diff options
Diffstat (limited to 'arch/ia64')
95 files changed, 1377 insertions, 682 deletions
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 18318749884..0b7f701d5cf 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -66,7 +66,7 @@ config IA64_UNCACHED_ALLOCATOR bool select GENERIC_ALLOCATOR -config DMA_IS_DMA32 +config AUDIT_ARCH bool default y @@ -253,7 +253,7 @@ config NR_CPUS int "Maximum number of CPUs (2-1024)" range 2 1024 depends on SMP - default "64" + default "1024" help You should set this to the number of CPUs in your system, but keep in mind that a kernel compiled for, e.g., 2 CPUs will boot but @@ -271,6 +271,9 @@ config HOTPLUG_CPU can be controlled through /sys/devices/system/cpu/cpu#. Say N if you want to disable CPU hotplug. +config ARCH_ENABLE_MEMORY_HOTPLUG + def_bool y + config SCHED_SMT bool "SMT scheduler support" depends on SMP @@ -346,13 +349,16 @@ config NUMA config NODES_SHIFT int "Max num nodes shift(3-10)" range 3 10 - default "8" + default "10" depends on NEED_MULTIPLE_NODES help This option specifies the maximum number of nodes in your SSI system. MAX_NUMNODES will be 2^(This value). If in doubt, use the default. +config ARCH_POPULATES_NODE_MAP + def_bool y + # VIRTUAL_MEM_MAP and FLAT_NODE_MEM_MAP are functionally equivalent. # VIRTUAL_MEM_MAP has been retained for historical reasons. config VIRTUAL_MEM_MAP @@ -374,6 +380,10 @@ config HAVE_ARCH_EARLY_PFN_TO_NID def_bool y depends on NEED_MULTIPLE_NODES +config HAVE_ARCH_NODEDATA_EXTENSION + def_bool y + depends on NUMA + config IA32_SUPPORT bool "Support for Linux/x86 binaries" help @@ -413,6 +423,14 @@ config IA64_PALINFO config SGI_SN def_bool y if (IA64_SGI_SN2 || IA64_GENERIC) +config IA64_ESI + bool "ESI (Extensible SAL Interface) support" + help + If you say Y here, support is built into the kernel to + make ESI calls. ESI calls are used to support vendor-specific + firmware extensions, such as the ability to inject memory-errors + for test-purposes. If you're unsure, say N. + source "drivers/sn/Kconfig" source "drivers/firmware/Kconfig" @@ -485,6 +503,10 @@ config GENERIC_PENDING_IRQ depends on GENERIC_HARDIRQS && SMP default y +config IRQ_PER_CPU + bool + default y + source "arch/ia64/hp/sim/Kconfig" menu "Instrumentation Support" diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig index 9ea35398e10..0f14a82b856 100644 --- a/arch/ia64/configs/sn2_defconfig +++ b/arch/ia64/configs/sn2_defconfig @@ -363,7 +363,7 @@ CONFIG_BLK_DEV_IDECD=y # CONFIG_IDE_GENERIC=y CONFIG_BLK_DEV_IDEPCI=y -# CONFIG_IDEPCI_SHARE_IRQ is not set +CONFIG_IDEPCI_SHARE_IRQ=y # CONFIG_BLK_DEV_OFFBOARD is not set # CONFIG_BLK_DEV_GENERIC is not set # CONFIG_BLK_DEV_OPTI621 is not set diff --git a/arch/ia64/configs/tiger_defconfig b/arch/ia64/configs/tiger_defconfig index 766bf495543..9d1cffb57cd 100644 --- a/arch/ia64/configs/tiger_defconfig +++ b/arch/ia64/configs/tiger_defconfig @@ -114,7 +114,7 @@ CONFIG_IA64_CYCLONE=y CONFIG_IOSAPIC=y CONFIG_FORCE_MAX_ZONEORDER=17 CONFIG_SMP=y -CONFIG_NR_CPUS=4 +CONFIG_NR_CPUS=16 CONFIG_HOTPLUG_CPU=y CONFIG_PERMIT_BSP_REMOVE=y CONFIG_FORCE_CPEI_RETARGET=y diff --git a/arch/ia64/defconfig b/arch/ia64/defconfig index 6cba55da572..9001b3fbaa3 100644 --- a/arch/ia64/defconfig +++ b/arch/ia64/defconfig @@ -366,7 +366,7 @@ CONFIG_BLK_DEV_IDESCSI=m # CONFIG_IDE_GENERIC is not set # CONFIG_BLK_DEV_IDEPNP is not set CONFIG_BLK_DEV_IDEPCI=y -# CONFIG_IDEPCI_SHARE_IRQ is not set +CONFIG_IDEPCI_SHARE_IRQ=y # CONFIG_BLK_DEV_OFFBOARD is not set CONFIG_BLK_DEV_GENERIC=y # CONFIG_BLK_DEV_OPTI621 is not set diff --git a/arch/ia64/dig/setup.c b/arch/ia64/dig/setup.c index 38aa9c10885..9196b330ff7 100644 --- a/arch/ia64/dig/setup.c +++ b/arch/ia64/dig/setup.c @@ -8,14 +8,13 @@ * Copyright (C) 1999 Walt Drummond <drummond@valinux.com> * Copyright (C) 1999 Vijay Chander <vijay@engr.sgi.com> */ -#include <linux/config.h> #include <linux/init.h> #include <linux/delay.h> #include <linux/kernel.h> #include <linux/kdev_t.h> #include <linux/string.h> -#include <linux/tty.h> +#include <linux/screen_info.h> #include <linux/console.h> #include <linux/timex.h> #include <linux/sched.h> diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c index 5825ddee58d..db8e1fcfa04 100644 --- a/arch/ia64/hp/common/sba_iommu.c +++ b/arch/ia64/hp/common/sba_iommu.c @@ -19,7 +19,6 @@ ** */ -#include <linux/config.h> #include <linux/types.h> #include <linux/kernel.h> #include <linux/module.h> diff --git a/arch/ia64/hp/sim/boot/bootloader.c b/arch/ia64/hp/sim/boot/bootloader.c index 51a7b7b4dd0..c5e9baafafe 100644 --- a/arch/ia64/hp/sim/boot/bootloader.c +++ b/arch/ia64/hp/sim/boot/bootloader.c @@ -11,7 +11,6 @@ */ struct task_struct; /* forward declaration for elf.h */ -#include <linux/config.h> #include <linux/elf.h> #include <linux/init.h> #include <linux/kernel.h> diff --git a/arch/ia64/hp/sim/boot/fw-emu.c b/arch/ia64/hp/sim/boot/fw-emu.c index 30fdfb1d0a5..5a0a7afcfc3 100644 --- a/arch/ia64/hp/sim/boot/fw-emu.c +++ b/arch/ia64/hp/sim/boot/fw-emu.c @@ -4,7 +4,6 @@ * Copyright (C) 1998-2001 Hewlett-Packard Co * David Mosberger-Tang <davidm@hpl.hp.com> */ -#include <linux/config.h> #ifdef CONFIG_PCI # include <linux/pci.h> diff --git a/arch/ia64/hp/sim/hpsim_console.c b/arch/ia64/hp/sim/hpsim_console.c index 5deff21e587..6e149c8ab83 100644 --- a/arch/ia64/hp/sim/hpsim_console.c +++ b/arch/ia64/hp/sim/hpsim_console.c @@ -5,7 +5,6 @@ * David Mosberger-Tang <davidm@hpl.hp.com> * Copyright (C) 1999 Vijay Chander <vijay@engr.sgi.com> */ -#include <linux/config.h> #include <linux/init.h> #include <linux/kernel.h> diff --git a/arch/ia64/hp/sim/hpsim_irq.c b/arch/ia64/hp/sim/hpsim_irq.c index c0d25a2a3e9..8145547bb52 100644 --- a/arch/ia64/hp/sim/hpsim_irq.c +++ b/arch/ia64/hp/sim/hpsim_irq.c @@ -44,8 +44,8 @@ hpsim_irq_init (void) int i; for (i = 0; i < NR_IRQS; ++i) { - idesc = irq_descp(i); - if (idesc->handler == &no_irq_type) - idesc->handler = &irq_type_hp_sim; + idesc = irq_desc + i; + if (idesc->chip == &no_irq_type) + idesc->chip = &irq_type_hp_sim; } } diff --git a/arch/ia64/hp/sim/hpsim_setup.c b/arch/ia64/hp/sim/hpsim_setup.c index 694fc86bfbd..f2297192a58 100644 --- a/arch/ia64/hp/sim/hpsim_setup.c +++ b/arch/ia64/hp/sim/hpsim_setup.c @@ -5,7 +5,6 @@ * David Mosberger-Tang <davidm@hpl.hp.com> * Copyright (C) 1999 Vijay Chander <vijay@engr.sgi.com> */ -#include <linux/config.h> #include <linux/console.h> #include <linux/init.h> #include <linux/kdev_t.h> diff --git a/arch/ia64/hp/sim/simeth.c b/arch/ia64/hp/sim/simeth.c index 0639ec0ed01..e1a1b11473e 100644 --- a/arch/ia64/hp/sim/simeth.c +++ b/arch/ia64/hp/sim/simeth.c @@ -4,7 +4,6 @@ * Copyright (C) 1999-2001, 2003 Hewlett-Packard Co * Stephane Eranian <eranian@hpl.hp.com> */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/sched.h> #include <linux/types.h> @@ -321,7 +320,7 @@ simeth_device_event(struct notifier_block *this,unsigned long event, void *ptr) } printk(KERN_INFO "simeth_device_event: %s ipaddr=0x%x\n", - dev->name, htonl(ifa->ifa_local)); + dev->name, ntohl(ifa->ifa_local)); /* * XXX Fix me @@ -332,7 +331,7 @@ simeth_device_event(struct notifier_block *this,unsigned long event, void *ptr) local = dev->priv; /* now do it for real */ r = event == NETDEV_UP ? - netdev_attach(local->simfd, dev->irq, htonl(ifa->ifa_local)): + netdev_attach(local->simfd, dev->irq, ntohl(ifa->ifa_local)): netdev_detach(local->simfd); printk(KERN_INFO "simeth: netdev_attach/detach: event=%s ->%d\n", diff --git a/arch/ia64/hp/sim/simscsi.c b/arch/ia64/hp/sim/simscsi.c index a3fe9753113..8f0a16a79a6 100644 --- a/arch/ia64/hp/sim/simscsi.c +++ b/arch/ia64/hp/sim/simscsi.c @@ -151,7 +151,7 @@ static void simscsi_sg_readwrite (struct scsi_cmnd *sc, int mode, unsigned long offset) { int list_len = sc->use_sg; - struct scatterlist *sl = (struct scatterlist *)sc->buffer; + struct scatterlist *sl = (struct scatterlist *)sc->request_buffer; struct disk_stat stat; struct disk_req req; @@ -244,7 +244,8 @@ static void simscsi_fillresult(struct scsi_cmnd *sc, char *buf, unsigned len) if (scatterlen == 0) memcpy(sc->request_buffer, buf, len); - else for (slp = (struct scatterlist *)sc->buffer; scatterlen-- > 0 && len > 0; slp++) { + else for (slp = (struct scatterlist *)sc->request_buffer; + scatterlen-- > 0 && len > 0; slp++) { unsigned thislen = min(len, slp->length); memcpy(page_address(slp->page) + slp->offset, buf, thislen); diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c index 0e5c6ae5022..0daacc20ed3 100644 --- a/arch/ia64/hp/sim/simserial.c +++ b/arch/ia64/hp/sim/simserial.c @@ -16,7 +16,6 @@ * 07/30/02 D. Mosberger Replace sti()/cli() with explicit spinlocks & local irq masking */ -#include <linux/config.h> #include <linux/init.h> #include <linux/errno.h> #include <linux/sched.h> @@ -47,7 +46,7 @@ #define NR_PORTS 1 /* only one port for now */ -#define IRQ_T(info) ((info->flags & ASYNC_SHARE_IRQ) ? SA_SHIRQ : SA_INTERRUPT) +#define IRQ_T(info) ((info->flags & ASYNC_SHARE_IRQ) ? IRQF_SHARED : IRQF_DISABLED) #define SSC_GETCHAR 21 diff --git a/arch/ia64/ia32/Makefile b/arch/ia64/ia32/Makefile index 61cb60affd9..baad8c7699c 100644 --- a/arch/ia64/ia32/Makefile +++ b/arch/ia64/ia32/Makefile @@ -4,6 +4,7 @@ obj-y := ia32_entry.o sys_ia32.o ia32_signal.o \ ia32_support.o ia32_traps.o binfmt_elf32.o ia32_ldt.o +obj-$(CONFIG_AUDIT) += audit.o # Don't let GCC uses f16-f31 so that save_ia32_fpstate_live() and # restore_ia32_fpstate_live() can be sure the live register contain user-level state. diff --git a/arch/ia64/ia32/audit.c b/arch/ia64/ia32/audit.c new file mode 100644 index 00000000000..92d7d0c8d93 --- /dev/null +++ b/arch/ia64/ia32/audit.c @@ -0,0 +1,37 @@ +#include <asm-i386/unistd.h> + +unsigned ia32_dir_class[] = { +#include <asm-generic/audit_dir_write.h> +~0U +}; + +unsigned ia32_chattr_class[] = { +#include <asm-generic/audit_change_attr.h> +~0U +}; + +unsigned ia32_write_class[] = { +#include <asm-generic/audit_write.h> +~0U +}; + +unsigned ia32_read_class[] = { +#include <asm-generic/audit_read.h> +~0U +}; + +int ia32_classify_syscall(unsigned syscall) +{ + switch(syscall) { + case __NR_open: + return 2; + case __NR_openat: + return 3; + case __NR_socketcall: + return 4; + case __NR_execve: + return 5; + default: + return 1; + } +} diff --git a/arch/ia64/ia32/binfmt_elf32.c b/arch/ia64/ia32/binfmt_elf32.c index da03c06744f..daa6b91bc92 100644 --- a/arch/ia64/ia32/binfmt_elf32.c +++ b/arch/ia64/ia32/binfmt_elf32.c @@ -9,7 +9,6 @@ * 04/13/01 D. Mosberger dropped saving tssd in ar.k1---it's not needed * 09/14/01 D. Mosberger fixed memory management for gdt/tss page */ -#include <linux/config.h> #include <linux/types.h> #include <linux/mm.h> diff --git a/arch/ia64/ia32/ia32priv.h b/arch/ia64/ia32/ia32priv.h index ccb98ed48e5..703a67c934f 100644 --- a/arch/ia64/ia32/ia32priv.h +++ b/arch/ia64/ia32/ia32priv.h @@ -1,7 +1,6 @@ #ifndef _ASM_IA64_IA32_PRIV_H #define _ASM_IA64_IA32_PRIV_H -#include <linux/config.h> #include <asm/ia32.h> diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 5366b3b23d0..bddbd22706e 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -14,7 +14,6 @@ * environment. */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/syscalls.h> #include <linux/sysctl.h> @@ -1943,7 +1942,7 @@ struct sysctl32 { unsigned int __unused[4]; }; -#ifdef CONFIG_SYSCTL +#ifdef CONFIG_SYSCTL_SYSCALL asmlinkage long sys32_sysctl (struct sysctl32 __user *args) { diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile index 09a0dbc17fb..31497496eb4 100644 --- a/arch/ia64/kernel/Makefile +++ b/arch/ia64/kernel/Makefile @@ -29,8 +29,14 @@ obj-$(CONFIG_CPU_FREQ) += cpufreq/ obj-$(CONFIG_IA64_MCA_RECOVERY) += mca_recovery.o obj-$(CONFIG_KPROBES) += kprobes.o jprobes.o obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR) += uncached.o +obj-$(CONFIG_AUDIT) += audit.o mca_recovery-y += mca_drv.o mca_drv_asm.o +obj-$(CONFIG_IA64_ESI) += esi.o +ifneq ($(CONFIG_IA64_ESI),) +obj-y += esi_stub.o # must be in kernel proper +endif + # The gate DSO image is built using a special linker script. targets += gate.so gate-syms.o @@ -44,7 +50,8 @@ CPPFLAGS_gate.lds := -P -C -U$(ARCH) quiet_cmd_gate = GATE $@ cmd_gate = $(CC) -nostdlib $(GATECFLAGS_$(@F)) -Wl,-T,$(filter-out FORCE,$^) -o $@ -GATECFLAGS_gate.so = -shared -s -Wl,-soname=linux-gate.so.1 +GATECFLAGS_gate.so = -shared -s -Wl,-soname=linux-gate.so.1 \ + $(call ld-option, -Wl$(comma)--hash-style=sysv) $(obj)/gate.so: $(obj)/gate.lds $(obj)/gate.o FORCE $(call if_changed,gate) diff --git a/arch/ia64/kernel/acpi-ext.c b/arch/ia64/kernel/acpi-ext.c index fff82929d22..b7515bc808a 100644 --- a/arch/ia64/kernel/acpi-ext.c +++ b/arch/ia64/kernel/acpi-ext.c @@ -8,7 +8,6 @@ * published by the Free Software Foundation. */ -#include <linux/config.h> #include <linux/module.h> #include <linux/types.h> #include <linux/acpi.h> @@ -51,7 +50,7 @@ static acpi_status hp_ccsr_locate(acpi_handle obj, u64 *base, u64 *length) memcpy(length, vendor->byte_data + 8, sizeof(*length)); exit: - acpi_os_free(buffer.pointer); + kfree(buffer.pointer); return status; } diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index ca16d9556bd..32c3abededc 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -32,7 +32,6 @@ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ */ -#include <linux/config.h> #include <linux/module.h> #include <linux/init.h> #include <linux/kernel.h> @@ -56,7 +55,7 @@ #define BAD_MADT_ENTRY(entry, end) ( \ (!entry) || (unsigned long)entry + sizeof(*entry) > end || \ - ((acpi_table_entry_header *)entry)->length != sizeof(*entry)) + ((acpi_table_entry_header *)entry)->length < sizeof(*entry)) #define PREFIX "ACPI: " @@ -772,16 +771,19 @@ int acpi_map_cpu2node(acpi_handle handle, int cpu, long physid) { #ifdef CONFIG_ACPI_NUMA int pxm_id; + int nid; pxm_id = acpi_get_pxm(handle); - /* - * Assuming that the container driver would have set the proximity - * domain and would have initialized pxm_to_node(pxm_id) && pxm_flag + * We don't have cpu-only-node hotadd. But if the system equips + * SRAT table, pxm is already found and node is ready. + * So, just pxm_to_nid(pxm) is OK. + * This code here is for the system which doesn't have full SRAT + * table for possible cpus. */ - node_cpuid[cpu].nid = (pxm_id < 0) ? 0 : pxm_to_node(pxm_id); - + nid = acpi_map_pxm_to_node(pxm_id); node_cpuid[cpu].phys_id = physid; + node_cpuid[cpu].nid = nid; #endif return (0); } @@ -857,7 +859,7 @@ int acpi_map_lsapic(acpi_handle handle, int *pcpu) obj = buffer.pointer; if (obj->type != ACPI_TYPE_BUFFER || obj->buffer.length < sizeof(*lsapic)) { - acpi_os_free(buffer.pointer); + kfree(buffer.pointer); return -EINVAL; } @@ -865,13 +867,13 @@ int acpi_map_lsapic(acpi_handle handle, int *pcpu) if ((lsapic->header.type != ACPI_MADT_LSAPIC) || (!lsapic->flags.enabled)) { - acpi_os_free(buffer.pointer); + kfree(buffer.pointer); return -EINVAL; } physid = ((lsapic->id << 8) | (lsapic->eid)); - acpi_os_free(buffer.pointer); + kfree(buffer.pointer); buffer.length = ACPI_ALLOCATE_BUFFER; buffer.pointer = NULL; @@ -935,20 +937,20 @@ acpi_map_iosapic(acpi_handle handle, u32 depth, void *context, void **ret) obj = buffer.pointer; if (obj->type != ACPI_TYPE_BUFFER || obj->buffer.length < sizeof(*iosapic)) { - acpi_os_free(buffer.pointer); + kfree(buffer.pointer); return AE_OK; } iosapic = (struct acpi_table_iosapic *)obj->buffer.pointer; if (iosapic->header.type != ACPI_MADT_IOSAPIC) { - acpi_os_free(buffer.pointer); + kfree(buffer.pointer); return AE_OK; } gsi_base = iosapic->global_irq_base; - acpi_os_free(buffer.pointer); + kfree(buffer.pointer); /* * OK, it's an IOSAPIC MADT entry, look for a _PXM value to tell diff --git a/arch/ia64/kernel/asm-offsets.c b/arch/ia64/kernel/asm-offsets.c index 16e7b6600ae..75a2a2c1225 100644 --- a/arch/ia64/kernel/asm-offsets.c +++ b/arch/ia64/kernel/asm-offsets.c @@ -5,7 +5,6 @@ */ #define ASM_OFFSETS_C 1 -#include <linux/config.h> #include <linux/sched.h> diff --git a/arch/ia64/kernel/audit.c b/arch/ia64/kernel/audit.c new file mode 100644 index 00000000000..04682555a28 --- /dev/null +++ b/arch/ia64/kernel/audit.c @@ -0,0 +1,64 @@ +#include <linux/init.h> +#include <linux/types.h> +#include <linux/audit.h> +#include <asm/unistd.h> + +static unsigned dir_class[] = { +#include <asm-generic/audit_dir_write.h> +~0U +}; + +static unsigned read_class[] = { +#include <asm-generic/audit_read.h> +~0U +}; + +static unsigned write_class[] = { +#include <asm-generic/audit_write.h> +~0U +}; + +static unsigned chattr_class[] = { +#include <asm-generic/audit_change_attr.h> +~0U +}; + +int audit_classify_syscall(int abi, unsigned syscall) +{ +#ifdef CONFIG_IA32_SUPPORT + extern int ia32_classify_syscall(unsigned); + if (abi == AUDIT_ARCH_I386) + return ia32_classify_syscall(syscall); +#endif + switch(syscall) { + case __NR_open: + return 2; + case __NR_openat: + return 3; + case __NR_execve: + return 5; + default: + return 0; + } +} + +static int __init audit_classes_init(void) +{ +#ifdef CONFIG_IA32_SUPPORT + extern __u32 ia32_dir_class[]; + extern __u32 ia32_write_class[]; + extern __u32 ia32_read_class[]; + extern __u32 ia32_chattr_class[]; + audit_register_class(AUDIT_CLASS_WRITE_32, ia32_write_class); + audit_register_class(AUDIT_CLASS_READ_32, ia32_read_class); + audit_register_class(AUDIT_CLASS_DIR_WRITE_32, ia32_dir_class); + audit_register_class(AUDIT_CLASS_CHATTR_32, ia32_chattr_class); +#endif + audit_register_class(AUDIT_CLASS_WRITE, write_class); + audit_register_class(AUDIT_CLASS_READ, read_class); + audit_register_class(AUDIT_CLASS_DIR_WRITE, dir_class); + audit_register_class(AUDIT_CLASS_CHATTR, chattr_class); + return 0; +} + +__initcall(audit_classes_init); diff --git a/arch/ia64/kernel/cpufreq/acpi-cpufreq.c b/arch/ia64/kernel/cpufreq/acpi-cpufreq.c index 5a1bf815282..86faf221a07 100644 --- a/arch/ia64/kernel/cpufreq/acpi-cpufreq.c +++ b/arch/ia64/kernel/cpufreq/acpi-cpufreq.c @@ -9,7 +9,6 @@ * Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c index c33d0ba7e30..bb8770a177b 100644 --- a/arch/ia64/kernel/efi.c +++ b/arch/ia64/kernel/efi.c @@ -20,7 +20,6 @@ * Goutham Rao: <goutham.rao@intel.com> * Skip non-WB memory and ignore empty memory ranges. */ -#include <linux/config.h> #include <linux/module.h> #include <linux/kernel.h> #include <linux/init.h> @@ -633,7 +632,7 @@ kern_memory_descriptor (unsigned long phys_addr) if (phys_addr - md->start < (md->num_pages << EFI_PAGE_SHIFT)) return md; } - return 0; + return NULL; } static efi_memory_desc_t * @@ -653,7 +652,7 @@ efi_memory_descriptor (unsigned long phys_addr) if (phys_addr - md->phys_addr < (md->num_pages << EFI_PAGE_SHIFT)) return md; } - return 0; + return NULL; } u32 @@ -760,7 +759,7 @@ valid_phys_addr_range (unsigned long phys_addr, unsigned long size) } int -valid_mmap_phys_addr_range (unsigned long phys_addr, unsigned long size) +valid_mmap_phys_addr_range (unsigned long pfn, unsigned long size) { /* * MMIO regions are often missing from the EFI memory map. @@ -924,7 +923,7 @@ find_memmap_space (void) void efi_memmap_init(unsigned long *s, unsigned long *e) { - struct kern_memdesc *k, *prev = 0; + struct kern_memdesc *k, *prev = NULL; u64 contig_low=0, contig_high=0; u64 as, ae, lim; void *efi_map_start, *efi_map_end, *p, *q; diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 32c999f58d1..12701cf32d9 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -31,7 +31,6 @@ * pNonSys: !pSys */ -#include <linux/config.h> #include <asm/asmmacro.h> #include <asm/cache.h> diff --git a/arch/ia64/kernel/entry.h b/arch/ia64/kernel/entry.h index ebc3dfb8882..b83edac0296 100644 --- a/arch/ia64/kernel/entry.h +++ b/arch/ia64/kernel/entry.h @@ -1,4 +1,3 @@ -#include <linux/config.h> /* * Preserved registers that are shared between code in ivt.S and diff --git a/arch/ia64/kernel/esi.c b/arch/ia64/kernel/esi.c new file mode 100644 index 00000000000..ebf4e988e78 --- /dev/null +++ b/arch/ia64/kernel/esi.c @@ -0,0 +1,205 @@ +/* + * Extensible SAL Interface (ESI) support routines. + * + * Copyright (C) 2006 Hewlett-Packard Co + * Alex Williamson <alex.williamson@hp.com> + */ +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/string.h> + +#include <asm/esi.h> +#include <asm/sal.h> + +MODULE_AUTHOR("Alex Williamson <alex.williamson@hp.com>"); +MODULE_DESCRIPTION("Extensible SAL Interface (ESI) support"); +MODULE_LICENSE("GPL"); + +#define MODULE_NAME "esi" + +#define ESI_TABLE_GUID \ + EFI_GUID(0x43EA58DC, 0xCF28, 0x4b06, 0xB3, \ + 0x91, 0xB7, 0x50, 0x59, 0x34, 0x2B, 0xD4) + +enum esi_systab_entry_type { + ESI_DESC_ENTRY_POINT = 0 +}; + +/* + * Entry type: Size: + * 0 48 + */ +#define ESI_DESC_SIZE(type) "\060"[(unsigned) (type)] + +typedef struct ia64_esi_desc_entry_point { + u8 type; + u8 reserved1[15]; + u64 esi_proc; + u64 gp; + efi_guid_t guid; +} ia64_esi_desc_entry_point_t; + +struct pdesc { + void *addr; + void *gp; +}; + +static struct ia64_sal_systab *esi_systab; + +static int __init esi_init (void) +{ + efi_config_table_t *config_tables; + struct ia64_sal_systab *systab; + unsigned long esi = 0; + char *p; + int i; + + config_tables = __va(efi.systab->tables); + + for (i = 0; i < (int) efi.systab->nr_tables; ++i) { + if (efi_guidcmp(config_tables[i].guid, ESI_TABLE_GUID) == 0) { + esi = config_tables[i].table; + break; + } + } + + if (!esi) + return -ENODEV;; + + systab = __va(esi); + + if (strncmp(systab->signature, "ESIT", 4) != 0) { + printk(KERN_ERR "bad signature in ESI system table!"); + return -ENODEV; + } + + p = (char *) (systab + 1); + for (i = 0; i < systab->entry_count; i++) { + /* + * The first byte of each entry type contains the type + * descriptor. + */ + switch (*p) { + case ESI_DESC_ENTRY_POINT: + break; + default: + printk(KERN_WARNING "Unkown table type %d found in " + "ESI table, ignoring rest of table\n", *p); + return -ENODEV; + } + + p += ESI_DESC_SIZE(*p); + } + + esi_systab = systab; + return 0; +} + + +int ia64_esi_call (efi_guid_t guid, struct ia64_sal_retval *isrvp, + enum esi_proc_type proc_type, u64 func, + u64 arg1, u64 arg2, u64 arg3, u64 arg4, u64 arg5, u64 arg6, + u64 arg7) +{ + struct ia64_fpreg fr[6]; + unsigned long flags = 0; + int i; + char *p; + + if (!esi_systab) + return -1; + + p = (char *) (esi_systab + 1); + for (i = 0; i < esi_systab->entry_count; i++) { + if (*p == ESI_DESC_ENTRY_POINT) { + ia64_esi_desc_entry_point_t *esi = (void *)p; + if (!efi_guidcmp(guid, esi->guid)) { + ia64_sal_handler esi_proc; + struct pdesc pdesc; + + pdesc.addr = __va(esi->esi_proc); + pdesc.gp = __va(esi->gp); + + esi_proc = (ia64_sal_handler) &pdesc; + + ia64_save_scratch_fpregs(fr); + if (proc_type == ESI_PROC_SERIALIZED) + spin_lock_irqsave(&sal_lock, flags); + else if (proc_type == ESI_PROC_MP_SAFE) + local_irq_save(flags); + else + preempt_disable(); + *isrvp = (*esi_proc)(func, arg1, arg2, arg3, + arg4, arg5, arg6, arg7); + if (proc_type == ESI_PROC_SERIALIZED) + spin_unlock_irqrestore(&sal_lock, + flags); + else if (proc_type == ESI_PROC_MP_SAFE) + local_irq_restore(flags); + else + preempt_enable(); + ia64_load_scratch_fpregs(fr); + return 0; + } + } + p += ESI_DESC_SIZE(*p); + } + return -1; +} +EXPORT_SYMBOL_GPL(ia64_esi_call); + +int ia64_esi_call_phys (efi_guid_t guid, struct ia64_sal_retval *isrvp, + u64 func, u64 arg1, u64 arg2, u64 arg3, u64 arg4, + u64 arg5, u64 arg6, u64 arg7) +{ + struct ia64_fpreg fr[6]; + unsigned long flags; + u64 esi_params[8]; + char *p; + int i; + + if (!esi_systab) + return -1; + + p = (char *) (esi_systab + 1); + for (i = 0; i < esi_systab->entry_count; i++) { + if (*p == ESI_DESC_ENTRY_POINT) { + ia64_esi_desc_entry_point_t *esi = (void *)p; + if (!efi_guidcmp(guid, esi->guid)) { + ia64_sal_handler esi_proc; + struct pdesc pdesc; + + pdesc.addr = (void *)esi->esi_proc; + pdesc.gp = (void *)esi->gp; + + esi_proc = (ia64_sal_handler) &pdesc; + + esi_params[0] = func; + esi_params[1] = arg1; + esi_params[2] = arg2; + esi_params[3] = arg3; + esi_params[4] = arg4; + esi_params[5] = arg5; + esi_params[6] = arg6; + esi_params[7] = arg7; + ia64_save_scratch_fpregs(fr); + spin_lock_irqsave(&sal_lock, flags); + *isrvp = esi_call_phys(esi_proc, esi_params); + spin_unlock_irqrestore(&sal_lock, flags); + ia64_load_scratch_fpregs(fr); + return 0; + } + } + p += ESI_DESC_SIZE(*p); + } + return -1; +} +EXPORT_SYMBOL_GPL(ia64_esi_call_phys); + +static void __exit esi_exit (void) +{ +} + +module_init(esi_init); +module_exit(esi_exit); /* makes module removable... */ diff --git a/arch/ia64/kernel/esi_stub.S b/arch/ia64/kernel/esi_stub.S new file mode 100644 index 00000000000..6b3d6c1f99b --- /dev/null +++ b/arch/ia64/kernel/esi_stub.S @@ -0,0 +1,96 @@ +/* + * ESI call stub. + * + * Copyright (C) 2005 Hewlett-Packard Co + * Alex Williamson <alex.williamson@hp.com> + * + * Based on EFI call stub by David Mosberger. The stub is virtually + * identical to the one for EFI phys-mode calls, except that ESI + * calls may have up to 8 arguments, so they get passed to this routine + * through memory. + * + * This stub allows us to make ESI calls in physical mode with interrupts + * turned off. ESI calls may not support calling from virtual mode. + * + * Google for "Extensible SAL specification" for a document describing the + * ESI standard. + */ + +/* + * PSR settings as per SAL spec (Chapter 8 in the "IA-64 System + * Abstraction Layer Specification", revision 2.6e). Note that + * psr.dfl and psr.dfh MUST be cleared, despite what this manual says. + * Otherwise, SAL dies whenever it's trying to do an IA-32 BIOS call + * (the br.ia instruction fails unless psr.dfl and psr.dfh are + * cleared). Fortunately, SAL promises not to touch the floating + * point regs, so at least we don't have to save f2-f127. + */ +#define PSR_BITS_TO_CLEAR \ + (IA64_PSR_I | IA64_PSR_IT | IA64_PSR_DT | IA64_PSR_RT | \ + IA64_PSR_DD | IA64_PSR_SS | IA64_PSR_RI | IA64_PSR_ED | \ + IA64_PSR_DFL | IA64_PSR_DFH) + +#define PSR_BITS_TO_SET \ + (IA64_PSR_BN) + +#include <asm/processor.h> +#include <asm/asmmacro.h> + +/* + * Inputs: + * in0 = address of function descriptor of ESI routine to call + * in1 = address of array of ESI parameters + * + * Outputs: + * r8 = result returned by called function + */ +GLOBAL_ENTRY(esi_call_phys) + .prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(2) + alloc loc1=ar.pfs,2,7,8,0 + ld8 r2=[in0],8 // load ESI function's entry point + mov loc0=rp + .body + ;; + ld8 out0=[in1],8 // ESI params loaded from array + ;; // passing all as inputs doesn't work + ld8 out1=[in1],8 + ;; + ld8 out2=[in1],8 + ;; + ld8 out3=[in1],8 + ;; + ld8 out4=[in1],8 + ;; + ld8 out5=[in1],8 + ;; + ld8 out6=[in1],8 + ;; + ld8 out7=[in1] + mov loc2=gp // save global pointer + mov loc4=ar.rsc // save RSE configuration + mov ar.rsc=0 // put RSE in enforced lazy, LE mode + ;; + ld8 gp=[in0] // load ESI function's global pointer + movl r16=PSR_BITS_TO_CLEAR + mov loc3=psr // save processor status word + movl r17=PSR_BITS_TO_SET + ;; + or loc3=loc3,r17 + mov b6=r2 + ;; + andcm r16=loc3,r16 // get psr with IT, DT, and RT bits cleared + br.call.sptk.many rp=ia64_switch_mode_phys +.ret0: mov loc5=r19 // old ar.bsp + mov loc6=r20 // old sp + br.call.sptk.many rp=b6 // call the ESI function +.ret1: mov ar.rsc=0 // put RSE in enforced lazy, LE mode + mov r16=loc3 // save virtual mode psr + mov r19=loc5 // save virtual mode bspstore + mov r20=loc6 // save virtual mode sp + br.call.sptk.many rp=ia64_switch_mode_virt // return to virtual mode +.ret2: mov ar.rsc=loc4 // restore RSE configuration + mov ar.pfs=loc1 + mov rp=loc0 + mov gp=loc2 + br.ret.sptk.many rp +END(esi_call_phys) diff --git a/arch/ia64/kernel/gate.S b/arch/ia64/kernel/gate.S index 86064ca9895..3274850cf27 100644 --- a/arch/ia64/kernel/gate.S +++ b/arch/ia64/kernel/gate.S @@ -6,7 +6,6 @@ * David Mosberger-Tang <davidm@hpl.hp.com> */ -#include <linux/config.h> #include <asm/asmmacro.h> #include <asm/errno.h> diff --git a/arch/ia64/kernel/gate.lds.S b/arch/ia64/kernel/gate.lds.S index 7c99e6ec3da..6d198339bf8 100644 --- a/arch/ia64/kernel/gate.lds.S +++ b/arch/ia64/kernel/gate.lds.S @@ -4,7 +4,6 @@ * in one page). This script controls its layout. */ -#include <linux/config.h> #include <asm/system.h> @@ -13,6 +12,7 @@ SECTIONS . = GATE_ADDR + SIZEOF_HEADERS; .hash : { *(.hash) } :readable + .gnu.hash : { *(.gnu.hash) } .dynsym : { *(.dynsym) } .dynstr : { *(.dynstr) } .gnu.version : { *(.gnu.version) } diff --git a/arch/ia64/kernel/head.S b/arch/ia64/kernel/head.S index f1778a84ea6..44d540efa6d 100644 --- a/arch/ia64/kernel/head.S +++ b/arch/ia64/kernel/head.S @@ -19,7 +19,6 @@ * Support for CPU Hotplug */ -#include <linux/config.h> #include <asm/asmmacro.h> #include <asm/fpu.h> @@ -198,6 +197,11 @@ start_ap: ;; srlz.i ;; + { + flushrs // must be first insn in group + srlz.i + } + ;; /* * Save the region registers, predicate before they get clobbered */ @@ -854,7 +858,6 @@ END(__ia64_init_fpu) */ GLOBAL_ENTRY(ia64_switch_mode_phys) { - alloc r2=ar.pfs,0,0,0,0 rsm psr.i | psr.ic // disable interrupts and interrupt collection mov r15=ip } @@ -903,7 +906,6 @@ END(ia64_switch_mode_phys) */ GLOBAL_ENTRY(ia64_switch_mode_virt) { - alloc r2=ar.pfs,0,0,0,0 rsm psr.i | psr.ic // disable interrupts and interrupt collection mov r15=ip } diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c index bbcfd08378a..879c1817bd1 100644 --- a/arch/ia64/kernel/ia64_ksyms.c +++ b/arch/ia64/kernel/ia64_ksyms.c @@ -5,7 +5,6 @@ * All other exports should be put directly after the definition. */ -#include <linux/config.h> #include <linux/module.h> #include <linux/string.h> @@ -63,7 +62,7 @@ EXPORT_SYMBOL(__udivdi3); EXPORT_SYMBOL(__moddi3); EXPORT_SYMBOL(__umoddi3); -#if defined(CONFIG_MD_RAID5) || defined(CONFIG_MD_RAID5_MODULE) +#if defined(CONFIG_MD_RAID456) || defined(CONFIG_MD_RAID456_MODULE) extern void xor_ia64_2(void); extern void xor_ia64_3(void); extern void xor_ia64_4(void); @@ -106,5 +105,9 @@ EXPORT_SYMBOL(ia64_spinlock_contention); # endif #endif +#if defined(CONFIG_IA64_ESI) || defined(CONFIG_IA64_ESI_MODULE) +extern void esi_call_phys (void); +EXPORT_SYMBOL_GPL(esi_call_phys); +#endif extern char ia64_ivt[]; EXPORT_SYMBOL(ia64_ivt); diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index d58c1c5c903..9bf15fefa7e 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -79,7 +79,6 @@ * describeinterrupts. Now we use "IRQ" only for Linux IRQ's. ISA IRQ * (isa_irq) is the only exception in this source code. */ -#include <linux/config.h> #include <linux/acpi.h> #include <linux/init.h> @@ -456,7 +455,7 @@ iosapic_startup_edge_irq (unsigned int irq) static void iosapic_ack_edge_irq (unsigned int irq) { - irq_desc_t *idesc = irq_descp(irq); + irq_desc_t *idesc = irq_desc + irq; move_native_irq(irq); /* @@ -659,14 +658,14 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery, else irq_type = &irq_type_iosapic_level; - idesc = irq_descp(vector); - if (idesc->handler != irq_type) { - if (idesc->handler != &no_irq_type) + idesc = irq_desc + vector; + if (idesc->chip != irq_type) { + if (idesc->chip != &no_irq_type) printk(KERN_WARNING "%s: changing vector %d from %s to %s\n", __FUNCTION__, vector, - idesc->handler->typename, irq_type->typename); - idesc->handler = irq_type; + idesc->chip->typename, irq_type->typename); + idesc->chip = irq_type; } return 0; } @@ -793,14 +792,14 @@ again: return -ENOSPC; } - spin_lock_irqsave(&irq_descp(vector)->lock, flags); + spin_lock_irqsave(&irq_desc[vector].lock, flags); spin_lock(&iosapic_lock); { if (gsi_to_vector(gsi) > 0) { if (list_empty(&iosapic_intr_info[vector].rtes)) free_irq_vector(vector); spin_unlock(&iosapic_lock); - spin_unlock_irqrestore(&irq_descp(vector)->lock, + spin_unlock_irqrestore(&irq_desc[vector].lock, flags); goto again; } @@ -810,7 +809,7 @@ again: polarity, trigger); if (err < 0) { spin_unlock(&iosapic_lock); - spin_unlock_irqrestore(&irq_descp(vector)->lock, + spin_unlock_irqrestore(&irq_desc[vector].lock, flags); return err; } @@ -825,7 +824,7 @@ again: set_rte(gsi, vector, dest, mask); } spin_unlock(&iosapic_lock); - spin_unlock_irqrestore(&irq_descp(vector)->lock, flags); + spin_unlock_irqrestore(&irq_desc[vector].lock, flags); printk(KERN_INFO "GSI %u (%s, %s) -> CPU %d (0x%04x) vector %d\n", gsi, (trigger == IOSAPIC_EDGE ? "edge" : "level"), @@ -860,7 +859,7 @@ iosapic_unregister_intr (unsigned int gsi) } vector = irq_to_vector(irq); - idesc = irq_descp(irq); + idesc = irq_desc + irq; spin_lock_irqsave(&idesc->lock, flags); spin_lock(&iosapic_lock); { @@ -903,7 +902,7 @@ iosapic_unregister_intr (unsigned int gsi) BUG_ON(iosapic_intr_info[vector].count); /* Clear the interrupt controller descriptor */ - idesc->handler = &no_irq_type; + idesc->chip = &no_irq_type; /* Clear the interrupt information */ memset(&iosapic_intr_info[vector], 0, diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index 9c72ea3f643..7852382de2f 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -76,7 +76,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); } #endif - seq_printf(p, " %14s", irq_desc[i].handler->typename); + seq_printf(p, " %14s", irq_desc[i].chip->typename); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) @@ -100,7 +100,7 @@ void set_irq_affinity_info (unsigned int irq, int hwid, int redir) cpu_set(cpu_logical_id(hwid), mask); if (irq < NR_IRQS) { - irq_affinity[irq] = mask; + irq_desc[irq].affinity = mask; irq_redir[irq] = (char) (redir & 0xff); } } @@ -120,7 +120,7 @@ static void migrate_irqs(void) int irq, new_cpu; for (irq=0; irq < NR_IRQS; irq++) { - desc = irq_descp(irq); + desc = irq_desc + irq; /* * No handling for now. @@ -131,7 +131,7 @@ static void migrate_irqs(void) if (desc->status == IRQ_PER_CPU) continue; - cpus_and(mask, irq_affinity[irq], cpu_online_map); + cpus_and(mask, irq_desc[irq].affinity, cpu_online_map); if (any_online_cpu(mask) == NR_CPUS) { /* * Save it for phase 2 processing @@ -144,15 +144,15 @@ static void migrate_irqs(void) /* * Al three are essential, currently WARN_ON.. maybe panic? */ - if (desc->handler && desc->handler->disable && - desc->handler->enable && desc->handler->set_affinity) { - desc->handler->disable(irq); - desc->handler->set_affinity(irq, mask); - desc->handler->enable(irq); + if (desc->chip && desc->chip->disable && + desc->chip->enable && desc->chip->set_affinity) { + desc->chip->disable(irq); + desc->chip->set_affinity(irq, mask); + desc->chip->enable(irq); } else { - WARN_ON((!(desc->handler) || !(desc->handler->disable) || - !(desc->handler->enable) || - !(desc->handler->set_affinity))); + WARN_ON((!(desc->chip) || !(desc->chip->disable) || + !(desc->chip->enable) || + !(desc->chip->set_affinity))); } } } diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index ef9a2b49307..a041367f043 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -14,7 +14,6 @@ * Added CPU Hotplug handling for IPF. */ -#include <linux/config.h> #include <linux/module.h> #include <linux/jiffies.h> @@ -236,7 +235,7 @@ extern irqreturn_t handle_IPI (int irq, void *dev_id, struct pt_regs *regs); static struct irqaction ipi_irqaction = { .handler = handle_IPI, - .flags = SA_INTERRUPT, + .flags = IRQF_DISABLED, .name = "IPI" }; #endif @@ -249,9 +248,9 @@ register_percpu_irq (ia64_vector vec, struct irqaction *action) for (irq = 0; irq < NR_IRQS; ++irq) if (irq_to_vector(irq) == vec) { - desc = irq_descp(irq); + desc = irq_desc + irq; desc->status |= IRQ_PER_CPU; - desc->handler = &irq_type_ia64_lsapic; + desc->chip = &irq_type_ia64_lsapic; if (action) setup_irq(irq, action); } diff --git a/arch/ia64/kernel/irq_lsapic.c b/arch/ia64/kernel/irq_lsapic.c index ea14e6a0440..1ab58b09f3d 100644 --- a/arch/ia64/kernel/irq_lsapic.c +++ b/arch/ia64/kernel/irq_lsapic.c @@ -26,6 +26,13 @@ lsapic_noop (unsigned int irq) /* nuthing to do... */ } +static int lsapic_retrigger(unsigned int irq) +{ + ia64_resend_irq(irq); + + return 1; +} + struct hw_interrupt_type irq_type_ia64_lsapic = { .typename = "LSAPIC", .startup = lsapic_noop_startup, @@ -33,5 +40,6 @@ struct hw_interrupt_type irq_type_ia64_lsapic = { .enable = lsapic_noop, .disable = lsapic_noop, .ack = lsapic_noop, - .end = lsapic_noop + .end = lsapic_noop, + .retrigger = lsapic_retrigger, }; diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S index 829a43cab79..6b7fcbd3f6f 100644 --- a/arch/ia64/kernel/ivt.S +++ b/arch/ia64/kernel/ivt.S @@ -38,7 +38,6 @@ * Table is based upon EAS2.6 (Oct 1999) */ -#include <linux/config.h> #include <asm/asmmacro.h> #include <asm/break.h> diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index f9039f88d01..169ec3a7156 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -23,7 +23,6 @@ * <anil.s.keshavamurthy@intel.com> adapted from i386 */ -#include <linux/config.h> #include <linux/kprobes.h> #include <linux/ptrace.h> #include <linux/string.h> @@ -137,10 +136,8 @@ static void __kprobes update_kprobe_inst_flag(uint template, uint slot, static int __kprobes unsupported_inst(uint template, uint slot, uint major_opcode, unsigned long kprobe_inst, - struct kprobe *p) + unsigned long addr) { - unsigned long addr = (unsigned long)p->addr; - if (bundle_encoding[template][slot] == I) { switch (major_opcode) { case 0x0: //I_UNIT_MISC_OPCODE: @@ -218,7 +215,7 @@ static void __kprobes prepare_break_inst(uint template, uint slot, struct kprobe *p) { unsigned long break_inst = BREAK_INST; - bundle_t *bundle = &p->ainsn.insn.bundle; + bundle_t *bundle = &p->opcode.bundle; /* * Copy the original kprobe_inst qualifying predicate(qp) @@ -424,11 +421,9 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p) unsigned long *kprobe_addr = (unsigned long *)(addr & ~0xFULL); unsigned long kprobe_inst=0; unsigned int slot = addr & 0xf, template, major_opcode = 0; - bundle_t *bundle = &p->ainsn.insn.bundle; - - memcpy(&p->opcode.bundle, kprobe_addr, sizeof(bundle_t)); - memcpy(&p->ainsn.insn.bundle, kprobe_addr, sizeof(bundle_t)); + bundle_t *bundle; + bundle = &((kprobe_opcode_t *)kprobe_addr)->bundle; template = bundle->quad0.template; if(valid_kprobe_addr(template, slot, addr)) @@ -441,9 +436,16 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p) /* Get kprobe_inst and major_opcode from the bundle */ get_kprobe_inst(bundle, slot, &kprobe_inst, &major_opcode); - if (unsupported_inst(template, slot, major_opcode, kprobe_inst, p)) + if (unsupported_inst(template, slot, major_opcode, kprobe_inst, addr)) return -EINVAL; + + p->ainsn.insn = get_insn_slot(); + if (!p->ainsn.insn) + return -ENOMEM; + memcpy(&p->opcode, kprobe_addr, sizeof(kprobe_opcode_t)); + memcpy(p->ainsn.insn, kprobe_addr, sizeof(kprobe_opcode_t)); + prepare_break_inst(template, slot, major_opcode, kprobe_inst, p); return 0; @@ -454,8 +456,10 @@ void __kprobes arch_arm_kprobe(struct kprobe *p) unsigned long addr = (unsigned long)p->addr; unsigned long arm_addr = addr & ~0xFULL; - memcpy((char *)arm_addr, &p->ainsn.insn.bundle, sizeof(bundle_t)); - flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t)); + flush_icache_range((unsigned long)p->ainsn.insn, + (unsigned long)p->ainsn.insn + sizeof(kprobe_opcode_t)); + memcpy((char *)arm_addr, &p->opcode, sizeof(kprobe_opcode_t)); + flush_icache_range(arm_addr, arm_addr + sizeof(kprobe_opcode_t)); } void __kprobes arch_disarm_kprobe(struct kprobe *p) @@ -463,11 +467,18 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) unsigned long addr = (unsigned long)p->addr; unsigned long arm_addr = addr & ~0xFULL; - /* p->opcode contains the original unaltered bundle */ - memcpy((char *) arm_addr, (char *) &p->opcode.bundle, sizeof(bundle_t)); - flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t)); + /* p->ainsn.insn contains the original unaltered kprobe_opcode_t */ + memcpy((char *) arm_addr, (char *) p->ainsn.insn, + sizeof(kprobe_opcode_t)); + flush_icache_range(arm_addr, arm_addr + sizeof(kprobe_opcode_t)); } +void __kprobes arch_remove_kprobe(struct kprobe *p) +{ + mutex_lock(&kprobe_mutex); + free_insn_slot(p->ainsn.insn); + mutex_unlock(&kprobe_mutex); +} /* * We are resuming execution after a single step fault, so the pt_regs * structure reflects the register state after we executed the instruction @@ -478,12 +489,12 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) */ static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs) { - unsigned long bundle_addr = ((unsigned long) (&p->opcode.bundle)) & ~0xFULL; + unsigned long bundle_addr = (unsigned long) (&p->ainsn.insn->bundle); unsigned long resume_addr = (unsigned long)p->addr & ~0xFULL; unsigned long template; int slot = ((unsigned long)p->addr & 0xf); - template = p->opcode.bundle.quad0.template; + template = p->ainsn.insn->bundle.quad0.template; if (slot == 1 && bundle_encoding[template][1] == L) slot = 2; @@ -545,7 +556,7 @@ turn_ss_off: static void __kprobes prepare_ss(struct kprobe *p, struct pt_regs *regs) { - unsigned long bundle_addr = (unsigned long) &p->opcode.bundle; + unsigned long bundle_addr = (unsigned long) &p->ainsn.insn->bundle; unsigned long slot = (unsigned long)p->addr & 0xf; /* single step inline if break instruction */ @@ -760,6 +771,12 @@ static int __kprobes kprobes_fault_handler(struct pt_regs *regs, int trapnr) */ if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr)) return 1; + /* + * In case the user-specified fault handler returned + * zero, try to fix up. + */ + if (ia64_done_with_exception(regs)) + return 1; /* * Let ia64_do_page_fault() fix it. diff --git a/arch/ia64/kernel/machvec.c b/arch/ia64/kernel/machvec.c index 4b0b71d5aef..d4a546aa504 100644 --- a/arch/ia64/kernel/machvec.c +++ b/arch/ia64/kernel/machvec.c @@ -1,4 +1,3 @@ -#include <linux/config.h> #include <linux/module.h> #include <asm/machvec.h> diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 6a0880639bc..bfbd8986153 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -54,8 +54,10 @@ * * 2005-10-07 Keith Owens <kaos@sgi.com> * Add notify_die() hooks. + * + * 2006-09-15 Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com> + * Add printing support for MCA/INIT. */ -#include <linux/config.h> #include <linux/types.h> #include <linux/init.h> #include <linux/sched.h> @@ -137,11 +139,175 @@ extern void salinfo_log_wakeup(int type, u8 *buffer, u64 size, int irqsafe); static int mca_init __initdata; +/* + * limited & delayed printing support for MCA/INIT handler + */ + +#define mprintk(fmt...) ia64_mca_printk(fmt) + +#define MLOGBUF_SIZE (512+256*NR_CPUS) +#define MLOGBUF_MSGMAX 256 +static char mlogbuf[MLOGBUF_SIZE]; +static DEFINE_SPINLOCK(mlogbuf_wlock); /* mca context only */ +static DEFINE_SPINLOCK(mlogbuf_rlock); /* normal context only */ +static unsigned long mlogbuf_start; +static unsigned long mlogbuf_end; +static unsigned int mlogbuf_finished = 0; +static unsigned long mlogbuf_timestamp = 0; + +static int loglevel_save = -1; +#define BREAK_LOGLEVEL(__console_loglevel) \ + oops_in_progress = 1; \ + if (loglevel_save < 0) \ + loglevel_save = __console_loglevel; \ + __console_loglevel = 15; + +#define RESTORE_LOGLEVEL(__console_loglevel) \ + if (loglevel_save >= 0) { \ + __console_loglevel = loglevel_save; \ + loglevel_save = -1; \ + } \ + mlogbuf_finished = 0; \ + oops_in_progress = 0; + +/* + * Push messages into buffer, print them later if not urgent. + */ +void ia64_mca_printk(const char *fmt, ...) +{ + va_list args; + int printed_len; + char temp_buf[MLOGBUF_MSGMAX]; + char *p; + + va_start(args, fmt); + printed_len = vscnprintf(temp_buf, sizeof(temp_buf), fmt, args); + va_end(args); + + /* Copy the output into mlogbuf */ + if (oops_in_progress) { + /* mlogbuf was abandoned, use printk directly instead. */ + printk(temp_buf); + } else { + spin_lock(&mlogbuf_wlock); + for (p = temp_buf; *p; p++) { + unsigned long next = (mlogbuf_end + 1) % MLOGBUF_SIZE; + if (next != mlogbuf_start) { + mlogbuf[mlogbuf_end] = *p; + mlogbuf_end = next; + } else { + /* buffer full */ + break; + } + } + mlogbuf[mlogbuf_end] = '\0'; + spin_unlock(&mlogbuf_wlock); + } +} +EXPORT_SYMBOL(ia64_mca_printk); + +/* + * Print buffered messages. + * NOTE: call this after returning normal context. (ex. from salinfod) + */ +void ia64_mlogbuf_dump(void) +{ + char temp_buf[MLOGBUF_MSGMAX]; + char *p; + unsigned long index; + unsigned long flags; + unsigned int printed_len; + + /* Get output from mlogbuf */ + while (mlogbuf_start != mlogbuf_end) { + temp_buf[0] = '\0'; + p = temp_buf; + printed_len = 0; + + spin_lock_irqsave(&mlogbuf_rlock, flags); + + index = mlogbuf_start; + while (index != mlogbuf_end) { + *p = mlogbuf[index]; + index = (index + 1) % MLOGBUF_SIZE; + if (!*p) + break; + p++; + if (++printed_len >= MLOGBUF_MSGMAX - 1) + break; + } + *p = '\0'; + if (temp_buf[0]) + printk(temp_buf); + mlogbuf_start = index; + + mlogbuf_timestamp = 0; + spin_unlock_irqrestore(&mlogbuf_rlock, flags); + } +} +EXPORT_SYMBOL(ia64_mlogbuf_dump); + +/* + * Call this if system is going to down or if immediate flushing messages to + * console is required. (ex. recovery was failed, crash dump is going to be + * invoked, long-wait rendezvous etc.) + * NOTE: this should be called from monarch. + */ +static void ia64_mlogbuf_finish(int wait) +{ + BREAK_LOGLEVEL(console_loglevel); + + spin_lock_init(&mlogbuf_rlock); + ia64_mlogbuf_dump(); + printk(KERN_EMERG "mlogbuf_finish: printing switched to urgent mode, " + "MCA/INIT might be dodgy or fail.\n"); + + if (!wait) + return; + + /* wait for console */ + printk("Delaying for 5 seconds...\n"); + udelay(5*1000000); + + mlogbuf_finished = 1; +} +EXPORT_SYMBOL(ia64_mlogbuf_finish); + +/* + * Print buffered messages from INIT context. + */ +static void ia64_mlogbuf_dump_from_init(void) +{ + if (mlogbuf_finished) + return; + + if (mlogbuf_timestamp && (mlogbuf_timestamp + 30*HZ > jiffies)) { + printk(KERN_ERR "INIT: mlogbuf_dump is interrupted by INIT " + " and the system seems to be messed up.\n"); + ia64_mlogbuf_finish(0); + return; + } + + if (!spin_trylock(&mlogbuf_rlock)) { + printk(KERN_ERR "INIT: mlogbuf_dump is interrupted by INIT. " + "Generated messages other than stack dump will be " + "buffered to mlogbuf and will be printed later.\n"); + printk(KERN_ERR "INIT: If messages would not printed after " + "this INIT, wait 30sec and assert INIT again.\n"); + if (!mlogbuf_timestamp) + mlogbuf_timestamp = jiffies; + return; + } + spin_unlock(&mlogbuf_rlock); + ia64_mlogbuf_dump(); +} static void inline ia64_mca_spin(const char *func) { - printk(KERN_EMERG "%s: spinning here, not returning to SAL\n", func); + if (monarch_cpu == smp_processor_id()) + ia64_mlogbuf_finish(0); + mprintk(KERN_EMERG "%s: spinning here, not returning to SAL\n", func); while (1) cpu_relax(); } @@ -345,9 +511,6 @@ ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs) /* SAL spec states this should run w/ interrupts enabled */ local_irq_enable(); - /* Get the CPE error record and log it */ - ia64_mca_log_sal_error_record(SAL_INFO_TYPE_CPE); - spin_lock(&cpe_history_lock); if (!cpe_poll_enabled && cpe_vector >= 0) { @@ -376,7 +539,7 @@ ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs) mod_timer(&cpe_poll_timer, jiffies + MIN_CPE_POLL_INTERVAL); /* lock already released, get out now */ - return IRQ_HANDLED; + goto out; } else { cpe_history[index++] = now; if (index == CPE_HISTORY_LENGTH) @@ -384,6 +547,10 @@ ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs) } } spin_unlock(&cpe_history_lock); +out: + /* Get the CPE error record and log it */ + ia64_mca_log_sal_error_record(SAL_INFO_TYPE_CPE); + return IRQ_HANDLED; } @@ -679,7 +846,7 @@ copy_reg(const u64 *fr, u64 fnat, u64 *tr, u64 *tnat) */ static void -ia64_mca_modify_comm(const task_t *previous_current) +ia64_mca_modify_comm(const struct task_struct *previous_current) { char *p, comm[sizeof(current->comm)]; if (previous_current->pid) @@ -710,7 +877,7 @@ ia64_mca_modify_comm(const task_t *previous_current) * that we can do backtrace on the MCA/INIT handler code itself. */ -static task_t * +static struct task_struct * ia64_mca_modify_original_stack(struct pt_regs *regs, const struct switch_stack *sw, struct ia64_sal_os_state *sos, @@ -720,7 +887,7 @@ ia64_mca_modify_original_stack(struct pt_regs *regs, ia64_va va; extern char ia64_leave_kernel[]; /* Need asm address, not function descriptor */ const pal_min_state_area_t *ms = sos->pal_min_state; - task_t *previous_current; + struct task_struct *previous_current; struct pt_regs *old_regs; struct switch_stack *old_sw; unsigned size = sizeof(struct pt_regs) + @@ -989,18 +1156,22 @@ ia64_wait_for_slaves(int monarch, const char *type) } if (!missing) goto all_in; - printk(KERN_INFO "OS %s slave did not rendezvous on cpu", type); + /* + * Maybe slave(s) dead. Print buffered messages immediately. + */ + ia64_mlogbuf_finish(0); + mprintk(KERN_INFO "OS %s slave did not rendezvous on cpu", type); for_each_online_cpu(c) { if (c == monarch) continue; if (ia64_mc_info.imi_rendez_checkin[c] == IA64_MCA_RENDEZ_CHECKIN_NOTDONE) - printk(" %d", c); + mprintk(" %d", c); } - printk("\n"); + mprintk("\n"); return; all_in: - printk(KERN_INFO "All OS %s slaves have reached rendezvous\n", type); + mprintk(KERN_INFO "All OS %s slaves have reached rendezvous\n", type); return; } @@ -1024,14 +1195,12 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, pal_processor_state_info_t *psp = (pal_processor_state_info_t *) &sos->proc_state_param; int recover, cpu = smp_processor_id(); - task_t *previous_current; + struct task_struct *previous_current; struct ia64_mca_notify_die nd = { .sos = sos, .monarch_cpu = &monarch_cpu }; - oops_in_progress = 1; /* FIXME: make printk NMI/MCA/INIT safe */ - console_loglevel = 15; /* make sure printks make it to console */ - printk(KERN_INFO "Entered OS MCA handler. PSP=%lx cpu=%d monarch=%ld\n", - sos->proc_state_param, cpu, sos->monarch); + mprintk(KERN_INFO "Entered OS MCA handler. PSP=%lx cpu=%d " + "monarch=%ld\n", sos->proc_state_param, cpu, sos->monarch); previous_current = ia64_mca_modify_original_stack(regs, sw, sos, "MCA"); monarch_cpu = cpu; @@ -1067,6 +1236,9 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, rh->severity = sal_log_severity_corrected; ia64_sal_clear_state_info(SAL_INFO_TYPE_MCA); sos->os_status = IA64_MCA_CORRECTED; + } else { + /* Dump buffered message to console */ + ia64_mlogbuf_finish(1); } if (notify_die(DIE_MCA_MONARCH_LEAVE, "MCA", regs, (long)&nd, 0, recover) == NOTIFY_STOP) @@ -1107,9 +1279,6 @@ ia64_mca_cmc_int_handler(int cmc_irq, void *arg, struct pt_regs *ptregs) /* SAL spec states this should run w/ interrupts enabled */ local_irq_enable(); - /* Get the CMC error record and log it */ - ia64_mca_log_sal_error_record(SAL_INFO_TYPE_CMC); - spin_lock(&cmc_history_lock); if (!cmc_polling_enabled) { int i, count = 1; /* we know 1 happened now */ @@ -1142,7 +1311,7 @@ ia64_mca_cmc_int_handler(int cmc_irq, void *arg, struct pt_regs *ptregs) mod_timer(&cmc_poll_timer, jiffies + CMC_POLL_INTERVAL); /* lock already released, get out now */ - return IRQ_HANDLED; + goto out; } else { cmc_history[index++] = now; if (index == CMC_HISTORY_LENGTH) @@ -1150,6 +1319,10 @@ ia64_mca_cmc_int_handler(int cmc_irq, void *arg, struct pt_regs *ptregs) } } spin_unlock(&cmc_history_lock); +out: + /* Get the CMC error record and log it */ + ia64_mca_log_sal_error_record(SAL_INFO_TYPE_CMC); + return IRQ_HANDLED; } @@ -1306,6 +1479,15 @@ default_monarch_init_process(struct notifier_block *self, unsigned long val, voi struct task_struct *g, *t; if (val != DIE_INIT_MONARCH_PROCESS) return NOTIFY_DONE; + + /* + * FIXME: mlogbuf will brim over with INIT stack dumps. + * To enable show_stack from INIT, we use oops_in_progress which should + * be used in real oops. This would cause something wrong after INIT. + */ + BREAK_LOGLEVEL(console_loglevel); + ia64_mlogbuf_dump_from_init(); + printk(KERN_ERR "Processes interrupted by INIT -"); for_each_online_cpu(c) { struct ia64_sal_os_state *s; @@ -1327,6 +1509,8 @@ default_monarch_init_process(struct notifier_block *self, unsigned long val, voi } while_each_thread (g, t); read_unlock(&tasklist_lock); } + /* FIXME: This will not restore zapped printk locks. */ + RESTORE_LOGLEVEL(console_loglevel); return NOTIFY_DONE; } @@ -1353,17 +1537,14 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, { static atomic_t slaves; static atomic_t monarchs; - task_t *previous_current; + struct task_struct *previous_current; int cpu = smp_processor_id(); struct ia64_mca_notify_die nd = { .sos = sos, .monarch_cpu = &monarch_cpu }; - oops_in_progress = 1; /* FIXME: make printk NMI/MCA/INIT safe */ - console_loglevel = 15; /* make sure printks make it to console */ - (void) notify_die(DIE_INIT_ENTER, "INIT", regs, (long)&nd, 0, 0); - printk(KERN_INFO "Entered OS INIT handler. PSP=%lx cpu=%d monarch=%ld\n", + mprintk(KERN_INFO "Entered OS INIT handler. PSP=%lx cpu=%d monarch=%ld\n", sos->proc_state_param, cpu, sos->monarch); salinfo_log_wakeup(SAL_INFO_TYPE_INIT, NULL, 0, 0); @@ -1376,7 +1557,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, * fix their proms and get their customers updated. */ if (!sos->monarch && atomic_add_return(1, &slaves) == num_online_cpus()) { - printk(KERN_WARNING "%s: Promoting cpu %d to monarch.\n", + mprintk(KERN_WARNING "%s: Promoting cpu %d to monarch.\n", __FUNCTION__, cpu); atomic_dec(&slaves); sos->monarch = 1; @@ -1388,7 +1569,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, * fix their proms and get their customers updated. */ if (sos->monarch && atomic_add_return(1, &monarchs) > 1) { - printk(KERN_WARNING "%s: Demoting cpu %d to slave.\n", + mprintk(KERN_WARNING "%s: Demoting cpu %d to slave.\n", __FUNCTION__, cpu); atomic_dec(&monarchs); sos->monarch = 0; @@ -1409,7 +1590,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, if (notify_die(DIE_INIT_SLAVE_LEAVE, "INIT", regs, (long)&nd, 0, 0) == NOTIFY_STOP) ia64_mca_spin(__FUNCTION__); - printk("Slave on cpu %d returning to normal service.\n", cpu); + mprintk("Slave on cpu %d returning to normal service.\n", cpu); set_curr_task(cpu, previous_current); ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_NOTDONE; atomic_dec(&slaves); @@ -1427,7 +1608,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, * same serial line, the user will need some time to switch out of the BMC before * the dump begins. */ - printk("Delaying for 5 seconds...\n"); + mprintk("Delaying for 5 seconds...\n"); udelay(5*1000000); ia64_wait_for_slaves(cpu, "INIT"); /* If nobody intercepts DIE_INIT_MONARCH_PROCESS then we drop through @@ -1440,7 +1621,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, if (notify_die(DIE_INIT_MONARCH_LEAVE, "INIT", regs, (long)&nd, 0, 0) == NOTIFY_STOP) ia64_mca_spin(__FUNCTION__); - printk("\nINIT dump complete. Monarch on cpu %d returning to normal service.\n", cpu); + mprintk("\nINIT dump complete. Monarch on cpu %d returning to normal service.\n", cpu); atomic_dec(&monarchs); set_curr_task(cpu, previous_current); monarch_cpu = -1; @@ -1458,38 +1639,38 @@ __setup("disable_cpe_poll", ia64_mca_disable_cpe_polling); static struct irqaction cmci_irqaction = { .handler = ia64_mca_cmc_int_handler, - .flags = SA_INTERRUPT, + .flags = IRQF_DISABLED, .name = "cmc_hndlr" }; static struct irqaction cmcp_irqaction = { .handler = ia64_mca_cmc_int_caller, - .flags = SA_INTERRUPT, + .flags = IRQF_DISABLED, .name = "cmc_poll" }; static struct irqaction mca_rdzv_irqaction = { .handler = ia64_mca_rendez_int_handler, - .flags = SA_INTERRUPT, + .flags = IRQF_DISABLED, .name = "mca_rdzv" }; static struct irqaction mca_wkup_irqaction = { .handler = ia64_mca_wakeup_int_handler, - .flags = SA_INTERRUPT, + .flags = IRQF_DISABLED, .name = "mca_wkup" }; #ifdef CONFIG_ACPI static struct irqaction mca_cpe_irqaction = { .handler = ia64_mca_cpe_int_handler, - .flags = SA_INTERRUPT, + .flags = IRQF_DISABLED, .name = "cpe_hndlr" }; static struct irqaction mca_cpep_irqaction = { .handler = ia64_mca_cpe_int_caller, - .flags = SA_INTERRUPT, + .flags = IRQF_DISABLED, .name = "cpe_poll" }; #endif /* CONFIG_ACPI */ @@ -1788,7 +1969,7 @@ ia64_mca_late_init(void) cpe_poll_enabled = 0; for (irq = 0; irq < NR_IRQS; ++irq) if (irq_to_vector(irq) == cpe_vector) { - desc = irq_descp(irq); + desc = irq_desc + irq; desc->status |= IRQ_PER_CPU; setup_irq(irq, &mca_cpe_irqaction); ia64_cpe_irq = irq; diff --git a/arch/ia64/kernel/mca_asm.S b/arch/ia64/kernel/mca_asm.S index c1bd1feffab..c6b607c00de 100644 --- a/arch/ia64/kernel/mca_asm.S +++ b/arch/ia64/kernel/mca_asm.S @@ -19,7 +19,6 @@ // 12/08/05 Keith Owens <kaos@sgi.com> // Use per cpu MCA/INIT stacks for all data. // -#include <linux/config.h> #include <linux/threads.h> #include <asm/asmmacro.h> @@ -1026,18 +1025,13 @@ ia64_old_stack: ia64_set_kernel_registers: add temp3=MCA_SP_OFFSET, r3 - add temp4=MCA_SOS_OFFSET+SOS(OS_GP), r3 mov b0=r2 // save return address GET_IA64_MCA_DATA(temp1) ;; - add temp4=temp4, temp1 // &struct ia64_sal_os_state.os_gp add r12=temp1, temp3 // kernel stack pointer on MCA/INIT stack add r13=temp1, r3 // set current to start of MCA/INIT stack add r20=temp1, r3 // physical start of MCA/INIT stack ;; - ld8 r1=[temp4] // OS GP from SAL OS state - ;; - DATA_PA_TO_VA(r1,temp1) DATA_PA_TO_VA(r12,temp2) DATA_PA_TO_VA(r13,temp3) ;; @@ -1068,6 +1062,10 @@ ia64_set_kernel_registers: mov cr.itir=r18 mov cr.ifa=r13 mov r20=IA64_TR_CURRENT_STACK + + movl r17=FPSR_DEFAULT + ;; + mov.m ar.fpsr=r17 // set ar.fpsr to kernel default value ;; itr.d dtr[r20]=r21 ;; diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c index ca6666b51cc..a45009d2bc9 100644 --- a/arch/ia64/kernel/mca_drv.c +++ b/arch/ia64/kernel/mca_drv.c @@ -8,7 +8,6 @@ * Copyright (C) 2005 Keith Owens <kaos@sgi.com> * Copyright (C) 2006 Russ Anderson <rja@sgi.com> */ -#include <linux/config.h> #include <linux/types.h> #include <linux/init.h> #include <linux/sched.h> @@ -80,14 +79,30 @@ static int fatal_mca(const char *fmt, ...) { va_list args; + char buf[256]; va_start(args, fmt); - vprintk(fmt, args); + vsnprintf(buf, sizeof(buf), fmt, args); va_end(args); + ia64_mca_printk(KERN_ALERT "MCA: %s\n", buf); return MCA_NOT_RECOVERED; } +static int +mca_recovered(const char *fmt, ...) +{ + va_list args; + char buf[256]; + + va_start(args, fmt); + vsnprintf(buf, sizeof(buf), fmt, args); + va_end(args); + ia64_mca_printk(KERN_INFO "MCA: %s\n", buf); + + return MCA_RECOVERED; +} + /** * mca_page_isolate - isolate a poisoned page in order not to use it later * @paddr: poisoned memory location @@ -141,6 +156,7 @@ mca_page_isolate(unsigned long paddr) void mca_handler_bh(unsigned long paddr, void *iip, unsigned long ipsr) { + ia64_mlogbuf_dump(); printk(KERN_ERR "OS_MCA: process [cpu %d, pid: %d, uid: %d, " "iip: %p, psr: 0x%lx,paddr: 0x%lx](%s) encounters MCA.\n", raw_smp_processor_id(), current->pid, current->uid, @@ -441,7 +457,7 @@ recover_from_read_error(slidx_table_t *slidx, /* Is target address valid? */ if (!pbci->tv) - return fatal_mca(KERN_ALERT "MCA: target address not valid\n"); + return fatal_mca("target address not valid"); /* * cpu read or memory-mapped io read @@ -459,7 +475,7 @@ recover_from_read_error(slidx_table_t *slidx, /* Is minstate valid? */ if (!peidx_bottom(peidx) || !(peidx_bottom(peidx)->valid.minstate)) - return fatal_mca(KERN_ALERT "MCA: minstate not valid\n"); + return fatal_mca("minstate not valid"); psr1 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_ipsr); psr2 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_xpsr); @@ -493,13 +509,14 @@ recover_from_read_error(slidx_table_t *slidx, psr2->bn = 1; psr2->i = 0; - return MCA_RECOVERED; + return mca_recovered("user memory corruption. " + "kill affected process - recovered."); } } - return fatal_mca(KERN_ALERT "MCA: kernel context not recovered," - " iip 0x%lx\n", pmsa->pmsa_iip); + return fatal_mca("kernel context not recovered, iip 0x%lx\n", + pmsa->pmsa_iip); } /** @@ -585,13 +602,13 @@ recover_from_processor_error(int platform, slidx_table_t *slidx, * The machine check is corrected. */ if (psp->cm == 1) - return MCA_RECOVERED; + return mca_recovered("machine check is already corrected."); /* * The error was not contained. Software must be reset. */ if (psp->us || psp->ci == 0) - return fatal_mca(KERN_ALERT "MCA: error not contained\n"); + return fatal_mca("error not contained"); /* * The cache check and bus check bits have four possible states @@ -602,22 +619,22 @@ recover_from_processor_error(int platform, slidx_table_t *slidx, * 1 1 Memory error, attempt recovery */ if (psp->bc == 0 || pbci == NULL) - return fatal_mca(KERN_ALERT "MCA: No bus check\n"); + return fatal_mca("No bus check"); /* * Sorry, we cannot handle so many. */ if (peidx_bus_check_num(peidx) > 1) - return fatal_mca(KERN_ALERT "MCA: Too many bus checks\n"); + return fatal_mca("Too many bus checks"); /* * Well, here is only one bus error. */ if (pbci->ib) - return fatal_mca(KERN_ALERT "MCA: Internal Bus error\n"); + return fatal_mca("Internal Bus error"); if (pbci->cc) - return fatal_mca(KERN_ALERT "MCA: Cache-cache error\n"); + return fatal_mca("Cache-cache error"); if (pbci->eb && pbci->bsi > 0) - return fatal_mca(KERN_ALERT "MCA: External bus check fatal status\n"); + return fatal_mca("External bus check fatal status"); /* * This is a local MCA and estimated as recoverble external bus error. @@ -629,7 +646,7 @@ recover_from_processor_error(int platform, slidx_table_t *slidx, /* * On account of strange SAL error record, we cannot recover. */ - return fatal_mca(KERN_ALERT "MCA: Strange SAL record\n"); + return fatal_mca("Strange SAL record"); } /** @@ -658,10 +675,10 @@ mca_try_to_recover(void *rec, struct ia64_sal_os_state *sos) /* Now, OS can recover when there is one processor error section */ if (n_proc_err > 1) - return fatal_mca(KERN_ALERT "MCA: Too Many Errors\n"); + return fatal_mca("Too Many Errors"); else if (n_proc_err == 0) - /* Weird SAL record ... We need not to recover */ - return fatal_mca(KERN_ALERT "MCA: Weird SAL record\n"); + /* Weird SAL record ... We can't do anything */ + return fatal_mca("Weird SAL record"); /* Make index of processor error section */ mca_make_peidx((sal_log_processor_info_t*) @@ -672,7 +689,7 @@ mca_try_to_recover(void *rec, struct ia64_sal_os_state *sos) /* Check whether MCA is global or not */ if (is_mca_global(&peidx, &pbci, sos)) - return fatal_mca(KERN_ALERT "MCA: global MCA\n"); + return fatal_mca("global MCA"); /* Try to recover a processor error */ return recover_from_processor_error(platform_err, &slidx, &peidx, diff --git a/arch/ia64/kernel/mca_drv.h b/arch/ia64/kernel/mca_drv.h index 31a2e52bb16..c85e943ba5f 100644 --- a/arch/ia64/kernel/mca_drv.h +++ b/arch/ia64/kernel/mca_drv.h @@ -118,3 +118,7 @@ struct mca_table_entry { extern const struct mca_table_entry *search_mca_tables (unsigned long addr); extern int mca_recover_range(unsigned long); +extern void ia64_mca_printk(const char * fmt, ...) + __attribute__ ((format (printf, 1, 2))); +extern void ia64_mlogbuf_dump(void); + diff --git a/arch/ia64/kernel/mca_drv_asm.S b/arch/ia64/kernel/mca_drv_asm.S index e6a580d354b..f2d4900751b 100644 --- a/arch/ia64/kernel/mca_drv_asm.S +++ b/arch/ia64/kernel/mca_drv_asm.S @@ -5,7 +5,6 @@ * Copyright (C) 2004 FUJITSU LIMITED * Copyright (C) Hidetoshi Seto (seto.hidetoshi@jp.fujitsu.com) */ -#include <linux/config.h> #include <linux/threads.h> #include <asm/asmmacro.h> diff --git a/arch/ia64/kernel/minstate.h b/arch/ia64/kernel/minstate.h index 85ed54179af..c9ac8bada78 100644 --- a/arch/ia64/kernel/minstate.h +++ b/arch/ia64/kernel/minstate.h @@ -1,4 +1,3 @@ -#include <linux/config.h> #include <asm/cache.h> diff --git a/arch/ia64/kernel/module.c b/arch/ia64/kernel/module.c index 3a30cfc9574..158e3c51bb7 100644 --- a/arch/ia64/kernel/module.c +++ b/arch/ia64/kernel/module.c @@ -25,7 +25,6 @@ SEGREL64LSB */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/sched.h> diff --git a/arch/ia64/kernel/numa.c b/arch/ia64/kernel/numa.c index 0766493d4d0..20340631179 100644 --- a/arch/ia64/kernel/numa.c +++ b/arch/ia64/kernel/numa.c @@ -19,7 +19,6 @@ * Copyright (C) 2004 Silicon Graphics, Inc. * Jesse Barnes <jbarnes@sgi.com> */ -#include <linux/config.h> #include <linux/topology.h> #include <linux/module.h> #include <asm/processor.h> @@ -30,6 +29,36 @@ EXPORT_SYMBOL(cpu_to_node_map); cpumask_t node_to_cpu_mask[MAX_NUMNODES] __cacheline_aligned; +void __cpuinit map_cpu_to_node(int cpu, int nid) +{ + int oldnid; + if (nid < 0) { /* just initialize by zero */ + cpu_to_node_map[cpu] = 0; + return; + } + /* sanity check first */ + oldnid = cpu_to_node_map[cpu]; + if (cpu_isset(cpu, node_to_cpu_mask[oldnid])) { + return; /* nothing to do */ + } + /* we don't have cpu-driven node hot add yet... + In usual case, node is created from SRAT at boot time. */ + if (!node_online(nid)) + nid = first_online_node; + cpu_to_node_map[cpu] = nid; + cpu_set(cpu, node_to_cpu_mask[nid]); + return; +} + +void __cpuinit unmap_cpu_from_node(int cpu, int nid) +{ + WARN_ON(!cpu_isset(cpu, node_to_cpu_mask[nid])); + WARN_ON(cpu_to_node_map[cpu] != nid); + cpu_to_node_map[cpu] = 0; + cpu_clear(cpu, node_to_cpu_mask[nid]); +} + + /** * build_cpu_to_node_map - setup cpu to node and node to cpumask arrays * @@ -50,8 +79,6 @@ void __init build_cpu_to_node_map(void) node = node_cpuid[i].nid; break; } - cpu_to_node_map[cpu] = (node >= 0) ? node : 0; - if (node >= 0) - cpu_set(cpu, node_to_cpu_mask[node]); + map_cpu_to_node(cpu, node); } } diff --git a/arch/ia64/kernel/pal.S b/arch/ia64/kernel/pal.S index 5018c7f2e7a..ebaf1e685f5 100644 --- a/arch/ia64/kernel/pal.S +++ b/arch/ia64/kernel/pal.S @@ -217,12 +217,7 @@ GLOBAL_ENTRY(ia64_pal_call_phys_stacked) .body ;; ld8 loc2 = [loc2] // loc2 <- entry point - mov out0 = in0 // first argument - mov out1 = in1 // copy arg2 - mov out2 = in2 // copy arg3 - mov out3 = in3 // copy arg3 - ;; - mov loc3 = psr // save psr + mov loc3 = psr // save psr ;; mov loc4=ar.rsc // save RSE configuration dep.z loc2=loc2,0,61 // convert pal entry point to physical @@ -236,18 +231,23 @@ GLOBAL_ENTRY(ia64_pal_call_phys_stacked) ;; andcm r16=loc3,r16 // removes bits to clear from psr br.call.sptk.many rp=ia64_switch_mode_phys -.ret6: + + mov out0 = in0 // first argument + mov out1 = in1 // copy arg2 + mov out2 = in2 // copy arg3 + mov out3 = in3 // copy arg3 mov loc5 = r19 mov loc6 = r20 + br.call.sptk.many rp=b7 // now make the call -.ret7: + mov ar.rsc=0 // put RSE in enforced lazy, LE mode mov r16=loc3 // r16= original psr mov r19=loc5 mov r20=loc6 br.call.sptk.many rp=ia64_switch_mode_virt // return to virtual mode -.ret8: mov psr.l = loc3 // restore init PSR + mov psr.l = loc3 // restore init PSR mov ar.pfs = loc1 mov rp = loc0 ;; diff --git a/arch/ia64/kernel/palinfo.c b/arch/ia64/kernel/palinfo.c index 859fb37ff49..0b546e2b36a 100644 --- a/arch/ia64/kernel/palinfo.c +++ b/arch/ia64/kernel/palinfo.c @@ -17,7 +17,6 @@ * 10/23/2001 S.Eranian updated pal_perf_mon_info bug fixes * 03/24/2004 Ashok Raj updated to work with CPU Hotplug */ -#include <linux/config.h> #include <linux/types.h> #include <linux/errno.h> #include <linux/init.h> @@ -567,29 +566,23 @@ version_info(char *page) pal_version_u_t min_ver, cur_ver; char *p = page; - /* The PAL_VERSION call is advertised as being able to support - * both physical and virtual mode calls. This seems to be a documentation - * bug rather than firmware bug. In fact, it does only support physical mode. - * So now the code reflects this fact and the pal_version() has been updated - * accordingly. - */ - if (ia64_pal_version(&min_ver, &cur_ver) != 0) return 0; + if (ia64_pal_version(&min_ver, &cur_ver) != 0) + return 0; p += sprintf(p, "PAL_vendor : 0x%02x (min=0x%02x)\n" - "PAL_A : %x.%x.%x (min=%x.%x.%x)\n" - "PAL_B : %x.%x.%x (min=%x.%x.%x)\n", - cur_ver.pal_version_s.pv_pal_vendor, min_ver.pal_version_s.pv_pal_vendor, - - cur_ver.pal_version_s.pv_pal_a_model>>4, - cur_ver.pal_version_s.pv_pal_a_model&0xf, cur_ver.pal_version_s.pv_pal_a_rev, - min_ver.pal_version_s.pv_pal_a_model>>4, - min_ver.pal_version_s.pv_pal_a_model&0xf, min_ver.pal_version_s.pv_pal_a_rev, - - cur_ver.pal_version_s.pv_pal_b_model>>4, - cur_ver.pal_version_s.pv_pal_b_model&0xf, cur_ver.pal_version_s.pv_pal_b_rev, - min_ver.pal_version_s.pv_pal_b_model>>4, - min_ver.pal_version_s.pv_pal_b_model&0xf, min_ver.pal_version_s.pv_pal_b_rev); + "PAL_A : %02x.%02x (min=%02x.%02x)\n" + "PAL_B : %02x.%02x (min=%02x.%02x)\n", + cur_ver.pal_version_s.pv_pal_vendor, + min_ver.pal_version_s.pv_pal_vendor, + cur_ver.pal_version_s.pv_pal_a_model, + cur_ver.pal_version_s.pv_pal_a_rev, + min_ver.pal_version_s.pv_pal_a_model, + min_ver.pal_version_s.pv_pal_a_rev, + cur_ver.pal_version_s.pv_pal_b_model, + cur_ver.pal_version_s.pv_pal_b_rev, + min_ver.pal_version_s.pv_pal_b_model, + min_ver.pal_version_s.pv_pal_b_rev); return p - page; } @@ -959,9 +952,9 @@ remove_palinfo_proc_entries(unsigned int hcpu) } } +#ifdef CONFIG_HOTPLUG_CPU static int palinfo_cpu_callback(struct notifier_block *nfb, - unsigned long action, - void *hcpu) + unsigned long action, void *hcpu) { unsigned int hotcpu = (unsigned long)hcpu; @@ -969,11 +962,9 @@ static int palinfo_cpu_callback(struct notifier_block *nfb, case CPU_ONLINE: create_palinfo_proc_entries(hotcpu); break; -#ifdef CONFIG_HOTPLUG_CPU case CPU_DEAD: remove_palinfo_proc_entries(hotcpu); break; -#endif } return NOTIFY_OK; } @@ -983,6 +974,7 @@ static struct notifier_block palinfo_cpu_notifier = .notifier_call = palinfo_cpu_callback, .priority = 0, }; +#endif static int __init palinfo_init(void) @@ -998,7 +990,7 @@ palinfo_init(void) } /* Register for future delivery via notify registration */ - register_cpu_notifier(&palinfo_cpu_notifier); + register_hotcpu_notifier(&palinfo_cpu_notifier); return 0; } @@ -1021,7 +1013,7 @@ palinfo_exit(void) /* * Unregister from cpu notifier callbacks */ - unregister_cpu_notifier(&palinfo_cpu_notifier); + unregister_hotcpu_notifier(&palinfo_cpu_notifier); } module_init(palinfo_init); diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 6d7bc8ff7b3..281004ff7b0 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -19,7 +19,6 @@ * http://www.hpl.hp.com/research/linux/perfmon */ -#include <linux/config.h> #include <linux/module.h> #include <linux/kernel.h> #include <linux/sched.h> @@ -35,6 +34,7 @@ #include <linux/file.h> #include <linux/poll.h> #include <linux/vfs.h> +#include <linux/smp.h> #include <linux/pagemap.h> #include <linux/mount.h> #include <linux/bitops.h> @@ -63,6 +63,9 @@ #define PFM_INVALID_ACTIVATION (~0UL) +#define PFM_NUM_PMC_REGS 64 /* PMC save area for ctxsw */ +#define PFM_NUM_PMD_REGS 64 /* PMD save area for ctxsw */ + /* * depth of message queue */ @@ -297,14 +300,17 @@ typedef struct pfm_context { unsigned long ctx_reload_pmcs[4]; /* bitmask of force reload PMC on ctxsw in */ unsigned long ctx_used_monitors[4]; /* bitmask of monitor PMC being used */ - unsigned long ctx_pmcs[IA64_NUM_PMC_REGS]; /* saved copies of PMC values */ + unsigned long ctx_pmcs[PFM_NUM_PMC_REGS]; /* saved copies of PMC values */ unsigned int ctx_used_ibrs[1]; /* bitmask of used IBR (speedup ctxsw in) */ unsigned int ctx_used_dbrs[1]; /* bitmask of used DBR (speedup ctxsw in) */ unsigned long ctx_dbrs[IA64_NUM_DBG_REGS]; /* DBR values (cache) when not loaded */ unsigned long ctx_ibrs[IA64_NUM_DBG_REGS]; /* IBR values (cache) when not loaded */ - pfm_counter_t ctx_pmds[IA64_NUM_PMD_REGS]; /* software state for PMDS */ + pfm_counter_t ctx_pmds[PFM_NUM_PMD_REGS]; /* software state for PMDS */ + + unsigned long th_pmcs[PFM_NUM_PMC_REGS]; /* PMC thread save state */ + unsigned long th_pmds[PFM_NUM_PMD_REGS]; /* PMD thread save state */ u64 ctx_saved_psr_up; /* only contains psr.up value */ @@ -868,7 +874,6 @@ static void pfm_mask_monitoring(struct task_struct *task) { pfm_context_t *ctx = PFM_GET_CTX(task); - struct thread_struct *th = &task->thread; unsigned long mask, val, ovfl_mask; int i; @@ -889,7 +894,7 @@ pfm_mask_monitoring(struct task_struct *task) * So in both cases, the live register contains the owner's * state. We can ONLY touch the PMU registers and NOT the PSR. * - * As a consequence to this call, the thread->pmds[] array + * As a consequence to this call, the ctx->th_pmds[] array * contains stale information which must be ignored * when context is reloaded AND monitoring is active (see * pfm_restart). @@ -924,9 +929,9 @@ pfm_mask_monitoring(struct task_struct *task) mask = ctx->ctx_used_monitors[0] >> PMU_FIRST_COUNTER; for(i= PMU_FIRST_COUNTER; mask; i++, mask>>=1) { if ((mask & 0x1) == 0UL) continue; - ia64_set_pmc(i, th->pmcs[i] & ~0xfUL); - th->pmcs[i] &= ~0xfUL; - DPRINT_ovfl(("pmc[%d]=0x%lx\n", i, th->pmcs[i])); + ia64_set_pmc(i, ctx->th_pmcs[i] & ~0xfUL); + ctx->th_pmcs[i] &= ~0xfUL; + DPRINT_ovfl(("pmc[%d]=0x%lx\n", i, ctx->th_pmcs[i])); } /* * make all of this visible @@ -943,7 +948,6 @@ static void pfm_restore_monitoring(struct task_struct *task) { pfm_context_t *ctx = PFM_GET_CTX(task); - struct thread_struct *th = &task->thread; unsigned long mask, ovfl_mask; unsigned long psr, val; int i, is_system; @@ -1009,9 +1013,9 @@ pfm_restore_monitoring(struct task_struct *task) mask = ctx->ctx_used_monitors[0] >> PMU_FIRST_COUNTER; for(i= PMU_FIRST_COUNTER; mask; i++, mask>>=1) { if ((mask & 0x1) == 0UL) continue; - th->pmcs[i] = ctx->ctx_pmcs[i]; - ia64_set_pmc(i, th->pmcs[i]); - DPRINT(("[%d] pmc[%d]=0x%lx\n", task->pid, i, th->pmcs[i])); + ctx->th_pmcs[i] = ctx->ctx_pmcs[i]; + ia64_set_pmc(i, ctx->th_pmcs[i]); + DPRINT(("[%d] pmc[%d]=0x%lx\n", task->pid, i, ctx->th_pmcs[i])); } ia64_srlz_d(); @@ -1070,7 +1074,6 @@ pfm_restore_pmds(unsigned long *pmds, unsigned long mask) static inline void pfm_copy_pmds(struct task_struct *task, pfm_context_t *ctx) { - struct thread_struct *thread = &task->thread; unsigned long ovfl_val = pmu_conf->ovfl_val; unsigned long mask = ctx->ctx_all_pmds[0]; unsigned long val; @@ -1092,11 +1095,11 @@ pfm_copy_pmds(struct task_struct *task, pfm_context_t *ctx) ctx->ctx_pmds[i].val = val & ~ovfl_val; val &= ovfl_val; } - thread->pmds[i] = val; + ctx->th_pmds[i] = val; DPRINT(("pmd[%d]=0x%lx soft_val=0x%lx\n", i, - thread->pmds[i], + ctx->th_pmds[i], ctx->ctx_pmds[i].val)); } } @@ -1107,7 +1110,6 @@ pfm_copy_pmds(struct task_struct *task, pfm_context_t *ctx) static inline void pfm_copy_pmcs(struct task_struct *task, pfm_context_t *ctx) { - struct thread_struct *thread = &task->thread; unsigned long mask = ctx->ctx_all_pmcs[0]; int i; @@ -1115,8 +1117,8 @@ pfm_copy_pmcs(struct task_struct *task, pfm_context_t *ctx) for (i=0; mask; i++, mask>>=1) { /* masking 0 with ovfl_val yields 0 */ - thread->pmcs[i] = ctx->ctx_pmcs[i]; - DPRINT(("pmc[%d]=0x%lx\n", i, thread->pmcs[i])); + ctx->th_pmcs[i] = ctx->ctx_pmcs[i]; + DPRINT(("pmc[%d]=0x%lx\n", i, ctx->th_pmcs[i])); } } @@ -2860,7 +2862,6 @@ pfm_reset_regs(pfm_context_t *ctx, unsigned long *ovfl_regs, int is_long_reset) static int pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) { - struct thread_struct *thread = NULL; struct task_struct *task; pfarg_reg_t *req = (pfarg_reg_t *)arg; unsigned long value, pmc_pm; @@ -2881,7 +2882,6 @@ pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) if (state == PFM_CTX_ZOMBIE) return -EINVAL; if (is_loaded) { - thread = &task->thread; /* * In system wide and when the context is loaded, access can only happen * when the caller is running on the CPU being monitored by the session. @@ -3036,7 +3036,7 @@ pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) * * The value in ctx_pmcs[] can only be changed in pfm_write_pmcs(). * - * The value in thread->pmcs[] may be modified on overflow, i.e., when + * The value in th_pmcs[] may be modified on overflow, i.e., when * monitoring needs to be stopped. */ if (is_monitor) CTX_USED_MONITOR(ctx, 1UL << cnum); @@ -3050,7 +3050,7 @@ pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) /* * write thread state */ - if (is_system == 0) thread->pmcs[cnum] = value; + if (is_system == 0) ctx->th_pmcs[cnum] = value; /* * write hardware register if we can @@ -3102,7 +3102,6 @@ error: static int pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) { - struct thread_struct *thread = NULL; struct task_struct *task; pfarg_reg_t *req = (pfarg_reg_t *)arg; unsigned long value, hw_value, ovfl_mask; @@ -3126,7 +3125,6 @@ pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) * the owner of the local PMU. */ if (likely(is_loaded)) { - thread = &task->thread; /* * In system wide and when the context is loaded, access can only happen * when the caller is running on the CPU being monitored by the session. @@ -3234,7 +3232,7 @@ pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) /* * write thread state */ - if (is_system == 0) thread->pmds[cnum] = hw_value; + if (is_system == 0) ctx->th_pmds[cnum] = hw_value; /* * write hardware register if we can @@ -3300,7 +3298,6 @@ abort_mission: static int pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) { - struct thread_struct *thread = NULL; struct task_struct *task; unsigned long val = 0UL, lval, ovfl_mask, sval; pfarg_reg_t *req = (pfarg_reg_t *)arg; @@ -3324,7 +3321,6 @@ pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) if (state == PFM_CTX_ZOMBIE) return -EINVAL; if (likely(is_loaded)) { - thread = &task->thread; /* * In system wide and when the context is loaded, access can only happen * when the caller is running on the CPU being monitored by the session. @@ -3386,7 +3382,7 @@ pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) * if context is zombie, then task does not exist anymore. * In this case, we use the full value saved in the context (pfm_flush_regs()). */ - val = is_loaded ? thread->pmds[cnum] : 0UL; + val = is_loaded ? ctx->th_pmds[cnum] : 0UL; } rd_func = pmu_conf->pmd_desc[cnum].read_check; @@ -4355,8 +4351,8 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) pfm_copy_pmds(task, ctx); pfm_copy_pmcs(task, ctx); - pmcs_source = thread->pmcs; - pmds_source = thread->pmds; + pmcs_source = ctx->th_pmcs; + pmds_source = ctx->th_pmds; /* * always the case for system-wide @@ -4937,13 +4933,15 @@ abort_locked: if (likely(ctx)) { DPRINT(("context unlocked\n")); UNPROTECT_CTX(ctx, flags); - fput(file); } /* copy argument back to user, if needed */ if (call_made && PFM_CMD_RW_ARG(cmd) && copy_to_user(arg, args_k, base_sz*count)) ret = -EFAULT; error_args: + if (file) + fput(file); + kfree(args_k); DPRINT(("cmd=%s ret=%ld\n", PFM_CMD_NAME(cmd), ret)); @@ -5863,14 +5861,12 @@ void pfm_save_regs(struct task_struct *task) { pfm_context_t *ctx; - struct thread_struct *t; unsigned long flags; u64 psr; ctx = PFM_GET_CTX(task); if (ctx == NULL) return; - t = &task->thread; /* * we always come here with interrupts ALREADY disabled by @@ -5928,19 +5924,19 @@ pfm_save_regs(struct task_struct *task) * guarantee we will be schedule at that same * CPU again. */ - pfm_save_pmds(t->pmds, ctx->ctx_used_pmds[0]); + pfm_save_pmds(ctx->th_pmds, ctx->ctx_used_pmds[0]); /* * save pmc0 ia64_srlz_d() done in pfm_save_pmds() * we will need it on the restore path to check * for pending overflow. */ - t->pmcs[0] = ia64_get_pmc(0); + ctx->th_pmcs[0] = ia64_get_pmc(0); /* * unfreeze PMU if had pending overflows */ - if (t->pmcs[0] & ~0x1UL) pfm_unfreeze_pmu(); + if (ctx->th_pmcs[0] & ~0x1UL) pfm_unfreeze_pmu(); /* * finally, allow context access. @@ -5985,7 +5981,6 @@ static void pfm_lazy_save_regs (struct task_struct *task) { pfm_context_t *ctx; - struct thread_struct *t; unsigned long flags; { u64 psr = pfm_get_psr(); @@ -5993,7 +5988,6 @@ pfm_lazy_save_regs (struct task_struct *task) } ctx = PFM_GET_CTX(task); - t = &task->thread; /* * we need to mask PMU overflow here to @@ -6018,19 +6012,19 @@ pfm_lazy_save_regs (struct task_struct *task) /* * save all the pmds we use */ - pfm_save_pmds(t->pmds, ctx->ctx_used_pmds[0]); + pfm_save_pmds(ctx->th_pmds, ctx->ctx_used_pmds[0]); /* * save pmc0 ia64_srlz_d() done in pfm_save_pmds() * it is needed to check for pended overflow * on the restore path */ - t->pmcs[0] = ia64_get_pmc(0); + ctx->th_pmcs[0] = ia64_get_pmc(0); /* * unfreeze PMU if had pending overflows */ - if (t->pmcs[0] & ~0x1UL) pfm_unfreeze_pmu(); + if (ctx->th_pmcs[0] & ~0x1UL) pfm_unfreeze_pmu(); /* * now get can unmask PMU interrupts, they will @@ -6049,7 +6043,6 @@ void pfm_load_regs (struct task_struct *task) { pfm_context_t *ctx; - struct thread_struct *t; unsigned long pmc_mask = 0UL, pmd_mask = 0UL; unsigned long flags; u64 psr, psr_up; @@ -6060,11 +6053,10 @@ pfm_load_regs (struct task_struct *task) BUG_ON(GET_PMU_OWNER()); - t = &task->thread; /* * possible on unload */ - if (unlikely((t->flags & IA64_THREAD_PM_VALID) == 0)) return; + if (unlikely((task->thread.flags & IA64_THREAD_PM_VALID) == 0)) return; /* * we always come here with interrupts ALREADY disabled by @@ -6146,26 +6138,26 @@ pfm_load_regs (struct task_struct *task) * * XXX: optimize here */ - if (pmd_mask) pfm_restore_pmds(t->pmds, pmd_mask); - if (pmc_mask) pfm_restore_pmcs(t->pmcs, pmc_mask); + if (pmd_mask) pfm_restore_pmds(ctx->th_pmds, pmd_mask); + if (pmc_mask) pfm_restore_pmcs(ctx->th_pmcs, pmc_mask); /* * check for pending overflow at the time the state * was saved. */ - if (unlikely(PMC0_HAS_OVFL(t->pmcs[0]))) { + if (unlikely(PMC0_HAS_OVFL(ctx->th_pmcs[0]))) { /* * reload pmc0 with the overflow information * On McKinley PMU, this will trigger a PMU interrupt */ - ia64_set_pmc(0, t->pmcs[0]); + ia64_set_pmc(0, ctx->th_pmcs[0]); ia64_srlz_d(); - t->pmcs[0] = 0UL; + ctx->th_pmcs[0] = 0UL; /* * will replay the PMU interrupt */ - if (need_irq_resend) hw_resend_irq(NULL, IA64_PERFMON_VECTOR); + if (need_irq_resend) ia64_resend_irq(IA64_PERFMON_VECTOR); pfm_stats[smp_processor_id()].pfm_replay_ovfl_intr_count++; } @@ -6213,7 +6205,6 @@ pfm_load_regs (struct task_struct *task) void pfm_load_regs (struct task_struct *task) { - struct thread_struct *t; pfm_context_t *ctx; struct task_struct *owner; unsigned long pmd_mask, pmc_mask; @@ -6222,7 +6213,6 @@ pfm_load_regs (struct task_struct *task) owner = GET_PMU_OWNER(); ctx = PFM_GET_CTX(task); - t = &task->thread; psr = pfm_get_psr(); BUG_ON(psr & (IA64_PSR_UP|IA64_PSR_PP)); @@ -6285,27 +6275,27 @@ pfm_load_regs (struct task_struct *task) */ pmc_mask = ctx->ctx_all_pmcs[0]; - pfm_restore_pmds(t->pmds, pmd_mask); - pfm_restore_pmcs(t->pmcs, pmc_mask); + pfm_restore_pmds(ctx->th_pmds, pmd_mask); + pfm_restore_pmcs(ctx->th_pmcs, pmc_mask); /* * check for pending overflow at the time the state * was saved. */ - if (unlikely(PMC0_HAS_OVFL(t->pmcs[0]))) { + if (unlikely(PMC0_HAS_OVFL(ctx->th_pmcs[0]))) { /* * reload pmc0 with the overflow information * On McKinley PMU, this will trigger a PMU interrupt */ - ia64_set_pmc(0, t->pmcs[0]); + ia64_set_pmc(0, ctx->th_pmcs[0]); ia64_srlz_d(); - t->pmcs[0] = 0UL; + ctx->th_pmcs[0] = 0UL; /* * will replay the PMU interrupt */ - if (need_irq_resend) hw_resend_irq(NULL, IA64_PERFMON_VECTOR); + if (need_irq_resend) ia64_resend_irq(IA64_PERFMON_VECTOR); pfm_stats[smp_processor_id()].pfm_replay_ovfl_intr_count++; } @@ -6375,11 +6365,11 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx) */ pfm_unfreeze_pmu(); } else { - pmc0 = task->thread.pmcs[0]; + pmc0 = ctx->th_pmcs[0]; /* * clear whatever overflow status bits there were */ - task->thread.pmcs[0] = 0; + ctx->th_pmcs[0] = 0; } ovfl_val = pmu_conf->ovfl_val; /* @@ -6400,7 +6390,7 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx) /* * can access PMU always true in system wide mode */ - val = pmd_val = can_access_pmu ? ia64_get_pmd(i) : task->thread.pmds[i]; + val = pmd_val = can_access_pmu ? ia64_get_pmd(i) : ctx->th_pmds[i]; if (PMD_IS_COUNTING(i)) { DPRINT(("[%d] pmd[%d] ctx_pmd=0x%lx hw_pmd=0x%lx\n", @@ -6432,7 +6422,7 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx) DPRINT(("[%d] ctx_pmd[%d]=0x%lx pmd_val=0x%lx\n", task->pid, i, val, pmd_val)); - if (is_self) task->thread.pmds[i] = pmd_val; + if (is_self) ctx->th_pmds[i] = pmd_val; ctx->ctx_pmds[i].val = val; } @@ -6440,7 +6430,7 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx) static struct irqaction perfmon_irqaction = { .handler = pfm_interrupt_handler, - .flags = SA_INTERRUPT, + .flags = IRQF_DISABLED, .name = "perfmon" }; @@ -6676,7 +6666,7 @@ pfm_init(void) ffz(pmu_conf->ovfl_val)); /* sanity check */ - if (pmu_conf->num_pmds >= IA64_NUM_PMD_REGS || pmu_conf->num_pmcs >= IA64_NUM_PMC_REGS) { + if (pmu_conf->num_pmds >= PFM_NUM_PMD_REGS || pmu_conf->num_pmcs >= PFM_NUM_PMC_REGS) { printk(KERN_ERR "perfmon: not enough pmc/pmd, perfmon disabled\n"); pmu_conf = NULL; return -1; @@ -6751,7 +6741,6 @@ void dump_pmu_state(const char *from) { struct task_struct *task; - struct thread_struct *t; struct pt_regs *regs; pfm_context_t *ctx; unsigned long psr, dcr, info, flags; @@ -6796,16 +6785,14 @@ dump_pmu_state(const char *from) ia64_psr(regs)->up = 0; ia64_psr(regs)->pp = 0; - t = ¤t->thread; - for (i=1; PMC_IS_LAST(i) == 0; i++) { if (PMC_IS_IMPL(i) == 0) continue; - printk("->CPU%d pmc[%d]=0x%lx thread_pmc[%d]=0x%lx\n", this_cpu, i, ia64_get_pmc(i), i, t->pmcs[i]); + printk("->CPU%d pmc[%d]=0x%lx thread_pmc[%d]=0x%lx\n", this_cpu, i, ia64_get_pmc(i), i, ctx->th_pmcs[i]); } for (i=1; PMD_IS_LAST(i) == 0; i++) { if (PMD_IS_IMPL(i) == 0) continue; - printk("->CPU%d pmd[%d]=0x%lx thread_pmd[%d]=0x%lx\n", this_cpu, i, ia64_get_pmd(i), i, t->pmds[i]); + printk("->CPU%d pmd[%d]=0x%lx thread_pmd[%d]=0x%lx\n", this_cpu, i, ia64_get_pmd(i), i, ctx->th_pmds[i]); } if (ctx) { diff --git a/arch/ia64/kernel/perfmon_default_smpl.c b/arch/ia64/kernel/perfmon_default_smpl.c index 344941db0a9..ff80eab83b3 100644 --- a/arch/ia64/kernel/perfmon_default_smpl.c +++ b/arch/ia64/kernel/perfmon_default_smpl.c @@ -8,7 +8,6 @@ #include <linux/kernel.h> #include <linux/types.h> #include <linux/module.h> -#include <linux/config.h> #include <linux/init.h> #include <asm/delay.h> #include <linux/smp.h> diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c index 355d57970ba..ea914cc6812 100644 --- a/arch/ia64/kernel/process.c +++ b/arch/ia64/kernel/process.c @@ -9,7 +9,6 @@ * Add notify_die() hooks. */ #define __KERNEL_SYSCALLS__ /* see <asm/unistd.h> */ -#include <linux/config.h> #include <linux/cpu.h> #include <linux/pm.h> @@ -272,9 +271,9 @@ cpu_idle (void) /* endless idle loop with no priority at all */ while (1) { if (can_do_pal_halt) - clear_thread_flag(TIF_POLLING_NRFLAG); + current_thread_info()->status &= ~TS_POLLING; else - set_thread_flag(TIF_POLLING_NRFLAG); + current_thread_info()->status |= TS_POLLING; if (!need_resched()) { void (*idle)(void); diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c index e61e15e28d8..aa705e46b97 100644 --- a/arch/ia64/kernel/ptrace.c +++ b/arch/ia64/kernel/ptrace.c @@ -6,7 +6,6 @@ * * Derived from the x86 and Alpha versions. */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/sched.h> #include <linux/slab.h> diff --git a/arch/ia64/kernel/sal.c b/arch/ia64/kernel/sal.c index 77fa65903d9..642fdc7b969 100644 --- a/arch/ia64/kernel/sal.c +++ b/arch/ia64/kernel/sal.c @@ -6,7 +6,6 @@ * Copyright (C) 1999 VA Linux Systems * Copyright (C) 1999 Walt Drummond <drummond@valinux.com> */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/init.h> diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c index 663a186ad19..e63b8ca5344 100644 --- a/arch/ia64/kernel/salinfo.c +++ b/arch/ia64/kernel/salinfo.c @@ -266,6 +266,7 @@ salinfo_log_wakeup(int type, u8 *buffer, u64 size, int irqsafe) /* Check for outstanding MCA/INIT records every minute (arbitrary) */ #define SALINFO_TIMER_DELAY (60*HZ) static struct timer_list salinfo_timer; +extern void ia64_mlogbuf_dump(void); static void salinfo_timeout_check(struct salinfo_data *data) @@ -283,6 +284,7 @@ salinfo_timeout_check(struct salinfo_data *data) static void salinfo_timeout (unsigned long arg) { + ia64_mlogbuf_dump(); salinfo_timeout_check(salinfo_data + SAL_INFO_TYPE_MCA); salinfo_timeout_check(salinfo_data + SAL_INFO_TYPE_INIT); salinfo_timer.expires = jiffies + SALINFO_TIMER_DELAY; @@ -332,6 +334,8 @@ retry: if (cpu == -1) goto retry; + ia64_mlogbuf_dump(); + /* for next read, start checking at next CPU */ data->cpu_check = cpu; if (++data->cpu_check == NR_CPUS) @@ -572,7 +576,7 @@ static struct file_operations salinfo_data_fops = { }; #ifdef CONFIG_HOTPLUG_CPU -static int +static int __devinit salinfo_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) { unsigned int i, cpu = (unsigned long)hcpu; @@ -673,9 +677,7 @@ salinfo_init(void) salinfo_timer.function = &salinfo_timeout; add_timer(&salinfo_timer); -#ifdef CONFIG_HOTPLUG_CPU - register_cpu_notifier(&salinfo_cpu_notifier); -#endif + register_hotcpu_notifier(&salinfo_cpu_notifier); return 0; } diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index 6dba2d63f24..84f93c0f2c6 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -22,7 +22,6 @@ * 06/24/99 W.Drummond added boot_cpu_data. * 05/28/05 Z. Menyhart Dynamic stride size for "flush_icache_range()" */ -#include <linux/config.h> #include <linux/module.h> #include <linux/init.h> @@ -36,7 +35,7 @@ #include <linux/seq_file.h> #include <linux/string.h> #include <linux/threads.h> -#include <linux/tty.h> +#include <linux/screen_info.h> #include <linux/dmi.h> #include <linux/serial.h> #include <linux/serial_core.h> @@ -510,7 +509,7 @@ show_cpuinfo (struct seq_file *m, void *v) { 1UL << 1, "spontaneous deferral"}, { 1UL << 2, "16-byte atomic ops" } }; - char family[32], features[128], *cp, sep; + char features[128], *cp, sep; struct cpuinfo_ia64 *c = v; unsigned long mask; unsigned long proc_freq; @@ -518,12 +517,6 @@ show_cpuinfo (struct seq_file *m, void *v) mask = c->features; - switch (c->family) { - case 0x07: memcpy(family, "Itanium", 8); break; - case 0x1f: memcpy(family, "Itanium 2", 10); break; - default: sprintf(family, "%u", c->family); break; - } - /* build the feature string: */ memcpy(features, " standard", 10); cp = features; @@ -554,8 +547,9 @@ show_cpuinfo (struct seq_file *m, void *v) "processor : %d\n" "vendor : %s\n" "arch : IA-64\n" - "family : %s\n" + "family : %u\n" "model : %u\n" + "model name : %s\n" "revision : %u\n" "archrev : %u\n" "features :%s\n" /* don't change this---it _is_ right! */ @@ -564,7 +558,8 @@ show_cpuinfo (struct seq_file *m, void *v) "cpu MHz : %lu.%06lu\n" "itc MHz : %lu.%06lu\n" "BogoMIPS : %lu.%02lu\n", - cpunum, c->vendor, family, c->model, c->revision, c->archrev, + cpunum, c->vendor, c->family, c->model, + c->model_name, c->revision, c->archrev, features, c->ppn, c->number, proc_freq / 1000, proc_freq % 1000, c->itc_freq / 1000000, c->itc_freq % 1000000, @@ -612,6 +607,31 @@ struct seq_operations cpuinfo_op = { .show = show_cpuinfo }; +static char brandname[128]; + +static char * __cpuinit +get_model_name(__u8 family, __u8 model) +{ + char brand[128]; + + if (ia64_pal_get_brand_info(brand)) { + if (family == 0x7) + memcpy(brand, "Merced", 7); + else if (family == 0x1f) switch (model) { + case 0: memcpy(brand, "McKinley", 9); break; + case 1: memcpy(brand, "Madison", 8); break; + case 2: memcpy(brand, "Madison up to 9M cache", 23); break; + } else + memcpy(brand, "Unknown", 8); + } + if (brandname[0] == '\0') + return strcpy(brandname, brand); + else if (strcmp(brandname, brand) == 0) + return brandname; + else + return kstrdup(brand, GFP_KERNEL); +} + static void __cpuinit identify_cpu (struct cpuinfo_ia64 *c) { @@ -641,7 +661,6 @@ identify_cpu (struct cpuinfo_ia64 *c) pal_status_t status; unsigned long impl_va_msb = 50, phys_addr_size = 44; /* Itanium defaults */ int i; - for (i = 0; i < 5; ++i) cpuid.bits[i] = ia64_get_cpuid(i); @@ -664,6 +683,7 @@ identify_cpu (struct cpuinfo_ia64 *c) c->family = cpuid.field.family; c->archrev = cpuid.field.archrev; c->features = cpuid.field.features; + c->model_name = get_model_name(c->family, c->model); status = ia64_pal_vm_summary(&vm1, &vm2); if (status == PAL_STATUS_SUCCESS) { diff --git a/arch/ia64/kernel/signal.c b/arch/ia64/kernel/signal.c index 1d7903ee212..77f8b49c788 100644 --- a/arch/ia64/kernel/signal.c +++ b/arch/ia64/kernel/signal.c @@ -7,7 +7,6 @@ * Derived from i386 and Alpha versions. */ -#include <linux/config.h> #include <linux/errno.h> #include <linux/kernel.h> #include <linux/mm.h> diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index 44e9547878a..f7d7f566814 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -21,7 +21,6 @@ * 05/01/30 Suresh Siddha <suresh.b.siddha@intel.com> * Setup cpu_sibling_map and cpu_core_map */ -#include <linux/config.h> #include <linux/module.h> #include <linux/acpi.h> @@ -125,7 +124,7 @@ extern void __devinit calibrate_delay (void); extern void start_ap (void); extern unsigned long ia64_iobase; -task_t *task_for_booting_cpu; +struct task_struct *task_for_booting_cpu; /* * State for each CPU @@ -677,16 +676,16 @@ int migrate_platform_irqs(unsigned int cpu) new_cpei_cpu = any_online_cpu(cpu_online_map); mask = cpumask_of_cpu(new_cpei_cpu); set_cpei_target_cpu(new_cpei_cpu); - desc = irq_descp(ia64_cpe_irq); + desc = irq_desc + ia64_cpe_irq; /* * Switch for now, immediatly, we need to do fake intr * as other interrupts, but need to study CPEI behaviour with * polling before making changes. */ if (desc) { - desc->handler->disable(ia64_cpe_irq); - desc->handler->set_affinity(ia64_cpe_irq, mask); - desc->handler->enable(ia64_cpe_irq); + desc->chip->disable(ia64_cpe_irq); + desc->chip->set_affinity(ia64_cpe_irq, mask); + desc->chip->enable(ia64_cpe_irq); printk ("Re-targetting CPEI to cpu %d\n", new_cpei_cpu); } } @@ -880,3 +879,27 @@ identify_siblings(struct cpuinfo_ia64 *c) c->core_id = info.log1_cid; c->thread_id = info.log1_tid; } + +/* + * returns non zero, if multi-threading is enabled + * on at least one physical package. Due to hotplug cpu + * and (maxcpus=), all threads may not necessarily be enabled + * even though the processor supports multi-threading. + */ +int is_multithreading_enabled(void) +{ + int i, j; + + for_each_present_cpu(i) { + for_each_present_cpu(j) { + if (j == i) + continue; + if ((cpu_data(j)->socket_id == cpu_data(i)->socket_id)) { + if (cpu_data(j)->core_id == cpu_data(i)->core_id) + return 1; + } + } + } + return 0; +} +EXPORT_SYMBOL_GPL(is_multithreading_enabled); diff --git a/arch/ia64/kernel/sys_ia64.c b/arch/ia64/kernel/sys_ia64.c index c7b943f1019..9ef62a3fbfa 100644 --- a/arch/ia64/kernel/sys_ia64.c +++ b/arch/ia64/kernel/sys_ia64.c @@ -5,7 +5,6 @@ * Copyright (C) 1999-2000, 2002-2003, 2005 Hewlett-Packard Co * David Mosberger-Tang <davidm@hpl.hp.com> */ -#include <linux/config.h> #include <linux/errno.h> #include <linux/fs.h> #include <linux/mm.h> @@ -164,10 +163,25 @@ sys_pipe (void) return retval; } +int ia64_mmap_check(unsigned long addr, unsigned long len, + unsigned long flags) +{ + unsigned long roff; + + /* + * Don't permit mappings into unmapped space, the virtual page table + * of a region, or across a region boundary. Note: RGN_MAP_LIMIT is + * equal to 2^n-PAGE_SIZE (for some integer n <= 61) and len > 0. + */ + roff = REGION_OFFSET(addr); + if ((len > RGN_MAP_LIMIT) || (roff > (RGN_MAP_LIMIT - len))) + return -EINVAL; + return 0; +} + static inline unsigned long do_mmap2 (unsigned long addr, unsigned long len, int prot, int flags, int fd, unsigned long pgoff) { - unsigned long roff; struct file *file = NULL; flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); @@ -189,17 +203,6 @@ do_mmap2 (unsigned long addr, unsigned long len, int prot, int flags, int fd, un goto out; } - /* - * Don't permit mappings into unmapped space, the virtual page table of a region, - * or across a region boundary. Note: RGN_MAP_LIMIT is equal to 2^n-PAGE_SIZE - * (for some integer n <= 61) and len > 0. - */ - roff = REGION_OFFSET(addr); - if ((len > RGN_MAP_LIMIT) || (roff > (RGN_MAP_LIMIT - len))) { - addr = -EINVAL; - goto out; - } - down_write(¤t->mm->mmap_sem); addr = do_mmap_pgoff(file, addr, len, prot, flags, pgoff); up_write(¤t->mm->mmap_sem); diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index 49958904045..16262687a10 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -8,7 +8,6 @@ * Copyright (C) 1999-2000 VA Linux Systems * Copyright (C) 1999-2000 Walt Drummond <drummond@valinux.com> */ -#include <linux/config.h> #include <linux/cpu.h> #include <linux/init.h> @@ -79,7 +78,7 @@ timer_interrupt (int irq, void *dev_id, struct pt_regs *regs) * xtime_lock. */ write_seqlock(&xtime_lock); - do_timer(regs); + do_timer(1); local_cpu_data->itm_next = new_itm; write_sequnlock(&xtime_lock); } else @@ -232,7 +231,7 @@ ia64_init_itm (void) static struct irqaction timer_irqaction = { .handler = timer_interrupt, - .flags = SA_INTERRUPT, + .flags = IRQF_DISABLED, .name = "timer" }; diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index 4f3a16b37f8..5629b45e89c 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -13,7 +13,6 @@ * Populate cpu cache entries in sysfs for cpu cache info */ -#include <linux/config.h> #include <linux/cpu.h> #include <linux/kernel.h> #include <linux/mm.h> @@ -26,19 +25,10 @@ #include <asm/numa.h> #include <asm/cpu.h> -#ifdef CONFIG_NUMA -static struct node *sysfs_nodes; -#endif static struct ia64_cpu *sysfs_cpus; int arch_register_cpu(int num) { - struct node *parent = NULL; - -#ifdef CONFIG_NUMA - parent = &sysfs_nodes[cpu_to_node(num)]; -#endif /* CONFIG_NUMA */ - #if defined (CONFIG_ACPI) && defined (CONFIG_HOTPLUG_CPU) /* * If CPEI cannot be re-targetted, and this is @@ -46,23 +36,18 @@ int arch_register_cpu(int num) */ if (!can_cpei_retarget() && is_cpu_cpei_target(num)) sysfs_cpus[num].cpu.no_control = 1; + map_cpu_to_node(num, node_cpuid[num].nid); #endif - return register_cpu(&sysfs_cpus[num].cpu, num, parent); + return register_cpu(&sysfs_cpus[num].cpu, num); } #ifdef CONFIG_HOTPLUG_CPU void arch_unregister_cpu(int num) { - struct node *parent = NULL; - -#ifdef CONFIG_NUMA - int node = cpu_to_node(num); - parent = &sysfs_nodes[node]; -#endif /* CONFIG_NUMA */ - - return unregister_cpu(&sysfs_cpus[num].cpu, parent); + unregister_cpu(&sysfs_cpus[num].cpu); + unmap_cpu_from_node(num, cpu_to_node(num)); } EXPORT_SYMBOL(arch_register_cpu); EXPORT_SYMBOL(arch_unregister_cpu); @@ -74,26 +59,18 @@ static int __init topology_init(void) int i, err = 0; #ifdef CONFIG_NUMA - sysfs_nodes = kzalloc(sizeof(struct node) * MAX_NUMNODES, GFP_KERNEL); - if (!sysfs_nodes) { - err = -ENOMEM; - goto out; - } - /* * MCD - Do we want to register all ONLINE nodes, or all POSSIBLE nodes? */ for_each_online_node(i) { - if ((err = register_node(&sysfs_nodes[i], i, 0))) + if ((err = register_one_node(i))) goto out; } #endif sysfs_cpus = kzalloc(sizeof(struct ia64_cpu) * NR_CPUS, GFP_KERNEL); - if (!sysfs_cpus) { - err = -ENOMEM; - goto out; - } + if (!sysfs_cpus) + panic("kzalloc in topology_init failed - NR_CPUS too big?"); for_each_present_cpu(i) { if((err = arch_register_cpu(i))) @@ -166,7 +143,7 @@ static void cache_shared_cpu_map_setup( unsigned int cpu, num_shared = (int) csi.num_shared; do { - for_each_cpu(j) + for_each_possible_cpu(j) if (cpu_data(cpu)->socket_id == cpu_data(j)->socket_id && cpu_data(j)->core_id == csi.log1_cid && cpu_data(j)->thread_id == csi.log1_tid) @@ -426,7 +403,7 @@ static int __cpuinit cache_remove_dev(struct sys_device * sys_dev) * When a cpu is hot-plugged, do a check and initiate * cache kobject if necessary */ -static int cache_cpu_callback(struct notifier_block *nfb, +static int __cpuinit cache_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -444,7 +421,7 @@ static int cache_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -static struct notifier_block cache_cpu_notifier = +static struct notifier_block __cpuinitdata cache_cpu_notifier = { .notifier_call = cache_cpu_callback }; @@ -458,7 +435,7 @@ static int __cpuinit cache_sysfs_init(void) (void *)(long)i); } - register_cpu_notifier(&cache_cpu_notifier); + register_hotcpu_notifier(&cache_cpu_notifier); return 0; } diff --git a/arch/ia64/kernel/traps.c b/arch/ia64/kernel/traps.c index 7c1ddc8ac44..fffa9e0826b 100644 --- a/arch/ia64/kernel/traps.c +++ b/arch/ia64/kernel/traps.c @@ -7,7 +7,6 @@ * 05/12/00 grao <goutham.rao@intel.com> : added isr in siginfo for SIGFPE */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/init.h> #include <linux/sched.h> @@ -118,11 +117,8 @@ die (const char *str, struct pt_regs *regs, long err) die.lock_owner = -1; spin_unlock_irq(&die.lock); - if (panic_on_oops) { - printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n"); - ssleep(5); + if (panic_on_oops) panic("Fatal exception"); - } do_exit(SIGSEGV); } diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c index 5f03b9e524d..c58e933694d 100644 --- a/arch/ia64/kernel/uncached.c +++ b/arch/ia64/kernel/uncached.c @@ -32,32 +32,38 @@ extern void __init efi_memmap_walk_uc(efi_freemem_callback_t, void *); -#define MAX_UNCACHED_GRANULES 5 -static int allocated_granules; +struct uncached_pool { + struct gen_pool *pool; + struct mutex add_chunk_mutex; /* serialize adding a converted chunk */ + int nchunks_added; /* #of converted chunks added to pool */ + atomic_t status; /* smp called function's return status*/ +}; + +#define MAX_CONVERTED_CHUNKS_PER_NODE 2 -struct gen_pool *uncached_pool[MAX_NUMNODES]; +struct uncached_pool uncached_pools[MAX_NUMNODES]; static void uncached_ipi_visibility(void *data) { int status; + struct uncached_pool *uc_pool = (struct uncached_pool *)data; status = ia64_pal_prefetch_visibility(PAL_VISIBILITY_PHYSICAL); if ((status != PAL_VISIBILITY_OK) && (status != PAL_VISIBILITY_OK_REMOTE_NEEDED)) - printk(KERN_DEBUG "pal_prefetch_visibility() returns %i on " - "CPU %i\n", status, raw_smp_processor_id()); + atomic_inc(&uc_pool->status); } static void uncached_ipi_mc_drain(void *data) { int status; + struct uncached_pool *uc_pool = (struct uncached_pool *)data; status = ia64_pal_mc_drain(); - if (status) - printk(KERN_WARNING "ia64_pal_mc_drain() failed with %i on " - "CPU %i\n", status, raw_smp_processor_id()); + if (status != PAL_STATUS_SUCCESS) + atomic_inc(&uc_pool->status); } @@ -70,21 +76,34 @@ static void uncached_ipi_mc_drain(void *data) * This is accomplished by first allocating a granule of cached memory pages * and then converting them to uncached memory pages. */ -static int uncached_add_chunk(struct gen_pool *pool, int nid) +static int uncached_add_chunk(struct uncached_pool *uc_pool, int nid) { struct page *page; - int status, i; + int status, i, nchunks_added = uc_pool->nchunks_added; unsigned long c_addr, uc_addr; - if (allocated_granules >= MAX_UNCACHED_GRANULES) + if (mutex_lock_interruptible(&uc_pool->add_chunk_mutex) != 0) + return -1; /* interrupted by a signal */ + + if (uc_pool->nchunks_added > nchunks_added) { + /* someone added a new chunk while we were waiting */ + mutex_unlock(&uc_pool->add_chunk_mutex); + return 0; + } + + if (uc_pool->nchunks_added >= MAX_CONVERTED_CHUNKS_PER_NODE) { + mutex_unlock(&uc_pool->add_chunk_mutex); return -1; + } /* attempt to allocate a granule's worth of cached memory pages */ - page = alloc_pages_node(nid, GFP_KERNEL | __GFP_ZERO, + page = alloc_pages_node(nid, GFP_KERNEL | __GFP_ZERO | GFP_THISNODE, IA64_GRANULE_SHIFT-PAGE_SHIFT); - if (!page) + if (!page) { + mutex_unlock(&uc_pool->add_chunk_mutex); return -1; + } /* convert the memory pages from cached to uncached */ @@ -102,11 +121,14 @@ static int uncached_add_chunk(struct gen_pool *pool, int nid) flush_tlb_kernel_range(uc_addr, uc_adddr + IA64_GRANULE_SIZE); status = ia64_pal_prefetch_visibility(PAL_VISIBILITY_PHYSICAL); - if (!status) { - status = smp_call_function(uncached_ipi_visibility, NULL, 0, 1); - if (status) + if (status == PAL_VISIBILITY_OK_REMOTE_NEEDED) { + atomic_set(&uc_pool->status, 0); + status = smp_call_function(uncached_ipi_visibility, uc_pool, + 0, 1); + if (status || atomic_read(&uc_pool->status)) goto failed; - } + } else if (status != PAL_VISIBILITY_OK) + goto failed; preempt_disable(); @@ -120,20 +142,24 @@ static int uncached_add_chunk(struct gen_pool *pool, int nid) preempt_enable(); - ia64_pal_mc_drain(); - status = smp_call_function(uncached_ipi_mc_drain, NULL, 0, 1); - if (status) + status = ia64_pal_mc_drain(); + if (status != PAL_STATUS_SUCCESS) + goto failed; + atomic_set(&uc_pool->status, 0); + status = smp_call_function(uncached_ipi_mc_drain, uc_pool, 0, 1); + if (status || atomic_read(&uc_pool->status)) goto failed; /* * The chunk of memory pages has been converted to uncached so now we * can add it to the pool. */ - status = gen_pool_add(pool, uc_addr, IA64_GRANULE_SIZE, nid); + status = gen_pool_add(uc_pool->pool, uc_addr, IA64_GRANULE_SIZE, nid); if (status) goto failed; - allocated_granules++; + uc_pool->nchunks_added++; + mutex_unlock(&uc_pool->add_chunk_mutex); return 0; /* failed to convert or add the chunk so give it back to the kernel */ @@ -142,6 +168,7 @@ failed: ClearPageUncached(&page[i]); free_pages(c_addr, IA64_GRANULE_SHIFT-PAGE_SHIFT); + mutex_unlock(&uc_pool->add_chunk_mutex); return -1; } @@ -158,7 +185,7 @@ failed: unsigned long uncached_alloc_page(int starting_nid) { unsigned long uc_addr; - struct gen_pool *pool; + struct uncached_pool *uc_pool; int nid; if (unlikely(starting_nid >= MAX_NUMNODES)) @@ -171,14 +198,14 @@ unsigned long uncached_alloc_page(int starting_nid) do { if (!node_online(nid)) continue; - pool = uncached_pool[nid]; - if (pool == NULL) + uc_pool = &uncached_pools[nid]; + if (uc_pool->pool == NULL) continue; do { - uc_addr = gen_pool_alloc(pool, PAGE_SIZE); + uc_addr = gen_pool_alloc(uc_pool->pool, PAGE_SIZE); if (uc_addr != 0) return uc_addr; - } while (uncached_add_chunk(pool, nid) == 0); + } while (uncached_add_chunk(uc_pool, nid) == 0); } while ((nid = (nid + 1) % MAX_NUMNODES) != starting_nid); @@ -197,7 +224,7 @@ EXPORT_SYMBOL(uncached_alloc_page); void uncached_free_page(unsigned long uc_addr) { int nid = paddr_to_nid(uc_addr - __IA64_UNCACHED_OFFSET); - struct gen_pool *pool = uncached_pool[nid]; + struct gen_pool *pool = uncached_pools[nid].pool; if (unlikely(pool == NULL)) return; @@ -224,7 +251,7 @@ static int __init uncached_build_memmap(unsigned long uc_start, unsigned long uc_end, void *arg) { int nid = paddr_to_nid(uc_start - __IA64_UNCACHED_OFFSET); - struct gen_pool *pool = uncached_pool[nid]; + struct gen_pool *pool = uncached_pools[nid].pool; size_t size = uc_end - uc_start; touch_softlockup_watchdog(); @@ -242,7 +269,8 @@ static int __init uncached_init(void) int nid; for_each_online_node(nid) { - uncached_pool[nid] = gen_pool_create(PAGE_SHIFT, nid); + uncached_pools[nid].pool = gen_pool_create(PAGE_SHIFT, nid); + mutex_init(&uncached_pools[nid].add_chunk_mutex); } efi_memmap_walk_uc(uncached_build_memmap, NULL); diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S index 783600fe52b..b3b2e389d6b 100644 --- a/arch/ia64/kernel/vmlinux.lds.S +++ b/arch/ia64/kernel/vmlinux.lds.S @@ -1,4 +1,3 @@ -#include <linux/config.h> #include <asm/cache.h> #include <asm/ptrace.h> @@ -185,7 +184,9 @@ SECTIONS *(.data.gate) __stop_gate_section = .; } - . = ALIGN(PAGE_SIZE); /* make sure the gate page doesn't expose kernel data */ + . = ALIGN(PAGE_SIZE); /* make sure the gate page doesn't expose + * kernel data + */ .data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET) { *(.data.read_mostly) } @@ -203,7 +204,9 @@ SECTIONS *(.data.percpu) __per_cpu_end = .; } - . = __phys_per_cpu_start + PERCPU_PAGE_SIZE; /* ensure percpu data fits into percpu page size */ + . = __phys_per_cpu_start + PERCPU_PAGE_SIZE; /* ensure percpu data fits + * into percpu page size + */ data : { } :data .data : AT(ADDR(.data) - LOAD_OFFSET) diff --git a/arch/ia64/lib/Makefile b/arch/ia64/lib/Makefile index d8536a2c22a..38fa6e49e79 100644 --- a/arch/ia64/lib/Makefile +++ b/arch/ia64/lib/Makefile @@ -14,7 +14,7 @@ lib-y := __divsi3.o __udivsi3.o __modsi3.o __umodsi3.o \ lib-$(CONFIG_ITANIUM) += copy_page.o copy_user.o memcpy.o lib-$(CONFIG_MCKINLEY) += copy_page_mck.o memcpy_mck.o lib-$(CONFIG_PERFMON) += carta_random.o -lib-$(CONFIG_MD_RAID5) += xor.o +lib-$(CONFIG_MD_RAID456) += xor.o AFLAGS___divdi3.o = AFLAGS___udivdi3.o = -DUNSIGNED diff --git a/arch/ia64/lib/clear_page.S b/arch/ia64/lib/clear_page.S index d4987061dda..2d814e7ed19 100644 --- a/arch/ia64/lib/clear_page.S +++ b/arch/ia64/lib/clear_page.S @@ -8,7 +8,6 @@ * 2/12/02 kchen Tuned for both Itanium and McKinley * 3/08/02 davidm Some more tweaking */ -#include <linux/config.h> #include <asm/asmmacro.h> #include <asm/page.h> diff --git a/arch/ia64/lib/io.c b/arch/ia64/lib/io.c index 8949e44091a..bcd16f8ad9d 100644 --- a/arch/ia64/lib/io.c +++ b/arch/ia64/lib/io.c @@ -1,4 +1,3 @@ -#include <linux/config.h> #include <linux/module.h> #include <linux/types.h> diff --git a/arch/ia64/lib/memcpy_mck.S b/arch/ia64/lib/memcpy_mck.S index 9e534d52b1d..ab0f8763972 100644 --- a/arch/ia64/lib/memcpy_mck.S +++ b/arch/ia64/lib/memcpy_mck.S @@ -13,7 +13,6 @@ * Copyright (C) 2002 Intel Corp. * Copyright (C) 2002 Ken Chen <kenneth.w.chen@intel.com> */ -#include <linux/config.h> #include <asm/asmmacro.h> #include <asm/page.h> diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c index 84fd1c14c8a..daf977ff292 100644 --- a/arch/ia64/mm/contig.c +++ b/arch/ia64/mm/contig.c @@ -14,7 +14,6 @@ * Routines used by ia64 machines with contiguous (or virtually contiguous) * memory. */ -#include <linux/config.h> #include <linux/bootmem.h> #include <linux/efi.h> #include <linux/mm.h> @@ -27,7 +26,7 @@ #include <asm/mca.h> #ifdef CONFIG_VIRTUAL_MEM_MAP -static unsigned long num_dma_physpages; +static unsigned long max_gap; #endif /** @@ -41,14 +40,21 @@ show_mem (void) int i, total = 0, reserved = 0; int shared = 0, cached = 0; - printk("Mem-info:\n"); + printk(KERN_INFO "Mem-info:\n"); show_free_areas(); - printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); + printk(KERN_INFO "Free swap: %6ldkB\n", + nr_swap_pages<<(PAGE_SHIFT-10)); i = max_mapnr; - while (i-- > 0) { - if (!pfn_valid(i)) + for (i = 0; i < max_mapnr; i++) { + if (!pfn_valid(i)) { +#ifdef CONFIG_VIRTUAL_MEM_MAP + if (max_gap < LARGE_GAP) + continue; + i = vmemmap_find_next_valid_pfn(0, i) - 1; +#endif continue; + } total++; if (PageReserved(mem_map+i)) reserved++; @@ -57,12 +63,12 @@ show_mem (void) else if (page_count(mem_map + i)) shared += page_count(mem_map + i) - 1; } - printk("%d pages of RAM\n", total); - printk("%d reserved pages\n", reserved); - printk("%d pages shared\n", shared); - printk("%d pages swap cached\n", cached); - printk("%ld pages in page table cache\n", - pgtable_quicklist_total_size()); + printk(KERN_INFO "%d pages of RAM\n", total); + printk(KERN_INFO "%d reserved pages\n", reserved); + printk(KERN_INFO "%d pages shared\n", shared); + printk(KERN_INFO "%d pages swap cached\n", cached); + printk(KERN_INFO "%ld pages in page table cache\n", + pgtable_quicklist_total_size()); } /* physical address where the bootmem map is located */ @@ -212,18 +218,6 @@ count_pages (u64 start, u64 end, void *arg) return 0; } -#ifdef CONFIG_VIRTUAL_MEM_MAP -static int -count_dma_pages (u64 start, u64 end, void *arg) -{ - unsigned long *count = arg; - - if (start < MAX_DMA_ADDRESS) - *count += (min(end, MAX_DMA_ADDRESS) - start) >> PAGE_SHIFT; - return 0; -} -#endif - /* * Set up the page tables. */ @@ -232,71 +226,46 @@ void __init paging_init (void) { unsigned long max_dma; - unsigned long zones_size[MAX_NR_ZONES]; -#ifdef CONFIG_VIRTUAL_MEM_MAP - unsigned long zholes_size[MAX_NR_ZONES]; - unsigned long max_gap; -#endif - - /* initialize mem_map[] */ - - memset(zones_size, 0, sizeof(zones_size)); + unsigned long nid = 0; + unsigned long max_zone_pfns[MAX_NR_ZONES]; num_physpages = 0; efi_memmap_walk(count_pages, &num_physpages); max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT; + max_zone_pfns[ZONE_DMA] = max_dma; + max_zone_pfns[ZONE_NORMAL] = max_low_pfn; #ifdef CONFIG_VIRTUAL_MEM_MAP - memset(zholes_size, 0, sizeof(zholes_size)); - - num_dma_physpages = 0; - efi_memmap_walk(count_dma_pages, &num_dma_physpages); - - if (max_low_pfn < max_dma) { - zones_size[ZONE_DMA] = max_low_pfn; - zholes_size[ZONE_DMA] = max_low_pfn - num_dma_physpages; - } else { - zones_size[ZONE_DMA] = max_dma; - zholes_size[ZONE_DMA] = max_dma - num_dma_physpages; - if (num_physpages > num_dma_physpages) { - zones_size[ZONE_NORMAL] = max_low_pfn - max_dma; - zholes_size[ZONE_NORMAL] = - ((max_low_pfn - max_dma) - - (num_physpages - num_dma_physpages)); - } - } - - max_gap = 0; + efi_memmap_walk(register_active_ranges, &nid); efi_memmap_walk(find_largest_hole, (u64 *)&max_gap); if (max_gap < LARGE_GAP) { vmem_map = (struct page *) 0; - free_area_init_node(0, NODE_DATA(0), zones_size, 0, - zholes_size); + free_area_init_nodes(max_zone_pfns); } else { unsigned long map_size; /* allocate virtual_mem_map */ - map_size = PAGE_ALIGN(max_low_pfn * sizeof(struct page)); + map_size = PAGE_ALIGN(ALIGN(max_low_pfn, MAX_ORDER_NR_PAGES) * + sizeof(struct page)); vmalloc_end -= map_size; vmem_map = (struct page *) vmalloc_end; efi_memmap_walk(create_mem_map_page_table, NULL); - NODE_DATA(0)->node_mem_map = vmem_map; - free_area_init_node(0, NODE_DATA(0), zones_size, - 0, zholes_size); + /* + * alloc_node_mem_map makes an adjustment for mem_map + * which isn't compatible with vmem_map. + */ + NODE_DATA(0)->node_mem_map = vmem_map + + find_min_pfn_with_active_regions(); + free_area_init_nodes(max_zone_pfns); printk("Virtual mem_map starts at 0x%p\n", mem_map); } #else /* !CONFIG_VIRTUAL_MEM_MAP */ - if (max_low_pfn < max_dma) - zones_size[ZONE_DMA] = max_low_pfn; - else { - zones_size[ZONE_DMA] = max_dma; - zones_size[ZONE_NORMAL] = max_low_pfn - max_dma; - } - free_area_init(zones_size); + add_active_range(0, 0, max_low_pfn); + free_area_init_nodes(max_zone_pfns); #endif /* !CONFIG_VIRTUAL_MEM_MAP */ zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page)); } diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index b6bcc9fa360..d497b6b0f5b 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -33,7 +33,6 @@ */ struct early_node_data { struct ia64_node_data *node_data; - pg_data_t *pgdat; unsigned long pernode_addr; unsigned long pernode_size; struct bootmem_data bootmem_data; @@ -46,6 +45,8 @@ struct early_node_data { static struct early_node_data mem_data[MAX_NUMNODES] __initdata; static nodemask_t memory_less_mask __initdata; +static pg_data_t *pgdat_list[MAX_NUMNODES]; + /* * To prevent cache aliasing effects, align per-node structures so that they * start at addresses that are strided by node number. @@ -99,7 +100,7 @@ static int __init build_node_maps(unsigned long start, unsigned long len, * acpi_boot_init() (which builds the node_to_cpu_mask array) hasn't been * called yet. Note that node 0 will also count all non-existent cpus. */ -static int __init early_nr_cpus_node(int node) +static int __meminit early_nr_cpus_node(int node) { int cpu, n = 0; @@ -114,7 +115,7 @@ static int __init early_nr_cpus_node(int node) * compute_pernodesize - compute size of pernode data * @node: the node id. */ -static unsigned long __init compute_pernodesize(int node) +static unsigned long __meminit compute_pernodesize(int node) { unsigned long pernodesize = 0, cpus; @@ -175,13 +176,13 @@ static void __init fill_pernode(int node, unsigned long pernode, pernode += PERCPU_PAGE_SIZE * cpus; pernode += node * L1_CACHE_BYTES; - mem_data[node].pgdat = __va(pernode); + pgdat_list[node] = __va(pernode); pernode += L1_CACHE_ALIGN(sizeof(pg_data_t)); mem_data[node].node_data = __va(pernode); pernode += L1_CACHE_ALIGN(sizeof(struct ia64_node_data)); - mem_data[node].pgdat->bdata = bdp; + pgdat_list[node]->bdata = bdp; pernode += L1_CACHE_ALIGN(sizeof(pg_data_t)); cpu_data = per_cpu_node_setup(cpu_data, node); @@ -268,7 +269,7 @@ static int __init find_pernode_space(unsigned long start, unsigned long len, static int __init free_node_bootmem(unsigned long start, unsigned long len, int node) { - free_bootmem_node(mem_data[node].pgdat, start, len); + free_bootmem_node(pgdat_list[node], start, len); return 0; } @@ -287,7 +288,7 @@ static void __init reserve_pernode_space(void) int node; for_each_online_node(node) { - pg_data_t *pdp = mem_data[node].pgdat; + pg_data_t *pdp = pgdat_list[node]; if (node_isset(node, memory_less_mask)) continue; @@ -307,6 +308,27 @@ static void __init reserve_pernode_space(void) } } +static void __meminit scatter_node_data(void) +{ + pg_data_t **dst; + int node; + + /* + * for_each_online_node() can't be used at here. + * node_online_map is not set for hot-added nodes at this time, + * because we are halfway through initialization of the new node's + * structures. If for_each_online_node() is used, a new node's + * pg_data_ptrs will be not initialized. Insted of using it, + * pgdat_list[] is checked. + */ + for_each_node(node) { + if (pgdat_list[node]) { + dst = LOCAL_DATA_ADDR(pgdat_list[node])->pg_data_ptrs; + memcpy(dst, pgdat_list, sizeof(pgdat_list)); + } + } +} + /** * initialize_pernode_data - fixup per-cpu & per-node pointers * @@ -317,17 +339,10 @@ static void __init reserve_pernode_space(void) */ static void __init initialize_pernode_data(void) { - pg_data_t *pgdat_list[MAX_NUMNODES]; int cpu, node; - for_each_online_node(node) - pgdat_list[node] = mem_data[node].pgdat; + scatter_node_data(); - /* Copy the pg_data_t list to each node and init the node field */ - for_each_online_node(node) { - memcpy(mem_data[node].node_data->pg_data_ptrs, pgdat_list, - sizeof(pgdat_list)); - } #ifdef CONFIG_SMP /* Set the node_data pointer for each per-cpu struct */ for (cpu = 0; cpu < NR_CPUS; cpu++) { @@ -372,7 +387,7 @@ static void __init *memory_less_node_alloc(int nid, unsigned long pernodesize) if (bestnode == -1) bestnode = anynode; - ptr = __alloc_bootmem_node(mem_data[bestnode].pgdat, pernodesize, + ptr = __alloc_bootmem_node(pgdat_list[bestnode], pernodesize, PERCPU_PAGE_SIZE, __pa(MAX_DMA_ADDRESS)); return ptr; @@ -476,7 +491,7 @@ void __init find_memory(void) pernodesize = mem_data[node].pernode_size; map = pernode + pernodesize; - init_bootmem_node(mem_data[node].pgdat, + init_bootmem_node(pgdat_list[node], map>>PAGE_SHIFT, bdp->node_boot_start>>PAGE_SHIFT, bdp->node_low_pfn); @@ -519,68 +534,6 @@ void __cpuinit *per_cpu_init(void) } #endif /* CONFIG_SMP */ -#ifdef CONFIG_VIRTUAL_MEM_MAP -static inline int find_next_valid_pfn_for_pgdat(pg_data_t *pgdat, int i) -{ - unsigned long end_address, hole_next_pfn; - unsigned long stop_address; - - end_address = (unsigned long) &vmem_map[pgdat->node_start_pfn + i]; - end_address = PAGE_ALIGN(end_address); - - stop_address = (unsigned long) &vmem_map[ - pgdat->node_start_pfn + pgdat->node_spanned_pages]; - - do { - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; - pte_t *pte; - - pgd = pgd_offset_k(end_address); - if (pgd_none(*pgd)) { - end_address += PGDIR_SIZE; - continue; - } - - pud = pud_offset(pgd, end_address); - if (pud_none(*pud)) { - end_address += PUD_SIZE; - continue; - } - - pmd = pmd_offset(pud, end_address); - if (pmd_none(*pmd)) { - end_address += PMD_SIZE; - continue; - } - - pte = pte_offset_kernel(pmd, end_address); -retry_pte: - if (pte_none(*pte)) { - end_address += PAGE_SIZE; - pte++; - if ((end_address < stop_address) && - (end_address != ALIGN(end_address, 1UL << PMD_SHIFT))) - goto retry_pte; - continue; - } - /* Found next valid vmem_map page */ - break; - } while (end_address < stop_address); - - end_address = min(end_address, stop_address); - end_address = end_address - (unsigned long) vmem_map + sizeof(struct page) - 1; - hole_next_pfn = end_address / sizeof(struct page); - return hole_next_pfn - pgdat->node_start_pfn; -} -#else -static inline int find_next_valid_pfn_for_pgdat(pg_data_t *pgdat, int i) -{ - return i + 1; -} -#endif - /** * show_mem - give short summary of memory stats * @@ -594,15 +547,16 @@ void show_mem(void) unsigned long total_present = 0; pg_data_t *pgdat; - printk("Mem-info:\n"); + printk(KERN_INFO "Mem-info:\n"); show_free_areas(); - printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); + printk(KERN_INFO "Free swap: %6ldkB\n", + nr_swap_pages<<(PAGE_SHIFT-10)); + printk(KERN_INFO "Node memory in pages:\n"); for_each_online_pgdat(pgdat) { unsigned long present; unsigned long flags; int shared = 0, cached = 0, reserved = 0; - printk("Node ID: %d\n", pgdat->node_id); pgdat_resize_lock(pgdat, &flags); present = pgdat->node_present_pages; for(i = 0; i < pgdat->node_spanned_pages; i++) { @@ -610,7 +564,8 @@ void show_mem(void) if (pfn_valid(pgdat->node_start_pfn + i)) page = pfn_to_page(pgdat->node_start_pfn + i); else { - i = find_next_valid_pfn_for_pgdat(pgdat, i) - 1; + i = vmemmap_find_next_valid_pfn(pgdat->node_id, + i) - 1; continue; } if (PageReserved(page)) @@ -625,18 +580,17 @@ void show_mem(void) total_reserved += reserved; total_cached += cached; total_shared += shared; - printk("\t%ld pages of RAM\n", present); - printk("\t%d reserved pages\n", reserved); - printk("\t%d pages shared\n", shared); - printk("\t%d pages swap cached\n", cached); + printk(KERN_INFO "Node %4d: RAM: %11ld, rsvd: %8d, " + "shrd: %10d, swpd: %10d\n", pgdat->node_id, + present, reserved, shared, cached); } - printk("%ld pages of RAM\n", total_present); - printk("%d reserved pages\n", total_reserved); - printk("%d pages shared\n", total_shared); - printk("%d pages swap cached\n", total_cached); - printk("Total of %ld pages in page table cache\n", - pgtable_quicklist_total_size()); - printk("%d free buffer pages\n", nr_free_buffer_pages()); + printk(KERN_INFO "%ld pages of RAM\n", total_present); + printk(KERN_INFO "%d reserved pages\n", total_reserved); + printk(KERN_INFO "%d pages shared\n", total_shared); + printk(KERN_INFO "%d pages swap cached\n", total_cached); + printk(KERN_INFO "Total of %ld pages in page table cache\n", + pgtable_quicklist_total_size()); + printk(KERN_INFO "%d free buffer pages\n", nr_free_buffer_pages()); } /** @@ -700,6 +654,7 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n { unsigned long end = start + len; + add_active_range(node, start >> PAGE_SHIFT, end >> PAGE_SHIFT); mem_data[node].num_physpages += len >> PAGE_SHIFT; if (start <= __pa(MAX_DMA_ADDRESS)) mem_data[node].num_dma_physpages += @@ -724,10 +679,10 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n void __init paging_init(void) { unsigned long max_dma; - unsigned long zones_size[MAX_NR_ZONES]; - unsigned long zholes_size[MAX_NR_ZONES]; unsigned long pfn_offset = 0; + unsigned long max_pfn = 0; int node; + unsigned long max_zone_pfns[MAX_NR_ZONES]; max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT; @@ -736,53 +691,45 @@ void __init paging_init(void) efi_memmap_walk(filter_rsvd_memory, count_node_pages); #ifdef CONFIG_VIRTUAL_MEM_MAP - vmalloc_end -= PAGE_ALIGN(max_low_pfn * sizeof(struct page)); + vmalloc_end -= PAGE_ALIGN(ALIGN(max_low_pfn, MAX_ORDER_NR_PAGES) * + sizeof(struct page)); vmem_map = (struct page *) vmalloc_end; efi_memmap_walk(create_mem_map_page_table, NULL); printk("Virtual mem_map starts at 0x%p\n", vmem_map); #endif for_each_online_node(node) { - memset(zones_size, 0, sizeof(zones_size)); - memset(zholes_size, 0, sizeof(zholes_size)); - num_physpages += mem_data[node].num_physpages; - - if (mem_data[node].min_pfn >= max_dma) { - /* All of this node's memory is above ZONE_DMA */ - zones_size[ZONE_NORMAL] = mem_data[node].max_pfn - - mem_data[node].min_pfn; - zholes_size[ZONE_NORMAL] = mem_data[node].max_pfn - - mem_data[node].min_pfn - - mem_data[node].num_physpages; - } else if (mem_data[node].max_pfn < max_dma) { - /* All of this node's memory is in ZONE_DMA */ - zones_size[ZONE_DMA] = mem_data[node].max_pfn - - mem_data[node].min_pfn; - zholes_size[ZONE_DMA] = mem_data[node].max_pfn - - mem_data[node].min_pfn - - mem_data[node].num_dma_physpages; - } else { - /* This node has memory in both zones */ - zones_size[ZONE_DMA] = max_dma - - mem_data[node].min_pfn; - zholes_size[ZONE_DMA] = zones_size[ZONE_DMA] - - mem_data[node].num_dma_physpages; - zones_size[ZONE_NORMAL] = mem_data[node].max_pfn - - max_dma; - zholes_size[ZONE_NORMAL] = zones_size[ZONE_NORMAL] - - (mem_data[node].num_physpages - - mem_data[node].num_dma_physpages); - } - pfn_offset = mem_data[node].min_pfn; #ifdef CONFIG_VIRTUAL_MEM_MAP NODE_DATA(node)->node_mem_map = vmem_map + pfn_offset; #endif - free_area_init_node(node, NODE_DATA(node), zones_size, - pfn_offset, zholes_size); + if (mem_data[node].max_pfn > max_pfn) + max_pfn = mem_data[node].max_pfn; } + max_zone_pfns[ZONE_DMA] = max_dma; + max_zone_pfns[ZONE_NORMAL] = max_pfn; + free_area_init_nodes(max_zone_pfns); + zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page)); } + +pg_data_t *arch_alloc_nodedata(int nid) +{ + unsigned long size = compute_pernodesize(nid); + + return kzalloc(size, GFP_KERNEL); +} + +void arch_free_nodedata(pg_data_t *pgdat) +{ + kfree(pgdat); +} + +void arch_refresh_nodedata(int update_node, pg_data_t *update_pgdat) +{ + pgdat_list[update_node] = update_pgdat; + scatter_node_data(); +} diff --git a/arch/ia64/mm/extable.c b/arch/ia64/mm/extable.c index 6d259e34f35..71c50dd8f87 100644 --- a/arch/ia64/mm/extable.c +++ b/arch/ia64/mm/extable.c @@ -5,7 +5,6 @@ * David Mosberger-Tang <davidm@hpl.hp.com> */ -#include <linux/config.h> #include <linux/sort.h> #include <asm/uaccess.h> diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c index d98ec49570b..59f3ab93761 100644 --- a/arch/ia64/mm/fault.c +++ b/arch/ia64/mm/fault.c @@ -19,6 +19,40 @@ extern void die (char *, struct pt_regs *, long); +#ifdef CONFIG_KPROBES +ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain); + +/* Hook to register for page fault notifications */ +int register_page_fault_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_register(¬ify_page_fault_chain, nb); +} + +int unregister_page_fault_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(¬ify_page_fault_chain, nb); +} + +static inline int notify_page_fault(enum die_val val, const char *str, + struct pt_regs *regs, long err, int trap, int sig) +{ + struct die_args args = { + .regs = regs, + .str = str, + .err = err, + .trapnr = trap, + .signr = sig + }; + return atomic_notifier_call_chain(¬ify_page_fault_chain, val, &args); +} +#else +static inline int notify_page_fault(enum die_val val, const char *str, + struct pt_regs *regs, long err, int trap, int sig) +{ + return NOTIFY_DONE; +} +#endif + /* * Return TRUE if ADDRESS points at a page in the kernel's mapped segment * (inside region 5, on ia64) and that page is present. @@ -84,7 +118,7 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re /* * This is to handle the kprobes on user space access instructions */ - if (notify_die(DIE_PAGE_FAULT, "page fault", regs, code, TRAP_BRKPT, + if (notify_page_fault(DIE_PAGE_FAULT, "page fault", regs, code, TRAP_BRKPT, SIGSEGV) == NOTIFY_STOP) return; @@ -112,9 +146,11 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re # error File is out of sync with <linux/mm.h>. Please update. # endif + if (((isr >> IA64_ISR_R_BIT) & 1UL) && (!(vma->vm_flags & (VM_READ | VM_WRITE)))) + goto bad_area; + mask = ( (((isr >> IA64_ISR_X_BIT) & 1UL) << VM_EXEC_BIT) - | (((isr >> IA64_ISR_W_BIT) & 1UL) << VM_WRITE_BIT) - | (((isr >> IA64_ISR_R_BIT) & 1UL) << VM_READ_BIT)); + | (((isr >> IA64_ISR_W_BIT) & 1UL) << VM_WRITE_BIT)); if ((vma->vm_flags & mask) != mask) goto bad_area; @@ -244,7 +280,7 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re out_of_memory: up_read(&mm->mmap_sem); - if (current->pid == 1) { + if (is_init(current)) { yield(); down_read(&mm->mmap_sem); goto survive; diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index 8d506710fdb..eee5c1cfbe3 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -8,7 +8,6 @@ * Feb, 2004: dynamic hugetlb page size via boot parameter */ -#include <linux/config.h> #include <linux/init.h> #include <linux/fs.h> #include <linux/mm.h> diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index 11f08001f8c..ff87a5cba39 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -4,7 +4,6 @@ * Copyright (C) 1998-2003 Hewlett-Packard Co * David Mosberger-Tang <davidm@hpl.hp.com> */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/init.h> @@ -416,6 +415,61 @@ ia64_mmu_init (void *my_cpu_data) } #ifdef CONFIG_VIRTUAL_MEM_MAP +int vmemmap_find_next_valid_pfn(int node, int i) +{ + unsigned long end_address, hole_next_pfn; + unsigned long stop_address; + pg_data_t *pgdat = NODE_DATA(node); + + end_address = (unsigned long) &vmem_map[pgdat->node_start_pfn + i]; + end_address = PAGE_ALIGN(end_address); + + stop_address = (unsigned long) &vmem_map[ + pgdat->node_start_pfn + pgdat->node_spanned_pages]; + + do { + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + pgd = pgd_offset_k(end_address); + if (pgd_none(*pgd)) { + end_address += PGDIR_SIZE; + continue; + } + + pud = pud_offset(pgd, end_address); + if (pud_none(*pud)) { + end_address += PUD_SIZE; + continue; + } + + pmd = pmd_offset(pud, end_address); + if (pmd_none(*pmd)) { + end_address += PMD_SIZE; + continue; + } + + pte = pte_offset_kernel(pmd, end_address); +retry_pte: + if (pte_none(*pte)) { + end_address += PAGE_SIZE; + pte++; + if ((end_address < stop_address) && + (end_address != ALIGN(end_address, 1UL << PMD_SHIFT))) + goto retry_pte; + continue; + } + /* Found next valid vmem_map page */ + break; + } while (end_address < stop_address); + + end_address = min(end_address, stop_address); + end_address = end_address - (unsigned long) vmem_map + sizeof(struct page) - 1; + hole_next_pfn = end_address / sizeof(struct page); + return hole_next_pfn - pgdat->node_start_pfn; +} int __init create_mem_map_page_table (u64 start, u64 end, void *arg) @@ -539,6 +593,18 @@ find_largest_hole (u64 start, u64 end, void *arg) last_end = end; return 0; } + +int __init +register_active_ranges(u64 start, u64 end, void *nid) +{ + BUG_ON(nid == NULL); + BUG_ON(*(unsigned long *)nid >= MAX_NUMNODES); + + add_active_range(*(unsigned long *)nid, + __pa(start) >> PAGE_SHIFT, + __pa(end) >> PAGE_SHIFT); + return 0; +} #endif /* CONFIG_VIRTUAL_MEM_MAP */ static int __init @@ -652,7 +718,7 @@ void online_page(struct page *page) num_physpages++; } -int add_memory(u64 start, u64 size) +int arch_add_memory(int nid, u64 start, u64 size) { pg_data_t *pgdat; struct zone *zone; @@ -660,7 +726,7 @@ int add_memory(u64 start, u64 size) unsigned long nr_pages = size >> PAGE_SHIFT; int ret; - pgdat = NODE_DATA(0); + pgdat = NODE_DATA(nid); zone = pgdat->node_zones + ZONE_NORMAL; ret = __add_pages(zone, start_pfn, nr_pages); @@ -671,7 +737,6 @@ int add_memory(u64 start, u64 size) return ret; } -EXPORT_SYMBOL_GPL(add_memory); int remove_memory(u64 start, u64 size) { diff --git a/arch/ia64/mm/ioremap.c b/arch/ia64/mm/ioremap.c index 07bd02b6c37..4280c074d64 100644 --- a/arch/ia64/mm/ioremap.c +++ b/arch/ia64/mm/ioremap.c @@ -32,7 +32,7 @@ ioremap (unsigned long offset, unsigned long size) */ attr = kern_mem_attribute(offset, size); if (attr & EFI_MEMORY_WB) - return phys_to_virt(offset); + return (void __iomem *) phys_to_virt(offset); else if (attr & EFI_MEMORY_UC) return __ioremap(offset, size); @@ -43,7 +43,7 @@ ioremap (unsigned long offset, unsigned long size) gran_base = GRANULEROUNDDOWN(offset); gran_size = GRANULEROUNDUP(offset + size) - gran_base; if (efi_mem_attribute(gran_base, gran_size) & EFI_MEMORY_WB) - return phys_to_virt(offset); + return (void __iomem *) phys_to_virt(offset); return __ioremap(offset, size); } @@ -53,7 +53,7 @@ void __iomem * ioremap_nocache (unsigned long offset, unsigned long size) { if (kern_mem_attribute(offset, size) & EFI_MEMORY_WB) - return 0; + return NULL; return __ioremap(offset, size); } diff --git a/arch/ia64/mm/numa.c b/arch/ia64/mm/numa.c index 4e5c8b36ad9..64e4c21f311 100644 --- a/arch/ia64/mm/numa.c +++ b/arch/ia64/mm/numa.c @@ -10,7 +10,6 @@ * 2002/08/07 Erich Focht <efocht@ess.nec.de> */ -#include <linux/config.h> #include <linux/cpu.h> #include <linux/kernel.h> #include <linux/mm.h> diff --git a/arch/ia64/mm/tlb.c b/arch/ia64/mm/tlb.c index 4dbbca0b5e9..ffad7624436 100644 --- a/arch/ia64/mm/tlb.c +++ b/arch/ia64/mm/tlb.c @@ -11,7 +11,6 @@ * Rohit Seth <rohit.seth@intel.com> * Ken Chen <kenneth.w.chen@intel.com> */ -#include <linux/config.h> #include <linux/module.h> #include <linux/init.h> #include <linux/kernel.h> diff --git a/arch/ia64/oprofile/perfmon.c b/arch/ia64/oprofile/perfmon.c index b7975a469fb..bc41dd32fec 100644 --- a/arch/ia64/oprofile/perfmon.c +++ b/arch/ia64/oprofile/perfmon.c @@ -8,7 +8,6 @@ */ #include <linux/kernel.h> -#include <linux/config.h> #include <linux/oprofile.h> #include <linux/sched.h> #include <asm/perfmon.h> diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c index 61dd8608da4..15c7c670da3 100644 --- a/arch/ia64/pci/pci.c +++ b/arch/ia64/pci/pci.c @@ -10,7 +10,6 @@ * * Note: Above list of copyright holders is incomplete... */ -#include <linux/config.h> #include <linux/acpi.h> #include <linux/types.h> @@ -563,12 +562,13 @@ pcibios_enable_device (struct pci_dev *dev, int mask) void pcibios_disable_device (struct pci_dev *dev) { - acpi_pci_irq_disable(dev); + if (dev->is_enabled) + acpi_pci_irq_disable(dev); } void pcibios_align_resource (void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { } @@ -602,8 +602,6 @@ pci_mmap_page_range (struct pci_dev *dev, struct vm_area_struct *vma, * Leave vm_pgoff as-is, the PCI space address is the physical * address on this platform. */ - vma->vm_flags |= (VM_SHM | VM_RESERVED | VM_IO); - if (write_combine && efi_range_is_wc(vma->vm_start, vma->vm_end - vma->vm_start)) vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); @@ -653,7 +651,7 @@ pci_mmap_legacy_page_range(struct pci_bus *bus, struct vm_area_struct *vma) * Avoid attribute aliasing. See Documentation/ia64/aliasing.txt * for more details. */ - if (!valid_mmap_phys_addr_range(vma->vm_pgoff << PAGE_SHIFT, size)) + if (!valid_mmap_phys_addr_range(vma->vm_pgoff, size)) return -EINVAL; prot = phys_mem_access_prot(NULL, vma->vm_pgoff, size, vma->vm_page_prot); @@ -666,7 +664,6 @@ pci_mmap_legacy_page_range(struct pci_bus *bus, struct vm_area_struct *vma) vma->vm_pgoff += (unsigned long)addr >> PAGE_SHIFT; vma->vm_page_prot = prot; - vma->vm_flags |= (VM_SHM | VM_RESERVED | VM_IO); if (remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff, size, vma->vm_page_prot)) diff --git a/arch/ia64/sn/kernel/bte.c b/arch/ia64/sn/kernel/bte.c index e952ef4f6d9..7f73ad4408a 100644 --- a/arch/ia64/sn/kernel/bte.c +++ b/arch/ia64/sn/kernel/bte.c @@ -6,7 +6,6 @@ * Copyright (c) 2000-2006 Silicon Graphics, Inc. All Rights Reserved. */ -#include <linux/config.h> #include <linux/module.h> #include <asm/sn/nodepda.h> #include <asm/sn/addrs.h> @@ -278,8 +277,7 @@ bte_result_t bte_unaligned_copy(u64 src, u64 dest, u64 len, u64 mode) } /* temporary buffer used during unaligned transfers */ - bteBlock_unaligned = kmalloc(len + 3 * L1_CACHE_BYTES, - GFP_KERNEL | GFP_DMA); + bteBlock_unaligned = kmalloc(len + 3 * L1_CACHE_BYTES, GFP_KERNEL); if (bteBlock_unaligned == NULL) { return BTEFAIL_NOTAVAIL; } diff --git a/arch/ia64/sn/kernel/huberror.c b/arch/ia64/sn/kernel/huberror.c index 56ab6bae00e..96fb81e6321 100644 --- a/arch/ia64/sn/kernel/huberror.c +++ b/arch/ia64/sn/kernel/huberror.c @@ -178,7 +178,7 @@ void hubiio_crb_error_handler(struct hubdev_info *hubdev_info) */ void hub_error_init(struct hubdev_info *hubdev_info) { - if (request_irq(SGI_II_ERROR, (void *)hub_eint_handler, SA_SHIRQ, + if (request_irq(SGI_II_ERROR, (void *)hub_eint_handler, IRQF_SHARED, "SN_hub_error", (void *)hubdev_info)) printk("hub_error_init: Failed to request_irq for 0x%p\n", hubdev_info); @@ -196,7 +196,7 @@ void hub_error_init(struct hubdev_info *hubdev_info) void ice_error_init(struct hubdev_info *hubdev_info) { if (request_irq - (SGI_TIO_ERROR, (void *)hub_eint_handler, SA_SHIRQ, "SN_TIO_error", + (SGI_TIO_ERROR, (void *)hub_eint_handler, IRQF_SHARED, "SN_TIO_error", (void *)hubdev_info)) printk("ice_error_init: request_irq() error hubdev_info 0x%p\n", hubdev_info); diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c index dc8e2b69671..7bb6ad188ba 100644 --- a/arch/ia64/sn/kernel/irq.c +++ b/arch/ia64/sn/kernel/irq.c @@ -27,7 +27,7 @@ static void unregister_intr_pda(struct sn_irq_info *sn_irq_info); int sn_force_interrupt_flag = 1; extern int sn_ioif_inited; struct list_head **sn_irq_lh; -static spinlock_t sn_irq_info_lock = SPIN_LOCK_UNLOCKED; /* non-IRQ lock */ +static DEFINE_SPINLOCK(sn_irq_info_lock); /* non-IRQ lock */ u64 sn_intr_alloc(nasid_t local_nasid, int local_widget, struct sn_irq_info *sn_irq_info, @@ -225,8 +225,8 @@ void sn_irq_init(void) ia64_last_device_vector = IA64_SN2_LAST_DEVICE_VECTOR; for (i = 0; i < NR_IRQS; i++) { - if (base_desc[i].handler == &no_irq_type) { - base_desc[i].handler = &irq_type_sn; + if (base_desc[i].chip == &no_irq_type) { + base_desc[i].chip = &irq_type_sn; } } } diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index 93577abae36..5f2dcba7fa8 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -6,14 +6,13 @@ * Copyright (C) 1999,2001-2006 Silicon Graphics, Inc. All rights reserved. */ -#include <linux/config.h> #include <linux/module.h> #include <linux/init.h> #include <linux/delay.h> #include <linux/kernel.h> #include <linux/kdev_t.h> #include <linux/string.h> -#include <linux/tty.h> +#include <linux/screen_info.h> #include <linux/console.h> #include <linux/timex.h> #include <linux/sched.h> @@ -458,7 +457,7 @@ void __init sn_setup(char **cmdline_p) * support here so we don't have to listen to failed keyboard probe * messages. */ - if (version <= 0x0209 && acpi_kbd_controller_present) { + if (is_shub1() && version <= 0x0209 && acpi_kbd_controller_present) { printk(KERN_INFO "Disabling legacy keyboard support as prom " "is too old and doesn't provide FADT\n"); acpi_kbd_controller_present = 0; @@ -566,7 +565,7 @@ static void __init sn_init_pdas(char **cmdline_p) * Also sets up a few fields in the nodepda. Also known as * platform_cpu_init() by the ia64 machvec code. */ -void __init sn_cpu_init(void) +void __cpuinit sn_cpu_init(void) { int cpuid; int cpuphyid; @@ -577,7 +576,8 @@ void __init sn_cpu_init(void) int i; static int wars_have_been_checked; - if (smp_processor_id() == 0 && IS_MEDUSA()) { + cpuid = smp_processor_id(); + if (cpuid == 0 && IS_MEDUSA()) { if (ia64_sn_is_fake_prom()) sn_prom_type = 2; else @@ -597,6 +597,12 @@ void __init sn_cpu_init(void) sn_hub_info->as_shift = sn_hub_info->nasid_shift - 2; /* + * Don't check status. The SAL call is not supported on all PROMs + * but a failure is harmless. + */ + (void) ia64_sn_set_cpu_number(cpuid); + + /* * The boot cpu makes this call again after platform initialization is * complete. */ @@ -607,7 +613,6 @@ void __init sn_cpu_init(void) if (ia64_sn_get_prom_feature_set(i, &sn_prom_features[i]) != 0) break; - cpuid = smp_processor_id(); cpuphyid = get_sapicid(); if (ia64_sn_get_sapic_info(cpuphyid, &nasid, &subnode, &slice)) diff --git a/arch/ia64/sn/kernel/sn2/prominfo_proc.c b/arch/ia64/sn/kernel/sn2/prominfo_proc.c index 6ae276d5d50..4dcce3d0e04 100644 --- a/arch/ia64/sn/kernel/sn2/prominfo_proc.c +++ b/arch/ia64/sn/kernel/sn2/prominfo_proc.c @@ -8,7 +8,6 @@ * Module to export the system's Firmware Interface Tables, including * PROM revision numbers and banners, in /proc */ -#include <linux/config.h> #include <linux/module.h> #include <linux/slab.h> #include <linux/proc_fs.h> diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c index 9a8a29339d2..b632b9c1e3b 100644 --- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c +++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c @@ -32,9 +32,10 @@ #include <linux/cpumask.h> #include <linux/smp_lock.h> #include <linux/nodemask.h> +#include <linux/smp.h> + #include <asm/processor.h> #include <asm/topology.h> -#include <asm/smp.h> #include <asm/semaphore.h> #include <asm/uaccess.h> #include <asm/sal.h> diff --git a/arch/ia64/sn/kernel/sn2/sn_proc_fs.c b/arch/ia64/sn/kernel/sn2/sn_proc_fs.c index 5100261310f..43ddc2eccb9 100644 --- a/arch/ia64/sn/kernel/sn2/sn_proc_fs.c +++ b/arch/ia64/sn/kernel/sn2/sn_proc_fs.c @@ -5,7 +5,6 @@ * * Copyright (C) 2000-2005 Silicon Graphics, Inc. All rights reserved. */ -#include <linux/config.h> #ifdef CONFIG_PROC_FS #include <linux/proc_fs.h> diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c index 8255a9be463..1f3540826e6 100644 --- a/arch/ia64/sn/kernel/xpc_channel.c +++ b/arch/ia64/sn/kernel/xpc_channel.c @@ -202,7 +202,7 @@ xpc_setup_infrastructure(struct xpc_partition *part) init_waitqueue_head(&part->channel_mgr_wq); sprintf(part->IPI_owner, "xpc%02d", partid); - ret = request_irq(SGI_XPC_NOTIFY, xpc_notify_IRQ_handler, SA_SHIRQ, + ret = request_irq(SGI_XPC_NOTIFY, xpc_notify_IRQ_handler, IRQF_SHARED, part->IPI_owner, (void *) (u64) partid); if (ret != 0) { dev_err(xpc_chan, "can't register NOTIFY IRQ handler, " @@ -279,8 +279,8 @@ xpc_pull_remote_cachelines(struct xpc_partition *part, void *dst, return part->reason; } - bte_ret = xp_bte_copy((u64) src, (u64) ia64_tpa((u64) dst), - (u64) cnt, (BTE_NORMAL | BTE_WACQUIRE), NULL); + bte_ret = xp_bte_copy((u64) src, (u64) dst, (u64) cnt, + (BTE_NORMAL | BTE_WACQUIRE), NULL); if (bte_ret == BTE_SUCCESS) { return xpcSuccess; } diff --git a/arch/ia64/sn/kernel/xpc_main.c b/arch/ia64/sn/kernel/xpc_main.c index 99b123a6421..4d026f9dd98 100644 --- a/arch/ia64/sn/kernel/xpc_main.c +++ b/arch/ia64/sn/kernel/xpc_main.c @@ -480,7 +480,7 @@ xpc_activating(void *__partid) partid_t partid = (u64) __partid; struct xpc_partition *part = &xpc_partitions[partid]; unsigned long irq_flags; - struct sched_param param = { sched_priority: MAX_RT_PRIO - 1 }; + struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; int ret; @@ -1052,6 +1052,8 @@ xpc_do_exit(enum xpc_retval reason) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); } @@ -1212,24 +1214,20 @@ xpc_init(void) partid_t partid; struct xpc_partition *part; pid_t pid; + size_t buf_size; if (!ia64_platform_is("sn2")) { return -ENODEV; } - /* - * xpc_remote_copy_buffer is used as a temporary buffer for bte_copy'ng - * various portions of a partition's reserved page. Its size is based - * on the size of the reserved page header and part_nasids mask. So we - * need to ensure that the other items will fit as well. - */ - if (XPC_RP_VARS_SIZE > XPC_RP_HEADER_SIZE + XP_NASID_MASK_BYTES) { - dev_err(xpc_part, "xpc_remote_copy_buffer is not big enough\n"); - return -EPERM; - } - DBUG_ON((u64) xpc_remote_copy_buffer != - L1_CACHE_ALIGN((u64) xpc_remote_copy_buffer)); + + buf_size = max(XPC_RP_VARS_SIZE, + XPC_RP_HEADER_SIZE + XP_NASID_MASK_BYTES); + xpc_remote_copy_buffer = xpc_kmalloc_cacheline_aligned(buf_size, + GFP_KERNEL, &xpc_remote_copy_buffer_base); + if (xpc_remote_copy_buffer == NULL) + return -ENOMEM; snprintf(xpc_part->bus_id, BUS_ID_SIZE, "part"); snprintf(xpc_chan->bus_id, BUS_ID_SIZE, "chan"); @@ -1293,6 +1291,8 @@ xpc_init(void) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); return -EBUSY; } @@ -1311,6 +1311,8 @@ xpc_init(void) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); return -EBUSY; } @@ -1362,6 +1364,8 @@ xpc_init(void) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); return -EBUSY; } diff --git a/arch/ia64/sn/kernel/xpc_partition.c b/arch/ia64/sn/kernel/xpc_partition.c index 2a89cfce495..57c723f5cba 100644 --- a/arch/ia64/sn/kernel/xpc_partition.c +++ b/arch/ia64/sn/kernel/xpc_partition.c @@ -71,19 +71,15 @@ struct xpc_partition xpc_partitions[XP_MAX_PARTITIONS + 1]; * Generic buffer used to store a local copy of portions of a remote * partition's reserved page (either its header and part_nasids mask, * or its vars). - * - * xpc_discovery runs only once and is a seperate thread that is - * very likely going to be processing in parallel with receiving - * interrupts. */ -char ____cacheline_aligned xpc_remote_copy_buffer[XPC_RP_HEADER_SIZE + - XP_NASID_MASK_BYTES]; +char *xpc_remote_copy_buffer; +void *xpc_remote_copy_buffer_base; /* * Guarantee that the kmalloc'd memory is cacheline aligned. */ -static void * +void * xpc_kmalloc_cacheline_aligned(size_t size, gfp_t flags, void **base) { /* see if kmalloc will give us cachline aligned memory by default */ @@ -148,7 +144,7 @@ xpc_get_rsvd_page_pa(int nasid) } } - bte_res = xp_bte_copy(rp_pa, ia64_tpa(buf), buf_len, + bte_res = xp_bte_copy(rp_pa, buf, buf_len, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bte_res != BTE_SUCCESS) { dev_dbg(xpc_part, "xp_bte_copy failed %i\n", bte_res); @@ -447,7 +443,7 @@ xpc_check_remote_hb(void) /* pull the remote_hb cache line */ bres = xp_bte_copy(part->remote_vars_pa, - ia64_tpa((u64) remote_vars), + (u64) remote_vars, XPC_RP_VARS_SIZE, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bres != BTE_SUCCESS) { @@ -498,8 +494,7 @@ xpc_get_remote_rp(int nasid, u64 *discovered_nasids, /* pull over the reserved page header and part_nasids mask */ - - bres = xp_bte_copy(*remote_rp_pa, ia64_tpa((u64) remote_rp), + bres = xp_bte_copy(*remote_rp_pa, (u64) remote_rp, XPC_RP_HEADER_SIZE + xp_nasid_mask_bytes, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bres != BTE_SUCCESS) { @@ -554,11 +549,8 @@ xpc_get_remote_vars(u64 remote_vars_pa, struct xpc_vars *remote_vars) return xpcVarsNotSet; } - /* pull over the cross partition variables */ - - bres = xp_bte_copy(remote_vars_pa, ia64_tpa((u64) remote_vars), - XPC_RP_VARS_SIZE, + bres = xp_bte_copy(remote_vars_pa, (u64) remote_vars, XPC_RP_VARS_SIZE, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bres != BTE_SUCCESS) { return xpc_map_bte_errors(bres); @@ -1239,7 +1231,7 @@ xpc_initiate_partid_to_nasids(partid_t partid, void *nasid_mask) part_nasid_pa = (u64) XPC_RP_PART_NASIDS(part->remote_rp_pa); - bte_res = xp_bte_copy(part_nasid_pa, ia64_tpa((u64) nasid_mask), + bte_res = xp_bte_copy(part_nasid_pa, (u64) nasid_mask, xp_nasid_mask_bytes, (BTE_NOTIFY | BTE_WACQUIRE), NULL); return xpc_map_bte_errors(bte_res); diff --git a/arch/ia64/sn/kernel/xpnet.c b/arch/ia64/sn/kernel/xpnet.c index e5c6d3c0a8e..007703c494a 100644 --- a/arch/ia64/sn/kernel/xpnet.c +++ b/arch/ia64/sn/kernel/xpnet.c @@ -22,7 +22,6 @@ */ -#include <linux/config.h> #include <linux/module.h> #include <linux/kernel.h> #include <linux/pci.h> diff --git a/arch/ia64/sn/pci/pcibr/pcibr_provider.c b/arch/ia64/sn/pci/pcibr/pcibr_provider.c index ab1211ef017..838c93c9a16 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_provider.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_provider.c @@ -139,7 +139,7 @@ pcibr_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont * register the bridge's error interrupt handler */ if (request_irq(SGI_PCIASIC_ERROR, (void *)pcibr_error_intr_handler, - SA_SHIRQ, "PCIBR error", (void *)(soft))) { + IRQF_SHARED, "PCIBR error", (void *)(soft))) { printk(KERN_WARNING "pcibr cannot allocate interrupt for error handler\n"); } diff --git a/arch/ia64/sn/pci/tioca_provider.c b/arch/ia64/sn/pci/tioca_provider.c index 20de72791b9..c36b0f5affb 100644 --- a/arch/ia64/sn/pci/tioca_provider.c +++ b/arch/ia64/sn/pci/tioca_provider.c @@ -595,7 +595,7 @@ tioca_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont /* sanity check prom rev */ - if (sn_sal_rev() < 0x0406) { + if (is_shub1() && sn_sal_rev() < 0x0406) { printk (KERN_ERR "%s: SGI prom rev 4.06 or greater required " "for tioca support\n", __FUNCTION__); @@ -646,7 +646,7 @@ tioca_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont if (request_irq(SGI_TIOCA_ERROR, tioca_error_intr_handler, - SA_SHIRQ, "TIOCA error", (void *)tioca_common)) + IRQF_SHARED, "TIOCA error", (void *)tioca_common)) printk(KERN_WARNING "%s: Unable to get irq %d. " "Error interrupts won't be routed for TIOCA bus %d\n", diff --git a/arch/ia64/sn/pci/tioce_provider.c b/arch/ia64/sn/pci/tioce_provider.c index 2d7948567eb..af7171adcd2 100644 --- a/arch/ia64/sn/pci/tioce_provider.c +++ b/arch/ia64/sn/pci/tioce_provider.c @@ -74,7 +74,7 @@ tioce_mmr_war_pre(struct tioce_kernel *kern, void *mmr_addr) else mmr_war_offset = 0x158; - readq_relaxed((void *)(mmr_base + mmr_war_offset)); + readq_relaxed((void __iomem *)(mmr_base + mmr_war_offset)); } } @@ -92,8 +92,8 @@ tioce_mmr_war_post(struct tioce_kernel *kern, void *mmr_addr) if (mmr_offset < 0x45000) { if (mmr_offset == 0x100) - readq_relaxed((void *)(mmr_base + 0x38)); - readq_relaxed((void *)(mmr_base + 0xb050)); + readq_relaxed((void __iomem *)(mmr_base + 0x38)); + readq_relaxed((void __iomem *)(mmr_base + 0xb050)); } } @@ -1027,7 +1027,7 @@ tioce_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont if (request_irq(SGI_PCIASIC_ERROR, tioce_error_intr_handler, - SA_SHIRQ, "TIOCE error", (void *)tioce_common)) + IRQF_SHARED, "TIOCE error", (void *)tioce_common)) printk(KERN_WARNING "%s: Unable to get irq %d. " "Error interrupts won't be routed for " |