax_d_l2l1; for (i = 0; i < 2; i++) { cs->bcs[i].BC_SetStack = hisax_bc_setstack; @@ -1618,8 +1617,10 @@ static void hisax_sched_event(struct IsdnCardState *cs, int event) schedule_work(&cs->tqueue); } -static void hisax_bh(struct IsdnCardState *cs) +static void hisax_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *st; int pr; diff --git a/drivers/isdn/hisax/hfc4s8s_l1.c b/drivers/isdn/hisax/hfc4s8s_l1.c index d852c9d998b..de9b1a4d6ba 100644 --- a/drivers/isdn/hisax/hfc4s8s_l1.c +++ b/drivers/isdn/hisax/hfc4s8s_l1.c @@ -1083,8 +1083,9 @@ tx_b_frame(struct hfc4s8s_btype *bch) /* bottom half handler for interrupt */ /*************************************/ static void -hfc4s8s_bh(hfc4s8s_hw * hw) +hfc4s8s_bh(struct work_struct *work) { + hfc4s8s_hw *hw = container_of(work, hfc4s8s_hw, tqueue); u_char b; struct hfc4s8s_l1 *l1p; volatile u_char *fifo_stat; @@ -1550,7 +1551,7 @@ setup_instance(hfc4s8s_hw * hw) goto out; } - INIT_WORK(&hw->tqueue, (void *) (void *) hfc4s8s_bh, hw); + INIT_WORK(&hw->tqueue, hfc4s8s_bh); if (request_irq (hw->irq, hfc4s8s_interrupt, IRQF_SHARED, hw->card_name, hw)) { diff --git a/drivers/isdn/hisax/hfc_2bds0.c b/drivers/isdn/hisax/hfc_2bds0.c index 6360e821472..8d9864453a2 100644 --- a/drivers/isdn/hisax/hfc_2bds0.c +++ b/drivers/isdn/hisax/hfc_2bds0.c @@ -549,10 +549,11 @@ setstack_2b(struct PStack *st, struct BCState *bcs) } static void -hfcd_bh(struct IsdnCardState *cs) +hfcd_bh(struct work_struct *work) { - if (!cs) - return; + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); + if (test_and_clear_bit(D_L1STATECHANGE, &cs->event)) { switch (cs->dc.hfcd.ph_state) { case (0): @@ -1072,5 +1073,5 @@ set_cs_func(struct IsdnCardState *cs) cs->dbusytimer.function = (void *) hfc_dbusy_timer; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); - INIT_WORK(&cs->tqueue, (void *)(void *) hfcd_bh, cs); + INIT_WORK(&cs->tqueue, hfcd_bh); } diff --git a/drivers/isdn/hisax/hfc_pci.c b/drivers/isdn/hisax/hfc_pci.c index 93f60b56351..5db0a85b827 100644 --- a/drivers/isdn/hisax/hfc_pci.c +++ b/drivers/isdn/hisax/hfc_pci.c @@ -1506,8 +1506,10 @@ setstack_2b(struct PStack *st, struct BCState *bcs) /* handle L1 state changes */ /***************************/ static void -hfcpci_bh(struct IsdnCardState *cs) +hfcpci_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); u_long flags; // struct PStack *stptr; @@ -1722,7 +1724,7 @@ setup_hfcpci(struct IsdnCard *card) Write_hfc(cs, HFCPCI_INT_M2, cs->hw.hfcpci.int_m2); /* At this point the needed PCI config is done */ /* fifos are still not enabled */ - INIT_WORK(&cs->tqueue, (void *)(void *) hfcpci_bh, cs); + INIT_WORK(&cs->tqueue, hfcpci_bh); cs->setstack_d = setstack_hfcpci; cs->BC_Send_Data = &hfcpci_send_data; cs->readisac = NULL; diff --git a/drivers/isdn/hisax/hfc_sx.c b/drivers/isdn/hisax/hfc_sx.c index 954d1536db1..4fd09d21a27 100644 --- a/drivers/isdn/hisax/hfc_sx.c +++ b/drivers/isdn/hisax/hfc_sx.c @@ -1251,8 +1251,10 @@ setstack_2b(struct PStack *st, struct BCState *bcs) /* handle L1 state changes */ /***************************/ static void -hfcsx_bh(struct IsdnCardState *cs) +hfcsx_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); u_long flags; if (!cs) @@ -1499,7 +1501,7 @@ setup_hfcsx(struct IsdnCard *card) cs->dbusytimer.function = (void *) hfcsx_dbusy_timer; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); - INIT_WORK(&cs->tqueue, (void *)(void *) hfcsx_bh, cs); + INIT_WORK(&cs->tqueue, hfcsx_bh); cs->readisac = NULL; cs->writeisac = NULL; cs->readisacfifo = NULL; diff --git a/drivers/isdn/hisax/icc.c b/drivers/isdn/hisax/icc.c index da706925d54..682cac32f25 100644 --- a/drivers/isdn/hisax/icc.c +++ b/drivers/isdn/hisax/icc.c @@ -77,8 +77,10 @@ icc_new_ph(struct IsdnCardState *cs) } static void -icc_bh(struct IsdnCardState *cs) +icc_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *stptr; if (!cs) @@ -674,7 +676,7 @@ clear_pending_icc_ints(struct IsdnCardState *cs) void __devinit setup_icc(struct IsdnCardState *cs) { - INIT_WORK(&cs->tqueue, (void *)(void *) icc_bh, cs); + INIT_WORK(&cs->tqueue, icc_bh); cs->dbusytimer.function = (void *) dbusy_timer_handler; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); diff --git a/drivers/isdn/hisax/isac.c b/drivers/isdn/hisax/isac.c index 282f349408b..4e9f23803da 100644 --- a/drivers/isdn/hisax/isac.c +++ b/drivers/isdn/hisax/isac.c @@ -81,8 +81,10 @@ isac_new_ph(struct IsdnCardState *cs) } static void -isac_bh(struct IsdnCardState *cs) +isac_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *stptr; if (!cs) @@ -674,7 +676,7 @@ clear_pending_isac_ints(struct IsdnCardState *cs) void __devinit setup_isac(struct IsdnCardState *cs) { - INIT_WORK(&cs->tqueue, (void *)(void *) isac_bh, cs); + INIT_WORK(&cs->tqueue, isac_bh); cs->dbusytimer.function = (void *) dbusy_timer_handler; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); diff --git a/drivers/isdn/hisax/isar.c b/drivers/isdn/hisax/isar.c index 674af673ff9..6f1a6583b17 100644 --- a/drivers/isdn/hisax/isar.c +++ b/drivers/isdn/hisax/isar.c @@ -437,8 +437,10 @@ extern void BChannel_bh(struct BCState *); #define B_LL_OK 10 static void -isar_bh(struct BCState *bcs) +isar_bh(struct work_struct *work) { + struct BCState *bcs = container_of(work, struct BCState, tqueue); + BChannel_bh(bcs); if (test_and_clear_bit(B_LL_NOCARRIER, &bcs->event)) ll_deliver_faxstat(bcs, ISDN_FAX_CLASS1_NOCARR); @@ -1580,7 +1582,7 @@ isar_setup(struct IsdnCardState *cs) cs->bcs[i].mode = 0; cs->bcs[i].hw.isar.dpath = i + 1; modeisar(&cs->bcs[i], 0, 0); - INIT_WORK(&cs->bcs[i].tqueue, (void *)(void *) isar_bh, &cs->bcs[i]); + INIT_WORK(&cs->bcs[i].tqueue, isar_bh); } } diff --git a/drivers/isdn/hisax/isdnl1.c b/drivers/isdn/hisax/isdnl1.c index bab35688648..a14204ec88e 100644 --- a/drivers/isdn/hisax/isdnl1.c +++ b/drivers/isdn/hisax/isdnl1.c @@ -315,8 +315,10 @@ BChannel_proc_ack(struct BCState *bcs) } void -BChannel_bh(struct BCState *bcs) +BChannel_bh(struct work_struct *work) { + struct BCState *bcs = container_of(work, struct BCState, tqueue); + if (!bcs) return; if (test_and_clear_bit(B_RCVBUFREADY, &bcs->event)) @@ -362,7 +364,7 @@ init_bcstate(struct IsdnCardState *cs, int bc) bcs->cs = cs; bcs->channel = bc; - INIT_WORK(&bcs->tqueue, (void *)(void *) BChannel_bh, bcs); + INIT_WORK(&bcs->tqueue, BChannel_bh); spin_lock_init(&bcs->aclock); bcs->BC_SetStack = NULL; bcs->BC_Close = NULL; diff --git a/drivers/isdn/hisax/w6692.c b/drivers/isdn/hisax/w6692.c index 1655341797a..3aeceaf9769 100644 --- a/drivers/isdn/hisax/w6692.c +++ b/drivers/isdn/hisax/w6692.c @@ -101,8 +101,10 @@ W6692_new_ph(struct IsdnCardState *cs) } static void -W6692_bh(struct IsdnCardState *cs) +W6692_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *stptr; if (!cs) @@ -1070,7 +1072,7 @@ setup_w6692(struct IsdnCard *card) id_list[cs->subtyp].card_name, cs->irq, cs->hw.w6692.iobase); - INIT_WORK(&cs->tqueue, (void *)(void *) W6692_bh, cs); + INIT_WORK(&cs->tqueue, W6692_bh); cs->readW6692 = &ReadW6692; cs->writeW6692 = &WriteW6692; cs->readisacfifo = &ReadISACfifo; diff --git a/drivers/isdn/i4l/isdn_net.c b/drivers/isdn/i4l/isdn_net.c index 1f8d6ae66b4..2e4daebfb7e 100644 --- a/drivers/isdn/i4l/isdn_net.c +++ b/drivers/isdn/i4l/isdn_net.c @@ -984,9 +984,9 @@ void isdn_net_write_super(isdn_net_local *lp, struct sk_buff *skb) /* * called from tq_immediate */ -static void isdn_net_softint(void *private) +static void isdn_net_softint(struct work_struct *work) { - isdn_net_local *lp = private; + isdn_net_local *lp = container_of(work, isdn_net_local, tqueue); struct sk_buff *skb; spin_lock_bh(&lp->xmit_lock); @@ -2596,7 +2596,7 @@ isdn_net_new(char *name, struct net_device *master) netdev->local->netdev = netdev; netdev->local->next = netdev->local; - INIT_WORK(&netdev->local->tqueue, (void *)(void *) isdn_net_softint, netdev->local); + INIT_WORK(&netdev->local->tqueue, isdn_net_softint); spin_lock_init(&netdev->local->xmit_lock); netdev->local->isdn_device = -1; diff --git a/drivers/isdn/pcbit/drv.c b/drivers/isdn/pcbit/drv.c index 6ead5e1508b..1966f3410a1 100644 --- a/drivers/isdn/pcbit/drv.c +++ b/drivers/isdn/pcbit/drv.c @@ -68,8 +68,6 @@ static void pcbit_set_msn(struct pcbit_dev *dev, char *list); static int pcbit_check_msn(struct pcbit_dev *dev, char *msn); -extern void pcbit_deliver(void * data); - int pcbit_init_dev(int board, int mem_base, int irq) { struct pcbit_dev *dev; @@ -129,7 +127,7 @@ int pcbit_init_dev(int board, int mem_base, int irq) memset(dev->b2, 0, sizeof(struct pcbit_chan)); dev->b2->id = 1; - INIT_WORK(&dev->qdelivery, pcbit_deliver, dev); + INIT_WORK(&dev->qdelivery, pcbit_deliver); /* * interrupts diff --git a/drivers/isdn/pcbit/layer2.c b/drivers/isdn/pcbit/layer2.c index 937fd212038..0c9f6df873f 100644 --- a/drivers/isdn/pcbit/layer2.c +++ b/drivers/isdn/pcbit/layer2.c @@ -67,7 +67,6 @@ extern void pcbit_l3_receive(struct pcbit_dev *dev, ulong msg, * Prototypes */ -void pcbit_deliver(void *data); static void pcbit_transmit(struct pcbit_dev *dev); static void pcbit_recv_ack(struct pcbit_dev *dev, unsigned char ack); @@ -299,11 +298,12 @@ pcbit_transmit(struct pcbit_dev *dev) */ void -pcbit_deliver(void *data) +pcbit_deliver(struct work_struct *work) { struct frame_buf *frame; unsigned long flags, msg; - struct pcbit_dev *dev = (struct pcbit_dev *) data; + struct pcbit_dev *dev = + container_of(work, struct pcbit_dev, qdelivery); spin_lock_irqsave(&dev->lock, flags); diff --git a/drivers/isdn/pcbit/pcbit.h b/drivers/isdn/pcbit/pcbit.h index 388bacefd23..19c18e88ff1 100644 --- a/drivers/isdn/pcbit/pcbit.h +++ b/drivers/isdn/pcbit/pcbit.h @@ -166,4 +166,6 @@ struct pcbit_ioctl { #define L2_RUNNING 5 #define L2_ERROR 6 +extern void pcbit_deliver(struct work_struct *work); + #endif diff --git a/drivers/macintosh/smu.c b/drivers/macintosh/smu.c index ade25b3fbb3..4871158aca3 100644 --- a/drivers/macintosh/smu.c +++ b/drivers/macintosh/smu.c @@ -600,7 +600,7 @@ core_initcall(smu_late_init); * sysfs visibility */ -static void smu_expose_childs(void *unused) +static void smu_expose_childs(struct work_struct *unused) { struct device_node *np; @@ -610,7 +610,7 @@ static void smu_expose_childs(void *unused) &smu->of_dev->dev); } -static DECLARE_WORK(smu_expose_childs_work, smu_expose_childs, NULL); +static DECLARE_WORK(smu_expose_childs_work, smu_expose_childs); static int smu_platform_probe(struct of_device* dev, const struct of_device_id *match) diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 08a40f4e4f6..ed2d4ef27fd 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -458,11 +458,11 @@ static void dec_pending(struct crypt_io *io, int error) * interrupt context. */ static struct workqueue_struct *_kcryptd_workqueue; -static void kcryptd_do_work(void *data); +static void kcryptd_do_work(struct work_struct *work); static void kcryptd_queue_io(struct crypt_io *io) { - INIT_WORK(&io->work, kcryptd_do_work, io); + INIT_WORK(&io->work, kcryptd_do_work); queue_work(_kcryptd_workqueue, &io->work); } @@ -618,9 +618,9 @@ static void process_read_endio(struct crypt_io *io) dec_pending(io, crypt_convert(cc, &ctx)); } -static void kcryptd_do_work(void *data) +static void kcryptd_do_work(struct work_struct *work) { - struct crypt_io *io = data; + struct crypt_io *io = container_of(work, struct crypt_io, work); if (io->post_process) process_read_endio(io); diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index d754e0bc6e9..e77ee6fd104 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -104,8 +104,8 @@ typedef int (*action_fn) (struct pgpath *pgpath); static kmem_cache_t *_mpio_cache; struct workqueue_struct *kmultipathd; -static void process_queued_ios(void *data); -static void trigger_event(void *data); +static void process_queued_ios(struct work_struct *work); +static void trigger_event(struct work_struct *work); /*----------------------------------------------- @@ -173,8 +173,8 @@ static struct multipath *alloc_multipath(struct dm_target *ti) INIT_LIST_HEAD(&m->priority_groups); spin_lock_init(&m->lock); m->queue_io = 1; - INIT_WORK(&m->process_queued_ios, process_queued_ios, m); - INIT_WORK(&m->trigger_event, trigger_event, m); + INIT_WORK(&m->process_queued_ios, process_queued_ios); + INIT_WORK(&m->trigger_event, trigger_event); m->mpio_pool = mempool_create_slab_pool(MIN_IOS, _mpio_cache); if (!m->mpio_pool) { kfree(m); @@ -379,9 +379,10 @@ static void dispatch_queued_ios(struct multipath *m) } } -static void process_queued_ios(void *data) +static void process_queued_ios(struct work_struct *work) { - struct multipath *m = (struct multipath *) data; + struct multipath *m = + container_of(work, struct multipath, process_queued_ios); struct hw_handler *hwh = &m->hw_handler; struct pgpath *pgpath = NULL; unsigned init_required = 0, must_queue = 1; @@ -421,9 +422,10 @@ out: * An event is triggered whenever a path is taken out of use. * Includes path failure and PG bypass. */ -static void trigger_event(void *data) +static void trigger_event(struct work_struct *work) { - struct multipath *m = (struct multipath *) data; + struct multipath *m = + container_of(work, struct multipath, trigger_event); dm_table_event(m->ti->table); } diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index 48a653b3f51..fc8cbb168e3 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -883,7 +883,7 @@ static void do_mirror(struct mirror_set *ms) do_writes(ms, &writes); } -static void do_work(void *ignored) +static void do_work(struct work_struct *ignored) { struct mirror_set *ms; @@ -1269,7 +1269,7 @@ static int __init dm_mirror_init(void) dm_dirty_log_exit(); return r; } - INIT_WORK(&_kmirrord_work, do_work, NULL); + INIT_WORK(&_kmirrord_work, do_work); r = dm_register_target(&mirror_target); if (r < 0) { diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 5281e009407..91c7aa1fed0 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -40,7 +40,7 @@ #define SNAPSHOT_PAGES 256 struct workqueue_struct *ksnapd; -static void flush_queued_bios(void *data); +static void flush_queued_bios(struct work_struct *work); struct pending_exception { struct exception e; @@ -528,7 +528,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) } bio_list_init(&s->queued_bios); - INIT_WORK(&s->queued_bios_work, flush_queued_bios, s); + INIT_WORK(&s->queued_bios_work, flush_queued_bios); /* Add snapshot to the list of snapshots for this origin */ /* Exceptions aren't triggered till snapshot_resume() is called */ @@ -603,9 +603,10 @@ static void flush_bios(struct bio *bio) } } -static void flush_queued_bios(void *data) +static void flush_queued_bios(struct work_struct *work) { - struct dm_snapshot *s = (struct dm_snapshot *) data; + struct dm_snapshot *s = + container_of(work, struct dm_snapshot, queued_bios_work); struct bio *queued_bios; unsigned long flags; diff --git a/drivers/md/kcopyd.c b/drivers/md/kcopyd.c index f1db6eff485..b3c01496c73 100644 --- a/drivers/md/kcopyd.c +++ b/drivers/md/kcopyd.c @@ -417,7 +417,7 @@ static int process_jobs(struct list_head *jobs, int (*fn) (struct kcopyd_job *)) /* * kcopyd does this every time it's woken up. */ -static void do_work(void *ignored) +static void do_work(struct work_struct *ignored) { /* * The order that these are called is *very* important. @@ -628,7 +628,7 @@ static int kcopyd_init(void) } kcopyd_clients++; - INIT_WORK(&_kcopyd_work, do_work, NULL); + INIT_WORK(&_kcopyd_work, do_work); mutex_unlock(&kcopyd_init_lock); return 0; } diff --git a/drivers/media/dvb/b2c2/flexcop-pci.c b/drivers/media/dvb/b2c2/flexcop-pci.c index 06893243f3d..6e166801505 100644 --- a/drivers/media/dvb/b2c2/flexcop-pci.c +++ b/drivers/media/dvb/b2c2/flexcop-pci.c @@ -63,7 +63,7 @@ struct flexcop_pci { unsigned long last_irq; - struct work_struct irq_check_work; + struct delayed_work irq_check_work; struct flexcop_device *fc_dev; }; @@ -97,9 +97,10 @@ static int flexcop_pci_write_ibi_reg(struct flexcop_device *fc, flexcop_ibi_regi return 0; } -static void flexcop_pci_irq_check_work(void *data) +static void flexcop_pci_irq_check_work(struct work_struct *work) { - struct flexcop_pci *fc_pci = data; + struct flexcop_pci *fc_pci = + container_of(work, struct flexcop_pci, irq_check_work.work); struct flexcop_device *fc = fc_pci->fc_dev; flexcop_ibi_value v = fc->read_ibi_reg(fc,sram_dest_reg_714); @@ -371,7 +372,7 @@ static int flexcop_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e if ((ret = flexcop_pci_dma_init(fc_pci)) != 0) goto err_fc_exit; - INIT_WORK(&fc_pci->irq_check_work, flexcop_pci_irq_check_work, fc_pci); + INIT_DELAYED_WORK(&fc_pci->irq_check_work, flexcop_pci_irq_check_work); return ret; diff --git a/drivers/media/dvb/cinergyT2/cinergyT2.c b/drivers/media/dvb/cinergyT2/cinergyT2.c index ff7d4f56ced..dd0bcbe140b 100644 --- a/drivers/media/dvb/cinergyT2/cinergyT2.c +++ b/drivers/media/dvb/cinergyT2/cinergyT2.c @@ -127,7 +127,7 @@ struct cinergyt2 { struct dvbt_set_parameters_msg param; struct dvbt_get_status_msg status; - struct work_struct query_work; + struct delayed_work query_work; wait_queue_head_t poll_wq; int pending_fe_events; @@ -141,7 +141,7 @@ struct cinergyt2 { #ifdef ENABLE_RC struct input_dev *rc_input_dev; char phys[64]; - struct work_struct rc_query_work; + struct delayed_work rc_query_work; int rc_input_event; u32 rc_last_code; unsigned long last_event_jiffies; @@ -724,9 +724,10 @@ static struct dvb_device cinergyt2_fe_template = { #ifdef ENABLE_RC -static void cinergyt2_query_rc (void *data) +static void cinergyt2_query_rc (struct work_struct *work) { - struct cinergyt2 *cinergyt2 = data; + struct cinergyt2 *cinergyt2 = + container_of(work, struct cinergyt2, rc_query_work.work); char buf[1] = { CINERGYT2_EP1_GET_RC_EVENTS }; struct cinergyt2_rc_event rc_events[12]; int n, len, i; @@ -807,7 +808,7 @@ static int cinergyt2_register_rc(struct cinergyt2 *cinergyt2) strlcat(cinergyt2->phys, "/input0", sizeof(cinergyt2->phys)); cinergyt2->rc_input_event = KEY_MAX; cinergyt2->rc_last_code = ~0; - INIT_WORK(&cinergyt2->rc_query_work, cinergyt2_query_rc, cinergyt2); + INIT_DELAYED_WORK(&cinergyt2->rc_query_work, cinergyt2_query_rc); input_dev->name = DRIVER_NAME " remote control"; input_dev->phys = cinergyt2->phys; @@ -848,9 +849,10 @@ static inline void cinergyt2_resume_rc(struct cinergyt2 *cinergyt2) { } #endif /* ENABLE_RC */ -static void cinergyt2_query (void *data) +static void cinergyt2_query (struct work_struct *work) { - struct cinergyt2 *cinergyt2 = (struct cinergyt2 *) data; + struct cinergyt2 *cinergyt2 = + container_of(work, struct cinergyt2, query_work.work); char cmd [] = { CINERGYT2_EP1_GET_TUNER_STATUS }; struct dvbt_get_status_msg *s = &cinergyt2->status; uint8_t lock_bits; @@ -894,7 +896,7 @@ static int cinergyt2_probe (struct usb_interface *intf, mutex_init(&cinergyt2->sem); init_waitqueue_head (&cinergyt2->poll_wq); - INIT_WORK(&cinergyt2->query_work, cinergyt2_query, cinergyt2); + INIT_DELAYED_WORK(&cinergyt2->query_work, cinergyt2_query); cinergyt2->udev = interface_to_usbdev(intf); cinergyt2->param.cmd = CINERGYT2_EP1_SET_TUNER_PARAMETERS; diff --git a/drivers/media/dvb/dvb-core/dvb_net.c b/drivers/media/dvb/dvb-core/dvb_net.c index 8859ab74f0f..ebf4dc5190f 100644 --- a/drivers/media/dvb/dvb-core/dvb_net.c +++ b/drivers/media/dvb/dvb-core/dvb_net.c @@ -127,6 +127,7 @@ struct dvb_net_priv { int in_use; struct net_device_stats stats; u16 pid; + struct net_device *net; struct dvb_net *host; struct dmx_demux *demux; struct dmx_section_feed *secfeed; @@ -1123,10 +1124,11 @@ static int dvb_set_mc_filter (struct net_device *dev, struct dev_mc_list *mc) } -static void wq_set_multicast_list (void *data) +static void wq_set_multicast_list (struct work_struct *work) { - struct net_device *dev = data; - struct dvb_net_priv *priv = dev->priv; + struct dvb_net_priv *priv = + container_of(work, struct dvb_net_priv, set_multicast_list_wq); + struct net_device *dev = priv->net; dvb_net_feed_stop(dev); priv->rx_mode = RX_MODE_UNI; @@ -1167,9 +1169,11 @@ static void dvb_net_set_multicast_list (struct net_device *dev) } -static void wq_restart_net_feed (void *data) +static void wq_restart_net_feed (struct work_struct *work) { - struct net_device *dev = data; + struct dvb_net_priv *priv = + container_of(work, struct dvb_net_priv, restart_net_feed_wq); + struct net_device *dev = priv->net; if (netif_running(dev)) { dvb_net_feed_stop(dev); @@ -1276,6 +1280,7 @@ static int dvb_net_add_if(struct dvb_net *dvbnet, u16 pid, u8 feedtype) dvbnet->device[if_num] = net; priv = net->priv; + priv->net = net; priv->demux = dvbnet->demux; priv->pid = pid; priv->rx_mode = RX_MODE_UNI; @@ -1284,8 +1289,8 @@ static int dvb_net_add_if(struct dvb_net *dvbnet, u16 pid, u8 feedtype) priv->feedtype = feedtype; reset_ule(priv); - INIT_WORK(&priv->set_multicast_list_wq, wq_set_multicast_list, net); - INIT_WORK(&priv->restart_net_feed_wq, wq_restart_net_feed, net); + INIT_WORK(&priv->set_multicast_list_wq, wq_set_multicast_list); + INIT_WORK(&priv->restart_net_feed_wq, wq_restart_net_feed); mutex_init(&priv->mutex); net->base_addr = pid; diff --git a/drivers/media/dvb/dvb-usb/dvb-usb-remote.c b/drivers/media/dvb/dvb-usb/dvb-usb-remote.c index 0a3a0b6c235..794e4471561 100644 --- a/drivers/media/dvb/dvb-usb/dvb-usb-remote.c +++ b/drivers/media/dvb/dvb-usb/dvb-usb-remote.c @@ -13,9 +13,10 @@ * * TODO: Fix the repeat rate of the input device. */ -static void dvb_usb_read_remote_control(void *data) +static void dvb_usb_read_remote_control(struct work_struct *work) { - struct dvb_usb_device *d = data; + struct dvb_usb_device *d = + container_of(work, struct dvb_usb_device, rc_query_work.work); u32 event; int state; @@ -128,7 +129,7 @@ int dvb_usb_remote_init(struct dvb_usb_device *d) input_register_device(d->rc_input_dev); - INIT_WORK(&d->rc_query_work, dvb_usb_read_remote_control, d); + INIT_DELAYED_WORK(&d->rc_query_work, dvb_usb_read_remote_control); info("schedule remote query interval to %d msecs.", d->props.rc_interval); schedule_delayed_work(&d->rc_query_work,msecs_to_jiffies(d->props.rc_interval)); diff --git a/drivers/media/dvb/dvb-usb/dvb-usb.h b/drivers/media/dvb/dvb-usb/dvb-usb.h index 376c45a8e77..0d721731a52 100644 --- a/drivers/media/dvb/dvb-usb/dvb-usb.h +++ b/drivers/media/dvb/dvb-usb/dvb-usb.h @@ -369,7 +369,7 @@ struct dvb_usb_device { /* remote control */ struct input_dev *rc_input_dev; char rc_phys[64]; - struct work_struct rc_query_work; + struct delayed_work rc_query_work; u32 last_event; int last_state; diff --git a/drivers/media/video/cpia_pp.c b/drivers/media/video/cpia_pp.c index 41f4b8d1755..b12cec94f4c 100644 --- a/drivers/media/video/cpia_pp.c +++ b/drivers/media/video/cpia_pp.c @@ -82,6 +82,8 @@ struct pp_cam_entry { struct pardevice *pdev; struct parport *port; struct work_struct cb_task; + void (*cb_func)(void *cbdata); + void *cb_data; int open_count; wait_queue_head_t wq_stream; /* image state flags */ @@ -130,6 +132,20 @@ static void cpia_parport_disable_irq( struct parport *port ) { #define PARPORT_CHUNK_SIZE PAGE_SIZE +static void cpia_pp_run_callback(struct work_struct *work) +{ + void (*cb_func)(void *cbdata); + void *cb_data; + struct pp_cam_entry *cam; + + cam = container_of(work, struct pp_cam_entry, cb_task); + cb_func = cam->cb_func; + cb_data = cam->cb_data; + work_release(work); + + cb_func(cb_data); +} + /**************************************************************************** * * CPiA-specific low-level parport functions for nibble uploads @@ -664,7 +680,9 @@ static int cpia_pp_registerCallback(void *privdata, void (*cb)(void *cbdata), vo int retval = 0; if(cam->port->irq != PARPORT_IRQ_NONE) { - INIT_WORK(&cam->cb_task, cb, cbdata); + cam->cb_func = cb; + cam->cb_data = cbdata; + INIT_WORK_NAR(&cam->cb_task, cpia_pp_run_callback); } else { retval = -1; } diff --git a/drivers/media/video/cx88/cx88-input.c b/drivers/media/video/cx88/cx88-input.c index 57e1c024a54..e60a0a52e4b 100644 --- a/drivers/media/video/cx88/cx88-input.c +++ b/drivers/media/video/cx88/cx88-input.c @@ -145,9 +145,9 @@ static void ir_timer(unsigned long data) schedule_work(&ir->work); } -static void cx88_ir_work(void *data) +static void cx88_ir_work(struct work_struct *work) { - struct cx88_IR *ir = data; + struct cx88_IR *ir = container_of(work, struct cx88_IR, work); unsigned long timeout; cx88_ir_handle_key(ir); @@ -308,7 +308,7 @@ int cx88_ir_init(struct cx88_core *core, struct pci_dev *pci) core->ir = ir; if (ir->polling) { - INIT_WORK(&ir->work, cx88_ir_work, ir); + INIT_WORK(&ir->work, cx88_ir_work); init_timer(&ir->timer); ir->timer.function = ir_timer; ir->timer.data = (unsigned long)ir; diff --git a/drivers/media/video/ir-kbd-i2c.c b/drivers/media/video/ir-kbd-i2c.c index 1457b160222..ab87e7bfe84 100644 --- a/drivers/media/video/ir-kbd-i2c.c +++ b/drivers/media/video/ir-kbd-i2c.c @@ -268,9 +268,9 @@ static void ir_timer(unsigned long data) schedule_work(&ir->work); } -static void ir_work(void *data) +static void ir_work(struct work_struct *work) { - struct IR_i2c *ir = data; + struct IR_i2c *ir = container_of(work, struct IR_i2c, work); ir_key_poll(ir); mod_timer(&ir->timer, jiffies+HZ/10); } @@ -400,7 +400,7 @@ static int ir_attach(struct i2c_adapter *adap, int addr, ir->input->name,ir->input->phys,adap->name); /* start polling via eventd */ - INIT_WORK(&ir->work, ir_work, ir); + INIT_WORK(&ir->work, ir_work); init_timer(&ir->timer); ir->timer.function = ir_timer; ir->timer.data = (unsigned long)ir; diff --git a/drivers/media/video/pvrusb2/pvrusb2-context.c b/drivers/media/video/pvrusb2/pvrusb2-context.c index f129f316d20..cf129746205 100644 --- a/drivers/media/video/pvrusb2/pvrusb2-context.c +++ b/drivers/media/video/pvrusb2/pvrusb2-context.c @@ -45,16 +45,21 @@ static void pvr2_context_trigger_poll(struct pvr2_context *mp) } -static void pvr2_context_poll(struct pvr2_context *mp) +static void pvr2_context_poll(struct work_struct *work) { + struct pvr2_context *mp = + container_of(work, struct pvr2_context, workpoll); pvr2_context_enter(mp); do { pvr2_hdw_poll(mp->hdw); } while (0); pvr2_context_exit(mp); } -static void pvr2_context_setup(struct pvr2_context *mp) +static void pvr2_context_setup(struct work_struct *work) { + struct pvr2_context *mp = + container_of(work, struct pvr2_context, workinit); + pvr2_context_enter(mp); do { if (!pvr2_hdw_dev_ok(mp->hdw)) break; pvr2_hdw_setup(mp->hdw); @@ -92,8 +97,8 @@ struct pvr2_context *pvr2_context_create( } mp->workqueue = create_singlethread_workqueue("pvrusb2"); - INIT_WORK(&mp->workinit,(void (*)(void*))pvr2_context_setup,mp); - INIT_WORK(&mp->workpoll,(void (*)(void*))pvr2_context_poll,mp); + INIT_WORK(&mp->workinit, pvr2_context_setup); + INIT_WORK(&mp->workpoll, pvr2_context_poll); queue_work(mp->workqueue,&mp->workinit); done: return mp; diff --git a/drivers/media/video/saa6588.c b/drivers/media/video/saa6588.c index a81285ca7d5..8ba05c214ca 100644 --- a/drivers/media/video/saa6588.c +++ b/drivers/media/video/saa6588.c @@ -322,9 +322,9 @@ static void saa6588_timer(unsigned long data) schedule_work(&s->work); } -static void saa6588_work(void *data) +static void saa6588_work(struct work_struct *work) { - struct saa6588 *s = (struct saa6588 *)data; + struct saa6588 *s = container_of(work, struct saa6588, work); saa6588_i2c_poll(s); mod_timer(&s->timer, jiffies + msecs_to_jiffies(20)); @@ -417,7 +417,7 @@ static int saa6588_attach(struct i2c_adapter *adap, int addr, int kind) saa6588_configure(s); /* start polling via eventd */ - INIT_WORK(&s->work, saa6588_work, s); + INIT_WORK(&s->work, saa6588_work); init_timer(&s->timer); s->timer.function = saa6588_timer; s->timer.data = (unsigned long)s; diff --git a/drivers/media/video/saa7134/saa7134-empress.c b/drivers/media/video/saa7134/saa7134-empress.c index 65d044086ce..daaae870a2c 100644 --- a/drivers/media/video/saa7134/saa7134-empress.c +++ b/drivers/media/video/saa7134/saa7134-empress.c @@ -343,9 +343,10 @@ static struct video_device saa7134_empress_template = .minor = -1, }; -static void empress_signal_update(void* data) +static void empress_signal_update(struct work_struct *work) { - struct saa7134_dev* dev = (struct saa7134_dev*) data; + struct saa7134_dev* dev = + container_of(work, struct saa7134_dev, empress_workqueue); if (dev->nosignal) { dprintk("no video signal\n"); @@ -378,7 +379,7 @@ static int empress_init(struct saa7134_dev *dev) "%s empress (%s)", dev->name, saa7134_boards[dev->board].name); - INIT_WORK(&dev->empress_workqueue, empress_signal_update, (void*) dev); + INIT_WORK(&dev->empress_workqueue, empress_signal_update); err = video_register_device(dev->empress_dev,VFL_TYPE_GRABBER, empress_nr[dev->nr]); @@ -399,7 +400,7 @@ static int empress_init(struct saa7134_dev *dev) sizeof(struct saa7134_buf), dev); - empress_signal_update(dev); + empress_signal_update(&dev->empress_workqueue); return 0; } diff --git a/drivers/message/fusion/mptfc.c b/drivers/message/fusion/mptfc.c index 1dd49177315..ef2b55e1991 100644 --- a/drivers/message/fusion/mptfc.c +++ b/drivers/message/fusion/mptfc.c @@ -1018,9 +1018,10 @@ mptfc_init_host_attr(MPT_ADAPTER *ioc,int portnum) } static void -mptfc_setup_reset(void *arg) +mptfc_setup_reset(struct work_struct *work) { - MPT_ADAPTER *ioc = (MPT_ADAPTER *)arg; + MPT_ADAPTER *ioc = + container_of(work, MPT_ADAPTER, fc_setup_reset_work); u64 pn; struct mptfc_rport_info *ri; @@ -1043,9 +1044,10 @@ mptfc_setup_reset(void *arg) } static void -mptfc_rescan_devices(void *arg) +mptfc_rescan_devices(struct work_struct *work) { - MPT_ADAPTER *ioc = (MPT_ADAPTER *)arg; + MPT_ADAPTER *ioc = + container_of(work, MPT_ADAPTER, fc_rescan_work); int ii; u64 pn; struct mptfc_rport_info *ri; @@ -1154,8 +1156,8 @@ mptfc_probe(struct pci_dev *pdev, const struct pci_device_id *id) } spin_lock_init(&ioc->fc_rescan_work_lock); - INIT_WORK(&ioc->fc_rescan_work, mptfc_rescan_devices,(void *)ioc); - INIT_WORK(&ioc->fc_setup_reset_work, mptfc_setup_reset, (void *)ioc); + INIT_WORK(&ioc->fc_rescan_work, mptfc_rescan_devices); + INIT_WORK(&ioc->fc_setup_reset_work, mptfc_setup_reset); spin_lock_irqsave(&ioc->FreeQlock, flags); diff --git a/drivers/message/fusion/mptlan.c b/drivers/message/fusion/mptlan.c index 314c3a27585..b7c4407c5e3 100644 --- a/drivers/message/fusion/mptlan.c +++ b/drivers/message/fusion/mptlan.c @@ -111,7 +111,8 @@ struct mpt_lan_priv { u32 total_received; struct net_device_stats stats; /* Per device statistics */ - struct work_struct post_buckets_task; + struct delayed_work post_buckets_task; + struct net_device *dev; unsigned long post_buckets_active; }; @@ -132,7 +133,7 @@ static int lan_reply (MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, static int mpt_lan_open(struct net_device *dev); static int mpt_lan_reset(struct net_device *dev); static int mpt_lan_close(struct net_device *dev); -static void mpt_lan_post_receive_buckets(void *dev_id); +static void mpt_lan_post_receive_buckets(struct mpt_lan_priv *priv); static void mpt_lan_wake_post_buckets_task(struct net_device *dev, int priority); static int mpt_lan_receive_post_turbo(struct net_device *dev, u32 tmsg); @@ -345,7 +346,7 @@ mpt_lan_ioc_reset(MPT_ADAPTER *ioc, int reset_phase) priv->mpt_rxfidx[++priv->mpt_rxfidx_tail] = i; spin_unlock_irqrestore(&priv->rxfidx_lock, flags); } else { - mpt_lan_post_receive_buckets(dev); + mpt_lan_post_receive_buckets(priv); netif_wake_queue(dev); } @@ -441,7 +442,7 @@ mpt_lan_open(struct net_device *dev) dlprintk((KERN_INFO MYNAM "/lo: Finished initializing RcvCtl\n")); - mpt_lan_post_receive_buckets(dev); + mpt_lan_post_receive_buckets(priv); printk(KERN_INFO MYNAM ": %s/%s: interface up & active\n", IOC_AND_NETDEV_NAMES_s_s(dev)); @@ -854,7 +855,7 @@ mpt_lan_wake_post_buckets_task(struct net_device *dev, int priority) if (test_and_set_bit(0, &priv->post_buckets_active) == 0) { if (priority) { - schedule_work(&priv->post_buckets_task); + schedule_delayed_work(&priv->post_buckets_task, 0); } else { schedule_delayed_work(&priv->post_buckets_task, 1); dioprintk((KERN_INFO MYNAM ": post_buckets queued on " @@ -1188,10 +1189,9 @@ mpt_lan_receive_post_reply(struct net_device *dev, /* Simple SGE's only at the moment */ static void -mpt_lan_post_receive_buckets(void *dev_id) +mpt_lan_post_receive_buckets(struct mpt_lan_priv *priv) { - struct net_device *dev = dev_id; - struct mpt_lan_priv *priv = dev->priv; + struct net_device *dev = priv->dev; MPT_ADAPTER *mpt_dev = priv->mpt_dev; MPT_FRAME_HDR *mf; LANReceivePostRequest_t *pRecvReq; @@ -1335,6 +1335,13 @@ out: clear_bit(0, &priv->post_buckets_active); } +static void +mpt_lan_post_receive_buckets_work(struct work_struct *work) +{ + mpt_lan_post_receive_buckets(container_of(work, struct mpt_lan_priv, + post_buckets_task.work)); +} + /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ static struct net_device * mpt_register_lan_device (MPT_ADAPTER *mpt_dev, int pnum) @@ -1350,11 +1357,13 @@ mpt_register_lan_device (MPT_ADAPTER *mpt_dev, int pnum) priv = netdev_priv(dev); + priv->dev = dev; priv->mpt_dev = mpt_dev; priv->pnum = pnum; - memset(&priv->post_buckets_task, 0, sizeof(struct work_struct)); - INIT_WORK(&priv->post_buckets_task, mpt_lan_post_receive_buckets, dev); + memset(&priv->post_buckets_task, 0, sizeof(priv->post_buckets_task)); + INIT_DELAYED_WORK(&priv->post_buckets_task, + mpt_lan_post_receive_buckets_work); priv->post_buckets_active = 0; dlprintk((KERN_INFO MYNAM "@%d: bucketlen = %d\n", diff --git a/drivers/message/fusion/mptsas.c b/drivers/message/fusion/mptsas.c index b752a479f6d..4f0c530e47b 100644 --- a/drivers/message/fusion/mptsas.c +++ b/drivers/message/fusion/mptsas.c @@ -2006,9 +2006,10 @@ __mptsas_discovery_work(MPT_ADAPTER *ioc) *(Mutex LOCKED) */ static void -mptsas_discovery_work(void * arg) +mptsas_discovery_work(struct work_struct *work) { - struct mptsas_discovery_event *ev = arg; + struct mptsas_discovery_event *ev = + container_of(work, struct mptsas_discovery_event, work); MPT_ADAPTER *ioc = ev->ioc; mutex_lock(&ioc->sas_discovery_mutex); @@ -2068,9 +2069,9 @@ mptsas_find_phyinfo_by_target(MPT_ADAPTER *ioc, u32 id) * Work queue thread to clear the persitency table */ static void -mptsas_persist_clear_table(void * arg) +mptsas_persist_clear_table(struct work_struct *work) { - MPT_ADAPTER *ioc = (MPT_ADAPTER *)arg; + MPT_ADAPTER *ioc = container_of(work, MPT_ADAPTER, sas_persist_task); mptbase_sas_persist_operation(ioc, MPI_SAS_OP_CLEAR_NOT_PRESENT); } @@ -2093,9 +2094,10 @@ mptsas_reprobe_target(struct scsi_target *starget, int uld_attach) * Work queue thread to handle SAS hotplug events */ static void -mptsas_hotplug_work(void *arg) +mptsas_hotplug_work(struct work_struct *work) { - struct mptsas_hotplug_event *ev = arg; + struct mptsas_hotplug_event *ev = + container_of(work, struct mptsas_hotplug_event, work); MPT_ADAPTER *ioc = ev->ioc; struct mptsas_phyinfo *phy_info; struct sas_rphy *rphy; @@ -2341,7 +2343,7 @@ mptsas_send_sas_event(MPT_ADAPTER *ioc, break; } - INIT_WORK(&ev->work, mptsas_hotplug_work, ev); + INIT_WORK(&ev->work, mptsas_hotplug_work); ev->ioc = ioc; ev->handle = le16_to_cpu(sas_event_data->DevHandle); ev->parent_handle = @@ -2366,7 +2368,7 @@ mptsas_send_sas_event(MPT_ADAPTER *ioc, * Persistent table is full. */ INIT_WORK(&ioc->sas_persist_task, - mptsas_persist_clear_table, (void *)ioc); + mptsas_persist_clear_table); schedule_work(&ioc->sas_persist_task); break; case MPI_EVENT_SAS_DEV_STAT_RC_SMART_DATA: @@ -2395,7 +2397,7 @@ mptsas_send_raid_event(MPT_ADAPTER *ioc, return; } - INIT_WORK(&ev->work, mptsas_hotplug_work, ev); + INIT_WORK(&ev->work, mptsas_hotplug_work); ev->ioc = ioc; ev->id = raid_event_data->VolumeID; ev->event_type = MPTSAS_IGNORE_EVENT; @@ -2474,7 +2476,7 @@ mptsas_send_discovery_event(MPT_ADAPTER *ioc, ev = kzalloc(sizeof(*ev), GFP_ATOMIC); if (!ev) return; - INIT_WORK(&ev->work, mptsas_discovery_work, ev); + INIT_WORK(&ev->work, mptsas_discovery_work); ev->ioc = ioc; schedule_work(&ev->work); }; @@ -2511,8 +2513,7 @@ mptsas_event_process(MPT_ADAPTER *ioc, EventNotificationReply_t *reply) break; case MPI_EVENT_PERSISTENT_TABLE_FULL: INIT_WORK(&ioc->sas_persist_task, - mptsas_persist_clear_table, - (void *)ioc); + mptsas_persist_clear_table); schedule_work(&ioc->sas_persist_task); break; case MPI_EVENT_SAS_DISCOVERY: diff --git a/drivers/message/fusion/mptspi.c b/drivers/message/fusion/mptspi.c index e4cc3dd5fc9..f422c0d0621 100644 --- a/drivers/message/fusion/mptspi.c +++ b/drivers/message/fusion/mptspi.c @@ -646,9 +646,10 @@ struct work_queue_wrapper { int disk; }; -static void mpt_work_wrapper(void *data) +static void mpt_work_wrapper(struct work_struct *work) { - struct work_queue_wrapper *wqw = (struct work_queue_wrapper *)data; + struct work_queue_wrapper *wqw = + container_of(work, struct work_queue_wrapper, work); struct _MPT_SCSI_HOST *hd = wqw->hd; struct Scsi_Host *shost = hd->ioc->sh; struct scsi_device *sdev; @@ -695,7 +696,7 @@ static void mpt_dv_raid(struct _MPT_SCSI_HOST *hd, int disk) disk); return; } - INIT_WORK(&wqw->work, mpt_work_wrapper, wqw); + INIT_WORK(&wqw->work, mpt_work_wrapper); wqw->hd = hd; wqw->disk = disk; @@ -784,9 +785,10 @@ MODULE_DEVICE_TABLE(pci, mptspi_pci_table); * renegotiate for a given target */ static void -mptspi_dv_renegotiate_work(void *data) +mptspi_dv_renegotiate_work(struct work_struct *work) { - struct work_queue_wrapper *wqw = (struct work_queue_wrapper *)data; + struct work_queue_wrapper *wqw = + container_of(work, struct work_queue_wrapper, work); struct _MPT_SCSI_HOST *hd = wqw->hd; struct scsi_device *sdev; @@ -804,7 +806,7 @@ mptspi_dv_renegotiate(struct _MPT_SCSI_HOST *hd) if (!wqw) return; - INIT_WORK(&wqw->work, mptspi_dv_renegotiate_work, wqw); + INIT_WORK(&wqw->work, mptspi_dv_renegotiate_work); wqw->hd = hd; schedule_work(&wqw->work); diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c index 64130227574..7fc7399bd2e 100644 --- a/drivers/message/i2o/driver.c +++ b/drivers/message/i2o/driver.c @@ -232,7 +232,7 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m) break; } - INIT_WORK(&evt->work, (void (*)(void *))drv->event, evt); + INIT_WORK(&evt->work, drv->event); queue_work(drv->event_queue, &evt->work); return 1; } diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index a2350640384..9e529d8dd5c 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -371,8 +371,10 @@ static int i2o_exec_remove(struct device *dev) * new LCT and if the buffer for the LCT was to small sends a LCT NOTIFY * again, otherwise send LCT NOTIFY to get informed on next LCT change. */ -static void i2o_exec_lct_modified(struct i2o_exec_lct_notify_work *work) +static void i2o_exec_lct_modified(struct work_struct *_work) { + struct i2o_exec_lct_notify_work *work = + container_of(_work, struct i2o_exec_lct_notify_work, work); u32 change_ind = 0; struct i2o_controller *c = work->c; @@ -439,8 +441,7 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m, work->c = c; - INIT_WORK(&work->work, (void (*)(void *))i2o_exec_lct_modified, - work); + INIT_WORK(&work->work, i2o_exec_lct_modified); queue_work(i2o_exec_driver.event_queue, &work->work); return 1; } @@ -460,13 +461,15 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m, /** * i2o_exec_event - Event handling function - * @evt: Event which occurs + * @work: Work item in occurring event * * Handles events send by the Executive device. At the moment does not do * anything useful. */ -static void i2o_exec_event(struct i2o_event *evt) +static void i2o_exec_event(struct work_struct *work) { + struct i2o_event *evt = container_of(work, struct i2o_event, work); + if (likely(evt->i2o_dev)) osm_debug("Event received from device: %d\n", evt->i2o_dev->lct_data.tid); diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c index eaba81bf2ec..70ae0025332 100644 --- a/drivers/message/i2o/i2o_block.c +++ b/drivers/message/i2o/i2o_block.c @@ -419,16 +419,18 @@ static int i2o_block_prep_req_fn(struct request_queue *q, struct request *req) /** * i2o_block_delayed_request_fn - delayed request queue function - * delayed_request: the delayed request with the queue to start + * @work: the delayed request with the queue to start * * If the request queue is stopped for a disk, and there is no open * request, a new event is created, which calls this function to start * the queue after I2O_BLOCK_REQUEST_TIME. Otherwise the queue will never * be started again. */ -static void i2o_block_delayed_request_fn(void *delayed_request) +static void i2o_block_delayed_request_fn(struct work_struct *work) { - struct i2o_block_delayed_request *dreq = delayed_request; + struct i2o_block_delayed_request *dreq = + container_of(work, struct i2o_block_delayed_request, + work.work); struct request_queue *q = dreq->queue; unsigned long flags; @@ -538,8 +540,9 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m, return 1; }; -static void i2o_block_event(struct i2o_event *evt) +static void i2o_block_event(struct work_struct *work) { + struct i2o_event *evt = container_of(work, struct i2o_event, work); osm_debug("event received\n"); kfree(evt); }; @@ -938,8 +941,8 @@ static void i2o_block_request_fn(struct request_queue *q) continue; dreq->queue = q; - INIT_WORK(&dreq->work, i2o_block_delayed_request_fn, - dreq); + INIT_DELAYED_WORK(&dreq->work, + i2o_block_delayed_request_fn); if (!queue_delayed_work(i2o_block_driver.event_queue, &dreq->work, diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h index 4fdaa5bda41..d9fdc95b440 100644 --- a/drivers/message/i2o/i2o_block.h +++ b/drivers/message/i2o/i2o_block.h @@ -96,7 +96,7 @@ struct i2o_block_request { /* I2O Block device delayed request */ struct i2o_block_delayed_request { - struct work_struct work; + struct delayed_work work; struct request_queue *queue; }; diff --git a/drivers/misc/tifm_7xx1.c b/drivers/misc/tifm_7xx1.c index 1ba8754e938..2ab7add78f9 100644 --- a/drivers/misc/tifm_7xx1.c +++ b/drivers/misc/tifm_7xx1.c @@ -33,9 +33,10 @@ static void tifm_7xx1_eject(struct tifm_adapter *fm, struct tifm_dev *sock) spin_unlock_irqrestore(&fm->lock, flags); } -static void tifm_7xx1_remove_media(void *adapter) +static void tifm_7xx1_remove_media(struct work_struct *work) { - struct tifm_adapter *fm = adapter; + struct tifm_adapter *fm = + container_of(work, struct tifm_adapter, media_remover); unsigned long flags; int cnt; struct tifm_dev *sock; @@ -169,9 +170,10 @@ tifm_7xx1_sock_addr(char __iomem *base_addr, unsigned int sock_num) return base_addr + ((sock_num + 1) << 10); } -static void tifm_7xx1_insert_media(void *adapter) +static void tifm_7xx1_insert_media(struct work_struct *work) { - struct tifm_adapter *fm = adapter; + struct tifm_adapter *fm = + container_of(work, struct tifm_adapter, media_inserter); unsigned long flags; tifm_media_id media_id; char *card_name = "xx"; @@ -261,7 +263,7 @@ static int tifm_7xx1_suspend(struct pci_dev *dev, pm_message_t state) spin_unlock_irqrestore(&fm->lock, flags); flush_workqueue(fm->wq); - tifm_7xx1_remove_media(fm); + tifm_7xx1_remove_media(&fm->media_remover); pci_set_power_state(dev, PCI_D3hot); pci_disable_device(dev); @@ -328,8 +330,8 @@ static int tifm_7xx1_probe(struct pci_dev *dev, if (!fm->sockets) goto err_out_free; - INIT_WORK(&fm->media_inserter, tifm_7xx1_insert_media, fm); - INIT_WORK(&fm->media_remover, tifm_7xx1_remove_media, fm); + INIT_WORK(&fm->media_inserter, tifm_7xx1_insert_media); + INIT_WORK(&fm->media_remover, tifm_7xx1_remove_media); fm->eject = tifm_7xx1_eject; pci_set_drvdata(dev, fm); @@ -384,7 +386,7 @@ static void tifm_7xx1_remove(struct pci_dev *dev) flush_workqueue(fm->wq); - tifm_7xx1_remove_media(fm); + tifm_7xx1_remove_media(&fm->media_remover); writel(TIFM_IRQ_SETALL, fm->addr + FM_CLEAR_INTERRUPT_ENABLE); free_irq(dev->irq, fm); diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index 766bc54406e..21fd39e4a20 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -1165,18 +1165,16 @@ static void mmc_setup(struct mmc_host *host) */ void mmc_detect_change(struct mmc_host *host, unsigned long delay) { - if (delay) - mmc_schedule_delayed_work(&host->detect, delay); - else - mmc_schedule_work(&host->detect); + mmc_schedule_delayed_work(&host->detect, delay); } EXPORT_SYMBOL(mmc_detect_change); -static void mmc_rescan(void *data) +static void mmc_rescan(struct work_struct *work) { - struct mmc_host *host = data; + struct mmc_host *host = + container_of(work, struct mmc_host, detect.work); struct list_head *l, *n; unsigned char power_mode; @@ -1259,7 +1257,7 @@ struct mmc_host *mmc_alloc_host(int extra, struct device *dev) spin_lock_init(&host->lock); init_waitqueue_head(&host->wq); INIT_LIST_HEAD(&host->cards); - INIT_WORK(&host->detect, mmc_rescan, host); + INIT_DELAYED_WORK(&host->detect, mmc_rescan); /* * By default, hosts do not support SGIO or large requests. @@ -1357,7 +1355,7 @@ EXPORT_SYMBOL(mmc_suspend_host); */ int mmc_resume_host(struct mmc_host *host) { - mmc_rescan(host); + mmc_rescan(&host->detect.work); return 0; } diff --git a/drivers/mmc/mmc.h b/drivers/mmc/mmc.h index cd5e0ab3d84..149affe0b68 100644 --- a/drivers/mmc/mmc.h +++ b/drivers/mmc/mmc.h @@ -20,6 +20,6 @@ void mmc_remove_host_sysfs(struct mmc_host *host); void mmc_free_host_sysfs(struct mmc_host *host); int mmc_schedule_work(struct work_struct *work); -int mmc_schedule_delayed_work(struct work_struct *work, unsigned long delay); +int mmc_schedule_delayed_work(struct delayed_work *work, unsigned long delay); void mmc_flush_scheduled_work(void); #endif diff --git a/drivers/mmc/mmc_sysfs.c b/drivers/mmc/mmc_sysfs.c index 10cc9734eaa..fd9a5fc6db7 100644 --- a/drivers/mmc/mmc_sysfs.c +++ b/drivers/mmc/mmc_sysfs.c @@ -320,18 +320,10 @@ void mmc_free_host_sysfs(struct mmc_host *host) static struct workqueue_struct *workqueue; -/* - * Internal function. Schedule work in the MMC work queue. - */ -int mmc_schedule_work(struct work_struct *work) -{ - return queue_work(workqueue, work); -} - /* * Internal function. Schedule delayed work in the MMC work queue. */ -int mmc_schedule_delayed_work(struct work_struct *work, unsigned long delay) +int mmc_schedule_delayed_work(struct delayed_work *work, unsigned long delay) { return queue_delayed_work(workqueue, work, delay); } diff --git a/drivers/mmc/tifm_sd.c b/drivers/mmc/tifm_sd.c index 0fdc55b08a6..e846499a004 100644 --- a/drivers/mmc/tifm_sd.c +++ b/drivers/mmc/tifm_sd.c @@ -99,7 +99,7 @@ struct tifm_sd { struct mmc_request *req; struct work_struct cmd_handler; - struct work_struct abort_handler; + struct delayed_work abort_handler; wait_queue_head_t can_eject; size_t written_blocks; @@ -496,9 +496,9 @@ err_out: mmc_request_done(mmc, mrq); } -static void tifm_sd_end_cmd(void *data) +static void tifm_sd_end_cmd(struct work_struct *work) { - struct tifm_sd *host = data; + struct tifm_sd *host = container_of(work, struct tifm_sd, cmd_handler); struct tifm_dev *sock = host->dev; struct mmc_host *mmc = tifm_get_drvdata(sock); struct mmc_request *mrq; @@ -608,9 +608,9 @@ err_out: mmc_request_done(mmc, mrq); } -static void tifm_sd_end_cmd_nodma(void *data) +static void tifm_sd_end_cmd_nodma(struct work_struct *work) { - struct tifm_sd *host = (struct tifm_sd*)data; + struct tifm_sd *host = container_of(work, struct tifm_sd, cmd_handler); struct tifm_dev *sock = host->dev; struct mmc_host *mmc = tifm_get_drvdata(sock); struct mmc_request *mrq; @@ -661,11 +661,14 @@ static void tifm_sd_end_cmd_nodma(void *data) mmc_request_done(mmc, mrq); } -static void tifm_sd_abort(void *data) +static void tifm_sd_abort(struct work_struct *work) { + struct tifm_sd *host = + container_of(work, struct tifm_sd, abort_handler.work); + printk(KERN_ERR DRIVER_NAME ": card failed to respond for a long period of time"); - tifm_eject(((struct tifm_sd*)data)->dev); + tifm_eject(host->dev); } static void tifm_sd_ios(struct mmc_host *mmc, struct mmc_ios *ios) @@ -762,9 +765,9 @@ static struct mmc_host_ops tifm_sd_ops = { .get_ro = tifm_sd_ro }; -static void tifm_sd_register_host(void *data) +static void tifm_sd_register_host(struct work_struct *work) { - struct tifm_sd *host = (struct tifm_sd*)data; + struct tifm_sd *host = container_of(work, struct tifm_sd, cmd_handler); struct tifm_dev *sock = host->dev; struct mmc_host *mmc = tifm_get_drvdata(sock); unsigned long flags; @@ -772,8 +775,7 @@ static void tifm_sd_register_host(void *data) spin_lock_irqsave(&sock->lock, flags); host->flags |= HOST_REG; PREPARE_WORK(&host->cmd_handler, - no_dma ? tifm_sd_end_cmd_nodma : tifm_sd_end_cmd, - data); + no_dma ? tifm_sd_end_cmd_nodma : tifm_sd_end_cmd); spin_unlock_irqrestore(&sock->lock, flags); dev_dbg(&sock->dev, "adding host\n"); mmc_add_host(mmc); @@ -799,8 +801,8 @@ static int tifm_sd_probe(struct tifm_dev *sock) host->dev = sock; host->clk_div = 61; init_waitqueue_head(&host->can_eject); - INIT_WORK(&host->cmd_handler, tifm_sd_register_host, host); - INIT_WORK(&host->abort_handler, tifm_sd_abort, host); + INIT_WORK(&host->cmd_handler, tifm_sd_register_host); + INIT_DELAYED_WORK(&host->abort_handler, tifm_sd_abort); tifm_set_drvdata(sock, mmc); sock->signal_irq = tifm_sd_signal_irq; diff --git a/drivers/net/8139too.c b/drivers/net/8139too.c index d02ed51abfc..931028f672d 100644 --- a/drivers/net/8139too.c +++ b/drivers/net/8139too.c @@ -594,7 +594,7 @@ struct rtl8139_private { u32 rx_config; struct rtl_extra_stats xstats; - struct work_struct thread; + struct delayed_work thread; struct mii_if_info mii; unsigned int regs_len; @@ -636,8 +636,8 @@ static struct net_device_stats *rtl8139_get_stats (struct net_device *dev); static void rtl8139_set_rx_mode (struct net_device *dev); static void __set_rx_mode (struct net_device *dev); static void rtl8139_hw_start (struct net_device *dev); -static void rtl8139_thread (void *_data); -static void rtl8139_tx_timeout_task(void *_data); +static void rtl8139_thread (struct work_struct *work); +static void rtl8139_tx_timeout_task(struct work_struct *work); static const struct ethtool_ops rtl8139_ethtool_ops; /* write MMIO register, with flush */ @@ -1010,7 +1010,7 @@ static int __devinit rtl8139_init_one (struct pci_dev *pdev, (debug < 0 ? RTL8139_DEF_MSG_ENABLE : ((1 << debug) - 1)); spin_lock_init (&tp->lock); spin_lock_init (&tp->rx_lock); - INIT_WORK(&tp->thread, rtl8139_thread, dev); + INIT_DELAYED_WORK(&tp->thread, rtl8139_thread); tp->mii.dev = dev; tp->mii.mdio_read = mdio_read; tp->mii.mdio_write = mdio_write; @@ -1596,15 +1596,16 @@ static inline void rtl8139_thread_iter (struct net_device *dev, RTL_R8 (Config1)); } -static void rtl8139_thread (void *_data) +static void rtl8139_thread (struct work_struct *work) { - struct net_device *dev = _data; - struct rtl8139_private *tp = netdev_priv(dev); + struct rtl8139_private *tp = + container_of(work, struct rtl8139_private, thread.work); + struct net_device *dev = tp->mii.dev; unsigned long thr_delay = next_tick; if (tp->watchdog_fired) { tp->watchdog_fired = 0; - rtl8139_tx_timeout_task(_data); + rtl8139_tx_timeout_task(work); } else if (rtnl_trylock()) { rtl8139_thread_iter (dev, tp, tp->mmio_addr); rtnl_unlock (); @@ -1646,10 +1647,11 @@ static inline void rtl8139_tx_clear (struct rtl8139_private *tp) /* XXX account for unsent Tx packets in tp->stats.tx_dropped */ } -static void rtl8139_tx_timeout_task (void *_data) +static void rtl8139_tx_timeout_task (struct work_struct *work) { - struct net_device *dev = _data; - struct rtl8139_private *tp = netdev_priv(dev); + struct rtl8139_private *tp = + container_of(work, struct rtl8139_private, thread.work); + struct net_device *dev = tp->mii.dev; void __iomem *ioaddr = tp->mmio_addr; int i; u8 tmp8; @@ -1695,7 +1697,7 @@ static void rtl8139_tx_timeout (struct net_device *dev) struct rtl8139_private *tp = netdev_priv(dev); if (!tp->have_thread) { - INIT_WORK(&tp->thread, rtl8139_tx_timeout_task, dev); + INIT_DELAYED_WORK(&tp->thread, rtl8139_tx_timeout_task); schedule_delayed_work(&tp->thread, next_tick); } else tp->watchdog_fired = 1; diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 01b76d3aa42..b12cc4596b8 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -4339,9 +4339,9 @@ bnx2_open(struct net_device *dev) } static void -bnx2_reset_task(void *data) +bnx2_reset_task(struct work_struct *work) { - struct bnx2 *bp = data; + struct bnx2 *bp = container_of(work, struct bnx2, reset_task); if (!netif_running(bp->dev)) return; @@ -5630,7 +5630,7 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) bp->pdev = pdev; spin_lock_init(&bp->phy_lock); - INIT_WORK(&bp->reset_task, bnx2_reset_task, bp); + INIT_WORK(&bp->reset_task, bnx2_reset_task); dev->base_addr = dev->mem_start = pci_resource_start(pdev, 0); mem_len = MB_GET_CID_ADDR(17); diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c index 521c5b71023..fe08f384549 100644 --- a/drivers/net/cassini.c +++ b/drivers/net/cassini.c @@ -4066,9 +4066,9 @@ static int cas_alloc_rxds(struct cas *cp) return 0; } -static void cas_reset_task(void *data) +static void cas_reset_task(struct work_struct *work) { - struct cas *cp = (struct cas *) data; + struct cas *cp = container_of(work, struct cas, reset_task); #if 0 int pending = atomic_read(&cp->reset_task_pending); #else @@ -5006,7 +5006,7 @@ static int __devinit cas_init_one(struct pci_dev *pdev, atomic_set(&cp->reset_task_pending_spare, 0); atomic_set(&cp->reset_task_pending_mtu, 0); #endif - INIT_WORK(&cp->reset_task, cas_reset_task, cp); + INIT_WORK(&cp->reset_task, cas_reset_task); /* Default link parameters */ if (link_mode >= 0 && link_mode <= 6) diff --git a/drivers/net/chelsio/common.h b/drivers/net/chelsio/common.h index 5d9dd14427c..8b1bedbce0d 100644 --- a/drivers/net/chelsio/common.h +++ b/drivers/net/chelsio/common.h @@ -209,7 +209,7 @@ struct adapter { struct peespi *espi; struct port_info port[MAX_NPORTS]; - struct work_struct stats_update_task; + struct delayed_work stats_update_task; struct timer_list stats_update_timer; struct semaphore mib_mutex; diff --git a/drivers/net/chelsio/cxgb2.c b/drivers/net/chelsio/cxgb2.c index ad7ff9641a7..f607cc6a276 100644 --- a/drivers/net/chelsio/cxgb2.c +++ b/drivers/net/chelsio/cxgb2.c @@ -927,10 +927,11 @@ static void t1_netpoll(struct net_device *dev) * Periodic accumulation of MAC statistics. This is used only if the MAC * does not have any other way to prevent stats counter overflow. */ -static void mac_stats_task(void *data) +static void mac_stats_task(struct work_struct *work) { int i; - struct adapter *adapter = data; + struct adapter *adapter = + container_of(work, struct adapter, stats_update_task.work); for_each_port(adapter, i) { struct port_info *p = &adapter->port[i]; @@ -951,9 +952,10 @@ static void mac_stats_task(void *data) /* * Processes elmer0 external interrupts in process context. */ -static void ext_intr_task(void *data) +static void ext_intr_task(struct work_struct *work) { - struct adapter *adapter = data; + struct adapter *adapter = + container_of(work, struct adapter, ext_intr_handler_task); elmer0_ext_intr_handler(adapter); @@ -1087,9 +1089,9 @@ static int __devinit init_one(struct pci_dev *pdev, spin_lock_init(&adapter->async_lock); INIT_WORK(&adapter->ext_intr_handler_task, - ext_intr_task, adapter); - INIT_WORK(&adapter->stats_update_task, mac_stats_task, - adapter); + ext_intr_task); + INIT_DELAYED_WORK(&adapter->stats_update_task, + mac_stats_task); #ifdef work_struct init_timer(&adapter->stats_update_timer); adapter->stats_update_timer.function = mac_stats_timer; diff --git a/drivers/net/e100.c b/drivers/net/e100.c index 19ab3441269..e7737d02bb0 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -2102,9 +2102,10 @@ static void e100_tx_timeout(struct net_device *netdev) schedule_work(&nic->tx_timeout_task); } -static void e100_tx_timeout_task(struct net_device *netdev) +static void e100_tx_timeout_task(struct work_struct *work) { - struct nic *nic = netdev_priv(netdev); + struct nic *nic = container_of(work, struct nic, tx_timeout_task); + struct net_device *netdev = nic->netdev; DPRINTK(TX_ERR, DEBUG, "scb.status=0x%02X\n", readb(&nic->csr->scb.status)); @@ -2637,8 +2638,7 @@ static int __devinit e100_probe(struct pci_dev *pdev, nic->blink_timer.function = e100_blink_led; nic->blink_timer.data = (unsigned long)nic; - INIT_WORK(&nic->tx_timeout_task, - (void (*)(void *))e100_tx_timeout_task, netdev); + INIT_WORK(&nic->tx_timeout_task, e100_tx_timeout_task); if((err = e100_alloc(nic))) { DPRINTK(PROBE, ERR, "Cannot alloc driver memory, aborting.\n"); diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 6ad69610141..83fa32f7239 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -2224,11 +2224,12 @@ static int ehea_stop(struct net_device *dev) return ret; } -static void ehea_reset_port(void *data) +static void ehea_reset_port(struct work_struct *work) { int ret; - struct net_device *dev = data; - struct ehea_port *port = netdev_priv(dev); + struct ehea_port *port = + container_of(work, struct ehea_port, reset_task); + struct net_device *dev = port->netdev; port->resets++; down(&port->port_lock); @@ -2379,7 +2380,7 @@ static int ehea_setup_single_port(struct ehea_port *port, dev->tx_timeout = &ehea_tx_watchdog; dev->watchdog_timeo = EHEA_WATCH_DOG_TIMEOUT; - INIT_WORK(&port->reset_task, ehea_reset_port, dev); + INIT_WORK(&port->reset_task, ehea_reset_port); ehea_set_ethtool_ops(dev); diff --git a/drivers/net/hamradio/baycom_epp.c b/drivers/net/hamradio/baycom_epp.c index 1ed9cccd3c1..3c33d6f6a6a 100644 --- a/drivers/net/hamradio/baycom_epp.c +++ b/drivers/net/hamradio/baycom_epp.c @@ -168,8 +168,9 @@ struct baycom_state { int magic; struct pardevice *pdev; + struct net_device *dev; unsigned int work_running; - struct work_struct run_work; + struct delayed_work run_work; unsigned int modem; unsigned int bitrate; unsigned char stat; @@ -659,16 +660,18 @@ static int receive(struct net_device *dev, int cnt) #define GETTICK(x) #endif /* __i386__ */ -static void epp_bh(struct net_device *dev) +static void epp_bh(struct work_struct *work) { + struct net_device *dev; struct baycom_state *bc; struct parport *pp; unsigned char stat; unsigned char tmp[2]; unsigned int time1 = 0, time2 = 0, time3 = 0; int cnt, cnt2; - - bc = netdev_priv(dev); + + bc = container_of(work, struct baycom_state, run_work.work); + dev = bc->dev; if (!bc->work_running) return; baycom_int_freq(bc); @@ -889,7 +892,7 @@ static int epp_open(struct net_device *dev) return -EBUSY; } dev->irq = /*pp->irq*/ 0; - INIT_WORK(&bc->run_work, (void *)(void *)epp_bh, dev); + INIT_DELAYED_WORK(&bc->run_work, epp_bh); bc->work_running = 1; bc->modem = EPP_CONVENTIONAL; if (eppconfig(bc)) @@ -1213,6 +1216,7 @@ static void __init baycom_epp_dev_setup(struct net_device *dev) /* * initialize part of the baycom_state struct */ + bc->dev = dev; bc->magic = BAYCOM_MAGIC; bc->cfg.fclk = 19666600; bc->cfg.bps = 9600; diff --git a/drivers/net/irda/mcs7780.c b/drivers/net/irda/mcs7780.c index b32c52ed19d..f0c61f3b2a8 100644 --- a/drivers/net/irda/mcs7780.c +++ b/drivers/net/irda/mcs7780.c @@ -560,9 +560,9 @@ static inline int mcs_find_endpoints(struct mcs_cb *mcs, return ret; } -static void mcs_speed_work(void *arg) +static void mcs_speed_work(struct work_struct *work) { - struct mcs_cb *mcs = arg; + struct mcs_cb *mcs = container_of(work, struct mcs_cb, work); struct net_device *netdev = mcs->netdev; mcs_speed_change(mcs); @@ -927,7 +927,7 @@ static int mcs_probe(struct usb_interface *intf, irda_qos_bits_to_value(&mcs->qos); /* Speed change work initialisation*/ - INIT_WORK(&mcs->work, mcs_speed_work, mcs); + INIT_WORK(&mcs->work, mcs_speed_work); /* Override the network functions we need to use */ ndev->hard_start_xmit = mcs_hard_xmit; diff --git a/drivers/net/irda/sir-dev.h b/drivers/net/irda/sir-dev.h index 9fa294a546d..2a57bc67ce3 100644 --- a/drivers/net/irda/sir-dev.h +++ b/drivers/net/irda/sir-dev.h @@ -22,7 +22,7 @@ struct sir_fsm { struct semaphore sem; - struct work_struct work; + struct delayed_work work; unsigned state, substate; int param; int result; diff --git a/drivers/net/irda/sir_dev.c b/drivers/net/irda/sir_dev.c index 3b5854d10c1..17b0c3ab620 100644 --- a/drivers/net/irda/sir_dev.c +++ b/drivers/net/irda/sir_dev.c @@ -100,9 +100,9 @@ static int sirdev_tx_complete_fsm(struct sir_dev *dev) * Both must be unlocked/restarted on completion - but only on final exit. */ -static void sirdev_config_fsm(void *data) +static void sirdev_config_fsm(struct work_struct *work) { - struct sir_dev *dev = data; + struct sir_dev *dev = container_of(work, struct sir_dev, fsm.work.work); struct sir_fsm *fsm = &dev->fsm; int next_state; int ret = -1; @@ -309,8 +309,8 @@ int sirdev_schedule_request(struct sir_dev *dev, int initial_state, unsigned par fsm->param = param; fsm->result = 0; - INIT_WORK(&fsm->work, sirdev_config_fsm, dev); - queue_work(irda_sir_wq, &fsm->work); + INIT_DELAYED_WORK(&fsm->work, sirdev_config_fsm); + queue_delayed_work(irda_sir_wq, &fsm->work, 0); return 0; } diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c index 2284e2ce169..d6f4f185bf3 100644 --- a/drivers/net/iseries_veth.c +++ b/drivers/net/iseries_veth.c @@ -166,7 +166,7 @@ struct veth_msg { struct veth_lpar_connection { HvLpIndex remote_lp; - struct work_struct statemachine_wq; + struct delayed_work statemachine_wq; struct veth_msg *msgs; int num_events; struct veth_cap_data local_caps; @@ -456,7 +456,7 @@ static struct kobj_type veth_port_ktype = { static inline void veth_kick_statemachine(struct veth_lpar_connection *cnx) { - schedule_work(&cnx->statemachine_wq); + schedule_delayed_work(&cnx->statemachine_wq, 0); } static void veth_take_cap(struct veth_lpar_connection *cnx, @@ -638,9 +638,11 @@ static int veth_process_caps(struct veth_lpar_connection *cnx) } /* FIXME: The gotos here are a bit dubious */ -static void veth_statemachine(void *p) +static void veth_statemachine(struct work_struct *work) { - struct veth_lpar_connection *cnx = (struct veth_lpar_connection *)p; + struct veth_lpar_connection *cnx = + container_of(work, struct veth_lpar_connection, + statemachine_wq.work); int rlp = cnx->remote_lp; int rc; @@ -827,7 +829,7 @@ static int veth_init_connection(u8 rlp) cnx->remote_lp = rlp; spin_lock_init(&cnx->lock); - INIT_WORK(&cnx->statemachine_wq, veth_statemachine, cnx); + INIT_DELAYED_WORK(&cnx->statemachine_wq, veth_statemachine); init_timer(&cnx->ack_timer); cnx->ack_timer.function = veth_timed_ack; diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c index e09f575a3a3..d1ebb91ed27 100644 --- a/drivers/net/ixgb/ixgb_main.c +++ b/drivers/net/ixgb/ixgb_main.c @@ -106,7 +106,7 @@ static boolean_t ixgb_clean_rx_irq(struct ixgb_adapter *adapter); static void ixgb_alloc_rx_buffers(struct ixgb_adapter *adapter); void ixgb_set_ethtool_ops(struct net_device *netdev); static void ixgb_tx_timeout(struct net_device *dev); -static void ixgb_tx_timeout_task(struct net_device *dev); +static void ixgb_tx_timeout_task(struct work_struct *work); static void ixgb_vlan_rx_register(struct net_device *netdev, struct vlan_group *grp); static void ixgb_vlan_rx_add_vid(struct net_device *netdev, uint16_t vid); @@ -489,8 +489,7 @@ ixgb_probe(struct pci_dev *pdev, adapter->watchdog_timer.function = &ixgb_watchdog; adapter->watchdog_timer.data = (unsigned long)adapter; - INIT_WORK(&adapter->tx_timeout_task, - (void (*)(void *))ixgb_tx_timeout_task, netdev); + INIT_WORK(&adapter->tx_timeout_task, ixgb_tx_timeout_task); strcpy(netdev->name, "eth%d"); if((err = register_netdev(netdev))) @@ -1493,9 +1492,10 @@ ixgb_tx_timeout(struct net_device *netdev) } static void -ixgb_tx_timeout_task(struct net_device *netdev) +ixgb_tx_timeout_task(struct work_struct *work) { - struct ixgb_adapter *adapter = netdev_priv(netdev); + struct ixgb_adapter *adapter = + container_of(work, struct ixgb_adapter, tx_timeout_task); adapter->tx_timeout_count++; ixgb_down(adapter, TRUE); diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index 806081b5973..98703e086ee 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -2615,9 +2615,10 @@ static u32 myri10ge_read_reboot(struct myri10ge_priv *mgp) * This watchdog is used to check whether the board has suffered * from a parity error and needs to be recovered. */ -static void myri10ge_watchdog(void *arg) +static void myri10ge_watchdog(struct work_struct *work) { - struct myri10ge_priv *mgp = arg; + struct myri10ge_priv *mgp = + container_of(work, struct myri10ge_priv, watchdog_work); u32 reboot; int status; u16 cmd, vendor; @@ -2887,7 +2888,7 @@ static int myri10ge_probe(struct pci_dev *pdev, const struct pci_device_id *ent) (unsigned long)mgp); SET_ETHTOOL_OPS(netdev, &myri10ge_ethtool_ops); - INIT_WORK(&mgp->watchdog_work, myri10ge_watchdog, mgp); + INIT_WORK(&mgp->watchdog_work, myri10ge_watchdog); status = register_netdev(netdev); if (status != 0) { dev_err(&pdev->dev, "register_netdev failed: %d\n", status); diff --git a/drivers/net/ns83820.c b/drivers/net/ns83820.c index b0127c71a5b..312e0e33171 100644 --- a/drivers/net/ns83820.c +++ b/drivers/net/ns83820.c @@ -427,6 +427,7 @@ struct ns83820 { u8 __iomem *base; struct pci_dev *pci_dev; + struct net_device *ndev; #ifdef NS83820_VLAN_ACCEL_SUPPORT struct vlan_group *vlgrp; @@ -631,10 +632,10 @@ static void fastcall rx_refill_atomic(struct net_device *ndev) } /* REFILL */ -static inline void queue_refill(void *_dev) +static inline void queue_refill(struct work_struct *work) { - struct net_device *ndev = _dev; - struct ns83820 *dev = PRIV(ndev); + struct ns83820 *dev = container_of(work, struct ns83820, tq_refill); + struct net_device *ndev = dev->ndev; rx_refill(ndev, GFP_KERNEL); if (dev->rx_info.up) @@ -1841,6 +1842,7 @@ static int __devinit ns83820_init_one(struct pci_dev *pci_dev, const struct pci_ ndev = alloc_etherdev(sizeof(struct ns83820)); dev = PRIV(ndev); + dev->ndev = ndev; err = -ENOMEM; if (!dev) goto out; @@ -1853,7 +1855,7 @@ static int __devinit ns83820_init_one(struct pci_dev *pci_dev, const struct pci_ SET_MODULE_OWNER(ndev); SET_NETDEV_DEV(ndev, &pci_dev->dev); - INIT_WORK(&dev->tq_refill, queue_refill, ndev); + INIT_WORK(&dev->tq_refill, queue_refill); tasklet_init(&dev->rx_tasklet, rx_action, (unsigned long)ndev); err = pci_enable_device(pci_dev); diff --git a/drivers/net/pcmcia/xirc2ps_cs.c b/drivers/net/pcmcia/xirc2ps_cs.c index f3914f58d67..5de8850f232 100644 --- a/drivers/net/pcmcia/xirc2ps_cs.c +++ b/drivers/net/pcmcia/xirc2ps_cs.c @@ -332,6 +332,7 @@ static irqreturn_t xirc2ps_interrupt(int irq, void *dev_id); */ typedef struct local_info_t { + struct net_device *dev; struct pcmcia_device *p_dev; dev_node_t node; struct net_device_stats stats; @@ -353,7 +354,7 @@ typedef struct local_info_t { */ static int do_start_xmit(struct sk_buff *skb, struct net_device *dev); static void do_tx_timeout(struct net_device *dev); -static void xirc2ps_tx_timeout_task(void *data); +static void xirc2ps_tx_timeout_task(struct work_struct *work); static struct net_device_stats *do_get_stats(struct net_device *dev); static void set_addresses(struct net_device *dev); static void set_multicast_list(struct net_device *dev); @@ -567,6 +568,7 @@ xirc2ps_probe(struct pcmcia_device *link) if (!dev) return -ENOMEM; local = netdev_priv(dev); + local->dev = dev; local->p_dev = link; link->priv = dev; @@ -591,7 +593,7 @@ xirc2ps_probe(struct pcmcia_device *link) #ifdef HAVE_TX_TIMEOUT dev->tx_timeout = do_tx_timeout; dev->watchdog_timeo = TX_TIMEOUT; - INIT_WORK(&local->tx_timeout_task, xirc2ps_tx_timeout_task, dev); + INIT_WORK(&local->tx_timeout_task, xirc2ps_tx_timeout_task); #endif return xirc2ps_config(link); @@ -1344,9 +1346,11 @@ xirc2ps_interrupt(int irq, void *dev_id) /*====================================================================*/ static void -xirc2ps_tx_timeout_task(void *data) +xirc2ps_tx_timeout_task(struct work_struct *work) { - struct net_device *dev = data; + local_info_t *local = + container_of(work, local_info_t, tx_timeout_task); + struct net_device *dev = local->dev; /* reset the card */ do_reset(dev,1); dev->trans_start = jiffies; diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 3af9fcf76c8..a443976d5dc 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -394,7 +394,7 @@ out_unlock: EXPORT_SYMBOL(phy_start_aneg); -static void phy_change(void *data); +static void phy_change(struct work_struct *work); static void phy_timer(unsigned long data); /* phy_start_machine: @@ -549,7 +549,7 @@ int phy_start_interrupts(struct phy_device *phydev) { int err = 0; - INIT_WORK(&phydev->phy_queue, phy_change, phydev); + INIT_WORK(&phydev->phy_queue, phy_change); if (request_irq(phydev->irq, phy_interrupt, IRQF_SHARED, @@ -585,10 +585,11 @@ EXPORT_SYMBOL(phy_stop_interrupts); /* Scheduled by the phy_interrupt/timer to handle PHY changes */ -static void phy_change(void *data) +static void phy_change(struct work_struct *work) { int err; - struct phy_device *phydev = data; + struct phy_device *phydev = + container_of(work, struct phy_device, phy_queue); err = phy_disable_interrupts(phydev); diff --git a/drivers/net/plip.c b/drivers/net/plip.c index 71afb274498..6bb085f5443 100644 --- a/drivers/net/plip.c +++ b/drivers/net/plip.c @@ -138,9 +138,9 @@ static const unsigned int net_debug = NET_DEBUG; #define PLIP_NIBBLE_WAIT 3000 /* Bottom halves */ -static void plip_kick_bh(struct net_device *dev); -static void plip_bh(struct net_device *dev); -static void plip_timer_bh(struct net_device *dev); +static void plip_kick_bh(struct work_struct *work); +static void plip_bh(struct work_struct *work); +static void plip_timer_bh(struct work_struct *work); /* Interrupt handler */ static void plip_interrupt(int irq, void *dev_id); @@ -207,9 +207,10 @@ struct plip_local { struct net_local { struct net_device_stats enet_stats; + struct net_device *dev; struct work_struct immediate; - struct work_struct deferred; - struct work_struct timer; + struct delayed_work deferred; + struct delayed_work timer; struct plip_local snd_data; struct plip_local rcv_data; struct pardevice *pardev; @@ -306,11 +307,11 @@ plip_init_netdev(struct net_device *dev) nl->nibble = PLIP_NIBBLE_WAIT; /* Initialize task queue structures */ - INIT_WORK(&nl->immediate, (void (*)(void *))plip_bh, dev); - INIT_WORK(&nl->deferred, (void (*)(void *))plip_kick_bh, dev); + INIT_WORK(&nl->immediate, plip_bh); + INIT_DELAYED_WORK(&nl->deferred, plip_kick_bh); if (dev->irq == -1) - INIT_WORK(&nl->timer, (void (*)(void *))plip_timer_bh, dev); + INIT_DELAYED_WORK(&nl->timer, plip_timer_bh); spin_lock_init(&nl->lock); } @@ -319,9 +320,10 @@ plip_init_netdev(struct net_device *dev) This routine is kicked by do_timer(). Request `plip_bh' to be invoked. */ static void -plip_kick_bh(struct net_device *dev) +plip_kick_bh(struct work_struct *work) { - struct net_local *nl = netdev_priv(dev); + struct net_local *nl = + container_of(work, struct net_local, deferred.work); if (nl->is_deferred) schedule_work(&nl->immediate); @@ -362,9 +364,9 @@ static const plip_func connection_state_table[] = /* Bottom half handler of PLIP. */ static void -plip_bh(struct net_device *dev) +plip_bh(struct work_struct *work) { - struct net_local *nl = netdev_priv(dev); + struct net_local *nl = container_of(work, struct net_local, immediate); struct plip_local *snd = &nl->snd_data; struct plip_local *rcv = &nl->rcv_data; plip_func f; @@ -372,20 +374,21 @@ plip_bh(struct net_device *dev) nl->is_deferred = 0; f = connection_state_table[nl->connection]; - if ((r = (*f)(dev, nl, snd, rcv)) != OK - && (r = plip_bh_timeout_error(dev, nl, snd, rcv, r)) != OK) { + if ((r = (*f)(nl->dev, nl, snd, rcv)) != OK + && (r = plip_bh_timeout_error(nl->dev, nl, snd, rcv, r)) != OK) { nl->is_deferred = 1; schedule_delayed_work(&nl->deferred, 1); } } static void -plip_timer_bh(struct net_device *dev) +plip_timer_bh(struct work_struct *work) { - struct net_local *nl = netdev_priv(dev); + struct net_local *nl = + container_of(work, struct net_local, timer.work); if (!(atomic_read (&nl->kill_timer))) { - plip_interrupt (-1, dev); + plip_interrupt (-1, nl->dev); schedule_delayed_work(&nl->timer, 1); } @@ -1284,6 +1287,7 @@ static void plip_attach (struct parport *port) } nl = netdev_priv(dev); + nl->dev = dev; nl->pardev = parport_register_device(port, name, plip_preempt, plip_wakeup, plip_interrupt, 0, dev); diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c index ec640f6229a..d79d141a601 100644 --- a/drivers/net/qla3xxx.c +++ b/drivers/net/qla3xxx.c @@ -2008,7 +2008,7 @@ static irqreturn_t ql3xxx_isr(int irq, void *dev_id) "%s: Another function issued a reset to the " "chip. ISR value = %x.\n", ndev->name, value); } - queue_work(qdev->workqueue, &qdev->reset_work); + queue_delayed_work(qdev->workqueue, &qdev->reset_work, 0); spin_unlock(&qdev->adapter_lock); } else if (value & ISP_IMR_DISABLE_CMPL_INT) { ql_disable_interrupts(qdev); @@ -3182,11 +3182,13 @@ static void ql3xxx_tx_timeout(struct net_device *ndev) /* * Wake up the worker to process this event. */ - queue_work(qdev->workqueue, &qdev->tx_timeout_work); + queue_delayed_work(qdev->workqueue, &qdev->tx_timeout_work, 0); } -static void ql_reset_work(struct ql3_adapter *qdev) +static void ql_reset_work(struct work_struct *work) { + struct ql3_adapter *qdev = + container_of(work, struct ql3_adapter, reset_work.work); struct net_device *ndev = qdev->ndev; u32 value; struct ql_tx_buf_cb *tx_cb; @@ -3278,9 +3280,12 @@ static void ql_reset_work(struct ql3_adapter *qdev) } } -static void ql_tx_timeout_work(struct ql3_adapter *qdev) +static void ql_tx_timeout_work(struct work_struct *work) { - ql_cycle_adapter(qdev,QL_DO_RESET); + struct ql3_adapter *qdev = + container_of(work, struct ql3_adapter, tx_timeout_work.work); + + ql_cycle_adapter(qdev, QL_DO_RESET); } static void ql_get_board_info(struct ql3_adapter *qdev) @@ -3459,9 +3464,8 @@ static int __devinit ql3xxx_probe(struct pci_dev *pdev, netif_stop_queue(ndev); qdev->workqueue = create_singlethread_workqueue(ndev->name); - INIT_WORK(&qdev->reset_work, (void (*)(void *))ql_reset_work, qdev); - INIT_WORK(&qdev->tx_timeout_work, - (void (*)(void *))ql_tx_timeout_work, qdev); + INIT_DELAYED_WORK(&qdev->reset_work, ql_reset_work); + INIT_DELAYED_WORK(&qdev->tx_timeout_work, ql_tx_timeout_work); init_timer(&qdev->adapter_timer); qdev->adapter_timer.function = ql3xxx_timer; diff --git a/drivers/net/qla3xxx.h b/drivers/net/qla3xxx.h index 65da2c0bfda..ea94de7fd07 100644 --- a/drivers/net/qla3xxx.h +++ b/drivers/net/qla3xxx.h @@ -1186,8 +1186,8 @@ struct ql3_adapter { u32 numPorts; struct net_device_stats stats; struct workqueue_struct *workqueue; - struct work_struct reset_work; - struct work_struct tx_timeout_work; + struct delayed_work reset_work; + struct delayed_work tx_timeout_work; u32 max_frame_size; }; diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index 27f90b2139c..1f9663a7082 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -424,6 +424,7 @@ struct ring_info { struct rtl8169_private { void __iomem *mmio_addr; /* memory map physical address */ struct pci_dev *pci_dev; /* Index of PCI device */ + struct net_device *dev; struct net_device_stats stats; /* statistics of net device */ spinlock_t lock; /* spin lock flag */ u32 msg_enable; @@ -455,7 +456,7 @@ struct rtl8169_private { void (*phy_reset_enable)(void __iomem *); unsigned int (*phy_reset_pending)(void __iomem *); unsigned int (*link_ok)(void __iomem *); - struct work_struct task; + struct delayed_work task; unsigned wol_enabled : 1; }; @@ -1492,6 +1493,7 @@ rtl8169_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) SET_MODULE_OWNER(dev); SET_NETDEV_DEV(dev, &pdev->dev); tp = netdev_priv(dev); + tp->dev = dev; tp->msg_enable = netif_msg_init(debug.msg_enable, R8169_MSG_DEFAULT); /* enable device (incl. PCI PM wakeup and hotplug setup) */ @@ -1764,7 +1766,7 @@ static int rtl8169_open(struct net_device *dev) if (retval < 0) goto err_free_rx; - INIT_WORK(&tp->task, NULL, dev); + INIT_DELAYED_WORK(&tp->task, NULL); rtl8169_hw_start(dev); @@ -2087,11 +2089,11 @@ static void rtl8169_tx_clear(struct rtl8169_private *tp) tp->cur_tx = tp->dirty_tx = 0; } -static void rtl8169_schedule_work(struct net_device *dev, void (*task)(void *)) +static void rtl8169_schedule_work(struct net_device *dev, work_func_t task) { struct rtl8169_private *tp = netdev_priv(dev); - PREPARE_WORK(&tp->task, task, dev); + PREPARE_DELAYED_WORK(&tp->task, task); schedule_delayed_work(&tp->task, 4); } @@ -2110,9 +2112,11 @@ static void rtl8169_wait_for_quiescence(struct net_device *dev) netif_poll_enable(dev); } -static void rtl8169_reinit_task(void *_data) +static void rtl8169_reinit_task(struct work_struct *work) { - struct net_device *dev = _data; + struct rtl8169_private *tp = + container_of(work, struct rtl8169_private, task.work); + struct net_device *dev = tp->dev; int ret; if (netif_running(dev)) { @@ -2135,10 +2139,11 @@ static void rtl8169_reinit_task(void *_data) } } -static void rtl8169_reset_task(void *_data) +static void rtl8169_reset_task(struct work_struct *work) { - struct net_device *dev = _data; - struct rtl8169_private *tp = netdev_priv(dev); + struct rtl8169_private *tp = + container_of(work, struct rtl8169_private, task.work); + struct net_device *dev = tp->dev; if (!netif_running(dev)) return; diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c index 33569ec9dbf..250cdbeefdf 100644 --- a/drivers/net/s2io.c +++ b/drivers/net/s2io.c @@ -5872,9 +5872,9 @@ static void s2io_tasklet(unsigned long dev_addr) * Description: Sets the link status for the adapter */ -static void s2io_set_link(unsigned long data) +static void s2io_set_link(struct work_struct *work) { - nic_t *nic = (nic_t *) data; + nic_t *nic = container_of(work, nic_t, set_link_task); struct net_device *dev = nic->dev; XENA_dev_config_t __iomem *bar0 = nic->bar0; register u64 val64; @@ -6379,10 +6379,10 @@ static int s2io_card_up(nic_t * sp) * spin lock. */ -static void s2io_restart_nic(unsigned long data) +static void s2io_restart_nic(struct work_struct *work) { - struct net_device *dev = (struct net_device *) data; - nic_t *sp = dev->priv; + nic_t *sp = container_of(work, nic_t, rst_timer_task); + struct net_device *dev = sp->dev; s2io_card_down(sp); if (s2io_card_up(sp)) { @@ -6992,10 +6992,8 @@ s2io_init_nic(struct pci_dev *pdev, const struct pci_device_id *pre) dev->tx_timeout = &s2io_tx_watchdog; dev->watchdog_timeo = WATCH_DOG_TIMEOUT; - INIT_WORK(&sp->rst_timer_task, - (void (*)(void *)) s2io_restart_nic, dev); - INIT_WORK(&sp->set_link_task, - (void (*)(void *)) s2io_set_link, sp); + INIT_WORK(&sp->rst_timer_task, s2io_restart_nic); + INIT_WORK(&sp->set_link_task, s2io_set_link); pci_save_state(sp->pdev); diff --git a/drivers/net/s2io.h b/drivers/net/s2io.h index 12b719f4d00..3b0bafd273c 100644 --- a/drivers/net/s2io.h +++ b/drivers/net/s2io.h @@ -1000,7 +1000,7 @@ s2io_msix_fifo_handle(int irq, void *dev_id); static irqreturn_t s2io_isr(int irq, void *dev_id); static int verify_xena_quiescence(nic_t *sp, u64 val64, int flag); static const struct ethtool_ops netdev_ethtool_ops; -static void s2io_set_link(unsigned long data); +static void s2io_set_link(struct work_struct *work); static int s2io_set_swapper(nic_t * sp); static void s2io_card_down(nic_t *nic); static int s2io_card_up(nic_t *nic); diff --git a/drivers/net/sis190.c b/drivers/net/sis190.c index aaba458584f..b70ed79d412 100644 --- a/drivers/net/sis190.c +++ b/drivers/net/sis190.c @@ -280,6 +280,7 @@ enum sis190_feature { struct sis190_private { void __iomem *mmio_addr; struct pci_dev *pci_dev; + struct net_device *dev; struct net_device_stats stats; spinlock_t lock; u32 rx_buf_sz; @@ -897,10 +898,11 @@ static void sis190_hw_start(struct net_device *dev) netif_start_queue(dev); } -static void sis190_phy_task(void * data) +static void sis190_phy_task(struct work_struct *work) { - struct net_device *dev = data; - struct sis190_private *tp = netdev_priv(dev); + struct sis190_private *tp = + container_of(work, struct sis190_private, phy_task); + struct net_device *dev = tp->dev; void __iomem *ioaddr = tp->mmio_addr; int phy_id = tp->mii_if.phy_id; u16 val; @@ -1047,7 +1049,7 @@ static int sis190_open(struct net_device *dev) if (rc < 0) goto err_free_rx_1; - INIT_WORK(&tp->phy_task, sis190_phy_task, dev); + INIT_WORK(&tp->phy_task, sis190_phy_task); sis190_request_timer(dev); @@ -1436,6 +1438,7 @@ static struct net_device * __devinit sis190_init_board(struct pci_dev *pdev) SET_NETDEV_DEV(dev, &pdev->dev); tp = netdev_priv(dev); + tp->dev = dev; tp->msg_enable = netif_msg_init(debug.msg_enable, SIS190_MSG_DEFAULT); rc = pci_enable_device(pdev); @@ -1798,7 +1801,7 @@ static int __devinit sis190_init_one(struct pci_dev *pdev, sis190_init_rxfilter(dev); - INIT_WORK(&tp->phy_task, sis190_phy_task, dev); + INIT_WORK(&tp->phy_task, sis190_phy_task); dev->open = sis190_open; dev->stop = sis190_close; diff --git a/drivers/net/skge.c b/drivers/net/skge.c index b2949035f66..3b67614372a 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -1327,10 +1327,11 @@ static void xm_check_link(struct net_device *dev) * Since internal PHY is wired to a level triggered pin, can't * get an interrupt when carrier is detected. */ -static void xm_link_timer(void *arg) +static void xm_link_timer(struct work_struct *work) { - struct net_device *dev = arg; - struct skge_port *skge = netdev_priv(arg); + struct skge_port *skge = + container_of(work, struct skge_port, link_thread.work); + struct net_device *dev = skge->netdev; struct skge_hw *hw = skge->hw; int port = skge->port; @@ -3072,9 +3073,9 @@ static void skge_error_irq(struct skge_hw *hw) * because accessing phy registers requires spin wait which might * cause excess interrupt latency. */ -static void skge_extirq(void *arg) +static void skge_extirq(struct work_struct *work) { - struct skge_hw *hw = arg; + struct skge_hw *hw = container_of(work, struct skge_hw, phy_work); int port; mutex_lock(&hw->phy_mutex); @@ -3456,7 +3457,7 @@ static struct net_device *skge_devinit(struct skge_hw *hw, int port, skge->port = port; /* Only used for Genesis XMAC */ - INIT_WORK(&skge->link_thread, xm_link_timer, dev); + INIT_DELAYED_WORK(&skge->link_thread, xm_link_timer); if (hw->chip_id != CHIP_ID_GENESIS) { dev->features |= NETIF_F_IP_CSUM | NETIF_F_SG; @@ -3543,7 +3544,7 @@ static int __devinit skge_probe(struct pci_dev *pdev, hw->pdev = pdev; mutex_init(&hw->phy_mutex); - INIT_WORK(&hw->phy_work, skge_extirq, hw); + INIT_WORK(&hw->phy_work, skge_extirq); spin_lock_init(&hw->hw_lock); hw->regs = ioremap_nocache(pci_resource_start(pdev, 0), 0x4000); diff --git a/drivers/net/skge.h b/drivers/net/skge.h index 537c0aaa1db..23e5275d920 100644 --- a/drivers/net/skge.h +++ b/drivers/net/skge.h @@ -2456,7 +2456,7 @@ struct skge_port { struct net_device_stats net_stats; - struct work_struct link_thread; + struct delayed_work link_thread; enum pause_control flow_control; enum pause_status flow_status; u8 rx_csum; diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 418138dd6c6..f88fcac0e46 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -1945,10 +1945,11 @@ spider_net_stop(struct net_device *netdev) * called as task when tx hangs, resets interface (if interface is up) */ static void -spider_net_tx_timeout_task(void *data) +spider_net_tx_timeout_task(struct work_struct *work) { - struct net_device *netdev = data; - struct spider_net_card *card = netdev_priv(netdev); + struct spider_net_card *card = + container_of(work, struct spider_net_card, tx_timeout_task); + struct net_device *netdev = card->netdev; if (!(netdev->flags & IFF_UP)) goto out; @@ -2122,7 +2123,7 @@ spider_net_alloc_card(void) card = netdev_priv(netdev); card->netdev = netdev; card->msg_enable = SPIDER_NET_DEFAULT_MSG; - INIT_WORK(&card->tx_timeout_task, spider_net_tx_timeout_task, netdev); + INIT_WORK(&card->tx_timeout_task, spider_net_tx_timeout_task); init_waitqueue_head(&card->waitq); atomic_set(&card->tx_timeout_task_counter, 0); diff --git a/drivers/net/sungem.c b/drivers/net/sungem.c index 253e96e7ad2..004d651681a 100644 --- a/drivers/net/sungem.c +++ b/drivers/net/sungem.c @@ -2281,9 +2281,9 @@ static void gem_do_stop(struct net_device *dev, int wol) } } -static void gem_reset_task(void *data) +static void gem_reset_task(struct work_struct *work) { - struct gem *gp = (struct gem *) data; + struct gem *gp = container_of(work, struct gem, reset_task); mutex_lock(&gp->pm_mutex); @@ -3043,7 +3043,7 @@ static int __devinit gem_init_one(struct pci_dev *pdev, gp->link_timer.function = gem_link_timer; gp->link_timer.data = (unsigned long) gp; - INIT_WORK(&gp->reset_task, gem_reset_task, gp); + INIT_WORK(&gp->reset_task, gem_reset_task); gp->lstate = link_down; gp->timer_ticks = 0; diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index c20bb998e0e..d9123c9adc1 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -3654,9 +3654,9 @@ static void tg3_poll_controller(struct net_device *dev) } #endif -static void tg3_reset_task(void *_data) +static void tg3_reset_task(struct work_struct *work) { - struct tg3 *tp = _data; + struct tg3 *tp = container_of(work, struct tg3, reset_task); unsigned int restart_timer; tg3_full_lock(tp, 0); @@ -11734,7 +11734,7 @@ static int __devinit tg3_init_one(struct pci_dev *pdev, #endif spin_lock_init(&tp->lock); spin_lock_init(&tp->indirect_lock); - INIT_WORK(&tp->reset_task, tg3_reset_task, tp); + INIT_WORK(&tp->reset_task, tg3_reset_task); tp->regs = ioremap_nocache(tg3reg_base, tg3reg_len); if (tp->regs == 0UL) { diff --git a/drivers/net/tlan.c b/drivers/net/tlan.c index e14f5a00f65..f85f0025112 100644 --- a/drivers/net/tlan.c +++ b/drivers/net/tlan.c @@ -296,6 +296,7 @@ static void TLan_SetMulticastList( struct net_device *); static int TLan_ioctl( struct net_device *dev, struct ifreq *rq, int cmd); static int TLan_probe1( struct pci_dev *pdev, long ioaddr, int irq, int rev, const struct pci_device_id *ent); static void TLan_tx_timeout( struct net_device *dev); +static void TLan_tx_timeout_work(struct work_struct *work); static int tlan_init_one( struct pci_dev *pdev, const struct pci_device_id *ent); static u32 TLan_HandleInvalid( struct net_device *, u16 ); @@ -562,6 +563,7 @@ static int __devinit TLan_probe1(struct pci_dev *pdev, priv = netdev_priv(dev); priv->pciDev = pdev; + priv->dev = dev; /* Is this a PCI device? */ if (pdev) { @@ -634,7 +636,7 @@ static int __devinit TLan_probe1(struct pci_dev *pdev, /* This will be used when we get an adapter error from * within our irq handler */ - INIT_WORK(&priv->tlan_tqueue, (void *)(void*)TLan_tx_timeout, dev); + INIT_WORK(&priv->tlan_tqueue, TLan_tx_timeout_work); spin_lock_init(&priv->lock); @@ -1040,6 +1042,25 @@ static void TLan_tx_timeout(struct net_device *dev) } + /*************************************************************** + * TLan_tx_timeout_work + * + * Returns: nothing + * + * Params: + * work work item of device which timed out + * + **************************************************************/ + +static void TLan_tx_timeout_work(struct work_struct *work) +{ + TLanPrivateInfo *priv = + container_of(work, TLanPrivateInfo, tlan_tqueue); + + TLan_tx_timeout(priv->dev); +} + + /*************************************************************** * TLan_StartTx diff --git a/drivers/net/tlan.h b/drivers/net/tlan.h index a44e2f2ef62..41ce0b66593 100644 --- a/drivers/net/tlan.h +++ b/drivers/net/tlan.h @@ -170,6 +170,7 @@ typedef u8 TLanBuffer[TLAN_MAX_FRAME_SIZE]; typedef struct tlan_private_tag { struct net_device *nextDevice; struct pci_dev *pciDev; + struct net_device *dev; void *dmaStorage; dma_addr_t dmaStorageDMA; unsigned int dmaSize; diff --git a/drivers/net/tulip/21142.c b/drivers/net/tulip/21142.c index fa3a2bb105a..942b839ccc5 100644 --- a/drivers/net/tulip/21142.c +++ b/drivers/net/tulip/21142.c @@ -26,10 +26,11 @@ static u16 t21142_csr15[] = { 0x0008, 0x0006, 0x000E, 0x0008, 0x0008, }; /* Handle the 21143 uniquely: do autoselect with NWay, not the EEPROM list of available transceivers. */ -void t21142_media_task(void *data) +void t21142_media_task(struct work_struct *work) { - struct net_device *dev = data; - struct tulip_private *tp = netdev_priv(dev); + struct tulip_private *tp = + container_of(work, struct tulip_private, media_work); + struct net_device *dev = tp->dev; void __iomem *ioaddr = tp->base_addr; int csr12 = ioread32(ioaddr + CSR12); int next_tick = 60*HZ; diff --git a/drivers/net/tulip/timer.c b/drivers/net/tulip/timer.c index 066e5d6bcbd..df326fe1cc8 100644 --- a/drivers/net/tulip/timer.c +++ b/drivers/net/tulip/timer.c @@ -18,10 +18,11 @@ #include "tulip.h" -void tulip_media_task(void *data) +void tulip_media_task(struct work_struct *work) { - struct net_device *dev = data; - struct tulip_private *tp = netdev_priv(dev); + struct tulip_private *tp = + container_of(work, struct tulip_private, media_work); + struct net_device *dev = tp->dev; void __iomem *ioaddr = tp->base_addr; u32 csr12 = ioread32(ioaddr + CSR12); int next_tick = 2*HZ; diff --git a/drivers/net/tulip/tulip.h b/drivers/net/tulip/tulip.h index ad107f45c7b..25f25da7691 100644 --- a/drivers/net/tulip/tulip.h +++ b/drivers/net/tulip/tulip.h @@ -44,7 +44,7 @@ struct tulip_chip_table { int valid_intrs; /* CSR7 interrupt enable settings */ int flags; void (*media_timer) (unsigned long); - void (*media_task) (void *); + work_func_t media_task; }; @@ -392,6 +392,7 @@ struct tulip_private { int csr12_shadow; int pad0; /* Used for 8-byte alignment */ struct work_struct media_work; + struct net_device *dev; }; @@ -406,7 +407,7 @@ struct eeprom_fixup { /* 21142.c */ extern u16 t21142_csr14[]; -void t21142_media_task(void *data); +void t21142_media_task(struct work_struct *work); void t21142_start_nway(struct net_device *dev); void t21142_lnk_change(struct net_device *dev, int csr5); @@ -444,7 +445,7 @@ void pnic_lnk_change(struct net_device *dev, int csr5); void pnic_timer(unsigned long data); /* timer.c */ -void tulip_media_task(void *data); +void tulip_media_task(struct work_struct *work); void mxic_timer(unsigned long data); void comet_timer(unsigned long data); diff --git a/drivers/net/tulip/tulip_core.c b/drivers/net/tulip/tulip_core.c index 0aee618f883..5a35354aa52 100644 --- a/drivers/net/tulip/tulip_core.c +++ b/drivers/net/tulip/tulip_core.c @@ -1367,6 +1367,7 @@ static int __devinit tulip_init_one (struct pci_dev *pdev, * it is zeroed and aligned in alloc_etherdev */ tp = netdev_priv(dev); + tp->dev = dev; tp->rx_ring = pci_alloc_consistent(pdev, sizeof(struct tulip_rx_desc) * RX_RING_SIZE + @@ -1389,7 +1390,7 @@ static int __devinit tulip_init_one (struct pci_dev *pdev, tp->timer.data = (unsigned long)dev; tp->timer.function = tulip_tbl[tp->chip_id].media_timer; - INIT_WORK(&tp->media_work, tulip_tbl[tp->chip_id].media_task, dev); + INIT_WORK(&tp->media_work, tulip_tbl[tp->chip_id].media_task); dev->base_addr = (unsigned long)ioaddr; diff --git a/drivers/net/wan/pc300_tty.c b/drivers/net/wan/pc300_tty.c index 931cbdf6d79..b2a23aed442 100644 --- a/drivers/net/wan/pc300_tty.c +++ b/drivers/net/wan/pc300_tty.c @@ -125,8 +125,8 @@ static int cpc_tty_write_room(struct tty_struct *tty); static int cpc_tty_chars_in_buffer(struct tty_struct *tty); static void cpc_tty_flush_buffer(struct tty_struct *tty); static void cpc_tty_hangup(struct tty_struct *tty); -static void cpc_tty_rx_work(void *data); -static void cpc_tty_tx_work(void *data); +static void cpc_tty_rx_work(struct work_struct *work); +static void cpc_tty_tx_work(struct work_struct *work); static int cpc_tty_send_to_card(pc300dev_t *dev,void *buf, int len); static void cpc_tty_trace(pc300dev_t *dev, char* buf, int len, char rxtx); static void cpc_tty_signal_off(pc300dev_t *pc300dev, unsigned char); @@ -261,8 +261,8 @@ void cpc_tty_init(pc300dev_t *pc300dev) cpc_tty->tty_minor = port + CPC_TTY_MINOR_START; cpc_tty->pc300dev = pc300dev; - INIT_WORK(&cpc_tty->tty_tx_work, cpc_tty_tx_work, (void *)cpc_tty); - INIT_WORK(&cpc_tty->tty_rx_work, cpc_tty_rx_work, (void *)port); + INIT_WORK(&cpc_tty->tty_tx_work, cpc_tty_tx_work); + INIT_WORK(&cpc_tty->tty_rx_work, cpc_tty_rx_work); cpc_tty->buf_rx.first = cpc_tty->buf_rx.last = NULL; @@ -659,21 +659,23 @@ static void cpc_tty_hangup(struct tty_struct *tty) * o call the line disc. read * o free memory */ -static void cpc_tty_rx_work(void * data) +static void cpc_tty_rx_work(struct work_struct *work) { + st_cpc_tty_area *cpc_tty; unsigned long port; int i, j; - st_cpc_tty_area *cpc_tty; volatile st_cpc_rx_buf *buf; char flags=0,flg_rx=1; struct tty_ldisc *ld; if (cpc_tty_cnt == 0) return; - for (i=0; (i < 4) && flg_rx ; i++) { flg_rx = 0; - port = (unsigned long)data; + + cpc_tty = container_of(work, st_cpc_tty_area, tty_rx_work); + port = cpc_tty - cpc_tty_area; + for (j=0; j < CPC_TTY_NPORTS; j++) { cpc_tty = &cpc_tty_area[port]; @@ -882,9 +884,10 @@ void cpc_tty_receive(pc300dev_t *pc300dev) * o if need call line discipline wakeup * o call wake_up_interruptible */ -static void cpc_tty_tx_work(void *data) +static void cpc_tty_tx_work(struct work_struct *work) { - st_cpc_tty_area *cpc_tty = (st_cpc_tty_area *) data; + st_cpc_tty_area *cpc_tty = + container_of(work, st_cpc_tty_area, tty_tx_work); struct tty_struct *tty; CPC_TTY_DBG("%s: cpc_tty_tx_work init\n",cpc_tty->name); diff --git a/drivers/net/wireless/bcm43xx/bcm43xx.h b/drivers/net/wireless/bcm43xx/bcm43xx.h index d6a8bf09878..fbc0c087f53 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx.h +++ b/drivers/net/wireless/bcm43xx/bcm43xx.h @@ -787,7 +787,7 @@ struct bcm43xx_private { struct tasklet_struct isr_tasklet; /* Periodic tasks */ - struct work_struct periodic_work; + struct delayed_work periodic_work; unsigned int periodic_state; struct work_struct restart_work; diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c index a1b783813d8..728a9b789fd 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c @@ -3177,9 +3177,10 @@ static int estimate_periodic_work_badness(unsigned int state) return badness; } -static void bcm43xx_periodic_work_handler(void *d) +static void bcm43xx_periodic_work_handler(struct work_struct *work) { - struct bcm43xx_private *bcm = d; + struct bcm43xx_private *bcm = + container_of(work, struct bcm43xx_private, periodic_work.work); struct net_device *net_dev = bcm->net_dev; unsigned long flags; u32 savedirqs = 0; @@ -3242,11 +3243,11 @@ void bcm43xx_periodic_tasks_delete(struct bcm43xx_private *bcm) void bcm43xx_periodic_tasks_setup(struct bcm43xx_private *bcm) { - struct work_struct *work = &(bcm->periodic_work); + struct delayed_work *work = &bcm->periodic_work; assert(bcm43xx_status(bcm) == BCM43xx_STAT_INITIALIZED); - INIT_WORK(work, bcm43xx_periodic_work_handler, bcm); - schedule_work(work); + INIT_DELAYED_WORK(work, bcm43xx_periodic_work_handler); + schedule_delayed_work(work, 0); } static void bcm43xx_security_init(struct bcm43xx_private *bcm) @@ -3598,7 +3599,7 @@ static int bcm43xx_init_board(struct bcm43xx_private *bcm) bcm43xx_periodic_tasks_setup(bcm); /*FIXME: This should be handled by softmac instead. */ - schedule_work(&bcm->softmac->associnfo.work); + schedule_delayed_work(&bcm->softmac->associnfo.work, 0); out: mutex_unlock(&(bcm)->mutex); @@ -4149,9 +4150,10 @@ static void __devexit bcm43xx_remove_one(struct pci_dev *pdev) /* Hard-reset the chip. Do not call this directly. * Use bcm43xx_controller_restart() */ -static void bcm43xx_chip_reset(void *_bcm) +static void bcm43xx_chip_reset(struct work_struct *work) { - struct bcm43xx_private *bcm = _bcm; + struct bcm43xx_private *bcm = + container_of(work, struct bcm43xx_private, restart_work); struct bcm43xx_phyinfo *phy; int err = -ENODEV; @@ -4178,7 +4180,7 @@ void bcm43xx_controller_restart(struct bcm43xx_private *bcm, const char *reason) if (bcm43xx_status(bcm) != BCM43xx_STAT_INITIALIZED) return; printk(KERN_ERR PFX "Controller RESET (%s) ...\n", reason); - INIT_WORK(&bcm->restart_work, bcm43xx_chip_reset, bcm); + INIT_WORK(&bcm->restart_work, bcm43xx_chip_reset); schedule_work(&bcm->restart_work); } diff --git a/drivers/net/wireless/hostap/hostap.h b/drivers/net/wireless/hostap/hostap.h index e663518bd57..e89c890d16f 100644 --- a/drivers/net/wireless/hostap/hostap.h +++ b/drivers/net/wireless/hostap/hostap.h @@ -35,7 +35,7 @@ int hostap_80211_get_hdrlen(u16 fc); struct net_device_stats *hostap_get_stats(struct net_device *dev); void hostap_setup_dev(struct net_device *dev, local_info_t *local, int main_dev); -void hostap_set_multicast_list_queue(void *data); +void hostap_set_multicast_list_queue(struct work_struct *work); int hostap_set_hostapd(local_info_t *local, int val, int rtnl_locked); int hostap_set_hostapd_sta(local_info_t *local, int val, int rtnl_locked); void hostap_cleanup(local_info_t *local); diff --git a/drivers/net/wireless/hostap/hostap_ap.c b/drivers/net/wireless/hostap/hostap_ap.c index ba13125024c..08bc57a4b89 100644 --- a/drivers/net/wireless/hostap/hostap_ap.c +++ b/drivers/net/wireless/hostap/hostap_ap.c @@ -49,10 +49,10 @@ MODULE_PARM_DESC(autom_ap_wds, "Add WDS connections to other APs " static struct sta_info* ap_get_sta(struct ap_data *ap, u8 *sta); static void hostap_event_expired_sta(struct net_device *dev, struct sta_info *sta); -static void handle_add_proc_queue(void *data); +static void handle_add_proc_queue(struct work_struct *work); #ifndef PRISM2_NO_KERNEL_IEEE80211_MGMT -static void handle_wds_oper_queue(void *data); +static void handle_wds_oper_queue(struct work_struct *work); static void prism2_send_mgmt(struct net_device *dev, u16 type_subtype, char *body, int body_len, u8 *addr, u16 tx_cb_idx); @@ -807,7 +807,7 @@ void hostap_init_data(local_info_t *local) INIT_LIST_HEAD(&ap->sta_list); /* Initialize task queue structure for AP management */ - INIT_WORK(&local->ap->add_sta_proc_queue, handle_add_proc_queue, ap); + INIT_WORK(&local->ap->add_sta_proc_queue, handle_add_proc_queue); ap->tx_callback_idx = hostap_tx_callback_register(local, hostap_ap_tx_cb, ap); @@ -815,7 +815,7 @@ void hostap_init_data(local_info_t *local) printk(KERN_WARNING "%s: failed to register TX callback for " "AP\n", local->dev->name); #ifndef PRISM2_NO_KERNEL_IEEE80211_MGMT - INIT_WORK(&local->ap->wds_oper_queue, handle_wds_oper_queue, local); + INIT_WORK(&local->ap->wds_oper_queue, handle_wds_oper_queue); ap->tx_callback_auth = hostap_tx_callback_register(local, hostap_ap_tx_cb_auth, ap); @@ -1062,9 +1062,10 @@ static int prism2_sta_proc_read(char *page, char **start, off_t off, } -static void handle_add_proc_queue(void *data) +static void handle_add_proc_queue(struct work_struct *work) { - struct ap_data *ap = (struct ap_data *) data; + struct ap_data *ap = container_of(work, struct ap_data, + add_sta_proc_queue); struct sta_info *sta; char name[20]; struct add_sta_proc_data *entry, *prev; @@ -1952,9 +1953,11 @@ static void handle_pspoll(local_info_t *local, #ifndef PRISM2_NO_KERNEL_IEEE80211_MGMT -static void handle_wds_oper_queue(void *data) +static void handle_wds_oper_queue(struct work_struct *work) { - local_info_t *local = data; + struct ap_data *ap = container_of(work, struct ap_data, + wds_oper_queue); + local_info_t *local = ap->local; struct wds_oper_data *entry, *prev; spin_lock_bh(&local->lock); diff --git a/drivers/net/wireless/hostap/hostap_hw.c b/drivers/net/wireless/hostap/hostap_hw.c index ed00ebb6e7f..c19e68636a1 100644 --- a/drivers/net/wireless/hostap/hostap_hw.c +++ b/drivers/net/wireless/hostap/hostap_hw.c @@ -1645,9 +1645,9 @@ static void prism2_schedule_reset(local_info_t *local) /* Called only as scheduled task after noticing card timeout in interrupt * context */ -static void handle_reset_queue(void *data) +static void handle_reset_queue(struct work_struct *work) { - local_info_t *local = (local_info_t *) data; + local_info_t *local = container_of(work, local_info_t, reset_queue); printk(KERN_DEBUG "%s: scheduled card reset\n", local->dev->name); prism2_hw_reset(local->dev); @@ -2896,9 +2896,10 @@ static void hostap_passive_scan(unsigned long data) /* Called only as a scheduled task when communications quality values should * be updated. */ -static void handle_comms_qual_update(void *data) +static void handle_comms_qual_update(struct work_struct *work) { - local_info_t *local = data; + local_info_t *local = + container_of(work, local_info_t, comms_qual_update); prism2_update_comms_qual(local->dev); } @@ -3050,9 +3051,9 @@ static int prism2_set_tim(struct net_device *dev, int aid, int set) } -static void handle_set_tim_queue(void *data) +static void handle_set_tim_queue(struct work_struct *work) { - local_info_t *local = (local_info_t *) data; + local_info_t *local = container_of(work, local_info_t, set_tim_queue); struct set_tim_data *entry; u16 val; @@ -3209,15 +3210,15 @@ prism2_init_local_data(struct prism2_helper_functions *funcs, int card_idx, local->scan_channel_mask = 0xffff; /* Initialize task queue structures */ - INIT_WORK(&local->reset_queue, handle_reset_queue, local); + INIT_WORK(&local->reset_queue, handle_reset_queue); INIT_WORK(&local->set_multicast_list_queue, - hostap_set_multicast_list_queue, local->dev); + hostap_set_multicast_list_queue); - INIT_WORK(&local->set_tim_queue, handle_set_tim_queue, local); + INIT_WORK(&local->set_tim_queue, handle_set_tim_queue); INIT_LIST_HEAD(&local->set_tim_list); spin_lock_init(&local->set_tim_lock); - INIT_WORK(&local->comms_qual_update, handle_comms_qual_update, local); + INIT_WORK(&local->comms_qual_update, handle_comms_qual_update); /* Initialize tasklets for handling hardware IRQ related operations * outside hw IRQ handler */ diff --git a/drivers/net/wireless/hostap/hostap_info.c b/drivers/net/wireless/hostap/hostap_info.c index 50f72d831cf..5fd2b1ad7f5 100644 --- a/drivers/net/wireless/hostap/hostap_info.c +++ b/drivers/net/wireless/hostap/hostap_info.c @@ -474,9 +474,9 @@ static void handle_info_queue_scanresults(local_info_t *local) /* Called only as scheduled task after receiving info frames (used to avoid * pending too much time in HW IRQ handler). */ -static void handle_info_queue(void *data) +static void handle_info_queue(struct work_struct *work) { - local_info_t *local = (local_info_t *) data; + local_info_t *local = container_of(work, local_info_t, info_queue); if (test_and_clear_bit(PRISM2_INFO_PENDING_LINKSTATUS, &local->pending_info)) @@ -493,7 +493,7 @@ void hostap_info_init(local_info_t *local) { skb_queue_head_init(&local->info_list); #ifndef PRISM2_NO_STATION_MODES - INIT_WORK(&local->info_queue, handle_info_queue, local); + INIT_WORK(&local->info_queue, handle_info_queue); #endif /* PRISM2_NO_STATION_MODES */ } diff --git a/drivers/net/wireless/hostap/hostap_main.c b/drivers/net/wireless/hostap/hostap_main.c index 53374fcba77..0796be9d9e7 100644 --- a/drivers/net/wireless/hostap/hostap_main.c +++ b/drivers/net/wireless/hostap/hostap_main.c @@ -767,14 +767,14 @@ static int prism2_set_mac_address(struct net_device *dev, void *p) /* TODO: to be further implemented as soon as Prism2 fully supports * GroupAddresses and correct documentation is available */ -void hostap_set_multicast_list_queue(void *data) +void hostap_set_multicast_list_queue(struct work_struct *work) { - struct net_device *dev = (struct net_device *) data; + local_info_t *local = + container_of(work, local_info_t, set_multicast_list_queue); + struct net_device *dev = local->dev; struct hostap_interface *iface; - local_info_t *local; iface = netdev_priv(dev); - local = iface->local; if (hostap_set_word(dev, HFA384X_RID_PROMISCUOUSMODE, local->is_promisc)) { printk(KERN_INFO "%s: %sabling promiscuous mode failed\n", diff --git a/drivers/net/wireless/ipw2100.c b/drivers/net/wireless/ipw2100.c index 4e4eaa2a99c..0f554373a60 100644 --- a/drivers/net/wireless/ipw2100.c +++ b/drivers/net/wireless/ipw2100.c @@ -316,7 +316,7 @@ static void ipw2100_release_firmware(struct ipw2100_priv *priv, struct ipw2100_fw *fw); static int ipw2100_ucode_download(struct ipw2100_priv *priv, struct ipw2100_fw *fw); -static void ipw2100_wx_event_work(struct ipw2100_priv *priv); +static void ipw2100_wx_event_work(struct work_struct *work); static struct iw_statistics *ipw2100_wx_wireless_stats(struct net_device *dev); static struct iw_handler_def ipw2100_wx_handler_def; @@ -679,7 +679,8 @@ static void schedule_reset(struct ipw2100_priv *priv) queue_delayed_work(priv->workqueue, &priv->reset_work, priv->reset_backoff * HZ); else - queue_work(priv->workqueue, &priv->reset_work); + queue_delayed_work(priv->workqueue, &priv->reset_work, + 0); if (priv->reset_backoff < MAX_RESET_BACKOFF) priv->reset_backoff++; @@ -1873,8 +1874,10 @@ static void ipw2100_down(struct ipw2100_priv *priv) netif_stop_queue(priv->net_dev); } -static void ipw2100_reset_adapter(struct ipw2100_priv *priv) +static void ipw2100_reset_adapter(struct work_struct *work) { + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, reset_work.work); unsigned long flags; union iwreq_data wrqu = { .ap_addr = { @@ -2071,9 +2074,9 @@ static void isr_indicate_association_lost(struct ipw2100_priv *priv, u32 status) return; if (priv->status & STATUS_SECURITY_UPDATED) - queue_work(priv->workqueue, &priv->security_work); + queue_delayed_work(priv->workqueue, &priv->security_work, 0); - queue_work(priv->workqueue, &priv->wx_event_work); + queue_delayed_work(priv->workqueue, &priv->wx_event_work, 0); } static void isr_indicate_rf_kill(struct ipw2100_priv *priv, u32 status) @@ -5524,8 +5527,11 @@ static int ipw2100_configure_security(struct ipw2100_priv *priv, int batch_mode) return err; } -static void ipw2100_security_work(struct ipw2100_priv *priv) +static void ipw2100_security_work(struct work_struct *work) { + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, security_work.work); + /* If we happen to have reconnected before we get a chance to * process this, then update the security settings--which causes * a disassociation to occur */ @@ -5748,7 +5754,7 @@ static int ipw2100_set_address(struct net_device *dev, void *p) priv->reset_backoff = 0; mutex_unlock(&priv->action_mutex); - ipw2100_reset_adapter(priv); + ipw2100_reset_adapter(&priv->reset_work.work); return 0; done: @@ -5923,9 +5929,10 @@ static const struct ethtool_ops ipw2100_ethtool_ops = { .get_drvinfo = ipw_ethtool_get_drvinfo, }; -static void ipw2100_hang_check(void *adapter) +static void ipw2100_hang_check(struct work_struct *work) { - struct ipw2100_priv *priv = adapter; + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, hang_check.work); unsigned long flags; u32 rtc = 0xa5a5a5a5; u32 len = sizeof(rtc); @@ -5965,9 +5972,10 @@ static void ipw2100_hang_check(void *adapter) spin_unlock_irqrestore(&priv->low_lock, flags); } -static void ipw2100_rf_kill(void *adapter) +static void ipw2100_rf_kill(struct work_struct *work) { - struct ipw2100_priv *priv = adapter; + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, rf_kill.work); unsigned long flags; spin_lock_irqsave(&priv->low_lock, flags); @@ -6117,14 +6125,11 @@ static struct net_device *ipw2100_alloc_device(struct pci_dev *pci_dev, priv->workqueue = create_workqueue(DRV_NAME); - INIT_WORK(&priv->reset_work, - (void (*)(void *))ipw2100_reset_adapter, priv); - INIT_WORK(&priv->security_work, - (void (*)(void *))ipw2100_security_work, priv); - INIT_WORK(&priv->wx_event_work, - (void (*)(void *))ipw2100_wx_event_work, priv); - INIT_WORK(&priv->hang_check, ipw2100_hang_check, priv); - INIT_WORK(&priv->rf_kill, ipw2100_rf_kill, priv); + INIT_DELAYED_WORK(&priv->reset_work, ipw2100_reset_adapter); + INIT_DELAYED_WORK(&priv->security_work, ipw2100_security_work); + INIT_DELAYED_WORK(&priv->wx_event_work, ipw2100_wx_event_work); + INIT_DELAYED_WORK(&priv->hang_check, ipw2100_hang_check); + INIT_DELAYED_WORK(&priv->rf_kill, ipw2100_rf_kill); tasklet_init(&priv->irq_tasklet, (void (*)(unsigned long)) ipw2100_irq_tasklet, (unsigned long)priv); @@ -8290,8 +8295,10 @@ static struct iw_handler_def ipw2100_wx_handler_def = { .get_wireless_stats = ipw2100_wx_wireless_stats, }; -static void ipw2100_wx_event_work(struct ipw2100_priv *priv) +static void ipw2100_wx_event_work(struct work_struct *work) { + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, wx_event_work.work); union iwreq_data wrqu; int len = ETH_ALEN; diff --git a/drivers/net/wireless/ipw2100.h b/drivers/net/wireless/ipw2100.h index 55b7227198d..de7d384d38a 100644 --- a/drivers/net/wireless/ipw2100.h +++ b/drivers/net/wireless/ipw2100.h @@ -583,11 +583,11 @@ struct ipw2100_priv { struct tasklet_struct irq_tasklet; struct workqueue_struct *workqueue; - struct work_struct reset_work; - struct work_struct security_work; - struct work_struct wx_event_work; - struct work_struct hang_check; - struct work_struct rf_kill; + struct delayed_work reset_work; + struct delayed_work security_work; + struct delayed_work wx_event_work; + struct delayed_work hang_check; + struct delayed_work rf_kill; u32 interrupts; int tx_interrupts; diff --git a/drivers/net/wireless/ipw2200.c b/drivers/net/wireless/ipw2200.c index 1f742814a01..587a0918fa5 100644 --- a/drivers/net/wireless/ipw2200.c +++ b/drivers/net/wireless/ipw2200.c @@ -187,9 +187,9 @@ static struct ipw_rx_queue *ipw_rx_queue_alloc(struct ipw_priv *); static void ipw_rx_queue_free(struct ipw_priv *, struct ipw_rx_queue *); static void ipw_rx_queue_replenish(void *); static int ipw_up(struct ipw_priv *); -static void ipw_bg_up(void *); +static void ipw_bg_up(struct work_struct *work); static void ipw_down(struct ipw_priv *); -static void ipw_bg_down(void *); +static void ipw_bg_down(struct work_struct *work); static int ipw_config(struct ipw_priv *); static int init_supported_rates(struct ipw_priv *priv, struct ipw_supported_rates *prates); @@ -862,11 +862,12 @@ static void ipw_led_link_on(struct ipw_priv *priv) spin_unlock_irqrestore(&priv->lock, flags); } -static void ipw_bg_led_link_on(void *data) +static void ipw_bg_led_link_on(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, led_link_on.work); mutex_lock(&priv->mutex); - ipw_led_link_on(data); + ipw_led_link_on(priv); mutex_unlock(&priv->mutex); } @@ -906,11 +907,12 @@ static void ipw_led_link_off(struct ipw_priv *priv) spin_unlock_irqrestore(&priv->lock, flags); } -static void ipw_bg_led_link_off(void *data) +static void ipw_bg_led_link_off(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, led_link_off.work); mutex_lock(&priv->mutex); - ipw_led_link_off(data); + ipw_led_link_off(priv); mutex_unlock(&priv->mutex); } @@ -985,11 +987,12 @@ static void ipw_led_activity_off(struct ipw_priv *priv) spin_unlock_irqrestore(&priv->lock, flags); } -static void ipw_bg_led_activity_off(void *data) +static void ipw_bg_led_activity_off(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, led_act_off.work); mutex_lock(&priv->mutex); - ipw_led_activity_off(data); + ipw_led_activity_off(priv); mutex_unlock(&priv->mutex); } @@ -2228,11 +2231,12 @@ static void ipw_adapter_restart(void *adapter) } } -static void ipw_bg_adapter_restart(void *data) +static void ipw_bg_adapter_restart(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, adapter_restart); mutex_lock(&priv->mutex); - ipw_adapter_restart(data); + ipw_adapter_restart(priv); mutex_unlock(&priv->mutex); } @@ -2249,11 +2253,12 @@ static void ipw_scan_check(void *data) } } -static void ipw_bg_scan_check(void *data) +static void ipw_bg_scan_check(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, scan_check.work); mutex_lock(&priv->mutex); - ipw_scan_check(data); + ipw_scan_check(priv); mutex_unlock(&priv->mutex); } @@ -3831,17 +3836,19 @@ static int ipw_disassociate(void *data) return 1; } -static void ipw_bg_disassociate(void *data) +static void ipw_bg_disassociate(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, disassociate); mutex_lock(&priv->mutex); - ipw_disassociate(data); + ipw_disassociate(priv); mutex_unlock(&priv->mutex); } -static void ipw_system_config(void *data) +static void ipw_system_config(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, system_config); #ifdef CONFIG_IPW2200_PROMISCUOUS if (priv->prom_net_dev && netif_running(priv->prom_net_dev)) { @@ -4208,11 +4215,12 @@ static void ipw_gather_stats(struct ipw_priv *priv) IPW_STATS_INTERVAL); } -static void ipw_bg_gather_stats(void *data) +static void ipw_bg_gather_stats(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, gather_stats.work); mutex_lock(&priv->mutex); - ipw_gather_stats(data); + ipw_gather_stats(priv); mutex_unlock(&priv->mutex); } @@ -4268,8 +4276,8 @@ static void ipw_handle_missed_beacon(struct ipw_priv *priv, if (!(priv->status & STATUS_ROAMING)) { priv->status |= STATUS_ROAMING; if (!(priv->status & STATUS_SCANNING)) - queue_work(priv->workqueue, - &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); } return; } @@ -4607,8 +4615,8 @@ static void ipw_rx_notification(struct ipw_priv *priv, #ifdef CONFIG_IPW2200_MONITOR if (priv->ieee->iw_mode == IW_MODE_MONITOR) { priv->status |= STATUS_SCAN_FORCED; - queue_work(priv->workqueue, - &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); break; } priv->status &= ~STATUS_SCAN_FORCED; @@ -4631,8 +4639,8 @@ static void ipw_rx_notification(struct ipw_priv *priv, /* Don't schedule if we aborted the scan */ priv->status &= ~STATUS_ROAMING; } else if (priv->status & STATUS_SCAN_PENDING) - queue_work(priv->workqueue, - &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); else if (priv->config & CFG_BACKGROUND_SCAN && priv->status & STATUS_ASSOCIATED) queue_delayed_work(priv->workqueue, @@ -5055,11 +5063,12 @@ static void ipw_rx_queue_replenish(void *data) ipw_rx_queue_restock(priv); } -static void ipw_bg_rx_queue_replenish(void *data) +static void ipw_bg_rx_queue_replenish(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, rx_replenish); mutex_lock(&priv->mutex); - ipw_rx_queue_replenish(data); + ipw_rx_queue_replenish(priv); mutex_unlock(&priv->mutex); } @@ -5489,9 +5498,10 @@ static int ipw_find_adhoc_network(struct ipw_priv *priv, return 1; } -static void ipw_merge_adhoc_network(void *data) +static void ipw_merge_adhoc_network(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, merge_networks); struct ieee80211_network *network = NULL; struct ipw_network_match match = { .network = priv->assoc_network @@ -5948,11 +5958,12 @@ static void ipw_adhoc_check(void *data) priv->assoc_request.beacon_interval); } -static void ipw_bg_adhoc_check(void *data) +static void ipw_bg_adhoc_check(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, adhoc_check.work); mutex_lock(&priv->mutex); - ipw_adhoc_check(data); + ipw_adhoc_check(priv); mutex_unlock(&priv->mutex); } @@ -6299,19 +6310,26 @@ done: return err; } -static int ipw_request_passive_scan(struct ipw_priv *priv) { - return ipw_request_scan_helper(priv, IW_SCAN_TYPE_PASSIVE); +static void ipw_request_passive_scan(struct work_struct *work) +{ + struct ipw_priv *priv = + container_of(work, struct ipw_priv, request_passive_scan); + ipw_request_scan_helper(priv, IW_SCAN_TYPE_PASSIVE); } -static int ipw_request_scan(struct ipw_priv *priv) { - return ipw_request_scan_helper(priv, IW_SCAN_TYPE_ACTIVE); +static void ipw_request_scan(struct work_struct *work) +{ + struct ipw_priv *priv = + container_of(work, struct ipw_priv, request_scan.work); + ipw_request_scan_helper(priv, IW_SCAN_TYPE_ACTIVE); } -static void ipw_bg_abort_scan(void *data) +static void ipw_bg_abort_scan(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, abort_scan); mutex_lock(&priv->mutex); - ipw_abort_scan(data); + ipw_abort_scan(priv); mutex_unlock(&priv->mutex); } @@ -7084,9 +7102,10 @@ static int ipw_qos_set_tx_queue_command(struct ipw_priv *priv, /* * background support to run QoS activate functionality */ -static void ipw_bg_qos_activate(void *data) +static void ipw_bg_qos_activate(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, qos_activate); if (priv == NULL) return; @@ -7394,11 +7413,12 @@ static void ipw_roam(void *data) priv->status &= ~STATUS_ROAMING; } -static void ipw_bg_roam(void *data) +static void ipw_bg_roam(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, roam); mutex_lock(&priv->mutex); - ipw_roam(data); + ipw_roam(priv); mutex_unlock(&priv->mutex); } @@ -7479,8 +7499,8 @@ static int ipw_associate(void *data) &priv->request_scan, SCAN_INTERVAL); else - queue_work(priv->workqueue, - &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); } return 0; @@ -7491,11 +7511,12 @@ static int ipw_associate(void *data) return 1; } -static void ipw_bg_associate(void *data) +static void ipw_bg_associate(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, associate); mutex_lock(&priv->mutex); - ipw_associate(data); + ipw_associate(priv); mutex_unlock(&priv->mutex); } @@ -9410,7 +9431,7 @@ static int ipw_wx_set_scan(struct net_device *dev, IPW_DEBUG_WX("Start scan\n"); - queue_work(priv->workqueue, &priv->request_scan); + queue_delayed_work(priv->workqueue, &priv->request_scan, 0); return 0; } @@ -10547,11 +10568,12 @@ static void ipw_rf_kill(void *adapter) spin_unlock_irqrestore(&priv->lock, flags); } -static void ipw_bg_rf_kill(void *data) +static void ipw_bg_rf_kill(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, rf_kill.work); mutex_lock(&priv->mutex); - ipw_rf_kill(data); + ipw_rf_kill(priv); mutex_unlock(&priv->mutex); } @@ -10582,11 +10604,12 @@ static void ipw_link_up(struct ipw_priv *priv) queue_delayed_work(priv->workqueue, &priv->request_scan, HZ); } -static void ipw_bg_link_up(void *data) +static void ipw_bg_link_up(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, link_up); mutex_lock(&priv->mutex); - ipw_link_up(data); + ipw_link_up(priv); mutex_unlock(&priv->mutex); } @@ -10606,15 +10629,16 @@ static void ipw_link_down(struct ipw_priv *priv) if (!(priv->status & STATUS_EXIT_PENDING)) { /* Queue up another scan... */ - queue_work(priv->workqueue, &priv->request_scan); + queue_delayed_work(priv->workqueue, &priv->request_scan, 0); } } -static void ipw_bg_link_down(void *data) +static void ipw_bg_link_down(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, link_down); mutex_lock(&priv->mutex); - ipw_link_down(data); + ipw_link_down(priv); mutex_unlock(&priv->mutex); } @@ -10626,38 +10650,30 @@ static int ipw_setup_deferred_work(struct ipw_priv *priv) init_waitqueue_head(&priv->wait_command_queue); init_waitqueue_head(&priv->wait_state); - INIT_WORK(&priv->adhoc_check, ipw_bg_adhoc_check, priv); - INIT_WORK(&priv->associate, ipw_bg_associate, priv); - INIT_WORK(&priv->disassociate, ipw_bg_disassociate, priv); - INIT_WORK(&priv->system_config, ipw_system_config, priv); - INIT_WORK(&priv->rx_replenish, ipw_bg_rx_queue_replenish, priv); - INIT_WORK(&priv->adapter_restart, ipw_bg_adapter_restart, priv); - INIT_WORK(&priv->rf_kill, ipw_bg_rf_kill, priv); - INIT_WORK(&priv->up, (void (*)(void *))ipw_bg_up, priv); - INIT_WORK(&priv->down, (void (*)(void *))ipw_bg_down, priv); - INIT_WORK(&priv->request_scan, - (void (*)(void *))ipw_request_scan, priv); - INIT_WORK(&priv->request_passive_scan, - (void (*)(void *))ipw_request_passive_scan, priv); - INIT_WORK(&priv->gather_stats, - (void (*)(void *))ipw_bg_gather_stats, priv); - INIT_WORK(&priv->abort_scan, (void (*)(void *))ipw_bg_abort_scan, priv); - INIT_WORK(&priv->roam, ipw_bg_roam, priv); - INIT_WORK(&priv->scan_check, ipw_bg_scan_check, priv); - INIT_WORK(&priv->link_up, (void (*)(void *))ipw_bg_link_up, priv); - INIT_WORK(&priv->link_down, (void (*)(void *))ipw_bg_link_down, priv); - INIT_WORK(&priv->led_link_on, (void (*)(void *))ipw_bg_led_link_on, - priv); - INIT_WORK(&priv->led_link_off, (void (*)(void *))ipw_bg_led_link_off, - priv); - INIT_WORK(&priv->led_act_off, (void (*)(void *))ipw_bg_led_activity_off, - priv); - INIT_WORK(&priv->merge_networks, - (void (*)(void *))ipw_merge_adhoc_network, priv); + INIT_DELAYED_WORK(&priv->adhoc_check, ipw_bg_adhoc_check); + INIT_WORK(&priv->associate, ipw_bg_associate); + INIT_WORK(&priv->disassociate, ipw_bg_disassociate); + INIT_WORK(&priv->system_config, ipw_system_config); + INIT_WORK(&priv->rx_replenish, ipw_bg_rx_queue_replenish); + INIT_WORK(&priv->adapter_restart, ipw_bg_adapter_restart); + INIT_DELAYED_WORK(&priv->rf_kill, ipw_bg_rf_kill); + INIT_WORK(&priv->up, ipw_bg_up); + INIT_WORK(&priv->down, ipw_bg_down); + INIT_DELAYED_WORK(&priv->request_scan, ipw_request_scan); + INIT_WORK(&priv->request_passive_scan, ipw_request_passive_scan); + INIT_DELAYED_WORK(&priv->gather_stats, ipw_bg_gather_stats); + INIT_WORK(&priv->abort_scan, ipw_bg_abort_scan); + INIT_WORK(&priv->roam, ipw_bg_roam); + INIT_DELAYED_WORK(&priv->scan_check, ipw_bg_scan_check); + INIT_WORK(&priv->link_up, ipw_bg_link_up); + INIT_WORK(&priv->link_down, ipw_bg_link_down); + INIT_DELAYED_WORK(&priv->led_link_on, ipw_bg_led_link_on); + INIT_DELAYED_WORK(&priv->led_link_off, ipw_bg_led_link_off); + INIT_DELAYED_WORK(&priv->led_act_off, ipw_bg_led_activity_off); + INIT_WORK(&priv->merge_networks, ipw_merge_adhoc_network); #ifdef CONFIG_IPW2200_QOS - INIT_WORK(&priv->qos_activate, (void (*)(void *))ipw_bg_qos_activate, - priv); + INIT_WORK(&priv->qos_activate, ipw_bg_qos_activate); #endif /* CONFIG_IPW2200_QOS */ tasklet_init(&priv->irq_tasklet, (void (*)(unsigned long)) @@ -11190,7 +11206,8 @@ static int ipw_up(struct ipw_priv *priv) /* If configure to try and auto-associate, kick * off a scan. */ - queue_work(priv->workqueue, &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); return 0; } @@ -11211,11 +11228,12 @@ static int ipw_up(struct ipw_priv *priv) return -EIO; } -static void ipw_bg_up(void *data) +static void ipw_bg_up(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, up); mutex_lock(&priv->mutex); - ipw_up(data); + ipw_up(priv); mutex_unlock(&priv->mutex); } @@ -11282,11 +11300,12 @@ static void ipw_down(struct ipw_priv *priv) ipw_led_radio_off(priv); } -static void ipw_bg_down(void *data) +static void ipw_bg_down(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, down); mutex_lock(&priv->mutex); - ipw_down(data); + ipw_down(priv); mutex_unlock(&priv->mutex); } diff --git a/drivers/net/wireless/ipw2200.h b/drivers/net/wireless/ipw2200.h index dad5eedefbf..626a240a87d 100644 --- a/drivers/net/wireless/ipw2200.h +++ b/drivers/net/wireless/ipw2200.h @@ -1290,21 +1290,21 @@ struct ipw_priv { struct workqueue_struct *workqueue; - struct work_struct adhoc_check; + struct delayed_work adhoc_check; struct work_struct associate; struct work_struct disassociate; struct work_struct system_config; struct work_struct rx_replenish; - struct work_struct request_scan; + struct delayed_work request_scan; struct work_struct request_passive_scan; struct work_struct adapter_restart; - struct work_struct rf_kill; + struct delayed_work rf_kill; struct work_struct up; struct work_struct down; - struct work_struct gather_stats; + struct delayed_work gather_stats; struct work_struct abort_scan; struct work_struct roam; - struct work_struct scan_check; + struct delayed_work scan_check; struct work_struct link_up; struct work_struct link_down; @@ -1319,9 +1319,9 @@ struct ipw_priv { u32 led_ofdm_on; u32 led_ofdm_off; - struct work_struct led_link_on; - struct work_struct led_link_off; - struct work_struct led_act_off; + struct delayed_work led_link_on; + struct delayed_work led_link_off; + struct delayed_work led_act_off; struct work_struct merge_networks; struct ipw_cmd_log *cmdlog; diff --git a/drivers/net/wireless/orinoco.c b/drivers/net/wireless/orinoco.c index 336cabac13b..936c888e03e 100644 --- a/drivers/net/wireless/orinoco.c +++ b/drivers/net/wireless/orinoco.c @@ -980,9 +980,11 @@ static void print_linkstatus(struct net_device *dev, u16 status) } /* Search scan results for requested BSSID, join it if found */ -static void orinoco_join_ap(struct net_device *dev) +static void orinoco_join_ap(struct work_struct *work) { - struct orinoco_private *priv = netdev_priv(dev); + struct orinoco_private *priv = + container_of(work, struct orinoco_private, join_work); + struct net_device *dev = priv->ndev; struct hermes *hw = &priv->hw; int err; unsigned long flags; @@ -1055,9 +1057,11 @@ static void orinoco_join_ap(struct net_device *dev) } /* Send new BSSID to userspace */ -static void orinoco_send_wevents(struct net_device *dev) +static void orinoco_send_wevents(struct work_struct *work) { - struct orinoco_private *priv = netdev_priv(dev); + struct orinoco_private *priv = + container_of(work, struct orinoco_private, wevent_work); + struct net_device *dev = priv->ndev; struct hermes *hw = &priv->hw; union iwreq_data wrqu; int err; @@ -1864,9 +1868,11 @@ __orinoco_set_multicast_list(struct net_device *dev) /* This must be called from user context, without locks held - use * schedule_work() */ -static void orinoco_reset(struct net_device *dev) +static void orinoco_reset(struct work_struct *work) { - struct orinoco_private *priv = netdev_priv(dev); + struct orinoco_private *priv = + container_of(work, struct orinoco_private, reset_work); + struct net_device *dev = priv->ndev; struct hermes *hw = &priv->hw; int err; unsigned long flags; @@ -2434,9 +2440,9 @@ struct net_device *alloc_orinocodev(int sizeof_card, priv->hw_unavailable = 1; /* orinoco_init() must clear this * before anything else touches the * hardware */ - INIT_WORK(&priv->reset_work, (void (*)(void *))orinoco_reset, dev); - INIT_WORK(&priv->join_work, (void (*)(void *))orinoco_join_ap, dev); - INIT_WORK(&priv->wevent_work, (void (*)(void *))orinoco_send_wevents, dev); + INIT_WORK(&priv->reset_work, orinoco_reset); + INIT_WORK(&priv->join_work, orinoco_join_ap); + INIT_WORK(&priv->wevent_work, orinoco_send_wevents); netif_carrier_off(dev); priv->last_linkstatus = 0xffff; @@ -3608,7 +3614,7 @@ static int orinoco_ioctl_reset(struct net_device *dev, printk(KERN_DEBUG "%s: Forcing reset!\n", dev->name); /* Firmware reset */ - orinoco_reset(dev); + orinoco_reset(&priv->reset_work); } else { printk(KERN_DEBUG "%s: Force scheduling reset!\n", dev->name); @@ -4154,7 +4160,7 @@ static int orinoco_ioctl_commit(struct net_device *dev, return 0; if (priv->broken_disableport) { - orinoco_reset(dev); + orinoco_reset(&priv->reset_work); return 0; } diff --git a/drivers/net/wireless/prism54/isl_ioctl.c b/drivers/net/wireless/prism54/isl_ioctl.c index 286325ca329..e7700b4257e 100644 --- a/drivers/net/wireless/prism54/isl_ioctl.c +++ b/drivers/net/wireless/prism54/isl_ioctl.c @@ -158,8 +158,9 @@ prism54_mib_init(islpci_private *priv) * schedule_work(), thus we can as well use sleeping semaphore * locking */ void -prism54_update_stats(islpci_private *priv) +prism54_update_stats(struct work_struct *work) { + islpci_private *priv = container_of(work, islpci_private, stats_work); char *data; int j; struct obj_bss bss, *bss2; @@ -2494,9 +2495,10 @@ prism54_process_trap_helper(islpci_private *priv, enum oid_num_t oid, * interrupt context, no locks held. */ void -prism54_process_trap(void *data) +prism54_process_trap(struct work_struct *work) { - struct islpci_mgmtframe *frame = data; + struct islpci_mgmtframe *frame = + container_of(work, struct islpci_mgmtframe, ws); struct net_device *ndev = frame->ndev; enum oid_num_t n = mgt_oidtonum(frame->header->oid); diff --git a/drivers/net/wireless/prism54/isl_ioctl.h b/drivers/net/wireless/prism54/isl_ioctl.h index 65f33acd0a4..0802fa64996 100644 --- a/drivers/net/wireless/prism54/isl_ioctl.h +++ b/drivers/net/wireless/prism54/isl_ioctl.h @@ -32,12 +32,12 @@ void prism54_mib_init(islpci_private *); struct iw_statistics *prism54_get_wireless_stats(struct net_device *); -void prism54_update_stats(islpci_private *); +void prism54_update_stats(struct work_struct *); void prism54_acl_init(struct islpci_acl *); void prism54_acl_clean(struct islpci_acl *); -void prism54_process_trap(void *); +void prism54_process_trap(struct work_struct *); void prism54_wpa_bss_ie_init(islpci_private *priv); void prism54_wpa_bss_ie_clean(islpci_private *priv); diff --git a/drivers/net/wireless/prism54/islpci_dev.c b/drivers/net/wireless/prism54/islpci_dev.c index ec1c00f19eb..e35fcb2543c 100644 --- a/drivers/net/wireless/prism54/islpci_dev.c +++ b/drivers/net/wireless/prism54/islpci_dev.c @@ -861,11 +861,10 @@ islpci_setup(struct pci_dev *pdev) priv->state_off = 1; /* initialize workqueue's */ - INIT_WORK(&priv->stats_work, - (void (*)(void *)) prism54_update_stats, priv); + INIT_WORK(&priv->stats_work, prism54_update_stats); priv->stats_timestamp = 0; - INIT_WORK(&priv->reset_task, islpci_do_reset_and_wake, priv); + INIT_WORK(&priv->reset_task, islpci_do_reset_and_wake); priv->reset_task_pending = 0; /* allocate various memory areas */ diff --git a/drivers/net/wireless/prism54/islpci_eth.c b/drivers/net/wireless/prism54/islpci_eth.c index a8261d8454d..103a3787773 100644 --- a/drivers/net/wireless/prism54/islpci_eth.c +++ b/drivers/net/wireless/prism54/islpci_eth.c @@ -482,9 +482,9 @@ islpci_eth_receive(islpci_private *priv) } void -islpci_do_reset_and_wake(void *data) +islpci_do_reset_and_wake(struct work_struct *work) { - islpci_private *priv = (islpci_private *) data; + islpci_private *priv = container_of(work, islpci_private, reset_task); islpci_reset(priv, 1); netif_wake_queue(priv->ndev); priv->reset_task_pending = 0; diff --git a/drivers/net/wireless/prism54/islpci_eth.h b/drivers/net/wireless/prism54/islpci_eth.h index bc9d7a60b8d..99d37eda9f0 100644 --- a/drivers/net/wireless/prism54/islpci_eth.h +++ b/drivers/net/wireless/prism54/islpci_eth.h @@ -68,6 +68,6 @@ void islpci_eth_cleanup_transmit(islpci_private *, isl38xx_control_block *); int islpci_eth_transmit(struct sk_buff *, struct net_device *); int islpci_eth_receive(islpci_private *); void islpci_eth_tx_timeout(struct net_device *); -void islpci_do_reset_and_wake(void *data); +void islpci_do_reset_and_wake(struct work_struct *); #endif /* _ISL_GEN_H */ diff --git a/drivers/net/wireless/prism54/islpci_mgt.c b/drivers/net/wireless/prism54/islpci_mgt.c index 2e061a80b29..656ec9fa712 100644 --- a/drivers/net/wireless/prism54/islpci_mgt.c +++ b/drivers/net/wireless/prism54/islpci_mgt.c @@ -387,7 +387,7 @@ islpci_mgt_receive(struct net_device *ndev) /* Create work to handle trap out of interrupt * context. */ - INIT_WORK(&frame->ws, prism54_process_trap, frame); + INIT_WORK(&frame->ws, prism54_process_trap); schedule_work(&frame->ws); } else { diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c index a7d29bddb29..5e4f4b70737 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.c +++ b/drivers/net/wireless/zd1211rw/zd_mac.c @@ -1090,9 +1090,10 @@ void zd_dump_rx_status(const struct rx_status *status) #define LINK_LED_WORK_DELAY HZ -static void link_led_handler(void *p) +static void link_led_handler(struct work_struct *work) { - struct zd_mac *mac = p; + struct zd_mac *mac = + container_of(work, struct zd_mac, housekeeping.link_led_work.work); struct zd_chip *chip = &mac->chip; struct ieee80211softmac_device *sm = ieee80211_priv(mac->netdev); int is_associated; @@ -1113,7 +1114,7 @@ static void link_led_handler(void *p) static void housekeeping_init(struct zd_mac *mac) { - INIT_WORK(&mac->housekeeping.link_led_work, link_led_handler, mac); + INIT_DELAYED_WORK(&mac->housekeeping.link_led_work, link_led_handler); } static void housekeeping_enable(struct zd_mac *mac) diff --git a/drivers/net/wireless/zd1211rw/zd_mac.h b/drivers/net/wireless/zd1211rw/zd_mac.h index b8ea3de7924..7957cac3de2 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.h +++ b/drivers/net/wireless/zd1211rw/zd_mac.h @@ -121,7 +121,7 @@ enum mac_flags { }; struct housekeeping { - struct work_struct link_led_work; + struct delayed_work link_led_work; }; #define ZD_MAC_STATS_BUFFER_SIZE 16 diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c index fc4bc9b94c7..a83c3db7d18 100644 --- a/drivers/oprofile/cpu_buffer.c +++ b/drivers/oprofile/cpu_buffer.c @@ -29,7 +29,7 @@ struct oprofile_cpu_buffer cpu_buffer[NR_CPUS] __cacheline_aligned; -static void wq_sync_buffer(void *); +static void wq_sync_buffer(struct work_struct *work); #define DEFAULT_TIMER_EXPIRE (HZ / 10) static int work_enabled; @@ -65,7 +65,7 @@ int alloc_cpu_buffers(void) b->sample_received = 0; b->sample_lost_overflow = 0; b->cpu = i; - INIT_WORK(&b->work, wq_sync_buffer, b); + INIT_DELAYED_WORK(&b->work, wq_sync_buffer); } return 0; @@ -282,9 +282,10 @@ void oprofile_add_trace(unsigned long pc) * By using schedule_delayed_work_on and then schedule_delayed_work * we guarantee this will stay on the correct cpu */ -static void wq_sync_buffer(void * data) +static void wq_sync_buffer(struct work_struct *work) { - struct oprofile_cpu_buffer * b = data; + struct oprofile_cpu_buffer * b = + container_of(work, struct oprofile_cpu_buffer, work.work); if (b->cpu != smp_processor_id()) { printk("WQ on CPU%d, prefer CPU%d\n", smp_processor_id(), b->cpu); diff --git a/drivers/oprofile/cpu_buffer.h b/drivers/oprofile/cpu_buffer.h index 09abb80e057..49900d9e323 100644 --- a/drivers/oprofile/cpu_buffer.h +++ b/drivers/oprofile/cpu_buffer.h @@ -43,7 +43,7 @@ struct oprofile_cpu_buffer { unsigned long sample_lost_overflow; unsigned long backtrace_aborted; int cpu; - struct work_struct work; + struct delayed_work work; } ____cacheline_aligned; extern struct oprofile_cpu_buffer cpu_buffer[]; diff --git a/drivers/pci/hotplug/shpchp.h b/drivers/pci/hotplug/shpchp.h index ea2087c3414..50757695844 100644 --- a/drivers/pci/hotplug/shpchp.h +++ b/drivers/pci/hotplug/shpchp.h @@ -70,7 +70,7 @@ struct slot { struct hotplug_slot *hotplug_slot; struct list_head slot_list; char name[SLOT_NAME_SIZE]; - struct work_struct work; /* work for button event */ + struct delayed_work work; /* work for button event */ struct mutex lock; }; @@ -187,7 +187,7 @@ extern int shpchp_configure_device(struct slot *p_slot); extern int shpchp_unconfigure_device(struct slot *p_slot); extern void shpchp_remove_ctrl_files(struct controller *ctrl); extern void cleanup_slots(struct controller *ctrl); -extern void queue_pushbutton_work(void *data); +extern void queue_pushbutton_work(struct work_struct *work); #ifdef CONFIG_ACPI diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c index 235c18a2239..4eac85b3d90 100644 --- a/drivers/pci/hotplug/shpchp_core.c +++ b/drivers/pci/hotplug/shpchp_core.c @@ -159,7 +159,7 @@ static int init_slots(struct controller *ctrl) goto error_info; slot->number = sun; - INIT_WORK(&slot->work, queue_pushbutton_work, slot); + INIT_DELAYED_WORK(&slot->work, queue_pushbutton_work); /* register this slot with the hotplug pci core */ hotplug_slot->private = slot; diff --git a/drivers/pci/hotplug/shpchp_ctrl.c b/drivers/pci/hotplug/shpchp_ctrl.c index c39901dbff2..158ac783609 100644 --- a/drivers/pci/hotplug/shpchp_ctrl.c +++ b/drivers/pci/hotplug/shpchp_ctrl.c @@ -36,7 +36,7 @@ #include "../pci.h" #include "shpchp.h" -static void interrupt_event_handler(void *data); +static void interrupt_event_handler(struct work_struct *work); static int shpchp_enable_slot(struct slot *p_slot); static int shpchp_disable_slot(struct slot *p_slot); @@ -50,7 +50,7 @@ static int queue_interrupt_event(struct slot *p_slot, u32 event_type) info->event_type = event_type; info->p_slot = p_slot; - INIT_WORK(&info->work, interrupt_event_handler, info); + INIT_WORK(&info->work, interrupt_event_handler); schedule_work(&info->work); @@ -408,9 +408,10 @@ struct pushbutton_work_info { * Handles all pending events and exits. * */ -static void shpchp_pushbutton_thread(void *data) +static void shpchp_pushbutton_thread(struct work_struct *work) { - struct pushbutton_work_info *info = data; + struct pushbutton_work_info *info = + container_of(work, struct pushbutton_work_info, work); struct slot *p_slot = info->p_slot; mutex_lock(&p_slot->lock); @@ -436,9 +437,9 @@ static void shpchp_pushbutton_thread(void *data) kfree(info); } -void queue_pushbutton_work(void *data) +void queue_pushbutton_work(struct work_struct *work) { - struct slot *p_slot = data; + struct slot *p_slot = container_of(work, struct slot, work.work); struct pushbutton_work_info *info; info = kmalloc(sizeof(*info), GFP_KERNEL); @@ -447,7 +448,7 @@ void queue_pushbutton_work(void *data) return; } info->p_slot = p_slot; - INIT_WORK(&info->work, shpchp_pushbutton_thread, info); + INIT_WORK(&info->work, shpchp_pushbutton_thread); mutex_lock(&p_slot->lock); switch (p_slot->state) { @@ -541,9 +542,9 @@ static void handle_button_press_event(struct slot *p_slot) } } -static void interrupt_event_handler(void *data) +static void interrupt_event_handler(struct work_struct *work) { - struct event_info *info = data; + struct event_info *info = container_of(work, struct event_info, work); struct slot *p_slot = info->p_slot; mutex_lock(&p_slot->lock); diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c index a20d84d707d..e469a46a388 100644 --- a/drivers/pcmcia/ds.c +++ b/drivers/pcmcia/ds.c @@ -698,9 +698,10 @@ static int pcmcia_card_add(struct pcmcia_socket *s) } -static void pcmcia_delayed_add_pseudo_device(void *data) +static void pcmcia_delayed_add_pseudo_device(struct work_struct *work) { - struct pcmcia_socket *s = data; + struct pcmcia_socket *s = + container_of(work, struct pcmcia_socket, device_add); pcmcia_device_add(s, 0); s->pcmcia_state.device_add_pending = 0; } @@ -1246,7 +1247,7 @@ static int __devinit pcmcia_bus_add_socket(struct class_device *class_dev, init_waitqueue_head(&socket->queue); #endif INIT_LIST_HEAD(&socket->devices_list); - INIT_WORK(&socket->device_add, pcmcia_delayed_add_pseudo_device, socket); + INIT_WORK(&socket->device_add, pcmcia_delayed_add_pseudo_device); memset(&socket->pcmcia_state, 0, sizeof(u8)); socket->device_count = 0; diff --git a/drivers/rtc/rtc-dev.c b/drivers/rtc/rtc-dev.c index 583789c66cd..dcf5f86461f 100644 --- a/drivers/rtc/rtc-dev.c +++ b/drivers/rtc/rtc-dev.c @@ -53,9 +53,10 @@ static int rtc_dev_open(struct inode *inode, struct file *file) * Routine to poll RTC seconds field for change as often as possible, * after first RTC_UIE use timer to reduce polling */ -static void rtc_uie_task(void *data) +static void rtc_uie_task(struct work_struct *work) { - struct rtc_device *rtc = data; + struct rtc_device *rtc = + container_of(work, struct rtc_device, uie_task); struct rtc_time tm; int num = 0; int err; @@ -398,7 +399,7 @@ static int rtc_dev_add_device(struct class_device *class_dev, spin_lock_init(&rtc->irq_lock); init_waitqueue_head(&rtc->irq_queue); #ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL - INIT_WORK(&rtc->uie_task, rtc_uie_task, rtc); + INIT_WORK(&rtc->uie_task, rtc_uie_task); setup_timer(&rtc->uie_timer, rtc_uie_timer, (unsigned long)rtc); #endif diff --git a/drivers/scsi/NCR5380.c b/drivers/scsi/NCR5380.c index a6aa9107288..bb3cb336054 100644 --- a/drivers/scsi/NCR5380.c +++ b/drivers/scsi/NCR5380.c @@ -849,7 +849,7 @@ static int __devinit NCR5380_init(struct Scsi_Host *instance, int flags) hostdata->issue_queue = NULL; hostdata->disconnected_queue = NULL; - INIT_WORK(&hostdata->coroutine, NCR5380_main, hostdata); + INIT_DELAYED_WORK(&hostdata->coroutine, NCR5380_main); #ifdef NCR5380_STATS for (i = 0; i < 8; ++i) { @@ -1016,7 +1016,7 @@ static int NCR5380_queue_command(Scsi_Cmnd * cmd, void (*done) (Scsi_Cmnd *)) /* Run the coroutine if it isn't already running. */ /* Kick off command processing */ - schedule_work(&hostdata->coroutine); + schedule_delayed_work(&hostdata->coroutine, 0); return 0; } @@ -1033,9 +1033,10 @@ static int NCR5380_queue_command(Scsi_Cmnd * cmd, void (*done) (Scsi_Cmnd *)) * host lock and called routines may take the isa dma lock. */ -static void NCR5380_main(void *p) +static void NCR5380_main(struct work_struct *work) { - struct NCR5380_hostdata *hostdata = p; + struct NCR5380_hostdata *hostdata = + container_of(work, struct NCR5380_hostdata, coroutine.work); struct Scsi_Host *instance = hostdata->host; Scsi_Cmnd *tmp, *prev; int done; @@ -1221,7 +1222,7 @@ static irqreturn_t NCR5380_intr(int irq, void *dev_id) } /* if BASR_IRQ */ spin_unlock_irqrestore(instance->host_lock, flags); if(!done) - schedule_work(&hostdata->coroutine); + schedule_delayed_work(&hostdata->coroutine, 0); } while (!done); return IRQ_HANDLED; } diff --git a/drivers/scsi/NCR5380.h b/drivers/scsi/NCR5380.h index 1bc73de496b..713a108c02e 100644 --- a/drivers/scsi/NCR5380.h +++ b/drivers/scsi/NCR5380.h @@ -271,7 +271,7 @@ struct NCR5380_hostdata { unsigned long time_expires; /* in jiffies, set prior to sleeping */ int select_time; /* timer in select for target response */ volatile Scsi_Cmnd *selecting; - struct work_struct coroutine; /* our co-routine */ + struct delayed_work coroutine; /* our co-routine */ #ifdef NCR5380_STATS unsigned timebase; /* Base for time calcs */ long time_read[8]; /* time to do reads */ @@ -298,7 +298,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance); #ifndef DONT_USE_INTR static irqreturn_t NCR5380_intr(int irq, void *dev_id); #endif -static void NCR5380_main(void *ptr); +static void NCR5380_main(struct work_struct *work); static void NCR5380_print_options(struct Scsi_Host *instance); #ifdef NDEBUG static void NCR5380_print_phase(struct Scsi_Host *instance); diff --git a/drivers/scsi/aha152x.c b/drivers/scsi/aha152x.c index 306f46b85a5..0cec742d12e 100644 --- a/drivers/scsi/aha152x.c +++ b/drivers/scsi/aha152x.c @@ -1443,7 +1443,7 @@ static struct work_struct aha152x_tq; * Run service completions on the card with interrupts enabled. * */ -static void run(void) +static void run(struct work_struct *work) { struct aha152x_hostdata *hd; @@ -1499,7 +1499,7 @@ static irqreturn_t intr(int irqno, void *dev_id) HOSTDATA(shpnt)->service=1; /* Poke the BH handler */ - INIT_WORK(&aha152x_tq, (void *) run, NULL); + INIT_WORK(&aha152x_tq, run); schedule_work(&aha152x_tq); } DO_UNLOCK(flags); diff --git a/drivers/scsi/imm.c b/drivers/scsi/imm.c index e31f6122106..0464c182c57 100644 --- a/drivers/scsi/imm.c +++ b/drivers/scsi/imm.c @@ -36,7 +36,7 @@ typedef struct { int base_hi; /* Hi Base address for ECP-ISA chipset */ int mode; /* Transfer mode */ struct scsi_cmnd *cur_cmd; /* Current queued command */ - struct work_struct imm_tq; /* Polling interrupt stuff */ + struct delayed_work imm_tq; /* Polling interrupt stuff */ unsigned long jstart; /* Jiffies at start */ unsigned failed:1; /* Failure flag */ unsigned dp:1; /* Data phase present */ @@ -733,9 +733,9 @@ static int imm_completion(struct scsi_cmnd *cmd) * the scheduler's task queue to generate a stream of call-backs and * complete the request when the drive is ready. */ -static void imm_interrupt(void *data) +static void imm_interrupt(struct work_struct *work) { - imm_struct *dev = (imm_struct *) data; + imm_struct *dev = container_of(work, imm_struct, imm_tq.work); struct scsi_cmnd *cmd = dev->cur_cmd; struct Scsi_Host *host = cmd->device->host; unsigned long flags; @@ -745,7 +745,6 @@ static void imm_interrupt(void *data) return; } if (imm_engine(dev, cmd)) { - INIT_WORK(&dev->imm_tq, imm_interrupt, (void *) dev); schedule_delayed_work(&dev->imm_tq, 1); return; } @@ -953,8 +952,7 @@ static int imm_queuecommand(struct scsi_cmnd *cmd, cmd->result = DID_ERROR << 16; /* default return code */ cmd->SCp.phase = 0; /* bus free */ - INIT_WORK(&dev->imm_tq, imm_interrupt, dev); - schedule_work(&dev->imm_tq); + schedule_delayed_work(&dev->imm_tq, 0); imm_pb_claim(dev); @@ -1225,7 +1223,7 @@ static int __imm_attach(struct parport *pb) else ports = 8; - INIT_WORK(&dev->imm_tq, imm_interrupt, dev); + INIT_DELAYED_WORK(&dev->imm_tq, imm_interrupt); err = -ENOMEM; host = scsi_host_alloc(&imm_template, sizeof(imm_struct *)); diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 2dde821025f..d51c3e764bb 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -2093,7 +2093,7 @@ static void ipr_release_dump(struct kref *kref) /** * ipr_worker_thread - Worker thread - * @data: ioa config struct + * @work: ioa config struct * * Called at task level from a work thread. This function takes care * of adding and removing device from the mid-layer as configuration @@ -2102,13 +2102,14 @@ static void ipr_release_dump(struct kref *kref) * Return value: * nothing **/ -static void ipr_worker_thread(void *data) +static void ipr_worker_thread(struct work_struct *work) { unsigned long lock_flags; struct ipr_resource_entry *res; struct scsi_device *sdev; struct ipr_dump *dump; - struct ipr_ioa_cfg *ioa_cfg = data; + struct ipr_ioa_cfg *ioa_cfg = + container_of(work, struct ipr_ioa_cfg, work_q); u8 bus, target, lun; int did_work; @@ -6926,7 +6927,7 @@ static void __devinit ipr_init_ioa_cfg(struct ipr_ioa_cfg *ioa_cfg, INIT_LIST_HEAD(&ioa_cfg->hostrcb_pending_q); INIT_LIST_HEAD(&ioa_cfg->free_res_q); INIT_LIST_HEAD(&ioa_cfg->used_res_q); - INIT_WORK(&ioa_cfg->work_q, ipr_worker_thread, ioa_cfg); + INIT_WORK(&ioa_cfg->work_q, ipr_worker_thread); init_waitqueue_head(&ioa_cfg->reset_wait_q); ioa_cfg->sdt_state = INACTIVE; if (ipr_enable_cache) diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index 5d886218948..e11b23c641e 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -719,9 +719,10 @@ again: return rc; } -static void iscsi_xmitworker(void *data) +static void iscsi_xmitworker(struct work_struct *work) { - struct iscsi_conn *conn = data; + struct iscsi_conn *conn = + container_of(work, struct iscsi_conn, xmitwork); int rc; /* * serialize Xmit worker on a per-connection basis. @@ -1512,7 +1513,7 @@ iscsi_conn_setup(struct iscsi_cls_session *cls_session, uint32_t conn_idx) if (conn->mgmtqueue == ERR_PTR(-ENOMEM)) goto mgmtqueue_alloc_fail; - INIT_WORK(&conn->xmitwork, iscsi_xmitworker, conn); + INIT_WORK(&conn->xmitwork, iscsi_xmitworker); /* allocate login_mtask used for the login/text sequences */ spin_lock_bh(&session->lock); diff --git a/drivers/scsi/libsas/sas_discover.c b/drivers/scsi/libsas/sas_discover.c index d977bd492d8..fb7df7b7581 100644 --- a/drivers/scsi/libsas/sas_discover.c +++ b/drivers/scsi/libsas/sas_discover.c @@ -647,10 +647,12 @@ void sas_unregister_domain_devices(struct asd_sas_port *port) * Discover process only interrogates devices in order to discover the * domain. */ -static void sas_discover_domain(void *data) +static void sas_discover_domain(struct work_struct *work) { int error = 0; - struct asd_sas_port *port = data; + struct sas_discovery_event *ev = + container_of(work, struct sas_discovery_event, work); + struct asd_sas_port *port = ev->port; sas_begin_event(DISCE_DISCOVER_DOMAIN, &port->disc.disc_event_lock, &port->disc.pending); @@ -692,10 +694,12 @@ static void sas_discover_domain(void *data) current->pid, error); } -static void sas_revalidate_domain(void *data) +static void sas_revalidate_domain(struct work_struct *work) { int res = 0; - struct asd_sas_port *port = data; + struct sas_discovery_event *ev = + container_of(work, struct sas_discovery_event, work); + struct asd_sas_port *port = ev->port; sas_begin_event(DISCE_REVALIDATE_DOMAIN, &port->disc.disc_event_lock, &port->disc.pending); @@ -722,7 +726,7 @@ int sas_discover_event(struct asd_sas_port *port, enum discover_event ev) BUG_ON(ev >= DISC_NUM_EVENTS); sas_queue_event(ev, &disc->disc_event_lock, &disc->pending, - &disc->disc_work[ev], port->ha->core.shost); + &disc->disc_work[ev].work, port->ha->core.shost); return 0; } @@ -737,13 +741,15 @@ void sas_init_disc(struct sas_discovery *disc, struct asd_sas_port *port) { int i; - static void (*sas_event_fns[DISC_NUM_EVENTS])(void *) = { + static const work_func_t sas_event_fns[DISC_NUM_EVENTS] = { [DISCE_DISCOVER_DOMAIN] = sas_discover_domain, [DISCE_REVALIDATE_DOMAIN] = sas_revalidate_domain, }; spin_lock_init(&disc->disc_event_lock); disc->pending = 0; - for (i = 0; i < DISC_NUM_EVENTS; i++) - INIT_WORK(&disc->disc_work[i], sas_event_fns[i], port); + for (i = 0; i < DISC_NUM_EVENTS; i++) { + INIT_WORK(&disc->disc_work[i].work, sas_event_fns[i]); + disc->disc_work[i].port = port; + } } diff --git a/drivers/scsi/libsas/sas_event.c b/drivers/scsi/libsas/sas_event.c index 19110ed1c89..d83392ee682 100644 --- a/drivers/scsi/libsas/sas_event.c +++ b/drivers/scsi/libsas/sas_event.c @@ -31,7 +31,7 @@ static void notify_ha_event(struct sas_ha_struct *sas_ha, enum ha_event event) BUG_ON(event >= HA_NUM_EVENTS); sas_queue_event(event, &sas_ha->event_lock, &sas_ha->pending, - &sas_ha->ha_events[event], sas_ha->core.shost); + &sas_ha->ha_events[event].work, sas_ha->core.shost); } static void notify_port_event(struct asd_sas_phy *phy, enum port_event event) @@ -41,7 +41,7 @@ static void notify_port_event(struct asd_sas_phy *phy, enum port_event event) BUG_ON(event >= PORT_NUM_EVENTS); sas_queue_event(event, &ha->event_lock, &phy->port_events_pending, - &phy->port_events[event], ha->core.shost); + &phy->port_events[event].work, ha->core.shost); } static void notify_phy_event(struct asd_sas_phy *phy, enum phy_event event) @@ -51,12 +51,12 @@ static void notify_phy_event(struct asd_sas_phy *phy, enum phy_event event) BUG_ON(event >= PHY_NUM_EVENTS); sas_queue_event(event, &ha->event_lock, &phy->phy_events_pending, - &phy->phy_events[event], ha->core.shost); + &phy->phy_events[event].work, ha->core.shost); } int sas_init_events(struct sas_ha_struct *sas_ha) { - static void (*sas_ha_event_fns[HA_NUM_EVENTS])(void *) = { + static const work_func_t sas_ha_event_fns[HA_NUM_EVENTS] = { [HAE_RESET] = sas_hae_reset, }; @@ -64,8 +64,10 @@ int sas_init_events(struct sas_ha_struct *sas_ha) spin_lock_init(&sas_ha->event_lock); - for (i = 0; i < HA_NUM_EVENTS; i++) - INIT_WORK(&sas_ha->ha_events[i], sas_ha_event_fns[i], sas_ha); + for (i = 0; i < HA_NUM_EVENTS; i++) { + INIT_WORK(&sas_ha->ha_events[i].work, sas_ha_event_fns[i]); + sas_ha->ha_events[i].ha = sas_ha; + } sas_ha->notify_ha_event = notify_ha_event; sas_ha->notify_port_event = notify_port_event; diff --git a/drivers/scsi/libsas/sas_init.c b/drivers/scsi/libsas/sas_init.c index c836a237fb7..7b4e9169f44 100644 --- a/drivers/scsi/libsas/sas_init.c +++ b/drivers/scsi/libsas/sas_init.c @@ -65,9 +65,11 @@ void sas_hash_addr(u8 *hashed, const u8 *sas_addr) /* ---------- HA events ---------- */ -void sas_hae_reset(void *data) +void sas_hae_reset(struct work_struct *work) { - struct sas_ha_struct *ha = data; + struct sas_ha_event *ev = + container_of(work, struct sas_ha_event, work); + struct sas_ha_struct *ha = ev->ha; sas_begin_event(HAE_RESET, &ha->event_lock, &ha->pending); diff --git a/drivers/scsi/libsas/sas_internal.h b/drivers/scsi/libsas/sas_internal.h index bffcee47492..137d7e496b6 100644 --- a/drivers/scsi/libsas/sas_internal.h +++ b/drivers/scsi/libsas/sas_internal.h @@ -60,11 +60,11 @@ void sas_shutdown_queue(struct sas_ha_struct *sas_ha); void sas_deform_port(struct asd_sas_phy *phy); -void sas_porte_bytes_dmaed(void *); -void sas_porte_broadcast_rcvd(void *); -void sas_porte_link_reset_err(void *); -void sas_porte_timer_event(void *); -void sas_porte_hard_reset(void *); +void sas_porte_bytes_dmaed(struct work_struct *work); +void sas_porte_broadcast_rcvd(struct work_struct *work); +void sas_porte_link_reset_err(struct work_struct *work); +void sas_porte_timer_event(struct work_struct *work); +void sas_porte_hard_reset(struct work_struct *work); int sas_notify_lldd_dev_found(struct domain_device *); void sas_notify_lldd_dev_gone(struct domain_device *); @@ -75,7 +75,7 @@ int sas_smp_get_phy_events(struct sas_phy *phy); struct domain_device *sas_find_dev_by_rphy(struct sas_rphy *rphy); -void sas_hae_reset(void *); +void sas_hae_reset(struct work_struct *work); static inline void sas_queue_event(int event, spinlock_t *lock, unsigned long *pending, diff --git a/drivers/scsi/libsas/sas_phy.c b/drivers/scsi/libsas/sas_phy.c index 9340cdbae4a..b459c4b635b 100644 --- a/drivers/scsi/libsas/sas_phy.c +++ b/drivers/scsi/libsas/sas_phy.c @@ -30,9 +30,11 @@ /* ---------- Phy events ---------- */ -static void sas_phye_loss_of_signal(void *data) +static void sas_phye_loss_of_signal(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PHYE_LOSS_OF_SIGNAL, &phy->ha->event_lock, &phy->phy_events_pending); @@ -40,18 +42,22 @@ static void sas_phye_loss_of_signal(void *data) sas_deform_port(phy); } -static void sas_phye_oob_done(void *data) +static void sas_phye_oob_done(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PHYE_OOB_DONE, &phy->ha->event_lock, &phy->phy_events_pending); phy->error = 0; } -static void sas_phye_oob_error(void *data) +static void sas_phye_oob_error(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; struct sas_ha_struct *sas_ha = phy->ha; struct asd_sas_port *port = phy->port; struct sas_internal *i = @@ -80,9 +86,11 @@ static void sas_phye_oob_error(void *data) } } -static void sas_phye_spinup_hold(void *data) +static void sas_phye_spinup_hold(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; struct sas_ha_struct *sas_ha = phy->ha; struct sas_internal *i = to_sas_internal(sas_ha->core.shost->transportt); @@ -100,14 +108,14 @@ int sas_register_phys(struct sas_ha_struct *sas_ha) { int i; - static void (*sas_phy_event_fns[PHY_NUM_EVENTS])(void *) = { + static const work_func_t sas_phy_event_fns[PHY_NUM_EVENTS] = { [PHYE_LOSS_OF_SIGNAL] = sas_phye_loss_of_signal, [PHYE_OOB_DONE] = sas_phye_oob_done, [PHYE_OOB_ERROR] = sas_phye_oob_error, [PHYE_SPINUP_HOLD] = sas_phye_spinup_hold, }; - static void (*sas_port_event_fns[PORT_NUM_EVENTS])(void *) = { + static const work_func_t sas_port_event_fns[PORT_NUM_EVENTS] = { [PORTE_BYTES_DMAED] = sas_porte_bytes_dmaed, [PORTE_BROADCAST_RCVD] = sas_porte_broadcast_rcvd, [PORTE_LINK_RESET_ERR] = sas_porte_link_reset_err, @@ -122,13 +130,18 @@ int sas_register_phys(struct sas_ha_struct *sas_ha) phy->error = 0; INIT_LIST_HEAD(&phy->port_phy_el); - for (k = 0; k < PORT_NUM_EVENTS; k++) - INIT_WORK(&phy->port_events[k], sas_port_event_fns[k], - phy); + for (k = 0; k < PORT_NUM_EVENTS; k++) { + INIT_WORK(&phy->port_events[k].work, + sas_port_event_fns[k]); + phy->port_events[k].phy = phy; + } + + for (k = 0; k < PHY_NUM_EVENTS; k++) { + INIT_WORK(&phy->phy_events[k].work, + sas_phy_event_fns[k]); + phy->phy_events[k].phy = phy; + } - for (k = 0; k < PHY_NUM_EVENTS; k++) - INIT_WORK(&phy->phy_events[k], sas_phy_event_fns[k], - phy); phy->port = NULL; phy->ha = sas_ha; spin_lock_init(&phy->frame_rcvd_lock); diff --git a/drivers/scsi/libsas/sas_port.c b/drivers/scsi/libsas/sas_port.c index 253cdcf306a..971c37ceecb 100644 --- a/drivers/scsi/libsas/sas_port.c +++ b/drivers/scsi/libsas/sas_port.c @@ -181,9 +181,11 @@ void sas_deform_port(struct asd_sas_phy *phy) /* ---------- SAS port events ---------- */ -void sas_porte_bytes_dmaed(void *data) +void sas_porte_bytes_dmaed(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PORTE_BYTES_DMAED, &phy->ha->event_lock, &phy->port_events_pending); @@ -191,11 +193,13 @@ void sas_porte_bytes_dmaed(void *data) sas_form_port(phy); } -void sas_porte_broadcast_rcvd(void *data) +void sas_porte_broadcast_rcvd(struct work_struct *work) { + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; unsigned long flags; u32 prim; - struct asd_sas_phy *phy = data; sas_begin_event(PORTE_BROADCAST_RCVD, &phy->ha->event_lock, &phy->port_events_pending); @@ -208,9 +212,11 @@ void sas_porte_broadcast_rcvd(void *data) sas_discover_event(phy->port, DISCE_REVALIDATE_DOMAIN); } -void sas_porte_link_reset_err(void *data) +void sas_porte_link_reset_err(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PORTE_LINK_RESET_ERR, &phy->ha->event_lock, &phy->port_events_pending); @@ -218,9 +224,11 @@ void sas_porte_link_reset_err(void *data) sas_deform_port(phy); } -void sas_porte_timer_event(void *data) +void sas_porte_timer_event(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PORTE_TIMER_EVENT, &phy->ha->event_lock, &phy->port_events_pending); @@ -228,9 +236,11 @@ void sas_porte_timer_event(void *data) sas_deform_port(phy); } -void sas_porte_hard_reset(void *data) +void sas_porte_hard_reset(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PORTE_HARD_RESET, &phy->ha->event_lock, &phy->port_events_pending); diff --git a/drivers/scsi/ppa.c b/drivers/scsi/ppa.c index 89a2a9f11e4..584ba4d6e03 100644 --- a/drivers/scsi/ppa.c +++ b/drivers/scsi/ppa.c @@ -31,7 +31,7 @@ typedef struct { int base; /* Actual port address */ int mode; /* Transfer mode */ struct scsi_cmnd *cur_cmd; /* Current queued command */ - struct work_struct ppa_tq; /* Polling interrupt stuff */ + struct delayed_work ppa_tq; /* Polling interrupt stuff */ unsigned long jstart; /* Jiffies at start */ unsigned long recon_tmo; /* How many usecs to wait for reconnection (6th bit) */ unsigned int failed:1; /* Failure flag */ @@ -627,9 +627,9 @@ static int ppa_completion(struct scsi_cmnd *cmd) * the scheduler's task queue to generate a stream of call-backs and * complete the request when the drive is ready. */ -static void ppa_interrupt(void *data) +static void ppa_interrupt(struct work_struct *work) { - ppa_struct *dev = (ppa_struct *) data; + ppa_struct *dev = container_of(work, ppa_struct, ppa_tq.work); struct scsi_cmnd *cmd = dev->cur_cmd; if (!cmd) { @@ -637,7 +637,6 @@ static void ppa_interrupt(void *data) return; } if (ppa_engine(dev, cmd)) { - dev->ppa_tq.data = (void *) dev; schedule_delayed_work(&dev->ppa_tq, 1); return; } @@ -822,8 +821,7 @@ static int ppa_queuecommand(struct scsi_cmnd *cmd, cmd->result = DID_ERROR << 16; /* default return code */ cmd->SCp.phase = 0; /* bus free */ - dev->ppa_tq.data = dev; - schedule_work(&dev->ppa_tq); + schedule_delayed_work(&dev->ppa_tq, 0); ppa_pb_claim(dev); @@ -1086,7 +1084,7 @@ static int __ppa_attach(struct parport *pb) else ports = 8; - INIT_WORK(&dev->ppa_tq, ppa_interrupt, dev); + INIT_DELAYED_WORK(&dev->ppa_tq, ppa_interrupt); err = -ENOMEM; host = scsi_host_alloc(&ppa_template, sizeof(ppa_struct *)); diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 5b8db610953..bbbc9d039ba 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -1011,9 +1011,10 @@ static int qla4xxx_recover_adapter(struct scsi_qla_host *ha, * the mid-level tries to sleep when it reaches the driver threshold * "host->can_queue". This can cause a panic if we were in our interrupt code. **/ -static void qla4xxx_do_dpc(void *data) +static void qla4xxx_do_dpc(struct work_struct *work) { - struct scsi_qla_host *ha = (struct scsi_qla_host *) data; + struct scsi_qla_host *ha = + container_of(work, struct scsi_qla_host, dpc_work); struct ddb_entry *ddb_entry, *dtemp; DEBUG2(printk("scsi%ld: %s: DPC handler waking up.\n", @@ -1315,7 +1316,7 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, ret = -ENODEV; goto probe_failed; } - INIT_WORK(&ha->dpc_work, qla4xxx_do_dpc, ha); + INIT_WORK(&ha->dpc_work, qla4xxx_do_dpc); ret = request_irq(pdev->irq, qla4xxx_intr_handler, SA_INTERRUPT|SA_SHIRQ, "qla4xxx", ha); diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c index 38c215a78f6..3571ce8934e 100644 --- a/drivers/scsi/scsi_transport_fc.c +++ b/drivers/scsi/scsi_transport_fc.c @@ -241,9 +241,9 @@ fc_bitfield_name_search(remote_port_roles, fc_remote_port_role_names) #define FC_MGMTSRVR_PORTID 0x00000a -static void fc_timeout_deleted_rport(void *data); -static void fc_timeout_fail_rport_io(void *data); -static void fc_scsi_scan_rport(void *data); +static void fc_timeout_deleted_rport(struct work_struct *work); +static void fc_timeout_fail_rport_io(struct work_struct *work); +static void fc_scsi_scan_rport(struct work_struct *work); /* * Attribute counts pre object type... @@ -1613,7 +1613,7 @@ fc_flush_work(struct Scsi_Host *shost) * 1 on success / 0 already queued / < 0 for error **/ static int -fc_queue_devloss_work(struct Scsi_Host *shost, struct work_struct *work, +fc_queue_devloss_work(struct Scsi_Host *shost, struct delayed_work *work, unsigned long delay) { if (unlikely(!fc_host_devloss_work_q(shost))) { @@ -1625,9 +1625,6 @@ fc_queue_devloss_work(struct Scsi_Host *shost, struct work_struct *work, return -EINVAL; } - if (delay == 0) - return queue_work(fc_host_devloss_work_q(shost), work); - return queue_delayed_work(fc_host_devloss_work_q(shost), work, delay); } @@ -1712,12 +1709,13 @@ EXPORT_SYMBOL(fc_remove_host); * fc_starget_delete - called to delete the scsi decendents of an rport * (target and all sdevs) * - * @data: remote port to be operated on. + * @work: remote port to be operated on. **/ static void -fc_starget_delete(void *data) +fc_starget_delete(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, stgt_delete_work); struct Scsi_Host *shost = rport_to_shost(rport); unsigned long flags; struct fc_internal *i = to_fc_internal(shost->transportt); @@ -1751,12 +1749,13 @@ fc_starget_delete(void *data) /** * fc_rport_final_delete - finish rport termination and delete it. * - * @data: remote port to be deleted. + * @work: remote port to be deleted. **/ static void -fc_rport_final_delete(void *data) +fc_rport_final_delete(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, rport_delete_work); struct device *dev = &rport->dev; struct Scsi_Host *shost = rport_to_shost(rport); struct fc_internal *i = to_fc_internal(shost->transportt); @@ -1770,7 +1769,7 @@ fc_rport_final_delete(void *data) /* Delete SCSI target and sdevs */ if (rport->scsi_target_id != -1) - fc_starget_delete(data); + fc_starget_delete(&rport->stgt_delete_work); else if (i->f->dev_loss_tmo_callbk) i->f->dev_loss_tmo_callbk(rport); else if (i->f->terminate_rport_io) @@ -1829,11 +1828,11 @@ fc_rport_create(struct Scsi_Host *shost, int channel, rport->channel = channel; rport->fast_io_fail_tmo = -1; - INIT_WORK(&rport->dev_loss_work, fc_timeout_deleted_rport, rport); - INIT_WORK(&rport->fail_io_work, fc_timeout_fail_rport_io, rport); - INIT_WORK(&rport->scan_work, fc_scsi_scan_rport, rport); - INIT_WORK(&rport->stgt_delete_work, fc_starget_delete, rport); - INIT_WORK(&rport->rport_delete_work, fc_rport_final_delete, rport); + INIT_DELAYED_WORK(&rport->dev_loss_work, fc_timeout_deleted_rport); + INIT_DELAYED_WORK(&rport->fail_io_work, fc_timeout_fail_rport_io); + INIT_WORK(&rport->scan_work, fc_scsi_scan_rport); + INIT_WORK(&rport->stgt_delete_work, fc_starget_delete); + INIT_WORK(&rport->rport_delete_work, fc_rport_final_delete); spin_lock_irqsave(shost->host_lock, flags); @@ -1963,7 +1962,7 @@ fc_remote_port_add(struct Scsi_Host *shost, int channel, } if (match) { - struct work_struct *work = + struct delayed_work *work = &rport->dev_loss_work; memcpy(&rport->node_name, &ids->node_name, @@ -2267,12 +2266,13 @@ EXPORT_SYMBOL(fc_remote_port_rolechg); * was a SCSI target (thus was blocked), and failed * to return in the alloted time. * - * @data: rport target that failed to reappear in the alloted time. + * @work: rport target that failed to reappear in the alloted time. **/ static void -fc_timeout_deleted_rport(void *data) +fc_timeout_deleted_rport(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, dev_loss_work.work); struct Scsi_Host *shost = rport_to_shost(rport); struct fc_host_attrs *fc_host = shost_to_fc_host(shost); unsigned long flags; @@ -2366,15 +2366,16 @@ fc_timeout_deleted_rport(void *data) * fc_timeout_fail_rport_io - Timeout handler for a fast io failing on a * disconnected SCSI target. * - * @data: rport to terminate io on. + * @work: rport to terminate io on. * * Notes: Only requests the failure of the io, not that all are flushed * prior to returning. **/ static void -fc_timeout_fail_rport_io(void *data) +fc_timeout_fail_rport_io(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, fail_io_work.work); struct Scsi_Host *shost = rport_to_shost(rport); struct fc_internal *i = to_fc_internal(shost->transportt); @@ -2387,12 +2388,13 @@ fc_timeout_fail_rport_io(void *data) /** * fc_scsi_scan_rport - called to perform a scsi scan on a remote port. * - * @data: remote port to be scanned. + * @work: remote port to be scanned. **/ static void -fc_scsi_scan_rport(void *data) +fc_scsi_scan_rport(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, scan_work); struct Scsi_Host *shost = rport_to_shost(rport); unsigned long flags; diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 9b25124a989..9c22f134271 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -234,9 +234,11 @@ static int iscsi_user_scan(struct Scsi_Host *shost, uint channel, return 0; } -static void session_recovery_timedout(void *data) +static void session_recovery_timedout(struct work_struct *work) { - struct iscsi_cls_session *session = data; + struct iscsi_cls_session *session = + container_of(work, struct iscsi_cls_session, + recovery_work.work); dev_printk(KERN_INFO, &session->dev, "iscsi: session recovery timed " "out after %d secs\n", session->recovery_tmo); @@ -276,7 +278,7 @@ iscsi_alloc_session(struct Scsi_Host *shost, session->transport = transport; session->recovery_tmo = 120; - INIT_WORK(&session->recovery_work, session_recovery_timedout, session); + INIT_DELAYED_WORK(&session->recovery_work, session_recovery_timedout); INIT_LIST_HEAD(&session->host_list); INIT_LIST_HEAD(&session->sess_list); diff --git a/drivers/scsi/scsi_transport_spi.c b/drivers/scsi/scsi_transport_spi.c index 9f070f0d0f2..3fded483146 100644 --- a/drivers/scsi/scsi_transport_spi.c +++ b/drivers/scsi/scsi_transport_spi.c @@ -964,9 +964,10 @@ struct work_queue_wrapper { }; static void -spi_dv_device_work_wrapper(void *data) +spi_dv_device_work_wrapper(struct work_struct *work) { - struct work_queue_wrapper *wqw = (struct work_queue_wrapper *)data; + struct work_queue_wrapper *wqw = + container_of(work, struct work_queue_wrapper, work); struct scsi_device *sdev = wqw->sdev; kfree(wqw); @@ -1006,7 +1007,7 @@ spi_schedule_dv_device(struct scsi_device *sdev) return; } - INIT_WORK(&wqw->work, spi_dv_device_work_wrapper, wqw); + INIT_WORK(&wqw->work, spi_dv_device_work_wrapper); wqw->sdev = sdev; schedule_work(&wqw->work); diff --git a/drivers/spi/spi_bitbang.c b/drivers/spi/spi_bitbang.c index a23862ef72b..08c1c57c612 100644 --- a/drivers/spi/spi_bitbang.c +++ b/drivers/spi/spi_bitbang.c @@ -265,9 +265,10 @@ static int spi_bitbang_bufs(struct spi_device *spi, struct spi_transfer *t) * Drivers can provide word-at-a-time i/o primitives, or provide * transfer-at-a-time ones to leverage dma or fifo hardware. */ -static void bitbang_work(void *_bitbang) +static void bitbang_work(struct work_struct *work) { - struct spi_bitbang *bitbang = _bitbang; + struct spi_bitbang *bitbang = + container_of(work, struct spi_bitbang, work); unsigned long flags; spin_lock_irqsave(&bitbang->lock, flags); @@ -456,7 +457,7 @@ int spi_bitbang_start(struct spi_bitbang *bitbang) if (!bitbang->master || !bitbang->chipselect) return -EINVAL; - INIT_WORK(&bitbang->work, bitbang_work, bitbang); + INIT_WORK(&bitbang->work, bitbang_work); spin_lock_init(&bitbang->lock); INIT_LIST_HEAD(&bitbang->queue); diff --git a/drivers/usb/atm/cxacru.c b/drivers/usb/atm/cxacru.c index e6565633ba0..3dfa3e40e14 100644 --- a/drivers/usb/atm/cxacru.c +++ b/drivers/usb/atm/cxacru.c @@ -158,7 +158,7 @@ struct cxacru_data { const struct cxacru_modem_type *modem_type; int line_status; - struct work_struct poll_work; + struct delayed_work poll_work; /* contol handles */ struct mutex cm_serialize; @@ -347,7 +347,7 @@ static int cxacru_card_status(struct cxacru_data *instance) return 0; } -static void cxacru_poll_status(struct cxacru_data *instance); +static void cxacru_poll_status(struct work_struct *work); static int cxacru_atm_start(struct usbatm_data *usbatm_instance, struct atm_dev *atm_dev) @@ -376,12 +376,14 @@ static int cxacru_atm_start(struct usbatm_data *usbatm_instance, } /* Start status polling */ - cxacru_poll_status(instance); + cxacru_poll_status(&instance->poll_work.work); return 0; } -static void cxacru_poll_status(struct cxacru_data *instance) +static void cxacru_poll_status(struct work_struct *work) { + struct cxacru_data *instance = + container_of(work, struct cxacru_data, poll_work.work); u32 buf[CXINF_MAX] = {}; struct usbatm_data *usbatm = instance->usbatm; struct atm_dev *atm_dev = usbatm->atm_dev; @@ -720,7 +722,7 @@ static int cxacru_bind(struct usbatm_data *usbatm_instance, mutex_init(&instance->cm_serialize); - INIT_WORK(&instance->poll_work, (void *)cxacru_poll_status, instance); + INIT_DELAYED_WORK(&instance->poll_work, cxacru_poll_status); usbatm_instance->driver_data = instance; diff --git a/drivers/usb/atm/speedtch.c b/drivers/usb/atm/speedtch.c index c870c804470..7ed34bb1c50 100644 --- a/drivers/usb/atm/speedtch.c +++ b/drivers/usb/atm/speedtch.c @@ -142,7 +142,7 @@ struct speedtch_instance_data { struct speedtch_params params; /* set in probe, constant afterwards */ - struct work_struct status_checker; + struct delayed_work status_checker; unsigned char last_status; @@ -498,8 +498,11 @@ static int speedtch_start_synchro(struct speedtch_instance_data *instance) return ret; } -static void speedtch_check_status(struct speedtch_instance_data *instance) +static void speedtch_check_status(struct work_struct *work) { + struct speedtch_instance_data *instance = + container_of(work, struct speedtch_instance_data, + status_checker.work); struct usbatm_data *usbatm = instance->usbatm; struct atm_dev *atm_dev = usbatm->atm_dev; unsigned char *buf = instance->scratch_buffer; @@ -576,7 +579,7 @@ static void speedtch_status_poll(unsigned long data) { struct speedtch_instance_data *instance = (void *)data; - schedule_work(&instance->status_checker); + schedule_delayed_work(&instance->status_checker, 0); /* The following check is racy, but the race is harmless */ if (instance->poll_delay < MAX_POLL_DELAY) @@ -596,7 +599,7 @@ static void speedtch_resubmit_int(unsigned long data) if (int_urb) { ret = usb_submit_urb(int_urb, GFP_ATOMIC); if (!ret) - schedule_work(&instance->status_checker); + schedule_delayed_work(&instance->status_checker, 0); else { atm_dbg(instance->usbatm, "%s: usb_submit_urb failed with result %d\n", __func__, ret); mod_timer(&instance->resubmit_timer, jiffies + msecs_to_jiffies(RESUBMIT_DELAY)); @@ -640,7 +643,7 @@ static void speedtch_handle_int(struct urb *int_urb) if ((int_urb = instance->int_urb)) { ret = usb_submit_urb(int_urb, GFP_ATOMIC); - schedule_work(&instance->status_checker); + schedule_delayed_work(&instance->status_checker, 0); if (ret < 0) { atm_dbg(usbatm, "%s: usb_submit_urb failed with result %d\n", __func__, ret); goto fail; @@ -855,7 +858,7 @@ static int speedtch_bind(struct usbatm_data *usbatm, usbatm->flags |= (use_isoc ? UDSL_USE_ISOC : 0); - INIT_WORK(&instance->status_checker, (void *)speedtch_check_status, instance); + INIT_DELAYED_WORK(&instance->status_checker, speedtch_check_status); instance->status_checker.timer.function = speedtch_status_poll; instance->status_checker.timer.data = (unsigned long)instance; diff --git a/drivers/usb/atm/ueagle-atm.c b/drivers/usb/atm/ueagle-atm.c index f6b9f7e1f71..e39bb09f5af 100644 --- a/drivers/usb/atm/ueagle-atm.c +++ b/drivers/usb/atm/ueagle-atm.c @@ -658,9 +658,9 @@ static int request_dsp(struct uea_softc *sc) /* * The uea_load_page() function must be called within a process context */ -static void uea_load_page(void *xsc) +static void uea_load_page(struct work_struct *work) { - struct uea_softc *sc = xsc; + struct uea_softc *sc = container_of(work, struct uea_softc, task); u16 pageno = sc->pageno; u16 ovl = sc->ovl; struct block_info bi; @@ -1352,7 +1352,7 @@ static int uea_boot(struct uea_softc *sc) uea_enters(INS_TO_USBDEV(sc)); - INIT_WORK(&sc->task, uea_load_page, sc); + INIT_WORK(&sc->task, uea_load_page); init_waitqueue_head(&sc->sync_q); init_waitqueue_head(&sc->cmv_ack_wait); diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 9a9012fd284..6408e10fdbf 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -421,9 +421,9 @@ static void acm_write_bulk(struct urb *urb) schedule_work(&acm->work); } -static void acm_softint(void *private) +static void acm_softint(struct work_struct *work) { - struct acm *acm = private; + struct acm *acm = container_of(work, struct acm, work); dbg("Entering acm_softint."); if (!ACM_READY(acm)) @@ -927,7 +927,7 @@ skip_normal_probe: acm->rx_buflimit = num_rx_buf; acm->urb_task.func = acm_rx_tasklet; acm->urb_task.data = (unsigned long) acm; - INIT_WORK(&acm->work, acm_softint, acm); + INIT_WORK(&acm->work, acm_softint); spin_lock_init(&acm->throttle_lock); spin_lock_init(&acm->write_lock); spin_lock_init(&acm->read_lock); diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index ba165aff9ea..ad0ffbe8f7d 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -167,9 +167,10 @@ static void set_port_led( #define LED_CYCLE_PERIOD ((2*HZ)/3) -static void led_work (void *__hub) +static void led_work (struct work_struct *work) { - struct usb_hub *hub = __hub; + struct usb_hub *hub = + container_of(work, struct usb_hub, leds.work); struct usb_device *hdev = hub->hdev; unsigned i; unsigned changed = 0; @@ -351,9 +352,10 @@ hub_clear_tt_buffer (struct usb_device *hdev, u16 devinfo, u16 tt) * talking to TTs must queue control transfers (not just bulk and iso), so * both can talk to the same hub concurrently. */ -static void hub_tt_kevent (void *arg) +static void hub_tt_kevent (struct work_struct *work) { - struct usb_hub *hub = arg; + struct usb_hub *hub = + container_of(work, struct usb_hub, tt.kevent); unsigned long flags; spin_lock_irqsave (&hub->tt.lock, flags); @@ -641,7 +643,7 @@ static int hub_configure(struct usb_hub *hub, spin_lock_init (&hub->tt.lock); INIT_LIST_HEAD (&hub->tt.clear_list); - INIT_WORK (&hub->tt.kevent, hub_tt_kevent, hub); + INIT_WORK (&hub->tt.kevent, hub_tt_kevent); switch (hdev->descriptor.bDeviceProtocol) { case 0: break; @@ -880,7 +882,7 @@ descriptor_error: INIT_LIST_HEAD(&hub->event_list); hub->intfdev = &intf->dev; hub->hdev = hdev; - INIT_WORK(&hub->leds, led_work, hub); + INIT_DELAYED_WORK(&hub->leds, led_work); usb_set_intfdata (intf, hub); @@ -2281,7 +2283,7 @@ check_highspeed (struct usb_hub *hub, struct usb_device *udev, int port1) /* hub LEDs are probably harder to miss than syslog */ if (hub->has_indicators) { hub->indicator[port1-1] = INDICATOR_GREEN_BLINK; - schedule_work (&hub->leds); + schedule_delayed_work (&hub->leds, 0); } } kfree(qual); @@ -2455,7 +2457,7 @@ static void hub_port_connect_change(struct usb_hub *hub, int port1, if (hub->has_indicators) { hub->indicator[port1-1] = INDICATOR_AMBER_BLINK; - schedule_work (&hub->leds); + schedule_delayed_work (&hub->leds, 0); } status = -ENOTCONN; /* Don't retry */ goto loop_disable; diff --git a/drivers/usb/core/hub.h b/drivers/usb/core/hub.h index 0f8e82a4d48..035d3ef3588 100644 --- a/drivers/usb/core/hub.h +++ b/drivers/usb/core/hub.h @@ -230,7 +230,7 @@ struct usb_hub { unsigned has_indicators:1; enum hub_led_mode indicator[USB_MAXCHILDREN]; - struct work_struct leds; + struct delayed_work leds; }; #endif /* __LINUX_HUB_H */ diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index 7729c074488..89572bc021b 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -1501,9 +1501,10 @@ struct set_config_request { }; /* Worker routine for usb_driver_set_configuration() */ -static void driver_set_config_work(void *_req) +static void driver_set_config_work(struct work_struct *work) { - struct set_config_request *req = _req; + struct set_config_request *req = + container_of(work, struct set_config_request, work); usb_lock_device(req->udev); usb_set_configuration(req->udev, req->config); @@ -1541,7 +1542,7 @@ int usb_driver_set_configuration(struct usb_device *udev, int config) return -ENOMEM; req->udev = udev; req->config = config; - INIT_WORK(&req->work, driver_set_config_work, req); + INIT_WORK(&req->work, driver_set_config_work); usb_get_dev(udev); if (!schedule_work(&req->work)) { diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c index 467cb02832f..ab2f68fc7d2 100644 --- a/drivers/usb/core/usb.c +++ b/drivers/usb/core/usb.c @@ -210,9 +210,10 @@ static void ksuspend_usb_cleanup(void) #ifdef CONFIG_USB_SUSPEND /* usb_autosuspend_work - callback routine to autosuspend a USB device */ -static void usb_autosuspend_work(void *_udev) +static void usb_autosuspend_work(struct work_struct *work) { - struct usb_device *udev = _udev; + struct usb_device *udev = + container_of(work, struct usb_device, autosuspend.work); usb_pm_lock(udev); udev->auto_pm = 1; @@ -222,7 +223,7 @@ static void usb_autosuspend_work(void *_udev) #else -static void usb_autosuspend_work(void *_udev) +static void usb_autosuspend_work(struct work_struct *work) {} #endif @@ -304,7 +305,7 @@ usb_alloc_dev(struct usb_device *parent, struct usb_bus *bus, unsigned port1) #ifdef CONFIG_PM mutex_init(&dev->pm_mutex); - INIT_WORK(&dev->autosuspend, usb_autosuspend_work, dev); + INIT_DELAYED_WORK(&dev->autosuspend, usb_autosuspend_work); #endif return dev; } diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c index 1c17d26d03b..107119c5430 100644 --- a/drivers/usb/gadget/ether.c +++ b/drivers/usb/gadget/ether.c @@ -1833,9 +1833,9 @@ static void rx_fill (struct eth_dev *dev, gfp_t gfp_flags) spin_unlock_irqrestore(&dev->req_lock, flags); } -static void eth_work (void *_dev) +static void eth_work (struct work_struct *work) { - struct eth_dev *dev = _dev; + struct eth_dev *dev = container_of(work, struct eth_dev, work); if (test_and_clear_bit (WORK_RX_MEMORY, &dev->todo)) { if (netif_running (dev->net)) @@ -2398,7 +2398,7 @@ autoconf_fail: dev = netdev_priv(net); spin_lock_init (&dev->lock); spin_lock_init (&dev->req_lock); - INIT_WORK (&dev->work, eth_work, dev); + INIT_WORK (&dev->work, eth_work); INIT_LIST_HEAD (&dev->tx_reqs); INIT_LIST_HEAD (&dev->rx_reqs); diff --git a/drivers/usb/host/u132-hcd.c b/drivers/usb/host/u132-hcd.c index 32c635ecbf3..4f95a249c91 100644 --- a/drivers/usb/host/u132-hcd.c +++ b/drivers/usb/host/u132-hcd.c @@ -163,7 +163,7 @@ struct u132_endp { u16 queue_next; struct urb *urb_list[ENDP_QUEUE_SIZE]; struct list_head urb_more; - struct work_struct scheduler; + struct delayed_work scheduler; }; struct u132_ring { unsigned in_use:1; @@ -171,7 +171,7 @@ struct u132_ring { u8 number; struct u132 *u132; struct u132_endp *curr_endp; - struct work_struct scheduler; + struct delayed_work scheduler; }; #define OHCI_QUIRK_AMD756 0x01 #define OHCI_QUIRK_SUPERIO 0x02 @@ -198,7 +198,7 @@ struct u132 { u32 hc_roothub_portstatus[MAX_ROOT_PORTS]; int flags; unsigned long next_statechange; - struct work_struct monitor; + struct delayed_work monitor; int num_endpoints; struct u132_addr addr[MAX_U132_ADDRS]; struct u132_udev udev[MAX_U132_UDEVS]; @@ -314,7 +314,7 @@ static void u132_ring_requeue_work(struct u132 *u132, struct u132_ring *ring, if (delta > 0) { if (queue_delayed_work(workqueue, &ring->scheduler, delta)) return; - } else if (queue_work(workqueue, &ring->scheduler)) + } else if (queue_delayed_work(workqueue, &ring->scheduler, 0)) return; kref_put(&u132->kref, u132_hcd_delete); return; @@ -393,12 +393,8 @@ static inline void u132_endp_init_kref(struct u132 *u132, static void u132_endp_queue_work(struct u132 *u132, struct u132_endp *endp, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(workqueue, &endp->scheduler, delta)) - kref_get(&endp->kref); - } else if (queue_work(workqueue, &endp->scheduler)) - kref_get(&endp->kref); - return; + if (queue_delayed_work(workqueue, &endp->scheduler, delta)) + kref_get(&endp->kref); } static void u132_endp_cancel_work(struct u132 *u132, struct u132_endp *endp) @@ -414,24 +410,14 @@ static inline void u132_monitor_put_kref(struct u132 *u132) static void u132_monitor_queue_work(struct u132 *u132, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(workqueue, &u132->monitor, delta)) { - kref_get(&u132->kref); - } - } else if (queue_work(workqueue, &u132->monitor)) - kref_get(&u132->kref); - return; + if (queue_delayed_work(workqueue, &u132->monitor, delta)) + kref_get(&u132->kref); } static void u132_monitor_requeue_work(struct u132 *u132, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(workqueue, &u132->monitor, delta)) - return; - } else if (queue_work(workqueue, &u132->monitor)) - return; - kref_put(&u132->kref, u132_hcd_delete); - return; + if (!queue_delayed_work(workqueue, &u132->monitor, delta)) + kref_put(&u132->kref, u132_hcd_delete); } static void u132_monitor_cancel_work(struct u132 *u132) @@ -493,9 +479,9 @@ static int read_roothub_info(struct u132 *u132) return 0; } -static void u132_hcd_monitor_work(void *data) +static void u132_hcd_monitor_work(struct work_struct *work) { - struct u132 *u132 = data; + struct u132 *u132 = container_of(work, struct u132, monitor.work); if (u132->going > 1) { dev_err(&u132->platform_dev->dev, "device has been removed %d\n" , u132->going); @@ -1319,15 +1305,14 @@ static void u132_hcd_initial_setup_sent(void *data, struct urb *urb, u8 *buf, } } -static void u132_hcd_ring_work_scheduler(void *data); -static void u132_hcd_endp_work_scheduler(void *data); /* * this work function is only executed from the work queue * */ -static void u132_hcd_ring_work_scheduler(void *data) +static void u132_hcd_ring_work_scheduler(struct work_struct *work) { - struct u132_ring *ring = data; + struct u132_ring *ring = + container_of(work, struct u132_ring, scheduler.work); struct u132 *u132 = ring->u132; down(&u132->scheduler_lock); if (ring->in_use) { @@ -1386,10 +1371,11 @@ static void u132_hcd_ring_work_scheduler(void *data) } } -static void u132_hcd_endp_work_scheduler(void *data) +static void u132_hcd_endp_work_scheduler(struct work_struct *work) { struct u132_ring *ring; - struct u132_endp *endp = data; + struct u132_endp *endp = + container_of(work, struct u132_endp, scheduler.work); struct u132 *u132 = endp->u132; down(&u132->scheduler_lock); ring = endp->ring; @@ -1947,7 +1933,7 @@ static int create_endpoint_and_queue_int(struct u132 *u132, if (!endp) { return -ENOMEM; } - INIT_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler, (void *)endp); + INIT_DELAYED_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler); spin_lock_init(&endp->queue_lock.slock); INIT_LIST_HEAD(&endp->urb_more); ring = endp->ring = &u132->ring[0]; @@ -2036,7 +2022,7 @@ static int create_endpoint_and_queue_bulk(struct u132 *u132, if (!endp) { return -ENOMEM; } - INIT_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler, (void *)endp); + INIT_DELAYED_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler); spin_lock_init(&endp->queue_lock.slock); INIT_LIST_HEAD(&endp->urb_more); endp->dequeueing = 0; @@ -2121,7 +2107,7 @@ static int create_endpoint_and_queue_control(struct u132 *u132, if (!endp) { return -ENOMEM; } - INIT_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler, (void *)endp); + INIT_DELAYED_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler); spin_lock_init(&endp->queue_lock.slock); INIT_LIST_HEAD(&endp->urb_more); ring = endp->ring = &u132->ring[0]; @@ -3100,10 +3086,10 @@ static void u132_initialise(struct u132 *u132, struct platform_device *pdev) ring->number = rings + 1; ring->length = 0; ring->curr_endp = NULL; - INIT_WORK(&ring->scheduler, u132_hcd_ring_work_scheduler, - (void *)ring); + INIT_DELAYED_WORK(&ring->scheduler, + u132_hcd_ring_work_scheduler); } down(&u132->sw_lock); - INIT_WORK(&u132->monitor, u132_hcd_monitor_work, (void *)u132); + INIT_DELAYED_WORK(&u132->monitor, u132_hcd_monitor_work); while (ports-- > 0) { struct u132_port *port = &u132->port[ports]; port->u132 = u132; diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 6d08a3bcc95..ebc9e823a46 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -969,9 +969,10 @@ static void hid_retry_timeout(unsigned long _hid) } /* Workqueue routine to reset the device */ -static void hid_reset(void *_hid) +static void hid_reset(struct work_struct *work) { - struct hid_device *hid = (struct hid_device *) _hid; + struct hid_device *hid = + container_of(work, struct hid_device, reset_work); int rc_lock, rc; dev_dbg(&hid->intf->dev, "resetting device\n"); @@ -2015,7 +2016,7 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) init_waitqueue_head(&hid->wait); - INIT_WORK(&hid->reset_work, hid_reset, hid); + INIT_WORK(&hid->reset_work, hid_reset); setup_timer(&hid->io_retry, hid_retry_timeout, (unsigned long) hid); spin_lock_init(&hid->inlock); diff --git a/drivers/usb/misc/ftdi-elan.c b/drivers/usb/misc/ftdi-elan.c index 9b591b8b936..e4e2cf2ba91 100644 --- a/drivers/usb/misc/ftdi-elan.c +++ b/drivers/usb/misc/ftdi-elan.c @@ -156,9 +156,9 @@ struct usb_ftdi { struct usb_device *udev; struct usb_interface *interface; struct usb_class_driver *class; - struct work_struct status_work; - struct work_struct command_work; - struct work_struct respond_work; + struct delayed_work status_work; + struct delayed_work command_work; + struct delayed_work respond_work; struct u132_platform_data platform_data; struct resource resources[0]; struct platform_device platform_dev; @@ -210,23 +210,14 @@ static void ftdi_elan_init_kref(struct usb_ftdi *ftdi) static void ftdi_status_requeue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(status_queue, &ftdi->status_work, delta)) - return; - } else if (queue_work(status_queue, &ftdi->status_work)) - return; - kref_put(&ftdi->kref, ftdi_elan_delete); - return; + if (!queue_delayed_work(status_queue, &ftdi->status_work, delta)) + kref_put(&ftdi->kref, ftdi_elan_delete); } static void ftdi_status_queue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(status_queue, &ftdi->status_work, delta)) - kref_get(&ftdi->kref); - } else if (queue_work(status_queue, &ftdi->status_work)) - kref_get(&ftdi->kref); - return; + if (queue_delayed_work(status_queue, &ftdi->status_work, delta)) + kref_get(&ftdi->kref); } static void ftdi_status_cancel_work(struct usb_ftdi *ftdi) @@ -237,25 +228,14 @@ static void ftdi_status_cancel_work(struct usb_ftdi *ftdi) static void ftdi_command_requeue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(command_queue, &ftdi->command_work, - delta)) - return; - } else if (queue_work(command_queue, &ftdi->command_work)) - return; - kref_put(&ftdi->kref, ftdi_elan_delete); - return; + if (!queue_delayed_work(command_queue, &ftdi->command_work, delta)) + kref_put(&ftdi->kref, ftdi_elan_delete); } static void ftdi_command_queue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(command_queue, &ftdi->command_work, - delta)) - kref_get(&ftdi->kref); - } else if (queue_work(command_queue, &ftdi->command_work)) - kref_get(&ftdi->kref); - return; + if (queue_delayed_work(command_queue, &ftdi->command_work, delta)) + kref_get(&ftdi->kref); } static void ftdi_command_cancel_work(struct usb_ftdi *ftdi) @@ -267,25 +247,14 @@ static void ftdi_command_cancel_work(struct usb_ftdi *ftdi) static void ftdi_response_requeue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(respond_queue, &ftdi->respond_work, - delta)) - return; - } else if (queue_work(respond_queue, &ftdi->respond_work)) - return; - kref_put(&ftdi->kref, ftdi_elan_delete); - return; + if (!queue_delayed_work(respond_queue, &ftdi->respond_work, delta)) + kref_put(&ftdi->kref, ftdi_elan_delete); } static void ftdi_respond_queue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(respond_queue, &ftdi->respond_work, - delta)) - kref_get(&ftdi->kref); - } else if (queue_work(respond_queue, &ftdi->respond_work)) - kref_get(&ftdi->kref); - return; + if (queue_delayed_work(respond_queue, &ftdi->respond_work, delta)) + kref_get(&ftdi->kref); } static void ftdi_response_cancel_work(struct usb_ftdi *ftdi) @@ -475,9 +444,11 @@ static void ftdi_elan_kick_command_queue(struct usb_ftdi *ftdi) return; } -static void ftdi_elan_command_work(void *data) +static void ftdi_elan_command_work(struct work_struct *work) { - struct usb_ftdi *ftdi = data; + struct usb_ftdi *ftdi = + container_of(work, struct usb_ftdi, command_work.work); + if (ftdi->disconnected > 0) { ftdi_elan_put_kref(ftdi); return; @@ -500,9 +471,10 @@ static void ftdi_elan_kick_respond_queue(struct usb_ftdi *ftdi) return; } -static void ftdi_elan_respond_work(void *data) +static void ftdi_elan_respond_work(struct work_struct *work) { - struct usb_ftdi *ftdi = data; + struct usb_ftdi *ftdi = + container_of(work, struct usb_ftdi, respond_work.work); if (ftdi->disconnected > 0) { ftdi_elan_put_kref(ftdi); return; @@ -534,9 +506,10 @@ static void ftdi_elan_respond_work(void *data) * after the FTDI has been synchronized * */ -static void ftdi_elan_status_work(void *data) +static void ftdi_elan_status_work(struct work_struct *work) { - struct usb_ftdi *ftdi = data; + struct usb_ftdi *ftdi = + container_of(work, struct usb_ftdi, status_work.work); int work_delay_in_msec = 0; if (ftdi->disconnected > 0) { ftdi_elan_put_kref(ftdi); @@ -2691,12 +2664,9 @@ static int ftdi_elan_probe(struct usb_interface *interface, ftdi->class = NULL; dev_info(&ftdi->udev->dev, "USB FDTI=%p ELAN interface %d now a" "ctivated\n", ftdi, iface_desc->desc.bInterfaceNumber); - INIT_WORK(&ftdi->status_work, ftdi_elan_status_work, - (void *)ftdi); - INIT_WORK(&ftdi->command_work, ftdi_elan_command_work, - (void *)ftdi); - INIT_WORK(&ftdi->respond_work, ftdi_elan_respond_work, - (void *)ftdi); + INIT_DELAYED_WORK(&ftdi->status_work, ftdi_elan_status_work); + INIT_DELAYED_WORK(&ftdi->command_work, ftdi_elan_command_work); + INIT_DELAYED_WORK(&ftdi->respond_work, ftdi_elan_respond_work); ftdi_status_queue_work(ftdi, msecs_to_jiffies(3 *1000)); return 0; } else { diff --git a/drivers/usb/misc/phidgetkit.c b/drivers/usb/misc/phidgetkit.c index abb4dcd811a..33e716c6a79 100644 --- a/drivers/usb/misc/phidgetkit.c +++ b/drivers/usb/misc/phidgetkit.c @@ -81,8 +81,8 @@ struct interfacekit { unsigned char *data; dma_addr_t data_dma; - struct work_struct do_notify; - struct work_struct do_resubmit; + struct delayed_work do_notify; + struct delayed_work do_resubmit; unsigned long input_events; unsigned long sensor_events; }; @@ -374,7 +374,7 @@ static void interfacekit_irq(struct urb *urb) } if (kit->input_events || kit->sensor_events) - schedule_work(&kit->do_notify); + schedule_delayed_work(&kit->do_notify, 0); resubmit: status = usb_submit_urb(urb, SLAB_ATOMIC); @@ -384,9 +384,10 @@ resubmit: kit->udev->devpath, status); } -static void do_notify(void *data) +static void do_notify(struct work_struct *work) { - struct interfacekit *kit = data; + struct interfacekit *kit = + container_of(work, struct interfacekit, do_notify.work); int i; char sysfs_file[8]; @@ -405,9 +406,11 @@ static void do_notify(void *data) } } -static void do_resubmit(void *data) +static void do_resubmit(struct work_struct *work) { - set_outputs(data); + struct interfacekit *kit = + container_of(work, struct interfacekit, do_resubmit.work); + set_outputs(kit); } #define show_set_output(value) \ @@ -575,8 +578,8 @@ static int interfacekit_probe(struct usb_interface *intf, const struct usb_devic kit->udev = usb_get_dev(dev); kit->intf = intf; - INIT_WORK(&kit->do_notify, do_notify, kit); - INIT_WORK(&kit->do_resubmit, do_resubmit, kit); + INIT_DELAYED_WORK(&kit->do_notify, do_notify); + INIT_DELAYED_WORK(&kit->do_resubmit, do_resubmit); usb_fill_int_urb(kit->irq, kit->udev, pipe, kit->data, maxp > URB_INT_SIZE ? URB_INT_SIZE : maxp, interfacekit_irq, kit, endpoint->bInterval); diff --git a/drivers/usb/misc/phidgetmotorcontrol.c b/drivers/usb/misc/phidgetmotorcontrol.c index 5c780cab92e..0385ffcc741 100644 --- a/drivers/usb/misc/phidgetmotorcontrol.c +++ b/drivers/usb/misc/phidgetmotorcontrol.c @@ -41,7 +41,7 @@ struct motorcontrol { unsigned char *data; dma_addr_t data_dma; - struct work_struct do_notify; + struct delayed_work do_notify; unsigned long input_events; unsigned long speed_events; unsigned long exceed_events; @@ -148,7 +148,7 @@ static void motorcontrol_irq(struct urb *urb) set_bit(1, &mc->exceed_events); if (mc->input_events || mc->exceed_events || mc->speed_events) - schedule_work(&mc->do_notify); + schedule_delayed_work(&mc->do_notify, 0); resubmit: status = usb_submit_urb(urb, SLAB_ATOMIC); @@ -159,9 +159,10 @@ resubmit: mc->udev->devpath, status); } -static void do_notify(void *data) +static void do_notify(struct work_struct *work) { - struct motorcontrol *mc = data; + struct motorcontrol *mc = + container_of(work, struct motorcontrol, do_notify.work); int i; char sysfs_file[8]; @@ -348,7 +349,7 @@ static int motorcontrol_probe(struct usb_interface *intf, const struct usb_devic mc->udev = usb_get_dev(dev); mc->intf = intf; mc->acceleration[0] = mc->acceleration[1] = 10; - INIT_WORK(&mc->do_notify, do_notify, mc); + INIT_DELAYED_WORK(&mc->do_notify, do_notify); usb_fill_int_urb(mc->irq, mc->udev, pipe, mc->data, maxp > URB_INT_SIZE ? URB_INT_SIZE : maxp, motorcontrol_irq, mc, endpoint->bInterval); diff --git a/drivers/usb/net/kaweth.c b/drivers/usb/net/kaweth.c index 7c906a43e49..fa78326d0bf 100644 --- a/drivers/usb/net/kaweth.c +++ b/drivers/usb/net/kaweth.c @@ -222,7 +222,7 @@ struct kaweth_device int suspend_lowmem_ctrl; int linkstate; int opened; - struct work_struct lowmem_work; + struct delayed_work lowmem_work; struct usb_device *dev; struct net_device *net; @@ -530,9 +530,10 @@ resubmit: kaweth_resubmit_int_urb(kaweth, GFP_ATOMIC); } -static void kaweth_resubmit_tl(void *d) +static void kaweth_resubmit_tl(struct work_struct *work) { - struct kaweth_device *kaweth = (struct kaweth_device *)d; + struct kaweth_device *kaweth = + container_of(work, struct kaweth_device, lowmem_work.work); if (IS_BLOCKED(kaweth->status)) return; @@ -1126,7 +1127,7 @@ err_fw: /* kaweth is zeroed as part of alloc_netdev */ - INIT_WORK(&kaweth->lowmem_work, kaweth_resubmit_tl, (void *)kaweth); + INIT_DELAYED_WORK(&kaweth->lowmem_work, kaweth_resubmit_tl); SET_MODULE_OWNER(netdev); diff --git a/drivers/usb/net/pegasus.c b/drivers/usb/net/pegasus.c index 33abbd2176b..78cf6f09128 100644 --- a/drivers/usb/net/pegasus.c +++ b/drivers/usb/net/pegasus.c @@ -1280,9 +1280,9 @@ static inline void setup_pegasus_II(pegasus_t * pegasus) static struct workqueue_struct *pegasus_workqueue = NULL; #define CARRIER_CHECK_DELAY (2 * HZ) -static void check_carrier(void *data) +static void check_carrier(struct work_struct *work) { - pegasus_t *pegasus = data; + pegasus_t *pegasus = container_of(work, pegasus_t, carrier_check.work); set_carrier(pegasus->net); if (!(pegasus->flags & PEGASUS_UNPLUG)) { queue_delayed_work(pegasus_workqueue, &pegasus->carrier_check, @@ -1318,7 +1318,7 @@ static int pegasus_probe(struct usb_interface *intf, tasklet_init(&pegasus->rx_tl, rx_fixup, (unsigned long) pegasus); - INIT_WORK(&pegasus->carrier_check, check_carrier, pegasus); + INIT_DELAYED_WORK(&pegasus->carrier_check, check_carrier); pegasus->intf = intf; pegasus->usb = dev; diff --git a/drivers/usb/net/pegasus.h b/drivers/usb/net/pegasus.h index 006438069b6..98f6898cae1 100644 --- a/drivers/usb/net/pegasus.h +++ b/drivers/usb/net/pegasus.h @@ -95,7 +95,7 @@ typedef struct pegasus { int dev_index; int intr_interval; struct tasklet_struct rx_tl; - struct work_struct carrier_check; + struct delayed_work carrier_check; struct urb *ctrl_urb, *rx_urb, *tx_urb, *intr_urb; struct sk_buff *rx_pool[RX_SKBS]; struct sk_buff *rx_skb; diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index 760b5327b81..79b5474fe23 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -782,9 +782,10 @@ static struct ethtool_ops usbnet_ethtool_ops = { * especially now that control transfers can be queued. */ static void -kevent (void *data) +kevent (struct work_struct *work) { - struct usbnet *dev = data; + struct usbnet *dev = + container_of(work, struct usbnet, kevent); int status; /* usb_clear_halt() needs a thread context */ @@ -1146,7 +1147,7 @@ usbnet_probe (struct usb_interface *udev, const struct usb_device_id *prod) skb_queue_head_init (&dev->done); dev->bh.func = usbnet_bh; dev->bh.data = (unsigned long) dev; - INIT_WORK (&dev->kevent, kevent, dev); + INIT_WORK (&dev->kevent, kevent); dev->delay.function = usbnet_bh; dev->delay.data = (unsigned long) dev; init_timer (&dev->delay); diff --git a/drivers/usb/serial/aircable.c b/drivers/usb/serial/aircable.c index 81227550913..2a4ac9bd6a3 100644 --- a/drivers/usb/serial/aircable.c +++ b/drivers/usb/serial/aircable.c @@ -92,6 +92,7 @@ struct aircable_private { struct circ_buf *rx_buf; /* read buffer */ int rx_flags; /* for throttilng */ struct work_struct rx_work; /* work cue for the receiving line */ + struct usb_serial_port *port; /* USB port with which associated */ }; /* Private methods */ @@ -251,10 +252,11 @@ static void aircable_send(struct usb_serial_port *port) schedule_work(&port->work); } -static void aircable_read(void *params) +static void aircable_read(struct work_struct *work) { - struct usb_serial_port *port = params; - struct aircable_private *priv = usb_get_serial_port_data(port); + struct aircable_private *priv = + container_of(work, struct aircable_private, rx_work); + struct usb_serial_port *port = priv->port; struct tty_struct *tty; unsigned char *data; int count; @@ -348,7 +350,8 @@ static int aircable_attach (struct usb_serial *serial) } priv->rx_flags &= ~(THROTTLED | ACTUALLY_THROTTLED); - INIT_WORK(&priv->rx_work, aircable_read, port); + priv->port = port; + INIT_WORK(&priv->rx_work, aircable_read); usb_set_serial_port_data(serial->port[0], priv); @@ -515,7 +518,7 @@ static void aircable_read_bulk_callback(struct urb *urb) package_length - shift); } } - aircable_read(port); + aircable_read(&priv->rx_work); } /* Schedule the next read _if_ we are still open */ diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c index bdb58100fc1..fd159b040bf 100644 --- a/drivers/usb/serial/digi_acceleport.c +++ b/drivers/usb/serial/digi_acceleport.c @@ -430,13 +430,14 @@ struct digi_port { int dp_in_close; /* close in progress */ wait_queue_head_t dp_close_wait; /* wait queue for close */ struct work_struct dp_wakeup_work; + struct usb_serial_port *dp_port; }; /* Local Function Declarations */ static void digi_wakeup_write( struct usb_serial_port *port ); -static void digi_wakeup_write_lock(void *); +static void digi_wakeup_write_lock(struct work_struct *work); static int digi_write_oob_command( struct usb_serial_port *port, unsigned char *buf, int count, int interruptible ); static int digi_write_inb_command( struct usb_serial_port *port, @@ -598,11 +599,12 @@ static inline long cond_wait_interruptible_timeout_irqrestore( * on writes. */ -static void digi_wakeup_write_lock(void *arg) +static void digi_wakeup_write_lock(struct work_struct *work) { - struct usb_serial_port *port = arg; + struct digi_port *priv = + container_of(work, struct digi_port, dp_wakeup_work); + struct usb_serial_port *port = priv->dp_port; unsigned long flags; - struct digi_port *priv = usb_get_serial_port_data(port); spin_lock_irqsave( &priv->dp_port_lock, flags ); @@ -1702,8 +1704,8 @@ dbg( "digi_startup: TOP" ); init_waitqueue_head( &priv->dp_flush_wait ); priv->dp_in_close = 0; init_waitqueue_head( &priv->dp_close_wait ); - INIT_WORK(&priv->dp_wakeup_work, - digi_wakeup_write_lock, serial->port[i]); + INIT_WORK(&priv->dp_wakeup_work, digi_wakeup_write_lock); + priv->dp_port = serial->port[i]; /* initialize write wait queue for this port */ init_waitqueue_head( &serial->port[i]->write_wait ); diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index c186b4e73c7..88ed5c1d236 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -559,7 +559,8 @@ struct ftdi_private { char prev_status, diff_status; /* Used for TIOCMIWAIT */ __u8 rx_flags; /* receive state flags (throttling) */ spinlock_t rx_lock; /* spinlock for receive state */ - struct work_struct rx_work; + struct delayed_work rx_work; + struct usb_serial_port *port; int rx_processed; unsigned long rx_bytes; @@ -593,7 +594,7 @@ static int ftdi_write_room (struct usb_serial_port *port); static int ftdi_chars_in_buffer (struct usb_serial_port *port); static void ftdi_write_bulk_callback (struct urb *urb); static void ftdi_read_bulk_callback (struct urb *urb); -static void ftdi_process_read (void *param); +static void ftdi_process_read (struct work_struct *work); static void ftdi_set_termios (struct usb_serial_port *port, struct termios * old); static int ftdi_tiocmget (struct usb_serial_port *port, struct file *file); static int ftdi_tiocmset (struct usb_serial_port *port, struct file * file, unsigned int set, unsigned int clear); @@ -1201,7 +1202,8 @@ static int ftdi_sio_attach (struct usb_serial *serial) port->read_urb->transfer_buffer_length = BUFSZ; } - INIT_WORK(&priv->rx_work, ftdi_process_read, port); + INIT_DELAYED_WORK(&priv->rx_work, ftdi_process_read); + priv->port = port; /* Free port's existing write urb and transfer buffer. */ if (port->write_urb) { @@ -1641,17 +1643,18 @@ static void ftdi_read_bulk_callback (struct urb *urb) priv->rx_bytes += countread; spin_unlock_irqrestore(&priv->rx_lock, flags); - ftdi_process_read(port); + ftdi_process_read(&priv->rx_work.work); } /* ftdi_read_bulk_callback */ -static void ftdi_process_read (void *param) +static void ftdi_process_read (struct work_struct *work) { /* ftdi_process_read */ - struct usb_serial_port *port = (struct usb_serial_port*)param; + struct ftdi_private *priv = + container_of(work, struct ftdi_private, rx_work.work); + struct usb_serial_port *port = priv->port; struct urb *urb; struct tty_struct *tty; - struct ftdi_private *priv; char error_flag; unsigned char *data; @@ -2180,7 +2183,7 @@ static void ftdi_unthrottle (struct usb_serial_port *port) spin_unlock_irqrestore(&priv->rx_lock, flags); if (actually_throttled) - schedule_work(&priv->rx_work); + schedule_delayed_work(&priv->rx_work, 0); } static int __init ftdi_init (void) diff --git a/drivers/usb/serial/keyspan_pda.c b/drivers/usb/serial/keyspan_pda.c index 909005107ea..e09a0bfe623 100644 --- a/drivers/usb/serial/keyspan_pda.c +++ b/drivers/usb/serial/keyspan_pda.c @@ -120,6 +120,8 @@ struct keyspan_pda_private { int tx_throttled; struct work_struct wakeup_work; struct work_struct unthrottle_work; + struct usb_serial *serial; + struct usb_serial_port *port; }; @@ -175,9 +177,11 @@ static struct usb_device_id id_table_fake_xircom [] = { }; #endif -static void keyspan_pda_wakeup_write( struct usb_serial_port *port ) +static void keyspan_pda_wakeup_write(struct work_struct *work) { - + struct keyspan_pda_private *priv = + container_of(work, struct keyspan_pda_private, wakeup_work); + struct usb_serial_port *port = priv->port; struct tty_struct *tty = port->tty; /* wake up port processes */ @@ -187,8 +191,11 @@ static void keyspan_pda_wakeup_write( struct usb_serial_port *port ) tty_wakeup(tty); } -static void keyspan_pda_request_unthrottle( struct usb_serial *serial ) +static void keyspan_pda_request_unthrottle(struct work_struct *work) { + struct keyspan_pda_private *priv = + container_of(work, struct keyspan_pda_private, unthrottle_work); + struct usb_serial *serial = priv->serial; int result; dbg(" request_unthrottle"); @@ -765,11 +772,10 @@ static int keyspan_pda_startup (struct usb_serial *serial) return (1); /* error */ usb_set_serial_port_data(serial->port[0], priv); init_waitqueue_head(&serial->port[0]->write_wait); - INIT_WORK(&priv->wakeup_work, (void *)keyspan_pda_wakeup_write, - (void *)(serial->port[0])); - INIT_WORK(&priv->unthrottle_work, - (void *)keyspan_pda_request_unthrottle, - (void *)(serial)); + INIT_WORK(&priv->wakeup_work, keyspan_pda_wakeup_write); + INIT_WORK(&priv->unthrottle_work, keyspan_pda_request_unthrottle); + priv->serial = serial; + priv->port = serial->port[0]; return (0); } diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c index 8006e51c34b..2cfba8488a9 100644 --- a/drivers/usb/serial/usb-serial.c +++ b/drivers/usb/serial/usb-serial.c @@ -533,9 +533,10 @@ void usb_serial_port_softint(struct usb_serial_port *port) schedule_work(&port->work); } -static void usb_serial_port_work(void *private) +static void usb_serial_port_work(struct work_struct *work) { - struct usb_serial_port *port = private; + struct usb_serial_port *port = + container_of(work, struct usb_serial_port, work); struct tty_struct *tty; dbg("%s - port %d", __FUNCTION__, port->number); @@ -799,7 +800,7 @@ int usb_serial_probe(struct usb_interface *interface, port->serial = serial; spin_lock_init(&port->lock); mutex_init(&port->mutex); - INIT_WORK(&port->work, usb_serial_port_work, port); + INIT_WORK(&port->work, usb_serial_port_work); serial->port[i] = port; } diff --git a/drivers/usb/serial/whiteheat.c b/drivers/usb/serial/whiteheat.c index 4d1cd7aeccd..154c7d29059 100644 --- a/drivers/usb/serial/whiteheat.c +++ b/drivers/usb/serial/whiteheat.c @@ -227,6 +227,7 @@ struct whiteheat_private { struct list_head rx_urbs_submitted; struct list_head rx_urb_q; struct work_struct rx_work; + struct usb_serial_port *port; struct list_head tx_urbs_free; struct list_head tx_urbs_submitted; }; @@ -241,7 +242,7 @@ static void command_port_read_callback(struct urb *urb); static int start_port_read(struct usb_serial_port *port); static struct whiteheat_urb_wrap *urb_to_wrap(struct urb *urb, struct list_head *head); static struct list_head *list_first(struct list_head *head); -static void rx_data_softint(void *private); +static void rx_data_softint(struct work_struct *work); static int firm_send_command(struct usb_serial_port *port, __u8 command, __u8 *data, __u8 datasize); static int firm_open(struct usb_serial_port *port); @@ -424,7 +425,8 @@ static int whiteheat_attach (struct usb_serial *serial) spin_lock_init(&info->lock); info->flags = 0; info->mcr = 0; - INIT_WORK(&info->rx_work, rx_data_softint, port); + INIT_WORK(&info->rx_work, rx_data_softint); + info->port = port; INIT_LIST_HEAD(&info->rx_urbs_free); INIT_LIST_HEAD(&info->rx_urbs_submitted); @@ -949,7 +951,7 @@ static void whiteheat_unthrottle (struct usb_serial_port *port) spin_unlock_irqrestore(&info->lock, flags); if (actually_throttled) - rx_data_softint(port); + rx_data_softint(&info->rx_work); return; } @@ -1400,10 +1402,11 @@ static struct list_head *list_first(struct list_head *head) } -static void rx_data_softint(void *private) +static void rx_data_softint(struct work_struct *work) { - struct usb_serial_port *port = (struct usb_serial_port *)private; - struct whiteheat_private *info = usb_get_serial_port_data(port); + struct whiteheat_private *info = + container_of(work, struct whiteheat_private, rx_work); + struct usb_serial_port *port = info->port; struct tty_struct *tty = port->tty; struct whiteheat_urb_wrap *wrap; struct urb *urb; diff --git a/drivers/video/console/fbcon.c b/drivers/video/console/fbcon.c index 302174b8e47..31f476a6479 100644 --- a/drivers/video/console/fbcon.c +++ b/drivers/video/console/fbcon.c @@ -383,9 +383,9 @@ static void fbcon_update_softback(struct vc_data *vc) softback_top = 0; } -static void fb_flashcursor(void *private) +static void fb_flashcursor(struct work_struct *work) { - struct fb_info *info = private; + struct fb_info *info = container_of(work, struct fb_info, queue); struct fbcon_ops *ops = info->fbcon_par; struct display *p; struct vc_data *vc = NULL; @@ -442,7 +442,7 @@ static void fbcon_add_cursor_timer(struct fb_info *info) if ((!info->queue.func || info->queue.func == fb_flashcursor) && !(ops->flags & FBCON_FLAGS_CURSOR_TIMER)) { if (!info->queue.func) - INIT_WORK(&info->queue, fb_flashcursor, info); + INIT_WORK(&info->queue, fb_flashcursor); init_timer(&ops->cursor_timer); ops->cursor_timer.function = cursor_timer_handler; diff --git a/fs/9p/mux.c b/fs/9p/mux.c index 90a79c78454..944273c3dbf 100644 --- a/fs/9p/mux.c +++ b/fs/9p/mux.c @@ -110,8 +110,8 @@ struct v9fs_mux_rpc { }; static int v9fs_poll_proc(void *); -static void v9fs_read_work(void *); -static void v9fs_write_work(void *); +static void v9fs_read_work(struct work_struct *work); +static void v9fs_write_work(struct work_struct *work); static void v9fs_pollwait(struct file *filp, wait_queue_head_t * wait_address, poll_table * p); static u16 v9fs_mux_get_tag(struct v9fs_mux_data *); @@ -297,8 +297,8 @@ struct v9fs_mux_data *v9fs_mux_init(struct v9fs_transport *trans, int msize, m->rbuf = NULL; m->wpos = m->wsize = 0; m->wbuf = NULL; - INIT_WORK(&m->rq, v9fs_read_work, m); - INIT_WORK(&m->wq, v9fs_write_work, m); + INIT_WORK(&m->rq, v9fs_read_work); + INIT_WORK(&m->wq, v9fs_write_work); m->wsched = 0; memset(&m->poll_waddr, 0, sizeof(m->poll_waddr)); m->poll_task = NULL; @@ -458,13 +458,13 @@ static int v9fs_poll_proc(void *a) /** * v9fs_write_work - called when a transport can send some data */ -static void v9fs_write_work(void *a) +static void v9fs_write_work(struct work_struct *work) { int n, err; struct v9fs_mux_data *m; struct v9fs_req *req; - m = a; + m = container_of(work, struct v9fs_mux_data, wq); if (m->err < 0) { clear_bit(Wworksched, &m->wsched); @@ -564,7 +564,7 @@ static void process_request(struct v9fs_mux_data *m, struct v9fs_req *req) /** * v9fs_read_work - called when there is some data to be read from a transport */ -static void v9fs_read_work(void *a) +static void v9fs_read_work(struct work_struct *work) { int n, err; struct v9fs_mux_data *m; @@ -572,7 +572,7 @@ static void v9fs_read_work(void *a) struct v9fs_fcall *rcall; char *rbuf; - m = a; + m = container_of(work, struct v9fs_mux_data, rq); if (m->err < 0) return; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 78fe0fae23f..55f5333dae9 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -35,7 +35,7 @@ struct greedy { struct gfs2_holder gr_gh; - struct work_struct gr_work; + struct delayed_work gr_work; }; struct gfs2_gl_hash_bucket { @@ -1368,9 +1368,9 @@ static void gfs2_glock_prefetch(struct gfs2_glock *gl, unsigned int state, glops->go_xmote_th(gl, state, flags); } -static void greedy_work(void *data) +static void greedy_work(struct work_struct *work) { - struct greedy *gr = data; + struct greedy *gr = container_of(work, struct greedy, gr_work.work); struct gfs2_holder *gh = &gr->gr_gh; struct gfs2_glock *gl = gh->gh_gl; const struct gfs2_glock_operations *glops = gl->gl_ops; @@ -1422,7 +1422,7 @@ int gfs2_glock_be_greedy(struct gfs2_glock *gl, unsigned int time) gfs2_holder_init(gl, 0, 0, gh); set_bit(HIF_GREEDY, &gh->gh_iflags); - INIT_WORK(&gr->gr_work, greedy_work, gr); + INIT_DELAYED_WORK(&gr->gr_work, greedy_work); set_bit(GLF_SKIP_WAITERS2, &gl->gl_flags); schedule_delayed_work(&gr->gr_work, time); diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 42e3bef270c..72dad552aa0 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -577,12 +577,12 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) server->rcv.ptr = (unsigned char*)&server->rcv.buf; server->rcv.len = 10; server->rcv.state = 0; - INIT_WORK(&server->rcv.tq, ncp_tcp_rcv_proc, server); - INIT_WORK(&server->tx.tq, ncp_tcp_tx_proc, server); + INIT_WORK(&server->rcv.tq, ncp_tcp_rcv_proc); + INIT_WORK(&server->tx.tq, ncp_tcp_tx_proc); sock->sk->sk_write_space = ncp_tcp_write_space; } else { - INIT_WORK(&server->rcv.tq, ncpdgram_rcv_proc, server); - INIT_WORK(&server->timeout_tq, ncpdgram_timeout_proc, server); + INIT_WORK(&server->rcv.tq, ncpdgram_rcv_proc); + INIT_WORK(&server->timeout_tq, ncpdgram_timeout_proc); server->timeout_tm.data = (unsigned long)server; server->timeout_tm.function = ncpdgram_timeout_call; } diff --git a/fs/ncpfs/sock.c b/fs/ncpfs/sock.c index 11c2b252ebe..e496d8b65e9 100644 --- a/fs/ncpfs/sock.c +++ b/fs/ncpfs/sock.c @@ -350,9 +350,10 @@ static void info_server(struct ncp_server *server, unsigned int id, const void * } } -void ncpdgram_rcv_proc(void *s) +void ncpdgram_rcv_proc(struct work_struct *work) { - struct ncp_server *server = s; + struct ncp_server *server = + container_of(work, struct ncp_server, rcv.tq); struct socket* sock; sock = server->ncp_sock; @@ -468,9 +469,10 @@ static void __ncpdgram_timeout_proc(struct ncp_server *server) } } -void ncpdgram_timeout_proc(void *s) +void ncpdgram_timeout_proc(struct work_struct *work) { - struct ncp_server *server = s; + struct ncp_server *server = + container_of(work, struct ncp_server, timeout_tq); mutex_lock(&server->rcv.creq_mutex); __ncpdgram_timeout_proc(server); mutex_unlock(&server->rcv.creq_mutex); @@ -652,18 +654,20 @@ skipdata:; } } -void ncp_tcp_rcv_proc(void *s) +void ncp_tcp_rcv_proc(struct work_struct *work) { - struct ncp_server *server = s; + struct ncp_server *server = + container_of(work, struct ncp_server, rcv.tq); mutex_lock(&server->rcv.creq_mutex); __ncptcp_rcv_proc(server); mutex_unlock(&server->rcv.creq_mutex); } -void ncp_tcp_tx_proc(void *s) +void ncp_tcp_tx_proc(struct work_struct *work) { - struct ncp_server *server = s; + struct ncp_server *server = + container_of(work, struct ncp_server, tx.tq); mutex_lock(&server->rcv.creq_mutex); __ncptcp_try_send(server); diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 293b6495829..e431e93ab50 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1829,9 +1829,8 @@ out: } static struct workqueue_struct *laundry_wq; -static struct work_struct laundromat_work; -static void laundromat_main(void *); -static DECLARE_WORK(laundromat_work, laundromat_main, NULL); +static void laundromat_main(struct work_struct *); +static DECLARE_DELAYED_WORK(laundromat_work, laundromat_main); __be32 nfsd4_renew(clientid_t *clid) @@ -1940,7 +1939,7 @@ nfs4_laundromat(void) } void -laundromat_main(void *not_used) +laundromat_main(struct work_struct *not_used) { time_t t; diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index f43bc5f18a3..0b2ad163005 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -1205,10 +1205,12 @@ int ocfs2_flush_truncate_log(struct ocfs2_super *osb) return status; } -static void ocfs2_truncate_log_worker(void *data) +static void ocfs2_truncate_log_worker(struct work_struct *work) { int status; - struct ocfs2_super *osb = data; + struct ocfs2_super *osb = + container_of(work, struct ocfs2_super, + osb_truncate_log_wq.work); mlog_entry_void(); @@ -1441,7 +1443,8 @@ int ocfs2_truncate_log_init(struct ocfs2_super *osb) /* ocfs2_truncate_log_shutdown keys on the existence of * osb->osb_tl_inode so we don't set any of the osb variables * until we're sure all is well. */ - INIT_WORK(&osb->osb_truncate_log_wq, ocfs2_truncate_log_worker, osb); + INIT_DELAYED_WORK(&osb->osb_truncate_log_wq, + ocfs2_truncate_log_worker); osb->osb_tl_bh = tl_bh; osb->osb_tl_inode = tl_inode; diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 305cba3681f..4cd9a958045 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -141,7 +141,7 @@ struct o2hb_region { * recognizes a node going up and down in one iteration */ u64 hr_generation; - struct work_struct hr_write_timeout_work; + struct delayed_work hr_write_timeout_work; unsigned long hr_last_timeout_start; /* Used during o2hb_check_slot to hold a copy of the block @@ -156,9 +156,11 @@ struct o2hb_bio_wait_ctxt { int wc_error; }; -static void o2hb_write_timeout(void *arg) +static void o2hb_write_timeout(struct work_struct *work) { - struct o2hb_region *reg = arg; + struct o2hb_region *reg = + container_of(work, struct o2hb_region, + hr_write_timeout_work.work); mlog(ML_ERROR, "Heartbeat write timeout to device %s after %u " "milliseconds\n", reg->hr_dev_name, @@ -1404,7 +1406,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, goto out; } - INIT_WORK(®->hr_write_timeout_work, o2hb_write_timeout, reg); + INIT_DELAYED_WORK(®->hr_write_timeout_work, o2hb_write_timeout); /* * A node is considered live after it has beat LIVE_THRESHOLD diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c index 7bba98fbfc1..4705d659fe5 100644 --- a/fs/ocfs2/cluster/quorum.c +++ b/fs/ocfs2/cluster/quorum.c @@ -88,7 +88,7 @@ void o2quo_disk_timeout(void) o2quo_fence_self(); } -static void o2quo_make_decision(void *arg) +static void o2quo_make_decision(struct work_struct *work) { int quorum; int lowest_hb, lowest_reachable = 0, fence = 0; @@ -306,7 +306,7 @@ void o2quo_init(void) struct o2quo_state *qs = &o2quo_state; spin_lock_init(&qs->qs_lock); - INIT_WORK(&qs->qs_work, o2quo_make_decision, NULL); + INIT_WORK(&qs->qs_work, o2quo_make_decision); } void o2quo_exit(void) diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index b650efa8c8b..9b3209dc0b1 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -140,11 +140,11 @@ static int o2net_sys_err_translations[O2NET_ERR_MAX] = [O2NET_ERR_DIED] = -EHOSTDOWN,}; /* can't quite avoid *all* internal declarations :/ */ -static void o2net_sc_connect_completed(void *arg); -static void o2net_rx_until_empty(void *arg); -static void o2net_shutdown_sc(void *arg); +static void o2net_sc_connect_completed(struct work_struct *work); +static void o2net_rx_until_empty(struct work_struct *work); +static void o2net_shutdown_sc(struct work_struct *work); static void o2net_listen_data_ready(struct sock *sk, int bytes); -static void o2net_sc_send_keep_req(void *arg); +static void o2net_sc_send_keep_req(struct work_struct *work); static void o2net_idle_timer(unsigned long data); static void o2net_sc_postpone_idle(struct o2net_sock_container *sc); @@ -308,10 +308,10 @@ static struct o2net_sock_container *sc_alloc(struct o2nm_node *node) o2nm_node_get(node); sc->sc_node = node; - INIT_WORK(&sc->sc_connect_work, o2net_sc_connect_completed, sc); - INIT_WORK(&sc->sc_rx_work, o2net_rx_until_empty, sc); - INIT_WORK(&sc->sc_shutdown_work, o2net_shutdown_sc, sc); - INIT_WORK(&sc->sc_keepalive_work, o2net_sc_send_keep_req, sc); + INIT_WORK(&sc->sc_connect_work, o2net_sc_connect_completed); + INIT_WORK(&sc->sc_rx_work, o2net_rx_until_empty); + INIT_WORK(&sc->sc_shutdown_work, o2net_shutdown_sc); + INIT_DELAYED_WORK(&sc->sc_keepalive_work, o2net_sc_send_keep_req); init_timer(&sc->sc_idle_timeout); sc->sc_idle_timeout.function = o2net_idle_timer; @@ -342,7 +342,7 @@ static void o2net_sc_queue_work(struct o2net_sock_container *sc, sc_put(sc); } static void o2net_sc_queue_delayed_work(struct o2net_sock_container *sc, - struct work_struct *work, + struct delayed_work *work, int delay) { sc_get(sc); @@ -350,7 +350,7 @@ static void o2net_sc_queue_delayed_work(struct o2net_sock_container *sc, sc_put(sc); } static void o2net_sc_cancel_delayed_work(struct o2net_sock_container *sc, - struct work_struct *work) + struct delayed_work *work) { if (cancel_delayed_work(work)) sc_put(sc); @@ -564,9 +564,11 @@ static void o2net_ensure_shutdown(struct o2net_node *nn, * ourselves as state_change couldn't get the nn_lock and call set_nn_state * itself. */ -static void o2net_shutdown_sc(void *arg) +static void o2net_shutdown_sc(struct work_struct *work) { - struct o2net_sock_container *sc = arg; + struct o2net_sock_container *sc = + container_of(work, struct o2net_sock_container, + sc_shutdown_work); struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num); sclog(sc, "shutting down\n"); @@ -1201,9 +1203,10 @@ out: /* this work func is triggerd by data ready. it reads until it can read no * more. it interprets 0, eof, as fatal. if data_ready hits while we're doing * our work the work struct will be marked and we'll be called again. */ -static void o2net_rx_until_empty(void *arg) +static void o2net_rx_until_empty(struct work_struct *work) { - struct o2net_sock_container *sc = arg; + struct o2net_sock_container *sc = + container_of(work, struct o2net_sock_container, sc_rx_work); int ret; do { @@ -1249,9 +1252,11 @@ static int o2net_set_nodelay(struct socket *sock) /* called when a connect completes and after a sock is accepted. the * rx path will see the response and mark the sc valid */ -static void o2net_sc_connect_completed(void *arg) +static void o2net_sc_connect_completed(struct work_struct *work) { - struct o2net_sock_container *sc = arg; + struct o2net_sock_container *sc = + container_of(work, struct o2net_sock_container, + sc_connect_work); mlog(ML_MSG, "sc sending handshake with ver %llu id %llx\n", (unsigned long long)O2NET_PROTOCOL_VERSION, @@ -1262,9 +1267,11 @@ static void o2net_sc_connect_completed(void *arg) } /* this is called as a work_struct func. */ -static void o2net_sc_send_keep_req(void *arg) +static void o2net_sc_send_keep_req(struct work_struct *work) { - struct o2net_sock_container *sc = arg; + struct o2net_sock_container *sc = + container_of(work, struct o2net_sock_container, + sc_keepalive_work.work); o2net_sendpage(sc, o2net_keep_req, sizeof(*o2net_keep_req)); sc_put(sc); @@ -1314,14 +1321,15 @@ static void o2net_sc_postpone_idle(struct o2net_sock_container *sc) * having a connect attempt fail, etc. This centralizes the logic which decides * if a connect attempt should be made or if we should give up and all future * transmit attempts should fail */ -static void o2net_start_connect(void *arg) +static void o2net_start_connect(struct work_struct *work) { - struct o2net_node *nn = arg; + struct o2net_node *nn = + container_of(work, struct o2net_node, nn_connect_work.work); struct o2net_sock_container *sc = NULL; struct o2nm_node *node = NULL, *mynode = NULL; struct socket *sock = NULL; struct sockaddr_in myaddr = {0, }, remoteaddr = {0, }; - int ret = 0; + int ret = 0, stop; /* if we're greater we initiate tx, otherwise we accept */ if (o2nm_this_node() <= o2net_num_from_nn(nn)) @@ -1342,10 +1350,9 @@ static void o2net_start_connect(void *arg) spin_lock(&nn->nn_lock); /* see if we already have one pending or have given up */ - if (nn->nn_sc || nn->nn_persistent_error) - arg = NULL; + stop = (nn->nn_sc || nn->nn_persistent_error); spin_unlock(&nn->nn_lock); - if (arg == NULL) /* *shrug*, needed some indicator */ + if (stop) goto out; nn->nn_last_connect_attempt = jiffies; @@ -1421,9 +1428,10 @@ out: return; } -static void o2net_connect_expired(void *arg) +static void o2net_connect_expired(struct work_struct *work) { - struct o2net_node *nn = arg; + struct o2net_node *nn = + container_of(work, struct o2net_node, nn_connect_expired.work); spin_lock(&nn->nn_lock); if (!nn->nn_sc_valid) { @@ -1436,9 +1444,10 @@ static void o2net_connect_expired(void *arg) spin_unlock(&nn->nn_lock); } -static void o2net_still_up(void *arg) +static void o2net_still_up(struct work_struct *work) { - struct o2net_node *nn = arg; + struct o2net_node *nn = + container_of(work, struct o2net_node, nn_still_up.work); o2quo_hb_still_up(o2net_num_from_nn(nn)); } @@ -1644,9 +1653,9 @@ out: return ret; } -static void o2net_accept_many(void *arg) +static void o2net_accept_many(struct work_struct *work) { - struct socket *sock = arg; + struct socket *sock = o2net_listen_sock; while (o2net_accept_one(sock) == 0) cond_resched(); } @@ -1700,7 +1709,7 @@ static int o2net_open_listening_sock(__be16 port) write_unlock_bh(&sock->sk->sk_callback_lock); o2net_listen_sock = sock; - INIT_WORK(&o2net_listen_work, o2net_accept_many, sock); + INIT_WORK(&o2net_listen_work, o2net_accept_many); sock->sk->sk_reuse = 1; ret = sock->ops->bind(sock, (struct sockaddr *)&sin, sizeof(sin)); @@ -1819,9 +1828,10 @@ int o2net_init(void) struct o2net_node *nn = o2net_nn_from_num(i); spin_lock_init(&nn->nn_lock); - INIT_WORK(&nn->nn_connect_work, o2net_start_connect, nn); - INIT_WORK(&nn->nn_connect_expired, o2net_connect_expired, nn); - INIT_WORK(&nn->nn_still_up, o2net_still_up, nn); + INIT_DELAYED_WORK(&nn->nn_connect_work, o2net_start_connect); + INIT_DELAYED_WORK(&nn->nn_connect_expired, + o2net_connect_expired); + INIT_DELAYED_WORK(&nn->nn_still_up, o2net_still_up); /* until we see hb from a node we'll return einval */ nn->nn_persistent_error = -ENOTCONN; init_waitqueue_head(&nn->nn_sc_wq); diff --git a/fs/ocfs2/cluster/tcp_internal.h b/fs/ocfs2/cluster/tcp_internal.h index 4b46aac7d24..daebbd3a2c8 100644 --- a/fs/ocfs2/cluster/tcp_internal.h +++ b/fs/ocfs2/cluster/tcp_internal.h @@ -86,18 +86,18 @@ struct o2net_node { * connect attempt fails and so can be self-arming. shutdown is * careful to first mark the nn such that no connects will be attempted * before canceling delayed connect work and flushing the queue. */ - struct work_struct nn_connect_work; + struct delayed_work nn_connect_work; unsigned long nn_last_connect_attempt; /* this is queued as nodes come up and is canceled when a connection is * established. this expiring gives up on the node and errors out * transmits */ - struct work_struct nn_connect_expired; + struct delayed_work nn_connect_expired; /* after we give up on a socket we wait a while before deciding * that it is still heartbeating and that we should do some * quorum work */ - struct work_struct nn_still_up; + struct delayed_work nn_still_up; }; struct o2net_sock_container { @@ -129,7 +129,7 @@ struct o2net_sock_container { struct work_struct sc_shutdown_work; struct timer_list sc_idle_timeout; - struct work_struct sc_keepalive_work; + struct delayed_work sc_keepalive_work; unsigned sc_handshake_ok:1; diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index fa968180b07..6b6ff76538c 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -153,7 +153,7 @@ static inline struct hlist_head *dlm_lockres_hash(struct dlm_ctxt *dlm, unsigned * called functions that cannot be directly called from the * net message handlers for some reason, usually because * they need to send net messages of their own. */ -void dlm_dispatch_work(void *data); +void dlm_dispatch_work(struct work_struct *work); struct dlm_lock_resource; struct dlm_work_item; diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 8d1065f8b3b..637646e6922 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -1296,7 +1296,7 @@ static struct dlm_ctxt *dlm_alloc_ctxt(const char *domain, spin_lock_init(&dlm->work_lock); INIT_LIST_HEAD(&dlm->work_list); - INIT_WORK(&dlm->dispatched_work, dlm_dispatch_work, dlm); + INIT_WORK(&dlm->dispatched_work, dlm_dispatch_work); kref_init(&dlm->dlm_refs); dlm->dlm_state = DLM_CTXT_NEW; diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index 9d950d7cea3..fb3e2b0817f 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -153,9 +153,10 @@ static inline void dlm_reset_recovery(struct dlm_ctxt *dlm) } /* Worker function used during recovery. */ -void dlm_dispatch_work(void *data) +void dlm_dispatch_work(struct work_struct *work) { - struct dlm_ctxt *dlm = (struct dlm_ctxt *)data; + struct dlm_ctxt *dlm = + container_of(work, struct dlm_ctxt, dispatched_work); LIST_HEAD(tmp_list); struct list_head *iter, *iter2; struct dlm_work_item *item; diff --git a/fs/ocfs2/dlm/userdlm.c b/fs/ocfs2/dlm/userdlm.c index eead48bbfac..7d2f578b267 100644 --- a/fs/ocfs2/dlm/userdlm.c +++ b/fs/ocfs2/dlm/userdlm.c @@ -171,15 +171,14 @@ static inline void user_dlm_grab_inode_ref(struct user_lock_res *lockres) BUG(); } -static void user_dlm_unblock_lock(void *opaque); +static void user_dlm_unblock_lock(struct work_struct *work); static void __user_dlm_queue_lockres(struct user_lock_res *lockres) { if (!(lockres->l_flags & USER_LOCK_QUEUED)) { user_dlm_grab_inode_ref(lockres); - INIT_WORK(&lockres->l_work, user_dlm_unblock_lock, - lockres); + INIT_WORK(&lockres->l_work, user_dlm_unblock_lock); queue_work(user_dlm_worker, &lockres->l_work); lockres->l_flags |= USER_LOCK_QUEUED; @@ -279,10 +278,11 @@ static inline void user_dlm_drop_inode_ref(struct user_lock_res *lockres) iput(inode); } -static void user_dlm_unblock_lock(void *opaque) +static void user_dlm_unblock_lock(struct work_struct *work) { int new_level, status; - struct user_lock_res *lockres = (struct user_lock_res *) opaque; + struct user_lock_res *lockres = + container_of(work, struct user_lock_res, l_work); struct dlm_ctxt *dlm = dlm_ctxt_from_user_lockres(lockres); mlog(0, "processing lockres %.*s\n", lockres->l_namelen, diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index fd9734def55..d95ee2720e6 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -911,11 +911,12 @@ struct ocfs2_la_recovery_item { * NOTE: This function can and will sleep on recovery of other nodes * during cluster locking, just like any other ocfs2 process. */ -void ocfs2_complete_recovery(void *data) +void ocfs2_complete_recovery(struct work_struct *work) { int ret; - struct ocfs2_super *osb = data; - struct ocfs2_journal *journal = osb->journal; + struct ocfs2_journal *journal = + container_of(work, struct ocfs2_journal, j_recovery_work); + struct ocfs2_super *osb = journal->j_osb; struct ocfs2_dinode *la_dinode, *tl_dinode; struct ocfs2_la_recovery_item *item; struct list_head *p, *n; diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index 2f3a6acdac4..5be161a4ad9 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -172,7 +172,7 @@ static inline void ocfs2_handle_set_sync(struct ocfs2_journal_handle *handle, in } /* Exported only for the journal struct init code in super.c. Do not call. */ -void ocfs2_complete_recovery(void *data); +void ocfs2_complete_recovery(struct work_struct *work); /* * Journal Control: diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 0462a7f4e21..9b1bad1d48e 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -283,7 +283,7 @@ struct ocfs2_super /* Truncate log info */ struct inode *osb_tl_inode; struct buffer_head *osb_tl_bh; - struct work_struct osb_truncate_log_wq; + struct delayed_work osb_truncate_log_wq; struct ocfs2_node_map osb_recovering_orphan_dirs; unsigned int *osb_orphan_wipes; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 76b46ebbb10..9a8089030f5 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1365,7 +1365,7 @@ static int ocfs2_initialize_super(struct super_block *sb, spin_lock_init(&journal->j_lock); journal->j_trans_id = (unsigned long) 1; INIT_LIST_HEAD(&journal->j_la_cleanups); - INIT_WORK(&journal->j_recovery_work, ocfs2_complete_recovery, osb); + INIT_WORK(&journal->j_recovery_work, ocfs2_complete_recovery); journal->j_state = OCFS2_JOURNAL_FREE; /* get some pseudo constants for clustersize bits */ diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 85ce2326830..cd1bb75ceb2 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -104,7 +104,7 @@ static int release_journal_dev(struct super_block *super, struct reiserfs_journal *journal); static int dirty_one_transaction(struct super_block *s, struct reiserfs_journal_list *jl); -static void flush_async_commits(void *p); +static void flush_async_commits(struct work_struct *work); static void queue_log_writer(struct super_block *s); /* values for join in do_journal_begin_r */ @@ -2836,7 +2836,8 @@ int journal_init(struct super_block *p_s_sb, const char *j_dev_name, if (reiserfs_mounted_fs_count <= 1) commit_wq = create_workqueue("reiserfs"); - INIT_WORK(&journal->j_work, flush_async_commits, p_s_sb); + INIT_DELAYED_WORK(&journal->j_work, flush_async_commits); + journal->j_work_sb = p_s_sb; return 0; free_and_return: free_journal_ram(p_s_sb); @@ -3447,10 +3448,11 @@ int journal_end_sync(struct reiserfs_transaction_handle *th, /* ** writeback the pending async commits to disk */ -static void flush_async_commits(void *p) +static void flush_async_commits(struct work_struct *work) { - struct super_block *p_s_sb = p; - struct reiserfs_journal *journal = SB_JOURNAL(p_s_sb); + struct reiserfs_journal *journal = + container_of(work, struct reiserfs_journal, j_work.work); + struct super_block *p_s_sb = journal->j_work_sb; struct reiserfs_journal_list *jl; struct list_head *entry; diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 09360cf1e1f..8e6b56fc1ca 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -149,9 +149,10 @@ xfs_destroy_ioend( */ STATIC void xfs_end_bio_delalloc( - void *data) + struct work_struct *work) { - xfs_ioend_t *ioend = data; + xfs_ioend_t *ioend = + container_of(work, xfs_ioend_t, io_work); xfs_destroy_ioend(ioend); } @@ -161,9 +162,10 @@ xfs_end_bio_delalloc( */ STATIC void xfs_end_bio_written( - void *data) + struct work_struct *work) { - xfs_ioend_t *ioend = data; + xfs_ioend_t *ioend = + container_of(work, xfs_ioend_t, io_work); xfs_destroy_ioend(ioend); } @@ -176,9 +178,10 @@ xfs_end_bio_written( */ STATIC void xfs_end_bio_unwritten( - void *data) + struct work_struct *work) { - xfs_ioend_t *ioend = data; + xfs_ioend_t *ioend = + container_of(work, xfs_ioend_t, io_work); bhv_vnode_t *vp = ioend->io_vnode; xfs_off_t offset = ioend->io_offset; size_t size = ioend->io_size; @@ -220,11 +223,11 @@ xfs_alloc_ioend( ioend->io_size = 0; if (type == IOMAP_UNWRITTEN) - INIT_WORK(&ioend->io_work, xfs_end_bio_unwritten, ioend); + INIT_WORK(&ioend->io_work, xfs_end_bio_unwritten); else if (type == IOMAP_DELAY) - INIT_WORK(&ioend->io_work, xfs_end_bio_delalloc, ioend); + INIT_WORK(&ioend->io_work, xfs_end_bio_delalloc); else - INIT_WORK(&ioend->io_work, xfs_end_bio_written, ioend); + INIT_WORK(&ioend->io_work, xfs_end_bio_written); return ioend; } diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index d3382843698..eef4a0ba11e 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -994,9 +994,10 @@ xfs_buf_wait_unpin( STATIC void xfs_buf_iodone_work( - void *v) + struct work_struct *work) { - xfs_buf_t *bp = (xfs_buf_t *)v; + xfs_buf_t *bp = + container_of(work, xfs_buf_t, b_iodone_work); if (bp->b_iodone) (*(bp->b_iodone))(bp); @@ -1017,10 +1018,10 @@ xfs_buf_ioend( if ((bp->b_iodone) || (bp->b_flags & XBF_ASYNC)) { if (schedule) { - INIT_WORK(&bp->b_iodone_work, xfs_buf_iodone_work, bp); + INIT_WORK(&bp->b_iodone_work, xfs_buf_iodone_work); queue_work(xfslogd_workqueue, &bp->b_iodone_work); } else { - xfs_buf_iodone_work(bp); + xfs_buf_iodone_work(&bp->b_iodone_work); } } else { up(&bp->b_iodonesema); diff --git a/include/linux/connector.h b/include/linux/connector.h index 4c02119c6ab..3ea1cd58de9 100644 --- a/include/linux/connector.h +++ b/include/linux/connector.h @@ -133,7 +133,7 @@ struct cn_callback_data { struct cn_callback_entry { struct list_head callback_entry; struct cn_callback *cb; - struct work_struct work; + struct delayed_work work; struct cn_queue_dev *pdev; struct cn_callback_id id; @@ -170,7 +170,7 @@ void cn_queue_free_dev(struct cn_queue_dev *dev); int cn_cb_equal(struct cb_id *, struct cb_id *); -void cn_queue_wrapper(void *data); +void cn_queue_wrapper(struct work_struct *work); extern int cn_already_initialized; diff --git a/include/linux/i2o.h b/include/linux/i2o.h index c115e9e840b..1fb02e17f6f 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -461,7 +461,7 @@ struct i2o_driver { int (*reply) (struct i2o_controller *, u32, struct i2o_message *); /* Event handler */ - void (*event) (struct i2o_event *); + work_func_t event; struct workqueue_struct *event_queue; /* Event queue */ diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h index 587264a58d5..8b08ef3820f 100644 --- a/include/linux/mmc/host.h +++ b/include/linux/mmc/host.h @@ -110,7 +110,7 @@ struct mmc_host { struct mmc_card *card_busy; /* the MMC card claiming host */ struct mmc_card *card_selected; /* the selected MMC card */ - struct work_struct detect; + struct delayed_work detect; unsigned long private[0] ____cacheline_aligned; }; diff --git a/include/linux/ncp_fs_sb.h b/include/linux/ncp_fs_sb.h index b089d950628..a503052138b 100644 --- a/include/linux/ncp_fs_sb.h +++ b/include/linux/ncp_fs_sb.h @@ -127,10 +127,10 @@ struct ncp_server { } unexpected_packet; }; -extern void ncp_tcp_rcv_proc(void *server); -extern void ncp_tcp_tx_proc(void *server); -extern void ncpdgram_rcv_proc(void *server); -extern void ncpdgram_timeout_proc(void *server); +extern void ncp_tcp_rcv_proc(struct work_struct *work); +extern void ncp_tcp_tx_proc(struct work_struct *work); +extern void ncpdgram_rcv_proc(struct work_struct *work); +extern void ncpdgram_timeout_proc(struct work_struct *work); extern void ncpdgram_timeout_call(unsigned long server); extern void ncp_tcp_data_ready(struct sock* sk, int len); extern void ncp_tcp_write_space(struct sock* sk); diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h index 73e0becec08..6610103f23e 100644 --- a/include/linux/reiserfs_fs_sb.h +++ b/include/linux/reiserfs_fs_sb.h @@ -249,7 +249,8 @@ struct reiserfs_journal { int j_errno; /* when flushing ordered buffers, throttle new ordered writers */ - struct work_struct j_work; + struct delayed_work j_work; + struct super_block *j_work_sb; atomic_t j_async_throttle; }; diff --git a/include/linux/relay.h b/include/linux/relay.h index 24accb48384..0e3d91b7699 100644 --- a/include/linux/relay.h +++ b/include/linux/relay.h @@ -38,7 +38,7 @@ struct rchan_buf size_t subbufs_consumed; /* count of sub-buffers consumed */ struct rchan *chan; /* associated channel */ wait_queue_head_t read_wait; /* reader wait queue */ - struct work_struct wake_readers; /* reader wake-up work struct */ + struct delayed_work wake_readers; /* reader wake-up work struct */ struct dentry *dentry; /* channel file dentry */ struct kref kref; /* channel buffer refcount */ struct page **page_array; /* array of current buffer pages */ diff --git a/include/linux/usb.h b/include/linux/usb.h index 5482bfb3303..06ce7a62604 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -382,7 +382,7 @@ struct usb_device { int pm_usage_cnt; /* usage counter for autosuspend */ #ifdef CONFIG_PM - struct work_struct autosuspend; /* for delayed autosuspends */ + struct delayed_work autosuspend; /* for delayed autosuspends */ struct mutex pm_mutex; /* protects PM operations */ unsigned auto_pm:1; /* autosuspend/resume in progress */ diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h index 617b672b113..89119277553 100644 --- a/include/net/ieee80211softmac.h +++ b/include/net/ieee80211softmac.h @@ -108,8 +108,8 @@ struct ieee80211softmac_assoc_info { /* Scan retries remaining */ int scan_retry; - struct work_struct work; - struct work_struct timeout; + struct delayed_work work; + struct delayed_work timeout; }; struct ieee80211softmac_bss_info { diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index c6d93bb0dcd..60b684470db 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1030,7 +1030,7 @@ void sctp_inq_init(struct sctp_inq *); void sctp_inq_free(struct sctp_inq *); void sctp_inq_push(struct sctp_inq *, struct sctp_chunk *packet); struct sctp_chunk *sctp_inq_pop(struct sctp_inq *); -void sctp_inq_set_th_handler(struct sctp_inq *, void (*)(void *), void *); +void sctp_inq_set_th_handler(struct sctp_inq *, work_func_t); /* This is the structure we use to hold outbound chunks. You push * chunks in and they automatically pop out the other end as bundled diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h index 9582e840166..9ccc0365aa8 100644 --- a/include/scsi/libsas.h +++ b/include/scsi/libsas.h @@ -200,9 +200,14 @@ struct domain_device { void *lldd_dev; }; +struct sas_discovery_event { + struct work_struct work; + struct asd_sas_port *port; +}; + struct sas_discovery { spinlock_t disc_event_lock; - struct work_struct disc_work[DISC_NUM_EVENTS]; + struct sas_discovery_event disc_work[DISC_NUM_EVENTS]; unsigned long pending; u8 fanout_sas_addr[8]; u8 eeds_a[8]; @@ -248,14 +253,19 @@ struct asd_sas_port { void *lldd_port; /* not touched by the sas class code */ }; +struct asd_sas_event { + struct work_struct work; + struct asd_sas_phy *phy; +}; + /* The phy pretty much is controlled by the LLDD. * The class only reads those fields. */ struct asd_sas_phy { /* private: */ /* protected by ha->event_lock */ - struct work_struct port_events[PORT_NUM_EVENTS]; - struct work_struct phy_events[PHY_NUM_EVENTS]; + struct asd_sas_event port_events[PORT_NUM_EVENTS]; + struct asd_sas_event phy_events[PHY_NUM_EVENTS]; unsigned long port_events_pending; unsigned long phy_events_pending; @@ -307,10 +317,15 @@ struct scsi_core { int queue_thread_kill; }; +struct sas_ha_event { + struct work_struct work; + struct sas_ha_struct *ha; +}; + struct sas_ha_struct { /* private: */ spinlock_t event_lock; - struct work_struct ha_events[HA_NUM_EVENTS]; + struct sas_ha_event ha_events[HA_NUM_EVENTS]; unsigned long pending; struct scsi_core core; diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h index fd352323378..798f7c7ee42 100644 --- a/include/scsi/scsi_transport_fc.h +++ b/include/scsi/scsi_transport_fc.h @@ -206,9 +206,9 @@ struct fc_rport { /* aka fc_starget_attrs */ u8 flags; struct list_head peers; struct device dev; - struct work_struct dev_loss_work; + struct delayed_work dev_loss_work; struct work_struct scan_work; - struct work_struct fail_io_work; + struct delayed_work fail_io_work; struct work_struct stgt_delete_work; struct work_struct rport_delete_work; } __attribute__((aligned(sizeof(unsigned long)))); diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h index 4b95c89c95c..d5c218ddc52 100644 --- a/include/scsi/scsi_transport_iscsi.h +++ b/include/scsi/scsi_transport_iscsi.h @@ -176,7 +176,7 @@ struct iscsi_cls_session { /* recovery fields */ int recovery_tmo; - struct work_struct recovery_work; + struct delayed_work recovery_work; int target_id; diff --git a/include/sound/ac97_codec.h b/include/sound/ac97_codec.h index 4c43521cc49..33720397a90 100644 --- a/include/sound/ac97_codec.h +++ b/include/sound/ac97_codec.h @@ -511,7 +511,7 @@ struct snd_ac97 { #ifdef CONFIG_SND_AC97_POWER_SAVE unsigned int power_up; /* power states */ struct workqueue_struct *power_workq; - struct work_struct power_work; + struct delayed_work power_work; #endif struct device dev; }; diff --git a/include/sound/ak4114.h b/include/sound/ak4114.h index 11702aa0bea..2ee061625fd 100644 --- a/include/sound/ak4114.h +++ b/include/sound/ak4114.h @@ -182,7 +182,7 @@ struct ak4114 { unsigned char rcs0; unsigned char rcs1; struct workqueue_struct *workqueue; - struct work_struct work; + struct delayed_work work; void *change_callback_private; void (*change_callback)(struct ak4114 *ak4114, unsigned char c0, unsigned char c1); }; diff --git a/kernel/relay.c b/kernel/relay.c index f04bbdb56ac..2b92e8ece85 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -308,9 +308,10 @@ static struct rchan_callbacks default_channel_callbacks = { * reason waking is deferred is that calling directly from write * causes problems if you're writing from say the scheduler. */ -static void wakeup_readers(void *private) +static void wakeup_readers(struct work_struct *work) { - struct rchan_buf *buf = private; + struct rchan_buf *buf = + container_of(work, struct rchan_buf, wake_readers.work); wake_up_interruptible(&buf->read_wait); } @@ -328,7 +329,7 @@ static inline void __relay_reset(struct rchan_buf *buf, unsigned int init) if (init) { init_waitqueue_head(&buf->read_wait); kref_init(&buf->kref); - INIT_WORK(&buf->wake_readers, NULL, NULL); + INIT_DELAYED_WORK(&buf->wake_readers, NULL); } else { cancel_delayed_work(&buf->wake_readers); flush_scheduled_work(); @@ -549,7 +550,8 @@ size_t relay_switch_subbuf(struct rchan_buf *buf, size_t length) buf->padding[old_subbuf]; smp_mb(); if (waitqueue_active(&buf->read_wait)) { - PREPARE_WORK(&buf->wake_readers, wakeup_readers, buf); + PREPARE_DELAYED_WORK(&buf->wake_readers, + wakeup_readers); schedule_delayed_work(&buf->wake_readers, 1); } } diff --git a/mm/swap.c b/mm/swap.c index 2e0e871f542..d9a3770d8f3 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -216,7 +216,7 @@ void lru_add_drain(void) } #ifdef CONFIG_NUMA -static void lru_add_drain_per_cpu(void *dummy) +static void lru_add_drain_per_cpu(struct work_struct *dummy) { lru_add_drain(); } @@ -226,7 +226,7 @@ static void lru_add_drain_per_cpu(void *dummy) */ int lru_add_drain_all(void) { - return schedule_on_each_cpu(lru_add_drain_per_cpu, NULL); + return schedule_on_each_cpu(lru_add_drain_per_cpu); } #else diff --git a/net/atm/lec.c b/net/atm/lec.c index 66c57c1091a..e801fff69dc 100644 --- a/net/atm/lec.c +++ b/net/atm/lec.c @@ -1458,7 +1458,7 @@ static void lane2_associate_ind(struct net_device *dev, u8 *mac_addr, #define LEC_ARP_REFRESH_INTERVAL (3*HZ) -static void lec_arp_check_expire(void *data); +static void lec_arp_check_expire(struct work_struct *work); static void lec_arp_expire_arp(unsigned long data); /* @@ -1481,7 +1481,7 @@ static void lec_arp_init(struct lec_priv *priv) INIT_HLIST_HEAD(&priv->lec_no_forward); INIT_HLIST_HEAD(&priv->mcast_fwds); spin_lock_init(&priv->lec_arp_lock); - INIT_WORK(&priv->lec_arp_work, lec_arp_check_expire, priv); + INIT_DELAYED_WORK(&priv->lec_arp_work, lec_arp_check_expire); schedule_delayed_work(&priv->lec_arp_work, LEC_ARP_REFRESH_INTERVAL); } @@ -1879,10 +1879,11 @@ static void lec_arp_expire_vcc(unsigned long data) * to ESI_FORWARD_DIRECT. This causes the flush period to end * regardless of the progress of the flush protocol. */ -static void lec_arp_check_expire(void *data) +static void lec_arp_check_expire(struct work_struct *work) { unsigned long flags; - struct lec_priv *priv = data; + struct lec_priv *priv = + container_of(work, struct lec_priv, lec_arp_work.work); struct hlist_node *node, *next; struct lec_arp_table *entry; unsigned long now; diff --git a/net/atm/lec.h b/net/atm/lec.h index 877f5093969..984e8e6e083 100644 --- a/net/atm/lec.h +++ b/net/atm/lec.h @@ -92,7 +92,7 @@ struct lec_priv { spinlock_t lec_arp_lock; struct atm_vcc *mcast_vcc; /* Default Multicast Send VCC */ struct atm_vcc *lecd; - struct work_struct lec_arp_work; /* C10 */ + struct delayed_work lec_arp_work; /* C10 */ unsigned int maximum_unknown_frame_count; /* * Within the period of time defined by this variable, the client will send diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 3eeeb7a86e7..d4c935692cc 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -237,9 +237,9 @@ static void bt_release(struct device *dev) kfree(data); } -static void add_conn(void *data) +static void add_conn(struct work_struct *work) { - struct hci_conn *conn = data; + struct hci_conn *conn = container_of(work, struct hci_conn, work); int i; if (device_register(&conn->dev) < 0) { @@ -272,14 +272,14 @@ void hci_conn_add_sysfs(struct hci_conn *conn) dev_set_drvdata(&conn->dev, conn); - INIT_WORK(&conn->work, add_conn, (void *) conn); + INIT_WORK(&conn->work, add_conn); schedule_work(&conn->work); } -static void del_conn(void *data) +static void del_conn(struct work_struct *work) { - struct hci_conn *conn = data; + struct hci_conn *conn = container_of(work, struct hci_conn, work); device_del(&conn->dev); } @@ -287,7 +287,7 @@ void hci_conn_del_sysfs(struct hci_conn *conn) { BT_DBG("conn %p", conn); - INIT_WORK(&conn->work, del_conn, (void *) conn); + INIT_WORK(&conn->work, del_conn); schedule_work(&conn->work); } diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index f753c40c11d..55bb2634c08 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -77,12 +77,16 @@ static int port_cost(struct net_device *dev) * Called from work queue to allow for calling functions that * might sleep (such as speed check), and to debounce. */ -static void port_carrier_check(void *arg) +static void port_carrier_check(struct work_struct *work) { - struct net_device *dev = arg; struct net_bridge_port *p; + struct net_device *dev; struct net_bridge *br; + dev = container_of(work, struct net_bridge_port, + carrier_check.work)->dev; + work_release(work); + rtnl_lock(); p = dev->br_port; if (!p) @@ -276,7 +280,7 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br, p->port_no = index; br_init_port(p); p->state = BR_STATE_DISABLED; - INIT_WORK(&p->carrier_check, port_carrier_check, dev); + INIT_DELAYED_WORK_NAR(&p->carrier_check, port_carrier_check); br_stp_port_timer_init(p); kobject_init(&p->kobj); diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 74258d86f25..3a534e94c7f 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -82,7 +82,7 @@ struct net_bridge_port struct timer_list hold_timer; struct timer_list message_age_timer; struct kobject kobj; - struct work_struct carrier_check; + struct delayed_work carrier_check; struct rcu_head rcu; }; diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 6589adb14cb..63f24c914dd 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -56,7 +56,7 @@ static atomic_t trapped; static void zap_completion_queue(void); static void arp_reply(struct sk_buff *skb); -static void queue_process(void *p) +static void queue_process(struct work_struct *work) { unsigned long flags; struct sk_buff *skb; @@ -77,7 +77,7 @@ static void queue_process(void *p) } } -static DECLARE_WORK(send_queue, queue_process, NULL); +static DECLARE_WORK(send_queue, queue_process); void netpoll_queue(struct sk_buff *skb) { diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 9045438d6b3..36db5be2a9e 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -31,8 +31,7 @@ struct inet_timewait_death_row dccp_death_row = { .tw_timer = TIMER_INITIALIZER(inet_twdr_hangman, 0, (unsigned long)&dccp_death_row), .twkill_work = __WORK_INITIALIZER(dccp_death_row.twkill_work, - inet_twdr_twkill_work, - &dccp_death_row), + inet_twdr_twkill_work), /* Short-time timewait calendar */ .twcal_hand = -1, diff --git a/net/ieee80211/softmac/ieee80211softmac_assoc.c b/net/ieee80211/softmac/ieee80211softmac_assoc.c index cf51c87a971..08386c10295 100644 --- a/net/ieee80211/softmac/ieee80211softmac_assoc.c +++ b/net/ieee80211/softmac/ieee80211softmac_assoc.c @@ -58,9 +58,11 @@ ieee80211softmac_assoc(struct ieee80211softmac_device *mac, struct ieee80211soft } void -ieee80211softmac_assoc_timeout(void *d) +ieee80211softmac_assoc_timeout(struct work_struct *work) { - struct ieee80211softmac_device *mac = (struct ieee80211softmac_device *)d; + struct ieee80211softmac_device *mac = + container_of(work, struct ieee80211softmac_device, + associnfo.timeout.work); struct ieee80211softmac_network *n; mutex_lock(&mac->associnfo.mutex); @@ -186,9 +188,11 @@ ieee80211softmac_assoc_notify_auth(struct net_device *dev, int event_type, void /* This function is called to handle userspace requests (asynchronously) */ void -ieee80211softmac_assoc_work(void *d) +ieee80211softmac_assoc_work(struct work_struct *work) { - struct ieee80211softmac_device *mac = (struct ieee80211softmac_device *)d; + struct ieee80211softmac_device *mac = + container_of(work, struct ieee80211softmac_device, + associnfo.work.work); struct ieee80211softmac_network *found = NULL; struct ieee80211_network *net = NULL, *best = NULL; int bssvalid; @@ -412,7 +416,7 @@ ieee80211softmac_handle_assoc_response(struct net_device * dev, network->authenticated = 0; /* we don't want to do this more than once ... */ network->auth_desynced_once = 1; - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); break; } default: @@ -446,7 +450,7 @@ ieee80211softmac_handle_disassoc(struct net_device * dev, ieee80211softmac_disassoc(mac); /* try to reassociate */ - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); return 0; } @@ -466,7 +470,7 @@ ieee80211softmac_handle_reassoc_req(struct net_device * dev, dprintkl(KERN_INFO PFX "reassoc request from unknown network\n"); return 0; } - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); return 0; } diff --git a/net/ieee80211/softmac/ieee80211softmac_auth.c b/net/ieee80211/softmac/ieee80211softmac_auth.c index 4cef39e171d..2ae1833b657 100644 --- a/net/ieee80211/softmac/ieee80211softmac_auth.c +++ b/net/ieee80211/softmac/ieee80211softmac_auth.c @@ -26,7 +26,7 @@ #include "ieee80211softmac_priv.h" -static void ieee80211softmac_auth_queue(void *data); +static void ieee80211softmac_auth_queue(struct work_struct *work); /* Queues an auth request to the desired AP */ int @@ -54,14 +54,14 @@ ieee80211softmac_auth_req(struct ieee80211softmac_device *mac, auth->mac = mac; auth->retry = IEEE80211SOFTMAC_AUTH_RETRY_LIMIT; auth->state = IEEE80211SOFTMAC_AUTH_OPEN_REQUEST; - INIT_WORK(&auth->work, &ieee80211softmac_auth_queue, (void *)auth); + INIT_DELAYED_WORK(&auth->work, ieee80211softmac_auth_queue); /* Lock (for list) */ spin_lock_irqsave(&mac->lock, flags); /* add to list */ list_add_tail(&auth->list, &mac->auth_queue); - schedule_work(&auth->work); + schedule_delayed_work(&auth->work, 0); spin_unlock_irqrestore(&mac->lock, flags); return 0; @@ -70,14 +70,15 @@ ieee80211softmac_auth_req(struct ieee80211softmac_device *mac, /* Sends an auth request to the desired AP and handles timeouts */ static void -ieee80211softmac_auth_queue(void *data) +ieee80211softmac_auth_queue(struct work_struct *work) { struct ieee80211softmac_device *mac; struct ieee80211softmac_auth_queue_item *auth; struct ieee80211softmac_network *net; unsigned long flags; - auth = (struct ieee80211softmac_auth_queue_item *)data; + auth = container_of(work, struct ieee80211softmac_auth_queue_item, + work.work); net = auth->net; mac = auth->mac; @@ -118,9 +119,11 @@ ieee80211softmac_auth_queue(void *data) /* Sends a response to an auth challenge (for shared key auth). */ static void -ieee80211softmac_auth_challenge_response(void *_aq) +ieee80211softmac_auth_challenge_response(struct work_struct *work) { - struct ieee80211softmac_auth_queue_item *aq = _aq; + struct ieee80211softmac_auth_queue_item *aq = + container_of(work, struct ieee80211softmac_auth_queue_item, + work.work); /* Send our response */ ieee80211softmac_send_mgt_frame(aq->mac, aq->net, IEEE80211_STYPE_AUTH, aq->state); @@ -228,8 +231,8 @@ ieee80211softmac_auth_resp(struct net_device *dev, struct ieee80211_auth *auth) * we have obviously already sent the initial auth * request. */ cancel_delayed_work(&aq->work); - INIT_WORK(&aq->work, &ieee80211softmac_auth_challenge_response, (void *)aq); - schedule_work(&aq->work); + INIT_DELAYED_WORK(&aq->work, &ieee80211softmac_auth_challenge_response); + schedule_delayed_work(&aq->work, 0); spin_unlock_irqrestore(&mac->lock, flags); return 0; case IEEE80211SOFTMAC_AUTH_SHARED_PASS: @@ -392,6 +395,6 @@ ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *de ieee80211softmac_deauth_from_net(mac, net); /* let's try to re-associate */ - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); return 0; } diff --git a/net/ieee80211/softmac/ieee80211softmac_event.c b/net/ieee80211/softmac/ieee80211softmac_event.c index f34fa2ef666..b9015656cfb 100644 --- a/net/ieee80211/softmac/ieee80211softmac_event.c +++ b/net/ieee80211/softmac/ieee80211softmac_event.c @@ -73,10 +73,12 @@ static char *event_descriptions[IEEE80211SOFTMAC_EVENT_LAST+1] = { static void -ieee80211softmac_notify_callback(void *d) +ieee80211softmac_notify_callback(struct work_struct *work) { - struct ieee80211softmac_event event = *(struct ieee80211softmac_event*) d; - kfree(d); + struct ieee80211softmac_event *pevent = + container_of(work, struct ieee80211softmac_event, work.work); + struct ieee80211softmac_event event = *pevent; + kfree(pevent); event.fun(event.mac->dev, event.event_type, event.context); } @@ -99,7 +101,7 @@ ieee80211softmac_notify_internal(struct ieee80211softmac_device *mac, return -ENOMEM; eventptr->event_type = event; - INIT_WORK(&eventptr->work, ieee80211softmac_notify_callback, eventptr); + INIT_DELAYED_WORK(&eventptr->work, ieee80211softmac_notify_callback); eventptr->fun = fun; eventptr->context = context; eventptr->mac = mac; @@ -170,7 +172,7 @@ ieee80211softmac_call_events_locked(struct ieee80211softmac_device *mac, int eve /* User may have subscribed to ANY event, so * we tell them which event triggered it. */ eventptr->event_type = event; - schedule_work(&eventptr->work); + schedule_delayed_work(&eventptr->work, 0); } } } diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c index 33aff4f4a47..256207b71dc 100644 --- a/net/ieee80211/softmac/ieee80211softmac_module.c +++ b/net/ieee80211/softmac/ieee80211softmac_module.c @@ -58,8 +58,8 @@ struct net_device *alloc_ieee80211softmac(int sizeof_priv) INIT_LIST_HEAD(&softmac->events); mutex_init(&softmac->associnfo.mutex); - INIT_WORK(&softmac->associnfo.work, ieee80211softmac_assoc_work, softmac); - INIT_WORK(&softmac->associnfo.timeout, ieee80211softmac_assoc_timeout, softmac); + INIT_DELAYED_WORK(&softmac->associnfo.work, ieee80211softmac_assoc_work); + INIT_DELAYED_WORK(&softmac->associnfo.timeout, ieee80211softmac_assoc_timeout); softmac->start_scan = ieee80211softmac_start_scan_implementation; softmac->wait_for_scan = ieee80211softmac_wait_for_scan_implementation; softmac->stop_scan = ieee80211softmac_stop_scan_implementation; diff --git a/net/ieee80211/softmac/ieee80211softmac_priv.h b/net/ieee80211/softmac/ieee80211softmac_priv.h index 0642e090b8a..c0dbe070e54 100644 --- a/net/ieee80211/softmac/ieee80211softmac_priv.h +++ b/net/ieee80211/softmac/ieee80211softmac_priv.h @@ -78,7 +78,7 @@ /* private definitions and prototypes */ /*** prototypes from _scan.c */ -void ieee80211softmac_scan(void *sm); +void ieee80211softmac_scan(struct work_struct *work); /* for internal use if scanning is needed */ int ieee80211softmac_start_scan(struct ieee80211softmac_device *mac); void ieee80211softmac_stop_scan(struct ieee80211softmac_device *mac); @@ -149,7 +149,7 @@ int ieee80211softmac_auth_resp(struct net_device *dev, struct ieee80211_auth *au int ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *deauth); /*** prototypes from _assoc.c */ -void ieee80211softmac_assoc_work(void *d); +void ieee80211softmac_assoc_work(struct work_struct *work); int ieee80211softmac_handle_assoc_response(struct net_device * dev, struct ieee80211_assoc_response * resp, struct ieee80211_network * network); @@ -157,7 +157,7 @@ int ieee80211softmac_handle_disassoc(struct net_device * dev, struct ieee80211_disassoc * disassoc); int ieee80211softmac_handle_reassoc_req(struct net_device * dev, struct ieee80211_reassoc_request * reassoc); -void ieee80211softmac_assoc_timeout(void *d); +void ieee80211softmac_assoc_timeout(struct work_struct *work); void ieee80211softmac_send_disassoc_req(struct ieee80211softmac_device *mac, u16 reason); void ieee80211softmac_disassoc(struct ieee80211softmac_device *mac); @@ -207,7 +207,7 @@ struct ieee80211softmac_auth_queue_item { struct ieee80211softmac_device *mac; /* SoftMAC device */ u8 retry; /* Retry limit */ u8 state; /* Auth State */ - struct work_struct work; /* Work queue */ + struct delayed_work work; /* Work queue */ }; /* scanning information */ @@ -219,7 +219,8 @@ struct ieee80211softmac_scaninfo { stop:1; u8 skip_flags; struct completion finished; - struct work_struct softmac_scan; + struct delayed_work softmac_scan; + struct ieee80211softmac_device *mac; }; /* private event struct */ @@ -227,7 +228,7 @@ struct ieee80211softmac_event { struct list_head list; int event_type; void *event_context; - struct work_struct work; + struct delayed_work work; notify_function_ptr fun; void *context; struct ieee80211softmac_device *mac; diff --git a/net/ieee80211/softmac/ieee80211softmac_scan.c b/net/ieee80211/softmac/ieee80211softmac_scan.c index d31cf77498c..a8326076581 100644 --- a/net/ieee80211/softmac/ieee80211softmac_scan.c +++ b/net/ieee80211/softmac/ieee80211softmac_scan.c @@ -91,12 +91,14 @@ ieee80211softmac_wait_for_scan(struct ieee80211softmac_device *sm) /* internal scanning implementation follows */ -void ieee80211softmac_scan(void *d) +void ieee80211softmac_scan(struct work_struct *work) { int invalid_channel; u8 current_channel_idx; - struct ieee80211softmac_device *sm = (struct ieee80211softmac_device *)d; - struct ieee80211softmac_scaninfo *si = sm->scaninfo; + struct ieee80211softmac_scaninfo *si = + container_of(work, struct ieee80211softmac_scaninfo, + softmac_scan.work); + struct ieee80211softmac_device *sm = si->mac; unsigned long flags; while (!(si->stop) && (si->current_channel_idx < si->number_channels)) { @@ -146,7 +148,8 @@ static inline struct ieee80211softmac_scaninfo *allocate_scaninfo(struct ieee802 struct ieee80211softmac_scaninfo *info = kmalloc(sizeof(struct ieee80211softmac_scaninfo), GFP_ATOMIC); if (unlikely(!info)) return NULL; - INIT_WORK(&info->softmac_scan, ieee80211softmac_scan, mac); + INIT_DELAYED_WORK(&info->softmac_scan, ieee80211softmac_scan); + info->mac = mac; init_completion(&info->finished); return info; } @@ -189,7 +192,7 @@ int ieee80211softmac_start_scan_implementation(struct net_device *dev) sm->scaninfo->started = 1; sm->scaninfo->stop = 0; INIT_COMPLETION(sm->scaninfo->finished); - schedule_work(&sm->scaninfo->softmac_scan); + schedule_delayed_work(&sm->scaninfo->softmac_scan, 0); spin_unlock_irqrestore(&sm->lock, flags); return 0; } diff --git a/net/ieee80211/softmac/ieee80211softmac_wx.c b/net/ieee80211/softmac/ieee80211softmac_wx.c index 23068a830f7..2ffaebd21c5 100644 --- a/net/ieee80211/softmac/ieee80211softmac_wx.c +++ b/net/ieee80211/softmac/ieee80211softmac_wx.c @@ -122,7 +122,7 @@ ieee80211softmac_wx_set_essid(struct net_device *net_dev, sm->associnfo.associating = 1; /* queue lower level code to do work (if necessary) */ - schedule_work(&sm->associnfo.work); + schedule_delayed_work(&sm->associnfo.work, 0); out: mutex_unlock(&sm->associnfo.mutex); @@ -356,7 +356,7 @@ ieee80211softmac_wx_set_wap(struct net_device *net_dev, /* force reassociation */ mac->associnfo.bssvalid = 0; if (mac->associnfo.associated) - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); } else if (is_zero_ether_addr(data->ap_addr.sa_data)) { /* the bssid we have is no longer fixed */ mac->associnfo.bssfixed = 0; @@ -373,7 +373,7 @@ ieee80211softmac_wx_set_wap(struct net_device *net_dev, /* tell the other code that this bssid should be used no matter what */ mac->associnfo.bssfixed = 1; /* queue associate if new bssid or (old one again and not associated) */ - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); } out: diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c index f261616e460..9b933381ebb 100644 --- a/net/ipv4/ipvs/ip_vs_ctl.c +++ b/net/ipv4/ipvs/ip_vs_ctl.c @@ -221,10 +221,10 @@ static void update_defense_level(void) * Timer for checking the defense */ #define DEFENSE_TIMER_PERIOD 1*HZ -static void defense_work_handler(void *data); -static DECLARE_WORK(defense_work, defense_work_handler, NULL); +static void defense_work_handler(struct work_struct *work); +static DECLARE_DELAYED_WORK(defense_work, defense_work_handler); -static void defense_work_handler(void *data) +static void defense_work_handler(struct work_struct *work) { update_defense_level(); if (atomic_read(&ip_vs_dropentry)) diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c index d50a02030ad..262bda808d9 100644 --- a/net/irda/ircomm/ircomm_tty.c +++ b/net/irda/ircomm/ircomm_tty.c @@ -61,7 +61,7 @@ static void ircomm_tty_flush_buffer(struct tty_struct *tty); static void ircomm_tty_send_xchar(struct tty_struct *tty, char ch); static void ircomm_tty_wait_until_sent(struct tty_struct *tty, int timeout); static void ircomm_tty_hangup(struct tty_struct *tty); -static void ircomm_tty_do_softint(void *private_); +static void ircomm_tty_do_softint(struct work_struct *work); static void ircomm_tty_shutdown(struct ircomm_tty_cb *self); static void ircomm_tty_stop(struct tty_struct *tty); @@ -389,7 +389,7 @@ static int ircomm_tty_open(struct tty_struct *tty, struct file *filp) self->flow = FLOW_STOP; self->line = line; - INIT_WORK(&self->tqueue, ircomm_tty_do_softint, self); + INIT_WORK(&self->tqueue, ircomm_tty_do_softint); self->max_header_size = IRCOMM_TTY_HDR_UNINITIALISED; self->max_data_size = IRCOMM_TTY_DATA_UNINITIALISED; self->close_delay = 5*HZ/10; @@ -594,15 +594,16 @@ static void ircomm_tty_flush_buffer(struct tty_struct *tty) } /* - * Function ircomm_tty_do_softint (private_) + * Function ircomm_tty_do_softint (work) * * We use this routine to give the write wakeup to the user at at a * safe time (as fast as possible after write have completed). This * can be compared to the Tx interrupt. */ -static void ircomm_tty_do_softint(void *private_) +static void ircomm_tty_do_softint(struct work_struct *work) { - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) private_; + struct ircomm_tty_cb *self = + container_of(work, struct ircomm_tty_cb, tqueue); struct tty_struct *tty; unsigned long flags; struct sk_buff *skb, *ctrl_skb; diff --git a/net/sctp/associola.c b/net/sctp/associola.c index ed0445fe85e..88124696ba6 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -61,7 +61,7 @@ #include /* Forward declarations for internal functions. */ -static void sctp_assoc_bh_rcv(struct sctp_association *asoc); +static void sctp_assoc_bh_rcv(struct work_struct *work); /* 1st Level Abstractions. */ @@ -269,9 +269,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a /* Create an input queue. */ sctp_inq_init(&asoc->base.inqueue); - sctp_inq_set_th_handler(&asoc->base.inqueue, - (void (*)(void *))sctp_assoc_bh_rcv, - asoc); + sctp_inq_set_th_handler(&asoc->base.inqueue, sctp_assoc_bh_rcv); /* Create an output queue. */ sctp_outq_init(asoc, &asoc->outqueue); @@ -944,8 +942,11 @@ out: } /* Do delayed input processing. This is scheduled by sctp_rcv(). */ -static void sctp_assoc_bh_rcv(struct sctp_association *asoc) +static void sctp_assoc_bh_rcv(struct work_struct *work) { + struct sctp_association *asoc = + container_of(work, struct sctp_association, + base.inqueue.immediate); struct sctp_endpoint *ep; struct sctp_chunk *chunk; struct sock *sk; diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 9b6b394b66f..a2b55372151 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -61,7 +61,7 @@ #include /* Forward declarations for internal helpers. */ -static void sctp_endpoint_bh_rcv(struct sctp_endpoint *ep); +static void sctp_endpoint_bh_rcv(struct work_struct *work); /* * Initialize the base fields of the endpoint structure. @@ -85,8 +85,7 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, sctp_inq_init(&ep->base.inqueue); /* Set its top-half handler */ - sctp_inq_set_th_handler(&ep->base.inqueue, - (void (*)(void *))sctp_endpoint_bh_rcv, ep); + sctp_inq_set_th_handler(&ep->base.inqueue, sctp_endpoint_bh_rcv); /* Initialize the bind addr area */ sctp_bind_addr_init(&ep->base.bind_addr, 0); @@ -311,8 +310,11 @@ int sctp_endpoint_is_peeled_off(struct sctp_endpoint *ep, /* Do delayed input processing. This is scheduled by sctp_rcv(). * This may be called on BH or task time. */ -static void sctp_endpoint_bh_rcv(struct sctp_endpoint *ep) +static void sctp_endpoint_bh_rcv(struct work_struct *work) { + struct sctp_endpoint *ep = + container_of(work, struct sctp_endpoint, + base.inqueue.immediate); struct sctp_association *asoc; struct sock *sk; struct sctp_transport *transport; diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index cf6deed7e84..71b07466e88 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -54,7 +54,7 @@ void sctp_inq_init(struct sctp_inq *queue) queue->in_progress = NULL; /* Create a task for delivering data. */ - INIT_WORK(&queue->immediate, NULL, NULL); + INIT_WORK(&queue->immediate, NULL); queue->malloced = 0; } @@ -97,7 +97,7 @@ void sctp_inq_push(struct sctp_inq *q, struct sctp_chunk *chunk) * on the BH related data structures. */ list_add_tail(&chunk->list, &q->in_chunk_list); - q->immediate.func(q->immediate.data); + q->immediate.func(&q->immediate); } /* Extract a chunk from an SCTP inqueue. @@ -205,9 +205,8 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) * The intent is that this routine will pull stuff out of the * inqueue and process it. */ -void sctp_inq_set_th_handler(struct sctp_inq *q, - void (*callback)(void *), void *arg) +void sctp_inq_set_th_handler(struct sctp_inq *q, work_func_t callback) { - INIT_WORK(&q->immediate, callback, arg); + INIT_WORK(&q->immediate, callback); } diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 7736b23c3f0..ba924d40df7 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -358,7 +358,7 @@ static void xfrm_policy_gc_kill(struct xfrm_policy *policy) xfrm_pol_put(policy); } -static void xfrm_policy_gc_task(void *data) +static void xfrm_policy_gc_task(struct work_struct *work) { struct xfrm_policy *policy; struct hlist_node *entry, *tmp; @@ -546,7 +546,7 @@ static inline int xfrm_byidx_should_resize(int total) static DEFINE_MUTEX(hash_resize_mutex); -static void xfrm_hash_resize(void *__unused) +static void xfrm_hash_resize(struct work_struct *__unused) { int dir, total; @@ -563,7 +563,7 @@ static void xfrm_hash_resize(void *__unused) mutex_unlock(&hash_resize_mutex); } -static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize, NULL); +static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize); /* Generate new index... KAME seems to generate them ordered by cost * of an absolute inpredictability of ordering of rules. This will not pass. */ @@ -2080,7 +2080,7 @@ static void __init xfrm_policy_init(void) panic("XFRM: failed to allocate bydst hash\n"); } - INIT_WORK(&xfrm_policy_gc_work, xfrm_policy_gc_task, NULL); + INIT_WORK(&xfrm_policy_gc_work, xfrm_policy_gc_task); register_netdevice_notifier(&xfrm_dev_notifier); } diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 899de9ed22a..40c52717984 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -115,7 +115,7 @@ static unsigned long xfrm_hash_new_size(void) static DEFINE_MUTEX(hash_resize_mutex); -static void xfrm_hash_resize(void *__unused) +static void xfrm_hash_resize(struct work_struct *__unused) { struct hlist_head *ndst, *nsrc, *nspi, *odst, *osrc, *ospi; unsigned long nsize, osize; @@ -168,7 +168,7 @@ out_unlock: mutex_unlock(&hash_resize_mutex); } -static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize, NULL); +static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize); DECLARE_WAIT_QUEUE_HEAD(km_waitq); EXPORT_SYMBOL(km_waitq); @@ -207,7 +207,7 @@ static void xfrm_state_gc_destroy(struct xfrm_state *x) kfree(x); } -static void xfrm_state_gc_task(void *data) +static void xfrm_state_gc_task(struct work_struct *data) { struct xfrm_state *x; struct hlist_node *entry, *tmp; @@ -1568,6 +1568,6 @@ void __init xfrm_state_init(void) panic("XFRM: Cannot allocate bydst/bysrc/byspi hashes."); xfrm_state_hmask = ((sz / sizeof(struct hlist_head)) - 1); - INIT_WORK(&xfrm_state_gc_work, xfrm_state_gc_task, NULL); + INIT_WORK(&xfrm_state_gc_work, xfrm_state_gc_task); } diff --git a/sound/aoa/aoa-gpio.h b/sound/aoa/aoa-gpio.h index 3a61f311557..ee64f5de896 100644 --- a/sound/aoa/aoa-gpio.h +++ b/sound/aoa/aoa-gpio.h @@ -59,10 +59,10 @@ struct gpio_methods { }; struct gpio_notification { + struct delayed_work work; notify_func_t notify; void *data; void *gpio_private; - struct work_struct work; struct mutex mutex; }; diff --git a/sound/aoa/core/snd-aoa-gpio-feature.c b/sound/aoa/core/snd-aoa-gpio-feature.c index 40eb47eccf9..2b03bc798bc 100644 --- a/sound/aoa/core/snd-aoa-gpio-feature.c +++ b/sound/aoa/core/snd-aoa-gpio-feature.c @@ -195,9 +195,10 @@ static void ftr_gpio_all_amps_restore(struct gpio_runtime *rt) ftr_gpio_set_lineout(rt, (s>>2)&1); } -static void ftr_handle_notify(void *data) +static void ftr_handle_notify(struct work_struct *work) { - struct gpio_notification *notif = data; + struct gpio_notification *notif = + container_of(work, struct gpio_notification, work.work); mutex_lock(¬if->mutex); if (notif->notify) @@ -253,12 +254,9 @@ static void ftr_gpio_init(struct gpio_runtime *rt) ftr_gpio_all_amps_off(rt); rt->implementation_private = 0; - INIT_WORK(&rt->headphone_notify.work, ftr_handle_notify, - &rt->headphone_notify); - INIT_WORK(&rt->line_in_notify.work, ftr_handle_notify, - &rt->line_in_notify); - INIT_WORK(&rt->line_out_notify.work, ftr_handle_notify, - &rt->line_out_notify); + INIT_DELAYED_WORK(&rt->headphone_notify.work, ftr_handle_notify); + INIT_DELAYED_WORK(&rt->line_in_notify.work, ftr_handle_notify); + INIT_DELAYED_WORK(&rt->line_out_notify.work, ftr_handle_notify); mutex_init(&rt->headphone_notify.mutex); mutex_init(&rt->line_in_notify.mutex); mutex_init(&rt->line_out_notify.mutex); @@ -287,7 +285,7 @@ static irqreturn_t ftr_handle_notify_irq(int xx, void *data) { struct gpio_notification *notif = data; - schedule_work(¬if->work); + schedule_delayed_work(¬if->work, 0); return IRQ_HANDLED; } diff --git a/sound/aoa/core/snd-aoa-gpio-pmf.c b/sound/aoa/core/snd-aoa-gpio-pmf.c index 2836c321839..5ca2220eac7 100644 --- a/sound/aoa/core/snd-aoa-gpio-pmf.c +++ b/sound/aoa/core/snd-aoa-gpio-pmf.c @@ -69,9 +69,10 @@ static void pmf_gpio_all_amps_restore(struct gpio_runtime *rt) pmf_gpio_set_lineout(rt, (s>>2)&1); } -static void pmf_handle_notify(void *data) +static void pmf_handle_notify(struct work_struct *work) { - struct gpio_notification *notif = data; + struct gpio_notification *notif = + container_of(work, struct gpio_notification, work.work); mutex_lock(¬if->mutex); if (notif->notify) @@ -83,12 +84,9 @@ static void pmf_gpio_init(struct gpio_runtime *rt) { pmf_gpio_all_amps_off(rt); rt->implementation_private = 0; - INIT_WORK(&rt->headphone_notify.work, pmf_handle_notify, - &rt->headphone_notify); - INIT_WORK(&rt->line_in_notify.work, pmf_handle_notify, - &rt->line_in_notify); - INIT_WORK(&rt->line_out_notify.work, pmf_handle_notify, - &rt->line_out_notify); + INIT_DELAYED_WORK(&rt->headphone_notify.work, pmf_handle_notify); + INIT_DELAYED_WORK(&rt->line_in_notify.work, pmf_handle_notify); + INIT_DELAYED_WORK(&rt->line_out_notify.work, pmf_handle_notify); mutex_init(&rt->headphone_notify.mutex); mutex_init(&rt->line_in_notify.mutex); mutex_init(&rt->line_out_notify.mutex); @@ -129,7 +127,7 @@ static void pmf_handle_notify_irq(void *data) { struct gpio_notification *notif = data; - schedule_work(¬if->work); + schedule_delayed_work(¬if->work, 0); } static int pmf_set_notify(struct gpio_runtime *rt, diff --git a/sound/i2c/other/ak4114.c b/sound/i2c/other/ak4114.c index 12ffffc9e81..d2f2c5078e6 100644 --- a/sound/i2c/other/ak4114.c +++ b/sound/i2c/other/ak4114.c @@ -35,7 +35,7 @@ MODULE_LICENSE("GPL"); #define AK4114_ADDR 0x00 /* fixed address */ -static void ak4114_stats(void *); +static void ak4114_stats(struct work_struct *work); static void reg_write(struct ak4114 *ak4114, unsigned char reg, unsigned char val) { @@ -158,7 +158,7 @@ void snd_ak4114_reinit(struct ak4114 *chip) reg_write(chip, AK4114_REG_PWRDN, old | AK4114_RST | AK4114_PWN); /* bring up statistics / event queing */ chip->init = 0; - INIT_WORK(&chip->work, ak4114_stats, chip); + INIT_DELAYED_WORK(&chip->work, ak4114_stats); queue_delayed_work(chip->workqueue, &chip->work, HZ / 10); } @@ -561,9 +561,9 @@ int snd_ak4114_check_rate_and_errors(struct ak4114 *ak4114, unsigned int flags) return res; } -static void ak4114_stats(void *data) +static void ak4114_stats(struct work_struct *work) { - struct ak4114 *chip = (struct ak4114 *)data; + struct ak4114 *chip = container_of(work, struct ak4114, work.work); if (chip->init) return; diff --git a/sound/pci/ac97/ac97_codec.c b/sound/pci/ac97/ac97_codec.c index 6577b232535..7abcb10b275 100644 --- a/sound/pci/ac97/ac97_codec.c +++ b/sound/pci/ac97/ac97_codec.c @@ -1927,9 +1927,10 @@ static int snd_ac97_dev_disconnect(struct snd_device *device) static struct snd_ac97_build_ops null_build_ops; #ifdef CONFIG_SND_AC97_POWER_SAVE -static void do_update_power(void *data) +static void do_update_power(struct work_struct *work) { - update_power_regs(data); + update_power_regs( + container_of(work, struct snd_ac97, power_work.work)); } #endif @@ -1989,7 +1990,7 @@ int snd_ac97_mixer(struct snd_ac97_bus *bus, struct snd_ac97_template *template, mutex_init(&ac97->page_mutex); #ifdef CONFIG_SND_AC97_POWER_SAVE ac97->power_workq = create_workqueue("ac97"); - INIT_WORK(&ac97->power_work, do_update_power, ac97); + INIT_DELAYED_WORK(&ac97->power_work, do_update_power); #endif #ifdef CONFIG_PCI diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c index 9c3d7ac0806..71482c15a85 100644 --- a/sound/pci/hda/hda_codec.c +++ b/sound/pci/hda/hda_codec.c @@ -272,10 +272,11 @@ EXPORT_SYMBOL(snd_hda_queue_unsol_event); /* * process queueud unsolicited events */ -static void process_unsol_events(void *data) +static void process_unsol_events(struct work_struct *work) { - struct hda_bus *bus = data; - struct hda_bus_unsolicited *unsol = bus->unsol; + struct hda_bus_unsolicited *unsol = + container_of(work, struct hda_bus_unsolicited, work); + struct hda_bus *bus = unsol->bus; struct hda_codec *codec; unsigned int rp, caddr, res; @@ -314,7 +315,8 @@ static int init_unsol_queue(struct hda_bus *bus) kfree(unsol); return -ENOMEM; } - INIT_WORK(&unsol->work, process_unsol_events, bus); + INIT_WORK(&unsol->work, process_unsol_events); + unsol->bus = bus; bus->unsol = unsol; return 0; } diff --git a/sound/pci/hda/hda_local.h b/sound/pci/hda/hda_local.h index f9416c36396..9ca1baf860b 100644 --- a/sound/pci/hda/hda_local.h +++ b/sound/pci/hda/hda_local.h @@ -206,6 +206,7 @@ struct hda_bus_unsolicited { /* workqueue */ struct workqueue_struct *workq; struct work_struct work; + struct hda_bus *bus; }; /* diff --git a/sound/ppc/tumbler.c b/sound/ppc/tumbler.c index 2fbe1d183fc..8f074c7936e 100644 --- a/sound/ppc/tumbler.c +++ b/sound/ppc/tumbler.c @@ -942,10 +942,11 @@ static void check_mute(struct snd_pmac *chip, struct pmac_gpio *gp, int val, int } static struct work_struct device_change; +static struct snd_pmac *device_change_chip; -static void device_change_handler(void *self) +static void device_change_handler(struct work_struct *work) { - struct snd_pmac *chip = self; + struct snd_pmac *chip = device_change_chip; struct pmac_tumbler *mix; int headphone, lineout; @@ -1417,7 +1418,8 @@ int __init snd_pmac_tumbler_init(struct snd_pmac *chip) chip->resume = tumbler_resume; #endif - INIT_WORK(&device_change, device_change_handler, (void *)chip); + INIT_WORK(&device_change, device_change_handler); + device_change_chip = chip; #ifdef PMAC_SUPPORT_AUTOMUTE if ((mix->headphone_irq >=0 || mix->lineout_irq >= 0) -- cgit v1.2.3-70-g09d2 From 368c73d4f689dae0807d0a2aa74c61fd2b9b075f Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 4 Oct 2006 00:41:26 +0100 Subject: PCI: quirks: fix the festering mess that claims to handle IDE quirks The number of permutations of crap we do is amazing and almost all of it has the wrong effect in 2.6. At the heart of this is the PCI SFF magic which says that compatibility mode PCI IDE controllers use ISA IRQ routing and hard coded addresses not the BAR values. The old quirks variously clears them, sets them, adjusts them and then IDE ignores the result. In order to drive all this garbage out and to do it portably we need to handle the SFF rules directly and properly. Because we know the device BAR 0-3 are not used in compatibility mode we load them with the values that are implied (and indeed which many controllers actually thoughtfully put there in this mode anyway). This removes special cases in the IDE layer and libata which now knows that bar 0/1/2/3 always contain the correct address. It means our resource allocation map is accurate from boot, not "mostly accurate" after ide is loaded, and it shoots lots of code. There is also lots more code and magic constant knowledge to shoot once this is in and settled. Been in my test tree for a while both with drivers/ide and with libata. Wants some -mm shakedown in case I've missed something dumb or there are corner cases lurking. Signed-off-by: Alan Cox Signed-off-by: Greg Kroah-Hartman --- arch/i386/pci/fixup.c | 46 --------------------------------------- drivers/pci/probe.c | 27 +++++++++++++++++++++++ drivers/pci/quirks.c | 59 +++------------------------------------------------ 3 files changed, 30 insertions(+), 102 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/pci/fixup.c b/arch/i386/pci/fixup.c index c1949ff38d6..cde1170b01a 100644 --- a/arch/i386/pci/fixup.c +++ b/arch/i386/pci/fixup.c @@ -74,52 +74,6 @@ static void __devinit pci_fixup_ncr53c810(struct pci_dev *d) } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NCR, PCI_DEVICE_ID_NCR_53C810, pci_fixup_ncr53c810); -static void __devinit pci_fixup_ide_bases(struct pci_dev *d) -{ - int i; - - /* - * PCI IDE controllers use non-standard I/O port decoding, respect it. - */ - if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE) - return; - DBG("PCI: IDE base address fixup for %s\n", pci_name(d)); - for(i=0; i<4; i++) { - struct resource *r = &d->resource[i]; - if ((r->start & ~0x80) == 0x374) { - r->start |= 2; - r->end = r->start; - } - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases); - -static void __devinit pci_fixup_ide_trash(struct pci_dev *d) -{ - int i; - - /* - * Runs the fixup only for the first IDE controller - * (Shai Fultheim - shai@ftcon.com) - */ - static int called = 0; - if (called) - return; - called = 1; - - /* - * There exist PCI IDE controllers which have utter garbage - * in first four base registers. Ignore that. - */ - DBG("PCI: IDE base address trash cleared for %s\n", pci_name(d)); - for(i=0; i<4; i++) - d->resource[i].start = d->resource[i].end = d->resource[i].flags = 0; -} -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_5513, pci_fixup_ide_trash); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_10, pci_fixup_ide_trash); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_11, pci_fixup_ide_trash); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801DB_9, pci_fixup_ide_trash); - static void __devinit pci_fixup_latency(struct pci_dev *d) { /* diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index e159d660449..0eeac60042b 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -679,6 +679,33 @@ static int pci_setup_device(struct pci_dev * dev) pci_read_bases(dev, 6, PCI_ROM_ADDRESS); pci_read_config_word(dev, PCI_SUBSYSTEM_VENDOR_ID, &dev->subsystem_vendor); pci_read_config_word(dev, PCI_SUBSYSTEM_ID, &dev->subsystem_device); + + /* + * Do the ugly legacy mode stuff here rather than broken chip + * quirk code. Legacy mode ATA controllers have fixed + * addresses. These are not always echoed in BAR0-3, and + * BAR0-3 in a few cases contain junk! + */ + if (class == PCI_CLASS_STORAGE_IDE) { + u8 progif; + pci_read_config_byte(dev, PCI_CLASS_PROG, &progif); + if ((progif & 1) == 0) { + dev->resource[0].start = 0x1F0; + dev->resource[0].end = 0x1F7; + dev->resource[0].flags = IORESOURCE_IO; + dev->resource[1].start = 0x3F6; + dev->resource[1].end = 0x3F6; + dev->resource[1].flags = IORESOURCE_IO; + } + if ((progif & 4) == 0) { + dev->resource[2].start = 0x170; + dev->resource[2].end = 0x177; + dev->resource[2].flags = IORESOURCE_IO; + dev->resource[3].start = 0x376; + dev->resource[3].end = 0x376; + dev->resource[3].flags = IORESOURCE_IO; + } + } break; case PCI_HEADER_TYPE_BRIDGE: /* bridge header */ diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 5b448381169..9ca9b9bf616 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -796,56 +796,6 @@ static void __init quirk_mediagx_master(struct pci_dev *dev) } DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_PCI_MASTER, quirk_mediagx_master ); -/* - * As per PCI spec, ignore base address registers 0-3 of the IDE controllers - * running in Compatible mode (bits 0 and 2 in the ProgIf for primary and - * secondary channels respectively). If the device reports Compatible mode - * but does use BAR0-3 for address decoding, we assume that firmware has - * programmed these BARs with standard values (0x1f0,0x3f4 and 0x170,0x374). - * Exceptions (if they exist) must be handled in chip/architecture specific - * fixups. - * - * Note: for non x86 people. You may need an arch specific quirk to handle - * moving IDE devices to native mode as well. Some plug in card devices power - * up in compatible mode and assume the BIOS will adjust them. - * - * Q: should we load the 0x1f0,0x3f4 into the registers or zap them as - * we do now ? We don't want is pci_enable_device to come along - * and assign new resources. Both approaches work for that. - */ -static void __devinit quirk_ide_bases(struct pci_dev *dev) -{ - struct resource *res; - int first_bar = 2, last_bar = 0; - - if ((dev->class >> 8) != PCI_CLASS_STORAGE_IDE) - return; - - res = &dev->resource[0]; - - /* primary channel: ProgIf bit 0, BAR0, BAR1 */ - if (!(dev->class & 1) && (res[0].flags || res[1].flags)) { - res[0].start = res[0].end = res[0].flags = 0; - res[1].start = res[1].end = res[1].flags = 0; - first_bar = 0; - last_bar = 1; - } - - /* secondary channel: ProgIf bit 2, BAR2, BAR3 */ - if (!(dev->class & 4) && (res[2].flags || res[3].flags)) { - res[2].start = res[2].end = res[2].flags = 0; - res[3].start = res[3].end = res[3].flags = 0; - last_bar = 3; - } - - if (!last_bar) - return; - - printk(KERN_INFO "PCI: Ignoring BAR%d-%d of IDE controller %s\n", - first_bar, last_bar, pci_name(dev)); -} -DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, quirk_ide_bases); - /* * Ensure C0 rev restreaming is off. This is normally done by * the BIOS but in the odd case it is not the results are corruption @@ -880,11 +830,10 @@ static void __devinit quirk_svwks_csb5ide(struct pci_dev *pdev) prog &= ~5; pdev->class &= ~5; pci_write_config_byte(pdev, PCI_CLASS_PROG, prog); - /* need to re-assign BARs for compat mode */ - quirk_ide_bases(pdev); + /* PCI layer will sort out resources */ } } -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_CSB5IDE, quirk_svwks_csb5ide ); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_CSB5IDE, quirk_svwks_csb5ide ); /* * Intel 82801CAM ICH3-M datasheet says IDE modes must be the same @@ -900,11 +849,9 @@ static void __init quirk_ide_samemode(struct pci_dev *pdev) prog &= ~5; pdev->class &= ~5; pci_write_config_byte(pdev, PCI_CLASS_PROG, prog); - /* need to re-assign BARs for compat mode */ - quirk_ide_bases(pdev); } } -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_10, quirk_ide_samemode); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_10, quirk_ide_samemode); /* This was originally an Alpha specific thing, but it really fits here. * The i82375 PCI/EISA bridge appears as non-classified. Fix that. -- cgit v1.2.3-70-g09d2 From 7edab2f0876ff6a38e10a88c9aca20180aad307c Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 17 Oct 2006 10:17:58 -0700 Subject: pci/i386: style cleanups Mostly CodingStyle cleanups for arch/i386/pci/i386.c: - fit in 80 columns; - use a #defined value instead of an inline constant; Also change one resource_size_t (DBG) printk from %08lx to %lx since it can be more than 32 bits (more than 8 hexits). Signed-off-by: Randy Dunlap Signed-off-by: Greg Kroah-Hartman --- arch/i386/pci/i386.c | 64 +++++++++++++++++++++++++++++++++++----------------- 1 file changed, 43 insertions(+), 21 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/pci/i386.c b/arch/i386/pci/i386.c index 98580292f0d..43005f04442 100644 --- a/arch/i386/pci/i386.c +++ b/arch/i386/pci/i386.c @@ -104,16 +104,24 @@ static void __init pcibios_allocate_bus_resources(struct list_head *bus_list) /* Depth-First Search on bus tree */ list_for_each_entry(bus, bus_list, node) { if ((dev = bus->self)) { - for (idx = PCI_BRIDGE_RESOURCES; idx < PCI_NUM_RESOURCES; idx++) { + for (idx = PCI_BRIDGE_RESOURCES; + idx < PCI_NUM_RESOURCES; idx++) { r = &dev->resource[idx]; if (!r->flags) continue; pr = pci_find_parent_resource(dev, r); - if (!r->start || !pr || request_resource(pr, r) < 0) { - printk(KERN_ERR "PCI: Cannot allocate resource region %d of bridge %s\n", idx, pci_name(dev)); - /* Something is wrong with the region. - Invalidate the resource to prevent child - resource allocations in this range. */ + if (!r->start || !pr || + request_resource(pr, r) < 0) { + printk(KERN_ERR "PCI: Cannot allocate " + "resource region %d " + "of bridge %s\n", + idx, pci_name(dev)); + /* + * Something is wrong with the region. + * Invalidate the resource to prevent + * child resource allocations in this + * range. + */ r->flags = 0; } } @@ -131,7 +139,7 @@ static void __init pcibios_allocate_resources(int pass) for_each_pci_dev(dev) { pci_read_config_word(dev, PCI_COMMAND, &command); - for(idx = 0; idx < 6; idx++) { + for (idx = 0; idx < PCI_ROM_RESOURCE; idx++) { r = &dev->resource[idx]; if (r->parent) /* Already allocated */ continue; @@ -142,11 +150,15 @@ static void __init pcibios_allocate_resources(int pass) else disabled = !(command & PCI_COMMAND_MEMORY); if (pass == disabled) { - DBG("PCI: Resource %08lx-%08lx (f=%lx, d=%d, p=%d)\n", + DBG("PCI: Resource %08lx-%08lx " + "(f=%lx, d=%d, p=%d)\n", r->start, r->end, r->flags, disabled, pass); pr = pci_find_parent_resource(dev, r); if (!pr || request_resource(pr, r) < 0) { - printk(KERN_ERR "PCI: Cannot allocate resource region %d of device %s\n", idx, pci_name(dev)); + printk(KERN_ERR "PCI: Cannot allocate " + "resource region %d " + "of device %s\n", + idx, pci_name(dev)); /* We'll assign a new address later */ r->end -= r->start; r->start = 0; @@ -156,12 +168,16 @@ static void __init pcibios_allocate_resources(int pass) if (!pass) { r = &dev->resource[PCI_ROM_RESOURCE]; if (r->flags & IORESOURCE_ROM_ENABLE) { - /* Turn the ROM off, leave the resource region, but keep it unregistered. */ + /* Turn the ROM off, leave the resource region, + * but keep it unregistered. */ u32 reg; - DBG("PCI: Switching off ROM of %s\n", pci_name(dev)); + DBG("PCI: Switching off ROM of %s\n", + pci_name(dev)); r->flags &= ~IORESOURCE_ROM_ENABLE; - pci_read_config_dword(dev, dev->rom_base_reg, ®); - pci_write_config_dword(dev, dev->rom_base_reg, reg & ~PCI_ROM_ADDRESS_ENABLE); + pci_read_config_dword(dev, + dev->rom_base_reg, ®); + pci_write_config_dword(dev, dev->rom_base_reg, + reg & ~PCI_ROM_ADDRESS_ENABLE); } } } @@ -173,9 +189,11 @@ static int __init pcibios_assign_resources(void) struct resource *r, *pr; if (!(pci_probe & PCI_ASSIGN_ROMS)) { - /* Try to use BIOS settings for ROMs, otherwise let - pci_assign_unassigned_resources() allocate the new - addresses. */ + /* + * Try to use BIOS settings for ROMs, otherwise let + * pci_assign_unassigned_resources() allocate the new + * addresses. + */ for_each_pci_dev(dev) { r = &dev->resource[PCI_ROM_RESOURCE]; if (!r->flags || !r->start) @@ -215,9 +233,9 @@ int pcibios_enable_resources(struct pci_dev *dev, int mask) pci_read_config_word(dev, PCI_COMMAND, &cmd); old_cmd = cmd; - for(idx = 0; idx < PCI_NUM_RESOURCES; idx++) { + for (idx = 0; idx < PCI_NUM_RESOURCES; idx++) { /* Only set up the requested stuff */ - if (!(mask & (1<resource[idx]; @@ -227,7 +245,9 @@ int pcibios_enable_resources(struct pci_dev *dev, int mask) (!(r->flags & IORESOURCE_ROM_ENABLE))) continue; if (!r->start && r->end) { - printk(KERN_ERR "PCI: Device %s not available because of resource collisions\n", pci_name(dev)); + printk(KERN_ERR "PCI: Device %s not available " + "because of resource collisions\n", + pci_name(dev)); return -EINVAL; } if (r->flags & IORESOURCE_IO) @@ -236,7 +256,8 @@ int pcibios_enable_resources(struct pci_dev *dev, int mask) cmd |= PCI_COMMAND_MEMORY; } if (cmd != old_cmd) { - printk("PCI: Enabling device %s (%04x -> %04x)\n", pci_name(dev), old_cmd, cmd); + printk("PCI: Enabling device %s (%04x -> %04x)\n", + pci_name(dev), old_cmd, cmd); pci_write_config_word(dev, PCI_COMMAND, cmd); } return 0; @@ -258,7 +279,8 @@ void pcibios_set_master(struct pci_dev *dev) lat = pcibios_max_latency; else return; - printk(KERN_DEBUG "PCI: Setting latency timer of device %s to %d\n", pci_name(dev), lat); + printk(KERN_DEBUG "PCI: Setting latency timer of device %s to %d\n", + pci_name(dev), lat); pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat); } -- cgit v1.2.3-70-g09d2 From 039d09a845209122c5193e650ab2d8b3c849ca7c Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Tue, 17 Oct 2006 10:02:50 +0530 Subject: PCI: arch/i386/kernel/pci-dma.c: ioremap balanced with iounmap ioremap must be balanced by an iounmap and failing to do so can result in a memory leak. Tested (compilation only): - using allmodconfig - making sure the files are compiling without any warning/error due to new changes Signed-off-by: Amol Lad Signed-off-by: Greg Kroah-Hartman --- arch/i386/kernel/pci-dma.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/pci-dma.c b/arch/i386/kernel/pci-dma.c index 25fe6685393..5c8c6ef1fc5 100644 --- a/arch/i386/kernel/pci-dma.c +++ b/arch/i386/kernel/pci-dma.c @@ -75,7 +75,7 @@ EXPORT_SYMBOL(dma_free_coherent); int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr, dma_addr_t device_addr, size_t size, int flags) { - void __iomem *mem_base; + void __iomem *mem_base = NULL; int pages = size >> PAGE_SHIFT; int bitmap_size = (pages + 31)/32; @@ -114,6 +114,8 @@ int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr, free1_out: kfree(dev->dma_mem->bitmap); out: + if (mem_base) + iounmap(mem_base); return 0; } EXPORT_SYMBOL(dma_declare_coherent_memory); -- cgit v1.2.3-70-g09d2 From 3b59d52d8c7925e7a9a396f2e31a66eb060c6c37 Mon Sep 17 00:00:00 2001 From: Jason Gaston Date: Wed, 22 Nov 2006 15:15:08 -0800 Subject: PCI: irq: irq and pci_ids patch for Intel ICH9 This updated patch adds the Intel ICH9 LPC and SMBus Controller DID's. Signed-off-by: Jason Gaston Signed-off-by: Greg Kroah-Hartman --- arch/i386/pci/irq.c | 6 ++++++ include/linux/pci_ids.h | 7 +++++++ 2 files changed, 13 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c index 69163998ade..e65551cd821 100644 --- a/arch/i386/pci/irq.c +++ b/arch/i386/pci/irq.c @@ -543,6 +543,12 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH8_2: case PCI_DEVICE_ID_INTEL_ICH8_3: case PCI_DEVICE_ID_INTEL_ICH8_4: + case PCI_DEVICE_ID_INTEL_ICH9_0: + case PCI_DEVICE_ID_INTEL_ICH9_1: + case PCI_DEVICE_ID_INTEL_ICH9_2: + case PCI_DEVICE_ID_INTEL_ICH9_3: + case PCI_DEVICE_ID_INTEL_ICH9_4: + case PCI_DEVICE_ID_INTEL_ICH9_5: r->name = "PIIX/ICH"; r->get = pirq_piix_get; r->set = pirq_piix_set; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index fa4e1d79978..e060a763794 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2211,6 +2211,13 @@ #define PCI_DEVICE_ID_INTEL_ICH8_4 0x2815 #define PCI_DEVICE_ID_INTEL_ICH8_5 0x283e #define PCI_DEVICE_ID_INTEL_ICH8_6 0x2850 +#define PCI_DEVICE_ID_INTEL_ICH9_0 0x2910 +#define PCI_DEVICE_ID_INTEL_ICH9_1 0x2911 +#define PCI_DEVICE_ID_INTEL_ICH9_2 0x2912 +#define PCI_DEVICE_ID_INTEL_ICH9_3 0x2913 +#define PCI_DEVICE_ID_INTEL_ICH9_4 0x2914 +#define PCI_DEVICE_ID_INTEL_ICH9_5 0x2915 +#define PCI_DEVICE_ID_INTEL_ICH9_6 0x2930 #define PCI_DEVICE_ID_INTEL_82855PM_HB 0x3340 #define PCI_DEVICE_ID_INTEL_82830_HB 0x3575 #define PCI_DEVICE_ID_INTEL_82830_CGC 0x3577 -- cgit v1.2.3-70-g09d2 From 2b290da053608692ea206507d993b70c39d2cdea Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 16 Nov 2006 13:16:23 +0100 Subject: PCI: make arch/i386/pci/common.c:pci_bf_sort static This patch makes the needlessly global pci_bf_sort static. Signed-off-by: Adrian Bunk Acked-by: Matt Domsch Signed-off-by: Greg Kroah-Hartman --- arch/i386/pci/common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c index cdfcf971098..53ca6e89798 100644 --- a/arch/i386/pci/common.c +++ b/arch/i386/pci/common.c @@ -20,7 +20,7 @@ unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 | PCI_PROBE_MMCONF; -int pci_bf_sort; +static int pci_bf_sort; int pci_routeirq; int pcibios_last_bus = -1; unsigned long pirq_table_addr; -- cgit v1.2.3-70-g09d2 From a271aaf15f492d202def1b34e447107b60fe4ece Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 7 Aug 2006 22:19:37 -0700 Subject: Driver core: convert msr code to use struct device Converts from using struct "class_device" to "struct device" making everything show up properly in /sys/devices/ with symlinks from the /sys/class directory. Signed-off-by: Greg Kroah-Hartman --- arch/i386/kernel/msr.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/msr.c b/arch/i386/kernel/msr.c index d535cdbbfd2..a773f776c9e 100644 --- a/arch/i386/kernel/msr.c +++ b/arch/i386/kernel/msr.c @@ -239,14 +239,14 @@ static struct file_operations msr_fops = { .open = msr_open, }; -static int msr_class_device_create(int i) +static int msr_device_create(int i) { int err = 0; - struct class_device *class_err; + struct device *dev; - class_err = class_device_create(msr_class, NULL, MKDEV(MSR_MAJOR, i), NULL, "msr%d",i); - if (IS_ERR(class_err)) - err = PTR_ERR(class_err); + dev = device_create(msr_class, NULL, MKDEV(MSR_MAJOR, i), "msr%d",i); + if (IS_ERR(dev)) + err = PTR_ERR(dev); return err; } @@ -258,10 +258,10 @@ static int msr_class_cpu_callback(struct notifier_block *nfb, switch (action) { case CPU_ONLINE: - msr_class_device_create(cpu); + msr_device_create(cpu); break; case CPU_DEAD: - class_device_destroy(msr_class, MKDEV(MSR_MAJOR, cpu)); + device_destroy(msr_class, MKDEV(MSR_MAJOR, cpu)); break; } return NOTIFY_OK; @@ -290,7 +290,7 @@ static int __init msr_init(void) goto out_chrdev; } for_each_online_cpu(i) { - err = msr_class_device_create(i); + err = msr_device_create(i); if (err != 0) goto out_class; } @@ -302,7 +302,7 @@ static int __init msr_init(void) out_class: i = 0; for_each_online_cpu(i) - class_device_destroy(msr_class, MKDEV(MSR_MAJOR, i)); + device_destroy(msr_class, MKDEV(MSR_MAJOR, i)); class_destroy(msr_class); out_chrdev: unregister_chrdev(MSR_MAJOR, "cpu/msr"); @@ -314,7 +314,7 @@ static void __exit msr_exit(void) { int cpu = 0; for_each_online_cpu(cpu) - class_device_destroy(msr_class, MKDEV(MSR_MAJOR, cpu)); + device_destroy(msr_class, MKDEV(MSR_MAJOR, cpu)); class_destroy(msr_class); unregister_chrdev(MSR_MAJOR, "cpu/msr"); unregister_hotcpu_notifier(&msr_class_cpu_notifier); -- cgit v1.2.3-70-g09d2 From 07accdc18e014cad945013886ee34e09f859f88a Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 7 Aug 2006 22:19:37 -0700 Subject: Driver core: convert cpuid code to use struct device Converts from using struct "class_device" to "struct device" making everything show up properly in /sys/devices/ with symlinks from the /sys/class directory. Signed-off-by: Greg Kroah-Hartman --- arch/i386/kernel/cpuid.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpuid.c b/arch/i386/kernel/cpuid.c index fde8bea85ce..ab0c327e79d 100644 --- a/arch/i386/kernel/cpuid.c +++ b/arch/i386/kernel/cpuid.c @@ -156,14 +156,14 @@ static struct file_operations cpuid_fops = { .open = cpuid_open, }; -static int cpuid_class_device_create(int i) +static int cpuid_device_create(int i) { int err = 0; - struct class_device *class_err; + struct device *dev; - class_err = class_device_create(cpuid_class, NULL, MKDEV(CPUID_MAJOR, i), NULL, "cpu%d",i); - if (IS_ERR(class_err)) - err = PTR_ERR(class_err); + dev = device_create(cpuid_class, NULL, MKDEV(CPUID_MAJOR, i), "cpu%d",i); + if (IS_ERR(dev)) + err = PTR_ERR(dev); return err; } @@ -174,10 +174,10 @@ static int cpuid_class_cpu_callback(struct notifier_block *nfb, unsigned long ac switch (action) { case CPU_ONLINE: - cpuid_class_device_create(cpu); + cpuid_device_create(cpu); break; case CPU_DEAD: - class_device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, cpu)); + device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, cpu)); break; } return NOTIFY_OK; @@ -206,7 +206,7 @@ static int __init cpuid_init(void) goto out_chrdev; } for_each_online_cpu(i) { - err = cpuid_class_device_create(i); + err = cpuid_device_create(i); if (err != 0) goto out_class; } @@ -218,7 +218,7 @@ static int __init cpuid_init(void) out_class: i = 0; for_each_online_cpu(i) { - class_device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, i)); + device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, i)); } class_destroy(cpuid_class); out_chrdev: @@ -232,7 +232,7 @@ static void __exit cpuid_exit(void) int cpu = 0; for_each_online_cpu(cpu) - class_device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, cpu)); + device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, cpu)); class_destroy(cpuid_class); unregister_chrdev(CPUID_MAJOR, "cpu/cpuid"); unregister_hotcpu_notifier(&cpuid_class_cpu_notifier); -- cgit v1.2.3-70-g09d2 From 7bdd21cef9e5dbc3d3a718c55bb3d0da024644da Mon Sep 17 00:00:00 2001 From: Len Brown Date: Sat, 2 Dec 2006 02:27:46 -0500 Subject: Revert "ACPI: SCI interrupt source override" This reverts commit 281ea49b0c294649a6de47a6f8fbe5611137726b, which broke ACPI Interrupt source overrides that move the SCI from one IRQ in PIC mode to another in IOAPIC mode. If the SCI shared an interrupt line with another device, this would result in a "irq 18: nobody cared" type failure. http://bugzilla.kernel.org/show_bug.cgi?id=7601 Signed-off-by: Len Brown --- arch/i386/kernel/acpi/boot.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c index d12fb97a533..c8f96cff07c 100644 --- a/arch/i386/kernel/acpi/boot.c +++ b/arch/i386/kernel/acpi/boot.c @@ -333,7 +333,7 @@ acpi_parse_ioapic(acpi_table_entry_header * header, const unsigned long end) /* * Parse Interrupt Source Override for the ACPI SCI */ -static void acpi_sci_ioapic_setup(u32 bus_irq, u32 gsi, u16 polarity, u16 trigger) +static void acpi_sci_ioapic_setup(u32 gsi, u16 polarity, u16 trigger) { if (trigger == 0) /* compatible SCI trigger is level */ trigger = 3; @@ -353,13 +353,13 @@ static void acpi_sci_ioapic_setup(u32 bus_irq, u32 gsi, u16 polarity, u16 trigge * If GSI is < 16, this will update its flags, * else it will create a new mp_irqs[] entry. */ - mp_override_legacy_irq(bus_irq, polarity, trigger, gsi); + mp_override_legacy_irq(gsi, polarity, trigger, gsi); /* * stash over-ride to indicate we've been here * and for later update of acpi_fadt */ - acpi_sci_override_gsi = bus_irq; + acpi_sci_override_gsi = gsi; return; } @@ -377,7 +377,7 @@ acpi_parse_int_src_ovr(acpi_table_entry_header * header, acpi_table_print_madt_entry(header); if (intsrc->bus_irq == acpi_fadt.sci_int) { - acpi_sci_ioapic_setup(intsrc->bus_irq, intsrc->global_irq, + acpi_sci_ioapic_setup(intsrc->global_irq, intsrc->flags.polarity, intsrc->flags.trigger); return 0; @@ -880,7 +880,7 @@ static int __init acpi_parse_madt_ioapic_entries(void) * pretend we got one so we can set the SCI flags. */ if (!acpi_sci_override_gsi) - acpi_sci_ioapic_setup(acpi_fadt.sci_int, acpi_fadt.sci_int, 0, 0); + acpi_sci_ioapic_setup(acpi_fadt.sci_int, 0, 0); /* Fill in identity legacy mapings where no override */ mp_config_acpi_legacy_irqs(); -- cgit v1.2.3-70-g09d2 From f6a570333e554b48ad589e7137c77c57809eee81 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 18 Oct 2006 01:47:25 -0400 Subject: [PATCH] severing module.h->sched.h Signed-off-by: Al Viro --- arch/i386/kernel/alternative.c | 1 + arch/i386/kernel/cpu/mcheck/therm_throt.c | 1 + drivers/base/cpu.c | 1 + drivers/hwmon/abituguru.c | 1 + drivers/leds/ledtrig-ide-disk.c | 1 + drivers/leds/ledtrig-timer.c | 1 + drivers/pci/access.c | 1 + drivers/scsi/scsi_transport_sas.c | 1 + drivers/w1/slaves/w1_therm.c | 1 + include/asm-x86_64/elf.h | 1 - include/linux/acct.h | 1 + include/linux/module.h | 13 +------------ include/scsi/libiscsi.h | 2 ++ kernel/latency.c | 1 + kernel/module.c | 15 ++++++++++++++- lib/random32.c | 1 + 16 files changed, 29 insertions(+), 14 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c index 583c238e17f..535f9794fba 100644 --- a/arch/i386/kernel/alternative.c +++ b/arch/i386/kernel/alternative.c @@ -1,4 +1,5 @@ #include +#include #include #include #include diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c index 2d8703b7ce6..bad8b442070 100644 --- a/arch/i386/kernel/cpu/mcheck/therm_throt.c +++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c @@ -20,6 +20,7 @@ #include #include #include +#include #include /* How long to wait between reporting thermal events */ diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 4bef76a2f3f..1f745f12f94 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/hwmon/abituguru.c b/drivers/hwmon/abituguru.c index e5cb0fdab9b..b1dc63e4ac7 100644 --- a/drivers/hwmon/abituguru.c +++ b/drivers/hwmon/abituguru.c @@ -21,6 +21,7 @@ etc voltage & frequency control is not supported! */ #include +#include #include #include #include diff --git a/drivers/leds/ledtrig-ide-disk.c b/drivers/leds/ledtrig-ide-disk.c index fa651886ab4..54b155c7026 100644 --- a/drivers/leds/ledtrig-ide-disk.c +++ b/drivers/leds/ledtrig-ide-disk.c @@ -12,6 +12,7 @@ */ #include +#include #include #include #include diff --git a/drivers/leds/ledtrig-timer.c b/drivers/leds/ledtrig-timer.c index 29a8818a32e..d756bdb01c5 100644 --- a/drivers/leds/ledtrig-timer.c +++ b/drivers/leds/ledtrig-timer.c @@ -12,6 +12,7 @@ */ #include +#include #include #include #include diff --git a/drivers/pci/access.c b/drivers/pci/access.c index 73a58c73d52..fc405f0165d 100644 --- a/drivers/pci/access.c +++ b/drivers/pci/access.c @@ -1,5 +1,6 @@ #include #include +#include #include #include diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c index b5b0c2cba96..5c0b75bbfa1 100644 --- a/drivers/scsi/scsi_transport_sas.c +++ b/drivers/scsi/scsi_transport_sas.c @@ -25,6 +25,7 @@ #include #include +#include #include #include #include diff --git a/drivers/w1/slaves/w1_therm.c b/drivers/w1/slaves/w1_therm.c index 5372cfcbd05..b022fffd8c5 100644 --- a/drivers/w1/slaves/w1_therm.c +++ b/drivers/w1/slaves/w1_therm.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include diff --git a/include/asm-x86_64/elf.h b/include/asm-x86_64/elf.h index a406fcb1e92..6d24ea7c4d9 100644 --- a/include/asm-x86_64/elf.h +++ b/include/asm-x86_64/elf.h @@ -45,7 +45,6 @@ typedef struct user_i387_struct elf_fpregset_t; #ifdef __KERNEL__ #include -#include /* * This is used to ensure we don't load something for the wrong architecture. diff --git a/include/linux/acct.h b/include/linux/acct.h index 0496d1f0995..302eb727ecb 100644 --- a/include/linux/acct.h +++ b/include/linux/acct.h @@ -119,6 +119,7 @@ struct acct_v3 #ifdef CONFIG_BSD_PROCESS_ACCT struct vfsmount; struct super_block; +struct pacct_struct; extern void acct_auto_close_mnt(struct vfsmount *m); extern void acct_auto_close(struct super_block *sb); extern void acct_init_pacct(struct pacct_struct *pacct); diff --git a/include/linux/module.h b/include/linux/module.h index 9258ffd8a7f..d33df2408e0 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -6,7 +6,6 @@ * Rewritten by Richard Henderson Dec 1996 * Rewritten again by Rusty Russell, 2002 */ -#include #include #include #include @@ -411,17 +410,7 @@ static inline int try_module_get(struct module *module) return ret; } -static inline void module_put(struct module *module) -{ - if (module) { - unsigned int cpu = get_cpu(); - local_dec(&module->ref[cpu].count); - /* Maybe they're waiting for us to drop reference? */ - if (unlikely(!module_is_live(module))) - wake_up_process(module->waiter); - put_cpu(); - } -} +extern void module_put(struct module *module); #else /*!CONFIG_MODULE_UNLOAD*/ static inline int try_module_get(struct module *module) diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h index 61eebec00a7..ea0816d4904 100644 --- a/include/scsi/libiscsi.h +++ b/include/scsi/libiscsi.h @@ -25,6 +25,8 @@ #include #include +#include +#include #include #include diff --git a/kernel/latency.c b/kernel/latency.c index 258f2555abb..e63fcacb61a 100644 --- a/kernel/latency.c +++ b/kernel/latency.c @@ -36,6 +36,7 @@ #include #include #include +#include #include struct latency_info { diff --git a/kernel/module.c b/kernel/module.c index 45e01cb6010..e2d09d604ca 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -34,10 +34,10 @@ #include #include #include +#include #include #include #include -#include #include #include #include @@ -790,6 +790,19 @@ static struct module_attribute refcnt = { .show = show_refcnt, }; +void module_put(struct module *module) +{ + if (module) { + unsigned int cpu = get_cpu(); + local_dec(&module->ref[cpu].count); + /* Maybe they're waiting for us to drop reference? */ + if (unlikely(!module_is_live(module))) + wake_up_process(module->waiter); + put_cpu(); + } +} +EXPORT_SYMBOL(module_put); + #else /* !CONFIG_MODULE_UNLOAD */ static void print_unload_info(struct seq_file *m, struct module *mod) { diff --git a/lib/random32.c b/lib/random32.c index 4a15ce51cea..ec7f81d3fb1 100644 --- a/lib/random32.c +++ b/lib/random32.c @@ -36,6 +36,7 @@ #include #include #include +#include #include struct rnd_state { -- cgit v1.2.3-70-g09d2 From 914e26379decf1fd984b22e51fd2e4209b7a7f1b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 18 Oct 2006 13:55:46 -0400 Subject: [PATCH] severing fs.h, radix-tree.h -> sched.h Signed-off-by: Al Viro --- arch/i386/kernel/acpi/cstate.c | 1 + drivers/acpi/dock.c | 1 + drivers/char/hw_random/core.c | 1 + drivers/char/tpm/tpm.h | 1 + drivers/hwmon/hdaps.c | 1 + drivers/spi/spi_butterfly.c | 1 + fs/9p/conv.c | 1 + fs/9p/fcall.c | 1 + fs/9p/fid.c | 1 + fs/9p/v9fs.c | 1 + fs/9p/vfs_dir.c | 1 + fs/9p/vfs_file.c | 1 + fs/inotify.c | 1 + fs/jffs2/acl.c | 1 + fs/jffs2/wbuf.c | 1 + fs/jfs/ioctl.c | 1 + fs/proc/root.c | 1 + fs/super.c | 18 ++++++++++++++++++ fs/sync.c | 1 + fs/utimes.c | 1 + include/linux/fs.h | 37 +++++++------------------------------ include/linux/radix-tree.h | 1 - 22 files changed, 44 insertions(+), 31 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/acpi/cstate.c b/arch/i386/kernel/acpi/cstate.c index 20563e52c62..4664b55f623 100644 --- a/arch/i386/kernel/acpi/cstate.c +++ b/arch/i386/kernel/acpi/cstate.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c index 578b99b71d9..bf5b79ed361 100644 --- a/drivers/acpi/dock.c +++ b/drivers/acpi/dock.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include diff --git a/drivers/char/hw_random/core.c b/drivers/char/hw_random/core.c index ebace201bec..26a860adcb3 100644 --- a/drivers/char/hw_random/core.c +++ b/drivers/char/hw_random/core.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/char/tpm/tpm.h b/drivers/char/tpm/tpm.h index 050ced247f6..bb9a43c6cf3 100644 --- a/drivers/char/tpm/tpm.h +++ b/drivers/char/tpm/tpm.h @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/hwmon/hdaps.c b/drivers/hwmon/hdaps.c index 26be4ea8a38..e8ef62b83d6 100644 --- a/drivers/hwmon/hdaps.c +++ b/drivers/hwmon/hdaps.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #define HDAPS_LOW_PORT 0x1600 /* first port used by hdaps */ diff --git a/drivers/spi/spi_butterfly.c b/drivers/spi/spi_butterfly.c index 39d9b20f203..c2f601f8e4f 100644 --- a/drivers/spi/spi_butterfly.c +++ b/drivers/spi/spi_butterfly.c @@ -23,6 +23,7 @@ #include #include +#include #include #include #include diff --git a/fs/9p/conv.c b/fs/9p/conv.c index 56d88c1a09c..a3ed571eee3 100644 --- a/fs/9p/conv.c +++ b/fs/9p/conv.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include "debug.h" diff --git a/fs/9p/fcall.c b/fs/9p/fcall.c index 8556097fcda..dc336a67592 100644 --- a/fs/9p/fcall.c +++ b/fs/9p/fcall.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include "debug.h" diff --git a/fs/9p/fid.c b/fs/9p/fid.c index 70492ccb438..27507201f9e 100644 --- a/fs/9p/fid.c +++ b/fs/9p/fid.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include "debug.h" diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index 0f628041e3f..0b96fae8b47 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c index e32d5971039..905c882f4e2 100644 --- a/fs/9p/vfs_dir.c +++ b/fs/9p/vfs_dir.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index c3c47eda757..79e6f9cd734 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/inotify.c b/fs/inotify.c index 723836a1f71..f5099d86fd9 100644 --- a/fs/inotify.c +++ b/fs/inotify.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c index 0ae3cd10702..73f0d60f73a 100644 --- a/fs/jffs2/acl.c +++ b/fs/jffs2/acl.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index b9b700730df..70707309dfa 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -19,6 +19,7 @@ #include #include #include +#include #include "nodelist.h" diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c index 37db5248826..ed814b1ff4d 100644 --- a/fs/jfs/ioctl.c +++ b/fs/jfs/ioctl.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include diff --git a/fs/proc/root.c b/fs/proc/root.c index ffe66c38488..64d242b6dcf 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/super.c b/fs/super.c index 47e554c12e7..84c320f6ad7 100644 --- a/fs/super.c +++ b/fs/super.c @@ -220,6 +220,24 @@ static int grab_super(struct super_block *s) __releases(sb_lock) return 0; } +/* + * Superblock locking. We really ought to get rid of these two. + */ +void lock_super(struct super_block * sb) +{ + get_fs_excl(); + mutex_lock(&sb->s_lock); +} + +void unlock_super(struct super_block * sb) +{ + put_fs_excl(); + mutex_unlock(&sb->s_lock); +} + +EXPORT_SYMBOL(lock_super); +EXPORT_SYMBOL(unlock_super); + /* * Write out and wait upon all dirty data associated with this * superblock. Filesystem data as well as the underlying block diff --git a/fs/sync.c b/fs/sync.c index 1de747b5ddb..865f32be386 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/utimes.c b/fs/utimes.c index 1bcd852fc4a..99cf2cb11fe 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include diff --git a/include/linux/fs.h b/include/linux/fs.h index 2fe6e3f900b..cac7b1ef954 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -276,7 +276,7 @@ extern int dir_notify_enable; #include #include #include -#include +#include #include #include @@ -977,36 +977,13 @@ enum { #define vfs_check_frozen(sb, level) \ wait_event((sb)->s_wait_unfrozen, ((sb)->s_frozen < (level))) -static inline void get_fs_excl(void) -{ - atomic_inc(¤t->fs_excl); -} - -static inline void put_fs_excl(void) -{ - atomic_dec(¤t->fs_excl); -} - -static inline int has_fs_excl(void) -{ - return atomic_read(¤t->fs_excl); -} +#define get_fs_excl() atomic_inc(¤t->fs_excl) +#define put_fs_excl() atomic_dec(¤t->fs_excl) +#define has_fs_excl() atomic_read(¤t->fs_excl) - -/* - * Superblock locking. - */ -static inline void lock_super(struct super_block * sb) -{ - get_fs_excl(); - mutex_lock(&sb->s_lock); -} - -static inline void unlock_super(struct super_block * sb) -{ - put_fs_excl(); - mutex_unlock(&sb->s_lock); -} +/* not quite ready to be deprecated, but... */ +extern void lock_super(struct super_block *); +extern void unlock_super(struct super_block *); /* * VFS helper functions.. diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h index 9158a68140c..cbfa1153742 100644 --- a/include/linux/radix-tree.h +++ b/include/linux/radix-tree.h @@ -19,7 +19,6 @@ #ifndef _LINUX_RADIX_TREE_H #define _LINUX_RADIX_TREE_H -#include #include #include -- cgit v1.2.3-70-g09d2 From dc3d1742543fffc79dc4d680ab64d2059e97d809 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:00 +0100 Subject: [PATCH] i386: Update defconfig Signed-off-by: Andi Kleen --- arch/i386/defconfig | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/defconfig b/arch/i386/defconfig index 97aacd6bd7d..65891f11ace 100644 --- a/arch/i386/defconfig +++ b/arch/i386/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-rc2-git4 -# Sat Oct 21 03:38:56 2006 +# Linux kernel version: 2.6.19-git7 +# Wed Dec 6 23:50:49 2006 # CONFIG_X86_32=y CONFIG_GENERIC_TIME=y @@ -40,13 +40,14 @@ CONFIG_POSIX_MQUEUE=y CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y # CONFIG_CPUSETS is not set +CONFIG_SYSFS_DEPRECATED=y # CONFIG_RELAY is not set CONFIG_INITRAMFS_SOURCE="" CONFIG_CC_OPTIMIZE_FOR_SIZE=y CONFIG_SYSCTL=y # CONFIG_EMBEDDED is not set CONFIG_UID16=y -# CONFIG_SYSCTL_SYSCALL is not set +CONFIG_SYSCTL_SYSCALL=y CONFIG_KALLSYMS=y CONFIG_KALLSYMS_ALL=y # CONFIG_KALLSYMS_EXTRA_PASS is not set @@ -110,6 +111,7 @@ CONFIG_SMP=y # CONFIG_X86_VISWS is not set CONFIG_X86_GENERICARCH=y # CONFIG_X86_ES7000 is not set +# CONFIG_PARAVIRT is not set CONFIG_X86_CYCLONE_TIMER=y # CONFIG_M386 is not set # CONFIG_M486 is not set @@ -120,6 +122,7 @@ CONFIG_X86_CYCLONE_TIMER=y # CONFIG_MPENTIUMII is not set CONFIG_MPENTIUMIII=y # CONFIG_MPENTIUMM is not set +# CONFIG_MCORE2 is not set # CONFIG_MPENTIUM4 is not set # CONFIG_MK6 is not set # CONFIG_MK7 is not set @@ -197,7 +200,6 @@ CONFIG_RESOURCES_64BIT=y CONFIG_MTRR=y # CONFIG_EFI is not set # CONFIG_IRQBALANCE is not set -CONFIG_REGPARM=y CONFIG_SECCOMP=y # CONFIG_HZ_100 is not set CONFIG_HZ_250=y @@ -205,7 +207,8 @@ CONFIG_HZ_250=y CONFIG_HZ=250 # CONFIG_KEXEC is not set # CONFIG_CRASH_DUMP is not set -CONFIG_PHYSICAL_START=0x100000 +# CONFIG_RELOCATABLE is not set +CONFIG_PHYSICAL_ALIGN=0x100000 # CONFIG_HOTPLUG_CPU is not set CONFIG_COMPAT_VDSO=y CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y @@ -367,6 +370,7 @@ CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set CONFIG_TCP_CONG_CUBIC=y CONFIG_DEFAULT_TCP_CONG="cubic" +# CONFIG_TCP_MD5SIG is not set CONFIG_IPV6=y # CONFIG_IPV6_PRIVACY is not set # CONFIG_IPV6_ROUTER_PREF is not set @@ -677,6 +681,7 @@ CONFIG_SATA_INTEL_COMBINED=y # CONFIG_PATA_IT821X is not set # CONFIG_PATA_JMICRON is not set # CONFIG_PATA_TRIFLEX is not set +# CONFIG_PATA_MARVELL is not set # CONFIG_PATA_MPIIX is not set # CONFIG_PATA_OLDPIIX is not set # CONFIG_PATA_NETCELL is not set @@ -850,6 +855,7 @@ CONFIG_BNX2=y # CONFIG_IXGB is not set # CONFIG_S2IO is not set # CONFIG_MYRI10GE is not set +# CONFIG_NETXEN_NIC is not set # # Token Ring devices @@ -984,10 +990,6 @@ CONFIG_RTC=y # CONFIG_R3964 is not set # CONFIG_APPLICOM is not set # CONFIG_SONYPI is not set - -# -# Ftape, the floppy tape device driver -# CONFIG_AGP=y # CONFIG_AGP_ALI is not set # CONFIG_AGP_ATI is not set @@ -1108,6 +1110,7 @@ CONFIG_USB_DEVICEFS=y # CONFIG_USB_BANDWIDTH is not set # CONFIG_USB_DYNAMIC_MINORS is not set # CONFIG_USB_SUSPEND is not set +# CONFIG_USB_MULTITHREAD_PROBE is not set # CONFIG_USB_OTG is not set # @@ -1185,6 +1188,7 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_KAWETH is not set # CONFIG_USB_PEGASUS is not set # CONFIG_USB_RTL8150 is not set +# CONFIG_USB_USBNET_MII is not set # CONFIG_USB_USBNET is not set CONFIG_USB_MON=y -- cgit v1.2.3-70-g09d2 From 9b483417527f2e47985856867c5716df013227c7 Mon Sep 17 00:00:00 2001 From: Andreas Mohr Date: Thu, 7 Dec 2006 02:14:00 +0100 Subject: [PATCH] i386: fix buggy MTRR address checks Fix checks that failed to realize that values are 4-kB-unit-sized (note the format strings in this same diff context which *do* realize the unit size, via appended "000"!). Also fix an incorrect below-1MB area check (as gathered from Jan Beulich's unapplied patch at http://www.ussg.iu.edu/hypermail/linux/kernel/0411.1/1378.html ) Update mtrr_add_page() docu to make 4-kB-sized calculation more obvious. Given several further items mentioned in Jan's patch mail, all in all MTRR code seems surprisingly buggy, for a surprisingly long period of time (many years). Further work/investigation would be useful. TBD Note that my patch is pretty much UNTESTED, since I can only verify that it TBD successfully boots my machine, but I cannot test against actual buggy TBD hardware which would require these (formerly broken) checks. Long -mm TBD simmering would make sense, especially since these now-working checks might TBD turn out to have adverse effects on unaffected hardware. Signed-off-by: Andreas Mohr Signed-off-by: Andi Kleen Acked-by: Jan Beulich Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/cpu/mtrr/generic.c | 4 ++-- arch/i386/kernel/cpu/mtrr/main.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/mtrr/generic.c b/arch/i386/kernel/cpu/mtrr/generic.c index 0b61eed8bbd..ee8dc675395 100644 --- a/arch/i386/kernel/cpu/mtrr/generic.c +++ b/arch/i386/kernel/cpu/mtrr/generic.c @@ -366,7 +366,7 @@ int generic_validate_add_page(unsigned long base, unsigned long size, unsigned i printk(KERN_WARNING "mtrr: base(0x%lx000) is not 4 MiB aligned\n", base); return -EINVAL; } - if (!(base + size < 0x70000000 || base > 0x7003FFFF) && + if (!(base + size < 0x70000 || base > 0x7003F) && (type == MTRR_TYPE_WRCOMB || type == MTRR_TYPE_WRBACK)) { printk(KERN_WARNING "mtrr: writable mtrr between 0x70000000 and 0x7003FFFF may hang the CPU.\n"); @@ -374,7 +374,7 @@ int generic_validate_add_page(unsigned long base, unsigned long size, unsigned i } } - if (base + size < 0x100) { + if (base < 0x100) { printk(KERN_WARNING "mtrr: cannot set region below 1 MiB (0x%lx000,0x%lx000)\n", base, size); return -EINVAL; diff --git a/arch/i386/kernel/cpu/mtrr/main.c b/arch/i386/kernel/cpu/mtrr/main.c index fff90bda473..2b8b0b361cc 100644 --- a/arch/i386/kernel/cpu/mtrr/main.c +++ b/arch/i386/kernel/cpu/mtrr/main.c @@ -263,8 +263,8 @@ static void set_mtrr(unsigned int reg, unsigned long base, /** * mtrr_add_page - Add a memory type region - * @base: Physical base address of region in pages (4 KB) - * @size: Physical size of region in pages (4 KB) + * @base: Physical base address of region in pages (in units of 4 kB!) + * @size: Physical size of region in pages (4 kB) * @type: Type of MTRR desired * @increment: If this is true do usage counting on the region * -- cgit v1.2.3-70-g09d2 From b615ebdac97c648a2ae7d23c5a0bbb3972adf928 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:00 +0100 Subject: [PATCH] x86: shorten lines in unwinder to be <= 80 characters Andrew complained about > 80 character lines in the new unwinder. Fix that. Signed-off-by: Andi Kleen --- arch/i386/kernel/traps.c | 16 ++++++++++------ arch/x86_64/kernel/traps.c | 21 +++++++++++++-------- 2 files changed, 23 insertions(+), 14 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index fe9c5e8e7e6..fe81d89c50d 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -173,6 +173,8 @@ dump_trace_unwind(struct unwind_frame_info *info, void *data) return n; } +#define MSG(msg) ops->warning(data, msg) + void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, struct stacktrace_ops *ops, void *data) @@ -191,29 +193,31 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, if (unwind_init_frame_info(&info, task, regs) == 0) unw_ret = dump_trace_unwind(&info, &oad); } else if (task == current) - unw_ret = unwind_init_running(&info, dump_trace_unwind, &oad); + unw_ret = unwind_init_running(&info, dump_trace_unwind, + &oad); else { if (unwind_init_blocked(&info, task) == 0) unw_ret = dump_trace_unwind(&info, &oad); } if (unw_ret > 0) { if (call_trace == 1 && !arch_unw_user_mode(&info)) { - ops->warning_symbol(data, "DWARF2 unwinder stuck at %s\n", + ops->warning_symbol(data, + "DWARF2 unwinder stuck at %s\n", UNW_PC(&info)); if (UNW_SP(&info) >= PAGE_OFFSET) { - ops->warning(data, "Leftover inexact backtrace:\n"); + MSG("Leftover inexact backtrace:\n"); stack = (void *)UNW_SP(&info); if (!stack) return; ebp = UNW_FP(&info); } else - ops->warning(data, "Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:\n"); } else if (call_trace >= 1) return; else - ops->warning(data, "Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:\n"); } else - ops->warning(data, "Inexact backtrace:\n"); + MSG("Inexact backtrace:\n"); } if (!stack) { unsigned long dummy; diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 0d65b22f229..d3f43c958ca 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -235,6 +235,8 @@ static int dump_trace_unwind(struct unwind_frame_info *info, void *context) return n; } +#define MSG(txt) ops->warning(data, txt) + /* * x86-64 can have upto three kernel stacks: * process stack @@ -248,11 +250,12 @@ static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) return p > t && p < t + THREAD_SIZE - 3; } -void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack, +void dump_trace(struct task_struct *tsk, struct pt_regs *regs, + unsigned long *stack, struct stacktrace_ops *ops, void *data) { const unsigned cpu = smp_processor_id(); - unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr; + unsigned long *irqstack_end = (unsigned long*)cpu_pda(cpu)->irqstackptr; unsigned used = 0; struct thread_info *tinfo; @@ -268,28 +271,30 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s if (unwind_init_frame_info(&info, tsk, regs) == 0) unw_ret = dump_trace_unwind(&info, &oad); } else if (tsk == current) - unw_ret = unwind_init_running(&info, dump_trace_unwind, &oad); + unw_ret = unwind_init_running(&info, dump_trace_unwind, + &oad); else { if (unwind_init_blocked(&info, tsk) == 0) unw_ret = dump_trace_unwind(&info, &oad); } if (unw_ret > 0) { if (call_trace == 1 && !arch_unw_user_mode(&info)) { - ops->warning_symbol(data, "DWARF2 unwinder stuck at %s\n", + ops->warning_symbol(data, + "DWARF2 unwinder stuck at %s\n", UNW_PC(&info)); if ((long)UNW_SP(&info) < 0) { - ops->warning(data, "Leftover inexact backtrace:\n"); + MSG("Leftover inexact backtrace:"); stack = (unsigned long *)UNW_SP(&info); if (!stack) return; } else - ops->warning(data, "Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:\n"); } else if (call_trace >= 1) return; else - ops->warning(data, "Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:\n"); } else - ops->warning(data, "Inexact backtrace:\n"); + MSG("Inexact backtrace:\n"); } if (!stack) { unsigned long dummy; -- cgit v1.2.3-70-g09d2 From dd315df1767cf56bd4fb8d730fdff4a3d7e15d84 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:00 +0100 Subject: [PATCH] x86: Compress stack unwinder output The unwinder has some extra newlines, which eat up loads of screen space when it spews. (See https://bugzilla.redhat.com/bugzilla/attachment.cgi?id=137900 for a nasty example). warning_symbol-> and warning-> already printk a newline, so don't add one in the strings passed to them. [AK: redone for new code] Signed-off-by: Dave Jones Signed-off-by: Andi Kleen --- arch/i386/kernel/traps.c | 10 +++++----- arch/x86_64/kernel/traps.c | 8 ++++---- 2 files changed, 9 insertions(+), 9 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index fe81d89c50d..396041a46e3 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -202,22 +202,22 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, if (unw_ret > 0) { if (call_trace == 1 && !arch_unw_user_mode(&info)) { ops->warning_symbol(data, - "DWARF2 unwinder stuck at %s\n", + "DWARF2 unwinder stuck at %s", UNW_PC(&info)); if (UNW_SP(&info) >= PAGE_OFFSET) { - MSG("Leftover inexact backtrace:\n"); + MSG("Leftover inexact backtrace:"); stack = (void *)UNW_SP(&info); if (!stack) return; ebp = UNW_FP(&info); } else - MSG("Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:"); } else if (call_trace >= 1) return; else - MSG("Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:"); } else - MSG("Inexact backtrace:\n"); + MSG("Inexact backtrace:"); } if (!stack) { unsigned long dummy; diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index d3f43c958ca..eedd4e759c3 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -280,7 +280,7 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, if (unw_ret > 0) { if (call_trace == 1 && !arch_unw_user_mode(&info)) { ops->warning_symbol(data, - "DWARF2 unwinder stuck at %s\n", + "DWARF2 unwinder stuck at %s", UNW_PC(&info)); if ((long)UNW_SP(&info) < 0) { MSG("Leftover inexact backtrace:"); @@ -288,13 +288,13 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, if (!stack) return; } else - MSG("Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:"); } else if (call_trace >= 1) return; else - MSG("Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:"); } else - MSG("Inexact backtrace:\n"); + MSG("Inexact backtrace:"); } if (!stack) { unsigned long dummy; -- cgit v1.2.3-70-g09d2 From d606f88fa5e529e9dc72be97e79db1e36a6261cb Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 7 Dec 2006 02:14:00 +0100 Subject: [PATCH] i386: fix must_checks Fix __must_check warnings in i386/math-emu. Signed-off-by: Randy Dunlap Signed-off-by: Andi Kleen --- arch/i386/math-emu/fpu_emu.h | 1 + arch/i386/math-emu/fpu_entry.c | 3 ++- arch/i386/math-emu/fpu_system.h | 1 + arch/i386/math-emu/load_store.c | 2 ++ arch/i386/math-emu/reg_ld_str.c | 15 ++++++++++----- 5 files changed, 16 insertions(+), 6 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/math-emu/fpu_emu.h b/arch/i386/math-emu/fpu_emu.h index d62b20a3e66..65120f52385 100644 --- a/arch/i386/math-emu/fpu_emu.h +++ b/arch/i386/math-emu/fpu_emu.h @@ -57,6 +57,7 @@ #define TAG_Special Const(2) /* De-normal, + or - infinity, or Not a Number */ #define TAG_Empty Const(3) /* empty */ +#define TAG_Error Const(0x80) /* probably need to abort */ #define LOADED_DATA Const(10101) /* Special st() number to identify loaded data (not on stack). */ diff --git a/arch/i386/math-emu/fpu_entry.c b/arch/i386/math-emu/fpu_entry.c index d93f16ef828..ddf8fa3bbd0 100644 --- a/arch/i386/math-emu/fpu_entry.c +++ b/arch/i386/math-emu/fpu_entry.c @@ -742,7 +742,8 @@ int save_i387_soft(void *s387, struct _fpstate __user * buf) S387->fcs &= ~0xf8000000; S387->fos |= 0xffff0000; #endif /* PECULIAR_486 */ - __copy_to_user(d, &S387->cwd, 7*4); + if (__copy_to_user(d, &S387->cwd, 7*4)) + return -1; RE_ENTRANT_CHECK_ON; d += 7*4; diff --git a/arch/i386/math-emu/fpu_system.h b/arch/i386/math-emu/fpu_system.h index bf26341c8bd..a3ae28c49dd 100644 --- a/arch/i386/math-emu/fpu_system.h +++ b/arch/i386/math-emu/fpu_system.h @@ -68,6 +68,7 @@ #define FPU_access_ok(x,y,z) if ( !access_ok(x,y,z) ) \ math_abort(FPU_info,SIGSEGV) +#define FPU_abort math_abort(FPU_info, SIGSEGV) #undef FPU_IGNORE_CODE_SEGV #ifdef FPU_IGNORE_CODE_SEGV diff --git a/arch/i386/math-emu/load_store.c b/arch/i386/math-emu/load_store.c index 85314be2fef..eebd6fb1c8a 100644 --- a/arch/i386/math-emu/load_store.c +++ b/arch/i386/math-emu/load_store.c @@ -227,6 +227,8 @@ int FPU_load_store(u_char type, fpu_addr_modes addr_modes, case 027: /* fild m64int */ clear_C1(); loaded_tag = FPU_load_int64((long long __user *)data_address); + if (loaded_tag == TAG_Error) + return 0; FPU_settag0(loaded_tag); break; case 030: /* fstenv m14/28byte */ diff --git a/arch/i386/math-emu/reg_ld_str.c b/arch/i386/math-emu/reg_ld_str.c index f06ed41d191..e976caef649 100644 --- a/arch/i386/math-emu/reg_ld_str.c +++ b/arch/i386/math-emu/reg_ld_str.c @@ -244,7 +244,8 @@ int FPU_load_int64(long long __user *_s) RE_ENTRANT_CHECK_OFF; FPU_access_ok(VERIFY_READ, _s, 8); - copy_from_user(&s,_s,8); + if (copy_from_user(&s,_s,8)) + FPU_abort; RE_ENTRANT_CHECK_ON; if (s == 0) @@ -907,7 +908,8 @@ int FPU_store_int64(FPU_REG *st0_ptr, u_char st0_tag, long long __user *d) RE_ENTRANT_CHECK_OFF; FPU_access_ok(VERIFY_WRITE,d,8); - copy_to_user(d, &tll, 8); + if (copy_to_user(d, &tll, 8)) + FPU_abort; RE_ENTRANT_CHECK_ON; return 1; @@ -1336,7 +1338,8 @@ u_char __user *fstenv(fpu_addr_modes addr_modes, u_char __user *d) I387.soft.fcs &= ~0xf8000000; I387.soft.fos |= 0xffff0000; #endif /* PECULIAR_486 */ - __copy_to_user(d, &control_word, 7*4); + if (__copy_to_user(d, &control_word, 7*4)) + FPU_abort; RE_ENTRANT_CHECK_ON; d += 0x1c; } @@ -1359,9 +1362,11 @@ void fsave(fpu_addr_modes addr_modes, u_char __user *data_address) FPU_access_ok(VERIFY_WRITE,d,80); /* Copy all registers in stack order. */ - __copy_to_user(d, register_base+offset, other); + if (__copy_to_user(d, register_base+offset, other)) + FPU_abort; if ( offset ) - __copy_to_user(d+other, register_base, offset); + if (__copy_to_user(d+other, register_base, offset)) + FPU_abort; RE_ENTRANT_CHECK_ON; finit(); -- cgit v1.2.3-70-g09d2 From a63954b5cad5765e52870bb649992bf636f32a6b Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 7 Dec 2006 02:14:00 +0100 Subject: [PATCH] i386: remove pointless printk from i386 oops output This just got removed on x86-64, do the same on 32bit. It always annoyed me when this ate a line of oops output pushing interesting stuff off the screen. Signed-off-by: Dave Jones Signed-off-by: Andi Kleen --- arch/i386/kernel/traps.c | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 396041a46e3..48ebfab661b 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -777,7 +777,6 @@ void __kprobes die_nmi(struct pt_regs *regs, const char *msg) printk(" on CPU%d, eip %08lx, registers:\n", smp_processor_id(), regs->eip); show_registers(regs); - printk(KERN_EMERG "console shuts up ...\n"); console_silent(); spin_unlock(&nmi_print_lock); bust_spinlocks(0); -- cgit v1.2.3-70-g09d2 From 42ed458aa51337357d7632c64aed4528f923e829 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] i386: i386 add X86_FEATURE_PEBS and detection Here is a patch (used by perfmon2) to detect the presence of the Precise Event Based Sampling (PEBS) feature for i386. The patch also adds the cpu_has_pebs macro. - adds X86_FEATURE_PEBS - adds cpu_has_pebs to test for X86_FEATURE_PEBS Signed-off-by: stephane eranian Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/cpu/intel.c | 8 +++++++- include/asm-i386/cpufeature.h | 2 ++ 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c index 94a95aa5227..798c2f617e8 100644 --- a/arch/i386/kernel/cpu/intel.c +++ b/arch/i386/kernel/cpu/intel.c @@ -195,8 +195,14 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) if ((c->x86 == 0xf && c->x86_model >= 0x03) || (c->x86 == 0x6 && c->x86_model >= 0x0e)) set_bit(X86_FEATURE_CONSTANT_TSC, c->x86_capability); -} + if (cpu_has_ds) { + unsigned int l1; + rdmsr(MSR_IA32_MISC_ENABLE, l1, l2); + if (!(l1 & (1<<12))) + set_bit(X86_FEATURE_PEBS, c->x86_capability); + } +} static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 * c, unsigned int size) { diff --git a/include/asm-i386/cpufeature.h b/include/asm-i386/cpufeature.h index 69ce35049a0..231672558c1 100644 --- a/include/asm-i386/cpufeature.h +++ b/include/asm-i386/cpufeature.h @@ -73,6 +73,7 @@ #define X86_FEATURE_UP (3*32+ 9) /* smp kernel running on up */ #define X86_FEATURE_FXSAVE_LEAK (3*32+10) /* FXSAVE leaks FOP/FIP/FOP */ #define X86_FEATURE_ARCH_PERFMON (3*32+11) /* Intel Architectural PerfMon */ +#define X86_FEATURE_PEBS (3*32+12) /* Precise-Event Based Sampling */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* Streaming SIMD Extensions-3 */ @@ -135,6 +136,7 @@ #define cpu_has_pmm boot_cpu_has(X86_FEATURE_PMM) #define cpu_has_pmm_enabled boot_cpu_has(X86_FEATURE_PMM_EN) #define cpu_has_ds boot_cpu_has(X86_FEATURE_DS) +#define cpu_has_pebs boot_cpu_has(X86_FEATURE_PEBS) #endif /* __ASM_I386_CPUFEATURE_H */ -- cgit v1.2.3-70-g09d2 From e5e3a0428968dcc1f9318ce1c941a918e99f8b84 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] i386: remove default_ldt, and simplify ldt-setting. This patch removes the default_ldt[] array, as it has been unused since iBCS stopped being supported. This means it is now possible to actually set an empty LDT segment. In order to deal with this, the set_ldt_desc/load_LDT pair has been replaced with a single set_ldt() operation which is responsible for both setting up the LDT descriptor in the GDT, and reloading the LDT register. If there are no LDT entries, the LDT register is loaded with a NULL descriptor. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Andi Kleen Acked-by: Zachary Amsden Signed-off-by: Andrew Morton --- arch/i386/kernel/ldt.c | 4 +--- arch/i386/kernel/traps.c | 3 --- include/asm-i386/desc.h | 50 ++++++++++++++++-------------------------- include/asm-i386/mmu_context.h | 4 ++-- 4 files changed, 22 insertions(+), 39 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/ldt.c b/arch/i386/kernel/ldt.c index 445211eb2d5..b410e5fb034 100644 --- a/arch/i386/kernel/ldt.c +++ b/arch/i386/kernel/ldt.c @@ -160,16 +160,14 @@ static int read_default_ldt(void __user * ptr, unsigned long bytecount) { int err; unsigned long size; - void *address; err = 0; - address = &default_ldt[0]; size = 5*sizeof(struct desc_struct); if (size > bytecount) size = bytecount; err = size; - if (copy_to_user(ptr, address, size)) + if (clear_user(ptr, size)) err = -EFAULT; return err; diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 48ebfab661b..56655ea8d98 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -61,9 +61,6 @@ int panic_on_unrecovered_nmi; asmlinkage int system_call(void); -struct desc_struct default_ldt[] = { { 0, 0 }, { 0, 0 }, { 0, 0 }, - { 0, 0 }, { 0, 0 } }; - /* Do we ignore FPU interrupts ? */ char ignore_fpu_irq = 0; diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h index 5874ef119ff..a0398f780ca 100644 --- a/include/asm-i386/desc.h +++ b/include/asm-i386/desc.h @@ -33,11 +33,6 @@ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu) return (struct desc_struct *)per_cpu(cpu_gdt_descr, cpu).address; } -/* - * This is the ldt that every process will get unless we need - * something other than this. - */ -extern struct desc_struct default_ldt[]; extern struct desc_struct idt_table[]; extern void set_intr_gate(unsigned int irq, void * addr); @@ -65,7 +60,6 @@ static inline void pack_gate(__u32 *a, __u32 *b, #define DESCTYPE_S 0x10 /* !system */ #define load_TR_desc() __asm__ __volatile__("ltr %w0"::"q" (GDT_ENTRY_TSS*8)) -#define load_LDT_desc() __asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8)) #define load_gdt(dtr) __asm__ __volatile("lgdt %0"::"m" (*dtr)) #define load_idt(dtr) __asm__ __volatile("lidt %0"::"m" (*dtr)) @@ -115,13 +109,20 @@ static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const vo write_gdt_entry(get_cpu_gdt_table(cpu), entry, a, b); } -static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int entries) +static inline void set_ldt(void *addr, unsigned int entries) { - __u32 a, b; - pack_descriptor(&a, &b, (unsigned long)addr, - entries * sizeof(struct desc_struct) - 1, - DESCTYPE_LDT, 0); - write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, a, b); + if (likely(entries == 0)) + __asm__ __volatile__("lldt %w0"::"q" (0)); + else { + unsigned cpu = smp_processor_id(); + __u32 a, b; + + pack_descriptor(&a, &b, (unsigned long)addr, + entries * sizeof(struct desc_struct) - 1, + DESCTYPE_LDT, 0); + write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, a, b); + __asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8)); + } } #define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr) @@ -153,35 +154,22 @@ static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int entri static inline void clear_LDT(void) { - int cpu = get_cpu(); - - set_ldt_desc(cpu, &default_ldt[0], 5); - load_LDT_desc(); - put_cpu(); + set_ldt(NULL, 0); } /* * load one particular LDT into the current CPU */ -static inline void load_LDT_nolock(mm_context_t *pc, int cpu) +static inline void load_LDT_nolock(mm_context_t *pc) { - void *segments = pc->ldt; - int count = pc->size; - - if (likely(!count)) { - segments = &default_ldt[0]; - count = 5; - } - - set_ldt_desc(cpu, segments, count); - load_LDT_desc(); + set_ldt(pc->ldt, pc->size); } static inline void load_LDT(mm_context_t *pc) { - int cpu = get_cpu(); - load_LDT_nolock(pc, cpu); - put_cpu(); + preempt_disable(); + load_LDT_nolock(pc); + preempt_enable(); } static inline unsigned long get_desc_base(unsigned long *desc) diff --git a/include/asm-i386/mmu_context.h b/include/asm-i386/mmu_context.h index 62b7bf18409..1b1495372c4 100644 --- a/include/asm-i386/mmu_context.h +++ b/include/asm-i386/mmu_context.h @@ -44,7 +44,7 @@ static inline void switch_mm(struct mm_struct *prev, * load the LDT, if the LDT is different: */ if (unlikely(prev->context.ldt != next->context.ldt)) - load_LDT_nolock(&next->context, cpu); + load_LDT_nolock(&next->context); } #ifdef CONFIG_SMP else { @@ -56,7 +56,7 @@ static inline void switch_mm(struct mm_struct *prev, * tlb flush IPI delivery. We must reload %cr3. */ load_cr3(next->pgd); - load_LDT_nolock(&next->context, cpu); + load_LDT_nolock(&next->context); } } #endif -- cgit v1.2.3-70-g09d2 From bb81a09e55eaf7e5f798468ab971469b6f66a259 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] x86: all cpu backtrace When a spinlock lockup occurs, arrange for the NMI code to emit an all-cpu backtrace, so we get to see which CPU is holding the lock, and where. Cc: Andi Kleen Cc: Ingo Molnar Cc: Badari Pulavarty Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/kernel/nmi.c | 26 ++++++++++++++++++++++++++ arch/x86_64/kernel/nmi.c | 29 ++++++++++++++++++++++++++++- include/asm-i386/nmi.h | 8 ++++++++ include/asm-x86_64/nmi.h | 3 +++ include/linux/nmi.h | 5 +++++ lib/spinlock_debug.c | 4 ++++ 6 files changed, 74 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index eaafe233a5d..171194ccb7b 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include @@ -42,6 +43,8 @@ int nmi_watchdog_enabled; static DEFINE_PER_CPU(unsigned long, perfctr_nmi_owner); static DEFINE_PER_CPU(unsigned long, evntsel_nmi_owner[3]); +static cpumask_t backtrace_mask = CPU_MASK_NONE; + /* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's * offset from MSR_P4_BSU_ESCR0. It will be the max for all platforms (for now) */ @@ -907,6 +910,16 @@ __kprobes int nmi_watchdog_tick(struct pt_regs * regs, unsigned reason) touched = 1; } + if (cpu_isset(cpu, backtrace_mask)) { + static DEFINE_SPINLOCK(lock); /* Serialise the printks */ + + spin_lock(&lock); + printk("NMI backtrace for cpu %d\n", cpu); + dump_stack(); + spin_unlock(&lock); + cpu_clear(cpu, backtrace_mask); + } + sum = per_cpu(irq_stat, cpu).apic_timer_irqs; /* if the apic timer isn't firing, this cpu isn't doing much */ @@ -1033,6 +1046,19 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file, #endif +void __trigger_all_cpu_backtrace(void) +{ + int i; + + backtrace_mask = cpu_online_map; + /* Wait for up to 10 seconds for all CPUs to do the backtrace */ + for (i = 0; i < 10 * 1000; i++) { + if (cpus_empty(backtrace_mask)) + break; + mdelay(1); + } +} + EXPORT_SYMBOL(nmi_active); EXPORT_SYMBOL(nmi_watchdog); EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi); diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index 7af9cb3e2d9..27e95e7922c 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -12,14 +12,15 @@ * Mikael Pettersson : PM converted to driver model. Disable/enable API. */ +#include #include #include #include #include #include -#include #include #include +#include #include #include @@ -41,6 +42,8 @@ int panic_on_unrecovered_nmi; static DEFINE_PER_CPU(unsigned, perfctr_nmi_owner); static DEFINE_PER_CPU(unsigned, evntsel_nmi_owner[2]); +static cpumask_t backtrace_mask = CPU_MASK_NONE; + /* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's * offset from MSR_P4_BSU_ESCR0. It will be the max for all platforms (for now) */ @@ -782,6 +785,7 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason) { int sum; int touched = 0; + int cpu = smp_processor_id(); struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); u64 dummy; int rc=0; @@ -799,6 +803,16 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason) touched = 1; } + if (cpu_isset(cpu, backtrace_mask)) { + static DEFINE_SPINLOCK(lock); /* Serialise the printks */ + + spin_lock(&lock); + printk("NMI backtrace for cpu %d\n", cpu); + dump_stack(); + spin_unlock(&lock); + cpu_clear(cpu, backtrace_mask); + } + #ifdef CONFIG_X86_MCE /* Could check oops_in_progress here too, but it's safer not too */ @@ -931,6 +945,19 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file, #endif +void __trigger_all_cpu_backtrace(void) +{ + int i; + + backtrace_mask = cpu_online_map; + /* Wait for up to 10 seconds for all CPUs to do the backtrace */ + for (i = 0; i < 10 * 1000; i++) { + if (cpus_empty(backtrace_mask)) + break; + mdelay(1); + } +} + EXPORT_SYMBOL(nmi_active); EXPORT_SYMBOL(nmi_watchdog); EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi); diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h index 269d315719c..b04333ea6f3 100644 --- a/include/asm-i386/nmi.h +++ b/include/asm-i386/nmi.h @@ -5,6 +5,9 @@ #define ASM_NMI_H #include +#include + +#ifdef ARCH_HAS_NMI_WATCHDOG /** * do_nmi_callback @@ -42,4 +45,9 @@ extern int proc_nmi_enabled(struct ctl_table *, int , struct file *, void __user *, size_t *, loff_t *); extern int unknown_nmi_panic; +void __trigger_all_cpu_backtrace(void); +#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace() + +#endif + #endif /* ASM_NMI_H */ diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h index f367d4014b4..72375e7d32a 100644 --- a/include/asm-x86_64/nmi.h +++ b/include/asm-x86_64/nmi.h @@ -77,4 +77,7 @@ extern int proc_nmi_enabled(struct ctl_table *, int , struct file *, extern int unknown_nmi_panic; +void __trigger_all_cpu_backtrace(void); +#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace() + #endif /* ASM_NMI_H */ diff --git a/include/linux/nmi.h b/include/linux/nmi.h index e16904e28c3..acb4ed13024 100644 --- a/include/linux/nmi.h +++ b/include/linux/nmi.h @@ -15,9 +15,14 @@ * disables interrupts for a long time. This call is stateless. */ #ifdef ARCH_HAS_NMI_WATCHDOG +#include extern void touch_nmi_watchdog(void); #else # define touch_nmi_watchdog() touch_softlockup_watchdog() #endif +#ifndef trigger_all_cpu_backtrace +#define trigger_all_cpu_backtrace() do { } while (0) +#endif + #endif diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index b6c4f898197..479fd462eaa 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -7,6 +7,7 @@ */ #include +#include #include #include #include @@ -117,6 +118,9 @@ static void __spin_lock_debug(spinlock_t *lock) raw_smp_processor_id(), current->comm, current->pid, lock); dump_stack(); +#ifdef CONFIG_SMP + trigger_all_cpu_backtrace(); +#endif } } } -- cgit v1.2.3-70-g09d2 From be44d2aabce2d62f72d5751d1871b6212bf7a1c7 Mon Sep 17 00:00:00 2001 From: Stas Sergeev Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] i386: espfix cleanup Clean up the espfix code: - Introduced PER_CPU() macro to be used from asm - Introduced GET_DESC_BASE() macro to be used from asm - Rewrote the fixup code in asm, as calling a C code with the altered %ss appeared to be unsafe - No longer altering the stack from a .fixup section - 16bit per-cpu stack is no longer used, instead the stack segment base is patched the way so that the high word of the kernel and user %esp are the same. - Added the limit-patching for the espfix segment. (Chuck Ebbert) [jeremy@goop.org: use the x86 scaling addressing mode rather than shifting] Signed-off-by: Stas Sergeev Signed-off-by: Andi Kleen Acked-by: Zachary Amsden Acked-by: Chuck Ebbert <76306.1226@compuserve.com> Acked-by: Jan Beulich Cc: Andi Kleen Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andrew Morton --- arch/i386/kernel/asm-offsets.c | 5 +++ arch/i386/kernel/cpu/common.c | 11 ------- arch/i386/kernel/entry.S | 73 +++++++++++++++++++----------------------- arch/i386/kernel/head.S | 2 +- arch/i386/kernel/traps.c | 57 +++++++++------------------------ include/asm-i386/desc.h | 27 +++++++++++++--- include/asm-i386/percpu.h | 25 +++++++++++++++ 7 files changed, 103 insertions(+), 97 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index c80271f8f08..e94d910a28b 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -58,6 +58,11 @@ void foo(void) OFFSET(TI_sysenter_return, thread_info, sysenter_return); BLANK(); + OFFSET(GDS_size, Xgt_desc_struct, size); + OFFSET(GDS_address, Xgt_desc_struct, address); + OFFSET(GDS_pad, Xgt_desc_struct, pad); + BLANK(); + OFFSET(EXEC_DOMAIN_handler, exec_domain, handler); OFFSET(RT_SIGFRAME_sigcontext, rt_sigframe, uc.uc_mcontext); BLANK(); diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index d9f3e3c31f0..5532fc4e1bf 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -24,9 +24,6 @@ DEFINE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr); EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr); -DEFINE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]); -EXPORT_PER_CPU_SYMBOL(cpu_16bit_stack); - static int cachesize_override __cpuinitdata = -1; static int disable_x86_fxsr __cpuinitdata; static int disable_x86_serial_nr __cpuinitdata = 1; @@ -603,7 +600,6 @@ void __cpuinit cpu_init(void) struct tss_struct * t = &per_cpu(init_tss, cpu); struct thread_struct *thread = ¤t->thread; struct desc_struct *gdt; - __u32 stk16_off = (__u32)&per_cpu(cpu_16bit_stack, cpu); struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); if (cpu_test_and_set(cpu, cpu_initialized)) { @@ -651,13 +647,6 @@ old_gdt: * and set up the GDT descriptor: */ memcpy(gdt, cpu_gdt_table, GDT_SIZE); - - /* Set up GDT entry for 16bit stack */ - *(__u64 *)(&gdt[GDT_ENTRY_ESPFIX_SS]) |= - ((((__u64)stk16_off) << 16) & 0x000000ffffff0000ULL) | - ((((__u64)stk16_off) << 32) & 0xff00000000000000ULL) | - (CPU_16BIT_STACK_SIZE - 1); - cpu_gdt_descr->size = GDT_SIZE - 1; cpu_gdt_descr->address = (unsigned long)gdt; diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index 5a63d6fdb70..c38d801ba0b 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -48,6 +48,7 @@ #include #include #include +#include #include #include "irq_vectors.h" @@ -418,23 +419,18 @@ ldt_ss: * This is an "official" bug of all the x86-compatible * CPUs, which we can try to work around to make * dosemu and wine happy. */ - subl $8, %esp # reserve space for switch16 pointer - CFI_ADJUST_CFA_OFFSET 8 + movl OLDESP(%esp), %eax + movl %esp, %edx + call patch_espfix_desc + pushl $__ESPFIX_SS + CFI_ADJUST_CFA_OFFSET 4 + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 DISABLE_INTERRUPTS TRACE_IRQS_OFF - movl %esp, %eax - /* Set up the 16bit stack frame with switch32 pointer on top, - * and a switch16 pointer on top of the current frame. */ - call setup_x86_bogus_stack - CFI_ADJUST_CFA_OFFSET -8 # frame has moved - TRACE_IRQS_IRET - RESTORE_REGS - lss 20+4(%esp), %esp # switch to 16bit stack -1: INTERRUPT_RETURN -.section __ex_table,"a" - .align 4 - .long 1b,iret_exc -.previous + lss (%esp), %esp + CFI_ADJUST_CFA_OFFSET -8 + jmp restore_nocheck CFI_ENDPROC # perform work that needs to be done immediately before resumption @@ -524,30 +520,30 @@ syscall_badsys: CFI_ENDPROC #define FIXUP_ESPFIX_STACK \ - movl %esp, %eax; \ - /* switch to 32bit stack using the pointer on top of 16bit stack */ \ - lss %ss:CPU_16BIT_STACK_SIZE-8, %esp; \ - /* copy data from 16bit stack to 32bit stack */ \ - call fixup_x86_bogus_stack; \ - /* put ESP to the proper location */ \ - movl %eax, %esp; -#define UNWIND_ESPFIX_STACK \ + /* since we are on a wrong stack, we cant make it a C code :( */ \ + GET_THREAD_INFO(%ebp); \ + movl TI_cpu(%ebp), %ebx; \ + PER_CPU(cpu_gdt_descr, %ebx); \ + movl GDS_address(%ebx), %ebx; \ + GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah); \ + addl %esp, %eax; \ + pushl $__KERNEL_DS; \ + CFI_ADJUST_CFA_OFFSET 4; \ pushl %eax; \ CFI_ADJUST_CFA_OFFSET 4; \ + lss (%esp), %esp; \ + CFI_ADJUST_CFA_OFFSET -8; +#define UNWIND_ESPFIX_STACK \ movl %ss, %eax; \ - /* see if on 16bit stack */ \ + /* see if on espfix stack */ \ cmpw $__ESPFIX_SS, %ax; \ - je 28f; \ -27: popl %eax; \ - CFI_ADJUST_CFA_OFFSET -4; \ -.section .fixup,"ax"; \ -28: movl $__KERNEL_DS, %eax; \ + jne 27f; \ + movl $__KERNEL_DS, %eax; \ movl %eax, %ds; \ movl %eax, %es; \ - /* switch to 32bit stack */ \ + /* switch to normal stack */ \ FIXUP_ESPFIX_STACK; \ - jmp 27b; \ -.previous +27:; /* * Build the entry stubs and pointer table with @@ -614,7 +610,6 @@ error_code: pushl %eax CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET eax, 0 - xorl %eax, %eax pushl %ebp CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET ebp, 0 @@ -627,7 +622,6 @@ error_code: pushl %edx CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET edx, 0 - decl %eax # eax = -1 pushl %ecx CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET ecx, 0 @@ -644,7 +638,7 @@ error_code: /*CFI_REGISTER es, ecx*/ movl ES(%esp), %edi # get the function address movl ORIG_EAX(%esp), %edx # get the error code - movl %eax, ORIG_EAX(%esp) + movl $-1, ORIG_EAX(%esp) movl %ecx, ES(%esp) /*CFI_REL_OFFSET es, ES*/ movl $(__USER_DS), %ecx @@ -754,7 +748,7 @@ KPROBE_ENTRY(nmi) cmpw $__ESPFIX_SS, %ax popl %eax CFI_ADJUST_CFA_OFFSET -4 - je nmi_16bit_stack + je nmi_espfix_stack cmpl $sysenter_entry,(%esp) je nmi_stack_fixup pushl %eax @@ -797,7 +791,7 @@ nmi_debug_stack_check: FIX_STACK(24,nmi_stack_correct, 1) jmp nmi_stack_correct -nmi_16bit_stack: +nmi_espfix_stack: /* We have a RING0_INT_FRAME here. * * create the pointer to lss back @@ -806,7 +800,6 @@ nmi_16bit_stack: CFI_ADJUST_CFA_OFFSET 4 pushl %esp CFI_ADJUST_CFA_OFFSET 4 - movzwl %sp, %esp addw $4, (%esp) /* copy the iret frame of 12 bytes */ .rept 3 @@ -817,11 +810,11 @@ nmi_16bit_stack: CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL FIXUP_ESPFIX_STACK # %eax == %esp - CFI_ADJUST_CFA_OFFSET -20 # the frame has now moved xorl %edx,%edx # zero error code call do_nmi RESTORE_REGS - lss 12+4(%esp), %esp # back to 16bit stack + lss 12+4(%esp), %esp # back to espfix stack + CFI_ADJUST_CFA_OFFSET -24 1: INTERRUPT_RETURN CFI_ENDPROC .section __ex_table,"a" diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index ca31f18d277..b1f1df11fcc 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -584,7 +584,7 @@ ENTRY(cpu_gdt_table) .quad 0x00009a000000ffff /* 0xc0 APM CS 16 code (16 bit) */ .quad 0x004092000000ffff /* 0xc8 APM DS data */ - .quad 0x0000920000000000 /* 0xd0 - ESPFIX 16-bit SS */ + .quad 0x00c0920000000000 /* 0xd0 - ESPFIX SS */ .quad 0x0000000000000000 /* 0xd8 - unused */ .quad 0x0000000000000000 /* 0xe0 - unused */ .quad 0x0000000000000000 /* 0xe8 - unused */ diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 56655ea8d98..f9bb1f89d68 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -1088,49 +1088,24 @@ fastcall void do_spurious_interrupt_bug(struct pt_regs * regs, #endif } -fastcall void setup_x86_bogus_stack(unsigned char * stk) +fastcall unsigned long patch_espfix_desc(unsigned long uesp, + unsigned long kesp) { - unsigned long *switch16_ptr, *switch32_ptr; - struct pt_regs *regs; - unsigned long stack_top, stack_bot; - unsigned short iret_frame16_off; int cpu = smp_processor_id(); - /* reserve the space on 32bit stack for the magic switch16 pointer */ - memmove(stk, stk + 8, sizeof(struct pt_regs)); - switch16_ptr = (unsigned long *)(stk + sizeof(struct pt_regs)); - regs = (struct pt_regs *)stk; - /* now the switch32 on 16bit stack */ - stack_bot = (unsigned long)&per_cpu(cpu_16bit_stack, cpu); - stack_top = stack_bot + CPU_16BIT_STACK_SIZE; - switch32_ptr = (unsigned long *)(stack_top - 8); - iret_frame16_off = CPU_16BIT_STACK_SIZE - 8 - 20; - /* copy iret frame on 16bit stack */ - memcpy((void *)(stack_bot + iret_frame16_off), ®s->eip, 20); - /* fill in the switch pointers */ - switch16_ptr[0] = (regs->esp & 0xffff0000) | iret_frame16_off; - switch16_ptr[1] = __ESPFIX_SS; - switch32_ptr[0] = (unsigned long)stk + sizeof(struct pt_regs) + - 8 - CPU_16BIT_STACK_SIZE; - switch32_ptr[1] = __KERNEL_DS; -} - -fastcall unsigned char * fixup_x86_bogus_stack(unsigned short sp) -{ - unsigned long *switch32_ptr; - unsigned char *stack16, *stack32; - unsigned long stack_top, stack_bot; - int len; - int cpu = smp_processor_id(); - stack_bot = (unsigned long)&per_cpu(cpu_16bit_stack, cpu); - stack_top = stack_bot + CPU_16BIT_STACK_SIZE; - switch32_ptr = (unsigned long *)(stack_top - 8); - /* copy the data from 16bit stack to 32bit stack */ - len = CPU_16BIT_STACK_SIZE - 8 - sp; - stack16 = (unsigned char *)(stack_bot + sp); - stack32 = (unsigned char *) - (switch32_ptr[0] + CPU_16BIT_STACK_SIZE - 8 - len); - memcpy(stack32, stack16, len); - return stack32; + struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); + struct desc_struct *gdt = (struct desc_struct *)cpu_gdt_descr->address; + unsigned long base = (kesp - uesp) & -THREAD_SIZE; + unsigned long new_kesp = kesp - base; + unsigned long lim_pages = (new_kesp | (THREAD_SIZE - 1)) >> PAGE_SHIFT; + __u64 desc = *(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS]; + /* Set up base for espfix segment */ + desc &= 0x00f0ff0000000000ULL; + desc |= ((((__u64)base) << 16) & 0x000000ffffff0000ULL) | + ((((__u64)base) << 32) & 0xff00000000000000ULL) | + ((((__u64)lim_pages) << 32) & 0x000f000000000000ULL) | + (lim_pages & 0xffff); + *(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS] = desc; + return new_kesp; } /* diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h index a0398f780ca..6cf2ac2bfde 100644 --- a/include/asm-i386/desc.h +++ b/include/asm-i386/desc.h @@ -4,8 +4,6 @@ #include #include -#define CPU_16BIT_STACK_SIZE 1024 - #ifndef __ASSEMBLY__ #include @@ -16,8 +14,6 @@ extern struct desc_struct cpu_gdt_table[GDT_ENTRIES]; -DECLARE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]); - struct Xgt_desc_struct { unsigned short size; unsigned long address __attribute__((packed)); @@ -181,6 +177,29 @@ static inline unsigned long get_desc_base(unsigned long *desc) return base; } +#else /* __ASSEMBLY__ */ + +/* + * GET_DESC_BASE reads the descriptor base of the specified segment. + * + * Args: + * idx - descriptor index + * gdt - GDT pointer + * base - 32bit register to which the base will be written + * lo_w - lo word of the "base" register + * lo_b - lo byte of the "base" register + * hi_b - hi byte of the low word of the "base" register + * + * Example: + * GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah) + * Will read the base address of GDT_ENTRY_ESPFIX_SS and put it into %eax. + */ +#define GET_DESC_BASE(idx, gdt, base, lo_w, lo_b, hi_b) \ + movb idx*8+4(gdt), lo_b; \ + movb idx*8+7(gdt), hi_b; \ + shll $16, base; \ + movw idx*8+2(gdt), lo_w; + #endif /* !__ASSEMBLY__ */ #endif diff --git a/include/asm-i386/percpu.h b/include/asm-i386/percpu.h index 5764afa4b6a..510ae1d3486 100644 --- a/include/asm-i386/percpu.h +++ b/include/asm-i386/percpu.h @@ -1,6 +1,31 @@ #ifndef __ARCH_I386_PERCPU__ #define __ARCH_I386_PERCPU__ +#ifndef __ASSEMBLY__ #include +#else + +/* + * PER_CPU finds an address of a per-cpu variable. + * + * Args: + * var - variable name + * cpu - 32bit register containing the current CPU number + * + * The resulting address is stored in the "cpu" argument. + * + * Example: + * PER_CPU(cpu_gdt_descr, %ebx) + */ +#ifdef CONFIG_SMP +#define PER_CPU(var, cpu) \ + movl __per_cpu_offset(,cpu,4), cpu; \ + addl $per_cpu__/**/var, cpu; +#else /* ! SMP */ +#define PER_CPU(var, cpu) \ + movl $per_cpu__/**/var, cpu; +#endif /* SMP */ + +#endif /* !__ASSEMBLY__ */ #endif /* __ARCH_I386_PERCPU__ */ -- cgit v1.2.3-70-g09d2 From acc207616a91a413a50fdd8847a747c4a7324167 Mon Sep 17 00:00:00 2001 From: Chuck Ebbert <76306.1226@compuserve.com> Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] i386: add sleazy FPU optimization i386 port of the sLeAZY-fpu feature. Chuck reports that this gives him a +/- 0.4% improvement on his simple benchmark x86_64 description follows: Right now the kernel on x86-64 has a 100% lazy fpu behavior: after *every* context switch a trap is taken for the first FPU use to restore the FPU context lazily. This is of course great for applications that have very sporadic or no FPU use (since then you avoid doing the expensive save/restore all the time). However for very frequent FPU users... you take an extra trap every context switch. The patch below adds a simple heuristic to this code: After 5 consecutive context switches of FPU use, the lazy behavior is disabled and the context gets restored every context switch. If the app indeed uses the FPU, the trap is avoided. (the chance of the 6th time slice using FPU after the previous 5 having done so are quite high obviously). After 256 switches, this is reset and lazy behavior is returned (until there are 5 consecutive ones again). The reason for this is to give apps that do longer bursts of FPU use still the lazy behavior back after some time. Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com> Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/kernel/process.c | 12 ++++++++++++ arch/i386/kernel/traps.c | 3 ++- include/asm-i386/i387.h | 5 ++++- 3 files changed, 18 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index dd53c58f64f..ae924c416b6 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -648,6 +648,11 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas __unlazy_fpu(prev_p); + + /* we're going to use this soon, after a few expensive things */ + if (next_p->fpu_counter > 5) + prefetch(&next->i387.fxsave); + /* * Reload esp0. */ @@ -697,6 +702,13 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas disable_tsc(prev_p, next_p); + /* If the task has used fpu the last 5 timeslices, just do a full + * restore of the math state immediately to avoid the trap; the + * chances of needing FPU soon are obviously high now + */ + if (next_p->fpu_counter > 5) + math_state_restore(); + return prev_p; } diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index f9bb1f89d68..4a6fa2837df 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -1118,7 +1118,7 @@ fastcall unsigned long patch_espfix_desc(unsigned long uesp, * Must be called with kernel preemption disabled (in this case, * local interrupts are disabled at the call-site in entry.S). */ -asmlinkage void math_state_restore(struct pt_regs regs) +asmlinkage void math_state_restore(void) { struct thread_info *thread = current_thread_info(); struct task_struct *tsk = thread->task; @@ -1128,6 +1128,7 @@ asmlinkage void math_state_restore(struct pt_regs regs) init_fpu(tsk); restore_fpu(tsk); thread->status |= TS_USEDFPU; /* So we fnsave on switch_to() */ + tsk->fpu_counter++; } #ifndef CONFIG_MATH_EMULATION diff --git a/include/asm-i386/i387.h b/include/asm-i386/i387.h index bc1d6edae1e..434936c732d 100644 --- a/include/asm-i386/i387.h +++ b/include/asm-i386/i387.h @@ -76,7 +76,9 @@ static inline void __save_init_fpu( struct task_struct *tsk ) #define __unlazy_fpu( tsk ) do { \ if (task_thread_info(tsk)->status & TS_USEDFPU) \ - save_init_fpu( tsk ); \ + save_init_fpu( tsk ); \ + else \ + tsk->fpu_counter = 0; \ } while (0) #define __clear_fpu( tsk ) \ @@ -118,6 +120,7 @@ static inline void save_init_fpu( struct task_struct *tsk ) extern unsigned short get_fpu_cwd( struct task_struct *tsk ); extern unsigned short get_fpu_swd( struct task_struct *tsk ); extern unsigned short get_fpu_mxcsr( struct task_struct *tsk ); +extern asmlinkage void math_state_restore(void); /* * Signal frame handlers... -- cgit v1.2.3-70-g09d2 From c0e84b9901c0924e2503c0aab3772a4469ba4aef Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Thu, 7 Dec 2006 02:14:02 +0100 Subject: [PATCH] i386: Add iounmap in error paths in hpet code Signed-off-by: Amol Lad Signed-off-by: Andi Kleen --- arch/i386/kernel/time_hpet.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/time_hpet.c b/arch/i386/kernel/time_hpet.c index 1a2a979cf6a..1e4702dfcd0 100644 --- a/arch/i386/kernel/time_hpet.c +++ b/arch/i386/kernel/time_hpet.c @@ -132,14 +132,20 @@ int __init hpet_enable(void) * the single HPET timer for system time. */ #ifdef CONFIG_HPET_EMULATE_RTC - if (!(id & HPET_ID_NUMBER)) + if (!(id & HPET_ID_NUMBER)) { + iounmap(hpet_virt_address); + hpet_virt_address = NULL; return -1; + } #endif hpet_period = hpet_readl(HPET_PERIOD); - if ((hpet_period < HPET_MIN_PERIOD) || (hpet_period > HPET_MAX_PERIOD)) + if ((hpet_period < HPET_MIN_PERIOD) || (hpet_period > HPET_MAX_PERIOD)) { + iounmap(hpet_virt_address); + hpet_virt_address = NULL; return -1; + } /* * 64 bit math @@ -156,8 +162,11 @@ int __init hpet_enable(void) hpet_use_timer = id & HPET_ID_LEGSUP; - if (hpet_timer_stop_set_go(hpet_tick)) + if (hpet_timer_stop_set_go(hpet_tick)) { + iounmap(hpet_virt_address); + hpet_virt_address = NULL; return -1; + } use_hpet = 1; -- cgit v1.2.3-70-g09d2 From fa5cecd111d235819a1d807d43216ae459a0dd6f Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Thu, 7 Dec 2006 02:14:02 +0100 Subject: [PATCH] i386: add missing iounmap in i386 hpet clocksource code ioremap must be balanced by an iounmap and failing to do so can result in a memory leak. Tested (compilation only): - using allmodconfig - making sure the files are compiling without any warning/error due to new changes Signed-off-by: Amol Lad Signed-off-by: Andi Kleen --- arch/i386/kernel/hpet.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/hpet.c b/arch/i386/kernel/hpet.c index 17647a530b2..45a8685bb60 100644 --- a/arch/i386/kernel/hpet.c +++ b/arch/i386/kernel/hpet.c @@ -34,6 +34,7 @@ static int __init init_hpet_clocksource(void) unsigned long hpet_period; void __iomem* hpet_base; u64 tmp; + int err; if (!is_hpet_enabled()) return -ENODEV; @@ -61,7 +62,11 @@ static int __init init_hpet_clocksource(void) do_div(tmp, FSEC_PER_NSEC); clocksource_hpet.mult = (u32)tmp; - return clocksource_register(&clocksource_hpet); + err = clocksource_register(&clocksource_hpet); + if (err) + iounmap(hpet_base); + + return err; } module_init(init_hpet_clocksource); -- cgit v1.2.3-70-g09d2 From eb5b7b9d86f46b45ba1f986302fdf7df84fb8297 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 7 Dec 2006 02:14:02 +0100 Subject: [PATCH] i386: Use asm-offsets for the offsets of registers into the pt_regs struct Use asm-offsets for the offsets of registers into the pt_regs struct, rather than having hard-coded constants I left the constants in the comments of entry.S because they're useful for reference; the code in entry.S is very dependent on the layout of pt_regs, even when using asm-offsets. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Keith Owens Signed-off-by: Andrew Morton --- arch/i386/kernel/asm-offsets.c | 17 ++++++ arch/i386/kernel/entry.S | 120 ++++++++++++++++++----------------------- 2 files changed, 69 insertions(+), 68 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index e94d910a28b..70b19807acf 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -63,6 +63,23 @@ void foo(void) OFFSET(GDS_pad, Xgt_desc_struct, pad); BLANK(); + OFFSET(PT_EBX, pt_regs, ebx); + OFFSET(PT_ECX, pt_regs, ecx); + OFFSET(PT_EDX, pt_regs, edx); + OFFSET(PT_ESI, pt_regs, esi); + OFFSET(PT_EDI, pt_regs, edi); + OFFSET(PT_EBP, pt_regs, ebp); + OFFSET(PT_EAX, pt_regs, eax); + OFFSET(PT_DS, pt_regs, xds); + OFFSET(PT_ES, pt_regs, xes); + OFFSET(PT_ORIG_EAX, pt_regs, orig_eax); + OFFSET(PT_EIP, pt_regs, eip); + OFFSET(PT_CS, pt_regs, xcs); + OFFSET(PT_EFLAGS, pt_regs, eflags); + OFFSET(PT_OLDESP, pt_regs, esp); + OFFSET(PT_OLDSS, pt_regs, xss); + BLANK(); + OFFSET(EXEC_DOMAIN_handler, exec_domain, handler); OFFSET(RT_SIGFRAME_sigcontext, rt_sigframe, uc.uc_mcontext); BLANK(); diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index c38d801ba0b..0069bf01603 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -54,22 +54,6 @@ #define nr_syscalls ((syscall_table_size)/4) -EBX = 0x00 -ECX = 0x04 -EDX = 0x08 -ESI = 0x0C -EDI = 0x10 -EBP = 0x14 -EAX = 0x18 -DS = 0x1C -ES = 0x20 -ORIG_EAX = 0x24 -EIP = 0x28 -CS = 0x2C -EFLAGS = 0x30 -OLDESP = 0x34 -OLDSS = 0x38 - CF_MASK = 0x00000001 TF_MASK = 0x00000100 IF_MASK = 0x00000200 @@ -93,7 +77,7 @@ VM_MASK = 0x00020000 .macro TRACE_IRQS_IRET #ifdef CONFIG_TRACE_IRQFLAGS - testl $IF_MASK,EFLAGS(%esp) # interrupts off? + testl $IF_MASK,PT_EFLAGS(%esp) # interrupts off? jz 1f TRACE_IRQS_ON 1: @@ -199,18 +183,18 @@ VM_MASK = 0x00020000 #define RING0_PTREGS_FRAME \ CFI_STARTPROC simple;\ CFI_SIGNAL_FRAME;\ - CFI_DEF_CFA esp, OLDESP-EBX;\ - /*CFI_OFFSET cs, CS-OLDESP;*/\ - CFI_OFFSET eip, EIP-OLDESP;\ - /*CFI_OFFSET es, ES-OLDESP;*/\ - /*CFI_OFFSET ds, DS-OLDESP;*/\ - CFI_OFFSET eax, EAX-OLDESP;\ - CFI_OFFSET ebp, EBP-OLDESP;\ - CFI_OFFSET edi, EDI-OLDESP;\ - CFI_OFFSET esi, ESI-OLDESP;\ - CFI_OFFSET edx, EDX-OLDESP;\ - CFI_OFFSET ecx, ECX-OLDESP;\ - CFI_OFFSET ebx, EBX-OLDESP + CFI_DEF_CFA esp, PT_OLDESP-PT_EBX;\ + /*CFI_OFFSET cs, PT_CS-PT_OLDESP;*/\ + CFI_OFFSET eip, PT_EIP-PT_OLDESP;\ + /*CFI_OFFSET es, PT_ES-PT_OLDESP;*/\ + /*CFI_OFFSET ds, PT_DS-PT_OLDESP;*/\ + CFI_OFFSET eax, PT_EAX-PT_OLDESP;\ + CFI_OFFSET ebp, PT_EBP-PT_OLDESP;\ + CFI_OFFSET edi, PT_EDI-PT_OLDESP;\ + CFI_OFFSET esi, PT_ESI-PT_OLDESP;\ + CFI_OFFSET edx, PT_EDX-PT_OLDESP;\ + CFI_OFFSET ecx, PT_ECX-PT_OLDESP;\ + CFI_OFFSET ebx, PT_EBX-PT_OLDESP ENTRY(ret_from_fork) CFI_STARTPROC @@ -242,8 +226,8 @@ ret_from_exception: ret_from_intr: GET_THREAD_INFO(%ebp) check_userspace: - movl EFLAGS(%esp), %eax # mix EFLAGS and CS - movb CS(%esp), %al + movl PT_EFLAGS(%esp), %eax # mix EFLAGS and CS + movb PT_CS(%esp), %al andl $(VM_MASK | SEGMENT_RPL_MASK), %eax cmpl $USER_RPL, %eax jb resume_kernel # not returning to v8086 or userspace @@ -266,7 +250,7 @@ need_resched: movl TI_flags(%ebp), %ecx # need_resched set ? testb $_TIF_NEED_RESCHED, %cl jz restore_all - testl $IF_MASK,EFLAGS(%esp) # interrupts off (exception path) ? + testl $IF_MASK,PT_EFLAGS(%esp) # interrupts off (exception path) ? jz restore_all call preempt_schedule_irq jmp need_resched @@ -332,15 +316,15 @@ sysenter_past_esp: cmpl $(nr_syscalls), %eax jae syscall_badsys call *sys_call_table(,%eax,4) - movl %eax,EAX(%esp) + movl %eax,PT_EAX(%esp) DISABLE_INTERRUPTS TRACE_IRQS_OFF movl TI_flags(%ebp), %ecx testw $_TIF_ALLWORK_MASK, %cx jne syscall_exit_work /* if something modifies registers it must also disable sysexit */ - movl EIP(%esp), %edx - movl OLDESP(%esp), %ecx + movl PT_EIP(%esp), %edx + movl PT_OLDESP(%esp), %ecx xorl %ebp,%ebp TRACE_IRQS_ON ENABLE_INTERRUPTS_SYSEXIT @@ -354,7 +338,7 @@ ENTRY(system_call) CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL GET_THREAD_INFO(%ebp) - testl $TF_MASK,EFLAGS(%esp) + testl $TF_MASK,PT_EFLAGS(%esp) jz no_singlestep orl $_TIF_SINGLESTEP,TI_flags(%ebp) no_singlestep: @@ -366,7 +350,7 @@ no_singlestep: jae syscall_badsys syscall_call: call *sys_call_table(,%eax,4) - movl %eax,EAX(%esp) # store the return value + movl %eax,PT_EAX(%esp) # store the return value syscall_exit: DISABLE_INTERRUPTS # make sure we don't miss an interrupt # setting need_resched or sigpending @@ -377,12 +361,12 @@ syscall_exit: jne syscall_exit_work restore_all: - movl EFLAGS(%esp), %eax # mix EFLAGS, SS and CS - # Warning: OLDSS(%esp) contains the wrong/random values if we + movl PT_EFLAGS(%esp), %eax # mix EFLAGS, SS and CS + # Warning: PT_OLDSS(%esp) contains the wrong/random values if we # are returning to the kernel. # See comments in process.c:copy_thread() for details. - movb OLDSS(%esp), %ah - movb CS(%esp), %al + movb PT_OLDSS(%esp), %ah + movb PT_CS(%esp), %al andl $(VM_MASK | (SEGMENT_TI_MASK << 8) | SEGMENT_RPL_MASK), %eax cmpl $((SEGMENT_LDT << 8) | USER_RPL), %eax CFI_REMEMBER_STATE @@ -409,7 +393,7 @@ iret_exc: CFI_RESTORE_STATE ldt_ss: - larl OLDSS(%esp), %eax + larl PT_OLDSS(%esp), %eax jnz restore_nocheck testl $0x00400000, %eax # returning to 32bit stack? jnz restore_nocheck # allright, normal return @@ -419,7 +403,7 @@ ldt_ss: * This is an "official" bug of all the x86-compatible * CPUs, which we can try to work around to make * dosemu and wine happy. */ - movl OLDESP(%esp), %eax + movl PT_OLDESP(%esp), %eax movl %esp, %edx call patch_espfix_desc pushl $__ESPFIX_SS @@ -454,7 +438,7 @@ work_resched: work_notifysig: # deal with pending signals and # notify-resume requests - testl $VM_MASK, EFLAGS(%esp) + testl $VM_MASK, PT_EFLAGS(%esp) movl %esp, %eax jne work_notifysig_v86 # returning to kernel-space or # vm86-space @@ -479,14 +463,14 @@ work_notifysig_v86: # perform syscall exit tracing ALIGN syscall_trace_entry: - movl $-ENOSYS,EAX(%esp) + movl $-ENOSYS,PT_EAX(%esp) movl %esp, %eax xorl %edx,%edx call do_syscall_trace cmpl $0, %eax jne resume_userspace # ret != 0 -> running under PTRACE_SYSEMU, # so must skip actual syscall - movl ORIG_EAX(%esp), %eax + movl PT_ORIG_EAX(%esp), %eax cmpl $(nr_syscalls), %eax jnae syscall_call jmp syscall_exit @@ -511,11 +495,11 @@ syscall_fault: CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL GET_THREAD_INFO(%ebp) - movl $-EFAULT,EAX(%esp) + movl $-EFAULT,PT_EAX(%esp) jmp resume_userspace syscall_badsys: - movl $-ENOSYS,EAX(%esp) + movl $-ENOSYS,PT_EAX(%esp) jmp resume_userspace CFI_ENDPROC @@ -636,10 +620,10 @@ error_code: popl %ecx CFI_ADJUST_CFA_OFFSET -4 /*CFI_REGISTER es, ecx*/ - movl ES(%esp), %edi # get the function address - movl ORIG_EAX(%esp), %edx # get the error code - movl $-1, ORIG_EAX(%esp) - movl %ecx, ES(%esp) + movl PT_ES(%esp), %edi # get the function address + movl PT_ORIG_EAX(%esp), %edx # get the error code + movl $-1, PT_ORIG_EAX(%esp) + movl %ecx, PT_ES(%esp) /*CFI_REL_OFFSET es, ES*/ movl $(__USER_DS), %ecx movl %ecx, %ds @@ -942,26 +926,26 @@ ENTRY(arch_unwind_init_running) movl 4(%esp), %edx movl (%esp), %ecx leal 4(%esp), %eax - movl %ebx, EBX(%edx) + movl %ebx, PT_EBX(%edx) xorl %ebx, %ebx - movl %ebx, ECX(%edx) - movl %ebx, EDX(%edx) - movl %esi, ESI(%edx) - movl %edi, EDI(%edx) - movl %ebp, EBP(%edx) - movl %ebx, EAX(%edx) - movl $__USER_DS, DS(%edx) - movl $__USER_DS, ES(%edx) - movl %ebx, ORIG_EAX(%edx) - movl %ecx, EIP(%edx) + movl %ebx, PT_ECX(%edx) + movl %ebx, PT_EDX(%edx) + movl %esi, PT_ESI(%edx) + movl %edi, PT_EDI(%edx) + movl %ebp, PT_EBP(%edx) + movl %ebx, PT_EAX(%edx) + movl $__USER_DS, PT_DS(%edx) + movl $__USER_DS, PT_ES(%edx) + movl %ebx, PT_ORIG_EAX(%edx) + movl %ecx, PT_EIP(%edx) movl 12(%esp), %ecx - movl $__KERNEL_CS, CS(%edx) - movl %ebx, EFLAGS(%edx) - movl %eax, OLDESP(%edx) + movl $__KERNEL_CS, PT_CS(%edx) + movl %ebx, PT_EFLAGS(%edx) + movl %eax, PT_OLDESP(%edx) movl 8(%esp), %eax movl %ecx, 8(%esp) - movl EBX(%edx), %ebx - movl $__KERNEL_DS, OLDSS(%edx) + movl PT_EBX(%edx), %ebx + movl $__KERNEL_DS, PT_OLDSS(%edx) jmpl *%eax CFI_ENDPROC ENDPROC(arch_unwind_init_running) -- cgit v1.2.3-70-g09d2 From 9ca36101a8d74704d78f10910f89d62de96f9dc8 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 7 Dec 2006 02:14:02 +0100 Subject: [PATCH] i386: Basic definitions for i386-pda This patch has the basic definitions of struct i386_pda, and the segment selector in the GDT. asm-i386/pda.h is more or less a direct copy of asm-x86_64/pda.h. The most interesting difference is the use of _proxy_pda, which is used to give gcc a model for the actual memory operations on the real pda structure. No actual reference is ever made to _proxy_pda, so it is never defined. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Chuck Ebbert <76306.1226@compuserve.com> Cc: Zachary Amsden Cc: Jan Beulich Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/head.S | 2 +- include/asm-i386/pda.h | 95 ++++++++++++++++++++++++++++++++++++++++++++++ include/asm-i386/segment.h | 5 ++- 3 files changed, 100 insertions(+), 2 deletions(-) create mode 100644 include/asm-i386/pda.h (limited to 'arch/i386') diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index b1f1df11fcc..4a83384c5a6 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -585,7 +585,7 @@ ENTRY(cpu_gdt_table) .quad 0x004092000000ffff /* 0xc8 APM DS data */ .quad 0x00c0920000000000 /* 0xd0 - ESPFIX SS */ - .quad 0x0000000000000000 /* 0xd8 - unused */ + .quad 0x0000000000000000 /* 0xd8 - PDA */ .quad 0x0000000000000000 /* 0xe0 - unused */ .quad 0x0000000000000000 /* 0xe8 - unused */ .quad 0x0000000000000000 /* 0xf0 - unused */ diff --git a/include/asm-i386/pda.h b/include/asm-i386/pda.h new file mode 100644 index 00000000000..4c39ccb1305 --- /dev/null +++ b/include/asm-i386/pda.h @@ -0,0 +1,95 @@ +/* + Per-processor Data Areas + Jeremy Fitzhardinge 2006 + Based on asm-x86_64/pda.h by Andi Kleen. + */ +#ifndef _I386_PDA_H +#define _I386_PDA_H + +#include + +struct i386_pda +{ + struct i386_pda *_pda; /* pointer to self */ +}; + +extern struct i386_pda *_cpu_pda[]; + +#define cpu_pda(i) (_cpu_pda[i]) + +#define pda_offset(field) offsetof(struct i386_pda, field) + +extern void __bad_pda_field(void); + +/* This variable is never instantiated. It is only used as a stand-in + for the real per-cpu PDA memory, so that gcc can understand what + memory operations the inline asms() below are performing. This + eliminates the need to make the asms volatile or have memory + clobbers, so gcc can readily analyse them. */ +extern struct i386_pda _proxy_pda; + +#define pda_to_op(op,field,val) \ + do { \ + typedef typeof(_proxy_pda.field) T__; \ + if (0) { T__ tmp__; tmp__ = (val); } \ + switch (sizeof(_proxy_pda.field)) { \ + case 1: \ + asm(op "b %1,%%gs:%c2" \ + : "+m" (_proxy_pda.field) \ + :"ri" ((T__)val), \ + "i"(pda_offset(field))); \ + break; \ + case 2: \ + asm(op "w %1,%%gs:%c2" \ + : "+m" (_proxy_pda.field) \ + :"ri" ((T__)val), \ + "i"(pda_offset(field))); \ + break; \ + case 4: \ + asm(op "l %1,%%gs:%c2" \ + : "+m" (_proxy_pda.field) \ + :"ri" ((T__)val), \ + "i"(pda_offset(field))); \ + break; \ + default: __bad_pda_field(); \ + } \ + } while (0) + +#define pda_from_op(op,field) \ + ({ \ + typeof(_proxy_pda.field) ret__; \ + switch (sizeof(_proxy_pda.field)) { \ + case 1: \ + asm(op "b %%gs:%c1,%0" \ + : "=r" (ret__) \ + : "i" (pda_offset(field)), \ + "m" (_proxy_pda.field)); \ + break; \ + case 2: \ + asm(op "w %%gs:%c1,%0" \ + : "=r" (ret__) \ + : "i" (pda_offset(field)), \ + "m" (_proxy_pda.field)); \ + break; \ + case 4: \ + asm(op "l %%gs:%c1,%0" \ + : "=r" (ret__) \ + : "i" (pda_offset(field)), \ + "m" (_proxy_pda.field)); \ + break; \ + default: __bad_pda_field(); \ + } \ + ret__; }) + +/* Return a pointer to a pda field */ +#define pda_addr(field) \ + ((typeof(_proxy_pda.field) *)((unsigned char *)read_pda(_pda) + \ + pda_offset(field))) + +#define read_pda(field) pda_from_op("mov",field) +#define write_pda(field,val) pda_to_op("mov",field,val) +#define add_pda(field,val) pda_to_op("add",field,val) +#define sub_pda(field,val) pda_to_op("sub",field,val) +#define or_pda(field,val) pda_to_op("or",field,val) + +#endif /* _I386_PDA_H */ diff --git a/include/asm-i386/segment.h b/include/asm-i386/segment.h index b7ab59685ba..5bdda79b6b5 100644 --- a/include/asm-i386/segment.h +++ b/include/asm-i386/segment.h @@ -39,7 +39,7 @@ * 25 - APM BIOS support * * 26 - ESPFIX small SS - * 27 - unused + * 27 - PDA [ per-cpu private data area ] * 28 - unused * 29 - unused * 30 - unused @@ -74,6 +74,9 @@ #define GDT_ENTRY_ESPFIX_SS (GDT_ENTRY_KERNEL_BASE + 14) #define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8) +#define GDT_ENTRY_PDA (GDT_ENTRY_KERNEL_BASE + 15) +#define __KERNEL_PDA (GDT_ENTRY_PDA * 8) + #define GDT_ENTRY_DOUBLEFAULT_TSS 31 /* -- cgit v1.2.3-70-g09d2 From 62111195800d80c66cdc69063ea3145878c99fbf Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 7 Dec 2006 02:14:02 +0100 Subject: [PATCH] i386: Initialize the per-CPU data area When a CPU is brought up, a PDA and GDT are allocated for it. The GDT's __KERNEL_PDA entry is pointed to the allocated PDA memory, so that all references using this segment descriptor will refer to the PDA. This patch rearranges CPU initialization a bit, so that the GDT/PDA are set up as early as possible in cpu_init(). Also for secondary CPUs, GDT+PDA are preallocated and initialized so all the secondary CPU needs to do is set up the ldt and load %gs. This will be important once smp_processor_id() and current use the PDA. In all cases, the PDA is set up in head.S, before a CPU starts running C code, so the PDA is always available. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Chuck Ebbert <76306.1226@compuserve.com> Cc: Zachary Amsden Cc: Jan Beulich Cc: Andi Kleen Cc: James Bottomley Cc: Matt Tolentino Signed-off-by: Andrew Morton --- arch/i386/kernel/cpu/common.c | 177 ++++++++++++++++++++++++++--------- arch/i386/kernel/smpboot.c | 28 ++++-- arch/i386/mach-voyager/voyager_smp.c | 14 ++- include/asm-i386/processor.h | 3 + 4 files changed, 172 insertions(+), 50 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index 5532fc4e1bf..2534e25ed74 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -18,12 +18,16 @@ #include #include #endif +#include #include "cpu.h" DEFINE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr); EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr); +struct i386_pda *_cpu_pda[NR_CPUS] __read_mostly; +EXPORT_SYMBOL(_cpu_pda); + static int cachesize_override __cpuinitdata = -1; static int disable_x86_fxsr __cpuinitdata; static int disable_x86_serial_nr __cpuinitdata = 1; @@ -588,41 +592,16 @@ void __init early_cpu_init(void) disable_pse = 1; #endif } -/* - * cpu_init() initializes state that is per-CPU. Some data is already - * initialized (naturally) in the bootstrap process, such as the GDT - * and IDT. We reload them nevertheless, this function acts as a - * 'CPU state barrier', nothing should get across. - */ -void __cpuinit cpu_init(void) + +__cpuinit int alloc_gdt(int cpu) { - int cpu = smp_processor_id(); - struct tss_struct * t = &per_cpu(init_tss, cpu); - struct thread_struct *thread = ¤t->thread; - struct desc_struct *gdt; struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); + struct desc_struct *gdt; + struct i386_pda *pda; - if (cpu_test_and_set(cpu, cpu_initialized)) { - printk(KERN_WARNING "CPU#%d already initialized!\n", cpu); - for (;;) local_irq_enable(); - } - printk(KERN_INFO "Initializing CPU#%d\n", cpu); + gdt = (struct desc_struct *)cpu_gdt_descr->address; + pda = cpu_pda(cpu); - if (cpu_has_vme || cpu_has_tsc || cpu_has_de) - clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE); - if (tsc_disable && cpu_has_tsc) { - printk(KERN_NOTICE "Disabling TSC...\n"); - /**** FIX-HPA: DOES THIS REALLY BELONG HERE? ****/ - clear_bit(X86_FEATURE_TSC, boot_cpu_data.x86_capability); - set_in_cr4(X86_CR4_TSD); - } - - /* The CPU hotplug case */ - if (cpu_gdt_descr->address) { - gdt = (struct desc_struct *)cpu_gdt_descr->address; - memset(gdt, 0, PAGE_SIZE); - goto old_gdt; - } /* * This is a horrible hack to allocate the GDT. The problem * is that cpu_init() is called really early for the boot CPU @@ -630,36 +609,117 @@ void __cpuinit cpu_init(void) * CPUs, when bootmem will have gone away */ if (NODE_DATA(0)->bdata->node_bootmem_map) { - gdt = (struct desc_struct *)alloc_bootmem_pages(PAGE_SIZE); - /* alloc_bootmem_pages panics on failure, so no check */ + BUG_ON(gdt != NULL || pda != NULL); + + gdt = alloc_bootmem_pages(PAGE_SIZE); + pda = alloc_bootmem(sizeof(*pda)); + /* alloc_bootmem(_pages) panics on failure, so no check */ + memset(gdt, 0, PAGE_SIZE); + memset(pda, 0, sizeof(*pda)); } else { - gdt = (struct desc_struct *)get_zeroed_page(GFP_KERNEL); - if (unlikely(!gdt)) { - printk(KERN_CRIT "CPU%d failed to allocate GDT\n", cpu); - for (;;) - local_irq_enable(); + /* GDT and PDA might already have been allocated if + this is a CPU hotplug re-insertion. */ + if (gdt == NULL) + gdt = (struct desc_struct *)get_zeroed_page(GFP_KERNEL); + + if (pda == NULL) + pda = kmalloc_node(sizeof(*pda), GFP_KERNEL, cpu_to_node(cpu)); + + if (unlikely(!gdt || !pda)) { + free_pages((unsigned long)gdt, 0); + kfree(pda); + return 0; } } -old_gdt: + + cpu_gdt_descr->address = (unsigned long)gdt; + cpu_pda(cpu) = pda; + + return 1; +} + +/* Initial PDA used by boot CPU */ +struct i386_pda boot_pda = { + ._pda = &boot_pda, +}; + +/* Initialize the CPU's GDT and PDA. The boot CPU does this for + itself, but secondaries find this done for them. */ +__cpuinit int init_gdt(int cpu, struct task_struct *idle) +{ + struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); + struct desc_struct *gdt; + struct i386_pda *pda; + + /* For non-boot CPUs, the GDT and PDA should already have been + allocated. */ + if (!alloc_gdt(cpu)) { + printk(KERN_CRIT "CPU%d failed to allocate GDT or PDA\n", cpu); + return 0; + } + + gdt = (struct desc_struct *)cpu_gdt_descr->address; + pda = cpu_pda(cpu); + + BUG_ON(gdt == NULL || pda == NULL); + /* * Initialize the per-CPU GDT with the boot GDT, * and set up the GDT descriptor: */ memcpy(gdt, cpu_gdt_table, GDT_SIZE); cpu_gdt_descr->size = GDT_SIZE - 1; - cpu_gdt_descr->address = (unsigned long)gdt; + pack_descriptor((u32 *)&gdt[GDT_ENTRY_PDA].a, + (u32 *)&gdt[GDT_ENTRY_PDA].b, + (unsigned long)pda, sizeof(*pda) - 1, + 0x80 | DESCTYPE_S | 0x2, 0); /* present read-write data segment */ + + memset(pda, 0, sizeof(*pda)); + pda->_pda = pda; + + return 1; +} + +/* Common CPU init for both boot and secondary CPUs */ +static void __cpuinit _cpu_init(int cpu, struct task_struct *curr) +{ + struct tss_struct * t = &per_cpu(init_tss, cpu); + struct thread_struct *thread = &curr->thread; + struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); + + /* Reinit these anyway, even if they've already been done (on + the boot CPU, this will transition from the boot gdt+pda to + the real ones). */ load_gdt(cpu_gdt_descr); + + if (cpu_test_and_set(cpu, cpu_initialized)) { + printk(KERN_WARNING "CPU#%d already initialized!\n", cpu); + for (;;) local_irq_enable(); + } + + printk(KERN_INFO "Initializing CPU#%d\n", cpu); + + if (cpu_has_vme || cpu_has_tsc || cpu_has_de) + clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE); + if (tsc_disable && cpu_has_tsc) { + printk(KERN_NOTICE "Disabling TSC...\n"); + /**** FIX-HPA: DOES THIS REALLY BELONG HERE? ****/ + clear_bit(X86_FEATURE_TSC, boot_cpu_data.x86_capability); + set_in_cr4(X86_CR4_TSD); + } + load_idt(&idt_descr); /* * Set up and load the per-CPU TSS and LDT */ atomic_inc(&init_mm.mm_count); - current->active_mm = &init_mm; - BUG_ON(current->mm); - enter_lazy_tlb(&init_mm, current); + curr->active_mm = &init_mm; + if (curr->mm) + BUG(); + enter_lazy_tlb(&init_mm, curr); load_esp0(t, thread); set_tss_desc(cpu,t); @@ -690,6 +750,37 @@ old_gdt: mxcsr_feature_mask_init(); } +/* Entrypoint to initialize secondary CPU */ +void __cpuinit secondary_cpu_init(void) +{ + int cpu = smp_processor_id(); + struct task_struct *curr = current; + + _cpu_init(cpu, curr); +} + +/* + * cpu_init() initializes state that is per-CPU. Some data is already + * initialized (naturally) in the bootstrap process, such as the GDT + * and IDT. We reload them nevertheless, this function acts as a + * 'CPU state barrier', nothing should get across. + */ +void __cpuinit cpu_init(void) +{ + int cpu = smp_processor_id(); + struct task_struct *curr = current; + + /* Set up the real GDT and PDA, so we can transition from the + boot versions. */ + if (!init_gdt(cpu, curr)) { + /* failed to allocate something; not much we can do... */ + for (;;) + local_irq_enable(); + } + + _cpu_init(cpu, curr); +} + #ifdef CONFIG_HOTPLUG_CPU void __cpuinit cpu_uninit(void) { diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 4bb8b77cd65..095636620fa 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -52,6 +52,7 @@ #include #include #include +#include #include #include @@ -536,11 +537,11 @@ set_cpu_sibling_map(int cpu) static void __devinit start_secondary(void *unused) { /* - * Dont put anything before smp_callin(), SMP + * Don't put *anything* before secondary_cpu_init(), SMP * booting is too fragile that we want to limit the * things done here to the most necessary things. */ - cpu_init(); + secondary_cpu_init(); preempt_disable(); smp_callin(); while (!cpu_isset(smp_processor_id(), smp_commenced_mask)) @@ -599,13 +600,16 @@ void __devinit initialize_secondary(void) "movl %0,%%esp\n\t" "jmp *%1" : - :"r" (current->thread.esp),"r" (current->thread.eip)); + :"m" (current->thread.esp),"m" (current->thread.eip)); } +/* Static state in head.S used to set up a CPU */ extern struct { void * esp; unsigned short ss; } stack_start; +extern struct i386_pda *start_pda; +extern struct Xgt_desc_struct cpu_gdt_descr; #ifdef CONFIG_NUMA @@ -936,9 +940,6 @@ static int __devinit do_boot_cpu(int apicid, int cpu) unsigned long start_eip; unsigned short nmi_high = 0, nmi_low = 0; - ++cpucount; - alternatives_smp_switch(1); - /* * We can't use kernel_thread since we must avoid to * reschedule the child. @@ -946,15 +947,30 @@ static int __devinit do_boot_cpu(int apicid, int cpu) idle = alloc_idle_task(cpu); if (IS_ERR(idle)) panic("failed fork for CPU %d", cpu); + + /* Pre-allocate and initialize the CPU's GDT and PDA so it + doesn't have to do any memory allocation during the + delicate CPU-bringup phase. */ + if (!init_gdt(cpu, idle)) { + printk(KERN_INFO "Couldn't allocate GDT/PDA for CPU %d\n", cpu); + return -1; /* ? */ + } + idle->thread.eip = (unsigned long) start_secondary; /* start_eip had better be page-aligned! */ start_eip = setup_trampoline(); + ++cpucount; + alternatives_smp_switch(1); + /* So we see what's up */ printk("Booting processor %d/%d eip %lx\n", cpu, apicid, start_eip); /* Stack for startup_32 can be just as for start_secondary onwards */ stack_start.esp = (void *) idle->thread.esp; + start_pda = cpu_pda(cpu); + cpu_gdt_descr = per_cpu(cpu_gdt_descr, cpu); + irq_ctx_init(cpu); x86_cpu_to_apicid[cpu] = apicid; diff --git a/arch/i386/mach-voyager/voyager_smp.c b/arch/i386/mach-voyager/voyager_smp.c index f3fea2ad50f..55428e656a3 100644 --- a/arch/i386/mach-voyager/voyager_smp.c +++ b/arch/i386/mach-voyager/voyager_smp.c @@ -28,6 +28,7 @@ #include #include #include +#include /* TLB state -- visible externally, indexed physically */ DEFINE_PER_CPU(struct tlb_state, cpu_tlbstate) ____cacheline_aligned = { &init_mm, 0 }; @@ -422,6 +423,7 @@ find_smp_config(void) VOYAGER_SUS_IN_CONTROL_PORT); current_thread_info()->cpu = boot_cpu_id; + write_pda(cpu_number, boot_cpu_id); } /* @@ -458,7 +460,7 @@ start_secondary(void *unused) /* external functions not defined in the headers */ extern void calibrate_delay(void); - cpu_init(); + secondary_cpu_init(); /* OK, we're in the routine */ ack_CPI(VIC_CPU_BOOT_CPI); @@ -578,6 +580,15 @@ do_boot_cpu(__u8 cpu) /* init_tasks (in sched.c) is indexed logically */ stack_start.esp = (void *) idle->thread.esp; + /* Pre-allocate and initialize the CPU's GDT and PDA so it + doesn't have to do any memory allocation during the + delicate CPU-bringup phase. */ + if (!init_gdt(cpu, idle)) { + printk(KERN_INFO "Couldn't allocate GDT/PDA for CPU %d\n", cpu); + cpucount--; + return; + } + irq_ctx_init(cpu); /* Note: Don't modify initial ss override */ @@ -1963,4 +1974,5 @@ void __init smp_setup_processor_id(void) { current_thread_info()->cpu = hard_smp_processor_id(); + write_pda(cpu_number, hard_smp_processor_id()); } diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h index e0ddca94d50..a9f2041c7c8 100644 --- a/include/asm-i386/processor.h +++ b/include/asm-i386/processor.h @@ -727,4 +727,7 @@ extern unsigned long boot_option_idle_override; extern void enable_sep_cpu(void); extern int sysenter_setup(void); +extern int init_gdt(int cpu, struct task_struct *idle); +extern void secondary_cpu_init(void); + #endif /* __ASM_I386_PROCESSOR_H */ -- cgit v1.2.3-70-g09d2 From f95d47caae5302a63d92be9a0292abc90e2a14e1 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 7 Dec 2006 02:14:02 +0100 Subject: [PATCH] i386: Use %gs as the PDA base-segment in the kernel This patch is the meat of the PDA change. This patch makes several related changes: 1: Most significantly, %gs is now used in the kernel. This means that on entry, the old value of %gs is saved away, and it is reloaded with __KERNEL_PDA. 2: entry.S constructs the stack in the shape of struct pt_regs, and this is passed around the kernel so that the process's saved register state can be accessed. Unfortunately struct pt_regs doesn't currently have space for %gs (or %fs). This patch extends pt_regs to add space for gs (no space is allocated for %fs, since it won't be used, and it would just complicate the code in entry.S to work around the space). 3: Because %gs is now saved on the stack like %ds, %es and the integer registers, there are a number of places where it no longer needs to be handled specially; namely context switch, and saving/restoring the register state in a signal context. 4: And since kernel threads run in kernel space and call normal kernel code, they need to be created with their %gs == __KERNEL_PDA. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Chuck Ebbert <76306.1226@compuserve.com> Cc: Zachary Amsden Cc: Jan Beulich Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/asm-offsets.c | 1 + arch/i386/kernel/cpu/common.c | 21 +++++++++++-- arch/i386/kernel/entry.S | 70 +++++++++++++++++++++++++++++------------- arch/i386/kernel/head.S | 31 ++++++++++++++++--- arch/i386/kernel/process.c | 26 ++++++++-------- arch/i386/kernel/signal.c | 6 ++-- include/asm-i386/mmu_context.h | 4 +-- include/asm-i386/processor.h | 4 ++- include/asm-i386/ptrace.h | 2 ++ kernel/fork.c | 2 +- 10 files changed, 117 insertions(+), 50 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index 70b19807acf..9620872d353 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -72,6 +72,7 @@ void foo(void) OFFSET(PT_EAX, pt_regs, eax); OFFSET(PT_DS, pt_regs, xds); OFFSET(PT_ES, pt_regs, xes); + OFFSET(PT_GS, pt_regs, xgs); OFFSET(PT_ORIG_EAX, pt_regs, orig_eax); OFFSET(PT_EIP, pt_regs, eip); OFFSET(PT_CS, pt_regs, xcs); diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index 2534e25ed74..4e63d8ce602 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -593,6 +593,14 @@ void __init early_cpu_init(void) #endif } +/* Make sure %gs is initialized properly in idle threads */ +struct pt_regs * __devinit idle_regs(struct pt_regs *regs) +{ + memset(regs, 0, sizeof(struct pt_regs)); + regs->xgs = __KERNEL_PDA; + return regs; +} + __cpuinit int alloc_gdt(int cpu) { struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); @@ -644,6 +652,14 @@ struct i386_pda boot_pda = { ._pda = &boot_pda, }; +static inline void set_kernel_gs(void) +{ + /* Set %gs for this CPU's PDA. Memory clobber is to create a + barrier with respect to any PDA operations, so the compiler + doesn't move any before here. */ + asm volatile ("mov %0, %%gs" : : "r" (__KERNEL_PDA) : "memory"); +} + /* Initialize the CPU's GDT and PDA. The boot CPU does this for itself, but secondaries find this done for them. */ __cpuinit int init_gdt(int cpu, struct task_struct *idle) @@ -693,6 +709,7 @@ static void __cpuinit _cpu_init(int cpu, struct task_struct *curr) the boot CPU, this will transition from the boot gdt+pda to the real ones). */ load_gdt(cpu_gdt_descr); + set_kernel_gs(); if (cpu_test_and_set(cpu, cpu_initialized)) { printk(KERN_WARNING "CPU#%d already initialized!\n", cpu); @@ -731,8 +748,8 @@ static void __cpuinit _cpu_init(int cpu, struct task_struct *curr) __set_tss_desc(cpu, GDT_ENTRY_DOUBLEFAULT_TSS, &doublefault_tss); #endif - /* Clear %fs and %gs. */ - asm volatile ("movl %0, %%fs; movl %0, %%gs" : : "r" (0)); + /* Clear %fs. */ + asm volatile ("mov %0, %%fs" : : "r" (0)); /* Clear all 6 debug registers: */ set_debugreg(0, 0); diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index 0069bf01603..b99d4a16007 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -30,12 +30,13 @@ * 18(%esp) - %eax * 1C(%esp) - %ds * 20(%esp) - %es - * 24(%esp) - orig_eax - * 28(%esp) - %eip - * 2C(%esp) - %cs - * 30(%esp) - %eflags - * 34(%esp) - %oldesp - * 38(%esp) - %oldss + * 24(%esp) - %gs + * 28(%esp) - orig_eax + * 2C(%esp) - %eip + * 30(%esp) - %cs + * 34(%esp) - %eflags + * 38(%esp) - %oldesp + * 3C(%esp) - %oldss * * "current" is in register %ebx during any slow entries. */ @@ -92,6 +93,9 @@ VM_MASK = 0x00020000 #define SAVE_ALL \ cld; \ + pushl %gs; \ + CFI_ADJUST_CFA_OFFSET 4;\ + /*CFI_REL_OFFSET gs, 0;*/\ pushl %es; \ CFI_ADJUST_CFA_OFFSET 4;\ /*CFI_REL_OFFSET es, 0;*/\ @@ -121,7 +125,9 @@ VM_MASK = 0x00020000 CFI_REL_OFFSET ebx, 0;\ movl $(__USER_DS), %edx; \ movl %edx, %ds; \ - movl %edx, %es; + movl %edx, %es; \ + movl $(__KERNEL_PDA), %edx; \ + movl %edx, %gs #define RESTORE_INT_REGS \ popl %ebx; \ @@ -154,17 +160,22 @@ VM_MASK = 0x00020000 2: popl %es; \ CFI_ADJUST_CFA_OFFSET -4;\ /*CFI_RESTORE es;*/\ -.section .fixup,"ax"; \ -3: movl $0,(%esp); \ - jmp 1b; \ +3: popl %gs; \ + CFI_ADJUST_CFA_OFFSET -4;\ + /*CFI_RESTORE gs;*/\ +.pushsection .fixup,"ax"; \ 4: movl $0,(%esp); \ + jmp 1b; \ +5: movl $0,(%esp); \ jmp 2b; \ -.previous; \ +6: movl $0,(%esp); \ + jmp 3b; \ .section __ex_table,"a";\ .align 4; \ - .long 1b,3b; \ - .long 2b,4b; \ -.previous + .long 1b,4b; \ + .long 2b,5b; \ + .long 3b,6b; \ +.popsection #define RING0_INT_FRAME \ CFI_STARTPROC simple;\ @@ -231,6 +242,7 @@ check_userspace: andl $(VM_MASK | SEGMENT_RPL_MASK), %eax cmpl $USER_RPL, %eax jb resume_kernel # not returning to v8086 or userspace + ENTRY(resume_userspace) DISABLE_INTERRUPTS # make sure we don't miss an interrupt # setting need_resched or sigpending @@ -327,9 +339,16 @@ sysenter_past_esp: movl PT_OLDESP(%esp), %ecx xorl %ebp,%ebp TRACE_IRQS_ON +1: mov PT_GS(%esp), %gs ENABLE_INTERRUPTS_SYSEXIT CFI_ENDPROC - +.pushsection .fixup,"ax" +2: movl $0,PT_GS(%esp) + jmp 1b +.section __ex_table,"a" + .align 4 + .long 1b,2b +.popsection # system call handler stub ENTRY(system_call) @@ -375,7 +394,7 @@ restore_nocheck: TRACE_IRQS_IRET restore_nocheck_notrace: RESTORE_REGS - addl $4, %esp + addl $4, %esp # skip orig_eax/error_code CFI_ADJUST_CFA_OFFSET -4 1: INTERRUPT_RETURN .section .fixup,"ax" @@ -588,6 +607,10 @@ KPROBE_ENTRY(page_fault) CFI_ADJUST_CFA_OFFSET 4 ALIGN error_code: + /* the function address is in %gs's slot on the stack */ + pushl %es + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET es, 0*/ pushl %ds CFI_ADJUST_CFA_OFFSET 4 /*CFI_REL_OFFSET ds, 0*/ @@ -613,18 +636,20 @@ error_code: CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET ebx, 0 cld - pushl %es + pushl %gs CFI_ADJUST_CFA_OFFSET 4 - /*CFI_REL_OFFSET es, 0*/ + /*CFI_REL_OFFSET gs, 0*/ + movl $(__KERNEL_PDA), %ecx + movl %ecx, %gs UNWIND_ESPFIX_STACK popl %ecx CFI_ADJUST_CFA_OFFSET -4 /*CFI_REGISTER es, ecx*/ - movl PT_ES(%esp), %edi # get the function address + movl PT_GS(%esp), %edi # get the function address movl PT_ORIG_EAX(%esp), %edx # get the error code - movl $-1, PT_ORIG_EAX(%esp) - movl %ecx, PT_ES(%esp) - /*CFI_REL_OFFSET es, ES*/ + movl $-1, PT_ORIG_EAX(%esp) # no syscall to restart + mov %ecx, PT_GS(%esp) + /*CFI_REL_OFFSET gs, ES*/ movl $(__USER_DS), %ecx movl %ecx, %ds movl %ecx, %es @@ -936,6 +961,7 @@ ENTRY(arch_unwind_init_running) movl %ebx, PT_EAX(%edx) movl $__USER_DS, PT_DS(%edx) movl $__USER_DS, PT_ES(%edx) + movl $0, PT_GS(%edx) movl %ebx, PT_ORIG_EAX(%edx) movl %ecx, PT_EIP(%edx) movl 12(%esp), %ecx diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index 4a83384c5a6..5b14e95ac8b 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -302,6 +302,7 @@ is386: movl $2,%ecx # set MP movl %eax,%cr0 call check_x87 + call setup_pda lgdt cpu_gdt_descr lidt idt_descr ljmp $(__KERNEL_CS),$1f @@ -312,10 +313,13 @@ is386: movl $2,%ecx # set MP movl %eax,%ds movl %eax,%es - xorl %eax,%eax # Clear FS/GS and LDT + xorl %eax,%eax # Clear FS and LDT movl %eax,%fs - movl %eax,%gs lldt %ax + + movl $(__KERNEL_PDA),%eax + mov %eax,%gs + cld # gcc2 wants the direction flag cleared at all times pushl $0 # fake return address for unwinder #ifdef CONFIG_SMP @@ -345,6 +349,23 @@ check_x87: .byte 0xDB,0xE4 /* fsetpm for 287, ignored by 387 */ ret +/* + * Point the GDT at this CPU's PDA. On boot this will be + * cpu_gdt_table and boot_pda; for secondary CPUs, these will be + * that CPU's GDT and PDA. + */ +setup_pda: + /* get the PDA pointer */ + movl start_pda, %eax + + /* slot the PDA address into the GDT */ + mov cpu_gdt_descr+2, %ecx + mov %ax, (__KERNEL_PDA+0+2)(%ecx) /* base & 0x0000ffff */ + shr $16, %eax + mov %al, (__KERNEL_PDA+4+0)(%ecx) /* base & 0x00ff0000 */ + mov %ah, (__KERNEL_PDA+4+3)(%ecx) /* base & 0xff000000 */ + ret + /* * setup_idt * @@ -484,6 +505,8 @@ ENTRY(empty_zero_page) * This starts the data section. */ .data +ENTRY(start_pda) + .long boot_pda ENTRY(stack_start) .long init_thread_union+THREAD_SIZE @@ -525,7 +548,7 @@ idt_descr: # boot GDT descriptor (later on used by CPU#0): .word 0 # 32 bit align gdt_desc.address -cpu_gdt_descr: +ENTRY(cpu_gdt_descr) .word GDT_ENTRIES*8-1 .long cpu_gdt_table @@ -585,7 +608,7 @@ ENTRY(cpu_gdt_table) .quad 0x004092000000ffff /* 0xc8 APM DS data */ .quad 0x00c0920000000000 /* 0xd0 - ESPFIX SS */ - .quad 0x0000000000000000 /* 0xd8 - PDA */ + .quad 0x00cf92000000ffff /* 0xd8 - PDA */ .quad 0x0000000000000000 /* 0xe0 - unused */ .quad 0x0000000000000000 /* 0xe8 - unused */ .quad 0x0000000000000000 /* 0xf0 - unused */ diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index ae924c416b6..905364d4284 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -56,6 +56,7 @@ #include #include +#include asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); @@ -346,6 +347,7 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) regs.xds = __USER_DS; regs.xes = __USER_DS; + regs.xgs = __KERNEL_PDA; regs.orig_eax = -1; regs.eip = (unsigned long) kernel_thread_helper; regs.xcs = __KERNEL_CS | get_kernel_rpl(); @@ -431,7 +433,6 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long esp, p->thread.eip = (unsigned long) ret_from_fork; savesegment(fs,p->thread.fs); - savesegment(gs,p->thread.gs); tsk = current; if (unlikely(test_tsk_thread_flag(tsk, TIF_IO_BITMAP))) { @@ -659,16 +660,16 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas load_esp0(tss, next); /* - * Save away %fs and %gs. No need to save %es and %ds, as - * those are always kernel segments while inside the kernel. - * Doing this before setting the new TLS descriptors avoids - * the situation where we temporarily have non-reloadable - * segments in %fs and %gs. This could be an issue if the - * NMI handler ever used %fs or %gs (it does not today), or - * if the kernel is running inside of a hypervisor layer. + * Save away %fs. No need to save %gs, as it was saved on the + * stack on entry. No need to save %es and %ds, as those are + * always kernel segments while inside the kernel. Doing this + * before setting the new TLS descriptors avoids the situation + * where we temporarily have non-reloadable segments in %fs + * and %gs. This could be an issue if the NMI handler ever + * used %fs or %gs (it does not today), or if the kernel is + * running inside of a hypervisor layer. */ savesegment(fs, prev->fs); - savesegment(gs, prev->gs); /* * Load the per-thread Thread-Local Storage descriptor. @@ -676,16 +677,13 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas load_TLS(next, cpu); /* - * Restore %fs and %gs if needed. + * Restore %fs if needed. * - * Glibc normally makes %fs be zero, and %gs is one of - * the TLS segments. + * Glibc normally makes %fs be zero. */ if (unlikely(prev->fs | next->fs)) loadsegment(fs, next->fs); - if (prev->gs | next->gs) - loadsegment(gs, next->gs); /* * Restore IOPL if needed. diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c index 43002cfb40c..65d7620eaa0 100644 --- a/arch/i386/kernel/signal.c +++ b/arch/i386/kernel/signal.c @@ -128,7 +128,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax X86_EFLAGS_TF | X86_EFLAGS_SF | X86_EFLAGS_ZF | \ X86_EFLAGS_AF | X86_EFLAGS_PF | X86_EFLAGS_CF) - GET_SEG(gs); + COPY_SEG(gs); GET_SEG(fs); COPY_SEG(es); COPY_SEG(ds); @@ -244,9 +244,7 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate, { int tmp, err = 0; - tmp = 0; - savesegment(gs, tmp); - err |= __put_user(tmp, (unsigned int __user *)&sc->gs); + err |= __put_user(regs->xgs, (unsigned int __user *)&sc->gs); savesegment(fs, tmp); err |= __put_user(tmp, (unsigned int __user *)&sc->fs); diff --git a/include/asm-i386/mmu_context.h b/include/asm-i386/mmu_context.h index 1b1495372c4..68ff102d6f5 100644 --- a/include/asm-i386/mmu_context.h +++ b/include/asm-i386/mmu_context.h @@ -62,8 +62,8 @@ static inline void switch_mm(struct mm_struct *prev, #endif } -#define deactivate_mm(tsk, mm) \ - asm("movl %0,%%fs ; movl %0,%%gs": :"r" (0)) +#define deactivate_mm(tsk, mm) \ + asm("movl %0,%%fs": :"r" (0)); #define activate_mm(prev, next) \ switch_mm((prev),(next),NULL) diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h index a9f2041c7c8..f73cf836e64 100644 --- a/include/asm-i386/processor.h +++ b/include/asm-i386/processor.h @@ -473,6 +473,7 @@ struct thread_struct { .vm86_info = NULL, \ .sysenter_cs = __KERNEL_CS, \ .io_bitmap_ptr = NULL, \ + .gs = __KERNEL_PDA, \ } /* @@ -500,7 +501,8 @@ static inline void load_esp0(struct tss_struct *tss, struct thread_struct *threa } #define start_thread(regs, new_eip, new_esp) do { \ - __asm__("movl %0,%%fs ; movl %0,%%gs": :"r" (0)); \ + __asm__("movl %0,%%fs": :"r" (0)); \ + regs->xgs = 0; \ set_fs(USER_DS); \ regs->xds = __USER_DS; \ regs->xes = __USER_DS; \ diff --git a/include/asm-i386/ptrace.h b/include/asm-i386/ptrace.h index d505f501077..bdbc894339b 100644 --- a/include/asm-i386/ptrace.h +++ b/include/asm-i386/ptrace.h @@ -16,6 +16,8 @@ struct pt_regs { long eax; int xds; int xes; + /* int xfs; */ + int xgs; long orig_eax; long eip; int xcs; diff --git a/kernel/fork.c b/kernel/fork.c index 8cdd3e72ba5..fd22245e388 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1303,7 +1303,7 @@ fork_out: return ERR_PTR(retval); } -struct pt_regs * __devinit __attribute__((weak)) idle_regs(struct pt_regs *regs) +noinline struct pt_regs * __devinit __attribute__((weak)) idle_regs(struct pt_regs *regs) { memset(regs, 0, sizeof(struct pt_regs)); return regs; -- cgit v1.2.3-70-g09d2 From 66e10a44d724f1464b5e8b5a3eae1e2cbbc2cca6 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 7 Dec 2006 02:14:02 +0100 Subject: [PATCH] i386: Fix places where using %gs changes the usermode ABI There are a few places where the change in struct pt_regs and the use of %gs affect the userspace ABI. These are primarily debugging interfaces where thread state can be inspected or extracted. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Chuck Ebbert <76306.1226@compuserve.com> Cc: Zachary Amsden Cc: Jan Beulich Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/process.c | 6 +++--- arch/i386/kernel/ptrace.c | 18 ++++++------------ include/asm-i386/elf.h | 2 +- include/asm-i386/unwind.h | 1 + 4 files changed, 11 insertions(+), 16 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 905364d4284..dc427254517 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -315,8 +315,8 @@ void show_regs(struct pt_regs * regs) regs->eax,regs->ebx,regs->ecx,regs->edx); printk("ESI: %08lx EDI: %08lx EBP: %08lx", regs->esi, regs->edi, regs->ebp); - printk(" DS: %04x ES: %04x\n", - 0xffff & regs->xds,0xffff & regs->xes); + printk(" DS: %04x ES: %04x GS: %04x\n", + 0xffff & regs->xds,0xffff & regs->xes, 0xffff & regs->xgs); cr0 = read_cr0(); cr2 = read_cr2(); @@ -509,7 +509,7 @@ void dump_thread(struct pt_regs * regs, struct user * dump) dump->regs.ds = regs->xds; dump->regs.es = regs->xes; savesegment(fs,dump->regs.fs); - savesegment(gs,dump->regs.gs); + dump->regs.gs = regs->xgs; dump->regs.orig_eax = regs->orig_eax; dump->regs.eip = regs->eip; dump->regs.cs = regs->xcs; diff --git a/arch/i386/kernel/ptrace.c b/arch/i386/kernel/ptrace.c index 775f50e9395..f3f94ac5736 100644 --- a/arch/i386/kernel/ptrace.c +++ b/arch/i386/kernel/ptrace.c @@ -94,13 +94,9 @@ static int putreg(struct task_struct *child, return -EIO; child->thread.fs = value; return 0; - case GS: - if (value && (value & 3) != 3) - return -EIO; - child->thread.gs = value; - return 0; case DS: case ES: + case GS: if (value && (value & 3) != 3) return -EIO; value &= 0xffff; @@ -116,8 +112,8 @@ static int putreg(struct task_struct *child, value |= get_stack_long(child, EFL_OFFSET) & ~FLAG_MASK; break; } - if (regno > GS*4) - regno -= 2*4; + if (regno > ES*4) + regno -= 1*4; put_stack_long(child, regno - sizeof(struct pt_regs), value); return 0; } @@ -131,18 +127,16 @@ static unsigned long getreg(struct task_struct *child, case FS: retval = child->thread.fs; break; - case GS: - retval = child->thread.gs; - break; case DS: case ES: + case GS: case SS: case CS: retval = 0xffff; /* fall through */ default: - if (regno > GS*4) - regno -= 2*4; + if (regno > ES*4) + regno -= 1*4; regno = regno - sizeof(struct pt_regs); retval &= get_stack_long(child, regno); } diff --git a/include/asm-i386/elf.h b/include/asm-i386/elf.h index 3a05436f31c..45d21a0c95b 100644 --- a/include/asm-i386/elf.h +++ b/include/asm-i386/elf.h @@ -91,7 +91,7 @@ typedef struct user_fxsr_struct elf_fpxregset_t; pr_reg[7] = regs->xds; \ pr_reg[8] = regs->xes; \ savesegment(fs,pr_reg[9]); \ - savesegment(gs,pr_reg[10]); \ + pr_reg[10] = regs->xgs; \ pr_reg[11] = regs->orig_eax; \ pr_reg[12] = regs->eip; \ pr_reg[13] = regs->xcs; \ diff --git a/include/asm-i386/unwind.h b/include/asm-i386/unwind.h index 5031d693b89..601fc67bd77 100644 --- a/include/asm-i386/unwind.h +++ b/include/asm-i386/unwind.h @@ -71,6 +71,7 @@ static inline void arch_unw_init_blocked(struct unwind_frame_info *info) info->regs.xss = __KERNEL_DS; info->regs.xds = __USER_DS; info->regs.xes = __USER_DS; + info->regs.xgs = __KERNEL_PDA; } extern asmlinkage int arch_unwind_init_running(struct unwind_frame_info *, -- cgit v1.2.3-70-g09d2 From 49d26b6eaa8e970c8cf6e299e6ccba2474191bf5 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] i386: Update sys_vm86 to cope with changed pt_regs and %gs usage sys_vm86 uses a struct kernel_vm86_regs, which is identical to pt_regs, but adds an extra space for all the segment registers. Previously this structure was completely independent, so changes in pt_regs had to be reflected in kernel_vm86_regs. This changes just embeds pt_regs in kernel_vm86_regs, and makes the appropriate changes to vm86.c to deal with the new naming. Also, since %gs is dealt with differently in the kernel, this change adjusts vm86.c to reflect this. While making these changes, I also cleaned up some frankly bizarre code which was added when auditing was added to sys_vm86. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Chuck Ebbert <76306.1226@compuserve.com> Cc: Zachary Amsden Cc: Jan Beulich Cc: Andi Kleen Cc: Al Viro Cc: Jason Baron Cc: Chris Wright Signed-off-by: Andrew Morton --- arch/i386/kernel/vm86.c | 121 +++++++++++++++++++++++++++++------------------- include/asm-i386/vm86.h | 17 +------ 2 files changed, 76 insertions(+), 62 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/vm86.c b/arch/i386/kernel/vm86.c index cbcd61d6120..be2f96e67f7 100644 --- a/arch/i386/kernel/vm86.c +++ b/arch/i386/kernel/vm86.c @@ -43,6 +43,7 @@ #include #include #include +#include #include #include @@ -72,10 +73,10 @@ /* * 8- and 16-bit register defines.. */ -#define AL(regs) (((unsigned char *)&((regs)->eax))[0]) -#define AH(regs) (((unsigned char *)&((regs)->eax))[1]) -#define IP(regs) (*(unsigned short *)&((regs)->eip)) -#define SP(regs) (*(unsigned short *)&((regs)->esp)) +#define AL(regs) (((unsigned char *)&((regs)->pt.eax))[0]) +#define AH(regs) (((unsigned char *)&((regs)->pt.eax))[1]) +#define IP(regs) (*(unsigned short *)&((regs)->pt.eip)) +#define SP(regs) (*(unsigned short *)&((regs)->pt.esp)) /* * virtual flags (16 and 32-bit versions) @@ -89,10 +90,37 @@ #define SAFE_MASK (0xDD5) #define RETURN_MASK (0xDFF) -#define VM86_REGS_PART2 orig_eax -#define VM86_REGS_SIZE1 \ - ( (unsigned)( & (((struct kernel_vm86_regs *)0)->VM86_REGS_PART2) ) ) -#define VM86_REGS_SIZE2 (sizeof(struct kernel_vm86_regs) - VM86_REGS_SIZE1) +/* convert kernel_vm86_regs to vm86_regs */ +static int copy_vm86_regs_to_user(struct vm86_regs __user *user, + const struct kernel_vm86_regs *regs) +{ + int ret = 0; + + /* kernel_vm86_regs is missing xfs, so copy everything up to + (but not including) xgs, and then rest after xgs. */ + ret += copy_to_user(user, regs, offsetof(struct kernel_vm86_regs, pt.xgs)); + ret += copy_to_user(&user->__null_gs, ®s->pt.xgs, + sizeof(struct kernel_vm86_regs) - + offsetof(struct kernel_vm86_regs, pt.xgs)); + + return ret; +} + +/* convert vm86_regs to kernel_vm86_regs */ +static int copy_vm86_regs_from_user(struct kernel_vm86_regs *regs, + const struct vm86_regs __user *user, + unsigned extra) +{ + int ret = 0; + + ret += copy_from_user(regs, user, offsetof(struct kernel_vm86_regs, pt.xgs)); + ret += copy_from_user(®s->pt.xgs, &user->__null_gs, + sizeof(struct kernel_vm86_regs) - + offsetof(struct kernel_vm86_regs, pt.xgs) + + extra); + + return ret; +} struct pt_regs * FASTCALL(save_v86_state(struct kernel_vm86_regs * regs)); struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs) @@ -112,10 +140,8 @@ struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs) printk("no vm86_info: BAD\n"); do_exit(SIGSEGV); } - set_flags(regs->eflags, VEFLAGS, VIF_MASK | current->thread.v86mask); - tmp = copy_to_user(¤t->thread.vm86_info->regs,regs, VM86_REGS_SIZE1); - tmp += copy_to_user(¤t->thread.vm86_info->regs.VM86_REGS_PART2, - ®s->VM86_REGS_PART2, VM86_REGS_SIZE2); + set_flags(regs->pt.eflags, VEFLAGS, VIF_MASK | current->thread.v86mask); + tmp = copy_vm86_regs_to_user(¤t->thread.vm86_info->regs,regs); tmp += put_user(current->thread.screen_bitmap,¤t->thread.vm86_info->screen_bitmap); if (tmp) { printk("vm86: could not access userspace vm86_info\n"); @@ -129,9 +155,11 @@ struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs) current->thread.saved_esp0 = 0; put_cpu(); - loadsegment(fs, current->thread.saved_fs); - loadsegment(gs, current->thread.saved_gs); ret = KVM86->regs32; + + loadsegment(fs, current->thread.saved_fs); + ret->xgs = current->thread.saved_gs; + return ret; } @@ -183,9 +211,9 @@ asmlinkage int sys_vm86old(struct pt_regs regs) tsk = current; if (tsk->thread.saved_esp0) goto out; - tmp = copy_from_user(&info, v86, VM86_REGS_SIZE1); - tmp += copy_from_user(&info.regs.VM86_REGS_PART2, &v86->regs.VM86_REGS_PART2, - (long)&info.vm86plus - (long)&info.regs.VM86_REGS_PART2); + tmp = copy_vm86_regs_from_user(&info.regs, &v86->regs, + offsetof(struct kernel_vm86_struct, vm86plus) - + sizeof(info.regs)); ret = -EFAULT; if (tmp) goto out; @@ -233,9 +261,9 @@ asmlinkage int sys_vm86(struct pt_regs regs) if (tsk->thread.saved_esp0) goto out; v86 = (struct vm86plus_struct __user *)regs.ecx; - tmp = copy_from_user(&info, v86, VM86_REGS_SIZE1); - tmp += copy_from_user(&info.regs.VM86_REGS_PART2, &v86->regs.VM86_REGS_PART2, - (long)&info.regs32 - (long)&info.regs.VM86_REGS_PART2); + tmp = copy_vm86_regs_from_user(&info.regs, &v86->regs, + offsetof(struct kernel_vm86_struct, regs32) - + sizeof(info.regs)); ret = -EFAULT; if (tmp) goto out; @@ -252,15 +280,15 @@ out: static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk) { struct tss_struct *tss; - long eax; /* * make sure the vm86() system call doesn't try to do anything silly */ - info->regs.__null_ds = 0; - info->regs.__null_es = 0; + info->regs.pt.xds = 0; + info->regs.pt.xes = 0; + info->regs.pt.xgs = 0; -/* we are clearing fs,gs later just before "jmp resume_userspace", - * because starting with Linux 2.1.x they aren't no longer saved/restored +/* we are clearing fs later just before "jmp resume_userspace", + * because it is not saved/restored. */ /* @@ -268,10 +296,10 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk * has set it up safely, so this makes sure interrupt etc flags are * inherited from protected mode. */ - VEFLAGS = info->regs.eflags; - info->regs.eflags &= SAFE_MASK; - info->regs.eflags |= info->regs32->eflags & ~SAFE_MASK; - info->regs.eflags |= VM_MASK; + VEFLAGS = info->regs.pt.eflags; + info->regs.pt.eflags &= SAFE_MASK; + info->regs.pt.eflags |= info->regs32->eflags & ~SAFE_MASK; + info->regs.pt.eflags |= VM_MASK; switch (info->cpu_type) { case CPU_286: @@ -294,7 +322,7 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk info->regs32->eax = 0; tsk->thread.saved_esp0 = tsk->thread.esp0; savesegment(fs, tsk->thread.saved_fs); - savesegment(gs, tsk->thread.saved_gs); + tsk->thread.saved_gs = info->regs32->xgs; tss = &per_cpu(init_tss, get_cpu()); tsk->thread.esp0 = (unsigned long) &info->VM86_TSS_ESP0; @@ -306,19 +334,18 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk tsk->thread.screen_bitmap = info->screen_bitmap; if (info->flags & VM86_SCREEN_BITMAP) mark_screen_rdonly(tsk->mm); - __asm__ __volatile__("xorl %eax,%eax; movl %eax,%fs; movl %eax,%gs\n\t"); - __asm__ __volatile__("movl %%eax, %0\n" :"=r"(eax)); /*call audit_syscall_exit since we do not exit via the normal paths */ if (unlikely(current->audit_context)) - audit_syscall_exit(AUDITSC_RESULT(eax), eax); + audit_syscall_exit(AUDITSC_RESULT(0), 0); __asm__ __volatile__( "movl %0,%%esp\n\t" "movl %1,%%ebp\n\t" + "mov %2, %%fs\n\t" "jmp resume_userspace" : /* no outputs */ - :"r" (&info->regs), "r" (task_thread_info(tsk))); + :"r" (&info->regs), "r" (task_thread_info(tsk)), "r" (0)); /* we never return here */ } @@ -348,12 +375,12 @@ static inline void clear_IF(struct kernel_vm86_regs * regs) static inline void clear_TF(struct kernel_vm86_regs * regs) { - regs->eflags &= ~TF_MASK; + regs->pt.eflags &= ~TF_MASK; } static inline void clear_AC(struct kernel_vm86_regs * regs) { - regs->eflags &= ~AC_MASK; + regs->pt.eflags &= ~AC_MASK; } /* It is correct to call set_IF(regs) from the set_vflags_* @@ -370,7 +397,7 @@ static inline void clear_AC(struct kernel_vm86_regs * regs) static inline void set_vflags_long(unsigned long eflags, struct kernel_vm86_regs * regs) { set_flags(VEFLAGS, eflags, current->thread.v86mask); - set_flags(regs->eflags, eflags, SAFE_MASK); + set_flags(regs->pt.eflags, eflags, SAFE_MASK); if (eflags & IF_MASK) set_IF(regs); else @@ -380,7 +407,7 @@ static inline void set_vflags_long(unsigned long eflags, struct kernel_vm86_regs static inline void set_vflags_short(unsigned short flags, struct kernel_vm86_regs * regs) { set_flags(VFLAGS, flags, current->thread.v86mask); - set_flags(regs->eflags, flags, SAFE_MASK); + set_flags(regs->pt.eflags, flags, SAFE_MASK); if (flags & IF_MASK) set_IF(regs); else @@ -389,7 +416,7 @@ static inline void set_vflags_short(unsigned short flags, struct kernel_vm86_reg static inline unsigned long get_vflags(struct kernel_vm86_regs * regs) { - unsigned long flags = regs->eflags & RETURN_MASK; + unsigned long flags = regs->pt.eflags & RETURN_MASK; if (VEFLAGS & VIF_MASK) flags |= IF_MASK; @@ -493,7 +520,7 @@ static void do_int(struct kernel_vm86_regs *regs, int i, unsigned long __user *intr_ptr; unsigned long segoffs; - if (regs->cs == BIOSSEG) + if (regs->pt.xcs == BIOSSEG) goto cannot_handle; if (is_revectored(i, &KVM86->int_revectored)) goto cannot_handle; @@ -505,9 +532,9 @@ static void do_int(struct kernel_vm86_regs *regs, int i, if ((segoffs >> 16) == BIOSSEG) goto cannot_handle; pushw(ssp, sp, get_vflags(regs), cannot_handle); - pushw(ssp, sp, regs->cs, cannot_handle); + pushw(ssp, sp, regs->pt.xcs, cannot_handle); pushw(ssp, sp, IP(regs), cannot_handle); - regs->cs = segoffs >> 16; + regs->pt.xcs = segoffs >> 16; SP(regs) -= 6; IP(regs) = segoffs & 0xffff; clear_TF(regs); @@ -524,7 +551,7 @@ int handle_vm86_trap(struct kernel_vm86_regs * regs, long error_code, int trapno if (VMPI.is_vm86pus) { if ( (trapno==3) || (trapno==1) ) return_to_32bit(regs, VM86_TRAP + (trapno << 8)); - do_int(regs, trapno, (unsigned char __user *) (regs->ss << 4), SP(regs)); + do_int(regs, trapno, (unsigned char __user *) (regs->pt.xss << 4), SP(regs)); return 0; } if (trapno !=1) @@ -560,10 +587,10 @@ void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code) handle_vm86_trap(regs, 0, 1); \ return; } while (0) - orig_flags = *(unsigned short *)®s->eflags; + orig_flags = *(unsigned short *)®s->pt.eflags; - csp = (unsigned char __user *) (regs->cs << 4); - ssp = (unsigned char __user *) (regs->ss << 4); + csp = (unsigned char __user *) (regs->pt.xcs << 4); + ssp = (unsigned char __user *) (regs->pt.xss << 4); sp = SP(regs); ip = IP(regs); @@ -650,7 +677,7 @@ void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code) SP(regs) += 6; } IP(regs) = newip; - regs->cs = newcs; + regs->pt.xcs = newcs; CHECK_IF_IN_TRAP; if (data32) { set_vflags_long(newflags, regs); diff --git a/include/asm-i386/vm86.h b/include/asm-i386/vm86.h index 952fd695738..a5edf517b99 100644 --- a/include/asm-i386/vm86.h +++ b/include/asm-i386/vm86.h @@ -145,26 +145,13 @@ struct vm86plus_struct { * at the end of the structure. Look at ptrace.h to see the "normal" * setup. For user space layout see 'struct vm86_regs' above. */ +#include struct kernel_vm86_regs { /* * normal regs, with special meaning for the segment descriptors.. */ - long ebx; - long ecx; - long edx; - long esi; - long edi; - long ebp; - long eax; - long __null_ds; - long __null_es; - long orig_eax; - long eip; - unsigned short cs, __csh; - long eflags; - long esp; - unsigned short ss, __ssh; + struct pt_regs pt; /* * these are specific to v86 mode: */ -- cgit v1.2.3-70-g09d2 From b2938f880890ebfcccad356275e0000193153623 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] i386: Implement smp_processor_id() with the PDA Use the cpu_number in the PDA to implement raw_smp_processor_id. This is a little simpler than using thread_info, though the cpu field in thread_info cannot be removed since it is used for things other than getting the current CPU in common code. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Chuck Ebbert <76306.1226@compuserve.com> Cc: Zachary Amsden Cc: Jan Beulich Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/asm-offsets.c | 4 +++- arch/i386/kernel/cpu/common.c | 2 ++ arch/i386/kernel/entry.S | 3 +-- include/asm-i386/pda.h | 2 ++ include/asm-i386/smp.h | 3 ++- 5 files changed, 10 insertions(+), 4 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index 9620872d353..85f1b038e9c 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -51,7 +51,6 @@ void foo(void) OFFSET(TI_exec_domain, thread_info, exec_domain); OFFSET(TI_flags, thread_info, flags); OFFSET(TI_status, thread_info, status); - OFFSET(TI_cpu, thread_info, cpu); OFFSET(TI_preempt_count, thread_info, preempt_count); OFFSET(TI_addr_limit, thread_info, addr_limit); OFFSET(TI_restart_block, thread_info, restart_block); @@ -97,4 +96,7 @@ void foo(void) DEFINE(VDSO_PRELINK, VDSO_PRELINK); OFFSET(crypto_tfm_ctx_offset, crypto_tfm, __crt_ctx); + + BLANK(); + OFFSET(PDA_cpu, i386_pda, cpu_number); } diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index 4e63d8ce602..e476202b887 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -650,6 +650,7 @@ __cpuinit int alloc_gdt(int cpu) /* Initial PDA used by boot CPU */ struct i386_pda boot_pda = { ._pda = &boot_pda, + .cpu_number = 0, }; static inline void set_kernel_gs(void) @@ -694,6 +695,7 @@ __cpuinit int init_gdt(int cpu, struct task_struct *idle) memset(pda, 0, sizeof(*pda)); pda->_pda = pda; + pda->cpu_number = cpu; return 1; } diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index b99d4a16007..d7423efaeea 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -524,8 +524,7 @@ syscall_badsys: #define FIXUP_ESPFIX_STACK \ /* since we are on a wrong stack, we cant make it a C code :( */ \ - GET_THREAD_INFO(%ebp); \ - movl TI_cpu(%ebp), %ebx; \ + movl %gs:PDA_cpu, %ebx; \ PER_CPU(cpu_gdt_descr, %ebx); \ movl GDS_address(%ebx), %ebx; \ GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah); \ diff --git a/include/asm-i386/pda.h b/include/asm-i386/pda.h index 4c39ccb1305..f90fde22566 100644 --- a/include/asm-i386/pda.h +++ b/include/asm-i386/pda.h @@ -11,6 +11,8 @@ struct i386_pda { struct i386_pda *_pda; /* pointer to self */ + + int cpu_number; }; extern struct i386_pda *_cpu_pda[]; diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h index bd59c1508e7..64fe624c02c 100644 --- a/include/asm-i386/smp.h +++ b/include/asm-i386/smp.h @@ -8,6 +8,7 @@ #include #include #include +#include #endif #ifdef CONFIG_X86_LOCAL_APIC @@ -56,7 +57,7 @@ extern void cpu_uninit(void); * from the initial startup. We map APIC_BASE very early in page_setup(), * so this is correct in the x86 case. */ -#define raw_smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (read_pda(cpu_number)) extern cpumask_t cpu_callout_map; extern cpumask_t cpu_callin_map; -- cgit v1.2.3-70-g09d2 From ec7fcaabbfb3c5bd5189f857b6ac7bb9745ef291 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] i386: Implement "current" with the PDA Use the pcurrent field in the PDA to implement the "current" macro. This ends up compiling down to a single instruction to get the current task. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Chuck Ebbert <76306.1226@compuserve.com> Cc: Zachary Amsden Cc: Jan Beulich Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/asm-offsets.c | 2 ++ arch/i386/kernel/cpu/common.c | 2 ++ arch/i386/kernel/process.c | 1 + include/asm-i386/current.h | 7 ++++--- include/asm-i386/pda.h | 2 ++ 5 files changed, 11 insertions(+), 3 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index 85f1b038e9c..0666eb0ed7b 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -15,6 +15,7 @@ #include #include #include +#include #define DEFINE(sym, val) \ asm volatile("\n->" #sym " %0 " #val : : "i" (val)) @@ -99,4 +100,5 @@ void foo(void) BLANK(); OFFSET(PDA_cpu, i386_pda, cpu_number); + OFFSET(PDA_pcurrent, i386_pda, pcurrent); } diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index e476202b887..6958ae5e2fa 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -651,6 +651,7 @@ __cpuinit int alloc_gdt(int cpu) struct i386_pda boot_pda = { ._pda = &boot_pda, .cpu_number = 0, + .pcurrent = &init_task, }; static inline void set_kernel_gs(void) @@ -696,6 +697,7 @@ __cpuinit int init_gdt(int cpu, struct task_struct *idle) memset(pda, 0, sizeof(*pda)); pda->_pda = pda; pda->cpu_number = cpu; + pda->pcurrent = idle; return 1; } diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index dc427254517..8749b10d380 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -684,6 +684,7 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas if (unlikely(prev->fs | next->fs)) loadsegment(fs, next->fs); + write_pda(pcurrent, next_p); /* * Restore IOPL if needed. diff --git a/include/asm-i386/current.h b/include/asm-i386/current.h index 3cbbecd7901..5252ee0f6d7 100644 --- a/include/asm-i386/current.h +++ b/include/asm-i386/current.h @@ -1,13 +1,14 @@ #ifndef _I386_CURRENT_H #define _I386_CURRENT_H -#include +#include +#include struct task_struct; -static __always_inline struct task_struct * get_current(void) +static __always_inline struct task_struct *get_current(void) { - return current_thread_info()->task; + return read_pda(pcurrent); } #define current get_current() diff --git a/include/asm-i386/pda.h b/include/asm-i386/pda.h index f90fde22566..08a35c478af 100644 --- a/include/asm-i386/pda.h +++ b/include/asm-i386/pda.h @@ -7,12 +7,14 @@ #define _I386_PDA_H #include +#include struct i386_pda { struct i386_pda *_pda; /* pointer to self */ int cpu_number; + struct task_struct *pcurrent; /* current process */ }; extern struct i386_pda *_cpu_pda[]; -- cgit v1.2.3-70-g09d2 From 72690a21188586022a9e65cb6f1cc8845167555a Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] x86: Don't use nested idle loops Currently the idle loop has two nested loops -- one high level in cpu_idle and in some low level idle functions another one. Looping in the low level idle functions breaks the idle notifiers because interrupts waking up sleep states need to execute exit_idle() which is only in cpu_idle(). So don't do that, only loop in cpu_idle(). This only removes code. In some cases e.g. poll_idle the idle loop is a little longer now because cpu_idle checks more things. I hope that isn't a problem ACPI idle doesn't change behaviour because it never looped anyways. Cc: len.brown@intel.com Cc: eranian@hpl.hp.com Signed-off-by: Andi Kleen --- arch/i386/kernel/process.c | 30 +++++++++--------------------- arch/x86_64/kernel/process.c | 30 +++++++++--------------------- 2 files changed, 18 insertions(+), 42 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 8749b10d380..8f42659ef9d 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -100,22 +100,18 @@ EXPORT_SYMBOL(enable_hlt); */ void default_idle(void) { - local_irq_enable(); - if (!hlt_counter && boot_cpu_data.hlt_works_ok) { current_thread_info()->status &= ~TS_POLLING; smp_mb__after_clear_bit(); - while (!need_resched()) { - local_irq_disable(); - if (!need_resched()) - safe_halt(); - else - local_irq_enable(); - } + local_irq_disable(); + if (!need_resched()) + safe_halt(); /* enables interrupts racelessly */ + else + local_irq_enable(); current_thread_info()->status |= TS_POLLING; } else { - while (!need_resched()) - cpu_relax(); + /* loop is done by the caller */ + cpu_relax(); } } #ifdef CONFIG_APM_MODULE @@ -129,14 +125,7 @@ EXPORT_SYMBOL(default_idle); */ static void poll_idle (void) { - local_irq_enable(); - - asm volatile( - "2:" - "testl %0, %1;" - "rep; nop;" - "je 2b;" - : : "i"(_TIF_NEED_RESCHED), "m" (current_thread_info()->flags)); + cpu_relax(); } #ifdef CONFIG_HOTPLUG_CPU @@ -257,8 +246,7 @@ void mwait_idle_with_hints(unsigned long eax, unsigned long ecx) static void mwait_idle(void) { local_irq_enable(); - while (!need_resched()) - mwait_idle_with_hints(0, 0); + mwait_idle_with_hints(0, 0); } void __devinit select_idle_routine(const struct cpuinfo_x86 *c) diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 7451a4c43c1..0b7b4caa4f7 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -108,17 +108,15 @@ void exit_idle(void) */ static void default_idle(void) { - local_irq_enable(); - current_thread_info()->status &= ~TS_POLLING; smp_mb__after_clear_bit(); - while (!need_resched()) { - local_irq_disable(); - if (!need_resched()) - safe_halt(); - else - local_irq_enable(); - } + local_irq_disable(); + if (!need_resched()) { + /* Enables interrupts one instruction before HLT. + x86 special cases this so there is no race. */ + safe_halt(); + } else + local_irq_enable(); current_thread_info()->status |= TS_POLLING; } @@ -129,16 +127,7 @@ static void default_idle(void) */ static void poll_idle (void) { - local_irq_enable(); - - asm volatile( - "2:" - "testl %0,%1;" - "rep; nop;" - "je 2b;" - : : - "i" (_TIF_NEED_RESCHED), - "m" (current_thread_info()->flags)); + cpu_relax(); } void cpu_idle_wait(void) @@ -257,8 +246,7 @@ void mwait_idle_with_hints(unsigned long eax, unsigned long ecx) static void mwait_idle(void) { local_irq_enable(); - while (!need_resched()) - mwait_idle_with_hints(0,0); + mwait_idle_with_hints(0,0); } void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) -- cgit v1.2.3-70-g09d2 From 7cd8b6861eb586aabe4c725cc0c259ce2e653695 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] x86: remove last two pci_find offenders in the core code Resending as I believe the discussion about them established they were correct. Signed-off-by: Alan Cox Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/pci/irq.c | 4 +++- arch/x86_64/kernel/pci-calgary.c | 2 +- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c index e65551cd821..f2cb942f828 100644 --- a/arch/i386/pci/irq.c +++ b/arch/i386/pci/irq.c @@ -764,7 +764,7 @@ static void __init pirq_find_router(struct irq_router *r) DBG(KERN_DEBUG "PCI: Attempting to find IRQ router for %04x:%04x\n", rt->rtr_vendor, rt->rtr_device); - pirq_router_dev = pci_find_slot(rt->rtr_bus, rt->rtr_devfn); + pirq_router_dev = pci_get_bus_and_slot(rt->rtr_bus, rt->rtr_devfn); if (!pirq_router_dev) { DBG(KERN_DEBUG "PCI: Interrupt router not found at " "%02x:%02x\n", rt->rtr_bus, rt->rtr_devfn); @@ -784,6 +784,8 @@ static void __init pirq_find_router(struct irq_router *r) pirq_router_dev->vendor, pirq_router_dev->device, pci_name(pirq_router_dev)); + + /* The device remains referenced for the kernel lifetime */ } static struct irq_info *pirq_get_info(struct pci_dev *dev) diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index 37a770859e7..d2ea87a9526 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -921,7 +921,7 @@ static int __init calgary_init(void) error: do { - dev = pci_find_device_reverse(PCI_VENDOR_ID_IBM, + dev = pci_get_device_reverse(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CALGARY, dev); if (!dev) -- cgit v1.2.3-70-g09d2 From 9c5f8be4625e73f17e28fea89399ed871a30e064 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] x86: Mention PCI instead of RAM in NMI parity error message On modern systems RAM errors don't cause NMIs, but it's usually caused by PCI SERR. Mention PCI instead of RAM in the printk. Reported by r_hayashi@ctc-g.co.jp (Ryutaro Hayashi) Cc: r_hayashi@ctc-g.co.jp Signed-off-by: Andi Kleen --- arch/i386/kernel/traps.c | 3 +-- arch/x86_64/kernel/traps.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 4a6fa2837df..237f4884a1e 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -708,8 +708,7 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs) { printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x on " "CPU %d.\n", reason, smp_processor_id()); - printk(KERN_EMERG "You probably have a hardware problem with your RAM " - "chips\n"); + printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n"); if (panic_on_unrecovered_nmi) panic("NMI: Not continuing"); diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index e37b4d77d5a..70bfaab9822 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -793,8 +793,7 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs) { printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n", reason); - printk(KERN_EMERG "You probably have a hardware problem with your " - "RAM chips\n"); + printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n"); if (panic_on_unrecovered_nmi) panic("NMI: Not continuing"); -- cgit v1.2.3-70-g09d2 From 6569580de7ae367def89b7671029cb97c1965574 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] i386: Distinguish absolute symbols Ld knows about 2 kinds of symbols, absolute and section relative. Section relative symbols symbols change value when a section is moved and absolute symbols do not. Currently in the linker script we have several labels marking the beginning and ending of sections that are outside of sections, making them absolute symbols. Having a mixture of absolute and section relative symbols refereing to the same data is currently harmless but it is confusing. This must be done carefully as newer revs of ld do not place symbols that appear in sections without data and instead ld makes those symbols global :( My ultimate goal is to build a relocatable kernel. The safest and least intrusive technique is to generate relocation entries so the kernel can be relocated at load time. The only penalty would be an increase in the size of the kernel binary. The problem is that if absolute and relocatable symbols are not properly specified absolute symbols will be relocated or section relative symbols won't be, which is fatal. The practical motivation is that when generating kernels that will run from a reserved area for analyzing what caused a kernel panic, it is simpler if you don't need to hard code the physical memory location they will run at, especially for the distributions. [AK: and merged:] o Also put a message so that in future people can be aware of it and avoid introducing absolute symbols. Signed-off-by: Eric W. Biederman Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/i386/kernel/vmlinux.lds.S | 113 +++++++++++++++++++++----------------- include/asm-generic/vmlinux.lds.h | 10 ++-- 2 files changed, 68 insertions(+), 55 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index c6f84a0322b..cbd24860fbb 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -1,5 +1,11 @@ /* ld script to make i386 Linux kernel * Written by Martin Mares ; + * + * Don't define absolute symbols until and unless you know that symbol + * value is should remain constant even if kernel image is relocated + * at run time. Absolute symbols are not relocated. If symbol value should + * change if kernel is relocated, make the symbol section relative and + * put it inside the section definition. */ #define LOAD_OFFSET __PAGE_OFFSET @@ -24,31 +30,32 @@ SECTIONS . = __KERNEL_START; phys_startup_32 = startup_32 - LOAD_OFFSET; /* read-only */ - _text = .; /* Text and read-only data */ .text : AT(ADDR(.text) - LOAD_OFFSET) { + _text = .; /* Text and read-only data */ *(.text) SCHED_TEXT LOCK_TEXT KPROBES_TEXT *(.fixup) *(.gnu.warning) - } :text = 0x9090 - - _etext = .; /* End of text section */ + _etext = .; /* End of text section */ + } :text = 0x9090 . = ALIGN(16); /* Exception table */ - __start___ex_table = .; - __ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) { *(__ex_table) } - __stop___ex_table = .; + __ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) { + __start___ex_table = .; + *(__ex_table) + __stop___ex_table = .; + } RODATA . = ALIGN(4); - __tracedata_start = .; .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) { + __tracedata_start = .; *(.tracedata) + __tracedata_end = .; } - __tracedata_end = .; /* writeable */ . = ALIGN(4096); @@ -58,10 +65,12 @@ SECTIONS } :data . = ALIGN(4096); - __nosave_begin = .; - .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { *(.data.nosave) } - . = ALIGN(4096); - __nosave_end = .; + .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { + __nosave_begin = .; + *(.data.nosave) + . = ALIGN(4096); + __nosave_end = .; + } . = ALIGN(4096); .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) { @@ -75,8 +84,10 @@ SECTIONS /* rarely changed data like cpu maps */ . = ALIGN(32); - .data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET) { *(.data.read_mostly) } - _edata = .; /* End of data section */ + .data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET) { + *(.data.read_mostly) + _edata = .; /* End of data section */ + } #ifdef CONFIG_STACK_UNWIND . = ALIGN(4); @@ -94,54 +105,56 @@ SECTIONS /* might get freed after init */ . = ALIGN(4096); - __smp_alt_begin = .; - __smp_alt_instructions = .; .smp_altinstructions : AT(ADDR(.smp_altinstructions) - LOAD_OFFSET) { + __smp_alt_begin = .; + __smp_alt_instructions = .; *(.smp_altinstructions) + __smp_alt_instructions_end = .; } - __smp_alt_instructions_end = .; . = ALIGN(4); - __smp_locks = .; .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) { + __smp_locks = .; *(.smp_locks) + __smp_locks_end = .; } - __smp_locks_end = .; .smp_altinstr_replacement : AT(ADDR(.smp_altinstr_replacement) - LOAD_OFFSET) { *(.smp_altinstr_replacement) + __smp_alt_end = .; } . = ALIGN(4096); - __smp_alt_end = .; /* will be freed after init */ . = ALIGN(4096); /* Init code and data */ - __init_begin = .; .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) { + __init_begin = .; _sinittext = .; *(.init.text) _einittext = .; } .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { *(.init.data) } . = ALIGN(16); - __setup_start = .; - .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { *(.init.setup) } - __setup_end = .; - __initcall_start = .; + .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { + __setup_start = .; + *(.init.setup) + __setup_end = .; + } .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { + __initcall_start = .; INITCALLS + __initcall_end = .; } - __initcall_end = .; - __con_initcall_start = .; .con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) { + __con_initcall_start = .; *(.con_initcall.init) + __con_initcall_end = .; } - __con_initcall_end = .; SECURITY_INIT . = ALIGN(4); - __alt_instructions = .; .altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) { + __alt_instructions = .; *(.altinstructions) + __alt_instructions_end = .; } - __alt_instructions_end = .; .altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) { *(.altinstr_replacement) } @@ -150,32 +163,32 @@ SECTIONS .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { *(.exit.text) } .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { *(.exit.data) } . = ALIGN(4096); - __initramfs_start = .; - .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) } - __initramfs_end = .; + .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { + __initramfs_start = .; + *(.init.ramfs) + __initramfs_end = .; + } . = ALIGN(L1_CACHE_BYTES); - __per_cpu_start = .; - .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) } - __per_cpu_end = .; + .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { + __per_cpu_start = .; + *(.data.percpu) + __per_cpu_end = .; + } . = ALIGN(4096); - __init_end = .; /* freed after init ends here */ - __bss_start = .; /* BSS */ - .bss.page_aligned : AT(ADDR(.bss.page_aligned) - LOAD_OFFSET) { - *(.bss.page_aligned) - } .bss : AT(ADDR(.bss) - LOAD_OFFSET) { + __init_end = .; + __bss_start = .; /* BSS */ + *(.bss.page_aligned) *(.bss) + . = ALIGN(4); + __bss_stop = .; + _end = . ; + /* This is where the kernel creates the early boot page tables */ + . = ALIGN(4096); + pg0 = . ; } - . = ALIGN(4); - __bss_stop = .; - - _end = . ; - - /* This is where the kernel creates the early boot page tables */ - . = ALIGN(4096); - pg0 = .; /* Sections to be discarded */ /DISCARD/ : { diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index e60d6f21fa6..9f4747780da 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -11,8 +11,8 @@ #define RODATA \ . = ALIGN(4096); \ - __start_rodata = .; \ .rodata : AT(ADDR(.rodata) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start_rodata) = .; \ *(.rodata) *(.rodata.*) \ *(__vermagic) /* Kernel version magic */ \ } \ @@ -119,17 +119,17 @@ *(__ksymtab_strings) \ } \ \ + /* Unwind data binary search table */ \ + EH_FRAME_HDR \ + \ /* Built-in module parameters. */ \ __param : AT(ADDR(__param) - LOAD_OFFSET) { \ VMLINUX_SYMBOL(__start___param) = .; \ *(__param) \ VMLINUX_SYMBOL(__stop___param) = .; \ + VMLINUX_SYMBOL(__end_rodata) = .; \ } \ \ - /* Unwind data binary search table */ \ - EH_FRAME_HDR \ - \ - __end_rodata = .; \ . = ALIGN(4096); #define SECURITY_INIT \ -- cgit v1.2.3-70-g09d2 From 6ed018845f1172cdc94f8a20ad807df901c6b7eb Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] i386: Add comment for align to vmlinux.lds Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/i386/kernel/vmlinux.lds.S | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index cbd24860fbb..c217e18f108 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -121,6 +121,12 @@ SECTIONS *(.smp_altinstr_replacement) __smp_alt_end = .; } + /* will be freed after init + * Following ALIGN() is required to make sure no other data falls on the + * same page where __smp_alt_end is pointing as that page might be freed + * after boot. Always make sure that ALIGN() directive is present after + * the section which contains __smp_alt_end. + */ . = ALIGN(4096); /* will be freed after init */ -- cgit v1.2.3-70-g09d2 From 8621b81c744ff8880a1efe095a4dcd09763ddb5a Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] i386: Reserve kernel memory starting from _text Currently when we are reserving the memory the kernel text resides in we start at __PHYSICAL_START which happens to be correct but not very obvious. In addition when we start relocating the kernel __PHYSICAL_START is the wrong value, as it is an absolute symbol that does not get relocated. By starting the reservation at __pa_symbol(_text) the code is clearer and will be correct when relocated. Signed-off-by: Eric W. Biederman Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/i386/kernel/setup.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 141041dde74..61539afbdf2 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -1118,8 +1118,8 @@ void __init setup_bootmem_allocator(void) * the (very unlikely) case of us accidentally initializing the * bootmem allocator with an invalid RAM area. */ - reserve_bootmem(__PHYSICAL_START, (PFN_PHYS(min_low_pfn) + - bootmap_size + PAGE_SIZE-1) - (__PHYSICAL_START)); + reserve_bootmem(__pa_symbol(_text), (PFN_PHYS(min_low_pfn) + + bootmap_size + PAGE_SIZE-1) - __pa_symbol(_text)); /* * reserve physical page 0 - it's a special BIOS page on many boxes, -- cgit v1.2.3-70-g09d2 From 2a43f3ede48ea3d5790b863b719a1e21c90a3697 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Thu, 7 Dec 2006 02:14:04 +0100 Subject: [PATCH] i386: CONFIG_PHYSICAL_START cleanup Defining __PHYSICAL_START and __KERNEL_START in asm-i386/page.h works but it triggers a full kernel rebuild for the silliest of reasons. This modifies the users to directly use CONFIG_PHYSICAL_START and linux/config.h which prevents the full rebuild problem, which makes the code much more maintainer and hopefully user friendly. Signed-off-by: Eric W. Biederman Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/i386/boot/compressed/head.S | 7 +++---- arch/i386/boot/compressed/misc.c | 7 +++---- arch/i386/kernel/vmlinux.lds.S | 2 +- include/asm-i386/page.h | 3 --- 4 files changed, 7 insertions(+), 12 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/boot/compressed/head.S b/arch/i386/boot/compressed/head.S index b5893e4ecd3..40a8de8270a 100644 --- a/arch/i386/boot/compressed/head.S +++ b/arch/i386/boot/compressed/head.S @@ -25,7 +25,6 @@ #include #include -#include .globl startup_32 @@ -75,7 +74,7 @@ startup_32: popl %esi # discard address popl %esi # real mode pointer xorl %ebx,%ebx - ljmp $(__BOOT_CS), $__PHYSICAL_START + ljmp $(__BOOT_CS), $CONFIG_PHYSICAL_START /* * We come here, if we were loaded high. @@ -100,7 +99,7 @@ startup_32: popl %ecx # lcount popl %edx # high_buffer_start popl %eax # hcount - movl $__PHYSICAL_START,%edi + movl $CONFIG_PHYSICAL_START,%edi cli # make sure we don't get interrupted ljmp $(__BOOT_CS), $0x1000 # and jump to the move routine @@ -125,5 +124,5 @@ move_routine_start: movsl movl %ebx,%esi # Restore setup pointer xorl %ebx,%ebx - ljmp $(__BOOT_CS), $__PHYSICAL_START + ljmp $(__BOOT_CS), $CONFIG_PHYSICAL_START move_routine_end: diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c index b2ccd543410..20970ff4411 100644 --- a/arch/i386/boot/compressed/misc.c +++ b/arch/i386/boot/compressed/misc.c @@ -13,7 +13,6 @@ #include #include #include -#include /* * gzip declarations @@ -303,7 +302,7 @@ static void setup_normal_output_buffer(void) #else if ((RM_ALT_MEM_K > RM_EXT_MEM_K ? RM_ALT_MEM_K : RM_EXT_MEM_K) < 1024) error("Less than 2MB of memory"); #endif - output_data = (unsigned char *)__PHYSICAL_START; /* Normally Points to 1M */ + output_data = (unsigned char *)CONFIG_PHYSICAL_START; /* Normally Points to 1M */ free_mem_end_ptr = (long)real_mode; } @@ -326,8 +325,8 @@ static void setup_output_buffer_if_we_run_high(struct moveparams *mv) low_buffer_size = low_buffer_end - LOW_BUFFER_START; high_loaded = 1; free_mem_end_ptr = (long)high_buffer_start; - if ( (__PHYSICAL_START + low_buffer_size) > ((ulg)high_buffer_start)) { - high_buffer_start = (uch *)(__PHYSICAL_START + low_buffer_size); + if ( (CONFIG_PHYSICAL_START + low_buffer_size) > ((ulg)high_buffer_start)) { + high_buffer_start = (uch *)(CONFIG_PHYSICAL_START + low_buffer_size); mv->hcount = 0; /* say: we need not to move high_buffer */ } else mv->hcount = -1; diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index c217e18f108..f8d61ec4c8c 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -27,7 +27,7 @@ PHDRS { } SECTIONS { - . = __KERNEL_START; + . = LOAD_OFFSET + CONFIG_PHYSICAL_START; phys_startup_32 = startup_32 - LOAD_OFFSET; /* read-only */ .text : AT(ADDR(.text) - LOAD_OFFSET) { diff --git a/include/asm-i386/page.h b/include/asm-i386/page.h index 5a70501291d..2b69686107a 100644 --- a/include/asm-i386/page.h +++ b/include/asm-i386/page.h @@ -112,12 +112,9 @@ extern int page_is_ram(unsigned long pagenr); #ifdef __ASSEMBLY__ #define __PAGE_OFFSET CONFIG_PAGE_OFFSET -#define __PHYSICAL_START CONFIG_PHYSICAL_START #else #define __PAGE_OFFSET ((unsigned long)CONFIG_PAGE_OFFSET) -#define __PHYSICAL_START ((unsigned long)CONFIG_PHYSICAL_START) #endif -#define __KERNEL_START (__PAGE_OFFSET + __PHYSICAL_START) #define PAGE_OFFSET ((unsigned long)__PAGE_OFFSET) -- cgit v1.2.3-70-g09d2 From 968de4f02621db35b8ae5239c8cfc6664fb872d8 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Thu, 7 Dec 2006 02:14:04 +0100 Subject: [PATCH] i386: Relocatable kernel support This patch modifies the i386 kernel so that if CONFIG_RELOCATABLE is selected it will be able to be loaded at any 4K aligned address below 1G. The technique used is to compile the decompressor with -fPIC and modify it so the decompressor is fully relocatable. For the main kernel relocations are generated. Resulting in a kernel that is relocatable with no runtime overhead and no need to modify the source code. A reserved 32bit word in the parameters has been assigned to serve as a stack so we figure out where are running. Signed-off-by: Eric W. Biederman Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/i386/Kconfig | 12 + arch/i386/Makefile | 4 +- arch/i386/boot/compressed/Makefile | 28 +- arch/i386/boot/compressed/head.S | 184 +++++++---- arch/i386/boot/compressed/misc.c | 261 ++++++++-------- arch/i386/boot/compressed/relocs.c | 563 ++++++++++++++++++++++++++++++++++ arch/i386/boot/compressed/vmlinux.lds | 43 +++ arch/i386/boot/compressed/vmlinux.scr | 3 +- arch/i386/boot/setup.S | 29 +- include/linux/screen_info.h | 3 +- 10 files changed, 920 insertions(+), 210 deletions(-) create mode 100644 arch/i386/boot/compressed/relocs.c create mode 100644 arch/i386/boot/compressed/vmlinux.lds (limited to 'arch/i386') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 8ff1c6fb5aa..d588ca874bb 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -773,6 +773,18 @@ config CRASH_DUMP PHYSICAL_START. For more details see Documentation/kdump/kdump.txt +config RELOCATABLE + bool "Build a relocatable kernel" + help + This build a kernel image that retains relocation information + so it can be loaded someplace besides the default 1MB. + The relocations tend to the kernel binary about 10% larger, + but are discarded at runtime. + + One use is for the kexec on panic case where the recovery kernel + must live at a different physical address than the primary + kernel. + config PHYSICAL_START hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP) diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 0677908dfa0..d1aca52bf69 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -26,7 +26,9 @@ endif LDFLAGS := -m elf_i386 OBJCOPYFLAGS := -O binary -R .note -R .comment -S -LDFLAGS_vmlinux := +ifdef CONFIG_RELOCATABLE +LDFLAGS_vmlinux := --emit-relocs +endif CHECKFLAGS += -D__i386__ CFLAGS += -pipe -msoft-float diff --git a/arch/i386/boot/compressed/Makefile b/arch/i386/boot/compressed/Makefile index 258ea95224f..cc28da3a881 100644 --- a/arch/i386/boot/compressed/Makefile +++ b/arch/i386/boot/compressed/Makefile @@ -4,22 +4,42 @@ # create a compressed vmlinux image from the original vmlinux # -targets := vmlinux vmlinux.bin vmlinux.bin.gz head.o misc.o piggy.o +targets := vmlinux vmlinux.bin vmlinux.bin.gz head.o misc.o piggy.o \ + vmlinux.bin.all vmlinux.relocs EXTRA_AFLAGS := -traditional -LDFLAGS_vmlinux := -Ttext $(IMAGE_OFFSET) -e startup_32 +LDFLAGS_vmlinux := -T +CFLAGS_misc.o += -fPIC +hostprogs-y := relocs -$(obj)/vmlinux: $(obj)/head.o $(obj)/misc.o $(obj)/piggy.o FORCE +$(obj)/vmlinux: $(src)/vmlinux.lds $(obj)/head.o $(obj)/misc.o $(obj)/piggy.o FORCE $(call if_changed,ld) @: $(obj)/vmlinux.bin: vmlinux FORCE $(call if_changed,objcopy) +quiet_cmd_relocs = RELOCS $@ + cmd_relocs = $(obj)/relocs $< > $@ +$(obj)/vmlinux.relocs: vmlinux $(obj)/relocs FORCE + $(call if_changed,relocs) + +vmlinux.bin.all-y := $(obj)/vmlinux.bin +vmlinux.bin.all-$(CONFIG_RELOCATABLE) += $(obj)/vmlinux.relocs +quiet_cmd_relocbin = BUILD $@ + cmd_relocbin = cat $(filter-out FORCE,$^) > $@ +$(obj)/vmlinux.bin.all: $(vmlinux.bin.all-y) FORCE + $(call if_changed,relocbin) + +ifdef CONFIG_RELOCATABLE +$(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin.all FORCE + $(call if_changed,gzip) +else $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin FORCE $(call if_changed,gzip) +endif LDFLAGS_piggy.o := -r --format binary --oformat elf32-i386 -T -$(obj)/piggy.o: $(obj)/vmlinux.scr $(obj)/vmlinux.bin.gz FORCE +$(obj)/piggy.o: $(src)/vmlinux.scr $(obj)/vmlinux.bin.gz FORCE $(call if_changed,ld) diff --git a/arch/i386/boot/compressed/head.S b/arch/i386/boot/compressed/head.S index 40a8de8270a..e4dd7a6b9b0 100644 --- a/arch/i386/boot/compressed/head.S +++ b/arch/i386/boot/compressed/head.S @@ -25,9 +25,11 @@ #include #include +#include +.section ".text.head" .globl startup_32 - + startup_32: cld cli @@ -36,93 +38,141 @@ startup_32: movl %eax,%es movl %eax,%fs movl %eax,%gs + movl %eax,%ss - lss stack_start,%esp - xorl %eax,%eax -1: incl %eax # check that A20 really IS enabled - movl %eax,0x000000 # loop forever if it isn't - cmpl %eax,0x100000 - je 1b +/* Calculate the delta between where we were compiled to run + * at and where we were actually loaded at. This can only be done + * with a short local call on x86. Nothing else will tell us what + * address we are running at. The reserved chunk of the real-mode + * data at 0x34-0x3f are used as the stack for this calculation. + * Only 4 bytes are needed. + */ + leal 0x40(%esi), %esp + call 1f +1: popl %ebp + subl $1b, %ebp + +/* Compute the delta between where we were compiled to run at + * and where the code will actually run at. + */ + /* Start with the delta to where the kernel will run at. If we are + * a relocatable kernel this is the delta to our load address otherwise + * this is the delta to CONFIG_PHYSICAL start. + */ +#ifdef CONFIG_RELOCATABLE + movl %ebp, %ebx +#else + movl $(CONFIG_PHYSICAL_START - startup_32), %ebx +#endif + + /* Replace the compressed data size with the uncompressed size */ + subl input_len(%ebp), %ebx + movl output_len(%ebp), %eax + addl %eax, %ebx + /* Add 8 bytes for every 32K input block */ + shrl $12, %eax + addl %eax, %ebx + /* Add 32K + 18 bytes of extra slack */ + addl $(32768 + 18), %ebx + /* Align on a 4K boundary */ + addl $4095, %ebx + andl $~4095, %ebx + +/* Copy the compressed kernel to the end of our buffer + * where decompression in place becomes safe. + */ + pushl %esi + leal _end(%ebp), %esi + leal _end(%ebx), %edi + movl $(_end - startup_32), %ecx + std + rep + movsb + cld + popl %esi + +/* Compute the kernel start address. + */ +#ifdef CONFIG_RELOCATABLE + leal startup_32(%ebp), %ebp +#else + movl $CONFIG_PHYSICAL_START, %ebp +#endif /* - * Initialize eflags. Some BIOS's leave bits like NT set. This would - * confuse the debugger if this code is traced. - * XXX - best to initialize before switching to protected mode. + * Jump to the relocated address. */ - pushl $0 - popfl + leal relocated(%ebx), %eax + jmp *%eax +.section ".text" +relocated: + /* * Clear BSS */ xorl %eax,%eax - movl $_edata,%edi - movl $_end,%ecx + leal _edata(%ebx),%edi + leal _end(%ebx), %ecx subl %edi,%ecx cld rep stosb + +/* + * Setup the stack for the decompressor + */ + leal stack_end(%ebx), %esp + /* * Do the decompression, and jump to the new kernel.. */ - subl $16,%esp # place for structure on the stack - movl %esp,%eax + movl output_len(%ebx), %eax + pushl %eax + pushl %ebp # output address + movl input_len(%ebx), %eax + pushl %eax # input_len + leal input_data(%ebx), %eax + pushl %eax # input_data + leal _end(%ebx), %eax + pushl %eax # end of the image as third argument pushl %esi # real mode pointer as second arg - pushl %eax # address of structure as first arg call decompress_kernel - orl %eax,%eax - jnz 3f - popl %esi # discard address - popl %esi # real mode pointer - xorl %ebx,%ebx - ljmp $(__BOOT_CS), $CONFIG_PHYSICAL_START + addl $20, %esp + popl %ecx + +#if CONFIG_RELOCATABLE +/* Find the address of the relocations. + */ + movl %ebp, %edi + addl %ecx, %edi + +/* Calculate the delta between where vmlinux was compiled to run + * and where it was actually loaded. + */ + movl %ebp, %ebx + subl $CONFIG_PHYSICAL_START, %ebx /* - * We come here, if we were loaded high. - * We need to move the move-in-place routine down to 0x1000 - * and then start it with the buffer addresses in registers, - * which we got from the stack. + * Process relocations. */ -3: - movl $move_routine_start,%esi - movl $0x1000,%edi - movl $move_routine_end,%ecx - subl %esi,%ecx - addl $3,%ecx - shrl $2,%ecx - cld - rep - movsl - - popl %esi # discard the address - popl %ebx # real mode pointer - popl %esi # low_buffer_start - popl %ecx # lcount - popl %edx # high_buffer_start - popl %eax # hcount - movl $CONFIG_PHYSICAL_START,%edi - cli # make sure we don't get interrupted - ljmp $(__BOOT_CS), $0x1000 # and jump to the move routine + +1: subl $4, %edi + movl 0(%edi), %ecx + testl %ecx, %ecx + jz 2f + addl %ebx, -__PAGE_OFFSET(%ebx, %ecx) + jmp 1b +2: +#endif /* - * Routine (template) for moving the decompressed kernel in place, - * if we were high loaded. This _must_ PIC-code ! + * Jump to the decompressed kernel. */ -move_routine_start: - movl %ecx,%ebp - shrl $2,%ecx - rep - movsl - movl %ebp,%ecx - andl $3,%ecx - rep - movsb - movl %edx,%esi - movl %eax,%ecx # NOTE: rep movsb won't move if %ecx == 0 - addl $3,%ecx - shrl $2,%ecx - rep - movsl - movl %ebx,%esi # Restore setup pointer xorl %ebx,%ebx - ljmp $(__BOOT_CS), $CONFIG_PHYSICAL_START -move_routine_end: + jmp *%ebp + +.bss +.balign 4 +stack: + .fill 4096, 1, 0 +stack_end: diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c index 20970ff4411..4eac24e95a1 100644 --- a/arch/i386/boot/compressed/misc.c +++ b/arch/i386/boot/compressed/misc.c @@ -13,6 +13,88 @@ #include #include #include +#include + +/* WARNING!! + * This code is compiled with -fPIC and it is relocated dynamically + * at run time, but no relocation processing is performed. + * This means that it is not safe to place pointers in static structures. + */ + +/* + * Getting to provable safe in place decompression is hard. + * Worst case behaviours need to be analized. + * Background information: + * + * The file layout is: + * magic[2] + * method[1] + * flags[1] + * timestamp[4] + * extraflags[1] + * os[1] + * compressed data blocks[N] + * crc[4] orig_len[4] + * + * resulting in 18 bytes of non compressed data overhead. + * + * Files divided into blocks + * 1 bit (last block flag) + * 2 bits (block type) + * + * 1 block occurs every 32K -1 bytes or when there 50% compression has been achieved. + * The smallest block type encoding is always used. + * + * stored: + * 32 bits length in bytes. + * + * fixed: + * magic fixed tree. + * symbols. + * + * dynamic: + * dynamic tree encoding. + * symbols. + * + * + * The buffer for decompression in place is the length of the + * uncompressed data, plus a small amount extra to keep the algorithm safe. + * The compressed data is placed at the end of the buffer. The output + * pointer is placed at the start of the buffer and the input pointer + * is placed where the compressed data starts. Problems will occur + * when the output pointer overruns the input pointer. + * + * The output pointer can only overrun the input pointer if the input + * pointer is moving faster than the output pointer. A condition only + * triggered by data whose compressed form is larger than the uncompressed + * form. + * + * The worst case at the block level is a growth of the compressed data + * of 5 bytes per 32767 bytes. + * + * The worst case internal to a compressed block is very hard to figure. + * The worst case can at least be boundined by having one bit that represents + * 32764 bytes and then all of the rest of the bytes representing the very + * very last byte. + * + * All of which is enough to compute an amount of extra data that is required + * to be safe. To avoid problems at the block level allocating 5 extra bytes + * per 32767 bytes of data is sufficient. To avoind problems internal to a block + * adding an extra 32767 bytes (the worst case uncompressed block size) is + * sufficient, to ensure that in the worst case the decompressed data for + * block will stop the byte before the compressed data for a block begins. + * To avoid problems with the compressed data's meta information an extra 18 + * bytes are needed. Leading to the formula: + * + * extra_bytes = (uncompressed_size >> 12) + 32768 + 18 + decompressor_size. + * + * Adding 8 bytes per 32K is a bit excessive but much easier to calculate. + * Adding 32768 instead of 32767 just makes for round numbers. + * Adding the decompressor_size is necessary as it musht live after all + * of the data as well. Last I measured the decompressor is about 14K. + * 10K of actuall data and 4K of bss. + * + */ /* * gzip declarations @@ -29,15 +111,20 @@ typedef unsigned char uch; typedef unsigned short ush; typedef unsigned long ulg; -#define WSIZE 0x8000 /* Window size must be at least 32k, */ - /* and a power of two */ +#define WSIZE 0x80000000 /* Window size must be at least 32k, + * and a power of two + * We don't actually have a window just + * a huge output buffer so I report + * a 2G windows size, as that should + * always be larger than our output buffer. + */ -static uch *inbuf; /* input buffer */ -static uch window[WSIZE]; /* Sliding window buffer */ +static uch *inbuf; /* input buffer */ +static uch *window; /* Sliding window buffer, (and final output buffer) */ -static unsigned insize = 0; /* valid bytes in inbuf */ -static unsigned inptr = 0; /* index of next byte to be processed in inbuf */ -static unsigned outcnt = 0; /* bytes in output buffer */ +static unsigned insize; /* valid bytes in inbuf */ +static unsigned inptr; /* index of next byte to be processed in inbuf */ +static unsigned outcnt; /* bytes in output buffer */ /* gzip flag byte */ #define ASCII_FLAG 0x01 /* bit 0 set: file probably ASCII text */ @@ -88,8 +175,6 @@ extern unsigned char input_data[]; extern int input_len; static long bytes_out = 0; -static uch *output_data; -static unsigned long output_ptr = 0; static void *malloc(int size); static void free(void *where); @@ -99,17 +184,10 @@ static void *memcpy(void *dest, const void *src, unsigned n); static void putstr(const char *); -extern int end; -static long free_mem_ptr = (long)&end; -static long free_mem_end_ptr; +static unsigned long free_mem_ptr; +static unsigned long free_mem_end_ptr; -#define INPLACE_MOVE_ROUTINE 0x1000 -#define LOW_BUFFER_START 0x2000 -#define LOW_BUFFER_MAX 0x90000 #define HEAP_SIZE 0x3000 -static unsigned int low_buffer_end, low_buffer_size; -static int high_loaded =0; -static uch *high_buffer_start /* = (uch *)(((ulg)&end) + HEAP_SIZE)*/; static char *vidmem = (char *)0xb8000; static int vidport; @@ -150,7 +228,7 @@ static void gzip_mark(void **ptr) static void gzip_release(void **ptr) { - free_mem_ptr = (long) *ptr; + free_mem_ptr = (unsigned long) *ptr; } static void scroll(void) @@ -178,7 +256,7 @@ static void putstr(const char *s) y--; } } else { - vidmem [ ( x + cols * y ) * 2 ] = c; + vidmem [ ( x + cols * y ) * 2 ] = c; if ( ++x >= cols ) { x = 0; if ( ++y >= lines ) { @@ -223,58 +301,31 @@ static void* memcpy(void* dest, const void* src, unsigned n) */ static int fill_inbuf(void) { - if (insize != 0) { - error("ran out of input data"); - } - - inbuf = input_data; - insize = input_len; - inptr = 1; - return inbuf[0]; + error("ran out of input data"); + return 0; } /* =========================================================================== * Write the output window window[0..outcnt-1] and update crc and bytes_out. * (Used for the decompressed data only.) */ -static void flush_window_low(void) -{ - ulg c = crc; /* temporary variable */ - unsigned n; - uch *in, *out, ch; - - in = window; - out = &output_data[output_ptr]; - for (n = 0; n < outcnt; n++) { - ch = *out++ = *in++; - c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8); - } - crc = c; - bytes_out += (ulg)outcnt; - output_ptr += (ulg)outcnt; - outcnt = 0; -} - -static void flush_window_high(void) -{ - ulg c = crc; /* temporary variable */ - unsigned n; - uch *in, ch; - in = window; - for (n = 0; n < outcnt; n++) { - ch = *output_data++ = *in++; - if ((ulg)output_data == low_buffer_end) output_data=high_buffer_start; - c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8); - } - crc = c; - bytes_out += (ulg)outcnt; - outcnt = 0; -} - static void flush_window(void) { - if (high_loaded) flush_window_high(); - else flush_window_low(); + /* With my window equal to my output buffer + * I only need to compute the crc here. + */ + ulg c = crc; /* temporary variable */ + unsigned n; + uch *in, ch; + + in = window; + for (n = 0; n < outcnt; n++) { + ch = *in++; + c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8); + } + crc = c; + bytes_out += (ulg)outcnt; + outcnt = 0; } static void error(char *x) @@ -286,66 +337,8 @@ static void error(char *x) while(1); /* Halt */ } -#define STACK_SIZE (4096) - -long user_stack [STACK_SIZE]; - -struct { - long * a; - short b; - } stack_start = { & user_stack [STACK_SIZE] , __BOOT_DS }; - -static void setup_normal_output_buffer(void) -{ -#ifdef STANDARD_MEMORY_BIOS_CALL - if (RM_EXT_MEM_K < 1024) error("Less than 2MB of memory"); -#else - if ((RM_ALT_MEM_K > RM_EXT_MEM_K ? RM_ALT_MEM_K : RM_EXT_MEM_K) < 1024) error("Less than 2MB of memory"); -#endif - output_data = (unsigned char *)CONFIG_PHYSICAL_START; /* Normally Points to 1M */ - free_mem_end_ptr = (long)real_mode; -} - -struct moveparams { - uch *low_buffer_start; int lcount; - uch *high_buffer_start; int hcount; -}; - -static void setup_output_buffer_if_we_run_high(struct moveparams *mv) -{ - high_buffer_start = (uch *)(((ulg)&end) + HEAP_SIZE); -#ifdef STANDARD_MEMORY_BIOS_CALL - if (RM_EXT_MEM_K < (3*1024)) error("Less than 4MB of memory"); -#else - if ((RM_ALT_MEM_K > RM_EXT_MEM_K ? RM_ALT_MEM_K : RM_EXT_MEM_K) < (3*1024)) error("Less than 4MB of memory"); -#endif - mv->low_buffer_start = output_data = (unsigned char *)LOW_BUFFER_START; - low_buffer_end = ((unsigned int)real_mode > LOW_BUFFER_MAX - ? LOW_BUFFER_MAX : (unsigned int)real_mode) & ~0xfff; - low_buffer_size = low_buffer_end - LOW_BUFFER_START; - high_loaded = 1; - free_mem_end_ptr = (long)high_buffer_start; - if ( (CONFIG_PHYSICAL_START + low_buffer_size) > ((ulg)high_buffer_start)) { - high_buffer_start = (uch *)(CONFIG_PHYSICAL_START + low_buffer_size); - mv->hcount = 0; /* say: we need not to move high_buffer */ - } - else mv->hcount = -1; - mv->high_buffer_start = high_buffer_start; -} - -static void close_output_buffer_if_we_run_high(struct moveparams *mv) -{ - if (bytes_out > low_buffer_size) { - mv->lcount = low_buffer_size; - if (mv->hcount) - mv->hcount = bytes_out - low_buffer_size; - } else { - mv->lcount = bytes_out; - mv->hcount = 0; - } -} - -asmlinkage int decompress_kernel(struct moveparams *mv, void *rmode) +asmlinkage void decompress_kernel(void *rmode, unsigned long end, + uch *input_data, unsigned long input_len, uch *output) { real_mode = rmode; @@ -360,13 +353,25 @@ asmlinkage int decompress_kernel(struct moveparams *mv, void *rmode) lines = RM_SCREEN_INFO.orig_video_lines; cols = RM_SCREEN_INFO.orig_video_cols; - if (free_mem_ptr < 0x100000) setup_normal_output_buffer(); - else setup_output_buffer_if_we_run_high(mv); + window = output; /* Output buffer (Normally at 1M) */ + free_mem_ptr = end; /* Heap */ + free_mem_end_ptr = end + HEAP_SIZE; + inbuf = input_data; /* Input buffer */ + insize = input_len; + inptr = 0; + + if (((u32)output - CONFIG_PHYSICAL_START) & 0x3fffff) + error("Destination address not 4M aligned"); + if (end > ((-__PAGE_OFFSET-(512 <<20)-1) & 0x7fffffff)) + error("Destination address too large"); +#ifndef CONFIG_RELOCATABLE + if ((u32)output != CONFIG_PHYSICAL_START) + error("Wrong destination address"); +#endif makecrc(); putstr("Uncompressing Linux... "); gunzip(); putstr("Ok, booting the kernel.\n"); - if (high_loaded) close_output_buffer_if_we_run_high(mv); - return high_loaded; + return; } diff --git a/arch/i386/boot/compressed/relocs.c b/arch/i386/boot/compressed/relocs.c new file mode 100644 index 00000000000..0551ceb21be --- /dev/null +++ b/arch/i386/boot/compressed/relocs.c @@ -0,0 +1,563 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#define USE_BSD +#include + +#define MAX_SHDRS 100 +static Elf32_Ehdr ehdr; +static Elf32_Shdr shdr[MAX_SHDRS]; +static Elf32_Sym *symtab[MAX_SHDRS]; +static Elf32_Rel *reltab[MAX_SHDRS]; +static char *strtab[MAX_SHDRS]; +static unsigned long reloc_count, reloc_idx; +static unsigned long *relocs; + +static void die(char *fmt, ...) +{ + va_list ap; + va_start(ap, fmt); + vfprintf(stderr, fmt, ap); + va_end(ap); + exit(1); +} + +static const char *sym_type(unsigned type) +{ + static const char *type_name[] = { +#define SYM_TYPE(X) [X] = #X + SYM_TYPE(STT_NOTYPE), + SYM_TYPE(STT_OBJECT), + SYM_TYPE(STT_FUNC), + SYM_TYPE(STT_SECTION), + SYM_TYPE(STT_FILE), + SYM_TYPE(STT_COMMON), + SYM_TYPE(STT_TLS), +#undef SYM_TYPE + }; + const char *name = "unknown sym type name"; + if (type < sizeof(type_name)/sizeof(type_name[0])) { + name = type_name[type]; + } + return name; +} + +static const char *sym_bind(unsigned bind) +{ + static const char *bind_name[] = { +#define SYM_BIND(X) [X] = #X + SYM_BIND(STB_LOCAL), + SYM_BIND(STB_GLOBAL), + SYM_BIND(STB_WEAK), +#undef SYM_BIND + }; + const char *name = "unknown sym bind name"; + if (bind < sizeof(bind_name)/sizeof(bind_name[0])) { + name = bind_name[bind]; + } + return name; +} + +static const char *sym_visibility(unsigned visibility) +{ + static const char *visibility_name[] = { +#define SYM_VISIBILITY(X) [X] = #X + SYM_VISIBILITY(STV_DEFAULT), + SYM_VISIBILITY(STV_INTERNAL), + SYM_VISIBILITY(STV_HIDDEN), + SYM_VISIBILITY(STV_PROTECTED), +#undef SYM_VISIBILITY + }; + const char *name = "unknown sym visibility name"; + if (visibility < sizeof(visibility_name)/sizeof(visibility_name[0])) { + name = visibility_name[visibility]; + } + return name; +} + +static const char *rel_type(unsigned type) +{ + static const char *type_name[] = { +#define REL_TYPE(X) [X] = #X + REL_TYPE(R_386_NONE), + REL_TYPE(R_386_32), + REL_TYPE(R_386_PC32), + REL_TYPE(R_386_GOT32), + REL_TYPE(R_386_PLT32), + REL_TYPE(R_386_COPY), + REL_TYPE(R_386_GLOB_DAT), + REL_TYPE(R_386_JMP_SLOT), + REL_TYPE(R_386_RELATIVE), + REL_TYPE(R_386_GOTOFF), + REL_TYPE(R_386_GOTPC), +#undef REL_TYPE + }; + const char *name = "unknown type rel type name"; + if (type < sizeof(type_name)/sizeof(type_name[0])) { + name = type_name[type]; + } + return name; +} + +static const char *sec_name(unsigned shndx) +{ + const char *sec_strtab; + const char *name; + sec_strtab = strtab[ehdr.e_shstrndx]; + name = ""; + if (shndx < ehdr.e_shnum) { + name = sec_strtab + shdr[shndx].sh_name; + } + else if (shndx == SHN_ABS) { + name = "ABSOLUTE"; + } + else if (shndx == SHN_COMMON) { + name = "COMMON"; + } + return name; +} + +static const char *sym_name(const char *sym_strtab, Elf32_Sym *sym) +{ + const char *name; + name = ""; + if (sym->st_name) { + name = sym_strtab + sym->st_name; + } + else { + name = sec_name(shdr[sym->st_shndx].sh_name); + } + return name; +} + + + +#if BYTE_ORDER == LITTLE_ENDIAN +#define le16_to_cpu(val) (val) +#define le32_to_cpu(val) (val) +#endif +#if BYTE_ORDER == BIG_ENDIAN +#define le16_to_cpu(val) bswap_16(val) +#define le32_to_cpu(val) bswap_32(val) +#endif + +static uint16_t elf16_to_cpu(uint16_t val) +{ + return le16_to_cpu(val); +} + +static uint32_t elf32_to_cpu(uint32_t val) +{ + return le32_to_cpu(val); +} + +static void read_ehdr(FILE *fp) +{ + if (fread(&ehdr, sizeof(ehdr), 1, fp) != 1) { + die("Cannot read ELF header: %s\n", + strerror(errno)); + } + if (memcmp(ehdr.e_ident, ELFMAG, 4) != 0) { + die("No ELF magic\n"); + } + if (ehdr.e_ident[EI_CLASS] != ELFCLASS32) { + die("Not a 32 bit executable\n"); + } + if (ehdr.e_ident[EI_DATA] != ELFDATA2LSB) { + die("Not a LSB ELF executable\n"); + } + if (ehdr.e_ident[EI_VERSION] != EV_CURRENT) { + die("Unknown ELF version\n"); + } + /* Convert the fields to native endian */ + ehdr.e_type = elf16_to_cpu(ehdr.e_type); + ehdr.e_machine = elf16_to_cpu(ehdr.e_machine); + ehdr.e_version = elf32_to_cpu(ehdr.e_version); + ehdr.e_entry = elf32_to_cpu(ehdr.e_entry); + ehdr.e_phoff = elf32_to_cpu(ehdr.e_phoff); + ehdr.e_shoff = elf32_to_cpu(ehdr.e_shoff); + ehdr.e_flags = elf32_to_cpu(ehdr.e_flags); + ehdr.e_ehsize = elf16_to_cpu(ehdr.e_ehsize); + ehdr.e_phentsize = elf16_to_cpu(ehdr.e_phentsize); + ehdr.e_phnum = elf16_to_cpu(ehdr.e_phnum); + ehdr.e_shentsize = elf16_to_cpu(ehdr.e_shentsize); + ehdr.e_shnum = elf16_to_cpu(ehdr.e_shnum); + ehdr.e_shstrndx = elf16_to_cpu(ehdr.e_shstrndx); + + if ((ehdr.e_type != ET_EXEC) && (ehdr.e_type != ET_DYN)) { + die("Unsupported ELF header type\n"); + } + if (ehdr.e_machine != EM_386) { + die("Not for x86\n"); + } + if (ehdr.e_version != EV_CURRENT) { + die("Unknown ELF version\n"); + } + if (ehdr.e_ehsize != sizeof(Elf32_Ehdr)) { + die("Bad Elf header size\n"); + } + if (ehdr.e_phentsize != sizeof(Elf32_Phdr)) { + die("Bad program header entry\n"); + } + if (ehdr.e_shentsize != sizeof(Elf32_Shdr)) { + die("Bad section header entry\n"); + } + if (ehdr.e_shstrndx >= ehdr.e_shnum) { + die("String table index out of bounds\n"); + } +} + +static void read_shdrs(FILE *fp) +{ + int i; + if (ehdr.e_shnum > MAX_SHDRS) { + die("%d section headers supported: %d\n", + ehdr.e_shnum, MAX_SHDRS); + } + if (fseek(fp, ehdr.e_shoff, SEEK_SET) < 0) { + die("Seek to %d failed: %s\n", + ehdr.e_shoff, strerror(errno)); + } + if (fread(&shdr, sizeof(shdr[0]), ehdr.e_shnum, fp) != ehdr.e_shnum) { + die("Cannot read ELF section headers: %s\n", + strerror(errno)); + } + for(i = 0; i < ehdr.e_shnum; i++) { + shdr[i].sh_name = elf32_to_cpu(shdr[i].sh_name); + shdr[i].sh_type = elf32_to_cpu(shdr[i].sh_type); + shdr[i].sh_flags = elf32_to_cpu(shdr[i].sh_flags); + shdr[i].sh_addr = elf32_to_cpu(shdr[i].sh_addr); + shdr[i].sh_offset = elf32_to_cpu(shdr[i].sh_offset); + shdr[i].sh_size = elf32_to_cpu(shdr[i].sh_size); + shdr[i].sh_link = elf32_to_cpu(shdr[i].sh_link); + shdr[i].sh_info = elf32_to_cpu(shdr[i].sh_info); + shdr[i].sh_addralign = elf32_to_cpu(shdr[i].sh_addralign); + shdr[i].sh_entsize = elf32_to_cpu(shdr[i].sh_entsize); + } + +} + +static void read_strtabs(FILE *fp) +{ + int i; + for(i = 0; i < ehdr.e_shnum; i++) { + if (shdr[i].sh_type != SHT_STRTAB) { + continue; + } + strtab[i] = malloc(shdr[i].sh_size); + if (!strtab[i]) { + die("malloc of %d bytes for strtab failed\n", + shdr[i].sh_size); + } + if (fseek(fp, shdr[i].sh_offset, SEEK_SET) < 0) { + die("Seek to %d failed: %s\n", + shdr[i].sh_offset, strerror(errno)); + } + if (fread(strtab[i], 1, shdr[i].sh_size, fp) != shdr[i].sh_size) { + die("Cannot read symbol table: %s\n", + strerror(errno)); + } + } +} + +static void read_symtabs(FILE *fp) +{ + int i,j; + for(i = 0; i < ehdr.e_shnum; i++) { + if (shdr[i].sh_type != SHT_SYMTAB) { + continue; + } + symtab[i] = malloc(shdr[i].sh_size); + if (!symtab[i]) { + die("malloc of %d bytes for symtab failed\n", + shdr[i].sh_size); + } + if (fseek(fp, shdr[i].sh_offset, SEEK_SET) < 0) { + die("Seek to %d failed: %s\n", + shdr[i].sh_offset, strerror(errno)); + } + if (fread(symtab[i], 1, shdr[i].sh_size, fp) != shdr[i].sh_size) { + die("Cannot read symbol table: %s\n", + strerror(errno)); + } + for(j = 0; j < shdr[i].sh_size/sizeof(symtab[i][0]); j++) { + symtab[i][j].st_name = elf32_to_cpu(symtab[i][j].st_name); + symtab[i][j].st_value = elf32_to_cpu(symtab[i][j].st_value); + symtab[i][j].st_size = elf32_to_cpu(symtab[i][j].st_size); + symtab[i][j].st_shndx = elf16_to_cpu(symtab[i][j].st_shndx); + } + } +} + + +static void read_relocs(FILE *fp) +{ + int i,j; + for(i = 0; i < ehdr.e_shnum; i++) { + if (shdr[i].sh_type != SHT_REL) { + continue; + } + reltab[i] = malloc(shdr[i].sh_size); + if (!reltab[i]) { + die("malloc of %d bytes for relocs failed\n", + shdr[i].sh_size); + } + if (fseek(fp, shdr[i].sh_offset, SEEK_SET) < 0) { + die("Seek to %d failed: %s\n", + shdr[i].sh_offset, strerror(errno)); + } + if (fread(reltab[i], 1, shdr[i].sh_size, fp) != shdr[i].sh_size) { + die("Cannot read symbol table: %s\n", + strerror(errno)); + } + for(j = 0; j < shdr[i].sh_size/sizeof(reltab[0][0]); j++) { + reltab[i][j].r_offset = elf32_to_cpu(reltab[i][j].r_offset); + reltab[i][j].r_info = elf32_to_cpu(reltab[i][j].r_info); + } + } +} + + +static void print_absolute_symbols(void) +{ + int i; + printf("Absolute symbols\n"); + printf(" Num: Value Size Type Bind Visibility Name\n"); + for(i = 0; i < ehdr.e_shnum; i++) { + char *sym_strtab; + Elf32_Sym *sh_symtab; + int j; + if (shdr[i].sh_type != SHT_SYMTAB) { + continue; + } + sh_symtab = symtab[i]; + sym_strtab = strtab[shdr[i].sh_link]; + for(j = 0; j < shdr[i].sh_size/sizeof(symtab[0][0]); j++) { + Elf32_Sym *sym; + const char *name; + sym = &symtab[i][j]; + name = sym_name(sym_strtab, sym); + if (sym->st_shndx != SHN_ABS) { + continue; + } + printf("%5d %08x %5d %10s %10s %12s %s\n", + j, sym->st_value, sym->st_size, + sym_type(ELF32_ST_TYPE(sym->st_info)), + sym_bind(ELF32_ST_BIND(sym->st_info)), + sym_visibility(ELF32_ST_VISIBILITY(sym->st_other)), + name); + } + } + printf("\n"); +} + +static void print_absolute_relocs(void) +{ + int i; + printf("Absolute relocations\n"); + printf("Offset Info Type Sym.Value Sym.Name\n"); + for(i = 0; i < ehdr.e_shnum; i++) { + char *sym_strtab; + Elf32_Sym *sh_symtab; + unsigned sec_applies, sec_symtab; + int j; + if (shdr[i].sh_type != SHT_REL) { + continue; + } + sec_symtab = shdr[i].sh_link; + sec_applies = shdr[i].sh_info; + if (!(shdr[sec_applies].sh_flags & SHF_ALLOC)) { + continue; + } + sh_symtab = symtab[sec_symtab]; + sym_strtab = strtab[shdr[sec_symtab].sh_link]; + for(j = 0; j < shdr[i].sh_size/sizeof(reltab[0][0]); j++) { + Elf32_Rel *rel; + Elf32_Sym *sym; + const char *name; + rel = &reltab[i][j]; + sym = &sh_symtab[ELF32_R_SYM(rel->r_info)]; + name = sym_name(sym_strtab, sym); + if (sym->st_shndx != SHN_ABS) { + continue; + } + printf("%08x %08x %10s %08x %s\n", + rel->r_offset, + rel->r_info, + rel_type(ELF32_R_TYPE(rel->r_info)), + sym->st_value, + name); + } + } + printf("\n"); +} + +static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym)) +{ + int i; + /* Walk through the relocations */ + for(i = 0; i < ehdr.e_shnum; i++) { + char *sym_strtab; + Elf32_Sym *sh_symtab; + unsigned sec_applies, sec_symtab; + int j; + if (shdr[i].sh_type != SHT_REL) { + continue; + } + sec_symtab = shdr[i].sh_link; + sec_applies = shdr[i].sh_info; + if (!(shdr[sec_applies].sh_flags & SHF_ALLOC)) { + continue; + } + sh_symtab = symtab[sec_symtab]; + sym_strtab = strtab[shdr[sec_symtab].sh_link]; + for(j = 0; j < shdr[i].sh_size/sizeof(reltab[0][0]); j++) { + Elf32_Rel *rel; + Elf32_Sym *sym; + unsigned r_type; + rel = &reltab[i][j]; + sym = &sh_symtab[ELF32_R_SYM(rel->r_info)]; + r_type = ELF32_R_TYPE(rel->r_info); + /* Don't visit relocations to absolute symbols */ + if (sym->st_shndx == SHN_ABS) { + continue; + } + if (r_type == R_386_PC32) { + /* PC relative relocations don't need to be adjusted */ + } + else if (r_type == R_386_32) { + /* Visit relocations that need to be adjusted */ + visit(rel, sym); + } + else { + die("Unsupported relocation type: %d\n", r_type); + } + } + } +} + +static void count_reloc(Elf32_Rel *rel, Elf32_Sym *sym) +{ + reloc_count += 1; +} + +static void collect_reloc(Elf32_Rel *rel, Elf32_Sym *sym) +{ + /* Remember the address that needs to be adjusted. */ + relocs[reloc_idx++] = rel->r_offset; +} + +static int cmp_relocs(const void *va, const void *vb) +{ + const unsigned long *a, *b; + a = va; b = vb; + return (*a == *b)? 0 : (*a > *b)? 1 : -1; +} + +static void emit_relocs(int as_text) +{ + int i; + /* Count how many relocations I have and allocate space for them. */ + reloc_count = 0; + walk_relocs(count_reloc); + relocs = malloc(reloc_count * sizeof(relocs[0])); + if (!relocs) { + die("malloc of %d entries for relocs failed\n", + reloc_count); + } + /* Collect up the relocations */ + reloc_idx = 0; + walk_relocs(collect_reloc); + + /* Order the relocations for more efficient processing */ + qsort(relocs, reloc_count, sizeof(relocs[0]), cmp_relocs); + + /* Print the relocations */ + if (as_text) { + /* Print the relocations in a form suitable that + * gas will like. + */ + printf(".section \".data.reloc\",\"a\"\n"); + printf(".balign 4\n"); + for(i = 0; i < reloc_count; i++) { + printf("\t .long 0x%08lx\n", relocs[i]); + } + printf("\n"); + } + else { + unsigned char buf[4]; + buf[0] = buf[1] = buf[2] = buf[3] = 0; + /* Print a stop */ + printf("%c%c%c%c", buf[0], buf[1], buf[2], buf[3]); + /* Now print each relocation */ + for(i = 0; i < reloc_count; i++) { + buf[0] = (relocs[i] >> 0) & 0xff; + buf[1] = (relocs[i] >> 8) & 0xff; + buf[2] = (relocs[i] >> 16) & 0xff; + buf[3] = (relocs[i] >> 24) & 0xff; + printf("%c%c%c%c", buf[0], buf[1], buf[2], buf[3]); + } + } +} + +static void usage(void) +{ + die("i386_reloc [--abs | --text] vmlinux\n"); +} + +int main(int argc, char **argv) +{ + int show_absolute; + int as_text; + const char *fname; + FILE *fp; + int i; + + show_absolute = 0; + as_text = 0; + fname = NULL; + for(i = 1; i < argc; i++) { + char *arg = argv[i]; + if (*arg == '-') { + if (strcmp(argv[1], "--abs") == 0) { + show_absolute = 1; + continue; + } + else if (strcmp(argv[1], "--text") == 0) { + as_text = 1; + continue; + } + } + else if (!fname) { + fname = arg; + continue; + } + usage(); + } + if (!fname) { + usage(); + } + fp = fopen(fname, "r"); + if (!fp) { + die("Cannot open %s: %s\n", + fname, strerror(errno)); + } + read_ehdr(fp); + read_shdrs(fp); + read_strtabs(fp); + read_symtabs(fp); + read_relocs(fp); + if (show_absolute) { + print_absolute_symbols(); + print_absolute_relocs(); + return 0; + } + emit_relocs(as_text); + return 0; +} diff --git a/arch/i386/boot/compressed/vmlinux.lds b/arch/i386/boot/compressed/vmlinux.lds new file mode 100644 index 00000000000..cc4854f6c6c --- /dev/null +++ b/arch/i386/boot/compressed/vmlinux.lds @@ -0,0 +1,43 @@ +OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386") +OUTPUT_ARCH(i386) +ENTRY(startup_32) +SECTIONS +{ + /* Be careful parts of head.S assume startup_32 is at + * address 0. + */ + . = 0 ; + .text.head : { + _head = . ; + *(.text.head) + _ehead = . ; + } + .data.compressed : { + *(.data.compressed) + } + .text : { + _text = .; /* Text */ + *(.text) + *(.text.*) + _etext = . ; + } + .rodata : { + _rodata = . ; + *(.rodata) /* read-only data */ + *(.rodata.*) + _erodata = . ; + } + .data : { + _data = . ; + *(.data) + *(.data.*) + _edata = . ; + } + .bss : { + _bss = . ; + *(.bss) + *(.bss.*) + *(COMMON) + _end = . ; + } +} diff --git a/arch/i386/boot/compressed/vmlinux.scr b/arch/i386/boot/compressed/vmlinux.scr index 1ed9d791f86..707a88f7f29 100644 --- a/arch/i386/boot/compressed/vmlinux.scr +++ b/arch/i386/boot/compressed/vmlinux.scr @@ -1,9 +1,10 @@ SECTIONS { - .data : { + .data.compressed : { input_len = .; LONG(input_data_end - input_data) input_data = .; *(.data) + output_len = . - 4; input_data_end = .; } } diff --git a/arch/i386/boot/setup.S b/arch/i386/boot/setup.S index 3aec4538a11..9aa8b051818 100644 --- a/arch/i386/boot/setup.S +++ b/arch/i386/boot/setup.S @@ -588,11 +588,6 @@ rmodeswtch_normal: call default_switch rmodeswtch_end: -# we get the code32 start address and modify the below 'jmpi' -# (loader may have changed it) - movl %cs:code32_start, %eax - movl %eax, %cs:code32 - # Now we move the system to its rightful place ... but we check if we have a # big-kernel. In that case we *must* not move it ... testb $LOADED_HIGH, %cs:loadflags @@ -788,11 +783,12 @@ a20_err_msg: a20_done: #endif /* CONFIG_X86_VOYAGER */ -# set up gdt and idt +# set up gdt and idt and 32bit start address lidt idt_48 # load idt with 0,0 xorl %eax, %eax # Compute gdt_base movw %ds, %ax # (Convert %ds:gdt to a linear ptr) shll $4, %eax + addl %eax, code32 addl $gdt, %eax movl %eax, (gdt_48+2) lgdt gdt_48 # load gdt with whatever is @@ -851,9 +847,26 @@ flush_instr: # Manual, Mixing 16-bit and 32-bit code, page 16-6) .byte 0x66, 0xea # prefix + jmpi-opcode -code32: .long 0x1000 # will be set to 0x100000 - # for big kernels +code32: .long startup_32 # will be set to %cs+startup_32 .word __BOOT_CS +.code32 +startup_32: + movl $(__BOOT_DS), %eax + movl %eax, %ds + movl %eax, %es + movl %eax, %fs + movl %eax, %gs + movl %eax, %ss + + xorl %eax, %eax +1: incl %eax # check that A20 really IS enabled + movl %eax, 0x00000000 # loop forever if it isn't + cmpl %eax, 0x00100000 + je 1b + + # Jump to the 32bit entry point + jmpl *(code32_start - start + (DELTA_INITSEG << 4))(%esi) +.code16 # Here's a bunch of information about your current kernel.. kernel_version: .ascii UTS_RELEASE diff --git a/include/linux/screen_info.h b/include/linux/screen_info.h index 2925e66a673..b02308ee766 100644 --- a/include/linux/screen_info.h +++ b/include/linux/screen_info.h @@ -42,7 +42,8 @@ struct screen_info { u16 pages; /* 0x32 */ u16 vesa_attributes; /* 0x34 */ u32 capabilities; /* 0x36 */ - /* 0x3a -- 0x3f reserved for future expansion */ + /* 0x3a -- 0x3b reserved for future expansion */ + /* 0x3c -- 0x3f micro stack for relocatable kernels */ }; extern struct screen_info screen_info; -- cgit v1.2.3-70-g09d2 From 6a044b3a0a1829ef19bb29548ffe553f48e8d80c Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Thu, 7 Dec 2006 02:14:04 +0100 Subject: [PATCH] i386: Warn upon absolute relocations being present o Relocations generated w.r.t absolute symbols are not processed as by definition, absolute symbols are not to be relocated. Explicitly warn user about absolutions relocations present at compile time. o These relocations get introduced either due to linker optimizations or some programming oversights. o Also create a list of symbols which have been audited to be safe and don't emit warnings for these. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/i386/boot/compressed/Makefile | 2 +- arch/i386/boot/compressed/relocs.c | 82 +++++++++++++++++++++++++++++++++----- 2 files changed, 73 insertions(+), 11 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/boot/compressed/Makefile b/arch/i386/boot/compressed/Makefile index cc28da3a881..a661217f33e 100644 --- a/arch/i386/boot/compressed/Makefile +++ b/arch/i386/boot/compressed/Makefile @@ -20,7 +20,7 @@ $(obj)/vmlinux.bin: vmlinux FORCE $(call if_changed,objcopy) quiet_cmd_relocs = RELOCS $@ - cmd_relocs = $(obj)/relocs $< > $@ + cmd_relocs = $(obj)/relocs $< > $@;$(obj)/relocs --abs-relocs $< $(obj)/vmlinux.relocs: vmlinux $(obj)/relocs FORCE $(call if_changed,relocs) diff --git a/arch/i386/boot/compressed/relocs.c b/arch/i386/boot/compressed/relocs.c index 0551ceb21be..468da89153c 100644 --- a/arch/i386/boot/compressed/relocs.c +++ b/arch/i386/boot/compressed/relocs.c @@ -19,6 +19,33 @@ static char *strtab[MAX_SHDRS]; static unsigned long reloc_count, reloc_idx; static unsigned long *relocs; +/* + * Following symbols have been audited. There values are constant and do + * not change if bzImage is loaded at a different physical address than + * the address for which it has been compiled. Don't warn user about + * absolute relocations present w.r.t these symbols. + */ +static const char* safe_abs_relocs[] = { + "__kernel_vsyscall", + "__kernel_rt_sigreturn", + "__kernel_sigreturn", + "SYSENTER_RETURN", +}; + +static int is_safe_abs_reloc(const char* sym_name) +{ + int i, array_size; + + array_size = sizeof(safe_abs_relocs)/sizeof(char*); + + for(i = 0; i < array_size; i++) { + if (!strcmp(sym_name, safe_abs_relocs[i])) + /* Match found */ + return 1; + } + return 0; +} + static void die(char *fmt, ...) { va_list ap; @@ -359,9 +386,8 @@ static void print_absolute_symbols(void) static void print_absolute_relocs(void) { - int i; - printf("Absolute relocations\n"); - printf("Offset Info Type Sym.Value Sym.Name\n"); + int i, printed = 0; + for(i = 0; i < ehdr.e_shnum; i++) { char *sym_strtab; Elf32_Sym *sh_symtab; @@ -387,6 +413,31 @@ static void print_absolute_relocs(void) if (sym->st_shndx != SHN_ABS) { continue; } + + /* Absolute symbols are not relocated if bzImage is + * loaded at a non-compiled address. Display a warning + * to user at compile time about the absolute + * relocations present. + * + * User need to audit the code to make sure + * some symbols which should have been section + * relative have not become absolute because of some + * linker optimization or wrong programming usage. + * + * Before warning check if this absolute symbol + * relocation is harmless. + */ + if (is_safe_abs_reloc(name)) + continue; + + if (!printed) { + printf("WARNING: Absolute relocations" + " present\n"); + printf("Offset Info Type Sym.Value " + "Sym.Name\n"); + printed = 1; + } + printf("%08x %08x %10s %08x %s\n", rel->r_offset, rel->r_info, @@ -395,7 +446,9 @@ static void print_absolute_relocs(void) name); } } - printf("\n"); + + if (printed) + printf("\n"); } static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym)) @@ -508,25 +561,31 @@ static void emit_relocs(int as_text) static void usage(void) { - die("i386_reloc [--abs | --text] vmlinux\n"); + die("relocs [--abs-syms |--abs-relocs | --text] vmlinux\n"); } int main(int argc, char **argv) { - int show_absolute; + int show_absolute_syms, show_absolute_relocs; int as_text; const char *fname; FILE *fp; int i; - show_absolute = 0; + show_absolute_syms = 0; + show_absolute_relocs = 0; as_text = 0; fname = NULL; for(i = 1; i < argc; i++) { char *arg = argv[i]; if (*arg == '-') { - if (strcmp(argv[1], "--abs") == 0) { - show_absolute = 1; + if (strcmp(argv[1], "--abs-syms") == 0) { + show_absolute_syms = 1; + continue; + } + + if (strcmp(argv[1], "--abs-relocs") == 0) { + show_absolute_relocs = 1; continue; } else if (strcmp(argv[1], "--text") == 0) { @@ -553,8 +612,11 @@ int main(int argc, char **argv) read_strtabs(fp); read_symtabs(fp); read_relocs(fp); - if (show_absolute) { + if (show_absolute_syms) { print_absolute_symbols(); + return 0; + } + if (show_absolute_relocs) { print_absolute_relocs(); return 0; } -- cgit v1.2.3-70-g09d2 From e69f202d0a1419219198566e1c22218a5c71a9a6 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Thu, 7 Dec 2006 02:14:04 +0100 Subject: [PATCH] i386: Implement CONFIG_PHYSICAL_ALIGN o Now CONFIG_PHYSICAL_START is being replaced with CONFIG_PHYSICAL_ALIGN. Hardcoding the kernel physical start value creates a problem in relocatable kernel context due to boot loader limitations. For ex, if somebody compiles a relocatable kernel to be run from address 4MB, but this kernel will run from location 1MB as grub loads the kernel at physical address 1MB. Kernel thinks that I am a relocatable kernel and I should run from the address I have been loaded at. So somebody wanting to run kernel from 4MB alignment location (for improved performance regions) can't do that. o Hence, Eric proposed that probably CONFIG_PHYSICAL_ALIGN will make more sense in relocatable kernel context. At run time kernel will move itself to a physical addr location which meets user specified alignment restrictions. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/i386/Kconfig | 33 ++++++++++++++++++--------------- arch/i386/boot/compressed/head.S | 26 ++++++++++++++------------ arch/i386/boot/compressed/misc.c | 7 ++++--- arch/i386/kernel/vmlinux.lds.S | 3 ++- include/asm-i386/boot.h | 6 +++++- 5 files changed, 43 insertions(+), 32 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index d588ca874bb..fd2fa7a7ec5 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -785,23 +785,26 @@ config RELOCATABLE must live at a different physical address than the primary kernel. -config PHYSICAL_START - hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP) - - default "0x1000000" if CRASH_DUMP +config PHYSICAL_ALIGN + hex "Alignment value to which kernel should be aligned" default "0x100000" + range 0x2000 0x400000 help - This gives the physical address where the kernel is loaded. Normally - for regular kernels this value is 0x100000 (1MB). But in the case - of kexec on panic the fail safe kernel needs to run at a different - address than the panic-ed kernel. This option is used to set the load - address for kernels used to capture crash dump on being kexec'ed - after panic. The default value for crash dump kernels is - 0x1000000 (16MB). This can also be set based on the "X" value as - specified in the "crashkernel=YM@XM" command line boot parameter - passed to the panic-ed kernel. Typically this parameter is set as - crashkernel=64M@16M. Please take a look at - Documentation/kdump/kdump.txt for more details about crash dumps. + This value puts the alignment restrictions on physical address + where kernel is loaded and run from. Kernel is compiled for an + address which meets above alignment restriction. + + If bootloader loads the kernel at a non-aligned address and + CONFIG_RELOCATABLE is set, kernel will move itself to nearest + address aligned to above value and run from there. + + If bootloader loads the kernel at a non-aligned address and + CONFIG_RELOCATABLE is not set, kernel will ignore the run time + load address and decompress itself to the address it has been + compiled for and run from there. The address for which kernel is + compiled already meets above alignment restrictions. Hence the + end result is that kernel runs from a physical address meeting + above alignment restrictions. Don't change this unless you know what you are doing. diff --git a/arch/i386/boot/compressed/head.S b/arch/i386/boot/compressed/head.S index e4dd7a6b9b0..f395a4bb38b 100644 --- a/arch/i386/boot/compressed/head.S +++ b/arch/i386/boot/compressed/head.S @@ -26,6 +26,7 @@ #include #include #include +#include .section ".text.head" .globl startup_32 @@ -52,17 +53,17 @@ startup_32: 1: popl %ebp subl $1b, %ebp -/* Compute the delta between where we were compiled to run at - * and where the code will actually run at. +/* %ebp contains the address we are loaded at by the boot loader and %ebx + * contains the address where we should move the kernel image temporarily + * for safe in-place decompression. */ - /* Start with the delta to where the kernel will run at. If we are - * a relocatable kernel this is the delta to our load address otherwise - * this is the delta to CONFIG_PHYSICAL start. - */ + #ifdef CONFIG_RELOCATABLE - movl %ebp, %ebx + movl %ebp, %ebx + addl $(CONFIG_PHYSICAL_ALIGN - 1), %ebx + andl $(~(CONFIG_PHYSICAL_ALIGN - 1)), %ebx #else - movl $(CONFIG_PHYSICAL_START - startup_32), %ebx + movl $LOAD_PHYSICAL_ADDR, %ebx #endif /* Replace the compressed data size with the uncompressed size */ @@ -94,9 +95,10 @@ startup_32: /* Compute the kernel start address. */ #ifdef CONFIG_RELOCATABLE - leal startup_32(%ebp), %ebp + addl $(CONFIG_PHYSICAL_ALIGN - 1), %ebp + andl $(~(CONFIG_PHYSICAL_ALIGN - 1)), %ebp #else - movl $CONFIG_PHYSICAL_START, %ebp + movl $LOAD_PHYSICAL_ADDR, %ebp #endif /* @@ -150,8 +152,8 @@ relocated: * and where it was actually loaded. */ movl %ebp, %ebx - subl $CONFIG_PHYSICAL_START, %ebx - + subl $LOAD_PHYSICAL_ADDR, %ebx + jz 2f /* Nothing to be done if loaded at compiled addr. */ /* * Process relocations. */ diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c index 4eac24e95a1..dc153893155 100644 --- a/arch/i386/boot/compressed/misc.c +++ b/arch/i386/boot/compressed/misc.c @@ -14,6 +14,7 @@ #include #include #include +#include /* WARNING!! * This code is compiled with -fPIC and it is relocated dynamically @@ -360,12 +361,12 @@ asmlinkage void decompress_kernel(void *rmode, unsigned long end, insize = input_len; inptr = 0; - if (((u32)output - CONFIG_PHYSICAL_START) & 0x3fffff) - error("Destination address not 4M aligned"); + if ((u32)output & (CONFIG_PHYSICAL_ALIGN -1)) + error("Destination address not CONFIG_PHYSICAL_ALIGN aligned"); if (end > ((-__PAGE_OFFSET-(512 <<20)-1) & 0x7fffffff)) error("Destination address too large"); #ifndef CONFIG_RELOCATABLE - if ((u32)output != CONFIG_PHYSICAL_START) + if ((u32)output != LOAD_PHYSICAL_ADDR) error("Wrong destination address"); #endif diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index f8d61ec4c8c..6860f20aa57 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -14,6 +14,7 @@ #include #include #include +#include OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386") OUTPUT_ARCH(i386) @@ -27,7 +28,7 @@ PHDRS { } SECTIONS { - . = LOAD_OFFSET + CONFIG_PHYSICAL_START; + . = LOAD_OFFSET + LOAD_PHYSICAL_ADDR; phys_startup_32 = startup_32 - LOAD_OFFSET; /* read-only */ .text : AT(ADDR(.text) - LOAD_OFFSET) { diff --git a/include/asm-i386/boot.h b/include/asm-i386/boot.h index 96b228e6e79..8ce79a6fa89 100644 --- a/include/asm-i386/boot.h +++ b/include/asm-i386/boot.h @@ -12,4 +12,8 @@ #define EXTENDED_VGA 0xfffe /* 80x50 mode */ #define ASK_VGA 0xfffd /* ask for it at bootup */ -#endif +/* Physical address where kenrel should be loaded. */ +#define LOAD_PHYSICAL_ADDR ((0x100000 + CONFIG_PHYSICAL_ALIGN - 1) \ + & ~(CONFIG_PHYSICAL_ALIGN - 1)) + +#endif /* _LINUX_BOOT_H */ -- cgit v1.2.3-70-g09d2 From be274eeaf20b4c7155242645d5e2c48b023e609b Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Thu, 7 Dec 2006 02:14:04 +0100 Subject: [PATCH] i386: extend bzImage protocol for relocatable protected mode kernel Extend bzImage protocol to enable bootloaders to load a completely relocatable bzImage. Now protected mode component of kernel is also relocatable and a boot-loader can load the protected mode component at a differnt physical address than 1MB. (If kernel was built with CONFIG_RELOCATABLE) Kexec can make use of it to load this kernel at a different physical address to capture kernel crash dumps. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton --- Documentation/i386/boot.txt | 4 ++++ arch/i386/boot/setup.S | 13 ++++++++++++- 2 files changed, 16 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/Documentation/i386/boot.txt b/Documentation/i386/boot.txt index c51314b1a46..cb28254f155 100644 --- a/Documentation/i386/boot.txt +++ b/Documentation/i386/boot.txt @@ -35,6 +35,8 @@ Protocol 2.03: (Kernel 2.4.18-pre1) Explicitly makes the highest possible initrd address available to the bootloader. Protocol 2.04: (Kernel 2.6.14) Extend the syssize field to four bytes. +Protocol 2.05: (Kernel 2.6.20) Make protected mode kernel relocatable. + Introduce relocatable_kernel and kernel_alignment fields. **** MEMORY LAYOUT @@ -129,6 +131,8 @@ Offset Proto Name Meaning 0226/2 N/A pad1 Unused 0228/4 2.02+ cmd_line_ptr 32-bit pointer to the kernel command line 022C/4 2.03+ initrd_addr_max Highest legal initrd address +0230/4 2.04+ kernel_alignment Physical addr alignment required for kernel +0234/1 2.04+ relocatable_kernel Whether kernel is relocatable or not (1) For backwards compatibility, if the setup_sects field contains 0, the real value is 4. diff --git a/arch/i386/boot/setup.S b/arch/i386/boot/setup.S index 9aa8b051818..06edf1c6624 100644 --- a/arch/i386/boot/setup.S +++ b/arch/i386/boot/setup.S @@ -81,7 +81,7 @@ start: # This is the setup header, and it must start at %cs:2 (old 0x9020:2) .ascii "HdrS" # header signature - .word 0x0204 # header version number (>= 0x0105) + .word 0x0205 # header version number (>= 0x0105) # or else old loadlin-1.5 will fail) realmode_swtch: .word 0, 0 # default_switch, SETUPSEG start_sys_seg: .word SYSSEG @@ -160,6 +160,17 @@ ramdisk_max: .long (-__PAGE_OFFSET-(512 << 20)-1) & 0x7fffffff # The highest safe address for # the contents of an initrd +kernel_alignment: .long CONFIG_PHYSICAL_ALIGN #physical addr alignment + #required for protected mode + #kernel +#ifdef CONFIG_RELOCATABLE +relocatable_kernel: .byte 1 +#else +relocatable_kernel: .byte 0 +#endif +pad2: .byte 0 +pad3: .word 0 + trampoline: call start_of_setup .align 16 # The offset at this point is 0x240 -- cgit v1.2.3-70-g09d2 From 4c7aa6c3b25ef96bc1b723238041195e3d8bf047 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Thu, 7 Dec 2006 02:14:04 +0100 Subject: [PATCH] i386: Mark CONFIG_RELOCATABLE EXPERIMENTAL Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index fd2fa7a7ec5..1f0f7b60995 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -774,7 +774,8 @@ config CRASH_DUMP For more details see Documentation/kdump/kdump.txt config RELOCATABLE - bool "Build a relocatable kernel" + bool "Build a relocatable kernel(EXPERIMENTAL)" + depends on EXPERIMENTAL help This build a kernel image that retains relocation information so it can be loaded someplace besides the default 1MB. -- cgit v1.2.3-70-g09d2 From 74b47a7844501445d41d704fe7c626f4b1819508 Mon Sep 17 00:00:00 2001 From: Joe Korty Date: Thu, 7 Dec 2006 02:14:04 +0100 Subject: [PATCH] i386: Fix entry.S code with !CONFIG_VM86 The entry.S code at work_notifysig is surely wrong. It drops into unrelated code if the branch to work_notifysig_v86 is taken, and CONFIG_VM86=n. [PATCH] Make vm86 support optional tree 9b5daef5280800a0006343a17f63072658d91a1d pushed to git Jan 8, 2006, and first appears in 2.6.16 The 'fix' here is to also compile out the vm86 test & branch when CONFIG_VM86=n. Signed-off-by: Joe Korty Signed-off-by: Andi Kleen --- arch/i386/kernel/entry.S | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index d7423efaeea..0220bc8cbb4 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -457,6 +457,7 @@ work_resched: work_notifysig: # deal with pending signals and # notify-resume requests +#ifdef CONFIG_VM86 testl $VM_MASK, PT_EFLAGS(%esp) movl %esp, %eax jne work_notifysig_v86 # returning to kernel-space or @@ -467,17 +468,18 @@ work_notifysig: # deal with pending signals and ALIGN work_notifysig_v86: -#ifdef CONFIG_VM86 pushl %ecx # save ti_flags for do_notify_resume CFI_ADJUST_CFA_OFFSET 4 call save_v86_state # %eax contains pt_regs pointer popl %ecx CFI_ADJUST_CFA_OFFSET -4 movl %eax, %esp +#else + movl %esp, %eax +#endif xorl %edx, %edx call do_notify_resume jmp resume_userspace_sig -#endif # perform syscall exit tracing ALIGN -- cgit v1.2.3-70-g09d2 From 770d132f03ac15b12919f1bac481f4beda13e094 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:05 +0100 Subject: [PATCH] i386: Retrieve CLFLUSH size from CPUID Also report it in /proc/cpuinfo similar to x86-64. Needed for followon patch Signed-off-by: Andi Kleen --- arch/i386/kernel/cpu/common.c | 3 +++ arch/i386/kernel/cpu/proc.c | 3 ++- include/asm-i386/processor.h | 1 + 3 files changed, 6 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index 6958ae5e2fa..cda41aef79a 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -309,6 +309,8 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 * c) #else c->apicid = (ebx >> 24) & 0xFF; #endif + if (c->x86_capability[0] & (1<<19)) + c->x86_clflush_size = ((ebx >> 8) & 0xff) * 8; } else { /* Have CPUID level 0 only - unheard of */ c->x86 = 4; @@ -373,6 +375,7 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c) c->x86_vendor_id[0] = '\0'; /* Unset */ c->x86_model_id[0] = '\0'; /* Unset */ c->x86_max_cores = 1; + c->x86_clflush_size = 32; memset(&c->x86_capability, 0, sizeof c->x86_capability); if (!have_cpuid_p()) { diff --git a/arch/i386/kernel/cpu/proc.c b/arch/i386/kernel/cpu/proc.c index 76aac088a32..6624d8583c4 100644 --- a/arch/i386/kernel/cpu/proc.c +++ b/arch/i386/kernel/cpu/proc.c @@ -152,9 +152,10 @@ static int show_cpuinfo(struct seq_file *m, void *v) seq_printf(m, " [%d]", i); } - seq_printf(m, "\nbogomips\t: %lu.%02lu\n\n", + seq_printf(m, "\nbogomips\t: %lu.%02lu\n", c->loops_per_jiffy/(500000/HZ), (c->loops_per_jiffy/(5000/HZ)) % 100); + seq_printf(m, "clflush size\t: %u\n\n", c->x86_clflush_size); return 0; } diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h index f73cf836e64..98fa73b7176 100644 --- a/include/asm-i386/processor.h +++ b/include/asm-i386/processor.h @@ -72,6 +72,7 @@ struct cpuinfo_x86 { #endif unsigned char x86_max_cores; /* cpuid returned max cores value */ unsigned char apicid; + unsigned short x86_clflush_size; #ifdef CONFIG_SMP unsigned char booted_cores; /* number of cores as seen by OS */ __u8 phys_proc_id; /* Physical processor id. */ -- cgit v1.2.3-70-g09d2 From 3760dd6efa75c98e223643da2eb7040406433053 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:05 +0100 Subject: [PATCH] i386: Use CLFLUSH instead of WBINVD in change_page_attr CLFLUSH is a lot faster than WBINVD so try to use that. Signed-off-by: Andi Kleen --- arch/i386/mm/pageattr.c | 24 +++++++++++++++++------- include/asm-i386/cpufeature.h | 1 + 2 files changed, 18 insertions(+), 7 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/mm/pageattr.c b/arch/i386/mm/pageattr.c index 8564b6ae17e..ad91528bdc1 100644 --- a/arch/i386/mm/pageattr.c +++ b/arch/i386/mm/pageattr.c @@ -67,11 +67,17 @@ static struct page *split_large_page(unsigned long address, pgprot_t prot, return base; } -static void flush_kernel_map(void *dummy) +static void flush_kernel_map(void *arg) { - /* Could use CLFLUSH here if the CPU supports it (Hammer,P4) */ - if (boot_cpu_data.x86_model >= 4) + unsigned long adr = (unsigned long)arg; + + if (adr && cpu_has_clflush) { + int i; + for (i = 0; i < PAGE_SIZE; i += boot_cpu_data.x86_clflush_size) + asm volatile("clflush (%0)" :: "r" (adr + i)); + } else if (boot_cpu_data.x86_model >= 4) wbinvd(); + /* Flush all to work around Errata in early athlons regarding * large page flushing. */ @@ -173,9 +179,9 @@ __change_page_attr(struct page *page, pgprot_t prot) return 0; } -static inline void flush_map(void) +static inline void flush_map(void *adr) { - on_each_cpu(flush_kernel_map, NULL, 1, 1); + on_each_cpu(flush_kernel_map, adr, 1, 1); } /* @@ -217,9 +223,13 @@ void global_flush_tlb(void) spin_lock_irq(&cpa_lock); list_replace_init(&df_list, &l); spin_unlock_irq(&cpa_lock); - flush_map(); - list_for_each_entry_safe(pg, next, &l, lru) + if (!cpu_has_clflush) + flush_map(0); + list_for_each_entry_safe(pg, next, &l, lru) { + if (cpu_has_clflush) + flush_map(page_address(pg)); __free_page(pg); + } } #ifdef CONFIG_DEBUG_PAGEALLOC diff --git a/include/asm-i386/cpufeature.h b/include/asm-i386/cpufeature.h index 231672558c1..4c83e059228 100644 --- a/include/asm-i386/cpufeature.h +++ b/include/asm-i386/cpufeature.h @@ -137,6 +137,7 @@ #define cpu_has_pmm_enabled boot_cpu_has(X86_FEATURE_PMM_EN) #define cpu_has_ds boot_cpu_has(X86_FEATURE_DS) #define cpu_has_pebs boot_cpu_has(X86_FEATURE_PEBS) +#define cpu_has_clflush boot_cpu_has(X86_FEATURE_CLFLSH) #endif /* __ASM_I386_CPUFEATURE_H */ -- cgit v1.2.3-70-g09d2 From 11a4180c0b03e2ee0c948fd8430ee092dc1625b3 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] i386: Use probe_kernel_address instead of __get_user in fault paths Makes the intention of the code cleaner to read and avoids a potential deadlock on mmap_sem. Also change the types of the arguments to not include __user because they're really not user addresses. Signed-off-by: Andi Kleen --- arch/i386/kernel/traps.c | 24 +++++++++++++----------- arch/i386/mm/fault.c | 12 ++++++------ 2 files changed, 19 insertions(+), 17 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 237f4884a1e..7b2f9f02208 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -380,7 +380,7 @@ void show_registers(struct pt_regs *regs) * time of the fault.. */ if (in_kernel) { - u8 __user *eip; + u8 *eip; int code_bytes = 64; unsigned char c; @@ -389,18 +389,20 @@ void show_registers(struct pt_regs *regs) printk(KERN_EMERG "Code: "); - eip = (u8 __user *)regs->eip - 43; - if (eip < (u8 __user *)PAGE_OFFSET || __get_user(c, eip)) { + eip = (u8 *)regs->eip - 43; + if (eip < (u8 *)PAGE_OFFSET || + probe_kernel_address(eip, c)) { /* try starting at EIP */ - eip = (u8 __user *)regs->eip; + eip = (u8 *)regs->eip; code_bytes = 32; } for (i = 0; i < code_bytes; i++, eip++) { - if (eip < (u8 __user *)PAGE_OFFSET || __get_user(c, eip)) { + if (eip < (u8 *)PAGE_OFFSET || + probe_kernel_address(eip, c)) { printk(" Bad EIP value."); break; } - if (eip == (u8 __user *)regs->eip) + if (eip == (u8 *)regs->eip) printk("<%02x> ", c); else printk("%02x ", c); @@ -416,7 +418,7 @@ static void handle_BUG(struct pt_regs *regs) if (eip < PAGE_OFFSET) return; - if (probe_kernel_address((unsigned short __user *)eip, ud2)) + if (probe_kernel_address((unsigned short *)eip, ud2)) return; if (ud2 != 0x0b0f) return; @@ -429,11 +431,11 @@ static void handle_BUG(struct pt_regs *regs) char *file; char c; - if (probe_kernel_address((unsigned short __user *)(eip + 2), - line)) + if (probe_kernel_address((unsigned short *)(eip + 2), line)) break; - if (__get_user(file, (char * __user *)(eip + 4)) || - (unsigned long)file < PAGE_OFFSET || __get_user(c, file)) + if (probe_kernel_address((char **)(eip + 4), file) || + (unsigned long)file < PAGE_OFFSET || + probe_kernel_address(file, c)) file = ""; printk(KERN_EMERG "kernel BUG at %s:%d!\n", file, line); diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c index 2581575786c..aaaa4d225f7 100644 --- a/arch/i386/mm/fault.c +++ b/arch/i386/mm/fault.c @@ -22,9 +22,9 @@ #include #include #include +#include #include -#include #include #include #include @@ -167,7 +167,7 @@ static inline unsigned long get_segment_eip(struct pt_regs *regs, static int __is_prefetch(struct pt_regs *regs, unsigned long addr) { unsigned long limit; - unsigned long instr = get_segment_eip (regs, &limit); + unsigned char *instr = (unsigned char *)get_segment_eip (regs, &limit); int scan_more = 1; int prefetch = 0; int i; @@ -177,9 +177,9 @@ static int __is_prefetch(struct pt_regs *regs, unsigned long addr) unsigned char instr_hi; unsigned char instr_lo; - if (instr > limit) + if (instr > (unsigned char *)limit) break; - if (__get_user(opcode, (unsigned char __user *) instr)) + if (probe_kernel_address(instr, opcode)) break; instr_hi = opcode & 0xf0; @@ -204,9 +204,9 @@ static int __is_prefetch(struct pt_regs *regs, unsigned long addr) case 0x00: /* Prefetch instruction is 0x0F0D or 0x0F18 */ scan_more = 0; - if (instr > limit) + if (instr > (unsigned char *)limit) break; - if (__get_user(opcode, (unsigned char __user *) instr)) + if (probe_kernel_address(instr, opcode)) break; prefetch = (instr_lo == 0xF) && (opcode == 0x0D || opcode == 0x18); -- cgit v1.2.3-70-g09d2 From 269c2d81ed66af7c09a1619ffe165f03e7470a5b Mon Sep 17 00:00:00 2001 From: "bibo,mao" Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] i386: i386 create e820.c to handle standard io/mem resources This patch creates new file named e820.c to hanle standard io/mem resources, moving request_standard_resources function from setup.c to e820.c. Also this patch modifies Makfile to compile file e820.c. Signed-off-by: bibo,mao Signed-off-by: Andi Kleen Makefile | 2 arch/i386/kernel/Makefile | 2 arch/i386/kernel/e820.c | 289 ++++++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 276 ------------------------------------------- 3 files changed, 293 insertions(+), 274 deletions(-) --- arch/i386/kernel/Makefile | 2 +- arch/i386/kernel/e820.c | 289 ++++++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 276 +------------------------------------------ 3 files changed, 293 insertions(+), 274 deletions(-) create mode 100644 arch/i386/kernel/e820.c (limited to 'arch/i386') diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile index 1a884b6e6e5..f614854bd71 100644 --- a/arch/i386/kernel/Makefile +++ b/arch/i386/kernel/Makefile @@ -6,7 +6,7 @@ extra-y := head.o init_task.o vmlinux.lds obj-y := process.o signal.o entry.o traps.o irq.o \ ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_i386.o \ - pci-dma.o i386_ksyms.o i387.o bootflag.o \ + pci-dma.o i386_ksyms.o i387.o bootflag.o e820.o\ quirks.o i8237.o topology.o alternative.o i8253.o tsc.o obj-$(CONFIG_STACKTRACE) += stacktrace.o diff --git a/arch/i386/kernel/e820.c b/arch/i386/kernel/e820.c new file mode 100644 index 00000000000..cce70604948 --- /dev/null +++ b/arch/i386/kernel/e820.c @@ -0,0 +1,289 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#ifdef CONFIG_EFI +int efi_enabled = 0; +EXPORT_SYMBOL(efi_enabled); +#endif + +struct e820map e820; +struct resource data_resource = { + .name = "Kernel data", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_MEM +}; + +struct resource code_resource = { + .name = "Kernel code", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_MEM +}; + +static struct resource system_rom_resource = { + .name = "System ROM", + .start = 0xf0000, + .end = 0xfffff, + .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM +}; + +static struct resource extension_rom_resource = { + .name = "Extension ROM", + .start = 0xe0000, + .end = 0xeffff, + .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM +}; + +static struct resource adapter_rom_resources[] = { { + .name = "Adapter ROM", + .start = 0xc8000, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM +}, { + .name = "Adapter ROM", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM +}, { + .name = "Adapter ROM", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM +}, { + .name = "Adapter ROM", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM +}, { + .name = "Adapter ROM", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM +}, { + .name = "Adapter ROM", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM +} }; + +static struct resource video_rom_resource = { + .name = "Video ROM", + .start = 0xc0000, + .end = 0xc7fff, + .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM +}; + +static struct resource video_ram_resource = { + .name = "Video RAM area", + .start = 0xa0000, + .end = 0xbffff, + .flags = IORESOURCE_BUSY | IORESOURCE_MEM +}; + +static struct resource standard_io_resources[] = { { + .name = "dma1", + .start = 0x0000, + .end = 0x001f, + .flags = IORESOURCE_BUSY | IORESOURCE_IO +}, { + .name = "pic1", + .start = 0x0020, + .end = 0x0021, + .flags = IORESOURCE_BUSY | IORESOURCE_IO +}, { + .name = "timer0", + .start = 0x0040, + .end = 0x0043, + .flags = IORESOURCE_BUSY | IORESOURCE_IO +}, { + .name = "timer1", + .start = 0x0050, + .end = 0x0053, + .flags = IORESOURCE_BUSY | IORESOURCE_IO +}, { + .name = "keyboard", + .start = 0x0060, + .end = 0x006f, + .flags = IORESOURCE_BUSY | IORESOURCE_IO +}, { + .name = "dma page reg", + .start = 0x0080, + .end = 0x008f, + .flags = IORESOURCE_BUSY | IORESOURCE_IO +}, { + .name = "pic2", + .start = 0x00a0, + .end = 0x00a1, + .flags = IORESOURCE_BUSY | IORESOURCE_IO +}, { + .name = "dma2", + .start = 0x00c0, + .end = 0x00df, + .flags = IORESOURCE_BUSY | IORESOURCE_IO +}, { + .name = "fpu", + .start = 0x00f0, + .end = 0x00ff, + .flags = IORESOURCE_BUSY | IORESOURCE_IO +} }; + +#define romsignature(x) (*(unsigned short *)(x) == 0xaa55) + +static int __init romchecksum(unsigned char *rom, unsigned long length) +{ + unsigned char *p, sum = 0; + + for (p = rom; p < rom + length; p++) + sum += *p; + return sum == 0; +} + +static void __init probe_roms(void) +{ + unsigned long start, length, upper; + unsigned char *rom; + int i; + + /* video rom */ + upper = adapter_rom_resources[0].start; + for (start = video_rom_resource.start; start < upper; start += 2048) { + rom = isa_bus_to_virt(start); + if (!romsignature(rom)) + continue; + + video_rom_resource.start = start; + + /* 0 < length <= 0x7f * 512, historically */ + length = rom[2] * 512; + + /* if checksum okay, trust length byte */ + if (length && romchecksum(rom, length)) + video_rom_resource.end = start + length - 1; + + request_resource(&iomem_resource, &video_rom_resource); + break; + } + + start = (video_rom_resource.end + 1 + 2047) & ~2047UL; + if (start < upper) + start = upper; + + /* system rom */ + request_resource(&iomem_resource, &system_rom_resource); + upper = system_rom_resource.start; + + /* check for extension rom (ignore length byte!) */ + rom = isa_bus_to_virt(extension_rom_resource.start); + if (romsignature(rom)) { + length = extension_rom_resource.end - extension_rom_resource.start + 1; + if (romchecksum(rom, length)) { + request_resource(&iomem_resource, &extension_rom_resource); + upper = extension_rom_resource.start; + } + } + + /* check for adapter roms on 2k boundaries */ + for (i = 0; i < ARRAY_SIZE(adapter_rom_resources) && start < upper; start += 2048) { + rom = isa_bus_to_virt(start); + if (!romsignature(rom)) + continue; + + /* 0 < length <= 0x7f * 512, historically */ + length = rom[2] * 512; + + /* but accept any length that fits if checksum okay */ + if (!length || start + length > upper || !romchecksum(rom, length)) + continue; + + adapter_rom_resources[i].start = start; + adapter_rom_resources[i].end = start + length - 1; + request_resource(&iomem_resource, &adapter_rom_resources[i]); + + start = adapter_rom_resources[i++].end & ~2047UL; + } +} + +/* + * Request address space for all standard RAM and ROM resources + * and also for regions reported as reserved by the e820. + */ +static void __init +legacy_init_iomem_resources(struct resource *code_resource, struct resource *data_resource) +{ + int i; + + probe_roms(); + for (i = 0; i < e820.nr_map; i++) { + struct resource *res; +#ifndef CONFIG_RESOURCES_64BIT + if (e820.map[i].addr + e820.map[i].size > 0x100000000ULL) + continue; +#endif + res = kzalloc(sizeof(struct resource), GFP_ATOMIC); + switch (e820.map[i].type) { + case E820_RAM: res->name = "System RAM"; break; + case E820_ACPI: res->name = "ACPI Tables"; break; + case E820_NVS: res->name = "ACPI Non-volatile Storage"; break; + default: res->name = "reserved"; + } + res->start = e820.map[i].addr; + res->end = res->start + e820.map[i].size - 1; + res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; + if (request_resource(&iomem_resource, res)) { + kfree(res); + continue; + } + if (e820.map[i].type == E820_RAM) { + /* + * We don't know which RAM region contains kernel data, + * so we try it repeatedly and let the resource manager + * test it. + */ + request_resource(res, code_resource); + request_resource(res, data_resource); +#ifdef CONFIG_KEXEC + request_resource(res, &crashk_res); +#endif + } + } +} + +/* + * Request address space for all standard resources + * + * This is called just before pcibios_init(), which is also a + * subsys_initcall, but is linked in later (in arch/i386/pci/common.c). + */ +static int __init request_standard_resources(void) +{ + int i; + + printk("Setting up standard PCI resources\n"); + if (efi_enabled) + efi_initialize_iomem_resources(&code_resource, &data_resource); + else + legacy_init_iomem_resources(&code_resource, &data_resource); + + /* EFI systems may still have VGA */ + request_resource(&iomem_resource, &video_ram_resource); + + /* request I/O space for devices used on all i[345]86 PCs */ + for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++) + request_resource(&ioport_resource, &standard_io_resources[i]); + return 0; +} + +subsys_initcall(request_standard_resources); diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 61539afbdf2..acd2d9392ab 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -76,11 +76,9 @@ int disable_pse __devinitdata = 0; /* * Machine setup.. */ - -#ifdef CONFIG_EFI -int efi_enabled = 0; -EXPORT_SYMBOL(efi_enabled); -#endif +extern struct e820map e820; +extern struct resource code_resource; +extern struct resource data_resource; /* cpu data as detected by the assembly code in head.S */ struct cpuinfo_x86 new_cpu_data __initdata = { 0, 0, 0, 0, -1, 1, 0, 0, -1 }; @@ -134,7 +132,6 @@ struct ist_info ist_info; defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE) EXPORT_SYMBOL(ist_info); #endif -struct e820map e820; extern void early_cpu_init(void); extern int root_mountflags; @@ -149,203 +146,6 @@ static char command_line[COMMAND_LINE_SIZE]; unsigned char __initdata boot_params[PARAM_SIZE]; -static struct resource data_resource = { - .name = "Kernel data", - .start = 0, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_MEM -}; - -static struct resource code_resource = { - .name = "Kernel code", - .start = 0, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_MEM -}; - -static struct resource system_rom_resource = { - .name = "System ROM", - .start = 0xf0000, - .end = 0xfffff, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -}; - -static struct resource extension_rom_resource = { - .name = "Extension ROM", - .start = 0xe0000, - .end = 0xeffff, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -}; - -static struct resource adapter_rom_resources[] = { { - .name = "Adapter ROM", - .start = 0xc8000, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -}, { - .name = "Adapter ROM", - .start = 0, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -}, { - .name = "Adapter ROM", - .start = 0, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -}, { - .name = "Adapter ROM", - .start = 0, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -}, { - .name = "Adapter ROM", - .start = 0, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -}, { - .name = "Adapter ROM", - .start = 0, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -} }; - -static struct resource video_rom_resource = { - .name = "Video ROM", - .start = 0xc0000, - .end = 0xc7fff, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -}; - -static struct resource video_ram_resource = { - .name = "Video RAM area", - .start = 0xa0000, - .end = 0xbffff, - .flags = IORESOURCE_BUSY | IORESOURCE_MEM -}; - -static struct resource standard_io_resources[] = { { - .name = "dma1", - .start = 0x0000, - .end = 0x001f, - .flags = IORESOURCE_BUSY | IORESOURCE_IO -}, { - .name = "pic1", - .start = 0x0020, - .end = 0x0021, - .flags = IORESOURCE_BUSY | IORESOURCE_IO -}, { - .name = "timer0", - .start = 0x0040, - .end = 0x0043, - .flags = IORESOURCE_BUSY | IORESOURCE_IO -}, { - .name = "timer1", - .start = 0x0050, - .end = 0x0053, - .flags = IORESOURCE_BUSY | IORESOURCE_IO -}, { - .name = "keyboard", - .start = 0x0060, - .end = 0x006f, - .flags = IORESOURCE_BUSY | IORESOURCE_IO -}, { - .name = "dma page reg", - .start = 0x0080, - .end = 0x008f, - .flags = IORESOURCE_BUSY | IORESOURCE_IO -}, { - .name = "pic2", - .start = 0x00a0, - .end = 0x00a1, - .flags = IORESOURCE_BUSY | IORESOURCE_IO -}, { - .name = "dma2", - .start = 0x00c0, - .end = 0x00df, - .flags = IORESOURCE_BUSY | IORESOURCE_IO -}, { - .name = "fpu", - .start = 0x00f0, - .end = 0x00ff, - .flags = IORESOURCE_BUSY | IORESOURCE_IO -} }; - -#define romsignature(x) (*(unsigned short *)(x) == 0xaa55) - -static int __init romchecksum(unsigned char *rom, unsigned long length) -{ - unsigned char *p, sum = 0; - - for (p = rom; p < rom + length; p++) - sum += *p; - return sum == 0; -} - -static void __init probe_roms(void) -{ - unsigned long start, length, upper; - unsigned char *rom; - int i; - - /* video rom */ - upper = adapter_rom_resources[0].start; - for (start = video_rom_resource.start; start < upper; start += 2048) { - rom = isa_bus_to_virt(start); - if (!romsignature(rom)) - continue; - - video_rom_resource.start = start; - - /* 0 < length <= 0x7f * 512, historically */ - length = rom[2] * 512; - - /* if checksum okay, trust length byte */ - if (length && romchecksum(rom, length)) - video_rom_resource.end = start + length - 1; - - request_resource(&iomem_resource, &video_rom_resource); - break; - } - - start = (video_rom_resource.end + 1 + 2047) & ~2047UL; - if (start < upper) - start = upper; - - /* system rom */ - request_resource(&iomem_resource, &system_rom_resource); - upper = system_rom_resource.start; - - /* check for extension rom (ignore length byte!) */ - rom = isa_bus_to_virt(extension_rom_resource.start); - if (romsignature(rom)) { - length = extension_rom_resource.end - extension_rom_resource.start + 1; - if (romchecksum(rom, length)) { - request_resource(&iomem_resource, &extension_rom_resource); - upper = extension_rom_resource.start; - } - } - - /* check for adapter roms on 2k boundaries */ - for (i = 0; i < ARRAY_SIZE(adapter_rom_resources) && start < upper; start += 2048) { - rom = isa_bus_to_virt(start); - if (!romsignature(rom)) - continue; - - /* 0 < length <= 0x7f * 512, historically */ - length = rom[2] * 512; - - /* but accept any length that fits if checksum okay */ - if (!length || start + length > upper || !romchecksum(rom, length)) - continue; - - adapter_rom_resources[i].start = start; - adapter_rom_resources[i].end = start + length - 1; - request_resource(&iomem_resource, &adapter_rom_resources[i]); - - start = adapter_rom_resources[i++].end & ~2047UL; - } -} - static void __init limit_regions(unsigned long long size) { unsigned long long current_addr = 0; @@ -1200,77 +1000,7 @@ void __init remapped_pgdat_init(void) } } -/* - * Request address space for all standard RAM and ROM resources - * and also for regions reported as reserved by the e820. - */ -static void __init -legacy_init_iomem_resources(struct resource *code_resource, struct resource *data_resource) -{ - int i; - - probe_roms(); - for (i = 0; i < e820.nr_map; i++) { - struct resource *res; -#ifndef CONFIG_RESOURCES_64BIT - if (e820.map[i].addr + e820.map[i].size > 0x100000000ULL) - continue; -#endif - res = kzalloc(sizeof(struct resource), GFP_ATOMIC); - switch (e820.map[i].type) { - case E820_RAM: res->name = "System RAM"; break; - case E820_ACPI: res->name = "ACPI Tables"; break; - case E820_NVS: res->name = "ACPI Non-volatile Storage"; break; - default: res->name = "reserved"; - } - res->start = e820.map[i].addr; - res->end = res->start + e820.map[i].size - 1; - res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; - if (request_resource(&iomem_resource, res)) { - kfree(res); - continue; - } - if (e820.map[i].type == E820_RAM) { - /* - * We don't know which RAM region contains kernel data, - * so we try it repeatedly and let the resource manager - * test it. - */ - request_resource(res, code_resource); - request_resource(res, data_resource); -#ifdef CONFIG_KEXEC - request_resource(res, &crashk_res); -#endif - } - } -} - -/* - * Request address space for all standard resources - * - * This is called just before pcibios_init(), which is also a - * subsys_initcall, but is linked in later (in arch/i386/pci/common.c). - */ -static int __init request_standard_resources(void) -{ - int i; - - printk("Setting up standard PCI resources\n"); - if (efi_enabled) - efi_initialize_iomem_resources(&code_resource, &data_resource); - else - legacy_init_iomem_resources(&code_resource, &data_resource); - - /* EFI systems may still have VGA */ - request_resource(&iomem_resource, &video_ram_resource); - - /* request I/O space for devices used on all i[345]86 PCs */ - for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++) - request_resource(&ioport_resource, &standard_io_resources[i]); - return 0; -} -subsys_initcall(request_standard_resources); static void __init register_memory(void) { -- cgit v1.2.3-70-g09d2 From 8e3342f736dd1c19ce7c28625dedd7d8730fc7ad Mon Sep 17 00:00:00 2001 From: "bibo,mao" Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] i386: create e820.c for e820 map sanitize and copy function This patch moves bios e820 map sanitize and copy function from setup.c to e820.c Signed-off-by: bibo,mao Signed-off-by: Andi Kleen arch/i386/kernel/e820.c | 252 +++++++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 240 -------------------------------------------- 2 files changed, 252 insertions(+), 240 deletions(-) --- arch/i386/kernel/e820.c | 252 +++++++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 240 -------------------------------------------- 2 files changed, 252 insertions(+), 240 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/e820.c b/arch/i386/kernel/e820.c index cce70604948..0db95760b07 100644 --- a/arch/i386/kernel/e820.c +++ b/arch/i386/kernel/e820.c @@ -19,6 +19,14 @@ EXPORT_SYMBOL(efi_enabled); #endif struct e820map e820; +struct change_member { + struct e820entry *pbios; /* pointer to original bios entry */ + unsigned long long addr; /* address for this change point */ +}; +static struct change_member change_point_list[2*E820MAX] __initdata; +static struct change_member *change_point[2*E820MAX] __initdata; +static struct e820entry *overlap_list[E820MAX] __initdata; +static struct e820entry new_bios[E820MAX] __initdata; struct resource data_resource = { .name = "Kernel data", .start = 0, @@ -287,3 +295,247 @@ static int __init request_standard_resources(void) } subsys_initcall(request_standard_resources); + +void __init add_memory_region(unsigned long long start, + unsigned long long size, int type) +{ + int x; + + if (!efi_enabled) { + x = e820.nr_map; + + if (x == E820MAX) { + printk(KERN_ERR "Ooops! Too many entries in the memory map!\n"); + return; + } + + e820.map[x].addr = start; + e820.map[x].size = size; + e820.map[x].type = type; + e820.nr_map++; + } +} /* add_memory_region */ + +/* + * Sanitize the BIOS e820 map. + * + * Some e820 responses include overlapping entries. The following + * replaces the original e820 map with a new one, removing overlaps. + * + */ +int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) +{ + struct change_member *change_tmp; + unsigned long current_type, last_type; + unsigned long long last_addr; + int chgidx, still_changing; + int overlap_entries; + int new_bios_entry; + int old_nr, new_nr, chg_nr; + int i; + + /* + Visually we're performing the following (1,2,3,4 = memory types)... + + Sample memory map (w/overlaps): + ____22__________________ + ______________________4_ + ____1111________________ + _44_____________________ + 11111111________________ + ____________________33__ + ___________44___________ + __________33333_________ + ______________22________ + ___________________2222_ + _________111111111______ + _____________________11_ + _________________4______ + + Sanitized equivalent (no overlap): + 1_______________________ + _44_____________________ + ___1____________________ + ____22__________________ + ______11________________ + _________1______________ + __________3_____________ + ___________44___________ + _____________33_________ + _______________2________ + ________________1_______ + _________________4______ + ___________________2____ + ____________________33__ + ______________________4_ + */ + printk("sanitize start\n"); + /* if there's only one memory region, don't bother */ + if (*pnr_map < 2) { + printk("sanitize bail 0\n"); + return -1; + } + + old_nr = *pnr_map; + + /* bail out if we find any unreasonable addresses in bios map */ + for (i=0; iaddr = biosmap[i].addr; + change_point[chgidx++]->pbios = &biosmap[i]; + change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size; + change_point[chgidx++]->pbios = &biosmap[i]; + } + } + chg_nr = chgidx; /* true number of change-points */ + + /* sort change-point list by memory addresses (low -> high) */ + still_changing = 1; + while (still_changing) { + still_changing = 0; + for (i=1; i < chg_nr; i++) { + /* if > , swap */ + /* or, if current= & last=, swap */ + if ((change_point[i]->addr < change_point[i-1]->addr) || + ((change_point[i]->addr == change_point[i-1]->addr) && + (change_point[i]->addr == change_point[i]->pbios->addr) && + (change_point[i-1]->addr != change_point[i-1]->pbios->addr)) + ) + { + change_tmp = change_point[i]; + change_point[i] = change_point[i-1]; + change_point[i-1] = change_tmp; + still_changing=1; + } + } + } + + /* create a new bios memory map, removing overlaps */ + overlap_entries=0; /* number of entries in the overlap table */ + new_bios_entry=0; /* index for creating new bios map entries */ + last_type = 0; /* start with undefined memory type */ + last_addr = 0; /* start with 0 as last starting address */ + /* loop through change-points, determining affect on the new bios map */ + for (chgidx=0; chgidx < chg_nr; chgidx++) + { + /* keep track of all overlapping bios entries */ + if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr) + { + /* add map entry to overlap list (> 1 entry implies an overlap) */ + overlap_list[overlap_entries++]=change_point[chgidx]->pbios; + } + else + { + /* remove entry from list (order independent, so swap with last) */ + for (i=0; ipbios) + overlap_list[i] = overlap_list[overlap_entries-1]; + } + overlap_entries--; + } + /* if there are overlapping entries, decide which "type" to use */ + /* (larger value takes precedence -- 1=usable, 2,3,4,4+=unusable) */ + current_type = 0; + for (i=0; itype > current_type) + current_type = overlap_list[i]->type; + /* continue building up new bios map based on this information */ + if (current_type != last_type) { + if (last_type != 0) { + new_bios[new_bios_entry].size = + change_point[chgidx]->addr - last_addr; + /* move forward only if the new size was non-zero */ + if (new_bios[new_bios_entry].size != 0) + if (++new_bios_entry >= E820MAX) + break; /* no more space left for new bios entries */ + } + if (current_type != 0) { + new_bios[new_bios_entry].addr = change_point[chgidx]->addr; + new_bios[new_bios_entry].type = current_type; + last_addr=change_point[chgidx]->addr; + } + last_type = current_type; + } + } + new_nr = new_bios_entry; /* retain count for new bios entries */ + + /* copy new bios mapping into original location */ + memcpy(biosmap, new_bios, new_nr*sizeof(struct e820entry)); + *pnr_map = new_nr; + + printk("sanitize end\n"); + return 0; +} + +/* + * Copy the BIOS e820 map into a safe place. + * + * Sanity-check it while we're at it.. + * + * If we're lucky and live on a modern system, the setup code + * will have given us a memory map that we can use to properly + * set up memory. If we aren't, we'll fake a memory map. + * + * We check to see that the memory map contains at least 2 elements + * before we'll use it, because the detection code in setup.S may + * not be perfect and most every PC known to man has two memory + * regions: one from 0 to 640k, and one from 1mb up. (The IBM + * thinkpad 560x, for example, does not cooperate with the memory + * detection code.) + */ +int __init copy_e820_map(struct e820entry * biosmap, int nr_map) +{ + /* Only one memory region (or negative)? Ignore it */ + if (nr_map < 2) + return -1; + + do { + unsigned long long start = biosmap->addr; + unsigned long long size = biosmap->size; + unsigned long long end = start + size; + unsigned long type = biosmap->type; + printk("copy_e820_map() start: %016Lx size: %016Lx end: %016Lx type: %ld\n", start, size, end, type); + + /* Overflow in 64 bits? Ignore the memory map. */ + if (start > end) + return -1; + + /* + * Some BIOSes claim RAM in the 640k - 1M region. + * Not right. Fix it up. + */ + if (type == E820_RAM) { + printk("copy_e820_map() type is E820_RAM\n"); + if (start < 0x100000ULL && end > 0xA0000ULL) { + printk("copy_e820_map() lies in range...\n"); + if (start < 0xA0000ULL) { + printk("copy_e820_map() start < 0xA0000ULL\n"); + add_memory_region(start, 0xA0000ULL-start, type); + } + if (end <= 0x100000ULL) { + printk("copy_e820_map() end <= 0x100000ULL\n"); + continue; + } + start = 0x100000ULL; + size = end - start; + } + } + add_memory_region(start, size, type); + } while (biosmap++,--nr_map); + return 0; +} + diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index acd2d9392ab..b7509aec0eb 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -191,26 +191,6 @@ static void __init limit_regions(unsigned long long size) } } -void __init add_memory_region(unsigned long long start, - unsigned long long size, int type) -{ - int x; - - if (!efi_enabled) { - x = e820.nr_map; - - if (x == E820MAX) { - printk(KERN_ERR "Ooops! Too many entries in the memory map!\n"); - return; - } - - e820.map[x].addr = start; - e820.map[x].size = size; - e820.map[x].type = type; - e820.nr_map++; - } -} /* add_memory_region */ - #define E820_DEBUG 1 static void __init print_memory_map(char *who) @@ -239,226 +219,6 @@ static void __init print_memory_map(char *who) } } -/* - * Sanitize the BIOS e820 map. - * - * Some e820 responses include overlapping entries. The following - * replaces the original e820 map with a new one, removing overlaps. - * - */ -struct change_member { - struct e820entry *pbios; /* pointer to original bios entry */ - unsigned long long addr; /* address for this change point */ -}; -static struct change_member change_point_list[2*E820MAX] __initdata; -static struct change_member *change_point[2*E820MAX] __initdata; -static struct e820entry *overlap_list[E820MAX] __initdata; -static struct e820entry new_bios[E820MAX] __initdata; - -int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) -{ - struct change_member *change_tmp; - unsigned long current_type, last_type; - unsigned long long last_addr; - int chgidx, still_changing; - int overlap_entries; - int new_bios_entry; - int old_nr, new_nr, chg_nr; - int i; - - /* - Visually we're performing the following (1,2,3,4 = memory types)... - - Sample memory map (w/overlaps): - ____22__________________ - ______________________4_ - ____1111________________ - _44_____________________ - 11111111________________ - ____________________33__ - ___________44___________ - __________33333_________ - ______________22________ - ___________________2222_ - _________111111111______ - _____________________11_ - _________________4______ - - Sanitized equivalent (no overlap): - 1_______________________ - _44_____________________ - ___1____________________ - ____22__________________ - ______11________________ - _________1______________ - __________3_____________ - ___________44___________ - _____________33_________ - _______________2________ - ________________1_______ - _________________4______ - ___________________2____ - ____________________33__ - ______________________4_ - */ - - /* if there's only one memory region, don't bother */ - if (*pnr_map < 2) - return -1; - - old_nr = *pnr_map; - - /* bail out if we find any unreasonable addresses in bios map */ - for (i=0; iaddr = biosmap[i].addr; - change_point[chgidx++]->pbios = &biosmap[i]; - change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size; - change_point[chgidx++]->pbios = &biosmap[i]; - } - } - chg_nr = chgidx; /* true number of change-points */ - - /* sort change-point list by memory addresses (low -> high) */ - still_changing = 1; - while (still_changing) { - still_changing = 0; - for (i=1; i < chg_nr; i++) { - /* if > , swap */ - /* or, if current= & last=, swap */ - if ((change_point[i]->addr < change_point[i-1]->addr) || - ((change_point[i]->addr == change_point[i-1]->addr) && - (change_point[i]->addr == change_point[i]->pbios->addr) && - (change_point[i-1]->addr != change_point[i-1]->pbios->addr)) - ) - { - change_tmp = change_point[i]; - change_point[i] = change_point[i-1]; - change_point[i-1] = change_tmp; - still_changing=1; - } - } - } - - /* create a new bios memory map, removing overlaps */ - overlap_entries=0; /* number of entries in the overlap table */ - new_bios_entry=0; /* index for creating new bios map entries */ - last_type = 0; /* start with undefined memory type */ - last_addr = 0; /* start with 0 as last starting address */ - /* loop through change-points, determining affect on the new bios map */ - for (chgidx=0; chgidx < chg_nr; chgidx++) - { - /* keep track of all overlapping bios entries */ - if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr) - { - /* add map entry to overlap list (> 1 entry implies an overlap) */ - overlap_list[overlap_entries++]=change_point[chgidx]->pbios; - } - else - { - /* remove entry from list (order independent, so swap with last) */ - for (i=0; ipbios) - overlap_list[i] = overlap_list[overlap_entries-1]; - } - overlap_entries--; - } - /* if there are overlapping entries, decide which "type" to use */ - /* (larger value takes precedence -- 1=usable, 2,3,4,4+=unusable) */ - current_type = 0; - for (i=0; itype > current_type) - current_type = overlap_list[i]->type; - /* continue building up new bios map based on this information */ - if (current_type != last_type) { - if (last_type != 0) { - new_bios[new_bios_entry].size = - change_point[chgidx]->addr - last_addr; - /* move forward only if the new size was non-zero */ - if (new_bios[new_bios_entry].size != 0) - if (++new_bios_entry >= E820MAX) - break; /* no more space left for new bios entries */ - } - if (current_type != 0) { - new_bios[new_bios_entry].addr = change_point[chgidx]->addr; - new_bios[new_bios_entry].type = current_type; - last_addr=change_point[chgidx]->addr; - } - last_type = current_type; - } - } - new_nr = new_bios_entry; /* retain count for new bios entries */ - - /* copy new bios mapping into original location */ - memcpy(biosmap, new_bios, new_nr*sizeof(struct e820entry)); - *pnr_map = new_nr; - - return 0; -} - -/* - * Copy the BIOS e820 map into a safe place. - * - * Sanity-check it while we're at it.. - * - * If we're lucky and live on a modern system, the setup code - * will have given us a memory map that we can use to properly - * set up memory. If we aren't, we'll fake a memory map. - * - * We check to see that the memory map contains at least 2 elements - * before we'll use it, because the detection code in setup.S may - * not be perfect and most every PC known to man has two memory - * regions: one from 0 to 640k, and one from 1mb up. (The IBM - * thinkpad 560x, for example, does not cooperate with the memory - * detection code.) - */ -int __init copy_e820_map(struct e820entry * biosmap, int nr_map) -{ - /* Only one memory region (or negative)? Ignore it */ - if (nr_map < 2) - return -1; - - do { - unsigned long long start = biosmap->addr; - unsigned long long size = biosmap->size; - unsigned long long end = start + size; - unsigned long type = biosmap->type; - - /* Overflow in 64 bits? Ignore the memory map. */ - if (start > end) - return -1; - - /* - * Some BIOSes claim RAM in the 640k - 1M region. - * Not right. Fix it up. - */ - if (type == E820_RAM) { - if (start < 0x100000ULL && end > 0xA0000ULL) { - if (start < 0xA0000ULL) - add_memory_region(start, 0xA0000ULL-start, type); - if (end <= 0x100000ULL) - continue; - start = 0x100000ULL; - size = end - start; - } - } - add_memory_region(start, size, type); - } while (biosmap++,--nr_map); - return 0; -} - #if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE) struct edd edd; #ifdef CONFIG_EDD_MODULE -- cgit v1.2.3-70-g09d2 From b2dff6a88cbed59d787a8ca7367c76ba385e1187 Mon Sep 17 00:00:00 2001 From: "bibo,mao" Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] i386: Move find_max_pfn function to e820.c Move more code from setup.c into e820.c Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/kernel/e820.c | 52 +++++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 55 ------------------------------------------------ include/asm-i386/e820.h | 1 + 3 files changed, 53 insertions(+), 55 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/e820.c b/arch/i386/kernel/e820.c index 0db95760b07..be4934f6f85 100644 --- a/arch/i386/kernel/e820.c +++ b/arch/i386/kernel/e820.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -539,3 +540,54 @@ int __init copy_e820_map(struct e820entry * biosmap, int nr_map) return 0; } +/* + * Callback for efi_memory_walk. + */ +static int __init +efi_find_max_pfn(unsigned long start, unsigned long end, void *arg) +{ + unsigned long *max_pfn = arg, pfn; + + if (start < end) { + pfn = PFN_UP(end -1); + if (pfn > *max_pfn) + *max_pfn = pfn; + } + return 0; +} + +static int __init +efi_memory_present_wrapper(unsigned long start, unsigned long end, void *arg) +{ + memory_present(0, PFN_UP(start), PFN_DOWN(end)); + return 0; +} + +/* + * Find the highest page frame number we have available + */ +void __init find_max_pfn(void) +{ + int i; + + max_pfn = 0; + if (efi_enabled) { + efi_memmap_walk(efi_find_max_pfn, &max_pfn); + efi_memmap_walk(efi_memory_present_wrapper, NULL); + return; + } + + for (i = 0; i < e820.nr_map; i++) { + unsigned long start, end; + /* RAM? */ + if (e820.map[i].type != E820_RAM) + continue; + start = PFN_UP(e820.map[i].addr); + end = PFN_DOWN(e820.map[i].addr + e820.map[i].size); + if (start >= end) + continue; + if (end > max_pfn) + max_pfn = end; + memory_present(0, start, end); + } +} diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index b7509aec0eb..3d808054fdf 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -63,9 +63,6 @@ #include #include -/* Forward Declaration. */ -void __init find_max_pfn(void); - /* This value is set up by the early boot code to point to the value immediately after the boot time page tables. It contains a *physical* address, and must not be in the .bss segment! */ @@ -387,29 +384,6 @@ static int __init parse_reservetop(char *arg) } early_param("reservetop", parse_reservetop); -/* - * Callback for efi_memory_walk. - */ -static int __init -efi_find_max_pfn(unsigned long start, unsigned long end, void *arg) -{ - unsigned long *max_pfn = arg, pfn; - - if (start < end) { - pfn = PFN_UP(end -1); - if (pfn > *max_pfn) - *max_pfn = pfn; - } - return 0; -} - -static int __init -efi_memory_present_wrapper(unsigned long start, unsigned long end, void *arg) -{ - memory_present(0, PFN_UP(start), PFN_DOWN(end)); - return 0; -} - /* * This function checks if the entire range is mapped with type. * @@ -442,35 +416,6 @@ e820_all_mapped(unsigned long s, unsigned long e, unsigned type) return 0; } -/* - * Find the highest page frame number we have available - */ -void __init find_max_pfn(void) -{ - int i; - - max_pfn = 0; - if (efi_enabled) { - efi_memmap_walk(efi_find_max_pfn, &max_pfn); - efi_memmap_walk(efi_memory_present_wrapper, NULL); - return; - } - - for (i = 0; i < e820.nr_map; i++) { - unsigned long start, end; - /* RAM? */ - if (e820.map[i].type != E820_RAM) - continue; - start = PFN_UP(e820.map[i].addr); - end = PFN_DOWN(e820.map[i].addr + e820.map[i].size); - if (start >= end) - continue; - if (end > max_pfn) - max_pfn = end; - memory_present(0, start, end); - } -} - /* * Determine low and high memory ranges: */ diff --git a/include/asm-i386/e820.h b/include/asm-i386/e820.h index f7514fb6e8e..14756942515 100644 --- a/include/asm-i386/e820.h +++ b/include/asm-i386/e820.h @@ -38,6 +38,7 @@ extern struct e820map e820; extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type); +extern void find_max_pfn(void); #endif/*!__ASSEMBLY__*/ -- cgit v1.2.3-70-g09d2 From b5b2405706005cc7765f6ecd00965d29e93f090a Mon Sep 17 00:00:00 2001 From: "bibo,mao" Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] i386: Move e820/efi memmap walking code to e820.c This patch moves e820/efi memmap table walking function from setup.c to e820.c, also this patch adds extern declaration in header file. Signed-off-by: bibo,mao Signed-off-by: Andi Kleen arch/i386/kernel/e820.c | 115 +++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 118 ----------------------------------- include/asm-i386/e820.h | 2 arch/i386/kernel/e820.c | 115 +++++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 118 ----------------------------------------------- include/asm-i386/e820.h | 2 3 files changed, 117 insertions(+), 118 deletions(-) --- arch/i386/kernel/e820.c | 115 +++++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 118 ----------------------------------------------- include/asm-i386/e820.h | 2 + 3 files changed, 117 insertions(+), 118 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/e820.c b/arch/i386/kernel/e820.c index be4934f6f85..47c495bf0cb 100644 --- a/arch/i386/kernel/e820.c +++ b/arch/i386/kernel/e820.c @@ -28,6 +28,11 @@ static struct change_member change_point_list[2*E820MAX] __initdata; static struct change_member *change_point[2*E820MAX] __initdata; static struct e820entry *overlap_list[E820MAX] __initdata; static struct e820entry new_bios[E820MAX] __initdata; +/* For PCI or other memory-mapped resources */ +unsigned long pci_mem_start = 0x10000000; +#ifdef CONFIG_PCI +EXPORT_SYMBOL(pci_mem_start); +#endif struct resource data_resource = { .name = "Kernel data", .start = 0, @@ -591,3 +596,113 @@ void __init find_max_pfn(void) memory_present(0, start, end); } } + +/* + * Free all available memory for boot time allocation. Used + * as a callback function by efi_memory_walk() + */ + +static int __init +free_available_memory(unsigned long start, unsigned long end, void *arg) +{ + /* check max_low_pfn */ + if (start >= (max_low_pfn << PAGE_SHIFT)) + return 0; + if (end >= (max_low_pfn << PAGE_SHIFT)) + end = max_low_pfn << PAGE_SHIFT; + if (start < end) + free_bootmem(start, end - start); + + return 0; +} +/* + * Register fully available low RAM pages with the bootmem allocator. + */ +void __init register_bootmem_low_pages(unsigned long max_low_pfn) +{ + int i; + + if (efi_enabled) { + efi_memmap_walk(free_available_memory, NULL); + return; + } + for (i = 0; i < e820.nr_map; i++) { + unsigned long curr_pfn, last_pfn, size; + /* + * Reserve usable low memory + */ + if (e820.map[i].type != E820_RAM) + continue; + /* + * We are rounding up the start address of usable memory: + */ + curr_pfn = PFN_UP(e820.map[i].addr); + if (curr_pfn >= max_low_pfn) + continue; + /* + * ... and at the end of the usable range downwards: + */ + last_pfn = PFN_DOWN(e820.map[i].addr + e820.map[i].size); + + if (last_pfn > max_low_pfn) + last_pfn = max_low_pfn; + + /* + * .. finally, did all the rounding and playing + * around just make the area go away? + */ + if (last_pfn <= curr_pfn) + continue; + + size = last_pfn - curr_pfn; + free_bootmem(PFN_PHYS(curr_pfn), PFN_PHYS(size)); + } +} + +void __init register_memory(void) +{ + unsigned long gapstart, gapsize, round; + unsigned long long last; + int i; + + /* + * Search for the bigest gap in the low 32 bits of the e820 + * memory space. + */ + last = 0x100000000ull; + gapstart = 0x10000000; + gapsize = 0x400000; + i = e820.nr_map; + while (--i >= 0) { + unsigned long long start = e820.map[i].addr; + unsigned long long end = start + e820.map[i].size; + + /* + * Since "last" is at most 4GB, we know we'll + * fit in 32 bits if this condition is true + */ + if (last > end) { + unsigned long gap = last - end; + + if (gap > gapsize) { + gapsize = gap; + gapstart = end; + } + } + if (start < last) + last = start; + } + + /* + * See how much we want to round up: start off with + * rounding to the next 1MB area. + */ + round = 0x100000; + while ((gapsize >> 4) > round) + round += round; + /* Fun with two's complement */ + pci_mem_start = (gapstart + round) & -round; + + printk("Allocating PCI resources starting at %08lx (gap: %08lx:%08lx)\n", + pci_mem_start, gapstart, gapsize); +} diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 3d808054fdf..51ed015a1f3 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -94,12 +94,6 @@ unsigned int machine_submodel_id; unsigned int BIOS_revision; unsigned int mca_pentium_flag; -/* For PCI or other memory-mapped resources */ -unsigned long pci_mem_start = 0x10000000; -#ifdef CONFIG_PCI -EXPORT_SYMBOL(pci_mem_start); -#endif - /* Boot loader ID as an integer, for the benefit of proc_dointvec */ int bootloader_type; @@ -475,68 +469,6 @@ unsigned long __init find_max_low_pfn(void) return max_low_pfn; } -/* - * Free all available memory for boot time allocation. Used - * as a callback function by efi_memory_walk() - */ - -static int __init -free_available_memory(unsigned long start, unsigned long end, void *arg) -{ - /* check max_low_pfn */ - if (start >= (max_low_pfn << PAGE_SHIFT)) - return 0; - if (end >= (max_low_pfn << PAGE_SHIFT)) - end = max_low_pfn << PAGE_SHIFT; - if (start < end) - free_bootmem(start, end - start); - - return 0; -} -/* - * Register fully available low RAM pages with the bootmem allocator. - */ -static void __init register_bootmem_low_pages(unsigned long max_low_pfn) -{ - int i; - - if (efi_enabled) { - efi_memmap_walk(free_available_memory, NULL); - return; - } - for (i = 0; i < e820.nr_map; i++) { - unsigned long curr_pfn, last_pfn, size; - /* - * Reserve usable low memory - */ - if (e820.map[i].type != E820_RAM) - continue; - /* - * We are rounding up the start address of usable memory: - */ - curr_pfn = PFN_UP(e820.map[i].addr); - if (curr_pfn >= max_low_pfn) - continue; - /* - * ... and at the end of the usable range downwards: - */ - last_pfn = PFN_DOWN(e820.map[i].addr + e820.map[i].size); - - if (last_pfn > max_low_pfn) - last_pfn = max_low_pfn; - - /* - * .. finally, did all the rounding and playing - * around just make the area go away? - */ - if (last_pfn <= curr_pfn) - continue; - - size = last_pfn - curr_pfn; - free_bootmem(PFN_PHYS(curr_pfn), PFN_PHYS(size)); - } -} - /* * workaround for Dell systems that neglect to reserve EBDA */ @@ -705,56 +637,6 @@ void __init remapped_pgdat_init(void) } } - - -static void __init register_memory(void) -{ - unsigned long gapstart, gapsize, round; - unsigned long long last; - int i; - - /* - * Search for the bigest gap in the low 32 bits of the e820 - * memory space. - */ - last = 0x100000000ull; - gapstart = 0x10000000; - gapsize = 0x400000; - i = e820.nr_map; - while (--i >= 0) { - unsigned long long start = e820.map[i].addr; - unsigned long long end = start + e820.map[i].size; - - /* - * Since "last" is at most 4GB, we know we'll - * fit in 32 bits if this condition is true - */ - if (last > end) { - unsigned long gap = last - end; - - if (gap > gapsize) { - gapsize = gap; - gapstart = end; - } - } - if (start < last) - last = start; - } - - /* - * See how much we want to round up: start off with - * rounding to the next 1MB area. - */ - round = 0x100000; - while ((gapsize >> 4) > round) - round += round; - /* Fun with two's complement */ - pci_mem_start = (gapstart + round) & -round; - - printk("Allocating PCI resources starting at %08lx (gap: %08lx:%08lx)\n", - pci_mem_start, gapstart, gapsize); -} - #ifdef CONFIG_MCA static void set_mca_bus(int x) { diff --git a/include/asm-i386/e820.h b/include/asm-i386/e820.h index 14756942515..8da4175a553 100644 --- a/include/asm-i386/e820.h +++ b/include/asm-i386/e820.h @@ -39,6 +39,8 @@ extern struct e820map e820; extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type); extern void find_max_pfn(void); +extern void register_bootmem_low_pages(unsigned long max_low_pfn); +extern void register_memory(void); #endif/*!__ASSEMBLY__*/ -- cgit v1.2.3-70-g09d2 From cef518e88b8ed94ea483c436ef5e5b151a3fabc6 Mon Sep 17 00:00:00 2001 From: "bibo,mao" Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] i386: Move memory map printing and other code to e820.c This patch moves e820 memory map print and memmap boot param parsing function from setup.c to e820.c, also adds limit_regions and print_memory_map declaration in header file. Signed-off-by: bibo,mao Signed-off-by: Andi Kleen arch/i386/kernel/e820.c | 152 +++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 158 --------------------------------- include/asm-i386/e820.h | 2 arch/i386/kernel/e820.c | 152 ++++++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 153 ----------------------------------------------- include/asm-i386/e820.h | 2 3 files changed, 155 insertions(+), 152 deletions(-) --- arch/i386/kernel/e820.c | 152 ++++++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 153 +---------------------------------------------- include/asm-i386/e820.h | 2 + 3 files changed, 155 insertions(+), 152 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/e820.c b/arch/i386/kernel/e820.c index 47c495bf0cb..b755255f272 100644 --- a/arch/i386/kernel/e820.c +++ b/arch/i386/kernel/e820.c @@ -33,6 +33,7 @@ unsigned long pci_mem_start = 0x10000000; #ifdef CONFIG_PCI EXPORT_SYMBOL(pci_mem_start); #endif +extern int user_defined_memmap; struct resource data_resource = { .name = "Kernel data", .start = 0, @@ -706,3 +707,154 @@ void __init register_memory(void) printk("Allocating PCI resources starting at %08lx (gap: %08lx:%08lx)\n", pci_mem_start, gapstart, gapsize); } + +void __init print_memory_map(char *who) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + printk(" %s: %016Lx - %016Lx ", who, + e820.map[i].addr, + e820.map[i].addr + e820.map[i].size); + switch (e820.map[i].type) { + case E820_RAM: printk("(usable)\n"); + break; + case E820_RESERVED: + printk("(reserved)\n"); + break; + case E820_ACPI: + printk("(ACPI data)\n"); + break; + case E820_NVS: + printk("(ACPI NVS)\n"); + break; + default: printk("type %lu\n", e820.map[i].type); + break; + } + } +} + +void __init limit_regions(unsigned long long size) +{ + unsigned long long current_addr = 0; + int i; + + print_memory_map("limit_regions start"); + if (efi_enabled) { + efi_memory_desc_t *md; + void *p; + + for (p = memmap.map, i = 0; p < memmap.map_end; + p += memmap.desc_size, i++) { + md = p; + current_addr = md->phys_addr + (md->num_pages << 12); + if (md->type == EFI_CONVENTIONAL_MEMORY) { + if (current_addr >= size) { + md->num_pages -= + (((current_addr-size) + PAGE_SIZE-1) >> PAGE_SHIFT); + memmap.nr_map = i + 1; + return; + } + } + } + } + for (i = 0; i < e820.nr_map; i++) { + current_addr = e820.map[i].addr + e820.map[i].size; + if (current_addr < size) + continue; + + if (e820.map[i].type != E820_RAM) + continue; + + if (e820.map[i].addr >= size) { + /* + * This region starts past the end of the + * requested size, skip it completely. + */ + e820.nr_map = i; + } else { + e820.nr_map = i + 1; + e820.map[i].size -= current_addr - size; + } + print_memory_map("limit_regions endfor"); + return; + } + print_memory_map("limit_regions endfunc"); +} + + /* + * This function checks if the entire range is mapped with type. + * + * Note: this function only works correct if the e820 table is sorted and + * not-overlapping, which is the case + */ +int __init +e820_all_mapped(unsigned long s, unsigned long e, unsigned type) +{ + u64 start = s; + u64 end = e; + int i; + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + if (type && ei->type != type) + continue; + /* is the region (part) in overlap with the current region ?*/ + if (ei->addr >= end || ei->addr + ei->size <= start) + continue; + /* if the region is at the beginning of we move + * start to the end of the region since it's ok until there + */ + if (ei->addr <= start) + start = ei->addr + ei->size; + /* if start is now at or beyond end, we're done, full + * coverage */ + if (start >= end) + return 1; /* we're done */ + } + return 0; +} + +static int __init parse_memmap(char *arg) +{ + if (!arg) + return -EINVAL; + + if (strcmp(arg, "exactmap") == 0) { +#ifdef CONFIG_CRASH_DUMP + /* If we are doing a crash dump, we + * still need to know the real mem + * size before original memory map is + * reset. + */ + find_max_pfn(); + saved_max_pfn = max_pfn; +#endif + e820.nr_map = 0; + user_defined_memmap = 1; + } else { + /* If the user specifies memory size, we + * limit the BIOS-provided memory map to + * that size. exactmap can be used to specify + * the exact map. mem=number can be used to + * trim the existing memory map. + */ + unsigned long long start_at, mem_size; + + mem_size = memparse(arg, &arg); + if (*arg == '@') { + start_at = memparse(arg+1, &arg); + add_memory_region(start_at, mem_size, E820_RAM); + } else if (*arg == '#') { + start_at = memparse(arg+1, &arg); + add_memory_region(start_at, mem_size, E820_ACPI); + } else if (*arg == '$') { + start_at = memparse(arg+1, &arg); + add_memory_region(start_at, mem_size, E820_RESERVED); + } else { + limit_regions(mem_size); + user_defined_memmap = 1; + } + } + return 0; +} +early_param("memmap", parse_memmap); diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 51ed015a1f3..e5bb87aa5a4 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -73,7 +73,6 @@ int disable_pse __devinitdata = 0; /* * Machine setup.. */ -extern struct e820map e820; extern struct resource code_resource; extern struct resource data_resource; @@ -137,79 +136,6 @@ static char command_line[COMMAND_LINE_SIZE]; unsigned char __initdata boot_params[PARAM_SIZE]; -static void __init limit_regions(unsigned long long size) -{ - unsigned long long current_addr = 0; - int i; - - if (efi_enabled) { - efi_memory_desc_t *md; - void *p; - - for (p = memmap.map, i = 0; p < memmap.map_end; - p += memmap.desc_size, i++) { - md = p; - current_addr = md->phys_addr + (md->num_pages << 12); - if (md->type == EFI_CONVENTIONAL_MEMORY) { - if (current_addr >= size) { - md->num_pages -= - (((current_addr-size) + PAGE_SIZE-1) >> PAGE_SHIFT); - memmap.nr_map = i + 1; - return; - } - } - } - } - for (i = 0; i < e820.nr_map; i++) { - current_addr = e820.map[i].addr + e820.map[i].size; - if (current_addr < size) - continue; - - if (e820.map[i].type != E820_RAM) - continue; - - if (e820.map[i].addr >= size) { - /* - * This region starts past the end of the - * requested size, skip it completely. - */ - e820.nr_map = i; - } else { - e820.nr_map = i + 1; - e820.map[i].size -= current_addr - size; - } - return; - } -} - -#define E820_DEBUG 1 - -static void __init print_memory_map(char *who) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - printk(" %s: %016Lx - %016Lx ", who, - e820.map[i].addr, - e820.map[i].addr + e820.map[i].size); - switch (e820.map[i].type) { - case E820_RAM: printk("(usable)\n"); - break; - case E820_RESERVED: - printk("(reserved)\n"); - break; - case E820_ACPI: - printk("(ACPI data)\n"); - break; - case E820_NVS: - printk("(ACPI NVS)\n"); - break; - default: printk("type %lu\n", e820.map[i].type); - break; - } - } -} - #if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE) struct edd edd; #ifdef CONFIG_EDD_MODULE @@ -233,7 +159,7 @@ static inline void copy_edd(void) } #endif -static int __initdata user_defined_memmap = 0; +int __initdata user_defined_memmap = 0; /* * "mem=nopentium" disables the 4MB page tables. @@ -270,51 +196,6 @@ static int __init parse_mem(char *arg) } early_param("mem", parse_mem); -static int __init parse_memmap(char *arg) -{ - if (!arg) - return -EINVAL; - - if (strcmp(arg, "exactmap") == 0) { -#ifdef CONFIG_CRASH_DUMP - /* If we are doing a crash dump, we - * still need to know the real mem - * size before original memory map is - * reset. - */ - find_max_pfn(); - saved_max_pfn = max_pfn; -#endif - e820.nr_map = 0; - user_defined_memmap = 1; - } else { - /* If the user specifies memory size, we - * limit the BIOS-provided memory map to - * that size. exactmap can be used to specify - * the exact map. mem=number can be used to - * trim the existing memory map. - */ - unsigned long long start_at, mem_size; - - mem_size = memparse(arg, &arg); - if (*arg == '@') { - start_at = memparse(arg+1, &arg); - add_memory_region(start_at, mem_size, E820_RAM); - } else if (*arg == '#') { - start_at = memparse(arg+1, &arg); - add_memory_region(start_at, mem_size, E820_ACPI); - } else if (*arg == '$') { - start_at = memparse(arg+1, &arg); - add_memory_region(start_at, mem_size, E820_RESERVED); - } else { - limit_regions(mem_size); - user_defined_memmap = 1; - } - } - return 0; -} -early_param("memmap", parse_memmap); - #ifdef CONFIG_PROC_VMCORE /* elfcorehdr= specifies the location of elf core header * stored by the crashed kernel. @@ -378,38 +259,6 @@ static int __init parse_reservetop(char *arg) } early_param("reservetop", parse_reservetop); - /* - * This function checks if the entire range is mapped with type. - * - * Note: this function only works correct if the e820 table is sorted and - * not-overlapping, which is the case - */ -int __init -e820_all_mapped(unsigned long s, unsigned long e, unsigned type) -{ - u64 start = s; - u64 end = e; - int i; - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - if (type && ei->type != type) - continue; - /* is the region (part) in overlap with the current region ?*/ - if (ei->addr >= end || ei->addr + ei->size <= start) - continue; - /* if the region is at the beginning of we move - * start to the end of the region since it's ok until there - */ - if (ei->addr <= start) - start = ei->addr + ei->size; - /* if start is now at or beyond end, we're done, full - * coverage */ - if (start >= end) - return 1; /* we're done */ - } - return 0; -} - /* * Determine low and high memory ranges: */ diff --git a/include/asm-i386/e820.h b/include/asm-i386/e820.h index 8da4175a553..395077aba58 100644 --- a/include/asm-i386/e820.h +++ b/include/asm-i386/e820.h @@ -41,6 +41,8 @@ extern int e820_all_mapped(unsigned long start, unsigned long end, extern void find_max_pfn(void); extern void register_bootmem_low_pages(unsigned long max_low_pfn); extern void register_memory(void); +extern void limit_regions(unsigned long long size); +extern void print_memory_map(char *who); #endif/*!__ASSEMBLY__*/ -- cgit v1.2.3-70-g09d2 From d15512f442ef1ea60f6195b0444fb27b3cf8d0e6 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:07 +0100 Subject: [PATCH] i386: Fix race in IO-APIC routing entry setup. Interrupt could happen between setting the IO-APIC entry and setting its interrupt data. Pointed out by Linus. Signed-off-by: Andi Kleen --- arch/i386/kernel/io_apic.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 3b7a63e0ed1..e33b7a84529 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -153,14 +153,20 @@ static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) * the interrupt, and we need to make sure the entry is fully populated * before that happens. */ -static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) +static void +__ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { - unsigned long flags; union entry_union eu; eu.entry = e; - spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic, 0x11 + 2*pin, eu.w2); io_apic_write(apic, 0x10 + 2*pin, eu.w1); +} + +static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) +{ + unsigned long flags; + spin_lock_irqsave(&ioapic_lock, flags); + __ioapic_write_entry(apic, pin, e); spin_unlock_irqrestore(&ioapic_lock, flags); } @@ -1360,8 +1366,8 @@ static void __init setup_IO_APIC_irqs(void) if (!apic && (irq < 16)) disable_8259A_irq(irq); } - ioapic_write_entry(apic, pin, entry); spin_lock_irqsave(&ioapic_lock, flags); + __ioapic_write_entry(apic, pin, entry); set_native_irq_info(irq, TARGET_CPUS); spin_unlock_irqrestore(&ioapic_lock, flags); } @@ -2856,8 +2862,8 @@ int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int a if (!ioapic && (irq < 16)) disable_8259A_irq(irq); - ioapic_write_entry(ioapic, pin, entry); spin_lock_irqsave(&ioapic_lock, flags); + __ioapic_write_entry(ioapic, pin, entry); set_native_irq_info(irq, TARGET_CPUS); spin_unlock_irqrestore(&ioapic_lock, flags); -- cgit v1.2.3-70-g09d2 From 8c89812684de3b47066d800031dfd7098abbdc74 Mon Sep 17 00:00:00 2001 From: Chuck Ebbert <76306.1226@compuserve.com> Date: Thu, 7 Dec 2006 02:14:07 +0100 Subject: [PATCH] i386: remove IOPL check on task switch IOPL is implicitly saved and restored on task switch, so explicit check is no longer needed. Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com> Signed-off-by: Andi Kleen --- arch/i386/kernel/process.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 8f42659ef9d..99308510a17 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -674,12 +674,6 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas write_pda(pcurrent, next_p); - /* - * Restore IOPL if needed. - */ - if (unlikely(prev->iopl != next->iopl)) - set_iopl_mask(next->iopl); - /* * Now maybe handle debug registers and/or IO bitmaps */ -- cgit v1.2.3-70-g09d2 From 3529833f1ca8790df06ce218b5d9d438776696ed Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 7 Dec 2006 02:14:07 +0100 Subject: [PATCH] i386: substitute __va lookup with pfn_to_kaddr Substitutes allocate_pgdat virtual address lookup with pfn_to_kaddr macro. Signed-off-by: David Rientjes Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/mm/discontig.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c index ddbdb0336f2..103b76e56a9 100644 --- a/arch/i386/mm/discontig.c +++ b/arch/i386/mm/discontig.c @@ -168,7 +168,7 @@ static void __init allocate_pgdat(int nid) if (nid && node_has_online_mem(nid)) NODE_DATA(nid) = (pg_data_t *)node_remap_start_vaddr[nid]; else { - NODE_DATA(nid) = (pg_data_t *)(__va(min_low_pfn << PAGE_SHIFT)); + NODE_DATA(nid) = (pg_data_t *)(pfn_to_kaddr(min_low_pfn)); min_low_pfn += PFN_UP(sizeof(pg_data_t)); } } -- cgit v1.2.3-70-g09d2 From db91b882aabd0b3b55a87cbfb344f2798bb740b4 Mon Sep 17 00:00:00 2001 From: Nicolas Kaiser Date: Thu, 7 Dec 2006 02:14:07 +0100 Subject: [PATCH] i386: Fix double #includes in arch/i386 Fix double #includes in arch/i386 Signed-off-by: Nicolas Kaiser Signed-off-by: Andi Kleen --- arch/i386/kernel/cpuid.c | 1 - arch/i386/kernel/tsc.c | 1 - 2 files changed, 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpuid.c b/arch/i386/kernel/cpuid.c index ab0c327e79d..5c5d4507ee7 100644 --- a/arch/i386/kernel/cpuid.c +++ b/arch/i386/kernel/cpuid.c @@ -34,7 +34,6 @@ #include #include #include -#include #include #include #include diff --git a/arch/i386/kernel/tsc.c b/arch/i386/kernel/tsc.c index fbc95828cd7..7f22e03253e 100644 --- a/arch/i386/kernel/tsc.c +++ b/arch/i386/kernel/tsc.c @@ -13,7 +13,6 @@ #include #include -#include #include #include "mach_timer.h" -- cgit v1.2.3-70-g09d2 From d3561b7fa0fb0fc583bab0eeda32bec9e4c4056d Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:07 +0100 Subject: [PATCH] paravirt: header and stubs for paravirtualisation Create a paravirt.h header for all the critical operations which need to be replaced with hypervisor calls, and include that instead of defining native operations, when CONFIG_PARAVIRT. This patch does the dumbest possible replacement of paravirtualized instructions: calls through a "paravirt_ops" structure. Currently these are function implementations of native hardware: hypervisors will override the ops structure with their own variants. All the pv-ops functions are declared "fastcall" so that a specific register-based ABI is used, to make inlining assember easier. And: +From: Andy Whitcroft The paravirt ops introduce a 'weak' attribute onto memory_setup(). Code ordering leads to the following warnings on x86: arch/i386/kernel/setup.c:651: warning: weak declaration of `memory_setup' after first use results in unspecified behavior Move memory_setup() to avoid this. Signed-off-by: Rusty Russell Signed-off-by: Chris Wright Signed-off-by: Andi Kleen Cc: Jeremy Fitzhardinge Cc: Zachary Amsden Signed-off-by: Andrew Morton Signed-off-by: Andy Whitcroft --- arch/i386/Kconfig | 11 + arch/i386/boot/compressed/misc.c | 1 + arch/i386/kernel/Makefile | 1 + arch/i386/kernel/asm-offsets.c | 10 + arch/i386/kernel/entry.S | 34 ++- arch/i386/kernel/i8259.c | 5 +- arch/i386/kernel/paravirt.c | 404 +++++++++++++++++++++++++++++ arch/i386/kernel/setup.c | 8 +- arch/i386/kernel/smpboot.c | 5 + arch/i386/kernel/time.c | 15 +- arch/i386/power/cpu.c | 8 +- drivers/net/de600.c | 1 - include/asm-i386/delay.h | 8 + include/asm-i386/desc.h | 9 +- include/asm-i386/io.h | 8 +- include/asm-i386/irq.h | 3 + include/asm-i386/irqflags.h | 42 +-- include/asm-i386/mach-default/setup_arch.h | 2 + include/asm-i386/msr.h | 5 + include/asm-i386/paravirt.h | 281 ++++++++++++++++++++ include/asm-i386/processor.h | 15 +- include/asm-i386/segment.h | 2 + include/asm-i386/setup.h | 1 + include/asm-i386/spinlock.h | 4 + include/asm-i386/suspend.h | 8 +- include/asm-i386/system.h | 16 +- include/asm-i386/time.h | 41 +++ 27 files changed, 890 insertions(+), 58 deletions(-) create mode 100644 arch/i386/kernel/paravirt.c create mode 100644 include/asm-i386/paravirt.h create mode 100644 include/asm-i386/time.h (limited to 'arch/i386') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 1f0f7b60995..bb1fa061c6c 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -182,6 +182,17 @@ config X86_ES7000 endchoice +config PARAVIRT + bool "Paravirtualization support (EXPERIMENTAL)" + depends on EXPERIMENTAL + help + Paravirtualization is a way of running multiple instances of + Linux on the same machine, under a hypervisor. This option + changes the kernel so it can modify itself when it is run + under a hypervisor, improving performance significantly. + However, when run without a hypervisor the kernel is + theoretically slower. If in doubt, say N. + config ACPI_SRAT bool default y diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c index dc153893155..c6798c75c67 100644 --- a/arch/i386/boot/compressed/misc.c +++ b/arch/i386/boot/compressed/misc.c @@ -9,6 +9,7 @@ * High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996 */ +#undef CONFIG_PARAVIRT #include #include #include diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile index f614854bd71..40661213604 100644 --- a/arch/i386/kernel/Makefile +++ b/arch/i386/kernel/Makefile @@ -39,6 +39,7 @@ obj-$(CONFIG_VM86) += vm86.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o obj-$(CONFIG_HPET_TIMER) += hpet.o obj-$(CONFIG_K8_NB) += k8.o +obj-$(CONFIG_PARAVIRT) += paravirt.o EXTRA_AFLAGS := -traditional diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index 0666eb0ed7b..1b2f3cd3327 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -101,4 +101,14 @@ void foo(void) BLANK(); OFFSET(PDA_cpu, i386_pda, cpu_number); OFFSET(PDA_pcurrent, i386_pda, pcurrent); + +#ifdef CONFIG_PARAVIRT + BLANK(); + OFFSET(PARAVIRT_enabled, paravirt_ops, paravirt_enabled); + OFFSET(PARAVIRT_irq_disable, paravirt_ops, irq_disable); + OFFSET(PARAVIRT_irq_enable, paravirt_ops, irq_enable); + OFFSET(PARAVIRT_irq_enable_sysexit, paravirt_ops, irq_enable_sysexit); + OFFSET(PARAVIRT_iret, paravirt_ops, iret); + OFFSET(PARAVIRT_read_cr0, paravirt_ops, read_cr0); +#endif } diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index 0220bc8cbb4..d274612e05c 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -62,13 +62,6 @@ DF_MASK = 0x00000400 NT_MASK = 0x00004000 VM_MASK = 0x00020000 -/* These are replaces for paravirtualization */ -#define DISABLE_INTERRUPTS cli -#define ENABLE_INTERRUPTS sti -#define ENABLE_INTERRUPTS_SYSEXIT sti; sysexit -#define INTERRUPT_RETURN iret -#define GET_CR0_INTO_EAX movl %cr0, %eax - #ifdef CONFIG_PREEMPT #define preempt_stop DISABLE_INTERRUPTS; TRACE_IRQS_OFF #else @@ -416,6 +409,20 @@ ldt_ss: jnz restore_nocheck testl $0x00400000, %eax # returning to 32bit stack? jnz restore_nocheck # allright, normal return + +#ifdef CONFIG_PARAVIRT + /* + * The kernel can't run on a non-flat stack if paravirt mode + * is active. Rather than try to fixup the high bits of + * ESP, bypass this code entirely. This may break DOSemu + * and/or Wine support in a paravirt VM, although the option + * is still available to implement the setting of the high + * 16-bits in the INTERRUPT_RETURN paravirt-op. + */ + cmpl $0, paravirt_ops+PARAVIRT_enabled + jne restore_nocheck +#endif + /* If returning to userspace with 16bit stack, * try to fix the higher word of ESP, as the CPU * won't restore it. @@ -833,6 +840,19 @@ nmi_espfix_stack: .previous KPROBE_END(nmi) +#ifdef CONFIG_PARAVIRT +ENTRY(native_iret) +1: iret +.section __ex_table,"a" + .align 4 + .long 1b,iret_exc +.previous + +ENTRY(native_irq_enable_sysexit) + sti + sysexit +#endif + KPROBE_ENTRY(int3) RING0_INT_FRAME pushl $-1 # mark this as an int diff --git a/arch/i386/kernel/i8259.c b/arch/i386/kernel/i8259.c index 62996cd1708..c8d45821c78 100644 --- a/arch/i386/kernel/i8259.c +++ b/arch/i386/kernel/i8259.c @@ -381,7 +381,10 @@ void __init init_ISA_irqs (void) } } -void __init init_IRQ(void) +/* Overridden in paravirt.c */ +void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ"))); + +void __init native_init_IRQ(void) { int i; diff --git a/arch/i386/kernel/paravirt.c b/arch/i386/kernel/paravirt.c new file mode 100644 index 00000000000..478192cd4b9 --- /dev/null +++ b/arch/i386/kernel/paravirt.c @@ -0,0 +1,404 @@ +/* Paravirtualization interfaces + Copyright (C) 2006 Rusty Russell IBM Corporation + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA +*/ +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +/* nop stub */ +static void native_nop(void) +{ +} + +static void __init default_banner(void) +{ + printk(KERN_INFO "Booting paravirtualized kernel on %s\n", + paravirt_ops.name); +} + +char *memory_setup(void) +{ + return paravirt_ops.memory_setup(); +} + +static fastcall unsigned long native_get_debugreg(int regno) +{ + unsigned long val = 0; /* Damn you, gcc! */ + + switch (regno) { + case 0: + asm("movl %%db0, %0" :"=r" (val)); break; + case 1: + asm("movl %%db1, %0" :"=r" (val)); break; + case 2: + asm("movl %%db2, %0" :"=r" (val)); break; + case 3: + asm("movl %%db3, %0" :"=r" (val)); break; + case 6: + asm("movl %%db6, %0" :"=r" (val)); break; + case 7: + asm("movl %%db7, %0" :"=r" (val)); break; + default: + BUG(); + } + return val; +} + +static fastcall void native_set_debugreg(int regno, unsigned long value) +{ + switch (regno) { + case 0: + asm("movl %0,%%db0" : /* no output */ :"r" (value)); + break; + case 1: + asm("movl %0,%%db1" : /* no output */ :"r" (value)); + break; + case 2: + asm("movl %0,%%db2" : /* no output */ :"r" (value)); + break; + case 3: + asm("movl %0,%%db3" : /* no output */ :"r" (value)); + break; + case 6: + asm("movl %0,%%db6" : /* no output */ :"r" (value)); + break; + case 7: + asm("movl %0,%%db7" : /* no output */ :"r" (value)); + break; + default: + BUG(); + } +} + +void init_IRQ(void) +{ + paravirt_ops.init_IRQ(); +} + +static fastcall void native_clts(void) +{ + asm volatile ("clts"); +} + +static fastcall unsigned long native_read_cr0(void) +{ + unsigned long val; + asm volatile("movl %%cr0,%0\n\t" :"=r" (val)); + return val; +} + +static fastcall void native_write_cr0(unsigned long val) +{ + asm volatile("movl %0,%%cr0": :"r" (val)); +} + +static fastcall unsigned long native_read_cr2(void) +{ + unsigned long val; + asm volatile("movl %%cr2,%0\n\t" :"=r" (val)); + return val; +} + +static fastcall void native_write_cr2(unsigned long val) +{ + asm volatile("movl %0,%%cr2": :"r" (val)); +} + +static fastcall unsigned long native_read_cr3(void) +{ + unsigned long val; + asm volatile("movl %%cr3,%0\n\t" :"=r" (val)); + return val; +} + +static fastcall void native_write_cr3(unsigned long val) +{ + asm volatile("movl %0,%%cr3": :"r" (val)); +} + +static fastcall unsigned long native_read_cr4(void) +{ + unsigned long val; + asm volatile("movl %%cr4,%0\n\t" :"=r" (val)); + return val; +} + +static fastcall unsigned long native_read_cr4_safe(void) +{ + unsigned long val; + /* This could fault if %cr4 does not exist */ + asm("1: movl %%cr4, %0 \n" + "2: \n" + ".section __ex_table,\"a\" \n" + ".long 1b,2b \n" + ".previous \n" + : "=r" (val): "0" (0)); + return val; +} + +static fastcall void native_write_cr4(unsigned long val) +{ + asm volatile("movl %0,%%cr4": :"r" (val)); +} + +static fastcall unsigned long native_save_fl(void) +{ + unsigned long f; + asm volatile("pushfl ; popl %0":"=g" (f): /* no input */); + return f; +} + +static fastcall void native_restore_fl(unsigned long f) +{ + asm volatile("pushl %0 ; popfl": /* no output */ + :"g" (f) + :"memory", "cc"); +} + +static fastcall void native_irq_disable(void) +{ + asm volatile("cli": : :"memory"); +} + +static fastcall void native_irq_enable(void) +{ + asm volatile("sti": : :"memory"); +} + +static fastcall void native_safe_halt(void) +{ + asm volatile("sti; hlt": : :"memory"); +} + +static fastcall void native_halt(void) +{ + asm volatile("hlt": : :"memory"); +} + +static fastcall void native_wbinvd(void) +{ + asm volatile("wbinvd": : :"memory"); +} + +static fastcall unsigned long long native_read_msr(unsigned int msr, int *err) +{ + unsigned long long val; + + asm volatile("2: rdmsr ; xorl %0,%0\n" + "1:\n\t" + ".section .fixup,\"ax\"\n\t" + "3: movl %3,%0 ; jmp 1b\n\t" + ".previous\n\t" + ".section __ex_table,\"a\"\n" + " .align 4\n\t" + " .long 2b,3b\n\t" + ".previous" + : "=r" (*err), "=A" (val) + : "c" (msr), "i" (-EFAULT)); + + return val; +} + +static fastcall int native_write_msr(unsigned int msr, unsigned long long val) +{ + int err; + asm volatile("2: wrmsr ; xorl %0,%0\n" + "1:\n\t" + ".section .fixup,\"ax\"\n\t" + "3: movl %4,%0 ; jmp 1b\n\t" + ".previous\n\t" + ".section __ex_table,\"a\"\n" + " .align 4\n\t" + " .long 2b,3b\n\t" + ".previous" + : "=a" (err) + : "c" (msr), "0" ((u32)val), "d" ((u32)(val>>32)), + "i" (-EFAULT)); + return err; +} + +static fastcall unsigned long long native_read_tsc(void) +{ + unsigned long long val; + asm volatile("rdtsc" : "=A" (val)); + return val; +} + +static fastcall unsigned long long native_read_pmc(void) +{ + unsigned long long val; + asm volatile("rdpmc" : "=A" (val)); + return val; +} + +static fastcall void native_load_tr_desc(void) +{ + asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8)); +} + +static fastcall void native_load_gdt(const struct Xgt_desc_struct *dtr) +{ + asm volatile("lgdt %0"::"m" (*dtr)); +} + +static fastcall void native_load_idt(const struct Xgt_desc_struct *dtr) +{ + asm volatile("lidt %0"::"m" (*dtr)); +} + +static fastcall void native_store_gdt(struct Xgt_desc_struct *dtr) +{ + asm ("sgdt %0":"=m" (*dtr)); +} + +static fastcall void native_store_idt(struct Xgt_desc_struct *dtr) +{ + asm ("sidt %0":"=m" (*dtr)); +} + +static fastcall unsigned long native_store_tr(void) +{ + unsigned long tr; + asm ("str %0":"=r" (tr)); + return tr; +} + +static fastcall void native_load_tls(struct thread_struct *t, unsigned int cpu) +{ +#define C(i) get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i] + C(0); C(1); C(2); +#undef C +} + +static inline void native_write_dt_entry(void *dt, int entry, u32 entry_low, u32 entry_high) +{ + u32 *lp = (u32 *)((char *)dt + entry*8); + lp[0] = entry_low; + lp[1] = entry_high; +} + +static fastcall void native_write_ldt_entry(void *dt, int entrynum, u32 low, u32 high) +{ + native_write_dt_entry(dt, entrynum, low, high); +} + +static fastcall void native_write_gdt_entry(void *dt, int entrynum, u32 low, u32 high) +{ + native_write_dt_entry(dt, entrynum, low, high); +} + +static fastcall void native_write_idt_entry(void *dt, int entrynum, u32 low, u32 high) +{ + native_write_dt_entry(dt, entrynum, low, high); +} + +static fastcall void native_load_esp0(struct tss_struct *tss, + struct thread_struct *thread) +{ + tss->esp0 = thread->esp0; + + /* This can only happen when SEP is enabled, no need to test "SEP"arately */ + if (unlikely(tss->ss1 != thread->sysenter_cs)) { + tss->ss1 = thread->sysenter_cs; + wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0); + } +} + +static fastcall void native_io_delay(void) +{ + asm volatile("outb %al,$0x80"); +} + +/* These are in entry.S */ +extern fastcall void native_iret(void); +extern fastcall void native_irq_enable_sysexit(void); + +static int __init print_banner(void) +{ + paravirt_ops.banner(); + return 0; +} +core_initcall(print_banner); + +struct paravirt_ops paravirt_ops = { + .name = "bare hardware", + .paravirt_enabled = 0, + .kernel_rpl = 0, + + .banner = default_banner, + .arch_setup = native_nop, + .memory_setup = machine_specific_memory_setup, + .get_wallclock = native_get_wallclock, + .set_wallclock = native_set_wallclock, + .time_init = time_init_hook, + .init_IRQ = native_init_IRQ, + + .cpuid = native_cpuid, + .get_debugreg = native_get_debugreg, + .set_debugreg = native_set_debugreg, + .clts = native_clts, + .read_cr0 = native_read_cr0, + .write_cr0 = native_write_cr0, + .read_cr2 = native_read_cr2, + .write_cr2 = native_write_cr2, + .read_cr3 = native_read_cr3, + .write_cr3 = native_write_cr3, + .read_cr4 = native_read_cr4, + .read_cr4_safe = native_read_cr4_safe, + .write_cr4 = native_write_cr4, + .save_fl = native_save_fl, + .restore_fl = native_restore_fl, + .irq_disable = native_irq_disable, + .irq_enable = native_irq_enable, + .safe_halt = native_safe_halt, + .halt = native_halt, + .wbinvd = native_wbinvd, + .read_msr = native_read_msr, + .write_msr = native_write_msr, + .read_tsc = native_read_tsc, + .read_pmc = native_read_pmc, + .load_tr_desc = native_load_tr_desc, + .set_ldt = native_set_ldt, + .load_gdt = native_load_gdt, + .load_idt = native_load_idt, + .store_gdt = native_store_gdt, + .store_idt = native_store_idt, + .store_tr = native_store_tr, + .load_tls = native_load_tls, + .write_ldt_entry = native_write_ldt_entry, + .write_gdt_entry = native_write_gdt_entry, + .write_idt_entry = native_write_idt_entry, + .load_esp0 = native_load_esp0, + + .set_iopl_mask = native_set_iopl_mask, + .io_delay = native_io_delay, + .const_udelay = __const_udelay, + + .irq_enable_sysexit = native_irq_enable_sysexit, + .iret = native_iret, +}; +EXPORT_SYMBOL(paravirt_ops); diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index e5bb87aa5a4..695d53fd14d 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -495,6 +495,12 @@ static void set_mca_bus(int x) static void set_mca_bus(int x) { } #endif +/* Overridden in paravirt.c if CONFIG_PARAVIRT */ +char * __attribute__((weak)) memory_setup(void) +{ + return machine_specific_memory_setup(); +} + /* * Determine if we were loaded by an EFI loader. If so, then we have also been * passed the efi memmap, systab, etc., so we should use these data structures @@ -547,7 +553,7 @@ void __init setup_arch(char **cmdline_p) efi_init(); else { printk(KERN_INFO "BIOS-provided physical RAM map:\n"); - print_memory_map(machine_specific_memory_setup()); + print_memory_map(memory_setup()); } copy_edd(); diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 095636620fa..cd7de9c9654 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -33,6 +33,11 @@ * Dave Jones : Report invalid combinations of Athlon CPUs. * Rusty Russell : Hacked into shape for new "hotplug" boot process. */ + +/* SMP boot always wants to use real time delay to allow sufficient time for + * the APs to come online */ +#define USE_REAL_TIME_DELAY + #include #include #include diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c index 78af572fd17..c505b16c099 100644 --- a/arch/i386/kernel/time.c +++ b/arch/i386/kernel/time.c @@ -56,6 +56,7 @@ #include #include #include +#include #include "mach_time.h" @@ -116,10 +117,7 @@ static int set_rtc_mmss(unsigned long nowtime) /* gets recalled with irq locally disabled */ /* XXX - does irqsave resolve this? -johnstul */ spin_lock_irqsave(&rtc_lock, flags); - if (efi_enabled) - retval = efi_set_rtc_mmss(nowtime); - else - retval = mach_set_rtc_mmss(nowtime); + retval = set_wallclock(nowtime); spin_unlock_irqrestore(&rtc_lock, flags); return retval; @@ -223,10 +221,7 @@ unsigned long get_cmos_time(void) spin_lock_irqsave(&rtc_lock, flags); - if (efi_enabled) - retval = efi_get_time(); - else - retval = mach_get_cmos_time(); + retval = get_wallclock(); spin_unlock_irqrestore(&rtc_lock, flags); @@ -370,7 +365,7 @@ static void __init hpet_time_init(void) printk("Using HPET for base-timer\n"); } - time_init_hook(); + do_time_init(); } #endif @@ -392,5 +387,5 @@ void __init time_init(void) do_settimeofday(&ts); - time_init_hook(); + do_time_init(); } diff --git a/arch/i386/power/cpu.c b/arch/i386/power/cpu.c index 5a1abeff033..2c15500f871 100644 --- a/arch/i386/power/cpu.c +++ b/arch/i386/power/cpu.c @@ -26,8 +26,8 @@ void __save_processor_state(struct saved_context *ctxt) /* * descriptor tables */ - store_gdt(&ctxt->gdt_limit); - store_idt(&ctxt->idt_limit); + store_gdt(&ctxt->gdt); + store_idt(&ctxt->idt); store_tr(ctxt->tr); /* @@ -99,8 +99,8 @@ void __restore_processor_state(struct saved_context *ctxt) * now restore the descriptor tables to their proper values * ltr is done i fix_processor_context(). */ - load_gdt(&ctxt->gdt_limit); - load_idt(&ctxt->idt_limit); + load_gdt(&ctxt->gdt); + load_idt(&ctxt->idt); /* * segment registers diff --git a/drivers/net/de600.c b/drivers/net/de600.c index 690bb40b353..8396e411f1c 100644 --- a/drivers/net/de600.c +++ b/drivers/net/de600.c @@ -43,7 +43,6 @@ static const char version[] = "de600.c: $Revision: 1.41-2.5 $, Bjorn Ekwall (bj * modify the following "#define": (see for more info) #define REALLY_SLOW_IO */ -#define SLOW_IO_BY_JUMPING /* Looks "better" than dummy write to port 0x80 :-) */ /* use 0 for production, 1 for verification, >2 for debug */ #ifdef DE600_DEBUG diff --git a/include/asm-i386/delay.h b/include/asm-i386/delay.h index 9ae5e3782ed..32d6678d0bb 100644 --- a/include/asm-i386/delay.h +++ b/include/asm-i386/delay.h @@ -16,6 +16,13 @@ extern void __ndelay(unsigned long nsecs); extern void __const_udelay(unsigned long usecs); extern void __delay(unsigned long loops); +#if defined(CONFIG_PARAVIRT) && !defined(USE_REAL_TIME_DELAY) +#define udelay(n) paravirt_ops.const_udelay((n) * 0x10c7ul) + +#define ndelay(n) paravirt_ops.const_udelay((n) * 5ul) + +#else /* !PARAVIRT || USE_REAL_TIME_DELAY */ + /* 0x10c7 is 2**32 / 1000000 (rounded up) */ #define udelay(n) (__builtin_constant_p(n) ? \ ((n) > 20000 ? __bad_udelay() : __const_udelay((n) * 0x10c7ul)) : \ @@ -25,6 +32,7 @@ extern void __delay(unsigned long loops); #define ndelay(n) (__builtin_constant_p(n) ? \ ((n) > 20000 ? __bad_ndelay() : __const_udelay((n) * 5ul)) : \ __ndelay(n)) +#endif void use_tsc_delay(void); diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h index 6cf2ac2bfde..f19820f0834 100644 --- a/include/asm-i386/desc.h +++ b/include/asm-i386/desc.h @@ -55,6 +55,9 @@ static inline void pack_gate(__u32 *a, __u32 *b, #define DESCTYPE_DPL3 0x60 /* DPL-3 */ #define DESCTYPE_S 0x10 /* !system */ +#ifdef CONFIG_PARAVIRT +#include +#else #define load_TR_desc() __asm__ __volatile__("ltr %w0"::"q" (GDT_ENTRY_TSS*8)) #define load_gdt(dtr) __asm__ __volatile("lgdt %0"::"m" (*dtr)) @@ -105,7 +108,11 @@ static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const vo write_gdt_entry(get_cpu_gdt_table(cpu), entry, a, b); } -static inline void set_ldt(void *addr, unsigned int entries) +#define set_ldt native_set_ldt +#endif /* CONFIG_PARAVIRT */ + +static inline fastcall void native_set_ldt(const void *addr, + unsigned int entries) { if (likely(entries == 0)) __asm__ __volatile__("lldt %w0"::"q" (0)); diff --git a/include/asm-i386/io.h b/include/asm-i386/io.h index 68df0dc3ab8..86ff5e83be2 100644 --- a/include/asm-i386/io.h +++ b/include/asm-i386/io.h @@ -256,11 +256,11 @@ static inline void flush_write_buffers(void) #endif /* __KERNEL__ */ -#ifdef SLOW_IO_BY_JUMPING -#define __SLOW_DOWN_IO "jmp 1f; 1: jmp 1f; 1:" +#if defined(CONFIG_PARAVIRT) +#include #else + #define __SLOW_DOWN_IO "outb %%al,$0x80;" -#endif static inline void slow_down_io(void) { __asm__ __volatile__( @@ -271,6 +271,8 @@ static inline void slow_down_io(void) { : : ); } +#endif + #ifdef CONFIG_X86_NUMAQ extern void *xquad_portio; /* Where the IO area was mapped */ #define XQUAD_PORT_ADDR(port, quad) (xquad_portio + (XQUAD_PORTIO_QUAD*quad) + port) diff --git a/include/asm-i386/irq.h b/include/asm-i386/irq.h index 331726b4112..9e15ce0006e 100644 --- a/include/asm-i386/irq.h +++ b/include/asm-i386/irq.h @@ -41,4 +41,7 @@ extern int irqbalance_disable(char *str); extern void fixup_irqs(cpumask_t map); #endif +void init_IRQ(void); +void __init native_init_IRQ(void); + #endif /* _ASM_IRQ_H */ diff --git a/include/asm-i386/irqflags.h b/include/asm-i386/irqflags.h index e1bdb97c07f..9ce01f3fb7b 100644 --- a/include/asm-i386/irqflags.h +++ b/include/asm-i386/irqflags.h @@ -10,6 +10,9 @@ #ifndef _ASM_IRQFLAGS_H #define _ASM_IRQFLAGS_H +#ifdef CONFIG_PARAVIRT +#include +#else #ifndef __ASSEMBLY__ static inline unsigned long __raw_local_save_flags(void) @@ -25,9 +28,6 @@ static inline unsigned long __raw_local_save_flags(void) return flags; } -#define raw_local_save_flags(flags) \ - do { (flags) = __raw_local_save_flags(); } while (0) - static inline void raw_local_irq_restore(unsigned long flags) { __asm__ __volatile__( @@ -66,18 +66,6 @@ static inline void halt(void) __asm__ __volatile__("hlt": : :"memory"); } -static inline int raw_irqs_disabled_flags(unsigned long flags) -{ - return !(flags & (1 << 9)); -} - -static inline int raw_irqs_disabled(void) -{ - unsigned long flags = __raw_local_save_flags(); - - return raw_irqs_disabled_flags(flags); -} - /* * For spinlocks, etc: */ @@ -90,9 +78,33 @@ static inline unsigned long __raw_local_irq_save(void) return flags; } +#else +#define DISABLE_INTERRUPTS cli +#define ENABLE_INTERRUPTS sti +#define ENABLE_INTERRUPTS_SYSEXIT sti; sysexit +#define INTERRUPT_RETURN iret +#define GET_CR0_INTO_EAX movl %cr0, %eax +#endif /* __ASSEMBLY__ */ +#endif /* CONFIG_PARAVIRT */ + +#ifndef __ASSEMBLY__ +#define raw_local_save_flags(flags) \ + do { (flags) = __raw_local_save_flags(); } while (0) + #define raw_local_irq_save(flags) \ do { (flags) = __raw_local_irq_save(); } while (0) +static inline int raw_irqs_disabled_flags(unsigned long flags) +{ + return !(flags & (1 << 9)); +} + +static inline int raw_irqs_disabled(void) +{ + unsigned long flags = __raw_local_save_flags(); + + return raw_irqs_disabled_flags(flags); +} #endif /* __ASSEMBLY__ */ /* diff --git a/include/asm-i386/mach-default/setup_arch.h b/include/asm-i386/mach-default/setup_arch.h index fb42099e7bd..605e3ccb991 100644 --- a/include/asm-i386/mach-default/setup_arch.h +++ b/include/asm-i386/mach-default/setup_arch.h @@ -2,4 +2,6 @@ /* no action for generic */ +#ifndef ARCH_SETUP #define ARCH_SETUP +#endif diff --git a/include/asm-i386/msr.h b/include/asm-i386/msr.h index 1820d9d73af..5679d499307 100644 --- a/include/asm-i386/msr.h +++ b/include/asm-i386/msr.h @@ -1,6 +1,10 @@ #ifndef __ASM_MSR_H #define __ASM_MSR_H +#ifdef CONFIG_PARAVIRT +#include +#else + /* * Access to machine-specific registers (available on 586 and better only) * Note: the rd* operations modify the parameters directly (without using @@ -77,6 +81,7 @@ static inline void wrmsrl (unsigned long msr, unsigned long long val) __asm__ __volatile__("rdpmc" \ : "=a" (low), "=d" (high) \ : "c" (counter)) +#endif /* !CONFIG_PARAVIRT */ /* symbolic names for some interesting MSRs */ /* Intel defined MSRs. */ diff --git a/include/asm-i386/paravirt.h b/include/asm-i386/paravirt.h new file mode 100644 index 00000000000..a7551a44686 --- /dev/null +++ b/include/asm-i386/paravirt.h @@ -0,0 +1,281 @@ +#ifndef __ASM_PARAVIRT_H +#define __ASM_PARAVIRT_H +/* Various instructions on x86 need to be replaced for + * para-virtualization: those hooks are defined here. */ +#include + +#ifdef CONFIG_PARAVIRT +#ifndef __ASSEMBLY__ +struct thread_struct; +struct Xgt_desc_struct; +struct tss_struct; +struct paravirt_ops +{ + unsigned int kernel_rpl; + int paravirt_enabled; + const char *name; + + void (*arch_setup)(void); + char *(*memory_setup)(void); + void (*init_IRQ)(void); + + void (*banner)(void); + + unsigned long (*get_wallclock)(void); + int (*set_wallclock)(unsigned long); + void (*time_init)(void); + + /* All the function pointers here are declared as "fastcall" + so that we get a specific register-based calling + convention. This makes it easier to implement inline + assembler replacements. */ + + void (fastcall *cpuid)(unsigned int *eax, unsigned int *ebx, + unsigned int *ecx, unsigned int *edx); + + unsigned long (fastcall *get_debugreg)(int regno); + void (fastcall *set_debugreg)(int regno, unsigned long value); + + void (fastcall *clts)(void); + + unsigned long (fastcall *read_cr0)(void); + void (fastcall *write_cr0)(unsigned long); + + unsigned long (fastcall *read_cr2)(void); + void (fastcall *write_cr2)(unsigned long); + + unsigned long (fastcall *read_cr3)(void); + void (fastcall *write_cr3)(unsigned long); + + unsigned long (fastcall *read_cr4_safe)(void); + unsigned long (fastcall *read_cr4)(void); + void (fastcall *write_cr4)(unsigned long); + + unsigned long (fastcall *save_fl)(void); + void (fastcall *restore_fl)(unsigned long); + void (fastcall *irq_disable)(void); + void (fastcall *irq_enable)(void); + void (fastcall *safe_halt)(void); + void (fastcall *halt)(void); + void (fastcall *wbinvd)(void); + + /* err = 0/-EFAULT. wrmsr returns 0/-EFAULT. */ + u64 (fastcall *read_msr)(unsigned int msr, int *err); + int (fastcall *write_msr)(unsigned int msr, u64 val); + + u64 (fastcall *read_tsc)(void); + u64 (fastcall *read_pmc)(void); + + void (fastcall *load_tr_desc)(void); + void (fastcall *load_gdt)(const struct Xgt_desc_struct *); + void (fastcall *load_idt)(const struct Xgt_desc_struct *); + void (fastcall *store_gdt)(struct Xgt_desc_struct *); + void (fastcall *store_idt)(struct Xgt_desc_struct *); + void (fastcall *set_ldt)(const void *desc, unsigned entries); + unsigned long (fastcall *store_tr)(void); + void (fastcall *load_tls)(struct thread_struct *t, unsigned int cpu); + void (fastcall *write_ldt_entry)(void *dt, int entrynum, + u32 low, u32 high); + void (fastcall *write_gdt_entry)(void *dt, int entrynum, + u32 low, u32 high); + void (fastcall *write_idt_entry)(void *dt, int entrynum, + u32 low, u32 high); + void (fastcall *load_esp0)(struct tss_struct *tss, + struct thread_struct *thread); + + void (fastcall *set_iopl_mask)(unsigned mask); + + void (fastcall *io_delay)(void); + void (*const_udelay)(unsigned long loops); + + /* These two are jmp to, not actually called. */ + void (fastcall *irq_enable_sysexit)(void); + void (fastcall *iret)(void); +}; + +extern struct paravirt_ops paravirt_ops; + +#define paravirt_enabled() (paravirt_ops.paravirt_enabled) + +static inline void load_esp0(struct tss_struct *tss, + struct thread_struct *thread) +{ + paravirt_ops.load_esp0(tss, thread); +} + +#define ARCH_SETUP paravirt_ops.arch_setup(); +static inline unsigned long get_wallclock(void) +{ + return paravirt_ops.get_wallclock(); +} + +static inline int set_wallclock(unsigned long nowtime) +{ + return paravirt_ops.set_wallclock(nowtime); +} + +static inline void do_time_init(void) +{ + return paravirt_ops.time_init(); +} + +/* The paravirtualized CPUID instruction. */ +static inline void __cpuid(unsigned int *eax, unsigned int *ebx, + unsigned int *ecx, unsigned int *edx) +{ + paravirt_ops.cpuid(eax, ebx, ecx, edx); +} + +/* + * These special macros can be used to get or set a debugging register + */ +#define get_debugreg(var, reg) var = paravirt_ops.get_debugreg(reg) +#define set_debugreg(val, reg) paravirt_ops.set_debugreg(reg, val) + +#define clts() paravirt_ops.clts() + +#define read_cr0() paravirt_ops.read_cr0() +#define write_cr0(x) paravirt_ops.write_cr0(x) + +#define read_cr2() paravirt_ops.read_cr2() +#define write_cr2(x) paravirt_ops.write_cr2(x) + +#define read_cr3() paravirt_ops.read_cr3() +#define write_cr3(x) paravirt_ops.write_cr3(x) + +#define read_cr4() paravirt_ops.read_cr4() +#define read_cr4_safe(x) paravirt_ops.read_cr4_safe() +#define write_cr4(x) paravirt_ops.write_cr4(x) + +static inline unsigned long __raw_local_save_flags(void) +{ + return paravirt_ops.save_fl(); +} + +static inline void raw_local_irq_restore(unsigned long flags) +{ + return paravirt_ops.restore_fl(flags); +} + +static inline void raw_local_irq_disable(void) +{ + paravirt_ops.irq_disable(); +} + +static inline void raw_local_irq_enable(void) +{ + paravirt_ops.irq_enable(); +} + +static inline unsigned long __raw_local_irq_save(void) +{ + unsigned long flags = paravirt_ops.save_fl(); + + paravirt_ops.irq_disable(); + + return flags; +} + +static inline void raw_safe_halt(void) +{ + paravirt_ops.safe_halt(); +} + +static inline void halt(void) +{ + paravirt_ops.safe_halt(); +} +#define wbinvd() paravirt_ops.wbinvd() + +#define get_kernel_rpl() (paravirt_ops.kernel_rpl) + +#define rdmsr(msr,val1,val2) do { \ + int _err; \ + u64 _l = paravirt_ops.read_msr(msr,&_err); \ + val1 = (u32)_l; \ + val2 = _l >> 32; \ +} while(0) + +#define wrmsr(msr,val1,val2) do { \ + u64 _l = ((u64)(val2) << 32) | (val1); \ + paravirt_ops.write_msr((msr), _l); \ +} while(0) + +#define rdmsrl(msr,val) do { \ + int _err; \ + val = paravirt_ops.read_msr((msr),&_err); \ +} while(0) + +#define wrmsrl(msr,val) (paravirt_ops.write_msr((msr),(val))) +#define wrmsr_safe(msr,a,b) ({ \ + u64 _l = ((u64)(b) << 32) | (a); \ + paravirt_ops.write_msr((msr),_l); \ +}) + +/* rdmsr with exception handling */ +#define rdmsr_safe(msr,a,b) ({ \ + int _err; \ + u64 _l = paravirt_ops.read_msr(msr,&_err); \ + (*a) = (u32)_l; \ + (*b) = _l >> 32; \ + _err; }) + +#define rdtsc(low,high) do { \ + u64 _l = paravirt_ops.read_tsc(); \ + low = (u32)_l; \ + high = _l >> 32; \ +} while(0) + +#define rdtscl(low) do { \ + u64 _l = paravirt_ops.read_tsc(); \ + low = (int)_l; \ +} while(0) + +#define rdtscll(val) (val = paravirt_ops.read_tsc()) + +#define write_tsc(val1,val2) wrmsr(0x10, val1, val2) + +#define rdpmc(counter,low,high) do { \ + u64 _l = paravirt_ops.read_pmc(); \ + low = (u32)_l; \ + high = _l >> 32; \ +} while(0) + +#define load_TR_desc() (paravirt_ops.load_tr_desc()) +#define load_gdt(dtr) (paravirt_ops.load_gdt(dtr)) +#define load_idt(dtr) (paravirt_ops.load_idt(dtr)) +#define set_ldt(addr, entries) (paravirt_ops.set_ldt((addr), (entries))) +#define store_gdt(dtr) (paravirt_ops.store_gdt(dtr)) +#define store_idt(dtr) (paravirt_ops.store_idt(dtr)) +#define store_tr(tr) ((tr) = paravirt_ops.store_tr()) +#define load_TLS(t,cpu) (paravirt_ops.load_tls((t),(cpu))) +#define write_ldt_entry(dt, entry, low, high) \ + (paravirt_ops.write_ldt_entry((dt), (entry), (low), (high))) +#define write_gdt_entry(dt, entry, low, high) \ + (paravirt_ops.write_gdt_entry((dt), (entry), (low), (high))) +#define write_idt_entry(dt, entry, low, high) \ + (paravirt_ops.write_idt_entry((dt), (entry), (low), (high))) +#define set_iopl_mask(mask) (paravirt_ops.set_iopl_mask(mask)) + +/* The paravirtualized I/O functions */ +static inline void slow_down_io(void) { + paravirt_ops.io_delay(); +#ifdef REALLY_SLOW_IO + paravirt_ops.io_delay(); + paravirt_ops.io_delay(); + paravirt_ops.io_delay(); +#endif +} + +#define CLI_STRING "pushl %eax; pushl %ecx; pushl %edx; call *paravirt_ops+PARAVIRT_irq_disable; popl %edx; popl %ecx; popl %eax" +#define STI_STRING "pushl %eax; pushl %ecx; pushl %edx; call *paravirt_ops+PARAVIRT_irq_enable; popl %edx; popl %ecx; popl %eax" +#else /* __ASSEMBLY__ */ + +#define INTERRUPT_RETURN jmp *%cs:paravirt_ops+PARAVIRT_iret +#define DISABLE_INTERRUPTS pushl %eax; pushl %ecx; pushl %edx; call *paravirt_ops+PARAVIRT_irq_disable; popl %edx; popl %ecx; popl %eax +#define ENABLE_INTERRUPTS pushl %eax; pushl %ecx; pushl %edx; call *%cs:paravirt_ops+PARAVIRT_irq_enable; popl %edx; popl %ecx; popl %eax +#define ENABLE_INTERRUPTS_SYSEXIT jmp *%cs:paravirt_ops+PARAVIRT_irq_enable_sysexit +#define GET_CR0_INTO_EAX call *paravirt_ops+PARAVIRT_read_cr0 +#endif /* __ASSEMBLY__ */ +#endif /* CONFIG_PARAVIRT */ +#endif /* __ASM_PARAVIRT_H */ diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h index 98fa73b7176..6c2c4457be0 100644 --- a/include/asm-i386/processor.h +++ b/include/asm-i386/processor.h @@ -144,8 +144,8 @@ static inline void detect_ht(struct cpuinfo_x86 *c) {} #define X86_EFLAGS_VIP 0x00100000 /* Virtual Interrupt Pending */ #define X86_EFLAGS_ID 0x00200000 /* CPUID detection flag */ -static inline void __cpuid(unsigned int *eax, unsigned int *ebx, - unsigned int *ecx, unsigned int *edx) +static inline fastcall void native_cpuid(unsigned int *eax, unsigned int *ebx, + unsigned int *ecx, unsigned int *edx) { /* ecx is often an input as well as an output. */ __asm__("cpuid" @@ -491,6 +491,12 @@ struct thread_struct { .io_bitmap = { [ 0 ... IO_BITMAP_LONGS] = ~0 }, \ } +#ifdef CONFIG_PARAVIRT +#include +#else +#define paravirt_enabled() 0 +#define __cpuid native_cpuid + static inline void load_esp0(struct tss_struct *tss, struct thread_struct *thread) { tss->esp0 = thread->esp0; @@ -524,10 +530,13 @@ static inline void load_esp0(struct tss_struct *tss, struct thread_struct *threa : /* no output */ \ :"r" (value)) +#define set_iopl_mask native_set_iopl_mask +#endif /* CONFIG_PARAVIRT */ + /* * Set IOPL bits in EFLAGS from given mask */ -static inline void set_iopl_mask(unsigned mask) +static fastcall inline void native_set_iopl_mask(unsigned mask) { unsigned int reg; __asm__ __volatile__ ("pushfl;" diff --git a/include/asm-i386/segment.h b/include/asm-i386/segment.h index 5bdda79b6b5..3c796af3377 100644 --- a/include/asm-i386/segment.h +++ b/include/asm-i386/segment.h @@ -131,5 +131,7 @@ #define SEGMENT_LDT 0x4 #define SEGMENT_GDT 0x0 +#ifndef CONFIG_PARAVIRT #define get_kernel_rpl() 0 #endif +#endif diff --git a/include/asm-i386/setup.h b/include/asm-i386/setup.h index 2734909eff8..9930c5a355f 100644 --- a/include/asm-i386/setup.h +++ b/include/asm-i386/setup.h @@ -70,6 +70,7 @@ extern unsigned char boot_params[PARAM_SIZE]; struct e820entry; char * __init machine_specific_memory_setup(void); +char *memory_setup(void); int __init copy_e820_map(struct e820entry * biosmap, int nr_map); int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map); diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h index c18b71fae6b..dea60709db2 100644 --- a/include/asm-i386/spinlock.h +++ b/include/asm-i386/spinlock.h @@ -7,8 +7,12 @@ #include #include +#ifdef CONFIG_PARAVIRT +#include +#else #define CLI_STRING "cli" #define STI_STRING "sti" +#endif /* CONFIG_PARAVIRT */ /* * Your basic SMP spinlocks, allowing only a single CPU anywhere diff --git a/include/asm-i386/suspend.h b/include/asm-i386/suspend.h index 08be1e5009d..30361526d56 100644 --- a/include/asm-i386/suspend.h +++ b/include/asm-i386/suspend.h @@ -23,12 +23,8 @@ arch_prepare_suspend(void) struct saved_context { u16 es, fs, gs, ss; unsigned long cr0, cr2, cr3, cr4; - u16 gdt_pad; - u16 gdt_limit; - unsigned long gdt_base; - u16 idt_pad; - u16 idt_limit; - unsigned long idt_base; + struct Xgt_desc_struct gdt; + struct Xgt_desc_struct idt; u16 ldt; u16 tss; unsigned long tr; diff --git a/include/asm-i386/system.h b/include/asm-i386/system.h index a6dabbcd6e6..a6d20d9a1a3 100644 --- a/include/asm-i386/system.h +++ b/include/asm-i386/system.h @@ -88,6 +88,9 @@ __asm__ __volatile__ ("movw %%dx,%1\n\t" \ #define savesegment(seg, value) \ asm volatile("mov %%" #seg ",%0":"=rm" (value)) +#ifdef CONFIG_PARAVIRT +#include +#else #define read_cr0() ({ \ unsigned int __dummy; \ __asm__ __volatile__( \ @@ -139,17 +142,18 @@ __asm__ __volatile__ ("movw %%dx,%1\n\t" \ #define write_cr4(x) \ __asm__ __volatile__("movl %0,%%cr4": :"r" (x)) -/* - * Clear and set 'TS' bit respectively - */ +#define wbinvd() \ + __asm__ __volatile__ ("wbinvd": : :"memory") + +/* Clear the 'TS' bit */ #define clts() __asm__ __volatile__ ("clts") +#endif/* CONFIG_PARAVIRT */ + +/* Set the 'TS' bit */ #define stts() write_cr0(8 | read_cr0()) #endif /* __KERNEL__ */ -#define wbinvd() \ - __asm__ __volatile__ ("wbinvd": : :"memory") - static inline unsigned long get_limit(unsigned long segment) { unsigned long __limit; diff --git a/include/asm-i386/time.h b/include/asm-i386/time.h new file mode 100644 index 00000000000..ea8065af825 --- /dev/null +++ b/include/asm-i386/time.h @@ -0,0 +1,41 @@ +#ifndef _ASMi386_TIME_H +#define _ASMi386_TIME_H + +#include +#include "mach_time.h" + +static inline unsigned long native_get_wallclock(void) +{ + unsigned long retval; + + if (efi_enabled) + retval = efi_get_time(); + else + retval = mach_get_cmos_time(); + + return retval; +} + +static inline int native_set_wallclock(unsigned long nowtime) +{ + int retval; + + if (efi_enabled) + retval = efi_set_rtc_mmss(nowtime); + else + retval = mach_set_rtc_mmss(nowtime); + + return retval; +} + +#ifdef CONFIG_PARAVIRT +#include +#else /* !CONFIG_PARAVIRT */ + +#define get_wallclock() native_get_wallclock() +#define set_wallclock(x) native_set_wallclock(x) +#define do_time_init() time_init_hook() + +#endif /* CONFIG_PARAVIRT */ + +#endif -- cgit v1.2.3-70-g09d2 From 139ec7c416248b9ea227d21839235344edfee1e0 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] paravirt: Patch inline replacements for paravirt intercepts It turns out that the most called ops, by several orders of magnitude, are the interrupt manipulation ops. These are obvious candidates for patching, so mark them up and create infrastructure for it. The method used is that the ops structure has a patch function, which is called for each place which needs to be patched: this returns a number of instructions (the rest are NOP-padded). Usually we can spare a register (%eax) for the binary patched code to use, but in a couple of critical places in entry.S we can't: we make the clobbers explicit at the call site, and manually clobber the allowed registers in debug mode as an extra check. And: Don't abuse CONFIG_DEBUG_KERNEL, add CONFIG_DEBUG_PARAVIRT. And: AK: Fix warnings in x86-64 alternative.c build And: AK: Fix compilation with defconfig And: ^From: Andrew Morton Some binutlises still like to emit references to __stop_parainstructions and __start_parainstructions. And: AK: Fix warnings about unused variables when PARAVIRT is disabled. Signed-off-by: Rusty Russell Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Chris Wright Signed-off-by: Zachary Amsden Signed-off-by: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/Kconfig.debug | 10 ++ arch/i386/kernel/alternative.c | 63 ++++++++++--- arch/i386/kernel/entry.S | 39 +++++--- arch/i386/kernel/module.c | 11 ++- arch/i386/kernel/paravirt.c | 44 +++++++++ arch/i386/kernel/vmlinux.lds.S | 6 ++ include/asm-i386/alternative.h | 13 ++- include/asm-i386/desc.h | 41 ++++---- include/asm-i386/irqflags.h | 4 +- include/asm-i386/paravirt.h | 189 ++++++++++++++++++++++++++++++------- include/asm-i386/processor.h | 198 +++++++++++++++++++-------------------- include/asm-i386/spinlock.h | 15 ++- include/asm-x86_64/alternative.h | 12 +++ scripts/mod/modpost.c | 2 + 14 files changed, 459 insertions(+), 188 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/Kconfig.debug b/arch/i386/Kconfig.debug index b31c0802e1c..f68cc6f215f 100644 --- a/arch/i386/Kconfig.debug +++ b/arch/i386/Kconfig.debug @@ -85,4 +85,14 @@ config DOUBLEFAULT option saves about 4k and might cause you much additional grey hair. +config DEBUG_PARAVIRT + bool "Enable some paravirtualization debugging" + default y + depends on PARAVIRT && DEBUG_KERNEL + help + Currently deliberately clobbers regs which are allowed to be + clobbered in inlined paravirt hooks, even in native mode. + If turning this off solves a problem, then DISABLE_INTERRUPTS() or + ENABLE_INTERRUPTS() is lying about what registers can be clobbered. + endmenu diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c index 535f9794fba..9eca21b49f6 100644 --- a/arch/i386/kernel/alternative.c +++ b/arch/i386/kernel/alternative.c @@ -124,6 +124,20 @@ static unsigned char** find_nop_table(void) #endif /* CONFIG_X86_64 */ +static void nop_out(void *insns, unsigned int len) +{ + unsigned char **noptable = find_nop_table(); + + while (len > 0) { + unsigned int noplen = len; + if (noplen > ASM_NOP_MAX) + noplen = ASM_NOP_MAX; + memcpy(insns, noptable[noplen], noplen); + insns += noplen; + len -= noplen; + } +} + extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; extern struct alt_instr __smp_alt_instructions[], __smp_alt_instructions_end[]; extern u8 *__smp_locks[], *__smp_locks_end[]; @@ -138,10 +152,9 @@ extern u8 __smp_alt_begin[], __smp_alt_end[]; void apply_alternatives(struct alt_instr *start, struct alt_instr *end) { - unsigned char **noptable = find_nop_table(); struct alt_instr *a; u8 *instr; - int diff, i, k; + int diff; DPRINTK("%s: alt table %p -> %p\n", __FUNCTION__, start, end); for (a = start; a < end; a++) { @@ -159,13 +172,7 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end) #endif memcpy(instr, a->replacement, a->replacementlen); diff = a->instrlen - a->replacementlen; - /* Pad the rest with nops */ - for (i = a->replacementlen; diff > 0; diff -= k, i += k) { - k = diff; - if (k > ASM_NOP_MAX) - k = ASM_NOP_MAX; - memcpy(a->instr + i, noptable[k], k); - } + nop_out(instr + a->replacementlen, diff); } } @@ -209,7 +216,6 @@ static void alternatives_smp_lock(u8 **start, u8 **end, u8 *text, u8 *text_end) static void alternatives_smp_unlock(u8 **start, u8 **end, u8 *text, u8 *text_end) { - unsigned char **noptable = find_nop_table(); u8 **ptr; for (ptr = start; ptr < end; ptr++) { @@ -217,7 +223,7 @@ static void alternatives_smp_unlock(u8 **start, u8 **end, u8 *text, u8 *text_end continue; if (*ptr > text_end) continue; - **ptr = noptable[1][0]; + nop_out(*ptr, 1); }; } @@ -343,6 +349,40 @@ void alternatives_smp_switch(int smp) #endif +#ifdef CONFIG_PARAVIRT +void apply_paravirt(struct paravirt_patch *start, struct paravirt_patch *end) +{ + struct paravirt_patch *p; + + for (p = start; p < end; p++) { + unsigned int used; + + used = paravirt_ops.patch(p->instrtype, p->clobbers, p->instr, + p->len); +#ifdef CONFIG_DEBUG_PARAVIRT + { + int i; + /* Deliberately clobber regs using "not %reg" to find bugs. */ + for (i = 0; i < 3; i++) { + if (p->len - used >= 2 && (p->clobbers & (1 << i))) { + memcpy(p->instr + used, "\xf7\xd0", 2); + p->instr[used+1] |= i; + used += 2; + } + } + } +#endif + /* Pad the rest with nops */ + nop_out(p->instr + used, p->len - used); + } + + /* Sync to be conservative, in case we patched following instructions */ + sync_core(); +} +extern struct paravirt_patch __start_parainstructions[], + __stop_parainstructions[]; +#endif /* CONFIG_PARAVIRT */ + void __init alternative_instructions(void) { unsigned long flags; @@ -390,5 +430,6 @@ void __init alternative_instructions(void) alternatives_smp_switch(0); } #endif + apply_paravirt(__start_parainstructions, __stop_parainstructions); local_irq_restore(flags); } diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index d274612e05c..de34b7fed3c 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -53,6 +53,19 @@ #include #include "irq_vectors.h" +/* + * We use macros for low-level operations which need to be overridden + * for paravirtualization. The following will never clobber any registers: + * INTERRUPT_RETURN (aka. "iret") + * GET_CR0_INTO_EAX (aka. "movl %cr0, %eax") + * ENABLE_INTERRUPTS_SYSEXIT (aka "sti; sysexit"). + * + * For DISABLE_INTERRUPTS/ENABLE_INTERRUPTS (aka "cli"/"sti"), you must + * specify what registers can be overwritten (CLBR_NONE, CLBR_EAX/EDX/ECX/ANY). + * Allowing a register to be clobbered can shrink the paravirt replacement + * enough to patch inline, increasing performance. + */ + #define nr_syscalls ((syscall_table_size)/4) CF_MASK = 0x00000001 @@ -63,9 +76,9 @@ NT_MASK = 0x00004000 VM_MASK = 0x00020000 #ifdef CONFIG_PREEMPT -#define preempt_stop DISABLE_INTERRUPTS; TRACE_IRQS_OFF +#define preempt_stop(clobbers) DISABLE_INTERRUPTS(clobbers); TRACE_IRQS_OFF #else -#define preempt_stop +#define preempt_stop(clobbers) #define resume_kernel restore_nocheck #endif @@ -226,7 +239,7 @@ ENTRY(ret_from_fork) ALIGN RING0_PTREGS_FRAME ret_from_exception: - preempt_stop + preempt_stop(CLBR_ANY) ret_from_intr: GET_THREAD_INFO(%ebp) check_userspace: @@ -237,7 +250,7 @@ check_userspace: jb resume_kernel # not returning to v8086 or userspace ENTRY(resume_userspace) - DISABLE_INTERRUPTS # make sure we don't miss an interrupt + DISABLE_INTERRUPTS(CLBR_ANY) # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret movl TI_flags(%ebp), %ecx @@ -248,7 +261,7 @@ ENTRY(resume_userspace) #ifdef CONFIG_PREEMPT ENTRY(resume_kernel) - DISABLE_INTERRUPTS + DISABLE_INTERRUPTS(CLBR_ANY) cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ? jnz restore_nocheck need_resched: @@ -277,7 +290,7 @@ sysenter_past_esp: * No need to follow this irqs on/off section: the syscall * disabled irqs and here we enable it straight after entry: */ - ENABLE_INTERRUPTS + ENABLE_INTERRUPTS(CLBR_NONE) pushl $(__USER_DS) CFI_ADJUST_CFA_OFFSET 4 /*CFI_REL_OFFSET ss, 0*/ @@ -322,7 +335,7 @@ sysenter_past_esp: jae syscall_badsys call *sys_call_table(,%eax,4) movl %eax,PT_EAX(%esp) - DISABLE_INTERRUPTS + DISABLE_INTERRUPTS(CLBR_ECX|CLBR_EDX) TRACE_IRQS_OFF movl TI_flags(%ebp), %ecx testw $_TIF_ALLWORK_MASK, %cx @@ -364,7 +377,7 @@ syscall_call: call *sys_call_table(,%eax,4) movl %eax,PT_EAX(%esp) # store the return value syscall_exit: - DISABLE_INTERRUPTS # make sure we don't miss an interrupt + DISABLE_INTERRUPTS(CLBR_ANY) # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret TRACE_IRQS_OFF @@ -393,7 +406,7 @@ restore_nocheck_notrace: .section .fixup,"ax" iret_exc: TRACE_IRQS_ON - ENABLE_INTERRUPTS + ENABLE_INTERRUPTS(CLBR_NONE) pushl $0 # no error code pushl $do_iret_error jmp error_code @@ -436,7 +449,7 @@ ldt_ss: CFI_ADJUST_CFA_OFFSET 4 pushl %eax CFI_ADJUST_CFA_OFFSET 4 - DISABLE_INTERRUPTS + DISABLE_INTERRUPTS(CLBR_EAX) TRACE_IRQS_OFF lss (%esp), %esp CFI_ADJUST_CFA_OFFSET -8 @@ -451,7 +464,7 @@ work_pending: jz work_notifysig work_resched: call schedule - DISABLE_INTERRUPTS # make sure we don't miss an interrupt + DISABLE_INTERRUPTS(CLBR_ANY) # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret TRACE_IRQS_OFF @@ -509,7 +522,7 @@ syscall_exit_work: testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP), %cl jz work_pending TRACE_IRQS_ON - ENABLE_INTERRUPTS # could let do_syscall_trace() call + ENABLE_INTERRUPTS(CLBR_ANY) # could let do_syscall_trace() call # schedule() instead movl %esp, %eax movl $1, %edx @@ -693,7 +706,7 @@ ENTRY(device_not_available) GET_CR0_INTO_EAX testl $0x4, %eax # EM (math emulation bit) jne device_not_available_emulate - preempt_stop + preempt_stop(CLBR_ANY) call math_state_restore jmp ret_from_exception device_not_available_emulate: diff --git a/arch/i386/kernel/module.c b/arch/i386/kernel/module.c index 470cf97e7cd..d7d9c8b23f7 100644 --- a/arch/i386/kernel/module.c +++ b/arch/i386/kernel/module.c @@ -108,7 +108,8 @@ int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, struct module *me) { - const Elf_Shdr *s, *text = NULL, *alt = NULL, *locks = NULL; + const Elf_Shdr *s, *text = NULL, *alt = NULL, *locks = NULL, + *para = NULL; char *secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; for (s = sechdrs; s < sechdrs + hdr->e_shnum; s++) { @@ -118,6 +119,8 @@ int module_finalize(const Elf_Ehdr *hdr, alt = s; if (!strcmp(".smp_locks", secstrings + s->sh_name)) locks= s; + if (!strcmp(".parainstructions", secstrings + s->sh_name)) + para = s; } if (alt) { @@ -132,6 +135,12 @@ int module_finalize(const Elf_Ehdr *hdr, lseg, lseg + locks->sh_size, tseg, tseg + text->sh_size); } + + if (para) { + void *pseg = (void *)para->sh_addr; + apply_paravirt(pseg, pseg + para->sh_size); + } + return 0; } diff --git a/arch/i386/kernel/paravirt.c b/arch/i386/kernel/paravirt.c index 478192cd4b9..d4646042644 100644 --- a/arch/i386/kernel/paravirt.c +++ b/arch/i386/kernel/paravirt.c @@ -45,6 +45,49 @@ char *memory_setup(void) return paravirt_ops.memory_setup(); } +/* Simple instruction patching code. */ +#define DEF_NATIVE(name, code) \ + extern const char start_##name[], end_##name[]; \ + asm("start_" #name ": " code "; end_" #name ":") +DEF_NATIVE(cli, "cli"); +DEF_NATIVE(sti, "sti"); +DEF_NATIVE(popf, "push %eax; popf"); +DEF_NATIVE(pushf, "pushf; pop %eax"); +DEF_NATIVE(pushf_cli, "pushf; pop %eax; cli"); +DEF_NATIVE(iret, "iret"); +DEF_NATIVE(sti_sysexit, "sti; sysexit"); + +static const struct native_insns +{ + const char *start, *end; +} native_insns[] = { + [PARAVIRT_IRQ_DISABLE] = { start_cli, end_cli }, + [PARAVIRT_IRQ_ENABLE] = { start_sti, end_sti }, + [PARAVIRT_RESTORE_FLAGS] = { start_popf, end_popf }, + [PARAVIRT_SAVE_FLAGS] = { start_pushf, end_pushf }, + [PARAVIRT_SAVE_FLAGS_IRQ_DISABLE] = { start_pushf_cli, end_pushf_cli }, + [PARAVIRT_INTERRUPT_RETURN] = { start_iret, end_iret }, + [PARAVIRT_STI_SYSEXIT] = { start_sti_sysexit, end_sti_sysexit }, +}; + +static unsigned native_patch(u8 type, u16 clobbers, void *insns, unsigned len) +{ + unsigned int insn_len; + + /* Don't touch it if we don't have a replacement */ + if (type >= ARRAY_SIZE(native_insns) || !native_insns[type].start) + return len; + + insn_len = native_insns[type].end - native_insns[type].start; + + /* Similarly if we can't fit replacement. */ + if (len < insn_len) + return len; + + memcpy(insns, native_insns[type].start, insn_len); + return insn_len; +} + static fastcall unsigned long native_get_debugreg(int regno) { unsigned long val = 0; /* Damn you, gcc! */ @@ -349,6 +392,7 @@ struct paravirt_ops paravirt_ops = { .paravirt_enabled = 0, .kernel_rpl = 0, + .patch = native_patch, .banner = default_banner, .arch_setup = native_nop, .memory_setup = machine_specific_memory_setup, diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 6860f20aa57..5c69cf0e594 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -165,6 +165,12 @@ SECTIONS .altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) { *(.altinstr_replacement) } + . = ALIGN(4); + __start_parainstructions = .; + .parainstructions : AT(ADDR(.parainstructions) - LOAD_OFFSET) { + *(.parainstructions) + } + __stop_parainstructions = .; /* .exit.text is discard at runtime, not link time, to deal with references from .altinstructions and .eh_frame */ .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { *(.exit.text) } diff --git a/include/asm-i386/alternative.h b/include/asm-i386/alternative.h index b01a7ec409c..b8fa9557c53 100644 --- a/include/asm-i386/alternative.h +++ b/include/asm-i386/alternative.h @@ -4,7 +4,7 @@ #ifdef __KERNEL__ #include - +#include #include struct alt_instr { @@ -118,4 +118,15 @@ static inline void alternatives_smp_switch(int smp) {} #define LOCK_PREFIX "" #endif +struct paravirt_patch; +#ifdef CONFIG_PARAVIRT +void apply_paravirt(struct paravirt_patch *start, struct paravirt_patch *end); +#else +static inline void +apply_paravirt(struct paravirt_patch *start, struct paravirt_patch *end) +{} +#define __start_parainstructions NULL +#define __stop_parainstructions NULL +#endif + #endif /* _I386_ALTERNATIVE_H */ diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h index f19820f0834..f398cc45644 100644 --- a/include/asm-i386/desc.h +++ b/include/asm-i386/desc.h @@ -81,31 +81,15 @@ static inline void load_TLS(struct thread_struct *t, unsigned int cpu) #undef C } -static inline void write_dt_entry(void *dt, int entry, __u32 entry_a, __u32 entry_b) -{ - __u32 *lp = (__u32 *)((char *)dt + entry*8); - *lp = entry_a; - *(lp+1) = entry_b; -} - #define write_ldt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b) #define write_gdt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b) #define write_idt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b) -static inline void _set_gate(int gate, unsigned int type, void *addr, unsigned short seg) -{ - __u32 a, b; - pack_gate(&a, &b, (unsigned long)addr, seg, type, 0); - write_idt_entry(idt_table, gate, a, b); -} - -static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const void *addr) +static inline void write_dt_entry(void *dt, int entry, __u32 entry_a, __u32 entry_b) { - __u32 a, b; - pack_descriptor(&a, &b, (unsigned long)addr, - offsetof(struct tss_struct, __cacheline_filler) - 1, - DESCTYPE_TSS, 0); - write_gdt_entry(get_cpu_gdt_table(cpu), entry, a, b); + __u32 *lp = (__u32 *)((char *)dt + entry*8); + *lp = entry_a; + *(lp+1) = entry_b; } #define set_ldt native_set_ldt @@ -128,6 +112,23 @@ static inline fastcall void native_set_ldt(const void *addr, } } +static inline void _set_gate(int gate, unsigned int type, void *addr, unsigned short seg) +{ + __u32 a, b; + pack_gate(&a, &b, (unsigned long)addr, seg, type, 0); + write_idt_entry(idt_table, gate, a, b); +} + +static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const void *addr) +{ + __u32 a, b; + pack_descriptor(&a, &b, (unsigned long)addr, + offsetof(struct tss_struct, __cacheline_filler) - 1, + DESCTYPE_TSS, 0); + write_gdt_entry(get_cpu_gdt_table(cpu), entry, a, b); +} + + #define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr) #define LDT_entry_a(info) \ diff --git a/include/asm-i386/irqflags.h b/include/asm-i386/irqflags.h index 9ce01f3fb7b..17b18cf4fe9 100644 --- a/include/asm-i386/irqflags.h +++ b/include/asm-i386/irqflags.h @@ -79,8 +79,8 @@ static inline unsigned long __raw_local_irq_save(void) } #else -#define DISABLE_INTERRUPTS cli -#define ENABLE_INTERRUPTS sti +#define DISABLE_INTERRUPTS(clobbers) cli +#define ENABLE_INTERRUPTS(clobbers) sti #define ENABLE_INTERRUPTS_SYSEXIT sti; sysexit #define INTERRUPT_RETURN iret #define GET_CR0_INTO_EAX movl %cr0, %eax diff --git a/include/asm-i386/paravirt.h b/include/asm-i386/paravirt.h index a7551a44686..081194751ad 100644 --- a/include/asm-i386/paravirt.h +++ b/include/asm-i386/paravirt.h @@ -3,8 +3,26 @@ /* Various instructions on x86 need to be replaced for * para-virtualization: those hooks are defined here. */ #include +#include #ifdef CONFIG_PARAVIRT +/* These are the most performance critical ops, so we want to be able to patch + * callers */ +#define PARAVIRT_IRQ_DISABLE 0 +#define PARAVIRT_IRQ_ENABLE 1 +#define PARAVIRT_RESTORE_FLAGS 2 +#define PARAVIRT_SAVE_FLAGS 3 +#define PARAVIRT_SAVE_FLAGS_IRQ_DISABLE 4 +#define PARAVIRT_INTERRUPT_RETURN 5 +#define PARAVIRT_STI_SYSEXIT 6 + +/* Bitmask of what can be clobbered: usually at least eax. */ +#define CLBR_NONE 0x0 +#define CLBR_EAX 0x1 +#define CLBR_ECX 0x2 +#define CLBR_EDX 0x4 +#define CLBR_ANY 0x7 + #ifndef __ASSEMBLY__ struct thread_struct; struct Xgt_desc_struct; @@ -15,6 +33,15 @@ struct paravirt_ops int paravirt_enabled; const char *name; + /* + * Patch may replace one of the defined code sequences with arbitrary + * code, subject to the same register constraints. This generally + * means the code is not free to clobber any registers other than EAX. + * The patch function should return the number of bytes of code + * generated, as we nop pad the rest in generic code. + */ + unsigned (*patch)(u8 type, u16 clobber, void *firstinsn, unsigned len); + void (*arch_setup)(void); char *(*memory_setup)(void); void (*init_IRQ)(void); @@ -147,35 +174,6 @@ static inline void __cpuid(unsigned int *eax, unsigned int *ebx, #define read_cr4_safe(x) paravirt_ops.read_cr4_safe() #define write_cr4(x) paravirt_ops.write_cr4(x) -static inline unsigned long __raw_local_save_flags(void) -{ - return paravirt_ops.save_fl(); -} - -static inline void raw_local_irq_restore(unsigned long flags) -{ - return paravirt_ops.restore_fl(flags); -} - -static inline void raw_local_irq_disable(void) -{ - paravirt_ops.irq_disable(); -} - -static inline void raw_local_irq_enable(void) -{ - paravirt_ops.irq_enable(); -} - -static inline unsigned long __raw_local_irq_save(void) -{ - unsigned long flags = paravirt_ops.save_fl(); - - paravirt_ops.irq_disable(); - - return flags; -} - static inline void raw_safe_halt(void) { paravirt_ops.safe_halt(); @@ -267,15 +265,134 @@ static inline void slow_down_io(void) { #endif } -#define CLI_STRING "pushl %eax; pushl %ecx; pushl %edx; call *paravirt_ops+PARAVIRT_irq_disable; popl %edx; popl %ecx; popl %eax" -#define STI_STRING "pushl %eax; pushl %ecx; pushl %edx; call *paravirt_ops+PARAVIRT_irq_enable; popl %edx; popl %ecx; popl %eax" +/* These all sit in the .parainstructions section to tell us what to patch. */ +struct paravirt_patch { + u8 *instr; /* original instructions */ + u8 instrtype; /* type of this instruction */ + u8 len; /* length of original instruction */ + u16 clobbers; /* what registers you may clobber */ +}; + +#define paravirt_alt(insn_string, typenum, clobber) \ + "771:\n\t" insn_string "\n" "772:\n" \ + ".pushsection .parainstructions,\"a\"\n" \ + " .long 771b\n" \ + " .byte " __stringify(typenum) "\n" \ + " .byte 772b-771b\n" \ + " .short " __stringify(clobber) "\n" \ + ".popsection" + +static inline unsigned long __raw_local_save_flags(void) +{ + unsigned long f; + + __asm__ __volatile__(paravirt_alt( "pushl %%ecx; pushl %%edx;" + "call *%1;" + "popl %%edx; popl %%ecx", + PARAVIRT_SAVE_FLAGS, CLBR_NONE) + : "=a"(f): "m"(paravirt_ops.save_fl) + : "memory", "cc"); + return f; +} + +static inline void raw_local_irq_restore(unsigned long f) +{ + __asm__ __volatile__(paravirt_alt( "pushl %%ecx; pushl %%edx;" + "call *%1;" + "popl %%edx; popl %%ecx", + PARAVIRT_RESTORE_FLAGS, CLBR_EAX) + : "=a"(f) : "m" (paravirt_ops.restore_fl), "0"(f) + : "memory", "cc"); +} + +static inline void raw_local_irq_disable(void) +{ + __asm__ __volatile__(paravirt_alt( "pushl %%ecx; pushl %%edx;" + "call *%0;" + "popl %%edx; popl %%ecx", + PARAVIRT_IRQ_DISABLE, CLBR_EAX) + : : "m" (paravirt_ops.irq_disable) + : "memory", "eax", "cc"); +} + +static inline void raw_local_irq_enable(void) +{ + __asm__ __volatile__(paravirt_alt( "pushl %%ecx; pushl %%edx;" + "call *%0;" + "popl %%edx; popl %%ecx", + PARAVIRT_IRQ_ENABLE, CLBR_EAX) + : : "m" (paravirt_ops.irq_enable) + : "memory", "eax", "cc"); +} + +static inline unsigned long __raw_local_irq_save(void) +{ + unsigned long f; + + __asm__ __volatile__(paravirt_alt( "pushl %%ecx; pushl %%edx;" + "call *%1; pushl %%eax;" + "call *%2; popl %%eax;" + "popl %%edx; popl %%ecx", + PARAVIRT_SAVE_FLAGS_IRQ_DISABLE, + CLBR_NONE) + : "=a"(f) + : "m" (paravirt_ops.save_fl), + "m" (paravirt_ops.irq_disable) + : "memory", "cc"); + return f; +} + +#define CLI_STRING paravirt_alt("pushl %%ecx; pushl %%edx;" \ + "call *paravirt_ops+%c[irq_disable];" \ + "popl %%edx; popl %%ecx", \ + PARAVIRT_IRQ_DISABLE, CLBR_EAX) + +#define STI_STRING paravirt_alt("pushl %%ecx; pushl %%edx;" \ + "call *paravirt_ops+%c[irq_enable];" \ + "popl %%edx; popl %%ecx", \ + PARAVIRT_IRQ_ENABLE, CLBR_EAX) +#define CLI_STI_CLOBBERS , "%eax" +#define CLI_STI_INPUT_ARGS \ + , \ + [irq_disable] "i" (offsetof(struct paravirt_ops, irq_disable)), \ + [irq_enable] "i" (offsetof(struct paravirt_ops, irq_enable)) + #else /* __ASSEMBLY__ */ -#define INTERRUPT_RETURN jmp *%cs:paravirt_ops+PARAVIRT_iret -#define DISABLE_INTERRUPTS pushl %eax; pushl %ecx; pushl %edx; call *paravirt_ops+PARAVIRT_irq_disable; popl %edx; popl %ecx; popl %eax -#define ENABLE_INTERRUPTS pushl %eax; pushl %ecx; pushl %edx; call *%cs:paravirt_ops+PARAVIRT_irq_enable; popl %edx; popl %ecx; popl %eax -#define ENABLE_INTERRUPTS_SYSEXIT jmp *%cs:paravirt_ops+PARAVIRT_irq_enable_sysexit -#define GET_CR0_INTO_EAX call *paravirt_ops+PARAVIRT_read_cr0 +#define PARA_PATCH(ptype, clobbers, ops) \ +771:; \ + ops; \ +772:; \ + .pushsection .parainstructions,"a"; \ + .long 771b; \ + .byte ptype; \ + .byte 772b-771b; \ + .short clobbers; \ + .popsection + +#define INTERRUPT_RETURN \ + PARA_PATCH(PARAVIRT_INTERRUPT_RETURN, CLBR_ANY, \ + jmp *%cs:paravirt_ops+PARAVIRT_iret) + +#define DISABLE_INTERRUPTS(clobbers) \ + PARA_PATCH(PARAVIRT_IRQ_DISABLE, clobbers, \ + pushl %ecx; pushl %edx; \ + call *paravirt_ops+PARAVIRT_irq_disable; \ + popl %edx; popl %ecx) \ + +#define ENABLE_INTERRUPTS(clobbers) \ + PARA_PATCH(PARAVIRT_IRQ_ENABLE, clobbers, \ + pushl %ecx; pushl %edx; \ + call *%cs:paravirt_ops+PARAVIRT_irq_enable; \ + popl %edx; popl %ecx) + +#define ENABLE_INTERRUPTS_SYSEXIT \ + PARA_PATCH(PARAVIRT_STI_SYSEXIT, CLBR_ANY, \ + jmp *%cs:paravirt_ops+PARAVIRT_irq_enable_sysexit) + +#define GET_CR0_INTO_EAX \ + call *paravirt_ops+PARAVIRT_read_cr0 + #endif /* __ASSEMBLY__ */ #endif /* CONFIG_PARAVIRT */ #endif /* __ASM_PARAVIRT_H */ diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h index 6c2c4457be0..5f0418d0078 100644 --- a/include/asm-i386/processor.h +++ b/include/asm-i386/processor.h @@ -156,59 +156,6 @@ static inline fastcall void native_cpuid(unsigned int *eax, unsigned int *ebx, : "0" (*eax), "2" (*ecx)); } -/* - * Generic CPUID function - * clear %ecx since some cpus (Cyrix MII) do not set or clear %ecx - * resulting in stale register contents being returned. - */ -static inline void cpuid(unsigned int op, unsigned int *eax, unsigned int *ebx, unsigned int *ecx, unsigned int *edx) -{ - *eax = op; - *ecx = 0; - __cpuid(eax, ebx, ecx, edx); -} - -/* Some CPUID calls want 'count' to be placed in ecx */ -static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx, - int *edx) -{ - *eax = op; - *ecx = count; - __cpuid(eax, ebx, ecx, edx); -} - -/* - * CPUID functions returning a single datum - */ -static inline unsigned int cpuid_eax(unsigned int op) -{ - unsigned int eax, ebx, ecx, edx; - - cpuid(op, &eax, &ebx, &ecx, &edx); - return eax; -} -static inline unsigned int cpuid_ebx(unsigned int op) -{ - unsigned int eax, ebx, ecx, edx; - - cpuid(op, &eax, &ebx, &ecx, &edx); - return ebx; -} -static inline unsigned int cpuid_ecx(unsigned int op) -{ - unsigned int eax, ebx, ecx, edx; - - cpuid(op, &eax, &ebx, &ecx, &edx); - return ecx; -} -static inline unsigned int cpuid_edx(unsigned int op) -{ - unsigned int eax, ebx, ecx, edx; - - cpuid(op, &eax, &ebx, &ecx, &edx); - return edx; -} - #define load_cr3(pgdir) write_cr3(__pa(pgdir)) /* @@ -491,22 +438,6 @@ struct thread_struct { .io_bitmap = { [ 0 ... IO_BITMAP_LONGS] = ~0 }, \ } -#ifdef CONFIG_PARAVIRT -#include -#else -#define paravirt_enabled() 0 -#define __cpuid native_cpuid - -static inline void load_esp0(struct tss_struct *tss, struct thread_struct *thread) -{ - tss->esp0 = thread->esp0; - /* This can only happen when SEP is enabled, no need to test "SEP"arately */ - if (unlikely(tss->ss1 != thread->sysenter_cs)) { - tss->ss1 = thread->sysenter_cs; - wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0); - } -} - #define start_thread(regs, new_eip, new_esp) do { \ __asm__("movl %0,%%fs": :"r" (0)); \ regs->xgs = 0; \ @@ -519,36 +450,6 @@ static inline void load_esp0(struct tss_struct *tss, struct thread_struct *threa regs->esp = new_esp; \ } while (0) -/* - * These special macros can be used to get or set a debugging register - */ -#define get_debugreg(var, register) \ - __asm__("movl %%db" #register ", %0" \ - :"=r" (var)) -#define set_debugreg(value, register) \ - __asm__("movl %0,%%db" #register \ - : /* no output */ \ - :"r" (value)) - -#define set_iopl_mask native_set_iopl_mask -#endif /* CONFIG_PARAVIRT */ - -/* - * Set IOPL bits in EFLAGS from given mask - */ -static fastcall inline void native_set_iopl_mask(unsigned mask) -{ - unsigned int reg; - __asm__ __volatile__ ("pushfl;" - "popl %0;" - "andl %1, %0;" - "orl %2, %0;" - "pushl %0;" - "popfl" - : "=&r" (reg) - : "i" (~X86_EFLAGS_IOPL), "r" (mask)); -} - /* Forward declaration, a strange C thing */ struct task_struct; struct mm_struct; @@ -640,6 +541,105 @@ static inline void rep_nop(void) #define cpu_relax() rep_nop() +#ifdef CONFIG_PARAVIRT +#include +#else +#define paravirt_enabled() 0 +#define __cpuid native_cpuid + +static inline void load_esp0(struct tss_struct *tss, struct thread_struct *thread) +{ + tss->esp0 = thread->esp0; + /* This can only happen when SEP is enabled, no need to test "SEP"arately */ + if (unlikely(tss->ss1 != thread->sysenter_cs)) { + tss->ss1 = thread->sysenter_cs; + wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0); + } +} + +/* + * These special macros can be used to get or set a debugging register + */ +#define get_debugreg(var, register) \ + __asm__("movl %%db" #register ", %0" \ + :"=r" (var)) +#define set_debugreg(value, register) \ + __asm__("movl %0,%%db" #register \ + : /* no output */ \ + :"r" (value)) + +#define set_iopl_mask native_set_iopl_mask +#endif /* CONFIG_PARAVIRT */ + +/* + * Set IOPL bits in EFLAGS from given mask + */ +static fastcall inline void native_set_iopl_mask(unsigned mask) +{ + unsigned int reg; + __asm__ __volatile__ ("pushfl;" + "popl %0;" + "andl %1, %0;" + "orl %2, %0;" + "pushl %0;" + "popfl" + : "=&r" (reg) + : "i" (~X86_EFLAGS_IOPL), "r" (mask)); +} + +/* + * Generic CPUID function + * clear %ecx since some cpus (Cyrix MII) do not set or clear %ecx + * resulting in stale register contents being returned. + */ +static inline void cpuid(unsigned int op, unsigned int *eax, unsigned int *ebx, unsigned int *ecx, unsigned int *edx) +{ + *eax = op; + *ecx = 0; + __cpuid(eax, ebx, ecx, edx); +} + +/* Some CPUID calls want 'count' to be placed in ecx */ +static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx, + int *edx) +{ + *eax = op; + *ecx = count; + __cpuid(eax, ebx, ecx, edx); +} + +/* + * CPUID functions returning a single datum + */ +static inline unsigned int cpuid_eax(unsigned int op) +{ + unsigned int eax, ebx, ecx, edx; + + cpuid(op, &eax, &ebx, &ecx, &edx); + return eax; +} +static inline unsigned int cpuid_ebx(unsigned int op) +{ + unsigned int eax, ebx, ecx, edx; + + cpuid(op, &eax, &ebx, &ecx, &edx); + return ebx; +} +static inline unsigned int cpuid_ecx(unsigned int op) +{ + unsigned int eax, ebx, ecx, edx; + + cpuid(op, &eax, &ebx, &ecx, &edx); + return ecx; +} +static inline unsigned int cpuid_edx(unsigned int op) +{ + unsigned int eax, ebx, ecx, edx; + + cpuid(op, &eax, &ebx, &ecx, &edx); + return edx; +} + /* generic versions from gas */ #define GENERIC_NOP1 ".byte 0x90\n" #define GENERIC_NOP2 ".byte 0x89,0xf6\n" diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h index dea60709db2..d3bcebed60c 100644 --- a/include/asm-i386/spinlock.h +++ b/include/asm-i386/spinlock.h @@ -12,6 +12,8 @@ #else #define CLI_STRING "cli" #define STI_STRING "sti" +#define CLI_STI_CLOBBERS +#define CLI_STI_INPUT_ARGS #endif /* CONFIG_PARAVIRT */ /* @@ -57,25 +59,28 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long fla { asm volatile( "\n1:\t" - LOCK_PREFIX " ; decb %0\n\t" + LOCK_PREFIX " ; decb %[slock]\n\t" "jns 5f\n" "2:\t" - "testl $0x200, %1\n\t" + "testl $0x200, %[flags]\n\t" "jz 4f\n\t" STI_STRING "\n" "3:\t" "rep;nop\n\t" - "cmpb $0, %0\n\t" + "cmpb $0, %[slock]\n\t" "jle 3b\n\t" CLI_STRING "\n\t" "jmp 1b\n" "4:\t" "rep;nop\n\t" - "cmpb $0, %0\n\t" + "cmpb $0, %[slock]\n\t" "jg 1b\n\t" "jmp 4b\n" "5:\n\t" - : "+m" (lock->slock) : "r" (flags) : "memory"); + : [slock] "+m" (lock->slock) + : [flags] "r" (flags) + CLI_STI_INPUT_ARGS + : "memory" CLI_STI_CLOBBERS); } #endif diff --git a/include/asm-x86_64/alternative.h b/include/asm-x86_64/alternative.h index a584826cc57..a6657b4f3e0 100644 --- a/include/asm-x86_64/alternative.h +++ b/include/asm-x86_64/alternative.h @@ -4,6 +4,7 @@ #ifdef __KERNEL__ #include +#include #include struct alt_instr { @@ -133,4 +134,15 @@ static inline void alternatives_smp_switch(int smp) {} #define LOCK_PREFIX "" #endif +struct paravirt_patch; +#ifdef CONFIG_PARAVIRT +void apply_paravirt(struct paravirt_patch *start, struct paravirt_patch *end); +#else +static inline void +apply_paravirt(struct paravirt_patch *start, struct paravirt_patch *end) +{} +#define __start_parainstructions NULL +#define __stop_parainstructions NULL +#endif + #endif /* _X86_64_ALTERNATIVE_H */ diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c index 2e114162314..ac0a5822299 100644 --- a/scripts/mod/modpost.c +++ b/scripts/mod/modpost.c @@ -911,6 +911,7 @@ static int init_section_ref_ok(const char *name) ".toc1", /* used by ppc64 */ ".stab", ".rodata", + ".parainstructions", ".text.lock", "__bug_table", /* used by powerpc for BUG() */ ".pci_fixup_header", @@ -931,6 +932,7 @@ static int init_section_ref_ok(const char *name) ".altinstructions", ".eh_frame", ".debug", + ".parainstructions", NULL }; /* part of section name */ -- cgit v1.2.3-70-g09d2 From d7cd56111f30259e1b532a12e06f59f8e0a20355 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] i386: cpu_detect extraction Both lhype and Xen want to call the core of the x86 cpu detect code before calling start_kernel. (extracted from larger patch) AK: folded in start_kernel header patch Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Rusty Russell Signed-off-by: Andi Kleen Cc: Jeremy Fitzhardinge Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/cpu/common.c | 37 +++++++++++++++++++++---------------- include/asm-i386/processor.h | 3 +++ include/linux/start_kernel.h | 12 ++++++++++++ init/main.c | 1 + 4 files changed, 37 insertions(+), 16 deletions(-) create mode 100644 include/linux/start_kernel.h (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index cda41aef79a..68bcb687019 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -236,29 +236,14 @@ static int __cpuinit have_cpuid_p(void) return flag_is_changeable_p(X86_EFLAGS_ID); } -/* Do minimum CPU detection early. - Fields really needed: vendor, cpuid_level, family, model, mask, cache alignment. - The others are not touched to avoid unwanted side effects. - - WARNING: this function is only called on the BP. Don't add code here - that is supposed to run on all CPUs. */ -static void __init early_cpu_detect(void) +void __init cpu_detect(struct cpuinfo_x86 *c) { - struct cpuinfo_x86 *c = &boot_cpu_data; - - c->x86_cache_alignment = 32; - - if (!have_cpuid_p()) - return; - /* Get vendor name */ cpuid(0x00000000, &c->cpuid_level, (int *)&c->x86_vendor_id[0], (int *)&c->x86_vendor_id[8], (int *)&c->x86_vendor_id[4]); - get_cpu_vendor(c, 1); - c->x86 = 4; if (c->cpuid_level >= 0x00000001) { u32 junk, tfms, cap0, misc; @@ -275,6 +260,26 @@ static void __init early_cpu_detect(void) } } +/* Do minimum CPU detection early. + Fields really needed: vendor, cpuid_level, family, model, mask, cache alignment. + The others are not touched to avoid unwanted side effects. + + WARNING: this function is only called on the BP. Don't add code here + that is supposed to run on all CPUs. */ +static void __init early_cpu_detect(void) +{ + struct cpuinfo_x86 *c = &boot_cpu_data; + + c->x86_cache_alignment = 32; + + if (!have_cpuid_p()) + return; + + cpu_detect(c); + + get_cpu_vendor(c, 1); +} + static void __cpuinit generic_identify(struct cpuinfo_x86 * c) { u32 tfms, xlvl; diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h index 5f0418d0078..a52d6544042 100644 --- a/include/asm-i386/processor.h +++ b/include/asm-i386/processor.h @@ -20,6 +20,7 @@ #include #include #include +#include /* flag for disabling the tsc */ extern int tsc_disable; @@ -112,6 +113,8 @@ extern struct cpuinfo_x86 cpu_data[]; extern int cpu_llc_id[NR_CPUS]; extern char ignore_fpu_irq; +void __init cpu_detect(struct cpuinfo_x86 *c); + extern void identify_cpu(struct cpuinfo_x86 *); extern void print_cpu_info(struct cpuinfo_x86 *); extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c); diff --git a/include/linux/start_kernel.h b/include/linux/start_kernel.h new file mode 100644 index 00000000000..d3e5f275654 --- /dev/null +++ b/include/linux/start_kernel.h @@ -0,0 +1,12 @@ +#ifndef _LINUX_START_KERNEL_H +#define _LINUX_START_KERNEL_H + +#include +#include + +/* Define the prototype for start_kernel here, rather than cluttering + up something else. */ + +extern asmlinkage void __init start_kernel(void); + +#endif /* _LINUX_START_KERNEL_H */ diff --git a/init/main.c b/init/main.c index 36f608a7cfb..985c9ed8605 100644 --- a/init/main.c +++ b/init/main.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3-70-g09d2 From c9ccf30d77f04064fe5436027ab9d2230c7cdd94 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] paravirt: Add startup infrastructure for paravirtualization 1) Each hypervisor writes a probe function to detect whether we are running under that hypervisor. paravirt_probe() registers this function. 2) If vmlinux is booted with ring != 0, we call all the probe functions (with registers except %esp intact) in link order: the winner will not return. Signed-off-by: Rusty Russell Signed-off-by: Chris Wright Signed-off-by: Andi Kleen Cc: Jeremy Fitzhardinge Cc: Zachary Amsden Signed-off-by: Andrew Morton --- arch/i386/kernel/Makefile | 2 ++ arch/i386/kernel/head.S | 33 +++++++++++++++++++++++++++++++++ arch/i386/kernel/paravirt.c | 4 ++++ arch/i386/kernel/vmlinux.lds.S | 6 ++++++ include/asm-i386/paravirt.h | 5 +++++ 5 files changed, 50 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile index 40661213604..1e8988e558c 100644 --- a/arch/i386/kernel/Makefile +++ b/arch/i386/kernel/Makefile @@ -39,6 +39,8 @@ obj-$(CONFIG_VM86) += vm86.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o obj-$(CONFIG_HPET_TIMER) += hpet.o obj-$(CONFIG_K8_NB) += k8.o + +# Make sure this is linked after any other paravirt_ops structs: see head.S obj-$(CONFIG_PARAVIRT) += paravirt.o EXTRA_AFLAGS := -traditional diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index 5b14e95ac8b..edef5084ce1 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -55,6 +55,12 @@ */ ENTRY(startup_32) +#ifdef CONFIG_PARAVIRT + movl %cs, %eax + testl $0x3, %eax + jnz startup_paravirt +#endif + /* * Set segments to known values. */ @@ -486,6 +492,33 @@ ignore_int: #endif iret +#ifdef CONFIG_PARAVIRT +startup_paravirt: + cld + movl $(init_thread_union+THREAD_SIZE),%esp + + /* We take pains to preserve all the regs. */ + pushl %edx + pushl %ecx + pushl %eax + + /* paravirt.o is last in link, and that probe fn never returns */ + pushl $__start_paravirtprobe +1: + movl 0(%esp), %eax + pushl (%eax) + movl 8(%esp), %eax + call *(%esp) + popl %eax + + movl 4(%esp), %eax + movl 8(%esp), %ecx + movl 12(%esp), %edx + + addl $4, (%esp) + jmp 1b +#endif + /* * Real beginning of normal "text" segment */ diff --git a/arch/i386/kernel/paravirt.c b/arch/i386/kernel/paravirt.c index d4646042644..5a9bd3250a1 100644 --- a/arch/i386/kernel/paravirt.c +++ b/arch/i386/kernel/paravirt.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -387,6 +388,9 @@ static int __init print_banner(void) } core_initcall(print_banner); +/* We simply declare start_kernel to be the paravirt probe of last resort. */ +paravirt_probe(start_kernel); + struct paravirt_ops paravirt_ops = { .name = "bare hardware", .paravirt_enabled = 0, diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 5c69cf0e594..877dc5cfe3a 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -65,6 +65,12 @@ SECTIONS CONSTRUCTORS } :data + __start_paravirtprobe = .; + .paravirtprobe : AT(ADDR(.paravirtprobe) - LOAD_OFFSET) { + *(.paravirtprobe) + } + __stop_paravirtprobe = .; + . = ALIGN(4096); .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { __nosave_begin = .; diff --git a/include/asm-i386/paravirt.h b/include/asm-i386/paravirt.h index 081194751ad..dd707d8c827 100644 --- a/include/asm-i386/paravirt.h +++ b/include/asm-i386/paravirt.h @@ -120,6 +120,11 @@ struct paravirt_ops void (fastcall *iret)(void); }; +/* Mark a paravirt probe function. */ +#define paravirt_probe(fn) \ + static asmlinkage void (*__paravirtprobe_##fn)(void) __attribute_used__ \ + __attribute__((__section__(".paravirtprobe"))) = fn + extern struct paravirt_ops paravirt_ops; #define paravirt_enabled() (paravirt_ops.paravirt_enabled) -- cgit v1.2.3-70-g09d2 From 4f205fd45a5c192907188d6f8f6d7e66db859248 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] paravirt: Allow selected bug checks to be Allow selected bug checks to be skipped by paravirt kernels. The two most important are the F00F workaround (which is either done by the hypervisor, or not required), and the 'hlt' instruction check, which can break under some hypervisors. Signed-off-by: Zachary Amsden Signed-off-by: Chris Wright Signed-off-by: Andi Kleen Cc: Rusty Russell Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton --- arch/i386/kernel/cpu/intel.c | 2 +- include/asm-i386/bugs.h | 4 ++++ 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c index 798c2f617e8..3ae795e9056 100644 --- a/arch/i386/kernel/cpu/intel.c +++ b/arch/i386/kernel/cpu/intel.c @@ -107,7 +107,7 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) * Note that the workaround only should be initialized once... */ c->f00f_bug = 0; - if ( c->x86 == 5 ) { + if (!paravirt_enabled() && c->x86 == 5) { static int f00f_workaround_enabled = 0; c->f00f_bug = 1; diff --git a/include/asm-i386/bugs.h b/include/asm-i386/bugs.h index 592ffeeda45..38f1aebbbdb 100644 --- a/include/asm-i386/bugs.h +++ b/include/asm-i386/bugs.h @@ -21,6 +21,7 @@ #include #include #include +#include static int __init no_halt(char *s) { @@ -91,6 +92,9 @@ static void __init check_fpu(void) static void __init check_hlt(void) { + if (paravirt_enabled()) + return; + printk(KERN_INFO "Checking 'hlt' instruction... "); if (!boot_cpu_data.hlt_works_ok) { printk("disabled\n"); -- cgit v1.2.3-70-g09d2 From 3bbf54725467d604698721384d858b5983b87e8f Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] paravirt: Disable vdso by default when CONFIG_PARAVIRT is enabled They don't work together and this way even glibc still works. Signed-off-by: Andi Kleen --- arch/i386/kernel/sysenter.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c index 713ba39d32c..92849c7def5 100644 --- a/arch/i386/kernel/sysenter.c +++ b/arch/i386/kernel/sysenter.c @@ -27,7 +27,11 @@ * Should the kernel map a VDSO page into processes and pass its * address down to glibc upon exec()? */ +#ifdef CONFIG_PARAVIRT +unsigned int __read_mostly vdso_enabled = 0; +#else unsigned int __read_mostly vdso_enabled = 1; +#endif EXPORT_SYMBOL_GPL(vdso_enabled); -- cgit v1.2.3-70-g09d2 From 6020c8f315709a508b027ef6749e85b125190947 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] paravirt: Allow disable power management under hypervisor Two legacy power management modes are much easier to just explicitly disable when running in paravirtualized mode - neither APM nor PnP is still relevant. The status of ACPI is still debatable, and noacpi is still a common enough boot parameter that it is not necessary to explicitly disable ACPI. Signed-off-by: Zachary Amsden Signed-off-by: Chris Wright Signed-off-by: Andi Kleen Cc: Rusty Russell Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton --- arch/i386/kernel/apm.c | 3 ++- drivers/pnp/pnpbios/core.c | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c index a60358fe9a4..a97847da9ed 100644 --- a/arch/i386/kernel/apm.c +++ b/arch/i386/kernel/apm.c @@ -231,6 +231,7 @@ #include #include #include +#include #include "io_ports.h" @@ -2235,7 +2236,7 @@ static int __init apm_init(void) dmi_check_system(apm_dmi_table); - if (apm_info.bios.version == 0) { + if (apm_info.bios.version == 0 || paravirt_enabled()) { printk(KERN_INFO "apm: BIOS not found.\n"); return -ENODEV; } diff --git a/drivers/pnp/pnpbios/core.c b/drivers/pnp/pnpbios/core.c index 81a6c83d89a..80066ad792f 100644 --- a/drivers/pnp/pnpbios/core.c +++ b/drivers/pnp/pnpbios/core.c @@ -530,7 +530,8 @@ static int __init pnpbios_init(void) if (check_legacy_ioport(PNPBIOS_BASE)) return -ENODEV; #endif - if (pnpbios_disabled || dmi_check_system(pnpbios_dmi_table)) { + if (pnpbios_disabled || dmi_check_system(pnpbios_dmi_table) || + paravirt_enabled()) { printk(KERN_INFO "PnPBIOS: Disabled\n"); return -ENODEV; } -- cgit v1.2.3-70-g09d2 From 13623d79309dd82e1964458fa017979d16f33fa8 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] paravirt: Add APIC accessors to paravirt-ops. Add APIC accessors to paravirt-ops. Unfortunately, we need two write functions, as some older broken hardware requires workarounds for Pentium APIC errata - this is the purpose of apic_write_atomic. AK: replaced __inline with inline Signed-off-by: Zachary Amsden Signed-off-by: Chris Wright Signed-off-by: Andi Kleen Cc: Rusty Russell Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton --- arch/i386/kernel/paravirt.c | 8 ++++++++ include/asm-i386/apic.h | 15 ++++++++++++--- include/asm-i386/paravirt.h | 27 +++++++++++++++++++++++++++ 3 files changed, 47 insertions(+), 3 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/paravirt.c b/arch/i386/kernel/paravirt.c index 5a9bd3250a1..fe82eb3adf4 100644 --- a/arch/i386/kernel/paravirt.c +++ b/arch/i386/kernel/paravirt.c @@ -29,6 +29,8 @@ #include #include #include +#include +#include /* nop stub */ static void native_nop(void) @@ -446,6 +448,12 @@ struct paravirt_ops paravirt_ops = { .io_delay = native_io_delay, .const_udelay = __const_udelay, +#ifdef CONFIG_X86_LOCAL_APIC + .apic_write = native_apic_write, + .apic_write_atomic = native_apic_write_atomic, + .apic_read = native_apic_read, +#endif + .irq_enable_sysexit = native_irq_enable_sysexit, .iret = native_iret, }; diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h index b9529578fc3..41a44319905 100644 --- a/include/asm-i386/apic.h +++ b/include/asm-i386/apic.h @@ -37,18 +37,27 @@ extern void generic_apic_probe(void); /* * Basic functions accessing APICs. */ +#ifdef CONFIG_PARAVIRT +#include +#else +#define apic_write native_apic_write +#define apic_write_atomic native_apic_write_atomic +#define apic_read native_apic_read +#endif -static __inline void apic_write(unsigned long reg, unsigned long v) +static __inline fastcall void native_apic_write(unsigned long reg, + unsigned long v) { *((volatile unsigned long *)(APIC_BASE+reg)) = v; } -static __inline void apic_write_atomic(unsigned long reg, unsigned long v) +static __inline fastcall void native_apic_write_atomic(unsigned long reg, + unsigned long v) { xchg((volatile unsigned long *)(APIC_BASE+reg), v); } -static __inline unsigned long apic_read(unsigned long reg) +static __inline fastcall unsigned long native_apic_read(unsigned long reg) { return *((volatile unsigned long *)(APIC_BASE+reg)); } diff --git a/include/asm-i386/paravirt.h b/include/asm-i386/paravirt.h index dd707d8c827..e2c803fadb1 100644 --- a/include/asm-i386/paravirt.h +++ b/include/asm-i386/paravirt.h @@ -115,6 +115,12 @@ struct paravirt_ops void (fastcall *io_delay)(void); void (*const_udelay)(unsigned long loops); +#ifdef CONFIG_X86_LOCAL_APIC + void (fastcall *apic_write)(unsigned long reg, unsigned long v); + void (fastcall *apic_write_atomic)(unsigned long reg, unsigned long v); + unsigned long (fastcall *apic_read)(unsigned long reg); +#endif + /* These two are jmp to, not actually called. */ void (fastcall *irq_enable_sysexit)(void); void (fastcall *iret)(void); @@ -270,6 +276,27 @@ static inline void slow_down_io(void) { #endif } +#ifdef CONFIG_X86_LOCAL_APIC +/* + * Basic functions accessing APICs. + */ +static inline void apic_write(unsigned long reg, unsigned long v) +{ + paravirt_ops.apic_write(reg,v); +} + +static inline void apic_write_atomic(unsigned long reg, unsigned long v) +{ + paravirt_ops.apic_write_atomic(reg,v); +} + +static inline unsigned long apic_read(unsigned long reg) +{ + return paravirt_ops.apic_read(reg); +} +#endif + + /* These all sit in the .parainstructions section to tell us what to patch. */ struct paravirt_patch { u8 *instr; /* original instructions */ -- cgit v1.2.3-70-g09d2 From da181a8b3916aa7f2e3c5775d2bd2fe3454cf82d Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] paravirt: Add MMU virtualization to paravirt_ops Add the three bare TLB accessor functions to paravirt-ops. Most amusingly, flush_tlb is redefined on SMP, so I can't call the paravirt op flush_tlb. Instead, I chose to indicate the actual flush type, kernel (global) vs. user (non-global). Global in this sense means using the global bit in the page table entry, which makes TLB entries persistent across CR3 reloads, not global as in the SMP sense of invoking remote shootdowns, so the term is confusingly overloaded. AK: folded in fix from Zach for PAE compilation Signed-off-by: Zachary Amsden Signed-off-by: Chris Wright Signed-off-by: Andi Kleen Cc: Rusty Russell Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton --- arch/i386/kernel/paravirt.c | 109 ++++++++++++++++++++++++++++++++++++++ arch/i386/mm/boot_ioremap.c | 1 + include/asm-i386/paravirt.h | 75 ++++++++++++++++++++++++++ include/asm-i386/pgtable-2level.h | 5 +- include/asm-i386/pgtable-3level.h | 40 +++++++------- include/asm-i386/pgtable.h | 4 +- include/asm-i386/tlbflush.h | 18 +++++-- 7 files changed, 226 insertions(+), 26 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/paravirt.c b/arch/i386/kernel/paravirt.c index fe82eb3adf4..3dceab5828f 100644 --- a/arch/i386/kernel/paravirt.c +++ b/arch/i386/kernel/paravirt.c @@ -31,6 +31,7 @@ #include #include #include +#include /* nop stub */ static void native_nop(void) @@ -379,6 +380,97 @@ static fastcall void native_io_delay(void) asm volatile("outb %al,$0x80"); } +static fastcall void native_flush_tlb(void) +{ + __native_flush_tlb(); +} + +/* + * Global pages have to be flushed a bit differently. Not a real + * performance problem because this does not happen often. + */ +static fastcall void native_flush_tlb_global(void) +{ + __native_flush_tlb_global(); +} + +static fastcall void native_flush_tlb_single(u32 addr) +{ + __native_flush_tlb_single(addr); +} + +#ifndef CONFIG_X86_PAE +static fastcall void native_set_pte(pte_t *ptep, pte_t pteval) +{ + *ptep = pteval; +} + +static fastcall void native_set_pte_at(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pteval) +{ + *ptep = pteval; +} + +static fastcall void native_set_pmd(pmd_t *pmdp, pmd_t pmdval) +{ + *pmdp = pmdval; +} + +#else /* CONFIG_X86_PAE */ + +static fastcall void native_set_pte(pte_t *ptep, pte_t pte) +{ + ptep->pte_high = pte.pte_high; + smp_wmb(); + ptep->pte_low = pte.pte_low; +} + +static fastcall void native_set_pte_at(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pte) +{ + ptep->pte_high = pte.pte_high; + smp_wmb(); + ptep->pte_low = pte.pte_low; +} + +static fastcall void native_set_pte_present(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte) +{ + ptep->pte_low = 0; + smp_wmb(); + ptep->pte_high = pte.pte_high; + smp_wmb(); + ptep->pte_low = pte.pte_low; +} + +static fastcall void native_set_pte_atomic(pte_t *ptep, pte_t pteval) +{ + set_64bit((unsigned long long *)ptep,pte_val(pteval)); +} + +static fastcall void native_set_pmd(pmd_t *pmdp, pmd_t pmdval) +{ + set_64bit((unsigned long long *)pmdp,pmd_val(pmdval)); +} + +static fastcall void native_set_pud(pud_t *pudp, pud_t pudval) +{ + *pudp = pudval; +} + +static fastcall void native_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) +{ + ptep->pte_low = 0; + smp_wmb(); + ptep->pte_high = 0; +} + +static fastcall void native_pmd_clear(pmd_t *pmd) +{ + u32 *tmp = (u32 *)pmd; + *tmp = 0; + smp_wmb(); + *(tmp + 1) = 0; +} +#endif /* CONFIG_X86_PAE */ + /* These are in entry.S */ extern fastcall void native_iret(void); extern fastcall void native_irq_enable_sysexit(void); @@ -454,6 +546,23 @@ struct paravirt_ops paravirt_ops = { .apic_read = native_apic_read, #endif + .flush_tlb_user = native_flush_tlb, + .flush_tlb_kernel = native_flush_tlb_global, + .flush_tlb_single = native_flush_tlb_single, + + .set_pte = native_set_pte, + .set_pte_at = native_set_pte_at, + .set_pmd = native_set_pmd, + .pte_update = (void *)native_nop, + .pte_update_defer = (void *)native_nop, +#ifdef CONFIG_X86_PAE + .set_pte_atomic = native_set_pte_atomic, + .set_pte_present = native_set_pte_present, + .set_pud = native_set_pud, + .pte_clear = native_pte_clear, + .pmd_clear = native_pmd_clear, +#endif + .irq_enable_sysexit = native_irq_enable_sysexit, .iret = native_iret, }; diff --git a/arch/i386/mm/boot_ioremap.c b/arch/i386/mm/boot_ioremap.c index 4de11f508c3..4de95a17a7d 100644 --- a/arch/i386/mm/boot_ioremap.c +++ b/arch/i386/mm/boot_ioremap.c @@ -16,6 +16,7 @@ */ #undef CONFIG_X86_PAE +#undef CONFIG_PARAVIRT #include #include #include diff --git a/include/asm-i386/paravirt.h b/include/asm-i386/paravirt.h index e2c803fadb1..9f06265065f 100644 --- a/include/asm-i386/paravirt.h +++ b/include/asm-i386/paravirt.h @@ -4,6 +4,7 @@ * para-virtualization: those hooks are defined here. */ #include #include +#include #ifdef CONFIG_PARAVIRT /* These are the most performance critical ops, so we want to be able to patch @@ -27,6 +28,7 @@ struct thread_struct; struct Xgt_desc_struct; struct tss_struct; +struct mm_struct; struct paravirt_ops { unsigned int kernel_rpl; @@ -121,6 +123,23 @@ struct paravirt_ops unsigned long (fastcall *apic_read)(unsigned long reg); #endif + void (fastcall *flush_tlb_user)(void); + void (fastcall *flush_tlb_kernel)(void); + void (fastcall *flush_tlb_single)(u32 addr); + + void (fastcall *set_pte)(pte_t *ptep, pte_t pteval); + void (fastcall *set_pte_at)(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pteval); + void (fastcall *set_pmd)(pmd_t *pmdp, pmd_t pmdval); + void (fastcall *pte_update)(struct mm_struct *mm, u32 addr, pte_t *ptep); + void (fastcall *pte_update_defer)(struct mm_struct *mm, u32 addr, pte_t *ptep); +#ifdef CONFIG_X86_PAE + void (fastcall *set_pte_atomic)(pte_t *ptep, pte_t pteval); + void (fastcall *set_pte_present)(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte); + void (fastcall *set_pud)(pud_t *pudp, pud_t pudval); + void (fastcall *pte_clear)(struct mm_struct *mm, unsigned long addr, pte_t *ptep); + void (fastcall *pmd_clear)(pmd_t *pmdp); +#endif + /* These two are jmp to, not actually called. */ void (fastcall *irq_enable_sysexit)(void); void (fastcall *iret)(void); @@ -297,6 +316,62 @@ static inline unsigned long apic_read(unsigned long reg) #endif +#define __flush_tlb() paravirt_ops.flush_tlb_user() +#define __flush_tlb_global() paravirt_ops.flush_tlb_kernel() +#define __flush_tlb_single(addr) paravirt_ops.flush_tlb_single(addr) + +static inline void set_pte(pte_t *ptep, pte_t pteval) +{ + paravirt_ops.set_pte(ptep, pteval); +} + +static inline void set_pte_at(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pteval) +{ + paravirt_ops.set_pte_at(mm, addr, ptep, pteval); +} + +static inline void set_pmd(pmd_t *pmdp, pmd_t pmdval) +{ + paravirt_ops.set_pmd(pmdp, pmdval); +} + +static inline void pte_update(struct mm_struct *mm, u32 addr, pte_t *ptep) +{ + paravirt_ops.pte_update(mm, addr, ptep); +} + +static inline void pte_update_defer(struct mm_struct *mm, u32 addr, pte_t *ptep) +{ + paravirt_ops.pte_update_defer(mm, addr, ptep); +} + +#ifdef CONFIG_X86_PAE +static inline void set_pte_atomic(pte_t *ptep, pte_t pteval) +{ + paravirt_ops.set_pte_atomic(ptep, pteval); +} + +static inline void set_pte_present(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte) +{ + paravirt_ops.set_pte_present(mm, addr, ptep, pte); +} + +static inline void set_pud(pud_t *pudp, pud_t pudval) +{ + paravirt_ops.set_pud(pudp, pudval); +} + +static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) +{ + paravirt_ops.pte_clear(mm, addr, ptep); +} + +static inline void pmd_clear(pmd_t *pmdp) +{ + paravirt_ops.pmd_clear(pmdp); +} +#endif + /* These all sit in the .parainstructions section to tell us what to patch. */ struct paravirt_patch { u8 *instr; /* original instructions */ diff --git a/include/asm-i386/pgtable-2level.h b/include/asm-i386/pgtable-2level.h index 8d8d3b9ecdb..04d6186abc2 100644 --- a/include/asm-i386/pgtable-2level.h +++ b/include/asm-i386/pgtable-2level.h @@ -13,11 +13,14 @@ * within a page table are directly modified. Thus, the following * hook is made available. */ +#ifndef CONFIG_PARAVIRT #define set_pte(pteptr, pteval) (*(pteptr) = pteval) #define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval) +#define set_pmd(pmdptr, pmdval) (*(pmdptr) = (pmdval)) +#endif + #define set_pte_atomic(pteptr, pteval) set_pte(pteptr,pteval) #define set_pte_present(mm,addr,ptep,pteval) set_pte_at(mm,addr,ptep,pteval) -#define set_pmd(pmdptr, pmdval) (*(pmdptr) = (pmdval)) #define pte_clear(mm,addr,xp) do { set_pte_at(mm, addr, xp, __pte(0)); } while (0) #define pmd_clear(xp) do { set_pmd(xp, __pmd(0)); } while (0) diff --git a/include/asm-i386/pgtable-3level.h b/include/asm-i386/pgtable-3level.h index c2d701ea35b..2a6e67db8bc 100644 --- a/include/asm-i386/pgtable-3level.h +++ b/include/asm-i386/pgtable-3level.h @@ -44,6 +44,7 @@ static inline int pte_exec_kernel(pte_t pte) return pte_x(pte); } +#ifndef CONFIG_PARAVIRT /* Rules for using set_pte: the pte being assigned *must* be * either not present or in a state where the hardware will * not attempt to update the pte. In places where this is @@ -80,25 +81,6 @@ static inline void set_pte_present(struct mm_struct *mm, unsigned long addr, pte #define set_pud(pudptr,pudval) \ (*(pudptr) = (pudval)) -/* - * Pentium-II erratum A13: in PAE mode we explicitly have to flush - * the TLB via cr3 if the top-level pgd is changed... - * We do not let the generic code free and clear pgd entries due to - * this erratum. - */ -static inline void pud_clear (pud_t * pud) { } - -#define pud_page(pud) \ -((struct page *) __va(pud_val(pud) & PAGE_MASK)) - -#define pud_page_vaddr(pud) \ -((unsigned long) __va(pud_val(pud) & PAGE_MASK)) - - -/* Find an entry in the second-level page table.. */ -#define pmd_offset(pud, address) ((pmd_t *) pud_page(*(pud)) + \ - pmd_index(address)) - /* * For PTEs and PDEs, we must clear the P-bit first when clearing a page table * entry, so clear the bottom half first and enforce ordering with a compiler @@ -118,6 +100,26 @@ static inline void pmd_clear(pmd_t *pmd) smp_wmb(); *(tmp + 1) = 0; } +#endif + +/* + * Pentium-II erratum A13: in PAE mode we explicitly have to flush + * the TLB via cr3 if the top-level pgd is changed... + * We do not let the generic code free and clear pgd entries due to + * this erratum. + */ +static inline void pud_clear (pud_t * pud) { } + +#define pud_page(pud) \ +((struct page *) __va(pud_val(pud) & PAGE_MASK)) + +#define pud_page_vaddr(pud) \ +((unsigned long) __va(pud_val(pud) & PAGE_MASK)) + + +/* Find an entry in the second-level page table.. */ +#define pmd_offset(pud, address) ((pmd_t *) pud_page(*(pud)) + \ + pmd_index(address)) #define __HAVE_ARCH_PTEP_GET_AND_CLEAR static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h index 7d398f493dd..efd7d90789d 100644 --- a/include/asm-i386/pgtable.h +++ b/include/asm-i386/pgtable.h @@ -15,6 +15,7 @@ #include #include #include +#include #ifndef _I386_BITOPS_H #include @@ -246,6 +247,7 @@ static inline pte_t pte_mkhuge(pte_t pte) { (pte).pte_low |= _PAGE_PSE; return p # include #endif +#ifndef CONFIG_PARAVIRT /* * Rules for using pte_update - it must be called after any PTE update which * has not been done using the set_pte / clear_pte interfaces. It is used by @@ -261,7 +263,7 @@ static inline pte_t pte_mkhuge(pte_t pte) { (pte).pte_low |= _PAGE_PSE; return p */ #define pte_update(mm, addr, ptep) do { } while (0) #define pte_update_defer(mm, addr, ptep) do { } while (0) - +#endif /* * We only update the dirty/accessed state if we set diff --git a/include/asm-i386/tlbflush.h b/include/asm-i386/tlbflush.h index 360648b0f2b..4dd82840d53 100644 --- a/include/asm-i386/tlbflush.h +++ b/include/asm-i386/tlbflush.h @@ -4,7 +4,15 @@ #include #include -#define __flush_tlb() \ +#ifdef CONFIG_PARAVIRT +#include +#else +#define __flush_tlb() __native_flush_tlb() +#define __flush_tlb_global() __native_flush_tlb_global() +#define __flush_tlb_single(addr) __native_flush_tlb_single(addr) +#endif + +#define __native_flush_tlb() \ do { \ unsigned int tmpreg; \ \ @@ -19,7 +27,7 @@ * Global pages have to be flushed a bit differently. Not a real * performance problem because this does not happen often. */ -#define __flush_tlb_global() \ +#define __native_flush_tlb_global() \ do { \ unsigned int tmpreg, cr4, cr4_orig; \ \ @@ -36,6 +44,9 @@ : "memory"); \ } while (0) +#define __native_flush_tlb_single(addr) \ + __asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory") + # define __flush_tlb_all() \ do { \ if (cpu_has_pge) \ @@ -46,9 +57,6 @@ #define cpu_has_invlpg (boot_cpu_data.x86 > 3) -#define __flush_tlb_single(addr) \ - __asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory") - #ifdef CONFIG_X86_INVLPG # define __flush_tlb_one(addr) __flush_tlb_single(addr) #else -- cgit v1.2.3-70-g09d2 From bd472c794bbf6771c3fc1c58f188bc16c393d2fe Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] paravirt: Be careful about touching BIOS address space BIOS ROM areas may not be mapped into the guest address space, so be careful when touching those addresses to make sure they appear to be mapped. [akpm@osdl.org: fix unused var warning] AK: Changed __get_user to probe_kernel_address Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Rusty Russell Signed-off-by: Andi Kleen Cc: Jeremy Fitzhardinge Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/e820.c | 10 +++++++++- arch/i386/pci/pcbios.c | 11 +++++++++-- 2 files changed, 18 insertions(+), 3 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/e820.c b/arch/i386/kernel/e820.c index b755255f272..b704790f796 100644 --- a/arch/i386/kernel/e820.c +++ b/arch/i386/kernel/e820.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include @@ -155,7 +156,14 @@ static struct resource standard_io_resources[] = { { .flags = IORESOURCE_BUSY | IORESOURCE_IO } }; -#define romsignature(x) (*(unsigned short *)(x) == 0xaa55) +static int romsignature(const unsigned char *x) +{ + unsigned short sig; + int ret = 0; + if (probe_kernel_address((const unsigned short *)x, sig) == 0) + ret = (sig == 0xaa55); + return ret; +} static int __init romchecksum(unsigned char *rom, unsigned long length) { diff --git a/arch/i386/pci/pcbios.c b/arch/i386/pci/pcbios.c index ed1512a175a..5f5193401be 100644 --- a/arch/i386/pci/pcbios.c +++ b/arch/i386/pci/pcbios.c @@ -5,6 +5,7 @@ #include #include #include +#include #include "pci.h" #include "pci-functions.h" @@ -314,6 +315,10 @@ static struct pci_raw_ops * __devinit pci_find_bios(void) for (check = (union bios32 *) __va(0xe0000); check <= (union bios32 *) __va(0xffff0); ++check) { + long sig; + if (probe_kernel_address(&check->fields.signature, sig)) + continue; + if (check->fields.signature != BIOS32_SIGNATURE) continue; length = check->fields.length * 16; @@ -331,11 +336,13 @@ static struct pci_raw_ops * __devinit pci_find_bios(void) } DBG("PCI: BIOS32 Service Directory structure at 0x%p\n", check); if (check->fields.entry >= 0x100000) { - printk("PCI: BIOS32 entry (0x%p) in high memory, cannot use.\n", check); + printk("PCI: BIOS32 entry (0x%p) in high memory, " + "cannot use.\n", check); return NULL; } else { unsigned long bios32_entry = check->fields.entry; - DBG("PCI: BIOS32 Service Directory entry at 0x%lx\n", bios32_entry); + DBG("PCI: BIOS32 Service Directory entry at 0x%lx\n", + bios32_entry); bios32_indirect.address = bios32_entry + PAGE_OFFSET; if (check_pcibios()) return &pci_bios_access; -- cgit v1.2.3-70-g09d2 From 8542b200cbe5609edd7aae0c304c091a1c290452 Mon Sep 17 00:00:00 2001 From: Zachary Amsden Date: Thu, 7 Dec 2006 02:14:09 +0100 Subject: [PATCH] paravirt: Add option to allow skipping the timer check Add a way to disable the timer IRQ routing check via a boot option. The VMI timer code uses this to avoid triggering the pester Mingo code, which probes for some very unusual and broken motherboard routings. It fires 100% of the time when using a paravirtual delay mechanism instead of using a realtime delay, since there is no elapsed real time, and the 4 timer IRQs have not yet been delivered. In addition, it is entirely possible, though improbable, that this bug could surface on real hardware which picks a particularly bad time to enter SMM mode, causing a long latency during one of the timer IRQs. While here, make check_timer be __init. Signed-off-by: Zachary Amsden Signed-off-by: Andi Kleen [chrisw: use no_timer_check to bring inline with x86_64 as per Andi's request] Signed-off-by: Chris Wright Cc: Andi Kleen Signed-off-by: Andrew Morton --- Documentation/kernel-parameters.txt | 7 +++++-- arch/i386/kernel/io_apic.c | 16 ++++++++++++++-- 2 files changed, 19 insertions(+), 4 deletions(-) (limited to 'arch/i386') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 2e1898e4e8f..4e90aa427ae 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -599,8 +599,6 @@ and is between 256 and 4096 characters. It is defined in the file hugepages= [HW,IA-32,IA-64] Maximal number of HugeTLB pages. - noirqbalance [IA-32,SMP,KNL] Disable kernel irq balancing - i8042.direct [HW] Put keyboard port into non-translated mode i8042.dumbkbd [HW] Pretend that controller can only read data from keyboard and cannot control its state @@ -1052,9 +1050,14 @@ and is between 256 and 4096 characters. It is defined in the file in certain environments such as networked servers or real-time systems. + noirqbalance [IA-32,SMP,KNL] Disable kernel irq balancing + noirqdebug [IA-32] Disables the code which attempts to detect and disable unhandled interrupt sources. + no_timer_check [IA-32,X86_64,APIC] Disables the code which tests for + broken timer IRQ sources. + noisapnp [ISAPNP] Disables ISA PnP code. noinitrd [RAM] Tells the kernel not to load any configured diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index e33b7a84529..993150f206e 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -1932,6 +1932,15 @@ static void __init setup_ioapic_ids_from_mpc(void) static void __init setup_ioapic_ids_from_mpc(void) { } #endif +static int no_timer_check __initdata; + +static int __init notimercheck(char *s) +{ + no_timer_check = 1; + return 1; +} +__setup("no_timer_check", notimercheck); + /* * There is a nasty bug in some older SMP boards, their mptable lies * about the timer IRQ. We do the following to work around the situation: @@ -1940,10 +1949,13 @@ static void __init setup_ioapic_ids_from_mpc(void) { } * - if this function detects that timer IRQs are defunct, then we fall * back to ISA timer IRQs */ -static int __init timer_irq_works(void) +int __init timer_irq_works(void) { unsigned long t1 = jiffies; + if (no_timer_check) + return 1; + local_irq_enable(); /* Let ten ticks pass... */ mdelay((10 * 1000) / HZ); @@ -2214,7 +2226,7 @@ int timer_uses_ioapic_pin_0; * is so screwy. Thanks to Brian Perkins for testing/hacking this beast * fanatically on his truly buggy board. */ -static inline void check_timer(void) +static inline void __init check_timer(void) { int apic1, pin1, apic2, pin2; int vector; -- cgit v1.2.3-70-g09d2 From c55d92d141b9c40c67db249de91f5c224eb49859 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:09 +0100 Subject: [PATCH] i386: Add support for compilation for Core2 gcc doesn't support -mtune=core2 yet, but will be soon. Use -mtune=generic or -mtune=i686 as fallback TBD need benchmarking for INTEL_USERCOPY etc. So far I used the same defaults as MPENTIUMM Signed-off-by: Andi Kleen --- arch/i386/Kconfig.cpu | 19 +++++++++++++------ arch/i386/Makefile.cpu | 1 + include/asm-i386/module.h | 2 ++ 3 files changed, 16 insertions(+), 6 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/Kconfig.cpu b/arch/i386/Kconfig.cpu index fc4f2abccf0..821fd269ca5 100644 --- a/arch/i386/Kconfig.cpu +++ b/arch/i386/Kconfig.cpu @@ -103,8 +103,15 @@ config MPENTIUMM Select this for Intel Pentium M (not Pentium-4 M) notebook chips. +config MCORE2 + bool "Core 2/newer Xeon" + help + Select this for Intel Core 2 and newer Core 2 Xeons (Xeon 51xx and 53xx) + CPUs. You can distingush newer from older Xeons by the CPU family + in /proc/cpuinfo. Newer ones have 6. + config MPENTIUM4 - bool "Pentium-4/Celeron(P4-based)/Pentium-4 M/Xeon" + bool "Pentium-4/Celeron(P4-based)/Pentium-4 M/older Xeon" help Select this for Intel Pentium 4 chips. This includes the Pentium 4, P4-based Celeron and Xeon, and Pentium-4 M @@ -229,7 +236,7 @@ config X86_L1_CACHE_SHIFT default "7" if MPENTIUM4 || X86_GENERIC default "4" if X86_ELAN || M486 || M386 || MGEODEGX1 default "5" if MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE_LX - default "6" if MK7 || MK8 || MPENTIUMM + default "6" if MK7 || MK8 || MPENTIUMM || MCORE2 config RWSEM_GENERIC_SPINLOCK bool @@ -287,17 +294,17 @@ config X86_ALIGNMENT_16 config X86_GOOD_APIC bool - depends on MK7 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || MK8 || MEFFICEON + depends on MK7 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || MK8 || MEFFICEON || MCORE2 default y config X86_INTEL_USERCOPY bool - depends on MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M586MMX || X86_GENERIC || MK8 || MK7 || MEFFICEON + depends on MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M586MMX || X86_GENERIC || MK8 || MK7 || MEFFICEON || MCORE2 default y config X86_USE_PPRO_CHECKSUM bool - depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX + depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2 default y config X86_USE_3DNOW @@ -312,5 +319,5 @@ config X86_OOSTORE config X86_TSC bool - depends on (MWINCHIP3D || MWINCHIP2 || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MGEODEGX1 || MGEODE_LX) && !X86_NUMAQ + depends on (MWINCHIP3D || MWINCHIP2 || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MGEODEGX1 || MGEODE_LX || MCORE2) && !X86_NUMAQ default y diff --git a/arch/i386/Makefile.cpu b/arch/i386/Makefile.cpu index a11befba26d..a32c031c90d 100644 --- a/arch/i386/Makefile.cpu +++ b/arch/i386/Makefile.cpu @@ -32,6 +32,7 @@ cflags-$(CONFIG_MWINCHIP2) += $(call cc-option,-march=winchip2,-march=i586) cflags-$(CONFIG_MWINCHIP3D) += $(call cc-option,-march=winchip2,-march=i586) cflags-$(CONFIG_MCYRIXIII) += $(call cc-option,-march=c3,-march=i486) $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0 cflags-$(CONFIG_MVIAC3_2) += $(call cc-option,-march=c3-2,-march=i686) +cflags-$(CONFIG_MCORE2) += -march=i686 $(call cc-option,-mtune=core2,$(call cc-option,-mtune=generic,-mtune=i686)) # AMD Elan support cflags-$(CONFIG_X86_ELAN) += -march=i486 diff --git a/include/asm-i386/module.h b/include/asm-i386/module.h index 424661d25bd..fe5ae42e027 100644 --- a/include/asm-i386/module.h +++ b/include/asm-i386/module.h @@ -20,6 +20,8 @@ struct mod_arch_specific #define MODULE_PROC_FAMILY "586TSC " #elif defined CONFIG_M586MMX #define MODULE_PROC_FAMILY "586MMX " +#elif defined CONFIG_MCORE2 +#define MODULE_PROC_FAMILY "CORE2 " #elif defined CONFIG_M686 #define MODULE_PROC_FAMILY "686 " #elif defined CONFIG_MPENTIUMII -- cgit v1.2.3-70-g09d2 From b0bfece40b1988aa8e3d910938691dce7859d82d Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 7 Dec 2006 02:14:09 +0100 Subject: [PATCH] i386: clear_fixmap() should not use set_pte() While not strictly required with the current code (as the upper half of page table entries generated by __set_fixmap() cannot be non-zero due to the second parameter of this function being 'unsigned long'), the use of set_pte() in __set_fixmap() in the context of clear_fixmap() is still improper with CONFIG_X86_PAE (see the respective comment in include/asm-i386/pgtable-3level.h) and would turn into a bug if that second parameter ever gets changed to a 64-bit type. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/i386/mm/pgtable.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c index 10126e3f817..65b5c095903 100644 --- a/arch/i386/mm/pgtable.c +++ b/arch/i386/mm/pgtable.c @@ -95,8 +95,11 @@ static void set_pte_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags) return; } pte = pte_offset_kernel(pmd, vaddr); - /* stored as-is, to permit clearing entries */ - set_pte(pte, pfn_pte(pfn, flags)); + if (pgprot_val(flags)) + /* stored as-is, to permit clearing entries */ + set_pte(pte, pfn_pte(pfn, flags)); + else + pte_clear(&init_mm, vaddr, pte); /* * It's enough to flush this one mapping. -- cgit v1.2.3-70-g09d2 From c6ea396de6836bdeb2d2433368130642bf0f6e15 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 7 Dec 2006 02:14:09 +0100 Subject: [PATCH] i386: Don't touch per cpu memory of offline CPUs in touch_nmi_watchdog Just like on x86-64, don't touch foreign CPUs' memory if the watchdog isn't enabled at all. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/i386/kernel/nmi.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index 171194ccb7b..f5bc7e1be80 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -870,14 +870,16 @@ static unsigned int void touch_nmi_watchdog (void) { - int i; + if (nmi_watchdog > 0) { + unsigned cpu; - /* - * Just reset the alert counters, (other CPUs might be - * spinning on locks we hold): - */ - for_each_possible_cpu(i) - alert_counter[i] = 0; + /* + * Just reset the alert counters, (other CPUs might be + * spinning on locks we hold): + */ + for_each_present_cpu (cpu) + alert_counter[cpu] = 0; + } /* * Tickle the softlockup detector too: -- cgit v1.2.3-70-g09d2 From 475850c86b908ae026d5a4be02a1b1e9c408c75a Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 7 Dec 2006 02:14:09 +0100 Subject: [PATCH] i386: conditionalize inclusion of some MTRR flavors Avoid inclusion of code that's dead for x86-64. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/i386/kernel/cpu/mtrr/Makefile | 4 +--- arch/i386/kernel/cpu/mtrr/main.c | 6 ++++++ 2 files changed, 7 insertions(+), 3 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/mtrr/Makefile b/arch/i386/kernel/cpu/mtrr/Makefile index a25b701ab84..191fc053364 100644 --- a/arch/i386/kernel/cpu/mtrr/Makefile +++ b/arch/i386/kernel/cpu/mtrr/Makefile @@ -1,5 +1,3 @@ obj-y := main.o if.o generic.o state.o -obj-y += amd.o -obj-y += cyrix.o -obj-y += centaur.o +obj-$(CONFIG_X86_32) += amd.o cyrix.o centaur.o diff --git a/arch/i386/kernel/cpu/mtrr/main.c b/arch/i386/kernel/cpu/mtrr/main.c index 2b8b0b361cc..a4de30b9d3d 100644 --- a/arch/i386/kernel/cpu/mtrr/main.c +++ b/arch/i386/kernel/cpu/mtrr/main.c @@ -59,7 +59,11 @@ struct mtrr_ops * mtrr_if = NULL; static void set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type); +#ifndef CONFIG_X86_64 extern int arr3_protected; +#else +#define arr3_protected 0 +#endif void set_mtrr_ops(struct mtrr_ops * ops) { @@ -544,9 +548,11 @@ extern void centaur_init_mtrr(void); static void __init init_ifs(void) { +#ifndef CONFIG_X86_64 amd_init_mtrr(); cyrix_init_mtrr(); centaur_init_mtrr(); +#endif } /* The suspend/resume methods are only for CPU without MTRR. CPU using generic -- cgit v1.2.3-70-g09d2 From 365bff806e9faba000fb4956c7486fbf3a746d96 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 7 Dec 2006 02:14:09 +0100 Subject: [PATCH] i386: fix MTRR code Until not so long ago, there were system log messages pointing to inconsistent MTRR setup of the video frame buffer caused by the way vesafb and X worked. While vesafb was fixed meanwhile, I believe fixing it there only hides a shortcoming in the MTRR code itself, in that that code is not symmetric with respect to the ordering of attempts to set up two (or more) regions where one contains the other. In the current shape, it permits only setting up sub-regions of pre-exisiting ones. The patch below makes this symmetric. While working on that I noticed a few more inconsistencies in that code, namely - use of 'unsigned int' for sizes in many, but not all places (the patch is converting this to use 'unsigned long' everywhere, which specifically might be necessary for x86-64 once a processor supporting more than 44 physical address bits would become available) - the code to correct inconsistent settings during secondary processor startup tried (if necessary) to correct, among other things, the value in IA32_MTRR_DEF_TYPE, however the newly computed value would never get used (i.e. stored in the respective MSR) - the generic range validation code checked that the end of the to-be-added range would be above 1MB; the value checked should have been the start of the range - when contained regions are detected, previously this was allowed only when the old region was uncacheable; this can be symmetric (i.e. the new region can also be uncacheable) and even further as per Intel's documentation write-trough and write-back for either region is also compatible with the respective opposite in the other Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/i386/kernel/cpu/mtrr/amd.c | 2 +- arch/i386/kernel/cpu/mtrr/centaur.c | 9 +++-- arch/i386/kernel/cpu/mtrr/cyrix.c | 25 +++++++++--- arch/i386/kernel/cpu/mtrr/generic.c | 78 +++++++++++++++++++++++++++++++------ arch/i386/kernel/cpu/mtrr/if.c | 28 +++++++------ arch/i386/kernel/cpu/mtrr/main.c | 55 +++++++++++++++++++------- arch/i386/kernel/cpu/mtrr/mtrr.h | 25 ++++++------ 7 files changed, 162 insertions(+), 60 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/mtrr/amd.c b/arch/i386/kernel/cpu/mtrr/amd.c index 1a1e04b6fd0..0949cdbf848 100644 --- a/arch/i386/kernel/cpu/mtrr/amd.c +++ b/arch/i386/kernel/cpu/mtrr/amd.c @@ -7,7 +7,7 @@ static void amd_get_mtrr(unsigned int reg, unsigned long *base, - unsigned int *size, mtrr_type * type) + unsigned long *size, mtrr_type * type) { unsigned long low, high; diff --git a/arch/i386/kernel/cpu/mtrr/centaur.c b/arch/i386/kernel/cpu/mtrr/centaur.c index 33f00ac314e..cb9aa3a7a7a 100644 --- a/arch/i386/kernel/cpu/mtrr/centaur.c +++ b/arch/i386/kernel/cpu/mtrr/centaur.c @@ -17,7 +17,7 @@ static u8 centaur_mcr_type; /* 0 for winchip, 1 for winchip2 */ */ static int -centaur_get_free_region(unsigned long base, unsigned long size) +centaur_get_free_region(unsigned long base, unsigned long size, int replace_reg) /* [SUMMARY] Get a free MTRR. The starting (base) address of the region. The size (in bytes) of the region. @@ -26,10 +26,11 @@ centaur_get_free_region(unsigned long base, unsigned long size) { int i, max; mtrr_type ltype; - unsigned long lbase; - unsigned int lsize; + unsigned long lbase, lsize; max = num_var_ranges; + if (replace_reg >= 0 && replace_reg < max) + return replace_reg; for (i = 0; i < max; ++i) { if (centaur_mcr_reserved & (1 << i)) continue; @@ -49,7 +50,7 @@ mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi) static void centaur_get_mcr(unsigned int reg, unsigned long *base, - unsigned int *size, mtrr_type * type) + unsigned long *size, mtrr_type * type) { *base = centaur_mcr[reg].high >> PAGE_SHIFT; *size = -(centaur_mcr[reg].low & 0xfffff000) >> PAGE_SHIFT; diff --git a/arch/i386/kernel/cpu/mtrr/cyrix.c b/arch/i386/kernel/cpu/mtrr/cyrix.c index 9027a987006..0737a596db4 100644 --- a/arch/i386/kernel/cpu/mtrr/cyrix.c +++ b/arch/i386/kernel/cpu/mtrr/cyrix.c @@ -9,7 +9,7 @@ int arr3_protected; static void cyrix_get_arr(unsigned int reg, unsigned long *base, - unsigned int *size, mtrr_type * type) + unsigned long *size, mtrr_type * type) { unsigned long flags; unsigned char arr, ccr3, rcr, shift; @@ -77,7 +77,7 @@ cyrix_get_arr(unsigned int reg, unsigned long *base, } static int -cyrix_get_free_region(unsigned long base, unsigned long size) +cyrix_get_free_region(unsigned long base, unsigned long size, int replace_reg) /* [SUMMARY] Get a free ARR. The starting (base) address of the region. The size (in bytes) of the region. @@ -86,9 +86,24 @@ cyrix_get_free_region(unsigned long base, unsigned long size) { int i; mtrr_type ltype; - unsigned long lbase; - unsigned int lsize; + unsigned long lbase, lsize; + switch (replace_reg) { + case 7: + if (size < 0x40) + break; + case 6: + case 5: + case 4: + return replace_reg; + case 3: + if (arr3_protected) + break; + case 2: + case 1: + case 0: + return replace_reg; + } /* If we are to set up a region >32M then look at ARR7 immediately */ if (size > 0x2000) { cyrix_get_arr(7, &lbase, &lsize, <ype); @@ -214,7 +229,7 @@ static void cyrix_set_arr(unsigned int reg, unsigned long base, typedef struct { unsigned long base; - unsigned int size; + unsigned long size; mtrr_type type; } arr_state_t; diff --git a/arch/i386/kernel/cpu/mtrr/generic.c b/arch/i386/kernel/cpu/mtrr/generic.c index ee8dc675395..f77fc53db65 100644 --- a/arch/i386/kernel/cpu/mtrr/generic.c +++ b/arch/i386/kernel/cpu/mtrr/generic.c @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -15,12 +16,19 @@ struct mtrr_state { struct mtrr_var_range *var_ranges; mtrr_type fixed_ranges[NUM_FIXED_RANGES]; unsigned char enabled; + unsigned char have_fixed; mtrr_type def_type; }; static unsigned long smp_changes_mask; static struct mtrr_state mtrr_state = {}; +#undef MODULE_PARAM_PREFIX +#define MODULE_PARAM_PREFIX "mtrr." + +static __initdata int mtrr_show; +module_param_named(show, mtrr_show, bool, 0); + /* Get the MSR pair relating to a var range */ static void __init get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr) @@ -43,6 +51,14 @@ get_fixed_ranges(mtrr_type * frs) rdmsr(MTRRfix4K_C0000_MSR + i, p[6 + i * 2], p[7 + i * 2]); } +static void __init print_fixed(unsigned base, unsigned step, const mtrr_type*types) +{ + unsigned i; + + for (i = 0; i < 8; ++i, ++types, base += step) + printk(KERN_INFO "MTRR %05X-%05X %s\n", base, base + step - 1, mtrr_attrib_to_str(*types)); +} + /* Grab all of the MTRR state for this CPU into *state */ void __init get_mtrr_state(void) { @@ -58,13 +74,49 @@ void __init get_mtrr_state(void) } vrs = mtrr_state.var_ranges; + rdmsr(MTRRcap_MSR, lo, dummy); + mtrr_state.have_fixed = (lo >> 8) & 1; + for (i = 0; i < num_var_ranges; i++) get_mtrr_var_range(i, &vrs[i]); - get_fixed_ranges(mtrr_state.fixed_ranges); + if (mtrr_state.have_fixed) + get_fixed_ranges(mtrr_state.fixed_ranges); rdmsr(MTRRdefType_MSR, lo, dummy); mtrr_state.def_type = (lo & 0xff); mtrr_state.enabled = (lo & 0xc00) >> 10; + + if (mtrr_show) { + int high_width; + + printk(KERN_INFO "MTRR default type: %s\n", mtrr_attrib_to_str(mtrr_state.def_type)); + if (mtrr_state.have_fixed) { + printk(KERN_INFO "MTRR fixed ranges %sabled:\n", + mtrr_state.enabled & 1 ? "en" : "dis"); + print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0); + for (i = 0; i < 2; ++i) + print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8); + for (i = 0; i < 8; ++i) + print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8); + } + printk(KERN_INFO "MTRR variable ranges %sabled:\n", + mtrr_state.enabled & 2 ? "en" : "dis"); + high_width = ((size_or_mask ? ffs(size_or_mask) - 1 : 32) - (32 - PAGE_SHIFT) + 3) / 4; + for (i = 0; i < num_var_ranges; ++i) { + if (mtrr_state.var_ranges[i].mask_lo & (1 << 11)) + printk(KERN_INFO "MTRR %u base %0*X%05X000 mask %0*X%05X000 %s\n", + i, + high_width, + mtrr_state.var_ranges[i].base_hi, + mtrr_state.var_ranges[i].base_lo >> 12, + high_width, + mtrr_state.var_ranges[i].mask_hi, + mtrr_state.var_ranges[i].mask_lo >> 12, + mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff)); + else + printk(KERN_INFO "MTRR %u disabled\n", i); + } + } } /* Some BIOS's are fucked and don't set all MTRRs the same! */ @@ -95,7 +147,7 @@ void mtrr_wrmsr(unsigned msr, unsigned a, unsigned b) smp_processor_id(), msr, a, b); } -int generic_get_free_region(unsigned long base, unsigned long size) +int generic_get_free_region(unsigned long base, unsigned long size, int replace_reg) /* [SUMMARY] Get a free MTRR. The starting (base) address of the region. The size (in bytes) of the region. @@ -104,10 +156,11 @@ int generic_get_free_region(unsigned long base, unsigned long size) { int i, max; mtrr_type ltype; - unsigned long lbase; - unsigned lsize; + unsigned long lbase, lsize; max = num_var_ranges; + if (replace_reg >= 0 && replace_reg < max) + return replace_reg; for (i = 0; i < max; ++i) { mtrr_if->get(i, &lbase, &lsize, <ype); if (lsize == 0) @@ -117,7 +170,7 @@ int generic_get_free_region(unsigned long base, unsigned long size) } static void generic_get_mtrr(unsigned int reg, unsigned long *base, - unsigned int *size, mtrr_type * type) + unsigned long *size, mtrr_type *type) { unsigned int mask_lo, mask_hi, base_lo, base_hi; @@ -202,7 +255,9 @@ static int set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr) return changed; } -static unsigned long set_mtrr_state(u32 deftype_lo, u32 deftype_hi) +static u32 deftype_lo, deftype_hi; + +static unsigned long set_mtrr_state(void) /* [SUMMARY] Set the MTRR state for this CPU. The MTRR state information to read. Some relevant CPU context. @@ -217,14 +272,14 @@ static unsigned long set_mtrr_state(u32 deftype_lo, u32 deftype_hi) if (set_mtrr_var_ranges(i, &mtrr_state.var_ranges[i])) change_mask |= MTRR_CHANGE_MASK_VARIABLE; - if (set_fixed_ranges(mtrr_state.fixed_ranges)) + if (mtrr_state.have_fixed && set_fixed_ranges(mtrr_state.fixed_ranges)) change_mask |= MTRR_CHANGE_MASK_FIXED; /* Set_mtrr_restore restores the old value of MTRRdefType, so to set it we fiddle with the saved value */ if ((deftype_lo & 0xff) != mtrr_state.def_type || ((deftype_lo & 0xc00) >> 10) != mtrr_state.enabled) { - deftype_lo |= (mtrr_state.def_type | mtrr_state.enabled << 10); + deftype_lo = (deftype_lo & ~0xcff) | mtrr_state.def_type | (mtrr_state.enabled << 10); change_mask |= MTRR_CHANGE_MASK_DEFTYPE; } @@ -233,7 +288,6 @@ static unsigned long set_mtrr_state(u32 deftype_lo, u32 deftype_hi) static unsigned long cr4 = 0; -static u32 deftype_lo, deftype_hi; static DEFINE_SPINLOCK(set_atomicity_lock); /* @@ -271,7 +325,7 @@ static void prepare_set(void) __acquires(set_atomicity_lock) rdmsr(MTRRdefType_MSR, deftype_lo, deftype_hi); /* Disable MTRRs, and set the default type to uncached */ - mtrr_wrmsr(MTRRdefType_MSR, deftype_lo & 0xf300UL, deftype_hi); + mtrr_wrmsr(MTRRdefType_MSR, deftype_lo & ~0xcff, deftype_hi); } static void post_set(void) __releases(set_atomicity_lock) @@ -300,7 +354,7 @@ static void generic_set_all(void) prepare_set(); /* Actually set the state */ - mask = set_mtrr_state(deftype_lo,deftype_hi); + mask = set_mtrr_state(); post_set(); local_irq_restore(flags); @@ -374,7 +428,7 @@ int generic_validate_add_page(unsigned long base, unsigned long size, unsigned i } } - if (base < 0x100) { + if (base + size < 0x100) { printk(KERN_WARNING "mtrr: cannot set region below 1 MiB (0x%lx000,0x%lx000)\n", base, size); return -EINVAL; diff --git a/arch/i386/kernel/cpu/mtrr/if.c b/arch/i386/kernel/cpu/mtrr/if.c index 5ac051bb9d5..9753bc6a1f3 100644 --- a/arch/i386/kernel/cpu/mtrr/if.c +++ b/arch/i386/kernel/cpu/mtrr/if.c @@ -17,7 +17,7 @@ extern unsigned int *usage_table; #define FILE_FCOUNT(f) (((struct seq_file *)((f)->private_data))->private) -static char *mtrr_strings[MTRR_NUM_TYPES] = +static const char *const mtrr_strings[MTRR_NUM_TYPES] = { "uncachable", /* 0 */ "write-combining", /* 1 */ @@ -28,7 +28,7 @@ static char *mtrr_strings[MTRR_NUM_TYPES] = "write-back", /* 6 */ }; -char *mtrr_attrib_to_str(int x) +const char *mtrr_attrib_to_str(int x) { return (x <= 6) ? mtrr_strings[x] : "?"; } @@ -155,6 +155,7 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg) { int err = 0; mtrr_type type; + unsigned long size; struct mtrr_sentry sentry; struct mtrr_gentry gentry; void __user *arg = (void __user *) __arg; @@ -235,15 +236,15 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg) case MTRRIOC_GET_ENTRY: if (gentry.regnum >= num_var_ranges) return -EINVAL; - mtrr_if->get(gentry.regnum, &gentry.base, &gentry.size, &type); + mtrr_if->get(gentry.regnum, &gentry.base, &size, &type); /* Hide entries that go above 4GB */ - if (gentry.base + gentry.size > 0x100000 - || gentry.size == 0x100000) + if (gentry.base + size - 1 >= (1UL << (8 * sizeof(gentry.size) - PAGE_SHIFT)) + || size >= (1UL << (8 * sizeof(gentry.size) - PAGE_SHIFT))) gentry.base = gentry.size = gentry.type = 0; else { gentry.base <<= PAGE_SHIFT; - gentry.size <<= PAGE_SHIFT; + gentry.size = size << PAGE_SHIFT; gentry.type = type; } @@ -273,8 +274,14 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg) case MTRRIOC_GET_PAGE_ENTRY: if (gentry.regnum >= num_var_ranges) return -EINVAL; - mtrr_if->get(gentry.regnum, &gentry.base, &gentry.size, &type); - gentry.type = type; + mtrr_if->get(gentry.regnum, &gentry.base, &size, &type); + /* Hide entries that would overflow */ + if (size != (__typeof__(gentry.size))size) + gentry.base = gentry.size = gentry.type = 0; + else { + gentry.size = size; + gentry.type = type; + } break; } @@ -353,8 +360,7 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset) char factor; int i, max, len; mtrr_type type; - unsigned long base; - unsigned int size; + unsigned long base, size; len = 0; max = num_var_ranges; @@ -373,7 +379,7 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset) } /* RED-PEN: base can be > 32bit */ len += seq_printf(seq, - "reg%02i: base=0x%05lx000 (%4liMB), size=%4i%cB: %s, count=%d\n", + "reg%02i: base=0x%05lx000 (%4luMB), size=%4lu%cB: %s, count=%d\n", i, base, base >> (20 - PAGE_SHIFT), size, factor, mtrr_attrib_to_str(type), usage_table[i]); } diff --git a/arch/i386/kernel/cpu/mtrr/main.c b/arch/i386/kernel/cpu/mtrr/main.c index a4de30b9d3d..aeea23e8a05 100644 --- a/arch/i386/kernel/cpu/mtrr/main.c +++ b/arch/i386/kernel/cpu/mtrr/main.c @@ -172,6 +172,13 @@ static void ipi_handler(void *info) #endif +static inline int types_compatible(mtrr_type type1, mtrr_type type2) { + return type1 == MTRR_TYPE_UNCACHABLE || + type2 == MTRR_TYPE_UNCACHABLE || + (type1 == MTRR_TYPE_WRTHROUGH && type2 == MTRR_TYPE_WRBACK) || + (type1 == MTRR_TYPE_WRBACK && type2 == MTRR_TYPE_WRTHROUGH); +} + /** * set_mtrr - update mtrrs on all processors * @reg: mtrr in question @@ -304,11 +311,9 @@ static void set_mtrr(unsigned int reg, unsigned long base, int mtrr_add_page(unsigned long base, unsigned long size, unsigned int type, char increment) { - int i; + int i, replace, error; mtrr_type ltype; - unsigned long lbase; - unsigned int lsize; - int error; + unsigned long lbase, lsize; if (!mtrr_if) return -ENXIO; @@ -328,12 +333,18 @@ int mtrr_add_page(unsigned long base, unsigned long size, return -ENOSYS; } + if (!size) { + printk(KERN_WARNING "mtrr: zero sized request\n"); + return -EINVAL; + } + if (base & size_or_mask || size & size_or_mask) { printk(KERN_WARNING "mtrr: base or size exceeds the MTRR width\n"); return -EINVAL; } error = -EINVAL; + replace = -1; /* No CPU hotplug when we change MTRR entries */ lock_cpu_hotplug(); @@ -341,21 +352,28 @@ int mtrr_add_page(unsigned long base, unsigned long size, mutex_lock(&mtrr_mutex); for (i = 0; i < num_var_ranges; ++i) { mtrr_if->get(i, &lbase, &lsize, <ype); - if (base >= lbase + lsize) - continue; - if ((base < lbase) && (base + size <= lbase)) + if (!lsize || base > lbase + lsize - 1 || base + size - 1 < lbase) continue; /* At this point we know there is some kind of overlap/enclosure */ - if ((base < lbase) || (base + size > lbase + lsize)) { + if (base < lbase || base + size - 1 > lbase + lsize - 1) { + if (base <= lbase && base + size - 1 >= lbase + lsize - 1) { + /* New region encloses an existing region */ + if (type == ltype) { + replace = replace == -1 ? i : -2; + continue; + } + else if (types_compatible(type, ltype)) + continue; + } printk(KERN_WARNING "mtrr: 0x%lx000,0x%lx000 overlaps existing" - " 0x%lx000,0x%x000\n", base, size, lbase, + " 0x%lx000,0x%lx000\n", base, size, lbase, lsize); goto out; } /* New region is enclosed by an existing region */ if (ltype != type) { - if (type == MTRR_TYPE_UNCACHABLE) + if (types_compatible(type, ltype)) continue; printk (KERN_WARNING "mtrr: type mismatch for %lx000,%lx000 old: %s new: %s\n", base, size, mtrr_attrib_to_str(ltype), @@ -368,10 +386,18 @@ int mtrr_add_page(unsigned long base, unsigned long size, goto out; } /* Search for an empty MTRR */ - i = mtrr_if->get_free_region(base, size); + i = mtrr_if->get_free_region(base, size, replace); if (i >= 0) { set_mtrr(i, base, size, type); - usage_table[i] = 1; + if (likely(replace < 0)) + usage_table[i] = 1; + else { + usage_table[i] = usage_table[replace] + !!increment; + if (unlikely(replace != i)) { + set_mtrr(replace, 0, 0, 0); + usage_table[replace] = 0; + } + } } else printk(KERN_INFO "mtrr: no more MTRRs available\n"); error = i; @@ -459,8 +485,7 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size) { int i, max; mtrr_type ltype; - unsigned long lbase; - unsigned int lsize; + unsigned long lbase, lsize; int error = -EINVAL; if (!mtrr_if) @@ -561,7 +586,7 @@ static void __init init_ifs(void) struct mtrr_value { mtrr_type ltype; unsigned long lbase; - unsigned int lsize; + unsigned long lsize; }; static struct mtrr_value * mtrr_state; diff --git a/arch/i386/kernel/cpu/mtrr/mtrr.h b/arch/i386/kernel/cpu/mtrr/mtrr.h index 99c9f268204..d61ea9db6cf 100644 --- a/arch/i386/kernel/cpu/mtrr/mtrr.h +++ b/arch/i386/kernel/cpu/mtrr/mtrr.h @@ -43,15 +43,16 @@ struct mtrr_ops { void (*set_all)(void); void (*get)(unsigned int reg, unsigned long *base, - unsigned int *size, mtrr_type * type); - int (*get_free_region) (unsigned long base, unsigned long size); - + unsigned long *size, mtrr_type * type); + int (*get_free_region)(unsigned long base, unsigned long size, + int replace_reg); int (*validate_add_page)(unsigned long base, unsigned long size, unsigned int type); int (*have_wrcomb)(void); }; -extern int generic_get_free_region(unsigned long base, unsigned long size); +extern int generic_get_free_region(unsigned long base, unsigned long size, + int replace_reg); extern int generic_validate_add_page(unsigned long base, unsigned long size, unsigned int type); @@ -62,17 +63,17 @@ extern int positive_have_wrcomb(void); /* library functions for processor-specific routines */ struct set_mtrr_context { unsigned long flags; - unsigned long deftype_lo; - unsigned long deftype_hi; unsigned long cr4val; - unsigned long ccr3; + u32 deftype_lo; + u32 deftype_hi; + u32 ccr3; }; struct mtrr_var_range { - unsigned long base_lo; - unsigned long base_hi; - unsigned long mask_lo; - unsigned long mask_hi; + u32 base_lo; + u32 base_hi; + u32 mask_lo; + u32 mask_hi; }; void set_mtrr_done(struct set_mtrr_context *ctxt); @@ -92,6 +93,6 @@ extern struct mtrr_ops * mtrr_if; extern unsigned int num_var_ranges; void mtrr_state_warn(void); -char *mtrr_attrib_to_str(int x); +const char *mtrr_attrib_to_str(int x); void mtrr_wrmsr(unsigned, unsigned, unsigned); -- cgit v1.2.3-70-g09d2 From ba10650a880c2df23bd1db6c0570ddb66f389641 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 7 Dec 2006 02:14:10 +0100 Subject: [PATCH] i386: alloc_gdt() static Make the needlessly global alloc_gdt() static. (against) pda-percpu-init Signed-off-by: Adrian Bunk Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton --- arch/i386/kernel/cpu/common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index 68bcb687019..1b34c56f812 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -609,7 +609,7 @@ struct pt_regs * __devinit idle_regs(struct pt_regs *regs) return regs; } -__cpuinit int alloc_gdt(int cpu) +static __cpuinit int alloc_gdt(int cpu) { struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); struct desc_struct *gdt; -- cgit v1.2.3-70-g09d2 From 79929fd1c1887d2a057cbb80d487a2e2f1c01a02 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Thu, 7 Dec 2006 02:14:10 +0100 Subject: [PATCH] i386: Convert more absolute symbols to section relative o Convert more absolute symbols to section relative to keep the theme in vmlinux.lds.S file and to avoid problem if kernel is relocated. o Also put a message so that in future people can be aware of it and avoid introducing absolute symbols. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/vmlinux.lds.S | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 877dc5cfe3a..25581e87c60 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -8,6 +8,12 @@ * put it inside the section definition. */ +/* Don't define absolute symbols until and unless you know that symbol + * value is should remain constant even if kernel image is relocated + * at run time. Absolute symbols are not relocated. If symbol value should + * change if kernel is relocated, make the symbol section relative and + * put it inside the section definition. + */ #define LOAD_OFFSET __PAGE_OFFSET #include @@ -65,11 +71,11 @@ SECTIONS CONSTRUCTORS } :data - __start_paravirtprobe = .; .paravirtprobe : AT(ADDR(.paravirtprobe) - LOAD_OFFSET) { + __start_paravirtprobe = .; *(.paravirtprobe) + __stop_paravirtprobe = .; } - __stop_paravirtprobe = .; . = ALIGN(4096); .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { @@ -172,11 +178,11 @@ SECTIONS *(.altinstr_replacement) } . = ALIGN(4); - __start_parainstructions = .; .parainstructions : AT(ADDR(.parainstructions) - LOAD_OFFSET) { + __start_parainstructions = .; *(.parainstructions) + __stop_parainstructions = .; } - __stop_parainstructions = .; /* .exit.text is discard at runtime, not link time, to deal with references from .altinstructions and .eh_frame */ .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { *(.exit.text) } -- cgit v1.2.3-70-g09d2 From 274e1bbdeeaf16e71418f11f5f305ab26061f2c2 Mon Sep 17 00:00:00 2001 From: "Siddha, Suresh B" Date: Thu, 7 Dec 2006 02:14:10 +0100 Subject: [PATCH] x86: add write_pci_config_byte() to direct PCI access routines Mechanism of selecting physical mode in genapic when cpu hotplug is enabled on x86_64, broke the quirk(quirk_intel_irqbalance()) introduced for working around the transposing interrupt message errata in E7520/E7320/E7525 (revision ID 0x9 and below. errata #23 in http://download.intel.com/design/chipsets/specupdt/30304203.pdf). This errata requires the mode to be in logical flat, so that interrupts can be directed to more than one cpu(and thus use hardware IRQ balancing enabled by BIOS on these platforms). Following four patches fixes this by moving the quirk to early quirk and forcing the x86_64 genapic selection to logical flat on these platforms. Thanks to Shaohua for pointing out the breakage. This patch: Add write_pci_config_byte() to direct PCI access routines Signed-off-by: Suresh Siddha Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: "Li, Shaohua" Signed-off-by: Andrew Morton --- arch/i386/pci/early.c | 7 +++++++ include/asm-x86_64/pci-direct.h | 1 + 2 files changed, 8 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/pci/early.c b/arch/i386/pci/early.c index 713d6c866ca..42df4b6606d 100644 --- a/arch/i386/pci/early.c +++ b/arch/i386/pci/early.c @@ -45,6 +45,13 @@ void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset, outl(val, 0xcfc); } +void write_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset, u8 val) +{ + PDprintk("%x writing to %x: %x\n", slot, offset, val); + outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8); + outb(val, 0xcfc); +} + int early_pci_allowed(void) { return (pci_probe & (PCI_PROBE_CONF1|PCI_PROBE_NOEARLY)) == diff --git a/include/asm-x86_64/pci-direct.h b/include/asm-x86_64/pci-direct.h index eba9cb471df..6823fa4f1af 100644 --- a/include/asm-x86_64/pci-direct.h +++ b/include/asm-x86_64/pci-direct.h @@ -10,6 +10,7 @@ extern u32 read_pci_config(u8 bus, u8 slot, u8 func, u8 offset); extern u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset); extern u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset); extern void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset, u32 val); +extern void write_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset, u8 val); extern int early_pci_allowed(void); -- cgit v1.2.3-70-g09d2 From fd6d7d26897dec834d0b9fbdc59819b0332a1257 Mon Sep 17 00:00:00 2001 From: "Siddha, Suresh B" Date: Thu, 7 Dec 2006 02:14:10 +0100 Subject: [PATCH] i386: introduce the mechanism of disabling cpu hotplug control Add 'enable_cpu_hotplug' flag and when cleared, the hotplug control file ("online") will not be added under /sys/devices/system/cpu/cpuX/ Next patch doing PCI quirks will use this. Signed-off-by: Suresh Siddha Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: "Li, Shaohua" Signed-off-by: Andrew Morton --- arch/i386/kernel/topology.c | 6 +++++- include/asm-i386/cpu.h | 3 +++ 2 files changed, 8 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/topology.c b/arch/i386/kernel/topology.c index 07d6da36a82..844c08fdb22 100644 --- a/arch/i386/kernel/topology.c +++ b/arch/i386/kernel/topology.c @@ -40,14 +40,18 @@ int arch_register_cpu(int num) * restrictions and assumptions in kernel. This basically * doesnt add a control file, one cannot attempt to offline * BSP. + * + * Also certain PCI quirks require not to enable hotplug control + * for all CPU's. */ - if (!num) + if (!num || !enable_cpu_hotplug) cpu_devices[num].cpu.no_control = 1; return register_cpu(&cpu_devices[num].cpu, num); } #ifdef CONFIG_HOTPLUG_CPU +int enable_cpu_hotplug = 1; void arch_unregister_cpu(int num) { return unregister_cpu(&cpu_devices[num].cpu); diff --git a/include/asm-i386/cpu.h b/include/asm-i386/cpu.h index b1bc7b1b64b..9d914e1e4aa 100644 --- a/include/asm-i386/cpu.h +++ b/include/asm-i386/cpu.h @@ -13,6 +13,9 @@ struct i386_cpu { extern int arch_register_cpu(int num); #ifdef CONFIG_HOTPLUG_CPU extern void arch_unregister_cpu(int); +extern int enable_cpu_hotplug; +#else +#define enable_cpu_hotplug 0 #endif DECLARE_PER_CPU(int, cpu_state); -- cgit v1.2.3-70-g09d2 From 72486f1f8f0a2bc828b9d30cf4690cf2dd6807fc Mon Sep 17 00:00:00 2001 From: "Siddha, Suresh B" Date: Thu, 7 Dec 2006 02:14:10 +0100 Subject: [PATCH] i386: change the 'no_control' field to 'hotpluggable' in the struct cpu Change the 'no_control' field in the cpu struct to a more positive and better term 'hotpluggable'. And change(/cleanup) the logic accordingly. Signed-off-by: Suresh Siddha Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: "Li, Shaohua" Signed-off-by: Andrew Morton --- arch/i386/kernel/topology.c | 4 ++-- arch/ia64/kernel/topology.c | 8 ++++---- arch/powerpc/kernel/sysfs.c | 8 ++++---- drivers/base/cpu.c | 6 +++--- include/linux/cpu.h | 2 +- 5 files changed, 14 insertions(+), 14 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/topology.c b/arch/i386/kernel/topology.c index 844c08fdb22..79cf608e14c 100644 --- a/arch/i386/kernel/topology.c +++ b/arch/i386/kernel/topology.c @@ -44,8 +44,8 @@ int arch_register_cpu(int num) * Also certain PCI quirks require not to enable hotplug control * for all CPU's. */ - if (!num || !enable_cpu_hotplug) - cpu_devices[num].cpu.no_control = 1; + if (num && enable_cpu_hotplug) + cpu_devices[num].cpu.hotpluggable = 1; return register_cpu(&cpu_devices[num].cpu, num); } diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index 5629b45e89c..687500ddb4b 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -31,11 +31,11 @@ int arch_register_cpu(int num) { #if defined (CONFIG_ACPI) && defined (CONFIG_HOTPLUG_CPU) /* - * If CPEI cannot be re-targetted, and this is - * CPEI target, then dont create the control file + * If CPEI can be re-targetted or if this is not + * CPEI target, then it is hotpluggable */ - if (!can_cpei_retarget() && is_cpu_cpei_target(num)) - sysfs_cpus[num].cpu.no_control = 1; + if (can_cpei_retarget() || !is_cpu_cpei_target(num)) + sysfs_cpus[num].cpu.hotpluggable = 1; map_cpu_to_node(num, node_cpuid[num].nid); #endif diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index 22123a0d541..63ed265b7f0 100644 --- a/arch/powerpc/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -239,7 +239,7 @@ static void unregister_cpu_online(unsigned int cpu) struct cpu *c = &per_cpu(cpu_devices, cpu); struct sys_device *s = &c->sysdev; - BUG_ON(c->no_control); + BUG_ON(!c->hotpluggable); if (!firmware_has_feature(FW_FEATURE_ISERIES) && cpu_has_feature(CPU_FTR_SMT)) @@ -424,10 +424,10 @@ static int __init topology_init(void) * CPU. For instance, the boot cpu might never be valid * for hotplugging. */ - if (!ppc_md.cpu_die) - c->no_control = 1; + if (ppc_md.cpu_die) + c->hotpluggable = 1; - if (cpu_online(cpu) || (c->no_control == 0)) { + if (cpu_online(cpu) || c->hotpluggable) { register_cpu(c, cpu); sysdev_create_file(&c->sysdev, &attr_physical_id); diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 1f745f12f94..7fd095efaeb 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -104,8 +104,8 @@ static SYSDEV_ATTR(crash_notes, 0400, show_crash_notes, NULL); /* * register_cpu - Setup a driverfs device for a CPU. - * @cpu - Callers can set the cpu->no_control field to 1, to indicate not to - * generate a control file in sysfs for this CPU. + * @cpu - cpu->hotpluggable field set to 1 will generate a control file in + * sysfs for this CPU. * @num - CPU number to use when creating the device. * * Initialize and register the CPU device. @@ -119,7 +119,7 @@ int __devinit register_cpu(struct cpu *cpu, int num) error = sysdev_register(&cpu->sysdev); - if (!error && !cpu->no_control) + if (!error && cpu->hotpluggable) register_cpu_control(cpu); if (!error) cpu_sys_devices[num] = &cpu->sysdev; diff --git a/include/linux/cpu.h b/include/linux/cpu.h index f02d71bf689..ad90340e7db 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -27,7 +27,7 @@ struct cpu { int node_id; /* The node which contains the CPU */ - int no_control; /* Should the sysfs control file be created? */ + int hotpluggable; /* creates sysfs control file if hotpluggable */ struct sys_device sysdev; }; -- cgit v1.2.3-70-g09d2 From b0d0a4ba45760b10ecee9035ed45b442c1a6cc84 Mon Sep 17 00:00:00 2001 From: "Siddha, Suresh B" Date: Thu, 7 Dec 2006 02:14:10 +0100 Subject: [PATCH] x86: fix the irqbalance quirk for E7320/E7520/E7525 Move the irqbalance quirks for E7320/E7520/E7525(Errata 23 in http://download.intel.com/design/chipsets/specupdt/30304203.pdf) to early quirks. And add a PCI quirk for these platforms to check(which happens very late during the boot) if the APIC routing is indeed set to default flat mode. This fixes the breakage(in x86_64) of this quirk due to cpu hotplug which selects physical mode instead of the logical flat(as needed for this errata workaround). Signed-off-by: Suresh Siddha Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: "Li, Shaohua" Signed-off-by: Andrew Morton --- arch/i386/kernel/acpi/earlyquirk.c | 21 +++++++++++++++++ arch/i386/kernel/quirks.c | 46 +++++++++++++++++++++++++++++--------- arch/i386/kernel/smpboot.c | 7 ++++++ arch/x86_64/kernel/early-quirks.c | 13 +++++++++++ arch/x86_64/kernel/smpboot.c | 8 +++++++ include/asm-i386/genapic.h | 2 +- include/asm-i386/irq.h | 2 ++ include/asm-x86_64/proto.h | 1 + 8 files changed, 88 insertions(+), 12 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/acpi/earlyquirk.c b/arch/i386/kernel/acpi/earlyquirk.c index c9841692bb7..4b60af7f91d 100644 --- a/arch/i386/kernel/acpi/earlyquirk.c +++ b/arch/i386/kernel/acpi/earlyquirk.c @@ -10,6 +10,7 @@ #include #include #include +#include #ifdef CONFIG_ACPI @@ -49,6 +50,24 @@ static int __init check_bridge(int vendor, int device) return 0; } +static void check_intel(void) +{ + u16 vendor, device; + + vendor = read_pci_config_16(0, 0, 0, PCI_VENDOR_ID); + + if (vendor != PCI_VENDOR_ID_INTEL) + return; + + device = read_pci_config_16(0, 0, 0, PCI_DEVICE_ID); +#ifdef CONFIG_SMP + if (device == PCI_DEVICE_ID_INTEL_E7320_MCH || + device == PCI_DEVICE_ID_INTEL_E7520_MCH || + device == PCI_DEVICE_ID_INTEL_E7525_MCH) + quirk_intel_irqbalance(); +#endif +} + void __init check_acpi_pci(void) { int num, slot, func; @@ -60,6 +79,8 @@ void __init check_acpi_pci(void) if (!early_pci_allowed()) return; + check_intel(); + /* Poor man's PCI discovery */ for (num = 0; num < 32; num++) { for (slot = 0; slot < 32; slot++) { diff --git a/arch/i386/kernel/quirks.c b/arch/i386/kernel/quirks.c index 9f6ab1789bb..a01320a7b63 100644 --- a/arch/i386/kernel/quirks.c +++ b/arch/i386/kernel/quirks.c @@ -3,10 +3,23 @@ */ #include #include +#include +#include +#include #if defined(CONFIG_X86_IO_APIC) && defined(CONFIG_SMP) && defined(CONFIG_PCI) +static void __devinit verify_quirk_intel_irqbalance(struct pci_dev *dev) +{ +#ifdef CONFIG_X86_64 + if (genapic != &apic_flat) + panic("APIC mode must be flat on this system\n"); +#elif defined(CONFIG_X86_GENERICARCH) + if (genapic != &apic_default) + panic("APIC mode must be default(flat) on this system. Use apic=default\n"); +#endif +} -static void __devinit quirk_intel_irqbalance(struct pci_dev *dev) +void __init quirk_intel_irqbalance(void) { u8 config, rev; u32 word; @@ -16,18 +29,18 @@ static void __devinit quirk_intel_irqbalance(struct pci_dev *dev) * based platforms. * Disable SW irqbalance/affinity on those platforms. */ - pci_read_config_byte(dev, PCI_CLASS_REVISION, &rev); + rev = read_pci_config_byte(0, 0, 0, PCI_CLASS_REVISION); if (rev > 0x9) return; printk(KERN_INFO "Intel E7520/7320/7525 detected."); - /* enable access to config space*/ - pci_read_config_byte(dev, 0xf4, &config); - pci_write_config_byte(dev, 0xf4, config|0x2); + /* enable access to config space */ + config = read_pci_config_byte(0, 0, 0, 0xf4); + write_pci_config_byte(0, 0, 0, 0xf4, config|0x2); /* read xTPR register */ - raw_pci_ops->read(0, 0, 0x40, 0x4c, 2, &word); + word = read_pci_config_16(0, 0, 0x40, 0x4c); if (!(word & (1 << 13))) { printk(KERN_INFO "Disabling irq balancing and affinity\n"); @@ -37,14 +50,25 @@ static void __devinit quirk_intel_irqbalance(struct pci_dev *dev) noirqdebug_setup(""); #ifdef CONFIG_PROC_FS no_irq_affinity = 1; +#endif +#ifdef CONFIG_HOTPLUG_CPU + printk(KERN_INFO "Disabling cpu hotplug control\n"); + enable_cpu_hotplug = 0; +#endif +#ifdef CONFIG_X86_64 + /* force the genapic selection to flat mode so that + * interrupts can be redirected to more than one CPU. + */ + genapic_force = &apic_flat; #endif } - /* put back the original value for config space*/ + /* put back the original value for config space */ if (!(config & 0x2)) - pci_write_config_byte(dev, 0xf4, config); + write_pci_config_byte(0, 0, 0, 0xf4, config); } -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7320_MCH, quirk_intel_irqbalance); -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7525_MCH, quirk_intel_irqbalance); -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7520_MCH, quirk_intel_irqbalance); +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7320_MCH, verify_quirk_intel_irqbalance); +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7525_MCH, verify_quirk_intel_irqbalance); +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7520_MCH, verify_quirk_intel_irqbalance); + #endif diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index cd7de9c9654..346f27f4c79 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -58,6 +58,7 @@ #include #include #include +#include #include #include @@ -1482,6 +1483,12 @@ int __devinit __cpu_up(unsigned int cpu) cpu_set(cpu, smp_commenced_mask); while (!cpu_isset(cpu, cpu_online_map)) cpu_relax(); + +#ifdef CONFIG_X86_GENERICARCH + if (num_online_cpus() > 8 && genapic == &apic_default) + panic("Default flat APIC routing can't be used with > 8 cpus\n"); +#endif + return 0; } diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c index fb0c6da41b7..829698f6d04 100644 --- a/arch/x86_64/kernel/early-quirks.c +++ b/arch/x86_64/kernel/early-quirks.c @@ -71,6 +71,18 @@ static void ati_bugs(void) { } +static void intel_bugs(void) +{ + u16 device = read_pci_config_16(0, 0, 0, PCI_DEVICE_ID); + +#ifdef CONFIG_SMP + if (device == PCI_DEVICE_ID_INTEL_E7320_MCH || + device == PCI_DEVICE_ID_INTEL_E7520_MCH || + device == PCI_DEVICE_ID_INTEL_E7525_MCH) + quirk_intel_irqbalance(); +#endif +} + struct chipset { u16 vendor; void (*f)(void); @@ -80,6 +92,7 @@ static struct chipset early_qrk[] = { { PCI_VENDOR_ID_NVIDIA, nvidia_bugs }, { PCI_VENDOR_ID_VIA, via_bugs }, { PCI_VENDOR_ID_ATI, ati_bugs }, + { PCI_VENDOR_ID_INTEL, intel_bugs}, {} }; diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index 62c2e747af5..4c161c208d5 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -60,6 +60,7 @@ #include #include #include +#include /* Number of siblings per CPU package */ int smp_num_siblings = 1; @@ -1167,6 +1168,13 @@ int __cpuinit __cpu_up(unsigned int cpu) while (!cpu_isset(cpu, cpu_online_map)) cpu_relax(); + + if (num_online_cpus() > 8 && genapic == &apic_flat) { + printk(KERN_WARNING + "flat APIC routing can't be used with > 8 cpus\n"); + BUG(); + } + err = 0; return err; diff --git a/include/asm-i386/genapic.h b/include/asm-i386/genapic.h index 8ffbb0f0745..fd2be593b06 100644 --- a/include/asm-i386/genapic.h +++ b/include/asm-i386/genapic.h @@ -122,6 +122,6 @@ struct genapic { APICFUNC(phys_pkg_id) \ } -extern struct genapic *genapic; +extern struct genapic *genapic, apic_default; #endif diff --git a/include/asm-i386/irq.h b/include/asm-i386/irq.h index 9e15ce0006e..11761cdaae1 100644 --- a/include/asm-i386/irq.h +++ b/include/asm-i386/irq.h @@ -37,6 +37,8 @@ static __inline__ int irq_canonicalize(int irq) extern int irqbalance_disable(char *str); #endif +extern void quirk_intel_irqbalance(void); + #ifdef CONFIG_HOTPLUG_CPU extern void fixup_irqs(cpumask_t map); #endif diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h index 21fa5afa232..6d324b83897 100644 --- a/include/asm-x86_64/proto.h +++ b/include/asm-x86_64/proto.h @@ -87,6 +87,7 @@ extern void syscall32_cpu_init(void); extern void setup_node_bootmem(int nodeid, unsigned long start, unsigned long end); extern void early_quirks(void); +extern void quirk_intel_irqbalance(void); extern void check_efer(void); extern int unhandled_signal(struct task_struct *tsk, int sig); -- cgit v1.2.3-70-g09d2 From e1cccf48b182dd743c3c83a4fdf8dc570a43b393 Mon Sep 17 00:00:00 2001 From: Artiom Myaskouvskey Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] i386: call efi_get_time during suspend Function efi_get_time called not only during init kernel phase but also during suspend (from get_cmos_time). When it is called from get_cmos_time the corresponding runtime service should be called in virtual and not in physical mode. Signed-off-by: Artiom Myaskouvskey Signed-off-by: Andi Kleen Cc: "Narayanan, Chandramouli" Cc: "Jiossy, Rami" Cc: "Satt, Shai" Cc: Andi Kleen Cc: Matt Domsch Signed-off-by: Andrew Morton --- arch/i386/kernel/efi.c | 17 ++++++++++++----- include/linux/efi.h | 2 +- 2 files changed, 13 insertions(+), 6 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/efi.c b/arch/i386/kernel/efi.c index 8b40648d0ef..b92c7f0a358 100644 --- a/arch/i386/kernel/efi.c +++ b/arch/i386/kernel/efi.c @@ -194,17 +194,24 @@ inline int efi_set_rtc_mmss(unsigned long nowtime) return 0; } /* - * This should only be used during kernel init and before runtime - * services have been remapped, therefore, we'll need to call in physical - * mode. Note, this call isn't used later, so mark it __init. + * This is used during kernel init before runtime + * services have been remapped and also during suspend, therefore, + * we'll need to call both in physical and virtual modes. */ -inline unsigned long __init efi_get_time(void) +inline unsigned long efi_get_time(void) { efi_status_t status; efi_time_t eft; efi_time_cap_t cap; - status = phys_efi_get_time(&eft, &cap); + if (efi.get_time) { + /* if we are in virtual mode use remapped function */ + status = efi.get_time(&eft, &cap); + } else { + /* we are in physical mode */ + status = phys_efi_get_time(&eft, &cap); + } + if (status != EFI_SUCCESS) printk("Oops: efitime: can't read time status: 0x%lx\n",status); diff --git a/include/linux/efi.h b/include/linux/efi.h index 66d621dbcb6..91ecf49fbf2 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -300,7 +300,7 @@ extern int efi_mem_attribute_range (unsigned long phys_addr, unsigned long size, extern int __init efi_uart_console_only (void); extern void efi_initialize_iomem_resources(struct resource *code_resource, struct resource *data_resource); -extern unsigned long __init efi_get_time(void); +extern unsigned long efi_get_time(void); extern int __init efi_set_rtc_mmss(unsigned long nowtime); extern struct efi_memory_map memmap; -- cgit v1.2.3-70-g09d2 From 956fb53197f82257974f1f9835485aeeef4510b3 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] i386: handle a negative return value The Coverity checker noted that bad things might happen if find_isa_irq_apic() returned -1. [akpm@osdl.org: add debugging checks] Signed-off-by: Adrian Bunk Signed-off-by: Andi Kleen Cc: Andi Kleen Acked-by: Ingo Molnar Signed-off-by: Andrew Morton --- arch/i386/kernel/io_apic.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 993150f206e..7bfd6c3ec87 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -2179,9 +2179,15 @@ static inline void unlock_ExtINT_logic(void) unsigned char save_control, save_freq_select; pin = find_isa_irq_pin(8, mp_INT); + if (pin == -1) { + WARN_ON_ONCE(1); + return; + } apic = find_isa_irq_apic(8, mp_INT); - if (pin == -1) + if (apic == -1) { + WARN_ON_ONCE(1); return; + } entry0 = ioapic_read_entry(apic, pin); clear_IO_APIC_pin(apic, pin); -- cgit v1.2.3-70-g09d2 From 7e95b593a1aeb6fe1d3904e799d23a45261f2c19 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] i386: Make irq_vector static irq_vector[] can now become static. Signed-off-by: Adrian Bunk Signed-off-by: Andi Kleen Acked-by: Eric W. Biederman Acked-by: Ingo Molnar Signed-off-by: Andrew Morton --- arch/i386/kernel/io_apic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 7bfd6c3ec87..56f571c9fc0 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -1241,7 +1241,7 @@ static inline int IO_APIC_irq_trigger(int irq) } /* irq_vectors is indexed by the sum of all RTEs in all I/O APICs. */ -u8 irq_vector[NR_IRQ_VECTORS] __read_mostly = { FIRST_DEVICE_VECTOR , 0 }; +static u8 irq_vector[NR_IRQ_VECTORS] __read_mostly = { FIRST_DEVICE_VECTOR , 0 }; static int __assign_irq_vector(int irq) { -- cgit v1.2.3-70-g09d2 From 538f188e03c821c93b355c9fc346806cdd34e286 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] i386: i386 add Intel BTS cpufeature bit and detection (take 2) Here is a small patch for i386 which adds a cpufeature flag and detection code for Intel's Branch Trace Store (BTS) feature. This feature can be found on Intel P4 and Core 2 processors among others. It can also be used by perfmon. changelog: - add CPU_FEATURE_BTS - add Branch Trace Store detection signed-off-by: stephane eranian Signed-off-by: Andi Kleen --- arch/i386/kernel/cpu/intel.c | 2 ++ include/asm-i386/cpufeature.h | 2 ++ 2 files changed, 4 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c index 3ae795e9056..56fe2658495 100644 --- a/arch/i386/kernel/cpu/intel.c +++ b/arch/i386/kernel/cpu/intel.c @@ -199,6 +199,8 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) if (cpu_has_ds) { unsigned int l1; rdmsr(MSR_IA32_MISC_ENABLE, l1, l2); + if (!(l1 & (1<<11))) + set_bit(X86_FEATURE_BTS, c->x86_capability); if (!(l1 & (1<<12))) set_bit(X86_FEATURE_PEBS, c->x86_capability); } diff --git a/include/asm-i386/cpufeature.h b/include/asm-i386/cpufeature.h index 4c83e059228..3f92b94e0d7 100644 --- a/include/asm-i386/cpufeature.h +++ b/include/asm-i386/cpufeature.h @@ -74,6 +74,7 @@ #define X86_FEATURE_FXSAVE_LEAK (3*32+10) /* FXSAVE leaks FOP/FIP/FOP */ #define X86_FEATURE_ARCH_PERFMON (3*32+11) /* Intel Architectural PerfMon */ #define X86_FEATURE_PEBS (3*32+12) /* Precise-Event Based Sampling */ +#define X86_FEATURE_BTS (3*32+13) /* Branch Trace Store */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* Streaming SIMD Extensions-3 */ @@ -138,6 +139,7 @@ #define cpu_has_ds boot_cpu_has(X86_FEATURE_DS) #define cpu_has_pebs boot_cpu_has(X86_FEATURE_PEBS) #define cpu_has_clflush boot_cpu_has(X86_FEATURE_CLFLSH) +#define cpu_has_bts boot_cpu_has(X86_FEATURE_BTS) #endif /* __ASM_I386_CPUFEATURE_H */ -- cgit v1.2.3-70-g09d2 From 9a8cb626a08f2c8251291f3c0a049b29665895d2 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] i386: Avoid boot warning with apic=debug There are two consumers of apic=: the apic debug level and the low level generic architecture code. early_param would warn when the low level code rejected "debug". Avoid this. Signed-off-by: Andi Kleen --- arch/i386/mach-generic/probe.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/mach-generic/probe.c b/arch/i386/mach-generic/probe.c index 94b1fd9cbe3..a7b3999bb37 100644 --- a/arch/i386/mach-generic/probe.c +++ b/arch/i386/mach-generic/probe.c @@ -45,7 +45,9 @@ static int __init parse_apic(char *arg) return 0; } } - return -ENOENT; + + /* Parsed again by __setup for debug/verbose */ + return 0; } early_param("apic", parse_apic); -- cgit v1.2.3-70-g09d2 From f990fff427d68af3e4e1d16fe799c106abc0bf53 Mon Sep 17 00:00:00 2001 From: Karsten Wiese Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] x86: Regard MSRs in lapic_suspend()/lapic_resume() Read/Write APIC_LVTPC and APIC_LVTTHMR only, if get_maxlvt() returns certain values. This is done like everywhere else in i386/kernel/apic.c, so I guess its correct. Suspends/Resumes to disk fine and eleminates an smp_error_interrupt() here on a K8. AK: ported to x86-64 too Signed-off-by: Karsten Wiese Signed-off-by: Andi Kleen --- arch/i386/kernel/apic.c | 22 ++++++++++++++++++---- arch/x86_64/kernel/apic.c | 22 ++++++++++++++++++---- 2 files changed, 36 insertions(+), 8 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c index 2fd4b7d927c..776d9be26af 100644 --- a/arch/i386/kernel/apic.c +++ b/arch/i386/kernel/apic.c @@ -647,23 +647,30 @@ static struct { static int lapic_suspend(struct sys_device *dev, pm_message_t state) { unsigned long flags; + int maxlvt; if (!apic_pm_state.active) return 0; + maxlvt = get_maxlvt(); + apic_pm_state.apic_id = apic_read(APIC_ID); apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI); apic_pm_state.apic_ldr = apic_read(APIC_LDR); apic_pm_state.apic_dfr = apic_read(APIC_DFR); apic_pm_state.apic_spiv = apic_read(APIC_SPIV); apic_pm_state.apic_lvtt = apic_read(APIC_LVTT); - apic_pm_state.apic_lvtpc = apic_read(APIC_LVTPC); + if (maxlvt >= 4) + apic_pm_state.apic_lvtpc = apic_read(APIC_LVTPC); apic_pm_state.apic_lvt0 = apic_read(APIC_LVT0); apic_pm_state.apic_lvt1 = apic_read(APIC_LVT1); apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR); apic_pm_state.apic_tmict = apic_read(APIC_TMICT); apic_pm_state.apic_tdcr = apic_read(APIC_TDCR); - apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); +#ifdef CONFIG_X86_MCE_P4THERMAL + if (maxlvt >= 5) + apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); +#endif local_irq_save(flags); disable_local_APIC(); @@ -675,10 +682,13 @@ static int lapic_resume(struct sys_device *dev) { unsigned int l, h; unsigned long flags; + int maxlvt; if (!apic_pm_state.active) return 0; + maxlvt = get_maxlvt(); + local_irq_save(flags); /* @@ -700,8 +710,12 @@ static int lapic_resume(struct sys_device *dev) apic_write(APIC_SPIV, apic_pm_state.apic_spiv); apic_write(APIC_LVT0, apic_pm_state.apic_lvt0); apic_write(APIC_LVT1, apic_pm_state.apic_lvt1); - apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr); - apic_write(APIC_LVTPC, apic_pm_state.apic_lvtpc); +#ifdef CONFIG_X86_MCE_P4THERMAL + if (maxlvt >= 5) + apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr); +#endif + if (maxlvt >= 4) + apic_write(APIC_LVTPC, apic_pm_state.apic_lvtpc); apic_write(APIC_LVTT, apic_pm_state.apic_lvtt); apic_write(APIC_TDCR, apic_pm_state.apic_tdcr); apic_write(APIC_TMICT, apic_pm_state.apic_tmict); diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c index 5c468971e64..f0b00d8731c 100644 --- a/arch/x86_64/kernel/apic.c +++ b/arch/x86_64/kernel/apic.c @@ -459,23 +459,30 @@ static struct { static int lapic_suspend(struct sys_device *dev, pm_message_t state) { unsigned long flags; + int maxlvt; if (!apic_pm_state.active) return 0; + maxlvt = get_maxlvt(); + apic_pm_state.apic_id = apic_read(APIC_ID); apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI); apic_pm_state.apic_ldr = apic_read(APIC_LDR); apic_pm_state.apic_dfr = apic_read(APIC_DFR); apic_pm_state.apic_spiv = apic_read(APIC_SPIV); apic_pm_state.apic_lvtt = apic_read(APIC_LVTT); - apic_pm_state.apic_lvtpc = apic_read(APIC_LVTPC); + if (maxlvt >= 4) + apic_pm_state.apic_lvtpc = apic_read(APIC_LVTPC); apic_pm_state.apic_lvt0 = apic_read(APIC_LVT0); apic_pm_state.apic_lvt1 = apic_read(APIC_LVT1); apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR); apic_pm_state.apic_tmict = apic_read(APIC_TMICT); apic_pm_state.apic_tdcr = apic_read(APIC_TDCR); - apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); +#ifdef CONFIG_X86_MCE_INTEL + if (maxlvt >= 5) + apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); +#endif local_irq_save(flags); disable_local_APIC(); local_irq_restore(flags); @@ -486,10 +493,13 @@ static int lapic_resume(struct sys_device *dev) { unsigned int l, h; unsigned long flags; + int maxlvt; if (!apic_pm_state.active) return 0; + maxlvt = get_maxlvt(); + local_irq_save(flags); rdmsr(MSR_IA32_APICBASE, l, h); l &= ~MSR_IA32_APICBASE_BASE; @@ -503,8 +513,12 @@ static int lapic_resume(struct sys_device *dev) apic_write(APIC_SPIV, apic_pm_state.apic_spiv); apic_write(APIC_LVT0, apic_pm_state.apic_lvt0); apic_write(APIC_LVT1, apic_pm_state.apic_lvt1); - apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr); - apic_write(APIC_LVTPC, apic_pm_state.apic_lvtpc); +#ifdef CONFIG_X86_MCE_INTEL + if (maxlvt >= 5) + apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr); +#endif + if (maxlvt >= 4) + apic_write(APIC_LVTPC, apic_pm_state.apic_lvtpc); apic_write(APIC_LVTT, apic_pm_state.apic_lvtt); apic_write(APIC_TDCR, apic_pm_state.apic_tdcr); apic_write(APIC_TMICT, apic_pm_state.apic_tmict); -- cgit v1.2.3-70-g09d2 From bf7e6a196318316e921f357557fca9d11d15f486 Mon Sep 17 00:00:00 2001 From: Artiom Myaskouvskey Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] i386: Preserve EFI run time regions with memmap parameter When using memmap kernel parameter in EFI boot we should also add to memory map memory regions of runtime services to enable their mapping later. AK: merged and cleaned up the patch Signed-off-by: Artiom Myaskouvskey Signed-off-by: Andi Kleen --- arch/i386/kernel/e820.c | 60 +++++++++++++++++++++++++++++++++++-------------- arch/i386/mm/init.c | 2 -- include/linux/efi.h | 1 + 3 files changed, 44 insertions(+), 19 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/e820.c b/arch/i386/kernel/e820.c index b704790f796..2f7d0a92fd7 100644 --- a/arch/i386/kernel/e820.c +++ b/arch/i386/kernel/e820.c @@ -742,29 +742,55 @@ void __init print_memory_map(char *who) } } -void __init limit_regions(unsigned long long size) +static __init __always_inline void efi_limit_regions(unsigned long long size) { unsigned long long current_addr = 0; + efi_memory_desc_t *md, *next_md; + void *p, *p1; + int i, j; + + j = 0; + p1 = memmap.map; + for (p = p1, i = 0; p < memmap.map_end; p += memmap.desc_size, i++) { + md = p; + next_md = p1; + current_addr = md->phys_addr + + PFN_PHYS(md->num_pages); + if (is_available_memory(md)) { + if (md->phys_addr >= size) continue; + memcpy(next_md, md, memmap.desc_size); + if (current_addr >= size) { + next_md->num_pages -= + PFN_UP(current_addr-size); + } + p1 += memmap.desc_size; + next_md = p1; + j++; + } else if ((md->attribute & EFI_MEMORY_RUNTIME) == + EFI_MEMORY_RUNTIME) { + /* In order to make runtime services + * available we have to include runtime + * memory regions in memory map */ + memcpy(next_md, md, memmap.desc_size); + p1 += memmap.desc_size; + next_md = p1; + j++; + } + } + memmap.nr_map = j; + memmap.map_end = memmap.map + + (memmap.nr_map * memmap.desc_size); +} + +void __init limit_regions(unsigned long long size) +{ + unsigned long long current_addr; int i; print_memory_map("limit_regions start"); if (efi_enabled) { - efi_memory_desc_t *md; - void *p; - - for (p = memmap.map, i = 0; p < memmap.map_end; - p += memmap.desc_size, i++) { - md = p; - current_addr = md->phys_addr + (md->num_pages << 12); - if (md->type == EFI_CONVENTIONAL_MEMORY) { - if (current_addr >= size) { - md->num_pages -= - (((current_addr-size) + PAGE_SIZE-1) >> PAGE_SHIFT); - memmap.nr_map = i + 1; - return; - } - } - } + efi_limit_regions(size); + return; } for (i = 0; i < e820.nr_map; i++) { current_addr = e820.map[i].addr + e820.map[i].size; diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c index 167416155ee..f4dd048187f 100644 --- a/arch/i386/mm/init.c +++ b/arch/i386/mm/init.c @@ -192,8 +192,6 @@ static inline int page_kills_ppro(unsigned long pagenr) return 0; } -extern int is_available_memory(efi_memory_desc_t *); - int page_is_ram(unsigned long pagenr) { int i; diff --git a/include/linux/efi.h b/include/linux/efi.h index 91ecf49fbf2..df1c91855f0 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -302,6 +302,7 @@ extern void efi_initialize_iomem_resources(struct resource *code_resource, struct resource *data_resource); extern unsigned long efi_get_time(void); extern int __init efi_set_rtc_mmss(unsigned long nowtime); +extern int is_available_memory(efi_memory_desc_t * md); extern struct efi_memory_map memmap; /** -- cgit v1.2.3-70-g09d2 From 6df0532eef0187c293d3ab1d4c158f92e8f24f8a Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] i386: remove duplicate printk We do the exact same printk about a dozen lines above with no intermediate printk's. Signed-off-by: Dave Jones Signed-off-by: Andi Kleen --- arch/i386/kernel/cpu/amd.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/amd.c b/arch/i386/kernel/cpu/amd.c index e4758095d87..41cfea57232 100644 --- a/arch/i386/kernel/cpu/amd.c +++ b/arch/i386/kernel/cpu/amd.c @@ -104,10 +104,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) f_vide(); rdtscl(d2); d = d2-d; - - /* Knock these two lines out if it debugs out ok */ - printk(KERN_INFO "AMD K6 stepping B detected - "); - /* -- cut here -- */ + if (d > 20*K6_BUG_LOOP) printk("system stability may be impaired when more than 32 MB are used.\n"); else -- cgit v1.2.3-70-g09d2 From 0741f4d207a644482d7a040f05cd264c98cf7ee8 Mon Sep 17 00:00:00 2001 From: Chuck Ebbert <76306.1226@compuserve.com> Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] x86: add sysctl for kstack_depth_to_print Add sysctl for kstack_depth_to_print. This lets users change the amount of raw stack data printed in dump_stack() without having to reboot. Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com> Signed-off-by: Andi Kleen --- Documentation/sysctl/kernel.txt | 8 ++++++++ arch/i386/kernel/traps.c | 2 +- arch/x86_64/kernel/traps.c | 2 +- include/asm-x86_64/stacktrace.h | 2 ++ kernel/sysctl.c | 9 +++++++++ 5 files changed, 21 insertions(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt index 0bc7f1e3c9e..5922e84d913 100644 --- a/Documentation/sysctl/kernel.txt +++ b/Documentation/sysctl/kernel.txt @@ -27,6 +27,7 @@ show up in /proc/sys/kernel: - hotplug - java-appletviewer [ binfmt_java, obsolete ] - java-interpreter [ binfmt_java, obsolete ] +- kstack_depth_to_print [ X86 only ] - l2cr [ PPC only ] - modprobe ==> Documentation/kmod.txt - msgmax @@ -170,6 +171,13 @@ This flag controls the L2 cache of G3 processor boards. If ============================================================== +kstack_depth_to_print: (X86 only) + +Controls the number of words to print when dumping the raw +kernel stack. + +============================================================== + osrelease, ostype & version: # cat osrelease diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 7b2f9f02208..1d48a75fa33 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -91,7 +91,7 @@ asmlinkage void alignment_check(void); asmlinkage void spurious_interrupt_bug(void); asmlinkage void machine_check(void); -static int kstack_depth_to_print = 24; +int kstack_depth_to_print = 24; #ifdef CONFIG_STACK_UNWIND static int call_trace = 1; #else diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 264db33476a..75ceccee178 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -108,7 +108,7 @@ static inline void preempt_conditional_cli(struct pt_regs *regs) preempt_enable_no_resched(); } -static int kstack_depth_to_print = 12; +int kstack_depth_to_print = 12; #ifdef CONFIG_STACK_UNWIND static int call_trace = 1; #else diff --git a/include/asm-x86_64/stacktrace.h b/include/asm-x86_64/stacktrace.h index 5eb9799bef7..6f0b5459430 100644 --- a/include/asm-x86_64/stacktrace.h +++ b/include/asm-x86_64/stacktrace.h @@ -1,6 +1,8 @@ #ifndef _ASM_STACKTRACE_H #define _ASM_STACKTRACE_H 1 +extern int kstack_depth_to_print; + /* Generic stack tracer with callbacks */ struct stacktrace_ops { diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 09e569f4792..6fc5e17086f 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -54,6 +54,7 @@ extern int proc_nr_files(ctl_table *table, int write, struct file *filp, #ifdef CONFIG_X86 #include +#include #endif #if defined(CONFIG_SYSCTL) @@ -707,6 +708,14 @@ static ctl_table kern_table[] = { .mode = 0444, .proc_handler = &proc_dointvec, }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "kstack_depth_to_print", + .data = &kstack_depth_to_print, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, #endif #if defined(CONFIG_MMU) { -- cgit v1.2.3-70-g09d2 From a36df98ab1cdd8a9e7daa4c1b5c48ffa2ad6ea09 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] i386: touch softlockup during backtracing Sometimes the soft watchdog fires after we're done oopsing. See http://projects.info-pull.com/mokb/MOKB-25-11-2006.html for an example. AK: changed to touch_nmi_watchdog() Signed-off-by: Dave Jones Signed-off-by: Andi Kleen --- arch/i386/kernel/traps.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 1d48a75fa33..86d8476be4f 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -29,6 +29,7 @@ #include #include #include +#include #ifdef CONFIG_EISA #include @@ -248,6 +249,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, stack = (unsigned long*)context->previous_esp; if (!stack) break; + touch_nmi_watchdog(); } } EXPORT_SYMBOL(dump_trace); -- cgit v1.2.3-70-g09d2 From a1a70c25bed75ed36ed48bbe18b9029428d2452d Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] i386: always enable regparm -mregparm=3 has been enabled by default for some time on i386, and AFAIK there aren't any problems with it left. This patch removes the REGPARM config option and sets -mregparm=3 unconditionally. Signed-off-by: Adrian Bunk Signed-off-by: Andi Kleen --- Documentation/stable_api_nonsense.txt | 3 --- arch/i386/Kconfig | 14 -------------- arch/i386/Makefile | 4 +--- include/asm-i386/module.h | 8 +------- 4 files changed, 2 insertions(+), 27 deletions(-) (limited to 'arch/i386') diff --git a/Documentation/stable_api_nonsense.txt b/Documentation/stable_api_nonsense.txt index f39c9d714db..a2afca3b2ba 100644 --- a/Documentation/stable_api_nonsense.txt +++ b/Documentation/stable_api_nonsense.txt @@ -62,9 +62,6 @@ consider the following facts about the Linux kernel: - different structures can contain different fields - Some functions may not be implemented at all, (i.e. some locks compile away to nothing for non-SMP builds.) - - Parameter passing of variables from function to function can be - done in different ways (the CONFIG_REGPARM option controls - this.) - Memory within the kernel can be aligned in different ways, depending on the build options. - Linux runs on a wide range of different processor architectures. diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index bb1fa061c6c..b6b2df40ca7 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -721,20 +721,6 @@ config BOOT_IOREMAP depends on (((X86_SUMMIT || X86_GENERICARCH) && NUMA) || (X86 && EFI)) default y -config REGPARM - bool "Use register arguments" - default y - help - Compile the kernel with -mregparm=3. This instructs gcc to use - a more efficient function call ABI which passes the first three - arguments of a function call via registers, which results in denser - and faster code. - - If this option is disabled, then the default ABI of passing - arguments via the stack is used. - - If unsure, say Y. - config SECCOMP bool "Enable seccomp to safely compute untrusted bytecode" depends on PROC_FS diff --git a/arch/i386/Makefile b/arch/i386/Makefile index d1aca52bf69..f7ac1aea1d8 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -31,7 +31,7 @@ LDFLAGS_vmlinux := --emit-relocs endif CHECKFLAGS += -D__i386__ -CFLAGS += -pipe -msoft-float +CFLAGS += -pipe -msoft-float -mregparm=3 # prevent gcc from keeping the stack 16 byte aligned CFLAGS += $(call cc-option,-mpreferred-stack-boundary=2) @@ -39,8 +39,6 @@ CFLAGS += $(call cc-option,-mpreferred-stack-boundary=2) # CPU-specific tuning. Anything which can be shared with UML should go here. include $(srctree)/arch/i386/Makefile.cpu -cflags-$(CONFIG_REGPARM) += -mregparm=3 - # temporary until string.h is fixed cflags-y += -ffreestanding diff --git a/include/asm-i386/module.h b/include/asm-i386/module.h index fe5ae42e027..02f8f541cbe 100644 --- a/include/asm-i386/module.h +++ b/include/asm-i386/module.h @@ -62,18 +62,12 @@ struct mod_arch_specific #error unknown processor family #endif -#ifdef CONFIG_REGPARM -#define MODULE_REGPARM "REGPARM " -#else -#define MODULE_REGPARM "" -#endif - #ifdef CONFIG_4KSTACKS #define MODULE_STACKSIZE "4KSTACKS " #else #define MODULE_STACKSIZE "" #endif -#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_REGPARM MODULE_STACKSIZE +#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE #endif /* _ASM_I386_MODULE_H */ -- cgit v1.2.3-70-g09d2 From 359ad0d4015a9ab39243f2ebc4eb07915bd618b2 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 7 Dec 2006 02:14:13 +0100 Subject: [PATCH] unwinder: more sanity checks in Dwarf2 unwinder Tighten the requirements on both input to and output from the Dwarf2 unwinder. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/i386/kernel/traps.c | 7 +++++++ arch/x86_64/kernel/traps.c | 7 +++++++ include/asm-i386/unwind.h | 12 ++++-------- include/asm-x86_64/unwind.h | 8 ++------ kernel/unwind.c | 16 +++++++++++++++- 5 files changed, 35 insertions(+), 15 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 86d8476be4f..c447807e2a4 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -161,12 +161,19 @@ dump_trace_unwind(struct unwind_frame_info *info, void *data) { struct ops_and_data *oad = (struct ops_and_data *)data; int n = 0; + unsigned long sp = UNW_SP(info); + if (arch_unw_user_mode(info)) + return -1; while (unwind(info) == 0 && UNW_PC(info)) { n++; oad->ops->address(oad->data, UNW_PC(info)); if (arch_unw_user_mode(info)) break; + if ((sp & ~(PAGE_SIZE - 1)) == (UNW_SP(info) & ~(PAGE_SIZE - 1)) + && sp > UNW_SP(info)) + break; + sp = UNW_SP(info); } return n; } diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 9864d195c40..4fdd162f0be 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -225,12 +225,19 @@ static int dump_trace_unwind(struct unwind_frame_info *info, void *context) { struct ops_and_data *oad = (struct ops_and_data *)context; int n = 0; + unsigned long sp = UNW_SP(info); + if (arch_unw_user_mode(info)) + return -1; while (unwind(info) == 0 && UNW_PC(info)) { n++; oad->ops->address(oad->data, UNW_PC(info)); if (arch_unw_user_mode(info)) break; + if ((sp & ~(PAGE_SIZE - 1)) == (UNW_SP(info) & ~(PAGE_SIZE - 1)) + && sp > UNW_SP(info)) + break; + sp = UNW_SP(info); } return n; } diff --git a/include/asm-i386/unwind.h b/include/asm-i386/unwind.h index 601fc67bd77..aa2c931e30d 100644 --- a/include/asm-i386/unwind.h +++ b/include/asm-i386/unwind.h @@ -79,17 +79,13 @@ extern asmlinkage int arch_unwind_init_running(struct unwind_frame_info *, void *arg), void *arg); -static inline int arch_unw_user_mode(const struct unwind_frame_info *info) +static inline int arch_unw_user_mode(/*const*/ struct unwind_frame_info *info) { -#if 0 /* This can only work when selector register and EFLAGS saves/restores - are properly annotated (and tracked in UNW_REGISTER_INFO). */ - return user_mode_vm(&info->regs); -#else - return info->regs.eip < PAGE_OFFSET + return user_mode_vm(&info->regs) + || info->regs.eip < PAGE_OFFSET || (info->regs.eip >= __fix_to_virt(FIX_VDSO) - && info->regs.eip < __fix_to_virt(FIX_VDSO) + PAGE_SIZE) + && info->regs.eip < __fix_to_virt(FIX_VDSO) + PAGE_SIZE) || info->regs.esp < PAGE_OFFSET; -#endif } #else diff --git a/include/asm-x86_64/unwind.h b/include/asm-x86_64/unwind.h index 2e7ff10fd77..2f6349e4871 100644 --- a/include/asm-x86_64/unwind.h +++ b/include/asm-x86_64/unwind.h @@ -87,14 +87,10 @@ extern int arch_unwind_init_running(struct unwind_frame_info *, static inline int arch_unw_user_mode(const struct unwind_frame_info *info) { -#if 0 /* This can only work when selector register saves/restores - are properly annotated (and tracked in UNW_REGISTER_INFO). */ - return user_mode(&info->regs); -#else - return (long)info->regs.rip >= 0 + return user_mode(&info->regs) + || (long)info->regs.rip >= 0 || (info->regs.rip >= VSYSCALL_START && info->regs.rip < VSYSCALL_END) || (long)info->regs.rsp >= 0; -#endif } #else diff --git a/kernel/unwind.c b/kernel/unwind.c index af48168a3af..7e721f10410 100644 --- a/kernel/unwind.c +++ b/kernel/unwind.c @@ -95,6 +95,7 @@ static const struct { typedef unsigned long uleb128_t; typedef signed long sleb128_t; +#define sleb128abs __builtin_labs static struct unwind_table { struct { @@ -787,7 +788,7 @@ int unwind(struct unwind_frame_info *frame) #define FRAME_REG(r, t) (((t *)frame)[reg_info[r].offs]) const u32 *fde = NULL, *cie = NULL; const u8 *ptr = NULL, *end = NULL; - unsigned long pc = UNW_PC(frame) - frame->call_frame; + unsigned long pc = UNW_PC(frame) - frame->call_frame, sp; unsigned long startLoc = 0, endLoc = 0, cfa; unsigned i; signed ptrType = -1; @@ -936,6 +937,9 @@ int unwind(struct unwind_frame_info *frame) state.dataAlign = get_sleb128(&ptr, end); if (state.codeAlign == 0 || state.dataAlign == 0 || ptr >= end) cie = NULL; + else if (UNW_PC(frame) % state.codeAlign + || UNW_SP(frame) % sleb128abs(state.dataAlign)) + return -EPERM; else { retAddrReg = state.version <= 1 ? *ptr++ : get_uleb128(&ptr, end); /* skip augmentation */ @@ -968,6 +972,8 @@ int unwind(struct unwind_frame_info *frame) #ifdef CONFIG_FRAME_POINTER unsigned long top, bottom; + if ((UNW_SP(frame) | UNW_FP(frame)) % sizeof(unsigned long)) + return -EPERM; top = STACK_TOP(frame->task); bottom = STACK_BOTTOM(frame->task); # if FRAME_RETADDR_OFFSET < 0 @@ -1018,6 +1024,7 @@ int unwind(struct unwind_frame_info *frame) || state.regs[retAddrReg].where == Nowhere || state.cfa.reg >= ARRAY_SIZE(reg_info) || reg_info[state.cfa.reg].width != sizeof(unsigned long) + || FRAME_REG(state.cfa.reg, unsigned long) % sizeof(unsigned long) || state.cfa.offs % sizeof(unsigned long)) return -EIO; /* update frame */ @@ -1038,6 +1045,8 @@ int unwind(struct unwind_frame_info *frame) #else # define CASES CASE(8); CASE(16); CASE(32); CASE(64) #endif + pc = UNW_PC(frame); + sp = UNW_SP(frame); for (i = 0; i < ARRAY_SIZE(state.regs); ++i) { if (REG_INVALID(i)) { if (state.regs[i].where == Nowhere) @@ -1118,6 +1127,11 @@ int unwind(struct unwind_frame_info *frame) } } + if (UNW_PC(frame) % state.codeAlign + || UNW_SP(frame) % sleb128abs(state.dataAlign) + || (pc == UNW_PC(frame) && sp == UNW_SP(frame))) + return -EIO; + return 0; #undef CASES #undef FRAME_REG -- cgit v1.2.3-70-g09d2 From d5d2448d896fbb9a427ee12eb8e5f6309f2473f7 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 7 Dec 2006 02:14:13 +0100 Subject: [PATCH] x86-64: Fix numaq build error CC arch/i386/boot/compressed/misc.o arch/i386/boot/compressed/misc.c:120: error: static declaration of 'xquad_portio' follows non-static declaration include/asm/io.h:275: error: previous declaration of 'xquad_portio' was here make[2]: *** [arch/i386/boot/compressed/misc.o] Error 1 Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/boot/compressed/misc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c index c6798c75c67..1ce7017fd62 100644 --- a/arch/i386/boot/compressed/misc.c +++ b/arch/i386/boot/compressed/misc.c @@ -196,7 +196,7 @@ static int vidport; static int lines, cols; #ifdef CONFIG_X86_NUMAQ -static void * xquad_portio = NULL; +void *xquad_portio; #endif #include "../../../../lib/inflate.c" -- cgit v1.2.3-70-g09d2 From 9cfa5b5dfafcfe64c1a48906f243cdd302f82471 Mon Sep 17 00:00:00 2001 From: Burman Yan Date: Thu, 7 Dec 2006 02:14:13 +0100 Subject: [PATCH] x86-64: replace kmalloc+memset with kzalloc in MTRR code Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/kernel/cpu/mtrr/if.c | 3 +-- arch/i386/kernel/cpu/mtrr/main.c | 6 ++---- 2 files changed, 3 insertions(+), 6 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/mtrr/if.c b/arch/i386/kernel/cpu/mtrr/if.c index 9753bc6a1f3..5ae1705eafa 100644 --- a/arch/i386/kernel/cpu/mtrr/if.c +++ b/arch/i386/kernel/cpu/mtrr/if.c @@ -44,10 +44,9 @@ mtrr_file_add(unsigned long base, unsigned long size, max = num_var_ranges; if (fcount == NULL) { - fcount = kmalloc(max * sizeof *fcount, GFP_KERNEL); + fcount = kzalloc(max * sizeof *fcount, GFP_KERNEL); if (!fcount) return -ENOMEM; - memset(fcount, 0, max * sizeof *fcount); FILE_FCOUNT(file) = fcount; } if (!page) { diff --git a/arch/i386/kernel/cpu/mtrr/main.c b/arch/i386/kernel/cpu/mtrr/main.c index aeea23e8a05..16bb7ea8714 100644 --- a/arch/i386/kernel/cpu/mtrr/main.c +++ b/arch/i386/kernel/cpu/mtrr/main.c @@ -596,10 +596,8 @@ static int mtrr_save(struct sys_device * sysdev, pm_message_t state) int i; int size = num_var_ranges * sizeof(struct mtrr_value); - mtrr_state = kmalloc(size,GFP_ATOMIC); - if (mtrr_state) - memset(mtrr_state,0,size); - else + mtrr_state = kzalloc(size,GFP_ATOMIC); + if (!mtrr_state) return -ENOMEM; for (i = 0; i < num_var_ranges; i++) { -- cgit v1.2.3-70-g09d2 From f475ff352c5e05d473c462b97c3a13a5b803af5a Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 7 Dec 2006 02:14:13 +0100 Subject: [PATCH] x86-64: remove unused variable Remove unused variable in msr_write(). Reported by D Binderman . Cc: H. Peter Anvin Signed-off-by: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/kernel/msr.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/msr.c b/arch/i386/kernel/msr.c index a773f776c9e..fd45059c908 100644 --- a/arch/i386/kernel/msr.c +++ b/arch/i386/kernel/msr.c @@ -195,7 +195,6 @@ static ssize_t msr_write(struct file *file, const char __user *buf, { const u32 __user *tmp = (const u32 __user *)buf; u32 data[2]; - size_t rv; u32 reg = *ppos; int cpu = iminor(file->f_dentry->d_inode); int err; @@ -203,7 +202,7 @@ static ssize_t msr_write(struct file *file, const char __user *buf, if (count % 8) return -EINVAL; /* Invalid chunk size */ - for (rv = 0; count; count -= 8) { + for (; count; count -= 8) { if (copy_from_user(&data, tmp, 8)) return -EFAULT; err = do_wrmsr(cpu, reg, data[0], data[1]); -- cgit v1.2.3-70-g09d2 From d7fb02712818643bab79a6b3cb8270a747d0227b Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 7 Dec 2006 02:14:19 +0100 Subject: [PATCH] x86-64: remove remaining pc98 code Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/kernel/io_apic.c | 23 ++--------------------- arch/i386/kernel/mpparse.c | 2 -- include/asm-i386/mpspec_def.h | 2 -- 3 files changed, 2 insertions(+), 25 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 56f571c9fc0..7f015a71ab5 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -842,8 +842,7 @@ static int __init find_isa_irq_pin(int irq, int type) if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA || mp_bus_id_to_type[lbus] == MP_BUS_EISA || - mp_bus_id_to_type[lbus] == MP_BUS_MCA || - mp_bus_id_to_type[lbus] == MP_BUS_NEC98 + mp_bus_id_to_type[lbus] == MP_BUS_MCA ) && (mp_irqs[i].mpc_irqtype == type) && (mp_irqs[i].mpc_srcbusirq == irq)) @@ -862,8 +861,7 @@ static int __init find_isa_irq_apic(int irq, int type) if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA || mp_bus_id_to_type[lbus] == MP_BUS_EISA || - mp_bus_id_to_type[lbus] == MP_BUS_MCA || - mp_bus_id_to_type[lbus] == MP_BUS_NEC98 + mp_bus_id_to_type[lbus] == MP_BUS_MCA ) && (mp_irqs[i].mpc_irqtype == type) && (mp_irqs[i].mpc_srcbusirq == irq)) @@ -993,12 +991,6 @@ static int EISA_ELCR(unsigned int irq) #define default_MCA_trigger(idx) (1) #define default_MCA_polarity(idx) (0) -/* NEC98 interrupts are always polarity zero edge triggered, - * when listed as conforming in the MP table. */ - -#define default_NEC98_trigger(idx) (0) -#define default_NEC98_polarity(idx) (0) - static int __init MPBIOS_polarity(int idx) { int bus = mp_irqs[idx].mpc_srcbus; @@ -1033,11 +1025,6 @@ static int __init MPBIOS_polarity(int idx) polarity = default_MCA_polarity(idx); break; } - case MP_BUS_NEC98: /* NEC 98 pin */ - { - polarity = default_NEC98_polarity(idx); - break; - } default: { printk(KERN_WARNING "broken BIOS!!\n"); @@ -1107,11 +1094,6 @@ static int MPBIOS_trigger(int idx) trigger = default_MCA_trigger(idx); break; } - case MP_BUS_NEC98: /* NEC 98 pin */ - { - trigger = default_NEC98_trigger(idx); - break; - } default: { printk(KERN_WARNING "broken BIOS!!\n"); @@ -1173,7 +1155,6 @@ static int pin_2_irq(int idx, int apic, int pin) case MP_BUS_ISA: /* ISA pin */ case MP_BUS_EISA: case MP_BUS_MCA: - case MP_BUS_NEC98: { irq = mp_irqs[idx].mpc_srcbusirq; break; diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c index 442aaf8c77e..2ce67228dff 100644 --- a/arch/i386/kernel/mpparse.c +++ b/arch/i386/kernel/mpparse.c @@ -249,8 +249,6 @@ static void __init MP_bus_info (struct mpc_config_bus *m) mp_current_pci_id++; } else if (strncmp(str, BUSTYPE_MCA, sizeof(BUSTYPE_MCA)-1) == 0) { mp_bus_id_to_type[m->mpc_busid] = MP_BUS_MCA; - } else if (strncmp(str, BUSTYPE_NEC98, sizeof(BUSTYPE_NEC98)-1) == 0) { - mp_bus_id_to_type[m->mpc_busid] = MP_BUS_NEC98; } else { printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str); } diff --git a/include/asm-i386/mpspec_def.h b/include/asm-i386/mpspec_def.h index 76feedf85a8..13bafb16e7a 100644 --- a/include/asm-i386/mpspec_def.h +++ b/include/asm-i386/mpspec_def.h @@ -97,7 +97,6 @@ struct mpc_config_bus #define BUSTYPE_TC "TC" #define BUSTYPE_VME "VME" #define BUSTYPE_XPRESS "XPRESS" -#define BUSTYPE_NEC98 "NEC98" struct mpc_config_ioapic { @@ -182,7 +181,6 @@ enum mp_bustype { MP_BUS_EISA, MP_BUS_PCI, MP_BUS_MCA, - MP_BUS_NEC98 }; #endif -- cgit v1.2.3-70-g09d2 From 116780fc04d9f6cd3ceeab0251681f1dfda53367 Mon Sep 17 00:00:00 2001 From: Burman Yan Date: Thu, 7 Dec 2006 02:14:19 +0100 Subject: [PATCH] i386: replace kmalloc+memset with kzalloc Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/kernel/cpu/intel_cacheinfo.c | 11 +++-------- arch/i386/kernel/mca.c | 13 ++++--------- arch/i386/kernel/pci-dma.c | 6 ++---- arch/i386/mach-voyager/voyager_cat.c | 6 ++---- include/asm-i386/thread_info.h | 10 +--------- 5 files changed, 12 insertions(+), 34 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/intel_cacheinfo.c b/arch/i386/kernel/cpu/intel_cacheinfo.c index 5c43be47587..80b4c5d421b 100644 --- a/arch/i386/kernel/cpu/intel_cacheinfo.c +++ b/arch/i386/kernel/cpu/intel_cacheinfo.c @@ -480,12 +480,10 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu) if (num_cache_leaves == 0) return -ENOENT; - cpuid4_info[cpu] = kmalloc( + cpuid4_info[cpu] = kzalloc( sizeof(struct _cpuid4_info) * num_cache_leaves, GFP_KERNEL); if (unlikely(cpuid4_info[cpu] == NULL)) return -ENOMEM; - memset(cpuid4_info[cpu], 0, - sizeof(struct _cpuid4_info) * num_cache_leaves); oldmask = current->cpus_allowed; retval = set_cpus_allowed(current, cpumask_of_cpu(cpu)); @@ -658,17 +656,14 @@ static int __cpuinit cpuid4_cache_sysfs_init(unsigned int cpu) return -ENOENT; /* Allocate all required memory */ - cache_kobject[cpu] = kmalloc(sizeof(struct kobject), GFP_KERNEL); + cache_kobject[cpu] = kzalloc(sizeof(struct kobject), GFP_KERNEL); if (unlikely(cache_kobject[cpu] == NULL)) goto err_out; - memset(cache_kobject[cpu], 0, sizeof(struct kobject)); - index_kobject[cpu] = kmalloc( + index_kobject[cpu] = kzalloc( sizeof(struct _index_kobject ) * num_cache_leaves, GFP_KERNEL); if (unlikely(index_kobject[cpu] == NULL)) goto err_out; - memset(index_kobject[cpu], 0, - sizeof(struct _index_kobject) * num_cache_leaves); return 0; diff --git a/arch/i386/kernel/mca.c b/arch/i386/kernel/mca.c index eb57a851789..b83672b8952 100644 --- a/arch/i386/kernel/mca.c +++ b/arch/i386/kernel/mca.c @@ -283,10 +283,9 @@ static int __init mca_init(void) bus->f.mca_transform_memory = mca_dummy_transform_memory; /* get the motherboard device */ - mca_dev = kmalloc(sizeof(struct mca_device), GFP_KERNEL); + mca_dev = kzalloc(sizeof(struct mca_device), GFP_KERNEL); if(unlikely(!mca_dev)) goto out_nomem; - memset(mca_dev, 0, sizeof(struct mca_device)); /* * We do not expect many MCA interrupts during initialization, @@ -310,11 +309,9 @@ static int __init mca_init(void) mca_dev->slot = MCA_MOTHERBOARD; mca_register_device(MCA_PRIMARY_BUS, mca_dev); - mca_dev = kmalloc(sizeof(struct mca_device), GFP_ATOMIC); + mca_dev = kzalloc(sizeof(struct mca_device), GFP_ATOMIC); if(unlikely(!mca_dev)) goto out_unlock_nomem; - memset(mca_dev, 0, sizeof(struct mca_device)); - /* Put motherboard into video setup mode, read integrated video * POS registers, and turn motherboard setup off. @@ -349,10 +346,9 @@ static int __init mca_init(void) } if(which_scsi) { /* found a scsi card */ - mca_dev = kmalloc(sizeof(struct mca_device), GFP_ATOMIC); + mca_dev = kzalloc(sizeof(struct mca_device), GFP_ATOMIC); if(unlikely(!mca_dev)) goto out_unlock_nomem; - memset(mca_dev, 0, sizeof(struct mca_device)); for(j = 0; j < 8; j++) mca_dev->pos[j] = pos[j]; @@ -378,10 +374,9 @@ static int __init mca_init(void) if(!mca_read_and_store_pos(pos)) continue; - mca_dev = kmalloc(sizeof(struct mca_device), GFP_ATOMIC); + mca_dev = kzalloc(sizeof(struct mca_device), GFP_ATOMIC); if(unlikely(!mca_dev)) goto out_unlock_nomem; - memset(mca_dev, 0, sizeof(struct mca_device)); for(j=0; j<8; j++) mca_dev->pos[j]=pos[j]; diff --git a/arch/i386/kernel/pci-dma.c b/arch/i386/kernel/pci-dma.c index 5c8c6ef1fc5..41af692c158 100644 --- a/arch/i386/kernel/pci-dma.c +++ b/arch/i386/kernel/pci-dma.c @@ -92,14 +92,12 @@ int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr, if (!mem_base) goto out; - dev->dma_mem = kmalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL); + dev->dma_mem = kzalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL); if (!dev->dma_mem) goto out; - memset(dev->dma_mem, 0, sizeof(struct dma_coherent_mem)); - dev->dma_mem->bitmap = kmalloc(bitmap_size, GFP_KERNEL); + dev->dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL); if (!dev->dma_mem->bitmap) goto free1_out; - memset(dev->dma_mem->bitmap, 0, bitmap_size); dev->dma_mem->virt_base = mem_base; dev->dma_mem->device_base = device_addr; diff --git a/arch/i386/mach-voyager/voyager_cat.c b/arch/i386/mach-voyager/voyager_cat.c index f50c6c6ad68..943a9473b13 100644 --- a/arch/i386/mach-voyager/voyager_cat.c +++ b/arch/i386/mach-voyager/voyager_cat.c @@ -776,7 +776,7 @@ voyager_cat_init(void) for(asic=0; asic < (*modpp)->num_asics; asic++) { int j; voyager_asic_t *asicp = *asicpp - = kmalloc(sizeof(voyager_asic_t), GFP_KERNEL); /*&voyager_asic_storage[asic_count++];*/ + = kzalloc(sizeof(voyager_asic_t), GFP_KERNEL); /*&voyager_asic_storage[asic_count++];*/ voyager_sp_table_t *sp_table; voyager_at_t *asic_table; voyager_jtt_t *jtag_table; @@ -785,7 +785,6 @@ voyager_cat_init(void) printk("**WARNING** kmalloc failure in cat_init\n"); continue; } - memset(asicp, 0, sizeof(voyager_asic_t)); asicpp = &(asicp->next); asicp->asic_location = asic; sp_table = (voyager_sp_table_t *)(eprom_buf + sp_offset); @@ -851,8 +850,7 @@ voyager_cat_init(void) #endif { - struct resource *res = kmalloc(sizeof(struct resource),GFP_KERNEL); - memset(res, 0, sizeof(struct resource)); + struct resource *res = kzalloc(sizeof(struct resource),GFP_KERNEL); res->name = kmalloc(128, GFP_KERNEL); sprintf((char *)res->name, "Voyager %s Quad CPI", cat_module_name(i)); res->start = qic_addr; diff --git a/include/asm-i386/thread_info.h b/include/asm-i386/thread_info.h index 54d6d7aea93..46d32ad9208 100644 --- a/include/asm-i386/thread_info.h +++ b/include/asm-i386/thread_info.h @@ -95,15 +95,7 @@ static inline struct thread_info *current_thread_info(void) /* thread information allocation */ #ifdef CONFIG_DEBUG_STACK_USAGE -#define alloc_thread_info(tsk) \ - ({ \ - struct thread_info *ret; \ - \ - ret = kmalloc(THREAD_SIZE, GFP_KERNEL); \ - if (ret) \ - memset(ret, 0, THREAD_SIZE); \ - ret; \ - }) +#define alloc_thread_info(tsk) kzalloc(THREAD_SIZE, GFP_KERNEL) #else #define alloc_thread_info(tsk) kmalloc(THREAD_SIZE, GFP_KERNEL) #endif -- cgit v1.2.3-70-g09d2 From f6ca8083c261864fc9de94ef99c3311ea259c5c3 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 7 Dec 2006 02:14:19 +0100 Subject: [PATCH] x86-64: Make ix86 default to HIGHMEM4G instead of NOHIGHMEM Generally better for allmodconfig coverage. Signed-off-by: Randy Dunlap Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index b6b2df40ca7..ea70359b02d 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -454,7 +454,8 @@ source "drivers/firmware/Kconfig" choice prompt "High Memory Support" - default NOHIGHMEM + default HIGHMEM4G if !X86_NUMAQ + default HIGHMEM64G if X86_NUMAQ config NOHIGHMEM bool "off" -- cgit v1.2.3-70-g09d2 From 6bedb2ccb02dcc70ffc8eb76df71c746378190ad Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 7 Dec 2006 02:14:19 +0100 Subject: [PATCH] x86-64: don't use set_irq_regs() We don't need to setup _irq_regs in smp_xxx_interrupt (except apic timer). These handlers run with irqs disabled and do not call functions which need "struct pt_regs". Signed-off-by: Oleg Nesterov Signed-off-by: Andi Kleen Acked-by: Ingo Molnar Cc: Andi Kleen Acked-By: David Howells Signed-off-by: Andrew Morton --- arch/i386/kernel/smp.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c index 31e5c6573aa..1b080ab8a49 100644 --- a/arch/i386/kernel/smp.c +++ b/arch/i386/kernel/smp.c @@ -321,7 +321,6 @@ static inline void leave_mm (unsigned long cpu) fastcall void smp_invalidate_interrupt(struct pt_regs *regs) { - struct pt_regs *old_regs = set_irq_regs(regs); unsigned long cpu; cpu = get_cpu(); @@ -352,7 +351,6 @@ fastcall void smp_invalidate_interrupt(struct pt_regs *regs) smp_mb__after_clear_bit(); out: put_cpu_no_resched(); - set_irq_regs(old_regs); } static void flush_tlb_others(cpumask_t cpumask, struct mm_struct *mm, @@ -607,14 +605,11 @@ void smp_send_stop(void) */ fastcall void smp_reschedule_interrupt(struct pt_regs *regs) { - struct pt_regs *old_regs = set_irq_regs(regs); ack_APIC_irq(); - set_irq_regs(old_regs); } fastcall void smp_call_function_interrupt(struct pt_regs *regs) { - struct pt_regs *old_regs = set_irq_regs(regs); void (*func) (void *info) = call_data->func; void *info = call_data->info; int wait = call_data->wait; @@ -637,7 +632,6 @@ fastcall void smp_call_function_interrupt(struct pt_regs *regs) mb(); atomic_inc(&call_data->finished); } - set_irq_regs(old_regs); } /* -- cgit v1.2.3-70-g09d2 From b65780e123ba9b762276482bbfb52836e4d41fd9 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 7 Dec 2006 02:14:19 +0100 Subject: [PATCH] unwinder: move .eh_frame to RODATA The .eh_frame section contents is never written to, so it can as well benefit from CONFIG_DEBUG_RODATA. Diff-ed against firstfloor tree. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/i386/kernel/vmlinux.lds.S | 9 --------- arch/x86_64/kernel/vmlinux.lds.S | 9 --------- include/asm-generic/vmlinux.lds.h | 17 ++++++++++++----- kernel/unwind.c | 2 +- 4 files changed, 13 insertions(+), 24 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 25581e87c60..56e6ad5cb04 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -102,15 +102,6 @@ SECTIONS _edata = .; /* End of data section */ } -#ifdef CONFIG_STACK_UNWIND - . = ALIGN(4); - .eh_frame : AT(ADDR(.eh_frame) - LOAD_OFFSET) { - __start_unwind = .; - *(.eh_frame) - __end_unwind = .; - } -#endif - . = ALIGN(THREAD_SIZE); /* init_task */ .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) { *(.data.init_task) diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index d9534e750d4..6a1f8f491e5 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -51,15 +51,6 @@ SECTIONS RODATA -#ifdef CONFIG_STACK_UNWIND - . = ALIGN(8); - .eh_frame : AT(ADDR(.eh_frame) - LOAD_OFFSET) { - __start_unwind = .; - *(.eh_frame) - __end_unwind = .; - } -#endif - . = ALIGN(PAGE_SIZE); /* Align data segment to page size boundary */ /* Data */ .data : AT(ADDR(.data) - LOAD_OFFSET) { diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 9f4747780da..4d4c62d1105 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -119,8 +119,7 @@ *(__ksymtab_strings) \ } \ \ - /* Unwind data binary search table */ \ - EH_FRAME_HDR \ + EH_FRAME \ \ /* Built-in module parameters. */ \ __param : AT(ADDR(__param) - LOAD_OFFSET) { \ @@ -162,15 +161,23 @@ VMLINUX_SYMBOL(__kprobes_text_end) = .; #ifdef CONFIG_STACK_UNWIND - /* Unwind data binary search table */ -#define EH_FRAME_HDR \ +#define EH_FRAME \ + /* Unwind data binary search table */ \ + . = ALIGN(8); \ .eh_frame_hdr : AT(ADDR(.eh_frame_hdr) - LOAD_OFFSET) { \ VMLINUX_SYMBOL(__start_unwind_hdr) = .; \ *(.eh_frame_hdr) \ VMLINUX_SYMBOL(__end_unwind_hdr) = .; \ + } \ + /* Unwind data */ \ + . = ALIGN(8); \ + .eh_frame : AT(ADDR(.eh_frame) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start_unwind) = .; \ + *(.eh_frame) \ + VMLINUX_SYMBOL(__end_unwind) = .; \ } #else -#define EH_FRAME_HDR +#define EH_FRAME #endif /* DWARF debug sections. diff --git a/kernel/unwind.c b/kernel/unwind.c index 08645aa7c2d..09c26132924 100644 --- a/kernel/unwind.c +++ b/kernel/unwind.c @@ -19,7 +19,7 @@ #include #include -extern char __start_unwind[], __end_unwind[]; +extern const char __start_unwind[], __end_unwind[]; extern const u8 __start_unwind_hdr[], __end_unwind_hdr[]; #define MAX_STACK_DEPTH 8 -- cgit v1.2.3-70-g09d2 From d9408cefe677636bc1c100fdcfac0b2ab9ff87bf Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 7 Dec 2006 02:14:19 +0100 Subject: [PATCH] i386: Clean up smp_tune_scheduling() - remove the write-only local variable "bandwidth" - don't set "max_cache_size" in the (cachesize < 0) case: that's already handled in kernel/sched.c:measure_migration_cost() Signed-off-by: Adrian Bunk Signed-off-by: Andi Kleen Acked-by: Ingo Molnar Signed-off-by: Andrew Morton --- arch/i386/kernel/smpboot.c | 29 +++++------------------------ 1 file changed, 5 insertions(+), 24 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 346f27f4c79..b4e6f32de45 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -1130,34 +1130,15 @@ exit: } #endif -static void smp_tune_scheduling (void) +static void smp_tune_scheduling(void) { unsigned long cachesize; /* kB */ - unsigned long bandwidth = 350; /* MB/s */ - /* - * Rough estimation for SMP scheduling, this is the number of - * cycles it takes for a fully memory-limited process to flush - * the SMP-local cache. - * - * (For a P5 this pretty much means we will choose another idle - * CPU almost always at wakeup time (this is due to the small - * L1 cache), on PIIs it's around 50-100 usecs, depending on - * the cache size) - */ - if (!cpu_khz) { - /* - * this basically disables processor-affinity - * scheduling on SMP without a TSC. - */ - return; - } else { + if (cpu_khz) { cachesize = boot_cpu_data.x86_cache_size; - if (cachesize == -1) { - cachesize = 16; /* Pentiums, 2x8kB cache */ - bandwidth = 100; - } - max_cache_size = cachesize * 1024; + + if (cachesize > 0) + max_cache_size = cachesize * 1024; } } -- cgit v1.2.3-70-g09d2 From 39dde65c9940c97fcd178a3d2b1c57ed8b7b68aa Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Wed, 6 Dec 2006 20:32:03 -0800 Subject: [PATCH] shared page table for hugetlb page Following up with the work on shared page table done by Dave McCracken. This set of patch target shared page table for hugetlb memory only. The shared page table is particular useful in the situation of large number of independent processes sharing large shared memory segments. In the normal page case, the amount of memory saved from process' page table is quite significant. For hugetlb, the saving on page table memory is not the primary objective (as hugetlb itself already cuts down page table overhead significantly), instead, the purpose of using shared page table on hugetlb is to allow faster TLB refill and smaller cache pollution upon TLB miss. With PT sharing, pte entries are shared among hundreds of processes, the cache consumption used by all the page table is smaller and in return, application gets much higher cache hit ratio. One other effect is that cache hit ratio with hardware page walker hitting on pte in cache will be higher and this helps to reduce tlb miss latency. These two effects contribute to higher application performance. Signed-off-by: Ken Chen Acked-by: Hugh Dickins Cc: Dave McCracken Cc: William Lee Irwin III Cc: "Luck, Tony" Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: David Gibson Cc: Adam Litke Cc: Paul Mundt Cc: "David S. Miller" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/hugetlbpage.c | 112 +++++++++++++++++++++++++++++++++++++++++- arch/ia64/mm/hugetlbpage.c | 5 ++ arch/powerpc/mm/hugetlbpage.c | 5 ++ arch/sh/mm/hugetlbpage.c | 5 ++ arch/sh64/mm/hugetlbpage.c | 5 ++ arch/sparc64/mm/hugetlbpage.c | 5 ++ include/linux/hugetlb.h | 1 + mm/hugetlb.c | 7 +++ 8 files changed, 144 insertions(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/mm/hugetlbpage.c b/arch/i386/mm/hugetlbpage.c index 1719a8141f8..34728e4afe4 100644 --- a/arch/i386/mm/hugetlbpage.c +++ b/arch/i386/mm/hugetlbpage.c @@ -17,6 +17,113 @@ #include #include +static unsigned long page_table_shareable(struct vm_area_struct *svma, + struct vm_area_struct *vma, + unsigned long addr, pgoff_t idx) +{ + unsigned long saddr = ((idx - svma->vm_pgoff) << PAGE_SHIFT) + + svma->vm_start; + unsigned long sbase = saddr & PUD_MASK; + unsigned long s_end = sbase + PUD_SIZE; + + /* + * match the virtual addresses, permission and the alignment of the + * page table page. + */ + if (pmd_index(addr) != pmd_index(saddr) || + vma->vm_flags != svma->vm_flags || + sbase < svma->vm_start || svma->vm_end < s_end) + return 0; + + return saddr; +} + +static int vma_shareable(struct vm_area_struct *vma, unsigned long addr) +{ + unsigned long base = addr & PUD_MASK; + unsigned long end = base + PUD_SIZE; + + /* + * check on proper vm_flags and page table alignment + */ + if (vma->vm_flags & VM_MAYSHARE && + vma->vm_start <= base && end <= vma->vm_end) + return 1; + return 0; +} + +/* + * search for a shareable pmd page for hugetlb. + */ +static void huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud) +{ + struct vm_area_struct *vma = find_vma(mm, addr); + struct address_space *mapping = vma->vm_file->f_mapping; + pgoff_t idx = ((addr - vma->vm_start) >> PAGE_SHIFT) + + vma->vm_pgoff; + struct prio_tree_iter iter; + struct vm_area_struct *svma; + unsigned long saddr; + pte_t *spte = NULL; + + if (!vma_shareable(vma, addr)) + return; + + spin_lock(&mapping->i_mmap_lock); + vma_prio_tree_foreach(svma, &iter, &mapping->i_mmap, idx, idx) { + if (svma == vma) + continue; + + saddr = page_table_shareable(svma, vma, addr, idx); + if (saddr) { + spte = huge_pte_offset(svma->vm_mm, saddr); + if (spte) { + get_page(virt_to_page(spte)); + break; + } + } + } + + if (!spte) + goto out; + + spin_lock(&mm->page_table_lock); + if (pud_none(*pud)) + pud_populate(mm, pud, (unsigned long) spte & PAGE_MASK); + else + put_page(virt_to_page(spte)); + spin_unlock(&mm->page_table_lock); +out: + spin_unlock(&mapping->i_mmap_lock); +} + +/* + * unmap huge page backed by shared pte. + * + * Hugetlb pte page is ref counted at the time of mapping. If pte is shared + * indicated by page_count > 1, unmap is achieved by clearing pud and + * decrementing the ref count. If count == 1, the pte page is not shared. + * + * called with vma->vm_mm->page_table_lock held. + * + * returns: 1 successfully unmapped a shared pte page + * 0 the underlying pte page is not shared, or it is the last user + */ +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + pgd_t *pgd = pgd_offset(mm, *addr); + pud_t *pud = pud_offset(pgd, *addr); + + BUG_ON(page_count(virt_to_page(ptep)) == 0); + if (page_count(virt_to_page(ptep)) == 1) + return 0; + + pud_clear(pud); + put_page(virt_to_page(ptep)); + *addr = ALIGN(*addr, HPAGE_SIZE * PTRS_PER_PTE) - HPAGE_SIZE; + return 1; +} + pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; @@ -25,8 +132,11 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) pgd = pgd_offset(mm, addr); pud = pud_alloc(mm, pgd, addr); - if (pud) + if (pud) { + if (pud_none(*pud)) + huge_pmd_share(mm, addr, pud); pte = (pte_t *) pmd_alloc(mm, pud, addr); + } BUG_ON(pte && !pte_none(*pte) && !pte_huge(*pte)); return pte; diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index f3a9585e98a..0c7e94edc20 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -64,6 +64,11 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr) return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + #define mk_pte_huge(entry) { pte_val(entry) |= _PAGE_P; } /* diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 506d89768d4..424a8f57e15 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -146,6 +146,11 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) return hugepte_offset(hpdp, addr); } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + static void free_hugepte_range(struct mmu_gather *tlb, hugepd_t *hpdp) { pte_t *hugepte = hugepd_page(*hpdp); diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c index 329059d6b54..cf2c2ee35a3 100644 --- a/arch/sh/mm/hugetlbpage.c +++ b/arch/sh/mm/hugetlbpage.c @@ -63,6 +63,11 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { diff --git a/arch/sh64/mm/hugetlbpage.c b/arch/sh64/mm/hugetlbpage.c index 187cf01750b..4b455f61114 100644 --- a/arch/sh64/mm/hugetlbpage.c +++ b/arch/sh64/mm/hugetlbpage.c @@ -53,6 +53,11 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t entry) { diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c index 53b9b1f528e..33fd0b265e7 100644 --- a/arch/sparc64/mm/hugetlbpage.c +++ b/arch/sparc64/mm/hugetlbpage.c @@ -235,6 +235,11 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t entry) { diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index ace64e57e17..a60995afe33 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -35,6 +35,7 @@ extern int sysctl_hugetlb_shm_group; pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr); pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr); +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep); struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write); struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, diff --git a/mm/hugetlb.c b/mm/hugetlb.c index f7355bf2f28..9244971b679 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -386,6 +386,9 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, if (!ptep) continue; + if (huge_pmd_unshare(mm, &address, ptep)) + continue; + pte = huge_ptep_get_and_clear(mm, address, ptep); if (pte_none(pte)) continue; @@ -658,11 +661,14 @@ void hugetlb_change_protection(struct vm_area_struct *vma, BUG_ON(address >= end); flush_cache_range(vma, address, end); + spin_lock(&vma->vm_file->f_mapping->i_mmap_lock); spin_lock(&mm->page_table_lock); for (; address < end; address += HPAGE_SIZE) { ptep = huge_pte_offset(mm, address); if (!ptep) continue; + if (huge_pmd_unshare(mm, &address, ptep)) + continue; if (!pte_none(*ptep)) { pte = huge_ptep_get_and_clear(mm, address, ptep); pte = pte_mkhuge(pte_modify(pte, newprot)); @@ -671,6 +677,7 @@ void hugetlb_change_protection(struct vm_area_struct *vma, } } spin_unlock(&mm->page_table_lock); + spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock); flush_tlb_range(vma, start, end); } -- cgit v1.2.3-70-g09d2 From a866374aecc90c7d90619727ccd851ac096b2fc7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 6 Dec 2006 20:32:20 -0800 Subject: [PATCH] mm: pagefault_{disable,enable}() Introduce pagefault_{disable,enable}() and use these where previously we did manual preempt increments/decrements to make the pagefault handler do the atomic thing. Currently they still rely on the increased preempt count, but do not rely on the disabled preemption, this might go away in the future. (NOTE: the extra barrier() in pagefault_disable might fix some holes on machines which have too many registers for their own good) [heiko.carstens@de.ibm.com: s390 fix] Signed-off-by: Peter Zijlstra Acked-by: Nick Piggin Cc: Martin Schwidefsky Signed-off-by: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/kernel/futex.c | 4 ++-- arch/i386/mm/highmem.c | 10 ++++------ arch/mips/mm/highmem.c | 10 ++++------ arch/s390/lib/uaccess_std.c | 6 +++--- arch/sparc/mm/highmem.c | 8 +++----- include/asm-frv/highmem.h | 5 ++--- include/asm-generic/futex.h | 4 ++-- include/asm-i386/futex.h | 4 ++-- include/asm-ia64/futex.h | 4 ++-- include/asm-mips/futex.h | 4 ++-- include/asm-parisc/futex.h | 4 ++-- include/asm-powerpc/futex.h | 4 ++-- include/asm-ppc/highmem.h | 8 +++----- include/asm-sparc64/futex.h | 4 ++-- include/asm-x86_64/futex.h | 4 ++-- include/linux/uaccess.h | 39 +++++++++++++++++++++++++++++++++++++-- kernel/futex.c | 28 ++++++++++++++-------------- 17 files changed, 88 insertions(+), 62 deletions(-) (limited to 'arch/i386') diff --git a/arch/frv/kernel/futex.c b/arch/frv/kernel/futex.c index eae874a970c..53dc5ed1ebd 100644 --- a/arch/frv/kernel/futex.c +++ b/arch/frv/kernel/futex.c @@ -200,7 +200,7 @@ int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -223,7 +223,7 @@ int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) break; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/arch/i386/mm/highmem.c b/arch/i386/mm/highmem.c index f9f647cdbc7..178bbfe6cba 100644 --- a/arch/i386/mm/highmem.c +++ b/arch/i386/mm/highmem.c @@ -32,7 +32,7 @@ void *kmap_atomic(struct page *page, enum km_type type) unsigned long vaddr; /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */ - inc_preempt_count(); + pagefault_disable(); if (!PageHighMem(page)) return page_address(page); @@ -52,8 +52,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type) #ifdef CONFIG_DEBUG_HIGHMEM if (vaddr >= PAGE_OFFSET && vaddr < (unsigned long)high_memory) { - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); return; } @@ -68,8 +67,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type) */ kpte_clear_flush(kmap_pte-idx, vaddr); - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } /* This is the same as kmap_atomic() but can map memory that doesn't @@ -80,7 +78,7 @@ void *kmap_atomic_pfn(unsigned long pfn, enum km_type type) enum fixed_addresses idx; unsigned long vaddr; - inc_preempt_count(); + pagefault_disable(); idx = type + KM_TYPE_NR*smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); diff --git a/arch/mips/mm/highmem.c b/arch/mips/mm/highmem.c index 99ebf3ccc22..675502ada5a 100644 --- a/arch/mips/mm/highmem.c +++ b/arch/mips/mm/highmem.c @@ -39,7 +39,7 @@ void *__kmap_atomic(struct page *page, enum km_type type) unsigned long vaddr; /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */ - inc_preempt_count(); + pagefault_disable(); if (!PageHighMem(page)) return page_address(page); @@ -62,8 +62,7 @@ void __kunmap_atomic(void *kvaddr, enum km_type type) enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id(); if (vaddr < FIXADDR_START) { // FIXME - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); return; } @@ -78,8 +77,7 @@ void __kunmap_atomic(void *kvaddr, enum km_type type) local_flush_tlb_one(vaddr); #endif - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } #ifndef CONFIG_LIMITED_DMA @@ -92,7 +90,7 @@ void *kmap_atomic_pfn(unsigned long pfn, enum km_type type) enum fixed_addresses idx; unsigned long vaddr; - inc_preempt_count(); + pagefault_disable(); idx = type + KM_TYPE_NR*smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); diff --git a/arch/s390/lib/uaccess_std.c b/arch/s390/lib/uaccess_std.c index 2d549ed2e11..bbaca66fa29 100644 --- a/arch/s390/lib/uaccess_std.c +++ b/arch/s390/lib/uaccess_std.c @@ -11,7 +11,7 @@ #include #include -#include +#include #include #ifndef __s390x__ @@ -258,7 +258,7 @@ int futex_atomic_op(int op, int __user *uaddr, int oparg, int *old) { int oldval = 0, newval, ret; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -284,7 +284,7 @@ int futex_atomic_op(int op, int __user *uaddr, int oparg, int *old) default: ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); *old = oldval; return ret; } diff --git a/arch/sparc/mm/highmem.c b/arch/sparc/mm/highmem.c index 4d8ed9c6518..01fc6c25429 100644 --- a/arch/sparc/mm/highmem.c +++ b/arch/sparc/mm/highmem.c @@ -35,7 +35,7 @@ void *kmap_atomic(struct page *page, enum km_type type) unsigned long vaddr; /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */ - inc_preempt_count(); + pagefault_disable(); if (!PageHighMem(page)) return page_address(page); @@ -70,8 +70,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type) unsigned long idx = type + KM_TYPE_NR*smp_processor_id(); if (vaddr < FIXADDR_START) { // FIXME - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); return; } @@ -97,8 +96,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type) #endif #endif - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } /* We may be fed a pagetable here by ptep_to_xxx and others. */ diff --git a/include/asm-frv/highmem.h b/include/asm-frv/highmem.h index 0f390f41f81..ff4d6cdeb15 100644 --- a/include/asm-frv/highmem.h +++ b/include/asm-frv/highmem.h @@ -115,7 +115,7 @@ static inline void *kmap_atomic(struct page *page, enum km_type type) { unsigned long paddr; - inc_preempt_count(); + pagefault_disable(); paddr = page_to_phys(page); switch (type) { @@ -170,8 +170,7 @@ static inline void kunmap_atomic(void *kvaddr, enum km_type type) default: BUG(); } - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } #endif /* !__ASSEMBLY__ */ diff --git a/include/asm-generic/futex.h b/include/asm-generic/futex.h index df893c16031..f422df0956a 100644 --- a/include/asm-generic/futex.h +++ b/include/asm-generic/futex.h @@ -21,7 +21,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -33,7 +33,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-i386/futex.h b/include/asm-i386/futex.h index 946d97cfea2..438ef0ec710 100644 --- a/include/asm-i386/futex.h +++ b/include/asm-i386/futex.h @@ -56,7 +56,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); if (op == FUTEX_OP_SET) __futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg); @@ -88,7 +88,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) } } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-ia64/futex.h b/include/asm-ia64/futex.h index 07d77f3a8cb..8a98a265413 100644 --- a/include/asm-ia64/futex.h +++ b/include/asm-ia64/futex.h @@ -59,7 +59,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -83,7 +83,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-mips/futex.h b/include/asm-mips/futex.h index 927a216bd53..47e5679c235 100644 --- a/include/asm-mips/futex.h +++ b/include/asm-mips/futex.h @@ -88,7 +88,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -115,7 +115,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-parisc/futex.h b/include/asm-parisc/futex.h index d84bbb283fd..dbee6e60aa8 100644 --- a/include/asm-parisc/futex.h +++ b/include/asm-parisc/futex.h @@ -21,7 +21,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -33,7 +33,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-powerpc/futex.h b/include/asm-powerpc/futex.h index 936422e5489..3f3673fd3ff 100644 --- a/include/asm-powerpc/futex.h +++ b/include/asm-powerpc/futex.h @@ -43,7 +43,7 @@ static inline int futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -65,7 +65,7 @@ static inline int futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-ppc/highmem.h b/include/asm-ppc/highmem.h index 1d2c4ef81c2..f7b21ee302b 100644 --- a/include/asm-ppc/highmem.h +++ b/include/asm-ppc/highmem.h @@ -79,7 +79,7 @@ static inline void *kmap_atomic(struct page *page, enum km_type type) unsigned long vaddr; /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */ - inc_preempt_count(); + pagefault_disable(); if (!PageHighMem(page)) return page_address(page); @@ -101,8 +101,7 @@ static inline void kunmap_atomic(void *kvaddr, enum km_type type) unsigned int idx = type + KM_TYPE_NR*smp_processor_id(); if (vaddr < KMAP_FIX_BEGIN) { // FIXME - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); return; } @@ -115,8 +114,7 @@ static inline void kunmap_atomic(void *kvaddr, enum km_type type) pte_clear(&init_mm, vaddr, kmap_pte+idx); flush_tlb_page(NULL, vaddr); #endif - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } static inline struct page *kmap_atomic_to_page(void *ptr) diff --git a/include/asm-sparc64/futex.h b/include/asm-sparc64/futex.h index 7392fc4a954..876312fe82c 100644 --- a/include/asm-sparc64/futex.h +++ b/include/asm-sparc64/futex.h @@ -45,7 +45,7 @@ static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -67,7 +67,7 @@ static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-x86_64/futex.h b/include/asm-x86_64/futex.h index 9804bf07b09..5cdfb08013c 100644 --- a/include/asm-x86_64/futex.h +++ b/include/asm-x86_64/futex.h @@ -55,7 +55,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -78,7 +78,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h index a48d7f11c7b..67918c22339 100644 --- a/include/linux/uaccess.h +++ b/include/linux/uaccess.h @@ -1,8 +1,43 @@ #ifndef __LINUX_UACCESS_H__ #define __LINUX_UACCESS_H__ +#include #include +/* + * These routines enable/disable the pagefault handler in that + * it will not take any locks and go straight to the fixup table. + * + * They have great resemblance to the preempt_disable/enable calls + * and in fact they are identical; this is because currently there is + * no other way to make the pagefault handlers do this. So we do + * disable preemption but we don't necessarily care about that. + */ +static inline void pagefault_disable(void) +{ + inc_preempt_count(); + /* + * make sure to have issued the store before a pagefault + * can hit. + */ + barrier(); +} + +static inline void pagefault_enable(void) +{ + /* + * make sure to issue those last loads/stores before enabling + * the pagefault handler again. + */ + barrier(); + dec_preempt_count(); + /* + * make sure we do.. + */ + barrier(); + preempt_check_resched(); +} + #ifndef ARCH_HAS_NOCACHE_UACCESS static inline unsigned long __copy_from_user_inatomic_nocache(void *to, @@ -35,9 +70,9 @@ static inline unsigned long __copy_from_user_nocache(void *to, ({ \ long ret; \ \ - inc_preempt_count(); \ + pagefault_disable(); \ ret = __get_user(retval, addr); \ - dec_preempt_count(); \ + pagefault_enable(); \ ret; \ }) diff --git a/kernel/futex.c b/kernel/futex.c index 93ef30ba209..af7b81cbde3 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -282,9 +282,9 @@ static inline int get_futex_value_locked(u32 *dest, u32 __user *from) { int ret; - inc_preempt_count(); + pagefault_disable(); ret = __copy_from_user_inatomic(dest, from, sizeof(u32)); - dec_preempt_count(); + pagefault_enable(); return ret ? -EFAULT : 0; } @@ -585,9 +585,9 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) if (!(uval & FUTEX_OWNER_DIED)) { newval = FUTEX_WAITERS | new_owner->pid; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); + pagefault_enable(); if (curval == -EFAULT) return -EFAULT; if (curval != uval) @@ -618,9 +618,9 @@ static int unlock_futex_pi(u32 __user *uaddr, u32 uval) * There is no waiter, so we unlock the futex. The owner died * bit has not to be preserved here. We are the owner: */ - inc_preempt_count(); + pagefault_disable(); oldval = futex_atomic_cmpxchg_inatomic(uaddr, uval, 0); - dec_preempt_count(); + pagefault_enable(); if (oldval == -EFAULT) return oldval; @@ -1158,9 +1158,9 @@ static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, */ newval = current->pid; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, 0, newval); - dec_preempt_count(); + pagefault_enable(); if (unlikely(curval == -EFAULT)) goto uaddr_faulted; @@ -1183,9 +1183,9 @@ static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, uval = curval; newval = uval | FUTEX_WAITERS; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); + pagefault_enable(); if (unlikely(curval == -EFAULT)) goto uaddr_faulted; @@ -1215,10 +1215,10 @@ static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, newval = current->pid | FUTEX_OWNER_DIED | FUTEX_WAITERS; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); + pagefault_enable(); if (unlikely(curval == -EFAULT)) goto uaddr_faulted; @@ -1390,9 +1390,9 @@ retry_locked: * anyone else up: */ if (!(uval & FUTEX_OWNER_DIED)) { - inc_preempt_count(); + pagefault_disable(); uval = futex_atomic_cmpxchg_inatomic(uaddr, current->pid, 0); - dec_preempt_count(); + pagefault_enable(); } if (unlikely(uval == -EFAULT)) -- cgit v1.2.3-70-g09d2 From 3b17979bda74493633364c2c263b452b7788e350 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 6 Dec 2006 20:32:22 -0800 Subject: [PATCH] Fix kunmap_atomic's use of kpte_clear_flush() kunmap_atomic() will call kpte_clear_flush with vaddr/ptep arguments which don't correspond if the vaddr is just a normal lowmem address (ie, not in the KMAP area). This patch makes sure that the pte is only cleared if kmap area was actually used for the mapping. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Rusty Russell Cc: Zachary Amsden Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/highmem.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/mm/highmem.c b/arch/i386/mm/highmem.c index 178bbfe6cba..e0fa6cb655a 100644 --- a/arch/i386/mm/highmem.c +++ b/arch/i386/mm/highmem.c @@ -50,22 +50,20 @@ void kunmap_atomic(void *kvaddr, enum km_type type) unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK; enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id(); -#ifdef CONFIG_DEBUG_HIGHMEM - if (vaddr >= PAGE_OFFSET && vaddr < (unsigned long)high_memory) { - pagefault_enable(); - return; - } - - if (vaddr != __fix_to_virt(FIX_KMAP_BEGIN+idx)) - BUG(); -#endif /* * Force other mappings to Oops if they'll try to access this pte * without first remap it. Keeping stale mappings around is a bad idea * also, in case the page changes cacheability attributes or becomes * a protected page in a hypervisor. */ - kpte_clear_flush(kmap_pte-idx, vaddr); + if (vaddr == __fix_to_virt(FIX_KMAP_BEGIN+idx)) + kpte_clear_flush(kmap_pte-idx, vaddr); + else { +#ifdef CONFIG_DEBUG_HIGHMEM + BUG_ON(vaddr < PAGE_OFFSET); + BUG_ON(vaddr >= (unsigned long)high_memory); +#endif + } pagefault_enable(); } -- cgit v1.2.3-70-g09d2 From a120586873d3d64de93bd6d593d237e131994e58 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 6 Dec 2006 20:32:37 -0800 Subject: [PATCH] Allow NULL pointers in percpu_free The patch (as824b) makes percpu_free() ignore NULL arguments, as one would expect for a deallocation routine. (Note that free_percpu is #defined as percpu_free in include/linux/percpu.h.) A few callers are updated to remove now-unneeded tests for NULL. A few other callers already seem to assume that passing a NULL pointer to percpu_free() is okay! The patch also removes an unnecessary NULL check in percpu_depopulate(). Signed-off-by: Alan Stern Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/acpi/cstate.c | 6 ++---- block/blktrace.c | 3 +-- mm/allocpercpu.c | 9 +++++---- net/ipv6/af_inet6.c | 6 ++---- 4 files changed, 10 insertions(+), 14 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/acpi/cstate.c b/arch/i386/kernel/acpi/cstate.c index 4664b55f623..12e937c1ce4 100644 --- a/arch/i386/kernel/acpi/cstate.c +++ b/arch/i386/kernel/acpi/cstate.c @@ -156,10 +156,8 @@ static int __init ffh_cstate_init(void) static void __exit ffh_cstate_exit(void) { - if (cpu_cstate_entry) { - free_percpu(cpu_cstate_entry); - cpu_cstate_entry = NULL; - } + free_percpu(cpu_cstate_entry); + cpu_cstate_entry = NULL; } arch_initcall(ffh_cstate_init); diff --git a/block/blktrace.c b/block/blktrace.c index 74e02c04b2d..d3679dd1d22 100644 --- a/block/blktrace.c +++ b/block/blktrace.c @@ -394,8 +394,7 @@ err: if (bt) { if (bt->dropped_file) debugfs_remove(bt->dropped_file); - if (bt->sequence) - free_percpu(bt->sequence); + free_percpu(bt->sequence); if (bt->rchan) relay_close(bt->rchan); kfree(bt); diff --git a/mm/allocpercpu.c b/mm/allocpercpu.c index eaa9abeea53..b2486cf887a 100644 --- a/mm/allocpercpu.c +++ b/mm/allocpercpu.c @@ -17,10 +17,9 @@ void percpu_depopulate(void *__pdata, int cpu) { struct percpu_data *pdata = __percpu_disguise(__pdata); - if (pdata->ptrs[cpu]) { - kfree(pdata->ptrs[cpu]); - pdata->ptrs[cpu] = NULL; - } + + kfree(pdata->ptrs[cpu]); + pdata->ptrs[cpu] = NULL; } EXPORT_SYMBOL_GPL(percpu_depopulate); @@ -123,6 +122,8 @@ EXPORT_SYMBOL_GPL(__percpu_alloc_mask); */ void percpu_free(void *__pdata) { + if (unlikely(!__pdata)) + return; __percpu_depopulate_mask(__pdata, &cpu_possible_map); kfree(__percpu_disguise(__pdata)); } diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 87c8f54872b..e5cd83b2205 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -720,10 +720,8 @@ snmp6_mib_free(void *ptr[2]) { if (ptr == NULL) return; - if (ptr[0]) - free_percpu(ptr[0]); - if (ptr[1]) - free_percpu(ptr[1]); + free_percpu(ptr[0]); + free_percpu(ptr[1]); ptr[0] = ptr[1] = NULL; } -- cgit v1.2.3-70-g09d2 From e94b1766097d53e6f3ccfb36c8baa562ffeda3fc Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:17 -0800 Subject: [PATCH] slab: remove SLAB_KERNEL SLAB_KERNEL is an alias of GFP_KERNEL. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/sysenter.c | 2 +- arch/ia64/ia32/binfmt_elf32.c | 8 ++++---- arch/ia64/kernel/perfmon.c | 2 +- arch/ia64/mm/init.c | 4 ++-- arch/powerpc/kernel/vdso.c | 2 +- arch/powerpc/platforms/cell/spufs/inode.c | 2 +- arch/sh/kernel/vsyscall/vsyscall.c | 2 +- arch/x86_64/ia32/ia32_binfmt.c | 2 +- arch/x86_64/ia32/syscall32.c | 2 +- drivers/atm/he.c | 4 ++-- drivers/base/dmapool.c | 2 +- drivers/dma/ioatdma.c | 4 ++-- drivers/ieee1394/hosts.c | 2 +- drivers/ieee1394/ohci1394.c | 8 ++++---- drivers/ieee1394/pcilynx.c | 2 +- drivers/ieee1394/raw1394.c | 10 +++++----- drivers/infiniband/hw/ehca/ehca_av.c | 2 +- drivers/infiniband/hw/ehca/ehca_cq.c | 2 +- drivers/infiniband/hw/ehca/ehca_main.c | 2 +- drivers/infiniband/hw/ehca/ehca_mrmw.c | 4 ++-- drivers/infiniband/hw/ehca/ehca_pd.c | 2 +- drivers/infiniband/hw/ehca/ehca_qp.c | 2 +- drivers/input/touchscreen/ads7846.c | 2 +- drivers/isdn/gigaset/bas-gigaset.c | 14 +++++++------- drivers/isdn/gigaset/usb-gigaset.c | 6 +++--- drivers/media/dvb/cinergyT2/cinergyT2.c | 2 +- drivers/mtd/devices/m25p80.c | 2 +- drivers/scsi/ipr.c | 2 +- drivers/spi/spi.c | 4 ++-- drivers/spi/spi_bitbang.c | 2 +- drivers/usb/core/hub.c | 4 ++-- drivers/usb/gadget/gmidi.c | 2 +- drivers/usb/gadget/goku_udc.c | 2 +- drivers/usb/gadget/inode.c | 6 +++--- drivers/usb/gadget/net2280.c | 2 +- drivers/usb/gadget/omap_udc.c | 2 +- drivers/usb/gadget/zero.c | 2 +- drivers/usb/host/hc_crisv10.c | 2 +- drivers/usb/host/ohci-pnx4008.c | 2 +- drivers/usb/input/acecad.c | 2 +- drivers/usb/input/usbtouchscreen.c | 2 +- drivers/usb/misc/usbtest.c | 28 ++++++++++++++-------------- drivers/usb/net/rndis_host.c | 2 +- drivers/usb/net/usbnet.c | 4 ++-- fs/adfs/super.c | 2 +- fs/affs/super.c | 2 +- fs/afs/super.c | 2 +- fs/befs/linuxvfs.c | 2 +- fs/bfs/inode.c | 2 +- fs/block_dev.c | 2 +- fs/cifs/cifsfs.c | 2 +- fs/cifs/misc.c | 4 ++-- fs/cifs/transport.c | 4 ++-- fs/coda/inode.c | 2 +- fs/dnotify.c | 2 +- fs/ecryptfs/crypto.c | 2 +- fs/ecryptfs/file.c | 2 +- fs/ecryptfs/inode.c | 4 ++-- fs/ecryptfs/keystore.c | 2 +- fs/ecryptfs/main.c | 4 ++-- fs/ecryptfs/super.c | 2 +- fs/efs/super.c | 2 +- fs/eventpoll.c | 4 ++-- fs/exec.c | 2 +- fs/ext2/super.c | 2 +- fs/fat/cache.c | 2 +- fs/fat/inode.c | 2 +- fs/fcntl.c | 2 +- fs/freevxfs/vxfs_inode.c | 4 ++-- fs/fuse/dev.c | 2 +- fs/fuse/inode.c | 2 +- fs/hfs/super.c | 2 +- fs/hfsplus/super.c | 2 +- fs/hugetlbfs/inode.c | 2 +- fs/inode.c | 2 +- fs/isofs/inode.c | 2 +- fs/jffs2/super.c | 2 +- fs/locks.c | 2 +- fs/minix/inode.c | 2 +- fs/ncpfs/inode.c | 2 +- fs/nfs/direct.c | 2 +- fs/nfs/inode.c | 2 +- fs/nfs/pagelist.c | 2 +- fs/openpromfs/inode.c | 2 +- fs/proc/inode.c | 2 +- fs/qnx4/inode.c | 2 +- fs/reiserfs/super.c | 2 +- fs/romfs/inode.c | 2 +- fs/smbfs/inode.c | 2 +- fs/smbfs/request.c | 2 +- fs/sysv/inode.c | 2 +- fs/udf/super.c | 2 +- fs/ufs/super.c | 2 +- include/linux/fs.h | 2 +- include/linux/rmap.h | 2 +- include/linux/slab.h | 1 - include/linux/taskstats_kern.h | 2 +- ipc/mqueue.c | 2 +- kernel/delayacct.c | 2 +- kernel/fork.c | 6 +++--- kernel/taskstats.c | 2 +- kernel/user.c | 2 +- mm/mempolicy.c | 2 +- mm/mmap.c | 4 ++-- mm/shmem.c | 2 +- mm/slab.c | 2 +- net/decnet/dn_table.c | 2 +- net/ipv4/fib_hash.c | 4 ++-- net/ipv4/fib_trie.c | 4 ++-- net/socket.c | 2 +- net/sunrpc/rpc_pipe.c | 2 +- security/keys/key.c | 2 +- security/selinux/hooks.c | 2 +- security/selinux/ss/avtab.c | 2 +- 114 files changed, 164 insertions(+), 165 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c index 713ba39d32c..0bbacd0ec17 100644 --- a/arch/i386/kernel/sysenter.c +++ b/arch/i386/kernel/sysenter.c @@ -132,7 +132,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack) goto up_fail; } - vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (!vma) { ret = -ENOMEM; goto up_fail; diff --git a/arch/ia64/ia32/binfmt_elf32.c b/arch/ia64/ia32/binfmt_elf32.c index daa6b91bc92..578737ec762 100644 --- a/arch/ia64/ia32/binfmt_elf32.c +++ b/arch/ia64/ia32/binfmt_elf32.c @@ -91,7 +91,7 @@ ia64_elf32_init (struct pt_regs *regs) * it with privilege level 3 because the IVE uses non-privileged accesses to these * tables. IA-32 segmentation is used to protect against IA-32 accesses to them. */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; @@ -117,7 +117,7 @@ ia64_elf32_init (struct pt_regs *regs) * code is locked in specific gate page, which is pointed by pretcode * when setup_frame_ia32 */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; @@ -142,7 +142,7 @@ ia64_elf32_init (struct pt_regs *regs) * Install LDT as anonymous memory. This gives us all-zero segment descriptors * until a task modifies them via modify_ldt(). */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; @@ -214,7 +214,7 @@ ia32_setup_arg_pages (struct linux_binprm *bprm, int executable_stack) bprm->loader += stack_base; bprm->exec += stack_base; - mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + mpnt = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!mpnt) return -ENOMEM; diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 3aaede0d698..e2321536ee4 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -2302,7 +2302,7 @@ pfm_smpl_buffer_alloc(struct task_struct *task, pfm_context_t *ctx, unsigned lon DPRINT(("smpl_buf @%p\n", smpl_buf)); /* allocate vma */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!vma) { DPRINT(("Cannot allocate vma\n")); goto error_kmem; diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index ff87a5cba39..56dc2024220 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -156,7 +156,7 @@ ia64_init_addr_space (void) * the problem. When the process attempts to write to the register backing store * for the first time, it will get a SEGFAULT in this case. */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; @@ -175,7 +175,7 @@ ia64_init_addr_space (void) /* map NaT-page at address zero to speed up speculative dereferencing of NULL: */ if (!(current->personality & MMAP_PAGE_ZERO)) { - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; diff --git a/arch/powerpc/kernel/vdso.c b/arch/powerpc/kernel/vdso.c index c913ad5cad2..a4b28c73bba 100644 --- a/arch/powerpc/kernel/vdso.c +++ b/arch/powerpc/kernel/vdso.c @@ -264,7 +264,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, /* Allocate a VMA structure and fill it up */ - vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (vma == NULL) { rc = -ENOMEM; goto fail_mmapsem; diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index c7d010749a1..7edfcc9d285 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -48,7 +48,7 @@ spufs_alloc_inode(struct super_block *sb) { struct spufs_inode_info *ei; - ei = kmem_cache_alloc(spufs_inode_cache, SLAB_KERNEL); + ei = kmem_cache_alloc(spufs_inode_cache, GFP_KERNEL); if (!ei) return NULL; diff --git a/arch/sh/kernel/vsyscall/vsyscall.c b/arch/sh/kernel/vsyscall/vsyscall.c index 075d6cc1a2d..deb46941f31 100644 --- a/arch/sh/kernel/vsyscall/vsyscall.c +++ b/arch/sh/kernel/vsyscall/vsyscall.c @@ -97,7 +97,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, goto up_fail; } - vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (!vma) { ret = -ENOMEM; goto up_fail; diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index 82ef182de6a..932a62ad6c8 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -351,7 +351,7 @@ int ia32_setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, bprm->loader += stack_base; bprm->exec += stack_base; - mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + mpnt = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!mpnt) return -ENOMEM; diff --git a/arch/x86_64/ia32/syscall32.c b/arch/x86_64/ia32/syscall32.c index 3a01329473a..3e5ed20cba4 100644 --- a/arch/x86_64/ia32/syscall32.c +++ b/arch/x86_64/ia32/syscall32.c @@ -49,7 +49,7 @@ int syscall32_setup_pages(struct linux_binprm *bprm, int exstack) struct mm_struct *mm = current->mm; int ret; - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!vma) return -ENOMEM; diff --git a/drivers/atm/he.c b/drivers/atm/he.c index 2a2f0fc2288..ec8a7a633e6 100644 --- a/drivers/atm/he.c +++ b/drivers/atm/he.c @@ -820,7 +820,7 @@ he_init_group(struct he_dev *he_dev, int group) void *cpuaddr; #ifdef USE_RBPS_POOL - cpuaddr = pci_pool_alloc(he_dev->rbps_pool, SLAB_KERNEL|SLAB_DMA, &dma_handle); + cpuaddr = pci_pool_alloc(he_dev->rbps_pool, GFP_KERNEL|SLAB_DMA, &dma_handle); if (cpuaddr == NULL) return -ENOMEM; #else @@ -884,7 +884,7 @@ he_init_group(struct he_dev *he_dev, int group) void *cpuaddr; #ifdef USE_RBPL_POOL - cpuaddr = pci_pool_alloc(he_dev->rbpl_pool, SLAB_KERNEL|SLAB_DMA, &dma_handle); + cpuaddr = pci_pool_alloc(he_dev->rbpl_pool, GFP_KERNEL|SLAB_DMA, &dma_handle); if (cpuaddr == NULL) return -ENOMEM; #else diff --git a/drivers/base/dmapool.c b/drivers/base/dmapool.c index fa4675254f6..dbe0735f8c9 100644 --- a/drivers/base/dmapool.c +++ b/drivers/base/dmapool.c @@ -126,7 +126,7 @@ dma_pool_create (const char *name, struct device *dev, } else if (allocation < size) return NULL; - if (!(retval = kmalloc (sizeof *retval, SLAB_KERNEL))) + if (!(retval = kmalloc (sizeof *retval, GFP_KERNEL))) return retval; strlcpy (retval->name, name, sizeof retval->name); diff --git a/drivers/dma/ioatdma.c b/drivers/dma/ioatdma.c index 0358419a0e4..8e872610461 100644 --- a/drivers/dma/ioatdma.c +++ b/drivers/dma/ioatdma.c @@ -636,10 +636,10 @@ static int ioat_self_test(struct ioat_device *device) dma_cookie_t cookie; int err = 0; - src = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, SLAB_KERNEL); + src = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, GFP_KERNEL); if (!src) return -ENOMEM; - dest = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, SLAB_KERNEL); + dest = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, GFP_KERNEL); if (!dest) { kfree(src); return -ENOMEM; diff --git a/drivers/ieee1394/hosts.c b/drivers/ieee1394/hosts.c index 8f4378a1631..b935e08695a 100644 --- a/drivers/ieee1394/hosts.c +++ b/drivers/ieee1394/hosts.c @@ -123,7 +123,7 @@ struct hpsb_host *hpsb_alloc_host(struct hpsb_host_driver *drv, size_t extra, int i; int hostnum = 0; - h = kzalloc(sizeof(*h) + extra, SLAB_KERNEL); + h = kzalloc(sizeof(*h) + extra, GFP_KERNEL); if (!h) return NULL; diff --git a/drivers/ieee1394/ohci1394.c b/drivers/ieee1394/ohci1394.c index 6e8ea9110c4..eae97d8dcf0 100644 --- a/drivers/ieee1394/ohci1394.c +++ b/drivers/ieee1394/ohci1394.c @@ -1225,7 +1225,7 @@ static int ohci_iso_recv_init(struct hpsb_iso *iso) int ctx; int ret = -ENOMEM; - recv = kmalloc(sizeof(*recv), SLAB_KERNEL); + recv = kmalloc(sizeof(*recv), GFP_KERNEL); if (!recv) return -ENOMEM; @@ -1918,7 +1918,7 @@ static int ohci_iso_xmit_init(struct hpsb_iso *iso) int ctx; int ret = -ENOMEM; - xmit = kmalloc(sizeof(*xmit), SLAB_KERNEL); + xmit = kmalloc(sizeof(*xmit), GFP_KERNEL); if (!xmit) return -ENOMEM; @@ -3021,7 +3021,7 @@ alloc_dma_rcv_ctx(struct ti_ohci *ohci, struct dma_rcv_ctx *d, return -ENOMEM; } - d->prg_cpu[i] = pci_pool_alloc(d->prg_pool, SLAB_KERNEL, d->prg_bus+i); + d->prg_cpu[i] = pci_pool_alloc(d->prg_pool, GFP_KERNEL, d->prg_bus+i); OHCI_DMA_ALLOC("pool dma_rcv prg[%d]", i); if (d->prg_cpu[i] != NULL) { @@ -3117,7 +3117,7 @@ alloc_dma_trm_ctx(struct ti_ohci *ohci, struct dma_trm_ctx *d, OHCI_DMA_ALLOC("dma_rcv prg pool"); for (i = 0; i < d->num_desc; i++) { - d->prg_cpu[i] = pci_pool_alloc(d->prg_pool, SLAB_KERNEL, d->prg_bus+i); + d->prg_cpu[i] = pci_pool_alloc(d->prg_pool, GFP_KERNEL, d->prg_bus+i); OHCI_DMA_ALLOC("pool dma_trm prg[%d]", i); if (d->prg_cpu[i] != NULL) { diff --git a/drivers/ieee1394/pcilynx.c b/drivers/ieee1394/pcilynx.c index 0a7412e27eb..9cab1d66147 100644 --- a/drivers/ieee1394/pcilynx.c +++ b/drivers/ieee1394/pcilynx.c @@ -1428,7 +1428,7 @@ static int __devinit add_card(struct pci_dev *dev, struct i2c_algo_bit_data i2c_adapter_data; error = -ENOMEM; - i2c_ad = kmalloc(sizeof(*i2c_ad), SLAB_KERNEL); + i2c_ad = kmalloc(sizeof(*i2c_ad), GFP_KERNEL); if (!i2c_ad) FAIL("failed to allocate I2C adapter memory"); memcpy(i2c_ad, &bit_ops, sizeof(struct i2c_adapter)); diff --git a/drivers/ieee1394/raw1394.c b/drivers/ieee1394/raw1394.c index 47f6a4e29b4..bf71e069eaf 100644 --- a/drivers/ieee1394/raw1394.c +++ b/drivers/ieee1394/raw1394.c @@ -112,7 +112,7 @@ static struct pending_request *__alloc_pending_request(gfp_t flags) static inline struct pending_request *alloc_pending_request(void) { - return __alloc_pending_request(SLAB_KERNEL); + return __alloc_pending_request(GFP_KERNEL); } static void free_pending_request(struct pending_request *req) @@ -1737,7 +1737,7 @@ static int arm_register(struct file_info *fi, struct pending_request *req) return (-EINVAL); } /* addr-list-entry for fileinfo */ - addr = kmalloc(sizeof(*addr), SLAB_KERNEL); + addr = kmalloc(sizeof(*addr), GFP_KERNEL); if (!addr) { req->req.length = 0; return (-ENOMEM); @@ -2103,7 +2103,7 @@ static int write_phypacket(struct file_info *fi, struct pending_request *req) static int get_config_rom(struct file_info *fi, struct pending_request *req) { int ret = sizeof(struct raw1394_request); - quadlet_t *data = kmalloc(req->req.length, SLAB_KERNEL); + quadlet_t *data = kmalloc(req->req.length, GFP_KERNEL); int status; if (!data) @@ -2133,7 +2133,7 @@ static int get_config_rom(struct file_info *fi, struct pending_request *req) static int update_config_rom(struct file_info *fi, struct pending_request *req) { int ret = sizeof(struct raw1394_request); - quadlet_t *data = kmalloc(req->req.length, SLAB_KERNEL); + quadlet_t *data = kmalloc(req->req.length, GFP_KERNEL); if (!data) return -ENOMEM; if (copy_from_user(data, int2ptr(req->req.sendb), req->req.length)) { @@ -2779,7 +2779,7 @@ static int raw1394_open(struct inode *inode, struct file *file) { struct file_info *fi; - fi = kzalloc(sizeof(*fi), SLAB_KERNEL); + fi = kzalloc(sizeof(*fi), GFP_KERNEL); if (!fi) return -ENOMEM; diff --git a/drivers/infiniband/hw/ehca/ehca_av.c b/drivers/infiniband/hw/ehca/ehca_av.c index 214e2fdddee..0d6e2c4bb24 100644 --- a/drivers/infiniband/hw/ehca/ehca_av.c +++ b/drivers/infiniband/hw/ehca/ehca_av.c @@ -57,7 +57,7 @@ struct ib_ah *ehca_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) struct ehca_shca *shca = container_of(pd->device, struct ehca_shca, ib_device); - av = kmem_cache_alloc(av_cache, SLAB_KERNEL); + av = kmem_cache_alloc(av_cache, GFP_KERNEL); if (!av) { ehca_err(pd->device, "Out of memory pd=%p ah_attr=%p", pd, ah_attr); diff --git a/drivers/infiniband/hw/ehca/ehca_cq.c b/drivers/infiniband/hw/ehca/ehca_cq.c index 458fe19648a..93995b658d9 100644 --- a/drivers/infiniband/hw/ehca/ehca_cq.c +++ b/drivers/infiniband/hw/ehca/ehca_cq.c @@ -134,7 +134,7 @@ struct ib_cq *ehca_create_cq(struct ib_device *device, int cqe, if (cqe >= 0xFFFFFFFF - 64 - additional_cqe) return ERR_PTR(-EINVAL); - my_cq = kmem_cache_alloc(cq_cache, SLAB_KERNEL); + my_cq = kmem_cache_alloc(cq_cache, GFP_KERNEL); if (!my_cq) { ehca_err(device, "Out of memory for ehca_cq struct device=%p", device); diff --git a/drivers/infiniband/hw/ehca/ehca_main.c b/drivers/infiniband/hw/ehca/ehca_main.c index 3d1c1c53503..cc47e4c13a1 100644 --- a/drivers/infiniband/hw/ehca/ehca_main.c +++ b/drivers/infiniband/hw/ehca/ehca_main.c @@ -108,7 +108,7 @@ static struct kmem_cache *ctblk_cache = NULL; void *ehca_alloc_fw_ctrlblock(void) { - void *ret = kmem_cache_zalloc(ctblk_cache, SLAB_KERNEL); + void *ret = kmem_cache_zalloc(ctblk_cache, GFP_KERNEL); if (!ret) ehca_gen_err("Out of memory for ctblk"); return ret; diff --git a/drivers/infiniband/hw/ehca/ehca_mrmw.c b/drivers/infiniband/hw/ehca/ehca_mrmw.c index abce676c0ae..0a5e2214cc5 100644 --- a/drivers/infiniband/hw/ehca/ehca_mrmw.c +++ b/drivers/infiniband/hw/ehca/ehca_mrmw.c @@ -53,7 +53,7 @@ static struct ehca_mr *ehca_mr_new(void) { struct ehca_mr *me; - me = kmem_cache_alloc(mr_cache, SLAB_KERNEL); + me = kmem_cache_alloc(mr_cache, GFP_KERNEL); if (me) { memset(me, 0, sizeof(struct ehca_mr)); spin_lock_init(&me->mrlock); @@ -72,7 +72,7 @@ static struct ehca_mw *ehca_mw_new(void) { struct ehca_mw *me; - me = kmem_cache_alloc(mw_cache, SLAB_KERNEL); + me = kmem_cache_alloc(mw_cache, GFP_KERNEL); if (me) { memset(me, 0, sizeof(struct ehca_mw)); spin_lock_init(&me->mwlock); diff --git a/drivers/infiniband/hw/ehca/ehca_pd.c b/drivers/infiniband/hw/ehca/ehca_pd.c index 2c3cdc6f7b3..d5345e5b3cd 100644 --- a/drivers/infiniband/hw/ehca/ehca_pd.c +++ b/drivers/infiniband/hw/ehca/ehca_pd.c @@ -50,7 +50,7 @@ struct ib_pd *ehca_alloc_pd(struct ib_device *device, { struct ehca_pd *pd; - pd = kmem_cache_alloc(pd_cache, SLAB_KERNEL); + pd = kmem_cache_alloc(pd_cache, GFP_KERNEL); if (!pd) { ehca_err(device, "device=%p context=%p out of memory", device, context); diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c index 8682aa50c70..c6c9cef203e 100644 --- a/drivers/infiniband/hw/ehca/ehca_qp.c +++ b/drivers/infiniband/hw/ehca/ehca_qp.c @@ -450,7 +450,7 @@ struct ib_qp *ehca_create_qp(struct ib_pd *pd, if (pd->uobject && udata) context = pd->uobject->context; - my_qp = kmem_cache_alloc(qp_cache, SLAB_KERNEL); + my_qp = kmem_cache_alloc(qp_cache, GFP_KERNEL); if (!my_qp) { ehca_err(pd->device, "pd=%p not enough memory to alloc qp", pd); return ERR_PTR(-ENOMEM); diff --git a/drivers/input/touchscreen/ads7846.c b/drivers/input/touchscreen/ads7846.c index f56d6a0f062..0517c7387d6 100644 --- a/drivers/input/touchscreen/ads7846.c +++ b/drivers/input/touchscreen/ads7846.c @@ -189,7 +189,7 @@ static int ads7846_read12_ser(struct device *dev, unsigned command) { struct spi_device *spi = to_spi_device(dev); struct ads7846 *ts = dev_get_drvdata(dev); - struct ser_req *req = kzalloc(sizeof *req, SLAB_KERNEL); + struct ser_req *req = kzalloc(sizeof *req, GFP_KERNEL); int status; int sample; int i; diff --git a/drivers/isdn/gigaset/bas-gigaset.c b/drivers/isdn/gigaset/bas-gigaset.c index 5857e7e23f8..63b629b1cdb 100644 --- a/drivers/isdn/gigaset/bas-gigaset.c +++ b/drivers/isdn/gigaset/bas-gigaset.c @@ -2218,21 +2218,21 @@ static int gigaset_probe(struct usb_interface *interface, * - three for the different uses of the default control pipe * - three for each isochronous pipe */ - if (!(ucs->urb_int_in = usb_alloc_urb(0, SLAB_KERNEL)) || - !(ucs->urb_cmd_in = usb_alloc_urb(0, SLAB_KERNEL)) || - !(ucs->urb_cmd_out = usb_alloc_urb(0, SLAB_KERNEL)) || - !(ucs->urb_ctrl = usb_alloc_urb(0, SLAB_KERNEL))) + if (!(ucs->urb_int_in = usb_alloc_urb(0, GFP_KERNEL)) || + !(ucs->urb_cmd_in = usb_alloc_urb(0, GFP_KERNEL)) || + !(ucs->urb_cmd_out = usb_alloc_urb(0, GFP_KERNEL)) || + !(ucs->urb_ctrl = usb_alloc_urb(0, GFP_KERNEL))) goto allocerr; for (j = 0; j < 2; ++j) { ubc = cs->bcs[j].hw.bas; for (i = 0; i < BAS_OUTURBS; ++i) if (!(ubc->isoouturbs[i].urb = - usb_alloc_urb(BAS_NUMFRAMES, SLAB_KERNEL))) + usb_alloc_urb(BAS_NUMFRAMES, GFP_KERNEL))) goto allocerr; for (i = 0; i < BAS_INURBS; ++i) if (!(ubc->isoinurbs[i] = - usb_alloc_urb(BAS_NUMFRAMES, SLAB_KERNEL))) + usb_alloc_urb(BAS_NUMFRAMES, GFP_KERNEL))) goto allocerr; } @@ -2246,7 +2246,7 @@ static int gigaset_probe(struct usb_interface *interface, (endpoint->bEndpointAddress) & 0x0f), ucs->int_in_buf, 3, read_int_callback, cs, endpoint->bInterval); - if ((rc = usb_submit_urb(ucs->urb_int_in, SLAB_KERNEL)) != 0) { + if ((rc = usb_submit_urb(ucs->urb_int_in, GFP_KERNEL)) != 0) { dev_err(cs->dev, "could not submit interrupt URB: %s\n", get_usb_rcmsg(rc)); goto error; diff --git a/drivers/isdn/gigaset/usb-gigaset.c b/drivers/isdn/gigaset/usb-gigaset.c index af89ce188f2..04f2ad7ba8b 100644 --- a/drivers/isdn/gigaset/usb-gigaset.c +++ b/drivers/isdn/gigaset/usb-gigaset.c @@ -763,7 +763,7 @@ static int gigaset_probe(struct usb_interface *interface, goto error; } - ucs->bulk_out_urb = usb_alloc_urb(0, SLAB_KERNEL); + ucs->bulk_out_urb = usb_alloc_urb(0, GFP_KERNEL); if (!ucs->bulk_out_urb) { dev_err(cs->dev, "Couldn't allocate bulk_out_urb\n"); retval = -ENOMEM; @@ -774,7 +774,7 @@ static int gigaset_probe(struct usb_interface *interface, atomic_set(&ucs->busy, 0); - ucs->read_urb = usb_alloc_urb(0, SLAB_KERNEL); + ucs->read_urb = usb_alloc_urb(0, GFP_KERNEL); if (!ucs->read_urb) { dev_err(cs->dev, "No free urbs available\n"); retval = -ENOMEM; @@ -797,7 +797,7 @@ static int gigaset_probe(struct usb_interface *interface, gigaset_read_int_callback, cs->inbuf + 0, endpoint->bInterval); - retval = usb_submit_urb(ucs->read_urb, SLAB_KERNEL); + retval = usb_submit_urb(ucs->read_urb, GFP_KERNEL); if (retval) { dev_err(cs->dev, "Could not submit URB (error %d)\n", -retval); goto error; diff --git a/drivers/media/dvb/cinergyT2/cinergyT2.c b/drivers/media/dvb/cinergyT2/cinergyT2.c index 206c13e47a0..9123147e376 100644 --- a/drivers/media/dvb/cinergyT2/cinergyT2.c +++ b/drivers/media/dvb/cinergyT2/cinergyT2.c @@ -287,7 +287,7 @@ static int cinergyt2_alloc_stream_urbs (struct cinergyt2 *cinergyt2) int i; cinergyt2->streambuf = usb_buffer_alloc(cinergyt2->udev, STREAM_URB_COUNT*STREAM_BUF_SIZE, - SLAB_KERNEL, &cinergyt2->streambuf_dmahandle); + GFP_KERNEL, &cinergyt2->streambuf_dmahandle); if (!cinergyt2->streambuf) { dprintk(1, "failed to alloc consistent stream memory area, bailing out!\n"); return -ENOMEM; diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c index ef4a731ca5c..334e078ffaf 100644 --- a/drivers/mtd/devices/m25p80.c +++ b/drivers/mtd/devices/m25p80.c @@ -451,7 +451,7 @@ static int __devinit m25p_probe(struct spi_device *spi) return -ENODEV; } - flash = kzalloc(sizeof *flash, SLAB_KERNEL); + flash = kzalloc(sizeof *flash, GFP_KERNEL); if (!flash) return -ENOMEM; diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index ccd4dafce8e..b318500785e 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -6940,7 +6940,7 @@ static int __devinit ipr_alloc_cmd_blks(struct ipr_ioa_cfg *ioa_cfg) return -ENOMEM; for (i = 0; i < IPR_NUM_CMD_BLKS; i++) { - ipr_cmd = pci_pool_alloc (ioa_cfg->ipr_cmd_pool, SLAB_KERNEL, &dma_addr); + ipr_cmd = pci_pool_alloc (ioa_cfg->ipr_cmd_pool, GFP_KERNEL, &dma_addr); if (!ipr_cmd) { ipr_free_cmd_blks(ioa_cfg); diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index c3c0626f550..09f2c74a40c 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -360,7 +360,7 @@ spi_alloc_master(struct device *dev, unsigned size) if (!dev) return NULL; - master = kzalloc(size + sizeof *master, SLAB_KERNEL); + master = kzalloc(size + sizeof *master, GFP_KERNEL); if (!master) return NULL; @@ -607,7 +607,7 @@ static int __init spi_init(void) { int status; - buf = kmalloc(SPI_BUFSIZ, SLAB_KERNEL); + buf = kmalloc(SPI_BUFSIZ, GFP_KERNEL); if (!buf) { status = -ENOMEM; goto err0; diff --git a/drivers/spi/spi_bitbang.c b/drivers/spi/spi_bitbang.c index 08c1c57c612..57289b61d0b 100644 --- a/drivers/spi/spi_bitbang.c +++ b/drivers/spi/spi_bitbang.c @@ -196,7 +196,7 @@ int spi_bitbang_setup(struct spi_device *spi) return -EINVAL; if (!cs) { - cs = kzalloc(sizeof *cs, SLAB_KERNEL); + cs = kzalloc(sizeof *cs, GFP_KERNEL); if (!cs) return -ENOMEM; spi->controller_state = cs; diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 0a46acf557a..77c05be5241 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -2371,7 +2371,7 @@ check_highspeed (struct usb_hub *hub, struct usb_device *udev, int port1) struct usb_qualifier_descriptor *qual; int status; - qual = kmalloc (sizeof *qual, SLAB_KERNEL); + qual = kmalloc (sizeof *qual, GFP_KERNEL); if (qual == NULL) return; @@ -2922,7 +2922,7 @@ static int config_descriptors_changed(struct usb_device *udev) if (len < le16_to_cpu(udev->config[index].desc.wTotalLength)) len = le16_to_cpu(udev->config[index].desc.wTotalLength); } - buf = kmalloc (len, SLAB_KERNEL); + buf = kmalloc (len, GFP_KERNEL); if (buf == NULL) { dev_err(&udev->dev, "no mem to re-read configs after reset\n"); /* assume the worst */ diff --git a/drivers/usb/gadget/gmidi.c b/drivers/usb/gadget/gmidi.c index 64554acad63..31351826f2b 100644 --- a/drivers/usb/gadget/gmidi.c +++ b/drivers/usb/gadget/gmidi.c @@ -1236,7 +1236,7 @@ autoconf_fail: /* ok, we made sense of the hardware ... */ - dev = kzalloc(sizeof(*dev), SLAB_KERNEL); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) { return -ENOMEM; } diff --git a/drivers/usb/gadget/goku_udc.c b/drivers/usb/gadget/goku_udc.c index a3076da3f4e..805a9826842 100644 --- a/drivers/usb/gadget/goku_udc.c +++ b/drivers/usb/gadget/goku_udc.c @@ -1864,7 +1864,7 @@ static int goku_probe(struct pci_dev *pdev, const struct pci_device_id *id) } /* alloc, and start init */ - dev = kmalloc (sizeof *dev, SLAB_KERNEL); + dev = kmalloc (sizeof *dev, GFP_KERNEL); if (dev == NULL){ pr_debug("enomem %s\n", pci_name(pdev)); retval = -ENOMEM; diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c index 86924f9cdd7..3fb1044a4db 100644 --- a/drivers/usb/gadget/inode.c +++ b/drivers/usb/gadget/inode.c @@ -412,7 +412,7 @@ ep_read (struct file *fd, char __user *buf, size_t len, loff_t *ptr) /* FIXME readahead for O_NONBLOCK and poll(); careful with ZLPs */ value = -ENOMEM; - kbuf = kmalloc (len, SLAB_KERNEL); + kbuf = kmalloc (len, GFP_KERNEL); if (unlikely (!kbuf)) goto free1; @@ -456,7 +456,7 @@ ep_write (struct file *fd, const char __user *buf, size_t len, loff_t *ptr) /* FIXME writebehind for O_NONBLOCK and poll(), qlen = 1 */ value = -ENOMEM; - kbuf = kmalloc (len, SLAB_KERNEL); + kbuf = kmalloc (len, GFP_KERNEL); if (!kbuf) goto free1; if (copy_from_user (kbuf, buf, len)) { @@ -1898,7 +1898,7 @@ dev_config (struct file *fd, const char __user *buf, size_t len, loff_t *ptr) buf += 4; length -= 4; - kbuf = kmalloc (length, SLAB_KERNEL); + kbuf = kmalloc (length, GFP_KERNEL); if (!kbuf) return -ENOMEM; if (copy_from_user (kbuf, buf, length)) { diff --git a/drivers/usb/gadget/net2280.c b/drivers/usb/gadget/net2280.c index 0b590831582..3024c679e38 100644 --- a/drivers/usb/gadget/net2280.c +++ b/drivers/usb/gadget/net2280.c @@ -2861,7 +2861,7 @@ static int net2280_probe (struct pci_dev *pdev, const struct pci_device_id *id) } /* alloc, and start init */ - dev = kzalloc (sizeof *dev, SLAB_KERNEL); + dev = kzalloc (sizeof *dev, GFP_KERNEL); if (dev == NULL){ retval = -ENOMEM; goto done; diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c index 48a09fd89d1..030d87c28c2 100644 --- a/drivers/usb/gadget/omap_udc.c +++ b/drivers/usb/gadget/omap_udc.c @@ -2581,7 +2581,7 @@ omap_udc_setup(struct platform_device *odev, struct otg_transceiver *xceiv) /* UDC_PULLUP_EN gates the chip clock */ // OTG_SYSCON_1_REG |= DEV_IDLE_EN; - udc = kzalloc(sizeof(*udc), SLAB_KERNEL); + udc = kzalloc(sizeof(*udc), GFP_KERNEL); if (!udc) return -ENOMEM; diff --git a/drivers/usb/gadget/zero.c b/drivers/usb/gadget/zero.c index 0f809dd6849..40710ea1b49 100644 --- a/drivers/usb/gadget/zero.c +++ b/drivers/usb/gadget/zero.c @@ -1190,7 +1190,7 @@ autoconf_fail: /* ok, we made sense of the hardware ... */ - dev = kzalloc(sizeof(*dev), SLAB_KERNEL); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) return -ENOMEM; spin_lock_init (&dev->lock); diff --git a/drivers/usb/host/hc_crisv10.c b/drivers/usb/host/hc_crisv10.c index 396dc69d4b4..7fd872aa654 100644 --- a/drivers/usb/host/hc_crisv10.c +++ b/drivers/usb/host/hc_crisv10.c @@ -188,7 +188,7 @@ static DEFINE_TIMER(bulk_eot_timer, NULL, 0, 0); #define CHECK_ALIGN(x) if (((__u32)(x)) & 0x00000003) \ {panic("Alignment check (DWORD) failed at %s:%s:%d\n", __FILE__, __FUNCTION__, __LINE__);} -#define SLAB_FLAG (in_interrupt() ? GFP_ATOMIC : SLAB_KERNEL) +#define SLAB_FLAG (in_interrupt() ? GFP_ATOMIC : GFP_KERNEL) #define KMALLOC_FLAG (in_interrupt() ? GFP_ATOMIC : GFP_KERNEL) /* Most helpful debugging aid */ diff --git a/drivers/usb/host/ohci-pnx4008.c b/drivers/usb/host/ohci-pnx4008.c index 2dbb7741490..7f26f9bdbaf 100644 --- a/drivers/usb/host/ohci-pnx4008.c +++ b/drivers/usb/host/ohci-pnx4008.c @@ -134,7 +134,7 @@ static int isp1301_attach(struct i2c_adapter *adap, int addr, int kind) { struct i2c_client *c; - c = (struct i2c_client *)kzalloc(sizeof(*c), SLAB_KERNEL); + c = (struct i2c_client *)kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return -ENOMEM; diff --git a/drivers/usb/input/acecad.c b/drivers/usb/input/acecad.c index 0096373b5f9..909138e5aa0 100644 --- a/drivers/usb/input/acecad.c +++ b/drivers/usb/input/acecad.c @@ -152,7 +152,7 @@ static int usb_acecad_probe(struct usb_interface *intf, const struct usb_device_ if (!acecad || !input_dev) goto fail1; - acecad->data = usb_buffer_alloc(dev, 8, SLAB_KERNEL, &acecad->data_dma); + acecad->data = usb_buffer_alloc(dev, 8, GFP_KERNEL, &acecad->data_dma); if (!acecad->data) goto fail1; diff --git a/drivers/usb/input/usbtouchscreen.c b/drivers/usb/input/usbtouchscreen.c index 49704d4ed0e..7f3c57da9bc 100644 --- a/drivers/usb/input/usbtouchscreen.c +++ b/drivers/usb/input/usbtouchscreen.c @@ -680,7 +680,7 @@ static int usbtouch_probe(struct usb_interface *intf, type->process_pkt = usbtouch_process_pkt; usbtouch->data = usb_buffer_alloc(udev, type->rept_size, - SLAB_KERNEL, &usbtouch->data_dma); + GFP_KERNEL, &usbtouch->data_dma); if (!usbtouch->data) goto out_free; diff --git a/drivers/usb/misc/usbtest.c b/drivers/usb/misc/usbtest.c index ea04dccdc65..fb321864a92 100644 --- a/drivers/usb/misc/usbtest.c +++ b/drivers/usb/misc/usbtest.c @@ -213,7 +213,7 @@ static struct urb *simple_alloc_urb ( if (bytes < 0) return NULL; - urb = usb_alloc_urb (0, SLAB_KERNEL); + urb = usb_alloc_urb (0, GFP_KERNEL); if (!urb) return urb; usb_fill_bulk_urb (urb, udev, pipe, NULL, bytes, simple_callback, NULL); @@ -223,7 +223,7 @@ static struct urb *simple_alloc_urb ( urb->transfer_flags = URB_NO_TRANSFER_DMA_MAP; if (usb_pipein (pipe)) urb->transfer_flags |= URB_SHORT_NOT_OK; - urb->transfer_buffer = usb_buffer_alloc (udev, bytes, SLAB_KERNEL, + urb->transfer_buffer = usb_buffer_alloc (udev, bytes, GFP_KERNEL, &urb->transfer_dma); if (!urb->transfer_buffer) { usb_free_urb (urb); @@ -315,7 +315,7 @@ static int simple_io ( init_completion (&completion); if (usb_pipeout (urb->pipe)) simple_fill_buf (urb); - if ((retval = usb_submit_urb (urb, SLAB_KERNEL)) != 0) + if ((retval = usb_submit_urb (urb, GFP_KERNEL)) != 0) break; /* NOTE: no timeouts; can't be broken out of by interrupt */ @@ -374,7 +374,7 @@ alloc_sglist (int nents, int max, int vary) unsigned i; unsigned size = max; - sg = kmalloc (nents * sizeof *sg, SLAB_KERNEL); + sg = kmalloc (nents * sizeof *sg, GFP_KERNEL); if (!sg) return NULL; @@ -382,7 +382,7 @@ alloc_sglist (int nents, int max, int vary) char *buf; unsigned j; - buf = kzalloc (size, SLAB_KERNEL); + buf = kzalloc (size, GFP_KERNEL); if (!buf) { free_sglist (sg, i); return NULL; @@ -428,7 +428,7 @@ static int perform_sglist ( (udev->speed == USB_SPEED_HIGH) ? (INTERRUPT_RATE << 3) : INTERRUPT_RATE, - sg, nents, 0, SLAB_KERNEL); + sg, nents, 0, GFP_KERNEL); if (retval) break; @@ -855,7 +855,7 @@ test_ctrl_queue (struct usbtest_dev *dev, struct usbtest_param *param) * as with bulk/intr sglists, sglen is the queue depth; it also * controls which subtests run (more tests than sglen) or rerun. */ - urb = kcalloc(param->sglen, sizeof(struct urb *), SLAB_KERNEL); + urb = kcalloc(param->sglen, sizeof(struct urb *), GFP_KERNEL); if (!urb) return -ENOMEM; for (i = 0; i < param->sglen; i++) { @@ -981,7 +981,7 @@ test_ctrl_queue (struct usbtest_dev *dev, struct usbtest_param *param) if (!u) goto cleanup; - reqp = usb_buffer_alloc (udev, sizeof *reqp, SLAB_KERNEL, + reqp = usb_buffer_alloc (udev, sizeof *reqp, GFP_KERNEL, &u->setup_dma); if (!reqp) goto cleanup; @@ -1067,7 +1067,7 @@ static int unlink1 (struct usbtest_dev *dev, int pipe, int size, int async) * FIXME want additional tests for when endpoint is STALLing * due to errors, or is just NAKing requests. */ - if ((retval = usb_submit_urb (urb, SLAB_KERNEL)) != 0) { + if ((retval = usb_submit_urb (urb, GFP_KERNEL)) != 0) { dev_dbg (&dev->intf->dev, "submit fail %d\n", retval); return retval; } @@ -1251,7 +1251,7 @@ static int ctrl_out (struct usbtest_dev *dev, if (length < 1 || length > 0xffff || vary >= length) return -EINVAL; - buf = kmalloc(length, SLAB_KERNEL); + buf = kmalloc(length, GFP_KERNEL); if (!buf) return -ENOMEM; @@ -1403,7 +1403,7 @@ static struct urb *iso_alloc_urb ( maxp *= 1 + (0x3 & (le16_to_cpu(desc->wMaxPacketSize) >> 11)); packets = (bytes + maxp - 1) / maxp; - urb = usb_alloc_urb (packets, SLAB_KERNEL); + urb = usb_alloc_urb (packets, GFP_KERNEL); if (!urb) return urb; urb->dev = udev; @@ -1411,7 +1411,7 @@ static struct urb *iso_alloc_urb ( urb->number_of_packets = packets; urb->transfer_buffer_length = bytes; - urb->transfer_buffer = usb_buffer_alloc (udev, bytes, SLAB_KERNEL, + urb->transfer_buffer = usb_buffer_alloc (udev, bytes, GFP_KERNEL, &urb->transfer_dma); if (!urb->transfer_buffer) { usb_free_urb (urb); @@ -1900,7 +1900,7 @@ usbtest_probe (struct usb_interface *intf, const struct usb_device_id *id) } #endif - dev = kzalloc(sizeof(*dev), SLAB_KERNEL); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) return -ENOMEM; info = (struct usbtest_info *) id->driver_info; @@ -1910,7 +1910,7 @@ usbtest_probe (struct usb_interface *intf, const struct usb_device_id *id) dev->intf = intf; /* cacheline-aligned scratch for i/o */ - if ((dev->buf = kmalloc (TBUF_SIZE, SLAB_KERNEL)) == NULL) { + if ((dev->buf = kmalloc (TBUF_SIZE, GFP_KERNEL)) == NULL) { kfree (dev); return -ENOMEM; } diff --git a/drivers/usb/net/rndis_host.c b/drivers/usb/net/rndis_host.c index c2a28d88ef3..99f26b3e502 100644 --- a/drivers/usb/net/rndis_host.c +++ b/drivers/usb/net/rndis_host.c @@ -469,7 +469,7 @@ static void rndis_unbind(struct usbnet *dev, struct usb_interface *intf) struct rndis_halt *halt; /* try to clear any rndis state/activity (no i/o from stack!) */ - halt = kcalloc(1, sizeof *halt, SLAB_KERNEL); + halt = kcalloc(1, sizeof *halt, GFP_KERNEL); if (halt) { halt->msg_type = RNDIS_MSG_HALT; halt->msg_len = ccpu2(sizeof *halt); diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index 327f9755567..6e39e998825 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -179,9 +179,9 @@ static int init_status (struct usbnet *dev, struct usb_interface *intf) period = max ((int) dev->status->desc.bInterval, (dev->udev->speed == USB_SPEED_HIGH) ? 7 : 3); - buf = kmalloc (maxp, SLAB_KERNEL); + buf = kmalloc (maxp, GFP_KERNEL); if (buf) { - dev->interrupt = usb_alloc_urb (0, SLAB_KERNEL); + dev->interrupt = usb_alloc_urb (0, GFP_KERNEL); if (!dev->interrupt) { kfree (buf); return -ENOMEM; diff --git a/fs/adfs/super.c b/fs/adfs/super.c index 9ade139086f..52eb10ca654 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -217,7 +217,7 @@ static kmem_cache_t *adfs_inode_cachep; static struct inode *adfs_alloc_inode(struct super_block *sb) { struct adfs_inode_info *ei; - ei = (struct adfs_inode_info *)kmem_cache_alloc(adfs_inode_cachep, SLAB_KERNEL); + ei = (struct adfs_inode_info *)kmem_cache_alloc(adfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/affs/super.c b/fs/affs/super.c index 5ea72c3a16c..81c73ec09f6 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -71,7 +71,7 @@ static kmem_cache_t * affs_inode_cachep; static struct inode *affs_alloc_inode(struct super_block *sb) { struct affs_inode_info *ei; - ei = (struct affs_inode_info *)kmem_cache_alloc(affs_inode_cachep, SLAB_KERNEL); + ei = (struct affs_inode_info *)kmem_cache_alloc(affs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; ei->vfs_inode.i_version = 1; diff --git a/fs/afs/super.c b/fs/afs/super.c index 67d1f5c819e..c6ead009bf7 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -412,7 +412,7 @@ static struct inode *afs_alloc_inode(struct super_block *sb) struct afs_vnode *vnode; vnode = (struct afs_vnode *) - kmem_cache_alloc(afs_inode_cachep, SLAB_KERNEL); + kmem_cache_alloc(afs_inode_cachep, GFP_KERNEL); if (!vnode) return NULL; diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 07f7144f0e2..995348df94a 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -277,7 +277,7 @@ befs_alloc_inode(struct super_block *sb) { struct befs_inode_info *bi; bi = (struct befs_inode_info *)kmem_cache_alloc(befs_inode_cachep, - SLAB_KERNEL); + GFP_KERNEL); if (!bi) return NULL; return &bi->vfs_inode; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index ed27ffb3459..2e45123c8f7 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -233,7 +233,7 @@ static kmem_cache_t * bfs_inode_cachep; static struct inode *bfs_alloc_inode(struct super_block *sb) { struct bfs_inode_info *bi; - bi = kmem_cache_alloc(bfs_inode_cachep, SLAB_KERNEL); + bi = kmem_cache_alloc(bfs_inode_cachep, GFP_KERNEL); if (!bi) return NULL; return &bi->vfs_inode; diff --git a/fs/block_dev.c b/fs/block_dev.c index 36c0e7af9d0..063506705f2 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -239,7 +239,7 @@ static kmem_cache_t * bdev_cachep __read_mostly; static struct inode *bdev_alloc_inode(struct super_block *sb) { - struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, SLAB_KERNEL); + struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 84976cdbe71..84168629cea 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -245,7 +245,7 @@ static struct inode * cifs_alloc_inode(struct super_block *sb) { struct cifsInodeInfo *cifs_inode; - cifs_inode = kmem_cache_alloc(cifs_inode_cachep, SLAB_KERNEL); + cifs_inode = kmem_cache_alloc(cifs_inode_cachep, GFP_KERNEL); if (!cifs_inode) return NULL; cifs_inode->cifsAttrs = 0x20; /* default */ diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 8355daff504..aedf683f011 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -153,7 +153,7 @@ cifs_buf_get(void) albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ ret_buf = - (struct smb_hdr *) mempool_alloc(cifs_req_poolp, SLAB_KERNEL | GFP_NOFS); + (struct smb_hdr *) mempool_alloc(cifs_req_poolp, GFP_KERNEL | GFP_NOFS); /* clear the first few header bytes */ /* for most paths, more is cleared in header_assemble */ @@ -192,7 +192,7 @@ cifs_small_buf_get(void) albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ ret_buf = - (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, SLAB_KERNEL | GFP_NOFS); + (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, GFP_KERNEL | GFP_NOFS); if (ret_buf) { /* No need to clear memory here, cleared in header assemble */ /* memset(ret_buf, 0, sizeof(struct smb_hdr) + 27);*/ diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 7514237cf31..1f727765a8e 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -51,7 +51,7 @@ AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) } temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp, - SLAB_KERNEL | GFP_NOFS); + GFP_KERNEL | GFP_NOFS); if (temp == NULL) return temp; else { @@ -118,7 +118,7 @@ AllocOplockQEntry(struct inode * pinode, __u16 fid, struct cifsTconInfo * tcon) return NULL; } temp = (struct oplock_q_entry *) kmem_cache_alloc(cifs_oplock_cachep, - SLAB_KERNEL); + GFP_KERNEL); if (temp == NULL) return temp; else { diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 88d12332116..50cedd2617d 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -43,7 +43,7 @@ static kmem_cache_t * coda_inode_cachep; static struct inode *coda_alloc_inode(struct super_block *sb) { struct coda_inode_info *ei; - ei = (struct coda_inode_info *)kmem_cache_alloc(coda_inode_cachep, SLAB_KERNEL); + ei = (struct coda_inode_info *)kmem_cache_alloc(coda_inode_cachep, GFP_KERNEL); if (!ei) return NULL; memset(&ei->c_fid, 0, sizeof(struct CodaFid)); diff --git a/fs/dnotify.c b/fs/dnotify.c index 2b0442db67e..e778b1737b7 100644 --- a/fs/dnotify.c +++ b/fs/dnotify.c @@ -77,7 +77,7 @@ int fcntl_dirnotify(int fd, struct file *filp, unsigned long arg) inode = filp->f_dentry->d_inode; if (!S_ISDIR(inode->i_mode)) return -ENOTDIR; - dn = kmem_cache_alloc(dn_cache, SLAB_KERNEL); + dn = kmem_cache_alloc(dn_cache, GFP_KERNEL); if (dn == NULL) return -ENOMEM; spin_lock(&inode->i_lock); diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 776b2eed371..7196f50fe15 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -628,7 +628,7 @@ int ecryptfs_decrypt_page(struct file *file, struct page *page) num_extents_per_page = PAGE_CACHE_SIZE / crypt_stat->extent_size; base_extent = (page->index * num_extents_per_page); lower_page_virt = kmem_cache_alloc(ecryptfs_lower_page_cache, - SLAB_KERNEL); + GFP_KERNEL); if (!lower_page_virt) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Error getting page for encrypted " diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index a92ef05eff8..42099e779a5 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -250,7 +250,7 @@ static int ecryptfs_open(struct inode *inode, struct file *file) int lower_flags; /* Released in ecryptfs_release or end of function if failure */ - file_info = kmem_cache_alloc(ecryptfs_file_info_cache, SLAB_KERNEL); + file_info = kmem_cache_alloc(ecryptfs_file_info_cache, GFP_KERNEL); ecryptfs_set_file_private(file, file_info); if (!file_info) { ecryptfs_printk(KERN_ERR, diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 70911412044..8a1945a84c3 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -369,7 +369,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, BUG_ON(!atomic_read(&lower_dentry->d_count)); ecryptfs_set_dentry_private(dentry, kmem_cache_alloc(ecryptfs_dentry_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (!ecryptfs_dentry_to_private(dentry)) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Out of memory whilst attempting " @@ -795,7 +795,7 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) /* Released at out_free: label */ ecryptfs_set_file_private(&fake_ecryptfs_file, kmem_cache_alloc(ecryptfs_file_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (unlikely(!ecryptfs_file_to_private(&fake_ecryptfs_file))) { rc = -ENOMEM; goto out; diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index c3746f56d16..745c0f1bfbb 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -207,7 +207,7 @@ parse_tag_3_packet(struct ecryptfs_crypt_stat *crypt_stat, /* Released: wipe_auth_tok_list called in ecryptfs_parse_packet_set or * at end of function upon failure */ auth_tok_list_item = - kmem_cache_alloc(ecryptfs_auth_tok_list_item_cache, SLAB_KERNEL); + kmem_cache_alloc(ecryptfs_auth_tok_list_item_cache, GFP_KERNEL); if (!auth_tok_list_item) { ecryptfs_printk(KERN_ERR, "Unable to allocate memory\n"); rc = -ENOMEM; diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index a78d87d14ba..a2c6ccbce30 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -378,7 +378,7 @@ ecryptfs_fill_super(struct super_block *sb, void *raw_data, int silent) /* Released in ecryptfs_put_super() */ ecryptfs_set_superblock_private(sb, kmem_cache_alloc(ecryptfs_sb_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (!ecryptfs_superblock_to_private(sb)) { ecryptfs_printk(KERN_WARNING, "Out of memory\n"); rc = -ENOMEM; @@ -402,7 +402,7 @@ ecryptfs_fill_super(struct super_block *sb, void *raw_data, int silent) /* through deactivate_super(sb) from get_sb_nodev() */ ecryptfs_set_dentry_private(sb->s_root, kmem_cache_alloc(ecryptfs_dentry_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (!ecryptfs_dentry_to_private(sb->s_root)) { ecryptfs_printk(KERN_ERR, "dentry_info_cache alloc failed\n"); diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index 825757ae486..eaa5daaf106 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -50,7 +50,7 @@ static struct inode *ecryptfs_alloc_inode(struct super_block *sb) struct inode *inode = NULL; ecryptfs_inode = kmem_cache_alloc(ecryptfs_inode_info_cache, - SLAB_KERNEL); + GFP_KERNEL); if (unlikely(!ecryptfs_inode)) goto out; ecryptfs_init_crypt_stat(&ecryptfs_inode->crypt_stat); diff --git a/fs/efs/super.c b/fs/efs/super.c index b3f50651eb6..69b15a996cf 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -57,7 +57,7 @@ static kmem_cache_t * efs_inode_cachep; static struct inode *efs_alloc_inode(struct super_block *sb) { struct efs_inode_info *ei; - ei = (struct efs_inode_info *)kmem_cache_alloc(efs_inode_cachep, SLAB_KERNEL); + ei = (struct efs_inode_info *)kmem_cache_alloc(efs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index ae228ec54e9..f5c88435c6b 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -961,7 +961,7 @@ static void ep_ptable_queue_proc(struct file *file, wait_queue_head_t *whead, struct epitem *epi = ep_item_from_epqueue(pt); struct eppoll_entry *pwq; - if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, SLAB_KERNEL))) { + if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, GFP_KERNEL))) { init_waitqueue_func_entry(&pwq->wait, ep_poll_callback); pwq->whead = whead; pwq->base = epi; @@ -1004,7 +1004,7 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, struct ep_pqueue epq; error = -ENOMEM; - if (!(epi = kmem_cache_alloc(epi_cache, SLAB_KERNEL))) + if (!(epi = kmem_cache_alloc(epi_cache, GFP_KERNEL))) goto eexit_1; /* Item initialization follow here ... */ diff --git a/fs/exec.c b/fs/exec.c index d993ea1a81a..2092bd20746 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -404,7 +404,7 @@ int setup_arg_pages(struct linux_binprm *bprm, bprm->loader += stack_base; bprm->exec += stack_base; - mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + mpnt = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!mpnt) return -ENOMEM; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index d8b9abd95d0..85c237e7385 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -140,7 +140,7 @@ static kmem_cache_t * ext2_inode_cachep; static struct inode *ext2_alloc_inode(struct super_block *sb) { struct ext2_inode_info *ei; - ei = (struct ext2_inode_info *)kmem_cache_alloc(ext2_inode_cachep, SLAB_KERNEL); + ei = (struct ext2_inode_info *)kmem_cache_alloc(ext2_inode_cachep, GFP_KERNEL); if (!ei) return NULL; #ifdef CONFIG_EXT2_FS_POSIX_ACL diff --git a/fs/fat/cache.c b/fs/fat/cache.c index 82cc4f59e3b..8c272278455 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -63,7 +63,7 @@ void fat_cache_destroy(void) static inline struct fat_cache *fat_cache_alloc(struct inode *inode) { - return kmem_cache_alloc(fat_cache_cachep, SLAB_KERNEL); + return kmem_cache_alloc(fat_cache_cachep, GFP_KERNEL); } static inline void fat_cache_free(struct fat_cache *cache) diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 78945b53b0f..b58fd0c9f3c 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -482,7 +482,7 @@ static kmem_cache_t *fat_inode_cachep; static struct inode *fat_alloc_inode(struct super_block *sb) { struct msdos_inode_info *ei; - ei = kmem_cache_alloc(fat_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(fat_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/fcntl.c b/fs/fcntl.c index e4f26165f12..c03dc9cb21c 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -567,7 +567,7 @@ int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fap int result = 0; if (on) { - new = kmem_cache_alloc(fasync_cache, SLAB_KERNEL); + new = kmem_cache_alloc(fasync_cache, GFP_KERNEL); if (!new) return -ENOMEM; } diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index 4786d51ad3b..d2dd0d70007 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -103,7 +103,7 @@ vxfs_blkiget(struct super_block *sbp, u_long extent, ino_t ino) struct vxfs_inode_info *vip; struct vxfs_dinode *dip; - if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, SLAB_KERNEL))) + if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, GFP_KERNEL))) goto fail; dip = (struct vxfs_dinode *)(bp->b_data + offset); memcpy(vip, dip, sizeof(*vip)); @@ -145,7 +145,7 @@ __vxfs_iget(ino_t ino, struct inode *ilistp) struct vxfs_dinode *dip; caddr_t kaddr = (char *)page_address(pp); - if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, SLAB_KERNEL))) + if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, GFP_KERNEL))) goto fail; dip = (struct vxfs_dinode *)(kaddr + offset); memcpy(vip, dip, sizeof(*vip)); diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 66571eafbb1..8c15139f275 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -41,7 +41,7 @@ static void fuse_request_init(struct fuse_req *req) struct fuse_req *fuse_request_alloc(void) { - struct fuse_req *req = kmem_cache_alloc(fuse_req_cachep, SLAB_KERNEL); + struct fuse_req *req = kmem_cache_alloc(fuse_req_cachep, GFP_KERNEL); if (req) fuse_request_init(req); return req; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index fc420357037..e039e2047cc 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -46,7 +46,7 @@ static struct inode *fuse_alloc_inode(struct super_block *sb) struct inode *inode; struct fuse_inode *fi; - inode = kmem_cache_alloc(fuse_inode_cachep, SLAB_KERNEL); + inode = kmem_cache_alloc(fuse_inode_cachep, GFP_KERNEL); if (!inode) return NULL; diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 85b17b3fa4a..ffc6409132c 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -145,7 +145,7 @@ static struct inode *hfs_alloc_inode(struct super_block *sb) { struct hfs_inode_info *i; - i = kmem_cache_alloc(hfs_inode_cachep, SLAB_KERNEL); + i = kmem_cache_alloc(hfs_inode_cachep, GFP_KERNEL); return i ? &i->vfs_inode : NULL; } diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 194eede52fa..4a0c70c76c8 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -440,7 +440,7 @@ static struct inode *hfsplus_alloc_inode(struct super_block *sb) { struct hfsplus_inode_info *i; - i = kmem_cache_alloc(hfsplus_inode_cachep, SLAB_KERNEL); + i = kmem_cache_alloc(hfsplus_inode_cachep, GFP_KERNEL); return i ? &i->vfs_inode : NULL; } diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 7f4756963d0..36e52173a54 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -522,7 +522,7 @@ static struct inode *hugetlbfs_alloc_inode(struct super_block *sb) if (unlikely(!hugetlbfs_dec_free_inodes(sbinfo))) return NULL; - p = kmem_cache_alloc(hugetlbfs_inode_cachep, SLAB_KERNEL); + p = kmem_cache_alloc(hugetlbfs_inode_cachep, GFP_KERNEL); if (unlikely(!p)) { hugetlbfs_inc_free_inodes(sbinfo); return NULL; diff --git a/fs/inode.c b/fs/inode.c index 26cdb115ce6..dd15984d51a 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -109,7 +109,7 @@ static struct inode *alloc_inode(struct super_block *sb) if (sb->s_op->alloc_inode) inode = sb->s_op->alloc_inode(sb); else - inode = (struct inode *) kmem_cache_alloc(inode_cachep, SLAB_KERNEL); + inode = (struct inode *) kmem_cache_alloc(inode_cachep, GFP_KERNEL); if (inode) { struct address_space * const mapping = &inode->i_data; diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index c34b862cdbf..4b6381cd2cf 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -62,7 +62,7 @@ static kmem_cache_t *isofs_inode_cachep; static struct inode *isofs_alloc_inode(struct super_block *sb) { struct iso_inode_info *ei; - ei = kmem_cache_alloc(isofs_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(isofs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index bc4b8106a49..77be534ce42 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -33,7 +33,7 @@ static kmem_cache_t *jffs2_inode_cachep; static struct inode *jffs2_alloc_inode(struct super_block *sb) { struct jffs2_inode_info *ei; - ei = (struct jffs2_inode_info *)kmem_cache_alloc(jffs2_inode_cachep, SLAB_KERNEL); + ei = (struct jffs2_inode_info *)kmem_cache_alloc(jffs2_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/locks.c b/fs/locks.c index e0b6a80649a..a7b97d50c1e 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -147,7 +147,7 @@ static kmem_cache_t *filelock_cache __read_mostly; /* Allocate an empty lock structure. */ static struct file_lock *locks_alloc_lock(void) { - return kmem_cache_alloc(filelock_cache, SLAB_KERNEL); + return kmem_cache_alloc(filelock_cache, GFP_KERNEL); } static void locks_release_private(struct file_lock *fl) diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 1e36bae4d0e..ce532c2deda 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -56,7 +56,7 @@ static kmem_cache_t * minix_inode_cachep; static struct inode *minix_alloc_inode(struct super_block *sb) { struct minix_inode_info *ei; - ei = (struct minix_inode_info *)kmem_cache_alloc(minix_inode_cachep, SLAB_KERNEL); + ei = (struct minix_inode_info *)kmem_cache_alloc(minix_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 72dad552aa0..ed84d899220 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -45,7 +45,7 @@ static kmem_cache_t * ncp_inode_cachep; static struct inode *ncp_alloc_inode(struct super_block *sb) { struct ncp_inode_info *ei; - ei = (struct ncp_inode_info *)kmem_cache_alloc(ncp_inode_cachep, SLAB_KERNEL); + ei = (struct ncp_inode_info *)kmem_cache_alloc(ncp_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index bdfabf854a5..769fd0a0c77 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -143,7 +143,7 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void) { struct nfs_direct_req *dreq; - dreq = kmem_cache_alloc(nfs_direct_cachep, SLAB_KERNEL); + dreq = kmem_cache_alloc(nfs_direct_cachep, GFP_KERNEL); if (!dreq) return NULL; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 08cc4c5919a..6b53aae4ed2 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1080,7 +1080,7 @@ void nfs4_clear_inode(struct inode *inode) struct inode *nfs_alloc_inode(struct super_block *sb) { struct nfs_inode *nfsi; - nfsi = (struct nfs_inode *)kmem_cache_alloc(nfs_inode_cachep, SLAB_KERNEL); + nfsi = (struct nfs_inode *)kmem_cache_alloc(nfs_inode_cachep, GFP_KERNEL); if (!nfsi) return NULL; nfsi->flags = 0UL; diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 829af323f28..a1561a820ab 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -26,7 +26,7 @@ static inline struct nfs_page * nfs_page_alloc(void) { struct nfs_page *p; - p = kmem_cache_alloc(nfs_page_cachep, SLAB_KERNEL); + p = kmem_cache_alloc(nfs_page_cachep, GFP_KERNEL); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->wb_list); diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 592a6402e85..911d1bcfc56 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -336,7 +336,7 @@ static struct inode *openprom_alloc_inode(struct super_block *sb) { struct op_inode_info *oi; - oi = kmem_cache_alloc(op_inode_cachep, SLAB_KERNEL); + oi = kmem_cache_alloc(op_inode_cachep, GFP_KERNEL); if (!oi) return NULL; diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 49dfb2ab783..b24cdb2f17c 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -88,7 +88,7 @@ static struct inode *proc_alloc_inode(struct super_block *sb) struct proc_inode *ei; struct inode *inode; - ei = (struct proc_inode *)kmem_cache_alloc(proc_inode_cachep, SLAB_KERNEL); + ei = (struct proc_inode *)kmem_cache_alloc(proc_inode_cachep, GFP_KERNEL); if (!ei) return NULL; ei->pid = NULL; diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 5a41db2a218..5b943eb11d7 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -520,7 +520,7 @@ static kmem_cache_t *qnx4_inode_cachep; static struct inode *qnx4_alloc_inode(struct super_block *sb) { struct qnx4_inode_info *ei; - ei = kmem_cache_alloc(qnx4_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(qnx4_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 17249994110..32332516d65 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -496,7 +496,7 @@ static struct inode *reiserfs_alloc_inode(struct super_block *sb) { struct reiserfs_inode_info *ei; ei = (struct reiserfs_inode_info *) - kmem_cache_alloc(reiserfs_inode_cachep, SLAB_KERNEL); + kmem_cache_alloc(reiserfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index ddcd9e1ef28..d1b455f9b66 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -555,7 +555,7 @@ static kmem_cache_t * romfs_inode_cachep; static struct inode *romfs_alloc_inode(struct super_block *sb) { struct romfs_inode_info *ei; - ei = (struct romfs_inode_info *)kmem_cache_alloc(romfs_inode_cachep, SLAB_KERNEL); + ei = (struct romfs_inode_info *)kmem_cache_alloc(romfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 2c122ee83ad..22161710348 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -55,7 +55,7 @@ static kmem_cache_t *smb_inode_cachep; static struct inode *smb_alloc_inode(struct super_block *sb) { struct smb_inode_info *ei; - ei = (struct smb_inode_info *)kmem_cache_alloc(smb_inode_cachep, SLAB_KERNEL); + ei = (struct smb_inode_info *)kmem_cache_alloc(smb_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/smbfs/request.c b/fs/smbfs/request.c index 0fb74697abc..3eb1402191b 100644 --- a/fs/smbfs/request.c +++ b/fs/smbfs/request.c @@ -61,7 +61,7 @@ static struct smb_request *smb_do_alloc_request(struct smb_sb_info *server, struct smb_request *req; unsigned char *buf = NULL; - req = kmem_cache_alloc(req_cachep, SLAB_KERNEL); + req = kmem_cache_alloc(req_cachep, GFP_KERNEL); VERBOSE("allocating request: %p\n", req); if (!req) goto out; diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index d63c5e48b05..a6ca12b747c 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -307,7 +307,7 @@ static struct inode *sysv_alloc_inode(struct super_block *sb) { struct sysv_inode_info *si; - si = kmem_cache_alloc(sysv_inode_cachep, SLAB_KERNEL); + si = kmem_cache_alloc(sysv_inode_cachep, GFP_KERNEL); if (!si) return NULL; return &si->vfs_inode; diff --git a/fs/udf/super.c b/fs/udf/super.c index 1aea6a4f9a4..e50f24221de 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -112,7 +112,7 @@ static kmem_cache_t * udf_inode_cachep; static struct inode *udf_alloc_inode(struct super_block *sb) { struct udf_inode_info *ei; - ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, SLAB_KERNEL); + ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, GFP_KERNEL); if (!ei) return NULL; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index ec79e3091d1..85a88c0c5e6 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1209,7 +1209,7 @@ static kmem_cache_t * ufs_inode_cachep; static struct inode *ufs_alloc_inode(struct super_block *sb) { struct ufs_inode_info *ei; - ei = (struct ufs_inode_info *)kmem_cache_alloc(ufs_inode_cachep, SLAB_KERNEL); + ei = (struct ufs_inode_info *)kmem_cache_alloc(ufs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; ei->vfs_inode.i_version = 1; diff --git a/include/linux/fs.h b/include/linux/fs.h index a8039c8d8cb..94b831b8157 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1483,7 +1483,7 @@ extern void __init vfs_caches_init(unsigned long); extern struct kmem_cache *names_cachep; -#define __getname() kmem_cache_alloc(names_cachep, SLAB_KERNEL) +#define __getname() kmem_cache_alloc(names_cachep, GFP_KERNEL) #define __putname(name) kmem_cache_free(names_cachep, (void *)(name)) #ifndef CONFIG_AUDITSYSCALL #define putname(name) __putname(name) diff --git a/include/linux/rmap.h b/include/linux/rmap.h index db2c1df4fef..61c2ab634b0 100644 --- a/include/linux/rmap.h +++ b/include/linux/rmap.h @@ -34,7 +34,7 @@ extern kmem_cache_t *anon_vma_cachep; static inline struct anon_vma *anon_vma_alloc(void) { - return kmem_cache_alloc(anon_vma_cachep, SLAB_KERNEL); + return kmem_cache_alloc(anon_vma_cachep, GFP_KERNEL); } static inline void anon_vma_free(struct anon_vma *anon_vma) diff --git a/include/linux/slab.h b/include/linux/slab.h index 34b046ea88f..639f65efa46 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -19,7 +19,6 @@ typedef struct kmem_cache kmem_cache_t; #include /* kmalloc_sizes.h needs L1_CACHE_BYTES */ /* flags for kmem_cache_alloc() */ -#define SLAB_KERNEL GFP_KERNEL #define SLAB_DMA GFP_DMA /* flags to pass to kmem_cache_create(). diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index 6562a2050a2..f81a5af8a4f 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -35,7 +35,7 @@ static inline void taskstats_tgid_alloc(struct task_struct *tsk) return; /* No problem if kmem_cache_zalloc() fails */ - stats = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); + stats = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); spin_lock_irq(&tsk->sighand->siglock); if (!sig->stats) { diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 7c274002c9f..813bb941342 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -224,7 +224,7 @@ static struct inode *mqueue_alloc_inode(struct super_block *sb) { struct mqueue_inode_info *ei; - ei = kmem_cache_alloc(mqueue_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(mqueue_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/kernel/delayacct.c b/kernel/delayacct.c index 66a0ea48751..70e9ec60308 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -41,7 +41,7 @@ void delayacct_init(void) void __delayacct_tsk_init(struct task_struct *tsk) { - tsk->delays = kmem_cache_zalloc(delayacct_cache, SLAB_KERNEL); + tsk->delays = kmem_cache_zalloc(delayacct_cache, GFP_KERNEL); if (tsk->delays) spin_lock_init(&tsk->delays->lock); } diff --git a/kernel/fork.c b/kernel/fork.c index 5678e6c61ef..711aa5f10da 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -237,7 +237,7 @@ static inline int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) goto fail_nomem; charge = len; } - tmp = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + tmp = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!tmp) goto fail_nomem; *tmp = *mpnt; @@ -319,7 +319,7 @@ static inline void mm_free_pgd(struct mm_struct * mm) __cacheline_aligned_in_smp DEFINE_SPINLOCK(mmlist_lock); -#define allocate_mm() (kmem_cache_alloc(mm_cachep, SLAB_KERNEL)) +#define allocate_mm() (kmem_cache_alloc(mm_cachep, GFP_KERNEL)) #define free_mm(mm) (kmem_cache_free(mm_cachep, (mm))) #include @@ -621,7 +621,7 @@ static struct files_struct *alloc_files(void) struct files_struct *newf; struct fdtable *fdt; - newf = kmem_cache_alloc(files_cachep, SLAB_KERNEL); + newf = kmem_cache_alloc(files_cachep, GFP_KERNEL); if (!newf) goto out; diff --git a/kernel/taskstats.c b/kernel/taskstats.c index d3d28919d4b..1b2b326cf70 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -425,7 +425,7 @@ void taskstats_exit_alloc(struct taskstats **ptidstats, unsigned int *mycpu) *mycpu = raw_smp_processor_id(); *ptidstats = NULL; - tmp = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); + tmp = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); if (!tmp) return; diff --git a/kernel/user.c b/kernel/user.c index 220e586127a..c1f93c164c9 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -132,7 +132,7 @@ struct user_struct * alloc_uid(uid_t uid) if (!up) { struct user_struct *new; - new = kmem_cache_alloc(uid_cachep, SLAB_KERNEL); + new = kmem_cache_alloc(uid_cachep, GFP_KERNEL); if (!new) return NULL; new->uid = uid; diff --git a/mm/mempolicy.c b/mm/mempolicy.c index e7b69c90cfd..ad864f8708b 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1326,7 +1326,7 @@ struct mempolicy *__mpol_copy(struct mempolicy *old) atomic_set(&new->refcnt, 1); if (new->policy == MPOL_BIND) { int sz = ksize(old->v.zonelist); - new->v.zonelist = kmemdup(old->v.zonelist, sz, SLAB_KERNEL); + new->v.zonelist = kmemdup(old->v.zonelist, sz, GFP_KERNEL); if (!new->v.zonelist) { kmem_cache_free(policy_cache, new); return ERR_PTR(-ENOMEM); diff --git a/mm/mmap.c b/mm/mmap.c index 7b40abd7cba..7be110e98d4 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1736,7 +1736,7 @@ int split_vma(struct mm_struct * mm, struct vm_area_struct * vma, if (mm->map_count >= sysctl_max_map_count) return -ENOMEM; - new = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!new) return -ENOMEM; @@ -2057,7 +2057,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, vma_start < new_vma->vm_end) *vmap = new_vma; } else { - new_vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + new_vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (new_vma) { *new_vma = *vma; pol = mpol_copy(vma_policy(vma)); diff --git a/mm/shmem.c b/mm/shmem.c index 4959535fc14..bdaecfdaabd 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2263,7 +2263,7 @@ static struct kmem_cache *shmem_inode_cachep; static struct inode *shmem_alloc_inode(struct super_block *sb) { struct shmem_inode_info *p; - p = (struct shmem_inode_info *)kmem_cache_alloc(shmem_inode_cachep, SLAB_KERNEL); + p = (struct shmem_inode_info *)kmem_cache_alloc(shmem_inode_cachep, GFP_KERNEL); if (!p) return NULL; return &p->vfs_inode; diff --git a/mm/slab.c b/mm/slab.c index 9f34b4946fb..1f374c1df01 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2237,7 +2237,7 @@ kmem_cache_create (const char *name, size_t size, size_t align, align = ralign; /* Get cache's description obj. */ - cachep = kmem_cache_zalloc(&cache_cache, SLAB_KERNEL); + cachep = kmem_cache_zalloc(&cache_cache, GFP_KERNEL); if (!cachep) goto oops; diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c index bdbc3f43166..101e5ccaf09 100644 --- a/net/decnet/dn_table.c +++ b/net/decnet/dn_table.c @@ -590,7 +590,7 @@ create: replace: err = -ENOBUFS; - new_f = kmem_cache_alloc(dn_hash_kmem, SLAB_KERNEL); + new_f = kmem_cache_alloc(dn_hash_kmem, GFP_KERNEL); if (new_f == NULL) goto out; diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c index 107bb6cbb0b..4463443e42c 100644 --- a/net/ipv4/fib_hash.c +++ b/net/ipv4/fib_hash.c @@ -485,13 +485,13 @@ static int fn_hash_insert(struct fib_table *tb, struct fib_config *cfg) goto out; err = -ENOBUFS; - new_fa = kmem_cache_alloc(fn_alias_kmem, SLAB_KERNEL); + new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); if (new_fa == NULL) goto out; new_f = NULL; if (!f) { - new_f = kmem_cache_alloc(fn_hash_kmem, SLAB_KERNEL); + new_f = kmem_cache_alloc(fn_hash_kmem, GFP_KERNEL); if (new_f == NULL) goto out_free_new_fa; diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index d17990ec724..6be6caf1af3 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1187,7 +1187,7 @@ static int fn_trie_insert(struct fib_table *tb, struct fib_config *cfg) u8 state; err = -ENOBUFS; - new_fa = kmem_cache_alloc(fn_alias_kmem, SLAB_KERNEL); + new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); if (new_fa == NULL) goto out; @@ -1232,7 +1232,7 @@ static int fn_trie_insert(struct fib_table *tb, struct fib_config *cfg) goto out; err = -ENOBUFS; - new_fa = kmem_cache_alloc(fn_alias_kmem, SLAB_KERNEL); + new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); if (new_fa == NULL) goto out; diff --git a/net/socket.c b/net/socket.c index e8db54702a6..4f417c2ddc1 100644 --- a/net/socket.c +++ b/net/socket.c @@ -236,7 +236,7 @@ static struct inode *sock_alloc_inode(struct super_block *sb) { struct socket_alloc *ei; - ei = kmem_cache_alloc(sock_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(sock_inode_cachep, GFP_KERNEL); if (!ei) return NULL; init_waitqueue_head(&ei->socket.wait); diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 49dba5febbb..df753d0a884 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -143,7 +143,7 @@ static struct inode * rpc_alloc_inode(struct super_block *sb) { struct rpc_inode *rpci; - rpci = (struct rpc_inode *)kmem_cache_alloc(rpc_inode_cachep, SLAB_KERNEL); + rpci = (struct rpc_inode *)kmem_cache_alloc(rpc_inode_cachep, GFP_KERNEL); if (!rpci) return NULL; return &rpci->vfs_inode; diff --git a/security/keys/key.c b/security/keys/key.c index 70eacbe5abd..157bac658bf 100644 --- a/security/keys/key.c +++ b/security/keys/key.c @@ -285,7 +285,7 @@ struct key *key_alloc(struct key_type *type, const char *desc, } /* allocate and initialise the key and its description */ - key = kmem_cache_alloc(key_jar, SLAB_KERNEL); + key = kmem_cache_alloc(key_jar, GFP_KERNEL); if (!key) goto no_memory_2; diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 78f98fe084e..ac1aeed0b28 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -181,7 +181,7 @@ static int inode_alloc_security(struct inode *inode) struct task_security_struct *tsec = current->security; struct inode_security_struct *isec; - isec = kmem_cache_alloc(sel_inode_cache, SLAB_KERNEL); + isec = kmem_cache_alloc(sel_inode_cache, GFP_KERNEL); if (!isec) return -ENOMEM; diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c index d049c7acbc8..2dfc6134c2c 100644 --- a/security/selinux/ss/avtab.c +++ b/security/selinux/ss/avtab.c @@ -36,7 +36,7 @@ avtab_insert_node(struct avtab *h, int hvalue, struct avtab_key *key, struct avtab_datum *datum) { struct avtab_node * newnode; - newnode = kmem_cache_alloc(avtab_node_cachep, SLAB_KERNEL); + newnode = kmem_cache_alloc(avtab_node_cachep, GFP_KERNEL); if (newnode == NULL) return NULL; memset(newnode, 0, sizeof(struct avtab_node)); -- cgit v1.2.3-70-g09d2 From e18b890bb0881bbab6f4f1a6cd20d9c60d66b003 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:20 -0800 Subject: [PATCH] slab: remove kmem_cache_t Replace all uses of kmem_cache_t with struct kmem_cache. The patch was generated using the following script: #!/bin/sh # # Replace one string by another in all the kernel sources. # set -e for file in `find * -name "*.c" -o -name "*.h"|xargs grep -l $1`; do quilt add $file sed -e "1,\$s/$1/$2/g" $file >/tmp/$$ mv /tmp/$$ $file quilt refresh done The script was run like this sh replace kmem_cache_t "struct kmem_cache" Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/DMA-API.txt | 4 ++-- arch/arm/mach-s3c2410/dma.c | 4 ++-- arch/arm26/mm/memc.c | 6 +++--- arch/frv/mm/pgalloc.c | 6 +++--- arch/i386/mm/init.c | 4 ++-- arch/i386/mm/pgtable.c | 6 +++--- arch/ia64/ia32/ia32_support.c | 2 +- arch/ia64/ia32/sys_ia32.c | 2 +- arch/powerpc/kernel/rtas_flash.c | 4 ++-- arch/powerpc/mm/hugetlbpage.c | 2 +- arch/powerpc/mm/init_64.c | 6 +++--- arch/powerpc/platforms/cell/spufs/inode.c | 4 ++-- arch/sh/kernel/cpu/sh4/sq.c | 2 +- arch/sh/mm/pmb.c | 6 +++--- arch/sparc64/mm/init.c | 4 ++-- arch/sparc64/mm/tsb.c | 2 +- block/cfq-iosched.c | 4 ++-- block/ll_rw_blk.c | 6 +++--- drivers/block/aoe/aoeblk.c | 2 +- drivers/ieee1394/eth1394.c | 2 +- drivers/md/dm-crypt.c | 2 +- drivers/md/dm-mpath.c | 2 +- drivers/md/dm-snap.c | 6 +++--- drivers/md/dm.c | 4 ++-- drivers/md/kcopyd.c | 2 +- drivers/md/raid5.c | 4 ++-- drivers/message/i2o/i2o_block.h | 2 +- drivers/pci/msi.c | 2 +- drivers/s390/block/dasd_devmap.c | 2 +- drivers/s390/block/dasd_int.h | 2 +- drivers/s390/scsi/zfcp_def.h | 6 +++--- drivers/scsi/aic94xx/aic94xx.h | 4 ++-- drivers/scsi/aic94xx/aic94xx_hwi.c | 2 +- drivers/scsi/aic94xx/aic94xx_init.c | 4 ++-- drivers/scsi/libsas/sas_init.c | 2 +- drivers/scsi/qla2xxx/qla_os.c | 2 +- drivers/scsi/qla4xxx/ql4_os.c | 2 +- drivers/scsi/scsi.c | 2 +- drivers/scsi/scsi_lib.c | 4 ++-- drivers/scsi/scsi_tgt_lib.c | 2 +- drivers/usb/host/hc_crisv10.c | 6 +++--- drivers/usb/host/uhci-hcd.c | 2 +- drivers/usb/mon/mon_text.c | 6 +++--- fs/adfs/super.c | 4 ++-- fs/affs/super.c | 4 ++-- fs/afs/super.c | 6 +++--- fs/aio.c | 4 ++-- fs/befs/linuxvfs.c | 4 ++-- fs/bfs/inode.c | 4 ++-- fs/bio.c | 4 ++-- fs/block_dev.c | 4 ++-- fs/buffer.c | 4 ++-- fs/cifs/cifsfs.c | 14 +++++++------- fs/cifs/transport.c | 2 +- fs/coda/inode.c | 4 ++-- fs/configfs/configfs_internal.h | 2 +- fs/configfs/mount.c | 2 +- fs/dcache.c | 6 +++--- fs/dcookies.c | 2 +- fs/dlm/memory.c | 2 +- fs/dnotify.c | 2 +- fs/dquot.c | 2 +- fs/ecryptfs/main.c | 2 +- fs/efs/super.c | 4 ++-- fs/eventpoll.c | 4 ++-- fs/ext2/super.c | 4 ++-- fs/ext3/super.c | 4 ++-- fs/ext4/super.c | 4 ++-- fs/fat/cache.c | 4 ++-- fs/fat/inode.c | 4 ++-- fs/fcntl.c | 2 +- fs/freevxfs/vxfs_inode.c | 2 +- fs/fuse/dev.c | 2 +- fs/fuse/inode.c | 4 ++-- fs/gfs2/main.c | 4 ++-- fs/gfs2/util.c | 6 +++--- fs/gfs2/util.h | 6 +++--- fs/hfs/super.c | 4 ++-- fs/hfsplus/super.c | 4 ++-- fs/hpfs/super.c | 4 ++-- fs/hugetlbfs/inode.c | 4 ++-- fs/inode.c | 4 ++-- fs/inotify_user.c | 4 ++-- fs/isofs/inode.c | 4 ++-- fs/jbd/journal.c | 6 +++--- fs/jbd/revoke.c | 4 ++-- fs/jbd2/journal.c | 6 +++--- fs/jbd2/revoke.c | 4 ++-- fs/jffs/inode-v23.c | 4 ++-- fs/jffs/jffs_fm.c | 4 ++-- fs/jffs2/malloc.c | 18 +++++++++--------- fs/jffs2/super.c | 4 ++-- fs/jfs/jfs_metapage.c | 4 ++-- fs/jfs/super.c | 4 ++-- fs/locks.c | 4 ++-- fs/mbcache.c | 2 +- fs/minix/inode.c | 4 ++-- fs/namespace.c | 2 +- fs/ncpfs/inode.c | 4 ++-- fs/nfs/direct.c | 2 +- fs/nfs/inode.c | 4 ++-- fs/nfs/pagelist.c | 2 +- fs/nfs/read.c | 2 +- fs/nfs/write.c | 2 +- fs/nfsd/nfs4state.c | 10 +++++----- fs/ocfs2/dlm/dlmfs.c | 4 ++-- fs/ocfs2/dlm/dlmmaster.c | 2 +- fs/ocfs2/extent_map.c | 2 +- fs/ocfs2/inode.h | 2 +- fs/ocfs2/super.c | 4 ++-- fs/ocfs2/uptodate.c | 2 +- fs/openpromfs/inode.c | 4 ++-- fs/proc/inode.c | 4 ++-- fs/qnx4/inode.c | 4 ++-- fs/reiserfs/super.c | 4 ++-- fs/romfs/inode.c | 4 ++-- fs/smbfs/inode.c | 4 ++-- fs/smbfs/request.c | 2 +- fs/sysfs/mount.c | 2 +- fs/sysfs/sysfs.h | 2 +- fs/sysv/inode.c | 4 ++-- fs/udf/super.c | 4 ++-- fs/ufs/super.c | 4 ++-- include/acpi/platform/aclinux.h | 2 +- include/asm-arm26/pgalloc.h | 2 +- include/asm-i386/pgtable.h | 10 +++++----- include/asm-powerpc/pgalloc.h | 2 +- include/asm-sparc64/pgalloc.h | 2 +- include/linux/delayacct.h | 2 +- include/linux/i2o.h | 2 +- include/linux/jbd.h | 2 +- include/linux/jbd2.h | 2 +- include/linux/raid/raid5.h | 2 +- include/linux/rmap.h | 2 +- include/linux/skbuff.h | 2 +- include/linux/taskstats_kern.h | 2 +- include/net/dst.h | 2 +- include/net/inet_hashtables.h | 6 +++--- include/net/neighbour.h | 2 +- include/net/netfilter/nf_conntrack_expect.h | 2 +- include/net/request_sock.h | 2 +- include/net/sock.h | 2 +- include/net/timewait_sock.h | 2 +- include/scsi/libsas.h | 4 ++-- ipc/mqueue.c | 4 ++-- kernel/delayacct.c | 2 +- kernel/fork.c | 16 ++++++++-------- kernel/pid.c | 2 +- kernel/posix-timers.c | 2 +- kernel/signal.c | 2 +- kernel/taskstats.c | 2 +- kernel/user.c | 2 +- lib/idr.c | 4 ++-- lib/radix-tree.c | 4 ++-- net/bridge/br_fdb.c | 2 +- net/core/flow.c | 2 +- net/core/skbuff.c | 8 ++++---- net/core/sock.c | 2 +- net/dccp/ackvec.c | 4 ++-- net/dccp/ccid.c | 6 +++--- net/dccp/ccid.h | 4 ++-- net/dccp/ccids/lib/loss_interval.h | 2 +- net/dccp/ccids/lib/packet_history.h | 4 ++-- net/decnet/dn_table.c | 2 +- net/ipv4/fib_hash.c | 4 ++-- net/ipv4/fib_trie.c | 2 +- net/ipv4/inet_hashtables.c | 4 ++-- net/ipv4/inetpeer.c | 2 +- net/ipv4/ipmr.c | 2 +- net/ipv4/ipvs/ip_vs_conn.c | 2 +- net/ipv4/netfilter/ip_conntrack_core.c | 4 ++-- net/ipv6/ip6_fib.c | 2 +- net/ipv6/xfrm6_tunnel.c | 2 +- net/netfilter/nf_conntrack_core.c | 6 +++--- net/netfilter/nf_conntrack_expect.c | 2 +- net/netfilter/xt_hashlimit.c | 2 +- net/sctp/protocol.c | 4 ++-- net/sctp/sm_make_chunk.c | 2 +- net/sctp/socket.c | 2 +- net/socket.c | 4 ++-- net/sunrpc/rpc_pipe.c | 4 ++-- net/sunrpc/sched.c | 4 ++-- net/tipc/handler.c | 2 +- net/xfrm/xfrm_input.c | 2 +- net/xfrm/xfrm_policy.c | 2 +- security/keys/key.c | 2 +- security/selinux/avc.c | 2 +- security/selinux/hooks.c | 2 +- security/selinux/ss/avtab.c | 2 +- 189 files changed, 332 insertions(+), 332 deletions(-) (limited to 'arch/i386') diff --git a/Documentation/DMA-API.txt b/Documentation/DMA-API.txt index 05431621c86..8621a064f7e 100644 --- a/Documentation/DMA-API.txt +++ b/Documentation/DMA-API.txt @@ -77,7 +77,7 @@ To get this part of the dma_ API, you must #include Many drivers need lots of small dma-coherent memory regions for DMA descriptors or I/O buffers. Rather than allocating in units of a page or more using dma_alloc_coherent(), you can use DMA pools. These work -much like a kmem_cache_t, except that they use the dma-coherent allocator +much like a struct kmem_cache, except that they use the dma-coherent allocator not __get_free_pages(). Also, they understand common hardware constraints for alignment, like queue heads needing to be aligned on N byte boundaries. @@ -94,7 +94,7 @@ The pool create() routines initialize a pool of dma-coherent buffers for use with a given device. It must be called in a context which can sleep. -The "name" is for diagnostics (like a kmem_cache_t name); dev and size +The "name" is for diagnostics (like a struct kmem_cache name); dev and size are like what you'd pass to dma_alloc_coherent(). The device's hardware alignment requirement for this type of data is "align" (which is expressed in bytes, and must be a power of two). If your device has no boundary diff --git a/arch/arm/mach-s3c2410/dma.c b/arch/arm/mach-s3c2410/dma.c index 3d211dc2f2f..01abb0ace23 100644 --- a/arch/arm/mach-s3c2410/dma.c +++ b/arch/arm/mach-s3c2410/dma.c @@ -40,7 +40,7 @@ /* io map for dma */ static void __iomem *dma_base; -static kmem_cache_t *dma_kmem; +static struct kmem_cache *dma_kmem; struct s3c24xx_dma_selection dma_sel; @@ -1271,7 +1271,7 @@ struct sysdev_class dma_sysclass = { /* kmem cache implementation */ -static void s3c2410_dma_cache_ctor(void *p, kmem_cache_t *c, unsigned long f) +static void s3c2410_dma_cache_ctor(void *p, struct kmem_cache *c, unsigned long f) { memset(p, 0, sizeof(struct s3c2410_dma_buf)); } diff --git a/arch/arm26/mm/memc.c b/arch/arm26/mm/memc.c index 34def6397c3..f2901581d4d 100644 --- a/arch/arm26/mm/memc.c +++ b/arch/arm26/mm/memc.c @@ -24,7 +24,7 @@ #define MEMC_TABLE_SIZE (256*sizeof(unsigned long)) -kmem_cache_t *pte_cache, *pgd_cache; +struct kmem_cache *pte_cache, *pgd_cache; int page_nr; /* @@ -162,12 +162,12 @@ void __init create_memmap_holes(struct meminfo *mi) { } -static void pte_cache_ctor(void *pte, kmem_cache_t *cache, unsigned long flags) +static void pte_cache_ctor(void *pte, struct kmem_cache *cache, unsigned long flags) { memzero(pte, sizeof(pte_t) * PTRS_PER_PTE); } -static void pgd_cache_ctor(void *pgd, kmem_cache_t *cache, unsigned long flags) +static void pgd_cache_ctor(void *pgd, struct kmem_cache *cache, unsigned long flags) { memzero(pgd + MEMC_TABLE_SIZE, USER_PTRS_PER_PGD * sizeof(pgd_t)); } diff --git a/arch/frv/mm/pgalloc.c b/arch/frv/mm/pgalloc.c index f76dd03ddd9..19b13be114a 100644 --- a/arch/frv/mm/pgalloc.c +++ b/arch/frv/mm/pgalloc.c @@ -18,7 +18,7 @@ #include pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__((aligned(PAGE_SIZE))); -kmem_cache_t *pgd_cache; +struct kmem_cache *pgd_cache; pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { @@ -100,7 +100,7 @@ static inline void pgd_list_del(pgd_t *pgd) set_page_private(next, (unsigned long) pprev); } -void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused) +void pgd_ctor(void *pgd, struct kmem_cache *cache, unsigned long unused) { unsigned long flags; @@ -120,7 +120,7 @@ void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused) } /* never called when PTRS_PER_PMD > 1 */ -void pgd_dtor(void *pgd, kmem_cache_t *cache, unsigned long unused) +void pgd_dtor(void *pgd, struct kmem_cache *cache, unsigned long unused) { unsigned long flags; /* can be called from interrupt context */ diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c index 167416155ee..a6a8e397dd8 100644 --- a/arch/i386/mm/init.c +++ b/arch/i386/mm/init.c @@ -699,8 +699,8 @@ int remove_memory(u64 start, u64 size) #endif #endif -kmem_cache_t *pgd_cache; -kmem_cache_t *pmd_cache; +struct kmem_cache *pgd_cache; +struct kmem_cache *pmd_cache; void __init pgtable_cache_init(void) { diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c index 10126e3f817..33be236fc6a 100644 --- a/arch/i386/mm/pgtable.c +++ b/arch/i386/mm/pgtable.c @@ -193,7 +193,7 @@ struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address) return pte; } -void pmd_ctor(void *pmd, kmem_cache_t *cache, unsigned long flags) +void pmd_ctor(void *pmd, struct kmem_cache *cache, unsigned long flags) { memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t)); } @@ -233,7 +233,7 @@ static inline void pgd_list_del(pgd_t *pgd) set_page_private(next, (unsigned long)pprev); } -void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused) +void pgd_ctor(void *pgd, struct kmem_cache *cache, unsigned long unused) { unsigned long flags; @@ -253,7 +253,7 @@ void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused) } /* never called when PTRS_PER_PMD > 1 */ -void pgd_dtor(void *pgd, kmem_cache_t *cache, unsigned long unused) +void pgd_dtor(void *pgd, struct kmem_cache *cache, unsigned long unused) { unsigned long flags; /* can be called from interrupt context */ diff --git a/arch/ia64/ia32/ia32_support.c b/arch/ia64/ia32/ia32_support.c index c187743965a..6af400a12ca 100644 --- a/arch/ia64/ia32/ia32_support.c +++ b/arch/ia64/ia32/ia32_support.c @@ -249,7 +249,7 @@ ia32_init (void) #if PAGE_SHIFT > IA32_PAGE_SHIFT { - extern kmem_cache_t *partial_page_cachep; + extern struct kmem_cache *partial_page_cachep; partial_page_cachep = kmem_cache_create("partial_page_cache", sizeof(struct partial_page), 0, 0, diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 9d6a3f21014..a4a6e1463af 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -254,7 +254,7 @@ mmap_subpage (struct file *file, unsigned long start, unsigned long end, int pro } /* SLAB cache for partial_page structures */ -kmem_cache_t *partial_page_cachep; +struct kmem_cache *partial_page_cachep; /* * init partial_page_list. diff --git a/arch/powerpc/kernel/rtas_flash.c b/arch/powerpc/kernel/rtas_flash.c index b9561d30051..7d0f13fecc0 100644 --- a/arch/powerpc/kernel/rtas_flash.c +++ b/arch/powerpc/kernel/rtas_flash.c @@ -101,7 +101,7 @@ struct flash_block_list_header { /* just the header of flash_block_list */ static struct flash_block_list_header rtas_firmware_flash_list = {0, NULL}; /* Use slab cache to guarantee 4k alignment */ -static kmem_cache_t *flash_block_cache = NULL; +static struct kmem_cache *flash_block_cache = NULL; #define FLASH_BLOCK_LIST_VERSION (1UL) @@ -286,7 +286,7 @@ static ssize_t rtas_flash_read(struct file *file, char __user *buf, } /* constructor for flash_block_cache */ -void rtas_block_ctor(void *ptr, kmem_cache_t *cache, unsigned long flags) +void rtas_block_ctor(void *ptr, struct kmem_cache *cache, unsigned long flags) { memset(ptr, 0, RTAS_BLK_SIZE); } diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 424a8f57e15..89c836d5480 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -1047,7 +1047,7 @@ repeat: return err; } -static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags) +static void zero_ctor(void *addr, struct kmem_cache *cache, unsigned long flags) { memset(addr, 0, kmem_cache_size(cache)); } diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index 9a178549cbc..d12a87ec5ae 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -141,7 +141,7 @@ static int __init setup_kcore(void) } module_init(setup_kcore); -static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags) +static void zero_ctor(void *addr, struct kmem_cache *cache, unsigned long flags) { memset(addr, 0, kmem_cache_size(cache)); } @@ -166,9 +166,9 @@ static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = { /* Hugepages need one extra cache, initialized in hugetlbpage.c. We * can't put into the tables above, because HPAGE_SHIFT is not compile * time constant. */ -kmem_cache_t *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)+1]; +struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)+1]; #else -kmem_cache_t *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)]; +struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)]; #endif void pgtable_cache_init(void) diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index 7edfcc9d285..e3af9112c02 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -40,7 +40,7 @@ #include "spufs.h" -static kmem_cache_t *spufs_inode_cache; +static struct kmem_cache *spufs_inode_cache; char *isolated_loader; static struct inode * @@ -65,7 +65,7 @@ spufs_destroy_inode(struct inode *inode) } static void -spufs_init_once(void *p, kmem_cache_t * cachep, unsigned long flags) +spufs_init_once(void *p, struct kmem_cache * cachep, unsigned long flags) { struct spufs_inode_info *ei = p; diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c index 55f43506995..0c9ea38d2ca 100644 --- a/arch/sh/kernel/cpu/sh4/sq.c +++ b/arch/sh/kernel/cpu/sh4/sq.c @@ -38,7 +38,7 @@ struct sq_mapping { static struct sq_mapping *sq_mapping_list; static DEFINE_SPINLOCK(sq_mapping_lock); -static kmem_cache_t *sq_cache; +static struct kmem_cache *sq_cache; static unsigned long *sq_bitmap; #define store_queue_barrier() \ diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c index 92e745341e4..b60ad83a763 100644 --- a/arch/sh/mm/pmb.c +++ b/arch/sh/mm/pmb.c @@ -30,7 +30,7 @@ #define NR_PMB_ENTRIES 16 -static kmem_cache_t *pmb_cache; +static struct kmem_cache *pmb_cache; static unsigned long pmb_map; static struct pmb_entry pmb_init_map[] = { @@ -283,7 +283,7 @@ void pmb_unmap(unsigned long addr) } while (pmbe); } -static void pmb_cache_ctor(void *pmb, kmem_cache_t *cachep, unsigned long flags) +static void pmb_cache_ctor(void *pmb, struct kmem_cache *cachep, unsigned long flags) { struct pmb_entry *pmbe = pmb; @@ -297,7 +297,7 @@ static void pmb_cache_ctor(void *pmb, kmem_cache_t *cachep, unsigned long flags) spin_unlock_irq(&pmb_list_lock); } -static void pmb_cache_dtor(void *pmb, kmem_cache_t *cachep, unsigned long flags) +static void pmb_cache_dtor(void *pmb, struct kmem_cache *cachep, unsigned long flags) { spin_lock_irq(&pmb_list_lock); pmb_list_del(pmb); diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c index 09cb7fccc03..a8e8802eed4 100644 --- a/arch/sparc64/mm/init.c +++ b/arch/sparc64/mm/init.c @@ -176,9 +176,9 @@ unsigned long sparc64_kern_sec_context __read_mostly; int bigkernel = 0; -kmem_cache_t *pgtable_cache __read_mostly; +struct kmem_cache *pgtable_cache __read_mostly; -static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags) +static void zero_ctor(void *addr, struct kmem_cache *cache, unsigned long flags) { clear_page(addr); } diff --git a/arch/sparc64/mm/tsb.c b/arch/sparc64/mm/tsb.c index beaa02810f0..236d02f41a0 100644 --- a/arch/sparc64/mm/tsb.c +++ b/arch/sparc64/mm/tsb.c @@ -239,7 +239,7 @@ static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_idx, unsign } } -static kmem_cache_t *tsb_caches[8] __read_mostly; +static struct kmem_cache *tsb_caches[8] __read_mostly; static const char *tsb_cache_names[8] = { "tsb_8KB", diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 84e9be07318..78c6b312bd3 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -43,8 +43,8 @@ static int cfq_slice_idle = HZ / 125; #define RQ_CIC(rq) ((struct cfq_io_context*)(rq)->elevator_private) #define RQ_CFQQ(rq) ((rq)->elevator_private2) -static kmem_cache_t *cfq_pool; -static kmem_cache_t *cfq_ioc_pool; +static struct kmem_cache *cfq_pool; +static struct kmem_cache *cfq_ioc_pool; static DEFINE_PER_CPU(unsigned long, ioc_count); static struct completion *ioc_gone; diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index cc6e95f8e5d..a4ff3271d4a 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -44,17 +44,17 @@ static struct io_context *current_io_context(gfp_t gfp_flags, int node); /* * For the allocated request tables */ -static kmem_cache_t *request_cachep; +static struct kmem_cache *request_cachep; /* * For queue allocation */ -static kmem_cache_t *requestq_cachep; +static struct kmem_cache *requestq_cachep; /* * For io context allocations */ -static kmem_cache_t *iocontext_cachep; +static struct kmem_cache *iocontext_cachep; /* * Controlling structure to kblockd diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index aa25f8b09fe..478489c568a 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -12,7 +12,7 @@ #include #include "aoe.h" -static kmem_cache_t *buf_pool_cache; +static struct kmem_cache *buf_pool_cache; static ssize_t aoedisk_show_state(struct gendisk * disk, char *page) { diff --git a/drivers/ieee1394/eth1394.c b/drivers/ieee1394/eth1394.c index 31e5cc49d61..27d6c642415 100644 --- a/drivers/ieee1394/eth1394.c +++ b/drivers/ieee1394/eth1394.c @@ -133,7 +133,7 @@ struct eth1394_node_info { #define ETH1394_DRIVER_NAME "eth1394" static const char driver_name[] = ETH1394_DRIVER_NAME; -static kmem_cache_t *packet_task_cache; +static struct kmem_cache *packet_task_cache; static struct hpsb_highlevel eth1394_highlevel; diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index ed2d4ef27fd..c7bee4f2eed 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -101,7 +101,7 @@ struct crypt_config { #define MIN_POOL_PAGES 32 #define MIN_BIO_PAGES 8 -static kmem_cache_t *_crypt_io_pool; +static struct kmem_cache *_crypt_io_pool; /* * Different IV generation algorithms: diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index e77ee6fd104..cf8bf052138 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -101,7 +101,7 @@ typedef int (*action_fn) (struct pgpath *pgpath); #define MIN_IOS 256 /* Mempool size */ -static kmem_cache_t *_mpio_cache; +static struct kmem_cache *_mpio_cache; struct workqueue_struct *kmultipathd; static void process_queued_ios(struct work_struct *work); diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 91c7aa1fed0..b0ce2ce8227 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -88,8 +88,8 @@ struct pending_exception { * Hash table mapping origin volumes to lists of snapshots and * a lock to protect it */ -static kmem_cache_t *exception_cache; -static kmem_cache_t *pending_cache; +static struct kmem_cache *exception_cache; +static struct kmem_cache *pending_cache; static mempool_t *pending_pool; /* @@ -228,7 +228,7 @@ static int init_exception_table(struct exception_table *et, uint32_t size) return 0; } -static void exit_exception_table(struct exception_table *et, kmem_cache_t *mem) +static void exit_exception_table(struct exception_table *et, struct kmem_cache *mem) { struct list_head *slot; struct exception *ex, *next; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index fc4f743f3b5..7ec1b112a6d 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -121,8 +121,8 @@ struct mapped_device { }; #define MIN_IOS 256 -static kmem_cache_t *_io_cache; -static kmem_cache_t *_tio_cache; +static struct kmem_cache *_io_cache; +static struct kmem_cache *_tio_cache; static int __init local_init(void) { diff --git a/drivers/md/kcopyd.c b/drivers/md/kcopyd.c index b3c01496c73..b46f6c575f7 100644 --- a/drivers/md/kcopyd.c +++ b/drivers/md/kcopyd.c @@ -203,7 +203,7 @@ struct kcopyd_job { /* FIXME: this should scale with the number of pages */ #define MIN_JOBS 512 -static kmem_cache_t *_job_cache; +static struct kmem_cache *_job_cache; static mempool_t *_job_pool; /* diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 69c3e201fa3..52914d5cec7 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -348,7 +348,7 @@ static int grow_one_stripe(raid5_conf_t *conf) static int grow_stripes(raid5_conf_t *conf, int num) { - kmem_cache_t *sc; + struct kmem_cache *sc; int devs = conf->raid_disks; sprintf(conf->cache_name[0], "raid5/%s", mdname(conf->mddev)); @@ -397,7 +397,7 @@ static int resize_stripes(raid5_conf_t *conf, int newsize) LIST_HEAD(newstripes); struct disk_info *ndisks; int err = 0; - kmem_cache_t *sc; + struct kmem_cache *sc; int i; if (newsize <= conf->pool_size) diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h index d9fdc95b440..67f921b4419 100644 --- a/drivers/message/i2o/i2o_block.h +++ b/drivers/message/i2o/i2o_block.h @@ -64,7 +64,7 @@ /* I2O Block OSM mempool struct */ struct i2o_block_mempool { - kmem_cache_t *slab; + struct kmem_cache *slab; mempool_t *pool; }; diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 9fc9a34ef24..9168401401b 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -26,7 +26,7 @@ static DEFINE_SPINLOCK(msi_lock); static struct msi_desc* msi_desc[NR_IRQS] = { [0 ... NR_IRQS-1] = NULL }; -static kmem_cache_t* msi_cachep; +static struct kmem_cache* msi_cachep; static int pci_msi_enable = 1; diff --git a/drivers/s390/block/dasd_devmap.c b/drivers/s390/block/dasd_devmap.c index 17fdd8c9f74..cf28ccc5794 100644 --- a/drivers/s390/block/dasd_devmap.c +++ b/drivers/s390/block/dasd_devmap.c @@ -25,7 +25,7 @@ #include "dasd_int.h" -kmem_cache_t *dasd_page_cache; +struct kmem_cache *dasd_page_cache; EXPORT_SYMBOL_GPL(dasd_page_cache); /* diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h index 9f52004f6fc..dc5dd509434 100644 --- a/drivers/s390/block/dasd_int.h +++ b/drivers/s390/block/dasd_int.h @@ -474,7 +474,7 @@ extern struct dasd_profile_info_t dasd_global_profile; extern unsigned int dasd_profile_level; extern struct block_device_operations dasd_device_operations; -extern kmem_cache_t *dasd_page_cache; +extern struct kmem_cache *dasd_page_cache; struct dasd_ccw_req * dasd_kmalloc_request(char *, int, int, struct dasd_device *); diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h index 74c0eac083e..32933ed54b8 100644 --- a/drivers/s390/scsi/zfcp_def.h +++ b/drivers/s390/scsi/zfcp_def.h @@ -1032,9 +1032,9 @@ struct zfcp_data { wwn_t init_wwpn; fcp_lun_t init_fcp_lun; char *driver_version; - kmem_cache_t *fsf_req_qtcb_cache; - kmem_cache_t *sr_buffer_cache; - kmem_cache_t *gid_pn_cache; + struct kmem_cache *fsf_req_qtcb_cache; + struct kmem_cache *sr_buffer_cache; + struct kmem_cache *gid_pn_cache; }; /** diff --git a/drivers/scsi/aic94xx/aic94xx.h b/drivers/scsi/aic94xx/aic94xx.h index 71a031df7a3..32f513b1b78 100644 --- a/drivers/scsi/aic94xx/aic94xx.h +++ b/drivers/scsi/aic94xx/aic94xx.h @@ -56,8 +56,8 @@ /* 2*ITNL timeout + 1 second */ #define AIC94XX_SCB_TIMEOUT (5*HZ) -extern kmem_cache_t *asd_dma_token_cache; -extern kmem_cache_t *asd_ascb_cache; +extern struct kmem_cache *asd_dma_token_cache; +extern struct kmem_cache *asd_ascb_cache; extern char sas_addr_str[2*SAS_ADDR_SIZE + 1]; static inline void asd_stringify_sas_addr(char *p, const u8 *sas_addr) diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.c b/drivers/scsi/aic94xx/aic94xx_hwi.c index af7e0113436..da94e126ca8 100644 --- a/drivers/scsi/aic94xx/aic94xx_hwi.c +++ b/drivers/scsi/aic94xx/aic94xx_hwi.c @@ -1047,7 +1047,7 @@ irqreturn_t asd_hw_isr(int irq, void *dev_id) static inline struct asd_ascb *asd_ascb_alloc(struct asd_ha_struct *asd_ha, gfp_t gfp_flags) { - extern kmem_cache_t *asd_ascb_cache; + extern struct kmem_cache *asd_ascb_cache; struct asd_seq_data *seq = &asd_ha->seq; struct asd_ascb *ascb; unsigned long flags; diff --git a/drivers/scsi/aic94xx/aic94xx_init.c b/drivers/scsi/aic94xx/aic94xx_init.c index 42302ef05ee..fbc82b00a41 100644 --- a/drivers/scsi/aic94xx/aic94xx_init.c +++ b/drivers/scsi/aic94xx/aic94xx_init.c @@ -450,8 +450,8 @@ static inline void asd_destroy_ha_caches(struct asd_ha_struct *asd_ha) asd_ha->scb_pool = NULL; } -kmem_cache_t *asd_dma_token_cache; -kmem_cache_t *asd_ascb_cache; +struct kmem_cache *asd_dma_token_cache; +struct kmem_cache *asd_ascb_cache; static int asd_create_global_caches(void) { diff --git a/drivers/scsi/libsas/sas_init.c b/drivers/scsi/libsas/sas_init.c index d65bc4e0f21..2f0c07fc3f4 100644 --- a/drivers/scsi/libsas/sas_init.c +++ b/drivers/scsi/libsas/sas_init.c @@ -36,7 +36,7 @@ #include "../scsi_sas_internal.h" -kmem_cache_t *sas_task_cache; +struct kmem_cache *sas_task_cache; /*------------ SAS addr hash -----------*/ void sas_hash_addr(u8 *hashed, const u8 *sas_addr) diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index cbe0cad83b6..d03523d3bf3 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -24,7 +24,7 @@ char qla2x00_version_str[40]; /* * SRB allocation cache */ -static kmem_cache_t *srb_cachep; +static struct kmem_cache *srb_cachep; /* * Ioctl related information. diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 969c9e43102..9ef693c8809 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -19,7 +19,7 @@ char qla4xxx_version_str[40]; /* * SRB allocation cache */ -static kmem_cache_t *srb_cachep; +static struct kmem_cache *srb_cachep; /* * Module parameter information and variables diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index fafc00deaad..24cffd98ee6 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -136,7 +136,7 @@ const char * scsi_device_type(unsigned type) EXPORT_SYMBOL(scsi_device_type); struct scsi_host_cmd_pool { - kmem_cache_t *slab; + struct kmem_cache *slab; unsigned int users; char *name; unsigned int slab_flags; diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index fb616c69151..1748e27501c 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -36,7 +36,7 @@ struct scsi_host_sg_pool { size_t size; char *name; - kmem_cache_t *slab; + struct kmem_cache *slab; mempool_t *pool; }; @@ -241,7 +241,7 @@ struct scsi_io_context { char sense[SCSI_SENSE_BUFFERSIZE]; }; -static kmem_cache_t *scsi_io_context_cache; +static struct kmem_cache *scsi_io_context_cache; static void scsi_end_async(struct request *req, int uptodate) { diff --git a/drivers/scsi/scsi_tgt_lib.c b/drivers/scsi/scsi_tgt_lib.c index 386dbae17b4..d402aff5f31 100644 --- a/drivers/scsi/scsi_tgt_lib.c +++ b/drivers/scsi/scsi_tgt_lib.c @@ -33,7 +33,7 @@ #include "scsi_tgt_priv.h" static struct workqueue_struct *scsi_tgtd; -static kmem_cache_t *scsi_tgt_cmd_cache; +static struct kmem_cache *scsi_tgt_cmd_cache; /* * TODO: this struct will be killed when the block layer supports large bios diff --git a/drivers/usb/host/hc_crisv10.c b/drivers/usb/host/hc_crisv10.c index 7fd872aa654..9325e46a68c 100644 --- a/drivers/usb/host/hc_crisv10.c +++ b/drivers/usb/host/hc_crisv10.c @@ -275,13 +275,13 @@ static volatile USB_SB_Desc_t TxIntrSB_zout __attribute__ ((aligned (4))); static int zout_buffer[4] __attribute__ ((aligned (4))); /* Cache for allocating new EP and SB descriptors. */ -static kmem_cache_t *usb_desc_cache; +static struct kmem_cache *usb_desc_cache; /* Cache for the registers allocated in the top half. */ -static kmem_cache_t *top_half_reg_cache; +static struct kmem_cache *top_half_reg_cache; /* Cache for the data allocated in the isoc descr top half. */ -static kmem_cache_t *isoc_compl_cache; +static struct kmem_cache *isoc_compl_cache; static struct usb_bus *etrax_usb_bus; diff --git a/drivers/usb/host/uhci-hcd.c b/drivers/usb/host/uhci-hcd.c index 226bf3de8ed..e87692c31be 100644 --- a/drivers/usb/host/uhci-hcd.c +++ b/drivers/usb/host/uhci-hcd.c @@ -81,7 +81,7 @@ MODULE_PARM_DESC(debug, "Debug level"); static char *errbuf; #define ERRBUF_LEN (32 * 1024) -static kmem_cache_t *uhci_up_cachep; /* urb_priv */ +static struct kmem_cache *uhci_up_cachep; /* urb_priv */ static void suspend_rh(struct uhci_hcd *uhci, enum uhci_rh_state new_state); static void wakeup_rh(struct uhci_hcd *uhci); diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c index 46aecc8754f..05cf2c9a8f8 100644 --- a/drivers/usb/mon/mon_text.c +++ b/drivers/usb/mon/mon_text.c @@ -50,7 +50,7 @@ struct mon_event_text { #define SLAB_NAME_SZ 30 struct mon_reader_text { - kmem_cache_t *e_slab; + struct kmem_cache *e_slab; int nevents; struct list_head e_list; struct mon_reader r; /* In C, parent class can be placed anywhere */ @@ -63,7 +63,7 @@ struct mon_reader_text { char slab_name[SLAB_NAME_SZ]; }; -static void mon_text_ctor(void *, kmem_cache_t *, unsigned long); +static void mon_text_ctor(void *, struct kmem_cache *, unsigned long); /* * mon_text_submit @@ -450,7 +450,7 @@ const struct file_operations mon_fops_text = { /* * Slab interface: constructor. */ -static void mon_text_ctor(void *mem, kmem_cache_t *slab, unsigned long sflags) +static void mon_text_ctor(void *mem, struct kmem_cache *slab, unsigned long sflags) { /* * Nothing to initialize. No, really! diff --git a/fs/adfs/super.c b/fs/adfs/super.c index 52eb10ca654..e5a205c9f42 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -212,7 +212,7 @@ static int adfs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static kmem_cache_t *adfs_inode_cachep; +static struct kmem_cache *adfs_inode_cachep; static struct inode *adfs_alloc_inode(struct super_block *sb) { @@ -228,7 +228,7 @@ static void adfs_destroy_inode(struct inode *inode) kmem_cache_free(adfs_inode_cachep, ADFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct adfs_inode_info *ei = (struct adfs_inode_info *) foo; diff --git a/fs/affs/super.c b/fs/affs/super.c index 81c73ec09f6..3de93e79994 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -66,7 +66,7 @@ affs_write_super(struct super_block *sb) pr_debug("AFFS: write_super() at %lu, clean=%d\n", get_seconds(), clean); } -static kmem_cache_t * affs_inode_cachep; +static struct kmem_cache * affs_inode_cachep; static struct inode *affs_alloc_inode(struct super_block *sb) { @@ -83,7 +83,7 @@ static void affs_destroy_inode(struct inode *inode) kmem_cache_free(affs_inode_cachep, AFFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct affs_inode_info *ei = (struct affs_inode_info *) foo; diff --git a/fs/afs/super.c b/fs/afs/super.c index c6ead009bf7..9a351c4c756 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -35,7 +35,7 @@ struct afs_mount_params { struct afs_volume *volume; }; -static void afs_i_init_once(void *foo, kmem_cache_t *cachep, +static void afs_i_init_once(void *foo, struct kmem_cache *cachep, unsigned long flags); static int afs_get_sb(struct file_system_type *fs_type, @@ -65,7 +65,7 @@ static struct super_operations afs_super_ops = { .put_super = afs_put_super, }; -static kmem_cache_t *afs_inode_cachep; +static struct kmem_cache *afs_inode_cachep; static atomic_t afs_count_active_inodes; /*****************************************************************************/ @@ -384,7 +384,7 @@ static void afs_put_super(struct super_block *sb) /* * initialise an inode cache slab element prior to any use */ -static void afs_i_init_once(void *_vnode, kmem_cache_t *cachep, +static void afs_i_init_once(void *_vnode, struct kmem_cache *cachep, unsigned long flags) { struct afs_vnode *vnode = (struct afs_vnode *) _vnode; diff --git a/fs/aio.c b/fs/aio.c index 287a1bc7a18..13aa9298abf 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -47,8 +47,8 @@ unsigned long aio_nr; /* current system wide number of aio requests */ unsigned long aio_max_nr = 0x10000; /* system wide maximum number of aio requests */ /*----end sysctl variables---*/ -static kmem_cache_t *kiocb_cachep; -static kmem_cache_t *kioctx_cachep; +static struct kmem_cache *kiocb_cachep; +static struct kmem_cache *kioctx_cachep; static struct workqueue_struct *aio_wq; diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 995348df94a..bce402eee55 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -61,7 +61,7 @@ static const struct super_operations befs_sops = { }; /* slab cache for befs_inode_info objects */ -static kmem_cache_t *befs_inode_cachep; +static struct kmem_cache *befs_inode_cachep; static const struct file_operations befs_dir_operations = { .read = generic_read_dir, @@ -289,7 +289,7 @@ befs_destroy_inode(struct inode *inode) kmem_cache_free(befs_inode_cachep, BEFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct befs_inode_info *bi = (struct befs_inode_info *) foo; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 2e45123c8f7..eac175ed9f4 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -228,7 +228,7 @@ static void bfs_write_super(struct super_block *s) unlock_kernel(); } -static kmem_cache_t * bfs_inode_cachep; +static struct kmem_cache * bfs_inode_cachep; static struct inode *bfs_alloc_inode(struct super_block *sb) { @@ -244,7 +244,7 @@ static void bfs_destroy_inode(struct inode *inode) kmem_cache_free(bfs_inode_cachep, BFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct bfs_inode_info *bi = foo; diff --git a/fs/bio.c b/fs/bio.c index 50c40ce2cea..7ec737eda72 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -30,7 +30,7 @@ #define BIO_POOL_SIZE 256 -static kmem_cache_t *bio_slab __read_mostly; +static struct kmem_cache *bio_slab __read_mostly; #define BIOVEC_NR_POOLS 6 @@ -44,7 +44,7 @@ mempool_t *bio_split_pool __read_mostly; struct biovec_slab { int nr_vecs; char *name; - kmem_cache_t *slab; + struct kmem_cache *slab; }; /* diff --git a/fs/block_dev.c b/fs/block_dev.c index 063506705f2..13816b4d76f 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -235,7 +235,7 @@ static int block_fsync(struct file *filp, struct dentry *dentry, int datasync) */ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(bdev_lock); -static kmem_cache_t * bdev_cachep __read_mostly; +static struct kmem_cache * bdev_cachep __read_mostly; static struct inode *bdev_alloc_inode(struct super_block *sb) { @@ -253,7 +253,7 @@ static void bdev_destroy_inode(struct inode *inode) kmem_cache_free(bdev_cachep, bdi); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct bdev_inode *ei = (struct bdev_inode *) foo; struct block_device *bdev = &ei->bdev; diff --git a/fs/buffer.c b/fs/buffer.c index 35527dca1db..a8ca0ac2148 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2908,7 +2908,7 @@ asmlinkage long sys_bdflush(int func, long data) /* * Buffer-head allocation */ -static kmem_cache_t *bh_cachep; +static struct kmem_cache *bh_cachep; /* * Once the number of bh's in the machine exceeds this level, we start @@ -2961,7 +2961,7 @@ void free_buffer_head(struct buffer_head *bh) EXPORT_SYMBOL(free_buffer_head); static void -init_buffer_head(void *data, kmem_cache_t *cachep, unsigned long flags) +init_buffer_head(void *data, struct kmem_cache *cachep, unsigned long flags) { if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == SLAB_CTOR_CONSTRUCTOR) { diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 84168629cea..e6b5866e500 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -81,7 +81,7 @@ extern mempool_t *cifs_sm_req_poolp; extern mempool_t *cifs_req_poolp; extern mempool_t *cifs_mid_poolp; -extern kmem_cache_t *cifs_oplock_cachep; +extern struct kmem_cache *cifs_oplock_cachep; static int cifs_read_super(struct super_block *sb, void *data, @@ -232,11 +232,11 @@ static int cifs_permission(struct inode * inode, int mask, struct nameidata *nd) return generic_permission(inode, mask, NULL); } -static kmem_cache_t *cifs_inode_cachep; -static kmem_cache_t *cifs_req_cachep; -static kmem_cache_t *cifs_mid_cachep; -kmem_cache_t *cifs_oplock_cachep; -static kmem_cache_t *cifs_sm_req_cachep; +static struct kmem_cache *cifs_inode_cachep; +static struct kmem_cache *cifs_req_cachep; +static struct kmem_cache *cifs_mid_cachep; +struct kmem_cache *cifs_oplock_cachep; +static struct kmem_cache *cifs_sm_req_cachep; mempool_t *cifs_sm_req_poolp; mempool_t *cifs_req_poolp; mempool_t *cifs_mid_poolp; @@ -668,7 +668,7 @@ const struct file_operations cifs_dir_ops = { }; static void -cifs_init_once(void *inode, kmem_cache_t * cachep, unsigned long flags) +cifs_init_once(void *inode, struct kmem_cache * cachep, unsigned long flags) { struct cifsInodeInfo *cifsi = inode; diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 1f727765a8e..f80007eaebf 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -34,7 +34,7 @@ #include "cifs_debug.h" extern mempool_t *cifs_mid_poolp; -extern kmem_cache_t *cifs_oplock_cachep; +extern struct kmem_cache *cifs_oplock_cachep; static struct mid_q_entry * AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 50cedd2617d..b64659fa82d 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -38,7 +38,7 @@ static void coda_clear_inode(struct inode *); static void coda_put_super(struct super_block *); static int coda_statfs(struct dentry *dentry, struct kstatfs *buf); -static kmem_cache_t * coda_inode_cachep; +static struct kmem_cache * coda_inode_cachep; static struct inode *coda_alloc_inode(struct super_block *sb) { @@ -58,7 +58,7 @@ static void coda_destroy_inode(struct inode *inode) kmem_cache_free(coda_inode_cachep, ITOC(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct coda_inode_info *ei = (struct coda_inode_info *) foo; diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h index 3f4ff7a242b..f92cd303d2c 100644 --- a/fs/configfs/configfs_internal.h +++ b/fs/configfs/configfs_internal.h @@ -49,7 +49,7 @@ struct configfs_dirent { #define CONFIGFS_NOT_PINNED (CONFIGFS_ITEM_ATTR) extern struct vfsmount * configfs_mount; -extern kmem_cache_t *configfs_dir_cachep; +extern struct kmem_cache *configfs_dir_cachep; extern int configfs_is_root(struct config_item *item); diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c index 68bd5c93ca5..ed678529ebb 100644 --- a/fs/configfs/mount.c +++ b/fs/configfs/mount.c @@ -38,7 +38,7 @@ struct vfsmount * configfs_mount = NULL; struct super_block * configfs_sb = NULL; -kmem_cache_t *configfs_dir_cachep; +struct kmem_cache *configfs_dir_cachep; static int configfs_mnt_count = 0; static struct super_operations configfs_ops = { diff --git a/fs/dcache.c b/fs/dcache.c index fd4a428998e..a7c67cee5d8 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -43,7 +43,7 @@ static __cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock); EXPORT_SYMBOL(dcache_lock); -static kmem_cache_t *dentry_cache __read_mostly; +static struct kmem_cache *dentry_cache __read_mostly; #define DNAME_INLINE_LEN (sizeof(struct dentry)-offsetof(struct dentry,d_iname)) @@ -2072,10 +2072,10 @@ static void __init dcache_init(unsigned long mempages) } /* SLAB cache for __getname() consumers */ -kmem_cache_t *names_cachep __read_mostly; +struct kmem_cache *names_cachep __read_mostly; /* SLAB cache for file structures */ -kmem_cache_t *filp_cachep __read_mostly; +struct kmem_cache *filp_cachep __read_mostly; EXPORT_SYMBOL(d_genocide); diff --git a/fs/dcookies.c b/fs/dcookies.c index 0c4b0674854..21af1629f9b 100644 --- a/fs/dcookies.c +++ b/fs/dcookies.c @@ -37,7 +37,7 @@ struct dcookie_struct { static LIST_HEAD(dcookie_users); static DEFINE_MUTEX(dcookie_mutex); -static kmem_cache_t *dcookie_cache __read_mostly; +static struct kmem_cache *dcookie_cache __read_mostly; static struct list_head *dcookie_hashtable __read_mostly; static size_t hash_size __read_mostly; diff --git a/fs/dlm/memory.c b/fs/dlm/memory.c index 989b608fd83..5352b03ff5a 100644 --- a/fs/dlm/memory.c +++ b/fs/dlm/memory.c @@ -15,7 +15,7 @@ #include "config.h" #include "memory.h" -static kmem_cache_t *lkb_cache; +static struct kmem_cache *lkb_cache; int dlm_memory_init(void) diff --git a/fs/dnotify.c b/fs/dnotify.c index e778b1737b7..1f26a2b9eee 100644 --- a/fs/dnotify.c +++ b/fs/dnotify.c @@ -23,7 +23,7 @@ int dir_notify_enable __read_mostly = 1; -static kmem_cache_t *dn_cache __read_mostly; +static struct kmem_cache *dn_cache __read_mostly; static void redo_inode_mask(struct inode *inode) { diff --git a/fs/dquot.c b/fs/dquot.c index c6ae6c0e0cf..f9cd5e23ebd 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -131,7 +131,7 @@ static struct quota_format_type *quota_formats; /* List of registered formats */ static struct quota_module_name module_names[] = INIT_QUOTA_MODULE_NAMES; /* SLAB cache for dquot structures */ -static kmem_cache_t *dquot_cachep; +static struct kmem_cache *dquot_cachep; int register_quota_format(struct quota_format_type *fmt) { diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index a2c6ccbce30..306f8fbd1a0 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -546,7 +546,7 @@ inode_info_init_once(void *vptr, struct kmem_cache *cachep, unsigned long flags) } static struct ecryptfs_cache_info { - kmem_cache_t **cache; + struct kmem_cache **cache; const char *name; size_t size; void (*ctor)(void*, struct kmem_cache *, unsigned long); diff --git a/fs/efs/super.c b/fs/efs/super.c index 69b15a996cf..dfebf21289f 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -52,7 +52,7 @@ static struct pt_types sgi_pt_types[] = { }; -static kmem_cache_t * efs_inode_cachep; +static struct kmem_cache * efs_inode_cachep; static struct inode *efs_alloc_inode(struct super_block *sb) { @@ -68,7 +68,7 @@ static void efs_destroy_inode(struct inode *inode) kmem_cache_free(efs_inode_cachep, INODE_INFO(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct efs_inode_info *ei = (struct efs_inode_info *) foo; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index f5c88435c6b..88a6f8d0b88 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -283,10 +283,10 @@ static struct mutex epmutex; static struct poll_safewake psw; /* Slab cache used to allocate "struct epitem" */ -static kmem_cache_t *epi_cache __read_mostly; +static struct kmem_cache *epi_cache __read_mostly; /* Slab cache used to allocate "struct eppoll_entry" */ -static kmem_cache_t *pwq_cache __read_mostly; +static struct kmem_cache *pwq_cache __read_mostly; /* Virtual fs used to allocate inodes for eventpoll files */ static struct vfsmount *eventpoll_mnt __read_mostly; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 85c237e7385..3aafb1dd917 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -135,7 +135,7 @@ static void ext2_put_super (struct super_block * sb) return; } -static kmem_cache_t * ext2_inode_cachep; +static struct kmem_cache * ext2_inode_cachep; static struct inode *ext2_alloc_inode(struct super_block *sb) { @@ -156,7 +156,7 @@ static void ext2_destroy_inode(struct inode *inode) kmem_cache_free(ext2_inode_cachep, EXT2_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ext2_inode_info *ei = (struct ext2_inode_info *) foo; diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 0cf633f0cfa..9856565d4dd 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -436,7 +436,7 @@ static void ext3_put_super (struct super_block * sb) return; } -static kmem_cache_t *ext3_inode_cachep; +static struct kmem_cache *ext3_inode_cachep; /* * Called inside transaction, so use GFP_NOFS @@ -462,7 +462,7 @@ static void ext3_destroy_inode(struct inode *inode) kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ext3_inode_info *ei = (struct ext3_inode_info *) foo; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index c730cbc8403..f2e8c4aa0fd 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -486,7 +486,7 @@ static void ext4_put_super (struct super_block * sb) return; } -static kmem_cache_t *ext4_inode_cachep; +static struct kmem_cache *ext4_inode_cachep; /* * Called inside transaction, so use GFP_NOFS @@ -513,7 +513,7 @@ static void ext4_destroy_inode(struct inode *inode) kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ext4_inode_info *ei = (struct ext4_inode_info *) foo; diff --git a/fs/fat/cache.c b/fs/fat/cache.c index 8c272278455..05c2941c74f 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -34,9 +34,9 @@ static inline int fat_max_cache(struct inode *inode) return FAT_MAX_CACHE; } -static kmem_cache_t *fat_cache_cachep; +static struct kmem_cache *fat_cache_cachep; -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct fat_cache *cache = (struct fat_cache *)foo; diff --git a/fs/fat/inode.c b/fs/fat/inode.c index b58fd0c9f3c..a9e4688582a 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -477,7 +477,7 @@ static void fat_put_super(struct super_block *sb) kfree(sbi); } -static kmem_cache_t *fat_inode_cachep; +static struct kmem_cache *fat_inode_cachep; static struct inode *fat_alloc_inode(struct super_block *sb) { @@ -493,7 +493,7 @@ static void fat_destroy_inode(struct inode *inode) kmem_cache_free(fat_inode_cachep, MSDOS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct msdos_inode_info *ei = (struct msdos_inode_info *)foo; diff --git a/fs/fcntl.c b/fs/fcntl.c index c03dc9cb21c..4740d35e52c 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -553,7 +553,7 @@ int send_sigurg(struct fown_struct *fown) } static DEFINE_RWLOCK(fasync_lock); -static kmem_cache_t *fasync_cache __read_mostly; +static struct kmem_cache *fasync_cache __read_mostly; /* * fasync_helper() is used by some character device drivers (mainly mice) diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index d2dd0d70007..0b7ae897cb7 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -46,7 +46,7 @@ extern const struct address_space_operations vxfs_immed_aops; extern struct inode_operations vxfs_immed_symlink_iops; -kmem_cache_t *vxfs_inode_cachep; +struct kmem_cache *vxfs_inode_cachep; #ifdef DIAGNOSTIC diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 8c15139f275..357764d85ff 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -19,7 +19,7 @@ MODULE_ALIAS_MISCDEV(FUSE_MINOR); -static kmem_cache_t *fuse_req_cachep; +static struct kmem_cache *fuse_req_cachep; static struct fuse_conn *fuse_get_conn(struct file *file) { diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index e039e2047cc..2bdc652b8b4 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -22,7 +22,7 @@ MODULE_AUTHOR("Miklos Szeredi "); MODULE_DESCRIPTION("Filesystem in Userspace"); MODULE_LICENSE("GPL"); -static kmem_cache_t *fuse_inode_cachep; +static struct kmem_cache *fuse_inode_cachep; struct list_head fuse_conn_list; DEFINE_MUTEX(fuse_mutex); @@ -601,7 +601,7 @@ static struct file_system_type fuse_fs_type = { static decl_subsys(fuse, NULL, NULL); static decl_subsys(connections, NULL, NULL); -static void fuse_inode_init_once(void *foo, kmem_cache_t *cachep, +static void fuse_inode_init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct inode * inode = foo; diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 9889c1eacec..7c1a9e22a52 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -25,7 +25,7 @@ #include "util.h" #include "glock.h" -static void gfs2_init_inode_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void gfs2_init_inode_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct gfs2_inode *ip = foo; if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == @@ -37,7 +37,7 @@ static void gfs2_init_inode_once(void *foo, kmem_cache_t *cachep, unsigned long } } -static void gfs2_init_glock_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void gfs2_init_glock_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct gfs2_glock *gl = foo; if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index 196c604faad..e5707a9f78c 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -23,9 +23,9 @@ #include "lm.h" #include "util.h" -kmem_cache_t *gfs2_glock_cachep __read_mostly; -kmem_cache_t *gfs2_inode_cachep __read_mostly; -kmem_cache_t *gfs2_bufdata_cachep __read_mostly; +struct kmem_cache *gfs2_glock_cachep __read_mostly; +struct kmem_cache *gfs2_inode_cachep __read_mostly; +struct kmem_cache *gfs2_bufdata_cachep __read_mostly; void gfs2_assert_i(struct gfs2_sbd *sdp) { diff --git a/fs/gfs2/util.h b/fs/gfs2/util.h index 76a50899fe9..7984dcf89ad 100644 --- a/fs/gfs2/util.h +++ b/fs/gfs2/util.h @@ -146,9 +146,9 @@ int gfs2_io_error_bh_i(struct gfs2_sbd *sdp, struct buffer_head *bh, gfs2_io_error_bh_i((sdp), (bh), __FUNCTION__, __FILE__, __LINE__); -extern kmem_cache_t *gfs2_glock_cachep; -extern kmem_cache_t *gfs2_inode_cachep; -extern kmem_cache_t *gfs2_bufdata_cachep; +extern struct kmem_cache *gfs2_glock_cachep; +extern struct kmem_cache *gfs2_inode_cachep; +extern struct kmem_cache *gfs2_bufdata_cachep; static inline unsigned int gfs2_tune_get_i(struct gfs2_tune *gt, unsigned int *p) diff --git a/fs/hfs/super.c b/fs/hfs/super.c index ffc6409132c..a3698796600 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -24,7 +24,7 @@ #include "hfs_fs.h" #include "btree.h" -static kmem_cache_t *hfs_inode_cachep; +static struct kmem_cache *hfs_inode_cachep; MODULE_LICENSE("GPL"); @@ -430,7 +430,7 @@ static struct file_system_type hfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfs_init_once(void *p, kmem_cache_t *cachep, unsigned long flags) +static void hfs_init_once(void *p, struct kmem_cache *cachep, unsigned long flags) { struct hfs_inode_info *i = p; diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 4a0c70c76c8..0f513c6bf84 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -434,7 +434,7 @@ MODULE_AUTHOR("Brad Boyer"); MODULE_DESCRIPTION("Extended Macintosh Filesystem"); MODULE_LICENSE("GPL"); -static kmem_cache_t *hfsplus_inode_cachep; +static struct kmem_cache *hfsplus_inode_cachep; static struct inode *hfsplus_alloc_inode(struct super_block *sb) { @@ -467,7 +467,7 @@ static struct file_system_type hfsplus_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfsplus_init_once(void *p, kmem_cache_t *cachep, unsigned long flags) +static void hfsplus_init_once(void *p, struct kmem_cache *cachep, unsigned long flags) { struct hfsplus_inode_info *i = p; diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 46ceadd6f16..34d68e21171 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -160,7 +160,7 @@ static int hpfs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static kmem_cache_t * hpfs_inode_cachep; +static struct kmem_cache * hpfs_inode_cachep; static struct inode *hpfs_alloc_inode(struct super_block *sb) { @@ -177,7 +177,7 @@ static void hpfs_destroy_inode(struct inode *inode) kmem_cache_free(hpfs_inode_cachep, hpfs_i(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct hpfs_inode_info *ei = (struct hpfs_inode_info *) foo; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 36e52173a54..0706f5aac6a 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -513,7 +513,7 @@ static void hugetlbfs_inc_free_inodes(struct hugetlbfs_sb_info *sbinfo) } -static kmem_cache_t *hugetlbfs_inode_cachep; +static struct kmem_cache *hugetlbfs_inode_cachep; static struct inode *hugetlbfs_alloc_inode(struct super_block *sb) { @@ -545,7 +545,7 @@ static const struct address_space_operations hugetlbfs_aops = { }; -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo; diff --git a/fs/inode.c b/fs/inode.c index dd15984d51a..699aa4f7aa7 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -97,7 +97,7 @@ static DEFINE_MUTEX(iprune_mutex); */ struct inodes_stat_t inodes_stat; -static kmem_cache_t * inode_cachep __read_mostly; +static struct kmem_cache * inode_cachep __read_mostly; static struct inode *alloc_inode(struct super_block *sb) { @@ -209,7 +209,7 @@ void inode_init_once(struct inode *inode) EXPORT_SYMBOL(inode_init_once); -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct inode * inode = (struct inode *) foo; diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 017cb0f134d..e1956e6f116 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -34,8 +34,8 @@ #include -static kmem_cache_t *watch_cachep __read_mostly; -static kmem_cache_t *event_cachep __read_mostly; +static struct kmem_cache *watch_cachep __read_mostly; +static struct kmem_cache *event_cachep __read_mostly; static struct vfsmount *inotify_mnt __read_mostly; diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 4b6381cd2cf..ea55b6c469e 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -57,7 +57,7 @@ static void isofs_put_super(struct super_block *sb) static void isofs_read_inode(struct inode *); static int isofs_statfs (struct dentry *, struct kstatfs *); -static kmem_cache_t *isofs_inode_cachep; +static struct kmem_cache *isofs_inode_cachep; static struct inode *isofs_alloc_inode(struct super_block *sb) { @@ -73,7 +73,7 @@ static void isofs_destroy_inode(struct inode *inode) kmem_cache_free(isofs_inode_cachep, ISOFS_I(inode)); } -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct iso_inode_info *ei = foo; diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index b85c686b60d..a8774bed20b 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -1630,7 +1630,7 @@ void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry) #define JBD_MAX_SLABS 5 #define JBD_SLAB_INDEX(size) (size >> 11) -static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; +static struct kmem_cache *jbd_slab[JBD_MAX_SLABS]; static const char *jbd_slab_names[JBD_MAX_SLABS] = { "jbd_1k", "jbd_2k", "jbd_4k", NULL, "jbd_8k" }; @@ -1693,7 +1693,7 @@ void jbd_slab_free(void *ptr, size_t size) /* * Journal_head storage management */ -static kmem_cache_t *journal_head_cache; +static struct kmem_cache *journal_head_cache; #ifdef CONFIG_JBD_DEBUG static atomic_t nr_journal_heads = ATOMIC_INIT(0); #endif @@ -1996,7 +1996,7 @@ static void __exit remove_jbd_proc_entry(void) #endif -kmem_cache_t *jbd_handle_cache; +struct kmem_cache *jbd_handle_cache; static int __init journal_init_handle_cache(void) { diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index c532429d8d9..d204ab394f3 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -70,8 +70,8 @@ #include #endif -static kmem_cache_t *revoke_record_cache; -static kmem_cache_t *revoke_table_cache; +static struct kmem_cache *revoke_record_cache; +static struct kmem_cache *revoke_table_cache; /* Each revoke record represents one single revoked block. During journal replay, this involves recording the transaction ID of the diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index c60f378b0f7..50356019ae3 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1641,7 +1641,7 @@ void * __jbd2_kmalloc (const char *where, size_t size, gfp_t flags, int retry) #define JBD_MAX_SLABS 5 #define JBD_SLAB_INDEX(size) (size >> 11) -static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; +static struct kmem_cache *jbd_slab[JBD_MAX_SLABS]; static const char *jbd_slab_names[JBD_MAX_SLABS] = { "jbd2_1k", "jbd2_2k", "jbd2_4k", NULL, "jbd2_8k" }; @@ -1704,7 +1704,7 @@ void jbd2_slab_free(void *ptr, size_t size) /* * Journal_head storage management */ -static kmem_cache_t *jbd2_journal_head_cache; +static struct kmem_cache *jbd2_journal_head_cache; #ifdef CONFIG_JBD_DEBUG static atomic_t nr_journal_heads = ATOMIC_INIT(0); #endif @@ -2007,7 +2007,7 @@ static void __exit jbd2_remove_jbd_proc_entry(void) #endif -kmem_cache_t *jbd2_handle_cache; +struct kmem_cache *jbd2_handle_cache; static int __init journal_init_handle_cache(void) { diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 380d19917f3..f506646ad0f 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -70,8 +70,8 @@ #include #endif -static kmem_cache_t *jbd2_revoke_record_cache; -static kmem_cache_t *jbd2_revoke_table_cache; +static struct kmem_cache *jbd2_revoke_record_cache; +static struct kmem_cache *jbd2_revoke_table_cache; /* Each revoke record represents one single revoked block. During journal replay, this involves recording the transaction ID of the diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c index 3f7899ea4cb..9f15bce9202 100644 --- a/fs/jffs/inode-v23.c +++ b/fs/jffs/inode-v23.c @@ -61,8 +61,8 @@ static const struct file_operations jffs_dir_operations; static struct inode_operations jffs_dir_inode_operations; static const struct address_space_operations jffs_address_operations; -kmem_cache_t *node_cache = NULL; -kmem_cache_t *fm_cache = NULL; +struct kmem_cache *node_cache = NULL; +struct kmem_cache *fm_cache = NULL; /* Called by the VFS at mount time to initialize the whole file system. */ static int jffs_fill_super(struct super_block *sb, void *data, int silent) diff --git a/fs/jffs/jffs_fm.c b/fs/jffs/jffs_fm.c index 29b68d939bd..077258b2103 100644 --- a/fs/jffs/jffs_fm.c +++ b/fs/jffs/jffs_fm.c @@ -29,8 +29,8 @@ static int jffs_mark_obsolete(struct jffs_fmcontrol *fmc, __u32 fm_offset); static struct jffs_fm *jffs_alloc_fm(void); static void jffs_free_fm(struct jffs_fm *n); -extern kmem_cache_t *fm_cache; -extern kmem_cache_t *node_cache; +extern struct kmem_cache *fm_cache; +extern struct kmem_cache *node_cache; #if CONFIG_JFFS_FS_VERBOSE > 0 void diff --git a/fs/jffs2/malloc.c b/fs/jffs2/malloc.c index 33f29100501..83f9881ec4c 100644 --- a/fs/jffs2/malloc.c +++ b/fs/jffs2/malloc.c @@ -19,16 +19,16 @@ /* These are initialised to NULL in the kernel startup code. If you're porting to other operating systems, beware */ -static kmem_cache_t *full_dnode_slab; -static kmem_cache_t *raw_dirent_slab; -static kmem_cache_t *raw_inode_slab; -static kmem_cache_t *tmp_dnode_info_slab; -static kmem_cache_t *raw_node_ref_slab; -static kmem_cache_t *node_frag_slab; -static kmem_cache_t *inode_cache_slab; +static struct kmem_cache *full_dnode_slab; +static struct kmem_cache *raw_dirent_slab; +static struct kmem_cache *raw_inode_slab; +static struct kmem_cache *tmp_dnode_info_slab; +static struct kmem_cache *raw_node_ref_slab; +static struct kmem_cache *node_frag_slab; +static struct kmem_cache *inode_cache_slab; #ifdef CONFIG_JFFS2_FS_XATTR -static kmem_cache_t *xattr_datum_cache; -static kmem_cache_t *xattr_ref_cache; +static struct kmem_cache *xattr_datum_cache; +static struct kmem_cache *xattr_ref_cache; #endif int __init jffs2_create_slab_caches(void) diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index 77be534ce42..7deb7825402 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -28,7 +28,7 @@ static void jffs2_put_super(struct super_block *); -static kmem_cache_t *jffs2_inode_cachep; +static struct kmem_cache *jffs2_inode_cachep; static struct inode *jffs2_alloc_inode(struct super_block *sb) { @@ -44,7 +44,7 @@ static void jffs2_destroy_inode(struct inode *inode) kmem_cache_free(jffs2_inode_cachep, JFFS2_INODE_INFO(inode)); } -static void jffs2_i_init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void jffs2_i_init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct jffs2_inode_info *ei = (struct jffs2_inode_info *) foo; diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 0cccd1c39d7..b1a1c729601 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -74,7 +74,7 @@ static inline void lock_metapage(struct metapage *mp) } #define METAPOOL_MIN_PAGES 32 -static kmem_cache_t *metapage_cache; +static struct kmem_cache *metapage_cache; static mempool_t *metapage_mempool; #define MPS_PER_PAGE (PAGE_CACHE_SIZE >> L2PSIZE) @@ -180,7 +180,7 @@ static inline void remove_metapage(struct page *page, struct metapage *mp) #endif -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct metapage *mp = (struct metapage *)foo; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 9c1c6e0e633..ca3e1912853 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -44,7 +44,7 @@ MODULE_DESCRIPTION("The Journaled Filesystem (JFS)"); MODULE_AUTHOR("Steve Best/Dave Kleikamp/Barry Arndt, IBM"); MODULE_LICENSE("GPL"); -static kmem_cache_t * jfs_inode_cachep; +static struct kmem_cache * jfs_inode_cachep; static struct super_operations jfs_super_operations; static struct export_operations jfs_export_operations; @@ -748,7 +748,7 @@ static struct file_system_type jfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct jfs_inode_info *jfs_ip = (struct jfs_inode_info *) foo; diff --git a/fs/locks.c b/fs/locks.c index a7b97d50c1e..1cb0c57fedb 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -142,7 +142,7 @@ int lease_break_time = 45; static LIST_HEAD(file_lock_list); static LIST_HEAD(blocked_list); -static kmem_cache_t *filelock_cache __read_mostly; +static struct kmem_cache *filelock_cache __read_mostly; /* Allocate an empty lock structure. */ static struct file_lock *locks_alloc_lock(void) @@ -199,7 +199,7 @@ EXPORT_SYMBOL(locks_init_lock); * Initialises the fields of the file lock which are invariant for * free file_locks. */ -static void init_once(void *foo, kmem_cache_t *cache, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cache, unsigned long flags) { struct file_lock *lock = (struct file_lock *) foo; diff --git a/fs/mbcache.c b/fs/mbcache.c index 0ff71256e65..deeb9dc062d 100644 --- a/fs/mbcache.c +++ b/fs/mbcache.c @@ -85,7 +85,7 @@ struct mb_cache { #ifndef MB_CACHE_INDEXES_COUNT int c_indexes_count; #endif - kmem_cache_t *c_entry_cache; + struct kmem_cache *c_entry_cache; struct list_head *c_block_hash; struct list_head *c_indexes_hash[0]; }; diff --git a/fs/minix/inode.c b/fs/minix/inode.c index ce532c2deda..629e09b38c5 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -51,7 +51,7 @@ static void minix_put_super(struct super_block *sb) return; } -static kmem_cache_t * minix_inode_cachep; +static struct kmem_cache * minix_inode_cachep; static struct inode *minix_alloc_inode(struct super_block *sb) { @@ -67,7 +67,7 @@ static void minix_destroy_inode(struct inode *inode) kmem_cache_free(minix_inode_cachep, minix_i(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct minix_inode_info *ei = (struct minix_inode_info *) foo; diff --git a/fs/namespace.c b/fs/namespace.c index 55442a6cf22..b00ac84ebbd 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -36,7 +36,7 @@ static int event; static struct list_head *mount_hashtable __read_mostly; static int hash_mask __read_mostly, hash_bits __read_mostly; -static kmem_cache_t *mnt_cache __read_mostly; +static struct kmem_cache *mnt_cache __read_mostly; static struct rw_semaphore namespace_sem; /* /sys/fs */ diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index ed84d899220..fae53243bb9 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -40,7 +40,7 @@ static void ncp_delete_inode(struct inode *); static void ncp_put_super(struct super_block *); static int ncp_statfs(struct dentry *, struct kstatfs *); -static kmem_cache_t * ncp_inode_cachep; +static struct kmem_cache * ncp_inode_cachep; static struct inode *ncp_alloc_inode(struct super_block *sb) { @@ -56,7 +56,7 @@ static void ncp_destroy_inode(struct inode *inode) kmem_cache_free(ncp_inode_cachep, NCP_FINFO(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ncp_inode_info *ei = (struct ncp_inode_info *) foo; diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 769fd0a0c77..2f488e1d9b6 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -58,7 +58,7 @@ #define NFSDBG_FACILITY NFSDBG_VFS -static kmem_cache_t *nfs_direct_cachep; +static struct kmem_cache *nfs_direct_cachep; /* * This represents a set of asynchronous requests that we're waiting on diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 6b53aae4ed2..15afa460e62 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -55,7 +55,7 @@ static int nfs_update_inode(struct inode *, struct nfs_fattr *); static void nfs_zap_acl_cache(struct inode *); -static kmem_cache_t * nfs_inode_cachep; +static struct kmem_cache * nfs_inode_cachep; static inline unsigned long nfs_fattr_to_ino_t(struct nfs_fattr *fattr) @@ -1111,7 +1111,7 @@ static inline void nfs4_init_once(struct nfs_inode *nfsi) #endif } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct nfs_inode *nfsi = (struct nfs_inode *) foo; diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index a1561a820ab..3fbfc2f0330 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -20,7 +20,7 @@ #define NFS_PARANOIA 1 -static kmem_cache_t *nfs_page_cachep; +static struct kmem_cache *nfs_page_cachep; static inline struct nfs_page * nfs_page_alloc(void) diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 56f66f0ccb6..244a8c45b68 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -38,7 +38,7 @@ static int nfs_pagein_one(struct list_head *, struct inode *); static const struct rpc_call_ops nfs_read_partial_ops; static const struct rpc_call_ops nfs_read_full_ops; -static kmem_cache_t *nfs_rdata_cachep; +static struct kmem_cache *nfs_rdata_cachep; static mempool_t *nfs_rdata_mempool; #define MIN_POOL_READ (32) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f7dd0d00595..41b07288f99 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -85,7 +85,7 @@ static const struct rpc_call_ops nfs_write_partial_ops; static const struct rpc_call_ops nfs_write_full_ops; static const struct rpc_call_ops nfs_commit_ops; -static kmem_cache_t *nfs_wdata_cachep; +static struct kmem_cache *nfs_wdata_cachep; static mempool_t *nfs_wdata_mempool; static mempool_t *nfs_commit_mempool; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index e431e93ab50..640c92b2a9f 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -84,10 +84,10 @@ static void nfs4_set_recdir(char *recdir); */ static DEFINE_MUTEX(client_mutex); -static kmem_cache_t *stateowner_slab = NULL; -static kmem_cache_t *file_slab = NULL; -static kmem_cache_t *stateid_slab = NULL; -static kmem_cache_t *deleg_slab = NULL; +static struct kmem_cache *stateowner_slab = NULL; +static struct kmem_cache *file_slab = NULL; +static struct kmem_cache *stateid_slab = NULL; +static struct kmem_cache *deleg_slab = NULL; void nfs4_lock_state(void) @@ -1003,7 +1003,7 @@ alloc_init_file(struct inode *ino) } static void -nfsd4_free_slab(kmem_cache_t **slab) +nfsd4_free_slab(struct kmem_cache **slab) { if (*slab == NULL) return; diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index 01f91501f70..941acf14e61 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -66,7 +66,7 @@ static struct file_operations dlmfs_file_operations; static struct inode_operations dlmfs_dir_inode_operations; static struct inode_operations dlmfs_root_inode_operations; static struct inode_operations dlmfs_file_inode_operations; -static kmem_cache_t *dlmfs_inode_cache; +static struct kmem_cache *dlmfs_inode_cache; struct workqueue_struct *user_dlm_worker; @@ -257,7 +257,7 @@ static ssize_t dlmfs_file_write(struct file *filp, } static void dlmfs_init_once(void *foo, - kmem_cache_t *cachep, + struct kmem_cache *cachep, unsigned long flags) { struct dlmfs_inode_private *ip = diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index f784177b624..856012b4fa4 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -221,7 +221,7 @@ EXPORT_SYMBOL_GPL(dlm_dump_all_mles); #endif /* 0 */ -static kmem_cache_t *dlm_mle_cache = NULL; +static struct kmem_cache *dlm_mle_cache = NULL; static void dlm_mle_release(struct kref *kref); diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index fcd4475d1f8..80ac69f11d9 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -61,7 +61,7 @@ struct ocfs2_em_insert_context { struct ocfs2_extent_map_entry *right_ent; }; -static kmem_cache_t *ocfs2_em_ent_cachep = NULL; +static struct kmem_cache *ocfs2_em_ent_cachep = NULL; static struct ocfs2_extent_map_entry * diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h index 46a378fbc40..1a7dd2945b3 100644 --- a/fs/ocfs2/inode.h +++ b/fs/ocfs2/inode.h @@ -106,7 +106,7 @@ static inline struct ocfs2_inode_info *OCFS2_I(struct inode *inode) #define INODE_JOURNAL(i) (OCFS2_I(i)->ip_flags & OCFS2_INODE_JOURNAL) #define SET_INODE_JOURNAL(i) (OCFS2_I(i)->ip_flags |= OCFS2_INODE_JOURNAL) -extern kmem_cache_t *ocfs2_inode_cache; +extern struct kmem_cache *ocfs2_inode_cache; extern const struct address_space_operations ocfs2_aops; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 7574d26ee0f..0524f8a04ad 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -68,7 +68,7 @@ #include "buffer_head_io.h" -static kmem_cache_t *ocfs2_inode_cachep = NULL; +static struct kmem_cache *ocfs2_inode_cachep = NULL; /* OCFS2 needs to schedule several differnt types of work which * require cluster locking, disk I/O, recovery waits, etc. Since these @@ -914,7 +914,7 @@ bail: } static void ocfs2_inode_init_once(void *data, - kmem_cache_t *cachep, + struct kmem_cache *cachep, unsigned long flags) { struct ocfs2_inode_info *oi = data; diff --git a/fs/ocfs2/uptodate.c b/fs/ocfs2/uptodate.c index 9707ed7a320..39814b900fc 100644 --- a/fs/ocfs2/uptodate.c +++ b/fs/ocfs2/uptodate.c @@ -69,7 +69,7 @@ struct ocfs2_meta_cache_item { sector_t c_block; }; -static kmem_cache_t *ocfs2_uptodate_cachep = NULL; +static struct kmem_cache *ocfs2_uptodate_cachep = NULL; void ocfs2_metadata_cache_init(struct inode *inode) { diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 911d1bcfc56..26f44e0074e 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -330,7 +330,7 @@ out: return 0; } -static kmem_cache_t *op_inode_cachep; +static struct kmem_cache *op_inode_cachep; static struct inode *openprom_alloc_inode(struct super_block *sb) { @@ -415,7 +415,7 @@ static struct file_system_type openprom_fs_type = { .kill_sb = kill_anon_super, }; -static void op_inode_init_once(void *data, kmem_cache_t * cachep, unsigned long flags) +static void op_inode_init_once(void *data, struct kmem_cache * cachep, unsigned long flags) { struct op_inode_info *oi = (struct op_inode_info *) data; diff --git a/fs/proc/inode.c b/fs/proc/inode.c index b24cdb2f17c..e26945ba685 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -81,7 +81,7 @@ static void proc_read_inode(struct inode * inode) inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; } -static kmem_cache_t * proc_inode_cachep; +static struct kmem_cache * proc_inode_cachep; static struct inode *proc_alloc_inode(struct super_block *sb) { @@ -105,7 +105,7 @@ static void proc_destroy_inode(struct inode *inode) kmem_cache_free(proc_inode_cachep, PROC_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct proc_inode *ei = (struct proc_inode *) foo; diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 5b943eb11d7..c047dc654d5 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -515,7 +515,7 @@ static void qnx4_read_inode(struct inode *inode) brelse(bh); } -static kmem_cache_t *qnx4_inode_cachep; +static struct kmem_cache *qnx4_inode_cachep; static struct inode *qnx4_alloc_inode(struct super_block *sb) { @@ -531,7 +531,7 @@ static void qnx4_destroy_inode(struct inode *inode) kmem_cache_free(qnx4_inode_cachep, qnx4_i(inode)); } -static void init_once(void *foo, kmem_cache_t * cachep, +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct qnx4_inode_info *ei = (struct qnx4_inode_info *) foo; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 32332516d65..745bc714ba9 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -490,7 +490,7 @@ static void reiserfs_put_super(struct super_block *s) return; } -static kmem_cache_t *reiserfs_inode_cachep; +static struct kmem_cache *reiserfs_inode_cachep; static struct inode *reiserfs_alloc_inode(struct super_block *sb) { @@ -507,7 +507,7 @@ static void reiserfs_destroy_inode(struct inode *inode) kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode)); } -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct reiserfs_inode_info *ei = (struct reiserfs_inode_info *)foo; diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index d1b455f9b66..c5af088d4a4 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -550,7 +550,7 @@ romfs_read_inode(struct inode *i) } } -static kmem_cache_t * romfs_inode_cachep; +static struct kmem_cache * romfs_inode_cachep; static struct inode *romfs_alloc_inode(struct super_block *sb) { @@ -566,7 +566,7 @@ static void romfs_destroy_inode(struct inode *inode) kmem_cache_free(romfs_inode_cachep, ROMFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct romfs_inode_info *ei = (struct romfs_inode_info *) foo; diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 22161710348..4af4cd729a5 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -50,7 +50,7 @@ static void smb_put_super(struct super_block *); static int smb_statfs(struct dentry *, struct kstatfs *); static int smb_show_options(struct seq_file *, struct vfsmount *); -static kmem_cache_t *smb_inode_cachep; +static struct kmem_cache *smb_inode_cachep; static struct inode *smb_alloc_inode(struct super_block *sb) { @@ -66,7 +66,7 @@ static void smb_destroy_inode(struct inode *inode) kmem_cache_free(smb_inode_cachep, SMB_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct smb_inode_info *ei = (struct smb_inode_info *) foo; unsigned long flagmask = SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR; diff --git a/fs/smbfs/request.c b/fs/smbfs/request.c index 3eb1402191b..a4bcae8a9af 100644 --- a/fs/smbfs/request.c +++ b/fs/smbfs/request.c @@ -25,7 +25,7 @@ #define ROUND_UP(x) (((x)+3) & ~3) /* cache for request structures */ -static kmem_cache_t *req_cachep; +static struct kmem_cache *req_cachep; static int smb_request_send_req(struct smb_request *req); diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index 20551a1b8a5..e503f858fba 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -16,7 +16,7 @@ struct vfsmount *sysfs_mount; struct super_block * sysfs_sb = NULL; -kmem_cache_t *sysfs_dir_cachep; +struct kmem_cache *sysfs_dir_cachep; static struct super_operations sysfs_ops = { .statfs = simple_statfs, diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index 6f3d6bd5288..bd7cec295da 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -1,6 +1,6 @@ extern struct vfsmount * sysfs_mount; -extern kmem_cache_t *sysfs_dir_cachep; +extern struct kmem_cache *sysfs_dir_cachep; extern struct inode * sysfs_new_inode(mode_t mode, struct sysfs_dirent *); extern int sysfs_create(struct dentry *, int mode, int (*init)(struct inode *)); diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index a6ca12b747c..ead9864567e 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -301,7 +301,7 @@ static void sysv_delete_inode(struct inode *inode) unlock_kernel(); } -static kmem_cache_t *sysv_inode_cachep; +static struct kmem_cache *sysv_inode_cachep; static struct inode *sysv_alloc_inode(struct super_block *sb) { @@ -318,7 +318,7 @@ static void sysv_destroy_inode(struct inode *inode) kmem_cache_free(sysv_inode_cachep, SYSV_I(inode)); } -static void init_once(void *p, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *p, struct kmem_cache *cachep, unsigned long flags) { struct sysv_inode_info *si = (struct sysv_inode_info *)p; diff --git a/fs/udf/super.c b/fs/udf/super.c index e50f24221de..397f54aaf66 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -107,7 +107,7 @@ static struct file_system_type udf_fstype = { .fs_flags = FS_REQUIRES_DEV, }; -static kmem_cache_t * udf_inode_cachep; +static struct kmem_cache * udf_inode_cachep; static struct inode *udf_alloc_inode(struct super_block *sb) { @@ -130,7 +130,7 @@ static void udf_destroy_inode(struct inode *inode) kmem_cache_free(udf_inode_cachep, UDF_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct udf_inode_info *ei = (struct udf_inode_info *) foo; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 85a88c0c5e6..0b18d2cc2ef 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1204,7 +1204,7 @@ static int ufs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static kmem_cache_t * ufs_inode_cachep; +static struct kmem_cache * ufs_inode_cachep; static struct inode *ufs_alloc_inode(struct super_block *sb) { @@ -1221,7 +1221,7 @@ static void ufs_destroy_inode(struct inode *inode) kmem_cache_free(ufs_inode_cachep, UFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ufs_inode_info *ei = (struct ufs_inode_info *) foo; diff --git a/include/acpi/platform/aclinux.h b/include/acpi/platform/aclinux.h index 47faf27913a..7f1e92930b6 100644 --- a/include/acpi/platform/aclinux.h +++ b/include/acpi/platform/aclinux.h @@ -64,7 +64,7 @@ /* Host-dependent types and defines */ #define ACPI_MACHINE_WIDTH BITS_PER_LONG -#define acpi_cache_t kmem_cache_t +#define acpi_cache_t struct kmem_cache #define acpi_spinlock spinlock_t * #define ACPI_EXPORT_SYMBOL(symbol) EXPORT_SYMBOL(symbol); #define strtoul simple_strtoul diff --git a/include/asm-arm26/pgalloc.h b/include/asm-arm26/pgalloc.h index 6437167b1ff..7725af3ddb4 100644 --- a/include/asm-arm26/pgalloc.h +++ b/include/asm-arm26/pgalloc.h @@ -15,7 +15,7 @@ #include #include -extern kmem_cache_t *pte_cache; +extern struct kmem_cache *pte_cache; static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long addr){ return kmem_cache_alloc(pte_cache, GFP_KERNEL); diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h index 7d398f493dd..bfee7ddfff5 100644 --- a/include/asm-i386/pgtable.h +++ b/include/asm-i386/pgtable.h @@ -34,14 +34,14 @@ struct vm_area_struct; #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page)) extern unsigned long empty_zero_page[1024]; extern pgd_t swapper_pg_dir[1024]; -extern kmem_cache_t *pgd_cache; -extern kmem_cache_t *pmd_cache; +extern struct kmem_cache *pgd_cache; +extern struct kmem_cache *pmd_cache; extern spinlock_t pgd_lock; extern struct page *pgd_list; -void pmd_ctor(void *, kmem_cache_t *, unsigned long); -void pgd_ctor(void *, kmem_cache_t *, unsigned long); -void pgd_dtor(void *, kmem_cache_t *, unsigned long); +void pmd_ctor(void *, struct kmem_cache *, unsigned long); +void pgd_ctor(void *, struct kmem_cache *, unsigned long); +void pgd_dtor(void *, struct kmem_cache *, unsigned long); void pgtable_cache_init(void); void paging_init(void); diff --git a/include/asm-powerpc/pgalloc.h b/include/asm-powerpc/pgalloc.h index ae63db7b3e7..b0830db68f8 100644 --- a/include/asm-powerpc/pgalloc.h +++ b/include/asm-powerpc/pgalloc.h @@ -11,7 +11,7 @@ #include #include -extern kmem_cache_t *pgtable_cache[]; +extern struct kmem_cache *pgtable_cache[]; #ifdef CONFIG_PPC_64K_PAGES #define PTE_CACHE_NUM 0 diff --git a/include/asm-sparc64/pgalloc.h b/include/asm-sparc64/pgalloc.h index 010f9cd0a67..5891ff7ba76 100644 --- a/include/asm-sparc64/pgalloc.h +++ b/include/asm-sparc64/pgalloc.h @@ -13,7 +13,7 @@ #include /* Page table allocation/freeing. */ -extern kmem_cache_t *pgtable_cache; +extern struct kmem_cache *pgtable_cache; static inline pgd_t *pgd_alloc(struct mm_struct *mm) { diff --git a/include/linux/delayacct.h b/include/linux/delayacct.h index 561e2a77805..55d1ca5e60f 100644 --- a/include/linux/delayacct.h +++ b/include/linux/delayacct.h @@ -30,7 +30,7 @@ #ifdef CONFIG_TASK_DELAY_ACCT extern int delayacct_on; /* Delay accounting turned on/off */ -extern kmem_cache_t *delayacct_cache; +extern struct kmem_cache *delayacct_cache; extern void delayacct_init(void); extern void __delayacct_tsk_init(struct task_struct *); extern void __delayacct_tsk_exit(struct task_struct *); diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 1fb02e17f6f..2514f4e286b 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -490,7 +490,7 @@ struct i2o_dma { */ struct i2o_pool { char *name; - kmem_cache_t *slab; + struct kmem_cache *slab; mempool_t *mempool; }; diff --git a/include/linux/jbd.h b/include/linux/jbd.h index fe89444b1c6..dacd566c9f6 100644 --- a/include/linux/jbd.h +++ b/include/linux/jbd.h @@ -949,7 +949,7 @@ void journal_put_journal_head(struct journal_head *jh); /* * handle management */ -extern kmem_cache_t *jbd_handle_cache; +extern struct kmem_cache *jbd_handle_cache; static inline handle_t *jbd_alloc_handle(gfp_t gfp_flags) { diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index ddb12879578..98a0ae52660 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -958,7 +958,7 @@ void jbd2_journal_put_journal_head(struct journal_head *jh); /* * handle management */ -extern kmem_cache_t *jbd2_handle_cache; +extern struct kmem_cache *jbd2_handle_cache; static inline handle_t *jbd_alloc_handle(gfp_t gfp_flags) { diff --git a/include/linux/raid/raid5.h b/include/linux/raid/raid5.h index f13299a1559..03636d7918f 100644 --- a/include/linux/raid/raid5.h +++ b/include/linux/raid/raid5.h @@ -235,7 +235,7 @@ struct raid5_private_data { */ int active_name; char cache_name[2][20]; - kmem_cache_t *slab_cache; /* for allocating stripes */ + struct kmem_cache *slab_cache; /* for allocating stripes */ int seq_flush, seq_write; int quiesce; diff --git a/include/linux/rmap.h b/include/linux/rmap.h index 61c2ab634b0..36f850373d2 100644 --- a/include/linux/rmap.h +++ b/include/linux/rmap.h @@ -30,7 +30,7 @@ struct anon_vma { #ifdef CONFIG_MMU -extern kmem_cache_t *anon_vma_cachep; +extern struct kmem_cache *anon_vma_cachep; static inline struct anon_vma *anon_vma_alloc(void) { diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 1d649f3eb00..4ff3940210d 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -345,7 +345,7 @@ static inline struct sk_buff *alloc_skb_fclone(unsigned int size, return __alloc_skb(size, priority, 1, -1); } -extern struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp, +extern struct sk_buff *alloc_skb_from_cache(struct kmem_cache *cp, unsigned int size, gfp_t priority); extern void kfree_skbmem(struct sk_buff *skb); diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index f81a5af8a4f..ce8a912e542 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -12,7 +12,7 @@ #include #ifdef CONFIG_TASKSTATS -extern kmem_cache_t *taskstats_cache; +extern struct kmem_cache *taskstats_cache; extern struct mutex taskstats_exit_mutex; static inline void taskstats_exit_free(struct taskstats *tidstats) diff --git a/include/net/dst.h b/include/net/dst.h index e156e38e4ac..62b7e7598e9 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -98,7 +98,7 @@ struct dst_ops int entry_size; atomic_t entries; - kmem_cache_t *kmem_cachep; + struct kmem_cache *kmem_cachep; }; #ifdef __KERNEL__ diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index a9eb2eaf094..34cc76e3ddb 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -125,7 +125,7 @@ struct inet_hashinfo { rwlock_t lhash_lock ____cacheline_aligned; atomic_t lhash_users; wait_queue_head_t lhash_wait; - kmem_cache_t *bind_bucket_cachep; + struct kmem_cache *bind_bucket_cachep; }; static inline struct inet_ehash_bucket *inet_ehash_bucket( @@ -136,10 +136,10 @@ static inline struct inet_ehash_bucket *inet_ehash_bucket( } extern struct inet_bind_bucket * - inet_bind_bucket_create(kmem_cache_t *cachep, + inet_bind_bucket_create(struct kmem_cache *cachep, struct inet_bind_hashbucket *head, const unsigned short snum); -extern void inet_bind_bucket_destroy(kmem_cache_t *cachep, +extern void inet_bind_bucket_destroy(struct kmem_cache *cachep, struct inet_bind_bucket *tb); static inline int inet_bhashfn(const __u16 lport, const int bhash_size) diff --git a/include/net/neighbour.h b/include/net/neighbour.h index c8aacbd2e33..23967031ddb 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -160,7 +160,7 @@ struct neigh_table atomic_t entries; rwlock_t lock; unsigned long last_rand; - kmem_cache_t *kmem_cachep; + struct kmem_cache *kmem_cachep; struct neigh_statistics *stats; struct neighbour **hash_buckets; unsigned int hash_mask; diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index cef3136e22a..41bcc9eb420 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -7,7 +7,7 @@ #include extern struct list_head nf_conntrack_expect_list; -extern kmem_cache_t *nf_conntrack_expect_cachep; +extern struct kmem_cache *nf_conntrack_expect_cachep; extern struct file_operations exp_file_ops; struct nf_conntrack_expect diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 426f0fe774e..7aed02ce2b6 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -29,7 +29,7 @@ struct proto; struct request_sock_ops { int family; int obj_size; - kmem_cache_t *slab; + struct kmem_cache *slab; int (*rtx_syn_ack)(struct sock *sk, struct request_sock *req, struct dst_entry *dst); diff --git a/include/net/sock.h b/include/net/sock.h index fe3a33fad03..730899ce516 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -571,7 +571,7 @@ struct proto { int *sysctl_rmem; int max_header; - kmem_cache_t *slab; + struct kmem_cache *slab; unsigned int obj_size; atomic_t *orphan_count; diff --git a/include/net/timewait_sock.h b/include/net/timewait_sock.h index d7a306ea560..1e1ee3253fd 100644 --- a/include/net/timewait_sock.h +++ b/include/net/timewait_sock.h @@ -15,7 +15,7 @@ #include struct timewait_sock_ops { - kmem_cache_t *twsk_slab; + struct kmem_cache *twsk_slab; unsigned int twsk_obj_size; int (*twsk_unique)(struct sock *sk, struct sock *sktw, void *twp); diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h index 9233ed5de66..0c775fceb67 100644 --- a/include/scsi/libsas.h +++ b/include/scsi/libsas.h @@ -557,7 +557,7 @@ struct sas_task { static inline struct sas_task *sas_alloc_task(gfp_t flags) { - extern kmem_cache_t *sas_task_cache; + extern struct kmem_cache *sas_task_cache; struct sas_task *task = kmem_cache_alloc(sas_task_cache, flags); if (task) { @@ -575,7 +575,7 @@ static inline struct sas_task *sas_alloc_task(gfp_t flags) static inline void sas_free_task(struct sas_task *task) { if (task) { - extern kmem_cache_t *sas_task_cache; + extern struct kmem_cache *sas_task_cache; BUG_ON(!list_empty(&task->list)); kmem_cache_free(sas_task_cache, task); } diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 813bb941342..3acc1661e51 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -90,7 +90,7 @@ static struct super_operations mqueue_super_ops; static void remove_notification(struct mqueue_inode_info *info); static spinlock_t mq_lock; -static kmem_cache_t *mqueue_inode_cachep; +static struct kmem_cache *mqueue_inode_cachep; static struct vfsmount *mqueue_mnt; static unsigned int queues_count; @@ -211,7 +211,7 @@ static int mqueue_get_sb(struct file_system_type *fs_type, return get_sb_single(fs_type, flags, data, mqueue_fill_super, mnt); } -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct mqueue_inode_info *p = (struct mqueue_inode_info *) foo; diff --git a/kernel/delayacct.c b/kernel/delayacct.c index 70e9ec60308..766d5912b26 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -20,7 +20,7 @@ #include int delayacct_on __read_mostly = 1; /* Delay accounting turned on/off */ -kmem_cache_t *delayacct_cache; +struct kmem_cache *delayacct_cache; static int __init delayacct_setup_disable(char *str) { diff --git a/kernel/fork.c b/kernel/fork.c index 711aa5f10da..2cf74edd329 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -82,26 +82,26 @@ int nr_processes(void) #ifndef __HAVE_ARCH_TASK_STRUCT_ALLOCATOR # define alloc_task_struct() kmem_cache_alloc(task_struct_cachep, GFP_KERNEL) # define free_task_struct(tsk) kmem_cache_free(task_struct_cachep, (tsk)) -static kmem_cache_t *task_struct_cachep; +static struct kmem_cache *task_struct_cachep; #endif /* SLAB cache for signal_struct structures (tsk->signal) */ -static kmem_cache_t *signal_cachep; +static struct kmem_cache *signal_cachep; /* SLAB cache for sighand_struct structures (tsk->sighand) */ -kmem_cache_t *sighand_cachep; +struct kmem_cache *sighand_cachep; /* SLAB cache for files_struct structures (tsk->files) */ -kmem_cache_t *files_cachep; +struct kmem_cache *files_cachep; /* SLAB cache for fs_struct structures (tsk->fs) */ -kmem_cache_t *fs_cachep; +struct kmem_cache *fs_cachep; /* SLAB cache for vm_area_struct structures */ -kmem_cache_t *vm_area_cachep; +struct kmem_cache *vm_area_cachep; /* SLAB cache for mm_struct structures (tsk->mm) */ -static kmem_cache_t *mm_cachep; +static struct kmem_cache *mm_cachep; void free_task(struct task_struct *tsk) { @@ -1421,7 +1421,7 @@ long do_fork(unsigned long clone_flags, #define ARCH_MIN_MMSTRUCT_ALIGN 0 #endif -static void sighand_ctor(void *data, kmem_cache_t *cachep, unsigned long flags) +static void sighand_ctor(void *data, struct kmem_cache *cachep, unsigned long flags) { struct sighand_struct *sighand = data; diff --git a/kernel/pid.c b/kernel/pid.c index b914392085f..a48879b0b92 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -31,7 +31,7 @@ #define pid_hashfn(nr) hash_long((unsigned long)nr, pidhash_shift) static struct hlist_head *pid_hash; static int pidhash_shift; -static kmem_cache_t *pid_cachep; +static struct kmem_cache *pid_cachep; int pid_max = PID_MAX_DEFAULT; diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index 9cbb5d1be06..5fe87de10ff 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -70,7 +70,7 @@ /* * Lets keep our timers in a slab cache :-) */ -static kmem_cache_t *posix_timers_cache; +static struct kmem_cache *posix_timers_cache; static struct idr posix_timers_id; static DEFINE_SPINLOCK(idr_lock); diff --git a/kernel/signal.c b/kernel/signal.c index df18c167a2a..8e19d278548 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -33,7 +33,7 @@ * SLAB caches for signal bits. */ -static kmem_cache_t *sigqueue_cachep; +static struct kmem_cache *sigqueue_cachep; /* * In POSIX a signal is sent either to a specific thread (Linux task) diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 1b2b326cf70..f5f92014ae9 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -34,7 +34,7 @@ static DEFINE_PER_CPU(__u32, taskstats_seqnum) = { 0 }; static int family_registered; -kmem_cache_t *taskstats_cache; +struct kmem_cache *taskstats_cache; static struct genl_family family = { .id = GENL_ID_GENERATE, diff --git a/kernel/user.c b/kernel/user.c index c1f93c164c9..4869563080e 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -26,7 +26,7 @@ #define __uidhashfn(uid) (((uid >> UIDHASH_BITS) + uid) & UIDHASH_MASK) #define uidhashentry(uid) (uidhash_table + __uidhashfn((uid))) -static kmem_cache_t *uid_cachep; +static struct kmem_cache *uid_cachep; static struct list_head uidhash_table[UIDHASH_SZ]; /* diff --git a/lib/idr.c b/lib/idr.c index 16d2143fea4..71853531d3b 100644 --- a/lib/idr.c +++ b/lib/idr.c @@ -33,7 +33,7 @@ #include #include -static kmem_cache_t *idr_layer_cache; +static struct kmem_cache *idr_layer_cache; static struct idr_layer *alloc_layer(struct idr *idp) { @@ -445,7 +445,7 @@ void *idr_replace(struct idr *idp, void *ptr, int id) } EXPORT_SYMBOL(idr_replace); -static void idr_cache_ctor(void * idr_layer, kmem_cache_t *idr_layer_cache, +static void idr_cache_ctor(void * idr_layer, struct kmem_cache *idr_layer_cache, unsigned long flags) { memset(idr_layer, 0, sizeof(struct idr_layer)); diff --git a/lib/radix-tree.c b/lib/radix-tree.c index aa9bfd0bdbd..9eb25955019 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -63,7 +63,7 @@ static unsigned long height_to_maxindex[RADIX_TREE_MAX_PATH] __read_mostly; /* * Radix tree node cache. */ -static kmem_cache_t *radix_tree_node_cachep; +static struct kmem_cache *radix_tree_node_cachep; /* * Per-cpu pool of preloaded nodes @@ -846,7 +846,7 @@ int radix_tree_tagged(struct radix_tree_root *root, unsigned int tag) EXPORT_SYMBOL(radix_tree_tagged); static void -radix_tree_node_ctor(void *node, kmem_cache_t *cachep, unsigned long flags) +radix_tree_node_ctor(void *node, struct kmem_cache *cachep, unsigned long flags) { memset(node, 0, sizeof(struct radix_tree_node)); } diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index d9f04864d15..8ca448db7a0 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -23,7 +23,7 @@ #include #include "br_private.h" -static kmem_cache_t *br_fdb_cache __read_mostly; +static struct kmem_cache *br_fdb_cache __read_mostly; static int fdb_insert(struct net_bridge *br, struct net_bridge_port *source, const unsigned char *addr); diff --git a/net/core/flow.c b/net/core/flow.c index 5df3e297f81..104c25d00a1 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -44,7 +44,7 @@ static DEFINE_PER_CPU(struct flow_cache_entry **, flow_tables) = { NULL }; #define flow_table(cpu) (per_cpu(flow_tables, cpu)) -static kmem_cache_t *flow_cachep __read_mostly; +static struct kmem_cache *flow_cachep __read_mostly; static int flow_lwm, flow_hwm; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 7217fb8928f..de7801d589e 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -68,8 +68,8 @@ #include "kmap_skb.h" -static kmem_cache_t *skbuff_head_cache __read_mostly; -static kmem_cache_t *skbuff_fclone_cache __read_mostly; +static struct kmem_cache *skbuff_head_cache __read_mostly; +static struct kmem_cache *skbuff_fclone_cache __read_mostly; /* * Keep out-of-line to prevent kernel bloat. @@ -144,7 +144,7 @@ EXPORT_SYMBOL(skb_truesize_bug); struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, int fclone, int node) { - kmem_cache_t *cache; + struct kmem_cache *cache; struct skb_shared_info *shinfo; struct sk_buff *skb; u8 *data; @@ -211,7 +211,7 @@ nodata: * Buffers may only be allocated from interrupts using a @gfp_mask of * %GFP_ATOMIC. */ -struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp, +struct sk_buff *alloc_skb_from_cache(struct kmem_cache *cp, unsigned int size, gfp_t gfp_mask) { diff --git a/net/core/sock.c b/net/core/sock.c index 419c7d3289c..4a432da441e 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -841,7 +841,7 @@ struct sock *sk_alloc(int family, gfp_t priority, struct proto *prot, int zero_it) { struct sock *sk = NULL; - kmem_cache_t *slab = prot->slab; + struct kmem_cache *slab = prot->slab; if (slab != NULL) sk = kmem_cache_alloc(slab, priority); diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c index bdf1bb7a82c..1f4727ddbdb 100644 --- a/net/dccp/ackvec.c +++ b/net/dccp/ackvec.c @@ -21,8 +21,8 @@ #include -static kmem_cache_t *dccp_ackvec_slab; -static kmem_cache_t *dccp_ackvec_record_slab; +static struct kmem_cache *dccp_ackvec_slab; +static struct kmem_cache *dccp_ackvec_record_slab; static struct dccp_ackvec_record *dccp_ackvec_record_new(void) { diff --git a/net/dccp/ccid.c b/net/dccp/ccid.c index ff05e59043c..d8cf92f09e6 100644 --- a/net/dccp/ccid.c +++ b/net/dccp/ccid.c @@ -55,9 +55,9 @@ static inline void ccids_read_unlock(void) #define ccids_read_unlock() do { } while(0) #endif -static kmem_cache_t *ccid_kmem_cache_create(int obj_size, const char *fmt,...) +static struct kmem_cache *ccid_kmem_cache_create(int obj_size, const char *fmt,...) { - kmem_cache_t *slab; + struct kmem_cache *slab; char slab_name_fmt[32], *slab_name; va_list args; @@ -75,7 +75,7 @@ static kmem_cache_t *ccid_kmem_cache_create(int obj_size, const char *fmt,...) return slab; } -static void ccid_kmem_cache_destroy(kmem_cache_t *slab) +static void ccid_kmem_cache_destroy(struct kmem_cache *slab) { if (slab != NULL) { const char *name = kmem_cache_name(slab); diff --git a/net/dccp/ccid.h b/net/dccp/ccid.h index c7c29514dce..bcc2d12ae81 100644 --- a/net/dccp/ccid.h +++ b/net/dccp/ccid.h @@ -27,9 +27,9 @@ struct ccid_operations { unsigned char ccid_id; const char *ccid_name; struct module *ccid_owner; - kmem_cache_t *ccid_hc_rx_slab; + struct kmem_cache *ccid_hc_rx_slab; __u32 ccid_hc_rx_obj_size; - kmem_cache_t *ccid_hc_tx_slab; + struct kmem_cache *ccid_hc_tx_slab; __u32 ccid_hc_tx_obj_size; int (*ccid_hc_rx_init)(struct ccid *ccid, struct sock *sk); int (*ccid_hc_tx_init)(struct ccid *ccid, struct sock *sk); diff --git a/net/dccp/ccids/lib/loss_interval.h b/net/dccp/ccids/lib/loss_interval.h index 0ae85f0340b..eb257014dd7 100644 --- a/net/dccp/ccids/lib/loss_interval.h +++ b/net/dccp/ccids/lib/loss_interval.h @@ -20,7 +20,7 @@ #define DCCP_LI_HIST_IVAL_F_LENGTH 8 struct dccp_li_hist { - kmem_cache_t *dccplih_slab; + struct kmem_cache *dccplih_slab; }; extern struct dccp_li_hist *dccp_li_hist_new(const char *name); diff --git a/net/dccp/ccids/lib/packet_history.h b/net/dccp/ccids/lib/packet_history.h index 067cf1c85a3..9a8bcf224aa 100644 --- a/net/dccp/ccids/lib/packet_history.h +++ b/net/dccp/ccids/lib/packet_history.h @@ -68,14 +68,14 @@ struct dccp_rx_hist_entry { }; struct dccp_tx_hist { - kmem_cache_t *dccptxh_slab; + struct kmem_cache *dccptxh_slab; }; extern struct dccp_tx_hist *dccp_tx_hist_new(const char *name); extern void dccp_tx_hist_delete(struct dccp_tx_hist *hist); struct dccp_rx_hist { - kmem_cache_t *dccprxh_slab; + struct kmem_cache *dccprxh_slab; }; extern struct dccp_rx_hist *dccp_rx_hist_new(const char *name); diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c index 101e5ccaf09..13b2421991b 100644 --- a/net/decnet/dn_table.c +++ b/net/decnet/dn_table.c @@ -79,7 +79,7 @@ for( ; ((f) = *(fp)) != NULL && dn_key_eq((f)->fn_key, (key)); (fp) = &(f)->fn_n static struct hlist_head dn_fib_table_hash[DN_FIB_TABLE_HASHSZ]; static DEFINE_RWLOCK(dn_fib_tables_lock); -static kmem_cache_t *dn_hash_kmem __read_mostly; +static struct kmem_cache *dn_hash_kmem __read_mostly; static int dn_fib_hash_zombies; static inline dn_fib_idx_t dn_hash(dn_fib_key_t key, struct dn_zone *dz) diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c index 4463443e42c..648f47c1c39 100644 --- a/net/ipv4/fib_hash.c +++ b/net/ipv4/fib_hash.c @@ -45,8 +45,8 @@ #include "fib_lookup.h" -static kmem_cache_t *fn_hash_kmem __read_mostly; -static kmem_cache_t *fn_alias_kmem __read_mostly; +static struct kmem_cache *fn_hash_kmem __read_mostly; +static struct kmem_cache *fn_alias_kmem __read_mostly; struct fib_node { struct hlist_node fn_hash; diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 6be6caf1af3..cfb249cc0a5 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -172,7 +172,7 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn); static struct tnode *halve(struct trie *t, struct tnode *tn); static void tnode_free(struct tnode *tn); -static kmem_cache_t *fn_alias_kmem __read_mostly; +static struct kmem_cache *fn_alias_kmem __read_mostly; static struct trie *trie_local = NULL, *trie_main = NULL; diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index bd6c9bc4189..8c79c8a4ea5 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -27,7 +27,7 @@ * Allocate and initialize a new local port bind bucket. * The bindhash mutex for snum's hash chain must be held here. */ -struct inet_bind_bucket *inet_bind_bucket_create(kmem_cache_t *cachep, +struct inet_bind_bucket *inet_bind_bucket_create(struct kmem_cache *cachep, struct inet_bind_hashbucket *head, const unsigned short snum) { @@ -45,7 +45,7 @@ struct inet_bind_bucket *inet_bind_bucket_create(kmem_cache_t *cachep, /* * Caller must hold hashbucket lock for this tb with local BH disabled */ -void inet_bind_bucket_destroy(kmem_cache_t *cachep, struct inet_bind_bucket *tb) +void inet_bind_bucket_destroy(struct kmem_cache *cachep, struct inet_bind_bucket *tb) { if (hlist_empty(&tb->owners)) { __hlist_del(&tb->node); diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index f072f3875af..711eb6d0285 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -73,7 +73,7 @@ /* Exported for inet_getid inline function. */ DEFINE_SPINLOCK(inet_peer_idlock); -static kmem_cache_t *peer_cachep __read_mostly; +static struct kmem_cache *peer_cachep __read_mostly; #define node_height(x) x->avl_height static struct inet_peer peer_fake_node = { diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index efcf45ecc81..ecb5422ea23 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -105,7 +105,7 @@ static DEFINE_SPINLOCK(mfc_unres_lock); In this case data path is free of exclusive locks at all. */ -static kmem_cache_t *mrt_cachep __read_mostly; +static struct kmem_cache *mrt_cachep __read_mostly; static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local); static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert); diff --git a/net/ipv4/ipvs/ip_vs_conn.c b/net/ipv4/ipvs/ip_vs_conn.c index 8832eb517d5..8086787a2c5 100644 --- a/net/ipv4/ipvs/ip_vs_conn.c +++ b/net/ipv4/ipvs/ip_vs_conn.c @@ -44,7 +44,7 @@ static struct list_head *ip_vs_conn_tab; /* SLAB cache for IPVS connections */ -static kmem_cache_t *ip_vs_conn_cachep __read_mostly; +static struct kmem_cache *ip_vs_conn_cachep __read_mostly; /* counter for current IPVS connections */ static atomic_t ip_vs_conn_count = ATOMIC_INIT(0); diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index f4b0e68a16d..8556a4f4f60 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -65,8 +65,8 @@ static LIST_HEAD(helpers); unsigned int ip_conntrack_htable_size __read_mostly = 0; int ip_conntrack_max __read_mostly; struct list_head *ip_conntrack_hash __read_mostly; -static kmem_cache_t *ip_conntrack_cachep __read_mostly; -static kmem_cache_t *ip_conntrack_expect_cachep __read_mostly; +static struct kmem_cache *ip_conntrack_cachep __read_mostly; +static struct kmem_cache *ip_conntrack_expect_cachep __read_mostly; struct ip_conntrack ip_conntrack_untracked; unsigned int ip_ct_log_invalid __read_mostly; static LIST_HEAD(unconfirmed); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 97a8cfbb61a..96d8310ae9c 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -50,7 +50,7 @@ struct rt6_statistics rt6_stats; -static kmem_cache_t * fib6_node_kmem __read_mostly; +static struct kmem_cache * fib6_node_kmem __read_mostly; enum fib_walk_state_t { diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index d4f68b0f27d..12e426b9aac 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -50,7 +50,7 @@ static u32 xfrm6_tunnel_spi; #define XFRM6_TUNNEL_SPI_MIN 1 #define XFRM6_TUNNEL_SPI_MAX 0xffffffff -static kmem_cache_t *xfrm6_tunnel_spi_kmem __read_mostly; +static struct kmem_cache *xfrm6_tunnel_spi_kmem __read_mostly; #define XFRM6_TUNNEL_SPI_BYADDR_HSIZE 256 #define XFRM6_TUNNEL_SPI_BYSPI_HSIZE 256 diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index eaa0f8a1adb..a9638ff52a7 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -108,7 +108,7 @@ static struct { size_t size; /* slab cache pointer */ - kmem_cache_t *cachep; + struct kmem_cache *cachep; /* allocated slab cache + modules which uses this slab cache */ int use; @@ -147,7 +147,7 @@ int nf_conntrack_register_cache(u_int32_t features, const char *name, { int ret = 0; char *cache_name; - kmem_cache_t *cachep; + struct kmem_cache *cachep; DEBUGP("nf_conntrack_register_cache: features=0x%x, name=%s, size=%d\n", features, name, size); @@ -226,7 +226,7 @@ EXPORT_SYMBOL_GPL(nf_conntrack_register_cache); /* FIXME: In the current, only nf_conntrack_cleanup() can call this function. */ void nf_conntrack_unregister_cache(u_int32_t features) { - kmem_cache_t *cachep; + struct kmem_cache *cachep; char *name; /* diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 588d3793704..c20f901fa17 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -29,7 +29,7 @@ LIST_HEAD(nf_conntrack_expect_list); EXPORT_SYMBOL_GPL(nf_conntrack_expect_list); -kmem_cache_t *nf_conntrack_expect_cachep __read_mostly; +struct kmem_cache *nf_conntrack_expect_cachep __read_mostly; static unsigned int nf_conntrack_expect_next_id; /* nf_conntrack_expect helper functions */ diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c index a98de0b54d6..a5a6e192ac2 100644 --- a/net/netfilter/xt_hashlimit.c +++ b/net/netfilter/xt_hashlimit.c @@ -92,7 +92,7 @@ struct xt_hashlimit_htable { static DEFINE_SPINLOCK(hashlimit_lock); /* protects htables list */ static DEFINE_MUTEX(hlimit_mutex); /* additional checkentry protection */ static HLIST_HEAD(hashlimit_htables); -static kmem_cache_t *hashlimit_cachep __read_mostly; +static struct kmem_cache *hashlimit_cachep __read_mostly; static inline int dst_cmp(const struct dsthash_ent *ent, struct dsthash_dst *b) { diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 11f3b549f4a..f2ba8615895 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -79,8 +79,8 @@ static struct sctp_pf *sctp_pf_inet_specific; static struct sctp_af *sctp_af_v4_specific; static struct sctp_af *sctp_af_v6_specific; -kmem_cache_t *sctp_chunk_cachep __read_mostly; -kmem_cache_t *sctp_bucket_cachep __read_mostly; +struct kmem_cache *sctp_chunk_cachep __read_mostly; +struct kmem_cache *sctp_bucket_cachep __read_mostly; /* Return the address of the control sock. */ struct sock *sctp_get_ctl_sock(void) diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 8d55d10041f..30927d3a597 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -65,7 +65,7 @@ #include #include -extern kmem_cache_t *sctp_chunk_cachep; +extern struct kmem_cache *sctp_chunk_cachep; SCTP_STATIC struct sctp_chunk *sctp_make_chunk(const struct sctp_association *asoc, diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 49607792cbd..1e8132b8c4d 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -107,7 +107,7 @@ static void sctp_sock_migrate(struct sock *, struct sock *, struct sctp_association *, sctp_socket_type_t); static char *sctp_hmac_alg = SCTP_COOKIE_HMAC_ALG; -extern kmem_cache_t *sctp_bucket_cachep; +extern struct kmem_cache *sctp_bucket_cachep; /* Get the sndbuf space available at the time on the association. */ static inline int sctp_wspace(struct sctp_association *asoc) diff --git a/net/socket.c b/net/socket.c index 4f417c2ddc1..43eff489c87 100644 --- a/net/socket.c +++ b/net/socket.c @@ -230,7 +230,7 @@ int move_addr_to_user(void *kaddr, int klen, void __user *uaddr, #define SOCKFS_MAGIC 0x534F434B -static kmem_cache_t *sock_inode_cachep __read_mostly; +static struct kmem_cache *sock_inode_cachep __read_mostly; static struct inode *sock_alloc_inode(struct super_block *sb) { @@ -257,7 +257,7 @@ static void sock_destroy_inode(struct inode *inode) container_of(inode, struct socket_alloc, vfs_inode)); } -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct socket_alloc *ei = (struct socket_alloc *)foo; diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index df753d0a884..19703aa9659 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -33,7 +33,7 @@ static int rpc_mount_count; static struct file_system_type rpc_pipe_fs_type; -static kmem_cache_t *rpc_inode_cachep __read_mostly; +static struct kmem_cache *rpc_inode_cachep __read_mostly; #define RPC_UPCALL_TIMEOUT (30*HZ) @@ -824,7 +824,7 @@ static struct file_system_type rpc_pipe_fs_type = { }; static void -init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct rpc_inode *rpci = (struct rpc_inode *) foo; diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index eff44bcdc95..225e6510b52 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -34,8 +34,8 @@ static int rpc_task_id; #define RPC_BUFFER_MAXSIZE (2048) #define RPC_BUFFER_POOLSIZE (8) #define RPC_TASK_POOLSIZE (8) -static kmem_cache_t *rpc_task_slabp __read_mostly; -static kmem_cache_t *rpc_buffer_slabp __read_mostly; +static struct kmem_cache *rpc_task_slabp __read_mostly; +static struct kmem_cache *rpc_buffer_slabp __read_mostly; static mempool_t *rpc_task_mempool __read_mostly; static mempool_t *rpc_buffer_mempool __read_mostly; diff --git a/net/tipc/handler.c b/net/tipc/handler.c index ae6ddf00a1a..eb80778d6d9 100644 --- a/net/tipc/handler.c +++ b/net/tipc/handler.c @@ -42,7 +42,7 @@ struct queue_item { unsigned long data; }; -static kmem_cache_t *tipc_queue_item_cache; +static struct kmem_cache *tipc_queue_item_cache; static struct list_head signal_queue_head; static DEFINE_SPINLOCK(qitem_lock); static int handler_enabled = 0; diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index a898a6a83a5..414f8907038 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -12,7 +12,7 @@ #include #include -static kmem_cache_t *secpath_cachep __read_mostly; +static struct kmem_cache *secpath_cachep __read_mostly; void __secpath_destroy(struct sec_path *sp) { diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index f6c77bd36fd..3f3f563eb4a 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -39,7 +39,7 @@ EXPORT_SYMBOL(xfrm_policy_count); static DEFINE_RWLOCK(xfrm_policy_afinfo_lock); static struct xfrm_policy_afinfo *xfrm_policy_afinfo[NPROTO]; -static kmem_cache_t *xfrm_dst_cache __read_mostly; +static struct kmem_cache *xfrm_dst_cache __read_mostly; static struct work_struct xfrm_policy_gc_work; static HLIST_HEAD(xfrm_policy_gc_list); diff --git a/security/keys/key.c b/security/keys/key.c index 157bac658bf..0db816f10f8 100644 --- a/security/keys/key.c +++ b/security/keys/key.c @@ -20,7 +20,7 @@ #include #include "internal.h" -static kmem_cache_t *key_jar; +static struct kmem_cache *key_jar; struct rb_root key_serial_tree; /* tree of keys indexed by serial */ DEFINE_SPINLOCK(key_serial_lock); diff --git a/security/selinux/avc.c b/security/selinux/avc.c index 65b4ec9c699..e7c0b5e2066 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -124,7 +124,7 @@ DEFINE_PER_CPU(struct avc_cache_stats, avc_cache_stats) = { 0 }; static struct avc_cache avc_cache; static struct avc_callback_node *avc_callbacks; -static kmem_cache_t *avc_node_cachep; +static struct kmem_cache *avc_node_cachep; static inline int avc_hash(u32 ssid, u32 tsid, u16 tclass) { diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index ac1aeed0b28..44e9cd47054 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -124,7 +124,7 @@ static struct security_operations *secondary_ops = NULL; static LIST_HEAD(superblock_security_head); static DEFINE_SPINLOCK(sb_security_lock); -static kmem_cache_t *sel_inode_cache; +static struct kmem_cache *sel_inode_cache; /* Return security context for a given sid or just the context length if the buffer is null or length is 0 */ diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c index 2dfc6134c2c..ebb993c5c24 100644 --- a/security/selinux/ss/avtab.c +++ b/security/selinux/ss/avtab.c @@ -28,7 +28,7 @@ (keyp->source_type << 9)) & \ AVTAB_HASH_MASK) -static kmem_cache_t *avtab_node_cachep; +static struct kmem_cache *avtab_node_cachep; static struct avtab_node* avtab_insert_node(struct avtab *h, int hvalue, -- cgit v1.2.3-70-g09d2 From 7dfb71030f7636a0d65200158113c37764552f93 Mon Sep 17 00:00:00 2001 From: Nigel Cunningham Date: Wed, 6 Dec 2006 20:34:23 -0800 Subject: [PATCH] Add include/linux/freezer.h and move definitions from sched.h Move process freezing functions from include/linux/sched.h to freezer.h, so that modifications to the freezer or the kernel configuration don't require recompiling just about everything. [akpm@osdl.org: fix ueagle driver] Signed-off-by: Nigel Cunningham Cc: "Rafael J. Wysocki" Cc: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/signal.c | 1 + arch/avr32/kernel/signal.c | 2 +- arch/frv/kernel/signal.c | 2 +- arch/h8300/kernel/signal.c | 2 +- arch/i386/kernel/io_apic.c | 1 + arch/m32r/kernel/signal.c | 2 +- arch/powerpc/kernel/signal_32.c | 2 +- arch/sh/kernel/signal.c | 1 + arch/sh64/kernel/signal.c | 2 +- drivers/block/pktcdvd.c | 2 +- drivers/char/hvc_console.c | 1 + drivers/edac/edac_mc.c | 1 + drivers/ieee1394/nodemgr.c | 1 + drivers/input/gameport/gameport.c | 1 + drivers/input/serio/serio.c | 1 + drivers/macintosh/therm_adt746x.c | 1 + drivers/macintosh/via-pmu.c | 2 +- drivers/macintosh/windfarm_core.c | 1 + drivers/md/md.c | 2 +- drivers/media/dvb/dvb-core/dvb_frontend.c | 2 +- drivers/media/video/msp3400-driver.c | 2 +- drivers/media/video/tvaudio.c | 1 + drivers/media/video/video-buf-dvb.c | 2 +- drivers/media/video/vivi.c | 1 + drivers/mfd/ucb1x00-ts.c | 2 +- drivers/net/irda/stir4200.c | 1 + drivers/net/wireless/airo.c | 1 + drivers/pcmcia/cs.c | 1 + drivers/pnp/pnpbios/core.c | 1 + drivers/usb/atm/ueagle-atm.c | 2 + drivers/usb/core/hub.c | 1 + drivers/usb/gadget/file_storage.c | 2 +- drivers/usb/storage/usb.c | 2 +- drivers/w1/w1.c | 1 + fs/afs/kafsasyncd.c | 1 + fs/afs/kafstimod.c | 1 + fs/cifs/cifsfs.c | 1 + fs/cifs/connect.c | 1 + fs/jbd/journal.c | 2 +- fs/jbd2/journal.c | 2 +- fs/jffs/intrep.c | 1 + fs/jffs2/background.c | 1 + fs/jfs/jfs_logmgr.c | 2 +- fs/jfs/jfs_txnmgr.c | 2 +- fs/lockd/clntproc.c | 1 + fs/xfs/linux-2.6/xfs_buf.c | 1 + fs/xfs/linux-2.6/xfs_super.c | 1 + include/linux/freezer.h | 84 +++++++++++++++++++++++++++++++ include/linux/sched.h | 81 ----------------------------- init/do_mounts_initrd.c | 1 + kernel/audit.c | 1 + kernel/power/disk.c | 1 + kernel/power/main.c | 1 + kernel/power/process.c | 1 + kernel/power/user.c | 1 + kernel/rtmutex-tester.c | 1 + kernel/sched.c | 2 +- kernel/signal.c | 1 + mm/pdflush.c | 1 + mm/vmscan.c | 1 + net/rxrpc/krxiod.c | 1 + net/rxrpc/krxsecd.c | 1 + net/rxrpc/krxtimod.c | 1 + net/sunrpc/svcsock.c | 1 + 64 files changed, 147 insertions(+), 101 deletions(-) create mode 100644 include/linux/freezer.h (limited to 'arch/i386') diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 48cf7fffddf..f38a60a03b8 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include diff --git a/arch/avr32/kernel/signal.c b/arch/avr32/kernel/signal.c index 33096651c24..0ec14854a20 100644 --- a/arch/avr32/kernel/signal.c +++ b/arch/avr32/kernel/signal.c @@ -15,7 +15,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/frv/kernel/signal.c b/arch/frv/kernel/signal.c index b8a5882b862..85baeae9666 100644 --- a/arch/frv/kernel/signal.c +++ b/arch/frv/kernel/signal.c @@ -21,7 +21,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/h8300/kernel/signal.c b/arch/h8300/kernel/signal.c index 7787f70a05b..02955604d76 100644 --- a/arch/h8300/kernel/signal.c +++ b/arch/h8300/kernel/signal.c @@ -38,7 +38,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 3b7a63e0ed1..44c5a3206b2 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include diff --git a/arch/m32r/kernel/signal.c b/arch/m32r/kernel/signal.c index b60cea4aeba..092ea86bb07 100644 --- a/arch/m32r/kernel/signal.c +++ b/arch/m32r/kernel/signal.c @@ -21,7 +21,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 320353f0926..e4ebe1a6228 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -36,7 +36,7 @@ #include #include #include -#include +#include #endif #include diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c index 50d7c4993be..bb1c480a59c 100644 --- a/arch/sh/kernel/signal.c +++ b/arch/sh/kernel/signal.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include diff --git a/arch/sh64/kernel/signal.c b/arch/sh64/kernel/signal.c index 9e2ffc45c0e..1666d3efb52 100644 --- a/arch/sh64/kernel/signal.c +++ b/arch/sh64/kernel/signal.c @@ -22,7 +22,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index f2904f67af4..e45eaa26411 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -54,7 +54,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c index 9902ffad3b1..cc2cd46bedc 100644 --- a/drivers/char/hvc_console.c +++ b/drivers/char/hvc_console.c @@ -38,6 +38,7 @@ #include #include #include +#include #include diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c index 75e9e38330f..1b4fc922180 100644 --- a/drivers/edac/edac_mc.c +++ b/drivers/edac/edac_mc.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/ieee1394/nodemgr.c b/drivers/ieee1394/nodemgr.c index 8e7b83f8448..e829c9336b3 100644 --- a/drivers/ieee1394/nodemgr.c +++ b/drivers/ieee1394/nodemgr.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include "csr.h" diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c index a0af97efe6a..79dfb4b25c9 100644 --- a/drivers/input/gameport/gameport.c +++ b/drivers/input/gameport/gameport.c @@ -23,6 +23,7 @@ #include #include /* HZ */ #include +#include /*#include */ diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c index 211943f85cb..5f1d4032fd5 100644 --- a/drivers/input/serio/serio.c +++ b/drivers/input/serio/serio.c @@ -35,6 +35,7 @@ #include #include #include +#include MODULE_AUTHOR("Vojtech Pavlik "); MODULE_DESCRIPTION("Serio abstraction core"); diff --git a/drivers/macintosh/therm_adt746x.c b/drivers/macintosh/therm_adt746x.c index 13b953ae8eb..3d3bf1643e7 100644 --- a/drivers/macintosh/therm_adt746x.c +++ b/drivers/macintosh/therm_adt746x.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c index e63ea1c1f3c..c8558d4ed50 100644 --- a/drivers/macintosh/via-pmu.c +++ b/drivers/macintosh/via-pmu.c @@ -42,7 +42,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/macintosh/windfarm_core.c b/drivers/macintosh/windfarm_core.c index ab3faa702d5..e947af982f9 100644 --- a/drivers/macintosh/windfarm_core.c +++ b/drivers/macintosh/windfarm_core.c @@ -34,6 +34,7 @@ #include #include #include +#include #include diff --git a/drivers/md/md.c b/drivers/md/md.c index 8cbf9c9df1c..6c4345bde07 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -39,10 +39,10 @@ #include #include #include /* for invalidate_bdev */ -#include #include #include #include +#include #include diff --git a/drivers/media/dvb/dvb-core/dvb_frontend.c b/drivers/media/dvb/dvb-core/dvb_frontend.c index a2ab2eebfc6..e85972222ab 100644 --- a/drivers/media/dvb/dvb-core/dvb_frontend.c +++ b/drivers/media/dvb/dvb-core/dvb_frontend.c @@ -34,7 +34,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/media/video/msp3400-driver.c b/drivers/media/video/msp3400-driver.c index cf43df3fe70..e1b56dc13c3 100644 --- a/drivers/media/video/msp3400-driver.c +++ b/drivers/media/video/msp3400-driver.c @@ -56,7 +56,7 @@ #include #include #include -#include +#include #include "msp3400-driver.h" /* ---------------------------------------------------------------------- */ diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c index fcaef4bf828..d506dfaa45a 100644 --- a/drivers/media/video/tvaudio.c +++ b/drivers/media/video/tvaudio.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include diff --git a/drivers/media/video/video-buf-dvb.c b/drivers/media/video/video-buf-dvb.c index f53edf1923b..fcc5467e763 100644 --- a/drivers/media/video/video-buf-dvb.c +++ b/drivers/media/video/video-buf-dvb.c @@ -20,7 +20,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/media/video/vivi.c b/drivers/media/video/vivi.c index 3c8dc72dc8e..9986de5cb3d 100644 --- a/drivers/media/video/vivi.c +++ b/drivers/media/video/vivi.c @@ -36,6 +36,7 @@ #include #include #include +#include /* Wake up at about 30 fps */ #define WAKE_NUMERATOR 30 diff --git a/drivers/mfd/ucb1x00-ts.c b/drivers/mfd/ucb1x00-ts.c index 82938ad6ddb..ce1a4810821 100644 --- a/drivers/mfd/ucb1x00-ts.c +++ b/drivers/mfd/ucb1x00-ts.c @@ -28,7 +28,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/net/irda/stir4200.c b/drivers/net/irda/stir4200.c index 3b4c4787593..c14a74634fd 100644 --- a/drivers/net/irda/stir4200.c +++ b/drivers/net/irda/stir4200.c @@ -50,6 +50,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c index efcdaf1c5f7..44a22701da9 100644 --- a/drivers/net/wireless/airo.c +++ b/drivers/net/wireless/airo.c @@ -49,6 +49,7 @@ #include #include #include +#include #include "airo.h" diff --git a/drivers/pcmcia/cs.c b/drivers/pcmcia/cs.c index f9cd831a3f3..606a4674033 100644 --- a/drivers/pcmcia/cs.c +++ b/drivers/pcmcia/cs.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include diff --git a/drivers/pnp/pnpbios/core.c b/drivers/pnp/pnpbios/core.c index 81a6c83d89a..81186f479a3 100644 --- a/drivers/pnp/pnpbios/core.c +++ b/drivers/pnp/pnpbios/core.c @@ -61,6 +61,7 @@ #include #include #include +#include #include #include diff --git a/drivers/usb/atm/ueagle-atm.c b/drivers/usb/atm/ueagle-atm.c index f2d196fa1e8..dae4ef1e8fe 100644 --- a/drivers/usb/atm/ueagle-atm.c +++ b/drivers/usb/atm/ueagle-atm.c @@ -64,6 +64,8 @@ #include #include #include +#include + #include #include "usbatm.h" diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 77c05be5241..2651c2e2a89 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c index 8b975d15538..c98316ce838 100644 --- a/drivers/usb/gadget/file_storage.c +++ b/drivers/usb/gadget/file_storage.c @@ -250,7 +250,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c index b401084b3d2..70644506651 100644 --- a/drivers/usb/storage/usb.c +++ b/drivers/usb/storage/usb.c @@ -49,7 +49,7 @@ #include #include -#include +#include #include #include #include diff --git a/drivers/w1/w1.c b/drivers/w1/w1.c index de3e9791f80..63c07243993 100644 --- a/drivers/w1/w1.c +++ b/drivers/w1/w1.c @@ -31,6 +31,7 @@ #include #include #include +#include #include diff --git a/fs/afs/kafsasyncd.c b/fs/afs/kafsasyncd.c index f09a794f248..615df2407cb 100644 --- a/fs/afs/kafsasyncd.c +++ b/fs/afs/kafsasyncd.c @@ -20,6 +20,7 @@ #include #include #include +#include #include "cell.h" #include "server.h" #include "volume.h" diff --git a/fs/afs/kafstimod.c b/fs/afs/kafstimod.c index 65bc05ab818..694344e4d3c 100644 --- a/fs/afs/kafstimod.c +++ b/fs/afs/kafstimod.c @@ -13,6 +13,7 @@ #include #include #include +#include #include "cell.h" #include "volume.h" #include "kafstimod.h" diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index e6b5866e500..71bc87a37fc 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -34,6 +34,7 @@ #include #include #include +#include #include "cifsfs.h" #include "cifspdu.h" #define DECLARE_GLOBALS_HERE diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 71f77914ce9..2caca06b4ba 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include "cifspdu.h" diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index a8774bed20b..10fff944393 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -31,7 +31,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 50356019ae3..44fc32bfd7f 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -31,7 +31,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/fs/jffs/intrep.c b/fs/jffs/intrep.c index 4a543e11497..478a74e2e9d 100644 --- a/fs/jffs/intrep.c +++ b/fs/jffs/intrep.c @@ -66,6 +66,7 @@ #include #include #include +#include #include "intrep.h" #include "jffs_fm.h" diff --git a/fs/jffs2/background.c b/fs/jffs2/background.c index ff2a872e80e..6eb3daebd56 100644 --- a/fs/jffs2/background.c +++ b/fs/jffs2/background.c @@ -16,6 +16,7 @@ #include #include #include +#include #include "nodelist.h" diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index b89c9aba046..5065baa530b 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -67,7 +67,7 @@ #include #include /* for sync_blockdev() */ #include -#include +#include #include #include #include "jfs_incore.h" diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index 81f6f04af19..d558e51b0df 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -46,7 +46,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 3d84f600b63..50643b6a555 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index eef4a0ba11e..b971237c5a9 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -32,6 +32,7 @@ #include #include #include +#include STATIC kmem_zone_t *xfs_buf_zone; STATIC kmem_shaker_t xfs_buf_shake; diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index de05abbbe7f..b93265b7c79 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -56,6 +56,7 @@ #include #include #include +#include STATIC struct quotactl_ops xfs_quotactl_operations; STATIC struct super_operations xfs_super_operations; diff --git a/include/linux/freezer.h b/include/linux/freezer.h new file mode 100644 index 00000000000..266373f7444 --- /dev/null +++ b/include/linux/freezer.h @@ -0,0 +1,84 @@ +/* Freezer declarations */ + +#ifdef CONFIG_PM +/* + * Check if a process has been frozen + */ +static inline int frozen(struct task_struct *p) +{ + return p->flags & PF_FROZEN; +} + +/* + * Check if there is a request to freeze a process + */ +static inline int freezing(struct task_struct *p) +{ + return p->flags & PF_FREEZE; +} + +/* + * Request that a process be frozen + * FIXME: SMP problem. We may not modify other process' flags! + */ +static inline void freeze(struct task_struct *p) +{ + p->flags |= PF_FREEZE; +} + +/* + * Sometimes we may need to cancel the previous 'freeze' request + */ +static inline void do_not_freeze(struct task_struct *p) +{ + p->flags &= ~PF_FREEZE; +} + +/* + * Wake up a frozen process + */ +static inline int thaw_process(struct task_struct *p) +{ + if (frozen(p)) { + p->flags &= ~PF_FROZEN; + wake_up_process(p); + return 1; + } + return 0; +} + +/* + * freezing is complete, mark process as frozen + */ +static inline void frozen_process(struct task_struct *p) +{ + p->flags = (p->flags & ~PF_FREEZE) | PF_FROZEN; +} + +extern void refrigerator(void); +extern int freeze_processes(void); +extern void thaw_processes(void); + +static inline int try_to_freeze(void) +{ + if (freezing(current)) { + refrigerator(); + return 1; + } else + return 0; +} +#else +static inline int frozen(struct task_struct *p) { return 0; } +static inline int freezing(struct task_struct *p) { return 0; } +static inline void freeze(struct task_struct *p) { BUG(); } +static inline int thaw_process(struct task_struct *p) { return 1; } +static inline void frozen_process(struct task_struct *p) { BUG(); } + +static inline void refrigerator(void) {} +static inline int freeze_processes(void) { BUG(); return 0; } +static inline void thaw_processes(void) {} + +static inline int try_to_freeze(void) { return 0; } + + +#endif diff --git a/include/linux/sched.h b/include/linux/sched.h index acfd2e15c5f..837a012f573 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1618,87 +1618,6 @@ extern int sched_create_sysfs_power_savings_entries(struct sysdev_class *cls); extern void normalize_rt_tasks(void); -#ifdef CONFIG_PM -/* - * Check if a process has been frozen - */ -static inline int frozen(struct task_struct *p) -{ - return p->flags & PF_FROZEN; -} - -/* - * Check if there is a request to freeze a process - */ -static inline int freezing(struct task_struct *p) -{ - return p->flags & PF_FREEZE; -} - -/* - * Request that a process be frozen - * FIXME: SMP problem. We may not modify other process' flags! - */ -static inline void freeze(struct task_struct *p) -{ - p->flags |= PF_FREEZE; -} - -/* - * Sometimes we may need to cancel the previous 'freeze' request - */ -static inline void do_not_freeze(struct task_struct *p) -{ - p->flags &= ~PF_FREEZE; -} - -/* - * Wake up a frozen process - */ -static inline int thaw_process(struct task_struct *p) -{ - if (frozen(p)) { - p->flags &= ~PF_FROZEN; - wake_up_process(p); - return 1; - } - return 0; -} - -/* - * freezing is complete, mark process as frozen - */ -static inline void frozen_process(struct task_struct *p) -{ - p->flags = (p->flags & ~PF_FREEZE) | PF_FROZEN; -} - -extern void refrigerator(void); -extern int freeze_processes(void); -extern void thaw_processes(void); - -static inline int try_to_freeze(void) -{ - if (freezing(current)) { - refrigerator(); - return 1; - } else - return 0; -} -#else -static inline int frozen(struct task_struct *p) { return 0; } -static inline int freezing(struct task_struct *p) { return 0; } -static inline void freeze(struct task_struct *p) { BUG(); } -static inline int thaw_process(struct task_struct *p) { return 1; } -static inline void frozen_process(struct task_struct *p) { BUG(); } - -static inline void refrigerator(void) {} -static inline int freeze_processes(void) { BUG(); return 0; } -static inline void thaw_processes(void) {} - -static inline int try_to_freeze(void) { return 0; } - -#endif /* CONFIG_PM */ #endif /* __KERNEL__ */ #endif diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c index 919a80cb322..2cfd7cb36e7 100644 --- a/init/do_mounts_initrd.c +++ b/init/do_mounts_initrd.c @@ -6,6 +6,7 @@ #include #include #include +#include #include "do_mounts.h" diff --git a/kernel/audit.c b/kernel/audit.c index 98106f6078b..d9b690ac684 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -57,6 +57,7 @@ #include #include #include +#include #include "audit.h" diff --git a/kernel/power/disk.c b/kernel/power/disk.c index f5079231383..53b3b57c022 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -20,6 +20,7 @@ #include #include #include +#include #include "power.h" diff --git a/kernel/power/main.c b/kernel/power/main.c index 873228c71da..6096c71b182 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -18,6 +18,7 @@ #include #include #include +#include #include "power.h" diff --git a/kernel/power/process.c b/kernel/power/process.c index 72e72d2c61e..29be608e834 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -13,6 +13,7 @@ #include #include #include +#include /* * Timeout for stopping processes diff --git a/kernel/power/user.c b/kernel/power/user.c index a63b25c63b4..26c66941c00 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -22,6 +22,7 @@ #include #include #include +#include #include diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c index 6dcea9dd8c9..015fc633c96 100644 --- a/kernel/rtmutex-tester.c +++ b/kernel/rtmutex-tester.c @@ -13,6 +13,7 @@ #include #include #include +#include #include "rtmutex.h" diff --git a/kernel/sched.c b/kernel/sched.c index 3399701c680..12fdbef1d9b 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -34,7 +34,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/signal.c b/kernel/signal.c index 8e19d278548..bc972d7e631 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include diff --git a/mm/pdflush.c b/mm/pdflush.c index b02102feeb4..8ce0900dc95 100644 --- a/mm/pdflush.c +++ b/mm/pdflush.c @@ -21,6 +21,7 @@ #include // Prototypes pdflush_operation() #include #include +#include /* diff --git a/mm/vmscan.c b/mm/vmscan.c index 2a6a79f6813..f6616e81fac 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include diff --git a/net/rxrpc/krxiod.c b/net/rxrpc/krxiod.c index dada34a77b2..49effd92144 100644 --- a/net/rxrpc/krxiod.c +++ b/net/rxrpc/krxiod.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/net/rxrpc/krxsecd.c b/net/rxrpc/krxsecd.c index cea4eb5e249..3ab0f77409f 100644 --- a/net/rxrpc/krxsecd.c +++ b/net/rxrpc/krxsecd.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include "internal.h" diff --git a/net/rxrpc/krxtimod.c b/net/rxrpc/krxtimod.c index 3e7466900bd..9a9b6132dba 100644 --- a/net/rxrpc/krxtimod.c +++ b/net/rxrpc/krxtimod.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 64ca1f61dd9..1c68956824e 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3-70-g09d2 From 2d4a34c9365c6e3f94a5b26ce296e1fce9b66c8b Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 6 Dec 2006 20:34:29 -0800 Subject: [PATCH] swsusp: Support i386 systems with PAE or without PSE Make swsusp support i386 systems with PAE or without PSE. This is done by creating temporary page tables located in resume-safe page frames before the suspend image is restored in the same way as x86_64 does it. Signed-off-by: Rafael J. Wysocki Cc: Andi Kleen Cc: Dave Jones Cc: Nigel Cunningham Cc: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/power/Makefile | 2 +- arch/i386/power/suspend.c | 158 +++++++++++++++++++++++++++++++++++++++++++++ arch/i386/power/swsusp.S | 9 ++- include/asm-i386/suspend.h | 13 +--- kernel/power/Kconfig | 2 +- 5 files changed, 168 insertions(+), 16 deletions(-) create mode 100644 arch/i386/power/suspend.c (limited to 'arch/i386') diff --git a/arch/i386/power/Makefile b/arch/i386/power/Makefile index 8cfa4e8a719..2de7bbf03cd 100644 --- a/arch/i386/power/Makefile +++ b/arch/i386/power/Makefile @@ -1,2 +1,2 @@ obj-$(CONFIG_PM) += cpu.o -obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o +obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o suspend.o diff --git a/arch/i386/power/suspend.c b/arch/i386/power/suspend.c new file mode 100644 index 00000000000..db5e98d2eb7 --- /dev/null +++ b/arch/i386/power/suspend.c @@ -0,0 +1,158 @@ +/* + * Suspend support specific for i386 - temporary page tables + * + * Distribute under GPLv2 + * + * Copyright (c) 2006 Rafael J. Wysocki + */ + +#include +#include + +#include +#include +#include + +/* Defined in arch/i386/power/swsusp.S */ +extern int restore_image(void); + +/* Pointer to the temporary resume page tables */ +pgd_t *resume_pg_dir; + +/* The following three functions are based on the analogous code in + * arch/i386/mm/init.c + */ + +/* + * Create a middle page table on a resume-safe page and put a pointer to it in + * the given global directory entry. This only returns the gd entry + * in non-PAE compilation mode, since the middle layer is folded. + */ +static pmd_t *resume_one_md_table_init(pgd_t *pgd) +{ + pud_t *pud; + pmd_t *pmd_table; + +#ifdef CONFIG_X86_PAE + pmd_table = (pmd_t *)get_safe_page(GFP_ATOMIC); + if (!pmd_table) + return NULL; + + set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT)); + pud = pud_offset(pgd, 0); + + BUG_ON(pmd_table != pmd_offset(pud, 0)); +#else + pud = pud_offset(pgd, 0); + pmd_table = pmd_offset(pud, 0); +#endif + + return pmd_table; +} + +/* + * Create a page table on a resume-safe page and place a pointer to it in + * a middle page directory entry. + */ +static pte_t *resume_one_page_table_init(pmd_t *pmd) +{ + if (pmd_none(*pmd)) { + pte_t *page_table = (pte_t *)get_safe_page(GFP_ATOMIC); + if (!page_table) + return NULL; + + set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE)); + + BUG_ON(page_table != pte_offset_kernel(pmd, 0)); + + return page_table; + } + + return pte_offset_kernel(pmd, 0); +} + +/* + * This maps the physical memory to kernel virtual address space, a total + * of max_low_pfn pages, by creating page tables starting from address + * PAGE_OFFSET. The page tables are allocated out of resume-safe pages. + */ +static int resume_physical_mapping_init(pgd_t *pgd_base) +{ + unsigned long pfn; + pgd_t *pgd; + pmd_t *pmd; + pte_t *pte; + int pgd_idx, pmd_idx; + + pgd_idx = pgd_index(PAGE_OFFSET); + pgd = pgd_base + pgd_idx; + pfn = 0; + + for (; pgd_idx < PTRS_PER_PGD; pgd++, pgd_idx++) { + pmd = resume_one_md_table_init(pgd); + if (!pmd) + return -ENOMEM; + + if (pfn >= max_low_pfn) + continue; + + for (pmd_idx = 0; pmd_idx < PTRS_PER_PMD; pmd++, pmd_idx++) { + if (pfn >= max_low_pfn) + break; + + /* Map with big pages if possible, otherwise create + * normal page tables. + * NOTE: We can mark everything as executable here + */ + if (cpu_has_pse) { + set_pmd(pmd, pfn_pmd(pfn, PAGE_KERNEL_LARGE_EXEC)); + pfn += PTRS_PER_PTE; + } else { + pte_t *max_pte; + + pte = resume_one_page_table_init(pmd); + if (!pte) + return -ENOMEM; + + max_pte = pte + PTRS_PER_PTE; + for (; pte < max_pte; pte++, pfn++) { + if (pfn >= max_low_pfn) + break; + + set_pte(pte, pfn_pte(pfn, PAGE_KERNEL_EXEC)); + } + } + } + } + return 0; +} + +static inline void resume_init_first_level_page_table(pgd_t *pg_dir) +{ +#ifdef CONFIG_X86_PAE + int i; + + /* Init entries of the first-level page table to the zero page */ + for (i = 0; i < PTRS_PER_PGD; i++) + set_pgd(pg_dir + i, + __pgd(__pa(empty_zero_page) | _PAGE_PRESENT)); +#endif +} + +int swsusp_arch_resume(void) +{ + int error; + + resume_pg_dir = (pgd_t *)get_safe_page(GFP_ATOMIC); + if (!resume_pg_dir) + return -ENOMEM; + + resume_init_first_level_page_table(resume_pg_dir); + error = resume_physical_mapping_init(resume_pg_dir); + if (error) + return error; + + /* We have got enough memory and from now on we cannot recover */ + restore_image(); + return 0; +} diff --git a/arch/i386/power/swsusp.S b/arch/i386/power/swsusp.S index 8a2b50a0aaa..53662e05b39 100644 --- a/arch/i386/power/swsusp.S +++ b/arch/i386/power/swsusp.S @@ -28,8 +28,9 @@ ENTRY(swsusp_arch_suspend) call swsusp_save ret -ENTRY(swsusp_arch_resume) - movl $swsusp_pg_dir-__PAGE_OFFSET, %ecx +ENTRY(restore_image) + movl resume_pg_dir, %ecx + subl $__PAGE_OFFSET, %ecx movl %ecx, %cr3 movl restore_pblist, %edx @@ -51,6 +52,10 @@ copy_loop: .p2align 4,,7 done: + /* go back to the original page tables */ + movl $swapper_pg_dir, %ecx + subl $__PAGE_OFFSET, %ecx + movl %ecx, %cr3 /* Flush TLB, including "global" things (vmalloc) */ movl mmu_cr4_features, %eax movl %eax, %edx diff --git a/include/asm-i386/suspend.h b/include/asm-i386/suspend.h index 08be1e5009d..c1da5caafaf 100644 --- a/include/asm-i386/suspend.h +++ b/include/asm-i386/suspend.h @@ -6,18 +6,7 @@ #include #include -static inline int -arch_prepare_suspend(void) -{ - /* If you want to make non-PSE machine work, turn off paging - in swsusp_arch_suspend. swsusp_pg_dir should have identity mapping, so - it could work... */ - if (!cpu_has_pse) { - printk(KERN_ERR "PSE is required for swsusp.\n"); - return -EPERM; - } - return 0; -} +static inline int arch_prepare_suspend(void) { return 0; } /* image of the saved processor state */ struct saved_context { diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 825068ca347..710ed084e7c 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -78,7 +78,7 @@ config PM_SYSFS_DEPRECATED config SOFTWARE_SUSPEND bool "Software Suspend" - depends on PM && SWAP && ((X86 && (!SMP || SUSPEND_SMP) && !X86_PAE) || ((FRV || PPC32) && !SMP)) + depends on PM && SWAP && ((X86 && (!SMP || SUSPEND_SMP)) || ((FRV || PPC32) && !SMP)) ---help--- Enable the possibility of suspending the machine. It doesn't need ACPI or APM. -- cgit v1.2.3-70-g09d2 From 6cfd76a26d9fe2ba54b9d496a48c1d9285e5c5ed Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 6 Dec 2006 20:37:22 -0800 Subject: [PATCH] lockdep: name some old style locks Name some of the remaning 'old_style_spin_init' locks Signed-off-by: Peter Zijlstra Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/traps.c | 2 +- include/asm-i386/rwsem.h | 4 ++-- include/linux/init_task.h | 2 +- include/linux/mutex.h | 2 +- include/linux/rtmutex.h | 2 +- include/linux/rwsem-spinlock.h | 3 ++- include/linux/sunrpc/sched.h | 4 ++-- kernel/acct.c | 3 ++- kernel/irq/handle.c | 2 +- net/sunrpc/svcauth.c | 3 ++- security/keys/process_keys.c | 2 +- 11 files changed, 16 insertions(+), 13 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index fe9c5e8e7e6..3124f1b04d6 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -452,7 +452,7 @@ void die(const char * str, struct pt_regs * regs, long err) u32 lock_owner; int lock_owner_depth; } die = { - .lock = SPIN_LOCK_UNLOCKED, + .lock = __SPIN_LOCK_UNLOCKED(die.lock), .lock_owner = -1, .lock_owner_depth = 0 }; diff --git a/include/asm-i386/rwsem.h b/include/asm-i386/rwsem.h index bc598d6388e..041906f3c6d 100644 --- a/include/asm-i386/rwsem.h +++ b/include/asm-i386/rwsem.h @@ -75,8 +75,8 @@ struct rw_semaphore { #define __RWSEM_INITIALIZER(name) \ -{ RWSEM_UNLOCKED_VALUE, SPIN_LOCK_UNLOCKED, LIST_HEAD_INIT((name).wait_list) \ - __RWSEM_DEP_MAP_INIT(name) } +{ RWSEM_UNLOCKED_VALUE, __SPIN_LOCK_UNLOCKED((name).wait_lock), \ + LIST_HEAD_INIT((name).wait_list) __RWSEM_DEP_MAP_INIT(name) } #define DECLARE_RWSEM(name) \ struct rw_semaphore name = __RWSEM_INITIALIZER(name) diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 33c5daacc74..733790d4f7d 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -73,7 +73,7 @@ extern struct nsproxy init_nsproxy; #define INIT_NSPROXY(nsproxy) { \ .count = ATOMIC_INIT(1), \ - .nslock = SPIN_LOCK_UNLOCKED, \ + .nslock = __SPIN_LOCK_UNLOCKED(nsproxy.nslock), \ .uts_ns = &init_uts_ns, \ .namespace = NULL, \ INIT_IPC_NS(ipc_ns) \ diff --git a/include/linux/mutex.h b/include/linux/mutex.h index 27c48daa318..b2b91c47756 100644 --- a/include/linux/mutex.h +++ b/include/linux/mutex.h @@ -94,7 +94,7 @@ do { \ #define __MUTEX_INITIALIZER(lockname) \ { .count = ATOMIC_INIT(1) \ - , .wait_lock = SPIN_LOCK_UNLOCKED \ + , .wait_lock = __SPIN_LOCK_UNLOCKED(lockname.wait_lock) \ , .wait_list = LIST_HEAD_INIT(lockname.wait_list) \ __DEBUG_MUTEX_INITIALIZER(lockname) \ __DEP_MAP_MUTEX_INITIALIZER(lockname) } diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h index 5d41dee82f8..b0090e9f788 100644 --- a/include/linux/rtmutex.h +++ b/include/linux/rtmutex.h @@ -63,7 +63,7 @@ struct hrtimer_sleeper; #endif #define __RT_MUTEX_INITIALIZER(mutexname) \ - { .wait_lock = SPIN_LOCK_UNLOCKED \ + { .wait_lock = __SPIN_LOCK_UNLOCKED(mutexname.wait_lock) \ , .wait_list = PLIST_HEAD_INIT(mutexname.wait_list, mutexname.wait_lock) \ , .owner = NULL \ __DEBUG_RT_MUTEX_INITIALIZER(mutexname)} diff --git a/include/linux/rwsem-spinlock.h b/include/linux/rwsem-spinlock.h index ae1fcadd598..813cee13da0 100644 --- a/include/linux/rwsem-spinlock.h +++ b/include/linux/rwsem-spinlock.h @@ -44,7 +44,8 @@ struct rw_semaphore { #endif #define __RWSEM_INITIALIZER(name) \ -{ 0, SPIN_LOCK_UNLOCKED, LIST_HEAD_INIT((name).wait_list) __RWSEM_DEP_MAP_INIT(name) } +{ 0, __SPIN_LOCK_UNLOCKED(name.wait_lock), LIST_HEAD_INIT((name).wait_list) \ + __RWSEM_DEP_MAP_INIT(name) } #define DECLARE_RWSEM(name) \ struct rw_semaphore name = __RWSEM_INITIALIZER(name) diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index f399c138f79..0746c3b16f3 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -222,7 +222,7 @@ struct rpc_wait_queue { #ifndef RPC_DEBUG # define RPC_WAITQ_INIT(var,qname) { \ - .lock = SPIN_LOCK_UNLOCKED, \ + .lock = __SPIN_LOCK_UNLOCKED(var.lock), \ .tasks = { \ [0] = LIST_HEAD_INIT(var.tasks[0]), \ [1] = LIST_HEAD_INIT(var.tasks[1]), \ @@ -231,7 +231,7 @@ struct rpc_wait_queue { } #else # define RPC_WAITQ_INIT(var,qname) { \ - .lock = SPIN_LOCK_UNLOCKED, \ + .lock = __SPIN_LOCK_UNLOCKED(var.lock), \ .tasks = { \ [0] = LIST_HEAD_INIT(var.tasks[0]), \ [1] = LIST_HEAD_INIT(var.tasks[1]), \ diff --git a/kernel/acct.c b/kernel/acct.c index 0aad5ca36a8..dc12db8600e 100644 --- a/kernel/acct.c +++ b/kernel/acct.c @@ -89,7 +89,8 @@ struct acct_glbs { struct timer_list timer; }; -static struct acct_glbs acct_globals __cacheline_aligned = {SPIN_LOCK_UNLOCKED}; +static struct acct_glbs acct_globals __cacheline_aligned = + {__SPIN_LOCK_UNLOCKED(acct_globals.lock)}; /* * Called whenever the timer says to check the free space. diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index a681912bc89..aff1f0fabb0 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -54,7 +54,7 @@ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned = { .chip = &no_irq_chip, .handle_irq = handle_bad_irq, .depth = 1, - .lock = SPIN_LOCK_UNLOCKED, + .lock = __SPIN_LOCK_UNLOCKED(irq_desc->lock), #ifdef CONFIG_SMP .affinity = CPU_MASK_ALL #endif diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c index ee9bb1522d5..c7bb5f7f21a 100644 --- a/net/sunrpc/svcauth.c +++ b/net/sunrpc/svcauth.c @@ -119,7 +119,8 @@ EXPORT_SYMBOL(svc_auth_unregister); #define DN_HASHMASK (DN_HASHMAX-1) static struct hlist_head auth_domain_table[DN_HASHMAX]; -static spinlock_t auth_domain_lock = SPIN_LOCK_UNLOCKED; +static spinlock_t auth_domain_lock = + __SPIN_LOCK_UNLOCKED(auth_domain_lock); void auth_domain_put(struct auth_domain *dom) { diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c index 32150cf7c37..b6f86808475 100644 --- a/security/keys/process_keys.c +++ b/security/keys/process_keys.c @@ -27,7 +27,7 @@ static DEFINE_MUTEX(key_session_mutex); struct key_user root_key_user = { .usage = ATOMIC_INIT(3), .consq = LIST_HEAD_INIT(root_key_user.consq), - .lock = SPIN_LOCK_UNLOCKED, + .lock = __SPIN_LOCK_UNLOCKED(root_key_user.lock), .nkeys = ATOMIC_INIT(2), .nikeys = ATOMIC_INIT(2), .uid = 0, -- cgit v1.2.3-70-g09d2 From 19e5d9c0d2194b4b47189cbec2921cbf72b0bd1c Mon Sep 17 00:00:00 2001 From: Henry Nestler Date: Wed, 6 Dec 2006 20:37:45 -0800 Subject: [PATCH] initrd: remove unused false condition for initrd_start After LOADER_TYPE && INITRD_START are true, the short if-condition for INITRD_START can never be false. Remove unused code from the else condition. Signed-off-by: Henry Nestler Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/kernel/setup.c | 2 +- arch/i386/kernel/setup.c | 3 +-- arch/m32r/kernel/setup.c | 4 +--- arch/m32r/mm/discontig.c | 4 +--- arch/sh/kernel/setup.c | 3 +-- arch/sh64/kernel/setup.c | 4 +--- arch/x86_64/kernel/setup.c | 3 +-- 7 files changed, 7 insertions(+), 16 deletions(-) (limited to 'arch/i386') diff --git a/arch/frv/kernel/setup.c b/arch/frv/kernel/setup.c index a8c61dac1ce..1a5eb6c301c 100644 --- a/arch/frv/kernel/setup.c +++ b/arch/frv/kernel/setup.c @@ -947,7 +947,7 @@ static void __init setup_linux_memory(void) if (LOADER_TYPE && INITRD_START) { if (INITRD_START + INITRD_SIZE <= (low_top_pfn << PAGE_SHIFT)) { reserve_bootmem(INITRD_START, INITRD_SIZE); - initrd_start = INITRD_START ? INITRD_START + PAGE_OFFSET : 0; + initrd_start = INITRD_START + PAGE_OFFSET; initrd_end = initrd_start + INITRD_SIZE; } else { diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 141041dde74..97bb869307b 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -1162,8 +1162,7 @@ void __init setup_bootmem_allocator(void) if (LOADER_TYPE && INITRD_START) { if (INITRD_START + INITRD_SIZE <= (max_low_pfn << PAGE_SHIFT)) { reserve_bootmem(INITRD_START, INITRD_SIZE); - initrd_start = - INITRD_START ? INITRD_START + PAGE_OFFSET : 0; + initrd_start = INITRD_START + PAGE_OFFSET; initrd_end = initrd_start+INITRD_SIZE; } else { diff --git a/arch/m32r/kernel/setup.c b/arch/m32r/kernel/setup.c index 0e7778be33c..936205f7aba 100644 --- a/arch/m32r/kernel/setup.c +++ b/arch/m32r/kernel/setup.c @@ -196,9 +196,7 @@ static unsigned long __init setup_memory(void) if (LOADER_TYPE && INITRD_START) { if (INITRD_START + INITRD_SIZE <= (max_low_pfn << PAGE_SHIFT)) { reserve_bootmem(INITRD_START, INITRD_SIZE); - initrd_start = INITRD_START ? - INITRD_START + PAGE_OFFSET : 0; - + initrd_start = INITRD_START + PAGE_OFFSET; initrd_end = initrd_start + INITRD_SIZE; printk("initrd:start[%08lx],size[%08lx]\n", initrd_start, INITRD_SIZE); diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c index abb34ccd598..c7efdb0aefc 100644 --- a/arch/m32r/mm/discontig.c +++ b/arch/m32r/mm/discontig.c @@ -105,9 +105,7 @@ unsigned long __init setup_memory(void) if (INITRD_START + INITRD_SIZE <= PFN_PHYS(max_low_pfn)) { reserve_bootmem_node(NODE_DATA(0), INITRD_START, INITRD_SIZE); - initrd_start = INITRD_START ? - INITRD_START + PAGE_OFFSET : 0; - + initrd_start = INITRD_START + PAGE_OFFSET; initrd_end = initrd_start + INITRD_SIZE; printk("initrd:start[%08lx],size[%08lx]\n", initrd_start, INITRD_SIZE); diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c index 696ca75752d..f8dd6b7bfab 100644 --- a/arch/sh/kernel/setup.c +++ b/arch/sh/kernel/setup.c @@ -332,8 +332,7 @@ void __init setup_arch(char **cmdline_p) if (LOADER_TYPE && INITRD_START) { if (INITRD_START + INITRD_SIZE <= (max_low_pfn << PAGE_SHIFT)) { reserve_bootmem_node(NODE_DATA(0), INITRD_START+__MEMORY_START, INITRD_SIZE); - initrd_start = - INITRD_START ? INITRD_START + PAGE_OFFSET + __MEMORY_START : 0; + initrd_start = INITRD_START + PAGE_OFFSET + __MEMORY_START; initrd_end = initrd_start + INITRD_SIZE; } else { printk("initrd extends beyond end of memory " diff --git a/arch/sh64/kernel/setup.c b/arch/sh64/kernel/setup.c index ffb310e33ce..b9e7d54d7b8 100644 --- a/arch/sh64/kernel/setup.c +++ b/arch/sh64/kernel/setup.c @@ -243,9 +243,7 @@ void __init setup_arch(char **cmdline_p) if (INITRD_START + INITRD_SIZE <= (PFN_PHYS(last_pfn))) { reserve_bootmem_node(NODE_DATA(0), INITRD_START + __MEMORY_START, INITRD_SIZE); - initrd_start = - (long) INITRD_START ? INITRD_START + PAGE_OFFSET + __MEMORY_START : 0; - + initrd_start = (long) INITRD_START + PAGE_OFFSET + __MEMORY_START; initrd_end = initrd_start + INITRD_SIZE; } else { printk("initrd extends beyond end of memory " diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index fc944b5e8f4..f12f266f3e9 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -471,8 +471,7 @@ void __init setup_arch(char **cmdline_p) if (LOADER_TYPE && INITRD_START) { if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) { reserve_bootmem_generic(INITRD_START, INITRD_SIZE); - initrd_start = - INITRD_START ? INITRD_START + PAGE_OFFSET : 0; + initrd_start = INITRD_START + PAGE_OFFSET; initrd_end = initrd_start+INITRD_SIZE; } else { -- cgit v1.2.3-70-g09d2 From b4c6c34a530b4d1c626f4ac0a884e0a9b849378c Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Wed, 6 Dec 2006 20:38:11 -0800 Subject: [PATCH] kprobes: enable booster on the preemptible kernel When we are unregistering a kprobe-booster, we can't release its instruction buffer immediately on the preemptive kernel, because some processes might be preempted on the buffer. The freeze_processes() and thaw_processes() functions can clean most of processes up from the buffer. There are still some non-frozen threads who have the PF_NOFREEZE flag. If those threads are sleeping (not preempted) at the known place outside the buffer, we can ensure safety of freeing. However, the processing of this check routine takes a long time. So, this patch introduces the garbage collection mechanism of insn_slot. It also introduces the "dirty" flag to free_insn_slot because of efficiency. The "clean" instruction slots (dirty flag is cleared) are released immediately. But the "dirty" slots which are used by boosted kprobes, are marked as garbages. collect_garbage_slots() will be invoked to release "dirty" slots if there are more than INSNS_PER_PAGE garbage slots or if there are no unused slots. Cc: "Keshavamurthy, Anil S" Cc: Ananth N Mavinakayanahalli Cc: "bibo,mao" Cc: Prasanna S Panchamukhi Cc: Yumiko Sugita Cc: Satoshi Oshima Cc: Hideo Aoki Signed-off-by: Masami Hiramatsu Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/kprobes.c | 4 +- arch/ia64/kernel/kprobes.c | 2 +- arch/powerpc/kernel/kprobes.c | 2 +- arch/s390/kernel/kprobes.c | 2 +- arch/x86_64/kernel/kprobes.c | 2 +- include/linux/kprobes.h | 2 +- kernel/kprobes.c | 117 ++++++++++++++++++++++++++++++++++-------- 7 files changed, 103 insertions(+), 28 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index fc79e1e859c..af1d5334499 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -184,7 +184,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1)); mutex_unlock(&kprobe_mutex); } @@ -333,7 +333,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) return 1; ss_probe: -#ifndef CONFIG_PREEMPT +#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM) if (p->ainsn.boostable == 1 && !p->post_handler){ /* Boost up -- we can execute copied instructions directly */ reset_current_kprobe(); diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index 51217d63285..4d592ee9300 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -481,7 +481,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, 0); mutex_unlock(&kprobe_mutex); } /* diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c index 7b8d12b9026..4657563f881 100644 --- a/arch/powerpc/kernel/kprobes.c +++ b/arch/powerpc/kernel/kprobes.c @@ -85,7 +85,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, 0); mutex_unlock(&kprobe_mutex); } diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index 67914fe7f31..576368c4f60 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -200,7 +200,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, 0); mutex_unlock(&kprobe_mutex); } diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c index ac241567e68..209c8c0bec7 100644 --- a/arch/x86_64/kernel/kprobes.c +++ b/arch/x86_64/kernel/kprobes.c @@ -224,7 +224,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, 0); mutex_unlock(&kprobe_mutex); } diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index ac4c0559f75..769be39b968 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -165,7 +165,7 @@ extern void arch_disarm_kprobe(struct kprobe *p); extern int arch_init_kprobes(void); extern void show_registers(struct pt_regs *regs); extern kprobe_opcode_t *get_insn_slot(void); -extern void free_insn_slot(kprobe_opcode_t *slot); +extern void free_insn_slot(kprobe_opcode_t *slot, int dirty); extern void kprobes_inc_nmissed_count(struct kprobe *p); /* Get the kprobe at this addr (if any) - called with preemption disabled */ diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 610c837ad9e..17ec4afb099 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include @@ -83,9 +84,36 @@ struct kprobe_insn_page { kprobe_opcode_t *insns; /* Page of instruction slots */ char slot_used[INSNS_PER_PAGE]; int nused; + int ngarbage; }; static struct hlist_head kprobe_insn_pages; +static int kprobe_garbage_slots; +static int collect_garbage_slots(void); + +static int __kprobes check_safety(void) +{ + int ret = 0; +#if defined(CONFIG_PREEMPT) && defined(CONFIG_PM) + ret = freeze_processes(); + if (ret == 0) { + struct task_struct *p, *q; + do_each_thread(p, q) { + if (p != current && p->state == TASK_RUNNING && + p->pid != 0) { + printk("Check failed: %s is running\n",p->comm); + ret = -1; + goto loop_end; + } + } while_each_thread(p, q); + } +loop_end: + thaw_processes(); +#else + synchronize_sched(); +#endif + return ret; +} /** * get_insn_slot() - Find a slot on an executable page for an instruction. @@ -96,6 +124,7 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) struct kprobe_insn_page *kip; struct hlist_node *pos; + retry: hlist_for_each(pos, &kprobe_insn_pages) { kip = hlist_entry(pos, struct kprobe_insn_page, hlist); if (kip->nused < INSNS_PER_PAGE) { @@ -112,7 +141,11 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) } } - /* All out of space. Need to allocate a new page. Use slot 0.*/ + /* If there are any garbage slots, collect it and try again. */ + if (kprobe_garbage_slots && collect_garbage_slots() == 0) { + goto retry; + } + /* All out of space. Need to allocate a new page. Use slot 0. */ kip = kmalloc(sizeof(struct kprobe_insn_page), GFP_KERNEL); if (!kip) { return NULL; @@ -133,10 +166,62 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) memset(kip->slot_used, 0, INSNS_PER_PAGE); kip->slot_used[0] = 1; kip->nused = 1; + kip->ngarbage = 0; return kip->insns; } -void __kprobes free_insn_slot(kprobe_opcode_t *slot) +/* Return 1 if all garbages are collected, otherwise 0. */ +static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx) +{ + kip->slot_used[idx] = 0; + kip->nused--; + if (kip->nused == 0) { + /* + * Page is no longer in use. Free it unless + * it's the last one. We keep the last one + * so as not to have to set it up again the + * next time somebody inserts a probe. + */ + hlist_del(&kip->hlist); + if (hlist_empty(&kprobe_insn_pages)) { + INIT_HLIST_NODE(&kip->hlist); + hlist_add_head(&kip->hlist, + &kprobe_insn_pages); + } else { + module_free(NULL, kip->insns); + kfree(kip); + } + return 1; + } + return 0; +} + +static int __kprobes collect_garbage_slots(void) +{ + struct kprobe_insn_page *kip; + struct hlist_node *pos, *next; + + /* Ensure no-one is preepmted on the garbages */ + if (check_safety() != 0) + return -EAGAIN; + + hlist_for_each_safe(pos, next, &kprobe_insn_pages) { + int i; + kip = hlist_entry(pos, struct kprobe_insn_page, hlist); + if (kip->ngarbage == 0) + continue; + kip->ngarbage = 0; /* we will collect all garbages */ + for (i = 0; i < INSNS_PER_PAGE; i++) { + if (kip->slot_used[i] == -1 && + collect_one_slot(kip, i)) + break; + } + } + kprobe_garbage_slots = 0; + return 0; +} + +void __kprobes free_insn_slot(kprobe_opcode_t * slot, int dirty) { struct kprobe_insn_page *kip; struct hlist_node *pos; @@ -146,28 +231,18 @@ void __kprobes free_insn_slot(kprobe_opcode_t *slot) if (kip->insns <= slot && slot < kip->insns + (INSNS_PER_PAGE * MAX_INSN_SIZE)) { int i = (slot - kip->insns) / MAX_INSN_SIZE; - kip->slot_used[i] = 0; - kip->nused--; - if (kip->nused == 0) { - /* - * Page is no longer in use. Free it unless - * it's the last one. We keep the last one - * so as not to have to set it up again the - * next time somebody inserts a probe. - */ - hlist_del(&kip->hlist); - if (hlist_empty(&kprobe_insn_pages)) { - INIT_HLIST_NODE(&kip->hlist); - hlist_add_head(&kip->hlist, - &kprobe_insn_pages); - } else { - module_free(NULL, kip->insns); - kfree(kip); - } + if (dirty) { + kip->slot_used[i] = -1; + kip->ngarbage++; + } else { + collect_one_slot(kip, i); } - return; + break; } } + if (dirty && (++kprobe_garbage_slots > INSNS_PER_PAGE)) { + collect_garbage_slots(); + } } #endif -- cgit v1.2.3-70-g09d2 From a38a44c1a93078fc5fadc4ac2df8dea4697069e2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 6 Dec 2006 20:38:16 -0800 Subject: [PATCH] smp_call_function_single() check that local interrupts are enabled smp_call_function_single() can deadlock if the caller disabled local interrupts (the target CPU could be spinning on call_lock). Check for that. Why on earth do these functions use spin_lock_bh()?? Cc: "Randy.Dunlap" Cc: Andi Kleen Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/smp.c | 4 ++++ arch/x86_64/kernel/smp.c | 4 ++++ 2 files changed, 8 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c index 31e5c6573aa..9827cf927ec 100644 --- a/arch/i386/kernel/smp.c +++ b/arch/i386/kernel/smp.c @@ -699,6 +699,10 @@ int smp_call_function_single(int cpu, void (*func) (void *info), void *info, put_cpu(); return -EBUSY; } + + /* Can deadlock when called with interrupts disabled */ + WARN_ON(irqs_disabled()); + spin_lock_bh(&call_lock); __smp_call_function_single(cpu, func, info, nonatomic, wait); spin_unlock_bh(&call_lock); diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c index 9f74c883568..32f4d7e2a06 100644 --- a/arch/x86_64/kernel/smp.c +++ b/arch/x86_64/kernel/smp.c @@ -379,6 +379,10 @@ int smp_call_function_single (int cpu, void (*func) (void *info), void *info, put_cpu(); return 0; } + + /* Can deadlock when called with interrupts disabled */ + WARN_ON(irqs_disabled()); + spin_lock_bh(&call_lock); __smp_call_function_single(cpu, func, info, nonatomic, wait); spin_unlock_bh(&call_lock); -- cgit v1.2.3-70-g09d2 From 02316067852187b8bec781bec07410e91af79627 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 6 Dec 2006 20:38:17 -0800 Subject: [PATCH] hotplug CPU: clean up hotcpu_notifier() use There was lots of #ifdef noise in the kernel due to hotcpu_notifier(fn, prio) not correctly marking 'fn' as used in the !HOTPLUG_CPU case, and thus generating compiler warnings of unused symbols, hence forcing people to add #ifdefs. the compiler can skip truly unused functions just fine: text data bss dec hex filename 1624412 728710 3674856 6027978 5bfaca vmlinux.before 1624412 728710 3674856 6027978 5bfaca vmlinux.after [akpm@osdl.org: topology.c fix] Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/cpu/mcheck/therm_throt.c | 2 -- arch/i386/kernel/cpuid.c | 2 -- arch/i386/kernel/microcode.c | 2 -- arch/i386/kernel/msr.c | 2 -- arch/ia64/kernel/palinfo.c | 2 -- arch/ia64/kernel/salinfo.c | 2 -- arch/s390/appldata/appldata_base.c | 2 -- arch/x86_64/kernel/mce.c | 2 -- arch/x86_64/kernel/mce_amd.c | 4 ++-- arch/x86_64/kernel/vsyscall.c | 2 -- block/ll_rw_blk.c | 4 ---- drivers/base/topology.c | 2 -- drivers/cpufreq/cpufreq.c | 2 -- fs/buffer.c | 2 -- include/linux/cpu.h | 6 +++--- kernel/cpuset.c | 4 ---- kernel/profile.c | 3 +-- kernel/sched.c | 3 --- kernel/workqueue.c | 2 -- lib/radix-tree.c | 2 -- mm/page_alloc.c | 4 ---- mm/swap.c | 2 ++ mm/vmscan.c | 2 -- net/core/dev.c | 2 -- net/core/flow.c | 2 -- 25 files changed, 8 insertions(+), 56 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c index bad8b442070..065005c3f16 100644 --- a/arch/i386/kernel/cpu/mcheck/therm_throt.c +++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c @@ -116,7 +116,6 @@ static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev) return sysfs_create_group(&sys_dev->kobj, &thermal_throttle_attr_group); } -#ifdef CONFIG_HOTPLUG_CPU static __cpuinit void thermal_throttle_remove_dev(struct sys_device *sys_dev) { return sysfs_remove_group(&sys_dev->kobj, &thermal_throttle_attr_group); @@ -153,7 +152,6 @@ static struct notifier_block thermal_throttle_cpu_notifier = { .notifier_call = thermal_throttle_cpu_callback, }; -#endif /* CONFIG_HOTPLUG_CPU */ static __init int thermal_throttle_init_device(void) { diff --git a/arch/i386/kernel/cpuid.c b/arch/i386/kernel/cpuid.c index ab0c327e79d..23b2cc748d4 100644 --- a/arch/i386/kernel/cpuid.c +++ b/arch/i386/kernel/cpuid.c @@ -167,7 +167,6 @@ static int cpuid_device_create(int i) return err; } -#ifdef CONFIG_HOTPLUG_CPU static int cpuid_class_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -187,7 +186,6 @@ static struct notifier_block __cpuinitdata cpuid_class_cpu_notifier = { .notifier_call = cpuid_class_cpu_callback, }; -#endif /* !CONFIG_HOTPLUG_CPU */ static int __init cpuid_init(void) { diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c index 23f5984d065..972346604f9 100644 --- a/arch/i386/kernel/microcode.c +++ b/arch/i386/kernel/microcode.c @@ -703,7 +703,6 @@ static struct sysdev_driver mc_sysdev_driver = { .resume = mc_sysdev_resume, }; -#ifdef CONFIG_HOTPLUG_CPU static __cpuinit int mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) { @@ -726,7 +725,6 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) static struct notifier_block mc_cpu_notifier = { .notifier_call = mc_cpu_callback, }; -#endif static int __init microcode_init (void) { diff --git a/arch/i386/kernel/msr.c b/arch/i386/kernel/msr.c index a773f776c9e..7763c67ca28 100644 --- a/arch/i386/kernel/msr.c +++ b/arch/i386/kernel/msr.c @@ -250,7 +250,6 @@ static int msr_device_create(int i) return err; } -#ifdef CONFIG_HOTPLUG_CPU static int msr_class_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { @@ -271,7 +270,6 @@ static struct notifier_block __cpuinitdata msr_class_cpu_notifier = { .notifier_call = msr_class_cpu_callback, }; -#endif static int __init msr_init(void) { diff --git a/arch/ia64/kernel/palinfo.c b/arch/ia64/kernel/palinfo.c index 0b546e2b36a..c4c10a0b99d 100644 --- a/arch/ia64/kernel/palinfo.c +++ b/arch/ia64/kernel/palinfo.c @@ -952,7 +952,6 @@ remove_palinfo_proc_entries(unsigned int hcpu) } } -#ifdef CONFIG_HOTPLUG_CPU static int palinfo_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { @@ -974,7 +973,6 @@ static struct notifier_block palinfo_cpu_notifier = .notifier_call = palinfo_cpu_callback, .priority = 0, }; -#endif static int __init palinfo_init(void) diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c index e63b8ca5344..fd607ca51a8 100644 --- a/arch/ia64/kernel/salinfo.c +++ b/arch/ia64/kernel/salinfo.c @@ -575,7 +575,6 @@ static struct file_operations salinfo_data_fops = { .write = salinfo_log_write, }; -#ifdef CONFIG_HOTPLUG_CPU static int __devinit salinfo_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) { @@ -620,7 +619,6 @@ static struct notifier_block salinfo_cpu_notifier = .notifier_call = salinfo_cpu_callback, .priority = 0, }; -#endif /* CONFIG_HOTPLUG_CPU */ static int __init salinfo_init(void) diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index 67d5cf9cba8..b8c23729026 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -561,7 +561,6 @@ appldata_offline_cpu(int cpu) spin_unlock(&appldata_timer_lock); } -#ifdef CONFIG_HOTPLUG_CPU static int __cpuinit appldata_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) @@ -582,7 +581,6 @@ appldata_cpu_notify(struct notifier_block *self, static struct notifier_block appldata_nb = { .notifier_call = appldata_cpu_notify, }; -#endif /* * appldata_init() diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index c7587fc3901..bc863c464a1 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -641,7 +641,6 @@ static __cpuinit int mce_create_device(unsigned int cpu) return err; } -#ifdef CONFIG_HOTPLUG_CPU static void mce_remove_device(unsigned int cpu) { int i; @@ -674,7 +673,6 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) static struct notifier_block mce_cpu_notifier = { .notifier_call = mce_cpu_callback, }; -#endif static __init int mce_init_device(void) { diff --git a/arch/x86_64/kernel/mce_amd.c b/arch/x86_64/kernel/mce_amd.c index 883fe747f64..fa09debad4b 100644 --- a/arch/x86_64/kernel/mce_amd.c +++ b/arch/x86_64/kernel/mce_amd.c @@ -551,7 +551,6 @@ out: return err; } -#ifdef CONFIG_HOTPLUG_CPU /* * let's be hotplug friendly. * in case of multiple core processors, the first core always takes ownership @@ -594,12 +593,14 @@ static void threshold_remove_bank(unsigned int cpu, int bank) sprintf(name, "threshold_bank%i", bank); +#ifdef CONFIG_SMP /* sibling symlink */ if (shared_bank[bank] && b->blocks->cpu != cpu) { sysfs_remove_link(&per_cpu(device_mce, cpu).kobj, name); per_cpu(threshold_banks, cpu)[bank] = NULL; return; } +#endif /* remove all sibling symlinks before unregistering */ for_each_cpu_mask(i, b->cpus) { @@ -656,7 +657,6 @@ static int threshold_cpu_callback(struct notifier_block *nfb, static struct notifier_block threshold_cpu_notifier = { .notifier_call = threshold_cpu_callback, }; -#endif /* CONFIG_HOTPLUG_CPU */ static __init int threshold_init_device(void) { diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index 630036c06c7..3785e495473 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -275,7 +275,6 @@ static void __cpuinit cpu_vsyscall_init(void *arg) vsyscall_set_cpu(raw_smp_processor_id()); } -#ifdef CONFIG_HOTPLUG_CPU static int __cpuinit cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) { @@ -284,7 +283,6 @@ cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1); return NOTIFY_DONE; } -#endif static void __init map_vsyscall(void) { diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index a4ff3271d4a..31512cd9f3a 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -3459,8 +3459,6 @@ static void blk_done_softirq(struct softirq_action *h) } } -#ifdef CONFIG_HOTPLUG_CPU - static int blk_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { @@ -3486,8 +3484,6 @@ static struct notifier_block __devinitdata blk_cpu_notifier = { .notifier_call = blk_cpu_notify, }; -#endif /* CONFIG_HOTPLUG_CPU */ - /** * blk_complete_request - end I/O on a request * @req: the request being processed diff --git a/drivers/base/topology.c b/drivers/base/topology.c index 3d12b85b096..067a9e8bc37 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -108,7 +108,6 @@ static int __cpuinit topology_add_dev(unsigned int cpu) return rc; } -#ifdef CONFIG_HOTPLUG_CPU static void __cpuinit topology_remove_dev(unsigned int cpu) { struct sys_device *sys_dev = get_cpu_sysdev(cpu); @@ -136,7 +135,6 @@ static int __cpuinit topology_cpu_callback(struct notifier_block *nfb, } return rc ? NOTIFY_BAD : NOTIFY_OK; } -#endif static int __cpuinit topology_sysfs_init(void) { diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 7a7c6e6dfe4..47ab42db122 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1537,7 +1537,6 @@ int cpufreq_update_policy(unsigned int cpu) } EXPORT_SYMBOL(cpufreq_update_policy); -#ifdef CONFIG_HOTPLUG_CPU static int cpufreq_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { @@ -1577,7 +1576,6 @@ static struct notifier_block __cpuinitdata cpufreq_cpu_notifier = { .notifier_call = cpufreq_cpu_callback, }; -#endif /* CONFIG_HOTPLUG_CPU */ /********************************************************************* * REGISTER / UNREGISTER CPUFREQ DRIVER * diff --git a/fs/buffer.c b/fs/buffer.c index a8ca0ac2148..517860f2d75 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2972,7 +2972,6 @@ init_buffer_head(void *data, struct kmem_cache *cachep, unsigned long flags) } } -#ifdef CONFIG_HOTPLUG_CPU static void buffer_exit_cpu(int cpu) { int i; @@ -2994,7 +2993,6 @@ static int buffer_cpu_notify(struct notifier_block *self, buffer_exit_cpu((unsigned long)hcpu); return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ void __init buffer_init(void) { diff --git a/include/linux/cpu.h b/include/linux/cpu.h index f02d71bf689..71dc6ba4f73 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -89,9 +89,9 @@ int cpu_down(unsigned int cpu); #define lock_cpu_hotplug() do { } while (0) #define unlock_cpu_hotplug() do { } while (0) #define lock_cpu_hotplug_interruptible() 0 -#define hotcpu_notifier(fn, pri) do { } while (0) -#define register_hotcpu_notifier(nb) do { } while (0) -#define unregister_hotcpu_notifier(nb) do { } while (0) +#define hotcpu_notifier(fn, pri) do { (void)(fn); } while (0) +#define register_hotcpu_notifier(nb) do { (void)(nb); } while (0) +#define unregister_hotcpu_notifier(nb) do { (void)(nb); } while (0) /* CPUs don't go offline once they're online w/o CONFIG_HOTPLUG_CPU */ static inline int cpu_is_offline(int cpu) { return 0; } diff --git a/kernel/cpuset.c b/kernel/cpuset.c index bd1e89c4c96..9b62b4c03ad 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -2044,7 +2044,6 @@ out: return err; } -#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_MEMORY_HOTPLUG) /* * If common_cpu_mem_hotplug_unplug(), below, unplugs any CPUs * or memory nodes, we need to walk over the cpuset hierarchy, @@ -2108,9 +2107,7 @@ static void common_cpu_mem_hotplug_unplug(void) mutex_unlock(&callback_mutex); mutex_unlock(&manage_mutex); } -#endif -#ifdef CONFIG_HOTPLUG_CPU /* * The top_cpuset tracks what CPUs and Memory Nodes are online, * period. This is necessary in order to make cpusets transparent @@ -2127,7 +2124,6 @@ static int cpuset_handle_cpuhp(struct notifier_block *nb, common_cpu_mem_hotplug_unplug(); return 0; } -#endif #ifdef CONFIG_MEMORY_HOTPLUG /* diff --git a/kernel/profile.c b/kernel/profile.c index 04fd84e8cdb..0961d93e1d9 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -319,7 +319,6 @@ out: put_cpu(); } -#ifdef CONFIG_HOTPLUG_CPU static int __devinit profile_cpu_callback(struct notifier_block *info, unsigned long action, void *__cpu) { @@ -372,10 +371,10 @@ static int __devinit profile_cpu_callback(struct notifier_block *info, } return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ #else /* !CONFIG_SMP */ #define profile_flip_buffers() do { } while (0) #define profile_discard_flip_buffers() do { } while (0) +#define profile_cpu_callback NULL void profile_hits(int type, void *__pc, unsigned int nr_hits) { diff --git a/kernel/sched.c b/kernel/sched.c index 75a005ed4ed..c83f531c288 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6740,8 +6740,6 @@ SYSDEV_ATTR(sched_smt_power_savings, 0644, sched_smt_power_savings_show, sched_smt_power_savings_store); #endif - -#ifdef CONFIG_HOTPLUG_CPU /* * Force a reinitialization of the sched domains hierarchy. The domains * and groups cannot be updated in place without racing with the balancing @@ -6774,7 +6772,6 @@ static int update_sched_domains(struct notifier_block *nfb, return NOTIFY_OK; } -#endif void __init sched_init_smp(void) { diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 5484d6e045c..c5257316f4b 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -655,7 +655,6 @@ int current_is_keventd(void) } -#ifdef CONFIG_HOTPLUG_CPU /* Take the work from this (downed) CPU. */ static void take_over_work(struct workqueue_struct *wq, unsigned int cpu) { @@ -738,7 +737,6 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -#endif void init_workqueues(void) { diff --git a/lib/radix-tree.c b/lib/radix-tree.c index e2cefabb5aa..d69ddbe4386 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -996,7 +996,6 @@ static __init void radix_tree_init_maxindex(void) height_to_maxindex[i] = __maxindex(i); } -#ifdef CONFIG_HOTPLUG_CPU static int radix_tree_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) @@ -1016,7 +1015,6 @@ static int radix_tree_callback(struct notifier_block *nfb, } return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ void __init radix_tree_init(void) { diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 2273952300d..27ec7a1b802 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -701,7 +701,6 @@ void drain_node_pages(int nodeid) } #endif -#if defined(CONFIG_PM) || defined(CONFIG_HOTPLUG_CPU) static void __drain_pages(unsigned int cpu) { unsigned long flags; @@ -723,7 +722,6 @@ static void __drain_pages(unsigned int cpu) } } } -#endif /* CONFIG_PM || CONFIG_HOTPLUG_CPU */ #ifdef CONFIG_PM @@ -2907,7 +2905,6 @@ void __init free_area_init(unsigned long *zones_size) __pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL); } -#ifdef CONFIG_HOTPLUG_CPU static int page_alloc_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { @@ -2922,7 +2919,6 @@ static int page_alloc_cpu_notify(struct notifier_block *self, } return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ void __init page_alloc_init(void) { diff --git a/mm/swap.c b/mm/swap.c index 017e72ca9bb..2ed7be39795 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -514,5 +514,7 @@ void __init swap_setup(void) * Right now other parts of the system means that we * _really_ don't want to cluster much more */ +#ifdef CONFIG_HOTPLUG_CPU hotcpu_notifier(cpu_swap_callback, 0); +#endif } diff --git a/mm/vmscan.c b/mm/vmscan.c index f6616e81fac..093f5fe6dd7 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1513,7 +1513,6 @@ out: } #endif -#ifdef CONFIG_HOTPLUG_CPU /* It's optimal to keep kswapds on the same CPUs as their memory, but not required for correctness. So if the last cpu in a node goes away, we get changed to run anywhere: as the first one comes back, @@ -1534,7 +1533,6 @@ static int __devinit cpu_callback(struct notifier_block *nfb, } return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ /* * This kswapd start function will be called by init and node-hot-add. diff --git a/net/core/dev.c b/net/core/dev.c index 59d058a3b50..e660cb57e42 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3340,7 +3340,6 @@ void unregister_netdev(struct net_device *dev) EXPORT_SYMBOL(unregister_netdev); -#ifdef CONFIG_HOTPLUG_CPU static int dev_cpu_callback(struct notifier_block *nfb, unsigned long action, void *ocpu) @@ -3384,7 +3383,6 @@ static int dev_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ #ifdef CONFIG_NET_DMA /** diff --git a/net/core/flow.c b/net/core/flow.c index 104c25d00a1..d137f971f97 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -340,7 +340,6 @@ static void __devinit flow_cache_cpu_prepare(int cpu) tasklet_init(tasklet, flow_cache_flush_tasklet, 0); } -#ifdef CONFIG_HOTPLUG_CPU static int flow_cache_cpu(struct notifier_block *nfb, unsigned long action, void *hcpu) @@ -349,7 +348,6 @@ static int flow_cache_cpu(struct notifier_block *nfb, __flow_cache_shrink((unsigned long)hcpu, 0); return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ static int __init flow_cache_init(void) { -- cgit v1.2.3-70-g09d2 From cd6ed52568e161ce924593ebc798050a2d23cca0 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 6 Dec 2006 20:40:06 -0800 Subject: [PATCH] arch/i386/kernel/reboot.c should #include Every file should #include the headers containing the prototypes for its global functions. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/reboot.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c index 84278e0093a..3514b4153f7 100644 --- a/arch/i386/kernel/reboot.c +++ b/arch/i386/kernel/reboot.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3-70-g09d2 From 85916f8166b59eeac63d2b4f7f1df8de849334b4 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 6 Dec 2006 20:40:41 -0800 Subject: [PATCH] Kexec / Kdump: Unify elf note code The elf note saving code is currently duplicated over several architectures. This cleanup patch simply adds code to a common file and then replaces the arch-specific code with calls to the newly added code. The only drawback with this approach is that s390 doesn't fully support kexec-on-panic which for that arch leads to introduction of unused code. Signed-off-by: Magnus Damm Cc: Vivek Goyal Cc: Andi Kleen Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/crash.c | 66 ++----------------------------------------- arch/powerpc/kernel/crash.c | 59 ++------------------------------------ arch/x86_64/kernel/crash.c | 69 ++------------------------------------------- include/linux/kexec.h | 1 + kernel/kexec.c | 56 ++++++++++++++++++++++++++++++++++++ 5 files changed, 63 insertions(+), 188 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c index 144b4328896..a5e0e990ea9 100644 --- a/arch/i386/kernel/crash.c +++ b/arch/i386/kernel/crash.c @@ -31,68 +31,6 @@ /* This keeps a track of which one is crashing cpu. */ static int crashing_cpu; -static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, - size_t data_len) -{ - struct elf_note note; - - note.n_namesz = strlen(name) + 1; - note.n_descsz = data_len; - note.n_type = type; - memcpy(buf, ¬e, sizeof(note)); - buf += (sizeof(note) +3)/4; - memcpy(buf, name, note.n_namesz); - buf += (note.n_namesz + 3)/4; - memcpy(buf, data, note.n_descsz); - buf += (note.n_descsz + 3)/4; - - return buf; -} - -static void final_note(u32 *buf) -{ - struct elf_note note; - - note.n_namesz = 0; - note.n_descsz = 0; - note.n_type = 0; - memcpy(buf, ¬e, sizeof(note)); -} - -static void crash_save_this_cpu(struct pt_regs *regs, int cpu) -{ - struct elf_prstatus prstatus; - u32 *buf; - - if ((cpu < 0) || (cpu >= NR_CPUS)) - return; - - /* Using ELF notes here is opportunistic. - * I need a well defined structure format - * for the data I pass, and I need tags - * on the data to indicate what information I have - * squirrelled away. ELF notes happen to provide - * all of that, so there is no need to invent something new. - */ - buf = (u32*)per_cpu_ptr(crash_notes, cpu); - if (!buf) - return; - memset(&prstatus, 0, sizeof(prstatus)); - prstatus.pr_pid = current->pid; - elf_core_copy_regs(&prstatus.pr_reg, regs); - buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, - sizeof(prstatus)); - final_note(buf); -} - -static void crash_save_self(struct pt_regs *regs) -{ - int cpu; - - cpu = safe_smp_processor_id(); - crash_save_this_cpu(regs, cpu); -} - #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) static atomic_t waiting_for_crash_ipi; @@ -121,7 +59,7 @@ static int crash_nmi_callback(struct notifier_block *self, crash_fixup_ss_esp(&fixed_regs, regs); regs = &fixed_regs; } - crash_save_this_cpu(regs, cpu); + crash_save_cpu(regs, cpu); disable_local_APIC(); atomic_dec(&waiting_for_crash_ipi); /* Assume hlt works */ @@ -195,5 +133,5 @@ void machine_crash_shutdown(struct pt_regs *regs) #if defined(CONFIG_X86_IO_APIC) disable_IO_APIC(); #endif - crash_save_self(regs); + crash_save_cpu(regs, safe_smp_processor_id()); } diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index 89b03c8da9d..d3f2080d2ee 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -46,61 +46,6 @@ int crashing_cpu = -1; static cpumask_t cpus_in_crash = CPU_MASK_NONE; cpumask_t cpus_in_sr = CPU_MASK_NONE; -static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, - size_t data_len) -{ - struct elf_note note; - - note.n_namesz = strlen(name) + 1; - note.n_descsz = data_len; - note.n_type = type; - memcpy(buf, ¬e, sizeof(note)); - buf += (sizeof(note) +3)/4; - memcpy(buf, name, note.n_namesz); - buf += (note.n_namesz + 3)/4; - memcpy(buf, data, note.n_descsz); - buf += (note.n_descsz + 3)/4; - - return buf; -} - -static void final_note(u32 *buf) -{ - struct elf_note note; - - note.n_namesz = 0; - note.n_descsz = 0; - note.n_type = 0; - memcpy(buf, ¬e, sizeof(note)); -} - -static void crash_save_this_cpu(struct pt_regs *regs, int cpu) -{ - struct elf_prstatus prstatus; - u32 *buf; - - if ((cpu < 0) || (cpu >= NR_CPUS)) - return; - - /* Using ELF notes here is opportunistic. - * I need a well defined structure format - * for the data I pass, and I need tags - * on the data to indicate what information I have - * squirrelled away. ELF notes happen to provide - * all of that that no need to invent something new. - */ - buf = (u32*)per_cpu_ptr(crash_notes, cpu); - if (!buf) - return; - - memset(&prstatus, 0, sizeof(prstatus)); - prstatus.pr_pid = current->pid; - elf_core_copy_regs(&prstatus.pr_reg, regs); - buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, - sizeof(prstatus)); - final_note(buf); -} - #ifdef CONFIG_SMP static atomic_t enter_on_soft_reset = ATOMIC_INIT(0); @@ -113,7 +58,7 @@ void crash_ipi_callback(struct pt_regs *regs) hard_irq_disable(); if (!cpu_isset(cpu, cpus_in_crash)) - crash_save_this_cpu(regs, cpu); + crash_save_cpu(regs, cpu); cpu_set(cpu, cpus_in_crash); /* @@ -306,7 +251,7 @@ void default_machine_crash_shutdown(struct pt_regs *regs) * such that another IPI will not be sent. */ crashing_cpu = smp_processor_id(); - crash_save_this_cpu(regs, crashing_cpu); + crash_save_cpu(regs, crashing_cpu); crash_kexec_prepare_cpus(crashing_cpu); cpu_set(crashing_cpu, cpus_in_crash); if (ppc_md.kexec_cpu_down) diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c index 3525f884af8..95a7a2c1313 100644 --- a/arch/x86_64/kernel/crash.c +++ b/arch/x86_64/kernel/crash.c @@ -28,71 +28,6 @@ /* This keeps a track of which one is crashing cpu. */ static int crashing_cpu; -static u32 *append_elf_note(u32 *buf, char *name, unsigned type, - void *data, size_t data_len) -{ - struct elf_note note; - - note.n_namesz = strlen(name) + 1; - note.n_descsz = data_len; - note.n_type = type; - memcpy(buf, ¬e, sizeof(note)); - buf += (sizeof(note) +3)/4; - memcpy(buf, name, note.n_namesz); - buf += (note.n_namesz + 3)/4; - memcpy(buf, data, note.n_descsz); - buf += (note.n_descsz + 3)/4; - - return buf; -} - -static void final_note(u32 *buf) -{ - struct elf_note note; - - note.n_namesz = 0; - note.n_descsz = 0; - note.n_type = 0; - memcpy(buf, ¬e, sizeof(note)); -} - -static void crash_save_this_cpu(struct pt_regs *regs, int cpu) -{ - struct elf_prstatus prstatus; - u32 *buf; - - if ((cpu < 0) || (cpu >= NR_CPUS)) - return; - - /* Using ELF notes here is opportunistic. - * I need a well defined structure format - * for the data I pass, and I need tags - * on the data to indicate what information I have - * squirrelled away. ELF notes happen to provide - * all of that, no need to invent something new. - */ - - buf = (u32*)per_cpu_ptr(crash_notes, cpu); - - if (!buf) - return; - - memset(&prstatus, 0, sizeof(prstatus)); - prstatus.pr_pid = current->pid; - elf_core_copy_regs(&prstatus.pr_reg, regs); - buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, - sizeof(prstatus)); - final_note(buf); -} - -static void crash_save_self(struct pt_regs *regs) -{ - int cpu; - - cpu = smp_processor_id(); - crash_save_this_cpu(regs, cpu); -} - #ifdef CONFIG_SMP static atomic_t waiting_for_crash_ipi; @@ -117,7 +52,7 @@ static int crash_nmi_callback(struct notifier_block *self, return NOTIFY_STOP; local_irq_disable(); - crash_save_this_cpu(regs, cpu); + crash_save_cpu(regs, cpu); disable_local_APIC(); atomic_dec(&waiting_for_crash_ipi); /* Assume hlt works */ @@ -196,5 +131,5 @@ void machine_crash_shutdown(struct pt_regs *regs) disable_IO_APIC(); - crash_save_self(regs); + crash_save_cpu(regs, smp_processor_id()); } diff --git a/include/linux/kexec.h b/include/linux/kexec.h index a4ede62b339..e3abcec6c51 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -105,6 +105,7 @@ extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order); extern void crash_kexec(struct pt_regs *); int kexec_should_crash(struct task_struct *); +void crash_save_cpu(struct pt_regs *regs, int cpu); extern struct kimage *kexec_image; extern struct kimage *kexec_crash_image; diff --git a/kernel/kexec.c b/kernel/kexec.c index d43692cf232..afbbbe981be 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -20,6 +20,8 @@ #include #include #include +#include +#include #include #include @@ -1066,6 +1068,60 @@ void crash_kexec(struct pt_regs *regs) } } +static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, + size_t data_len) +{ + struct elf_note note; + + note.n_namesz = strlen(name) + 1; + note.n_descsz = data_len; + note.n_type = type; + memcpy(buf, ¬e, sizeof(note)); + buf += (sizeof(note) + 3)/4; + memcpy(buf, name, note.n_namesz); + buf += (note.n_namesz + 3)/4; + memcpy(buf, data, note.n_descsz); + buf += (note.n_descsz + 3)/4; + + return buf; +} + +static void final_note(u32 *buf) +{ + struct elf_note note; + + note.n_namesz = 0; + note.n_descsz = 0; + note.n_type = 0; + memcpy(buf, ¬e, sizeof(note)); +} + +void crash_save_cpu(struct pt_regs *regs, int cpu) +{ + struct elf_prstatus prstatus; + u32 *buf; + + if ((cpu < 0) || (cpu >= NR_CPUS)) + return; + + /* Using ELF notes here is opportunistic. + * I need a well defined structure format + * for the data I pass, and I need tags + * on the data to indicate what information I have + * squirrelled away. ELF notes happen to provide + * all of that, so there is no need to invent something new. + */ + buf = (u32*)per_cpu_ptr(crash_notes, cpu); + if (!buf) + return; + memset(&prstatus, 0, sizeof(prstatus)); + prstatus.pr_pid = current->pid; + elf_core_copy_regs(&prstatus.pr_reg, regs); + buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, + sizeof(prstatus)); + final_note(buf); +} + static int __init crash_notes_memory_init(void) { /* Allocate memory for saving cpu registers. */ -- cgit v1.2.3-70-g09d2 From 91768d6c2bad0d2766a166f13f2f57e197de3458 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 8 Dec 2006 02:36:21 -0800 Subject: [PATCH] Generic BUG for i386 This makes i386 use the generic BUG machinery. There are no functional changes from the old i386 implementation. The main advantage in using the generic BUG machinery for i386 is that the inlined overhead of BUG is just the ud2a instruction; the file+line(+function) information are no longer inlined into the instruction stream. This reduces cache pollution, and makes disassembly work properly. Signed-off-by: Jeremy Fitzhardinge Cc: Andi Kleen Cc: Hugh Dickens Cc: Michael Ellerman Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/Kconfig | 5 +++++ arch/i386/kernel/module.c | 4 +++- arch/i386/kernel/traps.c | 41 +++++++++++------------------------------ arch/i386/kernel/vmlinux.lds.S | 2 ++ include/asm-i386/bug.h | 28 ++++++++++++++++++++-------- lib/Kconfig.debug | 2 +- 6 files changed, 42 insertions(+), 40 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index ea70359b02d..c2362c7ba74 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -49,6 +49,11 @@ config GENERIC_IOMAP bool default y +config GENERIC_BUG + bool + default y + depends on BUG + config GENERIC_HWEIGHT bool default y diff --git a/arch/i386/kernel/module.c b/arch/i386/kernel/module.c index d7d9c8b23f7..3db0a5442eb 100644 --- a/arch/i386/kernel/module.c +++ b/arch/i386/kernel/module.c @@ -21,6 +21,7 @@ #include #include #include +#include #if 0 #define DEBUGP printk @@ -141,10 +142,11 @@ int module_finalize(const Elf_Ehdr *hdr, apply_paravirt(pseg, pseg + para->sh_size); } - return 0; + return module_bug_finalize(hdr, sechdrs, me); } void module_arch_cleanup(struct module *mod) { alternatives_smp_module_del(mod); + module_bug_cleanup(mod); } diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 68de48e498c..2b30dbf8d11 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -30,6 +30,7 @@ #include #include #include +#include #ifdef CONFIG_EISA #include @@ -420,43 +421,22 @@ void show_registers(struct pt_regs *regs) printk("\n"); } -static void handle_BUG(struct pt_regs *regs) +int is_valid_bugaddr(unsigned long eip) { - unsigned long eip = regs->eip; unsigned short ud2; if (eip < PAGE_OFFSET) - return; + return 0; if (probe_kernel_address((unsigned short *)eip, ud2)) - return; - if (ud2 != 0x0b0f) - return; - - printk(KERN_EMERG "------------[ cut here ]------------\n"); - -#ifdef CONFIG_DEBUG_BUGVERBOSE - do { - unsigned short line; - char *file; - char c; - - if (probe_kernel_address((unsigned short *)(eip + 2), line)) - break; - if (probe_kernel_address((char **)(eip + 4), file) || - (unsigned long)file < PAGE_OFFSET || - probe_kernel_address(file, c)) - file = ""; + return 0; - printk(KERN_EMERG "kernel BUG at %s:%d!\n", file, line); - return; - } while (0); -#endif - printk(KERN_EMERG "Kernel BUG at [verbose debug info unavailable]\n"); + return ud2 == 0x0b0f; } -/* This is gone through when something in the kernel - * has done something bad and is about to be terminated. -*/ +/* + * This is gone through when something in the kernel has done something bad and + * is about to be terminated. + */ void die(const char * str, struct pt_regs * regs, long err) { static struct { @@ -488,7 +468,8 @@ void die(const char * str, struct pt_regs * regs, long err) unsigned long esp; unsigned short ss; - handle_BUG(regs); + report_bug(regs->eip); + printk(KERN_EMERG "%s: %04lx [#%d]\n", str, err & 0xffff, ++die_counter); #ifdef CONFIG_PREEMPT printk(KERN_EMERG "PREEMPT "); diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 56e6ad5cb04..16d3c7133ad 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -57,6 +57,8 @@ SECTIONS RODATA + BUG_TABLE + . = ALIGN(4); .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) { __tracedata_start = .; diff --git a/include/asm-i386/bug.h b/include/asm-i386/bug.h index 8062cdbf258..b0fd78ca261 100644 --- a/include/asm-i386/bug.h +++ b/include/asm-i386/bug.h @@ -4,20 +4,32 @@ /* * Tell the user there is some problem. - * The offending file and line are encoded after the "officially - * undefined" opcode for parsing in the trap handler. + * The offending file and line are encoded encoded in the __bug_table section. */ #ifdef CONFIG_BUG #define HAVE_ARCH_BUG + #ifdef CONFIG_DEBUG_BUGVERBOSE -#define BUG() \ - __asm__ __volatile__( "ud2\n" \ - "\t.word %c0\n" \ - "\t.long %c1\n" \ - : : "i" (__LINE__), "i" (__FILE__)) +#define BUG() \ + do { \ + asm volatile("1:\tud2\n" \ + ".pushsection __bug_table,\"a\"\n" \ + "2:\t.long 1b, %c0\n" \ + "\t.word %c1, 0\n" \ + "\t.org 2b+%c2\n" \ + ".popsection" \ + : : "i" (__FILE__), "i" (__LINE__), \ + "i" (sizeof(struct bug_entry))); \ + for(;;) ; \ + } while(0) + #else -#define BUG() __asm__ __volatile__("ud2\n") +#define BUG() \ + do { \ + asm volatile("ud2"); \ + for(;;) ; \ + } while(0) #endif #endif diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index ee46fb335d2..2c133c09860 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -285,7 +285,7 @@ config DEBUG_HIGHMEM config DEBUG_BUGVERBOSE bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EMBEDDED depends on BUG - depends on ARM || ARM26 || AVR32 || M32R || M68K || SPARC32 || SPARC64 || X86_32 || FRV || SUPERH || GENERIC_BUG + depends on ARM || ARM26 || AVR32 || M32R || M68K || SPARC32 || SPARC64 || FRV || SUPERH || GENERIC_BUG default !EMBEDDED help Say Y here to make BUG() panics output the file name and line number -- cgit v1.2.3-70-g09d2 From aab4c5a51cd6defc374ba1961c996781656d5283 Mon Sep 17 00:00:00 2001 From: "Josef \"Jeff\" Sipek" Date: Fri, 8 Dec 2006 02:36:42 -0800 Subject: [PATCH] i386: change uses of f_{dentry, vfsmnt} to use f_path Change all the uses of f_{dentry,vfsmnt} to f_path.{dentry,mnt} in the i386 arch code. Signed-off-by: Josef "Jeff" Sipek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/cpuid.c | 4 ++-- arch/i386/kernel/msr.c | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/cpuid.c b/arch/i386/kernel/cpuid.c index db6dd20c358..51130b39cd2 100644 --- a/arch/i386/kernel/cpuid.c +++ b/arch/i386/kernel/cpuid.c @@ -116,7 +116,7 @@ static ssize_t cpuid_read(struct file *file, char __user *buf, char __user *tmp = buf; u32 data[4]; u32 reg = *ppos; - int cpu = iminor(file->f_dentry->d_inode); + int cpu = iminor(file->f_path.dentry->d_inode); if (count % 16) return -EINVAL; /* Invalid chunk size */ @@ -134,7 +134,7 @@ static ssize_t cpuid_read(struct file *file, char __user *buf, static int cpuid_open(struct inode *inode, struct file *file) { - unsigned int cpu = iminor(file->f_dentry->d_inode); + unsigned int cpu = iminor(file->f_path.dentry->d_inode); struct cpuinfo_x86 *c = &(cpu_data)[cpu]; if (cpu >= NR_CPUS || !cpu_online(cpu)) diff --git a/arch/i386/kernel/msr.c b/arch/i386/kernel/msr.c index 1d1a56cae34..4a472a17d1c 100644 --- a/arch/i386/kernel/msr.c +++ b/arch/i386/kernel/msr.c @@ -172,7 +172,7 @@ static ssize_t msr_read(struct file *file, char __user * buf, u32 __user *tmp = (u32 __user *) buf; u32 data[2]; u32 reg = *ppos; - int cpu = iminor(file->f_dentry->d_inode); + int cpu = iminor(file->f_path.dentry->d_inode); int err; if (count % 8) @@ -196,7 +196,7 @@ static ssize_t msr_write(struct file *file, const char __user *buf, const u32 __user *tmp = (const u32 __user *)buf; u32 data[2]; u32 reg = *ppos; - int cpu = iminor(file->f_dentry->d_inode); + int cpu = iminor(file->f_path.dentry->d_inode); int err; if (count % 8) @@ -216,7 +216,7 @@ static ssize_t msr_write(struct file *file, const char __user *buf, static int msr_open(struct inode *inode, struct file *file) { - unsigned int cpu = iminor(file->f_dentry->d_inode); + unsigned int cpu = iminor(file->f_path.dentry->d_inode); struct cpuinfo_x86 *c = &(cpu_data)[cpu]; if (cpu >= NR_CPUS || !cpu_online(cpu)) -- cgit v1.2.3-70-g09d2 From f0d1b0b30d250a07627ad8b9fbbb5c7cc08422e8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 8 Dec 2006 02:37:49 -0800 Subject: [PATCH] LOG2: Implement a general integer log2 facility in the kernel This facility provides three entry points: ilog2() Log base 2 of unsigned long ilog2_u32() Log base 2 of u32 ilog2_u64() Log base 2 of u64 These facilities can either be used inside functions on dynamic data: int do_something(long q) { ...; y = ilog2(x) ...; } Or can be used to statically initialise global variables with constant values: unsigned n = ilog2(27); When performing static initialisation, the compiler will report "error: initializer element is not constant" if asked to take a log of zero or of something not reducible to a constant. They treat negative numbers as unsigned. When not dealing with a constant, they fall back to using fls() which permits them to use arch-specific log calculation instructions - such as BSR on x86/x86_64 or SCAN on FRV - if available. [akpm@osdl.org: MMC fix] Signed-off-by: David Howells Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Herbert Xu Cc: David Howells Cc: Wojtek Kaniewski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/Kconfig | 8 ++ arch/arm/Kconfig | 8 ++ arch/arm26/Kconfig | 8 ++ arch/avr32/Kconfig | 8 ++ arch/cris/Kconfig | 8 ++ arch/frv/Kconfig | 8 ++ arch/h8300/Kconfig | 8 ++ arch/i386/Kconfig.cpu | 8 ++ arch/ia64/Kconfig | 8 ++ arch/m32r/Kconfig | 8 ++ arch/m68k/Kconfig | 8 ++ arch/m68knommu/Kconfig | 8 ++ arch/mips/Kconfig | 8 ++ arch/parisc/Kconfig | 8 ++ arch/powerpc/Kconfig | 8 ++ arch/ppc/Kconfig | 8 ++ arch/s390/Kconfig | 8 ++ arch/sh/Kconfig | 8 ++ arch/sh64/Kconfig | 8 ++ arch/sparc/Kconfig | 8 ++ arch/sparc64/Kconfig | 8 ++ arch/v850/Kconfig | 8 ++ arch/x86_64/Kconfig | 8 ++ arch/xtensa/Kconfig | 8 ++ drivers/infiniband/hw/mthca/mthca_provider.c | 4 +- drivers/infiniband/hw/mthca/mthca_qp.c | 4 +- drivers/infiniband/hw/mthca/mthca_srq.c | 4 +- drivers/infiniband/ulp/iser/iser_memory.c | 4 +- drivers/md/dm-crypt.c | 2 +- drivers/mmc/tifm_sd.c | 2 +- fs/ext2/super.c | 6 +- fs/ext3/super.c | 6 +- include/asm-frv/bitops.h | 44 +++++++++ include/linux/kernel.h | 9 +- include/linux/log2.h | 131 +++++++++++++++++++++++++++ mm/page_alloc.c | 4 +- 36 files changed, 384 insertions(+), 28 deletions(-) create mode 100644 include/linux/log2.h (limited to 'arch/i386') diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index 7e55ea66c6d..84caf50725b 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -25,6 +25,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 8c05d4321ae..aa1d400d721 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -74,6 +74,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/arm26/Kconfig b/arch/arm26/Kconfig index c14fe918bc4..74eba8b5a8c 100644 --- a/arch/arm26/Kconfig +++ b/arch/arm26/Kconfig @@ -41,6 +41,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig index 5f1694eea84..bb059a4e1df 100644 --- a/arch/avr32/Kconfig +++ b/arch/avr32/Kconfig @@ -45,6 +45,14 @@ config GENERIC_TIME config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_BUST_SPINLOCK bool diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig index 6a1238a29d6..3474309e049 100644 --- a/arch/cris/Kconfig +++ b/arch/cris/Kconfig @@ -16,6 +16,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index cf1c446e003..7561d7b72e7 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -41,6 +41,14 @@ config TIME_LOW_RES bool default y +config ARCH_HAS_ILOG2_U32 + bool + default y + +config ARCH_HAS_ILOG2_U64 + bool + default y + mainmenu "Fujitsu FR-V Kernel Configuration" source "init/Kconfig" diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig index cabf0bfffc5..34a84bc4baf 100644 --- a/arch/h8300/Kconfig +++ b/arch/h8300/Kconfig @@ -29,6 +29,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default n +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/i386/Kconfig.cpu b/arch/i386/Kconfig.cpu index 821fd269ca5..2aecfba4ac4 100644 --- a/arch/i386/Kconfig.cpu +++ b/arch/i386/Kconfig.cpu @@ -248,6 +248,14 @@ config RWSEM_XCHGADD_ALGORITHM depends on !M386 default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_CALIBRATE_DELAY bool default y diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 75d839715b2..fcacfe291b9 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -34,6 +34,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig index 41fd490af3b..f383dab973f 100644 --- a/arch/m32r/Kconfig +++ b/arch/m32r/Kconfig @@ -214,6 +214,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default n +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig index 7bc14461a6a..70a577c89c7 100644 --- a/arch/m68k/Kconfig +++ b/arch/m68k/Kconfig @@ -17,6 +17,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig index aa70dde5422..25993c2a8fb 100644 --- a/arch/m68knommu/Kconfig +++ b/arch/m68knommu/Kconfig @@ -25,6 +25,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default n +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index d8af858fe3f..57af8d8cf46 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -819,6 +819,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index d2101237442..0f9ff618c6d 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -25,6 +25,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 291c95ac4b3..97c7a212cdd 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -41,6 +41,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig index f76a146cff1..692b5ba5320 100644 --- a/arch/ppc/Kconfig +++ b/arch/ppc/Kconfig @@ -19,6 +19,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default y + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 583d9ff0a57..45e47bfb68a 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -22,6 +22,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index d83d64af31f..8e24c40662e 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -59,6 +59,14 @@ config LOCKDEP_SUPPORT bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + source "init/Kconfig" menu "System type" diff --git a/arch/sh64/Kconfig b/arch/sh64/Kconfig index 58c678e0666..7bc0744b7ab 100644 --- a/arch/sh64/Kconfig +++ b/arch/sh64/Kconfig @@ -39,6 +39,14 @@ config RWSEM_XCHGADD_ALGORITHM config GENERIC_ISA_DMA bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + source init/Kconfig menu "System type" diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 92a7c8a636d..d0dec1ea2ee 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -166,6 +166,14 @@ config ARCH_MAY_HAVE_PC_FDC bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config SUN_PM bool default y diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig index b627f8dbcaa..d391d11f245 100644 --- a/arch/sparc64/Kconfig +++ b/arch/sparc64/Kconfig @@ -34,6 +34,14 @@ config ARCH_MAY_HAVE_PC_FDC bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config AUDIT_ARCH bool default y diff --git a/arch/v850/Kconfig b/arch/v850/Kconfig index 37ec644603a..bcf825875d1 100644 --- a/arch/v850/Kconfig +++ b/arch/v850/Kconfig @@ -38,6 +38,14 @@ config TIME_LOW_RES bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + # Turn off some random 386 crap that can affect device config config ISA bool diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 3254a616c69..3ac581d1720 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -101,6 +101,14 @@ config GENERIC_BUG default y depends on BUG +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + source "init/Kconfig" diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig index c1e69a1f92a..9eccfbd1b53 100644 --- a/arch/xtensa/Kconfig +++ b/arch/xtensa/Kconfig @@ -34,6 +34,14 @@ config GENERIC_HARDIRQS bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + source "init/Kconfig" menu "Processor type and features" diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c index 21422a3336a..7ec7c4b937f 100644 --- a/drivers/infiniband/hw/mthca/mthca_provider.c +++ b/drivers/infiniband/hw/mthca/mthca_provider.c @@ -124,7 +124,7 @@ static int mthca_query_device(struct ib_device *ibdev, props->max_map_per_fmr = 255; else props->max_map_per_fmr = - (1 << (32 - long_log2(mdev->limits.num_mpts))) - 1; + (1 << (32 - ilog2(mdev->limits.num_mpts))) - 1; err = 0; out: @@ -816,7 +816,7 @@ static int mthca_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *uda lkey = ucmd.lkey; } - ret = mthca_RESIZE_CQ(dev, cq->cqn, lkey, long_log2(entries), &status); + ret = mthca_RESIZE_CQ(dev, cq->cqn, lkey, ilog2(entries), &status); if (status) ret = -EINVAL; diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index 33e3ba7937f..d844a2569b4 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -636,11 +636,11 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask, if (mthca_is_memfree(dev)) { if (qp->rq.max) - qp_context->rq_size_stride = long_log2(qp->rq.max) << 3; + qp_context->rq_size_stride = ilog2(qp->rq.max) << 3; qp_context->rq_size_stride |= qp->rq.wqe_shift - 4; if (qp->sq.max) - qp_context->sq_size_stride = long_log2(qp->sq.max) << 3; + qp_context->sq_size_stride = ilog2(qp->sq.max) << 3; qp_context->sq_size_stride |= qp->sq.wqe_shift - 4; } diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c index 34d2c476896..10684da33d5 100644 --- a/drivers/infiniband/hw/mthca/mthca_srq.c +++ b/drivers/infiniband/hw/mthca/mthca_srq.c @@ -120,7 +120,7 @@ static void mthca_arbel_init_srq_context(struct mthca_dev *dev, memset(context, 0, sizeof *context); - logsize = long_log2(srq->max); + logsize = ilog2(srq->max); context->state_logsize_srqn = cpu_to_be32(logsize << 24 | srq->srqn); context->lkey = cpu_to_be32(srq->mr.ibmr.lkey); context->db_index = cpu_to_be32(srq->db_index); @@ -213,7 +213,7 @@ int mthca_alloc_srq(struct mthca_dev *dev, struct mthca_pd *pd, if (!mthca_is_memfree(dev) && (ds > dev->limits.max_desc_sz)) return -EINVAL; - srq->wqe_shift = long_log2(ds); + srq->wqe_shift = ilog2(ds); srq->srqn = mthca_alloc(&dev->srq_table.alloc); if (srq->srqn == -1) diff --git a/drivers/infiniband/ulp/iser/iser_memory.c b/drivers/infiniband/ulp/iser/iser_memory.c index 5e122501fd8..3aedd59b8a8 100644 --- a/drivers/infiniband/ulp/iser/iser_memory.c +++ b/drivers/infiniband/ulp/iser/iser_memory.c @@ -114,7 +114,7 @@ int iser_start_rdma_unaligned_sg(struct iscsi_iser_cmd_task *iser_ctask, if (cmd_data_len > ISER_KMALLOC_THRESHOLD) mem = (void *)__get_free_pages(GFP_NOIO, - long_log2(roundup_pow_of_two(cmd_data_len)) - PAGE_SHIFT); + ilog2(roundup_pow_of_two(cmd_data_len)) - PAGE_SHIFT); else mem = kmalloc(cmd_data_len, GFP_NOIO); @@ -211,7 +211,7 @@ void iser_finalize_rdma_unaligned_sg(struct iscsi_iser_cmd_task *iser_ctask, if (cmd_data_len > ISER_KMALLOC_THRESHOLD) free_pages((unsigned long)mem_copy->copy_buf, - long_log2(roundup_pow_of_two(cmd_data_len)) - PAGE_SHIFT); + ilog2(roundup_pow_of_two(cmd_data_len)) - PAGE_SHIFT); else kfree(mem_copy->copy_buf); diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index a1086ee8ccc..96152868525 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -220,7 +220,7 @@ static int crypt_iv_benbi_ctr(struct crypt_config *cc, struct dm_target *ti, const char *opts) { unsigned int bs = crypto_blkcipher_blocksize(cc->tfm); - int log = long_log2(bs); + int log = ilog2(bs); /* we need to calculate how far we must shift the sector count * to get the cipher block count, we use this shift in _gen */ diff --git a/drivers/mmc/tifm_sd.c b/drivers/mmc/tifm_sd.c index e846499a004..f18ad998b3c 100644 --- a/drivers/mmc/tifm_sd.c +++ b/drivers/mmc/tifm_sd.c @@ -387,7 +387,7 @@ static void tifm_sd_prepare_data(struct tifm_sd *card, struct mmc_command *cmd) writel(TIFM_FIFO_INT_SETALL, sock->addr + SOCK_DMA_FIFO_INT_ENABLE_CLEAR); - writel(long_log2(cmd->data->blksz) - 2, + writel(ilog2(cmd->data->blksz) - 2, sock->addr + SOCK_FIFO_PAGE_SIZE); writel(TIFM_FIFO_ENABLE, sock->addr + SOCK_FIFO_CONTROL); writel(TIFM_FIFO_INTMASK, sock->addr + SOCK_DMA_FIFO_INT_ENABLE_SET); diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 255cef5f742..6347c2dbdd8 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -597,8 +597,6 @@ static int ext2_check_descriptors (struct super_block * sb) return 1; } -#define log2(n) ffz(~(n)) - /* * Maximal file size. There is a direct, and {,double-,triple-}indirect * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks. @@ -834,9 +832,9 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) sbi->s_sbh = bh; sbi->s_mount_state = le16_to_cpu(es->s_state); sbi->s_addr_per_block_bits = - log2 (EXT2_ADDR_PER_BLOCK(sb)); + ilog2 (EXT2_ADDR_PER_BLOCK(sb)); sbi->s_desc_per_block_bits = - log2 (EXT2_DESC_PER_BLOCK(sb)); + ilog2 (EXT2_DESC_PER_BLOCK(sb)); if (sb->s_magic != EXT2_SUPER_MAGIC) goto cantfind_ext2; diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 580b8a6ca97..b34886734a4 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1347,8 +1347,6 @@ static void ext3_orphan_cleanup (struct super_block * sb, sb->s_flags = s_flags; /* Restore MS_RDONLY status */ } -#define log2(n) ffz(~(n)) - /* * Maximal file size. There is a direct, and {,double-,triple-}indirect * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks. @@ -1597,8 +1595,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) sbi->s_desc_per_block = blocksize / sizeof(struct ext3_group_desc); sbi->s_sbh = bh; sbi->s_mount_state = le16_to_cpu(es->s_state); - sbi->s_addr_per_block_bits = log2(EXT3_ADDR_PER_BLOCK(sb)); - sbi->s_desc_per_block_bits = log2(EXT3_DESC_PER_BLOCK(sb)); + sbi->s_addr_per_block_bits = ilog2(EXT3_ADDR_PER_BLOCK(sb)); + sbi->s_desc_per_block_bits = ilog2(EXT3_DESC_PER_BLOCK(sb)); for (i=0; i < 4; i++) sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]); sbi->s_def_hash_version = es->s_def_hash_version; diff --git a/include/asm-frv/bitops.h b/include/asm-frv/bitops.h index 1f70d47148b..f8560edf59f 100644 --- a/include/asm-frv/bitops.h +++ b/include/asm-frv/bitops.h @@ -256,6 +256,50 @@ int __ffs(unsigned long x) return 31 - bit; } +/* + * special slimline version of fls() for calculating ilog2_u32() + * - note: no protection against n == 0 + */ +#define ARCH_HAS_ILOG2_U32 +static inline __attribute__((const)) +int __ilog2_u32(u32 n) +{ + int bit; + asm("scan %1,gr0,%0" : "=r"(bit) : "r"(n)); + return 31 - bit; +} + +/* + * special slimline version of fls64() for calculating ilog2_u64() + * - note: no protection against n == 0 + */ +#define ARCH_HAS_ILOG2_U64 +static inline __attribute__((const)) +int __ilog2_u64(u64 n) +{ + union { + u64 ll; + struct { u32 h, l; }; + } _; + int bit, x, y; + + _.ll = n; + + asm(" subcc %3,gr0,gr0,icc0 \n" + " ckeq icc0,cc4 \n" + " cscan.p %3,gr0,%0 ,cc4,0 \n" + " setlos #63,%1 \n" + " cscan.p %4,gr0,%0 ,cc4,1 \n" + " setlos #31,%2 \n" + " csub.p %1,%0,%0 ,cc4,0 \n" + " csub %2,%0,%0 ,cc4,1 \n" + : "=&r"(bit), "=r"(x), "=r"(y) + : "0r"(_.h), "r"(_.l) + : "icc0", "cc4" + ); + return bit; +} + #include #include diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 6738283ac38..3710cce1664 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -13,6 +13,7 @@ #include #include #include +#include #include #include @@ -157,14 +158,6 @@ static inline int printk(const char *s, ...) { return 0; } unsigned long int_sqrt(unsigned long); -static inline int __attribute_pure__ long_log2(unsigned long x) -{ - int r = 0; - for (x >>= 1; x > 0; x >>= 1) - r++; - return r; -} - static inline unsigned long __attribute_const__ roundup_pow_of_two(unsigned long x) { diff --git a/include/linux/log2.h b/include/linux/log2.h new file mode 100644 index 00000000000..3979c60325f --- /dev/null +++ b/include/linux/log2.h @@ -0,0 +1,131 @@ +/* Integer base 2 logarithm calculation + * + * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#ifndef _LINUX_LOG2_H +#define _LINUX_LOG2_H + +#include +#include + +/* + * deal with unrepresentable constant logarithms + */ +extern __attribute__((const, noreturn)) +int ____ilog2_NaN(void); + +/* + * non-constant log of base 2 calculators + * - the arch may override these in asm/bitops.h if they can be implemented + * more efficiently than using fls() and fls64() + * - the arch is not required to handle n==0 if implementing the fallback + */ +#ifndef CONFIG_ARCH_HAS_ILOG2_U32 +static inline __attribute__((const)) +int __ilog2_u32(u32 n) +{ + return fls(n) - 1; +} +#endif + +#ifndef CONFIG_ARCH_HAS_ILOG2_U64 +static inline __attribute__((const)) +int __ilog2_u64(u64 n) +{ + return fls64(n) - 1; +} +#endif + +/** + * ilog2 - log of base 2 of 32-bit or a 64-bit unsigned value + * @n - parameter + * + * constant-capable log of base 2 calculation + * - this can be used to initialise global variables from constant data, hence + * the massive ternary operator construction + * + * selects the appropriately-sized optimised version depending on sizeof(n) + */ +#define ilog2(n) \ +( \ + __builtin_constant_p(n) ? ( \ + (n) < 1 ? ____ilog2_NaN() : \ + (n) & (1ULL << 63) ? 63 : \ + (n) & (1ULL << 62) ? 62 : \ + (n) & (1ULL << 61) ? 61 : \ + (n) & (1ULL << 60) ? 60 : \ + (n) & (1ULL << 59) ? 59 : \ + (n) & (1ULL << 58) ? 58 : \ + (n) & (1ULL << 57) ? 57 : \ + (n) & (1ULL << 56) ? 56 : \ + (n) & (1ULL << 55) ? 55 : \ + (n) & (1ULL << 54) ? 54 : \ + (n) & (1ULL << 53) ? 53 : \ + (n) & (1ULL << 52) ? 52 : \ + (n) & (1ULL << 51) ? 51 : \ + (n) & (1ULL << 50) ? 50 : \ + (n) & (1ULL << 49) ? 49 : \ + (n) & (1ULL << 48) ? 48 : \ + (n) & (1ULL << 47) ? 47 : \ + (n) & (1ULL << 46) ? 46 : \ + (n) & (1ULL << 45) ? 45 : \ + (n) & (1ULL << 44) ? 44 : \ + (n) & (1ULL << 43) ? 43 : \ + (n) & (1ULL << 42) ? 42 : \ + (n) & (1ULL << 41) ? 41 : \ + (n) & (1ULL << 40) ? 40 : \ + (n) & (1ULL << 39) ? 39 : \ + (n) & (1ULL << 38) ? 38 : \ + (n) & (1ULL << 37) ? 37 : \ + (n) & (1ULL << 36) ? 36 : \ + (n) & (1ULL << 35) ? 35 : \ + (n) & (1ULL << 34) ? 34 : \ + (n) & (1ULL << 33) ? 33 : \ + (n) & (1ULL << 32) ? 32 : \ + (n) & (1ULL << 31) ? 31 : \ + (n) & (1ULL << 30) ? 30 : \ + (n) & (1ULL << 29) ? 29 : \ + (n) & (1ULL << 28) ? 28 : \ + (n) & (1ULL << 27) ? 27 : \ + (n) & (1ULL << 26) ? 26 : \ + (n) & (1ULL << 25) ? 25 : \ + (n) & (1ULL << 24) ? 24 : \ + (n) & (1ULL << 23) ? 23 : \ + (n) & (1ULL << 22) ? 22 : \ + (n) & (1ULL << 21) ? 21 : \ + (n) & (1ULL << 20) ? 20 : \ + (n) & (1ULL << 19) ? 19 : \ + (n) & (1ULL << 18) ? 18 : \ + (n) & (1ULL << 17) ? 17 : \ + (n) & (1ULL << 16) ? 16 : \ + (n) & (1ULL << 15) ? 15 : \ + (n) & (1ULL << 14) ? 14 : \ + (n) & (1ULL << 13) ? 13 : \ + (n) & (1ULL << 12) ? 12 : \ + (n) & (1ULL << 11) ? 11 : \ + (n) & (1ULL << 10) ? 10 : \ + (n) & (1ULL << 9) ? 9 : \ + (n) & (1ULL << 8) ? 8 : \ + (n) & (1ULL << 7) ? 7 : \ + (n) & (1ULL << 6) ? 6 : \ + (n) & (1ULL << 5) ? 5 : \ + (n) & (1ULL << 4) ? 4 : \ + (n) & (1ULL << 3) ? 3 : \ + (n) & (1ULL << 2) ? 2 : \ + (n) & (1ULL << 1) ? 1 : \ + (n) & (1ULL << 0) ? 0 : \ + ____ilog2_NaN() \ + ) : \ + (sizeof(n) <= 4) ? \ + __ilog2_u32(n) : \ + __ilog2_u64(n) \ + ) + +#endif /* _LINUX_LOG2_H */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index cace22b3ac2..18f0e044c43 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3244,7 +3244,7 @@ void *__init alloc_large_system_hash(const char *tablename, if (numentries > max) numentries = max; - log2qty = long_log2(numentries); + log2qty = ilog2(numentries); do { size = bucketsize << log2qty; @@ -3266,7 +3266,7 @@ void *__init alloc_large_system_hash(const char *tablename, printk("%s hash table entries: %d (order: %d, %lu bytes)\n", tablename, (1U << log2qty), - long_log2(size) - PAGE_SHIFT, + ilog2(size) - PAGE_SHIFT, size); if (_hash_shift) -- cgit v1.2.3-70-g09d2 From b5bd1cec8999624fa2b32833ba9707eb4966df17 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Fri, 8 Dec 2006 02:40:46 -0800 Subject: [PATCH] visws: sgivwfb as module needs exports With CONFIG_FB_SGIVW=m: WARNING: "sgivwfb_mem_size" [drivers/video/sgivwfb.ko] undefined! WARNING: "sgivwfb_mem_phys" [drivers/video/sgivwfb.ko] undefined! (or don't allow FB_SGIVW=m in Kconfig) Signed-off-by: Randy Dunlap Acked-by: James Simmons Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mach-visws/setup.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/mach-visws/setup.c b/arch/i386/mach-visws/setup.c index 885c7cbfd47..233ee20907b 100644 --- a/arch/i386/mach-visws/setup.c +++ b/arch/i386/mach-visws/setup.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include @@ -142,6 +143,8 @@ void __init time_init_hook(void) unsigned long sgivwfb_mem_phys; unsigned long sgivwfb_mem_size; +EXPORT_SYMBOL(sgivwfb_mem_phys); +EXPORT_SYMBOL(sgivwfb_mem_size); long long mem_size __initdata = 0; -- cgit v1.2.3-70-g09d2 From 3b1bdf4e08d6a8d4fae5a30224ed2c55bf1e43fc Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Fri, 8 Dec 2006 02:41:13 -0800 Subject: [PATCH] CPU hotplug broken with 2GB VMSPLIT In VMSPLIT mode, kernel PGD might have more entries than user space. Acked-by: Jens Axboe Signed-off-by: Shaohua Li Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/smpboot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 4bf0e3c83b8..1e00b03163b 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -1118,7 +1118,7 @@ static int __cpuinit __smp_prepare_cpu(int cpu) /* init low mem mapping */ clone_pgd_range(swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS, - KERNEL_PGD_PTRS); + min_t(unsigned long, KERNEL_PGD_PTRS, USER_PGD_PTRS)); flush_tlb_all(); schedule_work(&info.task); wait_for_completion(&done); -- cgit v1.2.3-70-g09d2 From e0f27981f2d5fd93b2795e8348327b081c512f83 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 9 Dec 2006 21:33:35 +0100 Subject: [PATCH] i386: Update defconfig Signed-off-by: Andi Kleen --- arch/i386/defconfig | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/defconfig b/arch/i386/defconfig index 65891f11ace..3265208e589 100644 --- a/arch/i386/defconfig +++ b/arch/i386/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-git7 -# Wed Dec 6 23:50:49 2006 +# Linux kernel version: 2.6.19-git14 +# Sat Dec 9 21:23:14 2006 # CONFIG_X86_32=y CONFIG_GENERIC_TIME=y @@ -12,6 +12,7 @@ CONFIG_X86=y CONFIG_MMU=y CONFIG_GENERIC_ISA_DMA=y CONFIG_GENERIC_IOMAP=y +CONFIG_GENERIC_BUG=y CONFIG_GENERIC_HWEIGHT=y CONFIG_ARCH_MAY_HAVE_PC_FDC=y CONFIG_DMI=y @@ -141,6 +142,8 @@ CONFIG_X86_CMPXCHG=y CONFIG_X86_XADD=y CONFIG_X86_L1_CACHE_SHIFT=7 CONFIG_RWSEM_XCHGADD_ALGORITHM=y +# CONFIG_ARCH_HAS_ILOG2_U32 is not set +# CONFIG_ARCH_HAS_ILOG2_U64 is not set CONFIG_GENERIC_CALIBRATE_DELAY=y CONFIG_X86_WP_WORKS_OK=y CONFIG_X86_INVLPG=y @@ -203,6 +206,7 @@ CONFIG_MTRR=y CONFIG_SECCOMP=y # CONFIG_HZ_100 is not set CONFIG_HZ_250=y +# CONFIG_HZ_300 is not set # CONFIG_HZ_1000 is not set CONFIG_HZ=250 # CONFIG_KEXEC is not set @@ -563,6 +567,7 @@ CONFIG_IDEDMA_AUTO=y # # CONFIG_RAID_ATTRS is not set CONFIG_SCSI=y +# CONFIG_SCSI_TGT is not set CONFIG_SCSI_NETLINK=y # CONFIG_SCSI_PROC_FS is not set @@ -583,6 +588,7 @@ CONFIG_CHR_DEV_SG=y # CONFIG_SCSI_MULTI_LUN is not set # CONFIG_SCSI_CONSTANTS is not set # CONFIG_SCSI_LOGGING is not set +# CONFIG_SCSI_SCAN_ASYNC is not set # # SCSI Transports @@ -642,6 +648,7 @@ CONFIG_AIC79XX_DEBUG_MASK=0 # CONFIG_SCSI_DC390T is not set # CONFIG_SCSI_NSP32 is not set # CONFIG_SCSI_DEBUG is not set +# CONFIG_SCSI_SRP is not set # # Serial ATA (prod) and Parallel ATA (experimental) drivers @@ -1082,10 +1089,7 @@ CONFIG_SOUND=y # Open Sound System # CONFIG_SOUND_PRIME=y -CONFIG_OSS_OBSOLETE_DRIVER=y # CONFIG_SOUND_BT878 is not set -# CONFIG_SOUND_EMU10K1 is not set -# CONFIG_SOUND_FUSION is not set # CONFIG_SOUND_ES1371 is not set CONFIG_SOUND_ICH=y # CONFIG_SOUND_TRIDENT is not set @@ -1094,6 +1098,11 @@ CONFIG_SOUND_ICH=y # CONFIG_SOUND_VIA82CXXX is not set # CONFIG_SOUND_OSS is not set +# +# HID Devices +# +CONFIG_HID=y + # # USB support # @@ -1158,8 +1167,7 @@ CONFIG_USB_STORAGE=y # USB Input Devices # CONFIG_USB_HID=y -CONFIG_USB_HIDINPUT=y -# CONFIG_USB_HIDINPUT_POWERBOOK is not set +# CONFIG_USB_HID_POWERBOOK is not set # CONFIG_HID_FF is not set # CONFIG_USB_HIDDEV is not set # CONFIG_USB_AIPTEK is not set @@ -1443,6 +1451,11 @@ CONFIG_NLS_ISO8859_15=y # CONFIG_NLS_KOI8_U is not set CONFIG_NLS_UTF8=y +# +# Distributed Lock Manager +# +# CONFIG_DLM is not set + # # Instrumentation Support # @@ -1509,6 +1522,7 @@ CONFIG_DOUBLEFAULT=y # # Library routines # +CONFIG_BITREVERSE=y # CONFIG_CRC_CCITT is not set # CONFIG_CRC16 is not set CONFIG_CRC32=y -- cgit v1.2.3-70-g09d2 From 16d279d277aedd640d9dba5ddeb172b5e6bc7d75 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 9 Dec 2006 21:33:35 +0100 Subject: [PATCH] x86: Fix verify_quirk_intel_irqbalance() Fix verify_quirk_intel_irqbalance(). genapic checks should really happen only on affected versions of the E7520/E7320/E7525 based platforms. AK: This should akpm's Coyote SDV Signed-off-by: Suresh Siddha Signed-off-by: Andi Kleen --- arch/i386/kernel/quirks.c | 33 +++++++++++++++++++++++++++++---- 1 file changed, 29 insertions(+), 4 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/quirks.c b/arch/i386/kernel/quirks.c index a01320a7b63..34874c398b4 100644 --- a/arch/i386/kernel/quirks.c +++ b/arch/i386/kernel/quirks.c @@ -10,13 +10,38 @@ #if defined(CONFIG_X86_IO_APIC) && defined(CONFIG_SMP) && defined(CONFIG_PCI) static void __devinit verify_quirk_intel_irqbalance(struct pci_dev *dev) { + u8 config, rev; + u32 word; + + /* BIOS may enable hardware IRQ balancing for + * E7520/E7320/E7525(revision ID 0x9 and below) + * based platforms. + * For those platforms, make sure that the genapic is set to 'flat' + */ + pci_read_config_byte(dev, PCI_CLASS_REVISION, &rev); + if (rev > 0x9) + return; + + /* enable access to config space*/ + pci_read_config_byte(dev, 0xf4, &config); + pci_write_config_byte(dev, 0xf4, config|0x2); + + /* read xTPR register */ + raw_pci_ops->read(0, 0, 0x40, 0x4c, 2, &word); + + if (!(word & (1 << 13))) { #ifdef CONFIG_X86_64 - if (genapic != &apic_flat) - panic("APIC mode must be flat on this system\n"); + if (genapic != &apic_flat) + panic("APIC mode must be flat on this system\n"); #elif defined(CONFIG_X86_GENERICARCH) - if (genapic != &apic_default) - panic("APIC mode must be default(flat) on this system. Use apic=default\n"); + if (genapic != &apic_default) + panic("APIC mode must be default(flat) on this system. Use apic=default\n"); #endif + } + + /* put back the original value for config space*/ + if (!(config & 0x2)) + pci_write_config_byte(dev, 0xf4, config); } void __init quirk_intel_irqbalance(void) -- cgit v1.2.3-70-g09d2 From 92715e282be7c7488f892703c8d39b08976a833b Mon Sep 17 00:00:00 2001 From: Ravikiran G Thirumalai Date: Sat, 9 Dec 2006 21:33:35 +0100 Subject: [PATCH] x86: Fix boot hang due to nmi watchdog init code 2.6.19 stopped booting (or booted based on build/config) on our x86_64 systems due to a bug introduced in 2.6.19. check_nmi_watchdog schedules an IPI on all cpus to busy wait on a flag, but fails to set the busywait flag if NMI functionality is disabled. This causes the secondary cpus to spin in an endless loop, causing the kernel bootup to hang. Depending upon the build, the busywait flag got overwritten (stack variable) and caused the kernel to bootup on certain builds. Following patch fixes the bug by setting the busywait flag before returning from check_nmi_watchdog. I guess using a stack variable is not good here as the calling function could potentially return while the busy wait loop is still spinning on the flag. AK: I redid the patch significantly to be cleaner Signed-off-by: Ravikiran Thirumalai Signed-off-by: Shai Fultheim Signed-off-by: Andi Kleen --- arch/i386/kernel/nmi.c | 8 ++++---- arch/x86_64/kernel/nmi.c | 9 +++++---- 2 files changed, 9 insertions(+), 8 deletions(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index f5bc7e1be80..a5e34d65596 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -195,6 +195,8 @@ static __cpuinit inline int nmi_known_cpu(void) return 0; } +static int endflag __initdata = 0; + #ifdef CONFIG_SMP /* The performance counters used by NMI_LOCAL_APIC don't trigger when * the CPU is idle. To make sure the NMI watchdog really ticks on all @@ -202,7 +204,6 @@ static __cpuinit inline int nmi_known_cpu(void) */ static __init void nmi_cpu_busy(void *data) { - volatile int *endflag = data; local_irq_enable_in_hardirq(); /* Intentionally don't use cpu_relax here. This is to make sure that the performance counter really ticks, @@ -210,14 +211,13 @@ static __init void nmi_cpu_busy(void *data) pause instruction. On a real HT machine this is fine because all other CPUs are busy with "useless" delay loops and don't care if they get somewhat less cycles. */ - while (*endflag == 0) - barrier(); + while (endflag == 0) + mb(); } #endif static int __init check_nmi_watchdog(void) { - volatile int endflag = 0; unsigned int *prev_nmi_count; int cpu; diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index 27e95e7922c..186aebbae32 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -193,6 +193,8 @@ void nmi_watchdog_default(void) nmi_watchdog = NMI_IO_APIC; } +static int endflag __initdata = 0; + #ifdef CONFIG_SMP /* The performance counters used by NMI_LOCAL_APIC don't trigger when * the CPU is idle. To make sure the NMI watchdog really ticks on all @@ -200,7 +202,6 @@ void nmi_watchdog_default(void) */ static __init void nmi_cpu_busy(void *data) { - volatile int *endflag = data; local_irq_enable_in_hardirq(); /* Intentionally don't use cpu_relax here. This is to make sure that the performance counter really ticks, @@ -208,14 +209,13 @@ static __init void nmi_cpu_busy(void *data) pause instruction. On a real HT machine this is fine because all other CPUs are busy with "useless" delay loops and don't care if they get somewhat less cycles. */ - while (*endflag == 0) - barrier(); + while (endflag == 0) + mb(); } #endif int __init check_nmi_watchdog (void) { - volatile int endflag = 0; int *counts; int cpu; @@ -256,6 +256,7 @@ int __init check_nmi_watchdog (void) if (!atomic_read(&nmi_active)) { kfree(counts); atomic_set(&nmi_active, -1); + endflag = 1; return -1; } endflag = 1; -- cgit v1.2.3-70-g09d2 From 1bac3b383a93a4a920ffc57441eb133c78567fbd Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 9 Dec 2006 21:33:36 +0100 Subject: [PATCH] x86: Work around gcc 4.2 over aggressive optimizer The new PDA code uses a dummy _proxy_pda variable to describe memory references to the PDA. It is never referenced in inline assembly, but exists as input/output arguments. gcc 4.2 in some cases can CSE references to this which causes unresolved symbols. Define it to zero to avoid this. Signed-off-by: Andi Kleen --- arch/i386/kernel/vmlinux.lds.S | 1 + arch/x86_64/kernel/vmlinux.lds.S | 1 + 2 files changed, 2 insertions(+) (limited to 'arch/i386') diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 16d3c7133ad..a53c8b1854b 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -26,6 +26,7 @@ OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386") OUTPUT_ARCH(i386) ENTRY(phys_startup_32) jiffies = jiffies_64; +_proxy_pda = 0; PHDRS { text PT_LOAD FLAGS(5); /* R_E */ diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index 6c417788c54..514be5dd230 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -13,6 +13,7 @@ OUTPUT_FORMAT("elf64-x86-64", "elf64-x86-64", "elf64-x86-64") OUTPUT_ARCH(i386:x86-64) ENTRY(phys_startup_64) jiffies_64 = jiffies; +_proxy_pda = 0; PHDRS { text PT_LOAD FLAGS(5); /* R_E */ data PT_LOAD FLAGS(7); /* RWE */ -- cgit v1.2.3-70-g09d2 From 7e74437cf60cc84a655e301f1ee48027b3bcf96e Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 9 Dec 2006 21:33:36 +0100 Subject: [PATCH] i386: export smp_num_siblings for oprofile oprofile uses smp_num_siblings without testing for CONFIG_X86_HT. I looked at modifying oprofile, but this way is cleaner & simpler and I didn't see a good reason not to just export it when CONFIG_SMP. WARNING: "smp_num_siblings" [arch/i386/oprofile/oprofile.ko] undefined! Signed-off-by: Randy Dunlap Signed-off-by: Andi Kleen --- arch/i386/kernel/smpboot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 1e00b03163b..b0f84e5778a 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -69,7 +69,7 @@ static int __devinitdata smp_b_stepping; /* Number of siblings per CPU package */ int smp_num_siblings = 1; -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP EXPORT_SYMBOL(smp_num_siblings); #endif -- cgit v1.2.3-70-g09d2 From 306a22c2a20edf3a440bfb4cf2b94c19dfd81a53 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 9 Dec 2006 21:33:36 +0100 Subject: [PATCH] i386: Fix io_apic.c warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit gcc 4.2 warns linux/arch/i386/kernel/io_apic.c: In function ‘create_irq’: linux/arch/i386/kernel/io_apic.c:2488: warning: ‘vector’ may be used uninitialized in this function The warning is false, but somewhat legitimate so work around it. Signed-off-by: Andi Kleen --- arch/i386/kernel/io_apic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/i386') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index e21dcde0790..2424cc9c7b3 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -2485,7 +2485,7 @@ device_initcall(ioapic_init_sysfs); int create_irq(void) { /* Allocate an unused irq */ - int irq, new, vector; + int irq, new, vector = 0; unsigned long flags; irq = -ENOSPC; -- cgit v1.2.3-70-g09d2 From f0f32fccbffaaba8596d5c671153aa37aea9d4f0 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 9 Dec 2006 21:33:36 +0100 Subject: [PATCH] x86-64: no paravirt for X86_VOYAGER or X86_VISWS Since Voyager and Visual WS already define ARCH_SETUP, it looks like PARAVIRT shouldn't be offered for them. In file included from arch/i386/kernel/setup.c:63: include/asm-i386/mach-visws/setup_arch.h:8:1: warning: "ARCH_SETUP" redefin= ed In file included from include/asm/msr.h:5, from include/asm/processor.h:17, from include/asm/thread_info.h:16, from include/linux/thread_info.h:21, from include/linux/preempt.h:9, from include/linux/spinlock.h:49, from include/linux/capability.h:45, from include/linux/sched.h:46, from arch/i386/kernel/setup.c:26: include/asm/paravirt.h:163:1: warning: this is the location of the previous= definition In file included from arch/i386/kernel/setup.c:63: include/asm-i386/mach-visws/setup_arch.h:8:1: warning: "ARCH_SETUP" redefin= ed In file included from include/asm/msr.h:5, from include/asm/processor.h:17, from include/asm/thread_info.h:16, from include/linux/thread_info.h:21, from include/linux/preempt.h:9, from include/linux/spinlock.h:49, from include/linux/capability.h:45, from include/linux/sched.h:46, from arch/i386/kernel/setup.c:26: include/asm/paravirt.h:163:1: warning: this is the location of the previous= definition Signed-off-by: Randy Dunlap --- arch/i386/Kconfig | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/i386') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index c2362c7ba74..0d67a0a1151 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -190,6 +190,7 @@ endchoice config PARAVIRT bool "Paravirtualization support (EXPERIMENTAL)" depends on EXPERIMENTAL + depends on !(X86_VISWS || X86_VOYAGER) help Paravirtualization is a way of running multiple instances of Linux on the same machine, under a hypervisor. This option -- cgit v1.2.3-70-g09d2