diff options
Diffstat (limited to 'arch/x86/kernel')
101 files changed, 3735 insertions, 3791 deletions
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 0925676266b..2c833d8c414 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -11,6 +11,8 @@ ifdef CONFIG_FUNCTION_TRACER CFLAGS_REMOVE_tsc.o = -pg CFLAGS_REMOVE_rtc.o = -pg CFLAGS_REMOVE_paravirt-spinlocks.o = -pg +CFLAGS_REMOVE_pvclock.o = -pg +CFLAGS_REMOVE_kvmclock.o = -pg CFLAGS_REMOVE_ftrace.o = -pg CFLAGS_REMOVE_early_printk.o = -pg endif @@ -32,7 +34,8 @@ GCOV_PROFILE_paravirt.o := n 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.o ioport.o ldt.o dumpstack.o -obj-y += setup.o x86_init.o i8259.o irqinit.o +obj-y += setup.o x86_init.o i8259.o irqinit.o jump_label.o +obj-$(CONFIG_IRQ_WORK) += irq_work.o obj-$(CONFIG_X86_VISWS) += visws_quirks.o obj-$(CONFIG_X86_32) += probe_roms_32.o obj-$(CONFIG_X86_32) += sys_i386_32.o i386_ksyms_32.o @@ -42,6 +45,7 @@ obj-y += bootflag.o e820.o obj-y += pci-dma.o quirks.o i8237.o topology.o kdebugfs.o obj-y += alternative.o i8253.o pci-nommu.o hw_breakpoint.o obj-y += tsc.o io_delay.o rtc.o +obj-y += pci-iommu_table.o obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o obj-y += process.o @@ -83,15 +87,15 @@ obj-$(CONFIG_DOUBLEFAULT) += doublefault_32.o obj-$(CONFIG_KGDB) += kgdb.o obj-$(CONFIG_VM86) += vm86_32.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o +obj-$(CONFIG_EARLY_PRINTK_MRST) += early_printk_mrst.o obj-$(CONFIG_HPET_TIMER) += hpet.o obj-$(CONFIG_APB_TIMER) += apb_timer.o -obj-$(CONFIG_K8_NB) += k8.o +obj-$(CONFIG_AMD_NB) += amd_nb.o obj-$(CONFIG_DEBUG_RODATA_TEST) += test_rodata.o obj-$(CONFIG_DEBUG_NX_TEST) += test_nx.o -obj-$(CONFIG_VMI) += vmi_32.o vmiclock_32.o obj-$(CONFIG_KVM_GUEST) += kvm.o obj-$(CONFIG_KVM_CLOCK) += kvmclock.o obj-$(CONFIG_PARAVIRT) += paravirt.o paravirt_patch_$(BITS).o @@ -104,6 +108,7 @@ obj-$(CONFIG_SCx200) += scx200.o scx200-y += scx200_32.o obj-$(CONFIG_OLPC) += olpc.o +obj-$(CONFIG_OLPC_XO1) += olpc-xo1.o obj-$(CONFIG_OLPC_OPENFIRMWARE) += olpc_ofw.o obj-$(CONFIG_X86_MRST) += mrst.o @@ -120,7 +125,6 @@ obj-$(CONFIG_SWIOTLB) += pci-swiotlb.o # 64 bit specific files ifeq ($(CONFIG_X86_64),y) obj-$(CONFIG_X86_UV) += tlb_uv.o bios_uv.o uv_irq.o uv_sysfs.o uv_time.o - obj-$(CONFIG_X86_PM_TIMER) += pmtimer_64.o obj-$(CONFIG_AUDIT) += audit_64.o obj-$(CONFIG_GART_IOMMU) += pci-gart_64.o aperture_64.o diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c index fb7a5f052e2..5812404a0d4 100644 --- a/arch/x86/kernel/acpi/cstate.c +++ b/arch/x86/kernel/acpi/cstate.c @@ -13,6 +13,7 @@ #include <acpi/processor.h> #include <asm/acpi.h> +#include <asm/mwait.h> /* * Initialize bm_flags based on the CPU cache properties @@ -61,20 +62,10 @@ struct cstate_entry { unsigned int ecx; } states[ACPI_PROCESSOR_MAX_POWER]; }; -static struct cstate_entry *cpu_cstate_entry; /* per CPU ptr */ +static struct cstate_entry __percpu *cpu_cstate_entry; /* per CPU ptr */ static short mwait_supported[ACPI_PROCESSOR_MAX_POWER]; -#define MWAIT_SUBSTATE_MASK (0xf) -#define MWAIT_CSTATE_MASK (0xf) -#define MWAIT_SUBSTATE_SIZE (4) - -#define CPUID_MWAIT_LEAF (5) -#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1) -#define CPUID5_ECX_INTERRUPT_BREAK (0x2) - -#define MWAIT_ECX_INTERRUPT_BREAK (0x1) - #define NATIVE_CSTATE_BEYOND_HALT (2) static long acpi_processor_ffh_cstate_probe_cpu(void *_cx) diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 33cec152070..69fd72aa559 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -7,11 +7,16 @@ #include <linux/acpi.h> #include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/dmi.h> #include <linux/cpumask.h> #include <asm/segment.h> #include <asm/desc.h> +#ifdef CONFIG_X86_32 +#include <asm/pgtable.h> +#endif + #include "realmode/wakeup.h" #include "sleep.h" @@ -90,7 +95,7 @@ int acpi_save_state_mem(void) #ifndef CONFIG_64BIT header->pmode_entry = (u32)&wakeup_pmode_return; - header->pmode_cr3 = (u32)(swsusp_pg_dir - __PAGE_OFFSET); + header->pmode_cr3 = (u32)__pa(&initial_page_table); saved_magic = 0x12345678; #else /* CONFIG_64BIT */ header->trampoline_segment = setup_trampoline() >> 4; @@ -125,7 +130,7 @@ void acpi_restore_state_mem(void) */ void __init acpi_reserve_wakeup_memory(void) { - unsigned long mem; + phys_addr_t mem; if ((&wakeup_code_end - &wakeup_code_start) > WAKEUP_SIZE) { printk(KERN_ERR @@ -133,15 +138,15 @@ void __init acpi_reserve_wakeup_memory(void) return; } - mem = find_e820_area(0, 1<<20, WAKEUP_SIZE, PAGE_SIZE); + mem = memblock_find_in_range(0, 1<<20, WAKEUP_SIZE, PAGE_SIZE); - if (mem == -1L) { + if (mem == MEMBLOCK_ERROR) { printk(KERN_ERR "ACPI: Cannot allocate lowmem, S3 disabled.\n"); return; } acpi_realmode = (unsigned long) phys_to_virt(mem); acpi_wakeup_address = mem; - reserve_early(mem, mem + WAKEUP_SIZE, "ACPI WAKEUP"); + memblock_x86_reserve_range(mem, mem + WAKEUP_SIZE, "ACPI WAKEUP"); } diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index f65ab8b014c..a36bb90aef5 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -195,7 +195,7 @@ static void __init_or_module add_nops(void *insns, unsigned int len) extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; extern s32 __smp_locks[], __smp_locks_end[]; -static void *text_poke_early(void *addr, const void *opcode, size_t len); +void *text_poke_early(void *addr, const void *opcode, size_t len); /* Replace instructions with better alternatives for this CPU type. This runs before SMP is initialized to avoid SMP problems with @@ -522,7 +522,7 @@ void __init alternative_instructions(void) * instructions. And on the local CPU you need to be protected again NMI or MCE * handlers seeing an inconsistent instruction while you patch. */ -static void *__init_or_module text_poke_early(void *addr, const void *opcode, +void *__init_or_module text_poke_early(void *addr, const void *opcode, size_t len) { unsigned long flags; @@ -637,7 +637,72 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len) tpp.len = len; atomic_set(&stop_machine_first, 1); wrote_text = 0; - stop_machine(stop_machine_text_poke, (void *)&tpp, NULL); + /* Use __stop_machine() because the caller already got online_cpus. */ + __stop_machine(stop_machine_text_poke, (void *)&tpp, NULL); return addr; } +#if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL) + +unsigned char ideal_nop5[IDEAL_NOP_SIZE_5]; + +void __init arch_init_ideal_nop5(void) +{ + extern const unsigned char ftrace_test_p6nop[]; + extern const unsigned char ftrace_test_nop5[]; + extern const unsigned char ftrace_test_jmp[]; + int faulted = 0; + + /* + * 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. + */ + 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("converting mcount calls to 0f 1f 44 00 00\n"); + memcpy(ideal_nop5, ftrace_test_p6nop, IDEAL_NOP_SIZE_5); + break; + case 1: + pr_info("converting mcount calls to 66 66 66 66 90\n"); + memcpy(ideal_nop5, ftrace_test_nop5, IDEAL_NOP_SIZE_5); + break; + case 2: + pr_info("converting mcount calls to jmp . + 5\n"); + memcpy(ideal_nop5, ftrace_test_jmp, IDEAL_NOP_SIZE_5); + break; + } + +} +#endif diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index fa044e1e30a..d2fdb0826df 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2007-2009 Advanced Micro Devices, Inc. + * Copyright (C) 2007-2010 Advanced Micro Devices, Inc. * Author: Joerg Roedel <joerg.roedel@amd.com> * Leo Duran <leo.duran@amd.com> * @@ -1953,6 +1953,7 @@ static void __unmap_single(struct dma_ops_domain *dma_dom, size_t size, int dir) { + dma_addr_t flush_addr; dma_addr_t i, start; unsigned int pages; @@ -1960,6 +1961,7 @@ static void __unmap_single(struct dma_ops_domain *dma_dom, (dma_addr + size > dma_dom->aperture_size)) return; + flush_addr = dma_addr; pages = iommu_num_pages(dma_addr, size, PAGE_SIZE); dma_addr &= PAGE_MASK; start = dma_addr; @@ -1974,7 +1976,7 @@ static void __unmap_single(struct dma_ops_domain *dma_dom, dma_ops_free_addresses(dma_dom, dma_addr, pages); if (amd_iommu_unmap_flush || dma_dom->need_flush) { - iommu_flush_pages(&dma_dom->domain, dma_addr, size); + iommu_flush_pages(&dma_dom->domain, flush_addr, size); dma_dom->need_flush = false; } } diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 3cc63e2b8dd..6e11c813415 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2007-2009 Advanced Micro Devices, Inc. + * Copyright (C) 2007-2010 Advanced Micro Devices, Inc. * Author: Joerg Roedel <joerg.roedel@amd.com> * Leo Duran <leo.duran@amd.com> * @@ -31,7 +31,7 @@ #include <asm/iommu.h> #include <asm/gart.h> #include <asm/x86_init.h> - +#include <asm/iommu_table.h> /* * definitions for the ACPI scanning code */ @@ -194,6 +194,39 @@ static inline unsigned long tbl_size(int entry_size) return 1UL << shift; } +/* Access to l1 and l2 indexed register spaces */ + +static u32 iommu_read_l1(struct amd_iommu *iommu, u16 l1, u8 address) +{ + u32 val; + + pci_write_config_dword(iommu->dev, 0xf8, (address | l1 << 16)); + pci_read_config_dword(iommu->dev, 0xfc, &val); + return val; +} + +static void iommu_write_l1(struct amd_iommu *iommu, u16 l1, u8 address, u32 val) +{ + pci_write_config_dword(iommu->dev, 0xf8, (address | l1 << 16 | 1 << 31)); + pci_write_config_dword(iommu->dev, 0xfc, val); + pci_write_config_dword(iommu->dev, 0xf8, (address | l1 << 16)); +} + +static u32 iommu_read_l2(struct amd_iommu *iommu, u8 address) +{ + u32 val; + + pci_write_config_dword(iommu->dev, 0xf0, address); + pci_read_config_dword(iommu->dev, 0xf4, &val); + return val; +} + +static void iommu_write_l2(struct amd_iommu *iommu, u8 address, u32 val) +{ + pci_write_config_dword(iommu->dev, 0xf0, (address | 1 << 8)); + pci_write_config_dword(iommu->dev, 0xf4, val); +} + /**************************************************************************** * * AMD IOMMU MMIO register space handling functions @@ -619,6 +652,7 @@ static void __init init_iommu_from_pci(struct amd_iommu *iommu) { int cap_ptr = iommu->cap_ptr; u32 range, misc; + int i, j; pci_read_config_dword(iommu->dev, cap_ptr + MMIO_CAP_HDR_OFFSET, &iommu->cap); @@ -632,6 +666,30 @@ static void __init init_iommu_from_pci(struct amd_iommu *iommu) iommu->last_device = calc_devid(MMIO_GET_BUS(range), MMIO_GET_LD(range)); iommu->evt_msi_num = MMIO_MSI_NUM(misc); + + if (!is_rd890_iommu(iommu->dev)) + return; + + /* + * Some rd890 systems may not be fully reconfigured by the BIOS, so + * it's necessary for us to store this information so it can be + * reprogrammed on resume + */ + + pci_read_config_dword(iommu->dev, iommu->cap_ptr + 4, + &iommu->stored_addr_lo); + pci_read_config_dword(iommu->dev, iommu->cap_ptr + 8, + &iommu->stored_addr_hi); + + /* Low bit locks writes to configuration space */ + iommu->stored_addr_lo &= ~1; + + for (i = 0; i < 6; i++) + for (j = 0; j < 0x12; j++) + iommu->stored_l1[i][j] = iommu_read_l1(iommu, i, j); + + for (i = 0; i < 0x83; i++) + iommu->stored_l2[i] = iommu_read_l2(iommu, i); } /* @@ -649,29 +707,9 @@ static void __init init_iommu_from_acpi(struct amd_iommu *iommu, struct ivhd_entry *e; /* - * First set the recommended feature enable bits from ACPI - * into the IOMMU control registers - */ - h->flags & IVHD_FLAG_HT_TUN_EN_MASK ? - iommu_feature_enable(iommu, CONTROL_HT_TUN_EN) : - iommu_feature_disable(iommu, CONTROL_HT_TUN_EN); - - h->flags & IVHD_FLAG_PASSPW_EN_MASK ? - iommu_feature_enable(iommu, CONTROL_PASSPW_EN) : - iommu_feature_disable(iommu, CONTROL_PASSPW_EN); - - h->flags & IVHD_FLAG_RESPASSPW_EN_MASK ? - iommu_feature_enable(iommu, CONTROL_RESPASSPW_EN) : - iommu_feature_disable(iommu, CONTROL_RESPASSPW_EN); - - h->flags & IVHD_FLAG_ISOC_EN_MASK ? - iommu_feature_enable(iommu, CONTROL_ISOC_EN) : - iommu_feature_disable(iommu, CONTROL_ISOC_EN); - - /* - * make IOMMU memory accesses cache coherent + * First save the recommended feature enable bits from ACPI */ - iommu_feature_enable(iommu, CONTROL_COHERENT_EN); + iommu->acpi_flags = h->flags; /* * Done. Now parse the device entries @@ -1116,6 +1154,79 @@ static void init_device_table(void) } } +static void iommu_init_flags(struct amd_iommu *iommu) +{ + iommu->acpi_flags & IVHD_FLAG_HT_TUN_EN_MASK ? + iommu_feature_enable(iommu, CONTROL_HT_TUN_EN) : + iommu_feature_disable(iommu, CONTROL_HT_TUN_EN); + + iommu->acpi_flags & IVHD_FLAG_PASSPW_EN_MASK ? + iommu_feature_enable(iommu, CONTROL_PASSPW_EN) : + iommu_feature_disable(iommu, CONTROL_PASSPW_EN); + + iommu->acpi_flags & IVHD_FLAG_RESPASSPW_EN_MASK ? + iommu_feature_enable(iommu, CONTROL_RESPASSPW_EN) : + iommu_feature_disable(iommu, CONTROL_RESPASSPW_EN); + + iommu->acpi_flags & IVHD_FLAG_ISOC_EN_MASK ? + iommu_feature_enable(iommu, CONTROL_ISOC_EN) : + iommu_feature_disable(iommu, CONTROL_ISOC_EN); + + /* + * make IOMMU memory accesses cache coherent + */ + iommu_feature_enable(iommu, CONTROL_COHERENT_EN); +} + +static void iommu_apply_resume_quirks(struct amd_iommu *iommu) +{ + int i, j; + u32 ioc_feature_control; + struct pci_dev *pdev = NULL; + + /* RD890 BIOSes may not have completely reconfigured the iommu */ + if (!is_rd890_iommu(iommu->dev)) + return; + + /* + * First, we need to ensure that the iommu is enabled. This is + * controlled by a register in the northbridge + */ + pdev = pci_get_bus_and_slot(iommu->dev->bus->number, PCI_DEVFN(0, 0)); + + if (!pdev) + return; + + /* Select Northbridge indirect register 0x75 and enable writing */ + pci_write_config_dword(pdev, 0x60, 0x75 | (1 << 7)); + pci_read_config_dword(pdev, 0x64, &ioc_feature_control); + + /* Enable the iommu */ + if (!(ioc_feature_control & 0x1)) + pci_write_config_dword(pdev, 0x64, ioc_feature_control | 1); + + pci_dev_put(pdev); + + /* Restore the iommu BAR */ + pci_write_config_dword(iommu->dev, iommu->cap_ptr + 4, + iommu->stored_addr_lo); + pci_write_config_dword(iommu->dev, iommu->cap_ptr + 8, + iommu->stored_addr_hi); + + /* Restore the l1 indirect regs for each of the 6 l1s */ + for (i = 0; i < 6; i++) + for (j = 0; j < 0x12; j++) + iommu_write_l1(iommu, i, j, iommu->stored_l1[i][j]); + + /* Restore the l2 indirect regs */ + for (i = 0; i < 0x83; i++) + iommu_write_l2(iommu, i, iommu->stored_l2[i]); + + /* Lock PCI setup registers */ + pci_write_config_dword(iommu->dev, iommu->cap_ptr + 4, + iommu->stored_addr_lo | 1); +} + /* * This function finally enables all IOMMUs found in the system after * they have been initialized @@ -1126,6 +1237,7 @@ static void enable_iommus(void) for_each_iommu(iommu) { iommu_disable(iommu); + iommu_init_flags(iommu); iommu_set_device_table(iommu); iommu_enable_command_buffer(iommu); iommu_enable_event_buffer(iommu); @@ -1150,6 +1262,11 @@ static void disable_iommus(void) static int amd_iommu_resume(struct sys_device *dev) { + struct amd_iommu *iommu; + + for_each_iommu(iommu) + iommu_apply_resume_quirks(iommu); + /* re-load the hardware */ enable_iommus(); @@ -1382,13 +1499,13 @@ static int __init early_amd_iommu_detect(struct acpi_table_header *table) return 0; } -void __init amd_iommu_detect(void) +int __init amd_iommu_detect(void) { if (no_iommu || (iommu_detected && !gart_iommu_aperture)) - return; + return -ENODEV; if (amd_iommu_disabled) - return; + return -ENODEV; if (acpi_table_parse("IVRS", early_amd_iommu_detect) == 0) { iommu_detected = 1; @@ -1397,7 +1514,9 @@ void __init amd_iommu_detect(void) /* Make sure ACS will be enabled */ pci_request_acs(); + return 1; } + return -ENODEV; } /**************************************************************************** @@ -1428,3 +1547,8 @@ static int __init parse_amd_iommu_options(char *str) __setup("amd_iommu_dump", parse_amd_iommu_dump); __setup("amd_iommu=", parse_amd_iommu_options); + +IOMMU_INIT_FINISH(amd_iommu_detect, + gart_iommu_hole_init, + 0, + 0); diff --git a/arch/x86/kernel/k8.c b/arch/x86/kernel/amd_nb.c index 0f7bc20cfcd..8f6463d8ed0 100644 --- a/arch/x86/kernel/k8.c +++ b/arch/x86/kernel/amd_nb.c @@ -8,21 +8,19 @@ #include <linux/errno.h> #include <linux/module.h> #include <linux/spinlock.h> -#include <asm/k8.h> - -int num_k8_northbridges; -EXPORT_SYMBOL(num_k8_northbridges); +#include <asm/amd_nb.h> 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_15H_NB_MISC) }, {} }; EXPORT_SYMBOL(k8_nb_ids); -struct pci_dev **k8_northbridges; +struct k8_northbridge_info k8_northbridges; EXPORT_SYMBOL(k8_northbridges); static struct pci_dev *next_k8_northbridge(struct pci_dev *dev) @@ -40,36 +38,45 @@ int cache_k8_northbridges(void) int i; struct pci_dev *dev; - if (num_k8_northbridges) + if (k8_northbridges.num) return 0; dev = NULL; while ((dev = next_k8_northbridge(dev)) != NULL) - num_k8_northbridges++; + k8_northbridges.num++; + + /* some CPU families (e.g. family 0x11) do not support GART */ + if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10 || + boot_cpu_data.x86 == 0x15) + k8_northbridges.gart_supported = 1; - k8_northbridges = kmalloc((num_k8_northbridges + 1) * sizeof(void *), - GFP_KERNEL); - if (!k8_northbridges) + k8_northbridges.nb_misc = kmalloc((k8_northbridges.num + 1) * + sizeof(void *), GFP_KERNEL); + if (!k8_northbridges.nb_misc) return -ENOMEM; - if (!num_k8_northbridges) { - k8_northbridges[0] = NULL; + if (!k8_northbridges.num) { + k8_northbridges.nb_misc[0] = NULL; return 0; } - flush_words = kmalloc(num_k8_northbridges * sizeof(u32), GFP_KERNEL); - if (!flush_words) { - kfree(k8_northbridges); - return -ENOMEM; + if (k8_northbridges.gart_supported) { + flush_words = kmalloc(k8_northbridges.num * sizeof(u32), + GFP_KERNEL); + if (!flush_words) { + kfree(k8_northbridges.nb_misc); + return -ENOMEM; + } } dev = NULL; i = 0; while ((dev = next_k8_northbridge(dev)) != NULL) { - k8_northbridges[i] = dev; - pci_read_config_dword(dev, 0x9c, &flush_words[i++]); + k8_northbridges.nb_misc[i] = dev; + if (k8_northbridges.gart_supported) + pci_read_config_dword(dev, 0x9c, &flush_words[i++]); } - k8_northbridges[i] = NULL; + k8_northbridges.nb_misc[i] = NULL; return 0; } EXPORT_SYMBOL_GPL(cache_k8_northbridges); @@ -93,22 +100,25 @@ void k8_flush_garts(void) unsigned long flags; static DEFINE_SPINLOCK(gart_lock); + if (!k8_northbridges.gart_supported) + return; + /* Avoid races between AGP and IOMMU. In theory it's not needed but I'm not sure if the hardware won't lose flush requests when another is pending. This whole thing is so expensive anyways that it doesn't matter to serialize more. -AK */ spin_lock_irqsave(&gart_lock, flags); flushed = 0; - for (i = 0; i < num_k8_northbridges; i++) { - pci_write_config_dword(k8_northbridges[i], 0x9c, + for (i = 0; i < k8_northbridges.num; i++) { + pci_write_config_dword(k8_northbridges.nb_misc[i], 0x9c, flush_words[i]|1); flushed++; } - for (i = 0; i < num_k8_northbridges; i++) { + for (i = 0; i < k8_northbridges.num; i++) { u32 w; /* Make sure the hardware actually executed the flush*/ for (;;) { - pci_read_config_dword(k8_northbridges[i], + pci_read_config_dword(k8_northbridges.nb_misc[i], 0x9c, &w); if (!(w & 1)) break; diff --git a/arch/x86/kernel/apb_timer.c b/arch/x86/kernel/apb_timer.c index 8dd77800ff5..92543c73cf8 100644 --- a/arch/x86/kernel/apb_timer.c +++ b/arch/x86/kernel/apb_timer.c @@ -231,34 +231,6 @@ static void apbt_restart_clocksource(struct clocksource *cs) apbt_start_counter(phy_cs_timer_id); } -/* Setup IRQ routing via IOAPIC */ -#ifdef CONFIG_SMP -static void apbt_setup_irq(struct apbt_dev *adev) -{ - struct irq_chip *chip; - struct irq_desc *desc; - - /* timer0 irq has been setup early */ - if (adev->irq == 0) - return; - desc = irq_to_desc(adev->irq); - chip = get_irq_chip(adev->irq); - disable_irq(adev->irq); - desc->status |= IRQ_MOVE_PCNTXT; - irq_set_affinity(adev->irq, cpumask_of(adev->cpu)); - /* APB timer irqs are set up as mp_irqs, timer is edge triggerred */ - set_irq_chip_and_handler_name(adev->irq, chip, handle_edge_irq, "edge"); - enable_irq(adev->irq); - if (system_state == SYSTEM_BOOTING) - if (request_irq(adev->irq, apbt_interrupt_handler, - IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING, - adev->name, adev)) { - printk(KERN_ERR "Failed request IRQ for APBT%d\n", - adev->num); - } -} -#endif - static void apbt_enable_int(int n) { unsigned long ctrl = apbt_readl(n, APBTMR_N_CONTROL); @@ -334,6 +306,27 @@ static int __init apbt_clockevent_register(void) } #ifdef CONFIG_SMP + +static void apbt_setup_irq(struct apbt_dev *adev) +{ + /* timer0 irq has been setup early */ + if (adev->irq == 0) + return; + + if (system_state == SYSTEM_BOOTING) { + irq_modify_status(adev->irq, 0, IRQ_MOVE_PCNTXT); + /* APB timer irqs are set up as mp_irqs, timer is edge type */ + __set_irq_handler(adev->irq, handle_edge_irq, 0, "edge"); + if (request_irq(adev->irq, apbt_interrupt_handler, + IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING, + adev->name, adev)) { + printk(KERN_ERR "Failed request IRQ for APBT%d\n", + adev->num); + } + } else + enable_irq(adev->irq); +} + /* Should be called with per cpu */ void apbt_setup_secondary_clock(void) { @@ -343,7 +336,7 @@ void apbt_setup_secondary_clock(void) /* Don't register boot CPU clockevent */ cpu = smp_processor_id(); - if (cpu == boot_cpu_id) + if (!cpu) return; /* * We need to calculate the scaled math multiplication factor for @@ -389,16 +382,17 @@ static int apbt_cpuhp_notify(struct notifier_block *n, switch (action & 0xf) { case CPU_DEAD: + disable_irq(adev->irq); apbt_disable_int(cpu); - if (system_state == SYSTEM_RUNNING) + if (system_state == SYSTEM_RUNNING) { pr_debug("skipping APBT CPU %lu offline\n", cpu); - else if (adev) { + } else if (adev) { pr_debug("APBT clockevent for cpu %lu offline\n", cpu); free_irq(adev->irq, adev); } break; default: - pr_debug(KERN_INFO "APBT notified %lu, no action\n", action); + pr_debug("APBT notified %lu, no action\n", action); } return NOTIFY_OK; } @@ -552,7 +546,7 @@ bad_count: pr_debug("APB CS going back %lx:%lx:%lx ", t2, last_read, t2 - last_read); bad_count_x3: - pr_debug(KERN_INFO "tripple check enforced\n"); + pr_debug("triple check enforced\n"); t0 = apbt_readl(phy_cs_timer_id, APBTMR_N_CURRENT_VALUE); udelay(1); diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index a2e0caf26e1..b3a16e8f070 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -27,7 +27,7 @@ #include <asm/gart.h> #include <asm/pci-direct.h> #include <asm/dma.h> -#include <asm/k8.h> +#include <asm/amd_nb.h> #include <asm/x86_init.h> int gart_iommu_aperture; @@ -307,7 +307,7 @@ void __init early_gart_iommu_check(void) continue; ctl = read_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL); - aper_enabled = ctl & AMD64_GARTEN; + aper_enabled = ctl & GARTEN; aper_order = (ctl >> 1) & 7; aper_size = (32 * 1024 * 1024) << aper_order; aper_base = read_pci_config(bus, slot, 3, AMD64_GARTAPERTUREBASE) & 0x7fff; @@ -362,7 +362,7 @@ void __init early_gart_iommu_check(void) continue; ctl = read_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL); - ctl &= ~AMD64_GARTEN; + ctl &= ~GARTEN; write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, ctl); } } @@ -371,7 +371,7 @@ void __init early_gart_iommu_check(void) static int __initdata printed_gart_size_msg; -void __init gart_iommu_hole_init(void) +int __init gart_iommu_hole_init(void) { u32 agp_aper_base = 0, agp_aper_order = 0; u32 aper_size, aper_alloc = 0, aper_order = 0, last_aper_order = 0; @@ -381,7 +381,7 @@ void __init gart_iommu_hole_init(void) if (gart_iommu_aperture_disabled || !fix_aperture || !early_pci_allowed()) - return; + return -ENODEV; printk(KERN_INFO "Checking aperture...\n"); @@ -463,8 +463,9 @@ out: unsigned long n = (32 * 1024 * 1024) << last_aper_order; insert_aperture_resource((u32)last_aper_base, n); + return 1; } - return; + return 0; } if (!fallback_aper_force) { @@ -500,13 +501,18 @@ out: panic("Not enough memory for aperture"); } } else { - return; + return 0; } /* Fix up the north bridges */ for (i = 0; i < ARRAY_SIZE(bus_dev_ranges); i++) { - int bus; - int dev_base, dev_limit; + int bus, dev_base, dev_limit; + + /* + * Don't enable translation yet but enable GART IO and CPU + * accesses and set DISTLBWALKPRB since GART table memory is UC. + */ + u32 ctl = DISTLBWALKPRB | aper_order << 1; bus = bus_dev_ranges[i].bus; dev_base = bus_dev_ranges[i].dev_base; @@ -515,13 +521,12 @@ out: if (!early_is_k8_nb(read_pci_config(bus, slot, 3, 0x00))) continue; - /* Don't enable translation yet. That is done later. - Assume this BIOS didn't initialise the GART so - just overwrite all previous bits */ - write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, aper_order << 1); + write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, ctl); write_pci_config(bus, slot, 3, AMD64_GARTAPERTUREBASE, aper_alloc >> 25); } } set_up_gart_resume(aper_order, aper_alloc); + + return 1; } diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 980508c7908..850657d1b0e 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -52,6 +52,7 @@ #include <asm/mce.h> #include <asm/kvm_para.h> #include <asm/tsc.h> +#include <asm/atomic.h> unsigned int num_processors; @@ -370,38 +371,87 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen) } /* - * Setup extended LVT, AMD specific (K8, family 10h) + * Setup extended LVT, AMD specific * - * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and - * MCE interrupts are supported. Thus MCE offset must be set to 0. + * Software should use the LVT offsets the BIOS provides. The offsets + * are determined by the subsystems using it like those for MCE + * threshold or IBS. On K8 only offset 0 (APIC500) and MCE interrupts + * are supported. Beginning with family 10h at least 4 offsets are + * available. * - * If mask=1, the LVT entry does not generate interrupts while mask=0 - * enables the vector. See also the BKDGs. + * Since the offsets must be consistent for all cores, we keep track + * of the LVT offsets in software and reserve the offset for the same + * vector also to be used on other cores. An offset is freed by + * setting the entry to APIC_EILVT_MASKED. + * + * If the BIOS is right, there should be no conflicts. Otherwise a + * "[Firmware Bug]: ..." error message is generated. However, if + * software does not properly determines the offsets, it is not + * necessarily a BIOS bug. */ -#define APIC_EILVT_LVTOFF_MCE 0 -#define APIC_EILVT_LVTOFF_IBS 1 +static atomic_t eilvt_offsets[APIC_EILVT_NR_MAX]; -static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask) +static inline int eilvt_entry_is_changeable(unsigned int old, unsigned int new) { - unsigned long reg = (lvt_off << 4) + APIC_EILVTn(0); - unsigned int v = (mask << 16) | (msg_type << 8) | vector; - - apic_write(reg, v); + return (old & APIC_EILVT_MASKED) + || (new == APIC_EILVT_MASKED) + || ((new & ~APIC_EILVT_MASKED) == old); } -u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask) +static unsigned int reserve_eilvt_offset(int offset, unsigned int new) { - setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask); - return APIC_EILVT_LVTOFF_MCE; + unsigned int rsvd; /* 0: uninitialized */ + + if (offset >= APIC_EILVT_NR_MAX) + return ~0; + + rsvd = atomic_read(&eilvt_offsets[offset]) & ~APIC_EILVT_MASKED; + do { + if (rsvd && + !eilvt_entry_is_changeable(rsvd, new)) + /* may not change if vectors are different */ + return rsvd; + rsvd = atomic_cmpxchg(&eilvt_offsets[offset], rsvd, new); + } while (rsvd != new); + + return new; } -u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask) +/* + * If mask=1, the LVT entry does not generate interrupts while mask=0 + * enables the vector. See also the BKDGs. + */ + +int setup_APIC_eilvt(u8 offset, u8 vector, u8 msg_type, u8 mask) { - setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask); - return APIC_EILVT_LVTOFF_IBS; + unsigned long reg = APIC_EILVTn(offset); + unsigned int new, old, reserved; + + new = (mask << 16) | (msg_type << 8) | vector; + old = apic_read(reg); + reserved = reserve_eilvt_offset(offset, new); + + if (reserved != new) { + pr_err(FW_BUG "cpu %d, try to setup vector 0x%x, but " + "vector 0x%x was already reserved by another core, " + "APIC%lX=0x%x\n", + smp_processor_id(), new, reserved, reg, old); + return -EINVAL; + } + + if (!eilvt_entry_is_changeable(old, new)) { + pr_err(FW_BUG "cpu %d, try to setup vector 0x%x but " + "register already in use, APIC%lX=0x%x\n", + smp_processor_id(), new, reg, old); + return -EBUSY; + } + + apic_write(reg, new); + + return 0; } -EXPORT_SYMBOL_GPL(setup_APIC_eilvt_ibs); +EXPORT_SYMBOL_GPL(setup_APIC_eilvt); /* * Program the next event, relative to now @@ -1606,7 +1656,7 @@ void __init init_apic_mappings(void) * acpi lapic path already maps that address in * acpi_register_lapic_address() */ - if (!acpi_lapic) + if (!acpi_lapic && !smp_found_config) set_fixmap_nocache(FIX_APIC_BASE, apic_phys); apic_printk(APIC_VERBOSE, "mapped APIC to %08lx (%08lx)\n", @@ -1665,10 +1715,7 @@ int __init APIC_init_uniprocessor(void) } #endif -#ifndef CONFIG_SMP - enable_IR_x2apic(); default_setup_apic_routing(); -#endif verify_local_APIC(); connect_bsp_APIC(); diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 4dc0084ec1b..8ae808d110f 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -131,13 +131,9 @@ struct irq_pin_list { struct irq_pin_list *next; }; -static struct irq_pin_list *get_one_free_irq_2_pin(int node) +static struct irq_pin_list *alloc_irq_pin_list(int node) { - struct irq_pin_list *pin; - - pin = kzalloc_node(sizeof(*pin), GFP_ATOMIC, node); - - return pin; + return kzalloc_node(sizeof(struct irq_pin_list), GFP_KERNEL, node); } /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */ @@ -150,10 +146,7 @@ static struct irq_cfg irq_cfgx[NR_IRQS]; int __init arch_early_irq_init(void) { struct irq_cfg *cfg; - struct irq_desc *desc; - int count; - int node; - int i; + int count, node, i; if (!legacy_pic->nr_legacy_irqs) { nr_irqs_gsi = 0; @@ -162,13 +155,15 @@ int __init arch_early_irq_init(void) cfg = irq_cfgx; count = ARRAY_SIZE(irq_cfgx); - node= cpu_to_node(boot_cpu_id); + node = cpu_to_node(0); + + /* Make sure the legacy interrupts are marked in the bitmap */ + irq_reserve_irqs(0, legacy_pic->nr_legacy_irqs); for (i = 0; i < count; i++) { - desc = irq_to_desc(i); - desc->chip_data = &cfg[i]; - zalloc_cpumask_var_node(&cfg[i].domain, GFP_NOWAIT, node); - zalloc_cpumask_var_node(&cfg[i].old_domain, GFP_NOWAIT, node); + set_irq_chip_data(i, &cfg[i]); + zalloc_cpumask_var_node(&cfg[i].domain, GFP_KERNEL, node); + zalloc_cpumask_var_node(&cfg[i].old_domain, GFP_KERNEL, node); /* * For legacy IRQ's, start with assigning irq0 to irq15 to * IRQ0_VECTOR to IRQ15_VECTOR on cpu 0. @@ -183,165 +178,88 @@ int __init arch_early_irq_init(void) } #ifdef CONFIG_SPARSE_IRQ -struct irq_cfg *irq_cfg(unsigned int irq) +static struct irq_cfg *irq_cfg(unsigned int irq) { - struct irq_cfg *cfg = NULL; - struct irq_desc *desc; - - desc = irq_to_desc(irq); - if (desc) - cfg = desc->chip_data; - - return cfg; + return get_irq_chip_data(irq); } -static struct irq_cfg *get_one_free_irq_cfg(int node) +static struct irq_cfg *alloc_irq_cfg(unsigned int irq, int node) { struct irq_cfg *cfg; - cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node); - if (cfg) { - if (!zalloc_cpumask_var_node(&cfg->domain, GFP_ATOMIC, node)) { - kfree(cfg); - cfg = NULL; - } else if (!zalloc_cpumask_var_node(&cfg->old_domain, - GFP_ATOMIC, node)) { - free_cpumask_var(cfg->domain); - kfree(cfg); - cfg = NULL; - } - } - + cfg = kzalloc_node(sizeof(*cfg), GFP_KERNEL, node); + if (!cfg) + return NULL; + if (!zalloc_cpumask_var_node(&cfg->domain, GFP_KERNEL, node)) + goto out_cfg; + if (!zalloc_cpumask_var_node(&cfg->old_domain, GFP_KERNEL, node)) + goto out_domain; return cfg; +out_domain: + free_cpumask_var(cfg->domain); +out_cfg: + kfree(cfg); + return NULL; } -int arch_init_chip_data(struct irq_desc *desc, int node) -{ - struct irq_cfg *cfg; - - cfg = desc->chip_data; - if (!cfg) { - desc->chip_data = get_one_free_irq_cfg(node); - if (!desc->chip_data) { - printk(KERN_ERR "can not alloc irq_cfg\n"); - BUG_ON(1); - } - } - - return 0; -} - -/* for move_irq_desc */ -static void -init_copy_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg, int node) +static void free_irq_cfg(unsigned int at, struct irq_cfg *cfg) { - struct irq_pin_list *old_entry, *head, *tail, *entry; - - cfg->irq_2_pin = NULL; - old_entry = old_cfg->irq_2_pin; - if (!old_entry) - return; - - entry = get_one_free_irq_2_pin(node); - if (!entry) + if (!cfg) return; + set_irq_chip_data(at, NULL); + free_cpumask_var(cfg->domain); + free_cpumask_var(cfg->old_domain); + kfree(cfg); +} - 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(node); - 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; - } +#else - tail->next = NULL; - cfg->irq_2_pin = head; +struct irq_cfg *irq_cfg(unsigned int irq) +{ + return irq < nr_irqs ? irq_cfgx + irq : NULL; } -static void free_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg) +static struct irq_cfg *alloc_irq_cfg(unsigned int irq, int node) { - struct irq_pin_list *entry, *next; - - if (old_cfg->irq_2_pin == cfg->irq_2_pin) - return; + return irq_cfgx + irq; +} - entry = old_cfg->irq_2_pin; +static inline void free_irq_cfg(unsigned int at, struct irq_cfg *cfg) { } - while (entry) { - next = entry->next; - kfree(entry); - entry = next; - } - old_cfg->irq_2_pin = NULL; -} +#endif -void arch_init_copy_chip_data(struct irq_desc *old_desc, - struct irq_desc *desc, int node) +static struct irq_cfg *alloc_irq_and_cfg_at(unsigned int at, int node) { + int res = irq_alloc_desc_at(at, node); struct irq_cfg *cfg; - struct irq_cfg *old_cfg; - - cfg = get_one_free_irq_cfg(node); - - 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, node); -} + if (res < 0) { + if (res != -EEXIST) + return NULL; + cfg = get_irq_chip_data(at); + if (cfg) + return cfg; + } -static void free_irq_cfg(struct irq_cfg *old_cfg) -{ - kfree(old_cfg); + cfg = alloc_irq_cfg(at, node); + if (cfg) + set_irq_chip_data(at, cfg); + else + irq_free_desc(at); + return cfg; } -void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) +static int alloc_irq_from(unsigned int from, int node) { - 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; - } + return irq_alloc_desc_from(from, node); } -/* end for move_irq_desc */ -#else -struct irq_cfg *irq_cfg(unsigned int irq) +static void free_irq_at(unsigned int at, struct irq_cfg *cfg) { - return irq < nr_irqs ? irq_cfgx + irq : NULL; + free_irq_cfg(at, cfg); + irq_free_desc(at); } -#endif - struct io_apic { unsigned int index; unsigned int unused[3]; @@ -446,7 +364,7 @@ __ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) io_apic_write(apic, 0x10 + 2*pin, eu.w1); } -void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) +static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { unsigned long flags; raw_spin_lock_irqsave(&ioapic_lock, flags); @@ -476,7 +394,7 @@ static void ioapic_mask_entry(int apic, int pin) * fast in the common case, and fast for shared ISA-space IRQs. */ static int -add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin) +__add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin) { struct irq_pin_list **last, *entry; @@ -488,7 +406,7 @@ add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin) last = &entry->next; } - entry = get_one_free_irq_2_pin(node); + entry = alloc_irq_pin_list(node); if (!entry) { printk(KERN_ERR "can not alloc irq_pin_list (%d,%d,%d)\n", node, apic, pin); @@ -503,7 +421,7 @@ add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin) static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin) { - if (add_pin_to_irq_node_nopanic(cfg, node, apic, pin)) + if (__add_pin_to_irq_node(cfg, node, apic, pin)) panic("IO-APIC: failed to add irq-pin. Can not proceed\n"); } @@ -566,11 +484,6 @@ static void __unmask_and_level_IO_APIC_irq(struct irq_pin_list *entry) IO_APIC_REDIR_LEVEL_TRIGGER, NULL); } -static void __unmask_IO_APIC_irq(struct irq_cfg *cfg) -{ - io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL); -} - static void io_apic_sync(struct irq_pin_list *entry) { /* @@ -582,44 +495,37 @@ static void io_apic_sync(struct irq_pin_list *entry) readl(&io_apic->data); } -static void __mask_IO_APIC_irq(struct irq_cfg *cfg) +static void mask_ioapic(struct irq_cfg *cfg) { + unsigned long flags; + + raw_spin_lock_irqsave(&ioapic_lock, flags); io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); } -static void mask_IO_APIC_irq_desc(struct irq_desc *desc) +static void mask_ioapic_irq(struct irq_data *data) { - struct irq_cfg *cfg = desc->chip_data; - unsigned long flags; - - BUG_ON(!cfg); + mask_ioapic(data->chip_data); +} - raw_spin_lock_irqsave(&ioapic_lock, flags); - __mask_IO_APIC_irq(cfg); - raw_spin_unlock_irqrestore(&ioapic_lock, flags); +static void __unmask_ioapic(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL); } -static void unmask_IO_APIC_irq_desc(struct irq_desc *desc) +static void unmask_ioapic(struct irq_cfg *cfg) { - struct irq_cfg *cfg = desc->chip_data; unsigned long flags; raw_spin_lock_irqsave(&ioapic_lock, flags); - __unmask_IO_APIC_irq(cfg); + __unmask_ioapic(cfg); raw_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) +static void unmask_ioapic_irq(struct irq_data *data) { - struct irq_desc *desc = irq_to_desc(irq); - - unmask_IO_APIC_irq_desc(desc); + unmask_ioapic(data->chip_data); } static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) @@ -689,14 +595,14 @@ struct IO_APIC_route_entry **alloc_ioapic_entries(void) struct IO_APIC_route_entry **ioapic_entries; ioapic_entries = kzalloc(sizeof(*ioapic_entries) * nr_ioapics, - GFP_ATOMIC); + GFP_KERNEL); if (!ioapic_entries) return 0; for (apic = 0; apic < nr_ioapics; apic++) { ioapic_entries[apic] = kzalloc(sizeof(struct IO_APIC_route_entry) * - nr_ioapic_registers[apic], GFP_ATOMIC); + nr_ioapic_registers[apic], GFP_KERNEL); if (!ioapic_entries[apic]) goto nomem; } @@ -1254,7 +1160,6 @@ void __setup_vector_irq(int cpu) /* Initialize vector_irq on a new cpu */ int irq, vector; struct irq_cfg *cfg; - struct irq_desc *desc; /* * vector_lock will make sure that we don't run into irq vector @@ -1263,9 +1168,10 @@ void __setup_vector_irq(int cpu) */ raw_spin_lock(&vector_lock); /* Mark the inuse vectors */ - for_each_irq_desc(irq, desc) { - cfg = desc->chip_data; - + for_each_active_irq(irq) { + cfg = get_irq_chip_data(irq); + if (!cfg) + continue; /* * If it is a legacy IRQ handled by the legacy PIC, this cpu * will be part of the irq_cfg's domain. @@ -1322,17 +1228,17 @@ static inline int IO_APIC_irq_trigger(int irq) } #endif -static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long trigger) +static void ioapic_register_intr(unsigned int irq, unsigned long trigger) { if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || trigger == IOAPIC_LEVEL) - desc->status |= IRQ_LEVEL; + irq_set_status_flags(irq, IRQ_LEVEL); else - desc->status &= ~IRQ_LEVEL; + irq_clear_status_flags(irq, IRQ_LEVEL); - if (irq_remapped(irq)) { - desc->status |= IRQ_MOVE_PCNTXT; + if (irq_remapped(get_irq_chip_data(irq))) { + irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); if (trigger) set_irq_chip_and_handler_name(irq, &ir_ioapic_chip, handle_fasteoi_irq, @@ -1353,10 +1259,10 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t handle_edge_irq, "edge"); } -int setup_ioapic_entry(int apic_id, int irq, - struct IO_APIC_route_entry *entry, - unsigned int destination, int trigger, - int polarity, int vector, int pin) +static int setup_ioapic_entry(int apic_id, int irq, + struct IO_APIC_route_entry *entry, + unsigned int destination, int trigger, + int polarity, int vector, int pin) { /* * add it to the IO-APIC irq-routing table: @@ -1377,21 +1283,7 @@ int setup_ioapic_entry(int apic_id, int irq, if (index < 0) panic("Failed to allocate IRTE for ioapic %d\n", apic_id); - memset(&irte, 0, sizeof(irte)); - - irte.present = 1; - irte.dst_mode = apic->irq_dest_mode; - /* - * Trigger mode in the IRTE will always be edge, and the - * actual level or edge trigger will be setup in the IO-APIC - * RTE. This will help simplify level triggered irq migration. - * For more details, see the comments above explainig IO-APIC - * irq migration in the presence of interrupt-remapping. - */ - irte.trigger_mode = 0; - irte.dlvry_mode = apic->irq_delivery_mode; - irte.vector = vector; - irte.dest_id = IRTE_DEST(destination); + prepare_irte(&irte, vector, destination); /* Set source-id of interrupt request */ set_ioapic_sid(&irte, apic_id); @@ -1426,18 +1318,14 @@ int setup_ioapic_entry(int apic_id, int irq, return 0; } -static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq_desc *desc, - int trigger, int polarity) +static void setup_ioapic_irq(int apic_id, int pin, unsigned int irq, + struct irq_cfg *cfg, int trigger, int polarity) { - struct irq_cfg *cfg; struct IO_APIC_route_entry entry; unsigned int dest; if (!IO_APIC_IRQ(irq)) return; - - cfg = desc->chip_data; - /* * For legacy irqs, cfg->domain starts with cpu 0 for legacy * controllers like 8259. Now that IO-APIC can handle this irq, update @@ -1466,9 +1354,9 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq return; } - ioapic_register_intr(irq, desc, trigger); + ioapic_register_intr(irq, trigger); if (irq < legacy_pic->nr_legacy_irqs) - legacy_pic->chip->mask(irq); + legacy_pic->mask(irq); ioapic_write_entry(apic_id, pin, entry); } @@ -1479,11 +1367,9 @@ static struct { static void __init setup_IO_APIC_irqs(void) { - int apic_id, pin, idx, irq; - int notcon = 0; - struct irq_desc *desc; + int apic_id, pin, idx, irq, notcon = 0; + int node = cpu_to_node(0); struct irq_cfg *cfg; - int node = cpu_to_node(boot_cpu_id); apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); @@ -1520,19 +1406,17 @@ static void __init setup_IO_APIC_irqs(void) apic->multi_timer_check(apic_id, irq)) continue; - desc = irq_to_desc_alloc_node(irq, node); - if (!desc) { - printk(KERN_INFO "can not get irq_desc for %d\n", irq); + cfg = alloc_irq_and_cfg_at(irq, node); + if (!cfg) continue; - } - cfg = desc->chip_data; + add_pin_to_irq_node(cfg, node, apic_id, pin); /* * don't mark it in pin_programmed, so later acpi could * set it correctly when irq < 16 */ - setup_IO_APIC_irq(apic_id, pin, irq, desc, - irq_trigger(idx), irq_polarity(idx)); + setup_ioapic_irq(apic_id, pin, irq, cfg, irq_trigger(idx), + irq_polarity(idx)); } if (notcon) @@ -1547,9 +1431,7 @@ static void __init setup_IO_APIC_irqs(void) */ void setup_IO_APIC_irq_extra(u32 gsi) { - int apic_id = 0, pin, idx, irq; - int node = cpu_to_node(boot_cpu_id); - struct irq_desc *desc; + int apic_id = 0, pin, idx, irq, node = cpu_to_node(0); struct irq_cfg *cfg; /* @@ -1565,18 +1447,15 @@ void setup_IO_APIC_irq_extra(u32 gsi) return; irq = pin_2_irq(idx, apic_id, pin); -#ifdef CONFIG_SPARSE_IRQ - desc = irq_to_desc(irq); - if (desc) + + /* Only handle the non legacy irqs on secondary ioapics */ + if (apic_id == 0 || irq < NR_IRQS_LEGACY) return; -#endif - desc = irq_to_desc_alloc_node(irq, node); - if (!desc) { - printk(KERN_INFO "can not get irq_desc for %d\n", irq); + + cfg = alloc_irq_and_cfg_at(irq, node); + if (!cfg) return; - } - cfg = desc->chip_data; add_pin_to_irq_node(cfg, node, apic_id, pin); if (test_bit(pin, mp_ioapic_routing[apic_id].pin_programmed)) { @@ -1586,7 +1465,7 @@ void setup_IO_APIC_irq_extra(u32 gsi) } set_bit(pin, mp_ioapic_routing[apic_id].pin_programmed); - setup_IO_APIC_irq(apic_id, pin, irq, desc, + setup_ioapic_irq(apic_id, pin, irq, cfg, irq_trigger(idx), irq_polarity(idx)); } @@ -1637,7 +1516,6 @@ __apicdebuginit(void) print_IO_APIC(void) union IO_APIC_reg_03 reg_03; unsigned long flags; struct irq_cfg *cfg; - struct irq_desc *desc; unsigned int irq; printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); @@ -1724,10 +1602,12 @@ __apicdebuginit(void) print_IO_APIC(void) } } printk(KERN_DEBUG "IRQ to pin mappings:\n"); - for_each_irq_desc(irq, desc) { + for_each_active_irq(irq) { struct irq_pin_list *entry; - cfg = desc->chip_data; + cfg = get_irq_chip_data(irq); + if (!cfg) + continue; entry = cfg->irq_2_pin; if (!entry) continue; @@ -2232,29 +2112,26 @@ static int __init timer_irq_works(void) * an edge even if it isn't on the 8259A... */ -static unsigned int startup_ioapic_irq(unsigned int irq) +static unsigned int startup_ioapic_irq(struct irq_data *data) { - int was_pending = 0; + int was_pending = 0, irq = data->irq; unsigned long flags; - struct irq_cfg *cfg; raw_spin_lock_irqsave(&ioapic_lock, flags); if (irq < legacy_pic->nr_legacy_irqs) { - legacy_pic->chip->mask(irq); + legacy_pic->mask(irq); if (legacy_pic->irq_pending(irq)) was_pending = 1; } - cfg = irq_cfg(irq); - __unmask_IO_APIC_irq(cfg); + __unmask_ioapic(data->chip_data); raw_spin_unlock_irqrestore(&ioapic_lock, flags); return was_pending; } -static int ioapic_retrigger_irq(unsigned int irq) +static int ioapic_retrigger_irq(struct irq_data *data) { - - struct irq_cfg *cfg = irq_cfg(irq); + struct irq_cfg *cfg = data->chip_data; unsigned long flags; raw_spin_lock_irqsave(&vector_lock, flags); @@ -2305,7 +2182,7 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq * With interrupt-remapping, destination information comes * from interrupt-remapping table entry. */ - if (!irq_remapped(irq)) + if (!irq_remapped(cfg)) io_apic_write(apic, 0x11 + pin*2, dest); reg = io_apic_read(apic, 0x10 + pin*2); reg &= ~IO_APIC_REDIR_VECTOR_MASK; @@ -2315,65 +2192,46 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq } /* - * Either sets desc->affinity to a valid value, and returns + * Either sets data->affinity to a valid value, and returns * ->cpu_mask_to_apicid of that in dest_id, or returns -1 and - * leaves desc->affinity untouched. + * leaves data->affinity untouched. */ -unsigned int -set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask, - unsigned int *dest_id) +int __ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, + unsigned int *dest_id) { - struct irq_cfg *cfg; - unsigned int irq; + struct irq_cfg *cfg = data->chip_data; if (!cpumask_intersects(mask, cpu_online_mask)) return -1; - irq = desc->irq; - cfg = desc->chip_data; - if (assign_irq_vector(irq, cfg, mask)) + if (assign_irq_vector(data->irq, data->chip_data, mask)) return -1; - cpumask_copy(desc->affinity, mask); + cpumask_copy(data->affinity, mask); - *dest_id = apic->cpu_mask_to_apicid_and(desc->affinity, cfg->domain); + *dest_id = apic->cpu_mask_to_apicid_and(mask, cfg->domain); return 0; } static int -set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask) +ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_cfg *cfg; + unsigned int dest, irq = data->irq; unsigned long flags; - unsigned int dest; - unsigned int irq; - int ret = -1; - - irq = desc->irq; - cfg = desc->chip_data; + int ret; raw_spin_lock_irqsave(&ioapic_lock, flags); - ret = set_desc_affinity(desc, mask, &dest); + ret = __ioapic_set_affinity(data, mask, &dest); if (!ret) { /* Only the high 8 bits are valid. */ dest = SET_APIC_LOGICAL_ID(dest); - __target_IO_APIC_irq(irq, dest, cfg); + __target_IO_APIC_irq(irq, dest, data->chip_data); } raw_spin_unlock_irqrestore(&ioapic_lock, flags); - return ret; } -static int -set_ioapic_affinity_irq(unsigned int irq, const struct cpumask *mask) -{ - struct irq_desc *desc; - - desc = irq_to_desc(irq); - - return set_ioapic_affinity_irq_desc(desc, mask); -} - #ifdef CONFIG_INTR_REMAP /* @@ -2388,24 +2246,21 @@ set_ioapic_affinity_irq(unsigned int irq, const struct cpumask *mask) * the interrupt-remapping table entry. */ static int -migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) +ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_cfg *cfg; + struct irq_cfg *cfg = data->chip_data; + unsigned int dest, irq = data->irq; struct irte irte; - unsigned int dest; - unsigned int irq; - int ret = -1; if (!cpumask_intersects(mask, cpu_online_mask)) - return ret; + return -EINVAL; - irq = desc->irq; if (get_irte(irq, &irte)) - return ret; + return -EBUSY; - cfg = desc->chip_data; if (assign_irq_vector(irq, cfg, mask)) - return ret; + return -EBUSY; dest = apic->cpu_mask_to_apicid_and(cfg->domain, mask); @@ -2420,29 +2275,14 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) if (cfg->move_in_progress) send_cleanup_vector(cfg); - cpumask_copy(desc->affinity, mask); - + cpumask_copy(data->affinity, mask); return 0; } -/* - * Migrates the IRQ destination in the process context. - */ -static int set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, - const struct cpumask *mask) -{ - return migrate_ioapic_irq_desc(desc, mask); -} -static int set_ir_ioapic_affinity_irq(unsigned int irq, - const struct cpumask *mask) -{ - struct irq_desc *desc = irq_to_desc(irq); - - return set_ir_ioapic_affinity_irq_desc(desc, mask); -} #else -static inline int set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, - const struct cpumask *mask) +static inline int +ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { return 0; } @@ -2504,10 +2344,8 @@ unlock: irq_exit(); } -static void __irq_complete_move(struct irq_desc **descp, unsigned vector) +static void __irq_complete_move(struct irq_cfg *cfg, unsigned vector) { - struct irq_desc *desc = *descp; - struct irq_cfg *cfg = desc->chip_data; unsigned me; if (likely(!cfg->move_in_progress)) @@ -2519,31 +2357,28 @@ static void __irq_complete_move(struct irq_desc **descp, unsigned vector) send_cleanup_vector(cfg); } -static void irq_complete_move(struct irq_desc **descp) +static void irq_complete_move(struct irq_cfg *cfg) { - __irq_complete_move(descp, ~get_irq_regs()->orig_ax); + __irq_complete_move(cfg, ~get_irq_regs()->orig_ax); } void irq_force_complete_move(int irq) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = desc->chip_data; + struct irq_cfg *cfg = get_irq_chip_data(irq); if (!cfg) return; - __irq_complete_move(&desc, cfg->vector); + __irq_complete_move(cfg, cfg->vector); } #else -static inline void irq_complete_move(struct irq_desc **descp) {} +static inline void irq_complete_move(struct irq_cfg *cfg) { } #endif -static void ack_apic_edge(unsigned int irq) +static void ack_apic_edge(struct irq_data *data) { - struct irq_desc *desc = irq_to_desc(irq); - - irq_complete_move(&desc); - move_native_irq(irq); + irq_complete_move(data->chip_data); + move_native_irq(data->irq); ack_APIC_irq(); } @@ -2565,10 +2400,12 @@ atomic_t irq_mis_count; * Otherwise, we simulate the EOI message manually by changing the trigger * mode to edge and then back to level, with RTE being masked during this. */ -static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) +static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) { struct irq_pin_list *entry; + unsigned long flags; + raw_spin_lock_irqsave(&ioapic_lock, flags); for_each_irq_pin(entry, cfg->irq_2_pin) { if (mp_ioapics[entry->apic].apicver >= 0x20) { /* @@ -2577,7 +2414,7 @@ static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) * intr-remapping table entry. Hence for the io-apic * EOI we use the pin number. */ - if (irq_remapped(irq)) + if (irq_remapped(cfg)) io_apic_eoi(entry->apic, entry->pin); else io_apic_eoi(entry->apic, cfg->vector); @@ -2586,36 +2423,22 @@ static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) __unmask_and_level_IO_APIC_irq(entry); } } -} - -static void eoi_ioapic_irq(struct irq_desc *desc) -{ - struct irq_cfg *cfg; - unsigned long flags; - unsigned int irq; - - irq = desc->irq; - cfg = desc->chip_data; - - raw_spin_lock_irqsave(&ioapic_lock, flags); - __eoi_ioapic_irq(irq, cfg); raw_spin_unlock_irqrestore(&ioapic_lock, flags); } -static void ack_apic_level(unsigned int irq) +static void ack_apic_level(struct irq_data *data) { + struct irq_cfg *cfg = data->chip_data; + int i, do_unmask_irq = 0, irq = data->irq; struct irq_desc *desc = irq_to_desc(irq); unsigned long v; - int i; - struct irq_cfg *cfg; - int do_unmask_irq = 0; - irq_complete_move(&desc); + irq_complete_move(cfg); #ifdef CONFIG_GENERIC_PENDING_IRQ /* If we are moving the irq we need to mask it */ if (unlikely(desc->status & IRQ_MOVE_PENDING)) { do_unmask_irq = 1; - mask_IO_APIC_irq_desc(desc); + mask_ioapic(cfg); } #endif @@ -2651,7 +2474,6 @@ static void ack_apic_level(unsigned int irq) * we use the above logic (mask+edge followed by unmask+level) from * Manfred Spraul to clear the remote IRR. */ - cfg = desc->chip_data; i = cfg->vector; v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1)); @@ -2671,7 +2493,7 @@ static void ack_apic_level(unsigned int irq) if (!(v & (1 << (i & 0x1f)))) { atomic_inc(&irq_mis_count); - eoi_ioapic_irq(desc); + eoi_ioapic_irq(irq, cfg); } /* Now we can move and renable the irq */ @@ -2702,61 +2524,57 @@ 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. */ - cfg = desc->chip_data; if (!io_apic_level_ack_pending(cfg)) move_masked_irq(irq); - unmask_IO_APIC_irq_desc(desc); + unmask_ioapic(cfg); } } #ifdef CONFIG_INTR_REMAP -static void ir_ack_apic_edge(unsigned int irq) +static void ir_ack_apic_edge(struct irq_data *data) { ack_APIC_irq(); } -static void ir_ack_apic_level(unsigned int irq) +static void ir_ack_apic_level(struct irq_data *data) { - struct irq_desc *desc = irq_to_desc(irq); - ack_APIC_irq(); - eoi_ioapic_irq(desc); + eoi_ioapic_irq(data->irq, data->chip_data); } #endif /* CONFIG_INTR_REMAP */ 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", + .irq_startup = startup_ioapic_irq, + .irq_mask = mask_ioapic_irq, + .irq_unmask = unmask_ioapic_irq, + .irq_ack = ack_apic_edge, + .irq_eoi = ack_apic_level, #ifdef CONFIG_SMP - .set_affinity = set_ioapic_affinity_irq, + .irq_set_affinity = ioapic_set_affinity, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; 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, + .name = "IR-IO-APIC", + .irq_startup = startup_ioapic_irq, + .irq_mask = mask_ioapic_irq, + .irq_unmask = unmask_ioapic_irq, #ifdef CONFIG_INTR_REMAP - .ack = ir_ack_apic_edge, - .eoi = ir_ack_apic_level, + .irq_ack = ir_ack_apic_edge, + .irq_eoi = ir_ack_apic_level, #ifdef CONFIG_SMP - .set_affinity = set_ir_ioapic_affinity_irq, + .irq_set_affinity = ir_ioapic_set_affinity, #endif #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static inline void init_IO_APIC_traps(void) { - int irq; - struct irq_desc *desc; struct irq_cfg *cfg; + unsigned int irq; /* * NOTE! The local APIC isn't very good at handling @@ -2769,8 +2587,8 @@ 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_each_irq_desc(irq, desc) { - cfg = desc->chip_data; + for_each_active_irq(irq) { + cfg = get_irq_chip_data(irq); if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) { /* * Hmm.. We don't have an entry for this, @@ -2781,7 +2599,7 @@ static inline void init_IO_APIC_traps(void) legacy_pic->make_irq(irq); else /* Strange. Oh, well.. */ - desc->chip = &no_irq_chip; + set_irq_chip(irq, &no_irq_chip); } } } @@ -2790,7 +2608,7 @@ static inline void init_IO_APIC_traps(void) * The local APIC irq-chip implementation: */ -static void mask_lapic_irq(unsigned int irq) +static void mask_lapic_irq(struct irq_data *data) { unsigned long v; @@ -2798,7 +2616,7 @@ static void mask_lapic_irq(unsigned int irq) apic_write(APIC_LVT0, v | APIC_LVT_MASKED); } -static void unmask_lapic_irq(unsigned int irq) +static void unmask_lapic_irq(struct irq_data *data) { unsigned long v; @@ -2806,21 +2624,21 @@ static void unmask_lapic_irq(unsigned int irq) apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); } -static void ack_lapic_irq(unsigned int irq) +static void ack_lapic_irq(struct irq_data *data) { ack_APIC_irq(); } static struct irq_chip lapic_chip __read_mostly = { .name = "local-APIC", - .mask = mask_lapic_irq, - .unmask = unmask_lapic_irq, - .ack = ack_lapic_irq, + .irq_mask = mask_lapic_irq, + .irq_unmask = unmask_lapic_irq, + .irq_ack = ack_lapic_irq, }; -static void lapic_register_intr(int irq, struct irq_desc *desc) +static void lapic_register_intr(int irq) { - desc->status &= ~IRQ_LEVEL; + irq_clear_status_flags(irq, IRQ_LEVEL); set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq, "edge"); } @@ -2923,9 +2741,8 @@ int timer_through_8259 __initdata; */ static inline void __init check_timer(void) { - struct irq_desc *desc = irq_to_desc(0); - struct irq_cfg *cfg = desc->chip_data; - int node = cpu_to_node(boot_cpu_id); + struct irq_cfg *cfg = get_irq_chip_data(0); + int node = cpu_to_node(0); int apic1, pin1, apic2, pin2; unsigned long flags; int no_pin1 = 0; @@ -2935,7 +2752,7 @@ static inline void __init check_timer(void) /* * get/set the timer IRQ vector: */ - legacy_pic->chip->mask(0); + legacy_pic->mask(0); assign_irq_vector(0, cfg, apic->target_cpus()); /* @@ -2994,7 +2811,7 @@ static inline void __init check_timer(void) add_pin_to_irq_node(cfg, node, apic1, pin1); setup_timer_IRQ0_pin(apic1, pin1, cfg->vector); } else { - /* for edge trigger, setup_IO_APIC_irq already + /* for edge trigger, setup_ioapic_irq already * leave it unmasked. * so only need to unmask if it is level-trigger * do we really have level trigger timer? @@ -3002,12 +2819,12 @@ static inline void __init check_timer(void) int idx; idx = find_irq_entry(apic1, pin1, mp_INT); if (idx != -1 && irq_trigger(idx)) - unmask_IO_APIC_irq_desc(desc); + unmask_ioapic(cfg); } if (timer_irq_works()) { if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); } if (disable_timer_pin_1 > 0) clear_IO_APIC_pin(0, pin1); @@ -3030,14 +2847,14 @@ static inline void __init check_timer(void) */ replace_pin_at_irq_node(cfg, node, apic1, pin1, apic2, pin2); setup_timer_IRQ0_pin(apic2, pin2, cfg->vector); - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); if (timer_irq_works()) { apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); timer_through_8259 = 1; if (nmi_watchdog == NMI_IO_APIC) { - legacy_pic->chip->mask(0); + legacy_pic->mask(0); setup_nmi(); - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); } goto out; } @@ -3045,7 +2862,7 @@ static inline void __init check_timer(void) * Cleanup, just in case ... */ local_irq_disable(); - legacy_pic->chip->mask(0); + legacy_pic->mask(0); clear_IO_APIC_pin(apic2, pin2); apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); } @@ -3062,16 +2879,16 @@ static inline void __init check_timer(void) apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer as Virtual Wire IRQ...\n"); - lapic_register_intr(0, desc); + lapic_register_intr(0); apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector); /* Fixed mode */ - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); if (timer_irq_works()) { apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); goto out; } local_irq_disable(); - legacy_pic->chip->mask(0); + legacy_pic->mask(0); apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | cfg->vector); apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n"); @@ -3237,49 +3054,42 @@ device_initcall(ioapic_init_sysfs); /* * Dynamic irq allocate and deallocation */ -unsigned int create_irq_nr(unsigned int irq_want, int node) +unsigned int create_irq_nr(unsigned int from, int node) { - /* Allocate an unused irq */ - unsigned int irq; - unsigned int new; + struct irq_cfg *cfg; unsigned long flags; - struct irq_cfg *cfg_new = NULL; - struct irq_desc *desc_new = NULL; - - irq = 0; - if (irq_want < nr_irqs_gsi) - irq_want = nr_irqs_gsi; - - raw_spin_lock_irqsave(&vector_lock, flags); - for (new = irq_want; new < nr_irqs; new++) { - desc_new = irq_to_desc_alloc_node(new, node); - 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; + unsigned int ret = 0; + int irq; - desc_new = move_irq_desc(desc_new, node); - cfg_new = desc_new->chip_data; + if (from < nr_irqs_gsi) + from = nr_irqs_gsi; - if (__assign_irq_vector(new, cfg_new, apic->target_cpus()) == 0) - irq = new; - break; + irq = alloc_irq_from(from, node); + if (irq < 0) + return 0; + cfg = alloc_irq_cfg(irq, node); + if (!cfg) { + free_irq_at(irq, NULL); + return 0; } - raw_spin_unlock_irqrestore(&vector_lock, flags); - if (irq > 0) - dynamic_irq_init_keep_chip_data(irq); + raw_spin_lock_irqsave(&vector_lock, flags); + if (!__assign_irq_vector(irq, cfg, apic->target_cpus())) + ret = irq; + raw_spin_unlock_irqrestore(&vector_lock, flags); - return irq; + if (ret) { + set_irq_chip_data(irq, cfg); + irq_clear_status_flags(irq, IRQ_NOREQUEST); + } else { + free_irq_at(irq, cfg); + } + return ret; } int create_irq(void) { - int node = cpu_to_node(boot_cpu_id); + int node = cpu_to_node(0); unsigned int irq_want; int irq; @@ -3294,14 +3104,17 @@ int create_irq(void) void destroy_irq(unsigned int irq) { + struct irq_cfg *cfg = get_irq_chip_data(irq); unsigned long flags; - dynamic_irq_cleanup_keep_chip_data(irq); + irq_set_status_flags(irq, IRQ_NOREQUEST|IRQ_NOPROBE); - free_irte(irq); + if (intr_remapping_enabled) + free_irte(irq); raw_spin_lock_irqsave(&vector_lock, flags); - __clear_irq_vector(irq, get_irq_chip_data(irq)); + __clear_irq_vector(irq, cfg); raw_spin_unlock_irqrestore(&vector_lock, flags); + free_irq_at(irq, cfg); } /* @@ -3325,7 +3138,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus()); - if (irq_remapped(irq)) { + if (irq_remapped(get_irq_chip_data(irq))) { struct irte irte; int ir_index; u16 sub_handle; @@ -3333,14 +3146,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, ir_index = map_irq_to_irte_handle(irq, &sub_handle); BUG_ON(ir_index == -1); - memset (&irte, 0, sizeof(irte)); - - irte.present = 1; - irte.dst_mode = apic->irq_dest_mode; - irte.trigger_mode = 0; /* edge */ - irte.dlvry_mode = apic->irq_delivery_mode; - irte.vector = cfg->vector; - irte.dest_id = IRTE_DEST(dest); + prepare_irte(&irte, cfg->vector, dest); /* Set source-id of interrupt request */ if (pdev) @@ -3385,26 +3191,24 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, } #ifdef CONFIG_SMP -static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) +static int +msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; + struct irq_cfg *cfg = data->chip_data; struct msi_msg msg; unsigned int dest; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; - cfg = desc->chip_data; - - get_cached_msi_msg_desc(desc, &msg); + __get_cached_msi_msg(data->msi_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_desc(desc, &msg); + __write_msi_msg(data->msi_desc, &msg); return 0; } @@ -3414,17 +3218,17 @@ static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) * done in the process context using interrupt-remapping hardware. */ static int -ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) +ir_msi_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = desc->chip_data; - unsigned int dest; + struct irq_cfg *cfg = data->chip_data; + unsigned int dest, irq = data->irq; struct irte irte; if (get_irte(irq, &irte)) return -1; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; irte.vector = cfg->vector; @@ -3454,27 +3258,27 @@ ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) * 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_apic_edge, + .name = "PCI-MSI", + .irq_unmask = unmask_msi_irq, + .irq_mask = mask_msi_irq, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = set_msi_irq_affinity, + .irq_set_affinity = msi_set_affinity, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static struct irq_chip msi_ir_chip = { - .name = "IR-PCI-MSI", - .unmask = unmask_msi_irq, - .mask = mask_msi_irq, + .name = "IR-PCI-MSI", + .irq_unmask = unmask_msi_irq, + .irq_mask = mask_msi_irq, #ifdef CONFIG_INTR_REMAP - .ack = ir_ack_apic_edge, + .irq_ack = ir_ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = ir_set_msi_irq_affinity, + .irq_set_affinity = ir_msi_set_affinity, #endif #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; /* @@ -3506,8 +3310,8 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec) static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) { - int ret; struct msi_msg msg; + int ret; ret = msi_compose_msg(dev, irq, &msg, -1); if (ret < 0) @@ -3516,12 +3320,8 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) set_irq_msi(irq, msidesc); write_msi_msg(irq, &msg); - if (irq_remapped(irq)) { - struct irq_desc *desc = irq_to_desc(irq); - /* - * irq migration in process context - */ - desc->status |= IRQ_MOVE_PCNTXT; + if (irq_remapped(get_irq_chip_data(irq))) { + irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge"); } else set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge"); @@ -3533,13 +3333,10 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) { - unsigned int irq; - int ret, sub_handle; + int node, ret, sub_handle, index = 0; + unsigned int irq, irq_want; struct msi_desc *msidesc; - unsigned int irq_want; struct intel_iommu *iommu = NULL; - int index = 0; - int node; /* x86 doesn't support multiple MSI yet */ if (type == PCI_CAP_ID_MSI && nvec > 1) @@ -3599,18 +3396,17 @@ void arch_teardown_msi_irq(unsigned int irq) #if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP) #ifdef CONFIG_SMP -static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) +static int +dmar_msi_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; + struct irq_cfg *cfg = data->chip_data; + unsigned int dest, irq = data->irq; struct msi_msg msg; - unsigned int dest; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; - cfg = desc->chip_data; - dmar_msi_read(irq, &msg); msg.data &= ~MSI_DATA_VECTOR_MASK; @@ -3626,14 +3422,14 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) #endif /* CONFIG_SMP */ static struct irq_chip dmar_msi_type = { - .name = "DMAR_MSI", - .unmask = dmar_msi_unmask, - .mask = dmar_msi_mask, - .ack = ack_apic_edge, + .name = "DMAR_MSI", + .irq_unmask = dmar_msi_unmask, + .irq_mask = dmar_msi_mask, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = dmar_msi_set_affinity, + .irq_set_affinity = dmar_msi_set_affinity, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; int arch_setup_dmar_msi(unsigned int irq) @@ -3654,26 +3450,24 @@ int arch_setup_dmar_msi(unsigned int irq) #ifdef CONFIG_HPET_TIMER #ifdef CONFIG_SMP -static int hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask) +static int hpet_msi_set_affinity(struct irq_data *data, + const struct cpumask *mask, bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; + struct irq_cfg *cfg = data->chip_data; struct msi_msg msg; unsigned int dest; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; - cfg = desc->chip_data; - - hpet_msi_read(irq, &msg); + hpet_msi_read(data->handler_data, &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); + hpet_msi_write(data->handler_data, &msg); return 0; } @@ -3681,34 +3475,33 @@ static int hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask) #endif /* CONFIG_SMP */ static struct irq_chip ir_hpet_msi_type = { - .name = "IR-HPET_MSI", - .unmask = hpet_msi_unmask, - .mask = hpet_msi_mask, + .name = "IR-HPET_MSI", + .irq_unmask = hpet_msi_unmask, + .irq_mask = hpet_msi_mask, #ifdef CONFIG_INTR_REMAP - .ack = ir_ack_apic_edge, + .irq_ack = ir_ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = ir_set_msi_irq_affinity, + .irq_set_affinity = ir_msi_set_affinity, #endif #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static struct irq_chip hpet_msi_type = { .name = "HPET_MSI", - .unmask = hpet_msi_unmask, - .mask = hpet_msi_mask, - .ack = ack_apic_edge, + .irq_unmask = hpet_msi_unmask, + .irq_mask = hpet_msi_mask, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = hpet_msi_set_affinity, + .irq_set_affinity = hpet_msi_set_affinity, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; int arch_setup_hpet_msi(unsigned int irq, unsigned int id) { - int ret; struct msi_msg msg; - struct irq_desc *desc = irq_to_desc(irq); + int ret; if (intr_remapping_enabled) { struct intel_iommu *iommu = map_hpet_to_ir(id); @@ -3726,9 +3519,9 @@ int arch_setup_hpet_msi(unsigned int irq, unsigned int id) if (ret < 0) return ret; - hpet_msi_write(irq, &msg); - desc->status |= IRQ_MOVE_PCNTXT; - if (irq_remapped(irq)) + hpet_msi_write(get_irq_data(irq), &msg); + irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); + if (irq_remapped(get_irq_chip_data(irq))) set_irq_chip_and_handler_name(irq, &ir_hpet_msi_type, handle_edge_irq, "edge"); else @@ -3761,33 +3554,30 @@ static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) write_ht_irq_msg(irq, &msg); } -static int set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask) +static int +ht_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; + struct irq_cfg *cfg = data->chip_data; unsigned int dest; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; - cfg = desc->chip_data; - - target_ht_irq(irq, dest, cfg->vector); - + target_ht_irq(data->irq, dest, cfg->vector); return 0; } #endif static struct irq_chip ht_irq_chip = { - .name = "PCI-HT", - .mask = mask_ht_irq, - .unmask = unmask_ht_irq, - .ack = ack_apic_edge, + .name = "PCI-HT", + .irq_mask = mask_ht_irq, + .irq_unmask = unmask_ht_irq, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = set_ht_irq_affinity, + .irq_set_affinity = ht_set_affinity, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) @@ -3878,14 +3668,13 @@ int __init arch_probe_nr_irqs(void) if (nr < nr_irqs) nr_irqs = nr; - return 0; + return NR_IRQS_LEGACY; } #endif static int __io_apic_set_pci_routing(struct device *dev, int irq, struct io_apic_irq_attr *irq_attr) { - struct irq_desc *desc; struct irq_cfg *cfg; int node; int ioapic, pin; @@ -3901,13 +3690,11 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, if (dev) node = dev_to_node(dev); else - node = cpu_to_node(boot_cpu_id); + node = cpu_to_node(0); - desc = irq_to_desc_alloc_node(irq, node); - if (!desc) { - printk(KERN_INFO "can not get irq_desc %d\n", irq); + cfg = alloc_irq_and_cfg_at(irq, node); + if (!cfg) return 0; - } pin = irq_attr->ioapic_pin; trigger = irq_attr->trigger; @@ -3917,15 +3704,14 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, * IRQs < 16 are already in the irq_2_pin[] map */ if (irq >= legacy_pic->nr_legacy_irqs) { - cfg = desc->chip_data; - if (add_pin_to_irq_node_nopanic(cfg, node, ioapic, pin)) { + if (__add_pin_to_irq_node(cfg, node, ioapic, pin)) { printk(KERN_INFO "can not add pin %d for irq %d\n", pin, irq); return 0; } } - setup_IO_APIC_irq(ioapic, pin, irq, desc, trigger, polarity); + setup_ioapic_irq(ioapic, pin, irq, cfg, trigger, polarity); return 0; } @@ -4118,14 +3904,14 @@ void __init setup_ioapic_dest(void) */ if (desc->status & (IRQ_NO_BALANCING | IRQ_AFFINITY_SET)) - mask = desc->affinity; + mask = desc->irq_data.affinity; else mask = apic->target_cpus(); if (intr_remapping_enabled) - set_ir_ioapic_affinity_irq_desc(desc, mask); + ir_ioapic_set_affinity(&desc->irq_data, mask, false); else - set_ioapic_affinity_irq_desc(desc, mask); + ioapic_set_affinity(&desc->irq_data, mask, false); } } @@ -4309,19 +4095,18 @@ void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) void __init pre_init_apic_IRQ0(void) { struct irq_cfg *cfg; - struct irq_desc *desc; printk(KERN_INFO "Early APIC setup for system timer0\n"); #ifndef CONFIG_SMP phys_cpu_present_map = physid_mask_of_physid(boot_cpu_physical_apicid); #endif - desc = irq_to_desc_alloc_node(0, 0); + /* Make sure the irq descriptor is set up */ + cfg = alloc_irq_and_cfg_at(0, 0); setup_local_APIC(); - cfg = irq_cfg(0); add_pin_to_irq_node(cfg, 0, 0, 0); set_irq_chip_and_handler_name(0, &ioapic_chip, handle_edge_irq, "edge"); - setup_IO_APIC_irq(0, 0, 0, desc, 0, 0); + setup_ioapic_irq(0, 0, 0, cfg, 0, 0); } diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c index a43f71cb30f..c90041ccb74 100644 --- a/arch/x86/kernel/apic/nmi.c +++ b/arch/x86/kernel/apic/nmi.c @@ -178,7 +178,7 @@ int __init check_nmi_watchdog(void) error: if (nmi_watchdog == NMI_IO_APIC) { if (!timer_through_8259) - legacy_pic->chip->mask(0); + legacy_pic->mask(0); on_each_cpu(__acpi_nmi_disable, NULL, 1); } diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c index 3e28401f161..960f26ab5c9 100644 --- a/arch/x86/kernel/apic/numaq_32.c +++ b/arch/x86/kernel/apic/numaq_32.c @@ -26,6 +26,7 @@ #include <linux/nodemask.h> #include <linux/topology.h> #include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/threads.h> #include <linux/cpumask.h> #include <linux/kernel.h> @@ -88,7 +89,7 @@ static inline void numaq_register_node(int node, struct sys_cfg_data *scd) node_end_pfn[node] = MB_TO_PAGES(eq->hi_shrd_mem_start + eq->hi_shrd_mem_size); - e820_register_active_regions(node, node_start_pfn[node], + memblock_x86_register_active_regions(node, node_start_pfn[node], node_end_pfn[node]); memory_present(node, node_start_pfn[node], node_end_pfn[node]); diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c index 83e9be4778e..f9e4e6a5407 100644 --- a/arch/x86/kernel/apic/probe_64.c +++ b/arch/x86/kernel/apic/probe_64.c @@ -54,6 +54,9 @@ static int apicid_phys_pkg_id(int initial_apic_id, int index_msb) */ void __init default_setup_apic_routing(void) { + + enable_IR_x2apic(); + #ifdef CONFIG_X86_X2APIC if (x2apic_mode #ifdef CONFIG_X86_UV diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index e46f98f36e3..f744f54cb24 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -604,6 +604,10 @@ int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data) { if (reason != DIE_NMI_IPI) return NOTIFY_OK; + + if (in_crash_kexec) + /* do nothing if entering the crash kernel */ + return NOTIFY_OK; /* * Use a lock so only one cpu prints at a time * to prevent intermixed output. @@ -694,9 +698,11 @@ void __init uv_system_init(void) for (j = 0; j < 64; j++) { if (!test_bit(j, &present)) continue; - uv_blade_info[blade].pnode = (i * 64 + j); + pnode = (i * 64 + j); + uv_blade_info[blade].pnode = pnode; uv_blade_info[blade].nr_possible_cpus = 0; uv_blade_info[blade].nr_online_cpus = 0; + max_pnode = max(pnode, max_pnode); blade++; } } @@ -734,7 +740,6 @@ void __init uv_system_init(void) uv_cpu_hub_info(cpu)->scir.offset = uv_scir_offset(apicid); uv_node_to_blade[nid] = blade; uv_cpu_to_blade[cpu] = blade; - max_pnode = max(pnode, max_pnode); } /* Add blade/pnode info for nodes without cpus */ @@ -746,7 +751,6 @@ void __init uv_system_init(void) pnode = (paddr >> m_val) & pnode_mask; blade = boot_pnode_to_blade(pnode); uv_node_to_blade[nid] = blade; - max_pnode = max(pnode, max_pnode); } map_gru_high(max_pnode); diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 4c9c67bf09b..0e4f24c2a74 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -189,8 +189,8 @@ * Intel Order Number 241704-001. Microsoft Part Number 781-110-X01. * * [This document is available free from Intel by calling 800.628.8686 (fax - * 916.356.6100) or 800.548.4725; or via anonymous ftp from - * ftp://ftp.intel.com/pub/IAL/software_specs/apmv11.doc. It is also + * 916.356.6100) or 800.548.4725; or from + * http://www.microsoft.com/whdc/archive/amp_12.mspx It is also * available from Microsoft by calling 206.882.8080.] * * APM 1.2 Reference: @@ -1926,6 +1926,7 @@ static const struct file_operations apm_bios_fops = { .unlocked_ioctl = do_ioctl, .open = do_open, .release = do_release, + .llseek = noop_llseek, }; static struct miscdevice apm_device = { diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c index dfdbf640389..1a4088dda37 100644 --- a/arch/x86/kernel/asm-offsets_32.c +++ b/arch/x86/kernel/asm-offsets_32.c @@ -99,9 +99,7 @@ void foo(void) DEFINE(PAGE_SIZE_asm, PAGE_SIZE); DEFINE(PAGE_SHIFT_asm, PAGE_SHIFT); - DEFINE(PTRS_PER_PTE, PTRS_PER_PTE); - DEFINE(PTRS_PER_PMD, PTRS_PER_PMD); - DEFINE(PTRS_PER_PGD, PTRS_PER_PGD); + DEFINE(THREAD_SIZE_asm, THREAD_SIZE); OFFSET(crypto_tfm_ctx_offset, crypto_tfm, __crt_ctx); diff --git a/arch/x86/kernel/check.c b/arch/x86/kernel/check.c index fc999e6fc46..13a38917951 100644 --- a/arch/x86/kernel/check.c +++ b/arch/x86/kernel/check.c @@ -2,7 +2,8 @@ #include <linux/sched.h> #include <linux/kthread.h> #include <linux/workqueue.h> -#include <asm/e820.h> +#include <linux/memblock.h> + #include <asm/proto.h> /* @@ -18,10 +19,12 @@ 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 struct scan_area { + u64 addr; + u64 size; +} scan_areas[MAX_SCAN_AREAS]; static int num_scan_areas; - static __init int set_corruption_check(char *arg) { char *end; @@ -81,9 +84,9 @@ void __init setup_bios_corruption_check(void) while (addr < corruption_check_size && num_scan_areas < MAX_SCAN_AREAS) { u64 size; - addr = find_e820_area_size(addr, &size, PAGE_SIZE); + addr = memblock_x86_find_in_range_size(addr, &size, PAGE_SIZE); - if (!(addr + 1)) + if (addr == MEMBLOCK_ERROR) break; if (addr >= corruption_check_size) @@ -92,7 +95,7 @@ void __init setup_bios_corruption_check(void) if ((addr + size) > corruption_check_size) size = corruption_check_size - addr; - e820_update_range(addr, size, E820_RAM, E820_RESERVED); + memblock_x86_reserve_range(addr, addr + size, "SCAN RAM"); scan_areas[num_scan_areas].addr = addr; scan_areas[num_scan_areas].size = size; num_scan_areas++; @@ -105,7 +108,6 @@ void __init setup_bios_corruption_check(void) printk(KERN_INFO "Scanning %d areas for low memory corruption\n", num_scan_areas); - update_e820(); } diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 60a57b13082..9e093f8fe78 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -148,7 +148,7 @@ static void __cpuinit amd_k7_smp_check(struct cpuinfo_x86 *c) { #ifdef CONFIG_SMP /* calling is from identify_secondary_cpu() ? */ - if (c->cpu_index == boot_cpu_id) + if (!c->cpu_index) return; /* @@ -253,37 +253,51 @@ static int __cpuinit nearby_node(int apicid) #endif /* - * Fixup core topology information for AMD multi-node processors. - * Assumption: Number of cores in each internal node is the same. + * Fixup core topology information for + * (1) AMD multi-node processors + * Assumption: Number of cores in each internal node is the same. + * (2) AMD processors supporting compute units */ #ifdef CONFIG_X86_HT -static void __cpuinit amd_fixup_dcm(struct cpuinfo_x86 *c) +static void __cpuinit amd_get_topology(struct cpuinfo_x86 *c) { - unsigned long long value; - u32 nodes, cores_per_node; + u32 nodes; + u8 node_id; int cpu = smp_processor_id(); - if (!cpu_has(c, X86_FEATURE_NODEID_MSR)) - return; + /* get information required for multi-node processors */ + if (cpu_has(c, X86_FEATURE_TOPOEXT)) { + u32 eax, ebx, ecx, edx; - /* fixup topology information only once for a core */ - if (cpu_has(c, X86_FEATURE_AMD_DCM)) - return; + cpuid(0x8000001e, &eax, &ebx, &ecx, &edx); + nodes = ((ecx >> 8) & 7) + 1; + node_id = ecx & 7; - rdmsrl(MSR_FAM10H_NODE_ID, value); + /* get compute unit information */ + smp_num_siblings = ((ebx >> 8) & 3) + 1; + c->compute_unit_id = ebx & 0xff; + } else if (cpu_has(c, X86_FEATURE_NODEID_MSR)) { + u64 value; - nodes = ((value >> 3) & 7) + 1; - if (nodes == 1) + rdmsrl(MSR_FAM10H_NODE_ID, value); + nodes = ((value >> 3) & 7) + 1; + node_id = value & 7; + } else return; - set_cpu_cap(c, X86_FEATURE_AMD_DCM); - cores_per_node = c->x86_max_cores / nodes; + /* fixup multi-node processor information */ + if (nodes > 1) { + u32 cores_per_node; + + set_cpu_cap(c, X86_FEATURE_AMD_DCM); + cores_per_node = c->x86_max_cores / nodes; - /* store NodeID, use llc_shared_map to store sibling info */ - per_cpu(cpu_llc_id, cpu) = value & 7; + /* store NodeID, use llc_shared_map to store sibling info */ + per_cpu(cpu_llc_id, cpu) = node_id; - /* fixup core id to be in range from 0 to (cores_per_node - 1) */ - c->cpu_core_id = c->cpu_core_id % cores_per_node; + /* core id to be in range from 0 to (cores_per_node - 1) */ + c->cpu_core_id = c->cpu_core_id % cores_per_node; + } } #endif @@ -304,9 +318,7 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c) c->phys_proc_id = c->initial_apicid >> bits; /* use socket ID also for last level cache */ per_cpu(cpu_llc_id, cpu) = c->phys_proc_id; - /* fixup topology information on multi-node processors */ - if ((c->x86 == 0x10) && (c->x86_model == 9)) - amd_fixup_dcm(c); + amd_get_topology(c); #endif } @@ -412,6 +424,23 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_EXTD_APICID); } #endif + + /* We need to do the following only once */ + if (c != &boot_cpu_data) + return; + + if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) { + + if (c->x86 > 0x10 || + (c->x86 == 0x10 && c->x86_model >= 0x2)) { + u64 val; + + rdmsrl(MSR_K7_HWCR, val); + if (!(val & BIT(24))) + printk(KERN_WARNING FW_BUG "TSC doesn't count " + "with P0 frequency!\n"); + } + } } static void __cpuinit init_amd(struct cpuinfo_x86 *c) @@ -523,7 +552,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) #endif if (c->extended_cpuid_level >= 0x80000006) { - if ((c->x86 >= 0x0f) && (cpuid_edx(0x80000006) & 0xf000)) + if (cpuid_edx(0x80000006) & 0xf000) num_cache_leaves = 4; else num_cache_leaves = 3; @@ -669,7 +698,7 @@ bool cpu_has_amd_erratum(const int *erratum) } /* OSVW unavailable or ID unknown, match family-model-stepping range */ - ms = (cpu->x86_model << 8) | cpu->x86_mask; + ms = (cpu->x86_model << 4) | cpu->x86_mask; while ((range = *erratum++)) if ((cpu->x86 == AMD_MODEL_RANGE_FAMILY(range)) && (ms >= AMD_MODEL_RANGE_START(range)) && diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 490dac63c2d..4b68bda3093 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -545,7 +545,7 @@ void __cpuinit cpu_detect(struct cpuinfo_x86 *c) } } -static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) +void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) { u32 tfms, xlvl; u32 ebx; @@ -665,7 +665,7 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) this_cpu->c_early_init(c); #ifdef CONFIG_SMP - c->cpu_index = boot_cpu_id; + c->cpu_index = 0; #endif filter_cpuid_features(c, false); } @@ -704,16 +704,21 @@ void __init early_cpu_init(void) } /* - * The NOPL instruction is supposed to exist on all CPUs with - * family >= 6; unfortunately, that's not true in practice because - * of early VIA chips and (more importantly) broken virtualizers that - * are not easy to detect. In the latter case it doesn't even *fail* - * reliably, so probing for it doesn't even work. Disable it completely + * The NOPL instruction is supposed to exist on all CPUs of family >= 6; + * unfortunately, that's not true in practice because of early VIA + * chips and (more importantly) broken virtualizers that are not easy + * to detect. In the latter case it doesn't even *fail* reliably, so + * probing for it doesn't even work. Disable it completely on 32-bit * unless we can find a reliable way to detect all the broken cases. + * Enable it explicitly on 64-bit for non-constant inputs of cpu_has(). */ static void __cpuinit detect_nopl(struct cpuinfo_x86 *c) { +#ifdef CONFIG_X86_32 clear_cpu_cap(c, X86_FEATURE_NOPL); +#else + set_cpu_cap(c, X86_FEATURE_NOPL); +#endif } static void __cpuinit generic_identify(struct cpuinfo_x86 *c) @@ -1264,13 +1269,6 @@ void __cpuinit cpu_init(void) clear_all_debug_regs(); dbg_restore_debug_regs(); - /* - * Force FPU initialization: - */ - current_thread_info()->status = 0; - clear_used_math(); - mxcsr_feature_mask_init(); - fpu_init(); xsave_init(); } diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h index 3624e8a0f71..e765633f210 100644 --- a/arch/x86/kernel/cpu/cpu.h +++ b/arch/x86/kernel/cpu/cpu.h @@ -32,6 +32,8 @@ struct cpu_dev { extern const struct cpu_dev *const __x86_cpu_dev_start[], *const __x86_cpu_dev_end[]; +extern void get_cpu_cap(struct cpuinfo_x86 *c); extern void cpu_detect_cache_sizes(struct cpuinfo_x86 *c); +extern void get_cpu_cap(struct cpuinfo_x86 *c); #endif diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 246cd3afbb5..cd8da247dda 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -72,7 +72,7 @@ struct acpi_cpufreq_data { static DEFINE_PER_CPU(struct acpi_cpufreq_data *, acfreq_data); /* acpi_perf_data is a pointer to percpu data. */ -static struct acpi_processor_performance *acpi_perf_data; +static struct acpi_processor_performance __percpu *acpi_perf_data; static struct cpufreq_driver acpi_cpufreq_driver; diff --git a/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c index 994230d4dc4..4f6f679f279 100644 --- a/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c @@ -368,16 +368,22 @@ static int __init pcc_cpufreq_do_osc(acpi_handle *handle) return -ENODEV; out_obj = output.pointer; - if (out_obj->type != ACPI_TYPE_BUFFER) - return -ENODEV; + if (out_obj->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto out_free; + } errors = *((u32 *)out_obj->buffer.pointer) & ~(1 << 0); - if (errors) - return -ENODEV; + if (errors) { + ret = -ENODEV; + goto out_free; + } supported = *((u32 *)(out_obj->buffer.pointer + 4)); - if (!(supported & 0x1)) - return -ENODEV; + if (!(supported & 0x1)) { + ret = -ENODEV; + goto out_free; + } out_free: kfree(output.pointer); diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 85f69cdeae1..d16c2c53d6b 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -39,6 +39,7 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) misc_enable &= ~MSR_IA32_MISC_ENABLE_LIMIT_CPUID; wrmsrl(MSR_IA32_MISC_ENABLE, misc_enable); c->cpuid_level = cpuid_eax(0); + get_cpu_cap(c); } } @@ -169,7 +170,7 @@ static void __cpuinit intel_smp_check(struct cpuinfo_x86 *c) { #ifdef CONFIG_SMP /* calling is from identify_secondary_cpu() ? */ - if (c->cpu_index == boot_cpu_id) + if (!c->cpu_index) return; /* @@ -283,9 +284,7 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) /* Don't do the funky fallback heuristics the AMD version employs for now. */ node = apicid_to_node[apicid]; - if (node == NUMA_NO_NODE) - node = first_node(node_online_map); - else if (!node_online(node)) { + if (node == NUMA_NO_NODE || !node_online(node)) { /* reuse the value from init_cpu_to_node() */ node = cpu_to_node(cpu); } diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 898c2f4eab8..12cd823c8d0 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -17,7 +17,7 @@ #include <asm/processor.h> #include <linux/smp.h> -#include <asm/k8.h> +#include <asm/amd_nb.h> #include <asm/smp.h> #define LVL_1_INST 1 @@ -306,7 +306,7 @@ struct _cache_attr { ssize_t (*store)(struct _cpuid4_info *, const char *, size_t count); }; -#ifdef CONFIG_CPU_SUP_AMD +#ifdef CONFIG_AMD_NB /* * L3 cache descriptors @@ -369,7 +369,7 @@ static void __cpuinit amd_check_l3_disable(struct _cpuid4_info_regs *this_leaf, return; /* not in virtualized environments */ - if (num_k8_northbridges == 0) + if (k8_northbridges.num == 0) return; /* @@ -377,7 +377,7 @@ static void __cpuinit amd_check_l3_disable(struct _cpuid4_info_regs *this_leaf, * never freed but this is done only on shutdown so it doesn't matter. */ if (!l3_caches) { - int size = num_k8_northbridges * sizeof(struct amd_l3_cache *); + int size = k8_northbridges.num * sizeof(struct amd_l3_cache *); l3_caches = kzalloc(size, GFP_ATOMIC); if (!l3_caches) @@ -556,12 +556,12 @@ static struct _cache_attr cache_disable_0 = __ATTR(cache_disable_0, 0644, static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644, show_cache_disable_1, store_cache_disable_1); -#else /* CONFIG_CPU_SUP_AMD */ +#else /* CONFIG_AMD_NB */ static void __cpuinit amd_check_l3_disable(struct _cpuid4_info_regs *this_leaf, int index) { }; -#endif /* CONFIG_CPU_SUP_AMD */ +#endif /* CONFIG_AMD_NB */ static int __cpuinit cpuid4_cache_lookup_regs(int index, @@ -1000,7 +1000,7 @@ static struct attribute *default_attrs[] = { static struct attribute *default_l3_attrs[] = { DEFAULT_SYSFS_CACHE_ATTRS, -#ifdef CONFIG_CPU_SUP_AMD +#ifdef CONFIG_AMD_NB &cache_disable_0.attr, &cache_disable_1.attr, #endif diff --git a/arch/x86/kernel/cpu/mcheck/mce-apei.c b/arch/x86/kernel/cpu/mcheck/mce-apei.c index 745b54f9be8..8209472b27a 100644 --- a/arch/x86/kernel/cpu/mcheck/mce-apei.c +++ b/arch/x86/kernel/cpu/mcheck/mce-apei.c @@ -80,7 +80,7 @@ int apei_write_mce(struct mce *m) rcd.hdr.revision = CPER_RECORD_REV; rcd.hdr.signature_end = CPER_SIG_END; rcd.hdr.section_count = 1; - rcd.hdr.error_severity = CPER_SER_FATAL; + rcd.hdr.error_severity = CPER_SEV_FATAL; /* timestamp, platform_id, partition_id are all invalid */ rcd.hdr.validation_bits = 0; rcd.hdr.record_length = sizeof(rcd); @@ -96,7 +96,7 @@ int apei_write_mce(struct mce *m) rcd.sec_hdr.validation_bits = 0; rcd.sec_hdr.flags = CPER_SEC_PRIMARY; rcd.sec_hdr.section_type = CPER_SECTION_TYPE_MCE; - rcd.sec_hdr.section_severity = CPER_SER_FATAL; + rcd.sec_hdr.section_severity = CPER_SEV_FATAL; memcpy(&rcd.mce, m, sizeof(*m)); diff --git a/arch/x86/kernel/cpu/mcheck/mce-severity.c b/arch/x86/kernel/cpu/mcheck/mce-severity.c index 8a85dd1b1aa..1e8d66c1336 100644 --- a/arch/x86/kernel/cpu/mcheck/mce-severity.c +++ b/arch/x86/kernel/cpu/mcheck/mce-severity.c @@ -192,6 +192,7 @@ static const struct file_operations severities_coverage_fops = { .release = seq_release, .read = seq_read, .write = severities_coverage_write, + .llseek = seq_lseek, }; static int __init severities_debugfs_init(void) diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index ed41562909f..7a35b72d7c0 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1665,6 +1665,7 @@ struct file_operations mce_chrdev_ops = { .read = mce_read, .poll = mce_poll, .unlocked_ioctl = mce_ioctl, + .llseek = no_llseek, }; EXPORT_SYMBOL_GPL(mce_chrdev_ops); diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c index 224392d8fe8..80c482382d5 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c @@ -131,7 +131,8 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) u32 low = 0, high = 0, address = 0; unsigned int bank, block; struct thresh_restart tr; - u8 lvt_off; + int lvt_off = -1; + u8 offset; for (bank = 0; bank < NR_BANKS; ++bank) { for (block = 0; block < NR_BLOCKS; ++block) { @@ -141,6 +142,7 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) address = (low & MASK_BLKPTR_LO) >> 21; if (!address) break; + address += MCG_XBLK_ADDR; } else ++address; @@ -148,12 +150,8 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) if (rdmsr_safe(address, &low, &high)) break; - if (!(high & MASK_VALID_HI)) { - if (block) - continue; - else - break; - } + if (!(high & MASK_VALID_HI)) + continue; if (!(high & MASK_CNTP_HI) || (high & MASK_LOCKED_HI)) @@ -165,8 +163,28 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) if (shared_bank[bank] && c->cpu_core_id) break; #endif - lvt_off = setup_APIC_eilvt_mce(THRESHOLD_APIC_VECTOR, - APIC_EILVT_MSG_FIX, 0); + offset = (high & MASK_LVTOFF_HI) >> 20; + if (lvt_off < 0) { + if (setup_APIC_eilvt(offset, + THRESHOLD_APIC_VECTOR, + APIC_EILVT_MSG_FIX, 0)) { + pr_err(FW_BUG "cpu %d, failed to " + "setup threshold interrupt " + "for bank %d, block %d " + "(MSR%08X=0x%x%08x)", + smp_processor_id(), bank, block, + address, high, low); + continue; + } + lvt_off = offset; + } else if (lvt_off != offset) { + pr_err(FW_BUG "cpu %d, invalid threshold " + "interrupt offset %d for bank %d," + "block %d (MSR%08X=0x%x%08x)", + smp_processor_id(), lvt_off, bank, + block, address, high, low); + continue; + } high &= ~MASK_LVTOFF_HI; high |= lvt_off << 20; @@ -530,7 +548,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) err = -ENOMEM; goto out; } - if (!alloc_cpumask_var(&b->cpus, GFP_KERNEL)) { + if (!zalloc_cpumask_var(&b->cpus, GFP_KERNEL)) { kfree(b); err = -ENOMEM; goto out; @@ -543,7 +561,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) #ifndef CONFIG_SMP cpumask_setall(b->cpus); #else - cpumask_copy(b->cpus, c->llc_shared_map); + cpumask_set_cpu(cpu, b->cpus); #endif per_cpu(threshold_banks, cpu)[bank] = b; diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index c2a8b26d4fe..4b683267eca 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -202,10 +202,11 @@ static int therm_throt_process(bool new_event, int event, int level) #ifdef CONFIG_SYSFS /* Add/Remove thermal_throttle interface for CPU device: */ -static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev) +static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev, + unsigned int cpu) { int err; - struct cpuinfo_x86 *c = &cpu_data(smp_processor_id()); + struct cpuinfo_x86 *c = &cpu_data(cpu); err = sysfs_create_group(&sys_dev->kobj, &thermal_attr_group); if (err) @@ -215,7 +216,7 @@ static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev) err = sysfs_add_file_to_group(&sys_dev->kobj, &attr_core_power_limit_count.attr, thermal_attr_group.name); - if (cpu_has(c, X86_FEATURE_PTS)) + if (cpu_has(c, X86_FEATURE_PTS)) { err = sysfs_add_file_to_group(&sys_dev->kobj, &attr_package_throttle_count.attr, thermal_attr_group.name); @@ -223,6 +224,7 @@ static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev) err = sysfs_add_file_to_group(&sys_dev->kobj, &attr_package_power_limit_count.attr, thermal_attr_group.name); + } return err; } @@ -251,7 +253,7 @@ thermal_throttle_cpu_callback(struct notifier_block *nfb, case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: mutex_lock(&therm_cpu_lock); - err = thermal_throttle_add_dev(sys_dev); + err = thermal_throttle_add_dev(sys_dev, cpu); mutex_unlock(&therm_cpu_lock); WARN_ON(err); break; @@ -287,7 +289,7 @@ static __init int thermal_throttle_init_device(void) #endif /* connect live CPUs to sysfs */ for_each_online_cpu(cpu) { - err = thermal_throttle_add_dev(get_cpu_sysdev(cpu)); + err = thermal_throttle_add_dev(get_cpu_sysdev(cpu), cpu); WARN_ON(err); } #ifdef CONFIG_HOTPLUG_CPU @@ -348,7 +350,7 @@ static void intel_thermal_interrupt(void) static void unexpected_thermal_interrupt(void) { - printk(KERN_ERR "CPU%d: Unexpected LVT TMR interrupt!\n", + printk(KERN_ERR "CPU%d: Unexpected LVT thermal interrupt!\n", smp_processor_id()); add_taint(TAINT_MACHINE_CHECK); } diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c index c5f59d07142..ac140c7be39 100644 --- a/arch/x86/kernel/cpu/mtrr/cleanup.c +++ b/arch/x86/kernel/cpu/mtrr/cleanup.c @@ -827,7 +827,7 @@ int __init amd_special_default_mtrr(void) if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) return 0; - if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11) + if (boot_cpu_data.x86 < 0xf) return 0; /* In case some hypervisor doesn't pass SYSCFG through: */ if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0) diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 7d28d7d0388..9f27228ceff 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -64,18 +64,59 @@ static inline void k8_check_syscfg_dram_mod_en(void) } } +/* Get the size of contiguous MTRR range */ +static u64 get_mtrr_size(u64 mask) +{ + u64 size; + + mask >>= PAGE_SHIFT; + mask |= size_or_mask; + size = -mask; + size <<= PAGE_SHIFT; + return size; +} + /* - * Returns the effective MTRR type for the region - * Error returns: - * - 0xFE - when the range is "not entirely covered" by _any_ var range MTRR - * - 0xFF - when MTRR is not enabled + * Check and return the effective type for MTRR-MTRR type overlap. + * Returns 1 if the effective type is UNCACHEABLE, else returns 0 */ -u8 mtrr_type_lookup(u64 start, u64 end) +static int check_type_overlap(u8 *prev, u8 *curr) +{ + if (*prev == MTRR_TYPE_UNCACHABLE || *curr == MTRR_TYPE_UNCACHABLE) { + *prev = MTRR_TYPE_UNCACHABLE; + *curr = MTRR_TYPE_UNCACHABLE; + return 1; + } + + if ((*prev == MTRR_TYPE_WRBACK && *curr == MTRR_TYPE_WRTHROUGH) || + (*prev == MTRR_TYPE_WRTHROUGH && *curr == MTRR_TYPE_WRBACK)) { + *prev = MTRR_TYPE_WRTHROUGH; + *curr = MTRR_TYPE_WRTHROUGH; + } + + if (*prev != *curr) { + *prev = MTRR_TYPE_UNCACHABLE; + *curr = MTRR_TYPE_UNCACHABLE; + return 1; + } + + return 0; +} + +/* + * Error/Semi-error returns: + * 0xFF - when MTRR is not enabled + * *repeat == 1 implies [start:end] spanned across MTRR range and type returned + * corresponds only to [start:*partial_end]. + * Caller has to lookup again for [*partial_end:end]. + */ +static u8 __mtrr_type_lookup(u64 start, u64 end, u64 *partial_end, int *repeat) { int i; u64 base, mask; u8 prev_match, curr_match; + *repeat = 0; if (!mtrr_state_set) return 0xFF; @@ -126,8 +167,34 @@ u8 mtrr_type_lookup(u64 start, u64 end) start_state = ((start & mask) == (base & mask)); end_state = ((end & mask) == (base & mask)); - if (start_state != end_state) - return 0xFE; + + if (start_state != end_state) { + /* + * We have start:end spanning across an MTRR. + * We split the region into + * either + * (start:mtrr_end) (mtrr_end:end) + * or + * (start:mtrr_start) (mtrr_start:end) + * depending on kind of overlap. + * Return the type for first region and a pointer to + * the start of second region so that caller will + * lookup again on the second region. + * Note: This way we handle multiple overlaps as well. + */ + if (start_state) + *partial_end = base + get_mtrr_size(mask); + else + *partial_end = base; + + if (unlikely(*partial_end <= start)) { + WARN_ON(1); + *partial_end = start + PAGE_SIZE; + } + + end = *partial_end - 1; /* end is inclusive */ + *repeat = 1; + } if ((start & mask) != (base & mask)) continue; @@ -138,21 +205,8 @@ u8 mtrr_type_lookup(u64 start, u64 end) continue; } - if (prev_match == MTRR_TYPE_UNCACHABLE || - curr_match == MTRR_TYPE_UNCACHABLE) { - return MTRR_TYPE_UNCACHABLE; - } - - if ((prev_match == MTRR_TYPE_WRBACK && - curr_match == MTRR_TYPE_WRTHROUGH) || - (prev_match == MTRR_TYPE_WRTHROUGH && - curr_match == MTRR_TYPE_WRBACK)) { - prev_match = MTRR_TYPE_WRTHROUGH; - curr_match = MTRR_TYPE_WRTHROUGH; - } - - if (prev_match != curr_match) - return MTRR_TYPE_UNCACHABLE; + if (check_type_overlap(&prev_match, &curr_match)) + return curr_match; } if (mtrr_tom2) { @@ -166,6 +220,36 @@ u8 mtrr_type_lookup(u64 start, u64 end) return mtrr_state.def_type; } +/* + * Returns the effective MTRR type for the region + * Error return: + * 0xFF - when MTRR is not enabled + */ +u8 mtrr_type_lookup(u64 start, u64 end) +{ + u8 type, prev_type; + int repeat; + u64 partial_end; + + type = __mtrr_type_lookup(start, end, &partial_end, &repeat); + + /* + * Common path is with repeat = 0. + * However, we can have cases where [start:end] spans across some + * MTRR range. Do repeated lookups for that case here. + */ + while (repeat) { + prev_type = type; + start = partial_end; + type = __mtrr_type_lookup(start, end, &partial_end, &repeat); + + if (check_type_overlap(&prev_type, &type)) + return type; + } + + return type; +} + /* Get the MSR pair relating to a var range */ static void get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr) diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index f2da20fda02..fe73c1844a9 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -102,6 +102,7 @@ struct cpu_hw_events { */ struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */ unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + unsigned long running[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; int enabled; int n_events; @@ -530,7 +531,7 @@ static int x86_pmu_hw_config(struct perf_event *event) /* * Setup the hardware configuration for a given attr_type */ -static int __hw_perf_event_init(struct perf_event *event) +static int __x86_pmu_event_init(struct perf_event *event) { int err; @@ -583,7 +584,7 @@ static void x86_pmu_disable_all(void) } } -void hw_perf_disable(void) +static void x86_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -618,7 +619,7 @@ static void x86_pmu_enable_all(int added) } } -static const struct pmu pmu; +static struct pmu pmu; static inline int is_x86_event(struct perf_event *event) { @@ -800,10 +801,10 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc, hwc->last_tag == cpuc->tags[i]; } -static int x86_pmu_start(struct perf_event *event); -static void x86_pmu_stop(struct perf_event *event); +static void x86_pmu_start(struct perf_event *event, int flags); +static void x86_pmu_stop(struct perf_event *event, int flags); -void hw_perf_enable(void) +static void x86_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct perf_event *event; @@ -839,7 +840,14 @@ void hw_perf_enable(void) match_prev_assignment(hwc, cpuc, i)) continue; - x86_pmu_stop(event); + /* + * Ensure we don't accidentally enable a stopped + * counter simply because we rescheduled. + */ + if (hwc->state & PERF_HES_STOPPED) + hwc->state |= PERF_HES_ARCH; + + x86_pmu_stop(event, PERF_EF_UPDATE); } for (i = 0; i < cpuc->n_events; i++) { @@ -851,7 +859,10 @@ void hw_perf_enable(void) else if (i < n_running) continue; - x86_pmu_start(event); + if (hwc->state & PERF_HES_ARCH) + continue; + + x86_pmu_start(event, PERF_EF_RELOAD); } cpuc->n_added = 0; perf_events_lapic_init(); @@ -952,15 +963,12 @@ static void x86_pmu_enable_event(struct perf_event *event) } /* - * activate a single event + * Add a single event to the PMU. * * The event is added to the group of enabled events * but only if it can be scehduled with existing events. - * - * Called with PMU disabled. If successful and return value 1, - * then guaranteed to call perf_enable() and hw_perf_enable() */ -static int x86_pmu_enable(struct perf_event *event) +static int x86_pmu_add(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc; @@ -969,57 +977,67 @@ static int x86_pmu_enable(struct perf_event *event) hwc = &event->hw; + perf_pmu_disable(event->pmu); n0 = cpuc->n_events; - n = collect_events(cpuc, event, false); - if (n < 0) - return n; + ret = n = collect_events(cpuc, event, false); + if (ret < 0) + goto out; + + hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED; + if (!(flags & PERF_EF_START)) + hwc->state |= PERF_HES_ARCH; /* * If group events scheduling transaction was started, * skip the schedulability test here, it will be peformed - * at commit time(->commit_txn) as a whole + * at commit time (->commit_txn) as a whole */ if (cpuc->group_flag & PERF_EVENT_TXN) - goto out; + goto done_collect; ret = x86_pmu.schedule_events(cpuc, n, assign); if (ret) - return ret; + goto out; /* * copy new assignment, now we know it is possible * will be used by hw_perf_enable() */ memcpy(cpuc->assign, assign, n*sizeof(int)); -out: +done_collect: cpuc->n_events = n; cpuc->n_added += n - n0; cpuc->n_txn += n - n0; - return 0; + ret = 0; +out: + perf_pmu_enable(event->pmu); + return ret; } -static int x86_pmu_start(struct perf_event *event) +static void x86_pmu_start(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int idx = event->hw.idx; - if (idx == -1) - return -EAGAIN; + if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED))) + return; + + if (WARN_ON_ONCE(idx == -1)) + return; + + if (flags & PERF_EF_RELOAD) { + WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE)); + x86_perf_event_set_period(event); + } + + event->hw.state = 0; - x86_perf_event_set_period(event); cpuc->events[idx] = event; __set_bit(idx, cpuc->active_mask); + __set_bit(idx, cpuc->running); x86_pmu.enable(event); perf_event_update_userpage(event); - - return 0; -} - -static void x86_pmu_unthrottle(struct perf_event *event) -{ - int ret = x86_pmu_start(event); - WARN_ON_ONCE(ret); } void perf_event_print_debug(void) @@ -1076,27 +1094,29 @@ void perf_event_print_debug(void) local_irq_restore(flags); } -static void x86_pmu_stop(struct perf_event *event) +static void x86_pmu_stop(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; - int idx = hwc->idx; - - if (!__test_and_clear_bit(idx, cpuc->active_mask)) - return; - - x86_pmu.disable(event); - /* - * Drain the remaining delta count out of a event - * that we are disabling: - */ - x86_perf_event_update(event); + if (__test_and_clear_bit(hwc->idx, cpuc->active_mask)) { + x86_pmu.disable(event); + cpuc->events[hwc->idx] = NULL; + WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED); + hwc->state |= PERF_HES_STOPPED; + } - cpuc->events[idx] = NULL; + if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) { + /* + * Drain the remaining delta count out of a event + * that we are disabling: + */ + x86_perf_event_update(event); + hwc->state |= PERF_HES_UPTODATE; + } } -static void x86_pmu_disable(struct perf_event *event) +static void x86_pmu_del(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int i; @@ -1109,7 +1129,7 @@ static void x86_pmu_disable(struct perf_event *event) if (cpuc->group_flag & PERF_EVENT_TXN) return; - x86_pmu_stop(event); + x86_pmu_stop(event, PERF_EF_UPDATE); for (i = 0; i < cpuc->n_events; i++) { if (event == cpuc->event_list[i]) { @@ -1132,7 +1152,6 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) struct perf_sample_data data; struct cpu_hw_events *cpuc; struct perf_event *event; - struct hw_perf_event *hwc; int idx, handled = 0; u64 val; @@ -1141,11 +1160,18 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) cpuc = &__get_cpu_var(cpu_hw_events); for (idx = 0; idx < x86_pmu.num_counters; idx++) { - if (!test_bit(idx, cpuc->active_mask)) + if (!test_bit(idx, cpuc->active_mask)) { + /* + * Though we deactivated the counter some cpus + * might still deliver spurious interrupts still + * in flight. Catch them: + */ + if (__test_and_clear_bit(idx, cpuc->running)) + handled++; continue; + } event = cpuc->events[idx]; - hwc = &event->hw; val = x86_perf_event_update(event); if (val & (1ULL << (x86_pmu.cntval_bits - 1))) @@ -1154,14 +1180,14 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) /* * event overflow */ - handled = 1; + handled++; data.period = event->hw.last_period; if (!x86_perf_event_set_period(event)) continue; if (perf_event_overflow(event, 1, &data, regs)) - x86_pmu_stop(event); + x86_pmu_stop(event, 0); } if (handled) @@ -1170,25 +1196,6 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) return handled; } -void smp_perf_pending_interrupt(struct pt_regs *regs) -{ - irq_enter(); - ack_APIC_irq(); - inc_irq_stat(apic_pending_irqs); - perf_event_do_pending(); - irq_exit(); -} - -void set_perf_event_pending(void) -{ -#ifdef CONFIG_X86_LOCAL_APIC - if (!x86_pmu.apic || !x86_pmu_initialized()) - return; - - apic->send_IPI_self(LOCAL_PENDING_VECTOR); -#endif -} - void perf_events_lapic_init(void) { if (!x86_pmu.apic || !x86_pmu_initialized()) @@ -1200,12 +1207,20 @@ void perf_events_lapic_init(void) apic_write(APIC_LVTPC, APIC_DM_NMI); } +struct pmu_nmi_state { + unsigned int marked; + int handled; +}; + +static DEFINE_PER_CPU(struct pmu_nmi_state, pmu_nmi); + static int __kprobes perf_event_nmi_handler(struct notifier_block *self, unsigned long cmd, void *__args) { struct die_args *args = __args; - struct pt_regs *regs; + unsigned int this_nmi; + int handled; if (!atomic_read(&active_events)) return NOTIFY_DONE; @@ -1214,22 +1229,47 @@ perf_event_nmi_handler(struct notifier_block *self, case DIE_NMI: case DIE_NMI_IPI: break; - + case DIE_NMIUNKNOWN: + this_nmi = percpu_read(irq_stat.__nmi_count); + if (this_nmi != __get_cpu_var(pmu_nmi).marked) + /* let the kernel handle the unknown nmi */ + return NOTIFY_DONE; + /* + * This one is a PMU back-to-back nmi. Two events + * trigger 'simultaneously' raising two back-to-back + * NMIs. If the first NMI handles both, the latter + * will be empty and daze the CPU. So, we drop it to + * avoid false-positive 'unknown nmi' messages. + */ + return NOTIFY_STOP; default: return NOTIFY_DONE; } - regs = args->regs; - apic_write(APIC_LVTPC, APIC_DM_NMI); - /* - * Can't rely on the handled return value to say it was our NMI, two - * events could trigger 'simultaneously' raising two back-to-back NMIs. - * - * If the first NMI handles both, the latter will be empty and daze - * the CPU. - */ - x86_pmu.handle_irq(regs); + + handled = x86_pmu.handle_irq(args->regs); + if (!handled) + return NOTIFY_DONE; + + this_nmi = percpu_read(irq_stat.__nmi_count); + if ((handled > 1) || + /* the next nmi could be a back-to-back nmi */ + ((__get_cpu_var(pmu_nmi).marked == this_nmi) && + (__get_cpu_var(pmu_nmi).handled > 1))) { + /* + * We could have two subsequent back-to-back nmis: The + * first handles more than one counter, the 2nd + * handles only one counter and the 3rd handles no + * counter. + * + * This is the 2nd nmi because the previous was + * handling more than one counter. We will mark the + * next (3rd) and then drop it if unhandled. + */ + __get_cpu_var(pmu_nmi).marked = this_nmi + 1; + __get_cpu_var(pmu_nmi).handled = handled; + } return NOTIFY_STOP; } @@ -1345,7 +1385,6 @@ void __init init_hw_perf_events(void) x86_pmu.num_counters = X86_PMC_MAX_GENERIC; } x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1; - perf_max_events = x86_pmu.num_counters; if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) { WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!", @@ -1381,6 +1420,7 @@ void __init init_hw_perf_events(void) pr_info("... fixed-purpose events: %d\n", x86_pmu.num_counters_fixed); pr_info("... event mask: %016Lx\n", x86_pmu.intel_ctrl); + perf_pmu_register(&pmu); perf_cpu_notifier(x86_pmu_notifier); } @@ -1394,10 +1434,11 @@ static inline void x86_pmu_read(struct perf_event *event) * Set the flag to make pmu::enable() not perform the * schedulability test, it will be performed at commit time */ -static void x86_pmu_start_txn(const struct pmu *pmu) +static void x86_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + perf_pmu_disable(pmu); cpuc->group_flag |= PERF_EVENT_TXN; cpuc->n_txn = 0; } @@ -1407,7 +1448,7 @@ static void x86_pmu_start_txn(const struct pmu *pmu) * Clear the flag and pmu::enable() will perform the * schedulability test. */ -static void x86_pmu_cancel_txn(const struct pmu *pmu) +static void x86_pmu_cancel_txn(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -1417,6 +1458,7 @@ static void x86_pmu_cancel_txn(const struct pmu *pmu) */ cpuc->n_added -= cpuc->n_txn; cpuc->n_events -= cpuc->n_txn; + perf_pmu_enable(pmu); } /* @@ -1424,7 +1466,7 @@ static void x86_pmu_cancel_txn(const struct pmu *pmu) * Perform the group schedulability test as a whole * Return 0 if success */ -static int x86_pmu_commit_txn(const struct pmu *pmu) +static int x86_pmu_commit_txn(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int assign[X86_PMC_IDX_MAX]; @@ -1446,22 +1488,10 @@ static int x86_pmu_commit_txn(const struct pmu *pmu) memcpy(cpuc->assign, assign, n*sizeof(int)); cpuc->group_flag &= ~PERF_EVENT_TXN; - + perf_pmu_enable(pmu); return 0; } -static const struct pmu pmu = { - .enable = x86_pmu_enable, - .disable = x86_pmu_disable, - .start = x86_pmu_start, - .stop = x86_pmu_stop, - .read = x86_pmu_read, - .unthrottle = x86_pmu_unthrottle, - .start_txn = x86_pmu_start_txn, - .cancel_txn = x86_pmu_cancel_txn, - .commit_txn = x86_pmu_commit_txn, -}; - /* * validate that we can schedule this event */ @@ -1536,12 +1566,22 @@ out: return ret; } -const struct pmu *hw_perf_event_init(struct perf_event *event) +int x86_pmu_event_init(struct perf_event *event) { - const struct pmu *tmp; + struct pmu *tmp; int err; - err = __hw_perf_event_init(event); + switch (event->attr.type) { + case PERF_TYPE_RAW: + case PERF_TYPE_HARDWARE: + case PERF_TYPE_HW_CACHE: + break; + + default: + return -ENOENT; + } + + err = __x86_pmu_event_init(event); if (!err) { /* * we temporarily connect event to its pmu @@ -1561,26 +1601,31 @@ const struct pmu *hw_perf_event_init(struct perf_event *event) if (err) { if (event->destroy) event->destroy(event); - return ERR_PTR(err); } - return &pmu; + return err; } -/* - * callchain support - */ +static struct pmu pmu = { + .pmu_enable = x86_pmu_enable, + .pmu_disable = x86_pmu_disable, -static inline -void callchain_store(struct perf_callchain_entry *entry, u64 ip) -{ - if (entry->nr < PERF_MAX_STACK_DEPTH) - entry->ip[entry->nr++] = ip; -} + .event_init = x86_pmu_event_init, -static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry); -static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry); + .add = x86_pmu_add, + .del = x86_pmu_del, + .start = x86_pmu_start, + .stop = x86_pmu_stop, + .read = x86_pmu_read, + .start_txn = x86_pmu_start_txn, + .cancel_txn = x86_pmu_cancel_txn, + .commit_txn = x86_pmu_commit_txn, +}; + +/* + * callchain support + */ static void backtrace_warning_symbol(void *data, char *msg, unsigned long symbol) @@ -1602,7 +1647,7 @@ static void backtrace_address(void *data, unsigned long addr, int reliable) { struct perf_callchain_entry *entry = data; - callchain_store(entry, addr); + perf_callchain_store(entry, addr); } static const struct stacktrace_ops backtrace_ops = { @@ -1613,11 +1658,15 @@ static const struct stacktrace_ops backtrace_ops = { .walk_stack = print_context_stack_bp, }; -static void -perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) +void +perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { - callchain_store(entry, PERF_CONTEXT_KERNEL); - callchain_store(entry, regs->ip); + if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) { + /* TODO: We don't support guest os callchain now */ + return; + } + + perf_callchain_store(entry, regs->ip); dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry); } @@ -1646,7 +1695,7 @@ perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry *entry) if (fp < compat_ptr(regs->sp)) break; - callchain_store(entry, frame.return_address); + perf_callchain_store(entry, frame.return_address); fp = compat_ptr(frame.next_frame); } return 1; @@ -1659,19 +1708,20 @@ perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry *entry) } #endif -static void -perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry) +void +perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) { struct stack_frame frame; const void __user *fp; - if (!user_mode(regs)) - regs = task_pt_regs(current); + if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) { + /* TODO: We don't support guest os callchain now */ + return; + } fp = (void __user *)regs->bp; - callchain_store(entry, PERF_CONTEXT_USER); - callchain_store(entry, regs->ip); + perf_callchain_store(entry, regs->ip); if (perf_callchain_user32(regs, entry)) return; @@ -1688,52 +1738,11 @@ perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry) if ((unsigned long)fp < regs->sp) break; - callchain_store(entry, frame.return_address); + perf_callchain_store(entry, frame.return_address); fp = frame.next_frame; } } -static void -perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry) -{ - int is_user; - - if (!regs) - return; - - is_user = user_mode(regs); - - if (is_user && current->state != TASK_RUNNING) - return; - - if (!is_user) - perf_callchain_kernel(regs, entry); - - if (current->mm) - perf_callchain_user(regs, entry); -} - -struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) -{ - struct perf_callchain_entry *entry; - - if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) { - /* TODO: We don't support guest os callchain now */ - return NULL; - } - - if (in_nmi()) - entry = &__get_cpu_var(pmc_nmi_entry); - else - entry = &__get_cpu_var(pmc_irq_entry); - - entry->nr = 0; - - perf_do_callchain(regs, entry); - - return entry; -} - unsigned long perf_instruction_pointer(struct pt_regs *regs) { unsigned long ip; diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c index c2897b7b4a3..46d58448c3a 100644 --- a/arch/x86/kernel/cpu/perf_event_amd.c +++ b/arch/x86/kernel/cpu/perf_event_amd.c @@ -52,7 +52,7 @@ static __initconst const u64 amd_hw_cache_event_ids [ C(DTLB) ] = { [ C(OP_READ) ] = { [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ - [ C(RESULT_MISS) ] = 0x0046, /* L1 DTLB and L2 DLTB Miss */ + [ C(RESULT_MISS) ] = 0x0746, /* L1_DTLB_AND_L2_DLTB_MISS.ALL */ }, [ C(OP_WRITE) ] = { [ C(RESULT_ACCESS) ] = 0, @@ -66,7 +66,7 @@ static __initconst const u64 amd_hw_cache_event_ids [ C(ITLB) ] = { [ C(OP_READ) ] = { [ C(RESULT_ACCESS) ] = 0x0080, /* Instruction fecthes */ - [ C(RESULT_MISS) ] = 0x0085, /* Instr. fetch ITLB misses */ + [ C(RESULT_MISS) ] = 0x0385, /* L1_ITLB_AND_L2_ITLB_MISS.ALL */ }, [ C(OP_WRITE) ] = { [ C(RESULT_ACCESS) ] = -1, diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 214ac860ebe..c8f5c088cad 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -491,33 +491,78 @@ static void intel_pmu_enable_all(int added) * Intel Errata AAP53 (model 30) * Intel Errata BD53 (model 44) * - * These chips need to be 'reset' when adding counters by programming - * the magic three (non counting) events 0x4300D2, 0x4300B1 and 0x4300B5 - * either in sequence on the same PMC or on different PMCs. + * The official story: + * These chips need to be 'reset' when adding counters by programming the + * magic three (non-counting) events 0x4300B5, 0x4300D2, and 0x4300B1 either + * in sequence on the same PMC or on different PMCs. + * + * In practise it appears some of these events do in fact count, and + * we need to programm all 4 events. */ -static void intel_pmu_nhm_enable_all(int added) +static void intel_pmu_nhm_workaround(void) { - if (added) { - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - int i; + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + static const unsigned long nhm_magic[4] = { + 0x4300B5, + 0x4300D2, + 0x4300B1, + 0x4300B1 + }; + struct perf_event *event; + int i; - wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 0, 0x4300D2); - wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 1, 0x4300B1); - wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 2, 0x4300B5); + /* + * The Errata requires below steps: + * 1) Clear MSR_IA32_PEBS_ENABLE and MSR_CORE_PERF_GLOBAL_CTRL; + * 2) Configure 4 PERFEVTSELx with the magic events and clear + * the corresponding PMCx; + * 3) set bit0~bit3 of MSR_CORE_PERF_GLOBAL_CTRL; + * 4) Clear MSR_CORE_PERF_GLOBAL_CTRL; + * 5) Clear 4 pairs of ERFEVTSELx and PMCx; + */ - wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x3); - wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x0); + /* + * The real steps we choose are a little different from above. + * A) To reduce MSR operations, we don't run step 1) as they + * are already cleared before this function is called; + * B) Call x86_perf_event_update to save PMCx before configuring + * PERFEVTSELx with magic number; + * C) With step 5), we do clear only when the PERFEVTSELx is + * not used currently. + * D) Call x86_perf_event_set_period to restore PMCx; + */ + + /* We always operate 4 pairs of PERF Counters */ + for (i = 0; i < 4; i++) { + event = cpuc->events[i]; + if (event) + x86_perf_event_update(event); + } + + for (i = 0; i < 4; i++) { + wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + i, nhm_magic[i]); + wrmsrl(MSR_ARCH_PERFMON_PERFCTR0 + i, 0x0); + } - for (i = 0; i < 3; i++) { - struct perf_event *event = cpuc->events[i]; + wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0xf); + wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x0); - if (!event) - continue; + for (i = 0; i < 4; i++) { + event = cpuc->events[i]; + if (event) { + x86_perf_event_set_period(event); __x86_pmu_enable_event(&event->hw, - ARCH_PERFMON_EVENTSEL_ENABLE); - } + ARCH_PERFMON_EVENTSEL_ENABLE); + } else + wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + i, 0x0); } +} + +static void intel_pmu_nhm_enable_all(int added) +{ + if (added) + intel_pmu_nhm_workaround(); intel_pmu_enable_all(added); } @@ -667,22 +712,24 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) struct perf_sample_data data; struct cpu_hw_events *cpuc; int bit, loops; - u64 ack, status; + u64 status; + int handled; perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); intel_pmu_disable_all(); - intel_pmu_drain_bts_buffer(); + handled = intel_pmu_drain_bts_buffer(); status = intel_pmu_get_status(); if (!status) { intel_pmu_enable_all(0); - return 0; + return handled; } loops = 0; again: + intel_pmu_ack_status(status); if (++loops > 100) { WARN_ONCE(1, "perfevents: irq loop stuck!\n"); perf_event_print_debug(); @@ -691,19 +738,22 @@ again: } inc_irq_stat(apic_perf_irqs); - ack = status; intel_pmu_lbr_read(); /* * PEBS overflow sets bit 62 in the global status register */ - if (__test_and_clear_bit(62, (unsigned long *)&status)) + if (__test_and_clear_bit(62, (unsigned long *)&status)) { + handled++; x86_pmu.drain_pebs(regs); + } for_each_set_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) { struct perf_event *event = cpuc->events[bit]; + handled++; + if (!test_bit(bit, cpuc->active_mask)) continue; @@ -713,11 +763,9 @@ again: data.period = event->hw.last_period; if (perf_event_overflow(event, 1, &data, regs)) - x86_pmu_stop(event); + x86_pmu_stop(event, 0); } - intel_pmu_ack_status(ack); - /* * Repeat if there is more work to be done: */ @@ -727,7 +775,7 @@ again: done: intel_pmu_enable_all(0); - return 1; + return handled; } static struct event_constraint * diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index 18018d1311c..4977f9c400e 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -214,7 +214,7 @@ static void intel_pmu_disable_bts(void) update_debugctlmsr(debugctlmsr); } -static void intel_pmu_drain_bts_buffer(void) +static int intel_pmu_drain_bts_buffer(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct debug_store *ds = cpuc->ds; @@ -231,16 +231,16 @@ static void intel_pmu_drain_bts_buffer(void) struct pt_regs regs; if (!event) - return; + return 0; if (!ds) - return; + return 0; at = (struct bts_record *)(unsigned long)ds->bts_buffer_base; top = (struct bts_record *)(unsigned long)ds->bts_index; if (top <= at) - return; + return 0; ds->bts_index = ds->bts_buffer_base; @@ -256,7 +256,7 @@ static void intel_pmu_drain_bts_buffer(void) perf_prepare_sample(&header, &data, event, ®s); if (perf_output_begin(&handle, event, header.size * (top - at), 1, 1)) - return; + return 1; for (; at < top; at++) { data.ip = at->from; @@ -270,6 +270,7 @@ static void intel_pmu_drain_bts_buffer(void) /* There's new data available. */ event->hw.interrupts++; event->pending_kill = POLL_IN; + return 1; } /* @@ -491,7 +492,7 @@ static void __intel_pmu_pebs_event(struct perf_event *event, regs.flags &= ~PERF_EFLAGS_EXACT; if (perf_event_overflow(event, 1, &data, ®s)) - x86_pmu_stop(event); + x86_pmu_stop(event, 0); } static void intel_pmu_drain_pebs_core(struct pt_regs *iregs) diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index 107711bf0ee..81400b93e69 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -18,6 +18,8 @@ struct p4_event_bind { unsigned int opcode; /* Event code and ESCR selector */ unsigned int escr_msr[2]; /* ESCR MSR for this event */ + unsigned int escr_emask; /* valid ESCR EventMask bits */ + unsigned int shared; /* event is shared across threads */ char cntr[2][P4_CNTR_LIMIT]; /* counter index (offset), -1 on abscence */ }; @@ -66,231 +68,435 @@ static struct p4_event_bind p4_event_bind_map[] = { [P4_EVENT_TC_DELIVER_MODE] = { .opcode = P4_OPCODE(P4_EVENT_TC_DELIVER_MODE), .escr_msr = { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DD) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DB) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DI) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BD) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BB) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BI) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, ID), + .shared = 1, .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_BPU_FETCH_REQUEST] = { .opcode = P4_OPCODE(P4_EVENT_BPU_FETCH_REQUEST), .escr_msr = { MSR_P4_BPU_ESCR0, MSR_P4_BPU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BPU_FETCH_REQUEST, TCMISS), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_ITLB_REFERENCE] = { .opcode = P4_OPCODE(P4_EVENT_ITLB_REFERENCE), .escr_msr = { MSR_P4_ITLB_ESCR0, MSR_P4_ITLB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT) | + P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, MISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT_UK), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_MEMORY_CANCEL] = { .opcode = P4_OPCODE(P4_EVENT_MEMORY_CANCEL), .escr_msr = { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, ST_RB_FULL) | + P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, 64K_CONF), .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_MEMORY_COMPLETE] = { .opcode = P4_OPCODE(P4_EVENT_MEMORY_COMPLETE), .escr_msr = { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, LSC) | + P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, SSC), .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_LOAD_PORT_REPLAY] = { .opcode = P4_OPCODE(P4_EVENT_LOAD_PORT_REPLAY), .escr_msr = { MSR_P4_SAAT_ESCR0, MSR_P4_SAAT_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_LOAD_PORT_REPLAY, SPLIT_LD), .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_STORE_PORT_REPLAY] = { .opcode = P4_OPCODE(P4_EVENT_STORE_PORT_REPLAY), .escr_msr = { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_STORE_PORT_REPLAY, SPLIT_ST), .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_MOB_LOAD_REPLAY] = { .opcode = P4_OPCODE(P4_EVENT_MOB_LOAD_REPLAY), .escr_msr = { MSR_P4_MOB_ESCR0, MSR_P4_MOB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STA) | + P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STD) | + P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, PARTIAL_DATA) | + P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, UNALGN_ADDR), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_PAGE_WALK_TYPE] = { .opcode = P4_OPCODE(P4_EVENT_PAGE_WALK_TYPE), .escr_msr = { MSR_P4_PMH_ESCR0, MSR_P4_PMH_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, DTMISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, ITMISS), + .shared = 1, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_BSQ_CACHE_REFERENCE] = { .opcode = P4_OPCODE(P4_EVENT_BSQ_CACHE_REFERENCE), .escr_msr = { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_IOQ_ALLOCATION] = { .opcode = P4_OPCODE(P4_EVENT_IOQ_ALLOCATION), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, DEFAULT) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_READ) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_WRITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_UC) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WC) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WT) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WP) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WB) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OWN) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OTHER) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, PREFETCH), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_IOQ_ACTIVE_ENTRIES] = { /* shared ESCR */ .opcode = P4_OPCODE(P4_EVENT_IOQ_ACTIVE_ENTRIES), .escr_msr = { MSR_P4_FSB_ESCR1, MSR_P4_FSB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, DEFAULT) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_READ) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_WRITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_UC) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WC) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WT) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WP) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WB) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OWN) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OTHER) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, PREFETCH), .cntr = { {2, -1, -1}, {3, -1, -1} }, }, [P4_EVENT_FSB_DATA_ACTIVITY] = { .opcode = P4_OPCODE(P4_EVENT_FSB_DATA_ACTIVITY), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OTHER) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_DRV) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OWN) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OTHER), + .shared = 1, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_BSQ_ALLOCATION] = { /* shared ESCR, broken CCCR1 */ .opcode = P4_OPCODE(P4_EVENT_BSQ_ALLOCATION), .escr_msr = { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR0 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_IO_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LOCK_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_CACHE_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_SPLIT_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_DEM_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_ORD_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE2), .cntr = { {0, -1, -1}, {1, -1, -1} }, }, [P4_EVENT_BSQ_ACTIVE_ENTRIES] = { /* shared ESCR */ .opcode = P4_OPCODE(P4_EVENT_BSQ_ACTIVE_ENTRIES), .escr_msr = { MSR_P4_BSU_ESCR1 , MSR_P4_BSU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_IO_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LOCK_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_CACHE_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_SPLIT_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_DEM_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_ORD_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE2), .cntr = { {2, -1, -1}, {3, -1, -1} }, }, [P4_EVENT_SSE_INPUT_ASSIST] = { .opcode = P4_OPCODE(P4_EVENT_SSE_INPUT_ASSIST), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_SSE_INPUT_ASSIST, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_PACKED_SP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_PACKED_SP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_SP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_PACKED_DP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_PACKED_DP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_DP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_SCALAR_SP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_SCALAR_SP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_SP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_SCALAR_DP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_SCALAR_DP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_DP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_64BIT_MMX_UOP] = { .opcode = P4_OPCODE(P4_EVENT_64BIT_MMX_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_64BIT_MMX_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_128BIT_MMX_UOP] = { .opcode = P4_OPCODE(P4_EVENT_128BIT_MMX_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_128BIT_MMX_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_X87_FP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_X87_FP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_X87_FP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_TC_MISC] = { .opcode = P4_OPCODE(P4_EVENT_TC_MISC), .escr_msr = { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_TC_MISC, FLUSH), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_GLOBAL_POWER_EVENTS] = { .opcode = P4_OPCODE(P4_EVENT_GLOBAL_POWER_EVENTS), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_TC_MS_XFER] = { .opcode = P4_OPCODE(P4_EVENT_TC_MS_XFER), .escr_msr = { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_TC_MS_XFER, CISC), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_UOP_QUEUE_WRITES] = { .opcode = P4_OPCODE(P4_EVENT_UOP_QUEUE_WRITES), .escr_msr = { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_BUILD) | + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_DELIVER) | + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_ROM), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE] = { .opcode = P4_OPCODE(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE), .escr_msr = { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR0 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CONDITIONAL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CALL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, RETURN) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, INDIRECT), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_RETIRED_BRANCH_TYPE] = { .opcode = P4_OPCODE(P4_EVENT_RETIRED_BRANCH_TYPE), .escr_msr = { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CALL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_RESOURCE_STALL] = { .opcode = P4_OPCODE(P4_EVENT_RESOURCE_STALL), .escr_msr = { MSR_P4_ALF_ESCR0, MSR_P4_ALF_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_RESOURCE_STALL, SBFULL), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_WC_BUFFER] = { .opcode = P4_OPCODE(P4_EVENT_WC_BUFFER), .escr_msr = { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_EVICTS) | + P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_FULL_EVICTS), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_B2B_CYCLES] = { .opcode = P4_OPCODE(P4_EVENT_B2B_CYCLES), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = 0, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_BNR] = { .opcode = P4_OPCODE(P4_EVENT_BNR), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = 0, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_SNOOP] = { .opcode = P4_OPCODE(P4_EVENT_SNOOP), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = 0, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_RESPONSE] = { .opcode = P4_OPCODE(P4_EVENT_RESPONSE), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = 0, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_FRONT_END_EVENT] = { .opcode = P4_OPCODE(P4_EVENT_FRONT_END_EVENT), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, NBOGUS) | + P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, BOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_EXECUTION_EVENT] = { .opcode = P4_OPCODE(P4_EVENT_EXECUTION_EVENT), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS0) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS1) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS2) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS3) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS0) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS1) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS2) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS3), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_REPLAY_EVENT] = { .opcode = P4_OPCODE(P4_EVENT_REPLAY_EVENT), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, NBOGUS) | + P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, BOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_INSTR_RETIRED] = { .opcode = P4_OPCODE(P4_EVENT_INSTR_RETIRED), .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSTAG) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSNTAG) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSTAG), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_UOPS_RETIRED] = { .opcode = P4_OPCODE(P4_EVENT_UOPS_RETIRED), .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, NBOGUS) | + P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, BOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_UOP_TYPE] = { .opcode = P4_OPCODE(P4_EVENT_UOP_TYPE), .escr_msr = { MSR_P4_RAT_ESCR0, MSR_P4_RAT_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGLOADS) | + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGSTORES), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_BRANCH_RETIRED] = { .opcode = P4_OPCODE(P4_EVENT_BRANCH_RETIRED), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNP) | + P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNM) | + P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTP) | + P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTM), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_MISPRED_BRANCH_RETIRED] = { .opcode = P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED), .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_X87_ASSIST] = { .opcode = P4_OPCODE(P4_EVENT_X87_ASSIST), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSU) | + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSO) | + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAO) | + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAU) | + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, PREA), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_MACHINE_CLEAR] = { .opcode = P4_OPCODE(P4_EVENT_MACHINE_CLEAR), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, CLEAR) | + P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, MOCLEAR) | + P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, SMCLEAR), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_INSTR_COMPLETED] = { .opcode = P4_OPCODE(P4_EVENT_INSTR_COMPLETED), .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, NBOGUS) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, BOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, }; @@ -428,29 +634,73 @@ static u64 p4_pmu_event_map(int hw_event) return config; } +/* check cpu model specifics */ +static bool p4_event_match_cpu_model(unsigned int event_idx) +{ + /* INSTR_COMPLETED event only exist for model 3, 4, 6 (Prescott) */ + if (event_idx == P4_EVENT_INSTR_COMPLETED) { + if (boot_cpu_data.x86_model != 3 && + boot_cpu_data.x86_model != 4 && + boot_cpu_data.x86_model != 6) + return false; + } + + /* + * For info + * - IQ_ESCR0, IQ_ESCR1 only for models 1 and 2 + */ + + return true; +} + static int p4_validate_raw_event(struct perf_event *event) { - unsigned int v; + unsigned int v, emask; - /* user data may have out-of-bound event index */ + /* User data may have out-of-bound event index */ v = p4_config_unpack_event(event->attr.config); - if (v >= ARRAY_SIZE(p4_event_bind_map)) { - pr_warning("P4 PMU: Unknown event code: %d\n", v); + if (v >= ARRAY_SIZE(p4_event_bind_map)) + return -EINVAL; + + /* It may be unsupported: */ + if (!p4_event_match_cpu_model(v)) return -EINVAL; + + /* + * NOTE: P4_CCCR_THREAD_ANY has not the same meaning as + * in Architectural Performance Monitoring, it means not + * on _which_ logical cpu to count but rather _when_, ie it + * depends on logical cpu state -- count event if one cpu active, + * none, both or any, so we just allow user to pass any value + * desired. + * + * In turn we always set Tx_OS/Tx_USR bits bound to logical + * cpu without their propagation to another cpu + */ + + /* + * if an event is shared accross the logical threads + * the user needs special permissions to be able to use it + */ + if (p4_event_bind_map[v].shared) { + if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) + return -EACCES; } + /* ESCR EventMask bits may be invalid */ + emask = p4_config_unpack_escr(event->attr.config) & P4_ESCR_EVENTMASK_MASK; + if (emask & ~p4_event_bind_map[v].escr_emask) + return -EINVAL; + /* - * it may have some screwed PEBS bits + * it may have some invalid PEBS bits */ - if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE)) { - pr_warning("P4 PMU: PEBS are not supported yet\n"); + if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE)) return -EINVAL; - } + v = p4_config_unpack_metric(event->attr.config); - if (v >= ARRAY_SIZE(p4_pebs_bind_map)) { - pr_warning("P4 PMU: Unknown metric code: %d\n", v); + if (v >= ARRAY_SIZE(p4_pebs_bind_map)) return -EINVAL; - } return 0; } @@ -478,25 +728,21 @@ static int p4_hw_config(struct perf_event *event) if (event->attr.type == PERF_TYPE_RAW) { + /* + * Clear bits we reserve to be managed by kernel itself + * and never allowed from a user space + */ + event->attr.config &= P4_CONFIG_MASK; + rc = p4_validate_raw_event(event); if (rc) goto out; /* - * We don't control raw events so it's up to the caller - * to pass sane values (and we don't count the thread number - * on HT machine but allow HT-compatible specifics to be - * passed on) - * * Note that for RAW events we allow user to use P4_CCCR_RESERVED * bits since we keep additional info here (for cache events and etc) - * - * XXX: HT wide things should check perf_paranoid_cpu() && - * CAP_SYS_ADMIN */ - event->hw.config |= event->attr.config & - (p4_config_pack_escr(P4_ESCR_MASK_HT) | - p4_config_pack_cccr(P4_CCCR_MASK_HT | P4_CCCR_RESERVED)); + event->hw.config |= event->attr.config; } rc = x86_setup_perfctr(event); @@ -656,9 +902,14 @@ static int p4_pmu_handle_irq(struct pt_regs *regs) cpuc = &__get_cpu_var(cpu_hw_events); for (idx = 0; idx < x86_pmu.num_counters; idx++) { + int overflow; - if (!test_bit(idx, cpuc->active_mask)) + if (!test_bit(idx, cpuc->active_mask)) { + /* catch in-flight IRQs */ + if (__test_and_clear_bit(idx, cpuc->running)) + handled++; continue; + } event = cpuc->events[idx]; hwc = &event->hw; @@ -666,12 +917,14 @@ static int p4_pmu_handle_irq(struct pt_regs *regs) WARN_ON_ONCE(hwc->idx != idx); /* it might be unflagged overflow */ - handled = p4_pmu_clear_cccr_ovf(hwc); + overflow = p4_pmu_clear_cccr_ovf(hwc); val = x86_perf_event_update(event); - if (!handled && (val & (1ULL << (x86_pmu.cntval_bits - 1)))) + if (!overflow && (val & (1ULL << (x86_pmu.cntval_bits - 1)))) continue; + handled += overflow; + /* event overflow for sure */ data.period = event->hw.last_period; diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index fb329e9f849..d9f4ff8fcd6 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -700,11 +700,10 @@ static void probe_nmi_watchdog(void) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: - if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15 && - boot_cpu_data.x86 != 16 && boot_cpu_data.x86 != 17) - return; - wd_ops = &k7_wd_ops; - break; + if (boot_cpu_data.x86 == 6 || + (boot_cpu_data.x86 >= 0xf && boot_cpu_data.x86 <= 0x15)) + wd_ops = &k7_wd_ops; + return; case X86_VENDOR_INTEL: /* Work around where perfctr1 doesn't have a working enable * bit as described in the following errata: diff --git a/arch/x86/kernel/cpu/scattered.c b/arch/x86/kernel/cpu/scattered.c index 34b4dad6f0b..c7f64e6f537 100644 --- a/arch/x86/kernel/cpu/scattered.c +++ b/arch/x86/kernel/cpu/scattered.c @@ -31,6 +31,7 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) const struct cpuid_bit *cb; static const struct cpuid_bit __cpuinitconst cpuid_bits[] = { + { X86_FEATURE_DTS, CR_EAX, 0, 0x00000006, 0 }, { X86_FEATURE_IDA, CR_EAX, 1, 0x00000006, 0 }, { X86_FEATURE_ARAT, CR_EAX, 2, 0x00000006, 0 }, { X86_FEATURE_PLN, CR_EAX, 4, 0x00000006, 0 }, @@ -43,6 +44,12 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) { X86_FEATURE_LBRV, CR_EDX, 1, 0x8000000a, 0 }, { X86_FEATURE_SVML, CR_EDX, 2, 0x8000000a, 0 }, { X86_FEATURE_NRIPS, CR_EDX, 3, 0x8000000a, 0 }, + { X86_FEATURE_TSCRATEMSR, CR_EDX, 4, 0x8000000a, 0 }, + { X86_FEATURE_VMCBCLEAN, CR_EDX, 5, 0x8000000a, 0 }, + { X86_FEATURE_FLUSHBYASID, CR_EDX, 6, 0x8000000a, 0 }, + { X86_FEATURE_DECODEASSISTS, CR_EDX, 7, 0x8000000a, 0 }, + { X86_FEATURE_PAUSEFILTER, CR_EDX,10, 0x8000000a, 0 }, + { X86_FEATURE_PFTHRESHOLD, CR_EDX,12, 0x8000000a, 0 }, { 0, 0, 0, 0, 0 } }; diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index ebd4c51d096..764c7c2b181 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -28,6 +28,8 @@ #include <asm/reboot.h> #include <asm/virtext.h> +int in_crash_kexec; + #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) static void kdump_nmi_callback(int cpu, struct die_args *args) @@ -61,6 +63,7 @@ static void kdump_nmi_callback(int cpu, struct die_args *args) static void kdump_nmi_shootdown_cpus(void) { + in_crash_kexec = 1; nmi_shootdown_cpus(kdump_nmi_callback); disable_local_APIC(); diff --git a/arch/x86/kernel/crash_dump_64.c b/arch/x86/kernel/crash_dump_64.c index 045b36cada6..994828899e0 100644 --- a/arch/x86/kernel/crash_dump_64.c +++ b/arch/x86/kernel/crash_dump_64.c @@ -34,7 +34,7 @@ ssize_t copy_oldmem_page(unsigned long pfn, char *buf, if (!csize) return 0; - vaddr = ioremap(pfn << PAGE_SHIFT, PAGE_SIZE); + vaddr = ioremap_cache(pfn << PAGE_SHIFT, PAGE_SIZE); if (!vaddr) return -ENOMEM; @@ -46,6 +46,7 @@ ssize_t copy_oldmem_page(unsigned long pfn, char *buf, } else memcpy(buf, vaddr + offset, csize); + set_iounmap_nonlazy(); iounmap(vaddr); return csize; } diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 0d6fc71bedb..0c2b7ef7a34 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -15,6 +15,7 @@ #include <linux/pfn.h> #include <linux/suspend.h> #include <linux/firmware-map.h> +#include <linux/memblock.h> #include <asm/e820.h> #include <asm/proto.h> @@ -738,73 +739,7 @@ core_initcall(e820_mark_nvs_memory); #endif /* - * Find a free area with specified alignment in a specific range. - */ -u64 __init find_e820_area(u64 start, u64 end, u64 size, u64 align) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - u64 addr; - u64 ei_start, ei_last; - - if (ei->type != E820_RAM) - continue; - - ei_last = ei->addr + ei->size; - ei_start = ei->addr; - addr = find_early_area(ei_start, ei_last, start, end, - size, align); - - if (addr != -1ULL) - return addr; - } - return -1ULL; -} - -u64 __init find_fw_memmap_area(u64 start, u64 end, u64 size, u64 align) -{ - return find_e820_area(start, end, size, align); -} - -u64 __init get_max_mapped(void) -{ - u64 end = max_pfn_mapped; - - end <<= PAGE_SHIFT; - - return end; -} -/* - * Find next free range after *start - */ -u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - u64 addr; - u64 ei_start, ei_last; - - if (ei->type != E820_RAM) - continue; - - ei_last = ei->addr + ei->size; - ei_start = ei->addr; - addr = find_early_area_size(ei_start, ei_last, start, - sizep, align); - - if (addr != -1ULL) - return addr; - } - - return -1ULL; -} - -/* - * pre allocated 4k and reserved it in e820 + * pre allocated 4k and reserved it in memblock and e820_saved */ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) { @@ -813,8 +748,8 @@ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) u64 start; for (start = startt; ; start += size) { - start = find_e820_area_size(start, &size, align); - if (!(start + 1)) + start = memblock_x86_find_in_range_size(start, &size, align); + if (start == MEMBLOCK_ERROR) return 0; if (size >= sizet) break; @@ -830,10 +765,9 @@ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) addr = round_down(start + size - sizet, align); if (addr < start) return 0; - e820_update_range(addr, sizet, E820_RAM, E820_RESERVED); + memblock_x86_reserve_range(addr, addr + sizet, "new next"); e820_update_range_saved(addr, sizet, E820_RAM, E820_RESERVED); - printk(KERN_INFO "update e820 for early_reserve_e820\n"); - update_e820(); + printk(KERN_INFO "update e820_saved for early_reserve_e820\n"); update_e820_saved(); return addr; @@ -895,74 +829,6 @@ unsigned long __init e820_end_of_low_ram_pfn(void) { return e820_end_pfn(1UL<<(32 - PAGE_SHIFT), E820_RAM); } -/* - * Finds an active region in the address range from start_pfn to last_pfn and - * returns its range in ei_startpfn and ei_endpfn for the e820 entry. - */ -int __init e820_find_active_region(const struct e820entry *ei, - unsigned long start_pfn, - unsigned long last_pfn, - unsigned long *ei_startpfn, - unsigned long *ei_endpfn) -{ - u64 align = PAGE_SIZE; - - *ei_startpfn = round_up(ei->addr, align) >> PAGE_SHIFT; - *ei_endpfn = round_down(ei->addr + ei->size, align) >> PAGE_SHIFT; - - /* Skip map entries smaller than a page */ - if (*ei_startpfn >= *ei_endpfn) - return 0; - - /* Skip if map is outside the node */ - if (ei->type != E820_RAM || *ei_endpfn <= start_pfn || - *ei_startpfn >= last_pfn) - return 0; - - /* Check for overlaps */ - if (*ei_startpfn < start_pfn) - *ei_startpfn = start_pfn; - if (*ei_endpfn > last_pfn) - *ei_endpfn = last_pfn; - - return 1; -} - -/* Walk the e820 map and register active regions within a node */ -void __init e820_register_active_regions(int nid, unsigned long start_pfn, - unsigned long last_pfn) -{ - unsigned long ei_startpfn; - unsigned long ei_endpfn; - int i; - - for (i = 0; i < e820.nr_map; i++) - if (e820_find_active_region(&e820.map[i], - start_pfn, last_pfn, - &ei_startpfn, &ei_endpfn)) - add_active_range(nid, ei_startpfn, ei_endpfn); -} - -/* - * Find the hole size (in bytes) in the memory range. - * @start: starting address of the memory range to scan - * @end: ending address of the memory range to scan - */ -u64 __init e820_hole_size(u64 start, u64 end) -{ - unsigned long start_pfn = start >> PAGE_SHIFT; - unsigned long last_pfn = end >> PAGE_SHIFT; - unsigned long ei_startpfn, ei_endpfn, ram = 0; - int i; - - for (i = 0; i < e820.nr_map; i++) { - if (e820_find_active_region(&e820.map[i], - start_pfn, last_pfn, - &ei_startpfn, &ei_endpfn)) - ram += ei_endpfn - ei_startpfn; - } - return end - start - ((u64)ram << PAGE_SHIFT); -} static void early_panic(char *msg) { @@ -1210,3 +1076,48 @@ void __init setup_memory_map(void) printk(KERN_INFO "BIOS-provided physical RAM map:\n"); e820_print_map(who); } + +void __init memblock_x86_fill(void) +{ + int i; + u64 end; + + /* + * EFI may have more than 128 entries + * We are safe to enable resizing, beause memblock_x86_fill() + * is rather later for x86 + */ + memblock_can_resize = 1; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + + end = ei->addr + ei->size; + if (end != (resource_size_t)end) + continue; + + if (ei->type != E820_RAM && ei->type != E820_RESERVED_KERN) + continue; + + memblock_add(ei->addr, ei->size); + } + + memblock_analyze(); + memblock_dump_all(); +} + +void __init memblock_find_dma_reserve(void) +{ +#ifdef CONFIG_X86_64 + u64 free_size_pfn; + u64 mem_size_pfn; + /* + * need to find out used area below MAX_DMA_PFN + * need to use memblock to get free size in [0, MAX_DMA_PFN] + * at first, and assume boot_mem will not take below MAX_DMA_PFN + */ + mem_size_pfn = memblock_x86_memory_in_range(0, MAX_DMA_PFN << PAGE_SHIFT) >> PAGE_SHIFT; + free_size_pfn = memblock_x86_free_memory_in_range(0, MAX_DMA_PFN << PAGE_SHIFT) >> PAGE_SHIFT; + set_dma_reserve(mem_size_pfn - free_size_pfn); +#endif +} diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c index e5cc7e82e60..76b8cd953de 100644 --- a/arch/x86/kernel/early-quirks.c +++ b/arch/x86/kernel/early-quirks.c @@ -18,7 +18,6 @@ #include <asm/apic.h> #include <asm/iommu.h> #include <asm/gart.h> -#include <asm/hpet.h> static void __init fix_hypertransport_config(int num, int slot, int func) { @@ -98,7 +97,6 @@ static void __init nvidia_bugs(int num, int slot, int func) } #if defined(CONFIG_ACPI) && defined(CONFIG_X86_IO_APIC) -#if defined(CONFIG_ACPI) && defined(CONFIG_X86_IO_APIC) static u32 __init ati_ixp4x0_rev(int num, int slot, int func) { u32 d; @@ -116,7 +114,6 @@ static u32 __init ati_ixp4x0_rev(int num, int slot, int func) d &= 0xff; return d; } -#endif static void __init ati_bugs(int num, int slot, int func) { @@ -192,21 +189,6 @@ static void __init ati_bugs_contd(int num, int slot, int func) } #endif -/* - * Force the read back of the CMP register in hpet_next_event() - * to work around the problem that the CMP register write seems to be - * delayed. See hpet_next_event() for details. - * - * We do this on all SMBUS incarnations for now until we have more - * information about the affected chipsets. - */ -static void __init ati_hpet_bugs(int num, int slot, int func) -{ -#ifdef CONFIG_HPET_TIMER - hpet_readback_cmp = 1; -#endif -} - #define QFLAG_APPLY_ONCE 0x1 #define QFLAG_APPLIED 0x2 #define QFLAG_DONE (QFLAG_APPLY_ONCE|QFLAG_APPLIED) @@ -236,8 +218,6 @@ static struct chipset early_qrk[] __initdata = { PCI_CLASS_SERIAL_SMBUS, PCI_ANY_ID, 0, ati_bugs }, { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_SBX00_SMBUS, PCI_CLASS_SERIAL_SMBUS, PCI_ANY_ID, 0, ati_bugs_contd }, - { PCI_VENDOR_ID_ATI, PCI_ANY_ID, - PCI_CLASS_SERIAL_SMBUS, PCI_ANY_ID, 0, ati_hpet_bugs }, {} }; diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c index fa99bae75ac..4572f25f932 100644 --- a/arch/x86/kernel/early_printk.c +++ b/arch/x86/kernel/early_printk.c @@ -14,6 +14,7 @@ #include <xen/hvc-console.h> #include <asm/pci-direct.h> #include <asm/fixmap.h> +#include <asm/mrst.h> #include <asm/pgtable.h> #include <linux/usb/ehci_def.h> @@ -239,6 +240,18 @@ static int __init setup_early_printk(char *buf) if (!strncmp(buf, "xen", 3)) early_console_register(&xenboot_console, keep); #endif +#ifdef CONFIG_X86_MRST_EARLY_PRINTK + if (!strncmp(buf, "mrst", 4)) { + mrst_early_console_init(); + early_console_register(&early_mrst_console, keep); + } + + if (!strncmp(buf, "hsu", 3)) { + hsu_early_console_init(); + early_console_register(&early_hsu_console, keep); + } + +#endif buf++; } return 0; diff --git a/arch/x86/kernel/early_printk_mrst.c b/arch/x86/kernel/early_printk_mrst.c new file mode 100644 index 00000000000..65df603622b --- /dev/null +++ b/arch/x86/kernel/early_printk_mrst.c @@ -0,0 +1,319 @@ +/* + * early_printk_mrst.c - early consoles for Intel MID platforms + * + * Copyright (c) 2008-2010, Intel Corporation + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + */ + +/* + * This file implements two early consoles named mrst and hsu. + * mrst is based on Maxim3110 spi-uart device, it exists in both + * Moorestown and Medfield platforms, while hsu is based on a High + * Speed UART device which only exists in the Medfield platform + */ + +#include <linux/serial_reg.h> +#include <linux/serial_mfd.h> +#include <linux/kmsg_dump.h> +#include <linux/console.h> +#include <linux/kernel.h> +#include <linux/delay.h> +#include <linux/init.h> +#include <linux/io.h> + +#include <asm/fixmap.h> +#include <asm/pgtable.h> +#include <asm/mrst.h> + +#define MRST_SPI_TIMEOUT 0x200000 +#define MRST_REGBASE_SPI0 0xff128000 +#define MRST_REGBASE_SPI1 0xff128400 +#define MRST_CLK_SPI0_REG 0xff11d86c + +/* Bit fields in CTRLR0 */ +#define SPI_DFS_OFFSET 0 + +#define SPI_FRF_OFFSET 4 +#define SPI_FRF_SPI 0x0 +#define SPI_FRF_SSP 0x1 +#define SPI_FRF_MICROWIRE 0x2 +#define SPI_FRF_RESV 0x3 + +#define SPI_MODE_OFFSET 6 +#define SPI_SCPH_OFFSET 6 +#define SPI_SCOL_OFFSET 7 +#define SPI_TMOD_OFFSET 8 +#define SPI_TMOD_TR 0x0 /* xmit & recv */ +#define SPI_TMOD_TO 0x1 /* xmit only */ +#define SPI_TMOD_RO 0x2 /* recv only */ +#define SPI_TMOD_EPROMREAD 0x3 /* eeprom read mode */ + +#define SPI_SLVOE_OFFSET 10 +#define SPI_SRL_OFFSET 11 +#define SPI_CFS_OFFSET 12 + +/* Bit fields in SR, 7 bits */ +#define SR_MASK 0x7f /* cover 7 bits */ +#define SR_BUSY (1 << 0) +#define SR_TF_NOT_FULL (1 << 1) +#define SR_TF_EMPT (1 << 2) +#define SR_RF_NOT_EMPT (1 << 3) +#define SR_RF_FULL (1 << 4) +#define SR_TX_ERR (1 << 5) +#define SR_DCOL (1 << 6) + +struct dw_spi_reg { + u32 ctrl0; + u32 ctrl1; + u32 ssienr; + u32 mwcr; + u32 ser; + u32 baudr; + u32 txfltr; + u32 rxfltr; + u32 txflr; + u32 rxflr; + u32 sr; + u32 imr; + u32 isr; + u32 risr; + u32 txoicr; + u32 rxoicr; + u32 rxuicr; + u32 msticr; + u32 icr; + u32 dmacr; + u32 dmatdlr; + u32 dmardlr; + u32 idr; + u32 version; + + /* Currently operates as 32 bits, though only the low 16 bits matter */ + u32 dr; +} __packed; + +#define dw_readl(dw, name) __raw_readl(&(dw)->name) +#define dw_writel(dw, name, val) __raw_writel((val), &(dw)->name) + +/* Default use SPI0 register for mrst, we will detect Penwell and use SPI1 */ +static unsigned long mrst_spi_paddr = MRST_REGBASE_SPI0; + +static u32 *pclk_spi0; +/* Always contains an accessable address, start with 0 */ +static struct dw_spi_reg *pspi; + +static struct kmsg_dumper dw_dumper; +static int dumper_registered; + +static void dw_kmsg_dump(struct kmsg_dumper *dumper, + enum kmsg_dump_reason reason, + const char *s1, unsigned long l1, + const char *s2, unsigned long l2) +{ + int i; + + /* When run to this, we'd better re-init the HW */ + mrst_early_console_init(); + + for (i = 0; i < l1; i++) + early_mrst_console.write(&early_mrst_console, s1 + i, 1); + for (i = 0; i < l2; i++) + early_mrst_console.write(&early_mrst_console, s2 + i, 1); +} + +/* Set the ratio rate to 115200, 8n1, IRQ disabled */ +static void max3110_write_config(void) +{ + u16 config; + + config = 0xc001; + dw_writel(pspi, dr, config); +} + +/* Translate char to a eligible word and send to max3110 */ +static void max3110_write_data(char c) +{ + u16 data; + + data = 0x8000 | c; + dw_writel(pspi, dr, data); +} + +void mrst_early_console_init(void) +{ + u32 ctrlr0 = 0; + u32 spi0_cdiv; + u32 freq; /* Freqency info only need be searched once */ + + /* Base clk is 100 MHz, the actual clk = 100M / (clk_divider + 1) */ + pclk_spi0 = (void *)set_fixmap_offset_nocache(FIX_EARLYCON_MEM_BASE, + MRST_CLK_SPI0_REG); + spi0_cdiv = ((*pclk_spi0) & 0xe00) >> 9; + freq = 100000000 / (spi0_cdiv + 1); + + if (mrst_identify_cpu() == MRST_CPU_CHIP_PENWELL) + mrst_spi_paddr = MRST_REGBASE_SPI1; + + pspi = (void *)set_fixmap_offset_nocache(FIX_EARLYCON_MEM_BASE, + mrst_spi_paddr); + + /* Disable SPI controller */ + dw_writel(pspi, ssienr, 0); + + /* Set control param, 8 bits, transmit only mode */ + ctrlr0 = dw_readl(pspi, ctrl0); + + ctrlr0 &= 0xfcc0; + ctrlr0 |= 0xf | (SPI_FRF_SPI << SPI_FRF_OFFSET) + | (SPI_TMOD_TO << SPI_TMOD_OFFSET); + dw_writel(pspi, ctrl0, ctrlr0); + + /* + * Change the spi0 clk to comply with 115200 bps, use 100000 to + * calculate the clk dividor to make the clock a little slower + * than real baud rate. + */ + dw_writel(pspi, baudr, freq/100000); + + /* Disable all INT for early phase */ + dw_writel(pspi, imr, 0x0); + + /* Set the cs to spi-uart */ + dw_writel(pspi, ser, 0x2); + + /* Enable the HW, the last step for HW init */ + dw_writel(pspi, ssienr, 0x1); + + /* Set the default configuration */ + max3110_write_config(); + + /* Register the kmsg dumper */ + if (!dumper_registered) { + dw_dumper.dump = dw_kmsg_dump; + kmsg_dump_register(&dw_dumper); + dumper_registered = 1; + } +} + +/* Slave select should be called in the read/write function */ +static void early_mrst_spi_putc(char c) +{ + unsigned int timeout; + u32 sr; + + timeout = MRST_SPI_TIMEOUT; + /* Early putc needs to make sure the TX FIFO is not full */ + while (--timeout) { + sr = dw_readl(pspi, sr); + if (!(sr & SR_TF_NOT_FULL)) + cpu_relax(); + else + break; + } + + if (!timeout) + pr_warning("MRST earlycon: timed out\n"); + else + max3110_write_data(c); +} + +/* Early SPI only uses polling mode */ +static void early_mrst_spi_write(struct console *con, const char *str, unsigned n) +{ + int i; + + for (i = 0; i < n && *str; i++) { + if (*str == '\n') + early_mrst_spi_putc('\r'); + early_mrst_spi_putc(*str); + str++; + } +} + +struct console early_mrst_console = { + .name = "earlymrst", + .write = early_mrst_spi_write, + .flags = CON_PRINTBUFFER, + .index = -1, +}; + +/* + * Following is the early console based on Medfield HSU (High + * Speed UART) device. + */ +#define HSU_PORT2_PADDR 0xffa28180 + +static void __iomem *phsu; + +void hsu_early_console_init(void) +{ + u8 lcr; + + phsu = (void *)set_fixmap_offset_nocache(FIX_EARLYCON_MEM_BASE, + HSU_PORT2_PADDR); + + /* Disable FIFO */ + writeb(0x0, phsu + UART_FCR); + + /* Set to default 115200 bps, 8n1 */ + lcr = readb(phsu + UART_LCR); + writeb((0x80 | lcr), phsu + UART_LCR); + writeb(0x18, phsu + UART_DLL); + writeb(lcr, phsu + UART_LCR); + writel(0x3600, phsu + UART_MUL*4); + + writeb(0x8, phsu + UART_MCR); + writeb(0x7, phsu + UART_FCR); + writeb(0x3, phsu + UART_LCR); + + /* Clear IRQ status */ + readb(phsu + UART_LSR); + readb(phsu + UART_RX); + readb(phsu + UART_IIR); + readb(phsu + UART_MSR); + + /* Enable FIFO */ + writeb(0x7, phsu + UART_FCR); +} + +#define BOTH_EMPTY (UART_LSR_TEMT | UART_LSR_THRE) + +static void early_hsu_putc(char ch) +{ + unsigned int timeout = 10000; /* 10ms */ + u8 status; + + while (--timeout) { + status = readb(phsu + UART_LSR); + if (status & BOTH_EMPTY) + break; + udelay(1); + } + + /* Only write the char when there was no timeout */ + if (timeout) + writeb(ch, phsu + UART_TX); +} + +static void early_hsu_write(struct console *con, const char *str, unsigned n) +{ + int i; + + for (i = 0; i < n && *str; i++) { + if (*str == '\n') + early_hsu_putc('\r'); + early_hsu_putc(*str); + str++; + } +} + +struct console early_hsu_console = { + .name = "earlyhsu", + .write = early_hsu_write, + .flags = CON_PRINTBUFFER, + .index = -1, +}; diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index c2fa9b8b497..0fe27d7c625 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -30,6 +30,7 @@ #include <linux/init.h> #include <linux/efi.h> #include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/spinlock.h> #include <linux/uaccess.h> #include <linux/time.h> @@ -275,7 +276,7 @@ static void __init do_add_efi_memmap(void) sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); } -void __init efi_reserve_early(void) +void __init efi_memblock_x86_reserve_range(void) { unsigned long pmap; @@ -290,7 +291,7 @@ void __init efi_reserve_early(void) boot_params.efi_info.efi_memdesc_size; memmap.desc_version = boot_params.efi_info.efi_memdesc_version; memmap.desc_size = boot_params.efi_info.efi_memdesc_size; - reserve_early(pmap, pmap + memmap.nr_map * memmap.desc_size, + memblock_x86_reserve_range(pmap, pmap + memmap.nr_map * memmap.desc_size, "EFI memmap"); } diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index 227d00920d2..59e175e8959 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -115,8 +115,7 @@ /* unfortunately push/pop can't be no-op */ .macro PUSH_GS - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 .endm .macro POP_GS pop=0 addl $(4 + \pop), %esp @@ -140,14 +139,12 @@ #else /* CONFIG_X86_32_LAZY_GS */ .macro PUSH_GS - pushl %gs - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %gs /*CFI_REL_OFFSET gs, 0*/ .endm .macro POP_GS pop=0 -98: popl %gs - CFI_ADJUST_CFA_OFFSET -4 +98: popl_cfi %gs /*CFI_RESTORE gs*/ .if \pop <> 0 add $\pop, %esp @@ -195,35 +192,25 @@ .macro SAVE_ALL cld PUSH_GS - pushl %fs - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %fs /*CFI_REL_OFFSET fs, 0;*/ - pushl %es - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %es /*CFI_REL_OFFSET es, 0;*/ - pushl %ds - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ds /*CFI_REL_OFFSET ds, 0;*/ - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax CFI_REL_OFFSET eax, 0 - pushl %ebp - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebp CFI_REL_OFFSET ebp, 0 - pushl %edi - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %edi CFI_REL_OFFSET edi, 0 - pushl %esi - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %esi CFI_REL_OFFSET esi, 0 - pushl %edx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %edx CFI_REL_OFFSET edx, 0 - pushl %ecx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ecx CFI_REL_OFFSET ecx, 0 - pushl %ebx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebx CFI_REL_OFFSET ebx, 0 movl $(__USER_DS), %edx movl %edx, %ds @@ -234,39 +221,29 @@ .endm .macro RESTORE_INT_REGS - popl %ebx - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %ebx CFI_RESTORE ebx - popl %ecx - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %ecx CFI_RESTORE ecx - popl %edx - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %edx CFI_RESTORE edx - popl %esi - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %esi CFI_RESTORE esi - popl %edi - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %edi CFI_RESTORE edi - popl %ebp - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %ebp CFI_RESTORE ebp - popl %eax - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %eax CFI_RESTORE eax .endm .macro RESTORE_REGS pop=0 RESTORE_INT_REGS -1: popl %ds - CFI_ADJUST_CFA_OFFSET -4 +1: popl_cfi %ds /*CFI_RESTORE ds;*/ -2: popl %es - CFI_ADJUST_CFA_OFFSET -4 +2: popl_cfi %es /*CFI_RESTORE es;*/ -3: popl %fs - CFI_ADJUST_CFA_OFFSET -4 +3: popl_cfi %fs /*CFI_RESTORE fs;*/ POP_GS \pop .pushsection .fixup, "ax" @@ -320,16 +297,12 @@ ENTRY(ret_from_fork) CFI_STARTPROC - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax call schedule_tail GET_THREAD_INFO(%ebp) - popl %eax - CFI_ADJUST_CFA_OFFSET -4 - pushl $0x0202 # Reset kernel eflags - CFI_ADJUST_CFA_OFFSET 4 - popfl - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %eax + pushl_cfi $0x0202 # Reset kernel eflags + popfl_cfi jmp syscall_exit CFI_ENDPROC END(ret_from_fork) @@ -409,29 +382,23 @@ sysenter_past_esp: * enough kernel state to call TRACE_IRQS_OFF can be called - but * we immediately enable interrupts at that point anyway. */ - pushl $(__USER_DS) - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $__USER_DS /*CFI_REL_OFFSET ss, 0*/ - pushl %ebp - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebp CFI_REL_OFFSET esp, 0 - pushfl + pushfl_cfi orl $X86_EFLAGS_IF, (%esp) - CFI_ADJUST_CFA_OFFSET 4 - pushl $(__USER_CS) - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $__USER_CS /*CFI_REL_OFFSET cs, 0*/ /* * Push current_thread_info()->sysenter_return to the stack. * A tiny bit of offset fixup is necessary - 4*4 means the 4 words * pushed above; +8 corresponds to copy_thread's esp0 setting. */ - pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp) - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi (TI_sysenter_return-THREAD_SIZE_asm+8+4*4)(%esp) CFI_REL_OFFSET eip, 0 - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax SAVE_ALL ENABLE_INTERRUPTS(CLBR_NONE) @@ -486,8 +453,7 @@ sysenter_audit: movl %eax,%edx /* 2nd arg: syscall number */ movl $AUDIT_ARCH_I386,%eax /* 1st arg: audit arch */ call audit_syscall_entry - pushl %ebx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebx movl PT_EAX(%esp),%eax /* reload syscall number */ jmp sysenter_do_call @@ -529,8 +495,7 @@ ENDPROC(ia32_sysenter_target) # system call handler stub ENTRY(system_call) RING0_INT_FRAME # can't unwind into user space anyway - pushl %eax # save orig_eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax # save orig_eax SAVE_ALL GET_THREAD_INFO(%ebp) # system call tracing in operation / emulation @@ -566,7 +531,6 @@ restore_all_notrace: je ldt_ss # returning to user-space with LDT SS restore_nocheck: RESTORE_REGS 4 # skip orig_eax/error_code - CFI_ADJUST_CFA_OFFSET -4 irq_return: INTERRUPT_RETURN .section .fixup,"ax" @@ -619,10 +583,8 @@ ldt_ss: shr $16, %edx mov %dl, GDT_ESPFIX_SS + 4 /* bits 16..23 */ mov %dh, GDT_ESPFIX_SS + 7 /* bits 24..31 */ - pushl $__ESPFIX_SS - CFI_ADJUST_CFA_OFFSET 4 - push %eax /* new kernel esp */ - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $__ESPFIX_SS + pushl_cfi %eax /* new kernel esp */ /* Disable interrupts, but do not irqtrace this section: we * will soon execute iret and the tracer was already set to * the irqstate after the iret */ @@ -666,11 +628,9 @@ work_notifysig: # deal with pending signals and ALIGN work_notifysig_v86: - pushl %ecx # save ti_flags for do_notify_resume - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ecx # save ti_flags for do_notify_resume call save_v86_state # %eax contains pt_regs pointer - popl %ecx - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %ecx movl %eax, %esp #else movl %esp, %eax @@ -750,14 +710,18 @@ ptregs_##name: \ #define PTREGSCALL3(name) \ ALIGN; \ ptregs_##name: \ + CFI_STARTPROC; \ leal 4(%esp),%eax; \ - pushl %eax; \ + pushl_cfi %eax; \ movl PT_EDX(%eax),%ecx; \ movl PT_ECX(%eax),%edx; \ movl PT_EBX(%eax),%eax; \ call sys_##name; \ addl $4,%esp; \ - ret + CFI_ADJUST_CFA_OFFSET -4; \ + ret; \ + CFI_ENDPROC; \ +ENDPROC(ptregs_##name) PTREGSCALL1(iopl) PTREGSCALL0(fork) @@ -772,15 +736,19 @@ PTREGSCALL1(vm86old) /* Clone is an oddball. The 4th arg is in %edi */ ALIGN; ptregs_clone: + CFI_STARTPROC leal 4(%esp),%eax - pushl %eax - pushl PT_EDI(%eax) + pushl_cfi %eax + pushl_cfi PT_EDI(%eax) movl PT_EDX(%eax),%ecx movl PT_ECX(%eax),%edx movl PT_EBX(%eax),%eax call sys_clone addl $8,%esp + CFI_ADJUST_CFA_OFFSET -8 ret + CFI_ENDPROC +ENDPROC(ptregs_clone) .macro FIXUP_ESPFIX_STACK /* @@ -795,10 +763,8 @@ ptregs_clone: mov GDT_ESPFIX_SS + 7, %ah /* bits 24..31 */ shl $16, %eax addl %esp, %eax /* the adjusted stack pointer */ - pushl $__KERNEL_DS - CFI_ADJUST_CFA_OFFSET 4 - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $__KERNEL_DS + pushl_cfi %eax lss (%esp), %esp /* switch to the normal stack segment */ CFI_ADJUST_CFA_OFFSET -8 .endm @@ -835,8 +801,7 @@ vector=FIRST_EXTERNAL_VECTOR .if vector <> FIRST_EXTERNAL_VECTOR CFI_ADJUST_CFA_OFFSET -4 .endif -1: pushl $(~vector+0x80) /* Note: always in signed byte range */ - CFI_ADJUST_CFA_OFFSET 4 +1: pushl_cfi $(~vector+0x80) /* Note: always in signed byte range */ .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6 jmp 2f .endif @@ -876,8 +841,7 @@ ENDPROC(common_interrupt) #define BUILD_INTERRUPT3(name, nr, fn) \ ENTRY(name) \ RING0_INT_FRAME; \ - pushl $~(nr); \ - CFI_ADJUST_CFA_OFFSET 4; \ + pushl_cfi $~(nr); \ SAVE_ALL; \ TRACE_IRQS_OFF \ movl %esp,%eax; \ @@ -893,21 +857,18 @@ ENDPROC(name) ENTRY(coprocessor_error) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_coprocessor_error - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_coprocessor_error jmp error_code CFI_ENDPROC END(coprocessor_error) ENTRY(simd_coprocessor_error) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 #ifdef CONFIG_X86_INVD_BUG /* AMD 486 bug: invd from userspace calls exception 19 instead of #GP */ -661: pushl $do_general_protection +661: pushl_cfi $do_general_protection 662: .section .altinstructions,"a" .balign 4 @@ -922,19 +883,16 @@ ENTRY(simd_coprocessor_error) 664: .previous #else - pushl $do_simd_coprocessor_error + pushl_cfi $do_simd_coprocessor_error #endif - CFI_ADJUST_CFA_OFFSET 4 jmp error_code CFI_ENDPROC END(simd_coprocessor_error) ENTRY(device_not_available) RING0_INT_FRAME - pushl $-1 # mark this as an int - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_device_not_available - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $-1 # mark this as an int + pushl_cfi $do_device_not_available jmp error_code CFI_ENDPROC END(device_not_available) @@ -956,82 +914,68 @@ END(native_irq_enable_sysexit) ENTRY(overflow) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_overflow - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_overflow jmp error_code CFI_ENDPROC END(overflow) ENTRY(bounds) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_bounds - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_bounds jmp error_code CFI_ENDPROC END(bounds) ENTRY(invalid_op) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_invalid_op - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_invalid_op jmp error_code CFI_ENDPROC END(invalid_op) ENTRY(coprocessor_segment_overrun) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_coprocessor_segment_overrun - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_coprocessor_segment_overrun jmp error_code CFI_ENDPROC END(coprocessor_segment_overrun) ENTRY(invalid_TSS) RING0_EC_FRAME - pushl $do_invalid_TSS - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_invalid_TSS jmp error_code CFI_ENDPROC END(invalid_TSS) ENTRY(segment_not_present) RING0_EC_FRAME - pushl $do_segment_not_present - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_segment_not_present jmp error_code CFI_ENDPROC END(segment_not_present) ENTRY(stack_segment) RING0_EC_FRAME - pushl $do_stack_segment - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_stack_segment jmp error_code CFI_ENDPROC END(stack_segment) ENTRY(alignment_check) RING0_EC_FRAME - pushl $do_alignment_check - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_alignment_check jmp error_code CFI_ENDPROC END(alignment_check) ENTRY(divide_error) RING0_INT_FRAME - pushl $0 # no error code - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_divide_error - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 # no error code + pushl_cfi $do_divide_error jmp error_code CFI_ENDPROC END(divide_error) @@ -1039,10 +983,8 @@ END(divide_error) #ifdef CONFIG_X86_MCE ENTRY(machine_check) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl machine_check_vector - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi machine_check_vector jmp error_code CFI_ENDPROC END(machine_check) @@ -1050,10 +992,8 @@ END(machine_check) ENTRY(spurious_interrupt_bug) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_spurious_interrupt_bug - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_spurious_interrupt_bug jmp error_code CFI_ENDPROC END(spurious_interrupt_bug) @@ -1084,8 +1024,7 @@ ENTRY(xen_sysenter_target) ENTRY(xen_hypervisor_callback) CFI_STARTPROC - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 SAVE_ALL TRACE_IRQS_OFF @@ -1121,23 +1060,20 @@ ENDPROC(xen_hypervisor_callback) # We distinguish between categories by maintaining a status value in EAX. ENTRY(xen_failsafe_callback) CFI_STARTPROC - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax movl $1,%eax 1: mov 4(%esp),%ds 2: mov 8(%esp),%es 3: mov 12(%esp),%fs 4: mov 16(%esp),%gs testl %eax,%eax - popl %eax - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %eax lea 16(%esp),%esp CFI_ADJUST_CFA_OFFSET -16 jz 5f addl $16,%esp jmp iret_exc # EAX != 0 => Category 2 (Bad IRET) -5: pushl $0 # EAX == 0 => Category 1 (Bad segment) - CFI_ADJUST_CFA_OFFSET 4 +5: pushl_cfi $0 # EAX == 0 => Category 1 (Bad segment) SAVE_ALL jmp ret_from_exception CFI_ENDPROC @@ -1287,40 +1223,29 @@ syscall_table_size=(.-sys_call_table) ENTRY(page_fault) RING0_EC_FRAME - pushl $do_page_fault - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_page_fault ALIGN error_code: /* the function address is in %gs's slot on the stack */ - pushl %fs - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %fs /*CFI_REL_OFFSET fs, 0*/ - pushl %es - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %es /*CFI_REL_OFFSET es, 0*/ - pushl %ds - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ds /*CFI_REL_OFFSET ds, 0*/ - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax CFI_REL_OFFSET eax, 0 - pushl %ebp - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebp CFI_REL_OFFSET ebp, 0 - pushl %edi - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %edi CFI_REL_OFFSET edi, 0 - pushl %esi - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %esi CFI_REL_OFFSET esi, 0 - pushl %edx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %edx CFI_REL_OFFSET edx, 0 - pushl %ecx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ecx CFI_REL_OFFSET ecx, 0 - pushl %ebx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebx CFI_REL_OFFSET ebx, 0 cld movl $(__KERNEL_PERCPU), %ecx @@ -1362,12 +1287,9 @@ END(page_fault) 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 + pushfl_cfi + pushl_cfi $__KERNEL_CS + pushl_cfi $sysenter_past_esp CFI_REL_OFFSET eip, 0 .endm @@ -1377,8 +1299,7 @@ ENTRY(debug) 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 + pushl_cfi $-1 # mark this as an int SAVE_ALL TRACE_IRQS_OFF xorl %edx,%edx # error code 0 @@ -1398,32 +1319,27 @@ END(debug) */ ENTRY(nmi) RING0_INT_FRAME - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax movl %ss, %eax cmpw $__ESPFIX_SS, %ax - popl %eax - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %eax je nmi_espfix_stack cmpl $ia32_sysenter_target,(%esp) je nmi_stack_fixup - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax 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 + popl_cfi %eax 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 + pushl_cfi %eax SAVE_ALL xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer @@ -1452,18 +1368,14 @@ nmi_espfix_stack: * * create the pointer to lss back */ - pushl %ss - CFI_ADJUST_CFA_OFFSET 4 - pushl %esp - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ss + pushl_cfi %esp addl $4, (%esp) /* copy the iret frame of 12 bytes */ .rept 3 - pushl 16(%esp) - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi 16(%esp) .endr - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax SAVE_ALL FIXUP_ESPFIX_STACK # %eax == %esp xorl %edx,%edx # zero error code @@ -1477,8 +1389,7 @@ END(nmi) ENTRY(int3) RING0_INT_FRAME - pushl $-1 # mark this as an int - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $-1 # mark this as an int SAVE_ALL TRACE_IRQS_OFF xorl %edx,%edx # zero error code @@ -1490,8 +1401,7 @@ END(int3) ENTRY(general_protection) RING0_EC_FRAME - pushl $do_general_protection - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_general_protection jmp error_code CFI_ENDPROC END(general_protection) diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index c5ea5cdbe7b..fe2690d71c0 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -213,23 +213,17 @@ ENDPROC(native_usergs_sysret64) .macro FAKE_STACK_FRAME child_rip /* push in order ss, rsp, eflags, cs, rip */ xorl %eax, %eax - pushq $__KERNEL_DS /* ss */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $__KERNEL_DS /* ss */ /*CFI_REL_OFFSET ss,0*/ - pushq %rax /* rsp */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rax /* rsp */ CFI_REL_OFFSET rsp,0 - pushq $X86_EFLAGS_IF /* eflags - interrupts on */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $X86_EFLAGS_IF /* eflags - interrupts on */ /*CFI_REL_OFFSET rflags,0*/ - pushq $__KERNEL_CS /* cs */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $__KERNEL_CS /* cs */ /*CFI_REL_OFFSET cs,0*/ - pushq \child_rip /* rip */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi \child_rip /* rip */ CFI_REL_OFFSET rip,0 - pushq %rax /* orig rax */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rax /* orig rax */ .endm .macro UNFAKE_STACK_FRAME @@ -398,10 +392,8 @@ ENTRY(ret_from_fork) LOCK ; btr $TIF_FORK,TI_flags(%r8) - push kernel_eflags(%rip) - CFI_ADJUST_CFA_OFFSET 8 - popf # reset kernel eflags - CFI_ADJUST_CFA_OFFSET -8 + pushq_cfi kernel_eflags(%rip) + popfq_cfi # reset kernel eflags call schedule_tail # rdi: 'prev' task parameter @@ -521,11 +513,9 @@ sysret_careful: jnc sysret_signal TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) - pushq %rdi - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rdi call schedule - popq %rdi - CFI_ADJUST_CFA_OFFSET -8 + popq_cfi %rdi jmp sysret_check /* Handle a signal */ @@ -634,11 +624,9 @@ int_careful: jnc int_very_careful TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) - pushq %rdi - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rdi call schedule - popq %rdi - CFI_ADJUST_CFA_OFFSET -8 + popq_cfi %rdi DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF jmp int_with_check @@ -652,12 +640,10 @@ int_check_syscall_exit_work: /* Check for syscall exit trace */ testl $_TIF_WORK_SYSCALL_EXIT,%edx jz int_signal - pushq %rdi - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rdi leaq 8(%rsp),%rdi # &ptregs -> arg1 call syscall_trace_leave - popq %rdi - CFI_ADJUST_CFA_OFFSET -8 + popq_cfi %rdi andl $~(_TIF_WORK_SYSCALL_EXIT|_TIF_SYSCALL_EMU),%edi jmp int_restore_rest @@ -714,9 +700,8 @@ END(ptregscall_common) ENTRY(stub_execve) CFI_STARTPROC - popq %r11 - CFI_ADJUST_CFA_OFFSET -8 - CFI_REGISTER rip, r11 + addq $8, %rsp + PARTIAL_FRAME 0 SAVE_REST FIXUP_TOP_OF_STACK %r11 movq %rsp, %rcx @@ -735,7 +720,7 @@ END(stub_execve) ENTRY(stub_rt_sigreturn) CFI_STARTPROC addq $8, %rsp - CFI_ADJUST_CFA_OFFSET -8 + PARTIAL_FRAME 0 SAVE_REST movq %rsp,%rdi FIXUP_TOP_OF_STACK %r11 @@ -766,8 +751,7 @@ vector=FIRST_EXTERNAL_VECTOR .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 +1: pushq_cfi $(~vector+0x80) /* Note: always in signed byte range */ .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6 jmp 2f .endif @@ -796,8 +780,8 @@ END(interrupt) /* 0(%rsp): ~(interrupt number) */ .macro interrupt func - subq $10*8, %rsp - CFI_ADJUST_CFA_OFFSET 10*8 + subq $ORIG_RAX-ARGOFFSET+8, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-ARGOFFSET+8 call save_args PARTIAL_FRAME 0 call \func @@ -822,6 +806,7 @@ ret_from_intr: TRACE_IRQS_OFF decl PER_CPU_VAR(irq_count) leaveq + CFI_RESTORE rbp CFI_DEF_CFA_REGISTER rsp CFI_ADJUST_CFA_OFFSET -8 exit_intr: @@ -903,11 +888,9 @@ retint_careful: jnc retint_signal TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) - pushq %rdi - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rdi call schedule - popq %rdi - CFI_ADJUST_CFA_OFFSET -8 + popq_cfi %rdi GET_THREAD_INFO(%rcx) DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF @@ -956,8 +939,7 @@ END(common_interrupt) .macro apicinterrupt num sym do_sym ENTRY(\sym) INTR_FRAME - pushq $~(\num) - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $~(\num) interrupt \do_sym jmp ret_from_intr CFI_ENDPROC @@ -981,22 +963,10 @@ apicinterrupt X86_PLATFORM_IPI_VECTOR \ x86_platform_ipi smp_x86_platform_ipi #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 +.irpc idx, "01234567" +apicinterrupt (INVALIDATE_TLB_VECTOR_START)+\idx \ + invalidate_interrupt\idx smp_invalidate_interrupt +.endr #endif apicinterrupt THRESHOLD_APIC_VECTOR \ @@ -1023,9 +993,9 @@ apicinterrupt ERROR_APIC_VECTOR \ apicinterrupt SPURIOUS_APIC_VECTOR \ spurious_interrupt smp_spurious_interrupt -#ifdef CONFIG_PERF_EVENTS -apicinterrupt LOCAL_PENDING_VECTOR \ - perf_pending_interrupt smp_perf_pending_interrupt +#ifdef CONFIG_IRQ_WORK +apicinterrupt IRQ_WORK_VECTOR \ + irq_work_interrupt smp_irq_work_interrupt #endif /* @@ -1036,8 +1006,8 @@ ENTRY(\sym) INTR_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME pushq_cfi $-1 /* ORIG_RAX: no syscall to restart */ - subq $15*8,%rsp - CFI_ADJUST_CFA_OFFSET 15*8 + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call error_entry DEFAULT_FRAME 0 movq %rsp,%rdi /* pt_regs pointer */ @@ -1052,9 +1022,9 @@ END(\sym) 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 + pushq_cfi $-1 /* ORIG_RAX: no syscall to restart */ + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call save_paranoid TRACE_IRQS_OFF movq %rsp,%rdi /* pt_regs pointer */ @@ -1070,9 +1040,9 @@ END(\sym) 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 + pushq_cfi $-1 /* ORIG_RAX: no syscall to restart */ + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call save_paranoid TRACE_IRQS_OFF movq %rsp,%rdi /* pt_regs pointer */ @@ -1089,8 +1059,8 @@ END(\sym) ENTRY(\sym) XCPT_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME - subq $15*8,%rsp - CFI_ADJUST_CFA_OFFSET 15*8 + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call error_entry DEFAULT_FRAME 0 movq %rsp,%rdi /* pt_regs pointer */ @@ -1107,8 +1077,8 @@ END(\sym) ENTRY(\sym) XCPT_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME - subq $15*8,%rsp - CFI_ADJUST_CFA_OFFSET 15*8 + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call save_paranoid DEFAULT_FRAME 0 TRACE_IRQS_OFF @@ -1139,16 +1109,14 @@ zeroentry simd_coprocessor_error do_simd_coprocessor_error /* edi: new selector */ ENTRY(native_load_gs_index) CFI_STARTPROC - pushf - CFI_ADJUST_CFA_OFFSET 8 + pushfq_cfi DISABLE_INTERRUPTS(CLBR_ANY & ~CLBR_RDI) SWAPGS gs_change: movl %edi,%gs 2: mfence /* workaround */ SWAPGS - popf - CFI_ADJUST_CFA_OFFSET -8 + popfq_cfi ret CFI_ENDPROC END(native_load_gs_index) @@ -1185,13 +1153,13 @@ END(kernel_thread_helper) * execve(). This function needs to use IRET, not SYSRET, to set up all state properly. * * C extern interface: - * extern long execve(char *name, char **argv, char **envp) + * extern long execve(const char *name, char **argv, char **envp) * * asm input arguments: * rdi: name, rsi: argv, rdx: envp * * We want to fallback into: - * extern long sys_execve(char *name, char **argv,char **envp, struct pt_regs *regs) + * extern long sys_execve(const char *name, char **argv,char **envp, struct pt_regs *regs) * * do_sys_execve asm fallback arguments: * rdi: name, rsi: argv, rdx: envp, rcx: fake frame on the stack @@ -1215,8 +1183,7 @@ END(kernel_execve) /* Call softirq on interrupt stack. Interrupts are off. */ ENTRY(call_softirq) CFI_STARTPROC - push %rbp - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rbp CFI_REL_OFFSET rbp,0 mov %rsp,%rbp CFI_DEF_CFA_REGISTER rbp @@ -1225,6 +1192,7 @@ ENTRY(call_softirq) push %rbp # backlink for old unwinder call __do_softirq leaveq + CFI_RESTORE rbp CFI_DEF_CFA_REGISTER rsp CFI_ADJUST_CFA_OFFSET -8 decl PER_CPU_VAR(irq_count) @@ -1368,7 +1336,7 @@ paranoidzeroentry machine_check *machine_check_vector(%rip) /* ebx: no swapgs flag */ ENTRY(paranoid_exit) - INTR_FRAME + DEFAULT_FRAME DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF testl %ebx,%ebx /* swapgs needed? */ @@ -1445,7 +1413,6 @@ error_swapgs: error_sti: TRACE_IRQS_OFF ret - CFI_ENDPROC /* * There are two places in the kernel that can potentially fault with @@ -1470,6 +1437,7 @@ bstep_iret: /* Fix truncated RIP */ movq %rcx,RIP+8(%rsp) jmp error_swapgs + CFI_ENDPROC END(error_entry) @@ -1498,8 +1466,8 @@ ENTRY(nmi) INTR_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME pushq_cfi $-1 - subq $15*8, %rsp - CFI_ADJUST_CFA_OFFSET 15*8 + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call save_paranoid DEFAULT_FRAME 0 /* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */ diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index cd37469b54e..3afb33f14d2 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -257,14 +257,9 @@ do_ftrace_mod_code(unsigned long ip, void *new_code) return mod_code_status; } - - - -static unsigned char ftrace_nop[MCOUNT_INSN_SIZE]; - static unsigned char *ftrace_nop_replace(void) { - return ftrace_nop; + return ideal_nop5; } static int @@ -338,62 +333,6 @@ int ftrace_update_ftrace_func(ftrace_func_t func) int __init ftrace_dyn_arch_init(void *data) { - extern const unsigned char ftrace_test_p6nop[]; - extern const unsigned char ftrace_test_nop5[]; - extern const unsigned char ftrace_test_jmp[]; - int faulted = 0; - - /* - * 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. - */ - 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("converting mcount calls to 0f 1f 44 00 00\n"); - memcpy(ftrace_nop, ftrace_test_p6nop, MCOUNT_INSN_SIZE); - break; - case 1: - pr_info("converting mcount calls to 66 66 66 66 90\n"); - memcpy(ftrace_nop, ftrace_test_nop5, MCOUNT_INSN_SIZE); - break; - case 2: - pr_info("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; diff --git a/arch/x86/kernel/head.c b/arch/x86/kernel/head.c index 3e66bd364a9..af0699ba48c 100644 --- a/arch/x86/kernel/head.c +++ b/arch/x86/kernel/head.c @@ -1,5 +1,6 @@ #include <linux/kernel.h> #include <linux/init.h> +#include <linux/memblock.h> #include <asm/setup.h> #include <asm/bios_ebda.h> @@ -51,5 +52,5 @@ void __init reserve_ebda_region(void) lowmem = 0x9f000; /* reserve all memory between lowmem and the 1MB mark */ - reserve_early_overlap_ok(lowmem, 0x100000, "BIOS reserved"); + memblock_x86_reserve_range(lowmem, 0x100000, "* BIOS reserved"); } diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index 784360c0625..763310165fa 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -8,6 +8,7 @@ #include <linux/init.h> #include <linux/start_kernel.h> #include <linux/mm.h> +#include <linux/memblock.h> #include <asm/setup.h> #include <asm/sections.h> @@ -17,6 +18,7 @@ #include <asm/apic.h> #include <asm/io_apic.h> #include <asm/bios_ebda.h> +#include <asm/tlbflush.h> static void __init i386_default_early_setup(void) { @@ -30,17 +32,18 @@ static void __init i386_default_early_setup(void) void __init i386_start_kernel(void) { + memblock_init(); + #ifdef CONFIG_X86_TRAMPOLINE /* * 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_overlap_ok(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, - "EX TRAMPOLINE"); + memblock_x86_reserve_range(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE"); #endif - reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); + memblock_x86_reserve_range(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ @@ -49,7 +52,7 @@ void __init i386_start_kernel(void) u64 ramdisk_image = boot_params.hdr.ramdisk_image; u64 ramdisk_size = boot_params.hdr.ramdisk_size; u64 ramdisk_end = PAGE_ALIGN(ramdisk_image + ramdisk_size); - reserve_early(ramdisk_image, ramdisk_end, "RAMDISK"); + memblock_x86_reserve_range(ramdisk_image, ramdisk_end, "RAMDISK"); } #endif diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 7147143fd61..2d2673c28af 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -12,6 +12,7 @@ #include <linux/percpu.h> #include <linux/start_kernel.h> #include <linux/io.h> +#include <linux/memblock.h> #include <asm/processor.h> #include <asm/proto.h> @@ -79,6 +80,8 @@ void __init x86_64_start_kernel(char * real_mode_data) /* Cleanup the over mapped high alias */ cleanup_highmap(); + max_pfn_mapped = KERNEL_IMAGE_SIZE >> PAGE_SHIFT; + for (i = 0; i < NUM_EXCEPTION_VECTORS; i++) { #ifdef CONFIG_EARLY_PRINTK set_intr_gate(i, &early_idt_handlers[i]); @@ -98,7 +101,9 @@ void __init x86_64_start_reservations(char *real_mode_data) { copy_bootdata(__va(real_mode_data)); - reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); + memblock_init(); + + memblock_x86_reserve_range(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ @@ -107,7 +112,7 @@ void __init x86_64_start_reservations(char *real_mode_data) unsigned long ramdisk_image = boot_params.hdr.ramdisk_image; unsigned long ramdisk_size = boot_params.hdr.ramdisk_size; unsigned long ramdisk_end = PAGE_ALIGN(ramdisk_image + ramdisk_size); - reserve_early(ramdisk_image, ramdisk_end, "RAMDISK"); + memblock_x86_reserve_range(ramdisk_image, ramdisk_end, "RAMDISK"); } #endif diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index ff4c453e13f..bcece91dd31 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -183,13 +183,12 @@ default_entry: #ifdef CONFIG_X86_PAE /* - * In PAE mode swapper_pg_dir is statically defined to contain enough - * entries to cover the VMSPLIT option (that is the top 1, 2 or 3 - * entries). The identity mapping is handled by pointing two PGD - * entries to the first kernel PMD. + * In PAE mode initial_page_table is statically defined to contain + * enough entries to cover the VMSPLIT option (that is the top 1, 2 or 3 + * entries). The identity mapping is handled by pointing two PGD entries + * to the first kernel PMD. * - * Note the upper half of each PMD or PTE are always zero at - * this stage. + * Note the upper half of each PMD or PTE are always zero at this stage. */ #define KPMDS (((-__PAGE_OFFSET) >> 30) & 3) /* Number of kernel PMDs */ @@ -197,7 +196,7 @@ default_entry: xorl %ebx,%ebx /* %ebx is kept at zero */ movl $pa(__brk_base), %edi - movl $pa(swapper_pg_pmd), %edx + movl $pa(initial_pg_pmd), %edx movl $PTE_IDENT_ATTR, %eax 10: leal PDE_IDENT_ATTR(%edi),%ecx /* Create PMD entry */ @@ -226,14 +225,14 @@ default_entry: movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax - movl %eax,pa(swapper_pg_pmd+0x1000*KPMDS-8) + movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax + movl %eax,pa(initial_pg_pmd+0x1000*KPMDS-8) #else /* Not PAE */ page_pde_offset = (__PAGE_OFFSET >> 20); movl $pa(__brk_base), %edi - movl $pa(swapper_pg_dir), %edx + movl $pa(initial_page_table), %edx movl $PTE_IDENT_ATTR, %eax 10: leal PDE_IDENT_ATTR(%edi),%ecx /* Create PDE entry */ @@ -257,8 +256,8 @@ page_pde_offset = (__PAGE_OFFSET >> 20); movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax - movl %eax,pa(swapper_pg_dir+0xffc) + movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax + movl %eax,pa(initial_page_table+0xffc) #endif jmp 3f /* @@ -334,7 +333,7 @@ ENTRY(startup_32_smp) /* * Enable paging */ - movl $pa(swapper_pg_dir),%eax + movl $pa(initial_page_table), %eax movl %eax,%cr3 /* set the page table pointer.. */ movl %cr0,%eax orl $X86_CR0_PG,%eax @@ -621,16 +620,18 @@ ENTRY(initial_code) __PAGE_ALIGNED_BSS .align PAGE_SIZE_asm #ifdef CONFIG_X86_PAE -swapper_pg_pmd: +initial_pg_pmd: .fill 1024*KPMDS,4,0 #else -ENTRY(swapper_pg_dir) +ENTRY(initial_page_table) .fill 1024,4,0 #endif -swapper_pg_fixmap: +initial_pg_fixmap: .fill 1024,4,0 ENTRY(empty_zero_page) .fill 4096,1,0 +ENTRY(swapper_pg_dir) + .fill 1024,4,0 /* * This starts the data section. @@ -639,20 +640,20 @@ ENTRY(empty_zero_page) __PAGE_ALIGNED_DATA /* Page-aligned for the benefit of paravirt? */ .align PAGE_SIZE_asm -ENTRY(swapper_pg_dir) - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 /* low identity map */ +ENTRY(initial_page_table) + .long pa(initial_pg_pmd+PGD_IDENT_ATTR),0 /* low identity map */ # if KPMDS == 3 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x2000),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR+0x2000),0 # elif KPMDS == 2 .long 0,0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0 # elif KPMDS == 1 .long 0,0 .long 0,0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR),0 # else # error "Kernel PMDs should be 1, 2 or 3" # endif diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 351f9c0fea1..aff0b3c2750 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -35,7 +35,6 @@ unsigned long hpet_address; u8 hpet_blockid; /* OS timer block num */ u8 hpet_msi_disable; -u8 hpet_readback_cmp; #ifdef CONFIG_PCI_MSI static unsigned long hpet_num_timers; @@ -381,40 +380,35 @@ static int hpet_next_event(unsigned long delta, struct clock_event_device *evt, int timer) { u32 cnt; + s32 res; cnt = hpet_readl(HPET_COUNTER); cnt += (u32) delta; hpet_writel(cnt, HPET_Tn_CMP(timer)); /* - * We need to read back the CMP register on certain HPET - * implementations (ATI chipsets) which seem to delay the - * transfer of the compare register into the internal compare - * logic. With small deltas this might actually be too late as - * the counter could already be higher than the compare value - * at that point and we would wait for the next hpet interrupt - * forever. We found out that reading the CMP register back - * forces the transfer so we can rely on the comparison with - * the counter register below. - * - * That works fine on those ATI chipsets, but on newer Intel - * chipsets (ICH9...) this triggers due to an erratum: Reading - * the comparator immediately following a write is returning - * the old value. - * - * We restrict the read back to the affected ATI chipsets (set - * by quirks) and also run it with hpet=verbose for debugging - * purposes. + * HPETs are a complete disaster. The compare register is + * based on a equal comparison and neither provides a less + * than or equal functionality (which would require to take + * the wraparound into account) nor a simple count down event + * mode. Further the write to the comparator register is + * delayed internally up to two HPET clock cycles in certain + * chipsets (ATI, ICH9,10). We worked around that by reading + * back the compare register, but that required another + * workaround for ICH9,10 chips where the first readout after + * write can return the old stale value. We already have a + * minimum delta of 5us enforced, but a NMI or SMI hitting + * between the counter readout and the comparator write can + * move us behind that point easily. Now instead of reading + * the compare register back several times, we make the ETIME + * decision based on the following: Return ETIME if the + * counter value after the write is less than 8 HPET cycles + * away from the event or if the counter is already ahead of + * the event. */ - if (hpet_readback_cmp || hpet_verbose) { - u32 cmp = hpet_readl(HPET_Tn_CMP(timer)); - - if (cmp != cnt) - printk_once(KERN_WARNING - "hpet: compare register read back failed.\n"); - } + res = (s32)(cnt - hpet_readl(HPET_COUNTER)); - return (s32)(hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0; + return res < 8 ? -ETIME : 0; } static void hpet_legacy_set_mode(enum clock_event_mode mode, @@ -437,9 +431,9 @@ static int hpet_legacy_next_event(unsigned long delta, static DEFINE_PER_CPU(struct hpet_dev *, cpu_hpet_dev); static struct hpet_dev *hpet_devs; -void hpet_msi_unmask(unsigned int irq) +void hpet_msi_unmask(struct irq_data *data) { - struct hpet_dev *hdev = get_irq_data(irq); + struct hpet_dev *hdev = data->handler_data; unsigned int cfg; /* unmask it */ @@ -448,10 +442,10 @@ void hpet_msi_unmask(unsigned int irq) hpet_writel(cfg, HPET_Tn_CFG(hdev->num)); } -void hpet_msi_mask(unsigned int irq) +void hpet_msi_mask(struct irq_data *data) { + struct hpet_dev *hdev = data->handler_data; unsigned int cfg; - struct hpet_dev *hdev = get_irq_data(irq); /* mask it */ cfg = hpet_readl(HPET_Tn_CFG(hdev->num)); @@ -459,18 +453,14 @@ void hpet_msi_mask(unsigned int irq) hpet_writel(cfg, HPET_Tn_CFG(hdev->num)); } -void hpet_msi_write(unsigned int irq, struct msi_msg *msg) +void hpet_msi_write(struct hpet_dev *hdev, 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) +void hpet_msi_read(struct hpet_dev *hdev, 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; @@ -503,7 +493,7 @@ static int hpet_assign_irq(struct hpet_dev *dev) { unsigned int irq; - irq = create_irq(); + irq = create_irq_nr(0, -1); if (!irq) return -EINVAL; diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c index a474ec37c32..ff15c9dcc25 100644 --- a/arch/x86/kernel/hw_breakpoint.c +++ b/arch/x86/kernel/hw_breakpoint.c @@ -206,11 +206,27 @@ int arch_check_bp_in_kernelspace(struct perf_event *bp) int arch_bp_generic_fields(int x86_len, int x86_type, int *gen_len, int *gen_type) { - /* Len */ - switch (x86_len) { - case X86_BREAKPOINT_LEN_X: + /* Type */ + switch (x86_type) { + case X86_BREAKPOINT_EXECUTE: + if (x86_len != X86_BREAKPOINT_LEN_X) + return -EINVAL; + + *gen_type = HW_BREAKPOINT_X; *gen_len = sizeof(long); + return 0; + case X86_BREAKPOINT_WRITE: + *gen_type = HW_BREAKPOINT_W; break; + case X86_BREAKPOINT_RW: + *gen_type = HW_BREAKPOINT_W | HW_BREAKPOINT_R; + break; + default: + return -EINVAL; + } + + /* Len */ + switch (x86_len) { case X86_BREAKPOINT_LEN_1: *gen_len = HW_BREAKPOINT_LEN_1; break; @@ -229,21 +245,6 @@ int arch_bp_generic_fields(int x86_len, int x86_type, return -EINVAL; } - /* Type */ - switch (x86_type) { - case X86_BREAKPOINT_EXECUTE: - *gen_type = HW_BREAKPOINT_X; - break; - case X86_BREAKPOINT_WRITE: - *gen_type = HW_BREAKPOINT_W; - break; - case X86_BREAKPOINT_RW: - *gen_type = HW_BREAKPOINT_W | HW_BREAKPOINT_R; - break; - default: - return -EINVAL; - } - return 0; } @@ -316,9 +317,6 @@ int arch_validate_hwbkpt_settings(struct perf_event *bp) ret = -EINVAL; switch (info->len) { - case X86_BREAKPOINT_LEN_X: - align = sizeof(long) -1; - break; case X86_BREAKPOINT_LEN_1: align = 0; break; diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 1f11f5ce668..58bb239a2fd 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -40,6 +40,7 @@ static unsigned int mxcsr_feature_mask __read_mostly = 0xffffffffu; unsigned int xstate_size; +EXPORT_SYMBOL_GPL(xstate_size); unsigned int sig_xstate_ia32_size = sizeof(struct _fpstate_ia32); static struct i387_fxsave_struct fx_scratch __cpuinitdata; @@ -67,19 +68,22 @@ static void __cpuinit init_thread_xstate(void) */ if (!HAVE_HWFP) { + /* + * Disable xsave as we do not support it if i387 + * emulation is enabled. + */ + setup_clear_cpu_cap(X86_FEATURE_XSAVE); + setup_clear_cpu_cap(X86_FEATURE_XSAVEOPT); xstate_size = sizeof(struct i387_soft_struct); return; } if (cpu_has_fxsr) xstate_size = sizeof(struct i387_fxsave_struct); -#ifdef CONFIG_X86_32 else xstate_size = sizeof(struct i387_fsave_struct); -#endif } -#ifdef CONFIG_X86_64 /* * Called at bootup to set up the initial FPU state that is later cloned * into all processes. @@ -87,12 +91,21 @@ static void __cpuinit init_thread_xstate(void) void __cpuinit fpu_init(void) { - unsigned long oldcr0 = read_cr0(); - - set_in_cr4(X86_CR4_OSFXSR); - set_in_cr4(X86_CR4_OSXMMEXCPT); + unsigned long cr0; + unsigned long cr4_mask = 0; - write_cr0(oldcr0 & ~(X86_CR0_TS|X86_CR0_EM)); /* clear TS and EM */ + if (cpu_has_fxsr) + cr4_mask |= X86_CR4_OSFXSR; + if (cpu_has_xmm) + cr4_mask |= X86_CR4_OSXMMEXCPT; + if (cr4_mask) + set_in_cr4(cr4_mask); + + cr0 = read_cr0(); + cr0 &= ~(X86_CR0_TS|X86_CR0_EM); /* clear TS and EM */ + if (!HAVE_HWFP) + cr0 |= X86_CR0_EM; + write_cr0(cr0); if (!smp_processor_id()) init_thread_xstate(); @@ -103,24 +116,12 @@ void __cpuinit fpu_init(void) clear_used_math(); } -#else /* CONFIG_X86_64 */ - -void __cpuinit fpu_init(void) -{ - if (!smp_processor_id()) - init_thread_xstate(); -} - -#endif /* CONFIG_X86_32 */ - void fpu_finit(struct fpu *fpu) { -#ifdef CONFIG_X86_32 if (!HAVE_HWFP) { finit_soft_fpu(&fpu->state->soft); return; } -#endif if (cpu_has_fxsr) { struct i387_fxsave_struct *fx = &fpu->state->fxsave; @@ -385,19 +386,17 @@ convert_from_fxsr(struct user_i387_ia32_struct *env, struct task_struct *tsk) #ifdef CONFIG_X86_64 env->fip = fxsave->rip; env->foo = fxsave->rdp; + /* + * should be actually ds/cs at fpu exception time, but + * that information is not available in 64bit mode. + */ + env->fcs = task_pt_regs(tsk)->cs; if (tsk == current) { - /* - * should be actually ds/cs at fpu exception time, but - * that information is not available in 64bit mode. - */ - asm("mov %%ds, %[fos]" : [fos] "=r" (env->fos)); - asm("mov %%cs, %[fcs]" : [fcs] "=r" (env->fcs)); + savesegment(ds, env->fos); } else { - struct pt_regs *regs = task_pt_regs(tsk); - - env->fos = 0xffff0000 | tsk->thread.ds; - env->fcs = regs->cs; + env->fos = tsk->thread.ds; } + env->fos |= 0xffff0000; #else env->fip = fxsave->fip; env->fcs = (u16) fxsave->fcs | ((u32) fxsave->fop << 16); diff --git a/arch/x86/kernel/i8259.c b/arch/x86/kernel/i8259.c index cafa7c80ac9..20757cb2efa 100644 --- a/arch/x86/kernel/i8259.c +++ b/arch/x86/kernel/i8259.c @@ -29,24 +29,10 @@ * plus some generic x86 specific things if generic specifics makes * any sense at all. */ +static void init_8259A(int auto_eoi); static int i8259A_auto_eoi; DEFINE_RAW_SPINLOCK(i8259A_lock); -static void mask_and_ack_8259A(unsigned int); -static void mask_8259A(void); -static void unmask_8259A(void); -static void disable_8259A_irq(unsigned int irq); -static void enable_8259A_irq(unsigned int irq); -static void init_8259A(int auto_eoi); -static int i8259A_irq_pending(unsigned int irq); - -struct irq_chip i8259A_chip = { - .name = "XT-PIC", - .mask = disable_8259A_irq, - .disable = disable_8259A_irq, - .unmask = enable_8259A_irq, - .mask_ack = mask_and_ack_8259A, -}; /* * 8259A PIC functions to handle ISA devices: @@ -68,7 +54,7 @@ unsigned int cached_irq_mask = 0xffff; */ unsigned long io_apic_irqs; -static void disable_8259A_irq(unsigned int irq) +static void mask_8259A_irq(unsigned int irq) { unsigned int mask = 1 << irq; unsigned long flags; @@ -82,7 +68,12 @@ static void disable_8259A_irq(unsigned int irq) raw_spin_unlock_irqrestore(&i8259A_lock, flags); } -static void enable_8259A_irq(unsigned int irq) +static void disable_8259A_irq(struct irq_data *data) +{ + mask_8259A_irq(data->irq); +} + +static void unmask_8259A_irq(unsigned int irq) { unsigned int mask = ~(1 << irq); unsigned long flags; @@ -96,6 +87,11 @@ static void enable_8259A_irq(unsigned int irq) raw_spin_unlock_irqrestore(&i8259A_lock, flags); } +static void enable_8259A_irq(struct irq_data *data) +{ + unmask_8259A_irq(data->irq); +} + static int i8259A_irq_pending(unsigned int irq) { unsigned int mask = 1<<irq; @@ -117,7 +113,7 @@ static void make_8259A_irq(unsigned int irq) disable_irq_nosync(irq); io_apic_irqs &= ~(1<<irq); set_irq_chip_and_handler_name(irq, &i8259A_chip, handle_level_irq, - "XT"); + i8259A_chip.name); enable_irq(irq); } @@ -150,8 +146,9 @@ static inline int i8259A_irq_real(unsigned int irq) * first, _then_ send the EOI, and the order of EOI * to the two 8259s is important! */ -static void mask_and_ack_8259A(unsigned int irq) +static void mask_and_ack_8259A(struct irq_data *data) { + unsigned int irq = data->irq; unsigned int irqmask = 1 << irq; unsigned long flags; @@ -223,6 +220,14 @@ spurious_8259A_irq: } } +struct irq_chip i8259A_chip = { + .name = "XT-PIC", + .irq_mask = disable_8259A_irq, + .irq_disable = disable_8259A_irq, + .irq_unmask = enable_8259A_irq, + .irq_mask_ack = mask_and_ack_8259A, +}; + static char irq_trigger[2]; /** * ELCR registers (0x4d0, 0x4d1) control edge/level of IRQ @@ -342,9 +347,9 @@ static void init_8259A(int auto_eoi) * In AEOI mode we just have to mask the interrupt * when acking. */ - i8259A_chip.mask_ack = disable_8259A_irq; + i8259A_chip.irq_mask_ack = disable_8259A_irq; else - i8259A_chip.mask_ack = mask_and_ack_8259A; + i8259A_chip.irq_mask_ack = mask_and_ack_8259A; udelay(100); /* wait for 8259A to initialize */ @@ -363,14 +368,6 @@ static void init_8259A(int auto_eoi) static void legacy_pic_noop(void) { }; static void legacy_pic_uint_noop(unsigned int unused) { }; static void legacy_pic_int_noop(int unused) { }; - -static struct irq_chip dummy_pic_chip = { - .name = "dummy pic", - .mask = legacy_pic_uint_noop, - .unmask = legacy_pic_uint_noop, - .disable = legacy_pic_uint_noop, - .mask_ack = legacy_pic_uint_noop, -}; static int legacy_pic_irq_pending_noop(unsigned int irq) { return 0; @@ -378,7 +375,9 @@ static int legacy_pic_irq_pending_noop(unsigned int irq) struct legacy_pic null_legacy_pic = { .nr_legacy_irqs = 0, - .chip = &dummy_pic_chip, + .chip = &dummy_irq_chip, + .mask = legacy_pic_uint_noop, + .unmask = legacy_pic_uint_noop, .mask_all = legacy_pic_noop, .restore_mask = legacy_pic_noop, .init = legacy_pic_int_noop, @@ -389,7 +388,9 @@ struct legacy_pic null_legacy_pic = { struct legacy_pic default_legacy_pic = { .nr_legacy_irqs = NR_IRQS_LEGACY, .chip = &i8259A_chip, - .mask_all = mask_8259A, + .mask = mask_8259A_irq, + .unmask = unmask_8259A_irq, + .mask_all = mask_8259A, .restore_mask = unmask_8259A, .init = init_8259A, .irq_pending = i8259A_irq_pending, diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 91fd0c70a18..83ec0175f98 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -67,10 +67,10 @@ static int show_other_interrupts(struct seq_file *p, int prec) for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->apic_perf_irqs); seq_printf(p, " Performance monitoring interrupts\n"); - seq_printf(p, "%*s: ", prec, "PND"); + seq_printf(p, "%*s: ", prec, "IWI"); for_each_online_cpu(j) - seq_printf(p, "%10u ", irq_stats(j)->apic_pending_irqs); - seq_printf(p, " Performance pending work\n"); + seq_printf(p, "%10u ", irq_stats(j)->apic_irq_work_irqs); + seq_printf(p, " IRQ work interrupts\n"); #endif if (x86_platform_ipi_callback) { seq_printf(p, "%*s: ", prec, "PLT"); @@ -159,7 +159,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_printf(p, "%*d: ", prec, i); for_each_online_cpu(j) seq_printf(p, "%10u ", kstat_irqs_cpu(i, j)); - seq_printf(p, " %8s", desc->chip->name); + seq_printf(p, " %8s", desc->irq_data.chip->name); seq_printf(p, "-%-8s", desc->name); if (action) { @@ -185,7 +185,7 @@ u64 arch_irq_stat_cpu(unsigned int cpu) sum += irq_stats(cpu)->apic_timer_irqs; sum += irq_stats(cpu)->irq_spurious_count; sum += irq_stats(cpu)->apic_perf_irqs; - sum += irq_stats(cpu)->apic_pending_irqs; + sum += irq_stats(cpu)->apic_irq_work_irqs; #endif if (x86_platform_ipi_callback) sum += irq_stats(cpu)->x86_platform_ipis; @@ -282,6 +282,7 @@ void fixup_irqs(void) unsigned int irq, vector; static int warned; struct irq_desc *desc; + struct irq_data *data; for_each_irq_desc(irq, desc) { int break_affinity = 0; @@ -296,7 +297,8 @@ void fixup_irqs(void) /* interrupt's are disabled at this point */ raw_spin_lock(&desc->lock); - affinity = desc->affinity; + data = &desc->irq_data; + affinity = data->affinity; if (!irq_has_action(irq) || cpumask_equal(affinity, cpu_online_mask)) { raw_spin_unlock(&desc->lock); @@ -315,16 +317,16 @@ void fixup_irqs(void) affinity = cpu_all_mask; } - if (!(desc->status & IRQ_MOVE_PCNTXT) && desc->chip->mask) - desc->chip->mask(irq); + if (!(desc->status & IRQ_MOVE_PCNTXT) && data->chip->irq_mask) + data->chip->irq_mask(data); - if (desc->chip->set_affinity) - desc->chip->set_affinity(irq, affinity); + if (data->chip->irq_set_affinity) + data->chip->irq_set_affinity(data, affinity, true); else if (!(warned++)) set_affinity = 0; - if (!(desc->status & IRQ_MOVE_PCNTXT) && desc->chip->unmask) - desc->chip->unmask(irq); + if (!(desc->status & IRQ_MOVE_PCNTXT) && data->chip->irq_unmask) + data->chip->irq_unmask(data); raw_spin_unlock(&desc->lock); @@ -355,10 +357,10 @@ void fixup_irqs(void) if (irr & (1 << (vector % 32))) { irq = __get_cpu_var(vector_irq)[vector]; - desc = irq_to_desc(irq); + data = irq_get_irq_data(irq); raw_spin_lock(&desc->lock); - if (desc->chip->retrigger) - desc->chip->retrigger(irq); + if (data->chip->irq_retrigger) + data->chip->irq_retrigger(data); raw_spin_unlock(&desc->lock); } } diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index 10709f29d16..50fbbe60e50 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -49,20 +49,19 @@ static inline int check_stack_overflow(void) { return 0; } static inline void print_stack_overflow(void) { } #endif -#ifdef CONFIG_4KSTACKS /* * per-CPU IRQ handling contexts (thread information and stack) */ union irq_ctx { struct thread_info tinfo; u32 stack[THREAD_SIZE/sizeof(u32)]; -} __attribute__((aligned(PAGE_SIZE))); +} __attribute__((aligned(THREAD_SIZE))); static DEFINE_PER_CPU(union irq_ctx *, hardirq_ctx); static DEFINE_PER_CPU(union irq_ctx *, softirq_ctx); -static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, hardirq_stack); -static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, softirq_stack); +static DEFINE_PER_CPU_MULTIPAGE_ALIGNED(union irq_ctx, hardirq_stack, THREAD_SIZE); +static DEFINE_PER_CPU_MULTIPAGE_ALIGNED(union irq_ctx, softirq_stack, THREAD_SIZE); static void call_on_stack(void *func, void *stack) { @@ -187,11 +186,6 @@ asmlinkage void do_softirq(void) local_irq_restore(flags); } -#else -static inline int -execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq) { return 0; } -#endif - bool handle_irq(unsigned irq, struct pt_regs *regs) { struct irq_desc *desc; diff --git a/arch/x86/kernel/irq_work.c b/arch/x86/kernel/irq_work.c new file mode 100644 index 00000000000..ca8f703a1e7 --- /dev/null +++ b/arch/x86/kernel/irq_work.c @@ -0,0 +1,30 @@ +/* + * x86 specific code for irq_work + * + * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + */ + +#include <linux/kernel.h> +#include <linux/irq_work.h> +#include <linux/hardirq.h> +#include <asm/apic.h> + +void smp_irq_work_interrupt(struct pt_regs *regs) +{ + irq_enter(); + ack_APIC_irq(); + inc_irq_stat(apic_irq_work_irqs); + irq_work_run(); + irq_exit(); +} + +void arch_irq_work_raise(void) +{ +#ifdef CONFIG_X86_LOCAL_APIC + if (!cpu_has_apic) + return; + + apic->send_IPI_self(IRQ_WORK_VECTOR); + apic_wait_icr_idle(); +#endif +} diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c index 990ae7cfc57..c752e973958 100644 --- a/arch/x86/kernel/irqinit.c +++ b/arch/x86/kernel/irqinit.c @@ -100,6 +100,8 @@ int vector_used_by_percpu_irq(unsigned int vector) void __init init_ISA_irqs(void) { + struct irq_chip *chip = legacy_pic->chip; + const char *name = chip->name; int i; #if defined(CONFIG_X86_64) || defined(CONFIG_X86_LOCAL_APIC) @@ -107,19 +109,8 @@ void __init init_ISA_irqs(void) #endif legacy_pic->init(0); - /* - * 16 old-style INTA-cycle interrupts: - */ - for (i = 0; i < legacy_pic->nr_legacy_irqs; 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"); - } + for (i = 0; i < legacy_pic->nr_legacy_irqs; i++) + set_irq_chip_and_handler_name(i, chip, handle_level_irq, name); } void __init init_IRQ(void) @@ -224,9 +215,9 @@ static void __init apic_intr_init(void) alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt); alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt); - /* Performance monitoring interrupts: */ -# ifdef CONFIG_PERF_EVENTS - alloc_intr_gate(LOCAL_PENDING_VECTOR, perf_pending_interrupt); + /* IRQ work interrupts: */ +# ifdef CONFIG_IRQ_WORK + alloc_intr_gate(IRQ_WORK_VECTOR, irq_work_interrupt); # endif #endif diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c new file mode 100644 index 00000000000..961b6b30ba9 --- /dev/null +++ b/arch/x86/kernel/jump_label.c @@ -0,0 +1,50 @@ +/* + * jump label x86 support + * + * Copyright (C) 2009 Jason Baron <jbaron@redhat.com> + * + */ +#include <linux/jump_label.h> +#include <linux/memory.h> +#include <linux/uaccess.h> +#include <linux/module.h> +#include <linux/list.h> +#include <linux/jhash.h> +#include <linux/cpu.h> +#include <asm/kprobes.h> +#include <asm/alternative.h> + +#ifdef HAVE_JUMP_LABEL + +union jump_code_union { + char code[JUMP_LABEL_NOP_SIZE]; + struct { + char jump; + int offset; + } __attribute__((packed)); +}; + +void arch_jump_label_transform(struct jump_entry *entry, + enum jump_label_type type) +{ + union jump_code_union code; + + if (type == JUMP_LABEL_ENABLE) { + code.jump = 0xe9; + code.offset = entry->target - + (entry->code + JUMP_LABEL_NOP_SIZE); + } else + memcpy(&code, ideal_nop5, JUMP_LABEL_NOP_SIZE); + get_online_cpus(); + mutex_lock(&text_mutex); + text_poke_smp((void *)entry->code, &code, JUMP_LABEL_NOP_SIZE); + mutex_unlock(&text_mutex); + put_online_cpus(); +} + +void arch_jump_label_text_poke_early(jump_label_t addr) +{ + text_poke_early((void *)addr, ideal_nop5, JUMP_LABEL_NOP_SIZE); +} + +#endif diff --git a/arch/x86/kernel/kdebugfs.c b/arch/x86/kernel/kdebugfs.c index 8afd9f321f1..90fcf62854b 100644 --- a/arch/x86/kernel/kdebugfs.c +++ b/arch/x86/kernel/kdebugfs.c @@ -78,6 +78,7 @@ static int setup_data_open(struct inode *inode, struct file *file) static const struct file_operations fops_setup_data = { .read = setup_data_read, .open = setup_data_open, + .llseek = default_llseek, }; static int __init diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index ef10940e1af..d81cfebb848 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -194,7 +194,7 @@ static struct hw_breakpoint { unsigned long addr; int len; int type; - struct perf_event **pev; + struct perf_event * __percpu *pev; } breakinfo[HBP_NUM]; static unsigned long early_dr7; @@ -477,8 +477,6 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code, raw_smp_processor_id()); } - kgdb_correct_hw_break(); - return 0; } @@ -621,7 +619,12 @@ int kgdb_arch_init(void) static void kgdb_hw_overflow_handler(struct perf_event *event, int nmi, struct perf_sample_data *data, struct pt_regs *regs) { - kgdb_ll_trap(DIE_DEBUG, "debug", regs, 0, 0, SIGTRAP); + struct task_struct *tsk = current; + int i; + + for (i = 0; i < 4; i++) + if (breakinfo[i].enabled) + tsk->thread.debugreg6 |= (DR_TRAP0 << i); } void kgdb_arch_late(void) @@ -644,7 +647,7 @@ void kgdb_arch_late(void) if (breakinfo[i].pev) continue; breakinfo[i].pev = register_wide_hw_breakpoint(&attr, NULL); - if (IS_ERR(breakinfo[i].pev)) { + if (IS_ERR((void * __force)breakinfo[i].pev)) { printk(KERN_ERR "kgdb: Could not allocate hw" "breakpoints\nDisabling the kernel debugger\n"); breakinfo[i].pev = NULL; diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 1bfb6cf4dd5..1cbd54c0df9 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -230,9 +230,6 @@ static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr) return 0; } -/* Dummy buffers for kallsyms_lookup */ -static char __dummy_buf[KSYM_NAME_LEN]; - /* Check if paddr is at an instruction boundary */ static int __kprobes can_probe(unsigned long paddr) { @@ -241,7 +238,7 @@ static int __kprobes can_probe(unsigned long paddr) struct insn insn; kprobe_opcode_t buf[MAX_INSN_SIZE]; - if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf)) + if (!kallsyms_lookup_size_offset(paddr, NULL, &offset)) return 0; /* Decode instructions */ @@ -709,6 +706,7 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs) struct hlist_node *node, *tmp; unsigned long flags, orig_ret_address = 0; unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline; + kprobe_opcode_t *correct_ret_addr = NULL; INIT_HLIST_HEAD(&empty_rp); kretprobe_hash_lock(current, &head, &flags); @@ -740,14 +738,34 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs) /* another task is sharing our hash bucket */ continue; + orig_ret_address = (unsigned long)ri->ret_addr; + + if (orig_ret_address != trampoline_address) + /* + * This is the real return address. Any other + * instances associated with this task are for + * other calls deeper on the call stack + */ + break; + } + + kretprobe_assert(ri, orig_ret_address, trampoline_address); + + correct_ret_addr = ri->ret_addr; + hlist_for_each_entry_safe(ri, node, tmp, head, hlist) { + if (ri->task != current) + /* another task is sharing our hash bucket */ + continue; + + orig_ret_address = (unsigned long)ri->ret_addr; if (ri->rp && ri->rp->handler) { __get_cpu_var(current_kprobe) = &ri->rp->kp; get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE; + ri->ret_addr = correct_ret_addr; ri->rp->handler(ri, regs); __get_cpu_var(current_kprobe) = NULL; } - orig_ret_address = (unsigned long)ri->ret_addr; recycle_rp_inst(ri, &empty_rp); if (orig_ret_address != trampoline_address) @@ -759,8 +777,6 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs) break; } - kretprobe_assert(ri, orig_ret_address, trampoline_address); - kretprobe_hash_unlock(current, &flags); hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) { @@ -1110,7 +1126,7 @@ static void __kprobes synthesize_set_arg1(kprobe_opcode_t *addr, *(unsigned long *)addr = val; } -void __kprobes kprobes_optinsn_template_holder(void) +static void __used __kprobes kprobes_optinsn_template_holder(void) { asm volatile ( ".global optprobe_template_entry\n" @@ -1202,7 +1218,8 @@ static int __kprobes copy_optimized_instructions(u8 *dest, u8 *src) } /* Check whether the address range is reserved */ if (ftrace_text_reserved(src, src + len - 1) || - alternatives_text_reserved(src, src + len - 1)) + alternatives_text_reserved(src, src + len - 1) || + jump_label_text_reserved(src, src + len - 1)) return -EBUSY; return len; @@ -1250,11 +1267,9 @@ static int __kprobes can_optimize(unsigned long paddr) unsigned long addr, size = 0, offset = 0; struct insn insn; kprobe_opcode_t buf[MAX_INSN_SIZE]; - /* Dummy buffers for lookup_symbol_attrs */ - static char __dummy_buf[KSYM_NAME_LEN]; /* Lookup symbol including addr */ - if (!kallsyms_lookup(paddr, &size, &offset, NULL, __dummy_buf)) + if (!kallsyms_lookup_size_offset(paddr, &size, &offset)) return 0; /* Check there is enough space for a relative jump. */ diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index eb9b76c716c..ca43ce31a19 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -128,13 +128,15 @@ static struct clocksource kvm_clock = { static int kvm_register_clock(char *txt) { int cpu = smp_processor_id(); - int low, high; + int low, high, ret; + low = (int)__pa(&per_cpu(hv_clock, cpu)) | 1; high = ((u64)__pa(&per_cpu(hv_clock, cpu)) >> 32); + ret = native_write_msr_safe(msr_kvm_system_time, low, high); printk(KERN_INFO "kvm-clock: cpu %d, msr %x:%x, %s\n", cpu, high, low, txt); - return native_write_msr_safe(msr_kvm_system_time, low, high); + return ret; } #ifdef CONFIG_X86_LOCAL_APIC diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c index 035c8c52918..b3ea9db39db 100644 --- a/arch/x86/kernel/machine_kexec_64.c +++ b/arch/x86/kernel/machine_kexec_64.c @@ -36,7 +36,7 @@ static int init_one_level2_page(struct kimage *image, pgd_t *pgd, if (!page) goto out; pud = (pud_t *)page_address(page); - memset(pud, 0, PAGE_SIZE); + clear_page(pud); set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE)); } pud = pud_offset(pgd, addr); @@ -45,7 +45,7 @@ static int init_one_level2_page(struct kimage *image, pgd_t *pgd, if (!page) goto out; pmd = (pmd_t *)page_address(page); - memset(pmd, 0, PAGE_SIZE); + clear_page(pmd); set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE)); } pmd = pmd_offset(pud, addr); diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c index fa6551d36c1..1cca374a2ba 100644 --- a/arch/x86/kernel/microcode_core.c +++ b/arch/x86/kernel/microcode_core.c @@ -12,7 +12,7 @@ * Software Developer's Manual * Order Number 253668 or free download from: * - * http://developer.intel.com/design/pentium4/manuals/253668.htm + * http://developer.intel.com/Assets/PDF/manual/253668.pdf * * For more information, go to http://www.urbanmyth.org/microcode * @@ -232,6 +232,7 @@ static const struct file_operations microcode_fops = { .owner = THIS_MODULE, .write = microcode_write, .open = microcode_open, + .llseek = no_llseek, }; static struct miscdevice microcode_dev = { diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index 356170262a9..dcb65cc0a05 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -12,7 +12,7 @@ * Software Developer's Manual * Order Number 253668 or free download from: * - * http://developer.intel.com/design/pentium4/manuals/253668.htm + * http://developer.intel.com/Assets/PDF/manual/253668.pdf * * For more information, go to http://www.urbanmyth.org/microcode * diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c index e0bc186d750..8f295609173 100644 --- a/arch/x86/kernel/module.c +++ b/arch/x86/kernel/module.c @@ -239,11 +239,13 @@ int module_finalize(const Elf_Ehdr *hdr, apply_paravirt(pseg, pseg + para->sh_size); } - return module_bug_finalize(hdr, sechdrs, me); + /* make jump label nops */ + jump_label_apply_nops(me); + + return 0; } void module_arch_cleanup(struct module *mod) { alternatives_smp_module_del(mod); - module_bug_cleanup(mod); } diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index d86dbf7e54b..9af64d9c4b6 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -11,6 +11,7 @@ #include <linux/init.h> #include <linux/delay.h> #include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/kernel_stat.h> #include <linux/mc146818rtc.h> #include <linux/bitops.h> @@ -274,6 +275,18 @@ static void __init smp_dump_mptable(struct mpc_table *mpc, unsigned char *mpt) void __init default_smp_read_mpc_oem(struct mpc_table *mpc) { } +static void __init smp_register_lapic_address(unsigned long address) +{ + mp_lapic_addr = address; + + set_fixmap_nocache(FIX_APIC_BASE, address); + if (boot_cpu_physical_apicid == -1U) { + boot_cpu_physical_apicid = read_apic_id(); + apic_version[boot_cpu_physical_apicid] = + GET_APIC_VERSION(apic_read(APIC_LVR)); + } +} + static int __init smp_read_mpc(struct mpc_table *mpc, unsigned early) { char str[16]; @@ -295,6 +308,10 @@ static int __init smp_read_mpc(struct mpc_table *mpc, unsigned early) if (early) return 1; + /* Initialize the lapic mapping */ + if (!acpi_lapic) + smp_register_lapic_address(mpc->lapic); + if (mpc->oemptr) x86_init.mpparse.smp_read_mpc_oem(mpc); @@ -641,7 +658,7 @@ static void __init smp_reserve_memory(struct mpf_intel *mpf) { unsigned long size = get_mpc_size(mpf->physptr); - reserve_early_overlap_ok(mpf->physptr, mpf->physptr+size, "MP-table mpc"); + memblock_x86_reserve_range(mpf->physptr, mpf->physptr+size, "* MP-table mpc"); } static int __init smp_scan_config(unsigned long base, unsigned long length) @@ -670,7 +687,7 @@ static int __init smp_scan_config(unsigned long base, unsigned long length) mpf, (u64)virt_to_phys(mpf)); mem = virt_to_phys(mpf); - reserve_early_overlap_ok(mem, mem + sizeof(*mpf), "MP-table mpf"); + memblock_x86_reserve_range(mem, mem + sizeof(*mpf), "* MP-table mpf"); if (mpf->physptr) smp_reserve_memory(mpf); diff --git a/arch/x86/kernel/olpc-xo1.c b/arch/x86/kernel/olpc-xo1.c new file mode 100644 index 00000000000..f5442c03abc --- /dev/null +++ b/arch/x86/kernel/olpc-xo1.c @@ -0,0 +1,140 @@ +/* + * Support for features of the OLPC XO-1 laptop + * + * Copyright (C) 2010 One Laptop per Child + * Copyright (C) 2006 Red Hat, Inc. + * Copyright (C) 2006 Advanced Micro Devices, Inc. + * + * 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. + */ + +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/pci_ids.h> +#include <linux/platform_device.h> +#include <linux/pm.h> + +#include <asm/io.h> +#include <asm/olpc.h> + +#define DRV_NAME "olpc-xo1" + +#define PMS_BAR 4 +#define ACPI_BAR 5 + +/* PMC registers (PMS block) */ +#define PM_SCLK 0x10 +#define PM_IN_SLPCTL 0x20 +#define PM_WKXD 0x34 +#define PM_WKD 0x30 +#define PM_SSC 0x54 + +/* PM registers (ACPI block) */ +#define PM1_CNT 0x08 +#define PM_GPE0_STS 0x18 + +static unsigned long acpi_base; +static unsigned long pms_base; + +static void xo1_power_off(void) +{ + printk(KERN_INFO "OLPC XO-1 power off sequence...\n"); + + /* Enable all of these controls with 0 delay */ + outl(0x40000000, pms_base + PM_SCLK); + outl(0x40000000, pms_base + PM_IN_SLPCTL); + outl(0x40000000, pms_base + PM_WKXD); + outl(0x40000000, pms_base + PM_WKD); + + /* Clear status bits (possibly unnecessary) */ + outl(0x0002ffff, pms_base + PM_SSC); + outl(0xffffffff, acpi_base + PM_GPE0_STS); + + /* Write SLP_EN bit to start the machinery */ + outl(0x00002000, acpi_base + PM1_CNT); +} + +/* Read the base addresses from the PCI BAR info */ +static int __devinit setup_bases(struct pci_dev *pdev) +{ + int r; + + r = pci_enable_device_io(pdev); + if (r) { + dev_err(&pdev->dev, "can't enable device IO\n"); + return r; + } + + r = pci_request_region(pdev, ACPI_BAR, DRV_NAME); + if (r) { + dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", ACPI_BAR); + return r; + } + + r = pci_request_region(pdev, PMS_BAR, DRV_NAME); + if (r) { + dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", PMS_BAR); + pci_release_region(pdev, ACPI_BAR); + return r; + } + + acpi_base = pci_resource_start(pdev, ACPI_BAR); + pms_base = pci_resource_start(pdev, PMS_BAR); + + return 0; +} + +static int __devinit olpc_xo1_probe(struct platform_device *pdev) +{ + struct pci_dev *pcidev; + int r; + + pcidev = pci_get_device(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_ISA, + NULL); + if (!pdev) + return -ENODEV; + + r = setup_bases(pcidev); + if (r) + return r; + + pm_power_off = xo1_power_off; + + printk(KERN_INFO "OLPC XO-1 support registered\n"); + return 0; +} + +static int __devexit olpc_xo1_remove(struct platform_device *pdev) +{ + pm_power_off = NULL; + return 0; +} + +static struct platform_driver olpc_xo1_driver = { + .driver = { + .name = DRV_NAME, + .owner = THIS_MODULE, + }, + .probe = olpc_xo1_probe, + .remove = __devexit_p(olpc_xo1_remove), +}; + +static int __init olpc_xo1_init(void) +{ + return platform_driver_register(&olpc_xo1_driver); +} + +static void __exit olpc_xo1_exit(void) +{ + platform_driver_unregister(&olpc_xo1_driver); +} + +MODULE_AUTHOR("Daniel Drake <dsd@laptop.org>"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("platform:olpc-xo1"); + +module_init(olpc_xo1_init); +module_exit(olpc_xo1_exit); diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c index 0e0cdde519b..edaf3fe8dc5 100644 --- a/arch/x86/kernel/olpc.c +++ b/arch/x86/kernel/olpc.c @@ -17,6 +17,7 @@ #include <linux/spinlock.h> #include <linux/io.h> #include <linux/string.h> +#include <linux/platform_device.h> #include <asm/geode.h> #include <asm/setup.h> @@ -114,6 +115,7 @@ int olpc_ec_cmd(unsigned char cmd, unsigned char *inbuf, size_t inlen, unsigned long flags; int ret = -EIO; int i; + int restarts = 0; spin_lock_irqsave(&ec_lock, flags); @@ -169,7 +171,9 @@ restart: if (wait_on_obf(0x6c, 1)) { printk(KERN_ERR "olpc-ec: timeout waiting for" " EC to provide data!\n"); - goto restart; + if (restarts++ < 10) + goto restart; + goto err; } outbuf[i] = inb(0x68); pr_devel("olpc-ec: received 0x%x\n", outbuf[i]); @@ -183,8 +187,21 @@ err: } EXPORT_SYMBOL_GPL(olpc_ec_cmd); -#ifdef CONFIG_OLPC_OPENFIRMWARE -static void __init platform_detect(void) +static bool __init check_ofw_architecture(void) +{ + size_t propsize; + char olpc_arch[5]; + const void *args[] = { NULL, "architecture", olpc_arch, (void *)5 }; + void *res[] = { &propsize }; + + if (olpc_ofw("getprop", args, res)) { + printk(KERN_ERR "ofw: getprop call failed!\n"); + return false; + } + return propsize == 5 && strncmp("OLPC", olpc_arch, 5) == 0; +} + +static u32 __init get_board_revision(void) { size_t propsize; __be32 rev; @@ -193,45 +210,43 @@ static void __init platform_detect(void) if (olpc_ofw("getprop", args, res) || propsize != 4) { printk(KERN_ERR "ofw: getprop call failed!\n"); - rev = cpu_to_be32(0); + return cpu_to_be32(0); } - olpc_platform_info.boardrev = be32_to_cpu(rev); + return be32_to_cpu(rev); } -#else -static void __init platform_detect(void) + +static bool __init platform_detect(void) { - /* stopgap until OFW support is added to the kernel */ - olpc_platform_info.boardrev = olpc_board(0xc2); + if (!check_ofw_architecture()) + return false; + olpc_platform_info.flags |= OLPC_F_PRESENT; + olpc_platform_info.boardrev = get_board_revision(); + return true; } -#endif -static int __init olpc_init(void) +static int __init add_xo1_platform_devices(void) { - unsigned char *romsig; + struct platform_device *pdev; - /* The ioremap check is dangerous; limit what we run it on */ - if (!is_geode() || cs5535_has_vsa2()) - return 0; + pdev = platform_device_register_simple("xo1-rfkill", -1, NULL, 0); + if (IS_ERR(pdev)) + return PTR_ERR(pdev); - spin_lock_init(&ec_lock); + pdev = platform_device_register_simple("olpc-xo1", -1, NULL, 0); + if (IS_ERR(pdev)) + return PTR_ERR(pdev); - romsig = ioremap(0xffffffc0, 16); - if (!romsig) - return 0; + return 0; +} - if (strncmp(romsig, "CL1 Q", 7)) - goto unmap; - if (strncmp(romsig+6, romsig+13, 3)) { - printk(KERN_INFO "OLPC BIOS signature looks invalid. " - "Assuming not OLPC\n"); - goto unmap; - } +static int __init olpc_init(void) +{ + int r = 0; - printk(KERN_INFO "OLPC board with OpenFirmware %.16s\n", romsig); - olpc_platform_info.flags |= OLPC_F_PRESENT; + if (!olpc_ofw_present() || !platform_detect()) + return 0; - /* get the platform revision */ - platform_detect(); + spin_lock_init(&ec_lock); /* assume B1 and above models always have a DCON */ if (olpc_board_at_least(olpc_board(0xb1))) @@ -242,8 +257,10 @@ static int __init olpc_init(void) (unsigned char *) &olpc_platform_info.ecver, 1); #ifdef CONFIG_PCI_OLPC - /* If the VSA exists let it emulate PCI, if not emulate in kernel */ - if (!cs5535_has_vsa2()) + /* If the VSA exists let it emulate PCI, if not emulate in kernel. + * XO-1 only. */ + if (olpc_platform_info.boardrev < olpc_board_pre(0xd0) && + !cs5535_has_vsa2()) x86_init.pci.arch_init = pci_olpc_init; #endif @@ -252,8 +269,12 @@ static int __init olpc_init(void) olpc_platform_info.boardrev >> 4, olpc_platform_info.ecver); -unmap: - iounmap(romsig); + if (olpc_platform_info.boardrev < olpc_board_pre(0xd0)) { /* XO-1 */ + r = add_xo1_platform_devices(); + if (r) + return r; + } + return 0; } diff --git a/arch/x86/kernel/olpc_ofw.c b/arch/x86/kernel/olpc_ofw.c index 3218aa71ab5..78732046437 100644 --- a/arch/x86/kernel/olpc_ofw.c +++ b/arch/x86/kernel/olpc_ofw.c @@ -74,6 +74,12 @@ int __olpc_ofw(const char *name, int nr_args, const void **args, int nr_res, } EXPORT_SYMBOL_GPL(__olpc_ofw); +bool olpc_ofw_present(void) +{ + return olpc_ofw_cif != NULL; +} +EXPORT_SYMBOL_GPL(olpc_ofw_present); + /* OFW cif _should_ be above this address */ #define OFW_MIN 0xff000000 diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 1db183ed7c0..c5b250011fd 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -413,7 +413,6 @@ struct pv_mmu_ops pv_mmu_ops = { .alloc_pte = paravirt_nop, .alloc_pmd = paravirt_nop, - .alloc_pmd_clone = paravirt_nop, .alloc_pud = paravirt_nop, .release_pte = paravirt_nop, .release_pmd = paravirt_nop, diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c index 078d4ec1a9d..f56a117cef6 100644 --- a/arch/x86/kernel/pci-calgary_64.c +++ b/arch/x86/kernel/pci-calgary_64.c @@ -47,6 +47,7 @@ #include <asm/rio.h> #include <asm/bios_ebda.h> #include <asm/x86_init.h> +#include <asm/iommu_table.h> #ifdef CONFIG_CALGARY_IOMMU_ENABLED_BY_DEFAULT int use_calgary __read_mostly = 1; @@ -1364,7 +1365,7 @@ static int __init calgary_iommu_init(void) return 0; } -void __init detect_calgary(void) +int __init detect_calgary(void) { int bus; void *tbl; @@ -1378,13 +1379,13 @@ void __init detect_calgary(void) * another HW IOMMU already, bail out. */ if (no_iommu || iommu_detected) - return; + return -ENODEV; if (!use_calgary) - return; + return -ENODEV; if (!early_pci_allowed()) - return; + return -ENODEV; printk(KERN_DEBUG "Calgary: detecting Calgary via BIOS EBDA area\n"); @@ -1410,13 +1411,13 @@ void __init detect_calgary(void) if (!rio_table_hdr) { printk(KERN_DEBUG "Calgary: Unable to locate Rio Grande table " "in EBDA - bailing!\n"); - return; + return -ENODEV; } ret = build_detail_arrays(); if (ret) { printk(KERN_DEBUG "Calgary: build_detail_arrays ret %d\n", ret); - return; + return -ENOMEM; } specified_table_size = determine_tce_table_size((is_kdump_kernel() ? @@ -1464,7 +1465,7 @@ void __init detect_calgary(void) x86_init.iommu.iommu_init = calgary_iommu_init; } - return; + return calgary_found; cleanup: for (--bus; bus >= 0; --bus) { @@ -1473,6 +1474,7 @@ cleanup: if (info->tce_space) free_tce_table(info->tce_space); } + return -ENOMEM; } static int __init calgary_parse_options(char *p) @@ -1594,3 +1596,5 @@ static int __init calgary_fixup_tce_spaces(void) * and before device_initcall. */ rootfs_initcall(calgary_fixup_tce_spaces); + +IOMMU_INIT_POST(detect_calgary); diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 9f07cfcbd3a..9ea999a4dcc 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -11,9 +11,8 @@ #include <asm/iommu.h> #include <asm/gart.h> #include <asm/calgary.h> -#include <asm/amd_iommu.h> #include <asm/x86_init.h> -#include <asm/xen/swiotlb-xen.h> +#include <asm/iommu_table.h> static int forbid_dac __read_mostly; @@ -45,6 +44,8 @@ int iommu_detected __read_mostly = 0; */ int iommu_pass_through __read_mostly; +extern struct iommu_table_entry __iommu_table[], __iommu_table_end[]; + /* Dummy device used for NULL arguments (normally ISA). */ struct device x86_dma_fallback_dev = { .init_name = "fallback device", @@ -130,26 +131,24 @@ static void __init dma32_free_bootmem(void) void __init pci_iommu_alloc(void) { + struct iommu_table_entry *p; + /* free the range so iommu could get some range less than 4G */ dma32_free_bootmem(); - if (pci_xen_swiotlb_detect() || pci_swiotlb_detect()) - goto out; - - gart_iommu_hole_init(); - - detect_calgary(); - - detect_intel_iommu(); + sort_iommu_table(__iommu_table, __iommu_table_end); + check_iommu_entries(__iommu_table, __iommu_table_end); - /* needs to be called after gart_iommu_hole_init */ - amd_iommu_detect(); -out: - pci_xen_swiotlb_init(); - - pci_swiotlb_init(); + for (p = __iommu_table; p < __iommu_table_end; p++) { + if (p && p->detect && p->detect() > 0) { + p->flags |= IOMMU_DETECTED; + if (p->early_init) + p->early_init(); + if (p->flags & IOMMU_FINISH_IF_DETECTED) + break; + } + } } - void *dma_generic_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_addr, gfp_t flag) { @@ -292,6 +291,7 @@ EXPORT_SYMBOL(dma_supported); static int __init pci_iommu_init(void) { + struct iommu_table_entry *p; dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES); #ifdef CONFIG_PCI @@ -299,12 +299,10 @@ static int __init pci_iommu_init(void) #endif x86_init.iommu.iommu_init(); - if (swiotlb || xen_swiotlb) { - printk(KERN_INFO "PCI-DMA: " - "Using software bounce buffering for IO (SWIOTLB)\n"); - swiotlb_print_info(); - } else - swiotlb_free(); + for (p = __iommu_table; p < __iommu_table_end; p++) { + if (p && (p->flags & IOMMU_DETECTED) && p->late_init) + p->late_init(); + } return 0; } diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 0f7f130caa6..ba0f0ca9f28 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -39,8 +39,9 @@ #include <asm/cacheflush.h> #include <asm/swiotlb.h> #include <asm/dma.h> -#include <asm/k8.h> +#include <asm/amd_nb.h> #include <asm/x86_init.h> +#include <asm/iommu_table.h> static unsigned long iommu_bus_base; /* GART remapping area (physical) */ static unsigned long iommu_size; /* size of remapping area bytes */ @@ -560,8 +561,11 @@ static void enable_gart_translations(void) { int i; - for (i = 0; i < num_k8_northbridges; i++) { - struct pci_dev *dev = k8_northbridges[i]; + if (!k8_northbridges.gart_supported) + return; + + for (i = 0; i < k8_northbridges.num; i++) { + struct pci_dev *dev = k8_northbridges.nb_misc[i]; enable_gart_translation(dev, __pa(agp_gatt_table)); } @@ -592,16 +596,19 @@ static void gart_fixup_northbridges(struct sys_device *dev) if (!fix_up_north_bridges) return; + if (!k8_northbridges.gart_supported) + return; + pr_info("PCI-DMA: Restoring GART aperture settings\n"); - for (i = 0; i < num_k8_northbridges; i++) { - struct pci_dev *dev = k8_northbridges[i]; + for (i = 0; i < k8_northbridges.num; i++) { + struct pci_dev *dev = k8_northbridges.nb_misc[i]; /* * Don't enable translations just yet. That is the next * step. Restore the pre-suspend aperture settings. */ - pci_write_config_dword(dev, AMD64_GARTAPERTURECTL, aperture_order << 1); + gart_set_size_and_enable(dev, aperture_order); pci_write_config_dword(dev, AMD64_GARTAPERTUREBASE, aperture_alloc >> 25); } } @@ -649,8 +656,8 @@ static __init int init_k8_gatt(struct agp_kern_info *info) aper_size = aper_base = info->aper_size = 0; dev = NULL; - for (i = 0; i < num_k8_northbridges; i++) { - dev = k8_northbridges[i]; + for (i = 0; i < k8_northbridges.num; i++) { + dev = k8_northbridges.nb_misc[i]; new_aper_base = read_aperture(dev, &new_aper_size); if (!new_aper_base) goto nommu; @@ -718,10 +725,13 @@ static void gart_iommu_shutdown(void) if (!no_agp) return; - for (i = 0; i < num_k8_northbridges; i++) { + if (!k8_northbridges.gart_supported) + return; + + for (i = 0; i < k8_northbridges.num; i++) { u32 ctl; - dev = k8_northbridges[i]; + dev = k8_northbridges.nb_misc[i]; pci_read_config_dword(dev, AMD64_GARTAPERTURECTL, &ctl); ctl &= ~GARTEN; @@ -739,7 +749,7 @@ int __init gart_iommu_init(void) unsigned long scratch; long i; - if (num_k8_northbridges == 0) + if (!k8_northbridges.gart_supported) return 0; #ifndef CONFIG_AGP_AMD64 @@ -896,3 +906,4 @@ void __init gart_parse_options(char *p) } } } +IOMMU_INIT_POST(gart_iommu_hole_init); diff --git a/arch/x86/kernel/pci-iommu_table.c b/arch/x86/kernel/pci-iommu_table.c new file mode 100644 index 00000000000..55d745ec118 --- /dev/null +++ b/arch/x86/kernel/pci-iommu_table.c @@ -0,0 +1,89 @@ +#include <linux/dma-mapping.h> +#include <asm/iommu_table.h> +#include <linux/string.h> +#include <linux/kallsyms.h> + + +#define DEBUG 1 + +static struct iommu_table_entry * __init +find_dependents_of(struct iommu_table_entry *start, + struct iommu_table_entry *finish, + struct iommu_table_entry *q) +{ + struct iommu_table_entry *p; + + if (!q) + return NULL; + + for (p = start; p < finish; p++) + if (p->detect == q->depend) + return p; + + return NULL; +} + + +void __init sort_iommu_table(struct iommu_table_entry *start, + struct iommu_table_entry *finish) { + + struct iommu_table_entry *p, *q, tmp; + + for (p = start; p < finish; p++) { +again: + q = find_dependents_of(start, finish, p); + /* We are bit sneaky here. We use the memory address to figure + * out if the node we depend on is past our point, if so, swap. + */ + if (q > p) { + tmp = *p; + memmove(p, q, sizeof(*p)); + *q = tmp; + goto again; + } + } + +} + +#ifdef DEBUG +void __init check_iommu_entries(struct iommu_table_entry *start, + struct iommu_table_entry *finish) +{ + struct iommu_table_entry *p, *q, *x; + char sym_p[KSYM_SYMBOL_LEN]; + char sym_q[KSYM_SYMBOL_LEN]; + + /* Simple cyclic dependency checker. */ + for (p = start; p < finish; p++) { + q = find_dependents_of(start, finish, p); + x = find_dependents_of(start, finish, q); + if (p == x) { + sprint_symbol(sym_p, (unsigned long)p->detect); + sprint_symbol(sym_q, (unsigned long)q->detect); + + printk(KERN_ERR "CYCLIC DEPENDENCY FOUND! %s depends" \ + " on %s and vice-versa. BREAKING IT.\n", + sym_p, sym_q); + /* Heavy handed way..*/ + x->depend = 0; + } + } + + for (p = start; p < finish; p++) { + q = find_dependents_of(p, finish, p); + if (q && q > p) { + sprint_symbol(sym_p, (unsigned long)p->detect); + sprint_symbol(sym_q, (unsigned long)q->detect); + + printk(KERN_ERR "EXECUTION ORDER INVALID! %s "\ + "should be called before %s!\n", + sym_p, sym_q); + } + } +} +#else +inline void check_iommu_entries(struct iommu_table_entry *start, + struct iommu_table_entry *finish) +{ +} +#endif diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c index a5bc528d432..8f972cbddef 100644 --- a/arch/x86/kernel/pci-swiotlb.c +++ b/arch/x86/kernel/pci-swiotlb.c @@ -10,7 +10,8 @@ #include <asm/iommu.h> #include <asm/swiotlb.h> #include <asm/dma.h> - +#include <asm/xen/swiotlb-xen.h> +#include <asm/iommu_table.h> int swiotlb __read_mostly; static void *x86_swiotlb_alloc_coherent(struct device *hwdev, size_t size, @@ -41,25 +42,42 @@ static struct dma_map_ops swiotlb_dma_ops = { }; /* - * pci_swiotlb_detect - set swiotlb to 1 if necessary + * pci_swiotlb_detect_override - set swiotlb to 1 if necessary * * This returns non-zero if we are forced to use swiotlb (by the boot * option). */ -int __init pci_swiotlb_detect(void) +int __init pci_swiotlb_detect_override(void) { int use_swiotlb = swiotlb | swiotlb_force; + if (swiotlb_force) + swiotlb = 1; + + return use_swiotlb; +} +IOMMU_INIT_FINISH(pci_swiotlb_detect_override, + pci_xen_swiotlb_detect, + pci_swiotlb_init, + pci_swiotlb_late_init); + +/* + * if 4GB or more detected (and iommu=off not set) return 1 + * and set swiotlb to 1. + */ +int __init pci_swiotlb_detect_4gb(void) +{ /* don't initialize swiotlb if iommu=off (no_iommu=1) */ #ifdef CONFIG_X86_64 if (!no_iommu && max_pfn > MAX_DMA32_PFN) swiotlb = 1; #endif - if (swiotlb_force) - swiotlb = 1; - - return use_swiotlb; + return swiotlb; } +IOMMU_INIT(pci_swiotlb_detect_4gb, + pci_swiotlb_detect_override, + pci_swiotlb_init, + pci_swiotlb_late_init); void __init pci_swiotlb_init(void) { @@ -68,3 +86,15 @@ void __init pci_swiotlb_init(void) dma_ops = &swiotlb_dma_ops; } } + +void __init pci_swiotlb_late_init(void) +{ + /* An IOMMU turned us off. */ + if (!swiotlb) + swiotlb_free(); + else { + printk(KERN_INFO "PCI-DMA: " + "Using software bounce buffering for IO (SWIOTLB)\n"); + swiotlb_print_info(); + } +} diff --git a/arch/x86/kernel/pmtimer_64.c b/arch/x86/kernel/pmtimer_64.c deleted file mode 100644 index b112406f199..00000000000 --- a/arch/x86/kernel/pmtimer_64.c +++ /dev/null @@ -1,69 +0,0 @@ -/* Ported over from i386 by AK, original copyright was: - * - * (C) Dominik Brodowski <linux@brodo.de> 2003 - * - * Driver to use the Power Management Timer (PMTMR) available in some - * southbridges as primary timing source for the Linux kernel. - * - * Based on parts of linux/drivers/acpi/hardware/hwtimer.c, timer_pit.c, - * timer_hpet.c, and on Arjan van de Ven's implementation for 2.4. - * - * This file is licensed under the GPL v2. - * - * Dropped all the hardware bug workarounds for now. Hopefully they - * are not needed on 64bit chipsets. - */ - -#include <linux/jiffies.h> -#include <linux/kernel.h> -#include <linux/time.h> -#include <linux/init.h> -#include <linux/cpumask.h> -#include <linux/acpi_pmtmr.h> - -#include <asm/io.h> -#include <asm/proto.h> -#include <asm/msr.h> -#include <asm/vsyscall.h> - -static inline u32 cyc2us(u32 cycles) -{ - /* The Power Management Timer ticks at 3.579545 ticks per microsecond. - * 1 / PM_TIMER_FREQUENCY == 0.27936511 =~ 286/1024 [error: 0.024%] - * - * Even with HZ = 100, delta is at maximum 35796 ticks, so it can - * easily be multiplied with 286 (=0x11E) without having to fear - * u32 overflows. - */ - cycles *= 286; - return (cycles >> 10); -} - -static unsigned pmtimer_wait_tick(void) -{ - u32 a, b; - for (a = b = inl(pmtmr_ioport) & ACPI_PM_MASK; - a == b; - b = inl(pmtmr_ioport) & ACPI_PM_MASK) - cpu_relax(); - return b; -} - -/* note: wait time is rounded up to one tick */ -void pmtimer_wait(unsigned us) -{ - u32 a, b; - a = pmtimer_wait_tick(); - do { - b = inl(pmtmr_ioport); - cpu_relax(); - } while (cyc2us(b - a) < us); -} - -static int __init nopmtimer_setup(char *s) -{ - pmtmr_ioport = 0; - return 1; -} - -__setup("nopmtimer", nopmtimer_setup); diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index d401f1d2d06..57d1868a86a 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -301,8 +301,9 @@ EXPORT_SYMBOL(kernel_thread); /* * sys_execve() executes a new program. */ -long sys_execve(char __user *name, char __user * __user *argv, - char __user * __user *envp, struct pt_regs *regs) +long sys_execve(const char __user *name, + const char __user *const __user *argv, + const char __user *const __user *envp, struct pt_regs *regs) { long error; char *filename; diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 3d9ea531ddd..b3d7a3a04f3 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -424,7 +424,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) load_TLS(next, cpu); /* Must be after DS reload */ - unlazy_fpu(prev_p); + __unlazy_fpu(prev_p); /* Make sure cpu is ready for new context */ if (preload_fpu) diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index 239427ca02a..bab3b9e6f66 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -82,7 +82,8 @@ static inline u64 scale_delta(u64 delta, u32 mul_frac, int shift) static u64 pvclock_get_nsec_offset(struct pvclock_shadow_time *shadow) { u64 delta = native_read_tsc() - shadow->tsc_timestamp; - return scale_delta(delta, shadow->tsc_to_nsec_mul, shadow->tsc_shift); + return pvclock_scale_delta(delta, shadow->tsc_to_nsec_mul, + shadow->tsc_shift); } /* diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c index 939b9e98245..8bbe8c56916 100644 --- a/arch/x86/kernel/quirks.c +++ b/arch/x86/kernel/quirks.c @@ -344,6 +344,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8235, vt8237_force_enable_hpet); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237, vt8237_force_enable_hpet); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_CX700, + vt8237_force_enable_hpet); static void ati_force_hpet_resume(void) { diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index e3af342fe83..f7f53dcd3e0 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -84,7 +84,7 @@ static int __init reboot_setup(char *str) } /* we will leave sorting out the final value when we are ready to reboot, since we might not - have set up boot_cpu_id or smp_num_cpu */ + have detected BSP APIC ID or smp_num_cpu */ break; #endif /* CONFIG_SMP */ @@ -371,16 +371,10 @@ void machine_real_restart(const unsigned char *code, int length) CMOS_WRITE(0x00, 0x8f); spin_unlock(&rtc_lock); - /* Remap the kernel at virtual address zero, as well as offset zero - from the kernel segment. This assumes the kernel segment starts at - virtual address PAGE_OFFSET. */ - memcpy(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, - sizeof(swapper_pg_dir [0]) * KERNEL_PGD_PTRS); - /* - * Use `swapper_pg_dir' as our page directory. + * Switch back to the initial page table. */ - load_cr3(swapper_pg_dir); + load_cr3(initial_page_table); /* Write 0x1234 to absolute memory location 0x472. The BIOS reads this on booting to tell it to "Bypass memory test (also warm diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index b008e788320..95a32746fbf 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -31,6 +31,7 @@ #include <linux/apm_bios.h> #include <linux/initrd.h> #include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/seq_file.h> #include <linux/console.h> #include <linux/mca.h> @@ -83,7 +84,6 @@ #include <asm/dmi.h> #include <asm/io_apic.h> #include <asm/ist.h> -#include <asm/vmi.h> #include <asm/setup_arch.h> #include <asm/bios_ebda.h> #include <asm/cacheflush.h> @@ -107,11 +107,12 @@ #include <asm/percpu.h> #include <asm/topology.h> #include <asm/apicdef.h> -#include <asm/k8.h> +#include <asm/amd_nb.h> #ifdef CONFIG_X86_64 #include <asm/numa_64.h> #endif #include <asm/mce.h> +#include <asm/alternative.h> /* * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. @@ -125,7 +126,6 @@ unsigned long max_pfn_mapped; RESERVE_BRK(dmi_alloc, 65536); #endif -unsigned int boot_cpu_id __read_mostly; static __initdata unsigned long _brk_start = (unsigned long)__brk_base; unsigned long _brk_end = (unsigned long)__brk_base; @@ -302,7 +302,7 @@ static inline void init_gbpages(void) static void __init reserve_brk(void) { if (_brk_end > _brk_start) - reserve_early(__pa(_brk_start), __pa(_brk_end), "BRK"); + memblock_x86_reserve_range(__pa(_brk_start), __pa(_brk_end), "BRK"); /* Mark brk area as locked down and no longer taking any new allocations */ @@ -324,17 +324,16 @@ static void __init relocate_initrd(void) char *p, *q; /* We need to move the initrd down into lowmem */ - ramdisk_here = find_e820_area(0, end_of_lowmem, area_size, + ramdisk_here = memblock_find_in_range(0, end_of_lowmem, area_size, PAGE_SIZE); - if (ramdisk_here == -1ULL) + if (ramdisk_here == MEMBLOCK_ERROR) panic("Cannot find place for new RAMDISK of size %lld\n", ramdisk_size); /* Note: this includes all the lowmem currently occupied by the initrd, we rely on that fact to keep the data intact. */ - reserve_early(ramdisk_here, ramdisk_here + area_size, - "NEW RAMDISK"); + memblock_x86_reserve_range(ramdisk_here, ramdisk_here + area_size, "NEW RAMDISK"); initrd_start = ramdisk_here + PAGE_OFFSET; initrd_end = initrd_start + ramdisk_size; printk(KERN_INFO "Allocated new RAMDISK: %08llx - %08llx\n", @@ -390,7 +389,7 @@ static void __init reserve_initrd(void) initrd_start = 0; if (ramdisk_size >= (end_of_lowmem>>1)) { - free_early(ramdisk_image, ramdisk_end); + memblock_x86_free_range(ramdisk_image, ramdisk_end); printk(KERN_ERR "initrd too large to handle, " "disabling initrd\n"); return; @@ -413,7 +412,7 @@ static void __init reserve_initrd(void) relocate_initrd(); - free_early(ramdisk_image, ramdisk_end); + memblock_x86_free_range(ramdisk_image, ramdisk_end); } #else static void __init reserve_initrd(void) @@ -469,7 +468,7 @@ static void __init e820_reserve_setup_data(void) e820_print_map("reserve setup_data"); } -static void __init reserve_early_setup_data(void) +static void __init memblock_x86_reserve_range_setup_data(void) { struct setup_data *data; u64 pa_data; @@ -481,7 +480,7 @@ static void __init reserve_early_setup_data(void) while (pa_data) { data = early_memremap(pa_data, sizeof(*data)); sprintf(buf, "setup data %x", data->type); - reserve_early(pa_data, pa_data+sizeof(*data)+data->len, buf); + memblock_x86_reserve_range(pa_data, pa_data+sizeof(*data)+data->len, buf); pa_data = data->next; early_iounmap(data, sizeof(*data)); } @@ -502,6 +501,7 @@ static inline unsigned long long get_total_mem(void) return total << PAGE_SHIFT; } +#define DEFAULT_BZIMAGE_ADDR_MAX 0x37FFFFFF static void __init reserve_crashkernel(void) { unsigned long long total_mem; @@ -519,23 +519,27 @@ static void __init reserve_crashkernel(void) if (crash_base <= 0) { const unsigned long long alignment = 16<<20; /* 16M */ - crash_base = find_e820_area(alignment, ULONG_MAX, crash_size, - alignment); - if (crash_base == -1ULL) { + /* + * kexec want bzImage is below DEFAULT_BZIMAGE_ADDR_MAX + */ + crash_base = memblock_find_in_range(alignment, + DEFAULT_BZIMAGE_ADDR_MAX, crash_size, alignment); + + if (crash_base == MEMBLOCK_ERROR) { pr_info("crashkernel reservation failed - No suitable area found.\n"); return; } } else { unsigned long long start; - start = find_e820_area(crash_base, ULONG_MAX, crash_size, - 1<<20); + start = memblock_find_in_range(crash_base, + crash_base + crash_size, crash_size, 1<<20); if (start != crash_base) { pr_info("crashkernel reservation failed - memory is in use.\n"); return; } } - reserve_early(crash_base, crash_base + crash_size, "CRASH KERNEL"); + memblock_x86_reserve_range(crash_base, crash_base + crash_size, "CRASH KERNEL"); printk(KERN_INFO "Reserving %ldMB of memory at %ldMB " "for crashkernel (System RAM: %ldMB)\n", @@ -615,82 +619,10 @@ static __init void reserve_ibft_region(void) addr = find_ibft_region(&size); if (size) - reserve_early_overlap_ok(addr, addr + size, "ibft"); + memblock_x86_reserve_range(addr, addr + size, "* ibft"); } -#ifdef CONFIG_X86_RESERVE_LOW_64K -static int __init dmi_low_memory_corruption(const struct dmi_system_id *d) -{ - printk(KERN_NOTICE - "%s detected: BIOS may corrupt low RAM, working around it.\n", - d->ident); - - e820_update_range(0, 0x10000, E820_RAM, E820_RESERVED); - sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); - - return 0; -} -#endif - -/* List of systems that have known low memory corruption BIOS problems */ -static struct dmi_system_id __initdata bad_bios_dmi_table[] = { -#ifdef CONFIG_X86_RESERVE_LOW_64K - { - .callback = dmi_low_memory_corruption, - .ident = "AMI BIOS", - .matches = { - DMI_MATCH(DMI_BIOS_VENDOR, "American Megatrends Inc."), - }, - }, - { - .callback = dmi_low_memory_corruption, - .ident = "Phoenix BIOS", - .matches = { - DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies"), - }, - }, - { - .callback = dmi_low_memory_corruption, - .ident = "Phoenix/MSC BIOS", - .matches = { - DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix/MSC"), - }, - }, - /* - * AMI BIOS with low memory corruption was found on Intel DG45ID and - * DG45FC boards. - * It has a different DMI_BIOS_VENDOR = "Intel Corp.", for now we will - * match only DMI_BOARD_NAME and see if there is more bad products - * with this vendor. - */ - { - .callback = dmi_low_memory_corruption, - .ident = "AMI BIOS", - .matches = { - DMI_MATCH(DMI_BOARD_NAME, "DG45ID"), - }, - }, - { - .callback = dmi_low_memory_corruption, - .ident = "AMI BIOS", - .matches = { - DMI_MATCH(DMI_BOARD_NAME, "DG45FC"), - }, - }, - /* - * The Dell Inspiron Mini 1012 has DMI_BIOS_VENDOR = "Dell Inc.", so - * match on the product name. - */ - { - .callback = dmi_low_memory_corruption, - .ident = "Phoenix BIOS", - .matches = { - DMI_MATCH(DMI_PRODUCT_NAME, "Inspiron 1012"), - }, - }, -#endif - {} -}; +static unsigned reserve_low = CONFIG_X86_RESERVE_LOW << 10; static void __init trim_bios_range(void) { @@ -698,8 +630,14 @@ static void __init trim_bios_range(void) * A special case is the first 4Kb of memory; * This is a BIOS owned area, not kernel ram, but generally * not listed as such in the E820 table. + * + * This typically reserves additional memory (64KiB by default) + * since some BIOSes are known to corrupt low memory. See the + * Kconfig help text for X86_RESERVE_LOW. */ - e820_update_range(0, PAGE_SIZE, E820_RAM, E820_RESERVED); + e820_update_range(0, ALIGN(reserve_low, PAGE_SIZE), + E820_RAM, E820_RESERVED); + /* * special case: Some BIOSen report the PC BIOS * area (640->1Mb) as ram even though it is not. @@ -709,6 +647,37 @@ static void __init trim_bios_range(void) sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); } +static int __init parse_reservelow(char *p) +{ + unsigned long long size; + + if (!p) + return -EINVAL; + + size = memparse(p, &p); + + if (size < 4096) + size = 4096; + + if (size > 640*1024) + size = 640*1024; + + reserve_low = size; + + return 0; +} + +early_param("reservelow", parse_reservelow); + +static u64 __init get_max_mapped(void) +{ + u64 end = max_pfn_mapped; + + end <<= PAGE_SHIFT; + + return end; +} + /* * Determine if we were loaded by an EFI loader. If so, then we have also been * passed the efi memmap, systab, etc., so we should use these data structures @@ -726,18 +695,30 @@ void __init setup_arch(char **cmdline_p) { int acpi = 0; int k8 = 0; + unsigned long flags; #ifdef CONFIG_X86_32 memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); visws_early_detect(); + + /* + * copy kernel address range established so far and switch + * to the proper swapper page table + */ + clone_pgd_range(swapper_pg_dir + KERNEL_PGD_BOUNDARY, + initial_page_table + KERNEL_PGD_BOUNDARY, + KERNEL_PGD_PTRS); + + load_cr3(swapper_pg_dir); + __flush_tlb_all(); #else printk(KERN_INFO "Command line: %s\n", boot_command_line); #endif - /* VMI may relocate the fixmap; do this before touching ioremap area */ - vmi_init(); - - /* OFW also may relocate the fixmap */ + /* + * If we have OLPC OFW, we might end up relocating the fixmap due to + * reserve_top(), so do this before touching the ioremap area. + */ olpc_ofw_detect(); early_trap_init(); @@ -782,7 +763,7 @@ void __init setup_arch(char **cmdline_p) #endif 4)) { efi_enabled = 1; - efi_reserve_early(); + efi_memblock_x86_reserve_range(); } #endif @@ -838,11 +819,8 @@ void __init setup_arch(char **cmdline_p) x86_report_nx(); - /* Must be before kernel pagetables are setup */ - vmi_activate(); - /* after early param, so could get panic from serial */ - reserve_early_setup_data(); + memblock_x86_reserve_range_setup_data(); if (acpi_mps_check()) { #ifdef CONFIG_X86_LOCAL_APIC @@ -863,8 +841,6 @@ void __init setup_arch(char **cmdline_p) dmi_scan_machine(); - 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. @@ -897,8 +873,6 @@ void __init setup_arch(char **cmdline_p) */ max_pfn = e820_end_of_ram_pfn(); - /* preallocate 4k for mptable mpc */ - early_reserve_e820_mpc_new(); /* update e820 for memory not covered by WB MTRRs */ mtrr_bp_init(); if (mtrr_trim_uncached_memory(max_pfn)) @@ -920,18 +894,8 @@ void __init setup_arch(char **cmdline_p) max_low_pfn = max_pfn; high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1; - max_pfn_mapped = KERNEL_IMAGE_SIZE >> PAGE_SHIFT; #endif -#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION - setup_bios_corruption_check(); -#endif - - printk(KERN_DEBUG "initial memory mapped : 0 - %08lx\n", - max_pfn_mapped<<PAGE_SHIFT); - - reserve_brk(); - /* * Find and reserve possible boot-time SMP configuration: */ @@ -939,6 +903,26 @@ void __init setup_arch(char **cmdline_p) reserve_ibft_region(); + /* + * Need to conclude brk, before memblock_x86_fill() + * it could use memblock_find_in_range, could overlap with + * brk area. + */ + reserve_brk(); + + memblock.current_limit = get_max_mapped(); + memblock_x86_fill(); + + /* preallocate 4k for mptable mpc */ + early_reserve_e820_mpc_new(); + +#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION + setup_bios_corruption_check(); +#endif + + printk(KERN_DEBUG "initial memory mapped : 0 - %08lx\n", + max_pfn_mapped<<PAGE_SHIFT); + reserve_trampoline_memory(); #ifdef CONFIG_ACPI_SLEEP @@ -962,6 +946,7 @@ void __init setup_arch(char **cmdline_p) max_low_pfn = max_pfn; } #endif + memblock.current_limit = get_max_mapped(); /* * NOTE: On x86-32, only from this point on, fixmaps are ready for use. @@ -1000,10 +985,7 @@ void __init setup_arch(char **cmdline_p) #endif initmem_init(0, max_pfn, acpi, k8); -#ifndef CONFIG_NO_BOOTMEM - early_res_to_bootmem(0, max_low_pfn<<PAGE_SHIFT); -#endif - + memblock_find_dma_reserve(); dma32_reserve_bootmem(); #ifdef CONFIG_KVM_CLOCK @@ -1014,6 +996,13 @@ void __init setup_arch(char **cmdline_p) paging_init(); x86_init.paging.pagetable_setup_done(swapper_pg_dir); +#ifdef CONFIG_X86_32 + /* sync back kernel address range */ + clone_pgd_range(initial_page_table + KERNEL_PGD_BOUNDARY, + swapper_pg_dir + KERNEL_PGD_BOUNDARY, + KERNEL_PGD_PTRS); +#endif + tboot_probe(); #ifdef CONFIG_X86_64 @@ -1069,6 +1058,10 @@ void __init setup_arch(char **cmdline_p) x86_init.oem.banner(); mcheck_init(); + + local_irq_save(flags); + arch_init_ideal_nop5(); + local_irq_restore(flags); } #ifdef CONFIG_X86_32 diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index a60df9ae645..002b79685f7 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -131,13 +131,7 @@ static void * __init pcpu_fc_alloc(unsigned int cpu, size_t size, size_t align) static void __init pcpu_fc_free(void *ptr, size_t size) { -#ifdef CONFIG_NO_BOOTMEM - u64 start = __pa(ptr); - u64 end = start + size; - free_early_partial(start, end); -#else free_bootmem(__pa(ptr), size); -#endif } static int __init pcpu_cpu_distance(unsigned int from, unsigned int to) @@ -253,7 +247,7 @@ void __init setup_per_cpu_areas(void) * Up to this point, the boot CPU has been using .init.data * area. Reload any changed state for the boot CPU. */ - if (cpu == boot_cpu_id) + if (!cpu) switch_to_new_gdt(cpu); } diff --git a/arch/x86/kernel/sfi.c b/arch/x86/kernel/sfi.c index cb22acf3ed0..dd4c281ffe5 100644 --- a/arch/x86/kernel/sfi.c +++ b/arch/x86/kernel/sfi.c @@ -34,7 +34,7 @@ #ifdef CONFIG_X86_LOCAL_APIC static unsigned long sfi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE; -void __init mp_sfi_register_lapic_address(unsigned long address) +static void __init mp_sfi_register_lapic_address(unsigned long address) { mp_lapic_addr = address; @@ -46,7 +46,7 @@ void __init mp_sfi_register_lapic_address(unsigned long address) } /* All CPUs enumerated by SFI must be present and enabled */ -void __cpuinit mp_sfi_register_lapic(u8 id) +static void __cpuinit mp_sfi_register_lapic(u8 id) { if (MAX_APICS - id <= 0) { pr_warning("Processor #%d invalid (max %d)\n", diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index a5e928b0cb5..6af118511b4 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -62,7 +62,7 @@ #include <asm/pgtable.h> #include <asm/tlbflush.h> #include <asm/mtrr.h> -#include <asm/vmi.h> +#include <asm/mwait.h> #include <asm/apic.h> #include <asm/setup.h> #include <asm/uv/uv.h> @@ -73,7 +73,6 @@ #ifdef CONFIG_X86_32 u8 apicid_2_node[MAX_APICID]; -static int low_mappings; #endif /* State of each CPU */ @@ -91,6 +90,25 @@ DEFINE_PER_CPU(int, cpu_state) = { 0 }; static DEFINE_PER_CPU(struct task_struct *, idle_thread_array); #define get_idle_for_cpu(x) (per_cpu(idle_thread_array, x)) #define set_idle_for_cpu(x, p) (per_cpu(idle_thread_array, x) = (p)) + +/* + * We need this for trampoline_base protection from concurrent accesses when + * off- and onlining cores wildly. + */ +static DEFINE_MUTEX(x86_cpu_hotplug_driver_mutex); + +void cpu_hotplug_driver_lock() +{ + mutex_lock(&x86_cpu_hotplug_driver_mutex); +} + +void cpu_hotplug_driver_unlock() +{ + mutex_unlock(&x86_cpu_hotplug_driver_mutex); +} + +ssize_t arch_cpu_probe(const char *buf, size_t count) { return -1; } +ssize_t arch_cpu_release(const char *buf, size_t count) { return -1; } #else static struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ; #define get_idle_for_cpu(x) (idle_thread_array[(x)]) @@ -281,11 +299,16 @@ notrace static void __cpuinit start_secondary(void *unused) * fragile that we want to limit the things done here to the * most necessary things. */ - vmi_bringup(); cpu_init(); preempt_disable(); smp_callin(); +#ifdef CONFIG_X86_32 + /* switch away from the initial page table */ + load_cr3(swapper_pg_dir); + __flush_tlb_all(); +#endif + /* otherwise gcc will move up smp_processor_id before the cpu_init */ barrier(); /* @@ -294,17 +317,11 @@ notrace static void __cpuinit start_secondary(void *unused) check_tsc_sync_target(); if (nmi_watchdog == NMI_IO_APIC) { - legacy_pic->chip->mask(0); + legacy_pic->mask(0); enable_NMI_through_LVT0(); - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); } -#ifdef CONFIG_X86_32 - while (low_mappings) - cpu_relax(); - __flush_tlb_all(); -#endif - /* This must be done before setting cpu_online_mask */ set_cpu_sibling_map(raw_smp_processor_id()); wmb(); @@ -373,6 +390,19 @@ void __cpuinit smp_store_cpu_info(int id) identify_secondary_cpu(c); } +static void __cpuinit link_thread_siblings(int cpu1, int cpu2) +{ + struct cpuinfo_x86 *c1 = &cpu_data(cpu1); + struct cpuinfo_x86 *c2 = &cpu_data(cpu2); + + cpumask_set_cpu(cpu1, cpu_sibling_mask(cpu2)); + cpumask_set_cpu(cpu2, cpu_sibling_mask(cpu1)); + cpumask_set_cpu(cpu1, cpu_core_mask(cpu2)); + cpumask_set_cpu(cpu2, cpu_core_mask(cpu1)); + cpumask_set_cpu(cpu1, c2->llc_shared_map); + cpumask_set_cpu(cpu2, c1->llc_shared_map); +} + void __cpuinit set_cpu_sibling_map(int cpu) { @@ -385,14 +415,13 @@ void __cpuinit set_cpu_sibling_map(int cpu) for_each_cpu(i, cpu_sibling_setup_mask) { struct cpuinfo_x86 *o = &cpu_data(i); - if (c->phys_proc_id == o->phys_proc_id && - c->cpu_core_id == o->cpu_core_id) { - cpumask_set_cpu(i, cpu_sibling_mask(cpu)); - cpumask_set_cpu(cpu, cpu_sibling_mask(i)); - cpumask_set_cpu(i, cpu_core_mask(cpu)); - cpumask_set_cpu(cpu, cpu_core_mask(i)); - cpumask_set_cpu(i, c->llc_shared_map); - cpumask_set_cpu(cpu, o->llc_shared_map); + if (cpu_has(c, X86_FEATURE_TOPOEXT)) { + if (c->phys_proc_id == o->phys_proc_id && + c->compute_unit_id == o->compute_unit_id) + link_thread_siblings(cpu, i); + } else if (c->phys_proc_id == o->phys_proc_id && + c->cpu_core_id == o->cpu_core_id) { + link_thread_siblings(cpu, i); } } } else { @@ -897,20 +926,7 @@ int __cpuinit native_cpu_up(unsigned int cpu) per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; -#ifdef CONFIG_X86_32 - /* init low mem mapping */ - clone_pgd_range(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, - min_t(unsigned long, KERNEL_PGD_PTRS, KERNEL_PGD_BOUNDARY)); - flush_tlb_all(); - low_mappings = 1; - err = do_boot_cpu(apicid, cpu); - - zap_low_mappings(false); - low_mappings = 0; -#else - err = do_boot_cpu(apicid, cpu); -#endif if (err) { pr_debug("do_boot_cpu failed %d\n", err); return -EIO; @@ -1096,8 +1112,6 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) } set_cpu_sibling_map(0); - enable_IR_x2apic(); - default_setup_apic_routing(); if (smp_sanity_check(max_cpus) < 0) { printk(KERN_INFO "SMP disabled\n"); @@ -1105,6 +1119,8 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) goto out; } + default_setup_apic_routing(); + preempt_disable(); if (read_apic_id() != boot_cpu_physical_apicid) { panic("Boot APIC ID in local APIC unexpected (%d vs %d)", @@ -1370,11 +1386,88 @@ void play_dead_common(void) local_irq_disable(); } +/* + * We need to flush the caches before going to sleep, lest we have + * dirty data in our caches when we come back up. + */ +static inline void mwait_play_dead(void) +{ + unsigned int eax, ebx, ecx, edx; + unsigned int highest_cstate = 0; + unsigned int highest_subcstate = 0; + int i; + void *mwait_ptr; + + if (!cpu_has(¤t_cpu_data, X86_FEATURE_MWAIT)) + return; + if (!cpu_has(¤t_cpu_data, X86_FEATURE_CLFLSH)) + return; + if (current_cpu_data.cpuid_level < CPUID_MWAIT_LEAF) + return; + + eax = CPUID_MWAIT_LEAF; + ecx = 0; + native_cpuid(&eax, &ebx, &ecx, &edx); + + /* + * eax will be 0 if EDX enumeration is not valid. + * Initialized below to cstate, sub_cstate value when EDX is valid. + */ + if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED)) { + eax = 0; + } else { + edx >>= MWAIT_SUBSTATE_SIZE; + for (i = 0; i < 7 && edx; i++, edx >>= MWAIT_SUBSTATE_SIZE) { + if (edx & MWAIT_SUBSTATE_MASK) { + highest_cstate = i; + highest_subcstate = edx & MWAIT_SUBSTATE_MASK; + } + } + eax = (highest_cstate << MWAIT_SUBSTATE_SIZE) | + (highest_subcstate - 1); + } + + /* + * This should be a memory location in a cache line which is + * unlikely to be touched by other processors. The actual + * content is immaterial as it is not actually modified in any way. + */ + mwait_ptr = ¤t_thread_info()->flags; + + wbinvd(); + + while (1) { + /* + * The CLFLUSH is a workaround for erratum AAI65 for + * the Xeon 7400 series. It's not clear it is actually + * needed, but it should be harmless in either case. + * The WBINVD is insufficient due to the spurious-wakeup + * case where we return around the loop. + */ + clflush(mwait_ptr); + __monitor(mwait_ptr, 0, 0); + mb(); + __mwait(eax, 0); + } +} + +static inline void hlt_play_dead(void) +{ + if (current_cpu_data.x86 >= 4) + wbinvd(); + + while (1) { + native_halt(); + } +} + void native_play_dead(void) { play_dead_common(); tboot_shutdown(TB_SHUTDOWN_WFS); - wbinvd_halt(); + + mwait_play_dead(); /* Only returns on failure */ + hlt_play_dead(); } #else /* ... !CONFIG_HOTPLUG_CPU */ diff --git a/arch/x86/kernel/sys_i386_32.c b/arch/x86/kernel/sys_i386_32.c index 196552bb412..0b0cb5fede1 100644 --- a/arch/x86/kernel/sys_i386_32.c +++ b/arch/x86/kernel/sys_i386_32.c @@ -28,11 +28,13 @@ * Do a system call from kernel instead of calling sys_execve so we * end up with proper pt_regs. */ -int kernel_execve(const char *filename, char *const argv[], char *const envp[]) +int kernel_execve(const char *filename, + const char *const argv[], + const char *const envp[]) { long __res; - asm volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" + asm volatile ("int $0x80" : "=a" (__res) - : "0" (__NR_execve), "ri" (filename), "c" (argv), "d" (envp) : "memory"); + : "0" (__NR_execve), "b" (filename), "c" (argv), "d" (envp) : "memory"); return __res; } diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 7fea555929e..20ea20a39e2 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -8,6 +8,7 @@ */ #include <linux/seq_file.h> #include <linux/proc_fs.h> +#include <linux/debugfs.h> #include <linux/kernel.h> #include <linux/slab.h> @@ -22,19 +23,37 @@ #include <asm/irq_vectors.h> #include <asm/timer.h> -struct msg_desc { - struct bau_payload_queue_entry *msg; - int msg_slot; - int sw_ack_slot; - struct bau_payload_queue_entry *va_queue_first; - struct bau_payload_queue_entry *va_queue_last; +/* timeouts in nanoseconds (indexed by UVH_AGING_PRESCALE_SEL urgency7 30:28) */ +static int timeout_base_ns[] = { + 20, + 160, + 1280, + 10240, + 81920, + 655360, + 5242880, + 167772160 }; - -#define UV_INTD_SOFT_ACK_TIMEOUT_PERIOD 0x000000000bUL - -static int uv_bau_max_concurrent __read_mostly; - +static int timeout_us; static int nobau; +static int baudisabled; +static spinlock_t disable_lock; +static cycles_t congested_cycles; + +/* tunables: */ +static int max_bau_concurrent = MAX_BAU_CONCURRENT; +static int max_bau_concurrent_constant = MAX_BAU_CONCURRENT; +static int plugged_delay = PLUGGED_DELAY; +static int plugsb4reset = PLUGSB4RESET; +static int timeoutsb4reset = TIMEOUTSB4RESET; +static int ipi_reset_limit = IPI_RESET_LIMIT; +static int complete_threshold = COMPLETE_THRESHOLD; +static int congested_response_us = CONGESTED_RESPONSE_US; +static int congested_reps = CONGESTED_REPS; +static int congested_period = CONGESTED_PERIOD; +static struct dentry *tunables_dir; +static struct dentry *tunables_file; + static int __init setup_nobau(char *arg) { nobau = 1; @@ -52,10 +71,6 @@ static DEFINE_PER_CPU(struct ptc_stats, ptcstats); static DEFINE_PER_CPU(struct bau_control, bau_control); static DEFINE_PER_CPU(cpumask_var_t, uv_flush_tlb_mask); -struct reset_args { - int sender; -}; - /* * Determine the first node on a uvhub. 'Nodes' are used for kernel * memory allocation. @@ -126,7 +141,7 @@ static inline void uv_bau_process_retry_msg(struct msg_desc *mdp, struct ptc_stats *stat; msg = mdp->msg; - stat = &per_cpu(ptcstats, bcp->cpu); + stat = bcp->statp; stat->d_retries++; /* * cancel any message from msg+1 to the retry itself @@ -146,15 +161,14 @@ static inline void uv_bau_process_retry_msg(struct msg_desc *mdp, slot2 = msg2 - mdp->va_queue_first; mmr = uv_read_local_mmr (UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE); - msg_res = ((msg2->sw_ack_vector << 8) | - msg2->sw_ack_vector); + msg_res = msg2->sw_ack_vector; /* * This is a message retry; clear the resources held * by the previous message only if they timed out. * If it has not timed out we have an unexpected * situation to report. */ - if (mmr & (msg_res << 8)) { + if (mmr & (msg_res << UV_SW_ACK_NPENDING)) { /* * is the resource timed out? * make everyone ignore the cancelled message. @@ -164,9 +178,9 @@ static inline void uv_bau_process_retry_msg(struct msg_desc *mdp, cancel_count++; uv_write_local_mmr( UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, - (msg_res << 8) | msg_res); - } else - printk(KERN_INFO "note bau retry: no effect\n"); + (msg_res << UV_SW_ACK_NPENDING) | + msg_res); + } } } if (!cancel_count) @@ -190,7 +204,7 @@ static void uv_bau_process_message(struct msg_desc *mdp, * This must be a normal message, or retry of a normal message */ msg = mdp->msg; - stat = &per_cpu(ptcstats, bcp->cpu); + stat = bcp->statp; if (msg->address == TLB_FLUSH_ALL) { local_flush_tlb(); stat->d_alltlb++; @@ -274,7 +288,7 @@ uv_do_reset(void *ptr) bcp = &per_cpu(bau_control, smp_processor_id()); rap = (struct reset_args *)ptr; - stat = &per_cpu(ptcstats, bcp->cpu); + stat = bcp->statp; stat->d_resets++; /* @@ -302,13 +316,13 @@ uv_do_reset(void *ptr) */ mmr = uv_read_local_mmr (UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE); - msg_res = ((msg->sw_ack_vector << 8) | - msg->sw_ack_vector); + msg_res = msg->sw_ack_vector; if (mmr & msg_res) { stat->d_rcanceled++; uv_write_local_mmr( UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, - msg_res); + (msg_res << UV_SW_ACK_NPENDING) | + msg_res); } } } @@ -386,17 +400,12 @@ static int uv_wait_completion(struct bau_desc *bau_desc, unsigned long mmr_offset, int right_shift, int this_cpu, struct bau_control *bcp, struct bau_control *smaster, long try) { - int relaxes = 0; unsigned long descriptor_status; - unsigned long mmr; - unsigned long mask; cycles_t ttime; - cycles_t timeout_time; - struct ptc_stats *stat = &per_cpu(ptcstats, this_cpu); + struct ptc_stats *stat = bcp->statp; struct bau_control *hmaster; hmaster = bcp->uvhub_master; - timeout_time = get_cycles() + bcp->timeout_interval; /* spin on the status MMR, waiting for it to go idle */ while ((descriptor_status = (((unsigned long) @@ -423,7 +432,8 @@ static int uv_wait_completion(struct bau_desc *bau_desc, * pending. In that case hardware returns the * ERROR that looks like a destination timeout. */ - if (cycles_2_us(ttime - bcp->send_message) < BIOS_TO) { + if (cycles_2_us(ttime - bcp->send_message) < + timeout_us) { bcp->conseccompletes = 0; return FLUSH_RETRY_PLUGGED; } @@ -435,26 +445,6 @@ static int uv_wait_completion(struct bau_desc *bau_desc, * descriptor_status is still BUSY */ cpu_relax(); - relaxes++; - if (relaxes >= 10000) { - relaxes = 0; - if (get_cycles() > timeout_time) { - quiesce_local_uvhub(hmaster); - - /* single-thread the register change */ - spin_lock(&hmaster->masks_lock); - mmr = uv_read_local_mmr(mmr_offset); - mask = 0UL; - mask |= (3UL < right_shift); - mask = ~mask; - mmr &= mask; - uv_write_local_mmr(mmr_offset, mmr); - spin_unlock(&hmaster->masks_lock); - end_uvhub_quiesce(hmaster); - stat->s_busy++; - return FLUSH_GIVEUP; - } - } } } bcp->conseccompletes++; @@ -494,56 +484,116 @@ static inline int atomic_inc_unless_ge(spinlock_t *lock, atomic_t *v, int u) return 1; } +/* + * Our retries are blocked by all destination swack resources being + * in use, and a timeout is pending. In that case hardware immediately + * returns the ERROR that looks like a destination timeout. + */ +static void +destination_plugged(struct bau_desc *bau_desc, struct bau_control *bcp, + struct bau_control *hmaster, struct ptc_stats *stat) +{ + udelay(bcp->plugged_delay); + bcp->plugged_tries++; + if (bcp->plugged_tries >= bcp->plugsb4reset) { + bcp->plugged_tries = 0; + quiesce_local_uvhub(hmaster); + spin_lock(&hmaster->queue_lock); + uv_reset_with_ipi(&bau_desc->distribution, bcp->cpu); + spin_unlock(&hmaster->queue_lock); + end_uvhub_quiesce(hmaster); + bcp->ipi_attempts++; + stat->s_resets_plug++; + } +} + +static void +destination_timeout(struct bau_desc *bau_desc, struct bau_control *bcp, + struct bau_control *hmaster, struct ptc_stats *stat) +{ + hmaster->max_bau_concurrent = 1; + bcp->timeout_tries++; + if (bcp->timeout_tries >= bcp->timeoutsb4reset) { + bcp->timeout_tries = 0; + quiesce_local_uvhub(hmaster); + spin_lock(&hmaster->queue_lock); + uv_reset_with_ipi(&bau_desc->distribution, bcp->cpu); + spin_unlock(&hmaster->queue_lock); + end_uvhub_quiesce(hmaster); + bcp->ipi_attempts++; + stat->s_resets_timeout++; + } +} + +/* + * Completions are taking a very long time due to a congested numalink + * network. + */ +static void +disable_for_congestion(struct bau_control *bcp, struct ptc_stats *stat) +{ + int tcpu; + struct bau_control *tbcp; + + /* let only one cpu do this disabling */ + spin_lock(&disable_lock); + if (!baudisabled && bcp->period_requests && + ((bcp->period_time / bcp->period_requests) > congested_cycles)) { + /* it becomes this cpu's job to turn on the use of the + BAU again */ + baudisabled = 1; + bcp->set_bau_off = 1; + bcp->set_bau_on_time = get_cycles() + + sec_2_cycles(bcp->congested_period); + stat->s_bau_disabled++; + for_each_present_cpu(tcpu) { + tbcp = &per_cpu(bau_control, tcpu); + tbcp->baudisabled = 1; + } + } + spin_unlock(&disable_lock); +} + /** * uv_flush_send_and_wait * * Send a broadcast and wait for it to complete. * - * The flush_mask contains the cpus the broadcast is to be sent to, plus + * The flush_mask contains the cpus the broadcast is to be sent to including * cpus that are on the local uvhub. * - * Returns NULL if all flushing represented in the mask was done. The mask - * is zeroed. - * Returns @flush_mask if some remote flushing remains to be done. The - * mask will have some bits still set, representing any cpus on the local - * uvhub (not current cpu) and any on remote uvhubs if the broadcast failed. + * Returns 0 if all flushing represented in the mask was done. + * Returns 1 if it gives up entirely and the original cpu mask is to be + * returned to the kernel. */ -const struct cpumask *uv_flush_send_and_wait(struct bau_desc *bau_desc, - struct cpumask *flush_mask, - struct bau_control *bcp) +int uv_flush_send_and_wait(struct bau_desc *bau_desc, + struct cpumask *flush_mask, struct bau_control *bcp) { int right_shift; - int uvhub; - int bit; int completion_status = 0; int seq_number = 0; long try = 0; int cpu = bcp->uvhub_cpu; int this_cpu = bcp->cpu; - int this_uvhub = bcp->uvhub; unsigned long mmr_offset; unsigned long index; cycles_t time1; cycles_t time2; - struct ptc_stats *stat = &per_cpu(ptcstats, bcp->cpu); + cycles_t elapsed; + struct ptc_stats *stat = bcp->statp; struct bau_control *smaster = bcp->socket_master; struct bau_control *hmaster = bcp->uvhub_master; - /* - * Spin here while there are hmaster->max_concurrent or more active - * descriptors. This is the per-uvhub 'throttle'. - */ if (!atomic_inc_unless_ge(&hmaster->uvhub_lock, &hmaster->active_descriptor_count, - hmaster->max_concurrent)) { + hmaster->max_bau_concurrent)) { stat->s_throttles++; do { cpu_relax(); } while (!atomic_inc_unless_ge(&hmaster->uvhub_lock, &hmaster->active_descriptor_count, - hmaster->max_concurrent)); + hmaster->max_bau_concurrent)); } - while (hmaster->uvhub_quiesce) cpu_relax(); @@ -557,23 +607,10 @@ const struct cpumask *uv_flush_send_and_wait(struct bau_desc *bau_desc, } time1 = get_cycles(); do { - /* - * Every message from any given cpu gets a unique message - * sequence number. But retries use that same number. - * Our message may have timed out at the destination because - * all sw-ack resources are in use and there is a timeout - * pending there. In that case, our last send never got - * placed into the queue and we need to persist until it - * does. - * - * Make any retry a type MSG_RETRY so that the destination will - * free any resource held by a previous message from this cpu. - */ if (try == 0) { - /* use message type set by the caller the first time */ + bau_desc->header.msg_type = MSG_REGULAR; seq_number = bcp->message_number++; } else { - /* use RETRY type on all the rest; same sequence */ bau_desc->header.msg_type = MSG_RETRY; stat->s_retry_messages++; } @@ -581,50 +618,17 @@ const struct cpumask *uv_flush_send_and_wait(struct bau_desc *bau_desc, index = (1UL << UVH_LB_BAU_SB_ACTIVATION_CONTROL_PUSH_SHFT) | bcp->uvhub_cpu; bcp->send_message = get_cycles(); - uv_write_local_mmr(UVH_LB_BAU_SB_ACTIVATION_CONTROL, index); - try++; completion_status = uv_wait_completion(bau_desc, mmr_offset, right_shift, this_cpu, bcp, smaster, try); if (completion_status == FLUSH_RETRY_PLUGGED) { - /* - * Our retries may be blocked by all destination swack - * resources being consumed, and a timeout pending. In - * that case hardware immediately returns the ERROR - * that looks like a destination timeout. - */ - udelay(TIMEOUT_DELAY); - bcp->plugged_tries++; - if (bcp->plugged_tries >= PLUGSB4RESET) { - bcp->plugged_tries = 0; - quiesce_local_uvhub(hmaster); - spin_lock(&hmaster->queue_lock); - uv_reset_with_ipi(&bau_desc->distribution, - this_cpu); - spin_unlock(&hmaster->queue_lock); - end_uvhub_quiesce(hmaster); - bcp->ipi_attempts++; - stat->s_resets_plug++; - } + destination_plugged(bau_desc, bcp, hmaster, stat); } else if (completion_status == FLUSH_RETRY_TIMEOUT) { - hmaster->max_concurrent = 1; - bcp->timeout_tries++; - udelay(TIMEOUT_DELAY); - if (bcp->timeout_tries >= TIMEOUTSB4RESET) { - bcp->timeout_tries = 0; - quiesce_local_uvhub(hmaster); - spin_lock(&hmaster->queue_lock); - uv_reset_with_ipi(&bau_desc->distribution, - this_cpu); - spin_unlock(&hmaster->queue_lock); - end_uvhub_quiesce(hmaster); - bcp->ipi_attempts++; - stat->s_resets_timeout++; - } + destination_timeout(bau_desc, bcp, hmaster, stat); } - if (bcp->ipi_attempts >= 3) { + if (bcp->ipi_attempts >= bcp->ipi_reset_limit) { bcp->ipi_attempts = 0; completion_status = FLUSH_GIVEUP; break; @@ -633,49 +637,36 @@ const struct cpumask *uv_flush_send_and_wait(struct bau_desc *bau_desc, } while ((completion_status == FLUSH_RETRY_PLUGGED) || (completion_status == FLUSH_RETRY_TIMEOUT)); time2 = get_cycles(); - - if ((completion_status == FLUSH_COMPLETE) && (bcp->conseccompletes > 5) - && (hmaster->max_concurrent < hmaster->max_concurrent_constant)) - hmaster->max_concurrent++; - - /* - * hold any cpu not timing out here; no other cpu currently held by - * the 'throttle' should enter the activation code - */ + bcp->plugged_tries = 0; + bcp->timeout_tries = 0; + if ((completion_status == FLUSH_COMPLETE) && + (bcp->conseccompletes > bcp->complete_threshold) && + (hmaster->max_bau_concurrent < + hmaster->max_bau_concurrent_constant)) + hmaster->max_bau_concurrent++; while (hmaster->uvhub_quiesce) cpu_relax(); atomic_dec(&hmaster->active_descriptor_count); - - /* guard against cycles wrap */ - if (time2 > time1) - stat->s_time += (time2 - time1); - else - stat->s_requestor--; /* don't count this one */ + if (time2 > time1) { + elapsed = time2 - time1; + stat->s_time += elapsed; + if ((completion_status == FLUSH_COMPLETE) && (try == 1)) { + bcp->period_requests++; + bcp->period_time += elapsed; + if ((elapsed > congested_cycles) && + (bcp->period_requests > bcp->congested_reps)) { + disable_for_congestion(bcp, stat); + } + } + } else + stat->s_requestor--; if (completion_status == FLUSH_COMPLETE && try > 1) stat->s_retriesok++; else if (completion_status == FLUSH_GIVEUP) { - /* - * Cause the caller to do an IPI-style TLB shootdown on - * the target cpu's, all of which are still in the mask. - */ stat->s_giveup++; - return flush_mask; - } - - /* - * Success, so clear the remote cpu's from the mask so we don't - * use the IPI method of shootdown on them. - */ - for_each_cpu(bit, flush_mask) { - uvhub = uv_cpu_to_blade_id(bit); - if (uvhub == this_uvhub) - continue; - cpumask_clear_cpu(bit, flush_mask); + return 1; } - if (!cpumask_empty(flush_mask)) - return flush_mask; - - return NULL; + return 0; } /** @@ -707,70 +698,89 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, struct mm_struct *mm, unsigned long va, unsigned int cpu) { - int remotes; int tcpu; int uvhub; int locals = 0; + int remotes = 0; + int hubs = 0; struct bau_desc *bau_desc; struct cpumask *flush_mask; struct ptc_stats *stat; struct bau_control *bcp; + struct bau_control *tbcp; + /* kernel was booted 'nobau' */ if (nobau) return cpumask; bcp = &per_cpu(bau_control, cpu); + stat = bcp->statp; + + /* bau was disabled due to slow response */ + if (bcp->baudisabled) { + /* the cpu that disabled it must re-enable it */ + if (bcp->set_bau_off) { + if (get_cycles() >= bcp->set_bau_on_time) { + stat->s_bau_reenabled++; + baudisabled = 0; + for_each_present_cpu(tcpu) { + tbcp = &per_cpu(bau_control, tcpu); + tbcp->baudisabled = 0; + tbcp->period_requests = 0; + tbcp->period_time = 0; + } + } + } + return cpumask; + } + /* * Each sending cpu has a per-cpu mask which it fills from the caller's - * cpu mask. Only remote cpus are converted to uvhubs and copied. + * cpu mask. All cpus are converted to uvhubs and copied to the + * activation descriptor. */ flush_mask = (struct cpumask *)per_cpu(uv_flush_tlb_mask, cpu); - /* - * copy cpumask to flush_mask, removing current cpu - * (current cpu should already have been flushed by the caller and - * should never be returned if we return flush_mask) - */ + /* don't actually do a shootdown of the local cpu */ cpumask_andnot(flush_mask, cpumask, cpumask_of(cpu)); if (cpu_isset(cpu, *cpumask)) - locals++; /* current cpu was targeted */ + stat->s_ntargself++; bau_desc = bcp->descriptor_base; bau_desc += UV_ITEMS_PER_DESCRIPTOR * bcp->uvhub_cpu; - bau_uvhubs_clear(&bau_desc->distribution, UV_DISTRIBUTION_SIZE); - remotes = 0; + + /* cpu statistics */ for_each_cpu(tcpu, flush_mask) { uvhub = uv_cpu_to_blade_id(tcpu); - if (uvhub == bcp->uvhub) { - locals++; - continue; - } bau_uvhub_set(uvhub, &bau_desc->distribution); - remotes++; - } - if (remotes == 0) { - /* - * No off_hub flushing; return status for local hub. - * Return the caller's mask if all were local (the current - * cpu may be in that mask). - */ - if (locals) - return cpumask; + if (uvhub == bcp->uvhub) + locals++; else - return NULL; + remotes++; } - stat = &per_cpu(ptcstats, cpu); + if ((locals + remotes) == 0) + return NULL; stat->s_requestor++; - stat->s_ntargcpu += remotes; + stat->s_ntargcpu += remotes + locals; + stat->s_ntargremotes += remotes; + stat->s_ntarglocals += locals; remotes = bau_uvhub_weight(&bau_desc->distribution); - stat->s_ntarguvhub += remotes; - if (remotes >= 16) + + /* uvhub statistics */ + hubs = bau_uvhub_weight(&bau_desc->distribution); + if (locals) { + stat->s_ntarglocaluvhub++; + stat->s_ntargremoteuvhub += (hubs - 1); + } else + stat->s_ntargremoteuvhub += hubs; + stat->s_ntarguvhub += hubs; + if (hubs >= 16) stat->s_ntarguvhub16++; - else if (remotes >= 8) + else if (hubs >= 8) stat->s_ntarguvhub8++; - else if (remotes >= 4) + else if (hubs >= 4) stat->s_ntarguvhub4++; - else if (remotes >= 2) + else if (hubs >= 2) stat->s_ntarguvhub2++; else stat->s_ntarguvhub1++; @@ -779,10 +789,13 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, bau_desc->payload.sending_cpu = cpu; /* - * uv_flush_send_and_wait returns null if all cpu's were messaged, or - * the adjusted flush_mask if any cpu's were not messaged. + * uv_flush_send_and_wait returns 0 if all cpu's were messaged, + * or 1 if it gave up and the original cpumask should be returned. */ - return uv_flush_send_and_wait(bau_desc, flush_mask, bcp); + if (!uv_flush_send_and_wait(bau_desc, flush_mask, bcp)) + return NULL; + else + return cpumask; } /* @@ -810,7 +823,7 @@ void uv_bau_message_interrupt(struct pt_regs *regs) time_start = get_cycles(); bcp = &per_cpu(bau_control, smp_processor_id()); - stat = &per_cpu(ptcstats, smp_processor_id()); + stat = bcp->statp; msgdesc.va_queue_first = bcp->va_queue_first; msgdesc.va_queue_last = bcp->va_queue_last; msg = bcp->bau_msg_head; @@ -908,12 +921,12 @@ static void uv_ptc_seq_stop(struct seq_file *file, void *data) } static inline unsigned long long -millisec_2_cycles(unsigned long millisec) +microsec_2_cycles(unsigned long microsec) { unsigned long ns; unsigned long long cyc; - ns = millisec * 1000; + ns = microsec * 1000; cyc = (ns << CYC2NS_SCALE_FACTOR)/(per_cpu(cyc2ns, smp_processor_id())); return cyc; } @@ -931,15 +944,19 @@ static int uv_ptc_seq_show(struct seq_file *file, void *data) if (!cpu) { seq_printf(file, - "# cpu sent stime numuvhubs numuvhubs16 numuvhubs8 "); + "# cpu sent stime self locals remotes ncpus localhub "); + seq_printf(file, + "remotehub numuvhubs numuvhubs16 numuvhubs8 "); seq_printf(file, - "numuvhubs4 numuvhubs2 numuvhubs1 numcpus dto "); + "numuvhubs4 numuvhubs2 numuvhubs1 dto "); seq_printf(file, "retries rok resetp resett giveup sto bz throt "); seq_printf(file, "sw_ack recv rtime all "); seq_printf(file, - "one mult none retry canc nocan reset rcan\n"); + "one mult none retry canc nocan reset rcan "); + seq_printf(file, + "disable enable\n"); } if (cpu < num_possible_cpus() && cpu_online(cpu)) { stat = &per_cpu(ptcstats, cpu); @@ -947,18 +964,23 @@ static int uv_ptc_seq_show(struct seq_file *file, void *data) seq_printf(file, "cpu %d %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld ", cpu, stat->s_requestor, cycles_2_us(stat->s_time), - stat->s_ntarguvhub, stat->s_ntarguvhub16, + stat->s_ntargself, stat->s_ntarglocals, + stat->s_ntargremotes, stat->s_ntargcpu, + stat->s_ntarglocaluvhub, stat->s_ntargremoteuvhub, + stat->s_ntarguvhub, stat->s_ntarguvhub16); + seq_printf(file, "%ld %ld %ld %ld %ld ", stat->s_ntarguvhub8, stat->s_ntarguvhub4, stat->s_ntarguvhub2, stat->s_ntarguvhub1, - stat->s_ntargcpu, stat->s_dtimeout); + stat->s_dtimeout); seq_printf(file, "%ld %ld %ld %ld %ld %ld %ld %ld ", stat->s_retry_messages, stat->s_retriesok, stat->s_resets_plug, stat->s_resets_timeout, stat->s_giveup, stat->s_stimeout, stat->s_busy, stat->s_throttles); + /* destination side statistics */ seq_printf(file, - "%lx %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld\n", + "%lx %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld ", uv_read_global_mmr64(uv_cpu_to_pnode(cpu), UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE), stat->d_requestee, cycles_2_us(stat->d_time), @@ -966,15 +988,41 @@ static int uv_ptc_seq_show(struct seq_file *file, void *data) stat->d_nomsg, stat->d_retries, stat->d_canceled, stat->d_nocanceled, stat->d_resets, stat->d_rcanceled); + seq_printf(file, "%ld %ld\n", + stat->s_bau_disabled, stat->s_bau_reenabled); } return 0; } /* + * Display the tunables thru debugfs + */ +static ssize_t tunables_read(struct file *file, char __user *userbuf, + size_t count, loff_t *ppos) +{ + char *buf; + int ret; + + buf = kasprintf(GFP_KERNEL, "%s %s %s\n%d %d %d %d %d %d %d %d %d\n", + "max_bau_concurrent plugged_delay plugsb4reset", + "timeoutsb4reset ipi_reset_limit complete_threshold", + "congested_response_us congested_reps congested_period", + max_bau_concurrent, plugged_delay, plugsb4reset, + timeoutsb4reset, ipi_reset_limit, complete_threshold, + congested_response_us, congested_reps, congested_period); + + if (!buf) + return -ENOMEM; + + ret = simple_read_from_buffer(userbuf, count, ppos, buf, strlen(buf)); + kfree(buf); + return ret; +} + +/* * -1: resetf the statistics * 0: display meaning of the statistics - * >0: maximum concurrent active descriptors per uvhub (throttle) */ static ssize_t uv_ptc_proc_write(struct file *file, const char __user *user, size_t count, loff_t *data) @@ -983,7 +1031,6 @@ static ssize_t uv_ptc_proc_write(struct file *file, const char __user *user, long input_arg; char optstr[64]; struct ptc_stats *stat; - struct bau_control *bcp; if (count == 0 || count > sizeof(optstr)) return -EINVAL; @@ -1059,29 +1106,158 @@ static ssize_t uv_ptc_proc_write(struct file *file, const char __user *user, "reset: number of ipi-style reset requests processed\n"); printk(KERN_DEBUG "rcan: number messages canceled by reset requests\n"); + printk(KERN_DEBUG + "disable: number times use of the BAU was disabled\n"); + printk(KERN_DEBUG + "enable: number times use of the BAU was re-enabled\n"); } else if (input_arg == -1) { for_each_present_cpu(cpu) { stat = &per_cpu(ptcstats, cpu); memset(stat, 0, sizeof(struct ptc_stats)); } - } else { - uv_bau_max_concurrent = input_arg; - bcp = &per_cpu(bau_control, smp_processor_id()); - if (uv_bau_max_concurrent < 1 || - uv_bau_max_concurrent > bcp->cpus_in_uvhub) { - printk(KERN_DEBUG - "Error: BAU max concurrent %d; %d is invalid\n", - bcp->max_concurrent, uv_bau_max_concurrent); - return -EINVAL; - } - printk(KERN_DEBUG "Set BAU max concurrent:%d\n", - uv_bau_max_concurrent); - for_each_present_cpu(cpu) { - bcp = &per_cpu(bau_control, cpu); - bcp->max_concurrent = uv_bau_max_concurrent; + } + + return count; +} + +static int local_atoi(const char *name) +{ + int val = 0; + + for (;; name++) { + switch (*name) { + case '0' ... '9': + val = 10*val+(*name-'0'); + break; + default: + return val; } } +} + +/* + * set the tunables + * 0 values reset them to defaults + */ +static ssize_t tunables_write(struct file *file, const char __user *user, + size_t count, loff_t *data) +{ + int cpu; + int cnt = 0; + int val; + char *p; + char *q; + char instr[64]; + struct bau_control *bcp; + + if (count == 0 || count > sizeof(instr)-1) + return -EINVAL; + if (copy_from_user(instr, user, count)) + return -EFAULT; + instr[count] = '\0'; + /* count the fields */ + p = instr + strspn(instr, WHITESPACE); + q = p; + for (; *p; p = q + strspn(q, WHITESPACE)) { + q = p + strcspn(p, WHITESPACE); + cnt++; + if (q == p) + break; + } + if (cnt != 9) { + printk(KERN_INFO "bau tunable error: should be 9 numbers\n"); + return -EINVAL; + } + + p = instr + strspn(instr, WHITESPACE); + q = p; + for (cnt = 0; *p; p = q + strspn(q, WHITESPACE), cnt++) { + q = p + strcspn(p, WHITESPACE); + val = local_atoi(p); + switch (cnt) { + case 0: + if (val == 0) { + max_bau_concurrent = MAX_BAU_CONCURRENT; + max_bau_concurrent_constant = + MAX_BAU_CONCURRENT; + continue; + } + bcp = &per_cpu(bau_control, smp_processor_id()); + if (val < 1 || val > bcp->cpus_in_uvhub) { + printk(KERN_DEBUG + "Error: BAU max concurrent %d is invalid\n", + val); + return -EINVAL; + } + max_bau_concurrent = val; + max_bau_concurrent_constant = val; + continue; + case 1: + if (val == 0) + plugged_delay = PLUGGED_DELAY; + else + plugged_delay = val; + continue; + case 2: + if (val == 0) + plugsb4reset = PLUGSB4RESET; + else + plugsb4reset = val; + continue; + case 3: + if (val == 0) + timeoutsb4reset = TIMEOUTSB4RESET; + else + timeoutsb4reset = val; + continue; + case 4: + if (val == 0) + ipi_reset_limit = IPI_RESET_LIMIT; + else + ipi_reset_limit = val; + continue; + case 5: + if (val == 0) + complete_threshold = COMPLETE_THRESHOLD; + else + complete_threshold = val; + continue; + case 6: + if (val == 0) + congested_response_us = CONGESTED_RESPONSE_US; + else + congested_response_us = val; + continue; + case 7: + if (val == 0) + congested_reps = CONGESTED_REPS; + else + congested_reps = val; + continue; + case 8: + if (val == 0) + congested_period = CONGESTED_PERIOD; + else + congested_period = val; + continue; + } + if (q == p) + break; + } + for_each_present_cpu(cpu) { + bcp = &per_cpu(bau_control, cpu); + bcp->max_bau_concurrent = max_bau_concurrent; + bcp->max_bau_concurrent_constant = max_bau_concurrent; + bcp->plugged_delay = plugged_delay; + bcp->plugsb4reset = plugsb4reset; + bcp->timeoutsb4reset = timeoutsb4reset; + bcp->ipi_reset_limit = ipi_reset_limit; + bcp->complete_threshold = complete_threshold; + bcp->congested_response_us = congested_response_us; + bcp->congested_reps = congested_reps; + bcp->congested_period = congested_period; + } return count; } @@ -1097,6 +1273,11 @@ static int uv_ptc_proc_open(struct inode *inode, struct file *file) return seq_open(file, &uv_ptc_seq_ops); } +static int tunables_open(struct inode *inode, struct file *file) +{ + return 0; +} + static const struct file_operations proc_uv_ptc_operations = { .open = uv_ptc_proc_open, .read = seq_read, @@ -1105,6 +1286,13 @@ static const struct file_operations proc_uv_ptc_operations = { .release = seq_release, }; +static const struct file_operations tunables_fops = { + .open = tunables_open, + .read = tunables_read, + .write = tunables_write, + .llseek = default_llseek, +}; + static int __init uv_ptc_init(void) { struct proc_dir_entry *proc_uv_ptc; @@ -1119,6 +1307,20 @@ static int __init uv_ptc_init(void) UV_PTC_BASENAME); return -EINVAL; } + + tunables_dir = debugfs_create_dir(UV_BAU_TUNABLES_DIR, NULL); + if (!tunables_dir) { + printk(KERN_ERR "unable to create debugfs directory %s\n", + UV_BAU_TUNABLES_DIR); + return -EINVAL; + } + tunables_file = debugfs_create_file(UV_BAU_TUNABLES_FILE, 0600, + tunables_dir, NULL, &tunables_fops); + if (!tunables_file) { + printk(KERN_ERR "unable to create debugfs file %s\n", + UV_BAU_TUNABLES_FILE); + return -EINVAL; + } return 0; } @@ -1259,15 +1461,45 @@ static void __init uv_init_uvhub(int uvhub, int vector) } /* + * We will set BAU_MISC_CONTROL with a timeout period. + * But the BIOS has set UVH_AGING_PRESCALE_SEL and UVH_TRANSACTION_TIMEOUT. + * So the destination timeout period has be be calculated from them. + */ +static int +calculate_destination_timeout(void) +{ + unsigned long mmr_image; + int mult1; + int mult2; + int index; + int base; + int ret; + unsigned long ts_ns; + + mult1 = UV_INTD_SOFT_ACK_TIMEOUT_PERIOD & BAU_MISC_CONTROL_MULT_MASK; + mmr_image = uv_read_local_mmr(UVH_AGING_PRESCALE_SEL); + index = (mmr_image >> BAU_URGENCY_7_SHIFT) & BAU_URGENCY_7_MASK; + mmr_image = uv_read_local_mmr(UVH_TRANSACTION_TIMEOUT); + mult2 = (mmr_image >> BAU_TRANS_SHIFT) & BAU_TRANS_MASK; + base = timeout_base_ns[index]; + ts_ns = base * mult1 * mult2; + ret = ts_ns / 1000; + return ret; +} + +/* * initialize the bau_control structure for each cpu */ -static void uv_init_per_cpu(int nuvhubs) +static void __init uv_init_per_cpu(int nuvhubs) { - int i, j, k; + int i; int cpu; int pnode; int uvhub; + int have_hmaster; short socket = 0; + unsigned short socket_mask; + unsigned char *uvhub_mask; struct bau_control *bcp; struct uvhub_desc *bdp; struct socket_desc *sdp; @@ -1278,7 +1510,7 @@ static void uv_init_per_cpu(int nuvhubs) short cpu_number[16]; }; struct uvhub_desc { - short num_sockets; + unsigned short socket_mask; short num_cpus; short uvhub; short pnode; @@ -1286,57 +1518,84 @@ static void uv_init_per_cpu(int nuvhubs) }; struct uvhub_desc *uvhub_descs; + timeout_us = calculate_destination_timeout(); + uvhub_descs = (struct uvhub_desc *) kmalloc(nuvhubs * sizeof(struct uvhub_desc), GFP_KERNEL); memset(uvhub_descs, 0, nuvhubs * sizeof(struct uvhub_desc)); + uvhub_mask = kzalloc((nuvhubs+7)/8, GFP_KERNEL); for_each_present_cpu(cpu) { bcp = &per_cpu(bau_control, cpu); memset(bcp, 0, sizeof(struct bau_control)); - spin_lock_init(&bcp->masks_lock); - bcp->max_concurrent = uv_bau_max_concurrent; pnode = uv_cpu_hub_info(cpu)->pnode; uvhub = uv_cpu_hub_info(cpu)->numa_blade_id; + *(uvhub_mask + (uvhub/8)) |= (1 << (uvhub%8)); bdp = &uvhub_descs[uvhub]; bdp->num_cpus++; bdp->uvhub = uvhub; bdp->pnode = pnode; - /* time interval to catch a hardware stay-busy bug */ - bcp->timeout_interval = millisec_2_cycles(3); - /* kludge: assume uv_hub.h is constant */ - socket = (cpu_physical_id(cpu)>>5)&1; - if (socket >= bdp->num_sockets) - bdp->num_sockets = socket+1; + /* kludge: 'assuming' one node per socket, and assuming that + disabling a socket just leaves a gap in node numbers */ + socket = (cpu_to_node(cpu) & 1); + bdp->socket_mask |= (1 << socket); sdp = &bdp->socket[socket]; sdp->cpu_number[sdp->num_cpus] = cpu; sdp->num_cpus++; } - socket = 0; - for_each_possible_blade(uvhub) { + for (uvhub = 0; uvhub < nuvhubs; uvhub++) { + if (!(*(uvhub_mask + (uvhub/8)) & (1 << (uvhub%8)))) + continue; + have_hmaster = 0; bdp = &uvhub_descs[uvhub]; - for (i = 0; i < bdp->num_sockets; i++) { - sdp = &bdp->socket[i]; - for (j = 0; j < sdp->num_cpus; j++) { - cpu = sdp->cpu_number[j]; + socket_mask = bdp->socket_mask; + socket = 0; + while (socket_mask) { + if (!(socket_mask & 1)) + goto nextsocket; + sdp = &bdp->socket[socket]; + for (i = 0; i < sdp->num_cpus; i++) { + cpu = sdp->cpu_number[i]; bcp = &per_cpu(bau_control, cpu); bcp->cpu = cpu; - if (j == 0) { + if (i == 0) { smaster = bcp; - if (i == 0) + if (!have_hmaster) { + have_hmaster++; hmaster = bcp; + } } bcp->cpus_in_uvhub = bdp->num_cpus; bcp->cpus_in_socket = sdp->num_cpus; bcp->socket_master = smaster; + bcp->uvhub = bdp->uvhub; bcp->uvhub_master = hmaster; - for (k = 0; k < DEST_Q_SIZE; k++) - bcp->socket_acknowledge_count[k] = 0; - bcp->uvhub_cpu = - uv_cpu_hub_info(cpu)->blade_processor_id; + bcp->uvhub_cpu = uv_cpu_hub_info(cpu)-> + blade_processor_id; } +nextsocket: socket++; + socket_mask = (socket_mask >> 1); } } kfree(uvhub_descs); + kfree(uvhub_mask); + for_each_present_cpu(cpu) { + bcp = &per_cpu(bau_control, cpu); + bcp->baudisabled = 0; + bcp->statp = &per_cpu(ptcstats, cpu); + /* time interval to catch a hardware stay-busy bug */ + bcp->timeout_interval = microsec_2_cycles(2*timeout_us); + bcp->max_bau_concurrent = max_bau_concurrent; + bcp->max_bau_concurrent_constant = max_bau_concurrent; + bcp->plugged_delay = plugged_delay; + bcp->plugsb4reset = plugsb4reset; + bcp->timeoutsb4reset = timeoutsb4reset; + bcp->ipi_reset_limit = ipi_reset_limit; + bcp->complete_threshold = complete_threshold; + bcp->congested_response_us = congested_response_us; + bcp->congested_reps = congested_reps; + bcp->congested_period = congested_period; + } } /* @@ -1361,10 +1620,11 @@ static int __init uv_bau_init(void) zalloc_cpumask_var_node(&per_cpu(uv_flush_tlb_mask, cur_cpu), GFP_KERNEL, cpu_to_node(cur_cpu)); - uv_bau_max_concurrent = MAX_BAU_CONCURRENT; uv_nshift = uv_hub_info->m_val; uv_mmask = (1UL << uv_hub_info->m_val) - 1; nuvhubs = uv_num_possible_blades(); + spin_lock_init(&disable_lock); + congested_cycles = microsec_2_cycles(congested_response_us); uv_init_per_cpu(nuvhubs); @@ -1383,15 +1643,19 @@ static int __init uv_bau_init(void) alloc_intr_gate(vector, uv_bau_message_intr1); for_each_possible_blade(uvhub) { - pnode = uv_blade_to_pnode(uvhub); - /* INIT the bau */ - uv_write_global_mmr64(pnode, UVH_LB_BAU_SB_ACTIVATION_CONTROL, - ((unsigned long)1 << 63)); - mmr = 1; /* should be 1 to broadcast to both sockets */ - uv_write_global_mmr64(pnode, UVH_BAU_DATA_BROADCAST, mmr); + if (uv_blade_nr_possible_cpus(uvhub)) { + pnode = uv_blade_to_pnode(uvhub); + /* INIT the bau */ + uv_write_global_mmr64(pnode, + UVH_LB_BAU_SB_ACTIVATION_CONTROL, + ((unsigned long)1 << 63)); + mmr = 1; /* should be 1 to broadcast to both sockets */ + uv_write_global_mmr64(pnode, UVH_BAU_DATA_BROADCAST, + mmr); + } } return 0; } core_initcall(uv_bau_init); -core_initcall(uv_ptc_init); +fs_initcall(uv_ptc_init); diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c index c652ef62742..a375616d77f 100644 --- a/arch/x86/kernel/trampoline.c +++ b/arch/x86/kernel/trampoline.c @@ -1,7 +1,8 @@ #include <linux/io.h> +#include <linux/memblock.h> #include <asm/trampoline.h> -#include <asm/e820.h> +#include <asm/pgtable.h> #if defined(CONFIG_X86_64) && defined(CONFIG_ACPI_SLEEP) #define __trampinit @@ -16,15 +17,15 @@ unsigned char *__trampinitdata trampoline_base; void __init reserve_trampoline_memory(void) { - unsigned long mem; + phys_addr_t mem; /* Has to be in very low memory so we can execute real-mode AP code. */ - mem = find_e820_area(0, 1<<20, TRAMPOLINE_SIZE, PAGE_SIZE); - if (mem == -1L) + mem = memblock_find_in_range(0, 1<<20, TRAMPOLINE_SIZE, PAGE_SIZE); + if (mem == MEMBLOCK_ERROR) panic("Cannot allocate trampoline\n"); trampoline_base = __va(mem); - reserve_early(mem, mem + TRAMPOLINE_SIZE, "TRAMPOLINE"); + memblock_x86_reserve_range(mem, mem + TRAMPOLINE_SIZE, "TRAMPOLINE"); } /* diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 60788dee0f8..cb838ca42c9 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -575,6 +575,7 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) if (regs->flags & X86_VM_MASK) { handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1); + preempt_conditional_cli(regs); return; } @@ -776,21 +777,10 @@ asmlinkage void math_state_restore(void) } EXPORT_SYMBOL_GPL(math_state_restore); -#ifndef CONFIG_MATH_EMULATION -void math_emulate(struct math_emu_info *info) -{ - printk(KERN_EMERG - "math-emulation not enabled and no coprocessor found.\n"); - printk(KERN_EMERG "killing %s.\n", current->comm); - force_sig(SIGFPE, current); - schedule(); -} -#endif /* CONFIG_MATH_EMULATION */ - dotraplinkage void __kprobes do_device_not_available(struct pt_regs *regs, long error_code) { -#ifdef CONFIG_X86_32 +#ifdef CONFIG_MATH_EMULATION if (read_cr0() & X86_CR0_EM) { struct math_emu_info info = { }; @@ -798,12 +788,12 @@ do_device_not_available(struct pt_regs *regs, long error_code) info.regs = regs; math_emulate(&info); - } else { - math_state_restore(); /* interrupts still off */ - conditional_sti(regs); + return; } -#else - math_state_restore(); +#endif + math_state_restore(); /* interrupts still off */ +#ifdef CONFIG_X86_32 + conditional_sti(regs); #endif } @@ -881,18 +871,6 @@ void __init trap_init(void) #endif #ifdef CONFIG_X86_32 - if (cpu_has_fxsr) { - printk(KERN_INFO "Enabling fast FPU save and restore... "); - set_in_cr4(X86_CR4_OSFXSR); - printk("done.\n"); - } - if (cpu_has_xmm) { - printk(KERN_INFO - "Enabling unmasked SIMD FPU exception support... "); - set_in_cr4(X86_CR4_OSXMMEXCPT); - printk("done.\n"); - } - set_system_trap_gate(SYSCALL_VECTOR, &system_call); set_bit(SYSCALL_VECTOR, used_vectors); #endif diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index ce8e5023933..0c40d8b7241 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -104,10 +104,14 @@ int __init notsc_setup(char *str) __setup("notsc", notsc_setup); +static int no_sched_irq_time; + static int __init tsc_setup(char *str) { if (!strcmp(str, "reliable")) tsc_clocksource_reliable = 1; + if (!strncmp(str, "noirqtime", 9)) + no_sched_irq_time = 1; return 1; } @@ -626,6 +630,44 @@ static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu) local_irq_restore(flags); } +static unsigned long long cyc2ns_suspend; + +void save_sched_clock_state(void) +{ + if (!sched_clock_stable) + return; + + cyc2ns_suspend = sched_clock(); +} + +/* + * Even on processors with invariant TSC, TSC gets reset in some the + * ACPI system sleep states. And in some systems BIOS seem to reinit TSC to + * arbitrary value (still sync'd across cpu's) during resume from such sleep + * states. To cope up with this, recompute the cyc2ns_offset for each cpu so + * that sched_clock() continues from the point where it was left off during + * suspend. + */ +void restore_sched_clock_state(void) +{ + unsigned long long offset; + unsigned long flags; + int cpu; + + if (!sched_clock_stable) + return; + + local_irq_save(flags); + + __get_cpu_var(cyc2ns_offset) = 0; + offset = cyc2ns_suspend - sched_clock(); + + for_each_possible_cpu(cpu) + per_cpu(cyc2ns_offset, cpu) = offset; + + local_irq_restore(flags); +} + #ifdef CONFIG_CPU_FREQ /* Frequency scaling support. Adjust the TSC based timer when the cpu frequency @@ -763,6 +805,7 @@ void mark_tsc_unstable(char *reason) if (!tsc_unstable) { tsc_unstable = 1; sched_clock_stable = 0; + disable_sched_clock_irqtime(); printk(KERN_INFO "Marking TSC unstable due to %s\n", reason); /* Change only the rating, when not registered */ if (clocksource_tsc.mult) @@ -854,60 +897,6 @@ static void __init init_tsc_clocksource(void) clocksource_register_khz(&clocksource_tsc, tsc_khz); } -#ifdef CONFIG_X86_64 -/* - * calibrate_cpu is used on systems with fixed rate TSCs to determine - * processor frequency - */ -#define TICK_COUNT 100000000 -static unsigned long __init calibrate_cpu(void) -{ - int tsc_start, tsc_now; - int i, no_ctr_free; - unsigned long evntsel3 = 0, pmc3 = 0, pmc_now = 0; - unsigned long flags; - - for (i = 0; i < 4; i++) - if (avail_to_resrv_perfctr_nmi_bit(i)) - 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); - rdmsrl(MSR_K7_PERFCTR3, pmc3); - } else { - reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i); - reserve_evntsel_nmi(MSR_K7_EVNTSEL0 + i); - } - local_irq_save(flags); - /* start measuring cycles, incrementing from 0 */ - wrmsrl(MSR_K7_PERFCTR0 + i, 0); - wrmsrl(MSR_K7_EVNTSEL0 + i, 1 << 22 | 3 << 16 | 0x76); - rdtscl(tsc_start); - do { - rdmsrl(MSR_K7_PERFCTR0 + i, pmc_now); - tsc_now = get_cycles(); - } while ((tsc_now - tsc_start) < TICK_COUNT); - - local_irq_restore(flags); - if (no_ctr_free) { - wrmsrl(MSR_K7_EVNTSEL3, 0); - wrmsrl(MSR_K7_PERFCTR3, pmc3); - wrmsrl(MSR_K7_EVNTSEL3, evntsel3); - } else { - release_perfctr_nmi(MSR_K7_PERFCTR0 + i); - release_evntsel_nmi(MSR_K7_EVNTSEL0 + i); - } - - return pmc_now * tsc_khz / (tsc_now - tsc_start); -} -#else -static inline unsigned long calibrate_cpu(void) { return cpu_khz; } -#endif - void __init tsc_init(void) { u64 lpj; @@ -926,10 +915,6 @@ void __init tsc_init(void) return; } - if (cpu_has(&boot_cpu_data, X86_FEATURE_CONSTANT_TSC) && - (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)) - cpu_khz = calibrate_cpu(); - printk("Detected %lu.%03lu MHz processor.\n", (unsigned long)cpu_khz / 1000, (unsigned long)cpu_khz % 1000); @@ -949,6 +934,9 @@ void __init tsc_init(void) /* now allow native_sched_clock() to use rdtsc */ tsc_disabled = 0; + if (!no_sched_irq_time) + enable_sched_clock_irqtime(); + lpj = ((u64)tsc_khz * 1000); do_div(lpj, HZ); lpj_fine = lpj; diff --git a/arch/x86/kernel/uv_irq.c b/arch/x86/kernel/uv_irq.c index 1132129db79..7b24460917d 100644 --- a/arch/x86/kernel/uv_irq.c +++ b/arch/x86/kernel/uv_irq.c @@ -28,34 +28,21 @@ struct uv_irq_2_mmr_pnode{ static spinlock_t uv_irq_lock; static struct rb_root uv_irq_root; -static int uv_set_irq_affinity(unsigned int, const struct cpumask *); +static int uv_set_irq_affinity(struct irq_data *, const struct cpumask *, bool); -static void uv_noop(unsigned int irq) -{ -} - -static unsigned int uv_noop_ret(unsigned int irq) -{ - return 0; -} +static void uv_noop(struct irq_data *data) { } -static void uv_ack_apic(unsigned int irq) +static void uv_ack_apic(struct irq_data *data) { ack_APIC_irq(); } static 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_affinity = uv_set_irq_affinity, + .name = "UV-CORE", + .irq_mask = uv_noop, + .irq_unmask = uv_noop, + .irq_eoi = uv_ack_apic, + .irq_set_affinity = uv_set_irq_affinity, }; /* @@ -144,26 +131,22 @@ arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, unsigned long mmr_offset, int limit) { const struct cpumask *eligible_cpu = cpumask_of(cpu); - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; - int mmr_pnode; + struct irq_cfg *cfg = get_irq_chip_data(irq); unsigned long mmr_value; struct uv_IO_APIC_route_entry *entry; - int err; + int mmr_pnode, err; BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); - cfg = irq_cfg(irq); - err = assign_irq_vector(irq, cfg, eligible_cpu); if (err != 0) return err; if (limit == UV_AFFINITY_CPU) - desc->status |= IRQ_NO_BALANCING; + irq_set_status_flags(irq, IRQ_NO_BALANCING); else - desc->status |= IRQ_MOVE_PCNTXT; + irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); set_irq_chip_and_handler_name(irq, &uv_irq_chip, handle_percpu_irq, irq_name); @@ -206,17 +189,17 @@ static void arch_disable_uv_irq(int mmr_pnode, unsigned long mmr_offset) uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); } -static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask) +static int +uv_set_irq_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = desc->chip_data; + struct irq_cfg *cfg = data->chip_data; unsigned int dest; - unsigned long mmr_value; + unsigned long mmr_value, mmr_offset; struct uv_IO_APIC_route_entry *entry; - unsigned long mmr_offset; int mmr_pnode; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; mmr_value = 0; @@ -231,7 +214,7 @@ static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask) entry->dest = dest; /* Get previously stored MMR and pnode of hub sourcing interrupts */ - if (uv_irq_2_mmr_info(irq, &mmr_offset, &mmr_pnode)) + if (uv_irq_2_mmr_info(data->irq, &mmr_offset, &mmr_pnode)) return -1; uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); diff --git a/arch/x86/kernel/visws_quirks.c b/arch/x86/kernel/visws_quirks.c index e680ea52db9..3371bd053b8 100644 --- a/arch/x86/kernel/visws_quirks.c +++ b/arch/x86/kernel/visws_quirks.c @@ -66,10 +66,7 @@ static void __init visws_time_init(void) } /* Replaces the default init_ISA_irqs in the generic setup */ -static void __init visws_pre_intr_init(void) -{ - init_VISWS_APIC_irqs(); -} +static void __init visws_pre_intr_init(void); /* Quirk for machine specific memory setup. */ @@ -429,67 +426,34 @@ static int is_co_apic(unsigned int irq) /* * This is the SGI Cobalt (IO-)APIC: */ - -static void enable_cobalt_irq(unsigned int irq) +static void enable_cobalt_irq(struct irq_data *data) { - co_apic_set(is_co_apic(irq), irq); + co_apic_set(is_co_apic(data->irq), data->irq); } -static void disable_cobalt_irq(unsigned int irq) +static void disable_cobalt_irq(struct irq_data *data) { - int entry = is_co_apic(irq); + int entry = is_co_apic(data->irq); co_apic_write(CO_APIC_LO(entry), CO_APIC_MASK); co_apic_read(CO_APIC_LO(entry)); } -/* - * "irq" really just serves to identify the device. Here is where we - * map this to the Cobalt APIC entry where it's physically wired. - * This is called via request_irq -> setup_irq -> irq_desc->startup() - */ -static unsigned int startup_cobalt_irq(unsigned int irq) +static void ack_cobalt_irq(struct irq_data *data) { unsigned long flags; - struct irq_desc *desc = irq_to_desc(irq); spin_lock_irqsave(&cobalt_lock, flags); - 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; -} - -static void ack_cobalt_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&cobalt_lock, flags); - disable_cobalt_irq(irq); + disable_cobalt_irq(data); apic_write(APIC_EOI, APIC_EIO_ACK); spin_unlock_irqrestore(&cobalt_lock, flags); } -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 (!(desc->status & (IRQ_DISABLED | IRQ_INPROGRESS))) - enable_cobalt_irq(irq); - spin_unlock_irqrestore(&cobalt_lock, flags); -} - static struct irq_chip cobalt_irq_type = { - .name = "Cobalt-APIC", - .startup = startup_cobalt_irq, - .shutdown = disable_cobalt_irq, - .enable = enable_cobalt_irq, - .disable = disable_cobalt_irq, - .ack = ack_cobalt_irq, - .end = end_cobalt_irq, + .name = "Cobalt-APIC", + .irq_enable = enable_cobalt_irq, + .irq_disable = disable_cobalt_irq, + .irq_ack = ack_cobalt_irq, }; @@ -503,35 +467,34 @@ static struct irq_chip cobalt_irq_type = { * interrupt controller type, and through a special virtual interrupt- * controller. Device drivers only see the virtual interrupt sources. */ -static unsigned int startup_piix4_master_irq(unsigned int irq) +static unsigned int startup_piix4_master_irq(struct irq_data *data) { legacy_pic->init(0); - - return startup_cobalt_irq(irq); + enable_cobalt_irq(data); } -static void end_piix4_master_irq(unsigned int irq) +static void end_piix4_master_irq(struct irq_data *data) { unsigned long flags; spin_lock_irqsave(&cobalt_lock, flags); - enable_cobalt_irq(irq); + enable_cobalt_irq(data); spin_unlock_irqrestore(&cobalt_lock, flags); } static struct irq_chip piix4_master_irq_type = { - .name = "PIIX4-master", - .startup = startup_piix4_master_irq, - .ack = ack_cobalt_irq, - .end = end_piix4_master_irq, + .name = "PIIX4-master", + .irq_startup = startup_piix4_master_irq, + .irq_ack = ack_cobalt_irq, }; +static void pii4_mask(struct irq_data *data) { } static struct irq_chip piix4_virtual_irq_type = { - .name = "PIIX4-virtual", + .name = "PIIX4-virtual", + .mask = pii4_mask, }; - /* * PIIX4-8259 master/virtual functions to handle interrupt requests * from legacy devices: floppy, parallel, serial, rtc. @@ -549,9 +512,8 @@ static struct irq_chip piix4_virtual_irq_type = { */ static irqreturn_t piix4_master_intr(int irq, void *dev_id) { - int realirq; - struct irq_desc *desc; unsigned long flags; + int realirq; raw_spin_lock_irqsave(&i8259A_lock, flags); @@ -592,18 +554,10 @@ static irqreturn_t piix4_master_intr(int irq, void *dev_id) raw_spin_unlock_irqrestore(&i8259A_lock, flags); - desc = irq_to_desc(realirq); - /* * handle this 'virtual interrupt' as a Cobalt one now. */ - kstat_incr_irqs_this_cpu(realirq, desc); - - if (likely(desc->action != NULL)) - handle_IRQ_event(realirq, desc->action); - - if (!(desc->status & IRQ_DISABLED)) - legacy_pic->chip->unmask(realirq); + generic_handle_irq(realirq); return IRQ_HANDLED; @@ -624,41 +578,35 @@ static struct irqaction cascade_action = { static inline void set_piix4_virtual_irq_type(void) { - piix4_virtual_irq_type.shutdown = i8259A_chip.mask; piix4_virtual_irq_type.enable = i8259A_chip.unmask; piix4_virtual_irq_type.disable = i8259A_chip.mask; + piix4_virtual_irq_type.unmask = i8259A_chip.unmask; } -void init_VISWS_APIC_irqs(void) +static void __init visws_pre_intr_init(void) { int i; - for (i = 0; i < CO_IRQ_APIC0 + CO_APIC_LAST + 1; i++) { - struct irq_desc *desc = irq_to_desc(i); - - desc->status = IRQ_DISABLED; - desc->action = 0; - desc->depth = 1; + set_piix4_virtual_irq_type(); - if (i == 0) { - desc->chip = &cobalt_irq_type; - } - else if (i == CO_IRQ_IDE0) { - desc->chip = &cobalt_irq_type; - } - else if (i == CO_IRQ_IDE1) { - desc->chip = &cobalt_irq_type; - } - else if (i == CO_IRQ_8259) { - desc->chip = &piix4_master_irq_type; - } - else if (i < CO_IRQ_APIC0) { - set_piix4_virtual_irq_type(); - desc->chip = &piix4_virtual_irq_type; - } - else if (IS_CO_APIC(i)) { - desc->chip = &cobalt_irq_type; - } + for (i = 0; i < CO_IRQ_APIC0 + CO_APIC_LAST + 1; i++) { + struct irq_chip *chip = NULL; + + if (i == 0) + chip = &cobalt_irq_type; + else if (i == CO_IRQ_IDE0) + chip = &cobalt_irq_type; + else if (i == CO_IRQ_IDE1) + >chip = &cobalt_irq_type; + else if (i == CO_IRQ_8259) + chip = &piix4_master_irq_type; + else if (i < CO_IRQ_APIC0) + chip = &piix4_virtual_irq_type; + else if (IS_CO_APIC(i)) + chip = &cobalt_irq_type; + + if (chip) + set_irq_chip(i, chip); } setup_irq(CO_IRQ_8259, &master_action); diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c index 5ffb5622f79..61fb9851962 100644 --- a/arch/x86/kernel/vm86_32.c +++ b/arch/x86/kernel/vm86_32.c @@ -551,8 +551,14 @@ cannot_handle: int handle_vm86_trap(struct kernel_vm86_regs *regs, long error_code, int trapno) { if (VMPI.is_vm86pus) { - if ((trapno == 3) || (trapno == 1)) - return_to_32bit(regs, VM86_TRAP + (trapno << 8)); + if ((trapno == 3) || (trapno == 1)) { + KVM86->regs32->ax = VM86_TRAP + (trapno << 8); + /* setting this flag forces the code in entry_32.S to + call save_v86_state() and change the stack pointer + to KVM86->regs32 */ + set_thread_flag(TIF_IRET); + return 0; + } do_int(regs, trapno, (unsigned char __user *) (regs->pt.ss << 4), SP(regs)); return 0; } diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c deleted file mode 100644 index ce9fbacb752..00000000000 --- a/arch/x86/kernel/vmi_32.c +++ /dev/null @@ -1,893 +0,0 @@ -/* - * VMI specific paravirt-ops implementation - * - * Copyright (C) 2005, VMware, Inc. - * - * 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., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to zach@vmware.com - * - */ - -#include <linux/module.h> -#include <linux/cpu.h> -#include <linux/bootmem.h> -#include <linux/mm.h> -#include <linux/highmem.h> -#include <linux/sched.h> -#include <linux/gfp.h> -#include <asm/vmi.h> -#include <asm/io.h> -#include <asm/fixmap.h> -#include <asm/apicdef.h> -#include <asm/apic.h> -#include <asm/pgalloc.h> -#include <asm/processor.h> -#include <asm/timer.h> -#include <asm/vmi_time.h> -#include <asm/kmap_types.h> -#include <asm/setup.h> - -/* Convenient for calling VMI functions indirectly in the ROM */ -typedef u32 __attribute__((regparm(1))) (VROMFUNC)(void); -typedef u64 __attribute__((regparm(2))) (VROMLONGFUNC)(int); - -#define call_vrom_func(rom,func) \ - (((VROMFUNC *)(rom->func))()) - -#define call_vrom_long_func(rom,func,arg) \ - (((VROMLONGFUNC *)(rom->func)) (arg)) - -static struct vrom_header *vmi_rom; -static int disable_pge; -static int disable_pse; -static int disable_sep; -static int disable_tsc; -static int disable_mtrr; -static int disable_noidle; -static int disable_vmi_timer; - -/* Cached VMI operations */ -static struct { - void (*cpuid)(void /* non-c */); - void (*_set_ldt)(u32 selector); - void (*set_tr)(u32 selector); - void (*write_idt_entry)(struct desc_struct *, int, u32, u32); - void (*write_gdt_entry)(struct desc_struct *, int, u32, u32); - void (*write_ldt_entry)(struct desc_struct *, int, u32, u32); - void (*set_kernel_stack)(u32 selector, u32 sp0); - void (*allocate_page)(u32, u32, u32, u32, u32); - void (*release_page)(u32, u32); - void (*set_pte)(pte_t, pte_t *, unsigned); - void (*update_pte)(pte_t *, unsigned); - void (*set_linear_mapping)(int, void *, u32, u32); - void (*_flush_tlb)(int); - void (*set_initial_ap_state)(int, int); - void (*halt)(void); - void (*set_lazy_mode)(int mode); -} vmi_ops; - -/* Cached VMI operations */ -struct vmi_timer_ops vmi_timer_ops; - -/* - * VMI patching routines. - */ -#define MNEM_CALL 0xe8 -#define MNEM_JMP 0xe9 -#define MNEM_RET 0xc3 - -#define IRQ_PATCH_INT_MASK 0 -#define IRQ_PATCH_DISABLE 5 - -static inline void patch_offset(void *insnbuf, - unsigned long ip, unsigned long dest) -{ - *(unsigned long *)(insnbuf+1) = dest-ip-5; -} - -static unsigned patch_internal(int call, unsigned len, void *insnbuf, - unsigned long ip) -{ - u64 reloc; - struct vmi_relocation_info *const rel = (struct vmi_relocation_info *)&reloc; - reloc = call_vrom_long_func(vmi_rom, get_reloc, call); - switch(rel->type) { - case VMI_RELOCATION_CALL_REL: - BUG_ON(len < 5); - *(char *)insnbuf = MNEM_CALL; - patch_offset(insnbuf, ip, (unsigned long)rel->eip); - return 5; - - case VMI_RELOCATION_JUMP_REL: - BUG_ON(len < 5); - *(char *)insnbuf = MNEM_JMP; - patch_offset(insnbuf, ip, (unsigned long)rel->eip); - return 5; - - case VMI_RELOCATION_NOP: - /* obliterate the whole thing */ - return 0; - - case VMI_RELOCATION_NONE: - /* leave native code in place */ - break; - - default: - BUG(); - } - return len; -} - -/* - * Apply patch if appropriate, return length of new instruction - * sequence. The callee does nop padding for us. - */ -static unsigned vmi_patch(u8 type, u16 clobbers, void *insns, - unsigned long ip, unsigned len) -{ - switch (type) { - case PARAVIRT_PATCH(pv_irq_ops.irq_disable): - return patch_internal(VMI_CALL_DisableInterrupts, len, - insns, ip); - case PARAVIRT_PATCH(pv_irq_ops.irq_enable): - return patch_internal(VMI_CALL_EnableInterrupts, len, - insns, ip); - case PARAVIRT_PATCH(pv_irq_ops.restore_fl): - return patch_internal(VMI_CALL_SetInterruptMask, len, - insns, ip); - case PARAVIRT_PATCH(pv_irq_ops.save_fl): - return patch_internal(VMI_CALL_GetInterruptMask, len, - insns, ip); - case PARAVIRT_PATCH(pv_cpu_ops.iret): - return patch_internal(VMI_CALL_IRET, len, insns, ip); - case PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit): - return patch_internal(VMI_CALL_SYSEXIT, len, insns, ip); - default: - break; - } - return len; -} - -/* CPUID has non-C semantics, and paravirt-ops API doesn't match hardware ISA */ -static void vmi_cpuid(unsigned int *ax, unsigned int *bx, - unsigned int *cx, unsigned int *dx) -{ - int override = 0; - if (*ax == 1) - override = 1; - asm volatile ("call *%6" - : "=a" (*ax), - "=b" (*bx), - "=c" (*cx), - "=d" (*dx) - : "0" (*ax), "2" (*cx), "r" (vmi_ops.cpuid)); - if (override) { - if (disable_pse) - *dx &= ~X86_FEATURE_PSE; - if (disable_pge) - *dx &= ~X86_FEATURE_PGE; - if (disable_sep) - *dx &= ~X86_FEATURE_SEP; - if (disable_tsc) - *dx &= ~X86_FEATURE_TSC; - if (disable_mtrr) - *dx &= ~X86_FEATURE_MTRR; - } -} - -static inline void vmi_maybe_load_tls(struct desc_struct *gdt, int nr, struct desc_struct *new) -{ - if (gdt[nr].a != new->a || gdt[nr].b != new->b) - write_gdt_entry(gdt, nr, new, 0); -} - -static void vmi_load_tls(struct thread_struct *t, unsigned int cpu) -{ - struct desc_struct *gdt = get_cpu_gdt_table(cpu); - vmi_maybe_load_tls(gdt, GDT_ENTRY_TLS_MIN + 0, &t->tls_array[0]); - vmi_maybe_load_tls(gdt, GDT_ENTRY_TLS_MIN + 1, &t->tls_array[1]); - vmi_maybe_load_tls(gdt, GDT_ENTRY_TLS_MIN + 2, &t->tls_array[2]); -} - -static void vmi_set_ldt(const void *addr, unsigned entries) -{ - unsigned cpu = smp_processor_id(); - struct desc_struct desc; - - pack_descriptor(&desc, (unsigned long)addr, - entries * sizeof(struct desc_struct) - 1, - DESC_LDT, 0); - write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, &desc, DESC_LDT); - vmi_ops._set_ldt(entries ? GDT_ENTRY_LDT*sizeof(struct desc_struct) : 0); -} - -static void vmi_set_tr(void) -{ - vmi_ops.set_tr(GDT_ENTRY_TSS*sizeof(struct desc_struct)); -} - -static void vmi_write_idt_entry(gate_desc *dt, int entry, const gate_desc *g) -{ - u32 *idt_entry = (u32 *)g; - vmi_ops.write_idt_entry(dt, entry, idt_entry[0], idt_entry[1]); -} - -static void vmi_write_gdt_entry(struct desc_struct *dt, int entry, - const void *desc, int type) -{ - u32 *gdt_entry = (u32 *)desc; - vmi_ops.write_gdt_entry(dt, entry, gdt_entry[0], gdt_entry[1]); -} - -static void vmi_write_ldt_entry(struct desc_struct *dt, int entry, - const void *desc) -{ - u32 *ldt_entry = (u32 *)desc; - vmi_ops.write_ldt_entry(dt, entry, ldt_entry[0], ldt_entry[1]); -} - -static void vmi_load_sp0(struct tss_struct *tss, - struct thread_struct *thread) -{ - tss->x86_tss.sp0 = thread->sp0; - - /* This can only happen when SEP is enabled, no need to test "SEP"arately */ - if (unlikely(tss->x86_tss.ss1 != thread->sysenter_cs)) { - tss->x86_tss.ss1 = thread->sysenter_cs; - wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0); - } - vmi_ops.set_kernel_stack(__KERNEL_DS, tss->x86_tss.sp0); -} - -static void vmi_flush_tlb_user(void) -{ - vmi_ops._flush_tlb(VMI_FLUSH_TLB); -} - -static void vmi_flush_tlb_kernel(void) -{ - vmi_ops._flush_tlb(VMI_FLUSH_TLB | VMI_FLUSH_GLOBAL); -} - -/* Stub to do nothing at all; used for delays and unimplemented calls */ -static void vmi_nop(void) -{ -} - -static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn) -{ - vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0); -} - -static void vmi_allocate_pmd(struct mm_struct *mm, unsigned long pfn) -{ - /* - * This call comes in very early, before mem_map is setup. - * It is called only for swapper_pg_dir, which already has - * data on it. - */ - 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_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); -} - -static void vmi_release_pmd(unsigned long pfn) -{ - vmi_ops.release_page(pfn, VMI_PAGE_L2); -} - -/* - * We use the pgd_free hook for releasing the pgd page: - */ -static void vmi_pgd_free(struct mm_struct *mm, pgd_t *pgd) -{ - unsigned long pfn = __pa(pgd) >> PAGE_SHIFT; - - vmi_ops.release_page(pfn, VMI_PAGE_L2); -} - -/* - * Helper macros for MMU update flags. We can defer updates until a flush - * or page invalidation only if the update is to the current address space - * (otherwise, there is no flush). We must check against init_mm, since - * this could be a kernel update, which usually passes init_mm, although - * sometimes this check can be skipped if we know the particular function - * is only called on user mode PTEs. We could change the kernel to pass - * current->active_mm here, but in particular, I was unsure if changing - * mm/highmem.c to do this would still be correct on other architectures. - */ -#define is_current_as(mm, mustbeuser) ((mm) == current->active_mm || \ - (!mustbeuser && (mm) == &init_mm)) -#define vmi_flags_addr(mm, addr, level, user) \ - ((level) | (is_current_as(mm, user) ? \ - (VMI_PAGE_CURRENT_AS | ((addr) & VMI_PAGE_VA_MASK)) : 0)) -#define vmi_flags_addr_defer(mm, addr, level, user) \ - ((level) | (is_current_as(mm, user) ? \ - (VMI_PAGE_DEFER | VMI_PAGE_CURRENT_AS | ((addr) & VMI_PAGE_VA_MASK)) : 0)) - -static void vmi_update_pte(struct mm_struct *mm, unsigned long addr, pte_t *ptep) -{ - 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_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_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_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0)); -} - -static void vmi_set_pmd(pmd_t *pmdp, pmd_t pmdval) -{ -#ifdef CONFIG_X86_PAE - const pte_t pte = { .pte = pmdval.pmd }; -#else - const pte_t pte = { pmdval.pud.pgd.pgd }; -#endif - vmi_ops.set_pte(pte, (pte_t *)pmdp, VMI_PAGE_PD); -} - -#ifdef CONFIG_X86_PAE - -static void vmi_set_pte_atomic(pte_t *ptep, pte_t pteval) -{ - /* - * XXX This is called from set_pmd_pte, but at both PT - * and PD layers so the VMI_PAGE_PT flag is wrong. But - * it is only called for large page mapping changes, - * the Xen backend, doesn't support large pages, and the - * ESX backend doesn't depend on the flag. - */ - set_64bit((unsigned long long *)ptep,pte_val(pteval)); - vmi_ops.update_pte(ptep, VMI_PAGE_PT); -} - -static void vmi_set_pud(pud_t *pudp, pud_t pudval) -{ - /* Um, eww */ - const pte_t pte = { .pte = pudval.pgd.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_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_ops.set_pte(pte, (pte_t *)pmd, VMI_PAGE_PD); -} -#endif - -#ifdef CONFIG_SMP -static void __devinit -vmi_startup_ipi_hook(int phys_apicid, unsigned long start_eip, - unsigned long start_esp) -{ - struct vmi_ap_state ap; - - /* Default everything to zero. This is fine for most GPRs. */ - memset(&ap, 0, sizeof(struct vmi_ap_state)); - - ap.gdtr_limit = GDT_SIZE - 1; - ap.gdtr_base = (unsigned long) get_cpu_gdt_table(phys_apicid); - - ap.idtr_limit = IDT_ENTRIES * 8 - 1; - ap.idtr_base = (unsigned long) idt_table; - - ap.ldtr = 0; - - ap.cs = __KERNEL_CS; - ap.eip = (unsigned long) start_eip; - ap.ss = __KERNEL_DS; - ap.esp = (unsigned long) start_esp; - - ap.ds = __USER_DS; - ap.es = __USER_DS; - ap.fs = __KERNEL_PERCPU; - ap.gs = __KERNEL_STACK_CANARY; - - ap.eflags = 0; - -#ifdef CONFIG_X86_PAE - /* efer should match BSP efer. */ - if (cpu_has_nx) { - unsigned l, h; - rdmsr(MSR_EFER, l, h); - ap.efer = (unsigned long long) h << 32 | l; - } -#endif - - ap.cr3 = __pa(swapper_pg_dir); - /* Protected mode, paging, AM, WP, NE, MP. */ - ap.cr0 = 0x80050023; - ap.cr4 = mmu_cr4_features; - vmi_ops.set_initial_ap_state((u32)&ap, phys_apicid); -} -#endif - -static void vmi_start_context_switch(struct task_struct *prev) -{ - paravirt_start_context_switch(prev); - vmi_ops.set_lazy_mode(2); -} - -static void vmi_end_context_switch(struct task_struct *next) -{ - vmi_ops.set_lazy_mode(0); - paravirt_end_context_switch(next); -} - -static void vmi_enter_lazy_mmu(void) -{ - paravirt_enter_lazy_mmu(); - vmi_ops.set_lazy_mode(1); -} - -static void vmi_leave_lazy_mmu(void) -{ - vmi_ops.set_lazy_mode(0); - paravirt_leave_lazy_mmu(); -} - -static inline int __init check_vmi_rom(struct vrom_header *rom) -{ - struct pci_header *pci; - struct pnp_header *pnp; - const char *manufacturer = "UNKNOWN"; - const char *product = "UNKNOWN"; - const char *license = "unspecified"; - - if (rom->rom_signature != 0xaa55) - return 0; - if (rom->vrom_signature != VMI_SIGNATURE) - return 0; - if (rom->api_version_maj != VMI_API_REV_MAJOR || - rom->api_version_min+1 < VMI_API_REV_MINOR+1) { - printk(KERN_WARNING "VMI: Found mismatched rom version %d.%d\n", - rom->api_version_maj, - rom->api_version_min); - return 0; - } - - /* - * Relying on the VMI_SIGNATURE field is not 100% safe, so check - * the PCI header and device type to make sure this is really a - * VMI device. - */ - if (!rom->pci_header_offs) { - printk(KERN_WARNING "VMI: ROM does not contain PCI header.\n"); - return 0; - } - - pci = (struct pci_header *)((char *)rom+rom->pci_header_offs); - if (pci->vendorID != PCI_VENDOR_ID_VMWARE || - pci->deviceID != PCI_DEVICE_ID_VMWARE_VMI) { - /* Allow it to run... anyways, but warn */ - printk(KERN_WARNING "VMI: ROM from unknown manufacturer\n"); - } - - if (rom->pnp_header_offs) { - pnp = (struct pnp_header *)((char *)rom+rom->pnp_header_offs); - if (pnp->manufacturer_offset) - manufacturer = (const char *)rom+pnp->manufacturer_offset; - if (pnp->product_offset) - product = (const char *)rom+pnp->product_offset; - } - - if (rom->license_offs) - license = (char *)rom+rom->license_offs; - - printk(KERN_INFO "VMI: Found %s %s, API version %d.%d, ROM version %d.%d\n", - manufacturer, product, - rom->api_version_maj, rom->api_version_min, - pci->rom_version_maj, pci->rom_version_min); - - /* Don't allow BSD/MIT here for now because we don't want to end up - with any binary only shim layers */ - if (strcmp(license, "GPL") && strcmp(license, "GPL v2")) { - printk(KERN_WARNING "VMI: Non GPL license `%s' found for ROM. Not used.\n", - license); - return 0; - } - - return 1; -} - -/* - * Probe for the VMI option ROM - */ -static inline int __init probe_vmi_rom(void) -{ - unsigned long base; - - /* VMI ROM is in option ROM area, check signature */ - for (base = 0xC0000; base < 0xE0000; base += 2048) { - struct vrom_header *romstart; - romstart = (struct vrom_header *)isa_bus_to_virt(base); - if (check_vmi_rom(romstart)) { - vmi_rom = romstart; - return 1; - } - } - return 0; -} - -/* - * VMI setup common to all processors - */ -void vmi_bringup(void) -{ - /* We must establish the lowmem mapping for MMU ops to work */ - if (vmi_ops.set_linear_mapping) - vmi_ops.set_linear_mapping(0, (void *)__PAGE_OFFSET, MAXMEM_PFN, 0); -} - -/* - * Return a pointer to a VMI function or NULL if unimplemented - */ -static void *vmi_get_function(int vmicall) -{ - u64 reloc; - const struct vmi_relocation_info *rel = (struct vmi_relocation_info *)&reloc; - reloc = call_vrom_long_func(vmi_rom, get_reloc, vmicall); - BUG_ON(rel->type == VMI_RELOCATION_JUMP_REL); - if (rel->type == VMI_RELOCATION_CALL_REL) - return (void *)rel->eip; - else - return NULL; -} - -/* - * Helper macro for making the VMI paravirt-ops fill code readable. - * For unimplemented operations, fall back to default, unless nop - * is returned by the ROM. - */ -#define para_fill(opname, vmicall) \ -do { \ - reloc = call_vrom_long_func(vmi_rom, get_reloc, \ - VMI_CALL_##vmicall); \ - if (rel->type == VMI_RELOCATION_CALL_REL) \ - opname = (void *)rel->eip; \ - else if (rel->type == VMI_RELOCATION_NOP) \ - opname = (void *)vmi_nop; \ - else if (rel->type != VMI_RELOCATION_NONE) \ - printk(KERN_WARNING "VMI: Unknown relocation " \ - "type %d for " #vmicall"\n",\ - rel->type); \ -} while (0) - -/* - * Helper macro for making the VMI paravirt-ops fill code readable. - * For cached operations which do not match the VMI ROM ABI and must - * go through a tranlation stub. Ignore NOPs, since it is not clear - * a NOP * VMI function corresponds to a NOP paravirt-op when the - * functions are not in 1-1 correspondence. - */ -#define para_wrap(opname, wrapper, cache, vmicall) \ -do { \ - reloc = call_vrom_long_func(vmi_rom, get_reloc, \ - VMI_CALL_##vmicall); \ - BUG_ON(rel->type == VMI_RELOCATION_JUMP_REL); \ - if (rel->type == VMI_RELOCATION_CALL_REL) { \ - opname = wrapper; \ - vmi_ops.cache = (void *)rel->eip; \ - } \ -} while (0) - -/* - * Activate the VMI interface and switch into paravirtualized mode - */ -static inline int __init activate_vmi(void) -{ - short kernel_cs; - u64 reloc; - const struct vmi_relocation_info *rel = (struct vmi_relocation_info *)&reloc; - - /* - * Prevent page tables from being allocated in highmem, even if - * CONFIG_HIGHPTE is enabled. - */ - __userpte_alloc_gfp &= ~__GFP_HIGHMEM; - - if (call_vrom_func(vmi_rom, vmi_init) != 0) { - printk(KERN_ERR "VMI ROM failed to initialize!"); - return 0; - } - savesegment(cs, kernel_cs); - - pv_info.paravirt_enabled = 1; - pv_info.kernel_rpl = kernel_cs & SEGMENT_RPL_MASK; - pv_info.name = "vmi [deprecated]"; - - pv_init_ops.patch = vmi_patch; - - /* - * Many of these operations are ABI compatible with VMI. - * This means we can fill in the paravirt-ops with direct - * pointers into the VMI ROM. If the calling convention for - * these operations changes, this code needs to be updated. - * - * Exceptions - * CPUID paravirt-op uses pointers, not the native ISA - * halt has no VMI equivalent; all VMI halts are "safe" - * no MSR support yet - just trap and emulate. VMI uses the - * same ABI as the native ISA, but Linux wants exceptions - * from bogus MSR read / write handled - * rdpmc is not yet used in Linux - */ - - /* CPUID is special, so very special it gets wrapped like a present */ - para_wrap(pv_cpu_ops.cpuid, vmi_cpuid, cpuid, CPUID); - - para_fill(pv_cpu_ops.clts, CLTS); - para_fill(pv_cpu_ops.get_debugreg, GetDR); - para_fill(pv_cpu_ops.set_debugreg, SetDR); - para_fill(pv_cpu_ops.read_cr0, GetCR0); - para_fill(pv_mmu_ops.read_cr2, GetCR2); - para_fill(pv_mmu_ops.read_cr3, GetCR3); - para_fill(pv_cpu_ops.read_cr4, GetCR4); - para_fill(pv_cpu_ops.write_cr0, SetCR0); - para_fill(pv_mmu_ops.write_cr2, SetCR2); - para_fill(pv_mmu_ops.write_cr3, SetCR3); - para_fill(pv_cpu_ops.write_cr4, SetCR4); - - para_fill(pv_irq_ops.save_fl.func, GetInterruptMask); - para_fill(pv_irq_ops.restore_fl.func, SetInterruptMask); - para_fill(pv_irq_ops.irq_disable.func, DisableInterrupts); - para_fill(pv_irq_ops.irq_enable.func, EnableInterrupts); - - para_fill(pv_cpu_ops.wbinvd, WBINVD); - para_fill(pv_cpu_ops.read_tsc, RDTSC); - - /* The following we emulate with trap and emulate for now */ - /* paravirt_ops.read_msr = vmi_rdmsr */ - /* paravirt_ops.write_msr = vmi_wrmsr */ - /* paravirt_ops.rdpmc = vmi_rdpmc */ - - /* TR interface doesn't pass TR value, wrap */ - para_wrap(pv_cpu_ops.load_tr_desc, vmi_set_tr, set_tr, SetTR); - - /* LDT is special, too */ - para_wrap(pv_cpu_ops.set_ldt, vmi_set_ldt, _set_ldt, SetLDT); - - para_fill(pv_cpu_ops.load_gdt, SetGDT); - para_fill(pv_cpu_ops.load_idt, SetIDT); - para_fill(pv_cpu_ops.store_gdt, GetGDT); - para_fill(pv_cpu_ops.store_idt, GetIDT); - para_fill(pv_cpu_ops.store_tr, GetTR); - pv_cpu_ops.load_tls = vmi_load_tls; - para_wrap(pv_cpu_ops.write_ldt_entry, vmi_write_ldt_entry, - write_ldt_entry, WriteLDTEntry); - para_wrap(pv_cpu_ops.write_gdt_entry, vmi_write_gdt_entry, - write_gdt_entry, WriteGDTEntry); - para_wrap(pv_cpu_ops.write_idt_entry, vmi_write_idt_entry, - write_idt_entry, WriteIDTEntry); - para_wrap(pv_cpu_ops.load_sp0, vmi_load_sp0, set_kernel_stack, UpdateKernelStack); - para_fill(pv_cpu_ops.set_iopl_mask, SetIOPLMask); - para_fill(pv_cpu_ops.io_delay, IODelay); - - para_wrap(pv_cpu_ops.start_context_switch, vmi_start_context_switch, - set_lazy_mode, SetLazyMode); - para_wrap(pv_cpu_ops.end_context_switch, vmi_end_context_switch, - set_lazy_mode, SetLazyMode); - - para_wrap(pv_mmu_ops.lazy_mode.enter, vmi_enter_lazy_mmu, - set_lazy_mode, SetLazyMode); - para_wrap(pv_mmu_ops.lazy_mode.leave, vmi_leave_lazy_mmu, - set_lazy_mode, SetLazyMode); - - /* user and kernel flush are just handled with different flags to FlushTLB */ - para_wrap(pv_mmu_ops.flush_tlb_user, vmi_flush_tlb_user, _flush_tlb, FlushTLB); - para_wrap(pv_mmu_ops.flush_tlb_kernel, vmi_flush_tlb_kernel, _flush_tlb, FlushTLB); - para_fill(pv_mmu_ops.flush_tlb_single, InvalPage); - - /* - * Until a standard flag format can be agreed on, we need to - * implement these as wrappers in Linux. Get the VMI ROM - * function pointers for the two backend calls. - */ -#ifdef CONFIG_X86_PAE - vmi_ops.set_pte = vmi_get_function(VMI_CALL_SetPxELong); - vmi_ops.update_pte = vmi_get_function(VMI_CALL_UpdatePxELong); -#else - vmi_ops.set_pte = vmi_get_function(VMI_CALL_SetPxE); - vmi_ops.update_pte = vmi_get_function(VMI_CALL_UpdatePxE); -#endif - - if (vmi_ops.set_pte) { - pv_mmu_ops.set_pte = vmi_set_pte; - pv_mmu_ops.set_pte_at = vmi_set_pte_at; - pv_mmu_ops.set_pmd = vmi_set_pmd; -#ifdef CONFIG_X86_PAE - pv_mmu_ops.set_pte_atomic = vmi_set_pte_atomic; - pv_mmu_ops.set_pud = vmi_set_pud; - pv_mmu_ops.pte_clear = vmi_pte_clear; - pv_mmu_ops.pmd_clear = vmi_pmd_clear; -#endif - } - - if (vmi_ops.update_pte) { - pv_mmu_ops.pte_update = vmi_update_pte; - pv_mmu_ops.pte_update_defer = vmi_update_pte_defer; - } - - vmi_ops.allocate_page = vmi_get_function(VMI_CALL_AllocatePage); - if (vmi_ops.allocate_page) { - pv_mmu_ops.alloc_pte = vmi_allocate_pte; - pv_mmu_ops.alloc_pmd = vmi_allocate_pmd; - pv_mmu_ops.alloc_pmd_clone = vmi_allocate_pmd_clone; - } - - vmi_ops.release_page = vmi_get_function(VMI_CALL_ReleasePage); - if (vmi_ops.release_page) { - pv_mmu_ops.release_pte = vmi_release_pte; - pv_mmu_ops.release_pmd = vmi_release_pmd; - pv_mmu_ops.pgd_free = vmi_pgd_free; - } - - /* Set linear is needed in all cases */ - vmi_ops.set_linear_mapping = vmi_get_function(VMI_CALL_SetLinearMapping); - - /* - * These MUST always be patched. Don't support indirect jumps - * through these operations, as the VMI interface may use either - * a jump or a call to get to these operations, depending on - * the backend. They are performance critical anyway, so requiring - * a patch is not a big problem. - */ - pv_cpu_ops.irq_enable_sysexit = (void *)0xfeedbab0; - pv_cpu_ops.iret = (void *)0xbadbab0; - -#ifdef CONFIG_SMP - para_wrap(pv_apic_ops.startup_ipi_hook, vmi_startup_ipi_hook, set_initial_ap_state, SetInitialAPState); -#endif - -#ifdef CONFIG_X86_LOCAL_APIC - para_fill(apic->read, APICRead); - para_fill(apic->write, APICWrite); -#endif - - /* - * Check for VMI timer functionality by probing for a cycle frequency method - */ - reloc = call_vrom_long_func(vmi_rom, get_reloc, VMI_CALL_GetCycleFrequency); - if (!disable_vmi_timer && rel->type != VMI_RELOCATION_NONE) { - vmi_timer_ops.get_cycle_frequency = (void *)rel->eip; - vmi_timer_ops.get_cycle_counter = - vmi_get_function(VMI_CALL_GetCycleCounter); - vmi_timer_ops.get_wallclock = - vmi_get_function(VMI_CALL_GetWallclockTime); - vmi_timer_ops.wallclock_updated = - vmi_get_function(VMI_CALL_WallclockUpdated); - vmi_timer_ops.set_alarm = vmi_get_function(VMI_CALL_SetAlarm); - vmi_timer_ops.cancel_alarm = - vmi_get_function(VMI_CALL_CancelAlarm); - x86_init.timers.timer_init = vmi_time_init; -#ifdef CONFIG_X86_LOCAL_APIC - x86_init.timers.setup_percpu_clockev = vmi_time_bsp_init; - x86_cpuinit.setup_percpu_clockev = vmi_time_ap_init; -#endif - pv_time_ops.sched_clock = vmi_sched_clock; - x86_platform.calibrate_tsc = vmi_tsc_khz; - x86_platform.get_wallclock = vmi_get_wallclock; - x86_platform.set_wallclock = vmi_set_wallclock; - - /* We have true wallclock functions; disable CMOS clock sync */ - no_sync_cmos_clock = 1; - } else { - disable_noidle = 1; - disable_vmi_timer = 1; - } - - para_fill(pv_irq_ops.safe_halt, Halt); - - /* - * Alternative instruction rewriting doesn't happen soon enough - * to convert VMI_IRET to a call instead of a jump; so we have - * to do this before IRQs get reenabled. Fortunately, it is - * idempotent. - */ - apply_paravirt(__parainstructions, __parainstructions_end); - - vmi_bringup(); - - return 1; -} - -#undef para_fill - -void __init vmi_init(void) -{ - if (!vmi_rom) - probe_vmi_rom(); - else - check_vmi_rom(vmi_rom); - - /* In case probing for or validating the ROM failed, basil */ - if (!vmi_rom) - return; - - reserve_top_address(-vmi_rom->virtual_top); - -#ifdef CONFIG_X86_IO_APIC - /* This is virtual hardware; timer routing is wired correctly */ - no_timer_check = 1; -#endif -} - -void __init vmi_activate(void) -{ - unsigned long flags; - - if (!vmi_rom) - return; - - local_irq_save(flags); - activate_vmi(); - local_irq_restore(flags & X86_EFLAGS_IF); -} - -static int __init parse_vmi(char *arg) -{ - if (!arg) - return -EINVAL; - - if (!strcmp(arg, "disable_pge")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_PGE); - disable_pge = 1; - } else if (!strcmp(arg, "disable_pse")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_PSE); - disable_pse = 1; - } else if (!strcmp(arg, "disable_sep")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_SEP); - disable_sep = 1; - } else if (!strcmp(arg, "disable_tsc")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_TSC); - disable_tsc = 1; - } else if (!strcmp(arg, "disable_mtrr")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_MTRR); - disable_mtrr = 1; - } else if (!strcmp(arg, "disable_timer")) { - disable_vmi_timer = 1; - disable_noidle = 1; - } else if (!strcmp(arg, "disable_noidle")) - disable_noidle = 1; - return 0; -} - -early_param("vmi", parse_vmi); diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c deleted file mode 100644 index 5e1ff66ecd7..00000000000 --- a/arch/x86/kernel/vmiclock_32.c +++ /dev/null @@ -1,317 +0,0 @@ -/* - * VMI paravirtual timer support routines. - * - * Copyright (C) 2007, VMware, Inc. - * - * 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., 675 Mass Ave, Cambridge, MA 02139, USA. - * - */ - -#include <linux/smp.h> -#include <linux/interrupt.h> -#include <linux/cpumask.h> -#include <linux/clocksource.h> -#include <linux/clockchips.h> - -#include <asm/vmi.h> -#include <asm/vmi_time.h> -#include <asm/apicdef.h> -#include <asm/apic.h> -#include <asm/timer.h> -#include <asm/i8253.h> -#include <asm/irq_vectors.h> - -#define VMI_ONESHOT (VMI_ALARM_IS_ONESHOT | VMI_CYCLES_REAL | vmi_get_alarm_wiring()) -#define VMI_PERIODIC (VMI_ALARM_IS_PERIODIC | VMI_CYCLES_REAL | vmi_get_alarm_wiring()) - -static DEFINE_PER_CPU(struct clock_event_device, local_events); - -static inline u32 vmi_counter(u32 flags) -{ - /* Given VMI_ONESHOT or VMI_PERIODIC, return the corresponding - * cycle counter. */ - return flags & VMI_ALARM_COUNTER_MASK; -} - -/* paravirt_ops.get_wallclock = vmi_get_wallclock */ -unsigned long vmi_get_wallclock(void) -{ - unsigned long long wallclock; - wallclock = vmi_timer_ops.get_wallclock(); // nsec - (void)do_div(wallclock, 1000000000); // sec - - return wallclock; -} - -/* paravirt_ops.set_wallclock = vmi_set_wallclock */ -int vmi_set_wallclock(unsigned long now) -{ - return 0; -} - -/* paravirt_ops.sched_clock = vmi_sched_clock */ -unsigned long long vmi_sched_clock(void) -{ - return cycles_2_ns(vmi_timer_ops.get_cycle_counter(VMI_CYCLES_AVAILABLE)); -} - -/* x86_platform.calibrate_tsc = vmi_tsc_khz */ -unsigned long vmi_tsc_khz(void) -{ - unsigned long long khz; - khz = vmi_timer_ops.get_cycle_frequency(); - (void)do_div(khz, 1000); - return khz; -} - -static inline unsigned int vmi_get_timer_vector(void) -{ - return IRQ0_VECTOR; -} - -/** vmi clockchip */ -#ifdef CONFIG_X86_LOCAL_APIC -static unsigned int startup_timer_irq(unsigned int irq) -{ - unsigned long val = apic_read(APIC_LVTT); - apic_write(APIC_LVTT, vmi_get_timer_vector()); - - return (val & APIC_SEND_PENDING); -} - -static void mask_timer_irq(unsigned int irq) -{ - unsigned long val = apic_read(APIC_LVTT); - apic_write(APIC_LVTT, val | APIC_LVT_MASKED); -} - -static void unmask_timer_irq(unsigned int irq) -{ - unsigned long val = apic_read(APIC_LVTT); - apic_write(APIC_LVTT, val & ~APIC_LVT_MASKED); -} - -static void ack_timer_irq(unsigned int irq) -{ - ack_APIC_irq(); -} - -static struct irq_chip vmi_chip __read_mostly = { - .name = "VMI-LOCAL", - .startup = startup_timer_irq, - .mask = mask_timer_irq, - .unmask = unmask_timer_irq, - .ack = ack_timer_irq -}; -#endif - -/** vmi clockevent */ -#define VMI_ALARM_WIRED_IRQ0 0x00000000 -#define VMI_ALARM_WIRED_LVTT 0x00010000 -static int vmi_wiring = VMI_ALARM_WIRED_IRQ0; - -static inline int vmi_get_alarm_wiring(void) -{ - return vmi_wiring; -} - -static void vmi_timer_set_mode(enum clock_event_mode mode, - struct clock_event_device *evt) -{ - cycle_t now, cycles_per_hz; - BUG_ON(!irqs_disabled()); - - switch (mode) { - case CLOCK_EVT_MODE_ONESHOT: - case CLOCK_EVT_MODE_RESUME: - break; - case CLOCK_EVT_MODE_PERIODIC: - cycles_per_hz = vmi_timer_ops.get_cycle_frequency(); - (void)do_div(cycles_per_hz, HZ); - now = vmi_timer_ops.get_cycle_counter(vmi_counter(VMI_PERIODIC)); - vmi_timer_ops.set_alarm(VMI_PERIODIC, now, cycles_per_hz); - break; - case CLOCK_EVT_MODE_UNUSED: - case CLOCK_EVT_MODE_SHUTDOWN: - switch (evt->mode) { - case CLOCK_EVT_MODE_ONESHOT: - vmi_timer_ops.cancel_alarm(VMI_ONESHOT); - break; - case CLOCK_EVT_MODE_PERIODIC: - vmi_timer_ops.cancel_alarm(VMI_PERIODIC); - break; - default: - break; - } - break; - default: - break; - } -} - -static int vmi_timer_next_event(unsigned long delta, - struct clock_event_device *evt) -{ - /* Unfortunately, set_next_event interface only passes relative - * expiry, but we want absolute expiry. It'd be better if were - * were passed an absolute expiry, since a bunch of time may - * have been stolen between the time the delta is computed and - * when we set the alarm below. */ - cycle_t now = vmi_timer_ops.get_cycle_counter(vmi_counter(VMI_ONESHOT)); - - BUG_ON(evt->mode != CLOCK_EVT_MODE_ONESHOT); - vmi_timer_ops.set_alarm(VMI_ONESHOT, now + delta, 0); - return 0; -} - -static struct clock_event_device vmi_clockevent = { - .name = "vmi-timer", - .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT, - .shift = 22, - .set_mode = vmi_timer_set_mode, - .set_next_event = vmi_timer_next_event, - .rating = 1000, - .irq = 0, -}; - -static irqreturn_t vmi_timer_interrupt(int irq, void *dev_id) -{ - struct clock_event_device *evt = &__get_cpu_var(local_events); - evt->event_handler(evt); - return IRQ_HANDLED; -} - -static struct irqaction vmi_clock_action = { - .name = "vmi-timer", - .handler = vmi_timer_interrupt, - .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_TIMER, -}; - -static void __devinit vmi_time_init_clockevent(void) -{ - cycle_t cycles_per_msec; - struct clock_event_device *evt; - - int cpu = smp_processor_id(); - evt = &__get_cpu_var(local_events); - - /* Use cycles_per_msec since div_sc params are 32-bits. */ - cycles_per_msec = vmi_timer_ops.get_cycle_frequency(); - (void)do_div(cycles_per_msec, 1000); - - memcpy(evt, &vmi_clockevent, sizeof(*evt)); - /* Must pick .shift such that .mult fits in 32-bits. Choosing - * .shift to be 22 allows 2^(32-22) cycles per nano-seconds - * before overflow. */ - evt->mult = div_sc(cycles_per_msec, NSEC_PER_MSEC, evt->shift); - /* Upper bound is clockevent's use of ulong for cycle deltas. */ - evt->max_delta_ns = clockevent_delta2ns(ULONG_MAX, evt); - evt->min_delta_ns = clockevent_delta2ns(1, evt); - evt->cpumask = cpumask_of(cpu); - - printk(KERN_WARNING "vmi: registering clock event %s. mult=%u shift=%u\n", - evt->name, evt->mult, evt->shift); - clockevents_register_device(evt); -} - -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 -void __devinit vmi_time_bsp_init(void) -{ - /* - * On APIC systems, we want local timers to fire on each cpu. We do - * this by programming LVTT to deliver timer events to the IRQ handler - * for IRQ-0, since we can't re-use the APIC local timer handler - * without interfering with that code. - */ - clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL); - local_irq_disable(); -#ifdef CONFIG_SMP - /* - * XXX handle_percpu_irq only defined for SMP; we need to switch over - * to using it, since this is a local interrupt, which each CPU must - * handle individually without locking out or dropping simultaneous - * local timers on other CPUs. We also don't want to trigger the - * quirk workaround code for interrupts which gets invoked from - * handle_percpu_irq via eoi, so we use our own IRQ chip. - */ - set_irq_chip_and_handler_name(0, &vmi_chip, handle_percpu_irq, "lvtt"); -#else - set_irq_chip_and_handler_name(0, &vmi_chip, handle_edge_irq, "lvtt"); -#endif - vmi_wiring = VMI_ALARM_WIRED_LVTT; - apic_write(APIC_LVTT, vmi_get_timer_vector()); - local_irq_enable(); - clockevents_notify(CLOCK_EVT_NOTIFY_RESUME, NULL); -} - -void __devinit vmi_time_ap_init(void) -{ - vmi_time_init_clockevent(); - apic_write(APIC_LVTT, vmi_get_timer_vector()); -} -#endif - -/** vmi clocksource */ -static struct clocksource clocksource_vmi; - -static cycle_t read_real_cycles(struct clocksource *cs) -{ - cycle_t ret = (cycle_t)vmi_timer_ops.get_cycle_counter(VMI_CYCLES_REAL); - return max(ret, clocksource_vmi.cycle_last); -} - -static struct clocksource clocksource_vmi = { - .name = "vmi-timer", - .rating = 450, - .read = read_real_cycles, - .mask = CLOCKSOURCE_MASK(64), - .mult = 0, /* to be set */ - .shift = 22, - .flags = CLOCK_SOURCE_IS_CONTINUOUS, -}; - -static int __init init_vmi_clocksource(void) -{ - cycle_t cycles_per_msec; - - if (!vmi_timer_ops.get_cycle_frequency) - return 0; - /* Use khz2mult rather than hz2mult since hz arg is only 32-bits. */ - cycles_per_msec = vmi_timer_ops.get_cycle_frequency(); - (void)do_div(cycles_per_msec, 1000); - - /* Note that clocksource.{mult, shift} converts in the opposite direction - * as clockevents. */ - clocksource_vmi.mult = clocksource_khz2mult(cycles_per_msec, - clocksource_vmi.shift); - - printk(KERN_WARNING "vmi: registering clock source khz=%lld\n", cycles_per_msec); - return clocksource_register(&clocksource_vmi); - -} -module_init(init_vmi_clocksource); diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index d0bb52296fa..e03530aebfd 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -242,6 +242,12 @@ SECTIONS __x86_cpu_dev_end = .; } + /* + * start address and size of operations which during runtime + * can be patched with virtualization friendly instructions or + * baremetal native ones. Think page table operations. + * Details in paravirt_types.h + */ . = ALIGN(8); .parainstructions : AT(ADDR(.parainstructions) - LOAD_OFFSET) { __parainstructions = .; @@ -249,6 +255,11 @@ SECTIONS __parainstructions_end = .; } + /* + * struct alt_inst entries. From the header (alternative.h): + * "Alternative instructions for different CPU types or capabilities" + * Think locking instructions on spinlocks. + */ . = ALIGN(8); .altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) { __alt_instructions = .; @@ -256,11 +267,28 @@ SECTIONS __alt_instructions_end = .; } + /* + * And here are the replacement instructions. The linker sticks + * them as binary blobs. The .altinstructions has enough data to + * get the address and the length of them to patch the kernel safely. + */ .altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) { *(.altinstr_replacement) } /* + * struct iommu_table_entry entries are injected in this section. + * It is an array of IOMMUs which during run time gets sorted depending + * on its dependency order. After rootfs_initcall is complete + * this section can be safely removed. + */ + .iommu_table : AT(ADDR(.iommu_table) - LOAD_OFFSET) { + __iommu_table = .; + *(.iommu_table) + __iommu_table_end = .; + } + . = ALIGN(8); + /* * .exit.text is discard at runtime, not link time, to deal with * references from .altinstructions and .eh_frame */ @@ -273,7 +301,7 @@ SECTIONS } #if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP) - PERCPU(PAGE_SIZE) + PERCPU(THREAD_SIZE) #endif . = ALIGN(PAGE_SIZE); |