diff options
Diffstat (limited to 'arch/x86/kernel')
156 files changed, 9938 insertions, 7014 deletions
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index d8e5d0cdd678..4c58352209e0 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -40,7 +40,7 @@ obj-$(CONFIG_X86_64) += sys_x86_64.o x8664_ksyms_64.o obj-$(CONFIG_X86_64) += syscall_64.o vsyscall_64.o 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 +obj-y += alternative.o i8253.o pci-nommu.o hw_breakpoint.o obj-y += tsc.o io_delay.o rtc.o obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o @@ -87,9 +87,9 @@ obj-$(CONFIG_VM86) += vm86_32.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o obj-$(CONFIG_HPET_TIMER) += hpet.o +obj-$(CONFIG_APB_TIMER) += apb_timer.o obj-$(CONFIG_K8_NB) += k8.o -obj-$(CONFIG_MGEODE_LX) += geode_32.o mfgpt_32.o obj-$(CONFIG_DEBUG_RODATA_TEST) += test_rodata.o obj-$(CONFIG_DEBUG_NX_TEST) += test_nx.o diff --git a/arch/x86/kernel/acpi/Makefile b/arch/x86/kernel/acpi/Makefile index fd5ca97a2ad5..6f35260bb3ef 100644 --- a/arch/x86/kernel/acpi/Makefile +++ b/arch/x86/kernel/acpi/Makefile @@ -4,7 +4,7 @@ obj-$(CONFIG_ACPI) += boot.o obj-$(CONFIG_ACPI_SLEEP) += sleep.o wakeup_rm.o wakeup_$(BITS).o ifneq ($(CONFIG_ACPI_PROCESSOR),) -obj-y += cstate.o processor.o +obj-y += cstate.o endif $(obj)/wakeup_rm.o: $(obj)/realmode/wakeup.bin diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 67e929b89875..0061ea263061 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -35,6 +35,7 @@ #include <linux/ioport.h> #include <linux/pci.h> +#include <asm/pci_x86.h> #include <asm/pgtable.h> #include <asm/io_apic.h> #include <asm/apic.h> @@ -49,6 +50,7 @@ EXPORT_SYMBOL(acpi_disabled); #ifdef CONFIG_X86_64 # include <asm/proto.h> +# include <asm/numa_64.h> #endif /* X86 */ #define BAD_MADT_ENTRY(entry, end) ( \ @@ -446,6 +448,12 @@ void __init acpi_pic_sci_set_trigger(unsigned int irq, u16 trigger) int acpi_gsi_to_irq(u32 gsi, unsigned int *irq) { *irq = gsi; + +#ifdef CONFIG_X86_IO_APIC + if (acpi_irq_model == ACPI_IRQ_MODEL_IOAPIC) + setup_IO_APIC_irq_extra(gsi); +#endif + return 0; } @@ -473,7 +481,8 @@ int acpi_register_gsi(struct device *dev, u32 gsi, int trigger, int polarity) plat_gsi = mp_register_gsi(dev, gsi, trigger, polarity); } #endif - acpi_gsi_to_irq(plat_gsi, &irq); + irq = plat_gsi; + return irq; } @@ -481,6 +490,26 @@ int acpi_register_gsi(struct device *dev, u32 gsi, int trigger, int polarity) * ACPI based hotplug support for CPU */ #ifdef CONFIG_ACPI_HOTPLUG_CPU +#include <acpi/processor.h> + +static void acpi_map_cpu2node(acpi_handle handle, int cpu, int physid) +{ +#ifdef CONFIG_ACPI_NUMA + int nid; + + nid = acpi_get_node(handle); + if (nid == -1 || !node_online(nid)) + return; +#ifdef CONFIG_X86_64 + apicid_to_node[physid] = nid; + numa_set_node(cpu, nid); +#else /* CONFIG_X86_32 */ + apicid_2_node[physid] = nid; + cpu_to_node_map[cpu] = nid; +#endif + +#endif +} static int __cpuinit _acpi_map_lsapic(acpi_handle handle, int *pcpu) { @@ -539,7 +568,10 @@ static int __cpuinit _acpi_map_lsapic(acpi_handle handle, int *pcpu) goto free_new_map; } + acpi_processor_set_pdc(handle); + cpu = cpumask_first(new_map); + acpi_map_cpu2node(handle, cpu, physid); *pcpu = cpu; retval = 0; @@ -624,6 +656,7 @@ static int __init acpi_parse_hpet(struct acpi_table_header *table) } hpet_address = hpet_tbl->address.address; + hpet_blockid = hpet_tbl->sequence; /* * Some broken BIOSes advertise HPET at 0x0. We really do not @@ -1122,7 +1155,7 @@ static int __init acpi_parse_madt_ioapic_entries(void) if (!acpi_sci_override_gsi) acpi_sci_ioapic_setup(acpi_gbl_FADT.sci_interrupt, 0, 0); - /* Fill in identity legacy mapings where no override */ + /* Fill in identity legacy mappings where no override */ mp_config_acpi_legacy_irqs(); count = @@ -1184,9 +1217,6 @@ static void __init acpi_process_madt(void) if (!error) { acpi_lapic = 1; -#ifdef CONFIG_X86_BIGSMP - generic_bigsmp_probe(); -#endif /* * Parse MADT IO-APIC entries */ @@ -1196,8 +1226,6 @@ static void __init acpi_process_madt(void) acpi_ioapic = 1; smp_found_config = 1; - if (apic->setup_apic_routing) - apic->setup_apic_routing(); } } if (error == -EINVAL) { @@ -1268,23 +1296,6 @@ static int __init dmi_disable_acpi(const struct dmi_system_id *d) } /* - * Limit ACPI to CPU enumeration for HT - */ -static int __init force_acpi_ht(const struct dmi_system_id *d) -{ - if (!acpi_force) { - printk(KERN_NOTICE "%s detected: force use of acpi=ht\n", - d->ident); - disable_acpi(); - acpi_ht = 1; - } else { - printk(KERN_NOTICE - "Warning: acpi=force overrules DMI blacklist: acpi=ht\n"); - } - return 0; -} - -/* * Force ignoring BIOS IRQ0 pin2 override */ static int __init dmi_ignore_irq0_timer_override(const struct dmi_system_id *d) @@ -1320,90 +1331,6 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = { }, /* - * Boxes that need acpi=ht - */ - { - .callback = force_acpi_ht, - .ident = "FSC Primergy T850", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "FUJITSU SIEMENS"), - DMI_MATCH(DMI_PRODUCT_NAME, "PRIMERGY T850"), - }, - }, - { - .callback = force_acpi_ht, - .ident = "HP VISUALIZE NT Workstation", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "Hewlett-Packard"), - DMI_MATCH(DMI_PRODUCT_NAME, "HP VISUALIZE NT Workstation"), - }, - }, - { - .callback = force_acpi_ht, - .ident = "Compaq Workstation W8000", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Compaq"), - DMI_MATCH(DMI_PRODUCT_NAME, "Workstation W8000"), - }, - }, - { - .callback = force_acpi_ht, - .ident = "ASUS P2B-DS", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), - DMI_MATCH(DMI_BOARD_NAME, "P2B-DS"), - }, - }, - { - .callback = force_acpi_ht, - .ident = "ASUS CUR-DLS", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), - DMI_MATCH(DMI_BOARD_NAME, "CUR-DLS"), - }, - }, - { - .callback = force_acpi_ht, - .ident = "ABIT i440BX-W83977", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "ABIT <http://www.abit.com>"), - DMI_MATCH(DMI_BOARD_NAME, "i440BX-W83977 (BP6)"), - }, - }, - { - .callback = force_acpi_ht, - .ident = "IBM Bladecenter", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "IBM"), - DMI_MATCH(DMI_BOARD_NAME, "IBM eServer BladeCenter HS20"), - }, - }, - { - .callback = force_acpi_ht, - .ident = "IBM eServer xSeries 360", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "IBM"), - DMI_MATCH(DMI_BOARD_NAME, "eServer xSeries 360"), - }, - }, - { - .callback = force_acpi_ht, - .ident = "IBM eserver xSeries 330", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "IBM"), - DMI_MATCH(DMI_BOARD_NAME, "eserver xSeries 330"), - }, - }, - { - .callback = force_acpi_ht, - .ident = "IBM eserver xSeries 440", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "IBM"), - DMI_MATCH(DMI_PRODUCT_NAME, "eserver xSeries 440"), - }, - }, - - /* * Boxes that need ACPI PCI IRQ routing disabled */ { @@ -1528,16 +1455,10 @@ static struct dmi_system_id __initdata acpi_dmi_table_late[] = { * if acpi_blacklisted() acpi_disabled = 1; * acpi_irq_model=... * ... - * - * return value: (currently ignored) - * 0: success - * !0: failure */ -int __init acpi_boot_table_init(void) +void __init acpi_boot_table_init(void) { - int error; - dmi_check_system(acpi_dmi_table); /* @@ -1545,15 +1466,14 @@ int __init acpi_boot_table_init(void) * One exception: acpi=ht continues far enough to enumerate LAPICs */ if (acpi_disabled && !acpi_ht) - return 1; + return; /* * Initialize the ACPI boot-time table parser. */ - error = acpi_table_init(); - if (error) { + if (acpi_table_init()) { disable_acpi(); - return error; + return; } acpi_table_parse(ACPI_SIG_BOOT, acpi_parse_sbf); @@ -1561,18 +1481,15 @@ int __init acpi_boot_table_init(void) /* * blacklist may disable ACPI entirely */ - error = acpi_blacklisted(); - if (error) { + if (acpi_blacklisted()) { if (acpi_force) { printk(KERN_WARNING PREFIX "acpi=force override\n"); } else { printk(KERN_WARNING PREFIX "Disabling ACPI support\n"); disable_acpi(); - return error; + return; } } - - return 0; } int __init early_acpi_boot_init(void) @@ -1618,6 +1535,9 @@ int __init acpi_boot_init(void) acpi_table_parse(ACPI_SIG_HPET, acpi_parse_hpet); + if (!acpi_noirq) + x86_init.pci.init = pci_acpi_init; + return 0; } @@ -1642,8 +1562,10 @@ static int __init parse_acpi(char *arg) } /* Limit ACPI just to boot-time to enable HT */ else if (strcmp(arg, "ht") == 0) { - if (!acpi_force) + if (!acpi_force) { + printk(KERN_WARNING "acpi=ht will be removed in Linux-2.6.35\n"); disable_acpi(); + } acpi_ht = 1; } /* acpi=rsdt use RSDT instead of XSDT */ diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c index 59cdfa4686b2..2e837f5080fe 100644 --- a/arch/x86/kernel/acpi/cstate.c +++ b/arch/x86/kernel/acpi/cstate.c @@ -48,7 +48,7 @@ void acpi_processor_power_init_bm_check(struct acpi_processor_flags *flags, * P4, Core and beyond CPUs */ if (c->x86_vendor == X86_VENDOR_INTEL && - (c->x86 > 0xf || (c->x86 == 6 && c->x86_model >= 14))) + (c->x86 > 0xf || (c->x86 == 6 && c->x86_model >= 0x0f))) flags->bm_control = 0; } EXPORT_SYMBOL(acpi_processor_power_init_bm_check); diff --git a/arch/x86/kernel/acpi/processor.c b/arch/x86/kernel/acpi/processor.c deleted file mode 100644 index d296f4a195c9..000000000000 --- a/arch/x86/kernel/acpi/processor.c +++ /dev/null @@ -1,100 +0,0 @@ -/* - * Copyright (C) 2005 Intel Corporation - * Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> - * - Added _PDC for platforms with Intel CPUs - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/acpi.h> - -#include <acpi/processor.h> -#include <asm/acpi.h> - -static void init_intel_pdc(struct acpi_processor *pr, struct cpuinfo_x86 *c) -{ - struct acpi_object_list *obj_list; - union acpi_object *obj; - u32 *buf; - - /* allocate and initialize pdc. It will be used later. */ - obj_list = kmalloc(sizeof(struct acpi_object_list), GFP_KERNEL); - if (!obj_list) { - printk(KERN_ERR "Memory allocation error\n"); - return; - } - - obj = kmalloc(sizeof(union acpi_object), GFP_KERNEL); - if (!obj) { - printk(KERN_ERR "Memory allocation error\n"); - kfree(obj_list); - return; - } - - buf = kmalloc(12, GFP_KERNEL); - if (!buf) { - printk(KERN_ERR "Memory allocation error\n"); - kfree(obj); - kfree(obj_list); - return; - } - - buf[0] = ACPI_PDC_REVISION_ID; - buf[1] = 1; - buf[2] = ACPI_PDC_C_CAPABILITY_SMP; - - /* - * The default of PDC_SMP_T_SWCOORD bit is set for intel x86 cpu so - * that OSPM is capable of native ACPI throttling software - * coordination using BIOS supplied _TSD info. - */ - buf[2] |= ACPI_PDC_SMP_T_SWCOORD; - if (cpu_has(c, X86_FEATURE_EST)) - buf[2] |= ACPI_PDC_EST_CAPABILITY_SWSMP; - - if (cpu_has(c, X86_FEATURE_ACPI)) - buf[2] |= ACPI_PDC_T_FFH; - - /* - * If mwait/monitor is unsupported, C2/C3_FFH will be disabled - */ - if (!cpu_has(c, X86_FEATURE_MWAIT)) - buf[2] &= ~(ACPI_PDC_C_C2C3_FFH); - - obj->type = ACPI_TYPE_BUFFER; - obj->buffer.length = 12; - obj->buffer.pointer = (u8 *) buf; - obj_list->count = 1; - obj_list->pointer = obj; - pr->pdc = obj_list; - - return; -} - - -/* Initialize _PDC data based on the CPU vendor */ -void arch_acpi_processor_init_pdc(struct acpi_processor *pr) -{ - struct cpuinfo_x86 *c = &cpu_data(pr->id); - - pr->pdc = NULL; - if (c->x86_vendor == X86_VENDOR_INTEL) - init_intel_pdc(pr, c); - - return; -} - -EXPORT_SYMBOL(arch_acpi_processor_init_pdc); - -void arch_acpi_processor_cleanup_pdc(struct acpi_processor *pr) -{ - if (pr->pdc) { - kfree(pr->pdc->pointer->buffer.pointer); - kfree(pr->pdc->pointer); - kfree(pr->pdc); - pr->pdc = NULL; - } -} - -EXPORT_SYMBOL(arch_acpi_processor_cleanup_pdc); diff --git a/arch/x86/kernel/acpi/realmode/wakeup.lds.S b/arch/x86/kernel/acpi/realmode/wakeup.lds.S index 7da00b799cda..060fff8f5c5b 100644 --- a/arch/x86/kernel/acpi/realmode/wakeup.lds.S +++ b/arch/x86/kernel/acpi/realmode/wakeup.lds.S @@ -57,5 +57,8 @@ SECTIONS *(.note*) } + /* + * The ASSERT() sink to . is intentional, for binutils 2.14 compatibility: + */ . = ASSERT(_end <= WAKEUP_SIZE, "Wakeup too big!"); } diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index ca93638ba430..f9961034e557 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -78,12 +78,9 @@ int acpi_save_state_mem(void) #ifndef CONFIG_64BIT store_gdt((struct desc_ptr *)&header->pmode_gdt); - header->pmode_efer_low = nx_enabled; - if (header->pmode_efer_low & 1) { - /* This is strange, why not save efer, always? */ - rdmsr(MSR_EFER, header->pmode_efer_low, - header->pmode_efer_high); - } + if (rdmsr_safe(MSR_EFER, &header->pmode_efer_low, + &header->pmode_efer_high)) + header->pmode_efer_low = header->pmode_efer_high = 0; #endif /* !CONFIG_64BIT */ header->pmode_cr0 = read_cr0(); @@ -119,29 +116,32 @@ void acpi_restore_state_mem(void) /** - * acpi_reserve_bootmem - do _very_ early ACPI initialisation + * acpi_reserve_wakeup_memory - do _very_ early ACPI initialisation * * We allocate a page from the first 1MB of memory for the wakeup * routine for when we come back from a sleep state. The * runtime allocator allows specification of <16MB pages, but not * <1MB pages. */ -void __init acpi_reserve_bootmem(void) +void __init acpi_reserve_wakeup_memory(void) { + unsigned long mem; + if ((&wakeup_code_end - &wakeup_code_start) > WAKEUP_SIZE) { printk(KERN_ERR "ACPI: Wakeup code way too big, S3 disabled.\n"); return; } - acpi_realmode = (unsigned long)alloc_bootmem_low(WAKEUP_SIZE); + mem = find_e820_area(0, 1<<20, WAKEUP_SIZE, PAGE_SIZE); - if (!acpi_realmode) { + if (mem == -1L) { printk(KERN_ERR "ACPI: Cannot allocate lowmem, S3 disabled.\n"); return; } - - acpi_wakeup_address = virt_to_phys((void *)acpi_realmode); + acpi_realmode = (unsigned long) phys_to_virt(mem); + acpi_wakeup_address = mem; + reserve_early(mem, mem + WAKEUP_SIZE, "ACPI WAKEUP"); } @@ -162,6 +162,8 @@ static int __init acpi_sleep_setup(char *str) #endif if (strncmp(str, "old_ordering", 12) == 0) acpi_old_suspend_ordering(); + if (strncmp(str, "sci_force_enable", 16) == 0) + acpi_set_sci_en_on_resume(); str = strchr(str, ','); if (str != NULL) str += strspn(str, ", \t"); diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index de7353c0ce9c..3a4bf35c179b 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -7,6 +7,7 @@ #include <linux/mm.h> #include <linux/vmalloc.h> #include <linux/memory.h> +#include <linux/stop_machine.h> #include <asm/alternative.h> #include <asm/sections.h> #include <asm/pgtable.h> @@ -205,7 +206,7 @@ void __init_or_module apply_alternatives(struct alt_instr *start, struct alt_instr *end) { struct alt_instr *a; - char insnbuf[MAX_PATCH_LEN]; + u8 insnbuf[MAX_PATCH_LEN]; DPRINTK("%s: alt table %p -> %p\n", __func__, start, end); for (a = start; a < end; a++) { @@ -223,6 +224,8 @@ void __init_or_module apply_alternatives(struct alt_instr *start, } #endif memcpy(insnbuf, a->replacement, a->replacementlen); + if (*insnbuf == 0xe8 && a->replacementlen == 5) + *(s32 *)(insnbuf + 1) += a->replacement - a->instr; add_nops(insnbuf + a->replacementlen, a->instrlen - a->replacementlen); text_poke_early(instr, insnbuf, a->instrlen); @@ -390,6 +393,24 @@ void alternatives_smp_switch(int smp) mutex_unlock(&smp_alt); } +/* Return 1 if the address range is reserved for smp-alternatives */ +int alternatives_text_reserved(void *start, void *end) +{ + struct smp_alt_module *mod; + u8 **ptr; + u8 *text_start = start; + u8 *text_end = end; + + list_for_each_entry(mod, &smp_alt_modules, next) { + if (mod->text > text_end || mod->text_end < text_start) + continue; + for (ptr = mod->locks; ptr < mod->locks_end; ptr++) + if (text_start <= *ptr && text_end >= *ptr) + return 1; + } + + return 0; +} #endif #ifdef CONFIG_PARAVIRT @@ -552,3 +573,62 @@ void *__kprobes text_poke(void *addr, const void *opcode, size_t len) local_irq_restore(flags); return addr; } + +/* + * Cross-modifying kernel text with stop_machine(). + * This code originally comes from immediate value. + */ +static atomic_t stop_machine_first; +static int wrote_text; + +struct text_poke_params { + void *addr; + const void *opcode; + size_t len; +}; + +static int __kprobes stop_machine_text_poke(void *data) +{ + struct text_poke_params *tpp = data; + + if (atomic_dec_and_test(&stop_machine_first)) { + text_poke(tpp->addr, tpp->opcode, tpp->len); + smp_wmb(); /* Make sure other cpus see that this has run */ + wrote_text = 1; + } else { + while (!wrote_text) + cpu_relax(); + smp_mb(); /* Load wrote_text before following execution */ + } + + flush_icache_range((unsigned long)tpp->addr, + (unsigned long)tpp->addr + tpp->len); + return 0; +} + +/** + * text_poke_smp - Update instructions on a live kernel on SMP + * @addr: address to modify + * @opcode: source of the copy + * @len: length to copy + * + * Modify multi-byte instruction by using stop_machine() on SMP. This allows + * user to poke/set multi-byte text on SMP. Only non-NMI/MCE code modifying + * should be allowed, since stop_machine() does _not_ protect code against + * NMI and MCE. + * + * Note: Must be called under get_online_cpus() and text_mutex. + */ +void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len) +{ + struct text_poke_params tpp; + + tpp.addr = addr; + tpp.opcode = opcode; + tpp.len = len; + atomic_set(&stop_machine_first, 1); + wrote_text = 0; + stop_machine(stop_machine_text_poke, (void *)&tpp, NULL); + return addr; +} + diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 98f230f6a28d..adb0ba025702 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2007-2008 Advanced Micro Devices, Inc. + * Copyright (C) 2007-2009 Advanced Micro Devices, Inc. * Author: Joerg Roedel <joerg.roedel@amd.com> * Leo Duran <leo.duran@amd.com> * @@ -19,7 +19,7 @@ #include <linux/pci.h> #include <linux/gfp.h> -#include <linux/bitops.h> +#include <linux/bitmap.h> #include <linux/debugfs.h> #include <linux/scatterlist.h> #include <linux/dma-mapping.h> @@ -28,6 +28,7 @@ #include <asm/proto.h> #include <asm/iommu.h> #include <asm/gart.h> +#include <asm/amd_iommu_proto.h> #include <asm/amd_iommu_types.h> #include <asm/amd_iommu.h> @@ -56,20 +57,152 @@ struct iommu_cmd { u32 data[4]; }; -static int dma_ops_unity_map(struct dma_ops_domain *dma_dom, - struct unity_map_entry *e); -static struct dma_ops_domain *find_protection_domain(u16 devid); -static u64 *alloc_pte(struct protection_domain *domain, - unsigned long address, int end_lvl, - u64 **pte_page, gfp_t gfp); -static void dma_ops_reserve_addresses(struct dma_ops_domain *dom, - unsigned long start_page, - unsigned int pages); static void reset_iommu_command_buffer(struct amd_iommu *iommu); -static u64 *fetch_pte(struct protection_domain *domain, - unsigned long address, int map_size); static void update_domain(struct protection_domain *domain); +/**************************************************************************** + * + * Helper functions + * + ****************************************************************************/ + +static inline u16 get_device_id(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + + return calc_devid(pdev->bus->number, pdev->devfn); +} + +static struct iommu_dev_data *get_dev_data(struct device *dev) +{ + return dev->archdata.iommu; +} + +/* + * In this function the list of preallocated protection domains is traversed to + * find the domain for a specific device + */ +static struct dma_ops_domain *find_protection_domain(u16 devid) +{ + struct dma_ops_domain *entry, *ret = NULL; + unsigned long flags; + u16 alias = amd_iommu_alias_table[devid]; + + if (list_empty(&iommu_pd_list)) + return NULL; + + spin_lock_irqsave(&iommu_pd_list_lock, flags); + + list_for_each_entry(entry, &iommu_pd_list, list) { + if (entry->target_dev == devid || + entry->target_dev == alias) { + ret = entry; + break; + } + } + + spin_unlock_irqrestore(&iommu_pd_list_lock, flags); + + return ret; +} + +/* + * This function checks if the driver got a valid device from the caller to + * avoid dereferencing invalid pointers. + */ +static bool check_device(struct device *dev) +{ + u16 devid; + + if (!dev || !dev->dma_mask) + return false; + + /* No device or no PCI device */ + if (!dev || dev->bus != &pci_bus_type) + return false; + + devid = get_device_id(dev); + + /* Out of our scope? */ + if (devid > amd_iommu_last_bdf) + return false; + + if (amd_iommu_rlookup_table[devid] == NULL) + return false; + + return true; +} + +static int iommu_init_device(struct device *dev) +{ + struct iommu_dev_data *dev_data; + struct pci_dev *pdev; + u16 devid, alias; + + if (dev->archdata.iommu) + return 0; + + dev_data = kzalloc(sizeof(*dev_data), GFP_KERNEL); + if (!dev_data) + return -ENOMEM; + + dev_data->dev = dev; + + devid = get_device_id(dev); + alias = amd_iommu_alias_table[devid]; + pdev = pci_get_bus_and_slot(PCI_BUS(alias), alias & 0xff); + if (pdev) + dev_data->alias = &pdev->dev; + + atomic_set(&dev_data->bind, 0); + + dev->archdata.iommu = dev_data; + + + return 0; +} + +static void iommu_uninit_device(struct device *dev) +{ + kfree(dev->archdata.iommu); +} + +void __init amd_iommu_uninit_devices(void) +{ + struct pci_dev *pdev = NULL; + + for_each_pci_dev(pdev) { + + if (!check_device(&pdev->dev)) + continue; + + iommu_uninit_device(&pdev->dev); + } +} + +int __init amd_iommu_init_devices(void) +{ + struct pci_dev *pdev = NULL; + int ret = 0; + + for_each_pci_dev(pdev) { + + if (!check_device(&pdev->dev)) + continue; + + ret = iommu_init_device(&pdev->dev); + if (ret) + goto out_free; + } + + return 0; + +out_free: + + amd_iommu_uninit_devices(); + + return ret; +} #ifdef CONFIG_AMD_IOMMU_STATS /* @@ -90,7 +223,6 @@ DECLARE_STATS_COUNTER(alloced_io_mem); DECLARE_STATS_COUNTER(total_map_requests); static struct dentry *stats_dir; -static struct dentry *de_isolate; static struct dentry *de_fflush; static void amd_iommu_stats_add(struct __iommu_counter *cnt) @@ -108,9 +240,6 @@ static void amd_iommu_stats_init(void) if (stats_dir == NULL) return; - de_isolate = debugfs_create_bool("isolation", 0444, stats_dir, - (u32 *)&amd_iommu_isolate); - de_fflush = debugfs_create_bool("fullflush", 0444, stats_dir, (u32 *)&amd_iommu_unmap_flush); @@ -130,12 +259,6 @@ static void amd_iommu_stats_init(void) #endif -/* returns !0 if the IOMMU is caching non-present entries in its TLB */ -static int iommu_has_npcache(struct amd_iommu *iommu) -{ - return iommu->cap & (1UL << IOMMU_CAP_NPCACHE); -} - /**************************************************************************** * * Interrupt handling functions @@ -199,6 +322,7 @@ static void iommu_print_event(struct amd_iommu *iommu, void *__evt) break; case EVENT_TYPE_ILL_CMD: printk("ILLEGAL_COMMAND_ERROR address=0x%016llx]\n", address); + iommu->reset_in_progress = true; reset_iommu_command_buffer(iommu); dump_command(address); break; @@ -321,11 +445,8 @@ static void __iommu_wait_for_completion(struct amd_iommu *iommu) status &= ~MMIO_STATUS_COM_WAIT_INT_MASK; writel(status, iommu->mmio_base + MMIO_STATUS_OFFSET); - if (unlikely(i == EXIT_LOOP_COUNT)) { - spin_unlock(&iommu->lock); - reset_iommu_command_buffer(iommu); - spin_lock(&iommu->lock); - } + if (unlikely(i == EXIT_LOOP_COUNT)) + iommu->reset_in_progress = true; } /* @@ -372,26 +493,46 @@ static int iommu_completion_wait(struct amd_iommu *iommu) out: spin_unlock_irqrestore(&iommu->lock, flags); + if (iommu->reset_in_progress) + reset_iommu_command_buffer(iommu); + return 0; } +static void iommu_flush_complete(struct protection_domain *domain) +{ + int i; + + for (i = 0; i < amd_iommus_present; ++i) { + if (!domain->dev_iommu[i]) + continue; + + /* + * Devices of this domain are behind this IOMMU + * We need to wait for completion of all commands. + */ + iommu_completion_wait(amd_iommus[i]); + } +} + /* * Command send function for invalidating a device table entry */ -static int iommu_queue_inv_dev_entry(struct amd_iommu *iommu, u16 devid) +static int iommu_flush_device(struct device *dev) { + struct amd_iommu *iommu; struct iommu_cmd cmd; - int ret; + u16 devid; - BUG_ON(iommu == NULL); + devid = get_device_id(dev); + iommu = amd_iommu_rlookup_table[devid]; + /* Build command */ memset(&cmd, 0, sizeof(cmd)); CMD_SET_TYPE(&cmd, CMD_INV_DEV_ENTRY); cmd.data[0] = devid; - ret = iommu_queue_command(iommu, &cmd); - - return ret; + return iommu_queue_command(iommu, &cmd); } static void __iommu_build_inv_iommu_pages(struct iommu_cmd *cmd, u64 address, @@ -430,11 +571,11 @@ static int iommu_queue_inv_iommu_pages(struct amd_iommu *iommu, * It invalidates a single PTE if the range to flush is within a single * page. Otherwise it flushes the whole TLB of the IOMMU. */ -static int iommu_flush_pages(struct amd_iommu *iommu, u16 domid, - u64 address, size_t size) +static void __iommu_flush_pages(struct protection_domain *domain, + u64 address, size_t size, int pde) { - int s = 0; - unsigned pages = iommu_num_pages(address, size, PAGE_SIZE); + int s = 0, i; + unsigned long pages = iommu_num_pages(address, size, PAGE_SIZE); address &= PAGE_MASK; @@ -447,142 +588,212 @@ static int iommu_flush_pages(struct amd_iommu *iommu, u16 domid, s = 1; } - iommu_queue_inv_iommu_pages(iommu, address, domid, 0, s); - return 0; + for (i = 0; i < amd_iommus_present; ++i) { + if (!domain->dev_iommu[i]) + continue; + + /* + * Devices of this domain are behind this IOMMU + * We need a TLB flush + */ + iommu_queue_inv_iommu_pages(amd_iommus[i], address, + domain->id, pde, s); + } + + return; } -/* Flush the whole IO/TLB for a given protection domain */ -static void iommu_flush_tlb(struct amd_iommu *iommu, u16 domid) +static void iommu_flush_pages(struct protection_domain *domain, + u64 address, size_t size) { - u64 address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS; - - INC_STATS_COUNTER(domain_flush_single); + __iommu_flush_pages(domain, address, size, 0); +} - iommu_queue_inv_iommu_pages(iommu, address, domid, 0, 1); +/* Flush the whole IO/TLB for a given protection domain */ +static void iommu_flush_tlb(struct protection_domain *domain) +{ + __iommu_flush_pages(domain, 0, CMD_INV_IOMMU_ALL_PAGES_ADDRESS, 0); } /* Flush the whole IO/TLB for a given protection domain - including PDE */ -static void iommu_flush_tlb_pde(struct amd_iommu *iommu, u16 domid) +static void iommu_flush_tlb_pde(struct protection_domain *domain) { - u64 address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS; - - INC_STATS_COUNTER(domain_flush_single); - - iommu_queue_inv_iommu_pages(iommu, address, domid, 1, 1); + __iommu_flush_pages(domain, 0, CMD_INV_IOMMU_ALL_PAGES_ADDRESS, 1); } + /* - * This function flushes one domain on one IOMMU + * This function flushes the DTEs for all devices in domain */ -static void flush_domain_on_iommu(struct amd_iommu *iommu, u16 domid) +static void iommu_flush_domain_devices(struct protection_domain *domain) { - struct iommu_cmd cmd; + struct iommu_dev_data *dev_data; unsigned long flags; - __iommu_build_inv_iommu_pages(&cmd, CMD_INV_IOMMU_ALL_PAGES_ADDRESS, - domid, 1, 1); + spin_lock_irqsave(&domain->lock, flags); - spin_lock_irqsave(&iommu->lock, flags); - __iommu_queue_command(iommu, &cmd); - __iommu_completion_wait(iommu); - __iommu_wait_for_completion(iommu); - spin_unlock_irqrestore(&iommu->lock, flags); + list_for_each_entry(dev_data, &domain->dev_list, list) + iommu_flush_device(dev_data->dev); + + spin_unlock_irqrestore(&domain->lock, flags); } -static void flush_all_domains_on_iommu(struct amd_iommu *iommu) +static void iommu_flush_all_domain_devices(void) { - int i; + struct protection_domain *domain; + unsigned long flags; - for (i = 1; i < MAX_DOMAIN_ID; ++i) { - if (!test_bit(i, amd_iommu_pd_alloc_bitmap)) - continue; - flush_domain_on_iommu(iommu, i); + spin_lock_irqsave(&amd_iommu_pd_lock, flags); + + list_for_each_entry(domain, &amd_iommu_pd_list, list) { + iommu_flush_domain_devices(domain); + iommu_flush_complete(domain); } + spin_unlock_irqrestore(&amd_iommu_pd_lock, flags); +} + +void amd_iommu_flush_all_devices(void) +{ + iommu_flush_all_domain_devices(); } /* - * This function is used to flush the IO/TLB for a given protection domain - * on every IOMMU in the system + * This function uses heavy locking and may disable irqs for some time. But + * this is no issue because it is only called during resume. */ -static void iommu_flush_domain(u16 domid) +void amd_iommu_flush_all_domains(void) { - struct amd_iommu *iommu; + struct protection_domain *domain; + unsigned long flags; - INC_STATS_COUNTER(domain_flush_all); + spin_lock_irqsave(&amd_iommu_pd_lock, flags); - for_each_iommu(iommu) - flush_domain_on_iommu(iommu, domid); + list_for_each_entry(domain, &amd_iommu_pd_list, list) { + spin_lock(&domain->lock); + iommu_flush_tlb_pde(domain); + iommu_flush_complete(domain); + spin_unlock(&domain->lock); + } + + spin_unlock_irqrestore(&amd_iommu_pd_lock, flags); } -void amd_iommu_flush_all_domains(void) +static void reset_iommu_command_buffer(struct amd_iommu *iommu) { - struct amd_iommu *iommu; + pr_err("AMD-Vi: Resetting IOMMU command buffer\n"); - for_each_iommu(iommu) - flush_all_domains_on_iommu(iommu); + if (iommu->reset_in_progress) + panic("AMD-Vi: ILLEGAL_COMMAND_ERROR while resetting command buffer\n"); + + amd_iommu_reset_cmd_buffer(iommu); + amd_iommu_flush_all_devices(); + amd_iommu_flush_all_domains(); + + iommu->reset_in_progress = false; } -static void flush_all_devices_for_iommu(struct amd_iommu *iommu) +/**************************************************************************** + * + * The functions below are used the create the page table mappings for + * unity mapped regions. + * + ****************************************************************************/ + +/* + * This function is used to add another level to an IO page table. Adding + * another level increases the size of the address space by 9 bits to a size up + * to 64 bits. + */ +static bool increase_address_space(struct protection_domain *domain, + gfp_t gfp) { - int i; + u64 *pte; - for (i = 0; i <= amd_iommu_last_bdf; ++i) { - if (iommu != amd_iommu_rlookup_table[i]) - continue; + if (domain->mode == PAGE_MODE_6_LEVEL) + /* address space already 64 bit large */ + return false; - iommu_queue_inv_dev_entry(iommu, i); - iommu_completion_wait(iommu); - } + pte = (void *)get_zeroed_page(gfp); + if (!pte) + return false; + + *pte = PM_LEVEL_PDE(domain->mode, + virt_to_phys(domain->pt_root)); + domain->pt_root = pte; + domain->mode += 1; + domain->updated = true; + + return true; } -static void flush_devices_by_domain(struct protection_domain *domain) +static u64 *alloc_pte(struct protection_domain *domain, + unsigned long address, + int end_lvl, + u64 **pte_page, + gfp_t gfp) { - struct amd_iommu *iommu; - int i; + u64 *pte, *page; + int level; - for (i = 0; i <= amd_iommu_last_bdf; ++i) { - if ((domain == NULL && amd_iommu_pd_table[i] == NULL) || - (amd_iommu_pd_table[i] != domain)) - continue; + while (address > PM_LEVEL_SIZE(domain->mode)) + increase_address_space(domain, gfp); - iommu = amd_iommu_rlookup_table[i]; - if (!iommu) - continue; + level = domain->mode - 1; + pte = &domain->pt_root[PM_LEVEL_INDEX(level, address)]; + + while (level > end_lvl) { + if (!IOMMU_PTE_PRESENT(*pte)) { + page = (u64 *)get_zeroed_page(gfp); + if (!page) + return NULL; + *pte = PM_LEVEL_PDE(level, virt_to_phys(page)); + } - iommu_queue_inv_dev_entry(iommu, i); - iommu_completion_wait(iommu); + level -= 1; + + pte = IOMMU_PTE_PAGE(*pte); + + if (pte_page && level == end_lvl) + *pte_page = pte; + + pte = &pte[PM_LEVEL_INDEX(level, address)]; } + + return pte; } -static void reset_iommu_command_buffer(struct amd_iommu *iommu) +/* + * This function checks if there is a PTE for a given dma address. If + * there is one, it returns the pointer to it. + */ +static u64 *fetch_pte(struct protection_domain *domain, + unsigned long address, int map_size) { - pr_err("AMD-Vi: Resetting IOMMU command buffer\n"); + int level; + u64 *pte; - if (iommu->reset_in_progress) - panic("AMD-Vi: ILLEGAL_COMMAND_ERROR while resetting command buffer\n"); + level = domain->mode - 1; + pte = &domain->pt_root[PM_LEVEL_INDEX(level, address)]; - iommu->reset_in_progress = true; + while (level > map_size) { + if (!IOMMU_PTE_PRESENT(*pte)) + return NULL; - amd_iommu_reset_cmd_buffer(iommu); - flush_all_devices_for_iommu(iommu); - flush_all_domains_on_iommu(iommu); + level -= 1; - iommu->reset_in_progress = false; -} + pte = IOMMU_PTE_PAGE(*pte); + pte = &pte[PM_LEVEL_INDEX(level, address)]; -void amd_iommu_flush_all_devices(void) -{ - flush_devices_by_domain(NULL); -} + if ((PM_PTE_LEVEL(*pte) == 0) && level != map_size) { + pte = NULL; + break; + } + } -/**************************************************************************** - * - * The functions below are used the create the page table mappings for - * unity mapped regions. - * - ****************************************************************************/ + return pte; +} /* * Generic mapping functions. It maps a physical address into a DMA @@ -654,28 +865,6 @@ static int iommu_for_unity_map(struct amd_iommu *iommu, } /* - * Init the unity mappings for a specific IOMMU in the system - * - * Basically iterates over all unity mapping entries and applies them to - * the default domain DMA of that IOMMU if necessary. - */ -static int iommu_init_unity_mappings(struct amd_iommu *iommu) -{ - struct unity_map_entry *entry; - int ret; - - list_for_each_entry(entry, &amd_iommu_unity_map, list) { - if (!iommu_for_unity_map(iommu, entry)) - continue; - ret = dma_ops_unity_map(iommu->default_dom, entry); - if (ret) - return ret; - } - - return 0; -} - -/* * This function actually applies the mapping to the page table of the * dma_ops domain. */ @@ -704,6 +893,28 @@ static int dma_ops_unity_map(struct dma_ops_domain *dma_dom, } /* + * Init the unity mappings for a specific IOMMU in the system + * + * Basically iterates over all unity mapping entries and applies them to + * the default domain DMA of that IOMMU if necessary. + */ +static int iommu_init_unity_mappings(struct amd_iommu *iommu) +{ + struct unity_map_entry *entry; + int ret; + + list_for_each_entry(entry, &amd_iommu_unity_map, list) { + if (!iommu_for_unity_map(iommu, entry)) + continue; + ret = dma_ops_unity_map(iommu->default_dom, entry); + if (ret) + return ret; + } + + return 0; +} + +/* * Inits the unity mappings required for a specific device */ static int init_unity_mappings_for_device(struct dma_ops_domain *dma_dom, @@ -740,34 +951,23 @@ static int init_unity_mappings_for_device(struct dma_ops_domain *dma_dom, */ /* - * This function checks if there is a PTE for a given dma address. If - * there is one, it returns the pointer to it. + * Used to reserve address ranges in the aperture (e.g. for exclusion + * ranges. */ -static u64 *fetch_pte(struct protection_domain *domain, - unsigned long address, int map_size) +static void dma_ops_reserve_addresses(struct dma_ops_domain *dom, + unsigned long start_page, + unsigned int pages) { - int level; - u64 *pte; - - level = domain->mode - 1; - pte = &domain->pt_root[PM_LEVEL_INDEX(level, address)]; - - while (level > map_size) { - if (!IOMMU_PTE_PRESENT(*pte)) - return NULL; - - level -= 1; + unsigned int i, last_page = dom->aperture_size >> PAGE_SHIFT; - pte = IOMMU_PTE_PAGE(*pte); - pte = &pte[PM_LEVEL_INDEX(level, address)]; + if (start_page + pages > last_page) + pages = last_page - start_page; - if ((PM_PTE_LEVEL(*pte) == 0) && level != map_size) { - pte = NULL; - break; - } + for (i = start_page; i < start_page + pages; ++i) { + int index = i / APERTURE_RANGE_PAGES; + int page = i % APERTURE_RANGE_PAGES; + __set_bit(page, dom->aperture[index]->bitmap); } - - return pte; } /* @@ -775,12 +975,12 @@ static u64 *fetch_pte(struct protection_domain *domain, * aperture in case of dma_ops domain allocation or address allocation * failure. */ -static int alloc_new_range(struct amd_iommu *iommu, - struct dma_ops_domain *dma_dom, +static int alloc_new_range(struct dma_ops_domain *dma_dom, bool populate, gfp_t gfp) { int index = dma_dom->aperture_size >> APERTURE_RANGE_SHIFT; - int i; + struct amd_iommu *iommu; + unsigned long i; #ifdef CONFIG_IOMMU_STRESS populate = false; @@ -819,14 +1019,17 @@ static int alloc_new_range(struct amd_iommu *iommu, dma_dom->aperture_size += APERTURE_RANGE_SIZE; /* Intialize the exclusion range if necessary */ - if (iommu->exclusion_start && - iommu->exclusion_start >= dma_dom->aperture[index]->offset && - iommu->exclusion_start < dma_dom->aperture_size) { - unsigned long startpage = iommu->exclusion_start >> PAGE_SHIFT; - int pages = iommu_num_pages(iommu->exclusion_start, - iommu->exclusion_length, - PAGE_SIZE); - dma_ops_reserve_addresses(dma_dom, startpage, pages); + for_each_iommu(iommu) { + if (iommu->exclusion_start && + iommu->exclusion_start >= dma_dom->aperture[index]->offset + && iommu->exclusion_start < dma_dom->aperture_size) { + unsigned long startpage; + int pages = iommu_num_pages(iommu->exclusion_start, + iommu->exclusion_length, + PAGE_SIZE); + startpage = iommu->exclusion_start >> PAGE_SHIFT; + dma_ops_reserve_addresses(dma_dom, startpage, pages); + } } /* @@ -928,7 +1131,7 @@ static unsigned long dma_ops_alloc_addresses(struct device *dev, } if (unlikely(address == -1)) - address = bad_dma_address; + address = DMA_ERROR_CODE; WARN_ON((address + (PAGE_SIZE*pages)) > dom->aperture_size); @@ -959,7 +1162,7 @@ static void dma_ops_free_addresses(struct dma_ops_domain *dom, address = (address % APERTURE_RANGE_SIZE) >> PAGE_SHIFT; - iommu_area_free(range->bitmap, address, pages); + bitmap_clear(range->bitmap, address, pages); } @@ -973,6 +1176,31 @@ static void dma_ops_free_addresses(struct dma_ops_domain *dom, * ****************************************************************************/ +/* + * This function adds a protection domain to the global protection domain list + */ +static void add_domain_to_list(struct protection_domain *domain) +{ + unsigned long flags; + + spin_lock_irqsave(&amd_iommu_pd_lock, flags); + list_add(&domain->list, &amd_iommu_pd_list); + spin_unlock_irqrestore(&amd_iommu_pd_lock, flags); +} + +/* + * This function removes a protection domain to the global + * protection domain list + */ +static void del_domain_from_list(struct protection_domain *domain) +{ + unsigned long flags; + + spin_lock_irqsave(&amd_iommu_pd_lock, flags); + list_del(&domain->list); + spin_unlock_irqrestore(&amd_iommu_pd_lock, flags); +} + static u16 domain_id_alloc(void) { unsigned long flags; @@ -1000,26 +1228,6 @@ static void domain_id_free(int id) write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); } -/* - * Used to reserve address ranges in the aperture (e.g. for exclusion - * ranges. - */ -static void dma_ops_reserve_addresses(struct dma_ops_domain *dom, - unsigned long start_page, - unsigned int pages) -{ - unsigned int i, last_page = dom->aperture_size >> PAGE_SHIFT; - - if (start_page + pages > last_page) - pages = last_page - start_page; - - for (i = start_page; i < start_page + pages; ++i) { - int index = i / APERTURE_RANGE_PAGES; - int page = i % APERTURE_RANGE_PAGES; - __set_bit(page, dom->aperture[index]->bitmap); - } -} - static void free_pagetable(struct protection_domain *domain) { int i, j; @@ -1061,6 +1269,8 @@ static void dma_ops_domain_free(struct dma_ops_domain *dom) if (!dom) return; + del_domain_from_list(&dom->domain); + free_pagetable(&dom->domain); for (i = 0; i < APERTURE_MAX_RANGES; ++i) { @@ -1078,7 +1288,7 @@ static void dma_ops_domain_free(struct dma_ops_domain *dom) * It also intializes the page table and the address allocator data * structures required for the dma_ops interface */ -static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu) +static struct dma_ops_domain *dma_ops_domain_alloc(void) { struct dma_ops_domain *dma_dom; @@ -1091,6 +1301,7 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu) dma_dom->domain.id = domain_id_alloc(); if (dma_dom->domain.id == 0) goto free_dma_dom; + INIT_LIST_HEAD(&dma_dom->domain.dev_list); dma_dom->domain.mode = PAGE_MODE_2_LEVEL; dma_dom->domain.pt_root = (void *)get_zeroed_page(GFP_KERNEL); dma_dom->domain.flags = PD_DMA_OPS_MASK; @@ -1101,7 +1312,9 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu) dma_dom->need_flush = false; dma_dom->target_dev = 0xffff; - if (alloc_new_range(iommu, dma_dom, true, GFP_KERNEL)) + add_domain_to_list(&dma_dom->domain); + + if (alloc_new_range(dma_dom, true, GFP_KERNEL)) goto free_dma_dom; /* @@ -1129,22 +1342,6 @@ static bool dma_ops_domain(struct protection_domain *domain) return domain->flags & PD_DMA_OPS_MASK; } -/* - * Find out the protection domain structure for a given PCI device. This - * will give us the pointer to the page table root for example. - */ -static struct protection_domain *domain_for_device(u16 devid) -{ - struct protection_domain *dom; - unsigned long flags; - - read_lock_irqsave(&amd_iommu_devtable_lock, flags); - dom = amd_iommu_pd_table[devid]; - read_unlock_irqrestore(&amd_iommu_devtable_lock, flags); - - return dom; -} - static void set_dte_entry(u16 devid, struct protection_domain *domain) { u64 pte_root = virt_to_phys(domain->pt_root); @@ -1156,42 +1353,123 @@ static void set_dte_entry(u16 devid, struct protection_domain *domain) amd_iommu_dev_table[devid].data[2] = domain->id; amd_iommu_dev_table[devid].data[1] = upper_32_bits(pte_root); amd_iommu_dev_table[devid].data[0] = lower_32_bits(pte_root); +} + +static void clear_dte_entry(u16 devid) +{ + /* remove entry from the device table seen by the hardware */ + amd_iommu_dev_table[devid].data[0] = IOMMU_PTE_P | IOMMU_PTE_TV; + amd_iommu_dev_table[devid].data[1] = 0; + amd_iommu_dev_table[devid].data[2] = 0; + + amd_iommu_apply_erratum_63(devid); +} + +static void do_attach(struct device *dev, struct protection_domain *domain) +{ + struct iommu_dev_data *dev_data; + struct amd_iommu *iommu; + u16 devid; + + devid = get_device_id(dev); + iommu = amd_iommu_rlookup_table[devid]; + dev_data = get_dev_data(dev); + + /* Update data structures */ + dev_data->domain = domain; + list_add(&dev_data->list, &domain->dev_list); + set_dte_entry(devid, domain); + + /* Do reference counting */ + domain->dev_iommu[iommu->index] += 1; + domain->dev_cnt += 1; - amd_iommu_pd_table[devid] = domain; + /* Flush the DTE entry */ + iommu_flush_device(dev); +} + +static void do_detach(struct device *dev) +{ + struct iommu_dev_data *dev_data; + struct amd_iommu *iommu; + u16 devid; + + devid = get_device_id(dev); + iommu = amd_iommu_rlookup_table[devid]; + dev_data = get_dev_data(dev); + + /* decrease reference counters */ + dev_data->domain->dev_iommu[iommu->index] -= 1; + dev_data->domain->dev_cnt -= 1; + + /* Update data structures */ + dev_data->domain = NULL; + list_del(&dev_data->list); + clear_dte_entry(devid); + + /* Flush the DTE entry */ + iommu_flush_device(dev); } /* * If a device is not yet associated with a domain, this function does * assigns it visible for the hardware */ -static void __attach_device(struct amd_iommu *iommu, - struct protection_domain *domain, - u16 devid) +static int __attach_device(struct device *dev, + struct protection_domain *domain) { + struct iommu_dev_data *dev_data, *alias_data; + + dev_data = get_dev_data(dev); + alias_data = get_dev_data(dev_data->alias); + + if (!alias_data) + return -EINVAL; + /* lock domain */ spin_lock(&domain->lock); - /* update DTE entry */ - set_dte_entry(devid, domain); + /* Some sanity checks */ + if (alias_data->domain != NULL && + alias_data->domain != domain) + return -EBUSY; - domain->dev_cnt += 1; + if (dev_data->domain != NULL && + dev_data->domain != domain) + return -EBUSY; + + /* Do real assignment */ + if (dev_data->alias != dev) { + alias_data = get_dev_data(dev_data->alias); + if (alias_data->domain == NULL) + do_attach(dev_data->alias, domain); + + atomic_inc(&alias_data->bind); + } + + if (dev_data->domain == NULL) + do_attach(dev, domain); + + atomic_inc(&dev_data->bind); /* ready */ spin_unlock(&domain->lock); + + return 0; } /* * If a device is not yet associated with a domain, this function does * assigns it visible for the hardware */ -static void attach_device(struct amd_iommu *iommu, - struct protection_domain *domain, - u16 devid) +static int attach_device(struct device *dev, + struct protection_domain *domain) { unsigned long flags; + int ret; write_lock_irqsave(&amd_iommu_devtable_lock, flags); - __attach_device(iommu, domain, devid); + ret = __attach_device(dev, domain); write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); /* @@ -1199,96 +1477,130 @@ static void attach_device(struct amd_iommu *iommu, * left the caches in the IOMMU dirty. So we have to flush * here to evict all dirty stuff. */ - iommu_queue_inv_dev_entry(iommu, devid); - iommu_flush_tlb_pde(iommu, domain->id); + iommu_flush_tlb_pde(domain); + + return ret; } /* * Removes a device from a protection domain (unlocked) */ -static void __detach_device(struct protection_domain *domain, u16 devid) +static void __detach_device(struct device *dev) { + struct iommu_dev_data *dev_data = get_dev_data(dev); + struct iommu_dev_data *alias_data; + struct protection_domain *domain; + unsigned long flags; - /* lock domain */ - spin_lock(&domain->lock); + BUG_ON(!dev_data->domain); - /* remove domain from the lookup table */ - amd_iommu_pd_table[devid] = NULL; + domain = dev_data->domain; - /* remove entry from the device table seen by the hardware */ - amd_iommu_dev_table[devid].data[0] = IOMMU_PTE_P | IOMMU_PTE_TV; - amd_iommu_dev_table[devid].data[1] = 0; - amd_iommu_dev_table[devid].data[2] = 0; + spin_lock_irqsave(&domain->lock, flags); - /* decrease reference counter */ - domain->dev_cnt -= 1; + if (dev_data->alias != dev) { + alias_data = get_dev_data(dev_data->alias); + if (atomic_dec_and_test(&alias_data->bind)) + do_detach(dev_data->alias); + } - /* ready */ - spin_unlock(&domain->lock); + if (atomic_dec_and_test(&dev_data->bind)) + do_detach(dev); + + spin_unlock_irqrestore(&domain->lock, flags); /* * If we run in passthrough mode the device must be assigned to the - * passthrough domain if it is detached from any other domain + * passthrough domain if it is detached from any other domain. + * Make sure we can deassign from the pt_domain itself. */ - if (iommu_pass_through) { - struct amd_iommu *iommu = amd_iommu_rlookup_table[devid]; - __attach_device(iommu, pt_domain, devid); - } + if (iommu_pass_through && + (dev_data->domain == NULL && domain != pt_domain)) + __attach_device(dev, pt_domain); } /* * Removes a device from a protection domain (with devtable_lock held) */ -static void detach_device(struct protection_domain *domain, u16 devid) +static void detach_device(struct device *dev) { unsigned long flags; /* lock device table */ write_lock_irqsave(&amd_iommu_devtable_lock, flags); - __detach_device(domain, devid); + __detach_device(dev); write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); } +/* + * Find out the protection domain structure for a given PCI device. This + * will give us the pointer to the page table root for example. + */ +static struct protection_domain *domain_for_device(struct device *dev) +{ + struct protection_domain *dom; + struct iommu_dev_data *dev_data, *alias_data; + unsigned long flags; + u16 devid, alias; + + devid = get_device_id(dev); + alias = amd_iommu_alias_table[devid]; + dev_data = get_dev_data(dev); + alias_data = get_dev_data(dev_data->alias); + if (!alias_data) + return NULL; + + read_lock_irqsave(&amd_iommu_devtable_lock, flags); + dom = dev_data->domain; + if (dom == NULL && + alias_data->domain != NULL) { + __attach_device(dev, alias_data->domain); + dom = alias_data->domain; + } + + read_unlock_irqrestore(&amd_iommu_devtable_lock, flags); + + return dom; +} + static int device_change_notifier(struct notifier_block *nb, unsigned long action, void *data) { struct device *dev = data; - struct pci_dev *pdev = to_pci_dev(dev); - u16 devid = calc_devid(pdev->bus->number, pdev->devfn); + u16 devid; struct protection_domain *domain; struct dma_ops_domain *dma_domain; struct amd_iommu *iommu; unsigned long flags; - if (devid > amd_iommu_last_bdf) - goto out; - - devid = amd_iommu_alias_table[devid]; - - iommu = amd_iommu_rlookup_table[devid]; - if (iommu == NULL) - goto out; - - domain = domain_for_device(devid); + if (!check_device(dev)) + return 0; - if (domain && !dma_ops_domain(domain)) - WARN_ONCE(1, "AMD IOMMU WARNING: device %s already bound " - "to a non-dma-ops domain\n", dev_name(dev)); + devid = get_device_id(dev); + iommu = amd_iommu_rlookup_table[devid]; switch (action) { case BUS_NOTIFY_UNBOUND_DRIVER: + + domain = domain_for_device(dev); + if (!domain) goto out; if (iommu_pass_through) break; - detach_device(domain, devid); + detach_device(dev); break; case BUS_NOTIFY_ADD_DEVICE: + + iommu_init_device(dev); + + domain = domain_for_device(dev); + /* allocate a protection domain if a device is added */ dma_domain = find_protection_domain(devid); if (dma_domain) goto out; - dma_domain = dma_ops_domain_alloc(iommu); + dma_domain = dma_ops_domain_alloc(); if (!dma_domain) goto out; dma_domain->target_dev = devid; @@ -1298,11 +1610,15 @@ static int device_change_notifier(struct notifier_block *nb, spin_unlock_irqrestore(&iommu_pd_list_lock, flags); break; + case BUS_NOTIFY_DEL_DEVICE: + + iommu_uninit_device(dev); + default: goto out; } - iommu_queue_inv_dev_entry(iommu, devid); + iommu_flush_device(dev); iommu_completion_wait(iommu); out: @@ -1313,6 +1629,11 @@ static struct notifier_block device_nb = { .notifier_call = device_change_notifier, }; +void amd_iommu_init_notifier(void) +{ + bus_register_notifier(&pci_bus_type, &device_nb); +} + /***************************************************************************** * * The next functions belong to the dma_ops mapping/unmapping code. @@ -1320,106 +1641,46 @@ static struct notifier_block device_nb = { *****************************************************************************/ /* - * This function checks if the driver got a valid device from the caller to - * avoid dereferencing invalid pointers. - */ -static bool check_device(struct device *dev) -{ - if (!dev || !dev->dma_mask) - return false; - - return true; -} - -/* - * In this function the list of preallocated protection domains is traversed to - * find the domain for a specific device - */ -static struct dma_ops_domain *find_protection_domain(u16 devid) -{ - struct dma_ops_domain *entry, *ret = NULL; - unsigned long flags; - - if (list_empty(&iommu_pd_list)) - return NULL; - - spin_lock_irqsave(&iommu_pd_list_lock, flags); - - list_for_each_entry(entry, &iommu_pd_list, list) { - if (entry->target_dev == devid) { - ret = entry; - break; - } - } - - spin_unlock_irqrestore(&iommu_pd_list_lock, flags); - - return ret; -} - -/* * In the dma_ops path we only have the struct device. This function * finds the corresponding IOMMU, the protection domain and the * requestor id for a given device. * If the device is not yet associated with a domain this is also done * in this function. */ -static int get_device_resources(struct device *dev, - struct amd_iommu **iommu, - struct protection_domain **domain, - u16 *bdf) +static struct protection_domain *get_domain(struct device *dev) { + struct protection_domain *domain; struct dma_ops_domain *dma_dom; - struct pci_dev *pcidev; - u16 _bdf; - - *iommu = NULL; - *domain = NULL; - *bdf = 0xffff; + u16 devid = get_device_id(dev); - if (dev->bus != &pci_bus_type) - return 0; - - pcidev = to_pci_dev(dev); - _bdf = calc_devid(pcidev->bus->number, pcidev->devfn); + if (!check_device(dev)) + return ERR_PTR(-EINVAL); - /* device not translated by any IOMMU in the system? */ - if (_bdf > amd_iommu_last_bdf) - return 0; + domain = domain_for_device(dev); + if (domain != NULL && !dma_ops_domain(domain)) + return ERR_PTR(-EBUSY); - *bdf = amd_iommu_alias_table[_bdf]; + if (domain != NULL) + return domain; - *iommu = amd_iommu_rlookup_table[*bdf]; - if (*iommu == NULL) - return 0; - *domain = domain_for_device(*bdf); - if (*domain == NULL) { - dma_dom = find_protection_domain(*bdf); - if (!dma_dom) - dma_dom = (*iommu)->default_dom; - *domain = &dma_dom->domain; - attach_device(*iommu, *domain, *bdf); - DUMP_printk("Using protection domain %d for device %s\n", - (*domain)->id, dev_name(dev)); - } - - if (domain_for_device(_bdf) == NULL) - attach_device(*iommu, *domain, _bdf); + /* Device not bount yet - bind it */ + dma_dom = find_protection_domain(devid); + if (!dma_dom) + dma_dom = amd_iommu_rlookup_table[devid]->default_dom; + attach_device(dev, &dma_dom->domain); + DUMP_printk("Using protection domain %d for device %s\n", + dma_dom->domain.id, dev_name(dev)); - return 1; + return &dma_dom->domain; } static void update_device_table(struct protection_domain *domain) { - unsigned long flags; - int i; + struct iommu_dev_data *dev_data; - for (i = 0; i <= amd_iommu_last_bdf; ++i) { - if (amd_iommu_pd_table[i] != domain) - continue; - write_lock_irqsave(&amd_iommu_devtable_lock, flags); - set_dte_entry(i, domain); - write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); + list_for_each_entry(dev_data, &domain->dev_list, list) { + u16 devid = get_device_id(dev_data->dev); + set_dte_entry(devid, domain); } } @@ -1429,76 +1690,13 @@ static void update_domain(struct protection_domain *domain) return; update_device_table(domain); - flush_devices_by_domain(domain); - iommu_flush_domain(domain->id); + iommu_flush_domain_devices(domain); + iommu_flush_tlb_pde(domain); domain->updated = false; } /* - * This function is used to add another level to an IO page table. Adding - * another level increases the size of the address space by 9 bits to a size up - * to 64 bits. - */ -static bool increase_address_space(struct protection_domain *domain, - gfp_t gfp) -{ - u64 *pte; - - if (domain->mode == PAGE_MODE_6_LEVEL) - /* address space already 64 bit large */ - return false; - - pte = (void *)get_zeroed_page(gfp); - if (!pte) - return false; - - *pte = PM_LEVEL_PDE(domain->mode, - virt_to_phys(domain->pt_root)); - domain->pt_root = pte; - domain->mode += 1; - domain->updated = true; - - return true; -} - -static u64 *alloc_pte(struct protection_domain *domain, - unsigned long address, - int end_lvl, - u64 **pte_page, - gfp_t gfp) -{ - u64 *pte, *page; - int level; - - while (address > PM_LEVEL_SIZE(domain->mode)) - increase_address_space(domain, gfp); - - level = domain->mode - 1; - pte = &domain->pt_root[PM_LEVEL_INDEX(level, address)]; - - while (level > end_lvl) { - if (!IOMMU_PTE_PRESENT(*pte)) { - page = (u64 *)get_zeroed_page(gfp); - if (!page) - return NULL; - *pte = PM_LEVEL_PDE(level, virt_to_phys(page)); - } - - level -= 1; - - pte = IOMMU_PTE_PAGE(*pte); - - if (pte_page && level == end_lvl) - *pte_page = pte; - - pte = &pte[PM_LEVEL_INDEX(level, address)]; - } - - return pte; -} - -/* * This function fetches the PTE for a given address in the aperture */ static u64* dma_ops_get_pte(struct dma_ops_domain *dom, @@ -1528,8 +1726,7 @@ static u64* dma_ops_get_pte(struct dma_ops_domain *dom, * This is the generic map function. It maps one 4kb page at paddr to * the given address in the DMA address space for the domain. */ -static dma_addr_t dma_ops_domain_map(struct amd_iommu *iommu, - struct dma_ops_domain *dom, +static dma_addr_t dma_ops_domain_map(struct dma_ops_domain *dom, unsigned long address, phys_addr_t paddr, int direction) @@ -1542,7 +1739,7 @@ static dma_addr_t dma_ops_domain_map(struct amd_iommu *iommu, pte = dma_ops_get_pte(dom, address); if (!pte) - return bad_dma_address; + return DMA_ERROR_CODE; __pte = paddr | IOMMU_PTE_P | IOMMU_PTE_FC; @@ -1563,8 +1760,7 @@ static dma_addr_t dma_ops_domain_map(struct amd_iommu *iommu, /* * The generic unmapping function for on page in the DMA address space. */ -static void dma_ops_domain_unmap(struct amd_iommu *iommu, - struct dma_ops_domain *dom, +static void dma_ops_domain_unmap(struct dma_ops_domain *dom, unsigned long address) { struct aperture_range *aperture; @@ -1595,7 +1791,6 @@ static void dma_ops_domain_unmap(struct amd_iommu *iommu, * Must be called with the domain lock held. */ static dma_addr_t __map_single(struct device *dev, - struct amd_iommu *iommu, struct dma_ops_domain *dma_dom, phys_addr_t paddr, size_t size, @@ -1623,7 +1818,7 @@ static dma_addr_t __map_single(struct device *dev, retry: address = dma_ops_alloc_addresses(dev, dma_dom, pages, align_mask, dma_mask); - if (unlikely(address == bad_dma_address)) { + if (unlikely(address == DMA_ERROR_CODE)) { /* * setting next_address here will let the address * allocator only scan the new allocated range in the @@ -1631,11 +1826,11 @@ retry: */ dma_dom->next_address = dma_dom->aperture_size; - if (alloc_new_range(iommu, dma_dom, false, GFP_ATOMIC)) + if (alloc_new_range(dma_dom, false, GFP_ATOMIC)) goto out; /* - * aperture was sucessfully enlarged by 128 MB, try + * aperture was successfully enlarged by 128 MB, try * allocation again */ goto retry; @@ -1643,8 +1838,8 @@ retry: start = address; for (i = 0; i < pages; ++i) { - ret = dma_ops_domain_map(iommu, dma_dom, start, paddr, dir); - if (ret == bad_dma_address) + ret = dma_ops_domain_map(dma_dom, start, paddr, dir); + if (ret == DMA_ERROR_CODE) goto out_unmap; paddr += PAGE_SIZE; @@ -1655,10 +1850,10 @@ retry: ADD_STATS_COUNTER(alloced_io_mem, size); if (unlikely(dma_dom->need_flush && !amd_iommu_unmap_flush)) { - iommu_flush_tlb(iommu, dma_dom->domain.id); + iommu_flush_tlb(&dma_dom->domain); dma_dom->need_flush = false; - } else if (unlikely(iommu_has_npcache(iommu))) - iommu_flush_pages(iommu, dma_dom->domain.id, address, size); + } else if (unlikely(amd_iommu_np_cache)) + iommu_flush_pages(&dma_dom->domain, address, size); out: return address; @@ -1667,20 +1862,19 @@ out_unmap: for (--i; i >= 0; --i) { start -= PAGE_SIZE; - dma_ops_domain_unmap(iommu, dma_dom, start); + dma_ops_domain_unmap(dma_dom, start); } dma_ops_free_addresses(dma_dom, address, pages); - return bad_dma_address; + return DMA_ERROR_CODE; } /* * Does the reverse of the __map_single function. Must be called with * the domain lock held too */ -static void __unmap_single(struct amd_iommu *iommu, - struct dma_ops_domain *dma_dom, +static void __unmap_single(struct dma_ops_domain *dma_dom, dma_addr_t dma_addr, size_t size, int dir) @@ -1688,7 +1882,7 @@ static void __unmap_single(struct amd_iommu *iommu, dma_addr_t i, start; unsigned int pages; - if ((dma_addr == bad_dma_address) || + if ((dma_addr == DMA_ERROR_CODE) || (dma_addr + size > dma_dom->aperture_size)) return; @@ -1697,7 +1891,7 @@ static void __unmap_single(struct amd_iommu *iommu, start = dma_addr; for (i = 0; i < pages; ++i) { - dma_ops_domain_unmap(iommu, dma_dom, start); + dma_ops_domain_unmap(dma_dom, start); start += PAGE_SIZE; } @@ -1706,7 +1900,7 @@ static void __unmap_single(struct amd_iommu *iommu, dma_ops_free_addresses(dma_dom, dma_addr, pages); if (amd_iommu_unmap_flush || dma_dom->need_flush) { - iommu_flush_pages(iommu, dma_dom->domain.id, dma_addr, size); + iommu_flush_pages(&dma_dom->domain, dma_addr, size); dma_dom->need_flush = false; } } @@ -1720,36 +1914,29 @@ static dma_addr_t map_page(struct device *dev, struct page *page, struct dma_attrs *attrs) { unsigned long flags; - struct amd_iommu *iommu; struct protection_domain *domain; - u16 devid; dma_addr_t addr; u64 dma_mask; phys_addr_t paddr = page_to_phys(page) + offset; INC_STATS_COUNTER(cnt_map_single); - if (!check_device(dev)) - return bad_dma_address; - - dma_mask = *dev->dma_mask; - - get_device_resources(dev, &iommu, &domain, &devid); - - if (iommu == NULL || domain == NULL) - /* device not handled by any AMD IOMMU */ + domain = get_domain(dev); + if (PTR_ERR(domain) == -EINVAL) return (dma_addr_t)paddr; + else if (IS_ERR(domain)) + return DMA_ERROR_CODE; - if (!dma_ops_domain(domain)) - return bad_dma_address; + dma_mask = *dev->dma_mask; spin_lock_irqsave(&domain->lock, flags); - addr = __map_single(dev, iommu, domain->priv, paddr, size, dir, false, + + addr = __map_single(dev, domain->priv, paddr, size, dir, false, dma_mask); - if (addr == bad_dma_address) + if (addr == DMA_ERROR_CODE) goto out; - iommu_completion_wait(iommu); + iommu_flush_complete(domain); out: spin_unlock_irqrestore(&domain->lock, flags); @@ -1764,25 +1951,19 @@ static void unmap_page(struct device *dev, dma_addr_t dma_addr, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) { unsigned long flags; - struct amd_iommu *iommu; struct protection_domain *domain; - u16 devid; INC_STATS_COUNTER(cnt_unmap_single); - if (!check_device(dev) || - !get_device_resources(dev, &iommu, &domain, &devid)) - /* device not handled by any AMD IOMMU */ - return; - - if (!dma_ops_domain(domain)) + domain = get_domain(dev); + if (IS_ERR(domain)) return; spin_lock_irqsave(&domain->lock, flags); - __unmap_single(iommu, domain->priv, dma_addr, size, dir); + __unmap_single(domain->priv, dma_addr, size, dir); - iommu_completion_wait(iommu); + iommu_flush_complete(domain); spin_unlock_irqrestore(&domain->lock, flags); } @@ -1814,9 +1995,7 @@ static int map_sg(struct device *dev, struct scatterlist *sglist, struct dma_attrs *attrs) { unsigned long flags; - struct amd_iommu *iommu; struct protection_domain *domain; - u16 devid; int i; struct scatterlist *s; phys_addr_t paddr; @@ -1825,25 +2004,20 @@ static int map_sg(struct device *dev, struct scatterlist *sglist, INC_STATS_COUNTER(cnt_map_sg); - if (!check_device(dev)) + domain = get_domain(dev); + if (PTR_ERR(domain) == -EINVAL) + return map_sg_no_iommu(dev, sglist, nelems, dir); + else if (IS_ERR(domain)) return 0; dma_mask = *dev->dma_mask; - get_device_resources(dev, &iommu, &domain, &devid); - - if (!iommu || !domain) - return map_sg_no_iommu(dev, sglist, nelems, dir); - - if (!dma_ops_domain(domain)) - return 0; - spin_lock_irqsave(&domain->lock, flags); for_each_sg(sglist, s, nelems, i) { paddr = sg_phys(s); - s->dma_address = __map_single(dev, iommu, domain->priv, + s->dma_address = __map_single(dev, domain->priv, paddr, s->length, dir, false, dma_mask); @@ -1854,7 +2028,7 @@ static int map_sg(struct device *dev, struct scatterlist *sglist, goto unmap; } - iommu_completion_wait(iommu); + iommu_flush_complete(domain); out: spin_unlock_irqrestore(&domain->lock, flags); @@ -1863,7 +2037,7 @@ out: unmap: for_each_sg(sglist, s, mapped_elems, i) { if (s->dma_address) - __unmap_single(iommu, domain->priv, s->dma_address, + __unmap_single(domain->priv, s->dma_address, s->dma_length, dir); s->dma_address = s->dma_length = 0; } @@ -1882,30 +2056,25 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist, struct dma_attrs *attrs) { unsigned long flags; - struct amd_iommu *iommu; struct protection_domain *domain; struct scatterlist *s; - u16 devid; int i; INC_STATS_COUNTER(cnt_unmap_sg); - if (!check_device(dev) || - !get_device_resources(dev, &iommu, &domain, &devid)) - return; - - if (!dma_ops_domain(domain)) + domain = get_domain(dev); + if (IS_ERR(domain)) return; spin_lock_irqsave(&domain->lock, flags); for_each_sg(sglist, s, nelems, i) { - __unmap_single(iommu, domain->priv, s->dma_address, + __unmap_single(domain->priv, s->dma_address, s->dma_length, dir); s->dma_address = s->dma_length = 0; } - iommu_completion_wait(iommu); + iommu_flush_complete(domain); spin_unlock_irqrestore(&domain->lock, flags); } @@ -1918,49 +2087,44 @@ static void *alloc_coherent(struct device *dev, size_t size, { unsigned long flags; void *virt_addr; - struct amd_iommu *iommu; struct protection_domain *domain; - u16 devid; phys_addr_t paddr; u64 dma_mask = dev->coherent_dma_mask; INC_STATS_COUNTER(cnt_alloc_coherent); - if (!check_device(dev)) + domain = get_domain(dev); + if (PTR_ERR(domain) == -EINVAL) { + virt_addr = (void *)__get_free_pages(flag, get_order(size)); + *dma_addr = __pa(virt_addr); + return virt_addr; + } else if (IS_ERR(domain)) return NULL; - if (!get_device_resources(dev, &iommu, &domain, &devid)) - flag &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); + dma_mask = dev->coherent_dma_mask; + flag &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); + flag |= __GFP_ZERO; - flag |= __GFP_ZERO; virt_addr = (void *)__get_free_pages(flag, get_order(size)); if (!virt_addr) return NULL; paddr = virt_to_phys(virt_addr); - if (!iommu || !domain) { - *dma_addr = (dma_addr_t)paddr; - return virt_addr; - } - - if (!dma_ops_domain(domain)) - goto out_free; - if (!dma_mask) dma_mask = *dev->dma_mask; spin_lock_irqsave(&domain->lock, flags); - *dma_addr = __map_single(dev, iommu, domain->priv, paddr, + *dma_addr = __map_single(dev, domain->priv, paddr, size, DMA_BIDIRECTIONAL, true, dma_mask); - if (*dma_addr == bad_dma_address) { + if (*dma_addr == DMA_ERROR_CODE) { spin_unlock_irqrestore(&domain->lock, flags); goto out_free; } - iommu_completion_wait(iommu); + iommu_flush_complete(domain); spin_unlock_irqrestore(&domain->lock, flags); @@ -1980,28 +2144,19 @@ static void free_coherent(struct device *dev, size_t size, void *virt_addr, dma_addr_t dma_addr) { unsigned long flags; - struct amd_iommu *iommu; struct protection_domain *domain; - u16 devid; INC_STATS_COUNTER(cnt_free_coherent); - if (!check_device(dev)) - return; - - get_device_resources(dev, &iommu, &domain, &devid); - - if (!iommu || !domain) - goto free_mem; - - if (!dma_ops_domain(domain)) + domain = get_domain(dev); + if (IS_ERR(domain)) goto free_mem; spin_lock_irqsave(&domain->lock, flags); - __unmap_single(iommu, domain->priv, dma_addr, size, DMA_BIDIRECTIONAL); + __unmap_single(domain->priv, dma_addr, size, DMA_BIDIRECTIONAL); - iommu_completion_wait(iommu); + iommu_flush_complete(domain); spin_unlock_irqrestore(&domain->lock, flags); @@ -2015,22 +2170,7 @@ free_mem: */ static int amd_iommu_dma_supported(struct device *dev, u64 mask) { - u16 bdf; - struct pci_dev *pcidev; - - /* No device or no PCI device */ - if (!dev || dev->bus != &pci_bus_type) - return 0; - - pcidev = to_pci_dev(dev); - - bdf = calc_devid(pcidev->bus->number, pcidev->devfn); - - /* Out of our scope? */ - if (bdf > amd_iommu_last_bdf) - return 0; - - return 1; + return check_device(dev); } /* @@ -2044,25 +2184,28 @@ static void prealloc_protection_domains(void) { struct pci_dev *dev = NULL; struct dma_ops_domain *dma_dom; - struct amd_iommu *iommu; u16 devid; while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { - devid = calc_devid(dev->bus->number, dev->devfn); - if (devid > amd_iommu_last_bdf) - continue; - devid = amd_iommu_alias_table[devid]; - if (domain_for_device(devid)) + + /* Do we handle this device? */ + if (!check_device(&dev->dev)) continue; - iommu = amd_iommu_rlookup_table[devid]; - if (!iommu) + + /* Is there already any domain for it? */ + if (domain_for_device(&dev->dev)) continue; - dma_dom = dma_ops_domain_alloc(iommu); + + devid = get_device_id(&dev->dev); + + dma_dom = dma_ops_domain_alloc(); if (!dma_dom) continue; init_unity_mappings_for_device(dma_dom, devid); dma_dom->target_dev = devid; + attach_device(&dev->dev, &dma_dom->domain); + list_add_tail(&dma_dom->list, &iommu_pd_list); } } @@ -2080,6 +2223,12 @@ static struct dma_map_ops amd_iommu_dma_ops = { /* * The function which clues the AMD IOMMU driver into dma_ops. */ + +void __init amd_iommu_init_api(void) +{ + register_iommu(&amd_iommu_ops); +} + int __init amd_iommu_init_dma_ops(void) { struct amd_iommu *iommu; @@ -2091,7 +2240,7 @@ int __init amd_iommu_init_dma_ops(void) * protection domain will be assigned to the default one. */ for_each_iommu(iommu) { - iommu->default_dom = dma_ops_domain_alloc(iommu); + iommu->default_dom = dma_ops_domain_alloc(); if (iommu->default_dom == NULL) return -ENOMEM; iommu->default_dom->domain.flags |= PD_DEFAULT_MASK; @@ -2101,15 +2250,12 @@ int __init amd_iommu_init_dma_ops(void) } /* - * If device isolation is enabled, pre-allocate the protection - * domains for each device. + * Pre-allocate the protection domains for each device. */ - if (amd_iommu_isolate) - prealloc_protection_domains(); + prealloc_protection_domains(); iommu_detected = 1; - force_iommu = 1; - bad_dma_address = 0; + swiotlb = 0; #ifdef CONFIG_GART_IOMMU gart_iommu_aperture_disabled = 1; gart_iommu_aperture = 0; @@ -2118,10 +2264,6 @@ int __init amd_iommu_init_dma_ops(void) /* Make the driver finally visible to the drivers */ dma_ops = &amd_iommu_dma_ops; - register_iommu(&amd_iommu_ops); - - bus_register_notifier(&pci_bus_type, &device_nb); - amd_iommu_stats_init(); return 0; @@ -2148,14 +2290,17 @@ free_domains: static void cleanup_domain(struct protection_domain *domain) { + struct iommu_dev_data *dev_data, *next; unsigned long flags; - u16 devid; write_lock_irqsave(&amd_iommu_devtable_lock, flags); - for (devid = 0; devid <= amd_iommu_last_bdf; ++devid) - if (amd_iommu_pd_table[devid] == domain) - __detach_device(domain, devid); + list_for_each_entry_safe(dev_data, next, &domain->dev_list, list) { + struct device *dev = dev_data->dev; + + do_detach(dev); + atomic_set(&dev_data->bind, 0); + } write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); } @@ -2165,6 +2310,8 @@ static void protection_domain_free(struct protection_domain *domain) if (!domain) return; + del_domain_from_list(domain); + if (domain->id) domain_id_free(domain->id); @@ -2183,6 +2330,9 @@ static struct protection_domain *protection_domain_alloc(void) domain->id = domain_id_alloc(); if (!domain->id) goto out_err; + INIT_LIST_HEAD(&domain->dev_list); + + add_domain_to_list(domain); return domain; @@ -2239,26 +2389,23 @@ static void amd_iommu_domain_destroy(struct iommu_domain *dom) static void amd_iommu_detach_device(struct iommu_domain *dom, struct device *dev) { - struct protection_domain *domain = dom->priv; + struct iommu_dev_data *dev_data = dev->archdata.iommu; struct amd_iommu *iommu; - struct pci_dev *pdev; u16 devid; - if (dev->bus != &pci_bus_type) + if (!check_device(dev)) return; - pdev = to_pci_dev(dev); + devid = get_device_id(dev); - devid = calc_devid(pdev->bus->number, pdev->devfn); - - if (devid > 0) - detach_device(domain, devid); + if (dev_data->domain != NULL) + detach_device(dev); iommu = amd_iommu_rlookup_table[devid]; if (!iommu) return; - iommu_queue_inv_dev_entry(iommu, devid); + iommu_flush_device(dev); iommu_completion_wait(iommu); } @@ -2266,35 +2413,30 @@ static int amd_iommu_attach_device(struct iommu_domain *dom, struct device *dev) { struct protection_domain *domain = dom->priv; - struct protection_domain *old_domain; + struct iommu_dev_data *dev_data; struct amd_iommu *iommu; - struct pci_dev *pdev; + int ret; u16 devid; - if (dev->bus != &pci_bus_type) + if (!check_device(dev)) return -EINVAL; - pdev = to_pci_dev(dev); - - devid = calc_devid(pdev->bus->number, pdev->devfn); + dev_data = dev->archdata.iommu; - if (devid >= amd_iommu_last_bdf || - devid != amd_iommu_alias_table[devid]) - return -EINVAL; + devid = get_device_id(dev); iommu = amd_iommu_rlookup_table[devid]; if (!iommu) return -EINVAL; - old_domain = domain_for_device(devid); - if (old_domain) - detach_device(old_domain, devid); + if (dev_data->domain) + detach_device(dev); - attach_device(iommu, domain, devid); + ret = attach_device(dev, domain); iommu_completion_wait(iommu); - return 0; + return ret; } static int amd_iommu_map_range(struct iommu_domain *dom, @@ -2340,7 +2482,7 @@ static void amd_iommu_unmap_range(struct iommu_domain *dom, iova += PAGE_SIZE; } - iommu_flush_domain(domain->id); + iommu_flush_tlb_pde(domain); } static phys_addr_t amd_iommu_iova_to_phys(struct iommu_domain *dom, @@ -2391,10 +2533,11 @@ static struct iommu_ops amd_iommu_ops = { int __init amd_iommu_init_passthrough(void) { + struct amd_iommu *iommu; struct pci_dev *dev = NULL; - u16 devid, devid2; + u16 devid; - /* allocate passthroug domain */ + /* allocate passthrough domain */ pt_domain = protection_domain_alloc(); if (!pt_domain) return -ENOMEM; @@ -2402,20 +2545,17 @@ int __init amd_iommu_init_passthrough(void) pt_domain->mode |= PAGE_MODE_NONE; while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { - struct amd_iommu *iommu; - devid = calc_devid(dev->bus->number, dev->devfn); - if (devid > amd_iommu_last_bdf) + if (!check_device(&dev->dev)) continue; - devid2 = amd_iommu_alias_table[devid]; + devid = get_device_id(&dev->dev); - iommu = amd_iommu_rlookup_table[devid2]; + iommu = amd_iommu_rlookup_table[devid]; if (!iommu) continue; - __attach_device(iommu, pt_domain, devid); - __attach_device(iommu, pt_domain, devid2); + attach_device(&dev->dev, pt_domain); } pr_info("AMD-Vi: Initialized for Passthrough Mode\n"); diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index b4b61d462dcc..9dc91b431470 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2007-2008 Advanced Micro Devices, Inc. + * Copyright (C) 2007-2009 Advanced Micro Devices, Inc. * Author: Joerg Roedel <joerg.roedel@amd.com> * Leo Duran <leo.duran@amd.com> * @@ -25,10 +25,12 @@ #include <linux/interrupt.h> #include <linux/msi.h> #include <asm/pci-direct.h> +#include <asm/amd_iommu_proto.h> #include <asm/amd_iommu_types.h> #include <asm/amd_iommu.h> #include <asm/iommu.h> #include <asm/gart.h> +#include <asm/x86_init.h> /* * definitions for the ACPI scanning code @@ -123,18 +125,29 @@ u16 amd_iommu_last_bdf; /* largest PCI device id we have to handle */ LIST_HEAD(amd_iommu_unity_map); /* a list of required unity mappings we find in ACPI */ -#ifdef CONFIG_IOMMU_STRESS -bool amd_iommu_isolate = false; -#else -bool amd_iommu_isolate = true; /* if true, device isolation is - enabled */ -#endif - bool amd_iommu_unmap_flush; /* if true, flush on every unmap */ LIST_HEAD(amd_iommu_list); /* list of all AMD IOMMUs in the system */ +/* Array to assign indices to IOMMUs*/ +struct amd_iommu *amd_iommus[MAX_IOMMUS]; +int amd_iommus_present; + +/* IOMMUs have a non-present cache? */ +bool amd_iommu_np_cache __read_mostly; + +/* + * Set to true if ACPI table parsing and hardware intialization went properly + */ +static bool amd_iommu_initialized; + +/* + * List of protection domains - used during resume + */ +LIST_HEAD(amd_iommu_pd_list); +spinlock_t amd_iommu_pd_lock; + /* * Pointer to the device table which is shared by all AMD IOMMUs * it is indexed by the PCI device id or the HT unit id and contains @@ -157,12 +170,6 @@ u16 *amd_iommu_alias_table; struct amd_iommu **amd_iommu_rlookup_table; /* - * The pd table (protection domain table) is used to find the protection domain - * data structure a device belongs to. Indexed with the PCI device id too. - */ -struct protection_domain **amd_iommu_pd_table; - -/* * AMD IOMMU allows up to 2^16 differend protection domains. This is a bitmap * to know which ones are already in use. */ @@ -240,7 +247,7 @@ static void iommu_feature_enable(struct amd_iommu *iommu, u8 bit) writel(ctrl, iommu->mmio_base + MMIO_CONTROL_OFFSET); } -static void __init iommu_feature_disable(struct amd_iommu *iommu, u8 bit) +static void iommu_feature_disable(struct amd_iommu *iommu, u8 bit) { u32 ctrl; @@ -519,6 +526,26 @@ static void set_dev_entry_bit(u16 devid, u8 bit) amd_iommu_dev_table[devid].data[i] |= (1 << _bit); } +static int get_dev_entry_bit(u16 devid, u8 bit) +{ + int i = (bit >> 5) & 0x07; + int _bit = bit & 0x1f; + + return (amd_iommu_dev_table[devid].data[i] & (1 << _bit)) >> _bit; +} + + +void amd_iommu_apply_erratum_63(u16 devid) +{ + int sysmgt; + + sysmgt = get_dev_entry_bit(devid, DEV_ENTRY_SYSMGT1) | + (get_dev_entry_bit(devid, DEV_ENTRY_SYSMGT2) << 1); + + if (sysmgt == 0x01) + set_dev_entry_bit(devid, DEV_ENTRY_IW); +} + /* Writes the specific IOMMU for a device into the rlookup table */ static void __init set_iommu_for_device(struct amd_iommu *iommu, u16 devid) { @@ -547,6 +574,8 @@ static void __init set_dev_entry_from_acpi(struct amd_iommu *iommu, if (flags & ACPI_DEVFLAG_LINT1) set_dev_entry_bit(devid, DEV_ENTRY_LINT1_PASS); + amd_iommu_apply_erratum_63(devid); + set_iommu_for_device(iommu, devid); } @@ -816,7 +845,18 @@ static void __init free_iommu_all(void) static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h) { spin_lock_init(&iommu->lock); + + /* Add IOMMU to internal data structures */ list_add_tail(&iommu->list, &amd_iommu_list); + iommu->index = amd_iommus_present++; + + if (unlikely(iommu->index >= MAX_IOMMUS)) { + WARN(1, "AMD-Vi: System has more IOMMUs than supported by this driver\n"); + return -ENOSYS; + } + + /* Index is fine - add IOMMU to the array */ + amd_iommus[iommu->index] = iommu; /* * Copy data from ACPI table entry to the iommu struct @@ -846,6 +886,9 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h) init_iommu_from_acpi(iommu, h); init_iommu_devices(iommu); + if (iommu->cap & (1UL << IOMMU_CAP_NPCACHE)) + amd_iommu_np_cache = true; + return pci_enable_device(iommu->dev); } @@ -891,6 +934,8 @@ static int __init init_iommu_all(struct acpi_table_header *table) } WARN_ON(p != end); + amd_iommu_initialized = true; + return 0; } @@ -903,7 +948,7 @@ static int __init init_iommu_all(struct acpi_table_header *table) * ****************************************************************************/ -static int __init iommu_setup_msi(struct amd_iommu *iommu) +static int iommu_setup_msi(struct amd_iommu *iommu) { int r; @@ -1154,19 +1199,10 @@ static struct sys_device device_amd_iommu = { * functions. Finally it prints some information about AMD IOMMUs and * the driver state and enables the hardware. */ -int __init amd_iommu_init(void) +static int __init amd_iommu_init(void) { int i, ret = 0; - - if (no_iommu) { - printk(KERN_INFO "AMD-Vi disabled by kernel command line\n"); - return 0; - } - - if (!amd_iommu_detected) - return -ENODEV; - /* * First parse ACPI tables to find the largest Bus/Dev/Func * we need to handle. Upon this information the shared data @@ -1203,15 +1239,6 @@ int __init amd_iommu_init(void) if (amd_iommu_rlookup_table == NULL) goto free; - /* - * Protection Domain table - maps devices to protection domains - * This table has the same size as the rlookup_table - */ - amd_iommu_pd_table = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, - get_order(rlookup_table_size)); - if (amd_iommu_pd_table == NULL) - goto free; - amd_iommu_pd_alloc_bitmap = (void *)__get_free_pages( GFP_KERNEL | __GFP_ZERO, get_order(MAX_DOMAIN_ID/8)); @@ -1233,6 +1260,8 @@ int __init amd_iommu_init(void) */ amd_iommu_pd_alloc_bitmap[0] = 1; + spin_lock_init(&amd_iommu_pd_lock); + /* * now the data structures are allocated and basically initialized * start the real acpi table scan @@ -1241,6 +1270,9 @@ int __init amd_iommu_init(void) if (acpi_table_parse("IVRS", init_iommu_all) != 0) goto free; + if (!amd_iommu_initialized) + goto free; + if (acpi_table_parse("IVRS", init_memory_definitions) != 0) goto free; @@ -1252,39 +1284,43 @@ int __init amd_iommu_init(void) if (ret) goto free; + ret = amd_iommu_init_devices(); + if (ret) + goto free; + if (iommu_pass_through) ret = amd_iommu_init_passthrough(); else ret = amd_iommu_init_dma_ops(); + if (ret) goto free; + amd_iommu_init_api(); + + amd_iommu_init_notifier(); + enable_iommus(); if (iommu_pass_through) goto out; - printk(KERN_INFO "AMD-Vi: device isolation "); - if (amd_iommu_isolate) - printk("enabled\n"); - else - printk("disabled\n"); - if (amd_iommu_unmap_flush) printk(KERN_INFO "AMD-Vi: IO/TLB flush on unmap enabled\n"); else printk(KERN_INFO "AMD-Vi: Lazy IO/TLB flushing enabled\n"); + x86_platform.iommu_shutdown = disable_iommus; out: return ret; free: + + amd_iommu_uninit_devices(); + free_pages((unsigned long)amd_iommu_pd_alloc_bitmap, get_order(MAX_DOMAIN_ID/8)); - free_pages((unsigned long)amd_iommu_pd_table, - get_order(rlookup_table_size)); - free_pages((unsigned long)amd_iommu_rlookup_table, get_order(rlookup_table_size)); @@ -1301,11 +1337,6 @@ free: goto out; } -void amd_iommu_shutdown(void) -{ - disable_iommus(); -} - /**************************************************************************** * * Early detect code. This code runs at IOMMU detection time in the DMA @@ -1320,16 +1351,16 @@ static int __init early_amd_iommu_detect(struct acpi_table_header *table) void __init amd_iommu_detect(void) { - if (swiotlb || no_iommu || (iommu_detected && !gart_iommu_aperture)) + if (no_iommu || (iommu_detected && !gart_iommu_aperture)) return; if (acpi_table_parse("IVRS", early_amd_iommu_detect) == 0) { iommu_detected = 1; amd_iommu_detected = 1; -#ifdef CONFIG_GART_IOMMU - gart_iommu_aperture_disabled = 1; - gart_iommu_aperture = 0; -#endif + x86_init.iommu.iommu_init = amd_iommu_init; + + /* Make sure ACS will be enabled */ + pci_request_acs(); } } @@ -1350,10 +1381,6 @@ static int __init parse_amd_iommu_dump(char *str) static int __init parse_amd_iommu_options(char *str) { for (; *str; ++str) { - if (strncmp(str, "isolate", 7) == 0) - amd_iommu_isolate = true; - if (strncmp(str, "share", 5) == 0) - amd_iommu_isolate = false; if (strncmp(str, "fullflush", 9) == 0) amd_iommu_unmap_flush = true; } diff --git a/arch/x86/kernel/apb_timer.c b/arch/x86/kernel/apb_timer.c new file mode 100644 index 000000000000..4b7099526d2c --- /dev/null +++ b/arch/x86/kernel/apb_timer.c @@ -0,0 +1,784 @@ +/* + * apb_timer.c: Driver for Langwell APB timers + * + * (C) Copyright 2009 Intel Corporation + * Author: Jacob Pan (jacob.jun.pan@intel.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + * + * Note: + * Langwell is the south complex of Intel Moorestown MID platform. There are + * eight external timers in total that can be used by the operating system. + * The timer information, such as frequency and addresses, is provided to the + * OS via SFI tables. + * Timer interrupts are routed via FW/HW emulated IOAPIC independently via + * individual redirection table entries (RTE). + * Unlike HPET, there is no master counter, therefore one of the timers are + * used as clocksource. The overall allocation looks like: + * - timer 0 - NR_CPUs for per cpu timer + * - one timer for clocksource + * - one timer for watchdog driver. + * It is also worth notice that APB timer does not support true one-shot mode, + * free-running mode will be used here to emulate one-shot mode. + * APB timer can also be used as broadcast timer along with per cpu local APIC + * timer, but by default APB timer has higher rating than local APIC timers. + */ + +#include <linux/clocksource.h> +#include <linux/clockchips.h> +#include <linux/delay.h> +#include <linux/errno.h> +#include <linux/init.h> +#include <linux/sysdev.h> +#include <linux/pm.h> +#include <linux/pci.h> +#include <linux/sfi.h> +#include <linux/interrupt.h> +#include <linux/cpu.h> +#include <linux/irq.h> + +#include <asm/fixmap.h> +#include <asm/apb_timer.h> + +#define APBT_MASK CLOCKSOURCE_MASK(32) +#define APBT_SHIFT 22 +#define APBT_CLOCKEVENT_RATING 150 +#define APBT_CLOCKSOURCE_RATING 250 +#define APBT_MIN_DELTA_USEC 200 + +#define EVT_TO_APBT_DEV(evt) container_of(evt, struct apbt_dev, evt) +#define APBT_CLOCKEVENT0_NUM (0) +#define APBT_CLOCKEVENT1_NUM (1) +#define APBT_CLOCKSOURCE_NUM (2) + +static unsigned long apbt_address; +static int apb_timer_block_enabled; +static void __iomem *apbt_virt_address; +static int phy_cs_timer_id; + +/* + * Common DW APB timer info + */ +static uint64_t apbt_freq; + +static void apbt_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt); +static int apbt_next_event(unsigned long delta, + struct clock_event_device *evt); +static cycle_t apbt_read_clocksource(struct clocksource *cs); +static void apbt_restart_clocksource(struct clocksource *cs); + +struct apbt_dev { + struct clock_event_device evt; + unsigned int num; + int cpu; + unsigned int irq; + unsigned int tick; + unsigned int count; + unsigned int flags; + char name[10]; +}; + +int disable_apbt_percpu __cpuinitdata; + +static DEFINE_PER_CPU(struct apbt_dev, cpu_apbt_dev); + +#ifdef CONFIG_SMP +static unsigned int apbt_num_timers_used; +static struct apbt_dev *apbt_devs; +#endif + +static inline unsigned long apbt_readl_reg(unsigned long a) +{ + return readl(apbt_virt_address + a); +} + +static inline void apbt_writel_reg(unsigned long d, unsigned long a) +{ + writel(d, apbt_virt_address + a); +} + +static inline unsigned long apbt_readl(int n, unsigned long a) +{ + return readl(apbt_virt_address + a + n * APBTMRS_REG_SIZE); +} + +static inline void apbt_writel(int n, unsigned long d, unsigned long a) +{ + writel(d, apbt_virt_address + a + n * APBTMRS_REG_SIZE); +} + +static inline void apbt_set_mapping(void) +{ + struct sfi_timer_table_entry *mtmr; + + if (apbt_virt_address) { + pr_debug("APBT base already mapped\n"); + return; + } + mtmr = sfi_get_mtmr(APBT_CLOCKEVENT0_NUM); + if (mtmr == NULL) { + printk(KERN_ERR "Failed to get MTMR %d from SFI\n", + APBT_CLOCKEVENT0_NUM); + return; + } + apbt_address = (unsigned long)mtmr->phys_addr; + if (!apbt_address) { + printk(KERN_WARNING "No timer base from SFI, use default\n"); + apbt_address = APBT_DEFAULT_BASE; + } + apbt_virt_address = ioremap_nocache(apbt_address, APBT_MMAP_SIZE); + if (apbt_virt_address) { + pr_debug("Mapped APBT physical addr %p at virtual addr %p\n",\ + (void *)apbt_address, (void *)apbt_virt_address); + } else { + pr_debug("Failed mapping APBT phy address at %p\n",\ + (void *)apbt_address); + goto panic_noapbt; + } + apbt_freq = mtmr->freq_hz / USEC_PER_SEC; + sfi_free_mtmr(mtmr); + + /* Now figure out the physical timer id for clocksource device */ + mtmr = sfi_get_mtmr(APBT_CLOCKSOURCE_NUM); + if (mtmr == NULL) + goto panic_noapbt; + + /* Now figure out the physical timer id */ + phy_cs_timer_id = (unsigned int)(mtmr->phys_addr & 0xff) + / APBTMRS_REG_SIZE; + pr_debug("Use timer %d for clocksource\n", phy_cs_timer_id); + return; + +panic_noapbt: + panic("Failed to setup APB system timer\n"); + +} + +static inline void apbt_clear_mapping(void) +{ + iounmap(apbt_virt_address); + apbt_virt_address = NULL; +} + +/* + * APBT timer interrupt enable / disable + */ +static inline int is_apbt_capable(void) +{ + return apbt_virt_address ? 1 : 0; +} + +static struct clocksource clocksource_apbt = { + .name = "apbt", + .rating = APBT_CLOCKSOURCE_RATING, + .read = apbt_read_clocksource, + .mask = APBT_MASK, + .shift = APBT_SHIFT, + .flags = CLOCK_SOURCE_IS_CONTINUOUS, + .resume = apbt_restart_clocksource, +}; + +/* boot APB clock event device */ +static struct clock_event_device apbt_clockevent = { + .name = "apbt0", + .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT, + .set_mode = apbt_set_mode, + .set_next_event = apbt_next_event, + .shift = APBT_SHIFT, + .irq = 0, + .rating = APBT_CLOCKEVENT_RATING, +}; + +/* + * if user does not want to use per CPU apb timer, just give it a lower rating + * than local apic timer and skip the late per cpu timer init. + */ +static inline int __init setup_x86_mrst_timer(char *arg) +{ + if (!arg) + return -EINVAL; + + if (strcmp("apbt_only", arg) == 0) + disable_apbt_percpu = 0; + else if (strcmp("lapic_and_apbt", arg) == 0) + disable_apbt_percpu = 1; + else { + pr_warning("X86 MRST timer option %s not recognised" + " use x86_mrst_timer=apbt_only or lapic_and_apbt\n", + arg); + return -EINVAL; + } + return 0; +} +__setup("x86_mrst_timer=", setup_x86_mrst_timer); + +/* + * start count down from 0xffff_ffff. this is done by toggling the enable bit + * then load initial load count to ~0. + */ +static void apbt_start_counter(int n) +{ + unsigned long ctrl = apbt_readl(n, APBTMR_N_CONTROL); + + ctrl &= ~APBTMR_CONTROL_ENABLE; + apbt_writel(n, ctrl, APBTMR_N_CONTROL); + apbt_writel(n, ~0, APBTMR_N_LOAD_COUNT); + /* enable, mask interrupt */ + ctrl &= ~APBTMR_CONTROL_MODE_PERIODIC; + ctrl |= (APBTMR_CONTROL_ENABLE | APBTMR_CONTROL_INT); + apbt_writel(n, ctrl, APBTMR_N_CONTROL); + /* read it once to get cached counter value initialized */ + apbt_read_clocksource(&clocksource_apbt); +} + +static irqreturn_t apbt_interrupt_handler(int irq, void *data) +{ + struct apbt_dev *dev = (struct apbt_dev *)data; + struct clock_event_device *aevt = &dev->evt; + + if (!aevt->event_handler) { + printk(KERN_INFO "Spurious APBT timer interrupt on %d\n", + dev->num); + return IRQ_NONE; + } + aevt->event_handler(aevt); + return IRQ_HANDLED; +} + +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); + /* clear pending intr */ + apbt_readl(n, APBTMR_N_EOI); + ctrl &= ~APBTMR_CONTROL_INT; + apbt_writel(n, ctrl, APBTMR_N_CONTROL); +} + +static void apbt_disable_int(int n) +{ + unsigned long ctrl = apbt_readl(n, APBTMR_N_CONTROL); + + ctrl |= APBTMR_CONTROL_INT; + apbt_writel(n, ctrl, APBTMR_N_CONTROL); +} + + +static int __init apbt_clockevent_register(void) +{ + struct sfi_timer_table_entry *mtmr; + struct apbt_dev *adev = &__get_cpu_var(cpu_apbt_dev); + + mtmr = sfi_get_mtmr(APBT_CLOCKEVENT0_NUM); + if (mtmr == NULL) { + printk(KERN_ERR "Failed to get MTMR %d from SFI\n", + APBT_CLOCKEVENT0_NUM); + return -ENODEV; + } + + /* + * We need to calculate the scaled math multiplication factor for + * nanosecond to apbt tick conversion. + * mult = (nsec/cycle)*2^APBT_SHIFT + */ + apbt_clockevent.mult = div_sc((unsigned long) mtmr->freq_hz + , NSEC_PER_SEC, APBT_SHIFT); + + /* Calculate the min / max delta */ + apbt_clockevent.max_delta_ns = clockevent_delta2ns(0x7FFFFFFF, + &apbt_clockevent); + apbt_clockevent.min_delta_ns = clockevent_delta2ns( + APBT_MIN_DELTA_USEC*apbt_freq, + &apbt_clockevent); + /* + * Start apbt with the boot cpu mask and make it + * global if not used for per cpu timer. + */ + apbt_clockevent.cpumask = cpumask_of(smp_processor_id()); + adev->num = smp_processor_id(); + memcpy(&adev->evt, &apbt_clockevent, sizeof(struct clock_event_device)); + + if (disable_apbt_percpu) { + apbt_clockevent.rating = APBT_CLOCKEVENT_RATING - 100; + global_clock_event = &adev->evt; + printk(KERN_DEBUG "%s clockevent registered as global\n", + global_clock_event->name); + } + + if (request_irq(apbt_clockevent.irq, apbt_interrupt_handler, + IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING, + apbt_clockevent.name, adev)) { + printk(KERN_ERR "Failed request IRQ for APBT%d\n", + apbt_clockevent.irq); + } + + clockevents_register_device(&adev->evt); + /* Start APBT 0 interrupts */ + apbt_enable_int(APBT_CLOCKEVENT0_NUM); + + sfi_free_mtmr(mtmr); + return 0; +} + +#ifdef CONFIG_SMP +/* Should be called with per cpu */ +void apbt_setup_secondary_clock(void) +{ + struct apbt_dev *adev; + struct clock_event_device *aevt; + int cpu; + + /* Don't register boot CPU clockevent */ + cpu = smp_processor_id(); + if (cpu == boot_cpu_id) + return; + /* + * We need to calculate the scaled math multiplication factor for + * nanosecond to apbt tick conversion. + * mult = (nsec/cycle)*2^APBT_SHIFT + */ + printk(KERN_INFO "Init per CPU clockevent %d\n", cpu); + adev = &per_cpu(cpu_apbt_dev, cpu); + aevt = &adev->evt; + + memcpy(aevt, &apbt_clockevent, sizeof(*aevt)); + aevt->cpumask = cpumask_of(cpu); + aevt->name = adev->name; + aevt->mode = CLOCK_EVT_MODE_UNUSED; + + printk(KERN_INFO "Registering CPU %d clockevent device %s, mask %08x\n", + cpu, aevt->name, *(u32 *)aevt->cpumask); + + apbt_setup_irq(adev); + + clockevents_register_device(aevt); + + apbt_enable_int(cpu); + + return; +} + +/* + * this notify handler process CPU hotplug events. in case of S0i3, nonboot + * cpus are disabled/enabled frequently, for performance reasons, we keep the + * per cpu timer irq registered so that we do need to do free_irq/request_irq. + * + * TODO: it might be more reliable to directly disable percpu clockevent device + * without the notifier chain. currently, cpu 0 may get interrupts from other + * cpu timers during the offline process due to the ordering of notification. + * the extra interrupt is harmless. + */ +static int apbt_cpuhp_notify(struct notifier_block *n, + unsigned long action, void *hcpu) +{ + unsigned long cpu = (unsigned long)hcpu; + struct apbt_dev *adev = &per_cpu(cpu_apbt_dev, cpu); + + switch (action & 0xf) { + case CPU_DEAD: + apbt_disable_int(cpu); + if (system_state == SYSTEM_RUNNING) + pr_debug("skipping APBT CPU %lu offline\n", cpu); + 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); + } + return NOTIFY_OK; +} + +static __init int apbt_late_init(void) +{ + if (disable_apbt_percpu) + return 0; + /* This notifier should be called after workqueue is ready */ + hotcpu_notifier(apbt_cpuhp_notify, -20); + return 0; +} +fs_initcall(apbt_late_init); +#else + +void apbt_setup_secondary_clock(void) {} + +#endif /* CONFIG_SMP */ + +static void apbt_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + unsigned long ctrl; + uint64_t delta; + int timer_num; + struct apbt_dev *adev = EVT_TO_APBT_DEV(evt); + + timer_num = adev->num; + pr_debug("%s CPU %d timer %d mode=%d\n", + __func__, first_cpu(*evt->cpumask), timer_num, mode); + + switch (mode) { + case CLOCK_EVT_MODE_PERIODIC: + delta = ((uint64_t)(NSEC_PER_SEC/HZ)) * apbt_clockevent.mult; + delta >>= apbt_clockevent.shift; + ctrl = apbt_readl(timer_num, APBTMR_N_CONTROL); + ctrl |= APBTMR_CONTROL_MODE_PERIODIC; + apbt_writel(timer_num, ctrl, APBTMR_N_CONTROL); + /* + * DW APB p. 46, have to disable timer before load counter, + * may cause sync problem. + */ + ctrl &= ~APBTMR_CONTROL_ENABLE; + apbt_writel(timer_num, ctrl, APBTMR_N_CONTROL); + udelay(1); + pr_debug("Setting clock period %d for HZ %d\n", (int)delta, HZ); + apbt_writel(timer_num, delta, APBTMR_N_LOAD_COUNT); + ctrl |= APBTMR_CONTROL_ENABLE; + apbt_writel(timer_num, ctrl, APBTMR_N_CONTROL); + break; + /* APB timer does not have one-shot mode, use free running mode */ + case CLOCK_EVT_MODE_ONESHOT: + ctrl = apbt_readl(timer_num, APBTMR_N_CONTROL); + /* + * set free running mode, this mode will let timer reload max + * timeout which will give time (3min on 25MHz clock) to rearm + * the next event, therefore emulate the one-shot mode. + */ + ctrl &= ~APBTMR_CONTROL_ENABLE; + ctrl &= ~APBTMR_CONTROL_MODE_PERIODIC; + + apbt_writel(timer_num, ctrl, APBTMR_N_CONTROL); + /* write again to set free running mode */ + apbt_writel(timer_num, ctrl, APBTMR_N_CONTROL); + + /* + * DW APB p. 46, load counter with all 1s before starting free + * running mode. + */ + apbt_writel(timer_num, ~0, APBTMR_N_LOAD_COUNT); + ctrl &= ~APBTMR_CONTROL_INT; + ctrl |= APBTMR_CONTROL_ENABLE; + apbt_writel(timer_num, ctrl, APBTMR_N_CONTROL); + break; + + case CLOCK_EVT_MODE_UNUSED: + case CLOCK_EVT_MODE_SHUTDOWN: + apbt_disable_int(timer_num); + ctrl = apbt_readl(timer_num, APBTMR_N_CONTROL); + ctrl &= ~APBTMR_CONTROL_ENABLE; + apbt_writel(timer_num, ctrl, APBTMR_N_CONTROL); + break; + + case CLOCK_EVT_MODE_RESUME: + apbt_enable_int(timer_num); + break; + } +} + +static int apbt_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + unsigned long ctrl; + int timer_num; + + struct apbt_dev *adev = EVT_TO_APBT_DEV(evt); + + timer_num = adev->num; + /* Disable timer */ + ctrl = apbt_readl(timer_num, APBTMR_N_CONTROL); + ctrl &= ~APBTMR_CONTROL_ENABLE; + apbt_writel(timer_num, ctrl, APBTMR_N_CONTROL); + /* write new count */ + apbt_writel(timer_num, delta, APBTMR_N_LOAD_COUNT); + ctrl |= APBTMR_CONTROL_ENABLE; + apbt_writel(timer_num, ctrl, APBTMR_N_CONTROL); + return 0; +} + +/* + * APB timer clock is not in sync with pclk on Langwell, which translates to + * unreliable read value caused by sampling error. the error does not add up + * overtime and only happens when sampling a 0 as a 1 by mistake. so the time + * would go backwards. the following code is trying to prevent time traveling + * backwards. little bit paranoid. + */ +static cycle_t apbt_read_clocksource(struct clocksource *cs) +{ + unsigned long t0, t1, t2; + static unsigned long last_read; + +bad_count: + t1 = apbt_readl(phy_cs_timer_id, + APBTMR_N_CURRENT_VALUE); + t2 = apbt_readl(phy_cs_timer_id, + APBTMR_N_CURRENT_VALUE); + if (unlikely(t1 < t2)) { + pr_debug("APBT: read current count error %lx:%lx:%lx\n", + t1, t2, t2 - t1); + goto bad_count; + } + /* + * check against cached last read, makes sure time does not go back. + * it could be a normal rollover but we will do tripple check anyway + */ + if (unlikely(t2 > last_read)) { + /* check if we have a normal rollover */ + unsigned long raw_intr_status = + apbt_readl_reg(APBTMRS_RAW_INT_STATUS); + /* + * cs timer interrupt is masked but raw intr bit is set if + * rollover occurs. then we read EOI reg to clear it. + */ + if (raw_intr_status & (1 << phy_cs_timer_id)) { + apbt_readl(phy_cs_timer_id, APBTMR_N_EOI); + goto out; + } + 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"); + t0 = apbt_readl(phy_cs_timer_id, + APBTMR_N_CURRENT_VALUE); + udelay(1); + t1 = apbt_readl(phy_cs_timer_id, + APBTMR_N_CURRENT_VALUE); + udelay(1); + t2 = apbt_readl(phy_cs_timer_id, + APBTMR_N_CURRENT_VALUE); + if ((t2 > t1) || (t1 > t0)) { + printk(KERN_ERR "Error: APB CS tripple check failed\n"); + goto bad_count_x3; + } + } +out: + last_read = t2; + return (cycle_t)~t2; +} + +static int apbt_clocksource_register(void) +{ + u64 start, now; + cycle_t t1; + + /* Start the counter, use timer 2 as source, timer 0/1 for event */ + apbt_start_counter(phy_cs_timer_id); + + /* Verify whether apbt counter works */ + t1 = apbt_read_clocksource(&clocksource_apbt); + rdtscll(start); + + /* + * We don't know the TSC frequency yet, but waiting for + * 200000 TSC cycles is safe: + * 4 GHz == 50us + * 1 GHz == 200us + */ + do { + rep_nop(); + rdtscll(now); + } while ((now - start) < 200000UL); + + /* APBT is the only always on clocksource, it has to work! */ + if (t1 == apbt_read_clocksource(&clocksource_apbt)) + panic("APBT counter not counting. APBT disabled\n"); + + /* + * initialize and register APBT clocksource + * convert that to ns/clock cycle + * mult = (ns/c) * 2^APBT_SHIFT + */ + clocksource_apbt.mult = div_sc(MSEC_PER_SEC, + (unsigned long) apbt_freq, APBT_SHIFT); + clocksource_register(&clocksource_apbt); + + return 0; +} + +/* + * Early setup the APBT timer, only use timer 0 for booting then switch to + * per CPU timer if possible. + * returns 1 if per cpu apbt is setup + * returns 0 if no per cpu apbt is chosen + * panic if set up failed, this is the only platform timer on Moorestown. + */ +void __init apbt_time_init(void) +{ +#ifdef CONFIG_SMP + int i; + struct sfi_timer_table_entry *p_mtmr; + unsigned int percpu_timer; + struct apbt_dev *adev; +#endif + + if (apb_timer_block_enabled) + return; + apbt_set_mapping(); + if (apbt_virt_address) { + pr_debug("Found APBT version 0x%lx\n",\ + apbt_readl_reg(APBTMRS_COMP_VERSION)); + } else + goto out_noapbt; + /* + * Read the frequency and check for a sane value, for ESL model + * we extend the possible clock range to allow time scaling. + */ + + if (apbt_freq < APBT_MIN_FREQ || apbt_freq > APBT_MAX_FREQ) { + pr_debug("APBT has invalid freq 0x%llx\n", apbt_freq); + goto out_noapbt; + } + if (apbt_clocksource_register()) { + pr_debug("APBT has failed to register clocksource\n"); + goto out_noapbt; + } + if (!apbt_clockevent_register()) + apb_timer_block_enabled = 1; + else { + pr_debug("APBT has failed to register clockevent\n"); + goto out_noapbt; + } +#ifdef CONFIG_SMP + /* kernel cmdline disable apb timer, so we will use lapic timers */ + if (disable_apbt_percpu) { + printk(KERN_INFO "apbt: disabled per cpu timer\n"); + return; + } + pr_debug("%s: %d CPUs online\n", __func__, num_online_cpus()); + if (num_possible_cpus() <= sfi_mtimer_num) { + percpu_timer = 1; + apbt_num_timers_used = num_possible_cpus(); + } else { + percpu_timer = 0; + apbt_num_timers_used = 1; + adev = &per_cpu(cpu_apbt_dev, 0); + adev->flags &= ~APBT_DEV_USED; + } + pr_debug("%s: %d APB timers used\n", __func__, apbt_num_timers_used); + + /* here we set up per CPU timer data structure */ + apbt_devs = kzalloc(sizeof(struct apbt_dev) * apbt_num_timers_used, + GFP_KERNEL); + if (!apbt_devs) { + printk(KERN_ERR "Failed to allocate APB timer devices\n"); + return; + } + for (i = 0; i < apbt_num_timers_used; i++) { + adev = &per_cpu(cpu_apbt_dev, i); + adev->num = i; + adev->cpu = i; + p_mtmr = sfi_get_mtmr(i); + if (p_mtmr) { + adev->tick = p_mtmr->freq_hz; + adev->irq = p_mtmr->irq; + } else + printk(KERN_ERR "Failed to get timer for cpu %d\n", i); + adev->count = 0; + sprintf(adev->name, "apbt%d", i); + } +#endif + + return; + +out_noapbt: + apbt_clear_mapping(); + apb_timer_block_enabled = 0; + panic("failed to enable APB timer\n"); +} + +static inline void apbt_disable(int n) +{ + if (is_apbt_capable()) { + unsigned long ctrl = apbt_readl(n, APBTMR_N_CONTROL); + ctrl &= ~APBTMR_CONTROL_ENABLE; + apbt_writel(n, ctrl, APBTMR_N_CONTROL); + } +} + +/* called before apb_timer_enable, use early map */ +unsigned long apbt_quick_calibrate() +{ + int i, scale; + u64 old, new; + cycle_t t1, t2; + unsigned long khz = 0; + u32 loop, shift; + + apbt_set_mapping(); + apbt_start_counter(phy_cs_timer_id); + + /* check if the timer can count down, otherwise return */ + old = apbt_read_clocksource(&clocksource_apbt); + i = 10000; + while (--i) { + if (old != apbt_read_clocksource(&clocksource_apbt)) + break; + } + if (!i) + goto failed; + + /* count 16 ms */ + loop = (apbt_freq * 1000) << 4; + + /* restart the timer to ensure it won't get to 0 in the calibration */ + apbt_start_counter(phy_cs_timer_id); + + old = apbt_read_clocksource(&clocksource_apbt); + old += loop; + + t1 = __native_read_tsc(); + + do { + new = apbt_read_clocksource(&clocksource_apbt); + } while (new < old); + + t2 = __native_read_tsc(); + + shift = 5; + if (unlikely(loop >> shift == 0)) { + printk(KERN_INFO + "APBT TSC calibration failed, not enough resolution\n"); + return 0; + } + scale = (int)div_u64((t2 - t1), loop >> shift); + khz = (scale * apbt_freq * 1000) >> shift; + printk(KERN_INFO "TSC freq calculated by APB timer is %lu khz\n", khz); + return khz; +failed: + return 0; +} diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 128111d8ffe0..3704997e8b25 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -28,6 +28,7 @@ #include <asm/pci-direct.h> #include <asm/dma.h> #include <asm/k8.h> +#include <asm/x86_init.h> int gart_iommu_aperture; int gart_iommu_aperture_disabled __initdata; @@ -279,7 +280,8 @@ void __init early_gart_iommu_check(void) * or BIOS forget to put that in reserved. * try to update e820 to make that region as reserved. */ - int i, fix, slot; + u32 agp_aper_base = 0, agp_aper_order = 0; + int i, fix, slot, valid_agp = 0; u32 ctl; u32 aper_size = 0, aper_order = 0, last_aper_order = 0; u64 aper_base = 0, last_aper_base = 0; @@ -289,6 +291,8 @@ void __init early_gart_iommu_check(void) return; /* This is mostly duplicate of iommu_hole_init */ + agp_aper_base = search_agp_bridge(&agp_aper_order, &valid_agp); + fix = 0; for (i = 0; i < ARRAY_SIZE(bus_dev_ranges); i++) { int bus; @@ -341,10 +345,10 @@ void __init early_gart_iommu_check(void) } } - if (!fix) + if (valid_agp) return; - /* different nodes have different setting, disable them all at first*/ + /* disable them all at first */ for (i = 0; i < ARRAY_SIZE(bus_dev_ranges); i++) { int bus; int dev_base, dev_limit; @@ -400,6 +404,7 @@ void __init gart_iommu_hole_init(void) iommu_detected = 1; gart_iommu_aperture = 1; + x86_init.iommu.iommu_init = gart_iommu_init; aper_order = (read_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL) >> 1) & 7; aper_size = (32 * 1024 * 1024) << aper_order; @@ -456,8 +461,6 @@ out: if (aper_alloc) { /* Got the aperture from the AGP bridge */ - } else if (swiotlb && !valid_agp) { - /* Do nothing */ } else if ((!no_iommu && max_pfn > MAX_DMA32_PFN) || force_iommu || valid_agp || diff --git a/arch/x86/kernel/apic/Makefile b/arch/x86/kernel/apic/Makefile index da7b7b9f8bd8..565c1bfc507d 100644 --- a/arch/x86/kernel/apic/Makefile +++ b/arch/x86/kernel/apic/Makefile @@ -2,7 +2,7 @@ # Makefile for local APIC drivers and for the IO-APIC code # -obj-$(CONFIG_X86_LOCAL_APIC) += apic.o probe_$(BITS).o ipi.o nmi.o +obj-$(CONFIG_X86_LOCAL_APIC) += apic.o apic_noop.o probe_$(BITS).o ipi.o nmi.o obj-$(CONFIG_X86_IO_APIC) += io_apic.o obj-$(CONFIG_SMP) += ipi.o diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 894aa97f0717..00187f1fcfb7 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -61,12 +61,6 @@ unsigned int boot_cpu_physical_apicid = -1U; /* * The highest APIC ID seen during enumeration. - * - * On AMD, this determines the messaging protocol we can use: if all APIC IDs - * are in the 0 ... 7 range, then we can use logical addressing which - * has some performance advantages (better broadcasting). - * - * If there's an APIC ID above 8, we use physical addressing. */ unsigned int max_physical_apicid; @@ -241,28 +235,13 @@ static int modern_apic(void) } /* - * bare function to substitute write operation - * and it's _that_ fast :) - */ -static void native_apic_write_dummy(u32 reg, u32 v) -{ - WARN_ON_ONCE((cpu_has_apic || !disable_apic)); -} - -static u32 native_apic_read_dummy(u32 reg) -{ - WARN_ON_ONCE((cpu_has_apic && !disable_apic)); - return 0; -} - -/* - * right after this call apic->write/read doesn't do anything - * note that there is no restore operation it works one way + * right after this call apic become NOOP driven + * so apic->write/read doesn't do anything */ void apic_disable(void) { - apic->read = native_apic_read_dummy; - apic->write = native_apic_write_dummy; + pr_info("APIC: switched to apic NOOP\n"); + apic = &apic_noop; } void native_apic_wait_icr_idle(void) @@ -459,7 +438,7 @@ static void lapic_timer_setup(enum clock_event_mode mode, v = apic_read(APIC_LVTT); v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); apic_write(APIC_LVTT, v); - apic_write(APIC_TMICT, 0xffffffff); + apic_write(APIC_TMICT, 0); break; case CLOCK_EVT_MODE_RESUME: /* Nothing to do here */ @@ -602,7 +581,7 @@ calibrate_by_pmtimer(long deltapm, long *delta, long *deltatsc) res = (((u64)(*deltatsc)) * pm_100ms); do_div(res, deltapm); apic_printk(APIC_VERBOSE, "TSC delta adjusted to " - "PM-Timer: %lu (%ld) \n", + "PM-Timer: %lu (%ld)\n", (unsigned long)res, *deltatsc); *deltatsc = (long)res; } @@ -662,7 +641,7 @@ static int __init calibrate_APIC_clock(void) calibration_result = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS; apic_printk(APIC_VERBOSE, "..... delta %ld\n", delta); - apic_printk(APIC_VERBOSE, "..... mult: %ld\n", lapic_clockevent.mult); + apic_printk(APIC_VERBOSE, "..... mult: %u\n", lapic_clockevent.mult); apic_printk(APIC_VERBOSE, "..... calibration result: %u\n", calibration_result); @@ -1356,7 +1335,7 @@ void enable_x2apic(void) rdmsr(MSR_IA32_APICBASE, msr, msr2); if (!(msr & X2APIC_ENABLE)) { - pr_info("Enabling x2apic\n"); + printk_once(KERN_INFO "Enabling x2apic\n"); wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0); } } @@ -1392,14 +1371,11 @@ void __init enable_IR_x2apic(void) unsigned long flags; struct IO_APIC_route_entry **ioapic_entries = NULL; int ret, x2apic_enabled = 0; - int dmar_table_init_ret = 0; + int dmar_table_init_ret; -#ifdef CONFIG_INTR_REMAP dmar_table_init_ret = dmar_table_init(); - if (dmar_table_init_ret) - pr_debug("dmar_table_init() failed with %d:\n", - dmar_table_init_ret); -#endif + if (dmar_table_init_ret && !x2apic_supported()) + return; ioapic_entries = alloc_ioapic_entries(); if (!ioapic_entries) { @@ -1414,7 +1390,7 @@ void __init enable_IR_x2apic(void) } local_irq_save(flags); - mask_8259A(); + legacy_pic->mask_all(); mask_IO_APIC_setup(ioapic_entries); if (dmar_table_init_ret) @@ -1446,7 +1422,7 @@ void __init enable_IR_x2apic(void) nox2apic: if (!ret) /* IR enabling failed */ restore_IO_APIC_setup(ioapic_entries); - unmask_8259A(); + legacy_pic->restore_mask(); local_irq_restore(flags); out: @@ -1665,9 +1641,7 @@ int __init APIC_init_uniprocessor(void) #endif enable_IR_x2apic(); -#ifdef CONFIG_X86_64 default_setup_apic_routing(); -#endif verify_local_APIC(); connect_bsp_APIC(); @@ -1915,18 +1889,6 @@ void __cpuinit generic_processor_info(int apicid, int version) if (apicid > max_physical_apicid) max_physical_apicid = apicid; -#ifdef CONFIG_X86_32 - switch (boot_cpu_data.x86_vendor) { - case X86_VENDOR_INTEL: - if (num_processors > 8) - def_to_bigsmp = 1; - break; - case X86_VENDOR_AMD: - if (max_physical_apicid >= 8) - def_to_bigsmp = 1; - } -#endif - #if defined(CONFIG_SMP) || defined(CONFIG_X86_64) early_per_cpu(x86_cpu_to_apicid, cpu) = apicid; early_per_cpu(x86_bios_cpu_apicid, cpu) = apicid; @@ -2056,7 +2018,7 @@ static int lapic_resume(struct sys_device *dev) } mask_IO_APIC_setup(ioapic_entries); - mask_8259A(); + legacy_pic->mask_all(); } if (x2apic_mode) @@ -2100,7 +2062,7 @@ static int lapic_resume(struct sys_device *dev) if (intr_remapping_enabled) { reenable_intr_remapping(x2apic_mode); - unmask_8259A(); + legacy_pic->restore_mask(); restore_IO_APIC_setup(ioapic_entries); free_ioapic_entries(ioapic_entries); } diff --git a/arch/x86/kernel/apic/apic_flat_64.c b/arch/x86/kernel/apic/apic_flat_64.c index d0c99abc26c3..09d3b17ce0c2 100644 --- a/arch/x86/kernel/apic/apic_flat_64.c +++ b/arch/x86/kernel/apic/apic_flat_64.c @@ -223,7 +223,7 @@ struct apic apic_flat = { }; /* - * Physflat mode is used when there are more than 8 CPUs on a AMD system. + * Physflat mode is used when there are more than 8 CPUs on a system. * We cannot use logical delivery in this case because the mask * overflows, so use physical mode. */ @@ -240,6 +240,11 @@ static int physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) printk(KERN_DEBUG "system APIC only can use physical flat"); return 1; } + + if (!strncmp(oem_id, "IBM", 3) && !strncmp(oem_table_id, "EXA", 3)) { + printk(KERN_DEBUG "IBM Summit detected, will use apic physical"); + return 1; + } #endif return 0; @@ -306,10 +311,7 @@ physflat_cpu_mask_to_apicid_and(const struct cpumask *cpumask, if (cpumask_test_cpu(cpu, cpu_online_mask)) break; } - if (cpu < nr_cpu_ids) - return per_cpu(x86_cpu_to_apicid, cpu); - - return BAD_APICID; + return per_cpu(x86_cpu_to_apicid, cpu); } struct apic apic_physflat = { diff --git a/arch/x86/kernel/apic/apic_noop.c b/arch/x86/kernel/apic/apic_noop.c new file mode 100644 index 000000000000..e31b9ffe25f5 --- /dev/null +++ b/arch/x86/kernel/apic/apic_noop.c @@ -0,0 +1,200 @@ +/* + * NOOP APIC driver. + * + * Does almost nothing and should be substituted by a real apic driver via + * probe routine. + * + * Though in case if apic is disabled (for some reason) we try + * to not uglify the caller's code and allow to call (some) apic routines + * like self-ipi, etc... + */ + +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/module.h> +#include <linux/string.h> +#include <linux/kernel.h> +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <asm/fixmap.h> +#include <asm/mpspec.h> +#include <asm/apicdef.h> +#include <asm/apic.h> +#include <asm/setup.h> + +#include <linux/smp.h> +#include <asm/ipi.h> + +#include <linux/interrupt.h> +#include <asm/acpi.h> +#include <asm/e820.h> + +static void noop_init_apic_ldr(void) { } +static void noop_send_IPI_mask(const struct cpumask *cpumask, int vector) { } +static void noop_send_IPI_mask_allbutself(const struct cpumask *cpumask, int vector) { } +static void noop_send_IPI_allbutself(int vector) { } +static void noop_send_IPI_all(int vector) { } +static void noop_send_IPI_self(int vector) { } +static void noop_apic_wait_icr_idle(void) { } +static void noop_apic_icr_write(u32 low, u32 id) { } + +static int noop_wakeup_secondary_cpu(int apicid, unsigned long start_eip) +{ + return -1; +} + +static u32 noop_safe_apic_wait_icr_idle(void) +{ + return 0; +} + +static u64 noop_apic_icr_read(void) +{ + return 0; +} + +static int noop_cpu_to_logical_apicid(int cpu) +{ + return 0; +} + +static int noop_phys_pkg_id(int cpuid_apic, int index_msb) +{ + return 0; +} + +static unsigned int noop_get_apic_id(unsigned long x) +{ + return 0; +} + +static int noop_probe(void) +{ + /* + * NOOP apic should not ever be + * enabled via probe routine + */ + return 0; +} + +static int noop_apic_id_registered(void) +{ + /* + * if we would be really "pedantic" + * we should pass read_apic_id() here + * but since NOOP suppose APIC ID = 0 + * lets save a few cycles + */ + return physid_isset(0, phys_cpu_present_map); +} + +static const struct cpumask *noop_target_cpus(void) +{ + /* only BSP here */ + return cpumask_of(0); +} + +static unsigned long noop_check_apicid_used(physid_mask_t *map, int apicid) +{ + return physid_isset(apicid, *map); +} + +static unsigned long noop_check_apicid_present(int bit) +{ + return physid_isset(bit, phys_cpu_present_map); +} + +static void noop_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + if (cpu != 0) + pr_warning("APIC: Vector allocated for non-BSP cpu\n"); + cpumask_clear(retmask); + cpumask_set_cpu(cpu, retmask); +} + +int noop_apicid_to_node(int logical_apicid) +{ + /* we're always on node 0 */ + return 0; +} + +static u32 noop_apic_read(u32 reg) +{ + WARN_ON_ONCE((cpu_has_apic && !disable_apic)); + return 0; +} + +static void noop_apic_write(u32 reg, u32 v) +{ + WARN_ON_ONCE(cpu_has_apic && !disable_apic); +} + +struct apic apic_noop = { + .name = "noop", + .probe = noop_probe, + .acpi_madt_oem_check = NULL, + + .apic_id_registered = noop_apic_id_registered, + + .irq_delivery_mode = dest_LowestPrio, + /* logical delivery broadcast to all CPUs: */ + .irq_dest_mode = 1, + + .target_cpus = noop_target_cpus, + .disable_esr = 0, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = noop_check_apicid_used, + .check_apicid_present = noop_check_apicid_present, + + .vector_allocation_domain = noop_vector_allocation_domain, + .init_apic_ldr = noop_init_apic_ldr, + + .ioapic_phys_id_map = default_ioapic_phys_id_map, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = noop_apicid_to_node, + + .cpu_to_logical_apicid = noop_cpu_to_logical_apicid, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = physid_set_mask_of_physid, + + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + + .phys_pkg_id = noop_phys_pkg_id, + + .mps_oem_check = NULL, + + .get_apic_id = noop_get_apic_id, + .set_apic_id = NULL, + .apic_id_mask = 0x0F << 24, + + .cpu_mask_to_apicid = default_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = default_cpu_mask_to_apicid_and, + + .send_IPI_mask = noop_send_IPI_mask, + .send_IPI_mask_allbutself = noop_send_IPI_mask_allbutself, + .send_IPI_allbutself = noop_send_IPI_allbutself, + .send_IPI_all = noop_send_IPI_all, + .send_IPI_self = noop_send_IPI_self, + + .wakeup_secondary_cpu = noop_wakeup_secondary_cpu, + + /* should be safe */ + .trampoline_phys_low = DEFAULT_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = DEFAULT_TRAMPOLINE_PHYS_HIGH, + + .wait_for_init_deassert = NULL, + + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = NULL, + + .read = noop_apic_read, + .write = noop_apic_write, + .icr_read = noop_apic_icr_read, + .icr_write = noop_apic_icr_write, + .wait_icr_idle = noop_apic_wait_icr_idle, + .safe_wait_icr_idle = noop_safe_apic_wait_icr_idle, +}; diff --git a/arch/x86/kernel/apic/bigsmp_32.c b/arch/x86/kernel/apic/bigsmp_32.c index 77a06413b6b2..cb804c5091b9 100644 --- a/arch/x86/kernel/apic/bigsmp_32.c +++ b/arch/x86/kernel/apic/bigsmp_32.c @@ -35,7 +35,7 @@ static const struct cpumask *bigsmp_target_cpus(void) #endif } -static unsigned long bigsmp_check_apicid_used(physid_mask_t bitmap, int apicid) +static unsigned long bigsmp_check_apicid_used(physid_mask_t *map, int apicid) { return 0; } @@ -93,11 +93,6 @@ static int bigsmp_cpu_present_to_apicid(int mps_cpu) return BAD_APICID; } -static physid_mask_t bigsmp_apicid_to_cpu_present(int phys_apicid) -{ - return physid_mask_of_physid(phys_apicid); -} - /* Mapping from cpu number to logical apicid */ static inline int bigsmp_cpu_to_logical_apicid(int cpu) { @@ -106,10 +101,10 @@ static inline int bigsmp_cpu_to_logical_apicid(int cpu) return cpu_physical_id(cpu); } -static physid_mask_t bigsmp_ioapic_phys_id_map(physid_mask_t phys_map) +static void bigsmp_ioapic_phys_id_map(physid_mask_t *phys_map, physid_mask_t *retmap) { /* For clustered we don't have a good way to do this yet - hack */ - return physids_promote(0xFFL); + physids_promote(0xFFL, retmap); } static int bigsmp_check_phys_apicid_present(int phys_apicid) @@ -136,10 +131,7 @@ static unsigned int bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask, if (cpumask_test_cpu(cpu, cpu_online_mask)) break; } - if (cpu < nr_cpu_ids) - return bigsmp_cpu_to_logical_apicid(cpu); - - return BAD_APICID; + return bigsmp_cpu_to_logical_apicid(cpu); } static int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb) @@ -230,7 +222,7 @@ struct apic apic_bigsmp = { .apicid_to_node = bigsmp_apicid_to_node, .cpu_to_logical_apicid = bigsmp_cpu_to_logical_apicid, .cpu_present_to_apicid = bigsmp_cpu_present_to_apicid, - .apicid_to_cpu_present = bigsmp_apicid_to_cpu_present, + .apicid_to_cpu_present = physid_set_mask_of_physid, .setup_portio_remap = NULL, .check_phys_apicid_present = bigsmp_check_phys_apicid_present, .enable_apic_mode = NULL, diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c index 89174f847b49..dd2b5f264643 100644 --- a/arch/x86/kernel/apic/es7000_32.c +++ b/arch/x86/kernel/apic/es7000_32.c @@ -27,6 +27,9 @@ * * http://www.unisys.com */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/notifier.h> #include <linux/spinlock.h> #include <linux/cpumask.h> @@ -223,9 +226,9 @@ static int parse_unisys_oem(char *oemptr) mip_addr = val; mip = (struct mip_reg *)val; mip_reg = __va(mip); - pr_debug("es7000_mipcfg: host_reg = 0x%lx \n", + pr_debug("host_reg = 0x%lx\n", (unsigned long)host_reg); - pr_debug("es7000_mipcfg: mip_reg = 0x%lx \n", + pr_debug("mip_reg = 0x%lx\n", (unsigned long)mip_reg); success++; break; @@ -401,7 +404,7 @@ static void es7000_enable_apic_mode(void) if (!es7000_plat) return; - printk(KERN_INFO "ES7000: Enabling APIC mode.\n"); + pr_info("Enabling APIC mode.\n"); memset(&es7000_mip_reg, 0, sizeof(struct mip_reg)); es7000_mip_reg.off_0x00 = MIP_SW_APIC; es7000_mip_reg.off_0x38 = MIP_VALID; @@ -466,11 +469,11 @@ static const struct cpumask *es7000_target_cpus(void) return cpumask_of(smp_processor_id()); } -static unsigned long -es7000_check_apicid_used(physid_mask_t bitmap, int apicid) +static unsigned long es7000_check_apicid_used(physid_mask_t *map, int apicid) { return 0; } + static unsigned long es7000_check_apicid_present(int bit) { return physid_isset(bit, phys_cpu_present_map); @@ -514,8 +517,7 @@ static void es7000_setup_apic_routing(void) { int apic = per_cpu(x86_bios_cpu_apicid, smp_processor_id()); - printk(KERN_INFO - "Enabling APIC mode: %s. Using %d I/O APICs, target cpus %lx\n", + pr_info("Enabling APIC mode: %s. Using %d I/O APICs, target cpus %lx\n", (apic_version[apic] == 0x14) ? "Physical Cluster" : "Logical Cluster", nr_ioapics, cpumask_bits(es7000_target_cpus())[0]); @@ -539,14 +541,10 @@ static int es7000_cpu_present_to_apicid(int mps_cpu) static int cpu_id; -static physid_mask_t es7000_apicid_to_cpu_present(int phys_apicid) +static void es7000_apicid_to_cpu_present(int phys_apicid, physid_mask_t *retmap) { - physid_mask_t mask; - - mask = physid_mask_of_physid(cpu_id); + physid_set_mask_of_physid(cpu_id, retmap); ++cpu_id; - - return mask; } /* Mapping from cpu number to logical apicid */ @@ -561,10 +559,10 @@ static int es7000_cpu_to_logical_apicid(int cpu) #endif } -static physid_mask_t es7000_ioapic_phys_id_map(physid_mask_t phys_map) +static void es7000_ioapic_phys_id_map(physid_mask_t *phys_map, physid_mask_t *retmap) { /* For clustered we don't have a good way to do this yet - hack */ - return physids_promote(0xff); + physids_promote(0xFFL, retmap); } static int es7000_check_phys_apicid_present(int cpu_physical_apicid) diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index dc69f28489f5..e4e0ddcb1546 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -60,8 +60,6 @@ #include <asm/irq_remapping.h> #include <asm/hpet.h> #include <asm/hw_irq.h> -#include <asm/uv/uv_hub.h> -#include <asm/uv/uv_irq.h> #include <asm/apic.h> @@ -75,8 +73,8 @@ */ int sis_apic_bug = -1; -static DEFINE_SPINLOCK(ioapic_lock); -static DEFINE_SPINLOCK(vector_lock); +static DEFINE_RAW_SPINLOCK(ioapic_lock); +static DEFINE_RAW_SPINLOCK(vector_lock); /* * # of IRQ routing registers @@ -96,8 +94,6 @@ struct mpc_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* # of MP IRQ source entries */ int mp_irq_entries; -/* Number of legacy interrupts */ -static int nr_legacy_irqs __read_mostly = NR_IRQS_LEGACY; /* GSI interrupts */ static int nr_irqs_gsi = NR_IRQS_LEGACY; @@ -140,49 +136,12 @@ static struct irq_pin_list *get_one_free_irq_2_pin(int node) return pin; } -/* - * This is performance-critical, we want to do it O(1) - * - * Most irqs are mapped 1:1 with pins. - */ -struct irq_cfg { - struct irq_pin_list *irq_2_pin; - cpumask_var_t domain; - cpumask_var_t old_domain; - unsigned move_cleanup_count; - u8 vector; - u8 move_in_progress : 1; -}; - /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */ #ifdef CONFIG_SPARSE_IRQ -static struct irq_cfg irq_cfgx[] = { +static struct irq_cfg irq_cfgx[NR_IRQS_LEGACY]; #else -static struct irq_cfg irq_cfgx[NR_IRQS] = { +static struct irq_cfg irq_cfgx[NR_IRQS]; #endif - [0] = { .vector = IRQ0_VECTOR, }, - [1] = { .vector = IRQ1_VECTOR, }, - [2] = { .vector = IRQ2_VECTOR, }, - [3] = { .vector = IRQ3_VECTOR, }, - [4] = { .vector = IRQ4_VECTOR, }, - [5] = { .vector = IRQ5_VECTOR, }, - [6] = { .vector = IRQ6_VECTOR, }, - [7] = { .vector = IRQ7_VECTOR, }, - [8] = { .vector = IRQ8_VECTOR, }, - [9] = { .vector = IRQ9_VECTOR, }, - [10] = { .vector = IRQ10_VECTOR, }, - [11] = { .vector = IRQ11_VECTOR, }, - [12] = { .vector = IRQ12_VECTOR, }, - [13] = { .vector = IRQ13_VECTOR, }, - [14] = { .vector = IRQ14_VECTOR, }, - [15] = { .vector = IRQ15_VECTOR, }, -}; - -void __init io_apic_disable_legacy(void) -{ - nr_legacy_irqs = 0; - nr_irqs_gsi = 0; -} int __init arch_early_irq_init(void) { @@ -192,6 +151,11 @@ int __init arch_early_irq_init(void) int node; int i; + if (!legacy_pic->nr_legacy_irqs) { + nr_irqs_gsi = 0; + io_apic_irqs = ~0UL; + } + cfg = irq_cfgx; count = ARRAY_SIZE(irq_cfgx); node= cpu_to_node(boot_cpu_id); @@ -201,15 +165,21 @@ int __init arch_early_irq_init(void) 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); - if (i < nr_legacy_irqs) - cpumask_setall(cfg[i].domain); + /* + * For legacy IRQ's, start with assigning irq0 to irq15 to + * IRQ0_VECTOR to IRQ15_VECTOR on cpu 0. + */ + if (i < legacy_pic->nr_legacy_irqs) { + cfg[i].vector = IRQ0_VECTOR + i; + cpumask_set_cpu(0, cfg[i].domain); + } } return 0; } #ifdef CONFIG_SPARSE_IRQ -static struct irq_cfg *irq_cfg(unsigned int irq) +struct irq_cfg *irq_cfg(unsigned int irq) { struct irq_cfg *cfg = NULL; struct irq_desc *desc; @@ -361,7 +331,7 @@ void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) /* end for move_irq_desc */ #else -static struct irq_cfg *irq_cfg(unsigned int irq) +struct irq_cfg *irq_cfg(unsigned int irq) { return irq < nr_irqs ? irq_cfgx + irq : NULL; } @@ -422,7 +392,7 @@ static bool io_apic_level_ack_pending(struct irq_cfg *cfg) struct irq_pin_list *entry; unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); for_each_irq_pin(entry, cfg->irq_2_pin) { unsigned int reg; int pin; @@ -431,11 +401,11 @@ static bool io_apic_level_ack_pending(struct irq_cfg *cfg) reg = io_apic_read(entry->apic, 0x10 + pin*2); /* Is the remote IRR bit set? */ if (reg & IO_APIC_REDIR_REMOTE_IRR) { - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); return true; } } - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); return false; } @@ -449,10 +419,10 @@ static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) { union entry_union eu; unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); eu.w1 = io_apic_read(apic, 0x10 + 2 * pin); eu.w2 = io_apic_read(apic, 0x11 + 2 * pin); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); return eu.entry; } @@ -475,9 +445,9 @@ __ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); __ioapic_write_entry(apic, pin, e); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); } /* @@ -490,10 +460,10 @@ static void ioapic_mask_entry(int apic, int pin) unsigned long flags; union entry_union eu = { .entry.mask = 1 }; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic, 0x10 + 2*pin, eu.w1); io_apic_write(apic, 0x11 + 2*pin, eu.w2); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); } /* @@ -555,23 +525,41 @@ static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node, add_pin_to_irq_node(cfg, node, newapic, newpin); } +static void __io_apic_modify_irq(struct irq_pin_list *entry, + int mask_and, int mask_or, + void (*final)(struct irq_pin_list *entry)) +{ + unsigned int reg, pin; + + pin = entry->pin; + reg = io_apic_read(entry->apic, 0x10 + pin * 2); + reg &= mask_and; + reg |= mask_or; + io_apic_modify(entry->apic, 0x10 + pin * 2, reg); + if (final) + final(entry); +} + static void io_apic_modify_irq(struct irq_cfg *cfg, int mask_and, int mask_or, void (*final)(struct irq_pin_list *entry)) { - int pin; struct irq_pin_list *entry; - for_each_irq_pin(entry, cfg->irq_2_pin) { - unsigned int reg; - pin = entry->pin; - reg = io_apic_read(entry->apic, 0x10 + pin * 2); - reg &= mask_and; - reg |= mask_or; - io_apic_modify(entry->apic, 0x10 + pin * 2, reg); - if (final) - final(entry); - } + for_each_irq_pin(entry, cfg->irq_2_pin) + __io_apic_modify_irq(entry, mask_and, mask_or, final); +} + +static void __mask_and_edge_IO_APIC_irq(struct irq_pin_list *entry) +{ + __io_apic_modify_irq(entry, ~IO_APIC_REDIR_LEVEL_TRIGGER, + IO_APIC_REDIR_MASKED, NULL); +} + +static void __unmask_and_level_IO_APIC_irq(struct irq_pin_list *entry) +{ + __io_apic_modify_irq(entry, ~IO_APIC_REDIR_MASKED, + IO_APIC_REDIR_LEVEL_TRIGGER, NULL); } static void __unmask_IO_APIC_irq(struct irq_cfg *cfg) @@ -595,18 +583,6 @@ static void __mask_IO_APIC_irq(struct irq_cfg *cfg) io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync); } -static void __mask_and_edge_IO_APIC_irq(struct irq_cfg *cfg) -{ - io_apic_modify_irq(cfg, ~IO_APIC_REDIR_LEVEL_TRIGGER, - IO_APIC_REDIR_MASKED, NULL); -} - -static void __unmask_and_level_IO_APIC_irq(struct irq_cfg *cfg) -{ - io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, - IO_APIC_REDIR_LEVEL_TRIGGER, NULL); -} - static void mask_IO_APIC_irq_desc(struct irq_desc *desc) { struct irq_cfg *cfg = desc->chip_data; @@ -614,9 +590,9 @@ static void mask_IO_APIC_irq_desc(struct irq_desc *desc) BUG_ON(!cfg); - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); __mask_IO_APIC_irq(cfg); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); } static void unmask_IO_APIC_irq_desc(struct irq_desc *desc) @@ -624,9 +600,9 @@ static void unmask_IO_APIC_irq_desc(struct irq_desc *desc) struct irq_cfg *cfg = desc->chip_data; unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); __unmask_IO_APIC_irq(cfg); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); } static void mask_IO_APIC_irq(unsigned int irq) @@ -875,7 +851,7 @@ static int __init find_isa_irq_apic(int irq, int type) */ static int EISA_ELCR(unsigned int irq) { - if (irq < nr_legacy_irqs) { + if (irq < legacy_pic->nr_legacy_irqs) { unsigned int port = 0x4d0 + (irq >> 3); return (inb(port) >> (irq & 7)) & 1; } @@ -1150,12 +1126,12 @@ void lock_vector_lock(void) /* Used to the online set of cpus does not change * during assign_irq_vector. */ - spin_lock(&vector_lock); + raw_spin_lock(&vector_lock); } void unlock_vector_lock(void) { - spin_unlock(&vector_lock); + raw_spin_unlock(&vector_lock); } static int @@ -1172,12 +1148,13 @@ __assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask) * Also, we've got to be careful not to trash gate * 0x80, because int 0x80 is hm, kind of importantish. ;) */ - static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0; + static int current_vector = FIRST_EXTERNAL_VECTOR + VECTOR_OFFSET_START; + static int current_offset = VECTOR_OFFSET_START % 8; unsigned int old_vector; int cpu, err; cpumask_var_t tmp_mask; - if ((cfg->move_in_progress) || cfg->move_cleanup_count) + if (cfg->move_in_progress) return -EBUSY; if (!alloc_cpumask_var(&tmp_mask, GFP_ATOMIC)) @@ -1208,7 +1185,7 @@ next: if (vector >= first_system_vector) { /* If out of vectors on large boxen, must share them. */ offset = (offset + 1) % 8; - vector = FIRST_DEVICE_VECTOR + offset; + vector = FIRST_EXTERNAL_VECTOR + offset; } if (unlikely(current_vector == vector)) continue; @@ -1237,15 +1214,14 @@ next: return err; } -static int -assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask) +int assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask) { int err; unsigned long flags; - spin_lock_irqsave(&vector_lock, flags); + raw_spin_lock_irqsave(&vector_lock, flags); err = __assign_irq_vector(irq, cfg, mask); - spin_unlock_irqrestore(&vector_lock, flags); + raw_spin_unlock_irqrestore(&vector_lock, flags); return err; } @@ -1279,11 +1255,16 @@ static void __clear_irq_vector(int irq, struct irq_cfg *cfg) void __setup_vector_irq(int cpu) { /* Initialize vector_irq on a new cpu */ - /* This function must be called with vector_lock held */ int irq, vector; struct irq_cfg *cfg; struct irq_desc *desc; + /* + * vector_lock will make sure that we don't run into irq vector + * assignments that might be happening on another cpu in parallel, + * while we setup our initial vector to irq mappings. + */ + raw_spin_lock(&vector_lock); /* Mark the inuse vectors */ for_each_irq_desc(irq, desc) { cfg = desc->chip_data; @@ -1302,6 +1283,7 @@ void __setup_vector_irq(int cpu) if (!cpumask_test_cpu(cpu, cfg->domain)) per_cpu(vector_irq, cpu)[vector] = -1; } + raw_spin_unlock(&vector_lock); } static struct irq_chip ioapic_chip; @@ -1451,6 +1433,14 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq 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 + * the cfg->domain. + */ + if (irq < legacy_pic->nr_legacy_irqs && cpumask_test_cpu(0, cfg->domain)) + apic->vector_allocation_domain(0, cfg->domain); + if (assign_irq_vector(irq, cfg, apic->target_cpus())) return; @@ -1472,8 +1462,8 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq } ioapic_register_intr(irq, desc, trigger); - if (irq < nr_legacy_irqs) - disable_8259A_irq(irq); + if (irq < legacy_pic->nr_legacy_irqs) + legacy_pic->chip->mask(irq); ioapic_write_entry(apic_id, pin, entry); } @@ -1484,7 +1474,7 @@ static struct { static void __init setup_IO_APIC_irqs(void) { - int apic_id = 0, pin, idx, irq; + int apic_id, pin, idx, irq; int notcon = 0; struct irq_desc *desc; struct irq_cfg *cfg; @@ -1492,14 +1482,7 @@ static void __init setup_IO_APIC_irqs(void) apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); -#ifdef CONFIG_ACPI - if (!acpi_disabled && acpi_ioapic) { - apic_id = mp_find_ioapic(0); - if (apic_id < 0) - apic_id = 0; - } -#endif - + for (apic_id = 0; apic_id < nr_ioapics; apic_id++) for (pin = 0; pin < nr_ioapic_registers[apic_id]; pin++) { idx = find_irq_entry(apic_id, pin, mp_INT); if (idx == -1) { @@ -1521,6 +1504,9 @@ static void __init setup_IO_APIC_irqs(void) irq = pin_2_irq(idx, apic_id, pin); + if ((apic_id > 0) && (irq > 16)) + continue; + /* * Skip the timer IRQ if there's a quirk handler * installed and if it returns 1: @@ -1550,6 +1536,56 @@ static void __init setup_IO_APIC_irqs(void) } /* + * for the gsit that is not in first ioapic + * but could not use acpi_register_gsi() + * like some special sci in IBM x3330 + */ +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; + struct irq_cfg *cfg; + + /* + * Convert 'gsi' to 'ioapic.pin'. + */ + apic_id = mp_find_ioapic(gsi); + if (apic_id < 0) + return; + + pin = mp_find_ioapic_pin(apic_id, gsi); + idx = find_irq_entry(apic_id, pin, mp_INT); + if (idx == -1) + return; + + irq = pin_2_irq(idx, apic_id, pin); +#ifdef CONFIG_SPARSE_IRQ + desc = irq_to_desc(irq); + if (desc) + return; +#endif + desc = irq_to_desc_alloc_node(irq, node); + if (!desc) { + printk(KERN_INFO "can not get irq_desc for %d\n", irq); + 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)) { + pr_debug("Pin %d-%d already programmed\n", + mp_ioapics[apic_id].apicid, pin); + return; + } + set_bit(pin, mp_ioapic_routing[apic_id].pin_programmed); + + setup_IO_APIC_irq(apic_id, pin, irq, desc, + irq_trigger(idx), irq_polarity(idx)); +} + +/* * Set up the timer pin, possibly with the 8259A-master behind. */ static void __init setup_timer_IRQ0_pin(unsigned int apic_id, unsigned int pin, @@ -1599,9 +1635,6 @@ __apicdebuginit(void) print_IO_APIC(void) struct irq_desc *desc; unsigned int irq; - if (apic_verbosity == APIC_QUIET) - return; - printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); for (i = 0; i < nr_ioapics; i++) printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n", @@ -1615,14 +1648,14 @@ __apicdebuginit(void) print_IO_APIC(void) for (apic = 0; apic < nr_ioapics; apic++) { - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(apic, 0); reg_01.raw = io_apic_read(apic, 1); if (reg_01.bits.version >= 0x10) reg_02.raw = io_apic_read(apic, 2); if (reg_01.bits.version >= 0x20) reg_03.raw = io_apic_read(apic, 3); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); printk("\n"); printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].apicid); @@ -1661,7 +1694,7 @@ __apicdebuginit(void) print_IO_APIC(void) printk(KERN_DEBUG ".... IRQ redirection table:\n"); printk(KERN_DEBUG " NR Dst Mask Trig IRR Pol" - " Stat Dmod Deli Vect: \n"); + " Stat Dmod Deli Vect:\n"); for (i = 0; i <= reg_01.bits.entries; i++) { struct IO_APIC_route_entry entry; @@ -1708,9 +1741,6 @@ __apicdebuginit(void) print_APIC_field(int base) { int i; - if (apic_verbosity == APIC_QUIET) - return; - printk(KERN_DEBUG); for (i = 0; i < 8; i++) @@ -1724,9 +1754,6 @@ __apicdebuginit(void) print_local_APIC(void *dummy) unsigned int i, v, ver, maxlvt; u64 icr; - if (apic_verbosity == APIC_QUIET) - return; - printk(KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n", smp_processor_id(), hard_smp_processor_id()); v = apic_read(APIC_ID); @@ -1824,13 +1851,19 @@ __apicdebuginit(void) print_local_APIC(void *dummy) printk("\n"); } -__apicdebuginit(void) print_all_local_APICs(void) +__apicdebuginit(void) print_local_APICs(int maxcpu) { int cpu; + if (!maxcpu) + return; + preempt_disable(); - for_each_online_cpu(cpu) + for_each_online_cpu(cpu) { + if (cpu >= maxcpu) + break; smp_call_function_single(cpu, print_local_APIC, NULL, 1); + } preempt_enable(); } @@ -1839,12 +1872,12 @@ __apicdebuginit(void) print_PIC(void) unsigned int v; unsigned long flags; - if (apic_verbosity == APIC_QUIET || !nr_legacy_irqs) + if (!legacy_pic->nr_legacy_irqs) return; printk(KERN_DEBUG "\nprinting PIC contents\n"); - spin_lock_irqsave(&i8259A_lock, flags); + raw_spin_lock_irqsave(&i8259A_lock, flags); v = inb(0xa1) << 8 | inb(0x21); printk(KERN_DEBUG "... PIC IMR: %04x\n", v); @@ -1858,7 +1891,7 @@ __apicdebuginit(void) print_PIC(void) outb(0x0a,0xa0); outb(0x0a,0x20); - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); printk(KERN_DEBUG "... PIC ISR: %04x\n", v); @@ -1866,21 +1899,41 @@ __apicdebuginit(void) print_PIC(void) printk(KERN_DEBUG "... PIC ELCR: %04x\n", v); } -__apicdebuginit(int) print_all_ICs(void) +static int __initdata show_lapic = 1; +static __init int setup_show_lapic(char *arg) +{ + int num = -1; + + if (strcmp(arg, "all") == 0) { + show_lapic = CONFIG_NR_CPUS; + } else { + get_option(&arg, &num); + if (num >= 0) + show_lapic = num; + } + + return 1; +} +__setup("show_lapic=", setup_show_lapic); + +__apicdebuginit(int) print_ICs(void) { + if (apic_verbosity == APIC_QUIET) + return 0; + print_PIC(); /* don't print out if apic is not there */ if (!cpu_has_apic && !apic_from_smp_config()) return 0; - print_all_local_APICs(); + print_local_APICs(show_lapic); print_IO_APIC(); return 0; } -fs_initcall(print_all_ICs); +fs_initcall(print_ICs); /* Where if anywhere is the i8259 connect in external int mode */ @@ -1897,13 +1950,13 @@ void __init enable_IO_APIC(void) * The number of IO-APIC IRQ registers (== #pins): */ for (apic = 0; apic < nr_ioapics; apic++) { - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); reg_01.raw = io_apic_read(apic, 1); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); nr_ioapic_registers[apic] = reg_01.bits.entries+1; } - if (!nr_legacy_irqs) + if (!legacy_pic->nr_legacy_irqs) return; for(apic = 0; apic < nr_ioapics; apic++) { @@ -1960,7 +2013,7 @@ void disable_IO_APIC(void) */ clear_IO_APIC(); - if (!nr_legacy_irqs) + if (!legacy_pic->nr_legacy_irqs) return; /* @@ -2031,7 +2084,7 @@ void __init setup_ioapic_ids_from_mpc(void) * This is broken; anything with a real cpu count has to * circumvent this idiocy regardless. */ - phys_id_present_map = apic->ioapic_phys_id_map(phys_cpu_present_map); + apic->ioapic_phys_id_map(&phys_cpu_present_map, &phys_id_present_map); /* * Set the IOAPIC ID to the value stored in the MPC table. @@ -2039,9 +2092,9 @@ void __init setup_ioapic_ids_from_mpc(void) for (apic_id = 0; apic_id < nr_ioapics; apic_id++) { /* Read the register 0 value */ - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(apic_id, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); old_id = mp_ioapics[apic_id].apicid; @@ -2058,7 +2111,7 @@ void __init setup_ioapic_ids_from_mpc(void) * system must have a unique ID or we get lots of nice * 'stuck on smp_invalidate_needed IPI wait' messages. */ - if (apic->check_apicid_used(phys_id_present_map, + if (apic->check_apicid_used(&phys_id_present_map, mp_ioapics[apic_id].apicid)) { printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n", apic_id, mp_ioapics[apic_id].apicid); @@ -2073,7 +2126,7 @@ void __init setup_ioapic_ids_from_mpc(void) mp_ioapics[apic_id].apicid = i; } else { physid_mask_t tmp; - tmp = apic->apicid_to_cpu_present(mp_ioapics[apic_id].apicid); + apic->apicid_to_cpu_present(mp_ioapics[apic_id].apicid, &tmp); apic_printk(APIC_VERBOSE, "Setting %d in the " "phys_id_present_map\n", mp_ioapics[apic_id].apicid); @@ -2100,16 +2153,16 @@ void __init setup_ioapic_ids_from_mpc(void) mp_ioapics[apic_id].apicid); reg_00.bits.ID = mp_ioapics[apic_id].apicid; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic_id, 0, reg_00.raw); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); /* * Sanity check */ - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(apic_id, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); if (reg_00.bits.ID != mp_ioapics[apic_id].apicid) printk("could not set ID!\n"); else @@ -2192,15 +2245,15 @@ static unsigned int startup_ioapic_irq(unsigned int irq) unsigned long flags; struct irq_cfg *cfg; - spin_lock_irqsave(&ioapic_lock, flags); - if (irq < nr_legacy_irqs) { - disable_8259A_irq(irq); - if (i8259A_irq_pending(irq)) + raw_spin_lock_irqsave(&ioapic_lock, flags); + if (irq < legacy_pic->nr_legacy_irqs) { + legacy_pic->chip->mask(irq); + if (legacy_pic->irq_pending(irq)) was_pending = 1; } cfg = irq_cfg(irq); __unmask_IO_APIC_irq(cfg); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); return was_pending; } @@ -2211,9 +2264,9 @@ static int ioapic_retrigger_irq(unsigned int irq) struct irq_cfg *cfg = irq_cfg(irq); unsigned long flags; - spin_lock_irqsave(&vector_lock, flags); + raw_spin_lock_irqsave(&vector_lock, flags); apic->send_IPI_mask(cpumask_of(cpumask_first(cfg->domain)), cfg->vector); - spin_unlock_irqrestore(&vector_lock, flags); + raw_spin_unlock_irqrestore(&vector_lock, flags); return 1; } @@ -2228,20 +2281,16 @@ static int ioapic_retrigger_irq(unsigned int irq) */ #ifdef CONFIG_SMP -static void send_cleanup_vector(struct irq_cfg *cfg) +void send_cleanup_vector(struct irq_cfg *cfg) { cpumask_var_t cleanup_mask; if (unlikely(!alloc_cpumask_var(&cleanup_mask, GFP_ATOMIC))) { unsigned int i; - cfg->move_cleanup_count = 0; - for_each_cpu_and(i, cfg->old_domain, cpu_online_mask) - cfg->move_cleanup_count++; for_each_cpu_and(i, cfg->old_domain, cpu_online_mask) apic->send_IPI_mask(cpumask_of(i), IRQ_MOVE_CLEANUP_VECTOR); } else { cpumask_and(cleanup_mask, cfg->old_domain, cpu_online_mask); - cfg->move_cleanup_count = cpumask_weight(cleanup_mask); apic->send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR); free_cpumask_var(cleanup_mask); } @@ -2272,31 +2321,30 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq } } -static int -assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask); - /* * Either sets desc->affinity to a valid value, and returns - * ->cpu_mask_to_apicid of that, or returns BAD_APICID and + * ->cpu_mask_to_apicid of that in dest_id, or returns -1 and * leaves desc->affinity untouched. */ -static unsigned int -set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask) +unsigned int +set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask, + unsigned int *dest_id) { struct irq_cfg *cfg; unsigned int irq; if (!cpumask_intersects(mask, cpu_online_mask)) - return BAD_APICID; + return -1; irq = desc->irq; cfg = desc->chip_data; if (assign_irq_vector(irq, cfg, mask)) - return BAD_APICID; + return -1; cpumask_copy(desc->affinity, mask); - return apic->cpu_mask_to_apicid_and(desc->affinity, cfg->domain); + *dest_id = apic->cpu_mask_to_apicid_and(desc->affinity, cfg->domain); + return 0; } static int @@ -2311,15 +2359,14 @@ set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask) irq = desc->irq; cfg = desc->chip_data; - spin_lock_irqsave(&ioapic_lock, flags); - dest = set_desc_affinity(desc, mask); - if (dest != BAD_APICID) { + raw_spin_lock_irqsave(&ioapic_lock, flags); + ret = set_desc_affinity(desc, mask, &dest); + if (!ret) { /* Only the high 8 bits are valid. */ dest = SET_APIC_LOGICAL_ID(dest); __target_IO_APIC_irq(irq, dest, cfg); - ret = 0; } - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); return ret; } @@ -2432,8 +2479,13 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) continue; cfg = irq_cfg(irq); - spin_lock(&desc->lock); - if (!cfg->move_cleanup_count) + raw_spin_lock(&desc->lock); + + /* + * Check if the irq migration is in progress. If so, we + * haven't received the cleanup request yet for this irq. + */ + if (cfg->move_in_progress) goto unlock; if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) @@ -2452,29 +2504,40 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) goto unlock; } __get_cpu_var(vector_irq)[vector] = -1; - cfg->move_cleanup_count--; unlock: - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } irq_exit(); } -static void irq_complete_move(struct irq_desc **descp) +static void __irq_complete_move(struct irq_desc **descp, unsigned vector) { struct irq_desc *desc = *descp; struct irq_cfg *cfg = desc->chip_data; - unsigned vector, me; + unsigned me; if (likely(!cfg->move_in_progress)) return; - vector = ~get_irq_regs()->orig_ax; me = smp_processor_id(); if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) send_cleanup_vector(cfg); } + +static void irq_complete_move(struct irq_desc **descp) +{ + __irq_complete_move(descp, ~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; + + __irq_complete_move(&desc, cfg->vector); +} #else static inline void irq_complete_move(struct irq_desc **descp) {} #endif @@ -2490,6 +2553,59 @@ static void ack_apic_edge(unsigned int irq) atomic_t irq_mis_count; +/* + * IO-APIC versions below 0x20 don't support EOI register. + * For the record, here is the information about various versions: + * 0Xh 82489DX + * 1Xh I/OAPIC or I/O(x)APIC which are not PCI 2.2 Compliant + * 2Xh I/O(x)APIC which is PCI 2.2 Compliant + * 30h-FFh Reserved + * + * Some of the Intel ICH Specs (ICH2 to ICH5) documents the io-apic + * version as 0x2. This is an error with documentation and these ICH chips + * use io-apic's of version 0x20. + * + * For IO-APIC's with EOI register, we use that to do an explicit EOI. + * 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) +{ + struct irq_pin_list *entry; + + for_each_irq_pin(entry, cfg->irq_2_pin) { + if (mp_ioapics[entry->apic].apicver >= 0x20) { + /* + * Intr-remapping uses pin number as the virtual vector + * in the RTE. Actual vector is programmed in + * intr-remapping table entry. Hence for the io-apic + * EOI we use the pin number. + */ + if (irq_remapped(irq)) + io_apic_eoi(entry->apic, entry->pin); + else + io_apic_eoi(entry->apic, cfg->vector); + } else { + __mask_and_edge_IO_APIC_irq(entry); + __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) { struct irq_desc *desc = irq_to_desc(irq); @@ -2525,6 +2641,19 @@ static void ack_apic_level(unsigned int irq) * level-triggered interrupt. We mask the source for the time of the * operation to prevent an edge-triggered interrupt escaping meanwhile. * The idea is from Manfred Spraul. --macro + * + * Also in the case when cpu goes offline, fixup_irqs() will forward + * any unhandled interrupt on the offlined cpu to the new cpu + * destination that is handling the corresponding interrupt. This + * interrupt forwarding is done via IPI's. Hence, in this case also + * level-triggered io-apic interrupt will be seen as an edge + * interrupt in the IRR. And we can't rely on the cpu's EOI + * to be broadcasted to the IO-APIC's which will clear the remoteIRR + * corresponding to the level-triggered interrupt. Hence on IO-APIC's + * supporting EOI register, we do an explicit EOI to clear the + * remote IRR and on IO-APIC's which don't have an EOI register, + * 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; @@ -2536,6 +2665,19 @@ static void ack_apic_level(unsigned int irq) */ ack_APIC_irq(); + /* + * Tail end of clearing remote IRR bit (either by delivering the EOI + * message via io-apic EOI register write or simulating it using + * mask+edge followed by unnask+level logic) manually when the + * level triggered interrupt is seen as the edge triggered interrupt + * at the cpu. + */ + if (!(v & (1 << (i & 0x1f)))) { + atomic_inc(&irq_mis_count); + + eoi_ioapic_irq(desc); + } + /* Now we can move and renable the irq */ if (unlikely(do_unmask_irq)) { /* Only migrate the irq if the ack has been received. @@ -2569,41 +2711,9 @@ static void ack_apic_level(unsigned int irq) move_masked_irq(irq); unmask_IO_APIC_irq_desc(desc); } - - /* Tail end of version 0x11 I/O APIC bug workaround */ - if (!(v & (1 << (i & 0x1f)))) { - atomic_inc(&irq_mis_count); - spin_lock(&ioapic_lock); - __mask_and_edge_IO_APIC_irq(cfg); - __unmask_and_level_IO_APIC_irq(cfg); - spin_unlock(&ioapic_lock); - } } #ifdef CONFIG_INTR_REMAP -static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) -{ - struct irq_pin_list *entry; - - for_each_irq_pin(entry, cfg->irq_2_pin) - io_apic_eoi(entry->apic, entry->pin); -} - -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; - - spin_lock_irqsave(&ioapic_lock, flags); - __eoi_ioapic_irq(irq, cfg); - spin_unlock_irqrestore(&ioapic_lock, flags); -} - static void ir_ack_apic_edge(unsigned int irq) { ack_APIC_irq(); @@ -2671,8 +2781,8 @@ static inline void init_IO_APIC_traps(void) * so default to an old-fashioned 8259 * interrupt if we can.. */ - if (irq < nr_legacy_irqs) - make_8259A_irq(irq); + if (irq < legacy_pic->nr_legacy_irqs) + legacy_pic->make_irq(irq); else /* Strange. Oh, well.. */ desc->chip = &no_irq_chip; @@ -2829,7 +2939,7 @@ static inline void __init check_timer(void) /* * get/set the timer IRQ vector: */ - disable_8259A_irq(0); + legacy_pic->chip->mask(0); assign_irq_vector(0, cfg, apic->target_cpus()); /* @@ -2842,7 +2952,7 @@ static inline void __init check_timer(void) * automatically. */ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); - init_8259A(1); + legacy_pic->init(1); #ifdef CONFIG_X86_32 { unsigned int ver; @@ -2901,7 +3011,7 @@ static inline void __init check_timer(void) if (timer_irq_works()) { if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); - enable_8259A_irq(0); + legacy_pic->chip->unmask(0); } if (disable_timer_pin_1 > 0) clear_IO_APIC_pin(0, pin1); @@ -2924,14 +3034,14 @@ 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); - enable_8259A_irq(0); + legacy_pic->chip->unmask(0); if (timer_irq_works()) { apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); timer_through_8259 = 1; if (nmi_watchdog == NMI_IO_APIC) { - disable_8259A_irq(0); + legacy_pic->chip->mask(0); setup_nmi(); - enable_8259A_irq(0); + legacy_pic->chip->unmask(0); } goto out; } @@ -2939,7 +3049,7 @@ static inline void __init check_timer(void) * Cleanup, just in case ... */ local_irq_disable(); - disable_8259A_irq(0); + legacy_pic->chip->mask(0); clear_IO_APIC_pin(apic2, pin2); apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); } @@ -2958,22 +3068,22 @@ static inline void __init check_timer(void) lapic_register_intr(0, desc); apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector); /* Fixed mode */ - enable_8259A_irq(0); + legacy_pic->chip->unmask(0); if (timer_irq_works()) { apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); goto out; } local_irq_disable(); - disable_8259A_irq(0); + legacy_pic->chip->mask(0); apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | cfg->vector); apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n"); apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer as ExtINT IRQ...\n"); - init_8259A(0); - make_8259A_irq(0); + legacy_pic->init(0); + legacy_pic->make_irq(0); apic_write(APIC_LVT0, APIC_DM_EXTINT); unlock_ExtINT_logic(); @@ -3015,7 +3125,7 @@ void __init setup_IO_APIC(void) /* * calling enable_IO_APIC() is moved to setup_local_APIC for BP */ - io_apic_irqs = nr_legacy_irqs ? ~PIC_IRQS : ~0UL; + io_apic_irqs = legacy_pic->nr_legacy_irqs ? ~PIC_IRQS : ~0UL; apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n"); /* @@ -3026,7 +3136,7 @@ void __init setup_IO_APIC(void) sync_Arb_IDs(); setup_IO_APIC_irqs(); init_IO_APIC_traps(); - if (nr_legacy_irqs) + if (legacy_pic->nr_legacy_irqs) check_timer(); } @@ -3075,13 +3185,13 @@ static int ioapic_resume(struct sys_device *dev) data = container_of(dev, struct sysfs_ioapic_data, dev); entry = data->entry; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(dev->id, 0); if (reg_00.bits.ID != mp_ioapics[dev->id].apicid) { reg_00.bits.ID = mp_ioapics[dev->id].apicid; io_apic_write(dev->id, 0, reg_00.raw); } - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); for (i = 0; i < nr_ioapic_registers[dev->id]; i++) ioapic_write_entry(dev->id, i, entry[i]); @@ -3144,7 +3254,7 @@ unsigned int create_irq_nr(unsigned int irq_want, int node) if (irq_want < nr_irqs_gsi) irq_want = nr_irqs_gsi; - spin_lock_irqsave(&vector_lock, flags); + 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) { @@ -3157,19 +3267,17 @@ unsigned int create_irq_nr(unsigned int irq_want, int node) continue; desc_new = move_irq_desc(desc_new, node); + cfg_new = desc_new->chip_data; if (__assign_irq_vector(new, cfg_new, apic->target_cpus()) == 0) irq = new; break; } - spin_unlock_irqrestore(&vector_lock, flags); + raw_spin_unlock_irqrestore(&vector_lock, flags); + + if (irq > 0) + dynamic_irq_init_keep_chip_data(irq); - if (irq > 0) { - dynamic_irq_init(irq); - /* restore it, in case dynamic_irq_init clear it */ - if (desc_new) - desc_new->chip_data = cfg_new; - } return irq; } @@ -3191,27 +3299,21 @@ int create_irq(void) void destroy_irq(unsigned int irq) { unsigned long flags; - struct irq_cfg *cfg; - struct irq_desc *desc; - /* store it, in case dynamic_irq_cleanup clear it */ - desc = irq_to_desc(irq); - cfg = desc->chip_data; - dynamic_irq_cleanup(irq); - /* connect back irq_cfg */ - desc->chip_data = cfg; + dynamic_irq_cleanup_keep_chip_data(irq); free_irte(irq); - spin_lock_irqsave(&vector_lock, flags); - __clear_irq_vector(irq, cfg); - spin_unlock_irqrestore(&vector_lock, flags); + raw_spin_lock_irqsave(&vector_lock, flags); + __clear_irq_vector(irq, get_irq_chip_data(irq)); + raw_spin_unlock_irqrestore(&vector_lock, flags); } /* * MSI message composition */ #ifdef CONFIG_PCI_MSI -static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg) +static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, + struct msi_msg *msg, u8 hpet_id) { struct irq_cfg *cfg; int err; @@ -3245,7 +3347,10 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms irte.dest_id = IRTE_DEST(dest); /* Set source-id of interrupt request */ - set_msi_sid(&irte, pdev); + if (pdev) + set_msi_sid(&irte, pdev); + else + set_hpet_sid(&irte, hpet_id); modify_irte(irq, &irte); @@ -3291,8 +3396,7 @@ static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) struct msi_msg msg; unsigned int dest; - dest = set_desc_affinity(desc, mask); - if (dest == BAD_APICID) + if (set_desc_affinity(desc, mask, &dest)) return -1; cfg = desc->chip_data; @@ -3324,8 +3428,7 @@ ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) if (get_irte(irq, &irte)) return -1; - dest = set_desc_affinity(desc, mask); - if (dest == BAD_APICID) + if (set_desc_affinity(desc, mask, &dest)) return -1; irte.vector = cfg->vector; @@ -3410,7 +3513,7 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) int ret; struct msi_msg msg; - ret = msi_compose_msg(dev, irq, &msg); + ret = msi_compose_msg(dev, irq, &msg, -1); if (ret < 0) return ret; @@ -3507,8 +3610,7 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) struct msi_msg msg; unsigned int dest; - dest = set_desc_affinity(desc, mask); - if (dest == BAD_APICID) + if (set_desc_affinity(desc, mask, &dest)) return -1; cfg = desc->chip_data; @@ -3543,7 +3645,7 @@ int arch_setup_dmar_msi(unsigned int irq) int ret; struct msi_msg msg; - ret = msi_compose_msg(NULL, irq, &msg); + ret = msi_compose_msg(NULL, irq, &msg, -1); if (ret < 0) return ret; dmar_msi_write(irq, &msg); @@ -3563,8 +3665,7 @@ static int hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask) struct msi_msg msg; unsigned int dest; - dest = set_desc_affinity(desc, mask); - if (dest == BAD_APICID) + if (set_desc_affinity(desc, mask, &dest)) return -1; cfg = desc->chip_data; @@ -3583,6 +3684,19 @@ 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, +#ifdef CONFIG_INTR_REMAP + .ack = ir_ack_apic_edge, +#ifdef CONFIG_SMP + .set_affinity = ir_set_msi_irq_affinity, +#endif +#endif + .retrigger = ioapic_retrigger_irq, +}; + static struct irq_chip hpet_msi_type = { .name = "HPET_MSI", .unmask = hpet_msi_unmask, @@ -3594,20 +3708,36 @@ static struct irq_chip hpet_msi_type = { .retrigger = ioapic_retrigger_irq, }; -int arch_setup_hpet_msi(unsigned int 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); - ret = msi_compose_msg(NULL, irq, &msg); + if (intr_remapping_enabled) { + struct intel_iommu *iommu = map_hpet_to_ir(id); + int index; + + if (!iommu) + return -1; + + index = alloc_irte(iommu, irq, 1); + if (index < 0) + return -1; + } + + ret = msi_compose_msg(NULL, irq, &msg, id); if (ret < 0) return ret; hpet_msi_write(irq, &msg); desc->status |= IRQ_MOVE_PCNTXT; - set_irq_chip_and_handler_name(irq, &hpet_msi_type, handle_edge_irq, - "edge"); + if (irq_remapped(irq)) + set_irq_chip_and_handler_name(irq, &ir_hpet_msi_type, + handle_edge_irq, "edge"); + else + set_irq_chip_and_handler_name(irq, &hpet_msi_type, + handle_edge_irq, "edge"); return 0; } @@ -3641,8 +3771,7 @@ static int set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask) struct irq_cfg *cfg; unsigned int dest; - dest = set_desc_affinity(desc, mask); - if (dest == BAD_APICID) + if (set_desc_affinity(desc, mask, &dest)) return -1; cfg = desc->chip_data; @@ -3708,83 +3837,14 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) } #endif /* CONFIG_HT_IRQ */ -#ifdef CONFIG_X86_UV -/* - * Re-target the irq to the specified CPU and enable the specified MMR located - * on the specified blade to allow the sending of MSIs to the specified CPU. - */ -int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, - unsigned long mmr_offset) -{ - const struct cpumask *eligible_cpu = cpumask_of(cpu); - struct irq_cfg *cfg; - int mmr_pnode; - unsigned long mmr_value; - struct uv_IO_APIC_route_entry *entry; - unsigned long flags; - int 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; - - spin_lock_irqsave(&vector_lock, flags); - set_irq_chip_and_handler_name(irq, &uv_irq_chip, handle_percpu_irq, - irq_name); - spin_unlock_irqrestore(&vector_lock, flags); - - mmr_value = 0; - entry = (struct uv_IO_APIC_route_entry *)&mmr_value; - entry->vector = cfg->vector; - entry->delivery_mode = apic->irq_delivery_mode; - entry->dest_mode = apic->irq_dest_mode; - entry->polarity = 0; - entry->trigger = 0; - entry->mask = 0; - entry->dest = apic->cpu_mask_to_apicid(eligible_cpu); - - mmr_pnode = uv_blade_to_pnode(mmr_blade); - uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); - - if (cfg->move_in_progress) - send_cleanup_vector(cfg); - - return irq; -} - -/* - * Disable the specified MMR located on the specified blade so that MSIs are - * longer allowed to be sent. - */ -void arch_disable_uv_irq(int mmr_blade, unsigned long mmr_offset) -{ - unsigned long mmr_value; - struct uv_IO_APIC_route_entry *entry; - int mmr_pnode; - - BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); - - mmr_value = 0; - entry = (struct uv_IO_APIC_route_entry *)&mmr_value; - entry->mask = 1; - - mmr_pnode = uv_blade_to_pnode(mmr_blade); - uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); -} -#endif /* CONFIG_X86_64 */ - int __init io_apic_get_redir_entries (int ioapic) { union IO_APIC_reg_01 reg_01; unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); reg_01.raw = io_apic_read(ioapic, 1); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); return reg_01.bits.entries; } @@ -3867,7 +3927,7 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, /* * IRQs < 16 are already in the irq_2_pin[] map */ - if (irq >= nr_legacy_irqs) { + if (irq >= legacy_pic->nr_legacy_irqs) { cfg = desc->chip_data; if (add_pin_to_irq_node_nopanic(cfg, node, ioapic, pin)) { printk(KERN_INFO "can not add pin %d for irq %d\n", @@ -3944,11 +4004,11 @@ int __init io_apic_get_unique_id(int ioapic, int apic_id) */ if (physids_empty(apic_id_map)) - apic_id_map = apic->ioapic_phys_id_map(phys_cpu_present_map); + apic->ioapic_phys_id_map(&phys_cpu_present_map, &apic_id_map); - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(ioapic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); if (apic_id >= get_physical_broadcast()) { printk(KERN_WARNING "IOAPIC[%d]: Invalid apic_id %d, trying " @@ -3960,10 +4020,10 @@ int __init io_apic_get_unique_id(int ioapic, int apic_id) * Every APIC in a system must have a unique ID or we get lots of nice * 'stuck on smp_invalidate_needed IPI wait' messages. */ - if (apic->check_apicid_used(apic_id_map, apic_id)) { + if (apic->check_apicid_used(&apic_id_map, apic_id)) { for (i = 0; i < get_physical_broadcast(); i++) { - if (!apic->check_apicid_used(apic_id_map, i)) + if (!apic->check_apicid_used(&apic_id_map, i)) break; } @@ -3976,16 +4036,16 @@ int __init io_apic_get_unique_id(int ioapic, int apic_id) apic_id = i; } - tmp = apic->apicid_to_cpu_present(apic_id); + apic->apicid_to_cpu_present(apic_id, &tmp); physids_or(apic_id_map, apic_id_map, tmp); if (reg_00.bits.ID != apic_id) { reg_00.bits.ID = apic_id; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(ioapic, 0, reg_00.raw); reg_00.raw = io_apic_read(ioapic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); /* Sanity check */ if (reg_00.bits.ID != apic_id) { @@ -4006,9 +4066,9 @@ int __init io_apic_get_version(int ioapic) union IO_APIC_reg_01 reg_01; unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); + raw_spin_lock_irqsave(&ioapic_lock, flags); reg_01.raw = io_apic_read(ioapic, 1); - spin_unlock_irqrestore(&ioapic_lock, flags); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); return reg_01.bits.version; } @@ -4040,27 +4100,23 @@ int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) #ifdef CONFIG_SMP void __init setup_ioapic_dest(void) { - int pin, ioapic = 0, irq, irq_entry; + int pin, ioapic, irq, irq_entry; struct irq_desc *desc; const struct cpumask *mask; if (skip_ioapic_setup == 1) return; -#ifdef CONFIG_ACPI - if (!acpi_disabled && acpi_ioapic) { - ioapic = mp_find_ioapic(0); - if (ioapic < 0) - ioapic = 0; - } -#endif - + for (ioapic = 0; ioapic < nr_ioapics; ioapic++) for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) { irq_entry = find_irq_entry(ioapic, pin, mp_INT); if (irq_entry == -1) continue; irq = pin_2_irq(irq_entry, ioapic, pin); + if ((ioapic > 0) && (irq > 16)) + continue; + desc = irq_to_desc(irq); /* @@ -4106,7 +4162,7 @@ static struct resource * __init ioapic_setup_resources(int nr_ioapics) for (i = 0; i < nr_ioapics; i++) { res[i].name = mem; res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY; - sprintf(mem, "IOAPIC %u", i); + snprintf(mem, IOAPIC_RESOURCE_NAME_SIZE, "IOAPIC %u", i); mem += IOAPIC_RESOURCE_NAME_SIZE; } @@ -4140,18 +4196,17 @@ void __init ioapic_init_mappings(void) #ifdef CONFIG_X86_32 fake_ioapic_page: #endif - ioapic_phys = (unsigned long) - alloc_bootmem_pages(PAGE_SIZE); + ioapic_phys = (unsigned long)alloc_bootmem_pages(PAGE_SIZE); ioapic_phys = __pa(ioapic_phys); } set_fixmap_nocache(idx, ioapic_phys); - apic_printk(APIC_VERBOSE, - "mapped IOAPIC to %08lx (%08lx)\n", - __fix_to_virt(idx), ioapic_phys); + apic_printk(APIC_VERBOSE, "mapped IOAPIC to %08lx (%08lx)\n", + __fix_to_virt(idx) + (ioapic_phys & ~PAGE_MASK), + ioapic_phys); idx++; ioapic_res->start = ioapic_phys; - ioapic_res->end = ioapic_phys + (4 * 1024) - 1; + ioapic_res->end = ioapic_phys + IO_APIC_SLOT_SIZE - 1; ioapic_res++; } } @@ -4246,3 +4301,24 @@ void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) nr_ioapics++; } + +/* Enable IOAPIC early just for system timer */ +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); + + 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); +} diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c index 7ff61d6a188a..8aa65adbd25d 100644 --- a/arch/x86/kernel/apic/nmi.c +++ b/arch/x86/kernel/apic/nmi.c @@ -39,7 +39,8 @@ int unknown_nmi_panic; int nmi_watchdog_enabled; -static cpumask_t backtrace_mask __read_mostly; +/* For reliability, we're prepared to waste bits here. */ +static DECLARE_BITMAP(backtrace_mask, NR_CPUS) __read_mostly; /* nmi_active: * >0: the lapic NMI watchdog is active, but can be disabled @@ -176,7 +177,7 @@ int __init check_nmi_watchdog(void) error: if (nmi_watchdog == NMI_IO_APIC) { if (!timer_through_8259) - disable_8259A_irq(0); + legacy_pic->chip->mask(0); on_each_cpu(__acpi_nmi_disable, NULL, 1); } @@ -360,7 +361,7 @@ void stop_apic_nmi_watchdog(void *unused) */ static DEFINE_PER_CPU(unsigned, last_irq_sum); -static DEFINE_PER_CPU(local_t, alert_counter); +static DEFINE_PER_CPU(long, alert_counter); static DEFINE_PER_CPU(int, nmi_touch); void touch_nmi_watchdog(void) @@ -414,15 +415,15 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) } /* We can be called before check_nmi_watchdog, hence NULL check. */ - if (cpumask_test_cpu(cpu, &backtrace_mask)) { - static DEFINE_SPINLOCK(lock); /* Serialise the printks */ + if (cpumask_test_cpu(cpu, to_cpumask(backtrace_mask))) { + static DEFINE_RAW_SPINLOCK(lock); /* Serialise the printks */ - spin_lock(&lock); + raw_spin_lock(&lock); printk(KERN_WARNING "NMI backtrace for cpu %d\n", cpu); show_regs(regs); dump_stack(); - spin_unlock(&lock); - cpumask_clear_cpu(cpu, &backtrace_mask); + raw_spin_unlock(&lock); + cpumask_clear_cpu(cpu, to_cpumask(backtrace_mask)); rc = 1; } @@ -437,8 +438,8 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) * Ayiee, looks like this CPU is stuck ... * wait a few IRQs (5 seconds) before doing the oops ... */ - local_inc(&__get_cpu_var(alert_counter)); - if (local_read(&__get_cpu_var(alert_counter)) == 5 * nmi_hz) + __this_cpu_inc(alert_counter); + if (__this_cpu_read(alert_counter) == 5 * nmi_hz) /* * die_nmi will return ONLY if NOTIFY_STOP happens.. */ @@ -446,7 +447,7 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) regs, panic_on_timeout); } else { __get_cpu_var(last_irq_sum) = sum; - local_set(&__get_cpu_var(alert_counter), 0); + __this_cpu_write(alert_counter, 0); } /* see if the nmi watchdog went off */ @@ -558,14 +559,14 @@ void arch_trigger_all_cpu_backtrace(void) { int i; - cpumask_copy(&backtrace_mask, cpu_online_mask); + cpumask_copy(to_cpumask(backtrace_mask), cpu_online_mask); printk(KERN_INFO "sending NMI to all CPUs:\n"); apic->send_IPI_all(NMI_VECTOR); /* Wait for up to 10 seconds for all CPUs to do the backtrace */ for (i = 0; i < 10 * 1000; i++) { - if (cpumask_empty(&backtrace_mask)) + if (cpumask_empty(to_cpumask(backtrace_mask))) break; mdelay(1); } diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c index efa00e2b8505..3e28401f161c 100644 --- a/arch/x86/kernel/apic/numaq_32.c +++ b/arch/x86/kernel/apic/numaq_32.c @@ -225,7 +225,7 @@ static void __init smp_read_mpc_oem(struct mpc_table *mpc) mpc_record = 0; printk(KERN_INFO - "Found an OEM MPC table at %8p - parsing it ... \n", oemtable); + "Found an OEM MPC table at %8p - parsing it...\n", oemtable); if (memcmp(oemtable->signature, MPC_OEM_SIGNATURE, 4)) { printk(KERN_WARNING @@ -264,11 +264,6 @@ static void __init smp_read_mpc_oem(struct mpc_table *mpc) static __init void early_check_numaq(void) { /* - * Find possible boot-time SMP configuration: - */ - early_find_smp_config(); - - /* * get boot-time SMP configuration: */ if (smp_found_config) @@ -282,6 +277,7 @@ static __init void early_check_numaq(void) x86_init.mpparse.mpc_oem_pci_bus = mpc_oem_pci_bus; x86_init.mpparse.mpc_oem_bus_info = mpc_oem_bus_info; x86_init.timers.tsc_pre_init = numaq_tsc_init; + x86_init.pci.init = pci_numaq_init; } } @@ -334,10 +330,9 @@ static inline const struct cpumask *numaq_target_cpus(void) return cpu_all_mask; } -static inline unsigned long -numaq_check_apicid_used(physid_mask_t bitmap, int apicid) +static unsigned long numaq_check_apicid_used(physid_mask_t *map, int apicid) { - return physid_isset(apicid, bitmap); + return physid_isset(apicid, *map); } static inline unsigned long numaq_check_apicid_present(int bit) @@ -371,10 +366,10 @@ static inline int numaq_multi_timer_check(int apic, int irq) return apic != 0 && irq == 0; } -static inline physid_mask_t numaq_ioapic_phys_id_map(physid_mask_t phys_map) +static inline void numaq_ioapic_phys_id_map(physid_mask_t *phys_map, physid_mask_t *retmap) { /* We don't have a good way to do this yet - hack */ - return physids_promote(0xFUL); + return physids_promote(0xFUL, retmap); } static inline int numaq_cpu_to_logical_apicid(int cpu) @@ -402,12 +397,12 @@ static inline int numaq_apicid_to_node(int logical_apicid) return logical_apicid >> 4; } -static inline physid_mask_t numaq_apicid_to_cpu_present(int logical_apicid) +static void numaq_apicid_to_cpu_present(int logical_apicid, physid_mask_t *retmap) { int node = numaq_apicid_to_node(logical_apicid); int cpu = __ffs(logical_apicid & 0xf); - return physid_mask_of_physid(cpu + 4*node); + physid_set_mask_of_physid(cpu + 4*node, retmap); } /* Where the IO area was mapped on multiquad, always 0 otherwise */ diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c index 0c0182cc947d..99d2fe016084 100644 --- a/arch/x86/kernel/apic/probe_32.c +++ b/arch/x86/kernel/apic/probe_32.c @@ -52,7 +52,32 @@ static int __init print_ipi_mode(void) } late_initcall(print_ipi_mode); -void default_setup_apic_routing(void) +void __init default_setup_apic_routing(void) +{ + int version = apic_version[boot_cpu_physical_apicid]; + + if (num_possible_cpus() > 8) { + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_INTEL: + if (!APIC_XAPIC(version)) { + def_to_bigsmp = 0; + break; + } + /* If P4 and above fall through */ + case X86_VENDOR_AMD: + def_to_bigsmp = 1; + } + } + +#ifdef CONFIG_X86_BIGSMP + generic_bigsmp_probe(); +#endif + + if (apic->setup_apic_routing) + apic->setup_apic_routing(); +} + +static void setup_apic_flat_routing(void) { #ifdef CONFIG_X86_IO_APIC printk(KERN_INFO @@ -103,12 +128,12 @@ struct apic apic_default = { .init_apic_ldr = default_init_apic_ldr, .ioapic_phys_id_map = default_ioapic_phys_id_map, - .setup_apic_routing = default_setup_apic_routing, + .setup_apic_routing = setup_apic_flat_routing, .multi_timer_check = NULL, .apicid_to_node = default_apicid_to_node, .cpu_to_logical_apicid = default_cpu_to_logical_apicid, .cpu_present_to_apicid = default_cpu_present_to_apicid, - .apicid_to_cpu_present = default_apicid_to_cpu_present, + .apicid_to_cpu_present = physid_set_mask_of_physid, .setup_portio_remap = NULL, .check_phys_apicid_present = default_check_phys_apicid_present, .enable_apic_mode = NULL, diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c index c4cbd3080c1c..83e9be4778e2 100644 --- a/arch/x86/kernel/apic/probe_64.c +++ b/arch/x86/kernel/apic/probe_64.c @@ -67,17 +67,8 @@ void __init default_setup_apic_routing(void) } #endif - if (apic == &apic_flat) { - switch (boot_cpu_data.x86_vendor) { - case X86_VENDOR_INTEL: - if (num_processors > 8) - apic = &apic_physflat; - break; - case X86_VENDOR_AMD: - if (max_physical_apicid >= 8) - apic = &apic_physflat; - } - } + if (apic == &apic_flat && num_possible_cpus() > 8) + apic = &apic_physflat; printk(KERN_INFO "Setting APIC routing to %s\n", apic->name); diff --git a/arch/x86/kernel/apic/summit_32.c b/arch/x86/kernel/apic/summit_32.c index 645ecc4ff0be..9b419263d90d 100644 --- a/arch/x86/kernel/apic/summit_32.c +++ b/arch/x86/kernel/apic/summit_32.c @@ -183,7 +183,7 @@ static const struct cpumask *summit_target_cpus(void) return cpumask_of(0); } -static unsigned long summit_check_apicid_used(physid_mask_t bitmap, int apicid) +static unsigned long summit_check_apicid_used(physid_mask_t *map, int apicid) { return 0; } @@ -261,15 +261,15 @@ static int summit_cpu_present_to_apicid(int mps_cpu) return BAD_APICID; } -static physid_mask_t summit_ioapic_phys_id_map(physid_mask_t phys_id_map) +static void summit_ioapic_phys_id_map(physid_mask_t *phys_id_map, physid_mask_t *retmap) { /* For clustered we don't have a good way to do this yet - hack */ - return physids_promote(0x0F); + physids_promote(0x0FL, retmap); } -static physid_mask_t summit_apicid_to_cpu_present(int apicid) +static void summit_apicid_to_cpu_present(int apicid, physid_mask_t *retmap) { - return physid_mask_of_physid(0); + physid_set_mask_of_physid(0, retmap); } static int summit_check_phys_apicid_present(int physical_apicid) diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c index a5371ec36776..cf69c59f4910 100644 --- a/arch/x86/kernel/apic/x2apic_cluster.c +++ b/arch/x86/kernel/apic/x2apic_cluster.c @@ -148,10 +148,7 @@ x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, break; } - if (cpu < nr_cpu_ids) - return per_cpu(x86_cpu_to_logical_apicid, cpu); - - return BAD_APICID; + return per_cpu(x86_cpu_to_logical_apicid, cpu); } static unsigned int x2apic_cluster_phys_get_apic_id(unsigned long x) diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c index a8989aadc99a..8972f38c5ced 100644 --- a/arch/x86/kernel/apic/x2apic_phys.c +++ b/arch/x86/kernel/apic/x2apic_phys.c @@ -146,10 +146,7 @@ x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, break; } - if (cpu < nr_cpu_ids) - return per_cpu(x86_cpu_to_apicid, cpu); - - return BAD_APICID; + return per_cpu(x86_cpu_to_apicid, cpu); } static unsigned int x2apic_phys_get_apic_id(unsigned long x) diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index f5f5886a6b53..49dbeaef2a27 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -5,7 +5,7 @@ * * SGI UV APIC functions (note: not an Intel compatible APIC) * - * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 2007-2009 Silicon Graphics, Inc. All rights reserved. */ #include <linux/cpumask.h> #include <linux/hardirq.h> @@ -20,6 +20,8 @@ #include <linux/cpu.h> #include <linux/init.h> #include <linux/io.h> +#include <linux/pci.h> +#include <linux/kdebug.h> #include <asm/uv/uv_mmrs.h> #include <asm/uv/uv_hub.h> @@ -30,10 +32,27 @@ #include <asm/apic.h> #include <asm/ipi.h> #include <asm/smp.h> +#include <asm/x86_init.h> DEFINE_PER_CPU(int, x2apic_extra_bits); +#define PR_DEVEL(fmt, args...) pr_devel("%s: " fmt, __func__, args) + static enum uv_system_type uv_system_type; +static u64 gru_start_paddr, gru_end_paddr; +int uv_min_hub_revision_id; +EXPORT_SYMBOL_GPL(uv_min_hub_revision_id); +static DEFINE_SPINLOCK(uv_nmi_lock); + +static inline bool is_GRU_range(u64 start, u64 end) +{ + return start >= gru_start_paddr && end <= gru_end_paddr; +} + +static bool uv_is_untracked_pat_range(u64 start, u64 end) +{ + return is_ISA_range(start, end) || is_GRU_range(start, end); +} static int early_get_nodeid(void) { @@ -43,19 +62,28 @@ static int early_get_nodeid(void) mmr = early_ioremap(UV_LOCAL_MMR_BASE | UVH_NODE_ID, sizeof(*mmr)); node_id.v = *mmr; early_iounmap(mmr, sizeof(*mmr)); + + /* Currently, all blades have same revision number */ + uv_min_hub_revision_id = node_id.s.revision; + return node_id.s.node_id; } static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { + int nodeid; + if (!strcmp(oem_id, "SGI")) { + nodeid = early_get_nodeid(); + x86_platform.is_untracked_pat_range = uv_is_untracked_pat_range; + x86_platform.nmi_init = uv_nmi_init; if (!strcmp(oem_table_id, "UVL")) uv_system_type = UV_LEGACY_APIC; else if (!strcmp(oem_table_id, "UVX")) uv_system_type = UV_X2APIC; else if (!strcmp(oem_table_id, "UVH")) { __get_cpu_var(x2apic_extra_bits) = - early_get_nodeid() << (UV_APIC_PNODE_SHIFT - 1); + nodeid << (UV_APIC_PNODE_SHIFT - 1); uv_system_type = UV_NON_UNIQUE_APIC; return 1; } @@ -92,11 +120,9 @@ EXPORT_SYMBOL_GPL(uv_possible_blades); unsigned long sn_rtc_cycles_per_second; EXPORT_SYMBOL(sn_rtc_cycles_per_second); -/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ - static const struct cpumask *uv_target_cpus(void) { - return cpumask_of(0); + return cpu_online_mask; } static void uv_vector_allocation_domain(int cpu, struct cpumask *retmask) @@ -212,10 +238,7 @@ uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask, if (cpumask_test_cpu(cpu, cpu_online_mask)) break; } - if (cpu < nr_cpu_ids) - return per_cpu(x86_cpu_to_apicid, cpu); - - return BAD_APICID; + return per_cpu(x86_cpu_to_apicid, cpu); } static unsigned int x2apic_get_apic_id(unsigned long x) @@ -352,25 +375,25 @@ static __init void get_lowmem_redirect(unsigned long *base, unsigned long *size) for (i = 0; i < ARRAY_SIZE(redir_addrs); i++) { alias.v = uv_read_local_mmr(redir_addrs[i].alias); - if (alias.s.base == 0) { + if (alias.s.enable && alias.s.base == 0) { *size = (1UL << alias.s.m_alias); redirect.v = uv_read_local_mmr(redir_addrs[i].redirect); *base = (unsigned long)redirect.s.dest_base << DEST_SHIFT; return; } } - BUG(); + *base = *size = 0; } enum map_type {map_wb, map_uc}; -static __init void map_high(char *id, unsigned long base, int shift, - int max_pnode, enum map_type map_type) +static __init void map_high(char *id, unsigned long base, int pshift, + int bshift, int max_pnode, enum map_type map_type) { unsigned long bytes, paddr; - paddr = base << shift; - bytes = (1UL << shift) * (max_pnode + 1); + paddr = base << pshift; + bytes = (1UL << bshift) * (max_pnode + 1); printk(KERN_INFO "UV: Map %s_HI 0x%lx - 0x%lx\n", id, paddr, paddr + bytes); if (map_type == map_uc) @@ -385,8 +408,12 @@ static __init void map_gru_high(int max_pnode) int shift = UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_SHFT; gru.v = uv_read_local_mmr(UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR); - if (gru.s.enable) - map_high("GRU", gru.s.base, shift, max_pnode, map_wb); + if (gru.s.enable) { + map_high("GRU", gru.s.base, shift, shift, max_pnode, map_wb); + gru_start_paddr = ((u64)gru.s.base << shift); + gru_end_paddr = gru_start_paddr + (1UL << shift) * (max_pnode + 1); + + } } static __init void map_mmr_high(int max_pnode) @@ -396,7 +423,7 @@ static __init void map_mmr_high(int max_pnode) mmr.v = uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR); if (mmr.s.enable) - map_high("MMR", mmr.s.base, shift, max_pnode, map_uc); + map_high("MMR", mmr.s.base, shift, shift, max_pnode, map_uc); } static __init void map_mmioh_high(int max_pnode) @@ -406,7 +433,14 @@ static __init void map_mmioh_high(int max_pnode) mmioh.v = uv_read_local_mmr(UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR); if (mmioh.s.enable) - map_high("MMIOH", mmioh.s.base, shift, max_pnode, map_uc); + map_high("MMIOH", mmioh.s.base, shift, mmioh.s.m_io, + max_pnode, map_uc); +} + +static __init void map_low_mmrs(void) +{ + init_extra_mapping_uc(UV_GLOBAL_MMR32_BASE, UV_GLOBAL_MMR32_SIZE); + init_extra_mapping_uc(UV_LOCAL_MMR_BASE, UV_LOCAL_MMR_SIZE); } static __init void uv_rtc_init(void) @@ -452,7 +486,7 @@ static void uv_heartbeat(unsigned long ignored) static void __cpuinit uv_heartbeat_enable(int cpu) { - if (!uv_cpu_hub_info(cpu)->scir.enabled) { + while (!uv_cpu_hub_info(cpu)->scir.enabled) { struct timer_list *timer = &uv_cpu_hub_info(cpu)->scir.timer; uv_set_cpu_scir_bits(cpu, SCIR_CPU_HEARTBEAT|SCIR_CPU_ACTIVITY); @@ -460,11 +494,10 @@ static void __cpuinit uv_heartbeat_enable(int cpu) timer->expires = jiffies + SCIR_CPU_HB_INTERVAL; add_timer_on(timer, cpu); uv_cpu_hub_info(cpu)->scir.enabled = 1; - } - /* check boot cpu */ - if (!uv_cpu_hub_info(0)->scir.enabled) - uv_heartbeat_enable(0); + /* also ensure that boot cpu is enabled */ + cpu = 0; + } } #ifdef CONFIG_HOTPLUG_CPU @@ -523,6 +556,30 @@ late_initcall(uv_init_heartbeat); #endif /* !CONFIG_HOTPLUG_CPU */ +/* Direct Legacy VGA I/O traffic to designated IOH */ +int uv_set_vga_state(struct pci_dev *pdev, bool decode, + unsigned int command_bits, bool change_bridge) +{ + int domain, bus, rc; + + PR_DEVEL("devfn %x decode %d cmd %x chg_brdg %d\n", + pdev->devfn, decode, command_bits, change_bridge); + + if (!change_bridge) + return 0; + + if ((command_bits & PCI_COMMAND_IO) == 0) + return 0; + + domain = pci_domain_nr(pdev->bus); + bus = pdev->bus->number; + + rc = uv_bios_set_legacy_vga_target(decode, domain, bus); + PR_DEVEL("vga decode %d %x:%x, rc: %d\n", decode, domain, bus, rc); + + return rc; +} + /* * Called on each cpu to initialize the per_cpu UV data area. * FIXME: hotplug not supported yet @@ -539,6 +596,46 @@ void __cpuinit uv_cpu_init(void) set_x2apic_extra_bits(uv_hub_info->pnode); } +/* + * When NMI is received, print a stack trace. + */ +int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data) +{ + if (reason != DIE_NMI_IPI) + return NOTIFY_OK; + /* + * Use a lock so only one cpu prints at a time + * to prevent intermixed output. + */ + spin_lock(&uv_nmi_lock); + pr_info("NMI stack dump cpu %u:\n", smp_processor_id()); + dump_stack(); + spin_unlock(&uv_nmi_lock); + + return NOTIFY_STOP; +} + +static struct notifier_block uv_dump_stack_nmi_nb = { + .notifier_call = uv_handle_nmi +}; + +void uv_register_nmi_notifier(void) +{ + if (register_die_notifier(&uv_dump_stack_nmi_nb)) + printk(KERN_WARNING "UV NMI handler failed to register\n"); +} + +void uv_nmi_init(void) +{ + unsigned int value; + + /* + * Unmask NMI on all cpus + */ + value = apic_read(APIC_LVT1) | APIC_DM_NMI; + value &= ~APIC_LVT_MASKED; + apic_write(APIC_LVT1, value); +} void __init uv_system_init(void) { @@ -550,6 +647,8 @@ void __init uv_system_init(void) unsigned long mmr_base, present, paddr; unsigned short pnode_mask; + map_low_mmrs(); + m_n_config.v = uv_read_local_mmr(UVH_SI_ADDR_MAP_CONFIG); m_val = m_n_config.s.m_skt; n_val = m_n_config.s.n_skt; @@ -602,13 +701,15 @@ void __init uv_system_init(void) } uv_bios_init(); - uv_bios_get_sn_info(0, &uv_type, &sn_partition_id, - &sn_coherency_id, &sn_region_size); + uv_bios_get_sn_info(0, &uv_type, &sn_partition_id, &sn_coherency_id, + &sn_region_size, &system_serial_number); uv_rtc_init(); for_each_present_cpu(cpu) { + int apicid = per_cpu(x86_cpu_to_apicid, cpu); + nid = cpu_to_node(cpu); - pnode = uv_apicid_to_pnode(per_cpu(x86_cpu_to_apicid, cpu)); + pnode = uv_apicid_to_pnode(apicid); blade = boot_pnode_to_blade(pnode); lcpu = uv_blade_info[blade].nr_possible_cpus; uv_blade_info[blade].nr_possible_cpus++; @@ -619,25 +720,23 @@ void __init uv_system_init(void) uv_cpu_hub_info(cpu)->lowmem_remap_base = lowmem_redir_base; uv_cpu_hub_info(cpu)->lowmem_remap_top = lowmem_redir_size; uv_cpu_hub_info(cpu)->m_val = m_val; - uv_cpu_hub_info(cpu)->n_val = m_val; + uv_cpu_hub_info(cpu)->n_val = n_val; uv_cpu_hub_info(cpu)->numa_blade_id = blade; uv_cpu_hub_info(cpu)->blade_processor_id = lcpu; uv_cpu_hub_info(cpu)->pnode = pnode; uv_cpu_hub_info(cpu)->pnode_mask = pnode_mask; - uv_cpu_hub_info(cpu)->gpa_mask = (1 << (m_val + n_val)) - 1; + uv_cpu_hub_info(cpu)->gpa_mask = (1UL << (m_val + n_val)) - 1; uv_cpu_hub_info(cpu)->gnode_upper = gnode_upper; uv_cpu_hub_info(cpu)->gnode_extra = gnode_extra; uv_cpu_hub_info(cpu)->global_mmr_base = mmr_base; uv_cpu_hub_info(cpu)->coherency_domain_number = sn_coherency_id; - uv_cpu_hub_info(cpu)->scir.offset = SCIR_LOCAL_MMR_BASE + lcpu; + 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); - printk(KERN_DEBUG "UV: cpu %d, apicid 0x%x, pnode %d, nid %d, " - "lcpu %d, blade %d\n", - cpu, per_cpu(x86_cpu_to_apicid, cpu), pnode, nid, - lcpu, blade); + printk(KERN_DEBUG "UV: cpu %d, apicid 0x%x, pnode %d, nid %d, lcpu %d, blade %d\n", + cpu, apicid, pnode, nid, lcpu, blade); } /* Add blade/pnode info for nodes without cpus */ @@ -658,5 +757,9 @@ void __init uv_system_init(void) uv_cpu_init(); uv_scir_register_cpu_notifier(); + uv_register_nmi_notifier(); proc_mkdir("sgi_uv", NULL); + + /* register Legacy VGA I/O redirection handler */ + pci_register_set_vga_state(uv_set_vga_state); } diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 151ace69a5aa..031aa887b0eb 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -204,7 +204,6 @@ #include <linux/module.h> #include <linux/poll.h> -#include <linux/smp_lock.h> #include <linux/types.h> #include <linux/stddef.h> #include <linux/timer.h> @@ -403,6 +402,7 @@ static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue); static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue); static struct apm_user *user_list; static DEFINE_SPINLOCK(user_list_lock); +static DEFINE_MUTEX(apm_mutex); /* * Set up a segment that references the real mode segment 0x40 @@ -1531,7 +1531,7 @@ static long do_ioctl(struct file *filp, u_int cmd, u_long arg) return -EPERM; switch (cmd) { case APM_IOC_STANDBY: - lock_kernel(); + mutex_lock(&apm_mutex); if (as->standbys_read > 0) { as->standbys_read--; as->standbys_pending--; @@ -1540,10 +1540,10 @@ static long do_ioctl(struct file *filp, u_int cmd, u_long arg) queue_event(APM_USER_STANDBY, as); if (standbys_pending <= 0) standby(); - unlock_kernel(); + mutex_unlock(&apm_mutex); break; case APM_IOC_SUSPEND: - lock_kernel(); + mutex_lock(&apm_mutex); if (as->suspends_read > 0) { as->suspends_read--; as->suspends_pending--; @@ -1552,13 +1552,14 @@ static long do_ioctl(struct file *filp, u_int cmd, u_long arg) queue_event(APM_USER_SUSPEND, as); if (suspends_pending <= 0) { ret = suspend(1); + mutex_unlock(&apm_mutex); } else { as->suspend_wait = 1; + mutex_unlock(&apm_mutex); wait_event_interruptible(apm_suspend_waitqueue, as->suspend_wait == 0); ret = as->suspend_result; } - unlock_kernel(); return ret; default: return -ENOTTY; @@ -1608,12 +1609,10 @@ static int do_open(struct inode *inode, struct file *filp) { struct apm_user *as; - lock_kernel(); as = kmalloc(sizeof(*as), GFP_KERNEL); if (as == NULL) { printk(KERN_ERR "apm: cannot allocate struct of size %d bytes\n", sizeof(*as)); - unlock_kernel(); return -ENOMEM; } as->magic = APM_BIOS_MAGIC; @@ -1635,7 +1634,6 @@ static int do_open(struct inode *inode, struct file *filp) user_list = as; spin_unlock(&user_list_lock); filp->private_data = as; - unlock_kernel(); return 0; } @@ -1994,8 +1992,8 @@ static int __init apm_is_horked_d850md(const struct dmi_system_id *d) apm_info.disabled = 1; printk(KERN_INFO "%s machine detected. " "Disabling APM.\n", d->ident); - printk(KERN_INFO "This bug is fixed in bios P15 which is available for \n"); - printk(KERN_INFO "download from support.intel.com \n"); + printk(KERN_INFO "This bug is fixed in bios P15 which is available for\n"); + printk(KERN_INFO "download from support.intel.com\n"); } return 0; } diff --git a/arch/x86/kernel/bios_uv.c b/arch/x86/kernel/bios_uv.c index 63a88e1f987d..8bc57baaa9ad 100644 --- a/arch/x86/kernel/bios_uv.c +++ b/arch/x86/kernel/bios_uv.c @@ -15,8 +15,8 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA * - * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved. - * Copyright (c) Russ Anderson + * Copyright (c) 2008-2009 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) Russ Anderson <rja@sgi.com> */ #include <linux/efi.h> @@ -30,6 +30,7 @@ static struct uv_systab uv_systab; s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5) { struct uv_systab *tab = &uv_systab; + s64 ret; if (!tab->function) /* @@ -37,9 +38,11 @@ s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5) */ return BIOS_STATUS_UNIMPLEMENTED; - return efi_call6((void *)__va(tab->function), - (u64)which, a1, a2, a3, a4, a5); + ret = efi_call6((void *)__va(tab->function), (u64)which, + a1, a2, a3, a4, a5); + return ret; } +EXPORT_SYMBOL_GPL(uv_bios_call); s64 uv_bios_call_irqsave(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5) @@ -73,11 +76,14 @@ long sn_coherency_id; EXPORT_SYMBOL_GPL(sn_coherency_id); long sn_region_size; EXPORT_SYMBOL_GPL(sn_region_size); +long system_serial_number; +EXPORT_SYMBOL_GPL(system_serial_number); int uv_type; +EXPORT_SYMBOL_GPL(uv_type); s64 uv_bios_get_sn_info(int fc, int *uvtype, long *partid, long *coher, - long *region) + long *region, long *ssn) { s64 ret; u64 v0, v1; @@ -97,25 +103,24 @@ s64 uv_bios_get_sn_info(int fc, int *uvtype, long *partid, long *coher, *coher = part.coherence_id; if (region) *region = part.region_size; + if (ssn) + *ssn = v1; return ret; } +EXPORT_SYMBOL_GPL(uv_bios_get_sn_info); int -uv_bios_mq_watchlist_alloc(int blade, unsigned long addr, unsigned int mq_size, +uv_bios_mq_watchlist_alloc(unsigned long addr, unsigned int mq_size, unsigned long *intr_mmr_offset) { - union uv_watchlist_u size_blade; u64 watchlist; s64 ret; - size_blade.size = mq_size; - size_blade.blade = blade; - /* * bios returns watchlist number or negative error number. */ ret = (int)uv_bios_call_irqsave(UV_BIOS_WATCHLIST_ALLOC, addr, - size_blade.val, (u64)intr_mmr_offset, + mq_size, (u64)intr_mmr_offset, (u64)&watchlist, 0); if (ret < BIOS_STATUS_SUCCESS) return ret; @@ -158,6 +163,25 @@ s64 uv_bios_freq_base(u64 clock_type, u64 *ticks_per_second) } EXPORT_SYMBOL_GPL(uv_bios_freq_base); +/* + * uv_bios_set_legacy_vga_target - Set Legacy VGA I/O Target + * @decode: true to enable target, false to disable target + * @domain: PCI domain number + * @bus: PCI bus number + * + * Returns: + * 0: Success + * -EINVAL: Invalid domain or bus number + * -ENOSYS: Capability not available + * -EBUSY: Legacy VGA I/O cannot be retargeted at this time + */ +int uv_bios_set_legacy_vga_target(bool decode, int domain, int bus) +{ + return uv_bios_call(UV_BIOS_SET_LEGACY_VGA_TARGET, + (u64)decode, (u64)domain, (u64)bus, 0, 0); +} +EXPORT_SYMBOL_GPL(uv_bios_set_legacy_vga_target); + #ifdef CONFIG_EFI void uv_bios_init(void) @@ -189,4 +213,3 @@ void uv_bios_init(void) void uv_bios_init(void) { } #endif - diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 68537e957a9b..c202b62f3671 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -5,6 +5,7 @@ # Don't trace early stages of a secondary CPU boot ifdef CONFIG_FUNCTION_TRACER CFLAGS_REMOVE_common.o = -pg +CFLAGS_REMOVE_perf_event.o = -pg endif # Make sure load_percpu_segment has no stackprotector @@ -18,8 +19,6 @@ obj-y += vmware.o hypervisor.o sched.o obj-$(CONFIG_X86_32) += bugs.o cmpxchg.o obj-$(CONFIG_X86_64) += bugs_64.o -obj-$(CONFIG_X86_CPU_DEBUG) += cpu_debug.o - obj-$(CONFIG_CPU_SUP_INTEL) += intel.o obj-$(CONFIG_CPU_SUP_AMD) += amd.o obj-$(CONFIG_CPU_SUP_CYRIX_32) += cyrix.o diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c index c965e5212714..97ad79cdf688 100644 --- a/arch/x86/kernel/cpu/addon_cpuid_features.c +++ b/arch/x86/kernel/cpu/addon_cpuid_features.c @@ -32,6 +32,10 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) static const struct cpuid_bit __cpuinitconst cpuid_bits[] = { { X86_FEATURE_IDA, CR_EAX, 1, 0x00000006 }, { X86_FEATURE_ARAT, CR_EAX, 2, 0x00000006 }, + { X86_FEATURE_NPT, CR_EDX, 0, 0x8000000a }, + { X86_FEATURE_LBRV, CR_EDX, 1, 0x8000000a }, + { X86_FEATURE_SVML, CR_EDX, 2, 0x8000000a }, + { X86_FEATURE_NRIPS, CR_EDX, 3, 0x8000000a }, { 0, 0, 0, 0 } }; @@ -74,6 +78,7 @@ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c) unsigned int eax, ebx, ecx, edx, sub_index; unsigned int ht_mask_width, core_plus_mask_width; unsigned int core_select_mask, core_level_siblings; + static bool printed; if (c->cpuid_level < 0xb) return; @@ -127,12 +132,14 @@ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c) c->x86_max_cores = (core_level_siblings / smp_num_siblings); - - printk(KERN_INFO "CPU: Physical Processor ID: %d\n", - c->phys_proc_id); - if (c->x86_max_cores > 1) - printk(KERN_INFO "CPU: Processor Core ID: %d\n", - c->cpu_core_id); + if (!printed) { + printk(KERN_INFO "CPU: Physical Processor ID: %d\n", + c->phys_proc_id); + if (c->x86_max_cores > 1) + printk(KERN_INFO "CPU: Processor Core ID: %d\n", + c->cpu_core_id); + printed = 1; + } return; #endif } diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index c910a716a71c..e485825130d2 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -254,59 +254,36 @@ static int __cpuinit nearby_node(int apicid) /* * Fixup core topology information for AMD multi-node processors. - * Assumption 1: Number of cores in each internal node is the same. - * Assumption 2: Mixed systems with both single-node and dual-node - * processors are not supported. + * Assumption: Number of cores in each internal node is the same. */ #ifdef CONFIG_X86_HT static void __cpuinit amd_fixup_dcm(struct cpuinfo_x86 *c) { -#ifdef CONFIG_PCI - u32 t, cpn; - u8 n, n_id; + unsigned long long value; + u32 nodes, cores_per_node; int cpu = smp_processor_id(); + if (!cpu_has(c, X86_FEATURE_NODEID_MSR)) + return; + /* fixup topology information only once for a core */ if (cpu_has(c, X86_FEATURE_AMD_DCM)) return; - /* check for multi-node processor on boot cpu */ - t = read_pci_config(0, 24, 3, 0xe8); - if (!(t & (1 << 29))) + rdmsrl(MSR_FAM10H_NODE_ID, value); + + nodes = ((value >> 3) & 7) + 1; + if (nodes == 1) return; set_cpu_cap(c, X86_FEATURE_AMD_DCM); + cores_per_node = c->x86_max_cores / nodes; - /* cores per node: each internal node has half the number of cores */ - cpn = c->x86_max_cores >> 1; - - /* even-numbered NB_id of this dual-node processor */ - n = c->phys_proc_id << 1; - - /* - * determine internal node id and assign cores fifty-fifty to - * each node of the dual-node processor - */ - t = read_pci_config(0, 24 + n, 3, 0xe8); - n = (t>>30) & 0x3; - if (n == 0) { - if (c->cpu_core_id < cpn) - n_id = 0; - else - n_id = 1; - } else { - if (c->cpu_core_id < cpn) - n_id = 1; - else - n_id = 0; - } - - /* compute entire NodeID, use llc_shared_map to store sibling info */ - per_cpu(cpu_llc_id, cpu) = (c->phys_proc_id << 1) + n_id; + /* store NodeID, use llc_shared_map to store sibling info */ + per_cpu(cpu_llc_id, cpu) = value & 7; - /* fixup core id to be in range from 0 to cpn */ - c->cpu_core_id = c->cpu_core_id % cpn; -#endif + /* 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; } #endif @@ -375,8 +352,6 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) node = nearby_node(apicid); } numa_set_node(cpu, node); - - printk(KERN_INFO "CPU %d/0x%x -> Node %d\n", cpu, apicid, node); #endif } @@ -535,7 +510,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) } } - display_cacheinfo(c); + cpu_detect_cache_sizes(c); /* Multi core CPU? */ if (c->extended_cpuid_level >= 0x80000008) { diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c index c95e831bb095..e58d978e0758 100644 --- a/arch/x86/kernel/cpu/centaur.c +++ b/arch/x86/kernel/cpu/centaur.c @@ -294,7 +294,7 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_REP_GOOD); } - display_cacheinfo(c); + cpu_detect_cache_sizes(c); } enum { diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index cc25c2b4a567..4868e4a951ee 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -61,7 +61,7 @@ void __init setup_cpu_local_masks(void) static void __cpuinit default_init(struct cpuinfo_x86 *c) { #ifdef CONFIG_X86_64 - display_cacheinfo(c); + cpu_detect_cache_sizes(c); #else /* Not much we can do here... */ /* Check if at least it has cpuid */ @@ -383,7 +383,7 @@ static void __cpuinit get_model_name(struct cpuinfo_x86 *c) } } -void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) +void __cpuinit cpu_detect_cache_sizes(struct cpuinfo_x86 *c) { unsigned int n, dummy, ebx, ecx, edx, l2size; @@ -391,8 +391,6 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) if (n >= 0x80000005) { cpuid(0x80000005, &dummy, &ebx, &ecx, &edx); - printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n", - edx>>24, edx&0xFF, ecx>>24, ecx&0xFF); c->x86_cache_size = (ecx>>24) + (edx>>24); #ifdef CONFIG_X86_64 /* On K8 L1 TLB is inclusive, so don't count it */ @@ -422,9 +420,6 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) #endif c->x86_cache_size = l2size; - - printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n", - l2size, ecx & 0xFF); } void __cpuinit detect_ht(struct cpuinfo_x86 *c) @@ -432,6 +427,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) #ifdef CONFIG_X86_HT u32 eax, ebx, ecx, edx; int index_msb, core_bits; + static bool printed; if (!cpu_has(c, X86_FEATURE_HT)) return; @@ -447,7 +443,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) smp_num_siblings = (ebx & 0xff0000) >> 16; if (smp_num_siblings == 1) { - printk(KERN_INFO "CPU: Hyper-Threading is disabled\n"); + printk_once(KERN_INFO "CPU0: Hyper-Threading is disabled\n"); goto out; } @@ -474,11 +470,12 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) ((1 << core_bits) - 1); out: - if ((c->x86_max_cores * smp_num_siblings) > 1) { + if (!printed && (c->x86_max_cores * smp_num_siblings) > 1) { printk(KERN_INFO "CPU: Physical Processor ID: %d\n", c->phys_proc_id); printk(KERN_INFO "CPU: Processor Core ID: %d\n", c->cpu_core_id); + printed = 1; } #endif } @@ -659,24 +656,31 @@ void __init early_cpu_init(void) const struct cpu_dev *const *cdev; int count = 0; +#ifdef PROCESSOR_SELECT printk(KERN_INFO "KERNEL supported cpus:\n"); +#endif + for (cdev = __x86_cpu_dev_start; cdev < __x86_cpu_dev_end; cdev++) { const struct cpu_dev *cpudev = *cdev; - unsigned int j; if (count >= X86_VENDOR_NUM) break; cpu_devs[count] = cpudev; count++; - for (j = 0; j < 2; j++) { - if (!cpudev->c_ident[j]) - continue; - printk(KERN_INFO " %s %s\n", cpudev->c_vendor, - cpudev->c_ident[j]); +#ifdef PROCESSOR_SELECT + { + unsigned int j; + + for (j = 0; j < 2; j++) { + if (!cpudev->c_ident[j]) + continue; + printk(KERN_INFO " %s %s\n", cpudev->c_vendor, + cpudev->c_ident[j]); + } } +#endif } - early_identify_cpu(&boot_cpu_data); } @@ -837,10 +841,8 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) boot_cpu_data.x86_capability[i] &= c->x86_capability[i]; } -#ifdef CONFIG_X86_MCE /* Init Machine Check Exception if available. */ - mcheck_init(c); -#endif + mcheck_cpu_init(c); select_idle_routine(c); @@ -1093,7 +1095,7 @@ static void clear_all_debug_regs(void) void __cpuinit cpu_init(void) { - struct orig_ist *orig_ist; + struct orig_ist *oist; struct task_struct *me; struct tss_struct *t; unsigned long v; @@ -1102,7 +1104,7 @@ void __cpuinit cpu_init(void) cpu = stack_smp_processor_id(); t = &per_cpu(init_tss, cpu); - orig_ist = &per_cpu(orig_ist, cpu); + oist = &per_cpu(orig_ist, cpu); #ifdef CONFIG_NUMA if (cpu != 0 && percpu_read(node_number) == 0 && @@ -1115,7 +1117,7 @@ void __cpuinit cpu_init(void) if (cpumask_test_and_set_cpu(cpu, cpu_initialized_mask)) panic("CPU#%d already initialized!\n", cpu); - printk(KERN_INFO "Initializing CPU#%d\n", cpu); + pr_debug("Initializing CPU#%d\n", cpu); clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE); @@ -1136,19 +1138,19 @@ void __cpuinit cpu_init(void) wrmsrl(MSR_KERNEL_GS_BASE, 0); barrier(); - check_efer(); + x86_configure_nx(); if (cpu != 0) enable_x2apic(); /* * set up and load the per-CPU TSS */ - if (!orig_ist->ist[0]) { + if (!oist->ist[0]) { char *estacks = per_cpu(exception_stacks, cpu); for (v = 0; v < N_EXCEPTION_STACKS; v++) { estacks += exception_stack_sizes[v]; - orig_ist->ist[v] = t->x86_tss.ist[v] = + oist->ist[v] = t->x86_tss.ist[v] = (unsigned long)estacks; } } diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h index 6de9a908e400..3624e8a0f71b 100644 --- a/arch/x86/kernel/cpu/cpu.h +++ b/arch/x86/kernel/cpu/cpu.h @@ -32,6 +32,6 @@ struct cpu_dev { extern const struct cpu_dev *const __x86_cpu_dev_start[], *const __x86_cpu_dev_end[]; -extern void display_cacheinfo(struct cpuinfo_x86 *c); +extern void cpu_detect_cache_sizes(struct cpuinfo_x86 *c); #endif diff --git a/arch/x86/kernel/cpu/cpu_debug.c b/arch/x86/kernel/cpu/cpu_debug.c deleted file mode 100644 index dca325c03999..000000000000 --- a/arch/x86/kernel/cpu/cpu_debug.c +++ /dev/null @@ -1,688 +0,0 @@ -/* - * CPU x86 architecture debug code - * - * Copyright(C) 2009 Jaswinder Singh Rajput - * - * For licencing details see kernel-base/COPYING - */ - -#include <linux/interrupt.h> -#include <linux/compiler.h> -#include <linux/seq_file.h> -#include <linux/debugfs.h> -#include <linux/kprobes.h> -#include <linux/uaccess.h> -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/percpu.h> -#include <linux/signal.h> -#include <linux/errno.h> -#include <linux/sched.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/slab.h> -#include <linux/smp.h> - -#include <asm/cpu_debug.h> -#include <asm/paravirt.h> -#include <asm/system.h> -#include <asm/traps.h> -#include <asm/apic.h> -#include <asm/desc.h> - -static DEFINE_PER_CPU(struct cpu_cpuX_base [CPU_REG_ALL_BIT], cpu_arr); -static DEFINE_PER_CPU(struct cpu_private * [MAX_CPU_FILES], priv_arr); -static DEFINE_PER_CPU(int, cpu_priv_count); - -static DEFINE_MUTEX(cpu_debug_lock); - -static struct dentry *cpu_debugfs_dir; - -static struct cpu_debug_base cpu_base[] = { - { "mc", CPU_MC, 0 }, - { "monitor", CPU_MONITOR, 0 }, - { "time", CPU_TIME, 0 }, - { "pmc", CPU_PMC, 1 }, - { "platform", CPU_PLATFORM, 0 }, - { "apic", CPU_APIC, 0 }, - { "poweron", CPU_POWERON, 0 }, - { "control", CPU_CONTROL, 0 }, - { "features", CPU_FEATURES, 0 }, - { "lastbranch", CPU_LBRANCH, 0 }, - { "bios", CPU_BIOS, 0 }, - { "freq", CPU_FREQ, 0 }, - { "mtrr", CPU_MTRR, 0 }, - { "perf", CPU_PERF, 0 }, - { "cache", CPU_CACHE, 0 }, - { "sysenter", CPU_SYSENTER, 0 }, - { "therm", CPU_THERM, 0 }, - { "misc", CPU_MISC, 0 }, - { "debug", CPU_DEBUG, 0 }, - { "pat", CPU_PAT, 0 }, - { "vmx", CPU_VMX, 0 }, - { "call", CPU_CALL, 0 }, - { "base", CPU_BASE, 0 }, - { "ver", CPU_VER, 0 }, - { "conf", CPU_CONF, 0 }, - { "smm", CPU_SMM, 0 }, - { "svm", CPU_SVM, 0 }, - { "osvm", CPU_OSVM, 0 }, - { "tss", CPU_TSS, 0 }, - { "cr", CPU_CR, 0 }, - { "dt", CPU_DT, 0 }, - { "registers", CPU_REG_ALL, 0 }, -}; - -static struct cpu_file_base cpu_file[] = { - { "index", CPU_REG_ALL, 0 }, - { "value", CPU_REG_ALL, 1 }, -}; - -/* CPU Registers Range */ -static struct cpu_debug_range cpu_reg_range[] = { - { 0x00000000, 0x00000001, CPU_MC, }, - { 0x00000006, 0x00000007, CPU_MONITOR, }, - { 0x00000010, 0x00000010, CPU_TIME, }, - { 0x00000011, 0x00000013, CPU_PMC, }, - { 0x00000017, 0x00000017, CPU_PLATFORM, }, - { 0x0000001B, 0x0000001B, CPU_APIC, }, - { 0x0000002A, 0x0000002B, CPU_POWERON, }, - { 0x0000002C, 0x0000002C, CPU_FREQ, }, - { 0x0000003A, 0x0000003A, CPU_CONTROL, }, - { 0x00000040, 0x00000047, CPU_LBRANCH, }, - { 0x00000060, 0x00000067, CPU_LBRANCH, }, - { 0x00000079, 0x00000079, CPU_BIOS, }, - { 0x00000088, 0x0000008A, CPU_CACHE, }, - { 0x0000008B, 0x0000008B, CPU_BIOS, }, - { 0x0000009B, 0x0000009B, CPU_MONITOR, }, - { 0x000000C1, 0x000000C4, CPU_PMC, }, - { 0x000000CD, 0x000000CD, CPU_FREQ, }, - { 0x000000E7, 0x000000E8, CPU_PERF, }, - { 0x000000FE, 0x000000FE, CPU_MTRR, }, - - { 0x00000116, 0x0000011E, CPU_CACHE, }, - { 0x00000174, 0x00000176, CPU_SYSENTER, }, - { 0x00000179, 0x0000017B, CPU_MC, }, - { 0x00000186, 0x00000189, CPU_PMC, }, - { 0x00000198, 0x00000199, CPU_PERF, }, - { 0x0000019A, 0x0000019A, CPU_TIME, }, - { 0x0000019B, 0x0000019D, CPU_THERM, }, - { 0x000001A0, 0x000001A0, CPU_MISC, }, - { 0x000001C9, 0x000001C9, CPU_LBRANCH, }, - { 0x000001D7, 0x000001D8, CPU_LBRANCH, }, - { 0x000001D9, 0x000001D9, CPU_DEBUG, }, - { 0x000001DA, 0x000001E0, CPU_LBRANCH, }, - - { 0x00000200, 0x0000020F, CPU_MTRR, }, - { 0x00000250, 0x00000250, CPU_MTRR, }, - { 0x00000258, 0x00000259, CPU_MTRR, }, - { 0x00000268, 0x0000026F, CPU_MTRR, }, - { 0x00000277, 0x00000277, CPU_PAT, }, - { 0x000002FF, 0x000002FF, CPU_MTRR, }, - - { 0x00000300, 0x00000311, CPU_PMC, }, - { 0x00000345, 0x00000345, CPU_PMC, }, - { 0x00000360, 0x00000371, CPU_PMC, }, - { 0x0000038D, 0x00000390, CPU_PMC, }, - { 0x000003A0, 0x000003BE, CPU_PMC, }, - { 0x000003C0, 0x000003CD, CPU_PMC, }, - { 0x000003E0, 0x000003E1, CPU_PMC, }, - { 0x000003F0, 0x000003F2, CPU_PMC, }, - - { 0x00000400, 0x00000417, CPU_MC, }, - { 0x00000480, 0x0000048B, CPU_VMX, }, - - { 0x00000600, 0x00000600, CPU_DEBUG, }, - { 0x00000680, 0x0000068F, CPU_LBRANCH, }, - { 0x000006C0, 0x000006CF, CPU_LBRANCH, }, - - { 0x000107CC, 0x000107D3, CPU_PMC, }, - - { 0xC0000080, 0xC0000080, CPU_FEATURES, }, - { 0xC0000081, 0xC0000084, CPU_CALL, }, - { 0xC0000100, 0xC0000102, CPU_BASE, }, - { 0xC0000103, 0xC0000103, CPU_TIME, }, - - { 0xC0010000, 0xC0010007, CPU_PMC, }, - { 0xC0010010, 0xC0010010, CPU_CONF, }, - { 0xC0010015, 0xC0010015, CPU_CONF, }, - { 0xC0010016, 0xC001001A, CPU_MTRR, }, - { 0xC001001D, 0xC001001D, CPU_MTRR, }, - { 0xC001001F, 0xC001001F, CPU_CONF, }, - { 0xC0010030, 0xC0010035, CPU_BIOS, }, - { 0xC0010044, 0xC0010048, CPU_MC, }, - { 0xC0010050, 0xC0010056, CPU_SMM, }, - { 0xC0010058, 0xC0010058, CPU_CONF, }, - { 0xC0010060, 0xC0010060, CPU_CACHE, }, - { 0xC0010061, 0xC0010068, CPU_SMM, }, - { 0xC0010069, 0xC001006B, CPU_SMM, }, - { 0xC0010070, 0xC0010071, CPU_SMM, }, - { 0xC0010111, 0xC0010113, CPU_SMM, }, - { 0xC0010114, 0xC0010118, CPU_SVM, }, - { 0xC0010140, 0xC0010141, CPU_OSVM, }, - { 0xC0011022, 0xC0011023, CPU_CONF, }, -}; - -static int is_typeflag_valid(unsigned cpu, unsigned flag) -{ - int i; - - /* Standard Registers should be always valid */ - if (flag >= CPU_TSS) - return 1; - - for (i = 0; i < ARRAY_SIZE(cpu_reg_range); i++) { - if (cpu_reg_range[i].flag == flag) - return 1; - } - - /* Invalid */ - return 0; -} - -static unsigned get_cpu_range(unsigned cpu, unsigned *min, unsigned *max, - int index, unsigned flag) -{ - if (cpu_reg_range[index].flag == flag) { - *min = cpu_reg_range[index].min; - *max = cpu_reg_range[index].max; - } else - *max = 0; - - return *max; -} - -/* This function can also be called with seq = NULL for printk */ -static void print_cpu_data(struct seq_file *seq, unsigned type, - u32 low, u32 high) -{ - struct cpu_private *priv; - u64 val = high; - - if (seq) { - priv = seq->private; - if (priv->file) { - val = (val << 32) | low; - seq_printf(seq, "0x%llx\n", val); - } else - seq_printf(seq, " %08x: %08x_%08x\n", - type, high, low); - } else - printk(KERN_INFO " %08x: %08x_%08x\n", type, high, low); -} - -/* This function can also be called with seq = NULL for printk */ -static void print_msr(struct seq_file *seq, unsigned cpu, unsigned flag) -{ - unsigned msr, msr_min, msr_max; - struct cpu_private *priv; - u32 low, high; - int i; - - if (seq) { - priv = seq->private; - if (priv->file) { - if (!rdmsr_safe_on_cpu(priv->cpu, priv->reg, - &low, &high)) - print_cpu_data(seq, priv->reg, low, high); - return; - } - } - - for (i = 0; i < ARRAY_SIZE(cpu_reg_range); i++) { - if (!get_cpu_range(cpu, &msr_min, &msr_max, i, flag)) - continue; - - for (msr = msr_min; msr <= msr_max; msr++) { - if (rdmsr_safe_on_cpu(cpu, msr, &low, &high)) - continue; - print_cpu_data(seq, msr, low, high); - } - } -} - -static void print_tss(void *arg) -{ - struct pt_regs *regs = task_pt_regs(current); - struct seq_file *seq = arg; - unsigned int seg; - - seq_printf(seq, " RAX\t: %016lx\n", regs->ax); - seq_printf(seq, " RBX\t: %016lx\n", regs->bx); - seq_printf(seq, " RCX\t: %016lx\n", regs->cx); - seq_printf(seq, " RDX\t: %016lx\n", regs->dx); - - seq_printf(seq, " RSI\t: %016lx\n", regs->si); - seq_printf(seq, " RDI\t: %016lx\n", regs->di); - seq_printf(seq, " RBP\t: %016lx\n", regs->bp); - seq_printf(seq, " ESP\t: %016lx\n", regs->sp); - -#ifdef CONFIG_X86_64 - seq_printf(seq, " R08\t: %016lx\n", regs->r8); - seq_printf(seq, " R09\t: %016lx\n", regs->r9); - seq_printf(seq, " R10\t: %016lx\n", regs->r10); - seq_printf(seq, " R11\t: %016lx\n", regs->r11); - seq_printf(seq, " R12\t: %016lx\n", regs->r12); - seq_printf(seq, " R13\t: %016lx\n", regs->r13); - seq_printf(seq, " R14\t: %016lx\n", regs->r14); - seq_printf(seq, " R15\t: %016lx\n", regs->r15); -#endif - - asm("movl %%cs,%0" : "=r" (seg)); - seq_printf(seq, " CS\t: %04x\n", seg); - asm("movl %%ds,%0" : "=r" (seg)); - seq_printf(seq, " DS\t: %04x\n", seg); - seq_printf(seq, " SS\t: %04lx\n", regs->ss & 0xffff); - asm("movl %%es,%0" : "=r" (seg)); - seq_printf(seq, " ES\t: %04x\n", seg); - asm("movl %%fs,%0" : "=r" (seg)); - seq_printf(seq, " FS\t: %04x\n", seg); - asm("movl %%gs,%0" : "=r" (seg)); - seq_printf(seq, " GS\t: %04x\n", seg); - - seq_printf(seq, " EFLAGS\t: %016lx\n", regs->flags); - - seq_printf(seq, " EIP\t: %016lx\n", regs->ip); -} - -static void print_cr(void *arg) -{ - struct seq_file *seq = arg; - - seq_printf(seq, " cr0\t: %016lx\n", read_cr0()); - seq_printf(seq, " cr2\t: %016lx\n", read_cr2()); - seq_printf(seq, " cr3\t: %016lx\n", read_cr3()); - seq_printf(seq, " cr4\t: %016lx\n", read_cr4_safe()); -#ifdef CONFIG_X86_64 - seq_printf(seq, " cr8\t: %016lx\n", read_cr8()); -#endif -} - -static void print_desc_ptr(char *str, struct seq_file *seq, struct desc_ptr dt) -{ - seq_printf(seq, " %s\t: %016llx\n", str, (u64)(dt.address | dt.size)); -} - -static void print_dt(void *seq) -{ - struct desc_ptr dt; - unsigned long ldt; - - /* IDT */ - store_idt((struct desc_ptr *)&dt); - print_desc_ptr("IDT", seq, dt); - - /* GDT */ - store_gdt((struct desc_ptr *)&dt); - print_desc_ptr("GDT", seq, dt); - - /* LDT */ - store_ldt(ldt); - seq_printf(seq, " LDT\t: %016lx\n", ldt); - - /* TR */ - store_tr(ldt); - seq_printf(seq, " TR\t: %016lx\n", ldt); -} - -static void print_dr(void *arg) -{ - struct seq_file *seq = arg; - unsigned long dr; - int i; - - for (i = 0; i < 8; i++) { - /* Ignore db4, db5 */ - if ((i == 4) || (i == 5)) - continue; - get_debugreg(dr, i); - seq_printf(seq, " dr%d\t: %016lx\n", i, dr); - } - - seq_printf(seq, "\n MSR\t:\n"); -} - -static void print_apic(void *arg) -{ - struct seq_file *seq = arg; - -#ifdef CONFIG_X86_LOCAL_APIC - seq_printf(seq, " LAPIC\t:\n"); - seq_printf(seq, " ID\t\t: %08x\n", apic_read(APIC_ID) >> 24); - seq_printf(seq, " LVR\t\t: %08x\n", apic_read(APIC_LVR)); - seq_printf(seq, " TASKPRI\t: %08x\n", apic_read(APIC_TASKPRI)); - seq_printf(seq, " ARBPRI\t\t: %08x\n", apic_read(APIC_ARBPRI)); - seq_printf(seq, " PROCPRI\t: %08x\n", apic_read(APIC_PROCPRI)); - seq_printf(seq, " LDR\t\t: %08x\n", apic_read(APIC_LDR)); - seq_printf(seq, " DFR\t\t: %08x\n", apic_read(APIC_DFR)); - seq_printf(seq, " SPIV\t\t: %08x\n", apic_read(APIC_SPIV)); - seq_printf(seq, " ISR\t\t: %08x\n", apic_read(APIC_ISR)); - seq_printf(seq, " ESR\t\t: %08x\n", apic_read(APIC_ESR)); - seq_printf(seq, " ICR\t\t: %08x\n", apic_read(APIC_ICR)); - seq_printf(seq, " ICR2\t\t: %08x\n", apic_read(APIC_ICR2)); - seq_printf(seq, " LVTT\t\t: %08x\n", apic_read(APIC_LVTT)); - seq_printf(seq, " LVTTHMR\t: %08x\n", apic_read(APIC_LVTTHMR)); - seq_printf(seq, " LVTPC\t\t: %08x\n", apic_read(APIC_LVTPC)); - seq_printf(seq, " LVT0\t\t: %08x\n", apic_read(APIC_LVT0)); - seq_printf(seq, " LVT1\t\t: %08x\n", apic_read(APIC_LVT1)); - seq_printf(seq, " LVTERR\t\t: %08x\n", apic_read(APIC_LVTERR)); - seq_printf(seq, " TMICT\t\t: %08x\n", apic_read(APIC_TMICT)); - seq_printf(seq, " TMCCT\t\t: %08x\n", apic_read(APIC_TMCCT)); - seq_printf(seq, " TDCR\t\t: %08x\n", apic_read(APIC_TDCR)); - if (boot_cpu_has(X86_FEATURE_EXTAPIC)) { - unsigned int i, v, maxeilvt; - - v = apic_read(APIC_EFEAT); - maxeilvt = (v >> 16) & 0xff; - seq_printf(seq, " EFEAT\t\t: %08x\n", v); - seq_printf(seq, " ECTRL\t\t: %08x\n", apic_read(APIC_ECTRL)); - - for (i = 0; i < maxeilvt; i++) { - v = apic_read(APIC_EILVTn(i)); - seq_printf(seq, " EILVT%d\t\t: %08x\n", i, v); - } - } -#endif /* CONFIG_X86_LOCAL_APIC */ - seq_printf(seq, "\n MSR\t:\n"); -} - -static int cpu_seq_show(struct seq_file *seq, void *v) -{ - struct cpu_private *priv = seq->private; - - if (priv == NULL) - return -EINVAL; - - switch (cpu_base[priv->type].flag) { - case CPU_TSS: - smp_call_function_single(priv->cpu, print_tss, seq, 1); - break; - case CPU_CR: - smp_call_function_single(priv->cpu, print_cr, seq, 1); - break; - case CPU_DT: - smp_call_function_single(priv->cpu, print_dt, seq, 1); - break; - case CPU_DEBUG: - if (priv->file == CPU_INDEX_BIT) - smp_call_function_single(priv->cpu, print_dr, seq, 1); - print_msr(seq, priv->cpu, cpu_base[priv->type].flag); - break; - case CPU_APIC: - if (priv->file == CPU_INDEX_BIT) - smp_call_function_single(priv->cpu, print_apic, seq, 1); - print_msr(seq, priv->cpu, cpu_base[priv->type].flag); - break; - - default: - print_msr(seq, priv->cpu, cpu_base[priv->type].flag); - break; - } - seq_printf(seq, "\n"); - - return 0; -} - -static void *cpu_seq_start(struct seq_file *seq, loff_t *pos) -{ - if (*pos == 0) /* One time is enough ;-) */ - return seq; - - return NULL; -} - -static void *cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos) -{ - (*pos)++; - - return cpu_seq_start(seq, pos); -} - -static void cpu_seq_stop(struct seq_file *seq, void *v) -{ -} - -static const struct seq_operations cpu_seq_ops = { - .start = cpu_seq_start, - .next = cpu_seq_next, - .stop = cpu_seq_stop, - .show = cpu_seq_show, -}; - -static int cpu_seq_open(struct inode *inode, struct file *file) -{ - struct cpu_private *priv = inode->i_private; - struct seq_file *seq; - int err; - - err = seq_open(file, &cpu_seq_ops); - if (!err) { - seq = file->private_data; - seq->private = priv; - } - - return err; -} - -static int write_msr(struct cpu_private *priv, u64 val) -{ - u32 low, high; - - high = (val >> 32) & 0xffffffff; - low = val & 0xffffffff; - - if (!wrmsr_safe_on_cpu(priv->cpu, priv->reg, low, high)) - return 0; - - return -EPERM; -} - -static int write_cpu_register(struct cpu_private *priv, const char *buf) -{ - int ret = -EPERM; - u64 val; - - ret = strict_strtoull(buf, 0, &val); - if (ret < 0) - return ret; - - /* Supporting only MSRs */ - if (priv->type < CPU_TSS_BIT) - return write_msr(priv, val); - - return ret; -} - -static ssize_t cpu_write(struct file *file, const char __user *ubuf, - size_t count, loff_t *off) -{ - struct seq_file *seq = file->private_data; - struct cpu_private *priv = seq->private; - char buf[19]; - - if ((priv == NULL) || (count >= sizeof(buf))) - return -EINVAL; - - if (copy_from_user(&buf, ubuf, count)) - return -EFAULT; - - buf[count] = 0; - - if ((cpu_base[priv->type].write) && (cpu_file[priv->file].write)) - if (!write_cpu_register(priv, buf)) - return count; - - return -EACCES; -} - -static const struct file_operations cpu_fops = { - .owner = THIS_MODULE, - .open = cpu_seq_open, - .read = seq_read, - .write = cpu_write, - .llseek = seq_lseek, - .release = seq_release, -}; - -static int cpu_create_file(unsigned cpu, unsigned type, unsigned reg, - unsigned file, struct dentry *dentry) -{ - struct cpu_private *priv = NULL; - - /* Already intialized */ - if (file == CPU_INDEX_BIT) - if (per_cpu(cpu_arr[type].init, cpu)) - return 0; - - priv = kzalloc(sizeof(*priv), GFP_KERNEL); - if (priv == NULL) - return -ENOMEM; - - priv->cpu = cpu; - priv->type = type; - priv->reg = reg; - priv->file = file; - mutex_lock(&cpu_debug_lock); - per_cpu(priv_arr[type], cpu) = priv; - per_cpu(cpu_priv_count, cpu)++; - mutex_unlock(&cpu_debug_lock); - - if (file) - debugfs_create_file(cpu_file[file].name, S_IRUGO, - dentry, (void *)priv, &cpu_fops); - else { - debugfs_create_file(cpu_base[type].name, S_IRUGO, - per_cpu(cpu_arr[type].dentry, cpu), - (void *)priv, &cpu_fops); - mutex_lock(&cpu_debug_lock); - per_cpu(cpu_arr[type].init, cpu) = 1; - mutex_unlock(&cpu_debug_lock); - } - - return 0; -} - -static int cpu_init_regfiles(unsigned cpu, unsigned int type, unsigned reg, - struct dentry *dentry) -{ - unsigned file; - int err = 0; - - for (file = 0; file < ARRAY_SIZE(cpu_file); file++) { - err = cpu_create_file(cpu, type, reg, file, dentry); - if (err) - return err; - } - - return err; -} - -static int cpu_init_msr(unsigned cpu, unsigned type, struct dentry *dentry) -{ - struct dentry *cpu_dentry = NULL; - unsigned reg, reg_min, reg_max; - int i, err = 0; - char reg_dir[12]; - u32 low, high; - - for (i = 0; i < ARRAY_SIZE(cpu_reg_range); i++) { - if (!get_cpu_range(cpu, ®_min, ®_max, i, - cpu_base[type].flag)) - continue; - - for (reg = reg_min; reg <= reg_max; reg++) { - if (rdmsr_safe_on_cpu(cpu, reg, &low, &high)) - continue; - - sprintf(reg_dir, "0x%x", reg); - cpu_dentry = debugfs_create_dir(reg_dir, dentry); - err = cpu_init_regfiles(cpu, type, reg, cpu_dentry); - if (err) - return err; - } - } - - return err; -} - -static int cpu_init_allreg(unsigned cpu, struct dentry *dentry) -{ - struct dentry *cpu_dentry = NULL; - unsigned type; - int err = 0; - - for (type = 0; type < ARRAY_SIZE(cpu_base) - 1; type++) { - if (!is_typeflag_valid(cpu, cpu_base[type].flag)) - continue; - cpu_dentry = debugfs_create_dir(cpu_base[type].name, dentry); - per_cpu(cpu_arr[type].dentry, cpu) = cpu_dentry; - - if (type < CPU_TSS_BIT) - err = cpu_init_msr(cpu, type, cpu_dentry); - else - err = cpu_create_file(cpu, type, 0, CPU_INDEX_BIT, - cpu_dentry); - if (err) - return err; - } - - return err; -} - -static int cpu_init_cpu(void) -{ - struct dentry *cpu_dentry = NULL; - struct cpuinfo_x86 *cpui; - char cpu_dir[12]; - unsigned cpu; - int err = 0; - - for (cpu = 0; cpu < nr_cpu_ids; cpu++) { - cpui = &cpu_data(cpu); - if (!cpu_has(cpui, X86_FEATURE_MSR)) - continue; - - sprintf(cpu_dir, "cpu%d", cpu); - cpu_dentry = debugfs_create_dir(cpu_dir, cpu_debugfs_dir); - err = cpu_init_allreg(cpu, cpu_dentry); - - pr_info("cpu%d(%d) debug files %d\n", - cpu, nr_cpu_ids, per_cpu(cpu_priv_count, cpu)); - if (per_cpu(cpu_priv_count, cpu) > MAX_CPU_FILES) { - pr_err("Register files count %d exceeds limit %d\n", - per_cpu(cpu_priv_count, cpu), MAX_CPU_FILES); - per_cpu(cpu_priv_count, cpu) = MAX_CPU_FILES; - err = -ENFILE; - } - if (err) - return err; - } - - return err; -} - -static int __init cpu_debug_init(void) -{ - cpu_debugfs_dir = debugfs_create_dir("cpu", arch_debugfs_dir); - - return cpu_init_cpu(); -} - -static void __exit cpu_debug_exit(void) -{ - int i, cpu; - - if (cpu_debugfs_dir) - debugfs_remove_recursive(cpu_debugfs_dir); - - for (cpu = 0; cpu < nr_cpu_ids; cpu++) - for (i = 0; i < per_cpu(cpu_priv_count, cpu); i++) - kfree(per_cpu(priv_arr[i], cpu)); -} - -module_init(cpu_debug_init); -module_exit(cpu_debug_exit); - -MODULE_AUTHOR("Jaswinder Singh Rajput"); -MODULE_DESCRIPTION("CPU Debug module"); -MODULE_LICENSE("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/Kconfig b/arch/x86/kernel/cpu/cpufreq/Kconfig index f138c6c389b9..870e6cc6ad28 100644 --- a/arch/x86/kernel/cpu/cpufreq/Kconfig +++ b/arch/x86/kernel/cpu/cpufreq/Kconfig @@ -10,6 +10,20 @@ if CPU_FREQ comment "CPUFreq processor drivers" +config X86_PCC_CPUFREQ + tristate "Processor Clocking Control interface driver" + depends on ACPI && ACPI_PROCESSOR + help + This driver adds support for the PCC interface. + + For details, take a look at: + <file:Documentation/cpu-freq/pcc-cpufreq.txt>. + + To compile this driver as a module, choose M here: the + module will be called pcc-cpufreq. + + If in doubt, say N. + config X86_ACPI_CPUFREQ tristate "ACPI Processor P-States driver" select CPU_FREQ_TABLE diff --git a/arch/x86/kernel/cpu/cpufreq/Makefile b/arch/x86/kernel/cpu/cpufreq/Makefile index 509296df294d..1840c0a5170b 100644 --- a/arch/x86/kernel/cpu/cpufreq/Makefile +++ b/arch/x86/kernel/cpu/cpufreq/Makefile @@ -4,6 +4,7 @@ obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o +obj-$(CONFIG_X86_PCC_CPUFREQ) += pcc-cpufreq.o obj-$(CONFIG_X86_POWERNOW_K6) += powernow-k6.o obj-$(CONFIG_X86_POWERNOW_K7) += powernow-k7.o obj-$(CONFIG_X86_LONGHAUL) += longhaul.o diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 7d5c3b0ea8da..1b1920fa7c80 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -68,9 +68,9 @@ struct acpi_cpufreq_data { unsigned int cpu_feature; }; -static DEFINE_PER_CPU(struct acpi_cpufreq_data *, drv_data); +static DEFINE_PER_CPU(struct acpi_cpufreq_data *, acfreq_data); -static DEFINE_PER_CPU(struct aperfmperf, old_perf); +static DEFINE_PER_CPU(struct aperfmperf, acfreq_old_perf); /* acpi_perf_data is a pointer to percpu data. */ static struct acpi_processor_performance *acpi_perf_data; @@ -190,9 +190,11 @@ static void do_drv_write(void *_cmd) static void drv_read(struct drv_cmd *cmd) { + int err; cmd->val = 0; - smp_call_function_single(cpumask_any(cmd->mask), do_drv_read, cmd, 1); + err = smp_call_function_any(cmd->mask, do_drv_read, cmd, 1); + WARN_ON_ONCE(err); /* smp_call_function_any() was buggy? */ } static void drv_write(struct drv_cmd *cmd) @@ -214,14 +216,14 @@ static u32 get_cur_val(const struct cpumask *mask) if (unlikely(cpumask_empty(mask))) return 0; - switch (per_cpu(drv_data, cpumask_first(mask))->cpu_feature) { + switch (per_cpu(acfreq_data, cpumask_first(mask))->cpu_feature) { case SYSTEM_INTEL_MSR_CAPABLE: cmd.type = SYSTEM_INTEL_MSR_CAPABLE; cmd.addr.msr.reg = MSR_IA32_PERF_STATUS; break; case SYSTEM_IO_CAPABLE: cmd.type = SYSTEM_IO_CAPABLE; - perf = per_cpu(drv_data, cpumask_first(mask))->acpi_data; + perf = per_cpu(acfreq_data, cpumask_first(mask))->acpi_data; cmd.addr.io.port = perf->control_register.address; cmd.addr.io.bit_width = perf->control_register.bit_width; break; @@ -268,8 +270,8 @@ static unsigned int get_measured_perf(struct cpufreq_policy *policy, if (smp_call_function_single(cpu, read_measured_perf_ctrs, &perf, 1)) return 0; - ratio = calc_aperfmperf_ratio(&per_cpu(old_perf, cpu), &perf); - per_cpu(old_perf, cpu) = perf; + ratio = calc_aperfmperf_ratio(&per_cpu(acfreq_old_perf, cpu), &perf); + per_cpu(acfreq_old_perf, cpu) = perf; retval = (policy->cpuinfo.max_freq * ratio) >> APERFMPERF_SHIFT; @@ -278,7 +280,7 @@ static unsigned int get_measured_perf(struct cpufreq_policy *policy, static unsigned int get_cur_freq_on_cpu(unsigned int cpu) { - struct acpi_cpufreq_data *data = per_cpu(drv_data, cpu); + struct acpi_cpufreq_data *data = per_cpu(acfreq_data, cpu); unsigned int freq; unsigned int cached_freq; @@ -322,7 +324,7 @@ static unsigned int check_freqs(const struct cpumask *mask, unsigned int freq, static int acpi_cpufreq_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation) { - struct acpi_cpufreq_data *data = per_cpu(drv_data, policy->cpu); + struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu); struct acpi_processor_performance *perf; struct cpufreq_freqs freqs; struct drv_cmd cmd; @@ -416,7 +418,7 @@ out: static int acpi_cpufreq_verify(struct cpufreq_policy *policy) { - struct acpi_cpufreq_data *data = per_cpu(drv_data, policy->cpu); + struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu); dprintk("acpi_cpufreq_verify\n"); @@ -526,15 +528,21 @@ static const struct dmi_system_id sw_any_bug_dmi_table[] = { static int acpi_cpufreq_blacklist(struct cpuinfo_x86 *c) { - /* http://www.intel.com/Assets/PDF/specupdate/314554.pdf + /* Intel Xeon Processor 7100 Series Specification Update + * http://www.intel.com/Assets/PDF/specupdate/314554.pdf * AL30: A Machine Check Exception (MCE) Occurring during an * Enhanced Intel SpeedStep Technology Ratio Change May Cause - * Both Processor Cores to Lock Up when HT is enabled*/ + * Both Processor Cores to Lock Up. */ if (c->x86_vendor == X86_VENDOR_INTEL) { if ((c->x86 == 15) && (c->x86_model == 6) && - (c->x86_mask == 8) && smt_capable()) + (c->x86_mask == 8)) { + printk(KERN_INFO "acpi-cpufreq: Intel(R) " + "Xeon(R) 7100 Errata AL30, processors may " + "lock up on frequency changes: disabling " + "acpi-cpufreq.\n"); return -ENODEV; + } } return 0; } @@ -549,13 +557,18 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) unsigned int result = 0; struct cpuinfo_x86 *c = &cpu_data(policy->cpu); struct acpi_processor_performance *perf; +#ifdef CONFIG_SMP + static int blacklisted; +#endif dprintk("acpi_cpufreq_cpu_init\n"); #ifdef CONFIG_SMP - result = acpi_cpufreq_blacklist(c); - if (result) - return result; + if (blacklisted) + return blacklisted; + blacklisted = acpi_cpufreq_blacklist(c); + if (blacklisted) + return blacklisted; #endif data = kzalloc(sizeof(struct acpi_cpufreq_data), GFP_KERNEL); @@ -563,7 +576,7 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) return -ENOMEM; data->acpi_data = per_cpu_ptr(acpi_perf_data, cpu); - per_cpu(drv_data, cpu) = data; + per_cpu(acfreq_data, cpu) = data; if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) acpi_cpufreq_driver.flags |= CPUFREQ_CONST_LOOPS; @@ -714,20 +727,20 @@ err_unreg: acpi_processor_unregister_performance(perf, cpu); err_free: kfree(data); - per_cpu(drv_data, cpu) = NULL; + per_cpu(acfreq_data, cpu) = NULL; return result; } static int acpi_cpufreq_cpu_exit(struct cpufreq_policy *policy) { - struct acpi_cpufreq_data *data = per_cpu(drv_data, policy->cpu); + struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu); dprintk("acpi_cpufreq_cpu_exit\n"); if (data) { cpufreq_frequency_table_put_attr(policy->cpu); - per_cpu(drv_data, policy->cpu) = NULL; + per_cpu(acfreq_data, policy->cpu) = NULL; acpi_processor_unregister_performance(data->acpi_data, policy->cpu); kfree(data); @@ -738,7 +751,7 @@ static int acpi_cpufreq_cpu_exit(struct cpufreq_policy *policy) static int acpi_cpufreq_resume(struct cpufreq_policy *policy) { - struct acpi_cpufreq_data *data = per_cpu(drv_data, policy->cpu); + struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu); dprintk("acpi_cpufreq_resume\n"); @@ -753,14 +766,15 @@ static struct freq_attr *acpi_cpufreq_attr[] = { }; static struct cpufreq_driver acpi_cpufreq_driver = { - .verify = acpi_cpufreq_verify, - .target = acpi_cpufreq_target, - .init = acpi_cpufreq_cpu_init, - .exit = acpi_cpufreq_cpu_exit, - .resume = acpi_cpufreq_resume, - .name = "acpi-cpufreq", - .owner = THIS_MODULE, - .attr = acpi_cpufreq_attr, + .verify = acpi_cpufreq_verify, + .target = acpi_cpufreq_target, + .bios_limit = acpi_processor_get_bios_limit, + .init = acpi_cpufreq_cpu_init, + .exit = acpi_cpufreq_cpu_exit, + .resume = acpi_cpufreq_resume, + .name = "acpi-cpufreq", + .owner = THIS_MODULE, + .attr = acpi_cpufreq_attr, }; static int __init acpi_cpufreq_init(void) diff --git a/arch/x86/kernel/cpu/cpufreq/longhaul.c b/arch/x86/kernel/cpu/cpufreq/longhaul.c index ce2ed3e4aad9..7e7eea4f8261 100644 --- a/arch/x86/kernel/cpu/cpufreq/longhaul.c +++ b/arch/x86/kernel/cpu/cpufreq/longhaul.c @@ -813,7 +813,7 @@ static int __init longhaul_cpu_init(struct cpufreq_policy *policy) memcpy(eblcr, samuel2_eblcr, sizeof(samuel2_eblcr)); break; case 1 ... 15: - longhaul_version = TYPE_LONGHAUL_V1; + longhaul_version = TYPE_LONGHAUL_V2; if (c->x86_mask < 8) { cpu_model = CPU_SAMUEL2; cpuname = "C3 'Samuel 2' [C5B]"; @@ -885,7 +885,7 @@ static int __init longhaul_cpu_init(struct cpufreq_policy *policy) /* Find ACPI data for processor */ acpi_walk_namespace(ACPI_TYPE_PROCESSOR, ACPI_ROOT_OBJECT, - ACPI_UINT32_MAX, &longhaul_walk_callback, + ACPI_UINT32_MAX, &longhaul_walk_callback, NULL, NULL, (void *)&pr); /* Check ACPI support for C3 state */ diff --git a/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c new file mode 100644 index 000000000000..ff36d2979a90 --- /dev/null +++ b/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c @@ -0,0 +1,620 @@ +/* + * pcc-cpufreq.c - Processor Clocking Control firmware cpufreq interface + * + * Copyright (C) 2009 Red Hat, Matthew Garrett <mjg@redhat.com> + * Copyright (C) 2009 Hewlett-Packard Development Company, L.P. + * Nagananda Chumbalkar <nagananda.chumbalkar@hp.com> + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * 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/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/smp.h> +#include <linux/sched.h> +#include <linux/cpufreq.h> +#include <linux/compiler.h> + +#include <linux/acpi.h> +#include <linux/io.h> +#include <linux/spinlock.h> +#include <linux/uaccess.h> + +#include <acpi/processor.h> + +#define PCC_VERSION "1.00.00" +#define POLL_LOOPS 300 + +#define CMD_COMPLETE 0x1 +#define CMD_GET_FREQ 0x0 +#define CMD_SET_FREQ 0x1 + +#define BUF_SZ 4 + +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "pcc-cpufreq", msg) + +struct pcc_register_resource { + u8 descriptor; + u16 length; + u8 space_id; + u8 bit_width; + u8 bit_offset; + u8 access_size; + u64 address; +} __attribute__ ((packed)); + +struct pcc_memory_resource { + u8 descriptor; + u16 length; + u8 space_id; + u8 resource_usage; + u8 type_specific; + u64 granularity; + u64 minimum; + u64 maximum; + u64 translation_offset; + u64 address_length; +} __attribute__ ((packed)); + +static struct cpufreq_driver pcc_cpufreq_driver; + +struct pcc_header { + u32 signature; + u16 length; + u8 major; + u8 minor; + u32 features; + u16 command; + u16 status; + u32 latency; + u32 minimum_time; + u32 maximum_time; + u32 nominal; + u32 throttled_frequency; + u32 minimum_frequency; +}; + +static void __iomem *pcch_virt_addr; +static struct pcc_header __iomem *pcch_hdr; + +static DEFINE_SPINLOCK(pcc_lock); + +static struct acpi_generic_address doorbell; + +static u64 doorbell_preserve; +static u64 doorbell_write; + +static u8 OSC_UUID[16] = {0x63, 0x9B, 0x2C, 0x9F, 0x70, 0x91, 0x49, 0x1f, + 0xBB, 0x4F, 0xA5, 0x98, 0x2F, 0xA1, 0xB5, 0x46}; + +struct pcc_cpu { + u32 input_offset; + u32 output_offset; +}; + +static struct pcc_cpu *pcc_cpu_info; + +static int pcc_cpufreq_verify(struct cpufreq_policy *policy) +{ + cpufreq_verify_within_limits(policy, policy->cpuinfo.min_freq, + policy->cpuinfo.max_freq); + return 0; +} + +static inline void pcc_cmd(void) +{ + u64 doorbell_value; + int i; + + acpi_read(&doorbell_value, &doorbell); + acpi_write((doorbell_value & doorbell_preserve) | doorbell_write, + &doorbell); + + for (i = 0; i < POLL_LOOPS; i++) { + if (ioread16(&pcch_hdr->status) & CMD_COMPLETE) + break; + } +} + +static inline void pcc_clear_mapping(void) +{ + if (pcch_virt_addr) + iounmap(pcch_virt_addr); + pcch_virt_addr = NULL; +} + +static unsigned int pcc_get_freq(unsigned int cpu) +{ + struct pcc_cpu *pcc_cpu_data; + unsigned int curr_freq; + unsigned int freq_limit; + u16 status; + u32 input_buffer; + u32 output_buffer; + + spin_lock(&pcc_lock); + + dprintk("get: get_freq for CPU %d\n", cpu); + pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); + + input_buffer = 0x1; + iowrite32(input_buffer, + (pcch_virt_addr + pcc_cpu_data->input_offset)); + iowrite16(CMD_GET_FREQ, &pcch_hdr->command); + + pcc_cmd(); + + output_buffer = + ioread32(pcch_virt_addr + pcc_cpu_data->output_offset); + + /* Clear the input buffer - we are done with the current command */ + memset_io((pcch_virt_addr + pcc_cpu_data->input_offset), 0, BUF_SZ); + + status = ioread16(&pcch_hdr->status); + if (status != CMD_COMPLETE) { + dprintk("get: FAILED: for CPU %d, status is %d\n", + cpu, status); + goto cmd_incomplete; + } + iowrite16(0, &pcch_hdr->status); + curr_freq = (((ioread32(&pcch_hdr->nominal) * (output_buffer & 0xff)) + / 100) * 1000); + + dprintk("get: SUCCESS: (virtual) output_offset for cpu %d is " + "0x%x, contains a value of: 0x%x. Speed is: %d MHz\n", + cpu, (pcch_virt_addr + pcc_cpu_data->output_offset), + output_buffer, curr_freq); + + freq_limit = (output_buffer >> 8) & 0xff; + if (freq_limit != 0xff) { + dprintk("get: frequency for cpu %d is being temporarily" + " capped at %d\n", cpu, curr_freq); + } + + spin_unlock(&pcc_lock); + return curr_freq; + +cmd_incomplete: + iowrite16(0, &pcch_hdr->status); + spin_unlock(&pcc_lock); + return -EINVAL; +} + +static int pcc_cpufreq_target(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int relation) +{ + struct pcc_cpu *pcc_cpu_data; + struct cpufreq_freqs freqs; + u16 status; + u32 input_buffer; + int cpu; + + spin_lock(&pcc_lock); + cpu = policy->cpu; + pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); + + dprintk("target: CPU %d should go to target freq: %d " + "(virtual) input_offset is 0x%x\n", + cpu, target_freq, + (pcch_virt_addr + pcc_cpu_data->input_offset)); + + freqs.new = target_freq; + freqs.cpu = cpu; + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + + input_buffer = 0x1 | (((target_freq * 100) + / (ioread32(&pcch_hdr->nominal) * 1000)) << 8); + iowrite32(input_buffer, + (pcch_virt_addr + pcc_cpu_data->input_offset)); + iowrite16(CMD_SET_FREQ, &pcch_hdr->command); + + pcc_cmd(); + + /* Clear the input buffer - we are done with the current command */ + memset_io((pcch_virt_addr + pcc_cpu_data->input_offset), 0, BUF_SZ); + + status = ioread16(&pcch_hdr->status); + if (status != CMD_COMPLETE) { + dprintk("target: FAILED for cpu %d, with status: 0x%x\n", + cpu, status); + goto cmd_incomplete; + } + iowrite16(0, &pcch_hdr->status); + + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + dprintk("target: was SUCCESSFUL for cpu %d\n", cpu); + spin_unlock(&pcc_lock); + + return 0; + +cmd_incomplete: + iowrite16(0, &pcch_hdr->status); + spin_unlock(&pcc_lock); + return -EINVAL; +} + +static int pcc_get_offset(int cpu) +{ + acpi_status status; + struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; + union acpi_object *pccp, *offset; + struct pcc_cpu *pcc_cpu_data; + struct acpi_processor *pr; + int ret = 0; + + pr = per_cpu(processors, cpu); + pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); + + status = acpi_evaluate_object(pr->handle, "PCCP", NULL, &buffer); + if (ACPI_FAILURE(status)) + return -ENODEV; + + pccp = buffer.pointer; + if (!pccp || pccp->type != ACPI_TYPE_PACKAGE) { + ret = -ENODEV; + goto out_free; + }; + + offset = &(pccp->package.elements[0]); + if (!offset || offset->type != ACPI_TYPE_INTEGER) { + ret = -ENODEV; + goto out_free; + } + + pcc_cpu_data->input_offset = offset->integer.value; + + offset = &(pccp->package.elements[1]); + if (!offset || offset->type != ACPI_TYPE_INTEGER) { + ret = -ENODEV; + goto out_free; + } + + pcc_cpu_data->output_offset = offset->integer.value; + + memset_io((pcch_virt_addr + pcc_cpu_data->input_offset), 0, BUF_SZ); + memset_io((pcch_virt_addr + pcc_cpu_data->output_offset), 0, BUF_SZ); + + dprintk("pcc_get_offset: for CPU %d: pcc_cpu_data " + "input_offset: 0x%x, pcc_cpu_data output_offset: 0x%x\n", + cpu, pcc_cpu_data->input_offset, pcc_cpu_data->output_offset); +out_free: + kfree(buffer.pointer); + return ret; +} + +static int __init pcc_cpufreq_do_osc(acpi_handle *handle) +{ + acpi_status status; + struct acpi_object_list input; + struct acpi_buffer output = {ACPI_ALLOCATE_BUFFER, NULL}; + union acpi_object in_params[4]; + union acpi_object *out_obj; + u32 capabilities[2]; + u32 errors; + u32 supported; + int ret = 0; + + input.count = 4; + input.pointer = in_params; + input.count = 4; + input.pointer = in_params; + in_params[0].type = ACPI_TYPE_BUFFER; + in_params[0].buffer.length = 16; + in_params[0].buffer.pointer = OSC_UUID; + in_params[1].type = ACPI_TYPE_INTEGER; + in_params[1].integer.value = 1; + in_params[2].type = ACPI_TYPE_INTEGER; + in_params[2].integer.value = 2; + in_params[3].type = ACPI_TYPE_BUFFER; + in_params[3].buffer.length = 8; + in_params[3].buffer.pointer = (u8 *)&capabilities; + + capabilities[0] = OSC_QUERY_ENABLE; + capabilities[1] = 0x1; + + status = acpi_evaluate_object(*handle, "_OSC", &input, &output); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (!output.length) + return -ENODEV; + + out_obj = output.pointer; + if (out_obj->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto out_free; + } + + errors = *((u32 *)out_obj->buffer.pointer) & ~(1 << 0); + if (errors) { + ret = -ENODEV; + goto out_free; + } + + supported = *((u32 *)(out_obj->buffer.pointer + 4)); + if (!(supported & 0x1)) { + ret = -ENODEV; + goto out_free; + } + + kfree(output.pointer); + capabilities[0] = 0x0; + capabilities[1] = 0x1; + + status = acpi_evaluate_object(*handle, "_OSC", &input, &output); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (!output.length) + return -ENODEV; + + out_obj = output.pointer; + if (out_obj->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto out_free; + } + + errors = *((u32 *)out_obj->buffer.pointer) & ~(1 << 0); + if (errors) { + ret = -ENODEV; + goto out_free; + } + + supported = *((u32 *)(out_obj->buffer.pointer + 4)); + if (!(supported & 0x1)) { + ret = -ENODEV; + goto out_free; + } + +out_free: + kfree(output.pointer); + return ret; +} + +static int __init pcc_cpufreq_probe(void) +{ + acpi_status status; + struct acpi_buffer output = {ACPI_ALLOCATE_BUFFER, NULL}; + struct pcc_memory_resource *mem_resource; + struct pcc_register_resource *reg_resource; + union acpi_object *out_obj, *member; + acpi_handle handle, osc_handle; + int ret = 0; + + status = acpi_get_handle(NULL, "\\_SB", &handle); + if (ACPI_FAILURE(status)) + return -ENODEV; + + status = acpi_get_handle(handle, "_OSC", &osc_handle); + if (ACPI_SUCCESS(status)) { + ret = pcc_cpufreq_do_osc(&osc_handle); + if (ret) + dprintk("probe: _OSC evaluation did not succeed\n"); + /* Firmware's use of _OSC is optional */ + ret = 0; + } + + status = acpi_evaluate_object(handle, "PCCH", NULL, &output); + if (ACPI_FAILURE(status)) + return -ENODEV; + + out_obj = output.pointer; + if (out_obj->type != ACPI_TYPE_PACKAGE) { + ret = -ENODEV; + goto out_free; + } + + member = &out_obj->package.elements[0]; + if (member->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto out_free; + } + + mem_resource = (struct pcc_memory_resource *)member->buffer.pointer; + + dprintk("probe: mem_resource descriptor: 0x%x," + " length: %d, space_id: %d, resource_usage: %d," + " type_specific: %d, granularity: 0x%llx," + " minimum: 0x%llx, maximum: 0x%llx," + " translation_offset: 0x%llx, address_length: 0x%llx\n", + mem_resource->descriptor, mem_resource->length, + mem_resource->space_id, mem_resource->resource_usage, + mem_resource->type_specific, mem_resource->granularity, + mem_resource->minimum, mem_resource->maximum, + mem_resource->translation_offset, + mem_resource->address_length); + + if (mem_resource->space_id != ACPI_ADR_SPACE_SYSTEM_MEMORY) { + ret = -ENODEV; + goto out_free; + } + + pcch_virt_addr = ioremap_nocache(mem_resource->minimum, + mem_resource->address_length); + if (pcch_virt_addr == NULL) { + dprintk("probe: could not map shared mem region\n"); + goto out_free; + } + pcch_hdr = pcch_virt_addr; + + dprintk("probe: PCCH header (virtual) addr: 0x%p\n", pcch_hdr); + dprintk("probe: PCCH header is at physical address: 0x%llx," + " signature: 0x%x, length: %d bytes, major: %d, minor: %d," + " supported features: 0x%x, command field: 0x%x," + " status field: 0x%x, nominal latency: %d us\n", + mem_resource->minimum, ioread32(&pcch_hdr->signature), + ioread16(&pcch_hdr->length), ioread8(&pcch_hdr->major), + ioread8(&pcch_hdr->minor), ioread32(&pcch_hdr->features), + ioread16(&pcch_hdr->command), ioread16(&pcch_hdr->status), + ioread32(&pcch_hdr->latency)); + + dprintk("probe: min time between commands: %d us," + " max time between commands: %d us," + " nominal CPU frequency: %d MHz," + " minimum CPU frequency: %d MHz," + " minimum CPU frequency without throttling: %d MHz\n", + ioread32(&pcch_hdr->minimum_time), + ioread32(&pcch_hdr->maximum_time), + ioread32(&pcch_hdr->nominal), + ioread32(&pcch_hdr->throttled_frequency), + ioread32(&pcch_hdr->minimum_frequency)); + + member = &out_obj->package.elements[1]; + if (member->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto pcch_free; + } + + reg_resource = (struct pcc_register_resource *)member->buffer.pointer; + + doorbell.space_id = reg_resource->space_id; + doorbell.bit_width = reg_resource->bit_width; + doorbell.bit_offset = reg_resource->bit_offset; + doorbell.access_width = 64; + doorbell.address = reg_resource->address; + + dprintk("probe: doorbell: space_id is %d, bit_width is %d, " + "bit_offset is %d, access_width is %d, address is 0x%llx\n", + doorbell.space_id, doorbell.bit_width, doorbell.bit_offset, + doorbell.access_width, reg_resource->address); + + member = &out_obj->package.elements[2]; + if (member->type != ACPI_TYPE_INTEGER) { + ret = -ENODEV; + goto pcch_free; + } + + doorbell_preserve = member->integer.value; + + member = &out_obj->package.elements[3]; + if (member->type != ACPI_TYPE_INTEGER) { + ret = -ENODEV; + goto pcch_free; + } + + doorbell_write = member->integer.value; + + dprintk("probe: doorbell_preserve: 0x%llx," + " doorbell_write: 0x%llx\n", + doorbell_preserve, doorbell_write); + + pcc_cpu_info = alloc_percpu(struct pcc_cpu); + if (!pcc_cpu_info) { + ret = -ENOMEM; + goto pcch_free; + } + + printk(KERN_DEBUG "pcc-cpufreq: (v%s) driver loaded with frequency" + " limits: %d MHz, %d MHz\n", PCC_VERSION, + ioread32(&pcch_hdr->minimum_frequency), + ioread32(&pcch_hdr->nominal)); + kfree(output.pointer); + return ret; +pcch_free: + pcc_clear_mapping(); +out_free: + kfree(output.pointer); + return ret; +} + +static int pcc_cpufreq_cpu_init(struct cpufreq_policy *policy) +{ + unsigned int cpu = policy->cpu; + unsigned int result = 0; + + if (!pcch_virt_addr) { + result = -1; + goto pcch_null; + } + + result = pcc_get_offset(cpu); + if (result) { + dprintk("init: PCCP evaluation failed\n"); + goto free; + } + + policy->max = policy->cpuinfo.max_freq = + ioread32(&pcch_hdr->nominal) * 1000; + policy->min = policy->cpuinfo.min_freq = + ioread32(&pcch_hdr->minimum_frequency) * 1000; + policy->cur = pcc_get_freq(cpu); + + dprintk("init: policy->max is %d, policy->min is %d\n", + policy->max, policy->min); + + return 0; +free: + pcc_clear_mapping(); + free_percpu(pcc_cpu_info); +pcch_null: + return result; +} + +static int pcc_cpufreq_cpu_exit(struct cpufreq_policy *policy) +{ + return 0; +} + +static struct cpufreq_driver pcc_cpufreq_driver = { + .flags = CPUFREQ_CONST_LOOPS, + .get = pcc_get_freq, + .verify = pcc_cpufreq_verify, + .target = pcc_cpufreq_target, + .init = pcc_cpufreq_cpu_init, + .exit = pcc_cpufreq_cpu_exit, + .name = "pcc-cpufreq", + .owner = THIS_MODULE, +}; + +static int __init pcc_cpufreq_init(void) +{ + int ret; + + if (acpi_disabled) + return 0; + + ret = pcc_cpufreq_probe(); + if (ret) { + dprintk("pcc_cpufreq_init: PCCH evaluation failed\n"); + return ret; + } + + ret = cpufreq_register_driver(&pcc_cpufreq_driver); + + return ret; +} + +static void __exit pcc_cpufreq_exit(void) +{ + cpufreq_unregister_driver(&pcc_cpufreq_driver); + + pcc_clear_mapping(); + + free_percpu(pcc_cpu_info); +} + +MODULE_AUTHOR("Matthew Garrett, Naga Chumbalkar"); +MODULE_VERSION(PCC_VERSION); +MODULE_DESCRIPTION("Processor Clocking Control interface driver"); +MODULE_LICENSE("GPL"); + +late_initcall(pcc_cpufreq_init); +module_exit(pcc_cpufreq_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c index f10dea409f40..cb01dac267d3 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c @@ -164,7 +164,7 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy) } /* cpuinfo and default policy values */ - policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL; + policy->cpuinfo.transition_latency = 200000; policy->cur = busfreq * max_multiplier; result = cpufreq_frequency_table_cpuinfo(policy, clock_ratio); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c index d47c775eb0ab..9a97116f89e5 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c @@ -714,14 +714,17 @@ static struct freq_attr *powernow_table_attr[] = { }; static struct cpufreq_driver powernow_driver = { - .verify = powernow_verify, - .target = powernow_target, - .get = powernow_get, - .init = powernow_cpu_init, - .exit = powernow_cpu_exit, - .name = "powernow-k7", - .owner = THIS_MODULE, - .attr = powernow_table_attr, + .verify = powernow_verify, + .target = powernow_target, + .get = powernow_get, +#ifdef CONFIG_X86_POWERNOW_K7_ACPI + .bios_limit = acpi_processor_get_bios_limit, +#endif + .init = powernow_cpu_init, + .exit = powernow_cpu_exit, + .name = "powernow-k7", + .owner = THIS_MODULE, + .attr = powernow_table_attr, }; static int __init powernow_init(void) diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index 6394aa5c7985..d360b56e9825 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -806,7 +806,7 @@ static int find_psb_table(struct powernow_k8_data *data) static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index) { - acpi_integer control; + u64 control; if (!data->acpi_data.state_count || (cpu_family == CPU_HW_PSTATE)) return; @@ -824,7 +824,7 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) { struct cpufreq_frequency_table *powernow_table; int ret_val = -ENODEV; - acpi_integer control, status; + u64 control, status; if (acpi_processor_register_performance(&data->acpi_data, data->cpu)) { dprintk("register performance failed: bad ACPI data\n"); @@ -948,7 +948,7 @@ static int fill_powernow_table_fidvid(struct powernow_k8_data *data, u32 fid; u32 vid; u32 freq, index; - acpi_integer status, control; + u64 status, control; if (data->exttype) { status = data->acpi_data.states[i].status; @@ -1022,7 +1022,7 @@ static int get_transition_latency(struct powernow_k8_data *data) * set it to 1 to avoid problems in the future. * For all others it's a BIOS bug. */ - if (!boot_cpu_data.x86 == 0x11) + if (boot_cpu_data.x86 != 0x11) printk(KERN_ERR FW_WARN PFX "Invalid zero transition " "latency\n"); max_latency = 1; @@ -1118,7 +1118,7 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsigned relation) { - cpumask_t oldmask; + cpumask_var_t oldmask; struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu); u32 checkfid; u32 checkvid; @@ -1131,9 +1131,13 @@ static int powernowk8_target(struct cpufreq_policy *pol, checkfid = data->currfid; checkvid = data->currvid; - /* only run on specific CPU from here on */ - oldmask = current->cpus_allowed; - set_cpus_allowed_ptr(current, &cpumask_of_cpu(pol->cpu)); + /* only run on specific CPU from here on. */ + /* This is poor form: use a workqueue or smp_call_function_single */ + if (!alloc_cpumask_var(&oldmask, GFP_KERNEL)) + return -ENOMEM; + + cpumask_copy(oldmask, tsk_cpus_allowed(current)); + set_cpus_allowed_ptr(current, cpumask_of(pol->cpu)); if (smp_processor_id() != pol->cpu) { printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu); @@ -1193,7 +1197,8 @@ static int powernowk8_target(struct cpufreq_policy *pol, ret = 0; err_out: - set_cpus_allowed_ptr(current, &oldmask); + set_cpus_allowed_ptr(current, oldmask); + free_cpumask_var(oldmask); return ret; } @@ -1351,6 +1356,7 @@ static int __devexit powernowk8_cpu_exit(struct cpufreq_policy *pol) kfree(data->powernow_table); kfree(data); + per_cpu(powernow_data, pol->cpu) = NULL; return 0; } @@ -1370,7 +1376,7 @@ static unsigned int powernowk8_get(unsigned int cpu) int err; if (!data) - return -EINVAL; + return 0; smp_call_function_single(cpu, query_values_on_cpu, &err, true); if (err) @@ -1393,14 +1399,15 @@ static struct freq_attr *powernow_k8_attr[] = { }; static struct cpufreq_driver cpufreq_amd64_driver = { - .verify = powernowk8_verify, - .target = powernowk8_target, - .init = powernowk8_cpu_init, - .exit = __devexit_p(powernowk8_cpu_exit), - .get = powernowk8_get, - .name = "powernow-k8", - .owner = THIS_MODULE, - .attr = powernow_k8_attr, + .verify = powernowk8_verify, + .target = powernowk8_target, + .bios_limit = acpi_processor_get_bios_limit, + .init = powernowk8_cpu_init, + .exit = __devexit_p(powernowk8_cpu_exit), + .get = powernowk8_get, + .name = "powernow-k8", + .owner = THIS_MODULE, + .attr = powernow_k8_attr, }; /* driver entry point for init */ diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c index 6911e91fb4f6..2ce8e0b5cc54 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c @@ -39,7 +39,7 @@ static struct pci_dev *speedstep_chipset_dev; /* speedstep_processor */ -static unsigned int speedstep_processor; +static enum speedstep_processor speedstep_processor; static u32 pmbase; @@ -232,28 +232,23 @@ static unsigned int speedstep_detect_chipset(void) return 0; } -struct get_freq_data { - unsigned int speed; - unsigned int processor; -}; - -static void get_freq_data(void *_data) +static void get_freq_data(void *_speed) { - struct get_freq_data *data = _data; + unsigned int *speed = _speed; - data->speed = speedstep_get_frequency(data->processor); + *speed = speedstep_get_frequency(speedstep_processor); } static unsigned int speedstep_get(unsigned int cpu) { - struct get_freq_data data = { .processor = cpu }; + unsigned int speed; /* You're supposed to ensure CPU is online. */ - if (smp_call_function_single(cpu, get_freq_data, &data, 1) != 0) + if (smp_call_function_single(cpu, get_freq_data, &speed, 1) != 0) BUG(); - dprintk("detected %u kHz as current frequency\n", data.speed); - return data.speed; + dprintk("detected %u kHz as current frequency\n", speed); + return speed; } /** diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c index f4c290b8482f..ad0083abfa23 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c @@ -34,7 +34,7 @@ static int relaxed_check; * GET PROCESSOR CORE SPEED IN KHZ * *********************************************************************/ -static unsigned int pentium3_get_frequency(unsigned int processor) +static unsigned int pentium3_get_frequency(enum speedstep_processor processor) { /* See table 14 of p3_ds.pdf and table 22 of 29834003.pdf */ struct { @@ -227,7 +227,7 @@ static unsigned int pentium4_get_frequency(void) /* Warning: may get called from smp_call_function_single. */ -unsigned int speedstep_get_frequency(unsigned int processor) +unsigned int speedstep_get_frequency(enum speedstep_processor processor) { switch (processor) { case SPEEDSTEP_CPU_PCORE: @@ -380,7 +380,7 @@ EXPORT_SYMBOL_GPL(speedstep_detect_processor); * DETECT SPEEDSTEP SPEEDS * *********************************************************************/ -unsigned int speedstep_get_freqs(unsigned int processor, +unsigned int speedstep_get_freqs(enum speedstep_processor processor, unsigned int *low_speed, unsigned int *high_speed, unsigned int *transition_latency, diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.h b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.h index 2b6c04e5a304..70d9cea1219d 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.h +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.h @@ -11,18 +11,18 @@ /* processors */ - -#define SPEEDSTEP_CPU_PIII_C_EARLY 0x00000001 /* Coppermine core */ -#define SPEEDSTEP_CPU_PIII_C 0x00000002 /* Coppermine core */ -#define SPEEDSTEP_CPU_PIII_T 0x00000003 /* Tualatin core */ -#define SPEEDSTEP_CPU_P4M 0x00000004 /* P4-M */ - +enum speedstep_processor { + SPEEDSTEP_CPU_PIII_C_EARLY = 0x00000001, /* Coppermine core */ + SPEEDSTEP_CPU_PIII_C = 0x00000002, /* Coppermine core */ + SPEEDSTEP_CPU_PIII_T = 0x00000003, /* Tualatin core */ + SPEEDSTEP_CPU_P4M = 0x00000004, /* P4-M */ /* the following processors are not speedstep-capable and are not auto-detected * in speedstep_detect_processor(). However, their speed can be detected using * the speedstep_get_frequency() call. */ -#define SPEEDSTEP_CPU_PM 0xFFFFFF03 /* Pentium M */ -#define SPEEDSTEP_CPU_P4D 0xFFFFFF04 /* desktop P4 */ -#define SPEEDSTEP_CPU_PCORE 0xFFFFFF05 /* Core */ + SPEEDSTEP_CPU_PM = 0xFFFFFF03, /* Pentium M */ + SPEEDSTEP_CPU_P4D = 0xFFFFFF04, /* desktop P4 */ + SPEEDSTEP_CPU_PCORE = 0xFFFFFF05, /* Core */ +}; /* speedstep states -- only two of them */ @@ -31,10 +31,10 @@ /* detect a speedstep-capable processor */ -extern unsigned int speedstep_detect_processor (void); +extern enum speedstep_processor speedstep_detect_processor(void); /* detect the current speed (in khz) of the processor */ -extern unsigned int speedstep_get_frequency(unsigned int processor); +extern unsigned int speedstep_get_frequency(enum speedstep_processor processor); /* detect the low and high speeds of the processor. The callback @@ -42,7 +42,7 @@ extern unsigned int speedstep_get_frequency(unsigned int processor); * SPEEDSTEP_LOW; the second argument is zero so that no * cpufreq_notify_transition calls are initiated. */ -extern unsigned int speedstep_get_freqs(unsigned int processor, +extern unsigned int speedstep_get_freqs(enum speedstep_processor processor, unsigned int *low_speed, unsigned int *high_speed, unsigned int *transition_latency, diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c b/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c index befea088e4f5..04d73c114e49 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c @@ -35,7 +35,7 @@ static int smi_cmd; static unsigned int smi_sig; /* info about the processor */ -static unsigned int speedstep_processor; +static enum speedstep_processor speedstep_processor; /* * There are only two frequency states for each processor. Values diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c index 19807b89f058..4fbd384fb645 100644 --- a/arch/x86/kernel/cpu/cyrix.c +++ b/arch/x86/kernel/cpu/cyrix.c @@ -373,7 +373,7 @@ static void __cpuinit init_nsc(struct cpuinfo_x86 *c) /* Handle the GX (Formally known as the GX2) */ if (c->x86 == 5 && c->x86_model == 5) - display_cacheinfo(c); + cpu_detect_cache_sizes(c); else init_cyrix(c); } diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 40e1835b35e8..7e1cca13af35 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -70,8 +70,8 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) if (c->x86_power & (1 << 8)) { set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC); - set_cpu_cap(c, X86_FEATURE_TSC_RELIABLE); - sched_clock_stable = 1; + if (!check_tsc_unstable()) + sched_clock_stable = 1; } /* @@ -263,11 +263,13 @@ 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_online(node)) + if (node == NUMA_NO_NODE) node = first_node(node_online_map); + else if (!node_online(node)) { + /* reuse the value from init_cpu_to_node() */ + node = cpu_to_node(cpu); + } numa_set_node(cpu, node); - - printk(KERN_INFO "CPU %d/0x%x -> Node %d\n", cpu, apicid, node); #endif } diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 804c40e2bc3e..b3eeb66c0a51 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -18,6 +18,7 @@ #include <asm/processor.h> #include <linux/smp.h> #include <asm/k8.h> +#include <asm/smp.h> #define LVL_1_INST 1 #define LVL_1_DATA 2 @@ -31,6 +32,8 @@ struct _cache_table { short size; }; +#define MB(x) ((x) * 1024) + /* All the cache descriptor types we care about (no TLB or trace cache entries) */ @@ -44,9 +47,9 @@ static const struct _cache_table __cpuinitconst cache_table[] = { 0x0d, LVL_1_DATA, 16 }, /* 4-way set assoc, 64 byte line size */ { 0x21, LVL_2, 256 }, /* 8-way set assoc, 64 byte line size */ { 0x22, LVL_3, 512 }, /* 4-way set assoc, sectored cache, 64 byte line size */ - { 0x23, LVL_3, 1024 }, /* 8-way set assoc, sectored cache, 64 byte line size */ - { 0x25, LVL_3, 2048 }, /* 8-way set assoc, sectored cache, 64 byte line size */ - { 0x29, LVL_3, 4096 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x23, LVL_3, MB(1) }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x25, LVL_3, MB(2) }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x29, LVL_3, MB(4) }, /* 8-way set assoc, sectored cache, 64 byte line size */ { 0x2c, LVL_1_DATA, 32 }, /* 8-way set assoc, 64 byte line size */ { 0x30, LVL_1_INST, 32 }, /* 8-way set assoc, 64 byte line size */ { 0x39, LVL_2, 128 }, /* 4-way set assoc, sectored cache, 64 byte line size */ @@ -59,16 +62,16 @@ static const struct _cache_table __cpuinitconst cache_table[] = { 0x41, LVL_2, 128 }, /* 4-way set assoc, 32 byte line size */ { 0x42, LVL_2, 256 }, /* 4-way set assoc, 32 byte line size */ { 0x43, LVL_2, 512 }, /* 4-way set assoc, 32 byte line size */ - { 0x44, LVL_2, 1024 }, /* 4-way set assoc, 32 byte line size */ - { 0x45, LVL_2, 2048 }, /* 4-way set assoc, 32 byte line size */ - { 0x46, LVL_3, 4096 }, /* 4-way set assoc, 64 byte line size */ - { 0x47, LVL_3, 8192 }, /* 8-way set assoc, 64 byte line size */ - { 0x49, LVL_3, 4096 }, /* 16-way set assoc, 64 byte line size */ - { 0x4a, LVL_3, 6144 }, /* 12-way set assoc, 64 byte line size */ - { 0x4b, LVL_3, 8192 }, /* 16-way set assoc, 64 byte line size */ - { 0x4c, LVL_3, 12288 }, /* 12-way set assoc, 64 byte line size */ - { 0x4d, LVL_3, 16384 }, /* 16-way set assoc, 64 byte line size */ - { 0x4e, LVL_2, 6144 }, /* 24-way set assoc, 64 byte line size */ + { 0x44, LVL_2, MB(1) }, /* 4-way set assoc, 32 byte line size */ + { 0x45, LVL_2, MB(2) }, /* 4-way set assoc, 32 byte line size */ + { 0x46, LVL_3, MB(4) }, /* 4-way set assoc, 64 byte line size */ + { 0x47, LVL_3, MB(8) }, /* 8-way set assoc, 64 byte line size */ + { 0x49, LVL_3, MB(4) }, /* 16-way set assoc, 64 byte line size */ + { 0x4a, LVL_3, MB(6) }, /* 12-way set assoc, 64 byte line size */ + { 0x4b, LVL_3, MB(8) }, /* 16-way set assoc, 64 byte line size */ + { 0x4c, LVL_3, MB(12) }, /* 12-way set assoc, 64 byte line size */ + { 0x4d, LVL_3, MB(16) }, /* 16-way set assoc, 64 byte line size */ + { 0x4e, LVL_2, MB(6) }, /* 24-way set assoc, 64 byte line size */ { 0x60, LVL_1_DATA, 16 }, /* 8-way set assoc, sectored cache, 64 byte line size */ { 0x66, LVL_1_DATA, 8 }, /* 4-way set assoc, sectored cache, 64 byte line size */ { 0x67, LVL_1_DATA, 16 }, /* 4-way set assoc, sectored cache, 64 byte line size */ @@ -77,31 +80,34 @@ static const struct _cache_table __cpuinitconst cache_table[] = { 0x71, LVL_TRACE, 16 }, /* 8-way set assoc */ { 0x72, LVL_TRACE, 32 }, /* 8-way set assoc */ { 0x73, LVL_TRACE, 64 }, /* 8-way set assoc */ - { 0x78, LVL_2, 1024 }, /* 4-way set assoc, 64 byte line size */ - { 0x79, LVL_2, 128 }, /* 8-way set assoc, sectored cache, 64 byte line size */ - { 0x7a, LVL_2, 256 }, /* 8-way set assoc, sectored cache, 64 byte line size */ - { 0x7b, LVL_2, 512 }, /* 8-way set assoc, sectored cache, 64 byte line size */ - { 0x7c, LVL_2, 1024 }, /* 8-way set assoc, sectored cache, 64 byte line size */ - { 0x7d, LVL_2, 2048 }, /* 8-way set assoc, 64 byte line size */ - { 0x7f, LVL_2, 512 }, /* 2-way set assoc, 64 byte line size */ - { 0x82, LVL_2, 256 }, /* 8-way set assoc, 32 byte line size */ - { 0x83, LVL_2, 512 }, /* 8-way set assoc, 32 byte line size */ - { 0x84, LVL_2, 1024 }, /* 8-way set assoc, 32 byte line size */ - { 0x85, LVL_2, 2048 }, /* 8-way set assoc, 32 byte line size */ - { 0x86, LVL_2, 512 }, /* 4-way set assoc, 64 byte line size */ - { 0x87, LVL_2, 1024 }, /* 8-way set assoc, 64 byte line size */ - { 0xd0, LVL_3, 512 }, /* 4-way set assoc, 64 byte line size */ - { 0xd1, LVL_3, 1024 }, /* 4-way set assoc, 64 byte line size */ - { 0xd2, LVL_3, 2048 }, /* 4-way set assoc, 64 byte line size */ - { 0xd6, LVL_3, 1024 }, /* 8-way set assoc, 64 byte line size */ - { 0xd7, LVL_3, 2038 }, /* 8-way set assoc, 64 byte line size */ - { 0xd8, LVL_3, 4096 }, /* 12-way set assoc, 64 byte line size */ - { 0xdc, LVL_3, 2048 }, /* 12-way set assoc, 64 byte line size */ - { 0xdd, LVL_3, 4096 }, /* 12-way set assoc, 64 byte line size */ - { 0xde, LVL_3, 8192 }, /* 12-way set assoc, 64 byte line size */ - { 0xe2, LVL_3, 2048 }, /* 16-way set assoc, 64 byte line size */ - { 0xe3, LVL_3, 4096 }, /* 16-way set assoc, 64 byte line size */ - { 0xe4, LVL_3, 8192 }, /* 16-way set assoc, 64 byte line size */ + { 0x78, LVL_2, MB(1) }, /* 4-way set assoc, 64 byte line size */ + { 0x79, LVL_2, 128 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x7a, LVL_2, 256 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x7b, LVL_2, 512 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x7c, LVL_2, MB(1) }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x7d, LVL_2, MB(2) }, /* 8-way set assoc, 64 byte line size */ + { 0x7f, LVL_2, 512 }, /* 2-way set assoc, 64 byte line size */ + { 0x82, LVL_2, 256 }, /* 8-way set assoc, 32 byte line size */ + { 0x83, LVL_2, 512 }, /* 8-way set assoc, 32 byte line size */ + { 0x84, LVL_2, MB(1) }, /* 8-way set assoc, 32 byte line size */ + { 0x85, LVL_2, MB(2) }, /* 8-way set assoc, 32 byte line size */ + { 0x86, LVL_2, 512 }, /* 4-way set assoc, 64 byte line size */ + { 0x87, LVL_2, MB(1) }, /* 8-way set assoc, 64 byte line size */ + { 0xd0, LVL_3, 512 }, /* 4-way set assoc, 64 byte line size */ + { 0xd1, LVL_3, MB(1) }, /* 4-way set assoc, 64 byte line size */ + { 0xd2, LVL_3, MB(2) }, /* 4-way set assoc, 64 byte line size */ + { 0xd6, LVL_3, MB(1) }, /* 8-way set assoc, 64 byte line size */ + { 0xd7, LVL_3, MB(2) }, /* 8-way set assoc, 64 byte line size */ + { 0xd8, LVL_3, MB(4) }, /* 12-way set assoc, 64 byte line size */ + { 0xdc, LVL_3, MB(2) }, /* 12-way set assoc, 64 byte line size */ + { 0xdd, LVL_3, MB(4) }, /* 12-way set assoc, 64 byte line size */ + { 0xde, LVL_3, MB(8) }, /* 12-way set assoc, 64 byte line size */ + { 0xe2, LVL_3, MB(2) }, /* 16-way set assoc, 64 byte line size */ + { 0xe3, LVL_3, MB(4) }, /* 16-way set assoc, 64 byte line size */ + { 0xe4, LVL_3, MB(8) }, /* 16-way set assoc, 64 byte line size */ + { 0xea, LVL_3, MB(12) }, /* 24-way set assoc, 64 byte line size */ + { 0xeb, LVL_3, MB(18) }, /* 24-way set assoc, 64 byte line size */ + { 0xec, LVL_3, MB(24) }, /* 24-way set assoc, 64 byte line size */ { 0x00, 0, 0} }; @@ -147,7 +153,8 @@ struct _cpuid4_info { union _cpuid4_leaf_ebx ebx; union _cpuid4_leaf_ecx ecx; unsigned long size; - unsigned long can_disable; + bool can_disable; + unsigned int l3_indices; DECLARE_BITMAP(shared_cpu_map, NR_CPUS); }; @@ -157,7 +164,8 @@ struct _cpuid4_info_regs { union _cpuid4_leaf_ebx ebx; union _cpuid4_leaf_ecx ecx; unsigned long size; - unsigned long can_disable; + bool can_disable; + unsigned int l3_indices; }; unsigned short num_cache_leaves; @@ -287,6 +295,36 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax, (ebx->split.ways_of_associativity + 1) - 1; } +struct _cache_attr { + struct attribute attr; + ssize_t (*show)(struct _cpuid4_info *, char *); + ssize_t (*store)(struct _cpuid4_info *, const char *, size_t count); +}; + +#ifdef CONFIG_CPU_SUP_AMD +static unsigned int __cpuinit amd_calc_l3_indices(void) +{ + /* + * We're called over smp_call_function_single() and therefore + * are on the correct cpu. + */ + int cpu = smp_processor_id(); + int node = cpu_to_node(cpu); + struct pci_dev *dev = node_to_k8_nb_misc(node); + unsigned int sc0, sc1, sc2, sc3; + u32 val = 0; + + pci_read_config_dword(dev, 0x1C4, &val); + + /* calculate subcache sizes */ + sc0 = !(val & BIT(0)); + sc1 = !(val & BIT(4)); + sc2 = !(val & BIT(8)) + !(val & BIT(9)); + sc3 = !(val & BIT(12)) + !(val & BIT(13)); + + return (max(max(max(sc0, sc1), sc2), sc3) << 10) - 1; +} + static void __cpuinit amd_check_l3_disable(int index, struct _cpuid4_info_regs *this_leaf) { @@ -296,13 +334,104 @@ amd_check_l3_disable(int index, struct _cpuid4_info_regs *this_leaf) if (boot_cpu_data.x86 == 0x11) return; - /* see erratum #382 */ - if ((boot_cpu_data.x86 == 0x10) && (boot_cpu_data.x86_model < 0x8)) + /* see errata #382 and #388 */ + if ((boot_cpu_data.x86 == 0x10) && + ((boot_cpu_data.x86_model < 0x8) || + (boot_cpu_data.x86_mask < 0x1))) return; - this_leaf->can_disable = 1; + this_leaf->can_disable = true; + this_leaf->l3_indices = amd_calc_l3_indices(); +} + +static ssize_t show_cache_disable(struct _cpuid4_info *this_leaf, char *buf, + unsigned int index) +{ + int cpu = cpumask_first(to_cpumask(this_leaf->shared_cpu_map)); + int node = amd_get_nb_id(cpu); + struct pci_dev *dev = node_to_k8_nb_misc(node); + unsigned int reg = 0; + + if (!this_leaf->can_disable) + return -EINVAL; + + if (!dev) + return -EINVAL; + + pci_read_config_dword(dev, 0x1BC + index * 4, ®); + return sprintf(buf, "0x%08x\n", reg); +} + +#define SHOW_CACHE_DISABLE(index) \ +static ssize_t \ +show_cache_disable_##index(struct _cpuid4_info *this_leaf, char *buf) \ +{ \ + return show_cache_disable(this_leaf, buf, index); \ +} +SHOW_CACHE_DISABLE(0) +SHOW_CACHE_DISABLE(1) + +static ssize_t store_cache_disable(struct _cpuid4_info *this_leaf, + const char *buf, size_t count, unsigned int index) +{ + int cpu = cpumask_first(to_cpumask(this_leaf->shared_cpu_map)); + int node = amd_get_nb_id(cpu); + struct pci_dev *dev = node_to_k8_nb_misc(node); + unsigned long val = 0; + +#define SUBCACHE_MASK (3UL << 20) +#define SUBCACHE_INDEX 0xfff + + if (!this_leaf->can_disable) + return -EINVAL; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (!dev) + return -EINVAL; + + if (strict_strtoul(buf, 10, &val) < 0) + return -EINVAL; + + /* do not allow writes outside of allowed bits */ + if ((val & ~(SUBCACHE_MASK | SUBCACHE_INDEX)) || + ((val & SUBCACHE_INDEX) > this_leaf->l3_indices)) + return -EINVAL; + + val |= BIT(30); + pci_write_config_dword(dev, 0x1BC + index * 4, val); + /* + * We need to WBINVD on a core on the node containing the L3 cache which + * indices we disable therefore a simple wbinvd() is not sufficient. + */ + wbinvd_on_cpu(cpu); + pci_write_config_dword(dev, 0x1BC + index * 4, val | BIT(31)); + return count; } +#define STORE_CACHE_DISABLE(index) \ +static ssize_t \ +store_cache_disable_##index(struct _cpuid4_info *this_leaf, \ + const char *buf, size_t count) \ +{ \ + return store_cache_disable(this_leaf, buf, count, index); \ +} +STORE_CACHE_DISABLE(0) +STORE_CACHE_DISABLE(1) + +static struct _cache_attr cache_disable_0 = __ATTR(cache_disable_0, 0644, + show_cache_disable_0, store_cache_disable_0); +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 */ +static void __cpuinit +amd_check_l3_disable(int index, struct _cpuid4_info_regs *this_leaf) +{ +}; +#endif /* CONFIG_CPU_SUP_AMD */ + static int __cpuinit cpuid4_cache_lookup_regs(int index, struct _cpuid4_info_regs *this_leaf) @@ -488,22 +617,6 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) #endif } - if (trace) - printk(KERN_INFO "CPU: Trace cache: %dK uops", trace); - else if (l1i) - printk(KERN_INFO "CPU: L1 I cache: %dK", l1i); - - if (l1d) - printk(KERN_CONT ", L1 D cache: %dK\n", l1d); - else - printk(KERN_CONT "\n"); - - if (l2) - printk(KERN_INFO "CPU: L2 cache: %dK\n", l2); - - if (l3) - printk(KERN_INFO "CPU: L3 cache: %dK\n", l3); - c->x86_cache_size = l3 ? l3 : (l2 ? l2 : (l1i+l1d)); return l2; @@ -512,26 +625,27 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) #ifdef CONFIG_SYSFS /* pointer to _cpuid4_info array (for each cache leaf) */ -static DEFINE_PER_CPU(struct _cpuid4_info *, cpuid4_info); -#define CPUID4_INFO_IDX(x, y) (&((per_cpu(cpuid4_info, x))[y])) +static DEFINE_PER_CPU(struct _cpuid4_info *, ici_cpuid4_info); +#define CPUID4_INFO_IDX(x, y) (&((per_cpu(ici_cpuid4_info, x))[y])) #ifdef CONFIG_SMP static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index) { struct _cpuid4_info *this_leaf, *sibling_leaf; unsigned long num_threads_sharing; - int index_msb, i; + int index_msb, i, sibling; struct cpuinfo_x86 *c = &cpu_data(cpu); if ((index == 3) && (c->x86_vendor == X86_VENDOR_AMD)) { - struct cpuinfo_x86 *d; - for_each_online_cpu(i) { - if (!per_cpu(cpuid4_info, i)) + for_each_cpu(i, c->llc_shared_map) { + if (!per_cpu(ici_cpuid4_info, i)) continue; - d = &cpu_data(i); this_leaf = CPUID4_INFO_IDX(i, index); - cpumask_copy(to_cpumask(this_leaf->shared_cpu_map), - d->llc_shared_map); + for_each_cpu(sibling, c->llc_shared_map) { + if (!cpu_online(sibling)) + continue; + set_bit(sibling, this_leaf->shared_cpu_map); + } } return; } @@ -548,7 +662,7 @@ static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index) c->apicid >> index_msb) { cpumask_set_cpu(i, to_cpumask(this_leaf->shared_cpu_map)); - if (i != cpu && per_cpu(cpuid4_info, i)) { + if (i != cpu && per_cpu(ici_cpuid4_info, i)) { sibling_leaf = CPUID4_INFO_IDX(i, index); cpumask_set_cpu(cpu, to_cpumask( @@ -587,8 +701,8 @@ static void __cpuinit free_cache_attributes(unsigned int cpu) for (i = 0; i < num_cache_leaves; i++) cache_remove_shared_cpu_map(cpu, i); - kfree(per_cpu(cpuid4_info, cpu)); - per_cpu(cpuid4_info, cpu) = NULL; + kfree(per_cpu(ici_cpuid4_info, cpu)); + per_cpu(ici_cpuid4_info, cpu) = NULL; } static int @@ -627,15 +741,15 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu) if (num_cache_leaves == 0) return -ENOENT; - per_cpu(cpuid4_info, cpu) = kzalloc( + per_cpu(ici_cpuid4_info, cpu) = kzalloc( sizeof(struct _cpuid4_info) * num_cache_leaves, GFP_KERNEL); - if (per_cpu(cpuid4_info, cpu) == NULL) + if (per_cpu(ici_cpuid4_info, cpu) == NULL) return -ENOMEM; smp_call_function_single(cpu, get_cpu_leaves, &retval, true); if (retval) { - kfree(per_cpu(cpuid4_info, cpu)); - per_cpu(cpuid4_info, cpu) = NULL; + kfree(per_cpu(ici_cpuid4_info, cpu)); + per_cpu(ici_cpuid4_info, cpu) = NULL; } return retval; @@ -647,7 +761,7 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu) extern struct sysdev_class cpu_sysdev_class; /* from drivers/base/cpu.c */ /* pointer to kobject for cpuX/cache */ -static DEFINE_PER_CPU(struct kobject *, cache_kobject); +static DEFINE_PER_CPU(struct kobject *, ici_cache_kobject); struct _index_kobject { struct kobject kobj; @@ -656,8 +770,8 @@ struct _index_kobject { }; /* pointer to array of kobjects for cpuX/cache/indexY */ -static DEFINE_PER_CPU(struct _index_kobject *, index_kobject); -#define INDEX_KOBJECT_PTR(x, y) (&((per_cpu(index_kobject, x))[y])) +static DEFINE_PER_CPU(struct _index_kobject *, ici_index_kobject); +#define INDEX_KOBJECT_PTR(x, y) (&((per_cpu(ici_index_kobject, x))[y])) #define show_one_plus(file_name, object, val) \ static ssize_t show_##file_name \ @@ -723,82 +837,6 @@ static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf) #define to_object(k) container_of(k, struct _index_kobject, kobj) #define to_attr(a) container_of(a, struct _cache_attr, attr) -static ssize_t show_cache_disable(struct _cpuid4_info *this_leaf, char *buf, - unsigned int index) -{ - int cpu = cpumask_first(to_cpumask(this_leaf->shared_cpu_map)); - int node = cpu_to_node(cpu); - struct pci_dev *dev = node_to_k8_nb_misc(node); - unsigned int reg = 0; - - if (!this_leaf->can_disable) - return -EINVAL; - - if (!dev) - return -EINVAL; - - pci_read_config_dword(dev, 0x1BC + index * 4, ®); - return sprintf(buf, "%x\n", reg); -} - -#define SHOW_CACHE_DISABLE(index) \ -static ssize_t \ -show_cache_disable_##index(struct _cpuid4_info *this_leaf, char *buf) \ -{ \ - return show_cache_disable(this_leaf, buf, index); \ -} -SHOW_CACHE_DISABLE(0) -SHOW_CACHE_DISABLE(1) - -static ssize_t store_cache_disable(struct _cpuid4_info *this_leaf, - const char *buf, size_t count, unsigned int index) -{ - int cpu = cpumask_first(to_cpumask(this_leaf->shared_cpu_map)); - int node = cpu_to_node(cpu); - struct pci_dev *dev = node_to_k8_nb_misc(node); - unsigned long val = 0; - unsigned int scrubber = 0; - - if (!this_leaf->can_disable) - return -EINVAL; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - if (!dev) - return -EINVAL; - - if (strict_strtoul(buf, 10, &val) < 0) - return -EINVAL; - - val |= 0xc0000000; - - pci_read_config_dword(dev, 0x58, &scrubber); - scrubber &= ~0x1f000000; - pci_write_config_dword(dev, 0x58, scrubber); - - pci_write_config_dword(dev, 0x1BC + index * 4, val & ~0x40000000); - wbinvd(); - pci_write_config_dword(dev, 0x1BC + index * 4, val); - return count; -} - -#define STORE_CACHE_DISABLE(index) \ -static ssize_t \ -store_cache_disable_##index(struct _cpuid4_info *this_leaf, \ - const char *buf, size_t count) \ -{ \ - return store_cache_disable(this_leaf, buf, count, index); \ -} -STORE_CACHE_DISABLE(0) -STORE_CACHE_DISABLE(1) - -struct _cache_attr { - struct attribute attr; - ssize_t (*show)(struct _cpuid4_info *, char *); - ssize_t (*store)(struct _cpuid4_info *, const char *, size_t count); -}; - #define define_one_ro(_name) \ static struct _cache_attr _name = \ __ATTR(_name, 0444, show_##_name, NULL) @@ -813,23 +851,28 @@ define_one_ro(size); define_one_ro(shared_cpu_map); define_one_ro(shared_cpu_list); -static struct _cache_attr cache_disable_0 = __ATTR(cache_disable_0, 0644, - show_cache_disable_0, store_cache_disable_0); -static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644, - show_cache_disable_1, store_cache_disable_1); +#define DEFAULT_SYSFS_CACHE_ATTRS \ + &type.attr, \ + &level.attr, \ + &coherency_line_size.attr, \ + &physical_line_partition.attr, \ + &ways_of_associativity.attr, \ + &number_of_sets.attr, \ + &size.attr, \ + &shared_cpu_map.attr, \ + &shared_cpu_list.attr static struct attribute *default_attrs[] = { - &type.attr, - &level.attr, - &coherency_line_size.attr, - &physical_line_partition.attr, - &ways_of_associativity.attr, - &number_of_sets.attr, - &size.attr, - &shared_cpu_map.attr, - &shared_cpu_list.attr, + DEFAULT_SYSFS_CACHE_ATTRS, + NULL +}; + +static struct attribute *default_l3_attrs[] = { + DEFAULT_SYSFS_CACHE_ATTRS, +#ifdef CONFIG_CPU_SUP_AMD &cache_disable_0.attr, &cache_disable_1.attr, +#endif NULL }; @@ -860,7 +903,7 @@ static ssize_t store(struct kobject *kobj, struct attribute *attr, return ret; } -static struct sysfs_ops sysfs_ops = { +static const struct sysfs_ops sysfs_ops = { .show = show, .store = store, }; @@ -876,10 +919,10 @@ static struct kobj_type ktype_percpu_entry = { static void __cpuinit cpuid4_cache_sysfs_exit(unsigned int cpu) { - kfree(per_cpu(cache_kobject, cpu)); - kfree(per_cpu(index_kobject, cpu)); - per_cpu(cache_kobject, cpu) = NULL; - per_cpu(index_kobject, cpu) = NULL; + kfree(per_cpu(ici_cache_kobject, cpu)); + kfree(per_cpu(ici_index_kobject, cpu)); + per_cpu(ici_cache_kobject, cpu) = NULL; + per_cpu(ici_index_kobject, cpu) = NULL; free_cache_attributes(cpu); } @@ -895,14 +938,14 @@ static int __cpuinit cpuid4_cache_sysfs_init(unsigned int cpu) return err; /* Allocate all required memory */ - per_cpu(cache_kobject, cpu) = + per_cpu(ici_cache_kobject, cpu) = kzalloc(sizeof(struct kobject), GFP_KERNEL); - if (unlikely(per_cpu(cache_kobject, cpu) == NULL)) + if (unlikely(per_cpu(ici_cache_kobject, cpu) == NULL)) goto err_out; - per_cpu(index_kobject, cpu) = kzalloc( + per_cpu(ici_index_kobject, cpu) = kzalloc( sizeof(struct _index_kobject) * num_cache_leaves, GFP_KERNEL); - if (unlikely(per_cpu(index_kobject, cpu) == NULL)) + if (unlikely(per_cpu(ici_index_kobject, cpu) == NULL)) goto err_out; return 0; @@ -920,13 +963,14 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev) unsigned int cpu = sys_dev->id; unsigned long i, j; struct _index_kobject *this_object; + struct _cpuid4_info *this_leaf; int retval; retval = cpuid4_cache_sysfs_init(cpu); if (unlikely(retval < 0)) return retval; - retval = kobject_init_and_add(per_cpu(cache_kobject, cpu), + retval = kobject_init_and_add(per_cpu(ici_cache_kobject, cpu), &ktype_percpu_entry, &sys_dev->kobj, "%s", "cache"); if (retval < 0) { @@ -938,14 +982,22 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev) this_object = INDEX_KOBJECT_PTR(cpu, i); this_object->cpu = cpu; this_object->index = i; + + this_leaf = CPUID4_INFO_IDX(cpu, i); + + if (this_leaf->can_disable) + ktype_cache.default_attrs = default_l3_attrs; + else + ktype_cache.default_attrs = default_attrs; + retval = kobject_init_and_add(&(this_object->kobj), &ktype_cache, - per_cpu(cache_kobject, cpu), + per_cpu(ici_cache_kobject, cpu), "index%1lu", i); if (unlikely(retval)) { for (j = 0; j < i; j++) kobject_put(&(INDEX_KOBJECT_PTR(cpu, j)->kobj)); - kobject_put(per_cpu(cache_kobject, cpu)); + kobject_put(per_cpu(ici_cache_kobject, cpu)); cpuid4_cache_sysfs_exit(cpu); return retval; } @@ -953,7 +1005,7 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev) } cpumask_set_cpu(cpu, to_cpumask(cache_dev_map)); - kobject_uevent(per_cpu(cache_kobject, cpu), KOBJ_ADD); + kobject_uevent(per_cpu(ici_cache_kobject, cpu), KOBJ_ADD); return 0; } @@ -962,7 +1014,7 @@ static void __cpuinit cache_remove_dev(struct sys_device * sys_dev) unsigned int cpu = sys_dev->id; unsigned long i; - if (per_cpu(cpuid4_info, cpu) == NULL) + if (per_cpu(ici_cpuid4_info, cpu) == NULL) return; if (!cpumask_test_cpu(cpu, to_cpumask(cache_dev_map))) return; @@ -970,7 +1022,7 @@ static void __cpuinit cache_remove_dev(struct sys_device * sys_dev) for (i = 0; i < num_cache_leaves; i++) kobject_put(&(INDEX_KOBJECT_PTR(cpu, i)->kobj)); - kobject_put(per_cpu(cache_kobject, cpu)); + kobject_put(per_cpu(ici_cache_kobject, cpu)); cpuid4_cache_sysfs_exit(cpu); } diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c index 472763d92098..73734baa50f2 100644 --- a/arch/x86/kernel/cpu/mcheck/mce-inject.c +++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c @@ -74,7 +74,7 @@ static void raise_exception(struct mce *m, struct pt_regs *pregs) m->finished = 0; } -static cpumask_t mce_inject_cpumask; +static cpumask_var_t mce_inject_cpumask; static int mce_raise_notify(struct notifier_block *self, unsigned long val, void *data) @@ -82,9 +82,9 @@ static int mce_raise_notify(struct notifier_block *self, struct die_args *args = (struct die_args *)data; int cpu = smp_processor_id(); struct mce *m = &__get_cpu_var(injectm); - if (val != DIE_NMI_IPI || !cpu_isset(cpu, mce_inject_cpumask)) + if (val != DIE_NMI_IPI || !cpumask_test_cpu(cpu, mce_inject_cpumask)) return NOTIFY_DONE; - cpu_clear(cpu, mce_inject_cpumask); + cpumask_clear_cpu(cpu, mce_inject_cpumask); if (m->inject_flags & MCJ_EXCEPTION) raise_exception(m, args->regs); else if (m->status) @@ -148,22 +148,22 @@ static void raise_mce(struct mce *m) unsigned long start; int cpu; get_online_cpus(); - mce_inject_cpumask = cpu_online_map; - cpu_clear(get_cpu(), mce_inject_cpumask); + cpumask_copy(mce_inject_cpumask, cpu_online_mask); + cpumask_clear_cpu(get_cpu(), mce_inject_cpumask); for_each_online_cpu(cpu) { struct mce *mcpu = &per_cpu(injectm, cpu); if (!mcpu->finished || MCJ_CTX(mcpu->inject_flags) != MCJ_CTX_RANDOM) - cpu_clear(cpu, mce_inject_cpumask); + cpumask_clear_cpu(cpu, mce_inject_cpumask); } - if (!cpus_empty(mce_inject_cpumask)) - apic->send_IPI_mask(&mce_inject_cpumask, NMI_VECTOR); + if (!cpumask_empty(mce_inject_cpumask)) + apic->send_IPI_mask(mce_inject_cpumask, NMI_VECTOR); start = jiffies; - while (!cpus_empty(mce_inject_cpumask)) { + while (!cpumask_empty(mce_inject_cpumask)) { if (!time_before(jiffies, start + 2*HZ)) { printk(KERN_ERR "Timeout waiting for mce inject NMI %lx\n", - *cpus_addr(mce_inject_cpumask)); + *cpumask_bits(mce_inject_cpumask)); break; } cpu_relax(); @@ -210,6 +210,8 @@ static ssize_t mce_write(struct file *filp, const char __user *ubuf, static int inject_init(void) { + if (!alloc_cpumask_var(&mce_inject_cpumask, GFP_KERNEL)) + return -ENOMEM; printk(KERN_INFO "Machine check injector initialized\n"); mce_chrdev_ops.write = mce_write; register_die_notifier(&mce_raise_nb); diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 4b2af86e3e8d..3ab9c886b613 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -46,6 +46,16 @@ #include "mce-internal.h" +static DEFINE_MUTEX(mce_read_mutex); + +#define rcu_dereference_check_mce(p) \ + rcu_dereference_check((p), \ + rcu_read_lock_sched_held() || \ + lockdep_is_held(&mce_read_mutex)) + +#define CREATE_TRACE_POINTS +#include <trace/events/mce.h> + int mce_disabled __read_mostly; #define MISC_MCELOG_MINOR 227 @@ -85,6 +95,26 @@ static DECLARE_WAIT_QUEUE_HEAD(mce_wait); static DEFINE_PER_CPU(struct mce, mces_seen); static int cpu_missing; +/* + * CPU/chipset specific EDAC code can register a notifier call here to print + * MCE errors in a human-readable form. + */ +ATOMIC_NOTIFIER_HEAD(x86_mce_decoder_chain); +EXPORT_SYMBOL_GPL(x86_mce_decoder_chain); + +static int default_decode_mce(struct notifier_block *nb, unsigned long val, + void *data) +{ + pr_emerg("No human readable MCE decoding support on this CPU type.\n"); + pr_emerg("Run the message through 'mcelog --ascii' to decode.\n"); + + return NOTIFY_STOP; +} + +static struct notifier_block mce_dec_nb = { + .notifier_call = default_decode_mce, + .priority = -1, +}; /* MCA banks polled by the period polling timer for corrected events */ DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = { @@ -129,10 +159,13 @@ void mce_log(struct mce *mce) { unsigned next, entry; + /* Emit the trace record: */ + trace_mce_record(mce); + mce->finished = 0; wmb(); for (;;) { - entry = rcu_dereference(mcelog.next); + entry = rcu_dereference_check_mce(mcelog.next); for (;;) { /* * When the buffer fills up discard new entries. @@ -165,49 +198,46 @@ void mce_log(struct mce *mce) set_bit(0, &mce_need_notify); } -void __weak decode_mce(struct mce *m) -{ - return; -} - static void print_mce(struct mce *m) { - printk(KERN_EMERG - "CPU %d: Machine Check Exception: %16Lx Bank %d: %016Lx\n", + pr_emerg("CPU %d: Machine Check Exception: %16Lx Bank %d: %016Lx\n", m->extcpu, m->mcgstatus, m->bank, m->status); + if (m->ip) { - printk(KERN_EMERG "RIP%s %02x:<%016Lx> ", - !(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "", - m->cs, m->ip); + pr_emerg("RIP%s %02x:<%016Lx> ", + !(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "", + m->cs, m->ip); + if (m->cs == __KERNEL_CS) print_symbol("{%s}", m->ip); - printk(KERN_CONT "\n"); + pr_cont("\n"); } - printk(KERN_EMERG "TSC %llx ", m->tsc); + + pr_emerg("TSC %llx ", m->tsc); if (m->addr) - printk(KERN_CONT "ADDR %llx ", m->addr); + pr_cont("ADDR %llx ", m->addr); if (m->misc) - printk(KERN_CONT "MISC %llx ", m->misc); - printk(KERN_CONT "\n"); - printk(KERN_EMERG "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x\n", - m->cpuvendor, m->cpuid, m->time, m->socketid, - m->apicid); + pr_cont("MISC %llx ", m->misc); + + pr_cont("\n"); + pr_emerg("PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x\n", + m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid); - decode_mce(m); + /* + * Print out human-readable details about the MCE error, + * (if the CPU has an implementation for that) + */ + atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, m); } static void print_mce_head(void) { - printk(KERN_EMERG "\nHARDWARE ERROR\n"); + pr_emerg("\nHARDWARE ERROR\n"); } static void print_mce_tail(void) { - printk(KERN_EMERG "This is not a software problem!\n" -#if (!defined(CONFIG_EDAC) || !defined(CONFIG_CPU_SUP_AMD)) - "Run through mcelog --ascii to decode and contact your hardware vendor\n" -#endif - ); + pr_emerg("This is not a software problem!\n"); } #define PANIC_TIMEOUT 5 /* 5 seconds */ @@ -221,6 +251,7 @@ static atomic_t mce_fake_paniced; static void wait_for_panic(void) { long timeout = PANIC_TIMEOUT*USEC_PER_SEC; + preempt_disable(); local_irq_enable(); while (timeout-- > 0) @@ -288,6 +319,7 @@ static void mce_panic(char *msg, struct mce *final, char *exp) static int msr_to_offset(u32 msr) { unsigned bank = __get_cpu_var(injectm.bank); + if (msr == rip_msr) return offsetof(struct mce, ip); if (msr == MSR_IA32_MCx_STATUS(bank)) @@ -1111,7 +1143,7 @@ static int check_interval = 5 * 60; /* 5 minutes */ static DEFINE_PER_CPU(int, mce_next_interval); /* in jiffies */ static DEFINE_PER_CPU(struct timer_list, mce_timer); -static void mcheck_timer(unsigned long data) +static void mce_start_timer(unsigned long data) { struct timer_list *t = &per_cpu(mce_timer, data); int *n; @@ -1176,7 +1208,7 @@ int mce_notify_irq(void) } EXPORT_SYMBOL_GPL(mce_notify_irq); -static int mce_banks_init(void) +static int __cpuinit __mcheck_cpu_mce_banks_init(void) { int i; @@ -1195,7 +1227,7 @@ static int mce_banks_init(void) /* * Initialize Machine Checks for a CPU. */ -static int __cpuinit mce_cap_init(void) +static int __cpuinit __mcheck_cpu_cap_init(void) { unsigned b; u64 cap; @@ -1203,7 +1235,8 @@ static int __cpuinit mce_cap_init(void) rdmsrl(MSR_IA32_MCG_CAP, cap); b = cap & MCG_BANKCNT_MASK; - printk(KERN_INFO "mce: CPU supports %d MCE banks\n", b); + if (!banks) + printk(KERN_INFO "mce: CPU supports %d MCE banks\n", b); if (b > MAX_NR_BANKS) { printk(KERN_WARNING @@ -1216,7 +1249,7 @@ static int __cpuinit mce_cap_init(void) WARN_ON(banks != 0 && b != banks); banks = b; if (!mce_banks) { - int err = mce_banks_init(); + int err = __mcheck_cpu_mce_banks_init(); if (err) return err; @@ -1232,7 +1265,7 @@ static int __cpuinit mce_cap_init(void) return 0; } -static void mce_init(void) +static void __mcheck_cpu_init_generic(void) { mce_banks_t all_banks; u64 cap; @@ -1261,7 +1294,7 @@ static void mce_init(void) } /* Add per CPU specific workarounds here */ -static int __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c) +static int __cpuinit __mcheck_cpu_apply_quirks(struct cpuinfo_x86 *c) { if (c->x86_vendor == X86_VENDOR_UNKNOWN) { pr_info("MCE: unknown CPU type - not enabling MCE support.\n"); @@ -1329,7 +1362,7 @@ static int __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c) return 0; } -static void __cpuinit mce_ancient_init(struct cpuinfo_x86 *c) +static void __cpuinit __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c) { if (c->x86 != 5) return; @@ -1343,7 +1376,7 @@ static void __cpuinit mce_ancient_init(struct cpuinfo_x86 *c) } } -static void mce_cpu_features(struct cpuinfo_x86 *c) +static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c) { switch (c->x86_vendor) { case X86_VENDOR_INTEL: @@ -1357,18 +1390,19 @@ static void mce_cpu_features(struct cpuinfo_x86 *c) } } -static void mce_init_timer(void) +static void __mcheck_cpu_init_timer(void) { struct timer_list *t = &__get_cpu_var(mce_timer); int *n = &__get_cpu_var(mce_next_interval); + setup_timer(t, mce_start_timer, smp_processor_id()); + if (mce_ignore_ce) return; *n = check_interval * HZ; if (!*n) return; - setup_timer(t, mcheck_timer, smp_processor_id()); t->expires = round_jiffies(jiffies + *n); add_timer_on(t, smp_processor_id()); } @@ -1388,27 +1422,28 @@ void (*machine_check_vector)(struct pt_regs *, long error_code) = * Called for each booted CPU to set up machine checks. * Must be called with preempt off: */ -void __cpuinit mcheck_init(struct cpuinfo_x86 *c) +void __cpuinit mcheck_cpu_init(struct cpuinfo_x86 *c) { if (mce_disabled) return; - mce_ancient_init(c); + __mcheck_cpu_ancient_init(c); if (!mce_available(c)) return; - if (mce_cap_init() < 0 || mce_cpu_quirks(c) < 0) { + if (__mcheck_cpu_cap_init() < 0 || __mcheck_cpu_apply_quirks(c) < 0) { mce_disabled = 1; return; } machine_check_vector = do_machine_check; - mce_init(); - mce_cpu_features(c); - mce_init_timer(); + __mcheck_cpu_init_generic(); + __mcheck_cpu_init_vendor(c); + __mcheck_cpu_init_timer(); INIT_WORK(&__get_cpu_var(mce_work), mce_process_work); + } /* @@ -1457,8 +1492,6 @@ static void collect_tscs(void *data) rdtscll(cpu_tsc[smp_processor_id()]); } -static DEFINE_MUTEX(mce_read_mutex); - static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, loff_t *off) { @@ -1472,7 +1505,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, return -ENOMEM; mutex_lock(&mce_read_mutex); - next = rcu_dereference(mcelog.next); + next = rcu_dereference_check_mce(mcelog.next); /* Only supports full reads right now */ if (*off != 0 || usize < MCE_LOG_LEN*sizeof(struct mce)) { @@ -1537,7 +1570,7 @@ timeout: static unsigned int mce_poll(struct file *file, poll_table *wait) { poll_wait(file, &mce_wait, wait); - if (rcu_dereference(mcelog.next)) + if (rcu_dereference_check_mce(mcelog.next)) return POLLIN | POLLRDNORM; return 0; } @@ -1628,6 +1661,15 @@ static int __init mcheck_enable(char *str) } __setup("mce", mcheck_enable); +int __init mcheck_init(void) +{ + atomic_notifier_chain_register(&x86_mce_decoder_chain, &mce_dec_nb); + + mcheck_intel_therm_init(); + + return 0; +} + /* * Sysfs support */ @@ -1636,7 +1678,7 @@ __setup("mce", mcheck_enable); * Disable machine checks on suspend and shutdown. We can't really handle * them later. */ -static int mce_disable(void) +static int mce_disable_error_reporting(void) { int i; @@ -1651,12 +1693,12 @@ static int mce_disable(void) static int mce_suspend(struct sys_device *dev, pm_message_t state) { - return mce_disable(); + return mce_disable_error_reporting(); } static int mce_shutdown(struct sys_device *dev) { - return mce_disable(); + return mce_disable_error_reporting(); } /* @@ -1666,8 +1708,8 @@ static int mce_shutdown(struct sys_device *dev) */ static int mce_resume(struct sys_device *dev) { - mce_init(); - mce_cpu_features(¤t_cpu_data); + __mcheck_cpu_init_generic(); + __mcheck_cpu_init_vendor(¤t_cpu_data); return 0; } @@ -1677,8 +1719,8 @@ static void mce_cpu_restart(void *data) del_timer_sync(&__get_cpu_var(mce_timer)); if (!mce_available(¤t_cpu_data)) return; - mce_init(); - mce_init_timer(); + __mcheck_cpu_init_generic(); + __mcheck_cpu_init_timer(); } /* Reinit MCEs after user configuration changes */ @@ -1704,7 +1746,7 @@ static void mce_enable_ce(void *all) cmci_reenable(); cmci_recheck(); if (all) - mce_init_timer(); + __mcheck_cpu_init_timer(); } static struct sysdev_class mce_sysclass = { @@ -1892,7 +1934,7 @@ error2: sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[j].attr); error: while (--i >= 0) - sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[i].attr); + sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]); sysdev_unregister(&per_cpu(mce_dev, cpu)); @@ -1917,13 +1959,14 @@ static __cpuinit void mce_remove_device(unsigned int cpu) } /* Make sure there are no machine checks on offlined CPUs. */ -static void mce_disable_cpu(void *h) +static void __cpuinit mce_disable_cpu(void *h) { unsigned long action = *(unsigned long *)h; int i; if (!mce_available(¤t_cpu_data)) return; + if (!(action & CPU_TASKS_FROZEN)) cmci_clear(); for (i = 0; i < banks; i++) { @@ -1934,7 +1977,7 @@ static void mce_disable_cpu(void *h) } } -static void mce_reenable_cpu(void *h) +static void __cpuinit mce_reenable_cpu(void *h) { unsigned long action = *(unsigned long *)h; int i; @@ -1979,9 +2022,11 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) break; case CPU_DOWN_FAILED: case CPU_DOWN_FAILED_FROZEN: - t->expires = round_jiffies(jiffies + + if (!mce_ignore_ce && check_interval) { + t->expires = round_jiffies(jiffies + __get_cpu_var(mce_next_interval)); - add_timer_on(t, cpu); + add_timer_on(t, cpu); + } smp_call_function_single(cpu, mce_reenable_cpu, &action, 1); break; case CPU_POST_DEAD: @@ -2004,6 +2049,7 @@ static __init void mce_init_banks(void) struct mce_bank *b = &mce_banks[i]; struct sysdev_attribute *a = &b->attr; + sysfs_attr_init(&a->attr); a->attr.name = b->attrname; snprintf(b->attrname, ATTR_LEN, "bank%d", i); @@ -2013,7 +2059,7 @@ static __init void mce_init_banks(void) } } -static __init int mce_init_device(void) +static __init int mcheck_init_device(void) { int err; int i = 0; @@ -2041,7 +2087,7 @@ static __init int mce_init_device(void) return err; } -device_initcall(mce_init_device); +device_initcall(mcheck_init_device); /* * Old style boot options parsing. Only for compatibility. @@ -2089,7 +2135,7 @@ static int fake_panic_set(void *data, u64 val) DEFINE_SIMPLE_ATTRIBUTE(fake_panic_fops, fake_panic_get, fake_panic_set, "%llu\n"); -static int __init mce_debugfs_init(void) +static int __init mcheck_debugfs_init(void) { struct dentry *dmce, *ffake_panic; @@ -2103,5 +2149,5 @@ static int __init mce_debugfs_init(void) return 0; } -late_initcall(mce_debugfs_init); +late_initcall(mcheck_debugfs_init); #endif diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c index 83a3d1f4efca..cda932ca3ade 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c @@ -388,7 +388,7 @@ static ssize_t store(struct kobject *kobj, struct attribute *attr, return ret; } -static struct sysfs_ops threshold_ops = { +static const struct sysfs_ops threshold_ops = { .show = show, .store = store, }; diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c index 889f665fe93d..d15df6e49bf0 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c @@ -8,6 +8,7 @@ #include <linux/init.h> #include <linux/interrupt.h> #include <linux/percpu.h> +#include <linux/sched.h> #include <asm/apic.h> #include <asm/processor.h> #include <asm/msr.h> @@ -94,7 +95,7 @@ static void cmci_discover(int banks, int boot) /* Already owned by someone else? */ if (val & CMCI_EN) { - if (test_and_clear_bit(i, owned) || boot) + if (test_and_clear_bit(i, owned) && !boot) print_update("SHD", &hdr, i); __clear_bit(i, __get_cpu_var(mce_poll_banks)); continue; @@ -106,7 +107,7 @@ static void cmci_discover(int banks, int boot) /* Did the enable bit stick? -- the bank supports CMCI */ if (val & CMCI_EN) { - if (!test_and_set_bit(i, owned) || boot) + if (!test_and_set_bit(i, owned) && !boot) print_update("CMCI", &hdr, i); __clear_bit(i, __get_cpu_var(mce_poll_banks)); } else { diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index b3a1dba75330..81c499eceb21 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -49,6 +49,8 @@ static DEFINE_PER_CPU(struct thermal_state, thermal_state); static atomic_t therm_throt_en = ATOMIC_INIT(0); +static u32 lvtthmr_init __read_mostly; + #ifdef CONFIG_SYSFS #define define_therm_throt_sysdev_one_ro(_name) \ static SYSDEV_ATTR(_name, 0444, therm_throt_sysdev_show_##_name, NULL) @@ -254,14 +256,34 @@ asmlinkage void smp_thermal_interrupt(struct pt_regs *regs) ack_APIC_irq(); } +/* Thermal monitoring depends on APIC, ACPI and clock modulation */ +static int intel_thermal_supported(struct cpuinfo_x86 *c) +{ + if (!cpu_has_apic) + return 0; + if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC)) + return 0; + return 1; +} + +void __init mcheck_intel_therm_init(void) +{ + /* + * This function is only called on boot CPU. Save the init thermal + * LVT value on BSP and use that value to restore APs' thermal LVT + * entry BIOS programmed later + */ + if (intel_thermal_supported(&boot_cpu_data)) + lvtthmr_init = apic_read(APIC_LVTTHMR); +} + void intel_init_thermal(struct cpuinfo_x86 *c) { unsigned int cpu = smp_processor_id(); int tm2 = 0; u32 l, h; - /* Thermal monitoring depends on ACPI and clock modulation*/ - if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC)) + if (!intel_thermal_supported(c)) return; /* @@ -270,7 +292,20 @@ void intel_init_thermal(struct cpuinfo_x86 *c) * since it might be delivered via SMI already: */ rdmsr(MSR_IA32_MISC_ENABLE, l, h); - h = apic_read(APIC_LVTTHMR); + + /* + * The initial value of thermal LVT entries on all APs always reads + * 0x10000 because APs are woken up by BSP issuing INIT-SIPI-SIPI + * sequence to them and LVT registers are reset to 0s except for + * the mask bits which are set to 1s when APs receive INIT IPI. + * Always restore the value that BIOS has programmed on AP based on + * BSP's info we saved since BIOS is always setting the same value + * for all threads/cores + */ + apic_write(APIC_LVTTHMR, lvtthmr_init); + + h = lvtthmr_init; + if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { printk(KERN_DEBUG "CPU%d: Thermal monitoring handled by SMI\n", cpu); @@ -312,8 +347,8 @@ void intel_init_thermal(struct cpuinfo_x86 *c) l = apic_read(APIC_LVTTHMR); apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); - printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n", - cpu, tm2 ? "TM2" : "TM1"); + printk_once(KERN_INFO "CPU0: Thermal monitoring enabled (%s)\n", + tm2 ? "TM2" : "TM1"); /* enable thermal throttle processing */ atomic_set(&therm_throt_en, 1); diff --git a/arch/x86/kernel/cpu/mtrr/Makefile b/arch/x86/kernel/cpu/mtrr/Makefile index f4361b56f8e9..ad9e5ed81181 100644 --- a/arch/x86/kernel/cpu/mtrr/Makefile +++ b/arch/x86/kernel/cpu/mtrr/Makefile @@ -1,3 +1,3 @@ -obj-y := main.o if.o generic.o state.o cleanup.o +obj-y := main.o if.o generic.o cleanup.o obj-$(CONFIG_X86_32) += amd.o cyrix.o centaur.o diff --git a/arch/x86/kernel/cpu/mtrr/amd.c b/arch/x86/kernel/cpu/mtrr/amd.c index 33af14110dfd..92ba9cd31c9a 100644 --- a/arch/x86/kernel/cpu/mtrr/amd.c +++ b/arch/x86/kernel/cpu/mtrr/amd.c @@ -108,7 +108,7 @@ amd_validate_add_page(unsigned long base, unsigned long size, unsigned int type) return 0; } -static struct mtrr_ops amd_mtrr_ops = { +static const struct mtrr_ops amd_mtrr_ops = { .vendor = X86_VENDOR_AMD, .set = amd_set_mtrr, .get = amd_get_mtrr, diff --git a/arch/x86/kernel/cpu/mtrr/centaur.c b/arch/x86/kernel/cpu/mtrr/centaur.c index de89f14eff3a..316fe3e60a97 100644 --- a/arch/x86/kernel/cpu/mtrr/centaur.c +++ b/arch/x86/kernel/cpu/mtrr/centaur.c @@ -110,7 +110,7 @@ centaur_validate_add_page(unsigned long base, unsigned long size, unsigned int t return 0; } -static struct mtrr_ops centaur_mtrr_ops = { +static const struct mtrr_ops centaur_mtrr_ops = { .vendor = X86_VENDOR_CENTAUR, .set = centaur_set_mcr, .get = centaur_get_mcr, diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c index 315738c74aad..06130b52f012 100644 --- a/arch/x86/kernel/cpu/mtrr/cleanup.c +++ b/arch/x86/kernel/cpu/mtrr/cleanup.c @@ -22,10 +22,10 @@ #include <linux/pci.h> #include <linux/smp.h> #include <linux/cpu.h> -#include <linux/sort.h> #include <linux/mutex.h> #include <linux/uaccess.h> #include <linux/kvm_para.h> +#include <linux/range.h> #include <asm/processor.h> #include <asm/e820.h> @@ -34,11 +34,6 @@ #include "mtrr.h" -struct res_range { - unsigned long start; - unsigned long end; -}; - struct var_mtrr_range_state { unsigned long base_pfn; unsigned long size_pfn; @@ -56,7 +51,7 @@ struct var_mtrr_state { /* Should be related to MTRR_VAR_RANGES nums */ #define RANGE_NUM 256 -static struct res_range __initdata range[RANGE_NUM]; +static struct range __initdata range[RANGE_NUM]; static int __initdata nr_range; static struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; @@ -64,117 +59,11 @@ static struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; static int __initdata debug_print; #define Dprintk(x...) do { if (debug_print) printk(KERN_DEBUG x); } while (0) - -static int __init -add_range(struct res_range *range, int nr_range, - unsigned long start, unsigned long end) -{ - /* Out of slots: */ - if (nr_range >= RANGE_NUM) - return nr_range; - - range[nr_range].start = start; - range[nr_range].end = end; - - nr_range++; - - return nr_range; -} - -static int __init -add_range_with_merge(struct res_range *range, int nr_range, - unsigned long start, unsigned long end) -{ - int i; - - /* Try to merge it with old one: */ - for (i = 0; i < nr_range; i++) { - unsigned long final_start, final_end; - unsigned long common_start, common_end; - - if (!range[i].end) - continue; - - common_start = max(range[i].start, start); - common_end = min(range[i].end, end); - if (common_start > common_end + 1) - continue; - - final_start = min(range[i].start, start); - final_end = max(range[i].end, end); - - range[i].start = final_start; - range[i].end = final_end; - return nr_range; - } - - /* Need to add it: */ - return add_range(range, nr_range, start, end); -} - -static void __init -subtract_range(struct res_range *range, unsigned long start, unsigned long end) -{ - int i, j; - - for (j = 0; j < RANGE_NUM; j++) { - if (!range[j].end) - continue; - - if (start <= range[j].start && end >= range[j].end) { - range[j].start = 0; - range[j].end = 0; - continue; - } - - if (start <= range[j].start && end < range[j].end && - range[j].start < end + 1) { - range[j].start = end + 1; - continue; - } - - - if (start > range[j].start && end >= range[j].end && - range[j].end > start - 1) { - range[j].end = start - 1; - continue; - } - - if (start > range[j].start && end < range[j].end) { - /* Find the new spare: */ - for (i = 0; i < RANGE_NUM; i++) { - if (range[i].end == 0) - break; - } - if (i < RANGE_NUM) { - range[i].end = range[j].end; - range[i].start = end + 1; - } else { - printk(KERN_ERR "run of slot in ranges\n"); - } - range[j].end = start - 1; - continue; - } - } -} - -static int __init cmp_range(const void *x1, const void *x2) -{ - const struct res_range *r1 = x1; - const struct res_range *r2 = x2; - long start1, start2; - - start1 = r1->start; - start2 = r2->start; - - return start1 - start2; -} - #define BIOS_BUG_MSG KERN_WARNING \ "WARNING: BIOS bug: VAR MTRR %d contains strange UC entry under 1M, check with your system vendor!\n" static int __init -x86_get_mtrr_mem_range(struct res_range *range, int nr_range, +x86_get_mtrr_mem_range(struct range *range, int nr_range, unsigned long extra_remove_base, unsigned long extra_remove_size) { @@ -188,14 +77,14 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range, continue; base = range_state[i].base_pfn; size = range_state[i].size_pfn; - nr_range = add_range_with_merge(range, nr_range, base, - base + size - 1); + nr_range = add_range_with_merge(range, RANGE_NUM, nr_range, + base, base + size); } if (debug_print) { printk(KERN_DEBUG "After WB checking\n"); for (i = 0; i < nr_range; i++) - printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", - range[i].start, range[i].end + 1); + printk(KERN_DEBUG "MTRR MAP PFN: %016llx - %016llx\n", + range[i].start, range[i].end); } /* Take out UC ranges: */ @@ -217,51 +106,43 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range, size -= (1<<(20-PAGE_SHIFT)) - base; base = 1<<(20-PAGE_SHIFT); } - subtract_range(range, base, base + size - 1); + subtract_range(range, RANGE_NUM, base, base + size); } if (extra_remove_size) - subtract_range(range, extra_remove_base, - extra_remove_base + extra_remove_size - 1); + subtract_range(range, RANGE_NUM, extra_remove_base, + extra_remove_base + extra_remove_size); - /* get new range num */ - nr_range = 0; - for (i = 0; i < RANGE_NUM; i++) { - if (!range[i].end) - continue; - nr_range++; - } if (debug_print) { printk(KERN_DEBUG "After UC checking\n"); - for (i = 0; i < nr_range; i++) - printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", - range[i].start, range[i].end + 1); + for (i = 0; i < RANGE_NUM; i++) { + if (!range[i].end) + continue; + printk(KERN_DEBUG "MTRR MAP PFN: %016llx - %016llx\n", + range[i].start, range[i].end); + } } /* sort the ranges */ - sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); + nr_range = clean_sort_range(range, RANGE_NUM); if (debug_print) { printk(KERN_DEBUG "After sorting\n"); for (i = 0; i < nr_range; i++) - printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", - range[i].start, range[i].end + 1); + printk(KERN_DEBUG "MTRR MAP PFN: %016llx - %016llx\n", + range[i].start, range[i].end); } - /* clear those is not used */ - for (i = nr_range; i < RANGE_NUM; i++) - memset(&range[i], 0, sizeof(range[i])); - return nr_range; } #ifdef CONFIG_MTRR_SANITIZER -static unsigned long __init sum_ranges(struct res_range *range, int nr_range) +static unsigned long __init sum_ranges(struct range *range, int nr_range) { unsigned long sum = 0; int i; for (i = 0; i < nr_range; i++) - sum += range[i].end + 1 - range[i].start; + sum += range[i].end - range[i].start; return sum; } @@ -590,7 +471,7 @@ static int __init parse_mtrr_spare_reg(char *arg) early_param("mtrr_spare_reg_nr", parse_mtrr_spare_reg); static int __init -x86_setup_var_mtrrs(struct res_range *range, int nr_range, +x86_setup_var_mtrrs(struct range *range, int nr_range, u64 chunk_size, u64 gran_size) { struct var_mtrr_state var_state; @@ -608,7 +489,7 @@ x86_setup_var_mtrrs(struct res_range *range, int nr_range, /* Write the range: */ for (i = 0; i < nr_range; i++) { set_var_mtrr_range(&var_state, range[i].start, - range[i].end - range[i].start + 1); + range[i].end - range[i].start); } /* Write the last range: */ @@ -689,8 +570,6 @@ static int __init mtrr_need_cleanup(void) continue; if (!size) type = MTRR_NUM_TYPES; - if (type == MTRR_TYPE_WRPROT) - type = MTRR_TYPE_UNCACHABLE; num[type]++; } @@ -713,7 +592,7 @@ mtrr_calc_range_state(u64 chunk_size, u64 gran_size, unsigned long x_remove_base, unsigned long x_remove_size, int i) { - static struct res_range range_new[RANGE_NUM]; + static struct range range_new[RANGE_NUM]; unsigned long range_sums_new; static int nr_range_new; int num_reg; @@ -840,13 +719,13 @@ int __init mtrr_cleanup(unsigned address_bits) * [0, 1M) should always be covered by var mtrr with WB * and fixed mtrrs should take effect before var mtrr for it: */ - nr_range = add_range_with_merge(range, nr_range, 0, - (1ULL<<(20 - PAGE_SHIFT)) - 1); + nr_range = add_range_with_merge(range, RANGE_NUM, nr_range, 0, + 1ULL<<(20 - PAGE_SHIFT)); /* Sort the ranges: */ - sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); + sort_range(range, nr_range); range_sums = sum_ranges(range, nr_range); - printk(KERN_INFO "total RAM coverred: %ldM\n", + printk(KERN_INFO "total RAM covered: %ldM\n", range_sums >> (20 - PAGE_SHIFT)); if (mtrr_chunk_size && mtrr_gran_size) { @@ -1060,9 +939,9 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) nr_range = 0; if (mtrr_tom2) { range[nr_range].start = (1ULL<<(32 - PAGE_SHIFT)); - range[nr_range].end = (mtrr_tom2 >> PAGE_SHIFT) - 1; - if (highest_pfn < range[nr_range].end + 1) - highest_pfn = range[nr_range].end + 1; + range[nr_range].end = mtrr_tom2 >> PAGE_SHIFT; + if (highest_pfn < range[nr_range].end) + highest_pfn = range[nr_range].end; nr_range++; } nr_range = x86_get_mtrr_mem_range(range, nr_range, 0, 0); @@ -1074,15 +953,15 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) /* Check the holes: */ for (i = 0; i < nr_range - 1; i++) { - if (range[i].end + 1 < range[i+1].start) - total_trim_size += real_trim_memory(range[i].end + 1, + if (range[i].end < range[i+1].start) + total_trim_size += real_trim_memory(range[i].end, range[i+1].start); } /* Check the top: */ i = nr_range - 1; - if (range[i].end + 1 < end_pfn) - total_trim_size += real_trim_memory(range[i].end + 1, + if (range[i].end < end_pfn) + total_trim_size += real_trim_memory(range[i].end, end_pfn); if (total_trim_size) { diff --git a/arch/x86/kernel/cpu/mtrr/cyrix.c b/arch/x86/kernel/cpu/mtrr/cyrix.c index 228d982ce09c..68a3343e5798 100644 --- a/arch/x86/kernel/cpu/mtrr/cyrix.c +++ b/arch/x86/kernel/cpu/mtrr/cyrix.c @@ -265,7 +265,7 @@ static void cyrix_set_all(void) post_set(); } -static struct mtrr_ops cyrix_mtrr_ops = { +static const struct mtrr_ops cyrix_mtrr_ops = { .vendor = X86_VENDOR_CYRIX, .set_all = cyrix_set_all, .set = cyrix_set_arr, diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 55da0c5f68dd..9aa5dc76ff4a 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -464,7 +464,7 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base, tmp |= ~((1<<(hi - 1)) - 1); if (tmp != mask_lo) { - WARN_ONCE(1, KERN_INFO "mtrr: your BIOS has set up an incorrect mask, fixing it up.\n"); + printk(KERN_WARNING "mtrr: your BIOS has configured an incorrect mask, fixing it.\n"); mask_lo = tmp; } } @@ -570,7 +570,7 @@ static unsigned long set_mtrr_state(void) static unsigned long cr4; -static DEFINE_SPINLOCK(set_atomicity_lock); +static DEFINE_RAW_SPINLOCK(set_atomicity_lock); /* * Since we are disabling the cache don't allow any interrupts, @@ -590,7 +590,7 @@ static void prepare_set(void) __acquires(set_atomicity_lock) * changes to the way the kernel boots */ - spin_lock(&set_atomicity_lock); + raw_spin_lock(&set_atomicity_lock); /* Enter the no-fill (CD=1, NW=0) cache mode and flush caches. */ cr0 = read_cr0() | X86_CR0_CD; @@ -627,7 +627,7 @@ static void post_set(void) __releases(set_atomicity_lock) /* Restore value of CR4 */ if (cpu_has_pge) write_cr4(cr4); - spin_unlock(&set_atomicity_lock); + raw_spin_unlock(&set_atomicity_lock); } static void generic_set_all(void) @@ -752,7 +752,7 @@ int positive_have_wrcomb(void) /* * Generic structure... */ -struct mtrr_ops generic_mtrr_ops = { +const struct mtrr_ops generic_mtrr_ops = { .use_intel_if = 1, .set_all = generic_set_all, .get = generic_get_mtrr, diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c index f04e72527604..e006e56f699c 100644 --- a/arch/x86/kernel/cpu/mtrr/if.c +++ b/arch/x86/kernel/cpu/mtrr/if.c @@ -4,6 +4,7 @@ #include <linux/proc_fs.h> #include <linux/module.h> #include <linux/ctype.h> +#include <linux/string.h> #include <linux/init.h> #define LINE_SIZE 80 @@ -96,17 +97,24 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos) unsigned long long base, size; char *ptr; char line[LINE_SIZE]; + int length; size_t linelen; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (!len) - return -EINVAL; memset(line, 0, LINE_SIZE); - if (len > LINE_SIZE) - len = LINE_SIZE; - if (copy_from_user(line, buf, len - 1)) + + length = len; + length--; + + if (length > LINE_SIZE - 1) + length = LINE_SIZE - 1; + + if (length < 0) + return -EINVAL; + + if (copy_from_user(line, buf, length)) return -EFAULT; linelen = strlen(line); @@ -126,8 +134,7 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos) return -EINVAL; base = simple_strtoull(line + 5, &ptr, 0); - while (isspace(*ptr)) - ptr++; + ptr = skip_spaces(ptr); if (strncmp(ptr, "size=", 5)) return -EINVAL; @@ -135,14 +142,11 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos) size = simple_strtoull(ptr + 5, &ptr, 0); if ((base & 0xfff) || (size & 0xfff)) return -EINVAL; - while (isspace(*ptr)) - ptr++; + ptr = skip_spaces(ptr); if (strncmp(ptr, "type=", 5)) return -EINVAL; - ptr += 5; - while (isspace(*ptr)) - ptr++; + ptr = skip_spaces(ptr + 5); for (i = 0; i < MTRR_NUM_TYPES; ++i) { if (strcmp(ptr, mtrr_strings[i])) diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index 84e83de54575..79556bd9b602 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -60,14 +60,14 @@ static DEFINE_MUTEX(mtrr_mutex); u64 size_or_mask, size_and_mask; static bool mtrr_aps_delayed_init; -static struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM]; +static const struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM]; -struct mtrr_ops *mtrr_if; +const struct mtrr_ops *mtrr_if; static void set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type); -void set_mtrr_ops(struct mtrr_ops *ops) +void set_mtrr_ops(const struct mtrr_ops *ops) { if (ops->vendor && ops->vendor < X86_VENDOR_NUM) mtrr_ops[ops->vendor] = ops; @@ -145,6 +145,7 @@ struct set_mtrr_data { /** * ipi_handler - Synchronisation handler. Executed by "other" CPUs. + * @info: pointer to mtrr configuration data * * Returns nothing. */ diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h index a501dee9a87a..df5e41f31a27 100644 --- a/arch/x86/kernel/cpu/mtrr/mtrr.h +++ b/arch/x86/kernel/cpu/mtrr/mtrr.h @@ -32,7 +32,7 @@ extern int generic_get_free_region(unsigned long base, unsigned long size, extern int generic_validate_add_page(unsigned long base, unsigned long size, unsigned int type); -extern struct mtrr_ops generic_mtrr_ops; +extern const struct mtrr_ops generic_mtrr_ops; extern int positive_have_wrcomb(void); @@ -53,10 +53,10 @@ void fill_mtrr_var_range(unsigned int index, u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi); void get_mtrr_state(void); -extern void set_mtrr_ops(struct mtrr_ops *ops); +extern void set_mtrr_ops(const struct mtrr_ops *ops); extern u64 size_or_mask, size_and_mask; -extern struct mtrr_ops *mtrr_if; +extern const struct mtrr_ops *mtrr_if; #define is_cpu(vnd) (mtrr_if && mtrr_if->vendor == X86_VENDOR_##vnd) #define use_intel() (mtrr_if && mtrr_if->use_intel_if == 1) diff --git a/arch/x86/kernel/cpu/mtrr/state.c b/arch/x86/kernel/cpu/mtrr/state.c deleted file mode 100644 index dfc80b4e6b0d..000000000000 --- a/arch/x86/kernel/cpu/mtrr/state.c +++ /dev/null @@ -1,94 +0,0 @@ -#include <linux/init.h> -#include <linux/io.h> -#include <linux/mm.h> - -#include <asm/processor-cyrix.h> -#include <asm/processor-flags.h> -#include <asm/mtrr.h> -#include <asm/msr.h> - -#include "mtrr.h" - -/* Put the processor into a state where MTRRs can be safely set */ -void set_mtrr_prepare_save(struct set_mtrr_context *ctxt) -{ - unsigned int cr0; - - /* Disable interrupts locally */ - local_irq_save(ctxt->flags); - - if (use_intel() || is_cpu(CYRIX)) { - - /* Save value of CR4 and clear Page Global Enable (bit 7) */ - if (cpu_has_pge) { - ctxt->cr4val = read_cr4(); - write_cr4(ctxt->cr4val & ~X86_CR4_PGE); - } - - /* - * Disable and flush caches. Note that wbinvd flushes the TLBs - * as a side-effect - */ - cr0 = read_cr0() | X86_CR0_CD; - wbinvd(); - write_cr0(cr0); - wbinvd(); - - if (use_intel()) { - /* Save MTRR state */ - rdmsr(MSR_MTRRdefType, ctxt->deftype_lo, ctxt->deftype_hi); - } else { - /* - * Cyrix ARRs - - * everything else were excluded at the top - */ - ctxt->ccr3 = getCx86(CX86_CCR3); - } - } -} - -void set_mtrr_cache_disable(struct set_mtrr_context *ctxt) -{ - if (use_intel()) { - /* Disable MTRRs, and set the default type to uncached */ - mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo & 0xf300UL, - ctxt->deftype_hi); - } else { - if (is_cpu(CYRIX)) { - /* Cyrix ARRs - everything else were excluded at the top */ - setCx86(CX86_CCR3, (ctxt->ccr3 & 0x0f) | 0x10); - } - } -} - -/* Restore the processor after a set_mtrr_prepare */ -void set_mtrr_done(struct set_mtrr_context *ctxt) -{ - if (use_intel() || is_cpu(CYRIX)) { - - /* Flush caches and TLBs */ - wbinvd(); - - /* Restore MTRRdefType */ - if (use_intel()) { - /* Intel (P6) standard MTRRs */ - mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo, - ctxt->deftype_hi); - } else { - /* - * Cyrix ARRs - - * everything else was excluded at the top - */ - setCx86(CX86_CCR3, ctxt->ccr3); - } - - /* Enable caches */ - write_cr0(read_cr0() & 0xbfffffff); - - /* Restore value of CR4 */ - if (cpu_has_pge) - write_cr4(ctxt->cr4val); - } - /* Re-enable interrupts locally (if enabled previously) */ - local_irq_restore(ctxt->flags); -} diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index b5801c311846..60398a0d947c 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -7,6 +7,7 @@ * Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> * Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com> + * Copyright (C) 2009 Google, Inc., Stephane Eranian * * For licencing details see kernel-base/COPYING */ @@ -22,6 +23,7 @@ #include <linux/uaccess.h> #include <linux/highmem.h> #include <linux/cpu.h> +#include <linux/bitops.h> #include <asm/apic.h> #include <asm/stacktrace.h> @@ -68,15 +70,60 @@ struct debug_store { u64 pebs_event_reset[MAX_PEBS_EVENTS]; }; +struct event_constraint { + union { + unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + u64 idxmsk64; + }; + u64 code; + u64 cmask; + int weight; +}; + +struct amd_nb { + int nb_id; /* NorthBridge id */ + int refcnt; /* reference count */ + struct perf_event *owners[X86_PMC_IDX_MAX]; + struct event_constraint event_constraints[X86_PMC_IDX_MAX]; +}; + struct cpu_hw_events { - struct perf_event *events[X86_PMC_IDX_MAX]; - unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */ unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; unsigned long interrupts; int enabled; struct debug_store *ds; + + int n_events; + int n_added; + int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */ + u64 tags[X86_PMC_IDX_MAX]; + struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */ + struct amd_nb *amd_nb; }; +#define __EVENT_CONSTRAINT(c, n, m, w) {\ + { .idxmsk64 = (n) }, \ + .code = (c), \ + .cmask = (m), \ + .weight = (w), \ +} + +#define EVENT_CONSTRAINT(c, n, m) \ + __EVENT_CONSTRAINT(c, n, m, HWEIGHT(n)) + +#define INTEL_EVENT_CONSTRAINT(c, n) \ + EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVTSEL_MASK) + +#define FIXED_EVENT_CONSTRAINT(c, n) \ + EVENT_CONSTRAINT(c, (1ULL << (32+n)), INTEL_ARCH_FIXED_MASK) + +#define EVENT_CONSTRAINT_END \ + EVENT_CONSTRAINT(0, 0, 0) + +#define for_each_event_constraint(e, c) \ + for ((e) = (c); (e)->cmask; (e)++) + /* * struct x86_pmu - generic x86 pmu */ @@ -86,8 +133,8 @@ struct x86_pmu { int (*handle_irq)(struct pt_regs *); void (*disable_all)(void); void (*enable_all)(void); - void (*enable)(struct hw_perf_event *, int); - void (*disable)(struct hw_perf_event *, int); + void (*enable)(struct perf_event *); + void (*disable)(struct perf_event *); unsigned eventsel; unsigned perfctr; u64 (*event_map)(int); @@ -102,78 +149,28 @@ struct x86_pmu { u64 intel_ctrl; void (*enable_bts)(u64 config); void (*disable_bts)(void); -}; -static struct x86_pmu x86_pmu __read_mostly; + struct event_constraint * + (*get_event_constraints)(struct cpu_hw_events *cpuc, + struct perf_event *event); -static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { - .enabled = 1, -}; + void (*put_event_constraints)(struct cpu_hw_events *cpuc, + struct perf_event *event); + struct event_constraint *event_constraints; -/* - * Not sure about some of these - */ -static const u64 p6_perfmon_event_map[] = -{ - [PERF_COUNT_HW_CPU_CYCLES] = 0x0079, - [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, - [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0f2e, - [PERF_COUNT_HW_CACHE_MISSES] = 0x012e, - [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c4, - [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c5, - [PERF_COUNT_HW_BUS_CYCLES] = 0x0062, + void (*cpu_prepare)(int cpu); + void (*cpu_starting)(int cpu); + void (*cpu_dying)(int cpu); + void (*cpu_dead)(int cpu); }; -static u64 p6_pmu_event_map(int hw_event) -{ - return p6_perfmon_event_map[hw_event]; -} - -/* - * Event setting that is specified not to count anything. - * We use this to effectively disable a counter. - * - * L2_RQSTS with 0 MESI unit mask. - */ -#define P6_NOP_EVENT 0x0000002EULL - -static u64 p6_pmu_raw_event(u64 hw_event) -{ -#define P6_EVNTSEL_EVENT_MASK 0x000000FFULL -#define P6_EVNTSEL_UNIT_MASK 0x0000FF00ULL -#define P6_EVNTSEL_EDGE_MASK 0x00040000ULL -#define P6_EVNTSEL_INV_MASK 0x00800000ULL -#define P6_EVNTSEL_REG_MASK 0xFF000000ULL - -#define P6_EVNTSEL_MASK \ - (P6_EVNTSEL_EVENT_MASK | \ - P6_EVNTSEL_UNIT_MASK | \ - P6_EVNTSEL_EDGE_MASK | \ - P6_EVNTSEL_INV_MASK | \ - P6_EVNTSEL_REG_MASK) - - return hw_event & P6_EVNTSEL_MASK; -} - +static struct x86_pmu x86_pmu __read_mostly; -/* - * Intel PerfMon v3. Used on Core2 and later. - */ -static const u64 intel_perfmon_event_map[] = -{ - [PERF_COUNT_HW_CPU_CYCLES] = 0x003c, - [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, - [PERF_COUNT_HW_CACHE_REFERENCES] = 0x4f2e, - [PERF_COUNT_HW_CACHE_MISSES] = 0x412e, - [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c4, - [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c5, - [PERF_COUNT_HW_BUS_CYCLES] = 0x013c, +static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { + .enabled = 1, }; -static u64 intel_pmu_event_map(int hw_event) -{ - return intel_perfmon_event_map[hw_event]; -} +static int x86_perf_event_set_period(struct perf_event *event); /* * Generalized hw caching related hw_event table, filled @@ -190,435 +187,18 @@ static u64 __read_mostly hw_cache_event_ids [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX]; -static const u64 nehalem_hw_cache_event_ids - [PERF_COUNT_HW_CACHE_MAX] - [PERF_COUNT_HW_CACHE_OP_MAX] - [PERF_COUNT_HW_CACHE_RESULT_MAX] = -{ - [ C(L1D) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI */ - [ C(RESULT_MISS) ] = 0x0140, /* L1D_CACHE_LD.I_STATE */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI */ - [ C(RESULT_MISS) ] = 0x0141, /* L1D_CACHE_ST.I_STATE */ - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0x014e, /* L1D_PREFETCH.REQUESTS */ - [ C(RESULT_MISS) ] = 0x024e, /* L1D_PREFETCH.MISS */ - }, - }, - [ C(L1I ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS */ - [ C(RESULT_MISS) ] = 0x0280, /* L1I.MISSES */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0x0, - [ C(RESULT_MISS) ] = 0x0, - }, - }, - [ C(LL ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0324, /* L2_RQSTS.LOADS */ - [ C(RESULT_MISS) ] = 0x0224, /* L2_RQSTS.LD_MISS */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x0c24, /* L2_RQSTS.RFOS */ - [ C(RESULT_MISS) ] = 0x0824, /* L2_RQSTS.RFO_MISS */ - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0x4f2e, /* LLC Reference */ - [ C(RESULT_MISS) ] = 0x412e, /* LLC Misses */ - }, - }, - [ C(DTLB) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI (alias) */ - [ C(RESULT_MISS) ] = 0x0108, /* DTLB_LOAD_MISSES.ANY */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI (alias) */ - [ C(RESULT_MISS) ] = 0x010c, /* MEM_STORE_RETIRED.DTLB_MISS */ - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0x0, - [ C(RESULT_MISS) ] = 0x0, - }, - }, - [ C(ITLB) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x01c0, /* INST_RETIRED.ANY_P */ - [ C(RESULT_MISS) ] = 0x20c8, /* ITLB_MISS_RETIRED */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - }, - [ C(BPU ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ALL_BRANCHES */ - [ C(RESULT_MISS) ] = 0x03e8, /* BPU_CLEARS.ANY */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - }, -}; - -static const u64 core2_hw_cache_event_ids - [PERF_COUNT_HW_CACHE_MAX] - [PERF_COUNT_HW_CACHE_OP_MAX] - [PERF_COUNT_HW_CACHE_RESULT_MAX] = -{ - [ C(L1D) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI */ - [ C(RESULT_MISS) ] = 0x0140, /* L1D_CACHE_LD.I_STATE */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI */ - [ C(RESULT_MISS) ] = 0x0141, /* L1D_CACHE_ST.I_STATE */ - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0x104e, /* L1D_PREFETCH.REQUESTS */ - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(L1I ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0080, /* L1I.READS */ - [ C(RESULT_MISS) ] = 0x0081, /* L1I.MISSES */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(LL ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x4f29, /* L2_LD.MESI */ - [ C(RESULT_MISS) ] = 0x4129, /* L2_LD.ISTATE */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x4f2A, /* L2_ST.MESI */ - [ C(RESULT_MISS) ] = 0x412A, /* L2_ST.ISTATE */ - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(DTLB) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI (alias) */ - [ C(RESULT_MISS) ] = 0x0208, /* DTLB_MISSES.MISS_LD */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI (alias) */ - [ C(RESULT_MISS) ] = 0x0808, /* DTLB_MISSES.MISS_ST */ - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(ITLB) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x00c0, /* INST_RETIRED.ANY_P */ - [ C(RESULT_MISS) ] = 0x1282, /* ITLBMISSES */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - }, - [ C(BPU ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ANY */ - [ C(RESULT_MISS) ] = 0x00c5, /* BP_INST_RETIRED.MISPRED */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - }, -}; - -static const u64 atom_hw_cache_event_ids - [PERF_COUNT_HW_CACHE_MAX] - [PERF_COUNT_HW_CACHE_OP_MAX] - [PERF_COUNT_HW_CACHE_RESULT_MAX] = -{ - [ C(L1D) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x2140, /* L1D_CACHE.LD */ - [ C(RESULT_MISS) ] = 0, - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x2240, /* L1D_CACHE.ST */ - [ C(RESULT_MISS) ] = 0, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0x0, - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(L1I ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS */ - [ C(RESULT_MISS) ] = 0x0280, /* L1I.MISSES */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(LL ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x4f29, /* L2_LD.MESI */ - [ C(RESULT_MISS) ] = 0x4129, /* L2_LD.ISTATE */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x4f2A, /* L2_ST.MESI */ - [ C(RESULT_MISS) ] = 0x412A, /* L2_ST.ISTATE */ - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(DTLB) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x2140, /* L1D_CACHE_LD.MESI (alias) */ - [ C(RESULT_MISS) ] = 0x0508, /* DTLB_MISSES.MISS_LD */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x2240, /* L1D_CACHE_ST.MESI (alias) */ - [ C(RESULT_MISS) ] = 0x0608, /* DTLB_MISSES.MISS_ST */ - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(ITLB) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x00c0, /* INST_RETIRED.ANY_P */ - [ C(RESULT_MISS) ] = 0x0282, /* ITLB.MISSES */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - }, - [ C(BPU ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ANY */ - [ C(RESULT_MISS) ] = 0x00c5, /* BP_INST_RETIRED.MISPRED */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - }, -}; - -static u64 intel_pmu_raw_event(u64 hw_event) -{ -#define CORE_EVNTSEL_EVENT_MASK 0x000000FFULL -#define CORE_EVNTSEL_UNIT_MASK 0x0000FF00ULL -#define CORE_EVNTSEL_EDGE_MASK 0x00040000ULL -#define CORE_EVNTSEL_INV_MASK 0x00800000ULL -#define CORE_EVNTSEL_REG_MASK 0xFF000000ULL - -#define CORE_EVNTSEL_MASK \ - (CORE_EVNTSEL_EVENT_MASK | \ - CORE_EVNTSEL_UNIT_MASK | \ - CORE_EVNTSEL_EDGE_MASK | \ - CORE_EVNTSEL_INV_MASK | \ - CORE_EVNTSEL_REG_MASK) - - return hw_event & CORE_EVNTSEL_MASK; -} - -static const u64 amd_hw_cache_event_ids - [PERF_COUNT_HW_CACHE_MAX] - [PERF_COUNT_HW_CACHE_OP_MAX] - [PERF_COUNT_HW_CACHE_RESULT_MAX] = -{ - [ C(L1D) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ - [ C(RESULT_MISS) ] = 0x0041, /* Data Cache Misses */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x0142, /* Data Cache Refills :system */ - [ C(RESULT_MISS) ] = 0, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0x0267, /* Data Prefetcher :attempts */ - [ C(RESULT_MISS) ] = 0x0167, /* Data Prefetcher :cancelled */ - }, - }, - [ C(L1I ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0080, /* Instruction cache fetches */ - [ C(RESULT_MISS) ] = 0x0081, /* Instruction cache misses */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0x014B, /* Prefetch Instructions :Load */ - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(LL ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x037D, /* Requests to L2 Cache :IC+DC */ - [ C(RESULT_MISS) ] = 0x037E, /* L2 Cache Misses : IC+DC */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0x017F, /* L2 Fill/Writeback */ - [ C(RESULT_MISS) ] = 0, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(DTLB) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ - [ C(RESULT_MISS) ] = 0x0046, /* L1 DTLB and L2 DLTB Miss */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, - }, - }, - [ C(ITLB) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x0080, /* Instruction fecthes */ - [ C(RESULT_MISS) ] = 0x0085, /* Instr. fetch ITLB misses */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - }, - [ C(BPU ) ] = { - [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0x00c2, /* Retired Branch Instr. */ - [ C(RESULT_MISS) ] = 0x00c3, /* Retired Mispredicted BI */ - }, - [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = -1, - [ C(RESULT_MISS) ] = -1, - }, - }, -}; - -/* - * AMD Performance Monitor K7 and later. - */ -static const u64 amd_perfmon_event_map[] = -{ - [PERF_COUNT_HW_CPU_CYCLES] = 0x0076, - [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, - [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0080, - [PERF_COUNT_HW_CACHE_MISSES] = 0x0081, - [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c4, - [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c5, -}; - -static u64 amd_pmu_event_map(int hw_event) -{ - return amd_perfmon_event_map[hw_event]; -} - -static u64 amd_pmu_raw_event(u64 hw_event) -{ -#define K7_EVNTSEL_EVENT_MASK 0x7000000FFULL -#define K7_EVNTSEL_UNIT_MASK 0x00000FF00ULL -#define K7_EVNTSEL_EDGE_MASK 0x000040000ULL -#define K7_EVNTSEL_INV_MASK 0x000800000ULL -#define K7_EVNTSEL_REG_MASK 0x0FF000000ULL - -#define K7_EVNTSEL_MASK \ - (K7_EVNTSEL_EVENT_MASK | \ - K7_EVNTSEL_UNIT_MASK | \ - K7_EVNTSEL_EDGE_MASK | \ - K7_EVNTSEL_INV_MASK | \ - K7_EVNTSEL_REG_MASK) - - return hw_event & K7_EVNTSEL_MASK; -} - /* * Propagate event elapsed time into the generic event. * Can only be executed on the CPU where the event is active. * Returns the delta events processed. */ static u64 -x86_perf_event_update(struct perf_event *event, - struct hw_perf_event *hwc, int idx) +x86_perf_event_update(struct perf_event *event) { + struct hw_perf_event *hwc = &event->hw; int shift = 64 - x86_pmu.event_bits; u64 prev_raw_count, new_raw_count; + int idx = hwc->idx; s64 delta; if (idx == X86_PMC_IDX_FIXED_BTS) @@ -718,7 +298,7 @@ static inline bool bts_available(void) return x86_pmu.enable_bts != NULL; } -static inline void init_debug_store_on_cpu(int cpu) +static void init_debug_store_on_cpu(int cpu) { struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; @@ -730,7 +310,7 @@ static inline void init_debug_store_on_cpu(int cpu) (u32)((u64)(unsigned long)ds >> 32)); } -static inline void fini_debug_store_on_cpu(int cpu) +static void fini_debug_store_on_cpu(int cpu) { if (!per_cpu(cpu_hw_events, cpu).ds) return; @@ -859,42 +439,6 @@ set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event_attr *attr) return 0; } -static void intel_pmu_enable_bts(u64 config) -{ - unsigned long debugctlmsr; - - debugctlmsr = get_debugctlmsr(); - - debugctlmsr |= X86_DEBUGCTL_TR; - debugctlmsr |= X86_DEBUGCTL_BTS; - debugctlmsr |= X86_DEBUGCTL_BTINT; - - if (!(config & ARCH_PERFMON_EVENTSEL_OS)) - debugctlmsr |= X86_DEBUGCTL_BTS_OFF_OS; - - if (!(config & ARCH_PERFMON_EVENTSEL_USR)) - debugctlmsr |= X86_DEBUGCTL_BTS_OFF_USR; - - update_debugctlmsr(debugctlmsr); -} - -static void intel_pmu_disable_bts(void) -{ - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - unsigned long debugctlmsr; - - if (!cpuc->ds) - return; - - debugctlmsr = get_debugctlmsr(); - - debugctlmsr &= - ~(X86_DEBUGCTL_TR | X86_DEBUGCTL_BTS | X86_DEBUGCTL_BTINT | - X86_DEBUGCTL_BTS_OFF_OS | X86_DEBUGCTL_BTS_OFF_USR); - - update_debugctlmsr(debugctlmsr); -} - /* * Setup the hardware configuration for a given attr_type */ @@ -932,6 +476,10 @@ static int __hw_perf_event_init(struct perf_event *event) */ hwc->config = ARCH_PERFMON_EVENTSEL_INT; + hwc->idx = -1; + hwc->last_cpu = -1; + hwc->last_tag = ~0ULL; + /* * Count user and OS events unless requested not to. */ @@ -960,6 +508,9 @@ static int __hw_perf_event_init(struct perf_event *event) */ if (attr->type == PERF_TYPE_RAW) { hwc->config |= x86_pmu.raw_event(attr->config); + if ((hwc->config & ARCH_PERFMON_EVENTSEL_ANY) && + perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) + return -EACCES; return 0; } @@ -999,216 +550,314 @@ static int __hw_perf_event_init(struct perf_event *event) return 0; } -static void p6_pmu_disable_all(void) +static void x86_pmu_disable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - u64 val; - - if (!cpuc->enabled) - return; + int idx; - cpuc->enabled = 0; - barrier(); + for (idx = 0; idx < x86_pmu.num_events; idx++) { + u64 val; - /* p6 only has one enable register */ - rdmsrl(MSR_P6_EVNTSEL0, val); - val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE; - wrmsrl(MSR_P6_EVNTSEL0, val); + if (!test_bit(idx, cpuc->active_mask)) + continue; + rdmsrl(x86_pmu.eventsel + idx, val); + if (!(val & ARCH_PERFMON_EVENTSEL_ENABLE)) + continue; + val &= ~ARCH_PERFMON_EVENTSEL_ENABLE; + wrmsrl(x86_pmu.eventsel + idx, val); + } } -static void intel_pmu_disable_all(void) +void hw_perf_disable(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + if (!x86_pmu_initialized()) + return; + if (!cpuc->enabled) return; + cpuc->n_added = 0; cpuc->enabled = 0; barrier(); - wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0); - - if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) - intel_pmu_disable_bts(); + x86_pmu.disable_all(); } -static void amd_pmu_disable_all(void) +static void x86_pmu_enable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int idx; - if (!cpuc->enabled) - return; - - cpuc->enabled = 0; - /* - * ensure we write the disable before we start disabling the - * events proper, so that amd_pmu_enable_event() does the - * right thing. - */ - barrier(); - for (idx = 0; idx < x86_pmu.num_events; idx++) { + struct perf_event *event = cpuc->events[idx]; u64 val; if (!test_bit(idx, cpuc->active_mask)) continue; - rdmsrl(MSR_K7_EVNTSEL0 + idx, val); - if (!(val & ARCH_PERFMON_EVENTSEL0_ENABLE)) - continue; - val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE; - wrmsrl(MSR_K7_EVNTSEL0 + idx, val); + + val = event->hw.config; + val |= ARCH_PERFMON_EVENTSEL_ENABLE; + wrmsrl(x86_pmu.eventsel + idx, val); } } -void hw_perf_disable(void) +static const struct pmu pmu; + +static inline int is_x86_event(struct perf_event *event) { - if (!x86_pmu_initialized()) - return; - return x86_pmu.disable_all(); + return event->pmu == &pmu; } -static void p6_pmu_enable_all(void) +static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) { - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - unsigned long val; + struct event_constraint *c, *constraints[X86_PMC_IDX_MAX]; + unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + int i, j, w, wmax, num = 0; + struct hw_perf_event *hwc; - if (cpuc->enabled) - return; + bitmap_zero(used_mask, X86_PMC_IDX_MAX); - cpuc->enabled = 1; - barrier(); + for (i = 0; i < n; i++) { + c = x86_pmu.get_event_constraints(cpuc, cpuc->event_list[i]); + constraints[i] = c; + } - /* p6 only has one enable register */ - rdmsrl(MSR_P6_EVNTSEL0, val); - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; - wrmsrl(MSR_P6_EVNTSEL0, val); -} + /* + * fastpath, try to reuse previous register + */ + for (i = 0; i < n; i++) { + hwc = &cpuc->event_list[i]->hw; + c = constraints[i]; -static void intel_pmu_enable_all(void) -{ - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + /* never assigned */ + if (hwc->idx == -1) + break; - if (cpuc->enabled) - return; + /* constraint still honored */ + if (!test_bit(hwc->idx, c->idxmsk)) + break; - cpuc->enabled = 1; - barrier(); + /* not already used */ + if (test_bit(hwc->idx, used_mask)) + break; - wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl); + __set_bit(hwc->idx, used_mask); + if (assign) + assign[i] = hwc->idx; + } + if (i == n) + goto done; - if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) { - struct perf_event *event = - cpuc->events[X86_PMC_IDX_FIXED_BTS]; + /* + * begin slow path + */ - if (WARN_ON_ONCE(!event)) - return; + bitmap_zero(used_mask, X86_PMC_IDX_MAX); - intel_pmu_enable_bts(event->hw.config); - } -} + /* + * weight = number of possible counters + * + * 1 = most constrained, only works on one counter + * wmax = least constrained, works on any counter + * + * assign events to counters starting with most + * constrained events. + */ + wmax = x86_pmu.num_events; -static void amd_pmu_enable_all(void) -{ - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - int idx; + /* + * when fixed event counters are present, + * wmax is incremented by 1 to account + * for one more choice + */ + if (x86_pmu.num_events_fixed) + wmax++; - if (cpuc->enabled) - return; + for (w = 1, num = n; num && w <= wmax; w++) { + /* for each event */ + for (i = 0; num && i < n; i++) { + c = constraints[i]; + hwc = &cpuc->event_list[i]->hw; - cpuc->enabled = 1; - barrier(); + if (c->weight != w) + continue; - for (idx = 0; idx < x86_pmu.num_events; idx++) { - struct perf_event *event = cpuc->events[idx]; - u64 val; + for_each_set_bit(j, c->idxmsk, X86_PMC_IDX_MAX) { + if (!test_bit(j, used_mask)) + break; + } - if (!test_bit(idx, cpuc->active_mask)) - continue; + if (j == X86_PMC_IDX_MAX) + break; - val = event->hw.config; - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; - wrmsrl(MSR_K7_EVNTSEL0 + idx, val); + __set_bit(j, used_mask); + + if (assign) + assign[i] = j; + num--; + } } +done: + /* + * scheduling failed or is just a simulation, + * free resources if necessary + */ + if (!assign || num) { + for (i = 0; i < n; i++) { + if (x86_pmu.put_event_constraints) + x86_pmu.put_event_constraints(cpuc, cpuc->event_list[i]); + } + } + return num ? -ENOSPC : 0; } -void hw_perf_enable(void) +/* + * dogrp: true if must collect siblings events (group) + * returns total number of events and error code + */ +static int collect_events(struct cpu_hw_events *cpuc, struct perf_event *leader, bool dogrp) { - if (!x86_pmu_initialized()) - return; - x86_pmu.enable_all(); -} + struct perf_event *event; + int n, max_count; -static inline u64 intel_pmu_get_status(void) -{ - u64 status; + max_count = x86_pmu.num_events + x86_pmu.num_events_fixed; - rdmsrl(MSR_CORE_PERF_GLOBAL_STATUS, status); + /* current number of events already accepted */ + n = cpuc->n_events; - return status; -} + if (is_x86_event(leader)) { + if (n >= max_count) + return -ENOSPC; + cpuc->event_list[n] = leader; + n++; + } + if (!dogrp) + return n; -static inline void intel_pmu_ack_status(u64 ack) -{ - wrmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, ack); -} + list_for_each_entry(event, &leader->sibling_list, group_entry) { + if (!is_x86_event(event) || + event->state <= PERF_EVENT_STATE_OFF) + continue; -static inline void x86_pmu_enable_event(struct hw_perf_event *hwc, int idx) -{ - (void)checking_wrmsrl(hwc->config_base + idx, - hwc->config | ARCH_PERFMON_EVENTSEL0_ENABLE); -} + if (n >= max_count) + return -ENOSPC; -static inline void x86_pmu_disable_event(struct hw_perf_event *hwc, int idx) -{ - (void)checking_wrmsrl(hwc->config_base + idx, hwc->config); + cpuc->event_list[n] = event; + n++; + } + return n; } -static inline void -intel_pmu_disable_fixed(struct hw_perf_event *hwc, int __idx) +static inline void x86_assign_hw_event(struct perf_event *event, + struct cpu_hw_events *cpuc, int i) { - int idx = __idx - X86_PMC_IDX_FIXED; - u64 ctrl_val, mask; + struct hw_perf_event *hwc = &event->hw; - mask = 0xfULL << (idx * 4); + hwc->idx = cpuc->assign[i]; + hwc->last_cpu = smp_processor_id(); + hwc->last_tag = ++cpuc->tags[i]; - rdmsrl(hwc->config_base, ctrl_val); - ctrl_val &= ~mask; - (void)checking_wrmsrl(hwc->config_base, ctrl_val); + if (hwc->idx == X86_PMC_IDX_FIXED_BTS) { + hwc->config_base = 0; + hwc->event_base = 0; + } else if (hwc->idx >= X86_PMC_IDX_FIXED) { + hwc->config_base = MSR_ARCH_PERFMON_FIXED_CTR_CTRL; + /* + * We set it so that event_base + idx in wrmsr/rdmsr maps to + * MSR_ARCH_PERFMON_FIXED_CTR0 ... CTR2: + */ + hwc->event_base = + MSR_ARCH_PERFMON_FIXED_CTR0 - X86_PMC_IDX_FIXED; + } else { + hwc->config_base = x86_pmu.eventsel; + hwc->event_base = x86_pmu.perfctr; + } } -static inline void -p6_pmu_disable_event(struct hw_perf_event *hwc, int idx) +static inline int match_prev_assignment(struct hw_perf_event *hwc, + struct cpu_hw_events *cpuc, + int i) { - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - u64 val = P6_NOP_EVENT; - - if (cpuc->enabled) - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; - - (void)checking_wrmsrl(hwc->config_base + idx, val); + return hwc->idx == cpuc->assign[i] && + hwc->last_cpu == smp_processor_id() && + hwc->last_tag == cpuc->tags[i]; } -static inline void -intel_pmu_disable_event(struct hw_perf_event *hwc, int idx) +static int x86_pmu_start(struct perf_event *event); +static void x86_pmu_stop(struct perf_event *event); + +void hw_perf_enable(void) { - if (unlikely(idx == X86_PMC_IDX_FIXED_BTS)) { - intel_pmu_disable_bts(); + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct perf_event *event; + struct hw_perf_event *hwc; + int i; + + if (!x86_pmu_initialized()) return; - } - if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { - intel_pmu_disable_fixed(hwc, idx); + if (cpuc->enabled) return; + + if (cpuc->n_added) { + int n_running = cpuc->n_events - cpuc->n_added; + /* + * apply assignment obtained either from + * hw_perf_group_sched_in() or x86_pmu_enable() + * + * step1: save events moving to new counters + * step2: reprogram moved events into new counters + */ + for (i = 0; i < n_running; i++) { + event = cpuc->event_list[i]; + hwc = &event->hw; + + /* + * we can avoid reprogramming counter if: + * - assigned same counter as last time + * - running on same CPU as last time + * - no other event has used the counter since + */ + if (hwc->idx == -1 || + match_prev_assignment(hwc, cpuc, i)) + continue; + + x86_pmu_stop(event); + } + + for (i = 0; i < cpuc->n_events; i++) { + event = cpuc->event_list[i]; + hwc = &event->hw; + + if (!match_prev_assignment(hwc, cpuc, i)) + x86_assign_hw_event(event, cpuc, i); + else if (i < n_running) + continue; + + x86_pmu_start(event); + } + cpuc->n_added = 0; + perf_events_lapic_init(); } - x86_pmu_disable_event(hwc, idx); + cpuc->enabled = 1; + barrier(); + + x86_pmu.enable_all(); +} + +static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc) +{ + (void)checking_wrmsrl(hwc->config_base + hwc->idx, + hwc->config | ARCH_PERFMON_EVENTSEL_ENABLE); } -static inline void -amd_pmu_disable_event(struct hw_perf_event *hwc, int idx) +static inline void x86_pmu_disable_event(struct perf_event *event) { - x86_pmu_disable_event(hwc, idx); + struct hw_perf_event *hwc = &event->hw; + (void)checking_wrmsrl(hwc->config_base + hwc->idx, hwc->config); } static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left); @@ -1218,18 +867,18 @@ static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left); * To be called with the event disabled in hw: */ static int -x86_perf_event_set_period(struct perf_event *event, - struct hw_perf_event *hwc, int idx) +x86_perf_event_set_period(struct perf_event *event) { + struct hw_perf_event *hwc = &event->hw; s64 left = atomic64_read(&hwc->period_left); s64 period = hwc->sample_period; - int err, ret = 0; + int err, ret = 0, idx = hwc->idx; if (idx == X86_PMC_IDX_FIXED_BTS) return 0; /* - * If we are way outside a reasoable range then just skip forward: + * If we are way outside a reasonable range then just skip forward: */ if (unlikely(left <= -period)) { left = period; @@ -1269,157 +918,63 @@ x86_perf_event_set_period(struct perf_event *event, return ret; } -static inline void -intel_pmu_enable_fixed(struct hw_perf_event *hwc, int __idx) -{ - int idx = __idx - X86_PMC_IDX_FIXED; - u64 ctrl_val, bits, mask; - int err; - - /* - * Enable IRQ generation (0x8), - * and enable ring-3 counting (0x2) and ring-0 counting (0x1) - * if requested: - */ - bits = 0x8ULL; - if (hwc->config & ARCH_PERFMON_EVENTSEL_USR) - bits |= 0x2; - if (hwc->config & ARCH_PERFMON_EVENTSEL_OS) - bits |= 0x1; - bits <<= (idx * 4); - mask = 0xfULL << (idx * 4); - - rdmsrl(hwc->config_base, ctrl_val); - ctrl_val &= ~mask; - ctrl_val |= bits; - err = checking_wrmsrl(hwc->config_base, ctrl_val); -} - -static void p6_pmu_enable_event(struct hw_perf_event *hwc, int idx) +static void x86_pmu_enable_event(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - u64 val; - - val = hwc->config; if (cpuc->enabled) - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; - - (void)checking_wrmsrl(hwc->config_base + idx, val); + __x86_pmu_enable_event(&event->hw); } - -static void intel_pmu_enable_event(struct hw_perf_event *hwc, int idx) -{ - if (unlikely(idx == X86_PMC_IDX_FIXED_BTS)) { - if (!__get_cpu_var(cpu_hw_events).enabled) - return; - - intel_pmu_enable_bts(hwc->config); - return; - } - - if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { - intel_pmu_enable_fixed(hwc, idx); - return; - } - - x86_pmu_enable_event(hwc, idx); -} - -static void amd_pmu_enable_event(struct hw_perf_event *hwc, int idx) +/* + * activate a single event + * + * 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) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct hw_perf_event *hwc; + int assign[X86_PMC_IDX_MAX]; + int n, n0, ret; - if (cpuc->enabled) - x86_pmu_enable_event(hwc, idx); -} - -static int -fixed_mode_idx(struct perf_event *event, struct hw_perf_event *hwc) -{ - unsigned int hw_event; - - hw_event = hwc->config & ARCH_PERFMON_EVENT_MASK; + hwc = &event->hw; - if (unlikely((hw_event == - x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS)) && - (hwc->sample_period == 1))) - return X86_PMC_IDX_FIXED_BTS; + n0 = cpuc->n_events; + n = collect_events(cpuc, event, false); + if (n < 0) + return n; - if (!x86_pmu.num_events_fixed) - return -1; + ret = x86_schedule_events(cpuc, n, assign); + if (ret) + return ret; + /* + * copy new assignment, now we know it is possible + * will be used by hw_perf_enable() + */ + memcpy(cpuc->assign, assign, n*sizeof(int)); - if (unlikely(hw_event == x86_pmu.event_map(PERF_COUNT_HW_INSTRUCTIONS))) - return X86_PMC_IDX_FIXED_INSTRUCTIONS; - if (unlikely(hw_event == x86_pmu.event_map(PERF_COUNT_HW_CPU_CYCLES))) - return X86_PMC_IDX_FIXED_CPU_CYCLES; - if (unlikely(hw_event == x86_pmu.event_map(PERF_COUNT_HW_BUS_CYCLES))) - return X86_PMC_IDX_FIXED_BUS_CYCLES; + cpuc->n_events = n; + cpuc->n_added += n - n0; - return -1; + return 0; } -/* - * Find a PMC slot for the freshly enabled / scheduled in event: - */ -static int x86_pmu_enable(struct perf_event *event) +static int x86_pmu_start(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - struct hw_perf_event *hwc = &event->hw; - int idx; - - idx = fixed_mode_idx(event, hwc); - if (idx == X86_PMC_IDX_FIXED_BTS) { - /* BTS is already occupied. */ - if (test_and_set_bit(idx, cpuc->used_mask)) - return -EAGAIN; - - hwc->config_base = 0; - hwc->event_base = 0; - hwc->idx = idx; - } else if (idx >= 0) { - /* - * Try to get the fixed event, if that is already taken - * then try to get a generic event: - */ - if (test_and_set_bit(idx, cpuc->used_mask)) - goto try_generic; - - hwc->config_base = MSR_ARCH_PERFMON_FIXED_CTR_CTRL; - /* - * We set it so that event_base + idx in wrmsr/rdmsr maps to - * MSR_ARCH_PERFMON_FIXED_CTR0 ... CTR2: - */ - hwc->event_base = - MSR_ARCH_PERFMON_FIXED_CTR0 - X86_PMC_IDX_FIXED; - hwc->idx = idx; - } else { - idx = hwc->idx; - /* Try to get the previous generic event again */ - if (test_and_set_bit(idx, cpuc->used_mask)) { -try_generic: - idx = find_first_zero_bit(cpuc->used_mask, - x86_pmu.num_events); - if (idx == x86_pmu.num_events) - return -EAGAIN; - - set_bit(idx, cpuc->used_mask); - hwc->idx = idx; - } - hwc->config_base = x86_pmu.eventsel; - hwc->event_base = x86_pmu.perfctr; - } + int idx = event->hw.idx; - perf_events_lapic_init(); - - x86_pmu.disable(hwc, idx); + if (idx == -1) + return -EAGAIN; + x86_perf_event_set_period(event); cpuc->events[idx] = event; - set_bit(idx, cpuc->active_mask); - - x86_perf_event_set_period(event, hwc, idx); - x86_pmu.enable(hwc, idx); - + __set_bit(idx, cpuc->active_mask); + x86_pmu.enable(event); perf_event_update_userpage(event); return 0; @@ -1427,14 +982,8 @@ try_generic: static void x86_pmu_unthrottle(struct perf_event *event) { - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - struct hw_perf_event *hwc = &event->hw; - - if (WARN_ON_ONCE(hwc->idx >= X86_PMC_IDX_MAX || - cpuc->events[hwc->idx] != event)) - return; - - x86_pmu.enable(hwc, hwc->idx); + int ret = x86_pmu_start(event); + WARN_ON_ONCE(ret); } void perf_event_print_debug(void) @@ -1464,7 +1013,7 @@ void perf_event_print_debug(void) pr_info("CPU#%d: overflow: %016llx\n", cpu, overflow); pr_info("CPU#%d: fixed: %016llx\n", cpu, fixed); } - pr_info("CPU#%d: used: %016llx\n", cpu, *(u64 *)cpuc->used_mask); + pr_info("CPU#%d: active: %016llx\n", cpu, *(u64 *)cpuc->active_mask); for (idx = 0; idx < x86_pmu.num_events; idx++) { rdmsrl(x86_pmu.eventsel + idx, pmc_ctrl); @@ -1488,254 +1037,50 @@ void perf_event_print_debug(void) local_irq_restore(flags); } -static void intel_pmu_drain_bts_buffer(struct cpu_hw_events *cpuc) -{ - struct debug_store *ds = cpuc->ds; - struct bts_record { - u64 from; - u64 to; - u64 flags; - }; - struct perf_event *event = cpuc->events[X86_PMC_IDX_FIXED_BTS]; - struct bts_record *at, *top; - struct perf_output_handle handle; - struct perf_event_header header; - struct perf_sample_data data; - struct pt_regs regs; - - if (!event) - return; - - if (!ds) - return; - - at = (struct bts_record *)(unsigned long)ds->bts_buffer_base; - top = (struct bts_record *)(unsigned long)ds->bts_index; - - if (top <= at) - return; - - ds->bts_index = ds->bts_buffer_base; - - - data.period = event->hw.last_period; - data.addr = 0; - regs.ip = 0; - - /* - * Prepare a generic sample, i.e. fill in the invariant fields. - * We will overwrite the from and to address before we output - * the sample. - */ - perf_prepare_sample(&header, &data, event, ®s); - - if (perf_output_begin(&handle, event, - header.size * (top - at), 1, 1)) - return; - - for (; at < top; at++) { - data.ip = at->from; - data.addr = at->to; - - perf_output_sample(&handle, &header, &data, event); - } - - perf_output_end(&handle); - - /* There's new data available. */ - event->hw.interrupts++; - event->pending_kill = POLL_IN; -} - -static void x86_pmu_disable(struct perf_event *event) +static void x86_pmu_stop(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; int idx = hwc->idx; - /* - * Must be done before we disable, otherwise the nmi handler - * could reenable again: - */ - clear_bit(idx, cpuc->active_mask); - x86_pmu.disable(hwc, idx); + if (!__test_and_clear_bit(idx, cpuc->active_mask)) + return; - /* - * Make sure the cleared pointer becomes visible before we - * (potentially) free the event: - */ - barrier(); + x86_pmu.disable(event); /* * Drain the remaining delta count out of a event * that we are disabling: */ - x86_perf_event_update(event, hwc, idx); - - /* Drain the remaining BTS records. */ - if (unlikely(idx == X86_PMC_IDX_FIXED_BTS)) - intel_pmu_drain_bts_buffer(cpuc); + x86_perf_event_update(event); cpuc->events[idx] = NULL; - clear_bit(idx, cpuc->used_mask); - - perf_event_update_userpage(event); } -/* - * Save and restart an expired event. Called by NMI contexts, - * so it has to be careful about preempting normal event ops: - */ -static int intel_pmu_save_and_restart(struct perf_event *event) -{ - struct hw_perf_event *hwc = &event->hw; - int idx = hwc->idx; - int ret; - - x86_perf_event_update(event, hwc, idx); - ret = x86_perf_event_set_period(event, hwc, idx); - - if (event->state == PERF_EVENT_STATE_ACTIVE) - intel_pmu_enable_event(hwc, idx); - - return ret; -} - -static void intel_pmu_reset(void) -{ - struct debug_store *ds = __get_cpu_var(cpu_hw_events).ds; - unsigned long flags; - int idx; - - if (!x86_pmu.num_events) - return; - - local_irq_save(flags); - - printk("clearing PMU state on CPU#%d\n", smp_processor_id()); - - for (idx = 0; idx < x86_pmu.num_events; idx++) { - checking_wrmsrl(x86_pmu.eventsel + idx, 0ull); - checking_wrmsrl(x86_pmu.perfctr + idx, 0ull); - } - for (idx = 0; idx < x86_pmu.num_events_fixed; idx++) { - checking_wrmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, 0ull); - } - if (ds) - ds->bts_index = ds->bts_buffer_base; - - local_irq_restore(flags); -} - -static int p6_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; - - data.addr = 0; - - cpuc = &__get_cpu_var(cpu_hw_events); - - for (idx = 0; idx < x86_pmu.num_events; idx++) { - if (!test_bit(idx, cpuc->active_mask)) - continue; - - event = cpuc->events[idx]; - hwc = &event->hw; - - val = x86_perf_event_update(event, hwc, idx); - if (val & (1ULL << (x86_pmu.event_bits - 1))) - continue; - - /* - * event overflow - */ - handled = 1; - data.period = event->hw.last_period; - - if (!x86_perf_event_set_period(event, hwc, idx)) - continue; - - if (perf_event_overflow(event, 1, &data, regs)) - p6_pmu_disable_event(hwc, idx); - } - - if (handled) - inc_irq_stat(apic_perf_irqs); - - return handled; -} - -/* - * This handler is triggered by the local APIC, so the APIC IRQ handling - * rules apply: - */ -static int intel_pmu_handle_irq(struct pt_regs *regs) +static void x86_pmu_disable(struct perf_event *event) { - struct perf_sample_data data; - struct cpu_hw_events *cpuc; - int bit, loops; - u64 ack, status; - - data.addr = 0; - - cpuc = &__get_cpu_var(cpu_hw_events); - - perf_disable(); - intel_pmu_drain_bts_buffer(cpuc); - status = intel_pmu_get_status(); - if (!status) { - perf_enable(); - return 0; - } + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int i; - loops = 0; -again: - if (++loops > 100) { - WARN_ONCE(1, "perfevents: irq loop stuck!\n"); - perf_event_print_debug(); - intel_pmu_reset(); - perf_enable(); - return 1; - } + x86_pmu_stop(event); - inc_irq_stat(apic_perf_irqs); - ack = status; - for_each_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) { - struct perf_event *event = cpuc->events[bit]; + for (i = 0; i < cpuc->n_events; i++) { + if (event == cpuc->event_list[i]) { - clear_bit(bit, (unsigned long *) &status); - if (!test_bit(bit, cpuc->active_mask)) - continue; + if (x86_pmu.put_event_constraints) + x86_pmu.put_event_constraints(cpuc, event); - if (!intel_pmu_save_and_restart(event)) - continue; - - data.period = event->hw.last_period; + while (++i < cpuc->n_events) + cpuc->event_list[i-1] = cpuc->event_list[i]; - if (perf_event_overflow(event, 1, &data, regs)) - intel_pmu_disable_event(&event->hw, bit); + --cpuc->n_events; + break; + } } - - intel_pmu_ack_status(ack); - - /* - * Repeat if there is more work to be done: - */ - status = intel_pmu_get_status(); - if (status) - goto again; - - perf_enable(); - - return 1; + perf_event_update_userpage(event); } -static int amd_pmu_handle_irq(struct pt_regs *regs) +static int x86_pmu_handle_irq(struct pt_regs *regs) { struct perf_sample_data data; struct cpu_hw_events *cpuc; @@ -1744,7 +1089,7 @@ static int amd_pmu_handle_irq(struct pt_regs *regs) int idx, handled = 0; u64 val; - data.addr = 0; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); @@ -1755,7 +1100,7 @@ static int amd_pmu_handle_irq(struct pt_regs *regs) event = cpuc->events[idx]; hwc = &event->hw; - val = x86_perf_event_update(event, hwc, idx); + val = x86_perf_event_update(event); if (val & (1ULL << (x86_pmu.event_bits - 1))) continue; @@ -1765,11 +1110,11 @@ static int amd_pmu_handle_irq(struct pt_regs *regs) handled = 1; data.period = event->hw.last_period; - if (!x86_perf_event_set_period(event, hwc, idx)) + if (!x86_perf_event_set_period(event)) continue; if (perf_event_overflow(event, 1, &data, regs)) - amd_pmu_disable_event(hwc, idx); + x86_pmu_stop(event); } if (handled) @@ -1852,196 +1197,184 @@ static __read_mostly struct notifier_block perf_event_nmi_notifier = { .priority = 1 }; -static struct x86_pmu p6_pmu = { - .name = "p6", - .handle_irq = p6_pmu_handle_irq, - .disable_all = p6_pmu_disable_all, - .enable_all = p6_pmu_enable_all, - .enable = p6_pmu_enable_event, - .disable = p6_pmu_disable_event, - .eventsel = MSR_P6_EVNTSEL0, - .perfctr = MSR_P6_PERFCTR0, - .event_map = p6_pmu_event_map, - .raw_event = p6_pmu_raw_event, - .max_events = ARRAY_SIZE(p6_perfmon_event_map), - .apic = 1, - .max_period = (1ULL << 31) - 1, - .version = 0, - .num_events = 2, - /* - * Events have 40 bits implemented. However they are designed such - * that bits [32-39] are sign extensions of bit 31. As such the - * effective width of a event for P6-like PMU is 32 bits only. - * - * See IA-32 Intel Architecture Software developer manual Vol 3B - */ - .event_bits = 32, - .event_mask = (1ULL << 32) - 1, -}; +static struct event_constraint unconstrained; +static struct event_constraint emptyconstraint; -static struct x86_pmu intel_pmu = { - .name = "Intel", - .handle_irq = intel_pmu_handle_irq, - .disable_all = intel_pmu_disable_all, - .enable_all = intel_pmu_enable_all, - .enable = intel_pmu_enable_event, - .disable = intel_pmu_disable_event, - .eventsel = MSR_ARCH_PERFMON_EVENTSEL0, - .perfctr = MSR_ARCH_PERFMON_PERFCTR0, - .event_map = intel_pmu_event_map, - .raw_event = intel_pmu_raw_event, - .max_events = ARRAY_SIZE(intel_perfmon_event_map), - .apic = 1, - /* - * Intel PMCs cannot be accessed sanely above 32 bit width, - * so we install an artificial 1<<31 period regardless of - * the generic event period: - */ - .max_period = (1ULL << 31) - 1, - .enable_bts = intel_pmu_enable_bts, - .disable_bts = intel_pmu_disable_bts, -}; +static struct event_constraint * +x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) +{ + struct event_constraint *c; -static struct x86_pmu amd_pmu = { - .name = "AMD", - .handle_irq = amd_pmu_handle_irq, - .disable_all = amd_pmu_disable_all, - .enable_all = amd_pmu_enable_all, - .enable = amd_pmu_enable_event, - .disable = amd_pmu_disable_event, - .eventsel = MSR_K7_EVNTSEL0, - .perfctr = MSR_K7_PERFCTR0, - .event_map = amd_pmu_event_map, - .raw_event = amd_pmu_raw_event, - .max_events = ARRAY_SIZE(amd_perfmon_event_map), - .num_events = 4, - .event_bits = 48, - .event_mask = (1ULL << 48) - 1, - .apic = 1, - /* use highest bit to detect overflow */ - .max_period = (1ULL << 47) - 1, -}; + if (x86_pmu.event_constraints) { + for_each_event_constraint(c, x86_pmu.event_constraints) { + if ((event->hw.config & c->cmask) == c->code) + return c; + } + } -static int p6_pmu_init(void) + return &unconstrained; +} + +static int x86_event_sched_in(struct perf_event *event, + struct perf_cpu_context *cpuctx) { - switch (boot_cpu_data.x86_model) { - case 1: - case 3: /* Pentium Pro */ - case 5: - case 6: /* Pentium II */ - case 7: - case 8: - case 11: /* Pentium III */ - break; - case 9: - case 13: - /* Pentium M */ - break; - default: - pr_cont("unsupported p6 CPU model %d ", - boot_cpu_data.x86_model); - return -ENODEV; - } + int ret = 0; - x86_pmu = p6_pmu; + event->state = PERF_EVENT_STATE_ACTIVE; + event->oncpu = smp_processor_id(); + event->tstamp_running += event->ctx->time - event->tstamp_stopped; - if (!cpu_has_apic) { - pr_info("no APIC, boot with the \"lapic\" boot parameter to force-enable it.\n"); - pr_info("no hardware sampling interrupt available.\n"); - x86_pmu.apic = 0; - } + if (!is_x86_event(event)) + ret = event->pmu->enable(event); - return 0; + if (!ret && !is_software_event(event)) + cpuctx->active_oncpu++; + + if (!ret && event->attr.exclusive) + cpuctx->exclusive = 1; + + return ret; } -static int intel_pmu_init(void) +static void x86_event_sched_out(struct perf_event *event, + struct perf_cpu_context *cpuctx) { - union cpuid10_edx edx; - union cpuid10_eax eax; - unsigned int unused; - unsigned int ebx; - int version; - - if (!cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) { - /* check for P6 processor family */ - if (boot_cpu_data.x86 == 6) { - return p6_pmu_init(); - } else { - return -ENODEV; - } - } + event->state = PERF_EVENT_STATE_INACTIVE; + event->oncpu = -1; - /* - * Check whether the Architectural PerfMon supports - * Branch Misses Retired hw_event or not. - */ - cpuid(10, &eax.full, &ebx, &unused, &edx.full); - if (eax.split.mask_length <= ARCH_PERFMON_BRANCH_MISSES_RETIRED) - return -ENODEV; + if (!is_x86_event(event)) + event->pmu->disable(event); - version = eax.split.version_id; - if (version < 2) - return -ENODEV; + event->tstamp_running -= event->ctx->time - event->tstamp_stopped; - x86_pmu = intel_pmu; - x86_pmu.version = version; - x86_pmu.num_events = eax.split.num_events; - x86_pmu.event_bits = eax.split.bit_width; - x86_pmu.event_mask = (1ULL << eax.split.bit_width) - 1; + if (!is_software_event(event)) + cpuctx->active_oncpu--; + + if (event->attr.exclusive || !cpuctx->active_oncpu) + cpuctx->exclusive = 0; +} +/* + * Called to enable a whole group of events. + * Returns 1 if the group was enabled, or -EAGAIN if it could not be. + * Assumes the caller has disabled interrupts and has + * frozen the PMU with hw_perf_save_disable. + * + * called with PMU disabled. If successful and return value 1, + * then guaranteed to call perf_enable() and hw_perf_enable() + */ +int hw_perf_group_sched_in(struct perf_event *leader, + struct perf_cpu_context *cpuctx, + struct perf_event_context *ctx) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct perf_event *sub; + int assign[X86_PMC_IDX_MAX]; + int n0, n1, ret; + + /* n0 = total number of events */ + n0 = collect_events(cpuc, leader, true); + if (n0 < 0) + return n0; + + ret = x86_schedule_events(cpuc, n0, assign); + if (ret) + return ret; + + ret = x86_event_sched_in(leader, cpuctx); + if (ret) + return ret; + + n1 = 1; + list_for_each_entry(sub, &leader->sibling_list, group_entry) { + if (sub->state > PERF_EVENT_STATE_OFF) { + ret = x86_event_sched_in(sub, cpuctx); + if (ret) + goto undo; + ++n1; + } + } /* - * Quirk: v2 perfmon does not report fixed-purpose events, so - * assume at least 3 events: + * copy new assignment, now we know it is possible + * will be used by hw_perf_enable() */ - x86_pmu.num_events_fixed = max((int)edx.split.num_events_fixed, 3); + memcpy(cpuc->assign, assign, n0*sizeof(int)); + + cpuc->n_events = n0; + cpuc->n_added += n1; + ctx->nr_active += n1; /* - * Install the hw-cache-events table: + * 1 means successful and events are active + * This is not quite true because we defer + * actual activation until hw_perf_enable() but + * this way we* ensure caller won't try to enable + * individual events */ - switch (boot_cpu_data.x86_model) { - case 15: /* original 65 nm celeron/pentium/core2/xeon, "Merom"/"Conroe" */ - case 22: /* single-core 65 nm celeron/core2solo "Merom-L"/"Conroe-L" */ - case 23: /* current 45 nm celeron/core2/xeon "Penryn"/"Wolfdale" */ - case 29: /* six-core 45 nm xeon "Dunnington" */ - memcpy(hw_cache_event_ids, core2_hw_cache_event_ids, - sizeof(hw_cache_event_ids)); - - pr_cont("Core2 events, "); + return 1; +undo: + x86_event_sched_out(leader, cpuctx); + n0 = 1; + list_for_each_entry(sub, &leader->sibling_list, group_entry) { + if (sub->state == PERF_EVENT_STATE_ACTIVE) { + x86_event_sched_out(sub, cpuctx); + if (++n0 == n1) + break; + } + } + return ret; +} + +#include "perf_event_amd.c" +#include "perf_event_p6.c" +#include "perf_event_intel.c" + +static int __cpuinit +x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) +{ + unsigned int cpu = (long)hcpu; + + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_UP_PREPARE: + if (x86_pmu.cpu_prepare) + x86_pmu.cpu_prepare(cpu); + break; + + case CPU_STARTING: + if (x86_pmu.cpu_starting) + x86_pmu.cpu_starting(cpu); + break; + + case CPU_DYING: + if (x86_pmu.cpu_dying) + x86_pmu.cpu_dying(cpu); break; - default: - case 26: - memcpy(hw_cache_event_ids, nehalem_hw_cache_event_ids, - sizeof(hw_cache_event_ids)); - pr_cont("Nehalem/Corei7 events, "); + case CPU_DEAD: + if (x86_pmu.cpu_dead) + x86_pmu.cpu_dead(cpu); break; - case 28: - memcpy(hw_cache_event_ids, atom_hw_cache_event_ids, - sizeof(hw_cache_event_ids)); - pr_cont("Atom events, "); + default: break; } - return 0; + + return NOTIFY_OK; } -static int amd_pmu_init(void) +static void __init pmu_check_apic(void) { - /* Performance-monitoring supported from K7 and later: */ - if (boot_cpu_data.x86 < 6) - return -ENODEV; - - x86_pmu = amd_pmu; - - /* Events are common for all AMDs */ - memcpy(hw_cache_event_ids, amd_hw_cache_event_ids, - sizeof(hw_cache_event_ids)); + if (cpu_has_apic) + return; - return 0; + x86_pmu.apic = 0; + pr_info("no APIC, boot with the \"lapic\" boot parameter to force-enable it.\n"); + pr_info("no hardware sampling interrupt available.\n"); } void __init init_hw_perf_events(void) { + struct event_constraint *c; int err; pr_info("Performance Events: "); @@ -2061,6 +1394,8 @@ void __init init_hw_perf_events(void) return; } + pmu_check_apic(); + pr_cont("%s PMU driver.\n", x86_pmu.name); if (x86_pmu.num_events > X86_PMC_MAX_GENERIC) { @@ -2084,6 +1419,20 @@ void __init init_hw_perf_events(void) perf_events_lapic_init(); register_die_notifier(&perf_event_nmi_notifier); + unconstrained = (struct event_constraint) + __EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_events) - 1, + 0, x86_pmu.num_events); + + if (x86_pmu.event_constraints) { + for_each_event_constraint(c, x86_pmu.event_constraints) { + if (c->cmask != INTEL_ARCH_FIXED_MASK) + continue; + + c->idxmsk64 |= (1ULL << x86_pmu.num_events) - 1; + c->weight += x86_pmu.num_events; + } + } + pr_info("... version: %d\n", x86_pmu.version); pr_info("... bit width: %d\n", x86_pmu.event_bits); pr_info("... generic registers: %d\n", x86_pmu.num_events); @@ -2091,25 +1440,92 @@ void __init init_hw_perf_events(void) pr_info("... max period: %016Lx\n", x86_pmu.max_period); pr_info("... fixed-purpose events: %d\n", x86_pmu.num_events_fixed); pr_info("... event mask: %016Lx\n", perf_event_mask); + + perf_cpu_notifier(x86_pmu_notifier); } static inline void x86_pmu_read(struct perf_event *event) { - x86_perf_event_update(event, &event->hw, event->hw.idx); + x86_perf_event_update(event); } 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, }; +/* + * validate a single event group + * + * validation include: + * - check events are compatible which each other + * - events do not compete for the same counter + * - number of events <= number of counters + * + * validation ensures the group can be loaded onto the + * PMU if it was the only group available. + */ +static int validate_group(struct perf_event *event) +{ + struct perf_event *leader = event->group_leader; + struct cpu_hw_events *fake_cpuc; + int ret, n; + + ret = -ENOMEM; + fake_cpuc = kmalloc(sizeof(*fake_cpuc), GFP_KERNEL | __GFP_ZERO); + if (!fake_cpuc) + goto out; + + /* + * the event is not yet connected with its + * siblings therefore we must first collect + * existing siblings, then add the new event + * before we can simulate the scheduling + */ + ret = -ENOSPC; + n = collect_events(fake_cpuc, leader, true); + if (n < 0) + goto out_free; + + fake_cpuc->n_events = n; + n = collect_events(fake_cpuc, event, false); + if (n < 0) + goto out_free; + + fake_cpuc->n_events = n; + + ret = x86_schedule_events(fake_cpuc, n, NULL); + +out_free: + kfree(fake_cpuc); +out: + return ret; +} + const struct pmu *hw_perf_event_init(struct perf_event *event) { + const struct pmu *tmp; int err; err = __hw_perf_event_init(event); + if (!err) { + /* + * we temporarily connect event to its pmu + * such that validate_group() can classify + * it as an x86 event using is_x86_event() + */ + tmp = event->pmu; + event->pmu = &pmu; + + if (event->group_leader != event) + err = validate_group(event); + + event->pmu = tmp; + } if (err) { if (event->destroy) event->destroy(event); @@ -2132,7 +1548,6 @@ void callchain_store(struct perf_callchain_entry *entry, u64 ip) static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry); static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry); -static DEFINE_PER_CPU(int, in_nmi_frame); static void @@ -2148,9 +1563,6 @@ static void backtrace_warning(void *data, char *msg) static int backtrace_stack(void *data, char *name) { - per_cpu(in_nmi_frame, smp_processor_id()) = - x86_is_stack_id(NMI_STACK, name); - return 0; } @@ -2158,9 +1570,6 @@ static void backtrace_address(void *data, unsigned long addr, int reliable) { struct perf_callchain_entry *entry = data; - if (per_cpu(in_nmi_frame, smp_processor_id())) - return; - if (reliable) callchain_store(entry, addr); } @@ -2170,6 +1579,7 @@ static const struct stacktrace_ops backtrace_ops = { .warning_symbol = backtrace_warning_symbol, .stack = backtrace_stack, .address = backtrace_address, + .walk_stack = print_context_stack_bp, }; #include "../dumpstack.h" @@ -2180,7 +1590,7 @@ perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) callchain_store(entry, PERF_CONTEXT_KERNEL); callchain_store(entry, regs->ip); - dump_trace(NULL, regs, NULL, 0, &backtrace_ops, entry); + dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry); } /* @@ -2266,9 +1676,6 @@ perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry) is_user = user_mode(regs); - if (!current || current->pid == 0) - return; - if (is_user && current->state != TASK_RUNNING) return; @@ -2295,7 +1702,16 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) return entry; } -void hw_perf_event_setup_online(int cpu) +#ifdef CONFIG_EVENT_TRACING +void perf_arch_fetch_caller_regs(struct pt_regs *regs, unsigned long ip, int skip) { - init_debug_store_on_cpu(cpu); + regs->ip = ip; + /* + * perf_arch_fetch_caller_regs adds another call, we need to increment + * the skip level + */ + regs->bp = rewind_frame_pointer(skip + 1); + regs->cs = __KERNEL_CS; + local_save_flags(regs->flags); } +#endif diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c new file mode 100644 index 000000000000..573458f1caf2 --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event_amd.c @@ -0,0 +1,406 @@ +#ifdef CONFIG_CPU_SUP_AMD + +static DEFINE_RAW_SPINLOCK(amd_nb_lock); + +static __initconst u64 amd_hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = +{ + [ C(L1D) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ + [ C(RESULT_MISS) ] = 0x0041, /* Data Cache Misses */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x0142, /* Data Cache Refills :system */ + [ C(RESULT_MISS) ] = 0, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0267, /* Data Prefetcher :attempts */ + [ C(RESULT_MISS) ] = 0x0167, /* Data Prefetcher :cancelled */ + }, + }, + [ C(L1I ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0080, /* Instruction cache fetches */ + [ C(RESULT_MISS) ] = 0x0081, /* Instruction cache misses */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x014B, /* Prefetch Instructions :Load */ + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(LL ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x037D, /* Requests to L2 Cache :IC+DC */ + [ C(RESULT_MISS) ] = 0x037E, /* L2 Cache Misses : IC+DC */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x017F, /* L2 Fill/Writeback */ + [ C(RESULT_MISS) ] = 0, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(DTLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ + [ C(RESULT_MISS) ] = 0x0046, /* L1 DTLB and L2 DLTB Miss */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = 0, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(ITLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0080, /* Instruction fecthes */ + [ C(RESULT_MISS) ] = 0x0085, /* Instr. fetch ITLB misses */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(BPU ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x00c2, /* Retired Branch Instr. */ + [ C(RESULT_MISS) ] = 0x00c3, /* Retired Mispredicted BI */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, +}; + +/* + * AMD Performance Monitor K7 and later. + */ +static const u64 amd_perfmon_event_map[] = +{ + [PERF_COUNT_HW_CPU_CYCLES] = 0x0076, + [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, + [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0080, + [PERF_COUNT_HW_CACHE_MISSES] = 0x0081, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c4, + [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c5, +}; + +static u64 amd_pmu_event_map(int hw_event) +{ + return amd_perfmon_event_map[hw_event]; +} + +static u64 amd_pmu_raw_event(u64 hw_event) +{ +#define K7_EVNTSEL_EVENT_MASK 0xF000000FFULL +#define K7_EVNTSEL_UNIT_MASK 0x00000FF00ULL +#define K7_EVNTSEL_EDGE_MASK 0x000040000ULL +#define K7_EVNTSEL_INV_MASK 0x000800000ULL +#define K7_EVNTSEL_REG_MASK 0x0FF000000ULL + +#define K7_EVNTSEL_MASK \ + (K7_EVNTSEL_EVENT_MASK | \ + K7_EVNTSEL_UNIT_MASK | \ + K7_EVNTSEL_EDGE_MASK | \ + K7_EVNTSEL_INV_MASK | \ + K7_EVNTSEL_REG_MASK) + + return hw_event & K7_EVNTSEL_MASK; +} + +/* + * AMD64 events are detected based on their event codes. + */ +static inline int amd_is_nb_event(struct hw_perf_event *hwc) +{ + return (hwc->config & 0xe0) == 0xe0; +} + +static void amd_put_event_constraints(struct cpu_hw_events *cpuc, + struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + struct amd_nb *nb = cpuc->amd_nb; + int i; + + /* + * only care about NB events + */ + if (!(nb && amd_is_nb_event(hwc))) + return; + + /* + * need to scan whole list because event may not have + * been assigned during scheduling + * + * no race condition possible because event can only + * be removed on one CPU at a time AND PMU is disabled + * when we come here + */ + for (i = 0; i < x86_pmu.num_events; i++) { + if (nb->owners[i] == event) { + cmpxchg(nb->owners+i, event, NULL); + break; + } + } +} + + /* + * AMD64 NorthBridge events need special treatment because + * counter access needs to be synchronized across all cores + * of a package. Refer to BKDG section 3.12 + * + * NB events are events measuring L3 cache, Hypertransport + * traffic. They are identified by an event code >= 0xe00. + * They measure events on the NorthBride which is shared + * by all cores on a package. NB events are counted on a + * shared set of counters. When a NB event is programmed + * in a counter, the data actually comes from a shared + * counter. Thus, access to those counters needs to be + * synchronized. + * + * We implement the synchronization such that no two cores + * can be measuring NB events using the same counters. Thus, + * we maintain a per-NB allocation table. The available slot + * is propagated using the event_constraint structure. + * + * We provide only one choice for each NB event based on + * the fact that only NB events have restrictions. Consequently, + * if a counter is available, there is a guarantee the NB event + * will be assigned to it. If no slot is available, an empty + * constraint is returned and scheduling will eventually fail + * for this event. + * + * Note that all cores attached the same NB compete for the same + * counters to host NB events, this is why we use atomic ops. Some + * multi-chip CPUs may have more than one NB. + * + * Given that resources are allocated (cmpxchg), they must be + * eventually freed for others to use. This is accomplished by + * calling amd_put_event_constraints(). + * + * Non NB events are not impacted by this restriction. + */ +static struct event_constraint * +amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + struct amd_nb *nb = cpuc->amd_nb; + struct perf_event *old = NULL; + int max = x86_pmu.num_events; + int i, j, k = -1; + + /* + * if not NB event or no NB, then no constraints + */ + if (!(nb && amd_is_nb_event(hwc))) + return &unconstrained; + + /* + * detect if already present, if so reuse + * + * cannot merge with actual allocation + * because of possible holes + * + * event can already be present yet not assigned (in hwc->idx) + * because of successive calls to x86_schedule_events() from + * hw_perf_group_sched_in() without hw_perf_enable() + */ + for (i = 0; i < max; i++) { + /* + * keep track of first free slot + */ + if (k == -1 && !nb->owners[i]) + k = i; + + /* already present, reuse */ + if (nb->owners[i] == event) + goto done; + } + /* + * not present, so grab a new slot + * starting either at: + */ + if (hwc->idx != -1) { + /* previous assignment */ + i = hwc->idx; + } else if (k != -1) { + /* start from free slot found */ + i = k; + } else { + /* + * event not found, no slot found in + * first pass, try again from the + * beginning + */ + i = 0; + } + j = i; + do { + old = cmpxchg(nb->owners+i, NULL, event); + if (!old) + break; + if (++i == max) + i = 0; + } while (i != j); +done: + if (!old) + return &nb->event_constraints[i]; + + return &emptyconstraint; +} + +static struct amd_nb *amd_alloc_nb(int cpu, int nb_id) +{ + struct amd_nb *nb; + int i; + + nb = kmalloc(sizeof(struct amd_nb), GFP_KERNEL); + if (!nb) + return NULL; + + memset(nb, 0, sizeof(*nb)); + nb->nb_id = nb_id; + + /* + * initialize all possible NB constraints + */ + for (i = 0; i < x86_pmu.num_events; i++) { + __set_bit(i, nb->event_constraints[i].idxmsk); + nb->event_constraints[i].weight = 1; + } + return nb; +} + +static void amd_pmu_cpu_online(int cpu) +{ + struct cpu_hw_events *cpu1, *cpu2; + struct amd_nb *nb = NULL; + int i, nb_id; + + if (boot_cpu_data.x86_max_cores < 2) + return; + + /* + * function may be called too early in the + * boot process, in which case nb_id is bogus + */ + nb_id = amd_get_nb_id(cpu); + if (nb_id == BAD_APICID) + return; + + cpu1 = &per_cpu(cpu_hw_events, cpu); + cpu1->amd_nb = NULL; + + raw_spin_lock(&amd_nb_lock); + + for_each_online_cpu(i) { + cpu2 = &per_cpu(cpu_hw_events, i); + nb = cpu2->amd_nb; + if (!nb) + continue; + if (nb->nb_id == nb_id) + goto found; + } + + nb = amd_alloc_nb(cpu, nb_id); + if (!nb) { + pr_err("perf_events: failed NB allocation for CPU%d\n", cpu); + raw_spin_unlock(&amd_nb_lock); + return; + } +found: + nb->refcnt++; + cpu1->amd_nb = nb; + + raw_spin_unlock(&amd_nb_lock); +} + +static void amd_pmu_cpu_offline(int cpu) +{ + struct cpu_hw_events *cpuhw; + + if (boot_cpu_data.x86_max_cores < 2) + return; + + cpuhw = &per_cpu(cpu_hw_events, cpu); + + raw_spin_lock(&amd_nb_lock); + + if (--cpuhw->amd_nb->refcnt == 0) + kfree(cpuhw->amd_nb); + + cpuhw->amd_nb = NULL; + + raw_spin_unlock(&amd_nb_lock); +} + +static __initconst struct x86_pmu amd_pmu = { + .name = "AMD", + .handle_irq = x86_pmu_handle_irq, + .disable_all = x86_pmu_disable_all, + .enable_all = x86_pmu_enable_all, + .enable = x86_pmu_enable_event, + .disable = x86_pmu_disable_event, + .eventsel = MSR_K7_EVNTSEL0, + .perfctr = MSR_K7_PERFCTR0, + .event_map = amd_pmu_event_map, + .raw_event = amd_pmu_raw_event, + .max_events = ARRAY_SIZE(amd_perfmon_event_map), + .num_events = 4, + .event_bits = 48, + .event_mask = (1ULL << 48) - 1, + .apic = 1, + /* use highest bit to detect overflow */ + .max_period = (1ULL << 47) - 1, + .get_event_constraints = amd_get_event_constraints, + .put_event_constraints = amd_put_event_constraints, + + .cpu_prepare = amd_pmu_cpu_online, + .cpu_dead = amd_pmu_cpu_offline, +}; + +static __init int amd_pmu_init(void) +{ + /* Performance-monitoring supported from K7 and later: */ + if (boot_cpu_data.x86 < 6) + return -ENODEV; + + x86_pmu = amd_pmu; + + /* Events are common for all AMDs */ + memcpy(hw_cache_event_ids, amd_hw_cache_event_ids, + sizeof(hw_cache_event_ids)); + + return 0; +} + +#else /* CONFIG_CPU_SUP_AMD */ + +static int amd_pmu_init(void) +{ + return 0; +} + +#endif diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c new file mode 100644 index 000000000000..84bfde64a337 --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -0,0 +1,979 @@ +#ifdef CONFIG_CPU_SUP_INTEL + +/* + * Intel PerfMon, used on Core and later. + */ +static const u64 intel_perfmon_event_map[] = +{ + [PERF_COUNT_HW_CPU_CYCLES] = 0x003c, + [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, + [PERF_COUNT_HW_CACHE_REFERENCES] = 0x4f2e, + [PERF_COUNT_HW_CACHE_MISSES] = 0x412e, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c4, + [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c5, + [PERF_COUNT_HW_BUS_CYCLES] = 0x013c, +}; + +static struct event_constraint intel_core_event_constraints[] = +{ + INTEL_EVENT_CONSTRAINT(0x11, 0x2), /* FP_ASSIST */ + INTEL_EVENT_CONSTRAINT(0x12, 0x2), /* MUL */ + INTEL_EVENT_CONSTRAINT(0x13, 0x2), /* DIV */ + INTEL_EVENT_CONSTRAINT(0x14, 0x1), /* CYCLES_DIV_BUSY */ + INTEL_EVENT_CONSTRAINT(0x19, 0x2), /* DELAYED_BYPASS */ + INTEL_EVENT_CONSTRAINT(0xc1, 0x1), /* FP_COMP_INSTR_RET */ + EVENT_CONSTRAINT_END +}; + +static struct event_constraint intel_core2_event_constraints[] = +{ + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + /* + * Core2 has Fixed Counter 2 listed as CPU_CLK_UNHALTED.REF and event + * 0x013c as CPU_CLK_UNHALTED.BUS and specifies there is a fixed + * ratio between these counters. + */ + /* FIXED_EVENT_CONSTRAINT(0x013c, 2), CPU_CLK_UNHALTED.REF */ + INTEL_EVENT_CONSTRAINT(0x10, 0x1), /* FP_COMP_OPS_EXE */ + INTEL_EVENT_CONSTRAINT(0x11, 0x2), /* FP_ASSIST */ + INTEL_EVENT_CONSTRAINT(0x12, 0x2), /* MUL */ + INTEL_EVENT_CONSTRAINT(0x13, 0x2), /* DIV */ + INTEL_EVENT_CONSTRAINT(0x14, 0x1), /* CYCLES_DIV_BUSY */ + INTEL_EVENT_CONSTRAINT(0x18, 0x1), /* IDLE_DURING_DIV */ + INTEL_EVENT_CONSTRAINT(0x19, 0x2), /* DELAYED_BYPASS */ + INTEL_EVENT_CONSTRAINT(0xa1, 0x1), /* RS_UOPS_DISPATCH_CYCLES */ + INTEL_EVENT_CONSTRAINT(0xc9, 0x1), /* ITLB_MISS_RETIRED (T30-9) */ + INTEL_EVENT_CONSTRAINT(0xcb, 0x1), /* MEM_LOAD_RETIRED */ + EVENT_CONSTRAINT_END +}; + +static struct event_constraint intel_nehalem_event_constraints[] = +{ + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + /* FIXED_EVENT_CONSTRAINT(0x013c, 2), CPU_CLK_UNHALTED.REF */ + INTEL_EVENT_CONSTRAINT(0x40, 0x3), /* L1D_CACHE_LD */ + INTEL_EVENT_CONSTRAINT(0x41, 0x3), /* L1D_CACHE_ST */ + INTEL_EVENT_CONSTRAINT(0x42, 0x3), /* L1D_CACHE_LOCK */ + INTEL_EVENT_CONSTRAINT(0x43, 0x3), /* L1D_ALL_REF */ + INTEL_EVENT_CONSTRAINT(0x48, 0x3), /* L1D_PEND_MISS */ + INTEL_EVENT_CONSTRAINT(0x4e, 0x3), /* L1D_PREFETCH */ + INTEL_EVENT_CONSTRAINT(0x51, 0x3), /* L1D */ + INTEL_EVENT_CONSTRAINT(0x63, 0x3), /* CACHE_LOCK_CYCLES */ + EVENT_CONSTRAINT_END +}; + +static struct event_constraint intel_westmere_event_constraints[] = +{ + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + /* FIXED_EVENT_CONSTRAINT(0x013c, 2), CPU_CLK_UNHALTED.REF */ + INTEL_EVENT_CONSTRAINT(0x51, 0x3), /* L1D */ + INTEL_EVENT_CONSTRAINT(0x60, 0x1), /* OFFCORE_REQUESTS_OUTSTANDING */ + INTEL_EVENT_CONSTRAINT(0x63, 0x3), /* CACHE_LOCK_CYCLES */ + EVENT_CONSTRAINT_END +}; + +static struct event_constraint intel_gen_event_constraints[] = +{ + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + /* FIXED_EVENT_CONSTRAINT(0x013c, 2), CPU_CLK_UNHALTED.REF */ + EVENT_CONSTRAINT_END +}; + +static u64 intel_pmu_event_map(int hw_event) +{ + return intel_perfmon_event_map[hw_event]; +} + +static __initconst u64 westmere_hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = +{ + [ C(L1D) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x010b, /* MEM_INST_RETIRED.LOADS */ + [ C(RESULT_MISS) ] = 0x0151, /* L1D.REPL */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x020b, /* MEM_INST_RETURED.STORES */ + [ C(RESULT_MISS) ] = 0x0251, /* L1D.M_REPL */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x014e, /* L1D_PREFETCH.REQUESTS */ + [ C(RESULT_MISS) ] = 0x024e, /* L1D_PREFETCH.MISS */ + }, + }, + [ C(L1I ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS */ + [ C(RESULT_MISS) ] = 0x0280, /* L1I.MISSES */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, + [ C(LL ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0324, /* L2_RQSTS.LOADS */ + [ C(RESULT_MISS) ] = 0x0224, /* L2_RQSTS.LD_MISS */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x0c24, /* L2_RQSTS.RFOS */ + [ C(RESULT_MISS) ] = 0x0824, /* L2_RQSTS.RFO_MISS */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x4f2e, /* LLC Reference */ + [ C(RESULT_MISS) ] = 0x412e, /* LLC Misses */ + }, + }, + [ C(DTLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x010b, /* MEM_INST_RETIRED.LOADS */ + [ C(RESULT_MISS) ] = 0x0108, /* DTLB_LOAD_MISSES.ANY */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x020b, /* MEM_INST_RETURED.STORES */ + [ C(RESULT_MISS) ] = 0x010c, /* MEM_STORE_RETIRED.DTLB_MISS */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, + [ C(ITLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x01c0, /* INST_RETIRED.ANY_P */ + [ C(RESULT_MISS) ] = 0x0185, /* ITLB_MISSES.ANY */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(BPU ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ALL_BRANCHES */ + [ C(RESULT_MISS) ] = 0x03e8, /* BPU_CLEARS.ANY */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, +}; + +static __initconst u64 nehalem_hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = +{ + [ C(L1D) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI */ + [ C(RESULT_MISS) ] = 0x0140, /* L1D_CACHE_LD.I_STATE */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI */ + [ C(RESULT_MISS) ] = 0x0141, /* L1D_CACHE_ST.I_STATE */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x014e, /* L1D_PREFETCH.REQUESTS */ + [ C(RESULT_MISS) ] = 0x024e, /* L1D_PREFETCH.MISS */ + }, + }, + [ C(L1I ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS */ + [ C(RESULT_MISS) ] = 0x0280, /* L1I.MISSES */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, + [ C(LL ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0324, /* L2_RQSTS.LOADS */ + [ C(RESULT_MISS) ] = 0x0224, /* L2_RQSTS.LD_MISS */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x0c24, /* L2_RQSTS.RFOS */ + [ C(RESULT_MISS) ] = 0x0824, /* L2_RQSTS.RFO_MISS */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x4f2e, /* LLC Reference */ + [ C(RESULT_MISS) ] = 0x412e, /* LLC Misses */ + }, + }, + [ C(DTLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI (alias) */ + [ C(RESULT_MISS) ] = 0x0108, /* DTLB_LOAD_MISSES.ANY */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI (alias) */ + [ C(RESULT_MISS) ] = 0x010c, /* MEM_STORE_RETIRED.DTLB_MISS */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, + [ C(ITLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x01c0, /* INST_RETIRED.ANY_P */ + [ C(RESULT_MISS) ] = 0x20c8, /* ITLB_MISS_RETIRED */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(BPU ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ALL_BRANCHES */ + [ C(RESULT_MISS) ] = 0x03e8, /* BPU_CLEARS.ANY */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, +}; + +static __initconst u64 core2_hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = +{ + [ C(L1D) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI */ + [ C(RESULT_MISS) ] = 0x0140, /* L1D_CACHE_LD.I_STATE */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI */ + [ C(RESULT_MISS) ] = 0x0141, /* L1D_CACHE_ST.I_STATE */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x104e, /* L1D_PREFETCH.REQUESTS */ + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(L1I ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0080, /* L1I.READS */ + [ C(RESULT_MISS) ] = 0x0081, /* L1I.MISSES */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(LL ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x4f29, /* L2_LD.MESI */ + [ C(RESULT_MISS) ] = 0x4129, /* L2_LD.ISTATE */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x4f2A, /* L2_ST.MESI */ + [ C(RESULT_MISS) ] = 0x412A, /* L2_ST.ISTATE */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(DTLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI (alias) */ + [ C(RESULT_MISS) ] = 0x0208, /* DTLB_MISSES.MISS_LD */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI (alias) */ + [ C(RESULT_MISS) ] = 0x0808, /* DTLB_MISSES.MISS_ST */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(ITLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x00c0, /* INST_RETIRED.ANY_P */ + [ C(RESULT_MISS) ] = 0x1282, /* ITLBMISSES */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(BPU ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ANY */ + [ C(RESULT_MISS) ] = 0x00c5, /* BP_INST_RETIRED.MISPRED */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, +}; + +static __initconst u64 atom_hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = +{ + [ C(L1D) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x2140, /* L1D_CACHE.LD */ + [ C(RESULT_MISS) ] = 0, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x2240, /* L1D_CACHE.ST */ + [ C(RESULT_MISS) ] = 0, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(L1I ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS */ + [ C(RESULT_MISS) ] = 0x0280, /* L1I.MISSES */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(LL ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x4f29, /* L2_LD.MESI */ + [ C(RESULT_MISS) ] = 0x4129, /* L2_LD.ISTATE */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x4f2A, /* L2_ST.MESI */ + [ C(RESULT_MISS) ] = 0x412A, /* L2_ST.ISTATE */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(DTLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x2140, /* L1D_CACHE_LD.MESI (alias) */ + [ C(RESULT_MISS) ] = 0x0508, /* DTLB_MISSES.MISS_LD */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x2240, /* L1D_CACHE_ST.MESI (alias) */ + [ C(RESULT_MISS) ] = 0x0608, /* DTLB_MISSES.MISS_ST */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(ITLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x00c0, /* INST_RETIRED.ANY_P */ + [ C(RESULT_MISS) ] = 0x0282, /* ITLB.MISSES */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(BPU ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ANY */ + [ C(RESULT_MISS) ] = 0x00c5, /* BP_INST_RETIRED.MISPRED */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, +}; + +static u64 intel_pmu_raw_event(u64 hw_event) +{ +#define CORE_EVNTSEL_EVENT_MASK 0x000000FFULL +#define CORE_EVNTSEL_UNIT_MASK 0x0000FF00ULL +#define CORE_EVNTSEL_EDGE_MASK 0x00040000ULL +#define CORE_EVNTSEL_INV_MASK 0x00800000ULL +#define CORE_EVNTSEL_REG_MASK 0xFF000000ULL + +#define CORE_EVNTSEL_MASK \ + (INTEL_ARCH_EVTSEL_MASK | \ + INTEL_ARCH_UNIT_MASK | \ + INTEL_ARCH_EDGE_MASK | \ + INTEL_ARCH_INV_MASK | \ + INTEL_ARCH_CNT_MASK) + + return hw_event & CORE_EVNTSEL_MASK; +} + +static void intel_pmu_enable_bts(u64 config) +{ + unsigned long debugctlmsr; + + debugctlmsr = get_debugctlmsr(); + + debugctlmsr |= X86_DEBUGCTL_TR; + debugctlmsr |= X86_DEBUGCTL_BTS; + debugctlmsr |= X86_DEBUGCTL_BTINT; + + if (!(config & ARCH_PERFMON_EVENTSEL_OS)) + debugctlmsr |= X86_DEBUGCTL_BTS_OFF_OS; + + if (!(config & ARCH_PERFMON_EVENTSEL_USR)) + debugctlmsr |= X86_DEBUGCTL_BTS_OFF_USR; + + update_debugctlmsr(debugctlmsr); +} + +static void intel_pmu_disable_bts(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + unsigned long debugctlmsr; + + if (!cpuc->ds) + return; + + debugctlmsr = get_debugctlmsr(); + + debugctlmsr &= + ~(X86_DEBUGCTL_TR | X86_DEBUGCTL_BTS | X86_DEBUGCTL_BTINT | + X86_DEBUGCTL_BTS_OFF_OS | X86_DEBUGCTL_BTS_OFF_USR); + + update_debugctlmsr(debugctlmsr); +} + +static void intel_pmu_disable_all(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0); + + if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) + intel_pmu_disable_bts(); +} + +static void intel_pmu_enable_all(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl); + + if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) { + struct perf_event *event = + cpuc->events[X86_PMC_IDX_FIXED_BTS]; + + if (WARN_ON_ONCE(!event)) + return; + + intel_pmu_enable_bts(event->hw.config); + } +} + +static inline u64 intel_pmu_get_status(void) +{ + u64 status; + + rdmsrl(MSR_CORE_PERF_GLOBAL_STATUS, status); + + return status; +} + +static inline void intel_pmu_ack_status(u64 ack) +{ + wrmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, ack); +} + +static inline void +intel_pmu_disable_fixed(struct hw_perf_event *hwc) +{ + int idx = hwc->idx - X86_PMC_IDX_FIXED; + u64 ctrl_val, mask; + + mask = 0xfULL << (idx * 4); + + rdmsrl(hwc->config_base, ctrl_val); + ctrl_val &= ~mask; + (void)checking_wrmsrl(hwc->config_base, ctrl_val); +} + +static void intel_pmu_drain_bts_buffer(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct debug_store *ds = cpuc->ds; + struct bts_record { + u64 from; + u64 to; + u64 flags; + }; + struct perf_event *event = cpuc->events[X86_PMC_IDX_FIXED_BTS]; + struct bts_record *at, *top; + struct perf_output_handle handle; + struct perf_event_header header; + struct perf_sample_data data; + struct pt_regs regs; + + if (!event) + return; + + if (!ds) + return; + + at = (struct bts_record *)(unsigned long)ds->bts_buffer_base; + top = (struct bts_record *)(unsigned long)ds->bts_index; + + if (top <= at) + return; + + ds->bts_index = ds->bts_buffer_base; + + perf_sample_data_init(&data, 0); + + data.period = event->hw.last_period; + regs.ip = 0; + + /* + * Prepare a generic sample, i.e. fill in the invariant fields. + * We will overwrite the from and to address before we output + * the sample. + */ + perf_prepare_sample(&header, &data, event, ®s); + + if (perf_output_begin(&handle, event, + header.size * (top - at), 1, 1)) + return; + + for (; at < top; at++) { + data.ip = at->from; + data.addr = at->to; + + perf_output_sample(&handle, &header, &data, event); + } + + perf_output_end(&handle); + + /* There's new data available. */ + event->hw.interrupts++; + event->pending_kill = POLL_IN; +} + +static inline void +intel_pmu_disable_event(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) { + intel_pmu_disable_bts(); + intel_pmu_drain_bts_buffer(); + return; + } + + if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { + intel_pmu_disable_fixed(hwc); + return; + } + + x86_pmu_disable_event(event); +} + +static inline void +intel_pmu_enable_fixed(struct hw_perf_event *hwc) +{ + int idx = hwc->idx - X86_PMC_IDX_FIXED; + u64 ctrl_val, bits, mask; + int err; + + /* + * Enable IRQ generation (0x8), + * and enable ring-3 counting (0x2) and ring-0 counting (0x1) + * if requested: + */ + bits = 0x8ULL; + if (hwc->config & ARCH_PERFMON_EVENTSEL_USR) + bits |= 0x2; + if (hwc->config & ARCH_PERFMON_EVENTSEL_OS) + bits |= 0x1; + + /* + * ANY bit is supported in v3 and up + */ + if (x86_pmu.version > 2 && hwc->config & ARCH_PERFMON_EVENTSEL_ANY) + bits |= 0x4; + + bits <<= (idx * 4); + mask = 0xfULL << (idx * 4); + + rdmsrl(hwc->config_base, ctrl_val); + ctrl_val &= ~mask; + ctrl_val |= bits; + err = checking_wrmsrl(hwc->config_base, ctrl_val); +} + +static void intel_pmu_enable_event(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) { + if (!__get_cpu_var(cpu_hw_events).enabled) + return; + + intel_pmu_enable_bts(hwc->config); + return; + } + + if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { + intel_pmu_enable_fixed(hwc); + return; + } + + __x86_pmu_enable_event(hwc); +} + +/* + * Save and restart an expired event. Called by NMI contexts, + * so it has to be careful about preempting normal event ops: + */ +static int intel_pmu_save_and_restart(struct perf_event *event) +{ + x86_perf_event_update(event); + return x86_perf_event_set_period(event); +} + +static void intel_pmu_reset(void) +{ + struct debug_store *ds = __get_cpu_var(cpu_hw_events).ds; + unsigned long flags; + int idx; + + if (!x86_pmu.num_events) + return; + + local_irq_save(flags); + + printk("clearing PMU state on CPU#%d\n", smp_processor_id()); + + for (idx = 0; idx < x86_pmu.num_events; idx++) { + checking_wrmsrl(x86_pmu.eventsel + idx, 0ull); + checking_wrmsrl(x86_pmu.perfctr + idx, 0ull); + } + for (idx = 0; idx < x86_pmu.num_events_fixed; idx++) { + checking_wrmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, 0ull); + } + if (ds) + ds->bts_index = ds->bts_buffer_base; + + local_irq_restore(flags); +} + +/* + * This handler is triggered by the local APIC, so the APIC IRQ handling + * rules apply: + */ +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; + + perf_sample_data_init(&data, 0); + + cpuc = &__get_cpu_var(cpu_hw_events); + + intel_pmu_disable_all(); + intel_pmu_drain_bts_buffer(); + status = intel_pmu_get_status(); + if (!status) { + intel_pmu_enable_all(); + return 0; + } + + loops = 0; +again: + if (++loops > 100) { + WARN_ONCE(1, "perfevents: irq loop stuck!\n"); + perf_event_print_debug(); + intel_pmu_reset(); + goto done; + } + + inc_irq_stat(apic_perf_irqs); + ack = status; + for_each_set_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) { + struct perf_event *event = cpuc->events[bit]; + + if (!test_bit(bit, cpuc->active_mask)) + continue; + + if (!intel_pmu_save_and_restart(event)) + continue; + + data.period = event->hw.last_period; + + if (perf_event_overflow(event, 1, &data, regs)) + x86_pmu_stop(event); + } + + intel_pmu_ack_status(ack); + + /* + * Repeat if there is more work to be done: + */ + status = intel_pmu_get_status(); + if (status) + goto again; + +done: + intel_pmu_enable_all(); + return 1; +} + +static struct event_constraint bts_constraint = + EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0); + +static struct event_constraint * +intel_special_constraints(struct perf_event *event) +{ + unsigned int hw_event; + + hw_event = event->hw.config & INTEL_ARCH_EVENT_MASK; + + if (unlikely((hw_event == + x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS)) && + (event->hw.sample_period == 1))) { + + return &bts_constraint; + } + return NULL; +} + +static struct event_constraint * +intel_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) +{ + struct event_constraint *c; + + c = intel_special_constraints(event); + if (c) + return c; + + return x86_get_event_constraints(cpuc, event); +} + +static __initconst struct x86_pmu core_pmu = { + .name = "core", + .handle_irq = x86_pmu_handle_irq, + .disable_all = x86_pmu_disable_all, + .enable_all = x86_pmu_enable_all, + .enable = x86_pmu_enable_event, + .disable = x86_pmu_disable_event, + .eventsel = MSR_ARCH_PERFMON_EVENTSEL0, + .perfctr = MSR_ARCH_PERFMON_PERFCTR0, + .event_map = intel_pmu_event_map, + .raw_event = intel_pmu_raw_event, + .max_events = ARRAY_SIZE(intel_perfmon_event_map), + .apic = 1, + /* + * Intel PMCs cannot be accessed sanely above 32 bit width, + * so we install an artificial 1<<31 period regardless of + * the generic event period: + */ + .max_period = (1ULL << 31) - 1, + .get_event_constraints = intel_get_event_constraints, + .event_constraints = intel_core_event_constraints, +}; + +static __initconst struct x86_pmu intel_pmu = { + .name = "Intel", + .handle_irq = intel_pmu_handle_irq, + .disable_all = intel_pmu_disable_all, + .enable_all = intel_pmu_enable_all, + .enable = intel_pmu_enable_event, + .disable = intel_pmu_disable_event, + .eventsel = MSR_ARCH_PERFMON_EVENTSEL0, + .perfctr = MSR_ARCH_PERFMON_PERFCTR0, + .event_map = intel_pmu_event_map, + .raw_event = intel_pmu_raw_event, + .max_events = ARRAY_SIZE(intel_perfmon_event_map), + .apic = 1, + /* + * Intel PMCs cannot be accessed sanely above 32 bit width, + * so we install an artificial 1<<31 period regardless of + * the generic event period: + */ + .max_period = (1ULL << 31) - 1, + .enable_bts = intel_pmu_enable_bts, + .disable_bts = intel_pmu_disable_bts, + .get_event_constraints = intel_get_event_constraints, + + .cpu_starting = init_debug_store_on_cpu, + .cpu_dying = fini_debug_store_on_cpu, +}; + +static __init int intel_pmu_init(void) +{ + union cpuid10_edx edx; + union cpuid10_eax eax; + unsigned int unused; + unsigned int ebx; + int version; + + if (!cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) { + /* check for P6 processor family */ + if (boot_cpu_data.x86 == 6) { + return p6_pmu_init(); + } else { + return -ENODEV; + } + } + + /* + * Check whether the Architectural PerfMon supports + * Branch Misses Retired hw_event or not. + */ + cpuid(10, &eax.full, &ebx, &unused, &edx.full); + if (eax.split.mask_length <= ARCH_PERFMON_BRANCH_MISSES_RETIRED) + return -ENODEV; + + version = eax.split.version_id; + if (version < 2) + x86_pmu = core_pmu; + else + x86_pmu = intel_pmu; + + x86_pmu.version = version; + x86_pmu.num_events = eax.split.num_events; + x86_pmu.event_bits = eax.split.bit_width; + x86_pmu.event_mask = (1ULL << eax.split.bit_width) - 1; + + /* + * Quirk: v2 perfmon does not report fixed-purpose events, so + * assume at least 3 events: + */ + if (version > 1) + x86_pmu.num_events_fixed = max((int)edx.split.num_events_fixed, 3); + + /* + * Install the hw-cache-events table: + */ + switch (boot_cpu_data.x86_model) { + case 14: /* 65 nm core solo/duo, "Yonah" */ + pr_cont("Core events, "); + break; + + case 15: /* original 65 nm celeron/pentium/core2/xeon, "Merom"/"Conroe" */ + case 22: /* single-core 65 nm celeron/core2solo "Merom-L"/"Conroe-L" */ + case 23: /* current 45 nm celeron/core2/xeon "Penryn"/"Wolfdale" */ + case 29: /* six-core 45 nm xeon "Dunnington" */ + memcpy(hw_cache_event_ids, core2_hw_cache_event_ids, + sizeof(hw_cache_event_ids)); + + x86_pmu.event_constraints = intel_core2_event_constraints; + pr_cont("Core2 events, "); + break; + + case 26: /* 45 nm nehalem, "Bloomfield" */ + case 30: /* 45 nm nehalem, "Lynnfield" */ + memcpy(hw_cache_event_ids, nehalem_hw_cache_event_ids, + sizeof(hw_cache_event_ids)); + + x86_pmu.event_constraints = intel_nehalem_event_constraints; + pr_cont("Nehalem/Corei7 events, "); + break; + case 28: /* Atom */ + memcpy(hw_cache_event_ids, atom_hw_cache_event_ids, + sizeof(hw_cache_event_ids)); + + x86_pmu.event_constraints = intel_gen_event_constraints; + pr_cont("Atom events, "); + break; + + case 37: /* 32 nm nehalem, "Clarkdale" */ + case 44: /* 32 nm nehalem, "Gulftown" */ + memcpy(hw_cache_event_ids, westmere_hw_cache_event_ids, + sizeof(hw_cache_event_ids)); + + x86_pmu.event_constraints = intel_westmere_event_constraints; + pr_cont("Westmere events, "); + break; + + default: + /* + * default constraints for v2 and up + */ + x86_pmu.event_constraints = intel_gen_event_constraints; + pr_cont("generic architected perfmon, "); + } + return 0; +} + +#else /* CONFIG_CPU_SUP_INTEL */ + +static int intel_pmu_init(void) +{ + return 0; +} + +#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_p6.c b/arch/x86/kernel/cpu/perf_event_p6.c new file mode 100644 index 000000000000..a330485d14da --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event_p6.c @@ -0,0 +1,159 @@ +#ifdef CONFIG_CPU_SUP_INTEL + +/* + * Not sure about some of these + */ +static const u64 p6_perfmon_event_map[] = +{ + [PERF_COUNT_HW_CPU_CYCLES] = 0x0079, + [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, + [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0f2e, + [PERF_COUNT_HW_CACHE_MISSES] = 0x012e, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c4, + [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c5, + [PERF_COUNT_HW_BUS_CYCLES] = 0x0062, +}; + +static u64 p6_pmu_event_map(int hw_event) +{ + return p6_perfmon_event_map[hw_event]; +} + +/* + * Event setting that is specified not to count anything. + * We use this to effectively disable a counter. + * + * L2_RQSTS with 0 MESI unit mask. + */ +#define P6_NOP_EVENT 0x0000002EULL + +static u64 p6_pmu_raw_event(u64 hw_event) +{ +#define P6_EVNTSEL_EVENT_MASK 0x000000FFULL +#define P6_EVNTSEL_UNIT_MASK 0x0000FF00ULL +#define P6_EVNTSEL_EDGE_MASK 0x00040000ULL +#define P6_EVNTSEL_INV_MASK 0x00800000ULL +#define P6_EVNTSEL_REG_MASK 0xFF000000ULL + +#define P6_EVNTSEL_MASK \ + (P6_EVNTSEL_EVENT_MASK | \ + P6_EVNTSEL_UNIT_MASK | \ + P6_EVNTSEL_EDGE_MASK | \ + P6_EVNTSEL_INV_MASK | \ + P6_EVNTSEL_REG_MASK) + + return hw_event & P6_EVNTSEL_MASK; +} + +static struct event_constraint p6_event_constraints[] = +{ + INTEL_EVENT_CONSTRAINT(0xc1, 0x1), /* FLOPS */ + INTEL_EVENT_CONSTRAINT(0x10, 0x1), /* FP_COMP_OPS_EXE */ + INTEL_EVENT_CONSTRAINT(0x11, 0x1), /* FP_ASSIST */ + INTEL_EVENT_CONSTRAINT(0x12, 0x2), /* MUL */ + INTEL_EVENT_CONSTRAINT(0x13, 0x2), /* DIV */ + INTEL_EVENT_CONSTRAINT(0x14, 0x1), /* CYCLES_DIV_BUSY */ + EVENT_CONSTRAINT_END +}; + +static void p6_pmu_disable_all(void) +{ + u64 val; + + /* p6 only has one enable register */ + rdmsrl(MSR_P6_EVNTSEL0, val); + val &= ~ARCH_PERFMON_EVENTSEL_ENABLE; + wrmsrl(MSR_P6_EVNTSEL0, val); +} + +static void p6_pmu_enable_all(void) +{ + unsigned long val; + + /* p6 only has one enable register */ + rdmsrl(MSR_P6_EVNTSEL0, val); + val |= ARCH_PERFMON_EVENTSEL_ENABLE; + wrmsrl(MSR_P6_EVNTSEL0, val); +} + +static inline void +p6_pmu_disable_event(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + u64 val = P6_NOP_EVENT; + + if (cpuc->enabled) + val |= ARCH_PERFMON_EVENTSEL_ENABLE; + + (void)checking_wrmsrl(hwc->config_base + hwc->idx, val); +} + +static void p6_pmu_enable_event(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + u64 val; + + val = hwc->config; + if (cpuc->enabled) + val |= ARCH_PERFMON_EVENTSEL_ENABLE; + + (void)checking_wrmsrl(hwc->config_base + hwc->idx, val); +} + +static __initconst struct x86_pmu p6_pmu = { + .name = "p6", + .handle_irq = x86_pmu_handle_irq, + .disable_all = p6_pmu_disable_all, + .enable_all = p6_pmu_enable_all, + .enable = p6_pmu_enable_event, + .disable = p6_pmu_disable_event, + .eventsel = MSR_P6_EVNTSEL0, + .perfctr = MSR_P6_PERFCTR0, + .event_map = p6_pmu_event_map, + .raw_event = p6_pmu_raw_event, + .max_events = ARRAY_SIZE(p6_perfmon_event_map), + .apic = 1, + .max_period = (1ULL << 31) - 1, + .version = 0, + .num_events = 2, + /* + * Events have 40 bits implemented. However they are designed such + * that bits [32-39] are sign extensions of bit 31. As such the + * effective width of a event for P6-like PMU is 32 bits only. + * + * See IA-32 Intel Architecture Software developer manual Vol 3B + */ + .event_bits = 32, + .event_mask = (1ULL << 32) - 1, + .get_event_constraints = x86_get_event_constraints, + .event_constraints = p6_event_constraints, +}; + +static __init int p6_pmu_init(void) +{ + switch (boot_cpu_data.x86_model) { + case 1: + case 3: /* Pentium Pro */ + case 5: + case 6: /* Pentium II */ + case 7: + case 8: + case 11: /* Pentium III */ + case 9: + case 13: + /* Pentium M */ + break; + default: + pr_cont("unsupported p6 CPU model %d ", + boot_cpu_data.x86_model); + return -ENODEV; + } + + x86_pmu = p6_pmu; + + return 0; +} + +#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index fab786f60ed6..fb329e9f8494 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -115,17 +115,6 @@ int avail_to_resrv_perfctr_nmi_bit(unsigned int counter) return !test_bit(counter, perfctr_nmi_owner); } - -/* checks the an msr for availability */ -int avail_to_resrv_perfctr_nmi(unsigned int msr) -{ - unsigned int counter; - - counter = nmi_perfctr_msr_to_bit(msr); - BUG_ON(counter > NMI_MAX_COUNTER_BITS); - - return !test_bit(counter, perfctr_nmi_owner); -} EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit); int reserve_perfctr_nmi(unsigned int msr) @@ -691,7 +680,7 @@ static int setup_intel_arch_watchdog(unsigned nmi_hz) cpu_nmi_set_wd_enabled(); apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE; + evntsel |= ARCH_PERFMON_EVENTSEL_ENABLE; wrmsr(evntsel_msr, evntsel, 0); intel_arch_wd_ops.checkbit = 1ULL << (eax.split.bit_width - 1); return 1; @@ -712,7 +701,7 @@ 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 != 16 && boot_cpu_data.x86 != 17) return; wd_ops = &k7_wd_ops; break; diff --git a/arch/x86/kernel/cpu/transmeta.c b/arch/x86/kernel/cpu/transmeta.c index bb62b3e5caad..28000743bbb0 100644 --- a/arch/x86/kernel/cpu/transmeta.c +++ b/arch/x86/kernel/cpu/transmeta.c @@ -26,7 +26,7 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c) early_init_transmeta(c); - display_cacheinfo(c); + cpu_detect_cache_sizes(c); /* Print CMS and CPU revision */ max = cpuid_eax(0x80860000); diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c index 6a52d4b36a30..83e5e628de73 100644 --- a/arch/x86/kernel/cpuid.c +++ b/arch/x86/kernel/cpuid.c @@ -116,21 +116,16 @@ static int cpuid_open(struct inode *inode, struct file *file) { unsigned int cpu; struct cpuinfo_x86 *c; - int ret = 0; - - lock_kernel(); cpu = iminor(file->f_path.dentry->d_inode); - if (cpu >= nr_cpu_ids || !cpu_online(cpu)) { - ret = -ENXIO; /* No such CPU */ - goto out; - } + if (cpu >= nr_cpu_ids || !cpu_online(cpu)) + return -ENXIO; /* No such CPU */ + c = &cpu_data(cpu); if (c->cpuid_level < 0) - ret = -EIO; /* CPUID not supported */ -out: - unlock_kernel(); - return ret; + return -EIO; /* CPUID not supported */ + + return 0; } /* @@ -192,7 +187,8 @@ static int __init cpuid_init(void) int i, err = 0; i = 0; - if (register_chrdev(CPUID_MAJOR, "cpu/cpuid", &cpuid_fops)) { + if (__register_chrdev(CPUID_MAJOR, 0, NR_CPUS, + "cpu/cpuid", &cpuid_fops)) { printk(KERN_ERR "cpuid: unable to get major %d for cpuid\n", CPUID_MAJOR); err = -EBUSY; @@ -221,7 +217,7 @@ out_class: } class_destroy(cpuid_class); out_chrdev: - unregister_chrdev(CPUID_MAJOR, "cpu/cpuid"); + __unregister_chrdev(CPUID_MAJOR, 0, NR_CPUS, "cpu/cpuid"); out: return err; } @@ -233,7 +229,7 @@ static void __exit cpuid_exit(void) for_each_online_cpu(cpu) cpuid_device_destroy(cpu); class_destroy(cpuid_class); - unregister_chrdev(CPUID_MAJOR, "cpu/cpuid"); + __unregister_chrdev(CPUID_MAJOR, 0, NR_CPUS, "cpu/cpuid"); unregister_hotcpu_notifier(&cpuid_class_cpu_notifier); } diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index 5e409dc298a4..a4849c10a77e 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -27,8 +27,7 @@ #include <asm/cpu.h> #include <asm/reboot.h> #include <asm/virtext.h> -#include <asm/iommu.h> - +#include <asm/x86_init.h> #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) @@ -106,7 +105,7 @@ void native_machine_crash_shutdown(struct pt_regs *regs) #endif #ifdef CONFIG_X86_64 - pci_iommu_shutdown(); + x86_platform.iommu_shutdown(); #endif crash_save_cpu(regs, safe_smp_processor_id()); diff --git a/arch/x86/kernel/crash_dump_32.c b/arch/x86/kernel/crash_dump_32.c index f7cdb3b457aa..cd97ce18c29d 100644 --- a/arch/x86/kernel/crash_dump_32.c +++ b/arch/x86/kernel/crash_dump_32.c @@ -16,6 +16,22 @@ static void *kdump_buf_page; /* Stores the physical address of elf header of crash image. */ unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX; +static inline bool is_crashed_pfn_valid(unsigned long pfn) +{ +#ifndef CONFIG_X86_PAE + /* + * non-PAE kdump kernel executed from a PAE one will crop high pte + * bits and poke unwanted space counting again from address 0, we + * don't want that. pte must fit into unsigned long. In fact the + * test checks high 12 bits for being zero (pfn will be shifted left + * by PAGE_SHIFT). + */ + return pte_pfn(pfn_pte(pfn, __pgprot(0))) == pfn; +#else + return true; +#endif +} + /** * copy_oldmem_page - copy one page from "oldmem" * @pfn: page frame number to be copied @@ -41,6 +57,9 @@ ssize_t copy_oldmem_page(unsigned long pfn, char *buf, if (!csize) return 0; + if (!is_crashed_pfn_valid(pfn)) + return -EFAULT; + vaddr = kmap_atomic_pfn(pfn, KM_PTE0); if (!userbuf) { diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index ef42a038f1a6..1c47390dd0e5 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -265,13 +265,13 @@ struct ds_context { int cpu; }; -static DEFINE_PER_CPU(struct ds_context *, cpu_context); +static DEFINE_PER_CPU(struct ds_context *, cpu_ds_context); static struct ds_context *ds_get_context(struct task_struct *task, int cpu) { struct ds_context **p_context = - (task ? &task->thread.ds_ctx : &per_cpu(cpu_context, cpu)); + (task ? &task->thread.ds_ctx : &per_cpu(cpu_ds_context, cpu)); struct ds_context *context = NULL; struct ds_context *new_context = NULL; diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index 2d8a371d4339..6d817554780a 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -109,6 +109,32 @@ print_context_stack(struct thread_info *tinfo, } return bp; } +EXPORT_SYMBOL_GPL(print_context_stack); + +unsigned long +print_context_stack_bp(struct thread_info *tinfo, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data, + unsigned long *end, int *graph) +{ + struct stack_frame *frame = (struct stack_frame *)bp; + unsigned long *ret_addr = &frame->return_address; + + while (valid_stack_ptr(tinfo, ret_addr, sizeof(*ret_addr), end)) { + unsigned long addr = *ret_addr; + + if (!__kernel_text_address(addr)) + break; + + ops->address(data, addr, 1); + frame = frame->next_frame; + ret_addr = &frame->return_address; + print_ftrace_graph_addr(addr, data, ops, tinfo, graph); + } + + return (unsigned long)frame; +} +EXPORT_SYMBOL_GPL(print_context_stack_bp); static void @@ -141,10 +167,11 @@ static void print_trace_address(void *data, unsigned long addr, int reliable) } static const struct stacktrace_ops print_trace_ops = { - .warning = print_trace_warning, - .warning_symbol = print_trace_warning_symbol, - .stack = print_trace_stack, - .address = print_trace_address, + .warning = print_trace_warning, + .warning_symbol = print_trace_warning_symbol, + .stack = print_trace_stack, + .address = print_trace_address, + .walk_stack = print_context_stack, }; void @@ -188,7 +215,7 @@ void dump_stack(void) } EXPORT_SYMBOL(dump_stack); -static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t die_lock = __ARCH_SPIN_LOCK_UNLOCKED; static int die_owner = -1; static unsigned int die_nest_count; @@ -207,11 +234,11 @@ unsigned __kprobes long oops_begin(void) /* racy, but better than risking deadlock. */ raw_local_irq_save(flags); cpu = smp_processor_id(); - if (!__raw_spin_trylock(&die_lock)) { + if (!arch_spin_trylock(&die_lock)) { if (cpu == die_owner) /* nested oops. should stop eventually */; else - __raw_spin_lock(&die_lock); + arch_spin_lock(&die_lock); } die_nest_count++; die_owner = cpu; @@ -231,7 +258,7 @@ void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) die_nest_count--; if (!die_nest_count) /* Nest count reaches zero, release the lock. */ - __raw_spin_unlock(&die_lock); + arch_spin_unlock(&die_lock); raw_local_irq_restore(flags); oops_exit(); @@ -268,11 +295,12 @@ int __kprobes __die(const char *str, struct pt_regs *regs, long err) show_registers(regs); #ifdef CONFIG_X86_32 - sp = (unsigned long) (®s->sp); - savesegment(ss, ss); - if (user_mode(regs)) { + if (user_mode_vm(regs)) { sp = regs->sp; ss = regs->ss & 0xffff; + } else { + sp = kernel_stack_pointer(regs); + savesegment(ss, ss); } printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip); print_symbol("%s", regs->ip); diff --git a/arch/x86/kernel/dumpstack.h b/arch/x86/kernel/dumpstack.h index 81086c227ab7..29e5f7c845b2 100644 --- a/arch/x86/kernel/dumpstack.h +++ b/arch/x86/kernel/dumpstack.h @@ -14,12 +14,6 @@ #define get_bp(bp) asm("movq %%rbp, %0" : "=r" (bp) :) #endif -extern unsigned long -print_context_stack(struct thread_info *tinfo, - unsigned long *stack, unsigned long bp, - const struct stacktrace_ops *ops, void *data, - unsigned long *end, int *graph); - extern void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, char *log_lvl); @@ -35,4 +29,19 @@ struct stack_frame { struct stack_frame *next_frame; unsigned long return_address; }; + +static inline unsigned long rewind_frame_pointer(int n) +{ + struct stack_frame *frame; + + get_bp(frame); + +#ifdef CONFIG_FRAME_POINTER + while (n--) + frame = frame->next_frame; #endif + + return (unsigned long)frame; +} + +#endif /* DUMPSTACK_H */ diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c index f7dd2a7c3bf4..11540a189d93 100644 --- a/arch/x86/kernel/dumpstack_32.c +++ b/arch/x86/kernel/dumpstack_32.c @@ -10,19 +10,14 @@ #include <linux/module.h> #include <linux/ptrace.h> #include <linux/kexec.h> +#include <linux/sysfs.h> #include <linux/bug.h> #include <linux/nmi.h> -#include <linux/sysfs.h> #include <asm/stacktrace.h> #include "dumpstack.h" -/* Just a stub for now */ -int x86_is_stack_id(int id, char *name) -{ - return 0; -} void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, @@ -35,6 +30,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, if (!stack) { unsigned long dummy; + stack = &dummy; if (task && task != current) stack = (unsigned long *)task->thread.sp; @@ -57,8 +53,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, context = (struct thread_info *) ((unsigned long)stack & (~(THREAD_SIZE - 1))); - bp = print_context_stack(context, stack, bp, ops, - data, NULL, &graph); + bp = ops->walk_stack(context, stack, bp, ops, data, NULL, &graph); stack = (unsigned long *)context->previous_esp; if (!stack) @@ -72,7 +67,7 @@ EXPORT_SYMBOL(dump_trace); void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *sp, unsigned long bp, char *log_lvl) + unsigned long *sp, unsigned long bp, char *log_lvl) { unsigned long *stack; int i; @@ -156,4 +151,3 @@ int is_valid_bugaddr(unsigned long ip) return ud2 == 0x0b0f; } - diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c index a071e6be177e..272c9f1f05f3 100644 --- a/arch/x86/kernel/dumpstack_64.c +++ b/arch/x86/kernel/dumpstack_64.c @@ -10,34 +10,31 @@ #include <linux/module.h> #include <linux/ptrace.h> #include <linux/kexec.h> +#include <linux/sysfs.h> #include <linux/bug.h> #include <linux/nmi.h> -#include <linux/sysfs.h> #include <asm/stacktrace.h> #include "dumpstack.h" +#define N_EXCEPTION_STACKS_END \ + (N_EXCEPTION_STACKS + DEBUG_STKSZ/EXCEPTION_STKSZ - 2) static char x86_stack_ids[][8] = { - [DEBUG_STACK - 1] = "#DB", - [NMI_STACK - 1] = "NMI", - [DOUBLEFAULT_STACK - 1] = "#DF", - [STACKFAULT_STACK - 1] = "#SS", - [MCE_STACK - 1] = "#MC", + [ DEBUG_STACK-1 ] = "#DB", + [ NMI_STACK-1 ] = "NMI", + [ DOUBLEFAULT_STACK-1 ] = "#DF", + [ STACKFAULT_STACK-1 ] = "#SS", + [ MCE_STACK-1 ] = "#MC", #if DEBUG_STKSZ > EXCEPTION_STKSZ - [N_EXCEPTION_STACKS ... - N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]" + [ N_EXCEPTION_STACKS ... + N_EXCEPTION_STACKS_END ] = "#DB[?]" #endif - }; - -int x86_is_stack_id(int id, char *name) -{ - return x86_stack_ids[id - 1] == name; -} +}; static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, - unsigned *usedp, char **idp) + unsigned *usedp, char **idp) { unsigned k; @@ -101,6 +98,41 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, return NULL; } +static inline int +in_irq_stack(unsigned long *stack, unsigned long *irq_stack, + unsigned long *irq_stack_end) +{ + return (stack >= irq_stack && stack < irq_stack_end); +} + +/* + * We are returning from the irq stack and go to the previous one. + * If the previous stack is also in the irq stack, then bp in the first + * frame of the irq stack points to the previous, interrupted one. + * Otherwise we have another level of indirection: We first save + * the bp of the previous stack, then we switch the stack to the irq one + * and save a new bp that links to the previous one. + * (See save_args()) + */ +static inline unsigned long +fixup_bp_irq_link(unsigned long bp, unsigned long *stack, + unsigned long *irq_stack, unsigned long *irq_stack_end) +{ +#ifdef CONFIG_FRAME_POINTER + struct stack_frame *frame = (struct stack_frame *)bp; + unsigned long next; + + if (!in_irq_stack(stack, irq_stack, irq_stack_end)) { + if (!probe_kernel_address(&frame->next_frame, next)) + return next; + else + WARN_ONCE(1, "Perf: bad frame pointer = %p in " + "callchain\n", &frame->next_frame); + } +#endif + return bp; +} + /* * x86-64 can have up to three kernel stacks: * process stack @@ -157,8 +189,8 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, if (ops->stack(data, id) < 0) break; - bp = print_context_stack(tinfo, stack, bp, ops, - data, estack_end, &graph); + bp = ops->walk_stack(tinfo, stack, bp, ops, + data, estack_end, &graph); ops->stack(data, "<EOE>"); /* * We link to the next stack via the @@ -173,10 +205,10 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, irq_stack = irq_stack_end - (IRQ_STACK_SIZE - 64) / sizeof(*irq_stack); - if (stack >= irq_stack && stack < irq_stack_end) { + if (in_irq_stack(stack, irq_stack, irq_stack_end)) { if (ops->stack(data, "IRQ") < 0) break; - bp = print_context_stack(tinfo, stack, bp, + bp = ops->walk_stack(tinfo, stack, bp, ops, data, irq_stack_end, &graph); /* * We link to the next stack (which would be @@ -184,6 +216,8 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, * pointer (index -1 to end) in the IRQ stack: */ stack = (unsigned long *) (irq_stack_end[-1]); + bp = fixup_bp_irq_link(bp, stack, irq_stack, + irq_stack_end); irq_stack_end = NULL; ops->stack(data, "EOI"); continue; @@ -195,28 +229,31 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, /* * This handles the process stack: */ - bp = print_context_stack(tinfo, stack, bp, ops, data, NULL, &graph); + bp = ops->walk_stack(tinfo, stack, bp, ops, data, NULL, &graph); put_cpu(); } EXPORT_SYMBOL(dump_trace); void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *sp, unsigned long bp, char *log_lvl) + unsigned long *sp, unsigned long bp, char *log_lvl) { + unsigned long *irq_stack_end; + unsigned long *irq_stack; unsigned long *stack; + int cpu; int i; - const int cpu = smp_processor_id(); - unsigned long *irq_stack_end = - (unsigned long *)(per_cpu(irq_stack_ptr, cpu)); - unsigned long *irq_stack = - (unsigned long *)(per_cpu(irq_stack_ptr, cpu) - IRQ_STACK_SIZE); + + preempt_disable(); + cpu = smp_processor_id(); + + irq_stack_end = (unsigned long *)(per_cpu(irq_stack_ptr, cpu)); + irq_stack = (unsigned long *)(per_cpu(irq_stack_ptr, cpu) - IRQ_STACK_SIZE); /* - * debugging aid: "show_stack(NULL, NULL);" prints the - * back trace for this cpu. + * Debugging aid: "show_stack(NULL, NULL);" prints the + * back trace for this cpu: */ - if (sp == NULL) { if (task) sp = (unsigned long *)task->thread.sp; @@ -240,6 +277,8 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, printk(" %016lx", *stack++); touch_nmi_watchdog(); } + preempt_enable(); + printk("\n"); show_trace_log_lvl(task, regs, sp, bp, log_lvl); } @@ -253,6 +292,7 @@ void show_registers(struct pt_regs *regs) sp = regs->sp; printk("CPU %d ", cpu); + print_modules(); __show_regs(regs, 1); printk("Process %s (pid: %d, threadinfo %p, task %p)\n", cur->comm, cur->pid, task_thread_info(cur), cur); @@ -303,4 +343,3 @@ int is_valid_bugaddr(unsigned long ip) return ud2 == 0x0b0f; } - diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 85419bb7d4ab..740b440fbd73 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -12,21 +12,13 @@ #include <linux/types.h> #include <linux/init.h> #include <linux/bootmem.h> -#include <linux/ioport.h> -#include <linux/string.h> -#include <linux/kexec.h> -#include <linux/module.h> -#include <linux/mm.h> #include <linux/pfn.h> #include <linux/suspend.h> #include <linux/firmware-map.h> -#include <asm/pgtable.h> -#include <asm/page.h> #include <asm/e820.h> #include <asm/proto.h> #include <asm/setup.h> -#include <asm/trampoline.h> /* * The e820 map is the map that gets modified e.g. with command line parameters @@ -517,11 +509,19 @@ u64 __init e820_remove_range(u64 start, u64 size, unsigned old_type, int checktype) { int i; + u64 end; u64 real_removed_size = 0; if (size > (ULLONG_MAX - start)) size = ULLONG_MAX - start; + end = start + size; + printk(KERN_DEBUG "e820 remove range: %016Lx - %016Lx ", + (unsigned long long) start, + (unsigned long long) end); + e820_print_type(old_type); + printk(KERN_CONT "\n"); + for (i = 0; i < e820.nr_map; i++) { struct e820entry *ei = &e820.map[i]; u64 final_start, final_end; @@ -722,310 +722,44 @@ core_initcall(e820_mark_nvs_memory); #endif /* - * Early reserved memory areas. - */ -#define MAX_EARLY_RES 20 - -struct early_res { - u64 start, end; - char name[16]; - char overlap_ok; -}; -static struct early_res early_res[MAX_EARLY_RES] __initdata = { - { 0, PAGE_SIZE, "BIOS data page" }, /* BIOS data page */ - {} -}; - -static int __init find_overlapped_early(u64 start, u64 end) -{ - int i; - struct early_res *r; - - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - r = &early_res[i]; - if (end > r->start && start < r->end) - break; - } - - return i; -} - -/* - * Drop the i-th range from the early reservation map, - * by copying any higher ranges down one over it, and - * clearing what had been the last slot. - */ -static void __init drop_range(int i) -{ - int j; - - for (j = i + 1; j < MAX_EARLY_RES && early_res[j].end; j++) - ; - - memmove(&early_res[i], &early_res[i + 1], - (j - 1 - i) * sizeof(struct early_res)); - - early_res[j - 1].end = 0; -} - -/* - * Split any existing ranges that: - * 1) are marked 'overlap_ok', and - * 2) overlap with the stated range [start, end) - * into whatever portion (if any) of the existing range is entirely - * below or entirely above the stated range. Drop the portion - * of the existing range that overlaps with the stated range, - * which will allow the caller of this routine to then add that - * stated range without conflicting with any existing range. + * Find a free area with specified alignment in a specific range. */ -static void __init drop_overlaps_that_are_ok(u64 start, u64 end) +u64 __init find_e820_area(u64 start, u64 end, u64 size, u64 align) { int i; - struct early_res *r; - u64 lower_start, lower_end; - u64 upper_start, upper_end; - char name[16]; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - r = &early_res[i]; + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + u64 addr; + u64 ei_start, ei_last; - /* Continue past non-overlapping ranges */ - if (end <= r->start || start >= r->end) + if (ei->type != E820_RAM) continue; - /* - * Leave non-ok overlaps as is; let caller - * panic "Overlapping early reservations" - * when it hits this overlap. - */ - if (!r->overlap_ok) - return; - - /* - * We have an ok overlap. We will drop it from the early - * reservation map, and add back in any non-overlapping - * portions (lower or upper) as separate, overlap_ok, - * non-overlapping ranges. - */ - - /* 1. Note any non-overlapping (lower or upper) ranges. */ - strncpy(name, r->name, sizeof(name) - 1); - - lower_start = lower_end = 0; - upper_start = upper_end = 0; - if (r->start < start) { - lower_start = r->start; - lower_end = start; - } - if (r->end > end) { - upper_start = end; - upper_end = r->end; - } - - /* 2. Drop the original ok overlapping range */ - drop_range(i); - - i--; /* resume for-loop on copied down entry */ - - /* 3. Add back in any non-overlapping ranges. */ - if (lower_end) - reserve_early_overlap_ok(lower_start, lower_end, name); - if (upper_end) - reserve_early_overlap_ok(upper_start, upper_end, name); - } -} - -static void __init __reserve_early(u64 start, u64 end, char *name, - int overlap_ok) -{ - int i; - struct early_res *r; - - i = find_overlapped_early(start, end); - if (i >= MAX_EARLY_RES) - panic("Too many early reservations"); - r = &early_res[i]; - if (r->end) - panic("Overlapping early reservations " - "%llx-%llx %s to %llx-%llx %s\n", - start, end - 1, name?name:"", r->start, - r->end - 1, r->name); - r->start = start; - r->end = end; - r->overlap_ok = overlap_ok; - if (name) - strncpy(r->name, name, sizeof(r->name) - 1); -} - -/* - * A few early reservtations come here. - * - * The 'overlap_ok' in the name of this routine does -not- mean it - * is ok for these reservations to overlap an earlier reservation. - * Rather it means that it is ok for subsequent reservations to - * overlap this one. - * - * Use this entry point to reserve early ranges when you are doing - * so out of "Paranoia", reserving perhaps more memory than you need, - * just in case, and don't mind a subsequent overlapping reservation - * that is known to be needed. - * - * The drop_overlaps_that_are_ok() call here isn't really needed. - * It would be needed if we had two colliding 'overlap_ok' - * reservations, so that the second such would not panic on the - * overlap with the first. We don't have any such as of this - * writing, but might as well tolerate such if it happens in - * the future. - */ -void __init reserve_early_overlap_ok(u64 start, u64 end, char *name) -{ - drop_overlaps_that_are_ok(start, end); - __reserve_early(start, end, name, 1); -} - -/* - * Most early reservations come here. - * - * We first have drop_overlaps_that_are_ok() drop any pre-existing - * 'overlap_ok' ranges, so that we can then reserve this memory - * range without risk of panic'ing on an overlapping overlap_ok - * early reservation. - */ -void __init reserve_early(u64 start, u64 end, char *name) -{ - if (start >= end) - return; - - drop_overlaps_that_are_ok(start, end); - __reserve_early(start, end, name, 0); -} - -void __init free_early(u64 start, u64 end) -{ - struct early_res *r; - int i; - - i = find_overlapped_early(start, end); - r = &early_res[i]; - if (i >= MAX_EARLY_RES || r->end != end || r->start != start) - panic("free_early on not reserved area: %llx-%llx!", - start, end - 1); - - drop_range(i); -} + ei_last = ei->addr + ei->size; + ei_start = ei->addr; + addr = find_early_area(ei_start, ei_last, start, end, + size, align); -void __init early_res_to_bootmem(u64 start, u64 end) -{ - int i, count; - u64 final_start, final_end; - - count = 0; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) - count++; - - printk(KERN_INFO "(%d early reservations) ==> bootmem [%010llx - %010llx]\n", - count, start, end); - for (i = 0; i < count; i++) { - struct early_res *r = &early_res[i]; - printk(KERN_INFO " #%d [%010llx - %010llx] %16s", i, - r->start, r->end, r->name); - final_start = max(start, r->start); - final_end = min(end, r->end); - if (final_start >= final_end) { - printk(KERN_CONT "\n"); - continue; - } - printk(KERN_CONT " ==> [%010llx - %010llx]\n", - final_start, final_end); - reserve_bootmem_generic(final_start, final_end - final_start, - BOOTMEM_DEFAULT); + if (addr != -1ULL) + return addr; } + return -1ULL; } -/* Check for already reserved areas */ -static inline int __init bad_addr(u64 *addrp, u64 size, u64 align) +u64 __init find_fw_memmap_area(u64 start, u64 end, u64 size, u64 align) { - int i; - u64 addr = *addrp; - int changed = 0; - struct early_res *r; -again: - i = find_overlapped_early(addr, addr + size); - r = &early_res[i]; - if (i < MAX_EARLY_RES && r->end) { - *addrp = addr = round_up(r->end, align); - changed = 1; - goto again; - } - return changed; + return find_e820_area(start, end, size, align); } -/* Check for already reserved areas */ -static inline int __init bad_addr_size(u64 *addrp, u64 *sizep, u64 align) +u64 __init get_max_mapped(void) { - int i; - u64 addr = *addrp, last; - u64 size = *sizep; - int changed = 0; -again: - last = addr + size; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - struct early_res *r = &early_res[i]; - if (last > r->start && addr < r->start) { - size = r->start - addr; - changed = 1; - goto again; - } - if (last > r->end && addr < r->end) { - addr = round_up(r->end, align); - size = last - addr; - changed = 1; - goto again; - } - if (last <= r->end && addr >= r->start) { - (*sizep)++; - return 0; - } - } - if (changed) { - *addrp = addr; - *sizep = size; - } - return changed; -} + u64 end = max_pfn_mapped; -/* - * 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, last; - u64 ei_last; + end <<= PAGE_SHIFT; - if (ei->type != E820_RAM) - continue; - addr = round_up(ei->addr, align); - ei_last = ei->addr + ei->size; - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - continue; - while (bad_addr(&addr, size, align) && addr+size <= ei_last) - ; - last = addr + size; - if (last > ei_last) - continue; - if (last > end) - continue; - return addr; - } - return -1ULL; + return end; } - /* * Find next free range after *start */ @@ -1035,25 +769,19 @@ u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align) for (i = 0; i < e820.nr_map; i++) { struct e820entry *ei = &e820.map[i]; - u64 addr, last; - u64 ei_last; + u64 addr; + u64 ei_start, ei_last; if (ei->type != E820_RAM) continue; - addr = round_up(ei->addr, align); + ei_last = ei->addr + ei->size; - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - continue; - *sizep = ei_last - addr; - while (bad_addr_size(&addr, sizep, align) && - addr + *sizep <= ei_last) - ; - last = addr + *sizep; - if (last > ei_last) - continue; - return addr; + ei_start = ei->addr; + addr = find_early_area_size(ei_start, ei_last, start, + sizep, align); + + if (addr != -1ULL) + return addr; } return -1ULL; @@ -1378,8 +1106,8 @@ static unsigned long ram_alignment(resource_size_t pos) if (mb < 16) return 1024*1024; - /* To 32MB for anything above that */ - return 32*1024*1024; + /* To 64MB for anything above that */ + return 64*1024*1024; } #define MAX_RESOURCE_SIZE ((resource_size_t)-1) @@ -1412,6 +1140,8 @@ void __init e820_reserve_resources_late(void) end = MAX_RESOURCE_SIZE; if (start >= end) continue; + printk(KERN_DEBUG "reserve RAM buffer: %016llx - %016llx ", + start, end); reserve_region_with_split(&iomem_resource, start, end, "RAM buffer"); } diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c index 41fd965c80c6..b9c830c12b4a 100644 --- a/arch/x86/kernel/early_printk.c +++ b/arch/x86/kernel/early_printk.c @@ -206,8 +206,11 @@ static int __init setup_early_printk(char *buf) while (*buf != '\0') { if (!strncmp(buf, "serial", 6)) { - early_serial_init(buf + 6); + buf += 6; + early_serial_init(buf); early_console_register(&early_serial_console, keep); + if (!strncmp(buf, ",ttyS", 5)) + buf += 5; } if (!strncmp(buf, "ttyS", 4)) { early_serial_init(buf + 4); diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index ad5bd988fb79..c2fa9b8b497e 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -362,7 +362,7 @@ void __init efi_init(void) printk(KERN_ERR PFX "Could not map the firmware vendor!\n"); early_iounmap(tmp, 2); - printk(KERN_INFO "EFI v%u.%.02u by %s \n", + printk(KERN_INFO "EFI v%u.%.02u by %s\n", efi.systab->hdr.revision >> 16, efi.systab->hdr.revision & 0xffff, vendor); @@ -454,8 +454,10 @@ void __init efi_init(void) if (add_efi_memmap) do_add_efi_memmap(); +#ifdef CONFIG_X86_32 x86_platform.get_wallclock = efi_get_time; x86_platform.set_wallclock = efi_set_rtc_mmss; +#endif /* Setup for EFI runtime service */ reboot_type = BOOT_EFI; diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index c097e7d607c6..44a8e0dc6737 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -334,6 +334,10 @@ ENTRY(ret_from_fork) END(ret_from_fork) /* + * Interrupt exit functions should be protected against kprobes + */ + .pushsection .kprobes.text, "ax" +/* * Return to user mode is not as complex as all this looks, * but we want the default path for a system call return to * go as quickly as possible which is why some of this is @@ -383,6 +387,10 @@ need_resched: END(resume_kernel) #endif CFI_ENDPROC +/* + * End of kprobes section + */ + .popsection /* SYSENTER_RETURN points to after the "sysenter" instruction in the vsyscall page. See vsyscall-sysentry.S, which defines the symbol. */ @@ -513,6 +521,10 @@ sysexit_audit: PTGS_TO_GS_EX ENDPROC(ia32_sysenter_target) +/* + * syscall stub including irq exit should be protected against kprobes + */ + .pushsection .kprobes.text, "ax" # system call handler stub ENTRY(system_call) RING0_INT_FRAME # can't unwind into user space anyway @@ -705,26 +717,69 @@ syscall_badsys: jmp resume_userspace END(syscall_badsys) CFI_ENDPROC +/* + * End of kprobes section + */ + .popsection /* * System calls that need a pt_regs pointer. */ -#define PTREGSCALL(name) \ +#define PTREGSCALL0(name) \ ALIGN; \ ptregs_##name: \ leal 4(%esp),%eax; \ jmp sys_##name; -PTREGSCALL(iopl) -PTREGSCALL(fork) -PTREGSCALL(clone) -PTREGSCALL(vfork) -PTREGSCALL(execve) -PTREGSCALL(sigaltstack) -PTREGSCALL(sigreturn) -PTREGSCALL(rt_sigreturn) -PTREGSCALL(vm86) -PTREGSCALL(vm86old) +#define PTREGSCALL1(name) \ + ALIGN; \ +ptregs_##name: \ + leal 4(%esp),%edx; \ + movl (PT_EBX+4)(%esp),%eax; \ + jmp sys_##name; + +#define PTREGSCALL2(name) \ + ALIGN; \ +ptregs_##name: \ + leal 4(%esp),%ecx; \ + movl (PT_ECX+4)(%esp),%edx; \ + movl (PT_EBX+4)(%esp),%eax; \ + jmp sys_##name; + +#define PTREGSCALL3(name) \ + ALIGN; \ +ptregs_##name: \ + leal 4(%esp),%eax; \ + pushl %eax; \ + movl PT_EDX(%eax),%ecx; \ + movl PT_ECX(%eax),%edx; \ + movl PT_EBX(%eax),%eax; \ + call sys_##name; \ + addl $4,%esp; \ + ret + +PTREGSCALL1(iopl) +PTREGSCALL0(fork) +PTREGSCALL0(vfork) +PTREGSCALL3(execve) +PTREGSCALL2(sigaltstack) +PTREGSCALL0(sigreturn) +PTREGSCALL0(rt_sigreturn) +PTREGSCALL2(vm86) +PTREGSCALL1(vm86old) + +/* Clone is an oddball. The 4th arg is in %edi */ + ALIGN; +ptregs_clone: + leal 4(%esp),%eax + pushl %eax + pushl PT_EDI(%eax) + movl PT_EDX(%eax),%ecx + movl PT_ECX(%eax),%edx + movl PT_EBX(%eax),%eax + call sys_clone + addl $8,%esp + ret .macro FIXUP_ESPFIX_STACK /* @@ -814,6 +869,10 @@ common_interrupt: ENDPROC(common_interrupt) CFI_ENDPROC +/* + * Irq entries should be protected against kprobes + */ + .pushsection .kprobes.text, "ax" #define BUILD_INTERRUPT3(name, nr, fn) \ ENTRY(name) \ RING0_INT_FRAME; \ @@ -980,16 +1039,16 @@ ENTRY(spurious_interrupt_bug) jmp error_code CFI_ENDPROC END(spurious_interrupt_bug) +/* + * End of kprobes section + */ + .popsection ENTRY(kernel_thread_helper) pushl $0 # fake return address for unwinder CFI_STARTPROC - movl %edx,%eax - push %edx - CFI_ADJUST_CFA_OFFSET 4 - call *%ebx - push %eax - CFI_ADJUST_CFA_OFFSET 4 + movl %edi,%eax + call *%esi call do_exit ud2 # padding for call trace CFI_ENDPROC @@ -1185,17 +1244,14 @@ END(ftrace_graph_caller) .globl return_to_handler return_to_handler: - pushl $0 pushl %eax - pushl %ecx pushl %edx movl %ebp, %eax call ftrace_return_to_handler - movl %eax, 0xc(%esp) + movl %eax, %ecx popl %edx - popl %ecx popl %eax - ret + jmp *%ecx #endif .section .rodata,"a" diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index b5c061f8f358..0697ff139837 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -155,11 +155,11 @@ GLOBAL(return_to_handler) call ftrace_return_to_handler - movq %rax, 16(%rsp) + movq %rax, %rdi movq 8(%rsp), %rdx movq (%rsp), %rax - addq $16, %rsp - retq + addq $24, %rsp + jmp *%rdi #endif @@ -803,6 +803,10 @@ END(interrupt) call \func .endm +/* + * Interrupt entry/exit should be protected against kprobes + */ + .pushsection .kprobes.text, "ax" /* * The interrupt stubs push (~vector+0x80) onto the stack and * then jump to common_interrupt. @@ -941,6 +945,10 @@ ENTRY(retint_kernel) CFI_ENDPROC END(common_interrupt) +/* + * End of kprobes section + */ + .popsection /* * APIC interrupts. @@ -969,8 +977,8 @@ apicinterrupt UV_BAU_MESSAGE \ #endif apicinterrupt LOCAL_TIMER_VECTOR \ apic_timer_interrupt smp_apic_timer_interrupt -apicinterrupt GENERIC_INTERRUPT_VECTOR \ - generic_interrupt smp_generic_interrupt +apicinterrupt X86_PLATFORM_IPI_VECTOR \ + x86_platform_ipi smp_x86_platform_ipi #ifdef CONFIG_SMP apicinterrupt INVALIDATE_TLB_VECTOR_START+0 \ @@ -1068,10 +1076,10 @@ ENTRY(\sym) TRACE_IRQS_OFF movq %rsp,%rdi /* pt_regs pointer */ xorl %esi,%esi /* no error code */ - PER_CPU(init_tss, %rbp) - subq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%rbp) + PER_CPU(init_tss, %r12) + subq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%r12) call \do_sym - addq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%rbp) + addq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%r12) jmp paranoid_exit /* %ebx: no swapgs flag */ CFI_ENDPROC END(\sym) @@ -1158,63 +1166,20 @@ bad_gs: jmp 2b .previous -/* - * Create a kernel thread. - * - * C extern interface: - * extern long kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) - * - * asm input arguments: - * rdi: fn, rsi: arg, rdx: flags - */ -ENTRY(kernel_thread) - CFI_STARTPROC - FAKE_STACK_FRAME $child_rip - SAVE_ALL - - # rdi: flags, rsi: usp, rdx: will be &pt_regs - movq %rdx,%rdi - orq kernel_thread_flags(%rip),%rdi - movq $-1, %rsi - movq %rsp, %rdx - - xorl %r8d,%r8d - xorl %r9d,%r9d - - # clone now - call do_fork - movq %rax,RAX(%rsp) - xorl %edi,%edi - - /* - * It isn't worth to check for reschedule here, - * so internally to the x86_64 port you can rely on kernel_thread() - * not to reschedule the child before returning, this avoids the need - * of hacks for example to fork off the per-CPU idle tasks. - * [Hopefully no generic code relies on the reschedule -AK] - */ - RESTORE_ALL - UNFAKE_STACK_FRAME - ret - CFI_ENDPROC -END(kernel_thread) - -ENTRY(child_rip) +ENTRY(kernel_thread_helper) pushq $0 # fake return address CFI_STARTPROC /* * Here we are in the child and the registers are set as they were * at kernel_thread() invocation in the parent. */ - movq %rdi, %rax - movq %rsi, %rdi - call *%rax + call *%rsi # exit mov %eax, %edi call do_exit ud2 # padding for call trace CFI_ENDPROC -END(child_rip) +END(kernel_thread_helper) /* * execve(). This function needs to use IRET, not SYSRET, to set up all state properly. @@ -1491,12 +1456,17 @@ error_kernelspace: leaq irq_return(%rip),%rcx cmpq %rcx,RIP+8(%rsp) je error_swapgs - movl %ecx,%ecx /* zero extend */ - cmpq %rcx,RIP+8(%rsp) - je error_swapgs + movl %ecx,%eax /* zero extend */ + cmpq %rax,RIP+8(%rsp) + je bstep_iret cmpq $gs_change,RIP+8(%rsp) je error_swapgs jmp error_sti + +bstep_iret: + /* Fix truncated RIP */ + movq %rcx,RIP+8(%rsp) + jmp error_swapgs END(error_entry) diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index 9dbb527e1652..cd37469b54ee 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -9,6 +9,8 @@ * the dangers of modifying code on the run. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/spinlock.h> #include <linux/hardirq.h> #include <linux/uaccess.h> @@ -28,14 +30,32 @@ #ifdef CONFIG_DYNAMIC_FTRACE +/* + * modifying_code is set to notify NMIs that they need to use + * memory barriers when entering or exiting. But we don't want + * to burden NMIs with unnecessary memory barriers when code + * modification is not being done (which is most of the time). + * + * A mutex is already held when ftrace_arch_code_modify_prepare + * and post_process are called. No locks need to be taken here. + * + * Stop machine will make sure currently running NMIs are done + * and new NMIs will see the updated variable before we need + * to worry about NMIs doing memory barriers. + */ +static int modifying_code __read_mostly; +static DEFINE_PER_CPU(int, save_modifying_code); + int ftrace_arch_code_modify_prepare(void) { set_kernel_text_rw(); + modifying_code = 1; return 0; } int ftrace_arch_code_modify_post_process(void) { + modifying_code = 0; set_kernel_text_ro(); return 0; } @@ -147,6 +167,11 @@ static void ftrace_mod_code(void) void ftrace_nmi_enter(void) { + __get_cpu_var(save_modifying_code) = modifying_code; + + if (!__get_cpu_var(save_modifying_code)) + return; + if (atomic_inc_return(&nmi_running) & MOD_CODE_WRITE_FLAG) { smp_rmb(); ftrace_mod_code(); @@ -158,6 +183,9 @@ void ftrace_nmi_enter(void) void ftrace_nmi_exit(void) { + if (!__get_cpu_var(save_modifying_code)) + return; + /* Finish all executions before clearing nmi_running */ smp_mb(); atomic_dec(&nmi_running); @@ -187,9 +215,26 @@ static void wait_for_nmi(void) nmi_wait_count++; } +static inline int +within(unsigned long addr, unsigned long start, unsigned long end) +{ + return addr >= start && addr < end; +} + static int do_ftrace_mod_code(unsigned long ip, void *new_code) { + /* + * On x86_64, kernel text mappings are mapped read-only with + * CONFIG_DEBUG_RODATA. So we use the kernel identity mapping instead + * of the kernel text mapping to modify the kernel text. + * + * For 32bit kernels, these mappings are same and we can use + * kernel identity mapping to modify code. + */ + if (within(ip, (unsigned long)_text, (unsigned long)_etext)) + ip = (unsigned long)__va(__pa(ip)); + mod_code_ip = (void *)ip; mod_code_newcode = new_code; @@ -336,15 +381,15 @@ int __init ftrace_dyn_arch_init(void *data) switch (faulted) { case 0: - pr_info("ftrace: converting mcount calls to 0f 1f 44 00 00\n"); + 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("ftrace: converting mcount calls to 66 66 66 66 90\n"); + 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("ftrace: converting mcount calls to jmp . + 5\n"); + pr_info("converting mcount calls to jmp . + 5\n"); memcpy(ftrace_nop, ftrace_test_jmp, MCOUNT_INSN_SIZE); break; } @@ -465,85 +510,3 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, } } #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ - -#ifdef CONFIG_FTRACE_SYSCALLS - -extern unsigned long __start_syscalls_metadata[]; -extern unsigned long __stop_syscalls_metadata[]; -extern unsigned long *sys_call_table; - -static struct syscall_metadata **syscalls_metadata; - -static struct syscall_metadata *find_syscall_meta(unsigned long *syscall) -{ - struct syscall_metadata *start; - struct syscall_metadata *stop; - char str[KSYM_SYMBOL_LEN]; - - - start = (struct syscall_metadata *)__start_syscalls_metadata; - stop = (struct syscall_metadata *)__stop_syscalls_metadata; - kallsyms_lookup((unsigned long) syscall, NULL, NULL, NULL, str); - - for ( ; start < stop; start++) { - if (start->name && !strcmp(start->name, str)) - return start; - } - return NULL; -} - -struct syscall_metadata *syscall_nr_to_meta(int nr) -{ - if (!syscalls_metadata || nr >= NR_syscalls || nr < 0) - return NULL; - - return syscalls_metadata[nr]; -} - -int syscall_name_to_nr(char *name) -{ - int i; - - if (!syscalls_metadata) - return -1; - - for (i = 0; i < NR_syscalls; i++) { - if (syscalls_metadata[i]) { - if (!strcmp(syscalls_metadata[i]->name, name)) - return i; - } - } - return -1; -} - -void set_syscall_enter_id(int num, int id) -{ - syscalls_metadata[num]->enter_id = id; -} - -void set_syscall_exit_id(int num, int id) -{ - syscalls_metadata[num]->exit_id = id; -} - -static int __init arch_init_ftrace_syscalls(void) -{ - int i; - struct syscall_metadata *meta; - unsigned long **psys_syscall_table = &sys_call_table; - - syscalls_metadata = kzalloc(sizeof(*syscalls_metadata) * - NR_syscalls, GFP_KERNEL); - if (!syscalls_metadata) { - WARN_ON(1); - return -ENOMEM; - } - - for (i = 0; i < NR_syscalls; i++) { - meta = find_syscall_meta(psys_syscall_table[i]); - syscalls_metadata[i] = meta; - } - return 0; -} -arch_initcall(arch_init_ftrace_syscalls); -#endif diff --git a/arch/x86/kernel/geode_32.c b/arch/x86/kernel/geode_32.c deleted file mode 100644 index 9b08e852fd1a..000000000000 --- a/arch/x86/kernel/geode_32.c +++ /dev/null @@ -1,196 +0,0 @@ -/* - * AMD Geode southbridge support code - * Copyright (C) 2006, Advanced Micro Devices, Inc. - * Copyright (C) 2007, Andres Salomon <dilinger@debian.org> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of version 2 of the GNU General Public License - * as published by the Free Software Foundation. - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/ioport.h> -#include <linux/io.h> -#include <asm/msr.h> -#include <asm/geode.h> - -static struct { - char *name; - u32 msr; - int size; - u32 base; -} lbars[] = { - { "geode-pms", MSR_LBAR_PMS, LBAR_PMS_SIZE, 0 }, - { "geode-acpi", MSR_LBAR_ACPI, LBAR_ACPI_SIZE, 0 }, - { "geode-gpio", MSR_LBAR_GPIO, LBAR_GPIO_SIZE, 0 }, - { "geode-mfgpt", MSR_LBAR_MFGPT, LBAR_MFGPT_SIZE, 0 } -}; - -static void __init init_lbars(void) -{ - u32 lo, hi; - int i; - - for (i = 0; i < ARRAY_SIZE(lbars); i++) { - rdmsr(lbars[i].msr, lo, hi); - if (hi & 0x01) - lbars[i].base = lo & 0x0000ffff; - - if (lbars[i].base == 0) - printk(KERN_ERR "geode: Couldn't initialize '%s'\n", - lbars[i].name); - } -} - -int geode_get_dev_base(unsigned int dev) -{ - BUG_ON(dev >= ARRAY_SIZE(lbars)); - return lbars[dev].base; -} -EXPORT_SYMBOL_GPL(geode_get_dev_base); - -/* === GPIO API === */ - -void geode_gpio_set(u32 gpio, unsigned int reg) -{ - u32 base = geode_get_dev_base(GEODE_DEV_GPIO); - - if (!base) - return; - - /* low bank register */ - if (gpio & 0xFFFF) - outl(gpio & 0xFFFF, base + reg); - /* high bank register */ - gpio >>= 16; - if (gpio) - outl(gpio, base + 0x80 + reg); -} -EXPORT_SYMBOL_GPL(geode_gpio_set); - -void geode_gpio_clear(u32 gpio, unsigned int reg) -{ - u32 base = geode_get_dev_base(GEODE_DEV_GPIO); - - if (!base) - return; - - /* low bank register */ - if (gpio & 0xFFFF) - outl((gpio & 0xFFFF) << 16, base + reg); - /* high bank register */ - gpio &= (0xFFFF << 16); - if (gpio) - outl(gpio, base + 0x80 + reg); -} -EXPORT_SYMBOL_GPL(geode_gpio_clear); - -int geode_gpio_isset(u32 gpio, unsigned int reg) -{ - u32 base = geode_get_dev_base(GEODE_DEV_GPIO); - u32 val; - - if (!base) - return 0; - - /* low bank register */ - if (gpio & 0xFFFF) { - val = inl(base + reg) & (gpio & 0xFFFF); - if ((gpio & 0xFFFF) == val) - return 1; - } - /* high bank register */ - gpio >>= 16; - if (gpio) { - val = inl(base + 0x80 + reg) & gpio; - if (gpio == val) - return 1; - } - return 0; -} -EXPORT_SYMBOL_GPL(geode_gpio_isset); - -void geode_gpio_set_irq(unsigned int group, unsigned int irq) -{ - u32 lo, hi; - - if (group > 7 || irq > 15) - return; - - rdmsr(MSR_PIC_ZSEL_HIGH, lo, hi); - - lo &= ~(0xF << (group * 4)); - lo |= (irq & 0xF) << (group * 4); - - wrmsr(MSR_PIC_ZSEL_HIGH, lo, hi); -} -EXPORT_SYMBOL_GPL(geode_gpio_set_irq); - -void geode_gpio_setup_event(unsigned int gpio, int pair, int pme) -{ - u32 base = geode_get_dev_base(GEODE_DEV_GPIO); - u32 offset, shift, val; - - if (gpio >= 24) - offset = GPIO_MAP_W; - else if (gpio >= 16) - offset = GPIO_MAP_Z; - else if (gpio >= 8) - offset = GPIO_MAP_Y; - else - offset = GPIO_MAP_X; - - shift = (gpio % 8) * 4; - - val = inl(base + offset); - - /* Clear whatever was there before */ - val &= ~(0xF << shift); - - /* And set the new value */ - - val |= ((pair & 7) << shift); - - /* Set the PME bit if this is a PME event */ - - if (pme) - val |= (1 << (shift + 3)); - - outl(val, base + offset); -} -EXPORT_SYMBOL_GPL(geode_gpio_setup_event); - -int geode_has_vsa2(void) -{ - static int has_vsa2 = -1; - - if (has_vsa2 == -1) { - u16 val; - - /* - * The VSA has virtual registers that we can query for a - * signature. - */ - outw(VSA_VR_UNLOCK, VSA_VRC_INDEX); - outw(VSA_VR_SIGNATURE, VSA_VRC_INDEX); - - val = inw(VSA_VRC_DATA); - has_vsa2 = (val == AMD_VSA_SIG || val == GSW_VSA_SIG); - } - - return has_vsa2; -} -EXPORT_SYMBOL_GPL(geode_has_vsa2); - -static int __init geode_southbridge_init(void) -{ - if (!is_geode()) - return -ENODEV; - - init_lbars(); - (void) mfgpt_timer_setup(); - return 0; -} - -postcore_initcall(geode_southbridge_init); diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index 4f8e2507e8f3..adedeef1dedc 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -29,7 +29,15 @@ static void __init i386_default_early_setup(void) void __init i386_start_kernel(void) { - reserve_trampoline_memory(); +#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"); +#endif reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 0b06cd778fd9..b5a9896ca1e7 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -98,8 +98,6 @@ void __init x86_64_start_reservations(char *real_mode_data) { copy_bootdata(__va(real_mode_data)); - reserve_trampoline_memory(); - reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index 050c278481b1..37c3d4b17d85 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -18,6 +18,8 @@ #include <asm/asm-offsets.h> #include <asm/setup.h> #include <asm/processor-flags.h> +#include <asm/msr-index.h> +#include <asm/cpufeature.h> #include <asm/percpu.h> /* Physical address */ @@ -297,25 +299,27 @@ ENTRY(startup_32_smp) orl %edx,%eax movl %eax,%cr4 - btl $5, %eax # check if PAE is enabled - jnc 6f + testb $X86_CR4_PAE, %al # check if PAE is enabled + jz 6f /* Check if extended functions are implemented */ movl $0x80000000, %eax cpuid - cmpl $0x80000000, %eax - jbe 6f + /* Value must be in the range 0x80000001 to 0x8000ffff */ + subl $0x80000001, %eax + cmpl $(0x8000ffff-0x80000001), %eax + ja 6f mov $0x80000001, %eax cpuid /* Execute Disable bit supported? */ - btl $20, %edx + btl $(X86_FEATURE_NX & 31), %edx jnc 6f /* Setup EFER (Extended Feature Enable Register) */ - movl $0xc0000080, %ecx + movl $MSR_EFER, %ecx rdmsr - btsl $11, %eax + btsl $_EFER_NX, %eax /* Make changes effective */ wrmsr @@ -438,8 +442,8 @@ is386: movl $2,%ecx # set MP */ cmpb $0,ready jne 1f - movl $per_cpu__gdt_page,%eax - movl $per_cpu__stack_canary,%ecx + movl $gdt_page,%eax + movl $stack_canary,%ecx movw %cx, 8 * GDT_ENTRY_STACK_CANARY + 2(%eax) shrl $16, %ecx movb %cl, 8 * GDT_ENTRY_STACK_CANARY + 4(%eax) @@ -702,7 +706,7 @@ idt_descr: .word 0 # 32 bit align gdt_desc.address ENTRY(early_gdt_descr) .word GDT_ENTRIES*8-1 - .long per_cpu__gdt_page /* Overwritten for secondary CPUs */ + .long gdt_page /* Overwritten for secondary CPUs */ /* * The boot_gdt must mirror the equivalent in setup.S and is diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index 780cd928fcd5..3d1e6f16b7a6 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -27,7 +27,7 @@ #define GET_CR2_INTO_RCX movq %cr2, %rcx #endif -/* we are not able to switch in one step to the final KERNEL ADRESS SPACE +/* we are not able to switch in one step to the final KERNEL ADDRESS SPACE * because we need identity-mapped pages. * */ @@ -212,8 +212,8 @@ ENTRY(secondary_startup_64) */ lgdt early_gdt_descr(%rip) - /* set up data segments. actually 0 would do too */ - movl $__KERNEL_DS,%eax + /* set up data segments */ + xorl %eax,%eax movl %eax,%ds movl %eax,%ss movl %eax,%es @@ -262,11 +262,11 @@ ENTRY(secondary_startup_64) .quad x86_64_start_kernel ENTRY(initial_gs) .quad INIT_PER_CPU_VAR(irq_stack_union) - __FINITDATA ENTRY(stack_start) .quad init_thread_union+THREAD_SIZE-8 .word 0 + __FINITDATA bad_address: jmp bad_address @@ -340,6 +340,7 @@ ENTRY(name) i = i + 1 ; \ .endr + .data /* * This default setting generates an ident mapping at address 0x100000 * and a mapping for the kernel that precisely maps virtual address diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index dedc2bddf7a5..ee4fa1bfcb33 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -33,6 +33,9 @@ * HPET address is set in acpi/boot.c, when an ACPI entry exists */ unsigned long hpet_address; +u8 hpet_blockid; /* OS timer block num */ +u8 hpet_msi_disable; + #ifdef CONFIG_PCI_MSI static unsigned long hpet_num_timers; #endif @@ -47,12 +50,12 @@ struct hpet_dev { char name[10]; }; -unsigned long hpet_readl(unsigned long a) +inline unsigned int hpet_readl(unsigned int a) { return readl(hpet_virt_address + a); } -static inline void hpet_writel(unsigned long d, unsigned long a) +static inline void hpet_writel(unsigned int d, unsigned int a) { writel(d, hpet_virt_address + a); } @@ -167,7 +170,7 @@ do { \ static void hpet_reserve_msi_timers(struct hpet_data *hd); -static void hpet_reserve_platform_timers(unsigned long id) +static void hpet_reserve_platform_timers(unsigned int id) { struct hpet __iomem *hpet = hpet_virt_address; struct hpet_timer __iomem *timer = &hpet->hpet_timers[2]; @@ -205,7 +208,7 @@ static void hpet_reserve_platform_timers(unsigned long id) } #else -static void hpet_reserve_platform_timers(unsigned long id) { } +static void hpet_reserve_platform_timers(unsigned int id) { } #endif /* @@ -246,7 +249,7 @@ static void hpet_reset_counter(void) static void hpet_start_counter(void) { - unsigned long cfg = hpet_readl(HPET_CFG); + unsigned int cfg = hpet_readl(HPET_CFG); cfg |= HPET_CFG_ENABLE; hpet_writel(cfg, HPET_CFG); } @@ -263,7 +266,7 @@ static void hpet_resume_device(void) force_hpet_resume(); } -static void hpet_resume_counter(void) +static void hpet_resume_counter(struct clocksource *cs) { hpet_resume_device(); hpet_restart_counter(); @@ -271,7 +274,7 @@ static void hpet_resume_counter(void) static void hpet_enable_legacy_int(void) { - unsigned long cfg = hpet_readl(HPET_CFG); + unsigned int cfg = hpet_readl(HPET_CFG); cfg |= HPET_CFG_LEGACY; hpet_writel(cfg, HPET_CFG); @@ -314,7 +317,7 @@ static int hpet_setup_msi_irq(unsigned int irq); static void hpet_set_mode(enum clock_event_mode mode, struct clock_event_device *evt, int timer) { - unsigned long cfg, cmp, now; + unsigned int cfg, cmp, now; uint64_t delta; switch (mode) { @@ -323,7 +326,7 @@ static void hpet_set_mode(enum clock_event_mode mode, delta = ((uint64_t)(NSEC_PER_SEC/HZ)) * evt->mult; delta >>= evt->shift; now = hpet_readl(HPET_COUNTER); - cmp = now + (unsigned long) delta; + cmp = now + (unsigned int) delta; cfg = hpet_readl(HPET_Tn_CFG(timer)); /* Make sure we use edge triggered interrupts */ cfg &= ~HPET_TN_LEVEL; @@ -339,7 +342,7 @@ static void hpet_set_mode(enum clock_event_mode mode, * (See AMD-8111 HyperTransport I/O Hub Data Sheet, * Publication # 24674) */ - hpet_writel((unsigned long) delta, HPET_Tn_CMP(timer)); + hpet_writel((unsigned int) delta, HPET_Tn_CMP(timer)); hpet_start_counter(); hpet_print_config(); break; @@ -383,13 +386,24 @@ static int hpet_next_event(unsigned long delta, hpet_writel(cnt, HPET_Tn_CMP(timer)); /* - * We need to read back the CMP register to make sure that - * what we wrote hit the chip before we compare it to the - * counter. + * 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. If the read back from the + * compare register does not match the value we programmed + * then we might have a real hardware problem. We can not do + * much about it here, but at least alert the user/admin with + * a prominent warning. */ - WARN_ON_ONCE((u32)hpet_readl(HPET_Tn_CMP(timer)) != cnt); + WARN_ONCE(hpet_readl(HPET_Tn_CMP(timer)) != cnt, + KERN_WARNING "hpet: compare register read back failed.\n"); - return (s32)((u32)hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0; + return (s32)(hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0; } static void hpet_legacy_set_mode(enum clock_event_mode mode, @@ -415,7 +429,7 @@ static struct hpet_dev *hpet_devs; void hpet_msi_unmask(unsigned int irq) { struct hpet_dev *hdev = get_irq_data(irq); - unsigned long cfg; + unsigned int cfg; /* unmask it */ cfg = hpet_readl(HPET_Tn_CFG(hdev->num)); @@ -425,7 +439,7 @@ void hpet_msi_unmask(unsigned int irq) void hpet_msi_mask(unsigned int irq) { - unsigned long cfg; + unsigned int cfg; struct hpet_dev *hdev = get_irq_data(irq); /* mask it */ @@ -467,7 +481,7 @@ static int hpet_msi_next_event(unsigned long delta, static int hpet_setup_msi_irq(unsigned int irq) { - if (arch_setup_hpet_msi(irq)) { + if (arch_setup_hpet_msi(irq, hpet_blockid)) { destroy_irq(irq); return -EINVAL; } @@ -584,6 +598,11 @@ static void hpet_msi_capability_lookup(unsigned int start_timer) unsigned int num_timers_used = 0; int i; + if (hpet_msi_disable) + return; + + if (boot_cpu_has(X86_FEATURE_ARAT)) + return; id = hpet_readl(HPET_ID); num_timers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT); @@ -598,7 +617,7 @@ static void hpet_msi_capability_lookup(unsigned int start_timer) for (i = start_timer; i < num_timers - RESERVE_TIMERS; i++) { struct hpet_dev *hdev = &hpet_devs[num_timers_used]; - unsigned long cfg = hpet_readl(HPET_Tn_CFG(i)); + unsigned int cfg = hpet_readl(HPET_Tn_CFG(i)); /* Only consider HPET timer with MSI support */ if (!(cfg & HPET_TN_FSB_CAP)) @@ -813,7 +832,7 @@ static int hpet_clocksource_register(void) */ int __init hpet_enable(void) { - unsigned long id; + unsigned int id; int i; if (!is_hpet_capable()) @@ -872,10 +891,8 @@ int __init hpet_enable(void) if (id & HPET_ID_LEGSUP) { hpet_legacy_clockevent_register(); - hpet_msi_capability_lookup(2); return 1; } - hpet_msi_capability_lookup(0); return 0; out_nohpet: @@ -908,9 +925,20 @@ static __init int hpet_late_init(void) if (!hpet_virt_address) return -ENODEV; + if (hpet_readl(HPET_ID) & HPET_ID_LEGSUP) + hpet_msi_capability_lookup(2); + else + hpet_msi_capability_lookup(0); + hpet_reserve_platform_timers(hpet_readl(HPET_ID)); hpet_print_config(); + if (hpet_msi_disable) + return 0; + + if (boot_cpu_has(X86_FEATURE_ARAT)) + return 0; + for_each_online_cpu(cpu) { hpet_cpuhp_notify(NULL, CPU_ONLINE, (void *)(long)cpu); } @@ -925,7 +953,7 @@ fs_initcall(hpet_late_init); void hpet_disable(void) { if (is_hpet_capable()) { - unsigned long cfg = hpet_readl(HPET_CFG); + unsigned int cfg = hpet_readl(HPET_CFG); if (hpet_legacy_int_enabled) { cfg &= ~HPET_CFG_LEGACY; @@ -965,8 +993,8 @@ static int hpet_prev_update_sec; static struct rtc_time hpet_alarm_time; static unsigned long hpet_pie_count; static u32 hpet_t1_cmp; -static unsigned long hpet_default_delta; -static unsigned long hpet_pie_delta; +static u32 hpet_default_delta; +static u32 hpet_pie_delta; static unsigned long hpet_pie_limit; static rtc_irq_handler irq_handler; @@ -1017,7 +1045,8 @@ EXPORT_SYMBOL_GPL(hpet_unregister_irq_handler); */ int hpet_rtc_timer_init(void) { - unsigned long cfg, cnt, delta, flags; + unsigned int cfg, cnt, delta; + unsigned long flags; if (!is_hpet_enabled()) return 0; @@ -1027,7 +1056,7 @@ int hpet_rtc_timer_init(void) clc = (uint64_t) hpet_clockevent.mult * NSEC_PER_SEC; clc >>= hpet_clockevent.shift + DEFAULT_RTC_SHIFT; - hpet_default_delta = (unsigned long) clc; + hpet_default_delta = clc; } if (!(hpet_rtc_flags & RTC_PIE) || hpet_pie_limit) @@ -1113,7 +1142,7 @@ int hpet_set_periodic_freq(unsigned long freq) clc = (uint64_t) hpet_clockevent.mult * NSEC_PER_SEC; do_div(clc, freq); clc >>= hpet_clockevent.shift; - hpet_pie_delta = (unsigned long) clc; + hpet_pie_delta = clc; } return 1; } @@ -1127,7 +1156,7 @@ EXPORT_SYMBOL_GPL(hpet_rtc_dropped_irq); static void hpet_rtc_timer_reinit(void) { - unsigned long cfg, delta; + unsigned int cfg, delta; int lost_ints = -1; if (unlikely(!hpet_rtc_flags)) { diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c new file mode 100644 index 000000000000..d6cc065f519f --- /dev/null +++ b/arch/x86/kernel/hw_breakpoint.c @@ -0,0 +1,530 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) 2007 Alan Stern + * Copyright (C) 2009 IBM Corporation + * Copyright (C) 2009 Frederic Weisbecker <fweisbec@gmail.com> + * + * Authors: Alan Stern <stern@rowland.harvard.edu> + * K.Prasad <prasad@linux.vnet.ibm.com> + * Frederic Weisbecker <fweisbec@gmail.com> + */ + +/* + * HW_breakpoint: a unified kernel/user-space hardware breakpoint facility, + * using the CPU's debug registers. + */ + +#include <linux/perf_event.h> +#include <linux/hw_breakpoint.h> +#include <linux/irqflags.h> +#include <linux/notifier.h> +#include <linux/kallsyms.h> +#include <linux/kprobes.h> +#include <linux/percpu.h> +#include <linux/kdebug.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/init.h> +#include <linux/smp.h> + +#include <asm/hw_breakpoint.h> +#include <asm/processor.h> +#include <asm/debugreg.h> + +/* Per cpu debug control register value */ +DEFINE_PER_CPU(unsigned long, cpu_dr7); +EXPORT_PER_CPU_SYMBOL(cpu_dr7); + +/* Per cpu debug address registers values */ +static DEFINE_PER_CPU(unsigned long, cpu_debugreg[HBP_NUM]); + +/* + * Stores the breakpoints currently in use on each breakpoint address + * register for each cpus + */ +static DEFINE_PER_CPU(struct perf_event *, bp_per_reg[HBP_NUM]); + + +static inline unsigned long +__encode_dr7(int drnum, unsigned int len, unsigned int type) +{ + unsigned long bp_info; + + bp_info = (len | type) & 0xf; + bp_info <<= (DR_CONTROL_SHIFT + drnum * DR_CONTROL_SIZE); + bp_info |= (DR_GLOBAL_ENABLE << (drnum * DR_ENABLE_SIZE)); + + return bp_info; +} + +/* + * Encode the length, type, Exact, and Enable bits for a particular breakpoint + * as stored in debug register 7. + */ +unsigned long encode_dr7(int drnum, unsigned int len, unsigned int type) +{ + return __encode_dr7(drnum, len, type) | DR_GLOBAL_SLOWDOWN; +} + +/* + * Decode the length and type bits for a particular breakpoint as + * stored in debug register 7. Return the "enabled" status. + */ +int decode_dr7(unsigned long dr7, int bpnum, unsigned *len, unsigned *type) +{ + int bp_info = dr7 >> (DR_CONTROL_SHIFT + bpnum * DR_CONTROL_SIZE); + + *len = (bp_info & 0xc) | 0x40; + *type = (bp_info & 0x3) | 0x80; + + return (dr7 >> (bpnum * DR_ENABLE_SIZE)) & 0x3; +} + +/* + * Install a perf counter breakpoint. + * + * We seek a free debug address register and use it for this + * breakpoint. Eventually we enable it in the debug control register. + * + * Atomic: we hold the counter->ctx->lock and we only handle variables + * and registers local to this cpu. + */ +int arch_install_hw_breakpoint(struct perf_event *bp) +{ + struct arch_hw_breakpoint *info = counter_arch_bp(bp); + unsigned long *dr7; + int i; + + for (i = 0; i < HBP_NUM; i++) { + struct perf_event **slot = &__get_cpu_var(bp_per_reg[i]); + + if (!*slot) { + *slot = bp; + break; + } + } + + if (WARN_ONCE(i == HBP_NUM, "Can't find any breakpoint slot")) + return -EBUSY; + + set_debugreg(info->address, i); + __get_cpu_var(cpu_debugreg[i]) = info->address; + + dr7 = &__get_cpu_var(cpu_dr7); + *dr7 |= encode_dr7(i, info->len, info->type); + + set_debugreg(*dr7, 7); + + return 0; +} + +/* + * Uninstall the breakpoint contained in the given counter. + * + * First we search the debug address register it uses and then we disable + * it. + * + * Atomic: we hold the counter->ctx->lock and we only handle variables + * and registers local to this cpu. + */ +void arch_uninstall_hw_breakpoint(struct perf_event *bp) +{ + struct arch_hw_breakpoint *info = counter_arch_bp(bp); + unsigned long *dr7; + int i; + + for (i = 0; i < HBP_NUM; i++) { + struct perf_event **slot = &__get_cpu_var(bp_per_reg[i]); + + if (*slot == bp) { + *slot = NULL; + break; + } + } + + if (WARN_ONCE(i == HBP_NUM, "Can't find any breakpoint slot")) + return; + + dr7 = &__get_cpu_var(cpu_dr7); + *dr7 &= ~__encode_dr7(i, info->len, info->type); + + set_debugreg(*dr7, 7); +} + +static int get_hbp_len(u8 hbp_len) +{ + unsigned int len_in_bytes = 0; + + switch (hbp_len) { + case X86_BREAKPOINT_LEN_1: + len_in_bytes = 1; + break; + case X86_BREAKPOINT_LEN_2: + len_in_bytes = 2; + break; + case X86_BREAKPOINT_LEN_4: + len_in_bytes = 4; + break; +#ifdef CONFIG_X86_64 + case X86_BREAKPOINT_LEN_8: + len_in_bytes = 8; + break; +#endif + } + return len_in_bytes; +} + +/* + * Check for virtual address in user space. + */ +int arch_check_va_in_userspace(unsigned long va, u8 hbp_len) +{ + unsigned int len; + + len = get_hbp_len(hbp_len); + + return (va <= TASK_SIZE - len); +} + +/* + * Check for virtual address in kernel space. + */ +static int arch_check_va_in_kernelspace(unsigned long va, u8 hbp_len) +{ + unsigned int len; + + len = get_hbp_len(hbp_len); + + return (va >= TASK_SIZE) && ((va + len - 1) >= TASK_SIZE); +} + +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_1: + *gen_len = HW_BREAKPOINT_LEN_1; + break; + case X86_BREAKPOINT_LEN_2: + *gen_len = HW_BREAKPOINT_LEN_2; + break; + case X86_BREAKPOINT_LEN_4: + *gen_len = HW_BREAKPOINT_LEN_4; + break; +#ifdef CONFIG_X86_64 + case X86_BREAKPOINT_LEN_8: + *gen_len = HW_BREAKPOINT_LEN_8; + break; +#endif + default: + 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; +} + + +static int arch_build_bp_info(struct perf_event *bp) +{ + struct arch_hw_breakpoint *info = counter_arch_bp(bp); + + info->address = bp->attr.bp_addr; + + /* Len */ + switch (bp->attr.bp_len) { + case HW_BREAKPOINT_LEN_1: + info->len = X86_BREAKPOINT_LEN_1; + break; + case HW_BREAKPOINT_LEN_2: + info->len = X86_BREAKPOINT_LEN_2; + break; + case HW_BREAKPOINT_LEN_4: + info->len = X86_BREAKPOINT_LEN_4; + break; +#ifdef CONFIG_X86_64 + case HW_BREAKPOINT_LEN_8: + info->len = X86_BREAKPOINT_LEN_8; + break; +#endif + default: + return -EINVAL; + } + + /* Type */ + switch (bp->attr.bp_type) { + case HW_BREAKPOINT_W: + info->type = X86_BREAKPOINT_WRITE; + break; + case HW_BREAKPOINT_W | HW_BREAKPOINT_R: + info->type = X86_BREAKPOINT_RW; + break; + case HW_BREAKPOINT_X: + info->type = X86_BREAKPOINT_EXECUTE; + break; + default: + return -EINVAL; + } + + return 0; +} +/* + * Validate the arch-specific HW Breakpoint register settings + */ +int arch_validate_hwbkpt_settings(struct perf_event *bp, + struct task_struct *tsk) +{ + struct arch_hw_breakpoint *info = counter_arch_bp(bp); + unsigned int align; + int ret; + + + ret = arch_build_bp_info(bp); + if (ret) + return ret; + + ret = -EINVAL; + + if (info->type == X86_BREAKPOINT_EXECUTE) + /* + * Ptrace-refactoring code + * For now, we'll allow instruction breakpoint only for user-space + * addresses + */ + if ((!arch_check_va_in_userspace(info->address, info->len)) && + info->len != X86_BREAKPOINT_EXECUTE) + return ret; + + switch (info->len) { + case X86_BREAKPOINT_LEN_1: + align = 0; + break; + case X86_BREAKPOINT_LEN_2: + align = 1; + break; + case X86_BREAKPOINT_LEN_4: + align = 3; + break; +#ifdef CONFIG_X86_64 + case X86_BREAKPOINT_LEN_8: + align = 7; + break; +#endif + default: + return ret; + } + + /* + * Check that the low-order bits of the address are appropriate + * for the alignment implied by len. + */ + if (info->address & align) + return -EINVAL; + + /* Check that the virtual address is in the proper range */ + if (tsk) { + if (!arch_check_va_in_userspace(info->address, info->len)) + return -EFAULT; + } else { + if (!arch_check_va_in_kernelspace(info->address, info->len)) + return -EFAULT; + } + + return 0; +} + +/* + * Dump the debug register contents to the user. + * We can't dump our per cpu values because it + * may contain cpu wide breakpoint, something that + * doesn't belong to the current task. + * + * TODO: include non-ptrace user breakpoints (perf) + */ +void aout_dump_debugregs(struct user *dump) +{ + int i; + int dr7 = 0; + struct perf_event *bp; + struct arch_hw_breakpoint *info; + struct thread_struct *thread = ¤t->thread; + + for (i = 0; i < HBP_NUM; i++) { + bp = thread->ptrace_bps[i]; + + if (bp && !bp->attr.disabled) { + dump->u_debugreg[i] = bp->attr.bp_addr; + info = counter_arch_bp(bp); + dr7 |= encode_dr7(i, info->len, info->type); + } else { + dump->u_debugreg[i] = 0; + } + } + + dump->u_debugreg[4] = 0; + dump->u_debugreg[5] = 0; + dump->u_debugreg[6] = current->thread.debugreg6; + + dump->u_debugreg[7] = dr7; +} +EXPORT_SYMBOL_GPL(aout_dump_debugregs); + +/* + * Release the user breakpoints used by ptrace + */ +void flush_ptrace_hw_breakpoint(struct task_struct *tsk) +{ + int i; + struct thread_struct *t = &tsk->thread; + + for (i = 0; i < HBP_NUM; i++) { + unregister_hw_breakpoint(t->ptrace_bps[i]); + t->ptrace_bps[i] = NULL; + } +} + +void hw_breakpoint_restore(void) +{ + set_debugreg(__get_cpu_var(cpu_debugreg[0]), 0); + set_debugreg(__get_cpu_var(cpu_debugreg[1]), 1); + set_debugreg(__get_cpu_var(cpu_debugreg[2]), 2); + set_debugreg(__get_cpu_var(cpu_debugreg[3]), 3); + set_debugreg(current->thread.debugreg6, 6); + set_debugreg(__get_cpu_var(cpu_dr7), 7); +} +EXPORT_SYMBOL_GPL(hw_breakpoint_restore); + +/* + * Handle debug exception notifications. + * + * Return value is either NOTIFY_STOP or NOTIFY_DONE as explained below. + * + * NOTIFY_DONE returned if one of the following conditions is true. + * i) When the causative address is from user-space and the exception + * is a valid one, i.e. not triggered as a result of lazy debug register + * switching + * ii) When there are more bits than trap<n> set in DR6 register (such + * as BD, BS or BT) indicating that more than one debug condition is + * met and requires some more action in do_debug(). + * + * NOTIFY_STOP returned for all other cases + * + */ +static int __kprobes hw_breakpoint_handler(struct die_args *args) +{ + int i, cpu, rc = NOTIFY_STOP; + struct perf_event *bp; + unsigned long dr7, dr6; + unsigned long *dr6_p; + + /* The DR6 value is pointed by args->err */ + dr6_p = (unsigned long *)ERR_PTR(args->err); + dr6 = *dr6_p; + + /* Do an early return if no trap bits are set in DR6 */ + if ((dr6 & DR_TRAP_BITS) == 0) + return NOTIFY_DONE; + + get_debugreg(dr7, 7); + /* Disable breakpoints during exception handling */ + set_debugreg(0UL, 7); + /* + * Assert that local interrupts are disabled + * Reset the DRn bits in the virtualized register value. + * The ptrace trigger routine will add in whatever is needed. + */ + current->thread.debugreg6 &= ~DR_TRAP_BITS; + cpu = get_cpu(); + + /* Handle all the breakpoints that were triggered */ + for (i = 0; i < HBP_NUM; ++i) { + if (likely(!(dr6 & (DR_TRAP0 << i)))) + continue; + + /* + * The counter may be concurrently released but that can only + * occur from a call_rcu() path. We can then safely fetch + * the breakpoint, use its callback, touch its counter + * while we are in an rcu_read_lock() path. + */ + rcu_read_lock(); + + bp = per_cpu(bp_per_reg[i], cpu); + /* + * Reset the 'i'th TRAP bit in dr6 to denote completion of + * exception handling + */ + (*dr6_p) &= ~(DR_TRAP0 << i); + /* + * bp can be NULL due to lazy debug register switching + * or due to concurrent perf counter removing. + */ + if (!bp) { + rcu_read_unlock(); + break; + } + + perf_bp_event(bp, args->regs); + + rcu_read_unlock(); + } + /* + * Further processing in do_debug() is needed for a) user-space + * breakpoints (to generate signals) and b) when the system has + * taken exception due to multiple causes + */ + if ((current->thread.debugreg6 & DR_TRAP_BITS) || + (dr6 & (~DR_TRAP_BITS))) + rc = NOTIFY_DONE; + + set_debugreg(dr7, 7); + put_cpu(); + + return rc; +} + +/* + * Handle debug exception notifications. + */ +int __kprobes hw_breakpoint_exceptions_notify( + struct notifier_block *unused, unsigned long val, void *data) +{ + if (val != DIE_DEBUG) + return NOTIFY_DONE; + + return hw_breakpoint_handler(data); +} + +void hw_breakpoint_pmu_read(struct perf_event *bp) +{ + /* TODO */ +} diff --git a/arch/x86/kernel/i386_ksyms_32.c b/arch/x86/kernel/i386_ksyms_32.c index 43cec6bdda63..9c3bd4a2050e 100644 --- a/arch/x86/kernel/i386_ksyms_32.c +++ b/arch/x86/kernel/i386_ksyms_32.c @@ -10,6 +10,16 @@ EXPORT_SYMBOL(mcount); #endif +/* + * Note, this is a prototype to get at the symbol for + * the export, but dont use it from C code, it is used + * by assembly code and is not using C calling convention! + */ +#ifndef CONFIG_X86_CMPXCHG64 +extern void cmpxchg8b_emu(void); +EXPORT_SYMBOL(cmpxchg8b_emu); +#endif + /* Networking helper routines. */ EXPORT_SYMBOL(csum_partial_copy_generic); diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index f2f8540a7f3d..c01a2b846d47 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -164,6 +164,11 @@ int init_fpu(struct task_struct *tsk) return 0; } +/* + * The xstateregs_active() routine is the same as the fpregs_active() routine, + * as the "regset->n" for the xstate regset will be updated based on the feature + * capabilites supported by the xsave. + */ int fpregs_active(struct task_struct *target, const struct user_regset *regset) { return tsk_used_math(target) ? regset->n : 0; @@ -204,8 +209,6 @@ int xfpregs_set(struct task_struct *target, const struct user_regset *regset, if (ret) return ret; - set_stopped_child_used_math(target); - ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &target->thread.xstate->fxsave, 0, -1); @@ -224,6 +227,68 @@ int xfpregs_set(struct task_struct *target, const struct user_regset *regset, return ret; } +int xstateregs_get(struct task_struct *target, const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + int ret; + + if (!cpu_has_xsave) + return -ENODEV; + + ret = init_fpu(target); + if (ret) + return ret; + + /* + * Copy the 48bytes defined by the software first into the xstate + * memory layout in the thread struct, so that we can copy the entire + * xstateregs to the user using one user_regset_copyout(). + */ + memcpy(&target->thread.xstate->fxsave.sw_reserved, + xstate_fx_sw_bytes, sizeof(xstate_fx_sw_bytes)); + + /* + * Copy the xstate memory layout. + */ + ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf, + &target->thread.xstate->xsave, 0, -1); + return ret; +} + +int xstateregs_set(struct task_struct *target, const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + int ret; + struct xsave_hdr_struct *xsave_hdr; + + if (!cpu_has_xsave) + return -ENODEV; + + ret = init_fpu(target); + if (ret) + return ret; + + ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, + &target->thread.xstate->xsave, 0, -1); + + /* + * mxcsr reserved bits must be masked to zero for security reasons. + */ + target->thread.xstate->fxsave.mxcsr &= mxcsr_feature_mask; + + xsave_hdr = &target->thread.xstate->xsave.xsave_hdr; + + xsave_hdr->xstate_bv &= pcntxt_mask; + /* + * These bits must be zero. + */ + xsave_hdr->reserved1[0] = xsave_hdr->reserved1[1] = 0; + + return ret; +} + #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION /* @@ -404,8 +469,6 @@ int fpregs_set(struct task_struct *target, const struct user_regset *regset, if (ret) return ret; - set_stopped_child_used_math(target); - if (!HAVE_HWFP) return fpregs_soft_set(target, regset, pos, count, kbuf, ubuf); diff --git a/arch/x86/kernel/i8259.c b/arch/x86/kernel/i8259.c index df89102bef80..fb725ee15f55 100644 --- a/arch/x86/kernel/i8259.c +++ b/arch/x86/kernel/i8259.c @@ -32,8 +32,14 @@ */ static int i8259A_auto_eoi; -DEFINE_SPINLOCK(i8259A_lock); +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", @@ -63,51 +69,51 @@ unsigned int cached_irq_mask = 0xffff; */ unsigned long io_apic_irqs; -void disable_8259A_irq(unsigned int irq) +static void disable_8259A_irq(unsigned int irq) { unsigned int mask = 1 << irq; unsigned long flags; - spin_lock_irqsave(&i8259A_lock, flags); + raw_spin_lock_irqsave(&i8259A_lock, flags); cached_irq_mask |= mask; if (irq & 8) outb(cached_slave_mask, PIC_SLAVE_IMR); else outb(cached_master_mask, PIC_MASTER_IMR); - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); } -void enable_8259A_irq(unsigned int irq) +static void enable_8259A_irq(unsigned int irq) { unsigned int mask = ~(1 << irq); unsigned long flags; - spin_lock_irqsave(&i8259A_lock, flags); + raw_spin_lock_irqsave(&i8259A_lock, flags); cached_irq_mask &= mask; if (irq & 8) outb(cached_slave_mask, PIC_SLAVE_IMR); else outb(cached_master_mask, PIC_MASTER_IMR); - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); } -int i8259A_irq_pending(unsigned int irq) +static int i8259A_irq_pending(unsigned int irq) { unsigned int mask = 1<<irq; unsigned long flags; int ret; - spin_lock_irqsave(&i8259A_lock, flags); + raw_spin_lock_irqsave(&i8259A_lock, flags); if (irq < 8) ret = inb(PIC_MASTER_CMD) & mask; else ret = inb(PIC_SLAVE_CMD) & (mask >> 8); - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); return ret; } -void make_8259A_irq(unsigned int irq) +static void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); io_apic_irqs &= ~(1<<irq); @@ -150,7 +156,7 @@ static void mask_and_ack_8259A(unsigned int irq) unsigned int irqmask = 1 << irq; unsigned long flags; - spin_lock_irqsave(&i8259A_lock, flags); + raw_spin_lock_irqsave(&i8259A_lock, flags); /* * Lightweight spurious IRQ detection. We do not want * to overdo spurious IRQ handling - it's usually a sign @@ -183,7 +189,7 @@ handle_real_irq: outb(cached_master_mask, PIC_MASTER_IMR); outb(0x60+irq, PIC_MASTER_CMD); /* 'Specific EOI to master */ } - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); return; spurious_8259A_irq: @@ -281,37 +287,37 @@ static int __init i8259A_init_sysfs(void) device_initcall(i8259A_init_sysfs); -void mask_8259A(void) +static void mask_8259A(void) { unsigned long flags; - spin_lock_irqsave(&i8259A_lock, flags); + raw_spin_lock_irqsave(&i8259A_lock, flags); outb(0xff, PIC_MASTER_IMR); /* mask all of 8259A-1 */ outb(0xff, PIC_SLAVE_IMR); /* mask all of 8259A-2 */ - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); } -void unmask_8259A(void) +static void unmask_8259A(void) { unsigned long flags; - spin_lock_irqsave(&i8259A_lock, flags); + raw_spin_lock_irqsave(&i8259A_lock, flags); outb(cached_master_mask, PIC_MASTER_IMR); /* restore master IRQ mask */ outb(cached_slave_mask, PIC_SLAVE_IMR); /* restore slave IRQ mask */ - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); } -void init_8259A(int auto_eoi) +static void init_8259A(int auto_eoi) { unsigned long flags; i8259A_auto_eoi = auto_eoi; - spin_lock_irqsave(&i8259A_lock, flags); + raw_spin_lock_irqsave(&i8259A_lock, flags); outb(0xff, PIC_MASTER_IMR); /* mask all of 8259A-1 */ outb(0xff, PIC_SLAVE_IMR); /* mask all of 8259A-2 */ @@ -356,5 +362,49 @@ void init_8259A(int auto_eoi) outb(cached_master_mask, PIC_MASTER_IMR); /* restore master IRQ mask */ outb(cached_slave_mask, PIC_SLAVE_IMR); /* restore slave IRQ mask */ - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); } + +/* + * make i8259 a driver so that we can select pic functions at run time. the goal + * is to make x86 binary compatible among pc compatible and non-pc compatible + * platforms, such as x86 MID. + */ + +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; +} + +struct legacy_pic null_legacy_pic = { + .nr_legacy_irqs = 0, + .chip = &dummy_pic_chip, + .mask_all = legacy_pic_noop, + .restore_mask = legacy_pic_noop, + .init = legacy_pic_int_noop, + .irq_pending = legacy_pic_irq_pending_noop, + .make_irq = legacy_pic_uint_noop, +}; + +struct legacy_pic default_legacy_pic = { + .nr_legacy_irqs = NR_IRQS_LEGACY, + .chip = &i8259A_chip, + .mask_all = mask_8259A, + .restore_mask = unmask_8259A, + .init = init_8259A, + .irq_pending = i8259A_irq_pending, + .make_irq = make_8259A_irq, +}; + +struct legacy_pic *legacy_pic = &default_legacy_pic; diff --git a/arch/x86/kernel/ioport.c b/arch/x86/kernel/ioport.c index 99c4d308f16b..8eec0ec59af2 100644 --- a/arch/x86/kernel/ioport.c +++ b/arch/x86/kernel/ioport.c @@ -103,9 +103,10 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on) * on system-call entry - see also fork() and the signal handling * code. */ -static int do_iopl(unsigned int level, struct pt_regs *regs) +long sys_iopl(unsigned int level, struct pt_regs *regs) { unsigned int old = (regs->flags >> 12) & 3; + struct thread_struct *t = ¤t->thread; if (level > 3) return -EINVAL; @@ -115,29 +116,8 @@ static int do_iopl(unsigned int level, struct pt_regs *regs) return -EPERM; } regs->flags = (regs->flags & ~X86_EFLAGS_IOPL) | (level << 12); - - return 0; -} - -#ifdef CONFIG_X86_32 -long sys_iopl(struct pt_regs *regs) -{ - unsigned int level = regs->bx; - struct thread_struct *t = ¤t->thread; - int rc; - - rc = do_iopl(level, regs); - if (rc < 0) - goto out; - t->iopl = level << 12; set_iopl_mask(t->iopl); -out: - return rc; -} -#else -asmlinkage long sys_iopl(unsigned int level, struct pt_regs *regs) -{ - return do_iopl(level, regs); + + return 0; } -#endif diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 74656d1d4e30..91fd0c70a18a 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -18,7 +18,7 @@ atomic_t irq_err_count; /* Function pointer for generic interrupt vector handling */ -void (*generic_interrupt_extension)(void) = NULL; +void (*x86_platform_ipi_callback)(void) = NULL; /* * 'what should we do if we get a hw irq event on an illegal vector'. @@ -63,19 +63,19 @@ static int show_other_interrupts(struct seq_file *p, int prec) for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->irq_spurious_count); seq_printf(p, " Spurious interrupts\n"); - seq_printf(p, "%*s: ", prec, "CNT"); + seq_printf(p, "%*s: ", prec, "PMI"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->apic_perf_irqs); - seq_printf(p, " Performance counter interrupts\n"); + seq_printf(p, " Performance monitoring interrupts\n"); seq_printf(p, "%*s: ", prec, "PND"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->apic_pending_irqs); seq_printf(p, " Performance pending work\n"); #endif - if (generic_interrupt_extension) { + if (x86_platform_ipi_callback) { seq_printf(p, "%*s: ", prec, "PLT"); for_each_online_cpu(j) - seq_printf(p, "%10u ", irq_stats(j)->generic_irqs); + seq_printf(p, "%10u ", irq_stats(j)->x86_platform_ipis); seq_printf(p, " Platform interrupts\n"); } #ifdef CONFIG_SMP @@ -92,17 +92,17 @@ static int show_other_interrupts(struct seq_file *p, int prec) seq_printf(p, "%10u ", irq_stats(j)->irq_tlb_count); seq_printf(p, " TLB shootdowns\n"); #endif -#ifdef CONFIG_X86_MCE +#ifdef CONFIG_X86_THERMAL_VECTOR seq_printf(p, "%*s: ", prec, "TRM"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->irq_thermal_count); seq_printf(p, " Thermal event interrupts\n"); -# ifdef CONFIG_X86_MCE_THRESHOLD +#endif +#ifdef CONFIG_X86_MCE_THRESHOLD seq_printf(p, "%*s: ", prec, "THR"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->irq_threshold_count); seq_printf(p, " Threshold APIC interrupts\n"); -# endif #endif #ifdef CONFIG_X86_MCE seq_printf(p, "%*s: ", prec, "MCE"); @@ -149,7 +149,7 @@ int show_interrupts(struct seq_file *p, void *v) if (!desc) return 0; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); for_each_online_cpu(j) any_count |= kstat_irqs_cpu(i, j); action = desc->action; @@ -170,7 +170,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); out: - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } @@ -187,18 +187,18 @@ u64 arch_irq_stat_cpu(unsigned int cpu) sum += irq_stats(cpu)->apic_perf_irqs; sum += irq_stats(cpu)->apic_pending_irqs; #endif - if (generic_interrupt_extension) - sum += irq_stats(cpu)->generic_irqs; + if (x86_platform_ipi_callback) + sum += irq_stats(cpu)->x86_platform_ipis; #ifdef CONFIG_SMP sum += irq_stats(cpu)->irq_resched_count; sum += irq_stats(cpu)->irq_call_count; sum += irq_stats(cpu)->irq_tlb_count; #endif -#ifdef CONFIG_X86_MCE +#ifdef CONFIG_X86_THERMAL_VECTOR sum += irq_stats(cpu)->irq_thermal_count; -# ifdef CONFIG_X86_MCE_THRESHOLD +#endif +#ifdef CONFIG_X86_MCE_THRESHOLD sum += irq_stats(cpu)->irq_threshold_count; -# endif #endif #ifdef CONFIG_X86_MCE sum += per_cpu(mce_exception_count, cpu); @@ -251,9 +251,9 @@ unsigned int __irq_entry do_IRQ(struct pt_regs *regs) } /* - * Handler for GENERIC_INTERRUPT_VECTOR. + * Handler for X86_PLATFORM_IPI_VECTOR. */ -void smp_generic_interrupt(struct pt_regs *regs) +void smp_x86_platform_ipi(struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); @@ -263,10 +263,10 @@ void smp_generic_interrupt(struct pt_regs *regs) irq_enter(); - inc_irq_stat(generic_irqs); + inc_irq_stat(x86_platform_ipis); - if (generic_interrupt_extension) - generic_interrupt_extension(); + if (x86_platform_ipi_callback) + x86_platform_ipi_callback(); irq_exit(); @@ -274,3 +274,93 @@ void smp_generic_interrupt(struct pt_regs *regs) } EXPORT_SYMBOL_GPL(vector_used_by_percpu_irq); + +#ifdef CONFIG_HOTPLUG_CPU +/* A cpu has been removed from cpu_online_mask. Reset irq affinities. */ +void fixup_irqs(void) +{ + unsigned int irq, vector; + static int warned; + struct irq_desc *desc; + + for_each_irq_desc(irq, desc) { + int break_affinity = 0; + int set_affinity = 1; + const struct cpumask *affinity; + + if (!desc) + continue; + if (irq == 2) + continue; + + /* interrupt's are disabled at this point */ + raw_spin_lock(&desc->lock); + + affinity = desc->affinity; + if (!irq_has_action(irq) || + cpumask_equal(affinity, cpu_online_mask)) { + raw_spin_unlock(&desc->lock); + continue; + } + + /* + * Complete the irq move. This cpu is going down and for + * non intr-remapping case, we can't wait till this interrupt + * arrives at this cpu before completing the irq move. + */ + irq_force_complete_move(irq); + + if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) { + break_affinity = 1; + affinity = cpu_all_mask; + } + + if (!(desc->status & IRQ_MOVE_PCNTXT) && desc->chip->mask) + desc->chip->mask(irq); + + if (desc->chip->set_affinity) + desc->chip->set_affinity(irq, affinity); + else if (!(warned++)) + set_affinity = 0; + + if (!(desc->status & IRQ_MOVE_PCNTXT) && desc->chip->unmask) + desc->chip->unmask(irq); + + raw_spin_unlock(&desc->lock); + + if (break_affinity && set_affinity) + printk("Broke affinity for irq %i\n", irq); + else if (!set_affinity) + printk("Cannot set affinity for irq %i\n", irq); + } + + /* + * We can remove mdelay() and then send spuriuous interrupts to + * new cpu targets for all the irqs that were handled previously by + * this cpu. While it works, I have seen spurious interrupt messages + * (nothing wrong but still...). + * + * So for now, retain mdelay(1) and check the IRR and then send those + * interrupts to new targets as this cpu is already offlined... + */ + mdelay(1); + + for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) { + unsigned int irr; + + if (__get_cpu_var(vector_irq)[vector] < 0) + continue; + + irr = apic_read(APIC_IRR + (vector / 32 * 0x10)); + if (irr & (1 << (vector % 32))) { + irq = __get_cpu_var(vector_irq)[vector]; + + desc = irq_to_desc(irq); + raw_spin_lock(&desc->lock); + if (desc->chip->retrigger) + desc->chip->retrigger(irq); + raw_spin_unlock(&desc->lock); + } + } +} +#endif diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index 7d35d0fe2329..10709f29d166 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -211,48 +211,3 @@ bool handle_irq(unsigned irq, struct pt_regs *regs) return true; } - -#ifdef CONFIG_HOTPLUG_CPU - -/* A cpu has been removed from cpu_online_mask. Reset irq affinities. */ -void fixup_irqs(void) -{ - unsigned int irq; - struct irq_desc *desc; - - for_each_irq_desc(irq, desc) { - const struct cpumask *affinity; - - if (!desc) - continue; - if (irq == 2) - continue; - - affinity = desc->affinity; - if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) { - printk("Breaking affinity for irq %i\n", irq); - affinity = cpu_all_mask; - } - if (desc->chip->set_affinity) - desc->chip->set_affinity(irq, affinity); - else if (desc->action) - printk_once("Cannot set affinity for irq %i\n", irq); - } - -#if 0 - barrier(); - /* Ingo Molnar says: "after the IO-APIC masks have been redirected - [note the nop - the interrupt-enable boundary on x86 is two - instructions from sti] - to flush out pending hardirqs and - IPIs. After this point nothing is supposed to reach this CPU." */ - __asm__ __volatile__("sti; nop; cli"); - barrier(); -#else - /* That doesn't seem sufficient. Give it 1ms. */ - local_irq_enable(); - mdelay(1); - local_irq_disable(); -#endif -} -#endif - diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c index 977d8b43a0dd..acf8fbf8fbda 100644 --- a/arch/x86/kernel/irq_64.c +++ b/arch/x86/kernel/irq_64.c @@ -62,64 +62,6 @@ bool handle_irq(unsigned irq, struct pt_regs *regs) return true; } -#ifdef CONFIG_HOTPLUG_CPU -/* A cpu has been removed from cpu_online_mask. Reset irq affinities. */ -void fixup_irqs(void) -{ - unsigned int irq; - static int warned; - struct irq_desc *desc; - - for_each_irq_desc(irq, desc) { - int break_affinity = 0; - int set_affinity = 1; - const struct cpumask *affinity; - - if (!desc) - continue; - if (irq == 2) - continue; - - /* interrupt's are disabled at this point */ - spin_lock(&desc->lock); - - affinity = desc->affinity; - if (!irq_has_action(irq) || - cpumask_equal(affinity, cpu_online_mask)) { - spin_unlock(&desc->lock); - continue; - } - - if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) { - break_affinity = 1; - affinity = cpu_all_mask; - } - - if (desc->chip->mask) - desc->chip->mask(irq); - - if (desc->chip->set_affinity) - desc->chip->set_affinity(irq, affinity); - else if (!(warned++)) - set_affinity = 0; - - if (desc->chip->unmask) - desc->chip->unmask(irq); - - spin_unlock(&desc->lock); - - if (break_affinity && set_affinity) - printk("Broke affinity for irq %i\n", irq); - else if (!set_affinity) - printk("Cannot set affinity for irq %i\n", irq); - } - - /* That doesn't seem sufficient. Give it 1ms. */ - local_irq_enable(); - mdelay(1); - local_irq_disable(); -} -#endif extern void call_softirq(void); diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c index 40f30773fb29..ef257fc2921b 100644 --- a/arch/x86/kernel/irqinit.c +++ b/arch/x86/kernel/irqinit.c @@ -84,24 +84,7 @@ static struct irqaction irq2 = { }; DEFINE_PER_CPU(vector_irq_t, vector_irq) = { - [0 ... IRQ0_VECTOR - 1] = -1, - [IRQ0_VECTOR] = 0, - [IRQ1_VECTOR] = 1, - [IRQ2_VECTOR] = 2, - [IRQ3_VECTOR] = 3, - [IRQ4_VECTOR] = 4, - [IRQ5_VECTOR] = 5, - [IRQ6_VECTOR] = 6, - [IRQ7_VECTOR] = 7, - [IRQ8_VECTOR] = 8, - [IRQ9_VECTOR] = 9, - [IRQ10_VECTOR] = 10, - [IRQ11_VECTOR] = 11, - [IRQ12_VECTOR] = 12, - [IRQ13_VECTOR] = 13, - [IRQ14_VECTOR] = 14, - [IRQ15_VECTOR] = 15, - [IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1 + [0 ... NR_VECTORS - 1] = -1, }; int vector_used_by_percpu_irq(unsigned int vector) @@ -123,12 +106,12 @@ void __init init_ISA_irqs(void) #if defined(CONFIG_X86_64) || defined(CONFIG_X86_LOCAL_APIC) init_bsp_APIC(); #endif - init_8259A(0); + legacy_pic->init(0); /* * 16 old-style INTA-cycle interrupts: */ - for (i = 0; i < NR_IRQS_LEGACY; i++) { + for (i = 0; i < legacy_pic->nr_legacy_irqs; i++) { struct irq_desc *desc = irq_to_desc(i); desc->status = IRQ_DISABLED; @@ -142,6 +125,19 @@ void __init init_ISA_irqs(void) void __init init_IRQ(void) { + int i; + + /* + * On cpu 0, Assign IRQ0_VECTOR..IRQ15_VECTOR's to IRQ 0..15. + * If these IRQ's are handled by legacy interrupt-controllers like PIC, + * then this configuration will likely be static after the boot. If + * these IRQ's are handled by more mordern controllers like IO-APIC, + * then this vector space can be freed and re-used dynamically as the + * irq's migrate etc. + */ + for (i = 0; i < legacy_pic->nr_legacy_irqs; i++) + per_cpu(vector_irq, 0)[IRQ0_VECTOR + i] = i; + x86_init.irqs.intr_init(); } @@ -200,8 +196,8 @@ static void __init apic_intr_init(void) /* self generated IPI for local APIC timer */ alloc_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt); - /* generic IPI for platform specific use */ - alloc_intr_gate(GENERIC_INTERRUPT_VECTOR, generic_interrupt); + /* IPI for X86 platform specific use */ + alloc_intr_gate(X86_PLATFORM_IPI_VECTOR, x86_platform_ipi); /* IPI vectors for APIC spurious and error interrupts */ alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt); diff --git a/arch/x86/kernel/k8.c b/arch/x86/kernel/k8.c index cbc4332a77b2..9b895464dd03 100644 --- a/arch/x86/kernel/k8.c +++ b/arch/x86/kernel/k8.c @@ -121,3 +121,17 @@ void k8_flush_garts(void) } EXPORT_SYMBOL_GPL(k8_flush_garts); +static __init int init_k8_nbs(void) +{ + int err = 0; + + err = cache_k8_northbridges(); + + if (err < 0) + printk(KERN_NOTICE "K8 NB: Cannot enumerate AMD northbridges.\n"); + + return err; +} + +/* This has to go after the PCI subsystem */ +fs_initcall(init_k8_nbs); diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index 8d82a77a3f3b..bfba6019d762 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -42,7 +42,9 @@ #include <linux/init.h> #include <linux/smp.h> #include <linux/nmi.h> +#include <linux/hw_breakpoint.h> +#include <asm/debugreg.h> #include <asm/apicdef.h> #include <asm/system.h> @@ -85,10 +87,15 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) gdb_regs[GDB_DS] = regs->ds; gdb_regs[GDB_ES] = regs->es; gdb_regs[GDB_CS] = regs->cs; - gdb_regs[GDB_SS] = __KERNEL_DS; gdb_regs[GDB_FS] = 0xFFFF; gdb_regs[GDB_GS] = 0xFFFF; - gdb_regs[GDB_SP] = (int)®s->sp; + if (user_mode_vm(regs)) { + gdb_regs[GDB_SS] = regs->ss; + gdb_regs[GDB_SP] = regs->sp; + } else { + gdb_regs[GDB_SS] = __KERNEL_DS; + gdb_regs[GDB_SP] = kernel_stack_pointer(regs); + } #else gdb_regs[GDB_R8] = regs->r8; gdb_regs[GDB_R9] = regs->r9; @@ -101,7 +108,7 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) gdb_regs32[GDB_PS] = regs->flags; gdb_regs32[GDB_CS] = regs->cs; gdb_regs32[GDB_SS] = regs->ss; - gdb_regs[GDB_SP] = regs->sp; + gdb_regs[GDB_SP] = kernel_stack_pointer(regs); #endif } @@ -198,41 +205,81 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) static struct hw_breakpoint { unsigned enabled; - unsigned type; - unsigned len; unsigned long addr; + int len; + int type; + struct perf_event **pev; } breakinfo[4]; static void kgdb_correct_hw_break(void) { - unsigned long dr7; - int correctit = 0; - int breakbit; int breakno; - get_debugreg(dr7, 7); for (breakno = 0; breakno < 4; breakno++) { - breakbit = 2 << (breakno << 1); - if (!(dr7 & breakbit) && breakinfo[breakno].enabled) { - correctit = 1; - dr7 |= breakbit; - dr7 &= ~(0xf0000 << (breakno << 2)); - dr7 |= ((breakinfo[breakno].len << 2) | - breakinfo[breakno].type) << - ((breakno << 2) + 16); - if (breakno >= 0 && breakno <= 3) - set_debugreg(breakinfo[breakno].addr, breakno); - - } else { - if ((dr7 & breakbit) && !breakinfo[breakno].enabled) { - correctit = 1; - dr7 &= ~breakbit; - dr7 &= ~(0xf0000 << (breakno << 2)); - } - } + struct perf_event *bp; + struct arch_hw_breakpoint *info; + int val; + int cpu = raw_smp_processor_id(); + if (!breakinfo[breakno].enabled) + continue; + bp = *per_cpu_ptr(breakinfo[breakno].pev, cpu); + info = counter_arch_bp(bp); + if (bp->attr.disabled != 1) + continue; + bp->attr.bp_addr = breakinfo[breakno].addr; + bp->attr.bp_len = breakinfo[breakno].len; + bp->attr.bp_type = breakinfo[breakno].type; + info->address = breakinfo[breakno].addr; + info->len = breakinfo[breakno].len; + info->type = breakinfo[breakno].type; + val = arch_install_hw_breakpoint(bp); + if (!val) + bp->attr.disabled = 0; } - if (correctit) - set_debugreg(dr7, 7); + hw_breakpoint_restore(); +} + +static int hw_break_reserve_slot(int breakno) +{ + int cpu; + int cnt = 0; + struct perf_event **pevent; + + for_each_online_cpu(cpu) { + cnt++; + pevent = per_cpu_ptr(breakinfo[breakno].pev, cpu); + if (dbg_reserve_bp_slot(*pevent)) + goto fail; + } + + return 0; + +fail: + for_each_online_cpu(cpu) { + cnt--; + if (!cnt) + break; + pevent = per_cpu_ptr(breakinfo[breakno].pev, cpu); + dbg_release_bp_slot(*pevent); + } + return -1; +} + +static int hw_break_release_slot(int breakno) +{ + struct perf_event **pevent; + int cpu; + + for_each_online_cpu(cpu) { + pevent = per_cpu_ptr(breakinfo[breakno].pev, cpu); + if (dbg_release_bp_slot(*pevent)) + /* + * The debugger is responisble for handing the retry on + * remove failure. + */ + return -1; + } + return 0; } static int @@ -246,6 +293,10 @@ kgdb_remove_hw_break(unsigned long addr, int len, enum kgdb_bptype bptype) if (i == 4) return -1; + if (hw_break_release_slot(i)) { + printk(KERN_ERR "Cannot remove hw breakpoint at %lx\n", addr); + return -1; + } breakinfo[i].enabled = 0; return 0; @@ -254,15 +305,23 @@ kgdb_remove_hw_break(unsigned long addr, int len, enum kgdb_bptype bptype) static void kgdb_remove_all_hw_break(void) { int i; + int cpu = raw_smp_processor_id(); + struct perf_event *bp; - for (i = 0; i < 4; i++) - memset(&breakinfo[i], 0, sizeof(struct hw_breakpoint)); + for (i = 0; i < 4; i++) { + if (!breakinfo[i].enabled) + continue; + bp = *per_cpu_ptr(breakinfo[i].pev, cpu); + if (bp->attr.disabled == 1) + continue; + arch_uninstall_hw_breakpoint(bp); + bp->attr.disabled = 1; + } } static int kgdb_set_hw_break(unsigned long addr, int len, enum kgdb_bptype bptype) { - unsigned type; int i; for (i = 0; i < 4; i++) @@ -273,27 +332,42 @@ kgdb_set_hw_break(unsigned long addr, int len, enum kgdb_bptype bptype) switch (bptype) { case BP_HARDWARE_BREAKPOINT: - type = 0; - len = 1; + len = 1; + breakinfo[i].type = X86_BREAKPOINT_EXECUTE; break; case BP_WRITE_WATCHPOINT: - type = 1; + breakinfo[i].type = X86_BREAKPOINT_WRITE; break; case BP_ACCESS_WATCHPOINT: - type = 3; + breakinfo[i].type = X86_BREAKPOINT_RW; break; default: return -1; } - - if (len == 1 || len == 2 || len == 4) - breakinfo[i].len = len - 1; - else + switch (len) { + case 1: + breakinfo[i].len = X86_BREAKPOINT_LEN_1; + break; + case 2: + breakinfo[i].len = X86_BREAKPOINT_LEN_2; + break; + case 4: + breakinfo[i].len = X86_BREAKPOINT_LEN_4; + break; +#ifdef CONFIG_X86_64 + case 8: + breakinfo[i].len = X86_BREAKPOINT_LEN_8; + break; +#endif + default: return -1; - - breakinfo[i].enabled = 1; + } breakinfo[i].addr = addr; - breakinfo[i].type = type; + if (hw_break_reserve_slot(i)) { + breakinfo[i].addr = 0; + return -1; + } + breakinfo[i].enabled = 1; return 0; } @@ -308,8 +382,21 @@ kgdb_set_hw_break(unsigned long addr, int len, enum kgdb_bptype bptype) */ void kgdb_disable_hw_debug(struct pt_regs *regs) { + int i; + int cpu = raw_smp_processor_id(); + struct perf_event *bp; + /* Disable hardware debugging while we are in kgdb: */ set_debugreg(0UL, 7); + for (i = 0; i < 4; i++) { + if (!breakinfo[i].enabled) + continue; + bp = *per_cpu_ptr(breakinfo[i].pev, cpu); + if (bp->attr.disabled == 1) + continue; + arch_uninstall_hw_breakpoint(bp); + bp->attr.disabled = 1; + } } /** @@ -373,7 +460,6 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code, struct pt_regs *linux_regs) { unsigned long addr; - unsigned long dr6; char *ptr; int newPC; @@ -395,25 +481,10 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code, /* set the trace bit if we're stepping */ if (remcomInBuffer[0] == 's') { linux_regs->flags |= X86_EFLAGS_TF; - kgdb_single_step = 1; atomic_set(&kgdb_cpu_doing_single_step, raw_smp_processor_id()); } - get_debugreg(dr6, 6); - if (!(dr6 & 0x4000)) { - int breakno; - - for (breakno = 0; breakno < 4; breakno++) { - if (dr6 & (1 << breakno) && - breakinfo[breakno].type == 0) { - /* Set restore flag: */ - linux_regs->flags |= X86_EFLAGS_RF; - break; - } - } - } - set_debugreg(0UL, 6); kgdb_correct_hw_break(); return 0; @@ -434,6 +505,11 @@ single_step_cont(struct pt_regs *regs, struct die_args *args) "resuming...\n"); kgdb_arch_handle_exception(args->trapnr, args->signr, args->err, "c", "", regs); + /* + * Reset the BS bit in dr6 (pointed by args->err) to + * denote completion of processing + */ + (*(unsigned long *)ERR_PTR(args->err)) &= ~DR_STEP; return NOTIFY_STOP; } @@ -476,8 +552,7 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd) break; case DIE_DEBUG: - if (atomic_read(&kgdb_cpu_doing_single_step) == - raw_smp_processor_id()) { + if (atomic_read(&kgdb_cpu_doing_single_step) != -1) { if (user_mode(regs)) return single_step_cont(regs, args); break; @@ -530,7 +605,42 @@ static struct notifier_block kgdb_notifier = { */ int kgdb_arch_init(void) { - return register_die_notifier(&kgdb_notifier); + int i, cpu; + int ret; + struct perf_event_attr attr; + struct perf_event **pevent; + + ret = register_die_notifier(&kgdb_notifier); + if (ret != 0) + return ret; + /* + * Pre-allocate the hw breakpoint structions in the non-atomic + * portion of kgdb because this operation requires mutexs to + * complete. + */ + attr.bp_addr = (unsigned long)kgdb_arch_init; + attr.type = PERF_TYPE_BREAKPOINT; + attr.bp_len = HW_BREAKPOINT_LEN_1; + attr.bp_type = HW_BREAKPOINT_W; + attr.disabled = 1; + for (i = 0; i < 4; i++) { + breakinfo[i].pev = register_wide_hw_breakpoint(&attr, NULL); + if (IS_ERR(breakinfo[i].pev)) { + printk(KERN_ERR "kgdb: Could not allocate hw breakpoints\n"); + breakinfo[i].pev = NULL; + kgdb_arch_exit(); + return -1; + } + for_each_online_cpu(cpu) { + pevent = per_cpu_ptr(breakinfo[i].pev, cpu); + pevent[0]->hw.sample_period = 1; + if (pevent[0]->destroy != NULL) { + pevent[0]->destroy = NULL; + release_bp_slot(*pevent); + } + } + } + return ret; } /** @@ -541,6 +651,13 @@ int kgdb_arch_init(void) */ void kgdb_arch_exit(void) { + int i; + for (i = 0; i < 4; i++) { + if (breakinfo[i].pev) { + unregister_wide_hw_breakpoint(breakinfo[i].pev); + breakinfo[i].pev = NULL; + } + } unregister_die_notifier(&kgdb_notifier); } diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 7b5169d2b000..b43bbaebe2c0 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -48,31 +48,23 @@ #include <linux/preempt.h> #include <linux/module.h> #include <linux/kdebug.h> +#include <linux/kallsyms.h> +#include <linux/ftrace.h> #include <asm/cacheflush.h> #include <asm/desc.h> #include <asm/pgtable.h> #include <asm/uaccess.h> #include <asm/alternative.h> +#include <asm/insn.h> +#include <asm/debugreg.h> void jprobe_return_end(void); DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL; DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk); -#ifdef CONFIG_X86_64 -#define stack_addr(regs) ((unsigned long *)regs->sp) -#else -/* - * "®s->sp" looks wrong, but it's correct for x86_32. x86_32 CPUs - * don't save the ss and esp registers if the CPU is already in kernel - * mode when it traps. So for kprobes, regs->sp and regs->ss are not - * the [nonexistent] saved stack pointer and ss register, but rather - * the top 8 bytes of the pre-int3 stack. So ®s->sp happens to - * point to the top of the pre-int3 stack. - */ -#define stack_addr(regs) ((unsigned long *)®s->sp) -#endif +#define stack_addr(regs) ((unsigned long *)kernel_stack_pointer(regs)) #define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\ (((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) | \ @@ -106,50 +98,6 @@ static const u32 twobyte_is_boostable[256 / 32] = { /* ----------------------------------------------- */ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */ }; -static const u32 onebyte_has_modrm[256 / 32] = { - /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */ - /* ----------------------------------------------- */ - W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 00 */ - W(0x10, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 10 */ - W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 20 */ - W(0x30, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 30 */ - W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 40 */ - W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */ - W(0x60, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0) | /* 60 */ - W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 70 */ - W(0x80, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 80 */ - W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 90 */ - W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* a0 */ - W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* b0 */ - W(0xc0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* c0 */ - W(0xd0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) , /* d0 */ - W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* e0 */ - W(0xf0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1) /* f0 */ - /* ----------------------------------------------- */ - /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */ -}; -static const u32 twobyte_has_modrm[256 / 32] = { - /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */ - /* ----------------------------------------------- */ - W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1) | /* 0f */ - W(0x10, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0) , /* 1f */ - W(0x20, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* 2f */ - W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */ - W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 4f */ - W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */ - W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 6f */ - W(0x70, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1) , /* 7f */ - W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */ - W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 9f */ - W(0xa0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) | /* af */ - W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1) , /* bf */ - W(0xc0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */ - W(0xd0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */ - W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* ef */ - W(0xf0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0) /* ff */ - /* ----------------------------------------------- */ - /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */ -}; #undef W struct kretprobe_blackpoint kretprobe_blacklist[] = { @@ -159,16 +107,22 @@ struct kretprobe_blackpoint kretprobe_blacklist[] = { }; const int kretprobe_blacklist_size = ARRAY_SIZE(kretprobe_blacklist); -/* Insert a jump instruction at address 'from', which jumps to address 'to'.*/ -static void __kprobes set_jmp_op(void *from, void *to) +static void __kprobes __synthesize_relative_insn(void *from, void *to, u8 op) { - struct __arch_jmp_op { - char op; + struct __arch_relative_insn { + u8 op; s32 raddr; - } __attribute__((packed)) * jop; - jop = (struct __arch_jmp_op *)from; - jop->raddr = (s32)((long)(to) - ((long)(from) + 5)); - jop->op = RELATIVEJUMP_INSTRUCTION; + } __attribute__((packed)) *insn; + + insn = (struct __arch_relative_insn *)from; + insn->raddr = (s32)((long)(to) - ((long)(from) + 5)); + insn->op = op; +} + +/* Insert a jump instruction at address 'from', which jumps to address 'to'.*/ +static void __kprobes synthesize_reljump(void *from, void *to) +{ + __synthesize_relative_insn(from, to, RELATIVEJUMP_OPCODE); } /* @@ -244,6 +198,75 @@ retry: } } +/* Recover the probed instruction at addr for further analysis. */ +static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr) +{ + struct kprobe *kp; + kp = get_kprobe((void *)addr); + if (!kp) + return -EINVAL; + + /* + * Basically, kp->ainsn.insn has an original instruction. + * However, RIP-relative instruction can not do single-stepping + * at different place, __copy_instruction() tweaks the displacement of + * that instruction. In that case, we can't recover the instruction + * from the kp->ainsn.insn. + * + * On the other hand, kp->opcode has a copy of the first byte of + * the probed instruction, which is overwritten by int3. And + * the instruction at kp->addr is not modified by kprobes except + * for the first byte, we can recover the original instruction + * from it and kp->opcode. + */ + memcpy(buf, kp->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); + buf[0] = kp->opcode; + 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) +{ + int ret; + unsigned long addr, offset = 0; + struct insn insn; + kprobe_opcode_t buf[MAX_INSN_SIZE]; + + if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf)) + return 0; + + /* Decode instructions */ + addr = paddr - offset; + while (addr < paddr) { + kernel_insn_init(&insn, (void *)addr); + insn_get_opcode(&insn); + + /* + * Check if the instruction has been modified by another + * kprobe, in which case we replace the breakpoint by the + * original instruction in our buffer. + */ + if (insn.opcode.bytes[0] == BREAKPOINT_INSTRUCTION) { + ret = recover_probed_instruction(buf, addr); + if (ret) + /* + * Another debugging subsystem might insert + * this breakpoint. In that case, we can't + * recover it. + */ + return 0; + kernel_insn_init(&insn, buf); + } + insn_get_length(&insn); + addr += insn.length; + } + + return (addr == paddr); +} + /* * Returns non-zero if opcode modifies the interrupt flag. */ @@ -268,86 +291,67 @@ static int __kprobes is_IF_modifier(kprobe_opcode_t *insn) } /* - * Adjust the displacement if the instruction uses the %rip-relative - * addressing mode. + * Copy an instruction and adjust the displacement if the instruction + * uses the %rip-relative addressing mode. * If it does, Return the address of the 32-bit displacement word. * If not, return null. * Only applicable to 64-bit x86. */ -static void __kprobes fix_riprel(struct kprobe *p) +static int __kprobes __copy_instruction(u8 *dest, u8 *src, int recover) { -#ifdef CONFIG_X86_64 - u8 *insn = p->ainsn.insn; - s64 disp; - int need_modrm; - - /* Skip legacy instruction prefixes. */ - while (1) { - switch (*insn) { - case 0x66: - case 0x67: - case 0x2e: - case 0x3e: - case 0x26: - case 0x64: - case 0x65: - case 0x36: - case 0xf0: - case 0xf3: - case 0xf2: - ++insn; - continue; + struct insn insn; + int ret; + kprobe_opcode_t buf[MAX_INSN_SIZE]; + + kernel_insn_init(&insn, src); + if (recover) { + insn_get_opcode(&insn); + if (insn.opcode.bytes[0] == BREAKPOINT_INSTRUCTION) { + ret = recover_probed_instruction(buf, + (unsigned long)src); + if (ret) + return 0; + kernel_insn_init(&insn, buf); } - break; } + insn_get_length(&insn); + memcpy(dest, insn.kaddr, insn.length); - /* Skip REX instruction prefix. */ - if (is_REX_prefix(insn)) - ++insn; - - if (*insn == 0x0f) { - /* Two-byte opcode. */ - ++insn; - need_modrm = test_bit(*insn, - (unsigned long *)twobyte_has_modrm); - } else - /* One-byte opcode. */ - need_modrm = test_bit(*insn, - (unsigned long *)onebyte_has_modrm); - - if (need_modrm) { - u8 modrm = *++insn; - if ((modrm & 0xc7) == 0x05) { - /* %rip+disp32 addressing mode */ - /* Displacement follows ModRM byte. */ - ++insn; - /* - * The copied instruction uses the %rip-relative - * addressing mode. Adjust the displacement for the - * difference between the original location of this - * instruction and the location of the copy that will - * actually be run. The tricky bit here is making sure - * that the sign extension happens correctly in this - * calculation, since we need a signed 32-bit result to - * be sign-extended to 64 bits when it's added to the - * %rip value and yield the same 64-bit result that the - * sign-extension of the original signed 32-bit - * displacement would have given. - */ - disp = (u8 *) p->addr + *((s32 *) insn) - - (u8 *) p->ainsn.insn; - BUG_ON((s64) (s32) disp != disp); /* Sanity check. */ - *(s32 *)insn = (s32) disp; - } +#ifdef CONFIG_X86_64 + if (insn_rip_relative(&insn)) { + s64 newdisp; + u8 *disp; + kernel_insn_init(&insn, dest); + insn_get_displacement(&insn); + /* + * The copied instruction uses the %rip-relative addressing + * mode. Adjust the displacement for the difference between + * the original location of this instruction and the location + * of the copy that will actually be run. The tricky bit here + * is making sure that the sign extension happens correctly in + * this calculation, since we need a signed 32-bit result to + * be sign-extended to 64 bits when it's added to the %rip + * value and yield the same 64-bit result that the sign- + * extension of the original signed 32-bit displacement would + * have given. + */ + newdisp = (u8 *) src + (s64) insn.displacement.value - + (u8 *) dest; + BUG_ON((s64) (s32) newdisp != newdisp); /* Sanity check. */ + disp = (u8 *) dest + insn_offset_displacement(&insn); + *(s32 *) disp = (s32) newdisp; } #endif + return insn.length; } static void __kprobes arch_copy_kprobe(struct kprobe *p) { - memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); - - fix_riprel(p); + /* + * Copy an instruction without recovering int3, because it will be + * put by another subsystem. + */ + __copy_instruction(p->ainsn.insn, p->addr, 0); if (can_boost(p->addr)) p->ainsn.boostable = 0; @@ -359,6 +363,11 @@ static void __kprobes arch_copy_kprobe(struct kprobe *p) int __kprobes arch_prepare_kprobe(struct kprobe *p) { + if (alternatives_text_reserved(p->addr, p->addr)) + return -EINVAL; + + if (!can_probe((unsigned long)p->addr)) + return -EILSEQ; /* insn: must be on special executable page on x86. */ p->ainsn.insn = get_insn_slot(); if (!p->ainsn.insn) @@ -423,18 +432,6 @@ static void __kprobes restore_btf(void) update_debugctlmsr(current->thread.debugctlmsr); } -static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs) -{ - clear_btf(); - regs->flags |= X86_EFLAGS_TF; - regs->flags &= ~X86_EFLAGS_IF; - /* single step inline if the instruction is an int3 */ - if (p->opcode == BREAKPOINT_INSTRUCTION) - regs->ip = (unsigned long)p->addr; - else - regs->ip = (unsigned long)p->ainsn.insn; -} - void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, struct pt_regs *regs) { @@ -446,20 +443,50 @@ void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, *sara = (unsigned long) &kretprobe_trampoline; } +#ifdef CONFIG_OPTPROBES +static int __kprobes setup_detour_execution(struct kprobe *p, + struct pt_regs *regs, + int reenter); +#else +#define setup_detour_execution(p, regs, reenter) (0) +#endif + static void __kprobes setup_singlestep(struct kprobe *p, struct pt_regs *regs, - struct kprobe_ctlblk *kcb) + struct kprobe_ctlblk *kcb, int reenter) { -#if !defined(CONFIG_PREEMPT) || defined(CONFIG_FREEZER) + if (setup_detour_execution(p, regs, reenter)) + return; + +#if !defined(CONFIG_PREEMPT) if (p->ainsn.boostable == 1 && !p->post_handler) { /* Boost up -- we can execute copied instructions directly */ - reset_current_kprobe(); + if (!reenter) + reset_current_kprobe(); + /* + * Reentering boosted probe doesn't reset current_kprobe, + * nor set current_kprobe, because it doesn't use single + * stepping. + */ regs->ip = (unsigned long)p->ainsn.insn; preempt_enable_no_resched(); return; } #endif - prepare_singlestep(p, regs); - kcb->kprobe_status = KPROBE_HIT_SS; + if (reenter) { + save_previous_kprobe(kcb); + set_current_kprobe(p, regs, kcb); + kcb->kprobe_status = KPROBE_REENTER; + } else + kcb->kprobe_status = KPROBE_HIT_SS; + /* Prepare real single stepping */ + clear_btf(); + regs->flags |= X86_EFLAGS_TF; + regs->flags &= ~X86_EFLAGS_IF; + /* single step inline if the instruction is an int3 */ + if (p->opcode == BREAKPOINT_INSTRUCTION) + regs->ip = (unsigned long)p->addr; + else + regs->ip = (unsigned long)p->ainsn.insn; } /* @@ -472,37 +499,21 @@ static int __kprobes reenter_kprobe(struct kprobe *p, struct pt_regs *regs, { switch (kcb->kprobe_status) { case KPROBE_HIT_SSDONE: -#ifdef CONFIG_X86_64 - /* TODO: Provide re-entrancy from post_kprobes_handler() and - * avoid exception stack corruption while single-stepping on - * the instruction of the new probe. - */ - arch_disarm_kprobe(p); - regs->ip = (unsigned long)p->addr; - reset_current_kprobe(); - preempt_enable_no_resched(); - break; -#endif case KPROBE_HIT_ACTIVE: - save_previous_kprobe(kcb); - set_current_kprobe(p, regs, kcb); kprobes_inc_nmissed_count(p); - prepare_singlestep(p, regs); - kcb->kprobe_status = KPROBE_REENTER; + setup_singlestep(p, regs, kcb, 1); break; case KPROBE_HIT_SS: - if (p == kprobe_running()) { - regs->flags &= ~X86_EFLAGS_TF; - regs->flags |= kcb->kprobe_saved_flags; - return 0; - } else { - /* A probe has been hit in the codepath leading up - * to, or just after, single-stepping of a probed - * instruction. This entire codepath should strictly - * reside in .kprobes.text section. Raise a warning - * to highlight this peculiar case. - */ - } + /* A probe has been hit in the codepath leading up to, or just + * after, single-stepping of a probed instruction. This entire + * codepath should strictly reside in .kprobes.text section. + * Raise a BUG or we'll continue in an endless reentering loop + * and eventually a stack overflow. + */ + printk(KERN_WARNING "Unrecoverable kprobe detected at %p.\n", + p->addr); + dump_kprobe(p); + BUG(); default: /* impossible cases */ WARN_ON(1); @@ -514,7 +525,7 @@ static int __kprobes reenter_kprobe(struct kprobe *p, struct pt_regs *regs, /* * Interrupts are disabled on entry as trap3 is an interrupt gate and they - * remain disabled thorough out this function. + * remain disabled throughout this function. */ static int __kprobes kprobe_handler(struct pt_regs *regs) { @@ -565,13 +576,13 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) * more here. */ if (!p->pre_handler || !p->pre_handler(p, regs)) - setup_singlestep(p, regs, kcb); + setup_singlestep(p, regs, kcb, 0); return 1; } } else if (kprobe_running()) { p = __get_cpu_var(current_kprobe); if (p->break_handler && p->break_handler(p, regs)) { - setup_singlestep(p, regs, kcb); + setup_singlestep(p, regs, kcb, 0); return 1; } } /* else: not a kprobe fault; let the kernel handle it */ @@ -580,6 +591,69 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) return 0; } +#ifdef CONFIG_X86_64 +#define SAVE_REGS_STRING \ + /* Skip cs, ip, orig_ax. */ \ + " subq $24, %rsp\n" \ + " pushq %rdi\n" \ + " pushq %rsi\n" \ + " pushq %rdx\n" \ + " pushq %rcx\n" \ + " pushq %rax\n" \ + " pushq %r8\n" \ + " pushq %r9\n" \ + " pushq %r10\n" \ + " pushq %r11\n" \ + " pushq %rbx\n" \ + " pushq %rbp\n" \ + " pushq %r12\n" \ + " pushq %r13\n" \ + " pushq %r14\n" \ + " pushq %r15\n" +#define RESTORE_REGS_STRING \ + " popq %r15\n" \ + " popq %r14\n" \ + " popq %r13\n" \ + " popq %r12\n" \ + " popq %rbp\n" \ + " popq %rbx\n" \ + " popq %r11\n" \ + " popq %r10\n" \ + " popq %r9\n" \ + " popq %r8\n" \ + " popq %rax\n" \ + " popq %rcx\n" \ + " popq %rdx\n" \ + " popq %rsi\n" \ + " popq %rdi\n" \ + /* Skip orig_ax, ip, cs */ \ + " addq $24, %rsp\n" +#else +#define SAVE_REGS_STRING \ + /* Skip cs, ip, orig_ax and gs. */ \ + " subl $16, %esp\n" \ + " pushl %fs\n" \ + " pushl %ds\n" \ + " pushl %es\n" \ + " pushl %eax\n" \ + " pushl %ebp\n" \ + " pushl %edi\n" \ + " pushl %esi\n" \ + " pushl %edx\n" \ + " pushl %ecx\n" \ + " pushl %ebx\n" +#define RESTORE_REGS_STRING \ + " popl %ebx\n" \ + " popl %ecx\n" \ + " popl %edx\n" \ + " popl %esi\n" \ + " popl %edi\n" \ + " popl %ebp\n" \ + " popl %eax\n" \ + /* Skip ds, es, fs, gs, orig_ax, and ip. Note: don't pop cs here*/\ + " addl $24, %esp\n" +#endif + /* * When a retprobed function returns, this code saves registers and * calls trampoline_handler() runs, which calls the kretprobe's handler. @@ -593,65 +667,16 @@ static void __used __kprobes kretprobe_trampoline_holder(void) /* We don't bother saving the ss register */ " pushq %rsp\n" " pushfq\n" - /* - * Skip cs, ip, orig_ax. - * trampoline_handler() will plug in these values - */ - " subq $24, %rsp\n" - " pushq %rdi\n" - " pushq %rsi\n" - " pushq %rdx\n" - " pushq %rcx\n" - " pushq %rax\n" - " pushq %r8\n" - " pushq %r9\n" - " pushq %r10\n" - " pushq %r11\n" - " pushq %rbx\n" - " pushq %rbp\n" - " pushq %r12\n" - " pushq %r13\n" - " pushq %r14\n" - " pushq %r15\n" + SAVE_REGS_STRING " movq %rsp, %rdi\n" " call trampoline_handler\n" /* Replace saved sp with true return address. */ " movq %rax, 152(%rsp)\n" - " popq %r15\n" - " popq %r14\n" - " popq %r13\n" - " popq %r12\n" - " popq %rbp\n" - " popq %rbx\n" - " popq %r11\n" - " popq %r10\n" - " popq %r9\n" - " popq %r8\n" - " popq %rax\n" - " popq %rcx\n" - " popq %rdx\n" - " popq %rsi\n" - " popq %rdi\n" - /* Skip orig_ax, ip, cs */ - " addq $24, %rsp\n" + RESTORE_REGS_STRING " popfq\n" #else " pushf\n" - /* - * Skip cs, ip, orig_ax and gs. - * trampoline_handler() will plug in these values - */ - " subl $16, %esp\n" - " pushl %fs\n" - " pushl %es\n" - " pushl %ds\n" - " pushl %eax\n" - " pushl %ebp\n" - " pushl %edi\n" - " pushl %esi\n" - " pushl %edx\n" - " pushl %ecx\n" - " pushl %ebx\n" + SAVE_REGS_STRING " movl %esp, %eax\n" " call trampoline_handler\n" /* Move flags to cs */ @@ -659,15 +684,7 @@ static void __used __kprobes kretprobe_trampoline_holder(void) " movl %edx, 52(%esp)\n" /* Replace saved flags with true return address. */ " movl %eax, 56(%esp)\n" - " popl %ebx\n" - " popl %ecx\n" - " popl %edx\n" - " popl %esi\n" - " popl %edi\n" - " popl %ebp\n" - " popl %eax\n" - /* Skip ds, es, fs, gs, orig_ax and ip */ - " addl $24, %esp\n" + RESTORE_REGS_STRING " popf\n" #endif " ret\n"); @@ -835,8 +852,8 @@ static void __kprobes resume_execution(struct kprobe *p, * These instructions can be executed directly if it * jumps back to correct address. */ - set_jmp_op((void *)regs->ip, - (void *)orig_ip + (regs->ip - copy_ip)); + synthesize_reljump((void *)regs->ip, + (void *)orig_ip + (regs->ip - copy_ip)); p->ainsn.boostable = 1; } else { p->ainsn.boostable = -1; @@ -851,7 +868,7 @@ no_change: /* * Interrupts are disabled on entry as trap1 is an interrupt gate and they - * remain disabled thoroughout this function. + * remain disabled throughout this function. */ static int __kprobes post_kprobe_handler(struct pt_regs *regs) { @@ -967,8 +984,14 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, ret = NOTIFY_STOP; break; case DIE_DEBUG: - if (post_kprobe_handler(args->regs)) + if (post_kprobe_handler(args->regs)) { + /* + * Reset the BS bit in dr6 (pointed by args->err) to + * denote completion of processing + */ + (*(unsigned long *)ERR_PTR(args->err)) &= ~DR_STEP; ret = NOTIFY_STOP; + } break; case DIE_GPF: /* @@ -1057,6 +1080,358 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs) return 0; } + +#ifdef CONFIG_OPTPROBES + +/* Insert a call instruction at address 'from', which calls address 'to'.*/ +static void __kprobes synthesize_relcall(void *from, void *to) +{ + __synthesize_relative_insn(from, to, RELATIVECALL_OPCODE); +} + +/* Insert a move instruction which sets a pointer to eax/rdi (1st arg). */ +static void __kprobes synthesize_set_arg1(kprobe_opcode_t *addr, + unsigned long val) +{ +#ifdef CONFIG_X86_64 + *addr++ = 0x48; + *addr++ = 0xbf; +#else + *addr++ = 0xb8; +#endif + *(unsigned long *)addr = val; +} + +void __kprobes kprobes_optinsn_template_holder(void) +{ + asm volatile ( + ".global optprobe_template_entry\n" + "optprobe_template_entry: \n" +#ifdef CONFIG_X86_64 + /* We don't bother saving the ss register */ + " pushq %rsp\n" + " pushfq\n" + SAVE_REGS_STRING + " movq %rsp, %rsi\n" + ".global optprobe_template_val\n" + "optprobe_template_val: \n" + ASM_NOP5 + ASM_NOP5 + ".global optprobe_template_call\n" + "optprobe_template_call: \n" + ASM_NOP5 + /* Move flags to rsp */ + " movq 144(%rsp), %rdx\n" + " movq %rdx, 152(%rsp)\n" + RESTORE_REGS_STRING + /* Skip flags entry */ + " addq $8, %rsp\n" + " popfq\n" +#else /* CONFIG_X86_32 */ + " pushf\n" + SAVE_REGS_STRING + " movl %esp, %edx\n" + ".global optprobe_template_val\n" + "optprobe_template_val: \n" + ASM_NOP5 + ".global optprobe_template_call\n" + "optprobe_template_call: \n" + ASM_NOP5 + RESTORE_REGS_STRING + " addl $4, %esp\n" /* skip cs */ + " popf\n" +#endif + ".global optprobe_template_end\n" + "optprobe_template_end: \n"); +} + +#define TMPL_MOVE_IDX \ + ((long)&optprobe_template_val - (long)&optprobe_template_entry) +#define TMPL_CALL_IDX \ + ((long)&optprobe_template_call - (long)&optprobe_template_entry) +#define TMPL_END_IDX \ + ((long)&optprobe_template_end - (long)&optprobe_template_entry) + +#define INT3_SIZE sizeof(kprobe_opcode_t) + +/* Optimized kprobe call back function: called from optinsn */ +static void __kprobes optimized_callback(struct optimized_kprobe *op, + struct pt_regs *regs) +{ + struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + + preempt_disable(); + if (kprobe_running()) { + kprobes_inc_nmissed_count(&op->kp); + } else { + /* Save skipped registers */ +#ifdef CONFIG_X86_64 + regs->cs = __KERNEL_CS; +#else + regs->cs = __KERNEL_CS | get_kernel_rpl(); + regs->gs = 0; +#endif + regs->ip = (unsigned long)op->kp.addr + INT3_SIZE; + regs->orig_ax = ~0UL; + + __get_cpu_var(current_kprobe) = &op->kp; + kcb->kprobe_status = KPROBE_HIT_ACTIVE; + opt_pre_handler(&op->kp, regs); + __get_cpu_var(current_kprobe) = NULL; + } + preempt_enable_no_resched(); +} + +static int __kprobes copy_optimized_instructions(u8 *dest, u8 *src) +{ + int len = 0, ret; + + while (len < RELATIVEJUMP_SIZE) { + ret = __copy_instruction(dest + len, src + len, 1); + if (!ret || !can_boost(dest + len)) + return -EINVAL; + len += ret; + } + /* Check whether the address range is reserved */ + if (ftrace_text_reserved(src, src + len - 1) || + alternatives_text_reserved(src, src + len - 1)) + return -EBUSY; + + return len; +} + +/* Check whether insn is indirect jump */ +static int __kprobes insn_is_indirect_jump(struct insn *insn) +{ + return ((insn->opcode.bytes[0] == 0xff && + (X86_MODRM_REG(insn->modrm.value) & 6) == 4) || /* Jump */ + insn->opcode.bytes[0] == 0xea); /* Segment based jump */ +} + +/* Check whether insn jumps into specified address range */ +static int insn_jump_into_range(struct insn *insn, unsigned long start, int len) +{ + unsigned long target = 0; + + switch (insn->opcode.bytes[0]) { + case 0xe0: /* loopne */ + case 0xe1: /* loope */ + case 0xe2: /* loop */ + case 0xe3: /* jcxz */ + case 0xe9: /* near relative jump */ + case 0xeb: /* short relative jump */ + break; + case 0x0f: + if ((insn->opcode.bytes[1] & 0xf0) == 0x80) /* jcc near */ + break; + return 0; + default: + if ((insn->opcode.bytes[0] & 0xf0) == 0x70) /* jcc short */ + break; + return 0; + } + target = (unsigned long)insn->next_byte + insn->immediate.value; + + return (start <= target && target <= start + len); +} + +/* Decode whole function to ensure any instructions don't jump into target */ +static int __kprobes can_optimize(unsigned long paddr) +{ + int ret; + 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)) + return 0; + + /* Check there is enough space for a relative jump. */ + if (size - offset < RELATIVEJUMP_SIZE) + return 0; + + /* Decode instructions */ + addr = paddr - offset; + while (addr < paddr - offset + size) { /* Decode until function end */ + if (search_exception_tables(addr)) + /* + * Since some fixup code will jumps into this function, + * we can't optimize kprobe in this function. + */ + return 0; + kernel_insn_init(&insn, (void *)addr); + insn_get_opcode(&insn); + if (insn.opcode.bytes[0] == BREAKPOINT_INSTRUCTION) { + ret = recover_probed_instruction(buf, addr); + if (ret) + return 0; + kernel_insn_init(&insn, buf); + } + insn_get_length(&insn); + /* Recover address */ + insn.kaddr = (void *)addr; + insn.next_byte = (void *)(addr + insn.length); + /* Check any instructions don't jump into target */ + if (insn_is_indirect_jump(&insn) || + insn_jump_into_range(&insn, paddr + INT3_SIZE, + RELATIVE_ADDR_SIZE)) + return 0; + addr += insn.length; + } + + return 1; +} + +/* Check optimized_kprobe can actually be optimized. */ +int __kprobes arch_check_optimized_kprobe(struct optimized_kprobe *op) +{ + int i; + struct kprobe *p; + + for (i = 1; i < op->optinsn.size; i++) { + p = get_kprobe(op->kp.addr + i); + if (p && !kprobe_disabled(p)) + return -EEXIST; + } + + return 0; +} + +/* Check the addr is within the optimized instructions. */ +int __kprobes arch_within_optimized_kprobe(struct optimized_kprobe *op, + unsigned long addr) +{ + return ((unsigned long)op->kp.addr <= addr && + (unsigned long)op->kp.addr + op->optinsn.size > addr); +} + +/* Free optimized instruction slot */ +static __kprobes +void __arch_remove_optimized_kprobe(struct optimized_kprobe *op, int dirty) +{ + if (op->optinsn.insn) { + free_optinsn_slot(op->optinsn.insn, dirty); + op->optinsn.insn = NULL; + op->optinsn.size = 0; + } +} + +void __kprobes arch_remove_optimized_kprobe(struct optimized_kprobe *op) +{ + __arch_remove_optimized_kprobe(op, 1); +} + +/* + * Copy replacing target instructions + * Target instructions MUST be relocatable (checked inside) + */ +int __kprobes arch_prepare_optimized_kprobe(struct optimized_kprobe *op) +{ + u8 *buf; + int ret; + long rel; + + if (!can_optimize((unsigned long)op->kp.addr)) + return -EILSEQ; + + op->optinsn.insn = get_optinsn_slot(); + if (!op->optinsn.insn) + return -ENOMEM; + + /* + * Verify if the address gap is in 2GB range, because this uses + * a relative jump. + */ + rel = (long)op->optinsn.insn - (long)op->kp.addr + RELATIVEJUMP_SIZE; + if (abs(rel) > 0x7fffffff) + return -ERANGE; + + buf = (u8 *)op->optinsn.insn; + + /* Copy instructions into the out-of-line buffer */ + ret = copy_optimized_instructions(buf + TMPL_END_IDX, op->kp.addr); + if (ret < 0) { + __arch_remove_optimized_kprobe(op, 0); + return ret; + } + op->optinsn.size = ret; + + /* Copy arch-dep-instance from template */ + memcpy(buf, &optprobe_template_entry, TMPL_END_IDX); + + /* Set probe information */ + synthesize_set_arg1(buf + TMPL_MOVE_IDX, (unsigned long)op); + + /* Set probe function call */ + synthesize_relcall(buf + TMPL_CALL_IDX, optimized_callback); + + /* Set returning jmp instruction at the tail of out-of-line buffer */ + synthesize_reljump(buf + TMPL_END_IDX + op->optinsn.size, + (u8 *)op->kp.addr + op->optinsn.size); + + flush_icache_range((unsigned long) buf, + (unsigned long) buf + TMPL_END_IDX + + op->optinsn.size + RELATIVEJUMP_SIZE); + return 0; +} + +/* Replace a breakpoint (int3) with a relative jump. */ +int __kprobes arch_optimize_kprobe(struct optimized_kprobe *op) +{ + unsigned char jmp_code[RELATIVEJUMP_SIZE]; + s32 rel = (s32)((long)op->optinsn.insn - + ((long)op->kp.addr + RELATIVEJUMP_SIZE)); + + /* Backup instructions which will be replaced by jump address */ + memcpy(op->optinsn.copied_insn, op->kp.addr + INT3_SIZE, + RELATIVE_ADDR_SIZE); + + jmp_code[0] = RELATIVEJUMP_OPCODE; + *(s32 *)(&jmp_code[1]) = rel; + + /* + * text_poke_smp doesn't support NMI/MCE code modifying. + * However, since kprobes itself also doesn't support NMI/MCE + * code probing, it's not a problem. + */ + text_poke_smp(op->kp.addr, jmp_code, RELATIVEJUMP_SIZE); + return 0; +} + +/* Replace a relative jump with a breakpoint (int3). */ +void __kprobes arch_unoptimize_kprobe(struct optimized_kprobe *op) +{ + u8 buf[RELATIVEJUMP_SIZE]; + + /* Set int3 to first byte for kprobes */ + buf[0] = BREAKPOINT_INSTRUCTION; + memcpy(buf + 1, op->optinsn.copied_insn, RELATIVE_ADDR_SIZE); + text_poke_smp(op->kp.addr, buf, RELATIVEJUMP_SIZE); +} + +static int __kprobes setup_detour_execution(struct kprobe *p, + struct pt_regs *regs, + int reenter) +{ + struct optimized_kprobe *op; + + if (p->flags & KPROBE_FLAG_OPTIMIZED) { + /* This kprobe is really able to run optimized path. */ + op = container_of(p, struct optimized_kprobe, kp); + /* Detour through copied instructions */ + regs->ip = (unsigned long)op->optinsn.insn + TMPL_END_IDX; + if (!reenter) + reset_current_kprobe(); + preempt_enable_no_resched(); + return 1; + } + return 0; +} +#endif + int __init arch_init_kprobes(void) { return 0; diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c index c1c429d00130..a3fa43ba5d3b 100644 --- a/arch/x86/kernel/machine_kexec_32.c +++ b/arch/x86/kernel/machine_kexec_32.c @@ -25,6 +25,7 @@ #include <asm/desc.h> #include <asm/system.h> #include <asm/cacheflush.h> +#include <asm/debugreg.h> static void set_idt(void *newidt, __u16 limit) { @@ -157,8 +158,7 @@ int machine_kexec_prepare(struct kimage *image) { int error; - if (nx_enabled) - set_pages_x(image->control_code_page, 1); + set_pages_x(image->control_code_page, 1); error = machine_kexec_alloc_page_tables(image); if (error) return error; @@ -172,8 +172,7 @@ int machine_kexec_prepare(struct kimage *image) */ void machine_kexec_cleanup(struct kimage *image) { - if (nx_enabled) - set_pages_nx(image->control_code_page, 1); + set_pages_nx(image->control_code_page, 1); machine_kexec_free_page_tables(image); } @@ -202,6 +201,7 @@ void machine_kexec(struct kimage *image) /* Interrupts aren't acceptable while we reboot */ local_irq_disable(); + hw_breakpoint_disable(); if (image->preserve_context) { #ifdef CONFIG_X86_IO_APIC diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c index 84c3bf209e98..4a8bb82248ae 100644 --- a/arch/x86/kernel/machine_kexec_64.c +++ b/arch/x86/kernel/machine_kexec_64.c @@ -18,6 +18,7 @@ #include <asm/pgtable.h> #include <asm/tlbflush.h> #include <asm/mmu_context.h> +#include <asm/debugreg.h> static int init_one_level2_page(struct kimage *image, pgd_t *pgd, unsigned long addr) @@ -282,6 +283,7 @@ void machine_kexec(struct kimage *image) /* Interrupts aren't acceptable while we reboot */ local_irq_disable(); + hw_breakpoint_disable(); if (image->preserve_context) { #ifdef CONFIG_X86_IO_APIC diff --git a/arch/x86/kernel/mfgpt_32.c b/arch/x86/kernel/mfgpt_32.c deleted file mode 100644 index 2a62d843f015..000000000000 --- a/arch/x86/kernel/mfgpt_32.c +++ /dev/null @@ -1,410 +0,0 @@ -/* - * Driver/API for AMD Geode Multi-Function General Purpose Timers (MFGPT) - * - * Copyright (C) 2006, Advanced Micro Devices, Inc. - * Copyright (C) 2007, Andres Salomon <dilinger@debian.org> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of version 2 of the GNU General Public License - * as published by the Free Software Foundation. - * - * The MFGPTs are documented in AMD Geode CS5536 Companion Device Data Book. - */ - -/* - * We are using the 32.768kHz input clock - it's the only one that has the - * ranges we find desirable. The following table lists the suitable - * divisors and the associated Hz, minimum interval and the maximum interval: - * - * Divisor Hz Min Delta (s) Max Delta (s) - * 1 32768 .00048828125 2.000 - * 2 16384 .0009765625 4.000 - * 4 8192 .001953125 8.000 - * 8 4096 .00390625 16.000 - * 16 2048 .0078125 32.000 - * 32 1024 .015625 64.000 - * 64 512 .03125 128.000 - * 128 256 .0625 256.000 - * 256 128 .125 512.000 - */ - -#include <linux/kernel.h> -#include <linux/interrupt.h> -#include <linux/module.h> -#include <asm/geode.h> - -#define MFGPT_DEFAULT_IRQ 7 - -static struct mfgpt_timer_t { - unsigned int avail:1; -} mfgpt_timers[MFGPT_MAX_TIMERS]; - -/* Selected from the table above */ - -#define MFGPT_DIVISOR 16 -#define MFGPT_SCALE 4 /* divisor = 2^(scale) */ -#define MFGPT_HZ (32768 / MFGPT_DIVISOR) -#define MFGPT_PERIODIC (MFGPT_HZ / HZ) - -/* Allow for disabling of MFGPTs */ -static int disable; -static int __init mfgpt_disable(char *s) -{ - disable = 1; - return 1; -} -__setup("nomfgpt", mfgpt_disable); - -/* Reset the MFGPT timers. This is required by some broken BIOSes which already - * do the same and leave the system in an unstable state. TinyBIOS 0.98 is - * affected at least (0.99 is OK with MFGPT workaround left to off). - */ -static int __init mfgpt_fix(char *s) -{ - u32 val, dummy; - - /* The following udocumented bit resets the MFGPT timers */ - val = 0xFF; dummy = 0; - wrmsr(MSR_MFGPT_SETUP, val, dummy); - return 1; -} -__setup("mfgptfix", mfgpt_fix); - -/* - * Check whether any MFGPTs are available for the kernel to use. In most - * cases, firmware that uses AMD's VSA code will claim all timers during - * bootup; we certainly don't want to take them if they're already in use. - * In other cases (such as with VSAless OpenFirmware), the system firmware - * leaves timers available for us to use. - */ - - -static int timers = -1; - -static void geode_mfgpt_detect(void) -{ - int i; - u16 val; - - timers = 0; - - if (disable) { - printk(KERN_INFO "geode-mfgpt: MFGPT support is disabled\n"); - goto done; - } - - if (!geode_get_dev_base(GEODE_DEV_MFGPT)) { - printk(KERN_INFO "geode-mfgpt: MFGPT LBAR is not set up\n"); - goto done; - } - - for (i = 0; i < MFGPT_MAX_TIMERS; i++) { - val = geode_mfgpt_read(i, MFGPT_REG_SETUP); - if (!(val & MFGPT_SETUP_SETUP)) { - mfgpt_timers[i].avail = 1; - timers++; - } - } - -done: - printk(KERN_INFO "geode-mfgpt: %d MFGPT timers available.\n", timers); -} - -int geode_mfgpt_toggle_event(int timer, int cmp, int event, int enable) -{ - u32 msr, mask, value, dummy; - int shift = (cmp == MFGPT_CMP1) ? 0 : 8; - - if (timer < 0 || timer >= MFGPT_MAX_TIMERS) - return -EIO; - - /* - * The register maps for these are described in sections 6.17.1.x of - * the AMD Geode CS5536 Companion Device Data Book. - */ - switch (event) { - case MFGPT_EVENT_RESET: - /* - * XXX: According to the docs, we cannot reset timers above - * 6; that is, resets for 7 and 8 will be ignored. Is this - * a problem? -dilinger - */ - msr = MSR_MFGPT_NR; - mask = 1 << (timer + 24); - break; - - case MFGPT_EVENT_NMI: - msr = MSR_MFGPT_NR; - mask = 1 << (timer + shift); - break; - - case MFGPT_EVENT_IRQ: - msr = MSR_MFGPT_IRQ; - mask = 1 << (timer + shift); - break; - - default: - return -EIO; - } - - rdmsr(msr, value, dummy); - - if (enable) - value |= mask; - else - value &= ~mask; - - wrmsr(msr, value, dummy); - return 0; -} -EXPORT_SYMBOL_GPL(geode_mfgpt_toggle_event); - -int geode_mfgpt_set_irq(int timer, int cmp, int *irq, int enable) -{ - u32 zsel, lpc, dummy; - int shift; - - if (timer < 0 || timer >= MFGPT_MAX_TIMERS) - return -EIO; - - /* - * Unfortunately, MFGPTs come in pairs sharing their IRQ lines. If VSA - * is using the same CMP of the timer's Siamese twin, the IRQ is set to - * 2, and we mustn't use nor change it. - * XXX: Likewise, 2 Linux drivers might clash if the 2nd overwrites the - * IRQ of the 1st. This can only happen if forcing an IRQ, calling this - * with *irq==0 is safe. Currently there _are_ no 2 drivers. - */ - rdmsr(MSR_PIC_ZSEL_LOW, zsel, dummy); - shift = ((cmp == MFGPT_CMP1 ? 0 : 4) + timer % 4) * 4; - if (((zsel >> shift) & 0xF) == 2) - return -EIO; - - /* Choose IRQ: if none supplied, keep IRQ already set or use default */ - if (!*irq) - *irq = (zsel >> shift) & 0xF; - if (!*irq) - *irq = MFGPT_DEFAULT_IRQ; - - /* Can't use IRQ if it's 0 (=disabled), 2, or routed to LPC */ - if (*irq < 1 || *irq == 2 || *irq > 15) - return -EIO; - rdmsr(MSR_PIC_IRQM_LPC, lpc, dummy); - if (lpc & (1 << *irq)) - return -EIO; - - /* All chosen and checked - go for it */ - if (geode_mfgpt_toggle_event(timer, cmp, MFGPT_EVENT_IRQ, enable)) - return -EIO; - if (enable) { - zsel = (zsel & ~(0xF << shift)) | (*irq << shift); - wrmsr(MSR_PIC_ZSEL_LOW, zsel, dummy); - } - - return 0; -} - -static int mfgpt_get(int timer) -{ - mfgpt_timers[timer].avail = 0; - printk(KERN_INFO "geode-mfgpt: Registered timer %d\n", timer); - return timer; -} - -int geode_mfgpt_alloc_timer(int timer, int domain) -{ - int i; - - if (timers == -1) { - /* timers haven't been detected yet */ - geode_mfgpt_detect(); - } - - if (!timers) - return -1; - - if (timer >= MFGPT_MAX_TIMERS) - return -1; - - if (timer < 0) { - /* Try to find an available timer */ - for (i = 0; i < MFGPT_MAX_TIMERS; i++) { - if (mfgpt_timers[i].avail) - return mfgpt_get(i); - - if (i == 5 && domain == MFGPT_DOMAIN_WORKING) - break; - } - } else { - /* If they requested a specific timer, try to honor that */ - if (mfgpt_timers[timer].avail) - return mfgpt_get(timer); - } - - /* No timers available - too bad */ - return -1; -} -EXPORT_SYMBOL_GPL(geode_mfgpt_alloc_timer); - - -#ifdef CONFIG_GEODE_MFGPT_TIMER - -/* - * The MFPGT timers on the CS5536 provide us with suitable timers to use - * as clock event sources - not as good as a HPET or APIC, but certainly - * better than the PIT. This isn't a general purpose MFGPT driver, but - * a simplified one designed specifically to act as a clock event source. - * For full details about the MFGPT, please consult the CS5536 data sheet. - */ - -#include <linux/clocksource.h> -#include <linux/clockchips.h> - -static unsigned int mfgpt_tick_mode = CLOCK_EVT_MODE_SHUTDOWN; -static u16 mfgpt_event_clock; - -static int irq; -static int __init mfgpt_setup(char *str) -{ - get_option(&str, &irq); - return 1; -} -__setup("mfgpt_irq=", mfgpt_setup); - -static void mfgpt_disable_timer(u16 clock) -{ - /* avoid races by clearing CMP1 and CMP2 unconditionally */ - geode_mfgpt_write(clock, MFGPT_REG_SETUP, (u16) ~MFGPT_SETUP_CNTEN | - MFGPT_SETUP_CMP1 | MFGPT_SETUP_CMP2); -} - -static int mfgpt_next_event(unsigned long, struct clock_event_device *); -static void mfgpt_set_mode(enum clock_event_mode, struct clock_event_device *); - -static struct clock_event_device mfgpt_clockevent = { - .name = "mfgpt-timer", - .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT, - .set_mode = mfgpt_set_mode, - .set_next_event = mfgpt_next_event, - .rating = 250, - .cpumask = cpu_all_mask, - .shift = 32 -}; - -static void mfgpt_start_timer(u16 delta) -{ - geode_mfgpt_write(mfgpt_event_clock, MFGPT_REG_CMP2, (u16) delta); - geode_mfgpt_write(mfgpt_event_clock, MFGPT_REG_COUNTER, 0); - - geode_mfgpt_write(mfgpt_event_clock, MFGPT_REG_SETUP, - MFGPT_SETUP_CNTEN | MFGPT_SETUP_CMP2); -} - -static void mfgpt_set_mode(enum clock_event_mode mode, - struct clock_event_device *evt) -{ - mfgpt_disable_timer(mfgpt_event_clock); - - if (mode == CLOCK_EVT_MODE_PERIODIC) - mfgpt_start_timer(MFGPT_PERIODIC); - - mfgpt_tick_mode = mode; -} - -static int mfgpt_next_event(unsigned long delta, struct clock_event_device *evt) -{ - mfgpt_start_timer(delta); - return 0; -} - -static irqreturn_t mfgpt_tick(int irq, void *dev_id) -{ - u16 val = geode_mfgpt_read(mfgpt_event_clock, MFGPT_REG_SETUP); - - /* See if the interrupt was for us */ - if (!(val & (MFGPT_SETUP_SETUP | MFGPT_SETUP_CMP2 | MFGPT_SETUP_CMP1))) - return IRQ_NONE; - - /* Turn off the clock (and clear the event) */ - mfgpt_disable_timer(mfgpt_event_clock); - - if (mfgpt_tick_mode == CLOCK_EVT_MODE_SHUTDOWN) - return IRQ_HANDLED; - - /* Clear the counter */ - geode_mfgpt_write(mfgpt_event_clock, MFGPT_REG_COUNTER, 0); - - /* Restart the clock in periodic mode */ - - if (mfgpt_tick_mode == CLOCK_EVT_MODE_PERIODIC) { - geode_mfgpt_write(mfgpt_event_clock, MFGPT_REG_SETUP, - MFGPT_SETUP_CNTEN | MFGPT_SETUP_CMP2); - } - - mfgpt_clockevent.event_handler(&mfgpt_clockevent); - return IRQ_HANDLED; -} - -static struct irqaction mfgptirq = { - .handler = mfgpt_tick, - .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_TIMER, - .name = "mfgpt-timer" -}; - -int __init mfgpt_timer_setup(void) -{ - int timer, ret; - u16 val; - - timer = geode_mfgpt_alloc_timer(MFGPT_TIMER_ANY, MFGPT_DOMAIN_WORKING); - if (timer < 0) { - printk(KERN_ERR - "mfgpt-timer: Could not allocate a MFPGT timer\n"); - return -ENODEV; - } - - mfgpt_event_clock = timer; - - /* Set up the IRQ on the MFGPT side */ - if (geode_mfgpt_setup_irq(mfgpt_event_clock, MFGPT_CMP2, &irq)) { - printk(KERN_ERR "mfgpt-timer: Could not set up IRQ %d\n", irq); - return -EIO; - } - - /* And register it with the kernel */ - ret = setup_irq(irq, &mfgptirq); - - if (ret) { - printk(KERN_ERR - "mfgpt-timer: Unable to set up the interrupt.\n"); - goto err; - } - - /* Set the clock scale and enable the event mode for CMP2 */ - val = MFGPT_SCALE | (3 << 8); - - geode_mfgpt_write(mfgpt_event_clock, MFGPT_REG_SETUP, val); - - /* Set up the clock event */ - mfgpt_clockevent.mult = div_sc(MFGPT_HZ, NSEC_PER_SEC, - mfgpt_clockevent.shift); - mfgpt_clockevent.min_delta_ns = clockevent_delta2ns(0xF, - &mfgpt_clockevent); - mfgpt_clockevent.max_delta_ns = clockevent_delta2ns(0xFFFE, - &mfgpt_clockevent); - - printk(KERN_INFO - "mfgpt-timer: Registering MFGPT timer %d as a clock event, using IRQ %d\n", - timer, irq); - clockevents_register_device(&mfgpt_clockevent); - - return 0; - -err: - geode_mfgpt_release_irq(mfgpt_event_clock, MFGPT_CMP2, &irq); - printk(KERN_ERR - "mfgpt-timer: Unable to set up the MFGPT clock source\n"); - return -EIO; -} - -#endif diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c index 366baa179913..e1af7c055c7d 100644 --- a/arch/x86/kernel/microcode_amd.c +++ b/arch/x86/kernel/microcode_amd.c @@ -13,6 +13,9 @@ * Licensed under the terms of the GNU General Public * License version 2. See file COPYING for details. */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/firmware.h> #include <linux/pci_ids.h> #include <linux/uaccess.h> @@ -76,12 +79,12 @@ static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig) memset(csig, 0, sizeof(*csig)); if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) { - printk(KERN_WARNING "microcode: CPU%d: AMD CPU family 0x%x not " - "supported\n", cpu, c->x86); + pr_warning("microcode: CPU%d: AMD CPU family 0x%x not " + "supported\n", cpu, c->x86); return -1; } rdmsr(MSR_AMD64_PATCH_LEVEL, csig->rev, dummy); - printk(KERN_INFO "microcode: CPU%d: patch_level=0x%x\n", cpu, csig->rev); + pr_info("CPU%d: patch_level=0x%x\n", cpu, csig->rev); return 0; } @@ -103,23 +106,16 @@ static int get_matching_microcode(int cpu, void *mc, int rev) i++; } - if (!equiv_cpu_id) { - printk(KERN_WARNING "microcode: CPU%d: cpu revision " - "not listed in equivalent cpu table\n", cpu); + if (!equiv_cpu_id) return 0; - } - if (mc_header->processor_rev_id != equiv_cpu_id) { - printk(KERN_ERR "microcode: CPU%d: patch mismatch " - "(processor_rev_id: %x, equiv_cpu_id: %x)\n", - cpu, mc_header->processor_rev_id, equiv_cpu_id); + if (mc_header->processor_rev_id != equiv_cpu_id) return 0; - } /* ucode might be chipset specific -- currently we don't support this */ if (mc_header->nb_dev_id || mc_header->sb_dev_id) { - printk(KERN_ERR "microcode: CPU%d: loading of chipset " - "specific code not yet supported\n", cpu); + pr_err("CPU%d: loading of chipset specific code not yet supported\n", + cpu); return 0; } @@ -148,14 +144,12 @@ static int apply_microcode_amd(int cpu) /* check current patch id and patch's id for match */ if (rev != mc_amd->hdr.patch_id) { - printk(KERN_ERR "microcode: CPU%d: update failed " - "(for patch_level=0x%x)\n", cpu, mc_amd->hdr.patch_id); + pr_err("CPU%d: update failed (for patch_level=0x%x)\n", + cpu, mc_amd->hdr.patch_id); return -1; } - printk(KERN_INFO "microcode: CPU%d: updated (new patch_level=0x%x)\n", - cpu, rev); - + pr_info("CPU%d: updated (new patch_level=0x%x)\n", cpu, rev); uci->cpu_sig.rev = rev; return 0; @@ -178,18 +172,14 @@ get_next_ucode(const u8 *buf, unsigned int size, unsigned int *mc_size) return NULL; if (section_hdr[0] != UCODE_UCODE_TYPE) { - printk(KERN_ERR "microcode: error: invalid type field in " - "container file section header\n"); + pr_err("error: invalid type field in container file section header\n"); return NULL; } total_size = (unsigned long) (section_hdr[4] + (section_hdr[5] << 8)); - printk(KERN_DEBUG "microcode: size %u, total_size %u\n", - size, total_size); - if (total_size > size || total_size > UCODE_MAX_SIZE) { - printk(KERN_ERR "microcode: error: size mismatch\n"); + pr_err("error: size mismatch\n"); return NULL; } @@ -218,15 +208,13 @@ static int install_equiv_cpu_table(const u8 *buf) size = buf_pos[2]; if (buf_pos[1] != UCODE_EQUIV_CPU_TABLE_TYPE || !size) { - printk(KERN_ERR "microcode: error: invalid type field in " - "container file section header\n"); + pr_err("error: invalid type field in container file section header\n"); return 0; } equiv_cpu_table = (struct equiv_cpu_entry *) vmalloc(size); if (!equiv_cpu_table) { - printk(KERN_ERR "microcode: failed to allocate " - "equivalent CPU table\n"); + pr_err("failed to allocate equivalent CPU table\n"); return 0; } @@ -259,8 +247,7 @@ generic_load_microcode(int cpu, const u8 *data, size_t size) offset = install_equiv_cpu_table(ucode_ptr); if (!offset) { - printk(KERN_ERR "microcode: failed to create " - "equivalent cpu table\n"); + pr_err("failed to create equivalent cpu table\n"); return UCODE_ERROR; } @@ -291,8 +278,7 @@ generic_load_microcode(int cpu, const u8 *data, size_t size) if (!leftover) { vfree(uci->mc); uci->mc = new_mc; - pr_debug("microcode: CPU%d found a matching microcode " - "update with version 0x%x (current=0x%x)\n", + pr_debug("CPU%d found a matching microcode update with version 0x%x (current=0x%x)\n", cpu, new_rev, uci->cpu_sig.rev); } else { vfree(new_mc); @@ -317,6 +303,12 @@ static enum ucode_state request_microcode_fw(int cpu, struct device *device) return UCODE_NFOUND; } + if (*(u32 *)firmware->data != UCODE_MAGIC) { + pr_err("invalid UCODE_MAGIC (0x%08x)\n", + *(u32 *)firmware->data); + return UCODE_ERROR; + } + ret = generic_load_microcode(cpu, firmware->data, firmware->size); release_firmware(firmware); @@ -327,8 +319,7 @@ static enum ucode_state request_microcode_fw(int cpu, struct device *device) static enum ucode_state request_microcode_user(int cpu, const void __user *buf, size_t size) { - printk(KERN_INFO "microcode: AMD microcode update via " - "/dev/cpu/microcode not supported\n"); + pr_info("AMD microcode update via /dev/cpu/microcode not supported\n"); return UCODE_ERROR; } diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c index 378e9a8f1bf8..cceb5bc3c3c2 100644 --- a/arch/x86/kernel/microcode_core.c +++ b/arch/x86/kernel/microcode_core.c @@ -70,10 +70,12 @@ * Fix sigmatch() macro to handle old CPUs with pf == 0. * Thanks to Stuart Swales for pointing out this bug. */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/platform_device.h> #include <linux/miscdevice.h> #include <linux/capability.h> -#include <linux/smp_lock.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/mutex.h> @@ -201,7 +203,6 @@ static int do_microcode_update(const void __user *buf, size_t size) static int microcode_open(struct inode *unused1, struct file *unused2) { - cycle_kernel_lock(); return capable(CAP_SYS_RAWIO) ? 0 : -EPERM; } @@ -211,7 +212,7 @@ static ssize_t microcode_write(struct file *file, const char __user *buf, ssize_t ret = -EINVAL; if ((len >> PAGE_SHIFT) > totalram_pages) { - pr_err("microcode: too much data (max %ld pages)\n", totalram_pages); + pr_err("too much data (max %ld pages)\n", totalram_pages); return ret; } @@ -246,7 +247,7 @@ static int __init microcode_dev_init(void) error = misc_register(µcode_dev); if (error) { - pr_err("microcode: can't misc_register on minor=%d\n", MICROCODE_MINOR); + pr_err("can't misc_register on minor=%d\n", MICROCODE_MINOR); return error; } @@ -361,7 +362,7 @@ static enum ucode_state microcode_resume_cpu(int cpu) if (!uci->mc) return UCODE_NFOUND; - pr_debug("microcode: CPU%d updated upon resume\n", cpu); + pr_debug("CPU%d updated upon resume\n", cpu); apply_microcode_on_target(cpu); return UCODE_OK; @@ -381,7 +382,7 @@ static enum ucode_state microcode_init_cpu(int cpu) ustate = microcode_ops->request_microcode_fw(cpu, µcode_pdev->dev); if (ustate == UCODE_OK) { - pr_debug("microcode: CPU%d updated upon init\n", cpu); + pr_debug("CPU%d updated upon init\n", cpu); apply_microcode_on_target(cpu); } @@ -408,7 +409,7 @@ static int mc_sysdev_add(struct sys_device *sys_dev) if (!cpu_online(cpu)) return 0; - pr_debug("microcode: CPU%d added\n", cpu); + pr_debug("CPU%d added\n", cpu); err = sysfs_create_group(&sys_dev->kobj, &mc_attr_group); if (err) @@ -427,7 +428,7 @@ static int mc_sysdev_remove(struct sys_device *sys_dev) if (!cpu_online(cpu)) return 0; - pr_debug("microcode: CPU%d removed\n", cpu); + pr_debug("CPU%d removed\n", cpu); microcode_fini_cpu(cpu); sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); return 0; @@ -475,15 +476,15 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) microcode_update_cpu(cpu); case CPU_DOWN_FAILED: case CPU_DOWN_FAILED_FROZEN: - pr_debug("microcode: CPU%d added\n", cpu); + pr_debug("CPU%d added\n", cpu); if (sysfs_create_group(&sys_dev->kobj, &mc_attr_group)) - pr_err("microcode: Failed to create group for CPU%d\n", cpu); + pr_err("Failed to create group for CPU%d\n", cpu); break; case CPU_DOWN_PREPARE: case CPU_DOWN_PREPARE_FROZEN: /* Suspend is in progress, only remove the interface */ sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); - pr_debug("microcode: CPU%d removed\n", cpu); + pr_debug("CPU%d removed\n", cpu); break; case CPU_DEAD: case CPU_UP_CANCELED_FROZEN: @@ -509,7 +510,7 @@ static int __init microcode_init(void) microcode_ops = init_amd_microcode(); if (!microcode_ops) { - pr_err("microcode: no support for this CPU vendor\n"); + pr_err("no support for this CPU vendor\n"); return -ENODEV; } @@ -540,8 +541,7 @@ static int __init microcode_init(void) register_hotcpu_notifier(&mc_cpu_notifier); pr_info("Microcode Update Driver: v" MICROCODE_VERSION - " <tigran@aivazian.fsnet.co.uk>," - " Peter Oruba\n"); + " <tigran@aivazian.fsnet.co.uk>, Peter Oruba\n"); return 0; } diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index 0d334ddd0a96..85a343e28937 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -70,6 +70,9 @@ * Fix sigmatch() macro to handle old CPUs with pf == 0. * Thanks to Stuart Swales for pointing out this bug. */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/firmware.h> #include <linux/uaccess.h> #include <linux/kernel.h> @@ -146,8 +149,7 @@ static int collect_cpu_info(int cpu_num, struct cpu_signature *csig) if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 || cpu_has(c, X86_FEATURE_IA64)) { - printk(KERN_ERR "microcode: CPU%d not a capable Intel " - "processor\n", cpu_num); + pr_err("CPU%d not a capable Intel processor\n", cpu_num); return -1; } @@ -165,8 +167,8 @@ static int collect_cpu_info(int cpu_num, struct cpu_signature *csig) /* get the current revision from MSR 0x8B */ rdmsr(MSR_IA32_UCODE_REV, val[0], csig->rev); - printk(KERN_INFO "microcode: CPU%d sig=0x%x, pf=0x%x, revision=0x%x\n", - cpu_num, csig->sig, csig->pf, csig->rev); + pr_info("CPU%d sig=0x%x, pf=0x%x, revision=0x%x\n", + cpu_num, csig->sig, csig->pf, csig->rev); return 0; } @@ -194,28 +196,24 @@ static int microcode_sanity_check(void *mc) data_size = get_datasize(mc_header); if (data_size + MC_HEADER_SIZE > total_size) { - printk(KERN_ERR "microcode: error! " - "Bad data size in microcode data file\n"); + pr_err("error! Bad data size in microcode data file\n"); return -EINVAL; } if (mc_header->ldrver != 1 || mc_header->hdrver != 1) { - printk(KERN_ERR "microcode: error! " - "Unknown microcode update format\n"); + pr_err("error! Unknown microcode update format\n"); return -EINVAL; } ext_table_size = total_size - (MC_HEADER_SIZE + data_size); if (ext_table_size) { if ((ext_table_size < EXT_HEADER_SIZE) || ((ext_table_size - EXT_HEADER_SIZE) % EXT_SIGNATURE_SIZE)) { - printk(KERN_ERR "microcode: error! " - "Small exttable size in microcode data file\n"); + pr_err("error! Small exttable size in microcode data file\n"); return -EINVAL; } ext_header = mc + MC_HEADER_SIZE + data_size; if (ext_table_size != exttable_size(ext_header)) { - printk(KERN_ERR "microcode: error! " - "Bad exttable size in microcode data file\n"); + pr_err("error! Bad exttable size in microcode data file\n"); return -EFAULT; } ext_sigcount = ext_header->count; @@ -230,8 +228,7 @@ static int microcode_sanity_check(void *mc) while (i--) ext_table_sum += ext_tablep[i]; if (ext_table_sum) { - printk(KERN_WARNING "microcode: aborting, " - "bad extended signature table checksum\n"); + pr_warning("aborting, bad extended signature table checksum\n"); return -EINVAL; } } @@ -242,7 +239,7 @@ static int microcode_sanity_check(void *mc) while (i--) orig_sum += ((int *)mc)[i]; if (orig_sum) { - printk(KERN_ERR "microcode: aborting, bad checksum\n"); + pr_err("aborting, bad checksum\n"); return -EINVAL; } if (!ext_table_size) @@ -255,7 +252,7 @@ static int microcode_sanity_check(void *mc) - (mc_header->sig + mc_header->pf + mc_header->cksum) + (ext_sig->sig + ext_sig->pf + ext_sig->cksum); if (sum) { - printk(KERN_ERR "microcode: aborting, bad checksum\n"); + pr_err("aborting, bad checksum\n"); return -EINVAL; } } @@ -327,13 +324,11 @@ static int apply_microcode(int cpu) rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]); if (val[1] != mc_intel->hdr.rev) { - printk(KERN_ERR "microcode: CPU%d update " - "to revision 0x%x failed\n", - cpu_num, mc_intel->hdr.rev); + pr_err("CPU%d update to revision 0x%x failed\n", + cpu_num, mc_intel->hdr.rev); return -1; } - printk(KERN_INFO "microcode: CPU%d updated to revision " - "0x%x, date = %04x-%02x-%02x \n", + pr_info("CPU%d updated to revision 0x%x, date = %04x-%02x-%02x\n", cpu_num, val[1], mc_intel->hdr.date & 0xffff, mc_intel->hdr.date >> 24, @@ -362,8 +357,7 @@ static enum ucode_state generic_load_microcode(int cpu, void *data, size_t size, mc_size = get_totalsize(&mc_header); if (!mc_size || mc_size > leftover) { - printk(KERN_ERR "microcode: error!" - "Bad data in microcode data file\n"); + pr_err("error! Bad data in microcode data file\n"); break; } @@ -405,9 +399,8 @@ static enum ucode_state generic_load_microcode(int cpu, void *data, size_t size, vfree(uci->mc); uci->mc = (struct microcode_intel *)new_mc; - pr_debug("microcode: CPU%d found a matching microcode update with" - " version 0x%x (current=0x%x)\n", - cpu, new_rev, uci->cpu_sig.rev); + pr_debug("CPU%d found a matching microcode update with version 0x%x (current=0x%x)\n", + cpu, new_rev, uci->cpu_sig.rev); out: return state; } @@ -429,7 +422,7 @@ static enum ucode_state request_microcode_fw(int cpu, struct device *device) c->x86, c->x86_model, c->x86_mask); if (request_firmware(&firmware, name, device)) { - pr_debug("microcode: data file %s load failed\n", name); + pr_debug("data file %s load failed\n", name); return UCODE_NFOUND; } diff --git a/arch/x86/kernel/mmconf-fam10h_64.c b/arch/x86/kernel/mmconf-fam10h_64.c index 712d15fdc416..71825806cd44 100644 --- a/arch/x86/kernel/mmconf-fam10h_64.c +++ b/arch/x86/kernel/mmconf-fam10h_64.c @@ -7,6 +7,8 @@ #include <linux/string.h> #include <linux/pci.h> #include <linux/dmi.h> +#include <linux/range.h> + #include <asm/pci-direct.h> #include <linux/sort.h> #include <asm/io.h> @@ -30,11 +32,6 @@ static struct pci_hostbridge_probe pci_probes[] __cpuinitdata = { { 0xff, 0, PCI_VENDOR_ID_AMD, 0x1200 }, }; -struct range { - u64 start; - u64 end; -}; - static int __cpuinit cmp_range(const void *x1, const void *x2) { const struct range *r1 = x1; diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 5be95ef4ffec..a2c1edd2d3ac 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -359,13 +359,6 @@ static int __init smp_read_mpc(struct mpc_table *mpc, unsigned early) x86_init.mpparse.mpc_record(1); } -#ifdef CONFIG_X86_BIGSMP - generic_bigsmp_probe(); -#endif - - if (apic->setup_apic_routing) - apic->setup_apic_routing(); - if (!num_processors) printk(KERN_ERR "MPTABLE: no processors registered!\n"); return num_processors; @@ -667,36 +660,18 @@ void __init default_get_smp_config(unsigned int early) */ } -static void __init smp_reserve_bootmem(struct mpf_intel *mpf) +static void __init smp_reserve_memory(struct mpf_intel *mpf) { unsigned long size = get_mpc_size(mpf->physptr); -#ifdef CONFIG_X86_32 - /* - * We cannot access to MPC table to compute table size yet, - * as only few megabytes from the bottom is mapped now. - * PC-9800's MPC table places on the very last of physical - * memory; so that simply reserving PAGE_SIZE from mpf->physptr - * yields BUG() in reserve_bootmem. - * also need to make sure physptr is below than max_low_pfn - * we don't need reserve the area above max_low_pfn - */ - unsigned long end = max_low_pfn * PAGE_SIZE; - if (mpf->physptr < end) { - if (mpf->physptr + size > end) - size = end - mpf->physptr; - reserve_bootmem_generic(mpf->physptr, size, BOOTMEM_DEFAULT); - } -#else - reserve_bootmem_generic(mpf->physptr, size, BOOTMEM_DEFAULT); -#endif + reserve_early(mpf->physptr, mpf->physptr+size, "MP-table mpc"); } -static int __init smp_scan_config(unsigned long base, unsigned long length, - unsigned reserve) +static int __init smp_scan_config(unsigned long base, unsigned long length) { unsigned int *bp = phys_to_virt(base); struct mpf_intel *mpf; + unsigned long mem; apic_printk(APIC_VERBOSE, "Scan SMP from %p for %ld bytes.\n", bp, length); @@ -717,12 +692,10 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, printk(KERN_INFO "found SMP MP-table at [%p] %llx\n", mpf, (u64)virt_to_phys(mpf)); - if (!reserve) - return 1; - reserve_bootmem_generic(virt_to_phys(mpf), sizeof(*mpf), - BOOTMEM_DEFAULT); + mem = virt_to_phys(mpf); + reserve_early(mem, mem + sizeof(*mpf), "MP-table mpf"); if (mpf->physptr) - smp_reserve_bootmem(mpf); + smp_reserve_memory(mpf); return 1; } @@ -732,7 +705,7 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, return 0; } -void __init default_find_smp_config(unsigned int reserve) +void __init default_find_smp_config(void) { unsigned int address; @@ -744,9 +717,9 @@ void __init default_find_smp_config(unsigned int reserve) * 2) Scan the top 1K of base RAM * 3) Scan the 64K of bios */ - if (smp_scan_config(0x0, 0x400, reserve) || - smp_scan_config(639 * 0x400, 0x400, reserve) || - smp_scan_config(0xF0000, 0x10000, reserve)) + if (smp_scan_config(0x0, 0x400) || + smp_scan_config(639 * 0x400, 0x400) || + smp_scan_config(0xF0000, 0x10000)) return; /* * If it is an SMP machine we should know now, unless the @@ -767,7 +740,7 @@ void __init default_find_smp_config(unsigned int reserve) address = get_bios_ebda(); if (address) - smp_scan_config(address, 0x400, reserve); + smp_scan_config(address, 0x400); } #ifdef CONFIG_X86_IO_APIC @@ -965,9 +938,6 @@ void __init early_reserve_e820_mpc_new(void) { if (enable_update_mptable && alloc_mptable) { u64 startt = 0; -#ifdef CONFIG_X86_TRAMPOLINE - startt = TRAMPOLINE_BASE; -#endif mpc_new_phys = early_reserve_e820(startt, mpc_new_length, 4); } } diff --git a/arch/x86/kernel/mrst.c b/arch/x86/kernel/mrst.c index 3b7078abc871..0aad8670858e 100644 --- a/arch/x86/kernel/mrst.c +++ b/arch/x86/kernel/mrst.c @@ -10,8 +10,211 @@ * of the License. */ #include <linux/init.h> +#include <linux/kernel.h> +#include <linux/sfi.h> +#include <linux/irq.h> +#include <linux/module.h> #include <asm/setup.h> +#include <asm/mpspec_def.h> +#include <asm/hw_irq.h> +#include <asm/apic.h> +#include <asm/io_apic.h> +#include <asm/mrst.h> +#include <asm/io.h> +#include <asm/i8259.h> +#include <asm/apb_timer.h> + +static u32 sfi_mtimer_usage[SFI_MTMR_MAX_NUM]; +static struct sfi_timer_table_entry sfi_mtimer_array[SFI_MTMR_MAX_NUM]; +int sfi_mtimer_num; + +struct sfi_rtc_table_entry sfi_mrtc_array[SFI_MRTC_MAX]; +EXPORT_SYMBOL_GPL(sfi_mrtc_array); +int sfi_mrtc_num; + +static inline void assign_to_mp_irq(struct mpc_intsrc *m, + struct mpc_intsrc *mp_irq) +{ + memcpy(mp_irq, m, sizeof(struct mpc_intsrc)); +} + +static inline int mp_irq_cmp(struct mpc_intsrc *mp_irq, + struct mpc_intsrc *m) +{ + return memcmp(mp_irq, m, sizeof(struct mpc_intsrc)); +} + +static void save_mp_irq(struct mpc_intsrc *m) +{ + int i; + + for (i = 0; i < mp_irq_entries; i++) { + if (!mp_irq_cmp(&mp_irqs[i], m)) + return; + } + + assign_to_mp_irq(m, &mp_irqs[mp_irq_entries]); + if (++mp_irq_entries == MAX_IRQ_SOURCES) + panic("Max # of irq sources exceeded!!\n"); +} + +/* parse all the mtimer info to a static mtimer array */ +static int __init sfi_parse_mtmr(struct sfi_table_header *table) +{ + struct sfi_table_simple *sb; + struct sfi_timer_table_entry *pentry; + struct mpc_intsrc mp_irq; + int totallen; + + sb = (struct sfi_table_simple *)table; + if (!sfi_mtimer_num) { + sfi_mtimer_num = SFI_GET_NUM_ENTRIES(sb, + struct sfi_timer_table_entry); + pentry = (struct sfi_timer_table_entry *) sb->pentry; + totallen = sfi_mtimer_num * sizeof(*pentry); + memcpy(sfi_mtimer_array, pentry, totallen); + } + + printk(KERN_INFO "SFI: MTIMER info (num = %d):\n", sfi_mtimer_num); + pentry = sfi_mtimer_array; + for (totallen = 0; totallen < sfi_mtimer_num; totallen++, pentry++) { + printk(KERN_INFO "timer[%d]: paddr = 0x%08x, freq = %dHz," + " irq = %d\n", totallen, (u32)pentry->phys_addr, + pentry->freq_hz, pentry->irq); + if (!pentry->irq) + continue; + mp_irq.type = MP_IOAPIC; + mp_irq.irqtype = mp_INT; +/* triggering mode edge bit 2-3, active high polarity bit 0-1 */ + mp_irq.irqflag = 5; + mp_irq.srcbus = 0; + mp_irq.srcbusirq = pentry->irq; /* IRQ */ + mp_irq.dstapic = MP_APIC_ALL; + mp_irq.dstirq = pentry->irq; + save_mp_irq(&mp_irq); + } + + return 0; +} + +struct sfi_timer_table_entry *sfi_get_mtmr(int hint) +{ + int i; + if (hint < sfi_mtimer_num) { + if (!sfi_mtimer_usage[hint]) { + pr_debug("hint taken for timer %d irq %d\n",\ + hint, sfi_mtimer_array[hint].irq); + sfi_mtimer_usage[hint] = 1; + return &sfi_mtimer_array[hint]; + } + } + /* take the first timer available */ + for (i = 0; i < sfi_mtimer_num;) { + if (!sfi_mtimer_usage[i]) { + sfi_mtimer_usage[i] = 1; + return &sfi_mtimer_array[i]; + } + i++; + } + return NULL; +} + +void sfi_free_mtmr(struct sfi_timer_table_entry *mtmr) +{ + int i; + for (i = 0; i < sfi_mtimer_num;) { + if (mtmr->irq == sfi_mtimer_array[i].irq) { + sfi_mtimer_usage[i] = 0; + return; + } + i++; + } +} + +/* parse all the mrtc info to a global mrtc array */ +int __init sfi_parse_mrtc(struct sfi_table_header *table) +{ + struct sfi_table_simple *sb; + struct sfi_rtc_table_entry *pentry; + struct mpc_intsrc mp_irq; + + int totallen; + + sb = (struct sfi_table_simple *)table; + if (!sfi_mrtc_num) { + sfi_mrtc_num = SFI_GET_NUM_ENTRIES(sb, + struct sfi_rtc_table_entry); + pentry = (struct sfi_rtc_table_entry *)sb->pentry; + totallen = sfi_mrtc_num * sizeof(*pentry); + memcpy(sfi_mrtc_array, pentry, totallen); + } + + printk(KERN_INFO "SFI: RTC info (num = %d):\n", sfi_mrtc_num); + pentry = sfi_mrtc_array; + for (totallen = 0; totallen < sfi_mrtc_num; totallen++, pentry++) { + printk(KERN_INFO "RTC[%d]: paddr = 0x%08x, irq = %d\n", + totallen, (u32)pentry->phys_addr, pentry->irq); + mp_irq.type = MP_IOAPIC; + mp_irq.irqtype = mp_INT; + mp_irq.irqflag = 0; + mp_irq.srcbus = 0; + mp_irq.srcbusirq = pentry->irq; /* IRQ */ + mp_irq.dstapic = MP_APIC_ALL; + mp_irq.dstirq = pentry->irq; + save_mp_irq(&mp_irq); + } + return 0; +} + +/* + * the secondary clock in Moorestown can be APBT or LAPIC clock, default to + * APBT but cmdline option can also override it. + */ +static void __cpuinit mrst_setup_secondary_clock(void) +{ + /* restore default lapic clock if disabled by cmdline */ + if (disable_apbt_percpu) + return setup_secondary_APIC_clock(); + apbt_setup_secondary_clock(); +} + +static unsigned long __init mrst_calibrate_tsc(void) +{ + unsigned long flags, fast_calibrate; + + local_irq_save(flags); + fast_calibrate = apbt_quick_calibrate(); + local_irq_restore(flags); + + if (fast_calibrate) + return fast_calibrate; + + return 0; +} + +void __init mrst_time_init(void) +{ + sfi_table_parse(SFI_SIG_MTMR, NULL, NULL, sfi_parse_mtmr); + pre_init_apic_IRQ0(); + apbt_time_init(); +} + +void __init mrst_rtc_init(void) +{ + sfi_table_parse(SFI_SIG_MRTC, NULL, NULL, sfi_parse_mrtc); +} + +/* + * if we use per cpu apb timer, the bootclock already setup. if we use lapic + * timer and one apbt timer for broadcast, we need to set up lapic boot clock. + */ +static void __init mrst_setup_boot_clock(void) +{ + pr_info("%s: per cpu apbt flag %d \n", __func__, disable_apbt_percpu); + if (disable_apbt_percpu) + setup_boot_APIC_clock(); +}; /* * Moorestown specific x86_init function overrides and early setup @@ -21,4 +224,17 @@ void __init x86_mrst_early_setup(void) { x86_init.resources.probe_roms = x86_init_noop; x86_init.resources.reserve_resources = x86_init_noop; + + x86_init.timers.timer_init = mrst_time_init; + x86_init.timers.setup_percpu_clockev = mrst_setup_boot_clock; + + x86_init.irqs.pre_vector_init = x86_init_noop; + + x86_cpuinit.setup_percpu_clockev = mrst_setup_secondary_clock; + + x86_platform.calibrate_tsc = mrst_calibrate_tsc; + x86_init.pci.init = pci_mrst_init; + x86_init.pci.fixup_irqs = x86_init_noop; + + legacy_pic = &null_legacy_pic; } diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c index 6a3cefc7dda1..206735ac8cbd 100644 --- a/arch/x86/kernel/msr.c +++ b/arch/x86/kernel/msr.c @@ -172,23 +172,18 @@ static long msr_ioctl(struct file *file, unsigned int ioc, unsigned long arg) static int msr_open(struct inode *inode, struct file *file) { - unsigned int cpu = iminor(file->f_path.dentry->d_inode); - struct cpuinfo_x86 *c = &cpu_data(cpu); - int ret = 0; + unsigned int cpu; + struct cpuinfo_x86 *c; - lock_kernel(); cpu = iminor(file->f_path.dentry->d_inode); + if (cpu >= nr_cpu_ids || !cpu_online(cpu)) + return -ENXIO; /* No such CPU */ - if (cpu >= nr_cpu_ids || !cpu_online(cpu)) { - ret = -ENXIO; /* No such CPU */ - goto out; - } c = &cpu_data(cpu); if (!cpu_has(c, X86_FEATURE_MSR)) - ret = -EIO; /* MSR not supported */ -out: - unlock_kernel(); - return ret; + return -EIO; /* MSR not supported */ + + return 0; } /* @@ -251,7 +246,7 @@ static int __init msr_init(void) int i, err = 0; i = 0; - if (register_chrdev(MSR_MAJOR, "cpu/msr", &msr_fops)) { + if (__register_chrdev(MSR_MAJOR, 0, NR_CPUS, "cpu/msr", &msr_fops)) { printk(KERN_ERR "msr: unable to get major %d for msr\n", MSR_MAJOR); err = -EBUSY; @@ -279,7 +274,7 @@ out_class: msr_device_destroy(i); class_destroy(msr_class); out_chrdev: - unregister_chrdev(MSR_MAJOR, "cpu/msr"); + __unregister_chrdev(MSR_MAJOR, 0, NR_CPUS, "cpu/msr"); out: return err; } @@ -290,7 +285,7 @@ static void __exit msr_exit(void) for_each_online_cpu(cpu) msr_device_destroy(cpu); class_destroy(msr_class); - unregister_chrdev(MSR_MAJOR, "cpu/msr"); + __unregister_chrdev(MSR_MAJOR, 0, NR_CPUS, "cpu/msr"); unregister_hotcpu_notifier(&msr_class_cpu_notifier); } diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c index 4006c522adc7..8297160c41b3 100644 --- a/arch/x86/kernel/olpc.c +++ b/arch/x86/kernel/olpc.c @@ -17,7 +17,9 @@ #include <linux/spinlock.h> #include <linux/io.h> #include <linux/string.h> + #include <asm/geode.h> +#include <asm/setup.h> #include <asm/olpc.h> #ifdef CONFIG_OPEN_FIRMWARE @@ -212,7 +214,7 @@ static int __init olpc_init(void) unsigned char *romsig; /* The ioremap check is dangerous; limit what we run it on */ - if (!is_geode() || geode_has_vsa2()) + if (!is_geode() || cs5535_has_vsa2()) return 0; spin_lock_init(&ec_lock); @@ -243,9 +245,11 @@ static int __init olpc_init(void) olpc_ec_cmd(EC_FIRMWARE_REV, NULL, 0, (unsigned char *) &olpc_platform_info.ecver, 1); - /* check to see if the VSA exists */ - if (geode_has_vsa2()) - olpc_platform_info.flags |= OLPC_F_VSA; +#ifdef CONFIG_PCI_OLPC + /* If the VSA exists let it emulate PCI, if not emulate in kernel */ + if (!cs5535_has_vsa2()) + x86_init.pci.arch_init = pci_olpc_init; +#endif printk(KERN_INFO "OLPC board revision %s%X (EC=%x)\n", ((olpc_platform_info.boardrev & 0xf) < 8) ? "pre" : "", diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c index 3a7c5a44082e..676b8c77a976 100644 --- a/arch/x86/kernel/paravirt-spinlocks.c +++ b/arch/x86/kernel/paravirt-spinlocks.c @@ -8,9 +8,9 @@ #include <asm/paravirt.h> static inline void -default_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags) +default_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { - __raw_spin_lock(lock); + arch_spin_lock(lock); } struct pv_lock_ops pv_lock_ops = { diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 1b1739d16310..1db183ed7c01 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -428,10 +428,6 @@ struct pv_mmu_ops pv_mmu_ops = { .ptep_modify_prot_start = __ptep_modify_prot_start, .ptep_modify_prot_commit = __ptep_modify_prot_commit, -#ifdef CONFIG_HIGHPTE - .kmap_atomic_pte = kmap_atomic, -#endif - #if PAGETABLE_LEVELS >= 3 #ifdef CONFIG_X86_PAE .set_pte_atomic = native_set_pte_atomic, diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c index 971a3bec47a8..fb99f7edb341 100644 --- a/arch/x86/kernel/pci-calgary_64.c +++ b/arch/x86/kernel/pci-calgary_64.c @@ -31,7 +31,7 @@ #include <linux/string.h> #include <linux/crash_dump.h> #include <linux/dma-mapping.h> -#include <linux/bitops.h> +#include <linux/bitmap.h> #include <linux/pci_ids.h> #include <linux/pci.h> #include <linux/delay.h> @@ -46,6 +46,7 @@ #include <asm/dma.h> #include <asm/rio.h> #include <asm/bios_ebda.h> +#include <asm/x86_init.h> #ifdef CONFIG_CALGARY_IOMMU_ENABLED_BY_DEFAULT int use_calgary __read_mostly = 1; @@ -211,7 +212,7 @@ static void iommu_range_reserve(struct iommu_table *tbl, spin_lock_irqsave(&tbl->it_lock, flags); - iommu_area_reserve(tbl->it_map, index, npages); + bitmap_set(tbl->it_map, index, npages); spin_unlock_irqrestore(&tbl->it_lock, flags); } @@ -244,7 +245,7 @@ static unsigned long iommu_range_alloc(struct device *dev, if (panic_on_overflow) panic("Calgary: fix the allocator.\n"); else - return bad_dma_address; + return DMA_ERROR_CODE; } } @@ -260,12 +261,15 @@ static dma_addr_t iommu_alloc(struct device *dev, struct iommu_table *tbl, void *vaddr, unsigned int npages, int direction) { unsigned long entry; - dma_addr_t ret = bad_dma_address; + dma_addr_t ret; entry = iommu_range_alloc(dev, tbl, npages); - if (unlikely(entry == bad_dma_address)) - goto error; + if (unlikely(entry == DMA_ERROR_CODE)) { + printk(KERN_WARNING "Calgary: failed to allocate %u pages in " + "iommu %p\n", npages, tbl); + return DMA_ERROR_CODE; + } /* set the return dma address */ ret = (entry << PAGE_SHIFT) | ((unsigned long)vaddr & ~PAGE_MASK); @@ -273,13 +277,7 @@ static dma_addr_t iommu_alloc(struct device *dev, struct iommu_table *tbl, /* put the TCEs in the HW table */ tce_build(tbl, entry, npages, (unsigned long)vaddr & PAGE_MASK, direction); - return ret; - -error: - printk(KERN_WARNING "Calgary: failed to allocate %u pages in " - "iommu %p\n", npages, tbl); - return bad_dma_address; } static void iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr, @@ -290,8 +288,8 @@ static void iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr, unsigned long flags; /* were we called with bad_dma_address? */ - badend = bad_dma_address + (EMERGENCY_PAGES * PAGE_SIZE); - if (unlikely((dma_addr >= bad_dma_address) && (dma_addr < badend))) { + badend = DMA_ERROR_CODE + (EMERGENCY_PAGES * PAGE_SIZE); + if (unlikely((dma_addr >= DMA_ERROR_CODE) && (dma_addr < badend))) { WARN(1, KERN_ERR "Calgary: driver tried unmapping bad DMA " "address 0x%Lx\n", dma_addr); return; @@ -305,7 +303,7 @@ static void iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr, spin_lock_irqsave(&tbl->it_lock, flags); - iommu_area_free(tbl->it_map, entry, npages); + bitmap_clear(tbl->it_map, entry, npages); spin_unlock_irqrestore(&tbl->it_lock, flags); } @@ -318,13 +316,15 @@ static inline struct iommu_table *find_iommu_table(struct device *dev) pdev = to_pci_dev(dev); + /* search up the device tree for an iommu */ pbus = pdev->bus; - - /* is the device behind a bridge? Look for the root bus */ - while (pbus->parent) + do { + tbl = pci_iommu(pbus); + if (tbl && tbl->it_busno == pbus->number) + break; + tbl = NULL; pbus = pbus->parent; - - tbl = pci_iommu(pbus); + } while (pbus); BUG_ON(tbl && (tbl->it_busno != pbus->number)); @@ -373,7 +373,7 @@ static int calgary_map_sg(struct device *dev, struct scatterlist *sg, npages = iommu_num_pages(vaddr, s->length, PAGE_SIZE); entry = iommu_range_alloc(dev, tbl, npages); - if (entry == bad_dma_address) { + if (entry == DMA_ERROR_CODE) { /* makes sure unmap knows to stop */ s->dma_length = 0; goto error; @@ -391,7 +391,7 @@ static int calgary_map_sg(struct device *dev, struct scatterlist *sg, error: calgary_unmap_sg(dev, sg, nelems, dir, NULL); for_each_sg(sg, s, nelems, i) { - sg->dma_address = bad_dma_address; + sg->dma_address = DMA_ERROR_CODE; sg->dma_length = 0; } return 0; @@ -446,7 +446,7 @@ static void* calgary_alloc_coherent(struct device *dev, size_t size, /* set up tces to cover the allocated range */ mapping = iommu_alloc(dev, tbl, ret, npages, DMA_BIDIRECTIONAL); - if (mapping == bad_dma_address) + if (mapping == DMA_ERROR_CODE) goto free; *dma_handle = mapping; return ret; @@ -727,7 +727,7 @@ static void __init calgary_reserve_regions(struct pci_dev *dev) struct iommu_table *tbl = pci_iommu(dev->bus); /* reserve EMERGENCY_PAGES from bad_dma_address and up */ - iommu_range_reserve(tbl, bad_dma_address, EMERGENCY_PAGES); + iommu_range_reserve(tbl, DMA_ERROR_CODE, EMERGENCY_PAGES); /* avoid the BIOS/VGA first 640KB-1MB region */ /* for CalIOC2 - avoid the entire first MB */ @@ -1309,7 +1309,7 @@ static void calgary_init_bitmap_from_tce_table(struct iommu_table *tbl) /* * get_tce_space_from_tar(): * Function for kdump case. Get the tce tables from first kernel - * by reading the contents of the base adress register of calgary iommu + * by reading the contents of the base address register of calgary iommu */ static void __init get_tce_space_from_tar(void) { @@ -1344,6 +1344,23 @@ static void __init get_tce_space_from_tar(void) return; } +static int __init calgary_iommu_init(void) +{ + int ret; + + /* ok, we're trying to use Calgary - let's roll */ + printk(KERN_INFO "PCI-DMA: Using Calgary IOMMU\n"); + + ret = calgary_init(); + if (ret) { + printk(KERN_ERR "PCI-DMA: Calgary init failed %d, " + "falling back to no_iommu\n", ret); + return ret; + } + + return 0; +} + void __init detect_calgary(void) { int bus; @@ -1357,7 +1374,7 @@ void __init detect_calgary(void) * if the user specified iommu=off or iommu=soft or we found * another HW IOMMU already, bail out. */ - if (swiotlb || no_iommu || iommu_detected) + if (no_iommu || iommu_detected) return; if (!use_calgary) @@ -1442,9 +1459,7 @@ void __init detect_calgary(void) printk(KERN_INFO "PCI-DMA: Calgary TCE table spec is %d\n", specified_table_size); - /* swiotlb for devices that aren't behind the Calgary. */ - if (max_pfn > MAX_DMA32_PFN) - swiotlb = 1; + x86_init.iommu.iommu_init = calgary_iommu_init; } return; @@ -1457,35 +1472,6 @@ cleanup: } } -int __init calgary_iommu_init(void) -{ - int ret; - - if (no_iommu || (swiotlb && !calgary_detected)) - return -ENODEV; - - if (!calgary_detected) - return -ENODEV; - - /* ok, we're trying to use Calgary - let's roll */ - printk(KERN_INFO "PCI-DMA: Using Calgary IOMMU\n"); - - ret = calgary_init(); - if (ret) { - printk(KERN_ERR "PCI-DMA: Calgary init failed %d, " - "falling back to no_iommu\n", ret); - return ret; - } - - force_iommu = 1; - bad_dma_address = 0x0; - /* dma_ops is set to swiotlb or nommu */ - if (!dma_ops) - dma_ops = &nommu_dma_ops; - - return 0; -} - static int __init calgary_parse_options(char *p) { unsigned int bridge; diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 64b838eac18c..a4ac764a6880 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -11,10 +11,11 @@ #include <asm/gart.h> #include <asm/calgary.h> #include <asm/amd_iommu.h> +#include <asm/x86_init.h> static int forbid_dac __read_mostly; -struct dma_map_ops *dma_ops; +struct dma_map_ops *dma_ops = &nommu_dma_ops; EXPORT_SYMBOL(dma_ops); static int iommu_sac_force __read_mostly; @@ -35,22 +36,17 @@ int iommu_detected __read_mostly = 0; /* * This variable becomes 1 if iommu=pt is passed on the kernel command line. - * If this variable is 1, IOMMU implementations do no DMA ranslation for + * If this variable is 1, IOMMU implementations do no DMA translation for * devices and allow every device to access to whole physical memory. This is - * useful if a user want to use an IOMMU only for KVM device assignment to + * useful if a user wants to use an IOMMU only for KVM device assignment to * guests and not for driver dma translation. */ int iommu_pass_through __read_mostly; -dma_addr_t bad_dma_address __read_mostly = 0; -EXPORT_SYMBOL(bad_dma_address); - -/* Dummy device used for NULL arguments (normally ISA). Better would - be probably a smaller DMA mask, but this is bug-to-bug compatible - to older i386. */ +/* Dummy device used for NULL arguments (normally ISA). */ struct device x86_dma_fallback_dev = { .init_name = "fallback device", - .coherent_dma_mask = DMA_BIT_MASK(32), + .coherent_dma_mask = ISA_DMA_BIT_MASK, .dma_mask = &x86_dma_fallback_dev.coherent_dma_mask, }; EXPORT_SYMBOL(x86_dma_fallback_dev); @@ -69,7 +65,7 @@ int dma_set_mask(struct device *dev, u64 mask) } EXPORT_SYMBOL(dma_set_mask); -#ifdef CONFIG_X86_64 +#if defined(CONFIG_X86_64) && !defined(CONFIG_NUMA) static __initdata void *dma32_bootmem_ptr; static unsigned long dma32_bootmem_size __initdata = (128ULL<<20); @@ -120,27 +116,33 @@ static void __init dma32_free_bootmem(void) dma32_bootmem_ptr = NULL; dma32_bootmem_size = 0; } +#else +void __init dma32_reserve_bootmem(void) +{ +} +static void __init dma32_free_bootmem(void) +{ +} + #endif void __init pci_iommu_alloc(void) { -#ifdef CONFIG_X86_64 /* free the range so iommu could get some range less than 4G */ dma32_free_bootmem(); -#endif - /* - * The order of these functions is important for - * fall-back/fail-over reasons - */ + if (pci_swiotlb_detect()) + goto out; + gart_iommu_hole_init(); detect_calgary(); detect_intel_iommu(); + /* needs to be called after gart_iommu_hole_init */ amd_iommu_detect(); - +out: pci_swiotlb_init(); } @@ -216,7 +218,7 @@ static __init int iommu_setup(char *p) if (!strncmp(p, "allowdac", 8)) forbid_dac = 0; if (!strncmp(p, "nodac", 5)) - forbid_dac = -1; + forbid_dac = 1; if (!strncmp(p, "usedac", 6)) { forbid_dac = -1; return 1; @@ -291,27 +293,19 @@ static int __init pci_iommu_init(void) #ifdef CONFIG_PCI dma_debug_add_bus(&pci_bus_type); #endif + x86_init.iommu.iommu_init(); - calgary_iommu_init(); - - intel_iommu_init(); - - amd_iommu_init(); + if (swiotlb) { + printk(KERN_INFO "PCI-DMA: " + "Using software bounce buffering for IO (SWIOTLB)\n"); + swiotlb_print_info(); + } else + swiotlb_free(); - gart_iommu_init(); - - no_iommu_init(); return 0; } - -void pci_iommu_shutdown(void) -{ - gart_iommu_shutdown(); - - amd_iommu_shutdown(); -} /* Must execute after PCI subsystem */ -fs_initcall(pci_iommu_init); +rootfs_initcall(pci_iommu_init); #ifdef CONFIG_PCI /* Many VIA bridges seem to corrupt data for DAC. Disable it here */ diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 98a827ee9ed7..f3af115a573a 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -16,13 +16,14 @@ #include <linux/agp_backend.h> #include <linux/init.h> #include <linux/mm.h> +#include <linux/sched.h> #include <linux/string.h> #include <linux/spinlock.h> #include <linux/pci.h> #include <linux/module.h> #include <linux/topology.h> #include <linux/interrupt.h> -#include <linux/bitops.h> +#include <linux/bitmap.h> #include <linux/kdebug.h> #include <linux/scatterlist.h> #include <linux/iommu-helper.h> @@ -38,6 +39,7 @@ #include <asm/swiotlb.h> #include <asm/dma.h> #include <asm/k8.h> +#include <asm/x86_init.h> static unsigned long iommu_bus_base; /* GART remapping area (physical) */ static unsigned long iommu_size; /* size of remapping area bytes */ @@ -45,6 +47,8 @@ static unsigned long iommu_pages; /* .. and in pages */ static u32 *iommu_gatt_base; /* Remapping table */ +static dma_addr_t bad_dma_addr; + /* * If this is disabled the IOMMU will use an optimized flushing strategy * of only flushing when an mapping is reused. With it true the GART is @@ -91,7 +95,7 @@ static unsigned long alloc_iommu(struct device *dev, int size, base_index = ALIGN(iommu_bus_base & dma_get_seg_boundary(dev), PAGE_SIZE) >> PAGE_SHIFT; - boundary_size = ALIGN((unsigned long long)dma_get_seg_boundary(dev) + 1, + boundary_size = ALIGN((u64)dma_get_seg_boundary(dev) + 1, PAGE_SIZE) >> PAGE_SHIFT; spin_lock_irqsave(&iommu_bitmap_lock, flags); @@ -122,7 +126,7 @@ static void free_iommu(unsigned long offset, int size) unsigned long flags; spin_lock_irqsave(&iommu_bitmap_lock, flags); - iommu_area_free(iommu_gart_bitmap, offset, size); + bitmap_clear(iommu_gart_bitmap, offset, size); if (offset >= next_bit) next_bit = offset + size; spin_unlock_irqrestore(&iommu_bitmap_lock, flags); @@ -215,7 +219,7 @@ static dma_addr_t dma_map_area(struct device *dev, dma_addr_t phys_mem, if (panic_on_overflow) panic("dma_map_area overflow %lu bytes\n", size); iommu_full(dev, size, dir); - return bad_dma_address; + return bad_dma_addr; } for (i = 0; i < npages; i++) { @@ -293,7 +297,7 @@ static int dma_map_sg_nonforce(struct device *dev, struct scatterlist *sg, int i; #ifdef CONFIG_IOMMU_DEBUG - printk(KERN_DEBUG "dma_map_sg overflow\n"); + pr_debug("dma_map_sg overflow\n"); #endif for_each_sg(sg, s, nents, i) { @@ -301,7 +305,7 @@ static int dma_map_sg_nonforce(struct device *dev, struct scatterlist *sg, if (nonforced_iommu(dev, addr, s->length)) { addr = dma_map_area(dev, addr, s->length, dir, 0); - if (addr == bad_dma_address) { + if (addr == bad_dma_addr) { if (i > 0) gart_unmap_sg(dev, sg, i, dir, NULL); nents = 0; @@ -388,12 +392,14 @@ static int gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, if (!dev) dev = &x86_dma_fallback_dev; - out = 0; - start = 0; - start_sg = sgmap = sg; - seg_size = 0; - max_seg_size = dma_get_max_seg_size(dev); - ps = NULL; /* shut up gcc */ + out = 0; + start = 0; + start_sg = sg; + sgmap = sg; + seg_size = 0; + max_seg_size = dma_get_max_seg_size(dev); + ps = NULL; /* shut up gcc */ + for_each_sg(sg, s, nents, i) { dma_addr_t addr = sg_phys(s); @@ -416,11 +422,12 @@ static int gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, sgmap, pages, need) < 0) goto error; out++; - seg_size = 0; - sgmap = sg_next(sgmap); - pages = 0; - start = i; - start_sg = s; + + seg_size = 0; + sgmap = sg_next(sgmap); + pages = 0; + start = i; + start_sg = s; } } @@ -454,7 +461,7 @@ error: iommu_full(dev, pages << PAGE_SHIFT, dir); for_each_sg(sg, s, nents, i) - s->dma_address = bad_dma_address; + s->dma_address = bad_dma_addr; return 0; } @@ -478,7 +485,7 @@ gart_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_addr, DMA_BIDIRECTIONAL, align_mask); flush_gart(); - if (paddr != bad_dma_address) { + if (paddr != bad_dma_addr) { *dma_addr = paddr; return page_address(page); } @@ -498,6 +505,11 @@ gart_free_coherent(struct device *dev, size_t size, void *vaddr, free_pages((unsigned long)vaddr, get_order(size)); } +static int gart_mapping_error(struct device *dev, dma_addr_t dma_addr) +{ + return (dma_addr == bad_dma_addr); +} + static int no_agp; static __init unsigned long check_iommu_size(unsigned long aper, u64 aper_size) @@ -514,7 +526,7 @@ static __init unsigned long check_iommu_size(unsigned long aper, u64 aper_size) iommu_size -= round_up(a, PMD_PAGE_SIZE) - a; if (iommu_size < 64*1024*1024) { - printk(KERN_WARNING + pr_warning( "PCI-DMA: Warning: Small IOMMU %luMB." " Consider increasing the AGP aperture in BIOS\n", iommu_size >> 20); @@ -569,28 +581,32 @@ void set_up_gart_resume(u32 aper_order, u32 aper_alloc) aperture_alloc = aper_alloc; } -static int gart_resume(struct sys_device *dev) +static void gart_fixup_northbridges(struct sys_device *dev) { - printk(KERN_INFO "PCI-DMA: Resuming GART IOMMU\n"); + int i; - if (fix_up_north_bridges) { - int i; + if (!fix_up_north_bridges) + return; - printk(KERN_INFO "PCI-DMA: Restoring GART aperture settings\n"); + 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 < num_k8_northbridges; i++) { + struct pci_dev *dev = k8_northbridges[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); - pci_write_config_dword(dev, AMD64_GARTAPERTUREBASE, - aperture_alloc >> 25); - } + /* + * 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); + pci_write_config_dword(dev, AMD64_GARTAPERTUREBASE, aperture_alloc >> 25); } +} + +static int gart_resume(struct sys_device *dev) +{ + pr_info("PCI-DMA: Resuming GART IOMMU\n"); + + gart_fixup_northbridges(dev); enable_gart_translations(); @@ -603,15 +619,14 @@ static int gart_suspend(struct sys_device *dev, pm_message_t state) } static struct sysdev_class gart_sysdev_class = { - .name = "gart", - .suspend = gart_suspend, - .resume = gart_resume, + .name = "gart", + .suspend = gart_suspend, + .resume = gart_resume, }; static struct sys_device device_gart = { - .id = 0, - .cls = &gart_sysdev_class, + .cls = &gart_sysdev_class, }; /* @@ -626,7 +641,8 @@ static __init int init_k8_gatt(struct agp_kern_info *info) void *gatt; int i, error; - printk(KERN_INFO "PCI-DMA: Disabling AGP.\n"); + pr_info("PCI-DMA: Disabling AGP.\n"); + aper_size = aper_base = info->aper_size = 0; dev = NULL; for (i = 0; i < num_k8_northbridges; i++) { @@ -644,6 +660,7 @@ static __init int init_k8_gatt(struct agp_kern_info *info) } if (!aper_base) goto nommu; + info->aper_base = aper_base; info->aper_size = aper_size >> 20; @@ -666,14 +683,14 @@ static __init int init_k8_gatt(struct agp_kern_info *info) flush_gart(); - printk(KERN_INFO "PCI-DMA: aperture base @ %x size %u KB\n", + pr_info("PCI-DMA: aperture base @ %x size %u KB\n", aper_base, aper_size>>10); return 0; nommu: /* Should not happen anymore */ - printk(KERN_WARNING "PCI-DMA: More than 4GB of RAM and no IOMMU\n" + pr_warning("PCI-DMA: More than 4GB of RAM and no IOMMU\n" "falling back to iommu=soft.\n"); return -1; } @@ -685,14 +702,16 @@ static struct dma_map_ops gart_dma_ops = { .unmap_page = gart_unmap_page, .alloc_coherent = gart_alloc_coherent, .free_coherent = gart_free_coherent, + .mapping_error = gart_mapping_error, }; -void gart_iommu_shutdown(void) +static void gart_iommu_shutdown(void) { struct pci_dev *dev; int i; - if (no_agp && (dma_ops != &gart_dma_ops)) + /* don't shutdown it if there is AGP installed */ + if (!no_agp) return; for (i = 0; i < num_k8_northbridges; i++) { @@ -707,7 +726,7 @@ void gart_iommu_shutdown(void) } } -void __init gart_iommu_init(void) +int __init gart_iommu_init(void) { struct agp_kern_info info; unsigned long iommu_start; @@ -716,8 +735,8 @@ void __init gart_iommu_init(void) unsigned long scratch; long i; - if (cache_k8_northbridges() < 0 || num_k8_northbridges == 0) - return; + if (num_k8_northbridges == 0) + return 0; #ifndef CONFIG_AGP_AMD64 no_agp = 1; @@ -729,35 +748,28 @@ void __init gart_iommu_init(void) (agp_copy_info(agp_bridge, &info) < 0); #endif - if (swiotlb) - return; - - /* Did we detect a different HW IOMMU? */ - if (iommu_detected && !gart_iommu_aperture) - return; - if (no_iommu || (!force_iommu && max_pfn <= MAX_DMA32_PFN) || !gart_iommu_aperture || (no_agp && init_k8_gatt(&info) < 0)) { if (max_pfn > MAX_DMA32_PFN) { - printk(KERN_WARNING "More than 4GB of memory " - "but GART IOMMU not available.\n"); - printk(KERN_WARNING "falling back to iommu=soft.\n"); + pr_warning("More than 4GB of memory but GART IOMMU not available.\n"); + pr_warning("falling back to iommu=soft.\n"); } - return; + return 0; } /* need to map that range */ - aper_size = info.aper_size << 20; - aper_base = info.aper_base; - end_pfn = (aper_base>>PAGE_SHIFT) + (aper_size>>PAGE_SHIFT); + aper_size = info.aper_size << 20; + aper_base = info.aper_base; + end_pfn = (aper_base>>PAGE_SHIFT) + (aper_size>>PAGE_SHIFT); + if (end_pfn > max_low_pfn_mapped) { start_pfn = (aper_base>>PAGE_SHIFT); init_memory_mapping(start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT); } - printk(KERN_INFO "PCI-DMA: using GART IOMMU.\n"); + pr_info("PCI-DMA: using GART IOMMU.\n"); iommu_size = check_iommu_size(info.aper_base, aper_size); iommu_pages = iommu_size >> PAGE_SHIFT; @@ -772,8 +784,7 @@ void __init gart_iommu_init(void) ret = dma_debug_resize_entries(iommu_pages); if (ret) - printk(KERN_DEBUG - "PCI-DMA: Cannot trace all the entries\n"); + pr_debug("PCI-DMA: Cannot trace all the entries\n"); } #endif @@ -781,17 +792,16 @@ void __init gart_iommu_init(void) * Out of IOMMU space handling. * Reserve some invalid pages at the beginning of the GART. */ - iommu_area_reserve(iommu_gart_bitmap, 0, EMERGENCY_PAGES); + bitmap_set(iommu_gart_bitmap, 0, EMERGENCY_PAGES); - agp_memory_reserved = iommu_size; - printk(KERN_INFO - "PCI-DMA: Reserving %luMB of IOMMU area in the AGP aperture\n", + pr_info("PCI-DMA: Reserving %luMB of IOMMU area in the AGP aperture\n", iommu_size >> 20); - iommu_start = aper_size - iommu_size; - iommu_bus_base = info.aper_base + iommu_start; - bad_dma_address = iommu_bus_base; - iommu_gatt_base = agp_gatt_table + (iommu_start>>PAGE_SHIFT); + agp_memory_reserved = iommu_size; + iommu_start = aper_size - iommu_size; + iommu_bus_base = info.aper_base + iommu_start; + bad_dma_addr = iommu_bus_base; + iommu_gatt_base = agp_gatt_table + (iommu_start>>PAGE_SHIFT); /* * Unmap the IOMMU part of the GART. The alias of the page is @@ -813,7 +823,7 @@ void __init gart_iommu_init(void) * the pages as Not-Present: */ wbinvd(); - + /* * Now all caches are flushed and we can safely enable * GART hardware. Doing it early leaves the possibility @@ -837,6 +847,10 @@ void __init gart_iommu_init(void) flush_gart(); dma_ops = &gart_dma_ops; + x86_platform.iommu_shutdown = gart_iommu_shutdown; + swiotlb = 0; + + return 0; } void __init gart_parse_options(char *p) @@ -855,7 +869,7 @@ void __init gart_parse_options(char *p) #endif if (isdigit(*p) && get_option(&p, &arg)) iommu_size = arg; - if (!strncmp(p, "fullflush", 8)) + if (!strncmp(p, "fullflush", 9)) iommu_fullflush = 1; if (!strncmp(p, "nofullflush", 11)) iommu_fullflush = 0; diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c index a3933d4330cd..22be12b60a8f 100644 --- a/arch/x86/kernel/pci-nommu.c +++ b/arch/x86/kernel/pci-nommu.c @@ -33,7 +33,7 @@ static dma_addr_t nommu_map_page(struct device *dev, struct page *page, dma_addr_t bus = page_to_phys(page) + offset; WARN_ON(size == 0); if (!check_addr("map_single", dev, bus, size)) - return bad_dma_address; + return DMA_ERROR_CODE; flush_write_buffers(); return bus; } @@ -103,12 +103,3 @@ struct dma_map_ops nommu_dma_ops = { .sync_sg_for_device = nommu_sync_sg_for_device, .is_phys = 1, }; - -void __init no_iommu_init(void) -{ - if (dma_ops) - return; - - force_iommu = 0; /* no HW IOMMU */ - dma_ops = &nommu_dma_ops; -} diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c index aaa6b7839f1e..7d2829dde20e 100644 --- a/arch/x86/kernel/pci-swiotlb.c +++ b/arch/x86/kernel/pci-swiotlb.c @@ -42,18 +42,31 @@ static struct dma_map_ops swiotlb_dma_ops = { .dma_supported = NULL, }; -void __init pci_swiotlb_init(void) +/* + * pci_swiotlb_detect - 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 use_swiotlb = swiotlb | swiotlb_force; + /* don't initialize swiotlb if iommu=off (no_iommu=1) */ #ifdef CONFIG_X86_64 - if ((!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN)) + if (!no_iommu && max_pfn > MAX_DMA32_PFN) swiotlb = 1; #endif if (swiotlb_force) swiotlb = 1; + + return use_swiotlb; +} + +void __init pci_swiotlb_init(void) +{ if (swiotlb) { - printk(KERN_INFO "PCI-DMA: Using software bounce buffering for IO (SWIOTLB)\n"); - swiotlb_init(); + swiotlb_init(0); dma_ops = &swiotlb_dma_ops; } } diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 5284cd2b5776..ad9540676fcc 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -9,7 +9,11 @@ #include <linux/pm.h> #include <linux/clockchips.h> #include <linux/random.h> +#include <linux/user-return-notifier.h> +#include <linux/dmi.h> +#include <linux/utsname.h> #include <trace/events/power.h> +#include <linux/hw_breakpoint.h> #include <asm/system.h> #include <asm/apic.h> #include <asm/syscalls.h> @@ -17,6 +21,7 @@ #include <asm/uaccess.h> #include <asm/i387.h> #include <asm/ds.h> +#include <asm/debugreg.h> unsigned long idle_halt; EXPORT_SYMBOL(idle_halt); @@ -87,30 +92,37 @@ void exit_thread(void) } } -void flush_thread(void) +void show_regs(struct pt_regs *regs) { - struct task_struct *tsk = current; + show_registers(regs); + show_trace(NULL, regs, (unsigned long *)kernel_stack_pointer(regs), + regs->bp); +} -#ifdef CONFIG_X86_64 - if (test_tsk_thread_flag(tsk, TIF_ABI_PENDING)) { - clear_tsk_thread_flag(tsk, TIF_ABI_PENDING); - if (test_tsk_thread_flag(tsk, TIF_IA32)) { - clear_tsk_thread_flag(tsk, TIF_IA32); - } else { - set_tsk_thread_flag(tsk, TIF_IA32); - current_thread_info()->status |= TS_COMPAT; - } - } -#endif +void show_regs_common(void) +{ + const char *board, *product; + + board = dmi_get_system_info(DMI_BOARD_NAME); + if (!board) + board = ""; + product = dmi_get_system_info(DMI_PRODUCT_NAME); + if (!product) + product = ""; - clear_tsk_thread_flag(tsk, TIF_DEBUG); + printk(KERN_CONT "\n"); + printk(KERN_DEFAULT "Pid: %d, comm: %.20s %s %s %.*s %s/%s\n", + current->pid, current->comm, print_tainted(), + init_utsname()->release, + (int)strcspn(init_utsname()->version, " "), + init_utsname()->version, board, product); +} + +void flush_thread(void) +{ + struct task_struct *tsk = current; - tsk->thread.debugreg0 = 0; - tsk->thread.debugreg1 = 0; - tsk->thread.debugreg2 = 0; - tsk->thread.debugreg3 = 0; - tsk->thread.debugreg6 = 0; - tsk->thread.debugreg7 = 0; + flush_ptrace_hw_breakpoint(tsk); memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array)); /* * Forget coprocessor state.. @@ -192,16 +204,6 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, else if (next->debugctlmsr != prev->debugctlmsr) update_debugctlmsr(next->debugctlmsr); - if (test_tsk_thread_flag(next_p, TIF_DEBUG)) { - set_debugreg(next->debugreg0, 0); - set_debugreg(next->debugreg1, 1); - set_debugreg(next->debugreg2, 2); - set_debugreg(next->debugreg3, 3); - /* no 4 and 5 */ - set_debugreg(next->debugreg6, 6); - set_debugreg(next->debugreg7, 7); - } - if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^ test_tsk_thread_flag(next_p, TIF_NOTSC)) { /* prev and next are different */ @@ -224,6 +226,7 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, */ memset(tss->io_bitmap, 0xff, prev->io_bitmap_max); } + propagate_user_return_notify(prev_p, next_p); } int sys_fork(struct pt_regs *regs) @@ -247,6 +250,78 @@ int sys_vfork(struct pt_regs *regs) NULL, NULL); } +long +sys_clone(unsigned long clone_flags, unsigned long newsp, + void __user *parent_tid, void __user *child_tid, struct pt_regs *regs) +{ + if (!newsp) + newsp = regs->sp; + return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid); +} + +/* + * This gets run with %si containing the + * function to call, and %di containing + * the "args". + */ +extern void kernel_thread_helper(void); + +/* + * Create a kernel thread + */ +int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags) +{ + struct pt_regs regs; + + memset(®s, 0, sizeof(regs)); + + regs.si = (unsigned long) fn; + regs.di = (unsigned long) arg; + +#ifdef CONFIG_X86_32 + regs.ds = __USER_DS; + regs.es = __USER_DS; + regs.fs = __KERNEL_PERCPU; + regs.gs = __KERNEL_STACK_CANARY; +#else + regs.ss = __KERNEL_DS; +#endif + + regs.orig_ax = -1; + regs.ip = (unsigned long) kernel_thread_helper; + regs.cs = __KERNEL_CS | get_kernel_rpl(); + regs.flags = X86_EFLAGS_IF | 0x2; + + /* Ok, create the new process.. */ + return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL); +} +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 error; + char *filename; + + filename = getname(name); + error = PTR_ERR(filename); + if (IS_ERR(filename)) + return error; + error = do_execve(filename, argv, envp, regs); + +#ifdef CONFIG_X86_32 + if (error == 0) { + /* Make sure we don't return using sysenter.. */ + set_thread_flag(TIF_IRET); + } +#endif + + putname(filename); + return error; +} /* * Idle related variables and functions @@ -532,7 +607,7 @@ void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) { #ifdef CONFIG_SMP if (pm_idle == poll_idle && smp_num_siblings > 1) { - printk(KERN_WARNING "WARNING: polling idle and HT enabled," + printk_once(KERN_WARNING "WARNING: polling idle and HT enabled," " performance may degrade.\n"); } #endif diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 4cf79567cdab..f6c62667e30c 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -23,7 +23,6 @@ #include <linux/vmalloc.h> #include <linux/user.h> #include <linux/interrupt.h> -#include <linux/utsname.h> #include <linux/delay.h> #include <linux/reboot.h> #include <linux/init.h> @@ -35,7 +34,6 @@ #include <linux/tick.h> #include <linux/percpu.h> #include <linux/prctl.h> -#include <linux/dmi.h> #include <linux/ftrace.h> #include <linux/uaccess.h> #include <linux/io.h> @@ -58,6 +56,7 @@ #include <asm/idle.h> #include <asm/syscalls.h> #include <asm/ds.h> +#include <asm/debugreg.h> asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); @@ -127,39 +126,29 @@ void __show_regs(struct pt_regs *regs, int all) unsigned long d0, d1, d2, d3, d6, d7; unsigned long sp; unsigned short ss, gs; - const char *board; if (user_mode_vm(regs)) { sp = regs->sp; ss = regs->ss & 0xffff; gs = get_user_gs(regs); } else { - sp = (unsigned long) (®s->sp); + sp = kernel_stack_pointer(regs); savesegment(ss, ss); savesegment(gs, gs); } - printk("\n"); + show_regs_common(); - board = dmi_get_system_info(DMI_PRODUCT_NAME); - if (!board) - board = ""; - printk("Pid: %d, comm: %s %s (%s %.*s) %s\n", - task_pid_nr(current), current->comm, - print_tainted(), init_utsname()->release, - (int)strcspn(init_utsname()->version, " "), - init_utsname()->version, board); - - printk("EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n", + printk(KERN_DEFAULT "EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n", (u16)regs->cs, regs->ip, regs->flags, smp_processor_id()); print_symbol("EIP is at %s\n", regs->ip); - printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n", + printk(KERN_DEFAULT "EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n", regs->ax, regs->bx, regs->cx, regs->dx); - printk("ESI: %08lx EDI: %08lx EBP: %08lx ESP: %08lx\n", + printk(KERN_DEFAULT "ESI: %08lx EDI: %08lx EBP: %08lx ESP: %08lx\n", regs->si, regs->di, regs->bp, sp); - printk(" DS: %04x ES: %04x FS: %04x GS: %04x SS: %04x\n", + printk(KERN_DEFAULT " DS: %04x ES: %04x FS: %04x GS: %04x SS: %04x\n", (u16)regs->ds, (u16)regs->es, (u16)regs->fs, gs, ss); if (!all) @@ -169,61 +158,22 @@ void __show_regs(struct pt_regs *regs, int all) cr2 = read_cr2(); cr3 = read_cr3(); cr4 = read_cr4_safe(); - printk("CR0: %08lx CR2: %08lx CR3: %08lx CR4: %08lx\n", + printk(KERN_DEFAULT "CR0: %08lx CR2: %08lx CR3: %08lx CR4: %08lx\n", cr0, cr2, cr3, cr4); get_debugreg(d0, 0); get_debugreg(d1, 1); get_debugreg(d2, 2); get_debugreg(d3, 3); - printk("DR0: %08lx DR1: %08lx DR2: %08lx DR3: %08lx\n", + printk(KERN_DEFAULT "DR0: %08lx DR1: %08lx DR2: %08lx DR3: %08lx\n", d0, d1, d2, d3); get_debugreg(d6, 6); get_debugreg(d7, 7); - printk("DR6: %08lx DR7: %08lx\n", + printk(KERN_DEFAULT "DR6: %08lx DR7: %08lx\n", d6, d7); } -void show_regs(struct pt_regs *regs) -{ - __show_regs(regs, 1); - show_trace(NULL, regs, ®s->sp, regs->bp); -} - -/* - * This gets run with %bx containing the - * function to call, and %dx containing - * the "args". - */ -extern void kernel_thread_helper(void); - -/* - * Create a kernel thread - */ -int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags) -{ - struct pt_regs regs; - - memset(®s, 0, sizeof(regs)); - - regs.bx = (unsigned long) fn; - regs.dx = (unsigned long) arg; - - regs.ds = __USER_DS; - regs.es = __USER_DS; - regs.fs = __KERNEL_PERCPU; - regs.gs = __KERNEL_STACK_CANARY; - regs.orig_ax = -1; - regs.ip = (unsigned long) kernel_thread_helper; - regs.cs = __KERNEL_CS | get_kernel_rpl(); - regs.flags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2; - - /* Ok, create the new process.. */ - return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL); -} -EXPORT_SYMBOL(kernel_thread); - void release_thread(struct task_struct *dead_task) { BUG_ON(dead_task->mm); @@ -259,7 +209,12 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, task_user_gs(p) = get_user_gs(regs); + p->thread.io_bitmap_ptr = NULL; tsk = current; + err = -ENOMEM; + + memset(p->thread.ptrace_bps, 0, sizeof(p->thread.ptrace_bps)); + if (unlikely(test_tsk_thread_flag(tsk, TIF_IO_BITMAP))) { p->thread.io_bitmap_ptr = kmemdup(tsk->thread.io_bitmap_ptr, IO_BITMAP_BYTES, GFP_KERNEL); @@ -430,46 +385,6 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) return prev_p; } -int sys_clone(struct pt_regs *regs) -{ - unsigned long clone_flags; - unsigned long newsp; - int __user *parent_tidptr, *child_tidptr; - - clone_flags = regs->bx; - newsp = regs->cx; - parent_tidptr = (int __user *)regs->dx; - child_tidptr = (int __user *)regs->di; - if (!newsp) - newsp = regs->sp; - return do_fork(clone_flags, newsp, regs, 0, parent_tidptr, child_tidptr); -} - -/* - * sys_execve() executes a new program. - */ -int sys_execve(struct pt_regs *regs) -{ - int error; - char *filename; - - filename = getname((char __user *) regs->bx); - error = PTR_ERR(filename); - if (IS_ERR(filename)) - goto out; - error = do_execve(filename, - (char __user * __user *) regs->cx, - (char __user * __user *) regs->dx, - regs); - if (error == 0) { - /* Make sure we don't return using sysenter.. */ - set_thread_flag(TIF_IRET); - } - putname(filename); -out: - return error; -} - #define top_esp (THREAD_SIZE - sizeof(unsigned long)) #define top_ebp (THREAD_SIZE - 2*sizeof(unsigned long)) diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index ad535b683170..dc9690b4c4cc 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -26,7 +26,6 @@ #include <linux/slab.h> #include <linux/user.h> #include <linux/interrupt.h> -#include <linux/utsname.h> #include <linux/delay.h> #include <linux/module.h> #include <linux/ptrace.h> @@ -38,7 +37,6 @@ #include <linux/uaccess.h> #include <linux/io.h> #include <linux/ftrace.h> -#include <linux/dmi.h> #include <asm/pgtable.h> #include <asm/system.h> @@ -52,14 +50,13 @@ #include <asm/idle.h> #include <asm/syscalls.h> #include <asm/ds.h> +#include <asm/debugreg.h> asmlinkage extern void ret_from_fork(void); DEFINE_PER_CPU(unsigned long, old_rsp); static DEFINE_PER_CPU(unsigned char, is_idle); -unsigned long kernel_thread_flags = CLONE_VM | CLONE_UNTRACED; - static ATOMIC_NOTIFIER_HEAD(idle_notifier); void idle_notifier_register(struct notifier_block *n) @@ -162,31 +159,21 @@ void __show_regs(struct pt_regs *regs, int all) unsigned long d0, d1, d2, d3, d6, d7; unsigned int fsindex, gsindex; unsigned int ds, cs, es; - const char *board; - - printk("\n"); - print_modules(); - board = dmi_get_system_info(DMI_PRODUCT_NAME); - if (!board) - board = ""; - printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s %s\n", - current->pid, current->comm, print_tainted(), - init_utsname()->release, - (int)strcspn(init_utsname()->version, " "), - init_utsname()->version, board); - printk(KERN_INFO "RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip); + + show_regs_common(); + printk(KERN_DEFAULT "RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip); printk_address(regs->ip, 1); - printk(KERN_INFO "RSP: %04lx:%016lx EFLAGS: %08lx\n", regs->ss, + printk(KERN_DEFAULT "RSP: %04lx:%016lx EFLAGS: %08lx\n", regs->ss, regs->sp, regs->flags); - printk(KERN_INFO "RAX: %016lx RBX: %016lx RCX: %016lx\n", + printk(KERN_DEFAULT "RAX: %016lx RBX: %016lx RCX: %016lx\n", regs->ax, regs->bx, regs->cx); - printk(KERN_INFO "RDX: %016lx RSI: %016lx RDI: %016lx\n", + printk(KERN_DEFAULT "RDX: %016lx RSI: %016lx RDI: %016lx\n", regs->dx, regs->si, regs->di); - printk(KERN_INFO "RBP: %016lx R08: %016lx R09: %016lx\n", + printk(KERN_DEFAULT "RBP: %016lx R08: %016lx R09: %016lx\n", regs->bp, regs->r8, regs->r9); - printk(KERN_INFO "R10: %016lx R11: %016lx R12: %016lx\n", + printk(KERN_DEFAULT "R10: %016lx R11: %016lx R12: %016lx\n", regs->r10, regs->r11, regs->r12); - printk(KERN_INFO "R13: %016lx R14: %016lx R15: %016lx\n", + printk(KERN_DEFAULT "R13: %016lx R14: %016lx R15: %016lx\n", regs->r13, regs->r14, regs->r15); asm("movl %%ds,%0" : "=r" (ds)); @@ -207,28 +194,21 @@ void __show_regs(struct pt_regs *regs, int all) cr3 = read_cr3(); cr4 = read_cr4(); - printk(KERN_INFO "FS: %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n", + printk(KERN_DEFAULT "FS: %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n", fs, fsindex, gs, gsindex, shadowgs); - printk(KERN_INFO "CS: %04x DS: %04x ES: %04x CR0: %016lx\n", cs, ds, + printk(KERN_DEFAULT "CS: %04x DS: %04x ES: %04x CR0: %016lx\n", cs, ds, es, cr0); - printk(KERN_INFO "CR2: %016lx CR3: %016lx CR4: %016lx\n", cr2, cr3, + printk(KERN_DEFAULT "CR2: %016lx CR3: %016lx CR4: %016lx\n", cr2, cr3, cr4); get_debugreg(d0, 0); get_debugreg(d1, 1); get_debugreg(d2, 2); - printk(KERN_INFO "DR0: %016lx DR1: %016lx DR2: %016lx\n", d0, d1, d2); + printk(KERN_DEFAULT "DR0: %016lx DR1: %016lx DR2: %016lx\n", d0, d1, d2); get_debugreg(d3, 3); get_debugreg(d6, 6); get_debugreg(d7, 7); - printk(KERN_INFO "DR3: %016lx DR6: %016lx DR7: %016lx\n", d3, d6, d7); -} - -void show_regs(struct pt_regs *regs) -{ - printk(KERN_INFO "CPU %d:", smp_processor_id()); - __show_regs(regs, 1); - show_trace(NULL, regs, (void *)(regs + 1), regs->bp); + printk(KERN_DEFAULT "DR3: %016lx DR6: %016lx DR7: %016lx\n", d3, d6, d7); } void release_thread(struct task_struct *dead_task) @@ -285,8 +265,9 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, *childregs = *regs; childregs->ax = 0; - childregs->sp = sp; - if (sp == ~0UL) + if (user_mode(regs)) + childregs->sp = sp; + else childregs->sp = (unsigned long)childregs; p->thread.sp = (unsigned long) childregs; @@ -297,12 +278,16 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, p->thread.fs = me->thread.fs; p->thread.gs = me->thread.gs; + p->thread.io_bitmap_ptr = NULL; savesegment(gs, p->thread.gsindex); savesegment(fs, p->thread.fsindex); savesegment(es, p->thread.es); savesegment(ds, p->thread.ds); + err = -ENOMEM; + memset(p->thread.ptrace_bps, 0, sizeof(p->thread.ptrace_bps)); + if (unlikely(test_tsk_thread_flag(me, TIF_IO_BITMAP))) { p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL); if (!p->thread.io_bitmap_ptr) { @@ -341,29 +326,46 @@ out: kfree(p->thread.io_bitmap_ptr); p->thread.io_bitmap_max = 0; } + return err; } -void -start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) +static void +start_thread_common(struct pt_regs *regs, unsigned long new_ip, + unsigned long new_sp, + unsigned int _cs, unsigned int _ss, unsigned int _ds) { loadsegment(fs, 0); - loadsegment(es, 0); - loadsegment(ds, 0); + loadsegment(es, _ds); + loadsegment(ds, _ds); load_gs_index(0); regs->ip = new_ip; regs->sp = new_sp; percpu_write(old_rsp, new_sp); - regs->cs = __USER_CS; - regs->ss = __USER_DS; - regs->flags = 0x200; + regs->cs = _cs; + regs->ss = _ss; + regs->flags = X86_EFLAGS_IF; set_fs(USER_DS); /* * Free the old FP and other extended state */ free_thread_xstate(current); } -EXPORT_SYMBOL_GPL(start_thread); + +void +start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) +{ + start_thread_common(regs, new_ip, new_sp, + __USER_CS, __USER_DS, 0); +} + +#ifdef CONFIG_IA32_EMULATION +void start_thread_ia32(struct pt_regs *regs, u32 new_ip, u32 new_sp) +{ + start_thread_common(regs, new_ip, new_sp, + __USER32_CS, __USER32_DS, __USER32_DS); +} +#endif /* * switch_to(x,y) should switch tasks from x to y. @@ -495,26 +497,8 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) */ if (preload_fpu) __math_state_restore(); - return prev_p; -} -/* - * sys_execve() executes a new program. - */ -asmlinkage -long sys_execve(char __user *name, char __user * __user *argv, - char __user * __user *envp, struct pt_regs *regs) -{ - long error; - char *filename; - - filename = getname(name); - error = PTR_ERR(filename); - if (IS_ERR(filename)) - return error; - error = do_execve(filename, argv, envp, regs); - putname(filename); - return error; + return prev_p; } void set_personality_64bit(void) @@ -531,13 +515,16 @@ void set_personality_64bit(void) current->personality &= ~READ_IMPLIES_EXEC; } -asmlinkage long -sys_clone(unsigned long clone_flags, unsigned long newsp, - void __user *parent_tid, void __user *child_tid, struct pt_regs *regs) +void set_personality_ia32(void) { - if (!newsp) - newsp = regs->sp; - return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid); + /* inherit personality from parent */ + + /* Make sure to be in 32bit mode */ + set_thread_flag(TIF_IA32); + current->personality |= force_personality32; + + /* Prepare the first "return" to user space */ + current_thread_info()->status |= TS_COMPAT; } unsigned long get_wchan(struct task_struct *p) @@ -664,3 +651,8 @@ long sys_arch_prctl(int code, unsigned long addr) return do_arch_prctl(current, code, addr); } +unsigned long KSTK_ESP(struct task_struct *task) +{ + return (test_tsk_thread_flag(task, TIF_IA32)) ? + (task_pt_regs(task)->sp) : ((task)->thread.usersp); +} diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 7b058a2dc66a..a503b1fd04e5 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -22,6 +22,8 @@ #include <linux/seccomp.h> #include <linux/signal.h> #include <linux/workqueue.h> +#include <linux/perf_event.h> +#include <linux/hw_breakpoint.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -34,6 +36,7 @@ #include <asm/prctl.h> #include <asm/proto.h> #include <asm/ds.h> +#include <asm/hw_breakpoint.h> #include "tls.h" @@ -45,10 +48,99 @@ enum x86_regset { REGSET_FP, REGSET_XFP, REGSET_IOPERM64 = REGSET_XFP, + REGSET_XSTATE, REGSET_TLS, REGSET_IOPERM32, }; +struct pt_regs_offset { + const char *name; + int offset; +}; + +#define REG_OFFSET_NAME(r) {.name = #r, .offset = offsetof(struct pt_regs, r)} +#define REG_OFFSET_END {.name = NULL, .offset = 0} + +static const struct pt_regs_offset regoffset_table[] = { +#ifdef CONFIG_X86_64 + REG_OFFSET_NAME(r15), + REG_OFFSET_NAME(r14), + REG_OFFSET_NAME(r13), + REG_OFFSET_NAME(r12), + REG_OFFSET_NAME(r11), + REG_OFFSET_NAME(r10), + REG_OFFSET_NAME(r9), + REG_OFFSET_NAME(r8), +#endif + REG_OFFSET_NAME(bx), + REG_OFFSET_NAME(cx), + REG_OFFSET_NAME(dx), + REG_OFFSET_NAME(si), + REG_OFFSET_NAME(di), + REG_OFFSET_NAME(bp), + REG_OFFSET_NAME(ax), +#ifdef CONFIG_X86_32 + REG_OFFSET_NAME(ds), + REG_OFFSET_NAME(es), + REG_OFFSET_NAME(fs), + REG_OFFSET_NAME(gs), +#endif + REG_OFFSET_NAME(orig_ax), + REG_OFFSET_NAME(ip), + REG_OFFSET_NAME(cs), + REG_OFFSET_NAME(flags), + REG_OFFSET_NAME(sp), + REG_OFFSET_NAME(ss), + REG_OFFSET_END, +}; + +/** + * regs_query_register_offset() - query register offset from its name + * @name: the name of a register + * + * regs_query_register_offset() returns the offset of a register in struct + * pt_regs from its name. If the name is invalid, this returns -EINVAL; + */ +int regs_query_register_offset(const char *name) +{ + const struct pt_regs_offset *roff; + for (roff = regoffset_table; roff->name != NULL; roff++) + if (!strcmp(roff->name, name)) + return roff->offset; + return -EINVAL; +} + +/** + * regs_query_register_name() - query register name from its offset + * @offset: the offset of a register in struct pt_regs. + * + * regs_query_register_name() returns the name of a register from its + * offset in struct pt_regs. If the @offset is invalid, this returns NULL; + */ +const char *regs_query_register_name(unsigned int offset) +{ + const struct pt_regs_offset *roff; + for (roff = regoffset_table; roff->name != NULL; roff++) + if (roff->offset == offset) + return roff->name; + return NULL; +} + +static const int arg_offs_table[] = { +#ifdef CONFIG_X86_32 + [0] = offsetof(struct pt_regs, ax), + [1] = offsetof(struct pt_regs, dx), + [2] = offsetof(struct pt_regs, cx) +#else /* CONFIG_X86_64 */ + [0] = offsetof(struct pt_regs, di), + [1] = offsetof(struct pt_regs, si), + [2] = offsetof(struct pt_regs, dx), + [3] = offsetof(struct pt_regs, cx), + [4] = offsetof(struct pt_regs, r8), + [5] = offsetof(struct pt_regs, r9) +#endif +}; + /* * does not yet catch signals sent when the child dies. * in exit.c or in signal.c. @@ -137,11 +229,6 @@ static int set_segment_reg(struct task_struct *task, return 0; } -static unsigned long debugreg_addr_limit(struct task_struct *task) -{ - return TASK_SIZE - 3; -} - #else /* CONFIG_X86_64 */ #define FLAG_MASK (FLAG_MASK_32 | X86_EFLAGS_NT) @@ -266,15 +353,6 @@ static int set_segment_reg(struct task_struct *task, return 0; } -static unsigned long debugreg_addr_limit(struct task_struct *task) -{ -#ifdef CONFIG_IA32_EMULATION - if (test_tsk_thread_flag(task, TIF_IA32)) - return IA32_PAGE_OFFSET - 3; -#endif - return TASK_SIZE_MAX - 7; -} - #endif /* CONFIG_X86_32 */ static unsigned long get_flags(struct task_struct *task) @@ -408,14 +486,14 @@ static int genregs_get(struct task_struct *target, { if (kbuf) { unsigned long *k = kbuf; - while (count > 0) { + while (count >= sizeof(*k)) { *k++ = getreg(target, pos); count -= sizeof(*k); pos += sizeof(*k); } } else { unsigned long __user *u = ubuf; - while (count > 0) { + while (count >= sizeof(*u)) { if (__put_user(getreg(target, pos), u++)) return -EFAULT; count -= sizeof(*u); @@ -434,14 +512,14 @@ static int genregs_set(struct task_struct *target, int ret = 0; if (kbuf) { const unsigned long *k = kbuf; - while (count > 0 && !ret) { + while (count >= sizeof(*k) && !ret) { ret = putreg(target, pos, *k++); count -= sizeof(*k); pos += sizeof(*k); } } else { const unsigned long __user *u = ubuf; - while (count > 0 && !ret) { + while (count >= sizeof(*u) && !ret) { unsigned long word; ret = __get_user(word, u++); if (ret) @@ -454,99 +532,240 @@ static int genregs_set(struct task_struct *target, return ret; } +static void ptrace_triggered(struct perf_event *bp, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + int i; + struct thread_struct *thread = &(current->thread); + + /* + * Store in the virtual DR6 register the fact that the breakpoint + * was hit so the thread's debugger will see it. + */ + for (i = 0; i < HBP_NUM; i++) { + if (thread->ptrace_bps[i] == bp) + break; + } + + thread->debugreg6 |= (DR_TRAP0 << i); +} + /* - * This function is trivial and will be inlined by the compiler. - * Having it separates the implementation details of debug - * registers from the interface details of ptrace. + * Walk through every ptrace breakpoints for this thread and + * build the dr7 value on top of their attributes. + * */ -static unsigned long ptrace_get_debugreg(struct task_struct *child, int n) +static unsigned long ptrace_get_dr7(struct perf_event *bp[]) { - switch (n) { - case 0: return child->thread.debugreg0; - case 1: return child->thread.debugreg1; - case 2: return child->thread.debugreg2; - case 3: return child->thread.debugreg3; - case 6: return child->thread.debugreg6; - case 7: return child->thread.debugreg7; + int i; + int dr7 = 0; + struct arch_hw_breakpoint *info; + + for (i = 0; i < HBP_NUM; i++) { + if (bp[i] && !bp[i]->attr.disabled) { + info = counter_arch_bp(bp[i]); + dr7 |= encode_dr7(i, info->len, info->type); + } } - return 0; + + return dr7; } -static int ptrace_set_debugreg(struct task_struct *child, - int n, unsigned long data) +static int +ptrace_modify_breakpoint(struct perf_event *bp, int len, int type, + struct task_struct *tsk, int disabled) { - int i; + int err; + int gen_len, gen_type; + struct perf_event_attr attr; - if (unlikely(n == 4 || n == 5)) - return -EIO; + /* + * We should have at least an inactive breakpoint at this + * slot. It means the user is writing dr7 without having + * written the address register first + */ + if (!bp) + return -EINVAL; - if (n < 4 && unlikely(data >= debugreg_addr_limit(child))) - return -EIO; + err = arch_bp_generic_fields(len, type, &gen_len, &gen_type); + if (err) + return err; - switch (n) { - case 0: child->thread.debugreg0 = data; break; - case 1: child->thread.debugreg1 = data; break; - case 2: child->thread.debugreg2 = data; break; - case 3: child->thread.debugreg3 = data; break; + attr = bp->attr; + attr.bp_len = gen_len; + attr.bp_type = gen_type; + attr.disabled = disabled; - case 6: - if ((data & ~0xffffffffUL) != 0) - return -EIO; - child->thread.debugreg6 = data; - break; + return modify_user_hw_breakpoint(bp, &attr); +} - case 7: +/* + * Handle ptrace writes to debug register 7. + */ +static int ptrace_write_dr7(struct task_struct *tsk, unsigned long data) +{ + struct thread_struct *thread = &(tsk->thread); + unsigned long old_dr7; + int i, orig_ret = 0, rc = 0; + int enabled, second_pass = 0; + unsigned len, type; + struct perf_event *bp; + + data &= ~DR_CONTROL_RESERVED; + old_dr7 = ptrace_get_dr7(thread->ptrace_bps); +restore: + /* + * Loop through all the hardware breakpoints, making the + * appropriate changes to each. + */ + for (i = 0; i < HBP_NUM; i++) { + enabled = decode_dr7(data, i, &len, &type); + bp = thread->ptrace_bps[i]; + + if (!enabled) { + if (bp) { + /* + * Don't unregister the breakpoints right-away, + * unless all register_user_hw_breakpoint() + * requests have succeeded. This prevents + * any window of opportunity for debug + * register grabbing by other users. + */ + if (!second_pass) + continue; + + rc = ptrace_modify_breakpoint(bp, len, type, + tsk, 1); + if (rc) + break; + } + continue; + } + + rc = ptrace_modify_breakpoint(bp, len, type, tsk, 0); + if (rc) + break; + } + /* + * Make a second pass to free the remaining unused breakpoints + * or to restore the original breakpoints if an error occurred. + */ + if (!second_pass) { + second_pass = 1; + if (rc < 0) { + orig_ret = rc; + data = old_dr7; + } + goto restore; + } + return ((orig_ret < 0) ? orig_ret : rc); +} + +/* + * Handle PTRACE_PEEKUSR calls for the debug register area. + */ +static unsigned long ptrace_get_debugreg(struct task_struct *tsk, int n) +{ + struct thread_struct *thread = &(tsk->thread); + unsigned long val = 0; + + if (n < HBP_NUM) { + struct perf_event *bp; + bp = thread->ptrace_bps[n]; + if (!bp) + return 0; + val = bp->hw.info.address; + } else if (n == 6) { + val = thread->debugreg6; + } else if (n == 7) { + val = thread->ptrace_dr7; + } + return val; +} + +static int ptrace_set_breakpoint_addr(struct task_struct *tsk, int nr, + unsigned long addr) +{ + struct perf_event *bp; + struct thread_struct *t = &tsk->thread; + struct perf_event_attr attr; + + if (!t->ptrace_bps[nr]) { + hw_breakpoint_init(&attr); /* - * Sanity-check data. Take one half-byte at once with - * check = (val >> (16 + 4*i)) & 0xf. It contains the - * R/Wi and LENi bits; bits 0 and 1 are R/Wi, and bits - * 2 and 3 are LENi. Given a list of invalid values, - * we do mask |= 1 << invalid_value, so that - * (mask >> check) & 1 is a correct test for invalid - * values. - * - * R/Wi contains the type of the breakpoint / - * watchpoint, LENi contains the length of the watched - * data in the watchpoint case. - * - * The invalid values are: - * - LENi == 0x10 (undefined), so mask |= 0x0f00. [32-bit] - * - R/Wi == 0x10 (break on I/O reads or writes), so - * mask |= 0x4444. - * - R/Wi == 0x00 && LENi != 0x00, so we have mask |= - * 0x1110. - * - * Finally, mask = 0x0f00 | 0x4444 | 0x1110 == 0x5f54. - * - * See the Intel Manual "System Programming Guide", - * 15.2.4 - * - * Note that LENi == 0x10 is defined on x86_64 in long - * mode (i.e. even for 32-bit userspace software, but - * 64-bit kernel), so the x86_64 mask value is 0x5454. - * See the AMD manual no. 24593 (AMD64 System Programming) + * Put stub len and type to register (reserve) an inactive but + * correct bp */ -#ifdef CONFIG_X86_32 -#define DR7_MASK 0x5f54 -#else -#define DR7_MASK 0x5554 -#endif - data &= ~DR_CONTROL_RESERVED; - for (i = 0; i < 4; i++) - if ((DR7_MASK >> ((data >> (16 + 4*i)) & 0xf)) & 1) - return -EIO; - child->thread.debugreg7 = data; - if (data) - set_tsk_thread_flag(child, TIF_DEBUG); - else - clear_tsk_thread_flag(child, TIF_DEBUG); - break; + attr.bp_addr = addr; + attr.bp_len = HW_BREAKPOINT_LEN_1; + attr.bp_type = HW_BREAKPOINT_W; + attr.disabled = 1; + + bp = register_user_hw_breakpoint(&attr, ptrace_triggered, tsk); + + /* + * CHECKME: the previous code returned -EIO if the addr wasn't + * a valid task virtual addr. The new one will return -EINVAL in + * this case. + * -EINVAL may be what we want for in-kernel breakpoints users, + * but -EIO looks better for ptrace, since we refuse a register + * writing for the user. And anyway this is the previous + * behaviour. + */ + if (IS_ERR(bp)) + return PTR_ERR(bp); + + t->ptrace_bps[nr] = bp; + } else { + int err; + + bp = t->ptrace_bps[nr]; + + attr = bp->attr; + attr.bp_addr = addr; + err = modify_user_hw_breakpoint(bp, &attr); + if (err) + return err; } + return 0; } /* + * Handle PTRACE_POKEUSR calls for the debug register area. + */ +int ptrace_set_debugreg(struct task_struct *tsk, int n, unsigned long val) +{ + struct thread_struct *thread = &(tsk->thread); + int rc = 0; + + /* There are no DR4 or DR5 registers */ + if (n == 4 || n == 5) + return -EIO; + + if (n == 6) { + thread->debugreg6 = val; + goto ret_path; + } + if (n < HBP_NUM) { + rc = ptrace_set_breakpoint_addr(tsk, n, val); + if (rc) + return rc; + } + /* All that's left is DR7 */ + if (n == 7) { + rc = ptrace_write_dr7(tsk, val); + if (!rc) + thread->ptrace_dr7 = val; + } + +ret_path: + return rc; +} + +/* * These access the current or another (stopped) task's io permission * bitmap for debugging or core dump. */ @@ -1219,14 +1438,14 @@ static int genregs32_get(struct task_struct *target, { if (kbuf) { compat_ulong_t *k = kbuf; - while (count > 0) { + while (count >= sizeof(*k)) { getreg32(target, pos, k++); count -= sizeof(*k); pos += sizeof(*k); } } else { compat_ulong_t __user *u = ubuf; - while (count > 0) { + while (count >= sizeof(*u)) { compat_ulong_t word; getreg32(target, pos, &word); if (__put_user(word, u++)) @@ -1247,14 +1466,14 @@ static int genregs32_set(struct task_struct *target, int ret = 0; if (kbuf) { const compat_ulong_t *k = kbuf; - while (count > 0 && !ret) { + while (count >= sizeof(*k) && !ret) { ret = putreg32(target, pos, *k++); count -= sizeof(*k); pos += sizeof(*k); } } else { const compat_ulong_t __user *u = ubuf; - while (count > 0 && !ret) { + while (count >= sizeof(*u) && !ret) { compat_ulong_t word; ret = __get_user(word, u++); if (ret) @@ -1345,7 +1564,7 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, #ifdef CONFIG_X86_64 -static const struct user_regset x86_64_regsets[] = { +static struct user_regset x86_64_regsets[] __read_mostly = { [REGSET_GENERAL] = { .core_note_type = NT_PRSTATUS, .n = sizeof(struct user_regs_struct) / sizeof(long), @@ -1358,6 +1577,12 @@ static const struct user_regset x86_64_regsets[] = { .size = sizeof(long), .align = sizeof(long), .active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set }, + [REGSET_XSTATE] = { + .core_note_type = NT_X86_XSTATE, + .size = sizeof(u64), .align = sizeof(u64), + .active = xstateregs_active, .get = xstateregs_get, + .set = xstateregs_set + }, [REGSET_IOPERM64] = { .core_note_type = NT_386_IOPERM, .n = IO_BITMAP_LONGS, @@ -1383,7 +1608,7 @@ static const struct user_regset_view user_x86_64_view = { #endif /* CONFIG_X86_64 */ #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION -static const struct user_regset x86_32_regsets[] = { +static struct user_regset x86_32_regsets[] __read_mostly = { [REGSET_GENERAL] = { .core_note_type = NT_PRSTATUS, .n = sizeof(struct user_regs_struct32) / sizeof(u32), @@ -1402,6 +1627,12 @@ static const struct user_regset x86_32_regsets[] = { .size = sizeof(u32), .align = sizeof(u32), .active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set }, + [REGSET_XSTATE] = { + .core_note_type = NT_X86_XSTATE, + .size = sizeof(u64), .align = sizeof(u64), + .active = xstateregs_active, .get = xstateregs_get, + .set = xstateregs_set + }, [REGSET_TLS] = { .core_note_type = NT_386_TLS, .n = GDT_ENTRY_TLS_ENTRIES, .bias = GDT_ENTRY_TLS_MIN, @@ -1424,6 +1655,23 @@ static const struct user_regset_view user_x86_32_view = { }; #endif +/* + * This represents bytes 464..511 in the memory layout exported through + * the REGSET_XSTATE interface. + */ +u64 xstate_fx_sw_bytes[USER_XSTATE_FX_SW_WORDS]; + +void update_regset_xstate_info(unsigned int size, u64 xstate_mask) +{ +#ifdef CONFIG_X86_64 + x86_64_regsets[REGSET_XSTATE].n = size / sizeof(u64); +#endif +#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION + x86_32_regsets[REGSET_XSTATE].n = size / sizeof(u64); +#endif + xstate_fx_sw_bytes[USER_XSTATE_XCR0_WORD] = xstate_mask; +} + const struct user_regset_view *task_user_regset_view(struct task_struct *task) { #ifdef CONFIG_IA32_EMULATION @@ -1437,21 +1685,33 @@ const struct user_regset_view *task_user_regset_view(struct task_struct *task) #endif } -void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, - int error_code, int si_code) +static void fill_sigtrap_info(struct task_struct *tsk, + struct pt_regs *regs, + int error_code, int si_code, + struct siginfo *info) { - struct siginfo info; - tsk->thread.trap_no = 1; tsk->thread.error_code = error_code; - memset(&info, 0, sizeof(info)); - info.si_signo = SIGTRAP; - info.si_code = si_code; + memset(info, 0, sizeof(*info)); + info->si_signo = SIGTRAP; + info->si_code = si_code; + info->si_addr = user_mode_vm(regs) ? (void __user *)regs->ip : NULL; +} - /* User-mode ip? */ - info.si_addr = user_mode_vm(regs) ? (void __user *) regs->ip : NULL; +void user_single_step_siginfo(struct task_struct *tsk, + struct pt_regs *regs, + struct siginfo *info) +{ + fill_sigtrap_info(tsk, regs, 0, TRAP_BRKPT, info); +} +void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, + int error_code, int si_code) +{ + struct siginfo info; + + fill_sigtrap_info(tsk, regs, error_code, si_code, &info); /* Send us the fake SIGTRAP */ force_sig_info(SIGTRAP, &info, tsk); } @@ -1516,29 +1776,22 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs) asmregparm void syscall_trace_leave(struct pt_regs *regs) { + bool step; + if (unlikely(current->audit_context)) audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax); if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) trace_sys_exit(regs, regs->ax); - if (test_thread_flag(TIF_SYSCALL_TRACE)) - tracehook_report_syscall_exit(regs, 0); - /* * If TIF_SYSCALL_EMU is set, we only get here because of * TIF_SINGLESTEP (i.e. this is PTRACE_SYSEMU_SINGLESTEP). * We already reported this syscall instruction in - * syscall_trace_enter(), so don't do any more now. - */ - if (unlikely(test_thread_flag(TIF_SYSCALL_EMU))) - return; - - /* - * If we are single-stepping, synthesize a trap to follow the - * system call instruction. + * syscall_trace_enter(). */ - if (test_thread_flag(TIF_SINGLESTEP) && - tracehook_consider_fatal_signal(current, SIGTRAP)) - send_sigtrap(current, regs, 0, TRAP_BRKPT); + step = unlikely(test_thread_flag(TIF_SINGLESTEP)) && + !test_thread_flag(TIF_SYSCALL_EMU); + if (step || test_thread_flag(TIF_SYSCALL_TRACE)) + tracehook_report_syscall_exit(regs, step); } diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c index 6c3b2c6fd772..12e9feaa2f7a 100644 --- a/arch/x86/kernel/quirks.c +++ b/arch/x86/kernel/quirks.c @@ -491,6 +491,19 @@ void force_hpet_resume(void) break; } } + +/* + * HPET MSI on some boards (ATI SB700/SB800) has side effect on + * floppy DMA. Disable HPET MSI on such platforms. + */ +static void force_disable_hpet_msi(struct pci_dev *unused) +{ + hpet_msi_disable = 1; +} + +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_SBX00_SMBUS, + force_disable_hpet_msi); + #endif #if defined(CONFIG_PCI) && defined(CONFIG_NUMA) @@ -499,6 +512,7 @@ static void __init quirk_amd_nb_node(struct pci_dev *dev) { struct pci_dev *nb_ht; unsigned int devfn; + u32 node; u32 val; devfn = PCI_DEVFN(PCI_SLOT(dev->devfn), 0); @@ -507,7 +521,13 @@ static void __init quirk_amd_nb_node(struct pci_dev *dev) return; pci_read_config_dword(nb_ht, 0x60, &val); - set_dev_node(&dev->dev, val & 7); + node = val & 7; + /* + * Some hardware may return an invalid node ID, + * so check it first: + */ + if (node_online(node)) + set_dev_node(&dev->dev, node); pci_dev_put(nb_ht); } diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 27349f92a6d7..8e1aac86b50c 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -4,6 +4,7 @@ #include <linux/pm.h> #include <linux/efi.h> #include <linux/dmi.h> +#include <linux/sched.h> #include <linux/tboot.h> #include <acpi/reboot.h> #include <asm/io.h> @@ -22,7 +23,7 @@ # include <linux/ctype.h> # include <linux/mc146818rtc.h> #else -# include <asm/iommu.h> +# include <asm/x86_init.h> #endif /* @@ -202,6 +203,15 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_BOARD_NAME, "0T656F"), }, }, + { /* Handle problems with rebooting on Dell OptiPlex 760 with 0G919G*/ + .callback = set_bios_reboot, + .ident = "Dell OptiPlex 760", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 760"), + DMI_MATCH(DMI_BOARD_NAME, "0G919G"), + }, + }, { /* Handle problems with rebooting on Dell 2400's */ .callback = set_bios_reboot, .ident = "Dell PowerEdge 2400", @@ -258,6 +268,14 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "SBC-FITPC2"), }, }, + { /* Handle problems with rebooting on ASUS P4S800 */ + .callback = set_bios_reboot, + .ident = "ASUS P4S800", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), + DMI_MATCH(DMI_BOARD_NAME, "P4S800"), + }, + }, { } }; @@ -435,6 +453,22 @@ static struct dmi_system_id __initdata pci_reboot_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro5"), }, }, + { /* Handle problems with rebooting on Apple Macmini3,1 */ + .callback = set_pci_reboot, + .ident = "Apple Macmini3,1", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "Macmini3,1"), + }, + }, + { /* Handle problems with rebooting on the iMac9,1. */ + .callback = set_pci_reboot, + .ident = "Apple iMac9,1", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "iMac9,1"), + }, + }, { } }; @@ -613,7 +647,7 @@ void native_machine_shutdown(void) #endif #ifdef CONFIG_X86_64 - pci_iommu_shutdown(); + x86_platform.iommu_shutdown(); #endif } diff --git a/arch/x86/kernel/reboot_fixups_32.c b/arch/x86/kernel/reboot_fixups_32.c index 61a837743fe5..fda313ebbb03 100644 --- a/arch/x86/kernel/reboot_fixups_32.c +++ b/arch/x86/kernel/reboot_fixups_32.c @@ -12,7 +12,7 @@ #include <linux/interrupt.h> #include <asm/reboot_fixups.h> #include <asm/msr.h> -#include <asm/geode.h> +#include <linux/cs5535.h> static void cs5530a_warm_reset(struct pci_dev *dev) { @@ -80,6 +80,7 @@ void mach_reboot_fixups(void) continue; cur->reboot_fixup(dev); + pci_dev_put(dev); } } diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index e09f0e2c14b5..5d7ba1a449bd 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -73,6 +73,7 @@ #include <asm/mtrr.h> #include <asm/apic.h> +#include <asm/trampoline.h> #include <asm/e820.h> #include <asm/mpspec.h> #include <asm/setup.h> @@ -106,9 +107,11 @@ #include <asm/percpu.h> #include <asm/topology.h> #include <asm/apicdef.h> +#include <asm/k8.h> #ifdef CONFIG_X86_64 #include <asm/numa_64.h> #endif +#include <asm/mce.h> /* * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. @@ -118,7 +121,9 @@ unsigned long max_low_pfn_mapped; unsigned long max_pfn_mapped; +#ifdef CONFIG_DMI RESERVE_BRK(dmi_alloc, 65536); +#endif unsigned int boot_cpu_id __read_mostly; @@ -247,7 +252,7 @@ EXPORT_SYMBOL(edd); * from boot_params into a safe place. * */ -static inline void copy_edd(void) +static inline void __init copy_edd(void) { memcpy(edd.mbr_signature, boot_params.edd_mbr_sig_buffer, sizeof(edd.mbr_signature)); @@ -256,7 +261,7 @@ static inline void copy_edd(void) edd.edd_info_nr = boot_params.eddbuf_entries; } #else -static inline void copy_edd(void) +static inline void __init copy_edd(void) { } #endif @@ -486,42 +491,11 @@ static void __init reserve_early_setup_data(void) #ifdef CONFIG_KEXEC -/** - * Reserve @size bytes of crashkernel memory at any suitable offset. - * - * @size: Size of the crashkernel memory to reserve. - * Returns the base address on success, and -1ULL on failure. - */ -static -unsigned long long __init find_and_reserve_crashkernel(unsigned long long size) -{ - const unsigned long long alignment = 16<<20; /* 16M */ - unsigned long long start = 0LL; - - while (1) { - int ret; - - start = find_e820_area(start, ULONG_MAX, size, alignment); - if (start == -1ULL) - return start; - - /* try to reserve it */ - ret = reserve_bootmem_generic(start, size, BOOTMEM_EXCLUSIVE); - if (ret >= 0) - return start; - - start += alignment; - } -} - static inline unsigned long long get_total_mem(void) { unsigned long long total; - total = max_low_pfn - min_low_pfn; -#ifdef CONFIG_HIGHMEM - total += highend_pfn - highstart_pfn; -#endif + total = max_pfn - min_low_pfn; return total << PAGE_SHIFT; } @@ -541,21 +515,25 @@ static void __init reserve_crashkernel(void) /* 0 means: find the address automatically */ if (crash_base <= 0) { - crash_base = find_and_reserve_crashkernel(crash_size); + const unsigned long long alignment = 16<<20; /* 16M */ + + crash_base = find_e820_area(alignment, ULONG_MAX, crash_size, + alignment); if (crash_base == -1ULL) { - pr_info("crashkernel reservation failed. " - "No suitable area found.\n"); + pr_info("crashkernel reservation failed - No suitable area found.\n"); return; } } else { - ret = reserve_bootmem_generic(crash_base, crash_size, - BOOTMEM_EXCLUSIVE); - if (ret < 0) { - pr_info("crashkernel reservation failed - " - "memory is in use\n"); + unsigned long long start; + + start = find_e820_area(crash_base, ULONG_MAX, 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"); printk(KERN_INFO "Reserving %ldMB of memory at %ldMB " "for crashkernel (System RAM: %ldMB)\n", @@ -660,22 +638,54 @@ static struct dmi_system_id __initdata bad_bios_dmi_table[] = { }, }, { + .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 board. - * It hase different DMI_BIOS_VENDOR = "Intel Corp.", for now we will + * 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"), + }, + }, #endif {} }; +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. + */ + e820_update_range(0, PAGE_SIZE, E820_RAM, E820_RESERVED); + /* + * special case: Some BIOSen report the PC BIOS + * area (640->1Mb) as ram even though it is not. + * take them out. + */ + e820_remove_range(BIOS_BEGIN, BIOS_END - BIOS_BEGIN, E820_RAM, 1); + sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); +} + /* * 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 @@ -691,6 +701,9 @@ static struct dmi_system_id __initdata bad_bios_dmi_table[] = { void __init setup_arch(char **cmdline_p) { + int acpi = 0; + int k8 = 0; + #ifdef CONFIG_X86_32 memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); visws_early_detect(); @@ -783,21 +796,18 @@ void __init setup_arch(char **cmdline_p) strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE); *cmdline_p = command_line; -#ifdef CONFIG_X86_64 /* - * Must call this twice: Once just to detect whether hardware doesn't - * support NX (so that the early EHCI debug console setup can safely - * call set_fixmap(), and then again after parsing early parameters to - * honor the respective command line option. + * x86_configure_nx() is called before parse_early_param() to detect + * whether hardware doesn't support NX (so that the early EHCI debug + * console setup can safely call set_fixmap()). It may then be called + * again from within noexec_setup() during parsing early parameters + * to honor the respective command line option. */ - check_efer(); -#endif + x86_configure_nx(); parse_early_param(); -#ifdef CONFIG_X86_64 - check_efer(); -#endif + x86_report_nx(); /* Must be before kernel pagetables are setup */ vmi_activate(); @@ -839,7 +849,7 @@ void __init setup_arch(char **cmdline_p) insert_resource(&iomem_resource, &data_resource); insert_resource(&iomem_resource, &bss_resource); - + trim_bios_range(); #ifdef CONFIG_X86_32 if (ppro_with_ram_bug()) { e820_update_range(0x70000000ULL, 0x40000ULL, E820_RAM, @@ -893,6 +903,20 @@ void __init setup_arch(char **cmdline_p) reserve_brk(); + /* + * Find and reserve possible boot-time SMP configuration: + */ + find_smp_config(); + + reserve_trampoline_memory(); + +#ifdef CONFIG_ACPI_SLEEP + /* + * Reserve low memory region for sleep support. + * even before init_memory_mapping + */ + acpi_reserve_wakeup_memory(); +#endif init_gbpages(); /* max_pfn_mapped is updated here */ @@ -919,6 +943,8 @@ void __init setup_arch(char **cmdline_p) reserve_initrd(); + reserve_crashkernel(); + vsmp_init(); io_delay_init(); @@ -934,32 +960,20 @@ void __init setup_arch(char **cmdline_p) /* * Parse SRAT to discover nodes. */ - acpi_numa_init(); + acpi = acpi_numa_init(); #endif - initmem_init(0, max_pfn); - -#ifdef CONFIG_ACPI_SLEEP - /* - * Reserve low memory region for sleep support. - */ - acpi_reserve_bootmem(); +#ifdef CONFIG_K8_NUMA + if (!acpi) + k8 = !k8_numa_init(0, max_pfn); #endif - /* - * Find and reserve possible boot-time SMP configuration: - */ - find_smp_config(); - reserve_crashkernel(); + initmem_init(0, max_pfn, acpi, k8); +#ifndef CONFIG_NO_BOOTMEM + early_res_to_bootmem(0, max_low_pfn<<PAGE_SHIFT); +#endif -#ifdef CONFIG_X86_64 - /* - * dma32_reserve_bootmem() allocates bootmem which may conflict - * with the crashkernel command line, so do that after - * reserve_crashkernel() - */ dma32_reserve_bootmem(); -#endif reserve_ibft_region(); @@ -1024,6 +1038,8 @@ void __init setup_arch(char **cmdline_p) #endif #endif x86_init.oem.banner(); + + mcheck_init(); } #ifdef CONFIG_X86_32 diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index d559af913e1f..ef6370b00e70 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -1,3 +1,5 @@ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> @@ -20,9 +22,9 @@ #include <asm/stackprotector.h> #ifdef CONFIG_DEBUG_PER_CPU_MAPS -# define DBG(x...) printk(KERN_DEBUG x) +# define DBG(fmt, ...) pr_dbg(fmt, ##__VA_ARGS__) #else -# define DBG(x...) +# define DBG(fmt, ...) do { if (0) pr_dbg(fmt, ##__VA_ARGS__); } while (0) #endif DEFINE_PER_CPU(int, cpu_number); @@ -116,8 +118,8 @@ static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size, } else { ptr = __alloc_bootmem_node_nopanic(NODE_DATA(node), size, align, goal); - pr_debug("per cpu data for cpu%d %lu bytes on node%d at " - "%016lx\n", cpu, size, node, __pa(ptr)); + pr_debug("per cpu data for cpu%d %lu bytes on node%d at %016lx\n", + cpu, size, node, __pa(ptr)); } return ptr; #else @@ -135,7 +137,13 @@ 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) @@ -198,8 +206,7 @@ void __init setup_per_cpu_areas(void) pcpu_cpu_distance, pcpu_fc_alloc, pcpu_fc_free); if (rc < 0) - pr_warning("PERCPU: %s allocator failed (%d), " - "falling back to page size\n", + pr_warning("%s allocator failed (%d), falling back to page size\n", pcpu_fc_names[pcpu_chosen_fc], rc); } if (rc < 0) diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c index 6a44a76055ad..4fd173cd8e57 100644 --- a/arch/x86/kernel/signal.c +++ b/arch/x86/kernel/signal.c @@ -19,6 +19,7 @@ #include <linux/stddef.h> #include <linux/personality.h> #include <linux/uaccess.h> +#include <linux/user-return-notifier.h> #include <asm/processor.h> #include <asm/ucontext.h> @@ -544,22 +545,12 @@ sys_sigaction(int sig, const struct old_sigaction __user *act, } #endif /* CONFIG_X86_32 */ -#ifdef CONFIG_X86_32 -int sys_sigaltstack(struct pt_regs *regs) -{ - const stack_t __user *uss = (const stack_t __user *)regs->bx; - stack_t __user *uoss = (stack_t __user *)regs->cx; - - return do_sigaltstack(uss, uoss, regs->sp); -} -#else /* !CONFIG_X86_32 */ -asmlinkage long +long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, struct pt_regs *regs) { return do_sigaltstack(uss, uoss, regs->sp); } -#endif /* CONFIG_X86_32 */ /* * Do a signal return; undo the signal stack. @@ -799,15 +790,6 @@ static void do_signal(struct pt_regs *regs) signr = get_signal_to_deliver(&info, &ka, regs, NULL); if (signr > 0) { - /* - * Re-enable any watchpoints before delivering the - * signal to user space. The processor register will - * have been cleared if the watchpoint triggered - * inside the kernel. - */ - if (current->thread.debugreg7) - set_debugreg(current->thread.debugreg7, 7); - /* Whee! Actually deliver the signal. */ if (handle_signal(signr, &info, &ka, oldset, regs) == 0) { /* @@ -872,6 +854,8 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) if (current->replacement_session_keyring) key_replace_session_keyring(); } + if (thread_info_flags & _TIF_USER_RETURN_NOTIFY) + fire_user_return_notifiers(); #ifdef CONFIG_X86_32 clear_thread_flag(TIF_IRET); diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 565ebc65920e..a02e80c3c54b 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -48,6 +48,7 @@ #include <linux/err.h> #include <linux/nmi.h> #include <linux/tboot.h> +#include <linux/stackprotector.h> #include <asm/acpi.h> #include <asm/desc.h> @@ -67,6 +68,7 @@ #include <linux/mc146818rtc.h> #include <asm/smpboot_hooks.h> +#include <asm/i8259.h> #ifdef CONFIG_X86_32 u8 apicid_2_node[MAX_APICID]; @@ -241,6 +243,11 @@ static void __cpuinit smp_callin(void) map_cpu_to_logical_apicid(); notify_cpu_starting(cpuid); + + /* + * Need to setup vector mappings before we enable interrupts. + */ + __setup_vector_irq(smp_processor_id()); /* * Get our bogomips. * @@ -286,9 +293,9 @@ notrace static void __cpuinit start_secondary(void *unused) check_tsc_sync_target(); if (nmi_watchdog == NMI_IO_APIC) { - disable_8259A_irq(0); + legacy_pic->chip->mask(0); enable_NMI_through_LVT0(); - enable_8259A_irq(0); + legacy_pic->chip->unmask(0); } #ifdef CONFIG_X86_32 @@ -315,15 +322,18 @@ notrace static void __cpuinit start_secondary(void *unused) */ ipi_call_lock(); lock_vector_lock(); - __setup_vector_irq(smp_processor_id()); set_cpu_online(smp_processor_id(), true); unlock_vector_lock(); ipi_call_unlock(); per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE; + x86_platform.nmi_init(); /* enable local interrupts */ local_irq_enable(); + /* to prevent fake stack check failure in clock setup */ + boot_init_stack_canary(); + x86_cpuinit.setup_percpu_clockev(); wmb(); @@ -671,6 +681,26 @@ static void __cpuinit do_fork_idle(struct work_struct *work) complete(&c_idle->done); } +/* reduce the number of lines printed when booting a large cpu count system */ +static void __cpuinit announce_cpu(int cpu, int apicid) +{ + static int current_node = -1; + int node = cpu_to_node(cpu); + + if (system_state == SYSTEM_BOOTING) { + if (node != current_node) { + if (current_node > (-1)) + pr_cont(" Ok.\n"); + current_node = node; + pr_info("Booting Node %3d, Processors ", node); + } + pr_cont(" #%d%s", cpu, cpu == (nr_cpu_ids - 1) ? " Ok.\n" : ""); + return; + } else + pr_info("Booting Node %d Processor %d APIC 0x%x\n", + node, cpu, apicid); +} + /* * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad * (ie clustered apic addressing mode), this is a LOGICAL apic ID. @@ -687,7 +717,7 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu) .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done), }; - INIT_WORK(&c_idle.work, do_fork_idle); + INIT_WORK_ON_STACK(&c_idle.work, do_fork_idle); alternatives_smp_switch(1); @@ -713,6 +743,7 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu) if (IS_ERR(c_idle.idle)) { printk("failed fork for CPU %d\n", cpu); + destroy_work_on_stack(&c_idle.work); return PTR_ERR(c_idle.idle); } @@ -736,9 +767,8 @@ do_rest: /* start_ip had better be page-aligned! */ start_ip = setup_trampoline(); - /* So we see what's up */ - printk(KERN_INFO "Booting processor %d APIC 0x%x ip 0x%lx\n", - cpu, apicid, start_ip); + /* So we see what's up */ + announce_cpu(cpu, apicid); /* * This grunge runs the startup process for @@ -787,21 +817,17 @@ do_rest: udelay(100); } - if (cpumask_test_cpu(cpu, cpu_callin_mask)) { - /* number CPUs logically, starting from 1 (BSP is 0) */ - pr_debug("OK.\n"); - printk(KERN_INFO "CPU%d: ", cpu); - print_cpu_info(&cpu_data(cpu)); - pr_debug("CPU has booted.\n"); - } else { + if (cpumask_test_cpu(cpu, cpu_callin_mask)) + pr_debug("CPU%d: has booted.\n", cpu); + else { boot_error = 1; if (*((volatile unsigned char *)trampoline_base) == 0xA5) /* trampoline started but...? */ - printk(KERN_ERR "Stuck ??\n"); + pr_err("CPU%d: Stuck ??\n", cpu); else /* trampoline code not run */ - printk(KERN_ERR "Not responding.\n"); + pr_err("CPU%d: Not responding.\n", cpu); if (apic->inquire_remote_apic) apic->inquire_remote_apic(apicid); } @@ -831,6 +857,7 @@ do_rest: smpboot_restore_warm_reset_vector(); } + destroy_work_on_stack(&c_idle.work); return boot_error; } @@ -1066,9 +1093,7 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) set_cpu_sibling_map(0); enable_IR_x2apic(); -#ifdef CONFIG_X86_64 default_setup_apic_routing(); -#endif if (smp_sanity_check(max_cpus) < 0) { printk(KERN_INFO "SMP disabled\n"); @@ -1196,11 +1221,12 @@ __init void prefill_possible_map(void) total_cpus = max_t(int, possible, num_processors + disabled_cpus); - if (possible > CONFIG_NR_CPUS) { + /* nr_cpu_ids could be reduced via nr_cpus= */ + if (possible > nr_cpu_ids) { printk(KERN_WARNING "%d Processors exceeds NR_CPUS limit of %d\n", - possible, CONFIG_NR_CPUS); - possible = CONFIG_NR_CPUS; + possible, nr_cpu_ids); + possible = nr_cpu_ids; } printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n", @@ -1250,16 +1276,7 @@ static void __ref remove_cpu_from_maps(int cpu) void cpu_disable_common(void) { int cpu = smp_processor_id(); - /* - * HACK: - * Allow any queued timer interrupts to get serviced - * This is only a temporary solution until we cleanup - * fixup_irqs as we do for IA64. - */ - local_irq_enable(); - mdelay(1); - local_irq_disable(); remove_siblinginfo(cpu); /* It's now safe to remove this processor from the online map */ @@ -1300,14 +1317,16 @@ void native_cpu_die(unsigned int cpu) for (i = 0; i < 10; i++) { /* They ack this in play_dead by setting CPU_DEAD */ if (per_cpu(cpu_state, cpu) == CPU_DEAD) { - printk(KERN_INFO "CPU %d is now offline\n", cpu); + if (system_state == SYSTEM_RUNNING) + pr_info("CPU %u is now offline\n", cpu); + if (1 == num_online_cpus()) alternatives_smp_switch(0); return; } msleep(100); } - printk(KERN_ERR "CPU %u didn't die...\n", cpu); + pr_err("CPU %u didn't die...\n", cpu); } void play_dead_common(void) diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c index c3eb207181fe..922eefbb3f6c 100644 --- a/arch/x86/kernel/stacktrace.c +++ b/arch/x86/kernel/stacktrace.c @@ -53,17 +53,19 @@ save_stack_address_nosched(void *data, unsigned long addr, int reliable) } static const struct stacktrace_ops save_stack_ops = { - .warning = save_stack_warning, - .warning_symbol = save_stack_warning_symbol, - .stack = save_stack_stack, - .address = save_stack_address, + .warning = save_stack_warning, + .warning_symbol = save_stack_warning_symbol, + .stack = save_stack_stack, + .address = save_stack_address, + .walk_stack = print_context_stack, }; static const struct stacktrace_ops save_stack_ops_nosched = { - .warning = save_stack_warning, - .warning_symbol = save_stack_warning_symbol, - .stack = save_stack_stack, - .address = save_stack_address_nosched, + .warning = save_stack_warning, + .warning_symbol = save_stack_warning_symbol, + .stack = save_stack_stack, + .address = save_stack_address_nosched, + .walk_stack = print_context_stack, }; /* diff --git a/arch/x86/kernel/sys_i386_32.c b/arch/x86/kernel/sys_i386_32.c index 1884a8d12bfa..196552bb412c 100644 --- a/arch/x86/kernel/sys_i386_32.c +++ b/arch/x86/kernel/sys_i386_32.c @@ -24,216 +24,6 @@ #include <asm/syscalls.h> -asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, - unsigned long prot, unsigned long flags, - unsigned long fd, unsigned long pgoff) -{ - int error = -EBADF; - struct file *file = NULL; - struct mm_struct *mm = current->mm; - - flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); - if (!(flags & MAP_ANONYMOUS)) { - file = fget(fd); - if (!file) - goto out; - } - - down_write(&mm->mmap_sem); - error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff); - up_write(&mm->mmap_sem); - - if (file) - fput(file); -out: - return error; -} - -/* - * Perform the select(nd, in, out, ex, tv) and mmap() system - * calls. Linux/i386 didn't use to be able to handle more than - * 4 system call parameters, so these system calls used a memory - * block for parameter passing.. - */ - -struct mmap_arg_struct { - unsigned long addr; - unsigned long len; - unsigned long prot; - unsigned long flags; - unsigned long fd; - unsigned long offset; -}; - -asmlinkage int old_mmap(struct mmap_arg_struct __user *arg) -{ - struct mmap_arg_struct a; - int err = -EFAULT; - - if (copy_from_user(&a, arg, sizeof(a))) - goto out; - - err = -EINVAL; - if (a.offset & ~PAGE_MASK) - goto out; - - err = sys_mmap2(a.addr, a.len, a.prot, a.flags, - a.fd, a.offset >> PAGE_SHIFT); -out: - return err; -} - - -struct sel_arg_struct { - unsigned long n; - fd_set __user *inp, *outp, *exp; - struct timeval __user *tvp; -}; - -asmlinkage int old_select(struct sel_arg_struct __user *arg) -{ - struct sel_arg_struct a; - - if (copy_from_user(&a, arg, sizeof(a))) - return -EFAULT; - /* sys_select() does the appropriate kernel locking */ - return sys_select(a.n, a.inp, a.outp, a.exp, a.tvp); -} - -/* - * sys_ipc() is the de-multiplexer for the SysV IPC calls.. - * - * This is really horribly ugly. - */ -asmlinkage int sys_ipc(uint call, int first, int second, - int third, void __user *ptr, long fifth) -{ - int version, ret; - - version = call >> 16; /* hack for backward compatibility */ - call &= 0xffff; - - switch (call) { - case SEMOP: - return sys_semtimedop(first, (struct sembuf __user *)ptr, second, NULL); - case SEMTIMEDOP: - return sys_semtimedop(first, (struct sembuf __user *)ptr, second, - (const struct timespec __user *)fifth); - - case SEMGET: - return sys_semget(first, second, third); - case SEMCTL: { - union semun fourth; - if (!ptr) - return -EINVAL; - if (get_user(fourth.__pad, (void __user * __user *) ptr)) - return -EFAULT; - return sys_semctl(first, second, third, fourth); - } - - case MSGSND: - return sys_msgsnd(first, (struct msgbuf __user *) ptr, - second, third); - case MSGRCV: - switch (version) { - case 0: { - struct ipc_kludge tmp; - if (!ptr) - return -EINVAL; - - if (copy_from_user(&tmp, - (struct ipc_kludge __user *) ptr, - sizeof(tmp))) - return -EFAULT; - return sys_msgrcv(first, tmp.msgp, second, - tmp.msgtyp, third); - } - default: - return sys_msgrcv(first, - (struct msgbuf __user *) ptr, - second, fifth, third); - } - case MSGGET: - return sys_msgget((key_t) first, second); - case MSGCTL: - return sys_msgctl(first, second, (struct msqid_ds __user *) ptr); - - case SHMAT: - switch (version) { - default: { - ulong raddr; - ret = do_shmat(first, (char __user *) ptr, second, &raddr); - if (ret) - return ret; - return put_user(raddr, (ulong __user *) third); - } - case 1: /* iBCS2 emulator entry point */ - if (!segment_eq(get_fs(), get_ds())) - return -EINVAL; - /* The "(ulong *) third" is valid _only_ because of the kernel segment thing */ - return do_shmat(first, (char __user *) ptr, second, (ulong *) third); - } - case SHMDT: - return sys_shmdt((char __user *)ptr); - case SHMGET: - return sys_shmget(first, second, third); - case SHMCTL: - return sys_shmctl(first, second, - (struct shmid_ds __user *) ptr); - default: - return -ENOSYS; - } -} - -/* - * Old cruft - */ -asmlinkage int sys_uname(struct old_utsname __user *name) -{ - int err; - if (!name) - return -EFAULT; - down_read(&uts_sem); - err = copy_to_user(name, utsname(), sizeof(*name)); - up_read(&uts_sem); - return err? -EFAULT:0; -} - -asmlinkage int sys_olduname(struct oldold_utsname __user *name) -{ - int error; - - if (!name) - return -EFAULT; - if (!access_ok(VERIFY_WRITE, name, sizeof(struct oldold_utsname))) - return -EFAULT; - - down_read(&uts_sem); - - error = __copy_to_user(&name->sysname, &utsname()->sysname, - __OLD_UTS_LEN); - error |= __put_user(0, name->sysname + __OLD_UTS_LEN); - error |= __copy_to_user(&name->nodename, &utsname()->nodename, - __OLD_UTS_LEN); - error |= __put_user(0, name->nodename + __OLD_UTS_LEN); - error |= __copy_to_user(&name->release, &utsname()->release, - __OLD_UTS_LEN); - error |= __put_user(0, name->release + __OLD_UTS_LEN); - error |= __copy_to_user(&name->version, &utsname()->version, - __OLD_UTS_LEN); - error |= __put_user(0, name->version + __OLD_UTS_LEN); - error |= __copy_to_user(&name->machine, &utsname()->machine, - __OLD_UTS_LEN); - error |= __put_user(0, name->machine + __OLD_UTS_LEN); - - up_read(&uts_sem); - - error = error ? -EFAULT : 0; - - return error; -} - - /* * Do a system call from kernel instead of calling sys_execve so we * end up with proper pt_regs. diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c index 45e00eb09c3a..ff14a5044ce6 100644 --- a/arch/x86/kernel/sys_x86_64.c +++ b/arch/x86/kernel/sys_x86_64.c @@ -23,26 +23,11 @@ SYSCALL_DEFINE6(mmap, unsigned long, addr, unsigned long, len, unsigned long, fd, unsigned long, off) { long error; - struct file *file; - error = -EINVAL; if (off & ~PAGE_MASK) goto out; - error = -EBADF; - file = NULL; - flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); - if (!(flags & MAP_ANONYMOUS)) { - file = fget(fd); - if (!file) - goto out; - } - down_write(¤t->mm->mmap_sem); - error = do_mmap_pgoff(file, addr, len, prot, flags, off >> PAGE_SHIFT); - up_write(¤t->mm->mmap_sem); - - if (file) - fput(file); + error = sys_mmap_pgoff(addr, len, prot, flags, fd, off >> PAGE_SHIFT); out: return error; } @@ -224,15 +209,3 @@ bottomup: return addr; } - - -SYSCALL_DEFINE1(uname, struct new_utsname __user *, name) -{ - int err; - down_read(&uts_sem); - err = copy_to_user(name, utsname(), sizeof(*name)); - up_read(&uts_sem); - if (personality(current->personality) == PER_LINUX32) - err |= copy_to_user(&name->machine, "i686", 5); - return err ? -EFAULT : 0; -} diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index 0157cd26d7cc..8b3729341216 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -81,7 +81,7 @@ ENTRY(sys_call_table) .long sys_settimeofday .long sys_getgroups16 /* 80 */ .long sys_setgroups16 - .long old_select + .long sys_old_select .long sys_symlink .long sys_lstat .long sys_readlink /* 85 */ @@ -89,7 +89,7 @@ ENTRY(sys_call_table) .long sys_swapon .long sys_reboot .long sys_old_readdir - .long old_mmap /* 90 */ + .long sys_old_mmap /* 90 */ .long sys_munmap .long sys_truncate .long sys_ftruncate @@ -191,7 +191,7 @@ ENTRY(sys_call_table) .long sys_ni_syscall /* reserved for streams2 */ .long ptregs_vfork /* 190 */ .long sys_getrlimit - .long sys_mmap2 + .long sys_mmap_pgoff .long sys_truncate64 .long sys_ftruncate64 .long sys_stat64 /* 195 */ @@ -336,3 +336,4 @@ ENTRY(sys_call_table) .long sys_pwritev .long sys_rt_tgsigqueueinfo /* 335 */ .long sys_perf_event_open + .long sys_recvmmsg diff --git a/arch/x86/kernel/time.c b/arch/x86/kernel/time.c index dcb00d278512..fb5cc5e14cfa 100644 --- a/arch/x86/kernel/time.c +++ b/arch/x86/kernel/time.c @@ -38,7 +38,8 @@ unsigned long profile_pc(struct pt_regs *regs) #ifdef CONFIG_FRAME_POINTER return *(unsigned long *)(regs->bp + sizeof(long)); #else - unsigned long *sp = (unsigned long *)regs->sp; + unsigned long *sp = + (unsigned long *)kernel_stack_pointer(regs); /* * Return address is either directly at stack pointer * or above a saved flags. Eflags has bits 22-31 zero, @@ -69,11 +70,11 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id) * manually to deassert NMI lines for the watchdog if run * on an 82489DX-based system. */ - spin_lock(&i8259A_lock); + raw_spin_lock(&i8259A_lock); outb(0x0c, PIC_MASTER_OCW3); /* Ack the IRQ; AEOI will end it automatically. */ inb(PIC_MASTER_POLL); - spin_unlock(&i8259A_lock); + raw_spin_unlock(&i8259A_lock); } global_clock_event->event_handler(global_clock_event); diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 503c1f2e8835..364d015efebc 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -23,8 +23,6 @@ static struct bau_control **uv_bau_table_bases __read_mostly; static int uv_bau_retry_limit __read_mostly; -/* position of pnode (which is nasid>>1): */ -static int uv_nshift __read_mostly; /* base pnode in this partition */ static int uv_partition_base_pnode __read_mostly; @@ -723,7 +721,7 @@ uv_activation_descriptor_init(int node, int pnode) BUG_ON(!adp); pa = uv_gpa(adp); /* need the real nasid*/ - n = pa >> uv_nshift; + n = uv_gpa_to_pnode(pa); m = pa & uv_mmask; uv_write_global_mmr64(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE, @@ -778,7 +776,7 @@ uv_payload_queue_init(int node, int pnode, struct bau_control *bau_tablesp) * need the pnode of where the memory was really allocated */ pa = uv_gpa(pqp); - pn = pa >> uv_nshift; + pn = uv_gpa_to_pnode(pa); uv_write_global_mmr64(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST, ((unsigned long)pn << UV_PAYLOADQ_PNODE_SHIFT) | @@ -819,10 +817,8 @@ static int __init uv_init_blade(int blade) */ apicid = blade_to_first_apicid(blade); pa = uv_read_global_mmr64(pnode, UVH_BAU_DATA_CONFIG); - if ((pa & 0xff) != UV_BAU_MESSAGE) { - uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG, + uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG, ((apicid << 32) | UV_BAU_MESSAGE)); - } return 0; } @@ -843,8 +839,7 @@ static int __init uv_bau_init(void) GFP_KERNEL, cpu_to_node(cur_cpu)); uv_bau_retry_limit = 1; - uv_nshift = uv_hub_info->n_val; - uv_mmask = (1UL << uv_hub_info->n_val) - 1; + uv_mmask = (1UL << uv_hub_info->m_val) - 1; nblades = uv_num_possible_blades(); uv_bau_table_bases = (struct bau_control **) diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c index 699f7eeb896a..c652ef62742d 100644 --- a/arch/x86/kernel/trampoline.c +++ b/arch/x86/kernel/trampoline.c @@ -3,22 +3,28 @@ #include <asm/trampoline.h> #include <asm/e820.h> +#if defined(CONFIG_X86_64) && defined(CONFIG_ACPI_SLEEP) +#define __trampinit +#define __trampinitdata +#else +#define __trampinit __cpuinit +#define __trampinitdata __cpuinitdata +#endif + /* ready for x86_64 and x86 */ -unsigned char *__cpuinitdata trampoline_base = __va(TRAMPOLINE_BASE); +unsigned char *__trampinitdata trampoline_base; void __init reserve_trampoline_memory(void) { -#ifdef CONFIG_X86_32 - /* - * But first pinch a few for the stack/trampoline stuff - * FIXME: Don't need the extra page at 4K, but need to fix - * trampoline before removing it. (see the GDT stuff) - */ - reserve_early(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE"); -#endif + unsigned long mem; + /* Has to be in very low memory so we can execute real-mode AP code. */ - reserve_early(TRAMPOLINE_BASE, TRAMPOLINE_BASE + TRAMPOLINE_SIZE, - "TRAMPOLINE"); + mem = find_e820_area(0, 1<<20, TRAMPOLINE_SIZE, PAGE_SIZE); + if (mem == -1L) + panic("Cannot allocate trampoline\n"); + + trampoline_base = __va(mem); + reserve_early(mem, mem + TRAMPOLINE_SIZE, "TRAMPOLINE"); } /* @@ -26,7 +32,7 @@ void __init reserve_trampoline_memory(void) * bootstrap into the page concerned. The caller * has made sure it's suitably aligned. */ -unsigned long __cpuinit setup_trampoline(void) +unsigned long __trampinit setup_trampoline(void) { memcpy(trampoline_base, trampoline_data, TRAMPOLINE_SIZE); return virt_to_phys(trampoline_base); diff --git a/arch/x86/kernel/trampoline_64.S b/arch/x86/kernel/trampoline_64.S index 596d54c660a5..3af2dff58b21 100644 --- a/arch/x86/kernel/trampoline_64.S +++ b/arch/x86/kernel/trampoline_64.S @@ -32,8 +32,12 @@ #include <asm/segment.h> #include <asm/processor-flags.h> +#ifdef CONFIG_ACPI_SLEEP +.section .rodata, "a", @progbits +#else /* We can free up the trampoline after bootup if cpu hotplug is not supported. */ __CPUINITRODATA +#endif .code16 ENTRY(trampoline_data) diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 7e37dcee0cc3..1168e4454188 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -529,77 +529,59 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) { struct task_struct *tsk = current; - unsigned long condition; + unsigned long dr6; int si_code; - get_debugreg(condition, 6); + get_debugreg(dr6, 6); + + /* Filter out all the reserved bits which are preset to 1 */ + dr6 &= ~DR6_RESERVED; /* Catch kmemcheck conditions first of all! */ - if (condition & DR_STEP && kmemcheck_trap(regs)) + if ((dr6 & DR_STEP) && kmemcheck_trap(regs)) return; + /* DR6 may or may not be cleared by the CPU */ + set_debugreg(0, 6); /* * The processor cleared BTF, so don't mark that we need it set. */ clear_tsk_thread_flag(tsk, TIF_DEBUGCTLMSR); tsk->thread.debugctlmsr = 0; - if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code, - SIGTRAP) == NOTIFY_STOP) + /* Store the virtualized DR6 value */ + tsk->thread.debugreg6 = dr6; + + if (notify_die(DIE_DEBUG, "debug", regs, PTR_ERR(&dr6), error_code, + SIGTRAP) == NOTIFY_STOP) return; /* It's safe to allow irq's after DR6 has been saved */ preempt_conditional_sti(regs); - /* Mask out spurious debug traps due to lazy DR7 setting */ - if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) { - if (!tsk->thread.debugreg7) - goto clear_dr7; + if (regs->flags & X86_VM_MASK) { + handle_vm86_trap((struct kernel_vm86_regs *) regs, + error_code, 1); + return; } -#ifdef CONFIG_X86_32 - if (regs->flags & X86_VM_MASK) - goto debug_vm86; -#endif - - /* Save debug status register where ptrace can see it */ - tsk->thread.debugreg6 = condition; - /* - * Single-stepping through TF: make sure we ignore any events in - * kernel space (but re-enable TF when returning to user mode). + * Single-stepping through system calls: ignore any exceptions in + * kernel space, but re-enable TF when returning to user mode. + * + * We already checked v86 mode above, so we can check for kernel mode + * by just checking the CPL of CS. */ - if (condition & DR_STEP) { - if (!user_mode(regs)) - goto clear_TF_reenable; + if ((dr6 & DR_STEP) && !user_mode(regs)) { + tsk->thread.debugreg6 &= ~DR_STEP; + set_tsk_thread_flag(tsk, TIF_SINGLESTEP); + regs->flags &= ~X86_EFLAGS_TF; } - - si_code = get_si_code(condition); - /* Ok, finally something we can handle */ - send_sigtrap(tsk, regs, error_code, si_code); - - /* - * Disable additional traps. They'll be re-enabled when - * the signal is delivered. - */ -clear_dr7: - set_debugreg(0, 7); + si_code = get_si_code(tsk->thread.debugreg6); + if (tsk->thread.debugreg6 & (DR_STEP | DR_TRAP_BITS)) + send_sigtrap(tsk, regs, error_code, si_code); preempt_conditional_cli(regs); - return; - -#ifdef CONFIG_X86_32 -debug_vm86: - /* reenable preemption: handle_vm86_trap() might sleep */ - dec_preempt_count(); - handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1); - conditional_cli(regs); - return; -#endif -clear_TF_reenable: - set_tsk_thread_flag(tsk, TIF_SINGLESTEP); - regs->flags &= ~X86_EFLAGS_TF; - preempt_conditional_cli(regs); return; } diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index cd982f48e23e..9faf91ae1841 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -50,7 +50,7 @@ u64 native_sched_clock(void) * unstable. We do this because unlike Time Of Day, * the scheduler clock tolerates small errors and it's * very important for it to be as fast as the platform - * can achive it. ) + * can achieve it. ) */ if (unlikely(tsc_disabled)) { /* No locking but a rare wrong value is not a big deal: */ @@ -740,7 +740,7 @@ static cycle_t __vsyscall_fn vread_tsc(void) } #endif -static void resume_tsc(void) +static void resume_tsc(struct clocksource *cs) { clocksource_tsc.cycle_last = 0; } @@ -763,6 +763,7 @@ void mark_tsc_unstable(char *reason) { if (!tsc_unstable) { tsc_unstable = 1; + sched_clock_stable = 0; printk(KERN_INFO "Marking TSC unstable due to %s\n", reason); /* Change only the rating, when not registered */ if (clocksource_tsc.mult) @@ -805,7 +806,7 @@ static void __init check_system_tsc_reliable(void) unsigned long res_low, res_high; rdmsr_safe(MSR_GEODE_BUSCONT_CONF0, &res_low, &res_high); - /* Geode_LX - the OLPC CPU has a possibly a very reliable TSC */ + /* Geode_LX - the OLPC CPU has a very reliable TSC */ if (res_low & RTSC_SUSP) tsc_clocksource_reliable = 1; #endif diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c index f37930954d15..0aa5fed8b9e6 100644 --- a/arch/x86/kernel/tsc_sync.c +++ b/arch/x86/kernel/tsc_sync.c @@ -33,7 +33,7 @@ static __cpuinitdata atomic_t stop_count; * we want to have the fastest, inlined, non-debug version * of a critical section, to be able to prove TSC time-warps: */ -static __cpuinitdata raw_spinlock_t sync_lock = __RAW_SPIN_LOCK_UNLOCKED; +static __cpuinitdata arch_spinlock_t sync_lock = __ARCH_SPIN_LOCK_UNLOCKED; static __cpuinitdata cycles_t last_tsc; static __cpuinitdata cycles_t max_warp; @@ -62,13 +62,13 @@ static __cpuinit void check_tsc_warp(void) * previous TSC that was measured (possibly on * another CPU) and update the previous TSC timestamp. */ - __raw_spin_lock(&sync_lock); + arch_spin_lock(&sync_lock); prev = last_tsc; rdtsc_barrier(); now = get_cycles(); rdtsc_barrier(); last_tsc = now; - __raw_spin_unlock(&sync_lock); + arch_spin_unlock(&sync_lock); /* * Be nice every now and then (and also check whether @@ -87,10 +87,10 @@ static __cpuinit void check_tsc_warp(void) * we saw a time-warp of the TSC going backwards: */ if (unlikely(prev > now)) { - __raw_spin_lock(&sync_lock); + arch_spin_lock(&sync_lock); max_warp = max(max_warp, prev - now); nr_warps++; - __raw_spin_unlock(&sync_lock); + arch_spin_unlock(&sync_lock); } } WARN(!(now-start), @@ -114,13 +114,12 @@ void __cpuinit check_tsc_sync_source(int cpu) return; if (boot_cpu_has(X86_FEATURE_TSC_RELIABLE)) { - printk_once(KERN_INFO "Skipping synchronization checks as TSC is reliable.\n"); + if (cpu == (nr_cpu_ids-1) || system_state != SYSTEM_BOOTING) + pr_info( + "Skipped synchronization checks as TSC is reliable.\n"); return; } - pr_info("checking TSC synchronization [CPU#%d -> CPU#%d]:", - smp_processor_id(), cpu); - /* * Reset it - in case this is a second bootup: */ @@ -142,12 +141,14 @@ void __cpuinit check_tsc_sync_source(int cpu) cpu_relax(); if (nr_warps) { - printk("\n"); + pr_warning("TSC synchronization [CPU#%d -> CPU#%d]:\n", + smp_processor_id(), cpu); pr_warning("Measured %Ld cycles TSC warp between CPUs, " "turning off TSC clock.\n", max_warp); mark_tsc_unstable("check_tsc_sync_source failed"); } else { - printk(" passed.\n"); + pr_debug("TSC synchronization [CPU#%d -> CPU#%d]: passed\n", + smp_processor_id(), cpu); } /* diff --git a/arch/x86/kernel/uv_irq.c b/arch/x86/kernel/uv_irq.c index aeef529917e4..ece73d8e3240 100644 --- a/arch/x86/kernel/uv_irq.c +++ b/arch/x86/kernel/uv_irq.c @@ -9,10 +9,25 @@ */ #include <linux/module.h> +#include <linux/rbtree.h> #include <linux/irq.h> #include <asm/apic.h> #include <asm/uv/uv_irq.h> +#include <asm/uv/uv_hub.h> + +/* MMR offset and pnode of hub sourcing interrupts for a given irq */ +struct uv_irq_2_mmr_pnode{ + struct rb_node list; + unsigned long offset; + int pnode; + int irq; +}; + +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 void uv_noop(unsigned int irq) { @@ -39,25 +54,213 @@ struct irq_chip uv_irq_chip = { .unmask = uv_noop, .eoi = uv_ack_apic, .end = uv_noop, + .set_affinity = uv_set_irq_affinity, }; /* + * Add offset and pnode information of the hub sourcing interrupts to the + * rb tree for a specific irq. + */ +static int uv_set_irq_2_mmr_info(int irq, unsigned long offset, unsigned blade) +{ + struct rb_node **link = &uv_irq_root.rb_node; + struct rb_node *parent = NULL; + struct uv_irq_2_mmr_pnode *n; + struct uv_irq_2_mmr_pnode *e; + unsigned long irqflags; + + n = kmalloc_node(sizeof(struct uv_irq_2_mmr_pnode), GFP_KERNEL, + uv_blade_to_memory_nid(blade)); + if (!n) + return -ENOMEM; + + n->irq = irq; + n->offset = offset; + n->pnode = uv_blade_to_pnode(blade); + spin_lock_irqsave(&uv_irq_lock, irqflags); + /* Find the right place in the rbtree: */ + while (*link) { + parent = *link; + e = rb_entry(parent, struct uv_irq_2_mmr_pnode, list); + + if (unlikely(irq == e->irq)) { + /* irq entry exists */ + e->pnode = uv_blade_to_pnode(blade); + e->offset = offset; + spin_unlock_irqrestore(&uv_irq_lock, irqflags); + kfree(n); + return 0; + } + + if (irq < e->irq) + link = &(*link)->rb_left; + else + link = &(*link)->rb_right; + } + + /* Insert the node into the rbtree. */ + rb_link_node(&n->list, parent, link); + rb_insert_color(&n->list, &uv_irq_root); + + spin_unlock_irqrestore(&uv_irq_lock, irqflags); + return 0; +} + +/* Retrieve offset and pnode information from the rb tree for a specific irq */ +int uv_irq_2_mmr_info(int irq, unsigned long *offset, int *pnode) +{ + struct uv_irq_2_mmr_pnode *e; + struct rb_node *n; + unsigned long irqflags; + + spin_lock_irqsave(&uv_irq_lock, irqflags); + n = uv_irq_root.rb_node; + while (n) { + e = rb_entry(n, struct uv_irq_2_mmr_pnode, list); + + if (e->irq == irq) { + *offset = e->offset; + *pnode = e->pnode; + spin_unlock_irqrestore(&uv_irq_lock, irqflags); + return 0; + } + + if (irq < e->irq) + n = n->rb_left; + else + n = n->rb_right; + } + spin_unlock_irqrestore(&uv_irq_lock, irqflags); + return -1; +} + +/* + * Re-target the irq to the specified CPU and enable the specified MMR located + * on the specified blade to allow the sending of MSIs to the specified CPU. + */ +static int +arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, + unsigned long mmr_offset, int restrict) +{ + const struct cpumask *eligible_cpu = cpumask_of(cpu); + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; + int mmr_pnode; + unsigned long mmr_value; + struct uv_IO_APIC_route_entry *entry; + int 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 (restrict == UV_AFFINITY_CPU) + desc->status |= IRQ_NO_BALANCING; + else + desc->status |= IRQ_MOVE_PCNTXT; + + set_irq_chip_and_handler_name(irq, &uv_irq_chip, handle_percpu_irq, + irq_name); + + mmr_value = 0; + entry = (struct uv_IO_APIC_route_entry *)&mmr_value; + entry->vector = cfg->vector; + entry->delivery_mode = apic->irq_delivery_mode; + entry->dest_mode = apic->irq_dest_mode; + entry->polarity = 0; + entry->trigger = 0; + entry->mask = 0; + entry->dest = apic->cpu_mask_to_apicid(eligible_cpu); + + mmr_pnode = uv_blade_to_pnode(mmr_blade); + uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); + + if (cfg->move_in_progress) + send_cleanup_vector(cfg); + + return irq; +} + +/* + * Disable the specified MMR located on the specified blade so that MSIs are + * longer allowed to be sent. + */ +static void arch_disable_uv_irq(int mmr_pnode, unsigned long mmr_offset) +{ + unsigned long mmr_value; + struct uv_IO_APIC_route_entry *entry; + + BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != + sizeof(unsigned long)); + + mmr_value = 0; + entry = (struct uv_IO_APIC_route_entry *)&mmr_value; + entry->mask = 1; + + uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); +} + +static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask) +{ + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg = desc->chip_data; + unsigned int dest; + unsigned long mmr_value; + struct uv_IO_APIC_route_entry *entry; + unsigned long mmr_offset; + unsigned mmr_pnode; + + if (set_desc_affinity(desc, mask, &dest)) + return -1; + + mmr_value = 0; + entry = (struct uv_IO_APIC_route_entry *)&mmr_value; + + entry->vector = cfg->vector; + entry->delivery_mode = apic->irq_delivery_mode; + entry->dest_mode = apic->irq_dest_mode; + entry->polarity = 0; + entry->trigger = 0; + entry->mask = 0; + entry->dest = dest; + + /* Get previously stored MMR and pnode of hub sourcing interrupts */ + if (uv_irq_2_mmr_info(irq, &mmr_offset, &mmr_pnode)) + return -1; + + uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); + + if (cfg->move_in_progress) + send_cleanup_vector(cfg); + + return 0; +} + +/* * Set up a mapping of an available irq and vector, and enable the specified * MMR that defines the MSI that is to be sent to the specified CPU when an * interrupt is raised. */ int uv_setup_irq(char *irq_name, int cpu, int mmr_blade, - unsigned long mmr_offset) + unsigned long mmr_offset, int restrict) { - int irq; - int ret; + int irq, ret; + + irq = create_irq_nr(NR_IRQS_LEGACY, uv_blade_to_memory_nid(mmr_blade)); - irq = create_irq(); if (irq <= 0) return -EBUSY; - ret = arch_enable_uv_irq(irq_name, irq, cpu, mmr_blade, mmr_offset); - if (ret != irq) + ret = arch_enable_uv_irq(irq_name, irq, cpu, mmr_blade, mmr_offset, + restrict); + if (ret == irq) + uv_set_irq_2_mmr_info(irq, mmr_offset, mmr_blade); + else destroy_irq(irq); return ret; @@ -71,9 +274,28 @@ EXPORT_SYMBOL_GPL(uv_setup_irq); * * Set mmr_blade and mmr_offset to what was passed in on uv_setup_irq(). */ -void uv_teardown_irq(unsigned int irq, int mmr_blade, unsigned long mmr_offset) +void uv_teardown_irq(unsigned int irq) { - arch_disable_uv_irq(mmr_blade, mmr_offset); + struct uv_irq_2_mmr_pnode *e; + struct rb_node *n; + unsigned long irqflags; + + spin_lock_irqsave(&uv_irq_lock, irqflags); + n = uv_irq_root.rb_node; + while (n) { + e = rb_entry(n, struct uv_irq_2_mmr_pnode, list); + if (e->irq == irq) { + arch_disable_uv_irq(e->pnode, e->offset); + rb_erase(n, &uv_irq_root); + kfree(e); + break; + } + if (irq < e->irq) + n = n->rb_left; + else + n = n->rb_right; + } + spin_unlock_irqrestore(&uv_irq_lock, irqflags); destroy_irq(irq); } EXPORT_SYMBOL_GPL(uv_teardown_irq); diff --git a/arch/x86/kernel/uv_sysfs.c b/arch/x86/kernel/uv_sysfs.c index 36afb98675a4..309c70fb7759 100644 --- a/arch/x86/kernel/uv_sysfs.c +++ b/arch/x86/kernel/uv_sysfs.c @@ -54,19 +54,19 @@ static int __init sgi_uv_sysfs_init(void) if (!sgi_uv_kobj) sgi_uv_kobj = kobject_create_and_add("sgi_uv", firmware_kobj); if (!sgi_uv_kobj) { - printk(KERN_WARNING "kobject_create_and_add sgi_uv failed \n"); + printk(KERN_WARNING "kobject_create_and_add sgi_uv failed\n"); return -EINVAL; } ret = sysfs_create_file(sgi_uv_kobj, &partition_id_attr.attr); if (ret) { - printk(KERN_WARNING "sysfs_create_file partition_id failed \n"); + printk(KERN_WARNING "sysfs_create_file partition_id failed\n"); return ret; } ret = sysfs_create_file(sgi_uv_kobj, &coherence_id_attr.attr); if (ret) { - printk(KERN_WARNING "sysfs_create_file coherence_id failed \n"); + printk(KERN_WARNING "sysfs_create_file coherence_id failed\n"); return ret; } diff --git a/arch/x86/kernel/uv_time.c b/arch/x86/kernel/uv_time.c index 583f11d5c480..2b75ef638dbc 100644 --- a/arch/x86/kernel/uv_time.c +++ b/arch/x86/kernel/uv_time.c @@ -74,7 +74,7 @@ struct uv_rtc_timer_head { */ static struct uv_rtc_timer_head **blade_info __read_mostly; -static int uv_rtc_enable; +static int uv_rtc_evt_enable; /* * Hardware interface routines @@ -90,7 +90,7 @@ static void uv_rtc_send_IPI(int cpu) pnode = uv_apicid_to_pnode(apicid); val = (1UL << UVH_IPI_INT_SEND_SHFT) | (apicid << UVH_IPI_INT_APIC_ID_SHFT) | - (GENERIC_INTERRUPT_VECTOR << UVH_IPI_INT_VECTOR_SHFT); + (X86_PLATFORM_IPI_VECTOR << UVH_IPI_INT_VECTOR_SHFT); uv_write_global_mmr64(pnode, UVH_IPI_INT, val); } @@ -115,7 +115,7 @@ static int uv_setup_intr(int cpu, u64 expires) uv_write_global_mmr64(pnode, UVH_EVENT_OCCURRED0_ALIAS, UVH_EVENT_OCCURRED0_RTC1_MASK); - val = (GENERIC_INTERRUPT_VECTOR << UVH_RTC1_INT_CONFIG_VECTOR_SHFT) | + val = (X86_PLATFORM_IPI_VECTOR << UVH_RTC1_INT_CONFIG_VECTOR_SHFT) | ((u64)cpu_physical_id(cpu) << UVH_RTC1_INT_CONFIG_APIC_ID_SHFT); /* Set configuration */ @@ -123,7 +123,10 @@ static int uv_setup_intr(int cpu, u64 expires) /* Initialize comparator value */ uv_write_global_mmr64(pnode, UVH_INT_CMPB, expires); - return (expires < uv_read_rtc(NULL) && !uv_intr_pending(pnode)); + if (uv_read_rtc(NULL) <= expires) + return 0; + + return !uv_intr_pending(pnode); } /* @@ -223,6 +226,7 @@ static int uv_rtc_set_timer(int cpu, u64 expires) next_cpu = head->next_cpu; *t = expires; + /* Will this one be next to go off? */ if (next_cpu < 0 || bcpu == next_cpu || expires < head->cpu[next_cpu].expires) { @@ -231,7 +235,7 @@ static int uv_rtc_set_timer(int cpu, u64 expires) *t = ULLONG_MAX; uv_rtc_find_next_timer(head, pnode); spin_unlock_irqrestore(&head->lock, flags); - return 1; + return -ETIME; } } @@ -244,7 +248,7 @@ static int uv_rtc_set_timer(int cpu, u64 expires) * * Returns 1 if this timer was pending. */ -static int uv_rtc_unset_timer(int cpu) +static int uv_rtc_unset_timer(int cpu, int force) { int pnode = uv_cpu_to_pnode(cpu); int bid = uv_cpu_to_blade_id(cpu); @@ -256,14 +260,15 @@ static int uv_rtc_unset_timer(int cpu) spin_lock_irqsave(&head->lock, flags); - if (head->next_cpu == bcpu && uv_read_rtc(NULL) >= *t) + if ((head->next_cpu == bcpu && uv_read_rtc(NULL) >= *t) || force) rc = 1; - *t = ULLONG_MAX; - - /* Was the hardware setup for this timer? */ - if (head->next_cpu == bcpu) - uv_rtc_find_next_timer(head, pnode); + if (rc) { + *t = ULLONG_MAX; + /* Was the hardware setup for this timer? */ + if (head->next_cpu == bcpu) + uv_rtc_find_next_timer(head, pnode); + } spin_unlock_irqrestore(&head->lock, flags); @@ -277,10 +282,21 @@ static int uv_rtc_unset_timer(int cpu) /* * Read the RTC. + * + * Starting with HUB rev 2.0, the UV RTC register is replicated across all + * cachelines of it's own page. This allows faster simultaneous reads + * from a given socket. */ static cycle_t uv_read_rtc(struct clocksource *cs) { - return (cycle_t)uv_read_local_mmr(UVH_RTC); + unsigned long offset; + + if (uv_get_min_hub_revision_id() == 1) + offset = 0; + else + offset = (uv_blade_processor_id() * L1_CACHE_BYTES) % PAGE_SIZE; + + return (cycle_t)uv_read_local_mmr(UVH_RTC | offset); } /* @@ -310,32 +326,32 @@ static void uv_rtc_timer_setup(enum clock_event_mode mode, break; case CLOCK_EVT_MODE_UNUSED: case CLOCK_EVT_MODE_SHUTDOWN: - uv_rtc_unset_timer(ced_cpu); + uv_rtc_unset_timer(ced_cpu, 1); break; } } static void uv_rtc_interrupt(void) { - struct clock_event_device *ced = &__get_cpu_var(cpu_ced); int cpu = smp_processor_id(); + struct clock_event_device *ced = &per_cpu(cpu_ced, cpu); if (!ced || !ced->event_handler) return; - if (uv_rtc_unset_timer(cpu) != 1) + if (uv_rtc_unset_timer(cpu, 0) != 1) return; ced->event_handler(ced); } -static int __init uv_enable_rtc(char *str) +static int __init uv_enable_evt_rtc(char *str) { - uv_rtc_enable = 1; + uv_rtc_evt_enable = 1; return 1; } -__setup("uvrtc", uv_enable_rtc); +__setup("uvrtcevt", uv_enable_evt_rtc); static __init void uv_rtc_register_clockevents(struct work_struct *dummy) { @@ -350,27 +366,32 @@ static __init int uv_rtc_setup_clock(void) { int rc; - if (!uv_rtc_enable || !is_uv_system() || generic_interrupt_extension) + if (!is_uv_system()) return -ENODEV; - generic_interrupt_extension = uv_rtc_interrupt; - clocksource_uv.mult = clocksource_hz2mult(sn_rtc_cycles_per_second, clocksource_uv.shift); + /* If single blade, prefer tsc */ + if (uv_num_possible_blades() == 1) + clocksource_uv.rating = 250; + rc = clocksource_register(&clocksource_uv); - if (rc) { - generic_interrupt_extension = NULL; + if (rc) + printk(KERN_INFO "UV RTC clocksource failed rc %d\n", rc); + else + printk(KERN_INFO "UV RTC clocksource registered freq %lu MHz\n", + sn_rtc_cycles_per_second/(unsigned long)1E6); + + if (rc || !uv_rtc_evt_enable || x86_platform_ipi_callback) return rc; - } /* Setup and register clockevents */ rc = uv_rtc_allocate_timers(); - if (rc) { - clocksource_unregister(&clocksource_uv); - generic_interrupt_extension = NULL; - return rc; - } + if (rc) + goto error; + + x86_platform_ipi_callback = uv_rtc_interrupt; clock_event_device_uv.mult = div_sc(sn_rtc_cycles_per_second, NSEC_PER_SEC, clock_event_device_uv.shift); @@ -383,11 +404,19 @@ static __init int uv_rtc_setup_clock(void) rc = schedule_on_each_cpu(uv_rtc_register_clockevents); if (rc) { - clocksource_unregister(&clocksource_uv); - generic_interrupt_extension = NULL; + x86_platform_ipi_callback = NULL; uv_rtc_deallocate_timers(); + goto error; } + printk(KERN_INFO "UV RTC clockevents registered\n"); + + return 0; + +error: + clocksource_unregister(&clocksource_uv); + printk(KERN_INFO "UV RTC clockevents failed rc %d\n", rc); + return rc; } arch_initcall(uv_rtc_setup_clock); diff --git a/arch/x86/kernel/visws_quirks.c b/arch/x86/kernel/visws_quirks.c index f068553a1b17..e680ea52db9b 100644 --- a/arch/x86/kernel/visws_quirks.c +++ b/arch/x86/kernel/visws_quirks.c @@ -49,11 +49,6 @@ extern int no_broadcast; char visws_board_type = -1; char visws_board_rev = -1; -int is_visws_box(void) -{ - return visws_board_type >= 0; -} - static void __init visws_time_init(void) { printk(KERN_INFO "Starting Cobalt Timer system clock\n"); @@ -183,7 +178,7 @@ static void __init MP_processor_info(struct mpc_cpu *m) return; } - apic_cpus = apic->apicid_to_cpu_present(m->apicid); + apic->apicid_to_cpu_present(m->apicid, &apic_cpus); physids_or(phys_cpu_present_map, phys_cpu_present_map, apic_cpus); /* * Validate version @@ -197,7 +192,7 @@ static void __init MP_processor_info(struct mpc_cpu *m) apic_version[m->apicid] = ver; } -static void __init visws_find_smp_config(unsigned int reserve) +static void __init visws_find_smp_config(void) { struct mpc_cpu *mp = phys_to_virt(CO_CPU_TAB_PHYS); unsigned short ncpus = readw(phys_to_virt(CO_CPU_NUM_PHYS)); @@ -242,6 +237,8 @@ void __init visws_early_detect(void) x86_init.irqs.pre_vector_init = visws_pre_intr_init; x86_init.irqs.trap_init = visws_trap_init; x86_init.timers.timer_init = visws_time_init; + x86_init.pci.init = pci_visws_init; + x86_init.pci.init_irq = x86_init_noop; /* * Install reboot quirks: @@ -486,7 +483,7 @@ static void end_cobalt_irq(unsigned int irq) } static struct irq_chip cobalt_irq_type = { - .typename = "Cobalt-APIC", + .name = "Cobalt-APIC", .startup = startup_cobalt_irq, .shutdown = disable_cobalt_irq, .enable = enable_cobalt_irq, @@ -508,7 +505,7 @@ static struct irq_chip cobalt_irq_type = { */ static unsigned int startup_piix4_master_irq(unsigned int irq) { - init_8259A(0); + legacy_pic->init(0); return startup_cobalt_irq(irq); } @@ -523,7 +520,7 @@ static void end_piix4_master_irq(unsigned int irq) } static struct irq_chip piix4_master_irq_type = { - .typename = "PIIX4-master", + .name = "PIIX4-master", .startup = startup_piix4_master_irq, .ack = ack_cobalt_irq, .end = end_piix4_master_irq, @@ -531,10 +528,7 @@ static struct irq_chip piix4_master_irq_type = { static struct irq_chip piix4_virtual_irq_type = { - .typename = "PIIX4-virtual", - .shutdown = disable_8259A_irq, - .enable = enable_8259A_irq, - .disable = disable_8259A_irq, + .name = "PIIX4-virtual", }; @@ -559,7 +553,7 @@ static irqreturn_t piix4_master_intr(int irq, void *dev_id) struct irq_desc *desc; unsigned long flags; - spin_lock_irqsave(&i8259A_lock, flags); + raw_spin_lock_irqsave(&i8259A_lock, flags); /* Find out what's interrupting in the PIIX4 master 8259 */ outb(0x0c, 0x20); /* OCW3 Poll command */ @@ -596,7 +590,7 @@ static irqreturn_t piix4_master_intr(int irq, void *dev_id) outb(0x60 + realirq, 0x20); } - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); desc = irq_to_desc(realirq); @@ -609,12 +603,12 @@ static irqreturn_t piix4_master_intr(int irq, void *dev_id) handle_IRQ_event(realirq, desc->action); if (!(desc->status & IRQ_DISABLED)) - enable_8259A_irq(realirq); + legacy_pic->chip->unmask(realirq); return IRQ_HANDLED; out_unlock: - spin_unlock_irqrestore(&i8259A_lock, flags); + raw_spin_unlock_irqrestore(&i8259A_lock, flags); return IRQ_NONE; } @@ -628,6 +622,12 @@ static struct irqaction cascade_action = { .name = "cascade", }; +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; +} void init_VISWS_APIC_irqs(void) { @@ -653,6 +653,7 @@ void init_VISWS_APIC_irqs(void) 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)) { diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c index 9c4e62539058..5ffb5622f793 100644 --- a/arch/x86/kernel/vm86_32.c +++ b/arch/x86/kernel/vm86_32.c @@ -197,9 +197,8 @@ out: static int do_vm86_irq_handling(int subfunction, int irqnumber); static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk); -int sys_vm86old(struct pt_regs *regs) +int sys_vm86old(struct vm86_struct __user *v86, struct pt_regs *regs) { - struct vm86_struct __user *v86 = (struct vm86_struct __user *)regs->bx; struct kernel_vm86_struct info; /* declare this _on top_, * this avoids wasting of stack space. * This remains on the stack until we @@ -227,7 +226,7 @@ out: } -int sys_vm86(struct pt_regs *regs) +int sys_vm86(unsigned long cmd, unsigned long arg, struct pt_regs *regs) { struct kernel_vm86_struct info; /* declare this _on top_, * this avoids wasting of stack space. @@ -239,12 +238,12 @@ int sys_vm86(struct pt_regs *regs) struct vm86plus_struct __user *v86; tsk = current; - switch (regs->bx) { + switch (cmd) { case VM86_REQUEST_IRQ: case VM86_FREE_IRQ: case VM86_GET_IRQ_BITS: case VM86_GET_AND_RESET_IRQ: - ret = do_vm86_irq_handling(regs->bx, (int)regs->cx); + ret = do_vm86_irq_handling(cmd, (int)arg); goto out; case VM86_PLUS_INSTALL_CHECK: /* @@ -261,7 +260,7 @@ int sys_vm86(struct pt_regs *regs) ret = -EPERM; if (tsk->thread.saved_sp0) goto out; - v86 = (struct vm86plus_struct __user *)regs->cx; + v86 = (struct vm86plus_struct __user *)arg; tmp = copy_vm86_regs_from_user(&info.regs, &v86->regs, offsetof(struct kernel_vm86_struct, regs32) - sizeof(info.regs)); diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index 31e6f6cfe53e..7dd599deca4a 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -33,6 +33,7 @@ #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> @@ -266,30 +267,6 @@ static void vmi_nop(void) { } -#ifdef CONFIG_HIGHPTE -static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type) -{ - void *va = kmap_atomic(page, type); - - /* - * Internally, the VMI ROM must map virtual addresses to physical - * addresses for processing MMU updates. By the time MMU updates - * are issued, this information is typically already lost. - * Fortunately, the VMI provides a cache of mapping slots for active - * page tables. - * - * We use slot zero for the linear mapping of physical memory, and - * in HIGHPTE kernels, slot 1 and 2 for KM_PTE0 and KM_PTE1. - * - * args: SLOT VA COUNT PFN - */ - BUG_ON(type != KM_PTE0 && type != KM_PTE1); - vmi_ops.set_linear_mapping((type - KM_PTE0)+1, va, 1, page_to_pfn(page)); - - return va; -} -#endif - static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn) { vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0); @@ -640,6 +617,12 @@ static inline int __init activate_vmi(void) 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; @@ -648,7 +631,7 @@ static inline int __init activate_vmi(void) pv_info.paravirt_enabled = 1; pv_info.kernel_rpl = kernel_cs & SEGMENT_RPL_MASK; - pv_info.name = "vmi"; + pv_info.name = "vmi [deprecated]"; pv_init_ops.patch = vmi_patch; @@ -778,10 +761,6 @@ static inline int __init activate_vmi(void) /* Set linear is needed in all cases */ vmi_ops.set_linear_mapping = vmi_get_function(VMI_CALL_SetLinearMapping); -#ifdef CONFIG_HIGHPTE - if (vmi_ops.set_linear_mapping) - pv_mmu_ops.kmap_atomic_pte = vmi_kmap_atomic_pte; -#endif /* * These MUST always be patched. Don't support indirect jumps diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c index 611b9e2360d3..5e1ff66ecd73 100644 --- a/arch/x86/kernel/vmiclock_32.c +++ b/arch/x86/kernel/vmiclock_32.c @@ -79,11 +79,7 @@ unsigned long vmi_tsc_khz(void) static inline unsigned int vmi_get_timer_vector(void) { -#ifdef CONFIG_X86_IO_APIC - return FIRST_DEVICE_VECTOR; -#else - return FIRST_EXTERNAL_VECTOR; -#endif + return IRQ0_VECTOR; } /** vmi clockchip */ @@ -171,7 +167,7 @@ static int vmi_timer_next_event(unsigned long delta, { /* Unfortunately, set_next_event interface only passes relative * expiry, but we want absolute expiry. It'd be better if were - * were passed an aboslute expiry, since a bunch of time may + * 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)); @@ -226,7 +222,7 @@ static void __devinit vmi_time_init_clockevent(void) evt->min_delta_ns = clockevent_delta2ns(1, evt); evt->cpumask = cpumask_of(cpu); - printk(KERN_WARNING "vmi: registering clock event %s. mult=%lu shift=%u\n", + printk(KERN_WARNING "vmi: registering clock event %s. mult=%u shift=%u\n", evt->name, evt->mult, evt->shift); clockevents_register_device(evt); } diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index 92929fb3f9fa..44879df55696 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -41,6 +41,32 @@ ENTRY(phys_startup_64) jiffies_64 = jiffies; #endif +#if defined(CONFIG_X86_64) && defined(CONFIG_DEBUG_RODATA) +/* + * On 64-bit, align RODATA to 2MB so that even with CONFIG_DEBUG_RODATA + * we retain large page mappings for boundaries spanning kernel text, rodata + * and data sections. + * + * However, kernel identity mappings will have different RWX permissions + * to the pages mapping to text and to the pages padding (which are freed) the + * text section. Hence kernel identity mappings will be broken to smaller + * pages. For 64-bit, kernel text and kernel identity mappings are different, + * so we can enable protection checks that come with CONFIG_DEBUG_RODATA, + * as well as retain 2MB large page mappings for kernel text. + */ +#define X64_ALIGN_DEBUG_RODATA_BEGIN . = ALIGN(HPAGE_SIZE); + +#define X64_ALIGN_DEBUG_RODATA_END \ + . = ALIGN(HPAGE_SIZE); \ + __end_rodata_hpage_align = .; + +#else + +#define X64_ALIGN_DEBUG_RODATA_BEGIN +#define X64_ALIGN_DEBUG_RODATA_END + +#endif + PHDRS { text PT_LOAD FLAGS(5); /* R_E */ data PT_LOAD FLAGS(7); /* RWE */ @@ -90,7 +116,9 @@ SECTIONS EXCEPTION_TABLE(16) :text = 0x9090 + X64_ALIGN_DEBUG_RODATA_BEGIN RO_DATA(PAGE_SIZE) + X64_ALIGN_DEBUG_RODATA_END /* Data */ .data : AT(ADDR(.data) - LOAD_OFFSET) { @@ -107,13 +135,13 @@ SECTIONS PAGE_ALIGNED_DATA(PAGE_SIZE) - CACHELINE_ALIGNED_DATA(CONFIG_X86_L1_CACHE_BYTES) + CACHELINE_ALIGNED_DATA(L1_CACHE_BYTES) DATA_DATA CONSTRUCTORS /* rarely changed data like cpu maps */ - READ_MOSTLY_DATA(CONFIG_X86_INTERNODE_CACHE_BYTES) + READ_MOSTLY_DATA(INTERNODE_CACHE_BYTES) /* End of data section */ _edata = .; @@ -137,12 +165,12 @@ SECTIONS *(.vsyscall_0) } :user - . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); + . = ALIGN(L1_CACHE_BYTES); .vsyscall_fn : AT(VLOAD(.vsyscall_fn)) { *(.vsyscall_fn) } - . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); + . = ALIGN(L1_CACHE_BYTES); .vsyscall_gtod_data : AT(VLOAD(.vsyscall_gtod_data)) { *(.vsyscall_gtod_data) } @@ -166,7 +194,7 @@ SECTIONS } vgetcpu_mode = VVIRT(.vgetcpu_mode); - . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); + . = ALIGN(L1_CACHE_BYTES); .jiffies : AT(VLOAD(.jiffies)) { *(.jiffies) } @@ -291,9 +319,7 @@ SECTIONS __brk_limit = .; } - .end : AT(ADDR(.end) - LOAD_OFFSET) { - _end = .; - } + _end = .; STABS_DEBUG DWARF_DEBUG @@ -305,6 +331,9 @@ SECTIONS #ifdef CONFIG_X86_32 +/* + * The ASSERT() sink to . is intentional, for binutils 2.14 compatibility: + */ . = ASSERT((_end - LOAD_OFFSET <= KERNEL_IMAGE_SIZE), "kernel image bigger than KERNEL_IMAGE_SIZE"); #else @@ -312,7 +341,7 @@ SECTIONS * Per-cpu symbols which need to be offset from __per_cpu_load * for the boot processor. */ -#define INIT_PER_CPU(x) init_per_cpu__##x = per_cpu__##x + __per_cpu_load +#define INIT_PER_CPU(x) init_per_cpu__##x = x + __per_cpu_load INIT_PER_CPU(gdt_page); INIT_PER_CPU(irq_stack_union); @@ -323,7 +352,7 @@ INIT_PER_CPU(irq_stack_union); "kernel image bigger than KERNEL_IMAGE_SIZE"); #ifdef CONFIG_SMP -. = ASSERT((per_cpu__irq_stack_union == 0), +. = ASSERT((irq_stack_union == 0), "irq_stack_union is not at start of per-cpu area"); #endif diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index 8cb4974ff599..1c0c6ab9c60f 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -73,7 +73,8 @@ void update_vsyscall_tz(void) write_sequnlock_irqrestore(&vsyscall_gtod_data.lock, flags); } -void update_vsyscall(struct timespec *wall_time, struct clocksource *clock) +void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, + u32 mult) { unsigned long flags; @@ -82,7 +83,7 @@ void update_vsyscall(struct timespec *wall_time, struct clocksource *clock) vsyscall_gtod_data.clock.vread = clock->vread; vsyscall_gtod_data.clock.cycle_last = clock->cycle_last; vsyscall_gtod_data.clock.mask = clock->mask; - vsyscall_gtod_data.clock.mult = clock->mult; + vsyscall_gtod_data.clock.mult = mult; vsyscall_gtod_data.clock.shift = clock->shift; vsyscall_gtod_data.wall_time_sec = wall_time->tv_sec; vsyscall_gtod_data.wall_time_nsec = wall_time->tv_nsec; @@ -237,7 +238,7 @@ static ctl_table kernel_table2[] = { }; static ctl_table kernel_root_table2[] = { - { .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555, + { .procname = "kernel", .mode = 0555, .child = kernel_table2 }, {} }; @@ -300,7 +301,8 @@ static int __init vsyscall_init(void) register_sysctl_table(kernel_root_table2); #endif on_each_cpu(cpu_vsyscall_init, NULL, 1); - hotcpu_notifier(cpu_vsyscall_notifier, 0); + /* notifier priority > KVM */ + hotcpu_notifier(cpu_vsyscall_notifier, 30); return 0; } diff --git a/arch/x86/kernel/x8664_ksyms_64.c b/arch/x86/kernel/x8664_ksyms_64.c index 3909e3ba5ce3..693920b22496 100644 --- a/arch/x86/kernel/x8664_ksyms_64.c +++ b/arch/x86/kernel/x8664_ksyms_64.c @@ -17,8 +17,6 @@ EXPORT_SYMBOL(mcount); #endif -EXPORT_SYMBOL(kernel_thread); - EXPORT_SYMBOL(__get_user_1); EXPORT_SYMBOL(__get_user_2); EXPORT_SYMBOL(__get_user_4); @@ -28,11 +26,11 @@ EXPORT_SYMBOL(__put_user_2); EXPORT_SYMBOL(__put_user_4); EXPORT_SYMBOL(__put_user_8); -EXPORT_SYMBOL(copy_user_generic); +EXPORT_SYMBOL(copy_user_generic_string); +EXPORT_SYMBOL(copy_user_generic_unrolled); EXPORT_SYMBOL(__copy_user_nocache); -EXPORT_SYMBOL(copy_from_user); -EXPORT_SYMBOL(copy_to_user); -EXPORT_SYMBOL(__copy_from_user_inatomic); +EXPORT_SYMBOL(_copy_from_user); +EXPORT_SYMBOL(_copy_to_user); EXPORT_SYMBOL(copy_page); EXPORT_SYMBOL(clear_page); @@ -57,4 +55,6 @@ EXPORT_SYMBOL(__memcpy); EXPORT_SYMBOL(empty_zero_page); EXPORT_SYMBOL(init_level4_pgt); -EXPORT_SYMBOL(load_gs_index); +#ifndef CONFIG_PARAVIRT +EXPORT_SYMBOL(native_load_gs_index); +#endif diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c index 4449a4a2c2ed..61a1e8c7e19f 100644 --- a/arch/x86/kernel/x86_init.c +++ b/arch/x86/kernel/x86_init.c @@ -4,20 +4,26 @@ * For licencing details see kernel-base/COPYING */ #include <linux/init.h> +#include <linux/ioport.h> #include <asm/bios_ebda.h> #include <asm/paravirt.h> +#include <asm/pci_x86.h> #include <asm/mpspec.h> #include <asm/setup.h> #include <asm/apic.h> #include <asm/e820.h> #include <asm/time.h> #include <asm/irq.h> +#include <asm/pat.h> #include <asm/tsc.h> +#include <asm/iommu.h> void __cpuinit x86_init_noop(void) { } void __init x86_init_uint_noop(unsigned int unused) { } void __init x86_init_pgd_noop(pgd_t *unused) { } +int __init iommu_init_noop(void) { return 0; } +void iommu_shutdown_noop(void) { } /* * The platform setup functions are preset with the default functions @@ -62,14 +68,29 @@ struct x86_init_ops x86_init __initdata = { .tsc_pre_init = x86_init_noop, .timer_init = hpet_time_init, }, + + .iommu = { + .iommu_init = iommu_init_noop, + }, + + .pci = { + .init = x86_default_pci_init, + .init_irq = x86_default_pci_init_irq, + .fixup_irqs = x86_default_pci_fixup_irqs, + }, }; struct x86_cpuinit_ops x86_cpuinit __cpuinitdata = { .setup_percpu_clockev = setup_secondary_APIC_clock, }; +static void default_nmi_init(void) { }; + struct x86_platform_ops x86_platform = { .calibrate_tsc = native_calibrate_tsc, .get_wallclock = mach_get_cmos_time, .set_wallclock = mach_set_rtc_mmss, + .iommu_shutdown = iommu_shutdown_noop, + .is_untracked_pat_range = is_ISA_range, + .nmi_init = default_nmi_init }; diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c index c5ee17e8c6d9..782c3a362ec6 100644 --- a/arch/x86/kernel/xsave.c +++ b/arch/x86/kernel/xsave.c @@ -337,6 +337,7 @@ void __ref xsave_cntxt_init(void) cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx); xstate_size = ebx; + update_regset_xstate_info(xstate_size, pcntxt_mask); prepare_fx_sw_frame(); setup_xstate_init(); |