diff options
Diffstat (limited to 'arch/x86/kernel')
127 files changed, 8199 insertions, 10292 deletions
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 50632e16d01c..eb074530c7d3 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -6,11 +6,13 @@ extra-y := head_$(BITS).o head$(BITS).o head.o init_task.o vmlinu CPPFLAGS_vmlinux.lds += -U$(UTS_MACHINE) -ifdef CONFIG_FTRACE +ifdef CONFIG_FUNCTION_TRACER # Do not profile debug and lowlevel utilities CFLAGS_REMOVE_tsc.o = -pg CFLAGS_REMOVE_rtc.o = -pg CFLAGS_REMOVE_paravirt-spinlocks.o = -pg +CFLAGS_REMOVE_ftrace.o = -pg +CFLAGS_REMOVE_early_printk.o = -pg endif # @@ -23,9 +25,9 @@ CFLAGS_hpet.o := $(nostackp) CFLAGS_tsc.o := $(nostackp) CFLAGS_paravirt.o := $(nostackp) -obj-y := process_$(BITS).o signal_$(BITS).o entry_$(BITS).o -obj-y += traps.o irq_$(BITS).o dumpstack_$(BITS).o -obj-y += time_$(BITS).o ioport.o ldt.o +obj-y := process_$(BITS).o signal.o entry_$(BITS).o +obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o +obj-y += time_$(BITS).o ioport.o ldt.o dumpstack.o obj-y += setup.o i8259.o irqinit_$(BITS).o setup_percpu.o obj-$(CONFIG_X86_VISWS) += visws_quirks.o obj-$(CONFIG_X86_32) += probe_roms_32.o @@ -41,7 +43,7 @@ obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o obj-y += process.o obj-y += i387.o xsave.o obj-y += ptrace.o -obj-y += ds.o +obj-$(CONFIG_X86_DS) += ds.o obj-$(CONFIG_X86_32) += tls.o obj-$(CONFIG_IA32_EMULATION) += tls.o obj-y += step.o @@ -61,10 +63,11 @@ obj-$(CONFIG_X86_32_SMP) += smpcommon.o obj-$(CONFIG_X86_64_SMP) += tsc_sync.o smpcommon.o obj-$(CONFIG_X86_TRAMPOLINE) += trampoline_$(BITS).o obj-$(CONFIG_X86_MPPARSE) += mpparse.o -obj-$(CONFIG_X86_LOCAL_APIC) += apic_$(BITS).o nmi.o -obj-$(CONFIG_X86_IO_APIC) += io_apic_$(BITS).o +obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o +obj-$(CONFIG_X86_IO_APIC) += io_apic.o obj-$(CONFIG_X86_REBOOTFIXUPS) += reboot_fixups_32.o obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o +obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += ftrace.o obj-$(CONFIG_KEXEC) += machine_kexec_$(BITS).o obj-$(CONFIG_KEXEC) += relocate_kernel_$(BITS).o crash.o obj-$(CONFIG_CRASH_DUMP) += crash_dump_$(BITS).o @@ -105,11 +108,15 @@ microcode-$(CONFIG_MICROCODE_INTEL) += microcode_intel.o microcode-$(CONFIG_MICROCODE_AMD) += microcode_amd.o obj-$(CONFIG_MICROCODE) += microcode.o +obj-$(CONFIG_X86_CHECK_BIOS_CORRUPTION) += check.o + +obj-$(CONFIG_SWIOTLB) += pci-swiotlb_64.o # NB rename without _64 + ### # 64 bit specific files ifeq ($(CONFIG_X86_64),y) obj-y += genapic_64.o genapic_flat_64.o genx2apic_uv_x.o tlb_uv.o - obj-y += bios_uv.o + obj-y += bios_uv.o uv_irq.o uv_sysfs.o obj-y += genx2apic_cluster.o obj-y += genx2apic_phys.o obj-$(CONFIG_X86_PM_TIMER) += pmtimer_64.o @@ -118,7 +125,6 @@ ifeq ($(CONFIG_X86_64),y) obj-$(CONFIG_GART_IOMMU) += pci-gart_64.o aperture_64.o obj-$(CONFIG_CALGARY_IOMMU) += pci-calgary_64.o tce_64.o obj-$(CONFIG_AMD_IOMMU) += amd_iommu_init.o amd_iommu.o - obj-$(CONFIG_SWIOTLB) += pci-swiotlb_64.o obj-$(CONFIG_PCI_MMCONFIG) += mmconf-fam10h_64.o endif diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index eb875cdc7367..65d0b72777ea 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -153,12 +153,13 @@ char *__init __acpi_map_table(unsigned long phys, unsigned long 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 acpi_mcfg_64bit_base_addr __initdata = FALSE; - static int __init acpi_mcfg_oem_check(struct acpi_table_mcfg *mcfg) { if (!strcmp(mcfg->header.oem_id, "SGI")) @@ -1136,7 +1137,7 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity) return gsi; } if (test_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed)) { - pr_debug(KERN_DEBUG "Pin %d-%d already programmed\n", + pr_debug("Pin %d-%d already programmed\n", mp_ioapic_routing[ioapic].apic_id, ioapic_pin); #ifdef CONFIG_X86_32 return (gsi < IRQ_COMPRESSION_START ? gsi : gsi_to_irq[gsi]); @@ -1256,7 +1257,7 @@ static int __init acpi_parse_madt_ioapic_entries(void) count = acpi_table_parse_madt(ACPI_MADT_TYPE_INTERRUPT_OVERRIDE, acpi_parse_int_src_ovr, - NR_IRQ_VECTORS); + nr_irqs); if (count < 0) { printk(KERN_ERR PREFIX "Error parsing interrupt source overrides entry\n"); @@ -1276,7 +1277,7 @@ static int __init acpi_parse_madt_ioapic_entries(void) count = acpi_table_parse_madt(ACPI_MADT_TYPE_NMI_SOURCE, acpi_parse_nmi_src, - NR_IRQ_VECTORS); + nr_irqs); if (count < 0) { printk(KERN_ERR PREFIX "Error parsing NMI SRC entry\n"); /* TBD: Cleanup to allow fallback to MPS */ @@ -1342,7 +1343,6 @@ static void __init acpi_process_madt(void) error = acpi_parse_madt_ioapic_entries(); if (!error) { acpi_irq_model = ACPI_IRQ_MODEL_IOAPIC; - acpi_irq_balance_set(NULL); acpi_ioapic = 1; smp_found_config = 1; @@ -1360,6 +1360,17 @@ static void __init acpi_process_madt(void) disable_acpi(); } } + + /* + * ACPI supports both logical (e.g. Hyper-Threading) and physical + * processors, where MPS only supports physical. + */ + if (acpi_lapic && acpi_ioapic) + printk(KERN_INFO "Using ACPI (MADT) for SMP configuration " + "information\n"); + else if (acpi_lapic) + printk(KERN_INFO "Using ACPI for processor (LAPIC) " + "configuration information\n"); #endif return; } @@ -1598,6 +1609,11 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 360"), }, }, + {} +}; + +/* second table for DMI checks that should run after early-quirks */ +static struct dmi_system_id __initdata acpi_dmi_table_late[] = { /* * HP laptops which use a DSDT reporting as HP/SB400/10000, * which includes some code which overrides all temperature @@ -1726,6 +1742,9 @@ int __init early_acpi_boot_init(void) int __init acpi_boot_init(void) { + /* those are executed after early-quirks are executed */ + dmi_check_system(acpi_dmi_table_late); + /* * If acpi_disabled, bail out * One exception: acpi=ht continues far enough to enumerate LAPICs diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 426e5d91b63a..806b4e9051b4 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -10,6 +10,7 @@ #include <linux/dmi.h> #include <linux/cpumask.h> #include <asm/segment.h> +#include <asm/desc.h> #include "realmode/wakeup.h" #include "sleep.h" @@ -21,7 +22,7 @@ unsigned long acpi_realmode_flags; static unsigned long acpi_realmode; #if defined(CONFIG_SMP) && defined(CONFIG_64BIT) -static char temp_stack[10240]; +static char temp_stack[4096]; #endif /** @@ -97,7 +98,9 @@ int acpi_save_state_mem(void) #else /* CONFIG_64BIT */ header->trampoline_segment = setup_trampoline() >> 4; #ifdef CONFIG_SMP - stack_start.sp = temp_stack + 4096; + stack_start.sp = temp_stack + sizeof(temp_stack); + early_gdt_descr.address = + (unsigned long)get_cpu_gdt_table(smp_processor_id()); #endif initial_code = (unsigned long)wakeup_long64; saved_magic = 0x123456789abcdef0; diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 34e4d112b1ef..2e2da717b350 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -24,6 +24,7 @@ #include <linux/iommu-helper.h> #include <asm/proto.h> #include <asm/iommu.h> +#include <asm/gart.h> #include <asm/amd_iommu_types.h> #include <asm/amd_iommu.h> @@ -50,7 +51,7 @@ static int dma_ops_unity_map(struct dma_ops_domain *dma_dom, /* returns !0 if the IOMMU is caching non-present entries in its TLB */ static int iommu_has_npcache(struct amd_iommu *iommu) { - return iommu->cap & IOMMU_CAP_NPCACHE; + return iommu->cap & (1UL << IOMMU_CAP_NPCACHE); } /**************************************************************************** @@ -187,6 +188,8 @@ static int iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd) spin_lock_irqsave(&iommu->lock, flags); ret = __iommu_queue_command(iommu, cmd); + if (!ret) + iommu->need_sync = 1; spin_unlock_irqrestore(&iommu->lock, flags); return ret; @@ -210,10 +213,13 @@ static int iommu_completion_wait(struct amd_iommu *iommu) cmd.data[0] = CMD_COMPL_WAIT_INT_MASK; CMD_SET_TYPE(&cmd, CMD_COMPL_WAIT); - iommu->need_sync = 0; - spin_lock_irqsave(&iommu->lock, flags); + if (!iommu->need_sync) + goto out; + + iommu->need_sync = 0; + ret = __iommu_queue_command(iommu, &cmd); if (ret) @@ -230,8 +236,9 @@ static int iommu_completion_wait(struct amd_iommu *iommu) status &= ~MMIO_STATUS_COM_WAIT_INT_MASK; writel(status, iommu->mmio_base + MMIO_STATUS_OFFSET); - if (unlikely((i == EXIT_LOOP_COUNT) && printk_ratelimit())) - printk(KERN_WARNING "AMD IOMMU: Completion wait loop failed\n"); + if (unlikely(i == EXIT_LOOP_COUNT)) + panic("AMD IOMMU: Completion wait loop failed\n"); + out: spin_unlock_irqrestore(&iommu->lock, flags); @@ -254,8 +261,6 @@ static int iommu_queue_inv_dev_entry(struct amd_iommu *iommu, u16 devid) ret = iommu_queue_command(iommu, &cmd); - iommu->need_sync = 1; - return ret; } @@ -281,8 +286,6 @@ static int iommu_queue_inv_iommu_pages(struct amd_iommu *iommu, ret = iommu_queue_command(iommu, &cmd); - iommu->need_sync = 1; - return ret; } @@ -295,7 +298,7 @@ static int iommu_flush_pages(struct amd_iommu *iommu, u16 domid, u64 address, size_t size) { int s = 0; - unsigned pages = iommu_num_pages(address, size); + unsigned pages = iommu_num_pages(address, size, PAGE_SIZE); address &= PAGE_MASK; @@ -343,7 +346,7 @@ static int iommu_map(struct protection_domain *dom, u64 __pte, *pte, *page; bus_addr = PAGE_ALIGN(bus_addr); - phys_addr = PAGE_ALIGN(bus_addr); + phys_addr = PAGE_ALIGN(phys_addr); /* only support 512GB address spaces for now */ if (bus_addr > IOMMU_MAP_SIZE_L3 || !(prot & IOMMU_PROT_MASK)) @@ -536,6 +539,9 @@ static void dma_ops_free_addresses(struct dma_ops_domain *dom, { address >>= PAGE_SHIFT; iommu_area_free(dom->bitmap, address, pages); + + if (address >= dom->next_bit) + dom->need_flush = true; } /**************************************************************************** @@ -596,7 +602,7 @@ static void dma_ops_free_pagetable(struct dma_ops_domain *dma_dom) continue; p2 = IOMMU_PTE_PAGE(p1[i]); - for (j = 0; j < 512; ++i) { + for (j = 0; j < 512; ++j) { if (!IOMMU_PTE_PRESENT(p2[j])) continue; p3 = IOMMU_PTE_PAGE(p2[j]); @@ -680,7 +686,8 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu, iommu->exclusion_start < dma_dom->aperture_size) { unsigned long startpage = iommu->exclusion_start >> PAGE_SHIFT; int pages = iommu_num_pages(iommu->exclusion_start, - iommu->exclusion_length); + iommu->exclusion_length, + PAGE_SIZE); dma_ops_reserve_addresses(dma_dom, startpage, pages); } @@ -758,8 +765,6 @@ static void set_device_domain(struct amd_iommu *iommu, write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); iommu_queue_inv_dev_entry(iommu, devid); - - iommu->need_sync = 1; } /***************************************************************************** @@ -854,6 +859,9 @@ static int get_device_resources(struct device *dev, print_devid(_bdf, 1); } + if (domain_for_device(_bdf) == NULL) + set_device_domain(*iommu, *domain, _bdf); + return 1; } @@ -904,7 +912,7 @@ static void dma_ops_domain_unmap(struct amd_iommu *iommu, if (address >= dom->aperture_size) return; - WARN_ON(address & 0xfffULL || address > dom->aperture_size); + WARN_ON(address & ~PAGE_MASK || address >= dom->aperture_size); pte = dom->pte_pages[IOMMU_PTE_L1_INDEX(address)]; pte += IOMMU_PTE_L0_INDEX(address); @@ -916,8 +924,8 @@ static void dma_ops_domain_unmap(struct amd_iommu *iommu, /* * This function contains common code for mapping of a physically - * contiguous memory region into DMA address space. It is uses by all - * mapping functions provided by this IOMMU driver. + * contiguous memory region into DMA address space. It is used by all + * mapping functions provided with this IOMMU driver. * Must be called with the domain lock held. */ static dma_addr_t __map_single(struct device *dev, @@ -935,7 +943,7 @@ static dma_addr_t __map_single(struct device *dev, unsigned long align_mask = 0; int i; - pages = iommu_num_pages(paddr, size); + pages = iommu_num_pages(paddr, size, PAGE_SIZE); paddr &= PAGE_MASK; if (align) @@ -977,10 +985,11 @@ static void __unmap_single(struct amd_iommu *iommu, dma_addr_t i, start; unsigned int pages; - if ((dma_addr == 0) || (dma_addr + size > dma_dom->aperture_size)) + if ((dma_addr == bad_dma_address) || + (dma_addr + size > dma_dom->aperture_size)) return; - pages = iommu_num_pages(dma_addr, size); + pages = iommu_num_pages(dma_addr, size, PAGE_SIZE); dma_addr &= PAGE_MASK; start = dma_addr; @@ -991,8 +1000,10 @@ static void __unmap_single(struct amd_iommu *iommu, dma_ops_free_addresses(dma_dom, dma_addr, pages); - if (amd_iommu_unmap_flush) + if (amd_iommu_unmap_flush || dma_dom->need_flush) { iommu_flush_pages(iommu, dma_dom->domain.id, dma_addr, size); + dma_dom->need_flush = false; + } } /* @@ -1025,8 +1036,7 @@ static dma_addr_t map_single(struct device *dev, phys_addr_t paddr, if (addr == bad_dma_address) goto out; - if (unlikely(iommu->need_sync)) - iommu_completion_wait(iommu); + iommu_completion_wait(iommu); out: spin_unlock_irqrestore(&domain->lock, flags); @@ -1054,8 +1064,7 @@ static void unmap_single(struct device *dev, dma_addr_t dma_addr, __unmap_single(iommu, domain->priv, dma_addr, size, dir); - if (unlikely(iommu->need_sync)) - iommu_completion_wait(iommu); + iommu_completion_wait(iommu); spin_unlock_irqrestore(&domain->lock, flags); } @@ -1121,8 +1130,7 @@ static int map_sg(struct device *dev, struct scatterlist *sglist, goto unmap; } - if (unlikely(iommu->need_sync)) - iommu_completion_wait(iommu); + iommu_completion_wait(iommu); out: spin_unlock_irqrestore(&domain->lock, flags); @@ -1167,8 +1175,7 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist, s->dma_address = s->dma_length = 0; } - if (unlikely(iommu->need_sync)) - iommu_completion_wait(iommu); + iommu_completion_wait(iommu); spin_unlock_irqrestore(&domain->lock, flags); } @@ -1219,8 +1226,7 @@ static void *alloc_coherent(struct device *dev, size_t size, goto out; } - if (unlikely(iommu->need_sync)) - iommu_completion_wait(iommu); + iommu_completion_wait(iommu); out: spin_unlock_irqrestore(&domain->lock, flags); @@ -1251,8 +1257,7 @@ static void free_coherent(struct device *dev, size_t size, __unmap_single(iommu, domain->priv, dma_addr, size, DMA_BIDIRECTIONAL); - if (unlikely(iommu->need_sync)) - iommu_completion_wait(iommu); + iommu_completion_wait(iommu); spin_unlock_irqrestore(&domain->lock, flags); diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 4cd8083c58be..c625800c55ca 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -28,6 +28,7 @@ #include <asm/amd_iommu_types.h> #include <asm/amd_iommu.h> #include <asm/iommu.h> +#include <asm/gart.h> /* * definitions for the ACPI scanning code @@ -121,7 +122,7 @@ u16 amd_iommu_last_bdf; /* largest PCI device id we have LIST_HEAD(amd_iommu_unity_map); /* a list of required unity mappings we find in ACPI */ unsigned amd_iommu_aperture_order = 26; /* size of aperture in power of 2 */ -int amd_iommu_isolate; /* if 1, device isolation is enabled */ +int amd_iommu_isolate = 1; /* if 1, device isolation is enabled */ bool amd_iommu_unmap_flush; /* if true, flush on every unmap */ LIST_HEAD(amd_iommu_list); /* list of all AMD IOMMUs in the @@ -212,7 +213,7 @@ static void __init iommu_set_exclusion_range(struct amd_iommu *iommu) /* Programs the physical address of the device table into the IOMMU hardware */ static void __init iommu_set_device_table(struct amd_iommu *iommu) { - u32 entry; + u64 entry; BUG_ON(iommu->mmio_base == NULL); @@ -427,6 +428,10 @@ static u8 * __init alloc_command_buffer(struct amd_iommu *iommu) memcpy_toio(iommu->mmio_base + MMIO_CMD_BUF_OFFSET, &entry, sizeof(entry)); + /* set head and tail to zero manually */ + writel(0x00, iommu->mmio_base + MMIO_CMD_HEAD_OFFSET); + writel(0x00, iommu->mmio_base + MMIO_CMD_TAIL_OFFSET); + iommu_feature_enable(iommu, CONTROL_CMDBUF_EN); return cmd_buf; @@ -1074,7 +1079,8 @@ int __init amd_iommu_init(void) goto free; /* IOMMU rlookup table - find the IOMMU for a specific device */ - amd_iommu_rlookup_table = (void *)__get_free_pages(GFP_KERNEL, + amd_iommu_rlookup_table = (void *)__get_free_pages( + GFP_KERNEL | __GFP_ZERO, get_order(rlookup_table_size)); if (amd_iommu_rlookup_table == NULL) goto free; @@ -1213,7 +1219,9 @@ static int __init parse_amd_iommu_options(char *str) for (; *str; ++str) { if (strncmp(str, "isolate", 7) == 0) amd_iommu_isolate = 1; - if (strncmp(str, "fullflush", 11) == 0) + if (strncmp(str, "share", 5) == 0) + amd_iommu_isolate = 0; + if (strncmp(str, "fullflush", 9) == 0) amd_iommu_unmap_flush = true; } diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 9a32b37ee2ee..676debfc1702 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -1,8 +1,9 @@ /* * Firmware replacement code. * - * Work around broken BIOSes that don't set an aperture or only set the - * aperture in the AGP bridge. + * Work around broken BIOSes that don't set an aperture, only set the + * aperture in the AGP bridge, or set too small aperture. + * * If all fails map the aperture over some low memory. This is cheaper than * doing bounce buffering. The memory is lost. This is done at early boot * because only the bootmem allocator can allocate 32+MB. diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic.c index 21c831d96af3..b5229affb953 100644 --- a/arch/x86/kernel/apic_32.c +++ b/arch/x86/kernel/apic.c @@ -23,11 +23,14 @@ #include <linux/mc146818rtc.h> #include <linux/kernel_stat.h> #include <linux/sysdev.h> +#include <linux/ioport.h> #include <linux/cpu.h> #include <linux/clockchips.h> #include <linux/acpi_pmtmr.h> #include <linux/module.h> #include <linux/dmi.h> +#include <linux/dmar.h> +#include <linux/ftrace.h> #include <asm/atomic.h> #include <asm/smp.h> @@ -36,8 +39,14 @@ #include <asm/desc.h> #include <asm/arch_hooks.h> #include <asm/hpet.h> +#include <asm/pgalloc.h> #include <asm/i8253.h> #include <asm/nmi.h> +#include <asm/idle.h> +#include <asm/proto.h> +#include <asm/timex.h> +#include <asm/apic.h> +#include <asm/i8259.h> #include <mach_apic.h> #include <mach_apicdef.h> @@ -50,16 +59,58 @@ # error SPURIOUS_APIC_VECTOR definition error #endif -unsigned long mp_lapic_addr; - +#ifdef CONFIG_X86_32 /* * Knob to control our willingness to enable the local APIC. * * +1=force-enable */ static int force_enable_local_apic; -int disable_apic; +/* + * APIC command line parameters + */ +static int __init parse_lapic(char *arg) +{ + force_enable_local_apic = 1; + return 0; +} +early_param("lapic", parse_lapic); +/* Local APIC was disabled by the BIOS and enabled by the kernel */ +static int enabled_via_apicbase; + +#endif +#ifdef CONFIG_X86_64 +static int apic_calibrate_pmtmr __initdata; +static __init int setup_apicpmtimer(char *s) +{ + apic_calibrate_pmtmr = 1; + notsc_setup(NULL); + return 0; +} +__setup("apicpmtimer", setup_apicpmtimer); +#endif + +#ifdef CONFIG_X86_64 +#define HAVE_X2APIC +#endif + +#ifdef HAVE_X2APIC +int x2apic; +/* x2apic enabled before OS handover */ +int x2apic_preenabled; +int disable_x2apic; +static __init int setup_nox2apic(char *str) +{ + disable_x2apic = 1; + setup_clear_cpu_cap(X86_FEATURE_X2APIC); + return 0; +} +early_param("nox2apic", setup_nox2apic); +#endif + +unsigned long mp_lapic_addr; +int disable_apic; /* Disable local APIC timer from the kernel commandline or via dmi quirk */ static int disable_apic_timer __cpuinitdata; /* Local APIC timer works in C2 */ @@ -110,9 +161,6 @@ static struct clock_event_device lapic_clockevent = { }; static DEFINE_PER_CPU(struct clock_event_device, lapic_events); -/* Local APIC was disabled by the BIOS and enabled by the kernel */ -static int enabled_via_apicbase; - static unsigned long apic_phys; /* @@ -202,6 +250,42 @@ static struct apic_ops xapic_ops = { struct apic_ops __read_mostly *apic_ops = &xapic_ops; EXPORT_SYMBOL_GPL(apic_ops); +#ifdef HAVE_X2APIC +static void x2apic_wait_icr_idle(void) +{ + /* no need to wait for icr idle in x2apic */ + return; +} + +static u32 safe_x2apic_wait_icr_idle(void) +{ + /* no need to wait for icr idle in x2apic */ + return 0; +} + +void x2apic_icr_write(u32 low, u32 id) +{ + wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low); +} + +u64 x2apic_icr_read(void) +{ + unsigned long val; + + rdmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), val); + return val; +} + +static struct apic_ops x2apic_ops = { + .read = native_apic_msr_read, + .write = native_apic_msr_write, + .icr_read = x2apic_icr_read, + .icr_write = x2apic_icr_write, + .wait_icr_idle = x2apic_wait_icr_idle, + .safe_wait_icr_idle = safe_x2apic_wait_icr_idle, +}; +#endif + /** * enable_NMI_through_LVT0 - enable NMI through local vector table 0 */ @@ -219,6 +303,7 @@ void __cpuinit enable_NMI_through_LVT0(void) apic_write(APIC_LVT0, v); } +#ifdef CONFIG_X86_32 /** * get_physical_broadcast - Get number of physical broadcast IDs */ @@ -226,6 +311,7 @@ int get_physical_broadcast(void) { return modern_apic() ? 0xff : 0xf; } +#endif /** * lapic_get_maxlvt - get the maximum number of local vector table entries @@ -247,11 +333,7 @@ int lapic_get_maxlvt(void) */ /* Clock divisor */ -#ifdef CONFG_X86_64 -#define APIC_DIVISOR 1 -#else #define APIC_DIVISOR 16 -#endif /* * This function sets up the local APIC timer, with a timeout of @@ -360,6 +442,7 @@ static void lapic_timer_setup(enum clock_event_mode mode, v = apic_read(APIC_LVTT); v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); apic_write(APIC_LVTT, v); + apic_write(APIC_TMICT, 0xffffffff); break; case CLOCK_EVT_MODE_RESUME: /* Nothing to do here */ @@ -383,7 +466,7 @@ static void lapic_timer_broadcast(cpumask_t mask) * Setup the local APIC timer for this CPU. Copy the initilized values * of the boot CPU and register the clock event in the framework. */ -static void __devinit setup_APIC_timer(void) +static void __cpuinit setup_APIC_timer(void) { struct clock_event_device *levt = &__get_cpu_var(lapic_events); @@ -453,14 +536,51 @@ static void __init lapic_cal_handler(struct clock_event_device *dev) } } +static int __init calibrate_by_pmtimer(long deltapm, long *delta) +{ + const long pm_100ms = PMTMR_TICKS_PER_SEC / 10; + const long pm_thresh = pm_100ms / 100; + unsigned long mult; + u64 res; + +#ifndef CONFIG_X86_PM_TIMER + return -1; +#endif + + apic_printk(APIC_VERBOSE, "... PM timer delta = %ld\n", deltapm); + + /* Check, if the PM timer is available */ + if (!deltapm) + return -1; + + mult = clocksource_hz2mult(PMTMR_TICKS_PER_SEC, 22); + + if (deltapm > (pm_100ms - pm_thresh) && + deltapm < (pm_100ms + pm_thresh)) { + apic_printk(APIC_VERBOSE, "... PM timer result ok\n"); + } else { + res = (((u64)deltapm) * mult) >> 22; + do_div(res, 1000000); + pr_warning("APIC calibration not consistent " + "with PM Timer: %ldms instead of 100ms\n", + (long)res); + /* Correct the lapic counter value */ + res = (((u64)(*delta)) * pm_100ms); + do_div(res, deltapm); + pr_info("APIC delta adjusted to PM-Timer: " + "%lu (%ld)\n", (unsigned long)res, *delta); + *delta = (long)res; + } + + return 0; +} + static int __init calibrate_APIC_clock(void) { struct clock_event_device *levt = &__get_cpu_var(lapic_events); - const long pm_100ms = PMTMR_TICKS_PER_SEC/10; - const long pm_thresh = pm_100ms/100; void (*real_handler)(struct clock_event_device *dev); unsigned long deltaj; - long delta, deltapm; + long delta; int pm_referenced = 0; local_irq_disable(); @@ -470,10 +590,10 @@ static int __init calibrate_APIC_clock(void) global_clock_event->event_handler = lapic_cal_handler; /* - * Setup the APIC counter to 1e9. There is no way the lapic + * Setup the APIC counter to maximum. There is no way the lapic * can underflow in the 100ms detection time frame */ - __setup_APIC_LVTT(1000000000, 0, 0); + __setup_APIC_LVTT(0xffffffff, 0, 0); /* Let the interrupts run */ local_irq_enable(); @@ -490,34 +610,9 @@ static int __init calibrate_APIC_clock(void) delta = lapic_cal_t1 - lapic_cal_t2; apic_printk(APIC_VERBOSE, "... lapic delta = %ld\n", delta); - /* Check, if the PM timer is available */ - deltapm = lapic_cal_pm2 - lapic_cal_pm1; - apic_printk(APIC_VERBOSE, "... PM timer delta = %ld\n", deltapm); - - if (deltapm) { - unsigned long mult; - u64 res; - - mult = clocksource_hz2mult(PMTMR_TICKS_PER_SEC, 22); - - if (deltapm > (pm_100ms - pm_thresh) && - deltapm < (pm_100ms + pm_thresh)) { - apic_printk(APIC_VERBOSE, "... PM timer result ok\n"); - } else { - res = (((u64) deltapm) * mult) >> 22; - do_div(res, 1000000); - printk(KERN_WARNING "APIC calibration not consistent " - "with PM Timer: %ldms instead of 100ms\n", - (long)res); - /* Correct the lapic counter value */ - res = (((u64) delta) * pm_100ms); - do_div(res, deltapm); - printk(KERN_INFO "APIC delta adjusted to PM-Timer: " - "%lu (%ld)\n", (unsigned long) res, delta); - delta = (long) res; - } - pm_referenced = 1; - } + /* we trust the PM based calibration if possible */ + pm_referenced = !calibrate_by_pmtimer(lapic_cal_pm2 - lapic_cal_pm1, + &delta); /* Calculate the scaled math multiplication factor */ lapic_clockevent.mult = div_sc(delta, TICK_NSEC * LAPIC_CAL_LOOPS, @@ -552,14 +647,16 @@ static int __init calibrate_APIC_clock(void) */ if (calibration_result < (1000000 / HZ)) { local_irq_enable(); - printk(KERN_WARNING - "APIC frequency too slow, disabling apic timer\n"); + pr_warning("APIC frequency too slow, disabling apic timer\n"); return -1; } levt->features &= ~CLOCK_EVT_FEAT_DUMMY; - /* We trust the pm timer based calibration */ + /* + * PM timer calibration failed or not turned on + * so lets try APIC timer based calibration + */ if (!pm_referenced) { apic_printk(APIC_VERBOSE, "... verify APIC timer\n"); @@ -576,13 +673,9 @@ static int __init calibrate_APIC_clock(void) while (lapic_cal_loops <= LAPIC_CAL_LOOPS) cpu_relax(); - local_irq_disable(); - /* Stop the lapic timer */ lapic_timer_setup(CLOCK_EVT_MODE_SHUTDOWN, levt); - local_irq_enable(); - /* Jiffies delta */ deltaj = lapic_cal_j2 - lapic_cal_j1; apic_printk(APIC_VERBOSE, "... jiffies delta = %lu\n", deltaj); @@ -596,8 +689,7 @@ static int __init calibrate_APIC_clock(void) local_irq_enable(); if (levt->features & CLOCK_EVT_FEAT_DUMMY) { - printk(KERN_WARNING - "APIC timer disabled due to verification failure.\n"); + pr_warning("APIC timer disabled due to verification failure.\n"); return -1; } @@ -618,7 +710,7 @@ void __init setup_boot_APIC_clock(void) * broadcast mechanism is used. On UP systems simply ignore it. */ if (disable_apic_timer) { - printk(KERN_INFO "Disabling APIC timer\n"); + pr_info("Disabling APIC timer\n"); /* No broadcast on UP ! */ if (num_possible_cpus() > 1) { lapic_clockevent.mult = 1; @@ -645,14 +737,14 @@ void __init setup_boot_APIC_clock(void) if (nmi_watchdog != NMI_IO_APIC) lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY; else - printk(KERN_WARNING "APIC timer registered as dummy," + pr_warning("APIC timer registered as dummy," " due to nmi_watchdog=%d!\n", nmi_watchdog); /* Setup the lapic or request the broadcast */ setup_APIC_timer(); } -void __devinit setup_secondary_APIC_clock(void) +void __cpuinit setup_secondary_APIC_clock(void) { setup_APIC_timer(); } @@ -677,8 +769,7 @@ static void local_apic_timer_interrupt(void) * spurious. */ if (!evt->event_handler) { - printk(KERN_WARNING - "Spurious LAPIC timer interrupt on cpu %d\n", cpu); + pr_warning("Spurious LAPIC timer interrupt on cpu %d\n", cpu); /* Switch it off */ lapic_timer_setup(CLOCK_EVT_MODE_SHUTDOWN, evt); return; @@ -687,11 +778,7 @@ static void local_apic_timer_interrupt(void) /* * the NMI deadlock-detector uses this. */ -#ifdef CONFIG_X86_64 - add_pda(apic_timer_irqs, 1); -#else - per_cpu(irq_stat, cpu).apic_timer_irqs++; -#endif + inc_irq_stat(apic_timer_irqs); evt->event_handler(evt); } @@ -704,7 +791,7 @@ static void local_apic_timer_interrupt(void) * [ if a single-CPU system runs an SMP kernel then we call the local * interrupt as well. Thus we cannot inline the local irq ... ] */ -void smp_apic_timer_interrupt(struct pt_regs *regs) +void __irq_entry smp_apic_timer_interrupt(struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); @@ -718,6 +805,7 @@ void smp_apic_timer_interrupt(struct pt_regs *regs) * Besides, if we don't timer interrupts ignore the global * interrupt lock, which is the WrongThing (tm) to do. */ + exit_idle(); irq_enter(); local_apic_timer_interrupt(); irq_exit(); @@ -991,40 +1079,43 @@ void __init init_bsp_APIC(void) static void __cpuinit lapic_setup_esr(void) { - unsigned long oldvalue, value, maxlvt; - if (lapic_is_integrated() && !esr_disable) { - if (esr_disable) { - /* - * Something untraceable is creating bad interrupts on - * secondary quads ... for the moment, just leave the - * ESR disabled - we can't do anything useful with the - * errors anyway - mbligh - */ - printk(KERN_INFO "Leaving ESR disabled.\n"); - return; - } - /* !82489DX */ - maxlvt = lapic_get_maxlvt(); - if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ - apic_write(APIC_ESR, 0); - oldvalue = apic_read(APIC_ESR); + unsigned int oldvalue, value, maxlvt; - /* enables sending errors */ - value = ERROR_APIC_VECTOR; - apic_write(APIC_LVTERR, value); + if (!lapic_is_integrated()) { + pr_info("No ESR for 82489DX.\n"); + return; + } + + if (esr_disable) { /* - * spec says clear errors after enabling vector. + * Something untraceable is creating bad interrupts on + * secondary quads ... for the moment, just leave the + * ESR disabled - we can't do anything useful with the + * errors anyway - mbligh */ - if (maxlvt > 3) - apic_write(APIC_ESR, 0); - value = apic_read(APIC_ESR); - if (value != oldvalue) - apic_printk(APIC_VERBOSE, "ESR value before enabling " - "vector: 0x%08lx after: 0x%08lx\n", - oldvalue, value); - } else { - printk(KERN_INFO "No ESR for 82489DX.\n"); + pr_info("Leaving ESR disabled.\n"); + return; } + + maxlvt = lapic_get_maxlvt(); + if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ + apic_write(APIC_ESR, 0); + oldvalue = apic_read(APIC_ESR); + + /* enables sending errors */ + value = ERROR_APIC_VECTOR; + apic_write(APIC_LVTERR, value); + + /* + * spec says clear errors after enabling vector. + */ + if (maxlvt > 3) + apic_write(APIC_ESR, 0); + value = apic_read(APIC_ESR); + if (value != oldvalue) + apic_printk(APIC_VERBOSE, "ESR value before enabling " + "vector: 0x%08x after: 0x%08x\n", + oldvalue, value); } @@ -1033,24 +1124,27 @@ static void __cpuinit lapic_setup_esr(void) */ void __cpuinit setup_local_APIC(void) { - unsigned long value, integrated; + unsigned int value; int i, j; +#ifdef CONFIG_X86_32 /* Pound the ESR really hard over the head with a big hammer - mbligh */ - if (esr_disable) { + if (lapic_is_integrated() && esr_disable) { apic_write(APIC_ESR, 0); apic_write(APIC_ESR, 0); apic_write(APIC_ESR, 0); apic_write(APIC_ESR, 0); } +#endif - integrated = lapic_is_integrated(); + preempt_disable(); /* * Double-check whether this APIC is really registered. + * This is meaningless in clustered apic mode, so we skip it. */ if (!apic_id_registered()) - WARN_ON_ONCE(1); + BUG(); /* * Intel recommends to set DFR, LDR and TPR before enabling @@ -1096,6 +1190,7 @@ void __cpuinit setup_local_APIC(void) */ value |= APIC_SPIV_APIC_ENABLED; +#ifdef CONFIG_X86_32 /* * Some unknown Intel IO/APIC (or APIC) errata is biting us with * certain networking cards. If high frequency interrupts are @@ -1116,8 +1211,13 @@ void __cpuinit setup_local_APIC(void) * See also the comment in end_level_ioapic_irq(). --macro */ - /* Enable focus processor (bit==0) */ + /* + * - enable focus processor (bit==0) + * - 64bit mode always use processor focus + * so no need to set it + */ value &= ~APIC_SPIV_FOCUS_DISABLED; +#endif /* * Set spurious IRQ vector @@ -1154,9 +1254,11 @@ void __cpuinit setup_local_APIC(void) value = APIC_DM_NMI; else value = APIC_DM_NMI | APIC_LVT_MASKED; - if (!integrated) /* 82489DX */ + if (!lapic_is_integrated()) /* 82489DX */ value |= APIC_LVT_LEVEL_TRIGGER; apic_write(APIC_LVT1, value); + + preempt_enable(); } void __cpuinit end_local_APIC_setup(void) @@ -1177,6 +1279,146 @@ void __cpuinit end_local_APIC_setup(void) apic_pm_activate(); } +#ifdef HAVE_X2APIC +void check_x2apic(void) +{ + int msr, msr2; + + rdmsr(MSR_IA32_APICBASE, msr, msr2); + + if (msr & X2APIC_ENABLE) { + pr_info("x2apic enabled by BIOS, switching to x2apic ops\n"); + x2apic_preenabled = x2apic = 1; + apic_ops = &x2apic_ops; + } +} + +void enable_x2apic(void) +{ + int msr, msr2; + + rdmsr(MSR_IA32_APICBASE, msr, msr2); + if (!(msr & X2APIC_ENABLE)) { + pr_info("Enabling x2apic\n"); + wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0); + } +} + +void __init enable_IR_x2apic(void) +{ +#ifdef CONFIG_INTR_REMAP + int ret; + unsigned long flags; + + if (!cpu_has_x2apic) + return; + + if (!x2apic_preenabled && disable_x2apic) { + pr_info("Skipped enabling x2apic and Interrupt-remapping " + "because of nox2apic\n"); + return; + } + + if (x2apic_preenabled && disable_x2apic) + panic("Bios already enabled x2apic, can't enforce nox2apic"); + + if (!x2apic_preenabled && skip_ioapic_setup) { + pr_info("Skipped enabling x2apic and Interrupt-remapping " + "because of skipping io-apic setup\n"); + return; + } + + ret = dmar_table_init(); + if (ret) { + pr_info("dmar_table_init() failed with %d:\n", ret); + + if (x2apic_preenabled) + panic("x2apic enabled by bios. But IR enabling failed"); + else + pr_info("Not enabling x2apic,Intr-remapping\n"); + return; + } + + local_irq_save(flags); + mask_8259A(); + + ret = save_mask_IO_APIC_setup(); + if (ret) { + pr_info("Saving IO-APIC state failed: %d\n", ret); + goto end; + } + + ret = enable_intr_remapping(1); + + if (ret && x2apic_preenabled) { + local_irq_restore(flags); + panic("x2apic enabled by bios. But IR enabling failed"); + } + + if (ret) + goto end_restore; + + if (!x2apic) { + x2apic = 1; + apic_ops = &x2apic_ops; + enable_x2apic(); + } + +end_restore: + if (ret) + /* + * IR enabling failed + */ + restore_IO_APIC_setup(); + else + reinit_intr_remapped_IO_APIC(x2apic_preenabled); + +end: + unmask_8259A(); + local_irq_restore(flags); + + if (!ret) { + if (!x2apic_preenabled) + pr_info("Enabled x2apic and interrupt-remapping\n"); + else + pr_info("Enabled Interrupt-remapping\n"); + } else + pr_err("Failed to enable Interrupt-remapping and x2apic\n"); +#else + if (!cpu_has_x2apic) + return; + + if (x2apic_preenabled) + panic("x2apic enabled prior OS handover," + " enable CONFIG_INTR_REMAP"); + + pr_info("Enable CONFIG_INTR_REMAP for enabling intr-remapping " + " and x2apic\n"); +#endif + + return; +} +#endif /* HAVE_X2APIC */ + +#ifdef CONFIG_X86_64 +/* + * Detect and enable local APICs on non-SMP boards. + * Original code written by Keir Fraser. + * On AMD64 we trust the BIOS - if it says no APIC it is likely + * not correctly set up (usually the APIC timer won't work etc.) + */ +static int __init detect_init_APIC(void) +{ + if (!cpu_has_apic) { + pr_info("No local APIC present\n"); + return -1; + } + + mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; + boot_cpu_physical_apicid = 0; + return 0; +} +#else /* * Detect and initialize APIC */ @@ -1209,8 +1451,8 @@ static int __init detect_init_APIC(void) * "lapic" specified. */ if (!force_enable_local_apic) { - printk(KERN_INFO "Local APIC disabled by BIOS -- " - "you can enable it with \"lapic\"\n"); + pr_info("Local APIC disabled by BIOS -- " + "you can enable it with \"lapic\"\n"); return -1; } /* @@ -1220,8 +1462,7 @@ static int __init detect_init_APIC(void) */ rdmsr(MSR_IA32_APICBASE, l, h); if (!(l & MSR_IA32_APICBASE_ENABLE)) { - printk(KERN_INFO - "Local APIC disabled by BIOS -- reenabling.\n"); + pr_info("Local APIC disabled by BIOS -- reenabling.\n"); l &= ~MSR_IA32_APICBASE_BASE; l |= MSR_IA32_APICBASE_ENABLE | APIC_DEFAULT_PHYS_BASE; wrmsr(MSR_IA32_APICBASE, l, h); @@ -1234,7 +1475,7 @@ static int __init detect_init_APIC(void) */ features = cpuid_edx(1); if (!(features & (1 << X86_FEATURE_APIC))) { - printk(KERN_WARNING "Could not enable APIC!\n"); + pr_warning("Could not enable APIC!\n"); return -1; } set_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC); @@ -1245,22 +1486,56 @@ static int __init detect_init_APIC(void) if (l & MSR_IA32_APICBASE_ENABLE) mp_lapic_addr = l & MSR_IA32_APICBASE_BASE; - printk(KERN_INFO "Found and enabled local APIC!\n"); + pr_info("Found and enabled local APIC!\n"); apic_pm_activate(); return 0; no_apic: - printk(KERN_INFO "No local APIC present or hardware disabled\n"); + pr_info("No local APIC present or hardware disabled\n"); return -1; } +#endif + +#ifdef CONFIG_X86_64 +void __init early_init_lapic_mapping(void) +{ + unsigned long phys_addr; + + /* + * If no local APIC can be found then go out + * : it means there is no mpatable and MADT + */ + if (!smp_found_config) + return; + + phys_addr = mp_lapic_addr; + + set_fixmap_nocache(FIX_APIC_BASE, phys_addr); + apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", + APIC_BASE, phys_addr); + + /* + * Fetch the APIC ID of the BSP in case we have a + * default configuration (or the MP table is broken). + */ + boot_cpu_physical_apicid = read_apic_id(); +} +#endif /** * init_apic_mappings - initialize APIC mappings */ void __init init_apic_mappings(void) { +#ifdef HAVE_X2APIC + if (x2apic) { + boot_cpu_physical_apicid = read_apic_id(); + return; + } +#endif + /* * If no local APIC can be found then set up a fake all * zeroes page to simulate the local APIC and another @@ -1273,8 +1548,8 @@ void __init init_apic_mappings(void) apic_phys = mp_lapic_addr; set_fixmap_nocache(FIX_APIC_BASE, apic_phys); - printk(KERN_DEBUG "mapped APIC to %08lx (%08lx)\n", APIC_BASE, - apic_phys); + apic_printk(APIC_VERBOSE, "mapped APIC to %08lx (%08lx)\n", + APIC_BASE, apic_phys); /* * Fetch the APIC ID of the BSP in case we have a @@ -1282,18 +1557,27 @@ void __init init_apic_mappings(void) */ if (boot_cpu_physical_apicid == -1U) boot_cpu_physical_apicid = read_apic_id(); - } /* * This initializes the IO-APIC and APIC hardware if this is * a UP kernel. */ - int apic_version[MAX_APICS]; int __init APIC_init_uniprocessor(void) { +#ifdef CONFIG_X86_64 + if (disable_apic) { + pr_info("Apic disabled\n"); + return -1; + } + if (!cpu_has_apic) { + disable_apic = 1; + pr_info("Apic disabled by BIOS\n"); + return -1; + } +#else if (!smp_found_config && !cpu_has_apic) return -1; @@ -1302,39 +1586,68 @@ int __init APIC_init_uniprocessor(void) */ if (!cpu_has_apic && APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) { - printk(KERN_ERR "BIOS bug, local APIC #%d not detected!...\n", - boot_cpu_physical_apicid); + pr_err("BIOS bug, local APIC 0x%x not detected!...\n", + boot_cpu_physical_apicid); clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC); return -1; } +#endif - verify_local_APIC(); +#ifdef HAVE_X2APIC + enable_IR_x2apic(); +#endif +#ifdef CONFIG_X86_64 + setup_apic_routing(); +#endif + verify_local_APIC(); connect_bsp_APIC(); +#ifdef CONFIG_X86_64 + apic_write(APIC_ID, SET_APIC_ID(boot_cpu_physical_apicid)); +#else /* * Hack: In case of kdump, after a crash, kernel might be booting * on a cpu with non-zero lapic id. But boot_cpu_physical_apicid * might be zero if read from MP tables. Get it from LAPIC. */ -#ifdef CONFIG_CRASH_DUMP +# ifdef CONFIG_CRASH_DUMP boot_cpu_physical_apicid = read_apic_id(); +# endif #endif physid_set_mask_of_physid(boot_cpu_physical_apicid, &phys_cpu_present_map); - setup_local_APIC(); +#ifdef CONFIG_X86_64 + /* + * Now enable IO-APICs, actually call clear_IO_APIC + * We need clear_IO_APIC before enabling vector on BP + */ + if (!skip_ioapic_setup && nr_ioapics) + enable_IO_APIC(); +#endif + #ifdef CONFIG_X86_IO_APIC if (!smp_found_config || skip_ioapic_setup || !nr_ioapics) #endif localise_nmi_watchdog(); end_local_APIC_setup(); + #ifdef CONFIG_X86_IO_APIC - if (smp_found_config) - if (!skip_ioapic_setup && nr_ioapics) - setup_IO_APIC(); + if (smp_found_config && !skip_ioapic_setup && nr_ioapics) + setup_IO_APIC(); +# ifdef CONFIG_X86_64 + else + nr_ioapics = 0; +# endif #endif + +#ifdef CONFIG_X86_64 + setup_boot_APIC_clock(); + check_nmi_watchdog(); +#else setup_boot_clock(); +#endif return 0; } @@ -1348,8 +1661,9 @@ int __init APIC_init_uniprocessor(void) */ void smp_spurious_interrupt(struct pt_regs *regs) { - unsigned long v; + u32 v; + exit_idle(); irq_enter(); /* * Check if this really is a spurious interrupt and ACK it @@ -1360,10 +1674,11 @@ void smp_spurious_interrupt(struct pt_regs *regs) if (v & (1 << (SPURIOUS_APIC_VECTOR & 0x1f))) ack_APIC_irq(); + inc_irq_stat(irq_spurious_count); + /* see sw-dev-man vol 3, chapter 7.4.13.5 */ - printk(KERN_INFO "spurious APIC interrupt on CPU#%d, " - "should never happen.\n", smp_processor_id()); - __get_cpu_var(irq_stat).irq_spurious_count++; + pr_info("spurious APIC interrupt on CPU#%d, " + "should never happen.\n", smp_processor_id()); irq_exit(); } @@ -1372,8 +1687,9 @@ void smp_spurious_interrupt(struct pt_regs *regs) */ void smp_error_interrupt(struct pt_regs *regs) { - unsigned long v, v1; + u32 v, v1; + exit_idle(); irq_enter(); /* First tickle the hardware, only then report what went on. -- REW */ v = apic_read(APIC_ESR); @@ -1382,17 +1698,18 @@ void smp_error_interrupt(struct pt_regs *regs) ack_APIC_irq(); atomic_inc(&irq_err_count); - /* Here is what the APIC error bits mean: - 0: Send CS error - 1: Receive CS error - 2: Send accept error - 3: Receive accept error - 4: Reserved - 5: Send illegal vector - 6: Received illegal vector - 7: Illegal register address - */ - printk(KERN_DEBUG "APIC error on CPU%d: %02lx(%02lx)\n", + /* + * Here is what the APIC error bits mean: + * 0: Send CS error + * 1: Receive CS error + * 2: Send accept error + * 3: Receive accept error + * 4: Reserved + * 5: Send illegal vector + * 6: Received illegal vector + * 7: Illegal register address + */ + pr_debug("APIC error on CPU%d: %02x(%02x)\n", smp_processor_id(), v , v1); irq_exit(); } @@ -1496,15 +1813,15 @@ void __cpuinit generic_processor_info(int apicid, int version) * Validate version */ if (version == 0x0) { - printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! " - "fixing up to 0x10. (tell your hw vendor)\n", - version); + pr_warning("BIOS bug, APIC version is 0 for CPU#%d! " + "fixing up to 0x10. (tell your hw vendor)\n", + version); version = 0x10; } apic_version[apicid] = version; if (num_processors >= NR_CPUS) { - printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached." + pr_warning("WARNING: NR_CPUS limit of %i reached." " Processor ignored.\n", NR_CPUS); return; } @@ -1565,6 +1882,13 @@ void __cpuinit generic_processor_info(int apicid, int version) cpu_set(cpu, cpu_present_map); } +#ifdef CONFIG_X86_64 +int hard_smp_processor_id(void) +{ + return read_apic_id(); +} +#endif + /* * Power management */ @@ -1640,7 +1964,7 @@ static int lapic_resume(struct sys_device *dev) local_irq_save(flags); -#ifdef CONFIG_X86_64 +#ifdef HAVE_X2APIC if (x2apic) enable_x2apic(); else @@ -1702,7 +2026,7 @@ static struct sys_device device_lapic = { .cls = &lapic_sysclass, }; -static void __devinit apic_pm_activate(void) +static void __cpuinit apic_pm_activate(void) { apic_pm_state.active = 1; } @@ -1728,16 +2052,87 @@ static void apic_pm_activate(void) { } #endif /* CONFIG_PM */ +#ifdef CONFIG_X86_64 /* - * APIC command line parameters + * apic_is_clustered_box() -- Check if we can expect good TSC + * + * Thus far, the major user of this is IBM's Summit2 series: + * + * Clustered boxes may have unsynced TSC problems if they are + * multi-chassis. Use available data to take a good guess. + * If in doubt, go HPET. */ -static int __init parse_lapic(char *arg) +__cpuinit int apic_is_clustered_box(void) { - force_enable_local_apic = 1; - return 0; + int i, clusters, zeros; + unsigned id; + u16 *bios_cpu_apicid; + DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS); + + /* + * there is not this kind of box with AMD CPU yet. + * Some AMD box with quadcore cpu and 8 sockets apicid + * will be [4, 0x23] or [8, 0x27] could be thought to + * vsmp box still need checking... + */ + if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && !is_vsmp_box()) + return 0; + + bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); + bitmap_zero(clustermap, NUM_APIC_CLUSTERS); + + for (i = 0; i < NR_CPUS; i++) { + /* are we being called early in kernel startup? */ + if (bios_cpu_apicid) { + id = bios_cpu_apicid[i]; + } + else if (i < nr_cpu_ids) { + if (cpu_present(i)) + id = per_cpu(x86_bios_cpu_apicid, i); + else + continue; + } + else + break; + + if (id != BAD_APICID) + __set_bit(APIC_CLUSTERID(id), clustermap); + } + + /* Problem: Partially populated chassis may not have CPUs in some of + * the APIC clusters they have been allocated. Only present CPUs have + * x86_bios_cpu_apicid entries, thus causing zeroes in the bitmap. + * Since clusters are allocated sequentially, count zeros only if + * they are bounded by ones. + */ + clusters = 0; + zeros = 0; + for (i = 0; i < NUM_APIC_CLUSTERS; i++) { + if (test_bit(i, clustermap)) { + clusters += 1 + zeros; + zeros = 0; + } else + ++zeros; + } + + /* ScaleMP vSMPowered boxes have one cluster per board and TSCs are + * not guaranteed to be synced between boards + */ + if (is_vsmp_box() && clusters > 1) + return 1; + + /* + * If clusters > 2, then should be multi-chassis. + * May have to revisit this when multi-core + hyperthreaded CPUs come + * out, but AFAIK this will work even for them. + */ + return (clusters > 2); } -early_param("lapic", parse_lapic); +#endif +/* + * APIC command line parameters + */ static int __init setup_disableapic(char *arg) { disable_apic = 1; @@ -1779,7 +2174,6 @@ static int __init apic_set_verbosity(char *arg) if (!arg) { #ifdef CONFIG_X86_64 skip_ioapic_setup = 0; - ioapic_force = 1; return 0; #endif return -EINVAL; @@ -1790,7 +2184,7 @@ static int __init apic_set_verbosity(char *arg) else if (strcmp("verbose", arg) == 0) apic_verbosity = APIC_VERBOSE; else { - printk(KERN_WARNING "APIC Verbosity level %s not recognised" + pr_warning("APIC Verbosity level %s not recognised" " use apic=verbose or apic=debug\n", arg); return -EINVAL; } diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c deleted file mode 100644 index 94ddb69ae15e..000000000000 --- a/arch/x86/kernel/apic_64.c +++ /dev/null @@ -1,1848 +0,0 @@ -/* - * Local APIC handling, local APIC timers - * - * (c) 1999, 2000 Ingo Molnar <mingo@redhat.com> - * - * Fixes - * Maciej W. Rozycki : Bits for genuine 82489DX APICs; - * thanks to Eric Gilmore - * and Rolf G. Tews - * for testing these extensively. - * Maciej W. Rozycki : Various updates and fixes. - * Mikael Pettersson : Power Management for UP-APIC. - * Pavel Machek and - * Mikael Pettersson : PM converted to driver model. - */ - -#include <linux/init.h> - -#include <linux/mm.h> -#include <linux/delay.h> -#include <linux/bootmem.h> -#include <linux/interrupt.h> -#include <linux/mc146818rtc.h> -#include <linux/kernel_stat.h> -#include <linux/sysdev.h> -#include <linux/ioport.h> -#include <linux/clockchips.h> -#include <linux/acpi_pmtmr.h> -#include <linux/module.h> -#include <linux/dmar.h> - -#include <asm/atomic.h> -#include <asm/smp.h> -#include <asm/mtrr.h> -#include <asm/mpspec.h> -#include <asm/hpet.h> -#include <asm/pgalloc.h> -#include <asm/nmi.h> -#include <asm/idle.h> -#include <asm/proto.h> -#include <asm/timex.h> -#include <asm/apic.h> -#include <asm/i8259.h> - -#include <mach_ipi.h> -#include <mach_apic.h> - -/* Disable local APIC timer from the kernel commandline or via dmi quirk */ -static int disable_apic_timer __cpuinitdata; -static int apic_calibrate_pmtmr __initdata; -int disable_apic; -int disable_x2apic; -int x2apic; - -/* x2apic enabled before OS handover */ -int x2apic_preenabled; - -/* Local APIC timer works in C2 */ -int local_apic_timer_c2_ok; -EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok); - -/* - * Debug level, exported for io_apic.c - */ -unsigned int apic_verbosity; - -/* Have we found an MP table */ -int smp_found_config; - -static struct resource lapic_resource = { - .name = "Local APIC", - .flags = IORESOURCE_MEM | IORESOURCE_BUSY, -}; - -static unsigned int calibration_result; - -static int lapic_next_event(unsigned long delta, - struct clock_event_device *evt); -static void lapic_timer_setup(enum clock_event_mode mode, - struct clock_event_device *evt); -static void lapic_timer_broadcast(cpumask_t mask); -static void apic_pm_activate(void); - -/* - * The local apic timer can be used for any function which is CPU local. - */ -static struct clock_event_device lapic_clockevent = { - .name = "lapic", - .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT - | CLOCK_EVT_FEAT_C3STOP | CLOCK_EVT_FEAT_DUMMY, - .shift = 32, - .set_mode = lapic_timer_setup, - .set_next_event = lapic_next_event, - .broadcast = lapic_timer_broadcast, - .rating = 100, - .irq = -1, -}; -static DEFINE_PER_CPU(struct clock_event_device, lapic_events); - -static unsigned long apic_phys; - -unsigned long mp_lapic_addr; - -/* - * Get the LAPIC version - */ -static inline int lapic_get_version(void) -{ - return GET_APIC_VERSION(apic_read(APIC_LVR)); -} - -/* - * Check, if the APIC is integrated or a separate chip - */ -static inline int lapic_is_integrated(void) -{ -#ifdef CONFIG_X86_64 - return 1; -#else - return APIC_INTEGRATED(lapic_get_version()); -#endif -} - -/* - * Check, whether this is a modern or a first generation APIC - */ -static int modern_apic(void) -{ - /* AMD systems use old APIC versions, so check the CPU */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && - boot_cpu_data.x86 >= 0xf) - return 1; - return lapic_get_version() >= 0x14; -} - -/* - * Paravirt kernels also might be using these below ops. So we still - * use generic apic_read()/apic_write(), which might be pointing to different - * ops in PARAVIRT case. - */ -void xapic_wait_icr_idle(void) -{ - while (apic_read(APIC_ICR) & APIC_ICR_BUSY) - cpu_relax(); -} - -u32 safe_xapic_wait_icr_idle(void) -{ - u32 send_status; - int timeout; - - timeout = 0; - do { - send_status = apic_read(APIC_ICR) & APIC_ICR_BUSY; - if (!send_status) - break; - udelay(100); - } while (timeout++ < 1000); - - return send_status; -} - -void xapic_icr_write(u32 low, u32 id) -{ - apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(id)); - apic_write(APIC_ICR, low); -} - -u64 xapic_icr_read(void) -{ - u32 icr1, icr2; - - icr2 = apic_read(APIC_ICR2); - icr1 = apic_read(APIC_ICR); - - return icr1 | ((u64)icr2 << 32); -} - -static struct apic_ops xapic_ops = { - .read = native_apic_mem_read, - .write = native_apic_mem_write, - .icr_read = xapic_icr_read, - .icr_write = xapic_icr_write, - .wait_icr_idle = xapic_wait_icr_idle, - .safe_wait_icr_idle = safe_xapic_wait_icr_idle, -}; - -struct apic_ops __read_mostly *apic_ops = &xapic_ops; -EXPORT_SYMBOL_GPL(apic_ops); - -static void x2apic_wait_icr_idle(void) -{ - /* no need to wait for icr idle in x2apic */ - return; -} - -static u32 safe_x2apic_wait_icr_idle(void) -{ - /* no need to wait for icr idle in x2apic */ - return 0; -} - -void x2apic_icr_write(u32 low, u32 id) -{ - wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low); -} - -u64 x2apic_icr_read(void) -{ - unsigned long val; - - rdmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), val); - return val; -} - -static struct apic_ops x2apic_ops = { - .read = native_apic_msr_read, - .write = native_apic_msr_write, - .icr_read = x2apic_icr_read, - .icr_write = x2apic_icr_write, - .wait_icr_idle = x2apic_wait_icr_idle, - .safe_wait_icr_idle = safe_x2apic_wait_icr_idle, -}; - -/** - * enable_NMI_through_LVT0 - enable NMI through local vector table 0 - */ -void __cpuinit enable_NMI_through_LVT0(void) -{ - unsigned int v; - - /* unmask and set to NMI */ - v = APIC_DM_NMI; - - /* Level triggered for 82489DX (32bit mode) */ - if (!lapic_is_integrated()) - v |= APIC_LVT_LEVEL_TRIGGER; - - apic_write(APIC_LVT0, v); -} - -/** - * lapic_get_maxlvt - get the maximum number of local vector table entries - */ -int lapic_get_maxlvt(void) -{ - unsigned int v; - - v = apic_read(APIC_LVR); - /* - * - we always have APIC integrated on 64bit mode - * - 82489DXs do not report # of LVT entries - */ - return APIC_INTEGRATED(GET_APIC_VERSION(v)) ? GET_APIC_MAXLVT(v) : 2; -} - -/* - * Local APIC timer - */ - -/* Clock divisor */ -#ifdef CONFG_X86_64 -#define APIC_DIVISOR 1 -#else -#define APIC_DIVISOR 16 -#endif - -/* - * This function sets up the local APIC timer, with a timeout of - * 'clocks' APIC bus clock. During calibration we actually call - * this function twice on the boot CPU, once with a bogus timeout - * value, second time for real. The other (noncalibrating) CPUs - * call this function only once, with the real, calibrated value. - * - * We do reads before writes even if unnecessary, to get around the - * P5 APIC double write bug. - */ -static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen) -{ - unsigned int lvtt_value, tmp_value; - - lvtt_value = LOCAL_TIMER_VECTOR; - if (!oneshot) - lvtt_value |= APIC_LVT_TIMER_PERIODIC; - if (!lapic_is_integrated()) - lvtt_value |= SET_APIC_TIMER_BASE(APIC_TIMER_BASE_DIV); - - if (!irqen) - lvtt_value |= APIC_LVT_MASKED; - - apic_write(APIC_LVTT, lvtt_value); - - /* - * Divide PICLK by 16 - */ - tmp_value = apic_read(APIC_TDCR); - apic_write(APIC_TDCR, - (tmp_value & ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE)) | - APIC_TDR_DIV_16); - - if (!oneshot) - apic_write(APIC_TMICT, clocks / APIC_DIVISOR); -} - -/* - * Setup extended LVT, AMD specific (K8, family 10h) - * - * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and - * MCE interrupts are supported. Thus MCE offset must be set to 0. - * - * If mask=1, the LVT entry does not generate interrupts while mask=0 - * enables the vector. See also the BKDGs. - */ - -#define APIC_EILVT_LVTOFF_MCE 0 -#define APIC_EILVT_LVTOFF_IBS 1 - -static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask) -{ - unsigned long reg = (lvt_off << 4) + APIC_EILVT0; - unsigned int v = (mask << 16) | (msg_type << 8) | vector; - - apic_write(reg, v); -} - -u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask) -{ - setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask); - return APIC_EILVT_LVTOFF_MCE; -} - -u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask) -{ - setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask); - return APIC_EILVT_LVTOFF_IBS; -} -EXPORT_SYMBOL_GPL(setup_APIC_eilvt_ibs); - -/* - * Program the next event, relative to now - */ -static int lapic_next_event(unsigned long delta, - struct clock_event_device *evt) -{ - apic_write(APIC_TMICT, delta); - return 0; -} - -/* - * Setup the lapic timer in periodic or oneshot mode - */ -static void lapic_timer_setup(enum clock_event_mode mode, - struct clock_event_device *evt) -{ - unsigned long flags; - unsigned int v; - - /* Lapic used as dummy for broadcast ? */ - if (evt->features & CLOCK_EVT_FEAT_DUMMY) - return; - - local_irq_save(flags); - - switch (mode) { - case CLOCK_EVT_MODE_PERIODIC: - case CLOCK_EVT_MODE_ONESHOT: - __setup_APIC_LVTT(calibration_result, - mode != CLOCK_EVT_MODE_PERIODIC, 1); - break; - case CLOCK_EVT_MODE_UNUSED: - case CLOCK_EVT_MODE_SHUTDOWN: - v = apic_read(APIC_LVTT); - v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); - apic_write(APIC_LVTT, v); - break; - case CLOCK_EVT_MODE_RESUME: - /* Nothing to do here */ - break; - } - - local_irq_restore(flags); -} - -/* - * Local APIC timer broadcast function - */ -static void lapic_timer_broadcast(cpumask_t mask) -{ -#ifdef CONFIG_SMP - send_IPI_mask(mask, LOCAL_TIMER_VECTOR); -#endif -} - -/* - * Setup the local APIC timer for this CPU. Copy the initilized values - * of the boot CPU and register the clock event in the framework. - */ -static void setup_APIC_timer(void) -{ - struct clock_event_device *levt = &__get_cpu_var(lapic_events); - - memcpy(levt, &lapic_clockevent, sizeof(*levt)); - levt->cpumask = cpumask_of_cpu(smp_processor_id()); - - clockevents_register_device(levt); -} - -/* - * In this function we calibrate APIC bus clocks to the external - * timer. Unfortunately we cannot use jiffies and the timer irq - * to calibrate, since some later bootup code depends on getting - * the first irq? Ugh. - * - * We want to do the calibration only once since we - * want to have local timer irqs syncron. CPUs connected - * by the same APIC bus have the very same bus frequency. - * And we want to have irqs off anyways, no accidental - * APIC irq that way. - */ - -#define TICK_COUNT 100000000 - -static int __init calibrate_APIC_clock(void) -{ - unsigned apic, apic_start; - unsigned long tsc, tsc_start; - int result; - - local_irq_disable(); - - /* - * Put whatever arbitrary (but long enough) timeout - * value into the APIC clock, we just want to get the - * counter running for calibration. - * - * No interrupt enable ! - */ - __setup_APIC_LVTT(250000000, 0, 0); - - apic_start = apic_read(APIC_TMCCT); -#ifdef CONFIG_X86_PM_TIMER - if (apic_calibrate_pmtmr && pmtmr_ioport) { - pmtimer_wait(5000); /* 5ms wait */ - apic = apic_read(APIC_TMCCT); - result = (apic_start - apic) * 1000L / 5; - } else -#endif - { - rdtscll(tsc_start); - - do { - apic = apic_read(APIC_TMCCT); - rdtscll(tsc); - } while ((tsc - tsc_start) < TICK_COUNT && - (apic_start - apic) < TICK_COUNT); - - result = (apic_start - apic) * 1000L * tsc_khz / - (tsc - tsc_start); - } - - local_irq_enable(); - - printk(KERN_DEBUG "APIC timer calibration result %d\n", result); - - printk(KERN_INFO "Detected %d.%03d MHz APIC timer.\n", - result / 1000 / 1000, result / 1000 % 1000); - - /* Calculate the scaled math multiplication factor */ - lapic_clockevent.mult = div_sc(result, NSEC_PER_SEC, - lapic_clockevent.shift); - lapic_clockevent.max_delta_ns = - clockevent_delta2ns(0x7FFFFF, &lapic_clockevent); - lapic_clockevent.min_delta_ns = - clockevent_delta2ns(0xF, &lapic_clockevent); - - calibration_result = (result * APIC_DIVISOR) / HZ; - - /* - * Do a sanity check on the APIC calibration result - */ - if (calibration_result < (1000000 / HZ)) { - printk(KERN_WARNING - "APIC frequency too slow, disabling apic timer\n"); - return -1; - } - - return 0; -} - -/* - * Setup the boot APIC - * - * Calibrate and verify the result. - */ -void __init setup_boot_APIC_clock(void) -{ - /* - * The local apic timer can be disabled via the kernel - * commandline or from the CPU detection code. Register the lapic - * timer as a dummy clock event source on SMP systems, so the - * broadcast mechanism is used. On UP systems simply ignore it. - */ - if (disable_apic_timer) { - printk(KERN_INFO "Disabling APIC timer\n"); - /* No broadcast on UP ! */ - if (num_possible_cpus() > 1) { - lapic_clockevent.mult = 1; - setup_APIC_timer(); - } - return; - } - - apic_printk(APIC_VERBOSE, "Using local APIC timer interrupts.\n" - "calibrating APIC timer ...\n"); - - if (calibrate_APIC_clock()) { - /* No broadcast on UP ! */ - if (num_possible_cpus() > 1) - setup_APIC_timer(); - return; - } - - /* - * If nmi_watchdog is set to IO_APIC, we need the - * PIT/HPET going. Otherwise register lapic as a dummy - * device. - */ - if (nmi_watchdog != NMI_IO_APIC) - lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY; - else - printk(KERN_WARNING "APIC timer registered as dummy," - " due to nmi_watchdog=%d!\n", nmi_watchdog); - - /* Setup the lapic or request the broadcast */ - setup_APIC_timer(); -} - -void __cpuinit setup_secondary_APIC_clock(void) -{ - setup_APIC_timer(); -} - -/* - * The guts of the apic timer interrupt - */ -static void local_apic_timer_interrupt(void) -{ - int cpu = smp_processor_id(); - struct clock_event_device *evt = &per_cpu(lapic_events, cpu); - - /* - * Normally we should not be here till LAPIC has been initialized but - * in some cases like kdump, its possible that there is a pending LAPIC - * timer interrupt from previous kernel's context and is delivered in - * new kernel the moment interrupts are enabled. - * - * Interrupts are enabled early and LAPIC is setup much later, hence - * its possible that when we get here evt->event_handler is NULL. - * Check for event_handler being NULL and discard the interrupt as - * spurious. - */ - if (!evt->event_handler) { - printk(KERN_WARNING - "Spurious LAPIC timer interrupt on cpu %d\n", cpu); - /* Switch it off */ - lapic_timer_setup(CLOCK_EVT_MODE_SHUTDOWN, evt); - return; - } - - /* - * the NMI deadlock-detector uses this. - */ -#ifdef CONFIG_X86_64 - add_pda(apic_timer_irqs, 1); -#else - per_cpu(irq_stat, cpu).apic_timer_irqs++; -#endif - - evt->event_handler(evt); -} - -/* - * Local APIC timer interrupt. This is the most natural way for doing - * local interrupts, but local timer interrupts can be emulated by - * broadcast interrupts too. [in case the hw doesn't support APIC timers] - * - * [ if a single-CPU system runs an SMP kernel then we call the local - * interrupt as well. Thus we cannot inline the local irq ... ] - */ -void smp_apic_timer_interrupt(struct pt_regs *regs) -{ - struct pt_regs *old_regs = set_irq_regs(regs); - - /* - * NOTE! We'd better ACK the irq immediately, - * because timer handling can be slow. - */ - ack_APIC_irq(); - /* - * update_process_times() expects us to have done irq_enter(). - * Besides, if we don't timer interrupts ignore the global - * interrupt lock, which is the WrongThing (tm) to do. - */ - exit_idle(); - irq_enter(); - local_apic_timer_interrupt(); - irq_exit(); - - set_irq_regs(old_regs); -} - -int setup_profiling_timer(unsigned int multiplier) -{ - return -EINVAL; -} - - -/* - * Local APIC start and shutdown - */ - -/** - * clear_local_APIC - shutdown the local APIC - * - * This is called, when a CPU is disabled and before rebooting, so the state of - * the local APIC has no dangling leftovers. Also used to cleanout any BIOS - * leftovers during boot. - */ -void clear_local_APIC(void) -{ - int maxlvt; - u32 v; - - /* APIC hasn't been mapped yet */ - if (!apic_phys) - return; - - maxlvt = lapic_get_maxlvt(); - /* - * Masking an LVT entry can trigger a local APIC error - * if the vector is zero. Mask LVTERR first to prevent this. - */ - if (maxlvt >= 3) { - v = ERROR_APIC_VECTOR; /* any non-zero vector will do */ - apic_write(APIC_LVTERR, v | APIC_LVT_MASKED); - } - /* - * Careful: we have to set masks only first to deassert - * any level-triggered sources. - */ - v = apic_read(APIC_LVTT); - apic_write(APIC_LVTT, v | APIC_LVT_MASKED); - v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v | APIC_LVT_MASKED); - v = apic_read(APIC_LVT1); - apic_write(APIC_LVT1, v | APIC_LVT_MASKED); - if (maxlvt >= 4) { - v = apic_read(APIC_LVTPC); - apic_write(APIC_LVTPC, v | APIC_LVT_MASKED); - } - - /* lets not touch this if we didn't frob it */ -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(X86_MCE_INTEL) - if (maxlvt >= 5) { - v = apic_read(APIC_LVTTHMR); - apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED); - } -#endif - /* - * Clean APIC state for other OSs: - */ - apic_write(APIC_LVTT, APIC_LVT_MASKED); - apic_write(APIC_LVT0, APIC_LVT_MASKED); - apic_write(APIC_LVT1, APIC_LVT_MASKED); - if (maxlvt >= 3) - apic_write(APIC_LVTERR, APIC_LVT_MASKED); - if (maxlvt >= 4) - apic_write(APIC_LVTPC, APIC_LVT_MASKED); - - /* Integrated APIC (!82489DX) ? */ - if (lapic_is_integrated()) { - if (maxlvt > 3) - /* Clear ESR due to Pentium errata 3AP and 11AP */ - apic_write(APIC_ESR, 0); - apic_read(APIC_ESR); - } -} - -/** - * disable_local_APIC - clear and disable the local APIC - */ -void disable_local_APIC(void) -{ - unsigned int value; - - clear_local_APIC(); - - /* - * Disable APIC (implies clearing of registers - * for 82489DX!). - */ - value = apic_read(APIC_SPIV); - value &= ~APIC_SPIV_APIC_ENABLED; - apic_write(APIC_SPIV, value); - -#ifdef CONFIG_X86_32 - /* - * When LAPIC was disabled by the BIOS and enabled by the kernel, - * restore the disabled state. - */ - if (enabled_via_apicbase) { - unsigned int l, h; - - rdmsr(MSR_IA32_APICBASE, l, h); - l &= ~MSR_IA32_APICBASE_ENABLE; - wrmsr(MSR_IA32_APICBASE, l, h); - } -#endif -} - -/* - * If Linux enabled the LAPIC against the BIOS default disable it down before - * re-entering the BIOS on shutdown. Otherwise the BIOS may get confused and - * not power-off. Additionally clear all LVT entries before disable_local_APIC - * for the case where Linux didn't enable the LAPIC. - */ -void lapic_shutdown(void) -{ - unsigned long flags; - - if (!cpu_has_apic) - return; - - local_irq_save(flags); - -#ifdef CONFIG_X86_32 - if (!enabled_via_apicbase) - clear_local_APIC(); - else -#endif - disable_local_APIC(); - - - local_irq_restore(flags); -} - -/* - * This is to verify that we're looking at a real local APIC. - * Check these against your board if the CPUs aren't getting - * started for no apparent reason. - */ -int __init verify_local_APIC(void) -{ - unsigned int reg0, reg1; - - /* - * The version register is read-only in a real APIC. - */ - reg0 = apic_read(APIC_LVR); - apic_printk(APIC_DEBUG, "Getting VERSION: %x\n", reg0); - apic_write(APIC_LVR, reg0 ^ APIC_LVR_MASK); - reg1 = apic_read(APIC_LVR); - apic_printk(APIC_DEBUG, "Getting VERSION: %x\n", reg1); - - /* - * The two version reads above should print the same - * numbers. If the second one is different, then we - * poke at a non-APIC. - */ - if (reg1 != reg0) - return 0; - - /* - * Check if the version looks reasonably. - */ - reg1 = GET_APIC_VERSION(reg0); - if (reg1 == 0x00 || reg1 == 0xff) - return 0; - reg1 = lapic_get_maxlvt(); - if (reg1 < 0x02 || reg1 == 0xff) - return 0; - - /* - * The ID register is read/write in a real APIC. - */ - reg0 = apic_read(APIC_ID); - apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg0); - apic_write(APIC_ID, reg0 ^ APIC_ID_MASK); - reg1 = apic_read(APIC_ID); - apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg1); - apic_write(APIC_ID, reg0); - if (reg1 != (reg0 ^ APIC_ID_MASK)) - return 0; - - /* - * The next two are just to see if we have sane values. - * They're only really relevant if we're in Virtual Wire - * compatibility mode, but most boxes are anymore. - */ - reg0 = apic_read(APIC_LVT0); - apic_printk(APIC_DEBUG, "Getting LVT0: %x\n", reg0); - reg1 = apic_read(APIC_LVT1); - apic_printk(APIC_DEBUG, "Getting LVT1: %x\n", reg1); - - return 1; -} - -/** - * sync_Arb_IDs - synchronize APIC bus arbitration IDs - */ -void __init sync_Arb_IDs(void) -{ - /* - * Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 And not - * needed on AMD. - */ - if (modern_apic() || boot_cpu_data.x86_vendor == X86_VENDOR_AMD) - return; - - /* - * Wait for idle. - */ - apic_wait_icr_idle(); - - apic_printk(APIC_DEBUG, "Synchronizing Arb IDs.\n"); - apic_write(APIC_ICR, APIC_DEST_ALLINC | - APIC_INT_LEVELTRIG | APIC_DM_INIT); -} - -/* - * An initial setup of the virtual wire mode. - */ -void __init init_bsp_APIC(void) -{ - unsigned int value; - - /* - * Don't do the setup now if we have a SMP BIOS as the - * through-I/O-APIC virtual wire mode might be active. - */ - if (smp_found_config || !cpu_has_apic) - return; - - /* - * Do not trust the local APIC being empty at bootup. - */ - clear_local_APIC(); - - /* - * Enable APIC. - */ - value = apic_read(APIC_SPIV); - value &= ~APIC_VECTOR_MASK; - value |= APIC_SPIV_APIC_ENABLED; - -#ifdef CONFIG_X86_32 - /* This bit is reserved on P4/Xeon and should be cleared */ - if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && - (boot_cpu_data.x86 == 15)) - value &= ~APIC_SPIV_FOCUS_DISABLED; - else -#endif - value |= APIC_SPIV_FOCUS_DISABLED; - value |= SPURIOUS_APIC_VECTOR; - apic_write(APIC_SPIV, value); - - /* - * Set up the virtual wire mode. - */ - apic_write(APIC_LVT0, APIC_DM_EXTINT); - value = APIC_DM_NMI; - if (!lapic_is_integrated()) /* 82489DX */ - value |= APIC_LVT_LEVEL_TRIGGER; - apic_write(APIC_LVT1, value); -} - -static void __cpuinit lapic_setup_esr(void) -{ - unsigned long oldvalue, value, maxlvt; - if (lapic_is_integrated() && !esr_disable) { - if (esr_disable) { - /* - * Something untraceable is creating bad interrupts on - * secondary quads ... for the moment, just leave the - * ESR disabled - we can't do anything useful with the - * errors anyway - mbligh - */ - printk(KERN_INFO "Leaving ESR disabled.\n"); - return; - } - /* !82489DX */ - maxlvt = lapic_get_maxlvt(); - if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ - apic_write(APIC_ESR, 0); - oldvalue = apic_read(APIC_ESR); - - /* enables sending errors */ - value = ERROR_APIC_VECTOR; - apic_write(APIC_LVTERR, value); - /* - * spec says clear errors after enabling vector. - */ - if (maxlvt > 3) - apic_write(APIC_ESR, 0); - value = apic_read(APIC_ESR); - if (value != oldvalue) - apic_printk(APIC_VERBOSE, "ESR value before enabling " - "vector: 0x%08lx after: 0x%08lx\n", - oldvalue, value); - } else { - printk(KERN_INFO "No ESR for 82489DX.\n"); - } -} - - -/** - * setup_local_APIC - setup the local APIC - */ -void __cpuinit setup_local_APIC(void) -{ - unsigned int value; - int i, j; - - preempt_disable(); - value = apic_read(APIC_LVR); - - BUILD_BUG_ON((SPURIOUS_APIC_VECTOR & 0x0f) != 0x0f); - - /* - * Double-check whether this APIC is really registered. - * This is meaningless in clustered apic mode, so we skip it. - */ - if (!apic_id_registered()) - BUG(); - - /* - * Intel recommends to set DFR, LDR and TPR before enabling - * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel - * document number 292116). So here it goes... - */ - init_apic_ldr(); - - /* - * Set Task Priority to 'accept all'. We never change this - * later on. - */ - value = apic_read(APIC_TASKPRI); - value &= ~APIC_TPRI_MASK; - apic_write(APIC_TASKPRI, value); - - /* - * After a crash, we no longer service the interrupts and a pending - * interrupt from previous kernel might still have ISR bit set. - * - * Most probably by now CPU has serviced that pending interrupt and - * it might not have done the ack_APIC_irq() because it thought, - * interrupt came from i8259 as ExtInt. LAPIC did not get EOI so it - * does not clear the ISR bit and cpu thinks it has already serivced - * the interrupt. Hence a vector might get locked. It was noticed - * for timer irq (vector 0x31). Issue an extra EOI to clear ISR. - */ - for (i = APIC_ISR_NR - 1; i >= 0; i--) { - value = apic_read(APIC_ISR + i*0x10); - for (j = 31; j >= 0; j--) { - if (value & (1<<j)) - ack_APIC_irq(); - } - } - - /* - * Now that we are all set up, enable the APIC - */ - value = apic_read(APIC_SPIV); - value &= ~APIC_VECTOR_MASK; - /* - * Enable APIC - */ - value |= APIC_SPIV_APIC_ENABLED; - - /* We always use processor focus */ - - /* - * Set spurious IRQ vector - */ - value |= SPURIOUS_APIC_VECTOR; - apic_write(APIC_SPIV, value); - - /* - * Set up LVT0, LVT1: - * - * set up through-local-APIC on the BP's LINT0. This is not - * strictly necessary in pure symmetric-IO mode, but sometimes - * we delegate interrupts to the 8259A. - */ - /* - * TODO: set up through-local-APIC from through-I/O-APIC? --macro - */ - value = apic_read(APIC_LVT0) & APIC_LVT_MASKED; - if (!smp_processor_id() && !value) { - value = APIC_DM_EXTINT; - apic_printk(APIC_VERBOSE, "enabled ExtINT on CPU#%d\n", - smp_processor_id()); - } else { - value = APIC_DM_EXTINT | APIC_LVT_MASKED; - apic_printk(APIC_VERBOSE, "masked ExtINT on CPU#%d\n", - smp_processor_id()); - } - apic_write(APIC_LVT0, value); - - /* - * only the BP should see the LINT1 NMI signal, obviously. - */ - if (!smp_processor_id()) - value = APIC_DM_NMI; - else - value = APIC_DM_NMI | APIC_LVT_MASKED; - apic_write(APIC_LVT1, value); - preempt_enable(); -} - -void __cpuinit end_local_APIC_setup(void) -{ - lapic_setup_esr(); - -#ifdef CONFIG_X86_32 - { - unsigned int value; - /* Disable the local apic timer */ - value = apic_read(APIC_LVTT); - value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); - apic_write(APIC_LVTT, value); - } -#endif - - setup_apic_nmi_watchdog(NULL); - apic_pm_activate(); -} - -void check_x2apic(void) -{ - int msr, msr2; - - rdmsr(MSR_IA32_APICBASE, msr, msr2); - - if (msr & X2APIC_ENABLE) { - printk("x2apic enabled by BIOS, switching to x2apic ops\n"); - x2apic_preenabled = x2apic = 1; - apic_ops = &x2apic_ops; - } -} - -void enable_x2apic(void) -{ - int msr, msr2; - - rdmsr(MSR_IA32_APICBASE, msr, msr2); - if (!(msr & X2APIC_ENABLE)) { - printk("Enabling x2apic\n"); - wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0); - } -} - -void enable_IR_x2apic(void) -{ -#ifdef CONFIG_INTR_REMAP - int ret; - unsigned long flags; - - if (!cpu_has_x2apic) - return; - - if (!x2apic_preenabled && disable_x2apic) { - printk(KERN_INFO - "Skipped enabling x2apic and Interrupt-remapping " - "because of nox2apic\n"); - return; - } - - if (x2apic_preenabled && disable_x2apic) - panic("Bios already enabled x2apic, can't enforce nox2apic"); - - if (!x2apic_preenabled && skip_ioapic_setup) { - printk(KERN_INFO - "Skipped enabling x2apic and Interrupt-remapping " - "because of skipping io-apic setup\n"); - return; - } - - ret = dmar_table_init(); - if (ret) { - printk(KERN_INFO - "dmar_table_init() failed with %d:\n", ret); - - if (x2apic_preenabled) - panic("x2apic enabled by bios. But IR enabling failed"); - else - printk(KERN_INFO - "Not enabling x2apic,Intr-remapping\n"); - return; - } - - local_irq_save(flags); - mask_8259A(); - save_mask_IO_APIC_setup(); - - ret = enable_intr_remapping(1); - - if (ret && x2apic_preenabled) { - local_irq_restore(flags); - panic("x2apic enabled by bios. But IR enabling failed"); - } - - if (ret) - goto end; - - if (!x2apic) { - x2apic = 1; - apic_ops = &x2apic_ops; - enable_x2apic(); - } -end: - if (ret) - /* - * IR enabling failed - */ - restore_IO_APIC_setup(); - else - reinit_intr_remapped_IO_APIC(x2apic_preenabled); - - unmask_8259A(); - local_irq_restore(flags); - - if (!ret) { - if (!x2apic_preenabled) - printk(KERN_INFO - "Enabled x2apic and interrupt-remapping\n"); - else - printk(KERN_INFO - "Enabled Interrupt-remapping\n"); - } else - printk(KERN_ERR - "Failed to enable Interrupt-remapping and x2apic\n"); -#else - if (!cpu_has_x2apic) - return; - - if (x2apic_preenabled) - panic("x2apic enabled prior OS handover," - " enable CONFIG_INTR_REMAP"); - - printk(KERN_INFO "Enable CONFIG_INTR_REMAP for enabling intr-remapping " - " and x2apic\n"); -#endif - - return; -} - -/* - * Detect and enable local APICs on non-SMP boards. - * Original code written by Keir Fraser. - * On AMD64 we trust the BIOS - if it says no APIC it is likely - * not correctly set up (usually the APIC timer won't work etc.) - */ -static int __init detect_init_APIC(void) -{ - if (!cpu_has_apic) { - printk(KERN_INFO "No local APIC present\n"); - return -1; - } - - mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; - boot_cpu_physical_apicid = 0; - return 0; -} - -void __init early_init_lapic_mapping(void) -{ - unsigned long phys_addr; - - /* - * If no local APIC can be found then go out - * : it means there is no mpatable and MADT - */ - if (!smp_found_config) - return; - - phys_addr = mp_lapic_addr; - - set_fixmap_nocache(FIX_APIC_BASE, phys_addr); - apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", - APIC_BASE, phys_addr); - - /* - * Fetch the APIC ID of the BSP in case we have a - * default configuration (or the MP table is broken). - */ - boot_cpu_physical_apicid = read_apic_id(); -} - -/** - * init_apic_mappings - initialize APIC mappings - */ -void __init init_apic_mappings(void) -{ - if (x2apic) { - boot_cpu_physical_apicid = read_apic_id(); - return; - } - - /* - * If no local APIC can be found then set up a fake all - * zeroes page to simulate the local APIC and another - * one for the IO-APIC. - */ - if (!smp_found_config && detect_init_APIC()) { - apic_phys = (unsigned long) alloc_bootmem_pages(PAGE_SIZE); - apic_phys = __pa(apic_phys); - } else - apic_phys = mp_lapic_addr; - - set_fixmap_nocache(FIX_APIC_BASE, apic_phys); - apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", - APIC_BASE, apic_phys); - - /* - * Fetch the APIC ID of the BSP in case we have a - * default configuration (or the MP table is broken). - */ - boot_cpu_physical_apicid = read_apic_id(); -} - -/* - * This initializes the IO-APIC and APIC hardware if this is - * a UP kernel. - */ -int apic_version[MAX_APICS]; - -int __init APIC_init_uniprocessor(void) -{ - if (disable_apic) { - printk(KERN_INFO "Apic disabled\n"); - return -1; - } - if (!cpu_has_apic) { - disable_apic = 1; - printk(KERN_INFO "Apic disabled by BIOS\n"); - return -1; - } - - enable_IR_x2apic(); - setup_apic_routing(); - - verify_local_APIC(); - - connect_bsp_APIC(); - - physid_set_mask_of_physid(boot_cpu_physical_apicid, &phys_cpu_present_map); - apic_write(APIC_ID, SET_APIC_ID(boot_cpu_physical_apicid)); - - setup_local_APIC(); - - /* - * Now enable IO-APICs, actually call clear_IO_APIC - * We need clear_IO_APIC before enabling vector on BP - */ - if (!skip_ioapic_setup && nr_ioapics) - enable_IO_APIC(); - - if (!smp_found_config || skip_ioapic_setup || !nr_ioapics) - localise_nmi_watchdog(); - end_local_APIC_setup(); - - if (smp_found_config && !skip_ioapic_setup && nr_ioapics) - setup_IO_APIC(); - else - nr_ioapics = 0; - setup_boot_APIC_clock(); - check_nmi_watchdog(); - return 0; -} - -/* - * Local APIC interrupts - */ - -/* - * This interrupt should _never_ happen with our APIC/SMP architecture - */ -asmlinkage void smp_spurious_interrupt(void) -{ - unsigned int v; - exit_idle(); - irq_enter(); - /* - * Check if this really is a spurious interrupt and ACK it - * if it is a vectored one. Just in case... - * Spurious interrupts should not be ACKed. - */ - v = apic_read(APIC_ISR + ((SPURIOUS_APIC_VECTOR & ~0x1f) >> 1)); - if (v & (1 << (SPURIOUS_APIC_VECTOR & 0x1f))) - ack_APIC_irq(); - - add_pda(irq_spurious_count, 1); - irq_exit(); -} - -/* - * This interrupt should never happen with our APIC/SMP architecture - */ -asmlinkage void smp_error_interrupt(void) -{ - unsigned int v, v1; - - exit_idle(); - irq_enter(); - /* First tickle the hardware, only then report what went on. -- REW */ - v = apic_read(APIC_ESR); - apic_write(APIC_ESR, 0); - v1 = apic_read(APIC_ESR); - ack_APIC_irq(); - atomic_inc(&irq_err_count); - - /* Here is what the APIC error bits mean: - 0: Send CS error - 1: Receive CS error - 2: Send accept error - 3: Receive accept error - 4: Reserved - 5: Send illegal vector - 6: Received illegal vector - 7: Illegal register address - */ - printk(KERN_DEBUG "APIC error on CPU%d: %02x(%02x)\n", - smp_processor_id(), v , v1); - irq_exit(); -} - -/** - * connect_bsp_APIC - attach the APIC to the interrupt system - */ -void __init connect_bsp_APIC(void) -{ -#ifdef CONFIG_X86_32 - if (pic_mode) { - /* - * Do not trust the local APIC being empty at bootup. - */ - clear_local_APIC(); - /* - * PIC mode, enable APIC mode in the IMCR, i.e. connect BSP's - * local APIC to INT and NMI lines. - */ - apic_printk(APIC_VERBOSE, "leaving PIC mode, " - "enabling APIC mode.\n"); - outb(0x70, 0x22); - outb(0x01, 0x23); - } -#endif - enable_apic_mode(); -} - -/** - * disconnect_bsp_APIC - detach the APIC from the interrupt system - * @virt_wire_setup: indicates, whether virtual wire mode is selected - * - * Virtual wire mode is necessary to deliver legacy interrupts even when the - * APIC is disabled. - */ -void disconnect_bsp_APIC(int virt_wire_setup) -{ - unsigned int value; - -#ifdef CONFIG_X86_32 - if (pic_mode) { - /* - * Put the board back into PIC mode (has an effect only on - * certain older boards). Note that APIC interrupts, including - * IPIs, won't work beyond this point! The only exception are - * INIT IPIs. - */ - apic_printk(APIC_VERBOSE, "disabling APIC mode, " - "entering PIC mode.\n"); - outb(0x70, 0x22); - outb(0x00, 0x23); - return; - } -#endif - - /* Go back to Virtual Wire compatibility mode */ - - /* For the spurious interrupt use vector F, and enable it */ - value = apic_read(APIC_SPIV); - value &= ~APIC_VECTOR_MASK; - value |= APIC_SPIV_APIC_ENABLED; - value |= 0xf; - apic_write(APIC_SPIV, value); - - if (!virt_wire_setup) { - /* - * For LVT0 make it edge triggered, active high, - * external and enabled - */ - value = apic_read(APIC_LVT0); - value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING | - APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR | - APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED); - value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING; - value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT); - apic_write(APIC_LVT0, value); - } else { - /* Disable LVT0 */ - apic_write(APIC_LVT0, APIC_LVT_MASKED); - } - - /* - * For LVT1 make it edge triggered, active high, - * nmi and enabled - */ - value = apic_read(APIC_LVT1); - value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING | - APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR | - APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED); - value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING; - value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI); - apic_write(APIC_LVT1, value); -} - -void __cpuinit generic_processor_info(int apicid, int version) -{ - int cpu; - cpumask_t tmp_map; - - /* - * Validate version - */ - if (version == 0x0) { - printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! " - "fixing up to 0x10. (tell your hw vendor)\n", - version); - version = 0x10; - } - apic_version[apicid] = version; - - if (num_processors >= NR_CPUS) { - printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached." - " Processor ignored.\n", NR_CPUS); - return; - } - - num_processors++; - cpus_complement(tmp_map, cpu_present_map); - cpu = first_cpu(tmp_map); - - physid_set(apicid, phys_cpu_present_map); - if (apicid == boot_cpu_physical_apicid) { - /* - * x86_bios_cpu_apicid is required to have processors listed - * in same order as logical cpu numbers. Hence the first - * entry is BSP, and so on. - */ - cpu = 0; - } - if (apicid > max_physical_apicid) - max_physical_apicid = apicid; - -#ifdef CONFIG_X86_32 - /* - * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y - * but we need to work other dependencies like SMP_SUSPEND etc - * before this can be done without some confusion. - * if (CPU_HOTPLUG_ENABLED || num_processors > 8) - * - Ashok Raj <ashok.raj@intel.com> - */ - if (max_physical_apicid >= 8) { - switch (boot_cpu_data.x86_vendor) { - case X86_VENDOR_INTEL: - if (!APIC_XAPIC(version)) { - def_to_bigsmp = 0; - break; - } - /* If P4 and above fall through */ - case X86_VENDOR_AMD: - def_to_bigsmp = 1; - } - } -#endif - -#if defined(CONFIG_X86_SMP) || defined(CONFIG_X86_64) - /* are we being called early in kernel startup? */ - if (early_per_cpu_ptr(x86_cpu_to_apicid)) { - u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid); - u16 *bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); - - cpu_to_apicid[cpu] = apicid; - bios_cpu_apicid[cpu] = apicid; - } else { - per_cpu(x86_cpu_to_apicid, cpu) = apicid; - per_cpu(x86_bios_cpu_apicid, cpu) = apicid; - } -#endif - - cpu_set(cpu, cpu_possible_map); - cpu_set(cpu, cpu_present_map); -} - -int hard_smp_processor_id(void) -{ - return read_apic_id(); -} - -/* - * Power management - */ -#ifdef CONFIG_PM - -static struct { - /* - * 'active' is true if the local APIC was enabled by us and - * not the BIOS; this signifies that we are also responsible - * for disabling it before entering apm/acpi suspend - */ - int active; - /* r/w apic fields */ - unsigned int apic_id; - unsigned int apic_taskpri; - unsigned int apic_ldr; - unsigned int apic_dfr; - unsigned int apic_spiv; - unsigned int apic_lvtt; - unsigned int apic_lvtpc; - unsigned int apic_lvt0; - unsigned int apic_lvt1; - unsigned int apic_lvterr; - unsigned int apic_tmict; - unsigned int apic_tdcr; - unsigned int apic_thmr; -} apic_pm_state; - -static int lapic_suspend(struct sys_device *dev, pm_message_t state) -{ - unsigned long flags; - int maxlvt; - - if (!apic_pm_state.active) - return 0; - - maxlvt = lapic_get_maxlvt(); - - apic_pm_state.apic_id = apic_read(APIC_ID); - apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI); - apic_pm_state.apic_ldr = apic_read(APIC_LDR); - apic_pm_state.apic_dfr = apic_read(APIC_DFR); - apic_pm_state.apic_spiv = apic_read(APIC_SPIV); - apic_pm_state.apic_lvtt = apic_read(APIC_LVTT); - if (maxlvt >= 4) - apic_pm_state.apic_lvtpc = apic_read(APIC_LVTPC); - apic_pm_state.apic_lvt0 = apic_read(APIC_LVT0); - apic_pm_state.apic_lvt1 = apic_read(APIC_LVT1); - apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR); - apic_pm_state.apic_tmict = apic_read(APIC_TMICT); - apic_pm_state.apic_tdcr = apic_read(APIC_TDCR); -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) - if (maxlvt >= 5) - apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); -#endif - - local_irq_save(flags); - disable_local_APIC(); - local_irq_restore(flags); - return 0; -} - -static int lapic_resume(struct sys_device *dev) -{ - unsigned int l, h; - unsigned long flags; - int maxlvt; - - if (!apic_pm_state.active) - return 0; - - maxlvt = lapic_get_maxlvt(); - - local_irq_save(flags); - -#ifdef CONFIG_X86_64 - if (x2apic) - enable_x2apic(); - else -#endif - { - /* - * Make sure the APICBASE points to the right address - * - * FIXME! This will be wrong if we ever support suspend on - * SMP! We'll need to do this as part of the CPU restore! - */ - rdmsr(MSR_IA32_APICBASE, l, h); - l &= ~MSR_IA32_APICBASE_BASE; - l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr; - wrmsr(MSR_IA32_APICBASE, l, h); - } - - apic_write(APIC_LVTERR, ERROR_APIC_VECTOR | APIC_LVT_MASKED); - apic_write(APIC_ID, apic_pm_state.apic_id); - apic_write(APIC_DFR, apic_pm_state.apic_dfr); - apic_write(APIC_LDR, apic_pm_state.apic_ldr); - apic_write(APIC_TASKPRI, apic_pm_state.apic_taskpri); - apic_write(APIC_SPIV, apic_pm_state.apic_spiv); - apic_write(APIC_LVT0, apic_pm_state.apic_lvt0); - apic_write(APIC_LVT1, apic_pm_state.apic_lvt1); -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) - if (maxlvt >= 5) - apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr); -#endif - if (maxlvt >= 4) - apic_write(APIC_LVTPC, apic_pm_state.apic_lvtpc); - apic_write(APIC_LVTT, apic_pm_state.apic_lvtt); - apic_write(APIC_TDCR, apic_pm_state.apic_tdcr); - apic_write(APIC_TMICT, apic_pm_state.apic_tmict); - apic_write(APIC_ESR, 0); - apic_read(APIC_ESR); - apic_write(APIC_LVTERR, apic_pm_state.apic_lvterr); - apic_write(APIC_ESR, 0); - apic_read(APIC_ESR); - - local_irq_restore(flags); - - return 0; -} - -/* - * This device has no shutdown method - fully functioning local APICs - * are needed on every CPU up until machine_halt/restart/poweroff. - */ - -static struct sysdev_class lapic_sysclass = { - .name = "lapic", - .resume = lapic_resume, - .suspend = lapic_suspend, -}; - -static struct sys_device device_lapic = { - .id = 0, - .cls = &lapic_sysclass, -}; - -static void __cpuinit apic_pm_activate(void) -{ - apic_pm_state.active = 1; -} - -static int __init init_lapic_sysfs(void) -{ - int error; - - if (!cpu_has_apic) - return 0; - /* XXX: remove suspend/resume procs if !apic_pm_state.active? */ - - error = sysdev_class_register(&lapic_sysclass); - if (!error) - error = sysdev_register(&device_lapic); - return error; -} -device_initcall(init_lapic_sysfs); - -#else /* CONFIG_PM */ - -static void apic_pm_activate(void) { } - -#endif /* CONFIG_PM */ - -/* - * apic_is_clustered_box() -- Check if we can expect good TSC - * - * Thus far, the major user of this is IBM's Summit2 series: - * - * Clustered boxes may have unsynced TSC problems if they are - * multi-chassis. Use available data to take a good guess. - * If in doubt, go HPET. - */ -__cpuinit int apic_is_clustered_box(void) -{ - int i, clusters, zeros; - unsigned id; - u16 *bios_cpu_apicid; - DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS); - - /* - * there is not this kind of box with AMD CPU yet. - * Some AMD box with quadcore cpu and 8 sockets apicid - * will be [4, 0x23] or [8, 0x27] could be thought to - * vsmp box still need checking... - */ - if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && !is_vsmp_box()) - return 0; - - bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); - bitmap_zero(clustermap, NUM_APIC_CLUSTERS); - - for (i = 0; i < NR_CPUS; i++) { - /* are we being called early in kernel startup? */ - if (bios_cpu_apicid) { - id = bios_cpu_apicid[i]; - } - else if (i < nr_cpu_ids) { - if (cpu_present(i)) - id = per_cpu(x86_bios_cpu_apicid, i); - else - continue; - } - else - break; - - if (id != BAD_APICID) - __set_bit(APIC_CLUSTERID(id), clustermap); - } - - /* Problem: Partially populated chassis may not have CPUs in some of - * the APIC clusters they have been allocated. Only present CPUs have - * x86_bios_cpu_apicid entries, thus causing zeroes in the bitmap. - * Since clusters are allocated sequentially, count zeros only if - * they are bounded by ones. - */ - clusters = 0; - zeros = 0; - for (i = 0; i < NUM_APIC_CLUSTERS; i++) { - if (test_bit(i, clustermap)) { - clusters += 1 + zeros; - zeros = 0; - } else - ++zeros; - } - - /* ScaleMP vSMPowered boxes have one cluster per board and TSCs are - * not guaranteed to be synced between boards - */ - if (is_vsmp_box() && clusters > 1) - return 1; - - /* - * If clusters > 2, then should be multi-chassis. - * May have to revisit this when multi-core + hyperthreaded CPUs come - * out, but AFAIK this will work even for them. - */ - return (clusters > 2); -} - -static __init int setup_nox2apic(char *str) -{ - disable_x2apic = 1; - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_X2APIC); - return 0; -} -early_param("nox2apic", setup_nox2apic); - - -/* - * APIC command line parameters - */ -static int __init setup_disableapic(char *arg) -{ - disable_apic = 1; - setup_clear_cpu_cap(X86_FEATURE_APIC); - return 0; -} -early_param("disableapic", setup_disableapic); - -/* same as disableapic, for compatibility */ -static int __init setup_nolapic(char *arg) -{ - return setup_disableapic(arg); -} -early_param("nolapic", setup_nolapic); - -static int __init parse_lapic_timer_c2_ok(char *arg) -{ - local_apic_timer_c2_ok = 1; - return 0; -} -early_param("lapic_timer_c2_ok", parse_lapic_timer_c2_ok); - -static int __init parse_disable_apic_timer(char *arg) -{ - disable_apic_timer = 1; - return 0; -} -early_param("noapictimer", parse_disable_apic_timer); - -static int __init parse_nolapic_timer(char *arg) -{ - disable_apic_timer = 1; - return 0; -} -early_param("nolapic_timer", parse_nolapic_timer); - -static __init int setup_apicpmtimer(char *s) -{ - apic_calibrate_pmtmr = 1; - notsc_setup(NULL); - return 0; -} -__setup("apicpmtimer", setup_apicpmtimer); - -static int __init apic_set_verbosity(char *arg) -{ - if (!arg) { -#ifdef CONFIG_X86_64 - skip_ioapic_setup = 0; - ioapic_force = 1; - return 0; -#endif - return -EINVAL; - } - - if (strcmp("debug", arg) == 0) - apic_verbosity = APIC_DEBUG; - else if (strcmp("verbose", arg) == 0) - apic_verbosity = APIC_VERBOSE; - else { - printk(KERN_WARNING "APIC Verbosity level %s not recognised" - " use apic=verbose or apic=debug\n", arg); - return -EINVAL; - } - - return 0; -} -early_param("apic", apic_set_verbosity); - -static int __init lapic_insert_resource(void) -{ - if (!apic_phys) - return -1; - - /* Put local APIC into the resource map. */ - lapic_resource.start = apic_phys; - lapic_resource.end = lapic_resource.start + PAGE_SIZE - 1; - insert_resource(&iomem_resource, &lapic_resource); - - return 0; -} - -/* - * need call insert after e820_reserve_resources() - * that is using request_resource - */ -late_initcall(lapic_insert_resource); diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 5145a6e72bbb..3a26525a3f31 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -391,11 +391,7 @@ static int power_off; #else static int power_off = 1; #endif -#ifdef CONFIG_APM_REAL_MODE_POWER_OFF -static int realmode_power_off = 1; -#else static int realmode_power_off; -#endif #ifdef CONFIG_APM_ALLOW_INTS static int allow_ints = 1; #else diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c index 6649d09ad88f..ee4df08feee6 100644 --- a/arch/x86/kernel/asm-offsets_32.c +++ b/arch/x86/kernel/asm-offsets_32.c @@ -11,7 +11,7 @@ #include <linux/suspend.h> #include <linux/kbuild.h> #include <asm/ucontext.h> -#include "sigframe.h" +#include <asm/sigframe.h> #include <asm/pgtable.h> #include <asm/fixmap.h> #include <asm/processor.h> diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c index 505543a75a56..1d41d3f1edbc 100644 --- a/arch/x86/kernel/asm-offsets_64.c +++ b/arch/x86/kernel/asm-offsets_64.c @@ -20,9 +20,11 @@ #include <xen/interface/xen.h> +#include <asm/sigframe.h> + #define __NO_STUBS 1 #undef __SYSCALL -#undef ASM_X86__UNISTD_64_H +#undef _ASM_X86_UNISTD_64_H #define __SYSCALL(nr, sym) [nr] = 1, static char syscalls[] = { #include <asm/unistd.h> @@ -87,7 +89,7 @@ int main(void) BLANK(); #undef ENTRY DEFINE(IA32_RT_SIGFRAME_sigcontext, - offsetof (struct rt_sigframe32, uc.uc_mcontext)); + offsetof (struct rt_sigframe_ia32, uc.uc_mcontext)); BLANK(); #endif DEFINE(pbe_address, offsetof(struct pbe, address)); diff --git a/arch/x86/kernel/bios_uv.c b/arch/x86/kernel/bios_uv.c index fdd585f9c53d..2a0a2a3cac26 100644 --- a/arch/x86/kernel/bios_uv.c +++ b/arch/x86/kernel/bios_uv.c @@ -1,8 +1,6 @@ /* * BIOS run time interface routines. * - * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved. - * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or @@ -16,33 +14,178 @@ * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) Russ Anderson */ +#include <linux/efi.h> +#include <asm/efi.h> +#include <linux/io.h> #include <asm/uv/bios.h> +#include <asm/uv/uv_hub.h> -const char * -x86_bios_strerror(long status) +struct uv_systab uv_systab; + +s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5) { - const char *str; - switch (status) { - case 0: str = "Call completed without error"; break; - case -1: str = "Not implemented"; break; - case -2: str = "Invalid argument"; break; - case -3: str = "Call completed with error"; break; - default: str = "Unknown BIOS status code"; break; - } - return str; + struct uv_systab *tab = &uv_systab; + + if (!tab->function) + /* + * BIOS does not support UV systab + */ + return BIOS_STATUS_UNIMPLEMENTED; + + return efi_call6((void *)__va(tab->function), + (u64)which, a1, a2, a3, a4, a5); +} + +s64 uv_bios_call_irqsave(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, + u64 a4, u64 a5) +{ + unsigned long bios_flags; + s64 ret; + + local_irq_save(bios_flags); + ret = uv_bios_call(which, a1, a2, a3, a4, a5); + local_irq_restore(bios_flags); + + return ret; +} + +s64 uv_bios_call_reentrant(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, + u64 a4, u64 a5) +{ + s64 ret; + + preempt_disable(); + ret = uv_bios_call(which, a1, a2, a3, a4, a5); + preempt_enable(); + + return ret; +} + + +long sn_partition_id; +EXPORT_SYMBOL_GPL(sn_partition_id); +long sn_coherency_id; +EXPORT_SYMBOL_GPL(sn_coherency_id); +long sn_region_size; +EXPORT_SYMBOL_GPL(sn_region_size); +int uv_type; + + +s64 uv_bios_get_sn_info(int fc, int *uvtype, long *partid, long *coher, + long *region) +{ + s64 ret; + u64 v0, v1; + union partition_info_u part; + + ret = uv_bios_call_irqsave(UV_BIOS_GET_SN_INFO, fc, + (u64)(&v0), (u64)(&v1), 0, 0); + if (ret != BIOS_STATUS_SUCCESS) + return ret; + + part.val = v0; + if (uvtype) + *uvtype = part.hub_version; + if (partid) + *partid = part.partition_id; + if (coher) + *coher = part.coherence_id; + if (region) + *region = part.region_size; + return ret; } -long -x86_bios_freq_base(unsigned long which, unsigned long *ticks_per_second, - unsigned long *drift_info) +int +uv_bios_mq_watchlist_alloc(int blade, unsigned long addr, unsigned int mq_size, + unsigned long *intr_mmr_offset) { - struct uv_bios_retval isrv; + union uv_watchlist_u size_blade; + u64 watchlist; + s64 ret; - BIOS_CALL(isrv, BIOS_FREQ_BASE, which, 0, 0, 0, 0, 0, 0); - *ticks_per_second = isrv.v0; - *drift_info = isrv.v1; - return isrv.status; + size_blade.size = mq_size; + size_blade.blade = blade; + + /* + * bios returns watchlist number or negative error number. + */ + ret = (int)uv_bios_call_irqsave(UV_BIOS_WATCHLIST_ALLOC, addr, + size_blade.val, (u64)intr_mmr_offset, + (u64)&watchlist, 0); + if (ret < BIOS_STATUS_SUCCESS) + return ret; + + return watchlist; +} +EXPORT_SYMBOL_GPL(uv_bios_mq_watchlist_alloc); + +int +uv_bios_mq_watchlist_free(int blade, int watchlist_num) +{ + return (int)uv_bios_call_irqsave(UV_BIOS_WATCHLIST_FREE, + blade, watchlist_num, 0, 0, 0); +} +EXPORT_SYMBOL_GPL(uv_bios_mq_watchlist_free); + +s64 +uv_bios_change_memprotect(u64 paddr, u64 len, enum uv_memprotect perms) +{ + return uv_bios_call_irqsave(UV_BIOS_MEMPROTECT, paddr, len, + perms, 0, 0); +} +EXPORT_SYMBOL_GPL(uv_bios_change_memprotect); + +s64 +uv_bios_reserved_page_pa(u64 buf, u64 *cookie, u64 *addr, u64 *len) +{ + s64 ret; + + ret = uv_bios_call_irqsave(UV_BIOS_GET_PARTITION_ADDR, (u64)cookie, + (u64)addr, buf, (u64)len, 0); + return ret; +} +EXPORT_SYMBOL_GPL(uv_bios_reserved_page_pa); + +s64 uv_bios_freq_base(u64 clock_type, u64 *ticks_per_second) +{ + return uv_bios_call(UV_BIOS_FREQ_BASE, clock_type, + (u64)ticks_per_second, 0, 0, 0); } -EXPORT_SYMBOL_GPL(x86_bios_freq_base); +EXPORT_SYMBOL_GPL(uv_bios_freq_base); + + +#ifdef CONFIG_EFI +void uv_bios_init(void) +{ + struct uv_systab *tab; + + if ((efi.uv_systab == EFI_INVALID_TABLE_ADDR) || + (efi.uv_systab == (unsigned long)NULL)) { + printk(KERN_CRIT "No EFI UV System Table.\n"); + uv_systab.function = (unsigned long)NULL; + return; + } + + tab = (struct uv_systab *)ioremap(efi.uv_systab, + sizeof(struct uv_systab)); + if (strncmp(tab->signature, "UVST", 4) != 0) + printk(KERN_ERR "bad signature in UV system table!"); + + /* + * Copy table to permanent spot for later use. + */ + memcpy(&uv_systab, tab, sizeof(struct uv_systab)); + iounmap(tab); + + printk(KERN_INFO "EFI UV System Table Revision %d\n", tab->revision); +} +#else /* !CONFIG_EFI */ + +void uv_bios_init(void) { } +#endif + diff --git a/arch/x86/kernel/check.c b/arch/x86/kernel/check.c new file mode 100644 index 000000000000..2ac0ab71412a --- /dev/null +++ b/arch/x86/kernel/check.c @@ -0,0 +1,161 @@ +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/kthread.h> +#include <linux/workqueue.h> +#include <asm/e820.h> +#include <asm/proto.h> + +/* + * Some BIOSes seem to corrupt the low 64k of memory during events + * like suspend/resume and unplugging an HDMI cable. Reserve all + * remaining free memory in that area and fill it with a distinct + * pattern. + */ +#define MAX_SCAN_AREAS 8 + +static int __read_mostly memory_corruption_check = -1; + +static unsigned __read_mostly corruption_check_size = 64*1024; +static unsigned __read_mostly corruption_check_period = 60; /* seconds */ + +static struct e820entry scan_areas[MAX_SCAN_AREAS]; +static int num_scan_areas; + + +static __init int set_corruption_check(char *arg) +{ + char *end; + + memory_corruption_check = simple_strtol(arg, &end, 10); + + return (*end == 0) ? 0 : -EINVAL; +} +early_param("memory_corruption_check", set_corruption_check); + +static __init int set_corruption_check_period(char *arg) +{ + char *end; + + corruption_check_period = simple_strtoul(arg, &end, 10); + + return (*end == 0) ? 0 : -EINVAL; +} +early_param("memory_corruption_check_period", set_corruption_check_period); + +static __init int set_corruption_check_size(char *arg) +{ + char *end; + unsigned size; + + size = memparse(arg, &end); + + if (*end == '\0') + corruption_check_size = size; + + return (size == corruption_check_size) ? 0 : -EINVAL; +} +early_param("memory_corruption_check_size", set_corruption_check_size); + + +void __init setup_bios_corruption_check(void) +{ + u64 addr = PAGE_SIZE; /* assume first page is reserved anyway */ + + if (memory_corruption_check == -1) { + memory_corruption_check = +#ifdef CONFIG_X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK + 1 +#else + 0 +#endif + ; + } + + if (corruption_check_size == 0) + memory_corruption_check = 0; + + if (!memory_corruption_check) + return; + + corruption_check_size = round_up(corruption_check_size, PAGE_SIZE); + + while (addr < corruption_check_size && num_scan_areas < MAX_SCAN_AREAS) { + u64 size; + addr = find_e820_area_size(addr, &size, PAGE_SIZE); + + if (addr == 0) + break; + + if ((addr + size) > corruption_check_size) + size = corruption_check_size - addr; + + if (size == 0) + break; + + e820_update_range(addr, size, E820_RAM, E820_RESERVED); + scan_areas[num_scan_areas].addr = addr; + scan_areas[num_scan_areas].size = size; + num_scan_areas++; + + /* Assume we've already mapped this early memory */ + memset(__va(addr), 0, size); + + addr += size; + } + + printk(KERN_INFO "Scanning %d areas for low memory corruption\n", + num_scan_areas); + update_e820(); +} + + +void check_for_bios_corruption(void) +{ + int i; + int corruption = 0; + + if (!memory_corruption_check) + return; + + for (i = 0; i < num_scan_areas; i++) { + unsigned long *addr = __va(scan_areas[i].addr); + unsigned long size = scan_areas[i].size; + + for (; size; addr++, size -= sizeof(unsigned long)) { + if (!*addr) + continue; + printk(KERN_ERR "Corrupted low memory at %p (%lx phys) = %08lx\n", + addr, __pa(addr), *addr); + corruption = 1; + *addr = 0; + } + } + + WARN_ONCE(corruption, KERN_ERR "Memory corruption detected in low memory\n"); +} + +static void check_corruption(struct work_struct *dummy); +static DECLARE_DELAYED_WORK(bios_check_work, check_corruption); + +static void check_corruption(struct work_struct *dummy) +{ + check_for_bios_corruption(); + schedule_delayed_work(&bios_check_work, + round_jiffies_relative(corruption_check_period*HZ)); +} + +static int start_periodic_check_for_corruption(void) +{ + if (!memory_corruption_check || corruption_check_period == 0) + return 0; + + printk(KERN_INFO "Scanning for low memory corruption every %d seconds\n", + corruption_check_period); + + /* First time we run the checks right away */ + schedule_delayed_work(&bios_check_work, 0); + return 0; +} + +module_init(start_periodic_check_for_corruption); + diff --git a/arch/x86/kernel/cpu/.gitignore b/arch/x86/kernel/cpu/.gitignore new file mode 100644 index 000000000000..667df55a4399 --- /dev/null +++ b/arch/x86/kernel/cpu/.gitignore @@ -0,0 +1 @@ +capflags.c diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 7f0b45a5d788..82db7f45e2de 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -2,8 +2,14 @@ # Makefile for x86-compatible CPU details and quirks # +# Don't trace early stages of a secondary CPU boot +ifdef CONFIG_FUNCTION_TRACER +CFLAGS_REMOVE_common.o = -pg +endif + obj-y := intel_cacheinfo.o addon_cpuid_features.o obj-y += proc.o capflags.o powerflags.o common.o +obj-y += vmware.o hypervisor.o obj-$(CONFIG_X86_32) += bugs.o cmpxchg.o obj-$(CONFIG_X86_64) += bugs_64.o @@ -25,7 +31,7 @@ obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o quiet_cmd_mkcapflags = MKCAP $@ cmd_mkcapflags = $(PERL) $(srctree)/$(src)/mkcapflags.pl $< $@ -cpufeature = $(src)/../../../../include/asm-x86/cpufeature.h +cpufeature = $(src)/../../include/asm/cpufeature.h targets += capflags.c $(obj)/capflags.c: $(cpufeature) $(src)/mkcapflags.pl FORCE diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c index 0d9c993aa93e..2cf23634b6d9 100644 --- a/arch/x86/kernel/cpu/addon_cpuid_features.c +++ b/arch/x86/kernel/cpu/addon_cpuid_features.c @@ -69,7 +69,7 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) */ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c) { -#ifdef CONFIG_SMP +#ifdef CONFIG_X86_SMP unsigned int eax, ebx, ecx, edx, sub_index; unsigned int ht_mask_width, core_plus_mask_width; unsigned int core_select_mask, core_level_siblings; @@ -120,9 +120,17 @@ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c) c->cpu_core_id = phys_pkg_id(c->initial_apicid, ht_mask_width) & core_select_mask; c->phys_proc_id = phys_pkg_id(c->initial_apicid, core_plus_mask_width); + /* + * Reinit the apicid, now that we have extended initial_apicid. + */ + c->apicid = phys_pkg_id(c->initial_apicid, 0); #else c->cpu_core_id = phys_pkg_id(ht_mask_width) & core_select_mask; c->phys_proc_id = phys_pkg_id(core_plus_mask_width); + /* + * Reinit the apicid, now that we have extended initial_apicid. + */ + c->apicid = phys_pkg_id(0); #endif c->x86_max_cores = (core_level_siblings / smp_num_siblings); diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 32e73520adf7..7c878f6aa919 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -249,7 +249,7 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) } numa_set_node(cpu, node); - printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node); + printk(KERN_INFO "CPU %d/0x%x -> Node %d\n", cpu, apicid, node); #endif } @@ -283,9 +283,14 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c) { early_init_amd_mc(c); - /* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */ - if (c->x86_power & (1<<8)) + /* + * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate + * with P/T states and does not stop in deep C-states + */ + if (c->x86_power & (1 << 8)) { set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); + set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC); + } #ifdef CONFIG_X86_64 set_cpu_cap(c, X86_FEATURE_SYSCALL32); diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 25581dcb280e..42e0853030cb 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -20,6 +20,7 @@ #include <asm/pat.h> #include <asm/asm.h> #include <asm/numa.h> +#include <asm/smp.h> #ifdef CONFIG_X86_LOCAL_APIC #include <asm/mpspec.h> #include <asm/apic.h> @@ -35,6 +36,7 @@ #include <asm/proto.h> #include <asm/sections.h> #include <asm/setup.h> +#include <asm/hypervisor.h> #include "cpu.h" @@ -549,6 +551,10 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) this_cpu->c_early_init(c); validate_pat_support(c); + +#ifdef CONFIG_SMP + c->cpu_index = boot_cpu_id; +#endif } void __init early_cpu_init(void) @@ -698,6 +704,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) detect_ht(c); #endif + init_hypervisor(c); /* * On SMP, boot_cpu_data holds the common feature set between * all CPUs; so make sure that we indicate which features are @@ -857,7 +864,7 @@ EXPORT_SYMBOL(_cpu_pda); struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table }; -char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss; +static char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss; void __cpuinit pda_init(int cpu) { @@ -898,8 +905,8 @@ void __cpuinit pda_init(int cpu) } } -char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + - DEBUG_STKSZ] __page_aligned_bss; +static char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + + DEBUG_STKSZ] __page_aligned_bss; extern asmlinkage void ignore_sysret(void); @@ -1134,7 +1141,7 @@ void __cpuinit cpu_init(void) /* * Boot processor to setup the FP and extended state context info. */ - if (!smp_processor_id()) + if (smp_processor_id() == boot_cpu_id) init_thread_xstate(); xsave_init(); diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index c24c4a487b7c..88ea02dcb622 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -33,6 +33,7 @@ #include <linux/cpufreq.h> #include <linux/compiler.h> #include <linux/dmi.h> +#include <linux/ftrace.h> #include <linux/acpi.h> #include <acpi/processor.h> @@ -391,6 +392,7 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy, unsigned int next_perf_state = 0; /* Index into perf table */ unsigned int i; int result = 0; + struct power_trace it; dprintk("acpi_cpufreq_target %d (%d)\n", target_freq, policy->cpu); @@ -427,6 +429,8 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy, } } + trace_power_mark(&it, POWER_PSTATE, next_perf_state); + switch (data->cpu_feature) { case SYSTEM_INTEL_MSR_CAPABLE: cmd.type = SYSTEM_INTEL_MSR_CAPABLE; @@ -780,6 +784,9 @@ static int __init acpi_cpufreq_init(void) { int ret; + if (acpi_disabled) + return 0; + dprintk("acpi_cpufreq_init\n"); ret = acpi_cpufreq_early_init(); diff --git a/arch/x86/kernel/cpu/cpufreq/longhaul.c b/arch/x86/kernel/cpu/cpufreq/longhaul.c index 06fcce516d51..b0461856acfb 100644 --- a/arch/x86/kernel/cpu/cpufreq/longhaul.c +++ b/arch/x86/kernel/cpu/cpufreq/longhaul.c @@ -1,5 +1,5 @@ /* - * (C) 2001-2004 Dave Jones. <davej@codemonkey.org.uk> + * (C) 2001-2004 Dave Jones. <davej@redhat.com> * (C) 2002 Padraig Brady. <padraig@antefacto.com> * * Licensed under the terms of the GNU GPL License version 2. @@ -1019,7 +1019,7 @@ MODULE_PARM_DESC(scale_voltage, "Scale voltage of processor"); module_param(revid_errata, int, 0644); MODULE_PARM_DESC(revid_errata, "Ignore CPU Revision ID"); -MODULE_AUTHOR ("Dave Jones <davej@codemonkey.org.uk>"); +MODULE_AUTHOR ("Dave Jones <davej@redhat.com>"); MODULE_DESCRIPTION ("Longhaul driver for VIA Cyrix processors."); MODULE_LICENSE ("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c index b5ced806a316..c1ac5790c63e 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c @@ -246,7 +246,7 @@ static void __exit powernow_k6_exit(void) } -MODULE_AUTHOR("Arjan van de Ven <arjanv@redhat.com>, Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>"); +MODULE_AUTHOR("Arjan van de Ven, Dave Jones <davej@redhat.com>, Dominik Brodowski <linux@brodo.de>"); MODULE_DESCRIPTION("PowerNow! driver for AMD K6-2+ / K6-3+ processors."); MODULE_LICENSE("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c index 0a61159d7b71..7c7d56b43136 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c @@ -1,6 +1,6 @@ /* * AMD K7 Powernow driver. - * (C) 2003 Dave Jones <davej@codemonkey.org.uk> on behalf of SuSE Labs. + * (C) 2003 Dave Jones on behalf of SuSE Labs. * (C) 2003-2004 Dave Jones <davej@redhat.com> * * Licensed under the terms of the GNU GPL License version 2. @@ -692,7 +692,7 @@ static void __exit powernow_exit (void) module_param(acpi_force, int, 0444); MODULE_PARM_DESC(acpi_force, "Force ACPI to be used."); -MODULE_AUTHOR ("Dave Jones <davej@codemonkey.org.uk>"); +MODULE_AUTHOR ("Dave Jones <davej@redhat.com>"); MODULE_DESCRIPTION ("Powernow driver for AMD K7 processors."); MODULE_LICENSE ("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index 84bb395038d8..7f05f44b97e9 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -7,7 +7,7 @@ * Support : mark.langsdorf@amd.com * * Based on the powernow-k7.c module written by Dave Jones. - * (C) 2003 Dave Jones <davej@codemonkey.org.uk> on behalf of SuSE Labs + * (C) 2003 Dave Jones on behalf of SuSE Labs * (C) 2004 Dominik Brodowski <linux@brodo.de> * (C) 2004 Pavel Machek <pavel@suse.cz> * Licensed under the terms of the GNU GPL License version 2. @@ -45,7 +45,6 @@ #endif #define PFX "powernow-k8: " -#define BFX PFX "BIOS error: " #define VERSION "version 2.20.00" #include "powernow-k8.h" @@ -116,9 +115,20 @@ static int query_current_values_with_pending_wait(struct powernow_k8_data *data) u32 i = 0; if (cpu_family == CPU_HW_PSTATE) { - rdmsr(MSR_PSTATE_STATUS, lo, hi); - i = lo & HW_PSTATE_MASK; - data->currpstate = i; + if (data->currpstate == HW_PSTATE_INVALID) { + /* read (initial) hw pstate if not yet set */ + rdmsr(MSR_PSTATE_STATUS, lo, hi); + i = lo & HW_PSTATE_MASK; + + /* + * a workaround for family 11h erratum 311 might cause + * an "out-of-range Pstate if the core is in Pstate-0 + */ + if (i >= data->numps) + data->currpstate = HW_PSTATE_0; + else + data->currpstate = i; + } return 0; } do { @@ -536,35 +546,40 @@ static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst, u8 for (j = 0; j < data->numps; j++) { if (pst[j].vid > LEAST_VID) { - printk(KERN_ERR PFX "vid %d invalid : 0x%x\n", j, pst[j].vid); + printk(KERN_ERR FW_BUG PFX "vid %d invalid : 0x%x\n", + j, pst[j].vid); return -EINVAL; } if (pst[j].vid < data->rvo) { /* vid + rvo >= 0 */ - printk(KERN_ERR BFX "0 vid exceeded with pstate %d\n", j); + printk(KERN_ERR FW_BUG PFX "0 vid exceeded with pstate" + " %d\n", j); return -ENODEV; } if (pst[j].vid < maxvid + data->rvo) { /* vid + rvo >= maxvid */ - printk(KERN_ERR BFX "maxvid exceeded with pstate %d\n", j); + printk(KERN_ERR FW_BUG PFX "maxvid exceeded with pstate" + " %d\n", j); return -ENODEV; } if (pst[j].fid > MAX_FID) { - printk(KERN_ERR BFX "maxfid exceeded with pstate %d\n", j); + printk(KERN_ERR FW_BUG PFX "maxfid exceeded with pstate" + " %d\n", j); return -ENODEV; } if (j && (pst[j].fid < HI_FID_TABLE_BOTTOM)) { /* Only first fid is allowed to be in "low" range */ - printk(KERN_ERR BFX "two low fids - %d : 0x%x\n", j, pst[j].fid); + printk(KERN_ERR FW_BUG PFX "two low fids - %d : " + "0x%x\n", j, pst[j].fid); return -EINVAL; } if (pst[j].fid < lastfid) lastfid = pst[j].fid; } if (lastfid & 1) { - printk(KERN_ERR BFX "lastfid invalid\n"); + printk(KERN_ERR FW_BUG PFX "lastfid invalid\n"); return -EINVAL; } if (lastfid > LO_FID_TABLE_TOP) - printk(KERN_INFO BFX "first fid not from lo freq table\n"); + printk(KERN_INFO FW_BUG PFX "first fid not from lo freq table\n"); return 0; } @@ -672,13 +687,13 @@ static int find_psb_table(struct powernow_k8_data *data) dprintk("table vers: 0x%x\n", psb->tableversion); if (psb->tableversion != PSB_VERSION_1_4) { - printk(KERN_ERR BFX "PSB table is not v1.4\n"); + printk(KERN_ERR FW_BUG PFX "PSB table is not v1.4\n"); return -ENODEV; } dprintk("flags: 0x%x\n", psb->flags1); if (psb->flags1) { - printk(KERN_ERR BFX "unknown flags\n"); + printk(KERN_ERR FW_BUG PFX "unknown flags\n"); return -ENODEV; } @@ -705,7 +720,7 @@ static int find_psb_table(struct powernow_k8_data *data) } } if (cpst != 1) { - printk(KERN_ERR BFX "numpst must be 1\n"); + printk(KERN_ERR FW_BUG PFX "numpst must be 1\n"); return -ENODEV; } @@ -1117,6 +1132,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) } data->cpu = pol->cpu; + data->currpstate = HW_PSTATE_INVALID; if (powernow_k8_cpu_init_acpi(data)) { /* @@ -1130,17 +1146,19 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) "ACPI Processor module before starting this " "driver.\n"); #else - printk(KERN_ERR PFX "Your BIOS does not provide ACPI " - "_PSS objects in a way that Linux understands. " - "Please report this to the Linux ACPI maintainers" - " and complain to your BIOS vendor.\n"); + printk(KERN_ERR FW_BUG PFX "Your BIOS does not provide" + " ACPI _PSS objects in a way that Linux " + "understands. Please report this to the Linux " + "ACPI maintainers and complain to your BIOS " + "vendor.\n"); #endif kfree(data); return -ENODEV; } if (pol->cpu != 0) { - printk(KERN_ERR PFX "No ACPI _PSS objects for CPU other than " - "CPU0. Complain to your BIOS vendor.\n"); + printk(KERN_ERR FW_BUG PFX "No ACPI _PSS objects for " + "CPU other than CPU0. Complain to your BIOS " + "vendor.\n"); kfree(data); return -ENODEV; } @@ -1193,7 +1211,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) /* min/max the cpu is capable of */ if (cpufreq_frequency_table_cpuinfo(pol, data->powernow_table)) { - printk(KERN_ERR PFX "invalid powernow_table\n"); + printk(KERN_ERR FW_BUG PFX "invalid powernow_table\n"); powernow_k8_cpu_exit_acpi(data); kfree(data->powernow_table); kfree(data); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h index ab48cfed4d96..65cfb5d7f77f 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h @@ -5,6 +5,19 @@ * http://www.gnu.org/licenses/gpl.html */ + +enum pstate { + HW_PSTATE_INVALID = 0xff, + HW_PSTATE_0 = 0, + HW_PSTATE_1 = 1, + HW_PSTATE_2 = 2, + HW_PSTATE_3 = 3, + HW_PSTATE_4 = 4, + HW_PSTATE_5 = 5, + HW_PSTATE_6 = 6, + HW_PSTATE_7 = 7, +}; + struct powernow_k8_data { unsigned int cpu; @@ -23,7 +36,9 @@ struct powernow_k8_data { u32 exttype; /* extended interface = 1 */ /* keep track of the current fid / vid or pstate */ - u32 currvid, currfid, currpstate; + u32 currvid; + u32 currfid; + enum pstate currpstate; /* the powernow_table includes all frequency and vid/fid pairings: * fid are the lower 8 bits of the index, vid are the upper 8 bits. diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c index 191f7263c61d..04d0376b64b0 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c @@ -431,7 +431,7 @@ static void __exit speedstep_exit(void) } -MODULE_AUTHOR ("Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>"); +MODULE_AUTHOR ("Dave Jones <davej@redhat.com>, Dominik Brodowski <linux@brodo.de>"); MODULE_DESCRIPTION ("Speedstep driver for Intel mobile processors on chipsets with ICH-M southbridges."); MODULE_LICENSE ("GPL"); diff --git a/arch/x86/kernel/cpu/hypervisor.c b/arch/x86/kernel/cpu/hypervisor.c new file mode 100644 index 000000000000..fb5b86af0b01 --- /dev/null +++ b/arch/x86/kernel/cpu/hypervisor.c @@ -0,0 +1,58 @@ +/* + * Common hypervisor code + * + * Copyright (C) 2008, VMware, Inc. + * Author : Alok N Kataria <akataria@vmware.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. + * + */ + +#include <asm/processor.h> +#include <asm/vmware.h> +#include <asm/hypervisor.h> + +static inline void __cpuinit +detect_hypervisor_vendor(struct cpuinfo_x86 *c) +{ + if (vmware_platform()) { + c->x86_hyper_vendor = X86_HYPER_VENDOR_VMWARE; + } else { + c->x86_hyper_vendor = X86_HYPER_VENDOR_NONE; + } +} + +unsigned long get_hypervisor_tsc_freq(void) +{ + if (boot_cpu_data.x86_hyper_vendor == X86_HYPER_VENDOR_VMWARE) + return vmware_get_tsc_khz(); + return 0; +} + +static inline void __cpuinit +hypervisor_set_feature_bits(struct cpuinfo_x86 *c) +{ + if (boot_cpu_data.x86_hyper_vendor == X86_HYPER_VENDOR_VMWARE) { + vmware_set_feature_bits(c); + return; + } +} + +void __cpuinit init_hypervisor(struct cpuinfo_x86 *c) +{ + detect_hypervisor_vendor(c); + hypervisor_set_feature_bits(c); +} diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 99468dbd08da..8ea6929e974c 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -11,7 +11,6 @@ #include <asm/pgtable.h> #include <asm/msr.h> #include <asm/uaccess.h> -#include <asm/ptrace.h> #include <asm/ds.h> #include <asm/bugs.h> @@ -41,6 +40,16 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) if (c->x86 == 15 && c->x86_cache_alignment == 64) c->x86_cache_alignment = 128; #endif + + /* + * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate + * with P/T states and does not stop in deep C-states + */ + if (c->x86_power & (1 << 8)) { + set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); + set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC); + } + } #ifdef CONFIG_X86_32 @@ -174,7 +183,7 @@ static void __cpuinit srat_detect_node(void) node = first_node(node_online_map); numa_set_node(cpu, node); - printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node); + printk(KERN_INFO "CPU %d/0x%x -> Node %d\n", cpu, apicid, node); #endif } @@ -242,6 +251,13 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) intel_workarounds(c); + /* + * Detect the extended topology information if available. This + * will reinitialise the initial_apicid which will be used + * in init_intel_cacheinfo() + */ + detect_extended_topology(c); + l2 = init_intel_cacheinfo(c); if (c->cpuid_level > 9) { unsigned eax = cpuid_eax(10); @@ -307,13 +323,8 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_P4); if (c->x86 == 6) set_cpu_cap(c, X86_FEATURE_P3); - - if (cpu_has_bts) - ptrace_bts_init_intel(c); - #endif - detect_extended_topology(c); if (!cpu_has(c, X86_FEATURE_XTOPOLOGY)) { /* * let's use the legacy cpuid vector 0x1 and 0x4 for topology diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 3f46afbb1cf1..68b5d8681cbb 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -644,20 +644,17 @@ static inline ssize_t show_shared_cpu_list(struct _cpuid4_info *leaf, char *buf) return show_shared_cpu_map_func(leaf, 1, buf); } -static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf) { - switch(this_leaf->eax.split.type) { - case CACHE_TYPE_DATA: +static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf) +{ + switch (this_leaf->eax.split.type) { + case CACHE_TYPE_DATA: return sprintf(buf, "Data\n"); - break; - case CACHE_TYPE_INST: + case CACHE_TYPE_INST: return sprintf(buf, "Instruction\n"); - break; - case CACHE_TYPE_UNIFIED: + case CACHE_TYPE_UNIFIED: return sprintf(buf, "Unified\n"); - break; - default: + default: return sprintf(buf, "Unknown\n"); - break; } } diff --git a/arch/x86/kernel/cpu/mcheck/k7.c b/arch/x86/kernel/cpu/mcheck/k7.c index f390c9f66351..dd3af6e7b39a 100644 --- a/arch/x86/kernel/cpu/mcheck/k7.c +++ b/arch/x86/kernel/cpu/mcheck/k7.c @@ -1,6 +1,6 @@ /* - * Athlon/Hammer specific Machine Check Exception Reporting - * (C) Copyright 2002 Dave Jones <davej@codemonkey.org.uk> + * Athlon specific Machine Check Exception Reporting + * (C) Copyright 2002 Dave Jones <davej@redhat.com> */ #include <linux/init.h> diff --git a/arch/x86/kernel/cpu/mcheck/mce_32.c b/arch/x86/kernel/cpu/mcheck/mce_32.c index 774d87cfd8cd..0ebf3fc6a610 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_32.c +++ b/arch/x86/kernel/cpu/mcheck/mce_32.c @@ -1,6 +1,6 @@ /* * mce.c - x86 Machine Check Exception Reporting - * (c) 2002 Alan Cox <alan@redhat.com>, Dave Jones <davej@codemonkey.org.uk> + * (c) 2002 Alan Cox <alan@redhat.com>, Dave Jones <davej@redhat.com> */ #include <linux/init.h> diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c index 4b031a4ac856..1c838032fd37 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_64.c @@ -510,12 +510,9 @@ static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c) */ void __cpuinit mcheck_init(struct cpuinfo_x86 *c) { - static cpumask_t mce_cpus = CPU_MASK_NONE; - mce_cpu_quirks(c); if (mce_dont_init || - cpu_test_and_set(smp_processor_id(), mce_cpus) || !mce_available(c)) return; diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c index 5eb390a4b2e9..748c8f9e7a05 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c @@ -237,7 +237,7 @@ asmlinkage void mce_threshold_interrupt(void) } } out: - add_pda(irq_threshold_count, 1); + inc_irq_stat(irq_threshold_count); irq_exit(); } diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c index c17eaf5dd6dd..4b48f251fd39 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c @@ -26,7 +26,7 @@ asmlinkage void smp_thermal_interrupt(void) if (therm_throt_process(msr_val & 1)) mce_log_therm_throt_event(smp_processor_id(), msr_val); - add_pda(irq_thermal_count, 1); + inc_irq_stat(irq_thermal_count); irq_exit(); } diff --git a/arch/x86/kernel/cpu/mcheck/non-fatal.c b/arch/x86/kernel/cpu/mcheck/non-fatal.c index cc1fccdd31e0..a74af128efc9 100644 --- a/arch/x86/kernel/cpu/mcheck/non-fatal.c +++ b/arch/x86/kernel/cpu/mcheck/non-fatal.c @@ -1,7 +1,7 @@ /* * Non Fatal Machine Check Exception Reporting * - * (C) Copyright 2002 Dave Jones. <davej@codemonkey.org.uk> + * (C) Copyright 2002 Dave Jones. <davej@redhat.com> * * This file contains routines to check for non-fatal MCEs every 15s * diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index c78c04821ea1..1159e269e596 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -803,6 +803,7 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range, } static struct res_range __initdata range[RANGE_NUM]; +static int __initdata nr_range; #ifdef CONFIG_MTRR_SANITIZER @@ -1206,39 +1207,43 @@ struct mtrr_cleanup_result { #define PSHIFT (PAGE_SHIFT - 10) static struct mtrr_cleanup_result __initdata result[NUM_RESULT]; -static struct res_range __initdata range_new[RANGE_NUM]; static unsigned long __initdata min_loss_pfn[RANGE_NUM]; -static int __init mtrr_cleanup(unsigned address_bits) +static void __init print_out_mtrr_range_state(void) { - unsigned long extra_remove_base, extra_remove_size; - unsigned long base, size, def, dummy; - mtrr_type type; - int nr_range, nr_range_new; - u64 chunk_size, gran_size; - unsigned long range_sums, range_sums_new; - int index_good; - int num_reg_good; int i; + char start_factor = 'K', size_factor = 'K'; + unsigned long start_base, size_base; + mtrr_type type; - /* extra one for all 0 */ - int num[MTRR_NUM_TYPES + 1]; + for (i = 0; i < num_var_ranges; i++) { - if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1) - return 0; - rdmsr(MTRRdefType_MSR, def, dummy); - def &= 0xff; - if (def != MTRR_TYPE_UNCACHABLE) - return 0; + size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10); + if (!size_base) + continue; - /* get it and store it aside */ - memset(range_state, 0, sizeof(range_state)); - for (i = 0; i < num_var_ranges; i++) { - mtrr_if->get(i, &base, &size, &type); - range_state[i].base_pfn = base; - range_state[i].size_pfn = size; - range_state[i].type = type; + size_base = to_size_factor(size_base, &size_factor), + start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10); + start_base = to_size_factor(start_base, &start_factor), + type = range_state[i].type; + + printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n", + i, start_base, start_factor, + size_base, size_factor, + (type == MTRR_TYPE_UNCACHABLE) ? "UC" : + ((type == MTRR_TYPE_WRPROT) ? "WP" : + ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other")) + ); } +} + +static int __init mtrr_need_cleanup(void) +{ + int i; + mtrr_type type; + unsigned long size; + /* extra one for all 0 */ + int num[MTRR_NUM_TYPES + 1]; /* check entries number */ memset(num, 0, sizeof(num)); @@ -1263,29 +1268,133 @@ static int __init mtrr_cleanup(unsigned address_bits) num_var_ranges - num[MTRR_NUM_TYPES]) return 0; - /* print original var MTRRs at first, for debugging: */ - printk(KERN_DEBUG "original variable MTRRs\n"); - for (i = 0; i < num_var_ranges; i++) { - char start_factor = 'K', size_factor = 'K'; - unsigned long start_base, size_base; + return 1; +} - size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10); - if (!size_base) - continue; +static unsigned long __initdata range_sums; +static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size, + unsigned long extra_remove_base, + unsigned long extra_remove_size, + int i) +{ + int num_reg; + static struct res_range range_new[RANGE_NUM]; + static int nr_range_new; + unsigned long range_sums_new; + + /* convert ranges to var ranges state */ + num_reg = x86_setup_var_mtrrs(range, nr_range, + chunk_size, gran_size); + + /* we got new setting in range_state, check it */ + memset(range_new, 0, sizeof(range_new)); + nr_range_new = x86_get_mtrr_mem_range(range_new, 0, + extra_remove_base, extra_remove_size); + range_sums_new = sum_ranges(range_new, nr_range_new); + + result[i].chunk_sizek = chunk_size >> 10; + result[i].gran_sizek = gran_size >> 10; + result[i].num_reg = num_reg; + if (range_sums < range_sums_new) { + result[i].lose_cover_sizek = + (range_sums_new - range_sums) << PSHIFT; + result[i].bad = 1; + } else + result[i].lose_cover_sizek = + (range_sums - range_sums_new) << PSHIFT; - size_base = to_size_factor(size_base, &size_factor), - start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10); - start_base = to_size_factor(start_base, &start_factor), - type = range_state[i].type; + /* double check it */ + if (!result[i].bad && !result[i].lose_cover_sizek) { + if (nr_range_new != nr_range || + memcmp(range, range_new, sizeof(range))) + result[i].bad = 1; + } - printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n", - i, start_base, start_factor, - size_base, size_factor, - (type == MTRR_TYPE_UNCACHABLE) ? "UC" : - ((type == MTRR_TYPE_WRPROT) ? "WP" : - ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other")) - ); + if (!result[i].bad && (range_sums - range_sums_new < + min_loss_pfn[num_reg])) { + min_loss_pfn[num_reg] = + range_sums - range_sums_new; } +} + +static void __init mtrr_print_out_one_result(int i) +{ + char gran_factor, chunk_factor, lose_factor; + unsigned long gran_base, chunk_base, lose_base; + + gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), + chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), + lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), + printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", + result[i].bad ? "*BAD*" : " ", + gran_base, gran_factor, chunk_base, chunk_factor); + printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", + result[i].num_reg, result[i].bad ? "-" : "", + lose_base, lose_factor); +} + +static int __init mtrr_search_optimal_index(void) +{ + int i; + int num_reg_good; + int index_good; + + if (nr_mtrr_spare_reg >= num_var_ranges) + nr_mtrr_spare_reg = num_var_ranges - 1; + num_reg_good = -1; + for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) { + if (!min_loss_pfn[i]) + num_reg_good = i; + } + + index_good = -1; + if (num_reg_good != -1) { + for (i = 0; i < NUM_RESULT; i++) { + if (!result[i].bad && + result[i].num_reg == num_reg_good && + !result[i].lose_cover_sizek) { + index_good = i; + break; + } + } + } + + return index_good; +} + + +static int __init mtrr_cleanup(unsigned address_bits) +{ + unsigned long extra_remove_base, extra_remove_size; + unsigned long base, size, def, dummy; + mtrr_type type; + u64 chunk_size, gran_size; + int index_good; + int i; + + if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1) + return 0; + rdmsr(MTRRdefType_MSR, def, dummy); + def &= 0xff; + if (def != MTRR_TYPE_UNCACHABLE) + return 0; + + /* get it and store it aside */ + memset(range_state, 0, sizeof(range_state)); + for (i = 0; i < num_var_ranges; i++) { + mtrr_if->get(i, &base, &size, &type); + range_state[i].base_pfn = base; + range_state[i].size_pfn = size; + range_state[i].type = type; + } + + /* check if we need handle it and can handle it */ + if (!mtrr_need_cleanup()) + return 0; + + /* print original var MTRRs at first, for debugging: */ + printk(KERN_DEBUG "original variable MTRRs\n"); + print_out_mtrr_range_state(); memset(range, 0, sizeof(range)); extra_remove_size = 0; @@ -1309,176 +1418,64 @@ static int __init mtrr_cleanup(unsigned address_bits) range_sums >> (20 - PAGE_SHIFT)); if (mtrr_chunk_size && mtrr_gran_size) { - int num_reg; - char gran_factor, chunk_factor, lose_factor; - unsigned long gran_base, chunk_base, lose_base; - - debug_print++; - /* convert ranges to var ranges state */ - num_reg = x86_setup_var_mtrrs(range, nr_range, mtrr_chunk_size, - mtrr_gran_size); + i = 0; + mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size, + extra_remove_base, extra_remove_size, i); - /* we got new setting in range_state, check it */ - memset(range_new, 0, sizeof(range_new)); - nr_range_new = x86_get_mtrr_mem_range(range_new, 0, - extra_remove_base, - extra_remove_size); - range_sums_new = sum_ranges(range_new, nr_range_new); + mtrr_print_out_one_result(i); - i = 0; - result[i].chunk_sizek = mtrr_chunk_size >> 10; - result[i].gran_sizek = mtrr_gran_size >> 10; - result[i].num_reg = num_reg; - if (range_sums < range_sums_new) { - result[i].lose_cover_sizek = - (range_sums_new - range_sums) << PSHIFT; - result[i].bad = 1; - } else - result[i].lose_cover_sizek = - (range_sums - range_sums_new) << PSHIFT; - - gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), - chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), - lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), - printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", - result[i].bad?"*BAD*":" ", - gran_base, gran_factor, chunk_base, chunk_factor); - printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", - result[i].num_reg, result[i].bad?"-":"", - lose_base, lose_factor); if (!result[i].bad) { set_var_mtrr_all(address_bits); return 1; } printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, " "will find optimal one\n"); - debug_print--; - memset(result, 0, sizeof(result[0])); } i = 0; memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn)); memset(result, 0, sizeof(result)); for (gran_size = (1ULL<<16); gran_size < (1ULL<<32); gran_size <<= 1) { - char gran_factor; - unsigned long gran_base; - - if (debug_print) - gran_base = to_size_factor(gran_size >> 10, &gran_factor); for (chunk_size = gran_size; chunk_size < (1ULL<<32); chunk_size <<= 1) { - int num_reg; - if (debug_print) { - char chunk_factor; - unsigned long chunk_base; - - chunk_base = to_size_factor(chunk_size>>10, &chunk_factor), - printk(KERN_INFO "\n"); - printk(KERN_INFO "gran_size: %ld%c chunk_size: %ld%c \n", - gran_base, gran_factor, chunk_base, chunk_factor); - } if (i >= NUM_RESULT) continue; - /* convert ranges to var ranges state */ - num_reg = x86_setup_var_mtrrs(range, nr_range, - chunk_size, gran_size); - - /* we got new setting in range_state, check it */ - memset(range_new, 0, sizeof(range_new)); - nr_range_new = x86_get_mtrr_mem_range(range_new, 0, - extra_remove_base, extra_remove_size); - range_sums_new = sum_ranges(range_new, nr_range_new); - - result[i].chunk_sizek = chunk_size >> 10; - result[i].gran_sizek = gran_size >> 10; - result[i].num_reg = num_reg; - if (range_sums < range_sums_new) { - result[i].lose_cover_sizek = - (range_sums_new - range_sums) << PSHIFT; - result[i].bad = 1; - } else - result[i].lose_cover_sizek = - (range_sums - range_sums_new) << PSHIFT; - - /* double check it */ - if (!result[i].bad && !result[i].lose_cover_sizek) { - if (nr_range_new != nr_range || - memcmp(range, range_new, sizeof(range))) - result[i].bad = 1; + mtrr_calc_range_state(chunk_size, gran_size, + extra_remove_base, extra_remove_size, i); + if (debug_print) { + mtrr_print_out_one_result(i); + printk(KERN_INFO "\n"); } - if (!result[i].bad && (range_sums - range_sums_new < - min_loss_pfn[num_reg])) { - min_loss_pfn[num_reg] = - range_sums - range_sums_new; - } i++; } } - /* print out all */ - for (i = 0; i < NUM_RESULT; i++) { - char gran_factor, chunk_factor, lose_factor; - unsigned long gran_base, chunk_base, lose_base; - - gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), - chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), - lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), - printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", - result[i].bad?"*BAD*":" ", - gran_base, gran_factor, chunk_base, chunk_factor); - printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", - result[i].num_reg, result[i].bad?"-":"", - lose_base, lose_factor); - } - /* try to find the optimal index */ - if (nr_mtrr_spare_reg >= num_var_ranges) - nr_mtrr_spare_reg = num_var_ranges - 1; - num_reg_good = -1; - for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) { - if (!min_loss_pfn[i]) - num_reg_good = i; - } - - index_good = -1; - if (num_reg_good != -1) { - for (i = 0; i < NUM_RESULT; i++) { - if (!result[i].bad && - result[i].num_reg == num_reg_good && - !result[i].lose_cover_sizek) { - index_good = i; - break; - } - } - } + index_good = mtrr_search_optimal_index(); if (index_good != -1) { - char gran_factor, chunk_factor, lose_factor; - unsigned long gran_base, chunk_base, lose_base; - printk(KERN_INFO "Found optimal setting for mtrr clean up\n"); i = index_good; - gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), - chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), - lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), - printk(KERN_INFO "gran_size: %ld%c \tchunk_size: %ld%c \t", - gran_base, gran_factor, chunk_base, chunk_factor); - printk(KERN_CONT "num_reg: %d \tlose RAM: %ld%c\n", - result[i].num_reg, lose_base, lose_factor); + mtrr_print_out_one_result(i); + /* convert ranges to var ranges state */ chunk_size = result[i].chunk_sizek; chunk_size <<= 10; gran_size = result[i].gran_sizek; gran_size <<= 10; - debug_print++; x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size); - debug_print--; set_var_mtrr_all(address_bits); + printk(KERN_DEBUG "New variable MTRRs\n"); + print_out_mtrr_range_state(); return 1; + } else { + /* print out all */ + for (i = 0; i < NUM_RESULT; i++) + mtrr_print_out_one_result(i); } printk(KERN_INFO "mtrr_cleanup: can not find optimal value\n"); @@ -1562,7 +1559,6 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) { unsigned long i, base, size, highest_pfn = 0, def, dummy; mtrr_type type; - int nr_range; u64 total_trim_size; /* extra one for all 0 */ diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index 6bff382094f5..9abd48b22674 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -17,6 +17,8 @@ #include <linux/bitops.h> #include <linux/smp.h> #include <linux/nmi.h> +#include <linux/kprobes.h> + #include <asm/apic.h> #include <asm/intel_arch_perfmon.h> @@ -336,7 +338,8 @@ static void single_msr_unreserve(void) release_perfctr_nmi(wd_ops->perfctr); } -static void single_msr_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) +static void __kprobes +single_msr_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) { /* start the cycle over again */ write_watchdog_counter(wd->perfctr_msr, NULL, nmi_hz); @@ -401,7 +404,7 @@ static int setup_p6_watchdog(unsigned nmi_hz) return 1; } -static void p6_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) +static void __kprobes p6_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) { /* * P6 based Pentium M need to re-unmask @@ -605,7 +608,7 @@ static void p4_unreserve(void) release_perfctr_nmi(MSR_P4_IQ_PERFCTR0); } -static void p4_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) +static void __kprobes p4_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) { unsigned dummy; /* @@ -784,7 +787,7 @@ unsigned lapic_adjust_nmi_hz(unsigned hz) return hz; } -int lapic_wd_event(unsigned nmi_hz) +int __kprobes lapic_wd_event(unsigned nmi_hz) { struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); u64 ctr; diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c index a26c480b9491..01b1244ef1c0 100644 --- a/arch/x86/kernel/cpu/proc.c +++ b/arch/x86/kernel/cpu/proc.c @@ -160,14 +160,16 @@ static void *c_start(struct seq_file *m, loff_t *pos) { if (*pos == 0) /* just in case, cpu 0 is not the first */ *pos = first_cpu(cpu_online_map); - if ((*pos) < nr_cpu_ids && cpu_online(*pos)) + else + *pos = next_cpu_nr(*pos - 1, cpu_online_map); + if ((*pos) < nr_cpu_ids) return &cpu_data(*pos); return NULL; } static void *c_next(struct seq_file *m, void *v, loff_t *pos) { - *pos = next_cpu(*pos, cpu_online_map); + (*pos)++; return c_start(m, pos); } diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c new file mode 100644 index 000000000000..284c399e3234 --- /dev/null +++ b/arch/x86/kernel/cpu/vmware.c @@ -0,0 +1,112 @@ +/* + * VMware Detection code. + * + * Copyright (C) 2008, VMware, Inc. + * Author : Alok N Kataria <akataria@vmware.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. + * + */ + +#include <linux/dmi.h> +#include <asm/div64.h> +#include <asm/vmware.h> + +#define CPUID_VMWARE_INFO_LEAF 0x40000000 +#define VMWARE_HYPERVISOR_MAGIC 0x564D5868 +#define VMWARE_HYPERVISOR_PORT 0x5658 + +#define VMWARE_PORT_CMD_GETVERSION 10 +#define VMWARE_PORT_CMD_GETHZ 45 + +#define VMWARE_PORT(cmd, eax, ebx, ecx, edx) \ + __asm__("inl (%%dx)" : \ + "=a"(eax), "=c"(ecx), "=d"(edx), "=b"(ebx) : \ + "0"(VMWARE_HYPERVISOR_MAGIC), \ + "1"(VMWARE_PORT_CMD_##cmd), \ + "2"(VMWARE_HYPERVISOR_PORT), "3"(UINT_MAX) : \ + "memory"); + +static inline int __vmware_platform(void) +{ + uint32_t eax, ebx, ecx, edx; + VMWARE_PORT(GETVERSION, eax, ebx, ecx, edx); + return eax != (uint32_t)-1 && ebx == VMWARE_HYPERVISOR_MAGIC; +} + +static unsigned long __vmware_get_tsc_khz(void) +{ + uint64_t tsc_hz; + uint32_t eax, ebx, ecx, edx; + + VMWARE_PORT(GETHZ, eax, ebx, ecx, edx); + + if (ebx == UINT_MAX) + return 0; + tsc_hz = eax | (((uint64_t)ebx) << 32); + do_div(tsc_hz, 1000); + BUG_ON(tsc_hz >> 32); + return tsc_hz; +} + +/* + * While checking the dmi string infomation, just checking the product + * serial key should be enough, as this will always have a VMware + * specific string when running under VMware hypervisor. + */ +int vmware_platform(void) +{ + if (cpu_has_hypervisor) { + unsigned int eax, ebx, ecx, edx; + char hyper_vendor_id[13]; + + cpuid(CPUID_VMWARE_INFO_LEAF, &eax, &ebx, &ecx, &edx); + memcpy(hyper_vendor_id + 0, &ebx, 4); + memcpy(hyper_vendor_id + 4, &ecx, 4); + memcpy(hyper_vendor_id + 8, &edx, 4); + hyper_vendor_id[12] = '\0'; + if (!strcmp(hyper_vendor_id, "VMwareVMware")) + return 1; + } else if (dmi_available && dmi_name_in_serial("VMware") && + __vmware_platform()) + return 1; + + return 0; +} + +unsigned long vmware_get_tsc_khz(void) +{ + BUG_ON(!vmware_platform()); + return __vmware_get_tsc_khz(); +} + +/* + * VMware hypervisor takes care of exporting a reliable TSC to the guest. + * Still, due to timing difference when running on virtual cpus, the TSC can + * be marked as unstable in some cases. For example, the TSC sync check at + * bootup can fail due to a marginal offset between vcpus' TSCs (though the + * TSCs do not drift from each other). Also, the ACPI PM timer clocksource + * is not suitable as a watchdog when running on a hypervisor because the + * kernel may miss a wrap of the counter if the vcpu is descheduled for a + * long time. To skip these checks at runtime we set these capability bits, + * so that the kernel could just trust the hypervisor with providing a + * reliable virtual TSC that is suitable for timekeeping. + */ +void __cpuinit vmware_set_feature_bits(struct cpuinfo_x86 *c) +{ + set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); + set_cpu_cap(c, X86_FEATURE_TSC_RELIABLE); +} diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c index 6a44d6465991..72cefd1e649b 100644 --- a/arch/x86/kernel/cpuid.c +++ b/arch/x86/kernel/cpuid.c @@ -147,8 +147,8 @@ static __cpuinit int cpuid_device_create(int cpu) { struct device *dev; - dev = device_create_drvdata(cpuid_class, NULL, MKDEV(CPUID_MAJOR, cpu), - NULL, "cpu%d", cpu); + dev = device_create(cpuid_class, NULL, MKDEV(CPUID_MAJOR, cpu), NULL, + "cpu%d", cpu); return IS_ERR(dev) ? PTR_ERR(dev) : 0; } diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index 268553817909..d84a852e4cd7 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -29,34 +29,17 @@ #include <mach_ipi.h> -/* This keeps a track of which one is crashing cpu. */ -static int crashing_cpu; #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) -static atomic_t waiting_for_crash_ipi; -static int crash_nmi_callback(struct notifier_block *self, - unsigned long val, void *data) +static void kdump_nmi_callback(int cpu, struct die_args *args) { struct pt_regs *regs; #ifdef CONFIG_X86_32 struct pt_regs fixed_regs; #endif - int cpu; - if (val != DIE_NMI_IPI) - return NOTIFY_OK; - - regs = ((struct die_args *)data)->regs; - cpu = raw_smp_processor_id(); - - /* Don't do anything if this handler is invoked on crashing cpu. - * Otherwise, system will completely hang. Crashing cpu can get - * an NMI if system was initially booted with nmi_watchdog parameter. - */ - if (cpu == crashing_cpu) - return NOTIFY_STOP; - local_irq_disable(); + regs = args->regs; #ifdef CONFIG_X86_32 if (!user_mode_vm(regs)) { @@ -65,54 +48,19 @@ static int crash_nmi_callback(struct notifier_block *self, } #endif crash_save_cpu(regs, cpu); - disable_local_APIC(); - atomic_dec(&waiting_for_crash_ipi); - /* Assume hlt works */ - halt(); - for (;;) - cpu_relax(); - - return 1; -} -static void smp_send_nmi_allbutself(void) -{ - cpumask_t mask = cpu_online_map; - cpu_clear(safe_smp_processor_id(), mask); - if (!cpus_empty(mask)) - send_IPI_mask(mask, NMI_VECTOR); + disable_local_APIC(); } -static struct notifier_block crash_nmi_nb = { - .notifier_call = crash_nmi_callback, -}; - -static void nmi_shootdown_cpus(void) +static void kdump_nmi_shootdown_cpus(void) { - unsigned long msecs; - - atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1); - /* Would it be better to replace the trap vector here? */ - if (register_die_notifier(&crash_nmi_nb)) - return; /* return what? */ - /* Ensure the new callback function is set before sending - * out the NMI - */ - wmb(); + nmi_shootdown_cpus(kdump_nmi_callback); - smp_send_nmi_allbutself(); - - msecs = 1000; /* Wait at most a second for the other cpus to stop */ - while ((atomic_read(&waiting_for_crash_ipi) > 0) && msecs) { - mdelay(1); - msecs--; - } - - /* Leave the nmi callback set */ disable_local_APIC(); } + #else -static void nmi_shootdown_cpus(void) +static void kdump_nmi_shootdown_cpus(void) { /* There are no cpus to shootdown */ } @@ -131,9 +79,7 @@ void native_machine_crash_shutdown(struct pt_regs *regs) /* The kernel is broken so disable interrupts */ local_irq_disable(); - /* Make a note of crashing cpu. Will be used in NMI callback.*/ - crashing_cpu = safe_smp_processor_id(); - nmi_shootdown_cpus(); + kdump_nmi_shootdown_cpus(); lapic_shutdown(); #if defined(CONFIG_X86_IO_APIC) disable_IO_APIC(); diff --git a/arch/x86/kernel/crash_dump_32.c b/arch/x86/kernel/crash_dump_32.c index 72d0c56c1b48..f7cdb3b457aa 100644 --- a/arch/x86/kernel/crash_dump_32.c +++ b/arch/x86/kernel/crash_dump_32.c @@ -13,6 +13,9 @@ static void *kdump_buf_page; +/* Stores the physical address of elf header of crash image. */ +unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX; + /** * copy_oldmem_page - copy one page from "oldmem" * @pfn: page frame number to be copied diff --git a/arch/x86/kernel/crash_dump_64.c b/arch/x86/kernel/crash_dump_64.c index e90a60ef10c2..045b36cada65 100644 --- a/arch/x86/kernel/crash_dump_64.c +++ b/arch/x86/kernel/crash_dump_64.c @@ -10,6 +10,9 @@ #include <linux/uaccess.h> #include <linux/io.h> +/* Stores the physical address of elf header of crash image. */ +unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX; + /** * copy_oldmem_page - copy one page from "oldmem" * @pfn: page frame number to be copied diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index 2b69994fd3a8..da91701a2348 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -6,14 +6,13 @@ * precise-event based sampling (PEBS). * * It manages: - * - per-thread and per-cpu allocation of BTS and PEBS - * - buffer memory allocation (optional) - * - buffer overflow handling + * - DS and BTS hardware configuration + * - buffer overflow handling (to be done) * - buffer access * - * It assumes: - * - get_task_struct on all parameter tasks - * - current is allowed to trace parameter tasks + * It does not do: + * - security checking (is the caller allowed to trace the task) + * - buffer allocation (memory accounting) * * * Copyright (C) 2007-2008 Intel Corporation. @@ -21,8 +20,6 @@ */ -#ifdef CONFIG_X86_DS - #include <asm/ds.h> #include <linux/errno.h> @@ -30,22 +27,69 @@ #include <linux/slab.h> #include <linux/sched.h> #include <linux/mm.h> +#include <linux/kernel.h> /* * The configuration for a particular DS hardware implementation. */ struct ds_configuration { - /* the size of the DS structure in bytes */ - unsigned char sizeof_ds; - /* the size of one pointer-typed field in the DS structure in bytes; - this covers the first 8 fields related to buffer management. */ + /* the name of the configuration */ + const char *name; + /* the size of one pointer-typed field in the DS structure and + in the BTS and PEBS buffers in bytes; + this covers the first 8 DS fields related to buffer management. */ unsigned char sizeof_field; /* the size of a BTS/PEBS record in bytes */ unsigned char sizeof_rec[2]; + /* a series of bit-masks to control various features indexed + * by enum ds_feature */ + unsigned long ctl[dsf_ctl_max]; +}; +static DEFINE_PER_CPU(struct ds_configuration, ds_cfg_array); + +#define ds_cfg per_cpu(ds_cfg_array, smp_processor_id()) + +#define MAX_SIZEOF_DS (12 * 8) /* maximal size of a DS configuration */ +#define MAX_SIZEOF_BTS (3 * 8) /* maximal size of a BTS record */ +#define DS_ALIGNMENT (1 << 3) /* BTS and PEBS buffer alignment */ + +#define BTS_CONTROL \ + (ds_cfg.ctl[dsf_bts] | ds_cfg.ctl[dsf_bts_kernel] | ds_cfg.ctl[dsf_bts_user] |\ + ds_cfg.ctl[dsf_bts_overflow]) + + +/* + * A BTS or PEBS tracer. + * + * This holds the configuration of the tracer and serves as a handle + * to identify tracers. + */ +struct ds_tracer { + /* the DS context (partially) owned by this tracer */ + struct ds_context *context; + /* the buffer provided on ds_request() and its size in bytes */ + void *buffer; + size_t size; +}; + +struct bts_tracer { + /* the common DS part */ + struct ds_tracer ds; + /* the trace including the DS configuration */ + struct bts_trace trace; + /* buffer overflow notification function */ + bts_ovfl_callback_t ovfl; }; -static struct ds_configuration ds_cfg; +struct pebs_tracer { + /* the common DS part */ + struct ds_tracer ds; + /* the trace including the DS configuration */ + struct pebs_trace trace; + /* buffer overflow notification function */ + pebs_ovfl_callback_t ovfl; +}; /* * Debug Store (DS) save area configuration (see Intel64 and IA32 @@ -111,32 +155,9 @@ static inline void ds_set(unsigned char *base, enum ds_qualifier qual, /* - * Locking is done only for allocating BTS or PEBS resources and for - * guarding context and buffer memory allocation. - * - * Most functions require the current task to own the ds context part - * they are going to access. All the locking is done when validating - * access to the context. + * Locking is done only for allocating BTS or PEBS resources. */ -static spinlock_t ds_lock = __SPIN_LOCK_UNLOCKED(ds_lock); - -/* - * Validate that the current task is allowed to access the BTS/PEBS - * buffer of the parameter task. - * - * Returns 0, if access is granted; -Eerrno, otherwise. - */ -static inline int ds_validate_access(struct ds_context *context, - enum ds_qualifier qual) -{ - if (!context) - return -EPERM; - - if (context->owner[qual] == current) - return 0; - - return -EPERM; -} +static DEFINE_SPINLOCK(ds_lock); /* @@ -152,27 +173,32 @@ static inline int ds_validate_access(struct ds_context *context, * >0 number of per-thread tracers * <0 number of per-cpu tracers * - * The below functions to get and put tracers and to check the - * allocation type require the ds_lock to be held by the caller. - * * Tracers essentially gives the number of ds contexts for a certain * type of allocation. */ -static long tracers; +static atomic_t tracers = ATOMIC_INIT(0); static inline void get_tracer(struct task_struct *task) { - tracers += (task ? 1 : -1); + if (task) + atomic_inc(&tracers); + else + atomic_dec(&tracers); } static inline void put_tracer(struct task_struct *task) { - tracers -= (task ? 1 : -1); + if (task) + atomic_dec(&tracers); + else + atomic_inc(&tracers); } static inline int check_tracer(struct task_struct *task) { - return (task ? (tracers >= 0) : (tracers <= 0)); + return task ? + (atomic_read(&tracers) >= 0) : + (atomic_read(&tracers) <= 0); } @@ -185,100 +211,83 @@ static inline int check_tracer(struct task_struct *task) * * Contexts are use-counted. They are allocated on first access and * deallocated when the last user puts the context. - * - * We distinguish between an allocating and a non-allocating get of a - * context: - * - the allocating get is used for requesting BTS/PEBS resources. It - * requires the caller to hold the global ds_lock. - * - the non-allocating get is used for all other cases. A - * non-existing context indicates an error. It acquires and releases - * the ds_lock itself for obtaining the context. - * - * A context and its DS configuration are allocated and deallocated - * together. A context always has a DS configuration of the - * appropriate size. - */ -static DEFINE_PER_CPU(struct ds_context *, system_context); - -#define this_system_context per_cpu(system_context, smp_processor_id()) - -/* - * Returns the pointer to the parameter task's context or to the - * system-wide context, if task is NULL. - * - * Increases the use count of the returned context, if not NULL. */ -static inline struct ds_context *ds_get_context(struct task_struct *task) -{ - struct ds_context *context; - - spin_lock(&ds_lock); +struct ds_context { + /* pointer to the DS configuration; goes into MSR_IA32_DS_AREA */ + unsigned char ds[MAX_SIZEOF_DS]; + /* the owner of the BTS and PEBS configuration, respectively */ + struct bts_tracer *bts_master; + struct pebs_tracer *pebs_master; + /* use count */ + unsigned long count; + /* a pointer to the context location inside the thread_struct + * or the per_cpu context array */ + struct ds_context **this; + /* a pointer to the task owning this context, or NULL, if the + * context is owned by a cpu */ + struct task_struct *task; +}; - context = (task ? task->thread.ds_ctx : this_system_context); - if (context) - context->count++; +static DEFINE_PER_CPU(struct ds_context *, system_context_array); - spin_unlock(&ds_lock); +#define system_context per_cpu(system_context_array, smp_processor_id()) - return context; -} -/* - * Same as ds_get_context, but allocates the context and it's DS - * structure, if necessary; returns NULL; if out of memory. - * - * pre: requires ds_lock to be held - */ -static inline struct ds_context *ds_alloc_context(struct task_struct *task) +static inline struct ds_context *ds_get_context(struct task_struct *task) { struct ds_context **p_context = - (task ? &task->thread.ds_ctx : &this_system_context); - struct ds_context *context = *p_context; - - if (!context) { - context = kzalloc(sizeof(*context), GFP_KERNEL); - - if (!context) - return NULL; + (task ? &task->thread.ds_ctx : &system_context); + struct ds_context *context = NULL; + struct ds_context *new_context = NULL; + unsigned long irq; + + /* Chances are small that we already have a context. */ + new_context = kzalloc(sizeof(*new_context), GFP_KERNEL); + if (!new_context) + return NULL; - context->ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL); - if (!context->ds) { - kfree(context); - return NULL; - } + spin_lock_irqsave(&ds_lock, irq); - *p_context = context; + context = *p_context; + if (!context) { + context = new_context; context->this = p_context; context->task = task; + context->count = 0; if (task) set_tsk_thread_flag(task, TIF_DS_AREA_MSR); if (!task || (task == current)) - wrmsr(MSR_IA32_DS_AREA, (unsigned long)context->ds, 0); + wrmsrl(MSR_IA32_DS_AREA, (unsigned long)context->ds); - get_tracer(task); + *p_context = context; } context->count++; + spin_unlock_irqrestore(&ds_lock, irq); + + if (context != new_context) + kfree(new_context); + return context; } -/* - * Decreases the use count of the parameter context, if not NULL. - * Deallocates the context, if the use count reaches zero. - */ static inline void ds_put_context(struct ds_context *context) { + unsigned long irq; + if (!context) return; - spin_lock(&ds_lock); + spin_lock_irqsave(&ds_lock, irq); - if (--context->count) - goto out; + if (--context->count) { + spin_unlock_irqrestore(&ds_lock, irq); + return; + } *(context->this) = NULL; @@ -288,132 +297,263 @@ static inline void ds_put_context(struct ds_context *context) if (!context->task || (context->task == current)) wrmsrl(MSR_IA32_DS_AREA, 0); - put_tracer(context->task); + spin_unlock_irqrestore(&ds_lock, irq); - /* free any leftover buffers from tracers that did not - * deallocate them properly. */ - kfree(context->buffer[ds_bts]); - kfree(context->buffer[ds_pebs]); - kfree(context->ds); kfree(context); - out: - spin_unlock(&ds_lock); } /* - * Handle a buffer overflow + * Call the tracer's callback on a buffer overflow. * - * task: the task whose buffers are overflowing; - * NULL for a buffer overflow on the current cpu * context: the ds context * qual: the buffer type */ -static void ds_overflow(struct task_struct *task, struct ds_context *context, - enum ds_qualifier qual) +static void ds_overflow(struct ds_context *context, enum ds_qualifier qual) { - if (!context) - return; - - if (context->callback[qual]) - (*context->callback[qual])(task); - - /* todo: do some more overflow handling */ + switch (qual) { + case ds_bts: + if (context->bts_master && + context->bts_master->ovfl) + context->bts_master->ovfl(context->bts_master); + break; + case ds_pebs: + if (context->pebs_master && + context->pebs_master->ovfl) + context->pebs_master->ovfl(context->pebs_master); + break; + } } /* - * Allocate a non-pageable buffer of the parameter size. - * Checks the memory and the locked memory rlimit. + * Write raw data into the BTS or PEBS buffer. * - * Returns the buffer, if successful; - * NULL, if out of memory or rlimit exceeded. + * The remainder of any partially written record is zeroed out. * - * size: the requested buffer size in bytes - * pages (out): if not NULL, contains the number of pages reserved + * context: the DS context + * qual: the buffer type + * record: the data to write + * size: the size of the data */ -static inline void *ds_allocate_buffer(size_t size, unsigned int *pages) +static int ds_write(struct ds_context *context, enum ds_qualifier qual, + const void *record, size_t size) { - unsigned long rlim, vm, pgsz; - void *buffer; + int bytes_written = 0; - pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT; + if (!record) + return -EINVAL; - rlim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT; - vm = current->mm->total_vm + pgsz; - if (rlim < vm) - return NULL; + while (size) { + unsigned long base, index, end, write_end, int_th; + unsigned long write_size, adj_write_size; - rlim = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT; - vm = current->mm->locked_vm + pgsz; - if (rlim < vm) - return NULL; + /* + * write as much as possible without producing an + * overflow interrupt. + * + * interrupt_threshold must either be + * - bigger than absolute_maximum or + * - point to a record between buffer_base and absolute_maximum + * + * index points to a valid record. + */ + base = ds_get(context->ds, qual, ds_buffer_base); + index = ds_get(context->ds, qual, ds_index); + end = ds_get(context->ds, qual, ds_absolute_maximum); + int_th = ds_get(context->ds, qual, ds_interrupt_threshold); - buffer = kzalloc(size, GFP_KERNEL); - if (!buffer) - return NULL; + write_end = min(end, int_th); + + /* if we are already beyond the interrupt threshold, + * we fill the entire buffer */ + if (write_end <= index) + write_end = end; + + if (write_end <= index) + break; + + write_size = min((unsigned long) size, write_end - index); + memcpy((void *)index, record, write_size); + + record = (const char *)record + write_size; + size -= write_size; + bytes_written += write_size; + + adj_write_size = write_size / ds_cfg.sizeof_rec[qual]; + adj_write_size *= ds_cfg.sizeof_rec[qual]; + + /* zero out trailing bytes */ + memset((char *)index + write_size, 0, + adj_write_size - write_size); + index += adj_write_size; + + if (index >= end) + index = base; + ds_set(context->ds, qual, ds_index, index); + + if (index >= int_th) + ds_overflow(context, qual); + } + + return bytes_written; +} + + +/* + * Branch Trace Store (BTS) uses the following format. Different + * architectures vary in the size of those fields. + * - source linear address + * - destination linear address + * - flags + * + * Later architectures use 64bit pointers throughout, whereas earlier + * architectures use 32bit pointers in 32bit mode. + * + * We compute the base address for the first 8 fields based on: + * - the field size stored in the DS configuration + * - the relative field position + * + * In order to store additional information in the BTS buffer, we use + * a special source address to indicate that the record requires + * special interpretation. + * + * Netburst indicated via a bit in the flags field whether the branch + * was predicted; this is ignored. + * + * We use two levels of abstraction: + * - the raw data level defined here + * - an arch-independent level defined in ds.h + */ + +enum bts_field { + bts_from, + bts_to, + bts_flags, - current->mm->total_vm += pgsz; - current->mm->locked_vm += pgsz; + bts_qual = bts_from, + bts_jiffies = bts_to, + bts_pid = bts_flags, - if (pages) - *pages = pgsz; + bts_qual_mask = (bts_qual_max - 1), + bts_escape = ((unsigned long)-1 & ~bts_qual_mask) +}; - return buffer; +static inline unsigned long bts_get(const char *base, enum bts_field field) +{ + base += (ds_cfg.sizeof_field * field); + return *(unsigned long *)base; } -static int ds_request(struct task_struct *task, void *base, size_t size, - ds_ovfl_callback_t ovfl, enum ds_qualifier qual) +static inline void bts_set(char *base, enum bts_field field, unsigned long val) { - struct ds_context *context; - unsigned long buffer, adj; - const unsigned long alignment = (1 << 3); - int error = 0; + base += (ds_cfg.sizeof_field * field);; + (*(unsigned long *)base) = val; +} - if (!ds_cfg.sizeof_ds) - return -EOPNOTSUPP; - /* we require some space to do alignment adjustments below */ - if (size < (alignment + ds_cfg.sizeof_rec[qual])) +/* + * The raw BTS data is architecture dependent. + * + * For higher-level users, we give an arch-independent view. + * - ds.h defines struct bts_struct + * - bts_read translates one raw bts record into a bts_struct + * - bts_write translates one bts_struct into the raw format and + * writes it into the top of the parameter tracer's buffer. + * + * return: bytes read/written on success; -Eerrno, otherwise + */ +static int bts_read(struct bts_tracer *tracer, const void *at, + struct bts_struct *out) +{ + if (!tracer) return -EINVAL; - /* buffer overflow notification is not yet implemented */ - if (ovfl) - return -EOPNOTSUPP; + if (at < tracer->trace.ds.begin) + return -EINVAL; + if (tracer->trace.ds.end < (at + tracer->trace.ds.size)) + return -EINVAL; - spin_lock(&ds_lock); + memset(out, 0, sizeof(*out)); + if ((bts_get(at, bts_qual) & ~bts_qual_mask) == bts_escape) { + out->qualifier = (bts_get(at, bts_qual) & bts_qual_mask); + out->variant.timestamp.jiffies = bts_get(at, bts_jiffies); + out->variant.timestamp.pid = bts_get(at, bts_pid); + } else { + out->qualifier = bts_branch; + out->variant.lbr.from = bts_get(at, bts_from); + out->variant.lbr.to = bts_get(at, bts_to); + + if (!out->variant.lbr.from && !out->variant.lbr.to) + out->qualifier = bts_invalid; + } - if (!check_tracer(task)) - return -EPERM; + return ds_cfg.sizeof_rec[ds_bts]; +} - error = -ENOMEM; - context = ds_alloc_context(task); - if (!context) - goto out_unlock; +static int bts_write(struct bts_tracer *tracer, const struct bts_struct *in) +{ + unsigned char raw[MAX_SIZEOF_BTS]; - error = -EALREADY; - if (context->owner[qual] == current) - goto out_unlock; - error = -EPERM; - if (context->owner[qual] != NULL) - goto out_unlock; - context->owner[qual] = current; + if (!tracer) + return -EINVAL; + + if (MAX_SIZEOF_BTS < ds_cfg.sizeof_rec[ds_bts]) + return -EOVERFLOW; + + switch (in->qualifier) { + case bts_invalid: + bts_set(raw, bts_from, 0); + bts_set(raw, bts_to, 0); + bts_set(raw, bts_flags, 0); + break; + case bts_branch: + bts_set(raw, bts_from, in->variant.lbr.from); + bts_set(raw, bts_to, in->variant.lbr.to); + bts_set(raw, bts_flags, 0); + break; + case bts_task_arrives: + case bts_task_departs: + bts_set(raw, bts_qual, (bts_escape | in->qualifier)); + bts_set(raw, bts_jiffies, in->variant.timestamp.jiffies); + bts_set(raw, bts_pid, in->variant.timestamp.pid); + break; + default: + return -EINVAL; + } - spin_unlock(&ds_lock); + return ds_write(tracer->ds.context, ds_bts, raw, + ds_cfg.sizeof_rec[ds_bts]); +} - error = -ENOMEM; - if (!base) { - base = ds_allocate_buffer(size, &context->pages[qual]); - if (!base) - goto out_release; +static void ds_write_config(struct ds_context *context, + struct ds_trace *cfg, enum ds_qualifier qual) +{ + unsigned char *ds = context->ds; - context->buffer[qual] = base; - } - error = 0; + ds_set(ds, qual, ds_buffer_base, (unsigned long)cfg->begin); + ds_set(ds, qual, ds_index, (unsigned long)cfg->top); + ds_set(ds, qual, ds_absolute_maximum, (unsigned long)cfg->end); + ds_set(ds, qual, ds_interrupt_threshold, (unsigned long)cfg->ith); +} - context->callback[qual] = ovfl; +static void ds_read_config(struct ds_context *context, + struct ds_trace *cfg, enum ds_qualifier qual) +{ + unsigned char *ds = context->ds; + + cfg->begin = (void *)ds_get(ds, qual, ds_buffer_base); + cfg->top = (void *)ds_get(ds, qual, ds_index); + cfg->end = (void *)ds_get(ds, qual, ds_absolute_maximum); + cfg->ith = (void *)ds_get(ds, qual, ds_interrupt_threshold); +} + +static void ds_init_ds_trace(struct ds_trace *trace, enum ds_qualifier qual, + void *base, size_t size, size_t ith, + unsigned int flags) { + unsigned long buffer, adj; /* adjust the buffer address and size to meet alignment * constraints: @@ -425,395 +565,383 @@ static int ds_request(struct task_struct *task, void *base, size_t size, */ buffer = (unsigned long)base; - adj = ALIGN(buffer, alignment) - buffer; + adj = ALIGN(buffer, DS_ALIGNMENT) - buffer; buffer += adj; size -= adj; - size /= ds_cfg.sizeof_rec[qual]; - size *= ds_cfg.sizeof_rec[qual]; - - ds_set(context->ds, qual, ds_buffer_base, buffer); - ds_set(context->ds, qual, ds_index, buffer); - ds_set(context->ds, qual, ds_absolute_maximum, buffer + size); + trace->n = size / ds_cfg.sizeof_rec[qual]; + trace->size = ds_cfg.sizeof_rec[qual]; - if (ovfl) { - /* todo: select a suitable interrupt threshold */ - } else - ds_set(context->ds, qual, - ds_interrupt_threshold, buffer + size + 1); + size = (trace->n * trace->size); - /* we keep the context until ds_release */ - return error; - - out_release: - context->owner[qual] = NULL; - ds_put_context(context); - return error; + trace->begin = (void *)buffer; + trace->top = trace->begin; + trace->end = (void *)(buffer + size); + /* The value for 'no threshold' is -1, which will set the + * threshold outside of the buffer, just like we want it. + */ + trace->ith = (void *)(buffer + size - ith); - out_unlock: - spin_unlock(&ds_lock); - ds_put_context(context); - return error; + trace->flags = flags; } -int ds_request_bts(struct task_struct *task, void *base, size_t size, - ds_ovfl_callback_t ovfl) -{ - return ds_request(task, base, size, ovfl, ds_bts); -} -int ds_request_pebs(struct task_struct *task, void *base, size_t size, - ds_ovfl_callback_t ovfl) -{ - return ds_request(task, base, size, ovfl, ds_pebs); -} - -static int ds_release(struct task_struct *task, enum ds_qualifier qual) +static int ds_request(struct ds_tracer *tracer, struct ds_trace *trace, + enum ds_qualifier qual, struct task_struct *task, + void *base, size_t size, size_t th, unsigned int flags) { struct ds_context *context; int error; - context = ds_get_context(task); - error = ds_validate_access(context, qual); - if (error < 0) + error = -EINVAL; + if (!base) goto out; - kfree(context->buffer[qual]); - context->buffer[qual] = NULL; - - current->mm->total_vm -= context->pages[qual]; - current->mm->locked_vm -= context->pages[qual]; - context->pages[qual] = 0; - context->owner[qual] = NULL; - - /* - * we put the context twice: - * once for the ds_get_context - * once for the corresponding ds_request - */ - ds_put_context(context); - out: - ds_put_context(context); - return error; -} + /* we require some space to do alignment adjustments below */ + error = -EINVAL; + if (size < (DS_ALIGNMENT + ds_cfg.sizeof_rec[qual])) + goto out; -int ds_release_bts(struct task_struct *task) -{ - return ds_release(task, ds_bts); -} + if (th != (size_t)-1) { + th *= ds_cfg.sizeof_rec[qual]; -int ds_release_pebs(struct task_struct *task) -{ - return ds_release(task, ds_pebs); -} + error = -EINVAL; + if (size <= th) + goto out; + } -static int ds_get_index(struct task_struct *task, size_t *pos, - enum ds_qualifier qual) -{ - struct ds_context *context; - unsigned long base, index; - int error; + tracer->buffer = base; + tracer->size = size; + error = -ENOMEM; context = ds_get_context(task); - error = ds_validate_access(context, qual); - if (error < 0) + if (!context) goto out; + tracer->context = context; - base = ds_get(context->ds, qual, ds_buffer_base); - index = ds_get(context->ds, qual, ds_index); + ds_init_ds_trace(trace, qual, base, size, th, flags); - error = ((index - base) / ds_cfg.sizeof_rec[qual]); - if (pos) - *pos = error; + error = 0; out: - ds_put_context(context); return error; } -int ds_get_bts_index(struct task_struct *task, size_t *pos) +struct bts_tracer *ds_request_bts(struct task_struct *task, + void *base, size_t size, + bts_ovfl_callback_t ovfl, size_t th, + unsigned int flags) { - return ds_get_index(task, pos, ds_bts); -} - -int ds_get_pebs_index(struct task_struct *task, size_t *pos) -{ - return ds_get_index(task, pos, ds_pebs); -} - -static int ds_get_end(struct task_struct *task, size_t *pos, - enum ds_qualifier qual) -{ - struct ds_context *context; - unsigned long base, end; + struct bts_tracer *tracer; + unsigned long irq; int error; - context = ds_get_context(task); - error = ds_validate_access(context, qual); - if (error < 0) + error = -EOPNOTSUPP; + if (!ds_cfg.ctl[dsf_bts]) goto out; - base = ds_get(context->ds, qual, ds_buffer_base); - end = ds_get(context->ds, qual, ds_absolute_maximum); + /* buffer overflow notification is not yet implemented */ + error = -EOPNOTSUPP; + if (ovfl) + goto out; - error = ((end - base) / ds_cfg.sizeof_rec[qual]); - if (pos) - *pos = error; - out: - ds_put_context(context); - return error; -} + error = -ENOMEM; + tracer = kzalloc(sizeof(*tracer), GFP_KERNEL); + if (!tracer) + goto out; + tracer->ovfl = ovfl; -int ds_get_bts_end(struct task_struct *task, size_t *pos) -{ - return ds_get_end(task, pos, ds_bts); -} + error = ds_request(&tracer->ds, &tracer->trace.ds, + ds_bts, task, base, size, th, flags); + if (error < 0) + goto out_tracer; -int ds_get_pebs_end(struct task_struct *task, size_t *pos) -{ - return ds_get_end(task, pos, ds_pebs); -} -static int ds_access(struct task_struct *task, size_t index, - const void **record, enum ds_qualifier qual) -{ - struct ds_context *context; - unsigned long base, idx; - int error; + spin_lock_irqsave(&ds_lock, irq); - if (!record) - return -EINVAL; + error = -EPERM; + if (!check_tracer(task)) + goto out_unlock; + get_tracer(task); - context = ds_get_context(task); - error = ds_validate_access(context, qual); - if (error < 0) - goto out; + error = -EPERM; + if (tracer->ds.context->bts_master) + goto out_put_tracer; + tracer->ds.context->bts_master = tracer; - base = ds_get(context->ds, qual, ds_buffer_base); - idx = base + (index * ds_cfg.sizeof_rec[qual]); + spin_unlock_irqrestore(&ds_lock, irq); - error = -EINVAL; - if (idx > ds_get(context->ds, qual, ds_absolute_maximum)) - goto out; - *record = (const void *)idx; - error = ds_cfg.sizeof_rec[qual]; - out: - ds_put_context(context); - return error; -} + tracer->trace.read = bts_read; + tracer->trace.write = bts_write; -int ds_access_bts(struct task_struct *task, size_t index, const void **record) -{ - return ds_access(task, index, record, ds_bts); -} + ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_bts); + ds_resume_bts(tracer); -int ds_access_pebs(struct task_struct *task, size_t index, const void **record) -{ - return ds_access(task, index, record, ds_pebs); + return tracer; + + out_put_tracer: + put_tracer(task); + out_unlock: + spin_unlock_irqrestore(&ds_lock, irq); + ds_put_context(tracer->ds.context); + out_tracer: + kfree(tracer); + out: + return ERR_PTR(error); } -static int ds_write(struct task_struct *task, const void *record, size_t size, - enum ds_qualifier qual, int force) +struct pebs_tracer *ds_request_pebs(struct task_struct *task, + void *base, size_t size, + pebs_ovfl_callback_t ovfl, size_t th, + unsigned int flags) { - struct ds_context *context; + struct pebs_tracer *tracer; + unsigned long irq; int error; - if (!record) - return -EINVAL; + /* buffer overflow notification is not yet implemented */ + error = -EOPNOTSUPP; + if (ovfl) + goto out; - error = -EPERM; - context = ds_get_context(task); - if (!context) + error = -ENOMEM; + tracer = kzalloc(sizeof(*tracer), GFP_KERNEL); + if (!tracer) goto out; + tracer->ovfl = ovfl; - if (!force) { - error = ds_validate_access(context, qual); - if (error < 0) - goto out; - } + error = ds_request(&tracer->ds, &tracer->trace.ds, + ds_pebs, task, base, size, th, flags); + if (error < 0) + goto out_tracer; - error = 0; - while (size) { - unsigned long base, index, end, write_end, int_th; - unsigned long write_size, adj_write_size; + spin_lock_irqsave(&ds_lock, irq); - /* - * write as much as possible without producing an - * overflow interrupt. - * - * interrupt_threshold must either be - * - bigger than absolute_maximum or - * - point to a record between buffer_base and absolute_maximum - * - * index points to a valid record. - */ - base = ds_get(context->ds, qual, ds_buffer_base); - index = ds_get(context->ds, qual, ds_index); - end = ds_get(context->ds, qual, ds_absolute_maximum); - int_th = ds_get(context->ds, qual, ds_interrupt_threshold); + error = -EPERM; + if (!check_tracer(task)) + goto out_unlock; + get_tracer(task); - write_end = min(end, int_th); + error = -EPERM; + if (tracer->ds.context->pebs_master) + goto out_put_tracer; + tracer->ds.context->pebs_master = tracer; - /* if we are already beyond the interrupt threshold, - * we fill the entire buffer */ - if (write_end <= index) - write_end = end; + spin_unlock_irqrestore(&ds_lock, irq); - if (write_end <= index) - goto out; + ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_bts); + ds_resume_pebs(tracer); - write_size = min((unsigned long) size, write_end - index); - memcpy((void *)index, record, write_size); + return tracer; - record = (const char *)record + write_size; - size -= write_size; - error += write_size; + out_put_tracer: + put_tracer(task); + out_unlock: + spin_unlock_irqrestore(&ds_lock, irq); + ds_put_context(tracer->ds.context); + out_tracer: + kfree(tracer); + out: + return ERR_PTR(error); +} - adj_write_size = write_size / ds_cfg.sizeof_rec[qual]; - adj_write_size *= ds_cfg.sizeof_rec[qual]; +void ds_release_bts(struct bts_tracer *tracer) +{ + if (!tracer) + return; - /* zero out trailing bytes */ - memset((char *)index + write_size, 0, - adj_write_size - write_size); - index += adj_write_size; + ds_suspend_bts(tracer); - if (index >= end) - index = base; - ds_set(context->ds, qual, ds_index, index); + WARN_ON_ONCE(tracer->ds.context->bts_master != tracer); + tracer->ds.context->bts_master = NULL; - if (index >= int_th) - ds_overflow(task, context, qual); - } + put_tracer(tracer->ds.context->task); + ds_put_context(tracer->ds.context); - out: - ds_put_context(context); - return error; + kfree(tracer); } -int ds_write_bts(struct task_struct *task, const void *record, size_t size) +void ds_suspend_bts(struct bts_tracer *tracer) { - return ds_write(task, record, size, ds_bts, /* force = */ 0); -} + struct task_struct *task; -int ds_write_pebs(struct task_struct *task, const void *record, size_t size) -{ - return ds_write(task, record, size, ds_pebs, /* force = */ 0); -} + if (!tracer) + return; -int ds_unchecked_write_bts(struct task_struct *task, - const void *record, size_t size) -{ - return ds_write(task, record, size, ds_bts, /* force = */ 1); -} + task = tracer->ds.context->task; -int ds_unchecked_write_pebs(struct task_struct *task, - const void *record, size_t size) -{ - return ds_write(task, record, size, ds_pebs, /* force = */ 1); + if (!task || (task == current)) + update_debugctlmsr(get_debugctlmsr() & ~BTS_CONTROL); + + if (task) { + task->thread.debugctlmsr &= ~BTS_CONTROL; + + if (!task->thread.debugctlmsr) + clear_tsk_thread_flag(task, TIF_DEBUGCTLMSR); + } } -static int ds_reset_or_clear(struct task_struct *task, - enum ds_qualifier qual, int clear) +void ds_resume_bts(struct bts_tracer *tracer) { - struct ds_context *context; - unsigned long base, end; - int error; + struct task_struct *task; + unsigned long control; - context = ds_get_context(task); - error = ds_validate_access(context, qual); - if (error < 0) - goto out; + if (!tracer) + return; - base = ds_get(context->ds, qual, ds_buffer_base); - end = ds_get(context->ds, qual, ds_absolute_maximum); + task = tracer->ds.context->task; - if (clear) - memset((void *)base, 0, end - base); + control = ds_cfg.ctl[dsf_bts]; + if (!(tracer->trace.ds.flags & BTS_KERNEL)) + control |= ds_cfg.ctl[dsf_bts_kernel]; + if (!(tracer->trace.ds.flags & BTS_USER)) + control |= ds_cfg.ctl[dsf_bts_user]; - ds_set(context->ds, qual, ds_index, base); + if (task) { + task->thread.debugctlmsr |= control; + set_tsk_thread_flag(task, TIF_DEBUGCTLMSR); + } - error = 0; - out: - ds_put_context(context); - return error; + if (!task || (task == current)) + update_debugctlmsr(get_debugctlmsr() | control); } -int ds_reset_bts(struct task_struct *task) +void ds_release_pebs(struct pebs_tracer *tracer) { - return ds_reset_or_clear(task, ds_bts, /* clear = */ 0); + if (!tracer) + return; + + ds_suspend_pebs(tracer); + + WARN_ON_ONCE(tracer->ds.context->pebs_master != tracer); + tracer->ds.context->pebs_master = NULL; + + put_tracer(tracer->ds.context->task); + ds_put_context(tracer->ds.context); + + kfree(tracer); } -int ds_reset_pebs(struct task_struct *task) +void ds_suspend_pebs(struct pebs_tracer *tracer) { - return ds_reset_or_clear(task, ds_pebs, /* clear = */ 0); + } -int ds_clear_bts(struct task_struct *task) +void ds_resume_pebs(struct pebs_tracer *tracer) { - return ds_reset_or_clear(task, ds_bts, /* clear = */ 1); + } -int ds_clear_pebs(struct task_struct *task) +const struct bts_trace *ds_read_bts(struct bts_tracer *tracer) { - return ds_reset_or_clear(task, ds_pebs, /* clear = */ 1); + if (!tracer) + return NULL; + + ds_read_config(tracer->ds.context, &tracer->trace.ds, ds_bts); + return &tracer->trace; } -int ds_get_pebs_reset(struct task_struct *task, u64 *value) +const struct pebs_trace *ds_read_pebs(struct pebs_tracer *tracer) { - struct ds_context *context; - int error; + if (!tracer) + return NULL; + + ds_read_config(tracer->ds.context, &tracer->trace.ds, ds_pebs); + tracer->trace.reset_value = + *(u64 *)(tracer->ds.context->ds + (ds_cfg.sizeof_field * 8)); + + return &tracer->trace; +} - if (!value) +int ds_reset_bts(struct bts_tracer *tracer) +{ + if (!tracer) return -EINVAL; - context = ds_get_context(task); - error = ds_validate_access(context, ds_pebs); - if (error < 0) - goto out; + tracer->trace.ds.top = tracer->trace.ds.begin; - *value = *(u64 *)(context->ds + (ds_cfg.sizeof_field * 8)); + ds_set(tracer->ds.context->ds, ds_bts, ds_index, + (unsigned long)tracer->trace.ds.top); - error = 0; - out: - ds_put_context(context); - return error; + return 0; } -int ds_set_pebs_reset(struct task_struct *task, u64 value) +int ds_reset_pebs(struct pebs_tracer *tracer) { - struct ds_context *context; - int error; + if (!tracer) + return -EINVAL; - context = ds_get_context(task); - error = ds_validate_access(context, ds_pebs); - if (error < 0) - goto out; + tracer->trace.ds.top = tracer->trace.ds.begin; - *(u64 *)(context->ds + (ds_cfg.sizeof_field * 8)) = value; + ds_set(tracer->ds.context->ds, ds_bts, ds_index, + (unsigned long)tracer->trace.ds.top); - error = 0; - out: - ds_put_context(context); - return error; + return 0; +} + +int ds_set_pebs_reset(struct pebs_tracer *tracer, u64 value) +{ + if (!tracer) + return -EINVAL; + + *(u64 *)(tracer->ds.context->ds + (ds_cfg.sizeof_field * 8)) = value; + + return 0; } -static const struct ds_configuration ds_cfg_var = { - .sizeof_ds = sizeof(long) * 12, - .sizeof_field = sizeof(long), - .sizeof_rec[ds_bts] = sizeof(long) * 3, - .sizeof_rec[ds_pebs] = sizeof(long) * 10 +static const struct ds_configuration ds_cfg_netburst = { + .name = "netburst", + .ctl[dsf_bts] = (1 << 2) | (1 << 3), + .ctl[dsf_bts_kernel] = (1 << 5), + .ctl[dsf_bts_user] = (1 << 6), + + .sizeof_field = sizeof(long), + .sizeof_rec[ds_bts] = sizeof(long) * 3, +#ifdef __i386__ + .sizeof_rec[ds_pebs] = sizeof(long) * 10, +#else + .sizeof_rec[ds_pebs] = sizeof(long) * 18, +#endif }; -static const struct ds_configuration ds_cfg_64 = { - .sizeof_ds = 8 * 12, - .sizeof_field = 8, - .sizeof_rec[ds_bts] = 8 * 3, - .sizeof_rec[ds_pebs] = 8 * 10 +static const struct ds_configuration ds_cfg_pentium_m = { + .name = "pentium m", + .ctl[dsf_bts] = (1 << 6) | (1 << 7), + + .sizeof_field = sizeof(long), + .sizeof_rec[ds_bts] = sizeof(long) * 3, +#ifdef __i386__ + .sizeof_rec[ds_pebs] = sizeof(long) * 10, +#else + .sizeof_rec[ds_pebs] = sizeof(long) * 18, +#endif +}; +static const struct ds_configuration ds_cfg_core2 = { + .name = "core 2", + .ctl[dsf_bts] = (1 << 6) | (1 << 7), + .ctl[dsf_bts_kernel] = (1 << 9), + .ctl[dsf_bts_user] = (1 << 10), + + .sizeof_field = 8, + .sizeof_rec[ds_bts] = 8 * 3, + .sizeof_rec[ds_pebs] = 8 * 18, }; -static inline void +static void ds_configure(const struct ds_configuration *cfg) { + memset(&ds_cfg, 0, sizeof(ds_cfg)); ds_cfg = *cfg; + + printk(KERN_INFO "[ds] using %s configuration\n", ds_cfg.name); + + if (!cpu_has_bts) { + ds_cfg.ctl[dsf_bts] = 0; + printk(KERN_INFO "[ds] bts not available\n"); + } + if (!cpu_has_pebs) + printk(KERN_INFO "[ds] pebs not available\n"); + + WARN_ON_ONCE(MAX_SIZEOF_DS < (12 * ds_cfg.sizeof_field)); } void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) @@ -821,16 +949,15 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) switch (c->x86) { case 0x6: switch (c->x86_model) { + case 0 ... 0xC: + /* sorry, don't know about them */ + break; case 0xD: case 0xE: /* Pentium M */ - ds_configure(&ds_cfg_var); + ds_configure(&ds_cfg_pentium_m); break; - case 0xF: /* Core2 */ - case 0x1C: /* Atom */ - ds_configure(&ds_cfg_64); - break; - default: - /* sorry, don't know about them */ + default: /* Core2, Atom, ... */ + ds_configure(&ds_cfg_core2); break; } break; @@ -839,7 +966,7 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) case 0x0: case 0x1: case 0x2: /* Netburst */ - ds_configure(&ds_cfg_var); + ds_configure(&ds_cfg_netburst); break; default: /* sorry, don't know about them */ @@ -852,13 +979,52 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) } } -void ds_free(struct ds_context *context) +/* + * Change the DS configuration from tracing prev to tracing next. + */ +void ds_switch_to(struct task_struct *prev, struct task_struct *next) +{ + struct ds_context *prev_ctx = prev->thread.ds_ctx; + struct ds_context *next_ctx = next->thread.ds_ctx; + + if (prev_ctx) { + update_debugctlmsr(0); + + if (prev_ctx->bts_master && + (prev_ctx->bts_master->trace.ds.flags & BTS_TIMESTAMPS)) { + struct bts_struct ts = { + .qualifier = bts_task_departs, + .variant.timestamp.jiffies = jiffies_64, + .variant.timestamp.pid = prev->pid + }; + bts_write(prev_ctx->bts_master, &ts); + } + } + + if (next_ctx) { + if (next_ctx->bts_master && + (next_ctx->bts_master->trace.ds.flags & BTS_TIMESTAMPS)) { + struct bts_struct ts = { + .qualifier = bts_task_arrives, + .variant.timestamp.jiffies = jiffies_64, + .variant.timestamp.pid = next->pid + }; + bts_write(next_ctx->bts_master, &ts); + } + + wrmsrl(MSR_IA32_DS_AREA, (unsigned long)next_ctx->ds); + } + + update_debugctlmsr(next->thread.debugctlmsr); +} + +void ds_copy_thread(struct task_struct *tsk, struct task_struct *father) +{ + clear_tsk_thread_flag(tsk, TIF_DS_AREA_MSR); + tsk->thread.ds_ctx = NULL; +} + +void ds_exit_thread(struct task_struct *tsk) { - /* This is called when the task owning the parameter context - * is dying. There should not be any user of that context left - * to disturb us, anymore. */ - unsigned long leftovers = context->count; - while (leftovers--) - ds_put_context(context); + WARN_ON(tsk->thread.ds_ctx); } -#endif /* CONFIG_X86_DS */ diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c new file mode 100644 index 000000000000..6b1f6f6f8661 --- /dev/null +++ b/arch/x86/kernel/dumpstack.c @@ -0,0 +1,351 @@ +/* + * Copyright (C) 1991, 1992 Linus Torvalds + * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs + */ +#include <linux/kallsyms.h> +#include <linux/kprobes.h> +#include <linux/uaccess.h> +#include <linux/utsname.h> +#include <linux/hardirq.h> +#include <linux/kdebug.h> +#include <linux/module.h> +#include <linux/ptrace.h> +#include <linux/kexec.h> +#include <linux/bug.h> +#include <linux/nmi.h> +#include <linux/sysfs.h> + +#include <asm/stacktrace.h> + +#include "dumpstack.h" + +int panic_on_unrecovered_nmi; +unsigned int code_bytes = 64; +int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE; +static int die_counter; + +void printk_address(unsigned long address, int reliable) +{ + printk(" [<%p>] %s%pS\n", (void *) address, + reliable ? "" : "? ", (void *) address); +} + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +static void +print_ftrace_graph_addr(unsigned long addr, void *data, + const struct stacktrace_ops *ops, + struct thread_info *tinfo, int *graph) +{ + struct task_struct *task = tinfo->task; + unsigned long ret_addr; + int index = task->curr_ret_stack; + + if (addr != (unsigned long)return_to_handler) + return; + + if (!task->ret_stack || index < *graph) + return; + + index -= *graph; + ret_addr = task->ret_stack[index].ret; + + ops->address(data, ret_addr, 1); + + (*graph)++; +} +#else +static inline void +print_ftrace_graph_addr(unsigned long addr, void *data, + const struct stacktrace_ops *ops, + struct thread_info *tinfo, int *graph) +{ } +#endif + +/* + * x86-64 can have up to three kernel stacks: + * process stack + * interrupt stack + * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack + */ + +static inline int valid_stack_ptr(struct thread_info *tinfo, + void *p, unsigned int size, void *end) +{ + void *t = tinfo; + if (end) { + if (p < end && p >= (end-THREAD_SIZE)) + return 1; + else + return 0; + } + return p > t && p < t + THREAD_SIZE - size; +} + +unsigned long +print_context_stack(struct thread_info *tinfo, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data, + unsigned long *end, int *graph) +{ + struct stack_frame *frame = (struct stack_frame *)bp; + + while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) { + unsigned long addr; + + addr = *stack; + if (__kernel_text_address(addr)) { + if ((unsigned long) stack == bp + sizeof(long)) { + ops->address(data, addr, 1); + frame = frame->next_frame; + bp = (unsigned long) frame; + } else { + ops->address(data, addr, bp == 0); + } + print_ftrace_graph_addr(addr, data, ops, tinfo, graph); + } + stack++; + } + return bp; +} + + +static void +print_trace_warning_symbol(void *data, char *msg, unsigned long symbol) +{ + printk(data); + print_symbol(msg, symbol); + printk("\n"); +} + +static void print_trace_warning(void *data, char *msg) +{ + printk("%s%s\n", (char *)data, msg); +} + +static int print_trace_stack(void *data, char *name) +{ + printk("%s <%s> ", (char *)data, name); + return 0; +} + +/* + * Print one address/symbol entries per line. + */ +static void print_trace_address(void *data, unsigned long addr, int reliable) +{ + touch_nmi_watchdog(); + printk(data); + printk_address(addr, reliable); +} + +static const struct stacktrace_ops print_trace_ops = { + .warning = print_trace_warning, + .warning_symbol = print_trace_warning_symbol, + .stack = print_trace_stack, + .address = print_trace_address, +}; + +void +show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, + unsigned long *stack, unsigned long bp, char *log_lvl) +{ + printk("%sCall Trace:\n", log_lvl); + dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); +} + +void show_trace(struct task_struct *task, struct pt_regs *regs, + unsigned long *stack, unsigned long bp) +{ + show_trace_log_lvl(task, regs, stack, bp, ""); +} + +void show_stack(struct task_struct *task, unsigned long *sp) +{ + show_stack_log_lvl(task, NULL, sp, 0, ""); +} + +/* + * The architecture-independent dump_stack generator + */ +void dump_stack(void) +{ + unsigned long bp = 0; + unsigned long stack; + +#ifdef CONFIG_FRAME_POINTER + if (!bp) + get_bp(bp); +#endif + + printk("Pid: %d, comm: %.20s %s %s %.*s\n", + current->pid, current->comm, print_tainted(), + init_utsname()->release, + (int)strcspn(init_utsname()->version, " "), + init_utsname()->version); + show_trace(NULL, NULL, &stack, bp); +} +EXPORT_SYMBOL(dump_stack); + +static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; +static int die_owner = -1; +static unsigned int die_nest_count; + +unsigned __kprobes long oops_begin(void) +{ + int cpu; + unsigned long flags; + + oops_enter(); + + /* racy, but better than risking deadlock. */ + raw_local_irq_save(flags); + cpu = smp_processor_id(); + if (!__raw_spin_trylock(&die_lock)) { + if (cpu == die_owner) + /* nested oops. should stop eventually */; + else + __raw_spin_lock(&die_lock); + } + die_nest_count++; + die_owner = cpu; + console_verbose(); + bust_spinlocks(1); + return flags; +} + +void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) +{ + if (regs && kexec_should_crash(current)) + crash_kexec(regs); + + bust_spinlocks(0); + die_owner = -1; + add_taint(TAINT_DIE); + die_nest_count--; + if (!die_nest_count) + /* Nest count reaches zero, release the lock. */ + __raw_spin_unlock(&die_lock); + raw_local_irq_restore(flags); + oops_exit(); + + if (!signr) + return; + if (in_interrupt()) + panic("Fatal exception in interrupt"); + if (panic_on_oops) + panic("Fatal exception"); + do_exit(signr); +} + +int __kprobes __die(const char *str, struct pt_regs *regs, long err) +{ +#ifdef CONFIG_X86_32 + unsigned short ss; + unsigned long sp; +#endif + printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter); +#ifdef CONFIG_PREEMPT + printk("PREEMPT "); +#endif +#ifdef CONFIG_SMP + printk("SMP "); +#endif +#ifdef CONFIG_DEBUG_PAGEALLOC + printk("DEBUG_PAGEALLOC"); +#endif + printk("\n"); + sysfs_printk_last_file(); + if (notify_die(DIE_OOPS, str, regs, err, + current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) + return 1; + + show_registers(regs); +#ifdef CONFIG_X86_32 + sp = (unsigned long) (®s->sp); + savesegment(ss, ss); + if (user_mode(regs)) { + sp = regs->sp; + ss = regs->ss & 0xffff; + } + printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip); + print_symbol("%s", regs->ip); + printk(" SS:ESP %04x:%08lx\n", ss, sp); +#else + /* Executive summary in case the oops scrolled away */ + printk(KERN_ALERT "RIP "); + printk_address(regs->ip, 1); + printk(" RSP <%016lx>\n", regs->sp); +#endif + return 0; +} + +/* + * This is gone through when something in the kernel has done something bad + * and is about to be terminated: + */ +void die(const char *str, struct pt_regs *regs, long err) +{ + unsigned long flags = oops_begin(); + int sig = SIGSEGV; + + if (!user_mode_vm(regs)) + report_bug(regs->ip, regs); + + if (__die(str, regs, err)) + sig = 0; + oops_end(flags, regs, sig); +} + +void notrace __kprobes +die_nmi(char *str, struct pt_regs *regs, int do_panic) +{ + unsigned long flags; + + if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP) + return; + + /* + * We are in trouble anyway, lets at least try + * to get a message out. + */ + flags = oops_begin(); + printk(KERN_EMERG "%s", str); + printk(" on CPU%d, ip %08lx, registers:\n", + smp_processor_id(), regs->ip); + show_registers(regs); + oops_end(flags, regs, 0); + if (do_panic || panic_on_oops) + panic("Non maskable interrupt"); + nmi_exit(); + local_irq_enable(); + do_exit(SIGBUS); +} + +static int __init oops_setup(char *s) +{ + if (!s) + return -EINVAL; + if (!strcmp(s, "panic")) + panic_on_oops = 1; + return 0; +} +early_param("oops", oops_setup); + +static int __init kstack_setup(char *s) +{ + if (!s) + return -EINVAL; + kstack_depth_to_print = simple_strtoul(s, NULL, 0); + return 0; +} +early_param("kstack", kstack_setup); + +static int __init code_bytes_setup(char *s) +{ + code_bytes = simple_strtoul(s, NULL, 0); + if (code_bytes > 8192) + code_bytes = 8192; + + return 1; +} +__setup("code_bytes=", code_bytes_setup); diff --git a/arch/x86/kernel/dumpstack.h b/arch/x86/kernel/dumpstack.h new file mode 100644 index 000000000000..da87590b8698 --- /dev/null +++ b/arch/x86/kernel/dumpstack.h @@ -0,0 +1,39 @@ +/* + * Copyright (C) 1991, 1992 Linus Torvalds + * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs + */ + +#ifndef DUMPSTACK_H +#define DUMPSTACK_H + +#ifdef CONFIG_X86_32 +#define STACKSLOTS_PER_LINE 8 +#define get_bp(bp) asm("movl %%ebp, %0" : "=r" (bp) :) +#else +#define STACKSLOTS_PER_LINE 4 +#define get_bp(bp) asm("movq %%rbp, %0" : "=r" (bp) :) +#endif + +extern unsigned long +print_context_stack(struct thread_info *tinfo, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data, + unsigned long *end, int *graph); + +extern void +show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, + unsigned long *stack, unsigned long bp, char *log_lvl); + +extern void +show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, + unsigned long *sp, unsigned long bp, char *log_lvl); + +extern unsigned int code_bytes; +extern int kstack_depth_to_print; + +/* The form of the top of the frame on the stack */ +struct stack_frame { + struct stack_frame *next_frame; + unsigned long return_address; +}; +#endif diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c index 201ee359a1a9..d593cd1f58dc 100644 --- a/arch/x86/kernel/dumpstack_32.c +++ b/arch/x86/kernel/dumpstack_32.c @@ -13,72 +13,18 @@ #include <linux/kexec.h> #include <linux/bug.h> #include <linux/nmi.h> +#include <linux/sysfs.h> #include <asm/stacktrace.h> -#define STACKSLOTS_PER_LINE 8 -#define get_bp(bp) asm("movl %%ebp, %0" : "=r" (bp) :) - -int panic_on_unrecovered_nmi; -int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE; -static unsigned int code_bytes = 64; -static int die_counter; - -void printk_address(unsigned long address, int reliable) -{ - printk(" [<%p>] %s%pS\n", (void *) address, - reliable ? "" : "? ", (void *) address); -} - -static inline int valid_stack_ptr(struct thread_info *tinfo, - void *p, unsigned int size, void *end) -{ - void *t = tinfo; - if (end) { - if (p < end && p >= (end-THREAD_SIZE)) - return 1; - else - return 0; - } - return p > t && p < t + THREAD_SIZE - size; -} - -/* The form of the top of the frame on the stack */ -struct stack_frame { - struct stack_frame *next_frame; - unsigned long return_address; -}; - -static inline unsigned long -print_context_stack(struct thread_info *tinfo, - unsigned long *stack, unsigned long bp, - const struct stacktrace_ops *ops, void *data, - unsigned long *end) -{ - struct stack_frame *frame = (struct stack_frame *)bp; - - while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) { - unsigned long addr; - - addr = *stack; - if (__kernel_text_address(addr)) { - if ((unsigned long) stack == bp + sizeof(long)) { - ops->address(data, addr, 1); - frame = frame->next_frame; - bp = (unsigned long) frame; - } else { - ops->address(data, addr, bp == 0); - } - } - stack++; - } - return bp; -} +#include "dumpstack.h" void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, const struct stacktrace_ops *ops, void *data) { + int graph = 0; + if (!task) task = current; @@ -106,7 +52,8 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, context = (struct thread_info *) ((unsigned long)stack & (~(THREAD_SIZE - 1))); - bp = print_context_stack(context, stack, bp, ops, data, NULL); + bp = print_context_stack(context, stack, bp, ops, + data, NULL, &graph); stack = (unsigned long *)context->previous_esp; if (!stack) @@ -118,57 +65,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, } EXPORT_SYMBOL(dump_trace); -static void -print_trace_warning_symbol(void *data, char *msg, unsigned long symbol) -{ - printk(data); - print_symbol(msg, symbol); - printk("\n"); -} - -static void print_trace_warning(void *data, char *msg) -{ - printk("%s%s\n", (char *)data, msg); -} - -static int print_trace_stack(void *data, char *name) -{ - printk("%s <%s> ", (char *)data, name); - return 0; -} - -/* - * Print one address/symbol entries per line. - */ -static void print_trace_address(void *data, unsigned long addr, int reliable) -{ - touch_nmi_watchdog(); - printk(data); - printk_address(addr, reliable); -} - -static const struct stacktrace_ops print_trace_ops = { - .warning = print_trace_warning, - .warning_symbol = print_trace_warning_symbol, - .stack = print_trace_stack, - .address = print_trace_address, -}; - -static void -show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, char *log_lvl) -{ - printk("%sCall Trace:\n", log_lvl); - dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); -} - -void show_trace(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp) -{ - show_trace_log_lvl(task, regs, stack, bp, ""); -} - -static void +void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, unsigned long *sp, unsigned long bp, char *log_lvl) { @@ -195,33 +92,6 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, show_trace_log_lvl(task, regs, sp, bp, log_lvl); } -void show_stack(struct task_struct *task, unsigned long *sp) -{ - show_stack_log_lvl(task, NULL, sp, 0, ""); -} - -/* - * The architecture-independent dump_stack generator - */ -void dump_stack(void) -{ - unsigned long bp = 0; - unsigned long stack; - -#ifdef CONFIG_FRAME_POINTER - if (!bp) - get_bp(bp); -#endif - - printk("Pid: %d, comm: %.20s %s %s %.*s\n", - current->pid, current->comm, print_tainted(), - init_utsname()->release, - (int)strcspn(init_utsname()->version, " "), - init_utsname()->version); - show_trace(NULL, NULL, &stack, bp); -} - -EXPORT_SYMBOL(dump_stack); void show_registers(struct pt_regs *regs) { @@ -282,166 +152,3 @@ int is_valid_bugaddr(unsigned long ip) return ud2 == 0x0b0f; } -static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; -static int die_owner = -1; -static unsigned int die_nest_count; - -unsigned __kprobes long oops_begin(void) -{ - unsigned long flags; - - oops_enter(); - - if (die_owner != raw_smp_processor_id()) { - console_verbose(); - raw_local_irq_save(flags); - __raw_spin_lock(&die_lock); - die_owner = smp_processor_id(); - die_nest_count = 0; - bust_spinlocks(1); - } else { - raw_local_irq_save(flags); - } - die_nest_count++; - return flags; -} - -void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) -{ - bust_spinlocks(0); - die_owner = -1; - add_taint(TAINT_DIE); - __raw_spin_unlock(&die_lock); - raw_local_irq_restore(flags); - - if (!regs) - return; - - if (kexec_should_crash(current)) - crash_kexec(regs); - if (in_interrupt()) - panic("Fatal exception in interrupt"); - if (panic_on_oops) - panic("Fatal exception"); - oops_exit(); - do_exit(signr); -} - -int __kprobes __die(const char *str, struct pt_regs *regs, long err) -{ - unsigned short ss; - unsigned long sp; - - printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter); -#ifdef CONFIG_PREEMPT - printk("PREEMPT "); -#endif -#ifdef CONFIG_SMP - printk("SMP "); -#endif -#ifdef CONFIG_DEBUG_PAGEALLOC - printk("DEBUG_PAGEALLOC"); -#endif - printk("\n"); - if (notify_die(DIE_OOPS, str, regs, err, - current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) - return 1; - - show_registers(regs); - /* Executive summary in case the oops scrolled away */ - sp = (unsigned long) (®s->sp); - savesegment(ss, ss); - if (user_mode(regs)) { - sp = regs->sp; - ss = regs->ss & 0xffff; - } - printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip); - print_symbol("%s", regs->ip); - printk(" SS:ESP %04x:%08lx\n", ss, sp); - return 0; -} - -/* - * This is gone through when something in the kernel has done something bad - * and is about to be terminated: - */ -void die(const char *str, struct pt_regs *regs, long err) -{ - unsigned long flags = oops_begin(); - - if (die_nest_count < 3) { - report_bug(regs->ip, regs); - - if (__die(str, regs, err)) - regs = NULL; - } else { - printk(KERN_EMERG "Recursive die() failure, output suppressed\n"); - } - - oops_end(flags, regs, SIGSEGV); -} - -static DEFINE_SPINLOCK(nmi_print_lock); - -void notrace __kprobes -die_nmi(char *str, struct pt_regs *regs, int do_panic) -{ - if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP) - return; - - spin_lock(&nmi_print_lock); - /* - * We are in trouble anyway, lets at least try - * to get a message out: - */ - bust_spinlocks(1); - printk(KERN_EMERG "%s", str); - printk(" on CPU%d, ip %08lx, registers:\n", - smp_processor_id(), regs->ip); - show_registers(regs); - if (do_panic) - panic("Non maskable interrupt"); - console_silent(); - spin_unlock(&nmi_print_lock); - bust_spinlocks(0); - - /* - * If we are in kernel we are probably nested up pretty bad - * and might aswell get out now while we still can: - */ - if (!user_mode_vm(regs)) { - current->thread.trap_no = 2; - crash_kexec(regs); - } - - do_exit(SIGSEGV); -} - -static int __init oops_setup(char *s) -{ - if (!s) - return -EINVAL; - if (!strcmp(s, "panic")) - panic_on_oops = 1; - return 0; -} -early_param("oops", oops_setup); - -static int __init kstack_setup(char *s) -{ - if (!s) - return -EINVAL; - kstack_depth_to_print = simple_strtoul(s, NULL, 0); - return 0; -} -early_param("kstack", kstack_setup); - -static int __init code_bytes_setup(char *s) -{ - code_bytes = simple_strtoul(s, NULL, 0); - if (code_bytes > 8192) - code_bytes = 8192; - - return 1; -} -__setup("code_bytes=", code_bytes_setup); diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c index 086cc8118e39..c302d0707048 100644 --- a/arch/x86/kernel/dumpstack_64.c +++ b/arch/x86/kernel/dumpstack_64.c @@ -13,22 +13,11 @@ #include <linux/kexec.h> #include <linux/bug.h> #include <linux/nmi.h> +#include <linux/sysfs.h> #include <asm/stacktrace.h> -#define STACKSLOTS_PER_LINE 4 -#define get_bp(bp) asm("movq %%rbp, %0" : "=r" (bp) :) - -int panic_on_unrecovered_nmi; -int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE; -static unsigned int code_bytes = 64; -static int die_counter; - -void printk_address(unsigned long address, int reliable) -{ - printk(" [<%p>] %s%pS\n", (void *) address, - reliable ? "" : "? ", (void *) address); -} +#include "dumpstack.h" static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, unsigned *usedp, char **idp) @@ -112,51 +101,6 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack */ -static inline int valid_stack_ptr(struct thread_info *tinfo, - void *p, unsigned int size, void *end) -{ - void *t = tinfo; - if (end) { - if (p < end && p >= (end-THREAD_SIZE)) - return 1; - else - return 0; - } - return p > t && p < t + THREAD_SIZE - size; -} - -/* The form of the top of the frame on the stack */ -struct stack_frame { - struct stack_frame *next_frame; - unsigned long return_address; -}; - -static inline unsigned long -print_context_stack(struct thread_info *tinfo, - unsigned long *stack, unsigned long bp, - const struct stacktrace_ops *ops, void *data, - unsigned long *end) -{ - struct stack_frame *frame = (struct stack_frame *)bp; - - while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) { - unsigned long addr; - - addr = *stack; - if (__kernel_text_address(addr)) { - if ((unsigned long) stack == bp + sizeof(long)) { - ops->address(data, addr, 1); - frame = frame->next_frame; - bp = (unsigned long) frame; - } else { - ops->address(data, addr, bp == 0); - } - } - stack++; - } - return bp; -} - void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, const struct stacktrace_ops *ops, void *data) @@ -165,6 +109,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr; unsigned used = 0; struct thread_info *tinfo; + int graph = 0; if (!task) task = current; @@ -205,7 +150,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, break; bp = print_context_stack(tinfo, stack, bp, ops, - data, estack_end); + data, estack_end, &graph); ops->stack(data, "<EOE>"); /* * We link to the next stack via the @@ -224,7 +169,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, if (ops->stack(data, "IRQ") < 0) break; bp = print_context_stack(tinfo, stack, bp, - ops, data, irqstack_end); + ops, data, irqstack_end, &graph); /* * We link to the next stack (which would be * the process stack normally) the last @@ -242,62 +187,12 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, /* * This handles the process stack: */ - bp = print_context_stack(tinfo, stack, bp, ops, data, NULL); + bp = print_context_stack(tinfo, stack, bp, ops, data, NULL, &graph); put_cpu(); } EXPORT_SYMBOL(dump_trace); -static void -print_trace_warning_symbol(void *data, char *msg, unsigned long symbol) -{ - printk(data); - print_symbol(msg, symbol); - printk("\n"); -} - -static void print_trace_warning(void *data, char *msg) -{ - printk("%s%s\n", (char *)data, msg); -} - -static int print_trace_stack(void *data, char *name) -{ - printk("%s <%s> ", (char *)data, name); - return 0; -} - -/* - * Print one address/symbol entries per line. - */ -static void print_trace_address(void *data, unsigned long addr, int reliable) -{ - touch_nmi_watchdog(); - printk(data); - printk_address(addr, reliable); -} - -static const struct stacktrace_ops print_trace_ops = { - .warning = print_trace_warning, - .warning_symbol = print_trace_warning_symbol, - .stack = print_trace_stack, - .address = print_trace_address, -}; - -static void -show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, char *log_lvl) -{ - printk("%sCall Trace:\n", log_lvl); - dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); -} - -void show_trace(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp) -{ - show_trace_log_lvl(task, regs, stack, bp, ""); -} - -static void +void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, unsigned long *sp, unsigned long bp, char *log_lvl) { @@ -341,33 +236,6 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, show_trace_log_lvl(task, regs, sp, bp, log_lvl); } -void show_stack(struct task_struct *task, unsigned long *sp) -{ - show_stack_log_lvl(task, NULL, sp, 0, ""); -} - -/* - * The architecture-independent dump_stack generator - */ -void dump_stack(void) -{ - unsigned long bp = 0; - unsigned long stack; - -#ifdef CONFIG_FRAME_POINTER - if (!bp) - get_bp(bp); -#endif - - printk("Pid: %d, comm: %.20s %s %s %.*s\n", - current->pid, current->comm, print_tainted(), - init_utsname()->release, - (int)strcspn(init_utsname()->version, " "), - init_utsname()->version); - show_trace(NULL, NULL, &stack, bp); -} -EXPORT_SYMBOL(dump_stack); - void show_registers(struct pt_regs *regs) { int i; @@ -428,146 +296,3 @@ int is_valid_bugaddr(unsigned long ip) return ud2 == 0x0b0f; } -static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; -static int die_owner = -1; -static unsigned int die_nest_count; - -unsigned __kprobes long oops_begin(void) -{ - int cpu; - unsigned long flags; - - oops_enter(); - - /* racy, but better than risking deadlock. */ - raw_local_irq_save(flags); - cpu = smp_processor_id(); - if (!__raw_spin_trylock(&die_lock)) { - if (cpu == die_owner) - /* nested oops. should stop eventually */; - else - __raw_spin_lock(&die_lock); - } - die_nest_count++; - die_owner = cpu; - console_verbose(); - bust_spinlocks(1); - return flags; -} - -void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) -{ - die_owner = -1; - bust_spinlocks(0); - die_nest_count--; - if (!die_nest_count) - /* Nest count reaches zero, release the lock. */ - __raw_spin_unlock(&die_lock); - raw_local_irq_restore(flags); - if (!regs) { - oops_exit(); - return; - } - if (in_interrupt()) - panic("Fatal exception in interrupt"); - if (panic_on_oops) - panic("Fatal exception"); - oops_exit(); - do_exit(signr); -} - -int __kprobes __die(const char *str, struct pt_regs *regs, long err) -{ - printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter); -#ifdef CONFIG_PREEMPT - printk("PREEMPT "); -#endif -#ifdef CONFIG_SMP - printk("SMP "); -#endif -#ifdef CONFIG_DEBUG_PAGEALLOC - printk("DEBUG_PAGEALLOC"); -#endif - printk("\n"); - if (notify_die(DIE_OOPS, str, regs, err, - current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) - return 1; - - show_registers(regs); - add_taint(TAINT_DIE); - /* Executive summary in case the oops scrolled away */ - printk(KERN_ALERT "RIP "); - printk_address(regs->ip, 1); - printk(" RSP <%016lx>\n", regs->sp); - if (kexec_should_crash(current)) - crash_kexec(regs); - return 0; -} - -void die(const char *str, struct pt_regs *regs, long err) -{ - unsigned long flags = oops_begin(); - - if (!user_mode(regs)) - report_bug(regs->ip, regs); - - if (__die(str, regs, err)) - regs = NULL; - oops_end(flags, regs, SIGSEGV); -} - -notrace __kprobes void -die_nmi(char *str, struct pt_regs *regs, int do_panic) -{ - unsigned long flags; - - if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP) - return; - - flags = oops_begin(); - /* - * We are in trouble anyway, lets at least try - * to get a message out. - */ - printk(KERN_EMERG "%s", str); - printk(" on CPU%d, ip %08lx, registers:\n", - smp_processor_id(), regs->ip); - show_registers(regs); - if (kexec_should_crash(current)) - crash_kexec(regs); - if (do_panic || panic_on_oops) - panic("Non maskable interrupt"); - oops_end(flags, NULL, SIGBUS); - nmi_exit(); - local_irq_enable(); - do_exit(SIGBUS); -} - -static int __init oops_setup(char *s) -{ - if (!s) - return -EINVAL; - if (!strcmp(s, "panic")) - panic_on_oops = 1; - return 0; -} -early_param("oops", oops_setup); - -static int __init kstack_setup(char *s) -{ - if (!s) - return -EINVAL; - kstack_depth_to_print = simple_strtoul(s, NULL, 0); - return 0; -} -early_param("kstack", kstack_setup); - -static int __init code_bytes_setup(char *s) -{ - code_bytes = simple_strtoul(s, NULL, 0); - if (code_bytes > 8192) - code_bytes = 8192; - - return 1; -} -__setup("code_bytes=", code_bytes_setup); diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 78e642feac30..65a13943e098 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -677,22 +677,6 @@ struct early_res { }; static struct early_res early_res[MAX_EARLY_RES] __initdata = { { 0, PAGE_SIZE, "BIOS data page" }, /* BIOS data page */ -#if defined(CONFIG_X86_64) && defined(CONFIG_X86_TRAMPOLINE) - { TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" }, -#endif -#if defined(CONFIG_X86_32) && defined(CONFIG_SMP) - /* - * But first pinch a few for the stack/trampoline stuff - * FIXME: Don't need the extra page at 4K, but need to fix - * trampoline before removing it. (see the GDT stuff) - */ - { PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE" }, - /* - * Has to be in very low memory so we can execute - * real-mode AP code. - */ - { TRAMPOLINE_BASE, TRAMPOLINE_BASE + PAGE_SIZE, "TRAMPOLINE" }, -#endif {} }; @@ -1282,25 +1266,25 @@ void __init e820_reserve_resources(void) e820_res = res; for (i = 0; i < e820.nr_map; i++) { end = e820.map[i].addr + e820.map[i].size - 1; -#ifndef CONFIG_RESOURCES_64BIT - if (end > 0x100000000ULL) { + if (end != (resource_size_t)end) { res++; continue; } -#endif res->name = e820_type_to_string(e820.map[i].type); res->start = e820.map[i].addr; res->end = end; - res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; + res->flags = IORESOURCE_MEM; /* * don't register the region that could be conflicted with * pci device BAR resource and insert them later in * pcibios_resource_survey() */ - if (e820.map[i].type != E820_RESERVED || res->start < (1ULL<<20)) + if (e820.map[i].type != E820_RESERVED || res->start < (1ULL<<20)) { + res->flags |= IORESOURCE_BUSY; insert_resource(&iomem_resource, res); + } res++; } @@ -1320,7 +1304,7 @@ void __init e820_reserve_resources_late(void) res = e820_res; for (i = 0; i < e820.nr_map; i++) { if (!res->parent && res->end) - reserve_region_with_split(&iomem_resource, res->start, res->end, res->name); + insert_resource_expand_to_fit(&iomem_resource, res); res++; } } diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c index 733c4f8d42ea..744aa7fc49d5 100644 --- a/arch/x86/kernel/early-quirks.c +++ b/arch/x86/kernel/early-quirks.c @@ -17,6 +17,7 @@ #include <asm/io_apic.h> #include <asm/apic.h> #include <asm/iommu.h> +#include <asm/gart.h> static void __init fix_hypertransport_config(int num, int slot, int func) { @@ -95,7 +96,8 @@ static void __init nvidia_bugs(int num, int slot, int func) } -static u32 ati_ixp4x0_rev(int num, int slot, int func) +#if defined(CONFIG_ACPI) && defined(CONFIG_X86_IO_APIC) +static u32 __init ati_ixp4x0_rev(int num, int slot, int func) { u32 d; u8 b; @@ -115,7 +117,6 @@ static u32 ati_ixp4x0_rev(int num, int slot, int func) static void __init ati_bugs(int num, int slot, int func) { -#if defined(CONFIG_ACPI) && defined (CONFIG_X86_IO_APIC) u32 d; u8 b; @@ -138,21 +139,54 @@ static void __init ati_bugs(int num, int slot, int func) printk(KERN_INFO "If you got timer trouble " "try acpi_use_timer_override\n"); } -#endif } -#ifdef CONFIG_DMAR -static void __init intel_g33_dmar(int num, int slot, int func) +static u32 __init ati_sbx00_rev(int num, int slot, int func) { - struct acpi_table_header *dmar_tbl; - acpi_status status; + u32 old, d; + + d = read_pci_config(num, slot, func, 0x70); + old = d; + d &= ~(1<<8); + write_pci_config(num, slot, func, 0x70, d); + d = read_pci_config(num, slot, func, 0x8); + d &= 0xff; + write_pci_config(num, slot, func, 0x70, old); - status = acpi_get_table(ACPI_SIG_DMAR, 0, &dmar_tbl); - if (ACPI_SUCCESS(status)) { - printk(KERN_INFO "BIOS BUG: DMAR advertised on Intel G31/G33 chipset -- ignoring\n"); - dmar_disabled = 1; + return d; +} + +static void __init ati_bugs_contd(int num, int slot, int func) +{ + u32 d, rev; + + if (acpi_use_timer_override) + return; + + rev = ati_sbx00_rev(num, slot, func); + if (rev > 0x13) + return; + + /* check for IRQ0 interrupt swap */ + d = read_pci_config(num, slot, func, 0x64); + if (!(d & (1<<14))) + acpi_skip_timer_override = 1; + + if (acpi_skip_timer_override) { + printk(KERN_INFO "SB600 revision 0x%x\n", rev); + printk(KERN_INFO "Ignoring ACPI timer override.\n"); + printk(KERN_INFO "If you got timer trouble " + "try acpi_use_timer_override\n"); } } +#else +static void __init ati_bugs(int num, int slot, int func) +{ +} + +static void __init ati_bugs_contd(int num, int slot, int func) +{ +} #endif #define QFLAG_APPLY_ONCE 0x1 @@ -176,10 +210,8 @@ static struct chipset early_qrk[] __initdata = { PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, fix_hypertransport_config }, { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP400_SMBUS, PCI_CLASS_SERIAL_SMBUS, PCI_ANY_ID, 0, ati_bugs }, -#ifdef CONFIG_DMAR - { PCI_VENDOR_ID_INTEL, 0x29c0, - PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, intel_g33_dmar }, -#endif + { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_SBX00_SMBUS, + PCI_CLASS_SERIAL_SMBUS, PCI_ANY_ID, 0, ati_bugs_contd }, {} }; diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c index 34ad997d3834..23b138e31e9c 100644 --- a/arch/x86/kernel/early_printk.c +++ b/arch/x86/kernel/early_printk.c @@ -875,49 +875,6 @@ static struct console early_dbgp_console = { }; #endif -/* Console interface to a host file on AMD's SimNow! */ - -static int simnow_fd; - -enum { - MAGIC1 = 0xBACCD00A, - MAGIC2 = 0xCA110000, - XOPEN = 5, - XWRITE = 4, -}; - -static noinline long simnow(long cmd, long a, long b, long c) -{ - long ret; - - asm volatile("cpuid" : - "=a" (ret) : - "b" (a), "c" (b), "d" (c), "0" (MAGIC1), "D" (cmd + MAGIC2)); - return ret; -} - -static void __init simnow_init(char *str) -{ - char *fn = "klog"; - - if (*str == '=') - fn = ++str; - /* error ignored */ - simnow_fd = simnow(XOPEN, (unsigned long)fn, O_WRONLY|O_APPEND|O_CREAT, 0644); -} - -static void simnow_write(struct console *con, const char *s, unsigned n) -{ - simnow(XWRITE, simnow_fd, (unsigned long)s, n); -} - -static struct console simnow_console = { - .name = "simnow", - .write = simnow_write, - .flags = CON_PRINTBUFFER, - .index = -1, -}; - /* Direct interface for emergencies */ static struct console *early_console = &early_vga_console; static int __initdata early_console_initialized; @@ -960,10 +917,6 @@ static int __init setup_early_printk(char *buf) max_ypos = boot_params.screen_info.orig_video_lines; current_ypos = boot_params.screen_info.orig_y; early_console = &early_vga_console; - } else if (!strncmp(buf, "simnow", 6)) { - simnow_init(buf + 6); - early_console = &simnow_console; - keep_early = 1; #ifdef CONFIG_EARLY_PRINTK_DBGP } else if (!strncmp(buf, "dbgp", 4)) { if (early_dbgp_init(buf+4) < 0) diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index 945a31cdd81f..1119d247fe11 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -367,6 +367,10 @@ void __init efi_init(void) efi.smbios = config_tables[i].table; printk(" SMBIOS=0x%lx ", config_tables[i].table); } else if (!efi_guidcmp(config_tables[i].guid, + UV_SYSTEM_TABLE_GUID)) { + efi.uv_systab = config_tables[i].table; + printk(" UVsystab=0x%lx ", config_tables[i].table); + } else if (!efi_guidcmp(config_tables[i].guid, HCDP_TABLE_GUID)) { efi.hcdp = config_tables[i].table; printk(" HCDP=0x%lx ", config_tables[i].table); diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index b21fbfaffe39..d6f0490a7391 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -619,28 +619,37 @@ END(syscall_badsys) 27:; /* - * Build the entry stubs and pointer table with - * some assembler magic. + * Build the entry stubs and pointer table with some assembler magic. + * We pack 7 stubs into a single 32-byte chunk, which will fit in a + * single cache line on all modern x86 implementations. */ -.section .rodata,"a" +.section .init.rodata,"a" ENTRY(interrupt) .text - + .p2align 5 + .p2align CONFIG_X86_L1_CACHE_SHIFT ENTRY(irq_entries_start) RING0_INT_FRAME -vector=0 -.rept NR_IRQS - ALIGN - .if vector +vector=FIRST_EXTERNAL_VECTOR +.rept (NR_VECTORS-FIRST_EXTERNAL_VECTOR+6)/7 + .balign 32 + .rept 7 + .if vector < NR_VECTORS + .if vector <> FIRST_EXTERNAL_VECTOR CFI_ADJUST_CFA_OFFSET -4 - .endif -1: pushl $~(vector) + .endif +1: pushl $(~vector+0x80) /* Note: always in signed byte range */ CFI_ADJUST_CFA_OFFSET 4 - jmp common_interrupt - .previous + .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6 + jmp 2f + .endif + .previous .long 1b - .text + .text vector=vector+1 + .endif + .endr +2: jmp common_interrupt .endr END(irq_entries_start) @@ -652,8 +661,9 @@ END(interrupt) * the CPU automatically disables interrupts when executing an IRQ vector, * so IRQ-flags tracing has to follow that: */ - ALIGN + .p2align CONFIG_X86_L1_CACHE_SHIFT common_interrupt: + addl $-0x80,(%esp) /* Adjust vector into the [-256,-1] range */ SAVE_ALL TRACE_IRQS_OFF movl %esp,%eax @@ -678,65 +688,6 @@ ENDPROC(name) /* The include is where all of the SMP etc. interrupts come from */ #include "entry_arch.h" -KPROBE_ENTRY(page_fault) - RING0_EC_FRAME - pushl $do_page_fault - CFI_ADJUST_CFA_OFFSET 4 - ALIGN -error_code: - /* the function address is in %fs's slot on the stack */ - pushl %es - CFI_ADJUST_CFA_OFFSET 4 - /*CFI_REL_OFFSET es, 0*/ - pushl %ds - CFI_ADJUST_CFA_OFFSET 4 - /*CFI_REL_OFFSET ds, 0*/ - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 - CFI_REL_OFFSET eax, 0 - pushl %ebp - CFI_ADJUST_CFA_OFFSET 4 - CFI_REL_OFFSET ebp, 0 - pushl %edi - CFI_ADJUST_CFA_OFFSET 4 - CFI_REL_OFFSET edi, 0 - pushl %esi - CFI_ADJUST_CFA_OFFSET 4 - CFI_REL_OFFSET esi, 0 - pushl %edx - CFI_ADJUST_CFA_OFFSET 4 - CFI_REL_OFFSET edx, 0 - pushl %ecx - CFI_ADJUST_CFA_OFFSET 4 - CFI_REL_OFFSET ecx, 0 - pushl %ebx - CFI_ADJUST_CFA_OFFSET 4 - CFI_REL_OFFSET ebx, 0 - cld - pushl %fs - CFI_ADJUST_CFA_OFFSET 4 - /*CFI_REL_OFFSET fs, 0*/ - movl $(__KERNEL_PERCPU), %ecx - movl %ecx, %fs - UNWIND_ESPFIX_STACK - popl %ecx - CFI_ADJUST_CFA_OFFSET -4 - /*CFI_REGISTER es, ecx*/ - movl PT_FS(%esp), %edi # get the function address - movl PT_ORIG_EAX(%esp), %edx # get the error code - movl $-1, PT_ORIG_EAX(%esp) # no syscall to restart - mov %ecx, PT_FS(%esp) - /*CFI_REL_OFFSET fs, ES*/ - movl $(__USER_DS), %ecx - movl %ecx, %ds - movl %ecx, %es - TRACE_IRQS_OFF - movl %esp,%eax # pt_regs pointer - call *%edi - jmp ret_from_exception - CFI_ENDPROC -KPROBE_END(page_fault) - ENTRY(coprocessor_error) RING0_INT_FRAME pushl $0 @@ -767,140 +718,6 @@ ENTRY(device_not_available) CFI_ENDPROC END(device_not_available) -/* - * Debug traps and NMI can happen at the one SYSENTER instruction - * that sets up the real kernel stack. Check here, since we can't - * allow the wrong stack to be used. - * - * "TSS_sysenter_sp0+12" is because the NMI/debug handler will have - * already pushed 3 words if it hits on the sysenter instruction: - * eflags, cs and eip. - * - * We just load the right stack, and push the three (known) values - * by hand onto the new stack - while updating the return eip past - * the instruction that would have done it for sysenter. - */ -#define FIX_STACK(offset, ok, label) \ - cmpw $__KERNEL_CS,4(%esp); \ - jne ok; \ -label: \ - movl TSS_sysenter_sp0+offset(%esp),%esp; \ - CFI_DEF_CFA esp, 0; \ - CFI_UNDEFINED eip; \ - pushfl; \ - CFI_ADJUST_CFA_OFFSET 4; \ - pushl $__KERNEL_CS; \ - CFI_ADJUST_CFA_OFFSET 4; \ - pushl $sysenter_past_esp; \ - CFI_ADJUST_CFA_OFFSET 4; \ - CFI_REL_OFFSET eip, 0 - -KPROBE_ENTRY(debug) - RING0_INT_FRAME - cmpl $ia32_sysenter_target,(%esp) - jne debug_stack_correct - FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn) -debug_stack_correct: - pushl $-1 # mark this as an int - CFI_ADJUST_CFA_OFFSET 4 - SAVE_ALL - TRACE_IRQS_OFF - xorl %edx,%edx # error code 0 - movl %esp,%eax # pt_regs pointer - call do_debug - jmp ret_from_exception - CFI_ENDPROC -KPROBE_END(debug) - -/* - * NMI is doubly nasty. It can happen _while_ we're handling - * a debug fault, and the debug fault hasn't yet been able to - * clear up the stack. So we first check whether we got an - * NMI on the sysenter entry path, but after that we need to - * check whether we got an NMI on the debug path where the debug - * fault happened on the sysenter path. - */ -KPROBE_ENTRY(nmi) - RING0_INT_FRAME - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 - movl %ss, %eax - cmpw $__ESPFIX_SS, %ax - popl %eax - CFI_ADJUST_CFA_OFFSET -4 - je nmi_espfix_stack - cmpl $ia32_sysenter_target,(%esp) - je nmi_stack_fixup - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 - movl %esp,%eax - /* Do not access memory above the end of our stack page, - * it might not exist. - */ - andl $(THREAD_SIZE-1),%eax - cmpl $(THREAD_SIZE-20),%eax - popl %eax - CFI_ADJUST_CFA_OFFSET -4 - jae nmi_stack_correct - cmpl $ia32_sysenter_target,12(%esp) - je nmi_debug_stack_check -nmi_stack_correct: - /* We have a RING0_INT_FRAME here */ - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 - SAVE_ALL - TRACE_IRQS_OFF - xorl %edx,%edx # zero error code - movl %esp,%eax # pt_regs pointer - call do_nmi - jmp restore_nocheck_notrace - CFI_ENDPROC - -nmi_stack_fixup: - RING0_INT_FRAME - FIX_STACK(12,nmi_stack_correct, 1) - jmp nmi_stack_correct - -nmi_debug_stack_check: - /* We have a RING0_INT_FRAME here */ - cmpw $__KERNEL_CS,16(%esp) - jne nmi_stack_correct - cmpl $debug,(%esp) - jb nmi_stack_correct - cmpl $debug_esp_fix_insn,(%esp) - ja nmi_stack_correct - FIX_STACK(24,nmi_stack_correct, 1) - jmp nmi_stack_correct - -nmi_espfix_stack: - /* We have a RING0_INT_FRAME here. - * - * create the pointer to lss back - */ - pushl %ss - CFI_ADJUST_CFA_OFFSET 4 - pushl %esp - CFI_ADJUST_CFA_OFFSET 4 - addw $4, (%esp) - /* copy the iret frame of 12 bytes */ - .rept 3 - pushl 16(%esp) - CFI_ADJUST_CFA_OFFSET 4 - .endr - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 - SAVE_ALL - TRACE_IRQS_OFF - FIXUP_ESPFIX_STACK # %eax == %esp - xorl %edx,%edx # zero error code - call do_nmi - RESTORE_REGS - lss 12+4(%esp), %esp # back to espfix stack - CFI_ADJUST_CFA_OFFSET -24 - jmp irq_return - CFI_ENDPROC -KPROBE_END(nmi) - #ifdef CONFIG_PARAVIRT ENTRY(native_iret) iret @@ -916,19 +733,6 @@ ENTRY(native_irq_enable_sysexit) END(native_irq_enable_sysexit) #endif -KPROBE_ENTRY(int3) - RING0_INT_FRAME - pushl $-1 # mark this as an int - CFI_ADJUST_CFA_OFFSET 4 - SAVE_ALL - TRACE_IRQS_OFF - xorl %edx,%edx # zero error code - movl %esp,%eax # pt_regs pointer - call do_int3 - jmp ret_from_exception - CFI_ENDPROC -KPROBE_END(int3) - ENTRY(overflow) RING0_INT_FRAME pushl $0 @@ -993,14 +797,6 @@ ENTRY(stack_segment) CFI_ENDPROC END(stack_segment) -KPROBE_ENTRY(general_protection) - RING0_EC_FRAME - pushl $do_general_protection - CFI_ADJUST_CFA_OFFSET 4 - jmp error_code - CFI_ENDPROC -KPROBE_END(general_protection) - ENTRY(alignment_check) RING0_EC_FRAME pushl $do_alignment_check @@ -1024,7 +820,7 @@ ENTRY(machine_check) RING0_INT_FRAME pushl $0 CFI_ADJUST_CFA_OFFSET 4 - pushl $do_machine_check + pushl machine_check_vector CFI_ADJUST_CFA_OFFSET 4 jmp error_code CFI_ENDPROC @@ -1051,6 +847,7 @@ ENTRY(kernel_thread_helper) push %eax CFI_ADJUST_CFA_OFFSET 4 call do_exit + ud2 # padding for call trace CFI_ENDPROC ENDPROC(kernel_thread_helper) @@ -1149,28 +946,17 @@ ENDPROC(xen_failsafe_callback) #endif /* CONFIG_XEN */ -#ifdef CONFIG_FTRACE +#ifdef CONFIG_FUNCTION_TRACER #ifdef CONFIG_DYNAMIC_FTRACE ENTRY(mcount) - pushl %eax - pushl %ecx - pushl %edx - movl 0xc(%esp), %eax - subl $MCOUNT_INSN_SIZE, %eax - -.globl mcount_call -mcount_call: - call ftrace_stub - - popl %edx - popl %ecx - popl %eax - ret END(mcount) ENTRY(ftrace_caller) + cmpl $0, function_trace_stop + jne ftrace_stub + pushl %eax pushl %ecx pushl %edx @@ -1185,6 +971,11 @@ ftrace_call: popl %edx popl %ecx popl %eax +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +.globl ftrace_graph_call +ftrace_graph_call: + jmp ftrace_stub +#endif .globl ftrace_stub ftrace_stub: @@ -1194,8 +985,18 @@ END(ftrace_caller) #else /* ! CONFIG_DYNAMIC_FTRACE */ ENTRY(mcount) + cmpl $0, function_trace_stop + jne ftrace_stub + cmpl $ftrace_stub, ftrace_trace_function jnz trace +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + cmpl $ftrace_stub, ftrace_graph_return + jnz ftrace_graph_caller + + cmpl $ftrace_graph_entry_stub, ftrace_graph_entry + jnz ftrace_graph_caller +#endif .globl ftrace_stub ftrace_stub: ret @@ -1214,13 +1015,268 @@ trace: popl %edx popl %ecx popl %eax - jmp ftrace_stub END(mcount) #endif /* CONFIG_DYNAMIC_FTRACE */ -#endif /* CONFIG_FTRACE */ +#endif /* CONFIG_FUNCTION_TRACER */ + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +ENTRY(ftrace_graph_caller) + cmpl $0, function_trace_stop + jne ftrace_stub + + pushl %eax + pushl %ecx + pushl %edx + movl 0xc(%esp), %edx + lea 0x4(%ebp), %eax + subl $MCOUNT_INSN_SIZE, %edx + call prepare_ftrace_return + popl %edx + popl %ecx + popl %eax + ret +END(ftrace_graph_caller) + +.globl return_to_handler +return_to_handler: + pushl $0 + pushl %eax + pushl %ecx + pushl %edx + call ftrace_return_to_handler + movl %eax, 0xc(%esp) + popl %edx + popl %ecx + popl %eax + ret +#endif .section .rodata,"a" #include "syscall_table_32.S" syscall_table_size=(.-sys_call_table) + +/* + * Some functions should be protected against kprobes + */ + .pushsection .kprobes.text, "ax" + +ENTRY(page_fault) + RING0_EC_FRAME + pushl $do_page_fault + CFI_ADJUST_CFA_OFFSET 4 + ALIGN +error_code: + /* the function address is in %fs's slot on the stack */ + pushl %es + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET es, 0*/ + pushl %ds + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET ds, 0*/ + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET eax, 0 + pushl %ebp + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ebp, 0 + pushl %edi + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET edi, 0 + pushl %esi + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET esi, 0 + pushl %edx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET edx, 0 + pushl %ecx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ecx, 0 + pushl %ebx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ebx, 0 + cld + pushl %fs + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET fs, 0*/ + movl $(__KERNEL_PERCPU), %ecx + movl %ecx, %fs + UNWIND_ESPFIX_STACK + popl %ecx + CFI_ADJUST_CFA_OFFSET -4 + /*CFI_REGISTER es, ecx*/ + movl PT_FS(%esp), %edi # get the function address + movl PT_ORIG_EAX(%esp), %edx # get the error code + movl $-1, PT_ORIG_EAX(%esp) # no syscall to restart + mov %ecx, PT_FS(%esp) + /*CFI_REL_OFFSET fs, ES*/ + movl $(__USER_DS), %ecx + movl %ecx, %ds + movl %ecx, %es + TRACE_IRQS_OFF + movl %esp,%eax # pt_regs pointer + call *%edi + jmp ret_from_exception + CFI_ENDPROC +END(page_fault) + +/* + * Debug traps and NMI can happen at the one SYSENTER instruction + * that sets up the real kernel stack. Check here, since we can't + * allow the wrong stack to be used. + * + * "TSS_sysenter_sp0+12" is because the NMI/debug handler will have + * already pushed 3 words if it hits on the sysenter instruction: + * eflags, cs and eip. + * + * We just load the right stack, and push the three (known) values + * by hand onto the new stack - while updating the return eip past + * the instruction that would have done it for sysenter. + */ +#define FIX_STACK(offset, ok, label) \ + cmpw $__KERNEL_CS,4(%esp); \ + jne ok; \ +label: \ + movl TSS_sysenter_sp0+offset(%esp),%esp; \ + CFI_DEF_CFA esp, 0; \ + CFI_UNDEFINED eip; \ + pushfl; \ + CFI_ADJUST_CFA_OFFSET 4; \ + pushl $__KERNEL_CS; \ + CFI_ADJUST_CFA_OFFSET 4; \ + pushl $sysenter_past_esp; \ + CFI_ADJUST_CFA_OFFSET 4; \ + CFI_REL_OFFSET eip, 0 + +ENTRY(debug) + RING0_INT_FRAME + cmpl $ia32_sysenter_target,(%esp) + jne debug_stack_correct + FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn) +debug_stack_correct: + pushl $-1 # mark this as an int + CFI_ADJUST_CFA_OFFSET 4 + SAVE_ALL + TRACE_IRQS_OFF + xorl %edx,%edx # error code 0 + movl %esp,%eax # pt_regs pointer + call do_debug + jmp ret_from_exception + CFI_ENDPROC +END(debug) + +/* + * NMI is doubly nasty. It can happen _while_ we're handling + * a debug fault, and the debug fault hasn't yet been able to + * clear up the stack. So we first check whether we got an + * NMI on the sysenter entry path, but after that we need to + * check whether we got an NMI on the debug path where the debug + * fault happened on the sysenter path. + */ +ENTRY(nmi) + RING0_INT_FRAME + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + movl %ss, %eax + cmpw $__ESPFIX_SS, %ax + popl %eax + CFI_ADJUST_CFA_OFFSET -4 + je nmi_espfix_stack + cmpl $ia32_sysenter_target,(%esp) + je nmi_stack_fixup + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + movl %esp,%eax + /* Do not access memory above the end of our stack page, + * it might not exist. + */ + andl $(THREAD_SIZE-1),%eax + cmpl $(THREAD_SIZE-20),%eax + popl %eax + CFI_ADJUST_CFA_OFFSET -4 + jae nmi_stack_correct + cmpl $ia32_sysenter_target,12(%esp) + je nmi_debug_stack_check +nmi_stack_correct: + /* We have a RING0_INT_FRAME here */ + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + SAVE_ALL + TRACE_IRQS_OFF + xorl %edx,%edx # zero error code + movl %esp,%eax # pt_regs pointer + call do_nmi + jmp restore_nocheck_notrace + CFI_ENDPROC + +nmi_stack_fixup: + RING0_INT_FRAME + FIX_STACK(12,nmi_stack_correct, 1) + jmp nmi_stack_correct + +nmi_debug_stack_check: + /* We have a RING0_INT_FRAME here */ + cmpw $__KERNEL_CS,16(%esp) + jne nmi_stack_correct + cmpl $debug,(%esp) + jb nmi_stack_correct + cmpl $debug_esp_fix_insn,(%esp) + ja nmi_stack_correct + FIX_STACK(24,nmi_stack_correct, 1) + jmp nmi_stack_correct + +nmi_espfix_stack: + /* We have a RING0_INT_FRAME here. + * + * create the pointer to lss back + */ + pushl %ss + CFI_ADJUST_CFA_OFFSET 4 + pushl %esp + CFI_ADJUST_CFA_OFFSET 4 + addw $4, (%esp) + /* copy the iret frame of 12 bytes */ + .rept 3 + pushl 16(%esp) + CFI_ADJUST_CFA_OFFSET 4 + .endr + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + SAVE_ALL + TRACE_IRQS_OFF + FIXUP_ESPFIX_STACK # %eax == %esp + xorl %edx,%edx # zero error code + call do_nmi + RESTORE_REGS + lss 12+4(%esp), %esp # back to espfix stack + CFI_ADJUST_CFA_OFFSET -24 + jmp irq_return + CFI_ENDPROC +END(nmi) + +ENTRY(int3) + RING0_INT_FRAME + pushl $-1 # mark this as an int + CFI_ADJUST_CFA_OFFSET 4 + SAVE_ALL + TRACE_IRQS_OFF + xorl %edx,%edx # zero error code + movl %esp,%eax # pt_regs pointer + call do_int3 + jmp ret_from_exception + CFI_ENDPROC +END(int3) + +ENTRY(general_protection) + RING0_EC_FRAME + pushl $do_general_protection + CFI_ADJUST_CFA_OFFSET 4 + jmp error_code + CFI_ENDPROC +END(general_protection) + +/* + * End of kprobes section + */ + .popsection diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 1db6ce4314e1..e28c7a987793 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -11,15 +11,15 @@ * * NOTE: This code handles signal-recognition, which happens every time * after an interrupt and after each system call. - * - * Normal syscalls and interrupts don't save a full stack frame, this is + * + * Normal syscalls and interrupts don't save a full stack frame, this is * only done for syscall tracing, signals or fork/exec et.al. - * - * A note on terminology: - * - top of stack: Architecture defined interrupt frame from SS to RIP - * at the top of the kernel process stack. + * + * A note on terminology: + * - top of stack: Architecture defined interrupt frame from SS to RIP + * at the top of the kernel process stack. * - partial stack frame: partially saved registers upto R11. - * - full stack frame: Like partial stack frame, but all register saved. + * - full stack frame: Like partial stack frame, but all register saved. * * Some macro usage: * - CFI macros are used to generate dwarf2 unwind information for better @@ -60,50 +60,17 @@ #define __AUDIT_ARCH_LE 0x40000000 .code64 - -#ifdef CONFIG_FTRACE +#ifdef CONFIG_FUNCTION_TRACER #ifdef CONFIG_DYNAMIC_FTRACE ENTRY(mcount) - - subq $0x38, %rsp - movq %rax, (%rsp) - movq %rcx, 8(%rsp) - movq %rdx, 16(%rsp) - movq %rsi, 24(%rsp) - movq %rdi, 32(%rsp) - movq %r8, 40(%rsp) - movq %r9, 48(%rsp) - - movq 0x38(%rsp), %rdi - subq $MCOUNT_INSN_SIZE, %rdi - -.globl mcount_call -mcount_call: - call ftrace_stub - - movq 48(%rsp), %r9 - movq 40(%rsp), %r8 - movq 32(%rsp), %rdi - movq 24(%rsp), %rsi - movq 16(%rsp), %rdx - movq 8(%rsp), %rcx - movq (%rsp), %rax - addq $0x38, %rsp - retq END(mcount) ENTRY(ftrace_caller) + cmpl $0, function_trace_stop + jne ftrace_stub - /* taken from glibc */ - subq $0x38, %rsp - movq %rax, (%rsp) - movq %rcx, 8(%rsp) - movq %rdx, 16(%rsp) - movq %rsi, 24(%rsp) - movq %rdi, 32(%rsp) - movq %r8, 40(%rsp) - movq %r9, 48(%rsp) + MCOUNT_SAVE_FRAME movq 0x38(%rsp), %rdi movq 8(%rbp), %rsi @@ -113,14 +80,13 @@ ENTRY(ftrace_caller) ftrace_call: call ftrace_stub - movq 48(%rsp), %r9 - movq 40(%rsp), %r8 - movq 32(%rsp), %rdi - movq 24(%rsp), %rsi - movq 16(%rsp), %rdx - movq 8(%rsp), %rcx - movq (%rsp), %rax - addq $0x38, %rsp + MCOUNT_RESTORE_FRAME + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +.globl ftrace_graph_call +ftrace_graph_call: + jmp ftrace_stub +#endif .globl ftrace_stub ftrace_stub: @@ -129,15 +95,63 @@ END(ftrace_caller) #else /* ! CONFIG_DYNAMIC_FTRACE */ ENTRY(mcount) + cmpl $0, function_trace_stop + jne ftrace_stub + cmpq $ftrace_stub, ftrace_trace_function jnz trace + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + cmpq $ftrace_stub, ftrace_graph_return + jnz ftrace_graph_caller + + cmpq $ftrace_graph_entry_stub, ftrace_graph_entry + jnz ftrace_graph_caller +#endif + .globl ftrace_stub ftrace_stub: retq trace: - /* taken from glibc */ - subq $0x38, %rsp + MCOUNT_SAVE_FRAME + + movq 0x38(%rsp), %rdi + movq 8(%rbp), %rsi + subq $MCOUNT_INSN_SIZE, %rdi + + call *ftrace_trace_function + + MCOUNT_RESTORE_FRAME + + jmp ftrace_stub +END(mcount) +#endif /* CONFIG_DYNAMIC_FTRACE */ +#endif /* CONFIG_FUNCTION_TRACER */ + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +ENTRY(ftrace_graph_caller) + cmpl $0, function_trace_stop + jne ftrace_stub + + MCOUNT_SAVE_FRAME + + leaq 8(%rbp), %rdi + movq 0x38(%rsp), %rsi + subq $MCOUNT_INSN_SIZE, %rsi + + call prepare_ftrace_return + + MCOUNT_RESTORE_FRAME + + retq +END(ftrace_graph_caller) + + +.globl return_to_handler +return_to_handler: + subq $80, %rsp + movq %rax, (%rsp) movq %rcx, 8(%rsp) movq %rdx, 16(%rsp) @@ -145,13 +159,14 @@ trace: movq %rdi, 32(%rsp) movq %r8, 40(%rsp) movq %r9, 48(%rsp) + movq %r10, 56(%rsp) + movq %r11, 64(%rsp) - movq 0x38(%rsp), %rdi - movq 8(%rbp), %rsi - subq $MCOUNT_INSN_SIZE, %rdi - - call *ftrace_trace_function + call ftrace_return_to_handler + movq %rax, 72(%rsp) + movq 64(%rsp), %r11 + movq 56(%rsp), %r10 movq 48(%rsp), %r9 movq 40(%rsp), %r8 movq 32(%rsp), %rdi @@ -159,16 +174,14 @@ trace: movq 16(%rsp), %rdx movq 8(%rsp), %rcx movq (%rsp), %rax - addq $0x38, %rsp + addq $72, %rsp + retq +#endif - jmp ftrace_stub -END(mcount) -#endif /* CONFIG_DYNAMIC_FTRACE */ -#endif /* CONFIG_FTRACE */ #ifndef CONFIG_PREEMPT #define retint_kernel retint_restore_args -#endif +#endif #ifdef CONFIG_PARAVIRT ENTRY(native_usergs_sysret64) @@ -187,29 +200,29 @@ ENTRY(native_usergs_sysret64) .endm /* - * C code is not supposed to know about undefined top of stack. Every time - * a C function with an pt_regs argument is called from the SYSCALL based + * C code is not supposed to know about undefined top of stack. Every time + * a C function with an pt_regs argument is called from the SYSCALL based * fast path FIXUP_TOP_OF_STACK is needed. * RESTORE_TOP_OF_STACK syncs the syscall state after any possible ptregs * manipulation. - */ - - /* %rsp:at FRAMEEND */ - .macro FIXUP_TOP_OF_STACK tmp - movq %gs:pda_oldrsp,\tmp - movq \tmp,RSP(%rsp) - movq $__USER_DS,SS(%rsp) - movq $__USER_CS,CS(%rsp) - movq $-1,RCX(%rsp) - movq R11(%rsp),\tmp /* get eflags */ - movq \tmp,EFLAGS(%rsp) + */ + + /* %rsp:at FRAMEEND */ + .macro FIXUP_TOP_OF_STACK tmp offset=0 + movq %gs:pda_oldrsp,\tmp + movq \tmp,RSP+\offset(%rsp) + movq $__USER_DS,SS+\offset(%rsp) + movq $__USER_CS,CS+\offset(%rsp) + movq $-1,RCX+\offset(%rsp) + movq R11+\offset(%rsp),\tmp /* get eflags */ + movq \tmp,EFLAGS+\offset(%rsp) .endm - .macro RESTORE_TOP_OF_STACK tmp,offset=0 - movq RSP-\offset(%rsp),\tmp - movq \tmp,%gs:pda_oldrsp - movq EFLAGS-\offset(%rsp),\tmp - movq \tmp,R11-\offset(%rsp) + .macro RESTORE_TOP_OF_STACK tmp offset=0 + movq RSP+\offset(%rsp),\tmp + movq \tmp,%gs:pda_oldrsp + movq EFLAGS+\offset(%rsp),\tmp + movq \tmp,R11+\offset(%rsp) .endm .macro FAKE_STACK_FRAME child_rip @@ -221,7 +234,7 @@ ENTRY(native_usergs_sysret64) pushq %rax /* rsp */ CFI_ADJUST_CFA_OFFSET 8 CFI_REL_OFFSET rsp,0 - pushq $(1<<9) /* eflags - interrupts on */ + pushq $X86_EFLAGS_IF /* eflags - interrupts on */ CFI_ADJUST_CFA_OFFSET 8 /*CFI_REL_OFFSET rflags,0*/ pushq $__KERNEL_CS /* cs */ @@ -239,62 +252,184 @@ ENTRY(native_usergs_sysret64) CFI_ADJUST_CFA_OFFSET -(6*8) .endm - .macro CFI_DEFAULT_STACK start=1 +/* + * initial frame state for interrupts (and exceptions without error code) + */ + .macro EMPTY_FRAME start=1 offset=0 .if \start - CFI_STARTPROC simple + CFI_STARTPROC simple CFI_SIGNAL_FRAME - CFI_DEF_CFA rsp,SS+8 + CFI_DEF_CFA rsp,8+\offset .else - CFI_DEF_CFA_OFFSET SS+8 + CFI_DEF_CFA_OFFSET 8+\offset .endif - CFI_REL_OFFSET r15,R15 - CFI_REL_OFFSET r14,R14 - CFI_REL_OFFSET r13,R13 - CFI_REL_OFFSET r12,R12 - CFI_REL_OFFSET rbp,RBP - CFI_REL_OFFSET rbx,RBX - CFI_REL_OFFSET r11,R11 - CFI_REL_OFFSET r10,R10 - CFI_REL_OFFSET r9,R9 - CFI_REL_OFFSET r8,R8 - CFI_REL_OFFSET rax,RAX - CFI_REL_OFFSET rcx,RCX - CFI_REL_OFFSET rdx,RDX - CFI_REL_OFFSET rsi,RSI - CFI_REL_OFFSET rdi,RDI - CFI_REL_OFFSET rip,RIP - /*CFI_REL_OFFSET cs,CS*/ - /*CFI_REL_OFFSET rflags,EFLAGS*/ - CFI_REL_OFFSET rsp,RSP - /*CFI_REL_OFFSET ss,SS*/ .endm + +/* + * initial frame state for interrupts (and exceptions without error code) + */ + .macro INTR_FRAME start=1 offset=0 + EMPTY_FRAME \start, SS+8+\offset-RIP + /*CFI_REL_OFFSET ss, SS+\offset-RIP*/ + CFI_REL_OFFSET rsp, RSP+\offset-RIP + /*CFI_REL_OFFSET rflags, EFLAGS+\offset-RIP*/ + /*CFI_REL_OFFSET cs, CS+\offset-RIP*/ + CFI_REL_OFFSET rip, RIP+\offset-RIP + .endm + +/* + * initial frame state for exceptions with error code (and interrupts + * with vector already pushed) + */ + .macro XCPT_FRAME start=1 offset=0 + INTR_FRAME \start, RIP+\offset-ORIG_RAX + /*CFI_REL_OFFSET orig_rax, ORIG_RAX-ORIG_RAX*/ + .endm + /* - * A newly forked process directly context switches into this. - */ -/* rdi: prev */ + * frame that enables calling into C. + */ + .macro PARTIAL_FRAME start=1 offset=0 + XCPT_FRAME \start, ORIG_RAX+\offset-ARGOFFSET + CFI_REL_OFFSET rdi, RDI+\offset-ARGOFFSET + CFI_REL_OFFSET rsi, RSI+\offset-ARGOFFSET + CFI_REL_OFFSET rdx, RDX+\offset-ARGOFFSET + CFI_REL_OFFSET rcx, RCX+\offset-ARGOFFSET + CFI_REL_OFFSET rax, RAX+\offset-ARGOFFSET + CFI_REL_OFFSET r8, R8+\offset-ARGOFFSET + CFI_REL_OFFSET r9, R9+\offset-ARGOFFSET + CFI_REL_OFFSET r10, R10+\offset-ARGOFFSET + CFI_REL_OFFSET r11, R11+\offset-ARGOFFSET + .endm + +/* + * frame that enables passing a complete pt_regs to a C function. + */ + .macro DEFAULT_FRAME start=1 offset=0 + PARTIAL_FRAME \start, R11+\offset-R15 + CFI_REL_OFFSET rbx, RBX+\offset + CFI_REL_OFFSET rbp, RBP+\offset + CFI_REL_OFFSET r12, R12+\offset + CFI_REL_OFFSET r13, R13+\offset + CFI_REL_OFFSET r14, R14+\offset + CFI_REL_OFFSET r15, R15+\offset + .endm + +/* save partial stack frame */ +ENTRY(save_args) + XCPT_FRAME + cld + movq_cfi rdi, RDI+16-ARGOFFSET + movq_cfi rsi, RSI+16-ARGOFFSET + movq_cfi rdx, RDX+16-ARGOFFSET + movq_cfi rcx, RCX+16-ARGOFFSET + movq_cfi rax, RAX+16-ARGOFFSET + movq_cfi r8, R8+16-ARGOFFSET + movq_cfi r9, R9+16-ARGOFFSET + movq_cfi r10, R10+16-ARGOFFSET + movq_cfi r11, R11+16-ARGOFFSET + + leaq -ARGOFFSET+16(%rsp),%rdi /* arg1 for handler */ + movq_cfi rbp, 8 /* push %rbp */ + leaq 8(%rsp), %rbp /* mov %rsp, %ebp */ + testl $3, CS(%rdi) + je 1f + SWAPGS + /* + * irqcount is used to check if a CPU is already on an interrupt stack + * or not. While this is essentially redundant with preempt_count it is + * a little cheaper to use a separate counter in the PDA (short of + * moving irq_enter into assembly, which would be too much work) + */ +1: incl %gs:pda_irqcount + jne 2f + popq_cfi %rax /* move return address... */ + mov %gs:pda_irqstackptr,%rsp + EMPTY_FRAME 0 + pushq_cfi %rax /* ... to the new stack */ + /* + * We entered an interrupt context - irqs are off: + */ +2: TRACE_IRQS_OFF + ret + CFI_ENDPROC +END(save_args) + +ENTRY(save_rest) + PARTIAL_FRAME 1 REST_SKIP+8 + movq 5*8+16(%rsp), %r11 /* save return address */ + movq_cfi rbx, RBX+16 + movq_cfi rbp, RBP+16 + movq_cfi r12, R12+16 + movq_cfi r13, R13+16 + movq_cfi r14, R14+16 + movq_cfi r15, R15+16 + movq %r11, 8(%rsp) /* return address */ + FIXUP_TOP_OF_STACK %r11, 16 + ret + CFI_ENDPROC +END(save_rest) + +/* save complete stack frame */ +ENTRY(save_paranoid) + XCPT_FRAME 1 RDI+8 + cld + movq_cfi rdi, RDI+8 + movq_cfi rsi, RSI+8 + movq_cfi rdx, RDX+8 + movq_cfi rcx, RCX+8 + movq_cfi rax, RAX+8 + movq_cfi r8, R8+8 + movq_cfi r9, R9+8 + movq_cfi r10, R10+8 + movq_cfi r11, R11+8 + movq_cfi rbx, RBX+8 + movq_cfi rbp, RBP+8 + movq_cfi r12, R12+8 + movq_cfi r13, R13+8 + movq_cfi r14, R14+8 + movq_cfi r15, R15+8 + movl $1,%ebx + movl $MSR_GS_BASE,%ecx + rdmsr + testl %edx,%edx + js 1f /* negative -> in kernel */ + SWAPGS + xorl %ebx,%ebx +1: ret + CFI_ENDPROC +END(save_paranoid) + +/* + * A newly forked process directly context switches into this address. + * + * rdi: prev task we switched from + */ ENTRY(ret_from_fork) - CFI_DEFAULT_STACK + DEFAULT_FRAME + push kernel_eflags(%rip) CFI_ADJUST_CFA_OFFSET 8 - popf # reset kernel eflags + popf # reset kernel eflags CFI_ADJUST_CFA_OFFSET -8 - call schedule_tail + + call schedule_tail # rdi: 'prev' task parameter + GET_THREAD_INFO(%rcx) - testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),TI_flags(%rcx) - jnz rff_trace -rff_action: + + CFI_REMEMBER_STATE RESTORE_REST - testl $3,CS-ARGOFFSET(%rsp) # from kernel_thread? + + testl $3, CS-ARGOFFSET(%rsp) # from kernel_thread? je int_ret_from_sys_call - testl $_TIF_IA32,TI_flags(%rcx) + + testl $_TIF_IA32, TI_flags(%rcx) # 32-bit compat task needs IRET jnz int_ret_from_sys_call - RESTORE_TOP_OF_STACK %rdi,ARGOFFSET - jmp ret_from_sys_call -rff_trace: - movq %rsp,%rdi - call syscall_trace_leave - GET_THREAD_INFO(%rcx) - jmp rff_action + + RESTORE_TOP_OF_STACK %rdi, -ARGOFFSET + jmp ret_from_sys_call # go to the SYSRET fastpath + + CFI_RESTORE_STATE CFI_ENDPROC END(ret_from_fork) @@ -304,20 +439,20 @@ END(ret_from_fork) * SYSCALL does not save anything on the stack and does not change the * stack pointer. */ - + /* - * Register setup: + * Register setup: * rax system call number * rdi arg0 - * rcx return address for syscall/sysret, C arg3 + * rcx return address for syscall/sysret, C arg3 * rsi arg1 - * rdx arg2 + * rdx arg2 * r10 arg3 (--> moved to rcx for C) * r8 arg4 * r9 arg5 * r11 eflags for syscall/sysret, temporary for C - * r12-r15,rbp,rbx saved by C code, not touched. - * + * r12-r15,rbp,rbx saved by C code, not touched. + * * Interrupts are off on entry. * Only called from user space. * @@ -327,7 +462,7 @@ END(ret_from_fork) * When user can change the frames always force IRET. That is because * it deals with uncanonical addresses better. SYSRET has trouble * with them due to bugs in both AMD and Intel CPUs. - */ + */ ENTRY(system_call) CFI_STARTPROC simple @@ -343,7 +478,7 @@ ENTRY(system_call) */ ENTRY(system_call_after_swapgs) - movq %rsp,%gs:pda_oldrsp + movq %rsp,%gs:pda_oldrsp movq %gs:pda_kernelstack,%rsp /* * No need to follow this irqs off/on section - it's straight @@ -351,7 +486,7 @@ ENTRY(system_call_after_swapgs) */ ENABLE_INTERRUPTS(CLBR_NONE) SAVE_ARGS 8,1 - movq %rax,ORIG_RAX-ARGOFFSET(%rsp) + movq %rax,ORIG_RAX-ARGOFFSET(%rsp) movq %rcx,RIP-ARGOFFSET(%rsp) CFI_REL_OFFSET rip,RIP-ARGOFFSET GET_THREAD_INFO(%rcx) @@ -365,19 +500,19 @@ system_call_fastpath: movq %rax,RAX-ARGOFFSET(%rsp) /* * Syscall return path ending with SYSRET (fast path) - * Has incomplete stack frame and undefined top of stack. - */ + * Has incomplete stack frame and undefined top of stack. + */ ret_from_sys_call: movl $_TIF_ALLWORK_MASK,%edi /* edi: flagmask */ -sysret_check: +sysret_check: LOCKDEP_SYS_EXIT GET_THREAD_INFO(%rcx) DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF movl TI_flags(%rcx),%edx andl %edi,%edx - jnz sysret_careful + jnz sysret_careful CFI_REMEMBER_STATE /* * sysretq will re-enable interrupts: @@ -392,7 +527,7 @@ sysret_check: CFI_RESTORE_STATE /* Handle reschedules */ - /* edx: work, edi: workmask */ + /* edx: work, edi: workmask */ sysret_careful: bt $TIF_NEED_RESCHED,%edx jnc sysret_signal @@ -405,7 +540,7 @@ sysret_careful: CFI_ADJUST_CFA_OFFSET -8 jmp sysret_check - /* Handle a signal */ + /* Handle a signal */ sysret_signal: TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) @@ -414,17 +549,20 @@ sysret_signal: jc sysret_audit #endif /* edx: work flags (arg3) */ - leaq do_notify_resume(%rip),%rax leaq -ARGOFFSET(%rsp),%rdi # &pt_regs -> arg1 xorl %esi,%esi # oldset -> arg2 - call ptregscall_common + SAVE_REST + FIXUP_TOP_OF_STACK %r11 + call do_notify_resume + RESTORE_TOP_OF_STACK %r11 + RESTORE_REST movl $_TIF_WORK_MASK,%edi /* Use IRET because user could have changed frame. This works because ptregscall_common has called FIXUP_TOP_OF_STACK. */ DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF jmp int_with_check - + badsys: movq $-ENOSYS,RAX-ARGOFFSET(%rsp) jmp ret_from_sys_call @@ -463,7 +601,7 @@ sysret_audit: #endif /* CONFIG_AUDITSYSCALL */ /* Do syscall tracing */ -tracesys: +tracesys: #ifdef CONFIG_AUDITSYSCALL testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%rcx) jz auditsys @@ -486,8 +624,8 @@ tracesys: call *sys_call_table(,%rax,8) movq %rax,RAX-ARGOFFSET(%rsp) /* Use IRET because user could have changed frame */ - -/* + +/* * Syscall return path ending with IRET. * Has correct top of stack, but partial stack frame. */ @@ -531,18 +669,18 @@ int_very_careful: TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) SAVE_REST - /* Check for syscall exit trace */ + /* Check for syscall exit trace */ testl $_TIF_WORK_SYSCALL_EXIT,%edx jz int_signal pushq %rdi CFI_ADJUST_CFA_OFFSET 8 - leaq 8(%rsp),%rdi # &ptregs -> arg1 + leaq 8(%rsp),%rdi # &ptregs -> arg1 call syscall_trace_leave popq %rdi CFI_ADJUST_CFA_OFFSET -8 andl $~(_TIF_WORK_SYSCALL_EXIT|_TIF_SYSCALL_EMU),%edi jmp int_restore_rest - + int_signal: testl $_TIF_DO_NOTIFY_MASK,%edx jz 1f @@ -557,22 +695,24 @@ int_restore_rest: jmp int_with_check CFI_ENDPROC END(system_call) - -/* + +/* * Certain special system calls that need to save a complete full stack frame. - */ - + */ .macro PTREGSCALL label,func,arg - .globl \label -\label: - leaq \func(%rip),%rax - leaq -ARGOFFSET+8(%rsp),\arg /* 8 for return address */ - jmp ptregscall_common +ENTRY(\label) + PARTIAL_FRAME 1 8 /* offset 8: return address */ + subq $REST_SKIP, %rsp + CFI_ADJUST_CFA_OFFSET REST_SKIP + call save_rest + DEFAULT_FRAME 0 8 /* offset 8: return address */ + leaq 8(%rsp), \arg /* pt_regs pointer */ + call \func + jmp ptregscall_common + CFI_ENDPROC END(\label) .endm - CFI_STARTPROC - PTREGSCALL stub_clone, sys_clone, %r8 PTREGSCALL stub_fork, sys_fork, %rdi PTREGSCALL stub_vfork, sys_vfork, %rdi @@ -580,25 +720,18 @@ END(\label) PTREGSCALL stub_iopl, sys_iopl, %rsi ENTRY(ptregscall_common) - popq %r11 - CFI_ADJUST_CFA_OFFSET -8 - CFI_REGISTER rip, r11 - SAVE_REST - movq %r11, %r15 - CFI_REGISTER rip, r15 - FIXUP_TOP_OF_STACK %r11 - call *%rax - RESTORE_TOP_OF_STACK %r11 - movq %r15, %r11 - CFI_REGISTER rip, r11 - RESTORE_REST - pushq %r11 - CFI_ADJUST_CFA_OFFSET 8 - CFI_REL_OFFSET rip, 0 - ret + DEFAULT_FRAME 1 8 /* offset 8: return address */ + RESTORE_TOP_OF_STACK %r11, 8 + movq_cfi_restore R15+8, r15 + movq_cfi_restore R14+8, r14 + movq_cfi_restore R13+8, r13 + movq_cfi_restore R12+8, r12 + movq_cfi_restore RBP+8, rbp + movq_cfi_restore RBX+8, rbx + ret $REST_SKIP /* pop extended registers */ CFI_ENDPROC END(ptregscall_common) - + ENTRY(stub_execve) CFI_STARTPROC popq %r11 @@ -614,11 +747,11 @@ ENTRY(stub_execve) jmp int_ret_from_sys_call CFI_ENDPROC END(stub_execve) - + /* * sigreturn is special because it needs to restore all registers on return. * This cannot be done with SYSRET, so use the IRET return path instead. - */ + */ ENTRY(stub_rt_sigreturn) CFI_STARTPROC addq $8, %rsp @@ -634,70 +767,70 @@ ENTRY(stub_rt_sigreturn) END(stub_rt_sigreturn) /* - * initial frame state for interrupts and exceptions + * Build the entry stubs and pointer table with some assembler magic. + * We pack 7 stubs into a single 32-byte chunk, which will fit in a + * single cache line on all modern x86 implementations. */ - .macro _frame ref - CFI_STARTPROC simple - CFI_SIGNAL_FRAME - CFI_DEF_CFA rsp,SS+8-\ref - /*CFI_REL_OFFSET ss,SS-\ref*/ - CFI_REL_OFFSET rsp,RSP-\ref - /*CFI_REL_OFFSET rflags,EFLAGS-\ref*/ - /*CFI_REL_OFFSET cs,CS-\ref*/ - CFI_REL_OFFSET rip,RIP-\ref - .endm + .section .init.rodata,"a" +ENTRY(interrupt) + .text + .p2align 5 + .p2align CONFIG_X86_L1_CACHE_SHIFT +ENTRY(irq_entries_start) + INTR_FRAME +vector=FIRST_EXTERNAL_VECTOR +.rept (NR_VECTORS-FIRST_EXTERNAL_VECTOR+6)/7 + .balign 32 + .rept 7 + .if vector < NR_VECTORS + .if vector <> FIRST_EXTERNAL_VECTOR + CFI_ADJUST_CFA_OFFSET -8 + .endif +1: pushq $(~vector+0x80) /* Note: always in signed byte range */ + CFI_ADJUST_CFA_OFFSET 8 + .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6 + jmp 2f + .endif + .previous + .quad 1b + .text +vector=vector+1 + .endif + .endr +2: jmp common_interrupt +.endr + CFI_ENDPROC +END(irq_entries_start) -/* initial frame state for interrupts (and exceptions without error code) */ -#define INTR_FRAME _frame RIP -/* initial frame state for exceptions with error code (and interrupts with - vector already pushed) */ -#define XCPT_FRAME _frame ORIG_RAX +.previous +END(interrupt) +.previous -/* +/* * Interrupt entry/exit. * * Interrupt entry points save only callee clobbered registers in fast path. - * - * Entry runs with interrupts off. - */ + * + * Entry runs with interrupts off. + */ -/* 0(%rsp): interrupt number */ +/* 0(%rsp): ~(interrupt number) */ .macro interrupt func - cld - SAVE_ARGS - leaq -ARGOFFSET(%rsp),%rdi # arg1 for handler - pushq %rbp - /* - * Save rbp twice: One is for marking the stack frame, as usual, and the - * other, to fill pt_regs properly. This is because bx comes right - * before the last saved register in that structure, and not bp. If the - * base pointer were in the place bx is today, this would not be needed. - */ - movq %rbp, -8(%rsp) - CFI_ADJUST_CFA_OFFSET 8 - CFI_REL_OFFSET rbp, 0 - movq %rsp,%rbp - CFI_DEF_CFA_REGISTER rbp - testl $3,CS(%rdi) - je 1f - SWAPGS - /* irqcount is used to check if a CPU is already on an interrupt - stack or not. While this is essentially redundant with preempt_count - it is a little cheaper to use a separate counter in the PDA - (short of moving irq_enter into assembly, which would be too - much work) */ -1: incl %gs:pda_irqcount - cmoveq %gs:pda_irqstackptr,%rsp - push %rbp # backlink for old unwinder - /* - * We entered an interrupt context - irqs are off: - */ - TRACE_IRQS_OFF + subq $10*8, %rsp + CFI_ADJUST_CFA_OFFSET 10*8 + call save_args + PARTIAL_FRAME 0 call \func .endm -ENTRY(common_interrupt) + /* + * The interrupt stubs push (~vector+0x80) onto the stack and + * then jump to common_interrupt. + */ + .p2align CONFIG_X86_L1_CACHE_SHIFT +common_interrupt: XCPT_FRAME + addq $-0x80,(%rsp) /* Adjust vector to [-256,-1] range */ interrupt do_IRQ /* 0(%rsp): oldrsp-ARGOFFSET */ ret_from_intr: @@ -711,12 +844,12 @@ exit_intr: GET_THREAD_INFO(%rcx) testl $3,CS-ARGOFFSET(%rsp) je retint_kernel - + /* Interrupt came from user space */ /* * Has a correct top of stack, but a partial stack frame * %rcx: thread info. Interrupts off. - */ + */ retint_with_reschedule: movl $_TIF_WORK_MASK,%edi retint_check: @@ -789,20 +922,20 @@ retint_careful: pushq %rdi CFI_ADJUST_CFA_OFFSET 8 call schedule - popq %rdi + popq %rdi CFI_ADJUST_CFA_OFFSET -8 GET_THREAD_INFO(%rcx) DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF jmp retint_check - + retint_signal: testl $_TIF_DO_NOTIFY_MASK,%edx jz retint_swapgs TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) SAVE_REST - movq $-1,ORIG_RAX(%rsp) + movq $-1,ORIG_RAX(%rsp) xorl %esi,%esi # oldset movq %rsp,%rdi # &pt_regs call do_notify_resume @@ -824,324 +957,211 @@ ENTRY(retint_kernel) jnc retint_restore_args call preempt_schedule_irq jmp exit_intr -#endif +#endif CFI_ENDPROC END(common_interrupt) - + /* * APIC interrupts. - */ - .macro apicinterrupt num,func + */ +.macro apicinterrupt num sym do_sym +ENTRY(\sym) INTR_FRAME pushq $~(\num) CFI_ADJUST_CFA_OFFSET 8 - interrupt \func + interrupt \do_sym jmp ret_from_intr CFI_ENDPROC - .endm - -ENTRY(thermal_interrupt) - apicinterrupt THERMAL_APIC_VECTOR,smp_thermal_interrupt -END(thermal_interrupt) - -ENTRY(threshold_interrupt) - apicinterrupt THRESHOLD_APIC_VECTOR,mce_threshold_interrupt -END(threshold_interrupt) - -#ifdef CONFIG_SMP -ENTRY(reschedule_interrupt) - apicinterrupt RESCHEDULE_VECTOR,smp_reschedule_interrupt -END(reschedule_interrupt) - - .macro INVALIDATE_ENTRY num -ENTRY(invalidate_interrupt\num) - apicinterrupt INVALIDATE_TLB_VECTOR_START+\num,smp_invalidate_interrupt -END(invalidate_interrupt\num) - .endm +END(\sym) +.endm - INVALIDATE_ENTRY 0 - INVALIDATE_ENTRY 1 - INVALIDATE_ENTRY 2 - INVALIDATE_ENTRY 3 - INVALIDATE_ENTRY 4 - INVALIDATE_ENTRY 5 - INVALIDATE_ENTRY 6 - INVALIDATE_ENTRY 7 - -ENTRY(call_function_interrupt) - apicinterrupt CALL_FUNCTION_VECTOR,smp_call_function_interrupt -END(call_function_interrupt) -ENTRY(call_function_single_interrupt) - apicinterrupt CALL_FUNCTION_SINGLE_VECTOR,smp_call_function_single_interrupt -END(call_function_single_interrupt) -ENTRY(irq_move_cleanup_interrupt) - apicinterrupt IRQ_MOVE_CLEANUP_VECTOR,smp_irq_move_cleanup_interrupt -END(irq_move_cleanup_interrupt) +#ifdef CONFIG_SMP +apicinterrupt IRQ_MOVE_CLEANUP_VECTOR \ + irq_move_cleanup_interrupt smp_irq_move_cleanup_interrupt #endif -ENTRY(apic_timer_interrupt) - apicinterrupt LOCAL_TIMER_VECTOR,smp_apic_timer_interrupt -END(apic_timer_interrupt) +apicinterrupt UV_BAU_MESSAGE \ + uv_bau_message_intr1 uv_bau_message_interrupt +apicinterrupt LOCAL_TIMER_VECTOR \ + apic_timer_interrupt smp_apic_timer_interrupt + +#ifdef CONFIG_SMP +apicinterrupt INVALIDATE_TLB_VECTOR_START+0 \ + invalidate_interrupt0 smp_invalidate_interrupt +apicinterrupt INVALIDATE_TLB_VECTOR_START+1 \ + invalidate_interrupt1 smp_invalidate_interrupt +apicinterrupt INVALIDATE_TLB_VECTOR_START+2 \ + invalidate_interrupt2 smp_invalidate_interrupt +apicinterrupt INVALIDATE_TLB_VECTOR_START+3 \ + invalidate_interrupt3 smp_invalidate_interrupt +apicinterrupt INVALIDATE_TLB_VECTOR_START+4 \ + invalidate_interrupt4 smp_invalidate_interrupt +apicinterrupt INVALIDATE_TLB_VECTOR_START+5 \ + invalidate_interrupt5 smp_invalidate_interrupt +apicinterrupt INVALIDATE_TLB_VECTOR_START+6 \ + invalidate_interrupt6 smp_invalidate_interrupt +apicinterrupt INVALIDATE_TLB_VECTOR_START+7 \ + invalidate_interrupt7 smp_invalidate_interrupt +#endif -ENTRY(uv_bau_message_intr1) - apicinterrupt 220,uv_bau_message_interrupt -END(uv_bau_message_intr1) +apicinterrupt THRESHOLD_APIC_VECTOR \ + threshold_interrupt mce_threshold_interrupt +apicinterrupt THERMAL_APIC_VECTOR \ + thermal_interrupt smp_thermal_interrupt + +#ifdef CONFIG_SMP +apicinterrupt CALL_FUNCTION_SINGLE_VECTOR \ + call_function_single_interrupt smp_call_function_single_interrupt +apicinterrupt CALL_FUNCTION_VECTOR \ + call_function_interrupt smp_call_function_interrupt +apicinterrupt RESCHEDULE_VECTOR \ + reschedule_interrupt smp_reschedule_interrupt +#endif -ENTRY(error_interrupt) - apicinterrupt ERROR_APIC_VECTOR,smp_error_interrupt -END(error_interrupt) +apicinterrupt ERROR_APIC_VECTOR \ + error_interrupt smp_error_interrupt +apicinterrupt SPURIOUS_APIC_VECTOR \ + spurious_interrupt smp_spurious_interrupt -ENTRY(spurious_interrupt) - apicinterrupt SPURIOUS_APIC_VECTOR,smp_spurious_interrupt -END(spurious_interrupt) - /* * Exception entry points. - */ - .macro zeroentry sym + */ +.macro zeroentry sym do_sym +ENTRY(\sym) INTR_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME - pushq $0 /* push error code/oldrax */ - CFI_ADJUST_CFA_OFFSET 8 - pushq %rax /* push real oldrax to the rdi slot */ - CFI_ADJUST_CFA_OFFSET 8 - CFI_REL_OFFSET rax,0 - leaq \sym(%rip),%rax - jmp error_entry + pushq_cfi $-1 /* ORIG_RAX: no syscall to restart */ + subq $15*8,%rsp + CFI_ADJUST_CFA_OFFSET 15*8 + call error_entry + DEFAULT_FRAME 0 + movq %rsp,%rdi /* pt_regs pointer */ + xorl %esi,%esi /* no error code */ + call \do_sym + jmp error_exit /* %ebx: no swapgs flag */ CFI_ENDPROC - .endm +END(\sym) +.endm - .macro errorentry sym - XCPT_FRAME +.macro paranoidzeroentry sym do_sym +ENTRY(\sym) + INTR_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME - pushq %rax + pushq $-1 /* ORIG_RAX: no syscall to restart */ CFI_ADJUST_CFA_OFFSET 8 - CFI_REL_OFFSET rax,0 - leaq \sym(%rip),%rax - jmp error_entry + subq $15*8, %rsp + call save_paranoid + TRACE_IRQS_OFF + movq %rsp,%rdi /* pt_regs pointer */ + xorl %esi,%esi /* no error code */ + call \do_sym + jmp paranoid_exit /* %ebx: no swapgs flag */ CFI_ENDPROC - .endm +END(\sym) +.endm - /* error code is on the stack already */ - /* handle NMI like exceptions that can happen everywhere */ - .macro paranoidentry sym, ist=0, irqtrace=1 - SAVE_ALL - cld - movl $1,%ebx - movl $MSR_GS_BASE,%ecx - rdmsr - testl %edx,%edx - js 1f - SWAPGS - xorl %ebx,%ebx -1: - .if \ist - movq %gs:pda_data_offset, %rbp - .endif - .if \irqtrace - TRACE_IRQS_OFF - .endif - movq %rsp,%rdi - movq ORIG_RAX(%rsp),%rsi - movq $-1,ORIG_RAX(%rsp) - .if \ist - subq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp) - .endif - call \sym - .if \ist - addq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp) - .endif - DISABLE_INTERRUPTS(CLBR_NONE) - .if \irqtrace +.macro paranoidzeroentry_ist sym do_sym ist +ENTRY(\sym) + INTR_FRAME + PARAVIRT_ADJUST_EXCEPTION_FRAME + pushq $-1 /* ORIG_RAX: no syscall to restart */ + CFI_ADJUST_CFA_OFFSET 8 + subq $15*8, %rsp + call save_paranoid TRACE_IRQS_OFF - .endif - .endm + movq %rsp,%rdi /* pt_regs pointer */ + xorl %esi,%esi /* no error code */ + movq %gs:pda_data_offset, %rbp + subq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp) + call \do_sym + addq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp) + jmp paranoid_exit /* %ebx: no swapgs flag */ + CFI_ENDPROC +END(\sym) +.endm - /* - * "Paranoid" exit path from exception stack. - * Paranoid because this is used by NMIs and cannot take - * any kernel state for granted. - * We don't do kernel preemption checks here, because only - * NMI should be common and it does not enable IRQs and - * cannot get reschedule ticks. - * - * "trace" is 0 for the NMI handler only, because irq-tracing - * is fundamentally NMI-unsafe. (we cannot change the soft and - * hard flags at once, atomically) - */ - .macro paranoidexit trace=1 - /* ebx: no swapgs flag */ -paranoid_exit\trace: - testl %ebx,%ebx /* swapgs needed? */ - jnz paranoid_restore\trace - testl $3,CS(%rsp) - jnz paranoid_userspace\trace -paranoid_swapgs\trace: - .if \trace - TRACE_IRQS_IRETQ 0 - .endif - SWAPGS_UNSAFE_STACK -paranoid_restore\trace: - RESTORE_ALL 8 - jmp irq_return -paranoid_userspace\trace: - GET_THREAD_INFO(%rcx) - movl TI_flags(%rcx),%ebx - andl $_TIF_WORK_MASK,%ebx - jz paranoid_swapgs\trace - movq %rsp,%rdi /* &pt_regs */ - call sync_regs - movq %rax,%rsp /* switch stack for scheduling */ - testl $_TIF_NEED_RESCHED,%ebx - jnz paranoid_schedule\trace - movl %ebx,%edx /* arg3: thread flags */ - .if \trace - TRACE_IRQS_ON - .endif - ENABLE_INTERRUPTS(CLBR_NONE) - xorl %esi,%esi /* arg2: oldset */ - movq %rsp,%rdi /* arg1: &pt_regs */ - call do_notify_resume - DISABLE_INTERRUPTS(CLBR_NONE) - .if \trace - TRACE_IRQS_OFF - .endif - jmp paranoid_userspace\trace -paranoid_schedule\trace: - .if \trace - TRACE_IRQS_ON - .endif - ENABLE_INTERRUPTS(CLBR_ANY) - call schedule - DISABLE_INTERRUPTS(CLBR_ANY) - .if \trace - TRACE_IRQS_OFF - .endif - jmp paranoid_userspace\trace +.macro errorentry sym do_sym +ENTRY(\sym) + XCPT_FRAME + PARAVIRT_ADJUST_EXCEPTION_FRAME + subq $15*8,%rsp + CFI_ADJUST_CFA_OFFSET 15*8 + call error_entry + DEFAULT_FRAME 0 + movq %rsp,%rdi /* pt_regs pointer */ + movq ORIG_RAX(%rsp),%rsi /* get error code */ + movq $-1,ORIG_RAX(%rsp) /* no syscall to restart */ + call \do_sym + jmp error_exit /* %ebx: no swapgs flag */ CFI_ENDPROC - .endm +END(\sym) +.endm -/* - * Exception entry point. This expects an error code/orig_rax on the stack - * and the exception handler in %rax. - */ -KPROBE_ENTRY(error_entry) - _frame RDI - CFI_REL_OFFSET rax,0 - /* rdi slot contains rax, oldrax contains error code */ - cld - subq $14*8,%rsp - CFI_ADJUST_CFA_OFFSET (14*8) - movq %rsi,13*8(%rsp) - CFI_REL_OFFSET rsi,RSI - movq 14*8(%rsp),%rsi /* load rax from rdi slot */ - CFI_REGISTER rax,rsi - movq %rdx,12*8(%rsp) - CFI_REL_OFFSET rdx,RDX - movq %rcx,11*8(%rsp) - CFI_REL_OFFSET rcx,RCX - movq %rsi,10*8(%rsp) /* store rax */ - CFI_REL_OFFSET rax,RAX - movq %r8, 9*8(%rsp) - CFI_REL_OFFSET r8,R8 - movq %r9, 8*8(%rsp) - CFI_REL_OFFSET r9,R9 - movq %r10,7*8(%rsp) - CFI_REL_OFFSET r10,R10 - movq %r11,6*8(%rsp) - CFI_REL_OFFSET r11,R11 - movq %rbx,5*8(%rsp) - CFI_REL_OFFSET rbx,RBX - movq %rbp,4*8(%rsp) - CFI_REL_OFFSET rbp,RBP - movq %r12,3*8(%rsp) - CFI_REL_OFFSET r12,R12 - movq %r13,2*8(%rsp) - CFI_REL_OFFSET r13,R13 - movq %r14,1*8(%rsp) - CFI_REL_OFFSET r14,R14 - movq %r15,(%rsp) - CFI_REL_OFFSET r15,R15 - xorl %ebx,%ebx - testl $3,CS(%rsp) - je error_kernelspace -error_swapgs: - SWAPGS -error_sti: - TRACE_IRQS_OFF - movq %rdi,RDI(%rsp) - CFI_REL_OFFSET rdi,RDI - movq %rsp,%rdi - movq ORIG_RAX(%rsp),%rsi /* get error code */ - movq $-1,ORIG_RAX(%rsp) - call *%rax - /* ebx: no swapgs flag (1: don't need swapgs, 0: need it) */ -error_exit: - movl %ebx,%eax - RESTORE_REST - DISABLE_INTERRUPTS(CLBR_NONE) + /* error code is on the stack already */ +.macro paranoiderrorentry sym do_sym +ENTRY(\sym) + XCPT_FRAME + PARAVIRT_ADJUST_EXCEPTION_FRAME + subq $15*8,%rsp + CFI_ADJUST_CFA_OFFSET 15*8 + call save_paranoid + DEFAULT_FRAME 0 TRACE_IRQS_OFF - GET_THREAD_INFO(%rcx) - testl %eax,%eax - jne retint_kernel - LOCKDEP_SYS_EXIT_IRQ - movl TI_flags(%rcx),%edx - movl $_TIF_WORK_MASK,%edi - andl %edi,%edx - jnz retint_careful - jmp retint_swapgs + movq %rsp,%rdi /* pt_regs pointer */ + movq ORIG_RAX(%rsp),%rsi /* get error code */ + movq $-1,ORIG_RAX(%rsp) /* no syscall to restart */ + call \do_sym + jmp paranoid_exit /* %ebx: no swapgs flag */ CFI_ENDPROC +END(\sym) +.endm -error_kernelspace: - incl %ebx - /* There are two places in the kernel that can potentially fault with - usergs. Handle them here. The exception handlers after - iret run with kernel gs again, so don't set the user space flag. - B stepping K8s sometimes report an truncated RIP for IRET - exceptions returning to compat mode. Check for these here too. */ - leaq irq_return(%rip),%rcx - cmpq %rcx,RIP(%rsp) - je error_swapgs - movl %ecx,%ecx /* zero extend */ - cmpq %rcx,RIP(%rsp) - je error_swapgs - cmpq $gs_change,RIP(%rsp) - je error_swapgs - jmp error_sti -KPROBE_END(error_entry) - - /* Reload gs selector with exception handling */ - /* edi: new selector */ +zeroentry divide_error do_divide_error +zeroentry overflow do_overflow +zeroentry bounds do_bounds +zeroentry invalid_op do_invalid_op +zeroentry device_not_available do_device_not_available +paranoiderrorentry double_fault do_double_fault +zeroentry coprocessor_segment_overrun do_coprocessor_segment_overrun +errorentry invalid_TSS do_invalid_TSS +errorentry segment_not_present do_segment_not_present +zeroentry spurious_interrupt_bug do_spurious_interrupt_bug +zeroentry coprocessor_error do_coprocessor_error +errorentry alignment_check do_alignment_check +zeroentry simd_coprocessor_error do_simd_coprocessor_error + + /* Reload gs selector with exception handling */ + /* edi: new selector */ ENTRY(native_load_gs_index) CFI_STARTPROC pushf CFI_ADJUST_CFA_OFFSET 8 DISABLE_INTERRUPTS(CLBR_ANY | ~(CLBR_RDI)) - SWAPGS -gs_change: - movl %edi,%gs + SWAPGS +gs_change: + movl %edi,%gs 2: mfence /* workaround */ SWAPGS - popf + popf CFI_ADJUST_CFA_OFFSET -8 - ret + ret CFI_ENDPROC -ENDPROC(native_load_gs_index) - - .section __ex_table,"a" - .align 8 - .quad gs_change,bad_gs - .previous - .section .fixup,"ax" +END(native_load_gs_index) + + .section __ex_table,"a" + .align 8 + .quad gs_change,bad_gs + .previous + .section .fixup,"ax" /* running with kernelgs */ -bad_gs: +bad_gs: SWAPGS /* switch back to user gs */ xorl %eax,%eax - movl %eax,%gs - jmp 2b - .previous - + movl %eax,%gs + jmp 2b + .previous + /* * Create a kernel thread. * @@ -1164,7 +1184,7 @@ ENTRY(kernel_thread) xorl %r8d,%r8d xorl %r9d,%r9d - + # clone now call do_fork movq %rax,RAX(%rsp) @@ -1175,15 +1195,15 @@ ENTRY(kernel_thread) * so internally to the x86_64 port you can rely on kernel_thread() * not to reschedule the child before returning, this avoids the need * of hacks for example to fork off the per-CPU idle tasks. - * [Hopefully no generic code relies on the reschedule -AK] + * [Hopefully no generic code relies on the reschedule -AK] */ RESTORE_ALL UNFAKE_STACK_FRAME ret CFI_ENDPROC -ENDPROC(kernel_thread) - -child_rip: +END(kernel_thread) + +ENTRY(child_rip) pushq $0 # fake return address CFI_STARTPROC /* @@ -1196,8 +1216,9 @@ child_rip: # exit mov %eax, %edi call do_exit + ud2 # padding for call trace CFI_ENDPROC -ENDPROC(child_rip) +END(child_rip) /* * execve(). This function needs to use IRET, not SYSRET, to set up all state properly. @@ -1217,10 +1238,10 @@ ENDPROC(child_rip) ENTRY(kernel_execve) CFI_STARTPROC FAKE_STACK_FRAME $0 - SAVE_ALL + SAVE_ALL movq %rsp,%rcx call sys_execve - movq %rax, RAX(%rsp) + movq %rax, RAX(%rsp) RESTORE_REST testq %rax,%rax je int_ret_from_sys_call @@ -1228,129 +1249,7 @@ ENTRY(kernel_execve) UNFAKE_STACK_FRAME ret CFI_ENDPROC -ENDPROC(kernel_execve) - -KPROBE_ENTRY(page_fault) - errorentry do_page_fault -KPROBE_END(page_fault) - -ENTRY(coprocessor_error) - zeroentry do_coprocessor_error -END(coprocessor_error) - -ENTRY(simd_coprocessor_error) - zeroentry do_simd_coprocessor_error -END(simd_coprocessor_error) - -ENTRY(device_not_available) - zeroentry do_device_not_available -END(device_not_available) - - /* runs on exception stack */ -KPROBE_ENTRY(debug) - INTR_FRAME - PARAVIRT_ADJUST_EXCEPTION_FRAME - pushq $0 - CFI_ADJUST_CFA_OFFSET 8 - paranoidentry do_debug, DEBUG_STACK - paranoidexit -KPROBE_END(debug) - - /* runs on exception stack */ -KPROBE_ENTRY(nmi) - INTR_FRAME - PARAVIRT_ADJUST_EXCEPTION_FRAME - pushq $-1 - CFI_ADJUST_CFA_OFFSET 8 - paranoidentry do_nmi, 0, 0 -#ifdef CONFIG_TRACE_IRQFLAGS - paranoidexit 0 -#else - jmp paranoid_exit1 - CFI_ENDPROC -#endif -KPROBE_END(nmi) - -KPROBE_ENTRY(int3) - INTR_FRAME - PARAVIRT_ADJUST_EXCEPTION_FRAME - pushq $0 - CFI_ADJUST_CFA_OFFSET 8 - paranoidentry do_int3, DEBUG_STACK - jmp paranoid_exit1 - CFI_ENDPROC -KPROBE_END(int3) - -ENTRY(overflow) - zeroentry do_overflow -END(overflow) - -ENTRY(bounds) - zeroentry do_bounds -END(bounds) - -ENTRY(invalid_op) - zeroentry do_invalid_op -END(invalid_op) - -ENTRY(coprocessor_segment_overrun) - zeroentry do_coprocessor_segment_overrun -END(coprocessor_segment_overrun) - - /* runs on exception stack */ -ENTRY(double_fault) - XCPT_FRAME - PARAVIRT_ADJUST_EXCEPTION_FRAME - paranoidentry do_double_fault - jmp paranoid_exit1 - CFI_ENDPROC -END(double_fault) - -ENTRY(invalid_TSS) - errorentry do_invalid_TSS -END(invalid_TSS) - -ENTRY(segment_not_present) - errorentry do_segment_not_present -END(segment_not_present) - - /* runs on exception stack */ -ENTRY(stack_segment) - XCPT_FRAME - PARAVIRT_ADJUST_EXCEPTION_FRAME - paranoidentry do_stack_segment - jmp paranoid_exit1 - CFI_ENDPROC -END(stack_segment) - -KPROBE_ENTRY(general_protection) - errorentry do_general_protection -KPROBE_END(general_protection) - -ENTRY(alignment_check) - errorentry do_alignment_check -END(alignment_check) - -ENTRY(divide_error) - zeroentry do_divide_error -END(divide_error) - -ENTRY(spurious_interrupt_bug) - zeroentry do_spurious_interrupt_bug -END(spurious_interrupt_bug) - -#ifdef CONFIG_X86_MCE - /* runs on exception stack */ -ENTRY(machine_check) - INTR_FRAME - PARAVIRT_ADJUST_EXCEPTION_FRAME - pushq $0 - CFI_ADJUST_CFA_OFFSET 8 - paranoidentry do_machine_check - jmp paranoid_exit1 - CFI_ENDPROC -END(machine_check) -#endif +END(kernel_execve) /* Call softirq on interrupt stack. Interrupts are off. */ ENTRY(call_softirq) @@ -1370,40 +1269,33 @@ ENTRY(call_softirq) decl %gs:pda_irqcount ret CFI_ENDPROC -ENDPROC(call_softirq) - -KPROBE_ENTRY(ignore_sysret) - CFI_STARTPROC - mov $-ENOSYS,%eax - sysret - CFI_ENDPROC -ENDPROC(ignore_sysret) +END(call_softirq) #ifdef CONFIG_XEN -ENTRY(xen_hypervisor_callback) - zeroentry xen_do_hypervisor_callback -END(xen_hypervisor_callback) +zeroentry xen_hypervisor_callback xen_do_hypervisor_callback /* -# A note on the "critical region" in our callback handler. -# We want to avoid stacking callback handlers due to events occurring -# during handling of the last event. To do this, we keep events disabled -# until we've done all processing. HOWEVER, we must enable events before -# popping the stack frame (can't be done atomically) and so it would still -# be possible to get enough handler activations to overflow the stack. -# Although unlikely, bugs of that kind are hard to track down, so we'd -# like to avoid the possibility. -# So, on entry to the handler we detect whether we interrupted an -# existing activation in its critical region -- if so, we pop the current -# activation and restart the handler using the previous one. -*/ + * A note on the "critical region" in our callback handler. + * We want to avoid stacking callback handlers due to events occurring + * during handling of the last event. To do this, we keep events disabled + * until we've done all processing. HOWEVER, we must enable events before + * popping the stack frame (can't be done atomically) and so it would still + * be possible to get enough handler activations to overflow the stack. + * Although unlikely, bugs of that kind are hard to track down, so we'd + * like to avoid the possibility. + * So, on entry to the handler we detect whether we interrupted an + * existing activation in its critical region -- if so, we pop the current + * activation and restart the handler using the previous one. + */ ENTRY(xen_do_hypervisor_callback) # do_hypervisor_callback(struct *pt_regs) CFI_STARTPROC -/* Since we don't modify %rdi, evtchn_do_upall(struct *pt_regs) will - see the correct pointer to the pt_regs */ +/* + * Since we don't modify %rdi, evtchn_do_upall(struct *pt_regs) will + * see the correct pointer to the pt_regs + */ movq %rdi, %rsp # we don't return, adjust the stack frame CFI_ENDPROC - CFI_DEFAULT_STACK + DEFAULT_FRAME 11: incl %gs:pda_irqcount movq %rsp,%rbp CFI_DEF_CFA_REGISTER rbp @@ -1418,23 +1310,26 @@ ENTRY(xen_do_hypervisor_callback) # do_hypervisor_callback(struct *pt_regs) END(do_hypervisor_callback) /* -# Hypervisor uses this for application faults while it executes. -# We get here for two reasons: -# 1. Fault while reloading DS, ES, FS or GS -# 2. Fault while executing IRET -# Category 1 we do not need to fix up as Xen has already reloaded all segment -# registers that could be reloaded and zeroed the others. -# Category 2 we fix up by killing the current process. We cannot use the -# normal Linux return path in this case because if we use the IRET hypercall -# to pop the stack frame we end up in an infinite loop of failsafe callbacks. -# We distinguish between categories by comparing each saved segment register -# with its current contents: any discrepancy means we in category 1. -*/ + * Hypervisor uses this for application faults while it executes. + * We get here for two reasons: + * 1. Fault while reloading DS, ES, FS or GS + * 2. Fault while executing IRET + * Category 1 we do not need to fix up as Xen has already reloaded all segment + * registers that could be reloaded and zeroed the others. + * Category 2 we fix up by killing the current process. We cannot use the + * normal Linux return path in this case because if we use the IRET hypercall + * to pop the stack frame we end up in an infinite loop of failsafe callbacks. + * We distinguish between categories by comparing each saved segment register + * with its current contents: any discrepancy means we in category 1. + */ ENTRY(xen_failsafe_callback) - framesz = (RIP-0x30) /* workaround buggy gas */ - _frame framesz - CFI_REL_OFFSET rcx, 0 - CFI_REL_OFFSET r11, 8 + INTR_FRAME 1 (6*8) + /*CFI_REL_OFFSET gs,GS*/ + /*CFI_REL_OFFSET fs,FS*/ + /*CFI_REL_OFFSET es,ES*/ + /*CFI_REL_OFFSET ds,DS*/ + CFI_REL_OFFSET r11,8 + CFI_REL_OFFSET rcx,0 movw %ds,%cx cmpw %cx,0x10(%rsp) CFI_REMEMBER_STATE @@ -1455,12 +1350,9 @@ ENTRY(xen_failsafe_callback) CFI_RESTORE r11 addq $0x30,%rsp CFI_ADJUST_CFA_OFFSET -0x30 - pushq $0 - CFI_ADJUST_CFA_OFFSET 8 - pushq %r11 - CFI_ADJUST_CFA_OFFSET 8 - pushq %rcx - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $0 /* RIP */ + pushq_cfi %r11 + pushq_cfi %rcx jmp general_protection CFI_RESTORE_STATE 1: /* Segment mismatch => Category 1 (Bad segment). Retry the IRET. */ @@ -1470,11 +1362,223 @@ ENTRY(xen_failsafe_callback) CFI_RESTORE r11 addq $0x30,%rsp CFI_ADJUST_CFA_OFFSET -0x30 - pushq $0 - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $0 SAVE_ALL jmp error_exit CFI_ENDPROC END(xen_failsafe_callback) #endif /* CONFIG_XEN */ + +/* + * Some functions should be protected against kprobes + */ + .pushsection .kprobes.text, "ax" + +paranoidzeroentry_ist debug do_debug DEBUG_STACK +paranoidzeroentry_ist int3 do_int3 DEBUG_STACK +paranoiderrorentry stack_segment do_stack_segment +errorentry general_protection do_general_protection +errorentry page_fault do_page_fault +#ifdef CONFIG_X86_MCE +paranoidzeroentry machine_check do_machine_check +#endif + + /* + * "Paranoid" exit path from exception stack. + * Paranoid because this is used by NMIs and cannot take + * any kernel state for granted. + * We don't do kernel preemption checks here, because only + * NMI should be common and it does not enable IRQs and + * cannot get reschedule ticks. + * + * "trace" is 0 for the NMI handler only, because irq-tracing + * is fundamentally NMI-unsafe. (we cannot change the soft and + * hard flags at once, atomically) + */ + + /* ebx: no swapgs flag */ +ENTRY(paranoid_exit) + INTR_FRAME + DISABLE_INTERRUPTS(CLBR_NONE) + TRACE_IRQS_OFF + testl %ebx,%ebx /* swapgs needed? */ + jnz paranoid_restore + testl $3,CS(%rsp) + jnz paranoid_userspace +paranoid_swapgs: + TRACE_IRQS_IRETQ 0 + SWAPGS_UNSAFE_STACK +paranoid_restore: + RESTORE_ALL 8 + jmp irq_return +paranoid_userspace: + GET_THREAD_INFO(%rcx) + movl TI_flags(%rcx),%ebx + andl $_TIF_WORK_MASK,%ebx + jz paranoid_swapgs + movq %rsp,%rdi /* &pt_regs */ + call sync_regs + movq %rax,%rsp /* switch stack for scheduling */ + testl $_TIF_NEED_RESCHED,%ebx + jnz paranoid_schedule + movl %ebx,%edx /* arg3: thread flags */ + TRACE_IRQS_ON + ENABLE_INTERRUPTS(CLBR_NONE) + xorl %esi,%esi /* arg2: oldset */ + movq %rsp,%rdi /* arg1: &pt_regs */ + call do_notify_resume + DISABLE_INTERRUPTS(CLBR_NONE) + TRACE_IRQS_OFF + jmp paranoid_userspace +paranoid_schedule: + TRACE_IRQS_ON + ENABLE_INTERRUPTS(CLBR_ANY) + call schedule + DISABLE_INTERRUPTS(CLBR_ANY) + TRACE_IRQS_OFF + jmp paranoid_userspace + CFI_ENDPROC +END(paranoid_exit) + +/* + * Exception entry point. This expects an error code/orig_rax on the stack. + * returns in "no swapgs flag" in %ebx. + */ +ENTRY(error_entry) + XCPT_FRAME + CFI_ADJUST_CFA_OFFSET 15*8 + /* oldrax contains error code */ + cld + movq_cfi rdi, RDI+8 + movq_cfi rsi, RSI+8 + movq_cfi rdx, RDX+8 + movq_cfi rcx, RCX+8 + movq_cfi rax, RAX+8 + movq_cfi r8, R8+8 + movq_cfi r9, R9+8 + movq_cfi r10, R10+8 + movq_cfi r11, R11+8 + movq_cfi rbx, RBX+8 + movq_cfi rbp, RBP+8 + movq_cfi r12, R12+8 + movq_cfi r13, R13+8 + movq_cfi r14, R14+8 + movq_cfi r15, R15+8 + xorl %ebx,%ebx + testl $3,CS+8(%rsp) + je error_kernelspace +error_swapgs: + SWAPGS +error_sti: + TRACE_IRQS_OFF + ret + CFI_ENDPROC + +/* + * There are two places in the kernel that can potentially fault with + * usergs. Handle them here. The exception handlers after iret run with + * kernel gs again, so don't set the user space flag. B stepping K8s + * sometimes report an truncated RIP for IRET exceptions returning to + * compat mode. Check for these here too. + */ +error_kernelspace: + incl %ebx + leaq irq_return(%rip),%rcx + cmpq %rcx,RIP+8(%rsp) + je error_swapgs + movl %ecx,%ecx /* zero extend */ + cmpq %rcx,RIP+8(%rsp) + je error_swapgs + cmpq $gs_change,RIP+8(%rsp) + je error_swapgs + jmp error_sti +END(error_entry) + + +/* ebx: no swapgs flag (1: don't need swapgs, 0: need it) */ +ENTRY(error_exit) + DEFAULT_FRAME + movl %ebx,%eax + RESTORE_REST + DISABLE_INTERRUPTS(CLBR_NONE) + TRACE_IRQS_OFF + GET_THREAD_INFO(%rcx) + testl %eax,%eax + jne retint_kernel + LOCKDEP_SYS_EXIT_IRQ + movl TI_flags(%rcx),%edx + movl $_TIF_WORK_MASK,%edi + andl %edi,%edx + jnz retint_careful + jmp retint_swapgs + CFI_ENDPROC +END(error_exit) + + + /* runs on exception stack */ +ENTRY(nmi) + INTR_FRAME + PARAVIRT_ADJUST_EXCEPTION_FRAME + pushq_cfi $-1 + subq $15*8, %rsp + CFI_ADJUST_CFA_OFFSET 15*8 + call save_paranoid + DEFAULT_FRAME 0 + /* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */ + movq %rsp,%rdi + movq $-1,%rsi + call do_nmi +#ifdef CONFIG_TRACE_IRQFLAGS + /* paranoidexit; without TRACE_IRQS_OFF */ + /* ebx: no swapgs flag */ + DISABLE_INTERRUPTS(CLBR_NONE) + testl %ebx,%ebx /* swapgs needed? */ + jnz nmi_restore + testl $3,CS(%rsp) + jnz nmi_userspace +nmi_swapgs: + SWAPGS_UNSAFE_STACK +nmi_restore: + RESTORE_ALL 8 + jmp irq_return +nmi_userspace: + GET_THREAD_INFO(%rcx) + movl TI_flags(%rcx),%ebx + andl $_TIF_WORK_MASK,%ebx + jz nmi_swapgs + movq %rsp,%rdi /* &pt_regs */ + call sync_regs + movq %rax,%rsp /* switch stack for scheduling */ + testl $_TIF_NEED_RESCHED,%ebx + jnz nmi_schedule + movl %ebx,%edx /* arg3: thread flags */ + ENABLE_INTERRUPTS(CLBR_NONE) + xorl %esi,%esi /* arg2: oldset */ + movq %rsp,%rdi /* arg1: &pt_regs */ + call do_notify_resume + DISABLE_INTERRUPTS(CLBR_NONE) + jmp nmi_userspace +nmi_schedule: + ENABLE_INTERRUPTS(CLBR_ANY) + call schedule + DISABLE_INTERRUPTS(CLBR_ANY) + jmp nmi_userspace + CFI_ENDPROC +#else + jmp paranoid_exit + CFI_ENDPROC +#endif +END(nmi) + +ENTRY(ignore_sysret) + CFI_STARTPROC + mov $-ENOSYS,%eax + sysret + CFI_ENDPROC +END(ignore_sysret) + +/* + * End of kprobes section + */ + .popsection diff --git a/arch/x86/kernel/es7000_32.c b/arch/x86/kernel/es7000_32.c index f454c78fcef6..53699c931ad4 100644 --- a/arch/x86/kernel/es7000_32.c +++ b/arch/x86/kernel/es7000_32.c @@ -38,8 +38,11 @@ #include <asm/io.h> #include <asm/nmi.h> #include <asm/smp.h> +#include <asm/atomic.h> #include <asm/apicdef.h> #include <mach_mpparse.h> +#include <asm/genapic.h> +#include <asm/setup.h> /* * ES7000 chipsets @@ -161,6 +164,43 @@ es7000_rename_gsi(int ioapic, int gsi) return gsi; } +static int wakeup_secondary_cpu_via_mip(int cpu, unsigned long eip) +{ + unsigned long vect = 0, psaival = 0; + + if (psai == NULL) + return -1; + + vect = ((unsigned long)__pa(eip)/0x1000) << 16; + psaival = (0x1000000 | vect | cpu); + + while (*psai & 0x1000000) + ; + + *psai = psaival; + + return 0; +} + +static void noop_wait_for_deassert(atomic_t *deassert_not_used) +{ +} + +static int __init es7000_update_genapic(void) +{ + genapic->wakeup_cpu = wakeup_secondary_cpu_via_mip; + + /* MPENTIUMIII */ + if (boot_cpu_data.x86 == 6 && + (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11)) { + es7000_update_genapic_to_cluster(); + genapic->wait_for_init_deassert = noop_wait_for_deassert; + genapic->wakeup_cpu = wakeup_secondary_cpu_via_mip; + } + + return 0; +} + void __init setup_unisys(void) { @@ -176,6 +216,8 @@ setup_unisys(void) else es7000_plat = ES7000_CLASSIC; ioapic_renumber_irq = es7000_rename_gsi; + + x86_quirks->update_genapic = es7000_update_genapic; } /* @@ -250,31 +292,24 @@ int __init find_unisys_acpi_oem_table(unsigned long *oem_addr) { struct acpi_table_header *header = NULL; int i = 0; - acpi_size tbl_size; - while (ACPI_SUCCESS(acpi_get_table_with_size("OEM1", i++, &header, &tbl_size))) { + while (ACPI_SUCCESS(acpi_get_table("OEM1", i++, &header))) { if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) { struct oem_table *t = (struct oem_table *)header; oem_addrX = t->OEMTableAddr; oem_size = t->OEMTableSize; - early_acpi_os_unmap_memory(header, tbl_size); *oem_addr = (unsigned long)__acpi_map_table(oem_addrX, oem_size); return 0; } - early_acpi_os_unmap_memory(header, tbl_size); } return -1; } void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr) { - if (!oem_addr) - return; - - __acpi_unmap_table((char *)oem_addr, oem_size); } #endif @@ -324,26 +359,6 @@ es7000_mip_write(struct mip_reg *mip_reg) return status; } -int -es7000_start_cpu(int cpu, unsigned long eip) -{ - unsigned long vect = 0, psaival = 0; - - if (psai == NULL) - return -1; - - vect = ((unsigned long)__pa(eip)/0x1000) << 16; - psaival = (0x1000000 | vect | cpu); - - while (*psai & 0x1000000) - ; - - *psai = psaival; - - return 0; - -} - void __init es7000_sw_apic(void) { diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index ab115cd15fdf..1b43086b097a 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -11,17 +11,20 @@ #include <linux/spinlock.h> #include <linux/hardirq.h> +#include <linux/uaccess.h> #include <linux/ftrace.h> #include <linux/percpu.h> +#include <linux/sched.h> #include <linux/init.h> #include <linux/list.h> -#include <asm/alternative.h> #include <asm/ftrace.h> +#include <linux/ftrace.h> +#include <asm/nops.h> +#include <asm/nmi.h> -/* Long is fine, even if it is only 4 bytes ;-) */ -static long *ftrace_nop; +#ifdef CONFIG_DYNAMIC_FTRACE union ftrace_code_union { char code[MCOUNT_INSN_SIZE]; @@ -31,18 +34,12 @@ union ftrace_code_union { } __attribute__((packed)); }; - -static int notrace ftrace_calc_offset(long ip, long addr) +static int ftrace_calc_offset(long ip, long addr) { return (int)(addr - ip); } -notrace unsigned char *ftrace_nop_replace(void) -{ - return (char *)ftrace_nop; -} - -notrace unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) +static unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) { static union ftrace_code_union calc; @@ -56,48 +53,198 @@ notrace unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) return calc.code; } -notrace int +/* + * Modifying code must take extra care. On an SMP machine, if + * the code being modified is also being executed on another CPU + * that CPU will have undefined results and possibly take a GPF. + * We use kstop_machine to stop other CPUS from exectuing code. + * But this does not stop NMIs from happening. We still need + * to protect against that. We separate out the modification of + * the code to take care of this. + * + * Two buffers are added: An IP buffer and a "code" buffer. + * + * 1) Put the instruction pointer into the IP buffer + * and the new code into the "code" buffer. + * 2) Set a flag that says we are modifying code + * 3) Wait for any running NMIs to finish. + * 4) Write the code + * 5) clear the flag. + * 6) Wait for any running NMIs to finish. + * + * If an NMI is executed, the first thing it does is to call + * "ftrace_nmi_enter". This will check if the flag is set to write + * and if it is, it will write what is in the IP and "code" buffers. + * + * The trick is, it does not matter if everyone is writing the same + * content to the code location. Also, if a CPU is executing code + * it is OK to write to that code location if the contents being written + * are the same as what exists. + */ + +static atomic_t in_nmi = ATOMIC_INIT(0); +static int mod_code_status; /* holds return value of text write */ +static int mod_code_write; /* set when NMI should do the write */ +static void *mod_code_ip; /* holds the IP to write to */ +static void *mod_code_newcode; /* holds the text to write to the IP */ + +static unsigned nmi_wait_count; +static atomic_t nmi_update_count = ATOMIC_INIT(0); + +int ftrace_arch_read_dyn_info(char *buf, int size) +{ + int r; + + r = snprintf(buf, size, "%u %u", + nmi_wait_count, + atomic_read(&nmi_update_count)); + return r; +} + +static void ftrace_mod_code(void) +{ + /* + * Yes, more than one CPU process can be writing to mod_code_status. + * (and the code itself) + * But if one were to fail, then they all should, and if one were + * to succeed, then they all should. + */ + mod_code_status = probe_kernel_write(mod_code_ip, mod_code_newcode, + MCOUNT_INSN_SIZE); +} + +void ftrace_nmi_enter(void) +{ + atomic_inc(&in_nmi); + /* Must have in_nmi seen before reading write flag */ + smp_mb(); + if (mod_code_write) { + ftrace_mod_code(); + atomic_inc(&nmi_update_count); + } +} + +void ftrace_nmi_exit(void) +{ + /* Finish all executions before clearing in_nmi */ + smp_wmb(); + atomic_dec(&in_nmi); +} + +static void wait_for_nmi(void) +{ + int waited = 0; + + while (atomic_read(&in_nmi)) { + waited = 1; + cpu_relax(); + } + + if (waited) + nmi_wait_count++; +} + +static int +do_ftrace_mod_code(unsigned long ip, void *new_code) +{ + mod_code_ip = (void *)ip; + mod_code_newcode = new_code; + + /* The buffers need to be visible before we let NMIs write them */ + smp_wmb(); + + mod_code_write = 1; + + /* Make sure write bit is visible before we wait on NMIs */ + smp_mb(); + + wait_for_nmi(); + + /* Make sure all running NMIs have finished before we write the code */ + smp_mb(); + + ftrace_mod_code(); + + /* Make sure the write happens before clearing the bit */ + smp_wmb(); + + mod_code_write = 0; + + /* make sure NMIs see the cleared bit */ + smp_mb(); + + wait_for_nmi(); + + return mod_code_status; +} + + + + +static unsigned char ftrace_nop[MCOUNT_INSN_SIZE]; + +static unsigned char *ftrace_nop_replace(void) +{ + return ftrace_nop; +} + +static int ftrace_modify_code(unsigned long ip, unsigned char *old_code, unsigned char *new_code) { - unsigned replaced; - unsigned old = *(unsigned *)old_code; /* 4 bytes */ - unsigned new = *(unsigned *)new_code; /* 4 bytes */ - unsigned char newch = new_code[4]; - int faulted = 0; + unsigned char replaced[MCOUNT_INSN_SIZE]; /* * Note: Due to modules and __init, code can * disappear and change, we need to protect against faulting - * as well as code changing. + * as well as code changing. We do this by using the + * probe_kernel_* functions. * * No real locking needed, this code is run through - * kstop_machine. + * kstop_machine, or before SMP starts. */ - asm volatile ( - "1: lock\n" - " cmpxchg %3, (%2)\n" - " jnz 2f\n" - " movb %b4, 4(%2)\n" - "2:\n" - ".section .fixup, \"ax\"\n" - "3: movl $1, %0\n" - " jmp 2b\n" - ".previous\n" - _ASM_EXTABLE(1b, 3b) - : "=r"(faulted), "=a"(replaced) - : "r"(ip), "r"(new), "c"(newch), - "0"(faulted), "a"(old) - : "memory"); + + /* read the text we want to modify */ + if (probe_kernel_read(replaced, (void *)ip, MCOUNT_INSN_SIZE)) + return -EFAULT; + + /* Make sure it is what we expect it to be */ + if (memcmp(replaced, old_code, MCOUNT_INSN_SIZE) != 0) + return -EINVAL; + + /* replace the text with the new text */ + if (do_ftrace_mod_code(ip, new_code)) + return -EPERM; + sync_core(); - if (replaced != old && replaced != new) - faulted = 2; + return 0; +} + +int ftrace_make_nop(struct module *mod, + struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char *new, *old; + unsigned long ip = rec->ip; - return faulted; + old = ftrace_call_replace(ip, addr); + new = ftrace_nop_replace(); + + return ftrace_modify_code(rec->ip, old, new); } -notrace int ftrace_update_ftrace_func(ftrace_func_t func) +int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char *new, *old; + unsigned long ip = rec->ip; + + old = ftrace_nop_replace(); + new = ftrace_call_replace(ip, addr); + + return ftrace_modify_code(rec->ip, old, new); +} + +int ftrace_update_ftrace_func(ftrace_func_t func) { unsigned long ip = (unsigned long)(&ftrace_call); unsigned char old[MCOUNT_INSN_SIZE], *new; @@ -110,32 +257,281 @@ notrace int ftrace_update_ftrace_func(ftrace_func_t func) return ret; } -notrace int ftrace_mcount_set(unsigned long *data) +int __init ftrace_dyn_arch_init(void *data) { - unsigned long ip = (long)(&mcount_call); - unsigned long *addr = data; - unsigned char old[MCOUNT_INSN_SIZE], *new; + extern const unsigned char ftrace_test_p6nop[]; + extern const unsigned char ftrace_test_nop5[]; + extern const unsigned char ftrace_test_jmp[]; + int faulted = 0; /* - * Replace the mcount stub with a pointer to the - * ip recorder function. + * There is no good nop for all x86 archs. + * We will default to using the P6_NOP5, but first we + * will test to make sure that the nop will actually + * work on this CPU. If it faults, we will then + * go to a lesser efficient 5 byte nop. If that fails + * we then just use a jmp as our nop. This isn't the most + * efficient nop, but we can not use a multi part nop + * since we would then risk being preempted in the middle + * of that nop, and if we enabled tracing then, it might + * cause a system crash. + * + * TODO: check the cpuid to determine the best nop. */ - memcpy(old, &mcount_call, MCOUNT_INSN_SIZE); - new = ftrace_call_replace(ip, *addr); - *addr = ftrace_modify_code(ip, old, new); + asm volatile ( + "ftrace_test_jmp:" + "jmp ftrace_test_p6nop\n" + "nop\n" + "nop\n" + "nop\n" /* 2 byte jmp + 3 bytes */ + "ftrace_test_p6nop:" + P6_NOP5 + "jmp 1f\n" + "ftrace_test_nop5:" + ".byte 0x66,0x66,0x66,0x66,0x90\n" + "1:" + ".section .fixup, \"ax\"\n" + "2: movl $1, %0\n" + " jmp ftrace_test_nop5\n" + "3: movl $2, %0\n" + " jmp 1b\n" + ".previous\n" + _ASM_EXTABLE(ftrace_test_p6nop, 2b) + _ASM_EXTABLE(ftrace_test_nop5, 3b) + : "=r"(faulted) : "0" (faulted)); + + switch (faulted) { + case 0: + pr_info("ftrace: converting mcount calls to 0f 1f 44 00 00\n"); + memcpy(ftrace_nop, ftrace_test_p6nop, MCOUNT_INSN_SIZE); + break; + case 1: + pr_info("ftrace: converting mcount calls to 66 66 66 66 90\n"); + memcpy(ftrace_nop, ftrace_test_nop5, MCOUNT_INSN_SIZE); + break; + case 2: + pr_info("ftrace: converting mcount calls to jmp . + 5\n"); + memcpy(ftrace_nop, ftrace_test_jmp, MCOUNT_INSN_SIZE); + break; + } + + /* The return code is retured via data */ + *(unsigned long *)data = 0; return 0; } +#endif -int __init ftrace_dyn_arch_init(void *data) +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + +#ifdef CONFIG_DYNAMIC_FTRACE +extern void ftrace_graph_call(void); + +static int ftrace_mod_jmp(unsigned long ip, + int old_offset, int new_offset) +{ + unsigned char code[MCOUNT_INSN_SIZE]; + + if (probe_kernel_read(code, (void *)ip, MCOUNT_INSN_SIZE)) + return -EFAULT; + + if (code[0] != 0xe9 || old_offset != *(int *)(&code[1])) + return -EINVAL; + + *(int *)(&code[1]) = new_offset; + + if (do_ftrace_mod_code(ip, &code)) + return -EPERM; + + return 0; +} + +int ftrace_enable_ftrace_graph_caller(void) +{ + unsigned long ip = (unsigned long)(&ftrace_graph_call); + int old_offset, new_offset; + + old_offset = (unsigned long)(&ftrace_stub) - (ip + MCOUNT_INSN_SIZE); + new_offset = (unsigned long)(&ftrace_graph_caller) - (ip + MCOUNT_INSN_SIZE); + + return ftrace_mod_jmp(ip, old_offset, new_offset); +} + +int ftrace_disable_ftrace_graph_caller(void) +{ + unsigned long ip = (unsigned long)(&ftrace_graph_call); + int old_offset, new_offset; + + old_offset = (unsigned long)(&ftrace_graph_caller) - (ip + MCOUNT_INSN_SIZE); + new_offset = (unsigned long)(&ftrace_stub) - (ip + MCOUNT_INSN_SIZE); + + return ftrace_mod_jmp(ip, old_offset, new_offset); +} + +#else /* CONFIG_DYNAMIC_FTRACE */ + +/* + * These functions are picked from those used on + * this page for dynamic ftrace. They have been + * simplified to ignore all traces in NMI context. + */ +static atomic_t in_nmi; + +void ftrace_nmi_enter(void) +{ + atomic_inc(&in_nmi); +} + +void ftrace_nmi_exit(void) +{ + atomic_dec(&in_nmi); +} + +#endif /* !CONFIG_DYNAMIC_FTRACE */ + +/* Add a function return address to the trace stack on thread info.*/ +static int push_return_trace(unsigned long ret, unsigned long long time, + unsigned long func, int *depth) { - const unsigned char *const *noptable = find_nop_table(); + int index; - /* This is running in kstop_machine */ + if (!current->ret_stack) + return -EBUSY; - ftrace_mcount_set(data); + /* The return trace stack is full */ + if (current->curr_ret_stack == FTRACE_RETFUNC_DEPTH - 1) { + atomic_inc(¤t->trace_overrun); + return -EBUSY; + } - ftrace_nop = (unsigned long *)noptable[MCOUNT_INSN_SIZE]; + index = ++current->curr_ret_stack; + barrier(); + current->ret_stack[index].ret = ret; + current->ret_stack[index].func = func; + current->ret_stack[index].calltime = time; + *depth = index; return 0; } + +/* Retrieve a function return address to the trace stack on thread info.*/ +static void pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret) +{ + int index; + + index = current->curr_ret_stack; + + if (unlikely(index < 0)) { + ftrace_graph_stop(); + WARN_ON(1); + /* Might as well panic, otherwise we have no where to go */ + *ret = (unsigned long)panic; + return; + } + + *ret = current->ret_stack[index].ret; + trace->func = current->ret_stack[index].func; + trace->calltime = current->ret_stack[index].calltime; + trace->overrun = atomic_read(¤t->trace_overrun); + trace->depth = index; + barrier(); + current->curr_ret_stack--; + +} + +/* + * Send the trace to the ring-buffer. + * @return the original return address. + */ +unsigned long ftrace_return_to_handler(void) +{ + struct ftrace_graph_ret trace; + unsigned long ret; + + pop_return_trace(&trace, &ret); + trace.rettime = cpu_clock(raw_smp_processor_id()); + ftrace_graph_return(&trace); + + if (unlikely(!ret)) { + ftrace_graph_stop(); + WARN_ON(1); + /* Might as well panic. What else to do? */ + ret = (unsigned long)panic; + } + + return ret; +} + +/* + * 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) +{ + unsigned long old; + unsigned long long calltime; + int faulted; + struct ftrace_graph_ent trace; + unsigned long return_hooker = (unsigned long) + &return_to_handler; + + /* Nmi's are currently unsupported */ + if (unlikely(atomic_read(&in_nmi))) + return; + + if (unlikely(atomic_read(¤t->tracing_graph_pause))) + return; + + /* + * Protect against fault, even if it shouldn't + * happen. This tool is too much intrusive to + * ignore such a protection. + */ + asm volatile( + "1: " _ASM_MOV " (%[parent_old]), %[old]\n" + "2: " _ASM_MOV " %[return_hooker], (%[parent_replaced])\n" + " movl $0, %[faulted]\n" + + ".section .fixup, \"ax\"\n" + "3: movl $1, %[faulted]\n" + ".previous\n" + + _ASM_EXTABLE(1b, 3b) + _ASM_EXTABLE(2b, 3b) + + : [parent_replaced] "=r" (parent), [old] "=r" (old), + [faulted] "=r" (faulted) + : [parent_old] "0" (parent), [return_hooker] "r" (return_hooker) + : "memory" + ); + + if (unlikely(faulted)) { + ftrace_graph_stop(); + WARN_ON(1); + return; + } + + if (unlikely(!__kernel_text_address(old))) { + ftrace_graph_stop(); + *parent = old; + WARN_ON(1); + return; + } + + calltime = cpu_clock(raw_smp_processor_id()); + + if (push_return_trace(old, calltime, + self_addr, &trace.depth) == -EBUSY) { + *parent = old; + return; + } + + trace.func = self_addr; + + /* Only trace if the calling function expects to */ + if (!ftrace_graph_entry(&trace)) { + current->curr_ret_stack--; + *parent = old; + } +} +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c index 6c9bfc9e1e95..2bced78b0b8e 100644 --- a/arch/x86/kernel/genapic_64.c +++ b/arch/x86/kernel/genapic_64.c @@ -21,6 +21,7 @@ #include <asm/smp.h> #include <asm/ipi.h> #include <asm/genapic.h> +#include <asm/setup.h> extern struct genapic apic_flat; extern struct genapic apic_physflat; @@ -53,6 +54,9 @@ void __init setup_apic_routing(void) genapic = &apic_physflat; printk(KERN_INFO "Setting APIC routing to %s\n", genapic->name); } + + if (x86_quirks->update_genapic) + x86_quirks->update_genapic(); } /* Same for both flat and physical. */ diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c index 9eca5ba7a6b1..c0262791bda4 100644 --- a/arch/x86/kernel/genapic_flat_64.c +++ b/arch/x86/kernel/genapic_flat_64.c @@ -25,7 +25,7 @@ #include <acpi/acpi_bus.h> #endif -static int __init flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +static int flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 1; } @@ -170,7 +170,7 @@ struct genapic apic_flat = { * We cannot use logical delivery in this case because the mask * overflows, so use physical mode. */ -static int __init physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +static int physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { #ifdef CONFIG_ACPI /* @@ -179,8 +179,10 @@ static int __init physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) * is an example). */ if (acpi_gbl_FADT.header.revision > FADT2_REVISION_ID && - (acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL)) + (acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL)) { + printk(KERN_DEBUG "system APIC only can use physical flat"); return 1; + } #endif return 0; diff --git a/arch/x86/kernel/genx2apic_cluster.c b/arch/x86/kernel/genx2apic_cluster.c index e4bf2cc0d743..f6a2c8eb48a6 100644 --- a/arch/x86/kernel/genx2apic_cluster.c +++ b/arch/x86/kernel/genx2apic_cluster.c @@ -12,7 +12,7 @@ DEFINE_PER_CPU(u32, x86_cpu_to_logical_apicid); -static int __init x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { if (cpu_has_x2apic) return 1; diff --git a/arch/x86/kernel/genx2apic_phys.c b/arch/x86/kernel/genx2apic_phys.c index 8f1343df2627..d042211768b7 100644 --- a/arch/x86/kernel/genx2apic_phys.c +++ b/arch/x86/kernel/genx2apic_phys.c @@ -19,7 +19,7 @@ static int set_x2apic_phys_mode(char *arg) } early_param("x2apic_phys", set_x2apic_phys_mode); -static int __init x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { if (cpu_has_x2apic && x2apic_phys) return 1; diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c index 33581d94a90e..dece17289731 100644 --- a/arch/x86/kernel/genx2apic_uv_x.c +++ b/arch/x86/kernel/genx2apic_uv_x.c @@ -10,14 +10,17 @@ #include <linux/kernel.h> #include <linux/threads.h> +#include <linux/cpu.h> #include <linux/cpumask.h> #include <linux/string.h> #include <linux/ctype.h> #include <linux/init.h> #include <linux/sched.h> -#include <linux/bootmem.h> #include <linux/module.h> #include <linux/hardirq.h> +#include <linux/timer.h> +#include <linux/proc_fs.h> +#include <asm/current.h> #include <asm/smp.h> #include <asm/ipi.h> #include <asm/genapic.h> @@ -30,7 +33,7 @@ DEFINE_PER_CPU(int, x2apic_extra_bits); static enum uv_system_type uv_system_type; -static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +static int uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { if (!strcmp(oem_id, "SGI")) { if (!strcmp(oem_table_id, "UVL")) @@ -341,12 +344,12 @@ static __init void map_mmioh_high(int max_pnode) static __init void uv_rtc_init(void) { - long status, ticks_per_sec, drift; + long status; + u64 ticks_per_sec; - status = - x86_bios_freq_base(BIOS_FREQ_BASE_REALTIME_CLOCK, &ticks_per_sec, - &drift); - if (status != 0 || ticks_per_sec < 100000) { + status = uv_bios_freq_base(BIOS_FREQ_BASE_REALTIME_CLOCK, + &ticks_per_sec); + if (status != BIOS_STATUS_SUCCESS || ticks_per_sec < 100000) { printk(KERN_WARNING "unable to determine platform RTC clock frequency, " "guessing.\n"); @@ -356,7 +359,119 @@ static __init void uv_rtc_init(void) sn_rtc_cycles_per_second = ticks_per_sec; } -static bool uv_system_inited; +/* + * percpu heartbeat timer + */ +static void uv_heartbeat(unsigned long ignored) +{ + struct timer_list *timer = &uv_hub_info->scir.timer; + unsigned char bits = uv_hub_info->scir.state; + + /* flip heartbeat bit */ + bits ^= SCIR_CPU_HEARTBEAT; + + /* is this cpu idle? */ + if (idle_cpu(raw_smp_processor_id())) + bits &= ~SCIR_CPU_ACTIVITY; + else + bits |= SCIR_CPU_ACTIVITY; + + /* update system controller interface reg */ + uv_set_scir_bits(bits); + + /* enable next timer period */ + mod_timer(timer, jiffies + SCIR_CPU_HB_INTERVAL); +} + +static void __cpuinit uv_heartbeat_enable(int cpu) +{ + if (!uv_cpu_hub_info(cpu)->scir.enabled) { + struct timer_list *timer = &uv_cpu_hub_info(cpu)->scir.timer; + + uv_set_cpu_scir_bits(cpu, SCIR_CPU_HEARTBEAT|SCIR_CPU_ACTIVITY); + setup_timer(timer, uv_heartbeat, cpu); + timer->expires = jiffies + SCIR_CPU_HB_INTERVAL; + add_timer_on(timer, cpu); + uv_cpu_hub_info(cpu)->scir.enabled = 1; + } + + /* check boot cpu */ + if (!uv_cpu_hub_info(0)->scir.enabled) + uv_heartbeat_enable(0); +} + +#ifdef CONFIG_HOTPLUG_CPU +static void __cpuinit uv_heartbeat_disable(int cpu) +{ + if (uv_cpu_hub_info(cpu)->scir.enabled) { + uv_cpu_hub_info(cpu)->scir.enabled = 0; + del_timer(&uv_cpu_hub_info(cpu)->scir.timer); + } + uv_set_cpu_scir_bits(cpu, 0xff); +} + +/* + * cpu hotplug notifier + */ +static __cpuinit int uv_scir_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + long cpu = (long)hcpu; + + switch (action) { + case CPU_ONLINE: + uv_heartbeat_enable(cpu); + break; + case CPU_DOWN_PREPARE: + uv_heartbeat_disable(cpu); + break; + default: + break; + } + return NOTIFY_OK; +} + +static __init void uv_scir_register_cpu_notifier(void) +{ + hotcpu_notifier(uv_scir_cpu_notify, 0); +} + +#else /* !CONFIG_HOTPLUG_CPU */ + +static __init void uv_scir_register_cpu_notifier(void) +{ +} + +static __init int uv_init_heartbeat(void) +{ + int cpu; + + if (is_uv_system()) + for_each_online_cpu(cpu) + uv_heartbeat_enable(cpu); + return 0; +} + +late_initcall(uv_init_heartbeat); + +#endif /* !CONFIG_HOTPLUG_CPU */ + +/* + * Called on each cpu to initialize the per_cpu UV data area. + * ZZZ hotplug not supported yet + */ +void __cpuinit uv_cpu_init(void) +{ + /* CPU 0 initilization will be done via uv_system_init. */ + if (!uv_blade_info) + return; + + uv_blade_info[uv_numa_blade_id()].nr_online_cpus++; + + if (get_uv_system_type() == UV_NON_UNIQUE_APIC) + set_x2apic_extra_bits(uv_hub_info->pnode); +} + void __init uv_system_init(void) { @@ -383,16 +498,16 @@ void __init uv_system_init(void) printk(KERN_DEBUG "UV: Found %d blades\n", uv_num_possible_blades()); bytes = sizeof(struct uv_blade_info) * uv_num_possible_blades(); - uv_blade_info = alloc_bootmem_pages(bytes); + uv_blade_info = kmalloc(bytes, GFP_KERNEL); get_lowmem_redirect(&lowmem_redir_base, &lowmem_redir_size); bytes = sizeof(uv_node_to_blade[0]) * num_possible_nodes(); - uv_node_to_blade = alloc_bootmem_pages(bytes); + uv_node_to_blade = kmalloc(bytes, GFP_KERNEL); memset(uv_node_to_blade, 255, bytes); bytes = sizeof(uv_cpu_to_blade[0]) * num_possible_cpus(); - uv_cpu_to_blade = alloc_bootmem_pages(bytes); + uv_cpu_to_blade = kmalloc(bytes, GFP_KERNEL); memset(uv_cpu_to_blade, 255, bytes); blade = 0; @@ -412,6 +527,9 @@ void __init uv_system_init(void) gnode_upper = (((unsigned long)node_id.s.node_id) & ~((1 << n_val) - 1)) << m_val; + uv_bios_init(); + uv_bios_get_sn_info(0, &uv_type, &sn_partition_id, + &sn_coherency_id, &sn_region_size); uv_rtc_init(); for_each_present_cpu(cpu) { @@ -422,8 +540,7 @@ void __init uv_system_init(void) uv_blade_info[blade].nr_possible_cpus++; uv_cpu_hub_info(cpu)->lowmem_remap_base = lowmem_redir_base; - uv_cpu_hub_info(cpu)->lowmem_remap_top = - lowmem_redir_base + lowmem_redir_size; + uv_cpu_hub_info(cpu)->lowmem_remap_top = lowmem_redir_size; uv_cpu_hub_info(cpu)->m_val = m_val; uv_cpu_hub_info(cpu)->n_val = m_val; uv_cpu_hub_info(cpu)->numa_blade_id = blade; @@ -433,7 +550,8 @@ void __init uv_system_init(void) uv_cpu_hub_info(cpu)->gpa_mask = (1 << (m_val + n_val)) - 1; uv_cpu_hub_info(cpu)->gnode_upper = gnode_upper; uv_cpu_hub_info(cpu)->global_mmr_base = mmr_base; - uv_cpu_hub_info(cpu)->coherency_domain_number = 0;/* ZZZ */ + uv_cpu_hub_info(cpu)->coherency_domain_number = sn_coherency_id; + uv_cpu_hub_info(cpu)->scir.offset = SCIR_LOCAL_MMR_BASE + lcpu; uv_node_to_blade[nid] = blade; uv_cpu_to_blade[cpu] = blade; max_pnode = max(pnode, max_pnode); @@ -448,21 +566,8 @@ void __init uv_system_init(void) map_mmr_high(max_pnode); map_config_high(max_pnode); map_mmioh_high(max_pnode); - uv_system_inited = true; -} - -/* - * Called on each cpu to initialize the per_cpu UV data area. - * ZZZ hotplug not supported yet - */ -void __cpuinit uv_cpu_init(void) -{ - BUG_ON(!uv_system_inited); - - uv_blade_info[uv_numa_blade_id()].nr_online_cpus++; - if (get_uv_system_type() == UV_NON_UNIQUE_APIC) - set_x2apic_extra_bits(uv_hub_info->pnode); + uv_cpu_init(); + uv_scir_register_cpu_notifier(); + proc_mkdir("sgi_uv", NULL); } - - diff --git a/arch/x86/kernel/head.c b/arch/x86/kernel/head.c index 1dcb0f13897e..3e66bd364a9d 100644 --- a/arch/x86/kernel/head.c +++ b/arch/x86/kernel/head.c @@ -35,7 +35,6 @@ void __init reserve_ebda_region(void) /* start of EBDA area */ ebda_addr = get_bios_ebda(); - printk(KERN_INFO "BIOS EBDA/lowmem at: %08x/%08x\n", ebda_addr, lowmem); /* Fixup: bios puts an EBDA in the top 64K segment */ /* of conventional memory, but does not adjust lowmem. */ diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index fa1d25dd83e3..ac108d1fe182 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -12,9 +12,12 @@ #include <asm/sections.h> #include <asm/e820.h> #include <asm/bios_ebda.h> +#include <asm/trampoline.h> void __init i386_start_kernel(void) { + reserve_trampoline_memory(); + reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index d16084f90649..388e05a5fc17 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -24,6 +24,7 @@ #include <asm/kdebug.h> #include <asm/e820.h> #include <asm/bios_ebda.h> +#include <asm/trampoline.h> /* boot cpu pda */ static struct x8664_pda _boot_cpu_pda __read_mostly; @@ -120,6 +121,8 @@ void __init x86_64_start_reservations(char *real_mode_data) { copy_bootdata(__va(real_mode_data)); + reserve_trampoline_memory(); + reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index acf62fc233da..845ea097383e 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -1,29 +1,51 @@ #include <linux/clocksource.h> #include <linux/clockchips.h> +#include <linux/interrupt.h> +#include <linux/sysdev.h> #include <linux/delay.h> #include <linux/errno.h> #include <linux/hpet.h> #include <linux/init.h> -#include <linux/sysdev.h> +#include <linux/cpu.h> #include <linux/pm.h> +#include <linux/io.h> #include <asm/fixmap.h> -#include <asm/hpet.h> #include <asm/i8253.h> -#include <asm/io.h> +#include <asm/hpet.h> -#define HPET_MASK CLOCKSOURCE_MASK(32) -#define HPET_SHIFT 22 +#define HPET_MASK CLOCKSOURCE_MASK(32) +#define HPET_SHIFT 22 /* FSEC = 10^-15 NSEC = 10^-9 */ -#define FSEC_PER_NSEC 1000000L +#define FSEC_PER_NSEC 1000000L + +#define HPET_DEV_USED_BIT 2 +#define HPET_DEV_USED (1 << HPET_DEV_USED_BIT) +#define HPET_DEV_VALID 0x8 +#define HPET_DEV_FSB_CAP 0x1000 +#define HPET_DEV_PERI_CAP 0x2000 + +#define EVT_TO_HPET_DEV(evt) container_of(evt, struct hpet_dev, evt) /* * HPET address is set in acpi/boot.c, when an ACPI entry exists */ -unsigned long hpet_address; -static void __iomem *hpet_virt_address; +unsigned long hpet_address; +#ifdef CONFIG_PCI_MSI +static unsigned long hpet_num_timers; +#endif +static void __iomem *hpet_virt_address; + +struct hpet_dev { + struct clock_event_device evt; + unsigned int num; + int cpu; + unsigned int irq; + unsigned int flags; + char name[10]; +}; unsigned long hpet_readl(unsigned long a) { @@ -59,7 +81,7 @@ static inline void hpet_clear_mapping(void) static int boot_hpet_disable; int hpet_force_user; -static int __init hpet_setup(char* str) +static int __init hpet_setup(char *str) { if (str) { if (!strncmp("disable", str, 7)) @@ -80,7 +102,7 @@ __setup("nohpet", disable_hpet); static inline int is_hpet_capable(void) { - return (!boot_hpet_disable && hpet_address); + return !boot_hpet_disable && hpet_address; } /* @@ -102,6 +124,9 @@ EXPORT_SYMBOL_GPL(is_hpet_enabled); * timer 0 and timer 1 in case of RTC emulation. */ #ifdef CONFIG_HPET + +static void hpet_reserve_msi_timers(struct hpet_data *hd); + static void hpet_reserve_platform_timers(unsigned long id) { struct hpet __iomem *hpet = hpet_virt_address; @@ -111,10 +136,10 @@ static void hpet_reserve_platform_timers(unsigned long id) nrtimers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT) + 1; - memset(&hd, 0, sizeof (hd)); - hd.hd_phys_address = hpet_address; - hd.hd_address = hpet; - hd.hd_nirqs = nrtimers; + memset(&hd, 0, sizeof(hd)); + hd.hd_phys_address = hpet_address; + hd.hd_address = hpet; + hd.hd_nirqs = nrtimers; hpet_reserve_timer(&hd, 0); #ifdef CONFIG_HPET_EMULATE_RTC @@ -130,10 +155,12 @@ static void hpet_reserve_platform_timers(unsigned long id) hd.hd_irq[1] = HPET_LEGACY_RTC; for (i = 2; i < nrtimers; timer++, i++) { - hd.hd_irq[i] = (readl(&timer->hpet_config) & Tn_INT_ROUTE_CNF_MASK) >> - Tn_INT_ROUTE_CNF_SHIFT; + hd.hd_irq[i] = (readl(&timer->hpet_config) & + Tn_INT_ROUTE_CNF_MASK) >> Tn_INT_ROUTE_CNF_SHIFT; } + hpet_reserve_msi_timers(&hd); + hpet_alloc(&hd); } @@ -227,71 +254,422 @@ static void hpet_legacy_clockevent_register(void) printk(KERN_DEBUG "hpet clockevent registered\n"); } -static void hpet_legacy_set_mode(enum clock_event_mode mode, - struct clock_event_device *evt) +static int hpet_setup_msi_irq(unsigned int irq); + +static void hpet_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt, int timer) { unsigned long cfg, cmp, now; uint64_t delta; - switch(mode) { + switch (mode) { case CLOCK_EVT_MODE_PERIODIC: - delta = ((uint64_t)(NSEC_PER_SEC/HZ)) * hpet_clockevent.mult; - delta >>= hpet_clockevent.shift; + delta = ((uint64_t)(NSEC_PER_SEC/HZ)) * evt->mult; + delta >>= evt->shift; now = hpet_readl(HPET_COUNTER); cmp = now + (unsigned long) delta; - cfg = hpet_readl(HPET_T0_CFG); + cfg = hpet_readl(HPET_Tn_CFG(timer)); cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL | HPET_TN_32BIT; - hpet_writel(cfg, HPET_T0_CFG); + hpet_writel(cfg, HPET_Tn_CFG(timer)); /* * The first write after writing TN_SETVAL to the * config register sets the counter value, the second * write sets the period. */ - hpet_writel(cmp, HPET_T0_CMP); + hpet_writel(cmp, HPET_Tn_CMP(timer)); udelay(1); - hpet_writel((unsigned long) delta, HPET_T0_CMP); + hpet_writel((unsigned long) delta, HPET_Tn_CMP(timer)); break; case CLOCK_EVT_MODE_ONESHOT: - cfg = hpet_readl(HPET_T0_CFG); + cfg = hpet_readl(HPET_Tn_CFG(timer)); cfg &= ~HPET_TN_PERIODIC; cfg |= HPET_TN_ENABLE | HPET_TN_32BIT; - hpet_writel(cfg, HPET_T0_CFG); + hpet_writel(cfg, HPET_Tn_CFG(timer)); break; case CLOCK_EVT_MODE_UNUSED: case CLOCK_EVT_MODE_SHUTDOWN: - cfg = hpet_readl(HPET_T0_CFG); + cfg = hpet_readl(HPET_Tn_CFG(timer)); cfg &= ~HPET_TN_ENABLE; - hpet_writel(cfg, HPET_T0_CFG); + hpet_writel(cfg, HPET_Tn_CFG(timer)); break; case CLOCK_EVT_MODE_RESUME: - hpet_enable_legacy_int(); + if (timer == 0) { + hpet_enable_legacy_int(); + } else { + struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt); + hpet_setup_msi_irq(hdev->irq); + disable_irq(hdev->irq); + irq_set_affinity(hdev->irq, cpumask_of_cpu(hdev->cpu)); + enable_irq(hdev->irq); + } break; } } -static int hpet_legacy_next_event(unsigned long delta, - struct clock_event_device *evt) +static int hpet_next_event(unsigned long delta, + struct clock_event_device *evt, int timer) { u32 cnt; cnt = hpet_readl(HPET_COUNTER); cnt += (u32) delta; - hpet_writel(cnt, HPET_T0_CMP); + hpet_writel(cnt, HPET_Tn_CMP(timer)); /* * We need to read back the CMP register to make sure that * what we wrote hit the chip before we compare it to the * counter. */ - WARN_ON((u32)hpet_readl(HPET_T0_CMP) != cnt); + WARN_ON_ONCE((u32)hpet_readl(HPET_Tn_CMP(timer)) != cnt); return (s32)((u32)hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0; } +static void hpet_legacy_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + hpet_set_mode(mode, evt, 0); +} + +static int hpet_legacy_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + return hpet_next_event(delta, evt, 0); +} + +/* + * HPET MSI Support + */ +#ifdef CONFIG_PCI_MSI + +static DEFINE_PER_CPU(struct hpet_dev *, cpu_hpet_dev); +static struct hpet_dev *hpet_devs; + +void hpet_msi_unmask(unsigned int irq) +{ + struct hpet_dev *hdev = get_irq_data(irq); + unsigned long cfg; + + /* unmask it */ + cfg = hpet_readl(HPET_Tn_CFG(hdev->num)); + cfg |= HPET_TN_FSB; + hpet_writel(cfg, HPET_Tn_CFG(hdev->num)); +} + +void hpet_msi_mask(unsigned int irq) +{ + unsigned long cfg; + struct hpet_dev *hdev = get_irq_data(irq); + + /* mask it */ + cfg = hpet_readl(HPET_Tn_CFG(hdev->num)); + cfg &= ~HPET_TN_FSB; + hpet_writel(cfg, HPET_Tn_CFG(hdev->num)); +} + +void hpet_msi_write(unsigned int irq, struct msi_msg *msg) +{ + struct hpet_dev *hdev = get_irq_data(irq); + + hpet_writel(msg->data, HPET_Tn_ROUTE(hdev->num)); + hpet_writel(msg->address_lo, HPET_Tn_ROUTE(hdev->num) + 4); +} + +void hpet_msi_read(unsigned int irq, struct msi_msg *msg) +{ + struct hpet_dev *hdev = get_irq_data(irq); + + msg->data = hpet_readl(HPET_Tn_ROUTE(hdev->num)); + msg->address_lo = hpet_readl(HPET_Tn_ROUTE(hdev->num) + 4); + msg->address_hi = 0; +} + +static void hpet_msi_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt); + hpet_set_mode(mode, evt, hdev->num); +} + +static int hpet_msi_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt); + return hpet_next_event(delta, evt, hdev->num); +} + +static int hpet_setup_msi_irq(unsigned int irq) +{ + if (arch_setup_hpet_msi(irq)) { + destroy_irq(irq); + return -EINVAL; + } + return 0; +} + +static int hpet_assign_irq(struct hpet_dev *dev) +{ + unsigned int irq; + + irq = create_irq(); + if (!irq) + return -EINVAL; + + set_irq_data(irq, dev); + + if (hpet_setup_msi_irq(irq)) + return -EINVAL; + + dev->irq = irq; + return 0; +} + +static irqreturn_t hpet_interrupt_handler(int irq, void *data) +{ + struct hpet_dev *dev = (struct hpet_dev *)data; + struct clock_event_device *hevt = &dev->evt; + + if (!hevt->event_handler) { + printk(KERN_INFO "Spurious HPET timer interrupt on HPET timer %d\n", + dev->num); + return IRQ_HANDLED; + } + + hevt->event_handler(hevt); + return IRQ_HANDLED; +} + +static int hpet_setup_irq(struct hpet_dev *dev) +{ + + if (request_irq(dev->irq, hpet_interrupt_handler, + IRQF_DISABLED|IRQF_NOBALANCING, dev->name, dev)) + return -1; + + disable_irq(dev->irq); + irq_set_affinity(dev->irq, cpumask_of_cpu(dev->cpu)); + enable_irq(dev->irq); + + printk(KERN_DEBUG "hpet: %s irq %d for MSI\n", + dev->name, dev->irq); + + return 0; +} + +/* This should be called in specific @cpu */ +static void init_one_hpet_msi_clockevent(struct hpet_dev *hdev, int cpu) +{ + struct clock_event_device *evt = &hdev->evt; + uint64_t hpet_freq; + + WARN_ON(cpu != smp_processor_id()); + if (!(hdev->flags & HPET_DEV_VALID)) + return; + + if (hpet_setup_msi_irq(hdev->irq)) + return; + + hdev->cpu = cpu; + per_cpu(cpu_hpet_dev, cpu) = hdev; + evt->name = hdev->name; + hpet_setup_irq(hdev); + evt->irq = hdev->irq; + + evt->rating = 110; + evt->features = CLOCK_EVT_FEAT_ONESHOT; + if (hdev->flags & HPET_DEV_PERI_CAP) + evt->features |= CLOCK_EVT_FEAT_PERIODIC; + + evt->set_mode = hpet_msi_set_mode; + evt->set_next_event = hpet_msi_next_event; + evt->shift = 32; + + /* + * The period is a femto seconds value. We need to calculate the + * scaled math multiplication factor for nanosecond to hpet tick + * conversion. + */ + hpet_freq = 1000000000000000ULL; + do_div(hpet_freq, hpet_period); + evt->mult = div_sc((unsigned long) hpet_freq, + NSEC_PER_SEC, evt->shift); + /* Calculate the max delta */ + evt->max_delta_ns = clockevent_delta2ns(0x7FFFFFFF, evt); + /* 5 usec minimum reprogramming delta. */ + evt->min_delta_ns = 5000; + + evt->cpumask = cpumask_of_cpu(hdev->cpu); + clockevents_register_device(evt); +} + +#ifdef CONFIG_HPET +/* Reserve at least one timer for userspace (/dev/hpet) */ +#define RESERVE_TIMERS 1 +#else +#define RESERVE_TIMERS 0 +#endif + +static void hpet_msi_capability_lookup(unsigned int start_timer) +{ + unsigned int id; + unsigned int num_timers; + unsigned int num_timers_used = 0; + int i; + + id = hpet_readl(HPET_ID); + + num_timers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT); + num_timers++; /* Value read out starts from 0 */ + + hpet_devs = kzalloc(sizeof(struct hpet_dev) * num_timers, GFP_KERNEL); + if (!hpet_devs) + return; + + hpet_num_timers = num_timers; + + for (i = start_timer; i < num_timers - RESERVE_TIMERS; i++) { + struct hpet_dev *hdev = &hpet_devs[num_timers_used]; + unsigned long cfg = hpet_readl(HPET_Tn_CFG(i)); + + /* Only consider HPET timer with MSI support */ + if (!(cfg & HPET_TN_FSB_CAP)) + continue; + + hdev->flags = 0; + if (cfg & HPET_TN_PERIODIC_CAP) + hdev->flags |= HPET_DEV_PERI_CAP; + hdev->num = i; + + sprintf(hdev->name, "hpet%d", i); + if (hpet_assign_irq(hdev)) + continue; + + hdev->flags |= HPET_DEV_FSB_CAP; + hdev->flags |= HPET_DEV_VALID; + num_timers_used++; + if (num_timers_used == num_possible_cpus()) + break; + } + + printk(KERN_INFO "HPET: %d timers in total, %d timers will be used for per-cpu timer\n", + num_timers, num_timers_used); +} + +#ifdef CONFIG_HPET +static void hpet_reserve_msi_timers(struct hpet_data *hd) +{ + int i; + + if (!hpet_devs) + return; + + for (i = 0; i < hpet_num_timers; i++) { + struct hpet_dev *hdev = &hpet_devs[i]; + + if (!(hdev->flags & HPET_DEV_VALID)) + continue; + + hd->hd_irq[hdev->num] = hdev->irq; + hpet_reserve_timer(hd, hdev->num); + } +} +#endif + +static struct hpet_dev *hpet_get_unused_timer(void) +{ + int i; + + if (!hpet_devs) + return NULL; + + for (i = 0; i < hpet_num_timers; i++) { + struct hpet_dev *hdev = &hpet_devs[i]; + + if (!(hdev->flags & HPET_DEV_VALID)) + continue; + if (test_and_set_bit(HPET_DEV_USED_BIT, + (unsigned long *)&hdev->flags)) + continue; + return hdev; + } + return NULL; +} + +struct hpet_work_struct { + struct delayed_work work; + struct completion complete; +}; + +static void hpet_work(struct work_struct *w) +{ + struct hpet_dev *hdev; + int cpu = smp_processor_id(); + struct hpet_work_struct *hpet_work; + + hpet_work = container_of(w, struct hpet_work_struct, work.work); + + hdev = hpet_get_unused_timer(); + if (hdev) + init_one_hpet_msi_clockevent(hdev, cpu); + + complete(&hpet_work->complete); +} + +static int hpet_cpuhp_notify(struct notifier_block *n, + unsigned long action, void *hcpu) +{ + unsigned long cpu = (unsigned long)hcpu; + struct hpet_work_struct work; + struct hpet_dev *hdev = per_cpu(cpu_hpet_dev, cpu); + + switch (action & 0xf) { + case CPU_ONLINE: + INIT_DELAYED_WORK(&work.work, hpet_work); + init_completion(&work.complete); + /* FIXME: add schedule_work_on() */ + schedule_delayed_work_on(cpu, &work.work, 0); + wait_for_completion(&work.complete); + break; + case CPU_DEAD: + if (hdev) { + free_irq(hdev->irq, hdev); + hdev->flags &= ~HPET_DEV_USED; + per_cpu(cpu_hpet_dev, cpu) = NULL; + } + break; + } + return NOTIFY_OK; +} +#else + +static int hpet_setup_msi_irq(unsigned int irq) +{ + return 0; +} +static void hpet_msi_capability_lookup(unsigned int start_timer) +{ + return; +} + +#ifdef CONFIG_HPET +static void hpet_reserve_msi_timers(struct hpet_data *hd) +{ + return; +} +#endif + +static int hpet_cpuhp_notify(struct notifier_block *n, + unsigned long action, void *hcpu) +{ + return NOTIFY_OK; +} + +#endif + /* * Clock source related code */ @@ -427,13 +805,15 @@ int __init hpet_enable(void) if (id & HPET_ID_LEGSUP) { hpet_legacy_clockevent_register(); + hpet_msi_capability_lookup(2); return 1; } + hpet_msi_capability_lookup(0); return 0; out_nohpet: hpet_clear_mapping(); - boot_hpet_disable = 1; + hpet_address = 0; return 0; } @@ -445,6 +825,8 @@ out_nohpet: */ static __init int hpet_late_init(void) { + int cpu; + if (boot_hpet_disable) return -ENODEV; @@ -454,12 +836,20 @@ static __init int hpet_late_init(void) hpet_address = force_hpet_address; hpet_enable(); - if (!hpet_virt_address) - return -ENODEV; } + if (!hpet_virt_address) + return -ENODEV; + hpet_reserve_platform_timers(hpet_readl(HPET_ID)); + for_each_online_cpu(cpu) { + hpet_cpuhp_notify(NULL, CPU_ONLINE, (void *)(long)cpu); + } + + /* This notifier should be called after workqueue is ready */ + hotcpu_notifier(hpet_cpuhp_notify, -20); + return 0; } fs_initcall(hpet_late_init); diff --git a/arch/x86/kernel/i386_ksyms_32.c b/arch/x86/kernel/i386_ksyms_32.c index dd7ebee446af..43cec6bdda63 100644 --- a/arch/x86/kernel/i386_ksyms_32.c +++ b/arch/x86/kernel/i386_ksyms_32.c @@ -5,7 +5,7 @@ #include <asm/desc.h> #include <asm/ftrace.h> -#ifdef CONFIG_FTRACE +#ifdef CONFIG_FUNCTION_TRACER /* mcount is defined in assembly */ EXPORT_SYMBOL(mcount); #endif diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 1f20608d4ca8..b0f61f0dcd0a 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -58,7 +58,7 @@ void __cpuinit mxcsr_feature_mask_init(void) stts(); } -void __init init_thread_xstate(void) +void __cpuinit init_thread_xstate(void) { if (!HAVE_HWFP) { xstate_size = sizeof(struct i387_soft_struct); diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c index a4f93b4120c1..d39918076bb4 100644 --- a/arch/x86/kernel/init_task.c +++ b/arch/x86/kernel/init_task.c @@ -14,7 +14,6 @@ static struct fs_struct init_fs = INIT_FS; static struct signal_struct init_signals = INIT_SIGNALS(init_signals); static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand); struct mm_struct init_mm = INIT_MM(init_mm); -EXPORT_UNUSED_SYMBOL(init_mm); /* will be removed in 2.6.26 */ /* * Initial thread structure. diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic.c index 02063ae042f7..f6ea94b74da1 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic.c @@ -27,17 +27,21 @@ #include <linux/sched.h> #include <linux/pci.h> #include <linux/mc146818rtc.h> +#include <linux/compiler.h> #include <linux/acpi.h> +#include <linux/module.h> #include <linux/sysdev.h> #include <linux/msi.h> #include <linux/htirq.h> -#include <linux/dmar.h> -#include <linux/jiffies.h> +#include <linux/freezer.h> +#include <linux/kthread.h> +#include <linux/jiffies.h> /* time_after() */ #ifdef CONFIG_ACPI #include <acpi/acpi_bus.h> #endif #include <linux/bootmem.h> #include <linux/dmar.h> +#include <linux/hpet.h> #include <asm/idle.h> #include <asm/io.h> @@ -46,27 +50,108 @@ #include <asm/proto.h> #include <asm/acpi.h> #include <asm/dma.h> +#include <asm/timer.h> #include <asm/i8259.h> #include <asm/nmi.h> #include <asm/msidef.h> #include <asm/hypertransport.h> +#include <asm/setup.h> #include <asm/irq_remapping.h> +#include <asm/hpet.h> +#include <asm/uv/uv_hub.h> +#include <asm/uv/uv_irq.h> #include <mach_ipi.h> #include <mach_apic.h> +#include <mach_apicdef.h> #define __apicdebuginit(type) static type __init +/* + * Is the SiS APIC rmw bug present ? + * -1 = don't know, 0 = no, 1 = yes + */ +int sis_apic_bug = -1; + +static DEFINE_SPINLOCK(ioapic_lock); +static DEFINE_SPINLOCK(vector_lock); + +/* + * # of IRQ routing registers + */ +int nr_ioapic_registers[MAX_IO_APICS]; + +/* I/O APIC entries */ +struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; +int nr_ioapics; + +/* MP IRQ source entries */ +struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; + +/* # of MP IRQ source entries */ +int mp_irq_entries; + +#if defined (CONFIG_MCA) || defined (CONFIG_EISA) +int mp_bus_id_to_type[MAX_MP_BUSSES]; +#endif + +DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); + +int skip_ioapic_setup; + +static int __init parse_noapic(char *str) +{ + /* disable IO-APIC */ + disable_ioapic_setup(); + return 0; +} +early_param("noapic", parse_noapic); + +struct irq_pin_list; + +/* + * This is performance-critical, we want to do it O(1) + * + * the indexing order of this array favors 1:1 mappings + * between pins and IRQs. + */ + +struct irq_pin_list { + int apic, pin; + struct irq_pin_list *next; +}; + +static struct irq_pin_list *get_one_free_irq_2_pin(int cpu) +{ + struct irq_pin_list *pin; + int node; + + node = cpu_to_node(cpu); + + pin = kzalloc_node(sizeof(*pin), GFP_ATOMIC, node); + printk(KERN_DEBUG " alloc irq_2_pin on cpu %d node %d\n", cpu, node); + + return pin; +} + struct irq_cfg { + struct irq_pin_list *irq_2_pin; cpumask_t domain; cpumask_t old_domain; unsigned move_cleanup_count; u8 vector; u8 move_in_progress : 1; +#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC + u8 move_desc_pending : 1; +#endif }; /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */ -static struct irq_cfg irq_cfg[NR_IRQS] __read_mostly = { +#ifdef CONFIG_SPARSE_IRQ +static struct irq_cfg irq_cfgx[] = { +#else +static struct irq_cfg irq_cfgx[NR_IRQS] = { +#endif [0] = { .domain = CPU_MASK_ALL, .vector = IRQ0_VECTOR, }, [1] = { .domain = CPU_MASK_ALL, .vector = IRQ1_VECTOR, }, [2] = { .domain = CPU_MASK_ALL, .vector = IRQ2_VECTOR, }, @@ -85,63 +170,190 @@ static struct irq_cfg irq_cfg[NR_IRQS] __read_mostly = { [15] = { .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, }, }; -static int assign_irq_vector(int irq, cpumask_t mask); +void __init arch_early_irq_init(void) +{ + struct irq_cfg *cfg; + struct irq_desc *desc; + int count; + int i; -int first_system_vector = 0xfe; + cfg = irq_cfgx; + count = ARRAY_SIZE(irq_cfgx); -char system_vectors[NR_VECTORS] = { [0 ... NR_VECTORS-1] = SYS_VECTOR_FREE}; + for (i = 0; i < count; i++) { + desc = irq_to_desc(i); + desc->chip_data = &cfg[i]; + } +} -int sis_apic_bug; /* not actually supported, dummy for compile */ +#ifdef CONFIG_SPARSE_IRQ +static struct irq_cfg *irq_cfg(unsigned int irq) +{ + struct irq_cfg *cfg = NULL; + struct irq_desc *desc; -static int no_timer_check; + desc = irq_to_desc(irq); + if (desc) + cfg = desc->chip_data; -static int disable_timer_pin_1 __initdata; + return cfg; +} -int timer_through_8259 __initdata; +static struct irq_cfg *get_one_free_irq_cfg(int cpu) +{ + struct irq_cfg *cfg; + int node; -/* Where if anywhere is the i8259 connect in external int mode */ -static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; + node = cpu_to_node(cpu); -static DEFINE_SPINLOCK(ioapic_lock); -static DEFINE_SPINLOCK(vector_lock); + cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node); + printk(KERN_DEBUG " alloc irq_cfg on cpu %d node %d\n", cpu, node); -/* - * # of IRQ routing registers - */ -int nr_ioapic_registers[MAX_IO_APICS]; + return cfg; +} -/* I/O APIC RTE contents at the OS boot up */ -struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS]; +void arch_init_chip_data(struct irq_desc *desc, int cpu) +{ + struct irq_cfg *cfg; -/* I/O APIC entries */ -struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; -int nr_ioapics; + cfg = desc->chip_data; + if (!cfg) { + desc->chip_data = get_one_free_irq_cfg(cpu); + if (!desc->chip_data) { + printk(KERN_ERR "can not alloc irq_cfg\n"); + BUG_ON(1); + } + } +} -/* MP IRQ source entries */ -struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; +#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC -/* # of MP IRQ source entries */ -int mp_irq_entries; +static void +init_copy_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg, int cpu) +{ + struct irq_pin_list *old_entry, *head, *tail, *entry; -DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); + cfg->irq_2_pin = NULL; + old_entry = old_cfg->irq_2_pin; + if (!old_entry) + return; -/* - * Rough estimation of how many shared IRQs there are, can - * be changed anytime. - */ -#define MAX_PLUS_SHARED_IRQS NR_IRQS -#define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS) + entry = get_one_free_irq_2_pin(cpu); + if (!entry) + return; -/* - * This is performance-critical, we want to do it O(1) - * - * the indexing order of this array favors 1:1 mappings - * between pins and IRQs. - */ + entry->apic = old_entry->apic; + entry->pin = old_entry->pin; + head = entry; + tail = entry; + old_entry = old_entry->next; + while (old_entry) { + entry = get_one_free_irq_2_pin(cpu); + if (!entry) { + entry = head; + while (entry) { + head = entry->next; + kfree(entry); + entry = head; + } + /* still use the old one */ + return; + } + entry->apic = old_entry->apic; + entry->pin = old_entry->pin; + tail->next = entry; + tail = entry; + old_entry = old_entry->next; + } + + tail->next = NULL; + cfg->irq_2_pin = head; +} + +static void free_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg) +{ + struct irq_pin_list *entry, *next; + + if (old_cfg->irq_2_pin == cfg->irq_2_pin) + return; + + entry = old_cfg->irq_2_pin; + + while (entry) { + next = entry->next; + kfree(entry); + entry = next; + } + old_cfg->irq_2_pin = NULL; +} + +void arch_init_copy_chip_data(struct irq_desc *old_desc, + struct irq_desc *desc, int cpu) +{ + struct irq_cfg *cfg; + struct irq_cfg *old_cfg; + + cfg = get_one_free_irq_cfg(cpu); + + if (!cfg) + return; + + desc->chip_data = cfg; + + old_cfg = old_desc->chip_data; + + memcpy(cfg, old_cfg, sizeof(struct irq_cfg)); + + init_copy_irq_2_pin(old_cfg, cfg, cpu); +} + +static void free_irq_cfg(struct irq_cfg *old_cfg) +{ + kfree(old_cfg); +} + +void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) +{ + struct irq_cfg *old_cfg, *cfg; + + old_cfg = old_desc->chip_data; + cfg = desc->chip_data; + + if (old_cfg == cfg) + return; + + if (old_cfg) { + free_irq_2_pin(old_cfg, cfg); + free_irq_cfg(old_cfg); + old_desc->chip_data = NULL; + } +} + +static void set_extra_move_desc(struct irq_desc *desc, cpumask_t mask) +{ + struct irq_cfg *cfg = desc->chip_data; + + if (!cfg->move_in_progress) { + /* it means that domain is not changed */ + if (!cpus_intersects(desc->affinity, mask)) + cfg->move_desc_pending = 1; + } +} +#endif + +#else +static struct irq_cfg *irq_cfg(unsigned int irq) +{ + return irq < nr_irqs ? irq_cfgx + irq : NULL; +} + +#endif -static struct irq_pin_list { - short apic, pin, next; -} irq_2_pin[PIN_MAP_SIZE]; +#ifndef CONFIG_NUMA_MIGRATE_IRQ_DESC +static inline void set_extra_move_desc(struct irq_desc *desc, cpumask_t mask) +{ +} +#endif struct io_apic { unsigned int index; @@ -172,27 +384,32 @@ static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned i /* * Re-write a value: to be used for read-modify-write * cycles where the read already set up the index register. + * + * Older SiS APIC requires we rewrite the index register */ -static inline void io_apic_modify(unsigned int apic, unsigned int value) +static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value) { struct io_apic __iomem *io_apic = io_apic_base(apic); + + if (sis_apic_bug) + writel(reg, &io_apic->index); writel(value, &io_apic->data); } -static bool io_apic_level_ack_pending(unsigned int irq) +static bool io_apic_level_ack_pending(struct irq_cfg *cfg) { struct irq_pin_list *entry; unsigned long flags; spin_lock_irqsave(&ioapic_lock, flags); - entry = irq_2_pin + irq; + entry = cfg->irq_2_pin; for (;;) { unsigned int reg; int pin; - pin = entry->pin; - if (pin == -1) + if (!entry) break; + pin = entry->pin; reg = io_apic_read(entry->apic, 0x10 + pin*2); /* Is the remote IRR bit set? */ if (reg & IO_APIC_REDIR_REMOTE_IRR) { @@ -201,45 +418,13 @@ static bool io_apic_level_ack_pending(unsigned int irq) } if (!entry->next) break; - entry = irq_2_pin + entry->next; + entry = entry->next; } spin_unlock_irqrestore(&ioapic_lock, flags); return false; } -/* - * Synchronize the IO-APIC and the CPU by doing - * a dummy read from the IO-APIC - */ -static inline void io_apic_sync(unsigned int apic) -{ - struct io_apic __iomem *io_apic = io_apic_base(apic); - readl(&io_apic->data); -} - -#define __DO_ACTION(R, ACTION, FINAL) \ - \ -{ \ - int pin; \ - struct irq_pin_list *entry = irq_2_pin + irq; \ - \ - BUG_ON(irq >= NR_IRQS); \ - for (;;) { \ - unsigned int reg; \ - pin = entry->pin; \ - if (pin == -1) \ - break; \ - reg = io_apic_read(entry->apic, 0x10 + R + pin*2); \ - reg ACTION; \ - io_apic_modify(entry->apic, reg); \ - FINAL; \ - if (!entry->next) \ - break; \ - entry = irq_2_pin + entry->next; \ - } \ -} - union entry_union { struct { u32 w1, w2; }; struct IO_APIC_route_entry entry; @@ -296,83 +481,118 @@ static void ioapic_mask_entry(int apic, int pin) } #ifdef CONFIG_SMP -static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector) +static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq_cfg *cfg) { int apic, pin; - struct irq_pin_list *entry = irq_2_pin + irq; + struct irq_pin_list *entry; + u8 vector = cfg->vector; - BUG_ON(irq >= NR_IRQS); + entry = cfg->irq_2_pin; for (;;) { unsigned int reg; + + if (!entry) + break; + apic = entry->apic; pin = entry->pin; - if (pin == -1) - break; +#ifdef CONFIG_INTR_REMAP /* * With interrupt-remapping, destination information comes * from interrupt-remapping table entry. */ if (!irq_remapped(irq)) io_apic_write(apic, 0x11 + pin*2, dest); +#else + io_apic_write(apic, 0x11 + pin*2, dest); +#endif reg = io_apic_read(apic, 0x10 + pin*2); reg &= ~IO_APIC_REDIR_VECTOR_MASK; reg |= vector; - io_apic_modify(apic, reg); + io_apic_modify(apic, 0x10 + pin*2, reg); if (!entry->next) break; - entry = irq_2_pin + entry->next; + entry = entry->next; } } -static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) +static int assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask); + +static void set_ioapic_affinity_irq_desc(struct irq_desc *desc, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; unsigned long flags; unsigned int dest; cpumask_t tmp; + unsigned int irq; cpus_and(tmp, mask, cpu_online_map); if (cpus_empty(tmp)) return; - if (assign_irq_vector(irq, mask)) + irq = desc->irq; + cfg = desc->chip_data; + if (assign_irq_vector(irq, cfg, mask)) return; + set_extra_move_desc(desc, mask); + cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); - /* * Only the high 8 bits are valid. */ dest = SET_APIC_LOGICAL_ID(dest); spin_lock_irqsave(&ioapic_lock, flags); - __target_IO_APIC_irq(irq, dest, cfg->vector); - irq_desc[irq].affinity = mask; + __target_IO_APIC_irq(irq, dest, cfg); + desc->affinity = mask; spin_unlock_irqrestore(&ioapic_lock, flags); } -#endif + +static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) +{ + struct irq_desc *desc; + + desc = irq_to_desc(irq); + + set_ioapic_affinity_irq_desc(desc, mask); +} +#endif /* CONFIG_SMP */ /* * The common case is 1:1 IRQ<->pin mappings. Sometimes there are * shared ISA-space IRQs, so we have to support them. We are super * fast in the common case, and fast for shared ISA-space IRQs. */ -static void add_pin_to_irq(unsigned int irq, int apic, int pin) +static void add_pin_to_irq_cpu(struct irq_cfg *cfg, int cpu, int apic, int pin) { - static int first_free_entry = NR_IRQS; - struct irq_pin_list *entry = irq_2_pin + irq; + struct irq_pin_list *entry; - BUG_ON(irq >= NR_IRQS); - while (entry->next) - entry = irq_2_pin + entry->next; + entry = cfg->irq_2_pin; + if (!entry) { + entry = get_one_free_irq_2_pin(cpu); + if (!entry) { + printk(KERN_ERR "can not alloc irq_2_pin to add %d - %d\n", + apic, pin); + return; + } + cfg->irq_2_pin = entry; + entry->apic = apic; + entry->pin = pin; + return; + } + + while (entry->next) { + /* not again, please */ + if (entry->apic == apic && entry->pin == pin) + return; - if (entry->pin != -1) { - entry->next = first_free_entry; - entry = irq_2_pin + entry->next; - if (++first_free_entry >= PIN_MAP_SIZE) - panic("io_apic.c: ran out of irq_2_pin entries!"); + entry = entry->next; } + + entry->next = get_one_free_irq_2_pin(cpu); + entry = entry->next; entry->apic = apic; entry->pin = pin; } @@ -380,53 +600,123 @@ static void add_pin_to_irq(unsigned int irq, int apic, int pin) /* * Reroute an IRQ to a different pin. */ -static void __init replace_pin_at_irq(unsigned int irq, +static void __init replace_pin_at_irq_cpu(struct irq_cfg *cfg, int cpu, int oldapic, int oldpin, int newapic, int newpin) { - struct irq_pin_list *entry = irq_2_pin + irq; + struct irq_pin_list *entry = cfg->irq_2_pin; + int replaced = 0; - while (1) { + while (entry) { if (entry->apic == oldapic && entry->pin == oldpin) { entry->apic = newapic; entry->pin = newpin; - } - if (!entry->next) + replaced = 1; + /* every one is different, right? */ break; - entry = irq_2_pin + entry->next; + } + entry = entry->next; } + + /* why? call replace before add? */ + if (!replaced) + add_pin_to_irq_cpu(cfg, cpu, newapic, newpin); } +static inline void io_apic_modify_irq(struct irq_cfg *cfg, + int mask_and, int mask_or, + void (*final)(struct irq_pin_list *entry)) +{ + int pin; + struct irq_pin_list *entry; -#define DO_ACTION(name,R,ACTION, FINAL) \ - \ - static void name##_IO_APIC_irq (unsigned int irq) \ - __DO_ACTION(R, ACTION, FINAL) + for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) { + unsigned int reg; + pin = entry->pin; + reg = io_apic_read(entry->apic, 0x10 + pin * 2); + reg &= mask_and; + reg |= mask_or; + io_apic_modify(entry->apic, 0x10 + pin * 2, reg); + if (final) + final(entry); + } +} -/* mask = 1 */ -DO_ACTION(__mask, 0, |= IO_APIC_REDIR_MASKED, io_apic_sync(entry->apic)) +static void __unmask_IO_APIC_irq(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL); +} -/* mask = 0 */ -DO_ACTION(__unmask, 0, &= ~IO_APIC_REDIR_MASKED, ) +#ifdef CONFIG_X86_64 +void io_apic_sync(struct irq_pin_list *entry) +{ + /* + * Synchronize the IO-APIC and the CPU by doing + * a dummy read from the IO-APIC + */ + struct io_apic __iomem *io_apic; + io_apic = io_apic_base(entry->apic); + readl(&io_apic->data); +} -static void mask_IO_APIC_irq (unsigned int irq) +static void __mask_IO_APIC_irq(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync); +} +#else /* CONFIG_X86_32 */ +static void __mask_IO_APIC_irq(struct irq_cfg *cfg) { + io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, NULL); +} + +static void __mask_and_edge_IO_APIC_irq(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~IO_APIC_REDIR_LEVEL_TRIGGER, + IO_APIC_REDIR_MASKED, NULL); +} + +static void __unmask_and_level_IO_APIC_irq(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, + IO_APIC_REDIR_LEVEL_TRIGGER, NULL); +} +#endif /* CONFIG_X86_32 */ + +static void mask_IO_APIC_irq_desc(struct irq_desc *desc) +{ + struct irq_cfg *cfg = desc->chip_data; unsigned long flags; + BUG_ON(!cfg); + spin_lock_irqsave(&ioapic_lock, flags); - __mask_IO_APIC_irq(irq); + __mask_IO_APIC_irq(cfg); spin_unlock_irqrestore(&ioapic_lock, flags); } -static void unmask_IO_APIC_irq (unsigned int irq) +static void unmask_IO_APIC_irq_desc(struct irq_desc *desc) { + struct irq_cfg *cfg = desc->chip_data; unsigned long flags; spin_lock_irqsave(&ioapic_lock, flags); - __unmask_IO_APIC_irq(irq); + __unmask_IO_APIC_irq(cfg); spin_unlock_irqrestore(&ioapic_lock, flags); } +static void mask_IO_APIC_irq(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + + mask_IO_APIC_irq_desc(desc); +} +static void unmask_IO_APIC_irq(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + + unmask_IO_APIC_irq_desc(desc); +} + static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) { struct IO_APIC_route_entry entry; @@ -450,6 +740,68 @@ static void clear_IO_APIC (void) clear_IO_APIC_pin(apic, pin); } +#if !defined(CONFIG_SMP) && defined(CONFIG_X86_32) +void send_IPI_self(int vector) +{ + unsigned int cfg; + + /* + * Wait for idle. + */ + apic_wait_icr_idle(); + cfg = APIC_DM_FIXED | APIC_DEST_SELF | vector | APIC_DEST_LOGICAL; + /* + * Send the IPI. The write to APIC_ICR fires this off. + */ + apic_write(APIC_ICR, cfg); +} +#endif /* !CONFIG_SMP && CONFIG_X86_32*/ + +#ifdef CONFIG_X86_32 +/* + * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to + * specific CPU-side IRQs. + */ + +#define MAX_PIRQS 8 +static int pirq_entries [MAX_PIRQS]; +static int pirqs_enabled; + +static int __init ioapic_pirq_setup(char *str) +{ + int i, max; + int ints[MAX_PIRQS+1]; + + get_options(str, ARRAY_SIZE(ints), ints); + + for (i = 0; i < MAX_PIRQS; i++) + pirq_entries[i] = -1; + + pirqs_enabled = 1; + apic_printk(APIC_VERBOSE, KERN_INFO + "PIRQ redirection, working around broken MP-BIOS.\n"); + max = MAX_PIRQS; + if (ints[0] < MAX_PIRQS) + max = ints[0]; + + for (i = 0; i < max; i++) { + apic_printk(APIC_VERBOSE, KERN_DEBUG + "... PIRQ%d -> IRQ %d\n", i, ints[i+1]); + /* + * PIRQs are mapped upside down, usually. + */ + pirq_entries[MAX_PIRQS-i-1] = ints[i+1]; + } + return 1; +} + +__setup("pirq=", ioapic_pirq_setup); +#endif /* CONFIG_X86_32 */ + +#ifdef CONFIG_INTR_REMAP +/* I/O APIC RTE contents at the OS boot up */ +static struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS]; + /* * Saves and masks all the unmasked IO-APIC RTE's */ @@ -474,7 +826,7 @@ int save_mask_IO_APIC_setup(void) kzalloc(sizeof(struct IO_APIC_route_entry) * nr_ioapic_registers[apic], GFP_KERNEL); if (!early_ioapic_entries[apic]) - return -ENOMEM; + goto nomem; } for (apic = 0; apic < nr_ioapics; apic++) @@ -488,17 +840,31 @@ int save_mask_IO_APIC_setup(void) ioapic_write_entry(apic, pin, entry); } } + return 0; + +nomem: + while (apic >= 0) + kfree(early_ioapic_entries[apic--]); + memset(early_ioapic_entries, 0, + ARRAY_SIZE(early_ioapic_entries)); + + return -ENOMEM; } void restore_IO_APIC_setup(void) { int apic, pin; - for (apic = 0; apic < nr_ioapics; apic++) + for (apic = 0; apic < nr_ioapics; apic++) { + if (!early_ioapic_entries[apic]) + break; for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) ioapic_write_entry(apic, pin, early_ioapic_entries[apic][pin]); + kfree(early_ioapic_entries[apic]); + early_ioapic_entries[apic] = NULL; + } } void reinit_intr_remapped_IO_APIC(int intr_remapping) @@ -512,25 +878,7 @@ void reinit_intr_remapped_IO_APIC(int intr_remapping) */ restore_IO_APIC_setup(); } - -int skip_ioapic_setup; -int ioapic_force; - -static int __init parse_noapic(char *str) -{ - disable_ioapic_setup(); - return 0; -} -early_param("noapic", parse_noapic); - -/* Actually the next is obsolete, but keep it for paranoid reasons -AK */ -static int __init disable_timer_pin_setup(char *arg) -{ - disable_timer_pin_1 = 1; - return 1; -} -__setup("disable_timer_pin_1", disable_timer_pin_setup); - +#endif /* * Find the IRQ entry number of a certain pin. @@ -634,22 +982,54 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) best_guess = irq; } } - BUG_ON(best_guess >= NR_IRQS); return best_guess; } +EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector); + +#if defined(CONFIG_EISA) || defined(CONFIG_MCA) +/* + * EISA Edge/Level control register, ELCR + */ +static int EISA_ELCR(unsigned int irq) +{ + if (irq < NR_IRQS_LEGACY) { + unsigned int port = 0x4d0 + (irq >> 3); + return (inb(port) >> (irq & 7)) & 1; + } + apic_printk(APIC_VERBOSE, KERN_INFO + "Broken MPtable reports ISA irq %d\n", irq); + return 0; +} + +#endif + /* ISA interrupts are always polarity zero edge triggered, * when listed as conforming in the MP table. */ #define default_ISA_trigger(idx) (0) #define default_ISA_polarity(idx) (0) +/* EISA interrupts are always polarity zero and can be edge or level + * trigger depending on the ELCR value. If an interrupt is listed as + * EISA conforming in the MP table, that means its trigger type must + * be read in from the ELCR */ + +#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mp_srcbusirq)) +#define default_EISA_polarity(idx) default_ISA_polarity(idx) + /* PCI interrupts are always polarity one level triggered, * when listed as conforming in the MP table. */ #define default_PCI_trigger(idx) (1) #define default_PCI_polarity(idx) (1) +/* MCA interrupts are always polarity zero level triggered, + * when listed as conforming in the MP table. */ + +#define default_MCA_trigger(idx) (1) +#define default_MCA_polarity(idx) default_ISA_polarity(idx) + static int MPBIOS_polarity(int idx) { int bus = mp_irqs[idx].mp_srcbus; @@ -707,6 +1087,36 @@ static int MPBIOS_trigger(int idx) trigger = default_ISA_trigger(idx); else trigger = default_PCI_trigger(idx); +#if defined(CONFIG_EISA) || defined(CONFIG_MCA) + switch (mp_bus_id_to_type[bus]) { + case MP_BUS_ISA: /* ISA pin */ + { + /* set before the switch */ + break; + } + case MP_BUS_EISA: /* EISA pin */ + { + trigger = default_EISA_trigger(idx); + break; + } + case MP_BUS_PCI: /* PCI pin */ + { + /* set before the switch */ + break; + } + case MP_BUS_MCA: /* MCA pin */ + { + trigger = default_MCA_trigger(idx); + break; + } + default: + { + printk(KERN_WARNING "broken BIOS!!\n"); + trigger = 1; + break; + } + } +#endif break; case 1: /* edge */ { @@ -744,6 +1154,7 @@ static inline int irq_trigger(int idx) return MPBIOS_trigger(idx); } +int (*ioapic_renumber_irq)(int ioapic, int irq); static int pin_2_irq(int idx, int apic, int pin) { int irq, i; @@ -765,8 +1176,32 @@ static int pin_2_irq(int idx, int apic, int pin) while (i < apic) irq += nr_ioapic_registers[i++]; irq += pin; + /* + * For MPS mode, so far only needed by ES7000 platform + */ + if (ioapic_renumber_irq) + irq = ioapic_renumber_irq(apic, irq); + } + +#ifdef CONFIG_X86_32 + /* + * PCI IRQ command line redirection. Yes, limits are hardcoded. + */ + if ((pin >= 16) && (pin <= 23)) { + if (pirq_entries[pin-16] != -1) { + if (!pirq_entries[pin-16]) { + apic_printk(APIC_VERBOSE, KERN_DEBUG + "disabling PIRQ%d\n", pin-16); + } else { + irq = pirq_entries[pin-16]; + apic_printk(APIC_VERBOSE, KERN_DEBUG + "using PIRQ%d -> IRQ %d\n", + pin-16, irq); + } + } } - BUG_ON(irq >= NR_IRQS); +#endif + return irq; } @@ -783,7 +1218,7 @@ void unlock_vector_lock(void) spin_unlock(&vector_lock); } -static int __assign_irq_vector(int irq, cpumask_t mask) +static int __assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask) { /* * NOTE! The local APIC isn't very good at handling @@ -799,17 +1234,13 @@ static int __assign_irq_vector(int irq, cpumask_t mask) static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0; unsigned int old_vector; int cpu; - struct irq_cfg *cfg; - BUG_ON((unsigned)irq >= NR_IRQS); - cfg = &irq_cfg[irq]; + if ((cfg->move_in_progress) || cfg->move_cleanup_count) + return -EBUSY; /* Only try and allocate irqs on cpus that are present */ cpus_and(mask, mask, cpu_online_map); - if ((cfg->move_in_progress) || cfg->move_cleanup_count) - return -EBUSY; - old_vector = cfg->vector; if (old_vector) { cpumask_t tmp; @@ -837,8 +1268,13 @@ next: } if (unlikely(current_vector == vector)) continue; +#ifdef CONFIG_X86_64 if (vector == IA32_SYSCALL_VECTOR) goto next; +#else + if (vector == SYSCALL_VECTOR) + goto next; +#endif for_each_cpu_mask_nr(new_cpu, new_mask) if (per_cpu(vector_irq, new_cpu)[vector] != -1) goto next; @@ -858,25 +1294,22 @@ next: return -ENOSPC; } -static int assign_irq_vector(int irq, cpumask_t mask) +static int assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask) { int err; unsigned long flags; spin_lock_irqsave(&vector_lock, flags); - err = __assign_irq_vector(irq, mask); + err = __assign_irq_vector(irq, cfg, mask); spin_unlock_irqrestore(&vector_lock, flags); return err; } -static void __clear_irq_vector(int irq) +static void __clear_irq_vector(int irq, struct irq_cfg *cfg) { - struct irq_cfg *cfg; cpumask_t mask; int cpu, vector; - BUG_ON((unsigned)irq >= NR_IRQS); - cfg = &irq_cfg[irq]; BUG_ON(!cfg->vector); vector = cfg->vector; @@ -886,6 +1319,20 @@ static void __clear_irq_vector(int irq) cfg->vector = 0; cpus_clear(cfg->domain); + + if (likely(!cfg->move_in_progress)) + return; + cpus_and(mask, cfg->old_domain, cpu_online_map); + for_each_cpu_mask_nr(cpu, mask) { + for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; + vector++) { + if (per_cpu(vector_irq, cpu)[vector] != irq) + continue; + per_cpu(vector_irq, cpu)[vector] = -1; + break; + } + } + cfg->move_in_progress = 0; } void __setup_vector_irq(int cpu) @@ -893,12 +1340,17 @@ void __setup_vector_irq(int cpu) /* Initialize vector_irq on a new cpu */ /* This function must be called with vector_lock held */ int irq, vector; + struct irq_cfg *cfg; + struct irq_desc *desc; /* Mark the inuse vectors */ - for (irq = 0; irq < NR_IRQS; ++irq) { - if (!cpu_isset(cpu, irq_cfg[irq].domain)) + for_each_irq_desc(irq, desc) { + if (!desc) + continue; + cfg = desc->chip_data; + if (!cpu_isset(cpu, cfg->domain)) continue; - vector = irq_cfg[irq].vector; + vector = cfg->vector; per_cpu(vector_irq, cpu)[vector] = irq; } /* Mark the free vectors */ @@ -906,7 +1358,9 @@ void __setup_vector_irq(int cpu) irq = per_cpu(vector_irq, cpu)[vector]; if (irq < 0) continue; - if (!cpu_isset(cpu, irq_cfg[irq].domain)) + + cfg = irq_cfg(irq); + if (!cpu_isset(cpu, cfg->domain)) per_cpu(vector_irq, cpu)[vector] = -1; } } @@ -916,16 +1370,46 @@ static struct irq_chip ioapic_chip; static struct irq_chip ir_ioapic_chip; #endif -static void ioapic_register_intr(int irq, unsigned long trigger) +#define IOAPIC_AUTO -1 +#define IOAPIC_EDGE 0 +#define IOAPIC_LEVEL 1 + +#ifdef CONFIG_X86_32 +static inline int IO_APIC_irq_trigger(int irq) { - if (trigger) - irq_desc[irq].status |= IRQ_LEVEL; + int apic, idx, pin; + + for (apic = 0; apic < nr_ioapics; apic++) { + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + idx = find_irq_entry(apic, pin, mp_INT); + if ((idx != -1) && (irq == pin_2_irq(idx, apic, pin))) + return irq_trigger(idx); + } + } + /* + * nonexistent IRQs are edge default + */ + return 0; +} +#else +static inline int IO_APIC_irq_trigger(int irq) +{ + return 1; +} +#endif + +static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long trigger) +{ + + if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || + trigger == IOAPIC_LEVEL) + desc->status |= IRQ_LEVEL; else - irq_desc[irq].status &= ~IRQ_LEVEL; + desc->status &= ~IRQ_LEVEL; #ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { - irq_desc[irq].status |= IRQ_MOVE_PCNTXT; + desc->status |= IRQ_MOVE_PCNTXT; if (trigger) set_irq_chip_and_handler_name(irq, &ir_ioapic_chip, handle_fasteoi_irq, @@ -936,7 +1420,8 @@ static void ioapic_register_intr(int irq, unsigned long trigger) return; } #endif - if (trigger) + if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || + trigger == IOAPIC_LEVEL) set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_fasteoi_irq, "fasteoi"); @@ -1006,18 +1491,20 @@ static int setup_ioapic_entry(int apic, int irq, return 0; } -static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, +static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, struct irq_desc *desc, int trigger, int polarity) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; struct IO_APIC_route_entry entry; cpumask_t mask; if (!IO_APIC_IRQ(irq)) return; + cfg = desc->chip_data; + mask = TARGET_CPUS; - if (assign_irq_vector(irq, mask)) + if (assign_irq_vector(irq, cfg, mask)) return; cpus_and(mask, cfg->domain, mask); @@ -1034,12 +1521,12 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, cfg->vector)) { printk("Failed to setup ioapic entry for ioapic %d, pin %d\n", mp_ioapics[apic].mp_apicid, pin); - __clear_irq_vector(irq); + __clear_irq_vector(irq, cfg); return; } - ioapic_register_intr(irq, trigger); - if (irq < 16) + ioapic_register_intr(irq, desc, trigger); + if (irq < NR_IRQS_LEGACY) disable_8259A_irq(irq); ioapic_write_entry(apic, pin, entry); @@ -1047,37 +1534,58 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, static void __init setup_IO_APIC_irqs(void) { - int apic, pin, idx, irq, first_notcon = 1; + int apic, pin, idx, irq; + int notcon = 0; + struct irq_desc *desc; + struct irq_cfg *cfg; + int cpu = boot_cpu_id; apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); for (apic = 0; apic < nr_ioapics; apic++) { - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - - idx = find_irq_entry(apic,pin,mp_INT); - if (idx == -1) { - if (first_notcon) { - apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", mp_ioapics[apic].mp_apicid, pin); - first_notcon = 0; - } else - apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mp_apicid, pin); - continue; - } - if (!first_notcon) { - apic_printk(APIC_VERBOSE, " not connected.\n"); - first_notcon = 1; - } + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + + idx = find_irq_entry(apic, pin, mp_INT); + if (idx == -1) { + if (!notcon) { + notcon = 1; + apic_printk(APIC_VERBOSE, + KERN_DEBUG " %d-%d", + mp_ioapics[apic].mp_apicid, + pin); + } else + apic_printk(APIC_VERBOSE, " %d-%d", + mp_ioapics[apic].mp_apicid, + pin); + continue; + } + if (notcon) { + apic_printk(APIC_VERBOSE, + " (apicid-pin) not connected\n"); + notcon = 0; + } - irq = pin_2_irq(idx, apic, pin); - add_pin_to_irq(irq, apic, pin); + irq = pin_2_irq(idx, apic, pin); +#ifdef CONFIG_X86_32 + if (multi_timer_check(apic, irq)) + continue; +#endif + desc = irq_to_desc_alloc_cpu(irq, cpu); + if (!desc) { + printk(KERN_INFO "can not get irq_desc for %d\n", irq); + continue; + } + cfg = desc->chip_data; + add_pin_to_irq_cpu(cfg, cpu, apic, pin); - setup_IO_APIC_irq(apic, pin, irq, - irq_trigger(idx), irq_polarity(idx)); - } + setup_IO_APIC_irq(apic, pin, irq, desc, + irq_trigger(idx), irq_polarity(idx)); + } } - if (!first_notcon) - apic_printk(APIC_VERBOSE, " not connected.\n"); + if (notcon) + apic_printk(APIC_VERBOSE, + " (apicid-pin) not connected\n"); } /* @@ -1088,8 +1596,10 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, { struct IO_APIC_route_entry entry; +#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) return; +#endif memset(&entry, 0, sizeof(entry)); @@ -1124,7 +1634,11 @@ __apicdebuginit(void) print_IO_APIC(void) union IO_APIC_reg_00 reg_00; union IO_APIC_reg_01 reg_01; union IO_APIC_reg_02 reg_02; + union IO_APIC_reg_03 reg_03; unsigned long flags; + struct irq_cfg *cfg; + struct irq_desc *desc; + unsigned int irq; if (apic_verbosity == APIC_QUIET) return; @@ -1147,12 +1661,16 @@ __apicdebuginit(void) print_IO_APIC(void) reg_01.raw = io_apic_read(apic, 1); if (reg_01.bits.version >= 0x10) reg_02.raw = io_apic_read(apic, 2); + if (reg_01.bits.version >= 0x20) + reg_03.raw = io_apic_read(apic, 3); spin_unlock_irqrestore(&ioapic_lock, flags); printk("\n"); printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid); printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); + printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type); + printk(KERN_DEBUG "....... : LTS : %X\n", reg_00.bits.LTS); printk(KERN_DEBUG ".... register #01: %08X\n", *(int *)®_01); printk(KERN_DEBUG "....... : max redirection entries: %04X\n", reg_01.bits.entries); @@ -1160,11 +1678,27 @@ __apicdebuginit(void) print_IO_APIC(void) printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ); printk(KERN_DEBUG "....... : IO APIC version: %04X\n", reg_01.bits.version); - if (reg_01.bits.version >= 0x10) { + /* + * Some Intel chipsets with IO APIC VERSION of 0x1? don't have reg_02, + * but the value of reg_02 is read as the previous read register + * value, so ignore it if reg_02 == reg_01. + */ + if (reg_01.bits.version >= 0x10 && reg_02.raw != reg_01.raw) { printk(KERN_DEBUG ".... register #02: %08X\n", reg_02.raw); printk(KERN_DEBUG "....... : arbitration: %02X\n", reg_02.bits.arbitration); } + /* + * Some Intel chipsets with IO APIC VERSION of 0x2? don't have reg_02 + * or reg_03, but the value of reg_0[23] is read as the previous read + * register value, so ignore it if reg_03 == reg_0[12]. + */ + if (reg_01.bits.version >= 0x20 && reg_03.raw != reg_02.raw && + reg_03.raw != reg_01.raw) { + printk(KERN_DEBUG ".... register #03: %08X\n", reg_03.raw); + printk(KERN_DEBUG "....... : Boot DT : %X\n", reg_03.bits.boot_DT); + } + printk(KERN_DEBUG ".... IRQ redirection table:\n"); printk(KERN_DEBUG " NR Dst Mask Trig IRR Pol" @@ -1193,16 +1727,21 @@ __apicdebuginit(void) print_IO_APIC(void) } } printk(KERN_DEBUG "IRQ to pin mappings:\n"); - for (i = 0; i < NR_IRQS; i++) { - struct irq_pin_list *entry = irq_2_pin + i; - if (entry->pin < 0) + for_each_irq_desc(irq, desc) { + struct irq_pin_list *entry; + + if (!desc) + continue; + cfg = desc->chip_data; + entry = cfg->irq_2_pin; + if (!entry) continue; - printk(KERN_DEBUG "IRQ%d ", i); + printk(KERN_DEBUG "IRQ%d ", irq); for (;;) { printk("-> %d:%d", entry->apic, entry->pin); if (!entry->next) break; - entry = irq_2_pin + entry->next; + entry = entry->next; } printk("\n"); } @@ -1236,7 +1775,7 @@ __apicdebuginit(void) print_APIC_bitfield(int base) __apicdebuginit(void) print_local_APIC(void *dummy) { unsigned int v, ver, maxlvt; - unsigned long icr; + u64 icr; if (apic_verbosity == APIC_QUIET) return; @@ -1253,20 +1792,31 @@ __apicdebuginit(void) print_local_APIC(void *dummy) v = apic_read(APIC_TASKPRI); printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK); - v = apic_read(APIC_ARBPRI); - printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v, - v & APIC_ARBPRI_MASK); - v = apic_read(APIC_PROCPRI); - printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v); + if (APIC_INTEGRATED(ver)) { /* !82489DX */ + if (!APIC_XAPIC(ver)) { + v = apic_read(APIC_ARBPRI); + printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v, + v & APIC_ARBPRI_MASK); + } + v = apic_read(APIC_PROCPRI); + printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v); + } + + /* + * Remote read supported only in the 82489DX and local APIC for + * Pentium processors. + */ + if (!APIC_INTEGRATED(ver) || maxlvt == 3) { + v = apic_read(APIC_RRR); + printk(KERN_DEBUG "... APIC RRR: %08x\n", v); + } - v = apic_read(APIC_EOI); - printk(KERN_DEBUG "... APIC EOI: %08x\n", v); - v = apic_read(APIC_RRR); - printk(KERN_DEBUG "... APIC RRR: %08x\n", v); v = apic_read(APIC_LDR); printk(KERN_DEBUG "... APIC LDR: %08x\n", v); - v = apic_read(APIC_DFR); - printk(KERN_DEBUG "... APIC DFR: %08x\n", v); + if (!x2apic_enabled()) { + v = apic_read(APIC_DFR); + printk(KERN_DEBUG "... APIC DFR: %08x\n", v); + } v = apic_read(APIC_SPIV); printk(KERN_DEBUG "... APIC SPIV: %08x\n", v); @@ -1277,8 +1827,13 @@ __apicdebuginit(void) print_local_APIC(void *dummy) printk(KERN_DEBUG "... APIC IRR field:\n"); print_APIC_bitfield(APIC_IRR); - v = apic_read(APIC_ESR); - printk(KERN_DEBUG "... APIC ESR: %08x\n", v); + if (APIC_INTEGRATED(ver)) { /* !82489DX */ + if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ + apic_write(APIC_ESR, 0); + + v = apic_read(APIC_ESR); + printk(KERN_DEBUG "... APIC ESR: %08x\n", v); + } icr = apic_icr_read(); printk(KERN_DEBUG "... APIC ICR: %08x\n", (u32)icr); @@ -1312,7 +1867,12 @@ __apicdebuginit(void) print_local_APIC(void *dummy) __apicdebuginit(void) print_all_local_APICs(void) { - on_each_cpu(print_local_APIC, NULL, 1); + int cpu; + + preempt_disable(); + for_each_online_cpu(cpu) + smp_call_function_single(cpu, print_local_APIC, NULL, 1); + preempt_enable(); } __apicdebuginit(void) print_PIC(void) @@ -1359,17 +1919,22 @@ __apicdebuginit(int) print_all_ICs(void) fs_initcall(print_all_ICs); +/* Where if anywhere is the i8259 connect in external int mode */ +static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; + void __init enable_IO_APIC(void) { union IO_APIC_reg_01 reg_01; int i8259_apic, i8259_pin; - int i, apic; + int apic; unsigned long flags; - for (i = 0; i < PIN_MAP_SIZE; i++) { - irq_2_pin[i].pin = -1; - irq_2_pin[i].next = 0; - } +#ifdef CONFIG_X86_32 + int i; + if (!pirqs_enabled) + for (i = 0; i < MAX_PIRQS; i++) + pirq_entries[i] = -1; +#endif /* * The number of IO-APIC IRQ registers (== #pins): @@ -1399,6 +1964,10 @@ void __init enable_IO_APIC(void) } found_i8259: /* Look to see what if the MP table has reported the ExtINT */ + /* If we could not find the appropriate pin by looking at the ioapic + * the i8259 probably is not connected the ioapic but give the + * mptable a chance anyway. + */ i8259_pin = find_isa_irq_pin(0, mp_ExtINT); i8259_apic = find_isa_irq_apic(0, mp_ExtINT); /* Trust the MP table if nothing is setup in the hardware */ @@ -1458,6 +2027,133 @@ void disable_IO_APIC(void) disconnect_bsp_APIC(ioapic_i8259.pin != -1); } +#ifdef CONFIG_X86_32 +/* + * function to set the IO-APIC physical IDs based on the + * values stored in the MPC table. + * + * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999 + */ + +static void __init setup_ioapic_ids_from_mpc(void) +{ + union IO_APIC_reg_00 reg_00; + physid_mask_t phys_id_present_map; + int apic; + int i; + unsigned char old_id; + unsigned long flags; + + if (x86_quirks->setup_ioapic_ids && x86_quirks->setup_ioapic_ids()) + return; + + /* + * Don't check I/O APIC IDs for xAPIC systems. They have + * no meaning without the serial APIC bus. + */ + if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + || APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) + return; + /* + * This is broken; anything with a real cpu count has to + * circumvent this idiocy regardless. + */ + phys_id_present_map = ioapic_phys_id_map(phys_cpu_present_map); + + /* + * Set the IOAPIC ID to the value stored in the MPC table. + */ + for (apic = 0; apic < nr_ioapics; apic++) { + + /* Read the register 0 value */ + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(apic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + + old_id = mp_ioapics[apic].mp_apicid; + + if (mp_ioapics[apic].mp_apicid >= get_physical_broadcast()) { + printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n", + apic, mp_ioapics[apic].mp_apicid); + printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", + reg_00.bits.ID); + mp_ioapics[apic].mp_apicid = reg_00.bits.ID; + } + + /* + * Sanity check, is the ID really free? Every APIC in a + * system must have a unique ID or we get lots of nice + * 'stuck on smp_invalidate_needed IPI wait' messages. + */ + if (check_apicid_used(phys_id_present_map, + mp_ioapics[apic].mp_apicid)) { + printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n", + apic, mp_ioapics[apic].mp_apicid); + for (i = 0; i < get_physical_broadcast(); i++) + if (!physid_isset(i, phys_id_present_map)) + break; + if (i >= get_physical_broadcast()) + panic("Max APIC ID exceeded!\n"); + printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", + i); + physid_set(i, phys_id_present_map); + mp_ioapics[apic].mp_apicid = i; + } else { + physid_mask_t tmp; + tmp = apicid_to_cpu_present(mp_ioapics[apic].mp_apicid); + apic_printk(APIC_VERBOSE, "Setting %d in the " + "phys_id_present_map\n", + mp_ioapics[apic].mp_apicid); + physids_or(phys_id_present_map, phys_id_present_map, tmp); + } + + + /* + * We need to adjust the IRQ routing table + * if the ID changed. + */ + if (old_id != mp_ioapics[apic].mp_apicid) + for (i = 0; i < mp_irq_entries; i++) + if (mp_irqs[i].mp_dstapic == old_id) + mp_irqs[i].mp_dstapic + = mp_ioapics[apic].mp_apicid; + + /* + * Read the right value from the MPC table and + * write it into the ID register. + */ + apic_printk(APIC_VERBOSE, KERN_INFO + "...changing IO-APIC physical APIC ID to %d ...", + mp_ioapics[apic].mp_apicid); + + reg_00.bits.ID = mp_ioapics[apic].mp_apicid; + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(apic, 0, reg_00.raw); + spin_unlock_irqrestore(&ioapic_lock, flags); + + /* + * Sanity check + */ + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(apic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + if (reg_00.bits.ID != mp_ioapics[apic].mp_apicid) + printk("could not set ID!\n"); + else + apic_printk(APIC_VERBOSE, " ok.\n"); + } +} +#endif + +int no_timer_check __initdata; + +static int __init notimercheck(char *s) +{ + no_timer_check = 1; + return 1; +} +__setup("no_timer_check", notimercheck); + /* * There is a nasty bug in some older SMP boards, their mptable lies * about the timer IRQ. We do the following to work around the situation: @@ -1471,6 +2167,9 @@ static int __init timer_irq_works(void) unsigned long t1 = jiffies; unsigned long flags; + if (no_timer_check) + return 1; + local_save_flags(flags); local_irq_enable(); /* Let ten ticks pass... */ @@ -1518,22 +2217,26 @@ static unsigned int startup_ioapic_irq(unsigned int irq) { int was_pending = 0; unsigned long flags; + struct irq_cfg *cfg; spin_lock_irqsave(&ioapic_lock, flags); - if (irq < 16) { + if (irq < NR_IRQS_LEGACY) { disable_8259A_irq(irq); if (i8259A_irq_pending(irq)) was_pending = 1; } - __unmask_IO_APIC_irq(irq); + cfg = irq_cfg(irq); + __unmask_IO_APIC_irq(cfg); spin_unlock_irqrestore(&ioapic_lock, flags); return was_pending; } +#ifdef CONFIG_X86_64 static int ioapic_retrigger_irq(unsigned int irq) { - struct irq_cfg *cfg = &irq_cfg[irq]; + + struct irq_cfg *cfg = irq_cfg(irq); unsigned long flags; spin_lock_irqsave(&vector_lock, flags); @@ -1542,6 +2245,14 @@ static int ioapic_retrigger_irq(unsigned int irq) return 1; } +#else +static int ioapic_retrigger_irq(unsigned int irq) +{ + send_IPI_self(irq_cfg(irq)->vector); + + return 1; +} +#endif /* * Level and edge triggered IO-APIC interrupts need different handling, @@ -1578,32 +2289,37 @@ static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration); * as simple as edge triggered migration and we can do the irq migration * with a simple atomic update to IO-APIC RTE. */ -static void migrate_ioapic_irq(int irq, cpumask_t mask) +static void migrate_ioapic_irq_desc(struct irq_desc *desc, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; - struct irq_desc *desc = irq_desc + irq; + struct irq_cfg *cfg; cpumask_t tmp, cleanup_mask; struct irte irte; - int modify_ioapic_rte = desc->status & IRQ_LEVEL; + int modify_ioapic_rte; unsigned int dest; unsigned long flags; + unsigned int irq; cpus_and(tmp, mask, cpu_online_map); if (cpus_empty(tmp)) return; + irq = desc->irq; if (get_irte(irq, &irte)) return; - if (assign_irq_vector(irq, mask)) + cfg = desc->chip_data; + if (assign_irq_vector(irq, cfg, mask)) return; + set_extra_move_desc(desc, mask); + cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); + modify_ioapic_rte = desc->status & IRQ_LEVEL; if (modify_ioapic_rte) { spin_lock_irqsave(&ioapic_lock, flags); - __target_IO_APIC_irq(irq, dest, cfg->vector); + __target_IO_APIC_irq(irq, dest, cfg); spin_unlock_irqrestore(&ioapic_lock, flags); } @@ -1622,18 +2338,19 @@ static void migrate_ioapic_irq(int irq, cpumask_t mask) cfg->move_in_progress = 0; } - irq_desc[irq].affinity = mask; + desc->affinity = mask; } -static int migrate_irq_remapped_level(int irq) +static int migrate_irq_remapped_level_desc(struct irq_desc *desc) { int ret = -1; + struct irq_cfg *cfg = desc->chip_data; - mask_IO_APIC_irq(irq); + mask_IO_APIC_irq_desc(desc); - if (io_apic_level_ack_pending(irq)) { + if (io_apic_level_ack_pending(cfg)) { /* - * Interrupt in progress. Migrating irq now will change the + * Interrupt in progress. Migrating irq now will change the * vector information in the IO-APIC RTE and that will confuse * the EOI broadcast performed by cpu. * So, delay the irq migration to the next instance. @@ -1643,23 +2360,27 @@ static int migrate_irq_remapped_level(int irq) } /* everthing is clear. we have right of way */ - migrate_ioapic_irq(irq, irq_desc[irq].pending_mask); + migrate_ioapic_irq_desc(desc, desc->pending_mask); ret = 0; - irq_desc[irq].status &= ~IRQ_MOVE_PENDING; - cpus_clear(irq_desc[irq].pending_mask); + desc->status &= ~IRQ_MOVE_PENDING; + cpus_clear(desc->pending_mask); unmask: - unmask_IO_APIC_irq(irq); + unmask_IO_APIC_irq_desc(desc); + return ret; } static void ir_irq_migration(struct work_struct *work) { - int irq; + unsigned int irq; + struct irq_desc *desc; + + for_each_irq_desc(irq, desc) { + if (!desc) + continue; - for (irq = 0; irq < NR_IRQS; irq++) { - struct irq_desc *desc = irq_desc + irq; if (desc->status & IRQ_MOVE_PENDING) { unsigned long flags; @@ -1671,8 +2392,7 @@ static void ir_irq_migration(struct work_struct *work) continue; } - desc->chip->set_affinity(irq, - irq_desc[irq].pending_mask); + desc->chip->set_affinity(irq, desc->pending_mask); spin_unlock_irqrestore(&desc->lock, flags); } } @@ -1681,22 +2401,29 @@ static void ir_irq_migration(struct work_struct *work) /* * Migrates the IRQ destination in the process context. */ -static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) +static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, cpumask_t mask) { - if (irq_desc[irq].status & IRQ_LEVEL) { - irq_desc[irq].status |= IRQ_MOVE_PENDING; - irq_desc[irq].pending_mask = mask; - migrate_irq_remapped_level(irq); + if (desc->status & IRQ_LEVEL) { + desc->status |= IRQ_MOVE_PENDING; + desc->pending_mask = mask; + migrate_irq_remapped_level_desc(desc); return; } - migrate_ioapic_irq(irq, mask); + migrate_ioapic_irq_desc(desc, mask); +} +static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) +{ + struct irq_desc *desc = irq_to_desc(irq); + + set_ir_ioapic_affinity_irq_desc(desc, mask); } #endif asmlinkage void smp_irq_move_cleanup_interrupt(void) { unsigned vector, me; + ack_APIC_irq(); exit_idle(); irq_enter(); @@ -1707,11 +2434,15 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) struct irq_desc *desc; struct irq_cfg *cfg; irq = __get_cpu_var(vector_irq)[vector]; - if (irq >= NR_IRQS) + + if (irq == -1) continue; - desc = irq_desc + irq; - cfg = irq_cfg + irq; + desc = irq_to_desc(irq); + if (!desc) + continue; + + cfg = irq_cfg(irq); spin_lock(&desc->lock); if (!cfg->move_cleanup_count) goto unlock; @@ -1728,19 +2459,40 @@ unlock: irq_exit(); } -static void irq_complete_move(unsigned int irq) +static void irq_complete_move(struct irq_desc **descp) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_desc *desc = *descp; + struct irq_cfg *cfg = desc->chip_data; unsigned vector, me; - if (likely(!cfg->move_in_progress)) + if (likely(!cfg->move_in_progress)) { +#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC + if (likely(!cfg->move_desc_pending)) + return; + + /* domain has not changed, but affinity did */ + me = smp_processor_id(); + if (cpu_isset(me, desc->affinity)) { + *descp = desc = move_irq_desc(desc, me); + /* get the new one */ + cfg = desc->chip_data; + cfg->move_desc_pending = 0; + } +#endif return; + } vector = ~get_irq_regs()->orig_ax; me = smp_processor_id(); if ((vector == cfg->vector) && cpu_isset(me, cfg->domain)) { cpumask_t cleanup_mask; +#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC + *descp = desc = move_irq_desc(desc, me); + /* get the new one */ + cfg = desc->chip_data; +#endif + cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map); cfg->move_cleanup_count = cpus_weight(cleanup_mask); send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR); @@ -1748,8 +2500,9 @@ static void irq_complete_move(unsigned int irq) } } #else -static inline void irq_complete_move(unsigned int irq) {} +static inline void irq_complete_move(struct irq_desc **descp) {} #endif + #ifdef CONFIG_INTR_REMAP static void ack_x2apic_level(unsigned int irq) { @@ -1760,28 +2513,66 @@ static void ack_x2apic_edge(unsigned int irq) { ack_x2APIC_irq(); } + #endif static void ack_apic_edge(unsigned int irq) { - irq_complete_move(irq); + struct irq_desc *desc = irq_to_desc(irq); + + irq_complete_move(&desc); move_native_irq(irq); ack_APIC_irq(); } +atomic_t irq_mis_count; + static void ack_apic_level(unsigned int irq) { + struct irq_desc *desc = irq_to_desc(irq); + +#ifdef CONFIG_X86_32 + unsigned long v; + int i; +#endif + struct irq_cfg *cfg; int do_unmask_irq = 0; - irq_complete_move(irq); + irq_complete_move(&desc); #ifdef CONFIG_GENERIC_PENDING_IRQ /* If we are moving the irq we need to mask it */ - if (unlikely(irq_desc[irq].status & IRQ_MOVE_PENDING)) { + if (unlikely(desc->status & IRQ_MOVE_PENDING)) { do_unmask_irq = 1; - mask_IO_APIC_irq(irq); + mask_IO_APIC_irq_desc(desc); } #endif +#ifdef CONFIG_X86_32 + /* + * It appears there is an erratum which affects at least version 0x11 + * of I/O APIC (that's the 82093AA and cores integrated into various + * chipsets). Under certain conditions a level-triggered interrupt is + * erroneously delivered as edge-triggered one but the respective IRR + * bit gets set nevertheless. As a result the I/O unit expects an EOI + * message but it will never arrive and further interrupts are blocked + * from the source. The exact reason is so far unknown, but the + * phenomenon was observed when two consecutive interrupt requests + * from a given source get delivered to the same CPU and the source is + * temporarily disabled in between. + * + * A workaround is to simulate an EOI message manually. We achieve it + * by setting the trigger mode to edge and then to level when the edge + * trigger mode gets detected in the TMR of a local APIC for a + * level-triggered interrupt. We mask the source for the time of the + * operation to prevent an edge-triggered interrupt escaping meanwhile. + * The idea is from Manfred Spraul. --macro + */ + cfg = desc->chip_data; + i = cfg->vector; + + v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1)); +#endif + /* * We must acknowledge the irq before we move it or the acknowledge will * not propagate properly. @@ -1816,35 +2607,46 @@ static void ack_apic_level(unsigned int irq) * accurate and is causing problems then it is a hardware bug * and you can go talk to the chipset vendor about it. */ - if (!io_apic_level_ack_pending(irq)) + cfg = desc->chip_data; + if (!io_apic_level_ack_pending(cfg)) move_masked_irq(irq); - unmask_IO_APIC_irq(irq); + unmask_IO_APIC_irq_desc(desc); + } + +#ifdef CONFIG_X86_32 + if (!(v & (1 << (i & 0x1f)))) { + atomic_inc(&irq_mis_count); + spin_lock(&ioapic_lock); + __mask_and_edge_IO_APIC_irq(cfg); + __unmask_and_level_IO_APIC_irq(cfg); + spin_unlock(&ioapic_lock); } +#endif } static struct irq_chip ioapic_chip __read_mostly = { - .name = "IO-APIC", - .startup = startup_ioapic_irq, - .mask = mask_IO_APIC_irq, - .unmask = unmask_IO_APIC_irq, - .ack = ack_apic_edge, - .eoi = ack_apic_level, + .name = "IO-APIC", + .startup = startup_ioapic_irq, + .mask = mask_IO_APIC_irq, + .unmask = unmask_IO_APIC_irq, + .ack = ack_apic_edge, + .eoi = ack_apic_level, #ifdef CONFIG_SMP - .set_affinity = set_ioapic_affinity_irq, + .set_affinity = set_ioapic_affinity_irq, #endif .retrigger = ioapic_retrigger_irq, }; #ifdef CONFIG_INTR_REMAP static struct irq_chip ir_ioapic_chip __read_mostly = { - .name = "IR-IO-APIC", - .startup = startup_ioapic_irq, - .mask = mask_IO_APIC_irq, - .unmask = unmask_IO_APIC_irq, - .ack = ack_x2apic_edge, - .eoi = ack_x2apic_level, + .name = "IR-IO-APIC", + .startup = startup_ioapic_irq, + .mask = mask_IO_APIC_irq, + .unmask = unmask_IO_APIC_irq, + .ack = ack_x2apic_edge, + .eoi = ack_x2apic_level, #ifdef CONFIG_SMP - .set_affinity = set_ir_ioapic_affinity_irq, + .set_affinity = set_ir_ioapic_affinity_irq, #endif .retrigger = ioapic_retrigger_irq, }; @@ -1853,6 +2655,8 @@ static struct irq_chip ir_ioapic_chip __read_mostly = { static inline void init_IO_APIC_traps(void) { int irq; + struct irq_desc *desc; + struct irq_cfg *cfg; /* * NOTE! The local APIC isn't very good at handling @@ -1865,39 +2669,47 @@ static inline void init_IO_APIC_traps(void) * Also, we've got to be careful not to trash gate * 0x80, because int 0x80 is hm, kind of importantish. ;) */ - for (irq = 0; irq < NR_IRQS ; irq++) { - if (IO_APIC_IRQ(irq) && !irq_cfg[irq].vector) { + for_each_irq_desc(irq, desc) { + if (!desc) + continue; + + cfg = desc->chip_data; + if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) { /* * Hmm.. We don't have an entry for this, * so default to an old-fashioned 8259 * interrupt if we can.. */ - if (irq < 16) + if (irq < NR_IRQS_LEGACY) make_8259A_irq(irq); else /* Strange. Oh, well.. */ - irq_desc[irq].chip = &no_irq_chip; + desc->chip = &no_irq_chip; } } } -static void unmask_lapic_irq(unsigned int irq) +/* + * The local APIC irq-chip implementation: + */ + +static void mask_lapic_irq(unsigned int irq) { unsigned long v; v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); + apic_write(APIC_LVT0, v | APIC_LVT_MASKED); } -static void mask_lapic_irq(unsigned int irq) +static void unmask_lapic_irq(unsigned int irq) { unsigned long v; v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v | APIC_LVT_MASKED); + apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); } -static void ack_lapic_irq (unsigned int irq) +static void ack_lapic_irq(unsigned int irq) { ack_APIC_irq(); } @@ -1909,9 +2721,9 @@ static struct irq_chip lapic_chip __read_mostly = { .ack = ack_lapic_irq, }; -static void lapic_register_intr(int irq) +static void lapic_register_intr(int irq, struct irq_desc *desc) { - irq_desc[irq].status &= ~IRQ_LEVEL; + desc->status &= ~IRQ_LEVEL; set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq, "edge"); } @@ -1919,19 +2731,19 @@ static void lapic_register_intr(int irq) static void __init setup_nmi(void) { /* - * Dirty trick to enable the NMI watchdog ... + * Dirty trick to enable the NMI watchdog ... * We put the 8259A master into AEOI mode and * unmask on all local APICs LVT0 as NMI. * * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire') * is from Maciej W. Rozycki - so we do not have to EOI from * the NMI handler or the timer interrupt. - */ - printk(KERN_INFO "activating NMI Watchdog ..."); + */ + apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ..."); enable_NMI_through_LVT0(); - printk(" done.\n"); + apic_printk(APIC_VERBOSE, " done.\n"); } /* @@ -1948,12 +2760,17 @@ static inline void __init unlock_ExtINT_logic(void) unsigned char save_control, save_freq_select; pin = find_isa_irq_pin(8, mp_INT); + if (pin == -1) { + WARN_ON_ONCE(1); + return; + } apic = find_isa_irq_apic(8, mp_INT); - if (pin == -1) + if (apic == -1) { + WARN_ON_ONCE(1); return; + } entry0 = ioapic_read_entry(apic, pin); - clear_IO_APIC_pin(apic, pin); memset(&entry1, 0, sizeof(entry1)); @@ -1988,35 +2805,60 @@ static inline void __init unlock_ExtINT_logic(void) ioapic_write_entry(apic, pin, entry0); } +static int disable_timer_pin_1 __initdata; +/* Actually the next is obsolete, but keep it for paranoid reasons -AK */ +static int __init disable_timer_pin_setup(char *arg) +{ + disable_timer_pin_1 = 1; + return 0; +} +early_param("disable_timer_pin_1", disable_timer_pin_setup); + +int timer_through_8259 __initdata; + /* * This code may look a bit paranoid, but it's supposed to cooperate with * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ * is so screwy. Thanks to Brian Perkins for testing/hacking this beast * fanatically on his truly buggy board. * - * FIXME: really need to revamp this for modern platforms only. + * FIXME: really need to revamp this for all platforms. */ static inline void __init check_timer(void) { - struct irq_cfg *cfg = irq_cfg + 0; + struct irq_desc *desc = irq_to_desc(0); + struct irq_cfg *cfg = desc->chip_data; + int cpu = boot_cpu_id; int apic1, pin1, apic2, pin2; unsigned long flags; + unsigned int ver; int no_pin1 = 0; local_irq_save(flags); + ver = apic_read(APIC_LVR); + ver = GET_APIC_VERSION(ver); + /* * get/set the timer IRQ vector: */ disable_8259A_irq(0); - assign_irq_vector(0, TARGET_CPUS); + assign_irq_vector(0, cfg, TARGET_CPUS); /* * As IRQ0 is to be enabled in the 8259A, the virtual - * wire has to be disabled in the local APIC. + * wire has to be disabled in the local APIC. Also + * timer interrupts need to be acknowledged manually in + * the 8259A for the i82489DX when using the NMI + * watchdog as that APIC treats NMIs as level-triggered. + * The AEOI mode will finish them in the 8259A + * automatically. */ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); init_8259A(1); +#ifdef CONFIG_X86_32 + timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); +#endif pin1 = find_isa_irq_pin(0, mp_INT); apic1 = find_isa_irq_apic(0, mp_INT); @@ -2035,8 +2877,10 @@ static inline void __init check_timer(void) * 8259A. */ if (pin1 == -1) { +#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) panic("BIOS bug: timer not connected to IO-APIC"); +#endif pin1 = pin2; apic1 = apic2; no_pin1 = 1; @@ -2050,11 +2894,11 @@ static inline void __init check_timer(void) * Ok, does IRQ0 through the IOAPIC work? */ if (no_pin1) { - add_pin_to_irq(0, apic1, pin1); + add_pin_to_irq_cpu(cfg, cpu, apic1, pin1); setup_timer_IRQ0_pin(apic1, pin1, cfg->vector); } - unmask_IO_APIC_irq(0); - if (!no_timer_check && timer_irq_works()) { + unmask_IO_APIC_irq_desc(desc); + if (timer_irq_works()) { if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); enable_8259A_irq(0); @@ -2063,8 +2907,10 @@ static inline void __init check_timer(void) clear_IO_APIC_pin(0, pin1); goto out; } +#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) panic("timer doesn't work through Interrupt-remapped IO-APIC"); +#endif clear_IO_APIC_pin(apic1, pin1); if (!no_pin1) apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: " @@ -2077,9 +2923,9 @@ static inline void __init check_timer(void) /* * legacy devices should be connected to IO APIC #0 */ - replace_pin_at_irq(0, apic1, pin1, apic2, pin2); + replace_pin_at_irq_cpu(cfg, cpu, apic1, pin1, apic2, pin2); setup_timer_IRQ0_pin(apic2, pin2, cfg->vector); - unmask_IO_APIC_irq(0); + unmask_IO_APIC_irq_desc(desc); enable_8259A_irq(0); if (timer_irq_works()) { apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); @@ -2104,11 +2950,14 @@ static inline void __init check_timer(void) "through the IO-APIC - disabling NMI Watchdog!\n"); nmi_watchdog = NMI_NONE; } +#ifdef CONFIG_X86_32 + timer_ack = 0; +#endif apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer as Virtual Wire IRQ...\n"); - lapic_register_intr(0); + lapic_register_intr(0, desc); apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector); /* Fixed mode */ enable_8259A_irq(0); @@ -2140,13 +2989,6 @@ out: local_irq_restore(flags); } -static int __init notimercheck(char *s) -{ - no_timer_check = 1; - return 1; -} -__setup("no_timer_check", notimercheck); - /* * Traditionally ISA IRQ2 is the cascade IRQ, and is not available * to devices. However there may be an I/O APIC pin available for @@ -2164,25 +3006,49 @@ __setup("no_timer_check", notimercheck); * the I/O APIC in all cases now. No actual device should request * it anyway. --macro */ -#define PIC_IRQS (1<<2) +#define PIC_IRQS (1 << PIC_CASCADE_IR) void __init setup_IO_APIC(void) { +#ifdef CONFIG_X86_32 + enable_IO_APIC(); +#else /* * calling enable_IO_APIC() is moved to setup_local_APIC for BP */ +#endif io_apic_irqs = ~PIC_IRQS; apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n"); - + /* + * Set up IO-APIC IRQ routing. + */ +#ifdef CONFIG_X86_32 + if (!acpi_ioapic) + setup_ioapic_ids_from_mpc(); +#endif sync_Arb_IDs(); setup_IO_APIC_irqs(); init_IO_APIC_traps(); check_timer(); } +/* + * Called after all the initialization is done. If we didnt find any + * APIC bugs then we can allow the modify fast path + */ + +static int __init io_apic_bug_finalize(void) +{ + if (sis_apic_bug == -1) + sis_apic_bug = 0; + return 0; +} + +late_initcall(io_apic_bug_finalize); + struct sysfs_ioapic_data { struct sys_device dev; struct IO_APIC_route_entry entry[0]; @@ -2270,43 +3136,80 @@ device_initcall(ioapic_init_sysfs); /* * Dynamic irq allocate and deallocation */ -int create_irq(void) +unsigned int create_irq_nr(unsigned int irq_want) { /* Allocate an unused irq */ - int irq; - int new; + unsigned int irq; + unsigned int new; unsigned long flags; + struct irq_cfg *cfg_new = NULL; + int cpu = boot_cpu_id; + struct irq_desc *desc_new = NULL; - irq = -ENOSPC; + irq = 0; spin_lock_irqsave(&vector_lock, flags); - for (new = (NR_IRQS - 1); new >= 0; new--) { + for (new = irq_want; new < NR_IRQS; new++) { if (platform_legacy_irq(new)) continue; - if (irq_cfg[new].vector != 0) + + desc_new = irq_to_desc_alloc_cpu(new, cpu); + if (!desc_new) { + printk(KERN_INFO "can not get irq_desc for %d\n", new); + continue; + } + cfg_new = desc_new->chip_data; + + if (cfg_new->vector != 0) continue; - if (__assign_irq_vector(new, TARGET_CPUS) == 0) + if (__assign_irq_vector(new, cfg_new, TARGET_CPUS) == 0) irq = new; break; } spin_unlock_irqrestore(&vector_lock, flags); - if (irq >= 0) { + if (irq > 0) { dynamic_irq_init(irq); + /* restore it, in case dynamic_irq_init clear it */ + if (desc_new) + desc_new->chip_data = cfg_new; } return irq; } +static int nr_irqs_gsi = NR_IRQS_LEGACY; +int create_irq(void) +{ + unsigned int irq_want; + int irq; + + irq_want = nr_irqs_gsi; + irq = create_irq_nr(irq_want); + + if (irq == 0) + irq = -1; + + return irq; +} + void destroy_irq(unsigned int irq) { unsigned long flags; + struct irq_cfg *cfg; + struct irq_desc *desc; + /* store it, in case dynamic_irq_cleanup clear it */ + desc = irq_to_desc(irq); + cfg = desc->chip_data; dynamic_irq_cleanup(irq); + /* connect back irq_cfg */ + if (desc) + desc->chip_data = cfg; #ifdef CONFIG_INTR_REMAP free_irte(irq); #endif spin_lock_irqsave(&vector_lock, flags); - __clear_irq_vector(irq); + __clear_irq_vector(irq, cfg); spin_unlock_irqrestore(&vector_lock, flags); } @@ -2316,13 +3219,14 @@ void destroy_irq(unsigned int irq) #ifdef CONFIG_PCI_MSI static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; int err; unsigned dest; cpumask_t tmp; + cfg = irq_cfg(irq); tmp = TARGET_CPUS; - err = assign_irq_vector(irq, tmp); + err = assign_irq_vector(irq, cfg, tmp); if (err) return err; @@ -2383,7 +3287,8 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms #ifdef CONFIG_SMP static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; struct msi_msg msg; unsigned int dest; cpumask_t tmp; @@ -2392,23 +3297,25 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) if (cpus_empty(tmp)) return; - if (assign_irq_vector(irq, mask)) + cfg = desc->chip_data; + if (assign_irq_vector(irq, cfg, mask)) return; + set_extra_move_desc(desc, mask); + cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); - read_msi_msg(irq, &msg); + read_msi_msg_desc(desc, &msg); msg.data &= ~MSI_DATA_VECTOR_MASK; msg.data |= MSI_DATA_VECTOR(cfg->vector); msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; msg.address_lo |= MSI_ADDR_DEST_ID(dest); - write_msi_msg(irq, &msg); - irq_desc[irq].affinity = mask; + write_msi_msg_desc(desc, &msg); + desc->affinity = mask; } - #ifdef CONFIG_INTR_REMAP /* * Migrate the MSI irq to another cpumask. This migration is @@ -2416,7 +3323,8 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) */ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; unsigned int dest; cpumask_t tmp, cleanup_mask; struct irte irte; @@ -2428,9 +3336,12 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) if (get_irte(irq, &irte)) return; - if (assign_irq_vector(irq, mask)) + cfg = desc->chip_data; + if (assign_irq_vector(irq, cfg, mask)) return; + set_extra_move_desc(desc, mask); + cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); @@ -2454,8 +3365,9 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) cfg->move_in_progress = 0; } - irq_desc[irq].affinity = mask; + desc->affinity = mask; } + #endif #endif /* CONFIG_SMP */ @@ -2507,14 +3419,14 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec) if (index < 0) { printk(KERN_ERR "Unable to allocate %d IRTE for PCI %s\n", nvec, - pci_name(dev)); + pci_name(dev)); return -ENOSPC; } return index; } #endif -static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq) +static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) { int ret; struct msi_msg msg; @@ -2523,12 +3435,12 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq) if (ret < 0) return ret; - set_irq_msi(irq, desc); + set_irq_msi(irq, msidesc); write_msi_msg(irq, &msg); #ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { - struct irq_desc *desc = irq_desc + irq; + struct irq_desc *desc = irq_to_desc(irq); /* * irq migration in process context */ @@ -2538,16 +3450,21 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq) #endif set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge"); + dev_printk(KERN_DEBUG, &dev->dev, "irq %d for MSI/MSI-X\n", irq); + return 0; } -int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc) +int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc) { - int irq, ret; + unsigned int irq; + int ret; + unsigned int irq_want; - irq = create_irq(); - if (irq < 0) - return irq; + irq_want = nr_irqs_gsi; + irq = create_irq_nr(irq_want); + if (irq == 0) + return -1; #ifdef CONFIG_INTR_REMAP if (!intr_remapping_enabled) @@ -2558,7 +3475,7 @@ int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc) goto error; no_ir: #endif - ret = setup_msi_irq(dev, desc, irq); + ret = setup_msi_irq(dev, msidesc, irq); if (ret < 0) { destroy_irq(irq); return ret; @@ -2574,18 +3491,23 @@ error: int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) { - int irq, ret, sub_handle; - struct msi_desc *desc; + unsigned int irq; + int ret, sub_handle; + struct msi_desc *msidesc; + unsigned int irq_want; + #ifdef CONFIG_INTR_REMAP struct intel_iommu *iommu = 0; int index = 0; #endif + irq_want = nr_irqs_gsi; sub_handle = 0; - list_for_each_entry(desc, &dev->msi_list, list) { - irq = create_irq(); - if (irq < 0) - return irq; + list_for_each_entry(msidesc, &dev->msi_list, list) { + irq = create_irq_nr(irq_want); + irq_want++; + if (irq == 0) + return -1; #ifdef CONFIG_INTR_REMAP if (!intr_remapping_enabled) goto no_ir; @@ -2615,7 +3537,7 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) } no_ir: #endif - ret = setup_msi_irq(dev, desc, irq); + ret = setup_msi_irq(dev, msidesc, irq); if (ret < 0) goto error; sub_handle++; @@ -2636,7 +3558,8 @@ void arch_teardown_msi_irq(unsigned int irq) #ifdef CONFIG_SMP static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; struct msi_msg msg; unsigned int dest; cpumask_t tmp; @@ -2645,9 +3568,12 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) if (cpus_empty(tmp)) return; - if (assign_irq_vector(irq, mask)) + cfg = desc->chip_data; + if (assign_irq_vector(irq, cfg, mask)) return; + set_extra_move_desc(desc, mask); + cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); @@ -2659,8 +3585,9 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) msg.address_lo |= MSI_ADDR_DEST_ID(dest); dmar_msi_write(irq, &msg); - irq_desc[irq].affinity = mask; + desc->affinity = mask; } + #endif /* CONFIG_SMP */ struct irq_chip dmar_msi_type = { @@ -2689,6 +3616,71 @@ int arch_setup_dmar_msi(unsigned int irq) } #endif +#ifdef CONFIG_HPET_TIMER + +#ifdef CONFIG_SMP +static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask) +{ + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; + struct msi_msg msg; + unsigned int dest; + cpumask_t tmp; + + cpus_and(tmp, mask, cpu_online_map); + if (cpus_empty(tmp)) + return; + + cfg = desc->chip_data; + if (assign_irq_vector(irq, cfg, mask)) + return; + + set_extra_move_desc(desc, mask); + + cpus_and(tmp, cfg->domain, mask); + dest = cpu_mask_to_apicid(tmp); + + hpet_msi_read(irq, &msg); + + msg.data &= ~MSI_DATA_VECTOR_MASK; + msg.data |= MSI_DATA_VECTOR(cfg->vector); + msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; + msg.address_lo |= MSI_ADDR_DEST_ID(dest); + + hpet_msi_write(irq, &msg); + desc->affinity = mask; +} + +#endif /* CONFIG_SMP */ + +struct irq_chip hpet_msi_type = { + .name = "HPET_MSI", + .unmask = hpet_msi_unmask, + .mask = hpet_msi_mask, + .ack = ack_apic_edge, +#ifdef CONFIG_SMP + .set_affinity = hpet_msi_set_affinity, +#endif + .retrigger = ioapic_retrigger_irq, +}; + +int arch_setup_hpet_msi(unsigned int irq) +{ + int ret; + struct msi_msg msg; + + ret = msi_compose_msg(NULL, irq, &msg); + if (ret < 0) + return ret; + + hpet_msi_write(irq, &msg); + set_irq_chip_and_handler_name(irq, &hpet_msi_type, handle_edge_irq, + "edge"); + + return 0; +} +#endif + #endif /* CONFIG_PCI_MSI */ /* * Hypertransport interrupt support @@ -2713,7 +3705,8 @@ static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; unsigned int dest; cpumask_t tmp; @@ -2721,15 +3714,19 @@ static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) if (cpus_empty(tmp)) return; - if (assign_irq_vector(irq, mask)) + cfg = desc->chip_data; + if (assign_irq_vector(irq, cfg, mask)) return; + set_extra_move_desc(desc, mask); + cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); target_ht_irq(irq, dest, cfg->vector); - irq_desc[irq].affinity = mask; + desc->affinity = mask; } + #endif static struct irq_chip ht_irq_chip = { @@ -2745,12 +3742,13 @@ static struct irq_chip ht_irq_chip = { int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; int err; cpumask_t tmp; + cfg = irq_cfg(irq); tmp = TARGET_CPUS; - err = assign_irq_vector(irq, tmp); + err = assign_irq_vector(irq, cfg, tmp); if (!err) { struct ht_irq_msg msg; unsigned dest; @@ -2777,20 +3775,185 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) set_irq_chip_and_handler_name(irq, &ht_irq_chip, handle_edge_irq, "edge"); + + dev_printk(KERN_DEBUG, &dev->dev, "irq %d for HT\n", irq); } return err; } #endif /* CONFIG_HT_IRQ */ +#ifdef CONFIG_X86_64 +/* + * Re-target the irq to the specified CPU and enable the specified MMR located + * on the specified blade to allow the sending of MSIs to the specified CPU. + */ +int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, + unsigned long mmr_offset) +{ + const cpumask_t *eligible_cpu = get_cpu_mask(cpu); + struct irq_cfg *cfg; + int mmr_pnode; + unsigned long mmr_value; + struct uv_IO_APIC_route_entry *entry; + unsigned long flags; + int err; + + cfg = irq_cfg(irq); + + err = assign_irq_vector(irq, cfg, *eligible_cpu); + if (err != 0) + return err; + + spin_lock_irqsave(&vector_lock, flags); + set_irq_chip_and_handler_name(irq, &uv_irq_chip, handle_percpu_irq, + irq_name); + spin_unlock_irqrestore(&vector_lock, flags); + + mmr_value = 0; + entry = (struct uv_IO_APIC_route_entry *)&mmr_value; + BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); + + entry->vector = cfg->vector; + entry->delivery_mode = INT_DELIVERY_MODE; + entry->dest_mode = INT_DEST_MODE; + entry->polarity = 0; + entry->trigger = 0; + entry->mask = 0; + entry->dest = cpu_mask_to_apicid(*eligible_cpu); + + mmr_pnode = uv_blade_to_pnode(mmr_blade); + uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); + + return irq; +} + +/* + * Disable the specified MMR located on the specified blade so that MSIs are + * longer allowed to be sent. + */ +void arch_disable_uv_irq(int mmr_blade, unsigned long mmr_offset) +{ + unsigned long mmr_value; + struct uv_IO_APIC_route_entry *entry; + int mmr_pnode; + + mmr_value = 0; + entry = (struct uv_IO_APIC_route_entry *)&mmr_value; + BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); + + entry->mask = 1; + + mmr_pnode = uv_blade_to_pnode(mmr_blade); + uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); +} +#endif /* CONFIG_X86_64 */ + +int __init io_apic_get_redir_entries (int ioapic) +{ + union IO_APIC_reg_01 reg_01; + unsigned long flags; + + spin_lock_irqsave(&ioapic_lock, flags); + reg_01.raw = io_apic_read(ioapic, 1); + spin_unlock_irqrestore(&ioapic_lock, flags); + + return reg_01.bits.entries; +} + +void __init probe_nr_irqs_gsi(void) +{ + int idx; + int nr = 0; + + for (idx = 0; idx < nr_ioapics; idx++) + nr += io_apic_get_redir_entries(idx) + 1; + + if (nr > nr_irqs_gsi) + nr_irqs_gsi = nr; +} + /* -------------------------------------------------------------------------- ACPI-based IOAPIC Configuration -------------------------------------------------------------------------- */ #ifdef CONFIG_ACPI -#define IO_APIC_MAX_ID 0xFE +#ifdef CONFIG_X86_32 +int __init io_apic_get_unique_id(int ioapic, int apic_id) +{ + union IO_APIC_reg_00 reg_00; + static physid_mask_t apic_id_map = PHYSID_MASK_NONE; + physid_mask_t tmp; + unsigned long flags; + int i = 0; -int __init io_apic_get_redir_entries (int ioapic) + /* + * The P4 platform supports up to 256 APIC IDs on two separate APIC + * buses (one for LAPICs, one for IOAPICs), where predecessors only + * supports up to 16 on one shared APIC bus. + * + * TBD: Expand LAPIC/IOAPIC support on P4-class systems to take full + * advantage of new APIC bus architecture. + */ + + if (physids_empty(apic_id_map)) + apic_id_map = ioapic_phys_id_map(phys_cpu_present_map); + + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(ioapic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + + if (apic_id >= get_physical_broadcast()) { + printk(KERN_WARNING "IOAPIC[%d]: Invalid apic_id %d, trying " + "%d\n", ioapic, apic_id, reg_00.bits.ID); + apic_id = reg_00.bits.ID; + } + + /* + * Every APIC in a system must have a unique ID or we get lots of nice + * 'stuck on smp_invalidate_needed IPI wait' messages. + */ + if (check_apicid_used(apic_id_map, apic_id)) { + + for (i = 0; i < get_physical_broadcast(); i++) { + if (!check_apicid_used(apic_id_map, i)) + break; + } + + if (i == get_physical_broadcast()) + panic("Max apic_id exceeded!\n"); + + printk(KERN_WARNING "IOAPIC[%d]: apic_id %d already used, " + "trying %d\n", ioapic, apic_id, i); + + apic_id = i; + } + + tmp = apicid_to_cpu_present(apic_id); + physids_or(apic_id_map, apic_id_map, tmp); + + if (reg_00.bits.ID != apic_id) { + reg_00.bits.ID = apic_id; + + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(ioapic, 0, reg_00.raw); + reg_00.raw = io_apic_read(ioapic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + + /* Sanity check */ + if (reg_00.bits.ID != apic_id) { + printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic); + return -1; + } + } + + apic_printk(APIC_VERBOSE, KERN_INFO + "IOAPIC[%d]: Assigned apic_id %d\n", ioapic, apic_id); + + return apic_id; +} + +int __init io_apic_get_version(int ioapic) { union IO_APIC_reg_01 reg_01; unsigned long flags; @@ -2799,25 +3962,37 @@ int __init io_apic_get_redir_entries (int ioapic) reg_01.raw = io_apic_read(ioapic, 1); spin_unlock_irqrestore(&ioapic_lock, flags); - return reg_01.bits.entries; + return reg_01.bits.version; } - +#endif int io_apic_set_pci_routing (int ioapic, int pin, int irq, int triggering, int polarity) { + struct irq_desc *desc; + struct irq_cfg *cfg; + int cpu = boot_cpu_id; + if (!IO_APIC_IRQ(irq)) { apic_printk(APIC_QUIET,KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n", ioapic); return -EINVAL; } + desc = irq_to_desc_alloc_cpu(irq, cpu); + if (!desc) { + printk(KERN_INFO "can not get irq_desc %d\n", irq); + return 0; + } + /* * IRQs < 16 are already in the irq_2_pin[] map */ - if (irq >= 16) - add_pin_to_irq(irq, ioapic, pin); + if (irq >= NR_IRQS_LEGACY) { + cfg = desc->chip_data; + add_pin_to_irq_cpu(cfg, cpu, ioapic, pin); + } - setup_IO_APIC_irq(ioapic, pin, irq, triggering, polarity); + setup_IO_APIC_irq(ioapic, pin, irq, desc, triggering, polarity); return 0; } @@ -2853,6 +4028,9 @@ int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) void __init setup_ioapic_dest(void) { int pin, ioapic, irq, irq_entry; + struct irq_desc *desc; + struct irq_cfg *cfg; + cpumask_t mask; if (skip_ioapic_setup == 1) return; @@ -2868,16 +4046,31 @@ void __init setup_ioapic_dest(void) * when you have too many devices, because at that time only boot * cpu is online. */ - if (!irq_cfg[irq].vector) - setup_IO_APIC_irq(ioapic, pin, irq, + desc = irq_to_desc(irq); + cfg = desc->chip_data; + if (!cfg->vector) { + setup_IO_APIC_irq(ioapic, pin, irq, desc, irq_trigger(irq_entry), irq_polarity(irq_entry)); + continue; + + } + + /* + * Honour affinities which have been set in early boot + */ + if (desc->status & + (IRQ_NO_BALANCING | IRQ_AFFINITY_SET)) + mask = desc->affinity; + else + mask = TARGET_CPUS; + #ifdef CONFIG_INTR_REMAP - else if (intr_remapping_enabled) - set_ir_ioapic_affinity_irq(irq, TARGET_CPUS); -#endif + if (intr_remapping_enabled) + set_ir_ioapic_affinity_irq_desc(desc, mask); else - set_ioapic_affinity_irq(irq, TARGET_CPUS); +#endif + set_ioapic_affinity_irq_desc(desc, mask); } } @@ -2930,14 +4123,28 @@ void __init ioapic_init_mappings(void) for (i = 0; i < nr_ioapics; i++) { if (smp_found_config) { ioapic_phys = mp_ioapics[i].mp_apicaddr; +#ifdef CONFIG_X86_32 + if (!ioapic_phys) { + printk(KERN_ERR + "WARNING: bogus zero IO-APIC " + "address found in MPTABLE, " + "disabling IO/APIC support!\n"); + smp_found_config = 0; + skip_ioapic_setup = 1; + goto fake_ioapic_page; + } +#endif } else { +#ifdef CONFIG_X86_32 +fake_ioapic_page: +#endif ioapic_phys = (unsigned long) alloc_bootmem_pages(PAGE_SIZE); ioapic_phys = __pa(ioapic_phys); } set_fixmap_nocache(idx, ioapic_phys); apic_printk(APIC_VERBOSE, - "mapped IOAPIC to %016lx (%016lx)\n", + "mapped IOAPIC to %08lx (%08lx)\n", __fix_to_virt(idx), ioapic_phys); idx++; @@ -2971,4 +4178,3 @@ static int __init ioapic_insert_resources(void) /* Insert the IO APIC resources after PCI initialization has occured to handle * IO APICS that are mapped in on a BAR in PCI space. */ late_initcall(ioapic_insert_resources); - diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c deleted file mode 100644 index e710289f673e..000000000000 --- a/arch/x86/kernel/io_apic_32.c +++ /dev/null @@ -1,2908 +0,0 @@ -/* - * Intel IO-APIC support for multi-Pentium hosts. - * - * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar, Hajnalka Szabo - * - * Many thanks to Stig Venaas for trying out countless experimental - * patches and reporting/debugging problems patiently! - * - * (c) 1999, Multiple IO-APIC support, developed by - * Ken-ichi Yaku <yaku@css1.kbnes.nec.co.jp> and - * Hidemi Kishimoto <kisimoto@css1.kbnes.nec.co.jp>, - * further tested and cleaned up by Zach Brown <zab@redhat.com> - * and Ingo Molnar <mingo@redhat.com> - * - * Fixes - * Maciej W. Rozycki : Bits for genuine 82489DX APICs; - * thanks to Eric Gilmore - * and Rolf G. Tews - * for testing these extensively - * Paul Diefenbaugh : Added full ACPI support - */ - -#include <linux/mm.h> -#include <linux/interrupt.h> -#include <linux/init.h> -#include <linux/delay.h> -#include <linux/sched.h> -#include <linux/bootmem.h> -#include <linux/mc146818rtc.h> -#include <linux/compiler.h> -#include <linux/acpi.h> -#include <linux/module.h> -#include <linux/sysdev.h> -#include <linux/pci.h> -#include <linux/msi.h> -#include <linux/htirq.h> -#include <linux/freezer.h> -#include <linux/kthread.h> -#include <linux/jiffies.h> /* time_after() */ - -#include <asm/io.h> -#include <asm/smp.h> -#include <asm/desc.h> -#include <asm/timer.h> -#include <asm/i8259.h> -#include <asm/nmi.h> -#include <asm/msidef.h> -#include <asm/hypertransport.h> -#include <asm/setup.h> - -#include <mach_apic.h> -#include <mach_apicdef.h> - -#define __apicdebuginit(type) static type __init - -int (*ioapic_renumber_irq)(int ioapic, int irq); -atomic_t irq_mis_count; - -/* Where if anywhere is the i8259 connect in external int mode */ -static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; - -static DEFINE_SPINLOCK(ioapic_lock); -DEFINE_SPINLOCK(vector_lock); - -int timer_through_8259 __initdata; - -/* - * Is the SiS APIC rmw bug present ? - * -1 = don't know, 0 = no, 1 = yes - */ -int sis_apic_bug = -1; - -/* - * # of IRQ routing registers - */ -int nr_ioapic_registers[MAX_IO_APICS]; - -/* I/O APIC entries */ -struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; -int nr_ioapics; - -/* MP IRQ source entries */ -struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; - -/* # of MP IRQ source entries */ -int mp_irq_entries; - -#if defined (CONFIG_MCA) || defined (CONFIG_EISA) -int mp_bus_id_to_type[MAX_MP_BUSSES]; -#endif - -DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); - -static int disable_timer_pin_1 __initdata; - -/* - * Rough estimation of how many shared IRQs there are, can - * be changed anytime. - */ -#define MAX_PLUS_SHARED_IRQS NR_IRQS -#define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS) - -/* - * This is performance-critical, we want to do it O(1) - * - * the indexing order of this array favors 1:1 mappings - * between pins and IRQs. - */ - -static struct irq_pin_list { - int apic, pin, next; -} irq_2_pin[PIN_MAP_SIZE]; - -struct io_apic { - unsigned int index; - unsigned int unused[3]; - unsigned int data; -}; - -static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) -{ - return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) - + (mp_ioapics[idx].mp_apicaddr & ~PAGE_MASK); -} - -static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) -{ - struct io_apic __iomem *io_apic = io_apic_base(apic); - writel(reg, &io_apic->index); - return readl(&io_apic->data); -} - -static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) -{ - struct io_apic __iomem *io_apic = io_apic_base(apic); - writel(reg, &io_apic->index); - writel(value, &io_apic->data); -} - -/* - * Re-write a value: to be used for read-modify-write - * cycles where the read already set up the index register. - * - * Older SiS APIC requires we rewrite the index register - */ -static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value) -{ - volatile struct io_apic __iomem *io_apic = io_apic_base(apic); - if (sis_apic_bug) - writel(reg, &io_apic->index); - writel(value, &io_apic->data); -} - -union entry_union { - struct { u32 w1, w2; }; - struct IO_APIC_route_entry entry; -}; - -static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) -{ - union entry_union eu; - unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); - eu.w1 = io_apic_read(apic, 0x10 + 2 * pin); - eu.w2 = io_apic_read(apic, 0x11 + 2 * pin); - spin_unlock_irqrestore(&ioapic_lock, flags); - return eu.entry; -} - -/* - * When we write a new IO APIC routing entry, we need to write the high - * word first! If the mask bit in the low word is clear, we will enable - * the interrupt, and we need to make sure the entry is fully populated - * before that happens. - */ -static void -__ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) -{ - union entry_union eu; - eu.entry = e; - io_apic_write(apic, 0x11 + 2*pin, eu.w2); - io_apic_write(apic, 0x10 + 2*pin, eu.w1); -} - -static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) -{ - unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); - __ioapic_write_entry(apic, pin, e); - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -/* - * When we mask an IO APIC routing entry, we need to write the low - * word first, in order to set the mask bit before we change the - * high bits! - */ -static void ioapic_mask_entry(int apic, int pin) -{ - unsigned long flags; - union entry_union eu = { .entry.mask = 1 }; - - spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(apic, 0x10 + 2*pin, eu.w1); - io_apic_write(apic, 0x11 + 2*pin, eu.w2); - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -/* - * The common case is 1:1 IRQ<->pin mappings. Sometimes there are - * shared ISA-space IRQs, so we have to support them. We are super - * fast in the common case, and fast for shared ISA-space IRQs. - */ -static void add_pin_to_irq(unsigned int irq, int apic, int pin) -{ - static int first_free_entry = NR_IRQS; - struct irq_pin_list *entry = irq_2_pin + irq; - - while (entry->next) - entry = irq_2_pin + entry->next; - - if (entry->pin != -1) { - entry->next = first_free_entry; - entry = irq_2_pin + entry->next; - if (++first_free_entry >= PIN_MAP_SIZE) - panic("io_apic.c: whoops"); - } - entry->apic = apic; - entry->pin = pin; -} - -/* - * Reroute an IRQ to a different pin. - */ -static void __init replace_pin_at_irq(unsigned int irq, - int oldapic, int oldpin, - int newapic, int newpin) -{ - struct irq_pin_list *entry = irq_2_pin + irq; - - while (1) { - if (entry->apic == oldapic && entry->pin == oldpin) { - entry->apic = newapic; - entry->pin = newpin; - } - if (!entry->next) - break; - entry = irq_2_pin + entry->next; - } -} - -static void __modify_IO_APIC_irq(unsigned int irq, unsigned long enable, unsigned long disable) -{ - struct irq_pin_list *entry = irq_2_pin + irq; - unsigned int pin, reg; - - for (;;) { - pin = entry->pin; - if (pin == -1) - break; - reg = io_apic_read(entry->apic, 0x10 + pin*2); - reg &= ~disable; - reg |= enable; - io_apic_modify(entry->apic, 0x10 + pin*2, reg); - if (!entry->next) - break; - entry = irq_2_pin + entry->next; - } -} - -/* mask = 1 */ -static void __mask_IO_APIC_irq(unsigned int irq) -{ - __modify_IO_APIC_irq(irq, IO_APIC_REDIR_MASKED, 0); -} - -/* mask = 0 */ -static void __unmask_IO_APIC_irq(unsigned int irq) -{ - __modify_IO_APIC_irq(irq, 0, IO_APIC_REDIR_MASKED); -} - -/* mask = 1, trigger = 0 */ -static void __mask_and_edge_IO_APIC_irq(unsigned int irq) -{ - __modify_IO_APIC_irq(irq, IO_APIC_REDIR_MASKED, - IO_APIC_REDIR_LEVEL_TRIGGER); -} - -/* mask = 0, trigger = 1 */ -static void __unmask_and_level_IO_APIC_irq(unsigned int irq) -{ - __modify_IO_APIC_irq(irq, IO_APIC_REDIR_LEVEL_TRIGGER, - IO_APIC_REDIR_MASKED); -} - -static void mask_IO_APIC_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - __mask_IO_APIC_irq(irq); - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -static void unmask_IO_APIC_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - __unmask_IO_APIC_irq(irq); - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) -{ - struct IO_APIC_route_entry entry; - - /* Check delivery_mode to be sure we're not clearing an SMI pin */ - entry = ioapic_read_entry(apic, pin); - if (entry.delivery_mode == dest_SMI) - return; - - /* - * Disable it in the IO-APIC irq-routing table: - */ - ioapic_mask_entry(apic, pin); -} - -static void clear_IO_APIC(void) -{ - int apic, pin; - - for (apic = 0; apic < nr_ioapics; apic++) - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) - clear_IO_APIC_pin(apic, pin); -} - -#ifdef CONFIG_SMP -static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t cpumask) -{ - unsigned long flags; - int pin; - struct irq_pin_list *entry = irq_2_pin + irq; - unsigned int apicid_value; - cpumask_t tmp; - - cpus_and(tmp, cpumask, cpu_online_map); - if (cpus_empty(tmp)) - tmp = TARGET_CPUS; - - cpus_and(cpumask, tmp, CPU_MASK_ALL); - - apicid_value = cpu_mask_to_apicid(cpumask); - /* Prepare to do the io_apic_write */ - apicid_value = apicid_value << 24; - spin_lock_irqsave(&ioapic_lock, flags); - for (;;) { - pin = entry->pin; - if (pin == -1) - break; - io_apic_write(entry->apic, 0x10 + 1 + pin*2, apicid_value); - if (!entry->next) - break; - entry = irq_2_pin + entry->next; - } - irq_desc[irq].affinity = cpumask; - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -#if defined(CONFIG_IRQBALANCE) -# include <asm/processor.h> /* kernel_thread() */ -# include <linux/kernel_stat.h> /* kstat */ -# include <linux/slab.h> /* kmalloc() */ -# include <linux/timer.h> - -#define IRQBALANCE_CHECK_ARCH -999 -#define MAX_BALANCED_IRQ_INTERVAL (5*HZ) -#define MIN_BALANCED_IRQ_INTERVAL (HZ/2) -#define BALANCED_IRQ_MORE_DELTA (HZ/10) -#define BALANCED_IRQ_LESS_DELTA (HZ) - -static int irqbalance_disabled __read_mostly = IRQBALANCE_CHECK_ARCH; -static int physical_balance __read_mostly; -static long balanced_irq_interval __read_mostly = MAX_BALANCED_IRQ_INTERVAL; - -static struct irq_cpu_info { - unsigned long *last_irq; - unsigned long *irq_delta; - unsigned long irq; -} irq_cpu_data[NR_CPUS]; - -#define CPU_IRQ(cpu) (irq_cpu_data[cpu].irq) -#define LAST_CPU_IRQ(cpu, irq) (irq_cpu_data[cpu].last_irq[irq]) -#define IRQ_DELTA(cpu, irq) (irq_cpu_data[cpu].irq_delta[irq]) - -#define IDLE_ENOUGH(cpu,now) \ - (idle_cpu(cpu) && ((now) - per_cpu(irq_stat, (cpu)).idle_timestamp > 1)) - -#define IRQ_ALLOWED(cpu, allowed_mask) cpu_isset(cpu, allowed_mask) - -#define CPU_TO_PACKAGEINDEX(i) (first_cpu(per_cpu(cpu_sibling_map, i))) - -static cpumask_t balance_irq_affinity[NR_IRQS] = { - [0 ... NR_IRQS-1] = CPU_MASK_ALL -}; - -void set_balance_irq_affinity(unsigned int irq, cpumask_t mask) -{ - balance_irq_affinity[irq] = mask; -} - -static unsigned long move(int curr_cpu, cpumask_t allowed_mask, - unsigned long now, int direction) -{ - int search_idle = 1; - int cpu = curr_cpu; - - goto inside; - - do { - if (unlikely(cpu == curr_cpu)) - search_idle = 0; -inside: - if (direction == 1) { - cpu++; - if (cpu >= NR_CPUS) - cpu = 0; - } else { - cpu--; - if (cpu == -1) - cpu = NR_CPUS-1; - } - } while (!cpu_online(cpu) || !IRQ_ALLOWED(cpu, allowed_mask) || - (search_idle && !IDLE_ENOUGH(cpu, now))); - - return cpu; -} - -static inline void balance_irq(int cpu, int irq) -{ - unsigned long now = jiffies; - cpumask_t allowed_mask; - unsigned int new_cpu; - - if (irqbalance_disabled) - return; - - cpus_and(allowed_mask, cpu_online_map, balance_irq_affinity[irq]); - new_cpu = move(cpu, allowed_mask, now, 1); - if (cpu != new_cpu) - set_pending_irq(irq, cpumask_of_cpu(new_cpu)); -} - -static inline void rotate_irqs_among_cpus(unsigned long useful_load_threshold) -{ - int i, j; - - for_each_online_cpu(i) { - for (j = 0; j < NR_IRQS; j++) { - if (!irq_desc[j].action) - continue; - /* Is it a significant load ? */ - if (IRQ_DELTA(CPU_TO_PACKAGEINDEX(i), j) < - useful_load_threshold) - continue; - balance_irq(i, j); - } - } - balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL, - balanced_irq_interval - BALANCED_IRQ_LESS_DELTA); - return; -} - -static void do_irq_balance(void) -{ - int i, j; - unsigned long max_cpu_irq = 0, min_cpu_irq = (~0); - unsigned long move_this_load = 0; - int max_loaded = 0, min_loaded = 0; - int load; - unsigned long useful_load_threshold = balanced_irq_interval + 10; - int selected_irq; - int tmp_loaded, first_attempt = 1; - unsigned long tmp_cpu_irq; - unsigned long imbalance = 0; - cpumask_t allowed_mask, target_cpu_mask, tmp; - - for_each_possible_cpu(i) { - int package_index; - CPU_IRQ(i) = 0; - if (!cpu_online(i)) - continue; - package_index = CPU_TO_PACKAGEINDEX(i); - for (j = 0; j < NR_IRQS; j++) { - unsigned long value_now, delta; - /* Is this an active IRQ or balancing disabled ? */ - if (!irq_desc[j].action || irq_balancing_disabled(j)) - continue; - if (package_index == i) - IRQ_DELTA(package_index, j) = 0; - /* Determine the total count per processor per IRQ */ - value_now = (unsigned long) kstat_cpu(i).irqs[j]; - - /* Determine the activity per processor per IRQ */ - delta = value_now - LAST_CPU_IRQ(i, j); - - /* Update last_cpu_irq[][] for the next time */ - LAST_CPU_IRQ(i, j) = value_now; - - /* Ignore IRQs whose rate is less than the clock */ - if (delta < useful_load_threshold) - continue; - /* update the load for the processor or package total */ - IRQ_DELTA(package_index, j) += delta; - - /* Keep track of the higher numbered sibling as well */ - if (i != package_index) - CPU_IRQ(i) += delta; - /* - * We have sibling A and sibling B in the package - * - * cpu_irq[A] = load for cpu A + load for cpu B - * cpu_irq[B] = load for cpu B - */ - CPU_IRQ(package_index) += delta; - } - } - /* Find the least loaded processor package */ - for_each_online_cpu(i) { - if (i != CPU_TO_PACKAGEINDEX(i)) - continue; - if (min_cpu_irq > CPU_IRQ(i)) { - min_cpu_irq = CPU_IRQ(i); - min_loaded = i; - } - } - max_cpu_irq = ULONG_MAX; - -tryanothercpu: - /* - * Look for heaviest loaded processor. - * We may come back to get the next heaviest loaded processor. - * Skip processors with trivial loads. - */ - tmp_cpu_irq = 0; - tmp_loaded = -1; - for_each_online_cpu(i) { - if (i != CPU_TO_PACKAGEINDEX(i)) - continue; - if (max_cpu_irq <= CPU_IRQ(i)) - continue; - if (tmp_cpu_irq < CPU_IRQ(i)) { - tmp_cpu_irq = CPU_IRQ(i); - tmp_loaded = i; - } - } - - if (tmp_loaded == -1) { - /* - * In the case of small number of heavy interrupt sources, - * loading some of the cpus too much. We use Ingo's original - * approach to rotate them around. - */ - if (!first_attempt && imbalance >= useful_load_threshold) { - rotate_irqs_among_cpus(useful_load_threshold); - return; - } - goto not_worth_the_effort; - } - - first_attempt = 0; /* heaviest search */ - max_cpu_irq = tmp_cpu_irq; /* load */ - max_loaded = tmp_loaded; /* processor */ - imbalance = (max_cpu_irq - min_cpu_irq) / 2; - - /* - * if imbalance is less than approx 10% of max load, then - * observe diminishing returns action. - quit - */ - if (imbalance < (max_cpu_irq >> 3)) - goto not_worth_the_effort; - -tryanotherirq: - /* if we select an IRQ to move that can't go where we want, then - * see if there is another one to try. - */ - move_this_load = 0; - selected_irq = -1; - for (j = 0; j < NR_IRQS; j++) { - /* Is this an active IRQ? */ - if (!irq_desc[j].action) - continue; - if (imbalance <= IRQ_DELTA(max_loaded, j)) - continue; - /* Try to find the IRQ that is closest to the imbalance - * without going over. - */ - if (move_this_load < IRQ_DELTA(max_loaded, j)) { - move_this_load = IRQ_DELTA(max_loaded, j); - selected_irq = j; - } - } - if (selected_irq == -1) - goto tryanothercpu; - - imbalance = move_this_load; - - /* For physical_balance case, we accumulated both load - * values in the one of the siblings cpu_irq[], - * to use the same code for physical and logical processors - * as much as possible. - * - * NOTE: the cpu_irq[] array holds the sum of the load for - * sibling A and sibling B in the slot for the lowest numbered - * sibling (A), _AND_ the load for sibling B in the slot for - * the higher numbered sibling. - * - * We seek the least loaded sibling by making the comparison - * (A+B)/2 vs B - */ - load = CPU_IRQ(min_loaded) >> 1; - for_each_cpu_mask(j, per_cpu(cpu_sibling_map, min_loaded)) { - if (load > CPU_IRQ(j)) { - /* This won't change cpu_sibling_map[min_loaded] */ - load = CPU_IRQ(j); - min_loaded = j; - } - } - - cpus_and(allowed_mask, - cpu_online_map, - balance_irq_affinity[selected_irq]); - target_cpu_mask = cpumask_of_cpu(min_loaded); - cpus_and(tmp, target_cpu_mask, allowed_mask); - - if (!cpus_empty(tmp)) { - /* mark for change destination */ - set_pending_irq(selected_irq, cpumask_of_cpu(min_loaded)); - - /* Since we made a change, come back sooner to - * check for more variation. - */ - balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL, - balanced_irq_interval - BALANCED_IRQ_LESS_DELTA); - return; - } - goto tryanotherirq; - -not_worth_the_effort: - /* - * if we did not find an IRQ to move, then adjust the time interval - * upward - */ - balanced_irq_interval = min((long)MAX_BALANCED_IRQ_INTERVAL, - balanced_irq_interval + BALANCED_IRQ_MORE_DELTA); - return; -} - -static int balanced_irq(void *unused) -{ - int i; - unsigned long prev_balance_time = jiffies; - long time_remaining = balanced_irq_interval; - - /* push everything to CPU 0 to give us a starting point. */ - for (i = 0 ; i < NR_IRQS ; i++) { - irq_desc[i].pending_mask = cpumask_of_cpu(0); - set_pending_irq(i, cpumask_of_cpu(0)); - } - - set_freezable(); - for ( ; ; ) { - time_remaining = schedule_timeout_interruptible(time_remaining); - try_to_freeze(); - if (time_after(jiffies, - prev_balance_time+balanced_irq_interval)) { - preempt_disable(); - do_irq_balance(); - prev_balance_time = jiffies; - time_remaining = balanced_irq_interval; - preempt_enable(); - } - } - return 0; -} - -static int __init balanced_irq_init(void) -{ - int i; - struct cpuinfo_x86 *c; - cpumask_t tmp; - - cpus_shift_right(tmp, cpu_online_map, 2); - c = &boot_cpu_data; - /* When not overwritten by the command line ask subarchitecture. */ - if (irqbalance_disabled == IRQBALANCE_CHECK_ARCH) - irqbalance_disabled = NO_BALANCE_IRQ; - if (irqbalance_disabled) - return 0; - - /* disable irqbalance completely if there is only one processor online */ - if (num_online_cpus() < 2) { - irqbalance_disabled = 1; - return 0; - } - /* - * Enable physical balance only if more than 1 physical processor - * is present - */ - if (smp_num_siblings > 1 && !cpus_empty(tmp)) - physical_balance = 1; - - for_each_online_cpu(i) { - irq_cpu_data[i].irq_delta = kzalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL); - irq_cpu_data[i].last_irq = kzalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL); - if (irq_cpu_data[i].irq_delta == NULL || irq_cpu_data[i].last_irq == NULL) { - printk(KERN_ERR "balanced_irq_init: out of memory"); - goto failed; - } - } - - printk(KERN_INFO "Starting balanced_irq\n"); - if (!IS_ERR(kthread_run(balanced_irq, NULL, "kirqd"))) - return 0; - printk(KERN_ERR "balanced_irq_init: failed to spawn balanced_irq"); -failed: - for_each_possible_cpu(i) { - kfree(irq_cpu_data[i].irq_delta); - irq_cpu_data[i].irq_delta = NULL; - kfree(irq_cpu_data[i].last_irq); - irq_cpu_data[i].last_irq = NULL; - } - return 0; -} - -int __devinit irqbalance_disable(char *str) -{ - irqbalance_disabled = 1; - return 1; -} - -__setup("noirqbalance", irqbalance_disable); - -late_initcall(balanced_irq_init); -#endif /* CONFIG_IRQBALANCE */ -#endif /* CONFIG_SMP */ - -#ifndef CONFIG_SMP -void send_IPI_self(int vector) -{ - unsigned int cfg; - - /* - * Wait for idle. - */ - apic_wait_icr_idle(); - cfg = APIC_DM_FIXED | APIC_DEST_SELF | vector | APIC_DEST_LOGICAL; - /* - * Send the IPI. The write to APIC_ICR fires this off. - */ - apic_write(APIC_ICR, cfg); -} -#endif /* !CONFIG_SMP */ - - -/* - * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to - * specific CPU-side IRQs. - */ - -#define MAX_PIRQS 8 -static int pirq_entries [MAX_PIRQS]; -static int pirqs_enabled; -int skip_ioapic_setup; - -static int __init ioapic_pirq_setup(char *str) -{ - int i, max; - int ints[MAX_PIRQS+1]; - - get_options(str, ARRAY_SIZE(ints), ints); - - for (i = 0; i < MAX_PIRQS; i++) - pirq_entries[i] = -1; - - pirqs_enabled = 1; - apic_printk(APIC_VERBOSE, KERN_INFO - "PIRQ redirection, working around broken MP-BIOS.\n"); - max = MAX_PIRQS; - if (ints[0] < MAX_PIRQS) - max = ints[0]; - - for (i = 0; i < max; i++) { - apic_printk(APIC_VERBOSE, KERN_DEBUG - "... PIRQ%d -> IRQ %d\n", i, ints[i+1]); - /* - * PIRQs are mapped upside down, usually. - */ - pirq_entries[MAX_PIRQS-i-1] = ints[i+1]; - } - return 1; -} - -__setup("pirq=", ioapic_pirq_setup); - -/* - * Find the IRQ entry number of a certain pin. - */ -static int find_irq_entry(int apic, int pin, int type) -{ - int i; - - for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mp_irqtype == type && - (mp_irqs[i].mp_dstapic == mp_ioapics[apic].mp_apicid || - mp_irqs[i].mp_dstapic == MP_APIC_ALL) && - mp_irqs[i].mp_dstirq == pin) - return i; - - return -1; -} - -/* - * Find the pin to which IRQ[irq] (ISA) is connected - */ -static int __init find_isa_irq_pin(int irq, int type) -{ - int i; - - for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mp_srcbus; - - if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mp_irqtype == type) && - (mp_irqs[i].mp_srcbusirq == irq)) - - return mp_irqs[i].mp_dstirq; - } - return -1; -} - -static int __init find_isa_irq_apic(int irq, int type) -{ - int i; - - for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mp_srcbus; - - if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mp_irqtype == type) && - (mp_irqs[i].mp_srcbusirq == irq)) - break; - } - if (i < mp_irq_entries) { - int apic; - for (apic = 0; apic < nr_ioapics; apic++) { - if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic) - return apic; - } - } - - return -1; -} - -/* - * Find a specific PCI IRQ entry. - * Not an __init, possibly needed by modules - */ -static int pin_2_irq(int idx, int apic, int pin); - -int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) -{ - int apic, i, best_guess = -1; - - apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, " - "slot:%d, pin:%d.\n", bus, slot, pin); - if (test_bit(bus, mp_bus_not_pci)) { - printk(KERN_WARNING "PCI BIOS passed nonexistent PCI bus %d!\n", bus); - return -1; - } - for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mp_srcbus; - - for (apic = 0; apic < nr_ioapics; apic++) - if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic || - mp_irqs[i].mp_dstapic == MP_APIC_ALL) - break; - - if (!test_bit(lbus, mp_bus_not_pci) && - !mp_irqs[i].mp_irqtype && - (bus == lbus) && - (slot == ((mp_irqs[i].mp_srcbusirq >> 2) & 0x1f))) { - int irq = pin_2_irq(i, apic, mp_irqs[i].mp_dstirq); - - if (!(apic || IO_APIC_IRQ(irq))) - continue; - - if (pin == (mp_irqs[i].mp_srcbusirq & 3)) - return irq; - /* - * Use the first all-but-pin matching entry as a - * best-guess fuzzy result for broken mptables. - */ - if (best_guess < 0) - best_guess = irq; - } - } - return best_guess; -} -EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector); - -/* - * This function currently is only a helper for the i386 smp boot process where - * we need to reprogram the ioredtbls to cater for the cpus which have come online - * so mask in all cases should simply be TARGET_CPUS - */ -#ifdef CONFIG_SMP -void __init setup_ioapic_dest(void) -{ - int pin, ioapic, irq, irq_entry; - - if (skip_ioapic_setup == 1) - return; - - for (ioapic = 0; ioapic < nr_ioapics; ioapic++) { - for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) { - irq_entry = find_irq_entry(ioapic, pin, mp_INT); - if (irq_entry == -1) - continue; - irq = pin_2_irq(irq_entry, ioapic, pin); - set_ioapic_affinity_irq(irq, TARGET_CPUS); - } - - } -} -#endif - -#if defined(CONFIG_EISA) || defined(CONFIG_MCA) -/* - * EISA Edge/Level control register, ELCR - */ -static int EISA_ELCR(unsigned int irq) -{ - if (irq < 16) { - unsigned int port = 0x4d0 + (irq >> 3); - return (inb(port) >> (irq & 7)) & 1; - } - apic_printk(APIC_VERBOSE, KERN_INFO - "Broken MPtable reports ISA irq %d\n", irq); - return 0; -} -#endif - -/* ISA interrupts are always polarity zero edge triggered, - * when listed as conforming in the MP table. */ - -#define default_ISA_trigger(idx) (0) -#define default_ISA_polarity(idx) (0) - -/* EISA interrupts are always polarity zero and can be edge or level - * trigger depending on the ELCR value. If an interrupt is listed as - * EISA conforming in the MP table, that means its trigger type must - * be read in from the ELCR */ - -#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mp_srcbusirq)) -#define default_EISA_polarity(idx) default_ISA_polarity(idx) - -/* PCI interrupts are always polarity one level triggered, - * when listed as conforming in the MP table. */ - -#define default_PCI_trigger(idx) (1) -#define default_PCI_polarity(idx) (1) - -/* MCA interrupts are always polarity zero level triggered, - * when listed as conforming in the MP table. */ - -#define default_MCA_trigger(idx) (1) -#define default_MCA_polarity(idx) default_ISA_polarity(idx) - -static int MPBIOS_polarity(int idx) -{ - int bus = mp_irqs[idx].mp_srcbus; - int polarity; - - /* - * Determine IRQ line polarity (high active or low active): - */ - switch (mp_irqs[idx].mp_irqflag & 3) { - case 0: /* conforms, ie. bus-type dependent polarity */ - { - polarity = test_bit(bus, mp_bus_not_pci)? - default_ISA_polarity(idx): - default_PCI_polarity(idx); - break; - } - case 1: /* high active */ - { - polarity = 0; - break; - } - case 2: /* reserved */ - { - printk(KERN_WARNING "broken BIOS!!\n"); - polarity = 1; - break; - } - case 3: /* low active */ - { - polarity = 1; - break; - } - default: /* invalid */ - { - printk(KERN_WARNING "broken BIOS!!\n"); - polarity = 1; - break; - } - } - return polarity; -} - -static int MPBIOS_trigger(int idx) -{ - int bus = mp_irqs[idx].mp_srcbus; - int trigger; - - /* - * Determine IRQ trigger mode (edge or level sensitive): - */ - switch ((mp_irqs[idx].mp_irqflag>>2) & 3) { - case 0: /* conforms, ie. bus-type dependent */ - { - trigger = test_bit(bus, mp_bus_not_pci)? - default_ISA_trigger(idx): - default_PCI_trigger(idx); -#if defined(CONFIG_EISA) || defined(CONFIG_MCA) - switch (mp_bus_id_to_type[bus]) { - case MP_BUS_ISA: /* ISA pin */ - { - /* set before the switch */ - break; - } - case MP_BUS_EISA: /* EISA pin */ - { - trigger = default_EISA_trigger(idx); - break; - } - case MP_BUS_PCI: /* PCI pin */ - { - /* set before the switch */ - break; - } - case MP_BUS_MCA: /* MCA pin */ - { - trigger = default_MCA_trigger(idx); - break; - } - default: - { - printk(KERN_WARNING "broken BIOS!!\n"); - trigger = 1; - break; - } - } -#endif - break; - } - case 1: /* edge */ - { - trigger = 0; - break; - } - case 2: /* reserved */ - { - printk(KERN_WARNING "broken BIOS!!\n"); - trigger = 1; - break; - } - case 3: /* level */ - { - trigger = 1; - break; - } - default: /* invalid */ - { - printk(KERN_WARNING "broken BIOS!!\n"); - trigger = 0; - break; - } - } - return trigger; -} - -static inline int irq_polarity(int idx) -{ - return MPBIOS_polarity(idx); -} - -static inline int irq_trigger(int idx) -{ - return MPBIOS_trigger(idx); -} - -static int pin_2_irq(int idx, int apic, int pin) -{ - int irq, i; - int bus = mp_irqs[idx].mp_srcbus; - - /* - * Debugging check, we are in big trouble if this message pops up! - */ - if (mp_irqs[idx].mp_dstirq != pin) - printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n"); - - if (test_bit(bus, mp_bus_not_pci)) - irq = mp_irqs[idx].mp_srcbusirq; - else { - /* - * PCI IRQs are mapped in order - */ - i = irq = 0; - while (i < apic) - irq += nr_ioapic_registers[i++]; - irq += pin; - - /* - * For MPS mode, so far only needed by ES7000 platform - */ - if (ioapic_renumber_irq) - irq = ioapic_renumber_irq(apic, irq); - } - - /* - * PCI IRQ command line redirection. Yes, limits are hardcoded. - */ - if ((pin >= 16) && (pin <= 23)) { - if (pirq_entries[pin-16] != -1) { - if (!pirq_entries[pin-16]) { - apic_printk(APIC_VERBOSE, KERN_DEBUG - "disabling PIRQ%d\n", pin-16); - } else { - irq = pirq_entries[pin-16]; - apic_printk(APIC_VERBOSE, KERN_DEBUG - "using PIRQ%d -> IRQ %d\n", - pin-16, irq); - } - } - } - return irq; -} - -static inline int IO_APIC_irq_trigger(int irq) -{ - int apic, idx, pin; - - for (apic = 0; apic < nr_ioapics; apic++) { - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - idx = find_irq_entry(apic, pin, mp_INT); - if ((idx != -1) && (irq == pin_2_irq(idx, apic, pin))) - return irq_trigger(idx); - } - } - /* - * nonexistent IRQs are edge default - */ - return 0; -} - -/* irq_vectors is indexed by the sum of all RTEs in all I/O APICs. */ -static u8 irq_vector[NR_IRQ_VECTORS] __read_mostly = { FIRST_DEVICE_VECTOR , 0 }; - -static int __assign_irq_vector(int irq) -{ - static int current_vector = FIRST_DEVICE_VECTOR, current_offset; - int vector, offset; - - BUG_ON((unsigned)irq >= NR_IRQ_VECTORS); - - if (irq_vector[irq] > 0) - return irq_vector[irq]; - - vector = current_vector; - offset = current_offset; -next: - vector += 8; - if (vector >= first_system_vector) { - offset = (offset + 1) % 8; - vector = FIRST_DEVICE_VECTOR + offset; - } - if (vector == current_vector) - return -ENOSPC; - if (test_and_set_bit(vector, used_vectors)) - goto next; - - current_vector = vector; - current_offset = offset; - irq_vector[irq] = vector; - - return vector; -} - -static int assign_irq_vector(int irq) -{ - unsigned long flags; - int vector; - - spin_lock_irqsave(&vector_lock, flags); - vector = __assign_irq_vector(irq); - spin_unlock_irqrestore(&vector_lock, flags); - - return vector; -} - -static struct irq_chip ioapic_chip; - -#define IOAPIC_AUTO -1 -#define IOAPIC_EDGE 0 -#define IOAPIC_LEVEL 1 - -static void ioapic_register_intr(int irq, int vector, unsigned long trigger) -{ - if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || - trigger == IOAPIC_LEVEL) { - irq_desc[irq].status |= IRQ_LEVEL; - set_irq_chip_and_handler_name(irq, &ioapic_chip, - handle_fasteoi_irq, "fasteoi"); - } else { - irq_desc[irq].status &= ~IRQ_LEVEL; - set_irq_chip_and_handler_name(irq, &ioapic_chip, - handle_edge_irq, "edge"); - } - set_intr_gate(vector, interrupt[irq]); -} - -static void __init setup_IO_APIC_irqs(void) -{ - struct IO_APIC_route_entry entry; - int apic, pin, idx, irq, first_notcon = 1, vector; - - apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); - - for (apic = 0; apic < nr_ioapics; apic++) { - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - - /* - * add it to the IO-APIC irq-routing table: - */ - memset(&entry, 0, sizeof(entry)); - - entry.delivery_mode = INT_DELIVERY_MODE; - entry.dest_mode = INT_DEST_MODE; - entry.mask = 0; /* enable IRQ */ - entry.dest.logical.logical_dest = - cpu_mask_to_apicid(TARGET_CPUS); - - idx = find_irq_entry(apic, pin, mp_INT); - if (idx == -1) { - if (first_notcon) { - apic_printk(APIC_VERBOSE, KERN_DEBUG - " IO-APIC (apicid-pin) %d-%d", - mp_ioapics[apic].mp_apicid, - pin); - first_notcon = 0; - } else - apic_printk(APIC_VERBOSE, ", %d-%d", - mp_ioapics[apic].mp_apicid, pin); - continue; - } - - if (!first_notcon) { - apic_printk(APIC_VERBOSE, " not connected.\n"); - first_notcon = 1; - } - - entry.trigger = irq_trigger(idx); - entry.polarity = irq_polarity(idx); - - if (irq_trigger(idx)) { - entry.trigger = 1; - entry.mask = 1; - } - - irq = pin_2_irq(idx, apic, pin); - /* - * skip adding the timer int on secondary nodes, which causes - * a small but painful rift in the time-space continuum - */ - if (multi_timer_check(apic, irq)) - continue; - else - add_pin_to_irq(irq, apic, pin); - - if (!apic && !IO_APIC_IRQ(irq)) - continue; - - if (IO_APIC_IRQ(irq)) { - vector = assign_irq_vector(irq); - entry.vector = vector; - ioapic_register_intr(irq, vector, IOAPIC_AUTO); - - if (!apic && (irq < 16)) - disable_8259A_irq(irq); - } - ioapic_write_entry(apic, pin, entry); - } - } - - if (!first_notcon) - apic_printk(APIC_VERBOSE, " not connected.\n"); -} - -/* - * Set up the timer pin, possibly with the 8259A-master behind. - */ -static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, - int vector) -{ - struct IO_APIC_route_entry entry; - - memset(&entry, 0, sizeof(entry)); - - /* - * We use logical delivery to get the timer IRQ - * to the first CPU. - */ - entry.dest_mode = INT_DEST_MODE; - entry.mask = 1; /* mask IRQ now */ - entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS); - entry.delivery_mode = INT_DELIVERY_MODE; - entry.polarity = 0; - entry.trigger = 0; - entry.vector = vector; - - /* - * The timer IRQ doesn't have to know that behind the - * scene we may have a 8259A-master in AEOI mode ... - */ - ioapic_register_intr(0, vector, IOAPIC_EDGE); - - /* - * Add it to the IO-APIC irq-routing table: - */ - ioapic_write_entry(apic, pin, entry); -} - - -__apicdebuginit(void) print_IO_APIC(void) -{ - int apic, i; - union IO_APIC_reg_00 reg_00; - union IO_APIC_reg_01 reg_01; - union IO_APIC_reg_02 reg_02; - union IO_APIC_reg_03 reg_03; - unsigned long flags; - - if (apic_verbosity == APIC_QUIET) - return; - - printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); - for (i = 0; i < nr_ioapics; i++) - printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n", - mp_ioapics[i].mp_apicid, nr_ioapic_registers[i]); - - /* - * We are a bit conservative about what we expect. We have to - * know about every hardware change ASAP. - */ - printk(KERN_INFO "testing the IO APIC.......................\n"); - - for (apic = 0; apic < nr_ioapics; apic++) { - - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(apic, 0); - reg_01.raw = io_apic_read(apic, 1); - if (reg_01.bits.version >= 0x10) - reg_02.raw = io_apic_read(apic, 2); - if (reg_01.bits.version >= 0x20) - reg_03.raw = io_apic_read(apic, 3); - spin_unlock_irqrestore(&ioapic_lock, flags); - - printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid); - printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); - printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); - printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type); - printk(KERN_DEBUG "....... : LTS : %X\n", reg_00.bits.LTS); - - printk(KERN_DEBUG ".... register #01: %08X\n", reg_01.raw); - printk(KERN_DEBUG "....... : max redirection entries: %04X\n", reg_01.bits.entries); - - printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ); - printk(KERN_DEBUG "....... : IO APIC version: %04X\n", reg_01.bits.version); - - /* - * Some Intel chipsets with IO APIC VERSION of 0x1? don't have reg_02, - * but the value of reg_02 is read as the previous read register - * value, so ignore it if reg_02 == reg_01. - */ - if (reg_01.bits.version >= 0x10 && reg_02.raw != reg_01.raw) { - printk(KERN_DEBUG ".... register #02: %08X\n", reg_02.raw); - printk(KERN_DEBUG "....... : arbitration: %02X\n", reg_02.bits.arbitration); - } - - /* - * Some Intel chipsets with IO APIC VERSION of 0x2? don't have reg_02 - * or reg_03, but the value of reg_0[23] is read as the previous read - * register value, so ignore it if reg_03 == reg_0[12]. - */ - if (reg_01.bits.version >= 0x20 && reg_03.raw != reg_02.raw && - reg_03.raw != reg_01.raw) { - printk(KERN_DEBUG ".... register #03: %08X\n", reg_03.raw); - printk(KERN_DEBUG "....... : Boot DT : %X\n", reg_03.bits.boot_DT); - } - - printk(KERN_DEBUG ".... IRQ redirection table:\n"); - - printk(KERN_DEBUG " NR Log Phy Mask Trig IRR Pol" - " Stat Dest Deli Vect: \n"); - - for (i = 0; i <= reg_01.bits.entries; i++) { - struct IO_APIC_route_entry entry; - - entry = ioapic_read_entry(apic, i); - - printk(KERN_DEBUG " %02x %03X %02X ", - i, - entry.dest.logical.logical_dest, - entry.dest.physical.physical_dest - ); - - printk("%1d %1d %1d %1d %1d %1d %1d %02X\n", - entry.mask, - entry.trigger, - entry.irr, - entry.polarity, - entry.delivery_status, - entry.dest_mode, - entry.delivery_mode, - entry.vector - ); - } - } - printk(KERN_DEBUG "IRQ to pin mappings:\n"); - for (i = 0; i < NR_IRQS; i++) { - struct irq_pin_list *entry = irq_2_pin + i; - if (entry->pin < 0) - continue; - printk(KERN_DEBUG "IRQ%d ", i); - for (;;) { - printk("-> %d:%d", entry->apic, entry->pin); - if (!entry->next) - break; - entry = irq_2_pin + entry->next; - } - printk("\n"); - } - - printk(KERN_INFO ".................................... done.\n"); - - return; -} - -__apicdebuginit(void) print_APIC_bitfield(int base) -{ - unsigned int v; - int i, j; - - if (apic_verbosity == APIC_QUIET) - return; - - printk(KERN_DEBUG "0123456789abcdef0123456789abcdef\n" KERN_DEBUG); - for (i = 0; i < 8; i++) { - v = apic_read(base + i*0x10); - for (j = 0; j < 32; j++) { - if (v & (1<<j)) - printk("1"); - else - printk("0"); - } - printk("\n"); - } -} - -__apicdebuginit(void) print_local_APIC(void *dummy) -{ - unsigned int v, ver, maxlvt; - u64 icr; - - if (apic_verbosity == APIC_QUIET) - return; - - printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n", - smp_processor_id(), hard_smp_processor_id()); - v = apic_read(APIC_ID); - printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, - GET_APIC_ID(v)); - v = apic_read(APIC_LVR); - printk(KERN_INFO "... APIC VERSION: %08x\n", v); - ver = GET_APIC_VERSION(v); - maxlvt = lapic_get_maxlvt(); - - v = apic_read(APIC_TASKPRI); - printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK); - - if (APIC_INTEGRATED(ver)) { /* !82489DX */ - v = apic_read(APIC_ARBPRI); - printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v, - v & APIC_ARBPRI_MASK); - v = apic_read(APIC_PROCPRI); - printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v); - } - - v = apic_read(APIC_EOI); - printk(KERN_DEBUG "... APIC EOI: %08x\n", v); - v = apic_read(APIC_RRR); - printk(KERN_DEBUG "... APIC RRR: %08x\n", v); - v = apic_read(APIC_LDR); - printk(KERN_DEBUG "... APIC LDR: %08x\n", v); - v = apic_read(APIC_DFR); - printk(KERN_DEBUG "... APIC DFR: %08x\n", v); - v = apic_read(APIC_SPIV); - printk(KERN_DEBUG "... APIC SPIV: %08x\n", v); - - printk(KERN_DEBUG "... APIC ISR field:\n"); - print_APIC_bitfield(APIC_ISR); - printk(KERN_DEBUG "... APIC TMR field:\n"); - print_APIC_bitfield(APIC_TMR); - printk(KERN_DEBUG "... APIC IRR field:\n"); - print_APIC_bitfield(APIC_IRR); - - if (APIC_INTEGRATED(ver)) { /* !82489DX */ - if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ - apic_write(APIC_ESR, 0); - v = apic_read(APIC_ESR); - printk(KERN_DEBUG "... APIC ESR: %08x\n", v); - } - - icr = apic_icr_read(); - printk(KERN_DEBUG "... APIC ICR: %08x\n", icr); - printk(KERN_DEBUG "... APIC ICR2: %08x\n", icr >> 32); - - v = apic_read(APIC_LVTT); - printk(KERN_DEBUG "... APIC LVTT: %08x\n", v); - - if (maxlvt > 3) { /* PC is LVT#4. */ - v = apic_read(APIC_LVTPC); - printk(KERN_DEBUG "... APIC LVTPC: %08x\n", v); - } - v = apic_read(APIC_LVT0); - printk(KERN_DEBUG "... APIC LVT0: %08x\n", v); - v = apic_read(APIC_LVT1); - printk(KERN_DEBUG "... APIC LVT1: %08x\n", v); - - if (maxlvt > 2) { /* ERR is LVT#3. */ - v = apic_read(APIC_LVTERR); - printk(KERN_DEBUG "... APIC LVTERR: %08x\n", v); - } - - v = apic_read(APIC_TMICT); - printk(KERN_DEBUG "... APIC TMICT: %08x\n", v); - v = apic_read(APIC_TMCCT); - printk(KERN_DEBUG "... APIC TMCCT: %08x\n", v); - v = apic_read(APIC_TDCR); - printk(KERN_DEBUG "... APIC TDCR: %08x\n", v); - printk("\n"); -} - -__apicdebuginit(void) print_all_local_APICs(void) -{ - on_each_cpu(print_local_APIC, NULL, 1); -} - -__apicdebuginit(void) print_PIC(void) -{ - unsigned int v; - unsigned long flags; - - if (apic_verbosity == APIC_QUIET) - return; - - printk(KERN_DEBUG "\nprinting PIC contents\n"); - - spin_lock_irqsave(&i8259A_lock, flags); - - v = inb(0xa1) << 8 | inb(0x21); - printk(KERN_DEBUG "... PIC IMR: %04x\n", v); - - v = inb(0xa0) << 8 | inb(0x20); - printk(KERN_DEBUG "... PIC IRR: %04x\n", v); - - outb(0x0b, 0xa0); - outb(0x0b, 0x20); - v = inb(0xa0) << 8 | inb(0x20); - outb(0x0a, 0xa0); - outb(0x0a, 0x20); - - spin_unlock_irqrestore(&i8259A_lock, flags); - - printk(KERN_DEBUG "... PIC ISR: %04x\n", v); - - v = inb(0x4d1) << 8 | inb(0x4d0); - printk(KERN_DEBUG "... PIC ELCR: %04x\n", v); -} - -__apicdebuginit(int) print_all_ICs(void) -{ - print_PIC(); - print_all_local_APICs(); - print_IO_APIC(); - - return 0; -} - -fs_initcall(print_all_ICs); - - -static void __init enable_IO_APIC(void) -{ - union IO_APIC_reg_01 reg_01; - int i8259_apic, i8259_pin; - int i, apic; - unsigned long flags; - - for (i = 0; i < PIN_MAP_SIZE; i++) { - irq_2_pin[i].pin = -1; - irq_2_pin[i].next = 0; - } - if (!pirqs_enabled) - for (i = 0; i < MAX_PIRQS; i++) - pirq_entries[i] = -1; - - /* - * The number of IO-APIC IRQ registers (== #pins): - */ - for (apic = 0; apic < nr_ioapics; apic++) { - spin_lock_irqsave(&ioapic_lock, flags); - reg_01.raw = io_apic_read(apic, 1); - spin_unlock_irqrestore(&ioapic_lock, flags); - nr_ioapic_registers[apic] = reg_01.bits.entries+1; - } - for (apic = 0; apic < nr_ioapics; apic++) { - int pin; - /* See if any of the pins is in ExtINT mode */ - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - struct IO_APIC_route_entry entry; - entry = ioapic_read_entry(apic, pin); - - - /* If the interrupt line is enabled and in ExtInt mode - * I have found the pin where the i8259 is connected. - */ - if ((entry.mask == 0) && (entry.delivery_mode == dest_ExtINT)) { - ioapic_i8259.apic = apic; - ioapic_i8259.pin = pin; - goto found_i8259; - } - } - } - found_i8259: - /* Look to see what if the MP table has reported the ExtINT */ - /* If we could not find the appropriate pin by looking at the ioapic - * the i8259 probably is not connected the ioapic but give the - * mptable a chance anyway. - */ - i8259_pin = find_isa_irq_pin(0, mp_ExtINT); - i8259_apic = find_isa_irq_apic(0, mp_ExtINT); - /* Trust the MP table if nothing is setup in the hardware */ - if ((ioapic_i8259.pin == -1) && (i8259_pin >= 0)) { - printk(KERN_WARNING "ExtINT not setup in hardware but reported by MP table\n"); - ioapic_i8259.pin = i8259_pin; - ioapic_i8259.apic = i8259_apic; - } - /* Complain if the MP table and the hardware disagree */ - if (((ioapic_i8259.apic != i8259_apic) || (ioapic_i8259.pin != i8259_pin)) && - (i8259_pin >= 0) && (ioapic_i8259.pin >= 0)) - { - printk(KERN_WARNING "ExtINT in hardware and MP table differ\n"); - } - - /* - * Do not trust the IO-APIC being empty at bootup - */ - clear_IO_APIC(); -} - -/* - * Not an __init, needed by the reboot code - */ -void disable_IO_APIC(void) -{ - /* - * Clear the IO-APIC before rebooting: - */ - clear_IO_APIC(); - - /* - * If the i8259 is routed through an IOAPIC - * Put that IOAPIC in virtual wire mode - * so legacy interrupts can be delivered. - */ - if (ioapic_i8259.pin != -1) { - struct IO_APIC_route_entry entry; - - memset(&entry, 0, sizeof(entry)); - entry.mask = 0; /* Enabled */ - entry.trigger = 0; /* Edge */ - entry.irr = 0; - entry.polarity = 0; /* High */ - entry.delivery_status = 0; - entry.dest_mode = 0; /* Physical */ - entry.delivery_mode = dest_ExtINT; /* ExtInt */ - entry.vector = 0; - entry.dest.physical.physical_dest = read_apic_id(); - - /* - * Add it to the IO-APIC irq-routing table: - */ - ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry); - } - disconnect_bsp_APIC(ioapic_i8259.pin != -1); -} - -/* - * function to set the IO-APIC physical IDs based on the - * values stored in the MPC table. - * - * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999 - */ - -static void __init setup_ioapic_ids_from_mpc(void) -{ - union IO_APIC_reg_00 reg_00; - physid_mask_t phys_id_present_map; - int apic; - int i; - unsigned char old_id; - unsigned long flags; - - if (x86_quirks->setup_ioapic_ids && x86_quirks->setup_ioapic_ids()) - return; - - /* - * Don't check I/O APIC IDs for xAPIC systems. They have - * no meaning without the serial APIC bus. - */ - if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) - || APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) - return; - /* - * This is broken; anything with a real cpu count has to - * circumvent this idiocy regardless. - */ - phys_id_present_map = ioapic_phys_id_map(phys_cpu_present_map); - - /* - * Set the IOAPIC ID to the value stored in the MPC table. - */ - for (apic = 0; apic < nr_ioapics; apic++) { - - /* Read the register 0 value */ - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(apic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); - - old_id = mp_ioapics[apic].mp_apicid; - - if (mp_ioapics[apic].mp_apicid >= get_physical_broadcast()) { - printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n", - apic, mp_ioapics[apic].mp_apicid); - printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", - reg_00.bits.ID); - mp_ioapics[apic].mp_apicid = reg_00.bits.ID; - } - - /* - * Sanity check, is the ID really free? Every APIC in a - * system must have a unique ID or we get lots of nice - * 'stuck on smp_invalidate_needed IPI wait' messages. - */ - if (check_apicid_used(phys_id_present_map, - mp_ioapics[apic].mp_apicid)) { - printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n", - apic, mp_ioapics[apic].mp_apicid); - for (i = 0; i < get_physical_broadcast(); i++) - if (!physid_isset(i, phys_id_present_map)) - break; - if (i >= get_physical_broadcast()) - panic("Max APIC ID exceeded!\n"); - printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", - i); - physid_set(i, phys_id_present_map); - mp_ioapics[apic].mp_apicid = i; - } else { - physid_mask_t tmp; - tmp = apicid_to_cpu_present(mp_ioapics[apic].mp_apicid); - apic_printk(APIC_VERBOSE, "Setting %d in the " - "phys_id_present_map\n", - mp_ioapics[apic].mp_apicid); - physids_or(phys_id_present_map, phys_id_present_map, tmp); - } - - - /* - * We need to adjust the IRQ routing table - * if the ID changed. - */ - if (old_id != mp_ioapics[apic].mp_apicid) - for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mp_dstapic == old_id) - mp_irqs[i].mp_dstapic - = mp_ioapics[apic].mp_apicid; - - /* - * Read the right value from the MPC table and - * write it into the ID register. - */ - apic_printk(APIC_VERBOSE, KERN_INFO - "...changing IO-APIC physical APIC ID to %d ...", - mp_ioapics[apic].mp_apicid); - - reg_00.bits.ID = mp_ioapics[apic].mp_apicid; - spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(apic, 0, reg_00.raw); - spin_unlock_irqrestore(&ioapic_lock, flags); - - /* - * Sanity check - */ - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(apic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); - if (reg_00.bits.ID != mp_ioapics[apic].mp_apicid) - printk("could not set ID!\n"); - else - apic_printk(APIC_VERBOSE, " ok.\n"); - } -} - -int no_timer_check __initdata; - -static int __init notimercheck(char *s) -{ - no_timer_check = 1; - return 1; -} -__setup("no_timer_check", notimercheck); - -/* - * There is a nasty bug in some older SMP boards, their mptable lies - * about the timer IRQ. We do the following to work around the situation: - * - * - timer IRQ defaults to IO-APIC IRQ - * - if this function detects that timer IRQs are defunct, then we fall - * back to ISA timer IRQs - */ -static int __init timer_irq_works(void) -{ - unsigned long t1 = jiffies; - unsigned long flags; - - if (no_timer_check) - return 1; - - local_save_flags(flags); - local_irq_enable(); - /* Let ten ticks pass... */ - mdelay((10 * 1000) / HZ); - local_irq_restore(flags); - - /* - * Expect a few ticks at least, to be sure some possible - * glue logic does not lock up after one or two first - * ticks in a non-ExtINT mode. Also the local APIC - * might have cached one ExtINT interrupt. Finally, at - * least one tick may be lost due to delays. - */ - if (time_after(jiffies, t1 + 4)) - return 1; - - return 0; -} - -/* - * In the SMP+IOAPIC case it might happen that there are an unspecified - * number of pending IRQ events unhandled. These cases are very rare, - * so we 'resend' these IRQs via IPIs, to the same CPU. It's much - * better to do it this way as thus we do not have to be aware of - * 'pending' interrupts in the IRQ path, except at this point. - */ -/* - * Edge triggered needs to resend any interrupt - * that was delayed but this is now handled in the device - * independent code. - */ - -/* - * Startup quirk: - * - * Starting up a edge-triggered IO-APIC interrupt is - * nasty - we need to make sure that we get the edge. - * If it is already asserted for some reason, we need - * return 1 to indicate that is was pending. - * - * This is not complete - we should be able to fake - * an edge even if it isn't on the 8259A... - * - * (We do this for level-triggered IRQs too - it cannot hurt.) - */ -static unsigned int startup_ioapic_irq(unsigned int irq) -{ - int was_pending = 0; - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - if (irq < 16) { - disable_8259A_irq(irq); - if (i8259A_irq_pending(irq)) - was_pending = 1; - } - __unmask_IO_APIC_irq(irq); - spin_unlock_irqrestore(&ioapic_lock, flags); - - return was_pending; -} - -static void ack_ioapic_irq(unsigned int irq) -{ - move_native_irq(irq); - ack_APIC_irq(); -} - -static void ack_ioapic_quirk_irq(unsigned int irq) -{ - unsigned long v; - int i; - - move_native_irq(irq); -/* - * It appears there is an erratum which affects at least version 0x11 - * of I/O APIC (that's the 82093AA and cores integrated into various - * chipsets). Under certain conditions a level-triggered interrupt is - * erroneously delivered as edge-triggered one but the respective IRR - * bit gets set nevertheless. As a result the I/O unit expects an EOI - * message but it will never arrive and further interrupts are blocked - * from the source. The exact reason is so far unknown, but the - * phenomenon was observed when two consecutive interrupt requests - * from a given source get delivered to the same CPU and the source is - * temporarily disabled in between. - * - * A workaround is to simulate an EOI message manually. We achieve it - * by setting the trigger mode to edge and then to level when the edge - * trigger mode gets detected in the TMR of a local APIC for a - * level-triggered interrupt. We mask the source for the time of the - * operation to prevent an edge-triggered interrupt escaping meanwhile. - * The idea is from Manfred Spraul. --macro - */ - i = irq_vector[irq]; - - v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1)); - - ack_APIC_irq(); - - if (!(v & (1 << (i & 0x1f)))) { - atomic_inc(&irq_mis_count); - spin_lock(&ioapic_lock); - __mask_and_edge_IO_APIC_irq(irq); - __unmask_and_level_IO_APIC_irq(irq); - spin_unlock(&ioapic_lock); - } -} - -static int ioapic_retrigger_irq(unsigned int irq) -{ - send_IPI_self(irq_vector[irq]); - - return 1; -} - -static struct irq_chip ioapic_chip __read_mostly = { - .name = "IO-APIC", - .startup = startup_ioapic_irq, - .mask = mask_IO_APIC_irq, - .unmask = unmask_IO_APIC_irq, - .ack = ack_ioapic_irq, - .eoi = ack_ioapic_quirk_irq, -#ifdef CONFIG_SMP - .set_affinity = set_ioapic_affinity_irq, -#endif - .retrigger = ioapic_retrigger_irq, -}; - - -static inline void init_IO_APIC_traps(void) -{ - int irq; - - /* - * NOTE! The local APIC isn't very good at handling - * multiple interrupts at the same interrupt level. - * As the interrupt level is determined by taking the - * vector number and shifting that right by 4, we - * want to spread these out a bit so that they don't - * all fall in the same interrupt level. - * - * Also, we've got to be careful not to trash gate - * 0x80, because int 0x80 is hm, kind of importantish. ;) - */ - for (irq = 0; irq < NR_IRQS ; irq++) { - if (IO_APIC_IRQ(irq) && !irq_vector[irq]) { - /* - * Hmm.. We don't have an entry for this, - * so default to an old-fashioned 8259 - * interrupt if we can.. - */ - if (irq < 16) - make_8259A_irq(irq); - else - /* Strange. Oh, well.. */ - irq_desc[irq].chip = &no_irq_chip; - } - } -} - -/* - * The local APIC irq-chip implementation: - */ - -static void ack_lapic_irq(unsigned int irq) -{ - ack_APIC_irq(); -} - -static void mask_lapic_irq(unsigned int irq) -{ - unsigned long v; - - v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v | APIC_LVT_MASKED); -} - -static void unmask_lapic_irq(unsigned int irq) -{ - unsigned long v; - - v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); -} - -static struct irq_chip lapic_chip __read_mostly = { - .name = "local-APIC", - .mask = mask_lapic_irq, - .unmask = unmask_lapic_irq, - .ack = ack_lapic_irq, -}; - -static void lapic_register_intr(int irq, int vector) -{ - irq_desc[irq].status &= ~IRQ_LEVEL; - set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq, - "edge"); - set_intr_gate(vector, interrupt[irq]); -} - -static void __init setup_nmi(void) -{ - /* - * Dirty trick to enable the NMI watchdog ... - * We put the 8259A master into AEOI mode and - * unmask on all local APICs LVT0 as NMI. - * - * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire') - * is from Maciej W. Rozycki - so we do not have to EOI from - * the NMI handler or the timer interrupt. - */ - apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ..."); - - enable_NMI_through_LVT0(); - - apic_printk(APIC_VERBOSE, " done.\n"); -} - -/* - * This looks a bit hackish but it's about the only one way of sending - * a few INTA cycles to 8259As and any associated glue logic. ICR does - * not support the ExtINT mode, unfortunately. We need to send these - * cycles as some i82489DX-based boards have glue logic that keeps the - * 8259A interrupt line asserted until INTA. --macro - */ -static inline void __init unlock_ExtINT_logic(void) -{ - int apic, pin, i; - struct IO_APIC_route_entry entry0, entry1; - unsigned char save_control, save_freq_select; - - pin = find_isa_irq_pin(8, mp_INT); - if (pin == -1) { - WARN_ON_ONCE(1); - return; - } - apic = find_isa_irq_apic(8, mp_INT); - if (apic == -1) { - WARN_ON_ONCE(1); - return; - } - - entry0 = ioapic_read_entry(apic, pin); - clear_IO_APIC_pin(apic, pin); - - memset(&entry1, 0, sizeof(entry1)); - - entry1.dest_mode = 0; /* physical delivery */ - entry1.mask = 0; /* unmask IRQ now */ - entry1.dest.physical.physical_dest = hard_smp_processor_id(); - entry1.delivery_mode = dest_ExtINT; - entry1.polarity = entry0.polarity; - entry1.trigger = 0; - entry1.vector = 0; - - ioapic_write_entry(apic, pin, entry1); - - save_control = CMOS_READ(RTC_CONTROL); - save_freq_select = CMOS_READ(RTC_FREQ_SELECT); - CMOS_WRITE((save_freq_select & ~RTC_RATE_SELECT) | 0x6, - RTC_FREQ_SELECT); - CMOS_WRITE(save_control | RTC_PIE, RTC_CONTROL); - - i = 100; - while (i-- > 0) { - mdelay(10); - if ((CMOS_READ(RTC_INTR_FLAGS) & RTC_PF) == RTC_PF) - i -= 10; - } - - CMOS_WRITE(save_control, RTC_CONTROL); - CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT); - clear_IO_APIC_pin(apic, pin); - - ioapic_write_entry(apic, pin, entry0); -} - -/* - * This code may look a bit paranoid, but it's supposed to cooperate with - * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ - * is so screwy. Thanks to Brian Perkins for testing/hacking this beast - * fanatically on his truly buggy board. - */ -static inline void __init check_timer(void) -{ - int apic1, pin1, apic2, pin2; - int no_pin1 = 0; - int vector; - unsigned int ver; - unsigned long flags; - - local_irq_save(flags); - - ver = apic_read(APIC_LVR); - ver = GET_APIC_VERSION(ver); - - /* - * get/set the timer IRQ vector: - */ - disable_8259A_irq(0); - vector = assign_irq_vector(0); - set_intr_gate(vector, interrupt[0]); - - /* - * As IRQ0 is to be enabled in the 8259A, the virtual - * wire has to be disabled in the local APIC. Also - * timer interrupts need to be acknowledged manually in - * the 8259A for the i82489DX when using the NMI - * watchdog as that APIC treats NMIs as level-triggered. - * The AEOI mode will finish them in the 8259A - * automatically. - */ - apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); - init_8259A(1); - timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); - - pin1 = find_isa_irq_pin(0, mp_INT); - apic1 = find_isa_irq_apic(0, mp_INT); - pin2 = ioapic_i8259.pin; - apic2 = ioapic_i8259.apic; - - apic_printk(APIC_QUIET, KERN_INFO "..TIMER: vector=0x%02X " - "apic1=%d pin1=%d apic2=%d pin2=%d\n", - vector, apic1, pin1, apic2, pin2); - - /* - * Some BIOS writers are clueless and report the ExtINTA - * I/O APIC input from the cascaded 8259A as the timer - * interrupt input. So just in case, if only one pin - * was found above, try it both directly and through the - * 8259A. - */ - if (pin1 == -1) { - pin1 = pin2; - apic1 = apic2; - no_pin1 = 1; - } else if (pin2 == -1) { - pin2 = pin1; - apic2 = apic1; - } - - if (pin1 != -1) { - /* - * Ok, does IRQ0 through the IOAPIC work? - */ - if (no_pin1) { - add_pin_to_irq(0, apic1, pin1); - setup_timer_IRQ0_pin(apic1, pin1, vector); - } - unmask_IO_APIC_irq(0); - if (timer_irq_works()) { - if (nmi_watchdog == NMI_IO_APIC) { - setup_nmi(); - enable_8259A_irq(0); - } - if (disable_timer_pin_1 > 0) - clear_IO_APIC_pin(0, pin1); - goto out; - } - clear_IO_APIC_pin(apic1, pin1); - if (!no_pin1) - apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: " - "8254 timer not connected to IO-APIC\n"); - - apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer " - "(IRQ0) through the 8259A ...\n"); - apic_printk(APIC_QUIET, KERN_INFO - "..... (found apic %d pin %d) ...\n", apic2, pin2); - /* - * legacy devices should be connected to IO APIC #0 - */ - replace_pin_at_irq(0, apic1, pin1, apic2, pin2); - setup_timer_IRQ0_pin(apic2, pin2, vector); - unmask_IO_APIC_irq(0); - enable_8259A_irq(0); - if (timer_irq_works()) { - apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); - timer_through_8259 = 1; - if (nmi_watchdog == NMI_IO_APIC) { - disable_8259A_irq(0); - setup_nmi(); - enable_8259A_irq(0); - } - goto out; - } - /* - * Cleanup, just in case ... - */ - disable_8259A_irq(0); - clear_IO_APIC_pin(apic2, pin2); - apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); - } - - if (nmi_watchdog == NMI_IO_APIC) { - apic_printk(APIC_QUIET, KERN_WARNING "timer doesn't work " - "through the IO-APIC - disabling NMI Watchdog!\n"); - nmi_watchdog = NMI_NONE; - } - timer_ack = 0; - - apic_printk(APIC_QUIET, KERN_INFO - "...trying to set up timer as Virtual Wire IRQ...\n"); - - lapic_register_intr(0, vector); - apic_write(APIC_LVT0, APIC_DM_FIXED | vector); /* Fixed mode */ - enable_8259A_irq(0); - - if (timer_irq_works()) { - apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); - goto out; - } - disable_8259A_irq(0); - apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector); - apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n"); - - apic_printk(APIC_QUIET, KERN_INFO - "...trying to set up timer as ExtINT IRQ...\n"); - - init_8259A(0); - make_8259A_irq(0); - apic_write(APIC_LVT0, APIC_DM_EXTINT); - - unlock_ExtINT_logic(); - - if (timer_irq_works()) { - apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); - goto out; - } - apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n"); - panic("IO-APIC + timer doesn't work! Boot with apic=debug and send a " - "report. Then try booting with the 'noapic' option.\n"); -out: - local_irq_restore(flags); -} - -/* - * Traditionally ISA IRQ2 is the cascade IRQ, and is not available - * to devices. However there may be an I/O APIC pin available for - * this interrupt regardless. The pin may be left unconnected, but - * typically it will be reused as an ExtINT cascade interrupt for - * the master 8259A. In the MPS case such a pin will normally be - * reported as an ExtINT interrupt in the MP table. With ACPI - * there is no provision for ExtINT interrupts, and in the absence - * of an override it would be treated as an ordinary ISA I/O APIC - * interrupt, that is edge-triggered and unmasked by default. We - * used to do this, but it caused problems on some systems because - * of the NMI watchdog and sometimes IRQ0 of the 8254 timer using - * the same ExtINT cascade interrupt to drive the local APIC of the - * bootstrap processor. Therefore we refrain from routing IRQ2 to - * the I/O APIC in all cases now. No actual device should request - * it anyway. --macro - */ -#define PIC_IRQS (1 << PIC_CASCADE_IR) - -void __init setup_IO_APIC(void) -{ - int i; - - /* Reserve all the system vectors. */ - for (i = first_system_vector; i < NR_VECTORS; i++) - set_bit(i, used_vectors); - - enable_IO_APIC(); - - io_apic_irqs = ~PIC_IRQS; - - printk("ENABLING IO-APIC IRQs\n"); - - /* - * Set up IO-APIC IRQ routing. - */ - if (!acpi_ioapic) - setup_ioapic_ids_from_mpc(); - sync_Arb_IDs(); - setup_IO_APIC_irqs(); - init_IO_APIC_traps(); - check_timer(); -} - -/* - * Called after all the initialization is done. If we didnt find any - * APIC bugs then we can allow the modify fast path - */ - -static int __init io_apic_bug_finalize(void) -{ - if (sis_apic_bug == -1) - sis_apic_bug = 0; - return 0; -} - -late_initcall(io_apic_bug_finalize); - -struct sysfs_ioapic_data { - struct sys_device dev; - struct IO_APIC_route_entry entry[0]; -}; -static struct sysfs_ioapic_data *mp_ioapic_data[MAX_IO_APICS]; - -static int ioapic_suspend(struct sys_device *dev, pm_message_t state) -{ - struct IO_APIC_route_entry *entry; - struct sysfs_ioapic_data *data; - int i; - - data = container_of(dev, struct sysfs_ioapic_data, dev); - entry = data->entry; - for (i = 0; i < nr_ioapic_registers[dev->id]; i++) - entry[i] = ioapic_read_entry(dev->id, i); - - return 0; -} - -static int ioapic_resume(struct sys_device *dev) -{ - struct IO_APIC_route_entry *entry; - struct sysfs_ioapic_data *data; - unsigned long flags; - union IO_APIC_reg_00 reg_00; - int i; - - data = container_of(dev, struct sysfs_ioapic_data, dev); - entry = data->entry; - - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(dev->id, 0); - if (reg_00.bits.ID != mp_ioapics[dev->id].mp_apicid) { - reg_00.bits.ID = mp_ioapics[dev->id].mp_apicid; - io_apic_write(dev->id, 0, reg_00.raw); - } - spin_unlock_irqrestore(&ioapic_lock, flags); - for (i = 0; i < nr_ioapic_registers[dev->id]; i++) - ioapic_write_entry(dev->id, i, entry[i]); - - return 0; -} - -static struct sysdev_class ioapic_sysdev_class = { - .name = "ioapic", - .suspend = ioapic_suspend, - .resume = ioapic_resume, -}; - -static int __init ioapic_init_sysfs(void) -{ - struct sys_device *dev; - int i, size, error = 0; - - error = sysdev_class_register(&ioapic_sysdev_class); - if (error) - return error; - - for (i = 0; i < nr_ioapics; i++) { - size = sizeof(struct sys_device) + nr_ioapic_registers[i] - * sizeof(struct IO_APIC_route_entry); - mp_ioapic_data[i] = kzalloc(size, GFP_KERNEL); - if (!mp_ioapic_data[i]) { - printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i); - continue; - } - dev = &mp_ioapic_data[i]->dev; - dev->id = i; - dev->cls = &ioapic_sysdev_class; - error = sysdev_register(dev); - if (error) { - kfree(mp_ioapic_data[i]); - mp_ioapic_data[i] = NULL; - printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i); - continue; - } - } - - return 0; -} - -device_initcall(ioapic_init_sysfs); - -/* - * Dynamic irq allocate and deallocation - */ -int create_irq(void) -{ - /* Allocate an unused irq */ - int irq, new, vector = 0; - unsigned long flags; - - irq = -ENOSPC; - spin_lock_irqsave(&vector_lock, flags); - for (new = (NR_IRQS - 1); new >= 0; new--) { - if (platform_legacy_irq(new)) - continue; - if (irq_vector[new] != 0) - continue; - vector = __assign_irq_vector(new); - if (likely(vector > 0)) - irq = new; - break; - } - spin_unlock_irqrestore(&vector_lock, flags); - - if (irq >= 0) { - set_intr_gate(vector, interrupt[irq]); - dynamic_irq_init(irq); - } - return irq; -} - -void destroy_irq(unsigned int irq) -{ - unsigned long flags; - - dynamic_irq_cleanup(irq); - - spin_lock_irqsave(&vector_lock, flags); - clear_bit(irq_vector[irq], used_vectors); - irq_vector[irq] = 0; - spin_unlock_irqrestore(&vector_lock, flags); -} - -/* - * MSI message composition - */ -#ifdef CONFIG_PCI_MSI -static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg) -{ - int vector; - unsigned dest; - - vector = assign_irq_vector(irq); - if (vector >= 0) { - dest = cpu_mask_to_apicid(TARGET_CPUS); - - msg->address_hi = MSI_ADDR_BASE_HI; - msg->address_lo = - MSI_ADDR_BASE_LO | - ((INT_DEST_MODE == 0) ? -MSI_ADDR_DEST_MODE_PHYSICAL: - MSI_ADDR_DEST_MODE_LOGICAL) | - ((INT_DELIVERY_MODE != dest_LowestPrio) ? - MSI_ADDR_REDIRECTION_CPU: - MSI_ADDR_REDIRECTION_LOWPRI) | - MSI_ADDR_DEST_ID(dest); - - msg->data = - MSI_DATA_TRIGGER_EDGE | - MSI_DATA_LEVEL_ASSERT | - ((INT_DELIVERY_MODE != dest_LowestPrio) ? -MSI_DATA_DELIVERY_FIXED: - MSI_DATA_DELIVERY_LOWPRI) | - MSI_DATA_VECTOR(vector); - } - return vector; -} - -#ifdef CONFIG_SMP -static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) -{ - struct msi_msg msg; - unsigned int dest; - cpumask_t tmp; - int vector; - - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) - tmp = TARGET_CPUS; - - vector = assign_irq_vector(irq); - if (vector < 0) - return; - - dest = cpu_mask_to_apicid(mask); - - read_msi_msg(irq, &msg); - - msg.data &= ~MSI_DATA_VECTOR_MASK; - msg.data |= MSI_DATA_VECTOR(vector); - msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; - msg.address_lo |= MSI_ADDR_DEST_ID(dest); - - write_msi_msg(irq, &msg); - irq_desc[irq].affinity = mask; -} -#endif /* CONFIG_SMP */ - -/* - * IRQ Chip for MSI PCI/PCI-X/PCI-Express Devices, - * which implement the MSI or MSI-X Capability Structure. - */ -static struct irq_chip msi_chip = { - .name = "PCI-MSI", - .unmask = unmask_msi_irq, - .mask = mask_msi_irq, - .ack = ack_ioapic_irq, -#ifdef CONFIG_SMP - .set_affinity = set_msi_irq_affinity, -#endif - .retrigger = ioapic_retrigger_irq, -}; - -int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc) -{ - struct msi_msg msg; - int irq, ret; - irq = create_irq(); - if (irq < 0) - return irq; - - ret = msi_compose_msg(dev, irq, &msg); - if (ret < 0) { - destroy_irq(irq); - return ret; - } - - set_irq_msi(irq, desc); - write_msi_msg(irq, &msg); - - set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, - "edge"); - - return 0; -} - -void arch_teardown_msi_irq(unsigned int irq) -{ - destroy_irq(irq); -} - -#endif /* CONFIG_PCI_MSI */ - -/* - * Hypertransport interrupt support - */ -#ifdef CONFIG_HT_IRQ - -#ifdef CONFIG_SMP - -static void target_ht_irq(unsigned int irq, unsigned int dest) -{ - struct ht_irq_msg msg; - fetch_ht_irq_msg(irq, &msg); - - msg.address_lo &= ~(HT_IRQ_LOW_DEST_ID_MASK); - msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); - - msg.address_lo |= HT_IRQ_LOW_DEST_ID(dest); - msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); - - write_ht_irq_msg(irq, &msg); -} - -static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) -{ - unsigned int dest; - cpumask_t tmp; - - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) - tmp = TARGET_CPUS; - - cpus_and(mask, tmp, CPU_MASK_ALL); - - dest = cpu_mask_to_apicid(mask); - - target_ht_irq(irq, dest); - irq_desc[irq].affinity = mask; -} -#endif - -static struct irq_chip ht_irq_chip = { - .name = "PCI-HT", - .mask = mask_ht_irq, - .unmask = unmask_ht_irq, - .ack = ack_ioapic_irq, -#ifdef CONFIG_SMP - .set_affinity = set_ht_irq_affinity, -#endif - .retrigger = ioapic_retrigger_irq, -}; - -int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) -{ - int vector; - - vector = assign_irq_vector(irq); - if (vector >= 0) { - struct ht_irq_msg msg; - unsigned dest; - cpumask_t tmp; - - cpus_clear(tmp); - cpu_set(vector >> 8, tmp); - dest = cpu_mask_to_apicid(tmp); - - msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); - - msg.address_lo = - HT_IRQ_LOW_BASE | - HT_IRQ_LOW_DEST_ID(dest) | - HT_IRQ_LOW_VECTOR(vector) | - ((INT_DEST_MODE == 0) ? - HT_IRQ_LOW_DM_PHYSICAL : - HT_IRQ_LOW_DM_LOGICAL) | - HT_IRQ_LOW_RQEOI_EDGE | - ((INT_DELIVERY_MODE != dest_LowestPrio) ? - HT_IRQ_LOW_MT_FIXED : - HT_IRQ_LOW_MT_ARBITRATED) | - HT_IRQ_LOW_IRQ_MASKED; - - write_ht_irq_msg(irq, &msg); - - set_irq_chip_and_handler_name(irq, &ht_irq_chip, - handle_edge_irq, "edge"); - } - return vector; -} -#endif /* CONFIG_HT_IRQ */ - -/* -------------------------------------------------------------------------- - ACPI-based IOAPIC Configuration - -------------------------------------------------------------------------- */ - -#ifdef CONFIG_ACPI - -int __init io_apic_get_unique_id(int ioapic, int apic_id) -{ - union IO_APIC_reg_00 reg_00; - static physid_mask_t apic_id_map = PHYSID_MASK_NONE; - physid_mask_t tmp; - unsigned long flags; - int i = 0; - - /* - * The P4 platform supports up to 256 APIC IDs on two separate APIC - * buses (one for LAPICs, one for IOAPICs), where predecessors only - * supports up to 16 on one shared APIC bus. - * - * TBD: Expand LAPIC/IOAPIC support on P4-class systems to take full - * advantage of new APIC bus architecture. - */ - - if (physids_empty(apic_id_map)) - apic_id_map = ioapic_phys_id_map(phys_cpu_present_map); - - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(ioapic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); - - if (apic_id >= get_physical_broadcast()) { - printk(KERN_WARNING "IOAPIC[%d]: Invalid apic_id %d, trying " - "%d\n", ioapic, apic_id, reg_00.bits.ID); - apic_id = reg_00.bits.ID; - } - - /* - * Every APIC in a system must have a unique ID or we get lots of nice - * 'stuck on smp_invalidate_needed IPI wait' messages. - */ - if (check_apicid_used(apic_id_map, apic_id)) { - - for (i = 0; i < get_physical_broadcast(); i++) { - if (!check_apicid_used(apic_id_map, i)) - break; - } - - if (i == get_physical_broadcast()) - panic("Max apic_id exceeded!\n"); - - printk(KERN_WARNING "IOAPIC[%d]: apic_id %d already used, " - "trying %d\n", ioapic, apic_id, i); - - apic_id = i; - } - - tmp = apicid_to_cpu_present(apic_id); - physids_or(apic_id_map, apic_id_map, tmp); - - if (reg_00.bits.ID != apic_id) { - reg_00.bits.ID = apic_id; - - spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(ioapic, 0, reg_00.raw); - reg_00.raw = io_apic_read(ioapic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); - - /* Sanity check */ - if (reg_00.bits.ID != apic_id) { - printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic); - return -1; - } - } - - apic_printk(APIC_VERBOSE, KERN_INFO - "IOAPIC[%d]: Assigned apic_id %d\n", ioapic, apic_id); - - return apic_id; -} - - -int __init io_apic_get_version(int ioapic) -{ - union IO_APIC_reg_01 reg_01; - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - reg_01.raw = io_apic_read(ioapic, 1); - spin_unlock_irqrestore(&ioapic_lock, flags); - - return reg_01.bits.version; -} - - -int __init io_apic_get_redir_entries(int ioapic) -{ - union IO_APIC_reg_01 reg_01; - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - reg_01.raw = io_apic_read(ioapic, 1); - spin_unlock_irqrestore(&ioapic_lock, flags); - - return reg_01.bits.entries; -} - - -int io_apic_set_pci_routing(int ioapic, int pin, int irq, int edge_level, int active_high_low) -{ - struct IO_APIC_route_entry entry; - - if (!IO_APIC_IRQ(irq)) { - printk(KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n", - ioapic); - return -EINVAL; - } - - /* - * Generate a PCI IRQ routing entry and program the IOAPIC accordingly. - * Note that we mask (disable) IRQs now -- these get enabled when the - * corresponding device driver registers for this IRQ. - */ - - memset(&entry, 0, sizeof(entry)); - - entry.delivery_mode = INT_DELIVERY_MODE; - entry.dest_mode = INT_DEST_MODE; - entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS); - entry.trigger = edge_level; - entry.polarity = active_high_low; - entry.mask = 1; - - /* - * IRQs < 16 are already in the irq_2_pin[] map - */ - if (irq >= 16) - add_pin_to_irq(irq, ioapic, pin); - - entry.vector = assign_irq_vector(irq); - - apic_printk(APIC_DEBUG, KERN_DEBUG "IOAPIC[%d]: Set PCI routing entry " - "(%d-%d -> 0x%x -> IRQ %d Mode:%i Active:%i)\n", ioapic, - mp_ioapics[ioapic].mp_apicid, pin, entry.vector, irq, - edge_level, active_high_low); - - ioapic_register_intr(irq, entry.vector, edge_level); - - if (!ioapic && (irq < 16)) - disable_8259A_irq(irq); - - ioapic_write_entry(ioapic, pin, entry); - - return 0; -} - -int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) -{ - int i; - - if (skip_ioapic_setup) - return -1; - - for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mp_irqtype == mp_INT && - mp_irqs[i].mp_srcbusirq == bus_irq) - break; - if (i >= mp_irq_entries) - return -1; - - *trigger = irq_trigger(i); - *polarity = irq_polarity(i); - return 0; -} - -#endif /* CONFIG_ACPI */ - -static int __init parse_disable_timer_pin_1(char *arg) -{ - disable_timer_pin_1 = 1; - return 0; -} -early_param("disable_timer_pin_1", parse_disable_timer_pin_1); - -static int __init parse_enable_timer_pin_1(char *arg) -{ - disable_timer_pin_1 = -1; - return 0; -} -early_param("enable_timer_pin_1", parse_enable_timer_pin_1); - -static int __init parse_noapic(char *arg) -{ - /* disable IO-APIC */ - disable_ioapic_setup(); - return 0; -} -early_param("noapic", parse_noapic); - -void __init ioapic_init_mappings(void) -{ - unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0; - int i; - - for (i = 0; i < nr_ioapics; i++) { - if (smp_found_config) { - ioapic_phys = mp_ioapics[i].mp_apicaddr; - if (!ioapic_phys) { - printk(KERN_ERR - "WARNING: bogus zero IO-APIC " - "address found in MPTABLE, " - "disabling IO/APIC support!\n"); - smp_found_config = 0; - skip_ioapic_setup = 1; - goto fake_ioapic_page; - } - } else { -fake_ioapic_page: - ioapic_phys = (unsigned long) - alloc_bootmem_pages(PAGE_SIZE); - ioapic_phys = __pa(ioapic_phys); - } - set_fixmap_nocache(idx, ioapic_phys); - printk(KERN_DEBUG "mapped IOAPIC to %08lx (%08lx)\n", - __fix_to_virt(idx), ioapic_phys); - idx++; - } -} - diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c new file mode 100644 index 000000000000..3f1d9d18df67 --- /dev/null +++ b/arch/x86/kernel/irq.c @@ -0,0 +1,192 @@ +/* + * Common interrupt code for 32 and 64 bit + */ +#include <linux/cpu.h> +#include <linux/interrupt.h> +#include <linux/kernel_stat.h> +#include <linux/seq_file.h> + +#include <asm/apic.h> +#include <asm/io_apic.h> +#include <asm/smp.h> + +atomic_t irq_err_count; + +/* + * 'what should we do if we get a hw irq event on an illegal vector'. + * each architecture has to answer this themselves. + */ +void ack_bad_irq(unsigned int irq) +{ + printk(KERN_ERR "unexpected IRQ trap at vector %02x\n", irq); + +#ifdef CONFIG_X86_LOCAL_APIC + /* + * Currently unexpected vectors happen only on SMP and APIC. + * We _must_ ack these because every local APIC has only N + * irq slots per priority level, and a 'hanging, unacked' IRQ + * holds up an irq slot - in excessive cases (when multiple + * unexpected vectors occur) that might lock up the APIC + * completely. + * But only ack when the APIC is enabled -AK + */ + if (cpu_has_apic) + ack_APIC_irq(); +#endif +} + +#ifdef CONFIG_X86_32 +# define irq_stats(x) (&per_cpu(irq_stat, x)) +#else +# define irq_stats(x) cpu_pda(x) +#endif +/* + * /proc/interrupts printing: + */ +static int show_other_interrupts(struct seq_file *p) +{ + int j; + + seq_printf(p, "NMI: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->__nmi_count); + seq_printf(p, " Non-maskable interrupts\n"); +#ifdef CONFIG_X86_LOCAL_APIC + seq_printf(p, "LOC: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->apic_timer_irqs); + seq_printf(p, " Local timer interrupts\n"); +#endif +#ifdef CONFIG_SMP + seq_printf(p, "RES: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_resched_count); + seq_printf(p, " Rescheduling interrupts\n"); + seq_printf(p, "CAL: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_call_count); + seq_printf(p, " Function call interrupts\n"); + seq_printf(p, "TLB: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_tlb_count); + seq_printf(p, " TLB shootdowns\n"); +#endif +#ifdef CONFIG_X86_MCE + seq_printf(p, "TRM: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_thermal_count); + seq_printf(p, " Thermal event interrupts\n"); +# ifdef CONFIG_X86_64 + seq_printf(p, "THR: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_threshold_count); + seq_printf(p, " Threshold APIC interrupts\n"); +# endif +#endif +#ifdef CONFIG_X86_LOCAL_APIC + seq_printf(p, "SPU: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_spurious_count); + seq_printf(p, " Spurious interrupts\n"); +#endif + seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); +#if defined(CONFIG_X86_IO_APIC) + seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count)); +#endif + return 0; +} + +int show_interrupts(struct seq_file *p, void *v) +{ + unsigned long flags, any_count = 0; + int i = *(loff_t *) v, j; + struct irqaction *action; + struct irq_desc *desc; + + if (i > nr_irqs) + return 0; + + if (i == nr_irqs) + return show_other_interrupts(p); + + /* print header */ + if (i == 0) { + seq_printf(p, " "); + for_each_online_cpu(j) + seq_printf(p, "CPU%-8d", j); + seq_putc(p, '\n'); + } + + desc = irq_to_desc(i); + if (!desc) + return 0; + + spin_lock_irqsave(&desc->lock, flags); +#ifndef CONFIG_SMP + any_count = kstat_irqs(i); +#else + for_each_online_cpu(j) + any_count |= kstat_irqs_cpu(i, j); +#endif + action = desc->action; + if (!action && !any_count) + goto out; + + seq_printf(p, "%3d: ", i); +#ifndef CONFIG_SMP + seq_printf(p, "%10u ", kstat_irqs(i)); +#else + for_each_online_cpu(j) + seq_printf(p, "%10u ", kstat_irqs_cpu(i, j)); +#endif + seq_printf(p, " %8s", desc->chip->name); + seq_printf(p, "-%-8s", desc->name); + + if (action) { + seq_printf(p, " %s", action->name); + while ((action = action->next) != NULL) + seq_printf(p, ", %s", action->name); + } + + seq_putc(p, '\n'); +out: + spin_unlock_irqrestore(&desc->lock, flags); + return 0; +} + +/* + * /proc/stat helpers + */ +u64 arch_irq_stat_cpu(unsigned int cpu) +{ + u64 sum = irq_stats(cpu)->__nmi_count; + +#ifdef CONFIG_X86_LOCAL_APIC + sum += irq_stats(cpu)->apic_timer_irqs; +#endif +#ifdef CONFIG_SMP + sum += irq_stats(cpu)->irq_resched_count; + sum += irq_stats(cpu)->irq_call_count; + sum += irq_stats(cpu)->irq_tlb_count; +#endif +#ifdef CONFIG_X86_MCE + sum += irq_stats(cpu)->irq_thermal_count; +# ifdef CONFIG_X86_64 + sum += irq_stats(cpu)->irq_threshold_count; +#endif +#endif +#ifdef CONFIG_X86_LOCAL_APIC + sum += irq_stats(cpu)->irq_spurious_count; +#endif + return sum; +} + +u64 arch_irq_stat(void) +{ + u64 sum = atomic_read(&irq_err_count); + +#ifdef CONFIG_X86_IO_APIC + sum += atomic_read(&irq_mis_count); +#endif + return sum; +} diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index b71e02d42f4f..119fc9c8ff7f 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -25,29 +25,6 @@ EXPORT_PER_CPU_SYMBOL(irq_stat); DEFINE_PER_CPU(struct pt_regs *, irq_regs); EXPORT_PER_CPU_SYMBOL(irq_regs); -/* - * 'what should we do if we get a hw irq event on an illegal vector'. - * each architecture has to answer this themselves. - */ -void ack_bad_irq(unsigned int irq) -{ - printk(KERN_ERR "unexpected IRQ trap at vector %02x\n", irq); - -#ifdef CONFIG_X86_LOCAL_APIC - /* - * Currently unexpected vectors happen only on SMP and APIC. - * We _must_ ack these because every local APIC has only N - * irq slots per priority level, and a 'hanging, unacked' IRQ - * holds up an irq slot - in excessive cases (when multiple - * unexpected vectors occur) that might lock up the APIC - * completely. - * But only ack when the APIC is enabled -AK - */ - if (cpu_has_apic) - ack_APIC_irq(); -#endif -} - #ifdef CONFIG_DEBUG_STACKOVERFLOW /* Debugging check for stack overflow: is there less than 1KB free? */ static int check_stack_overflow(void) @@ -223,20 +200,25 @@ unsigned int do_IRQ(struct pt_regs *regs) { struct pt_regs *old_regs; /* high bit used in ret_from_ code */ - int overflow, irq = ~regs->orig_ax; - struct irq_desc *desc = irq_desc + irq; + int overflow; + unsigned vector = ~regs->orig_ax; + struct irq_desc *desc; + unsigned irq; - if (unlikely((unsigned)irq >= NR_IRQS)) { - printk(KERN_EMERG "%s: cannot handle IRQ %d\n", - __func__, irq); - BUG(); - } old_regs = set_irq_regs(regs); irq_enter(); + irq = __get_cpu_var(vector_irq)[vector]; overflow = check_stack_overflow(); + desc = irq_to_desc(irq); + if (unlikely(!desc)) { + printk(KERN_EMERG "%s: cannot handle IRQ %d vector %#x cpu %d\n", + __func__, irq, vector, smp_processor_id()); + BUG(); + } + if (!execute_on_irq_stack(overflow, desc, irq)) { if (unlikely(overflow)) print_stack_overflow(); @@ -248,146 +230,6 @@ unsigned int do_IRQ(struct pt_regs *regs) return 1; } -/* - * Interrupt statistics: - */ - -atomic_t irq_err_count; - -/* - * /proc/interrupts printing: - */ - -int show_interrupts(struct seq_file *p, void *v) -{ - int i = *(loff_t *) v, j; - struct irqaction * action; - unsigned long flags; - - if (i == 0) { - seq_printf(p, " "); - for_each_online_cpu(j) - seq_printf(p, "CPU%-8d",j); - seq_putc(p, '\n'); - } - - if (i < NR_IRQS) { - unsigned any_count = 0; - - spin_lock_irqsave(&irq_desc[i].lock, flags); -#ifndef CONFIG_SMP - any_count = kstat_irqs(i); -#else - for_each_online_cpu(j) - any_count |= kstat_cpu(j).irqs[i]; -#endif - action = irq_desc[i].action; - if (!action && !any_count) - goto skip; - seq_printf(p, "%3d: ",i); -#ifndef CONFIG_SMP - seq_printf(p, "%10u ", kstat_irqs(i)); -#else - for_each_online_cpu(j) - seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); -#endif - seq_printf(p, " %8s", irq_desc[i].chip->name); - seq_printf(p, "-%-8s", irq_desc[i].name); - - if (action) { - seq_printf(p, " %s", action->name); - while ((action = action->next) != NULL) - seq_printf(p, ", %s", action->name); - } - - seq_putc(p, '\n'); -skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); - } else if (i == NR_IRQS) { - seq_printf(p, "NMI: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", nmi_count(j)); - seq_printf(p, " Non-maskable interrupts\n"); -#ifdef CONFIG_X86_LOCAL_APIC - seq_printf(p, "LOC: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).apic_timer_irqs); - seq_printf(p, " Local timer interrupts\n"); -#endif -#ifdef CONFIG_SMP - seq_printf(p, "RES: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_resched_count); - seq_printf(p, " Rescheduling interrupts\n"); - seq_printf(p, "CAL: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_call_count); - seq_printf(p, " Function call interrupts\n"); - seq_printf(p, "TLB: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_tlb_count); - seq_printf(p, " TLB shootdowns\n"); -#endif -#ifdef CONFIG_X86_MCE - seq_printf(p, "TRM: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_thermal_count); - seq_printf(p, " Thermal event interrupts\n"); -#endif -#ifdef CONFIG_X86_LOCAL_APIC - seq_printf(p, "SPU: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_spurious_count); - seq_printf(p, " Spurious interrupts\n"); -#endif - seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); -#if defined(CONFIG_X86_IO_APIC) - seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count)); -#endif - } - return 0; -} - -/* - * /proc/stat helpers - */ -u64 arch_irq_stat_cpu(unsigned int cpu) -{ - u64 sum = nmi_count(cpu); - -#ifdef CONFIG_X86_LOCAL_APIC - sum += per_cpu(irq_stat, cpu).apic_timer_irqs; -#endif -#ifdef CONFIG_SMP - sum += per_cpu(irq_stat, cpu).irq_resched_count; - sum += per_cpu(irq_stat, cpu).irq_call_count; - sum += per_cpu(irq_stat, cpu).irq_tlb_count; -#endif -#ifdef CONFIG_X86_MCE - sum += per_cpu(irq_stat, cpu).irq_thermal_count; -#endif -#ifdef CONFIG_X86_LOCAL_APIC - sum += per_cpu(irq_stat, cpu).irq_spurious_count; -#endif - return sum; -} - -u64 arch_irq_stat(void) -{ - u64 sum = atomic_read(&irq_err_count); - -#ifdef CONFIG_X86_IO_APIC - sum += atomic_read(&irq_mis_count); -#endif - return sum; -} - #ifdef CONFIG_HOTPLUG_CPU #include <mach_apic.h> @@ -395,20 +237,24 @@ void fixup_irqs(cpumask_t map) { unsigned int irq; static int warned; + struct irq_desc *desc; - for (irq = 0; irq < NR_IRQS; irq++) { + for_each_irq_desc(irq, desc) { cpumask_t mask; + + if (!desc) + continue; if (irq == 2) continue; - cpus_and(mask, irq_desc[irq].affinity, map); + cpus_and(mask, desc->affinity, map); if (any_online_cpu(mask) == NR_CPUS) { printk("Breaking affinity for irq %i\n", irq); mask = map; } - if (irq_desc[irq].chip->set_affinity) - irq_desc[irq].chip->set_affinity(irq, mask); - else if (irq_desc[irq].action && !(warned++)) + if (desc->chip->set_affinity) + desc->chip->set_affinity(irq, mask); + else if (desc->action && !(warned++)) printk("Cannot set affinity for irq %i\n", irq); } diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c index f065fe9071b9..a174a217eb1a 100644 --- a/arch/x86/kernel/irq_64.c +++ b/arch/x86/kernel/irq_64.c @@ -13,34 +13,12 @@ #include <linux/seq_file.h> #include <linux/module.h> #include <linux/delay.h> +#include <linux/ftrace.h> #include <asm/uaccess.h> #include <asm/io_apic.h> #include <asm/idle.h> #include <asm/smp.h> -atomic_t irq_err_count; - -/* - * 'what should we do if we get a hw irq event on an illegal vector'. - * each architecture has to answer this themselves. - */ -void ack_bad_irq(unsigned int irq) -{ - printk(KERN_WARNING "unexpected IRQ trap at vector %02x\n", irq); - /* - * Currently unexpected vectors happen only on SMP and APIC. - * We _must_ ack these because every local APIC has only N - * irq slots per priority level, and a 'hanging, unacked' IRQ - * holds up an irq slot - in excessive cases (when multiple - * unexpected vectors occur) that might lock up the APIC - * completely. - * But don't ack when the APIC is disabled. -AK - */ - if (!disable_apic) - ack_APIC_irq(); -} - -#ifdef CONFIG_DEBUG_STACKOVERFLOW /* * Probabilistic stack overflow check: * @@ -50,134 +28,17 @@ void ack_bad_irq(unsigned int irq) */ static inline void stack_overflow_check(struct pt_regs *regs) { +#ifdef CONFIG_DEBUG_STACKOVERFLOW u64 curbase = (u64)task_stack_page(current); - static unsigned long warned = -60*HZ; - - if (regs->sp >= curbase && regs->sp <= curbase + THREAD_SIZE && - regs->sp < curbase + sizeof(struct thread_info) + 128 && - time_after(jiffies, warned + 60*HZ)) { - printk("do_IRQ: %s near stack overflow (cur:%Lx,sp:%lx)\n", - current->comm, curbase, regs->sp); - show_stack(NULL,NULL); - warned = jiffies; - } -} -#endif - -/* - * Generic, controller-independent functions: - */ -int show_interrupts(struct seq_file *p, void *v) -{ - int i = *(loff_t *) v, j; - struct irqaction * action; - unsigned long flags; - - if (i == 0) { - seq_printf(p, " "); - for_each_online_cpu(j) - seq_printf(p, "CPU%-8d",j); - seq_putc(p, '\n'); - } + WARN_ONCE(regs->sp >= curbase && + regs->sp <= curbase + THREAD_SIZE && + regs->sp < curbase + sizeof(struct thread_info) + + sizeof(struct pt_regs) + 128, - if (i < NR_IRQS) { - unsigned any_count = 0; - - spin_lock_irqsave(&irq_desc[i].lock, flags); -#ifndef CONFIG_SMP - any_count = kstat_irqs(i); -#else - for_each_online_cpu(j) - any_count |= kstat_cpu(j).irqs[i]; -#endif - action = irq_desc[i].action; - if (!action && !any_count) - goto skip; - seq_printf(p, "%3d: ",i); -#ifndef CONFIG_SMP - seq_printf(p, "%10u ", kstat_irqs(i)); -#else - for_each_online_cpu(j) - seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); + "do_IRQ: %s near stack overflow (cur:%Lx,sp:%lx)\n", + current->comm, curbase, regs->sp); #endif - seq_printf(p, " %8s", irq_desc[i].chip->name); - seq_printf(p, "-%-8s", irq_desc[i].name); - - if (action) { - seq_printf(p, " %s", action->name); - while ((action = action->next) != NULL) - seq_printf(p, ", %s", action->name); - } - seq_putc(p, '\n'); -skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); - } else if (i == NR_IRQS) { - seq_printf(p, "NMI: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count); - seq_printf(p, " Non-maskable interrupts\n"); - seq_printf(p, "LOC: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs); - seq_printf(p, " Local timer interrupts\n"); -#ifdef CONFIG_SMP - seq_printf(p, "RES: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_resched_count); - seq_printf(p, " Rescheduling interrupts\n"); - seq_printf(p, "CAL: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_call_count); - seq_printf(p, " Function call interrupts\n"); - seq_printf(p, "TLB: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_tlb_count); - seq_printf(p, " TLB shootdowns\n"); -#endif -#ifdef CONFIG_X86_MCE - seq_printf(p, "TRM: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_thermal_count); - seq_printf(p, " Thermal event interrupts\n"); - seq_printf(p, "THR: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_threshold_count); - seq_printf(p, " Threshold APIC interrupts\n"); -#endif - seq_printf(p, "SPU: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_spurious_count); - seq_printf(p, " Spurious interrupts\n"); - seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); - } - return 0; -} - -/* - * /proc/stat helpers - */ -u64 arch_irq_stat_cpu(unsigned int cpu) -{ - u64 sum = cpu_pda(cpu)->__nmi_count; - - sum += cpu_pda(cpu)->apic_timer_irqs; -#ifdef CONFIG_SMP - sum += cpu_pda(cpu)->irq_resched_count; - sum += cpu_pda(cpu)->irq_call_count; - sum += cpu_pda(cpu)->irq_tlb_count; -#endif -#ifdef CONFIG_X86_MCE - sum += cpu_pda(cpu)->irq_thermal_count; - sum += cpu_pda(cpu)->irq_threshold_count; -#endif - sum += cpu_pda(cpu)->irq_spurious_count; - return sum; -} - -u64 arch_irq_stat(void) -{ - return atomic_read(&irq_err_count); } /* @@ -185,9 +46,10 @@ u64 arch_irq_stat(void) * SMP cross-CPU interrupts have their own specific * handlers). */ -asmlinkage unsigned int do_IRQ(struct pt_regs *regs) +asmlinkage unsigned int __irq_entry do_IRQ(struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); + struct irq_desc *desc; /* high bit used in ret_from_ code */ unsigned vector = ~regs->orig_ax; @@ -197,12 +59,11 @@ asmlinkage unsigned int do_IRQ(struct pt_regs *regs) irq_enter(); irq = __get_cpu_var(vector_irq)[vector]; -#ifdef CONFIG_DEBUG_STACKOVERFLOW stack_overflow_check(regs); -#endif - if (likely(irq < NR_IRQS)) - generic_handle_irq(irq); + desc = irq_to_desc(irq); + if (likely(desc)) + generic_handle_irq_desc(irq, desc); else { if (!disable_apic) ack_APIC_irq(); @@ -223,42 +84,45 @@ void fixup_irqs(cpumask_t map) { unsigned int irq; static int warned; + struct irq_desc *desc; - for (irq = 0; irq < NR_IRQS; irq++) { + for_each_irq_desc(irq, desc) { cpumask_t mask; int break_affinity = 0; int set_affinity = 1; + if (!desc) + continue; if (irq == 2) continue; /* interrupt's are disabled at this point */ - spin_lock(&irq_desc[irq].lock); + spin_lock(&desc->lock); if (!irq_has_action(irq) || - cpus_equal(irq_desc[irq].affinity, map)) { - spin_unlock(&irq_desc[irq].lock); + cpus_equal(desc->affinity, map)) { + spin_unlock(&desc->lock); continue; } - cpus_and(mask, irq_desc[irq].affinity, map); + cpus_and(mask, desc->affinity, map); if (cpus_empty(mask)) { break_affinity = 1; mask = map; } - if (irq_desc[irq].chip->mask) - irq_desc[irq].chip->mask(irq); + if (desc->chip->mask) + desc->chip->mask(irq); - if (irq_desc[irq].chip->set_affinity) - irq_desc[irq].chip->set_affinity(irq, mask); + if (desc->chip->set_affinity) + desc->chip->set_affinity(irq, mask); else if (!(warned++)) set_affinity = 0; - if (irq_desc[irq].chip->unmask) - irq_desc[irq].chip->unmask(irq); + if (desc->chip->unmask) + desc->chip->unmask(irq); - spin_unlock(&irq_desc[irq].lock); + spin_unlock(&desc->lock); if (break_affinity && set_affinity) printk("Broke affinity for irq %i\n", irq); diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c index 9200a1e2752d..203384ed2b5d 100644 --- a/arch/x86/kernel/irqinit_32.c +++ b/arch/x86/kernel/irqinit_32.c @@ -68,7 +68,13 @@ void __init init_ISA_irqs (void) /* * 16 old-style INTA-cycle interrupts: */ - for (i = 0; i < 16; i++) { + for (i = 0; i < NR_IRQS_LEGACY; i++) { + struct irq_desc *desc = irq_to_desc(i); + + desc->status = IRQ_DISABLED; + desc->action = NULL; + desc->depth = 1; + set_irq_chip_and_handler_name(i, &i8259A_chip, handle_level_irq, "XT"); } @@ -83,6 +89,27 @@ static struct irqaction irq2 = { .name = "cascade", }; +DEFINE_PER_CPU(vector_irq_t, vector_irq) = { + [0 ... IRQ0_VECTOR - 1] = -1, + [IRQ0_VECTOR] = 0, + [IRQ1_VECTOR] = 1, + [IRQ2_VECTOR] = 2, + [IRQ3_VECTOR] = 3, + [IRQ4_VECTOR] = 4, + [IRQ5_VECTOR] = 5, + [IRQ6_VECTOR] = 6, + [IRQ7_VECTOR] = 7, + [IRQ8_VECTOR] = 8, + [IRQ9_VECTOR] = 9, + [IRQ10_VECTOR] = 10, + [IRQ11_VECTOR] = 11, + [IRQ12_VECTOR] = 12, + [IRQ13_VECTOR] = 13, + [IRQ14_VECTOR] = 14, + [IRQ15_VECTOR] = 15, + [IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1 +}; + /* Overridden in paravirt.c */ void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ"))); @@ -98,22 +125,14 @@ void __init native_init_IRQ(void) * us. (some of these will be overridden and become * 'special' SMP interrupts) */ - for (i = 0; i < (NR_VECTORS - FIRST_EXTERNAL_VECTOR); i++) { - int vector = FIRST_EXTERNAL_VECTOR + i; - if (i >= NR_IRQS) - break; + for (i = FIRST_EXTERNAL_VECTOR; i < NR_VECTORS; i++) { /* SYSCALL_VECTOR was reserved in trap_init. */ - if (!test_bit(vector, used_vectors)) - set_intr_gate(vector, interrupt[i]); + if (i != SYSCALL_VECTOR) + set_intr_gate(i, interrupt[i-FIRST_EXTERNAL_VECTOR]); } -#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_SMP) - /* - * IRQ0 must be given a fixed assignment and initialized, - * because it's used before the IO-APIC is set up. - */ - set_intr_gate(FIRST_DEVICE_VECTOR, interrupt[0]); +#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_SMP) /* * The reschedule interrupt is a CPU-to-CPU reschedule-helper * IPI, driven by wakeup. @@ -128,6 +147,9 @@ void __init native_init_IRQ(void) /* IPI for single call function */ set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR, call_function_single_interrupt); + + /* Low priority IPI to cleanup after moving an irq */ + set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt); #endif #ifdef CONFIG_X86_LOCAL_APIC diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c index 5b5be9d43c2a..6190e6ef546c 100644 --- a/arch/x86/kernel/irqinit_64.c +++ b/arch/x86/kernel/irqinit_64.c @@ -24,41 +24,6 @@ #include <asm/i8259.h> /* - * Common place to define all x86 IRQ vectors - * - * This builds up the IRQ handler stubs using some ugly macros in irq.h - * - * These macros create the low-level assembly IRQ routines that save - * register context and call do_IRQ(). do_IRQ() then does all the - * operations that are needed to keep the AT (or SMP IOAPIC) - * interrupt-controller happy. - */ - -#define IRQ_NAME2(nr) nr##_interrupt(void) -#define IRQ_NAME(nr) IRQ_NAME2(IRQ##nr) - -/* - * SMP has a few special interrupts for IPI messages - */ - -#define BUILD_IRQ(nr) \ - asmlinkage void IRQ_NAME(nr); \ - asm("\n.text\n.p2align\n" \ - "IRQ" #nr "_interrupt:\n\t" \ - "push $~(" #nr ") ; " \ - "jmp common_interrupt\n" \ - ".previous"); - -#define BI(x,y) \ - BUILD_IRQ(x##y) - -#define BUILD_16_IRQS(x) \ - BI(x,0) BI(x,1) BI(x,2) BI(x,3) \ - BI(x,4) BI(x,5) BI(x,6) BI(x,7) \ - BI(x,8) BI(x,9) BI(x,a) BI(x,b) \ - BI(x,c) BI(x,d) BI(x,e) BI(x,f) - -/* * ISA PIC or low IO-APIC triggered (INTA-cycle or APIC) interrupts: * (these are usually mapped to vectors 0x30-0x3f) */ @@ -73,37 +38,6 @@ * * (these are usually mapped into the 0x30-0xff vector range) */ - BUILD_16_IRQS(0x2) BUILD_16_IRQS(0x3) -BUILD_16_IRQS(0x4) BUILD_16_IRQS(0x5) BUILD_16_IRQS(0x6) BUILD_16_IRQS(0x7) -BUILD_16_IRQS(0x8) BUILD_16_IRQS(0x9) BUILD_16_IRQS(0xa) BUILD_16_IRQS(0xb) -BUILD_16_IRQS(0xc) BUILD_16_IRQS(0xd) BUILD_16_IRQS(0xe) BUILD_16_IRQS(0xf) - -#undef BUILD_16_IRQS -#undef BI - - -#define IRQ(x,y) \ - IRQ##x##y##_interrupt - -#define IRQLIST_16(x) \ - IRQ(x,0), IRQ(x,1), IRQ(x,2), IRQ(x,3), \ - IRQ(x,4), IRQ(x,5), IRQ(x,6), IRQ(x,7), \ - IRQ(x,8), IRQ(x,9), IRQ(x,a), IRQ(x,b), \ - IRQ(x,c), IRQ(x,d), IRQ(x,e), IRQ(x,f) - -/* for the irq vectors */ -static void (*__initdata interrupt[NR_VECTORS - FIRST_EXTERNAL_VECTOR])(void) = { - IRQLIST_16(0x2), IRQLIST_16(0x3), - IRQLIST_16(0x4), IRQLIST_16(0x5), IRQLIST_16(0x6), IRQLIST_16(0x7), - IRQLIST_16(0x8), IRQLIST_16(0x9), IRQLIST_16(0xa), IRQLIST_16(0xb), - IRQLIST_16(0xc), IRQLIST_16(0xd), IRQLIST_16(0xe), IRQLIST_16(0xf) -}; - -#undef IRQ -#undef IRQLIST_16 - - - /* * IRQ2 is cascade interrupt to second interrupt controller @@ -142,23 +76,18 @@ void __init init_ISA_irqs(void) init_bsp_APIC(); init_8259A(0); - for (i = 0; i < NR_IRQS; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; + for (i = 0; i < NR_IRQS_LEGACY; i++) { + struct irq_desc *desc = irq_to_desc(i); + + desc->status = IRQ_DISABLED; + desc->action = NULL; + desc->depth = 1; - if (i < 16) { - /* - * 16 old-style INTA-cycle interrupts: - */ - set_irq_chip_and_handler_name(i, &i8259A_chip, + /* + * 16 old-style INTA-cycle interrupts: + */ + set_irq_chip_and_handler_name(i, &i8259A_chip, handle_level_irq, "XT"); - } else { - /* - * 'high' PCI IRQs filled in on demand - */ - irq_desc[i].chip = &no_irq_chip; - } } } diff --git a/arch/x86/kernel/k8.c b/arch/x86/kernel/k8.c index 304d8bad6559..cbc4332a77b2 100644 --- a/arch/x86/kernel/k8.c +++ b/arch/x86/kernel/k8.c @@ -18,7 +18,6 @@ static u32 *flush_words; struct pci_device_id k8_nb_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) }, - { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_11H_NB_MISC) }, {} }; EXPORT_SYMBOL(k8_nb_ids); diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index d02def06ca91..e169ae9b6a62 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -78,6 +78,34 @@ static cycle_t kvm_clock_read(void) return ret; } +/* + * If we don't do that, there is the possibility that the guest + * will calibrate under heavy load - thus, getting a lower lpj - + * and execute the delays themselves without load. This is wrong, + * because no delay loop can finish beforehand. + * Any heuristics is subject to fail, because ultimately, a large + * poll of guests can be running and trouble each other. So we preset + * lpj here + */ +static unsigned long kvm_get_tsc_khz(void) +{ + return preset_lpj; +} + +static void kvm_get_preset_lpj(void) +{ + struct pvclock_vcpu_time_info *src; + unsigned long khz; + u64 lpj; + + src = &per_cpu(hv_clock, 0); + khz = pvclock_tsc_khz(src); + + lpj = ((u64)khz * 1000); + do_div(lpj, HZ); + preset_lpj = lpj; +} + static struct clocksource kvm_clock = { .name = "kvm-clock", .read = kvm_clock_read, @@ -100,7 +128,7 @@ static int kvm_register_clock(char *txt) } #ifdef CONFIG_X86_LOCAL_APIC -static void kvm_setup_secondary_clock(void) +static void __cpuinit kvm_setup_secondary_clock(void) { /* * Now that the first cpu already had this clocksource initialized, @@ -153,6 +181,7 @@ void __init kvmclock_init(void) pv_time_ops.get_wallclock = kvm_get_wallclock; pv_time_ops.set_wallclock = kvm_set_wallclock; pv_time_ops.sched_clock = kvm_clock_read; + pv_time_ops.get_tsc_khz = kvm_get_tsc_khz; #ifdef CONFIG_X86_LOCAL_APIC pv_apic_ops.setup_secondary_clock = kvm_setup_secondary_clock; #endif @@ -163,6 +192,7 @@ void __init kvmclock_init(void) #ifdef CONFIG_KEXEC machine_ops.crash_shutdown = kvm_crash_shutdown; #endif + kvm_get_preset_lpj(); clocksource_register(&kvm_clock); } } diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c index 0732adba05ca..37f420018a41 100644 --- a/arch/x86/kernel/machine_kexec_32.c +++ b/arch/x86/kernel/machine_kexec_32.c @@ -13,6 +13,7 @@ #include <linux/numa.h> #include <linux/ftrace.h> #include <linux/suspend.h> +#include <linux/gfp.h> #include <asm/pgtable.h> #include <asm/pgalloc.h> @@ -25,15 +26,6 @@ #include <asm/system.h> #include <asm/cacheflush.h> -#define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE))) -static u32 kexec_pgd[1024] PAGE_ALIGNED; -#ifdef CONFIG_X86_PAE -static u32 kexec_pmd0[1024] PAGE_ALIGNED; -static u32 kexec_pmd1[1024] PAGE_ALIGNED; -#endif -static u32 kexec_pte0[1024] PAGE_ALIGNED; -static u32 kexec_pte1[1024] PAGE_ALIGNED; - static void set_idt(void *newidt, __u16 limit) { struct desc_ptr curidt; @@ -76,6 +68,76 @@ static void load_segments(void) #undef __STR } +static void machine_kexec_free_page_tables(struct kimage *image) +{ + free_page((unsigned long)image->arch.pgd); +#ifdef CONFIG_X86_PAE + free_page((unsigned long)image->arch.pmd0); + free_page((unsigned long)image->arch.pmd1); +#endif + free_page((unsigned long)image->arch.pte0); + free_page((unsigned long)image->arch.pte1); +} + +static int machine_kexec_alloc_page_tables(struct kimage *image) +{ + image->arch.pgd = (pgd_t *)get_zeroed_page(GFP_KERNEL); +#ifdef CONFIG_X86_PAE + image->arch.pmd0 = (pmd_t *)get_zeroed_page(GFP_KERNEL); + image->arch.pmd1 = (pmd_t *)get_zeroed_page(GFP_KERNEL); +#endif + image->arch.pte0 = (pte_t *)get_zeroed_page(GFP_KERNEL); + image->arch.pte1 = (pte_t *)get_zeroed_page(GFP_KERNEL); + if (!image->arch.pgd || +#ifdef CONFIG_X86_PAE + !image->arch.pmd0 || !image->arch.pmd1 || +#endif + !image->arch.pte0 || !image->arch.pte1) { + machine_kexec_free_page_tables(image); + return -ENOMEM; + } + return 0; +} + +static void machine_kexec_page_table_set_one( + pgd_t *pgd, pmd_t *pmd, pte_t *pte, + unsigned long vaddr, unsigned long paddr) +{ + pud_t *pud; + + pgd += pgd_index(vaddr); +#ifdef CONFIG_X86_PAE + if (!(pgd_val(*pgd) & _PAGE_PRESENT)) + set_pgd(pgd, __pgd(__pa(pmd) | _PAGE_PRESENT)); +#endif + pud = pud_offset(pgd, vaddr); + pmd = pmd_offset(pud, vaddr); + if (!(pmd_val(*pmd) & _PAGE_PRESENT)) + set_pmd(pmd, __pmd(__pa(pte) | _PAGE_TABLE)); + pte = pte_offset_kernel(pmd, vaddr); + set_pte(pte, pfn_pte(paddr >> PAGE_SHIFT, PAGE_KERNEL_EXEC)); +} + +static void machine_kexec_prepare_page_tables(struct kimage *image) +{ + void *control_page; + pmd_t *pmd = 0; + + control_page = page_address(image->control_code_page); +#ifdef CONFIG_X86_PAE + pmd = image->arch.pmd0; +#endif + machine_kexec_page_table_set_one( + image->arch.pgd, pmd, image->arch.pte0, + (unsigned long)control_page, __pa(control_page)); +#ifdef CONFIG_X86_PAE + pmd = image->arch.pmd1; +#endif + machine_kexec_page_table_set_one( + image->arch.pgd, pmd, image->arch.pte1, + __pa(control_page), __pa(control_page)); +} + /* * A architecture hook called to validate the * proposed image and prepare the control pages @@ -87,12 +149,20 @@ static void load_segments(void) * reboot code buffer to allow us to avoid allocations * later. * - * Make control page executable. + * - Make control page executable. + * - Allocate page tables + * - Setup page tables */ int machine_kexec_prepare(struct kimage *image) { + int error; + if (nx_enabled) set_pages_x(image->control_code_page, 1); + error = machine_kexec_alloc_page_tables(image); + if (error) + return error; + machine_kexec_prepare_page_tables(image); return 0; } @@ -104,6 +174,7 @@ void machine_kexec_cleanup(struct kimage *image) { if (nx_enabled) set_pages_nx(image->control_code_page, 1); + machine_kexec_free_page_tables(image); } /* @@ -150,19 +221,11 @@ void machine_kexec(struct kimage *image) relocate_kernel_ptr = control_page; page_list[PA_CONTROL_PAGE] = __pa(control_page); page_list[VA_CONTROL_PAGE] = (unsigned long)control_page; - page_list[PA_PGD] = __pa(kexec_pgd); - page_list[VA_PGD] = (unsigned long)kexec_pgd; -#ifdef CONFIG_X86_PAE - page_list[PA_PMD_0] = __pa(kexec_pmd0); - page_list[VA_PMD_0] = (unsigned long)kexec_pmd0; - page_list[PA_PMD_1] = __pa(kexec_pmd1); - page_list[VA_PMD_1] = (unsigned long)kexec_pmd1; -#endif - page_list[PA_PTE_0] = __pa(kexec_pte0); - page_list[VA_PTE_0] = (unsigned long)kexec_pte0; - page_list[PA_PTE_1] = __pa(kexec_pte1); - page_list[VA_PTE_1] = (unsigned long)kexec_pte1; - page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page) << PAGE_SHIFT); + page_list[PA_PGD] = __pa(image->arch.pgd); + + if (image->type == KEXEC_TYPE_DEFAULT) + page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page) + << PAGE_SHIFT); /* The segment registers are funny things, they have both a * visible and an invisible part. Whenever the visible part is diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c index 7a1f8eeac2c7..c25fdb382292 100644 --- a/arch/x86/kernel/microcode_amd.c +++ b/arch/x86/kernel/microcode_amd.c @@ -10,7 +10,7 @@ * This driver allows to upgrade microcode on AMD * family 0x10 and 0x11 processors. * - * Licensed unter the terms of the GNU General Public + * Licensed under the terms of the GNU General Public * License version 2. See file COPYING for details. */ @@ -32,14 +32,14 @@ #include <linux/platform_device.h> #include <linux/pci.h> #include <linux/pci_ids.h> +#include <linux/uaccess.h> #include <asm/msr.h> -#include <asm/uaccess.h> #include <asm/processor.h> #include <asm/microcode.h> MODULE_DESCRIPTION("AMD Microcode Update Driver"); -MODULE_AUTHOR("Peter Oruba <peter.oruba@amd.com>"); +MODULE_AUTHOR("Peter Oruba"); MODULE_LICENSE("GPL v2"); #define UCODE_MAGIC 0x00414d44 @@ -47,43 +47,38 @@ MODULE_LICENSE("GPL v2"); #define UCODE_UCODE_TYPE 0x00000001 struct equiv_cpu_entry { - unsigned int installed_cpu; - unsigned int fixed_errata_mask; - unsigned int fixed_errata_compare; - unsigned int equiv_cpu; -}; + u32 installed_cpu; + u32 fixed_errata_mask; + u32 fixed_errata_compare; + u16 equiv_cpu; + u16 res; +} __attribute__((packed)); struct microcode_header_amd { - unsigned int data_code; - unsigned int patch_id; - unsigned char mc_patch_data_id[2]; - unsigned char mc_patch_data_len; - unsigned char init_flag; - unsigned int mc_patch_data_checksum; - unsigned int nb_dev_id; - unsigned int sb_dev_id; - unsigned char processor_rev_id[2]; - unsigned char nb_rev_id; - unsigned char sb_rev_id; - unsigned char bios_api_rev; - unsigned char reserved1[3]; - unsigned int match_reg[8]; -}; + u32 data_code; + u32 patch_id; + u16 mc_patch_data_id; + u8 mc_patch_data_len; + u8 init_flag; + u32 mc_patch_data_checksum; + u32 nb_dev_id; + u32 sb_dev_id; + u16 processor_rev_id; + u8 nb_rev_id; + u8 sb_rev_id; + u8 bios_api_rev; + u8 reserved1[3]; + u32 match_reg[8]; +} __attribute__((packed)); struct microcode_amd { struct microcode_header_amd hdr; unsigned int mpb[0]; }; -#define UCODE_MAX_SIZE (2048) -#define DEFAULT_UCODE_DATASIZE (896) -#define MC_HEADER_SIZE (sizeof(struct microcode_header_amd)) -#define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE) -#define DWSIZE (sizeof(u32)) -/* For now we support a fixed ucode total size only */ -#define get_totalsize(mc) \ - ((((struct microcode_amd *)mc)->hdr.mc_patch_data_len * 28) \ - + MC_HEADER_SIZE) +#define UCODE_MAX_SIZE 2048 +#define UCODE_CONTAINER_SECTION_HDR 8 +#define UCODE_CONTAINER_HEADER_SIZE 12 /* serialize access to the physical write */ static DEFINE_SPINLOCK(microcode_update_lock); @@ -93,31 +88,24 @@ static struct equiv_cpu_entry *equiv_cpu_table; static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig) { struct cpuinfo_x86 *c = &cpu_data(cpu); + u32 dummy; memset(csig, 0, sizeof(*csig)); - if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) { - printk(KERN_ERR "microcode: CPU%d not a capable AMD processor\n", - cpu); + printk(KERN_WARNING "microcode: CPU%d: AMD CPU family 0x%x not " + "supported\n", cpu, c->x86); return -1; } - - asm volatile("movl %1, %%ecx; rdmsr" - : "=a" (csig->rev) - : "i" (0x0000008B) : "ecx"); - - printk(KERN_INFO "microcode: collect_cpu_info_amd : patch_id=0x%x\n", - csig->rev); - + rdmsr(MSR_AMD64_PATCH_LEVEL, csig->rev, dummy); + printk(KERN_INFO "microcode: CPU%d: patch_level=0x%x\n", cpu, csig->rev); return 0; } static int get_matching_microcode(int cpu, void *mc, int rev) { struct microcode_header_amd *mc_header = mc; - struct pci_dev *nb_pci_dev, *sb_pci_dev; unsigned int current_cpu_id; - unsigned int equiv_cpu_id = 0x00; + u16 equiv_cpu_id = 0; unsigned int i = 0; BUG_ON(equiv_cpu_table == NULL); @@ -132,57 +120,25 @@ static int get_matching_microcode(int cpu, void *mc, int rev) } if (!equiv_cpu_id) { - printk(KERN_ERR "microcode: CPU%d cpu_id " - "not found in equivalent cpu table \n", cpu); + printk(KERN_WARNING "microcode: CPU%d: cpu revision " + "not listed in equivalent cpu table\n", cpu); return 0; } - if ((mc_header->processor_rev_id[0]) != (equiv_cpu_id & 0xff)) { - printk(KERN_ERR - "microcode: CPU%d patch does not match " - "(patch is %x, cpu extended is %x) \n", - cpu, mc_header->processor_rev_id[0], - (equiv_cpu_id & 0xff)); + if (mc_header->processor_rev_id != equiv_cpu_id) { + printk(KERN_ERR "microcode: CPU%d: patch mismatch " + "(processor_rev_id: %x, equiv_cpu_id: %x)\n", + cpu, mc_header->processor_rev_id, equiv_cpu_id); return 0; } - if ((mc_header->processor_rev_id[1]) != ((equiv_cpu_id >> 16) & 0xff)) { - printk(KERN_ERR "microcode: CPU%d patch does not match " - "(patch is %x, cpu base id is %x) \n", - cpu, mc_header->processor_rev_id[1], - ((equiv_cpu_id >> 16) & 0xff)); - + /* ucode might be chipset specific -- currently we don't support this */ + if (mc_header->nb_dev_id || mc_header->sb_dev_id) { + printk(KERN_ERR "microcode: CPU%d: loading of chipset " + "specific code not yet supported\n", cpu); return 0; } - /* ucode may be northbridge specific */ - if (mc_header->nb_dev_id) { - nb_pci_dev = pci_get_device(PCI_VENDOR_ID_AMD, - (mc_header->nb_dev_id & 0xff), - NULL); - if ((!nb_pci_dev) || - (mc_header->nb_rev_id != nb_pci_dev->revision)) { - printk(KERN_ERR "microcode: CPU%d NB mismatch \n", cpu); - pci_dev_put(nb_pci_dev); - return 0; - } - pci_dev_put(nb_pci_dev); - } - - /* ucode may be southbridge specific */ - if (mc_header->sb_dev_id) { - sb_pci_dev = pci_get_device(PCI_VENDOR_ID_AMD, - (mc_header->sb_dev_id & 0xff), - NULL); - if ((!sb_pci_dev) || - (mc_header->sb_rev_id != sb_pci_dev->revision)) { - printk(KERN_ERR "microcode: CPU%d SB mismatch \n", cpu); - pci_dev_put(sb_pci_dev); - return 0; - } - pci_dev_put(sb_pci_dev); - } - if (mc_header->patch_id <= rev) return 0; @@ -192,12 +148,10 @@ static int get_matching_microcode(int cpu, void *mc, int rev) static void apply_microcode_amd(int cpu) { unsigned long flags; - unsigned int eax, edx; - unsigned int rev; + u32 rev, dummy; int cpu_num = raw_smp_processor_id(); struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; struct microcode_amd *mc_amd = uci->mc; - unsigned long addr; /* We should bind the task to the CPU */ BUG_ON(cpu_num != cpu); @@ -206,42 +160,34 @@ static void apply_microcode_amd(int cpu) return; spin_lock_irqsave(µcode_update_lock, flags); - - addr = (unsigned long)&mc_amd->hdr.data_code; - edx = (unsigned int)(((unsigned long)upper_32_bits(addr))); - eax = (unsigned int)(((unsigned long)lower_32_bits(addr))); - - asm volatile("movl %0, %%ecx; wrmsr" : - : "i" (0xc0010020), "a" (eax), "d" (edx) : "ecx"); - + wrmsrl(MSR_AMD64_PATCH_LOADER, (u64)(long)&mc_amd->hdr.data_code); /* get patch id after patching */ - asm volatile("movl %1, %%ecx; rdmsr" - : "=a" (rev) - : "i" (0x0000008B) : "ecx"); - + rdmsr(MSR_AMD64_PATCH_LEVEL, rev, dummy); spin_unlock_irqrestore(µcode_update_lock, flags); /* check current patch id and patch's id for match */ if (rev != mc_amd->hdr.patch_id) { - printk(KERN_ERR "microcode: CPU%d update from revision " - "0x%x to 0x%x failed\n", cpu_num, - mc_amd->hdr.patch_id, rev); + printk(KERN_ERR "microcode: CPU%d: update failed " + "(for patch_level=0x%x)\n", cpu, mc_amd->hdr.patch_id); return; } - printk(KERN_INFO "microcode: CPU%d updated from revision " - "0x%x to 0x%x \n", - cpu_num, uci->cpu_sig.rev, mc_amd->hdr.patch_id); + printk(KERN_INFO "microcode: CPU%d: updated (new patch_level=0x%x)\n", + cpu, rev); uci->cpu_sig.rev = rev; } -static void * get_next_ucode(u8 *buf, unsigned int size, - int (*get_ucode_data)(void *, const void *, size_t), - unsigned int *mc_size) +static int get_ucode_data(void *to, const u8 *from, size_t n) +{ + memcpy(to, from, n); + return 0; +} + +static void *get_next_ucode(const u8 *buf, unsigned int size, + unsigned int *mc_size) { unsigned int total_size; -#define UCODE_CONTAINER_SECTION_HDR 8 u8 section_hdr[UCODE_CONTAINER_SECTION_HDR]; void *mc; @@ -249,39 +195,37 @@ static void * get_next_ucode(u8 *buf, unsigned int size, return NULL; if (section_hdr[0] != UCODE_UCODE_TYPE) { - printk(KERN_ERR "microcode: error! " - "Wrong microcode payload type field\n"); + printk(KERN_ERR "microcode: error: invalid type field in " + "container file section header\n"); return NULL; } total_size = (unsigned long) (section_hdr[4] + (section_hdr[5] << 8)); - printk(KERN_INFO "microcode: size %u, total_size %u\n", - size, total_size); + printk(KERN_DEBUG "microcode: size %u, total_size %u\n", + size, total_size); if (total_size > size || total_size > UCODE_MAX_SIZE) { - printk(KERN_ERR "microcode: error! Bad data in microcode data file\n"); + printk(KERN_ERR "microcode: error: size mismatch\n"); return NULL; } mc = vmalloc(UCODE_MAX_SIZE); if (mc) { memset(mc, 0, UCODE_MAX_SIZE); - if (get_ucode_data(mc, buf + UCODE_CONTAINER_SECTION_HDR, total_size)) { + if (get_ucode_data(mc, buf + UCODE_CONTAINER_SECTION_HDR, + total_size)) { vfree(mc); mc = NULL; } else *mc_size = total_size + UCODE_CONTAINER_SECTION_HDR; } -#undef UCODE_CONTAINER_SECTION_HDR return mc; } -static int install_equiv_cpu_table(u8 *buf, - int (*get_ucode_data)(void *, const void *, size_t)) +static int install_equiv_cpu_table(const u8 *buf) { -#define UCODE_CONTAINER_HEADER_SIZE 12 u8 *container_hdr[UCODE_CONTAINER_HEADER_SIZE]; unsigned int *buf_pos = (unsigned int *)container_hdr; unsigned long size; @@ -292,14 +236,15 @@ static int install_equiv_cpu_table(u8 *buf, size = buf_pos[2]; if (buf_pos[1] != UCODE_EQUIV_CPU_TABLE_TYPE || !size) { - printk(KERN_ERR "microcode: error! " - "Wrong microcode equivalnet cpu table\n"); + printk(KERN_ERR "microcode: error: invalid type field in " + "container file section header\n"); return 0; } equiv_cpu_table = (struct equiv_cpu_entry *) vmalloc(size); if (!equiv_cpu_table) { - printk(KERN_ERR "microcode: error, can't allocate memory for equiv CPU table\n"); + printk(KERN_ERR "microcode: failed to allocate " + "equivalent CPU table\n"); return 0; } @@ -310,7 +255,6 @@ static int install_equiv_cpu_table(u8 *buf, } return size + UCODE_CONTAINER_HEADER_SIZE; /* add header length */ -#undef UCODE_CONTAINER_HEADER_SIZE } static void free_equiv_cpu_table(void) @@ -321,18 +265,20 @@ static void free_equiv_cpu_table(void) } } -static int generic_load_microcode(int cpu, void *data, size_t size, - int (*get_ucode_data)(void *, const void *, size_t)) +static int generic_load_microcode(int cpu, const u8 *data, size_t size) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - u8 *ucode_ptr = data, *new_mc = NULL, *mc; + const u8 *ucode_ptr = data; + void *new_mc = NULL; + void *mc; int new_rev = uci->cpu_sig.rev; unsigned int leftover; unsigned long offset; - offset = install_equiv_cpu_table(ucode_ptr, get_ucode_data); + offset = install_equiv_cpu_table(ucode_ptr); if (!offset) { - printk(KERN_ERR "microcode: installing equivalent cpu table failed\n"); + printk(KERN_ERR "microcode: failed to create " + "equivalent cpu table\n"); return -EINVAL; } @@ -343,7 +289,7 @@ static int generic_load_microcode(int cpu, void *data, size_t size, unsigned int uninitialized_var(mc_size); struct microcode_header_amd *mc_header; - mc = get_next_ucode(ucode_ptr, leftover, get_ucode_data, &mc_size); + mc = get_next_ucode(ucode_ptr, leftover, &mc_size); if (!mc) break; @@ -353,7 +299,7 @@ static int generic_load_microcode(int cpu, void *data, size_t size, vfree(new_mc); new_rev = mc_header->patch_id; new_mc = mc; - } else + } else vfree(mc); ucode_ptr += mc_size; @@ -365,9 +311,9 @@ static int generic_load_microcode(int cpu, void *data, size_t size, if (uci->mc) vfree(uci->mc); uci->mc = new_mc; - pr_debug("microcode: CPU%d found a matching microcode update with" - " version 0x%x (current=0x%x)\n", - cpu, new_rev, uci->cpu_sig.rev); + pr_debug("microcode: CPU%d found a matching microcode " + "update with version 0x%x (current=0x%x)\n", + cpu, new_rev, uci->cpu_sig.rev); } else vfree(new_mc); } @@ -377,12 +323,6 @@ static int generic_load_microcode(int cpu, void *data, size_t size, return (int)leftover; } -static int get_ucode_fw(void *to, const void *from, size_t n) -{ - memcpy(to, from, n); - return 0; -} - static int request_microcode_fw(int cpu, struct device *device) { const char *fw_name = "amd-ucode/microcode_amd.bin"; @@ -394,12 +334,11 @@ static int request_microcode_fw(int cpu, struct device *device) ret = request_firmware(&firmware, fw_name, device); if (ret) { - printk(KERN_ERR "microcode: ucode data file %s load failed\n", fw_name); + printk(KERN_ERR "microcode: failed to load file %s\n", fw_name); return ret; } - ret = generic_load_microcode(cpu, (void*)firmware->data, firmware->size, - &get_ucode_fw); + ret = generic_load_microcode(cpu, firmware->data, firmware->size); release_firmware(firmware); @@ -408,8 +347,8 @@ static int request_microcode_fw(int cpu, struct device *device) static int request_microcode_user(int cpu, const void __user *buf, size_t size) { - printk(KERN_WARNING "microcode: AMD microcode update via /dev/cpu/microcode" - "is not supported\n"); + printk(KERN_INFO "microcode: AMD microcode update via " + "/dev/cpu/microcode not supported\n"); return -1; } @@ -433,3 +372,4 @@ struct microcode_ops * __init init_amd_microcode(void) { return µcode_amd_ops; } + diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c index 936d8d55f230..c9b721ba968c 100644 --- a/arch/x86/kernel/microcode_core.c +++ b/arch/x86/kernel/microcode_core.c @@ -99,7 +99,7 @@ MODULE_LICENSE("GPL"); #define MICROCODE_VERSION "2.00" -struct microcode_ops *microcode_ops; +static struct microcode_ops *microcode_ops; /* no concurrent ->write()s are allowed on /dev/cpu/microcode */ static DEFINE_MUTEX(microcode_mutex); @@ -203,7 +203,7 @@ MODULE_ALIAS_MISCDEV(MICROCODE_MINOR); #endif /* fake device for request_firmware */ -struct platform_device *microcode_pdev; +static struct platform_device *microcode_pdev; static ssize_t reload_store(struct sys_device *dev, struct sysdev_attribute *attr, @@ -272,13 +272,18 @@ static struct attribute_group mc_attr_group = { .name = "microcode", }; -static void microcode_fini_cpu(int cpu) +static void __microcode_fini_cpu(int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - mutex_lock(µcode_mutex); microcode_ops->microcode_fini_cpu(cpu); uci->valid = 0; +} + +static void microcode_fini_cpu(int cpu) +{ + mutex_lock(µcode_mutex); + __microcode_fini_cpu(cpu); mutex_unlock(µcode_mutex); } @@ -306,12 +311,16 @@ static int microcode_resume_cpu(int cpu) * to this cpu (a bit of paranoia): */ if (microcode_ops->collect_cpu_info(cpu, &nsig)) { - microcode_fini_cpu(cpu); + __microcode_fini_cpu(cpu); + printk(KERN_ERR "failed to collect_cpu_info for resuming cpu #%d\n", + cpu); return -1; } - if (memcmp(&nsig, &uci->cpu_sig, sizeof(nsig))) { - microcode_fini_cpu(cpu); + if ((nsig.sig != uci->cpu_sig.sig) || (nsig.pf != uci->cpu_sig.pf)) { + __microcode_fini_cpu(cpu); + printk(KERN_ERR "cached ucode doesn't match the resuming cpu #%d\n", + cpu); /* Should we look for a new ucode here? */ return 1; } @@ -319,7 +328,7 @@ static int microcode_resume_cpu(int cpu) return 0; } -void microcode_update_cpu(int cpu) +static void microcode_update_cpu(int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; int err = 0; @@ -480,8 +489,8 @@ static int __init microcode_init(void) printk(KERN_INFO "Microcode Update Driver: v" MICROCODE_VERSION - " <tigran@aivazian.fsnet.co.uk>" - " <peter.oruba@amd.com>\n"); + " <tigran@aivazian.fsnet.co.uk>," + " Peter Oruba\n"); return 0; } diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index 622dc4a21784..b7f4c929e615 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -155,6 +155,7 @@ static DEFINE_SPINLOCK(microcode_update_lock); static int collect_cpu_info(int cpu_num, struct cpu_signature *csig) { struct cpuinfo_x86 *c = &cpu_data(cpu_num); + unsigned long flags; unsigned int val[2]; memset(csig, 0, sizeof(*csig)); @@ -174,11 +175,16 @@ static int collect_cpu_info(int cpu_num, struct cpu_signature *csig) csig->pf = 1 << ((val[1] >> 18) & 7); } + /* serialize access to the physical write to MSR 0x79 */ + spin_lock_irqsave(µcode_update_lock, flags); + wrmsr(MSR_IA32_UCODE_REV, 0, 0); /* see notes above for revision 1.07. Apparent chip bug */ sync_core(); /* get the current revision from MSR 0x8B */ rdmsr(MSR_IA32_UCODE_REV, val[0], csig->rev); + spin_unlock_irqrestore(µcode_update_lock, flags); + pr_debug("microcode: collect_cpu_info : sig=0x%x, pf=0x%x, rev=0x%x\n", csig->sig, csig->pf, csig->rev); @@ -465,7 +471,7 @@ static void microcode_fini_cpu(int cpu) uci->mc = NULL; } -struct microcode_ops microcode_intel_ops = { +static struct microcode_ops microcode_intel_ops = { .request_microcode_user = request_microcode_user, .request_microcode_fw = request_microcode_fw, .collect_cpu_info = collect_cpu_info, diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index f98f4e1dba09..45e3b69808ba 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -586,23 +586,23 @@ static void __init __get_smp_config(unsigned int early) { struct intel_mp_floating *mpf = mpf_found; - if (x86_quirks->mach_get_smp_config) { - if (x86_quirks->mach_get_smp_config(early)) - return; - } + if (!mpf) + return; + if (acpi_lapic && early) return; + /* - * ACPI supports both logical (e.g. Hyper-Threading) and physical - * processors, where MPS only supports physical. + * MPS doesn't support hyperthreading, aka only have + * thread 0 apic id in MPS table */ - if (acpi_lapic && acpi_ioapic) { - printk(KERN_INFO "Using ACPI (MADT) for SMP configuration " - "information\n"); + if (acpi_lapic && acpi_ioapic) return; - } else if (acpi_lapic) - printk(KERN_INFO "Using ACPI for processor (LAPIC) " - "configuration information\n"); + + if (x86_quirks->mach_get_smp_config) { + if (x86_quirks->mach_get_smp_config(early)) + return; + } printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification); diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c index 2e2af5d18191..82a7c7ed6d45 100644 --- a/arch/x86/kernel/msr.c +++ b/arch/x86/kernel/msr.c @@ -163,8 +163,8 @@ static int __cpuinit msr_device_create(int cpu) { struct device *dev; - dev = device_create_drvdata(msr_class, NULL, MKDEV(MSR_MAJOR, cpu), - NULL, "msr%d", cpu); + dev = device_create(msr_class, NULL, MKDEV(MSR_MAJOR, cpu), NULL, + "msr%d", cpu); return IS_ERR(dev) ? PTR_ERR(dev) : 0; } diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c index 2c97f07f1c2c..8bd1bf9622a7 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/nmi.c @@ -131,6 +131,11 @@ static void report_broken_nmi(int cpu, int *prev_nmi_count) atomic_dec(&nmi_active); } +static void __acpi_nmi_disable(void *__unused) +{ + apic_write(APIC_LVT0, APIC_DM_NMI | APIC_LVT_MASKED); +} + int __init check_nmi_watchdog(void) { unsigned int *prev_nmi_count; @@ -179,8 +184,12 @@ int __init check_nmi_watchdog(void) kfree(prev_nmi_count); return 0; error: - if (nmi_watchdog == NMI_IO_APIC && !timer_through_8259) - disable_8259A_irq(0); + if (nmi_watchdog == NMI_IO_APIC) { + if (!timer_through_8259) + disable_8259A_irq(0); + on_each_cpu(__acpi_nmi_disable, NULL, 1); + } + #ifdef CONFIG_X86_32 timer_ack = 0; #endif @@ -199,12 +208,17 @@ static int __init setup_nmi_watchdog(char *str) ++str; } - get_option(&str, &nmi); - - if (nmi >= NMI_INVALID) - return 0; + if (!strncmp(str, "lapic", 5)) + nmi_watchdog = NMI_LOCAL_APIC; + else if (!strncmp(str, "ioapic", 6)) + nmi_watchdog = NMI_IO_APIC; + else { + get_option(&str, &nmi); + if (nmi >= NMI_INVALID) + return 0; + nmi_watchdog = nmi; + } - nmi_watchdog = nmi; return 1; } __setup("nmi_watchdog=", setup_nmi_watchdog); @@ -285,11 +299,6 @@ void acpi_nmi_enable(void) on_each_cpu(__acpi_nmi_enable, NULL, 1); } -static void __acpi_nmi_disable(void *__unused) -{ - apic_write(APIC_LVT0, APIC_DM_NMI | APIC_LVT_MASKED); -} - /* * Disable timer based NMIs on all CPUs: */ @@ -340,6 +349,8 @@ void stop_apic_nmi_watchdog(void *unused) return; if (nmi_watchdog == NMI_LOCAL_APIC) lapic_watchdog_stop(); + else + __acpi_nmi_disable(NULL); __get_cpu_var(wd_enabled) = 0; atomic_dec(&nmi_active); } @@ -465,6 +476,24 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) #ifdef CONFIG_SYSCTL +static void enable_ioapic_nmi_watchdog_single(void *unused) +{ + __get_cpu_var(wd_enabled) = 1; + atomic_inc(&nmi_active); + __acpi_nmi_enable(NULL); +} + +static void enable_ioapic_nmi_watchdog(void) +{ + on_each_cpu(enable_ioapic_nmi_watchdog_single, NULL, 1); + touch_nmi_watchdog(); +} + +static void disable_ioapic_nmi_watchdog(void) +{ + on_each_cpu(stop_apic_nmi_watchdog, NULL, 1); +} + static int __init setup_unknown_nmi_panic(char *str) { unknown_nmi_panic = 1; @@ -507,6 +536,11 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file, enable_lapic_nmi_watchdog(); else disable_lapic_nmi_watchdog(); + } else if (nmi_watchdog == NMI_IO_APIC) { + if (nmi_watchdog_enabled) + enable_ioapic_nmi_watchdog(); + else + disable_ioapic_nmi_watchdog(); } else { printk(KERN_WARNING "NMI watchdog doesn't know what hardware to touch\n"); diff --git a/arch/x86/kernel/numaq_32.c b/arch/x86/kernel/numaq_32.c index 4caff39078e0..0deea37a53cf 100644 --- a/arch/x86/kernel/numaq_32.c +++ b/arch/x86/kernel/numaq_32.c @@ -31,7 +31,7 @@ #include <asm/numaq.h> #include <asm/topology.h> #include <asm/processor.h> -#include <asm/mpspec.h> +#include <asm/genapic.h> #include <asm/e820.h> #include <asm/setup.h> @@ -235,6 +235,13 @@ static int __init numaq_setup_ioapic_ids(void) return 1; } +static int __init numaq_update_genapic(void) +{ + genapic->wakeup_cpu = wakeup_secondary_cpu_via_nmi; + + return 0; +} + static struct x86_quirks numaq_x86_quirks __initdata = { .arch_pre_time_init = numaq_pre_time_init, .arch_time_init = NULL, @@ -250,6 +257,7 @@ static struct x86_quirks numaq_x86_quirks __initdata = { .mpc_oem_pci_bus = mpc_oem_pci_bus, .smp_read_mpc_oem = smp_read_mpc_oem, .setup_ioapic_ids = numaq_setup_ioapic_ids, + .update_genapic = numaq_update_genapic, }; void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem, diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c index 0e9f1982b1dd..95777b0faa73 100644 --- a/arch/x86/kernel/paravirt-spinlocks.c +++ b/arch/x86/kernel/paravirt-spinlocks.c @@ -7,7 +7,8 @@ #include <asm/paravirt.h> -static void default_spin_lock_flags(struct raw_spinlock *lock, unsigned long flags) +static inline void +default_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags) { __raw_spin_lock(lock); } diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c index 080d1d27f37a..d28bbdc35e4e 100644 --- a/arch/x86/kernel/pci-calgary_64.c +++ b/arch/x86/kernel/pci-calgary_64.c @@ -217,16 +217,6 @@ static inline unsigned long verify_bit_range(unsigned long* bitmap, #endif /* CONFIG_IOMMU_DEBUG */ -static inline unsigned int num_dma_pages(unsigned long dma, unsigned int dmalen) -{ - unsigned int npages; - - npages = PAGE_ALIGN(dma + dmalen) - (dma & PAGE_MASK); - npages >>= PAGE_SHIFT; - - return npages; -} - static inline int translation_enabled(struct iommu_table *tbl) { /* only PHBs with translation enabled have an IOMMU table */ @@ -408,7 +398,7 @@ static void calgary_unmap_sg(struct device *dev, if (dmalen == 0) break; - npages = num_dma_pages(dma, dmalen); + npages = iommu_num_pages(dma, dmalen, PAGE_SIZE); iommu_free(tbl, dma, npages); } } @@ -427,7 +417,7 @@ static int calgary_map_sg(struct device *dev, struct scatterlist *sg, BUG_ON(!sg_page(s)); vaddr = (unsigned long) sg_virt(s); - npages = num_dma_pages(vaddr, s->length); + npages = iommu_num_pages(vaddr, s->length, PAGE_SIZE); entry = iommu_range_alloc(dev, tbl, npages); if (entry == bad_dma_address) { @@ -464,7 +454,7 @@ static dma_addr_t calgary_map_single(struct device *dev, phys_addr_t paddr, struct iommu_table *tbl = find_iommu_table(dev); uaddr = (unsigned long)vaddr; - npages = num_dma_pages(uaddr, size); + npages = iommu_num_pages(uaddr, size, PAGE_SIZE); return iommu_alloc(dev, tbl, vaddr, npages, direction); } @@ -475,7 +465,7 @@ static void calgary_unmap_single(struct device *dev, dma_addr_t dma_handle, struct iommu_table *tbl = find_iommu_table(dev); unsigned int npages; - npages = num_dma_pages(dma_handle, size); + npages = iommu_num_pages(dma_handle, size, PAGE_SIZE); iommu_free(tbl, dma_handle, npages); } @@ -1577,7 +1567,7 @@ static int __init calgary_parse_options(char *p) ++p; if (*p == '\0') break; - bridge = simple_strtol(p, &endp, 0); + bridge = simple_strtoul(p, &endp, 0); if (p == endp) break; diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 0a3824e837b4..19a1044a0cd9 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -6,6 +6,7 @@ #include <asm/proto.h> #include <asm/dma.h> #include <asm/iommu.h> +#include <asm/gart.h> #include <asm/calgary.h> #include <asm/amd_iommu.h> @@ -30,11 +31,6 @@ int no_iommu __read_mostly; /* Set this to 1 if there is a HW IOMMU in the system */ int iommu_detected __read_mostly = 0; -/* This tells the BIO block layer to assume merging. Default to off - because we cannot guarantee merging later. */ -int iommu_bio_merge __read_mostly = 0; -EXPORT_SYMBOL(iommu_bio_merge); - dma_addr_t bad_dma_address __read_mostly = 0; EXPORT_SYMBOL(bad_dma_address); @@ -105,11 +101,15 @@ static void __init dma32_free_bootmem(void) dma32_bootmem_ptr = NULL; dma32_bootmem_size = 0; } +#endif void __init pci_iommu_alloc(void) { +#ifdef CONFIG_X86_64 /* free the range so iommu could get some range less than 4G */ dma32_free_bootmem(); +#endif + /* * The order of these functions is important for * fall-back/fail-over reasons @@ -125,15 +125,6 @@ void __init pci_iommu_alloc(void) pci_swiotlb_init(); } -unsigned long iommu_num_pages(unsigned long addr, unsigned long len) -{ - unsigned long size = roundup((addr & ~PAGE_MASK) + len, PAGE_SIZE); - - return size >> PAGE_SHIFT; -} -EXPORT_SYMBOL(iommu_num_pages); -#endif - void *dma_generic_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_addr, gfp_t flag) { @@ -188,7 +179,6 @@ static __init int iommu_setup(char *p) } if (!strncmp(p, "biomerge", 8)) { - iommu_bio_merge = 4096; iommu_merge = 1; force_iommu = 1; } @@ -300,8 +290,8 @@ fs_initcall(pci_iommu_init); static __devinit void via_no_dac(struct pci_dev *dev) { if ((dev->class >> 8) == PCI_CLASS_BRIDGE_PCI && forbid_dac == 0) { - printk(KERN_INFO "PCI: VIA PCI bridge detected." - "Disabling DAC.\n"); + printk(KERN_INFO + "PCI: VIA PCI bridge detected. Disabling DAC.\n"); forbid_dac = 1; } } diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 145f1c83369f..a35eaa379ff6 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -123,6 +123,8 @@ static void free_iommu(unsigned long offset, int size) spin_lock_irqsave(&iommu_bitmap_lock, flags); iommu_area_free(iommu_gart_bitmap, offset, size); + if (offset >= next_bit) + next_bit = offset + size; spin_unlock_irqrestore(&iommu_bitmap_lock, flags); } @@ -231,7 +233,7 @@ nonforced_iommu(struct device *dev, unsigned long addr, size_t size) static dma_addr_t dma_map_area(struct device *dev, dma_addr_t phys_mem, size_t size, int dir, unsigned long align_mask) { - unsigned long npages = iommu_num_pages(phys_mem, size); + unsigned long npages = iommu_num_pages(phys_mem, size, PAGE_SIZE); unsigned long iommu_page = alloc_iommu(dev, npages, align_mask); int i; @@ -285,7 +287,7 @@ static void gart_unmap_single(struct device *dev, dma_addr_t dma_addr, return; iommu_page = (dma_addr - iommu_bus_base)>>PAGE_SHIFT; - npages = iommu_num_pages(dma_addr, size); + npages = iommu_num_pages(dma_addr, size, PAGE_SIZE); for (i = 0; i < npages; i++) { iommu_gatt_base[iommu_page + i] = gart_unmapped_entry; CLEAR_LEAK(iommu_page + i); @@ -368,7 +370,7 @@ static int __dma_map_cont(struct device *dev, struct scatterlist *start, } addr = phys_addr; - pages = iommu_num_pages(s->offset, s->length); + pages = iommu_num_pages(s->offset, s->length, PAGE_SIZE); while (pages--) { iommu_gatt_base[iommu_page] = GPTE_ENCODE(addr); SET_LEAK(iommu_page); @@ -451,7 +453,7 @@ gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, int dir) seg_size += s->length; need = nextneed; - pages += iommu_num_pages(s->offset, s->length); + pages += iommu_num_pages(s->offset, s->length, PAGE_SIZE); ps = s; } if (dma_map_cont(dev, start_sg, i - start, sgmap, pages, need) < 0) @@ -743,10 +745,8 @@ void __init gart_iommu_init(void) unsigned long scratch; long i; - if (cache_k8_northbridges() < 0 || num_k8_northbridges == 0) { - printk(KERN_INFO "PCI-GART: No AMD northbridge found.\n"); + if (cache_k8_northbridges() < 0 || num_k8_northbridges == 0) return; - } #ifndef CONFIG_AGP_AMD64 no_agp = 1; diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c index c4ce0332759e..242c3440687f 100644 --- a/arch/x86/kernel/pci-swiotlb_64.c +++ b/arch/x86/kernel/pci-swiotlb_64.c @@ -3,6 +3,8 @@ #include <linux/pci.h> #include <linux/cache.h> #include <linux/module.h> +#include <linux/swiotlb.h> +#include <linux/bootmem.h> #include <linux/dma-mapping.h> #include <asm/iommu.h> @@ -11,6 +13,31 @@ int swiotlb __read_mostly; +void *swiotlb_alloc_boot(size_t size, unsigned long nslabs) +{ + return alloc_bootmem_low_pages(size); +} + +void *swiotlb_alloc(unsigned order, unsigned long nslabs) +{ + return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order); +} + +dma_addr_t swiotlb_phys_to_bus(phys_addr_t paddr) +{ + return paddr; +} + +phys_addr_t swiotlb_bus_to_phys(dma_addr_t baddr) +{ + return baddr; +} + +int __weak swiotlb_arch_range_needs_mapping(void *ptr, size_t size) +{ + return 0; +} + static dma_addr_t swiotlb_map_single_phys(struct device *hwdev, phys_addr_t paddr, size_t size, int direction) @@ -18,9 +45,21 @@ swiotlb_map_single_phys(struct device *hwdev, phys_addr_t paddr, size_t size, return swiotlb_map_single(hwdev, phys_to_virt(paddr), size, direction); } +static void *x86_swiotlb_alloc_coherent(struct device *hwdev, size_t size, + dma_addr_t *dma_handle, gfp_t flags) +{ + void *vaddr; + + vaddr = dma_generic_alloc_coherent(hwdev, size, dma_handle, flags); + if (vaddr) + return vaddr; + + return swiotlb_alloc_coherent(hwdev, size, dma_handle, flags); +} + struct dma_mapping_ops swiotlb_dma_ops = { .mapping_error = swiotlb_dma_mapping_error, - .alloc_coherent = swiotlb_alloc_coherent, + .alloc_coherent = x86_swiotlb_alloc_coherent, .free_coherent = swiotlb_free_coherent, .map_single = swiotlb_map_single_phys, .unmap_single = swiotlb_unmap_single, @@ -38,8 +77,10 @@ struct dma_mapping_ops swiotlb_dma_ops = { 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) swiotlb = 1; +#endif if (swiotlb_force) swiotlb = 1; if (swiotlb) { diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index c622772744d8..e68bb9e30864 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -1,13 +1,16 @@ #include <linux/errno.h> #include <linux/kernel.h> #include <linux/mm.h> +#include <asm/idle.h> #include <linux/smp.h> #include <linux/slab.h> #include <linux/sched.h> #include <linux/module.h> #include <linux/pm.h> #include <linux/clockchips.h> +#include <linux/ftrace.h> #include <asm/system.h> +#include <asm/apic.h> unsigned long idle_halt; EXPORT_SYMBOL(idle_halt); @@ -100,6 +103,9 @@ static inline int hlt_use_halt(void) void default_idle(void) { if (hlt_use_halt()) { + struct power_trace it; + + trace_power_start(&it, POWER_CSTATE, 1); current_thread_info()->status &= ~TS_POLLING; /* * TS_POLLING-cleared state must be visible before we @@ -112,6 +118,7 @@ void default_idle(void) else local_irq_enable(); current_thread_info()->status |= TS_POLLING; + trace_power_end(&it); } else { local_irq_enable(); /* loop is done by the caller */ @@ -122,6 +129,21 @@ void default_idle(void) EXPORT_SYMBOL(default_idle); #endif +void stop_this_cpu(void *dummy) +{ + local_irq_disable(); + /* + * Remove this CPU: + */ + cpu_clear(smp_processor_id(), cpu_online_map); + disable_local_APIC(); + + for (;;) { + if (hlt_works(smp_processor_id())) + halt(); + } +} + static void do_nothing(void *unused) { } @@ -154,24 +176,31 @@ EXPORT_SYMBOL_GPL(cpu_idle_wait); */ void mwait_idle_with_hints(unsigned long ax, unsigned long cx) { + struct power_trace it; + + trace_power_start(&it, POWER_CSTATE, (ax>>4)+1); if (!need_resched()) { __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); if (!need_resched()) __mwait(ax, cx); } + trace_power_end(&it); } /* Default MONITOR/MWAIT with no hints, used for default C1 state */ static void mwait_idle(void) { + struct power_trace it; if (!need_resched()) { + trace_power_start(&it, POWER_CSTATE, 1); __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); if (!need_resched()) __sti_mwait(0, 0); else local_irq_enable(); + trace_power_end(&it); } else local_irq_enable(); } @@ -183,9 +212,13 @@ static void mwait_idle(void) */ static void poll_idle(void) { + struct power_trace it; + + trace_power_start(&it, POWER_CSTATE, 0); local_irq_enable(); while (!need_resched()) cpu_relax(); + trace_power_end(&it); } /* @@ -270,7 +303,7 @@ static void c1e_idle(void) rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi); if (lo & K8_INTP_C1E_ACTIVE_MASK) { c1e_detected = 1; - if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) + if (!boot_cpu_has(X86_FEATURE_NONSTOP_TSC)) mark_tsc_unstable("TSC halt in AMD C1E"); printk(KERN_INFO "System has AMD C1E enabled\n"); set_cpu_cap(&boot_cpu_data, X86_FEATURE_AMDC1E); diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 0a1302fe6d45..3ba155d24884 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -38,6 +38,7 @@ #include <linux/percpu.h> #include <linux/prctl.h> #include <linux/dmi.h> +#include <linux/ftrace.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -59,6 +60,7 @@ #include <asm/idle.h> #include <asm/syscalls.h> #include <asm/smp.h> +#include <asm/ds.h> asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); @@ -250,14 +252,8 @@ void exit_thread(void) tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET; put_cpu(); } -#ifdef CONFIG_X86_DS - /* Free any DS contexts that have not been properly released. */ - if (unlikely(current->thread.ds_ctx)) { - /* we clear debugctl to make sure DS is not used. */ - update_debugctlmsr(0); - ds_free(current->thread.ds_ctx); - } -#endif /* CONFIG_X86_DS */ + + ds_exit_thread(current); } void flush_thread(void) @@ -339,6 +335,12 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, kfree(p->thread.io_bitmap_ptr); p->thread.io_bitmap_max = 0; } + + ds_copy_thread(p, current); + + clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR); + p->thread.debugctlmsr = 0; + return err; } @@ -419,48 +421,19 @@ int set_tsc_mode(unsigned int val) return 0; } -#ifdef CONFIG_X86_DS -static int update_debugctl(struct thread_struct *prev, - struct thread_struct *next, unsigned long debugctl) -{ - unsigned long ds_prev = 0; - unsigned long ds_next = 0; - - if (prev->ds_ctx) - ds_prev = (unsigned long)prev->ds_ctx->ds; - if (next->ds_ctx) - ds_next = (unsigned long)next->ds_ctx->ds; - - if (ds_next != ds_prev) { - /* we clear debugctl to make sure DS - * is not in use when we change it */ - debugctl = 0; - update_debugctlmsr(0); - wrmsr(MSR_IA32_DS_AREA, ds_next, 0); - } - return debugctl; -} -#else -static int update_debugctl(struct thread_struct *prev, - struct thread_struct *next, unsigned long debugctl) -{ - return debugctl; -} -#endif /* CONFIG_X86_DS */ - static noinline void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, struct tss_struct *tss) { struct thread_struct *prev, *next; - unsigned long debugctl; prev = &prev_p->thread; next = &next_p->thread; - debugctl = update_debugctl(prev, next, prev->debugctlmsr); - - if (next->debugctlmsr != debugctl) + if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) || + test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR)) + ds_switch_to(prev_p, next_p); + else if (next->debugctlmsr != prev->debugctlmsr) update_debugctlmsr(next->debugctlmsr); if (test_tsk_thread_flag(next_p, TIF_DEBUG)) { @@ -482,15 +455,6 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, hard_enable_TSC(); } -#ifdef CONFIG_X86_PTRACE_BTS - if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS)) - ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS); - - if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS)) - ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES); -#endif /* CONFIG_X86_PTRACE_BTS */ - - if (!test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) { /* * Disable the bitmap via an invalid offset. We still cache @@ -548,7 +512,8 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, * the task-switch, and shows up in ret_from_fork in entry.S, * for example. */ -struct task_struct * __switch_to(struct task_struct *prev_p, struct task_struct *next_p) +__notrace_funcgraph struct task_struct * +__switch_to(struct task_struct *prev_p, struct task_struct *next_p) { struct thread_struct *prev = &prev_p->thread, *next = &next_p->thread; diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 749d5f888d4d..efb0396e19bf 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -40,6 +40,7 @@ #include <linux/prctl.h> #include <linux/uaccess.h> #include <linux/io.h> +#include <linux/ftrace.h> #include <asm/pgtable.h> #include <asm/system.h> @@ -53,6 +54,7 @@ #include <asm/ia32.h> #include <asm/idle.h> #include <asm/syscalls.h> +#include <asm/ds.h> asmlinkage extern void ret_from_fork(void); @@ -64,6 +66,13 @@ void idle_notifier_register(struct notifier_block *n) { atomic_notifier_chain_register(&idle_notifier, n); } +EXPORT_SYMBOL_GPL(idle_notifier_register); + +void idle_notifier_unregister(struct notifier_block *n) +{ + atomic_notifier_chain_unregister(&idle_notifier, n); +} +EXPORT_SYMBOL_GPL(idle_notifier_unregister); void enter_idle(void) { @@ -240,14 +249,8 @@ void exit_thread(void) t->io_bitmap_max = 0; put_cpu(); } -#ifdef CONFIG_X86_DS - /* Free any DS contexts that have not been properly released. */ - if (unlikely(t->ds_ctx)) { - /* we clear debugctl to make sure DS is not used. */ - update_debugctlmsr(0); - ds_free(t->ds_ctx); - } -#endif /* CONFIG_X86_DS */ + + ds_exit_thread(current); } void flush_thread(void) @@ -377,6 +380,12 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, if (err) goto out; } + + ds_copy_thread(p, me); + + clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR); + p->thread.debugctlmsr = 0; + err = 0; out: if (err && p->thread.io_bitmap_ptr) { @@ -475,35 +484,14 @@ static inline void __switch_to_xtra(struct task_struct *prev_p, struct tss_struct *tss) { struct thread_struct *prev, *next; - unsigned long debugctl; prev = &prev_p->thread, next = &next_p->thread; - debugctl = prev->debugctlmsr; - -#ifdef CONFIG_X86_DS - { - unsigned long ds_prev = 0, ds_next = 0; - - if (prev->ds_ctx) - ds_prev = (unsigned long)prev->ds_ctx->ds; - if (next->ds_ctx) - ds_next = (unsigned long)next->ds_ctx->ds; - - if (ds_next != ds_prev) { - /* - * We clear debugctl to make sure DS - * is not in use when we change it: - */ - debugctl = 0; - update_debugctlmsr(0); - wrmsrl(MSR_IA32_DS_AREA, ds_next); - } - } -#endif /* CONFIG_X86_DS */ - - if (next->debugctlmsr != debugctl) + if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) || + test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR)) + ds_switch_to(prev_p, next_p); + else if (next->debugctlmsr != prev->debugctlmsr) update_debugctlmsr(next->debugctlmsr); if (test_tsk_thread_flag(next_p, TIF_DEBUG)) { @@ -538,14 +526,6 @@ static inline void __switch_to_xtra(struct task_struct *prev_p, */ memset(tss->io_bitmap, 0xff, prev->io_bitmap_max); } - -#ifdef CONFIG_X86_PTRACE_BTS - if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS)) - ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS); - - if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS)) - ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES); -#endif /* CONFIG_X86_PTRACE_BTS */ } /* @@ -556,8 +536,9 @@ static inline void __switch_to_xtra(struct task_struct *prev_p, * - could test fs/gs bitsliced * * Kprobes not supported here. Set the probe on schedule instead. + * Function graph tracer not supported too. */ -struct task_struct * +__notrace_funcgraph struct task_struct * __switch_to(struct task_struct *prev_p, struct task_struct *next_p) { struct thread_struct *prev = &prev_p->thread; diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 0a6d8c12e10d..0a5df5f82fb9 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -581,158 +581,91 @@ static int ioperm_get(struct task_struct *target, } #ifdef CONFIG_X86_PTRACE_BTS -/* - * The configuration for a particular BTS hardware implementation. - */ -struct bts_configuration { - /* the size of a BTS record in bytes; at most BTS_MAX_RECORD_SIZE */ - unsigned char sizeof_bts; - /* the size of a field in the BTS record in bytes */ - unsigned char sizeof_field; - /* a bitmask to enable/disable BTS in DEBUGCTL MSR */ - unsigned long debugctl_mask; -}; -static struct bts_configuration bts_cfg; - -#define BTS_MAX_RECORD_SIZE (8 * 3) - - -/* - * Branch Trace Store (BTS) uses the following format. Different - * architectures vary in the size of those fields. - * - source linear address - * - destination linear address - * - flags - * - * Later architectures use 64bit pointers throughout, whereas earlier - * architectures use 32bit pointers in 32bit mode. - * - * We compute the base address for the first 8 fields based on: - * - the field size stored in the DS configuration - * - the relative field position - * - * In order to store additional information in the BTS buffer, we use - * a special source address to indicate that the record requires - * special interpretation. - * - * Netburst indicated via a bit in the flags field whether the branch - * was predicted; this is ignored. - */ - -enum bts_field { - bts_from = 0, - bts_to, - bts_flags, - - bts_escape = (unsigned long)-1, - bts_qual = bts_to, - bts_jiffies = bts_flags -}; - -static inline unsigned long bts_get(const char *base, enum bts_field field) -{ - base += (bts_cfg.sizeof_field * field); - return *(unsigned long *)base; -} - -static inline void bts_set(char *base, enum bts_field field, unsigned long val) -{ - base += (bts_cfg.sizeof_field * field);; - (*(unsigned long *)base) = val; -} - -/* - * Translate a BTS record from the raw format into the bts_struct format - * - * out (out): bts_struct interpretation - * raw: raw BTS record - */ -static void ptrace_bts_translate_record(struct bts_struct *out, const void *raw) -{ - memset(out, 0, sizeof(*out)); - if (bts_get(raw, bts_from) == bts_escape) { - out->qualifier = bts_get(raw, bts_qual); - out->variant.jiffies = bts_get(raw, bts_jiffies); - } else { - out->qualifier = BTS_BRANCH; - out->variant.lbr.from_ip = bts_get(raw, bts_from); - out->variant.lbr.to_ip = bts_get(raw, bts_to); - } -} - static int ptrace_bts_read_record(struct task_struct *child, size_t index, struct bts_struct __user *out) { - struct bts_struct ret; - const void *bts_record; - size_t bts_index, bts_end; + const struct bts_trace *trace; + struct bts_struct bts; + const unsigned char *at; int error; - error = ds_get_bts_end(child, &bts_end); - if (error < 0) - return error; - - if (bts_end <= index) - return -EINVAL; + trace = ds_read_bts(child->bts); + if (!trace) + return -EPERM; - error = ds_get_bts_index(child, &bts_index); - if (error < 0) - return error; + at = trace->ds.top - ((index + 1) * trace->ds.size); + if ((void *)at < trace->ds.begin) + at += (trace->ds.n * trace->ds.size); - /* translate the ptrace bts index into the ds bts index */ - bts_index += bts_end - (index + 1); - if (bts_end <= bts_index) - bts_index -= bts_end; + if (!trace->read) + return -EOPNOTSUPP; - error = ds_access_bts(child, bts_index, &bts_record); + error = trace->read(child->bts, at, &bts); if (error < 0) return error; - ptrace_bts_translate_record(&ret, bts_record); - - if (copy_to_user(out, &ret, sizeof(ret))) + if (copy_to_user(out, &bts, sizeof(bts))) return -EFAULT; - return sizeof(ret); + return sizeof(bts); } static int ptrace_bts_drain(struct task_struct *child, long size, struct bts_struct __user *out) { - struct bts_struct ret; - const unsigned char *raw; - size_t end, i; - int error; + const struct bts_trace *trace; + const unsigned char *at; + int error, drained = 0; - error = ds_get_bts_index(child, &end); - if (error < 0) - return error; + trace = ds_read_bts(child->bts); + if (!trace) + return -EPERM; - if (size < (end * sizeof(struct bts_struct))) + if (!trace->read) + return -EOPNOTSUPP; + + if (size < (trace->ds.top - trace->ds.begin)) return -EIO; - error = ds_access_bts(child, 0, (const void **)&raw); - if (error < 0) - return error; + for (at = trace->ds.begin; (void *)at < trace->ds.top; + out++, drained++, at += trace->ds.size) { + struct bts_struct bts; + int error; - for (i = 0; i < end; i++, out++, raw += bts_cfg.sizeof_bts) { - ptrace_bts_translate_record(&ret, raw); + error = trace->read(child->bts, at, &bts); + if (error < 0) + return error; - if (copy_to_user(out, &ret, sizeof(ret))) + if (copy_to_user(out, &bts, sizeof(bts))) return -EFAULT; } - error = ds_clear_bts(child); + memset(trace->ds.begin, 0, trace->ds.n * trace->ds.size); + + error = ds_reset_bts(child->bts); if (error < 0) return error; - return end; + return drained; } -static void ptrace_bts_ovfl(struct task_struct *child) +static int ptrace_bts_allocate_buffer(struct task_struct *child, size_t size) { - send_sig(child->thread.bts_ovfl_signal, child, 0); + child->bts_buffer = alloc_locked_buffer(size); + if (!child->bts_buffer) + return -ENOMEM; + + child->bts_size = size; + + return 0; +} + +static void ptrace_bts_free_buffer(struct task_struct *child) +{ + free_locked_buffer(child->bts_buffer, child->bts_size); + child->bts_buffer = NULL; + child->bts_size = 0; } static int ptrace_bts_config(struct task_struct *child, @@ -740,114 +673,86 @@ static int ptrace_bts_config(struct task_struct *child, const struct ptrace_bts_config __user *ucfg) { struct ptrace_bts_config cfg; - int error = 0; - - error = -EOPNOTSUPP; - if (!bts_cfg.sizeof_bts) - goto errout; + unsigned int flags = 0; - error = -EIO; if (cfg_size < sizeof(cfg)) - goto errout; + return -EIO; - error = -EFAULT; if (copy_from_user(&cfg, ucfg, sizeof(cfg))) - goto errout; + return -EFAULT; - error = -EINVAL; - if ((cfg.flags & PTRACE_BTS_O_SIGNAL) && - !(cfg.flags & PTRACE_BTS_O_ALLOC)) - goto errout; + if (child->bts) { + ds_release_bts(child->bts); + child->bts = NULL; + } - if (cfg.flags & PTRACE_BTS_O_ALLOC) { - ds_ovfl_callback_t ovfl = NULL; - unsigned int sig = 0; + if (cfg.flags & PTRACE_BTS_O_SIGNAL) { + if (!cfg.signal) + return -EINVAL; - /* we ignore the error in case we were not tracing child */ - (void)ds_release_bts(child); + return -EOPNOTSUPP; - if (cfg.flags & PTRACE_BTS_O_SIGNAL) { - if (!cfg.signal) - goto errout; + child->thread.bts_ovfl_signal = cfg.signal; + } - sig = cfg.signal; - ovfl = ptrace_bts_ovfl; - } + if ((cfg.flags & PTRACE_BTS_O_ALLOC) && + (cfg.size != child->bts_size)) { + int error; - error = ds_request_bts(child, /* base = */ NULL, cfg.size, ovfl); - if (error < 0) - goto errout; + ptrace_bts_free_buffer(child); - child->thread.bts_ovfl_signal = sig; + error = ptrace_bts_allocate_buffer(child, cfg.size); + if (error < 0) + return error; } - error = -EINVAL; - if (!child->thread.ds_ctx && cfg.flags) - goto errout; - if (cfg.flags & PTRACE_BTS_O_TRACE) - child->thread.debugctlmsr |= bts_cfg.debugctl_mask; - else - child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask; + flags |= BTS_USER; if (cfg.flags & PTRACE_BTS_O_SCHED) - set_tsk_thread_flag(child, TIF_BTS_TRACE_TS); - else - clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS); + flags |= BTS_TIMESTAMPS; - error = sizeof(cfg); + child->bts = ds_request_bts(child, child->bts_buffer, child->bts_size, + /* ovfl = */ NULL, /* th = */ (size_t)-1, + flags); + if (IS_ERR(child->bts)) { + int error = PTR_ERR(child->bts); -out: - if (child->thread.debugctlmsr) - set_tsk_thread_flag(child, TIF_DEBUGCTLMSR); - else - clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR); + ptrace_bts_free_buffer(child); + child->bts = NULL; - return error; + return error; + } -errout: - child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask; - clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS); - goto out; + return sizeof(cfg); } static int ptrace_bts_status(struct task_struct *child, long cfg_size, struct ptrace_bts_config __user *ucfg) { + const struct bts_trace *trace; struct ptrace_bts_config cfg; - size_t end; - const void *base, *max; - int error; if (cfg_size < sizeof(cfg)) return -EIO; - error = ds_get_bts_end(child, &end); - if (error < 0) - return error; - - error = ds_access_bts(child, /* index = */ 0, &base); - if (error < 0) - return error; - - error = ds_access_bts(child, /* index = */ end, &max); - if (error < 0) - return error; + trace = ds_read_bts(child->bts); + if (!trace) + return -EPERM; memset(&cfg, 0, sizeof(cfg)); - cfg.size = (max - base); + cfg.size = trace->ds.end - trace->ds.begin; cfg.signal = child->thread.bts_ovfl_signal; cfg.bts_size = sizeof(struct bts_struct); if (cfg.signal) cfg.flags |= PTRACE_BTS_O_SIGNAL; - if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) && - child->thread.debugctlmsr & bts_cfg.debugctl_mask) + if (trace->ds.flags & BTS_USER) cfg.flags |= PTRACE_BTS_O_TRACE; - if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS)) + if (trace->ds.flags & BTS_TIMESTAMPS) cfg.flags |= PTRACE_BTS_O_SCHED; if (copy_to_user(ucfg, &cfg, sizeof(cfg))) @@ -856,110 +761,77 @@ static int ptrace_bts_status(struct task_struct *child, return sizeof(cfg); } -static int ptrace_bts_write_record(struct task_struct *child, - const struct bts_struct *in) +static int ptrace_bts_clear(struct task_struct *child) { - unsigned char bts_record[BTS_MAX_RECORD_SIZE]; + const struct bts_trace *trace; - BUG_ON(BTS_MAX_RECORD_SIZE < bts_cfg.sizeof_bts); + trace = ds_read_bts(child->bts); + if (!trace) + return -EPERM; - memset(bts_record, 0, bts_cfg.sizeof_bts); - switch (in->qualifier) { - case BTS_INVALID: - break; + memset(trace->ds.begin, 0, trace->ds.n * trace->ds.size); - case BTS_BRANCH: - bts_set(bts_record, bts_from, in->variant.lbr.from_ip); - bts_set(bts_record, bts_to, in->variant.lbr.to_ip); - break; + return ds_reset_bts(child->bts); +} - case BTS_TASK_ARRIVES: - case BTS_TASK_DEPARTS: - bts_set(bts_record, bts_from, bts_escape); - bts_set(bts_record, bts_qual, in->qualifier); - bts_set(bts_record, bts_jiffies, in->variant.jiffies); - break; +static int ptrace_bts_size(struct task_struct *child) +{ + const struct bts_trace *trace; - default: - return -EINVAL; - } + trace = ds_read_bts(child->bts); + if (!trace) + return -EPERM; - /* The writing task will be the switched-to task on a context - * switch. It needs to write into the switched-from task's BTS - * buffer. */ - return ds_unchecked_write_bts(child, bts_record, bts_cfg.sizeof_bts); + return (trace->ds.top - trace->ds.begin) / trace->ds.size; } -void ptrace_bts_take_timestamp(struct task_struct *tsk, - enum bts_qualifier qualifier) +static void ptrace_bts_fork(struct task_struct *tsk) { - struct bts_struct rec = { - .qualifier = qualifier, - .variant.jiffies = jiffies_64 - }; - - ptrace_bts_write_record(tsk, &rec); + tsk->bts = NULL; + tsk->bts_buffer = NULL; + tsk->bts_size = 0; + tsk->thread.bts_ovfl_signal = 0; } -static const struct bts_configuration bts_cfg_netburst = { - .sizeof_bts = sizeof(long) * 3, - .sizeof_field = sizeof(long), - .debugctl_mask = (1<<2)|(1<<3)|(1<<5) -}; +static void ptrace_bts_untrace(struct task_struct *child) +{ + if (unlikely(child->bts)) { + ds_release_bts(child->bts); + child->bts = NULL; + + /* We cannot update total_vm and locked_vm since + child's mm is already gone. But we can reclaim the + memory. */ + kfree(child->bts_buffer); + child->bts_buffer = NULL; + child->bts_size = 0; + } +} -static const struct bts_configuration bts_cfg_pentium_m = { - .sizeof_bts = sizeof(long) * 3, - .sizeof_field = sizeof(long), - .debugctl_mask = (1<<6)|(1<<7) -}; +static void ptrace_bts_detach(struct task_struct *child) +{ + if (unlikely(child->bts)) { + ds_release_bts(child->bts); + child->bts = NULL; -static const struct bts_configuration bts_cfg_core2 = { - .sizeof_bts = 8 * 3, - .sizeof_field = 8, - .debugctl_mask = (1<<6)|(1<<7)|(1<<9) -}; + ptrace_bts_free_buffer(child); + } +} +#else +static inline void ptrace_bts_fork(struct task_struct *tsk) {} +static inline void ptrace_bts_detach(struct task_struct *child) {} +static inline void ptrace_bts_untrace(struct task_struct *child) {} +#endif /* CONFIG_X86_PTRACE_BTS */ -static inline void bts_configure(const struct bts_configuration *cfg) +void x86_ptrace_fork(struct task_struct *child, unsigned long clone_flags) { - bts_cfg = *cfg; + ptrace_bts_fork(child); } -void __cpuinit ptrace_bts_init_intel(struct cpuinfo_x86 *c) +void x86_ptrace_untrace(struct task_struct *child) { - switch (c->x86) { - case 0x6: - switch (c->x86_model) { - case 0xD: - case 0xE: /* Pentium M */ - bts_configure(&bts_cfg_pentium_m); - break; - case 0xF: /* Core2 */ - case 0x1C: /* Atom */ - bts_configure(&bts_cfg_core2); - break; - default: - /* sorry, don't know about them */ - break; - } - break; - case 0xF: - switch (c->x86_model) { - case 0x0: - case 0x1: - case 0x2: /* Netburst */ - bts_configure(&bts_cfg_netburst); - break; - default: - /* sorry, don't know about them */ - break; - } - break; - default: - /* sorry, don't know about them */ - break; - } + ptrace_bts_untrace(child); } -#endif /* CONFIG_X86_PTRACE_BTS */ /* * Called by kernel/ptrace.c when detaching.. @@ -972,15 +844,7 @@ void ptrace_disable(struct task_struct *child) #ifdef TIF_SYSCALL_EMU clear_tsk_thread_flag(child, TIF_SYSCALL_EMU); #endif -#ifdef CONFIG_X86_PTRACE_BTS - (void)ds_release_bts(child); - - child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask; - if (!child->thread.debugctlmsr) - clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR); - - clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS); -#endif /* CONFIG_X86_PTRACE_BTS */ + ptrace_bts_detach(child); } #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION @@ -1112,7 +976,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) break; case PTRACE_BTS_SIZE: - ret = ds_get_bts_index(child, /* pos = */ NULL); + ret = ptrace_bts_size(child); break; case PTRACE_BTS_GET: @@ -1121,7 +985,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) break; case PTRACE_BTS_CLEAR: - ret = ds_clear_bts(child); + ret = ptrace_bts_clear(child); break; case PTRACE_BTS_DRAIN: @@ -1384,6 +1248,14 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, case PTRACE_GET_THREAD_AREA: case PTRACE_SET_THREAD_AREA: +#ifdef CONFIG_X86_PTRACE_BTS + case PTRACE_BTS_CONFIG: + case PTRACE_BTS_STATUS: + case PTRACE_BTS_SIZE: + case PTRACE_BTS_GET: + case PTRACE_BTS_CLEAR: + case PTRACE_BTS_DRAIN: +#endif /* CONFIG_X86_PTRACE_BTS */ return arch_ptrace(child, request, addr, data); default: diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index 05fbe9a0325a..4f9c55f3a7c0 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -97,6 +97,18 @@ static unsigned pvclock_get_time_values(struct pvclock_shadow_time *dst, return dst->version; } +unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src) +{ + u64 pv_tsc_khz = 1000000ULL << 32; + + do_div(pv_tsc_khz, src->tsc_to_system_mul); + if (src->tsc_shift < 0) + pv_tsc_khz <<= -src->tsc_shift; + else + pv_tsc_khz >>= src->tsc_shift; + return pv_tsc_khz; +} + cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src) { struct pvclock_shadow_time shadow; diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c index f6a11b9b1f98..309949e9e1c1 100644 --- a/arch/x86/kernel/quirks.c +++ b/arch/x86/kernel/quirks.c @@ -35,9 +35,6 @@ static void __devinit quirk_intel_irqbalance(struct pci_dev *dev) if (!(word & (1 << 13))) { dev_info(&dev->dev, "Intel E7520/7320/7525 detected; " "disabling irq balancing and affinity\n"); -#ifdef CONFIG_IRQBALANCE - irqbalance_disable(""); -#endif noirqdebug_setup(""); #ifdef CONFIG_PROC_FS no_irq_affinity = 1; @@ -171,6 +168,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_31, ich_force_enable_hpet); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_1, ich_force_enable_hpet); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_4, + ich_force_enable_hpet); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH9_7, ich_force_enable_hpet); diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index f4c93f1cfc19..61f718df6eec 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -21,6 +21,9 @@ # include <asm/iommu.h> #endif +#include <mach_ipi.h> + + /* * Power off function, if any */ @@ -29,18 +32,17 @@ EXPORT_SYMBOL(pm_power_off); static const struct desc_ptr no_idt = {}; static int reboot_mode; -/* - * Keyboard reset and triple fault may result in INIT, not RESET, which - * doesn't work when we're in vmx root mode. Try ACPI first. - */ -enum reboot_type reboot_type = BOOT_ACPI; +enum reboot_type reboot_type = BOOT_KBD; int reboot_force; #if defined(CONFIG_X86_32) && defined(CONFIG_SMP) static int reboot_cpu = -1; #endif -/* reboot=b[ios] | s[mp] | t[riple] | k[bd] | e[fi] [, [w]arm | [c]old] +/* This is set by the PCI code if either type 1 or type 2 PCI is detected */ +bool port_cf9_safe = false; + +/* reboot=b[ios] | s[mp] | t[riple] | k[bd] | e[fi] [, [w]arm | [c]old] | p[ci] warm Don't set the cold reboot flag cold Set the cold reboot flag bios Reboot by jumping through the BIOS (only for X86_32) @@ -49,6 +51,7 @@ static int reboot_cpu = -1; kbd Use the keyboard controller. cold reset (default) acpi Use the RESET_REG in the FADT efi Use efi reset_system runtime service + pci Use the so-called "PCI reset register", CF9 force Avoid anything that could hang. */ static int __init reboot_setup(char *str) @@ -83,6 +86,7 @@ static int __init reboot_setup(char *str) case 'k': case 't': case 'e': + case 'p': reboot_type = *str; break; @@ -173,6 +177,15 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_BOARD_NAME, "0KW626"), }, }, + { /* Handle problems with rebooting on Dell Optiplex 330 with 0KP561 */ + .callback = set_bios_reboot, + .ident = "Dell OptiPlex 330", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 330"), + DMI_MATCH(DMI_BOARD_NAME, "0KP561"), + }, + }, { /* Handle problems with rebooting on Dell 2400's */ .callback = set_bios_reboot, .ident = "Dell PowerEdge 2400", @@ -399,12 +412,27 @@ static void native_machine_emergency_restart(void) reboot_type = BOOT_KBD; break; - case BOOT_EFI: if (efi_enabled) - efi.reset_system(reboot_mode ? EFI_RESET_WARM : EFI_RESET_COLD, + efi.reset_system(reboot_mode ? + EFI_RESET_WARM : + EFI_RESET_COLD, EFI_SUCCESS, 0, NULL); + reboot_type = BOOT_KBD; + break; + + case BOOT_CF9: + port_cf9_safe = true; + /* fall through */ + case BOOT_CF9_COND: + if (port_cf9_safe) { + u8 cf9 = inb(0xcf9) & ~6; + outb(cf9|2, 0xcf9); /* Request hard reset */ + udelay(50); + outb(cf9|6, 0xcf9); /* Actually do the reset */ + udelay(50); + } reboot_type = BOOT_KBD; break; } @@ -465,6 +493,11 @@ static void native_machine_restart(char *__unused) static void native_machine_halt(void) { + /* stop other cpus and apics */ + machine_shutdown(); + + /* stop this cpu */ + stop_this_cpu(NULL); } static void native_machine_power_off(void) @@ -518,3 +551,95 @@ void machine_crash_shutdown(struct pt_regs *regs) machine_ops.crash_shutdown(regs); } #endif + + +#if defined(CONFIG_SMP) + +/* This keeps a track of which one is crashing cpu. */ +static int crashing_cpu; +static nmi_shootdown_cb shootdown_callback; + +static atomic_t waiting_for_crash_ipi; + +static int crash_nmi_callback(struct notifier_block *self, + unsigned long val, void *data) +{ + int cpu; + + if (val != DIE_NMI_IPI) + return NOTIFY_OK; + + cpu = raw_smp_processor_id(); + + /* Don't do anything if this handler is invoked on crashing cpu. + * Otherwise, system will completely hang. Crashing cpu can get + * an NMI if system was initially booted with nmi_watchdog parameter. + */ + if (cpu == crashing_cpu) + return NOTIFY_STOP; + local_irq_disable(); + + shootdown_callback(cpu, (struct die_args *)data); + + atomic_dec(&waiting_for_crash_ipi); + /* Assume hlt works */ + halt(); + for (;;) + cpu_relax(); + + return 1; +} + +static void smp_send_nmi_allbutself(void) +{ + cpumask_t mask = cpu_online_map; + cpu_clear(safe_smp_processor_id(), mask); + if (!cpus_empty(mask)) + send_IPI_mask(mask, NMI_VECTOR); +} + +static struct notifier_block crash_nmi_nb = { + .notifier_call = crash_nmi_callback, +}; + +/* Halt all other CPUs, calling the specified function on each of them + * + * This function can be used to halt all other CPUs on crash + * or emergency reboot time. The function passed as parameter + * will be called inside a NMI handler on all CPUs. + */ +void nmi_shootdown_cpus(nmi_shootdown_cb callback) +{ + unsigned long msecs; + local_irq_disable(); + + /* Make a note of crashing cpu. Will be used in NMI callback.*/ + crashing_cpu = safe_smp_processor_id(); + + shootdown_callback = callback; + + atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1); + /* Would it be better to replace the trap vector here? */ + if (register_die_notifier(&crash_nmi_nb)) + return; /* return what? */ + /* Ensure the new callback function is set before sending + * out the NMI + */ + wmb(); + + smp_send_nmi_allbutself(); + + msecs = 1000; /* Wait at most a second for the other cpus to stop */ + while ((atomic_read(&waiting_for_crash_ipi) > 0) && msecs) { + mdelay(1); + msecs--; + } + + /* Leave the nmi callback set */ +} +#else /* !CONFIG_SMP */ +void nmi_shootdown_cpus(nmi_shootdown_cb callback) +{ + /* No other CPUs to shoot down */ +} +#endif diff --git a/arch/x86/kernel/relocate_kernel_32.S b/arch/x86/kernel/relocate_kernel_32.S index 6f50664b2ba5..a160f3119725 100644 --- a/arch/x86/kernel/relocate_kernel_32.S +++ b/arch/x86/kernel/relocate_kernel_32.S @@ -10,15 +10,12 @@ #include <asm/page.h> #include <asm/kexec.h> #include <asm/processor-flags.h> -#include <asm/pgtable.h> /* * Must be relocatable PIC code callable as a C function */ #define PTR(x) (x << 2) -#define PAGE_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY) -#define PAE_PGD_ATTR (_PAGE_PRESENT) /* control_page + KEXEC_CONTROL_CODE_MAX_SIZE * ~ control_page + PAGE_SIZE are used as data storage and stack for @@ -39,7 +36,6 @@ #define CP_PA_BACKUP_PAGES_MAP DATA(0x1c) .text - .align PAGE_SIZE .globl relocate_kernel relocate_kernel: /* Save the CPU context, used for jumping back */ @@ -60,117 +56,6 @@ relocate_kernel: movl %cr4, %eax movl %eax, CR4(%edi) -#ifdef CONFIG_X86_PAE - /* map the control page at its virtual address */ - - movl PTR(VA_PGD)(%ebp), %edi - movl PTR(VA_CONTROL_PAGE)(%ebp), %eax - andl $0xc0000000, %eax - shrl $27, %eax - addl %edi, %eax - - movl PTR(PA_PMD_0)(%ebp), %edx - orl $PAE_PGD_ATTR, %edx - movl %edx, (%eax) - - movl PTR(VA_PMD_0)(%ebp), %edi - movl PTR(VA_CONTROL_PAGE)(%ebp), %eax - andl $0x3fe00000, %eax - shrl $18, %eax - addl %edi, %eax - - movl PTR(PA_PTE_0)(%ebp), %edx - orl $PAGE_ATTR, %edx - movl %edx, (%eax) - - movl PTR(VA_PTE_0)(%ebp), %edi - movl PTR(VA_CONTROL_PAGE)(%ebp), %eax - andl $0x001ff000, %eax - shrl $9, %eax - addl %edi, %eax - - movl PTR(PA_CONTROL_PAGE)(%ebp), %edx - orl $PAGE_ATTR, %edx - movl %edx, (%eax) - - /* identity map the control page at its physical address */ - - movl PTR(VA_PGD)(%ebp), %edi - movl PTR(PA_CONTROL_PAGE)(%ebp), %eax - andl $0xc0000000, %eax - shrl $27, %eax - addl %edi, %eax - - movl PTR(PA_PMD_1)(%ebp), %edx - orl $PAE_PGD_ATTR, %edx - movl %edx, (%eax) - - movl PTR(VA_PMD_1)(%ebp), %edi - movl PTR(PA_CONTROL_PAGE)(%ebp), %eax - andl $0x3fe00000, %eax - shrl $18, %eax - addl %edi, %eax - - movl PTR(PA_PTE_1)(%ebp), %edx - orl $PAGE_ATTR, %edx - movl %edx, (%eax) - - movl PTR(VA_PTE_1)(%ebp), %edi - movl PTR(PA_CONTROL_PAGE)(%ebp), %eax - andl $0x001ff000, %eax - shrl $9, %eax - addl %edi, %eax - - movl PTR(PA_CONTROL_PAGE)(%ebp), %edx - orl $PAGE_ATTR, %edx - movl %edx, (%eax) -#else - /* map the control page at its virtual address */ - - movl PTR(VA_PGD)(%ebp), %edi - movl PTR(VA_CONTROL_PAGE)(%ebp), %eax - andl $0xffc00000, %eax - shrl $20, %eax - addl %edi, %eax - - movl PTR(PA_PTE_0)(%ebp), %edx - orl $PAGE_ATTR, %edx - movl %edx, (%eax) - - movl PTR(VA_PTE_0)(%ebp), %edi - movl PTR(VA_CONTROL_PAGE)(%ebp), %eax - andl $0x003ff000, %eax - shrl $10, %eax - addl %edi, %eax - - movl PTR(PA_CONTROL_PAGE)(%ebp), %edx - orl $PAGE_ATTR, %edx - movl %edx, (%eax) - - /* identity map the control page at its physical address */ - - movl PTR(VA_PGD)(%ebp), %edi - movl PTR(PA_CONTROL_PAGE)(%ebp), %eax - andl $0xffc00000, %eax - shrl $20, %eax - addl %edi, %eax - - movl PTR(PA_PTE_1)(%ebp), %edx - orl $PAGE_ATTR, %edx - movl %edx, (%eax) - - movl PTR(VA_PTE_1)(%ebp), %edi - movl PTR(PA_CONTROL_PAGE)(%ebp), %eax - andl $0x003ff000, %eax - shrl $10, %eax - addl %edi, %eax - - movl PTR(PA_CONTROL_PAGE)(%ebp), %edx - orl $PAGE_ATTR, %edx - movl %edx, (%eax) -#endif - -relocate_new_kernel: /* read the arguments and say goodbye to the stack */ movl 20+4(%esp), %ebx /* page_list */ movl 20+8(%esp), %ebp /* list of pages */ diff --git a/arch/x86/kernel/rtc.c b/arch/x86/kernel/rtc.c index 0a23b5795b25..dd6f2b71561b 100644 --- a/arch/x86/kernel/rtc.c +++ b/arch/x86/kernel/rtc.c @@ -52,7 +52,7 @@ int mach_set_rtc_mmss(unsigned long nowtime) cmos_minutes = CMOS_READ(RTC_MINUTES); if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD) - BCD_TO_BIN(cmos_minutes); + cmos_minutes = bcd2bin(cmos_minutes); /* * since we're only adjusting minutes and seconds, @@ -69,8 +69,8 @@ int mach_set_rtc_mmss(unsigned long nowtime) if (abs(real_minutes - cmos_minutes) < 30) { if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD) { - BIN_TO_BCD(real_seconds); - BIN_TO_BCD(real_minutes); + real_seconds = bin2bcd(real_seconds); + real_minutes = bin2bcd(real_minutes); } CMOS_WRITE(real_seconds,RTC_SECONDS); CMOS_WRITE(real_minutes,RTC_MINUTES); @@ -124,16 +124,16 @@ unsigned long mach_get_cmos_time(void) WARN_ON_ONCE(RTC_ALWAYS_BCD && (status & RTC_DM_BINARY)); if (RTC_ALWAYS_BCD || !(status & RTC_DM_BINARY)) { - BCD_TO_BIN(sec); - BCD_TO_BIN(min); - BCD_TO_BIN(hour); - BCD_TO_BIN(day); - BCD_TO_BIN(mon); - BCD_TO_BIN(year); + sec = bcd2bin(sec); + min = bcd2bin(min); + hour = bcd2bin(hour); + day = bcd2bin(day); + mon = bcd2bin(mon); + year = bcd2bin(year); } if (century) { - BCD_TO_BIN(century); + century = bcd2bin(century); year += century * 100; printk(KERN_INFO "Extended CMOS year: %d\n", century * 100); } else diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 2255782e8d4b..ae0d8042cf69 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -93,11 +93,13 @@ #include <asm/desc.h> #include <asm/dma.h> #include <asm/iommu.h> +#include <asm/gart.h> #include <asm/mmu_context.h> #include <asm/proto.h> #include <mach_apic.h> #include <asm/paravirt.h> +#include <asm/hypervisor.h> #include <asm/percpu.h> #include <asm/topology.h> @@ -448,6 +450,7 @@ static void __init reserve_early_setup_data(void) * @size: Size of the crashkernel memory to reserve. * Returns the base address on success, and -1ULL on failure. */ +static unsigned long long __init find_and_reserve_crashkernel(unsigned long long size) { const unsigned long long alignment = 16<<20; /* 16M */ @@ -561,7 +564,13 @@ static void __init reserve_standard_io_resources(void) } -#ifdef CONFIG_PROC_VMCORE +/* + * Note: elfcorehdr_addr is not just limited to vmcore. It is also used by + * is_kdump_kernel() to determine if we are booting after a panic. Hence + * ifdef it under CONFIG_CRASH_DUMP and not CONFIG_PROC_VMCORE. + */ + +#ifdef CONFIG_CRASH_DUMP /* elfcorehdr= specifies the location of elf core header * stored by the crashed kernel. This option will be passed * by kexec loader to the capture kernel. @@ -577,161 +586,24 @@ static int __init setup_elfcorehdr(char *arg) early_param("elfcorehdr", setup_elfcorehdr); #endif -static struct x86_quirks default_x86_quirks __initdata; - -struct x86_quirks *x86_quirks __initdata = &default_x86_quirks; - -/* - * Some BIOSes seem to corrupt the low 64k of memory during events - * like suspend/resume and unplugging an HDMI cable. Reserve all - * remaining free memory in that area and fill it with a distinct - * pattern. - */ -#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION -#define MAX_SCAN_AREAS 8 - -static int __read_mostly memory_corruption_check = -1; - -static unsigned __read_mostly corruption_check_size = 64*1024; -static unsigned __read_mostly corruption_check_period = 60; /* seconds */ - -static struct e820entry scan_areas[MAX_SCAN_AREAS]; -static int num_scan_areas; - - -static int set_corruption_check(char *arg) +static int __init default_update_genapic(void) { - char *end; - - memory_corruption_check = simple_strtol(arg, &end, 10); - - return (*end == 0) ? 0 : -EINVAL; -} -early_param("memory_corruption_check", set_corruption_check); - -static int set_corruption_check_period(char *arg) -{ - char *end; - - corruption_check_period = simple_strtoul(arg, &end, 10); - - return (*end == 0) ? 0 : -EINVAL; -} -early_param("memory_corruption_check_period", set_corruption_check_period); - -static int set_corruption_check_size(char *arg) -{ - char *end; - unsigned size; - - size = memparse(arg, &end); - - if (*end == '\0') - corruption_check_size = size; - - return (size == corruption_check_size) ? 0 : -EINVAL; -} -early_param("memory_corruption_check_size", set_corruption_check_size); - - -static void __init setup_bios_corruption_check(void) -{ - u64 addr = PAGE_SIZE; /* assume first page is reserved anyway */ - - if (memory_corruption_check == -1) { - memory_corruption_check = -#ifdef CONFIG_X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK - 1 -#else - 0 +#ifdef CONFIG_X86_SMP +# if defined(CONFIG_X86_GENERICARCH) || defined(CONFIG_X86_64) + genapic->wakeup_cpu = wakeup_secondary_cpu_via_init; +# endif #endif - ; - } - - if (corruption_check_size == 0) - memory_corruption_check = 0; - - if (!memory_corruption_check) - return; - - corruption_check_size = round_up(corruption_check_size, PAGE_SIZE); - - while(addr < corruption_check_size && num_scan_areas < MAX_SCAN_AREAS) { - u64 size; - addr = find_e820_area_size(addr, &size, PAGE_SIZE); - - if (addr == 0) - break; - - if ((addr + size) > corruption_check_size) - size = corruption_check_size - addr; - - if (size == 0) - break; - - e820_update_range(addr, size, E820_RAM, E820_RESERVED); - scan_areas[num_scan_areas].addr = addr; - scan_areas[num_scan_areas].size = size; - num_scan_areas++; - - /* Assume we've already mapped this early memory */ - memset(__va(addr), 0, size); - - addr += size; - } - - printk(KERN_INFO "Scanning %d areas for low memory corruption\n", - num_scan_areas); - update_e820(); -} - -static struct timer_list periodic_check_timer; -void check_for_bios_corruption(void) -{ - int i; - int corruption = 0; - - if (!memory_corruption_check) - return; - - for(i = 0; i < num_scan_areas; i++) { - unsigned long *addr = __va(scan_areas[i].addr); - unsigned long size = scan_areas[i].size; - - for(; size; addr++, size -= sizeof(unsigned long)) { - if (!*addr) - continue; - printk(KERN_ERR "Corrupted low memory at %p (%lx phys) = %08lx\n", - addr, __pa(addr), *addr); - corruption = 1; - *addr = 0; - } - } - - WARN(corruption, KERN_ERR "Memory corruption detected in low memory\n"); -} - -static void periodic_check_for_corruption(unsigned long data) -{ - check_for_bios_corruption(); - mod_timer(&periodic_check_timer, round_jiffies(jiffies + corruption_check_period*HZ)); + return 0; } -void start_periodic_check_for_corruption(void) -{ - if (!memory_corruption_check || corruption_check_period == 0) - return; - - printk(KERN_INFO "Scanning for low memory corruption every %d seconds\n", - corruption_check_period); +static struct x86_quirks default_x86_quirks __initdata = { + .update_genapic = default_update_genapic, +}; - init_timer(&periodic_check_timer); - periodic_check_timer.function = &periodic_check_for_corruption; - periodic_check_for_corruption(0); -} -#endif +struct x86_quirks *x86_quirks __initdata = &default_x86_quirks; +#ifdef CONFIG_X86_RESERVE_LOW_64K static int __init dmi_low_memory_corruption(const struct dmi_system_id *d) { printk(KERN_NOTICE @@ -743,6 +615,7 @@ static int __init dmi_low_memory_corruption(const struct dmi_system_id *d) return 0; } +#endif /* List of systems that have known low memory corruption BIOS problems */ static struct dmi_system_id __initdata bad_bios_dmi_table[] = { @@ -758,7 +631,7 @@ static struct dmi_system_id __initdata bad_bios_dmi_table[] = { .callback = dmi_low_memory_corruption, .ident = "Phoenix BIOS", .matches = { - DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies, LTD"), + DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies"), }, }, #endif @@ -788,6 +661,9 @@ void __init setup_arch(char **cmdline_p) printk(KERN_INFO "Command line: %s\n", boot_command_line); #endif + /* VMI may relocate the fixmap; do this before touching ioremap area */ + vmi_init(); + early_cpu_init(); early_ioremap_init(); @@ -874,13 +750,8 @@ void __init setup_arch(char **cmdline_p) check_efer(); #endif -#if defined(CONFIG_VMI) && defined(CONFIG_X86_32) - /* - * Must be before kernel pagetables are setup - * or fixmap area is touched. - */ - vmi_init(); -#endif + /* Must be before kernel pagetables are setup */ + vmi_activate(); /* after early param, so could get panic from serial */ reserve_early_setup_data(); @@ -903,6 +774,12 @@ void __init setup_arch(char **cmdline_p) dmi_check_system(bad_bios_dmi_table); + /* + * VMware detection requires dmi to be available, so this + * needs to be done after dmi_scan_machine, for the BP. + */ + init_hypervisor(&boot_cpu_data); + #ifdef CONFIG_X86_32 probe_roms(); #endif @@ -1067,6 +944,7 @@ void __init setup_arch(char **cmdline_p) #endif prefill_possible_map(); + #ifdef CONFIG_X86_64 init_cpu_to_node(); #endif @@ -1074,6 +952,9 @@ void __init setup_arch(char **cmdline_p) init_apic_mappings(); ioapic_init_mappings(); + /* need to wait for io_apic is mapped */ + probe_nr_irqs_gsi(); + kvm_guest_init(); e820_reserve_resources(); diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 0e67f72d9316..ae0c0d3bb770 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -140,25 +140,30 @@ static void __init setup_cpu_pda_map(void) */ void __init setup_per_cpu_areas(void) { - ssize_t size = PERCPU_ENOUGH_ROOM; + ssize_t size, old_size; char *ptr; int cpu; + unsigned long align = 1; /* Setup cpu_pda map */ setup_cpu_pda_map(); /* Copy section for each CPU (we discard the original) */ - size = PERCPU_ENOUGH_ROOM; + old_size = PERCPU_ENOUGH_ROOM; + align = max_t(unsigned long, PAGE_SIZE, align); + size = roundup(old_size, align); printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n", size); for_each_possible_cpu(cpu) { #ifndef CONFIG_NEED_MULTIPLE_NODES - ptr = alloc_bootmem_pages(size); + ptr = __alloc_bootmem(size, align, + __pa(MAX_DMA_ADDRESS)); #else int node = early_cpu_to_node(cpu); if (!node_online(node) || !NODE_DATA(node)) { - ptr = alloc_bootmem_pages(size); + ptr = __alloc_bootmem(size, align, + __pa(MAX_DMA_ADDRESS)); printk(KERN_INFO "cpu %d has no node %d or node-local memory\n", cpu, node); @@ -167,7 +172,8 @@ void __init setup_per_cpu_areas(void) cpu, __pa(ptr)); } else { - ptr = alloc_bootmem_pages_node(NODE_DATA(node), size); + ptr = __alloc_bootmem_node(NODE_DATA(node), size, align, + __pa(MAX_DMA_ADDRESS)); if (ptr) printk(KERN_DEBUG "per cpu data for cpu%d on node%d at %016lx\n", cpu, node, __pa(ptr)); @@ -175,7 +181,6 @@ void __init setup_per_cpu_areas(void) #endif per_cpu_offset(cpu) = ptr - __per_cpu_start; memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start); - } printk(KERN_DEBUG "NR_CPUS: %d, nr_cpu_ids: %d, nr_node_ids %d\n", @@ -213,7 +218,7 @@ static void __init setup_node_to_cpumask_map(void) /* allocate the map */ map = alloc_bootmem_low(nr_node_ids * sizeof(cpumask_t)); - pr_debug(KERN_DEBUG "Node to cpumask map at %p for %d nodes\n", + pr_debug("Node to cpumask map at %p for %d nodes\n", map, nr_node_ids); /* node_to_cpumask() will now work */ diff --git a/arch/x86/kernel/sigframe.h b/arch/x86/kernel/sigframe.h deleted file mode 100644 index cc673aa55ce4..000000000000 --- a/arch/x86/kernel/sigframe.h +++ /dev/null @@ -1,42 +0,0 @@ -#ifdef CONFIG_X86_32 -struct sigframe { - char __user *pretcode; - int sig; - struct sigcontext sc; - /* - * fpstate is unused. fpstate is moved/allocated after - * retcode[] below. This movement allows to have the FP state and the - * future state extensions (xsave) stay together. - * And at the same time retaining the unused fpstate, prevents changing - * the offset of extramask[] in the sigframe and thus prevent any - * legacy application accessing/modifying it. - */ - struct _fpstate fpstate_unused; - unsigned long extramask[_NSIG_WORDS-1]; - char retcode[8]; - /* fp state follows here */ -}; - -struct rt_sigframe { - char __user *pretcode; - int sig; - struct siginfo __user *pinfo; - void __user *puc; - struct siginfo info; - struct ucontext uc; - char retcode[8]; - /* fp state follows here */ -}; -#else -struct rt_sigframe { - char __user *pretcode; - struct ucontext uc; - struct siginfo info; - /* fp state follows here */ -}; - -int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, - sigset_t *set, struct pt_regs *regs); -int ia32_setup_frame(int sig, struct k_sigaction *ka, - sigset_t *set, struct pt_regs *regs); -#endif diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal.c index d6dd057d0f22..89bb7668041d 100644 --- a/arch/x86/kernel/signal_32.c +++ b/arch/x86/kernel/signal.c @@ -1,36 +1,41 @@ /* * Copyright (C) 1991, 1992 Linus Torvalds + * Copyright (C) 2000, 2001, 2002 Andi Kleen SuSE Labs * * 1997-11-28 Modified for POSIX.1b signals by Richard Henderson * 2000-06-20 Pentium III FXSR, SSE support by Gareth Hughes + * 2000-2002 x86-64 support by Andi Kleen */ -#include <linux/list.h> -#include <linux/personality.h> -#include <linux/binfmts.h> -#include <linux/suspend.h> +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/smp.h> #include <linux/kernel.h> -#include <linux/ptrace.h> #include <linux/signal.h> -#include <linux/stddef.h> -#include <linux/unistd.h> #include <linux/errno.h> -#include <linux/sched.h> #include <linux/wait.h> +#include <linux/ptrace.h> #include <linux/tracehook.h> -#include <linux/elf.h> -#include <linux/smp.h> -#include <linux/mm.h> +#include <linux/unistd.h> +#include <linux/stddef.h> +#include <linux/personality.h> +#include <linux/uaccess.h> #include <asm/processor.h> #include <asm/ucontext.h> -#include <asm/uaccess.h> #include <asm/i387.h> #include <asm/vdso.h> + +#ifdef CONFIG_X86_64 +#include <asm/proto.h> +#include <asm/ia32_unistd.h> +#include <asm/mce.h> +#endif /* CONFIG_X86_64 */ + #include <asm/syscall.h> #include <asm/syscalls.h> -#include "sigframe.h" +#include <asm/sigframe.h> #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) @@ -45,74 +50,6 @@ # define FIX_EFLAGS __FIX_EFLAGS #endif -/* - * Atomically swap in the new signal mask, and wait for a signal. - */ -asmlinkage int -sys_sigsuspend(int history0, int history1, old_sigset_t mask) -{ - mask &= _BLOCKABLE; - spin_lock_irq(¤t->sighand->siglock); - current->saved_sigmask = current->blocked; - siginitset(¤t->blocked, mask); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - current->state = TASK_INTERRUPTIBLE; - schedule(); - set_restore_sigmask(); - - return -ERESTARTNOHAND; -} - -asmlinkage int -sys_sigaction(int sig, const struct old_sigaction __user *act, - struct old_sigaction __user *oact) -{ - struct k_sigaction new_ka, old_ka; - int ret; - - if (act) { - old_sigset_t mask; - - if (!access_ok(VERIFY_READ, act, sizeof(*act)) || - __get_user(new_ka.sa.sa_handler, &act->sa_handler) || - __get_user(new_ka.sa.sa_restorer, &act->sa_restorer)) - return -EFAULT; - - __get_user(new_ka.sa.sa_flags, &act->sa_flags); - __get_user(mask, &act->sa_mask); - siginitset(&new_ka.sa.sa_mask, mask); - } - - ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); - - if (!ret && oact) { - if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) || - __put_user(old_ka.sa.sa_handler, &oact->sa_handler) || - __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer)) - return -EFAULT; - - __put_user(old_ka.sa.sa_flags, &oact->sa_flags); - __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); - } - - return ret; -} - -asmlinkage int sys_sigaltstack(unsigned long bx) -{ - /* - * This is needed to make gcc realize it doesn't own the - * "struct pt_regs" - */ - struct pt_regs *regs = (struct pt_regs *)&bx; - const stack_t __user *uss = (const stack_t __user *)bx; - stack_t __user *uoss = (stack_t __user *)regs->cx; - - return do_sigaltstack(uss, uoss, regs->sp); -} - #define COPY(x) { \ err |= __get_user(regs->x, &sc->x); \ } @@ -123,7 +60,7 @@ asmlinkage int sys_sigaltstack(unsigned long bx) regs->seg = tmp; \ } -#define COPY_SEG_STRICT(seg) { \ +#define COPY_SEG_CPL3(seg) { \ unsigned short tmp; \ err |= __get_user(tmp, &sc->seg); \ regs->seg = tmp | 3; \ @@ -135,9 +72,6 @@ asmlinkage int sys_sigaltstack(unsigned long bx) loadsegment(seg, tmp); \ } -/* - * Do a signal return; undo the signal stack. - */ static int restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, unsigned long *pax) @@ -149,14 +83,36 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, /* Always make any pending restarted system calls return -EINTR */ current_thread_info()->restart_block.fn = do_no_restart_syscall; +#ifdef CONFIG_X86_32 GET_SEG(gs); COPY_SEG(fs); COPY_SEG(es); COPY_SEG(ds); +#endif /* CONFIG_X86_32 */ + COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx); COPY(dx); COPY(cx); COPY(ip); - COPY_SEG_STRICT(cs); - COPY_SEG_STRICT(ss); + +#ifdef CONFIG_X86_64 + COPY(r8); + COPY(r9); + COPY(r10); + COPY(r11); + COPY(r12); + COPY(r13); + COPY(r14); + COPY(r15); +#endif /* CONFIG_X86_64 */ + +#ifdef CONFIG_X86_32 + COPY_SEG_CPL3(cs); + COPY_SEG_CPL3(ss); +#else /* !CONFIG_X86_32 */ + /* Kernel saves and restores only the CS segment register on signals, + * which is the bare minimum needed to allow mixed 32/64-bit code. + * App's signal handler can save/restore other segments if needed. */ + COPY_SEG_CPL3(cs); +#endif /* CONFIG_X86_32 */ err |= __get_user(tmpflags, &sc->flags); regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS); @@ -169,102 +125,24 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, return err; } -asmlinkage unsigned long sys_sigreturn(unsigned long __unused) -{ - struct sigframe __user *frame; - struct pt_regs *regs; - unsigned long ax; - sigset_t set; - - regs = (struct pt_regs *) &__unused; - frame = (struct sigframe __user *)(regs->sp - 8); - - if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) - goto badframe; - if (__get_user(set.sig[0], &frame->sc.oldmask) || (_NSIG_WORDS > 1 - && __copy_from_user(&set.sig[1], &frame->extramask, - sizeof(frame->extramask)))) - goto badframe; - - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - if (restore_sigcontext(regs, &frame->sc, &ax)) - goto badframe; - return ax; - -badframe: - if (show_unhandled_signals && printk_ratelimit()) { - printk("%s%s[%d] bad frame in sigreturn frame:" - "%p ip:%lx sp:%lx oeax:%lx", - task_pid_nr(current) > 1 ? KERN_INFO : KERN_EMERG, - current->comm, task_pid_nr(current), frame, regs->ip, - regs->sp, regs->orig_ax); - print_vma_addr(" in ", regs->ip); - printk(KERN_CONT "\n"); - } - - force_sig(SIGSEGV, current); - - return 0; -} - -static long do_rt_sigreturn(struct pt_regs *regs) -{ - struct rt_sigframe __user *frame; - unsigned long ax; - sigset_t set; - - frame = (struct rt_sigframe __user *)(regs->sp - sizeof(long)); - if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) - goto badframe; - if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) - goto badframe; - - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax)) - goto badframe; - - if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->sp) == -EFAULT) - goto badframe; - - return ax; - -badframe: - signal_fault(regs, frame, "rt_sigreturn"); - return 0; -} - -asmlinkage int sys_rt_sigreturn(unsigned long __unused) -{ - struct pt_regs *regs = (struct pt_regs *)&__unused; - - return do_rt_sigreturn(regs); -} - -/* - * Set up a signal frame. - */ static int setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate, struct pt_regs *regs, unsigned long mask) { - int tmp, err = 0; + int err = 0; - err |= __put_user(regs->fs, (unsigned int __user *)&sc->fs); - savesegment(gs, tmp); - err |= __put_user(tmp, (unsigned int __user *)&sc->gs); +#ifdef CONFIG_X86_32 + { + unsigned int tmp; + savesegment(gs, tmp); + err |= __put_user(tmp, (unsigned int __user *)&sc->gs); + } + err |= __put_user(regs->fs, (unsigned int __user *)&sc->fs); err |= __put_user(regs->es, (unsigned int __user *)&sc->es); err |= __put_user(regs->ds, (unsigned int __user *)&sc->ds); +#endif /* CONFIG_X86_32 */ + err |= __put_user(regs->di, &sc->di); err |= __put_user(regs->si, &sc->si); err |= __put_user(regs->bp, &sc->bp); @@ -273,19 +151,33 @@ setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate, err |= __put_user(regs->dx, &sc->dx); err |= __put_user(regs->cx, &sc->cx); err |= __put_user(regs->ax, &sc->ax); +#ifdef CONFIG_X86_64 + err |= __put_user(regs->r8, &sc->r8); + err |= __put_user(regs->r9, &sc->r9); + err |= __put_user(regs->r10, &sc->r10); + err |= __put_user(regs->r11, &sc->r11); + err |= __put_user(regs->r12, &sc->r12); + err |= __put_user(regs->r13, &sc->r13); + err |= __put_user(regs->r14, &sc->r14); + err |= __put_user(regs->r15, &sc->r15); +#endif /* CONFIG_X86_64 */ + err |= __put_user(current->thread.trap_no, &sc->trapno); err |= __put_user(current->thread.error_code, &sc->err); err |= __put_user(regs->ip, &sc->ip); +#ifdef CONFIG_X86_32 err |= __put_user(regs->cs, (unsigned int __user *)&sc->cs); err |= __put_user(regs->flags, &sc->flags); err |= __put_user(regs->sp, &sc->sp_at_signal); err |= __put_user(regs->ss, (unsigned int __user *)&sc->ss); +#else /* !CONFIG_X86_32 */ + err |= __put_user(regs->flags, &sc->flags); + err |= __put_user(regs->cs, &sc->cs); + err |= __put_user(0, &sc->gs); + err |= __put_user(0, &sc->fs); +#endif /* CONFIG_X86_32 */ - tmp = save_i387_xstate(fpstate); - if (tmp < 0) - err = 1; - else - err |= __put_user(tmp ? fpstate : NULL, &sc->fpstate); + err |= __put_user(fpstate, &sc->fpstate); /* non-iBCS2 extensions.. */ err |= __put_user(mask, &sc->oldmask); @@ -295,6 +187,32 @@ setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate, } /* + * Set up a signal frame. + */ +#ifdef CONFIG_X86_32 +static const struct { + u16 poplmovl; + u32 val; + u16 int80; +} __attribute__((packed)) retcode = { + 0xb858, /* popl %eax; movl $..., %eax */ + __NR_sigreturn, + 0x80cd, /* int $0x80 */ +}; + +static const struct { + u8 movl; + u32 val; + u16 int80; + u8 pad; +} __attribute__((packed)) rt_retcode = { + 0xb8, /* movl $..., %eax */ + __NR_rt_sigreturn, + 0x80cd, /* int $0x80 */ + 0 +}; + +/* * Determine which stack to use.. */ static inline void __user * @@ -328,6 +246,8 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size, if (used_math()) { sp = sp - sig_xstate_size; *fpstate = (struct _fpstate *) sp; + if (save_i387_xstate(*fpstate) < 0) + return (void __user *)-1L; } sp -= frame_size; @@ -383,9 +303,7 @@ __setup_frame(int sig, struct k_sigaction *ka, sigset_t *set, * reasons and because gdb uses it as a signature to notice * signal handler stack frames. */ - err |= __put_user(0xb858, (short __user *)(frame->retcode+0)); - err |= __put_user(__NR_sigreturn, (int __user *)(frame->retcode+2)); - err |= __put_user(0x80cd, (short __user *)(frame->retcode+6)); + err |= __put_user(*((u64 *)&retcode), (u64 *)frame->retcode); if (err) return -EFAULT; @@ -454,9 +372,7 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, * reasons and because gdb uses it as a signature to notice * signal handler stack frames. */ - err |= __put_user(0xb8, (char __user *)(frame->retcode+0)); - err |= __put_user(__NR_rt_sigreturn, (int __user *)(frame->retcode+1)); - err |= __put_user(0x80cd, (short __user *)(frame->retcode+5)); + err |= __put_user(*((u64 *)&rt_retcode), (u64 *)frame->retcode); if (err) return -EFAULT; @@ -475,23 +391,293 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, return 0; } +#else /* !CONFIG_X86_32 */ +/* + * Determine which stack to use.. + */ +static void __user * +get_stack(struct k_sigaction *ka, unsigned long sp, unsigned long size) +{ + /* Default to using normal stack - redzone*/ + sp -= 128; + + /* This is the X/Open sanctioned signal stack switching. */ + if (ka->sa.sa_flags & SA_ONSTACK) { + if (sas_ss_flags(sp) == 0) + sp = current->sas_ss_sp + current->sas_ss_size; + } + + return (void __user *)round_down(sp - size, 64); +} + +static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, + sigset_t *set, struct pt_regs *regs) +{ + struct rt_sigframe __user *frame; + void __user *fp = NULL; + int err = 0; + struct task_struct *me = current; + + if (used_math()) { + fp = get_stack(ka, regs->sp, sig_xstate_size); + frame = (void __user *)round_down( + (unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8; + + if (save_i387_xstate(fp) < 0) + return -EFAULT; + } else + frame = get_stack(ka, regs->sp, sizeof(struct rt_sigframe)) - 8; + + if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) + return -EFAULT; + + if (ka->sa.sa_flags & SA_SIGINFO) { + if (copy_siginfo_to_user(&frame->info, info)) + return -EFAULT; + } + + /* Create the ucontext. */ + if (cpu_has_xsave) + err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags); + else + err |= __put_user(0, &frame->uc.uc_flags); + err |= __put_user(0, &frame->uc.uc_link); + err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp); + err |= __put_user(sas_ss_flags(regs->sp), + &frame->uc.uc_stack.ss_flags); + err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size); + err |= setup_sigcontext(&frame->uc.uc_mcontext, fp, regs, set->sig[0]); + err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); + + /* Set up to return from userspace. If provided, use a stub + already in userspace. */ + /* x86-64 should always use SA_RESTORER. */ + if (ka->sa.sa_flags & SA_RESTORER) { + err |= __put_user(ka->sa.sa_restorer, &frame->pretcode); + } else { + /* could use a vstub here */ + return -EFAULT; + } + + if (err) + return -EFAULT; + + /* Set up registers for signal handler */ + regs->di = sig; + /* In case the signal handler was declared without prototypes */ + regs->ax = 0; + + /* This also works for non SA_SIGINFO handlers because they expect the + next argument after the signal number on the stack. */ + regs->si = (unsigned long)&frame->info; + regs->dx = (unsigned long)&frame->uc; + regs->ip = (unsigned long) ka->sa.sa_handler; + + regs->sp = (unsigned long)frame; + + /* Set up the CS register to run signal handlers in 64-bit mode, + even if the handler happens to be interrupting 32-bit code. */ + regs->cs = __USER_CS; + + return 0; +} +#endif /* CONFIG_X86_32 */ + +#ifdef CONFIG_X86_32 +/* + * Atomically swap in the new signal mask, and wait for a signal. + */ +asmlinkage int +sys_sigsuspend(int history0, int history1, old_sigset_t mask) +{ + mask &= _BLOCKABLE; + spin_lock_irq(¤t->sighand->siglock); + current->saved_sigmask = current->blocked; + siginitset(¤t->blocked, mask); + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_restore_sigmask(); + + return -ERESTARTNOHAND; +} + +asmlinkage int +sys_sigaction(int sig, const struct old_sigaction __user *act, + struct old_sigaction __user *oact) +{ + struct k_sigaction new_ka, old_ka; + int ret; + + if (act) { + old_sigset_t mask; + + if (!access_ok(VERIFY_READ, act, sizeof(*act)) || + __get_user(new_ka.sa.sa_handler, &act->sa_handler) || + __get_user(new_ka.sa.sa_restorer, &act->sa_restorer)) + return -EFAULT; + + __get_user(new_ka.sa.sa_flags, &act->sa_flags); + __get_user(mask, &act->sa_mask); + siginitset(&new_ka.sa.sa_mask, mask); + } + + ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); + + if (!ret && oact) { + if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) || + __put_user(old_ka.sa.sa_handler, &oact->sa_handler) || + __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer)) + return -EFAULT; + + __put_user(old_ka.sa.sa_flags, &oact->sa_flags); + __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); + } + + return ret; +} +#endif /* CONFIG_X86_32 */ + +#ifdef CONFIG_X86_32 +asmlinkage int sys_sigaltstack(unsigned long bx) +{ + /* + * This is needed to make gcc realize it doesn't own the + * "struct pt_regs" + */ + struct pt_regs *regs = (struct pt_regs *)&bx; + const stack_t __user *uss = (const stack_t __user *)bx; + stack_t __user *uoss = (stack_t __user *)regs->cx; + + return do_sigaltstack(uss, uoss, regs->sp); +} +#else /* !CONFIG_X86_32 */ +asmlinkage long +sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, + struct pt_regs *regs) +{ + return do_sigaltstack(uss, uoss, regs->sp); +} +#endif /* CONFIG_X86_32 */ + +/* + * Do a signal return; undo the signal stack. + */ +#ifdef CONFIG_X86_32 +asmlinkage unsigned long sys_sigreturn(unsigned long __unused) +{ + struct sigframe __user *frame; + struct pt_regs *regs; + unsigned long ax; + sigset_t set; + + regs = (struct pt_regs *) &__unused; + frame = (struct sigframe __user *)(regs->sp - 8); + + if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) + goto badframe; + if (__get_user(set.sig[0], &frame->sc.oldmask) || (_NSIG_WORDS > 1 + && __copy_from_user(&set.sig[1], &frame->extramask, + sizeof(frame->extramask)))) + goto badframe; + + sigdelsetmask(&set, ~_BLOCKABLE); + spin_lock_irq(¤t->sighand->siglock); + current->blocked = set; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + if (restore_sigcontext(regs, &frame->sc, &ax)) + goto badframe; + return ax; + +badframe: + signal_fault(regs, frame, "sigreturn"); + + return 0; +} +#endif /* CONFIG_X86_32 */ + +static long do_rt_sigreturn(struct pt_regs *regs) +{ + struct rt_sigframe __user *frame; + unsigned long ax; + sigset_t set; + + frame = (struct rt_sigframe __user *)(regs->sp - sizeof(long)); + if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) + goto badframe; + if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) + goto badframe; + + sigdelsetmask(&set, ~_BLOCKABLE); + spin_lock_irq(¤t->sighand->siglock); + current->blocked = set; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax)) + goto badframe; + + if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->sp) == -EFAULT) + goto badframe; + + return ax; + +badframe: + signal_fault(regs, frame, "rt_sigreturn"); + return 0; +} + +#ifdef CONFIG_X86_32 +asmlinkage int sys_rt_sigreturn(struct pt_regs regs) +{ + return do_rt_sigreturn(®s); +} +#else /* !CONFIG_X86_32 */ +asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) +{ + return do_rt_sigreturn(regs); +} +#endif /* CONFIG_X86_32 */ /* * OK, we're invoking a handler: */ static int signr_convert(int sig) { +#ifdef CONFIG_X86_32 struct thread_info *info = current_thread_info(); if (info->exec_domain && info->exec_domain->signal_invmap && sig < 32) return info->exec_domain->signal_invmap[sig]; +#endif /* CONFIG_X86_32 */ return sig; } +#ifdef CONFIG_X86_32 + #define is_ia32 1 #define ia32_setup_frame __setup_frame #define ia32_setup_rt_frame __setup_rt_frame +#else /* !CONFIG_X86_32 */ + +#ifdef CONFIG_IA32_EMULATION +#define is_ia32 test_thread_flag(TIF_IA32) +#else /* !CONFIG_IA32_EMULATION */ +#define is_ia32 0 +#endif /* CONFIG_IA32_EMULATION */ + +int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, + sigset_t *set, struct pt_regs *regs); +int ia32_setup_frame(int sig, struct k_sigaction *ka, + sigset_t *set, struct pt_regs *regs); + +#endif /* CONFIG_X86_32 */ + static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, struct pt_regs *regs) @@ -592,7 +778,13 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, return 0; } +#ifdef CONFIG_X86_32 #define NR_restart_syscall __NR_restart_syscall +#else /* !CONFIG_X86_32 */ +#define NR_restart_syscall \ + test_thread_flag(TIF_IA32) ? __NR_ia32_restart_syscall : __NR_restart_syscall +#endif /* CONFIG_X86_32 */ + /* * Note that 'init' is a special process: it doesn't get signals it doesn't * want to handle. Thus you cannot kill init even with a SIGKILL even by @@ -704,8 +896,9 @@ void signal_fault(struct pt_regs *regs, void __user *frame, char *where) struct task_struct *me = current; if (show_unhandled_signals && printk_ratelimit()) { - printk(KERN_INFO + printk("%s" "%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx", + task_pid_nr(current) > 1 ? KERN_INFO : KERN_EMERG, me->comm, me->pid, where, frame, regs->ip, regs->sp, regs->orig_ax); print_vma_addr(" in ", regs->ip); diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c deleted file mode 100644 index a5c9627f4db9..000000000000 --- a/arch/x86/kernel/signal_64.c +++ /dev/null @@ -1,516 +0,0 @@ -/* - * Copyright (C) 1991, 1992 Linus Torvalds - * Copyright (C) 2000, 2001, 2002 Andi Kleen SuSE Labs - * - * 1997-11-28 Modified for POSIX.1b signals by Richard Henderson - * 2000-06-20 Pentium III FXSR, SSE support by Gareth Hughes - * 2000-2002 x86-64 support by Andi Kleen - */ - -#include <linux/sched.h> -#include <linux/mm.h> -#include <linux/smp.h> -#include <linux/kernel.h> -#include <linux/signal.h> -#include <linux/errno.h> -#include <linux/wait.h> -#include <linux/ptrace.h> -#include <linux/tracehook.h> -#include <linux/unistd.h> -#include <linux/stddef.h> -#include <linux/personality.h> -#include <linux/compiler.h> -#include <linux/uaccess.h> - -#include <asm/processor.h> -#include <asm/ucontext.h> -#include <asm/i387.h> -#include <asm/proto.h> -#include <asm/ia32_unistd.h> -#include <asm/mce.h> -#include <asm/syscall.h> -#include <asm/syscalls.h> -#include "sigframe.h" - -#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) - -#define __FIX_EFLAGS (X86_EFLAGS_AC | X86_EFLAGS_OF | \ - X86_EFLAGS_DF | X86_EFLAGS_TF | X86_EFLAGS_SF | \ - X86_EFLAGS_ZF | X86_EFLAGS_AF | X86_EFLAGS_PF | \ - X86_EFLAGS_CF) - -#ifdef CONFIG_X86_32 -# define FIX_EFLAGS (__FIX_EFLAGS | X86_EFLAGS_RF) -#else -# define FIX_EFLAGS __FIX_EFLAGS -#endif - -asmlinkage long -sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, - struct pt_regs *regs) -{ - return do_sigaltstack(uss, uoss, regs->sp); -} - -#define COPY(x) { \ - err |= __get_user(regs->x, &sc->x); \ -} - -#define COPY_SEG_STRICT(seg) { \ - unsigned short tmp; \ - err |= __get_user(tmp, &sc->seg); \ - regs->seg = tmp | 3; \ -} - -/* - * Do a signal return; undo the signal stack. - */ -static int -restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, - unsigned long *pax) -{ - void __user *buf; - unsigned int tmpflags; - unsigned int err = 0; - - /* Always make any pending restarted system calls return -EINTR */ - current_thread_info()->restart_block.fn = do_no_restart_syscall; - - COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx); - COPY(dx); COPY(cx); COPY(ip); - COPY(r8); - COPY(r9); - COPY(r10); - COPY(r11); - COPY(r12); - COPY(r13); - COPY(r14); - COPY(r15); - - /* Kernel saves and restores only the CS segment register on signals, - * which is the bare minimum needed to allow mixed 32/64-bit code. - * App's signal handler can save/restore other segments if needed. */ - COPY_SEG_STRICT(cs); - - err |= __get_user(tmpflags, &sc->flags); - regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS); - regs->orig_ax = -1; /* disable syscall checks */ - - err |= __get_user(buf, &sc->fpstate); - err |= restore_i387_xstate(buf); - - err |= __get_user(*pax, &sc->ax); - return err; -} - -static long do_rt_sigreturn(struct pt_regs *regs) -{ - struct rt_sigframe __user *frame; - unsigned long ax; - sigset_t set; - - frame = (struct rt_sigframe __user *)(regs->sp - sizeof(long)); - if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) - goto badframe; - if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) - goto badframe; - - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax)) - goto badframe; - - if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->sp) == -EFAULT) - goto badframe; - - return ax; - -badframe: - signal_fault(regs, frame, "rt_sigreturn"); - return 0; -} - -asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) -{ - return do_rt_sigreturn(regs); -} - -/* - * Set up a signal frame. - */ - -static inline int -setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, - unsigned long mask, struct task_struct *me) -{ - int err = 0; - - err |= __put_user(regs->cs, &sc->cs); - err |= __put_user(0, &sc->gs); - err |= __put_user(0, &sc->fs); - - err |= __put_user(regs->di, &sc->di); - err |= __put_user(regs->si, &sc->si); - err |= __put_user(regs->bp, &sc->bp); - err |= __put_user(regs->sp, &sc->sp); - err |= __put_user(regs->bx, &sc->bx); - err |= __put_user(regs->dx, &sc->dx); - err |= __put_user(regs->cx, &sc->cx); - err |= __put_user(regs->ax, &sc->ax); - err |= __put_user(regs->r8, &sc->r8); - err |= __put_user(regs->r9, &sc->r9); - err |= __put_user(regs->r10, &sc->r10); - err |= __put_user(regs->r11, &sc->r11); - err |= __put_user(regs->r12, &sc->r12); - err |= __put_user(regs->r13, &sc->r13); - err |= __put_user(regs->r14, &sc->r14); - err |= __put_user(regs->r15, &sc->r15); - err |= __put_user(me->thread.trap_no, &sc->trapno); - err |= __put_user(me->thread.error_code, &sc->err); - err |= __put_user(regs->ip, &sc->ip); - err |= __put_user(regs->flags, &sc->flags); - err |= __put_user(mask, &sc->oldmask); - err |= __put_user(me->thread.cr2, &sc->cr2); - - return err; -} - -/* - * Determine which stack to use.. - */ - -static void __user * -get_stack(struct k_sigaction *ka, struct pt_regs *regs, unsigned long size) -{ - unsigned long sp; - - /* Default to using normal stack - redzone*/ - sp = regs->sp - 128; - - /* This is the X/Open sanctioned signal stack switching. */ - if (ka->sa.sa_flags & SA_ONSTACK) { - if (sas_ss_flags(sp) == 0) - sp = current->sas_ss_sp + current->sas_ss_size; - } - - return (void __user *)round_down(sp - size, 64); -} - -static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, - sigset_t *set, struct pt_regs *regs) -{ - struct rt_sigframe __user *frame; - void __user *fp = NULL; - int err = 0; - struct task_struct *me = current; - - if (used_math()) { - fp = get_stack(ka, regs, sig_xstate_size); - frame = (void __user *)round_down( - (unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8; - - if (save_i387_xstate(fp) < 0) - return -EFAULT; - } else - frame = get_stack(ka, regs, sizeof(struct rt_sigframe)) - 8; - - if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) - return -EFAULT; - - if (ka->sa.sa_flags & SA_SIGINFO) { - if (copy_siginfo_to_user(&frame->info, info)) - return -EFAULT; - } - - /* Create the ucontext. */ - if (cpu_has_xsave) - err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags); - else - err |= __put_user(0, &frame->uc.uc_flags); - err |= __put_user(0, &frame->uc.uc_link); - err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp); - err |= __put_user(sas_ss_flags(regs->sp), - &frame->uc.uc_stack.ss_flags); - err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size); - err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, set->sig[0], me); - err |= __put_user(fp, &frame->uc.uc_mcontext.fpstate); - if (sizeof(*set) == 16) { - __put_user(set->sig[0], &frame->uc.uc_sigmask.sig[0]); - __put_user(set->sig[1], &frame->uc.uc_sigmask.sig[1]); - } else - err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); - - /* Set up to return from userspace. If provided, use a stub - already in userspace. */ - /* x86-64 should always use SA_RESTORER. */ - if (ka->sa.sa_flags & SA_RESTORER) { - err |= __put_user(ka->sa.sa_restorer, &frame->pretcode); - } else { - /* could use a vstub here */ - return -EFAULT; - } - - if (err) - return -EFAULT; - - /* Set up registers for signal handler */ - regs->di = sig; - /* In case the signal handler was declared without prototypes */ - regs->ax = 0; - - /* This also works for non SA_SIGINFO handlers because they expect the - next argument after the signal number on the stack. */ - regs->si = (unsigned long)&frame->info; - regs->dx = (unsigned long)&frame->uc; - regs->ip = (unsigned long) ka->sa.sa_handler; - - regs->sp = (unsigned long)frame; - - /* Set up the CS register to run signal handlers in 64-bit mode, - even if the handler happens to be interrupting 32-bit code. */ - regs->cs = __USER_CS; - - return 0; -} - -/* - * OK, we're invoking a handler - */ -static int signr_convert(int sig) -{ - return sig; -} - -#ifdef CONFIG_IA32_EMULATION -#define is_ia32 test_thread_flag(TIF_IA32) -#else -#define is_ia32 0 -#endif - -static int -setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, - sigset_t *set, struct pt_regs *regs) -{ - int usig = signr_convert(sig); - int ret; - - /* Set up the stack frame */ - if (is_ia32) { - if (ka->sa.sa_flags & SA_SIGINFO) - ret = ia32_setup_rt_frame(usig, ka, info, set, regs); - else - ret = ia32_setup_frame(usig, ka, set, regs); - } else - ret = __setup_rt_frame(sig, ka, info, set, regs); - - if (ret) { - force_sigsegv(sig, current); - return -EFAULT; - } - - return ret; -} - -static int -handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, - sigset_t *oldset, struct pt_regs *regs) -{ - int ret; - - /* Are we from a system call? */ - if (syscall_get_nr(current, regs) >= 0) { - /* If so, check system call restarting.. */ - switch (syscall_get_error(current, regs)) { - case -ERESTART_RESTARTBLOCK: - case -ERESTARTNOHAND: - regs->ax = -EINTR; - break; - - case -ERESTARTSYS: - if (!(ka->sa.sa_flags & SA_RESTART)) { - regs->ax = -EINTR; - break; - } - /* fallthrough */ - case -ERESTARTNOINTR: - regs->ax = regs->orig_ax; - regs->ip -= 2; - break; - } - } - - /* - * If TF is set due to a debugger (TIF_FORCED_TF), clear the TF - * flag so that register information in the sigcontext is correct. - */ - if (unlikely(regs->flags & X86_EFLAGS_TF) && - likely(test_and_clear_thread_flag(TIF_FORCED_TF))) - regs->flags &= ~X86_EFLAGS_TF; - - ret = setup_rt_frame(sig, ka, info, oldset, regs); - - if (ret) - return ret; - -#ifdef CONFIG_X86_64 - /* - * This has nothing to do with segment registers, - * despite the name. This magic affects uaccess.h - * macros' behavior. Reset it to the normal setting. - */ - set_fs(USER_DS); -#endif - - /* - * Clear the direction flag as per the ABI for function entry. - */ - regs->flags &= ~X86_EFLAGS_DF; - - /* - * Clear TF when entering the signal handler, but - * notify any tracer that was single-stepping it. - * The tracer may want to single-step inside the - * handler too. - */ - regs->flags &= ~X86_EFLAGS_TF; - - spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked, ¤t->blocked, &ka->sa.sa_mask); - if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked, sig); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - tracehook_signal_handler(sig, info, ka, regs, - test_thread_flag(TIF_SINGLESTEP)); - - return 0; -} - -#define NR_restart_syscall \ - test_thread_flag(TIF_IA32) ? __NR_ia32_restart_syscall : __NR_restart_syscall -/* - * Note that 'init' is a special process: it doesn't get signals it doesn't - * want to handle. Thus you cannot kill init even with a SIGKILL even by - * mistake. - */ -static void do_signal(struct pt_regs *regs) -{ - struct k_sigaction ka; - siginfo_t info; - int signr; - sigset_t *oldset; - - /* - * We want the common case to go fast, which is why we may in certain - * cases get here from kernel mode. Just return without doing anything - * if so. - * X86_32: vm86 regs switched out by assembly code before reaching - * here, so testing against kernel CS suffices. - */ - if (!user_mode(regs)) - return; - - if (current_thread_info()->status & TS_RESTORE_SIGMASK) - oldset = ¤t->saved_sigmask; - else - oldset = ¤t->blocked; - - signr = get_signal_to_deliver(&info, &ka, regs, NULL); - if (signr > 0) { - /* - * Re-enable any watchpoints before delivering the - * signal to user space. The processor register will - * have been cleared if the watchpoint triggered - * inside the kernel. - */ - if (current->thread.debugreg7) - set_debugreg(current->thread.debugreg7, 7); - - /* Whee! Actually deliver the signal. */ - if (handle_signal(signr, &info, &ka, oldset, regs) == 0) { - /* - * A signal was successfully delivered; the saved - * sigmask will have been stored in the signal frame, - * and will be restored by sigreturn, so we can simply - * clear the TS_RESTORE_SIGMASK flag. - */ - current_thread_info()->status &= ~TS_RESTORE_SIGMASK; - } - return; - } - - /* Did we come from a system call? */ - if (syscall_get_nr(current, regs) >= 0) { - /* Restart the system call - no handlers present */ - switch (syscall_get_error(current, regs)) { - case -ERESTARTNOHAND: - case -ERESTARTSYS: - case -ERESTARTNOINTR: - regs->ax = regs->orig_ax; - regs->ip -= 2; - break; - - case -ERESTART_RESTARTBLOCK: - regs->ax = NR_restart_syscall; - regs->ip -= 2; - break; - } - } - - /* - * If there's no signal to deliver, we just put the saved sigmask - * back. - */ - if (current_thread_info()->status & TS_RESTORE_SIGMASK) { - current_thread_info()->status &= ~TS_RESTORE_SIGMASK; - sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); - } -} - -/* - * notification of userspace execution resumption - * - triggered by the TIF_WORK_MASK flags - */ -void -do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) -{ -#if defined(CONFIG_X86_64) && defined(CONFIG_X86_MCE) - /* notify userspace of pending MCEs */ - if (thread_info_flags & _TIF_MCE_NOTIFY) - mce_notify_user(); -#endif /* CONFIG_X86_64 && CONFIG_X86_MCE */ - - /* deal with pending signal delivery */ - if (thread_info_flags & _TIF_SIGPENDING) - do_signal(regs); - - if (thread_info_flags & _TIF_NOTIFY_RESUME) { - clear_thread_flag(TIF_NOTIFY_RESUME); - tracehook_notify_resume(regs); - } - -#ifdef CONFIG_X86_32 - clear_thread_flag(TIF_IRET); -#endif /* CONFIG_X86_32 */ -} - -void signal_fault(struct pt_regs *regs, void __user *frame, char *where) -{ - struct task_struct *me = current; - - if (show_unhandled_signals && printk_ratelimit()) { - printk(KERN_INFO - "%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx", - me->comm, me->pid, where, frame, - regs->ip, regs->sp, regs->orig_ax); - print_vma_addr(" in ", regs->ip); - printk(KERN_CONT "\n"); - } - - force_sig(SIGSEGV, me); -} diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c index 18f9b19f5f8f..7e558db362c1 100644 --- a/arch/x86/kernel/smp.c +++ b/arch/x86/kernel/smp.c @@ -140,19 +140,6 @@ void native_send_call_func_ipi(cpumask_t mask) send_IPI_mask(mask, CALL_FUNCTION_VECTOR); } -static void stop_this_cpu(void *dummy) -{ - local_irq_disable(); - /* - * Remove this CPU: - */ - cpu_clear(smp_processor_id(), cpu_online_map); - disable_local_APIC(); - if (hlt_works(smp_processor_id())) - for (;;) halt(); - for (;;); -} - /* * this function calls the 'stop' function on all other CPUs in the system. */ @@ -178,11 +165,7 @@ static void native_smp_send_stop(void) void smp_reschedule_interrupt(struct pt_regs *regs) { ack_APIC_irq(); -#ifdef CONFIG_X86_32 - __get_cpu_var(irq_stat).irq_resched_count++; -#else - add_pda(irq_resched_count, 1); -#endif + inc_irq_stat(irq_resched_count); } void smp_call_function_interrupt(struct pt_regs *regs) @@ -190,11 +173,7 @@ void smp_call_function_interrupt(struct pt_regs *regs) ack_APIC_irq(); irq_enter(); generic_smp_call_function_interrupt(); -#ifdef CONFIG_X86_32 - __get_cpu_var(irq_stat).irq_call_count++; -#else - add_pda(irq_call_count, 1); -#endif + inc_irq_stat(irq_call_count); irq_exit(); } @@ -203,11 +182,7 @@ void smp_call_function_single_interrupt(struct pt_regs *regs) ack_APIC_irq(); irq_enter(); generic_smp_call_function_single_interrupt(); -#ifdef CONFIG_X86_32 - __get_cpu_var(irq_stat).irq_call_count++; -#else - add_pda(irq_call_count, 1); -#endif + inc_irq_stat(irq_call_count); irq_exit(); } diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 8c3aca7cb343..f8500c969442 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -62,6 +62,7 @@ #include <asm/mtrr.h> #include <asm/vmi.h> #include <asm/genapic.h> +#include <asm/setup.h> #include <linux/mc146818rtc.h> #include <mach_apic.h> @@ -282,19 +283,19 @@ static void __cpuinit smp_callin(void) cpu_set(cpuid, cpu_callin_map); } +static int __cpuinitdata unsafe_smp; + /* * Activate a secondary processor. */ -static void __cpuinit start_secondary(void *unused) +notrace static void __cpuinit start_secondary(void *unused) { /* * Don't put *anything* before cpu_init(), SMP booting is too * fragile that we want to limit the things done here to the * most necessary things. */ -#ifdef CONFIG_VMI vmi_bringup(); -#endif cpu_init(); preempt_disable(); smp_callin(); @@ -397,7 +398,7 @@ static void __cpuinit smp_apply_quirks(struct cpuinfo_x86 *c) goto valid_k7; /* If we get here, not a certified SMP capable AMD system. */ - add_taint(TAINT_UNSAFE_SMP); + unsafe_smp = 1; } valid_k7: @@ -414,12 +415,10 @@ static void __cpuinit smp_checks(void) * Don't taint if we are running SMP kernel on a single non-MP * approved Athlon */ - if (tainted & TAINT_UNSAFE_SMP) { - if (num_online_cpus()) - printk(KERN_INFO "WARNING: This combination of AMD" - "processors is not suitable for SMP.\n"); - else - tainted &= ~TAINT_UNSAFE_SMP; + if (unsafe_smp && num_online_cpus() > 1) { + printk(KERN_INFO "WARNING: This combination of AMD" + "processors is not suitable for SMP.\n"); + add_taint(TAINT_UNSAFE_SMP); } } @@ -536,17 +535,17 @@ static void impress_friends(void) pr_debug("Before bogocount - setting activated=1.\n"); } -static inline void __inquire_remote_apic(int apicid) +void __inquire_remote_apic(int apicid) { unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 }; char *names[] = { "ID", "VERSION", "SPIV" }; int timeout; u32 status; - printk(KERN_INFO "Inquiring remote APIC #%d...\n", apicid); + printk(KERN_INFO "Inquiring remote APIC 0x%x...\n", apicid); for (i = 0; i < ARRAY_SIZE(regs); i++) { - printk(KERN_INFO "... APIC #%d %s: ", apicid, names[i]); + printk(KERN_INFO "... APIC 0x%x %s: ", apicid, names[i]); /* * Wait for idle. @@ -575,14 +574,13 @@ static inline void __inquire_remote_apic(int apicid) } } -#ifdef WAKE_SECONDARY_VIA_NMI /* * Poke the other CPU in the eye via NMI to wake it up. Remember that the normal * INIT, INIT, STARTUP sequence will reset the chip hard for us, and this * won't ... remember to clear down the APIC, etc later. */ -static int __devinit -wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip) +int __devinit +wakeup_secondary_cpu_via_nmi(int logical_apicid, unsigned long start_eip) { unsigned long send_status, accept_status = 0; int maxlvt; @@ -599,7 +597,7 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip) * Give the other CPU some time to accept the IPI. */ udelay(200); - if (APIC_INTEGRATED(apic_version[phys_apicid])) { + if (APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) { maxlvt = lapic_get_maxlvt(); if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ apic_write(APIC_ESR, 0); @@ -614,11 +612,9 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip) return (send_status | accept_status); } -#endif /* WAKE_SECONDARY_VIA_NMI */ -#ifdef WAKE_SECONDARY_VIA_INIT -static int __devinit -wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) +int __devinit +wakeup_secondary_cpu_via_init(int phys_apicid, unsigned long start_eip) { unsigned long send_status, accept_status = 0; int maxlvt, num_starts, j; @@ -737,7 +733,6 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) return (send_status | accept_status); } -#endif /* WAKE_SECONDARY_VIA_INIT */ struct create_idle { struct work_struct work; @@ -874,7 +869,7 @@ do_rest: start_ip = setup_trampoline(); /* So we see what's up */ - printk(KERN_INFO "Booting processor %d/%d ip %lx\n", + printk(KERN_INFO "Booting processor %d APIC 0x%x ip 0x%lx\n", cpu, apicid, start_ip); /* @@ -893,9 +888,11 @@ do_rest: smpboot_setup_warm_reset_vector(start_ip); /* * Be paranoid about clearing APIC errors. - */ - apic_write(APIC_ESR, 0); - apic_read(APIC_ESR); + */ + if (APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) { + apic_write(APIC_ESR, 0); + apic_read(APIC_ESR); + } } /* @@ -1084,8 +1081,10 @@ static int __init smp_sanity_check(unsigned max_cpus) #endif if (!physid_isset(hard_smp_processor_id(), phys_cpu_present_map)) { - printk(KERN_WARNING "weird, boot CPU (#%d) not listed" - "by the BIOS.\n", hard_smp_processor_id()); + printk(KERN_WARNING + "weird, boot CPU (#%d) not listed by the BIOS.\n", + hard_smp_processor_id()); + physid_set(hard_smp_processor_id(), phys_cpu_present_map); } diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c index a03e7f6d90c3..10786af95545 100644 --- a/arch/x86/kernel/stacktrace.c +++ b/arch/x86/kernel/stacktrace.c @@ -6,6 +6,7 @@ #include <linux/sched.h> #include <linux/stacktrace.h> #include <linux/module.h> +#include <linux/uaccess.h> #include <asm/stacktrace.h> static void save_stack_warning(void *data, char *msg) @@ -83,3 +84,66 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace) trace->entries[trace->nr_entries++] = ULONG_MAX; } EXPORT_SYMBOL_GPL(save_stack_trace_tsk); + +/* Userspace stacktrace - based on kernel/trace/trace_sysprof.c */ + +struct stack_frame { + const void __user *next_fp; + unsigned long ret_addr; +}; + +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; + + ret = 1; + pagefault_disable(); + if (__copy_from_user_inatomic(frame, fp, sizeof(*frame))) + ret = 0; + pagefault_enable(); + + return ret; +} + +static inline void __save_stack_trace_user(struct stack_trace *trace) +{ + const struct pt_regs *regs = task_pt_regs(current); + const void __user *fp = (const void __user *)regs->bp; + + if (trace->nr_entries < trace->max_entries) + trace->entries[trace->nr_entries++] = regs->ip; + + while (trace->nr_entries < trace->max_entries) { + struct stack_frame frame; + + frame.next_fp = NULL; + frame.ret_addr = 0; + if (!copy_stack_frame(fp, &frame)) + break; + if ((unsigned long)fp < regs->sp) + break; + if (frame.ret_addr) { + trace->entries[trace->nr_entries++] = + frame.ret_addr; + } + if (fp == frame.next_fp) + break; + fp = frame.next_fp; + } +} + +void save_stack_trace_user(struct stack_trace *trace) +{ + /* + * Trace user stack if we are not a kernel thread + */ + if (current->mm) { + __save_stack_trace_user(trace); + } + if (trace->nr_entries < trace->max_entries) + trace->entries[trace->nr_entries++] = ULONG_MAX; +} + diff --git a/arch/x86/kernel/syscall_64.c b/arch/x86/kernel/syscall_64.c index 3d1be4f0fac5..de87d6008295 100644 --- a/arch/x86/kernel/syscall_64.c +++ b/arch/x86/kernel/syscall_64.c @@ -8,12 +8,12 @@ #define __NO_STUBS #define __SYSCALL(nr, sym) extern asmlinkage void sym(void) ; -#undef ASM_X86__UNISTD_64_H +#undef _ASM_X86_UNISTD_64_H #include <asm/unistd_64.h> #undef __SYSCALL #define __SYSCALL(nr, sym) [nr] = sym, -#undef ASM_X86__UNISTD_64_H +#undef _ASM_X86_UNISTD_64_H typedef void (*sys_call_ptr_t)(void); diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c index 77b400f06ea2..65309e4cb1c0 100644 --- a/arch/x86/kernel/time_32.c +++ b/arch/x86/kernel/time_32.c @@ -75,7 +75,7 @@ EXPORT_SYMBOL(profile_pc); irqreturn_t timer_interrupt(int irq, void *dev_id) { /* Keep nmi watchdog up to date */ - per_cpu(irq_stat, smp_processor_id()).irq0_irqs++; + inc_irq_stat(irq0_irqs); #ifdef CONFIG_X86_IO_APIC if (timer_ack) { diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c index cb19d650c216..891e7a7c4334 100644 --- a/arch/x86/kernel/time_64.c +++ b/arch/x86/kernel/time_64.c @@ -49,9 +49,9 @@ unsigned long profile_pc(struct pt_regs *regs) } EXPORT_SYMBOL(profile_pc); -irqreturn_t timer_interrupt(int irq, void *dev_id) +static irqreturn_t timer_interrupt(int irq, void *dev_id) { - add_pda(irq0_irqs, 1); + inc_irq_stat(irq0_irqs); global_clock_event->event_handler(global_clock_event); @@ -80,6 +80,8 @@ unsigned long __init calibrate_cpu(void) break; no_ctr_free = (i == 4); if (no_ctr_free) { + WARN(1, KERN_WARNING "Warning: AMD perfctrs busy ... " + "cpu_khz value may be incorrect.\n"); i = 3; rdmsrl(MSR_K7_EVNTSEL3, evntsel3); wrmsrl(MSR_K7_EVNTSEL3, 0); diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c index e00534b33534..8da059f949be 100644 --- a/arch/x86/kernel/tlb_32.c +++ b/arch/x86/kernel/tlb_32.c @@ -34,9 +34,8 @@ static DEFINE_SPINLOCK(tlbstate_lock); */ void leave_mm(int cpu) { - if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) - BUG(); - cpu_clear(cpu, per_cpu(cpu_tlbstate, cpu).active_mm->cpu_vm_mask); + BUG_ON(x86_read_percpu(cpu_tlbstate.state) == TLBSTATE_OK); + cpu_clear(cpu, x86_read_percpu(cpu_tlbstate.active_mm)->cpu_vm_mask); load_cr3(swapper_pg_dir); } EXPORT_SYMBOL_GPL(leave_mm); @@ -104,8 +103,8 @@ void smp_invalidate_interrupt(struct pt_regs *regs) * BUG(); */ - if (flush_mm == per_cpu(cpu_tlbstate, cpu).active_mm) { - if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) { + if (flush_mm == x86_read_percpu(cpu_tlbstate.active_mm)) { + if (x86_read_percpu(cpu_tlbstate.state) == TLBSTATE_OK) { if (flush_va == TLB_FLUSH_ALL) local_flush_tlb(); else @@ -119,7 +118,7 @@ void smp_invalidate_interrupt(struct pt_regs *regs) smp_mb__after_clear_bit(); out: put_cpu_no_resched(); - __get_cpu_var(irq_stat).irq_tlb_count++; + inc_irq_stat(irq_tlb_count); } void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm, @@ -154,6 +153,12 @@ void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm, flush_mm = mm; flush_va = va; cpus_or(flush_cpumask, cpumask, flush_cpumask); + + /* + * Make the above memory operations globally visible before + * sending the IPI. + */ + smp_mb(); /* * We have to send the IPI only to * CPUs affected. @@ -232,7 +237,7 @@ static void do_flush_tlb_all(void *info) unsigned long cpu = smp_processor_id(); __flush_tlb_all(); - if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_LAZY) + if (x86_read_percpu(cpu_tlbstate.state) == TLBSTATE_LAZY) leave_mm(cpu); } diff --git a/arch/x86/kernel/tlb_64.c b/arch/x86/kernel/tlb_64.c index dcbf7a1159ea..29887d7081a9 100644 --- a/arch/x86/kernel/tlb_64.c +++ b/arch/x86/kernel/tlb_64.c @@ -154,7 +154,7 @@ asmlinkage void smp_invalidate_interrupt(struct pt_regs *regs) out: ack_APIC_irq(); cpu_clear(cpu, f->flush_cpumask); - add_pda(irq_tlb_count, 1); + inc_irq_stat(irq_tlb_count); } void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm, @@ -183,6 +183,11 @@ void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm, cpus_or(f->flush_cpumask, cpumask, f->flush_cpumask); /* + * Make the above memory operations globally visible before + * sending the IPI. + */ + smp_mb(); + /* * We have to send the IPI only to * CPUs affected. */ diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 8b8c0d6640fa..6a00e5faaa74 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -6,7 +6,7 @@ * This code is released under the GNU General Public License version 2 or * later. */ -#include <linux/mc146818rtc.h> +#include <linux/seq_file.h> #include <linux/proc_fs.h> #include <linux/kernel.h> @@ -566,14 +566,10 @@ static int __init uv_ptc_init(void) if (!is_uv_system()) return 0; - if (!proc_mkdir("sgi_uv", NULL)) - return -EINVAL; - proc_uv_ptc = create_proc_entry(UV_PTC_BASENAME, 0444, NULL); if (!proc_uv_ptc) { printk(KERN_ERR "unable to create %s proc entry\n", UV_PTC_BASENAME); - remove_proc_entry("sgi_uv", NULL); return -EINVAL; } proc_uv_ptc->proc_fops = &proc_uv_ptc_operations; diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c index 1106fac6024d..808031a5ba19 100644 --- a/arch/x86/kernel/trampoline.c +++ b/arch/x86/kernel/trampoline.c @@ -1,10 +1,26 @@ #include <linux/io.h> #include <asm/trampoline.h> +#include <asm/e820.h> /* ready for x86_64 and x86 */ unsigned char *trampoline_base = __va(TRAMPOLINE_BASE); +void __init reserve_trampoline_memory(void) +{ +#ifdef CONFIG_X86_32 + /* + * But first pinch a few for the stack/trampoline stuff + * FIXME: Don't need the extra page at 4K, but need to fix + * trampoline before removing it. (see the GDT stuff) + */ + reserve_early(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE"); +#endif + /* Has to be in very low memory so we can execute real-mode AP code. */ + reserve_early(TRAMPOLINE_BASE, TRAMPOLINE_BASE + TRAMPOLINE_SIZE, + "TRAMPOLINE"); +} + /* * Currently trivial. Write the real->protected mode * bootstrap into the page concerned. The caller @@ -12,7 +28,6 @@ unsigned char *trampoline_base = __va(TRAMPOLINE_BASE); */ unsigned long setup_trampoline(void) { - memcpy(trampoline_base, trampoline_data, - trampoline_end - trampoline_data); + memcpy(trampoline_base, trampoline_data, TRAMPOLINE_SIZE); return virt_to_phys(trampoline_base); } diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index e062974cce34..141907ab6e22 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -481,11 +481,7 @@ do_nmi(struct pt_regs *regs, long error_code) { nmi_enter(); -#ifdef CONFIG_X86_32 - { int cpu; cpu = smp_processor_id(); ++nmi_count(cpu); } -#else - add_pda(__nmi_count, 1); -#endif + inc_irq_stat(__nmi_count); if (!ignore_nmis) default_do_nmi(regs); @@ -664,7 +660,7 @@ void math_error(void __user *ip) { struct task_struct *task; siginfo_t info; - unsigned short cwd, swd; + unsigned short cwd, swd, err; /* * Save the info for the exception handler and clear the error. @@ -675,7 +671,6 @@ void math_error(void __user *ip) task->thread.error_code = 0; info.si_signo = SIGFPE; info.si_errno = 0; - info.si_code = __SI_FAULT; info.si_addr = ip; /* * (~cwd & swd) will mask out exceptions that are not set to unmasked @@ -689,34 +684,31 @@ void math_error(void __user *ip) */ cwd = get_fpu_cwd(task); swd = get_fpu_swd(task); - switch (swd & ~cwd & 0x3f) { - case 0x000: /* No unmasked exception */ + + err = swd & ~cwd & 0x3f; + #ifdef CONFIG_X86_32 + if (!err) return; #endif - default: /* Multiple exceptions */ - break; - case 0x001: /* Invalid Op */ + + if (err & 0x001) { /* Invalid op */ /* * swd & 0x240 == 0x040: Stack Underflow * swd & 0x240 == 0x240: Stack Overflow * User must clear the SF bit (0x40) if set */ info.si_code = FPE_FLTINV; - break; - case 0x002: /* Denormalize */ - case 0x010: /* Underflow */ - info.si_code = FPE_FLTUND; - break; - case 0x004: /* Zero Divide */ + } else if (err & 0x004) { /* Divide by Zero */ info.si_code = FPE_FLTDIV; - break; - case 0x008: /* Overflow */ + } else if (err & 0x008) { /* Overflow */ info.si_code = FPE_FLTOVF; - break; - case 0x020: /* Precision */ + } else if (err & 0x012) { /* Denormal, Underflow */ + info.si_code = FPE_FLTUND; + } else if (err & 0x020) { /* Precision */ info.si_code = FPE_FLTRES; - break; + } else { + info.si_code = __SI_FAULT|SI_KERNEL; /* WTF? */ } force_sig_info(SIGFPE, &info, task); } @@ -931,14 +923,6 @@ do_device_not_available(struct pt_regs *regs, long error) } #ifdef CONFIG_X86_32 -#ifdef CONFIG_X86_MCE -dotraplinkage void __kprobes do_machine_check(struct pt_regs *regs, long error) -{ - conditional_sti(regs); - machine_check_vector(regs, error); -} -#endif - dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code) { siginfo_t info; diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 161bb850fc47..599e58168631 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -15,6 +15,7 @@ #include <asm/vgtod.h> #include <asm/time.h> #include <asm/delay.h> +#include <asm/hypervisor.h> unsigned int cpu_khz; /* TSC clocks / usec, not used here */ EXPORT_SYMBOL(cpu_khz); @@ -31,6 +32,7 @@ static int tsc_unstable; erroneous rdtsc usage on !cpu_has_tsc processors */ static int tsc_disabled = -1; +static int tsc_clocksource_reliable; /* * Scheduler clock - returns current time in nanosec units. */ @@ -55,7 +57,7 @@ u64 native_sched_clock(void) rdtscll(this_offset); /* return the value in ns */ - return cycles_2_ns(this_offset); + return __cycles_2_ns(this_offset); } /* We need to define a real function for sched_clock, to override the @@ -98,6 +100,15 @@ int __init notsc_setup(char *str) __setup("notsc", notsc_setup); +static int __init tsc_setup(char *str) +{ + if (!strcmp(str, "reliable")) + tsc_clocksource_reliable = 1; + return 1; +} + +__setup("tsc=", tsc_setup); + #define MAX_RETRIES 5 #define SMI_TRESHOLD 50000 @@ -352,9 +363,15 @@ unsigned long native_calibrate_tsc(void) { u64 tsc1, tsc2, delta, ref1, ref2; unsigned long tsc_pit_min = ULONG_MAX, tsc_ref_min = ULONG_MAX; - unsigned long flags, latch, ms, fast_calibrate; + unsigned long flags, latch, ms, fast_calibrate, tsc_khz; int hpet = is_hpet_enabled(), i, loopmin; + tsc_khz = get_hypervisor_tsc_freq(); + if (tsc_khz) { + printk(KERN_INFO "TSC: Frequency read from the hypervisor\n"); + return tsc_khz; + } + local_irq_save(flags); fast_calibrate = quick_pit_calibrate(); local_irq_restore(flags); @@ -731,24 +748,21 @@ static struct dmi_system_id __initdata bad_tsc_dmi_table[] = { {} }; -/* - * Geode_LX - the OLPC CPU has a possibly a very reliable TSC - */ +static void __init check_system_tsc_reliable(void) +{ #ifdef CONFIG_MGEODE_LX -/* RTSC counts during suspend */ + /* RTSC counts during suspend */ #define RTSC_SUSP 0x100 - -static void __init check_geode_tsc_reliable(void) -{ unsigned long res_low, res_high; rdmsr_safe(MSR_GEODE_BUSCONT_CONF0, &res_low, &res_high); + /* Geode_LX - the OLPC CPU has a possibly a very reliable TSC */ if (res_low & RTSC_SUSP) - clocksource_tsc.flags &= ~CLOCK_SOURCE_MUST_VERIFY; -} -#else -static inline void check_geode_tsc_reliable(void) { } + tsc_clocksource_reliable = 1; #endif + if (boot_cpu_has(X86_FEATURE_TSC_RELIABLE)) + tsc_clocksource_reliable = 1; +} /* * Make an educated guess if the TSC is trustworthy and synchronized @@ -759,7 +773,7 @@ __cpuinit int unsynchronized_tsc(void) if (!cpu_has_tsc || tsc_unstable) return 1; -#ifdef CONFIG_SMP +#ifdef CONFIG_X86_SMP if (apic_is_clustered_box()) return 1; #endif @@ -783,6 +797,8 @@ static void __init init_tsc_clocksource(void) { clocksource_tsc.mult = clocksource_khz2mult(tsc_khz, clocksource_tsc.shift); + if (tsc_clocksource_reliable) + clocksource_tsc.flags &= ~CLOCK_SOURCE_MUST_VERIFY; /* lower the rating if we already know its unstable: */ if (check_tsc_unstable()) { clocksource_tsc.rating = 0; @@ -813,10 +829,6 @@ void __init tsc_init(void) cpu_khz = calibrate_cpu(); #endif - lpj = ((u64)tsc_khz * 1000); - do_div(lpj, HZ); - lpj_fine = lpj; - printk("Detected %lu.%03lu MHz processor.\n", (unsigned long)cpu_khz / 1000, (unsigned long)cpu_khz % 1000); @@ -836,6 +848,10 @@ void __init tsc_init(void) /* now allow native_sched_clock() to use rdtsc */ tsc_disabled = 0; + lpj = ((u64)tsc_khz * 1000); + do_div(lpj, HZ); + lpj_fine = lpj; + use_tsc_delay(); /* Check and install the TSC clocksource */ dmi_check_system(bad_tsc_dmi_table); @@ -843,7 +859,7 @@ void __init tsc_init(void) if (unsynchronized_tsc()) mark_tsc_unstable("TSCs unsynchronized"); - check_geode_tsc_reliable(); + check_system_tsc_reliable(); init_tsc_clocksource(); } diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c index 9ffb01c31c40..bf36328f6ef9 100644 --- a/arch/x86/kernel/tsc_sync.c +++ b/arch/x86/kernel/tsc_sync.c @@ -46,7 +46,9 @@ static __cpuinit void check_tsc_warp(void) cycles_t start, now, prev, end; int i; + rdtsc_barrier(); start = get_cycles(); + rdtsc_barrier(); /* * The measurement runs for 20 msecs: */ @@ -61,7 +63,9 @@ static __cpuinit void check_tsc_warp(void) */ __raw_spin_lock(&sync_lock); prev = last_tsc; + rdtsc_barrier(); now = get_cycles(); + rdtsc_barrier(); last_tsc = now; __raw_spin_unlock(&sync_lock); @@ -108,6 +112,12 @@ void __cpuinit check_tsc_sync_source(int cpu) if (unsynchronized_tsc()) return; + if (boot_cpu_has(X86_FEATURE_TSC_RELIABLE)) { + printk(KERN_INFO + "Skipping synchronization checks as TSC is reliable.\n"); + return; + } + printk(KERN_INFO "checking TSC synchronization [CPU#%d -> CPU#%d]:", smp_processor_id(), cpu); @@ -161,7 +171,7 @@ void __cpuinit check_tsc_sync_target(void) { int cpus = 2; - if (unsynchronized_tsc()) + if (unsynchronized_tsc() || boot_cpu_has(X86_FEATURE_TSC_RELIABLE)) return; /* diff --git a/arch/x86/kernel/uv_irq.c b/arch/x86/kernel/uv_irq.c new file mode 100644 index 000000000000..aeef529917e4 --- /dev/null +++ b/arch/x86/kernel/uv_irq.c @@ -0,0 +1,79 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * SGI UV IRQ functions + * + * Copyright (C) 2008 Silicon Graphics, Inc. All rights reserved. + */ + +#include <linux/module.h> +#include <linux/irq.h> + +#include <asm/apic.h> +#include <asm/uv/uv_irq.h> + +static void uv_noop(unsigned int irq) +{ +} + +static unsigned int uv_noop_ret(unsigned int irq) +{ + return 0; +} + +static void uv_ack_apic(unsigned int irq) +{ + ack_APIC_irq(); +} + +struct irq_chip uv_irq_chip = { + .name = "UV-CORE", + .startup = uv_noop_ret, + .shutdown = uv_noop, + .enable = uv_noop, + .disable = uv_noop, + .ack = uv_noop, + .mask = uv_noop, + .unmask = uv_noop, + .eoi = uv_ack_apic, + .end = uv_noop, +}; + +/* + * Set up a mapping of an available irq and vector, and enable the specified + * MMR that defines the MSI that is to be sent to the specified CPU when an + * interrupt is raised. + */ +int uv_setup_irq(char *irq_name, int cpu, int mmr_blade, + unsigned long mmr_offset) +{ + int irq; + int ret; + + irq = create_irq(); + if (irq <= 0) + return -EBUSY; + + ret = arch_enable_uv_irq(irq_name, irq, cpu, mmr_blade, mmr_offset); + if (ret != irq) + destroy_irq(irq); + + return ret; +} +EXPORT_SYMBOL_GPL(uv_setup_irq); + +/* + * Tear down a mapping of an irq and vector, and disable the specified MMR that + * defined the MSI that was to be sent to the specified CPU when an interrupt + * was raised. + * + * Set mmr_blade and mmr_offset to what was passed in on uv_setup_irq(). + */ +void uv_teardown_irq(unsigned int irq, int mmr_blade, unsigned long mmr_offset) +{ + arch_disable_uv_irq(mmr_blade, mmr_offset); + destroy_irq(irq); +} +EXPORT_SYMBOL_GPL(uv_teardown_irq); diff --git a/arch/x86/kernel/uv_sysfs.c b/arch/x86/kernel/uv_sysfs.c new file mode 100644 index 000000000000..67f9b9dbf800 --- /dev/null +++ b/arch/x86/kernel/uv_sysfs.c @@ -0,0 +1,72 @@ +/* + * This file supports the /sys/firmware/sgi_uv interfaces for SGI UV. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) Russ Anderson + */ + +#include <linux/sysdev.h> +#include <asm/uv/bios.h> + +struct kobject *sgi_uv_kobj; + +static ssize_t partition_id_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%ld\n", sn_partition_id); +} + +static ssize_t coherence_id_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%ld\n", partition_coherence_id()); +} + +static struct kobj_attribute partition_id_attr = + __ATTR(partition_id, S_IRUGO, partition_id_show, NULL); + +static struct kobj_attribute coherence_id_attr = + __ATTR(coherence_id, S_IRUGO, coherence_id_show, NULL); + + +static int __init sgi_uv_sysfs_init(void) +{ + unsigned long ret; + + if (!sgi_uv_kobj) + sgi_uv_kobj = kobject_create_and_add("sgi_uv", firmware_kobj); + if (!sgi_uv_kobj) { + printk(KERN_WARNING "kobject_create_and_add sgi_uv failed \n"); + return -EINVAL; + } + + ret = sysfs_create_file(sgi_uv_kobj, &partition_id_attr.attr); + if (ret) { + printk(KERN_WARNING "sysfs_create_file partition_id failed \n"); + return ret; + } + + ret = sysfs_create_file(sgi_uv_kobj, &coherence_id_attr.attr); + if (ret) { + printk(KERN_WARNING "sysfs_create_file coherence_id failed \n"); + return ret; + } + + return 0; +} + +device_initcall(sgi_uv_sysfs_init); diff --git a/arch/x86/kernel/visws_quirks.c b/arch/x86/kernel/visws_quirks.c index 61a97e616f70..0c9667f0752a 100644 --- a/arch/x86/kernel/visws_quirks.c +++ b/arch/x86/kernel/visws_quirks.c @@ -484,10 +484,11 @@ static void disable_cobalt_irq(unsigned int irq) static unsigned int startup_cobalt_irq(unsigned int irq) { unsigned long flags; + struct irq_desc *desc = irq_to_desc(irq); spin_lock_irqsave(&cobalt_lock, flags); - if ((irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING))) - irq_desc[irq].status &= ~(IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING); + if ((desc->status & (IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING))) + desc->status &= ~(IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING); enable_cobalt_irq(irq); spin_unlock_irqrestore(&cobalt_lock, flags); return 0; @@ -506,9 +507,10 @@ static void ack_cobalt_irq(unsigned int irq) static void end_cobalt_irq(unsigned int irq) { unsigned long flags; + struct irq_desc *desc = irq_to_desc(irq); spin_lock_irqsave(&cobalt_lock, flags); - if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) + if (!(desc->status & (IRQ_DISABLED | IRQ_INPROGRESS))) enable_cobalt_irq(irq); spin_unlock_irqrestore(&cobalt_lock, flags); } @@ -626,12 +628,12 @@ static irqreturn_t piix4_master_intr(int irq, void *dev_id) spin_unlock_irqrestore(&i8259A_lock, flags); - desc = irq_desc + realirq; + desc = irq_to_desc(realirq); /* * handle this 'virtual interrupt' as a Cobalt one now. */ - kstat_cpu(smp_processor_id()).irqs[realirq]++; + kstat_incr_irqs_this_cpu(realirq, desc); if (likely(desc->action != NULL)) handle_IRQ_event(realirq, desc->action); @@ -662,27 +664,29 @@ void init_VISWS_APIC_irqs(void) int i; for (i = 0; i < CO_IRQ_APIC0 + CO_APIC_LAST + 1; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; + struct irq_desc *desc = irq_to_desc(i); + + desc->status = IRQ_DISABLED; + desc->action = 0; + desc->depth = 1; if (i == 0) { - irq_desc[i].chip = &cobalt_irq_type; + desc->chip = &cobalt_irq_type; } else if (i == CO_IRQ_IDE0) { - irq_desc[i].chip = &cobalt_irq_type; + desc->chip = &cobalt_irq_type; } else if (i == CO_IRQ_IDE1) { - irq_desc[i].chip = &cobalt_irq_type; + desc->chip = &cobalt_irq_type; } else if (i == CO_IRQ_8259) { - irq_desc[i].chip = &piix4_master_irq_type; + desc->chip = &piix4_master_irq_type; } else if (i < CO_IRQ_APIC0) { - irq_desc[i].chip = &piix4_virtual_irq_type; + desc->chip = &piix4_virtual_irq_type; } else if (IS_CO_APIC(i)) { - irq_desc[i].chip = &cobalt_irq_type; + desc->chip = &cobalt_irq_type; } } diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index 8b6c393ab9fd..23206ba16874 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -266,109 +266,6 @@ static void vmi_nop(void) { } -#ifdef CONFIG_DEBUG_PAGE_TYPE - -#ifdef CONFIG_X86_PAE -#define MAX_BOOT_PTS (2048+4+1) -#else -#define MAX_BOOT_PTS (1024+1) -#endif - -/* - * During boot, mem_map is not yet available in paging_init, so stash - * all the boot page allocations here. - */ -static struct { - u32 pfn; - int type; -} boot_page_allocations[MAX_BOOT_PTS]; -static int num_boot_page_allocations; -static int boot_allocations_applied; - -void vmi_apply_boot_page_allocations(void) -{ - int i; - BUG_ON(!mem_map); - for (i = 0; i < num_boot_page_allocations; i++) { - struct page *page = pfn_to_page(boot_page_allocations[i].pfn); - page->type = boot_page_allocations[i].type; - page->type = boot_page_allocations[i].type & - ~(VMI_PAGE_ZEROED | VMI_PAGE_CLONE); - } - boot_allocations_applied = 1; -} - -static void record_page_type(u32 pfn, int type) -{ - BUG_ON(num_boot_page_allocations >= MAX_BOOT_PTS); - boot_page_allocations[num_boot_page_allocations].pfn = pfn; - boot_page_allocations[num_boot_page_allocations].type = type; - num_boot_page_allocations++; -} - -static void check_zeroed_page(u32 pfn, int type, struct page *page) -{ - u32 *ptr; - int i; - int limit = PAGE_SIZE / sizeof(int); - - if (page_address(page)) - ptr = (u32 *)page_address(page); - else - ptr = (u32 *)__va(pfn << PAGE_SHIFT); - /* - * When cloning the root in non-PAE mode, only the userspace - * pdes need to be zeroed. - */ - if (type & VMI_PAGE_CLONE) - limit = KERNEL_PGD_BOUNDARY; - for (i = 0; i < limit; i++) - BUG_ON(ptr[i]); -} - -/* - * We stash the page type into struct page so we can verify the page - * types are used properly. - */ -static void vmi_set_page_type(u32 pfn, int type) -{ - /* PAE can have multiple roots per page - don't track */ - if (PTRS_PER_PMD > 1 && (type & VMI_PAGE_PDP)) - return; - - if (boot_allocations_applied) { - struct page *page = pfn_to_page(pfn); - if (type != VMI_PAGE_NORMAL) - BUG_ON(page->type); - else - BUG_ON(page->type == VMI_PAGE_NORMAL); - page->type = type & ~(VMI_PAGE_ZEROED | VMI_PAGE_CLONE); - if (type & VMI_PAGE_ZEROED) - check_zeroed_page(pfn, type, page); - } else { - record_page_type(pfn, type); - } -} - -static void vmi_check_page_type(u32 pfn, int type) -{ - /* PAE can have multiple roots per page - skip checks */ - if (PTRS_PER_PMD > 1 && (type & VMI_PAGE_PDP)) - return; - - type &= ~(VMI_PAGE_ZEROED | VMI_PAGE_CLONE); - if (boot_allocations_applied) { - struct page *page = pfn_to_page(pfn); - BUG_ON((page->type ^ type) & VMI_PAGE_PAE); - BUG_ON(type == VMI_PAGE_NORMAL && page->type); - BUG_ON((type & page->type) == 0); - } -} -#else -#define vmi_set_page_type(p,t) do { } while (0) -#define vmi_check_page_type(p,t) do { } while (0) -#endif - #ifdef CONFIG_HIGHPTE static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type) { @@ -395,7 +292,6 @@ static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type) static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn) { - vmi_set_page_type(pfn, VMI_PAGE_L1); vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0); } @@ -406,27 +302,22 @@ static void vmi_allocate_pmd(struct mm_struct *mm, unsigned long pfn) * It is called only for swapper_pg_dir, which already has * data on it. */ - vmi_set_page_type(pfn, VMI_PAGE_L2); vmi_ops.allocate_page(pfn, VMI_PAGE_L2, 0, 0, 0); } static void vmi_allocate_pmd_clone(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count) { - vmi_set_page_type(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE); - vmi_check_page_type(clonepfn, VMI_PAGE_L2); vmi_ops.allocate_page(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE, clonepfn, start, count); } static void vmi_release_pte(unsigned long pfn) { vmi_ops.release_page(pfn, VMI_PAGE_L1); - vmi_set_page_type(pfn, VMI_PAGE_NORMAL); } static void vmi_release_pmd(unsigned long pfn) { vmi_ops.release_page(pfn, VMI_PAGE_L2); - vmi_set_page_type(pfn, VMI_PAGE_NORMAL); } /* @@ -450,26 +341,22 @@ static void vmi_release_pmd(unsigned long pfn) static void vmi_update_pte(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { - vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE); vmi_ops.update_pte(ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0)); } static void vmi_update_pte_defer(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { - vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE); vmi_ops.update_pte(ptep, vmi_flags_addr_defer(mm, addr, VMI_PAGE_PT, 0)); } static void vmi_set_pte(pte_t *ptep, pte_t pte) { /* XXX because of set_pmd_pte, this can be called on PT or PD layers */ - vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE | VMI_PAGE_PD); vmi_ops.set_pte(pte, ptep, VMI_PAGE_PT); } static void vmi_set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte) { - vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE); vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0)); } @@ -477,10 +364,8 @@ static void vmi_set_pmd(pmd_t *pmdp, pmd_t pmdval) { #ifdef CONFIG_X86_PAE const pte_t pte = { .pte = pmdval.pmd }; - vmi_check_page_type(__pa(pmdp) >> PAGE_SHIFT, VMI_PAGE_PMD); #else const pte_t pte = { pmdval.pud.pgd.pgd }; - vmi_check_page_type(__pa(pmdp) >> PAGE_SHIFT, VMI_PAGE_PGD); #endif vmi_ops.set_pte(pte, (pte_t *)pmdp, VMI_PAGE_PD); } @@ -502,7 +387,6 @@ static void vmi_set_pte_atomic(pte_t *ptep, pte_t pteval) static void vmi_set_pte_present(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte) { - vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE); vmi_ops.set_pte(pte, ptep, vmi_flags_addr_defer(mm, addr, VMI_PAGE_PT, 1)); } @@ -510,21 +394,18 @@ static void vmi_set_pud(pud_t *pudp, pud_t pudval) { /* Um, eww */ const pte_t pte = { .pte = pudval.pgd.pgd }; - vmi_check_page_type(__pa(pudp) >> PAGE_SHIFT, VMI_PAGE_PGD); vmi_ops.set_pte(pte, (pte_t *)pudp, VMI_PAGE_PDP); } static void vmi_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { const pte_t pte = { .pte = 0 }; - vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE); vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0)); } static void vmi_pmd_clear(pmd_t *pmd) { const pte_t pte = { .pte = 0 }; - vmi_check_page_type(__pa(pmd) >> PAGE_SHIFT, VMI_PAGE_PMD); vmi_ops.set_pte(pte, (pte_t *)pmd, VMI_PAGE_PD); } #endif @@ -960,8 +841,6 @@ static inline int __init activate_vmi(void) void __init vmi_init(void) { - unsigned long flags; - if (!vmi_rom) probe_vmi_rom(); else @@ -973,13 +852,21 @@ void __init vmi_init(void) reserve_top_address(-vmi_rom->virtual_top); - local_irq_save(flags); - activate_vmi(); - #ifdef CONFIG_X86_IO_APIC /* This is virtual hardware; timer routing is wired correctly */ no_timer_check = 1; #endif +} + +void vmi_activate(void) +{ + unsigned long flags; + + if (!vmi_rom) + return; + + local_irq_save(flags); + activate_vmi(); local_irq_restore(flags & X86_EFLAGS_IF); } diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c index 6953859fe289..254ee07f8635 100644 --- a/arch/x86/kernel/vmiclock_32.c +++ b/arch/x86/kernel/vmiclock_32.c @@ -235,11 +235,14 @@ static void __devinit vmi_time_init_clockevent(void) void __init vmi_time_init(void) { + unsigned int cpu; /* Disable PIT: BIOSes start PIT CH0 with 18.2hz peridic. */ outb_pit(0x3a, PIT_MODE); /* binary, mode 5, LSB/MSB, ch 0 */ vmi_time_init_clockevent(); setup_irq(0, &vmi_clock_action); + for_each_possible_cpu(cpu) + per_cpu(vector_irq, cpu)[vmi_get_timer_vector()] = 0; } #ifdef CONFIG_X86_LOCAL_APIC diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S index a9b8560adbc2..82c67559dde7 100644 --- a/arch/x86/kernel/vmlinux_32.lds.S +++ b/arch/x86/kernel/vmlinux_32.lds.S @@ -44,6 +44,7 @@ SECTIONS SCHED_TEXT LOCK_TEXT KPROBES_TEXT + IRQENTRY_TEXT *(.fixup) *(.gnu.warning) _etext = .; /* End of text section */ diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index 46e05447405b..1a614c0e6bef 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -35,6 +35,7 @@ SECTIONS SCHED_TEXT LOCK_TEXT KPROBES_TEXT + IRQENTRY_TEXT *(.fixup) *(.gnu.warning) _etext = .; /* End of text section */ diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c index 7766d36983fc..a688f3bfaec2 100644 --- a/arch/x86/kernel/vsmp_64.c +++ b/arch/x86/kernel/vsmp_64.c @@ -78,7 +78,7 @@ static unsigned __init_or_module vsmp_patch(u8 type, u16 clobbers, void *ibuf, static void __init set_vsmp_pv_ops(void) { - void *address; + void __iomem *address; unsigned int cap, ctl, cfg; /* set vSMP magic bits to indicate vSMP capable kernel */ diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index 0b8b6690a86d..44153afc9067 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -17,6 +17,9 @@ * want per guest time just set the kernel.vsyscall64 sysctl to 0. */ +/* Disable profiling for userspace code: */ +#define DISABLE_BRANCH_PROFILING + #include <linux/time.h> #include <linux/init.h> #include <linux/kernel.h> @@ -128,7 +131,16 @@ static __always_inline void do_vgettimeofday(struct timeval * tv) gettimeofday(tv,NULL); return; } + + /* + * Surround the RDTSC by barriers, to make sure it's not + * speculated to outside the seqlock critical section and + * does not cause time warps: + */ + rdtsc_barrier(); now = vread(); + rdtsc_barrier(); + base = __vsyscall_gtod_data.clock.cycle_last; mask = __vsyscall_gtod_data.clock.mask; mult = __vsyscall_gtod_data.clock.mult; diff --git a/arch/x86/kernel/x8664_ksyms_64.c b/arch/x86/kernel/x8664_ksyms_64.c index b545f371b5f5..695e426aa354 100644 --- a/arch/x86/kernel/x8664_ksyms_64.c +++ b/arch/x86/kernel/x8664_ksyms_64.c @@ -12,7 +12,7 @@ #include <asm/desc.h> #include <asm/ftrace.h> -#ifdef CONFIG_FTRACE +#ifdef CONFIG_FUNCTION_TRACER /* mcount is defined in assembly */ EXPORT_SYMBOL(mcount); #endif diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c index 9abac8a9d823..15c3e6999182 100644 --- a/arch/x86/kernel/xsave.c +++ b/arch/x86/kernel/xsave.c @@ -248,7 +248,7 @@ clear: * This will be saved when ever the FP and extended state context is * saved on the user stack during the signal handler delivery to the user. */ -void prepare_fx_sw_frame(void) +static void prepare_fx_sw_frame(void) { int size_extended = (xstate_size - sizeof(struct i387_fxsave_struct)) + FP_XSTATE_MAGIC2_SIZE; @@ -310,7 +310,7 @@ static void __init setup_xstate_init(void) /* * Enable and initialize the xsave feature. */ -void __init xsave_cntxt_init(void) +void __ref xsave_cntxt_init(void) { unsigned int eax, ebx, ecx, edx; |