diff options
| author | Ingo Molnar <mingo@elte.hu> | 2009-07-04 11:00:38 +0200 |
|---|---|---|
| committer | Ingo Molnar <mingo@elte.hu> | 2009-07-04 11:00:42 +0200 |
| commit | d7e57676e3ed7ab9b2c7c4bcb7873e51eacbdb84 (patch) | |
| tree | f7433f38cd407a0c35a8cbf2b7e3fd756087bce7 /arch/x86/kernel | |
| parent | feaa0457ec8351cae855edc9a3052ac49322538e (diff) | |
| parent | 746a99a5af60ee676afa2ba469ccd1373493c7e7 (diff) | |
Merge branch 'linus' into x86/cleanups
Merge reason: We were on an older pre-rc1 base, move to almost-rc2.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Diffstat (limited to 'arch/x86/kernel')
42 files changed, 946 insertions, 773 deletions
diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 631086159c5..6b8ca3a0285 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -44,11 +44,7 @@ static int __initdata acpi_force = 0; u32 acpi_rsdt_forced; -#ifdef CONFIG_ACPI -int acpi_disabled = 0; -#else -int acpi_disabled = 1; -#endif +int acpi_disabled; EXPORT_SYMBOL(acpi_disabled); #ifdef CONFIG_X86_64 @@ -122,72 +118,6 @@ void __init __acpi_unmap_table(char *map, unsigned long size) early_iounmap(map, size); } -#ifdef CONFIG_PCI_MMCONFIG - -static int acpi_mcfg_64bit_base_addr __initdata = FALSE; - -/* The physical address of the MMCONFIG aperture. Set from ACPI tables. */ -struct acpi_mcfg_allocation *pci_mmcfg_config; -int pci_mmcfg_config_num; - -static int __init acpi_mcfg_oem_check(struct acpi_table_mcfg *mcfg) -{ - if (!strcmp(mcfg->header.oem_id, "SGI")) - acpi_mcfg_64bit_base_addr = TRUE; - - return 0; -} - -int __init acpi_parse_mcfg(struct acpi_table_header *header) -{ - struct acpi_table_mcfg *mcfg; - unsigned long i; - int config_size; - - if (!header) - return -EINVAL; - - mcfg = (struct acpi_table_mcfg *)header; - - /* how many config structures do we have */ - pci_mmcfg_config_num = 0; - i = header->length - sizeof(struct acpi_table_mcfg); - while (i >= sizeof(struct acpi_mcfg_allocation)) { - ++pci_mmcfg_config_num; - i -= sizeof(struct acpi_mcfg_allocation); - }; - if (pci_mmcfg_config_num == 0) { - printk(KERN_ERR PREFIX "MMCONFIG has no entries\n"); - return -ENODEV; - } - - config_size = pci_mmcfg_config_num * sizeof(*pci_mmcfg_config); - pci_mmcfg_config = kmalloc(config_size, GFP_KERNEL); - if (!pci_mmcfg_config) { - printk(KERN_WARNING PREFIX - "No memory for MCFG config tables\n"); - return -ENOMEM; - } - - memcpy(pci_mmcfg_config, &mcfg[1], config_size); - - acpi_mcfg_oem_check(mcfg); - - for (i = 0; i < pci_mmcfg_config_num; ++i) { - if ((pci_mmcfg_config[i].address > 0xFFFFFFFF) && - !acpi_mcfg_64bit_base_addr) { - printk(KERN_ERR PREFIX - "MMCONFIG not in low 4GB of memory\n"); - kfree(pci_mmcfg_config); - pci_mmcfg_config_num = 0; - return -ENODEV; - } - } - - return 0; -} -#endif /* CONFIG_PCI_MMCONFIG */ - #ifdef CONFIG_X86_LOCAL_APIC static int __init acpi_parse_madt(struct acpi_table_header *table) { @@ -1519,14 +1449,6 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = { }, { .callback = force_acpi_ht, - .ident = "ASUS P4B266", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), - DMI_MATCH(DMI_BOARD_NAME, "P4B266"), - }, - }, - { - .callback = force_acpi_ht, .ident = "ASUS P2B-DS", .matches = { DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c index bbbe4bbb6f3..8c44c232efc 100644 --- a/arch/x86/kernel/acpi/cstate.c +++ b/arch/x86/kernel/acpi/cstate.c @@ -34,12 +34,22 @@ void acpi_processor_power_init_bm_check(struct acpi_processor_flags *flags, flags->bm_check = 1; else if (c->x86_vendor == X86_VENDOR_INTEL) { /* - * Today all CPUs that support C3 share cache. - * TBD: This needs to look at cache shared map, once - * multi-core detection patch makes to the base. + * Today all MP CPUs that support C3 share cache. + * And caches should not be flushed by software while + * entering C3 type state. */ flags->bm_check = 1; } + + /* + * On all recent Intel platforms, ARB_DISABLE is a nop. + * So, set bm_control to zero to indicate that ARB_DISABLE + * is not required while entering C3 type state on + * P4, Core and beyond CPUs + */ + if (c->x86_vendor == X86_VENDOR_INTEL && + (c->x86 > 0x6 || (c->x86 == 6 && c->x86_model >= 14))) + flags->bm_control = 0; } EXPORT_SYMBOL(acpi_processor_power_init_bm_check); diff --git a/arch/x86/kernel/acpi/processor.c b/arch/x86/kernel/acpi/processor.c index 7c074eec39f..d296f4a195c 100644 --- a/arch/x86/kernel/acpi/processor.c +++ b/arch/x86/kernel/acpi/processor.c @@ -72,6 +72,7 @@ static void init_intel_pdc(struct acpi_processor *pr, struct cpuinfo_x86 *c) return; } + /* Initialize _PDC data based on the CPU vendor */ void arch_acpi_processor_init_pdc(struct acpi_processor *pr) { @@ -85,3 +86,15 @@ void arch_acpi_processor_init_pdc(struct acpi_processor *pr) } EXPORT_SYMBOL(arch_acpi_processor_init_pdc); + +void arch_acpi_processor_cleanup_pdc(struct acpi_processor *pr) +{ + if (pr->pdc) { + kfree(pr->pdc->pointer->buffer.pointer); + kfree(pr->pdc->pointer); + kfree(pr->pdc); + pr->pdc = NULL; + } +} + +EXPORT_SYMBOL(arch_acpi_processor_cleanup_pdc); diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 1c60554537c..9372f0406ad 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -434,6 +434,16 @@ static void iommu_flush_tlb(struct amd_iommu *iommu, u16 domid) iommu_queue_inv_iommu_pages(iommu, address, domid, 0, 1); } +/* Flush the whole IO/TLB for a given protection domain - including PDE */ +static void iommu_flush_tlb_pde(struct amd_iommu *iommu, u16 domid) +{ + u64 address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS; + + INC_STATS_COUNTER(domain_flush_single); + + iommu_queue_inv_iommu_pages(iommu, address, domid, 1, 1); +} + /* * This function is used to flush the IO/TLB for a given protection domain * on every IOMMU in the system @@ -1078,7 +1088,13 @@ static void attach_device(struct amd_iommu *iommu, amd_iommu_pd_table[devid] = domain; write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); + /* + * We might boot into a crash-kernel here. The crashed kernel + * left the caches in the IOMMU dirty. So we have to flush + * here to evict all dirty stuff. + */ iommu_queue_inv_dev_entry(iommu, devid); + iommu_flush_tlb_pde(iommu, domain->id); } /* diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 238989ec077..10b2accd12e 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -260,6 +260,14 @@ static void iommu_enable(struct amd_iommu *iommu) static void iommu_disable(struct amd_iommu *iommu) { + /* Disable command buffer */ + iommu_feature_disable(iommu, CONTROL_CMDBUF_EN); + + /* Disable event logging and event interrupts */ + iommu_feature_disable(iommu, CONTROL_EVT_INT_EN); + iommu_feature_disable(iommu, CONTROL_EVT_LOG_EN); + + /* Disable IOMMU hardware itself */ iommu_feature_disable(iommu, CONTROL_IOMMU_EN); } @@ -478,6 +486,10 @@ static void iommu_enable_event_buffer(struct amd_iommu *iommu) memcpy_toio(iommu->mmio_base + MMIO_EVT_BUF_OFFSET, &entry, sizeof(entry)); + /* set head and tail to zero manually */ + writel(0x00, iommu->mmio_base + MMIO_EVT_HEAD_OFFSET); + writel(0x00, iommu->mmio_base + MMIO_EVT_TAIL_OFFSET); + iommu_feature_enable(iommu, CONTROL_EVT_LOG_EN); } @@ -1042,6 +1054,7 @@ static void enable_iommus(void) struct amd_iommu *iommu; for_each_iommu(iommu) { + iommu_disable(iommu); iommu_set_device_table(iommu); iommu_enable_command_buffer(iommu); iommu_enable_event_buffer(iommu); @@ -1066,12 +1079,6 @@ static void disable_iommus(void) static int amd_iommu_resume(struct sys_device *dev) { - /* - * Disable IOMMUs before reprogramming the hardware registers. - * IOMMU is still enabled from the resume kernel. - */ - disable_iommus(); - /* re-load the hardware */ enable_iommus(); @@ -1079,8 +1086,8 @@ static int amd_iommu_resume(struct sys_device *dev) * we have to flush after the IOMMUs are enabled because a * disabled IOMMU will never execute the commands we send */ - amd_iommu_flush_all_domains(); amd_iommu_flush_all_devices(); + amd_iommu_flush_all_domains(); return 0; } @@ -1273,6 +1280,11 @@ free: goto out; } +void amd_iommu_shutdown(void) +{ + disable_iommus(); +} + /**************************************************************************** * * Early detect code. This code runs at IOMMU detection time in the DMA diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index ef8d9290c7e..4d0216fcb36 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -462,7 +462,8 @@ static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) static void __ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { - union entry_union eu; + union entry_union eu = {{0, 0}}; + eu.entry = e; io_apic_write(apic, 0x11 + 2*pin, eu.w2); io_apic_write(apic, 0x10 + 2*pin, eu.w1); @@ -1413,6 +1414,9 @@ int setup_ioapic_entry(int apic_id, int irq, irte.vector = vector; irte.dest_id = IRTE_DEST(destination); + /* Set source-id of interrupt request */ + set_ioapic_sid(&irte, apic_id); + modify_irte(irq, &irte); ir_entry->index2 = (index >> 15) & 0x1; @@ -2003,7 +2007,9 @@ void disable_IO_APIC(void) /* * Use virtual wire A mode when interrupt remapping is enabled. */ - disconnect_bsp_APIC(!intr_remapping_enabled && ioapic_i8259.pin != -1); + if (cpu_has_apic) + disconnect_bsp_APIC(!intr_remapping_enabled && + ioapic_i8259.pin != -1); } #ifdef CONFIG_X86_32 @@ -3287,6 +3293,9 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms irte.vector = cfg->vector; irte.dest_id = IRTE_DEST(dest); + /* Set source-id of interrupt request */ + set_msi_sid(&irte, pdev); + modify_irte(irq, &irte); msg->address_hi = MSI_ADDR_BASE_HI; @@ -3567,7 +3576,7 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) #endif /* CONFIG_SMP */ -struct irq_chip dmar_msi_type = { +static struct irq_chip dmar_msi_type = { .name = "DMAR_MSI", .unmask = dmar_msi_unmask, .mask = dmar_msi_mask, diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c index 440a8bccd91..0c0182cc947 100644 --- a/arch/x86/kernel/apic/probe_32.c +++ b/arch/x86/kernel/apic/probe_32.c @@ -20,23 +20,12 @@ #include <asm/apic.h> #include <asm/setup.h> -#include <linux/threads.h> -#include <linux/cpumask.h> -#include <asm/mpspec.h> -#include <asm/fixmap.h> -#include <asm/apicdef.h> -#include <linux/kernel.h> -#include <linux/string.h> #include <linux/smp.h> -#include <linux/init.h> #include <asm/ipi.h> -#include <linux/smp.h> -#include <linux/init.h> #include <linux/interrupt.h> #include <asm/acpi.h> #include <asm/e820.h> -#include <asm/setup.h> #ifdef CONFIG_HOTPLUG_CPU #define DEFAULT_SEND_IPI (1) diff --git a/arch/x86/kernel/apic/summit_32.c b/arch/x86/kernel/apic/summit_32.c index 344eee4ac0a..eafdfbd1ea9 100644 --- a/arch/x86/kernel/apic/summit_32.c +++ b/arch/x86/kernel/apic/summit_32.c @@ -44,7 +44,6 @@ #include <asm/ipi.h> #include <linux/kernel.h> #include <linux/string.h> -#include <linux/init.h> #include <linux/gfp.h> #include <linux/smp.h> diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index e5b27d8f1b4..28e5f595604 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -258,13 +258,15 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c) { #ifdef CONFIG_X86_HT unsigned bits; + int cpu = smp_processor_id(); bits = c->x86_coreid_bits; - /* Low order bits define the core id (index of core in socket) */ c->cpu_core_id = c->initial_apicid & ((1 << bits)-1); /* Convert the initial APIC ID into the socket ID */ c->phys_proc_id = c->initial_apicid >> bits; + /* use socket ID also for last level cache */ + per_cpu(cpu_llc_id, cpu) = c->phys_proc_id; #endif } diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 9fa33886c0d..f1961c07af9 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -108,7 +108,7 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { /* data */ [GDT_ENTRY_APMBIOS_BASE+2] = { { { 0x0000ffff, 0x00409200 } } }, - [GDT_ENTRY_ESPFIX_SS] = { { { 0x00000000, 0x00c09200 } } }, + [GDT_ENTRY_ESPFIX_SS] = { { { 0x0000ffff, 0x00cf9200 } } }, [GDT_ENTRY_PERCPU] = { { { 0x0000ffff, 0x00cf9200 } } }, GDT_STACK_CANARY_INIT #endif diff --git a/arch/x86/kernel/cpu/mcheck/Makefile b/arch/x86/kernel/cpu/mcheck/Makefile index 45004faf67e..188a1ca5ad2 100644 --- a/arch/x86/kernel/cpu/mcheck/Makefile +++ b/arch/x86/kernel/cpu/mcheck/Makefile @@ -1,11 +1,12 @@ -obj-y = mce.o therm_throt.o +obj-y = mce.o obj-$(CONFIG_X86_NEW_MCE) += mce-severity.o obj-$(CONFIG_X86_OLD_MCE) += k7.o p4.o p6.o obj-$(CONFIG_X86_ANCIENT_MCE) += winchip.o p5.o -obj-$(CONFIG_X86_MCE_P4THERMAL) += mce_intel.o -obj-$(CONFIG_X86_MCE_INTEL) += mce_intel_64.o mce_intel.o -obj-$(CONFIG_X86_MCE_AMD) += mce_amd_64.o +obj-$(CONFIG_X86_MCE_INTEL) += mce_intel.o +obj-$(CONFIG_X86_MCE_AMD) += mce_amd.o obj-$(CONFIG_X86_MCE_NONFATAL) += non-fatal.o obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o obj-$(CONFIG_X86_MCE_INJECT) += mce-inject.o + +obj-$(CONFIG_X86_THERMAL_VECTOR) += therm_throt.o diff --git a/arch/x86/kernel/cpu/mcheck/k7.c b/arch/x86/kernel/cpu/mcheck/k7.c index 89e51042415..b945d5dbc60 100644 --- a/arch/x86/kernel/cpu/mcheck/k7.c +++ b/arch/x86/kernel/cpu/mcheck/k7.c @@ -10,10 +10,9 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* Machine Check Handler For AMD Athlon/Duron: */ static void k7_machine_check(struct pt_regs *regs, long error_code) { diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index fabba15e455..af425b83202 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -44,7 +44,6 @@ #include <asm/msr.h> #include "mce-internal.h" -#include "mce.h" /* Handle unconfigured int18 (should never happen) */ static void unexpected_machine_check(struct pt_regs *regs, long error_code) @@ -57,7 +56,7 @@ static void unexpected_machine_check(struct pt_regs *regs, long error_code) void (*machine_check_vector)(struct pt_regs *, long error_code) = unexpected_machine_check; -int mce_disabled; +int mce_disabled __read_mostly; #ifdef CONFIG_X86_NEW_MCE @@ -76,21 +75,22 @@ DEFINE_PER_CPU(unsigned, mce_exception_count); * 2: SIGBUS or log uncorrected errors (if possible), log corrected errors * 3: never panic or SIGBUS, log all errors (for testing only) */ -static int tolerant = 1; -static int banks; -static u64 *bank; -static unsigned long notify_user; -static int rip_msr; -static int mce_bootlog = -1; -static int monarch_timeout = -1; -static int mce_panic_timeout; -static int mce_dont_log_ce; -int mce_cmci_disabled; -int mce_ignore_ce; -int mce_ser; - -static char trigger[128]; -static char *trigger_argv[2] = { trigger, NULL }; +static int tolerant __read_mostly = 1; +static int banks __read_mostly; +static u64 *bank __read_mostly; +static int rip_msr __read_mostly; +static int mce_bootlog __read_mostly = -1; +static int monarch_timeout __read_mostly = -1; +static int mce_panic_timeout __read_mostly; +static int mce_dont_log_ce __read_mostly; +int mce_cmci_disabled __read_mostly; +int mce_ignore_ce __read_mostly; +int mce_ser __read_mostly; + +/* User mode helper program triggered by machine check event */ +static unsigned long mce_need_notify; +static char mce_helper[128]; +static char *mce_helper_argv[2] = { mce_helper, NULL }; static unsigned long dont_init_banks; @@ -180,7 +180,7 @@ void mce_log(struct mce *mce) wmb(); mce->finished = 1; - set_bit(0, ¬ify_user); + set_bit(0, &mce_need_notify); } static void print_mce(struct mce *m) @@ -691,18 +691,21 @@ static atomic_t global_nwo; * in the entry order. * TBD double check parallel CPU hotunplug */ -static int mce_start(int no_way_out, int *order) +static int mce_start(int *no_way_out) { - int nwo; + int order; int cpus = num_online_cpus(); u64 timeout = (u64)monarch_timeout * NSEC_PER_USEC; - if (!timeout) { - *order = -1; - return no_way_out; - } + if (!timeout) + return -1; - atomic_add(no_way_out, &global_nwo); + atomic_add(*no_way_out, &global_nwo); + /* + * global_nwo should be updated before mce_callin + */ + smp_wmb(); + order = atomic_add_return(1, &mce_callin); /* * Wait for everyone. @@ -710,40 +713,43 @@ static int mce_start(int no_way_out, int *order) while (atomic_read(&mce_callin) != cpus) { if (mce_timed_out(&timeout)) { atomic_set(&global_nwo, 0); - *order = -1; - return no_way_out; + return -1; } ndelay(SPINUNIT); } /* - * Cache the global no_way_out state. + * mce_callin should be read before global_nwo */ - nwo = atomic_read(&global_nwo); + smp_rmb(); - /* - * Monarch starts executing now, the others wait. - */ - if (*order == 1) { + if (order == 1) { + /* + * Monarch: Starts executing now, the others wait. + */ atomic_set(&mce_executing, 1); - return nwo; + } else { + /* + * Subject: Now start the scanning loop one by one in + * the original callin order. + * This way when there are any shared banks it will be + * only seen by one CPU before cleared, avoiding duplicates. + */ + while (atomic_read(&mce_executing) < order) { + if (mce_timed_out(&timeout)) { + atomic_set(&global_nwo, 0); + return -1; + } + ndelay(SPINUNIT); + } } /* - * Now start the scanning loop one by one - * in the original callin order. - * This way when there are any shared banks it will - * be only seen by one CPU before cleared, avoiding duplicates. + * Cache the global no_way_out state. */ - while (atomic_read(&mce_executing) < *order) { - if (mce_timed_out(&timeout)) { - atomic_set(&global_nwo, 0); - *order = -1; - return no_way_out; - } - ndelay(SPINUNIT); - } - return nwo; + *no_way_out = atomic_read(&global_nwo); + + return order; } /* @@ -863,7 +869,6 @@ void do_machine_check(struct pt_regs *regs, long error_code) * check handler. */ int order; - /* * If no_way_out gets set, there is no safe way to recover from this * MCE. If tolerant is cranked up, we'll try anyway. @@ -887,7 +892,6 @@ void do_machine_check(struct pt_regs *regs, long error_code) if (!banks) goto out; - order = atomic_add_return(1, &mce_callin); mce_setup(&m); m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS); @@ -909,7 +913,7 @@ void do_machine_check(struct pt_regs *regs, long error_code) * This way we don't report duplicated events on shared banks * because the first one to see it will clear it. */ - no_way_out = mce_start(no_way_out, &order); + order = mce_start(&no_way_out); for (i = 0; i < banks; i++) { __clear_bit(i, toclear); if (!bank[i]) @@ -1113,12 +1117,12 @@ static void mcheck_timer(unsigned long data) *n = min(*n*2, (int)round_jiffies_relative(check_interval*HZ)); t->expires = jiffies + *n; - add_timer(t); + add_timer_on(t, smp_processor_id()); } static void mce_do_trigger(struct work_struct *work) { - call_usermodehelper(trigger, trigger_argv, NULL, UMH_NO_WAIT); + call_usermodehelper(mce_helper, mce_helper_argv, NULL, UMH_NO_WAIT); } static DECLARE_WORK(mce_trigger_work, mce_do_trigger); @@ -1135,7 +1139,7 @@ int mce_notify_irq(void) clear_thread_flag(TIF_MCE_NOTIFY); - if (test_and_clear_bit(0, ¬ify_user)) { + if (test_and_clear_bit(0, &mce_need_notify)) { wake_up_interruptible(&mce_wait); /* @@ -1143,7 +1147,7 @@ int mce_notify_irq(void) * work_pending is always cleared before the function is * executed. */ - if (trigger[0] && !work_pending(&mce_trigger_work)) + if (mce_helper[0] && !work_pending(&mce_trigger_work)) schedule_work(&mce_trigger_work); if (__ratelimit(&ratelimit)) @@ -1245,7 +1249,7 @@ static void mce_cpu_quirks(struct cpuinfo_x86 *c) * Various K7s with broken bank 0 around. Always disable * by default. */ - if (c->x86 == 6) + if (c->x86 == 6 && banks > 0) bank[0] = 0; } @@ -1282,8 +1286,7 @@ static void __cpuinit mce_ancient_init(struct cpuinfo_x86 *c) return; switch (c->x86_vendor) { case X86_VENDOR_INTEL: - if (mce_p5_enabled()) - intel_p5_mcheck_init(c); + intel_p5_mcheck_init(c); break; case X86_VENDOR_CENTAUR: winchip_mcheck_init(c); @@ -1318,7 +1321,7 @@ static void mce_init_timer(void) return; setup_timer(t, mcheck_timer, smp_processor_id()); t->expires = round_jiffies(jiffies + *n); - add_timer(t); + add_timer_on(t, smp_processor_id()); } /* @@ -1609,8 +1612,9 @@ static int mce_resume(struct sys_device *dev) static void mce_cpu_restart(void *data) { del_timer_sync(&__get_cpu_var(mce_timer)); - if (mce_available(¤t_cpu_data)) - mce_init(); + if (!mce_available(¤t_cpu_data)) + return; + mce_init(); mce_init_timer(); } @@ -1620,6 +1624,26 @@ static void mce_restart(void) on_each_cpu(mce_cpu_restart, NULL, 1); } +/* Toggle features for corrected errors */ +static void mce_disable_ce(void *all) +{ + if (!mce_available(¤t_cpu_data)) + return; + if (all) + del_timer_sync(&__get_cpu_var(mce_timer)); + cmci_clear(); +} + +static void mce_enable_ce(void *all) +{ + if (!mce_available(¤t_cpu_data)) + return; + cmci_reenable(); + cmci_recheck(); + if (all) + mce_init_timer(); +} + static struct sysdev_class mce_sysclass = { .suspend = mce_suspend, .shutdown = mce_shutdown, @@ -1659,9 +1683,9 @@ static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr, static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr, char *buf) { - strcpy(buf, trigger); + strcpy(buf, mce_helper); strcat(buf, "\n"); - return strlen(trigger) + 1; + return strlen(mce_helper) + 1; } static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr, @@ -1670,10 +1694,10 @@ static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr, char *p; int len; - strncpy(trigger, buf, sizeof(trigger)); - trigger[sizeof(trigger)-1] = 0; - len = strlen(trigger); - p = strchr(trigger, '\n'); + strncpy(mce_helper, buf, sizeof(mce_helper)); + mce_helper[sizeof(mce_helper)-1] = 0; + len = strlen(mce_helper); + p = strchr(mce_helper, '\n'); if (*p) *p = 0; @@ -1681,6 +1705,52 @@ static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr, return len; } +static ssize_t set_ignore_ce(struct sys_device *s, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + u64 new; + + if (strict_strtoull(buf, 0, &new) < 0) + return -EINVAL; + + if (mce_ignore_ce ^ !!new) { + if (new) { + /* disable ce features */ + on_each_cpu(mce_disable_ce, (void *)1, 1); + mce_ignore_ce = 1; + } else { + /* enable ce features */ + mce_ignore_ce = 0; + on_each_cpu(mce_enable_ce, (void *)1, 1); + } + } + return size; +} + +static ssize_t set_cmci_disabled(struct sys_device *s, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + u64 new; + + if (strict_strtoull(buf, 0, &new) < 0) + return -EINVAL; + + if (mce_cmci_disabled ^ !!new) { + if (new) { + /* disable cmci */ + on_each_cpu(mce_disable_ce, NULL, 1); + mce_cmci_disabled = 1; + } else { + /* enable cmci */ + mce_cmci_disabled = 0; + on_each_cpu(mce_enable_ce, NULL, 1); + } + } + return size; +} + static ssize_t store_int_with_restart(struct sys_device *s, struct sysdev_attribute *attr, const char *buf, size_t size) @@ -1693,6 +1763,7 @@ static ssize_t store_int_with_restart(struct sys_device *s, static SYSDEV_ATTR(trigger, 0644, show_trigger, set_trigger); static SYSDEV_INT_ATTR(tolerant, 0644, tolerant); static SYSDEV_INT_ATTR(monarch_timeout, 0644, monarch_timeout); +static SYSDEV_INT_ATTR(dont_log_ce, 0644, mce_dont_log_ce); static struct sysdev_ext_attribute attr_check_interval = { _SYSDEV_ATTR(check_interval, 0644, sysdev_show_int, @@ -1700,9 +1771,24 @@ static struct sysdev_ext_attribute attr_check_interval = { &check_interval }; +static struct sysdev_ext_attribute attr_ignore_ce = { + _SYSDEV_ATTR(ignore_ce, 0644, sysdev_show_int, set_ignore_ce), + &mce_ignore_ce +}; + +static struct sysdev_ext_attribute attr_cmci_disabled = { + _SYSDEV_ATTR(cmci_disabled, 0644, sysdev_show_int, set_cmci_disabled), + &mce_cmci_disabled +}; + static struct sysdev_attribute *mce_attrs[] = { - &attr_tolerant.attr, &attr_check_interval.attr, &attr_trigger, + &attr_tolerant.attr, + &attr_check_interval.attr, + &attr_trigger, &attr_monarch_timeout.attr, + &attr_dont_log_ce.attr, + &attr_ignore_ce.attr, + &attr_cmci_disabled.attr, NULL }; @@ -1712,7 +1798,7 @@ static cpumask_var_t mce_dev_initialized; static __cpuinit int mce_create_device(unsigned int cpu) { int err; - int i; + int i, j; if (!mce_available(&boot_cpu_data)) return -EIO; @@ -1730,9 +1816,9 @@ static __cpuinit int mce_create_device(unsigned int cpu) if (err) goto error; } - for (i = 0; i < banks; i++) { + for (j = 0; j < banks; j++) { err = sysdev_create_file(&per_cpu(mce_dev, cpu), - &bank_attrs[i]); + &bank_attrs[j]); if (err) goto error2; } @@ -1740,8 +1826,8 @@ static __cpuinit int mce_create_device(unsigned int cpu) return 0; error2: - while (--i >= 0) - sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[i]); + while (--j >= 0) + sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[j]); error: while (--i >= 0) sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]); @@ -1883,7 +1969,7 @@ static __init int mce_init_device(void) if (!mce_available(&boot_cpu_data)) return -EIO; - alloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL); + zalloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL); err = mce_init_banks(); if (err) @@ -1915,7 +2001,7 @@ EXPORT_SYMBOL_GPL(nr_mce_banks); /* non-fatal.o */ /* This has to be run for each processor */ void mcheck_init(struct cpuinfo_x86 *c) { - if (mce_disabled == 1) + if (mce_disabled) return; switch (c->x86_vendor) { @@ -1945,10 +2031,9 @@ void mcheck_init(struct cpuinfo_x86 *c) static int __init mcheck_enable(char *str) { - mce_disabled = -1; + mce_p5_enabled = 1; return 1; } - __setup("mce", mcheck_enable); #endif /* CONFIG_X86_OLD_MCE */ diff --git a/arch/x86/kernel/cpu/mcheck/mce.h b/arch/x86/kernel/cpu/mcheck/mce.h deleted file mode 100644 index 84a552b458c..00000000000 --- a/arch/x86/kernel/cpu/mcheck/mce.h +++ /dev/null @@ -1,38 +0,0 @@ -#include <linux/init.h> -#include <asm/mce.h> - -#ifdef CONFIG_X86_OLD_MCE -void amd_mcheck_init(struct cpuinfo_x86 *c); -void intel_p4_mcheck_init(struct cpuinfo_x86 *c); -void intel_p6_mcheck_init(struct cpuinfo_x86 *c); -#endif - -#ifdef CONFIG_X86_ANCIENT_MCE -void intel_p5_mcheck_init(struct cpuinfo_x86 *c); -void winchip_mcheck_init(struct cpuinfo_x86 *c); -extern int mce_p5_enable; -static inline int mce_p5_enabled(void) { return mce_p5_enable; } -static inline void enable_p5_mce(void) { mce_p5_enable = 1; } -#else -static inline void intel_p5_mcheck_init(struct cpuinfo_x86 *c) {} -static inline void winchip_mcheck_init(struct cpuinfo_x86 *c) {} -static inline int mce_p5_enabled(void) { return 0; } -static inline void enable_p5_mce(void) { } -#endif - -/* Call the installed machine check handler for this CPU setup. */ -extern void (*machine_check_vector)(struct pt_regs *, long error_code); - -#ifdef CONFIG_X86_OLD_MCE - -extern int nr_mce_banks; - -void intel_set_thermal_handler(void); - -#else - -static inline void intel_set_thermal_handler(void) { } - -#endif - -void intel_init_thermal(struct cpuinfo_x86 *c); diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c index ddae21620bd..ddae21620bd 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c index 2b011d2d857..e1acec0f7a3 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c @@ -1,74 +1,226 @@ /* - * Common code for Intel machine checks + * Intel specific MCE features. + * Copyright 2004 Zwane Mwaikambo <zwane@linuxpower.ca> + * Copyright (C) 2008, 2009 Intel Corporation + * Author: Andi Kleen */ -#include <linux/interrupt.h> -#include <linux/kernel.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/smp.h> -#include <asm/therm_throt.h> -#include <asm/processor.h> -#include <asm/system.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/percpu.h> #include <asm/apic.h> +#include <asm/processor.h> #include <asm/msr.h> +#include <asm/mce.h> + +/* + * Support for Intel Correct Machine Check Interrupts. This allows + * the CPU to raise an interrupt when a corrected machine check happened. + * Normally we pick those up using a regular polling timer. + * Also supports reliable discovery of shared banks. + */ -#include "mce.h" +static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned); -void intel_init_thermal(struct cpuinfo_x86 *c) +/* + * cmci_discover_lock protects against parallel discovery attempts + * which could race against each other. + */ +static DEFINE_SPINLOCK(cmci_discover_lock); + +#define CMCI_THRESHOLD 1 + +static int cmci_supported(int *banks) { - unsigned int cpu = smp_processor_id(); - int tm2 = 0; - u32 l, h; + u64 cap; - /* Thermal monitoring depends on ACPI and clock modulation*/ - if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC)) - return; + if (mce_cmci_disabled || mce_ignore_ce) + return 0; /* - * First check if its enabled already, in which case there might - * be some SMM goo which handles it, so we can't even put a handler - * since it might be delivered via SMI already: + * Vendor check is not strictly needed, but the initial + * initialization is vendor keyed and this + * makes sure none of the backdoors are entered otherwise. */ - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - h = apic_read(APIC_LVTTHMR); - if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { - printk(KERN_DEBUG - "CPU%d: Thermal monitoring handled by SMI\n", cpu); - return; + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) + return 0; + if (!cpu_has_apic || lapic_get_maxlvt() < 6) + return 0; + rdmsrl(MSR_IA32_MCG_CAP, cap); + *banks = min_t(unsigned, MAX_NR_BANKS, cap & 0xff); + return !!(cap & MCG_CMCI_P); +} + +/* + * The interrupt handler. This is called on every event. + * Just call the poller directly to log any events. + * This could in theory increase the threshold under high load, + * but doesn't for now. + */ +static void intel_threshold_interrupt(void) +{ + machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); + mce_notify_irq(); +} + +static void print_update(char *type, int *hdr, int num) +{ + if (*hdr == 0) + printk(KERN_INFO "CPU %d MCA banks", smp_processor_id()); + *hdr = 1; + printk(KERN_CONT " %s:%d", type, num); +} + +/* + * Enable CMCI (Corrected Machine Check Interrupt) for available MCE banks + * on this CPU. Use the algorithm recommended in the SDM to discover shared + * banks. + */ +static void cmci_discover(int banks, int boot) +{ + unsigned long *owned = (void *)&__get_cpu_var(mce_banks_owned); + unsigned long flags; + int hdr = 0; + int i; + + spin_lock_irqsave(&cmci_discover_lock, flags); + for (i = 0; i < banks; i++) { + u64 val; + + if (test_bit(i, owned)) + continue; + + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + + /* Already owned by someone else? */ + if (val & CMCI_EN) { + if (test_and_clear_bit(i, owned) || boot) + print_update("SHD", &hdr, i); + __clear_bit(i, __get_cpu_var(mce_poll_banks)); + continue; + } + + val |= CMCI_EN | CMCI_THRESHOLD; + wrmsrl(MSR_IA32_MC0_CTL2 + i, val); + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + + /* Did the enable bit stick? -- the bank supports CMCI */ + if (val & CMCI_EN) { + if (!test_and_set_bit(i, owned) || boot) + print_update("CMCI", &hdr, i); + __clear_bit(i, __get_cpu_var(mce_poll_banks)); + } else { + WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks))); + } } + spin_unlock_irqrestore(&cmci_discover_lock, flags); + if (hdr) + printk(KERN_CONT "\n"); +} + +/* + * Just in case we missed an event during initialization check + * all the CMCI owned banks. + */ +void cmci_recheck(void) +{ + unsigned long flags; + int banks; + + if (!mce_available(¤t_cpu_data) || !cmci_supported(&banks)) + return; + local_irq_save(flags); + machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); + local_irq_restore(flags); +} - if (cpu_has(c, X86_FEATURE_TM2) && (l & MSR_IA32_MISC_ENABLE_TM2)) - tm2 = 1; +/* + * Disable CMCI on this CPU for all banks it owns when it goes down. + * This allows other CPUs to claim the banks on rediscovery. + */ +void cmci_clear(void) +{ + unsigned long flags; + int i; + int banks; + u64 val; - /* Check whether a vector already exists */ - if (h & APIC_VECTOR_MASK) { - printk(KERN_DEBUG - "CPU%d: Thermal LVT vector (%#x) already installed\n", - cpu, (h & APIC_VECTOR_MASK)); + if (!cmci_supported(&banks)) return; + spin_lock_irqsave(&cmci_discover_lock, flags); + for (i = 0; i < banks; i++) { + if (!test_bit(i, __get_cpu_var(mce_banks_owned))) + continue; + /* Disable CMCI */ + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + val &= ~(CMCI_EN|CMCI_THRESHOLD_MASK); + wrmsrl(MSR_IA32_MC0_CTL2 + i, val); + __clear_bit(i, __get_cpu_var(mce_banks_owned)); } + spin_unlock_irqrestore(&cmci_discover_lock, flags); +} + +/* + * After a CPU went down cycle through all the others and rediscover + * Must run in process context. + */ +void cmci_rediscover(int dying) +{ + int banks; + int cpu; + cpumask_var_t old; + + if (!cmci_supported(&banks)) + return; + if (!alloc_cpumask_var(&old, GFP_KERNEL)) + return; + cpumask_copy(old, ¤t->cpus_allowed); - /* We'll mask the thermal vector in the lapic till we're ready: */ - h = THERMAL_APIC_VECTOR | APIC_DM_FIXED | APIC_LVT_MASKED; - apic_write(APIC_LVTTHMR, h); + for_each_online_cpu(cpu) { + if (cpu == dying) + continue; + if (set_cpus_allowed_ptr(current, cpumask_of(cpu))) + continue; + /* Recheck banks in case CPUs don't all have the same */ + if (cmci_supported(&banks)) + cmci_discover(banks, 0); + } - rdmsr(MSR_IA32_THERM_INTERRUPT, l, h); - wrmsr(MSR_IA32_THERM_INTERRUPT, - l | (THERM_INT_LOW_ENABLE | THERM_INT_HIGH_ENABLE), h); + set_cpus_allowed_ptr(current, old); + free_cpumask_var(old); +} - intel_set_thermal_handler(); +/* + * Reenable CMCI on this CPU in case a CPU down failed. + */ +void cmci_reenable(void) +{ + int banks; + if (cmci_supported(&banks)) + cmci_discover(banks, 0); +} - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h); +static void intel_init_cmci(void) +{ + int banks; - /* Unmask the thermal vector: */ - l = apic_read(APIC_LVTTHMR); - apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); + if (!cmci_supported(&banks)) + return; - printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n", - cpu, tm2 ? "TM2" : "TM1"); + mce_threshold_vector = intel_threshold_interrupt; + cmci_discover(banks, 1); + /* + * For CPU #0 this runs with still disabled APIC, but that's + * ok because only the vector is set up. We still do another + * check for the banks later for CPU #0 just to make sure + * to not miss any events. + */ + apic_write(APIC_LVTCMCI, THRESHOLD_APIC_VECTOR|APIC_DM_FIXED); + cmci_recheck(); +} - /* enable thermal throttle processing */ - atomic_set(&therm_throt_en, 1); +void mce_intel_feature_init(struct cpuinfo_x86 *c) +{ + intel_init_thermal(c); + intel_init_cmci(); } diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c deleted file mode 100644 index f2ef6952c40..00000000000 --- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c +++ /dev/null @@ -1,248 +0,0 @@ -/* - * Intel specific MCE features. - * Copyright 2004 Zwane Mwaikambo <zwane@linuxpower.ca> - * Copyright (C) 2008, 2009 Intel Corporation - * Author: Andi Kleen - */ - -#include <linux/init.h> -#include <linux/interrupt.h> -#include <linux/percpu.h> -#include <asm/processor.h> -#include <asm/apic.h> -#include <asm/msr.h> -#include <asm/mce.h> -#include <asm/hw_irq.h> -#include <asm/idle.h> -#include <asm/therm_throt.h> - -#include "mce.h" - -asmlinkage void smp_thermal_interrupt(void) -{ - __u64 msr_val; - - ack_APIC_irq(); - - exit_idle(); - irq_enter(); - - rdmsrl(MSR_IA32_THERM_STATUS, msr_val); - if (therm_throt_process(msr_val & THERM_STATUS_PROCHOT)) - mce_log_therm_throt_event(msr_val); - - inc_irq_stat(irq_thermal_count); - irq_exit(); -} - -/* - * Support for Intel Correct Machine Check Interrupts. This allows - * the CPU to raise an interrupt when a corrected machine check happened. - * Normally we pick those up using a regular polling timer. - * Also supports reliable discovery of shared banks. - */ - -static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned); - -/* - * cmci_discover_lock protects against parallel discovery attempts - * which could race against each other. - */ -static DEFINE_SPINLOCK(cmci_discover_lock); - -#define CMCI_THRESHOLD 1 - -static int cmci_supported(int *banks) -{ - u64 cap; - - if (mce_cmci_disabled || mce_ignore_ce) - return 0; - - /* - * Vendor check is not strictly needed, but the initial - * initialization is vendor keyed and this - * makes sure none of the backdoors are entered otherwise. - */ - if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) - return 0; - if (!cpu_has_apic || lapic_get_maxlvt() < 6) - return 0; - rdmsrl(MSR_IA32_MCG_CAP, cap); - *banks = min_t(unsigned, MAX_NR_BANKS, cap & 0xff); - return !!(cap & MCG_CMCI_P); -} - -/* - * The interrupt handler. This is called on every event. - * Just call the poller directly to log any events. - * This could in theory increase the threshold under high load, - * but doesn't for now. - */ -static void intel_threshold_interrupt(void) -{ - machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); - mce_notify_irq(); -} - -static void print_update(char *type, int *hdr, int num) -{ - if (*hdr == 0) - printk(KERN_INFO "CPU %d MCA banks", smp_processor_id()); - *hdr = 1; - printk(KERN_CONT " %s:%d", type, num); -} - -/* - * Enable CMCI (Corrected Machine Check Interrupt) for available MCE banks - * on this CPU. Use the algorithm recommended in the SDM to discover shared - * banks. - */ -static void cmci_discover(int banks, int boot) -{ - unsigned long *owned = (void *)&__get_cpu_var(mce_banks_owned); - unsigned long flags; - int hdr = 0; - int i; - - spin_lock_irqsave(&cmci_discover_lock, flags); - for (i = 0; i < banks; i++) { - u64 val; - - if (test_bit(i, owned)) - continue; - - rdmsrl(MSR_IA32_MC0_CTL2 + i, val); - - /* Already owned by someone else? */ - if (val & CMCI_EN) { - if (test_and_clear_bit(i, owned) || boot) - print_update("SHD", &hdr, i); - __clear_bit(i, __get_cpu_var(mce_poll_banks)); - continue; - } - - val |= CMCI_EN | CMCI_THRESHOLD; - wrmsrl(MSR_IA32_MC0_CTL2 + i, val); - rdmsrl(MSR_IA32_MC0_CTL2 + i, val); - - /* Did the enable bit stick? -- the bank supports CMCI */ - if (val & CMCI_EN) { - if (!test_and_set_bit(i, owned) || boot) - print_update("CMCI", &hdr, i); - __clear_bit(i, __get_cpu_var(mce_poll_banks)); - } else { - WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks))); - } - } - spin_unlock_irqrestore(&cmci_discover_lock, flags); - if (hdr) - printk(KERN_CONT "\n"); -} - -/* - * Just in case we missed an event during initialization check - * all the CMCI owned banks. - */ -void cmci_recheck(void) -{ - unsigned long flags; - int banks; - - if (!mce_available(¤t_cpu_data) || !cmci_supported(&banks)) - return; - local_irq_save(flags); - machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); - local_irq_restore(flags); -} - -/* - * Disable CMCI on this CPU for all banks it owns when it goes down. - * This allows other CPUs to claim the banks on rediscovery. - */ -void cmci_clear(void) -{ - unsigned long flags; - int i; - int banks; - u64 val; - - if (!cmci_supported(&banks)) - return; - spin_lock_irqsave(&cmci_discover_lock, flags); - for (i = 0; i < banks; i++) { - if (!test_bit(i, __get_cpu_var(mce_banks_owned))) - continue; - /* Disable CMCI */ - rdmsrl(MSR_IA32_MC0_CTL2 + i, val); - val &= ~(CMCI_EN|CMCI_THRESHOLD_MASK); - wrmsrl(MSR_IA32_MC0_CTL2 + i, val); - __clear_bit(i, __get_cpu_var(mce_banks_owned)); - } - spin_unlock_irqrestore(&cmci_discover_lock, flags); -} - -/* - * After a CPU went down cycle through all the others and rediscover - * Must run in process context. - */ -void cmci_rediscover(int dying) -{ - int banks; - int cpu; - cpumask_var_t old; - - if (!cmci_supported(&banks)) - return; - if (!alloc_cpumask_var(&old, GFP_KERNEL)) - return; - cpumask_copy(old, ¤t->cpus_allowed); - - for_each_online_cpu(cpu) { - if (cpu == dying) - continue; - if (set_cpus_allowed_ptr(current, cpumask_of(cpu))) - continue; - /* Recheck banks in case CPUs don't all have the same */ - if (cmci_supported(&banks)) - cmci_discover(banks, 0); - } - - set_cpus_allowed_ptr(current, old); - free_cpumask_var(old); -} - -/* - * Reenable CMCI on this CPU in case a CPU down failed. - */ -void cmci_reenable(void) -{ - int banks; - if (cmci_supported(&banks)) - cmci_discover(banks, 0); -} - -static void intel_init_cmci(void) -{ - int banks; - - if (!cmci_supported(&banks)) - return; - - mce_threshold_vector = intel_threshold_interrupt; - cmci_discover(banks, 1); - /* - * For CPU #0 this runs with still disabled APIC, but that's - * ok because only the vector is set up. We still do another - * check for the banks later for CPU #0 just to make sure - * to not miss any events. - */ - apic_write(APIC_LVTCMCI, THRESHOLD_APIC_VECTOR|APIC_DM_FIXED); - cmci_recheck(); -} - -void mce_intel_feature_init(struct cpuinfo_x86 *c) -{ - intel_init_thermal(c); - intel_init_cmci(); -} diff --git a/arch/x86/kernel/cpu/mcheck/non-fatal.c b/arch/x86/kernel/cpu/mcheck/non-fatal.c index 70b710420f7..f5f2d6f71fb 100644 --- a/arch/x86/kernel/cpu/mcheck/non-fatal.c +++ b/arch/x86/kernel/cpu/mcheck/non-fatal.c @@ -17,10 +17,9 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - static int firstbank; #define MCE_RATE (15*HZ) /* timer rate is 15s */ diff --git a/arch/x86/kernel/cpu/mcheck/p4.c b/arch/x86/kernel/cpu/mcheck/p4.c index 82cee108a2d..4482aea9aa2 100644 --- a/arch/x86/kernel/cpu/mcheck/p4.c +++ b/arch/x86/kernel/cpu/mcheck/p4.c @@ -1,21 +1,15 @@ /* * P4 specific Machine Check Exception Reporting */ - -#include <linux/interrupt.h> #include <linux/kernel.h> #include <linux/types.h> #include <linux/init.h> #include <linux/smp.h> -#include <asm/therm_throt.h> #include <asm/processor.h> -#include <asm/system.h> -#include <asm/apic.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* as supported by the P4/Xeon family */ struct intel_mce_extended_msrs { u32 eax; @@ -33,46 +27,6 @@ struct intel_mce_extended_msrs { static int mce_num_extended_msrs; - -#ifdef CONFIG_X86_MCE_P4THERMAL - -static void unexpected_thermal_interrupt(struct pt_regs *regs) -{ - printk(KERN_ERR "CPU%d: Unexpected LVT TMR interrupt!\n", - smp_processor_id()); - add_taint(TAINT_MACHINE_CHECK); -} - -/* P4/Xeon Thermal transition interrupt handler: */ -static void intel_thermal_interrupt(struct pt_regs *regs) -{ - __u64 msr_val; - - ack_APIC_irq(); - - rdmsrl(MSR_IA32_THERM_STATUS, msr_val); - therm_throt_process(msr_val & THERM_STATUS_PROCHOT); -} - -/* Thermal interrupt handler for this CPU setup: */ -static void (*vendor_thermal_interrupt)(struct pt_regs *regs) = - unexpected_thermal_interrupt; - -void smp_thermal_interrupt(struct pt_regs *regs) -{ - irq_enter(); - vendor_thermal_interrupt(regs); - __get_cpu_var(irq_stat).irq_thermal_count++; - irq_exit(); -} - -void intel_set_thermal_handler(void) -{ - vendor_thermal_interrupt = intel_thermal_interrupt; -} - -#endif /* CONFIG_X86_MCE_P4THERMAL */ - /* P4/Xeon Extended MCE MSR retrieval, return 0 if unsupported */ static void intel_get_extended_msrs(struct intel_mce_extended_msrs *r) { diff --git a/arch/x86/kernel/cpu/mcheck/p5.c b/arch/x86/kernel/cpu/mcheck/p5.c index 015f481ab1b..5c0e6533d9b 100644 --- a/arch/x86/kernel/cpu/mcheck/p5.c +++ b/arch/x86/kernel/cpu/mcheck/p5.c @@ -10,12 +10,11 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* By default disabled */ -int mce_p5_enable; +int mce_p5_enabled __read_mostly; /* Machine check handler for Pentium class Intel CPUs: */ static void pentium_machine_check(struct pt_regs *regs, long error_code) @@ -43,15 +42,13 @@ void intel_p5_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; - /* Check for MCE support: */ - if (!cpu_has(c, X86_FEATURE_MCE)) + /* Default P5 to off as its often misconnected: */ + if (!mce_p5_enabled) return; -#ifdef CONFIG_X86_OLD_MCE - /* Default P5 to off as its often misconnected: */ - if (mce_disabled != -1) + /* Check for MCE support: */ + if (!cpu_has(c, X86_FEATURE_MCE)) return; -#endif machine_check_vector = pentium_machine_check; /* Make sure the vector pointer is visible before we enable MCEs: */ diff --git a/arch/x86/kernel/cpu/mcheck/p6.c b/arch/x86/kernel/cpu/mcheck/p6.c index 43c24e66745..01e4f817818 100644 --- a/arch/x86/kernel/cpu/mcheck/p6.c +++ b/arch/x86/kernel/cpu/mcheck/p6.c @@ -10,10 +10,9 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* Machine Check Handler For PII/PIII */ static void intel_machine_check(struct pt_regs *regs, long error_code) { diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index 7b1ae2e20ba..bff8dd191dd 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -13,13 +13,23 @@ * Credits: Adapted from Zwane Mwaikambo's original code in mce_intel.c. * Inspired by Ross Biro's and Al Borchers' counter code. */ +#include <linux/interrupt.h> #include <linux/notifier.h> #include <linux/jiffies.h> +#include <linux/kernel.h> #include <linux/percpu.h> #include <linux/sysdev.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/smp.h> #include <linux/cpu.h> -#include <asm/therm_throt.h> +#include <asm/processor.h> +#include <asm/system.h> +#include <asm/apic.h> +#include <asm/idle.h> +#include <asm/mce.h> +#include <asm/msr.h> /* How long to wait between reporting thermal events */ #define CHECK_INTERVAL (300 * HZ) @@ -27,7 +37,7 @@ static DEFINE_PER_CPU(__u64, next_check) = INITIAL_JIFFIES; static DEFINE_PER_CPU(unsigned long, thermal_throttle_count); -atomic_t therm_throt_en = ATOMIC_INIT(0); +static atomic_t therm_throt_en = ATOMIC_INIT(0); #ifdef CONFIG_SYSFS #define define_therm_throt_sysdev_one_ro(_name) \ @@ -82,7 +92,7 @@ static struct attribute_group thermal_throttle_attr_group = { * 1 : Event should be logged further, and a message has been * printed to the syslog. */ -int therm_throt_process(int curr) +static int therm_throt_process(int curr) { unsigned int cpu = smp_processor_id(); __u64 tmp_jiffs = get_jiffies_64(); @@ -186,6 +196,94 @@ static __init int thermal_throttle_init_device(void) return 0; } - device_initcall(thermal_throttle_init_device); + #endif /* CONFIG_SYSFS */ + +/* Thermal transition interrupt handler */ +static void intel_thermal_interrupt(void) +{ + __u64 msr_val; + + rdmsrl(MSR_IA32_THERM_STATUS, msr_val); + if (therm_throt_process(msr_val & THERM_STATUS_PROCHOT)) + mce_log_therm_throt_event(msr_val); +} + +static void unexpected_thermal_interrupt(void) +{ + printk(KERN_ERR "CPU%d: Unexpected LVT TMR interrupt!\n", + smp_processor_id()); + add_taint(TAINT_MACHINE_CHECK); +} + +static void (*smp_thermal_vector)(void) = unexpected_thermal_interrupt; + +asmlinkage void smp_thermal_interrupt(struct pt_regs *regs) +{ + exit_idle(); + irq_enter(); + inc_irq_stat(irq_thermal_count); + smp_thermal_vector(); + irq_exit(); + /* Ack only at the end to avoid potential reentry */ + ack_APIC_irq(); +} + +void intel_init_thermal(struct cpuinfo_x86 *c) +{ + unsigned int cpu = smp_processor_id(); + int tm2 = 0; + u32 l, h; + + /* Thermal monitoring depends on ACPI and clock modulation*/ + if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC)) + return; + + /* + * First check if its enabled already, in which case there might + * be some SMM goo which handles it, so we can't even put a handler + * since it might be delivered via SMI already: + */ + rdmsr(MSR_IA32_MISC_ENABLE, l, h); + h = apic_read(APIC_LVTTHMR); + if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { + printk(KERN_DEBUG + "CPU%d: Thermal monitoring handled by SMI\n", cpu); + return; + } + + if (cpu_has(c, X86_FEATURE_TM2) && (l & MSR_IA32_MISC_ENABLE_TM2)) + tm2 = 1; + + /* Check whether a vector already exists */ + if (h & APIC_VECTOR_MASK) { + printk(KERN_DEBUG + "CPU%d: Thermal LVT vector (%#x) already installed\n", + cpu, (h & APIC_VECTOR_MASK)); + return; + } + + /* We'll mask the thermal vector in the lapic till we're ready: */ + h = THERMAL_APIC_VECTOR | APIC_DM_FIXED | APIC_LVT_MASKED; + apic_write(APIC_LVTTHMR, h); + + rdmsr(MSR_IA32_THERM_INTERRUPT, l, h); + wrmsr(MSR_IA32_THERM_INTERRUPT, + l | (THERM_INT_LOW_ENABLE | THERM_INT_HIGH_ENABLE), h); + + smp_thermal_vector = intel_thermal_interrupt; + + rdmsr(MSR_IA32_MISC_ENABLE, l, h); + wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h); + + /* Unmask the thermal vector: */ + l = apic_read(APIC_LVTTHMR); + apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); + + printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n", + cpu, tm2 ? "TM2" : "TM1"); + + /* enable thermal throttle processing */ + atomic_set(&therm_throt_en, 1); +} diff --git a/arch/x86/kernel/cpu/mcheck/winchip.c b/arch/x86/kernel/cpu/mcheck/winchip.c index 81b02487090..54060f56597 100644 --- a/arch/x86/kernel/cpu/mcheck/winchip.c +++ b/arch/x86/kernel/cpu/mcheck/winchip.c @@ -9,10 +9,9 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* Machine check handler for WinChip C6: */ static void winchip_machine_check(struct pt_regs *regs, long error_code) { diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c index 275bc142cd5..d4cf4ce19aa 100644 --- a/arch/x86/kernel/cpu/perf_counter.c +++ b/arch/x86/kernel/cpu/perf_counter.c @@ -19,6 +19,7 @@ #include <linux/kdebug.h> #include <linux/sched.h> #include <linux/uaccess.h> +#include <linux/highmem.h> #include <asm/apic.h> #include <asm/stacktrace.h> @@ -389,23 +390,23 @@ static u64 intel_pmu_raw_event(u64 event) return event & CORE_EVNTSEL_MASK; } -static const u64 amd_0f_hw_cache_event_ids +static const u64 amd_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX] = { [ C(L1D) ] = { [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ + [ C(RESULT_MISS) ] = 0x0041, /* Data Cache Misses */ }, [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x0142, /* Data Cache Refills :system */ [ C(RESULT_MISS) ] = 0, }, [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x0267, /* Data Prefetcher :attempts */ + [ C(RESULT_MISS) ] = 0x0167, /* Data Prefetcher :cancelled */ }, }, [ C(L1I ) ] = { @@ -418,17 +419,17 @@ static const u64 amd_0f_hw_cache_event_ids [ C(RESULT_MISS) ] = -1, }, [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x014B, /* Prefetch Instructions :Load */ [ C(RESULT_MISS) ] = 0, }, }, [ C(LL ) ] = { [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x037D, /* Requests to L2 Cache :IC+DC */ + [ C(RESULT_MISS) ] = 0x037E, /* L2 Cache Misses : IC+DC */ }, [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x017F, /* L2 Fill/Writeback */ [ C(RESULT_MISS) ] = 0, }, [ C(OP_PREFETCH) ] = { @@ -438,8 +439,8 @@ static const u64 amd_0f_hw_cache_event_ids }, [ C(DTLB) ] = { [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ + [ C(RESULT_MISS) ] = 0x0046, /* L1 DTLB and L2 DLTB Miss */ }, [ C(OP_WRITE) ] = { [ C(RESULT_ACCESS) ] = 0, @@ -911,6 +912,8 @@ x86_perf_counter_set_period(struct perf_counter *counter, err = checking_wrmsrl(hwc->counter_base + idx, (u64)(-left) & x86_pmu.counter_mask); + perf_counter_update_userpage(counter); + return ret; } @@ -968,13 +971,6 @@ fixed_mode_idx(struct perf_counter *counter, struct hw_perf_counter *hwc) if (!x86_pmu.num_counters_fixed) return -1; - /* - * Quirk, IA32_FIXED_CTRs do not work on current Atom processors: - */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && - boot_cpu_data.x86_model == 28) - return -1; - event = hwc->config & ARCH_PERFMON_EVENT_MASK; if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_INSTRUCTIONS))) @@ -1040,6 +1036,8 @@ try_generic: x86_perf_counter_set_period(counter, hwc, idx); x86_pmu.enable(hwc, idx); + perf_counter_update_userpage(counter); + return 0; } @@ -1132,6 +1130,8 @@ static void x86_pmu_disable(struct perf_counter *counter) x86_perf_counter_update(counter, hwc, idx); cpuc->counters[idx] = NULL; clear_bit(idx, cpuc->used_mask); + + perf_counter_update_userpage(counter); } /* @@ -1223,6 +1223,8 @@ again: if (!intel_pmu_save_and_restart(counter)) continue; + data.period = counter->hw.last_period; + if (perf_counter_overflow(counter, 1, &data)) intel_pmu_disable_counter(&counter->hw, bit); } @@ -1425,8 +1427,6 @@ static int intel_pmu_init(void) */ x86_pmu.num_counters_fixed = max((int)edx.split.num_counters_fixed, 3); - rdmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl); - /* * Install the hw-cache-events table: */ @@ -1459,18 +1459,16 @@ static int intel_pmu_init(void) static int amd_pmu_init(void) { + /* Performance-monitoring supported from K7 and later: */ + if (boot_cpu_data.x86 < 6) + return -ENODEV; + x86_pmu = amd_pmu; - switch (boot_cpu_data.x86) { - case 0x0f: - case 0x10: - case 0x11: - memcpy(hw_cache_event_ids, amd_0f_hw_cache_event_ids, - sizeof(hw_cache_event_ids)); + /* Events are common for all AMDs */ + memcpy(hw_cache_event_ids, amd_hw_cache_event_ids, + sizeof(hw_cache_event_ids)); - pr_cont("AMD Family 0f/10/11 events, "); - break; - } return 0; } @@ -1498,21 +1496,22 @@ void __init init_hw_perf_counters(void) pr_cont("%s PMU driver.\n", x86_pmu.name); if (x86_pmu.num_counters > X86_PMC_MAX_GENERIC) { - x86_pmu.num_counters = X86_PMC_MAX_GENERIC; WARN(1, KERN_ERR "hw perf counters %d > max(%d), clipping!", x86_pmu.num_counters, X86_PMC_MAX_GENERIC); + x86_pmu.num_counters = X86_PMC_MAX_GENERIC; } perf_counter_mask = (1 << x86_pmu.num_counters) - 1; perf_max_counters = x86_pmu.num_counters; if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) { - x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED; WARN(1, KERN_ERR "hw perf counters fixed %d > max(%d), clipping!", x86_pmu.num_counters_fixed, X86_PMC_MAX_FIXED); + x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED; } perf_counter_mask |= ((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED; + x86_pmu.intel_ctrl = perf_counter_mask; perf_counters_lapic_init(); register_die_notifier(&perf_counter_nmi_notifier); @@ -1554,9 +1553,9 @@ const struct pmu *hw_perf_counter_init(struct perf_counter *counter) */ static inline -void callchain_store(struct perf_callchain_entry *entry, unsigned long ip) +void callchain_store(struct perf_callchain_entry *entry, u64 ip) { - if (entry->nr < MAX_STACK_DEPTH) + if (entry->nr < PERF_MAX_STACK_DEPTH) entry->ip[entry->nr++] = ip; } @@ -1577,8 +1576,8 @@ static void backtrace_warning(void *data, char *msg) static int backtrace_stack(void *data, char *name) { - /* Don't bother with IRQ stacks for now */ - return -1; + /* Process all stacks: */ + return 0; } static void backtrace_address(void *data, unsigned long addr, int reliable) @@ -1596,47 +1595,59 @@ static const struct stacktrace_ops backtrace_ops = { .address = backtrace_address, }; +#include "../dumpstack.h" + static void perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) { - unsigned long bp; - char *stack; - int nr = entry->nr; + callchain_store(entry, PERF_CONTEXT_KERNEL); + callchain_store(entry, regs->ip); - callchain_store(entry, instruction_pointer(regs)); + dump_trace(NULL, regs, NULL, 0, &backtrace_ops, entry); +} - stack = ((char *)regs + sizeof(struct pt_regs)); -#ifdef CONFIG_FRAME_POINTER - bp = frame_pointer(regs); -#else - bp = 0; -#endif +/* + * best effort, GUP based copy_from_user() that assumes IRQ or NMI context + */ +static unsigned long +copy_from_user_nmi(void *to, const void __user *from, unsigned long n) +{ + unsigned long offset, addr = (unsigned long)from; + int type = in_nmi() ? KM_NMI : KM_IRQ0; + unsigned long size, len = 0; + struct page *page; + void *map; + int ret; - dump_trace(NULL, regs, (void *)stack, bp, &backtrace_ops, entry); + do { + ret = __get_user_pages_fast(addr, 1, 0, &page); + if (!ret) + break; - entry->kernel = entry->nr - nr; -} + offset = addr & (PAGE_SIZE - 1); + size = min(PAGE_SIZE - offset, n - len); + map = kmap_atomic(page, type); + memcpy(to, map+offset, size); + kunmap_atomic(map, type); + put_page(page); -struct stack_frame { - const void __user *next_fp; - unsigned long return_address; -}; + len += size; + to += size; + addr += size; + + } while (len < n); + + return len; +} static int copy_stack_frame(const void __user *fp, struct stack_frame *frame) { - int ret; - - if (!access_ok(VERIFY_READ, fp, sizeof(*frame))) - return 0; + unsigned long bytes; - ret = 1; - pagefault_disable(); - if (__copy_from_user_inatomic(frame, fp, sizeof(*frame))) - ret = 0; - pagefault_enable(); + bytes = copy_from_user_nmi(frame, fp, sizeof(*frame)); - return ret; + return bytes == sizeof(*frame); } static void @@ -1644,28 +1655,28 @@ perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry) { struct stack_frame frame; const void __user *fp; - int nr = entry->nr; - regs = (struct pt_regs *)current->thread.sp0 - 1; - fp = (void __user *)regs->bp; + if (!user_mode(regs)) + regs = task_pt_regs(current); + + fp = (void __user *)regs->bp; + callchain_store(entry, PERF_CONTEXT_USER); callchain_store(entry, regs->ip); - while (entry->nr < MAX_STACK_DEPTH) { - frame.next_fp = NULL; + while (entry->nr < PERF_MAX_STACK_DEPTH) { + frame.next_frame = NULL; frame.return_address = 0; if (!copy_stack_frame(fp, &frame)) break; - if ((unsigned long)fp < user_stack_pointer(regs)) + if ((unsigned long)fp < regs->sp) break; callchain_store(entry, frame.return_address); - fp = frame.next_fp; + fp = frame.next_frame; } - - entry->user = entry->nr - nr; } static void @@ -1701,9 +1712,6 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) entry = &__get_cpu_var(irq_entry); entry->nr = 0; - entry->hv = 0; - entry->kernel = 0; - entry->user = 0; perf_do_callchain(regs, entry); diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index d6f5b9fbde3..5c481f6205b 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -716,11 +716,15 @@ static void probe_nmi_watchdog(void) wd_ops = &k7_wd_ops; break; case X86_VENDOR_INTEL: - /* - * Work around Core Duo (Yonah) errata AE49 where perfctr1 - * doesn't have a working enable bit. + /* Work around where perfctr1 doesn't have a working enable + * bit as described in the following errata: + * AE49 Core Duo and Intel Core Solo 65 nm + * AN49 Intel Pentium Dual-Core + * AF49 Dual-Core Intel Xeon Processor LV */ - if (boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 14) { + if ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 14) || + ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 15 && + boot_cpu_data.x86_mask == 4))) { intel_arch_wd_ops.perfctr = MSR_ARCH_PERFMON_PERFCTR0; intel_arch_wd_ops.evntsel = MSR_ARCH_PERFMON_EVENTSEL0; } diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index ff958248e61..5e409dc298a 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -27,6 +27,7 @@ #include <asm/cpu.h> #include <asm/reboot.h> #include <asm/virtext.h> +#include <asm/iommu.h> #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) @@ -103,5 +104,10 @@ void native_machine_crash_shutdown(struct pt_regs *regs) #ifdef CONFIG_HPET_TIMER hpet_disable(); #endif + +#ifdef CONFIG_X86_64 + pci_iommu_shutdown(); +#endif + crash_save_cpu(regs, safe_smp_processor_id()); } diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index 95ea5fa7d44..c8405718a4c 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -22,6 +22,7 @@ #include "dumpstack.h" int panic_on_unrecovered_nmi; +int panic_on_io_nmi; unsigned int code_bytes = 64; int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE; static int die_counter; diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 7271fa33d79..c4ca89d9aaf 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -1383,6 +1383,8 @@ static unsigned long ram_alignment(resource_size_t pos) return 32*1024*1024; } +#define MAX_RESOURCE_SIZE ((resource_size_t)-1) + void __init e820_reserve_resources_late(void) { int i; @@ -1400,17 +1402,19 @@ void __init e820_reserve_resources_late(void) * avoid stolen RAM: */ for (i = 0; i < e820.nr_map; i++) { - struct e820entry *entry = &e820_saved.map[i]; - resource_size_t start, end; + struct e820entry *entry = &e820.map[i]; + u64 start, end; if (entry->type != E820_RAM) continue; start = entry->addr + entry->size; - end = round_up(start, ram_alignment(start)); - if (start == end) + end = round_up(start, ram_alignment(start)) - 1; + if (end > MAX_RESOURCE_SIZE) + end = MAX_RESOURCE_SIZE; + if (start >= end) continue; - reserve_region_with_split(&iomem_resource, start, - end - 1, "RAM buffer"); + reserve_region_with_split(&iomem_resource, start, end, + "RAM buffer"); } } diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index 1736acc4d7a..96f7ac0bbf0 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -240,10 +240,35 @@ static void __init do_add_efi_memmap(void) unsigned long long size = md->num_pages << EFI_PAGE_SHIFT; int e820_type; - if (md->attribute & EFI_MEMORY_WB) - e820_type = E820_RAM; - else + switch (md->type) { + case EFI_LOADER_CODE: + case EFI_LOADER_DATA: + case EFI_BOOT_SERVICES_CODE: + case EFI_BOOT_SERVICES_DATA: + case EFI_CONVENTIONAL_MEMORY: + if (md->attribute & EFI_MEMORY_WB) + e820_type = E820_RAM; + else + e820_type = E820_RESERVED; + break; + case EFI_ACPI_RECLAIM_MEMORY: + e820_type = E820_ACPI; + break; + case EFI_ACPI_MEMORY_NVS: + e820_type = E820_NVS; + break; + case EFI_UNUSABLE_MEMORY: + e820_type = E820_UNUSABLE; + break; + default: + /* + * EFI_RESERVED_TYPE EFI_RUNTIME_SERVICES_CODE + * EFI_RUNTIME_SERVICES_DATA EFI_MEMORY_MAPPED_IO + * EFI_MEMORY_MAPPED_IO_PORT_SPACE EFI_PAL_CODE + */ e820_type = E820_RESERVED; + break; + } e820_add_region(start, size, e820_type); } sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index c929add475c..c097e7d607c 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -48,7 +48,6 @@ #include <asm/segment.h> #include <asm/smp.h> #include <asm/page_types.h> -#include <asm/desc.h> #include <asm/percpu.h> #include <asm/dwarf2.h> #include <asm/processor-flags.h> @@ -84,7 +83,7 @@ #define preempt_stop(clobbers) DISABLE_INTERRUPTS(clobbers); TRACE_IRQS_OFF #else #define preempt_stop(clobbers) -#define resume_kernel restore_nocheck +#define resume_kernel restore_all #endif .macro TRACE_IRQS_IRET @@ -372,7 +371,7 @@ END(ret_from_exception) ENTRY(resume_kernel) DISABLE_INTERRUPTS(CLBR_ANY) cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ? - jnz restore_nocheck + jnz restore_all need_resched: movl TI_flags(%ebp), %ecx # need_resched set ? testb $_TIF_NEED_RESCHED, %cl @@ -540,6 +539,8 @@ syscall_exit: jne syscall_exit_work restore_all: + TRACE_IRQS_IRET +restore_all_notrace: movl PT_EFLAGS(%esp), %eax # mix EFLAGS, SS and CS # Warning: PT_OLDSS(%esp) contains the wrong/random values if we # are returning to the kernel. @@ -551,8 +552,6 @@ restore_all: CFI_REMEMBER_STATE je ldt_ss # returning to user-space with LDT SS restore_nocheck: - TRACE_IRQS_IRET -restore_nocheck_notrace: RESTORE_REGS 4 # skip orig_eax/error_code CFI_ADJUST_CFA_OFFSET -4 irq_return: @@ -588,22 +587,34 @@ ldt_ss: jne restore_nocheck #endif - /* If returning to userspace with 16bit stack, - * try to fix the higher word of ESP, as the CPU - * won't restore it. - * This is an "official" bug of all the x86-compatible - * CPUs, which we can try to work around to make - * dosemu and wine happy. */ - movl PT_OLDESP(%esp), %eax - movl %esp, %edx - call patch_espfix_desc +/* + * Setup and switch to ESPFIX stack + * + * We're returning to userspace with a 16 bit stack. The CPU will not + * restore the high word of ESP for us on executing iret... This is an + * "official" bug of all the x86-compatible CPUs, which we can work + * around to make dosemu and wine happy. We do this by preloading the + * high word of ESP with the high word of the userspace ESP while + * compensating for the offset by changing to the ESPFIX segment with + * a base address that matches for the difference. + */ + mov %esp, %edx /* load kernel esp */ + mov PT_OLDESP(%esp), %eax /* load userspace esp */ + mov %dx, %ax /* eax: new kernel esp */ + sub %eax, %edx /* offset (low word is 0) */ + PER_CPU(gdt_page, %ebx) + shr $16, %edx + mov %dl, GDT_ENTRY_ESPFIX_SS * 8 + 4(%ebx) /* bits 16..23 */ + mov %dh, GDT_ENTRY_ESPFIX_SS * 8 + 7(%ebx) /* bits 24..31 */ pushl $__ESPFIX_SS CFI_ADJUST_CFA_OFFSET 4 - pushl %eax + push %eax /* new kernel esp */ CFI_ADJUST_CFA_OFFSET 4 + /* Disable interrupts, but do not irqtrace this section: we + * will soon execute iret and the tracer was already set to + * the irqstate after the iret */ DISABLE_INTERRUPTS(CLBR_EAX) - TRACE_IRQS_OFF - lss (%esp), %esp + lss (%esp), %esp /* switch to espfix segment */ CFI_ADJUST_CFA_OFFSET -8 jmp restore_nocheck CFI_ENDPROC @@ -716,15 +727,24 @@ PTREGSCALL(vm86) PTREGSCALL(vm86old) .macro FIXUP_ESPFIX_STACK - /* since we are on a wrong stack, we cant make it a C code :( */ +/* + * Switch back for ESPFIX stack to the normal zerobased stack + * + * We can't call C functions using the ESPFIX stack. This code reads + * the high word of the segment base from the GDT and swiches to the + * normal stack and adjusts ESP with the matching offset. + */ + /* fixup the stack */ PER_CPU(gdt_page, %ebx) - GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah) - addl %esp, %eax + mov GDT_ENTRY_ESPFIX_SS * 8 + 4(%ebx), %al /* bits 16..23 */ + mov GDT_ENTRY_ESPFIX_SS * 8 + 7(%ebx), %ah /* bits 24..31 */ + shl $16, %eax + addl %esp, %eax /* the adjusted stack pointer */ pushl $__KERNEL_DS CFI_ADJUST_CFA_OFFSET 4 pushl %eax CFI_ADJUST_CFA_OFFSET 4 - lss (%esp), %esp + lss (%esp), %esp /* switch to the normal stack segment */ CFI_ADJUST_CFA_OFFSET -8 .endm .macro UNWIND_ESPFIX_STACK @@ -1154,6 +1174,7 @@ ENTRY(ftrace_graph_caller) pushl %edx movl 0xc(%esp), %edx lea 0x4(%ebp), %eax + movl (%ebp), %ecx subl $MCOUNT_INSN_SIZE, %edx call prepare_ftrace_return popl %edx @@ -1168,6 +1189,7 @@ return_to_handler: pushl %eax pushl %ecx pushl %edx + movl %ebp, %eax call ftrace_return_to_handler movl %eax, 0xc(%esp) popl %edx @@ -1329,7 +1351,7 @@ nmi_stack_correct: xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_nmi - jmp restore_nocheck_notrace + jmp restore_all_notrace CFI_ENDPROC nmi_stack_fixup: diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index de74f0a3e0e..c251be74510 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -135,6 +135,7 @@ ENTRY(ftrace_graph_caller) leaq 8(%rbp), %rdi movq 0x38(%rsp), %rsi + movq (%rbp), %rdx subq $MCOUNT_INSN_SIZE, %rsi call prepare_ftrace_return @@ -150,6 +151,7 @@ GLOBAL(return_to_handler) /* Save the return values */ movq %rax, (%rsp) movq %rdx, 8(%rsp) + movq %rbp, %rdi call ftrace_return_to_handler diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index b79c5533c42..d94e1ea3b9f 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -408,7 +408,8 @@ int ftrace_disable_ftrace_graph_caller(void) * Hook the return address and push it in the stack of return addrs * in current thread info. */ -void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) +void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, + unsigned long frame_pointer) { unsigned long old; int faulted; @@ -453,7 +454,8 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) return; } - if (ftrace_push_return_trace(old, self_addr, &trace.depth) == -EBUSY) { + if (ftrace_push_return_trace(old, self_addr, &trace.depth, + frame_pointer) == -EBUSY) { *parent = old; return; } diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index dc5ed4bdd88..8663afb5653 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -13,7 +13,6 @@ #include <asm/segment.h> #include <asm/page_types.h> #include <asm/pgtable_types.h> -#include <asm/desc.h> #include <asm/cache.h> #include <asm/thread_info.h> #include <asm/asm-offsets.h> diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index 54b29bb24e7..fa54f78e2a0 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -12,7 +12,6 @@ #include <linux/linkage.h> #include <linux/threads.h> #include <linux/init.h> -#include <asm/desc.h> #include <asm/segment.h> #include <asm/pgtable.h> #include <asm/page.h> diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 81408b93f88..dedc2bddf7a 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -510,7 +510,8 @@ static int hpet_setup_irq(struct hpet_dev *dev) { if (request_irq(dev->irq, hpet_interrupt_handler, - IRQF_DISABLED|IRQF_NOBALANCING, dev->name, dev)) + IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING, + dev->name, dev)) return -1; disable_irq(dev->irq); diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 745579bc825..1a041bcf506 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -32,6 +32,8 @@ int no_iommu __read_mostly; /* Set this to 1 if there is a HW IOMMU in the system */ int iommu_detected __read_mostly = 0; +int iommu_pass_through; + dma_addr_t bad_dma_address __read_mostly = 0; EXPORT_SYMBOL(bad_dma_address); @@ -210,6 +212,10 @@ static __init int iommu_setup(char *p) if (!strncmp(p, "soft", 4)) swiotlb = 1; #endif + if (!strncmp(p, "pt", 2)) { + iommu_pass_through = 1; + return 1; + } gart_parse_options(p); @@ -290,6 +296,8 @@ static int __init pci_iommu_init(void) void pci_iommu_shutdown(void) { gart_iommu_shutdown(); + + amd_iommu_shutdown(); } /* Must execute after PCI subsystem */ fs_initcall(pci_iommu_init); diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c index a1712f2b50f..6af96ee4420 100644 --- a/arch/x86/kernel/pci-swiotlb.c +++ b/arch/x86/kernel/pci-swiotlb.c @@ -71,7 +71,8 @@ void __init pci_swiotlb_init(void) { /* don't initialize swiotlb if iommu=off (no_iommu=1) */ #ifdef CONFIG_X86_64 - if (!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN) + if ((!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN) || + iommu_pass_through) swiotlb = 1; #endif if (swiotlb_force) diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index be5ae80f897..de2cab13284 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -289,6 +289,20 @@ void * __init extend_brk(size_t size, size_t align) return ret; } +#ifdef CONFIG_X86_64 +static void __init init_gbpages(void) +{ + if (direct_gbpages && cpu_has_gbpages) + printk(KERN_INFO "Using GB pages for direct mapping\n"); + else + direct_gbpages = 0; +} +#else +static inline void init_gbpages(void) +{ +} +#endif + static void __init reserve_brk(void) { if (_brk_end > _brk_start) @@ -871,6 +885,8 @@ void __init setup_arch(char **cmdline_p) reserve_brk(); + init_gbpages(); + /* max_pfn_mapped is updated here */ max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT); max_pfn_mapped = max_low_pfn_mapped; diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 9c3f0823e6a..29a3eef7cf4 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -124,7 +124,7 @@ static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size, } /* - * Remap allocator + * Large page remap allocator * * This allocator uses PMD page as unit. A PMD page is allocated for * each cpu and each is remapped into vmalloc area using PMD mapping. @@ -137,105 +137,185 @@ static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size, * better than only using 4k mappings while still being NUMA friendly. */ #ifdef CONFIG_NEED_MULTIPLE_NODES -static size_t pcpur_size __initdata; -static void **pcpur_ptrs __initdata; +struct pcpul_ent { + unsigned int cpu; + void *ptr; +}; + +static size_t pcpul_size; +static struct pcpul_ent *pcpul_map; +static struct vm_struct pcpul_vm; -static struct page * __init pcpur_get_page(unsigned int cpu, int pageno) +static struct page * __init pcpul_get_page(unsigned int cpu, int pageno) { size_t off = (size_t)pageno << PAGE_SHIFT; - if (off >= pcpur_size) + if (off >= pcpul_size) return NULL; - return virt_to_page(pcpur_ptrs[cpu] + off); + return virt_to_page(pcpul_map[cpu].ptr + off); } -static ssize_t __init setup_pcpu_remap(size_t static_size) +static ssize_t __init setup_pcpu_lpage(size_t static_size, bool chosen) { - static struct vm_struct vm; - size_t ptrs_size, dyn_size; + size_t map_size, dyn_size; unsigned int cpu; + int i, j; ssize_t ret; - /* - * If large page isn't supported, there's no benefit in doing - * this. Also, on non-NUMA, embedding is better. - * - * NOTE: disabled for now. - */ - if (true || !cpu_has_pse || !pcpu_need_numa()) + if (!chosen) { + size_t vm_size = VMALLOC_END - VMALLOC_START; + size_t tot_size = num_possible_cpus() * PMD_SIZE; + + /* on non-NUMA, embedding is better */ + if (!pcpu_need_numa()) + return -EINVAL; + + /* don't consume more than 20% of vmalloc area */ + if (tot_size > vm_size / 5) { + pr_info("PERCPU: too large chunk size %zuMB for " + "large page remap\n", tot_size >> 20); + return -EINVAL; + } + } + + /* need PSE */ + if (!cpu_has_pse) { + pr_warning("PERCPU: lpage allocator requires PSE\n"); return -EINVAL; + } /* * Currently supports only single page. Supporting multiple * pages won't be too difficult if it ever becomes necessary. */ - pcpur_size = PFN_ALIGN(static_size + PERCPU_MODULE_RESERVE + + pcpul_size = PFN_ALIGN(static_size + PERCPU_MODULE_RESERVE + PERCPU_DYNAMIC_RESERVE); - if (pcpur_size > PMD_SIZE) { + if (pcpul_size > PMD_SIZE) { pr_warning("PERCPU: static data is larger than large page, " "can't use large page\n"); return -EINVAL; } - dyn_size = pcpur_size - static_size - PERCPU_FIRST_CHUNK_RESERVE; + dyn_size = pcpul_size - static_size - PERCPU_FIRST_CHUNK_RESERVE; /* allocate pointer array and alloc large pages */ - ptrs_size = PFN_ALIGN(num_possible_cpus() * sizeof(pcpur_ptrs[0])); - pcpur_ptrs = alloc_bootmem(ptrs_size); + map_size = PFN_ALIGN(num_possible_cpus() * sizeof(pcpul_map[0])); + pcpul_map = alloc_bootmem(map_size); for_each_possible_cpu(cpu) { - pcpur_ptrs[cpu] = pcpu_alloc_bootmem(cpu, PMD_SIZE, PMD_SIZE); - if (!pcpur_ptrs[cpu]) + pcpul_map[cpu].cpu = cpu; + pcpul_map[cpu].ptr = pcpu_alloc_bootmem(cpu, PMD_SIZE, + PMD_SIZE); + if (!pcpul_map[cpu].ptr) { + pr_warning("PERCPU: failed to allocate large page " + "for cpu%u\n", cpu); goto enomem; + } /* - * Only use pcpur_size bytes and give back the rest. + * Only use pcpul_size bytes and give back the rest. * * Ingo: The 2MB up-rounding bootmem is needed to make * sure the partial 2MB page is still fully RAM - it's * not well-specified to have a PAT-incompatible area * (unmapped RAM, device memory, etc.) in that hole. */ - free_bootmem(__pa(pcpur_ptrs[cpu] + pcpur_size), - PMD_SIZE - pcpur_size); + free_bootmem(__pa(pcpul_map[cpu].ptr + pcpul_size), + PMD_SIZE - pcpul_size); - memcpy(pcpur_ptrs[cpu], __per_cpu_load, static_size); + memcpy(pcpul_map[cpu].ptr, __per_cpu_load, static_size); } /* allocate address and map */ - vm.flags = VM_ALLOC; - vm.size = num_possible_cpus() * PMD_SIZE; - vm_area_register_early(&vm, PMD_SIZE); + pcpul_vm.flags = VM_ALLOC; + pcpul_vm.size = num_possible_cpus() * PMD_SIZE; + vm_area_register_early(&pcpul_vm, PMD_SIZE); for_each_possible_cpu(cpu) { - pmd_t *pmd; + pmd_t *pmd, pmd_v; - pmd = populate_extra_pmd((unsigned long)vm.addr - + cpu * PMD_SIZE); - set_pmd(pmd, pfn_pmd(page_to_pfn(virt_to_page(pcpur_ptrs[cpu])), - PAGE_KERNEL_LARGE)); + pmd = populate_extra_pmd((unsigned long)pcpul_vm.addr + + cpu * PMD_SIZE); + pmd_v = pfn_pmd(page_to_pfn(virt_to_page(pcpul_map[cpu].ptr)), + PAGE_KERNEL_LARGE); + set_pmd(pmd, pmd_v); } /* we're ready, commit */ pr_info("PERCPU: Remapped at %p with large pages, static data " - "%zu bytes\n", vm.addr, static_size); + "%zu bytes\n", pcpul_vm.addr, static_size); - ret = pcpu_setup_first_chunk(pcpur_get_page, static_size, + ret = pcpu_setup_first_chunk(pcpul_get_page, static_size, PERCPU_FIRST_CHUNK_RESERVE, dyn_size, - PMD_SIZE, vm.addr, NULL); - goto out_free_ar; + PMD_SIZE, pcpul_vm.addr, NULL); + + /* sort pcpul_map array for pcpu_lpage_remapped() */ + for (i = 0; i < num_possible_cpus() - 1; i++) + for (j = i + 1; j < num_possible_cpus(); j++) + if (pcpul_map[i].ptr > pcpul_map[j].ptr) { + struct pcpul_ent tmp = pcpul_map[i]; + pcpul_map[i] = pcpul_map[j]; + pcpul_map[j] = tmp; + } + + return ret; enomem: for_each_possible_cpu(cpu) - if (pcpur_ptrs[cpu]) - free_bootmem(__pa(pcpur_ptrs[cpu]), PMD_SIZE); - ret = -ENOMEM; -out_free_ar: - free_bootmem(__pa(pcpur_ptrs), ptrs_size); - return ret; + if (pcpul_map[cpu].ptr) + free_bootmem(__pa(pcpul_map[cpu].ptr), pcpul_size); + free_bootmem(__pa(pcpul_map), map_size); + return -ENOMEM; +} + +/** + * pcpu_lpage_remapped - determine whether a kaddr is in pcpul recycled area + * @kaddr: the kernel address in question + * + * Determine whether @kaddr falls in the pcpul recycled area. This is + * used by pageattr to detect VM aliases and break up the pcpu PMD + * mapping such that the same physical page is not mapped under + * different attributes. + * + * The recycled area is always at the tail of a partially used PMD + * page. + * + * RETURNS: + * Address of corresponding remapped pcpu address if match is found; + * otherwise, NULL. + */ +void *pcpu_lpage_remapped(void *kaddr) +{ + void *pmd_addr = (void *)((unsigned long)kaddr & PMD_MASK); + unsigned long offset = (unsigned long)kaddr & ~PMD_MASK; + int left = 0, right = num_possible_cpus() - 1; + int pos; + + /* pcpul in use at all? */ + if (!pcpul_map) + return NULL; + + /* okay, perform binary search */ + while (left <= right) { + pos = (left + right) / 2; + + if (pcpul_map[pos].ptr < pmd_addr) + left = pos + 1; + else if (pcpul_map[pos].ptr > pmd_addr) + right = pos - 1; + else { + /* it shouldn't be in the area for the first chunk */ + WARN_ON(offset < pcpul_size); + + return pcpul_vm.addr + + pcpul_map[pos].cpu * PMD_SIZE + offset; + } + } + + return NULL; } #else -static ssize_t __init setup_pcpu_remap(size_t static_size) +static ssize_t __init setup_pcpu_lpage(size_t static_size, bool chosen) { return -EINVAL; } @@ -249,7 +329,7 @@ static ssize_t __init setup_pcpu_remap(size_t static_size) * mapping so that it can use PMD mapping without additional TLB * pressure. */ -static ssize_t __init setup_pcpu_embed(size_t static_size) +static ssize_t __init setup_pcpu_embed(size_t static_size, bool chosen) { size_t reserve = PERCPU_MODULE_RESERVE + PERCPU_DYNAMIC_RESERVE; @@ -258,7 +338,7 @@ static ssize_t __init setup_pcpu_embed(size_t static_size) * this. Also, embedding allocation doesn't play well with * NUMA. */ - if (!cpu_has_pse || pcpu_need_numa()) + if (!chosen && (!cpu_has_pse || pcpu_need_numa())) return -EINVAL; return pcpu_embed_first_chunk(static_size, PERCPU_FIRST_CHUNK_RESERVE, @@ -308,8 +388,11 @@ static ssize_t __init setup_pcpu_4k(size_t static_size) void *ptr; ptr = pcpu_alloc_bootmem(cpu, PAGE_SIZE, PAGE_SIZE); - if (!ptr) + if (!ptr) { + pr_warning("PERCPU: failed to allocate " + "4k page for cpu%u\n", cpu); goto enomem; + } memcpy(ptr, __per_cpu_load + i * PAGE_SIZE, PAGE_SIZE); pcpu4k_pages[j++] = virt_to_page(ptr); @@ -333,6 +416,16 @@ out_free_ar: return ret; } +/* for explicit first chunk allocator selection */ +static char pcpu_chosen_alloc[16] __initdata; + +static int __init percpu_alloc_setup(char *str) +{ + strncpy(pcpu_chosen_alloc, str, sizeof(pcpu_chosen_alloc) - 1); + return 0; +} +early_param("percpu_alloc", percpu_alloc_setup); + static inline void setup_percpu_segment(int cpu) { #ifdef CONFIG_X86_32 @@ -346,11 +439,6 @@ static inline void setup_percpu_segment(int cpu) #endif } -/* - * Great future plan: - * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data. - * Always point %gs to its beginning - */ void __init setup_per_cpu_areas(void) { size_t static_size = __per_cpu_end - __per_cpu_start; @@ -367,9 +455,26 @@ void __init setup_per_cpu_areas(void) * of large page mappings. Please read comments on top of * each allocator for details. */ - ret = setup_pcpu_remap(static_size); - if (ret < 0) - ret = setup_pcpu_embed(static_size); + ret = -EINVAL; + if (strlen(pcpu_chosen_alloc)) { + if (strcmp(pcpu_chosen_alloc, "4k")) { + if (!strcmp(pcpu_chosen_alloc, "lpage")) + ret = setup_pcpu_lpage(static_size, true); + else if (!strcmp(pcpu_chosen_alloc, "embed")) + ret = setup_pcpu_embed(static_size, true); + else + pr_warning("PERCPU: unknown allocator %s " + "specified\n", pcpu_chosen_alloc); + if (ret < 0) + pr_warning("PERCPU: %s allocator failed (%zd), " + "falling back to 4k\n", + pcpu_chosen_alloc, ret); + } + } else { + ret = setup_pcpu_lpage(static_size, false); + if (ret < 0) + ret = setup_pcpu_embed(static_size, false); + } if (ret < 0) ret = setup_pcpu_4k(static_size); if (ret < 0) diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 124d40c575d..8ccabb8a2f6 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -711,7 +711,6 @@ uv_activation_descriptor_init(int node, int pnode) unsigned long pa; unsigned long m; unsigned long n; - unsigned long mmr_image; struct bau_desc *adp; struct bau_desc *ad2; @@ -727,12 +726,8 @@ uv_activation_descriptor_init(int node, int pnode) n = pa >> uv_nshift; m = pa & uv_mmask; - mmr_image = uv_read_global_mmr64(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE); - if (mmr_image) { - uv_write_global_mmr64(pnode, (unsigned long) - UVH_LB_BAU_SB_DESCRIPTOR_BASE, - (n << UV_DESC_BASE_PNODE_SHIFT | m)); - } + uv_write_global_mmr64(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE, + (n << UV_DESC_BASE_PNODE_SHIFT | m)); /* * initializing all 8 (UV_ITEMS_PER_DESCRIPTOR) descriptors for each diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 5f935f0d586..5204332f475 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -54,6 +54,7 @@ #include <asm/traps.h> #include <asm/desc.h> #include <asm/i387.h> +#include <asm/mce.h> #include <asm/mach_traps.h> @@ -65,8 +66,6 @@ #include <asm/setup.h> #include <asm/traps.h> -#include "cpu/mcheck/mce.h" - asmlinkage int system_call(void); /* Do we ignore FPU interrupts ? */ @@ -347,6 +346,9 @@ io_check_error(unsigned char reason, struct pt_regs *regs) printk(KERN_EMERG "NMI: IOCK error (debug interrupt?)\n"); show_registers(regs); + if (panic_on_io_nmi) + panic("NMI IOCK error: Not continuing"); + /* Re-enable the IOCK line, wait for a few seconds */ reason = (reason & 0xf) | 8; outb(reason, 0x61); diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index b0597ad02c9..6e1a368d21d 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -590,22 +590,26 @@ EXPORT_SYMBOL(recalibrate_cpu_khz); */ DEFINE_PER_CPU(unsigned long, cyc2ns); +DEFINE_PER_CPU(unsigned long long, cyc2ns_offset); static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu) { - unsigned long long tsc_now, ns_now; + unsigned long long tsc_now, ns_now, *offset; unsigned long flags, *scale; local_irq_save(flags); sched_clock_idle_sleep_event(); scale = &per_cpu(cyc2ns, cpu); + offset = &per_cpu(cyc2ns_offset, cpu); rdtscll(tsc_now); ns_now = __cycles_2_ns(tsc_now); - if (cpu_khz) + if (cpu_khz) { *scale = (NSEC_PER_MSEC << CYC2NS_SCALE_FACTOR)/cpu_khz; + *offset = ns_now - (tsc_now * *scale >> CYC2NS_SCALE_FACTOR); + } sched_clock_idle_wakeup_event(0); local_irq_restore(flags); |