From 3f22ab276b931b72ea04b184c155b34d0362bfc3 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jun 2005 00:07:43 -0700 Subject: [PATCH] make each arch use mm/Kconfig For all architectures, this just means that you'll see a "Memory Model" choice in your architecture menu. For those that implement DISCONTIGMEM, you may eventually want to make your ARCH_DISCONTIGMEM_ENABLE a "def_bool y" and make your users select DISCONTIGMEM right out of the new choice menu. The only disadvantage might be if you have some specific things that you need in your help option to explain something about DISCONTIGMEM. Signed-off-by: Dave Hansen Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/Kconfig | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index ab79af84699..32696c1d928 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -226,6 +226,8 @@ config WARN_STACK_SIZE This allows you to specify the maximum frame size a function may have without the compiler complaining about it. +source "mm/Kconfig" + comment "I/O subsystem configuration" config MACHCHK_WARNING -- cgit v1.2.3-70-g09d2 From 0f8e2d62fa04441cd12c08ce521e84e5bd3f8a46 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Thu, 23 Jun 2005 00:08:10 -0700 Subject: [PATCH] use ${CROSS_COMPILE}installkernel in arch/*/boot/install.sh The attached patch causes the various arch specific install.sh scripts to look for ${CROSS_COMPILE}installkernel rather than just installkernel (in both /sbin/ and ~/bin/ where the script already did this). This allows you to have e.g. arm-linux-installkernel as a handy way to install on your cross target. It also prevents the script picking up on the host /sbin/installkernel which causes the script to fall through and do the install itself (which is what I actually use myself, with $INSTALL_PATH set). I don't believe it causes back-compatibility problems since calling the host installkernel was never likely to work or be what you wanted when cross compiling anyway. If $CROSS_COMPILE isn't set then nothing changes. I only use ARM and i386 myself but I figured it couldn't hurt to do the whole lot. I've cc'd those who I hope are the arch maintainers for files that I've touched. Signed-off-by: Ian Campbell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/boot/install.sh | 4 ++-- arch/arm26/boot/install.sh | 4 ++-- arch/i386/boot/install.sh | 4 ++-- arch/ppc64/boot/install.sh | 4 ++-- arch/s390/boot/install.sh | 4 ++-- arch/x86_64/boot/install.sh | 4 ++-- 6 files changed, 12 insertions(+), 12 deletions(-) (limited to 'arch/s390') diff --git a/arch/arm/boot/install.sh b/arch/arm/boot/install.sh index 935bb27369e..9f9bed20734 100644 --- a/arch/arm/boot/install.sh +++ b/arch/arm/boot/install.sh @@ -21,8 +21,8 @@ # # User may have a custom install script -if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi -if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi +if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi +if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi if [ "$(basename $2)" = "zImage" ]; then # Compressed install diff --git a/arch/arm26/boot/install.sh b/arch/arm26/boot/install.sh index c628328dd9e..8a8399b26cf 100644 --- a/arch/arm26/boot/install.sh +++ b/arch/arm26/boot/install.sh @@ -23,8 +23,8 @@ # User may have a custom install script -if [ -x /sbin/installkernel ]; then - exec /sbin/installkernel "$@" +if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then + exec /sbin/${CROSS_COMPILE}installkernel "$@" fi if [ "$2" = "zImage" ]; then diff --git a/arch/i386/boot/install.sh b/arch/i386/boot/install.sh index 90f2452b3b9..f17b40dfc0f 100644 --- a/arch/i386/boot/install.sh +++ b/arch/i386/boot/install.sh @@ -21,8 +21,8 @@ # User may have a custom install script -if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi -if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi +if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi +if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi # Default install - same as make zlilo diff --git a/arch/ppc64/boot/install.sh b/arch/ppc64/boot/install.sh index 955c5681db6..cb2d6626b55 100644 --- a/arch/ppc64/boot/install.sh +++ b/arch/ppc64/boot/install.sh @@ -22,8 +22,8 @@ # User may have a custom install script -if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi -if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi +if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi +if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi # Default install diff --git a/arch/s390/boot/install.sh b/arch/s390/boot/install.sh index 278a8139cb1..d4026f62cb0 100644 --- a/arch/s390/boot/install.sh +++ b/arch/s390/boot/install.sh @@ -21,8 +21,8 @@ # User may have a custom install script -if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi -if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi +if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi +if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi # Default install - same as make zlilo diff --git a/arch/x86_64/boot/install.sh b/arch/x86_64/boot/install.sh index 90f2452b3b9..f17b40dfc0f 100644 --- a/arch/x86_64/boot/install.sh +++ b/arch/x86_64/boot/install.sh @@ -21,8 +21,8 @@ # User may have a custom install script -if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi -if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi +if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi +if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi # Default install - same as make zlilo -- cgit v1.2.3-70-g09d2 From 0d77e5a2c23da734f5a7925f64afa1c2ed92e0f9 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Thu, 23 Jun 2005 00:10:14 -0700 Subject: [PATCH] compat: introduce compat_time_t This patch is based on work by Carlos O'Donell and Matthew Wilcox. It introduces/updates the compat_time_t type and uses it for compat siginfo structures. I have built this on ppc64 and x86_64. Signed-off-by: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/ia32/ia32priv.h | 2 +- arch/s390/kernel/compat_linux.h | 2 +- arch/sparc64/kernel/signal32.c | 2 +- include/asm-ia64/compat.h | 1 + include/asm-mips/compat.h | 1 + include/asm-parisc/compat.h | 2 +- include/asm-ppc64/compat.h | 1 + include/asm-ppc64/ppc32.h | 2 +- include/asm-sparc64/compat.h | 1 + include/asm-x86_64/ia32.h | 2 +- 10 files changed, 10 insertions(+), 6 deletions(-) (limited to 'arch/s390') diff --git a/arch/ia64/ia32/ia32priv.h b/arch/ia64/ia32/ia32priv.h index b2de948bdae..e3e9290e3ff 100644 --- a/arch/ia64/ia32/ia32priv.h +++ b/arch/ia64/ia32/ia32priv.h @@ -241,7 +241,7 @@ typedef struct compat_siginfo { /* POSIX.1b timers */ struct { - timer_t _tid; /* timer id */ + compat_timer_t _tid; /* timer id */ int _overrun; /* overrun count */ char _pad[sizeof(unsigned int) - sizeof(int)]; compat_sigval_t _sigval; /* same as below */ diff --git a/arch/s390/kernel/compat_linux.h b/arch/s390/kernel/compat_linux.h index bf33dcfec7d..3898f66d0b2 100644 --- a/arch/s390/kernel/compat_linux.h +++ b/arch/s390/kernel/compat_linux.h @@ -45,7 +45,7 @@ typedef struct compat_siginfo { /* POSIX.1b timers */ struct { - timer_t _tid; /* timer id */ + compat_timer_t _tid; /* timer id */ int _overrun; /* overrun count */ compat_sigval_t _sigval; /* same as below */ int _sys_private; /* not to be passed to user */ diff --git a/arch/sparc64/kernel/signal32.c b/arch/sparc64/kernel/signal32.c index 9a375e975cf..f28428f4170 100644 --- a/arch/sparc64/kernel/signal32.c +++ b/arch/sparc64/kernel/signal32.c @@ -102,7 +102,7 @@ typedef struct compat_siginfo{ /* POSIX.1b timers */ struct { - timer_t _tid; /* timer id */ + compat_timer_t _tid; /* timer id */ int _overrun; /* overrun count */ compat_sigval_t _sigval; /* same as below */ int _sys_private; /* not to be passed to user */ diff --git a/include/asm-ia64/compat.h b/include/asm-ia64/compat.h index cc0ff0a4bdd..0c05e5bad8a 100644 --- a/include/asm-ia64/compat.h +++ b/include/asm-ia64/compat.h @@ -27,6 +27,7 @@ typedef u16 compat_ipc_pid_t; typedef s32 compat_daddr_t; typedef u32 compat_caddr_t; typedef __kernel_fsid_t compat_fsid_t; +typedef s32 compat_timer_t; typedef s32 compat_int_t; typedef s32 compat_long_t; diff --git a/include/asm-mips/compat.h b/include/asm-mips/compat.h index dce92079e7f..d78002afb1e 100644 --- a/include/asm-mips/compat.h +++ b/include/asm-mips/compat.h @@ -29,6 +29,7 @@ typedef s32 compat_caddr_t; typedef struct { s32 val[2]; } compat_fsid_t; +typedef s32 compat_timer_t; typedef s32 compat_int_t; typedef s32 compat_long_t; diff --git a/include/asm-parisc/compat.h b/include/asm-parisc/compat.h index ca0eac647a0..7630d1ad239 100644 --- a/include/asm-parisc/compat.h +++ b/include/asm-parisc/compat.h @@ -24,7 +24,7 @@ typedef u16 compat_nlink_t; typedef u16 compat_ipc_pid_t; typedef s32 compat_daddr_t; typedef u32 compat_caddr_t; -typedef u32 compat_timer_t; +typedef s32 compat_timer_t; typedef s32 compat_int_t; typedef s32 compat_long_t; diff --git a/include/asm-ppc64/compat.h b/include/asm-ppc64/compat.h index 09c28d28ce6..12414f5fc66 100644 --- a/include/asm-ppc64/compat.h +++ b/include/asm-ppc64/compat.h @@ -26,6 +26,7 @@ typedef s32 compat_daddr_t; typedef u32 compat_caddr_t; typedef __kernel_fsid_t compat_fsid_t; typedef s32 compat_key_t; +typedef s32 compat_timer_t; typedef s32 compat_int_t; typedef s32 compat_long_t; diff --git a/include/asm-ppc64/ppc32.h b/include/asm-ppc64/ppc32.h index 1d040489755..6b44a8caf39 100644 --- a/include/asm-ppc64/ppc32.h +++ b/include/asm-ppc64/ppc32.h @@ -32,7 +32,7 @@ typedef struct compat_siginfo { /* POSIX.1b timers */ struct { - timer_t _tid; /* timer id */ + compat_timer_t _tid; /* timer id */ int _overrun; /* overrun count */ compat_sigval_t _sigval; /* same as below */ int _sys_private; /* not to be passed to user */ diff --git a/include/asm-sparc64/compat.h b/include/asm-sparc64/compat.h index 22f58055b8a..b59122dd176 100644 --- a/include/asm-sparc64/compat.h +++ b/include/asm-sparc64/compat.h @@ -25,6 +25,7 @@ typedef s32 compat_daddr_t; typedef u32 compat_caddr_t; typedef __kernel_fsid_t compat_fsid_t; typedef s32 compat_key_t; +typedef s32 compat_timer_t; typedef s32 compat_int_t; typedef s32 compat_long_t; diff --git a/include/asm-x86_64/ia32.h b/include/asm-x86_64/ia32.h index c0a7717923e..6efa00fe4e7 100644 --- a/include/asm-x86_64/ia32.h +++ b/include/asm-x86_64/ia32.h @@ -94,7 +94,7 @@ typedef struct compat_siginfo{ /* POSIX.1b timers */ struct { - int _tid; /* timer id */ + compat_timer_t _tid; /* timer id */ int _overrun; /* overrun count */ compat_sigval_t _sigval; /* same as below */ int _sys_private; /* not to be passed to user */ -- cgit v1.2.3-70-g09d2 From f370513640492641b4046bfd9a6e4714f6ae530d Mon Sep 17 00:00:00 2001 From: Zwane Mwaikambo Date: Sat, 25 Jun 2005 14:54:50 -0700 Subject: [PATCH] i386 CPU hotplug (The i386 CPU hotplug patch provides infrastructure for some work which Pavel is doing as well as for ACPI S3 (suspend-to-RAM) work which Li Shaohua is doing) The following provides i386 architecture support for safely unregistering and registering processors during runtime, updated for the current -mm tree. In order to avoid dumping cpu hotplug code into kernel/irq/* i dropped the cpu_online check in do_IRQ() by modifying fixup_irqs(). The difference being that on cpu offline, fixup_irqs() is called before we clear the cpu from cpu_online_map and a long delay in order to ensure that we never have any queued external interrupts on the APICs. There are additional changes to s390 and ppc64 to account for this change. 1) Add CONFIG_HOTPLUG_CPU 2) disable local APIC timer on dead cpus. 3) Disable preempt around irq balancing to prevent CPUs going down. 4) Print irq stats for all possible cpus. 5) Debugging check for interrupts on offline cpus. 6) Hacky fixup_irqs() to redirect irqs when cpus go off/online. 7) play_dead() for offline cpus to spin inside. 8) Handle offline cpus set in flush_tlb_others(). 9) Grab lock earlier in smp_call_function() to prevent CPUs going down. 10) Implement __cpu_disable() and __cpu_die(). 11) Enable local interrupts in cpu_enable() after fixup_irqs() 12) Don't fiddle with NMI on dead cpu, but leave intact on other cpus. 13) Program IRQ affinity whilst cpu is still in cpu_online_map on offline. Signed-off-by: Zwane Mwaikambo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/Kconfig | 9 ++++ arch/i386/kernel/apic.c | 3 +- arch/i386/kernel/io_apic.c | 2 + arch/i386/kernel/irq.c | 67 ++++++++++++++++++++++------ arch/i386/kernel/process.c | 39 +++++++++++++++- arch/i386/kernel/smp.c | 24 ++++++---- arch/i386/kernel/smpboot.c | 98 ++++++++++++++++++++++++++++++++++++++--- arch/i386/kernel/traps.c | 8 ++++ arch/ia64/kernel/smpboot.c | 1 + arch/ppc64/kernel/pSeries_smp.c | 5 ++- arch/s390/kernel/smp.c | 4 +- include/asm-i386/cpu.h | 2 + include/asm-i386/irq.h | 4 ++ include/asm-i386/smp.h | 3 ++ kernel/cpu.c | 14 +++--- 15 files changed, 243 insertions(+), 40 deletions(-) (limited to 'arch/s390') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index d4ae5f9ceae..b4cd11e5845 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -1250,6 +1250,15 @@ config SCx200 This support is also available as a module. If compiled as a module, it will be called scx200. +config HOTPLUG_CPU + bool "Support for hot-pluggable CPUs (EXPERIMENTAL)" + depends on SMP && HOTPLUG && EXPERIMENTAL + ---help--- + Say Y here to experiment with turning CPUs off and on. CPUs + can be controlled through /sys/devices/system/cpu. + + Say N. + source "drivers/pcmcia/Kconfig" source "drivers/pci/hotplug/Kconfig" diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c index 8d993fa7175..a28a088f3e7 100644 --- a/arch/i386/kernel/apic.c +++ b/arch/i386/kernel/apic.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include @@ -1048,7 +1049,7 @@ void __init setup_secondary_APIC_clock(void) setup_APIC_timer(calibration_result); } -void __init disable_APIC_timer(void) +void __devinit disable_APIC_timer(void) { if (using_apic_timer) { unsigned long v; diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 08540bc4ba3..3c2b3bdfc80 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -576,9 +576,11 @@ static int balanced_irq(void *unused) try_to_freeze(PF_FREEZE); if (time_after(jiffies, prev_balance_time+balanced_irq_interval)) { + preempt_disable(); do_irq_balance(); prev_balance_time = jiffies; time_remaining = balanced_irq_interval; + preempt_enable(); } } return 0; diff --git a/arch/i386/kernel/irq.c b/arch/i386/kernel/irq.c index 73945a3c53c..af115004aec 100644 --- a/arch/i386/kernel/irq.c +++ b/arch/i386/kernel/irq.c @@ -15,6 +15,9 @@ #include #include #include +#include +#include +#include DEFINE_PER_CPU(irq_cpustat_t, irq_stat) ____cacheline_maxaligned_in_smp; EXPORT_PER_CPU_SYMBOL(irq_stat); @@ -210,9 +213,8 @@ int show_interrupts(struct seq_file *p, void *v) if (i == 0) { seq_printf(p, " "); - for (j=0; jtypename); seq_printf(p, " %s", action->name); @@ -240,16 +241,14 @@ skip: spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) { seq_printf(p, "NMI: "); - for (j = 0; j < NR_CPUS; j++) - if (cpu_online(j)) - seq_printf(p, "%10u ", nmi_count(j)); + for_each_cpu(j) + seq_printf(p, "%10u ", nmi_count(j)); seq_putc(p, '\n'); #ifdef CONFIG_X86_LOCAL_APIC seq_printf(p, "LOC: "); - for (j = 0; j < NR_CPUS; j++) - if (cpu_online(j)) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).apic_timer_irqs); + for_each_cpu(j) + seq_printf(p, "%10u ", + per_cpu(irq_stat,j).apic_timer_irqs); seq_putc(p, '\n'); #endif seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); @@ -259,3 +258,45 @@ skip: } return 0; } + +#ifdef CONFIG_HOTPLUG_CPU +#include + +void fixup_irqs(cpumask_t map) +{ + unsigned int irq; + static int warned; + + for (irq = 0; irq < NR_IRQS; irq++) { + cpumask_t mask; + if (irq == 2) + continue; + + cpus_and(mask, irq_affinity[irq], map); + if (any_online_cpu(mask) == NR_CPUS) { + printk("Breaking affinity for irq %i\n", irq); + mask = map; + } + if (irq_desc[irq].handler->set_affinity) + irq_desc[irq].handler->set_affinity(irq, mask); + else if (irq_desc[irq].action && !(warned++)) + printk("Cannot set affinity for irq %i\n", irq); + } + +#if 0 + barrier(); + /* Ingo Molnar says: "after the IO-APIC masks have been redirected + [note the nop - the interrupt-enable boundary on x86 is two + instructions from sti] - to flush out pending hardirqs and + IPIs. After this point nothing is supposed to reach this CPU." */ + __asm__ __volatile__("sti; nop; cli"); + barrier(); +#else + /* That doesn't seem sufficient. Give it 1ms. */ + local_irq_enable(); + mdelay(1); + local_irq_disable(); +#endif +} +#endif + diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index aea2ce1145d..c1b11e8df60 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -13,6 +13,7 @@ #include +#include #include #include #include @@ -55,6 +56,9 @@ #include #include +#include +#include + asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); static int hlt_counter; @@ -143,14 +147,44 @@ static void poll_idle (void) } } +#ifdef CONFIG_HOTPLUG_CPU +#include +/* We don't actually take CPU down, just spin without interrupts. */ +static inline void play_dead(void) +{ + /* Ack it */ + __get_cpu_var(cpu_state) = CPU_DEAD; + + /* We shouldn't have to disable interrupts while dead, but + * some interrupts just don't seem to go away, and this makes + * it "work" for testing purposes. */ + /* Death loop */ + while (__get_cpu_var(cpu_state) != CPU_UP_PREPARE) + cpu_relax(); + + local_irq_disable(); + __flush_tlb_all(); + cpu_set(smp_processor_id(), cpu_online_map); + enable_APIC_timer(); + local_irq_enable(); +} +#else +static inline void play_dead(void) +{ + BUG(); +} +#endif /* CONFIG_HOTPLUG_CPU */ + /* * The idle thread. There's no useful work to be * done, so just try to conserve power and have a * low exit latency (ie sit in a loop waiting for * somebody to say that they'd like to reschedule) */ -void cpu_idle (void) +void cpu_idle(void) { + int cpu = raw_smp_processor_id(); + /* endless idle loop with no priority at all */ while (1) { while (!need_resched()) { @@ -165,6 +199,9 @@ void cpu_idle (void) if (!idle) idle = default_idle; + if (cpu_is_offline(cpu)) + play_dead(); + __get_cpu_var(irq_stat).idle_timestamp = jiffies; idle(); } diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c index 68be7d0c723..35f521612b2 100644 --- a/arch/i386/kernel/smp.c +++ b/arch/i386/kernel/smp.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -164,7 +165,7 @@ void send_IPI_mask_bitmask(cpumask_t cpumask, int vector) unsigned long flags; local_irq_save(flags); - + WARN_ON(mask & ~cpus_addr(cpu_online_map)[0]); /* * Wait for idle. */ @@ -346,21 +347,21 @@ out: static void flush_tlb_others(cpumask_t cpumask, struct mm_struct *mm, unsigned long va) { - cpumask_t tmp; /* * A couple of (to be removed) sanity checks: * - * - we do not send IPIs to not-yet booted CPUs. * - current CPU must not be in mask * - mask must exist :) */ BUG_ON(cpus_empty(cpumask)); - - cpus_and(tmp, cpumask, cpu_online_map); - BUG_ON(!cpus_equal(cpumask, tmp)); BUG_ON(cpu_isset(smp_processor_id(), cpumask)); BUG_ON(!mm); + /* If a CPU which we ran on has gone down, OK. */ + cpus_and(cpumask, cpumask, cpu_online_map); + if (cpus_empty(cpumask)) + return; + /* * i'm not happy about this global shared spinlock in the * MM hot path, but we'll see how contended it is. @@ -476,6 +477,7 @@ void flush_tlb_all(void) */ void smp_send_reschedule(int cpu) { + WARN_ON(cpu_is_offline(cpu)); send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR); } @@ -516,10 +518,15 @@ int smp_call_function (void (*func) (void *info), void *info, int nonatomic, */ { struct call_data_struct data; - int cpus = num_online_cpus()-1; + int cpus; - if (!cpus) + /* Holding any lock stops cpus from going down. */ + spin_lock(&call_lock); + cpus = num_online_cpus() - 1; + if (!cpus) { + spin_unlock(&call_lock); return 0; + } /* Can deadlock when called with interrupts disabled */ WARN_ON(irqs_disabled()); @@ -531,7 +538,6 @@ int smp_call_function (void (*func) (void *info), void *info, int nonatomic, if (wait) atomic_set(&data.finished, 0); - spin_lock(&call_lock); call_data = &data; mb(); diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index c20d96d5c15..ad74a46e9ef 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -44,6 +44,9 @@ #include #include #include +#include +#include +#include #include #include @@ -96,6 +99,9 @@ static int trampoline_exec; static void map_cpu_to_logical_apicid(void); +/* State of each CPU. */ +DEFINE_PER_CPU(int, cpu_state) = { 0 }; + /* * Currently trivial. Write the real->protected mode * bootstrap into the page concerned. The caller @@ -1119,6 +1125,9 @@ static void __init smp_boot_cpus(unsigned int max_cpus) who understands all this stuff should rewrite it properly. --RR 15/Jul/02 */ void __init smp_prepare_cpus(unsigned int max_cpus) { + smp_commenced_mask = cpumask_of_cpu(0); + cpu_callin_map = cpumask_of_cpu(0); + mb(); smp_boot_cpus(max_cpus); } @@ -1128,20 +1137,99 @@ void __devinit smp_prepare_boot_cpu(void) cpu_set(smp_processor_id(), cpu_callout_map); } -int __devinit __cpu_up(unsigned int cpu) +#ifdef CONFIG_HOTPLUG_CPU + +/* must be called with the cpucontrol mutex held */ +static int __devinit cpu_enable(unsigned int cpu) { - /* This only works at boot for x86. See "rewrite" above. */ - if (cpu_isset(cpu, smp_commenced_mask)) { - local_irq_enable(); - return -ENOSYS; + /* get the target out of its holding state */ + per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; + wmb(); + + /* wait for the processor to ack it. timeout? */ + while (!cpu_online(cpu)) + cpu_relax(); + + fixup_irqs(cpu_online_map); + /* counter the disable in fixup_irqs() */ + local_irq_enable(); + return 0; +} + +int __cpu_disable(void) +{ + cpumask_t map = cpu_online_map; + int cpu = smp_processor_id(); + + /* + * Perhaps use cpufreq to drop frequency, but that could go + * into generic code. + * + * We won't take down the boot processor on i386 due to some + * interrupts only being able to be serviced by the BSP. + * Especially so if we're not using an IOAPIC -zwane + */ + if (cpu == 0) + return -EBUSY; + + /* We enable the timer again on the exit path of the death loop */ + disable_APIC_timer(); + /* Allow any queued timer interrupts to get serviced */ + local_irq_enable(); + mdelay(1); + local_irq_disable(); + + cpu_clear(cpu, map); + fixup_irqs(map); + /* It's now safe to remove this processor from the online map */ + cpu_clear(cpu, cpu_online_map); + return 0; +} + +void __cpu_die(unsigned int cpu) +{ + /* We don't do anything here: idle task is faking death itself. */ + unsigned int i; + + for (i = 0; i < 10; i++) { + /* They ack this in play_dead by setting CPU_DEAD */ + if (per_cpu(cpu_state, cpu) == CPU_DEAD) + return; + current->state = TASK_UNINTERRUPTIBLE; + schedule_timeout(HZ/10); } + printk(KERN_ERR "CPU %u didn't die...\n", cpu); +} +#else /* ... !CONFIG_HOTPLUG_CPU */ +int __cpu_disable(void) +{ + return -ENOSYS; +} +void __cpu_die(unsigned int cpu) +{ + /* We said "no" in __cpu_disable */ + BUG(); +} +#endif /* CONFIG_HOTPLUG_CPU */ + +int __devinit __cpu_up(unsigned int cpu) +{ /* In case one didn't come up */ if (!cpu_isset(cpu, cpu_callin_map)) { + printk(KERN_DEBUG "skipping cpu%d, didn't come online\n", cpu); local_irq_enable(); return -EIO; } +#ifdef CONFIG_HOTPLUG_CPU + /* Already up, and in cpu_quiescent now? */ + if (cpu_isset(cpu, smp_commenced_mask)) { + cpu_enable(cpu); + return 0; + } +#endif + local_irq_enable(); /* Unleash the CPU! */ cpu_set(cpu, smp_commenced_mask); diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index e4d4e2162c7..207ea8ba716 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -625,6 +625,14 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code) nmi_enter(); cpu = smp_processor_id(); + +#ifdef CONFIG_HOTPLUG_CPU + if (!cpu_online(cpu)) { + nmi_exit(); + return; + } +#endif + ++nmi_count(cpu); if (!nmi_callback(regs, cpu)) diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index 3865f088ffa..a888ddc10f7 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -688,6 +688,7 @@ int __cpu_disable(void) return -EBUSY; remove_siblinginfo(cpu); + cpu_clear(cpu, cpu_online_map); fixup_irqs(); local_flush_tlb_all(); cpu_clear(cpu, cpu_callin_map); diff --git a/arch/ppc64/kernel/pSeries_smp.c b/arch/ppc64/kernel/pSeries_smp.c index 30154140f7e..62c55a12356 100644 --- a/arch/ppc64/kernel/pSeries_smp.c +++ b/arch/ppc64/kernel/pSeries_smp.c @@ -93,10 +93,13 @@ static int query_cpu_stopped(unsigned int pcpu) int pSeries_cpu_disable(void) { + int cpu = smp_processor_id(); + + cpu_clear(cpu, cpu_online_map); systemcfg->processorCount--; /*fix boot_cpuid here*/ - if (smp_processor_id() == boot_cpuid) + if (cpu == boot_cpuid) boot_cpuid = any_online_cpu(cpu_online_map); /* FIXME: abstract this to not be platform specific later on */ diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index fdfcf0488b4..93c71fef99d 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -679,12 +679,14 @@ __cpu_disable(void) { unsigned long flags; ec_creg_mask_parms cr_parms; + int cpu = smp_processor_id(); spin_lock_irqsave(&smp_reserve_lock, flags); - if (smp_cpu_reserved[smp_processor_id()] != 0) { + if (smp_cpu_reserved[cpu] != 0) { spin_unlock_irqrestore(&smp_reserve_lock, flags); return -EBUSY; } + cpu_clear(cpu, cpu_online_map); #ifdef CONFIG_PFAULT /* Disable pfault pseudo page faults on this cpu. */ diff --git a/include/asm-i386/cpu.h b/include/asm-i386/cpu.h index 002740b2195..e7252c216ca 100644 --- a/include/asm-i386/cpu.h +++ b/include/asm-i386/cpu.h @@ -5,6 +5,7 @@ #include #include #include +#include #include @@ -16,4 +17,5 @@ extern int arch_register_cpu(int num); extern void arch_unregister_cpu(int); #endif +DECLARE_PER_CPU(int, cpu_state); #endif /* _ASM_I386_CPU_H_ */ diff --git a/include/asm-i386/irq.h b/include/asm-i386/irq.h index 05b9e61b0a7..e2d8bf23ad7 100644 --- a/include/asm-i386/irq.h +++ b/include/asm-i386/irq.h @@ -38,4 +38,8 @@ extern void release_vm86_irqs(struct task_struct *); extern int irqbalance_disable(char *str); #endif +#ifdef CONFIG_HOTPLUG_CPU +extern void fixup_irqs(cpumask_t map); +#endif + #endif /* _ASM_IRQ_H */ diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h index 55ef31f66bb..507f2fd39a6 100644 --- a/include/asm-i386/smp.h +++ b/include/asm-i386/smp.h @@ -83,6 +83,9 @@ static __inline int logical_smp_processor_id(void) } #endif + +extern int __cpu_disable(void); +extern void __cpu_die(unsigned int cpu); #endif /* !__ASSEMBLY__ */ #define NO_PROC_ID 0xFF /* No processor magic marker */ diff --git a/kernel/cpu.c b/kernel/cpu.c index 628f4ccda12..53d8263ae12 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -63,19 +63,15 @@ static int take_cpu_down(void *unused) { int err; - /* Take offline: makes arch_cpu_down somewhat easier. */ - cpu_clear(smp_processor_id(), cpu_online_map); - /* Ensure this CPU doesn't handle any more interrupts. */ err = __cpu_disable(); if (err < 0) - cpu_set(smp_processor_id(), cpu_online_map); - else - /* Force idle task to run as soon as we yield: it should - immediately notice cpu is offline and die quickly. */ - sched_idle_next(); + return err; - return err; + /* Force idle task to run as soon as we yield: it should + immediately notice cpu is offline and die quickly. */ + sched_idle_next(); + return 0; } int cpu_down(unsigned int cpu) -- cgit v1.2.3-70-g09d2 From 77fa22450de00d535de2cc8be653983560828000 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Sat, 25 Jun 2005 14:55:30 -0700 Subject: [PATCH] s390: improved machine check handling Improved machine check handling. Kernel is now able to receive machine checks while in kernel mode (system call, interrupt and program check handling). Also register validation is now performed. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/entry.S | 102 +++++++++++-- arch/s390/kernel/entry64.S | 97 +++++++++++-- arch/s390/kernel/process.c | 46 ++---- arch/s390/kernel/setup.c | 13 +- arch/s390/kernel/smp.c | 13 +- drivers/s390/s390mach.c | 321 ++++++++++++++++++++++++++++++++++++----- drivers/s390/s390mach.h | 35 ++++- include/asm-s390/lowcore.h | 7 +- include/asm-s390/processor.h | 52 +++---- include/asm-s390/ptrace.h | 2 +- include/asm-s390/system.h | 21 ++- include/asm-s390/thread_info.h | 2 + 12 files changed, 576 insertions(+), 135 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index c0e09b33feb..5b262b5d869 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -7,6 +7,7 @@ * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), * Hartmut Penner (hp@de.ibm.com), * Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com), + * Heiko Carstens */ #include @@ -49,9 +50,9 @@ SP_ILC = STACK_FRAME_OVERHEAD + __PT_ILC SP_TRAP = STACK_FRAME_OVERHEAD + __PT_TRAP SP_SIZE = STACK_FRAME_OVERHEAD + __PT_SIZE -_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | \ +_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING | \ _TIF_RESTART_SVC | _TIF_SINGLE_STEP ) -_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED) +_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING) STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER STACK_SIZE = 1 << STACK_SHIFT @@ -121,7 +122,11 @@ STACK_SIZE = 1 << STACK_SHIFT bz BASED(stack_overflow) 3: #endif -2: s %r15,BASED(.Lc_spsize) # make room for registers & psw +2: + .endm + + .macro CREATE_STACK_FRAME psworg,savearea + s %r15,BASED(.Lc_spsize) # make room for registers & psw mvc SP_PSW(8,%r15),0(%r12) # move user PSW to stack la %r12,\psworg st %r2,SP_ORIG_R2(%r15) # store original content of gpr 2 @@ -161,6 +166,13 @@ __switch_to_base: be __switch_to_noper-__switch_to_base(%r1) # we got away w/o bashing TLB's lctl %c9,%c11,__THREAD_per(%r3) # Nope we didn't __switch_to_noper: + l %r4,__THREAD_info(%r2) # get thread_info of prev + tm __TI_flags+3(%r4),_TIF_MCCK_PENDING # machine check pending? + bz __switch_to_no_mcck-__switch_to_base(%r1) + ni __TI_flags+3(%r4),255-_TIF_MCCK_PENDING # clear flag in prev + l %r4,__THREAD_info(%r3) # get thread_info of next + oi __TI_flags+3(%r4),_TIF_MCCK_PENDING # set it in next +__switch_to_no_mcck: stm %r6,%r15,__SF_GPRS(%r15)# store __switch_to registers of prev task st %r15,__THREAD_ksp(%r2) # store kernel stack to prev->tss.ksp l %r15,__THREAD_ksp(%r3) # load kernel stack from next->tss.ksp @@ -185,6 +197,7 @@ system_call: sysc_saveall: SAVE_ALL_BASE __LC_SAVE_AREA SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA lh %r7,0x8a # get svc number from lowcore #ifdef CONFIG_VIRT_CPU_ACCOUNTING sysc_vtime: @@ -234,6 +247,8 @@ sysc_work_loop: # One of the work bits is on. Find out which one. # sysc_work: + tm __TI_flags+3(%r9),_TIF_MCCK_PENDING + bo BASED(sysc_mcck_pending) tm __TI_flags+3(%r9),_TIF_NEED_RESCHED bo BASED(sysc_reschedule) tm __TI_flags+3(%r9),_TIF_SIGPENDING @@ -252,6 +267,14 @@ sysc_reschedule: la %r14,BASED(sysc_work_loop) br %r1 # call scheduler +# +# _TIF_MCCK_PENDING is set, call handler +# +sysc_mcck_pending: + l %r1,BASED(.Ls390_handle_mcck) + la %r14,BASED(sysc_work_loop) + br %r1 # TIF bit will be cleared by handler + # # _TIF_SIGPENDING is set, call do_signal # @@ -430,6 +453,7 @@ pgm_check_handler: tm __LC_PGM_INT_CODE+1,0x80 # check whether we got a per exception bnz BASED(pgm_per) # got per exception -> special case SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(pgm_no_vtime) @@ -468,6 +492,7 @@ pgm_per: # pgm_per_std: SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(pgm_no_vtime2) @@ -493,6 +518,7 @@ pgm_no_vtime2: # pgm_svcper: SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(pgm_no_vtime3) @@ -521,6 +547,7 @@ io_int_handler: stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+16 SAVE_ALL __LC_IO_OLD_PSW,__LC_SAVE_AREA+16,0 + CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+16 #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(io_no_vtime) @@ -578,15 +605,25 @@ io_work: lr %r15,%r1 # # One of the work bits is on. Find out which one. -# Checked are: _TIF_SIGPENDING and _TIF_NEED_RESCHED +# Checked are: _TIF_SIGPENDING, _TIF_NEED_RESCHED and _TIF_MCCK_PENDING # io_work_loop: + tm __TI_flags+3(%r9),_TIF_MCCK_PENDING + bo BASED(io_mcck_pending) tm __TI_flags+3(%r9),_TIF_NEED_RESCHED bo BASED(io_reschedule) tm __TI_flags+3(%r9),_TIF_SIGPENDING bo BASED(io_sigpending) b BASED(io_leave) +# +# _TIF_MCCK_PENDING is set, call handler +# +io_mcck_pending: + l %r1,BASED(.Ls390_handle_mcck) + l %r14,BASED(io_work_loop) + br %r1 # TIF bit will be cleared by handler + # # _TIF_NEED_RESCHED is set, call schedule # @@ -621,6 +658,7 @@ ext_int_handler: stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+16 SAVE_ALL __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16,0 + CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16 #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(ext_no_vtime) @@ -642,19 +680,62 @@ ext_no_vtime: .globl mcck_int_handler mcck_int_handler: - STORE_TIMER __LC_ASYNC_ENTER_TIMER + spt __LC_CPU_TIMER_SAVE_AREA # revalidate cpu timer + lm %r0,%r15,__LC_GPREGS_SAVE_AREA # revalidate gprs SAVE_ALL_BASE __LC_SAVE_AREA+32 - SAVE_ALL __LC_MCK_OLD_PSW,__LC_SAVE_AREA+32,0 + la %r12,__LC_MCK_OLD_PSW + tm __LC_MCCK_CODE,0x80 # system damage? + bo BASED(mcck_int_main) # yes -> rest of mcck code invalid + tm __LC_MCCK_CODE+5,0x02 # stored cpu timer value valid? + bo BASED(0f) + spt __LC_LAST_UPDATE_TIMER # revalidate cpu timer #ifdef CONFIG_VIRT_CPU_ACCOUNTING - tm SP_PSW+1(%r15),0x01 # interrupting from user ? + mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER + mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER + mvc __LC_LAST_UPDATE_TIMER(8),__LC_EXIT_TIMER +0: tm __LC_MCCK_CODE+2,0x08 # mwp of old psw valid? + bno BASED(mcck_no_vtime) # no -> skip cleanup critical + tm __LC_MCK_OLD_PSW+1,0x01 # interrupting from user ? bz BASED(mcck_no_vtime) UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER mcck_no_vtime: #endif +0: + tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? + bno BASED(mcck_int_main) # no -> skip cleanup critical + tm __LC_MCK_OLD_PSW+1,0x01 # test problem state bit + bnz BASED(mcck_int_main) # from user -> load async stack + clc __LC_MCK_OLD_PSW+4(4),BASED(.Lcritical_end) + bhe BASED(mcck_int_main) + clc __LC_MCK_OLD_PSW+4(4),BASED(.Lcritical_start) + bl BASED(mcck_int_main) + l %r14,BASED(.Lcleanup_critical) + basr %r14,%r14 +mcck_int_main: + l %r14,__LC_PANIC_STACK # are we already on the panic stack? + slr %r14,%r15 + sra %r14,PAGE_SHIFT + be BASED(0f) + l %r15,__LC_PANIC_STACK # load panic stack +0: CREATE_STACK_FRAME __LC_MCK_OLD_PSW,__LC_SAVE_AREA+32 + l %r9,__LC_THREAD_INFO # load pointer to thread_info struct + la %r2,SP_PTREGS(%r15) # load pt_regs l %r1,BASED(.Ls390_mcck) - basr %r14,%r1 # call machine check handler + basr %r14,%r1 # call machine check handler + tm SP_PSW+1(%r15),0x01 # returning to user ? + bno BASED(mcck_return) + l %r1,__LC_KERNEL_STACK # switch to kernel stack + s %r1,BASED(.Lc_spsize) + mvc SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15) + xc __SF_BACKCHAIN(4,%r1),__SF_BACKCHAIN(%r1) # clear back chain + lr %r15,%r1 + stosm __SF_EMPTY(%r15),0x04 # turn dat on + tm __TI_flags+3(%r9),_TIF_MCCK_PENDING + bno BASED(mcck_return) + l %r1,BASED(.Ls390_handle_mcck) + basr %r14,%r1 # call machine check handler mcck_return: RESTORE_ALL 0 @@ -742,7 +823,7 @@ cleanup_critical: clc 4(4,%r12),BASED(cleanup_table_sysc_work_loop) bl BASED(0f) clc 4(4,%r12),BASED(cleanup_table_sysc_work_loop+4) - bl BASED(cleanup_sysc_leave) + bl BASED(cleanup_sysc_return) 0: br %r14 @@ -760,6 +841,7 @@ cleanup_system_call: mvc __LC_SAVE_AREA(16),__LC_SAVE_AREA+16 0: st %r13,__LC_SAVE_AREA+20 SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA st %r15,__LC_SAVE_AREA+28 lh %r7,0x8a #ifdef CONFIG_VIRT_CPU_ACCOUNTING @@ -834,6 +916,8 @@ cleanup_sysc_leave_insn: * Symbol constants */ .Ls390_mcck: .long s390_do_machine_check +.Ls390_handle_mcck: + .long s390_handle_mcck .Ldo_IRQ: .long do_IRQ .Ldo_extint: .long do_extint .Ldo_signal: .long do_signal diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index 51527ab8c8f..57ca75d0ad7 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -7,6 +7,7 @@ * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), * Hartmut Penner (hp@de.ibm.com), * Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com), + * Heiko Carstens */ #include @@ -52,9 +53,9 @@ SP_SIZE = STACK_FRAME_OVERHEAD + __PT_SIZE STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER STACK_SIZE = 1 << STACK_SHIFT -_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | \ +_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING | \ _TIF_RESTART_SVC | _TIF_SINGLE_STEP ) -_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED) +_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING) #define BASED(name) name-system_call(%r13) @@ -114,7 +115,11 @@ _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED) jz stack_overflow 3: #endif -2: aghi %r15,-SP_SIZE # make room for registers & psw +2: + .endm + + .macro CREATE_STACK_FRAME psworg,savearea + aghi %r15,-SP_SIZE # make room for registers & psw mvc SP_PSW(16,%r15),0(%r12) # move user PSW to stack la %r12,\psworg stg %r2,SP_ORIG_R2(%r15) # store original content of gpr 2 @@ -152,6 +157,13 @@ __switch_to: je __switch_to_noper # we got away without bashing TLB's lctlg %c9,%c11,__THREAD_per(%r3) # Nope we didn't __switch_to_noper: + lg %r4,__THREAD_info(%r2) # get thread_info of prev + tm __TI_flags+7(%r4),_TIF_MCCK_PENDING # machine check pending? + jz __switch_to_no_mcck + ni __TI_flags+7(%r4),255-_TIF_MCCK_PENDING # clear flag in prev + lg %r4,__THREAD_info(%r3) # get thread_info of next + oi __TI_flags+7(%r4),_TIF_MCCK_PENDING # set it in next +__switch_to_no_mcck: stmg %r6,%r15,__SF_GPRS(%r15)# store __switch_to registers of prev task stg %r15,__THREAD_ksp(%r2) # store kernel stack to prev->tss.ksp lg %r15,__THREAD_ksp(%r3) # load kernel stack from next->tss.ksp @@ -176,6 +188,7 @@ system_call: sysc_saveall: SAVE_ALL_BASE __LC_SAVE_AREA SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA llgh %r7,__LC_SVC_INT_CODE # get svc number from lowcore #ifdef CONFIG_VIRT_CPU_ACCOUNTING sysc_vtime: @@ -232,6 +245,8 @@ sysc_work_loop: # One of the work bits is on. Find out which one. # sysc_work: + tm __TI_flags+7(%r9),_TIF_MCCK_PENDING + jo sysc_mcck_pending tm __TI_flags+7(%r9),_TIF_NEED_RESCHED jo sysc_reschedule tm __TI_flags+7(%r9),_TIF_SIGPENDING @@ -249,6 +264,13 @@ sysc_reschedule: larl %r14,sysc_work_loop jg schedule # return point is sysc_return +# +# _TIF_MCCK_PENDING is set, call handler +# +sysc_mcck_pending: + larl %r14,sysc_work_loop + jg s390_handle_mcck # TIF bit will be cleared by handler + # # _TIF_SIGPENDING is set, call do_signal # @@ -474,6 +496,7 @@ pgm_check_handler: tm __LC_PGM_INT_CODE+1,0x80 # check whether we got a per exception jnz pgm_per # got per exception -> special case SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz pgm_no_vtime @@ -512,6 +535,7 @@ pgm_per: # pgm_per_std: SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz pgm_no_vtime2 @@ -537,6 +561,7 @@ pgm_no_vtime2: # pgm_svcper: SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz pgm_no_vtime3 @@ -564,6 +589,7 @@ io_int_handler: stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+32 SAVE_ALL __LC_IO_OLD_PSW,__LC_SAVE_AREA+32,0 + CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+32 #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz io_no_vtime @@ -621,15 +647,24 @@ io_work: lgr %r15,%r1 # # One of the work bits is on. Find out which one. -# Checked are: _TIF_SIGPENDING and _TIF_NEED_RESCHED +# Checked are: _TIF_SIGPENDING, _TIF_NEED_RESCHED and _TIF_MCCK_PENDING # io_work_loop: + tm __TI_flags+7(%r9),_TIF_MCCK_PENDING + jo io_mcck_pending tm __TI_flags+7(%r9),_TIF_NEED_RESCHED jo io_reschedule tm __TI_flags+7(%r9),_TIF_SIGPENDING jo io_sigpending j io_leave +# +# _TIF_MCCK_PENDING is set, call handler +# +io_mcck_pending: + larl %r14,io_work_loop + jg s390_handle_mcck # TIF bit will be cleared by handler + # # _TIF_NEED_RESCHED is set, call schedule # @@ -661,6 +696,7 @@ ext_int_handler: stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+32 SAVE_ALL __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32,0 + CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32 #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz ext_no_vtime @@ -680,18 +716,60 @@ ext_no_vtime: */ .globl mcck_int_handler mcck_int_handler: - STORE_TIMER __LC_ASYNC_ENTER_TIMER + la %r1,4095 # revalidate r1 + spt __LC_CPU_TIMER_SAVE_AREA-4095(%r1) # revalidate cpu timer + lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1)# revalidate gprs SAVE_ALL_BASE __LC_SAVE_AREA+64 - SAVE_ALL __LC_MCK_OLD_PSW,__LC_SAVE_AREA+64,0 + la %r12,__LC_MCK_OLD_PSW + tm __LC_MCCK_CODE,0x80 # system damage? + jo mcck_int_main # yes -> rest of mcck code invalid + tm __LC_MCCK_CODE+5,0x02 # stored cpu timer value valid? + jo 0f + spt __LC_LAST_UPDATE_TIMER #ifdef CONFIG_VIRT_CPU_ACCOUNTING - tm SP_PSW+1(%r15),0x01 # interrupting from user ? + mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER + mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER + mvc __LC_LAST_UPDATE_TIMER(8),__LC_EXIT_TIMER +0: tm __LC_MCCK_CODE+2,0x08 # mwp of old psw valid? + jno mcck_no_vtime # no -> no timer update + tm __LC_MCK_OLD_PSW+1,0x01 # interrupting from user ? jz mcck_no_vtime UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER mcck_no_vtime: #endif - brasl %r14,s390_do_machine_check +0: + tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? + jno mcck_int_main # no -> skip cleanup critical + tm __LC_MCK_OLD_PSW+1,0x01 # test problem state bit + jnz mcck_int_main # from user -> load kernel stack + clc __LC_MCK_OLD_PSW+8(8),BASED(.Lcritical_end) + jhe mcck_int_main + clc __LC_MCK_OLD_PSW+8(8),BASED(.Lcritical_start) + jl mcck_int_main + brasl %r14,cleanup_critical +mcck_int_main: + lg %r14,__LC_PANIC_STACK # are we already on the panic stack? + slgr %r14,%r15 + srag %r14,%r14,PAGE_SHIFT + jz 0f + lg %r15,__LC_PANIC_STACK # load panic stack +0: CREATE_STACK_FRAME __LC_MCK_OLD_PSW,__LC_SAVE_AREA+64 + lg %r9,__LC_THREAD_INFO # load pointer to thread_info struct + la %r2,SP_PTREGS(%r15) # load pt_regs + brasl %r14,s390_do_machine_check + tm SP_PSW+1(%r15),0x01 # returning to user ? + jno mcck_return + lg %r1,__LC_KERNEL_STACK # switch to kernel stack + aghi %r1,-SP_SIZE + mvc SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15) + xc __SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1) # clear back chain + lgr %r15,%r1 + stosm __SF_EMPTY(%r15),0x04 # turn dat on + tm __TI_flags+7(%r9),_TIF_MCCK_PENDING + jno mcck_return + brasl %r14,s390_handle_mcck mcck_return: RESTORE_ALL 0 @@ -775,7 +853,7 @@ cleanup_critical: clc 8(8,%r12),BASED(cleanup_table_sysc_work_loop) jl 0f clc 8(8,%r12),BASED(cleanup_table_sysc_work_loop+8) - jl cleanup_sysc_leave + jl cleanup_sysc_return 0: br %r14 @@ -793,6 +871,7 @@ cleanup_system_call: mvc __LC_SAVE_AREA(32),__LC_SAVE_AREA+32 0: stg %r13,__LC_SAVE_AREA+40 SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA stg %r15,__LC_SAVE_AREA+56 llgh %r7,__LC_SVC_INT_CODE #ifdef CONFIG_VIRT_CPU_ACCOUNTING diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 7aea25d6e30..9f3dff6c0b7 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -91,13 +91,12 @@ void do_monitor_call(struct pt_regs *regs, long interruption_code) (void *)(long) smp_processor_id()); } +extern void s390_handle_mcck(void); /* * The idle loop on a S390... */ void default_idle(void) { - psw_t wait_psw; - unsigned long reg; int cpu, rc; local_irq_disable(); @@ -125,38 +124,17 @@ void default_idle(void) cpu_die(); #endif - /* - * Wait for external, I/O or machine check interrupt and - * switch off machine check bit after the wait has ended. - */ - wait_psw.mask = PSW_KERNEL_BITS | PSW_MASK_MCHECK | PSW_MASK_WAIT | - PSW_MASK_IO | PSW_MASK_EXT; -#ifndef CONFIG_ARCH_S390X - asm volatile ( - " basr %0,0\n" - "0: la %0,1f-0b(%0)\n" - " st %0,4(%1)\n" - " oi 4(%1),0x80\n" - " lpsw 0(%1)\n" - "1: la %0,2f-1b(%0)\n" - " st %0,4(%1)\n" - " oi 4(%1),0x80\n" - " ni 1(%1),0xf9\n" - " lpsw 0(%1)\n" - "2:" - : "=&a" (reg) : "a" (&wait_psw) : "memory", "cc" ); -#else /* CONFIG_ARCH_S390X */ - asm volatile ( - " larl %0,0f\n" - " stg %0,8(%1)\n" - " lpswe 0(%1)\n" - "0: larl %0,1f\n" - " stg %0,8(%1)\n" - " ni 1(%1),0xf9\n" - " lpswe 0(%1)\n" - "1:" - : "=&a" (reg) : "a" (&wait_psw) : "memory", "cc" ); -#endif /* CONFIG_ARCH_S390X */ + local_mcck_disable(); + if (test_thread_flag(TIF_MCCK_PENDING)) { + local_mcck_enable(); + local_irq_enable(); + s390_handle_mcck(); + return; + } + + /* Wait for external, I/O or machine check interrupt. */ + __load_psw_mask(PSW_KERNEL_BITS | PSW_MASK_WAIT | + PSW_MASK_IO | PSW_MASK_EXT); } void cpu_idle(void) diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index df83215beac..eb7be0ad717 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -414,7 +414,8 @@ setup_lowcore(void) lc->program_new_psw.mask = PSW_KERNEL_BITS; lc->program_new_psw.addr = PSW_ADDR_AMODE | (unsigned long)pgm_check_handler; - lc->mcck_new_psw.mask = PSW_KERNEL_BITS; + lc->mcck_new_psw.mask = + PSW_KERNEL_BITS & ~PSW_MASK_MCHECK & ~PSW_MASK_DAT; lc->mcck_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) mcck_int_handler; lc->io_new_psw.mask = PSW_KERNEL_BITS; @@ -424,12 +425,18 @@ setup_lowcore(void) lc->kernel_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE; lc->async_stack = (unsigned long) __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0) + ASYNC_SIZE; -#ifdef CONFIG_CHECK_STACK lc->panic_stack = (unsigned long) __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0) + PAGE_SIZE; -#endif lc->current_task = (unsigned long) init_thread_union.thread_info.task; lc->thread_info = (unsigned long) &init_thread_union; +#ifndef CONFIG_ARCH_S390X + if (MACHINE_HAS_IEEE) { + lc->extended_save_area_addr = (__u32) + __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0); + /* enable extended save area */ + ctl_set_bit(14, 29); + } +#endif #ifdef CONFIG_ARCH_S390X if (MACHINE_HAS_DIAG44) lc->diag44_opcode = 0x83000044; diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 93c71fef99d..50c335067cf 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -773,13 +773,24 @@ void __init smp_prepare_cpus(unsigned int max_cpus) *(lowcore_ptr[i]) = S390_lowcore; lowcore_ptr[i]->async_stack = stack + (ASYNC_SIZE); -#ifdef CONFIG_CHECK_STACK stack = __get_free_pages(GFP_KERNEL,0); if (stack == 0ULL) panic("smp_boot_cpus failed to allocate memory\n"); lowcore_ptr[i]->panic_stack = stack + (PAGE_SIZE); +#ifndef __s390x__ + if (MACHINE_HAS_IEEE) { + lowcore_ptr[i]->extended_save_area_addr = + (__u32) __get_free_pages(GFP_KERNEL,0); + if (lowcore_ptr[i]->extended_save_area_addr == 0) + panic("smp_boot_cpus failed to " + "allocate memory\n"); + } #endif } +#ifndef __s390x__ + if (MACHINE_HAS_IEEE) + ctl_set_bit(14, 29); /* enable extended save area */ +#endif set_prefix((u32)(unsigned long) lowcore_ptr[smp_processor_id()]); for_each_cpu(cpu) diff --git a/drivers/s390/s390mach.c b/drivers/s390/s390mach.c index ffa996c8a90..5bb255e02ac 100644 --- a/drivers/s390/s390mach.c +++ b/drivers/s390/s390mach.c @@ -31,14 +31,14 @@ extern void css_reiterate_subchannels(void); extern struct workqueue_struct *slow_path_wq; extern struct work_struct slow_path_work; -static void +static NORET_TYPE void s390_handle_damage(char *msg) { - printk(KERN_EMERG "%s\n", msg); #ifdef CONFIG_SMP smp_send_stop(); #endif disabled_wait((unsigned long) __builtin_return_address(0)); + for(;;); } /* @@ -122,40 +122,39 @@ repeat: return 0; } +struct mcck_struct { + int kill_task; + int channel_report; + int warning; + unsigned long long mcck_code; +}; + +static DEFINE_PER_CPU(struct mcck_struct, cpu_mcck); + /* - * machine check handler. + * Main machine check handler function. Will be called with interrupts enabled + * or disabled and machine checks enabled or disabled. */ void -s390_do_machine_check(void) +s390_handle_mcck(void) { - struct mci *mci; - - mci = (struct mci *) &S390_lowcore.mcck_interruption_code; + unsigned long flags; + struct mcck_struct mcck; - if (mci->sd) /* system damage */ - s390_handle_damage("received system damage machine check\n"); + /* + * Disable machine checks and get the current state of accumulated + * machine checks. Afterwards delete the old state and enable machine + * checks again. + */ + local_irq_save(flags); + local_mcck_disable(); + mcck = __get_cpu_var(cpu_mcck); + memset(&__get_cpu_var(cpu_mcck), 0, sizeof(struct mcck_struct)); + clear_thread_flag(TIF_MCCK_PENDING); + local_mcck_enable(); + local_irq_restore(flags); - if (mci->pd) /* instruction processing damage */ - s390_handle_damage("received instruction processing " - "damage machine check\n"); - - if (mci->se) /* storage error uncorrected */ - s390_handle_damage("received storage error uncorrected " - "machine check\n"); - - if (mci->sc) /* storage error corrected */ - printk(KERN_WARNING - "received storage error corrected machine check\n"); - - if (mci->ke) /* storage key-error uncorrected */ - s390_handle_damage("received storage key-error uncorrected " - "machine check\n"); - - if (mci->ds && mci->fa) /* storage degradation */ - s390_handle_damage("received storage degradation machine " - "check\n"); - - if (mci->cp) /* channel report word pending */ + if (mcck.channel_report) up(&m_sem); #ifdef CONFIG_MACHCHK_WARNING @@ -168,7 +167,7 @@ s390_do_machine_check(void) * On VM we only get one interrupt per virtally presented machinecheck. * Though one suffices, we may get one interrupt per (virtual) processor. */ - if (mci->w) { /* WARNING pending ? */ + if (mcck.warning) { /* WARNING pending ? */ static int mchchk_wng_posted = 0; /* * Use single machine clear, as we cannot handle smp right now @@ -178,6 +177,261 @@ s390_do_machine_check(void) kill_proc(1, SIGPWR, 1); } #endif + + if (mcck.kill_task) { + local_irq_enable(); + printk(KERN_EMERG "mcck: Terminating task because of machine " + "malfunction (code 0x%016llx).\n", mcck.mcck_code); + printk(KERN_EMERG "mcck: task: %s, pid: %d.\n", + current->comm, current->pid); + do_exit(SIGSEGV); + } +} + +/* + * returns 0 if all registers could be validated + * returns 1 otherwise + */ +static int +s390_revalidate_registers(struct mci *mci) +{ + int kill_task; + u64 tmpclock; + u64 zero; + void *fpt_save_area, *fpt_creg_save_area; + + kill_task = 0; + zero = 0; + /* General purpose registers */ + if (!mci->gr) + /* + * General purpose registers couldn't be restored and have + * unknown contents. Process needs to be terminated. + */ + kill_task = 1; + + /* Revalidate floating point registers */ + if (!mci->fp) + /* + * Floating point registers can't be restored and + * therefore the process needs to be terminated. + */ + kill_task = 1; + +#ifndef __s390x__ + asm volatile("ld 0,0(%0)\n" + "ld 2,8(%0)\n" + "ld 4,16(%0)\n" + "ld 6,24(%0)" + : : "a" (&S390_lowcore.floating_pt_save_area)); +#endif + + if (MACHINE_HAS_IEEE) { +#ifdef __s390x__ + fpt_save_area = &S390_lowcore.floating_pt_save_area; + fpt_creg_save_area = &S390_lowcore.fpt_creg_save_area; +#else + fpt_save_area = (void *) S390_lowcore.extended_save_area_addr; + fpt_creg_save_area = fpt_save_area+128; +#endif + /* Floating point control register */ + if (!mci->fc) { + /* + * Floating point control register can't be restored. + * Task will be terminated. + */ + asm volatile ("lfpc 0(%0)" : : "a" (&zero)); + kill_task = 1; + + } + else + asm volatile ( + "lfpc 0(%0)" + : : "a" (fpt_creg_save_area)); + + asm volatile("ld 0,0(%0)\n" + "ld 1,8(%0)\n" + "ld 2,16(%0)\n" + "ld 3,24(%0)\n" + "ld 4,32(%0)\n" + "ld 5,40(%0)\n" + "ld 6,48(%0)\n" + "ld 7,56(%0)\n" + "ld 8,64(%0)\n" + "ld 9,72(%0)\n" + "ld 10,80(%0)\n" + "ld 11,88(%0)\n" + "ld 12,96(%0)\n" + "ld 13,104(%0)\n" + "ld 14,112(%0)\n" + "ld 15,120(%0)\n" + : : "a" (fpt_save_area)); + } + + /* Revalidate access registers */ + asm volatile("lam 0,15,0(%0)" + : : "a" (&S390_lowcore.access_regs_save_area)); + if (!mci->ar) + /* + * Access registers have unknown contents. + * Terminating task. + */ + kill_task = 1; + + /* Revalidate control registers */ + if (!mci->cr) + /* + * Control registers have unknown contents. + * Can't recover and therefore stopping machine. + */ + s390_handle_damage("invalid control registers."); + else +#ifdef __s390x__ + asm volatile("lctlg 0,15,0(%0)" + : : "a" (&S390_lowcore.cregs_save_area)); +#else + asm volatile("lctl 0,15,0(%0)" + : : "a" (&S390_lowcore.cregs_save_area)); +#endif + + /* + * We don't even try to revalidate the TOD register, since we simply + * can't write something sensible into that register. + */ + +#ifdef __s390x__ + /* + * See if we can revalidate the TOD programmable register with its + * old contents (should be zero) otherwise set it to zero. + */ + if (!mci->pr) + asm volatile("sr 0,0\n" + "sckpf" + : : : "0", "cc"); + else + asm volatile( + "l 0,0(%0)\n" + "sckpf" + : : "a" (&S390_lowcore.tod_progreg_save_area) : "0", "cc"); +#endif + + /* Revalidate clock comparator register */ + asm volatile ("stck 0(%1)\n" + "sckc 0(%1)" + : "=m" (tmpclock) : "a" (&(tmpclock)) : "cc", "memory"); + + /* Check if old PSW is valid */ + if (!mci->wp) + /* + * Can't tell if we come from user or kernel mode + * -> stopping machine. + */ + s390_handle_damage("old psw invalid."); + + if (!mci->ms || !mci->pm || !mci->ia) + kill_task = 1; + + return kill_task; +} + +/* + * machine check handler. + */ +void +s390_do_machine_check(struct pt_regs *regs) +{ + struct mci *mci; + struct mcck_struct *mcck; + int umode; + + mci = (struct mci *) &S390_lowcore.mcck_interruption_code; + mcck = &__get_cpu_var(cpu_mcck); + umode = user_mode(regs); + + if (mci->sd) + /* System damage -> stopping machine */ + s390_handle_damage("received system damage machine check."); + + if (mci->pd) { + if (mci->b) { + /* Processing backup -> verify if we can survive this */ + u64 z_mcic, o_mcic, t_mcic; +#ifdef __s390x__ + z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<29); + o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 | + 1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 | + 1ULL<<30 | 1ULL<<21 | 1ULL<<20 | 1ULL<<17 | + 1ULL<<16); +#else + z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<57 | 1ULL<<50 | + 1ULL<<29); + o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 | + 1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 | + 1ULL<<30 | 1ULL<<20 | 1ULL<<17 | 1ULL<<16); +#endif + t_mcic = *(u64 *)mci; + + if (((t_mcic & z_mcic) != 0) || + ((t_mcic & o_mcic) != o_mcic)) { + s390_handle_damage("processing backup machine " + "check with damage."); + } + if (!umode) + s390_handle_damage("processing backup machine " + "check in kernel mode."); + mcck->kill_task = 1; + mcck->mcck_code = *(unsigned long long *) mci; + } + else { + /* Processing damage -> stopping machine */ + s390_handle_damage("received instruction processing " + "damage machine check."); + } + } + if (s390_revalidate_registers(mci)) { + if (umode) { + /* + * Couldn't restore all register contents while in + * user mode -> mark task for termination. + */ + mcck->kill_task = 1; + mcck->mcck_code = *(unsigned long long *) mci; + set_thread_flag(TIF_MCCK_PENDING); + } + else + /* + * Couldn't restore all register contents while in + * kernel mode -> stopping machine. + */ + s390_handle_damage("unable to revalidate registers."); + } + + if (mci->se) + /* Storage error uncorrected */ + s390_handle_damage("received storage error uncorrected " + "machine check."); + + if (mci->ke) + /* Storage key-error uncorrected */ + s390_handle_damage("received storage key-error uncorrected " + "machine check."); + + if (mci->ds && mci->fa) + /* Storage degradation */ + s390_handle_damage("received storage degradation machine " + "check."); + + if (mci->cp) { + /* Channel report word pending */ + mcck->channel_report = 1; + set_thread_flag(TIF_MCCK_PENDING); + } + + if (mci->w) { + /* Warning pending */ + mcck->warning = 1; + set_thread_flag(TIF_MCCK_PENDING); + } } /* @@ -189,9 +443,8 @@ static int machine_check_init(void) { init_MUTEX_LOCKED(&m_sem); - ctl_clear_bit(14, 25); /* disable damage MCH */ - ctl_set_bit(14, 26); /* enable degradation MCH */ - ctl_set_bit(14, 27); /* enable system recovery MCH */ + ctl_clear_bit(14, 25); /* disable external damage MCH */ + ctl_set_bit(14, 27); /* enable system recovery MCH */ #ifdef CONFIG_MACHCHK_WARNING ctl_set_bit(14, 24); /* enable warning MCH */ #endif diff --git a/drivers/s390/s390mach.h b/drivers/s390/s390mach.h index 7e26f0f1b0d..4eaa7017918 100644 --- a/drivers/s390/s390mach.h +++ b/drivers/s390/s390mach.h @@ -16,20 +16,45 @@ struct mci { __u32 sd : 1; /* 00 system damage */ __u32 pd : 1; /* 01 instruction-processing damage */ __u32 sr : 1; /* 02 system recovery */ - __u32 to_be_defined_1 : 4; /* 03-06 */ + __u32 to_be_defined_1 : 1; /* 03 */ + __u32 cd : 1; /* 04 timing-facility damage */ + __u32 ed : 1; /* 05 external damage */ + __u32 to_be_defined_2 : 1; /* 06 */ __u32 dg : 1; /* 07 degradation */ __u32 w : 1; /* 08 warning pending */ __u32 cp : 1; /* 09 channel-report pending */ - __u32 to_be_defined_2 : 6; /* 10-15 */ + __u32 sp : 1; /* 10 service-processor damage */ + __u32 ck : 1; /* 11 channel-subsystem damage */ + __u32 to_be_defined_3 : 2; /* 12-13 */ + __u32 b : 1; /* 14 backed up */ + __u32 to_be_defined_4 : 1; /* 15 */ __u32 se : 1; /* 16 storage error uncorrected */ __u32 sc : 1; /* 17 storage error corrected */ __u32 ke : 1; /* 18 storage-key error uncorrected */ __u32 ds : 1; /* 19 storage degradation */ - __u32 to_be_defined_3 : 4; /* 20-23 */ + __u32 wp : 1; /* 20 psw mwp validity */ + __u32 ms : 1; /* 21 psw mask and key validity */ + __u32 pm : 1; /* 22 psw program mask and cc validity */ + __u32 ia : 1; /* 23 psw instruction address validity */ __u32 fa : 1; /* 24 failing storage address validity */ - __u32 to_be_defined_4 : 7; /* 25-31 */ + __u32 to_be_defined_5 : 1; /* 25 */ + __u32 ec : 1; /* 26 external damage code validity */ + __u32 fp : 1; /* 27 floating point register validity */ + __u32 gr : 1; /* 28 general register validity */ + __u32 cr : 1; /* 29 control register validity */ + __u32 to_be_defined_6 : 1; /* 30 */ + __u32 st : 1; /* 31 storage logical validity */ __u32 ie : 1; /* 32 indirect storage error */ - __u32 to_be_defined_5 : 31; /* 33-63 */ + __u32 ar : 1; /* 33 access register validity */ + __u32 da : 1; /* 34 delayed access exception */ + __u32 to_be_defined_7 : 7; /* 35-41 */ + __u32 pr : 1; /* 42 tod programmable register validity */ + __u32 fc : 1; /* 43 fp control register validity */ + __u32 ap : 1; /* 44 ancillary report */ + __u32 to_be_defined_8 : 1; /* 45 */ + __u32 ct : 1; /* 46 cpu timer validity */ + __u32 cc : 1; /* 47 clock comparator validity */ + __u32 to_be_defined_9 : 16; /* 47-63 */ }; /* diff --git a/include/asm-s390/lowcore.h b/include/asm-s390/lowcore.h index df5172fc589..76b5b19c0ae 100644 --- a/include/asm-s390/lowcore.h +++ b/include/asm-s390/lowcore.h @@ -109,10 +109,14 @@ #ifndef __s390x__ #define __LC_PFAULT_INTPARM 0x080 +#define __LC_CPU_TIMER_SAVE_AREA 0x0D8 #define __LC_AREGS_SAVE_AREA 0x120 +#define __LC_GPREGS_SAVE_AREA 0x180 #define __LC_CREGS_SAVE_AREA 0x1C0 #else /* __s390x__ */ #define __LC_PFAULT_INTPARM 0x11B8 +#define __LC_GPREGS_SAVE_AREA 0x1280 +#define __LC_CPU_TIMER_SAVE_AREA 0x1328 #define __LC_AREGS_SAVE_AREA 0x1340 #define __LC_CREGS_SAVE_AREA 0x1380 #endif /* __s390x__ */ @@ -167,7 +171,8 @@ struct _lowcore __u16 subchannel_nr; /* 0x0ba */ __u32 io_int_parm; /* 0x0bc */ __u32 io_int_word; /* 0x0c0 */ - __u8 pad3[0xD8-0xC4]; /* 0x0c4 */ + __u8 pad3[0xD4-0xC4]; /* 0x0c4 */ + __u32 extended_save_area_addr; /* 0x0d4 */ __u32 cpu_timer_save_area[2]; /* 0x0d8 */ __u32 clock_comp_save_area[2]; /* 0x0e0 */ __u32 mcck_interruption_code[2]; /* 0x0e8 */ diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h index fb46e9090b5..8bd14de69e3 100644 --- a/include/asm-s390/processor.h +++ b/include/asm-s390/processor.h @@ -206,6 +206,18 @@ unsigned long get_wchan(struct task_struct *p); asm volatile ("ex 0,%0" : : "i" (__LC_DIAG44_OPCODE) : "memory") #endif /* __s390x__ */ +/* + * Set PSW to specified value. + */ +static inline void __load_psw(psw_t psw) +{ +#ifndef __s390x__ + asm volatile ("lpsw 0(%0)" : : "a" (&psw), "m" (psw) : "cc" ); +#else + asm volatile ("lpswe 0(%0)" : : "a" (&psw), "m" (psw) : "cc" ); +#endif +} + /* * Set PSW mask to specified value, while leaving the * PSW addr pointing to the next instruction. @@ -214,8 +226,8 @@ unsigned long get_wchan(struct task_struct *p); static inline void __load_psw_mask (unsigned long mask) { unsigned long addr; - psw_t psw; + psw.mask = mask; #ifndef __s390x__ @@ -241,30 +253,8 @@ static inline void __load_psw_mask (unsigned long mask) */ static inline void enabled_wait(void) { - unsigned long reg; - psw_t wait_psw; - - wait_psw.mask = PSW_BASE_BITS | PSW_MASK_IO | PSW_MASK_EXT | - PSW_MASK_MCHECK | PSW_MASK_WAIT | PSW_DEFAULT_KEY; -#ifndef __s390x__ - asm volatile ( - " basr %0,0\n" - "0: la %0,1f-0b(%0)\n" - " st %0,4(%1)\n" - " oi 4(%1),0x80\n" - " lpsw 0(%1)\n" - "1:" - : "=&a" (reg) : "a" (&wait_psw), "m" (wait_psw) - : "memory", "cc" ); -#else /* __s390x__ */ - asm volatile ( - " larl %0,0f\n" - " stg %0,8(%1)\n" - " lpswe 0(%1)\n" - "0:" - : "=&a" (reg) : "a" (&wait_psw), "m" (wait_psw) - : "memory", "cc" ); -#endif /* __s390x__ */ + __load_psw_mask(PSW_BASE_BITS | PSW_MASK_IO | PSW_MASK_EXT | + PSW_MASK_MCHECK | PSW_MASK_WAIT | PSW_DEFAULT_KEY); } /* @@ -273,13 +263,11 @@ static inline void enabled_wait(void) static inline void disabled_wait(unsigned long code) { - char psw_buffer[2*sizeof(psw_t)]; unsigned long ctl_buf; - psw_t *dw_psw = (psw_t *)(((unsigned long) &psw_buffer+sizeof(psw_t)-1) - & -sizeof(psw_t)); + psw_t dw_psw; - dw_psw->mask = PSW_BASE_BITS | PSW_MASK_WAIT; - dw_psw->addr = code; + dw_psw.mask = PSW_BASE_BITS | PSW_MASK_WAIT; + dw_psw.addr = code; /* * Store status and then load disabled wait psw, * the processor is dead afterwards @@ -301,7 +289,7 @@ static inline void disabled_wait(unsigned long code) " oi 0x1c0,0x10\n" /* fake protection bit */ " lpsw 0(%1)" : "=m" (ctl_buf) - : "a" (dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc" ); + : "a" (&dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc" ); #else /* __s390x__ */ asm volatile (" stctg 0,0,0(%2)\n" " ni 4(%2),0xef\n" /* switch off protection */ @@ -333,7 +321,7 @@ static inline void disabled_wait(unsigned long code) " oi 0x384(1),0x10\n" /* fake protection bit */ " lpswe 0(%1)" : "=m" (ctl_buf) - : "a" (dw_psw), "a" (&ctl_buf), + : "a" (&dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc", "0", "1"); #endif /* __s390x__ */ } diff --git a/include/asm-s390/ptrace.h b/include/asm-s390/ptrace.h index 4eff8f2e3bf..fc7c96edc69 100644 --- a/include/asm-s390/ptrace.h +++ b/include/asm-s390/ptrace.h @@ -276,7 +276,7 @@ typedef struct #endif /* __s390x__ */ #define PSW_KERNEL_BITS (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_PRIMARY | \ - PSW_DEFAULT_KEY) + PSW_MASK_MCHECK | PSW_DEFAULT_KEY) #define PSW_USER_BITS (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_HOME | \ PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK | \ PSW_MASK_PSTATE | PSW_DEFAULT_KEY) diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h index 81514d76edc..e3cb3ce1d24 100644 --- a/include/asm-s390/system.h +++ b/include/asm-s390/system.h @@ -16,6 +16,7 @@ #include #include #include +#include #ifdef __KERNEL__ @@ -331,9 +332,6 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size) #ifdef __s390x__ -#define __load_psw(psw) \ - __asm__ __volatile__("lpswe 0(%0)" : : "a" (&psw), "m" (psw) : "cc" ); - #define __ctl_load(array, low, high) ({ \ typedef struct { char _[sizeof(array)]; } addrtype; \ __asm__ __volatile__ ( \ @@ -390,9 +388,6 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size) #else /* __s390x__ */ -#define __load_psw(psw) \ - __asm__ __volatile__("lpsw 0(%0)" : : "a" (&psw) : "cc" ); - #define __ctl_load(array, low, high) ({ \ typedef struct { char _[sizeof(array)]; } addrtype; \ __asm__ __volatile__ ( \ @@ -451,6 +446,20 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size) /* For spinlocks etc */ #define local_irq_save(x) ((x) = local_irq_disable()) +/* + * Use to set psw mask except for the first byte which + * won't be changed by this function. + */ +static inline void +__set_psw_mask(unsigned long mask) +{ + local_save_flags(mask); + __load_psw_mask(mask); +} + +#define local_mcck_enable() __set_psw_mask(PSW_KERNEL_BITS) +#define local_mcck_disable() __set_psw_mask(PSW_KERNEL_BITS & ~PSW_MASK_MCHECK) + #ifdef CONFIG_SMP extern void smp_ctl_set_bit(int cr, int bit); diff --git a/include/asm-s390/thread_info.h b/include/asm-s390/thread_info.h index fe101d41e84..6c18a3f2431 100644 --- a/include/asm-s390/thread_info.h +++ b/include/asm-s390/thread_info.h @@ -96,6 +96,7 @@ static inline struct thread_info *current_thread_info(void) #define TIF_RESTART_SVC 4 /* restart svc with new svc number */ #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ #define TIF_SINGLE_STEP 6 /* deliver sigtrap on return to user */ +#define TIF_MCCK_PENDING 7 /* machine check handling is pending */ #define TIF_USEDFPU 16 /* FPU was used by this task this quantum (SMP) */ #define TIF_POLLING_NRFLAG 17 /* true if poll_idle() is polling TIF_NEED_RESCHED */ @@ -109,6 +110,7 @@ static inline struct thread_info *current_thread_info(void) #define _TIF_RESTART_SVC (1< Date: Sat, 25 Jun 2005 14:55:32 -0700 Subject: [PATCH] s390: add vmcp interface Add interface to issue VM control program commands. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/defconfig | 1 + arch/s390/kernel/cpcmd.c | 109 ++++++++++++--------- arch/s390/kernel/setup.c | 6 +- arch/s390/kernel/smp.c | 6 +- arch/s390/kernel/traps.c | 2 +- arch/s390/mm/extmem.c | 4 +- drivers/s390/Kconfig | 7 ++ drivers/s390/char/Makefile | 1 + drivers/s390/char/con3215.c | 4 +- drivers/s390/char/con3270.c | 4 +- drivers/s390/char/vmcp.c | 219 +++++++++++++++++++++++++++++++++++++++++++ drivers/s390/char/vmcp.h | 30 ++++++ drivers/s390/char/vmlogrdr.c | 10 +- drivers/s390/net/smsgiucv.c | 4 +- include/asm-s390/cpcmd.h | 18 +++- 15 files changed, 359 insertions(+), 66 deletions(-) create mode 100644 drivers/s390/char/vmcp.c create mode 100644 drivers/s390/char/vmcp.h (limited to 'arch/s390') diff --git a/arch/s390/defconfig b/arch/s390/defconfig index 07fd0414a4b..89850b2c27e 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -245,6 +245,7 @@ CONFIG_S390_TAPE_BLOCK=y # CONFIG_S390_TAPE_34XX=m # CONFIG_VMLOGRDR is not set +# CONFIG_VMCP is not set # CONFIG_MONREADER is not set # CONFIG_DCSS_SHM is not set diff --git a/arch/s390/kernel/cpcmd.c b/arch/s390/kernel/cpcmd.c index 44df8dc07c5..20062145e84 100644 --- a/arch/s390/kernel/cpcmd.c +++ b/arch/s390/kernel/cpcmd.c @@ -2,7 +2,7 @@ * arch/s390/kernel/cpcmd.c * * S390 version - * Copyright (C) 1999,2000 IBM Deutschland Entwicklung GmbH, IBM Corporation + * Copyright (C) 1999,2005 IBM Deutschland Entwicklung GmbH, IBM Corporation * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), * Christian Borntraeger (cborntra@de.ibm.com), */ @@ -18,93 +18,114 @@ #include static DEFINE_SPINLOCK(cpcmd_lock); -static char cpcmd_buf[240]; +static char cpcmd_buf[241]; /* * the caller of __cpcmd has to ensure that the response buffer is below 2 GB */ -void __cpcmd(char *cmd, char *response, int rlen) +int __cpcmd(const char *cmd, char *response, int rlen, int *response_code) { const int mask = 0x40000000L; unsigned long flags; + int return_code; + int return_len; int cmdlen; spin_lock_irqsave(&cpcmd_lock, flags); cmdlen = strlen(cmd); BUG_ON(cmdlen > 240); - strcpy(cpcmd_buf, cmd); + memcpy(cpcmd_buf, cmd, cmdlen); ASCEBC(cpcmd_buf, cmdlen); if (response != NULL && rlen > 0) { memset(response, 0, rlen); #ifndef CONFIG_ARCH_S390X - asm volatile ("LRA 2,0(%0)\n\t" - "LR 4,%1\n\t" - "O 4,%4\n\t" - "LRA 3,0(%2)\n\t" - "LR 5,%3\n\t" - ".long 0x83240008 # Diagnose X'08'\n\t" - : /* no output */ - : "a" (cpcmd_buf), "d" (cmdlen), - "a" (response), "d" (rlen), "m" (mask) - : "cc", "2", "3", "4", "5" ); + asm volatile ( "lra 2,0(%2)\n" + "lr 4,%3\n" + "o 4,%6\n" + "lra 3,0(%4)\n" + "lr 5,%5\n" + "diag 2,4,0x8\n" + "brc 8, .Litfits\n" + "ar 5, %5\n" + ".Litfits: \n" + "lr %0,4\n" + "lr %1,5\n" + : "=d" (return_code), "=d" (return_len) + : "a" (cpcmd_buf), "d" (cmdlen), + "a" (response), "d" (rlen), "m" (mask) + : "cc", "2", "3", "4", "5" ); #else /* CONFIG_ARCH_S390X */ - asm volatile (" lrag 2,0(%0)\n" - " lgr 4,%1\n" - " o 4,%4\n" - " lrag 3,0(%2)\n" - " lgr 5,%3\n" - " sam31\n" - " .long 0x83240008 # Diagnose X'08'\n" - " sam64" - : /* no output */ - : "a" (cpcmd_buf), "d" (cmdlen), - "a" (response), "d" (rlen), "m" (mask) - : "cc", "2", "3", "4", "5" ); + asm volatile ( "lrag 2,0(%2)\n" + "lgr 4,%3\n" + "o 4,%6\n" + "lrag 3,0(%4)\n" + "lgr 5,%5\n" + "sam31\n" + "diag 2,4,0x8\n" + "sam64\n" + "brc 8, .Litfits\n" + "agr 5, %5\n" + ".Litfits: \n" + "lgr %0,4\n" + "lgr %1,5\n" + : "=d" (return_code), "=d" (return_len) + : "a" (cpcmd_buf), "d" (cmdlen), + "a" (response), "d" (rlen), "m" (mask) + : "cc", "2", "3", "4", "5" ); #endif /* CONFIG_ARCH_S390X */ EBCASC(response, rlen); } else { + return_len = 0; #ifndef CONFIG_ARCH_S390X - asm volatile ("LRA 2,0(%0)\n\t" - "LR 3,%1\n\t" - ".long 0x83230008 # Diagnose X'08'\n\t" - : /* no output */ - : "a" (cpcmd_buf), "d" (cmdlen) - : "2", "3" ); + asm volatile ( "lra 2,0(%1)\n" + "lr 3,%2\n" + "diag 2,3,0x8\n" + "lr %0,3\n" + : "=d" (return_code) + : "a" (cpcmd_buf), "d" (cmdlen) + : "2", "3" ); #else /* CONFIG_ARCH_S390X */ - asm volatile (" lrag 2,0(%0)\n" - " lgr 3,%1\n" - " sam31\n" - " .long 0x83230008 # Diagnose X'08'\n" - " sam64" - : /* no output */ - : "a" (cpcmd_buf), "d" (cmdlen) - : "2", "3" ); + asm volatile ( "lrag 2,0(%1)\n" + "lgr 3,%2\n" + "sam31\n" + "diag 2,3,0x8\n" + "sam64\n" + "lgr %0,3\n" + : "=d" (return_code) + : "a" (cpcmd_buf), "d" (cmdlen) + : "2", "3" ); #endif /* CONFIG_ARCH_S390X */ } spin_unlock_irqrestore(&cpcmd_lock, flags); + if (response_code != NULL) + *response_code = return_code; + return return_len; } EXPORT_SYMBOL(__cpcmd); #ifdef CONFIG_ARCH_S390X -void cpcmd(char *cmd, char *response, int rlen) +int cpcmd(const char *cmd, char *response, int rlen, int *response_code) { char *lowbuf; + int len; + if ((rlen == 0) || (response == NULL) || !((unsigned long)response >> 31)) - __cpcmd(cmd, response, rlen); + len = __cpcmd(cmd, response, rlen, response_code); else { lowbuf = kmalloc(rlen, GFP_KERNEL | GFP_DMA); if (!lowbuf) { printk(KERN_WARNING "cpcmd: could not allocate response buffer\n"); - return; + return -ENOMEM; } - __cpcmd(cmd, lowbuf, rlen); + len = __cpcmd(cmd, lowbuf, rlen, response_code); memcpy(response, lowbuf, rlen); kfree(lowbuf); } + return len; } EXPORT_SYMBOL(cpcmd); diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index eb7be0ad717..b6d740ac0e6 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -198,11 +198,11 @@ static void __init conmode_default(void) char *ptr; if (MACHINE_IS_VM) { - __cpcmd("QUERY CONSOLE", query_buffer, 1024); + __cpcmd("QUERY CONSOLE", query_buffer, 1024, NULL); console_devno = simple_strtoul(query_buffer + 5, NULL, 16); ptr = strstr(query_buffer, "SUBCHANNEL ="); console_irq = simple_strtoul(ptr + 13, NULL, 16); - __cpcmd("QUERY TERM", query_buffer, 1024); + __cpcmd("QUERY TERM", query_buffer, 1024, NULL); ptr = strstr(query_buffer, "CONMODE"); /* * Set the conmode to 3215 so that the device recognition @@ -211,7 +211,7 @@ static void __init conmode_default(void) * 3215 and the 3270 driver will try to access the console * device (3215 as console and 3270 as normal tty). */ - __cpcmd("TERM CONMODE 3215", NULL, 0); + __cpcmd("TERM CONMODE 3215", NULL, 0, NULL); if (ptr == NULL) { #if defined(CONFIG_SCLP_CONSOLE) SET_CONSOLE_SCLP; diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 50c335067cf..642572a8e33 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -284,7 +284,7 @@ static void do_machine_restart(void * __unused) * locks are always held disabled). */ if (MACHINE_IS_VM) - cpcmd ("IPL", NULL, 0); + cpcmd ("IPL", NULL, 0, NULL); else reipl (0x10000 | S390_lowcore.ipl_device); } @@ -313,7 +313,7 @@ static void do_machine_halt(void * __unused) if (atomic_compare_and_swap(-1, smp_processor_id(), &cpuid) == 0) { smp_send_stop(); if (MACHINE_IS_VM && strlen(vmhalt_cmd) > 0) - cpcmd(vmhalt_cmd, NULL, 0); + cpcmd(vmhalt_cmd, NULL, 0, NULL); signal_processor(smp_processor_id(), sigp_stop_and_store_status); } @@ -332,7 +332,7 @@ static void do_machine_power_off(void * __unused) if (atomic_compare_and_swap(-1, smp_processor_id(), &cpuid) == 0) { smp_send_stop(); if (MACHINE_IS_VM && strlen(vmpoff_cmd) > 0) - cpcmd(vmpoff_cmd, NULL, 0); + cpcmd(vmpoff_cmd, NULL, 0, NULL); signal_processor(smp_processor_id(), sigp_stop_and_store_status); } diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c index ca34b6f34b3..bc7b7be7acb 100644 --- a/arch/s390/kernel/traps.c +++ b/arch/s390/kernel/traps.c @@ -735,7 +735,7 @@ void __init trap_init(void) &ext_int_pfault); #endif #ifndef CONFIG_ARCH_S390X - cpcmd("SET PAGEX ON", NULL, 0); + cpcmd("SET PAGEX ON", NULL, 0, NULL); #endif } } diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c index 648deed17e2..c5348108ca3 100644 --- a/arch/s390/mm/extmem.c +++ b/arch/s390/mm/extmem.c @@ -576,8 +576,8 @@ segment_save(char *name) segtype_string[seg->range[i].start & 0xff]); } sprintf(cmd2, "SAVESEG %s", name); - cpcmd(cmd1, NULL, 0); - cpcmd(cmd2, NULL, 0); + cpcmd(cmd1, NULL, 0, NULL); + cpcmd(cmd2, NULL, 0, NULL); spin_unlock(&dcss_lock); } diff --git a/drivers/s390/Kconfig b/drivers/s390/Kconfig index 96413c2cd1a..a86a650f3d6 100644 --- a/drivers/s390/Kconfig +++ b/drivers/s390/Kconfig @@ -187,6 +187,13 @@ config VMLOGRDR *SYMPTOM. This driver depends on the IUCV support driver. +config VMCP + tristate "Support for the z/VM CP interface (VM only)" + help + Select this option if you want to be able to interact with the control + program on z/VM + + config MONREADER tristate "API for reading z/VM monitor service records" depends on IUCV diff --git a/drivers/s390/char/Makefile b/drivers/s390/char/Makefile index 14e8cce9f86..6377a96735d 100644 --- a/drivers/s390/char/Makefile +++ b/drivers/s390/char/Makefile @@ -19,6 +19,7 @@ obj-$(CONFIG_SCLP_CPI) += sclp_cpi.o obj-$(CONFIG_ZVM_WATCHDOG) += vmwatchdog.o obj-$(CONFIG_VMLOGRDR) += vmlogrdr.o +obj-$(CONFIG_VMCP) += vmcp.o tape-$(CONFIG_S390_TAPE_BLOCK) += tape_block.o tape-$(CONFIG_PROC_FS) += tape_proc.o diff --git a/drivers/s390/char/con3215.c b/drivers/s390/char/con3215.c index 022f17bff73..f11a67fda40 100644 --- a/drivers/s390/char/con3215.c +++ b/drivers/s390/char/con3215.c @@ -860,8 +860,8 @@ con3215_init(void) /* Set the console mode for VM */ if (MACHINE_IS_VM) { - cpcmd("TERM CONMODE 3215", NULL, 0); - cpcmd("TERM AUTOCR OFF", NULL, 0); + cpcmd("TERM CONMODE 3215", NULL, 0, NULL); + cpcmd("TERM AUTOCR OFF", NULL, 0, NULL); } /* allocate 3215 request structures */ diff --git a/drivers/s390/char/con3270.c b/drivers/s390/char/con3270.c index d52fb57a6b1..fc7a213e591 100644 --- a/drivers/s390/char/con3270.c +++ b/drivers/s390/char/con3270.c @@ -591,8 +591,8 @@ con3270_init(void) /* Set the console mode for VM */ if (MACHINE_IS_VM) { - cpcmd("TERM CONMODE 3270", 0, 0); - cpcmd("TERM AUTOCR OFF", 0, 0); + cpcmd("TERM CONMODE 3270", NULL, 0, NULL); + cpcmd("TERM AUTOCR OFF", NULL, 0, NULL); } cdev = ccw_device_probe_console(); diff --git a/drivers/s390/char/vmcp.c b/drivers/s390/char/vmcp.c new file mode 100644 index 00000000000..dfbbf235ca2 --- /dev/null +++ b/drivers/s390/char/vmcp.c @@ -0,0 +1,219 @@ +/* + * Copyright (C) 2004,2005 IBM Corporation + * Interface implementation for communication with the v/VM control program + * Author(s): Christian Borntraeger + * + * + * z/VMs CP offers the possibility to issue commands via the diagnose code 8 + * this driver implements a character device that issues these commands and + * returns the answer of CP. + + * The idea of this driver is based on cpint from Neale Ferguson and #CP in CMS + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include "vmcp.h" + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Christian Borntraeger "); +MODULE_DESCRIPTION("z/VM CP interface"); + +static debug_info_t *vmcp_debug; + +static int vmcp_open(struct inode *inode, struct file *file) +{ + struct vmcp_session *session; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + session = kmalloc(sizeof(*session), GFP_KERNEL); + if (!session) + return -ENOMEM; + session->bufsize = PAGE_SIZE; + session->response = NULL; + session->resp_size = 0; + init_MUTEX(&session->mutex); + file->private_data = session; + return nonseekable_open(inode, file); +} + +static int vmcp_release(struct inode *inode, struct file *file) +{ + struct vmcp_session *session; + + session = (struct vmcp_session *)file->private_data; + file->private_data = NULL; + free_pages((unsigned long)session->response, get_order(session->bufsize)); + kfree(session); + return 0; +} + +static ssize_t +vmcp_read(struct file *file, char __user * buff, size_t count, loff_t * ppos) +{ + size_t tocopy; + struct vmcp_session *session; + + session = (struct vmcp_session *)file->private_data; + if (down_interruptible(&session->mutex)) + return -ERESTARTSYS; + if (!session->response) { + up(&session->mutex); + return 0; + } + if (*ppos > session->resp_size) { + up(&session->mutex); + return 0; + } + tocopy = min(session->resp_size - (size_t) (*ppos), count); + tocopy = min(tocopy,session->bufsize - (size_t) (*ppos)); + + if (copy_to_user(buff, session->response + (*ppos), tocopy)) { + up(&session->mutex); + return -EFAULT; + } + up(&session->mutex); + *ppos += tocopy; + return tocopy; +} + +static ssize_t +vmcp_write(struct file *file, const char __user * buff, size_t count, + loff_t * ppos) +{ + char *cmd; + struct vmcp_session *session; + + if (count > 240) + return -EINVAL; + cmd = kmalloc(count + 1, GFP_KERNEL); + if (!cmd) + return -ENOMEM; + if (copy_from_user(cmd, buff, count)) { + kfree(cmd); + return -EFAULT; + } + cmd[count] = '\0'; + session = (struct vmcp_session *)file->private_data; + if (down_interruptible(&session->mutex)) + return -ERESTARTSYS; + if (!session->response) + session->response = (char *)__get_free_pages(GFP_KERNEL + | __GFP_REPEAT | GFP_DMA, + get_order(session->bufsize)); + if (!session->response) { + up(&session->mutex); + kfree(cmd); + return -ENOMEM; + } + debug_text_event(vmcp_debug, 1, cmd); + session->resp_size = cpcmd(cmd, session->response, + session->bufsize, + &session->resp_code); + up(&session->mutex); + kfree(cmd); + *ppos = 0; /* reset the file pointer after a command */ + return count; +} + + +/* + * These ioctls are available, as the semantics of the diagnose 8 call + * does not fit very well into a Linux call. Diagnose X'08' is described in + * CP Programming Services SC24-6084-00 + * + * VMCP_GETCODE: gives the CP return code back to user space + * VMCP_SETBUF: sets the response buffer for the next write call. diagnose 8 + * expects adjacent pages in real storage and to make matters worse, we + * dont know the size of the response. Therefore we default to PAGESIZE and + * let userspace to change the response size, if userspace expects a bigger + * response + */ +static long vmcp_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + struct vmcp_session *session; + int temp; + + session = (struct vmcp_session *)file->private_data; + if (down_interruptible(&session->mutex)) + return -ERESTARTSYS; + switch (cmd) { + case VMCP_GETCODE: + temp = session->resp_code; + up(&session->mutex); + return put_user(temp, (int __user *)arg); + case VMCP_SETBUF: + free_pages((unsigned long)session->response, + get_order(session->bufsize)); + session->response=NULL; + temp = get_user(session->bufsize, (int __user *)arg); + if (get_order(session->bufsize) > 8) { + session->bufsize = PAGE_SIZE; + temp = -EINVAL; + } + up(&session->mutex); + return temp; + case VMCP_GETSIZE: + temp = session->resp_size; + up(&session->mutex); + return put_user(temp, (int __user *)arg); + default: + up(&session->mutex); + return -ENOIOCTLCMD; + } +} + +static struct file_operations vmcp_fops = { + .owner = THIS_MODULE, + .open = &vmcp_open, + .release = &vmcp_release, + .read = &vmcp_read, + .llseek = &no_llseek, + .write = &vmcp_write, + .unlocked_ioctl = &vmcp_ioctl, + .compat_ioctl = &vmcp_ioctl +}; + +static struct miscdevice vmcp_dev = { + .name = "vmcp", + .minor = MISC_DYNAMIC_MINOR, + .fops = &vmcp_fops, +}; + +static int __init vmcp_init(void) +{ + int ret; + + if (!MACHINE_IS_VM) { + printk(KERN_WARNING + "z/VM CP interface is only available under z/VM\n"); + return -ENODEV; + } + ret = misc_register(&vmcp_dev); + if (!ret) + printk(KERN_INFO "z/VM CP interface loaded\n"); + else + printk(KERN_WARNING + "z/VM CP interface not loaded. Could not register misc device.\n"); + vmcp_debug = debug_register("vmcp", 0, 1, 240); + debug_register_view(vmcp_debug, &debug_hex_ascii_view); + return ret; +} + +static void __exit vmcp_exit(void) +{ + WARN_ON(misc_deregister(&vmcp_dev) != 0); + debug_unregister(vmcp_debug); + printk(KERN_INFO "z/VM CP interface unloaded.\n"); +} + +module_init(vmcp_init); +module_exit(vmcp_exit); diff --git a/drivers/s390/char/vmcp.h b/drivers/s390/char/vmcp.h new file mode 100644 index 00000000000..87389e73046 --- /dev/null +++ b/drivers/s390/char/vmcp.h @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2004, 2005 IBM Corporation + * Interface implementation for communication with the v/VM control program + * Version 1.0 + * Author(s): Christian Borntraeger + * + * + * z/VMs CP offers the possibility to issue commands via the diagnose code 8 + * this driver implements a character device that issues these commands and + * returns the answer of CP. + * + * The idea of this driver is based on cpint from Neale Ferguson + */ + +#include +#include + +#define VMCP_GETCODE _IOR(0x10, 1, int) +#define VMCP_SETBUF _IOW(0x10, 2, int) +#define VMCP_GETSIZE _IOR(0x10, 3, int) + +struct vmcp_session { + unsigned int bufsize; + char *response; + int resp_size; + int resp_code; + /* As we use copy_from/to_user, which might * + * sleep and cannot use a spinlock */ + struct semaphore mutex; +}; diff --git a/drivers/s390/char/vmlogrdr.c b/drivers/s390/char/vmlogrdr.c index f7717327d15..491f00c032e 100644 --- a/drivers/s390/char/vmlogrdr.c +++ b/drivers/s390/char/vmlogrdr.c @@ -236,7 +236,7 @@ vmlogrdr_get_recording_class_AB(void) { int len,i; printk (KERN_DEBUG "vmlogrdr: query command: %s\n", cp_command); - cpcmd(cp_command, cp_response, sizeof(cp_response)); + cpcmd(cp_command, cp_response, sizeof(cp_response), NULL); printk (KERN_DEBUG "vmlogrdr: response: %s", cp_response); len = strnlen(cp_response,sizeof(cp_response)); // now the parsing @@ -288,7 +288,7 @@ vmlogrdr_recording(struct vmlogrdr_priv_t * logptr, int action, int purge) { printk (KERN_DEBUG "vmlogrdr: recording command: %s\n", cp_command); - cpcmd(cp_command, cp_response, sizeof(cp_response)); + cpcmd(cp_command, cp_response, sizeof(cp_response), NULL); printk (KERN_DEBUG "vmlogrdr: recording response: %s", cp_response); } @@ -301,7 +301,7 @@ vmlogrdr_recording(struct vmlogrdr_priv_t * logptr, int action, int purge) { qid_string); printk (KERN_DEBUG "vmlogrdr: recording command: %s\n", cp_command); - cpcmd(cp_command, cp_response, sizeof(cp_response)); + cpcmd(cp_command, cp_response, sizeof(cp_response), NULL); printk (KERN_DEBUG "vmlogrdr: recording response: %s", cp_response); /* The recording command will usually answer with 'Command complete' @@ -607,7 +607,7 @@ vmlogrdr_purge_store(struct device * dev, struct device_attribute *attr, const c priv->recording_name); printk (KERN_DEBUG "vmlogrdr: recording command: %s\n", cp_command); - cpcmd(cp_command, cp_response, sizeof(cp_response)); + cpcmd(cp_command, cp_response, sizeof(cp_response), NULL); printk (KERN_DEBUG "vmlogrdr: recording response: %s", cp_response); @@ -682,7 +682,7 @@ vmlogrdr_recording_status_show(struct device_driver *driver, char *buf) { char cp_command[] = "QUERY RECORDING "; int len; - cpcmd(cp_command, buf, 4096); + cpcmd(cp_command, buf, 4096, NULL); len = strlen(buf); return len; } diff --git a/drivers/s390/net/smsgiucv.c b/drivers/s390/net/smsgiucv.c index 1e3f7f3c662..d6469baa7e1 100644 --- a/drivers/s390/net/smsgiucv.c +++ b/drivers/s390/net/smsgiucv.c @@ -138,7 +138,7 @@ static void __exit smsg_exit(void) { if (smsg_handle > 0) { - cpcmd("SET SMSG OFF", 0, 0); + cpcmd("SET SMSG OFF", NULL, 0, NULL); iucv_sever(smsg_pathid, 0); iucv_unregister_program(smsg_handle); driver_unregister(&smsg_driver); @@ -177,7 +177,7 @@ smsg_init(void) smsg_handle = 0; return -EIO; } - cpcmd("SET SMSG IUCV", 0, 0); + cpcmd("SET SMSG IUCV", NULL, 0, NULL); return 0; } diff --git a/include/asm-s390/cpcmd.h b/include/asm-s390/cpcmd.h index 1d33c5da083..1fcf65be7a2 100644 --- a/include/asm-s390/cpcmd.h +++ b/include/asm-s390/cpcmd.h @@ -11,14 +11,28 @@ #define __CPCMD__ /* + * the lowlevel function for cpcmd * the caller of __cpcmd has to ensure that the response buffer is below 2 GB */ -extern void __cpcmd(char *cmd, char *response, int rlen); +extern int __cpcmd(const char *cmd, char *response, int rlen, int *response_code); #ifndef __s390x__ #define cpcmd __cpcmd #else -extern void cpcmd(char *cmd, char *response, int rlen); +/* + * cpcmd is the in-kernel interface for issuing CP commands + * + * cmd: null-terminated command string, max 240 characters + * response: response buffer for VM's textual response + * rlen: size of the response buffer, cpcmd will not exceed this size + * but will cap the output, if its too large. Everything that + * did not fit into the buffer will be silently dropped + * response_code: return pointer for VM's error code + * return value: the size of the response. The caller can check if the buffer + * was large enough by comparing the return value and rlen + * NOTE: If the response buffer is not below 2 GB, cpcmd can sleep + */ +extern int cpcmd(const char *cmd, char *response, int rlen, int *response_code); #endif /*__s390x__*/ #endif -- cgit v1.2.3-70-g09d2 From 66a464dbc8e0345b6f972b92bf1118e043d7c987 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Sat, 25 Jun 2005 14:55:33 -0700 Subject: [PATCH] s390: debug feature changes This patch changes the memory allocation method for the s390 debug feature. Trace buffers had been allocated using the get_free_pages() function before. Therefore it was not possible to get big memory areas in a running system due to memory fragmentation. Now the trace buffers are subdivided into several subbuffers with pagesize. Therefore it is now possible to allocate more memory for the trace buffers and more trace records can be written. In addition to that, dynamic specification of the size of the trace buffers is implemented. It is now possible to change the size of a trace buffer using a new debugfs file instance. When writing a number into this file, the trace buffer size is changed to 'number * pagesize'. In the past all the traces could be obtained from userspace by accessing files in the "proc" filesystem. Now with debugfs we have a new filesystem which should be used for debugging purposes. This patch moves the debug feature from procfs to debugfs. Since the interface of debug_register() changed, all device drivers, which use the debug feature had to be adjusted. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/s390/s390dbf.txt | 79 ++-- arch/s390/kernel/debug.c | 820 +++++++++++++++++++++++++++-------------- drivers/s390/block/dasd.c | 4 +- drivers/s390/block/dasd_proc.c | 3 +- drivers/s390/char/tape_34xx.c | 6 +- drivers/s390/char/tape_core.c | 2 +- drivers/s390/char/tape_proc.c | 1 + drivers/s390/char/vmcp.c | 2 +- drivers/s390/cio/cio.c | 8 +- drivers/s390/cio/qdio.c | 14 +- drivers/s390/cio/qdio.h | 16 +- drivers/s390/net/claw.c | 4 +- drivers/s390/net/ctcdbug.c | 10 +- drivers/s390/net/ctcdbug.h | 10 +- drivers/s390/net/iucv.h | 6 +- drivers/s390/net/lcs.c | 8 +- drivers/s390/net/netiucv.c | 12 +- drivers/s390/net/qeth.h | 14 +- drivers/s390/net/qeth_main.c | 14 +- include/asm-s390/debug.h | 48 ++- 20 files changed, 678 insertions(+), 403 deletions(-) (limited to 'arch/s390') diff --git a/Documentation/s390/s390dbf.txt b/Documentation/s390/s390dbf.txt index 2d1cd939b4d..e24fdeada97 100644 --- a/Documentation/s390/s390dbf.txt +++ b/Documentation/s390/s390dbf.txt @@ -12,8 +12,8 @@ where log records can be stored efficiently in memory, where each component One purpose of this is to inspect the debug logs after a production system crash in order to analyze the reason for the crash. If the system still runs but only a subcomponent which uses dbf failes, -it is possible to look at the debug logs on a live system via the Linux proc -filesystem. +it is possible to look at the debug logs on a live system via the Linux +debugfs filesystem. The debug feature may also very useful for kernel and driver development. Design: @@ -52,16 +52,18 @@ Each debug entry contains the following data: - Flag, if entry is an exception or not The debug logs can be inspected in a live system through entries in -the proc-filesystem. Under the path /proc/s390dbf there is +the debugfs-filesystem. Under the toplevel directory "s390dbf" there is a directory for each registered component, which is named like the -corresponding component. +corresponding component. The debugfs normally should be mounted to +/sys/kernel/debug therefore the debug feature can be accessed unter +/sys/kernel/debug/s390dbf. The content of the directories are files which represent different views to the debug log. Each component can decide which views should be used through registering them with the function debug_register_view(). Predefined views for hex/ascii, sprintf and raw binary data are provided. It is also possible to define other views. The content of -a view can be inspected simply by reading the corresponding proc file. +a view can be inspected simply by reading the corresponding debugfs file. All debug logs have an an actual debug level (range from 0 to 6). The default level is 3. Event and Exception functions have a 'level' @@ -69,14 +71,14 @@ parameter. Only debug entries with a level that is lower or equal than the actual level are written to the log. This means, when writing events, high priority log entries should have a low level value whereas low priority entries should have a high one. -The actual debug level can be changed with the help of the proc-filesystem -through writing a number string "x" to the 'level' proc file which is +The actual debug level can be changed with the help of the debugfs-filesystem +through writing a number string "x" to the 'level' debugfs file which is provided for every debug log. Debugging can be switched off completely -by using "-" on the 'level' proc file. +by using "-" on the 'level' debugfs file. Example: -> echo "-" > /proc/s390dbf/dasd/level +> echo "-" > /sys/kernel/debug/s390dbf/dasd/level It is also possible to deactivate the debug feature globally for every debug log. You can change the behavior using 2 sysctl parameters in @@ -99,11 +101,11 @@ Kernel Interfaces: ------------------ ---------------------------------------------------------------------------- -debug_info_t *debug_register(char *name, int pages_index, int nr_areas, +debug_info_t *debug_register(char *name, int pages, int nr_areas, int buf_size); -Parameter: name: Name of debug log (e.g. used for proc entry) - pages_index: 2^pages_index pages will be allocated per area +Parameter: name: Name of debug log (e.g. used for debugfs entry) + pages: number of pages, which will be allocated per area nr_areas: number of debug areas buf_size: size of data area in each debug entry @@ -134,7 +136,7 @@ Return Value: none Description: Sets new actual debug level if new_level is valid. --------------------------------------------------------------------------- -+void debug_stop_all(void); +void debug_stop_all(void); Parameter: none @@ -270,7 +272,7 @@ Parameter: id: handle for debug log Return Value: 0 : ok < 0: Error -Description: registers new debug view and creates proc dir entry +Description: registers new debug view and creates debugfs dir entry --------------------------------------------------------------------------- int debug_unregister_view (debug_info_t * id, struct debug_view *view); @@ -281,7 +283,7 @@ Parameter: id: handle for debug log Return Value: 0 : ok < 0: Error -Description: unregisters debug view and removes proc dir entry +Description: unregisters debug view and removes debugfs dir entry @@ -308,7 +310,7 @@ static int init(void) { /* register 4 debug areas with one page each and 4 byte data field */ - debug_info = debug_register ("test", 0, 4, 4 ); + debug_info = debug_register ("test", 1, 4, 4 ); debug_register_view(debug_info,&debug_hex_ascii_view); debug_register_view(debug_info,&debug_raw_view); @@ -343,7 +345,7 @@ static int init(void) /* register 4 debug areas with one page each and data field for */ /* format string pointer + 2 varargs (= 3 * sizeof(long)) */ - debug_info = debug_register ("test", 0, 4, sizeof(long) * 3); + debug_info = debug_register ("test", 1, 4, sizeof(long) * 3); debug_register_view(debug_info,&debug_sprintf_view); debug_sprintf_event(debug_info, 2 , "first event in %s:%i\n",__FILE__,__LINE__); @@ -362,16 +364,16 @@ module_exit(cleanup); -ProcFS Interface +Debugfs Interface ---------------- Views to the debug logs can be investigated through reading the corresponding -proc-files: +debugfs-files: Example: -> ls /proc/s390dbf/dasd -flush hex_ascii level raw -> cat /proc/s390dbf/dasd/hex_ascii | sort +1 +> ls /sys/kernel/debug/s390dbf/dasd +flush hex_ascii level pages raw +> cat /sys/kernel/debug/s390dbf/dasd/hex_ascii | sort +1 00 00974733272:680099 2 - 02 0006ad7e 07 ea 4a 90 | .... 00 00974733272:682210 2 - 02 0006ade6 46 52 45 45 | FREE 00 00974733272:682213 2 - 02 0006adf6 07 ea 4a 90 | .... @@ -391,25 +393,36 @@ Changing the debug level Example: -> cat /proc/s390dbf/dasd/level +> cat /sys/kernel/debug/s390dbf/dasd/level 3 -> echo "5" > /proc/s390dbf/dasd/level -> cat /proc/s390dbf/dasd/level +> echo "5" > /sys/kernel/debug/s390dbf/dasd/level +> cat /sys/kernel/debug/s390dbf/dasd/level 5 Flushing debug areas -------------------- Debug areas can be flushed with piping the number of the desired -area (0...n) to the proc file "flush". When using "-" all debug areas +area (0...n) to the debugfs file "flush". When using "-" all debug areas are flushed. Examples: 1. Flush debug area 0: -> echo "0" > /proc/s390dbf/dasd/flush +> echo "0" > /sys/kernel/debug/s390dbf/dasd/flush 2. Flush all debug areas: -> echo "-" > /proc/s390dbf/dasd/flush +> echo "-" > /sys/kernel/debug/s390dbf/dasd/flush + +Changing the size of debug areas +------------------------------------ +It is possible the change the size of debug areas through piping +the number of pages to the debugfs file "pages". The resize request will +also flush the debug areas. + +Example: + +Define 4 pages for the debug areas of debug feature "dasd": +> echo "4" > /sys/kernel/debug/s390dbf/dasd/pages Stooping the debug feature -------------------------- @@ -491,7 +504,7 @@ Defining views -------------- Views are specified with the 'debug_view' structure. There are defined -callback functions which are used for reading and writing the proc files: +callback functions which are used for reading and writing the debugfs files: struct debug_view { char name[DEBUG_MAX_PROCF_LEN]; @@ -525,7 +538,7 @@ typedef int (debug_input_proc_t) (debug_info_t* id, The "private_data" member can be used as pointer to view specific data. It is not used by the debug feature itself. -The output when reading a debug-proc file is structured like this: +The output when reading a debugfs file is structured like this: "prolog_proc output" @@ -534,13 +547,13 @@ The output when reading a debug-proc file is structured like this: "header_proc output 3" "format_proc output 3" ... -When a view is read from the proc fs, the Debug Feature calls the +When a view is read from the debugfs, the Debug Feature calls the 'prolog_proc' once for writing the prolog. Then 'header_proc' and 'format_proc' are called for each existing debug entry. The input_proc can be used to implement functionality when it is written to -the view (e.g. like with 'echo "0" > /proc/s390dbf/dasd/level). +the view (e.g. like with 'echo "0" > /sys/kernel/debug/s390dbf/dasd/level). For header_proc there can be used the default function debug_dflt_header_fn() which is defined in in debug.h. @@ -602,7 +615,7 @@ debug_info = debug_register ("test", 0, 4, 4 )); debug_register_view(debug_info, &debug_test_view); for(i = 0; i < 10; i ++) debug_int_event(debug_info, 1, i); -> cat /proc/s390dbf/test/myview +> cat /sys/kernel/debug/s390dbf/test/myview 00 00964419734:611402 1 - 00 88042ca This error........... 00 00964419734:611405 1 - 00 88042ca That error........... 00 00964419734:611408 1 - 00 88042ca Problem.............. diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index 91f8ce5543d..960ba6029c3 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -19,22 +19,27 @@ #include #include #include - #include #include +#include +#include #include #define DEBUG_PROLOG_ENTRY -1 +#define ALL_AREAS 0 /* copy all debug areas */ +#define NO_AREAS 1 /* copy no debug areas */ + /* typedefs */ typedef struct file_private_info { loff_t offset; /* offset of last read in file */ int act_area; /* number of last formated area */ + int act_page; /* act page in given area */ int act_entry; /* last formated entry (offset */ /* relative to beginning of last */ - /* formated area) */ + /* formated page) */ size_t act_entry_offset; /* up to this offset we copied */ /* in last read the last formated */ /* entry to userland */ @@ -51,8 +56,8 @@ typedef struct * This assumes that all args are converted into longs * on L/390 this is the case for all types of parameter * except of floats, and long long (32 bit) - * - */ + * + */ long args[0]; } debug_sprintf_entry_t; @@ -63,32 +68,38 @@ extern void tod_to_timeval(uint64_t todval, struct timeval *xtime); static int debug_init(void); static ssize_t debug_output(struct file *file, char __user *user_buf, - size_t user_len, loff_t * offset); + size_t user_len, loff_t * offset); static ssize_t debug_input(struct file *file, const char __user *user_buf, - size_t user_len, loff_t * offset); + size_t user_len, loff_t * offset); static int debug_open(struct inode *inode, struct file *file); static int debug_close(struct inode *inode, struct file *file); -static debug_info_t* debug_info_create(char *name, int page_order, int nr_areas, int buf_size); +static debug_info_t* debug_info_create(char *name, int pages_per_area, + int nr_areas, int buf_size); static void debug_info_get(debug_info_t *); static void debug_info_put(debug_info_t *); static int debug_prolog_level_fn(debug_info_t * id, - struct debug_view *view, char *out_buf); + struct debug_view *view, char *out_buf); static int debug_input_level_fn(debug_info_t * id, struct debug_view *view, - struct file *file, const char __user *user_buf, - size_t user_buf_size, loff_t * offset); + struct file *file, const char __user *user_buf, + size_t user_buf_size, loff_t * offset); +static int debug_prolog_pages_fn(debug_info_t * id, + struct debug_view *view, char *out_buf); +static int debug_input_pages_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_buf_size, loff_t * offset); static int debug_input_flush_fn(debug_info_t * id, struct debug_view *view, - struct file *file, const char __user *user_buf, - size_t user_buf_size, loff_t * offset); + struct file *file, const char __user *user_buf, + size_t user_buf_size, loff_t * offset); static int debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view, - char *out_buf, const char *in_buf); + char *out_buf, const char *in_buf); static int debug_raw_format_fn(debug_info_t * id, - struct debug_view *view, char *out_buf, - const char *in_buf); + struct debug_view *view, char *out_buf, + const char *in_buf); static int debug_raw_header_fn(debug_info_t * id, struct debug_view *view, - int area, debug_entry_t * entry, char *out_buf); + int area, debug_entry_t * entry, char *out_buf); static int debug_sprintf_format_fn(debug_info_t * id, struct debug_view *view, - char *out_buf, debug_sprintf_entry_t *curr_event); + char *out_buf, debug_sprintf_entry_t *curr_event); /* globals */ @@ -119,6 +130,15 @@ struct debug_view debug_level_view = { NULL }; +struct debug_view debug_pages_view = { + "pages", + &debug_prolog_pages_fn, + NULL, + NULL, + &debug_input_pages_fn, + NULL +}; + struct debug_view debug_flush_view = { "flush", NULL, @@ -149,98 +169,161 @@ DECLARE_MUTEX(debug_lock); static int initialized; static struct file_operations debug_file_ops = { - .owner = THIS_MODULE, + .owner = THIS_MODULE, .read = debug_output, - .write = debug_input, + .write = debug_input, .open = debug_open, .release = debug_close, }; -static struct proc_dir_entry *debug_proc_root_entry; +static struct dentry *debug_debugfs_root_entry; /* functions */ +/* + * debug_areas_alloc + * - Debug areas are implemented as a threedimensonal array: + * areas[areanumber][pagenumber][pageoffset] + */ + +static debug_entry_t*** +debug_areas_alloc(int pages_per_area, int nr_areas) +{ + debug_entry_t*** areas; + int i,j; + + areas = (debug_entry_t ***) kmalloc(nr_areas * + sizeof(debug_entry_t**), + GFP_KERNEL); + if (!areas) + goto fail_malloc_areas; + for (i = 0; i < nr_areas; i++) { + areas[i] = (debug_entry_t**) kmalloc(pages_per_area * + sizeof(debug_entry_t*),GFP_KERNEL); + if (!areas[i]) { + goto fail_malloc_areas2; + } + for(j = 0; j < pages_per_area; j++) { + areas[i][j] = (debug_entry_t*)kmalloc(PAGE_SIZE, + GFP_KERNEL); + if(!areas[i][j]) { + for(j--; j >=0 ; j--) { + kfree(areas[i][j]); + } + kfree(areas[i]); + goto fail_malloc_areas2; + } else { + memset(areas[i][j],0,PAGE_SIZE); + } + } + } + return areas; + +fail_malloc_areas2: + for(i--; i >= 0; i--){ + for(j=0; j < pages_per_area;j++){ + kfree(areas[i][j]); + } + kfree(areas[i]); + } + kfree(areas); +fail_malloc_areas: + return NULL; + +} + + /* * debug_info_alloc * - alloc new debug-info */ -static debug_info_t* debug_info_alloc(char *name, int page_order, - int nr_areas, int buf_size) +static debug_info_t* +debug_info_alloc(char *name, int pages_per_area, int nr_areas, int buf_size, + int level, int mode) { debug_info_t* rc; - int i; /* alloc everything */ - rc = (debug_info_t*) kmalloc(sizeof(debug_info_t), GFP_ATOMIC); + rc = (debug_info_t*) kmalloc(sizeof(debug_info_t), GFP_KERNEL); if(!rc) goto fail_malloc_rc; - rc->active_entry = (int*)kmalloc(nr_areas * sizeof(int), GFP_ATOMIC); - if(!rc->active_entry) - goto fail_malloc_active_entry; - memset(rc->active_entry, 0, nr_areas * sizeof(int)); - rc->areas = (debug_entry_t **) kmalloc(nr_areas * - sizeof(debug_entry_t *), - GFP_ATOMIC); - if (!rc->areas) - goto fail_malloc_areas; - for (i = 0; i < nr_areas; i++) { - rc->areas[i] = (debug_entry_t *) __get_free_pages(GFP_ATOMIC, - page_order); - if (!rc->areas[i]) { - for (i--; i >= 0; i--) { - free_pages((unsigned long) rc->areas[i], - page_order); - } - goto fail_malloc_areas2; - } else { - memset(rc->areas[i], 0, PAGE_SIZE << page_order); - } + rc->active_entries = (int*)kmalloc(nr_areas * sizeof(int), GFP_KERNEL); + if(!rc->active_entries) + goto fail_malloc_active_entries; + memset(rc->active_entries, 0, nr_areas * sizeof(int)); + rc->active_pages = (int*)kmalloc(nr_areas * sizeof(int), GFP_KERNEL); + if(!rc->active_pages) + goto fail_malloc_active_pages; + memset(rc->active_pages, 0, nr_areas * sizeof(int)); + if((mode == ALL_AREAS) && (pages_per_area != 0)){ + rc->areas = debug_areas_alloc(pages_per_area, nr_areas); + if(!rc->areas) + goto fail_malloc_areas; + } else { + rc->areas = NULL; } /* initialize members */ spin_lock_init(&rc->lock); - rc->page_order = page_order; - rc->nr_areas = nr_areas; - rc->active_area = 0; - rc->level = DEBUG_DEFAULT_LEVEL; - rc->buf_size = buf_size; - rc->entry_size = sizeof(debug_entry_t) + buf_size; - strlcpy(rc->name, name, sizeof(rc->name)); + rc->pages_per_area = pages_per_area; + rc->nr_areas = nr_areas; + rc->active_area = 0; + rc->level = level; + rc->buf_size = buf_size; + rc->entry_size = sizeof(debug_entry_t) + buf_size; + strlcpy(rc->name, name, sizeof(rc->name)-1); memset(rc->views, 0, DEBUG_MAX_VIEWS * sizeof(struct debug_view *)); -#ifdef CONFIG_PROC_FS - memset(rc->proc_entries, 0 ,DEBUG_MAX_VIEWS * - sizeof(struct proc_dir_entry*)); -#endif /* CONFIG_PROC_FS */ + memset(rc->debugfs_entries, 0 ,DEBUG_MAX_VIEWS * + sizeof(struct dentry*)); atomic_set(&(rc->ref_count), 0); return rc; -fail_malloc_areas2: - kfree(rc->areas); fail_malloc_areas: - kfree(rc->active_entry); -fail_malloc_active_entry: + kfree(rc->active_pages); +fail_malloc_active_pages: + kfree(rc->active_entries); +fail_malloc_active_entries: kfree(rc); fail_malloc_rc: return NULL; } /* - * debug_info_free - * - free memory debug-info + * debug_areas_free + * - free all debug areas */ -static void debug_info_free(debug_info_t* db_info){ - int i; +static void +debug_areas_free(debug_info_t* db_info) +{ + int i,j; + + if(!db_info->areas) + return; for (i = 0; i < db_info->nr_areas; i++) { - free_pages((unsigned long) db_info->areas[i], - db_info->page_order); + for(j = 0; j < db_info->pages_per_area; j++) { + kfree(db_info->areas[i][j]); + } + kfree(db_info->areas[i]); } kfree(db_info->areas); - kfree(db_info->active_entry); + db_info->areas = NULL; +} + +/* + * debug_info_free + * - free memory debug-info + */ + +static void +debug_info_free(debug_info_t* db_info){ + debug_areas_free(db_info); + kfree(db_info->active_entries); + kfree(db_info->active_pages); kfree(db_info); } @@ -249,21 +332,22 @@ static void debug_info_free(debug_info_t* db_info){ * - create new debug-info */ -static debug_info_t* debug_info_create(char *name, int page_order, - int nr_areas, int buf_size) +static debug_info_t* +debug_info_create(char *name, int pages_per_area, int nr_areas, int buf_size) { debug_info_t* rc; - rc = debug_info_alloc(name, page_order, nr_areas, buf_size); + rc = debug_info_alloc(name, pages_per_area, nr_areas, buf_size, + DEBUG_DEFAULT_LEVEL, ALL_AREAS); if(!rc) goto out; - - /* create proc rood directory */ - rc->proc_root_entry = proc_mkdir(rc->name, debug_proc_root_entry); + /* create root directory */ + rc->debugfs_root_entry = debugfs_create_dir(rc->name, + debug_debugfs_root_entry); /* append new element to linked list */ - if (debug_area_first == NULL) { + if (!debug_area_first) { /* first element in list */ debug_area_first = rc; rc->prev = NULL; @@ -285,17 +369,21 @@ out: * - copy debug-info */ -static debug_info_t* debug_info_copy(debug_info_t* in) +static debug_info_t* +debug_info_copy(debug_info_t* in, int mode) { - int i; + int i,j; debug_info_t* rc; - rc = debug_info_alloc(in->name, in->page_order, - in->nr_areas, in->buf_size); - if(!rc) + + rc = debug_info_alloc(in->name, in->pages_per_area, in->nr_areas, + in->buf_size, in->level, mode); + if(!rc || (mode == NO_AREAS)) goto out; for(i = 0; i < in->nr_areas; i++){ - memcpy(rc->areas[i],in->areas[i], PAGE_SIZE << in->page_order); + for(j = 0; j < in->pages_per_area; j++) { + memcpy(rc->areas[i][j], in->areas[i][j],PAGE_SIZE); + } } out: return rc; @@ -306,7 +394,8 @@ out: * - increments reference count for debug-info */ -static void debug_info_get(debug_info_t * db_info) +static void +debug_info_get(debug_info_t * db_info) { if (db_info) atomic_inc(&db_info->ref_count); @@ -317,29 +406,20 @@ static void debug_info_get(debug_info_t * db_info) * - decreases reference count for debug-info and frees it if necessary */ -static void debug_info_put(debug_info_t *db_info) +static void +debug_info_put(debug_info_t *db_info) { int i; if (!db_info) return; if (atomic_dec_and_test(&db_info->ref_count)) { -#ifdef DEBUG - printk(KERN_INFO "debug: freeing debug area %p (%s)\n", - db_info, db_info->name); -#endif for (i = 0; i < DEBUG_MAX_VIEWS; i++) { - if (db_info->views[i] == NULL) + if (!db_info->views[i]) continue; -#ifdef CONFIG_PROC_FS - remove_proc_entry(db_info->proc_entries[i]->name, - db_info->proc_root_entry); -#endif + debugfs_remove(db_info->debugfs_entries[i]); } -#ifdef CONFIG_PROC_FS - remove_proc_entry(db_info->proc_root_entry->name, - debug_proc_root_entry); -#endif + debugfs_remove(db_info->debugfs_root_entry); if(db_info == debug_area_first) debug_area_first = db_info->next; if(db_info == debug_area_last) @@ -355,9 +435,9 @@ static void debug_info_put(debug_info_t *db_info) * - format one debug entry and return size of formated data */ -static int debug_format_entry(file_private_info_t *p_info) +static int +debug_format_entry(file_private_info_t *p_info) { - debug_info_t *id_org = p_info->debug_info_org; debug_info_t *id_snap = p_info->debug_info_snap; struct debug_view *view = p_info->view; debug_entry_t *act_entry; @@ -365,22 +445,23 @@ static int debug_format_entry(file_private_info_t *p_info) if(p_info->act_entry == DEBUG_PROLOG_ENTRY){ /* print prolog */ if (view->prolog_proc) - len += view->prolog_proc(id_org, view,p_info->temp_buf); + len += view->prolog_proc(id_snap,view,p_info->temp_buf); goto out; } - - act_entry = (debug_entry_t *) ((char*)id_snap->areas[p_info->act_area] + - p_info->act_entry); + if (!id_snap->areas) /* this is true, if we have a prolog only view */ + goto out; /* or if 'pages_per_area' is 0 */ + act_entry = (debug_entry_t *) ((char*)id_snap->areas[p_info->act_area] + [p_info->act_page] + p_info->act_entry); if (act_entry->id.stck == 0LL) goto out; /* empty entry */ if (view->header_proc) - len += view->header_proc(id_org, view, p_info->act_area, + len += view->header_proc(id_snap, view, p_info->act_area, act_entry, p_info->temp_buf + len); if (view->format_proc) - len += view->format_proc(id_org, view, p_info->temp_buf + len, + len += view->format_proc(id_snap, view, p_info->temp_buf + len, DEBUG_DATA(act_entry)); - out: +out: return len; } @@ -389,20 +470,30 @@ static int debug_format_entry(file_private_info_t *p_info) * - goto next entry in p_info */ -extern inline int debug_next_entry(file_private_info_t *p_info) +extern inline int +debug_next_entry(file_private_info_t *p_info) { - debug_info_t *id = p_info->debug_info_snap; + debug_info_t *id; + + id = p_info->debug_info_snap; if(p_info->act_entry == DEBUG_PROLOG_ENTRY){ p_info->act_entry = 0; + p_info->act_page = 0; goto out; } - if ((p_info->act_entry += id->entry_size) - > ((PAGE_SIZE << (id->page_order)) - - id->entry_size)){ - - /* next area */ + if(!id->areas) + return 1; + p_info->act_entry += id->entry_size; + /* switch to next page, if we reached the end of the page */ + if (p_info->act_entry > (PAGE_SIZE - id->entry_size)){ + /* next page */ p_info->act_entry = 0; - p_info->act_area++; + p_info->act_page += 1; + if((p_info->act_page % id->pages_per_area) == 0) { + /* next area */ + p_info->act_area++; + p_info->act_page=0; + } if(p_info->act_area >= id->nr_areas) return 1; } @@ -416,13 +507,14 @@ out: * - copies formated debug entries to the user buffer */ -static ssize_t debug_output(struct file *file, /* file descriptor */ - char __user *user_buf, /* user buffer */ - size_t len, /* length of buffer */ - loff_t *offset) /* offset in the file */ +static ssize_t +debug_output(struct file *file, /* file descriptor */ + char __user *user_buf, /* user buffer */ + size_t len, /* length of buffer */ + loff_t *offset) /* offset in the file */ { size_t count = 0; - size_t entry_offset, size = 0; + size_t entry_offset; file_private_info_t *p_info; p_info = ((file_private_info_t *) file->private_data); @@ -430,27 +522,33 @@ static ssize_t debug_output(struct file *file, /* file descriptor */ return -EPIPE; if(p_info->act_area >= p_info->debug_info_snap->nr_areas) return 0; - entry_offset = p_info->act_entry_offset; - while(count < len){ - size = debug_format_entry(p_info); - size = min((len - count), (size - entry_offset)); - - if(size){ - if (copy_to_user(user_buf + count, - p_info->temp_buf + entry_offset, size)) - return -EFAULT; + int formatted_line_size; + int formatted_line_residue; + int user_buf_residue; + size_t copy_size; + + formatted_line_size = debug_format_entry(p_info); + formatted_line_residue = formatted_line_size - entry_offset; + user_buf_residue = len-count; + copy_size = min(user_buf_residue, formatted_line_residue); + if(copy_size){ + if (copy_to_user(user_buf + count, p_info->temp_buf + + entry_offset, copy_size)) + return -EFAULT; + count += copy_size; + entry_offset += copy_size; } - count += size; - entry_offset = 0; - if(count != len) - if(debug_next_entry(p_info)) + if(copy_size == formatted_line_residue){ + entry_offset = 0; + if(debug_next_entry(p_info)) goto out; + } } out: p_info->offset = *offset + count; - p_info->act_entry_offset = size; + p_info->act_entry_offset = entry_offset; *offset = p_info->offset; return count; } @@ -461,9 +559,9 @@ out: * - calls input function of view */ -static ssize_t debug_input(struct file *file, - const char __user *user_buf, size_t length, - loff_t *offset) +static ssize_t +debug_input(struct file *file, const char __user *user_buf, size_t length, + loff_t *offset) { int rc = 0; file_private_info_t *p_info; @@ -487,26 +585,23 @@ static ssize_t debug_input(struct file *file, * handle */ -static int debug_open(struct inode *inode, struct file *file) +static int +debug_open(struct inode *inode, struct file *file) { int i = 0, rc = 0; file_private_info_t *p_info; debug_info_t *debug_info, *debug_info_snapshot; -#ifdef DEBUG - printk("debug_open\n"); -#endif down(&debug_lock); /* find debug log and view */ - debug_info = debug_area_first; while(debug_info != NULL){ for (i = 0; i < DEBUG_MAX_VIEWS; i++) { - if (debug_info->views[i] == NULL) + if (!debug_info->views[i]) continue; - else if (debug_info->proc_entries[i] == - PDE(file->f_dentry->d_inode)) { + else if (debug_info->debugfs_entries[i] == + file->f_dentry) { goto found; /* found view ! */ } } @@ -516,41 +611,42 @@ static int debug_open(struct inode *inode, struct file *file) rc = -EINVAL; goto out; - found: +found: - /* make snapshot of current debug areas to get it consistent */ + /* Make snapshot of current debug areas to get it consistent. */ + /* To copy all the areas is only needed, if we have a view which */ + /* formats the debug areas. */ - debug_info_snapshot = debug_info_copy(debug_info); + if(!debug_info->views[i]->format_proc && + !debug_info->views[i]->header_proc){ + debug_info_snapshot = debug_info_copy(debug_info, NO_AREAS); + } else { + debug_info_snapshot = debug_info_copy(debug_info, ALL_AREAS); + } if(!debug_info_snapshot){ -#ifdef DEBUG - printk(KERN_ERR "debug_open: debug_info_copy failed (out of mem)\n"); -#endif rc = -ENOMEM; goto out; } - - if ((file->private_data = - kmalloc(sizeof(file_private_info_t), GFP_ATOMIC)) == 0) { -#ifdef DEBUG - printk(KERN_ERR "debug_open: kmalloc failed\n"); -#endif - debug_info_free(debug_info_snapshot); + p_info = (file_private_info_t *) kmalloc(sizeof(file_private_info_t), + GFP_KERNEL); + if(!p_info){ + if(debug_info_snapshot) + debug_info_free(debug_info_snapshot); rc = -ENOMEM; goto out; } - p_info = (file_private_info_t *) file->private_data; p_info->offset = 0; p_info->debug_info_snap = debug_info_snapshot; p_info->debug_info_org = debug_info; p_info->view = debug_info->views[i]; p_info->act_area = 0; + p_info->act_page = 0; p_info->act_entry = DEBUG_PROLOG_ENTRY; p_info->act_entry_offset = 0; - + file->private_data = p_info; debug_info_get(debug_info); - - out: +out: up(&debug_lock); return rc; } @@ -561,14 +657,13 @@ static int debug_open(struct inode *inode, struct file *file) * - deletes private_data area of the file handle */ -static int debug_close(struct inode *inode, struct file *file) +static int +debug_close(struct inode *inode, struct file *file) { file_private_info_t *p_info; -#ifdef DEBUG - printk("debug_close\n"); -#endif p_info = (file_private_info_t *) file->private_data; - debug_info_free(p_info->debug_info_snap); + if(p_info->debug_info_snap) + debug_info_free(p_info->debug_info_snap); debug_info_put(p_info->debug_info_org); kfree(file->private_data); return 0; /* success */ @@ -580,8 +675,8 @@ static int debug_close(struct inode *inode, struct file *file) * - returns handle for debug area */ -debug_info_t *debug_register - (char *name, int page_order, int nr_areas, int buf_size) +debug_info_t* +debug_register (char *name, int pages_per_area, int nr_areas, int buf_size) { debug_info_t *rc = NULL; @@ -591,18 +686,14 @@ debug_info_t *debug_register /* create new debug_info */ - rc = debug_info_create(name, page_order, nr_areas, buf_size); + rc = debug_info_create(name, pages_per_area, nr_areas, buf_size); if(!rc) goto out; debug_register_view(rc, &debug_level_view); debug_register_view(rc, &debug_flush_view); -#ifdef DEBUG - printk(KERN_INFO - "debug: reserved %d areas of %d pages for debugging %s\n", - nr_areas, 1 << page_order, rc->name); -#endif - out: - if (rc == NULL){ + debug_register_view(rc, &debug_pages_view); +out: + if (!rc){ printk(KERN_ERR "debug: debug_register failed for %s\n",name); } up(&debug_lock); @@ -614,27 +705,65 @@ debug_info_t *debug_register * - give back debug area */ -void debug_unregister(debug_info_t * id) +void +debug_unregister(debug_info_t * id) { if (!id) goto out; down(&debug_lock); -#ifdef DEBUG - printk(KERN_INFO "debug: unregistering %s\n", id->name); -#endif debug_info_put(id); up(&debug_lock); - out: +out: return; } +/* + * debug_set_size: + * - set area size (number of pages) and number of areas + */ +static int +debug_set_size(debug_info_t* id, int nr_areas, int pages_per_area) +{ + unsigned long flags; + debug_entry_t *** new_areas; + int rc=0; + + if(!id || (nr_areas <= 0) || (pages_per_area < 0)) + return -EINVAL; + if(pages_per_area > 0){ + new_areas = debug_areas_alloc(pages_per_area, nr_areas); + if(!new_areas) { + printk(KERN_WARNING "debug: could not allocate memory "\ + "for pagenumber: %i\n",pages_per_area); + rc = -ENOMEM; + goto out; + } + } else { + new_areas = NULL; + } + spin_lock_irqsave(&id->lock,flags); + debug_areas_free(id); + id->areas = new_areas; + id->nr_areas = nr_areas; + id->pages_per_area = pages_per_area; + id->active_area = 0; + memset(id->active_entries,0,sizeof(int)*id->nr_areas); + memset(id->active_pages, 0, sizeof(int)*id->nr_areas); + spin_unlock_irqrestore(&id->lock,flags); + printk(KERN_INFO "debug: %s: set new size (%i pages)\n"\ + ,id->name, pages_per_area); +out: + return rc; +} + /* * debug_set_level: * - set actual debug level */ -void debug_set_level(debug_info_t* id, int new_level) +void +debug_set_level(debug_info_t* id, int new_level) { unsigned long flags; if(!id) @@ -649,10 +778,6 @@ void debug_set_level(debug_info_t* id, int new_level) id->name, new_level, 0, DEBUG_MAX_LEVEL); } else { id->level = new_level; -#ifdef DEBUG - printk(KERN_INFO - "debug: %s: new level %i\n",id->name,id->level); -#endif } spin_unlock_irqrestore(&id->lock,flags); } @@ -663,11 +788,16 @@ void debug_set_level(debug_info_t* id, int new_level) * - set active entry to next in the ring buffer */ -extern inline void proceed_active_entry(debug_info_t * id) +extern inline void +proceed_active_entry(debug_info_t * id) { - if ((id->active_entry[id->active_area] += id->entry_size) - > ((PAGE_SIZE << (id->page_order)) - id->entry_size)) - id->active_entry[id->active_area] = 0; + if ((id->active_entries[id->active_area] += id->entry_size) + > (PAGE_SIZE - id->entry_size)){ + id->active_entries[id->active_area] = 0; + id->active_pages[id->active_area] = + (id->active_pages[id->active_area] + 1) % + id->pages_per_area; + } } /* @@ -675,7 +805,8 @@ extern inline void proceed_active_entry(debug_info_t * id) * - set active area to next in the ring buffer */ -extern inline void proceed_active_area(debug_info_t * id) +extern inline void +proceed_active_area(debug_info_t * id) { id->active_area++; id->active_area = id->active_area % id->nr_areas; @@ -685,10 +816,12 @@ extern inline void proceed_active_area(debug_info_t * id) * get_active_entry: */ -extern inline debug_entry_t *get_active_entry(debug_info_t * id) +extern inline debug_entry_t* +get_active_entry(debug_info_t * id) { - return (debug_entry_t *) ((char *) id->areas[id->active_area] + - id->active_entry[id->active_area]); + return (debug_entry_t *) (((char *) id->areas[id->active_area] + [id->active_pages[id->active_area]]) + + id->active_entries[id->active_area]); } /* @@ -696,8 +829,9 @@ extern inline debug_entry_t *get_active_entry(debug_info_t * id) * - set timestamp, caller address, cpu number etc. */ -extern inline void debug_finish_entry(debug_info_t * id, debug_entry_t* active, - int level, int exception) +extern inline void +debug_finish_entry(debug_info_t * id, debug_entry_t* active, int level, + int exception) { STCK(active->id.stck); active->id.fields.cpuid = smp_processor_id(); @@ -721,7 +855,8 @@ static int debug_active=1; * always allow read, allow write only if debug_stoppable is set or * if debug_active is already off */ -static int s390dbf_procactive(ctl_table *table, int write, struct file *filp, +static int +s390dbf_procactive(ctl_table *table, int write, struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos) { if (!write || debug_stoppable || !debug_active) @@ -766,7 +901,8 @@ static struct ctl_table s390dbf_dir_table[] = { struct ctl_table_header *s390dbf_sysctl_header; -void debug_stop_all(void) +void +debug_stop_all(void) { if (debug_stoppable) debug_active = 0; @@ -778,13 +914,13 @@ void debug_stop_all(void) * - write debug entry with given size */ -debug_entry_t *debug_event_common(debug_info_t * id, int level, const void *buf, - int len) +debug_entry_t* +debug_event_common(debug_info_t * id, int level, const void *buf, int len) { unsigned long flags; debug_entry_t *active; - if (!debug_active) + if (!debug_active || !id->areas) return NULL; spin_lock_irqsave(&id->lock, flags); active = get_active_entry(id); @@ -801,13 +937,13 @@ debug_entry_t *debug_event_common(debug_info_t * id, int level, const void *buf, * - write debug entry with given size and switch to next debug area */ -debug_entry_t *debug_exception_common(debug_info_t * id, int level, - const void *buf, int len) +debug_entry_t +*debug_exception_common(debug_info_t * id, int level, const void *buf, int len) { unsigned long flags; debug_entry_t *active; - if (!debug_active) + if (!debug_active || !id->areas) return NULL; spin_lock_irqsave(&id->lock, flags); active = get_active_entry(id); @@ -823,7 +959,8 @@ debug_entry_t *debug_exception_common(debug_info_t * id, int level, * counts arguments in format string for sprintf view */ -extern inline int debug_count_numargs(char *string) +extern inline int +debug_count_numargs(char *string) { int numargs=0; @@ -838,8 +975,8 @@ extern inline int debug_count_numargs(char *string) * debug_sprintf_event: */ -debug_entry_t *debug_sprintf_event(debug_info_t* id, - int level,char *string,...) +debug_entry_t* +debug_sprintf_event(debug_info_t* id, int level,char *string,...) { va_list ap; int numargs,idx; @@ -849,7 +986,7 @@ debug_entry_t *debug_sprintf_event(debug_info_t* id, if((!id) || (level > id->level)) return NULL; - if (!debug_active) + if (!debug_active || !id->areas) return NULL; numargs=debug_count_numargs(string); @@ -871,8 +1008,8 @@ debug_entry_t *debug_sprintf_event(debug_info_t* id, * debug_sprintf_exception: */ -debug_entry_t *debug_sprintf_exception(debug_info_t* id, - int level,char *string,...) +debug_entry_t* +debug_sprintf_exception(debug_info_t* id, int level,char *string,...) { va_list ap; int numargs,idx; @@ -882,7 +1019,7 @@ debug_entry_t *debug_sprintf_exception(debug_info_t* id, if((!id) || (level > id->level)) return NULL; - if (!debug_active) + if (!debug_active || !id->areas) return NULL; numargs=debug_count_numargs(string); @@ -906,15 +1043,14 @@ debug_entry_t *debug_sprintf_exception(debug_info_t* id, * - is called exactly once to initialize the debug feature */ -static int __init debug_init(void) +static int +__init debug_init(void) { int rc = 0; s390dbf_sysctl_header = register_sysctl_table(s390dbf_dir_table, 1); down(&debug_lock); -#ifdef CONFIG_PROC_FS - debug_proc_root_entry = proc_mkdir(DEBUG_DIR_ROOT, NULL); -#endif /* CONFIG_PROC_FS */ + debug_debugfs_root_entry = debugfs_create_dir(DEBUG_DIR_ROOT,NULL); printk(KERN_INFO "debug: Initialization complete\n"); initialized = 1; up(&debug_lock); @@ -926,13 +1062,14 @@ static int __init debug_init(void) * debug_register_view: */ -int debug_register_view(debug_info_t * id, struct debug_view *view) +int +debug_register_view(debug_info_t * id, struct debug_view *view) { int rc = 0; int i; unsigned long flags; mode_t mode = S_IFREG; - struct proc_dir_entry *pde; + struct dentry *pde; if (!id) goto out; @@ -940,16 +1077,17 @@ int debug_register_view(debug_info_t * id, struct debug_view *view) mode |= S_IRUSR; if (view->input_proc) mode |= S_IWUSR; - pde = create_proc_entry(view->name, mode, id->proc_root_entry); + pde = debugfs_create_file(view->name, mode, id->debugfs_root_entry, + NULL, &debug_file_ops); if (!pde){ - printk(KERN_WARNING "debug: create_proc_entry() failed! Cannot register view %s/%s\n", id->name,view->name); + printk(KERN_WARNING "debug: debugfs_create_file() failed!"\ + " Cannot register view %s/%s\n", id->name,view->name); rc = -1; goto out; } - spin_lock_irqsave(&id->lock, flags); for (i = 0; i < DEBUG_MAX_VIEWS; i++) { - if (id->views[i] == NULL) + if (!id->views[i]) break; } if (i == DEBUG_MAX_VIEWS) { @@ -957,16 +1095,14 @@ int debug_register_view(debug_info_t * id, struct debug_view *view) id->name,view->name); printk(KERN_WARNING "debug: maximum number of views reached (%i)!\n", i); - remove_proc_entry(pde->name, id->proc_root_entry); + debugfs_remove(pde); rc = -1; - } - else { + } else { id->views[i] = view; - pde->proc_fops = &debug_file_ops; - id->proc_entries[i] = pde; + id->debugfs_entries[i] = pde; } spin_unlock_irqrestore(&id->lock, flags); - out: +out: return rc; } @@ -974,7 +1110,8 @@ int debug_register_view(debug_info_t * id, struct debug_view *view) * debug_unregister_view: */ -int debug_unregister_view(debug_info_t * id, struct debug_view *view) +int +debug_unregister_view(debug_info_t * id, struct debug_view *view) { int rc = 0; int i; @@ -990,15 +1127,46 @@ int debug_unregister_view(debug_info_t * id, struct debug_view *view) if (i == DEBUG_MAX_VIEWS) rc = -1; else { -#ifdef CONFIG_PROC_FS - remove_proc_entry(id->proc_entries[i]->name, - id->proc_root_entry); -#endif + debugfs_remove(id->debugfs_entries[i]); id->views[i] = NULL; rc = 0; } spin_unlock_irqrestore(&id->lock, flags); - out: +out: + return rc; +} + +static inline char * +debug_get_user_string(const char __user *user_buf, size_t user_len) +{ + char* buffer; + + buffer = kmalloc(user_len + 1, GFP_KERNEL); + if (!buffer) + return ERR_PTR(-ENOMEM); + if (copy_from_user(buffer, user_buf, user_len) != 0) { + kfree(buffer); + return ERR_PTR(-EFAULT); + } + /* got the string, now strip linefeed. */ + if (buffer[user_len - 1] == '\n') + buffer[user_len - 1] = 0; + else + buffer[user_len] = 0; + return buffer; +} + +static inline int +debug_get_uint(char *buf) +{ + int rc; + + for(; isspace(*buf); buf++); + rc = simple_strtoul(buf, &buf, 10); + if(*buf){ + printk("debug: no integer specified!\n"); + rc = -EINVAL; + } return rc; } @@ -1011,13 +1179,69 @@ int debug_unregister_view(debug_info_t * id, struct debug_view *view) * prints out actual debug level */ -static int debug_prolog_level_fn(debug_info_t * id, +static int +debug_prolog_pages_fn(debug_info_t * id, struct debug_view *view, char *out_buf) +{ + return sprintf(out_buf, "%i\n", id->pages_per_area); +} + +/* + * reads new size (number of pages per debug area) + */ + +static int +debug_input_pages_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_len, loff_t * offset) +{ + char *str; + int rc,new_pages; + + if (user_len > 0x10000) + user_len = 0x10000; + if (*offset != 0){ + rc = -EPIPE; + goto out; + } + str = debug_get_user_string(user_buf,user_len); + if(IS_ERR(str)){ + rc = PTR_ERR(str); + goto out; + } + new_pages = debug_get_uint(str); + if(new_pages < 0){ + rc = -EINVAL; + goto free_str; + } + rc = debug_set_size(id,id->nr_areas, new_pages); + if(rc != 0){ + rc = -EINVAL; + goto free_str; + } + rc = user_len; +free_str: + kfree(str); +out: + *offset += user_len; + return rc; /* number of input characters */ +} + +/* + * prints out actual debug level + */ + +static int +debug_prolog_level_fn(debug_info_t * id, struct debug_view *view, char *out_buf) { int rc = 0; - if(id->level == -1) rc = sprintf(out_buf,"-\n"); - else rc = sprintf(out_buf, "%i\n", id->level); + if(id->level == DEBUG_OFF_LEVEL) { + rc = sprintf(out_buf,"-\n"); + } + else { + rc = sprintf(out_buf, "%i\n", id->level); + } return rc; } @@ -1025,30 +1249,43 @@ static int debug_prolog_level_fn(debug_info_t * id, * reads new debug level */ -static int debug_input_level_fn(debug_info_t * id, struct debug_view *view, - struct file *file, const char __user *user_buf, - size_t in_buf_size, loff_t * offset) +static int +debug_input_level_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_len, loff_t * offset) { - char input_buf[1]; - int rc = in_buf_size; + char *str; + int rc,new_level; - if (*offset != 0) + if (user_len > 0x10000) + user_len = 0x10000; + if (*offset != 0){ + rc = -EPIPE; goto out; - if (copy_from_user(input_buf, user_buf, 1)){ - rc = -EFAULT; + } + str = debug_get_user_string(user_buf,user_len); + if(IS_ERR(str)){ + rc = PTR_ERR(str); goto out; } - if (isdigit(input_buf[0])) { - int new_level = ((int) input_buf[0] - (int) '0'); - debug_set_level(id, new_level); - } else if(input_buf[0] == '-') { + if(str[0] == '-'){ debug_set_level(id, DEBUG_OFF_LEVEL); + rc = user_len; + goto free_str; } else { - printk(KERN_INFO "debug: level `%c` is not valid\n", - input_buf[0]); + new_level = debug_get_uint(str); } - out: - *offset += in_buf_size; + if(new_level < 0) { + printk(KERN_INFO "debug: level `%s` is not valid\n", str); + rc = -EINVAL; + } else { + debug_set_level(id, new_level); + rc = user_len; + } +free_str: + kfree(str); +out: + *offset += user_len; return rc; /* number of input characters */ } @@ -1057,29 +1294,36 @@ static int debug_input_level_fn(debug_info_t * id, struct debug_view *view, * flushes debug areas */ -void debug_flush(debug_info_t* id, int area) +void +debug_flush(debug_info_t* id, int area) { unsigned long flags; - int i; + int i,j; - if(!id) + if(!id || !id->areas) return; spin_lock_irqsave(&id->lock,flags); if(area == DEBUG_FLUSH_ALL){ id->active_area = 0; - memset(id->active_entry, 0, id->nr_areas * sizeof(int)); - for (i = 0; i < id->nr_areas; i++) - memset(id->areas[i], 0, PAGE_SIZE << id->page_order); + memset(id->active_entries, 0, id->nr_areas * sizeof(int)); + for (i = 0; i < id->nr_areas; i++) { + id->active_pages[i] = 0; + for(j = 0; j < id->pages_per_area; j++) { + memset(id->areas[i][j], 0, PAGE_SIZE); + } + } printk(KERN_INFO "debug: %s: all areas flushed\n",id->name); } else if(area >= 0 && area < id->nr_areas) { - id->active_entry[area] = 0; - memset(id->areas[area], 0, PAGE_SIZE << id->page_order); - printk(KERN_INFO - "debug: %s: area %i has been flushed\n", + id->active_entries[area] = 0; + id->active_pages[area] = 0; + for(i = 0; i < id->pages_per_area; i++) { + memset(id->areas[area][i],0,PAGE_SIZE); + } + printk(KERN_INFO "debug: %s: area %i has been flushed\n", id->name, area); } else { printk(KERN_INFO - "debug: %s: area %i cannot be flushed (range: %i - %i)\n", + "debug: %s: area %i cannot be flushed (range: %i - %i)\n", id->name, area, 0, id->nr_areas-1); } spin_unlock_irqrestore(&id->lock,flags); @@ -1089,15 +1333,20 @@ void debug_flush(debug_info_t* id, int area) * view function: flushes debug areas */ -static int debug_input_flush_fn(debug_info_t * id, struct debug_view *view, - struct file *file, const char __user *user_buf, - size_t in_buf_size, loff_t * offset) +static int +debug_input_flush_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_len, loff_t * offset) { char input_buf[1]; - int rc = in_buf_size; - - if (*offset != 0) + int rc = user_len; + + if (user_len > 0x10000) + user_len = 0x10000; + if (*offset != 0){ + rc = -EPIPE; goto out; + } if (copy_from_user(input_buf, user_buf, 1)){ rc = -EFAULT; goto out; @@ -1114,8 +1363,8 @@ static int debug_input_flush_fn(debug_info_t * id, struct debug_view *view, printk(KERN_INFO "debug: area `%c` is not valid\n", input_buf[0]); - out: - *offset += in_buf_size; +out: + *offset += user_len; return rc; /* number of input characters */ } @@ -1123,8 +1372,9 @@ static int debug_input_flush_fn(debug_info_t * id, struct debug_view *view, * prints debug header in raw format */ -int debug_raw_header_fn(debug_info_t * id, struct debug_view *view, - int area, debug_entry_t * entry, char *out_buf) +static int +debug_raw_header_fn(debug_info_t * id, struct debug_view *view, + int area, debug_entry_t * entry, char *out_buf) { int rc; @@ -1137,7 +1387,8 @@ int debug_raw_header_fn(debug_info_t * id, struct debug_view *view, * prints debug data in raw format */ -static int debug_raw_format_fn(debug_info_t * id, struct debug_view *view, +static int +debug_raw_format_fn(debug_info_t * id, struct debug_view *view, char *out_buf, const char *in_buf) { int rc; @@ -1151,8 +1402,9 @@ static int debug_raw_format_fn(debug_info_t * id, struct debug_view *view, * prints debug data in hex/ascii format */ -static int debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view, - char *out_buf, const char *in_buf) +static int +debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view, + char *out_buf, const char *in_buf) { int i, rc = 0; @@ -1176,7 +1428,8 @@ static int debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view, * prints header for debug entry */ -int debug_dflt_header_fn(debug_info_t * id, struct debug_view *view, +int +debug_dflt_header_fn(debug_info_t * id, struct debug_view *view, int area, debug_entry_t * entry, char *out_buf) { struct timeval time_val; @@ -1210,8 +1463,9 @@ int debug_dflt_header_fn(debug_info_t * id, struct debug_view *view, #define DEBUG_SPRINTF_MAX_ARGS 10 -int debug_sprintf_format_fn(debug_info_t * id, struct debug_view *view, - char *out_buf, debug_sprintf_entry_t *curr_event) +static int +debug_sprintf_format_fn(debug_info_t * id, struct debug_view *view, + char *out_buf, debug_sprintf_entry_t *curr_event) { int num_longs, num_used_args = 0,i, rc = 0; int index[DEBUG_SPRINTF_MAX_ARGS]; @@ -1251,14 +1505,10 @@ out: /* * clean up module */ -void __exit debug_exit(void) +void +__exit debug_exit(void) { -#ifdef DEBUG - printk("debug_cleanup_module: \n"); -#endif -#ifdef CONFIG_PROC_FS - remove_proc_entry(debug_proc_root_entry->name, NULL); -#endif /* CONFIG_PROC_FS */ + debugfs_remove(debug_debugfs_root_entry); unregister_sysctl_table(s390dbf_sysctl_header); return; } @@ -1266,7 +1516,7 @@ void __exit debug_exit(void) /* * module definitions */ -core_initcall(debug_init); +postcore_initcall(debug_init); module_exit(debug_exit); MODULE_LICENSE("GPL"); diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index 3e39508bd92..6527ff6f470 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -176,7 +176,7 @@ dasd_state_known_to_basic(struct dasd_device * device) return rc; /* register 'device' debug area, used for all DBF_DEV_XXX calls */ - device->debug_area = debug_register(device->cdev->dev.bus_id, 0, 2, + device->debug_area = debug_register(device->cdev->dev.bus_id, 1, 2, 8 * sizeof (long)); debug_register_view(device->debug_area, &debug_sprintf_view); debug_set_level(device->debug_area, DBF_EMERG); @@ -1981,7 +1981,7 @@ dasd_init(void) init_waitqueue_head(&dasd_init_waitq); /* register 'common' DASD debug area, used for all DBF_XXX calls */ - dasd_debug_area = debug_register("dasd", 0, 2, 8 * sizeof (long)); + dasd_debug_area = debug_register("dasd", 1, 2, 8 * sizeof (long)); if (dasd_debug_area == NULL) { rc = -ENOMEM; goto failed; diff --git a/drivers/s390/block/dasd_proc.c b/drivers/s390/block/dasd_proc.c index d7f19745911..43c34f8c5e6 100644 --- a/drivers/s390/block/dasd_proc.c +++ b/drivers/s390/block/dasd_proc.c @@ -9,13 +9,14 @@ * * /proc interface for the dasd driver. * - * $Revision: 1.31 $ + * $Revision: 1.32 $ */ #include #include #include #include +#include #include #include diff --git a/drivers/s390/char/tape_34xx.c b/drivers/s390/char/tape_34xx.c index 480ec87976f..20be88e91fa 100644 --- a/drivers/s390/char/tape_34xx.c +++ b/drivers/s390/char/tape_34xx.c @@ -1351,13 +1351,13 @@ tape_34xx_init (void) { int rc; - TAPE_DBF_AREA = debug_register ( "tape_34xx", 1, 2, 4*sizeof(long)); + TAPE_DBF_AREA = debug_register ( "tape_34xx", 2, 2, 4*sizeof(long)); debug_register_view(TAPE_DBF_AREA, &debug_sprintf_view); #ifdef DBF_LIKE_HELL debug_set_level(TAPE_DBF_AREA, 6); #endif - DBF_EVENT(3, "34xx init: $Revision: 1.21 $\n"); + DBF_EVENT(3, "34xx init: $Revision: 1.23 $\n"); /* Register driver for 3480/3490 tapes. */ rc = ccw_driver_register(&tape_34xx_driver); if (rc) @@ -1378,7 +1378,7 @@ tape_34xx_exit(void) MODULE_DEVICE_TABLE(ccw, tape_34xx_ids); MODULE_AUTHOR("(C) 2001-2002 IBM Deutschland Entwicklung GmbH"); MODULE_DESCRIPTION("Linux on zSeries channel attached 3480 tape " - "device driver ($Revision: 1.21 $)"); + "device driver ($Revision: 1.23 $)"); MODULE_LICENSE("GPL"); module_init(tape_34xx_init); diff --git a/drivers/s390/char/tape_core.c b/drivers/s390/char/tape_core.c index b4df4a515b1..0597aa0e27e 100644 --- a/drivers/s390/char/tape_core.c +++ b/drivers/s390/char/tape_core.c @@ -1186,7 +1186,7 @@ tape_mtop(struct tape_device *device, int mt_op, int mt_count) static int tape_init (void) { - TAPE_DBF_AREA = debug_register ( "tape", 1, 2, 4*sizeof(long)); + TAPE_DBF_AREA = debug_register ( "tape", 2, 2, 4*sizeof(long)); debug_register_view(TAPE_DBF_AREA, &debug_sprintf_view); #ifdef DBF_LIKE_HELL debug_set_level(TAPE_DBF_AREA, 6); diff --git a/drivers/s390/char/tape_proc.c b/drivers/s390/char/tape_proc.c index 801d17cca34..5fec0a10cc3 100644 --- a/drivers/s390/char/tape_proc.c +++ b/drivers/s390/char/tape_proc.c @@ -15,6 +15,7 @@ #include #include #include +#include #define TAPE_DBF_AREA tape_core_dbf diff --git a/drivers/s390/char/vmcp.c b/drivers/s390/char/vmcp.c index dfbbf235ca2..7f11a608a63 100644 --- a/drivers/s390/char/vmcp.c +++ b/drivers/s390/char/vmcp.c @@ -203,7 +203,7 @@ static int __init vmcp_init(void) else printk(KERN_WARNING "z/VM CP interface not loaded. Could not register misc device.\n"); - vmcp_debug = debug_register("vmcp", 0, 1, 240); + vmcp_debug = debug_register("vmcp", 1, 1, 240); debug_register_view(vmcp_debug, &debug_hex_ascii_view); return ret; } diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c index 1d9b3f18d8d..ea813bdce1d 100644 --- a/drivers/s390/cio/cio.c +++ b/drivers/s390/cio/cio.c @@ -1,7 +1,7 @@ /* * drivers/s390/cio/cio.c * S/390 common I/O routines -- low level i/o calls - * $Revision: 1.133 $ + * $Revision: 1.134 $ * * Copyright (C) 1999-2002 IBM Deutschland Entwicklung GmbH, * IBM Corporation @@ -63,17 +63,17 @@ __setup ("cio_msg=", cio_setup); static int __init cio_debug_init (void) { - cio_debug_msg_id = debug_register ("cio_msg", 4, 4, 16*sizeof (long)); + cio_debug_msg_id = debug_register ("cio_msg", 16, 4, 16*sizeof (long)); if (!cio_debug_msg_id) goto out_unregister; debug_register_view (cio_debug_msg_id, &debug_sprintf_view); debug_set_level (cio_debug_msg_id, 2); - cio_debug_trace_id = debug_register ("cio_trace", 4, 4, 8); + cio_debug_trace_id = debug_register ("cio_trace", 16, 4, 8); if (!cio_debug_trace_id) goto out_unregister; debug_register_view (cio_debug_trace_id, &debug_hex_ascii_view); debug_set_level (cio_debug_trace_id, 2); - cio_debug_crw_id = debug_register ("cio_crw", 2, 4, 16*sizeof (long)); + cio_debug_crw_id = debug_register ("cio_crw", 4, 4, 16*sizeof (long)); if (!cio_debug_crw_id) goto out_unregister; debug_register_view (cio_debug_crw_id, &debug_sprintf_view); diff --git a/drivers/s390/cio/qdio.c b/drivers/s390/cio/qdio.c index bbe9f45d143..82194c4eadf 100644 --- a/drivers/s390/cio/qdio.c +++ b/drivers/s390/cio/qdio.c @@ -56,7 +56,7 @@ #include "ioasm.h" #include "chsc.h" -#define VERSION_QDIO_C "$Revision: 1.98 $" +#define VERSION_QDIO_C "$Revision: 1.101 $" /****************** MODULE PARAMETER VARIABLES ********************/ MODULE_AUTHOR("Utz Bacher "); @@ -3342,7 +3342,7 @@ static int qdio_register_dbf_views(void) { qdio_dbf_setup=debug_register(QDIO_DBF_SETUP_NAME, - QDIO_DBF_SETUP_INDEX, + QDIO_DBF_SETUP_PAGES, QDIO_DBF_SETUP_NR_AREAS, QDIO_DBF_SETUP_LEN); if (!qdio_dbf_setup) @@ -3351,7 +3351,7 @@ qdio_register_dbf_views(void) debug_set_level(qdio_dbf_setup,QDIO_DBF_SETUP_LEVEL); qdio_dbf_sbal=debug_register(QDIO_DBF_SBAL_NAME, - QDIO_DBF_SBAL_INDEX, + QDIO_DBF_SBAL_PAGES, QDIO_DBF_SBAL_NR_AREAS, QDIO_DBF_SBAL_LEN); if (!qdio_dbf_sbal) @@ -3361,7 +3361,7 @@ qdio_register_dbf_views(void) debug_set_level(qdio_dbf_sbal,QDIO_DBF_SBAL_LEVEL); qdio_dbf_sense=debug_register(QDIO_DBF_SENSE_NAME, - QDIO_DBF_SENSE_INDEX, + QDIO_DBF_SENSE_PAGES, QDIO_DBF_SENSE_NR_AREAS, QDIO_DBF_SENSE_LEN); if (!qdio_dbf_sense) @@ -3371,7 +3371,7 @@ qdio_register_dbf_views(void) debug_set_level(qdio_dbf_sense,QDIO_DBF_SENSE_LEVEL); qdio_dbf_trace=debug_register(QDIO_DBF_TRACE_NAME, - QDIO_DBF_TRACE_INDEX, + QDIO_DBF_TRACE_PAGES, QDIO_DBF_TRACE_NR_AREAS, QDIO_DBF_TRACE_LEN); if (!qdio_dbf_trace) @@ -3382,7 +3382,7 @@ qdio_register_dbf_views(void) #ifdef CONFIG_QDIO_DEBUG qdio_dbf_slsb_out=debug_register(QDIO_DBF_SLSB_OUT_NAME, - QDIO_DBF_SLSB_OUT_INDEX, + QDIO_DBF_SLSB_OUT_PAGES, QDIO_DBF_SLSB_OUT_NR_AREAS, QDIO_DBF_SLSB_OUT_LEN); if (!qdio_dbf_slsb_out) @@ -3391,7 +3391,7 @@ qdio_register_dbf_views(void) debug_set_level(qdio_dbf_slsb_out,QDIO_DBF_SLSB_OUT_LEVEL); qdio_dbf_slsb_in=debug_register(QDIO_DBF_SLSB_IN_NAME, - QDIO_DBF_SLSB_IN_INDEX, + QDIO_DBF_SLSB_IN_PAGES, QDIO_DBF_SLSB_IN_NR_AREAS, QDIO_DBF_SLSB_IN_LEN); if (!qdio_dbf_slsb_in) diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h index b6daadac4e8..6b8aa6a852b 100644 --- a/drivers/s390/cio/qdio.h +++ b/drivers/s390/cio/qdio.h @@ -3,7 +3,7 @@ #include -#define VERSION_CIO_QDIO_H "$Revision: 1.32 $" +#define VERSION_CIO_QDIO_H "$Revision: 1.33 $" #ifdef CONFIG_QDIO_DEBUG #define QDIO_VERBOSE_LEVEL 9 @@ -132,7 +132,7 @@ enum qdio_irq_states { #define QDIO_DBF_SETUP_NAME "qdio_setup" #define QDIO_DBF_SETUP_LEN 8 -#define QDIO_DBF_SETUP_INDEX 2 +#define QDIO_DBF_SETUP_PAGES 4 #define QDIO_DBF_SETUP_NR_AREAS 1 #ifdef CONFIG_QDIO_DEBUG #define QDIO_DBF_SETUP_LEVEL 6 @@ -142,7 +142,7 @@ enum qdio_irq_states { #define QDIO_DBF_SBAL_NAME "qdio_labs" /* sbal */ #define QDIO_DBF_SBAL_LEN 256 -#define QDIO_DBF_SBAL_INDEX 2 +#define QDIO_DBF_SBAL_PAGES 4 #define QDIO_DBF_SBAL_NR_AREAS 2 #ifdef CONFIG_QDIO_DEBUG #define QDIO_DBF_SBAL_LEVEL 6 @@ -154,16 +154,16 @@ enum qdio_irq_states { #define QDIO_DBF_TRACE_LEN 8 #define QDIO_DBF_TRACE_NR_AREAS 2 #ifdef CONFIG_QDIO_DEBUG -#define QDIO_DBF_TRACE_INDEX 4 +#define QDIO_DBF_TRACE_PAGES 16 #define QDIO_DBF_TRACE_LEVEL 4 /* -------- could be even more verbose here */ #else /* CONFIG_QDIO_DEBUG */ -#define QDIO_DBF_TRACE_INDEX 2 +#define QDIO_DBF_TRACE_PAGES 4 #define QDIO_DBF_TRACE_LEVEL 2 #endif /* CONFIG_QDIO_DEBUG */ #define QDIO_DBF_SENSE_NAME "qdio_sense" #define QDIO_DBF_SENSE_LEN 64 -#define QDIO_DBF_SENSE_INDEX 1 +#define QDIO_DBF_SENSE_PAGES 2 #define QDIO_DBF_SENSE_NR_AREAS 1 #ifdef CONFIG_QDIO_DEBUG #define QDIO_DBF_SENSE_LEVEL 6 @@ -176,13 +176,13 @@ enum qdio_irq_states { #define QDIO_DBF_SLSB_OUT_NAME "qdio_slsb_out" #define QDIO_DBF_SLSB_OUT_LEN QDIO_MAX_BUFFERS_PER_Q -#define QDIO_DBF_SLSB_OUT_INDEX 8 +#define QDIO_DBF_SLSB_OUT_PAGES 256 #define QDIO_DBF_SLSB_OUT_NR_AREAS 1 #define QDIO_DBF_SLSB_OUT_LEVEL 6 #define QDIO_DBF_SLSB_IN_NAME "qdio_slsb_in" #define QDIO_DBF_SLSB_IN_LEN QDIO_MAX_BUFFERS_PER_Q -#define QDIO_DBF_SLSB_IN_INDEX 8 +#define QDIO_DBF_SLSB_IN_PAGES 256 #define QDIO_DBF_SLSB_IN_NR_AREAS 1 #define QDIO_DBF_SLSB_IN_LEVEL 6 #endif /* CONFIG_QDIO_DEBUG */ diff --git a/drivers/s390/net/claw.c b/drivers/s390/net/claw.c index a99927d54eb..60440dbe3a2 100644 --- a/drivers/s390/net/claw.c +++ b/drivers/s390/net/claw.c @@ -146,8 +146,8 @@ claw_unregister_debug_facility(void) static int claw_register_debug_facility(void) { - claw_dbf_setup = debug_register("claw_setup", 1, 1, 8); - claw_dbf_trace = debug_register("claw_trace", 1, 2, 8); + claw_dbf_setup = debug_register("claw_setup", 2, 1, 8); + claw_dbf_trace = debug_register("claw_trace", 2, 2, 8); if (claw_dbf_setup == NULL || claw_dbf_trace == NULL) { printk(KERN_WARNING "Not enough memory for debug facility.\n"); claw_unregister_debug_facility(); diff --git a/drivers/s390/net/ctcdbug.c b/drivers/s390/net/ctcdbug.c index 2c86bfa11b2..0e2a8bb9303 100644 --- a/drivers/s390/net/ctcdbug.c +++ b/drivers/s390/net/ctcdbug.c @@ -1,6 +1,6 @@ /* * - * linux/drivers/s390/net/ctcdbug.c ($Revision: 1.4 $) + * linux/drivers/s390/net/ctcdbug.c ($Revision: 1.6 $) * * CTC / ESCON network driver - s390 dbf exploit. * @@ -9,7 +9,7 @@ * Author(s): Original Code written by * Peter Tiedemann (ptiedem@de.ibm.com) * - * $Revision: 1.4 $ $Date: 2004/08/04 10:11:59 $ + * $Revision: 1.6 $ $Date: 2005/05/11 08:10:17 $ * * 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 @@ -51,15 +51,15 @@ int ctc_register_dbf_views(void) { ctc_dbf_setup = debug_register(CTC_DBF_SETUP_NAME, - CTC_DBF_SETUP_INDEX, + CTC_DBF_SETUP_PAGES, CTC_DBF_SETUP_NR_AREAS, CTC_DBF_SETUP_LEN); ctc_dbf_data = debug_register(CTC_DBF_DATA_NAME, - CTC_DBF_DATA_INDEX, + CTC_DBF_DATA_PAGES, CTC_DBF_DATA_NR_AREAS, CTC_DBF_DATA_LEN); ctc_dbf_trace = debug_register(CTC_DBF_TRACE_NAME, - CTC_DBF_TRACE_INDEX, + CTC_DBF_TRACE_PAGES, CTC_DBF_TRACE_NR_AREAS, CTC_DBF_TRACE_LEN); diff --git a/drivers/s390/net/ctcdbug.h b/drivers/s390/net/ctcdbug.h index 7fe2ebd1792..7d6afa1627c 100644 --- a/drivers/s390/net/ctcdbug.h +++ b/drivers/s390/net/ctcdbug.h @@ -1,6 +1,6 @@ /* * - * linux/drivers/s390/net/ctcdbug.h ($Revision: 1.5 $) + * linux/drivers/s390/net/ctcdbug.h ($Revision: 1.6 $) * * CTC / ESCON network driver - s390 dbf exploit. * @@ -9,7 +9,7 @@ * Author(s): Original Code written by * Peter Tiedemann (ptiedem@de.ibm.com) * - * $Revision: 1.5 $ $Date: 2005/02/27 19:46:44 $ + * $Revision: 1.6 $ $Date: 2005/05/11 08:10:17 $ * * 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 @@ -35,19 +35,19 @@ */ #define CTC_DBF_SETUP_NAME "ctc_setup" #define CTC_DBF_SETUP_LEN 16 -#define CTC_DBF_SETUP_INDEX 3 +#define CTC_DBF_SETUP_PAGES 8 #define CTC_DBF_SETUP_NR_AREAS 1 #define CTC_DBF_SETUP_LEVEL 3 #define CTC_DBF_DATA_NAME "ctc_data" #define CTC_DBF_DATA_LEN 128 -#define CTC_DBF_DATA_INDEX 3 +#define CTC_DBF_DATA_PAGES 8 #define CTC_DBF_DATA_NR_AREAS 1 #define CTC_DBF_DATA_LEVEL 3 #define CTC_DBF_TRACE_NAME "ctc_trace" #define CTC_DBF_TRACE_LEN 16 -#define CTC_DBF_TRACE_INDEX 2 +#define CTC_DBF_TRACE_PAGES 4 #define CTC_DBF_TRACE_NR_AREAS 2 #define CTC_DBF_TRACE_LEVEL 3 diff --git a/drivers/s390/net/iucv.h b/drivers/s390/net/iucv.h index 198330217ef..0c4644d3d2f 100644 --- a/drivers/s390/net/iucv.h +++ b/drivers/s390/net/iucv.h @@ -37,19 +37,19 @@ */ #define IUCV_DBF_SETUP_NAME "iucv_setup" #define IUCV_DBF_SETUP_LEN 32 -#define IUCV_DBF_SETUP_INDEX 1 +#define IUCV_DBF_SETUP_PAGES 2 #define IUCV_DBF_SETUP_NR_AREAS 1 #define IUCV_DBF_SETUP_LEVEL 3 #define IUCV_DBF_DATA_NAME "iucv_data" #define IUCV_DBF_DATA_LEN 128 -#define IUCV_DBF_DATA_INDEX 1 +#define IUCV_DBF_DATA_PAGES 2 #define IUCV_DBF_DATA_NR_AREAS 1 #define IUCV_DBF_DATA_LEVEL 2 #define IUCV_DBF_TRACE_NAME "iucv_trace" #define IUCV_DBF_TRACE_LEN 16 -#define IUCV_DBF_TRACE_INDEX 2 +#define IUCV_DBF_TRACE_PAGES 4 #define IUCV_DBF_TRACE_NR_AREAS 1 #define IUCV_DBF_TRACE_LEVEL 3 diff --git a/drivers/s390/net/lcs.c b/drivers/s390/net/lcs.c index ab086242d30..46f34ba93ac 100644 --- a/drivers/s390/net/lcs.c +++ b/drivers/s390/net/lcs.c @@ -11,7 +11,7 @@ * Frank Pavlic (pavlic@de.ibm.com) and * Martin Schwidefsky * - * $Revision: 1.98 $ $Date: 2005/04/18 13:41:29 $ + * $Revision: 1.99 $ $Date: 2005/05/11 08:10:17 $ * * 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 @@ -59,7 +59,7 @@ /** * initialization string for output */ -#define VERSION_LCS_C "$Revision: 1.98 $" +#define VERSION_LCS_C "$Revision: 1.99 $" static char version[] __initdata = "LCS driver ("VERSION_LCS_C "/" VERSION_LCS_H ")"; static char debug_buffer[255]; @@ -93,8 +93,8 @@ lcs_unregister_debug_facility(void) static int lcs_register_debug_facility(void) { - lcs_dbf_setup = debug_register("lcs_setup", 1, 1, 8); - lcs_dbf_trace = debug_register("lcs_trace", 1, 2, 8); + lcs_dbf_setup = debug_register("lcs_setup", 2, 1, 8); + lcs_dbf_trace = debug_register("lcs_trace", 2, 2, 8); if (lcs_dbf_setup == NULL || lcs_dbf_trace == NULL) { PRINT_ERR("Not enough memory for debug facility.\n"); lcs_unregister_debug_facility(); diff --git a/drivers/s390/net/netiucv.c b/drivers/s390/net/netiucv.c index 3fd4fb754b2..69425a7a6e9 100644 --- a/drivers/s390/net/netiucv.c +++ b/drivers/s390/net/netiucv.c @@ -1,5 +1,5 @@ /* - * $Id: netiucv.c,v 1.63 2004/07/27 13:36:05 mschwide Exp $ + * $Id: netiucv.c,v 1.66 2005/05/11 08:10:17 holzheu Exp $ * * IUCV network driver * @@ -30,7 +30,7 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. * - * RELEASE-TAG: IUCV network driver $Revision: 1.63 $ + * RELEASE-TAG: IUCV network driver $Revision: 1.66 $ * */ @@ -391,15 +391,15 @@ static int iucv_register_dbf_views(void) { iucv_dbf_setup = debug_register(IUCV_DBF_SETUP_NAME, - IUCV_DBF_SETUP_INDEX, + IUCV_DBF_SETUP_PAGES, IUCV_DBF_SETUP_NR_AREAS, IUCV_DBF_SETUP_LEN); iucv_dbf_data = debug_register(IUCV_DBF_DATA_NAME, - IUCV_DBF_DATA_INDEX, + IUCV_DBF_DATA_PAGES, IUCV_DBF_DATA_NR_AREAS, IUCV_DBF_DATA_LEN); iucv_dbf_trace = debug_register(IUCV_DBF_TRACE_NAME, - IUCV_DBF_TRACE_INDEX, + IUCV_DBF_TRACE_PAGES, IUCV_DBF_TRACE_NR_AREAS, IUCV_DBF_TRACE_LEN); @@ -2076,7 +2076,7 @@ DRIVER_ATTR(remove, 0200, NULL, remove_write); static void netiucv_banner(void) { - char vbuf[] = "$Revision: 1.63 $"; + char vbuf[] = "$Revision: 1.66 $"; char *version = vbuf; if ((version = strchr(version, ':'))) { diff --git a/drivers/s390/net/qeth.h b/drivers/s390/net/qeth.h index a755b57db46..008e0a5d2eb 100644 --- a/drivers/s390/net/qeth.h +++ b/drivers/s390/net/qeth.h @@ -42,44 +42,44 @@ */ #define QETH_DBF_SETUP_NAME "qeth_setup" #define QETH_DBF_SETUP_LEN 8 -#define QETH_DBF_SETUP_INDEX 3 +#define QETH_DBF_SETUP_PAGES 8 #define QETH_DBF_SETUP_NR_AREAS 1 #define QETH_DBF_SETUP_LEVEL 5 #define QETH_DBF_MISC_NAME "qeth_misc" #define QETH_DBF_MISC_LEN 128 -#define QETH_DBF_MISC_INDEX 1 +#define QETH_DBF_MISC_PAGES 2 #define QETH_DBF_MISC_NR_AREAS 1 #define QETH_DBF_MISC_LEVEL 2 #define QETH_DBF_DATA_NAME "qeth_data" #define QETH_DBF_DATA_LEN 96 -#define QETH_DBF_DATA_INDEX 3 +#define QETH_DBF_DATA_PAGES 8 #define QETH_DBF_DATA_NR_AREAS 1 #define QETH_DBF_DATA_LEVEL 2 #define QETH_DBF_CONTROL_NAME "qeth_control" #define QETH_DBF_CONTROL_LEN 256 -#define QETH_DBF_CONTROL_INDEX 3 +#define QETH_DBF_CONTROL_PAGES 8 #define QETH_DBF_CONTROL_NR_AREAS 2 #define QETH_DBF_CONTROL_LEVEL 5 #define QETH_DBF_TRACE_NAME "qeth_trace" #define QETH_DBF_TRACE_LEN 8 -#define QETH_DBF_TRACE_INDEX 2 +#define QETH_DBF_TRACE_PAGES 4 #define QETH_DBF_TRACE_NR_AREAS 2 #define QETH_DBF_TRACE_LEVEL 3 extern debug_info_t *qeth_dbf_trace; #define QETH_DBF_SENSE_NAME "qeth_sense" #define QETH_DBF_SENSE_LEN 64 -#define QETH_DBF_SENSE_INDEX 1 +#define QETH_DBF_SENSE_PAGES 2 #define QETH_DBF_SENSE_NR_AREAS 1 #define QETH_DBF_SENSE_LEVEL 2 #define QETH_DBF_QERR_NAME "qeth_qerr" #define QETH_DBF_QERR_LEN 8 -#define QETH_DBF_QERR_INDEX 1 +#define QETH_DBF_QERR_PAGES 2 #define QETH_DBF_QERR_NR_AREAS 2 #define QETH_DBF_QERR_LEVEL 2 diff --git a/drivers/s390/net/qeth_main.c b/drivers/s390/net/qeth_main.c index 208127a5033..3cb88c77003 100644 --- a/drivers/s390/net/qeth_main.c +++ b/drivers/s390/net/qeth_main.c @@ -7639,31 +7639,31 @@ static int qeth_register_dbf_views(void) { qeth_dbf_setup = debug_register(QETH_DBF_SETUP_NAME, - QETH_DBF_SETUP_INDEX, + QETH_DBF_SETUP_PAGES, QETH_DBF_SETUP_NR_AREAS, QETH_DBF_SETUP_LEN); qeth_dbf_misc = debug_register(QETH_DBF_MISC_NAME, - QETH_DBF_MISC_INDEX, + QETH_DBF_MISC_PAGES, QETH_DBF_MISC_NR_AREAS, QETH_DBF_MISC_LEN); qeth_dbf_data = debug_register(QETH_DBF_DATA_NAME, - QETH_DBF_DATA_INDEX, + QETH_DBF_DATA_PAGES, QETH_DBF_DATA_NR_AREAS, QETH_DBF_DATA_LEN); qeth_dbf_control = debug_register(QETH_DBF_CONTROL_NAME, - QETH_DBF_CONTROL_INDEX, + QETH_DBF_CONTROL_PAGES, QETH_DBF_CONTROL_NR_AREAS, QETH_DBF_CONTROL_LEN); qeth_dbf_sense = debug_register(QETH_DBF_SENSE_NAME, - QETH_DBF_SENSE_INDEX, + QETH_DBF_SENSE_PAGES, QETH_DBF_SENSE_NR_AREAS, QETH_DBF_SENSE_LEN); qeth_dbf_qerr = debug_register(QETH_DBF_QERR_NAME, - QETH_DBF_QERR_INDEX, + QETH_DBF_QERR_PAGES, QETH_DBF_QERR_NR_AREAS, QETH_DBF_QERR_LEN); qeth_dbf_trace = debug_register(QETH_DBF_TRACE_NAME, - QETH_DBF_TRACE_INDEX, + QETH_DBF_TRACE_PAGES, QETH_DBF_TRACE_NR_AREAS, QETH_DBF_TRACE_LEN); diff --git a/include/asm-s390/debug.h b/include/asm-s390/debug.h index 6bbcdea42a8..92360d90144 100644 --- a/include/asm-s390/debug.h +++ b/include/asm-s390/debug.h @@ -9,6 +9,8 @@ #ifndef DEBUG_H #define DEBUG_H +#include +#include #include /* Note: @@ -31,19 +33,18 @@ struct __debug_entry{ } __attribute__((packed)); -#define __DEBUG_FEATURE_VERSION 1 /* version of debug feature */ +#define __DEBUG_FEATURE_VERSION 2 /* version of debug feature */ #ifdef __KERNEL__ #include #include #include -#include #define DEBUG_MAX_LEVEL 6 /* debug levels range from 0 to 6 */ #define DEBUG_OFF_LEVEL -1 /* level where debug is switched off */ #define DEBUG_FLUSH_ALL -1 /* parameter to flush all areas */ #define DEBUG_MAX_VIEWS 10 /* max number of views in proc fs */ -#define DEBUG_MAX_PROCF_LEN 64 /* max length for a proc file name */ +#define DEBUG_MAX_NAME_LEN 64 /* max length for a debugfs file name */ #define DEBUG_DEFAULT_LEVEL 3 /* initial debug level */ #define DEBUG_DIR_ROOT "s390dbf" /* name of debug root directory in proc fs */ @@ -64,16 +65,17 @@ typedef struct debug_info { spinlock_t lock; int level; int nr_areas; - int page_order; + int pages_per_area; int buf_size; int entry_size; - debug_entry_t** areas; + debug_entry_t*** areas; int active_area; - int *active_entry; - struct proc_dir_entry* proc_root_entry; - struct proc_dir_entry* proc_entries[DEBUG_MAX_VIEWS]; + int *active_pages; + int *active_entries; + struct dentry* debugfs_root_entry; + struct dentry* debugfs_entries[DEBUG_MAX_VIEWS]; struct debug_view* views[DEBUG_MAX_VIEWS]; - char name[DEBUG_MAX_PROCF_LEN]; + char name[DEBUG_MAX_NAME_LEN]; } debug_info_t; typedef int (debug_header_proc_t) (debug_info_t* id, @@ -98,7 +100,7 @@ int debug_dflt_header_fn(debug_info_t* id, struct debug_view* view, int area, debug_entry_t* entry, char* out_buf); struct debug_view { - char name[DEBUG_MAX_PROCF_LEN]; + char name[DEBUG_MAX_NAME_LEN]; debug_prolog_proc_t* prolog_proc; debug_header_proc_t* header_proc; debug_format_proc_t* format_proc; @@ -120,7 +122,7 @@ debug_entry_t* debug_exception_common(debug_info_t* id, int level, /* Debug Feature API: */ -debug_info_t* debug_register(char* name, int pages_index, int nr_areas, +debug_info_t* debug_register(char* name, int pages, int nr_areas, int buf_size); void debug_unregister(debug_info_t* id); @@ -132,7 +134,8 @@ void debug_stop_all(void); extern inline debug_entry_t* debug_event(debug_info_t* id, int level, void* data, int length) { - if ((!id) || (level > id->level)) return NULL; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; return debug_event_common(id,level,data,length); } @@ -140,7 +143,8 @@ extern inline debug_entry_t* debug_int_event(debug_info_t* id, int level, unsigned int tag) { unsigned int t=tag; - if ((!id) || (level > id->level)) return NULL; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; return debug_event_common(id,level,&t,sizeof(unsigned int)); } @@ -148,14 +152,16 @@ extern inline debug_entry_t * debug_long_event (debug_info_t* id, int level, unsigned long tag) { unsigned long t=tag; - if ((!id) || (level > id->level)) return NULL; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; return debug_event_common(id,level,&t,sizeof(unsigned long)); } extern inline debug_entry_t* debug_text_event(debug_info_t* id, int level, const char* txt) { - if ((!id) || (level > id->level)) return NULL; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; return debug_event_common(id,level,txt,strlen(txt)); } @@ -167,7 +173,8 @@ debug_sprintf_event(debug_info_t* id,int level,char *string,...) extern inline debug_entry_t* debug_exception(debug_info_t* id, int level, void* data, int length) { - if ((!id) || (level > id->level)) return NULL; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; return debug_exception_common(id,level,data,length); } @@ -175,7 +182,8 @@ extern inline debug_entry_t* debug_int_exception(debug_info_t* id, int level, unsigned int tag) { unsigned int t=tag; - if ((!id) || (level > id->level)) return NULL; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; return debug_exception_common(id,level,&t,sizeof(unsigned int)); } @@ -183,14 +191,16 @@ extern inline debug_entry_t * debug_long_exception (debug_info_t* id, int level, unsigned long tag) { unsigned long t=tag; - if ((!id) || (level > id->level)) return NULL; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; return debug_exception_common(id,level,&t,sizeof(unsigned long)); } extern inline debug_entry_t* debug_text_exception(debug_info_t* id, int level, const char* txt) { - if ((!id) || (level > id->level)) return NULL; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; return debug_exception_common(id,level,txt,strlen(txt)); } -- cgit v1.2.3-70-g09d2 From cf13f0eaffa31bf6a145c53c589654b11c72ddc7 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Sat, 25 Jun 2005 14:58:11 -0700 Subject: [PATCH] kexec: s390 support Add kexec support for s390 architecture. From: Milton Miller - Fix passing of first argument to relocate_kernel assembly. - Fix Kconfig description. - Remove wrong comment and comments that describe obvious things. - Allow only KEXEC_TYPE_DEFAULT as image type -> dump not supported. Acked-by: Martin Schwidefsky Signed-off-by: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/Kconfig | 8 +++ arch/s390/kernel/Makefile | 10 ++++ arch/s390/kernel/compat_wrapper.S | 8 +++ arch/s390/kernel/crash.c | 17 +++++++ arch/s390/kernel/machine_kexec.c | 98 ++++++++++++++++++++++++++++++++++++ arch/s390/kernel/relocate_kernel.S | 81 +++++++++++++++++++++++++++++ arch/s390/kernel/relocate_kernel64.S | 82 ++++++++++++++++++++++++++++++ arch/s390/kernel/syscalls.S | 2 +- include/asm-s390/kexec.h | 42 ++++++++++++++++ include/asm-s390/unistd.h | 2 +- include/linux/kexec.h | 1 + 11 files changed, 349 insertions(+), 2 deletions(-) create mode 100644 arch/s390/kernel/crash.c create mode 100644 arch/s390/kernel/machine_kexec.c create mode 100644 arch/s390/kernel/relocate_kernel.S create mode 100644 arch/s390/kernel/relocate_kernel64.S create mode 100644 include/asm-s390/kexec.h (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 32696c1d928..6600ee87f89 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -455,6 +455,14 @@ config NO_IDLE_HZ_INIT The HZ timer is switched off in idle by default. That means the HZ timer is already disabled at boot time. +config KEXEC + bool "kexec system call (EXPERIMENTAL)" + depends on EXPERIMENTAL + help + kexec is a system call that implements the ability to shutdown your + current kernel, and to start another kernel. It is like a reboot + but is independent of hardware/microcode support. + endmenu config PCMCIA diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index b41e0e199a7..ab1e49d2e51 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -25,6 +25,16 @@ obj-$(CONFIG_ARCH_S390X) += entry64.o reipl64.o obj-$(CONFIG_VIRT_TIMER) += vtime.o +# Kexec part +S390_KEXEC_OBJS := machine_kexec.o crash.o +ifeq ($(CONFIG_ARCH_S390X),y) +S390_KEXEC_OBJS += relocate_kernel64.o +else +S390_KEXEC_OBJS += relocate_kernel.o +endif +obj-$(CONFIG_KEXEC) += $(S390_KEXEC_OBJS) + + # # This is just to get the dependencies... # diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index 7a607b1d038..bf529739c8a 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -1441,3 +1441,11 @@ compat_sys_waitid_wrapper: lgfr %r5,%r5 # int llgtr %r6,%r6 # struct rusage_emu31 * jg compat_sys_waitid + + .globl compat_sys_kexec_load_wrapper +compat_sys_kexec_load_wrapper: + llgfr %r2,%r2 # unsigned long + llgfr %r3,%r3 # unsigned long + llgtr %r4,%r4 # struct kexec_segment * + llgfr %r5,%r5 # unsigned long + jg compat_sys_kexec_load diff --git a/arch/s390/kernel/crash.c b/arch/s390/kernel/crash.c new file mode 100644 index 00000000000..db38283c1f2 --- /dev/null +++ b/arch/s390/kernel/crash.c @@ -0,0 +1,17 @@ +/* + * arch/s390/kernel/crash.c + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Heiko Carstens + * + */ + +#include +#include + +note_buf_t crash_notes[NR_CPUS]; + +void machine_crash_shutdown(void) +{ +} diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c new file mode 100644 index 00000000000..7a94db76df4 --- /dev/null +++ b/arch/s390/kernel/machine_kexec.c @@ -0,0 +1,98 @@ +/* + * arch/s390/kernel/machine_kexec.c + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger + * + */ + +/* + * s390_machine_kexec.c - handle the transition of Linux booting another kernel + * on the S390 architecture. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static void kexec_halt_all_cpus(void *); + +typedef void (*relocate_kernel_t) (kimage_entry_t *, unsigned long); + +const extern unsigned char relocate_kernel[]; +const extern unsigned long long relocate_kernel_len; + +int +machine_kexec_prepare(struct kimage *image) +{ + unsigned long reboot_code_buffer; + + /* We don't support anything but the default image type for now. */ + if (image->type != KEXEC_TYPE_DEFAULT) + return -EINVAL; + + /* Get the destination where the assembler code should be copied to.*/ + reboot_code_buffer = page_to_pfn(image->control_code_page)<control_code_page) << PAGE_SHIFT); + + /* Call the moving routine */ + (*data_mover) (&image->head, image->start); +} diff --git a/arch/s390/kernel/relocate_kernel.S b/arch/s390/kernel/relocate_kernel.S new file mode 100644 index 00000000000..d5e4a62fbb7 --- /dev/null +++ b/arch/s390/kernel/relocate_kernel.S @@ -0,0 +1,81 @@ +/* + * arch/s390/kernel/relocate_kernel.S + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger + * + */ + +/* + * moves the new kernel to its destination... + * %r2 = pointer to first kimage_entry_t + * %r3 = start address - where to jump to after the job is done... + * + * %r5 will be used as temp. storage + * %r6 holds the destination address + * %r7 = PAGE_SIZE + * %r8 holds the source address + * %r9 = PAGE_SIZE + * %r10 is a page mask + */ + + .text + .globl relocate_kernel + relocate_kernel: + basr %r13,0 #base address + .base: + spx zero64-.base(%r13) #absolute addressing mode + stnsm sys_msk-.base(%r13),0xf8 #disable DAT and IRQ (external) + lhi %r10,-1 #preparing the mask + sll %r10,12 #shift it such that it becomes 0xf000 + .top: + lhi %r7,4096 #load PAGE_SIZE in r7 + lhi %r9,4096 #load PAGE_SIZE in r9 + l %r5,0(%r2) #read another word for indirection page + ahi %r2,4 #increment pointer + tml %r5,0x1 #is it a destination page? + je .indir_check #NO, goto "indir_check" + lr %r6,%r5 #r6 = r5 + nr %r6,%r10 #mask it out and... + j .top #...next iteration + .indir_check: + tml %r5,0x2 #is it a indirection page? + je .done_test #NO, goto "done_test" + nr %r5,%r10 #YES, mask out, + lr %r2,%r5 #move it into the right register, + j .top #and read next... + .done_test: + tml %r5,0x4 #is it the done indicator? + je .source_test #NO! Well, then it should be the source indicator... + j .done #ok, lets finish it here... + .source_test: + tml %r5,0x8 #it should be a source indicator... + je .top #NO, ignore it... + lr %r8,%r5 #r8 = r5 + nr %r8,%r10 #masking + 0: mvcle %r6,%r8,0x0 #copy PAGE_SIZE bytes from r8 to r6 - pad with 0 + jo 0b + j .top + .done: + sr %r0,%r0 #clear register r0 + la %r4,load_psw-.base(%r13) #load psw-address into the register + o %r3,4(%r4) #or load address into psw + st %r3,4(%r4) + mvc 0(8,%r0),0(%r4) #copy psw to absolute address 0 + sr %r1,%r1 #clear %r1 + sr %r2,%r2 #clear %r2 + sigp %r1,%r2,0x12 #set cpuid to zero + lpsw 0 #hopefully start new kernel... + + .align 8 + zero64: + .quad 0 + load_psw: + .long 0x00080000,0x80000000 + sys_msk: + .quad 0 + relocate_kernel_end: + .globl relocate_kernel_len + relocate_kernel_len: + .quad relocate_kernel_end - relocate_kernel diff --git a/arch/s390/kernel/relocate_kernel64.S b/arch/s390/kernel/relocate_kernel64.S new file mode 100644 index 00000000000..96290cc4eb3 --- /dev/null +++ b/arch/s390/kernel/relocate_kernel64.S @@ -0,0 +1,82 @@ +/* + * arch/s390/kernel/relocate_kernel64.S + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger + * + */ + +/* + * moves the new kernel to its destination... + * %r2 = pointer to first kimage_entry_t + * %r3 = start address - where to jump to after the job is done... + * + * %r5 will be used as temp. storage + * %r6 holds the destination address + * %r7 = PAGE_SIZE + * %r8 holds the source address + * %r9 = PAGE_SIZE + * + * 0xf000 is a page_mask + */ + + .text + .globl relocate_kernel + relocate_kernel: + basr %r13,0 #base address + .base: + spx zero64-.base(%r13) #absolute addressing mode + stnsm sys_msk-.base(%r13),0xf8 #disable DAT and IRQ (external) + .top: + lghi %r7,4096 #load PAGE_SIZE in r7 + lghi %r9,4096 #load PAGE_SIZE in r9 + lg %r5,0(%r2) #read another word for indirection page + aghi %r2,8 #increment pointer + tml %r5,0x1 #is it a destination page? + je .indir_check #NO, goto "indir_check" + lgr %r6,%r5 #r6 = r5 + nill %r6,0xf000 #mask it out and... + j .top #...next iteration + .indir_check: + tml %r5,0x2 #is it a indirection page? + je .done_test #NO, goto "done_test" + nill %r5,0xf000 #YES, mask out, + lgr %r2,%r5 #move it into the right register, + j .top #and read next... + .done_test: + tml %r5,0x4 #is it the done indicator? + je .source_test #NO! Well, then it should be the source indicator... + j .done #ok, lets finish it here... + .source_test: + tml %r5,0x8 #it should be a source indicator... + je .top #NO, ignore it... + lgr %r8,%r5 #r8 = r5 + nill %r8,0xf000 #masking + 0: mvcle %r6,%r8,0x0 #copy PAGE_SIZE bytes from r8 to r6 - pad with 0 + jo 0b + j .top + .done: + sgr %r0,%r0 #clear register r0 + la %r4,load_psw-.base(%r13) #load psw-address into the register + o %r3,4(%r4) #or load address into psw + st %r3,4(%r4) + mvc 0(8,%r0),0(%r4) #copy psw to absolute address 0 + sam31 #31 bit mode + sr %r1,%r1 #erase register r1 + sr %r2,%r2 #erase register r2 + sigp %r1,%r2,0x12 #set cpuid to zero + lpsw 0 #hopefully start new kernel... + + .align 8 + zero64: + .quad 0 + load_psw: + .long 0x00080000,0x80000000 + sys_msk: + .quad 0 + relocate_kernel_end: + .globl relocate_kernel_len + relocate_kernel_len: + .quad relocate_kernel_end - relocate_kernel + diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index 515938628f8..a8668afb5f8 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -285,7 +285,7 @@ SYSCALL(sys_mq_timedsend,sys_mq_timedsend,compat_sys_mq_timedsend_wrapper) SYSCALL(sys_mq_timedreceive,sys_mq_timedreceive,compat_sys_mq_timedreceive_wrapper) SYSCALL(sys_mq_notify,sys_mq_notify,compat_sys_mq_notify_wrapper) /* 275 */ SYSCALL(sys_mq_getsetattr,sys_mq_getsetattr,compat_sys_mq_getsetattr_wrapper) -NI_SYSCALL /* reserved for kexec */ +SYSCALL(sys_kexec_load,sys_kexec_load,compat_sys_kexec_load_wrapper) SYSCALL(sys_add_key,sys_add_key,compat_sys_add_key_wrapper) SYSCALL(sys_request_key,sys_request_key,compat_sys_request_key_wrapper) SYSCALL(sys_keyctl,sys_keyctl,compat_sys_keyctl) /* 280 */ diff --git a/include/asm-s390/kexec.h b/include/asm-s390/kexec.h new file mode 100644 index 00000000000..54cf7d9f251 --- /dev/null +++ b/include/asm-s390/kexec.h @@ -0,0 +1,42 @@ +/* + * include/asm-s390/kexec.h + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger + * + */ + +#ifndef _S390_KEXEC_H +#define _S390_KEXEC_H + +#include +#include +/* + * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return. + * I.e. Maximum page that is mapped directly into kernel memory, + * and kmap is not required. + */ + +/* Maximum physical address we can use pages from */ +#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL) + +/* Maximum address we can reach in physical address mode */ +#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL) + +/* Maximum address we can use for the control pages */ +/* Not more than 2GB */ +#define KEXEC_CONTROL_MEMORY_LIMIT (1<<31) + +/* Allocate one page for the pdp and the second for the code */ +#define KEXEC_CONTROL_CODE_SIZE 4096 + +/* The native architecture */ +#define KEXEC_ARCH KEXEC_ARCH_S390 + +#define MAX_NOTE_BYTES 1024 +typedef u32 note_buf_t[MAX_NOTE_BYTES/4]; + +extern note_buf_t crash_notes[]; + +#endif /*_S390_KEXEC_H */ diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h index f1a204f7c0f..363db45f8d0 100644 --- a/include/asm-s390/unistd.h +++ b/include/asm-s390/unistd.h @@ -269,7 +269,7 @@ #define __NR_mq_timedreceive 274 #define __NR_mq_notify 275 #define __NR_mq_getsetattr 276 -/* Number 277 is reserved for new sys_kexec_load */ +#define __NR_kexec_load 277 #define __NR_add_key 278 #define __NR_request_key 279 #define __NR_keyctl 280 diff --git a/include/linux/kexec.h b/include/linux/kexec.h index e3fc35f4e35..0653a27c3d7 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -114,6 +114,7 @@ extern struct kimage *kexec_image; #define KEXEC_ARCH_PPC (20 << 16) #define KEXEC_ARCH_PPC64 (21 << 16) #define KEXEC_ARCH_IA_64 (50 << 16) +#define KEXEC_ARCH_S390 (22 << 16) #define KEXEC_FLAGS (KEXEC_ON_CRASH) /* List of defined/legal kexec flags */ -- cgit v1.2.3-70-g09d2 From 6e274d144302068a00794ec22e73520c0615cb6f Mon Sep 17 00:00:00 2001 From: Alexander Nyberg Date: Sat, 25 Jun 2005 14:58:26 -0700 Subject: [PATCH] kdump: Use real pt_regs from exception Makes kexec_crashdump() take a pt_regs * as an argument. This allows to get exact register state at the point of the crash. If we come from direct panic assertion NULL will be passed and the current registers saved before crashdump. This hooks into two places: die(): check the conditions under which we will panic when calling do_exit and go there directly with the pt_regs that caused the fatal fault. die_nmi(): If we receive an NMI lockup while in the kernel use the pt_regs and go directly to crash_kexec(). We're probably nested up badly at this point so this might be the only chance to escape with proper information. Signed-off-by: Alexander Nyberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/crash.c | 36 ++++++++++++++++++++++++------------ arch/i386/kernel/traps.c | 17 +++++++++++++++++ arch/ppc/kernel/machine_kexec.c | 2 +- arch/ppc64/kernel/machine_kexec.c | 2 +- arch/s390/kernel/crash.c | 2 +- arch/x86_64/kernel/crash.c | 2 +- drivers/char/sysrq.c | 2 +- include/linux/kexec.h | 8 ++++++-- include/linux/reboot.h | 3 ++- kernel/kexec.c | 13 +++++++++++-- kernel/panic.c | 2 +- 11 files changed, 66 insertions(+), 23 deletions(-) (limited to 'arch/s390') diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c index a021681d21f..8bdb4b6af0f 100644 --- a/arch/i386/kernel/crash.c +++ b/arch/i386/kernel/crash.c @@ -100,12 +100,31 @@ static void crash_get_current_regs(struct pt_regs *regs) regs->eip = (unsigned long)current_text_addr(); } -static void crash_save_self(void) +/* CPU does not save ss and esp on stack if execution is already + * running in kernel mode at the time of NMI occurrence. This code + * fixes it. + */ +static void crash_setup_regs(struct pt_regs *newregs, struct pt_regs *oldregs) +{ + memcpy(newregs, oldregs, sizeof(*newregs)); + newregs->esp = (unsigned long)&(oldregs->esp); + __asm__ __volatile__("xorl %eax, %eax;"); + __asm__ __volatile__ ("movw %%ss, %%ax;" :"=a"(newregs->xss)); +} + +/* We may have saved_regs from where the error came from + * or it is NULL if via a direct panic(). + */ +static void crash_save_self(struct pt_regs *saved_regs) { struct pt_regs regs; int cpu; cpu = smp_processor_id(); - crash_get_current_regs(®s); + + if (saved_regs) + crash_setup_regs(®s, saved_regs); + else + crash_get_current_regs(®s); crash_save_this_cpu(®s, cpu); } @@ -124,15 +143,8 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu) return 1; local_irq_disable(); - /* CPU does not save ss and esp on stack if execution is already - * running in kernel mode at the time of NMI occurrence. This code - * fixes it. - */ if (!user_mode(regs)) { - memcpy(&fixed_regs, regs, sizeof(*regs)); - fixed_regs.esp = (unsigned long)&(regs->esp); - __asm__ __volatile__("xorl %eax, %eax;"); - __asm__ __volatile__ ("movw %%ss, %%ax;" :"=a"(fixed_regs.xss)); + crash_setup_regs(&fixed_regs, regs); regs = &fixed_regs; } crash_save_this_cpu(regs, cpu); @@ -184,7 +196,7 @@ static void nmi_shootdown_cpus(void) } #endif -void machine_crash_shutdown(void) +void machine_crash_shutdown(struct pt_regs *regs) { /* This function is only called after the system * has paniced or is otherwise in a critical state. @@ -204,5 +216,5 @@ void machine_crash_shutdown(void) #if defined(CONFIG_X86_IO_APIC) disable_IO_APIC(); #endif - crash_save_self(); + crash_save_self(regs); } diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 207ea8ba716..e458463ebc0 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -27,6 +27,7 @@ #include #include #include +#include #ifdef CONFIG_EISA #include @@ -294,6 +295,9 @@ bug: printk("Kernel BUG\n"); } +/* 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 { @@ -341,6 +345,10 @@ void die(const char * str, struct pt_regs * regs, long err) bust_spinlocks(0); die.lock_owner = -1; spin_unlock_irq(&die.lock); + + if (kexec_should_crash(current)) + crash_kexec(regs); + if (in_interrupt()) panic("Fatal exception in interrupt"); @@ -570,6 +578,15 @@ void die_nmi (struct pt_regs *regs, const char *msg) console_silent(); spin_unlock(&nmi_print_lock); bust_spinlocks(0); + + /* If we are in kernel we are probably nested up pretty bad + * and might aswell get out now while we still can. + */ + if (!user_mode(regs)) { + current->thread.trap_no = 2; + crash_kexec(regs); + } + do_exit(SIGSEGV); } diff --git a/arch/ppc/kernel/machine_kexec.c b/arch/ppc/kernel/machine_kexec.c index 435ad9ea0a8..b82535357d6 100644 --- a/arch/ppc/kernel/machine_kexec.c +++ b/arch/ppc/kernel/machine_kexec.c @@ -34,7 +34,7 @@ void machine_shutdown(void) } } -void machine_crash_shutdown(void) +void machine_crash_shutdown(struct pt_regs *regs) { if (ppc_md.machine_crash_shutdown) { ppc_md.machine_crash_shutdown(); diff --git a/arch/ppc64/kernel/machine_kexec.c b/arch/ppc64/kernel/machine_kexec.c index 217965d60a4..06b25b59c8a 100644 --- a/arch/ppc64/kernel/machine_kexec.c +++ b/arch/ppc64/kernel/machine_kexec.c @@ -34,7 +34,7 @@ note_buf_t crash_notes[NR_CPUS]; * and if what it will achieve. Letting it be now to compile the code * in generic kexec environment */ -void machine_crash_shutdown(void) +void machine_crash_shutdown(struct pt_regs *regs) { /* do nothing right now */ /* smp_relase_cpus() if we want smp on panic kernel */ diff --git a/arch/s390/kernel/crash.c b/arch/s390/kernel/crash.c index db38283c1f2..7bd169c58b0 100644 --- a/arch/s390/kernel/crash.c +++ b/arch/s390/kernel/crash.c @@ -12,6 +12,6 @@ note_buf_t crash_notes[NR_CPUS]; -void machine_crash_shutdown(void) +void machine_crash_shutdown(struct pt_regs *regs) { } diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c index 6183bcb8525..d7fa4248501 100644 --- a/arch/x86_64/kernel/crash.c +++ b/arch/x86_64/kernel/crash.c @@ -22,7 +22,7 @@ note_buf_t crash_notes[NR_CPUS]; -void machine_crash_shutdown(void) +void machine_crash_shutdown(struct pt_regs *regs) { /* This function is only called after the system * has paniced or is otherwise in a critical state. diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c index 53b2c8fab00..af79805b557 100644 --- a/drivers/char/sysrq.c +++ b/drivers/char/sysrq.c @@ -100,7 +100,7 @@ static struct sysrq_key_op sysrq_unraw_op = { static void sysrq_handle_crashdump(int key, struct pt_regs *pt_regs, struct tty_struct *tty) { - crash_kexec(); + crash_kexec(pt_regs); } static struct sysrq_key_op sysrq_crashdump_op = { .handler = sysrq_handle_crashdump, diff --git a/include/linux/kexec.h b/include/linux/kexec.h index 0653a27c3d7..7383173a3a9 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -99,7 +99,8 @@ extern asmlinkage long compat_sys_kexec_load(unsigned long entry, unsigned long flags); #endif extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order); -extern void crash_kexec(void); +extern void crash_kexec(struct pt_regs *); +int kexec_should_crash(struct task_struct *); extern struct kimage *kexec_image; #define KEXEC_ON_CRASH 0x00000001 @@ -123,6 +124,9 @@ extern struct kimage *kexec_image; extern struct resource crashk_res; #else /* !CONFIG_KEXEC */ -static inline void crash_kexec(void) { } +struct pt_regs; +struct task_struct; +static inline void crash_kexec(struct pt_regs *regs) { } +static inline int kexec_should_crash(struct task_struct *p) { return 0; } #endif /* CONFIG_KEXEC */ #endif /* LINUX_KEXEC_H */ diff --git a/include/linux/reboot.h b/include/linux/reboot.h index c5a05e16edb..2d4dd23168d 100644 --- a/include/linux/reboot.h +++ b/include/linux/reboot.h @@ -52,7 +52,8 @@ extern void machine_halt(void); extern void machine_power_off(void); extern void machine_shutdown(void); -extern void machine_crash_shutdown(void); +struct pt_regs; +extern void machine_crash_shutdown(struct pt_regs *); #endif diff --git a/kernel/kexec.c b/kernel/kexec.c index a0411b3bd54..277f22afe74 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -18,6 +18,8 @@ #include #include #include +#include + #include #include #include @@ -32,6 +34,13 @@ struct resource crashk_res = { .flags = IORESOURCE_BUSY | IORESOURCE_MEM }; +int kexec_should_crash(struct task_struct *p) +{ + if (in_interrupt() || !p->pid || p->pid == 1 || panic_on_oops) + return 1; + return 0; +} + /* * When kexec transitions to the new kernel there is a one-to-one * mapping between physical and virtual addresses. On processors @@ -1010,7 +1019,7 @@ asmlinkage long compat_sys_kexec_load(unsigned long entry, } #endif -void crash_kexec(void) +void crash_kexec(struct pt_regs *regs) { struct kimage *image; int locked; @@ -1028,7 +1037,7 @@ void crash_kexec(void) if (!locked) { image = xchg(&kexec_crash_image, NULL); if (image) { - machine_crash_shutdown(); + machine_crash_shutdown(regs); machine_kexec(image); } xchg(&kexec_lock, 0); diff --git a/kernel/panic.c b/kernel/panic.c index 66f43d33cd8..74ba5f3e46c 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -83,7 +83,7 @@ NORET_TYPE void panic(const char * fmt, ...) * everything else. * Do we want to call this before we try to display a message? */ - crash_kexec(); + crash_kexec(NULL); #ifdef CONFIG_SMP /* -- cgit v1.2.3-70-g09d2 From 72414d3f1d22fc3e311b162fca95c430048d38ce Mon Sep 17 00:00:00 2001 From: Maneesh Soni Date: Sat, 25 Jun 2005 14:58:28 -0700 Subject: [PATCH] kexec code cleanup o Following patch provides purely cosmetic changes and corrects CodingStyle guide lines related certain issues like below in kexec related files o braces for one line "if" statements, "for" loops, o more than 80 column wide lines, o No space after "while", "for" and "switch" key words o Changes: o take-2: Removed the extra tab before "case" key words. o take-3: Put operator at the end of line and space before "*/" Signed-off-by: Maneesh Soni Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/crash.c | 23 +-- arch/i386/kernel/machine_kexec.c | 16 +- arch/ppc/kernel/machine_kexec.c | 30 ++-- arch/ppc64/kernel/machine_kexec.c | 9 +- arch/s390/kernel/machine_kexec.c | 4 +- arch/x86_64/kernel/machine_kexec.c | 49 +++--- drivers/char/mem.c | 2 +- include/linux/kexec.h | 13 +- include/linux/syscalls.h | 6 +- kernel/kexec.c | 302 ++++++++++++++++++++----------------- 10 files changed, 243 insertions(+), 211 deletions(-) (limited to 'arch/s390') diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c index 8bdb4b6af0f..e5fab12f792 100644 --- a/arch/i386/kernel/crash.c +++ b/arch/i386/kernel/crash.c @@ -31,10 +31,11 @@ note_buf_t crash_notes[NR_CPUS]; /* 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) +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; @@ -44,26 +45,28 @@ static u32 *append_elf_note(u32 *buf, 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)) { + + 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 @@ -75,9 +78,8 @@ static void crash_save_this_cpu(struct pt_regs *regs, int cpu) 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)); - + buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, + sizeof(prstatus)); final_note(buf); } @@ -119,8 +121,8 @@ static void crash_save_self(struct pt_regs *saved_regs) { struct pt_regs regs; int cpu; - cpu = smp_processor_id(); + cpu = smp_processor_id(); if (saved_regs) crash_setup_regs(®s, saved_regs); else @@ -153,6 +155,7 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu) /* Assume hlt works */ __asm__("hlt"); for(;;); + return 1; } @@ -169,8 +172,8 @@ static void smp_send_nmi_allbutself(void) static void nmi_shootdown_cpus(void) { unsigned long msecs; - atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1); + atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1); /* Would it be better to replace the trap vector here? */ set_nmi_callback(crash_nmi_callback); /* Ensure the new callback function is set before sending diff --git a/arch/i386/kernel/machine_kexec.c b/arch/i386/kernel/machine_kexec.c index 671880415d1..52ed18d8b51 100644 --- a/arch/i386/kernel/machine_kexec.c +++ b/arch/i386/kernel/machine_kexec.c @@ -80,7 +80,8 @@ static void identity_map_page(unsigned long address) /* Identity map the page table entry */ pgtable_level1[level1_index] = address | L0_ATTR; pgtable_level2[level2_index] = __pa(pgtable_level1) | L1_ATTR; - set_64bit(&pgtable_level3[level3_index], __pa(pgtable_level2) | L2_ATTR); + set_64bit(&pgtable_level3[level3_index], + __pa(pgtable_level2) | L2_ATTR); /* Flush the tlb so the new mapping takes effect. * Global tlb entries are not flushed but that is not an issue. @@ -139,8 +140,10 @@ static void load_segments(void) } typedef asmlinkage NORET_TYPE void (*relocate_new_kernel_t)( - unsigned long indirection_page, unsigned long reboot_code_buffer, - unsigned long start_address, unsigned int has_pae) ATTRIB_NORET; + unsigned long indirection_page, + unsigned long reboot_code_buffer, + unsigned long start_address, + unsigned int has_pae) ATTRIB_NORET; const extern unsigned char relocate_new_kernel[]; extern void relocate_new_kernel_end(void); @@ -180,20 +183,23 @@ NORET_TYPE void machine_kexec(struct kimage *image) { unsigned long page_list; unsigned long reboot_code_buffer; + relocate_new_kernel_t rnk; /* Interrupts aren't acceptable while we reboot */ local_irq_disable(); /* Compute some offsets */ - reboot_code_buffer = page_to_pfn(image->control_code_page) << PAGE_SHIFT; + reboot_code_buffer = page_to_pfn(image->control_code_page) + << PAGE_SHIFT; page_list = image->head; /* Set up an identity mapping for the reboot_code_buffer */ identity_map_page(reboot_code_buffer); /* copy it out */ - memcpy((void *)reboot_code_buffer, relocate_new_kernel, relocate_new_kernel_size); + memcpy((void *)reboot_code_buffer, relocate_new_kernel, + relocate_new_kernel_size); /* The segment registers are funny things, they are * automatically loaded from a table, in memory wherever you diff --git a/arch/ppc/kernel/machine_kexec.c b/arch/ppc/kernel/machine_kexec.c index b82535357d6..84d65a87191 100644 --- a/arch/ppc/kernel/machine_kexec.c +++ b/arch/ppc/kernel/machine_kexec.c @@ -21,24 +21,23 @@ #include typedef NORET_TYPE void (*relocate_new_kernel_t)( - unsigned long indirection_page, unsigned long reboot_code_buffer, - unsigned long start_address) ATTRIB_NORET; + unsigned long indirection_page, + unsigned long reboot_code_buffer, + unsigned long start_address) ATTRIB_NORET; const extern unsigned char relocate_new_kernel[]; const extern unsigned int relocate_new_kernel_size; void machine_shutdown(void) { - if (ppc_md.machine_shutdown) { + if (ppc_md.machine_shutdown) ppc_md.machine_shutdown(); - } } void machine_crash_shutdown(struct pt_regs *regs) { - if (ppc_md.machine_crash_shutdown) { + if (ppc_md.machine_crash_shutdown) ppc_md.machine_crash_shutdown(); - } } /* @@ -48,9 +47,8 @@ void machine_crash_shutdown(struct pt_regs *regs) */ int machine_kexec_prepare(struct kimage *image) { - if (ppc_md.machine_kexec_prepare) { + if (ppc_md.machine_kexec_prepare) return ppc_md.machine_kexec_prepare(image); - } /* * Fail if platform doesn't provide its own machine_kexec_prepare * implementation. @@ -60,9 +58,8 @@ int machine_kexec_prepare(struct kimage *image) void machine_kexec_cleanup(struct kimage *image) { - if (ppc_md.machine_kexec_cleanup) { + if (ppc_md.machine_kexec_cleanup) ppc_md.machine_kexec_cleanup(image); - } } /* @@ -71,9 +68,9 @@ void machine_kexec_cleanup(struct kimage *image) */ NORET_TYPE void machine_kexec(struct kimage *image) { - if (ppc_md.machine_kexec) { + if (ppc_md.machine_kexec) ppc_md.machine_kexec(image); - } else { + else { /* * Fall back to normal restart if platform doesn't provide * its own kexec function, and user insist to kexec... @@ -83,7 +80,6 @@ NORET_TYPE void machine_kexec(struct kimage *image) for(;;); } - /* * This is a generic machine_kexec function suitable at least for * non-OpenFirmware embedded platforms. @@ -104,15 +100,15 @@ void machine_kexec_simple(struct kimage *image) /* we need both effective and real address here */ reboot_code_buffer = - (unsigned long)page_address(image->control_code_page); + (unsigned long)page_address(image->control_code_page); reboot_code_buffer_phys = virt_to_phys((void *)reboot_code_buffer); /* copy our kernel relocation code to the control code page */ - memcpy((void *)reboot_code_buffer, - relocate_new_kernel, relocate_new_kernel_size); + memcpy((void *)reboot_code_buffer, relocate_new_kernel, + relocate_new_kernel_size); flush_icache_range(reboot_code_buffer, - reboot_code_buffer + KEXEC_CONTROL_CODE_SIZE); + reboot_code_buffer + KEXEC_CONTROL_CODE_SIZE); printk(KERN_INFO "Bye!\n"); /* now call it */ diff --git a/arch/ppc64/kernel/machine_kexec.c b/arch/ppc64/kernel/machine_kexec.c index 06b25b59c8a..fdb2fc649d7 100644 --- a/arch/ppc64/kernel/machine_kexec.c +++ b/arch/ppc64/kernel/machine_kexec.c @@ -58,7 +58,7 @@ int machine_kexec_prepare(struct kimage *image) * handle the virtual mode, we must make sure no destination * overlaps kernel static data or bss. */ - for(i = 0; i < image->nr_segments; i++) + for (i = 0; i < image->nr_segments; i++) if (image->segment[i].mem < __pa(_end)) return -ETXTBSY; @@ -76,7 +76,7 @@ int machine_kexec_prepare(struct kimage *image) low = __pa(htab_address); high = low + (htab_hash_mask + 1) * HASH_GROUP_SIZE; - for(i = 0; i < image->nr_segments; i++) { + for (i = 0; i < image->nr_segments; i++) { begin = image->segment[i].mem; end = begin + image->segment[i].memsz; @@ -98,7 +98,7 @@ int machine_kexec_prepare(struct kimage *image) low = *basep; high = low + (*sizep); - for(i = 0; i < image->nr_segments; i++) { + for (i = 0; i < image->nr_segments; i++) { begin = image->segment[i].mem; end = begin + image->segment[i].memsz; @@ -274,7 +274,8 @@ union thread_union kexec_stack /* Our assembly helper, in kexec_stub.S */ extern NORET_TYPE void kexec_sequence(void *newstack, unsigned long start, - void *image, void *control, void (*clear_all)(void)) ATTRIB_NORET; + void *image, void *control, + void (*clear_all)(void)) ATTRIB_NORET; /* too late to fail here */ void machine_kexec(struct kimage *image) diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c index 7a94db76df4..2721c3a32b8 100644 --- a/arch/s390/kernel/machine_kexec.c +++ b/arch/s390/kernel/machine_kexec.c @@ -67,7 +67,7 @@ machine_kexec(struct kimage *image) ctl_clear_bit(0,28); on_each_cpu(kexec_halt_all_cpus, image, 0, 0); - for(;;); + for (;;); } static void @@ -85,7 +85,7 @@ kexec_halt_all_cpus(void *kernel_image) for_each_online_cpu(cpu) { if (cpu == smp_processor_id()) continue; - while(!smp_cpu_not_running(cpu)) + while (!smp_cpu_not_running(cpu)) cpu_relax(); } diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c index 200b5993f8d..60d1eff4156 100644 --- a/arch/x86_64/kernel/machine_kexec.c +++ b/arch/x86_64/kernel/machine_kexec.c @@ -32,29 +32,31 @@ #define L2_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY) #define L3_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY) -static void init_level2_page( - u64 *level2p, unsigned long addr) +static void init_level2_page(u64 *level2p, unsigned long addr) { unsigned long end_addr; + addr &= PAGE_MASK; end_addr = addr + LEVEL2_SIZE; - while(addr < end_addr) { + while (addr < end_addr) { *(level2p++) = addr | L1_ATTR; addr += LEVEL1_SIZE; } } -static int init_level3_page(struct kimage *image, - u64 *level3p, unsigned long addr, unsigned long last_addr) +static int init_level3_page(struct kimage *image, u64 *level3p, + unsigned long addr, unsigned long last_addr) { unsigned long end_addr; int result; + result = 0; addr &= PAGE_MASK; end_addr = addr + LEVEL3_SIZE; - while((addr < last_addr) && (addr < end_addr)) { + while ((addr < last_addr) && (addr < end_addr)) { struct page *page; u64 *level2p; + page = kimage_alloc_control_pages(image, 0); if (!page) { result = -ENOMEM; @@ -66,7 +68,7 @@ static int init_level3_page(struct kimage *image, addr += LEVEL2_SIZE; } /* clear the unused entries */ - while(addr < end_addr) { + while (addr < end_addr) { *(level3p++) = 0; addr += LEVEL2_SIZE; } @@ -75,17 +77,19 @@ out: } -static int init_level4_page(struct kimage *image, - u64 *level4p, unsigned long addr, unsigned long last_addr) +static int init_level4_page(struct kimage *image, u64 *level4p, + unsigned long addr, unsigned long last_addr) { unsigned long end_addr; int result; + result = 0; addr &= PAGE_MASK; end_addr = addr + LEVEL4_SIZE; - while((addr < last_addr) && (addr < end_addr)) { + while ((addr < last_addr) && (addr < end_addr)) { struct page *page; u64 *level3p; + page = kimage_alloc_control_pages(image, 0); if (!page) { result = -ENOMEM; @@ -100,11 +104,11 @@ static int init_level4_page(struct kimage *image, addr += LEVEL3_SIZE; } /* clear the unused entries */ - while(addr < end_addr) { + while (addr < end_addr) { *(level4p++) = 0; addr += LEVEL3_SIZE; } - out: +out: return result; } @@ -113,7 +117,7 @@ static int init_pgtable(struct kimage *image, unsigned long start_pgtable) { u64 *level4p; level4p = (u64 *)__va(start_pgtable); - return init_level4_page(image, level4p, 0, end_pfn << PAGE_SHIFT); + return init_level4_page(image, level4p, 0, end_pfn << PAGE_SHIFT); } static void set_idt(void *newidt, u16 limit) @@ -159,9 +163,10 @@ static void load_segments(void) #undef __STR } -typedef NORET_TYPE void (*relocate_new_kernel_t)( - unsigned long indirection_page, unsigned long control_code_buffer, - unsigned long start_address, unsigned long pgtable) ATTRIB_NORET; +typedef NORET_TYPE void (*relocate_new_kernel_t)(unsigned long indirection_page, + unsigned long control_code_buffer, + unsigned long start_address, + unsigned long pgtable) ATTRIB_NORET; const extern unsigned char relocate_new_kernel[]; const extern unsigned long relocate_new_kernel_size; @@ -172,17 +177,17 @@ int machine_kexec_prepare(struct kimage *image) int result; /* Calculate the offsets */ - start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT; + start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT; control_code_buffer = start_pgtable + 4096UL; /* Setup the identity mapped 64bit page table */ result = init_pgtable(image, start_pgtable); - if (result) { + if (result) return result; - } /* Place the code in the reboot code buffer */ - memcpy(__va(control_code_buffer), relocate_new_kernel, relocate_new_kernel_size); + memcpy(__va(control_code_buffer), relocate_new_kernel, + relocate_new_kernel_size); return 0; } @@ -207,8 +212,8 @@ NORET_TYPE void machine_kexec(struct kimage *image) local_irq_disable(); /* Calculate the offsets */ - page_list = image->head; - start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT; + page_list = image->head; + start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT; control_code_buffer = start_pgtable + 4096UL; /* Set the low half of the page table to my identity mapped diff --git a/drivers/char/mem.c b/drivers/char/mem.c index b64108dd765..42187381506 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -287,7 +287,7 @@ static ssize_t read_oldmem(struct file *file, char __user *buf, size_t read = 0, csize; int rc = 0; - while(count) { + while (count) { pfn = *ppos / PAGE_SIZE; if (pfn > saved_max_pfn) return read; diff --git a/include/linux/kexec.h b/include/linux/kexec.h index 7383173a3a9..c8468472aec 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -91,14 +91,17 @@ extern NORET_TYPE void machine_kexec(struct kimage *image) ATTRIB_NORET; extern int machine_kexec_prepare(struct kimage *image); extern void machine_kexec_cleanup(struct kimage *image); extern asmlinkage long sys_kexec_load(unsigned long entry, - unsigned long nr_segments, struct kexec_segment __user *segments, - unsigned long flags); + unsigned long nr_segments, + struct kexec_segment __user *segments, + unsigned long flags); #ifdef CONFIG_COMPAT extern asmlinkage long compat_sys_kexec_load(unsigned long entry, - unsigned long nr_segments, struct compat_kexec_segment __user *segments, - unsigned long flags); + unsigned long nr_segments, + struct compat_kexec_segment __user *segments, + unsigned long flags); #endif -extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order); +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 *); extern struct kimage *kexec_image; diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 7ba8f8f747a..52830b6d94e 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -159,9 +159,9 @@ asmlinkage long sys_shutdown(int, int); asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user *arg); asmlinkage long sys_restart_syscall(void); -asmlinkage long sys_kexec_load(unsigned long entry, - unsigned long nr_segments, struct kexec_segment __user *segments, - unsigned long flags); +asmlinkage long sys_kexec_load(unsigned long entry, unsigned long nr_segments, + struct kexec_segment __user *segments, + unsigned long flags); asmlinkage long sys_exit(int error_code); asmlinkage void sys_exit_group(int error_code); diff --git a/kernel/kexec.c b/kernel/kexec.c index 277f22afe74..7843548cf2d 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -87,12 +87,15 @@ int kexec_should_crash(struct task_struct *p) */ #define KIMAGE_NO_DEST (-1UL) -static int kimage_is_destination_range( - struct kimage *image, unsigned long start, unsigned long end); -static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mask, unsigned long dest); +static int kimage_is_destination_range(struct kimage *image, + unsigned long start, unsigned long end); +static struct page *kimage_alloc_page(struct kimage *image, + unsigned int gfp_mask, + unsigned long dest); static int do_kimage_alloc(struct kimage **rimage, unsigned long entry, - unsigned long nr_segments, struct kexec_segment __user *segments) + unsigned long nr_segments, + struct kexec_segment __user *segments) { size_t segment_bytes; struct kimage *image; @@ -102,9 +105,9 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry, /* Allocate a controlling structure */ result = -ENOMEM; image = kmalloc(sizeof(*image), GFP_KERNEL); - if (!image) { + if (!image) goto out; - } + memset(image, 0, sizeof(*image)); image->head = 0; image->entry = &image->head; @@ -145,6 +148,7 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry, result = -EADDRNOTAVAIL; for (i = 0; i < nr_segments; i++) { unsigned long mstart, mend; + mstart = image->segment[i].mem; mend = mstart + image->segment[i].memsz; if ((mstart & ~PAGE_MASK) || (mend & ~PAGE_MASK)) @@ -159,12 +163,13 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry, * easy explanation as one segment stops on another. */ result = -EINVAL; - for(i = 0; i < nr_segments; i++) { + for (i = 0; i < nr_segments; i++) { unsigned long mstart, mend; unsigned long j; + mstart = image->segment[i].mem; mend = mstart + image->segment[i].memsz; - for(j = 0; j < i; j++) { + for (j = 0; j < i; j++) { unsigned long pstart, pend; pstart = image->segment[j].mem; pend = pstart + image->segment[j].memsz; @@ -180,25 +185,25 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry, * later on. */ result = -EINVAL; - for(i = 0; i < nr_segments; i++) { + for (i = 0; i < nr_segments; i++) { if (image->segment[i].bufsz > image->segment[i].memsz) goto out; } - result = 0; - out: - if (result == 0) { +out: + if (result == 0) *rimage = image; - } else { + else kfree(image); - } + return result; } static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry, - unsigned long nr_segments, struct kexec_segment __user *segments) + unsigned long nr_segments, + struct kexec_segment __user *segments) { int result; struct kimage *image; @@ -206,9 +211,9 @@ static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry, /* Allocate and initialize a controlling structure */ image = NULL; result = do_kimage_alloc(&image, entry, nr_segments, segments); - if (result) { + if (result) goto out; - } + *rimage = image; /* @@ -218,7 +223,7 @@ static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry, */ result = -ENOMEM; image->control_code_page = kimage_alloc_control_pages(image, - get_order(KEXEC_CONTROL_CODE_SIZE)); + get_order(KEXEC_CONTROL_CODE_SIZE)); if (!image->control_code_page) { printk(KERN_ERR "Could not allocate control_code_buffer\n"); goto out; @@ -226,16 +231,17 @@ static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry, result = 0; out: - if (result == 0) { + if (result == 0) *rimage = image; - } else { + else kfree(image); - } + return result; } static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry, - unsigned long nr_segments, struct kexec_segment *segments) + unsigned long nr_segments, + struct kexec_segment *segments) { int result; struct kimage *image; @@ -250,9 +256,8 @@ static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry, /* Allocate and initialize a controlling structure */ result = do_kimage_alloc(&image, entry, nr_segments, segments); - if (result) { + if (result) goto out; - } /* Enable the special crash kernel control page * allocation policy. @@ -272,6 +277,7 @@ static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry, result = -EADDRNOTAVAIL; for (i = 0; i < nr_segments; i++) { unsigned long mstart, mend; + mstart = image->segment[i].mem; mend = mstart + image->segment[i].memsz - 1; /* Ensure we are within the crash kernel limits */ @@ -279,7 +285,6 @@ static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry, goto out; } - /* * Find a location for the control code buffer, and add * the vector of segments so that it's pages will also be @@ -287,80 +292,84 @@ static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry, */ result = -ENOMEM; image->control_code_page = kimage_alloc_control_pages(image, - get_order(KEXEC_CONTROL_CODE_SIZE)); + get_order(KEXEC_CONTROL_CODE_SIZE)); if (!image->control_code_page) { printk(KERN_ERR "Could not allocate control_code_buffer\n"); goto out; } result = 0; - out: - if (result == 0) { +out: + if (result == 0) *rimage = image; - } else { + else kfree(image); - } + return result; } -static int kimage_is_destination_range( - struct kimage *image, unsigned long start, unsigned long end) +static int kimage_is_destination_range(struct kimage *image, + unsigned long start, + unsigned long end) { unsigned long i; for (i = 0; i < image->nr_segments; i++) { unsigned long mstart, mend; + mstart = image->segment[i].mem; - mend = mstart + image->segment[i].memsz; - if ((end > mstart) && (start < mend)) { + mend = mstart + image->segment[i].memsz; + if ((end > mstart) && (start < mend)) return 1; - } } + return 0; } -static struct page *kimage_alloc_pages(unsigned int gfp_mask, unsigned int order) +static struct page *kimage_alloc_pages(unsigned int gfp_mask, + unsigned int order) { struct page *pages; + pages = alloc_pages(gfp_mask, order); if (pages) { unsigned int count, i; pages->mapping = NULL; pages->private = order; count = 1 << order; - for(i = 0; i < count; i++) { + for (i = 0; i < count; i++) SetPageReserved(pages + i); - } } + return pages; } static void kimage_free_pages(struct page *page) { unsigned int order, count, i; + order = page->private; count = 1 << order; - for(i = 0; i < count; i++) { + for (i = 0; i < count; i++) ClearPageReserved(page + i); - } __free_pages(page, order); } static void kimage_free_page_list(struct list_head *list) { struct list_head *pos, *next; + list_for_each_safe(pos, next, list) { struct page *page; page = list_entry(pos, struct page, lru); list_del(&page->lru); - kimage_free_pages(page); } } -static struct page *kimage_alloc_normal_control_pages( - struct kimage *image, unsigned int order) +static struct page *kimage_alloc_normal_control_pages(struct kimage *image, + unsigned int order) { /* Control pages are special, they are the intermediaries * that are needed while we copy the rest of the pages @@ -387,6 +396,7 @@ static struct page *kimage_alloc_normal_control_pages( */ do { unsigned long pfn, epfn, addr, eaddr; + pages = kimage_alloc_pages(GFP_KERNEL, order); if (!pages) break; @@ -395,12 +405,12 @@ static struct page *kimage_alloc_normal_control_pages( addr = pfn << PAGE_SHIFT; eaddr = epfn << PAGE_SHIFT; if ((epfn >= (KEXEC_CONTROL_MEMORY_LIMIT >> PAGE_SHIFT)) || - kimage_is_destination_range(image, addr, eaddr)) - { + kimage_is_destination_range(image, addr, eaddr)) { list_add(&pages->lru, &extra_pages); pages = NULL; } - } while(!pages); + } while (!pages); + if (pages) { /* Remember the allocated page... */ list_add(&pages->lru, &image->control_pages); @@ -420,12 +430,12 @@ static struct page *kimage_alloc_normal_control_pages( * For now it is simpler to just free the pages. */ kimage_free_page_list(&extra_pages); - return pages; + return pages; } -static struct page *kimage_alloc_crash_control_pages( - struct kimage *image, unsigned int order) +static struct page *kimage_alloc_crash_control_pages(struct kimage *image, + unsigned int order) { /* Control pages are special, they are the intermediaries * that are needed while we copy the rest of the pages @@ -450,21 +460,22 @@ static struct page *kimage_alloc_crash_control_pages( */ unsigned long hole_start, hole_end, size; struct page *pages; + pages = NULL; size = (1 << order) << PAGE_SHIFT; hole_start = (image->control_page + (size - 1)) & ~(size - 1); hole_end = hole_start + size - 1; - while(hole_end <= crashk_res.end) { + while (hole_end <= crashk_res.end) { unsigned long i; - if (hole_end > KEXEC_CONTROL_MEMORY_LIMIT) { + + if (hole_end > KEXEC_CONTROL_MEMORY_LIMIT) break; - } - if (hole_end > crashk_res.end) { + if (hole_end > crashk_res.end) break; - } /* See if I overlap any of the segments */ - for(i = 0; i < image->nr_segments; i++) { + for (i = 0; i < image->nr_segments; i++) { unsigned long mstart, mend; + mstart = image->segment[i].mem; mend = mstart + image->segment[i].memsz - 1; if ((hole_end >= mstart) && (hole_start <= mend)) { @@ -480,18 +491,19 @@ static struct page *kimage_alloc_crash_control_pages( break; } } - if (pages) { + if (pages) image->control_page = hole_end; - } + return pages; } -struct page *kimage_alloc_control_pages( - struct kimage *image, unsigned int order) +struct page *kimage_alloc_control_pages(struct kimage *image, + unsigned int order) { struct page *pages = NULL; - switch(image->type) { + + switch (image->type) { case KEXEC_TYPE_DEFAULT: pages = kimage_alloc_normal_control_pages(image, order); break; @@ -499,43 +511,46 @@ struct page *kimage_alloc_control_pages( pages = kimage_alloc_crash_control_pages(image, order); break; } + return pages; } static int kimage_add_entry(struct kimage *image, kimage_entry_t entry) { - if (*image->entry != 0) { + if (*image->entry != 0) image->entry++; - } + if (image->entry == image->last_entry) { kimage_entry_t *ind_page; struct page *page; + page = kimage_alloc_page(image, GFP_KERNEL, KIMAGE_NO_DEST); - if (!page) { + if (!page) return -ENOMEM; - } + ind_page = page_address(page); *image->entry = virt_to_phys(ind_page) | IND_INDIRECTION; image->entry = ind_page; - image->last_entry = - ind_page + ((PAGE_SIZE/sizeof(kimage_entry_t)) - 1); + image->last_entry = ind_page + + ((PAGE_SIZE/sizeof(kimage_entry_t)) - 1); } *image->entry = entry; image->entry++; *image->entry = 0; + return 0; } -static int kimage_set_destination( - struct kimage *image, unsigned long destination) +static int kimage_set_destination(struct kimage *image, + unsigned long destination) { int result; destination &= PAGE_MASK; result = kimage_add_entry(image, destination | IND_DESTINATION); - if (result == 0) { + if (result == 0) image->destination = destination; - } + return result; } @@ -546,9 +561,9 @@ static int kimage_add_page(struct kimage *image, unsigned long page) page &= PAGE_MASK; result = kimage_add_entry(image, page | IND_SOURCE); - if (result == 0) { + if (result == 0) image->destination += PAGE_SIZE; - } + return result; } @@ -564,10 +579,11 @@ static void kimage_free_extra_pages(struct kimage *image) } static int kimage_terminate(struct kimage *image) { - if (*image->entry != 0) { + if (*image->entry != 0) image->entry++; - } + *image->entry = IND_DONE; + return 0; } @@ -591,26 +607,24 @@ static void kimage_free(struct kimage *image) if (!image) return; + kimage_free_extra_pages(image); for_each_kimage_entry(image, ptr, entry) { if (entry & IND_INDIRECTION) { /* Free the previous indirection page */ - if (ind & IND_INDIRECTION) { + if (ind & IND_INDIRECTION) kimage_free_entry(ind); - } /* Save this indirection page until we are * done with it. */ ind = entry; } - else if (entry & IND_SOURCE) { + else if (entry & IND_SOURCE) kimage_free_entry(entry); - } } /* Free the final indirection page */ - if (ind & IND_INDIRECTION) { + if (ind & IND_INDIRECTION) kimage_free_entry(ind); - } /* Handle any machine specific cleanup */ machine_kexec_cleanup(image); @@ -620,26 +634,28 @@ static void kimage_free(struct kimage *image) kfree(image); } -static kimage_entry_t *kimage_dst_used(struct kimage *image, unsigned long page) +static kimage_entry_t *kimage_dst_used(struct kimage *image, + unsigned long page) { kimage_entry_t *ptr, entry; unsigned long destination = 0; for_each_kimage_entry(image, ptr, entry) { - if (entry & IND_DESTINATION) { + if (entry & IND_DESTINATION) destination = entry & PAGE_MASK; - } else if (entry & IND_SOURCE) { - if (page == destination) { + if (page == destination) return ptr; - } destination += PAGE_SIZE; } } + return 0; } -static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mask, unsigned long destination) +static struct page *kimage_alloc_page(struct kimage *image, + unsigned int gfp_mask, + unsigned long destination) { /* * Here we implement safeguards to ensure that a source page @@ -679,11 +695,11 @@ static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mas /* Allocate a page, if we run out of memory give up */ page = kimage_alloc_pages(gfp_mask, 0); - if (!page) { + if (!page) return 0; - } /* If the page cannot be used file it away */ - if (page_to_pfn(page) > (KEXEC_SOURCE_MEMORY_LIMIT >> PAGE_SHIFT)) { + if (page_to_pfn(page) > + (KEXEC_SOURCE_MEMORY_LIMIT >> PAGE_SHIFT)) { list_add(&page->lru, &image->unuseable_pages); continue; } @@ -694,7 +710,8 @@ static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mas break; /* If the page is not a destination page use it */ - if (!kimage_is_destination_range(image, addr, addr + PAGE_SIZE)) + if (!kimage_is_destination_range(image, addr, + addr + PAGE_SIZE)) break; /* @@ -727,11 +744,12 @@ static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mas list_add(&page->lru, &image->dest_pages); } } + return page; } static int kimage_load_normal_segment(struct kimage *image, - struct kexec_segment *segment) + struct kexec_segment *segment) { unsigned long maddr; unsigned long ubytes, mbytes; @@ -745,34 +763,36 @@ static int kimage_load_normal_segment(struct kimage *image, maddr = segment->mem; result = kimage_set_destination(image, maddr); - if (result < 0) { + if (result < 0) goto out; - } - while(mbytes) { + + while (mbytes) { struct page *page; char *ptr; size_t uchunk, mchunk; + page = kimage_alloc_page(image, GFP_HIGHUSER, maddr); if (page == 0) { result = -ENOMEM; goto out; } - result = kimage_add_page(image, page_to_pfn(page) << PAGE_SHIFT); - if (result < 0) { + result = kimage_add_page(image, page_to_pfn(page) + << PAGE_SHIFT); + if (result < 0) goto out; - } + ptr = kmap(page); /* Start with a clear page */ memset(ptr, 0, PAGE_SIZE); ptr += maddr & ~PAGE_MASK; mchunk = PAGE_SIZE - (maddr & ~PAGE_MASK); - if (mchunk > mbytes) { + if (mchunk > mbytes) mchunk = mbytes; - } + uchunk = mchunk; - if (uchunk > ubytes) { + if (uchunk > ubytes) uchunk = ubytes; - } + result = copy_from_user(ptr, buf, uchunk); kunmap(page); if (result) { @@ -784,12 +804,12 @@ static int kimage_load_normal_segment(struct kimage *image, buf += mchunk; mbytes -= mchunk; } - out: +out: return result; } static int kimage_load_crash_segment(struct kimage *image, - struct kexec_segment *segment) + struct kexec_segment *segment) { /* For crash dumps kernels we simply copy the data from * user space to it's destination. @@ -805,10 +825,11 @@ static int kimage_load_crash_segment(struct kimage *image, ubytes = segment->bufsz; mbytes = segment->memsz; maddr = segment->mem; - while(mbytes) { + while (mbytes) { struct page *page; char *ptr; size_t uchunk, mchunk; + page = pfn_to_page(maddr >> PAGE_SHIFT); if (page == 0) { result = -ENOMEM; @@ -817,9 +838,9 @@ static int kimage_load_crash_segment(struct kimage *image, ptr = kmap(page); ptr += maddr & ~PAGE_MASK; mchunk = PAGE_SIZE - (maddr & ~PAGE_MASK); - if (mchunk > mbytes) { + if (mchunk > mbytes) mchunk = mbytes; - } + uchunk = mchunk; if (uchunk > ubytes) { uchunk = ubytes; @@ -837,15 +858,16 @@ static int kimage_load_crash_segment(struct kimage *image, buf += mchunk; mbytes -= mchunk; } - out: +out: return result; } static int kimage_load_segment(struct kimage *image, - struct kexec_segment *segment) + struct kexec_segment *segment) { int result = -ENOMEM; - switch(image->type) { + + switch (image->type) { case KEXEC_TYPE_DEFAULT: result = kimage_load_normal_segment(image, segment); break; @@ -853,6 +875,7 @@ static int kimage_load_segment(struct kimage *image, result = kimage_load_crash_segment(image, segment); break; } + return result; } @@ -885,9 +908,9 @@ static struct kimage *kexec_crash_image = NULL; */ static int kexec_lock = 0; -asmlinkage long sys_kexec_load(unsigned long entry, - unsigned long nr_segments, struct kexec_segment __user *segments, - unsigned long flags) +asmlinkage long sys_kexec_load(unsigned long entry, unsigned long nr_segments, + struct kexec_segment __user *segments, + unsigned long flags) { struct kimage **dest_image, *image; int locked; @@ -907,9 +930,7 @@ asmlinkage long sys_kexec_load(unsigned long entry, /* Verify we are on the appropriate architecture */ if (((flags & KEXEC_ARCH_MASK) != KEXEC_ARCH) && ((flags & KEXEC_ARCH_MASK) != KEXEC_ARCH_DEFAULT)) - { return -EINVAL; - } /* Put an artificial cap on the number * of segments passed to kexec_load. @@ -929,58 +950,59 @@ asmlinkage long sys_kexec_load(unsigned long entry, * KISS: always take the mutex. */ locked = xchg(&kexec_lock, 1); - if (locked) { + if (locked) return -EBUSY; - } + dest_image = &kexec_image; - if (flags & KEXEC_ON_CRASH) { + if (flags & KEXEC_ON_CRASH) dest_image = &kexec_crash_image; - } if (nr_segments > 0) { unsigned long i; + /* Loading another kernel to reboot into */ - if ((flags & KEXEC_ON_CRASH) == 0) { - result = kimage_normal_alloc(&image, entry, nr_segments, segments); - } + if ((flags & KEXEC_ON_CRASH) == 0) + result = kimage_normal_alloc(&image, entry, + nr_segments, segments); /* Loading another kernel to switch to if this one crashes */ else if (flags & KEXEC_ON_CRASH) { /* Free any current crash dump kernel before * we corrupt it. */ kimage_free(xchg(&kexec_crash_image, NULL)); - result = kimage_crash_alloc(&image, entry, nr_segments, segments); + result = kimage_crash_alloc(&image, entry, + nr_segments, segments); } - if (result) { + if (result) goto out; - } + result = machine_kexec_prepare(image); - if (result) { + if (result) goto out; - } - for(i = 0; i < nr_segments; i++) { + + for (i = 0; i < nr_segments; i++) { result = kimage_load_segment(image, &image->segment[i]); - if (result) { + if (result) goto out; - } } result = kimage_terminate(image); - if (result) { + if (result) goto out; - } } /* Install the new kernel, and Uninstall the old */ image = xchg(dest_image, image); - out: +out: xchg(&kexec_lock, 0); /* Release the mutex */ kimage_free(image); + return result; } #ifdef CONFIG_COMPAT asmlinkage long compat_sys_kexec_load(unsigned long entry, - unsigned long nr_segments, struct compat_kexec_segment __user *segments, - unsigned long flags) + unsigned long nr_segments, + struct compat_kexec_segment __user *segments, + unsigned long flags) { struct compat_kexec_segment in; struct kexec_segment out, __user *ksegments; @@ -989,20 +1011,17 @@ asmlinkage long compat_sys_kexec_load(unsigned long entry, /* Don't allow clients that don't understand the native * architecture to do anything. */ - if ((flags & KEXEC_ARCH_MASK) == KEXEC_ARCH_DEFAULT) { + if ((flags & KEXEC_ARCH_MASK) == KEXEC_ARCH_DEFAULT) return -EINVAL; - } - if (nr_segments > KEXEC_SEGMENT_MAX) { + if (nr_segments > KEXEC_SEGMENT_MAX) return -EINVAL; - } ksegments = compat_alloc_user_space(nr_segments * sizeof(out)); for (i=0; i < nr_segments; i++) { result = copy_from_user(&in, &segments[i], sizeof(in)); - if (result) { + if (result) return -EFAULT; - } out.buf = compat_ptr(in.buf); out.bufsz = in.bufsz; @@ -1010,9 +1029,8 @@ asmlinkage long compat_sys_kexec_load(unsigned long entry, out.memsz = in.memsz; result = copy_to_user(&ksegments[i], &out, sizeof(out)); - if (result) { + if (result) return -EFAULT; - } } return sys_kexec_load(entry, nr_segments, ksegments, flags); -- cgit v1.2.3-70-g09d2 From d5950b4355049092739bea97d1bdc14433126cc5 Mon Sep 17 00:00:00 2001 From: Sam Ravnborg Date: Mon, 11 Jul 2005 21:03:49 -0700 Subject: [NET]: add a top-level Networking menu to *config Create a new top-level menu named "Networking" thus moving net related options and protocol selection way from the drivers menu and up on the top-level where they belong. To implement this all architectures has to source "net/Kconfig" before drivers/*/Kconfig in their Kconfig file. This change has been implemented for all architectures. Device drivers for ordinary NIC's are still to be found in the Device Drivers section, but Bluetooth, IrDA and ax25 are located with their corresponding menu entries under the new networking menu item. Signed-off-by: Sam Ravnborg Signed-off-by: David S. Miller --- arch/alpha/Kconfig | 2 ++ arch/arm/Kconfig | 4 +++- arch/arm26/Kconfig | 4 +++- arch/cris/Kconfig | 4 +++- arch/frv/Kconfig | 2 ++ arch/h8300/Kconfig | 4 +++- arch/i386/Kconfig | 2 ++ arch/ia64/Kconfig | 2 ++ arch/m32r/Kconfig | 2 ++ arch/m68k/Kconfig | 2 ++ arch/m68knommu/Kconfig | 2 ++ arch/mips/Kconfig | 2 ++ arch/parisc/Kconfig | 2 ++ arch/ppc/Kconfig | 2 ++ arch/ppc64/Kconfig | 2 ++ arch/s390/Kconfig | 4 +++- arch/sh/Kconfig | 2 ++ arch/sh64/Kconfig | 2 ++ arch/sparc/Kconfig | 2 ++ arch/sparc64/Kconfig | 4 +++- arch/um/Kconfig | 4 +++- arch/v850/Kconfig | 4 +++- arch/x86_64/Kconfig | 2 ++ arch/xtensa/Kconfig | 2 ++ drivers/Kconfig | 2 +- drivers/net/Kconfig | 3 +++ net/Kconfig | 10 +++++----- 27 files changed, 65 insertions(+), 14 deletions(-) (limited to 'arch/s390') diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index c5739d6309d..083c5df42d3 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -596,6 +596,8 @@ source "fs/Kconfig.binfmt" endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 8752751f998..45462714caf 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -700,6 +700,8 @@ config APM endmenu +source "net/Kconfig" + menu "Device Drivers" source "drivers/base/Kconfig" @@ -732,7 +734,7 @@ source "drivers/ieee1394/Kconfig" source "drivers/message/i2o/Kconfig" -source "net/Kconfig" +source "drivers/net/Kconfig" source "drivers/isdn/Kconfig" diff --git a/arch/arm26/Kconfig b/arch/arm26/Kconfig index dc0c1936969..1f037326730 100644 --- a/arch/arm26/Kconfig +++ b/arch/arm26/Kconfig @@ -183,6 +183,8 @@ source "mm/Kconfig" endmenu +source "net/Kconfig" + source "drivers/base/Kconfig" source "drivers/parport/Kconfig" @@ -193,7 +195,7 @@ source "drivers/block/Kconfig" source "drivers/md/Kconfig" -source "net/Kconfig" +source "drivers/net/Kconfig" source "drivers/ide/Kconfig" diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig index f848e376149..e5979d68e35 100644 --- a/arch/cris/Kconfig +++ b/arch/cris/Kconfig @@ -122,6 +122,8 @@ source arch/cris/arch-v10/Kconfig endmenu +source "net/Kconfig" + # bring in ETRAX built-in drivers menu "Drivers for built-in interfaces" source arch/cris/arch-v10/drivers/Kconfig @@ -149,7 +151,7 @@ source "drivers/ieee1394/Kconfig" source "drivers/message/i2o/Kconfig" -source "net/Kconfig" +source "drivers/net/Kconfig" source "drivers/isdn/Kconfig" diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index c93f95146cc..ec85c0d6c6d 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -346,6 +346,8 @@ source "fs/Kconfig.binfmt" endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig index 62a89e812e3..375f2a8ff3b 100644 --- a/arch/h8300/Kconfig +++ b/arch/h8300/Kconfig @@ -55,6 +55,8 @@ source "fs/Kconfig.binfmt" endmenu +source "net/Kconfig" + source "drivers/base/Kconfig" source "drivers/mtd/Kconfig" @@ -65,7 +67,7 @@ source "drivers/ide/Kconfig" source "arch/h8300/Kconfig.ide" -source "net/Kconfig" +source "drivers/net/Kconfig" # # input - input/joystick depends on it. As does USB. diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 6c02336fe2e..a801d9d4860 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -1285,6 +1285,8 @@ source "fs/Kconfig.binfmt" endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 01b78e7f992..2e08942339a 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -423,6 +423,8 @@ endmenu endif +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig index 42ca8a39798..7772951df31 100644 --- a/arch/m32r/Kconfig +++ b/arch/m32r/Kconfig @@ -359,6 +359,8 @@ source "fs/Kconfig.binfmt" endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig index 691a2469ff3..178c4a3fbb7 100644 --- a/arch/m68k/Kconfig +++ b/arch/m68k/Kconfig @@ -450,6 +450,8 @@ source "drivers/zorro/Kconfig" endmenu +source "net/Kconfig" + source "drivers/Kconfig" menu "Character devices" diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig index dbfcdc8e608..117f183f0b4 100644 --- a/arch/m68knommu/Kconfig +++ b/arch/m68knommu/Kconfig @@ -575,6 +575,8 @@ config PM endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index bd9de7b00c0..b578239146b 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -1640,6 +1640,8 @@ config PM endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index ce327c799b4..1c2d8743523 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -190,6 +190,8 @@ source "fs/Kconfig.binfmt" endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig index 23b0d2f662c..b833cbcd77f 100644 --- a/arch/ppc/Kconfig +++ b/arch/ppc/Kconfig @@ -1355,6 +1355,8 @@ config PIN_TLB depends on ADVANCED_OPTIONS && 8xx endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig index f804f25232a..fdd8afba715 100644 --- a/arch/ppc64/Kconfig +++ b/arch/ppc64/Kconfig @@ -429,6 +429,8 @@ config CMDLINE endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 6600ee87f89..477ac2758bd 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -465,6 +465,8 @@ config KEXEC endmenu +source "net/Kconfig" + config PCMCIA bool default n @@ -475,7 +477,7 @@ source "drivers/scsi/Kconfig" source "drivers/s390/Kconfig" -source "net/Kconfig" +source "drivers/net/Kconfig" source "fs/Kconfig" diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index a7c8bfc1160..adc8109f8b7 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -784,6 +784,8 @@ config EMBEDDED_RAMDISK_IMAGE endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/sh64/Kconfig b/arch/sh64/Kconfig index 708e59736a4..4c3e5334adb 100644 --- a/arch/sh64/Kconfig +++ b/arch/sh64/Kconfig @@ -268,6 +268,8 @@ source "fs/Kconfig.binfmt" endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 7a117ef473c..aca028aa29b 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -268,6 +268,8 @@ source "mm/Kconfig" endmenu +source "net/Kconfig" + source "drivers/Kconfig" if !SUN4 diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig index 6a4733683f0..140607870f1 100644 --- a/arch/sparc64/Kconfig +++ b/arch/sparc64/Kconfig @@ -525,6 +525,8 @@ source "mm/Kconfig" endmenu +source "net/Kconfig" + source "drivers/base/Kconfig" source "drivers/video/Kconfig" @@ -551,7 +553,7 @@ endif source "drivers/ieee1394/Kconfig" -source "net/Kconfig" +source "drivers/net/Kconfig" source "drivers/isdn/Kconfig" diff --git a/arch/um/Kconfig b/arch/um/Kconfig index 6682c788364..f945444df49 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -275,6 +275,8 @@ endmenu source "init/Kconfig" +source "net/Kconfig" + source "drivers/base/Kconfig" source "arch/um/Kconfig_char" @@ -287,7 +289,7 @@ config NETDEVICES source "arch/um/Kconfig_net" -source "net/Kconfig" +source "drivers/net/Kconfig" source "fs/Kconfig" diff --git a/arch/v850/Kconfig b/arch/v850/Kconfig index 27febd6ffa8..89c053b6c2c 100644 --- a/arch/v850/Kconfig +++ b/arch/v850/Kconfig @@ -250,6 +250,8 @@ source "fs/Kconfig.binfmt" endmenu +source "net/Kconfig" + ############################################################################# source "drivers/base/Kconfig" @@ -283,7 +285,7 @@ source "drivers/ieee1394/Kconfig" source "drivers/message/i2o/Kconfig" -source "net/Kconfig" +source "drivers/net/Kconfig" source "drivers/isdn/Kconfig" diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index d09437b5c48..4b8326177c5 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -515,6 +515,8 @@ config UID16 endmenu +source "net/Kconfig" + source drivers/Kconfig source "drivers/firmware/Kconfig" diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig index c9b5d298e3c..2b6257bec4c 100644 --- a/arch/xtensa/Kconfig +++ b/arch/xtensa/Kconfig @@ -228,6 +228,8 @@ source "fs/Kconfig.binfmt" endmenu +source "net/Kconfig" + source "drivers/Kconfig" source "fs/Kconfig" diff --git a/drivers/Kconfig b/drivers/Kconfig index aed4a9b97c1..34efb2150e6 100644 --- a/drivers/Kconfig +++ b/drivers/Kconfig @@ -28,7 +28,7 @@ source "drivers/message/i2o/Kconfig" source "drivers/macintosh/Kconfig" -source "net/Kconfig" +source "drivers/net/Kconfig" source "drivers/isdn/Kconfig" diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 2b55687f6ee..9a07ff7a777 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -3,6 +3,8 @@ # Network device configuration # +menu "Network device support" + config NETDEVICES depends on NET bool "Network device support" @@ -2547,3 +2549,4 @@ config NETCONSOLE If you want to log kernel messages over the network, enable this. See for details. +endmenu diff --git a/net/Kconfig b/net/Kconfig index b9ca0a581bd..f46fc326c00 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -2,7 +2,7 @@ # Network configuration # -menu "Networking support" +menu "Networking" config NET bool "Networking support" @@ -10,7 +10,9 @@ config NET Unless you really know what you are doing, you should say Y here. The reason is that some programs need kernel networking support even when running on a stand-alone machine that isn't connected to any - other computer. If you are upgrading from an older kernel, you + other computer. + + If you are upgrading from an older kernel, you should consider updating your networking tools too because changes in the kernel and the tools often go hand in hand. The tools are contained in the package net-tools, the location and version number @@ -640,7 +642,5 @@ source "net/irda/Kconfig" source "net/bluetooth/Kconfig" -source "drivers/net/Kconfig" - -endmenu +endmenu # Networking -- cgit v1.2.3-70-g09d2 From 068e1b94bbd268f375349f68531829c8b7c210bc Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 13 Jul 2005 01:10:46 -0700 Subject: [PATCH] s390: fadvise hint values. Add special case for the POSIX_FADV_DONTNEED and POSIX_FADV_NOREUSE hint values for s390-64. The user space values in the s390-64 glibc headers for these two defines have always been 6 and 7 instead of 4 and 5. All 64 bit applications therefore use the "wrong" values. To get these applications working without recompiling the kernel needs to accept the "wrong" values. Since the values for s390-31 are 4 and 5 the compat wrapper for fadvise64 and fadvise64_64 need to rewrite the values for 31 bit system calls. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/compat_linux.c | 38 ++++++++++++++++++++++++++++++++++++++ arch/s390/kernel/compat_wrapper.S | 4 ++-- include/linux/fadvise.h | 10 ++++++++++ 3 files changed, 50 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index 61405622287..18610cea03a 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -58,6 +58,7 @@ #include #include #include +#include #include #include @@ -1043,3 +1044,40 @@ sys32_timer_create(clockid_t which_clock, struct compat_sigevent *se32, return ret; } + +/* + * 31 bit emulation wrapper functions for sys_fadvise64/fadvise64_64. + * These need to rewrite the advise values for POSIX_FADV_{DONTNEED,NOREUSE} + * because the 31 bit values differ from the 64 bit values. + */ + +asmlinkage long +sys32_fadvise64(int fd, loff_t offset, size_t len, int advise) +{ + if (advise == 4) + advise = POSIX_FADV_DONTNEED; + else if (advise == 5) + advise = POSIX_FADV_NOREUSE; + return sys_fadvise64(fd, offset, len, advise); +} + +struct fadvise64_64_args { + int fd; + long long offset; + long long len; + int advice; +}; + +asmlinkage long +sys32_fadvise64_64(struct fadvise64_64_args __user *args) +{ + struct fadvise64_64_args a; + + if ( copy_from_user(&a, args, sizeof(a)) ) + return -EFAULT; + if (a.advice == 4) + a.advice = POSIX_FADV_DONTNEED; + else if (a.advice == 5) + a.advice = POSIX_FADV_NOREUSE; + return sys_fadvise64_64(a.fd, a.offset, a.len, a.advice); +} diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index bf529739c8a..799a98eac92 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -1251,12 +1251,12 @@ sys32_fadvise64_wrapper: or %r3,%r4 # get low word of 64bit loff_t llgfr %r4,%r5 # size_t (unsigned long) lgfr %r5,%r6 # int - jg sys_fadvise64 + jg sys32_fadvise64 .globl sys32_fadvise64_64_wrapper sys32_fadvise64_64_wrapper: llgtr %r2,%r2 # struct fadvise64_64_args * - jg s390_fadvise64_64 + jg sys32_fadvise64_64 .globl sys32_clock_settime_wrapper sys32_clock_settime_wrapper: diff --git a/include/linux/fadvise.h b/include/linux/fadvise.h index 6fc656dfb93..e8e747139b9 100644 --- a/include/linux/fadvise.h +++ b/include/linux/fadvise.h @@ -5,7 +5,17 @@ #define POSIX_FADV_RANDOM 1 /* Expect random page references. */ #define POSIX_FADV_SEQUENTIAL 2 /* Expect sequential page references. */ #define POSIX_FADV_WILLNEED 3 /* Will need these pages. */ + +/* + * The advise values for POSIX_FADV_DONTNEED and POSIX_ADV_NOREUSE + * for s390-64 differ from the values for the rest of the world. + */ +#if defined(__s390x__) +#define POSIX_FADV_DONTNEED 6 /* Don't need these pages. */ +#define POSIX_FADV_NOREUSE 7 /* Data will be accessed once. */ +#else #define POSIX_FADV_DONTNEED 4 /* Don't need these pages. */ #define POSIX_FADV_NOREUSE 5 /* Data will be accessed once. */ +#endif #endif /* FADVISE_H_INCLUDED */ -- cgit v1.2.3-70-g09d2 From 59586e5a262a29361c45c929ea3253d4aec830b0 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Tue, 26 Jul 2005 11:36:01 -0600 Subject: [PATCH] Don't export machine_restart, machine_halt, or machine_power_off. machine_restart, machine_halt and machine_power_off are machine specific hooks deep into the reboot logic, that modules have no business messing with. Usually code should be calling kernel_restart, kernel_halt, kernel_power_off, or emergency_restart. So don't export machine_restart, machine_halt, and machine_power_off so we can catch buggy users. Signed-off-by: Eric W. Biederman Signed-off-by: Linus Torvalds --- arch/alpha/kernel/process.c | 3 --- arch/arm/kernel/process.c | 4 ---- arch/arm26/kernel/process.c | 5 ----- arch/cris/kernel/process.c | 6 ------ arch/h8300/kernel/process.c | 6 ------ arch/i386/kernel/reboot.c | 5 ----- arch/i386/mach-visws/reboot.c | 5 ----- arch/i386/mach-voyager/voyager_basic.c | 5 ----- arch/ia64/kernel/process.c | 5 ----- arch/m32r/kernel/process.c | 6 ------ arch/m68k/kernel/process.c | 6 ------ arch/m68knommu/kernel/process.c | 6 ------ arch/mips/kernel/reset.c | 5 ----- arch/parisc/kernel/process.c | 6 ------ arch/ppc/kernel/setup.c | 6 ------ arch/ppc64/kernel/setup.c | 3 --- arch/s390/kernel/setup.c | 6 ------ arch/sh/kernel/process.c | 6 ------ arch/sparc/kernel/process.c | 6 ------ arch/sparc64/kernel/power.c | 2 -- arch/sparc64/kernel/process.c | 4 ---- arch/um/kernel/reboot.c | 6 ------ arch/v850/kernel/anna.c | 6 ------ arch/v850/kernel/as85ep1.c | 6 ------ arch/v850/kernel/fpga85e2c.c | 6 ------ arch/v850/kernel/rte_cb.c | 6 ------ arch/v850/kernel/sim.c | 6 ------ arch/v850/kernel/sim85e2.c | 5 ----- arch/x86_64/kernel/reboot.c | 5 ----- 29 files changed, 152 deletions(-) (limited to 'arch/s390') diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c index 4933f3ce583..fa98dae3cd9 100644 --- a/arch/alpha/kernel/process.c +++ b/arch/alpha/kernel/process.c @@ -165,7 +165,6 @@ machine_restart(char *restart_cmd) common_shutdown(LINUX_REBOOT_CMD_RESTART, restart_cmd); } -EXPORT_SYMBOL(machine_restart); void machine_halt(void) @@ -173,7 +172,6 @@ machine_halt(void) common_shutdown(LINUX_REBOOT_CMD_HALT, NULL); } -EXPORT_SYMBOL(machine_halt); void machine_power_off(void) @@ -181,7 +179,6 @@ machine_power_off(void) common_shutdown(LINUX_REBOOT_CMD_POWER_OFF, NULL); } -EXPORT_SYMBOL(machine_power_off); /* Used by sysrq-p, among others. I don't believe r9-r15 are ever saved in the context it's used. */ diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index bbea636ff68..409db6d5ec9 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -131,7 +131,6 @@ void machine_halt(void) { } -EXPORT_SYMBOL(machine_halt); void machine_power_off(void) { @@ -139,7 +138,6 @@ void machine_power_off(void) pm_power_off(); } -EXPORT_SYMBOL(machine_power_off); void machine_restart(char * __unused) { @@ -169,8 +167,6 @@ void machine_restart(char * __unused) while (1); } -EXPORT_SYMBOL(machine_restart); - void __show_regs(struct pt_regs *regs) { unsigned long flags = condition_codes(regs); diff --git a/arch/arm26/kernel/process.c b/arch/arm26/kernel/process.c index 46aea6ac194..9eb9964d32a 100644 --- a/arch/arm26/kernel/process.c +++ b/arch/arm26/kernel/process.c @@ -103,9 +103,6 @@ void machine_power_off(void) { } -EXPORT_SYMBOL(machine_halt); -EXPORT_SYMBOL(machine_power_off); - void machine_restart(char * __unused) { /* @@ -136,8 +133,6 @@ void machine_restart(char * __unused) while (1); } -EXPORT_SYMBOL(machine_restart); - void show_regs(struct pt_regs * regs) { unsigned long flags; diff --git a/arch/cris/kernel/process.c b/arch/cris/kernel/process.c index 7645c4d4de6..a5ad2b67585 100644 --- a/arch/cris/kernel/process.c +++ b/arch/cris/kernel/process.c @@ -214,8 +214,6 @@ void machine_restart(char *cmd) hard_reset_now(); } -EXPORT_SYMBOL(machine_restart); - /* * Similar to machine_power_off, but don't shut off power. Add code * here to freeze the system for e.g. post-mortem debug purpose when @@ -226,16 +224,12 @@ void machine_halt(void) { } -EXPORT_SYMBOL(machine_halt); - /* If or when software power-off is implemented, add code here. */ void machine_power_off(void) { } -EXPORT_SYMBOL(machine_power_off); - /* * When a process does an "exec", machine state like FPU and debug * registers need to be reset. This is a hook function for that. diff --git a/arch/h8300/kernel/process.c b/arch/h8300/kernel/process.c index b5f83e9f04d..27f1fce64ce 100644 --- a/arch/h8300/kernel/process.c +++ b/arch/h8300/kernel/process.c @@ -90,8 +90,6 @@ void machine_restart(char * __unused) __asm__("jmp @@0"); } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { local_irq_disable(); @@ -99,8 +97,6 @@ void machine_halt(void) for (;;); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { local_irq_disable(); @@ -108,8 +104,6 @@ void machine_power_off(void) for (;;); } -EXPORT_SYMBOL(machine_power_off); - void show_regs(struct pt_regs * regs) { printk("\nPC: %08lx Status: %02x", diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c index b3e58484996..1b24bc7271a 100644 --- a/arch/i386/kernel/reboot.c +++ b/arch/i386/kernel/reboot.c @@ -337,14 +337,10 @@ void machine_restart(char * __unused) machine_real_restart(jump_to_bios, sizeof(jump_to_bios)); } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { lapic_shutdown(); @@ -355,5 +351,4 @@ void machine_power_off(void) pm_power_off(); } -EXPORT_SYMBOL(machine_power_off); diff --git a/arch/i386/mach-visws/reboot.c b/arch/i386/mach-visws/reboot.c index 95e4676594e..9e9296676f9 100644 --- a/arch/i386/mach-visws/reboot.c +++ b/arch/i386/mach-visws/reboot.c @@ -22,8 +22,6 @@ void machine_restart(char * __unused) outb(PIIX4_RESET_VAL, PIIX4_RESET_PORT); } -EXPORT_SYMBOL(machine_restart); - void machine_power_off(void) { unsigned short pm_status; @@ -43,10 +41,7 @@ void machine_power_off(void) outl(PIIX_SPECIAL_STOP, 0xCFC); } -EXPORT_SYMBOL(machine_power_off); - void machine_halt(void) { } -EXPORT_SYMBOL(machine_halt); diff --git a/arch/i386/mach-voyager/voyager_basic.c b/arch/i386/mach-voyager/voyager_basic.c index 8680080a6a8..b3eda46e0fe 100644 --- a/arch/i386/mach-voyager/voyager_basic.c +++ b/arch/i386/mach-voyager/voyager_basic.c @@ -278,8 +278,6 @@ machine_restart(char *cmd) } } -EXPORT_SYMBOL(machine_restart); - void mca_nmi_hook(void) { @@ -315,12 +313,9 @@ machine_halt(void) machine_power_off(); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { if (pm_power_off) pm_power_off(); } -EXPORT_SYMBOL(machine_power_off); diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c index e484910246a..66e84060980 100644 --- a/arch/ia64/kernel/process.c +++ b/arch/ia64/kernel/process.c @@ -807,16 +807,12 @@ machine_restart (char *restart_cmd) (*efi.reset_system)(EFI_RESET_WARM, 0, 0, NULL); } -EXPORT_SYMBOL(machine_restart); - void machine_halt (void) { cpu_halt(); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off (void) { @@ -825,4 +821,3 @@ machine_power_off (void) machine_halt(); } -EXPORT_SYMBOL(machine_power_off); diff --git a/arch/m32r/kernel/process.c b/arch/m32r/kernel/process.c index b556c3cf649..ea13a8f4d8b 100644 --- a/arch/m32r/kernel/process.c +++ b/arch/m32r/kernel/process.c @@ -115,8 +115,6 @@ void machine_restart(char *__unused) cpu_relax(); } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { printk("Please push reset button!\n"); @@ -124,15 +122,11 @@ void machine_halt(void) cpu_relax(); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { /* M32R_FIXME */ } -EXPORT_SYMBOL(machine_power_off); - static int __init idle_setup (char *str) { if (!strncmp(str, "poll", 4)) { diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c index 93b043e2a43..11b1b90ba6b 100644 --- a/arch/m68k/kernel/process.c +++ b/arch/m68k/kernel/process.c @@ -113,8 +113,6 @@ void machine_restart(char * __unused) for (;;); } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { if (mach_halt) @@ -122,8 +120,6 @@ void machine_halt(void) for (;;); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { if (mach_power_off) @@ -131,8 +127,6 @@ void machine_power_off(void) for (;;); } -EXPORT_SYMBOL(machine_power_off); - void show_regs(struct pt_regs * regs) { printk("\n"); diff --git a/arch/m68knommu/kernel/process.c b/arch/m68knommu/kernel/process.c index c4a33f265dc..82e7ec88880 100644 --- a/arch/m68knommu/kernel/process.c +++ b/arch/m68knommu/kernel/process.c @@ -80,8 +80,6 @@ void machine_restart(char * __unused) for (;;); } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { if (mach_halt) @@ -89,8 +87,6 @@ void machine_halt(void) for (;;); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { if (mach_power_off) @@ -98,8 +94,6 @@ void machine_power_off(void) for (;;); } -EXPORT_SYMBOL(machine_power_off); - void show_regs(struct pt_regs * regs) { printk(KERN_NOTICE "\n"); diff --git a/arch/mips/kernel/reset.c b/arch/mips/kernel/reset.c index 7e0a9821931..ae2ba67b7ef 100644 --- a/arch/mips/kernel/reset.c +++ b/arch/mips/kernel/reset.c @@ -26,18 +26,13 @@ void machine_restart(char *command) _machine_restart(command); } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { _machine_halt(); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { _machine_power_off(); } -EXPORT_SYMBOL(machine_power_off); diff --git a/arch/parisc/kernel/process.c b/arch/parisc/kernel/process.c index 46e4a6881f1..4fc04501d5e 100644 --- a/arch/parisc/kernel/process.c +++ b/arch/parisc/kernel/process.c @@ -150,8 +150,6 @@ void machine_restart(char *cmd) } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { /* @@ -160,8 +158,6 @@ void machine_halt(void) */ } -EXPORT_SYMBOL(machine_halt); - /* * This routine is called from sys_reboot to actually turn off the @@ -187,8 +183,6 @@ void machine_power_off(void) KERN_EMERG "Please power this system off now."); } -EXPORT_SYMBOL(machine_power_off); - /* * Create a kernel thread diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c index c42f7532693..929e5d1cc7f 100644 --- a/arch/ppc/kernel/setup.c +++ b/arch/ppc/kernel/setup.c @@ -121,8 +121,6 @@ void machine_restart(char *cmd) ppc_md.restart(cmd); } -EXPORT_SYMBOL(machine_restart); - void machine_power_off(void) { #ifdef CONFIG_NVRAM @@ -131,8 +129,6 @@ void machine_power_off(void) ppc_md.power_off(); } -EXPORT_SYMBOL(machine_power_off); - void machine_halt(void) { #ifdef CONFIG_NVRAM @@ -141,8 +137,6 @@ void machine_halt(void) ppc_md.halt(); } -EXPORT_SYMBOL(machine_halt); - void (*pm_power_off)(void) = machine_power_off; #ifdef CONFIG_TAU diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c index d1b33f0b26c..e80f10c8982 100644 --- a/arch/ppc64/kernel/setup.c +++ b/arch/ppc64/kernel/setup.c @@ -694,7 +694,6 @@ void machine_restart(char *cmd) local_irq_disable(); while (1) ; } -EXPORT_SYMBOL(machine_restart); void machine_power_off(void) { @@ -707,7 +706,6 @@ void machine_power_off(void) local_irq_disable(); while (1) ; } -EXPORT_SYMBOL(machine_power_off); void machine_halt(void) { @@ -720,7 +718,6 @@ void machine_halt(void) local_irq_disable(); while (1) ; } -EXPORT_SYMBOL(machine_halt); static int ppc64_panic_event(struct notifier_block *this, unsigned long event, void *ptr) diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index b6d740ac0e6..a12183989a7 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -299,24 +299,18 @@ void machine_restart(char *command) _machine_restart(command); } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { console_unblank(); _machine_halt(); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { console_unblank(); _machine_power_off(); } -EXPORT_SYMBOL(machine_power_off); - static void __init add_memory_hole(unsigned long start, unsigned long end) { diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c index 3d024590c24..6dce9d0b81f 100644 --- a/arch/sh/kernel/process.c +++ b/arch/sh/kernel/process.c @@ -80,8 +80,6 @@ void machine_restart(char * __unused) "mov.l @%1, %0" : : "r" (0x10000000), "r" (0x80000001)); } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { #if defined(CONFIG_SH_HS7751RVOIP) @@ -96,8 +94,6 @@ void machine_halt(void) cpu_sleep(); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { #if defined(CONFIG_SH_HS7751RVOIP) @@ -110,8 +106,6 @@ void machine_power_off(void) #endif } -EXPORT_SYMBOL(machine_power_off); - void show_regs(struct pt_regs * regs) { printk("\n"); diff --git a/arch/sparc/kernel/process.c b/arch/sparc/kernel/process.c index 2c216ffeea9..29e72b57d4f 100644 --- a/arch/sparc/kernel/process.c +++ b/arch/sparc/kernel/process.c @@ -158,8 +158,6 @@ void machine_halt(void) panic("Halt failed!"); } -EXPORT_SYMBOL(machine_halt); - void machine_restart(char * cmd) { char *p; @@ -180,8 +178,6 @@ void machine_restart(char * cmd) panic("Reboot failed!"); } -EXPORT_SYMBOL(machine_restart); - void machine_power_off(void) { #ifdef CONFIG_SUN_AUXIO @@ -191,8 +187,6 @@ void machine_power_off(void) machine_halt(); } -EXPORT_SYMBOL(machine_power_off); - static DEFINE_SPINLOCK(sparc_backtrace_lock); void __show_backtrace(unsigned long fp) diff --git a/arch/sparc64/kernel/power.c b/arch/sparc64/kernel/power.c index 533104c7907..946cee0257e 100644 --- a/arch/sparc64/kernel/power.c +++ b/arch/sparc64/kernel/power.c @@ -69,8 +69,6 @@ void machine_power_off(void) machine_halt(); } -EXPORT_SYMBOL(machine_power_off); - #ifdef CONFIG_PCI static int powerd(void *__unused) { diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c index cffb1c8ab4f..07424b07593 100644 --- a/arch/sparc64/kernel/process.c +++ b/arch/sparc64/kernel/process.c @@ -124,8 +124,6 @@ void machine_halt(void) panic("Halt failed!"); } -EXPORT_SYMBOL(machine_halt); - void machine_alt_power_off(void) { if (!serial_console && prom_palette) @@ -154,8 +152,6 @@ void machine_restart(char * cmd) panic("Reboot failed!"); } -EXPORT_SYMBOL(machine_restart); - static void show_regwindow32(struct pt_regs *regs) { struct reg_window32 __user *rw; diff --git a/arch/um/kernel/reboot.c b/arch/um/kernel/reboot.c index fcec51da1d3..a637e885c58 100644 --- a/arch/um/kernel/reboot.c +++ b/arch/um/kernel/reboot.c @@ -49,23 +49,17 @@ void machine_restart(char * __unused) CHOOSE_MODE(reboot_tt(), reboot_skas()); } -EXPORT_SYMBOL(machine_restart); - void machine_power_off(void) { uml_cleanup(); CHOOSE_MODE(halt_tt(), halt_skas()); } -EXPORT_SYMBOL(machine_power_off); - void machine_halt(void) { machine_power_off(); } -EXPORT_SYMBOL(machine_halt); - /* * Overrides for Emacs so that we follow Linus's tabbing style. * Emacs will notice this stuff at the end of the file and automatically diff --git a/arch/v850/kernel/anna.c b/arch/v850/kernel/anna.c index 6aaeab5e8a4..d0502e14243 100644 --- a/arch/v850/kernel/anna.c +++ b/arch/v850/kernel/anna.c @@ -132,8 +132,6 @@ void machine_restart (char *__unused) asm ("jmp r0"); /* Jump to the reset vector. */ } -EXPORT_SYMBOL(machine_restart); - void machine_halt (void) { #ifdef CONFIG_RESET_GUARD @@ -145,15 +143,11 @@ void machine_halt (void) asm ("halt; nop; nop; nop; nop; nop"); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off (void) { machine_halt (); } -EXPORT_SYMBOL(machine_power_off); - /* Called before configuring an on-chip UART. */ void anna_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud) { diff --git a/arch/v850/kernel/as85ep1.c b/arch/v850/kernel/as85ep1.c index 4059b1df11b..d78c5e4ea9b 100644 --- a/arch/v850/kernel/as85ep1.c +++ b/arch/v850/kernel/as85ep1.c @@ -160,8 +160,6 @@ void machine_restart (char *__unused) asm ("jmp r0"); /* Jump to the reset vector. */ } -EXPORT_SYMBOL(machine_restart); - void machine_halt (void) { #ifdef CONFIG_RESET_GUARD @@ -173,15 +171,11 @@ void machine_halt (void) asm ("halt; nop; nop; nop; nop; nop"); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off (void) { machine_halt (); } -EXPORT_SYMBOL(machine_power_off); - /* Called before configuring an on-chip UART. */ void as85ep1_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud) { diff --git a/arch/v850/kernel/fpga85e2c.c b/arch/v850/kernel/fpga85e2c.c index 4bac5149b3c..d8094519ad8 100644 --- a/arch/v850/kernel/fpga85e2c.c +++ b/arch/v850/kernel/fpga85e2c.c @@ -121,22 +121,16 @@ void machine_halt (void) } } -EXPORT_SYMBOL(machine_halt); - void machine_restart (char *__unused) { machine_halt (); } -EXPORT_SYMBOL(machine_restart); - void machine_power_off (void) { machine_halt (); } -EXPORT_SYMBOL(machine_power_off); - /* Interrupts */ diff --git a/arch/v850/kernel/rte_cb.c b/arch/v850/kernel/rte_cb.c index 7ba397f77ac..0c794b9e0f9 100644 --- a/arch/v850/kernel/rte_cb.c +++ b/arch/v850/kernel/rte_cb.c @@ -67,8 +67,6 @@ void machine_restart (char *__unused) asm ("jmp r0"); /* Jump to the reset vector. */ } -EXPORT_SYMBOL(machine_restart); - /* This says `HALt.' in LEDese. */ static unsigned char halt_leds_msg[] = { 0x76, 0x77, 0x38, 0xF8 }; @@ -89,15 +87,11 @@ void machine_halt (void) asm ("halt; nop; nop; nop; nop; nop"); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off (void) { machine_halt (); } -EXPORT_SYMBOL(machine_power_off); - /* Animated LED display for timer tick. */ diff --git a/arch/v850/kernel/sim.c b/arch/v850/kernel/sim.c index 4f31da96263..e2cc5580fa2 100644 --- a/arch/v850/kernel/sim.c +++ b/arch/v850/kernel/sim.c @@ -104,24 +104,18 @@ void machine_restart (char *__unused) V850_SIM_SYSCALL (exit, 0); } -EXPORT_SYMBOL(machine_restart); - void machine_halt (void) { V850_SIM_SYSCALL (write, 1, "HALT\n", 5); V850_SIM_SYSCALL (exit, 0); } -EXPORT_SYMBOL(machine_halt); - void machine_power_off (void) { V850_SIM_SYSCALL (write, 1, "POWER OFF\n", 10); V850_SIM_SYSCALL (exit, 0); } -EXPORT_SYMBOL(machine_power_off); - /* Load data from a file called NAME into ram. The address and length of the data image are returned in ADDR and LEN. */ diff --git a/arch/v850/kernel/sim85e2.c b/arch/v850/kernel/sim85e2.c index 93a722b516b..9111613fb53 100644 --- a/arch/v850/kernel/sim85e2.c +++ b/arch/v850/kernel/sim85e2.c @@ -184,18 +184,13 @@ void machine_halt (void) for (;;) {} } -EXPORT_SYMBOL(machine_halt); - void machine_restart (char *__unused) { machine_halt (); } -EXPORT_SYMBOL(machine_restart); - void machine_power_off (void) { machine_halt (); } -EXPORT_SYMBOL(machine_power_off); diff --git a/arch/x86_64/kernel/reboot.c b/arch/x86_64/kernel/reboot.c index 57e71dbdfd6..0515d326663 100644 --- a/arch/x86_64/kernel/reboot.c +++ b/arch/x86_64/kernel/reboot.c @@ -150,18 +150,13 @@ void machine_restart(char * __unused) } } -EXPORT_SYMBOL(machine_restart); - void machine_halt(void) { } -EXPORT_SYMBOL(machine_halt); - void machine_power_off(void) { if (pm_power_off) pm_power_off(); } -EXPORT_SYMBOL(machine_power_off); -- cgit v1.2.3-70-g09d2 From 951f22d5b1f0eaae35dafc669e3774a0c2084d10 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 27 Jul 2005 11:44:57 -0700 Subject: [PATCH] s390: spin lock retry Split spin lock and r/w lock implementation into a single try which is done inline and an out of line function that repeatedly tries to get the lock before doing the cpu_relax(). Add a system control to set the number of retries before a cpu is yielded. The reason for the spin lock retry is that the diagnose 0x44 that is used to give up the virtual cpu is quite expensive. For spin locks that are held only for a short period of time the costs of the diagnoses outweights the savings for spin locks that are held for a longer timer. The default retry count is 1000. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/head64.S | 3 - arch/s390/kernel/setup.c | 6 -- arch/s390/lib/Makefile | 4 +- arch/s390/lib/spinlock.c | 133 +++++++++++++++++++++++ include/asm-s390/lowcore.h | 4 +- include/asm-s390/processor.h | 5 +- include/asm-s390/spinlock.h | 252 ++++++++++++++----------------------------- include/linux/sysctl.h | 1 + kernel/sysctl.c | 12 ++- 9 files changed, 230 insertions(+), 190 deletions(-) create mode 100644 arch/s390/lib/spinlock.c (limited to 'arch/s390') diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index f525c0c2125..28c50bdf7d4 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -658,10 +658,8 @@ startup:basr %r13,0 # get base # la %r1,0f-.LPG1(%r13) # set program check address stg %r1,__LC_PGM_NEW_PSW+8 - mvc __LC_DIAG44_OPCODE(8),.Lnop-.LPG1(%r13) diag 0,0,0x44 # test diag 0x44 oi 7(%r12),32 # set diag44 flag - mvc __LC_DIAG44_OPCODE(8),.Ldiag44-.LPG1(%r13) 0: # @@ -702,7 +700,6 @@ startup:basr %r13,0 # get base .L4malign:.quad 0xffffffffffc00000 .Lscan2g:.quad 0x80000000 + 0x20000 - 8 # 2GB + 128K - 8 .Lnop: .long 0x07000700 -.Ldiag44:.long 0x83000044 .org PARMAREA-64 .Lduct: .long 0,0,0,0,0,0,0,0 diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index a12183989a7..5ba5a5485da 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -431,12 +431,6 @@ setup_lowcore(void) ctl_set_bit(14, 29); } #endif -#ifdef CONFIG_ARCH_S390X - if (MACHINE_HAS_DIAG44) - lc->diag44_opcode = 0x83000044; - else - lc->diag44_opcode = 0x07000700; -#endif /* CONFIG_ARCH_S390X */ set_prefix((u32)(unsigned long) lc); } diff --git a/arch/s390/lib/Makefile b/arch/s390/lib/Makefile index a8758b1d20a..b701efa1f00 100644 --- a/arch/s390/lib/Makefile +++ b/arch/s390/lib/Makefile @@ -5,5 +5,5 @@ EXTRA_AFLAGS := -traditional lib-y += delay.o string.o -lib-$(CONFIG_ARCH_S390_31) += uaccess.o -lib-$(CONFIG_ARCH_S390X) += uaccess64.o +lib-$(CONFIG_ARCH_S390_31) += uaccess.o spinlock.o +lib-$(CONFIG_ARCH_S390X) += uaccess64.o spinlock.o diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c new file mode 100644 index 00000000000..888b5596c19 --- /dev/null +++ b/arch/s390/lib/spinlock.c @@ -0,0 +1,133 @@ +/* + * arch/s390/lib/spinlock.c + * Out of line spinlock code. + * + * S390 version + * Copyright (C) 2004 IBM Deutschland Entwicklung GmbH, IBM Corporation + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#include +#include +#include +#include +#include + +atomic_t spin_retry_counter; +int spin_retry = 1000; + +/** + * spin_retry= parameter + */ +static int __init spin_retry_setup(char *str) +{ + spin_retry = simple_strtoul(str, &str, 0); + return 1; +} +__setup("spin_retry=", spin_retry_setup); + +static inline void +_diag44(void) +{ +#ifdef __s390x__ + if (MACHINE_HAS_DIAG44) +#endif + asm volatile("diag 0,0,0x44"); +} + +void +_raw_spin_lock_wait(spinlock_t *lp, unsigned int pc) +{ + int count = spin_retry; + + while (1) { + if (count-- <= 0) { + _diag44(); + count = spin_retry; + } + atomic_inc(&spin_retry_counter); + if (_raw_compare_and_swap(&lp->lock, 0, pc) == 0) + return; + } +} +EXPORT_SYMBOL(_raw_spin_lock_wait); + +int +_raw_spin_trylock_retry(spinlock_t *lp, unsigned int pc) +{ + int count = spin_retry; + + while (count-- > 0) { + atomic_inc(&spin_retry_counter); + if (_raw_compare_and_swap(&lp->lock, 0, pc) == 0) + return 1; + } + return 0; +} +EXPORT_SYMBOL(_raw_spin_trylock_retry); + +void +_raw_read_lock_wait(rwlock_t *rw) +{ + unsigned int old; + int count = spin_retry; + + while (1) { + if (count-- <= 0) { + _diag44(); + count = spin_retry; + } + atomic_inc(&spin_retry_counter); + old = rw->lock & 0x7fffffffU; + if (_raw_compare_and_swap(&rw->lock, old, old + 1) == old) + return; + } +} +EXPORT_SYMBOL(_raw_read_lock_wait); + +int +_raw_read_trylock_retry(rwlock_t *rw) +{ + unsigned int old; + int count = spin_retry; + + while (count-- > 0) { + atomic_inc(&spin_retry_counter); + old = rw->lock & 0x7fffffffU; + if (_raw_compare_and_swap(&rw->lock, old, old + 1) == old) + return 1; + } + return 0; +} +EXPORT_SYMBOL(_raw_read_trylock_retry); + +void +_raw_write_lock_wait(rwlock_t *rw) +{ + int count = spin_retry; + + while (1) { + if (count-- <= 0) { + _diag44(); + count = spin_retry; + } + atomic_inc(&spin_retry_counter); + if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0) + return; + } +} +EXPORT_SYMBOL(_raw_write_lock_wait); + +int +_raw_write_trylock_retry(rwlock_t *rw) +{ + int count = spin_retry; + + while (count-- > 0) { + atomic_inc(&spin_retry_counter); + if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0) + return 1; + } + return 0; +} +EXPORT_SYMBOL(_raw_write_trylock_retry); diff --git a/include/asm-s390/lowcore.h b/include/asm-s390/lowcore.h index 76b5b19c0ae..afe6a9f9b0a 100644 --- a/include/asm-s390/lowcore.h +++ b/include/asm-s390/lowcore.h @@ -90,7 +90,6 @@ #define __LC_SYSTEM_TIMER 0x278 #define __LC_LAST_UPDATE_CLOCK 0x280 #define __LC_STEAL_CLOCK 0x288 -#define __LC_DIAG44_OPCODE 0x290 #define __LC_KERNEL_STACK 0xD40 #define __LC_THREAD_INFO 0xD48 #define __LC_ASYNC_STACK 0xD50 @@ -286,8 +285,7 @@ struct _lowcore __u64 system_timer; /* 0x278 */ __u64 last_update_clock; /* 0x280 */ __u64 steal_clock; /* 0x288 */ - __u32 diag44_opcode; /* 0x290 */ - __u8 pad8[0xc00-0x294]; /* 0x294 */ + __u8 pad8[0xc00-0x290]; /* 0x290 */ /* System info area */ __u64 save_area[16]; /* 0xc00 */ __u8 pad9[0xd40-0xc80]; /* 0xc80 */ diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h index 8bd14de69e3..4ec652ebb3b 100644 --- a/include/asm-s390/processor.h +++ b/include/asm-s390/processor.h @@ -203,7 +203,10 @@ unsigned long get_wchan(struct task_struct *p); # define cpu_relax() asm volatile ("diag 0,0,68" : : : "memory") #else /* __s390x__ */ # define cpu_relax() \ - asm volatile ("ex 0,%0" : : "i" (__LC_DIAG44_OPCODE) : "memory") + do { \ + if (MACHINE_HAS_DIAG44) \ + asm volatile ("diag 0,0,68" : : : "memory"); \ + } while (0) #endif /* __s390x__ */ /* diff --git a/include/asm-s390/spinlock.h b/include/asm-s390/spinlock.h index 53cc736b982..8ff10300f7e 100644 --- a/include/asm-s390/spinlock.h +++ b/include/asm-s390/spinlock.h @@ -11,21 +11,16 @@ #ifndef __ASM_SPINLOCK_H #define __ASM_SPINLOCK_H -#ifdef __s390x__ -/* - * Grmph, take care of %&#! user space programs that include - * asm/spinlock.h. The diagnose is only available in kernel - * context. - */ -#ifdef __KERNEL__ -#include -#define __DIAG44_INSN "ex" -#define __DIAG44_OPERAND __LC_DIAG44_OPCODE -#else -#define __DIAG44_INSN "#" -#define __DIAG44_OPERAND 0 -#endif -#endif /* __s390x__ */ +static inline int +_raw_compare_and_swap(volatile unsigned int *lock, + unsigned int old, unsigned int new) +{ + asm volatile ("cs %0,%3,0(%4)" + : "=d" (old), "=m" (*lock) + : "0" (old), "d" (new), "a" (lock), "m" (*lock) + : "cc", "memory" ); + return old; +} /* * Simple spin lock operations. There are two variants, one clears IRQ's @@ -41,58 +36,35 @@ typedef struct { #endif } __attribute__ ((aligned (4))) spinlock_t; -#define SPIN_LOCK_UNLOCKED (spinlock_t) { 0 } -#define spin_lock_init(lp) do { (lp)->lock = 0; } while(0) +#define SPIN_LOCK_UNLOCKED (spinlock_t) { 0 } +#define spin_lock_init(lp) do { (lp)->lock = 0; } while(0) #define spin_unlock_wait(lp) do { barrier(); } while(((volatile spinlock_t *)(lp))->lock) -#define spin_is_locked(x) ((x)->lock != 0) +#define spin_is_locked(x) ((x)->lock != 0) #define _raw_spin_lock_flags(lock, flags) _raw_spin_lock(lock) -extern inline void _raw_spin_lock(spinlock_t *lp) +extern void _raw_spin_lock_wait(spinlock_t *lp, unsigned int pc); +extern int _raw_spin_trylock_retry(spinlock_t *lp, unsigned int pc); + +static inline void _raw_spin_lock(spinlock_t *lp) { -#ifndef __s390x__ - unsigned int reg1, reg2; - __asm__ __volatile__(" bras %0,1f\n" - "0: diag 0,0,68\n" - "1: slr %1,%1\n" - " cs %1,%0,0(%3)\n" - " jl 0b\n" - : "=&d" (reg1), "=&d" (reg2), "=m" (lp->lock) - : "a" (&lp->lock), "m" (lp->lock) - : "cc", "memory" ); -#else /* __s390x__ */ - unsigned long reg1, reg2; - __asm__ __volatile__(" bras %1,1f\n" - "0: " __DIAG44_INSN " 0,%4\n" - "1: slr %0,%0\n" - " cs %0,%1,0(%3)\n" - " jl 0b\n" - : "=&d" (reg1), "=&d" (reg2), "=m" (lp->lock) - : "a" (&lp->lock), "i" (__DIAG44_OPERAND), - "m" (lp->lock) : "cc", "memory" ); -#endif /* __s390x__ */ + unsigned long pc = (unsigned long) __builtin_return_address(0); + + if (unlikely(_raw_compare_and_swap(&lp->lock, 0, pc) != 0)) + _raw_spin_lock_wait(lp, pc); } -extern inline int _raw_spin_trylock(spinlock_t *lp) +static inline int _raw_spin_trylock(spinlock_t *lp) { - unsigned long reg; - unsigned int result; - - __asm__ __volatile__(" basr %1,0\n" - "0: cs %0,%1,0(%3)" - : "=d" (result), "=&d" (reg), "=m" (lp->lock) - : "a" (&lp->lock), "m" (lp->lock), "0" (0) - : "cc", "memory" ); - return !result; + unsigned long pc = (unsigned long) __builtin_return_address(0); + + if (likely(_raw_compare_and_swap(&lp->lock, 0, pc) == 0)) + return 1; + return _raw_spin_trylock_retry(lp, pc); } -extern inline void _raw_spin_unlock(spinlock_t *lp) +static inline void _raw_spin_unlock(spinlock_t *lp) { - unsigned int old; - - __asm__ __volatile__("cs %0,%3,0(%4)" - : "=d" (old), "=m" (lp->lock) - : "0" (lp->lock), "d" (0), "a" (lp) - : "cc", "memory" ); + _raw_compare_and_swap(&lp->lock, lp->lock, 0); } /* @@ -106,7 +78,7 @@ extern inline void _raw_spin_unlock(spinlock_t *lp) * read-locks. */ typedef struct { - volatile unsigned long lock; + volatile unsigned int lock; volatile unsigned long owner_pc; #ifdef CONFIG_PREEMPT unsigned int break_lock; @@ -129,123 +101,55 @@ typedef struct { */ #define write_can_lock(x) ((x)->lock == 0) -#ifndef __s390x__ -#define _raw_read_lock(rw) \ - asm volatile(" l 2,0(%1)\n" \ - " j 1f\n" \ - "0: diag 0,0,68\n" \ - "1: la 2,0(2)\n" /* clear high (=write) bit */ \ - " la 3,1(2)\n" /* one more reader */ \ - " cs 2,3,0(%1)\n" /* try to write new value */ \ - " jl 0b" \ - : "=m" ((rw)->lock) : "a" (&(rw)->lock), \ - "m" ((rw)->lock) : "2", "3", "cc", "memory" ) -#else /* __s390x__ */ -#define _raw_read_lock(rw) \ - asm volatile(" lg 2,0(%1)\n" \ - " j 1f\n" \ - "0: " __DIAG44_INSN " 0,%2\n" \ - "1: nihh 2,0x7fff\n" /* clear high (=write) bit */ \ - " la 3,1(2)\n" /* one more reader */ \ - " csg 2,3,0(%1)\n" /* try to write new value */ \ - " jl 0b" \ - : "=m" ((rw)->lock) \ - : "a" (&(rw)->lock), "i" (__DIAG44_OPERAND), \ - "m" ((rw)->lock) : "2", "3", "cc", "memory" ) -#endif /* __s390x__ */ - -#ifndef __s390x__ -#define _raw_read_unlock(rw) \ - asm volatile(" l 2,0(%1)\n" \ - " j 1f\n" \ - "0: diag 0,0,68\n" \ - "1: lr 3,2\n" \ - " ahi 3,-1\n" /* one less reader */ \ - " cs 2,3,0(%1)\n" \ - " jl 0b" \ - : "=m" ((rw)->lock) : "a" (&(rw)->lock), \ - "m" ((rw)->lock) : "2", "3", "cc", "memory" ) -#else /* __s390x__ */ -#define _raw_read_unlock(rw) \ - asm volatile(" lg 2,0(%1)\n" \ - " j 1f\n" \ - "0: " __DIAG44_INSN " 0,%2\n" \ - "1: lgr 3,2\n" \ - " bctgr 3,0\n" /* one less reader */ \ - " csg 2,3,0(%1)\n" \ - " jl 0b" \ - : "=m" ((rw)->lock) \ - : "a" (&(rw)->lock), "i" (__DIAG44_OPERAND), \ - "m" ((rw)->lock) : "2", "3", "cc", "memory" ) -#endif /* __s390x__ */ - -#ifndef __s390x__ -#define _raw_write_lock(rw) \ - asm volatile(" lhi 3,1\n" \ - " sll 3,31\n" /* new lock value = 0x80000000 */ \ - " j 1f\n" \ - "0: diag 0,0,68\n" \ - "1: slr 2,2\n" /* old lock value must be 0 */ \ - " cs 2,3,0(%1)\n" \ - " jl 0b" \ - : "=m" ((rw)->lock) : "a" (&(rw)->lock), \ - "m" ((rw)->lock) : "2", "3", "cc", "memory" ) -#else /* __s390x__ */ -#define _raw_write_lock(rw) \ - asm volatile(" llihh 3,0x8000\n" /* new lock value = 0x80...0 */ \ - " j 1f\n" \ - "0: " __DIAG44_INSN " 0,%2\n" \ - "1: slgr 2,2\n" /* old lock value must be 0 */ \ - " csg 2,3,0(%1)\n" \ - " jl 0b" \ - : "=m" ((rw)->lock) \ - : "a" (&(rw)->lock), "i" (__DIAG44_OPERAND), \ - "m" ((rw)->lock) : "2", "3", "cc", "memory" ) -#endif /* __s390x__ */ - -#ifndef __s390x__ -#define _raw_write_unlock(rw) \ - asm volatile(" slr 3,3\n" /* new lock value = 0 */ \ - " j 1f\n" \ - "0: diag 0,0,68\n" \ - "1: lhi 2,1\n" \ - " sll 2,31\n" /* old lock value must be 0x80000000 */ \ - " cs 2,3,0(%1)\n" \ - " jl 0b" \ - : "=m" ((rw)->lock) : "a" (&(rw)->lock), \ - "m" ((rw)->lock) : "2", "3", "cc", "memory" ) -#else /* __s390x__ */ -#define _raw_write_unlock(rw) \ - asm volatile(" slgr 3,3\n" /* new lock value = 0 */ \ - " j 1f\n" \ - "0: " __DIAG44_INSN " 0,%2\n" \ - "1: llihh 2,0x8000\n" /* old lock value must be 0x8..0 */\ - " csg 2,3,0(%1)\n" \ - " jl 0b" \ - : "=m" ((rw)->lock) \ - : "a" (&(rw)->lock), "i" (__DIAG44_OPERAND), \ - "m" ((rw)->lock) : "2", "3", "cc", "memory" ) -#endif /* __s390x__ */ - -#define _raw_read_trylock(lock) generic_raw_read_trylock(lock) - -extern inline int _raw_write_trylock(rwlock_t *rw) +extern void _raw_read_lock_wait(rwlock_t *lp); +extern int _raw_read_trylock_retry(rwlock_t *lp); +extern void _raw_write_lock_wait(rwlock_t *lp); +extern int _raw_write_trylock_retry(rwlock_t *lp); + +static inline void _raw_read_lock(rwlock_t *rw) +{ + unsigned int old; + old = rw->lock & 0x7fffffffU; + if (_raw_compare_and_swap(&rw->lock, old, old + 1) != old) + _raw_read_lock_wait(rw); +} + +static inline void _raw_read_unlock(rwlock_t *rw) +{ + unsigned int old, cmp; + + old = rw->lock; + do { + cmp = old; + old = _raw_compare_and_swap(&rw->lock, old, old - 1); + } while (cmp != old); +} + +static inline void _raw_write_lock(rwlock_t *rw) +{ + if (unlikely(_raw_compare_and_swap(&rw->lock, 0, 0x80000000) != 0)) + _raw_write_lock_wait(rw); +} + +static inline void _raw_write_unlock(rwlock_t *rw) +{ + _raw_compare_and_swap(&rw->lock, 0x80000000, 0); +} + +static inline int _raw_read_trylock(rwlock_t *rw) +{ + unsigned int old; + old = rw->lock & 0x7fffffffU; + if (likely(_raw_compare_and_swap(&rw->lock, old, old + 1) == old)) + return 1; + return _raw_read_trylock_retry(rw); +} + +static inline int _raw_write_trylock(rwlock_t *rw) { - unsigned long result, reg; - - __asm__ __volatile__( -#ifndef __s390x__ - " lhi %1,1\n" - " sll %1,31\n" - " cs %0,%1,0(%3)" -#else /* __s390x__ */ - " llihh %1,0x8000\n" - "0: csg %0,%1,0(%3)\n" -#endif /* __s390x__ */ - : "=d" (result), "=&d" (reg), "=m" (rw->lock) - : "a" (&rw->lock), "m" (rw->lock), "0" (0UL) - : "cc", "memory" ); - return result == 0; + if (likely(_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0)) + return 1; + return _raw_write_trylock_retry(rw); } #endif /* __ASM_SPINLOCK_H */ diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index bfbbe94b297..e82be96d490 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -145,6 +145,7 @@ enum KERN_BOOTLOADER_TYPE=67, /* int: boot loader type */ KERN_RANDOMIZE=68, /* int: randomize virtual address space */ KERN_SETUID_DUMPABLE=69, /* int: behaviour of dumps for setuid core */ + KERN_SPIN_RETRY=70, /* int: number of spinlock retries */ }; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index e60b9c36f1f..3e0bbee549e 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -114,6 +114,7 @@ extern int unaligned_enabled; extern int sysctl_ieee_emulation_warnings; #endif extern int sysctl_userprocess_debug; +extern int spin_retry; #endif extern int sysctl_hz_timer; @@ -647,7 +648,16 @@ static ctl_table kern_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, - +#if defined(CONFIG_ARCH_S390) + { + .ctl_name = KERN_SPIN_RETRY, + .procname = "spin_retry", + .data = &spin_retry, + .maxlen = sizeof (int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#endif { .ctl_name = 0 } }; -- cgit v1.2.3-70-g09d2 From 99b2d8df1d06f1072a949fc1e01a08b94b084d5f Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 27 Jul 2005 11:45:00 -0700 Subject: [PATCH] s390: external call performance The kernel uses the SIGP external call order code to signal other CPUs. When running with dedicated CPUs external calls don't get delivered immediately but within a fixed polling invervall. This can lead to delays where the system appears to do nothing. Replace the SIGP external call order with the SIGP emergency call order since this one gets delivered immediately. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/head.S | 2 +- arch/s390/kernel/head64.S | 2 +- arch/s390/kernel/s390_ext.c | 16 ++++++++++------ arch/s390/kernel/smp.c | 10 +++++----- 4 files changed, 17 insertions(+), 13 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S index fc8bf5e285f..928ca65e8bb 100644 --- a/arch/s390/kernel/head.S +++ b/arch/s390/kernel/head.S @@ -765,7 +765,7 @@ _stext: basr %r13,0 # get base # check control registers stctl %c0,%c15,0(%r15) - oi 2(%r15),0x20 # enable sigp external interrupts + oi 2(%r15),0x40 # enable sigp emergency signal oi 0(%r15),0x10 # switch on low address protection lctl %c0,%c15,0(%r15) diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index 28c50bdf7d4..10bc592c363 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -762,7 +762,7 @@ _stext: basr %r13,0 # get base # check control registers stctg %c0,%c15,0(%r15) - oi 6(%r15),0x20 # enable sigp external interrupts + oi 6(%r15),0x40 # enable sigp emergency signal oi 4(%r15),0x10 # switch on low address proctection lctlg %c0,%c15,0(%r15) diff --git a/arch/s390/kernel/s390_ext.c b/arch/s390/kernel/s390_ext.c index 3bdd38ec71d..207bc511a6e 100644 --- a/arch/s390/kernel/s390_ext.c +++ b/arch/s390/kernel/s390_ext.c @@ -19,7 +19,6 @@ #include /* - * Simple hash strategy: index = code & 0xff; * ext_int_hash[index] is the start of the list for all external interrupts * that hash to this index. With the current set of external interrupts * (0x1202 external call, 0x1004 cpu timer, 0x2401 hwc console, 0x4000 @@ -27,6 +26,11 @@ */ ext_int_info_t *ext_int_hash[256] = { 0, }; +static inline int ext_hash(__u16 code) +{ + return (code + (code >> 9)) & 0xff; +} + int register_external_interrupt(__u16 code, ext_int_handler_t handler) { ext_int_info_t *p; @@ -37,7 +41,7 @@ int register_external_interrupt(__u16 code, ext_int_handler_t handler) return -ENOMEM; p->code = code; p->handler = handler; - index = code & 0xff; + index = ext_hash(code); p->next = ext_int_hash[index]; ext_int_hash[index] = p; return 0; @@ -52,7 +56,7 @@ int register_early_external_interrupt(__u16 code, ext_int_handler_t handler, return -EINVAL; p->code = code; p->handler = handler; - index = code & 0xff; + index = ext_hash(code); p->next = ext_int_hash[index]; ext_int_hash[index] = p; return 0; @@ -63,7 +67,7 @@ int unregister_external_interrupt(__u16 code, ext_int_handler_t handler) ext_int_info_t *p, *q; int index; - index = code & 0xff; + index = ext_hash(code); q = NULL; p = ext_int_hash[index]; while (p != NULL) { @@ -90,7 +94,7 @@ int unregister_early_external_interrupt(__u16 code, ext_int_handler_t handler, if (p == NULL || p->code != code || p->handler != handler) return -EINVAL; - index = code & 0xff; + index = ext_hash(code); q = ext_int_hash[index]; if (p != q) { while (q != NULL) { @@ -120,7 +124,7 @@ void do_extint(struct pt_regs *regs, unsigned short code) */ account_ticks(regs); kstat_cpu(smp_processor_id()).irqs[EXTERNAL_INTERRUPT]++; - index = code & 0xff; + index = ext_hash(code); for (p = ext_int_hash[index]; p; p = p->next) { if (likely(p->code == code)) { if (likely(p->handler)) diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 642572a8e33..da77f001af8 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -375,7 +375,7 @@ static void smp_ext_bitcall(int cpu, ec_bit_sig sig) * Set signaling bit in lowcore of target cpu and kick it */ set_bit(sig, (unsigned long *) &lowcore_ptr[cpu]->ext_call_fast); - while(signal_processor(cpu, sigp_external_call) == sigp_busy) + while(signal_processor(cpu, sigp_emergency_signal) == sigp_busy) udelay(10); } @@ -394,7 +394,7 @@ static void smp_ext_bitcall_others(ec_bit_sig sig) * Set signaling bit in lowcore of target cpu and kick it */ set_bit(sig, (unsigned long *) &lowcore_ptr[cpu]->ext_call_fast); - while (signal_processor(cpu, sigp_external_call) == sigp_busy) + while (signal_processor(cpu, sigp_emergency_signal) == sigp_busy) udelay(10); } } @@ -751,9 +751,9 @@ void __init smp_prepare_cpus(unsigned int max_cpus) unsigned int cpu; int i; - /* request the 0x1202 external interrupt */ - if (register_external_interrupt(0x1202, do_ext_call_interrupt) != 0) - panic("Couldn't request external interrupt 0x1202"); + /* request the 0x1201 emergency signal external interrupt */ + if (register_external_interrupt(0x1201, do_ext_call_interrupt) != 0) + panic("Couldn't request external interrupt 0x1201"); smp_check_cpus(max_cpus); memset(lowcore_ptr,0,sizeof(lowcore_ptr)); /* -- cgit v1.2.3-70-g09d2 From f878e4367580f37327e68ce2c6861245077f0904 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 27 Jul 2005 11:45:05 -0700 Subject: [PATCH] s390: 31 bit memory size limit Limit reported memory size to 2GB if running in 31 bit mode. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/head.S | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S index 928ca65e8bb..d12cff11b4b 100644 --- a/arch/s390/kernel/head.S +++ b/arch/s390/kernel/head.S @@ -535,8 +535,13 @@ startup:basr %r13,0 # get base lhi %r1,0 icm %r1,3,.Lscpincr1-PARMAREA(%r4) # use this one if != 0 jnz .Lscnd - l %r1,.Lscpincr2-PARMAREA+4(%r4) # otherwise use this one + lhi %r1,0x800 # otherwise report 2GB .Lscnd: + lhi %r3,0x800 # limit reported memory size to 2GB + cr %r1,%r3 + jl .Lno2gb + lr %r1,%r3 +.Lno2gb: xr %r3,%r3 # same logic ic %r3,.Lscpa1-PARMAREA(%r4) chi %r3,0x00 -- cgit v1.2.3-70-g09d2 From 8ffa7405afe0eaf34db6254160b734f084601f68 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 27 Jul 2005 11:45:06 -0700 Subject: [PATCH] s390: cpu timer reset in machine check handler Fix wrong move direction of timer values for cpu accounting in case of a machine check that indicates a broken cpu timer. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/entry.S | 6 +++--- arch/s390/kernel/entry64.S | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 5b262b5d869..1a271b16cb5 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -690,9 +690,9 @@ mcck_int_handler: bo BASED(0f) spt __LC_LAST_UPDATE_TIMER # revalidate cpu timer #ifdef CONFIG_VIRT_CPU_ACCOUNTING - mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER - mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER - mvc __LC_LAST_UPDATE_TIMER(8),__LC_EXIT_TIMER + mvc __LC_ASYNC_ENTER_TIMER(8),__LC_LAST_UPDATE_TIMER + mvc __LC_SYNC_ENTER_TIMER(8),__LC_LAST_UPDATE_TIMER + mvc __LC_EXIT_TIMER(8),__LC_LAST_UPDATE_TIMER 0: tm __LC_MCCK_CODE+2,0x08 # mwp of old psw valid? bno BASED(mcck_no_vtime) # no -> skip cleanup critical tm __LC_MCK_OLD_PSW+1,0x01 # interrupting from user ? diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index 57ca75d0ad7..d9f22915008 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -727,9 +727,9 @@ mcck_int_handler: jo 0f spt __LC_LAST_UPDATE_TIMER #ifdef CONFIG_VIRT_CPU_ACCOUNTING - mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER - mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER - mvc __LC_LAST_UPDATE_TIMER(8),__LC_EXIT_TIMER + mvc __LC_ASYNC_ENTER_TIMER(8),__LC_LAST_UPDATE_TIMER + mvc __LC_SYNC_ENTER_TIMER(8),__LC_LAST_UPDATE_TIMER + mvc __LC_EXIT_TIMER(8),__LC_LAST_UPDATE_TIMER 0: tm __LC_MCCK_CODE+2,0x08 # mwp of old psw valid? jno mcck_no_vtime # no -> no timer update tm __LC_MCK_OLD_PSW+1,0x01 # interrupting from user ? -- cgit v1.2.3-70-g09d2 From 5712f52e8c024c6b7b82ae5c7b5b3d0f3e6f711b Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 29 Jul 2005 14:03:35 -0700 Subject: [PATCH] s390: default configuration Update default configuration of s390. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/defconfig | 253 +++++++++++++++++++++++++++++----------------------- 1 file changed, 142 insertions(+), 111 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/defconfig b/arch/s390/defconfig index 89850b2c27e..0865251a3f4 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.12-rc3 -# Fri Apr 22 15:30:58 2005 +# Linux kernel version: 2.6.13-rc4 +# Fri Jul 29 14:49:30 2005 # CONFIG_MMU=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y @@ -23,10 +23,11 @@ CONFIG_INIT_ENV_ARG_LIMIT=32 CONFIG_LOCALVERSION="" CONFIG_SWAP=y CONFIG_SYSVIPC=y -# CONFIG_POSIX_MQUEUE is not set +CONFIG_POSIX_MQUEUE=y # CONFIG_BSD_PROCESS_ACCT is not set CONFIG_SYSCTL=y -# CONFIG_AUDIT is not set +CONFIG_AUDIT=y +# CONFIG_AUDITSYSCALL is not set CONFIG_HOTPLUG=y CONFIG_KOBJECT_UEVENT=y CONFIG_IKCONFIG=y @@ -36,6 +37,8 @@ CONFIG_IKCONFIG_PROC=y CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_ALL is not set # CONFIG_KALLSYMS_EXTRA_PASS is not set +CONFIG_PRINTK=y +CONFIG_BUG=y CONFIG_BASE_FULL=y CONFIG_FUTEX=y CONFIG_EPOLL=y @@ -51,9 +54,10 @@ CONFIG_BASE_SMALL=0 # Loadable module support # CONFIG_MODULES=y -# CONFIG_MODULE_UNLOAD is not set +CONFIG_MODULE_UNLOAD=y +# CONFIG_MODULE_FORCE_UNLOAD is not set CONFIG_OBSOLETE_MODPARM=y -# CONFIG_MODVERSIONS is not set +CONFIG_MODVERSIONS=y # CONFIG_MODULE_SRCVERSION_ALL is not set CONFIG_KMOD=y CONFIG_STOP_MACHINE=y @@ -81,8 +85,15 @@ CONFIG_MARCH_G5=y # CONFIG_MARCH_Z990 is not set CONFIG_PACK_STACK=y # CONFIG_SMALL_STACK is not set -# CONFIG_CHECK_STACK is not set +CONFIG_CHECK_STACK=y +CONFIG_STACK_GUARD=256 # CONFIG_WARN_STACK is not set +CONFIG_SELECT_MEMORY_MODEL=y +CONFIG_FLATMEM_MANUAL=y +# CONFIG_DISCONTIGMEM_MANUAL is not set +# CONFIG_SPARSEMEM_MANUAL is not set +CONFIG_FLATMEM=y +CONFIG_FLAT_NODE_MEM_MAP=y # # I/O subsystem configuration @@ -95,7 +106,7 @@ CONFIG_QDIO=y # # Misc # -# CONFIG_PREEMPT is not set +CONFIG_PREEMPT=y CONFIG_IPL=y # CONFIG_IPL_TAPE is not set CONFIG_IPL_VM=y @@ -105,9 +116,110 @@ CONFIG_BINFMT_MISC=m CONFIG_PFAULT=y # CONFIG_SHARED_KERNEL is not set # CONFIG_CMM is not set -# CONFIG_VIRT_TIMER is not set +CONFIG_VIRT_TIMER=y +CONFIG_VIRT_CPU_ACCOUNTING=y +# CONFIG_APPLDATA_BASE is not set CONFIG_NO_IDLE_HZ=y CONFIG_NO_IDLE_HZ_INIT=y +# CONFIG_KEXEC is not set + +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +# CONFIG_PACKET_MMAP is not set +CONFIG_UNIX=y +CONFIG_XFRM=y +# CONFIG_XFRM_USER is not set +CONFIG_NET_KEY=y +CONFIG_INET=y +CONFIG_IP_MULTICAST=y +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +# CONFIG_IP_PNP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_IP_MROUTE is not set +# CONFIG_ARPD is not set +# CONFIG_SYN_COOKIES is not set +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_TUNNEL is not set +CONFIG_IP_TCPDIAG=y +CONFIG_IP_TCPDIAG_IPV6=y +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_BIC=y +CONFIG_IPV6=y +# CONFIG_IPV6_PRIVACY is not set +# CONFIG_INET6_AH is not set +# CONFIG_INET6_ESP is not set +# CONFIG_INET6_IPCOMP is not set +# CONFIG_INET6_TUNNEL is not set +# CONFIG_IPV6_TUNNEL is not set +# CONFIG_NETFILTER is not set + +# +# SCTP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_SCTP is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_NET_DIVERT is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +CONFIG_NET_SCHED=y +CONFIG_NET_SCH_CLK_JIFFIES=y +# CONFIG_NET_SCH_CLK_GETTIMEOFDAY is not set +# CONFIG_NET_SCH_CLK_CPU is not set +CONFIG_NET_SCH_CBQ=m +# CONFIG_NET_SCH_HTB is not set +# CONFIG_NET_SCH_HFSC is not set +CONFIG_NET_SCH_PRIO=m +CONFIG_NET_SCH_RED=m +CONFIG_NET_SCH_SFQ=m +CONFIG_NET_SCH_TEQL=m +CONFIG_NET_SCH_TBF=m +CONFIG_NET_SCH_GRED=m +CONFIG_NET_SCH_DSMARK=m +# CONFIG_NET_SCH_NETEM is not set +# CONFIG_NET_SCH_INGRESS is not set +CONFIG_NET_QOS=y +CONFIG_NET_ESTIMATOR=y +CONFIG_NET_CLS=y +# CONFIG_NET_CLS_BASIC is not set +CONFIG_NET_CLS_TCINDEX=m +CONFIG_NET_CLS_ROUTE4=m +CONFIG_NET_CLS_ROUTE=y +CONFIG_NET_CLS_FW=m +CONFIG_NET_CLS_U32=m +# CONFIG_CLS_U32_PERF is not set +# CONFIG_NET_CLS_IND is not set +CONFIG_NET_CLS_RSVP=m +CONFIG_NET_CLS_RSVP6=m +# CONFIG_NET_EMATCH is not set +# CONFIG_NET_CLS_ACT is not set +CONFIG_NET_CLS_POLICE=y + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set # CONFIG_PCMCIA is not set # @@ -133,6 +245,7 @@ CONFIG_CHR_DEV_ST=y CONFIG_BLK_DEV_SR=y CONFIG_BLK_DEV_SR_VENDOR=y CONFIG_CHR_DEV_SG=y +# CONFIG_CHR_DEV_SCH is not set # # Some SCSI devices (e.g. CD jukebox) support multiple LUNs @@ -205,7 +318,13 @@ CONFIG_MD_RAID5=m # CONFIG_MD_RAID6 is not set CONFIG_MD_MULTIPATH=m # CONFIG_MD_FAULTY is not set -# CONFIG_BLK_DEV_DM is not set +CONFIG_BLK_DEV_DM=y +CONFIG_DM_CRYPT=y +CONFIG_DM_SNAPSHOT=y +CONFIG_DM_MIRROR=y +CONFIG_DM_ZERO=y +CONFIG_DM_MULTIPATH=y +# CONFIG_DM_MULTIPATH_EMC is not set # # Character device drivers @@ -231,7 +350,8 @@ CONFIG_CCW_CONSOLE=y CONFIG_SCLP=y CONFIG_SCLP_TTY=y CONFIG_SCLP_CONSOLE=y -# CONFIG_SCLP_VT220_TTY is not set +CONFIG_SCLP_VT220_TTY=y +CONFIG_SCLP_VT220_CONSOLE=y CONFIG_SCLP_CPI=m CONFIG_S390_TAPE=m @@ -255,105 +375,8 @@ CONFIG_S390_TAPE_34XX=m CONFIG_Z90CRYPT=m # -# Networking support -# -CONFIG_NET=y - -# -# Networking options -# -CONFIG_PACKET=y -# CONFIG_PACKET_MMAP is not set -CONFIG_UNIX=y -CONFIG_NET_KEY=y -CONFIG_INET=y -CONFIG_IP_MULTICAST=y -# CONFIG_IP_ADVANCED_ROUTER is not set -# CONFIG_IP_PNP is not set -# CONFIG_NET_IPIP is not set -# CONFIG_NET_IPGRE is not set -# CONFIG_IP_MROUTE is not set -# CONFIG_ARPD is not set -# CONFIG_SYN_COOKIES is not set -# CONFIG_INET_AH is not set -# CONFIG_INET_ESP is not set -# CONFIG_INET_IPCOMP is not set -# CONFIG_INET_TUNNEL is not set -CONFIG_IP_TCPDIAG=y -CONFIG_IP_TCPDIAG_IPV6=y -CONFIG_IPV6=y -# CONFIG_IPV6_PRIVACY is not set -# CONFIG_INET6_AH is not set -# CONFIG_INET6_ESP is not set -# CONFIG_INET6_IPCOMP is not set -# CONFIG_INET6_TUNNEL is not set -# CONFIG_IPV6_TUNNEL is not set -# CONFIG_NETFILTER is not set -CONFIG_XFRM=y -# CONFIG_XFRM_USER is not set - -# -# SCTP Configuration (EXPERIMENTAL) -# -# CONFIG_IP_SCTP is not set -# CONFIG_ATM is not set -# CONFIG_BRIDGE is not set -# CONFIG_VLAN_8021Q is not set -# CONFIG_DECNET is not set -# CONFIG_LLC2 is not set -# CONFIG_IPX is not set -# CONFIG_ATALK is not set -# CONFIG_X25 is not set -# CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set -# CONFIG_ECONET is not set -# CONFIG_WAN_ROUTER is not set - -# -# QoS and/or fair queueing -# -CONFIG_NET_SCHED=y -CONFIG_NET_SCH_CLK_JIFFIES=y -# CONFIG_NET_SCH_CLK_GETTIMEOFDAY is not set -# CONFIG_NET_SCH_CLK_CPU is not set -CONFIG_NET_SCH_CBQ=m -# CONFIG_NET_SCH_HTB is not set -# CONFIG_NET_SCH_HFSC is not set -CONFIG_NET_SCH_PRIO=m -CONFIG_NET_SCH_RED=m -CONFIG_NET_SCH_SFQ=m -CONFIG_NET_SCH_TEQL=m -CONFIG_NET_SCH_TBF=m -CONFIG_NET_SCH_GRED=m -CONFIG_NET_SCH_DSMARK=m -# CONFIG_NET_SCH_NETEM is not set -# CONFIG_NET_SCH_INGRESS is not set -CONFIG_NET_QOS=y -CONFIG_NET_ESTIMATOR=y -CONFIG_NET_CLS=y -# CONFIG_NET_CLS_BASIC is not set -CONFIG_NET_CLS_TCINDEX=m -CONFIG_NET_CLS_ROUTE4=m -CONFIG_NET_CLS_ROUTE=y -CONFIG_NET_CLS_FW=m -CONFIG_NET_CLS_U32=m -# CONFIG_CLS_U32_PERF is not set -# CONFIG_NET_CLS_IND is not set -CONFIG_NET_CLS_RSVP=m -CONFIG_NET_CLS_RSVP6=m -# CONFIG_NET_EMATCH is not set -# CONFIG_NET_CLS_ACT is not set -CONFIG_NET_CLS_POLICE=y - -# -# Network testing +# Network device support # -# CONFIG_NET_PKTGEN is not set -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set -# CONFIG_HAMRADIO is not set -# CONFIG_IRDA is not set -# CONFIG_BT is not set CONFIG_NETDEVICES=y CONFIG_DUMMY=m CONFIG_BONDING=m @@ -411,12 +434,15 @@ CONFIG_CCWGROUP=y # CONFIG_SLIP is not set # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set # # File systems # CONFIG_EXT2_FS=y # CONFIG_EXT2_FS_XATTR is not set +# CONFIG_EXT2_FS_XIP is not set CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y # CONFIG_EXT3_FS_POSIX_ACL is not set @@ -426,6 +452,7 @@ CONFIG_JBD=y CONFIG_FS_MBCACHE=y # CONFIG_REISERFS_FS is not set # CONFIG_JFS_FS is not set +# CONFIG_FS_POSIX_ACL is not set # # XFS support @@ -433,6 +460,7 @@ CONFIG_FS_MBCACHE=y # CONFIG_XFS_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set +CONFIG_INOTIFY=y # CONFIG_QUOTA is not set CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set @@ -457,7 +485,6 @@ CONFIG_DNOTIFY=y CONFIG_PROC_FS=y CONFIG_PROC_KCORE=y CONFIG_SYSFS=y -# CONFIG_DEVFS_FS is not set # CONFIG_DEVPTS_FS_XATTR is not set CONFIG_TMPFS=y # CONFIG_TMPFS_XATTR is not set @@ -486,15 +513,18 @@ CONFIG_RAMFS=y # CONFIG_NFS_FS=y CONFIG_NFS_V3=y +# CONFIG_NFS_V3_ACL is not set # CONFIG_NFS_V4 is not set # CONFIG_NFS_DIRECTIO is not set CONFIG_NFSD=y CONFIG_NFSD_V3=y +# CONFIG_NFSD_V3_ACL is not set # CONFIG_NFSD_V4 is not set CONFIG_NFSD_TCP=y CONFIG_LOCKD=y CONFIG_LOCKD_V4=y CONFIG_EXPORTFS=y +CONFIG_NFS_COMMON=y CONFIG_SUNRPC=y # CONFIG_RPCSEC_GSS_KRB5 is not set # CONFIG_RPCSEC_GSS_SPKM3 is not set @@ -544,11 +574,12 @@ CONFIG_MAGIC_SYSRQ=y CONFIG_LOG_BUF_SHIFT=17 # CONFIG_SCHEDSTATS is not set # CONFIG_DEBUG_SLAB is not set +CONFIG_DEBUG_PREEMPT=y # CONFIG_DEBUG_SPINLOCK is not set # CONFIG_DEBUG_SPINLOCK_SLEEP is not set # CONFIG_DEBUG_KOBJECT is not set # CONFIG_DEBUG_INFO is not set -# CONFIG_DEBUG_FS is not set +CONFIG_DEBUG_FS=y # # Security options -- cgit v1.2.3-70-g09d2 From 350e3ade9ed0809a94c51db6ee66883a35f0d6c9 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 29 Jul 2005 14:03:36 -0700 Subject: [PATCH] s390: check for interrupt before waiting The patch that introduced waiting for interrupts after resetting the reader can cause the boot to fail because the system is waiting for an interrupt that will never arrive. Add code to check if an interrupt is supposed to arrive before waiting endlessly. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/head.S | 7 +++++++ arch/s390/kernel/head64.S | 7 +++++++ 2 files changed, 14 insertions(+) (limited to 'arch/s390') diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S index d12cff11b4b..2710e66fefb 100644 --- a/arch/s390/kernel/head.S +++ b/arch/s390/kernel/head.S @@ -346,6 +346,13 @@ iplstart: la %r2,.Lreset lhi %r3,26 diag %r2,%r3,8 + la %r5,.Lirb + stsch 0(%r5) # check if irq is pending + tm 30(%r5),0x0f # by verifying if any of the + bnz .Lwaitforirq # activity or status control + tm 31(%r5),0xff # bits is set in the schib + bz .Lnoreset +.Lwaitforirq: mvc 0x78(8),.Lrdrnewpsw # set up IO interrupt psw .Lwaitrdrirq: lpsw .Lrdrwaitpsw diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index 10bc592c363..9a8263a153c 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -345,6 +345,13 @@ iplstart: la %r2,.Lreset lhi %r3,26 diag %r2,%r3,8 + la %r5,.Lirb + stsch 0(%r5) # check if irq is pending + tm 30(%r5),0x0f # by verifying if any of the + bnz .Lwaitforirq # activity or status control + tm 31(%r5),0xff # bits is set in the schib + bz .Lnoreset +.Lwaitforirq: mvc 0x78(8),.Lrdrnewpsw # set up IO interrupt psw .Lwaitrdrirq: lpsw .Lrdrwaitpsw -- cgit v1.2.3-70-g09d2 From b87a1e506115e7df4c6bfd266267ee0088cba3b7 Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Fri, 29 Jul 2005 14:03:38 -0700 Subject: [PATCH] s390: fix inline assembly in appldata Fix inline assembly that gets miscompiled by gcc 4. Signed-off-by: Gerald Schaefer Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/appldata/appldata_base.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index c067435bae4..c9f2f60cfa5 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -232,7 +232,11 @@ static int appldata_diag(char record_nr, u16 function, unsigned long buffer, ry = -1; asm volatile( "diag %1,%0,0xDC\n\t" - : "=d" (ry) : "d" (&(appldata_parameter_list)) : "cc"); + : "=d" (ry) + : "d" (&appldata_parameter_list), + "m" (appldata_parameter_list), + "m" (appldata_product_id) + : "cc"); return (int) ry; } /************************ timer, work, DIAG ****************************/ -- cgit v1.2.3-70-g09d2 From 5d3f229fcd4409d3e182b204defc122eb7833535 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 1 Aug 2005 21:11:33 -0700 Subject: [PATCH] s390: kexec fixes and improvements. Disable pseudo page fault handling before starting the new kernel and try to use diag308 to reset the machine. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/machine_kexec.c | 7 ++++++ arch/s390/kernel/relocate_kernel.S | 41 +++++++++++++++++++++++++++++++- arch/s390/kernel/relocate_kernel64.S | 45 ++++++++++++++++++++++++++++++++++-- arch/s390/kernel/smp.c | 6 +++-- arch/s390/kernel/traps.c | 15 ++++++++++++ 5 files changed, 109 insertions(+), 5 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c index 2721c3a32b8..5aa71b05b8a 100644 --- a/arch/s390/kernel/machine_kexec.c +++ b/arch/s390/kernel/machine_kexec.c @@ -70,6 +70,8 @@ machine_kexec(struct kimage *image) for (;;); } +extern void pfault_fini(void); + static void kexec_halt_all_cpus(void *kernel_image) { @@ -78,6 +80,11 @@ kexec_halt_all_cpus(void *kernel_image) struct kimage *image; relocate_kernel_t data_mover; +#ifdef CONFIG_PFAULT + if (MACHINE_IS_VM) + pfault_fini(); +#endif + if (atomic_compare_and_swap(-1, smp_processor_id(), &cpuid)) signal_processor(smp_processor_id(), sigp_stop); diff --git a/arch/s390/kernel/relocate_kernel.S b/arch/s390/kernel/relocate_kernel.S index d5e4a62fbb7..2a25ec7147f 100644 --- a/arch/s390/kernel/relocate_kernel.S +++ b/arch/s390/kernel/relocate_kernel.S @@ -4,6 +4,7 @@ * (C) Copyright IBM Corp. 2005 * * Author(s): Rolf Adelsberger + * Heiko Carstens * */ @@ -25,8 +26,31 @@ relocate_kernel: basr %r13,0 #base address .base: - spx zero64-.base(%r13) #absolute addressing mode stnsm sys_msk-.base(%r13),0xf8 #disable DAT and IRQ (external) + spx zero64-.base(%r13) #absolute addressing mode + stctl %c0,%c15,ctlregs-.base(%r13) + stm %r0,%r15,gprregs-.base(%r13) + la %r1,load_psw-.base(%r13) + mvc 0(8,%r0),0(%r1) + la %r0,.back-.base(%r13) + st %r0,4(%r0) + oi 4(%r0),0x80 + mvc 0x68(8,%r0),0(%r1) + la %r0,.back_pgm-.base(%r13) + st %r0,0x6c(%r0) + oi 0x6c(%r0),0x80 + lhi %r0,0 + diag %r0,%r0,0x308 + .back: + basr %r13,0 + .back_base: + oi have_diag308-.back_base(%r13),0x01 + lctl %c0,%c15,ctlregs-.back_base(%r13) + lm %r0,%r15,gprregs-.back_base(%r13) + j .start_reloc + .back_pgm: + lm %r0,%r15,gprregs-.base(%r13) + .start_reloc: lhi %r10,-1 #preparing the mask sll %r10,12 #shift it such that it becomes 0xf000 .top: @@ -63,6 +87,10 @@ o %r3,4(%r4) #or load address into psw st %r3,4(%r4) mvc 0(8,%r0),0(%r4) #copy psw to absolute address 0 + tm have_diag308-.base(%r13),0x01 + jno .no_diag308 + diag %r0,%r0,0x308 + .no_diag308: sr %r1,%r1 #clear %r1 sr %r2,%r2 #clear %r2 sigp %r1,%r2,0x12 #set cpuid to zero @@ -75,6 +103,17 @@ .long 0x00080000,0x80000000 sys_msk: .quad 0 + ctlregs: + .rept 16 + .long 0 + .endr + gprregs: + .rept 16 + .long 0 + .endr + have_diag308: + .byte 0 + .align 8 relocate_kernel_end: .globl relocate_kernel_len relocate_kernel_len: diff --git a/arch/s390/kernel/relocate_kernel64.S b/arch/s390/kernel/relocate_kernel64.S index 96290cc4eb3..8cdb86e8911 100644 --- a/arch/s390/kernel/relocate_kernel64.S +++ b/arch/s390/kernel/relocate_kernel64.S @@ -4,6 +4,7 @@ * (C) Copyright IBM Corp. 2005 * * Author(s): Rolf Adelsberger + * Heiko Carstens * */ @@ -26,8 +27,34 @@ relocate_kernel: basr %r13,0 #base address .base: + stnsm sys_msk-.base(%r13),0xf8 #disable DAT and IRQs spx zero64-.base(%r13) #absolute addressing mode - stnsm sys_msk-.base(%r13),0xf8 #disable DAT and IRQ (external) + stctg %c0,%c15,ctlregs-.base(%r13) + stmg %r0,%r15,gprregs-.base(%r13) + lghi %r0,3 + sllg %r0,%r0,31 + stg %r0,0x1d0(%r0) + la %r0,.back_pgm-.base(%r13) + stg %r0,0x1d8(%r0) + la %r1,load_psw-.base(%r13) + mvc 0(8,%r0),0(%r1) + la %r0,.back-.base(%r13) + st %r0,4(%r0) + oi 4(%r0),0x80 + lghi %r0,0 + diag %r0,%r0,0x308 + .back: + lhi %r1,1 #mode 1 = esame + sigp %r1,%r0,0x12 #switch to esame mode + sam64 #switch to 64 bit addressing mode + basr %r13,0 + .back_base: + oi have_diag308-.back_base(%r13),0x01 + lctlg %c0,%c15,ctlregs-.back_base(%r13) + lmg %r0,%r15,gprregs-.back_base(%r13) + j .top + .back_pgm: + lmg %r0,%r15,gprregs-.base(%r13) .top: lghi %r7,4096 #load PAGE_SIZE in r7 lghi %r9,4096 #load PAGE_SIZE in r9 @@ -62,6 +89,10 @@ o %r3,4(%r4) #or load address into psw st %r3,4(%r4) mvc 0(8,%r0),0(%r4) #copy psw to absolute address 0 + tm have_diag308-.base(%r13),0x01 + jno .no_diag308 + diag %r0,%r0,0x308 + .no_diag308: sam31 #31 bit mode sr %r1,%r1 #erase register r1 sr %r2,%r2 #erase register r2 @@ -75,8 +106,18 @@ .long 0x00080000,0x80000000 sys_msk: .quad 0 + ctlregs: + .rept 16 + .quad 0 + .endr + gprregs: + .rept 16 + .quad 0 + .endr + have_diag308: + .byte 0 + .align 8 relocate_kernel_end: .globl relocate_kernel_len relocate_kernel_len: .quad relocate_kernel_end - relocate_kernel - diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index da77f001af8..85222fee436 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -537,7 +537,8 @@ int __devinit start_secondary(void *cpuvoid) #endif #ifdef CONFIG_PFAULT /* Enable pfault pseudo page faults on this cpu. */ - pfault_init(); + if (MACHINE_IS_VM) + pfault_init(); #endif /* Mark this cpu as online */ cpu_set(smp_processor_id(), cpu_online_map); @@ -690,7 +691,8 @@ __cpu_disable(void) #ifdef CONFIG_PFAULT /* Disable pfault pseudo page faults on this cpu. */ - pfault_fini(); + if (MACHINE_IS_VM) + pfault_fini(); #endif /* disable all external interrupts */ diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c index bc7b7be7acb..6b8703ec2ae 100644 --- a/arch/s390/kernel/traps.c +++ b/arch/s390/kernel/traps.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include @@ -675,6 +676,19 @@ asmlinkage void kernel_stack_overflow(struct pt_regs * regs) panic("Corrupt kernel stack, can't continue."); } +#ifndef CONFIG_ARCH_S390X +static int +pagex_reboot_event(struct notifier_block *this, unsigned long event, void *ptr) +{ + if (MACHINE_IS_VM) + cpcmd("SET PAGEX OFF", NULL, 0, NULL); + return NOTIFY_DONE; +} + +static struct notifier_block pagex_reboot_notifier = { + .notifier_call = &pagex_reboot_event, +}; +#endif /* init is done in lowcore.S and head.S */ @@ -735,6 +749,7 @@ void __init trap_init(void) &ext_int_pfault); #endif #ifndef CONFIG_ARCH_S390X + register_reboot_notifier(&pagex_reboot_notifier); cpcmd("SET PAGEX ON", NULL, 0, NULL); #endif } -- cgit v1.2.3-70-g09d2 From d2013485a52fb7ece48e5688b443cc098f4dbbdf Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Mon, 1 Aug 2005 21:11:34 -0700 Subject: [PATCH] s390: ioprio & inotify system calls. Add system calls for io priorities and inotify. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/compat_wrapper.S | 26 ++++++++++++++++++++++++++ arch/s390/kernel/syscalls.S | 5 +++++ include/asm-s390/unistd.h | 7 ++++++- 3 files changed, 37 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index 799a98eac92..23fe94e5868 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -1449,3 +1449,29 @@ compat_sys_kexec_load_wrapper: llgtr %r4,%r4 # struct kexec_segment * llgfr %r5,%r5 # unsigned long jg compat_sys_kexec_load + + .globl sys_ioprio_set_wrapper +sys_ioprio_set_wrapper: + lgfr %r2,%r2 # int + lgfr %r3,%r3 # int + lgfr %r4,%r4 # int + jg sys_ioprio_set + + .globl sys_ioprio_get_wrapper +sys_ioprio_get_wrapper: + lgfr %r2,%r2 # int + lgfr %r3,%r3 # int + jg sys_ioprio_get + + .globl sys_inotify_add_watch_wrapper +sys_inotify_add_watch_wrapper: + lgfr %r2,%r2 # int + llgtr %r3,%r3 # const char * + llgfr %r4,%r4 # u32 + jg sys_inotify_add_watch + + .globl sys_inotify_rm_watch_wrapper +sys_inotify_rm_watch_wrapper: + lgfr %r2,%r2 # int + llgfr %r3,%r3 # u32 + jg sys_inotify_rm_watch diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index a8668afb5f8..426d7cafdab 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -290,3 +290,8 @@ SYSCALL(sys_add_key,sys_add_key,compat_sys_add_key_wrapper) SYSCALL(sys_request_key,sys_request_key,compat_sys_request_key_wrapper) SYSCALL(sys_keyctl,sys_keyctl,compat_sys_keyctl) /* 280 */ SYSCALL(sys_waitid,sys_waitid,compat_sys_waitid_wrapper) +SYSCALL(sys_ioprio_set,sys_ioprio_set,sys_ioprio_set_wrapper) +SYSCALL(sys_ioprio_get,sys_ioprio_get,sys_ioprio_get_wrapper) +SYSCALL(sys_inotify_init,sys_inotify_init,sys_inotify_init) +SYSCALL(sys_inotify_add_watch,sys_inotify_add_watch,sys_inotify_add_watch_wrapper) +SYSCALL(sys_inotify_rm_watch,sys_inotify_rm_watch,sys_inotify_rm_watch_wrapper) diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h index 363db45f8d0..221e965da92 100644 --- a/include/asm-s390/unistd.h +++ b/include/asm-s390/unistd.h @@ -274,8 +274,13 @@ #define __NR_request_key 279 #define __NR_keyctl 280 #define __NR_waitid 281 +#define __NR_ioprio_set 282 +#define __NR_ioprio_get 283 +#define __NR_inotify_init 284 +#define __NR_inotify_add_watch 285 +#define __NR_inotify_rm_watch 286 -#define NR_syscalls 282 +#define NR_syscalls 287 /* * There are some system calls that are not present on 64 bit, some -- cgit v1.2.3-70-g09d2