diff options
Diffstat (limited to 'arch/x86/kernel')
134 files changed, 10297 insertions, 6867 deletions
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index d364df03c1d..339ce35648e 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -23,11 +23,12 @@ nostackp := $(call cc-option, -fno-stack-protector) CFLAGS_vsyscall_64.o := $(PROFILING) -g0 $(nostackp) CFLAGS_hpet.o := $(nostackp) CFLAGS_tsc.o := $(nostackp) +CFLAGS_paravirt.o := $(nostackp) obj-y := process_$(BITS).o signal.o entry_$(BITS).o obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o obj-y += time_$(BITS).o ioport.o ldt.o dumpstack.o -obj-y += setup.o i8259.o irqinit_$(BITS).o setup_percpu.o +obj-y += setup.o i8259.o irqinit_$(BITS).o obj-$(CONFIG_X86_VISWS) += visws_quirks.o obj-$(CONFIG_X86_32) += probe_roms_32.o obj-$(CONFIG_X86_32) += sys_i386_32.o i386_ksyms_32.o @@ -49,31 +50,27 @@ obj-y += step.o obj-$(CONFIG_STACKTRACE) += stacktrace.o obj-y += cpu/ obj-y += acpi/ -obj-$(CONFIG_X86_BIOS_REBOOT) += reboot.o +obj-y += reboot.o obj-$(CONFIG_MCA) += mca_32.o obj-$(CONFIG_X86_MSR) += msr.o obj-$(CONFIG_X86_CPUID) += cpuid.o obj-$(CONFIG_PCI) += early-quirks.o apm-y := apm_32.o obj-$(CONFIG_APM) += apm.o -obj-$(CONFIG_X86_SMP) += smp.o -obj-$(CONFIG_X86_SMP) += smpboot.o tsc_sync.o ipi.o tlb_$(BITS).o -obj-$(CONFIG_X86_32_SMP) += smpcommon.o -obj-$(CONFIG_X86_64_SMP) += tsc_sync.o smpcommon.o +obj-$(CONFIG_SMP) += smp.o +obj-$(CONFIG_SMP) += smpboot.o tsc_sync.o +obj-$(CONFIG_SMP) += setup_percpu.o +obj-$(CONFIG_X86_64_SMP) += tsc_sync.o obj-$(CONFIG_X86_TRAMPOLINE) += trampoline_$(BITS).o obj-$(CONFIG_X86_MPPARSE) += mpparse.o -obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o -obj-$(CONFIG_X86_IO_APIC) += io_apic.o +obj-y += apic/ obj-$(CONFIG_X86_REBOOTFIXUPS) += reboot_fixups_32.o obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += ftrace.o obj-$(CONFIG_KEXEC) += machine_kexec_$(BITS).o obj-$(CONFIG_KEXEC) += relocate_kernel_$(BITS).o crash.o obj-$(CONFIG_CRASH_DUMP) += crash_dump_$(BITS).o -obj-$(CONFIG_X86_NUMAQ) += numaq_32.o -obj-$(CONFIG_X86_ES7000) += es7000_32.o -obj-$(CONFIG_X86_SUMMIT_NUMA) += summit_32.o -obj-y += vsmp_64.o +obj-$(CONFIG_X86_VSMP) += vsmp_64.o obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_MODULES) += module_$(BITS).o obj-$(CONFIG_EFI) += efi.o efi_$(BITS).o efi_stub_$(BITS).o @@ -114,16 +111,13 @@ obj-$(CONFIG_SWIOTLB) += pci-swiotlb_64.o # NB rename without _64 ### # 64 bit specific files ifeq ($(CONFIG_X86_64),y) - obj-y += genapic_64.o genapic_flat_64.o genx2apic_uv_x.o tlb_uv.o - obj-y += bios_uv.o uv_irq.o uv_sysfs.o - obj-y += genx2apic_cluster.o - obj-y += genx2apic_phys.o - obj-$(CONFIG_X86_PM_TIMER) += pmtimer_64.o - obj-$(CONFIG_AUDIT) += audit_64.o - - obj-$(CONFIG_GART_IOMMU) += pci-gart_64.o aperture_64.o - obj-$(CONFIG_CALGARY_IOMMU) += pci-calgary_64.o tce_64.o - obj-$(CONFIG_AMD_IOMMU) += amd_iommu_init.o amd_iommu.o - - obj-$(CONFIG_PCI_MMCONFIG) += mmconf-fam10h_64.o + obj-$(CONFIG_X86_UV) += tlb_uv.o bios_uv.o uv_irq.o uv_sysfs.o uv_time.o + obj-$(CONFIG_X86_PM_TIMER) += pmtimer_64.o + obj-$(CONFIG_AUDIT) += audit_64.o + + obj-$(CONFIG_GART_IOMMU) += pci-gart_64.o aperture_64.o + obj-$(CONFIG_CALGARY_IOMMU) += pci-calgary_64.o tce_64.o + obj-$(CONFIG_AMD_IOMMU) += amd_iommu_init.o amd_iommu.o + + obj-$(CONFIG_PCI_MMCONFIG) += mmconf-fam10h_64.o endif diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 7678f10c456..a18eb7ce223 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -37,15 +37,10 @@ #include <asm/pgtable.h> #include <asm/io_apic.h> #include <asm/apic.h> -#include <asm/genapic.h> #include <asm/io.h> #include <asm/mpspec.h> #include <asm/smp.h> -#ifdef CONFIG_X86_LOCAL_APIC -# include <mach_apic.h> -#endif - static int __initdata acpi_force = 0; u32 acpi_rsdt_forced; #ifdef CONFIG_ACPI @@ -56,16 +51,7 @@ int acpi_disabled = 1; EXPORT_SYMBOL(acpi_disabled); #ifdef CONFIG_X86_64 - -#include <asm/proto.h> - -#else /* X86 */ - -#ifdef CONFIG_X86_LOCAL_APIC -#include <mach_apic.h> -#include <mach_mpparse.h> -#endif /* CONFIG_X86_LOCAL_APIC */ - +# include <asm/proto.h> #endif /* X86 */ #define BAD_MADT_ENTRY(entry, end) ( \ @@ -121,35 +107,18 @@ enum acpi_irq_model_id acpi_irq_model = ACPI_IRQ_MODEL_PIC; */ char *__init __acpi_map_table(unsigned long phys, unsigned long size) { - unsigned long base, offset, mapped_size; - int idx; if (!phys || !size) return NULL; - if (phys+size <= (max_low_pfn_mapped << PAGE_SHIFT)) - return __va(phys); - - offset = phys & (PAGE_SIZE - 1); - mapped_size = PAGE_SIZE - offset; - clear_fixmap(FIX_ACPI_END); - set_fixmap(FIX_ACPI_END, phys); - base = fix_to_virt(FIX_ACPI_END); - - /* - * Most cases can be covered by the below. - */ - idx = FIX_ACPI_END; - while (mapped_size < size) { - if (--idx < FIX_ACPI_BEGIN) - return NULL; /* cannot handle this */ - phys += PAGE_SIZE; - clear_fixmap(idx); - set_fixmap(idx, phys); - mapped_size += PAGE_SIZE; - } + return early_ioremap(phys, size); +} +void __init __acpi_unmap_table(char *map, unsigned long size) +{ + if (!map || !size) + return; - return ((unsigned char *)base + offset); + early_iounmap(map, size); } #ifdef CONFIG_PCI_MMCONFIG @@ -239,7 +208,8 @@ static int __init acpi_parse_madt(struct acpi_table_header *table) madt->address); } - acpi_madt_oem_check(madt->header.oem_id, madt->header.oem_table_id); + default_acpi_madt_oem_check(madt->header.oem_id, + madt->header.oem_table_id); return 0; } @@ -884,7 +854,7 @@ static struct { DECLARE_BITMAP(pin_programmed, MP_MAX_IOAPIC_PIN + 1); } mp_ioapic_routing[MAX_IO_APICS]; -static int mp_find_ioapic(int gsi) +int mp_find_ioapic(int gsi) { int i = 0; @@ -899,6 +869,16 @@ static int mp_find_ioapic(int gsi) return -1; } +int mp_find_ioapic_pin(int ioapic, int gsi) +{ + if (WARN_ON(ioapic == -1)) + return -1; + if (WARN_ON(gsi > mp_ioapic_routing[ioapic].gsi_end)) + return -1; + + return gsi - mp_ioapic_routing[ioapic].gsi_base; +} + static u8 __init uniq_ioapic_id(u8 id) { #ifdef CONFIG_X86_32 @@ -912,8 +892,8 @@ static u8 __init uniq_ioapic_id(u8 id) DECLARE_BITMAP(used, 256); bitmap_zero(used, 256); for (i = 0; i < nr_ioapics; i++) { - struct mp_config_ioapic *ia = &mp_ioapics[i]; - __set_bit(ia->mp_apicid, used); + struct mpc_ioapic *ia = &mp_ioapics[i]; + __set_bit(ia->apicid, used); } if (!test_bit(id, used)) return id; @@ -945,29 +925,29 @@ void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) idx = nr_ioapics; - mp_ioapics[idx].mp_type = MP_IOAPIC; - mp_ioapics[idx].mp_flags = MPC_APIC_USABLE; - mp_ioapics[idx].mp_apicaddr = address; + mp_ioapics[idx].type = MP_IOAPIC; + mp_ioapics[idx].flags = MPC_APIC_USABLE; + mp_ioapics[idx].apicaddr = address; set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address); - mp_ioapics[idx].mp_apicid = uniq_ioapic_id(id); + mp_ioapics[idx].apicid = uniq_ioapic_id(id); #ifdef CONFIG_X86_32 - mp_ioapics[idx].mp_apicver = io_apic_get_version(idx); + mp_ioapics[idx].apicver = io_apic_get_version(idx); #else - mp_ioapics[idx].mp_apicver = 0; + mp_ioapics[idx].apicver = 0; #endif /* * Build basic GSI lookup table to facilitate gsi->io_apic lookups * and to prevent reprogramming of IOAPIC pins (PCI GSIs). */ - mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mp_apicid; + mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].apicid; mp_ioapic_routing[idx].gsi_base = gsi_base; mp_ioapic_routing[idx].gsi_end = gsi_base + io_apic_get_redir_entries(idx); - printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%lx, " - "GSI %d-%d\n", idx, mp_ioapics[idx].mp_apicid, - mp_ioapics[idx].mp_apicver, mp_ioapics[idx].mp_apicaddr, + printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, " + "GSI %d-%d\n", idx, mp_ioapics[idx].apicid, + mp_ioapics[idx].apicver, mp_ioapics[idx].apicaddr, mp_ioapic_routing[idx].gsi_base, mp_ioapic_routing[idx].gsi_end); nr_ioapics++; @@ -996,19 +976,19 @@ int __init acpi_probe_gsi(void) return max_gsi + 1; } -static void assign_to_mp_irq(struct mp_config_intsrc *m, - struct mp_config_intsrc *mp_irq) +static void assign_to_mp_irq(struct mpc_intsrc *m, + struct mpc_intsrc *mp_irq) { - memcpy(mp_irq, m, sizeof(struct mp_config_intsrc)); + memcpy(mp_irq, m, sizeof(struct mpc_intsrc)); } -static int mp_irq_cmp(struct mp_config_intsrc *mp_irq, - struct mp_config_intsrc *m) +static int mp_irq_cmp(struct mpc_intsrc *mp_irq, + struct mpc_intsrc *m) { - return memcmp(mp_irq, m, sizeof(struct mp_config_intsrc)); + return memcmp(mp_irq, m, sizeof(struct mpc_intsrc)); } -static void save_mp_irq(struct mp_config_intsrc *m) +static void save_mp_irq(struct mpc_intsrc *m) { int i; @@ -1026,7 +1006,7 @@ void __init mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) { int ioapic; int pin; - struct mp_config_intsrc mp_irq; + struct mpc_intsrc mp_irq; /* * Convert 'gsi' to 'ioapic.pin'. @@ -1034,7 +1014,7 @@ void __init mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) ioapic = mp_find_ioapic(gsi); if (ioapic < 0) return; - pin = gsi - mp_ioapic_routing[ioapic].gsi_base; + pin = mp_find_ioapic_pin(ioapic, gsi); /* * TBD: This check is for faulty timer entries, where the override @@ -1044,13 +1024,13 @@ void __init mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) if ((bus_irq == 0) && (trigger == 3)) trigger = 1; - mp_irq.mp_type = MP_INTSRC; - mp_irq.mp_irqtype = mp_INT; - mp_irq.mp_irqflag = (trigger << 2) | polarity; - mp_irq.mp_srcbus = MP_ISA_BUS; - mp_irq.mp_srcbusirq = bus_irq; /* IRQ */ - mp_irq.mp_dstapic = mp_ioapics[ioapic].mp_apicid; /* APIC ID */ - mp_irq.mp_dstirq = pin; /* INTIN# */ + mp_irq.type = MP_INTSRC; + mp_irq.irqtype = mp_INT; + mp_irq.irqflag = (trigger << 2) | polarity; + mp_irq.srcbus = MP_ISA_BUS; + mp_irq.srcbusirq = bus_irq; /* IRQ */ + mp_irq.dstapic = mp_ioapics[ioapic].apicid; /* APIC ID */ + mp_irq.dstirq = pin; /* INTIN# */ save_mp_irq(&mp_irq); } @@ -1060,7 +1040,7 @@ void __init mp_config_acpi_legacy_irqs(void) int i; int ioapic; unsigned int dstapic; - struct mp_config_intsrc mp_irq; + struct mpc_intsrc mp_irq; #if defined (CONFIG_MCA) || defined (CONFIG_EISA) /* @@ -1085,7 +1065,7 @@ void __init mp_config_acpi_legacy_irqs(void) ioapic = mp_find_ioapic(0); if (ioapic < 0) return; - dstapic = mp_ioapics[ioapic].mp_apicid; + dstapic = mp_ioapics[ioapic].apicid; /* * Use the default configuration for the IRQs 0-15. Unless @@ -1095,16 +1075,14 @@ void __init mp_config_acpi_legacy_irqs(void) int idx; for (idx = 0; idx < mp_irq_entries; idx++) { - struct mp_config_intsrc *irq = mp_irqs + idx; + struct mpc_intsrc *irq = mp_irqs + idx; /* Do we already have a mapping for this ISA IRQ? */ - if (irq->mp_srcbus == MP_ISA_BUS - && irq->mp_srcbusirq == i) + if (irq->srcbus == MP_ISA_BUS && irq->srcbusirq == i) break; /* Do we already have a mapping for this IOAPIC pin */ - if (irq->mp_dstapic == dstapic && - irq->mp_dstirq == i) + if (irq->dstapic == dstapic && irq->dstirq == i) break; } @@ -1113,13 +1091,13 @@ void __init mp_config_acpi_legacy_irqs(void) continue; /* IRQ already used */ } - mp_irq.mp_type = MP_INTSRC; - mp_irq.mp_irqflag = 0; /* Conforming */ - mp_irq.mp_srcbus = MP_ISA_BUS; - mp_irq.mp_dstapic = dstapic; - mp_irq.mp_irqtype = mp_INT; - mp_irq.mp_srcbusirq = i; /* Identity mapped */ - mp_irq.mp_dstirq = i; + mp_irq.type = MP_INTSRC; + mp_irq.irqflag = 0; /* Conforming */ + mp_irq.srcbus = MP_ISA_BUS; + mp_irq.dstapic = dstapic; + mp_irq.irqtype = mp_INT; + mp_irq.srcbusirq = i; /* Identity mapped */ + mp_irq.dstirq = i; save_mp_irq(&mp_irq); } @@ -1156,7 +1134,7 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity) return gsi; } - ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_base; + ioapic_pin = mp_find_ioapic_pin(ioapic, gsi); #ifdef CONFIG_X86_32 if (ioapic_renumber_irq) @@ -1230,22 +1208,22 @@ int mp_config_acpi_gsi(unsigned char number, unsigned int devfn, u8 pin, u32 gsi, int triggering, int polarity) { #ifdef CONFIG_X86_MPPARSE - struct mp_config_intsrc mp_irq; + struct mpc_intsrc mp_irq; int ioapic; if (!acpi_ioapic) return 0; /* print the entry should happen on mptable identically */ - mp_irq.mp_type = MP_INTSRC; - mp_irq.mp_irqtype = mp_INT; - mp_irq.mp_irqflag = (triggering == ACPI_EDGE_SENSITIVE ? 4 : 0x0c) | + mp_irq.type = MP_INTSRC; + mp_irq.irqtype = mp_INT; + mp_irq.irqflag = (triggering == ACPI_EDGE_SENSITIVE ? 4 : 0x0c) | (polarity == ACPI_ACTIVE_HIGH ? 1 : 3); - mp_irq.mp_srcbus = number; - mp_irq.mp_srcbusirq = (((devfn >> 3) & 0x1f) << 2) | ((pin - 1) & 3); + mp_irq.srcbus = number; + mp_irq.srcbusirq = (((devfn >> 3) & 0x1f) << 2) | ((pin - 1) & 3); ioapic = mp_find_ioapic(gsi); - mp_irq.mp_dstapic = mp_ioapic_routing[ioapic].apic_id; - mp_irq.mp_dstirq = gsi - mp_ioapic_routing[ioapic].gsi_base; + mp_irq.dstapic = mp_ioapic_routing[ioapic].apic_id; + mp_irq.dstirq = mp_find_ioapic_pin(ioapic, gsi); save_mp_irq(&mp_irq); #endif @@ -1372,7 +1350,7 @@ static void __init acpi_process_madt(void) if (!error) { acpi_lapic = 1; -#ifdef CONFIG_X86_GENERICARCH +#ifdef CONFIG_X86_BIGSMP generic_bigsmp_probe(); #endif /* @@ -1384,9 +1362,8 @@ static void __init acpi_process_madt(void) acpi_ioapic = 1; smp_found_config = 1; -#ifdef CONFIG_X86_32 - setup_apic_routing(); -#endif + if (apic->setup_apic_routing) + apic->setup_apic_routing(); } } if (error == -EINVAL) { diff --git a/arch/x86/kernel/acpi/realmode/wakeup.S b/arch/x86/kernel/acpi/realmode/wakeup.S index 3355973b12a..580b4e29601 100644 --- a/arch/x86/kernel/acpi/realmode/wakeup.S +++ b/arch/x86/kernel/acpi/realmode/wakeup.S @@ -3,8 +3,8 @@ */ #include <asm/segment.h> #include <asm/msr-index.h> -#include <asm/page.h> -#include <asm/pgtable.h> +#include <asm/page_types.h> +#include <asm/pgtable_types.h> #include <asm/processor-flags.h> .code16 diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index a60c1f3bcb8..7c243a2c511 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -101,6 +101,7 @@ int acpi_save_state_mem(void) stack_start.sp = temp_stack + sizeof(temp_stack); early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(smp_processor_id()); + initial_gs = per_cpu_offset(smp_processor_id()); #endif initial_code = (unsigned long)wakeup_long64; saved_magic = 0x123456789abcdef0; diff --git a/arch/x86/kernel/acpi/wakeup_32.S b/arch/x86/kernel/acpi/wakeup_32.S index a12e6a9fb65..8ded418b059 100644 --- a/arch/x86/kernel/acpi/wakeup_32.S +++ b/arch/x86/kernel/acpi/wakeup_32.S @@ -1,7 +1,7 @@ .section .text.page_aligned #include <linux/linkage.h> #include <asm/segment.h> -#include <asm/page.h> +#include <asm/page_types.h> # Copyright 2003, 2008 Pavel Machek <pavel@suse.cz>, distribute under GPLv2 diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S index 96258d9dc97..8ea5164cbd0 100644 --- a/arch/x86/kernel/acpi/wakeup_64.S +++ b/arch/x86/kernel/acpi/wakeup_64.S @@ -1,8 +1,8 @@ .text #include <linux/linkage.h> #include <asm/segment.h> -#include <asm/pgtable.h> -#include <asm/page.h> +#include <asm/pgtable_types.h> +#include <asm/page_types.h> #include <asm/msr.h> #include <asm/asm-offsets.h> diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index a84ac7b570e..4c80f155743 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -414,9 +414,17 @@ void __init alternative_instructions(void) that might execute the to be patched code. Other CPUs are not running. */ stop_nmi(); -#ifdef CONFIG_X86_MCE - stop_mce(); -#endif + + /* + * Don't stop machine check exceptions while patching. + * MCEs only happen when something got corrupted and in this + * case we must do something about the corruption. + * Ignoring it is worse than a unlikely patching race. + * Also machine checks tend to be broadcast and if one CPU + * goes into machine check the others follow quickly, so we don't + * expect a machine check to cause undue problems during to code + * patching. + */ apply_alternatives(__alt_instructions, __alt_instructions_end); @@ -456,9 +464,6 @@ void __init alternative_instructions(void) (unsigned long)__smp_locks_end); restart_nmi(); -#ifdef CONFIG_X86_MCE - restart_mce(); -#endif } /** @@ -498,12 +503,12 @@ void *text_poke_early(void *addr, const void *opcode, size_t len) */ void *__kprobes text_poke(void *addr, const void *opcode, size_t len) { - unsigned long flags; char *vaddr; int nr_pages = 2; struct page *pages[2]; int i; + might_sleep(); if (!core_kernel_text((unsigned long)addr)) { pages[0] = vmalloc_to_page(addr); pages[1] = vmalloc_to_page(addr + PAGE_SIZE); @@ -517,9 +522,9 @@ void *__kprobes text_poke(void *addr, const void *opcode, size_t len) nr_pages = 1; vaddr = vmap(pages, nr_pages, VM_MAP, PAGE_KERNEL); BUG_ON(!vaddr); - local_irq_save(flags); + local_irq_disable(); memcpy(&vaddr[(unsigned long)addr & ~PAGE_MASK], opcode, len); - local_irq_restore(flags); + local_irq_enable(); vunmap(vaddr); sync_core(); /* Could also do a CLFLUSH here to speed up CPU recovery; but diff --git a/arch/x86/kernel/apic/Makefile b/arch/x86/kernel/apic/Makefile new file mode 100644 index 00000000000..da7b7b9f8bd --- /dev/null +++ b/arch/x86/kernel/apic/Makefile @@ -0,0 +1,19 @@ +# +# 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-$(CONFIG_X86_IO_APIC) += io_apic.o +obj-$(CONFIG_SMP) += ipi.o + +ifeq ($(CONFIG_X86_64),y) +obj-y += apic_flat_64.o +obj-$(CONFIG_X86_X2APIC) += x2apic_cluster.o +obj-$(CONFIG_X86_X2APIC) += x2apic_phys.o +obj-$(CONFIG_X86_UV) += x2apic_uv_x.o +endif + +obj-$(CONFIG_X86_BIGSMP) += bigsmp_32.o +obj-$(CONFIG_X86_NUMAQ) += numaq_32.o +obj-$(CONFIG_X86_ES7000) += es7000_32.o +obj-$(CONFIG_X86_SUMMIT) += summit_32.o diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic/apic.c index 570f36e44e5..85eb8e10081 100644 --- a/arch/x86/kernel/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -1,7 +1,7 @@ /* * Local APIC handling, local APIC timers * - * (c) 1999, 2000 Ingo Molnar <mingo@redhat.com> + * (c) 1999, 2000, 2009 Ingo Molnar <mingo@redhat.com> * * Fixes * Maciej W. Rozycki : Bits for genuine 82489DX APICs; @@ -14,51 +14,70 @@ * Mikael Pettersson : PM converted to driver model. */ -#include <linux/init.h> - -#include <linux/mm.h> -#include <linux/delay.h> -#include <linux/bootmem.h> -#include <linux/interrupt.h> -#include <linux/mc146818rtc.h> #include <linux/kernel_stat.h> -#include <linux/sysdev.h> -#include <linux/ioport.h> -#include <linux/cpu.h> -#include <linux/clockchips.h> +#include <linux/mc146818rtc.h> #include <linux/acpi_pmtmr.h> +#include <linux/clockchips.h> +#include <linux/interrupt.h> +#include <linux/bootmem.h> +#include <linux/ftrace.h> +#include <linux/ioport.h> #include <linux/module.h> -#include <linux/dmi.h> +#include <linux/sysdev.h> +#include <linux/delay.h> +#include <linux/timex.h> #include <linux/dmar.h> -#include <linux/ftrace.h> -#include <linux/smp.h> +#include <linux/init.h> +#include <linux/cpu.h> +#include <linux/dmi.h> #include <linux/nmi.h> -#include <linux/timex.h> +#include <linux/smp.h> +#include <linux/mm.h> +#include <asm/pgalloc.h> #include <asm/atomic.h> -#include <asm/mtrr.h> #include <asm/mpspec.h> -#include <asm/desc.h> -#include <asm/arch_hooks.h> -#include <asm/hpet.h> -#include <asm/pgalloc.h> #include <asm/i8253.h> -#include <asm/idle.h> +#include <asm/i8259.h> #include <asm/proto.h> #include <asm/apic.h> -#include <asm/i8259.h> +#include <asm/desc.h> +#include <asm/hpet.h> +#include <asm/idle.h> +#include <asm/mtrr.h> #include <asm/smp.h> +#include <asm/mce.h> + +unsigned int num_processors; + +unsigned disabled_cpus __cpuinitdata; -#include <mach_apic.h> -#include <mach_apicdef.h> -#include <mach_ipi.h> +/* Processor that is doing the boot up */ +unsigned int boot_cpu_physical_apicid = -1U; /* - * Sanity check + * The highest APIC ID seen during enumeration. + * + * 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. */ -#if ((SPURIOUS_APIC_VECTOR & 0x0F) != 0x0F) -# error SPURIOUS_APIC_VECTOR definition error -#endif +unsigned int max_physical_apicid; + +/* + * Bitmask of physically existing CPUs: + */ +physid_mask_t phys_cpu_present_map; + +/* + * Map cpu index to physical APIC ID + */ +DEFINE_EARLY_PER_CPU(u16, x86_cpu_to_apicid, BAD_APICID); +DEFINE_EARLY_PER_CPU(u16, x86_bios_cpu_apicid, BAD_APICID); +EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_apicid); +EXPORT_EARLY_PER_CPU_SYMBOL(x86_bios_cpu_apicid); #ifdef CONFIG_X86_32 /* @@ -92,11 +111,7 @@ static __init int setup_apicpmtimer(char *s) __setup("apicpmtimer", setup_apicpmtimer); #endif -#ifdef CONFIG_X86_64 -#define HAVE_X2APIC -#endif - -#ifdef HAVE_X2APIC +#ifdef CONFIG_X86_X2APIC int x2apic; /* x2apic enabled before OS handover */ static int x2apic_preenabled; @@ -194,18 +209,13 @@ static int modern_apic(void) return lapic_get_version() >= 0x14; } -/* - * Paravirt kernels also might be using these below ops. So we still - * use generic apic_read()/apic_write(), which might be pointing to different - * ops in PARAVIRT case. - */ -void xapic_wait_icr_idle(void) +void native_apic_wait_icr_idle(void) { while (apic_read(APIC_ICR) & APIC_ICR_BUSY) cpu_relax(); } -u32 safe_xapic_wait_icr_idle(void) +u32 native_safe_apic_wait_icr_idle(void) { u32 send_status; int timeout; @@ -221,13 +231,13 @@ u32 safe_xapic_wait_icr_idle(void) return send_status; } -void xapic_icr_write(u32 low, u32 id) +void native_apic_icr_write(u32 low, u32 id) { apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(id)); apic_write(APIC_ICR, low); } -static u64 xapic_icr_read(void) +u64 native_apic_icr_read(void) { u32 icr1, icr2; @@ -237,54 +247,6 @@ static u64 xapic_icr_read(void) return icr1 | ((u64)icr2 << 32); } -static struct apic_ops xapic_ops = { - .read = native_apic_mem_read, - .write = native_apic_mem_write, - .icr_read = xapic_icr_read, - .icr_write = xapic_icr_write, - .wait_icr_idle = xapic_wait_icr_idle, - .safe_wait_icr_idle = safe_xapic_wait_icr_idle, -}; - -struct apic_ops __read_mostly *apic_ops = &xapic_ops; -EXPORT_SYMBOL_GPL(apic_ops); - -#ifdef HAVE_X2APIC -static void x2apic_wait_icr_idle(void) -{ - /* no need to wait for icr idle in x2apic */ - return; -} - -static u32 safe_x2apic_wait_icr_idle(void) -{ - /* no need to wait for icr idle in x2apic */ - return 0; -} - -void x2apic_icr_write(u32 low, u32 id) -{ - wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low); -} - -static u64 x2apic_icr_read(void) -{ - unsigned long val; - - rdmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), val); - return val; -} - -static struct apic_ops x2apic_ops = { - .read = native_apic_msr_read, - .write = native_apic_msr_write, - .icr_read = x2apic_icr_read, - .icr_write = x2apic_icr_write, - .wait_icr_idle = x2apic_wait_icr_idle, - .safe_wait_icr_idle = safe_x2apic_wait_icr_idle, -}; -#endif - /** * enable_NMI_through_LVT0 - enable NMI through local vector table 0 */ @@ -457,7 +419,7 @@ static void lapic_timer_setup(enum clock_event_mode mode, static void lapic_timer_broadcast(const struct cpumask *mask) { #ifdef CONFIG_SMP - send_IPI_mask(mask, LOCAL_TIMER_VECTOR); + apic->send_IPI_mask(mask, LOCAL_TIMER_VECTOR); #endif } @@ -535,7 +497,8 @@ static void __init lapic_cal_handler(struct clock_event_device *dev) } } -static int __init calibrate_by_pmtimer(long deltapm, long *delta) +static int __init +calibrate_by_pmtimer(long deltapm, long *delta, long *deltatsc) { const long pm_100ms = PMTMR_TICKS_PER_SEC / 10; const long pm_thresh = pm_100ms / 100; @@ -546,7 +509,7 @@ static int __init calibrate_by_pmtimer(long deltapm, long *delta) return -1; #endif - apic_printk(APIC_VERBOSE, "... PM timer delta = %ld\n", deltapm); + apic_printk(APIC_VERBOSE, "... PM-Timer delta = %ld\n", deltapm); /* Check, if the PM timer is available */ if (!deltapm) @@ -556,19 +519,30 @@ static int __init calibrate_by_pmtimer(long deltapm, long *delta) if (deltapm > (pm_100ms - pm_thresh) && deltapm < (pm_100ms + pm_thresh)) { - apic_printk(APIC_VERBOSE, "... PM timer result ok\n"); - } else { - res = (((u64)deltapm) * mult) >> 22; - do_div(res, 1000000); - pr_warning("APIC calibration not consistent " - "with PM Timer: %ldms instead of 100ms\n", - (long)res); - /* Correct the lapic counter value */ - res = (((u64)(*delta)) * pm_100ms); + apic_printk(APIC_VERBOSE, "... PM-Timer result ok\n"); + return 0; + } + + res = (((u64)deltapm) * mult) >> 22; + do_div(res, 1000000); + pr_warning("APIC calibration not consistent " + "with PM-Timer: %ldms instead of 100ms\n",(long)res); + + /* Correct the lapic counter value */ + res = (((u64)(*delta)) * pm_100ms); + do_div(res, deltapm); + pr_info("APIC delta adjusted to PM-Timer: " + "%lu (%ld)\n", (unsigned long)res, *delta); + *delta = (long)res; + + /* Correct the tsc counter value */ + if (cpu_has_tsc) { + res = (((u64)(*deltatsc)) * pm_100ms); do_div(res, deltapm); - pr_info("APIC delta adjusted to PM-Timer: " - "%lu (%ld)\n", (unsigned long)res, *delta); - *delta = (long)res; + apic_printk(APIC_VERBOSE, "TSC delta adjusted to " + "PM-Timer: %lu (%ld) \n", + (unsigned long)res, *deltatsc); + *deltatsc = (long)res; } return 0; @@ -579,7 +553,7 @@ static int __init calibrate_APIC_clock(void) struct clock_event_device *levt = &__get_cpu_var(lapic_events); void (*real_handler)(struct clock_event_device *dev); unsigned long deltaj; - long delta; + long delta, deltatsc; int pm_referenced = 0; local_irq_disable(); @@ -609,9 +583,11 @@ static int __init calibrate_APIC_clock(void) delta = lapic_cal_t1 - lapic_cal_t2; apic_printk(APIC_VERBOSE, "... lapic delta = %ld\n", delta); + deltatsc = (long)(lapic_cal_tsc2 - lapic_cal_tsc1); + /* we trust the PM based calibration if possible */ pm_referenced = !calibrate_by_pmtimer(lapic_cal_pm2 - lapic_cal_pm1, - &delta); + &delta, &deltatsc); /* Calculate the scaled math multiplication factor */ lapic_clockevent.mult = div_sc(delta, TICK_NSEC * LAPIC_CAL_LOOPS, @@ -629,11 +605,10 @@ static int __init calibrate_APIC_clock(void) calibration_result); if (cpu_has_tsc) { - delta = (long)(lapic_cal_tsc2 - lapic_cal_tsc1); apic_printk(APIC_VERBOSE, "..... CPU clock speed is " "%ld.%04ld MHz.\n", - (delta / LAPIC_CAL_LOOPS) / (1000000 / HZ), - (delta / LAPIC_CAL_LOOPS) % (1000000 / HZ)); + (deltatsc / LAPIC_CAL_LOOPS) / (1000000 / HZ), + (deltatsc / LAPIC_CAL_LOOPS) % (1000000 / HZ)); } apic_printk(APIC_VERBOSE, "..... host bus clock speed is " @@ -834,7 +809,7 @@ void clear_local_APIC(void) u32 v; /* APIC hasn't been mapped yet */ - if (!apic_phys) + if (!x2apic && !apic_phys) return; maxlvt = lapic_get_maxlvt(); @@ -868,6 +843,14 @@ void clear_local_APIC(void) apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED); } #endif +#ifdef CONFIG_X86_MCE_INTEL + if (maxlvt >= 6) { + v = apic_read(APIC_LVTCMCI); + if (!(v & APIC_LVT_MASKED)) + apic_write(APIC_LVTCMCI, v | APIC_LVT_MASKED); + } +#endif + /* * Clean APIC state for other OSs: */ @@ -991,11 +974,11 @@ int __init verify_local_APIC(void) */ reg0 = apic_read(APIC_ID); apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg0); - apic_write(APIC_ID, reg0 ^ APIC_ID_MASK); + apic_write(APIC_ID, reg0 ^ apic->apic_id_mask); reg1 = apic_read(APIC_ID); apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg1); apic_write(APIC_ID, reg0); - if (reg1 != (reg0 ^ APIC_ID_MASK)) + if (reg1 != (reg0 ^ apic->apic_id_mask)) return 0; /* @@ -1089,7 +1072,7 @@ static void __cpuinit lapic_setup_esr(void) return; } - if (esr_disable) { + if (apic->disable_esr) { /* * Something untraceable is creating bad interrupts on * secondary quads ... for the moment, just leave the @@ -1130,9 +1113,14 @@ void __cpuinit setup_local_APIC(void) unsigned int value; int i, j; + if (disable_apic) { + arch_disable_smp_support(); + return; + } + #ifdef CONFIG_X86_32 /* Pound the ESR really hard over the head with a big hammer - mbligh */ - if (lapic_is_integrated() && esr_disable) { + if (lapic_is_integrated() && apic->disable_esr) { apic_write(APIC_ESR, 0); apic_write(APIC_ESR, 0); apic_write(APIC_ESR, 0); @@ -1146,7 +1134,7 @@ void __cpuinit setup_local_APIC(void) * Double-check whether this APIC is really registered. * This is meaningless in clustered apic mode, so we skip it. */ - if (!apic_id_registered()) + if (!apic->apic_id_registered()) BUG(); /* @@ -1154,7 +1142,7 @@ void __cpuinit setup_local_APIC(void) * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel * document number 292116). So here it goes... */ - init_apic_ldr(); + apic->init_apic_ldr(); /* * Set Task Priority to 'accept all'. We never change this @@ -1262,6 +1250,12 @@ void __cpuinit setup_local_APIC(void) apic_write(APIC_LVT1, value); preempt_enable(); + +#ifdef CONFIG_X86_MCE_INTEL + /* Recheck CMCI information after local APIC is up on CPU #0 */ + if (smp_processor_id() == 0) + cmci_recheck(); +#endif } void __cpuinit end_local_APIC_setup(void) @@ -1282,17 +1276,12 @@ void __cpuinit end_local_APIC_setup(void) apic_pm_activate(); } -#ifdef HAVE_X2APIC +#ifdef CONFIG_X86_X2APIC void check_x2apic(void) { - int msr, msr2; - - rdmsr(MSR_IA32_APICBASE, msr, msr2); - - if (msr & X2APIC_ENABLE) { + if (x2apic_enabled()) { pr_info("x2apic enabled by BIOS, switching to x2apic ops\n"); x2apic_preenabled = x2apic = 1; - apic_ops = &x2apic_ops; } } @@ -1300,6 +1289,9 @@ void enable_x2apic(void) { int msr, msr2; + if (!x2apic) + return; + rdmsr(MSR_IA32_APICBASE, msr, msr2); if (!(msr & X2APIC_ENABLE)) { pr_info("Enabling x2apic\n"); @@ -1342,15 +1334,16 @@ void __init enable_IR_x2apic(void) return; } - local_irq_save(flags); - mask_8259A(); - - ret = save_mask_IO_APIC_setup(); + ret = save_IO_APIC_setup(); if (ret) { pr_info("Saving IO-APIC state failed: %d\n", ret); goto end; } + local_irq_save(flags); + mask_IO_APIC_setup(); + mask_8259A(); + ret = enable_intr_remapping(1); if (ret && x2apic_preenabled) { @@ -1363,7 +1356,6 @@ void __init enable_IR_x2apic(void) if (!x2apic) { x2apic = 1; - apic_ops = &x2apic_ops; enable_x2apic(); } @@ -1376,10 +1368,10 @@ end_restore: else reinit_intr_remapped_IO_APIC(x2apic_preenabled); -end: unmask_8259A(); local_irq_restore(flags); +end: if (!ret) { if (!x2apic_preenabled) pr_info("Enabled x2apic and interrupt-remapping\n"); @@ -1401,7 +1393,7 @@ end: return; } -#endif /* HAVE_X2APIC */ +#endif /* CONFIG_X86_X2APIC */ #ifdef CONFIG_X86_64 /* @@ -1532,12 +1524,10 @@ void __init early_init_lapic_mapping(void) */ void __init init_apic_mappings(void) { -#ifdef HAVE_X2APIC if (x2apic) { boot_cpu_physical_apicid = read_apic_id(); return; } -#endif /* * If no local APIC can be found then set up a fake all @@ -1570,11 +1560,11 @@ int apic_version[MAX_APICS]; int __init APIC_init_uniprocessor(void) { -#ifdef CONFIG_X86_64 if (disable_apic) { pr_info("Apic disabled\n"); return -1; } +#ifdef CONFIG_X86_64 if (!cpu_has_apic) { disable_apic = 1; pr_info("Apic disabled by BIOS\n"); @@ -1596,11 +1586,9 @@ int __init APIC_init_uniprocessor(void) } #endif -#ifdef HAVE_X2APIC enable_IR_x2apic(); -#endif #ifdef CONFIG_X86_64 - setup_apic_routing(); + default_setup_apic_routing(); #endif verify_local_APIC(); @@ -1621,35 +1609,31 @@ int __init APIC_init_uniprocessor(void) physid_set_mask_of_physid(boot_cpu_physical_apicid, &phys_cpu_present_map); setup_local_APIC(); -#ifdef CONFIG_X86_64 +#ifdef CONFIG_X86_IO_APIC /* * Now enable IO-APICs, actually call clear_IO_APIC - * We need clear_IO_APIC before enabling vector on BP + * We need clear_IO_APIC before enabling error vector */ if (!skip_ioapic_setup && nr_ioapics) enable_IO_APIC(); #endif -#ifdef CONFIG_X86_IO_APIC - if (!smp_found_config || skip_ioapic_setup || !nr_ioapics) -#endif - localise_nmi_watchdog(); end_local_APIC_setup(); #ifdef CONFIG_X86_IO_APIC if (smp_found_config && !skip_ioapic_setup && nr_ioapics) setup_IO_APIC(); -# ifdef CONFIG_X86_64 - else + else { nr_ioapics = 0; -# endif + localise_nmi_watchdog(); + } +#else + localise_nmi_watchdog(); #endif + setup_boot_clock(); #ifdef CONFIG_X86_64 - setup_boot_APIC_clock(); check_nmi_watchdog(); -#else - setup_boot_clock(); #endif return 0; @@ -1738,7 +1722,8 @@ void __init connect_bsp_APIC(void) outb(0x01, 0x23); } #endif - enable_apic_mode(); + if (apic->enable_apic_mode) + apic->enable_apic_mode(); } /** @@ -1876,29 +1861,39 @@ void __cpuinit generic_processor_info(int apicid, int version) } #endif -#if defined(CONFIG_X86_SMP) || defined(CONFIG_X86_64) - /* are we being called early in kernel startup? */ - if (early_per_cpu_ptr(x86_cpu_to_apicid)) { - u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid); - u16 *bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); - - cpu_to_apicid[cpu] = apicid; - bios_cpu_apicid[cpu] = apicid; - } else { - per_cpu(x86_cpu_to_apicid, cpu) = apicid; - per_cpu(x86_bios_cpu_apicid, cpu) = apicid; - } +#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; #endif set_cpu_possible(cpu, true); set_cpu_present(cpu, true); } -#ifdef CONFIG_X86_64 int hard_smp_processor_id(void) { return read_apic_id(); } + +void default_init_apic_ldr(void) +{ + unsigned long val; + + apic_write(APIC_DFR, APIC_DFR_VALUE); + val = apic_read(APIC_LDR) & ~APIC_LDR_MASK; + val |= SET_APIC_LOGICAL_ID(1UL << smp_processor_id()); + apic_write(APIC_LDR, val); +} + +#ifdef CONFIG_X86_32 +int default_apicid_to_node(int logical_apicid) +{ +#ifdef CONFIG_SMP + return apicid_2_node[hard_smp_processor_id()]; +#else + return 0; +#endif +} #endif /* @@ -1976,12 +1971,9 @@ static int lapic_resume(struct sys_device *dev) local_irq_save(flags); -#ifdef HAVE_X2APIC if (x2apic) enable_x2apic(); - else -#endif - { + else { /* * Make sure the APICBASE points to the right address * diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/apic/apic_flat_64.c index 34185488e4f..f933822dba1 100644 --- a/arch/x86/kernel/genapic_flat_64.c +++ b/arch/x86/kernel/apic/apic_flat_64.c @@ -17,9 +17,8 @@ #include <linux/init.h> #include <linux/hardirq.h> #include <asm/smp.h> +#include <asm/apic.h> #include <asm/ipi.h> -#include <asm/genapic.h> -#include <mach_apicdef.h> #ifdef CONFIG_ACPI #include <acpi/acpi_bus.h> @@ -74,7 +73,7 @@ static inline void _flat_send_IPI_mask(unsigned long mask, int vector) unsigned long flags; local_irq_save(flags); - __send_IPI_dest_field(mask, vector, APIC_DEST_LOGICAL); + __default_send_IPI_dest_field(mask, vector, apic->dest_logical); local_irq_restore(flags); } @@ -85,14 +84,15 @@ static void flat_send_IPI_mask(const struct cpumask *cpumask, int vector) _flat_send_IPI_mask(mask, vector); } -static void flat_send_IPI_mask_allbutself(const struct cpumask *cpumask, - int vector) +static void + flat_send_IPI_mask_allbutself(const struct cpumask *cpumask, int vector) { unsigned long mask = cpumask_bits(cpumask)[0]; int cpu = smp_processor_id(); if (cpu < BITS_PER_LONG) clear_bit(cpu, &mask); + _flat_send_IPI_mask(mask, vector); } @@ -114,23 +114,27 @@ static void flat_send_IPI_allbutself(int vector) _flat_send_IPI_mask(mask, vector); } } else if (num_online_cpus() > 1) { - __send_IPI_shortcut(APIC_DEST_ALLBUT, vector,APIC_DEST_LOGICAL); + __default_send_IPI_shortcut(APIC_DEST_ALLBUT, + vector, apic->dest_logical); } } static void flat_send_IPI_all(int vector) { - if (vector == NMI_VECTOR) + if (vector == NMI_VECTOR) { flat_send_IPI_mask(cpu_online_mask, vector); - else - __send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL); + } else { + __default_send_IPI_shortcut(APIC_DEST_ALLINC, + vector, apic->dest_logical); + } } -static unsigned int get_apic_id(unsigned long x) +static unsigned int flat_get_apic_id(unsigned long x) { unsigned int id; id = (((x)>>24) & 0xFFu); + return id; } @@ -146,7 +150,7 @@ static unsigned int read_xapic_id(void) { unsigned int id; - id = get_apic_id(apic_read(APIC_ID)); + id = flat_get_apic_id(apic_read(APIC_ID)); return id; } @@ -169,31 +173,67 @@ static unsigned int flat_cpu_mask_to_apicid_and(const struct cpumask *cpumask, return mask1 & mask2; } -static unsigned int phys_pkg_id(int index_msb) +static int flat_phys_pkg_id(int initial_apic_id, int index_msb) { return hard_smp_processor_id() >> index_msb; } -struct genapic apic_flat = { - .name = "flat", - .acpi_madt_oem_check = flat_acpi_madt_oem_check, - .int_delivery_mode = dest_LowestPrio, - .int_dest_mode = (APIC_DEST_LOGICAL != 0), - .target_cpus = flat_target_cpus, - .vector_allocation_domain = flat_vector_allocation_domain, - .apic_id_registered = flat_apic_id_registered, - .init_apic_ldr = flat_init_apic_ldr, - .send_IPI_all = flat_send_IPI_all, - .send_IPI_allbutself = flat_send_IPI_allbutself, - .send_IPI_mask = flat_send_IPI_mask, - .send_IPI_mask_allbutself = flat_send_IPI_mask_allbutself, - .send_IPI_self = apic_send_IPI_self, - .cpu_mask_to_apicid = flat_cpu_mask_to_apicid, - .cpu_mask_to_apicid_and = flat_cpu_mask_to_apicid_and, - .phys_pkg_id = phys_pkg_id, - .get_apic_id = get_apic_id, - .set_apic_id = set_apic_id, - .apic_id_mask = (0xFFu<<24), +struct apic apic_flat = { + .name = "flat", + .probe = NULL, + .acpi_madt_oem_check = flat_acpi_madt_oem_check, + .apic_id_registered = flat_apic_id_registered, + + .irq_delivery_mode = dest_LowestPrio, + .irq_dest_mode = 1, /* logical */ + + .target_cpus = flat_target_cpus, + .disable_esr = 0, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = flat_vector_allocation_domain, + .init_apic_ldr = flat_init_apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = flat_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = flat_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFu << 24, + + .cpu_mask_to_apicid = flat_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = flat_cpu_mask_to_apicid_and, + + .send_IPI_mask = flat_send_IPI_mask, + .send_IPI_mask_allbutself = flat_send_IPI_mask_allbutself, + .send_IPI_allbutself = flat_send_IPI_allbutself, + .send_IPI_all = flat_send_IPI_all, + .send_IPI_self = apic_send_IPI_self, + + .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 = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, }; /* @@ -232,18 +272,18 @@ static void physflat_vector_allocation_domain(int cpu, struct cpumask *retmask) static void physflat_send_IPI_mask(const struct cpumask *cpumask, int vector) { - send_IPI_mask_sequence(cpumask, vector); + default_send_IPI_mask_sequence_phys(cpumask, vector); } static void physflat_send_IPI_mask_allbutself(const struct cpumask *cpumask, int vector) { - send_IPI_mask_allbutself(cpumask, vector); + default_send_IPI_mask_allbutself_phys(cpumask, vector); } static void physflat_send_IPI_allbutself(int vector) { - send_IPI_mask_allbutself(cpu_online_mask, vector); + default_send_IPI_mask_allbutself_phys(cpu_online_mask, vector); } static void physflat_send_IPI_all(int vector) @@ -276,32 +316,72 @@ physflat_cpu_mask_to_apicid_and(const struct cpumask *cpumask, * We're using fixed IRQ delivery, can only return one phys APIC ID. * May as well be the first. */ - for_each_cpu_and(cpu, cpumask, andmask) + for_each_cpu_and(cpu, cpumask, andmask) { 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; } -struct genapic apic_physflat = { - .name = "physical flat", - .acpi_madt_oem_check = physflat_acpi_madt_oem_check, - .int_delivery_mode = dest_Fixed, - .int_dest_mode = (APIC_DEST_PHYSICAL != 0), - .target_cpus = physflat_target_cpus, - .vector_allocation_domain = physflat_vector_allocation_domain, - .apic_id_registered = flat_apic_id_registered, - .init_apic_ldr = flat_init_apic_ldr,/*not needed, but shouldn't hurt*/ - .send_IPI_all = physflat_send_IPI_all, - .send_IPI_allbutself = physflat_send_IPI_allbutself, - .send_IPI_mask = physflat_send_IPI_mask, - .send_IPI_mask_allbutself = physflat_send_IPI_mask_allbutself, - .send_IPI_self = apic_send_IPI_self, - .cpu_mask_to_apicid = physflat_cpu_mask_to_apicid, - .cpu_mask_to_apicid_and = physflat_cpu_mask_to_apicid_and, - .phys_pkg_id = phys_pkg_id, - .get_apic_id = get_apic_id, - .set_apic_id = set_apic_id, - .apic_id_mask = (0xFFu<<24), +struct apic apic_physflat = { + + .name = "physical flat", + .probe = NULL, + .acpi_madt_oem_check = physflat_acpi_madt_oem_check, + .apic_id_registered = flat_apic_id_registered, + + .irq_delivery_mode = dest_Fixed, + .irq_dest_mode = 0, /* physical */ + + .target_cpus = physflat_target_cpus, + .disable_esr = 0, + .dest_logical = 0, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = physflat_vector_allocation_domain, + /* not needed, but shouldn't hurt: */ + .init_apic_ldr = flat_init_apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = flat_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = flat_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFu << 24, + + .cpu_mask_to_apicid = physflat_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = physflat_cpu_mask_to_apicid_and, + + .send_IPI_mask = physflat_send_IPI_mask, + .send_IPI_mask_allbutself = physflat_send_IPI_mask_allbutself, + .send_IPI_allbutself = physflat_send_IPI_allbutself, + .send_IPI_all = physflat_send_IPI_all, + .send_IPI_self = apic_send_IPI_self, + + .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 = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, }; diff --git a/arch/x86/kernel/apic/bigsmp_32.c b/arch/x86/kernel/apic/bigsmp_32.c new file mode 100644 index 00000000000..d806ecaa948 --- /dev/null +++ b/arch/x86/kernel/apic/bigsmp_32.c @@ -0,0 +1,267 @@ +/* + * APIC driver for "bigsmp" xAPIC machines with more than 8 virtual CPUs. + * + * Drives the local APIC in "clustered mode". + */ +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/dmi.h> +#include <linux/smp.h> + +#include <asm/apicdef.h> +#include <asm/fixmap.h> +#include <asm/mpspec.h> +#include <asm/apic.h> +#include <asm/ipi.h> + +static unsigned bigsmp_get_apic_id(unsigned long x) +{ + return (x >> 24) & 0xFF; +} + +static int bigsmp_apic_id_registered(void) +{ + return 1; +} + +static const cpumask_t *bigsmp_target_cpus(void) +{ +#ifdef CONFIG_SMP + return &cpu_online_map; +#else + return &cpumask_of_cpu(0); +#endif +} + +static unsigned long bigsmp_check_apicid_used(physid_mask_t bitmap, int apicid) +{ + return 0; +} + +static unsigned long bigsmp_check_apicid_present(int bit) +{ + return 1; +} + +static inline unsigned long calculate_ldr(int cpu) +{ + unsigned long val, id; + + val = apic_read(APIC_LDR) & ~APIC_LDR_MASK; + id = per_cpu(x86_bios_cpu_apicid, cpu); + val |= SET_APIC_LOGICAL_ID(id); + + return val; +} + +/* + * Set up the logical destination ID. + * + * Intel recommends to set DFR, LDR and TPR before enabling + * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel + * document number 292116). So here it goes... + */ +static void bigsmp_init_apic_ldr(void) +{ + unsigned long val; + int cpu = smp_processor_id(); + + apic_write(APIC_DFR, APIC_DFR_FLAT); + val = calculate_ldr(cpu); + apic_write(APIC_LDR, val); +} + +static void bigsmp_setup_apic_routing(void) +{ + printk(KERN_INFO + "Enabling APIC mode: Physflat. Using %d I/O APICs\n", + nr_ioapics); +} + +static int bigsmp_apicid_to_node(int logical_apicid) +{ + return apicid_2_node[hard_smp_processor_id()]; +} + +static int bigsmp_cpu_present_to_apicid(int mps_cpu) +{ + if (mps_cpu < nr_cpu_ids) + return (int) per_cpu(x86_bios_cpu_apicid, 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) +{ + if (cpu >= nr_cpu_ids) + return BAD_APICID; + return cpu_physical_id(cpu); +} + +static physid_mask_t bigsmp_ioapic_phys_id_map(physid_mask_t phys_map) +{ + /* For clustered we don't have a good way to do this yet - hack */ + return physids_promote(0xFFL); +} + +static int bigsmp_check_phys_apicid_present(int boot_cpu_physical_apicid) +{ + return 1; +} + +/* As we are using single CPU as destination, pick only one CPU here */ +static unsigned int bigsmp_cpu_mask_to_apicid(const cpumask_t *cpumask) +{ + return bigsmp_cpu_to_logical_apicid(first_cpu(*cpumask)); +} + +static unsigned int bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + for_each_cpu_and(cpu, cpumask, andmask) { + if (cpumask_test_cpu(cpu, cpu_online_mask)) + break; + } + if (cpu < nr_cpu_ids) + return bigsmp_cpu_to_logical_apicid(cpu); + + return BAD_APICID; +} + +static int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb) +{ + return cpuid_apic >> index_msb; +} + +static inline void bigsmp_send_IPI_mask(const struct cpumask *mask, int vector) +{ + default_send_IPI_mask_sequence_phys(mask, vector); +} + +static void bigsmp_send_IPI_allbutself(int vector) +{ + default_send_IPI_mask_allbutself_phys(cpu_online_mask, vector); +} + +static void bigsmp_send_IPI_all(int vector) +{ + bigsmp_send_IPI_mask(cpu_online_mask, vector); +} + +static int dmi_bigsmp; /* can be set by dmi scanners */ + +static int hp_ht_bigsmp(const struct dmi_system_id *d) +{ + printk(KERN_NOTICE "%s detected: force use of apic=bigsmp\n", d->ident); + dmi_bigsmp = 1; + + return 0; +} + + +static const struct dmi_system_id bigsmp_dmi_table[] = { + { hp_ht_bigsmp, "HP ProLiant DL760 G2", + { DMI_MATCH(DMI_BIOS_VENDOR, "HP"), + DMI_MATCH(DMI_BIOS_VERSION, "P44-"), + } + }, + + { hp_ht_bigsmp, "HP ProLiant DL740", + { DMI_MATCH(DMI_BIOS_VENDOR, "HP"), + DMI_MATCH(DMI_BIOS_VERSION, "P47-"), + } + }, + { } /* NULL entry stops DMI scanning */ +}; + +static void bigsmp_vector_allocation_domain(int cpu, cpumask_t *retmask) +{ + cpus_clear(*retmask); + cpu_set(cpu, *retmask); +} + +static int probe_bigsmp(void) +{ + if (def_to_bigsmp) + dmi_bigsmp = 1; + else + dmi_check_system(bigsmp_dmi_table); + + return dmi_bigsmp; +} + +struct apic apic_bigsmp = { + + .name = "bigsmp", + .probe = probe_bigsmp, + .acpi_madt_oem_check = NULL, + .apic_id_registered = bigsmp_apic_id_registered, + + .irq_delivery_mode = dest_Fixed, + /* phys delivery to target CPU: */ + .irq_dest_mode = 0, + + .target_cpus = bigsmp_target_cpus, + .disable_esr = 1, + .dest_logical = 0, + .check_apicid_used = bigsmp_check_apicid_used, + .check_apicid_present = bigsmp_check_apicid_present, + + .vector_allocation_domain = bigsmp_vector_allocation_domain, + .init_apic_ldr = bigsmp_init_apic_ldr, + + .ioapic_phys_id_map = bigsmp_ioapic_phys_id_map, + .setup_apic_routing = bigsmp_setup_apic_routing, + .multi_timer_check = NULL, + .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, + .setup_portio_remap = NULL, + .check_phys_apicid_present = bigsmp_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = bigsmp_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = bigsmp_get_apic_id, + .set_apic_id = NULL, + .apic_id_mask = 0xFF << 24, + + .cpu_mask_to_apicid = bigsmp_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = bigsmp_cpu_mask_to_apicid_and, + + .send_IPI_mask = bigsmp_send_IPI_mask, + .send_IPI_mask_allbutself = NULL, + .send_IPI_allbutself = bigsmp_send_IPI_allbutself, + .send_IPI_all = bigsmp_send_IPI_all, + .send_IPI_self = default_send_IPI_self, + + .trampoline_phys_low = DEFAULT_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = DEFAULT_TRAMPOLINE_PHYS_HIGH, + + .wait_for_init_deassert = default_wait_for_init_deassert, + + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = default_inquire_remote_apic, + + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, +}; diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c new file mode 100644 index 00000000000..19588f2770e --- /dev/null +++ b/arch/x86/kernel/apic/es7000_32.c @@ -0,0 +1,780 @@ +/* + * Written by: Garry Forsgren, Unisys Corporation + * Natalie Protasevich, Unisys Corporation + * + * This file contains the code to configure and interface + * with Unisys ES7000 series hardware system manager. + * + * Copyright (c) 2003 Unisys Corporation. + * Copyright (C) 2009, Red Hat, Inc., Ingo Molnar + * + * All Rights Reserved. + * + * 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. + * + * This program is distributed in the hope that it would be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write the Free Software Foundation, Inc., 59 + * Temple Place - Suite 330, Boston MA 02111-1307, USA. + * + * Contact information: Unisys Corporation, Township Line & Union Meeting + * Roads-A, Unisys Way, Blue Bell, Pennsylvania, 19424, or: + * + * http://www.unisys.com + */ +#include <linux/notifier.h> +#include <linux/spinlock.h> +#include <linux/cpumask.h> +#include <linux/threads.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/reboot.h> +#include <linux/string.h> +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/acpi.h> +#include <linux/init.h> +#include <linux/nmi.h> +#include <linux/smp.h> +#include <linux/io.h> + +#include <asm/apicdef.h> +#include <asm/atomic.h> +#include <asm/fixmap.h> +#include <asm/mpspec.h> +#include <asm/setup.h> +#include <asm/apic.h> +#include <asm/ipi.h> + +/* + * ES7000 chipsets + */ + +#define NON_UNISYS 0 +#define ES7000_CLASSIC 1 +#define ES7000_ZORRO 2 + +#define MIP_REG 1 +#define MIP_PSAI_REG 4 + +#define MIP_BUSY 1 +#define MIP_SPIN 0xf0000 +#define MIP_VALID 0x0100000000000000ULL +#define MIP_SW_APIC 0x1020b + +#define MIP_PORT(val) ((val >> 32) & 0xffff) + +#define MIP_RD_LO(val) (val & 0xffffffff) + +struct mip_reg { + unsigned long long off_0x00; + unsigned long long off_0x08; + unsigned long long off_0x10; + unsigned long long off_0x18; + unsigned long long off_0x20; + unsigned long long off_0x28; + unsigned long long off_0x30; + unsigned long long off_0x38; +}; + +struct mip_reg_info { + unsigned long long mip_info; + unsigned long long delivery_info; + unsigned long long host_reg; + unsigned long long mip_reg; +}; + +struct psai { + unsigned long long entry_type; + unsigned long long addr; + unsigned long long bep_addr; +}; + +#ifdef CONFIG_ACPI + +struct es7000_oem_table { + struct acpi_table_header Header; + u32 OEMTableAddr; + u32 OEMTableSize; +}; + +static unsigned long oem_addrX; +static unsigned long oem_size; + +#endif + +/* + * ES7000 Globals + */ + +static volatile unsigned long *psai; +static struct mip_reg *mip_reg; +static struct mip_reg *host_reg; +static int mip_port; +static unsigned long mip_addr; +static unsigned long host_addr; + +int es7000_plat; + +/* + * GSI override for ES7000 platforms. + */ + +static unsigned int base; + +static int +es7000_rename_gsi(int ioapic, int gsi) +{ + if (es7000_plat == ES7000_ZORRO) + return gsi; + + if (!base) { + int i; + for (i = 0; i < nr_ioapics; i++) + base += nr_ioapic_registers[i]; + } + + if (!ioapic && (gsi < 16)) + gsi += base; + + return gsi; +} + +static int wakeup_secondary_cpu_via_mip(int cpu, unsigned long eip) +{ + unsigned long vect = 0, psaival = 0; + + if (psai == NULL) + return -1; + + vect = ((unsigned long)__pa(eip)/0x1000) << 16; + psaival = (0x1000000 | vect | cpu); + + while (*psai & 0x1000000) + ; + + *psai = psaival; + + return 0; +} + +static int es7000_apic_is_cluster(void) +{ + /* MPENTIUMIII */ + if (boot_cpu_data.x86 == 6 && + (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11)) + return 1; + + return 0; +} + +static void setup_unisys(void) +{ + /* + * Determine the generation of the ES7000 currently running. + * + * es7000_plat = 1 if the machine is a 5xx ES7000 box + * es7000_plat = 2 if the machine is a x86_64 ES7000 box + * + */ + if (!(boot_cpu_data.x86 <= 15 && boot_cpu_data.x86_model <= 2)) + es7000_plat = ES7000_ZORRO; + else + es7000_plat = ES7000_CLASSIC; + ioapic_renumber_irq = es7000_rename_gsi; +} + +/* + * Parse the OEM Table: + */ +static int parse_unisys_oem(char *oemptr) +{ + int i; + int success = 0; + unsigned char type, size; + unsigned long val; + char *tp = NULL; + struct psai *psaip = NULL; + struct mip_reg_info *mi; + struct mip_reg *host, *mip; + + tp = oemptr; + + tp += 8; + + for (i = 0; i <= 6; i++) { + type = *tp++; + size = *tp++; + tp -= 2; + switch (type) { + case MIP_REG: + mi = (struct mip_reg_info *)tp; + val = MIP_RD_LO(mi->host_reg); + host_addr = val; + host = (struct mip_reg *)val; + host_reg = __va(host); + val = MIP_RD_LO(mi->mip_reg); + mip_port = MIP_PORT(mi->mip_info); + mip_addr = val; + mip = (struct mip_reg *)val; + mip_reg = __va(mip); + pr_debug("es7000_mipcfg: host_reg = 0x%lx \n", + (unsigned long)host_reg); + pr_debug("es7000_mipcfg: mip_reg = 0x%lx \n", + (unsigned long)mip_reg); + success++; + break; + case MIP_PSAI_REG: + psaip = (struct psai *)tp; + if (tp != NULL) { + if (psaip->addr) + psai = __va(psaip->addr); + else + psai = NULL; + success++; + } + break; + default: + break; + } + tp += size; + } + + if (success < 2) + es7000_plat = NON_UNISYS; + else + setup_unisys(); + + return es7000_plat; +} + +#ifdef CONFIG_ACPI +static int find_unisys_acpi_oem_table(unsigned long *oem_addr) +{ + struct acpi_table_header *header = NULL; + struct es7000_oem_table *table; + acpi_size tbl_size; + acpi_status ret; + int i = 0; + + for (;;) { + ret = acpi_get_table_with_size("OEM1", i++, &header, &tbl_size); + if (!ACPI_SUCCESS(ret)) + return -1; + + if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) + break; + + early_acpi_os_unmap_memory(header, tbl_size); + } + + table = (void *)header; + + oem_addrX = table->OEMTableAddr; + oem_size = table->OEMTableSize; + + early_acpi_os_unmap_memory(header, tbl_size); + + *oem_addr = (unsigned long)__acpi_map_table(oem_addrX, oem_size); + + return 0; +} + +static void unmap_unisys_acpi_oem_table(unsigned long oem_addr) +{ + if (!oem_addr) + return; + + __acpi_unmap_table((char *)oem_addr, oem_size); +} + +static int es7000_check_dsdt(void) +{ + struct acpi_table_header header; + + if (ACPI_SUCCESS(acpi_get_table_header(ACPI_SIG_DSDT, 0, &header)) && + !strncmp(header.oem_id, "UNISYS", 6)) + return 1; + return 0; +} + +static int es7000_acpi_ret; + +/* Hook from generic ACPI tables.c */ +static int es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + unsigned long oem_addr = 0; + int check_dsdt; + int ret = 0; + + /* check dsdt at first to avoid clear fix_map for oem_addr */ + check_dsdt = es7000_check_dsdt(); + + if (!find_unisys_acpi_oem_table(&oem_addr)) { + if (check_dsdt) { + ret = parse_unisys_oem((char *)oem_addr); + } else { + setup_unisys(); + ret = 1; + } + /* + * we need to unmap it + */ + unmap_unisys_acpi_oem_table(oem_addr); + } + + es7000_acpi_ret = ret; + + return ret && !es7000_apic_is_cluster(); +} + +static int es7000_acpi_madt_oem_check_cluster(char *oem_id, char *oem_table_id) +{ + int ret = es7000_acpi_ret; + + return ret && es7000_apic_is_cluster(); +} + +#else /* !CONFIG_ACPI: */ +static int es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + return 0; +} + +static int es7000_acpi_madt_oem_check_cluster(char *oem_id, char *oem_table_id) +{ + return 0; +} +#endif /* !CONFIG_ACPI */ + +static void es7000_spin(int n) +{ + int i = 0; + + while (i++ < n) + rep_nop(); +} + +static int es7000_mip_write(struct mip_reg *mip_reg) +{ + int status = 0; + int spin; + + spin = MIP_SPIN; + while ((host_reg->off_0x38 & MIP_VALID) != 0) { + if (--spin <= 0) { + WARN(1, "Timeout waiting for Host Valid Flag\n"); + return -1; + } + es7000_spin(MIP_SPIN); + } + + memcpy(host_reg, mip_reg, sizeof(struct mip_reg)); + outb(1, mip_port); + + spin = MIP_SPIN; + + while ((mip_reg->off_0x38 & MIP_VALID) == 0) { + if (--spin <= 0) { + WARN(1, "Timeout waiting for MIP Valid Flag\n"); + return -1; + } + es7000_spin(MIP_SPIN); + } + + status = (mip_reg->off_0x00 & 0xffff0000000000ULL) >> 48; + mip_reg->off_0x38 &= ~MIP_VALID; + + return status; +} + +static void es7000_enable_apic_mode(void) +{ + struct mip_reg es7000_mip_reg; + int mip_status; + + if (!es7000_plat) + return; + + printk(KERN_INFO "ES7000: 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; + + while ((mip_status = es7000_mip_write(&es7000_mip_reg)) != 0) + WARN(1, "Command failed, status = %x\n", mip_status); +} + +static void es7000_vector_allocation_domain(int cpu, cpumask_t *retmask) +{ + /* Careful. Some cpus do not strictly honor the set of cpus + * specified in the interrupt destination when using lowest + * priority interrupt delivery mode. + * + * In particular there was a hyperthreading cpu observed to + * deliver interrupts to the wrong hyperthread when only one + * hyperthread was specified in the interrupt desitination. + */ + *retmask = (cpumask_t){ { [0] = APIC_ALL_CPUS, } }; +} + + +static void es7000_wait_for_init_deassert(atomic_t *deassert) +{ + while (!atomic_read(deassert)) + cpu_relax(); +} + +static unsigned int es7000_get_apic_id(unsigned long x) +{ + return (x >> 24) & 0xFF; +} + +static void es7000_send_IPI_mask(const struct cpumask *mask, int vector) +{ + default_send_IPI_mask_sequence_phys(mask, vector); +} + +static void es7000_send_IPI_allbutself(int vector) +{ + default_send_IPI_mask_allbutself_phys(cpu_online_mask, vector); +} + +static void es7000_send_IPI_all(int vector) +{ + es7000_send_IPI_mask(cpu_online_mask, vector); +} + +static int es7000_apic_id_registered(void) +{ + return 1; +} + +static const cpumask_t *target_cpus_cluster(void) +{ + return &CPU_MASK_ALL; +} + +static const cpumask_t *es7000_target_cpus(void) +{ + return &cpumask_of_cpu(smp_processor_id()); +} + +static unsigned long +es7000_check_apicid_used(physid_mask_t bitmap, int apicid) +{ + return 0; +} +static unsigned long es7000_check_apicid_present(int bit) +{ + return physid_isset(bit, phys_cpu_present_map); +} + +static unsigned long calculate_ldr(int cpu) +{ + unsigned long id = per_cpu(x86_bios_cpu_apicid, cpu); + + return SET_APIC_LOGICAL_ID(id); +} + +/* + * Set up the logical destination ID. + * + * Intel recommends to set DFR, LdR and TPR before enabling + * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel + * document number 292116). So here it goes... + */ +static void es7000_init_apic_ldr_cluster(void) +{ + unsigned long val; + int cpu = smp_processor_id(); + + apic_write(APIC_DFR, APIC_DFR_CLUSTER); + val = calculate_ldr(cpu); + apic_write(APIC_LDR, val); +} + +static void es7000_init_apic_ldr(void) +{ + unsigned long val; + int cpu = smp_processor_id(); + + apic_write(APIC_DFR, APIC_DFR_FLAT); + val = calculate_ldr(cpu); + apic_write(APIC_LDR, val); +} + +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", + (apic_version[apic] == 0x14) ? + "Physical Cluster" : "Logical Cluster", + nr_ioapics, cpus_addr(*es7000_target_cpus())[0]); +} + +static int es7000_apicid_to_node(int logical_apicid) +{ + return 0; +} + + +static int es7000_cpu_present_to_apicid(int mps_cpu) +{ + if (!mps_cpu) + return boot_cpu_physical_apicid; + else if (mps_cpu < nr_cpu_ids) + return per_cpu(x86_bios_cpu_apicid, mps_cpu); + else + return BAD_APICID; +} + +static int cpu_id; + +static physid_mask_t es7000_apicid_to_cpu_present(int phys_apicid) +{ + physid_mask_t mask; + + mask = physid_mask_of_physid(cpu_id); + ++cpu_id; + + return mask; +} + +/* Mapping from cpu number to logical apicid */ +static int es7000_cpu_to_logical_apicid(int cpu) +{ +#ifdef CONFIG_SMP + if (cpu >= nr_cpu_ids) + return BAD_APICID; + return cpu_2_logical_apicid[cpu]; +#else + return logical_smp_processor_id(); +#endif +} + +static physid_mask_t es7000_ioapic_phys_id_map(physid_mask_t phys_map) +{ + /* For clustered we don't have a good way to do this yet - hack */ + return physids_promote(0xff); +} + +static int es7000_check_phys_apicid_present(int cpu_physical_apicid) +{ + boot_cpu_physical_apicid = read_apic_id(); + return 1; +} + +static unsigned int es7000_cpu_mask_to_apicid(const cpumask_t *cpumask) +{ + unsigned int round = 0; + int cpu, uninitialized_var(apicid); + + /* + * The cpus in the mask must all be on the apic cluster. + */ + for_each_cpu(cpu, cpumask) { + int new_apicid = es7000_cpu_to_logical_apicid(cpu); + + if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) { + WARN(1, "Not a valid mask!"); + + return BAD_APICID; + } + apicid = new_apicid; + round++; + } + return apicid; +} + +static unsigned int +es7000_cpu_mask_to_apicid_and(const struct cpumask *inmask, + const struct cpumask *andmask) +{ + int apicid = es7000_cpu_to_logical_apicid(0); + cpumask_var_t cpumask; + + if (!alloc_cpumask_var(&cpumask, GFP_ATOMIC)) + return apicid; + + cpumask_and(cpumask, inmask, andmask); + cpumask_and(cpumask, cpumask, cpu_online_mask); + apicid = es7000_cpu_mask_to_apicid(cpumask); + + free_cpumask_var(cpumask); + + return apicid; +} + +static int es7000_phys_pkg_id(int cpuid_apic, int index_msb) +{ + return cpuid_apic >> index_msb; +} + +static int probe_es7000(void) +{ + /* probed later in mptable/ACPI hooks */ + return 0; +} + +static int es7000_mps_ret; +static int es7000_mps_oem_check(struct mpc_table *mpc, char *oem, + char *productid) +{ + int ret = 0; + + if (mpc->oemptr) { + struct mpc_oemtable *oem_table = + (struct mpc_oemtable *)mpc->oemptr; + + if (!strncmp(oem, "UNISYS", 6)) + ret = parse_unisys_oem((char *)oem_table); + } + + es7000_mps_ret = ret; + + return ret && !es7000_apic_is_cluster(); +} + +static int es7000_mps_oem_check_cluster(struct mpc_table *mpc, char *oem, + char *productid) +{ + int ret = es7000_mps_ret; + + return ret && es7000_apic_is_cluster(); +} + +struct apic apic_es7000_cluster = { + + .name = "es7000", + .probe = probe_es7000, + .acpi_madt_oem_check = es7000_acpi_madt_oem_check_cluster, + .apic_id_registered = es7000_apic_id_registered, + + .irq_delivery_mode = dest_LowestPrio, + /* logical delivery broadcast to all procs: */ + .irq_dest_mode = 1, + + .target_cpus = target_cpus_cluster, + .disable_esr = 1, + .dest_logical = 0, + .check_apicid_used = es7000_check_apicid_used, + .check_apicid_present = es7000_check_apicid_present, + + .vector_allocation_domain = es7000_vector_allocation_domain, + .init_apic_ldr = es7000_init_apic_ldr_cluster, + + .ioapic_phys_id_map = es7000_ioapic_phys_id_map, + .setup_apic_routing = es7000_setup_apic_routing, + .multi_timer_check = NULL, + .apicid_to_node = es7000_apicid_to_node, + .cpu_to_logical_apicid = es7000_cpu_to_logical_apicid, + .cpu_present_to_apicid = es7000_cpu_present_to_apicid, + .apicid_to_cpu_present = es7000_apicid_to_cpu_present, + .setup_portio_remap = NULL, + .check_phys_apicid_present = es7000_check_phys_apicid_present, + .enable_apic_mode = es7000_enable_apic_mode, + .phys_pkg_id = es7000_phys_pkg_id, + .mps_oem_check = es7000_mps_oem_check_cluster, + + .get_apic_id = es7000_get_apic_id, + .set_apic_id = NULL, + .apic_id_mask = 0xFF << 24, + + .cpu_mask_to_apicid = es7000_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = es7000_cpu_mask_to_apicid_and, + + .send_IPI_mask = es7000_send_IPI_mask, + .send_IPI_mask_allbutself = NULL, + .send_IPI_allbutself = es7000_send_IPI_allbutself, + .send_IPI_all = es7000_send_IPI_all, + .send_IPI_self = default_send_IPI_self, + + .wakeup_secondary_cpu = wakeup_secondary_cpu_via_mip, + + .trampoline_phys_low = 0x467, + .trampoline_phys_high = 0x469, + + .wait_for_init_deassert = NULL, + + /* Nothing to do for most platforms, since cleared by the INIT cycle: */ + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = default_inquire_remote_apic, + + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, +}; + +struct apic apic_es7000 = { + + .name = "es7000", + .probe = probe_es7000, + .acpi_madt_oem_check = es7000_acpi_madt_oem_check, + .apic_id_registered = es7000_apic_id_registered, + + .irq_delivery_mode = dest_Fixed, + /* phys delivery to target CPUs: */ + .irq_dest_mode = 0, + + .target_cpus = es7000_target_cpus, + .disable_esr = 1, + .dest_logical = 0, + .check_apicid_used = es7000_check_apicid_used, + .check_apicid_present = es7000_check_apicid_present, + + .vector_allocation_domain = es7000_vector_allocation_domain, + .init_apic_ldr = es7000_init_apic_ldr, + + .ioapic_phys_id_map = es7000_ioapic_phys_id_map, + .setup_apic_routing = es7000_setup_apic_routing, + .multi_timer_check = NULL, + .apicid_to_node = es7000_apicid_to_node, + .cpu_to_logical_apicid = es7000_cpu_to_logical_apicid, + .cpu_present_to_apicid = es7000_cpu_present_to_apicid, + .apicid_to_cpu_present = es7000_apicid_to_cpu_present, + .setup_portio_remap = NULL, + .check_phys_apicid_present = es7000_check_phys_apicid_present, + .enable_apic_mode = es7000_enable_apic_mode, + .phys_pkg_id = es7000_phys_pkg_id, + .mps_oem_check = es7000_mps_oem_check, + + .get_apic_id = es7000_get_apic_id, + .set_apic_id = NULL, + .apic_id_mask = 0xFF << 24, + + .cpu_mask_to_apicid = es7000_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = es7000_cpu_mask_to_apicid_and, + + .send_IPI_mask = es7000_send_IPI_mask, + .send_IPI_mask_allbutself = NULL, + .send_IPI_allbutself = es7000_send_IPI_allbutself, + .send_IPI_all = es7000_send_IPI_all, + .send_IPI_self = default_send_IPI_self, + + .trampoline_phys_low = 0x467, + .trampoline_phys_high = 0x469, + + .wait_for_init_deassert = es7000_wait_for_init_deassert, + + /* Nothing to do for most platforms, since cleared by the INIT cycle: */ + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = default_inquire_remote_apic, + + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, +}; diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/apic/io_apic.c index bc7ac4da90d..42cdc78427a 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -1,7 +1,7 @@ /* * Intel IO-APIC support for multi-Pentium hosts. * - * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar, Hajnalka Szabo + * Copyright (C) 1997, 1998, 1999, 2000, 2009 Ingo Molnar, Hajnalka Szabo * * Many thanks to Stig Venaas for trying out countless experimental * patches and reporting/debugging problems patiently! @@ -46,6 +46,7 @@ #include <asm/idle.h> #include <asm/io.h> #include <asm/smp.h> +#include <asm/cpu.h> #include <asm/desc.h> #include <asm/proto.h> #include <asm/acpi.h> @@ -61,9 +62,7 @@ #include <asm/uv/uv_hub.h> #include <asm/uv/uv_irq.h> -#include <mach_ipi.h> -#include <mach_apic.h> -#include <mach_apicdef.h> +#include <asm/apic.h> #define __apicdebuginit(type) static type __init @@ -82,11 +81,11 @@ static DEFINE_SPINLOCK(vector_lock); int nr_ioapic_registers[MAX_IO_APICS]; /* I/O APIC entries */ -struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; +struct mpc_ioapic mp_ioapics[MAX_IO_APICS]; int nr_ioapics; /* MP IRQ source entries */ -struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; +struct mpc_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* # of MP IRQ source entries */ int mp_irq_entries; @@ -99,10 +98,19 @@ DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); int skip_ioapic_setup; +void arch_disable_smp_support(void) +{ +#ifdef CONFIG_PCI + noioapicquirk = 1; + noioapicreroute = -1; +#endif + skip_ioapic_setup = 1; +} + static int __init parse_noapic(char *str) { /* disable IO-APIC */ - disable_ioapic_setup(); + arch_disable_smp_support(); return 0; } early_param("noapic", parse_noapic); @@ -356,7 +364,7 @@ set_extra_move_desc(struct irq_desc *desc, const struct cpumask *mask) if (!cfg->move_in_progress) { /* it means that domain is not changed */ - if (!cpumask_intersects(&desc->affinity, mask)) + if (!cpumask_intersects(desc->affinity, mask)) cfg->move_desc_pending = 1; } } @@ -381,12 +389,20 @@ struct io_apic { unsigned int index; unsigned int unused[3]; unsigned int data; + unsigned int unused2[11]; + unsigned int eoi; }; static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) { return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) - + (mp_ioapics[idx].mp_apicaddr & ~PAGE_MASK); + + (mp_ioapics[idx].apicaddr & ~PAGE_MASK); +} + +static inline void io_apic_eoi(unsigned int apic, unsigned int vector) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(vector, &io_apic->eoi); } static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) @@ -478,7 +494,7 @@ __ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) io_apic_write(apic, 0x10 + 2*pin, eu.w1); } -static void 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); @@ -513,11 +529,11 @@ static void send_cleanup_vector(struct irq_cfg *cfg) 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) - send_IPI_mask(cpumask_of(i), IRQ_MOVE_CLEANUP_VECTOR); + 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); - send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR); + apic->send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR); free_cpumask_var(cleanup_mask); } cfg->move_in_progress = 0; @@ -538,16 +554,12 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq apic = entry->apic; pin = entry->pin; -#ifdef CONFIG_INTR_REMAP /* * With interrupt-remapping, destination information comes * from interrupt-remapping table entry. */ if (!irq_remapped(irq)) io_apic_write(apic, 0x11 + pin*2, dest); -#else - io_apic_write(apic, 0x11 + pin*2, dest); -#endif reg = io_apic_read(apic, 0x10 + pin*2); reg &= ~IO_APIC_REDIR_VECTOR_MASK; reg |= vector; @@ -562,8 +574,9 @@ 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 leaves desc->affinity untouched. + * Either sets desc->affinity to a valid value, and returns + * ->cpu_mask_to_apicid of that, or returns BAD_APICID and + * leaves desc->affinity untouched. */ static unsigned int set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask) @@ -579,9 +592,10 @@ set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask) if (assign_irq_vector(irq, cfg, mask)) return BAD_APICID; - cpumask_and(&desc->affinity, cfg->domain, mask); + cpumask_and(desc->affinity, cfg->domain, mask); set_extra_move_desc(desc, mask); - return cpu_mask_to_apicid_and(&desc->affinity, cpu_online_mask); + + return apic->cpu_mask_to_apicid_and(desc->affinity, cpu_online_mask); } static void @@ -796,23 +810,6 @@ static void clear_IO_APIC (void) clear_IO_APIC_pin(apic, pin); } -#if !defined(CONFIG_SMP) && defined(CONFIG_X86_32) -void send_IPI_self(int vector) -{ - unsigned int cfg; - - /* - * Wait for idle. - */ - apic_wait_icr_idle(); - cfg = APIC_DM_FIXED | APIC_DEST_SELF | vector | APIC_DEST_LOGICAL; - /* - * Send the IPI. The write to APIC_ICR fires this off. - */ - apic_write(APIC_ICR, cfg); -} -#endif /* !CONFIG_SMP && CONFIG_X86_32*/ - #ifdef CONFIG_X86_32 /* * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to @@ -820,8 +817,9 @@ void send_IPI_self(int vector) */ #define MAX_PIRQS 8 -static int pirq_entries [MAX_PIRQS]; -static int pirqs_enabled; +static int pirq_entries[MAX_PIRQS] = { + [0 ... MAX_PIRQS - 1] = -1 +}; static int __init ioapic_pirq_setup(char *str) { @@ -830,10 +828,6 @@ static int __init ioapic_pirq_setup(char *str) get_options(str, ARRAY_SIZE(ints), ints); - for (i = 0; i < MAX_PIRQS; i++) - pirq_entries[i] = -1; - - pirqs_enabled = 1; apic_printk(APIC_VERBOSE, KERN_INFO "PIRQ redirection, working around broken MP-BIOS.\n"); max = MAX_PIRQS; @@ -859,9 +853,9 @@ __setup("pirq=", ioapic_pirq_setup); static struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS]; /* - * Saves and masks all the unmasked IO-APIC RTE's + * Saves all the IO-APIC RTE's */ -int save_mask_IO_APIC_setup(void) +int save_IO_APIC_setup(void) { union IO_APIC_reg_01 reg_01; unsigned long flags; @@ -886,16 +880,9 @@ int save_mask_IO_APIC_setup(void) } for (apic = 0; apic < nr_ioapics; apic++) - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - struct IO_APIC_route_entry entry; - - entry = early_ioapic_entries[apic][pin] = + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) + early_ioapic_entries[apic][pin] = ioapic_read_entry(apic, pin); - if (!entry.mask) { - entry.mask = 1; - ioapic_write_entry(apic, pin, entry); - } - } return 0; @@ -908,6 +895,25 @@ nomem: return -ENOMEM; } +void mask_IO_APIC_setup(void) +{ + int apic, pin; + + for (apic = 0; apic < nr_ioapics; apic++) { + if (!early_ioapic_entries[apic]) + break; + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + struct IO_APIC_route_entry entry; + + entry = early_ioapic_entries[apic][pin]; + if (!entry.mask) { + entry.mask = 1; + ioapic_write_entry(apic, pin, entry); + } + } + } +} + void restore_IO_APIC_setup(void) { int apic, pin; @@ -944,10 +950,10 @@ static int find_irq_entry(int apic, int pin, int type) int i; for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mp_irqtype == type && - (mp_irqs[i].mp_dstapic == mp_ioapics[apic].mp_apicid || - mp_irqs[i].mp_dstapic == MP_APIC_ALL) && - mp_irqs[i].mp_dstirq == pin) + if (mp_irqs[i].irqtype == type && + (mp_irqs[i].dstapic == mp_ioapics[apic].apicid || + mp_irqs[i].dstapic == MP_APIC_ALL) && + mp_irqs[i].dstirq == pin) return i; return -1; @@ -961,13 +967,13 @@ static int __init find_isa_irq_pin(int irq, int type) int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mp_srcbus; + int lbus = mp_irqs[i].srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mp_irqtype == type) && - (mp_irqs[i].mp_srcbusirq == irq)) + (mp_irqs[i].irqtype == type) && + (mp_irqs[i].srcbusirq == irq)) - return mp_irqs[i].mp_dstirq; + return mp_irqs[i].dstirq; } return -1; } @@ -977,17 +983,17 @@ static int __init find_isa_irq_apic(int irq, int type) int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mp_srcbus; + int lbus = mp_irqs[i].srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mp_irqtype == type) && - (mp_irqs[i].mp_srcbusirq == irq)) + (mp_irqs[i].irqtype == type) && + (mp_irqs[i].srcbusirq == irq)) break; } if (i < mp_irq_entries) { int apic; for(apic = 0; apic < nr_ioapics; apic++) { - if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic) + if (mp_ioapics[apic].apicid == mp_irqs[i].dstapic) return apic; } } @@ -1012,23 +1018,23 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) return -1; } for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mp_srcbus; + int lbus = mp_irqs[i].srcbus; for (apic = 0; apic < nr_ioapics; apic++) - if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic || - mp_irqs[i].mp_dstapic == MP_APIC_ALL) + if (mp_ioapics[apic].apicid == mp_irqs[i].dstapic || + mp_irqs[i].dstapic == MP_APIC_ALL) break; if (!test_bit(lbus, mp_bus_not_pci) && - !mp_irqs[i].mp_irqtype && + !mp_irqs[i].irqtype && (bus == lbus) && - (slot == ((mp_irqs[i].mp_srcbusirq >> 2) & 0x1f))) { - int irq = pin_2_irq(i,apic,mp_irqs[i].mp_dstirq); + (slot == ((mp_irqs[i].srcbusirq >> 2) & 0x1f))) { + int irq = pin_2_irq(i, apic, mp_irqs[i].dstirq); if (!(apic || IO_APIC_IRQ(irq))) continue; - if (pin == (mp_irqs[i].mp_srcbusirq & 3)) + if (pin == (mp_irqs[i].srcbusirq & 3)) return irq; /* * Use the first all-but-pin matching entry as a @@ -1071,7 +1077,7 @@ static int EISA_ELCR(unsigned int irq) * EISA conforming in the MP table, that means its trigger type must * be read in from the ELCR */ -#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mp_srcbusirq)) +#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].srcbusirq)) #define default_EISA_polarity(idx) default_ISA_polarity(idx) /* PCI interrupts are always polarity one level triggered, @@ -1088,13 +1094,13 @@ static int EISA_ELCR(unsigned int irq) static int MPBIOS_polarity(int idx) { - int bus = mp_irqs[idx].mp_srcbus; + int bus = mp_irqs[idx].srcbus; int polarity; /* * Determine IRQ line polarity (high active or low active): */ - switch (mp_irqs[idx].mp_irqflag & 3) + switch (mp_irqs[idx].irqflag & 3) { case 0: /* conforms, ie. bus-type dependent polarity */ if (test_bit(bus, mp_bus_not_pci)) @@ -1130,13 +1136,13 @@ static int MPBIOS_polarity(int idx) static int MPBIOS_trigger(int idx) { - int bus = mp_irqs[idx].mp_srcbus; + int bus = mp_irqs[idx].srcbus; int trigger; /* * Determine IRQ trigger mode (edge or level sensitive): */ - switch ((mp_irqs[idx].mp_irqflag>>2) & 3) + switch ((mp_irqs[idx].irqflag>>2) & 3) { case 0: /* conforms, ie. bus-type dependent */ if (test_bit(bus, mp_bus_not_pci)) @@ -1214,16 +1220,16 @@ int (*ioapic_renumber_irq)(int ioapic, int irq); static int pin_2_irq(int idx, int apic, int pin) { int irq, i; - int bus = mp_irqs[idx].mp_srcbus; + int bus = mp_irqs[idx].srcbus; /* * Debugging check, we are in big trouble if this message pops up! */ - if (mp_irqs[idx].mp_dstirq != pin) + if (mp_irqs[idx].dstirq != pin) printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n"); if (test_bit(bus, mp_bus_not_pci)) { - irq = mp_irqs[idx].mp_srcbusirq; + irq = mp_irqs[idx].srcbusirq; } else { /* * PCI IRQs are mapped in order @@ -1315,7 +1321,7 @@ __assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask) int new_cpu; int vector, offset; - vector_allocation_domain(cpu, tmp_mask); + apic->vector_allocation_domain(cpu, tmp_mask); vector = current_vector; offset = current_offset; @@ -1421,9 +1427,8 @@ void __setup_vector_irq(int cpu) } static struct irq_chip ioapic_chip; -#ifdef CONFIG_INTR_REMAP static struct irq_chip ir_ioapic_chip; -#endif +static struct irq_chip msi_ir_chip; #define IOAPIC_AUTO -1 #define IOAPIC_EDGE 0 @@ -1462,7 +1467,6 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t else desc->status &= ~IRQ_LEVEL; -#ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { desc->status |= IRQ_MOVE_PCNTXT; if (trigger) @@ -1474,7 +1478,7 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t handle_edge_irq, "edge"); return; } -#endif + if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || trigger == IOAPIC_LEVEL) set_irq_chip_and_handler_name(irq, &ioapic_chip, @@ -1485,37 +1489,43 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t handle_edge_irq, "edge"); } -static int setup_ioapic_entry(int apic, int irq, - struct IO_APIC_route_entry *entry, - unsigned int destination, int trigger, - int polarity, int vector) +int setup_ioapic_entry(int apic_id, int irq, + struct IO_APIC_route_entry *entry, + unsigned int destination, int trigger, + int polarity, int vector, int pin) { /* * add it to the IO-APIC irq-routing table: */ memset(entry,0,sizeof(*entry)); -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) { - struct intel_iommu *iommu = map_ioapic_to_ir(apic); + struct intel_iommu *iommu = map_ioapic_to_ir(apic_id); struct irte irte; struct IR_IO_APIC_route_entry *ir_entry = (struct IR_IO_APIC_route_entry *) entry; int index; if (!iommu) - panic("No mapping iommu for ioapic %d\n", apic); + panic("No mapping iommu for ioapic %d\n", apic_id); index = alloc_irte(iommu, irq, 1); if (index < 0) - panic("Failed to allocate IRTE for ioapic %d\n", apic); + panic("Failed to allocate IRTE for ioapic %d\n", apic_id); memset(&irte, 0, sizeof(irte)); irte.present = 1; - irte.dst_mode = INT_DEST_MODE; - irte.trigger_mode = trigger; - irte.dlvry_mode = INT_DELIVERY_MODE; + irte.dst_mode = apic->irq_dest_mode; + /* + * Trigger mode in the IRTE will always be edge, and the + * actual level or edge trigger will be setup in the IO-APIC + * RTE. This will help simplify level triggered irq migration. + * For more details, see the comments above explainig IO-APIC + * irq migration in the presence of interrupt-remapping. + */ + irte.trigger_mode = 0; + irte.dlvry_mode = apic->irq_delivery_mode; irte.vector = vector; irte.dest_id = IRTE_DEST(destination); @@ -1525,18 +1535,21 @@ static int setup_ioapic_entry(int apic, int irq, ir_entry->zero = 0; ir_entry->format = 1; ir_entry->index = (index & 0x7fff); - } else -#endif - { - entry->delivery_mode = INT_DELIVERY_MODE; - entry->dest_mode = INT_DEST_MODE; + /* + * IO-APIC RTE will be configured with virtual vector. + * irq handler will do the explicit EOI to the io-apic. + */ + ir_entry->vector = pin; + } else { + entry->delivery_mode = apic->irq_delivery_mode; + entry->dest_mode = apic->irq_dest_mode; entry->dest = destination; + entry->vector = vector; } entry->mask = 0; /* enable IRQ */ entry->trigger = trigger; entry->polarity = polarity; - entry->vector = vector; /* Mask level triggered irqs. * Use IRQ_DELAYED_DISABLE for edge triggered irqs. @@ -1546,7 +1559,7 @@ static int setup_ioapic_entry(int apic, int irq, return 0; } -static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, struct irq_desc *desc, +static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq_desc *desc, int trigger, int polarity) { struct irq_cfg *cfg; @@ -1558,22 +1571,22 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, struct irq_de cfg = desc->chip_data; - if (assign_irq_vector(irq, cfg, TARGET_CPUS)) + if (assign_irq_vector(irq, cfg, apic->target_cpus())) return; - dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS); + dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus()); apic_printk(APIC_VERBOSE,KERN_DEBUG "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> " "IRQ %d Mode:%i Active:%i)\n", - apic, mp_ioapics[apic].mp_apicid, pin, cfg->vector, + apic_id, mp_ioapics[apic_id].apicid, pin, cfg->vector, irq, trigger, polarity); - if (setup_ioapic_entry(mp_ioapics[apic].mp_apicid, irq, &entry, - dest, trigger, polarity, cfg->vector)) { + if (setup_ioapic_entry(mp_ioapics[apic_id].apicid, irq, &entry, + dest, trigger, polarity, cfg->vector, pin)) { printk("Failed to setup ioapic entry for ioapic %d, pin %d\n", - mp_ioapics[apic].mp_apicid, pin); + mp_ioapics[apic_id].apicid, pin); __clear_irq_vector(irq, cfg); return; } @@ -1582,12 +1595,12 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, struct irq_de if (irq < NR_IRQS_LEGACY) disable_8259A_irq(irq); - ioapic_write_entry(apic, pin, entry); + ioapic_write_entry(apic_id, pin, entry); } static void __init setup_IO_APIC_irqs(void) { - int apic, pin, idx, irq; + int apic_id, pin, idx, irq; int notcon = 0; struct irq_desc *desc; struct irq_cfg *cfg; @@ -1595,21 +1608,19 @@ static void __init setup_IO_APIC_irqs(void) apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); - for (apic = 0; apic < nr_ioapics; apic++) { - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + 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, pin, mp_INT); + idx = find_irq_entry(apic_id, pin, mp_INT); if (idx == -1) { if (!notcon) { notcon = 1; apic_printk(APIC_VERBOSE, KERN_DEBUG " %d-%d", - mp_ioapics[apic].mp_apicid, - pin); + mp_ioapics[apic_id].apicid, pin); } else apic_printk(APIC_VERBOSE, " %d-%d", - mp_ioapics[apic].mp_apicid, - pin); + mp_ioapics[apic_id].apicid, pin); continue; } if (notcon) { @@ -1618,20 +1629,25 @@ static void __init setup_IO_APIC_irqs(void) notcon = 0; } - irq = pin_2_irq(idx, apic, pin); -#ifdef CONFIG_X86_32 - if (multi_timer_check(apic, irq)) + irq = pin_2_irq(idx, apic_id, pin); + + /* + * Skip the timer IRQ if there's a quirk handler + * installed and if it returns 1: + */ + if (apic->multi_timer_check && + apic->multi_timer_check(apic_id, irq)) continue; -#endif + desc = irq_to_desc_alloc_cpu(irq, cpu); if (!desc) { printk(KERN_INFO "can not get irq_desc for %d\n", irq); continue; } cfg = desc->chip_data; - add_pin_to_irq_cpu(cfg, cpu, apic, pin); + add_pin_to_irq_cpu(cfg, cpu, apic_id, pin); - setup_IO_APIC_irq(apic, pin, irq, desc, + setup_IO_APIC_irq(apic_id, pin, irq, desc, irq_trigger(idx), irq_polarity(idx)); } } @@ -1644,15 +1660,13 @@ static void __init setup_IO_APIC_irqs(void) /* * Set up the timer pin, possibly with the 8259A-master behind. */ -static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, +static void __init setup_timer_IRQ0_pin(unsigned int apic_id, unsigned int pin, int vector) { struct IO_APIC_route_entry entry; -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) return; -#endif memset(&entry, 0, sizeof(entry)); @@ -1660,10 +1674,10 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, * We use logical delivery to get the timer IRQ * to the first CPU. */ - entry.dest_mode = INT_DEST_MODE; - entry.mask = 1; /* mask IRQ now */ - entry.dest = cpu_mask_to_apicid(TARGET_CPUS); - entry.delivery_mode = INT_DELIVERY_MODE; + entry.dest_mode = apic->irq_dest_mode; + entry.mask = 0; /* don't mask IRQ for edge */ + entry.dest = apic->cpu_mask_to_apicid(apic->target_cpus()); + entry.delivery_mode = apic->irq_delivery_mode; entry.polarity = 0; entry.trigger = 0; entry.vector = vector; @@ -1677,7 +1691,7 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, /* * Add it to the IO-APIC irq-routing table: */ - ioapic_write_entry(apic, pin, entry); + ioapic_write_entry(apic_id, pin, entry); } @@ -1699,7 +1713,7 @@ __apicdebuginit(void) print_IO_APIC(void) printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); for (i = 0; i < nr_ioapics; i++) printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n", - mp_ioapics[i].mp_apicid, nr_ioapic_registers[i]); + mp_ioapics[i].apicid, nr_ioapic_registers[i]); /* * We are a bit conservative about what we expect. We have to @@ -1719,7 +1733,7 @@ __apicdebuginit(void) print_IO_APIC(void) spin_unlock_irqrestore(&ioapic_lock, flags); printk("\n"); - printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid); + printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].apicid); printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type); @@ -1980,13 +1994,6 @@ void __init enable_IO_APIC(void) int apic; unsigned long flags; -#ifdef CONFIG_X86_32 - int i; - if (!pirqs_enabled) - for (i = 0; i < MAX_PIRQS; i++) - pirq_entries[i] = -1; -#endif - /* * The number of IO-APIC IRQ registers (== #pins): */ @@ -2054,8 +2061,13 @@ void disable_IO_APIC(void) * If the i8259 is routed through an IOAPIC * Put that IOAPIC in virtual wire mode * so legacy interrupts can be delivered. + * + * With interrupt-remapping, for now we will use virtual wire A mode, + * as virtual wire B is little complex (need to configure both + * IOAPIC RTE aswell as interrupt-remapping table entry). + * As this gets called during crash dump, keep this simple for now. */ - if (ioapic_i8259.pin != -1) { + if (ioapic_i8259.pin != -1 && !intr_remapping_enabled) { struct IO_APIC_route_entry entry; memset(&entry, 0, sizeof(entry)); @@ -2075,7 +2087,10 @@ void disable_IO_APIC(void) ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry); } - disconnect_bsp_APIC(ioapic_i8259.pin != -1); + /* + * Use virtual wire A mode when interrupt remapping is enabled. + */ + disconnect_bsp_APIC(!intr_remapping_enabled && ioapic_i8259.pin != -1); } #ifdef CONFIG_X86_32 @@ -2090,7 +2105,7 @@ static void __init setup_ioapic_ids_from_mpc(void) { union IO_APIC_reg_00 reg_00; physid_mask_t phys_id_present_map; - int apic; + int apic_id; int i; unsigned char old_id; unsigned long flags; @@ -2109,26 +2124,26 @@ static 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 = ioapic_phys_id_map(phys_cpu_present_map); + phys_id_present_map = apic->ioapic_phys_id_map(phys_cpu_present_map); /* * Set the IOAPIC ID to the value stored in the MPC table. */ - for (apic = 0; apic < nr_ioapics; apic++) { + for (apic_id = 0; apic_id < nr_ioapics; apic_id++) { /* Read the register 0 value */ spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(apic, 0); + reg_00.raw = io_apic_read(apic_id, 0); spin_unlock_irqrestore(&ioapic_lock, flags); - old_id = mp_ioapics[apic].mp_apicid; + old_id = mp_ioapics[apic_id].apicid; - if (mp_ioapics[apic].mp_apicid >= get_physical_broadcast()) { + if (mp_ioapics[apic_id].apicid >= get_physical_broadcast()) { printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n", - apic, mp_ioapics[apic].mp_apicid); + apic_id, mp_ioapics[apic_id].apicid); printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", reg_00.bits.ID); - mp_ioapics[apic].mp_apicid = reg_00.bits.ID; + mp_ioapics[apic_id].apicid = reg_00.bits.ID; } /* @@ -2136,10 +2151,10 @@ static 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 (check_apicid_used(phys_id_present_map, - mp_ioapics[apic].mp_apicid)) { + 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, mp_ioapics[apic].mp_apicid); + apic_id, mp_ioapics[apic_id].apicid); for (i = 0; i < get_physical_broadcast(); i++) if (!physid_isset(i, phys_id_present_map)) break; @@ -2148,13 +2163,13 @@ static void __init setup_ioapic_ids_from_mpc(void) printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", i); physid_set(i, phys_id_present_map); - mp_ioapics[apic].mp_apicid = i; + mp_ioapics[apic_id].apicid = i; } else { physid_mask_t tmp; - tmp = apicid_to_cpu_present(mp_ioapics[apic].mp_apicid); + tmp = apic->apicid_to_cpu_present(mp_ioapics[apic_id].apicid); apic_printk(APIC_VERBOSE, "Setting %d in the " "phys_id_present_map\n", - mp_ioapics[apic].mp_apicid); + mp_ioapics[apic_id].apicid); physids_or(phys_id_present_map, phys_id_present_map, tmp); } @@ -2163,11 +2178,11 @@ static void __init setup_ioapic_ids_from_mpc(void) * We need to adjust the IRQ routing table * if the ID changed. */ - if (old_id != mp_ioapics[apic].mp_apicid) + if (old_id != mp_ioapics[apic_id].apicid) for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mp_dstapic == old_id) - mp_irqs[i].mp_dstapic - = mp_ioapics[apic].mp_apicid; + if (mp_irqs[i].dstapic == old_id) + mp_irqs[i].dstapic + = mp_ioapics[apic_id].apicid; /* * Read the right value from the MPC table and @@ -2175,20 +2190,20 @@ static void __init setup_ioapic_ids_from_mpc(void) */ apic_printk(APIC_VERBOSE, KERN_INFO "...changing IO-APIC physical APIC ID to %d ...", - mp_ioapics[apic].mp_apicid); + mp_ioapics[apic_id].apicid); - reg_00.bits.ID = mp_ioapics[apic].mp_apicid; + reg_00.bits.ID = mp_ioapics[apic_id].apicid; spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(apic, 0, reg_00.raw); + io_apic_write(apic_id, 0, reg_00.raw); spin_unlock_irqrestore(&ioapic_lock, flags); /* * Sanity check */ spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(apic, 0); + reg_00.raw = io_apic_read(apic_id, 0); spin_unlock_irqrestore(&ioapic_lock, flags); - if (reg_00.bits.ID != mp_ioapics[apic].mp_apicid) + if (reg_00.bits.ID != mp_ioapics[apic_id].apicid) printk("could not set ID!\n"); else apic_printk(APIC_VERBOSE, " ok.\n"); @@ -2291,7 +2306,7 @@ static int ioapic_retrigger_irq(unsigned int irq) unsigned long flags; spin_lock_irqsave(&vector_lock, flags); - send_IPI_mask(cpumask_of(cpumask_first(cfg->domain)), cfg->vector); + apic->send_IPI_mask(cpumask_of(cpumask_first(cfg->domain)), cfg->vector); spin_unlock_irqrestore(&vector_lock, flags); return 1; @@ -2299,7 +2314,7 @@ static int ioapic_retrigger_irq(unsigned int irq) #else static int ioapic_retrigger_irq(unsigned int irq) { - send_IPI_self(irq_cfg(irq)->vector); + apic->send_IPI_self(irq_cfg(irq)->vector); return 1; } @@ -2317,37 +2332,24 @@ static int ioapic_retrigger_irq(unsigned int irq) #ifdef CONFIG_SMP #ifdef CONFIG_INTR_REMAP -static void ir_irq_migration(struct work_struct *work); - -static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration); /* * Migrate the IO-APIC irq in the presence of intr-remapping. * - * For edge triggered, irq migration is a simple atomic update(of vector - * and cpu destination) of IRTE and flush the hardware cache. + * For both level and edge triggered, irq migration is a simple atomic + * update(of vector and cpu destination) of IRTE and flush the hardware cache. * - * For level triggered, we need to modify the io-apic RTE aswell with the update - * vector information, along with modifying IRTE with vector and destination. - * So irq migration for level triggered is little bit more complex compared to - * edge triggered migration. But the good news is, we use the same algorithm - * for level triggered migration as we have today, only difference being, - * we now initiate the irq migration from process context instead of the - * interrupt context. - * - * In future, when we do a directed EOI (combined with cpu EOI broadcast - * suppression) to the IO-APIC, level triggered irq migration will also be - * as simple as edge triggered migration and we can do the irq migration - * with a simple atomic update to IO-APIC RTE. + * For level triggered, we eliminate the io-apic RTE modification (with the + * updated vector information), by using a virtual vector (io-apic pin number). + * Real vector that is used for interrupting cpu will be coming from + * the interrupt-remapping table entry. */ static void migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) { struct irq_cfg *cfg; struct irte irte; - int modify_ioapic_rte; unsigned int dest; - unsigned long flags; unsigned int irq; if (!cpumask_intersects(mask, cpu_online_mask)) @@ -2363,14 +2365,7 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) set_extra_move_desc(desc, mask); - dest = cpu_mask_to_apicid_and(cfg->domain, mask); - - modify_ioapic_rte = desc->status & IRQ_LEVEL; - if (modify_ioapic_rte) { - spin_lock_irqsave(&ioapic_lock, flags); - __target_IO_APIC_irq(irq, dest, cfg); - spin_unlock_irqrestore(&ioapic_lock, flags); - } + dest = apic->cpu_mask_to_apicid_and(cfg->domain, mask); irte.vector = cfg->vector; irte.dest_id = IRTE_DEST(dest); @@ -2383,61 +2378,7 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) if (cfg->move_in_progress) send_cleanup_vector(cfg); - cpumask_copy(&desc->affinity, mask); -} - -static int migrate_irq_remapped_level_desc(struct irq_desc *desc) -{ - int ret = -1; - struct irq_cfg *cfg = desc->chip_data; - - mask_IO_APIC_irq_desc(desc); - - if (io_apic_level_ack_pending(cfg)) { - /* - * Interrupt in progress. Migrating irq now will change the - * vector information in the IO-APIC RTE and that will confuse - * the EOI broadcast performed by cpu. - * So, delay the irq migration to the next instance. - */ - schedule_delayed_work(&ir_migration_work, 1); - goto unmask; - } - - /* everthing is clear. we have right of way */ - migrate_ioapic_irq_desc(desc, &desc->pending_mask); - - ret = 0; - desc->status &= ~IRQ_MOVE_PENDING; - cpumask_clear(&desc->pending_mask); - -unmask: - unmask_IO_APIC_irq_desc(desc); - - return ret; -} - -static void ir_irq_migration(struct work_struct *work) -{ - unsigned int irq; - struct irq_desc *desc; - - for_each_irq_desc(irq, desc) { - if (desc->status & IRQ_MOVE_PENDING) { - unsigned long flags; - - spin_lock_irqsave(&desc->lock, flags); - if (!desc->chip->set_affinity || - !(desc->status & IRQ_MOVE_PENDING)) { - desc->status &= ~IRQ_MOVE_PENDING; - spin_unlock_irqrestore(&desc->lock, flags); - continue; - } - - desc->chip->set_affinity(irq, &desc->pending_mask); - spin_unlock_irqrestore(&desc->lock, flags); - } - } + cpumask_copy(desc->affinity, mask); } /* @@ -2446,13 +2387,6 @@ static void ir_irq_migration(struct work_struct *work) static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask) { - if (desc->status & IRQ_LEVEL) { - desc->status |= IRQ_MOVE_PENDING; - cpumask_copy(&desc->pending_mask, mask); - migrate_irq_remapped_level_desc(desc); - return; - } - migrate_ioapic_irq_desc(desc, mask); } static void set_ir_ioapic_affinity_irq(unsigned int irq, @@ -2462,6 +2396,11 @@ static void set_ir_ioapic_affinity_irq(unsigned int irq, set_ir_ioapic_affinity_irq_desc(desc, mask); } +#else +static inline void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, + const struct cpumask *mask) +{ +} #endif asmlinkage void smp_irq_move_cleanup_interrupt(void) @@ -2475,6 +2414,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) me = smp_processor_id(); for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) { unsigned int irq; + unsigned int irr; struct irq_desc *desc; struct irq_cfg *cfg; irq = __get_cpu_var(vector_irq)[vector]; @@ -2494,6 +2434,18 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) goto unlock; + irr = apic_read(APIC_IRR + (vector / 32 * 0x10)); + /* + * Check if the vector that needs to be cleanedup is + * registered at the cpu's IRR. If so, then this is not + * the best time to clean it up. Lets clean it up in the + * next attempt by sending another IRQ_MOVE_CLEANUP_VECTOR + * to myself. + */ + if (irr & (1 << (vector % 32))) { + apic->send_IPI_self(IRQ_MOVE_CLEANUP_VECTOR); + goto unlock; + } __get_cpu_var(vector_irq)[vector] = -1; cfg->move_cleanup_count--; unlock: @@ -2516,7 +2468,7 @@ static void irq_complete_move(struct irq_desc **descp) /* domain has not changed, but affinity did */ me = smp_processor_id(); - if (cpu_isset(me, desc->affinity)) { + if (cpumask_test_cpu(me, desc->affinity)) { *descp = desc = move_irq_desc(desc, me); /* get the new one */ cfg = desc->chip_data; @@ -2543,9 +2495,44 @@ static inline void irq_complete_move(struct irq_desc **descp) {} #endif #ifdef CONFIG_INTR_REMAP +static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) +{ + int apic, pin; + struct irq_pin_list *entry; + + entry = cfg->irq_2_pin; + for (;;) { + + if (!entry) + break; + + apic = entry->apic; + pin = entry->pin; + io_apic_eoi(apic, pin); + entry = entry->next; + } +} + +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 ack_x2apic_level(unsigned int irq) { + struct irq_desc *desc = irq_to_desc(irq); ack_x2APIC_irq(); + eoi_ioapic_irq(desc); } static void ack_x2apic_edge(unsigned int irq) @@ -2867,19 +2854,15 @@ static inline void __init check_timer(void) int cpu = boot_cpu_id; int apic1, pin1, apic2, pin2; unsigned long flags; - unsigned int ver; int no_pin1 = 0; local_irq_save(flags); - ver = apic_read(APIC_LVR); - ver = GET_APIC_VERSION(ver); - /* * get/set the timer IRQ vector: */ disable_8259A_irq(0); - assign_irq_vector(0, cfg, TARGET_CPUS); + assign_irq_vector(0, cfg, apic->target_cpus()); /* * As IRQ0 is to be enabled in the 8259A, the virtual @@ -2893,7 +2876,13 @@ static inline void __init check_timer(void) apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); init_8259A(1); #ifdef CONFIG_X86_32 - timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); + { + unsigned int ver; + + ver = apic_read(APIC_LVR); + ver = GET_APIC_VERSION(ver); + timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); + } #endif pin1 = find_isa_irq_pin(0, mp_INT); @@ -2913,10 +2902,8 @@ static inline void __init check_timer(void) * 8259A. */ if (pin1 == -1) { -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) panic("BIOS bug: timer not connected to IO-APIC"); -#endif pin1 = pin2; apic1 = apic2; no_pin1 = 1; @@ -2932,8 +2919,17 @@ static inline void __init check_timer(void) if (no_pin1) { add_pin_to_irq_cpu(cfg, cpu, apic1, pin1); setup_timer_IRQ0_pin(apic1, pin1, cfg->vector); + } else { + /* for edge trigger, setup_IO_APIC_irq already + * leave it unmasked. + * so only need to unmask if it is level-trigger + * do we really have level trigger timer? + */ + int idx; + idx = find_irq_entry(apic1, pin1, mp_INT); + if (idx != -1 && irq_trigger(idx)) + unmask_IO_APIC_irq_desc(desc); } - unmask_IO_APIC_irq_desc(desc); if (timer_irq_works()) { if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); @@ -2943,10 +2939,9 @@ static inline void __init check_timer(void) clear_IO_APIC_pin(0, pin1); goto out; } -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) panic("timer doesn't work through Interrupt-remapped IO-APIC"); -#endif + local_irq_disable(); clear_IO_APIC_pin(apic1, pin1); if (!no_pin1) apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: " @@ -2961,7 +2956,6 @@ static inline void __init check_timer(void) */ replace_pin_at_irq_cpu(cfg, cpu, apic1, pin1, apic2, pin2); setup_timer_IRQ0_pin(apic2, pin2, cfg->vector); - unmask_IO_APIC_irq_desc(desc); enable_8259A_irq(0); if (timer_irq_works()) { apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); @@ -2976,6 +2970,7 @@ static inline void __init check_timer(void) /* * Cleanup, just in case ... */ + local_irq_disable(); disable_8259A_irq(0); clear_IO_APIC_pin(apic2, pin2); apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); @@ -3001,6 +2996,7 @@ static inline void __init check_timer(void) apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); goto out; } + local_irq_disable(); disable_8259A_irq(0); apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | cfg->vector); apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n"); @@ -3018,6 +3014,7 @@ static inline void __init check_timer(void) apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); goto out; } + local_irq_disable(); apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n"); panic("IO-APIC + timer doesn't work! Boot with apic=debug and send a " "report. Then try booting with the 'noapic' option.\n"); @@ -3047,13 +3044,9 @@ out: void __init setup_IO_APIC(void) { -#ifdef CONFIG_X86_32 - enable_IO_APIC(); -#else /* * calling enable_IO_APIC() is moved to setup_local_APIC for BP */ -#endif io_apic_irqs = ~PIC_IRQS; @@ -3118,8 +3111,8 @@ static int ioapic_resume(struct sys_device *dev) spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(dev->id, 0); - if (reg_00.bits.ID != mp_ioapics[dev->id].mp_apicid) { - reg_00.bits.ID = mp_ioapics[dev->id].mp_apicid; + 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); @@ -3169,6 +3162,7 @@ static int __init ioapic_init_sysfs(void) device_initcall(ioapic_init_sysfs); +static int nr_irqs_gsi = NR_IRQS_LEGACY; /* * Dynamic irq allocate and deallocation */ @@ -3183,11 +3177,11 @@ unsigned int create_irq_nr(unsigned int irq_want) struct irq_desc *desc_new = NULL; irq = 0; - spin_lock_irqsave(&vector_lock, flags); - for (new = irq_want; new < NR_IRQS; new++) { - if (platform_legacy_irq(new)) - continue; + if (irq_want < nr_irqs_gsi) + irq_want = nr_irqs_gsi; + spin_lock_irqsave(&vector_lock, flags); + for (new = irq_want; new < nr_irqs; new++) { desc_new = irq_to_desc_alloc_cpu(new, cpu); if (!desc_new) { printk(KERN_INFO "can not get irq_desc for %d\n", new); @@ -3197,7 +3191,7 @@ unsigned int create_irq_nr(unsigned int irq_want) if (cfg_new->vector != 0) continue; - if (__assign_irq_vector(new, cfg_new, TARGET_CPUS) == 0) + if (__assign_irq_vector(new, cfg_new, apic->target_cpus()) == 0) irq = new; break; } @@ -3212,7 +3206,6 @@ unsigned int create_irq_nr(unsigned int irq_want) return irq; } -static int nr_irqs_gsi = NR_IRQS_LEGACY; int create_irq(void) { unsigned int irq_want; @@ -3241,9 +3234,7 @@ void destroy_irq(unsigned int irq) if (desc) desc->chip_data = cfg; -#ifdef CONFIG_INTR_REMAP free_irte(irq); -#endif spin_lock_irqsave(&vector_lock, flags); __clear_irq_vector(irq, cfg); spin_unlock_irqrestore(&vector_lock, flags); @@ -3259,14 +3250,16 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms int err; unsigned dest; + if (disable_apic) + return -ENXIO; + cfg = irq_cfg(irq); - err = assign_irq_vector(irq, cfg, TARGET_CPUS); + err = assign_irq_vector(irq, cfg, apic->target_cpus()); if (err) return err; - dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS); + dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus()); -#ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { struct irte irte; int ir_index; @@ -3278,9 +3271,9 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms memset (&irte, 0, sizeof(irte)); irte.present = 1; - irte.dst_mode = INT_DEST_MODE; + irte.dst_mode = apic->irq_dest_mode; irte.trigger_mode = 0; /* edge */ - irte.dlvry_mode = INT_DELIVERY_MODE; + irte.dlvry_mode = apic->irq_delivery_mode; irte.vector = cfg->vector; irte.dest_id = IRTE_DEST(dest); @@ -3292,16 +3285,19 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms MSI_ADDR_IR_SHV | MSI_ADDR_IR_INDEX1(ir_index) | MSI_ADDR_IR_INDEX2(ir_index); - } else -#endif - { - msg->address_hi = MSI_ADDR_BASE_HI; + } else { + if (x2apic_enabled()) + msg->address_hi = MSI_ADDR_BASE_HI | + MSI_ADDR_EXT_DEST_ID(dest); + else + msg->address_hi = MSI_ADDR_BASE_HI; + msg->address_lo = MSI_ADDR_BASE_LO | - ((INT_DEST_MODE == 0) ? + ((apic->irq_dest_mode == 0) ? MSI_ADDR_DEST_MODE_PHYSICAL: MSI_ADDR_DEST_MODE_LOGICAL) | - ((INT_DELIVERY_MODE != dest_LowestPrio) ? + ((apic->irq_delivery_mode != dest_LowestPrio) ? MSI_ADDR_REDIRECTION_CPU: MSI_ADDR_REDIRECTION_LOWPRI) | MSI_ADDR_DEST_ID(dest); @@ -3309,7 +3305,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms msg->data = MSI_DATA_TRIGGER_EDGE | MSI_DATA_LEVEL_ASSERT | - ((INT_DELIVERY_MODE != dest_LowestPrio) ? + ((apic->irq_delivery_mode != dest_LowestPrio) ? MSI_DATA_DELIVERY_FIXED: MSI_DATA_DELIVERY_LOWPRI) | MSI_DATA_VECTOR(cfg->vector); @@ -3406,6 +3402,7 @@ static struct irq_chip msi_ir_chip = { #endif .retrigger = ioapic_retrigger_irq, }; +#endif /* * Map the PCI dev to the corresponding remapping hardware unit @@ -3433,7 +3430,6 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec) } return index; } -#endif static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) { @@ -3447,7 +3443,6 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) set_irq_msi(irq, msidesc); write_msi_msg(irq, &msg); -#ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { struct irq_desc *desc = irq_to_desc(irq); /* @@ -3456,7 +3451,6 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) desc->status |= IRQ_MOVE_PCNTXT; set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge"); } else -#endif set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge"); dev_printk(KERN_DEBUG, &dev->dev, "irq %d for MSI/MSI-X\n", irq); @@ -3464,60 +3458,22 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) return 0; } -int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc) -{ - unsigned int irq; - int ret; - unsigned int irq_want; - - irq_want = nr_irqs_gsi; - irq = create_irq_nr(irq_want); - if (irq == 0) - return -1; - -#ifdef CONFIG_INTR_REMAP - if (!intr_remapping_enabled) - goto no_ir; - - ret = msi_alloc_irte(dev, irq, 1); - if (ret < 0) - goto error; -no_ir: -#endif - ret = setup_msi_irq(dev, msidesc, irq); - if (ret < 0) { - destroy_irq(irq); - return ret; - } - return 0; - -#ifdef CONFIG_INTR_REMAP -error: - destroy_irq(irq); - return ret; -#endif -} - int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) { unsigned int irq; int ret, sub_handle; struct msi_desc *msidesc; unsigned int irq_want; - -#ifdef CONFIG_INTR_REMAP struct intel_iommu *iommu = 0; int index = 0; -#endif irq_want = nr_irqs_gsi; sub_handle = 0; list_for_each_entry(msidesc, &dev->msi_list, list) { irq = create_irq_nr(irq_want); - irq_want++; if (irq == 0) return -1; -#ifdef CONFIG_INTR_REMAP + irq_want = irq + 1; if (!intr_remapping_enabled) goto no_ir; @@ -3545,7 +3501,6 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) set_irte_irq(irq, iommu, index, sub_handle); } no_ir: -#endif ret = setup_msi_irq(dev, msidesc, irq); if (ret < 0) goto error; @@ -3563,7 +3518,7 @@ void arch_teardown_msi_irq(unsigned int irq) destroy_irq(irq); } -#ifdef CONFIG_DMAR +#if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP) #ifdef CONFIG_SMP static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) { @@ -3727,13 +3682,17 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) struct irq_cfg *cfg; int err; + if (disable_apic) + return -ENXIO; + cfg = irq_cfg(irq); - err = assign_irq_vector(irq, cfg, TARGET_CPUS); + err = assign_irq_vector(irq, cfg, apic->target_cpus()); if (!err) { struct ht_irq_msg msg; unsigned dest; - dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS); + dest = apic->cpu_mask_to_apicid_and(cfg->domain, + apic->target_cpus()); msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); @@ -3741,11 +3700,11 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) HT_IRQ_LOW_BASE | HT_IRQ_LOW_DEST_ID(dest) | HT_IRQ_LOW_VECTOR(cfg->vector) | - ((INT_DEST_MODE == 0) ? + ((apic->irq_dest_mode == 0) ? HT_IRQ_LOW_DM_PHYSICAL : HT_IRQ_LOW_DM_LOGICAL) | HT_IRQ_LOW_RQEOI_EDGE | - ((INT_DELIVERY_MODE != dest_LowestPrio) ? + ((apic->irq_delivery_mode != dest_LowestPrio) ? HT_IRQ_LOW_MT_FIXED : HT_IRQ_LOW_MT_ARBITRATED) | HT_IRQ_LOW_IRQ_MASKED; @@ -3761,7 +3720,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) } #endif /* CONFIG_HT_IRQ */ -#ifdef CONFIG_X86_64 +#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. @@ -3793,12 +3752,12 @@ int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); entry->vector = cfg->vector; - entry->delivery_mode = INT_DELIVERY_MODE; - entry->dest_mode = INT_DEST_MODE; + entry->delivery_mode = apic->irq_delivery_mode; + entry->dest_mode = apic->irq_dest_mode; entry->polarity = 0; entry->trigger = 0; entry->mask = 0; - entry->dest = cpu_mask_to_apicid(eligible_cpu); + 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); @@ -3861,6 +3820,28 @@ void __init probe_nr_irqs_gsi(void) printk(KERN_DEBUG "nr_irqs_gsi: %d\n", nr_irqs_gsi); } +#ifdef CONFIG_SPARSE_IRQ +int __init arch_probe_nr_irqs(void) +{ + int nr; + + if (nr_irqs > (NR_VECTORS * nr_cpu_ids)) + nr_irqs = NR_VECTORS * nr_cpu_ids; + + nr = nr_irqs_gsi + 8 * nr_cpu_ids; +#if defined(CONFIG_PCI_MSI) || defined(CONFIG_HT_IRQ) + /* + * for MSI and HT dyn irq + */ + nr += nr_irqs_gsi * 16; +#endif + if (nr < nr_irqs) + nr_irqs = nr; + + return 0; +} +#endif + /* -------------------------------------------------------------------------- ACPI-based IOAPIC Configuration -------------------------------------------------------------------------- */ @@ -3886,7 +3867,7 @@ int __init io_apic_get_unique_id(int ioapic, int apic_id) */ if (physids_empty(apic_id_map)) - apic_id_map = ioapic_phys_id_map(phys_cpu_present_map); + apic_id_map = apic->ioapic_phys_id_map(phys_cpu_present_map); spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(ioapic, 0); @@ -3902,10 +3883,10 @@ 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 (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 (!check_apicid_used(apic_id_map, i)) + if (!apic->check_apicid_used(apic_id_map, i)) break; } @@ -3918,7 +3899,7 @@ int __init io_apic_get_unique_id(int ioapic, int apic_id) apic_id = i; } - tmp = apicid_to_cpu_present(apic_id); + tmp = apic->apicid_to_cpu_present(apic_id); physids_or(apic_id_map, apic_id_map, tmp); if (reg_00.bits.ID != apic_id) { @@ -3995,8 +3976,8 @@ int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) return -1; for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mp_irqtype == mp_INT && - mp_irqs[i].mp_srcbusirq == bus_irq) + if (mp_irqs[i].irqtype == mp_INT && + mp_irqs[i].srcbusirq == bus_irq) break; if (i >= mp_irq_entries) return -1; @@ -4011,7 +3992,7 @@ int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) /* * This function currently is only a helper for the i386 smp boot process where * we need to reprogram the ioredtbls to cater for the cpus which have come online - * so mask in all cases should simply be TARGET_CPUS + * so mask in all cases should simply be apic->target_cpus() */ #ifdef CONFIG_SMP void __init setup_ioapic_dest(void) @@ -4050,15 +4031,13 @@ void __init setup_ioapic_dest(void) */ if (desc->status & (IRQ_NO_BALANCING | IRQ_AFFINITY_SET)) - mask = &desc->affinity; + mask = desc->affinity; else - mask = TARGET_CPUS; + mask = apic->target_cpus(); -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) set_ir_ioapic_affinity_irq_desc(desc, mask); else -#endif set_ioapic_affinity_irq_desc(desc, mask); } @@ -4111,7 +4090,7 @@ void __init ioapic_init_mappings(void) ioapic_res = ioapic_setup_resources(); for (i = 0; i < nr_ioapics; i++) { if (smp_found_config) { - ioapic_phys = mp_ioapics[i].mp_apicaddr; + ioapic_phys = mp_ioapics[i].apicaddr; #ifdef CONFIG_X86_32 if (!ioapic_phys) { printk(KERN_ERR diff --git a/arch/x86/kernel/apic/ipi.c b/arch/x86/kernel/apic/ipi.c new file mode 100644 index 00000000000..dbf5445727a --- /dev/null +++ b/arch/x86/kernel/apic/ipi.c @@ -0,0 +1,164 @@ +#include <linux/cpumask.h> +#include <linux/interrupt.h> +#include <linux/init.h> + +#include <linux/mm.h> +#include <linux/delay.h> +#include <linux/spinlock.h> +#include <linux/kernel_stat.h> +#include <linux/mc146818rtc.h> +#include <linux/cache.h> +#include <linux/cpu.h> +#include <linux/module.h> + +#include <asm/smp.h> +#include <asm/mtrr.h> +#include <asm/tlbflush.h> +#include <asm/mmu_context.h> +#include <asm/apic.h> +#include <asm/proto.h> +#include <asm/ipi.h> + +void default_send_IPI_mask_sequence_phys(const struct cpumask *mask, int vector) +{ + unsigned long query_cpu; + unsigned long flags; + + /* + * Hack. The clustered APIC addressing mode doesn't allow us to send + * to an arbitrary mask, so I do a unicast to each CPU instead. + * - mbligh + */ + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + __default_send_IPI_dest_field(per_cpu(x86_cpu_to_apicid, + query_cpu), vector, APIC_DEST_PHYSICAL); + } + local_irq_restore(flags); +} + +void default_send_IPI_mask_allbutself_phys(const struct cpumask *mask, + int vector) +{ + unsigned int this_cpu = smp_processor_id(); + unsigned int query_cpu; + unsigned long flags; + + /* See Hack comment above */ + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + if (query_cpu == this_cpu) + continue; + __default_send_IPI_dest_field(per_cpu(x86_cpu_to_apicid, + query_cpu), vector, APIC_DEST_PHYSICAL); + } + local_irq_restore(flags); +} + +void default_send_IPI_mask_sequence_logical(const struct cpumask *mask, + int vector) +{ + unsigned long flags; + unsigned int query_cpu; + + /* + * Hack. The clustered APIC addressing mode doesn't allow us to send + * to an arbitrary mask, so I do a unicasts to each CPU instead. This + * should be modified to do 1 message per cluster ID - mbligh + */ + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) + __default_send_IPI_dest_field( + apic->cpu_to_logical_apicid(query_cpu), vector, + apic->dest_logical); + local_irq_restore(flags); +} + +void default_send_IPI_mask_allbutself_logical(const struct cpumask *mask, + int vector) +{ + unsigned long flags; + unsigned int query_cpu; + unsigned int this_cpu = smp_processor_id(); + + /* See Hack comment above */ + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + if (query_cpu == this_cpu) + continue; + __default_send_IPI_dest_field( + apic->cpu_to_logical_apicid(query_cpu), vector, + apic->dest_logical); + } + local_irq_restore(flags); +} + +#ifdef CONFIG_X86_32 + +/* + * This is only used on smaller machines. + */ +void default_send_IPI_mask_logical(const struct cpumask *cpumask, int vector) +{ + unsigned long mask = cpumask_bits(cpumask)[0]; + unsigned long flags; + + local_irq_save(flags); + WARN_ON(mask & ~cpumask_bits(cpu_online_mask)[0]); + __default_send_IPI_dest_field(mask, vector, apic->dest_logical); + local_irq_restore(flags); +} + +void default_send_IPI_allbutself(int vector) +{ + /* + * if there are no other CPUs in the system then we get an APIC send + * error if we try to broadcast, thus avoid sending IPIs in this case. + */ + if (!(num_online_cpus() > 1)) + return; + + __default_local_send_IPI_allbutself(vector); +} + +void default_send_IPI_all(int vector) +{ + __default_local_send_IPI_all(vector); +} + +void default_send_IPI_self(int vector) +{ + __default_send_IPI_shortcut(APIC_DEST_SELF, vector, apic->dest_logical); +} + +/* must come after the send_IPI functions above for inlining */ +static int convert_apicid_to_cpu(int apic_id) +{ + int i; + + for_each_possible_cpu(i) { + if (per_cpu(x86_cpu_to_apicid, i) == apic_id) + return i; + } + return -1; +} + +int safe_smp_processor_id(void) +{ + int apicid, cpuid; + + if (!boot_cpu_has(X86_FEATURE_APIC)) + return 0; + + apicid = hard_smp_processor_id(); + if (apicid == BAD_APICID) + return 0; + + cpuid = convert_apicid_to_cpu(apicid); + + return cpuid >= 0 ? cpuid : 0; +} +#endif diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/apic/nmi.c index 7228979f1e7..bdfad80c3cf 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/apic/nmi.c @@ -34,7 +34,7 @@ #include <asm/mce.h> -#include <mach_traps.h> +#include <asm/mach_traps.h> int unknown_nmi_panic; int nmi_watchdog_enabled; @@ -61,11 +61,7 @@ static int endflag __initdata; static inline unsigned int get_nmi_count(int cpu) { -#ifdef CONFIG_X86_64 - return cpu_pda(cpu)->__nmi_count; -#else - return nmi_count(cpu); -#endif + return per_cpu(irq_stat, cpu).__nmi_count; } static inline int mce_in_progress(void) @@ -82,12 +78,8 @@ static inline int mce_in_progress(void) */ static inline unsigned int get_timer_irqs(int cpu) { -#ifdef CONFIG_X86_64 - return read_pda(apic_timer_irqs) + read_pda(irq0_irqs); -#else return per_cpu(irq_stat, cpu).apic_timer_irqs + per_cpu(irq_stat, cpu).irq0_irqs; -#endif } #ifdef CONFIG_SMP diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c new file mode 100644 index 00000000000..ba2fc646553 --- /dev/null +++ b/arch/x86/kernel/apic/numaq_32.c @@ -0,0 +1,557 @@ +/* + * Written by: Patricia Gaughen, IBM Corporation + * + * Copyright (C) 2002, IBM Corp. + * Copyright (C) 2009, Red Hat, Inc., Ingo Molnar + * + * All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + * + * Send feedback to <gone@us.ibm.com> + */ +#include <linux/nodemask.h> +#include <linux/topology.h> +#include <linux/bootmem.h> +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/kernel.h> +#include <linux/mmzone.h> +#include <linux/module.h> +#include <linux/string.h> +#include <linux/init.h> +#include <linux/numa.h> +#include <linux/smp.h> +#include <linux/io.h> +#include <linux/mm.h> + +#include <asm/processor.h> +#include <asm/fixmap.h> +#include <asm/mpspec.h> +#include <asm/numaq.h> +#include <asm/setup.h> +#include <asm/apic.h> +#include <asm/e820.h> +#include <asm/ipi.h> + +#define MB_TO_PAGES(addr) ((addr) << (20 - PAGE_SHIFT)) + +int found_numaq; + +/* + * Have to match translation table entries to main table entries by counter + * hence the mpc_record variable .... can't see a less disgusting way of + * doing this .... + */ +struct mpc_trans { + unsigned char mpc_type; + unsigned char trans_len; + unsigned char trans_type; + unsigned char trans_quad; + unsigned char trans_global; + unsigned char trans_local; + unsigned short trans_reserved; +}; + +/* x86_quirks member */ +static int mpc_record; + +static struct mpc_trans *translation_table[MAX_MPC_ENTRY]; + +int mp_bus_id_to_node[MAX_MP_BUSSES]; +int mp_bus_id_to_local[MAX_MP_BUSSES]; +int quad_local_to_mp_bus_id[NR_CPUS/4][4]; + + +static inline void numaq_register_node(int node, struct sys_cfg_data *scd) +{ + struct eachquadmem *eq = scd->eq + node; + + node_set_online(node); + + /* Convert to pages */ + node_start_pfn[node] = + MB_TO_PAGES(eq->hi_shrd_mem_start - eq->priv_mem_size); + + node_end_pfn[node] = + MB_TO_PAGES(eq->hi_shrd_mem_start + eq->hi_shrd_mem_size); + + e820_register_active_regions(node, node_start_pfn[node], + node_end_pfn[node]); + + memory_present(node, node_start_pfn[node], node_end_pfn[node]); + + node_remap_size[node] = node_memmap_size_bytes(node, + node_start_pfn[node], + node_end_pfn[node]); +} + +/* + * Function: smp_dump_qct() + * + * Description: gets memory layout from the quad config table. This + * function also updates node_online_map with the nodes (quads) present. + */ +static void __init smp_dump_qct(void) +{ + struct sys_cfg_data *scd; + int node; + + scd = (void *)__va(SYS_CFG_DATA_PRIV_ADDR); + + nodes_clear(node_online_map); + for_each_node(node) { + if (scd->quads_present31_0 & (1 << node)) + numaq_register_node(node, scd); + } +} + +void __cpuinit numaq_tsc_disable(void) +{ + if (!found_numaq) + return; + + if (num_online_nodes() > 1) { + printk(KERN_DEBUG "NUMAQ: disabling TSC\n"); + setup_clear_cpu_cap(X86_FEATURE_TSC); + } +} + +static int __init numaq_pre_time_init(void) +{ + numaq_tsc_disable(); + return 0; +} + +static inline int generate_logical_apicid(int quad, int phys_apicid) +{ + return (quad << 4) + (phys_apicid ? phys_apicid << 1 : 1); +} + +/* x86_quirks member */ +static int mpc_apic_id(struct mpc_cpu *m) +{ + int quad = translation_table[mpc_record]->trans_quad; + int logical_apicid = generate_logical_apicid(quad, m->apicid); + + printk(KERN_DEBUG + "Processor #%d %u:%u APIC version %d (quad %d, apic %d)\n", + m->apicid, (m->cpufeature & CPU_FAMILY_MASK) >> 8, + (m->cpufeature & CPU_MODEL_MASK) >> 4, + m->apicver, quad, logical_apicid); + + return logical_apicid; +} + +/* x86_quirks member */ +static void mpc_oem_bus_info(struct mpc_bus *m, char *name) +{ + int quad = translation_table[mpc_record]->trans_quad; + int local = translation_table[mpc_record]->trans_local; + + mp_bus_id_to_node[m->busid] = quad; + mp_bus_id_to_local[m->busid] = local; + + printk(KERN_INFO "Bus #%d is %s (node %d)\n", m->busid, name, quad); +} + +/* x86_quirks member */ +static void mpc_oem_pci_bus(struct mpc_bus *m) +{ + int quad = translation_table[mpc_record]->trans_quad; + int local = translation_table[mpc_record]->trans_local; + + quad_local_to_mp_bus_id[quad][local] = m->busid; +} + +static void __init MP_translation_info(struct mpc_trans *m) +{ + printk(KERN_INFO + "Translation: record %d, type %d, quad %d, global %d, local %d\n", + mpc_record, m->trans_type, m->trans_quad, m->trans_global, + m->trans_local); + + if (mpc_record >= MAX_MPC_ENTRY) + printk(KERN_ERR "MAX_MPC_ENTRY exceeded!\n"); + else + translation_table[mpc_record] = m; /* stash this for later */ + + if (m->trans_quad < MAX_NUMNODES && !node_online(m->trans_quad)) + node_set_online(m->trans_quad); +} + +static int __init mpf_checksum(unsigned char *mp, int len) +{ + int sum = 0; + + while (len--) + sum += *mp++; + + return sum & 0xFF; +} + +/* + * Read/parse the MPC oem tables + */ +static void __init + smp_read_mpc_oem(struct mpc_oemtable *oemtable, unsigned short oemsize) +{ + int count = sizeof(*oemtable); /* the header size */ + unsigned char *oemptr = ((unsigned char *)oemtable) + count; + + mpc_record = 0; + printk(KERN_INFO + "Found an OEM MPC table at %8p - parsing it ... \n", oemtable); + + if (memcmp(oemtable->signature, MPC_OEM_SIGNATURE, 4)) { + printk(KERN_WARNING + "SMP mpc oemtable: bad signature [%c%c%c%c]!\n", + oemtable->signature[0], oemtable->signature[1], + oemtable->signature[2], oemtable->signature[3]); + return; + } + + if (mpf_checksum((unsigned char *)oemtable, oemtable->length)) { + printk(KERN_WARNING "SMP oem mptable: checksum error!\n"); + return; + } + + while (count < oemtable->length) { + switch (*oemptr) { + case MP_TRANSLATION: + { + struct mpc_trans *m = (void *)oemptr; + + MP_translation_info(m); + oemptr += sizeof(*m); + count += sizeof(*m); + ++mpc_record; + break; + } + default: + printk(KERN_WARNING + "Unrecognised OEM table entry type! - %d\n", + (int)*oemptr); + return; + } + } +} + +static int __init numaq_setup_ioapic_ids(void) +{ + /* so can skip it */ + return 1; +} + +static struct x86_quirks numaq_x86_quirks __initdata = { + .arch_pre_time_init = numaq_pre_time_init, + .arch_time_init = NULL, + .arch_pre_intr_init = NULL, + .arch_memory_setup = NULL, + .arch_intr_init = NULL, + .arch_trap_init = NULL, + .mach_get_smp_config = NULL, + .mach_find_smp_config = NULL, + .mpc_record = &mpc_record, + .mpc_apic_id = mpc_apic_id, + .mpc_oem_bus_info = mpc_oem_bus_info, + .mpc_oem_pci_bus = mpc_oem_pci_bus, + .smp_read_mpc_oem = smp_read_mpc_oem, + .setup_ioapic_ids = numaq_setup_ioapic_ids, +}; + +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) + early_get_smp_config(); + + if (found_numaq) + x86_quirks = &numaq_x86_quirks; +} + +int __init get_memcfg_numaq(void) +{ + early_check_numaq(); + if (!found_numaq) + return 0; + smp_dump_qct(); + + return 1; +} + +#define NUMAQ_APIC_DFR_VALUE (APIC_DFR_CLUSTER) + +static inline unsigned int numaq_get_apic_id(unsigned long x) +{ + return (x >> 24) & 0x0F; +} + +static inline void numaq_send_IPI_mask(const struct cpumask *mask, int vector) +{ + default_send_IPI_mask_sequence_logical(mask, vector); +} + +static inline void numaq_send_IPI_allbutself(int vector) +{ + default_send_IPI_mask_allbutself_logical(cpu_online_mask, vector); +} + +static inline void numaq_send_IPI_all(int vector) +{ + numaq_send_IPI_mask(cpu_online_mask, vector); +} + +#define NUMAQ_TRAMPOLINE_PHYS_LOW (0x8) +#define NUMAQ_TRAMPOLINE_PHYS_HIGH (0xa) + +/* + * Because we use NMIs rather than the INIT-STARTUP sequence to + * bootstrap the CPUs, the APIC may be in a weird state. Kick it: + */ +static inline void numaq_smp_callin_clear_local_apic(void) +{ + clear_local_APIC(); +} + +static inline const cpumask_t *numaq_target_cpus(void) +{ + return &CPU_MASK_ALL; +} + +static inline unsigned long +numaq_check_apicid_used(physid_mask_t bitmap, int apicid) +{ + return physid_isset(apicid, bitmap); +} + +static inline unsigned long numaq_check_apicid_present(int bit) +{ + return physid_isset(bit, phys_cpu_present_map); +} + +static inline int numaq_apic_id_registered(void) +{ + return 1; +} + +static inline void numaq_init_apic_ldr(void) +{ + /* Already done in NUMA-Q firmware */ +} + +static inline void numaq_setup_apic_routing(void) +{ + printk(KERN_INFO + "Enabling APIC mode: NUMA-Q. Using %d I/O APICs\n", + nr_ioapics); +} + +/* + * Skip adding the timer int on secondary nodes, which causes + * a small but painful rift in the time-space continuum. + */ +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) +{ + /* We don't have a good way to do this yet - hack */ + return physids_promote(0xFUL); +} + +static inline int numaq_cpu_to_logical_apicid(int cpu) +{ + if (cpu >= nr_cpu_ids) + return BAD_APICID; + return cpu_2_logical_apicid[cpu]; +} + +/* + * Supporting over 60 cpus on NUMA-Q requires a locality-dependent + * cpu to APIC ID relation to properly interact with the intelligent + * mode of the cluster controller. + */ +static inline int numaq_cpu_present_to_apicid(int mps_cpu) +{ + if (mps_cpu < 60) + return ((mps_cpu >> 2) << 4) | (1 << (mps_cpu & 0x3)); + else + return BAD_APICID; +} + +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) +{ + int node = numaq_apicid_to_node(logical_apicid); + int cpu = __ffs(logical_apicid & 0xf); + + return physid_mask_of_physid(cpu + 4*node); +} + +/* Where the IO area was mapped on multiquad, always 0 otherwise */ +void *xquad_portio; + +static inline int numaq_check_phys_apicid_present(int boot_cpu_physical_apicid) +{ + return 1; +} + +/* + * We use physical apicids here, not logical, so just return the default + * physical broadcast to stop people from breaking us + */ +static inline unsigned int numaq_cpu_mask_to_apicid(const cpumask_t *cpumask) +{ + return 0x0F; +} + +static inline unsigned int +numaq_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) +{ + return 0x0F; +} + +/* No NUMA-Q box has a HT CPU, but it can't hurt to use the default code. */ +static inline int numaq_phys_pkg_id(int cpuid_apic, int index_msb) +{ + return cpuid_apic >> index_msb; +} + +static int +numaq_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid) +{ + if (strncmp(oem, "IBM NUMA", 8)) + printk(KERN_ERR "Warning! Not a NUMA-Q system!\n"); + else + found_numaq = 1; + + return found_numaq; +} + +static int probe_numaq(void) +{ + /* already know from get_memcfg_numaq() */ + return found_numaq; +} + +static void numaq_vector_allocation_domain(int cpu, cpumask_t *retmask) +{ + /* Careful. Some cpus do not strictly honor the set of cpus + * specified in the interrupt destination when using lowest + * priority interrupt delivery mode. + * + * In particular there was a hyperthreading cpu observed to + * deliver interrupts to the wrong hyperthread when only one + * hyperthread was specified in the interrupt desitination. + */ + *retmask = (cpumask_t){ { [0] = APIC_ALL_CPUS, } }; +} + +static void numaq_setup_portio_remap(void) +{ + int num_quads = num_online_nodes(); + + if (num_quads <= 1) + return; + + printk(KERN_INFO + "Remapping cross-quad port I/O for %d quads\n", num_quads); + + xquad_portio = ioremap(XQUAD_PORTIO_BASE, num_quads*XQUAD_PORTIO_QUAD); + + printk(KERN_INFO + "xquad_portio vaddr 0x%08lx, len %08lx\n", + (u_long) xquad_portio, (u_long) num_quads*XQUAD_PORTIO_QUAD); +} + +struct apic apic_numaq = { + + .name = "NUMAQ", + .probe = probe_numaq, + .acpi_madt_oem_check = NULL, + .apic_id_registered = numaq_apic_id_registered, + + .irq_delivery_mode = dest_LowestPrio, + /* physical delivery on LOCAL quad: */ + .irq_dest_mode = 0, + + .target_cpus = numaq_target_cpus, + .disable_esr = 1, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = numaq_check_apicid_used, + .check_apicid_present = numaq_check_apicid_present, + + .vector_allocation_domain = numaq_vector_allocation_domain, + .init_apic_ldr = numaq_init_apic_ldr, + + .ioapic_phys_id_map = numaq_ioapic_phys_id_map, + .setup_apic_routing = numaq_setup_apic_routing, + .multi_timer_check = numaq_multi_timer_check, + .apicid_to_node = numaq_apicid_to_node, + .cpu_to_logical_apicid = numaq_cpu_to_logical_apicid, + .cpu_present_to_apicid = numaq_cpu_present_to_apicid, + .apicid_to_cpu_present = numaq_apicid_to_cpu_present, + .setup_portio_remap = numaq_setup_portio_remap, + .check_phys_apicid_present = numaq_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = numaq_phys_pkg_id, + .mps_oem_check = numaq_mps_oem_check, + + .get_apic_id = numaq_get_apic_id, + .set_apic_id = NULL, + .apic_id_mask = 0x0F << 24, + + .cpu_mask_to_apicid = numaq_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = numaq_cpu_mask_to_apicid_and, + + .send_IPI_mask = numaq_send_IPI_mask, + .send_IPI_mask_allbutself = NULL, + .send_IPI_allbutself = numaq_send_IPI_allbutself, + .send_IPI_all = numaq_send_IPI_all, + .send_IPI_self = default_send_IPI_self, + + .wakeup_secondary_cpu = wakeup_secondary_cpu_via_nmi, + .trampoline_phys_low = NUMAQ_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = NUMAQ_TRAMPOLINE_PHYS_HIGH, + + /* We don't do anything here because we use NMI's to boot instead */ + .wait_for_init_deassert = NULL, + + .smp_callin_clear_local_apic = numaq_smp_callin_clear_local_apic, + .inquire_remote_apic = NULL, + + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, +}; diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c new file mode 100644 index 00000000000..141c99a1c26 --- /dev/null +++ b/arch/x86/kernel/apic/probe_32.c @@ -0,0 +1,284 @@ +/* + * Default generic APIC driver. This handles up to 8 CPUs. + * + * Copyright 2003 Andi Kleen, SuSE Labs. + * Subject to the GNU Public License, v.2 + * + * Generic x86 APIC driver probe layer. + */ +#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/threads.h> +#include <linux/cpumask.h> +#include <asm/mpspec.h> +#include <asm/fixmap.h> +#include <asm/apicdef.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/smp.h> +#include <linux/init.h> +#include <asm/ipi.h> + +#include <linux/smp.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <asm/acpi.h> +#include <asm/e820.h> +#include <asm/setup.h> + +#ifdef CONFIG_HOTPLUG_CPU +#define DEFAULT_SEND_IPI (1) +#else +#define DEFAULT_SEND_IPI (0) +#endif + +int no_broadcast = DEFAULT_SEND_IPI; + +static __init int no_ipi_broadcast(char *str) +{ + get_option(&str, &no_broadcast); + pr_info("Using %s mode\n", + no_broadcast ? "No IPI Broadcast" : "IPI Broadcast"); + return 1; +} +__setup("no_ipi_broadcast=", no_ipi_broadcast); + +static int __init print_ipi_mode(void) +{ + pr_info("Using IPI %s mode\n", + no_broadcast ? "No-Shortcut" : "Shortcut"); + return 0; +} +late_initcall(print_ipi_mode); + +void default_setup_apic_routing(void) +{ +#ifdef CONFIG_X86_IO_APIC + printk(KERN_INFO + "Enabling APIC mode: Flat. Using %d I/O APICs\n", + nr_ioapics); +#endif +} + +static void default_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + /* + * Careful. Some cpus do not strictly honor the set of cpus + * specified in the interrupt destination when using lowest + * priority interrupt delivery mode. + * + * In particular there was a hyperthreading cpu observed to + * deliver interrupts to the wrong hyperthread when only one + * hyperthread was specified in the interrupt desitination. + */ + *retmask = (cpumask_t) { { [0] = APIC_ALL_CPUS } }; +} + +/* should be called last. */ +static int probe_default(void) +{ + return 1; +} + +struct apic apic_default = { + + .name = "default", + .probe = probe_default, + .acpi_madt_oem_check = NULL, + .apic_id_registered = default_apic_id_registered, + + .irq_delivery_mode = dest_LowestPrio, + /* logical delivery broadcast to all CPUs: */ + .irq_dest_mode = 1, + + .target_cpus = default_target_cpus, + .disable_esr = 0, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = default_check_apicid_used, + .check_apicid_present = default_check_apicid_present, + + .vector_allocation_domain = default_vector_allocation_domain, + .init_apic_ldr = default_init_apic_ldr, + + .ioapic_phys_id_map = default_ioapic_phys_id_map, + .setup_apic_routing = default_setup_apic_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, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = default_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = default_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 = default_send_IPI_mask_logical, + .send_IPI_mask_allbutself = default_send_IPI_mask_allbutself_logical, + .send_IPI_allbutself = default_send_IPI_allbutself, + .send_IPI_all = default_send_IPI_all, + .send_IPI_self = default_send_IPI_self, + + .trampoline_phys_low = DEFAULT_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = DEFAULT_TRAMPOLINE_PHYS_HIGH, + + .wait_for_init_deassert = default_wait_for_init_deassert, + + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = default_inquire_remote_apic, + + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, +}; + +extern struct apic apic_numaq; +extern struct apic apic_summit; +extern struct apic apic_bigsmp; +extern struct apic apic_es7000; +extern struct apic apic_es7000_cluster; +extern struct apic apic_default; + +struct apic *apic = &apic_default; +EXPORT_SYMBOL_GPL(apic); + +static struct apic *apic_probe[] __initdata = { +#ifdef CONFIG_X86_NUMAQ + &apic_numaq, +#endif +#ifdef CONFIG_X86_SUMMIT + &apic_summit, +#endif +#ifdef CONFIG_X86_BIGSMP + &apic_bigsmp, +#endif +#ifdef CONFIG_X86_ES7000 + &apic_es7000, + &apic_es7000_cluster, +#endif + &apic_default, /* must be last */ + NULL, +}; + +static int cmdline_apic __initdata; +static int __init parse_apic(char *arg) +{ + int i; + + if (!arg) + return -EINVAL; + + for (i = 0; apic_probe[i]; i++) { + if (!strcmp(apic_probe[i]->name, arg)) { + apic = apic_probe[i]; + cmdline_apic = 1; + return 0; + } + } + + /* Parsed again by __setup for debug/verbose */ + return 0; +} +early_param("apic", parse_apic); + +void __init generic_bigsmp_probe(void) +{ +#ifdef CONFIG_X86_BIGSMP + /* + * This routine is used to switch to bigsmp mode when + * - There is no apic= option specified by the user + * - generic_apic_probe() has chosen apic_default as the sub_arch + * - we find more than 8 CPUs in acpi LAPIC listing with xAPIC support + */ + + if (!cmdline_apic && apic == &apic_default) { + if (apic_bigsmp.probe()) { + apic = &apic_bigsmp; + printk(KERN_INFO "Overriding APIC driver with %s\n", + apic->name); + } + } +#endif +} + +void __init generic_apic_probe(void) +{ + if (!cmdline_apic) { + int i; + for (i = 0; apic_probe[i]; i++) { + if (apic_probe[i]->probe()) { + apic = apic_probe[i]; + break; + } + } + /* Not visible without early console */ + if (!apic_probe[i]) + panic("Didn't find an APIC driver"); + } + printk(KERN_INFO "Using APIC driver %s\n", apic->name); +} + +/* These functions can switch the APIC even after the initial ->probe() */ + +int __init +generic_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid) +{ + int i; + + for (i = 0; apic_probe[i]; ++i) { + if (!apic_probe[i]->mps_oem_check) + continue; + if (!apic_probe[i]->mps_oem_check(mpc, oem, productid)) + continue; + + if (!cmdline_apic) { + apic = apic_probe[i]; + printk(KERN_INFO "Switched to APIC driver `%s'.\n", + apic->name); + } + return 1; + } + return 0; +} + +int __init default_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + int i; + + for (i = 0; apic_probe[i]; ++i) { + if (!apic_probe[i]->acpi_madt_oem_check) + continue; + if (!apic_probe[i]->acpi_madt_oem_check(oem_id, oem_table_id)) + continue; + + if (!cmdline_apic) { + apic = apic_probe[i]; + printk(KERN_INFO "Switched to APIC driver `%s'.\n", + apic->name); + } + return 1; + } + return 0; +} diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c new file mode 100644 index 00000000000..1783652bb0e --- /dev/null +++ b/arch/x86/kernel/apic/probe_64.c @@ -0,0 +1,100 @@ +/* + * Copyright 2004 James Cleverdon, IBM. + * Subject to the GNU Public License, v.2 + * + * Generic APIC sub-arch probe layer. + * + * Hacked for x86-64 by James Cleverdon from i386 architecture code by + * Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and + * James Cleverdon. + */ +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/string.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/hardirq.h> +#include <linux/dmar.h> + +#include <asm/smp.h> +#include <asm/apic.h> +#include <asm/ipi.h> +#include <asm/setup.h> + +extern struct apic apic_flat; +extern struct apic apic_physflat; +extern struct apic apic_x2xpic_uv_x; +extern struct apic apic_x2apic_phys; +extern struct apic apic_x2apic_cluster; + +struct apic __read_mostly *apic = &apic_flat; +EXPORT_SYMBOL_GPL(apic); + +static struct apic *apic_probe[] __initdata = { +#ifdef CONFIG_X86_UV + &apic_x2apic_uv_x, +#endif +#ifdef CONFIG_X86_X2APIC + &apic_x2apic_phys, + &apic_x2apic_cluster, +#endif + &apic_physflat, + NULL, +}; + +/* + * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode. + */ +void __init default_setup_apic_routing(void) +{ +#ifdef CONFIG_X86_X2APIC + if (x2apic && (apic != &apic_x2apic_phys && +#ifdef CONFIG_X86_UV + apic != &apic_x2apic_uv_x && +#endif + apic != &apic_x2apic_cluster)) { + if (x2apic_phys) + apic = &apic_x2apic_phys; + else + apic = &apic_x2apic_cluster; + printk(KERN_INFO "Setting APIC routing to %s\n", apic->name); + } +#endif + + if (apic == &apic_flat) { + if (max_physical_apicid >= 8) + apic = &apic_physflat; + printk(KERN_INFO "Setting APIC routing to %s\n", apic->name); + } + + /* + * Now that apic routing model is selected, configure the + * fault handling for intr remapping. + */ + if (intr_remapping_enabled) + enable_drhd_fault_handling(); +} + +/* Same for both flat and physical. */ + +void apic_send_IPI_self(int vector) +{ + __default_send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL); +} + +int __init default_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + int i; + + for (i = 0; apic_probe[i]; ++i) { + if (apic_probe[i]->acpi_madt_oem_check(oem_id, oem_table_id)) { + apic = apic_probe[i]; + printk(KERN_INFO "Setting APIC routing to %s.\n", + apic->name); + return 1; + } + } + return 0; +} diff --git a/arch/x86/kernel/apic/summit_32.c b/arch/x86/kernel/apic/summit_32.c new file mode 100644 index 00000000000..aac52fa873f --- /dev/null +++ b/arch/x86/kernel/apic/summit_32.c @@ -0,0 +1,579 @@ +/* + * IBM Summit-Specific Code + * + * Written By: Matthew Dobson, IBM Corporation + * + * Copyright (c) 2003 IBM Corp. + * + * All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + * + * Send feedback to <colpatch@us.ibm.com> + * + */ + +#include <linux/mm.h> +#include <linux/init.h> +#include <asm/io.h> +#include <asm/bios_ebda.h> + +/* + * APIC driver for the IBM "Summit" chipset. + */ +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <asm/mpspec.h> +#include <asm/apic.h> +#include <asm/smp.h> +#include <asm/fixmap.h> +#include <asm/apicdef.h> +#include <asm/ipi.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/init.h> +#include <linux/gfp.h> +#include <linux/smp.h> + +static unsigned summit_get_apic_id(unsigned long x) +{ + return (x >> 24) & 0xFF; +} + +static inline void summit_send_IPI_mask(const cpumask_t *mask, int vector) +{ + default_send_IPI_mask_sequence_logical(mask, vector); +} + +static void summit_send_IPI_allbutself(int vector) +{ + cpumask_t mask = cpu_online_map; + cpu_clear(smp_processor_id(), mask); + + if (!cpus_empty(mask)) + summit_send_IPI_mask(&mask, vector); +} + +static void summit_send_IPI_all(int vector) +{ + summit_send_IPI_mask(&cpu_online_map, vector); +} + +#include <asm/tsc.h> + +extern int use_cyclone; + +#ifdef CONFIG_X86_SUMMIT_NUMA +static void setup_summit(void); +#else +static inline void setup_summit(void) {} +#endif + +static int summit_mps_oem_check(struct mpc_table *mpc, char *oem, + char *productid) +{ + if (!strncmp(oem, "IBM ENSW", 8) && + (!strncmp(productid, "VIGIL SMP", 9) + || !strncmp(productid, "EXA", 3) + || !strncmp(productid, "RUTHLESS SMP", 12))){ + mark_tsc_unstable("Summit based system"); + use_cyclone = 1; /*enable cyclone-timer*/ + setup_summit(); + return 1; + } + return 0; +} + +/* Hook from generic ACPI tables.c */ +static int summit_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + if (!strncmp(oem_id, "IBM", 3) && + (!strncmp(oem_table_id, "SERVIGIL", 8) + || !strncmp(oem_table_id, "EXA", 3))){ + mark_tsc_unstable("Summit based system"); + use_cyclone = 1; /*enable cyclone-timer*/ + setup_summit(); + return 1; + } + return 0; +} + +struct rio_table_hdr { + unsigned char version; /* Version number of this data structure */ + /* Version 3 adds chassis_num & WP_index */ + unsigned char num_scal_dev; /* # of Scalability devices (Twisters for Vigil) */ + unsigned char num_rio_dev; /* # of RIO I/O devices (Cyclones and Winnipegs) */ +} __attribute__((packed)); + +struct scal_detail { + unsigned char node_id; /* Scalability Node ID */ + unsigned long CBAR; /* Address of 1MB register space */ + unsigned char port0node; /* Node ID port connected to: 0xFF=None */ + unsigned char port0port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char port1node; /* Node ID port connected to: 0xFF = None */ + unsigned char port1port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char port2node; /* Node ID port connected to: 0xFF = None */ + unsigned char port2port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char chassis_num; /* 1 based Chassis number (1 = boot node) */ +} __attribute__((packed)); + +struct rio_detail { + unsigned char node_id; /* RIO Node ID */ + unsigned long BBAR; /* Address of 1MB register space */ + unsigned char type; /* Type of device */ + unsigned char owner_id; /* For WPEG: Node ID of Cyclone that owns this WPEG*/ + /* For CYC: Node ID of Twister that owns this CYC */ + unsigned char port0node; /* Node ID port connected to: 0xFF=None */ + unsigned char port0port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char port1node; /* Node ID port connected to: 0xFF=None */ + unsigned char port1port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char first_slot; /* For WPEG: Lowest slot number below this WPEG */ + /* For CYC: 0 */ + unsigned char status; /* For WPEG: Bit 0 = 1 : the XAPIC is used */ + /* = 0 : the XAPIC is not used, ie:*/ + /* ints fwded to another XAPIC */ + /* Bits1:7 Reserved */ + /* For CYC: Bits0:7 Reserved */ + unsigned char WP_index; /* For WPEG: WPEG instance index - lower ones have */ + /* lower slot numbers/PCI bus numbers */ + /* For CYC: No meaning */ + unsigned char chassis_num; /* 1 based Chassis number */ + /* For LookOut WPEGs this field indicates the */ + /* Expansion Chassis #, enumerated from Boot */ + /* Node WPEG external port, then Boot Node CYC */ + /* external port, then Next Vigil chassis WPEG */ + /* external port, etc. */ + /* Shared Lookouts have only 1 chassis number (the */ + /* first one assigned) */ +} __attribute__((packed)); + + +typedef enum { + CompatTwister = 0, /* Compatibility Twister */ + AltTwister = 1, /* Alternate Twister of internal 8-way */ + CompatCyclone = 2, /* Compatibility Cyclone */ + AltCyclone = 3, /* Alternate Cyclone of internal 8-way */ + CompatWPEG = 4, /* Compatibility WPEG */ + AltWPEG = 5, /* Second Planar WPEG */ + LookOutAWPEG = 6, /* LookOut WPEG */ + LookOutBWPEG = 7, /* LookOut WPEG */ +} node_type; + +static inline int is_WPEG(struct rio_detail *rio){ + return (rio->type == CompatWPEG || rio->type == AltWPEG || + rio->type == LookOutAWPEG || rio->type == LookOutBWPEG); +} + + +/* In clustered mode, the high nibble of APIC ID is a cluster number. + * The low nibble is a 4-bit bitmap. */ +#define XAPIC_DEST_CPUS_SHIFT 4 +#define XAPIC_DEST_CPUS_MASK ((1u << XAPIC_DEST_CPUS_SHIFT) - 1) +#define XAPIC_DEST_CLUSTER_MASK (XAPIC_DEST_CPUS_MASK << XAPIC_DEST_CPUS_SHIFT) + +#define SUMMIT_APIC_DFR_VALUE (APIC_DFR_CLUSTER) + +static const cpumask_t *summit_target_cpus(void) +{ + /* CPU_MASK_ALL (0xff) has undefined behaviour with + * dest_LowestPrio mode logical clustered apic interrupt routing + * Just start on cpu 0. IRQ balancing will spread load + */ + return &cpumask_of_cpu(0); +} + +static unsigned long summit_check_apicid_used(physid_mask_t bitmap, int apicid) +{ + return 0; +} + +/* we don't use the phys_cpu_present_map to indicate apicid presence */ +static unsigned long summit_check_apicid_present(int bit) +{ + return 1; +} + +static void summit_init_apic_ldr(void) +{ + unsigned long val, id; + int count = 0; + u8 my_id = (u8)hard_smp_processor_id(); + u8 my_cluster = APIC_CLUSTER(my_id); +#ifdef CONFIG_SMP + u8 lid; + int i; + + /* Create logical APIC IDs by counting CPUs already in cluster. */ + for (count = 0, i = nr_cpu_ids; --i >= 0; ) { + lid = cpu_2_logical_apicid[i]; + if (lid != BAD_APICID && APIC_CLUSTER(lid) == my_cluster) + ++count; + } +#endif + /* We only have a 4 wide bitmap in cluster mode. If a deranged + * BIOS puts 5 CPUs in one APIC cluster, we're hosed. */ + BUG_ON(count >= XAPIC_DEST_CPUS_SHIFT); + id = my_cluster | (1UL << count); + apic_write(APIC_DFR, SUMMIT_APIC_DFR_VALUE); + val = apic_read(APIC_LDR) & ~APIC_LDR_MASK; + val |= SET_APIC_LOGICAL_ID(id); + apic_write(APIC_LDR, val); +} + +static int summit_apic_id_registered(void) +{ + return 1; +} + +static void summit_setup_apic_routing(void) +{ + printk("Enabling APIC mode: Summit. Using %d I/O APICs\n", + nr_ioapics); +} + +static int summit_apicid_to_node(int logical_apicid) +{ +#ifdef CONFIG_SMP + return apicid_2_node[hard_smp_processor_id()]; +#else + return 0; +#endif +} + +/* Mapping from cpu number to logical apicid */ +static inline int summit_cpu_to_logical_apicid(int cpu) +{ +#ifdef CONFIG_SMP + if (cpu >= nr_cpu_ids) + return BAD_APICID; + return cpu_2_logical_apicid[cpu]; +#else + return logical_smp_processor_id(); +#endif +} + +static int summit_cpu_present_to_apicid(int mps_cpu) +{ + if (mps_cpu < nr_cpu_ids) + return (int)per_cpu(x86_bios_cpu_apicid, mps_cpu); + else + return BAD_APICID; +} + +static physid_mask_t summit_ioapic_phys_id_map(physid_mask_t phys_id_map) +{ + /* For clustered we don't have a good way to do this yet - hack */ + return physids_promote(0x0F); +} + +static physid_mask_t summit_apicid_to_cpu_present(int apicid) +{ + return physid_mask_of_physid(0); +} + +static int summit_check_phys_apicid_present(int boot_cpu_physical_apicid) +{ + return 1; +} + +static unsigned int summit_cpu_mask_to_apicid(const cpumask_t *cpumask) +{ + unsigned int round = 0; + int cpu, apicid = 0; + + /* + * The cpus in the mask must all be on the apic cluster. + */ + for_each_cpu(cpu, cpumask) { + int new_apicid = summit_cpu_to_logical_apicid(cpu); + + if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) { + printk("%s: Not a valid mask!\n", __func__); + return BAD_APICID; + } + apicid |= new_apicid; + round++; + } + return apicid; +} + +static unsigned int summit_cpu_mask_to_apicid_and(const struct cpumask *inmask, + const struct cpumask *andmask) +{ + int apicid = summit_cpu_to_logical_apicid(0); + cpumask_var_t cpumask; + + if (!alloc_cpumask_var(&cpumask, GFP_ATOMIC)) + return apicid; + + cpumask_and(cpumask, inmask, andmask); + cpumask_and(cpumask, cpumask, cpu_online_mask); + apicid = summit_cpu_mask_to_apicid(cpumask); + + free_cpumask_var(cpumask); + + return apicid; +} + +/* + * cpuid returns the value latched in the HW at reset, not the APIC ID + * register's value. For any box whose BIOS changes APIC IDs, like + * clustered APIC systems, we must use hard_smp_processor_id. + * + * See Intel's IA-32 SW Dev's Manual Vol2 under CPUID. + */ +static int summit_phys_pkg_id(int cpuid_apic, int index_msb) +{ + return hard_smp_processor_id() >> index_msb; +} + +static int probe_summit(void) +{ + /* probed later in mptable/ACPI hooks */ + return 0; +} + +static void summit_vector_allocation_domain(int cpu, cpumask_t *retmask) +{ + /* Careful. Some cpus do not strictly honor the set of cpus + * specified in the interrupt destination when using lowest + * priority interrupt delivery mode. + * + * In particular there was a hyperthreading cpu observed to + * deliver interrupts to the wrong hyperthread when only one + * hyperthread was specified in the interrupt desitination. + */ + *retmask = (cpumask_t){ { [0] = APIC_ALL_CPUS, } }; +} + +#ifdef CONFIG_X86_SUMMIT_NUMA +static struct rio_table_hdr *rio_table_hdr; +static struct scal_detail *scal_devs[MAX_NUMNODES]; +static struct rio_detail *rio_devs[MAX_NUMNODES*4]; + +#ifndef CONFIG_X86_NUMAQ +static int mp_bus_id_to_node[MAX_MP_BUSSES]; +#endif + +static int setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus) +{ + int twister = 0, node = 0; + int i, bus, num_buses; + + for (i = 0; i < rio_table_hdr->num_rio_dev; i++) { + if (rio_devs[i]->node_id == rio_devs[wpeg_num]->owner_id) { + twister = rio_devs[i]->owner_id; + break; + } + } + if (i == rio_table_hdr->num_rio_dev) { + printk(KERN_ERR "%s: Couldn't find owner Cyclone for Winnipeg!\n", __func__); + return last_bus; + } + + for (i = 0; i < rio_table_hdr->num_scal_dev; i++) { + if (scal_devs[i]->node_id == twister) { + node = scal_devs[i]->node_id; + break; + } + } + if (i == rio_table_hdr->num_scal_dev) { + printk(KERN_ERR "%s: Couldn't find owner Twister for Cyclone!\n", __func__); + return last_bus; + } + + switch (rio_devs[wpeg_num]->type) { + case CompatWPEG: + /* + * The Compatibility Winnipeg controls the 2 legacy buses, + * the 66MHz PCI bus [2 slots] and the 2 "extra" buses in case + * a PCI-PCI bridge card is used in either slot: total 5 buses. + */ + num_buses = 5; + break; + case AltWPEG: + /* + * The Alternate Winnipeg controls the 2 133MHz buses [1 slot + * each], their 2 "extra" buses, the 100MHz bus [2 slots] and + * the "extra" buses for each of those slots: total 7 buses. + */ + num_buses = 7; + break; + case LookOutAWPEG: + case LookOutBWPEG: + /* + * A Lookout Winnipeg controls 3 100MHz buses [2 slots each] + * & the "extra" buses for each of those slots: total 9 buses. + */ + num_buses = 9; + break; + default: + printk(KERN_INFO "%s: Unsupported Winnipeg type!\n", __func__); + return last_bus; + } + + for (bus = last_bus; bus < last_bus + num_buses; bus++) + mp_bus_id_to_node[bus] = node; + return bus; +} + +static int build_detail_arrays(void) +{ + unsigned long ptr; + int i, scal_detail_size, rio_detail_size; + + if (rio_table_hdr->num_scal_dev > MAX_NUMNODES) { + printk(KERN_WARNING "%s: MAX_NUMNODES too low! Defined as %d, but system has %d nodes.\n", __func__, MAX_NUMNODES, rio_table_hdr->num_scal_dev); + return 0; + } + + switch (rio_table_hdr->version) { + default: + printk(KERN_WARNING "%s: Invalid Rio Grande Table Version: %d\n", __func__, rio_table_hdr->version); + return 0; + case 2: + scal_detail_size = 11; + rio_detail_size = 13; + break; + case 3: + scal_detail_size = 12; + rio_detail_size = 15; + break; + } + + ptr = (unsigned long)rio_table_hdr + 3; + for (i = 0; i < rio_table_hdr->num_scal_dev; i++, ptr += scal_detail_size) + scal_devs[i] = (struct scal_detail *)ptr; + + for (i = 0; i < rio_table_hdr->num_rio_dev; i++, ptr += rio_detail_size) + rio_devs[i] = (struct rio_detail *)ptr; + + return 1; +} + +void setup_summit(void) +{ + unsigned long ptr; + unsigned short offset; + int i, next_wpeg, next_bus = 0; + + /* The pointer to the EBDA is stored in the word @ phys 0x40E(40:0E) */ + ptr = get_bios_ebda(); + ptr = (unsigned long)phys_to_virt(ptr); + + rio_table_hdr = NULL; + offset = 0x180; + while (offset) { + /* The block id is stored in the 2nd word */ + if (*((unsigned short *)(ptr + offset + 2)) == 0x4752) { + /* set the pointer past the offset & block id */ + rio_table_hdr = (struct rio_table_hdr *)(ptr + offset + 4); + break; + } + /* The next offset is stored in the 1st word. 0 means no more */ + offset = *((unsigned short *)(ptr + offset)); + } + if (!rio_table_hdr) { + printk(KERN_ERR "%s: Unable to locate Rio Grande Table in EBDA - bailing!\n", __func__); + return; + } + + if (!build_detail_arrays()) + return; + + /* The first Winnipeg we're looking for has an index of 0 */ + next_wpeg = 0; + do { + for (i = 0; i < rio_table_hdr->num_rio_dev; i++) { + if (is_WPEG(rio_devs[i]) && rio_devs[i]->WP_index == next_wpeg) { + /* It's the Winnipeg we're looking for! */ + next_bus = setup_pci_node_map_for_wpeg(i, next_bus); + next_wpeg++; + break; + } + } + /* + * If we go through all Rio devices and don't find one with + * the next index, it means we've found all the Winnipegs, + * and thus all the PCI buses. + */ + if (i == rio_table_hdr->num_rio_dev) + next_wpeg = 0; + } while (next_wpeg != 0); +} +#endif + +struct apic apic_summit = { + + .name = "summit", + .probe = probe_summit, + .acpi_madt_oem_check = summit_acpi_madt_oem_check, + .apic_id_registered = summit_apic_id_registered, + + .irq_delivery_mode = dest_LowestPrio, + /* logical delivery broadcast to all CPUs: */ + .irq_dest_mode = 1, + + .target_cpus = summit_target_cpus, + .disable_esr = 1, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = summit_check_apicid_used, + .check_apicid_present = summit_check_apicid_present, + + .vector_allocation_domain = summit_vector_allocation_domain, + .init_apic_ldr = summit_init_apic_ldr, + + .ioapic_phys_id_map = summit_ioapic_phys_id_map, + .setup_apic_routing = summit_setup_apic_routing, + .multi_timer_check = NULL, + .apicid_to_node = summit_apicid_to_node, + .cpu_to_logical_apicid = summit_cpu_to_logical_apicid, + .cpu_present_to_apicid = summit_cpu_present_to_apicid, + .apicid_to_cpu_present = summit_apicid_to_cpu_present, + .setup_portio_remap = NULL, + .check_phys_apicid_present = summit_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = summit_phys_pkg_id, + .mps_oem_check = summit_mps_oem_check, + + .get_apic_id = summit_get_apic_id, + .set_apic_id = NULL, + .apic_id_mask = 0xFF << 24, + + .cpu_mask_to_apicid = summit_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = summit_cpu_mask_to_apicid_and, + + .send_IPI_mask = summit_send_IPI_mask, + .send_IPI_mask_allbutself = NULL, + .send_IPI_allbutself = summit_send_IPI_allbutself, + .send_IPI_all = summit_send_IPI_all, + .send_IPI_self = default_send_IPI_self, + + .trampoline_phys_low = DEFAULT_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = DEFAULT_TRAMPOLINE_PHYS_HIGH, + + .wait_for_init_deassert = default_wait_for_init_deassert, + + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = default_inquire_remote_apic, + + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, +}; diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c new file mode 100644 index 00000000000..4a903e2f0d1 --- /dev/null +++ b/arch/x86/kernel/apic/x2apic_cluster.c @@ -0,0 +1,245 @@ +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/string.h> +#include <linux/kernel.h> +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/dmar.h> + +#include <asm/smp.h> +#include <asm/apic.h> +#include <asm/ipi.h> + +DEFINE_PER_CPU(u32, x86_cpu_to_logical_apicid); + +static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + return x2apic_enabled(); +} + +/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ + +static const struct cpumask *x2apic_target_cpus(void) +{ + return cpumask_of(0); +} + +/* + * for now each logical cpu is in its own vector allocation domain. + */ +static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + cpumask_clear(retmask); + cpumask_set_cpu(cpu, retmask); +} + +static void + __x2apic_send_IPI_dest(unsigned int apicid, int vector, unsigned int dest) +{ + unsigned long cfg; + + cfg = __prepare_ICR(0, vector, dest); + + /* + * send the IPI. + */ + native_x2apic_icr_write(cfg, apicid); +} + +/* + * for now, we send the IPI's one by one in the cpumask. + * TBD: Based on the cpu mask, we can send the IPI's to the cluster group + * at once. We have 16 cpu's in a cluster. This will minimize IPI register + * writes. + */ +static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) +{ + unsigned long query_cpu; + unsigned long flags; + + x2apic_wrmsr_fence(); + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + __x2apic_send_IPI_dest( + per_cpu(x86_cpu_to_logical_apicid, query_cpu), + vector, apic->dest_logical); + } + local_irq_restore(flags); +} + +static void + x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector) +{ + unsigned long this_cpu = smp_processor_id(); + unsigned long query_cpu; + unsigned long flags; + + x2apic_wrmsr_fence(); + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + if (query_cpu == this_cpu) + continue; + __x2apic_send_IPI_dest( + per_cpu(x86_cpu_to_logical_apicid, query_cpu), + vector, apic->dest_logical); + } + local_irq_restore(flags); +} + +static void x2apic_send_IPI_allbutself(int vector) +{ + unsigned long this_cpu = smp_processor_id(); + unsigned long query_cpu; + unsigned long flags; + + x2apic_wrmsr_fence(); + + local_irq_save(flags); + for_each_online_cpu(query_cpu) { + if (query_cpu == this_cpu) + continue; + __x2apic_send_IPI_dest( + per_cpu(x86_cpu_to_logical_apicid, query_cpu), + vector, apic->dest_logical); + } + local_irq_restore(flags); +} + +static void x2apic_send_IPI_all(int vector) +{ + x2apic_send_IPI_mask(cpu_online_mask, vector); +} + +static int x2apic_apic_id_registered(void) +{ + return 1; +} + +static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask) +{ + /* + * We're using fixed IRQ delivery, can only return one logical APIC ID. + * May as well be the first. + */ + int cpu = cpumask_first(cpumask); + + if ((unsigned)cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_logical_apicid, cpu); + else + return BAD_APICID; +} + +static unsigned int +x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one logical APIC ID. + * May as well be the first. + */ + for_each_cpu_and(cpu, cpumask, andmask) { + if (cpumask_test_cpu(cpu, cpu_online_mask)) + break; + } + + if (cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_logical_apicid, cpu); + + return BAD_APICID; +} + +static unsigned int x2apic_cluster_phys_get_apic_id(unsigned long x) +{ + unsigned int id; + + id = x; + return id; +} + +static unsigned long set_apic_id(unsigned int id) +{ + unsigned long x; + + x = id; + return x; +} + +static int x2apic_cluster_phys_pkg_id(int initial_apicid, int index_msb) +{ + return current_cpu_data.initial_apicid >> index_msb; +} + +static void x2apic_send_IPI_self(int vector) +{ + apic_write(APIC_SELF_IPI, vector); +} + +static void init_x2apic_ldr(void) +{ + int cpu = smp_processor_id(); + + per_cpu(x86_cpu_to_logical_apicid, cpu) = apic_read(APIC_LDR); +} + +struct apic apic_x2apic_cluster = { + + .name = "cluster x2apic", + .probe = NULL, + .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, + .apic_id_registered = x2apic_apic_id_registered, + + .irq_delivery_mode = dest_LowestPrio, + .irq_dest_mode = 1, /* logical */ + + .target_cpus = x2apic_target_cpus, + .disable_esr = 0, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = x2apic_vector_allocation_domain, + .init_apic_ldr = init_x2apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = x2apic_cluster_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = x2apic_cluster_phys_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFFFFFFFu, + + .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and, + + .send_IPI_mask = x2apic_send_IPI_mask, + .send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself, + .send_IPI_allbutself = x2apic_send_IPI_allbutself, + .send_IPI_all = x2apic_send_IPI_all, + .send_IPI_self = x2apic_send_IPI_self, + + .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 = native_apic_msr_read, + .write = native_apic_msr_write, + .icr_read = native_x2apic_icr_read, + .icr_write = native_x2apic_icr_write, + .wait_icr_idle = native_x2apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_x2apic_wait_icr_idle, +}; diff --git a/arch/x86/kernel/genx2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c index 21bcc0e098b..a284359627e 100644 --- a/arch/x86/kernel/genx2apic_phys.c +++ b/arch/x86/kernel/apic/x2apic_phys.c @@ -7,10 +7,10 @@ #include <linux/dmar.h> #include <asm/smp.h> +#include <asm/apic.h> #include <asm/ipi.h> -#include <asm/genapic.h> -static int x2apic_phys; +int x2apic_phys; static int set_x2apic_phys_mode(char *arg) { @@ -21,10 +21,10 @@ early_param("x2apic_phys", set_x2apic_phys_mode); static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { - if (cpu_has_x2apic && x2apic_phys) - return 1; - - return 0; + if (x2apic_phys) + return x2apic_enabled(); + else + return 0; } /* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ @@ -50,13 +50,15 @@ static void __x2apic_send_IPI_dest(unsigned int apicid, int vector, /* * send the IPI. */ - x2apic_icr_write(cfg, apicid); + native_x2apic_icr_write(cfg, apicid); } static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) { - unsigned long flags; unsigned long query_cpu; + unsigned long flags; + + x2apic_wrmsr_fence(); local_irq_save(flags); for_each_cpu(query_cpu, mask) { @@ -66,12 +68,14 @@ static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) local_irq_restore(flags); } -static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, - int vector) +static void + x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector) { - unsigned long flags; - unsigned long query_cpu; unsigned long this_cpu = smp_processor_id(); + unsigned long query_cpu; + unsigned long flags; + + x2apic_wrmsr_fence(); local_irq_save(flags); for_each_cpu(query_cpu, mask) { @@ -85,16 +89,19 @@ static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, static void x2apic_send_IPI_allbutself(int vector) { - unsigned long flags; - unsigned long query_cpu; unsigned long this_cpu = smp_processor_id(); + unsigned long query_cpu; + unsigned long flags; + + x2apic_wrmsr_fence(); local_irq_save(flags); - for_each_online_cpu(query_cpu) - if (query_cpu != this_cpu) - __x2apic_send_IPI_dest( - per_cpu(x86_cpu_to_apicid, query_cpu), - vector, APIC_DEST_PHYSICAL); + for_each_online_cpu(query_cpu) { + if (query_cpu == this_cpu) + continue; + __x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu), + vector, APIC_DEST_PHYSICAL); + } local_irq_restore(flags); } @@ -110,21 +117,21 @@ static int x2apic_apic_id_registered(void) static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask) { - int cpu; - /* * We're using fixed IRQ delivery, can only return one phys APIC ID. * May as well be the first. */ - cpu = cpumask_first(cpumask); + int cpu = cpumask_first(cpumask); + if ((unsigned)cpu < nr_cpu_ids) return per_cpu(x86_cpu_to_apicid, cpu); else return BAD_APICID; } -static unsigned int x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, - const struct cpumask *andmask) +static unsigned int +x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) { int cpu; @@ -132,31 +139,28 @@ static unsigned int x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, * We're using fixed IRQ delivery, can only return one phys APIC ID. * May as well be the first. */ - for_each_cpu_and(cpu, cpumask, andmask) + for_each_cpu_and(cpu, cpumask, andmask) { 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; } -static unsigned int get_apic_id(unsigned long x) +static unsigned int x2apic_phys_get_apic_id(unsigned long x) { - unsigned int id; - - id = x; - return id; + return x; } static unsigned long set_apic_id(unsigned int id) { - unsigned long x; - - x = id; - return x; + return id; } -static unsigned int phys_pkg_id(int index_msb) +static int x2apic_phys_pkg_id(int initial_apicid, int index_msb) { return current_cpu_data.initial_apicid >> index_msb; } @@ -168,27 +172,63 @@ static void x2apic_send_IPI_self(int vector) static void init_x2apic_ldr(void) { - return; -} - -struct genapic apic_x2apic_phys = { - .name = "physical x2apic", - .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, - .int_delivery_mode = dest_Fixed, - .int_dest_mode = (APIC_DEST_PHYSICAL != 0), - .target_cpus = x2apic_target_cpus, - .vector_allocation_domain = x2apic_vector_allocation_domain, - .apic_id_registered = x2apic_apic_id_registered, - .init_apic_ldr = init_x2apic_ldr, - .send_IPI_all = x2apic_send_IPI_all, - .send_IPI_allbutself = x2apic_send_IPI_allbutself, - .send_IPI_mask = x2apic_send_IPI_mask, - .send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself, - .send_IPI_self = x2apic_send_IPI_self, - .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, - .cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and, - .phys_pkg_id = phys_pkg_id, - .get_apic_id = get_apic_id, - .set_apic_id = set_apic_id, - .apic_id_mask = (0xFFFFFFFFu), +} + +struct apic apic_x2apic_phys = { + + .name = "physical x2apic", + .probe = NULL, + .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, + .apic_id_registered = x2apic_apic_id_registered, + + .irq_delivery_mode = dest_Fixed, + .irq_dest_mode = 0, /* physical */ + + .target_cpus = x2apic_target_cpus, + .disable_esr = 0, + .dest_logical = 0, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = x2apic_vector_allocation_domain, + .init_apic_ldr = init_x2apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = x2apic_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = x2apic_phys_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFFFFFFFu, + + .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and, + + .send_IPI_mask = x2apic_send_IPI_mask, + .send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself, + .send_IPI_allbutself = x2apic_send_IPI_allbutself, + .send_IPI_all = x2apic_send_IPI_all, + .send_IPI_self = x2apic_send_IPI_self, + + .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 = native_apic_msr_read, + .write = native_apic_msr_write, + .icr_read = native_x2apic_icr_read, + .icr_write = native_x2apic_icr_write, + .wait_icr_idle = native_x2apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_x2apic_wait_icr_idle, }; diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index b193e082f6c..1bd6da1f8fa 100644 --- a/arch/x86/kernel/genx2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -7,27 +7,28 @@ * * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved. */ - -#include <linux/kernel.h> -#include <linux/threads.h> -#include <linux/cpu.h> #include <linux/cpumask.h> +#include <linux/hardirq.h> +#include <linux/proc_fs.h> +#include <linux/threads.h> +#include <linux/kernel.h> +#include <linux/module.h> #include <linux/string.h> #include <linux/ctype.h> -#include <linux/init.h> #include <linux/sched.h> -#include <linux/module.h> -#include <linux/hardirq.h> #include <linux/timer.h> -#include <linux/proc_fs.h> -#include <asm/current.h> -#include <asm/smp.h> -#include <asm/ipi.h> -#include <asm/genapic.h> -#include <asm/pgtable.h> +#include <linux/cpu.h> +#include <linux/init.h> + #include <asm/uv/uv_mmrs.h> #include <asm/uv/uv_hub.h> +#include <asm/current.h> +#include <asm/pgtable.h> #include <asm/uv/bios.h> +#include <asm/uv/uv.h> +#include <asm/apic.h> +#include <asm/ipi.h> +#include <asm/smp.h> DEFINE_PER_CPU(int, x2apic_extra_bits); @@ -90,39 +91,43 @@ static void uv_vector_allocation_domain(int cpu, struct cpumask *retmask) cpumask_set_cpu(cpu, retmask); } -int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip) +static int uv_wakeup_secondary(int phys_apicid, unsigned long start_rip) { +#ifdef CONFIG_SMP unsigned long val; int pnode; pnode = uv_apicid_to_pnode(phys_apicid); val = (1UL << UVH_IPI_INT_SEND_SHFT) | (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) | - (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) | + ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) | APIC_DM_INIT; uv_write_global_mmr64(pnode, UVH_IPI_INT, val); mdelay(10); val = (1UL << UVH_IPI_INT_SEND_SHFT) | (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) | - (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) | + ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) | APIC_DM_STARTUP; uv_write_global_mmr64(pnode, UVH_IPI_INT, val); + + atomic_set(&init_deasserted, 1); +#endif return 0; } static void uv_send_IPI_one(int cpu, int vector) { - unsigned long val, apicid, lapicid; + unsigned long val, apicid; int pnode; apicid = per_cpu(x86_cpu_to_apicid, cpu); - lapicid = apicid & 0x3f; /* ZZZ macro needed */ pnode = uv_apicid_to_pnode(apicid); - val = - (1UL << UVH_IPI_INT_SEND_SHFT) | (lapicid << - UVH_IPI_INT_APIC_ID_SHFT) | - (vector << UVH_IPI_INT_VECTOR_SHFT); + + val = (1UL << UVH_IPI_INT_SEND_SHFT) | + (apicid << UVH_IPI_INT_APIC_ID_SHFT) | + (vector << UVH_IPI_INT_VECTOR_SHFT); + uv_write_global_mmr64(pnode, UVH_IPI_INT, val); } @@ -136,22 +141,24 @@ static void uv_send_IPI_mask(const struct cpumask *mask, int vector) static void uv_send_IPI_mask_allbutself(const struct cpumask *mask, int vector) { - unsigned int cpu; unsigned int this_cpu = smp_processor_id(); + unsigned int cpu; - for_each_cpu(cpu, mask) + for_each_cpu(cpu, mask) { if (cpu != this_cpu) uv_send_IPI_one(cpu, vector); + } } static void uv_send_IPI_allbutself(int vector) { - unsigned int cpu; unsigned int this_cpu = smp_processor_id(); + unsigned int cpu; - for_each_online_cpu(cpu) + for_each_online_cpu(cpu) { if (cpu != this_cpu) uv_send_IPI_one(cpu, vector); + } } static void uv_send_IPI_all(int vector) @@ -170,21 +177,21 @@ static void uv_init_apic_ldr(void) static unsigned int uv_cpu_mask_to_apicid(const struct cpumask *cpumask) { - int cpu; - /* * We're using fixed IRQ delivery, can only return one phys APIC ID. * May as well be the first. */ - cpu = cpumask_first(cpumask); + int cpu = cpumask_first(cpumask); + if ((unsigned)cpu < nr_cpu_ids) return per_cpu(x86_cpu_to_apicid, cpu); else return BAD_APICID; } -static unsigned int uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask, - const struct cpumask *andmask) +static unsigned int +uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) { int cpu; @@ -192,15 +199,17 @@ static unsigned int uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask, * We're using fixed IRQ delivery, can only return one phys APIC ID. * May as well be the first. */ - for_each_cpu_and(cpu, cpumask, andmask) + for_each_cpu_and(cpu, cpumask, andmask) { 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; } -static unsigned int get_apic_id(unsigned long x) +static unsigned int x2apic_get_apic_id(unsigned long x) { unsigned int id; @@ -222,10 +231,10 @@ static unsigned long set_apic_id(unsigned int id) static unsigned int uv_read_apic_id(void) { - return get_apic_id(apic_read(APIC_ID)); + return x2apic_get_apic_id(apic_read(APIC_ID)); } -static unsigned int phys_pkg_id(int index_msb) +static int uv_phys_pkg_id(int initial_apicid, int index_msb) { return uv_read_apic_id() >> index_msb; } @@ -235,26 +244,64 @@ static void uv_send_IPI_self(int vector) apic_write(APIC_SELF_IPI, vector); } -struct genapic apic_x2apic_uv_x = { - .name = "UV large system", - .acpi_madt_oem_check = uv_acpi_madt_oem_check, - .int_delivery_mode = dest_Fixed, - .int_dest_mode = (APIC_DEST_PHYSICAL != 0), - .target_cpus = uv_target_cpus, - .vector_allocation_domain = uv_vector_allocation_domain, - .apic_id_registered = uv_apic_id_registered, - .init_apic_ldr = uv_init_apic_ldr, - .send_IPI_all = uv_send_IPI_all, - .send_IPI_allbutself = uv_send_IPI_allbutself, - .send_IPI_mask = uv_send_IPI_mask, - .send_IPI_mask_allbutself = uv_send_IPI_mask_allbutself, - .send_IPI_self = uv_send_IPI_self, - .cpu_mask_to_apicid = uv_cpu_mask_to_apicid, - .cpu_mask_to_apicid_and = uv_cpu_mask_to_apicid_and, - .phys_pkg_id = phys_pkg_id, - .get_apic_id = get_apic_id, - .set_apic_id = set_apic_id, - .apic_id_mask = (0xFFFFFFFFu), +struct apic apic_x2apic_uv_x = { + + .name = "UV large system", + .probe = NULL, + .acpi_madt_oem_check = uv_acpi_madt_oem_check, + .apic_id_registered = uv_apic_id_registered, + + .irq_delivery_mode = dest_Fixed, + .irq_dest_mode = 1, /* logical */ + + .target_cpus = uv_target_cpus, + .disable_esr = 0, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = uv_vector_allocation_domain, + .init_apic_ldr = uv_init_apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = uv_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = x2apic_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFFFFFFFu, + + .cpu_mask_to_apicid = uv_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = uv_cpu_mask_to_apicid_and, + + .send_IPI_mask = uv_send_IPI_mask, + .send_IPI_mask_allbutself = uv_send_IPI_mask_allbutself, + .send_IPI_allbutself = uv_send_IPI_allbutself, + .send_IPI_all = uv_send_IPI_all, + .send_IPI_self = uv_send_IPI_self, + + .wakeup_secondary_cpu = uv_wakeup_secondary, + .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 = native_apic_msr_read, + .write = native_apic_msr_write, + .icr_read = native_x2apic_icr_read, + .icr_write = native_x2apic_icr_write, + .wait_icr_idle = native_x2apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_x2apic_wait_icr_idle, }; static __cpuinit void set_x2apic_extra_bits(int pnode) @@ -322,7 +369,7 @@ static __init void map_high(char *id, unsigned long base, int shift, paddr = base << shift; bytes = (1UL << shift) * (max_pnode + 1); printk(KERN_INFO "UV: Map %s_HI 0x%lx - 0x%lx\n", id, paddr, - paddr + bytes); + paddr + bytes); if (map_type == map_uc) init_extra_mapping_uc(paddr, bytes); else @@ -485,7 +532,7 @@ late_initcall(uv_init_heartbeat); /* * Called on each cpu to initialize the per_cpu UV data area. - * ZZZ hotplug not supported yet + * FIXME: hotplug not supported yet */ void __cpuinit uv_cpu_init(void) { diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 266ec6c18b6..10033fe718e 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -301,7 +301,7 @@ extern int (*console_blank_hook)(int); */ #define APM_ZERO_SEGS -#include "apm.h" +#include <asm/apm.h> /* * Define to re-initialize the interrupt 0 timer to 100 Hz after a suspend. diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c index ee4df08feee..fbf2f33e308 100644 --- a/arch/x86/kernel/asm-offsets_32.c +++ b/arch/x86/kernel/asm-offsets_32.c @@ -75,6 +75,7 @@ void foo(void) OFFSET(PT_DS, pt_regs, ds); OFFSET(PT_ES, pt_regs, es); OFFSET(PT_FS, pt_regs, fs); + OFFSET(PT_GS, pt_regs, gs); OFFSET(PT_ORIG_EAX, pt_regs, orig_ax); OFFSET(PT_EIP, pt_regs, ip); OFFSET(PT_CS, pt_regs, cs); diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c index 1d41d3f1edb..8793ab33e2c 100644 --- a/arch/x86/kernel/asm-offsets_64.c +++ b/arch/x86/kernel/asm-offsets_64.c @@ -11,7 +11,6 @@ #include <linux/hardirq.h> #include <linux/suspend.h> #include <linux/kbuild.h> -#include <asm/pda.h> #include <asm/processor.h> #include <asm/segment.h> #include <asm/thread_info.h> @@ -48,16 +47,6 @@ int main(void) #endif BLANK(); #undef ENTRY -#define ENTRY(entry) DEFINE(pda_ ## entry, offsetof(struct x8664_pda, entry)) - ENTRY(kernelstack); - ENTRY(oldrsp); - ENTRY(pcurrent); - ENTRY(irqcount); - ENTRY(cpunumber); - ENTRY(irqstackptr); - ENTRY(data_offset); - BLANK(); -#undef ENTRY #ifdef CONFIG_PARAVIRT BLANK(); OFFSET(PARAVIRT_enabled, pv_info, paravirt_enabled); diff --git a/arch/x86/kernel/check.c b/arch/x86/kernel/check.c index 2ac0ab71412..fc999e6fc46 100644 --- a/arch/x86/kernel/check.c +++ b/arch/x86/kernel/check.c @@ -83,15 +83,15 @@ void __init setup_bios_corruption_check(void) u64 size; addr = find_e820_area_size(addr, &size, PAGE_SIZE); - if (addr == 0) + if (!(addr + 1)) + break; + + if (addr >= corruption_check_size) break; if ((addr + size) > corruption_check_size) size = corruption_check_size - addr; - if (size == 0) - break; - e820_update_range(addr, size, E820_RAM, E820_RESERVED); scan_areas[num_scan_areas].addr = addr; scan_areas[num_scan_areas].size = size; diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 82db7f45e2d..4e242f9a06e 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -14,11 +14,12 @@ obj-y += vmware.o hypervisor.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 -obj-$(CONFIG_CPU_SUP_CENTAUR_32) += centaur.o -obj-$(CONFIG_CPU_SUP_CENTAUR_64) += centaur_64.o +obj-$(CONFIG_CPU_SUP_CENTAUR) += centaur.o obj-$(CONFIG_CPU_SUP_TRANSMETA_32) += transmeta.o obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c index 2cf23634b6d..8220ae69849 100644 --- a/arch/x86/kernel/cpu/addon_cpuid_features.c +++ b/arch/x86/kernel/cpu/addon_cpuid_features.c @@ -7,7 +7,7 @@ #include <asm/pat.h> #include <asm/processor.h> -#include <mach_apic.h> +#include <asm/apic.h> struct cpuid_bit { u16 feature; @@ -29,7 +29,7 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) u32 regs[4]; const struct cpuid_bit *cb; - static const struct cpuid_bit cpuid_bits[] = { + static const struct cpuid_bit __cpuinitconst cpuid_bits[] = { { X86_FEATURE_IDA, CR_EAX, 1, 0x00000006 }, { 0, 0, 0, 0 } }; @@ -69,7 +69,7 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) */ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c) { -#ifdef CONFIG_X86_SMP +#ifdef CONFIG_SMP unsigned int eax, ebx, ecx, edx, sub_index; unsigned int ht_mask_width, core_plus_mask_width; unsigned int core_select_mask, core_level_siblings; @@ -116,22 +116,14 @@ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c) core_select_mask = (~(-1 << core_plus_mask_width)) >> ht_mask_width; -#ifdef CONFIG_X86_32 - c->cpu_core_id = phys_pkg_id(c->initial_apicid, ht_mask_width) + c->cpu_core_id = apic->phys_pkg_id(c->initial_apicid, ht_mask_width) & core_select_mask; - c->phys_proc_id = phys_pkg_id(c->initial_apicid, core_plus_mask_width); + c->phys_proc_id = apic->phys_pkg_id(c->initial_apicid, core_plus_mask_width); /* * Reinit the apicid, now that we have extended initial_apicid. */ - c->apicid = phys_pkg_id(c->initial_apicid, 0); -#else - c->cpu_core_id = phys_pkg_id(ht_mask_width) & core_select_mask; - c->phys_proc_id = phys_pkg_id(core_plus_mask_width); - /* - * Reinit the apicid, now that we have extended initial_apicid. - */ - c->apicid = phys_pkg_id(0); -#endif + c->apicid = apic->phys_pkg_id(c->initial_apicid, 0); + c->x86_max_cores = (core_level_siblings / smp_num_siblings); @@ -143,37 +135,3 @@ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c) return; #endif } - -#ifdef CONFIG_X86_PAT -void __cpuinit validate_pat_support(struct cpuinfo_x86 *c) -{ - if (!cpu_has_pat) - pat_disable("PAT not supported by CPU."); - - switch (c->x86_vendor) { - case X86_VENDOR_INTEL: - /* - * There is a known erratum on Pentium III and Core Solo - * and Core Duo CPUs. - * " Page with PAT set to WC while associated MTRR is UC - * may consolidate to UC " - * Because of this erratum, it is better to stick with - * setting WC in MTRR rather than using PAT on these CPUs. - * - * Enable PAT WC only on P4, Core 2 or later CPUs. - */ - if (c->x86 > 0x6 || (c->x86 == 6 && c->x86_model >= 15)) - return; - - pat_disable("PAT WC disabled due to known CPU erratum."); - return; - - case X86_VENDOR_AMD: - case X86_VENDOR_CENTAUR: - case X86_VENDOR_TRANSMETA: - return; - } - - pat_disable("PAT disabled. Not yet verified on this CPU type."); -} -#endif diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 7c878f6aa91..7e4a459daa6 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -5,6 +5,7 @@ #include <asm/io.h> #include <asm/processor.h> #include <asm/apic.h> +#include <asm/cpu.h> #ifdef CONFIG_X86_64 # include <asm/numa_64.h> @@ -12,8 +13,6 @@ # include <asm/cacheflush.h> #endif -#include <mach_apic.h> - #include "cpu.h" #ifdef CONFIG_X86_32 @@ -143,6 +142,55 @@ static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c) } } +static void __cpuinit amd_k7_smp_check(struct cpuinfo_x86 *c) +{ +#ifdef CONFIG_SMP + /* calling is from identify_secondary_cpu() ? */ + if (c->cpu_index == boot_cpu_id) + return; + + /* + * Certain Athlons might work (for various values of 'work') in SMP + * but they are not certified as MP capable. + */ + /* Athlon 660/661 is valid. */ + if ((c->x86_model == 6) && ((c->x86_mask == 0) || + (c->x86_mask == 1))) + goto valid_k7; + + /* Duron 670 is valid */ + if ((c->x86_model == 7) && (c->x86_mask == 0)) + goto valid_k7; + + /* + * Athlon 662, Duron 671, and Athlon >model 7 have capability + * bit. It's worth noting that the A5 stepping (662) of some + * Athlon XP's have the MP bit set. + * See http://www.heise.de/newsticker/data/jow-18.10.01-000 for + * more. + */ + if (((c->x86_model == 6) && (c->x86_mask >= 2)) || + ((c->x86_model == 7) && (c->x86_mask >= 1)) || + (c->x86_model > 7)) + if (cpu_has_mp) + goto valid_k7; + + /* If we get here, not a certified SMP capable AMD system. */ + + /* + * Don't taint if we are running SMP kernel on a single non-MP + * approved Athlon + */ + WARN_ONCE(1, "WARNING: This combination of AMD" + "processors is not suitable for SMP.\n"); + if (!test_taint(TAINT_UNSAFE_SMP)) + add_taint(TAINT_UNSAFE_SMP); + +valid_k7: + ; +#endif +} + static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c) { u32 l, h; @@ -177,6 +225,8 @@ static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c) } set_cpu_cap(c, X86_FEATURE_K7); + + amd_k7_smp_check(c); } #endif @@ -452,7 +502,7 @@ static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int } #endif -static struct cpu_dev amd_cpu_dev __cpuinitdata = { +static const struct cpu_dev __cpuinitconst amd_cpu_dev = { .c_vendor = "AMD", .c_ident = { "AuthenticAMD" }, #ifdef CONFIG_X86_32 diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c index 89bfdd9cacc..c95e831bb09 100644 --- a/arch/x86/kernel/cpu/centaur.c +++ b/arch/x86/kernel/cpu/centaur.c @@ -1,11 +1,11 @@ +#include <linux/bitops.h> #include <linux/kernel.h> #include <linux/init.h> -#include <linux/bitops.h> #include <asm/processor.h> -#include <asm/msr.h> #include <asm/e820.h> #include <asm/mtrr.h> +#include <asm/msr.h> #include "cpu.h" @@ -276,7 +276,7 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c) */ c->x86_capability[5] = cpuid_edx(0xC0000001); } - +#ifdef CONFIG_X86_32 /* Cyrix III family needs CX8 & PGE explicitly enabled. */ if (c->x86_model >= 6 && c->x86_model <= 9) { rdmsr(MSR_VIA_FCR, lo, hi); @@ -288,6 +288,11 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c) /* Before Nehemiah, the C3's had 3dNOW! */ if (c->x86_model >= 6 && c->x86_model < 9) set_cpu_cap(c, X86_FEATURE_3DNOW); +#endif + if (c->x86 == 0x6 && c->x86_model >= 0xf) { + c->x86_cache_alignment = c->x86_clflush_size * 2; + set_cpu_cap(c, X86_FEATURE_REP_GOOD); + } display_cacheinfo(c); } @@ -316,16 +321,25 @@ enum { static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c) { switch (c->x86) { +#ifdef CONFIG_X86_32 case 5: /* Emulate MTRRs using Centaur's MCR. */ set_cpu_cap(c, X86_FEATURE_CENTAUR_MCR); break; +#endif + case 6: + if (c->x86_model >= 0xf) + set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); + break; } +#ifdef CONFIG_X86_64 + set_cpu_cap(c, X86_FEATURE_SYSENTER32); +#endif } static void __cpuinit init_centaur(struct cpuinfo_x86 *c) { - +#ifdef CONFIG_X86_32 char *name; u32 fcr_set = 0; u32 fcr_clr = 0; @@ -337,8 +351,10 @@ static void __cpuinit init_centaur(struct cpuinfo_x86 *c) * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */ clear_cpu_cap(c, 0*32+31); - +#endif + early_init_centaur(c); switch (c->x86) { +#ifdef CONFIG_X86_32 case 5: switch (c->x86_model) { case 4: @@ -442,16 +458,20 @@ static void __cpuinit init_centaur(struct cpuinfo_x86 *c) } sprintf(c->x86_model_id, "WinChip %s", name); break; - +#endif case 6: init_c3(c); break; } +#ifdef CONFIG_X86_64 + set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); +#endif } static unsigned int __cpuinit centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size) { +#ifdef CONFIG_X86_32 /* VIA C3 CPUs (670-68F) need further shifting. */ if ((c->x86 == 6) && ((c->x86_model == 7) || (c->x86_model == 8))) size >>= 8; @@ -464,11 +484,11 @@ centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size) if ((c->x86 == 6) && (c->x86_model == 9) && (c->x86_mask == 1) && (size == 65)) size -= 1; - +#endif return size; } -static struct cpu_dev centaur_cpu_dev __cpuinitdata = { +static const struct cpu_dev __cpuinitconst centaur_cpu_dev = { .c_vendor = "Centaur", .c_ident = { "CentaurHauls" }, .c_early_init = early_init_centaur, diff --git a/arch/x86/kernel/cpu/centaur_64.c b/arch/x86/kernel/cpu/centaur_64.c deleted file mode 100644 index a1625f5a1e7..00000000000 --- a/arch/x86/kernel/cpu/centaur_64.c +++ /dev/null @@ -1,37 +0,0 @@ -#include <linux/init.h> -#include <linux/smp.h> - -#include <asm/cpufeature.h> -#include <asm/processor.h> - -#include "cpu.h" - -static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c) -{ - if (c->x86 == 0x6 && c->x86_model >= 0xf) - set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); - - set_cpu_cap(c, X86_FEATURE_SYSENTER32); -} - -static void __cpuinit init_centaur(struct cpuinfo_x86 *c) -{ - early_init_centaur(c); - - if (c->x86 == 0x6 && c->x86_model >= 0xf) { - c->x86_cache_alignment = c->x86_clflush_size * 2; - set_cpu_cap(c, X86_FEATURE_REP_GOOD); - } - set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); -} - -static struct cpu_dev centaur_cpu_dev __cpuinitdata = { - .c_vendor = "Centaur", - .c_ident = { "CentaurHauls" }, - .c_early_init = early_init_centaur, - .c_init = init_centaur, - .c_x86_vendor = X86_VENDOR_CENTAUR, -}; - -cpu_dev_register(centaur_cpu_dev); - diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 83492b1f93b..e2962cc1e27 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1,118 +1,129 @@ -#include <linux/init.h> -#include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/string.h> #include <linux/bootmem.h> +#include <linux/linkage.h> #include <linux/bitops.h> +#include <linux/kernel.h> #include <linux/module.h> -#include <linux/kgdb.h> -#include <linux/topology.h> +#include <linux/percpu.h> +#include <linux/string.h> #include <linux/delay.h> +#include <linux/sched.h> +#include <linux/init.h> +#include <linux/kgdb.h> #include <linux/smp.h> -#include <linux/percpu.h> -#include <asm/i387.h> -#include <asm/msr.h> -#include <asm/io.h> -#include <asm/linkage.h> +#include <linux/io.h> + +#include <asm/stackprotector.h> #include <asm/mmu_context.h> +#include <asm/hypervisor.h> +#include <asm/processor.h> +#include <asm/sections.h> +#include <asm/topology.h> +#include <asm/cpumask.h> +#include <asm/pgtable.h> +#include <asm/atomic.h> +#include <asm/proto.h> +#include <asm/setup.h> +#include <asm/apic.h> +#include <asm/desc.h> +#include <asm/i387.h> #include <asm/mtrr.h> +#include <asm/numa.h> +#include <asm/asm.h> +#include <asm/cpu.h> #include <asm/mce.h> +#include <asm/msr.h> #include <asm/pat.h> -#include <asm/asm.h> -#include <asm/numa.h> #include <asm/smp.h> + #ifdef CONFIG_X86_LOCAL_APIC -#include <asm/mpspec.h> -#include <asm/apic.h> -#include <mach_apic.h> -#include <asm/genapic.h> +#include <asm/uv/uv.h> #endif -#include <asm/pda.h> -#include <asm/pgtable.h> -#include <asm/processor.h> -#include <asm/desc.h> -#include <asm/atomic.h> -#include <asm/proto.h> -#include <asm/sections.h> -#include <asm/setup.h> -#include <asm/hypervisor.h> - #include "cpu.h" #ifdef CONFIG_X86_64 /* all of these masks are initialized in setup_cpu_local_masks() */ -cpumask_var_t cpu_callin_mask; -cpumask_var_t cpu_callout_mask; cpumask_var_t cpu_initialized_mask; +cpumask_var_t cpu_callout_mask; +cpumask_var_t cpu_callin_mask; /* representing cpus for which sibling maps can be computed */ cpumask_var_t cpu_sibling_setup_mask; +/* correctly size the local cpu masks */ +void __init setup_cpu_local_masks(void) +{ + alloc_bootmem_cpumask_var(&cpu_initialized_mask); + alloc_bootmem_cpumask_var(&cpu_callin_mask); + alloc_bootmem_cpumask_var(&cpu_callout_mask); + alloc_bootmem_cpumask_var(&cpu_sibling_setup_mask); +} + #else /* CONFIG_X86_32 */ -cpumask_t cpu_callin_map; +cpumask_t cpu_sibling_setup_map; cpumask_t cpu_callout_map; cpumask_t cpu_initialized; -cpumask_t cpu_sibling_setup_map; +cpumask_t cpu_callin_map; #endif /* CONFIG_X86_32 */ -static struct cpu_dev *this_cpu __cpuinitdata; +static const struct cpu_dev *this_cpu __cpuinitdata; +DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { #ifdef CONFIG_X86_64 -/* We need valid kernel segments for data and code in long mode too - * IRET will check the segment types kkeil 2000/10/28 - * Also sysret mandates a special GDT layout - */ -/* The TLS descriptors are currently at a different place compared to i386. - Hopefully nobody expects them at a fixed place (Wine?) */ -DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = { - [GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } }, - [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } }, - [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } }, - [GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } }, - [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } }, - [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } }, -} }; + /* + * We need valid kernel segments for data and code in long mode too + * IRET will check the segment types kkeil 2000/10/28 + * Also sysret mandates a special GDT layout + * + * TLS descriptors are currently at a different place compared to i386. + * Hopefully nobody expects them at a fixed place (Wine?) + */ + [GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } }, + [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } }, + [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } }, + [GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } }, + [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } }, + [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } }, #else -DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { - [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00cf9a00 } } }, - [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9200 } } }, - [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00cffa00 } } }, - [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff200 } } }, + [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00cf9a00 } } }, + [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9200 } } }, + [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00cffa00 } } }, + [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff200 } } }, /* * Segments used for calling PnP BIOS have byte granularity. * They code segments and data segments have fixed 64k limits, * the transfer segment sizes are set at run time. */ /* 32-bit code */ - [GDT_ENTRY_PNPBIOS_CS32] = { { { 0x0000ffff, 0x00409a00 } } }, + [GDT_ENTRY_PNPBIOS_CS32] = { { { 0x0000ffff, 0x00409a00 } } }, /* 16-bit code */ - [GDT_ENTRY_PNPBIOS_CS16] = { { { 0x0000ffff, 0x00009a00 } } }, + [GDT_ENTRY_PNPBIOS_CS16] = { { { 0x0000ffff, 0x00009a00 } } }, /* 16-bit data */ - [GDT_ENTRY_PNPBIOS_DS] = { { { 0x0000ffff, 0x00009200 } } }, + [GDT_ENTRY_PNPBIOS_DS] = { { { 0x0000ffff, 0x00009200 } } }, /* 16-bit data */ - [GDT_ENTRY_PNPBIOS_TS1] = { { { 0x00000000, 0x00009200 } } }, + [GDT_ENTRY_PNPBIOS_TS1] = { { { 0x00000000, 0x00009200 } } }, /* 16-bit data */ - [GDT_ENTRY_PNPBIOS_TS2] = { { { 0x00000000, 0x00009200 } } }, + [GDT_ENTRY_PNPBIOS_TS2] = { { { 0x00000000, 0x00009200 } } }, /* * The APM segments have byte granularity and their bases * are set at run time. All have 64k limits. */ /* 32-bit code */ - [GDT_ENTRY_APMBIOS_BASE] = { { { 0x0000ffff, 0x00409a00 } } }, + [GDT_ENTRY_APMBIOS_BASE] = { { { 0x0000ffff, 0x00409a00 } } }, /* 16-bit code */ - [GDT_ENTRY_APMBIOS_BASE+1] = { { { 0x0000ffff, 0x00009a00 } } }, + [GDT_ENTRY_APMBIOS_BASE+1] = { { { 0x0000ffff, 0x00009a00 } } }, /* data */ - [GDT_ENTRY_APMBIOS_BASE+2] = { { { 0x0000ffff, 0x00409200 } } }, + [GDT_ENTRY_APMBIOS_BASE+2] = { { { 0x0000ffff, 0x00409200 } } }, - [GDT_ENTRY_ESPFIX_SS] = { { { 0x00000000, 0x00c09200 } } }, - [GDT_ENTRY_PERCPU] = { { { 0x00000000, 0x00000000 } } }, -} }; + [GDT_ENTRY_ESPFIX_SS] = { { { 0x00000000, 0x00c09200 } } }, + [GDT_ENTRY_PERCPU] = { { { 0x0000ffff, 0x00cf9200 } } }, + GDT_STACK_CANARY_INIT #endif +} }; EXPORT_PER_CPU_SYMBOL_GPL(gdt_page); #ifdef CONFIG_X86_32 @@ -153,16 +164,17 @@ static inline int flag_is_changeable_p(u32 flag) * the CPUID. Add "volatile" to not allow gcc to * optimize the subsequent calls to this function. */ - asm volatile ("pushfl\n\t" - "pushfl\n\t" - "popl %0\n\t" - "movl %0,%1\n\t" - "xorl %2,%0\n\t" - "pushl %0\n\t" - "popfl\n\t" - "pushfl\n\t" - "popl %0\n\t" - "popfl\n\t" + asm volatile ("pushfl \n\t" + "pushfl \n\t" + "popl %0 \n\t" + "movl %0, %1 \n\t" + "xorl %2, %0 \n\t" + "pushl %0 \n\t" + "popfl \n\t" + "pushfl \n\t" + "popl %0 \n\t" + "popfl \n\t" + : "=&r" (f1), "=&r" (f2) : "ir" (flag)); @@ -177,18 +189,22 @@ static int __cpuinit have_cpuid_p(void) static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c) { - if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr) { - /* Disable processor serial number */ - unsigned long lo, hi; - rdmsr(MSR_IA32_BBL_CR_CTL, lo, hi); - lo |= 0x200000; - wrmsr(MSR_IA32_BBL_CR_CTL, lo, hi); - printk(KERN_NOTICE "CPU serial number disabled.\n"); - clear_cpu_cap(c, X86_FEATURE_PN); - - /* Disabling the serial number may affect the cpuid level */ - c->cpuid_level = cpuid_eax(0); - } + unsigned long lo, hi; + + if (!cpu_has(c, X86_FEATURE_PN) || !disable_x86_serial_nr) + return; + + /* Disable processor serial number: */ + + rdmsr(MSR_IA32_BBL_CR_CTL, lo, hi); + lo |= 0x200000; + wrmsr(MSR_IA32_BBL_CR_CTL, lo, hi); + + printk(KERN_NOTICE "CPU serial number disabled.\n"); + clear_cpu_cap(c, X86_FEATURE_PN); + + /* Disabling the serial number may affect the cpuid level */ + c->cpuid_level = cpuid_eax(0); } static int __init x86_serial_nr_setup(char *s) @@ -213,16 +229,64 @@ static inline void squash_the_stupid_serial_number(struct cpuinfo_x86 *c) #endif /* + * Some CPU features depend on higher CPUID levels, which may not always + * be available due to CPUID level capping or broken virtualization + * software. Add those features to this table to auto-disable them. + */ +struct cpuid_dependent_feature { + u32 feature; + u32 level; +}; + +static const struct cpuid_dependent_feature __cpuinitconst +cpuid_dependent_features[] = { + { X86_FEATURE_MWAIT, 0x00000005 }, + { X86_FEATURE_DCA, 0x00000009 }, + { X86_FEATURE_XSAVE, 0x0000000d }, + { 0, 0 } +}; + +static void __cpuinit filter_cpuid_features(struct cpuinfo_x86 *c, bool warn) +{ + const struct cpuid_dependent_feature *df; + + for (df = cpuid_dependent_features; df->feature; df++) { + + if (!cpu_has(c, df->feature)) + continue; + /* + * Note: cpuid_level is set to -1 if unavailable, but + * extended_extended_level is set to 0 if unavailable + * and the legitimate extended levels are all negative + * when signed; hence the weird messing around with + * signs here... + */ + if (!((s32)df->level < 0 ? + (u32)df->level > (u32)c->extended_cpuid_level : + (s32)df->level > (s32)c->cpuid_level)) + continue; + + clear_cpu_cap(c, df->feature); + if (!warn) + continue; + + printk(KERN_WARNING + "CPU: CPU feature %s disabled, no CPUID level 0x%x\n", + x86_cap_flags[df->feature], df->level); + } +} + +/* * Naming convention should be: <Name> [(<Codename>)] * This table only is used unless init_<vendor>() below doesn't set it; - * in particular, if CPUID levels 0x80000002..4 are supported, this isn't used - * + * in particular, if CPUID levels 0x80000002..4 are supported, this + * isn't used */ /* Look up CPU names by table lookup. */ -static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c) +static const char *__cpuinit table_lookup_model(struct cpuinfo_x86 *c) { - struct cpu_model_info *info; + const struct cpu_model_info *info; if (c->x86_model >= 16) return NULL; /* Range check */ @@ -242,21 +306,34 @@ static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c) __u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata; -/* Current gdt points %fs at the "master" per-cpu area: after this, - * it's on the real one. */ -void switch_to_new_gdt(void) +void load_percpu_segment(int cpu) +{ +#ifdef CONFIG_X86_32 + loadsegment(fs, __KERNEL_PERCPU); +#else + loadsegment(gs, 0); + wrmsrl(MSR_GS_BASE, (unsigned long)per_cpu(irq_stack_union.gs_base, cpu)); +#endif + load_stack_canary_segment(); +} + +/* + * Current gdt points %fs at the "master" per-cpu area: after this, + * it's on the real one. + */ +void switch_to_new_gdt(int cpu) { struct desc_ptr gdt_descr; - gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id()); + gdt_descr.address = (long)get_cpu_gdt_table(cpu); gdt_descr.size = GDT_SIZE - 1; load_gdt(&gdt_descr); -#ifdef CONFIG_X86_32 - asm("mov %0, %%fs" : : "r" (__KERNEL_PERCPU) : "memory"); -#endif + /* Reload the per-cpu base */ + + load_percpu_segment(cpu); } -static struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {}; +static const struct cpu_dev *__cpuinitdata cpu_devs[X86_VENDOR_NUM] = {}; static void __cpuinit default_init(struct cpuinfo_x86 *c) { @@ -275,7 +352,7 @@ static void __cpuinit default_init(struct cpuinfo_x86 *c) #endif } -static struct cpu_dev __cpuinitdata default_cpu = { +static const struct cpu_dev __cpuinitconst default_cpu = { .c_init = default_init, .c_vendor = "Unknown", .c_x86_vendor = X86_VENDOR_UNKNOWN, @@ -289,22 +366,24 @@ static void __cpuinit get_model_name(struct cpuinfo_x86 *c) if (c->extended_cpuid_level < 0x80000004) return; - v = (unsigned int *) c->x86_model_id; + v = (unsigned int *)c->x86_model_id; cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]); cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]); cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]); c->x86_model_id[48] = 0; - /* Intel chips right-justify this string for some dumb reason; - undo that brain damage */ + /* + * Intel chips right-justify this string for some dumb reason; + * undo that brain damage: + */ p = q = &c->x86_model_id[0]; while (*p == ' ') - p++; + p++; if (p != q) { - while (*p) - *q++ = *p++; - while (q <= &c->x86_model_id[48]) - *q++ = '\0'; /* Zero-pad the rest */ + while (*p) + *q++ = *p++; + while (q <= &c->x86_model_id[48]) + *q++ = '\0'; /* Zero-pad the rest */ } } @@ -373,36 +452,30 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) if (smp_num_siblings == 1) { printk(KERN_INFO "CPU: Hyper-Threading is disabled\n"); - } else if (smp_num_siblings > 1) { + goto out; + } - if (smp_num_siblings > nr_cpu_ids) { - printk(KERN_WARNING "CPU: Unsupported number of siblings %d", - smp_num_siblings); - smp_num_siblings = 1; - return; - } + if (smp_num_siblings <= 1) + goto out; - index_msb = get_count_order(smp_num_siblings); -#ifdef CONFIG_X86_64 - c->phys_proc_id = phys_pkg_id(index_msb); -#else - c->phys_proc_id = phys_pkg_id(c->initial_apicid, index_msb); -#endif + if (smp_num_siblings > nr_cpu_ids) { + pr_warning("CPU: Unsupported number of siblings %d", + smp_num_siblings); + smp_num_siblings = 1; + return; + } - smp_num_siblings = smp_num_siblings / c->x86_max_cores; + index_msb = get_count_order(smp_num_siblings); + c->phys_proc_id = apic->phys_pkg_id(c->initial_apicid, index_msb); - index_msb = get_count_order(smp_num_siblings); + smp_num_siblings = smp_num_siblings / c->x86_max_cores; - core_bits = get_count_order(c->x86_max_cores); + index_msb = get_count_order(smp_num_siblings); -#ifdef CONFIG_X86_64 - c->cpu_core_id = phys_pkg_id(index_msb) & - ((1 << core_bits) - 1); -#else - c->cpu_core_id = phys_pkg_id(c->initial_apicid, index_msb) & - ((1 << core_bits) - 1); -#endif - } + core_bits = get_count_order(c->x86_max_cores); + + c->cpu_core_id = apic->phys_pkg_id(c->initial_apicid, index_msb) & + ((1 << core_bits) - 1); out: if ((c->x86_max_cores * smp_num_siblings) > 1) { @@ -417,8 +490,8 @@ out: static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) { char *v = c->x86_vendor_id; - int i; static int printed; + int i; for (i = 0; i < X86_VENDOR_NUM; i++) { if (!cpu_devs[i]) @@ -427,6 +500,7 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) if (!strcmp(v, cpu_devs[i]->c_ident[0]) || (cpu_devs[i]->c_ident[1] && !strcmp(v, cpu_devs[i]->c_ident[1]))) { + this_cpu = cpu_devs[i]; c->x86_vendor = this_cpu->c_x86_vendor; return; @@ -435,7 +509,9 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) if (!printed) { printed++; - printk(KERN_ERR "CPU: vendor_id '%s' unknown, using generic init.\n", v); + printk(KERN_ERR + "CPU: vendor_id '%s' unknown, using generic init.\n", v); + printk(KERN_ERR "CPU: Your system may be unstable.\n"); } @@ -455,14 +531,17 @@ void __cpuinit cpu_detect(struct cpuinfo_x86 *c) /* Intel-defined flags: level 0x00000001 */ if (c->cpuid_level >= 0x00000001) { u32 junk, tfms, cap0, misc; + cpuid(0x00000001, &tfms, &misc, &junk, &cap0); c->x86 = (tfms >> 8) & 0xf; c->x86_model = (tfms >> 4) & 0xf; c->x86_mask = tfms & 0xf; + if (c->x86 == 0xf) c->x86 += (tfms >> 20) & 0xff; if (c->x86 >= 0x6) c->x86_model += ((tfms >> 16) & 0xf) << 4; + if (cap0 & (1<<19)) { c->x86_clflush_size = ((misc >> 8) & 0xff) * 8; c->x86_cache_alignment = c->x86_clflush_size; @@ -478,6 +557,7 @@ static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) /* Intel-defined flags: level 0x00000001 */ if (c->cpuid_level >= 0x00000001) { u32 capability, excap; + cpuid(0x00000001, &tfms, &ebx, &excap, &capability); c->x86_capability[0] = capability; c->x86_capability[4] = excap; @@ -486,6 +566,7 @@ static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) /* AMD-defined flags: level 0x80000001 */ xlvl = cpuid_eax(0x80000000); c->extended_cpuid_level = xlvl; + if ((xlvl & 0xffff0000) == 0x80000000) { if (xlvl >= 0x80000001) { c->x86_capability[1] = cpuid_edx(0x80000001); @@ -493,13 +574,15 @@ static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) } } -#ifdef CONFIG_X86_64 if (c->extended_cpuid_level >= 0x80000008) { u32 eax = cpuid_eax(0x80000008); c->x86_virt_bits = (eax >> 8) & 0xff; c->x86_phys_bits = eax & 0xff; } +#ifdef CONFIG_X86_32 + else if (cpu_has(c, X86_FEATURE_PAE) || cpu_has(c, X86_FEATURE_PSE36)) + c->x86_phys_bits = 36; #endif if (c->extended_cpuid_level >= 0x80000007) @@ -546,8 +629,12 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) { #ifdef CONFIG_X86_64 c->x86_clflush_size = 64; + c->x86_phys_bits = 36; + c->x86_virt_bits = 48; #else c->x86_clflush_size = 32; + c->x86_phys_bits = 32; + c->x86_virt_bits = 32; #endif c->x86_cache_alignment = c->x86_clflush_size; @@ -570,21 +657,20 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) if (this_cpu->c_early_init) this_cpu->c_early_init(c); - validate_pat_support(c); - #ifdef CONFIG_SMP c->cpu_index = boot_cpu_id; #endif + filter_cpuid_features(c, false); } void __init early_cpu_init(void) { - struct cpu_dev **cdev; + const struct cpu_dev *const *cdev; int count = 0; - printk("KERNEL supported cpus:\n"); + printk(KERN_INFO "KERNEL supported cpus:\n"); for (cdev = __x86_cpu_dev_start; cdev < __x86_cpu_dev_end; cdev++) { - struct cpu_dev *cpudev = *cdev; + const struct cpu_dev *cpudev = *cdev; unsigned int j; if (count >= X86_VENDOR_NUM) @@ -595,7 +681,7 @@ void __init early_cpu_init(void) for (j = 0; j < 2; j++) { if (!cpudev->c_ident[j]) continue; - printk(" %s %s\n", cpudev->c_vendor, + printk(KERN_INFO " %s %s\n", cpudev->c_vendor, cpudev->c_ident[j]); } } @@ -637,7 +723,7 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 *c) c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xFF; #ifdef CONFIG_X86_32 # ifdef CONFIG_X86_HT - c->apicid = phys_pkg_id(c->initial_apicid, 0); + c->apicid = apic->phys_pkg_id(c->initial_apicid, 0); # else c->apicid = c->initial_apicid; # endif @@ -671,9 +757,13 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) c->x86_coreid_bits = 0; #ifdef CONFIG_X86_64 c->x86_clflush_size = 64; + c->x86_phys_bits = 36; + c->x86_virt_bits = 48; #else c->cpuid_level = -1; /* CPUID not detected */ c->x86_clflush_size = 32; + c->x86_phys_bits = 32; + c->x86_virt_bits = 32; #endif c->x86_cache_alignment = c->x86_clflush_size; memset(&c->x86_capability, 0, sizeof c->x86_capability); @@ -684,7 +774,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) this_cpu->c_identify(c); #ifdef CONFIG_X86_64 - c->apicid = phys_pkg_id(0); + c->apicid = apic->phys_pkg_id(c->initial_apicid, 0); #endif /* @@ -704,13 +794,16 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) squash_the_stupid_serial_number(c); /* - * The vendor-specific functions might have changed features. Now - * we do "generic changes." + * The vendor-specific functions might have changed features. + * Now we do "generic changes." */ + /* Filter out anything that depends on CPUID levels we don't have */ + filter_cpuid_features(c, true); + /* If the model name is still unset, do table lookup. */ if (!c->x86_model_id[0]) { - char *p; + const char *p; p = table_lookup_model(c); if (p) strcpy(c->x86_model_id, p); @@ -785,11 +878,11 @@ void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c) } struct msr_range { - unsigned min; - unsigned max; + unsigned min; + unsigned max; }; -static struct msr_range msr_range_array[] __cpuinitdata = { +static const struct msr_range msr_range_array[] __cpuinitconst = { { 0x00000000, 0x00000418}, { 0xc0000000, 0xc000040b}, { 0xc0010000, 0xc0010142}, @@ -798,14 +891,15 @@ static struct msr_range msr_range_array[] __cpuinitdata = { static void __cpuinit print_cpu_msr(void) { + unsigned index_min, index_max; unsigned index; u64 val; int i; - unsigned index_min, index_max; for (i = 0; i < ARRAY_SIZE(msr_range_array); i++) { index_min = msr_range_array[i].min; index_max = msr_range_array[i].max; + for (index = index_min; index < index_max; index++) { if (rdmsrl_amd_safe(index, &val)) continue; @@ -815,6 +909,7 @@ static void __cpuinit print_cpu_msr(void) } static int show_msr __cpuinitdata; + static __init int setup_show_msr(char *arg) { int num; @@ -836,12 +931,14 @@ __setup("noclflush", setup_noclflush); void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) { - char *vendor = NULL; + const char *vendor = NULL; - if (c->x86_vendor < X86_VENDOR_NUM) + if (c->x86_vendor < X86_VENDOR_NUM) { vendor = this_cpu->c_vendor; - else if (c->cpuid_level >= 0) - vendor = c->x86_vendor_id; + } else { + if (c->cpuid_level >= 0) + vendor = c->x86_vendor_id; + } if (vendor && !strstr(c->x86_model_id, vendor)) printk(KERN_CONT "%s ", vendor); @@ -868,65 +965,45 @@ void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) static __init int setup_disablecpuid(char *arg) { int bit; + if (get_option(&arg, &bit) && bit < NCAPINTS*32) setup_clear_cpu_cap(bit); else return 0; + return 1; } __setup("clearcpuid=", setup_disablecpuid); #ifdef CONFIG_X86_64 -struct x8664_pda **_cpu_pda __read_mostly; -EXPORT_SYMBOL(_cpu_pda); - struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table }; -static char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss; +DEFINE_PER_CPU_FIRST(union irq_stack_union, + irq_stack_union) __aligned(PAGE_SIZE); -void __cpuinit pda_init(int cpu) -{ - struct x8664_pda *pda = cpu_pda(cpu); +DEFINE_PER_CPU(char *, irq_stack_ptr) = + init_per_cpu_var(irq_stack_union.irq_stack) + IRQ_STACK_SIZE - 64; - /* Setup up data that may be needed in __get_free_pages early */ - loadsegment(fs, 0); - loadsegment(gs, 0); - /* Memory clobbers used to order PDA accessed */ - mb(); - wrmsrl(MSR_GS_BASE, pda); - mb(); - - pda->cpunumber = cpu; - pda->irqcount = -1; - pda->kernelstack = (unsigned long)stack_thread_info() - - PDA_STACKOFFSET + THREAD_SIZE; - pda->active_mm = &init_mm; - pda->mmu_state = 0; - - if (cpu == 0) { - /* others are initialized in smpboot.c */ - pda->pcurrent = &init_task; - pda->irqstackptr = boot_cpu_stack; - pda->irqstackptr += IRQSTACKSIZE - 64; - } else { - if (!pda->irqstackptr) { - pda->irqstackptr = (char *) - __get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER); - if (!pda->irqstackptr) - panic("cannot allocate irqstack for cpu %d", - cpu); - pda->irqstackptr += IRQSTACKSIZE - 64; - } +DEFINE_PER_CPU(unsigned long, kernel_stack) = + (unsigned long)&init_thread_union - KERNEL_STACK_OFFSET + THREAD_SIZE; +EXPORT_PER_CPU_SYMBOL(kernel_stack); - if (pda->nodenumber == 0 && cpu_to_node(cpu) != NUMA_NO_NODE) - pda->nodenumber = cpu_to_node(cpu); - } -} +DEFINE_PER_CPU(unsigned int, irq_count) = -1; -static char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + - DEBUG_STKSZ] __page_aligned_bss; +/* + * Special IST stacks which the CPU switches to when it calls + * an IST-marked descriptor entry. Up to 7 stacks (hardware + * limit), all of them are 4K, except the debug stack which + * is 8K. + */ +static const unsigned int exception_stack_sizes[N_EXCEPTION_STACKS] = { + [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STKSZ, + [DEBUG_STACK - 1] = DEBUG_STKSZ +}; -extern asmlinkage void ignore_sysret(void); +static DEFINE_PER_CPU_PAGE_ALIGNED(char, exception_stacks + [(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ]) + __aligned(PAGE_SIZE); /* May not be marked __init: used by software suspend */ void syscall_init(void) @@ -957,16 +1034,38 @@ unsigned long kernel_eflags; */ DEFINE_PER_CPU(struct orig_ist, orig_ist); -#else +#else /* CONFIG_X86_64 */ + +#ifdef CONFIG_CC_STACKPROTECTOR +DEFINE_PER_CPU(unsigned long, stack_canary); +#endif -/* Make sure %fs is initialized properly in idle threads */ +/* Make sure %fs and %gs are initialized properly in idle threads */ struct pt_regs * __cpuinit idle_regs(struct pt_regs *regs) { memset(regs, 0, sizeof(struct pt_regs)); regs->fs = __KERNEL_PERCPU; + regs->gs = __KERNEL_STACK_CANARY; + return regs; } -#endif +#endif /* CONFIG_X86_64 */ + +/* + * Clear all 6 debug registers: + */ +static void clear_all_debug_regs(void) +{ + int i; + + for (i = 0; i < 8; i++) { + /* Ignore db4, db5 */ + if ((i == 4) || (i == 5)) + continue; + + set_debugreg(0, i); + } +} /* * cpu_init() initializes state that is per-CPU. Some data is already @@ -976,21 +1075,25 @@ struct pt_regs * __cpuinit idle_regs(struct pt_regs *regs) * A lot of state is already set up in PDA init for 64 bit */ #ifdef CONFIG_X86_64 + void __cpuinit cpu_init(void) { - int cpu = stack_smp_processor_id(); - struct tss_struct *t = &per_cpu(init_tss, cpu); - struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu); - unsigned long v; - char *estacks = NULL; + struct orig_ist *orig_ist; struct task_struct *me; + struct tss_struct *t; + unsigned long v; + int cpu; int i; - /* CPU 0 is initialised in head64.c */ - if (cpu != 0) - pda_init(cpu); - else - estacks = boot_exception_stacks; + cpu = stack_smp_processor_id(); + t = &per_cpu(init_tss, cpu); + orig_ist = &per_cpu(orig_ist, cpu); + +#ifdef CONFIG_NUMA + if (cpu != 0 && percpu_read(node_number) == 0 && + cpu_to_node(cpu) != NUMA_NO_NODE) + percpu_write(node_number, cpu_to_node(cpu)); +#endif me = current; @@ -1006,7 +1109,9 @@ void __cpuinit cpu_init(void) * and set up the GDT descriptor: */ - switch_to_new_gdt(); + switch_to_new_gdt(cpu); + loadsegment(fs, 0); + load_idt((const struct desc_ptr *)&idt_descr); memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8); @@ -1017,31 +1122,24 @@ void __cpuinit cpu_init(void) barrier(); check_efer(); - if (cpu != 0 && x2apic) + if (cpu != 0) enable_x2apic(); /* * set up and load the per-CPU TSS */ if (!orig_ist->ist[0]) { - static const unsigned int order[N_EXCEPTION_STACKS] = { - [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER, - [DEBUG_STACK - 1] = DEBUG_STACK_ORDER - }; + char *estacks = per_cpu(exception_stacks, cpu); + for (v = 0; v < N_EXCEPTION_STACKS; v++) { - if (cpu) { - estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]); - if (!estacks) - panic("Cannot allocate exception " - "stack %ld %d\n", v, cpu); - } - estacks += PAGE_SIZE << order[v]; + estacks += exception_stack_sizes[v]; orig_ist->ist[v] = t->x86_tss.ist[v] = (unsigned long)estacks; } } t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap); + /* * <= is required because the CPU will access up to * 8 bits beyond the end of the IO permission bitmap. @@ -1051,8 +1149,7 @@ void __cpuinit cpu_init(void) atomic_inc(&init_mm.mm_count); me->active_mm = &init_mm; - if (me->mm) - BUG(); + BUG_ON(me->mm); enter_lazy_tlb(&init_mm, me); load_sp0(t, ¤t->thread); @@ -1069,22 +1166,9 @@ void __cpuinit cpu_init(void) */ if (kgdb_connected && arch_kgdb_ops.correct_hw_break) arch_kgdb_ops.correct_hw_break(); - else { -#endif - /* - * Clear all 6 debug registers: - */ - - set_debugreg(0UL, 0); - set_debugreg(0UL, 1); - set_debugreg(0UL, 2); - set_debugreg(0UL, 3); - set_debugreg(0UL, 6); - set_debugreg(0UL, 7); -#ifdef CONFIG_KGDB - /* If the kgdb is connected no debug regs should be altered. */ - } + else #endif + clear_all_debug_regs(); fpu_init(); @@ -1105,7 +1189,8 @@ void __cpuinit cpu_init(void) if (cpumask_test_and_set_cpu(cpu, cpu_initialized_mask)) { printk(KERN_WARNING "CPU#%d already initialized!\n", cpu); - for (;;) local_irq_enable(); + for (;;) + local_irq_enable(); } printk(KERN_INFO "Initializing CPU#%d\n", cpu); @@ -1114,15 +1199,14 @@ void __cpuinit cpu_init(void) clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE); load_idt(&idt_descr); - switch_to_new_gdt(); + switch_to_new_gdt(cpu); /* * Set up and load the per-CPU TSS and LDT */ atomic_inc(&init_mm.mm_count); curr->active_mm = &init_mm; - if (curr->mm) - BUG(); + BUG_ON(curr->mm); enter_lazy_tlb(&init_mm, curr); load_sp0(t, thread); @@ -1135,16 +1219,7 @@ void __cpuinit cpu_init(void) __set_tss_desc(cpu, GDT_ENTRY_DOUBLEFAULT_TSS, &doublefault_tss); #endif - /* Clear %gs. */ - asm volatile ("mov %0, %%gs" : : "r" (0)); - - /* Clear all 6 debug registers: */ - set_debugreg(0, 0); - set_debugreg(0, 1); - set_debugreg(0, 2); - set_debugreg(0, 3); - set_debugreg(0, 6); - set_debugreg(0, 7); + clear_all_debug_regs(); /* * Force FPU initialization: @@ -1164,6 +1239,4 @@ void __cpuinit cpu_init(void) xsave_init(); } - - #endif diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h index de4094a3921..9469ecb5aeb 100644 --- a/arch/x86/kernel/cpu/cpu.h +++ b/arch/x86/kernel/cpu/cpu.h @@ -5,15 +5,15 @@ struct cpu_model_info { int vendor; int family; - char *model_names[16]; + const char *model_names[16]; }; /* attempt to consolidate cpu attributes */ struct cpu_dev { - char * c_vendor; + const char * c_vendor; /* some have two possibilities for cpuid string */ - char * c_ident[2]; + const char * c_ident[2]; struct cpu_model_info c_models[4]; @@ -25,11 +25,12 @@ struct cpu_dev { }; #define cpu_dev_register(cpu_devX) \ - static struct cpu_dev *__cpu_dev_##cpu_devX __used \ + static const struct cpu_dev *const __cpu_dev_##cpu_devX __used \ __attribute__((__section__(".x86_cpu_dev.init"))) = \ &cpu_devX; -extern struct cpu_dev *__x86_cpu_dev_start[], *__x86_cpu_dev_end[]; +extern const struct cpu_dev *const __x86_cpu_dev_start[], + *const __x86_cpu_dev_end[]; extern void display_cacheinfo(struct cpuinfo_x86 *c); diff --git a/arch/x86/kernel/cpu/cpu_debug.c b/arch/x86/kernel/cpu/cpu_debug.c new file mode 100755 index 00000000000..46e29ab96c6 --- /dev/null +++ b/arch/x86/kernel/cpu/cpu_debug.c @@ -0,0 +1,901 @@ +/* + * 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_arr[CPU_REG_ALL_BIT]); +static DEFINE_PER_CPU(struct cpu_private *, priv_arr[MAX_CPU_FILES]); +static DEFINE_PER_CPU(unsigned, cpu_modelflag); +static DEFINE_PER_CPU(int, cpu_priv_count); +static DEFINE_PER_CPU(unsigned, cpu_model); + +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 }, +}; + +/* Intel Registers Range */ +static struct cpu_debug_range cpu_intel_range[] = { + { 0x00000000, 0x00000001, CPU_MC, CPU_INTEL_ALL }, + { 0x00000006, 0x00000007, CPU_MONITOR, CPU_CX_AT_XE }, + { 0x00000010, 0x00000010, CPU_TIME, CPU_INTEL_ALL }, + { 0x00000011, 0x00000013, CPU_PMC, CPU_INTEL_PENTIUM }, + { 0x00000017, 0x00000017, CPU_PLATFORM, CPU_PX_CX_AT_XE }, + { 0x0000001B, 0x0000001B, CPU_APIC, CPU_P6_CX_AT_XE }, + + { 0x0000002A, 0x0000002A, CPU_POWERON, CPU_PX_CX_AT_XE }, + { 0x0000002B, 0x0000002B, CPU_POWERON, CPU_INTEL_XEON }, + { 0x0000002C, 0x0000002C, CPU_FREQ, CPU_INTEL_XEON }, + { 0x0000003A, 0x0000003A, CPU_CONTROL, CPU_CX_AT_XE }, + + { 0x00000040, 0x00000043, CPU_LBRANCH, CPU_PM_CX_AT_XE }, + { 0x00000044, 0x00000047, CPU_LBRANCH, CPU_PM_CO_AT }, + { 0x00000060, 0x00000063, CPU_LBRANCH, CPU_C2_AT }, + { 0x00000064, 0x00000067, CPU_LBRANCH, CPU_INTEL_ATOM }, + + { 0x00000079, 0x00000079, CPU_BIOS, CPU_P6_CX_AT_XE }, + { 0x00000088, 0x0000008A, CPU_CACHE, CPU_INTEL_P6 }, + { 0x0000008B, 0x0000008B, CPU_BIOS, CPU_P6_CX_AT_XE }, + { 0x0000009B, 0x0000009B, CPU_MONITOR, CPU_INTEL_XEON }, + + { 0x000000C1, 0x000000C2, CPU_PMC, CPU_P6_CX_AT }, + { 0x000000CD, 0x000000CD, CPU_FREQ, CPU_CX_AT }, + { 0x000000E7, 0x000000E8, CPU_PERF, CPU_CX_AT }, + { 0x000000FE, 0x000000FE, CPU_MTRR, CPU_P6_CX_XE }, + + { 0x00000116, 0x00000116, CPU_CACHE, CPU_INTEL_P6 }, + { 0x00000118, 0x00000118, CPU_CACHE, CPU_INTEL_P6 }, + { 0x00000119, 0x00000119, CPU_CACHE, CPU_INTEL_PX }, + { 0x0000011A, 0x0000011B, CPU_CACHE, CPU_INTEL_P6 }, + { 0x0000011E, 0x0000011E, CPU_CACHE, CPU_PX_CX_AT }, + + { 0x00000174, 0x00000176, CPU_SYSENTER, CPU_P6_CX_AT_XE }, + { 0x00000179, 0x0000017A, CPU_MC, CPU_PX_CX_AT_XE }, + { 0x0000017B, 0x0000017B, CPU_MC, CPU_P6_XE }, + { 0x00000186, 0x00000187, CPU_PMC, CPU_P6_CX_AT }, + { 0x00000198, 0x00000199, CPU_PERF, CPU_PM_CX_AT_XE }, + { 0x0000019A, 0x0000019A, CPU_TIME, CPU_PM_CX_AT_XE }, + { 0x0000019B, 0x0000019D, CPU_THERM, CPU_PM_CX_AT_XE }, + { 0x000001A0, 0x000001A0, CPU_MISC, CPU_PM_CX_AT_XE }, + + { 0x000001C9, 0x000001C9, CPU_LBRANCH, CPU_PM_CX_AT }, + { 0x000001D7, 0x000001D8, CPU_LBRANCH, CPU_INTEL_XEON }, + { 0x000001D9, 0x000001D9, CPU_DEBUG, CPU_CX_AT_XE }, + { 0x000001DA, 0x000001DA, CPU_LBRANCH, CPU_INTEL_XEON }, + { 0x000001DB, 0x000001DB, CPU_LBRANCH, CPU_P6_XE }, + { 0x000001DC, 0x000001DC, CPU_LBRANCH, CPU_INTEL_P6 }, + { 0x000001DD, 0x000001DE, CPU_LBRANCH, CPU_PX_CX_AT_XE }, + { 0x000001E0, 0x000001E0, CPU_LBRANCH, CPU_INTEL_P6 }, + + { 0x00000200, 0x0000020F, CPU_MTRR, CPU_P6_CX_XE }, + { 0x00000250, 0x00000250, CPU_MTRR, CPU_P6_CX_XE }, + { 0x00000258, 0x00000259, CPU_MTRR, CPU_P6_CX_XE }, + { 0x00000268, 0x0000026F, CPU_MTRR, CPU_P6_CX_XE }, + { 0x00000277, 0x00000277, CPU_PAT, CPU_C2_AT_XE }, + { 0x000002FF, 0x000002FF, CPU_MTRR, CPU_P6_CX_XE }, + + { 0x00000300, 0x00000308, CPU_PMC, CPU_INTEL_XEON }, + { 0x00000309, 0x0000030B, CPU_PMC, CPU_C2_AT_XE }, + { 0x0000030C, 0x00000311, CPU_PMC, CPU_INTEL_XEON }, + { 0x00000345, 0x00000345, CPU_PMC, CPU_C2_AT }, + { 0x00000360, 0x00000371, CPU_PMC, CPU_INTEL_XEON }, + { 0x0000038D, 0x00000390, CPU_PMC, CPU_C2_AT }, + { 0x000003A0, 0x000003BE, CPU_PMC, CPU_INTEL_XEON }, + { 0x000003C0, 0x000003CD, CPU_PMC, CPU_INTEL_XEON }, + { 0x000003E0, 0x000003E1, CPU_PMC, CPU_INTEL_XEON }, + { 0x000003F0, 0x000003F0, CPU_PMC, CPU_INTEL_XEON }, + { 0x000003F1, 0x000003F1, CPU_PMC, CPU_C2_AT_XE }, + { 0x000003F2, 0x000003F2, CPU_PMC, CPU_INTEL_XEON }, + + { 0x00000400, 0x00000402, CPU_MC, CPU_PM_CX_AT_XE }, + { 0x00000403, 0x00000403, CPU_MC, CPU_INTEL_XEON }, + { 0x00000404, 0x00000406, CPU_MC, CPU_PM_CX_AT_XE }, + { 0x00000407, 0x00000407, CPU_MC, CPU_INTEL_XEON }, + { 0x00000408, 0x0000040A, CPU_MC, CPU_PM_CX_AT_XE }, + { 0x0000040B, 0x0000040B, CPU_MC, CPU_INTEL_XEON }, + { 0x0000040C, 0x0000040E, CPU_MC, CPU_PM_CX_XE }, + { 0x0000040F, 0x0000040F, CPU_MC, CPU_INTEL_XEON }, + { 0x00000410, 0x00000412, CPU_MC, CPU_PM_CX_AT_XE }, + { 0x00000413, 0x00000417, CPU_MC, CPU_CX_AT_XE }, + { 0x00000480, 0x0000048B, CPU_VMX, CPU_CX_AT_XE }, + + { 0x00000600, 0x00000600, CPU_DEBUG, CPU_PM_CX_AT_XE }, + { 0x00000680, 0x0000068F, CPU_LBRANCH, CPU_INTEL_XEON }, + { 0x000006C0, 0x000006CF, CPU_LBRANCH, CPU_INTEL_XEON }, + + { 0x000107CC, 0x000107D3, CPU_PMC, CPU_INTEL_XEON_MP }, + + { 0xC0000080, 0xC0000080, CPU_FEATURES, CPU_INTEL_XEON }, + { 0xC0000081, 0xC0000082, CPU_CALL, CPU_INTEL_XEON }, + { 0xC0000084, 0xC0000084, CPU_CALL, CPU_INTEL_XEON }, + { 0xC0000100, 0xC0000102, CPU_BASE, CPU_INTEL_XEON }, +}; + +/* AMD Registers Range */ +static struct cpu_debug_range cpu_amd_range[] = { + { 0x00000000, 0x00000001, CPU_MC, CPU_K10_PLUS, }, + { 0x00000010, 0x00000010, CPU_TIME, CPU_K8_PLUS, }, + { 0x0000001B, 0x0000001B, CPU_APIC, CPU_K8_PLUS, }, + { 0x0000002A, 0x0000002A, CPU_POWERON, CPU_K7_PLUS }, + { 0x0000008B, 0x0000008B, CPU_VER, CPU_K8_PLUS }, + { 0x000000FE, 0x000000FE, CPU_MTRR, CPU_K8_PLUS, }, + + { 0x00000174, 0x00000176, CPU_SYSENTER, CPU_K8_PLUS, }, + { 0x00000179, 0x0000017B, CPU_MC, CPU_K8_PLUS, }, + { 0x000001D9, 0x000001D9, CPU_DEBUG, CPU_K8_PLUS, }, + { 0x000001DB, 0x000001DE, CPU_LBRANCH, CPU_K8_PLUS, }, + + { 0x00000200, 0x0000020F, CPU_MTRR, CPU_K8_PLUS, }, + { 0x00000250, 0x00000250, CPU_MTRR, CPU_K8_PLUS, }, + { 0x00000258, 0x00000259, CPU_MTRR, CPU_K8_PLUS, }, + { 0x00000268, 0x0000026F, CPU_MTRR, CPU_K8_PLUS, }, + { 0x00000277, 0x00000277, CPU_PAT, CPU_K8_PLUS, }, + { 0x000002FF, 0x000002FF, CPU_MTRR, CPU_K8_PLUS, }, + + { 0x00000400, 0x00000413, CPU_MC, CPU_K8_PLUS, }, + + { 0xC0000080, 0xC0000080, CPU_FEATURES, CPU_AMD_ALL, }, + { 0xC0000081, 0xC0000084, CPU_CALL, CPU_K8_PLUS, }, + { 0xC0000100, 0xC0000102, CPU_BASE, CPU_K8_PLUS, }, + { 0xC0000103, 0xC0000103, CPU_TIME, CPU_K10_PLUS, }, + + { 0xC0010000, 0xC0010007, CPU_PMC, CPU_K8_PLUS, }, + { 0xC0010010, 0xC0010010, CPU_CONF, CPU_K7_PLUS, }, + { 0xC0010015, 0xC0010015, CPU_CONF, CPU_K7_PLUS, }, + { 0xC0010016, 0xC001001A, CPU_MTRR, CPU_K8_PLUS, }, + { 0xC001001D, 0xC001001D, CPU_MTRR, CPU_K8_PLUS, }, + { 0xC001001F, 0xC001001F, CPU_CONF, CPU_K8_PLUS, }, + { 0xC0010030, 0xC0010035, CPU_BIOS, CPU_K8_PLUS, }, + { 0xC0010044, 0xC0010048, CPU_MC, CPU_K8_PLUS, }, + { 0xC0010050, 0xC0010056, CPU_SMM, CPU_K0F_PLUS, }, + { 0xC0010058, 0xC0010058, CPU_CONF, CPU_K10_PLUS, }, + { 0xC0010060, 0xC0010060, CPU_CACHE, CPU_AMD_11, }, + { 0xC0010061, 0xC0010068, CPU_SMM, CPU_K10_PLUS, }, + { 0xC0010069, 0xC001006B, CPU_SMM, CPU_AMD_11, }, + { 0xC0010070, 0xC0010071, CPU_SMM, CPU_K10_PLUS, }, + { 0xC0010111, 0xC0010113, CPU_SMM, CPU_K8_PLUS, }, + { 0xC0010114, 0xC0010118, CPU_SVM, CPU_K10_PLUS, }, + { 0xC0010140, 0xC0010141, CPU_OSVM, CPU_K10_PLUS, }, + { 0xC0011022, 0xC0011023, CPU_CONF, CPU_K10_PLUS, }, +}; + + +/* Intel */ +static int get_intel_modelflag(unsigned model) +{ + int flag; + + switch (model) { + case 0x0501: + case 0x0502: + case 0x0504: + flag = CPU_INTEL_PENTIUM; + break; + case 0x0601: + case 0x0603: + case 0x0605: + case 0x0607: + case 0x0608: + case 0x060A: + case 0x060B: + flag = CPU_INTEL_P6; + break; + case 0x0609: + case 0x060D: + flag = CPU_INTEL_PENTIUM_M; + break; + case 0x060E: + flag = CPU_INTEL_CORE; + break; + case 0x060F: + case 0x0617: + flag = CPU_INTEL_CORE2; + break; + case 0x061C: + flag = CPU_INTEL_ATOM; + break; + case 0x0F00: + case 0x0F01: + case 0x0F02: + case 0x0F03: + case 0x0F04: + flag = CPU_INTEL_XEON_P4; + break; + case 0x0F06: + flag = CPU_INTEL_XEON_MP; + break; + default: + flag = CPU_NONE; + break; + } + + return flag; +} + +/* AMD */ +static int get_amd_modelflag(unsigned model) +{ + int flag; + + switch (model >> 8) { + case 0x6: + flag = CPU_AMD_K6; + break; + case 0x7: + flag = CPU_AMD_K7; + break; + case 0x8: + flag = CPU_AMD_K8; + break; + case 0xf: + flag = CPU_AMD_0F; + break; + case 0x10: + flag = CPU_AMD_10; + break; + case 0x11: + flag = CPU_AMD_11; + break; + default: + flag = CPU_NONE; + break; + } + + return flag; +} + +static int get_cpu_modelflag(unsigned cpu) +{ + int flag; + + flag = per_cpu(cpu_model, cpu); + + switch (flag >> 16) { + case X86_VENDOR_INTEL: + flag = get_intel_modelflag(flag); + break; + case X86_VENDOR_AMD: + flag = get_amd_modelflag(flag & 0xffff); + break; + default: + flag = CPU_NONE; + break; + } + + return flag; +} + +static int get_cpu_range_count(unsigned cpu) +{ + int index; + + switch (per_cpu(cpu_model, cpu) >> 16) { + case X86_VENDOR_INTEL: + index = ARRAY_SIZE(cpu_intel_range); + break; + case X86_VENDOR_AMD: + index = ARRAY_SIZE(cpu_amd_range); + break; + default: + index = 0; + break; + } + + return index; +} + +static int is_typeflag_valid(unsigned cpu, unsigned flag) +{ + unsigned vendor, modelflag; + int i, index; + + /* Standard Registers should be always valid */ + if (flag >= CPU_TSS) + return 1; + + modelflag = per_cpu(cpu_modelflag, cpu); + vendor = per_cpu(cpu_model, cpu) >> 16; + index = get_cpu_range_count(cpu); + + for (i = 0; i < index; i++) { + switch (vendor) { + case X86_VENDOR_INTEL: + if ((cpu_intel_range[i].model & modelflag) && + (cpu_intel_range[i].flag & flag)) + return 1; + break; + case X86_VENDOR_AMD: + if ((cpu_amd_range[i].model & modelflag) && + (cpu_amd_range[i].flag & flag)) + return 1; + break; + } + } + + /* Invalid */ + return 0; +} + +static unsigned get_cpu_range(unsigned cpu, unsigned *min, unsigned *max, + int index, unsigned flag) +{ + unsigned modelflag; + + modelflag = per_cpu(cpu_modelflag, cpu); + *max = 0; + switch (per_cpu(cpu_model, cpu) >> 16) { + case X86_VENDOR_INTEL: + if ((cpu_intel_range[index].model & modelflag) && + (cpu_intel_range[index].flag & flag)) { + *min = cpu_intel_range[index].min; + *max = cpu_intel_range[index].max; + } + break; + case X86_VENDOR_AMD: + if ((cpu_amd_range[index].model & modelflag) && + (cpu_amd_range[index].flag & flag)) { + *min = cpu_amd_range[index].min; + *max = cpu_amd_range[index].max; + } + break; + } + + 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, range; + + 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; + } + } + + range = get_cpu_range_count(cpu); + + for (i = 0; i < 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)); +#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, range, err = 0; + char reg_dir[12]; + u32 low, high; + + range = get_cpu_range_count(cpu); + + for (i = 0; i < 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; + per_cpu(cpu_model, cpu) = ((cpui->x86_vendor << 16) | + (cpui->x86 << 8) | + (cpui->x86_model)); + per_cpu(cpu_modelflag, cpu) = get_cpu_modelflag(cpu); + + 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/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 4b1c319d30c..22590cf688a 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -601,7 +601,7 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) if (!data) return -ENOMEM; - data->acpi_data = percpu_ptr(acpi_perf_data, cpu); + data->acpi_data = per_cpu_ptr(acpi_perf_data, cpu); per_cpu(drv_data, cpu) = data; if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) diff --git a/arch/x86/kernel/cpu/cpufreq/e_powersaver.c b/arch/x86/kernel/cpu/cpufreq/e_powersaver.c index c2f930d8664..41ab3f064cb 100644 --- a/arch/x86/kernel/cpu/cpufreq/e_powersaver.c +++ b/arch/x86/kernel/cpu/cpufreq/e_powersaver.c @@ -204,12 +204,12 @@ static int eps_cpu_init(struct cpufreq_policy *policy) } /* Enable Enhanced PowerSaver */ rdmsrl(MSR_IA32_MISC_ENABLE, val); - if (!(val & 1 << 16)) { - val |= 1 << 16; + if (!(val & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) { + val |= MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP; wrmsrl(MSR_IA32_MISC_ENABLE, val); /* Can be locked at 0 */ rdmsrl(MSR_IA32_MISC_ENABLE, val); - if (!(val & 1 << 16)) { + if (!(val & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) { printk(KERN_INFO "eps: Can't enable Enhanced PowerSaver\n"); return -ENODEV; } diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c index f08998278a3..c9f1fdc0283 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c @@ -390,14 +390,14 @@ static int centrino_cpu_init(struct cpufreq_policy *policy) enable it if not. */ rdmsr(MSR_IA32_MISC_ENABLE, l, h); - if (!(l & (1<<16))) { - l |= (1<<16); + if (!(l & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) { + l |= MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP; dprintk("trying to enable Enhanced SpeedStep (%x)\n", l); wrmsr(MSR_IA32_MISC_ENABLE, l, h); /* check to see if it stuck */ rdmsr(MSR_IA32_MISC_ENABLE, l, h); - if (!(l & (1<<16))) { + if (!(l & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) { printk(KERN_INFO PFX "couldn't enable Enhanced SpeedStep\n"); return -ENODEV; diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c index ffd0f5ed071..593171e967e 100644 --- a/arch/x86/kernel/cpu/cyrix.c +++ b/arch/x86/kernel/cpu/cyrix.c @@ -61,23 +61,23 @@ static void __cpuinit do_cyrix_devid(unsigned char *dir0, unsigned char *dir1) */ static unsigned char Cx86_dir0_msb __cpuinitdata = 0; -static char Cx86_model[][9] __cpuinitdata = { +static const char __cpuinitconst Cx86_model[][9] = { "Cx486", "Cx486", "5x86 ", "6x86", "MediaGX ", "6x86MX ", "M II ", "Unknown" }; -static char Cx486_name[][5] __cpuinitdata = { +static const char __cpuinitconst Cx486_name[][5] = { "SLC", "DLC", "SLC2", "DLC2", "SRx", "DRx", "SRx2", "DRx2" }; -static char Cx486S_name[][4] __cpuinitdata = { +static const char __cpuinitconst Cx486S_name[][4] = { "S", "S2", "Se", "S2e" }; -static char Cx486D_name[][4] __cpuinitdata = { +static const char __cpuinitconst Cx486D_name[][4] = { "DX", "DX2", "?", "?", "?", "DX4" }; static char Cx86_cb[] __cpuinitdata = "?.5x Core/Bus Clock"; -static char cyrix_model_mult1[] __cpuinitdata = "12??43"; -static char cyrix_model_mult2[] __cpuinitdata = "12233445"; +static const char __cpuinitconst cyrix_model_mult1[] = "12??43"; +static const char __cpuinitconst cyrix_model_mult2[] = "12233445"; /* * Reset the slow-loop (SLOP) bit on the 686(L) which is set by some old @@ -435,7 +435,7 @@ static void __cpuinit cyrix_identify(struct cpuinfo_x86 *c) } } -static struct cpu_dev cyrix_cpu_dev __cpuinitdata = { +static const struct cpu_dev __cpuinitconst cyrix_cpu_dev = { .c_vendor = "Cyrix", .c_ident = { "CyrixInstead" }, .c_early_init = early_init_cyrix, @@ -446,7 +446,7 @@ static struct cpu_dev cyrix_cpu_dev __cpuinitdata = { cpu_dev_register(cyrix_cpu_dev); -static struct cpu_dev nsc_cpu_dev __cpuinitdata = { +static const struct cpu_dev __cpuinitconst nsc_cpu_dev = { .c_vendor = "NSC", .c_ident = { "Geode by NSC" }, .c_init = init_nsc, diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 24ff26a38ad..b09d4eb52bb 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -13,6 +13,7 @@ #include <asm/uaccess.h> #include <asm/ds.h> #include <asm/bugs.h> +#include <asm/cpu.h> #ifdef CONFIG_X86_64 #include <asm/topology.h> @@ -24,7 +25,6 @@ #ifdef CONFIG_X86_LOCAL_APIC #include <asm/mpspec.h> #include <asm/apic.h> -#include <mach_apic.h> #endif static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) @@ -54,6 +54,11 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) c->x86_cache_alignment = 128; #endif + /* CPUID workaround for 0F33/0F34 CPU */ + if (c->x86 == 0xF && c->x86_model == 0x3 + && (c->x86_mask == 0x3 || c->x86_mask == 0x4)) + c->x86_phys_bits = 36; + /* * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate * with P/T states and does not stop in deep C-states @@ -63,6 +68,18 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC); } + /* + * There is a known erratum on Pentium III and Core Solo + * and Core Duo CPUs. + * " Page with PAT set to WC while associated MTRR is UC + * may consolidate to UC " + * Because of this erratum, it is better to stick with + * setting WC in MTRR rather than using PAT on these CPUs. + * + * Enable PAT WC only on P4, Core 2 or later CPUs. + */ + if (c->x86 == 6 && c->x86_model < 15) + clear_cpu_cap(c, X86_FEATURE_PAT); } #ifdef CONFIG_X86_32 @@ -99,6 +116,28 @@ static void __cpuinit trap_init_f00f_bug(void) } #endif +static void __cpuinit intel_smp_check(struct cpuinfo_x86 *c) +{ +#ifdef CONFIG_SMP + /* calling is from identify_secondary_cpu() ? */ + if (c->cpu_index == boot_cpu_id) + return; + + /* + * Mask B, Pentium, but not Pentium MMX + */ + if (c->x86 == 5 && + c->x86_mask >= 1 && c->x86_mask <= 4 && + c->x86_model <= 3) { + /* + * Remember we have B step Pentia with bugs + */ + WARN_ONCE(1, "WARNING: SMP operation may be unreliable" + "with B stepping processors.\n"); + } +#endif +} + static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c) { unsigned long lo, hi; @@ -135,10 +174,10 @@ static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c) */ if ((c->x86 == 15) && (c->x86_model == 1) && (c->x86_mask == 1)) { rdmsr(MSR_IA32_MISC_ENABLE, lo, hi); - if ((lo & (1<<9)) == 0) { + if ((lo & MSR_IA32_MISC_ENABLE_PREFETCH_DISABLE) == 0) { printk (KERN_INFO "CPU: C0 stepping P4 Xeon detected.\n"); printk (KERN_INFO "CPU: Disabling hardware prefetching (Errata 037)\n"); - lo |= (1<<9); /* Disable hw prefetching */ + lo |= MSR_IA32_MISC_ENABLE_PREFETCH_DISABLE; wrmsr (MSR_IA32_MISC_ENABLE, lo, hi); } } @@ -175,6 +214,8 @@ static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c) #ifdef CONFIG_X86_NUMAQ numaq_tsc_disable(); #endif + + intel_smp_check(c); } #else static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c) @@ -374,7 +415,7 @@ static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 *c, unsigned i } #endif -static struct cpu_dev intel_cpu_dev __cpuinitdata = { +static const struct cpu_dev __cpuinitconst intel_cpu_dev = { .c_vendor = "Intel", .c_ident = { "GenuineIntel" }, #ifdef CONFIG_X86_32 diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index da299eb85fc..c471eb1a389 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -32,7 +32,7 @@ struct _cache_table }; /* all the cache descriptor types we care about (no TLB or trace cache entries) */ -static struct _cache_table cache_table[] __cpuinitdata = +static const struct _cache_table __cpuinitconst cache_table[] = { { 0x06, LVL_1_INST, 8 }, /* 4-way set assoc, 32 byte line size */ { 0x08, LVL_1_INST, 16 }, /* 4-way set assoc, 32 byte line size */ @@ -147,7 +147,16 @@ struct _cpuid4_info { union _cpuid4_leaf_ecx ecx; unsigned long size; unsigned long can_disable; - cpumask_t shared_cpu_map; /* future?: only cpus/node is needed */ + DECLARE_BITMAP(shared_cpu_map, NR_CPUS); +}; + +/* subset of above _cpuid4_info w/o shared_cpu_map */ +struct _cpuid4_info_regs { + union _cpuid4_leaf_eax eax; + union _cpuid4_leaf_ebx ebx; + union _cpuid4_leaf_ecx ecx; + unsigned long size; + unsigned long can_disable; }; #ifdef CONFIG_PCI @@ -197,15 +206,15 @@ union l3_cache { unsigned val; }; -static unsigned short assocs[] __cpuinitdata = { +static const unsigned short __cpuinitconst assocs[] = { [1] = 1, [2] = 2, [4] = 4, [6] = 8, [8] = 16, [0xa] = 32, [0xb] = 48, [0xc] = 64, [0xf] = 0xffff // ?? }; -static unsigned char levels[] __cpuinitdata = { 1, 1, 2, 3 }; -static unsigned char types[] __cpuinitdata = { 1, 2, 3, 3 }; +static const unsigned char __cpuinitconst levels[] = { 1, 1, 2, 3 }; +static const unsigned char __cpuinitconst types[] = { 1, 2, 3, 3 }; static void __cpuinit amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax, @@ -278,7 +287,7 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax, } static void __cpuinit -amd_check_l3_disable(int index, struct _cpuid4_info *this_leaf) +amd_check_l3_disable(int index, struct _cpuid4_info_regs *this_leaf) { if (index < 3) return; @@ -286,7 +295,8 @@ amd_check_l3_disable(int index, struct _cpuid4_info *this_leaf) } static int -__cpuinit cpuid4_cache_lookup(int index, struct _cpuid4_info *this_leaf) +__cpuinit cpuid4_cache_lookup_regs(int index, + struct _cpuid4_info_regs *this_leaf) { union _cpuid4_leaf_eax eax; union _cpuid4_leaf_ebx ebx; @@ -314,6 +324,15 @@ __cpuinit cpuid4_cache_lookup(int index, struct _cpuid4_info *this_leaf) return 0; } +static int +__cpuinit cpuid4_cache_lookup(int index, struct _cpuid4_info *this_leaf) +{ + struct _cpuid4_info_regs *leaf_regs = + (struct _cpuid4_info_regs *)this_leaf; + + return cpuid4_cache_lookup_regs(index, leaf_regs); +} + static int __cpuinit find_num_cache_leaves(void) { unsigned int eax, ebx, ecx, edx; @@ -353,11 +372,10 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) * parameters cpuid leaf to find the cache details */ for (i = 0; i < num_cache_leaves; i++) { - struct _cpuid4_info this_leaf; - + struct _cpuid4_info_regs this_leaf; int retval; - retval = cpuid4_cache_lookup(i, &this_leaf); + retval = cpuid4_cache_lookup_regs(i, &this_leaf); if (retval >= 0) { switch(this_leaf.eax.split.level) { case 1: @@ -506,17 +524,20 @@ static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index) num_threads_sharing = 1 + this_leaf->eax.split.num_threads_sharing; if (num_threads_sharing == 1) - cpu_set(cpu, this_leaf->shared_cpu_map); + cpumask_set_cpu(cpu, to_cpumask(this_leaf->shared_cpu_map)); else { index_msb = get_count_order(num_threads_sharing); for_each_online_cpu(i) { if (cpu_data(i).apicid >> index_msb == c->apicid >> index_msb) { - cpu_set(i, this_leaf->shared_cpu_map); + cpumask_set_cpu(i, + to_cpumask(this_leaf->shared_cpu_map)); if (i != cpu && per_cpu(cpuid4_info, i)) { - sibling_leaf = CPUID4_INFO_IDX(i, index); - cpu_set(cpu, sibling_leaf->shared_cpu_map); + sibling_leaf = + CPUID4_INFO_IDX(i, index); + cpumask_set_cpu(cpu, to_cpumask( + sibling_leaf->shared_cpu_map)); } } } @@ -528,9 +549,10 @@ static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index) int sibling; this_leaf = CPUID4_INFO_IDX(cpu, index); - for_each_cpu_mask_nr(sibling, this_leaf->shared_cpu_map) { + for_each_cpu(sibling, to_cpumask(this_leaf->shared_cpu_map)) { sibling_leaf = CPUID4_INFO_IDX(sibling, index); - cpu_clear(cpu, sibling_leaf->shared_cpu_map); + cpumask_clear_cpu(cpu, + to_cpumask(sibling_leaf->shared_cpu_map)); } } #else @@ -635,8 +657,9 @@ static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf, int n = 0; if (len > 1) { - cpumask_t *mask = &this_leaf->shared_cpu_map; + const struct cpumask *mask; + mask = to_cpumask(this_leaf->shared_cpu_map); n = type? cpulist_scnprintf(buf, len-2, mask) : cpumask_scnprintf(buf, len-2, mask); @@ -699,7 +722,8 @@ static struct pci_dev *get_k8_northbridge(int node) static ssize_t show_cache_disable(struct _cpuid4_info *this_leaf, char *buf) { - int node = cpu_to_node(first_cpu(this_leaf->shared_cpu_map)); + const struct cpumask *mask = to_cpumask(this_leaf->shared_cpu_map); + int node = cpu_to_node(cpumask_first(mask)); struct pci_dev *dev = NULL; ssize_t ret = 0; int i; @@ -733,7 +757,8 @@ static ssize_t store_cache_disable(struct _cpuid4_info *this_leaf, const char *buf, size_t count) { - int node = cpu_to_node(first_cpu(this_leaf->shared_cpu_map)); + const struct cpumask *mask = to_cpumask(this_leaf->shared_cpu_map); + int node = cpu_to_node(cpumask_first(mask)); struct pci_dev *dev = NULL; unsigned int ret, index, val; @@ -878,7 +903,7 @@ err_out: return -ENOMEM; } -static cpumask_t cache_dev_map = CPU_MASK_NONE; +static DECLARE_BITMAP(cache_dev_map, NR_CPUS); /* Add/Remove cache interface for CPU device */ static int __cpuinit cache_add_dev(struct sys_device * sys_dev) @@ -918,7 +943,7 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev) } kobject_uevent(&(this_object->kobj), KOBJ_ADD); } - cpu_set(cpu, cache_dev_map); + cpumask_set_cpu(cpu, to_cpumask(cache_dev_map)); kobject_uevent(per_cpu(cache_kobject, cpu), KOBJ_ADD); return 0; @@ -931,9 +956,9 @@ static void __cpuinit cache_remove_dev(struct sys_device * sys_dev) if (per_cpu(cpuid4_info, cpu) == NULL) return; - if (!cpu_isset(cpu, cache_dev_map)) + if (!cpumask_test_cpu(cpu, to_cpumask(cache_dev_map))) return; - cpu_clear(cpu, cache_dev_map); + cpumask_clear_cpu(cpu, to_cpumask(cache_dev_map)); for (i = 0; i < num_cache_leaves; i++) kobject_put(&(INDEX_KOBJECT_PTR(cpu,i)->kobj)); diff --git a/arch/x86/kernel/cpu/mcheck/Makefile b/arch/x86/kernel/cpu/mcheck/Makefile index d7d2323bbb6..b2f89829bbe 100644 --- a/arch/x86/kernel/cpu/mcheck/Makefile +++ b/arch/x86/kernel/cpu/mcheck/Makefile @@ -4,3 +4,4 @@ obj-$(CONFIG_X86_32) += k7.o p4.o p5.o p6.o winchip.o obj-$(CONFIG_X86_MCE_INTEL) += mce_intel_64.o obj-$(CONFIG_X86_MCE_AMD) += mce_amd_64.o obj-$(CONFIG_X86_MCE_NONFATAL) += non-fatal.o +obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o diff --git a/arch/x86/kernel/cpu/mcheck/mce_32.c b/arch/x86/kernel/cpu/mcheck/mce_32.c index dfaebce3633..3552119b091 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_32.c +++ b/arch/x86/kernel/cpu/mcheck/mce_32.c @@ -60,20 +60,6 @@ void mcheck_init(struct cpuinfo_x86 *c) } } -static unsigned long old_cr4 __initdata; - -void __init stop_mce(void) -{ - old_cr4 = read_cr4(); - clear_in_cr4(X86_CR4_MCE); -} - -void __init restart_mce(void) -{ - if (old_cr4 & X86_CR4_MCE) - set_in_cr4(X86_CR4_MCE); -} - static int __init mcheck_disable(char *str) { mce_disabled = 1; diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c index fe79985ce0f..ca14604611e 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_64.c @@ -3,6 +3,8 @@ * K8 parts Copyright 2002,2003 Andi Kleen, SuSE Labs. * Rest from unknown author(s). * 2004 Andi Kleen. Rewrote most of it. + * Copyright 2008 Intel Corporation + * Author: Andi Kleen */ #include <linux/init.h> @@ -24,6 +26,9 @@ #include <linux/ctype.h> #include <linux/kmod.h> #include <linux/kdebug.h> +#include <linux/kobject.h> +#include <linux/sysfs.h> +#include <linux/ratelimit.h> #include <asm/processor.h> #include <asm/msr.h> #include <asm/mce.h> @@ -32,7 +37,6 @@ #include <asm/idle.h> #define MISC_MCELOG_MINOR 227 -#define NR_SYSFS_BANKS 6 atomic_t mce_entry; @@ -47,7 +51,7 @@ static int mce_dont_init; */ static int tolerant = 1; static int banks; -static unsigned long bank[NR_SYSFS_BANKS] = { [0 ... NR_SYSFS_BANKS-1] = ~0UL }; +static u64 *bank; static unsigned long notify_user; static int rip_msr; static int mce_bootlog = -1; @@ -58,6 +62,19 @@ static char *trigger_argv[2] = { trigger, NULL }; static DECLARE_WAIT_QUEUE_HEAD(mce_wait); +/* MCA banks polled by the period polling timer for corrected events */ +DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = { + [0 ... BITS_TO_LONGS(MAX_NR_BANKS)-1] = ~0UL +}; + +/* Do initial initialization of a struct mce */ +void mce_setup(struct mce *m) +{ + memset(m, 0, sizeof(struct mce)); + m->cpu = smp_processor_id(); + rdtscll(m->tsc); +} + /* * Lockless MCE logging infrastructure. * This avoids deadlocks on printk locks without having to break locks. Also @@ -119,11 +136,11 @@ static void print_mce(struct mce *m) print_symbol("{%s}", m->ip); printk("\n"); } - printk(KERN_EMERG "TSC %Lx ", m->tsc); + printk(KERN_EMERG "TSC %llx ", m->tsc); if (m->addr) - printk("ADDR %Lx ", m->addr); + printk("ADDR %llx ", m->addr); if (m->misc) - printk("MISC %Lx ", m->misc); + printk("MISC %llx ", m->misc); printk("\n"); printk(KERN_EMERG "This is not a software problem!\n"); printk(KERN_EMERG "Run through mcelog --ascii to decode " @@ -149,8 +166,10 @@ static void mce_panic(char *msg, struct mce *backup, unsigned long start) panic(msg); } -static int mce_available(struct cpuinfo_x86 *c) +int mce_available(struct cpuinfo_x86 *c) { + if (mce_dont_init) + return 0; return cpu_has(c, X86_FEATURE_MCE) && cpu_has(c, X86_FEATURE_MCA); } @@ -172,7 +191,77 @@ static inline void mce_get_rip(struct mce *m, struct pt_regs *regs) } /* - * The actual machine check handler + * Poll for corrected events or events that happened before reset. + * Those are just logged through /dev/mcelog. + * + * This is executed in standard interrupt context. + */ +void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) +{ + struct mce m; + int i; + + mce_setup(&m); + + rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus); + for (i = 0; i < banks; i++) { + if (!bank[i] || !test_bit(i, *b)) + continue; + + m.misc = 0; + m.addr = 0; + m.bank = i; + m.tsc = 0; + + barrier(); + rdmsrl(MSR_IA32_MC0_STATUS + i*4, m.status); + if (!(m.status & MCI_STATUS_VAL)) + continue; + + /* + * Uncorrected events are handled by the exception handler + * when it is enabled. But when the exception is disabled log + * everything. + * + * TBD do the same check for MCI_STATUS_EN here? + */ + if ((m.status & MCI_STATUS_UC) && !(flags & MCP_UC)) + continue; + + if (m.status & MCI_STATUS_MISCV) + rdmsrl(MSR_IA32_MC0_MISC + i*4, m.misc); + if (m.status & MCI_STATUS_ADDRV) + rdmsrl(MSR_IA32_MC0_ADDR + i*4, m.addr); + + if (!(flags & MCP_TIMESTAMP)) + m.tsc = 0; + /* + * Don't get the IP here because it's unlikely to + * have anything to do with the actual error location. + */ + + mce_log(&m); + add_taint(TAINT_MACHINE_CHECK); + + /* + * Clear state for this bank. + */ + wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + } + + /* + * Don't clear MCG_STATUS here because it's only defined for + * exceptions. + */ +} + +/* + * The actual machine check handler. This only handles real + * exceptions when something got corrupted coming in through int 18. + * + * This is executed in NMI context not subject to normal locking rules. This + * implies that most kernel services cannot be safely used. Don't even + * think about putting a printk in there! */ void do_machine_check(struct pt_regs * regs, long error_code) { @@ -190,17 +279,18 @@ void do_machine_check(struct pt_regs * regs, long error_code) * error. */ int kill_it = 0; + DECLARE_BITMAP(toclear, MAX_NR_BANKS); atomic_inc(&mce_entry); - if ((regs - && notify_die(DIE_NMI, "machine check", regs, error_code, + if (notify_die(DIE_NMI, "machine check", regs, error_code, 18, SIGKILL) == NOTIFY_STOP) - || !banks) + goto out2; + if (!banks) goto out2; - memset(&m, 0, sizeof(struct mce)); - m.cpu = smp_processor_id(); + mce_setup(&m); + rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus); /* if the restart IP is not valid, we're done for */ if (!(m.mcgstatus & MCG_STATUS_RIPV)) @@ -210,18 +300,32 @@ void do_machine_check(struct pt_regs * regs, long error_code) barrier(); for (i = 0; i < banks; i++) { - if (i < NR_SYSFS_BANKS && !bank[i]) + __clear_bit(i, toclear); + if (!bank[i]) continue; m.misc = 0; m.addr = 0; m.bank = i; - m.tsc = 0; rdmsrl(MSR_IA32_MC0_STATUS + i*4, m.status); if ((m.status & MCI_STATUS_VAL) == 0) continue; + /* + * Non uncorrected errors are handled by machine_check_poll + * Leave them alone. + */ + if ((m.status & MCI_STATUS_UC) == 0) + continue; + + /* + * Set taint even when machine check was not enabled. + */ + add_taint(TAINT_MACHINE_CHECK); + + __set_bit(i, toclear); + if (m.status & MCI_STATUS_EN) { /* if PCC was set, there's no way out */ no_way_out |= !!(m.status & MCI_STATUS_PCC); @@ -235,6 +339,12 @@ void do_machine_check(struct pt_regs * regs, long error_code) no_way_out = 1; kill_it = 1; } + } else { + /* + * Machine check event was not enabled. Clear, but + * ignore. + */ + continue; } if (m.status & MCI_STATUS_MISCV) @@ -243,10 +353,7 @@ void do_machine_check(struct pt_regs * regs, long error_code) rdmsrl(MSR_IA32_MC0_ADDR + i*4, m.addr); mce_get_rip(&m, regs); - if (error_code >= 0) - rdtscll(m.tsc); - if (error_code != -2) - mce_log(&m); + mce_log(&m); /* Did this bank cause the exception? */ /* Assume that the bank with uncorrectable errors did it, @@ -255,14 +362,8 @@ void do_machine_check(struct pt_regs * regs, long error_code) panicm = m; panicm_found = 1; } - - add_taint(TAINT_MACHINE_CHECK); } - /* Never do anything final in the polling timer */ - if (!regs) - goto out; - /* If we didn't find an uncorrectable error, pick the last one (shouldn't happen, just being safe). */ if (!panicm_found) @@ -309,10 +410,11 @@ void do_machine_check(struct pt_regs * regs, long error_code) /* notify userspace ASAP */ set_thread_flag(TIF_MCE_NOTIFY); - out: /* the last thing we do is clear state */ - for (i = 0; i < banks; i++) - wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + for (i = 0; i < banks; i++) { + if (test_bit(i, toclear)) + wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + } wrmsrl(MSR_IA32_MCG_STATUS, 0); out2: atomic_dec(&mce_entry); @@ -332,15 +434,13 @@ void do_machine_check(struct pt_regs * regs, long error_code) * and historically has been the register value of the * MSR_IA32_THERMAL_STATUS (Intel) msr. */ -void mce_log_therm_throt_event(unsigned int cpu, __u64 status) +void mce_log_therm_throt_event(__u64 status) { struct mce m; - memset(&m, 0, sizeof(m)); - m.cpu = cpu; + mce_setup(&m); m.bank = MCE_THERMAL_BANK; m.status = status; - rdtscll(m.tsc); mce_log(&m); } #endif /* CONFIG_X86_MCE_INTEL */ @@ -353,18 +453,18 @@ void mce_log_therm_throt_event(unsigned int cpu, __u64 status) static int check_interval = 5 * 60; /* 5 minutes */ static int next_interval; /* in jiffies */ -static void mcheck_timer(struct work_struct *work); -static DECLARE_DELAYED_WORK(mcheck_work, mcheck_timer); +static void mcheck_timer(unsigned long); +static DEFINE_PER_CPU(struct timer_list, mce_timer); -static void mcheck_check_cpu(void *info) +static void mcheck_timer(unsigned long data) { - if (mce_available(¤t_cpu_data)) - do_machine_check(NULL, 0); -} + struct timer_list *t = &per_cpu(mce_timer, data); -static void mcheck_timer(struct work_struct *work) -{ - on_each_cpu(mcheck_check_cpu, NULL, 1); + WARN_ON(smp_processor_id() != data); + + if (mce_available(¤t_cpu_data)) + machine_check_poll(MCP_TIMESTAMP, + &__get_cpu_var(mce_poll_banks)); /* * Alert userspace if needed. If we logged an MCE, reduce the @@ -377,31 +477,41 @@ static void mcheck_timer(struct work_struct *work) (int)round_jiffies_relative(check_interval*HZ)); } - schedule_delayed_work(&mcheck_work, next_interval); + t->expires = jiffies + next_interval; + add_timer(t); +} + +static void mce_do_trigger(struct work_struct *work) +{ + call_usermodehelper(trigger, trigger_argv, NULL, UMH_NO_WAIT); } +static DECLARE_WORK(mce_trigger_work, mce_do_trigger); + /* - * This is only called from process context. This is where we do - * anything we need to alert userspace about new MCEs. This is called - * directly from the poller and also from entry.S and idle, thanks to - * TIF_MCE_NOTIFY. + * Notify the user(s) about new machine check events. + * Can be called from interrupt context, but not from machine check/NMI + * context. */ int mce_notify_user(void) { + /* Not more than two messages every minute */ + static DEFINE_RATELIMIT_STATE(ratelimit, 60*HZ, 2); + clear_thread_flag(TIF_MCE_NOTIFY); if (test_and_clear_bit(0, ¬ify_user)) { - static unsigned long last_print; - unsigned long now = jiffies; - wake_up_interruptible(&mce_wait); - if (trigger[0]) - call_usermodehelper(trigger, trigger_argv, NULL, - UMH_NO_WAIT); - if (time_after_eq(now, last_print + (check_interval*HZ))) { - last_print = now; + /* + * There is no risk of missing notifications because + * work_pending is always cleared before the function is + * executed. + */ + if (trigger[0] && !work_pending(&mce_trigger_work)) + schedule_work(&mce_trigger_work); + + if (__ratelimit(&ratelimit)) printk(KERN_INFO "Machine check events logged\n"); - } return 1; } @@ -425,63 +535,78 @@ static struct notifier_block mce_idle_notifier = { static __init int periodic_mcheck_init(void) { - next_interval = check_interval * HZ; - if (next_interval) - schedule_delayed_work(&mcheck_work, - round_jiffies_relative(next_interval)); - idle_notifier_register(&mce_idle_notifier); - return 0; + idle_notifier_register(&mce_idle_notifier); + return 0; } __initcall(periodic_mcheck_init); - /* * Initialize Machine Checks for a CPU. */ -static void mce_init(void *dummy) +static int mce_cap_init(void) { u64 cap; - int i; + unsigned b; rdmsrl(MSR_IA32_MCG_CAP, cap); - banks = cap & 0xff; - if (banks > MCE_EXTENDED_BANK) { - banks = MCE_EXTENDED_BANK; - printk(KERN_INFO "MCE: warning: using only %d banks\n", - MCE_EXTENDED_BANK); + b = cap & 0xff; + if (b > MAX_NR_BANKS) { + printk(KERN_WARNING + "MCE: Using only %u machine check banks out of %u\n", + MAX_NR_BANKS, b); + b = MAX_NR_BANKS; } + + /* Don't support asymmetric configurations today */ + WARN_ON(banks != 0 && b != banks); + banks = b; + if (!bank) { + bank = kmalloc(banks * sizeof(u64), GFP_KERNEL); + if (!bank) + return -ENOMEM; + memset(bank, 0xff, banks * sizeof(u64)); + } + /* Use accurate RIP reporting if available. */ if ((cap & (1<<9)) && ((cap >> 16) & 0xff) >= 9) rip_msr = MSR_IA32_MCG_EIP; - /* Log the machine checks left over from the previous reset. - This also clears all registers */ - do_machine_check(NULL, mce_bootlog ? -1 : -2); + return 0; +} + +static void mce_init(void *dummy) +{ + u64 cap; + int i; + mce_banks_t all_banks; + + /* + * Log the machine checks left over from the previous reset. + */ + bitmap_fill(all_banks, MAX_NR_BANKS); + machine_check_poll(MCP_UC, &all_banks); set_in_cr4(X86_CR4_MCE); + rdmsrl(MSR_IA32_MCG_CAP, cap); if (cap & MCG_CTL_P) wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); for (i = 0; i < banks; i++) { - if (i < NR_SYSFS_BANKS) - wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]); - else - wrmsrl(MSR_IA32_MC0_CTL+4*i, ~0UL); - + wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]); wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); } } /* Add per CPU specific workarounds here */ -static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c) +static void mce_cpu_quirks(struct cpuinfo_x86 *c) { /* This should be disabled by the BIOS, but isn't always */ if (c->x86_vendor == X86_VENDOR_AMD) { - if(c->x86 == 15) + if (c->x86 == 15 && banks > 4) /* disable GART TBL walk error reporting, which trips off incorrectly with the IOMMU & 3ware & Cerberus. */ - clear_bit(10, &bank[4]); + clear_bit(10, (unsigned long *)&bank[4]); if(c->x86 <= 17 && mce_bootlog < 0) /* Lots of broken BIOS around that don't clear them by default and leave crap in there. Don't log. */ @@ -504,20 +629,38 @@ static void mce_cpu_features(struct cpuinfo_x86 *c) } } +static void mce_init_timer(void) +{ + struct timer_list *t = &__get_cpu_var(mce_timer); + + /* data race harmless because everyone sets to the same value */ + if (!next_interval) + next_interval = check_interval * HZ; + if (!next_interval) + return; + setup_timer(t, mcheck_timer, smp_processor_id()); + t->expires = round_jiffies(jiffies + next_interval); + add_timer(t); +} + /* * Called for each booted CPU to set up machine checks. * Must be called with preempt off. */ void __cpuinit mcheck_init(struct cpuinfo_x86 *c) { - mce_cpu_quirks(c); + if (!mce_available(c)) + return; - if (mce_dont_init || - !mce_available(c)) + if (mce_cap_init() < 0) { + mce_dont_init = 1; return; + } + mce_cpu_quirks(c); mce_init(NULL); mce_cpu_features(c); + mce_init_timer(); } /* @@ -573,7 +716,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, { unsigned long *cpu_tsc; static DEFINE_MUTEX(mce_read_mutex); - unsigned next; + unsigned prev, next; char __user *buf = ubuf; int i, err; @@ -592,25 +735,32 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, } err = 0; - for (i = 0; i < next; i++) { - unsigned long start = jiffies; - - while (!mcelog.entry[i].finished) { - if (time_after_eq(jiffies, start + 2)) { - memset(mcelog.entry + i,0, sizeof(struct mce)); - goto timeout; + prev = 0; + do { + for (i = prev; i < next; i++) { + unsigned long start = jiffies; + + while (!mcelog.entry[i].finished) { + if (time_after_eq(jiffies, start + 2)) { + memset(mcelog.entry + i, 0, + sizeof(struct mce)); + goto timeout; + } + cpu_relax(); } - cpu_relax(); + smp_rmb(); + err |= copy_to_user(buf, mcelog.entry + i, + sizeof(struct mce)); + buf += sizeof(struct mce); +timeout: + ; } - smp_rmb(); - err |= copy_to_user(buf, mcelog.entry + i, sizeof(struct mce)); - buf += sizeof(struct mce); - timeout: - ; - } - memset(mcelog.entry, 0, next * sizeof(struct mce)); - mcelog.next = 0; + memset(mcelog.entry + prev, 0, + (next - prev) * sizeof(struct mce)); + prev = next; + next = cmpxchg(&mcelog.next, prev, 0); + } while (next != prev); synchronize_sched(); @@ -680,20 +830,6 @@ static struct miscdevice mce_log_device = { &mce_chrdev_ops, }; -static unsigned long old_cr4 __initdata; - -void __init stop_mce(void) -{ - old_cr4 = read_cr4(); - clear_in_cr4(X86_CR4_MCE); -} - -void __init restart_mce(void) -{ - if (old_cr4 & X86_CR4_MCE) - set_in_cr4(X86_CR4_MCE); -} - /* * Old style boot options parsing. Only for compatibility. */ @@ -703,8 +839,7 @@ static int __init mcheck_disable(char *str) return 1; } -/* mce=off disables machine check. Note you can re-enable it later - using sysfs. +/* mce=off disables machine check. mce=TOLERANCELEVEL (number, see above) mce=bootlog Log MCEs from before booting. Disabled by default on AMD. mce=nobootlog Don't log MCEs from before booting. */ @@ -728,6 +863,29 @@ __setup("mce=", mcheck_enable); * Sysfs support */ +/* + * Disable machine checks on suspend and shutdown. We can't really handle + * them later. + */ +static int mce_disable(void) +{ + int i; + + for (i = 0; i < banks; i++) + wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); + return 0; +} + +static int mce_suspend(struct sys_device *dev, pm_message_t state) +{ + return mce_disable(); +} + +static int mce_shutdown(struct sys_device *dev) +{ + return mce_disable(); +} + /* On resume clear all MCE state. Don't want to see leftovers from the BIOS. Only one CPU is active at this time, the others get readded later using CPU hotplug. */ @@ -738,20 +896,24 @@ static int mce_resume(struct sys_device *dev) return 0; } +static void mce_cpu_restart(void *data) +{ + del_timer_sync(&__get_cpu_var(mce_timer)); + if (mce_available(¤t_cpu_data)) + mce_init(NULL); + mce_init_timer(); +} + /* Reinit MCEs after user configuration changes */ static void mce_restart(void) { - if (next_interval) - cancel_delayed_work(&mcheck_work); - /* Timer race is harmless here */ - on_each_cpu(mce_init, NULL, 1); next_interval = check_interval * HZ; - if (next_interval) - schedule_delayed_work(&mcheck_work, - round_jiffies_relative(next_interval)); + on_each_cpu(mce_cpu_restart, NULL, 1); } static struct sysdev_class mce_sysclass = { + .suspend = mce_suspend, + .shutdown = mce_shutdown, .resume = mce_resume, .name = "machinecheck", }; @@ -778,16 +940,26 @@ void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu) __cpuinit } \ static SYSDEV_ATTR(name, 0644, show_ ## name, set_ ## name); -/* - * TBD should generate these dynamically based on number of available banks. - * Have only 6 contol banks in /sysfs until then. - */ -ACCESSOR(bank0ctl,bank[0],mce_restart()) -ACCESSOR(bank1ctl,bank[1],mce_restart()) -ACCESSOR(bank2ctl,bank[2],mce_restart()) -ACCESSOR(bank3ctl,bank[3],mce_restart()) -ACCESSOR(bank4ctl,bank[4],mce_restart()) -ACCESSOR(bank5ctl,bank[5],mce_restart()) +static struct sysdev_attribute *bank_attrs; + +static ssize_t show_bank(struct sys_device *s, struct sysdev_attribute *attr, + char *buf) +{ + u64 b = bank[attr - bank_attrs]; + return sprintf(buf, "%llx\n", b); +} + +static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr, + const char *buf, size_t siz) +{ + char *end; + u64 new = simple_strtoull(buf, &end, 0); + if (end == buf) + return -EINVAL; + bank[attr - bank_attrs] = new; + mce_restart(); + return end-buf; +} static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr, char *buf) @@ -814,8 +986,6 @@ static SYSDEV_ATTR(trigger, 0644, show_trigger, set_trigger); static SYSDEV_INT_ATTR(tolerant, 0644, tolerant); ACCESSOR(check_interval,check_interval,mce_restart()) static struct sysdev_attribute *mce_attributes[] = { - &attr_bank0ctl, &attr_bank1ctl, &attr_bank2ctl, - &attr_bank3ctl, &attr_bank4ctl, &attr_bank5ctl, &attr_tolerant.attr, &attr_check_interval, &attr_trigger, NULL }; @@ -845,11 +1015,22 @@ static __cpuinit int mce_create_device(unsigned int cpu) if (err) goto error; } + for (i = 0; i < banks; i++) { + err = sysdev_create_file(&per_cpu(device_mce, cpu), + &bank_attrs[i]); + if (err) + goto error2; + } cpu_set(cpu, mce_device_initialized); return 0; +error2: + while (--i >= 0) { + sysdev_remove_file(&per_cpu(device_mce, cpu), + &bank_attrs[i]); + } error: - while (i--) { + while (--i >= 0) { sysdev_remove_file(&per_cpu(device_mce,cpu), mce_attributes[i]); } @@ -868,15 +1049,46 @@ static __cpuinit void mce_remove_device(unsigned int cpu) for (i = 0; mce_attributes[i]; i++) sysdev_remove_file(&per_cpu(device_mce,cpu), mce_attributes[i]); + for (i = 0; i < banks; i++) + sysdev_remove_file(&per_cpu(device_mce, cpu), + &bank_attrs[i]); sysdev_unregister(&per_cpu(device_mce,cpu)); cpu_clear(cpu, mce_device_initialized); } +/* Make sure there are no machine checks on offlined CPUs. */ +static void mce_disable_cpu(void *h) +{ + int i; + unsigned long action = *(unsigned long *)h; + + if (!mce_available(¤t_cpu_data)) + return; + if (!(action & CPU_TASKS_FROZEN)) + cmci_clear(); + for (i = 0; i < banks; i++) + wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); +} + +static void mce_reenable_cpu(void *h) +{ + int i; + unsigned long action = *(unsigned long *)h; + + if (!mce_available(¤t_cpu_data)) + return; + if (!(action & CPU_TASKS_FROZEN)) + cmci_reenable(); + for (i = 0; i < banks; i++) + wrmsrl(MSR_IA32_MC0_CTL + i*4, bank[i]); +} + /* Get notified when a cpu comes on/off. Be hotplug friendly. */ static int __cpuinit mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; + struct timer_list *t = &per_cpu(mce_timer, cpu); switch (action) { case CPU_ONLINE: @@ -891,6 +1103,21 @@ static int __cpuinit mce_cpu_callback(struct notifier_block *nfb, threshold_cpu_callback(action, cpu); mce_remove_device(cpu); break; + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + del_timer_sync(t); + smp_call_function_single(cpu, mce_disable_cpu, &action, 1); + break; + case CPU_DOWN_FAILED: + case CPU_DOWN_FAILED_FROZEN: + t->expires = round_jiffies(jiffies + next_interval); + add_timer_on(t, cpu); + smp_call_function_single(cpu, mce_reenable_cpu, &action, 1); + break; + case CPU_POST_DEAD: + /* intentionally ignoring frozen here */ + cmci_rediscover(cpu); + break; } return NOTIFY_OK; } @@ -899,6 +1126,34 @@ static struct notifier_block mce_cpu_notifier __cpuinitdata = { .notifier_call = mce_cpu_callback, }; +static __init int mce_init_banks(void) +{ + int i; + + bank_attrs = kzalloc(sizeof(struct sysdev_attribute) * banks, + GFP_KERNEL); + if (!bank_attrs) + return -ENOMEM; + + for (i = 0; i < banks; i++) { + struct sysdev_attribute *a = &bank_attrs[i]; + a->attr.name = kasprintf(GFP_KERNEL, "bank%d", i); + if (!a->attr.name) + goto nomem; + a->attr.mode = 0644; + a->show = show_bank; + a->store = set_bank; + } + return 0; + +nomem: + while (--i >= 0) + kfree(bank_attrs[i].attr.name); + kfree(bank_attrs); + bank_attrs = NULL; + return -ENOMEM; +} + static __init int mce_init_device(void) { int err; @@ -906,6 +1161,11 @@ static __init int mce_init_device(void) if (!mce_available(&boot_cpu_data)) return -EIO; + + err = mce_init_banks(); + if (err) + return err; + err = sysdev_class_register(&mce_sysclass); if (err) return err; diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c index f2ee0ae29bd..7d01be86887 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c @@ -67,7 +67,7 @@ static struct threshold_block threshold_defaults = { struct threshold_bank { struct kobject *kobj; struct threshold_block *blocks; - cpumask_t cpus; + cpumask_var_t cpus; }; static DEFINE_PER_CPU(struct threshold_bank *, threshold_banks[NR_BANKS]); @@ -79,6 +79,8 @@ static unsigned char shared_bank[NR_BANKS] = { static DEFINE_PER_CPU(unsigned char, bank_map); /* see which banks are on */ +static void amd_threshold_interrupt(void); + /* * CPU Initialization */ @@ -90,7 +92,8 @@ struct thresh_restart { }; /* must be called with correct cpu affinity */ -static long threshold_restart_bank(void *_tr) +/* Called via smp_call_function_single() */ +static void threshold_restart_bank(void *_tr) { struct thresh_restart *tr = _tr; u32 mci_misc_hi, mci_misc_lo; @@ -117,7 +120,6 @@ static long threshold_restart_bank(void *_tr) mci_misc_hi |= MASK_COUNT_EN_HI; wrmsr(tr->b->address, mci_misc_lo, mci_misc_hi); - return 0; } /* cpu init entry point, called from mce.c with preempt off */ @@ -174,6 +176,8 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) tr.reset = 0; tr.old_limit = 0; threshold_restart_bank(&tr); + + mce_threshold_vector = amd_threshold_interrupt; } } } @@ -187,19 +191,13 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) * the interrupt goes off when error_count reaches threshold_limit. * the handler will simply log mcelog w/ software defined bank number. */ -asmlinkage void mce_threshold_interrupt(void) +static void amd_threshold_interrupt(void) { unsigned int bank, block; struct mce m; u32 low = 0, high = 0, address = 0; - ack_APIC_irq(); - exit_idle(); - irq_enter(); - - memset(&m, 0, sizeof(m)); - rdtscll(m.tsc); - m.cpu = smp_processor_id(); + mce_setup(&m); /* assume first bank caused it */ for (bank = 0; bank < NR_BANKS; ++bank) { @@ -233,7 +231,8 @@ asmlinkage void mce_threshold_interrupt(void) /* Log the machine check that caused the threshold event. */ - do_machine_check(NULL, 0); + machine_check_poll(MCP_TIMESTAMP, + &__get_cpu_var(mce_poll_banks)); if (high & MASK_OVERFLOW_HI) { rdmsrl(address, m.misc); @@ -243,13 +242,10 @@ asmlinkage void mce_threshold_interrupt(void) + bank * NR_BLOCKS + block; mce_log(&m); - goto out; + return; } } } -out: - inc_irq_stat(irq_threshold_count); - irq_exit(); } /* @@ -283,7 +279,7 @@ static ssize_t store_interrupt_enable(struct threshold_block *b, tr.b = b; tr.reset = 0; tr.old_limit = 0; - work_on_cpu(b->cpu, threshold_restart_bank, &tr); + smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1); return end - buf; } @@ -305,23 +301,32 @@ static ssize_t store_threshold_limit(struct threshold_block *b, tr.b = b; tr.reset = 0; - work_on_cpu(b->cpu, threshold_restart_bank, &tr); + smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1); return end - buf; } -static long local_error_count(void *_b) +struct threshold_block_cross_cpu { + struct threshold_block *tb; + long retval; +}; + +static void local_error_count_handler(void *_tbcc) { - struct threshold_block *b = _b; + struct threshold_block_cross_cpu *tbcc = _tbcc; + struct threshold_block *b = tbcc->tb; u32 low, high; rdmsr(b->address, low, high); - return (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit); + tbcc->retval = (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit); } static ssize_t show_error_count(struct threshold_block *b, char *buf) { - return sprintf(buf, "%lx\n", work_on_cpu(b->cpu, local_error_count, b)); + struct threshold_block_cross_cpu tbcc = { .tb = b, }; + + smp_call_function_single(b->cpu, local_error_count_handler, &tbcc, 1); + return sprintf(buf, "%lx\n", tbcc.retval); } static ssize_t store_error_count(struct threshold_block *b, @@ -329,7 +334,7 @@ static ssize_t store_error_count(struct threshold_block *b, { struct thresh_restart tr = { .b = b, .reset = 1, .old_limit = 0 }; - work_on_cpu(b->cpu, threshold_restart_bank, &tr); + smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1); return 1; } @@ -398,7 +403,7 @@ static __cpuinit int allocate_threshold_blocks(unsigned int cpu, if ((bank >= NR_BANKS) || (block >= NR_BLOCKS)) return 0; - if (rdmsr_safe(address, &low, &high)) + if (rdmsr_safe_on_cpu(cpu, address, &low, &high)) return 0; if (!(high & MASK_VALID_HI)) { @@ -462,12 +467,11 @@ out_free: return err; } -static __cpuinit long local_allocate_threshold_blocks(void *_bank) +static __cpuinit long +local_allocate_threshold_blocks(int cpu, unsigned int bank) { - unsigned int *bank = _bank; - - return allocate_threshold_blocks(smp_processor_id(), *bank, 0, - MSR_IA32_MC0_MISC + *bank * 4); + return allocate_threshold_blocks(cpu, bank, 0, + MSR_IA32_MC0_MISC + bank * 4); } /* symlinks sibling shared banks to first core. first core owns dir/files. */ @@ -481,7 +485,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) #ifdef CONFIG_SMP if (cpu_data(cpu).cpu_core_id && shared_bank[bank]) { /* symlink */ - i = first_cpu(per_cpu(cpu_core_map, cpu)); + i = cpumask_first(&per_cpu(cpu_core_map, cpu)); /* first core not up yet */ if (cpu_data(i).cpu_core_id) @@ -501,7 +505,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) if (err) goto out; - b->cpus = per_cpu(cpu_core_map, cpu); + cpumask_copy(b->cpus, &per_cpu(cpu_core_map, cpu)); per_cpu(threshold_banks, cpu)[bank] = b; goto out; } @@ -512,24 +516,29 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) err = -ENOMEM; goto out; } + if (!alloc_cpumask_var(&b->cpus, GFP_KERNEL)) { + kfree(b); + err = -ENOMEM; + goto out; + } b->kobj = kobject_create_and_add(name, &per_cpu(device_mce, cpu).kobj); if (!b->kobj) goto out_free; #ifndef CONFIG_SMP - b->cpus = CPU_MASK_ALL; + cpumask_setall(b->cpus); #else - b->cpus = per_cpu(cpu_core_map, cpu); + cpumask_copy(b->cpus, &per_cpu(cpu_core_map, cpu)); #endif per_cpu(threshold_banks, cpu)[bank] = b; - err = work_on_cpu(cpu, local_allocate_threshold_blocks, &bank); + err = local_allocate_threshold_blocks(cpu, bank); if (err) goto out_free; - for_each_cpu_mask_nr(i, b->cpus) { + for_each_cpu(i, b->cpus) { if (i == cpu) continue; @@ -545,6 +554,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) out_free: per_cpu(threshold_banks, cpu)[bank] = NULL; + free_cpumask_var(b->cpus); kfree(b); out: return err; @@ -619,7 +629,7 @@ static void threshold_remove_bank(unsigned int cpu, int bank) #endif /* remove all sibling symlinks before unregistering */ - for_each_cpu_mask_nr(i, b->cpus) { + for_each_cpu(i, b->cpus) { if (i == cpu) continue; @@ -632,6 +642,7 @@ static void threshold_remove_bank(unsigned int cpu, int bank) free_out: kobject_del(b->kobj); kobject_put(b->kobj); + free_cpumask_var(b->cpus); kfree(b); per_cpu(threshold_banks, cpu)[bank] = NULL; } diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c index f44c3662436..57df3d38347 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c @@ -1,17 +1,21 @@ /* * Intel specific MCE features. * Copyright 2004 Zwane Mwaikambo <zwane@linuxpower.ca> + * Copyright (C) 2008, 2009 Intel Corporation + * Author: Andi Kleen */ #include <linux/init.h> #include <linux/interrupt.h> #include <linux/percpu.h> #include <asm/processor.h> +#include <asm/apic.h> #include <asm/msr.h> #include <asm/mce.h> #include <asm/hw_irq.h> #include <asm/idle.h> #include <asm/therm_throt.h> +#include <asm/apic.h> asmlinkage void smp_thermal_interrupt(void) { @@ -24,7 +28,7 @@ asmlinkage void smp_thermal_interrupt(void) rdmsrl(MSR_IA32_THERM_STATUS, msr_val); if (therm_throt_process(msr_val & 1)) - mce_log_therm_throt_event(smp_processor_id(), msr_val); + mce_log_therm_throt_event(msr_val); inc_irq_stat(irq_thermal_count); irq_exit(); @@ -48,13 +52,13 @@ static void intel_init_thermal(struct cpuinfo_x86 *c) */ rdmsr(MSR_IA32_MISC_ENABLE, l, h); h = apic_read(APIC_LVTTHMR); - if ((l & (1 << 3)) && (h & APIC_DM_SMI)) { + if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { printk(KERN_DEBUG "CPU%d: Thermal monitoring handled by SMI\n", cpu); return; } - if (cpu_has(c, X86_FEATURE_TM2) && (l & (1 << 13))) + if (cpu_has(c, X86_FEATURE_TM2) && (l & MSR_IA32_MISC_ENABLE_TM2)) tm2 = 1; if (h & APIC_VECTOR_MASK) { @@ -72,7 +76,7 @@ static void intel_init_thermal(struct cpuinfo_x86 *c) wrmsr(MSR_IA32_THERM_INTERRUPT, l | 0x03, h); rdmsr(MSR_IA32_MISC_ENABLE, l, h); - wrmsr(MSR_IA32_MISC_ENABLE, l | (1 << 3), h); + wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h); l = apic_read(APIC_LVTTHMR); apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); @@ -84,7 +88,209 @@ static void intel_init_thermal(struct cpuinfo_x86 *c) return; } +/* + * Support for Intel Correct Machine Check Interrupts. This allows + * the CPU to raise an interrupt when a corrected machine check happened. + * Normally we pick those up using a regular polling timer. + * Also supports reliable discovery of shared banks. + */ + +static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned); + +/* + * cmci_discover_lock protects against parallel discovery attempts + * which could race against each other. + */ +static DEFINE_SPINLOCK(cmci_discover_lock); + +#define CMCI_THRESHOLD 1 + +static int cmci_supported(int *banks) +{ + u64 cap; + + /* + * Vendor check is not strictly needed, but the initial + * initialization is vendor keyed and this + * makes sure none of the backdoors are entered otherwise. + */ + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) + return 0; + if (!cpu_has_apic || lapic_get_maxlvt() < 6) + return 0; + rdmsrl(MSR_IA32_MCG_CAP, cap); + *banks = min_t(unsigned, MAX_NR_BANKS, cap & 0xff); + return !!(cap & MCG_CMCI_P); +} + +/* + * The interrupt handler. This is called on every event. + * Just call the poller directly to log any events. + * This could in theory increase the threshold under high load, + * but doesn't for now. + */ +static void intel_threshold_interrupt(void) +{ + machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); + mce_notify_user(); +} + +static void print_update(char *type, int *hdr, int num) +{ + if (*hdr == 0) + printk(KERN_INFO "CPU %d MCA banks", smp_processor_id()); + *hdr = 1; + printk(KERN_CONT " %s:%d", type, num); +} + +/* + * Enable CMCI (Corrected Machine Check Interrupt) for available MCE banks + * on this CPU. Use the algorithm recommended in the SDM to discover shared + * banks. + */ +static void cmci_discover(int banks, int boot) +{ + unsigned long *owned = (void *)&__get_cpu_var(mce_banks_owned); + int hdr = 0; + int i; + + spin_lock(&cmci_discover_lock); + for (i = 0; i < banks; i++) { + u64 val; + + if (test_bit(i, owned)) + continue; + + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + + /* Already owned by someone else? */ + if (val & CMCI_EN) { + if (test_and_clear_bit(i, owned) || boot) + print_update("SHD", &hdr, i); + __clear_bit(i, __get_cpu_var(mce_poll_banks)); + continue; + } + + val |= CMCI_EN | CMCI_THRESHOLD; + wrmsrl(MSR_IA32_MC0_CTL2 + i, val); + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + + /* Did the enable bit stick? -- the bank supports CMCI */ + if (val & CMCI_EN) { + if (!test_and_set_bit(i, owned) || boot) + print_update("CMCI", &hdr, i); + __clear_bit(i, __get_cpu_var(mce_poll_banks)); + } else { + WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks))); + } + } + spin_unlock(&cmci_discover_lock); + if (hdr) + printk(KERN_CONT "\n"); +} + +/* + * Just in case we missed an event during initialization check + * all the CMCI owned banks. + */ +void cmci_recheck(void) +{ + unsigned long flags; + int banks; + + if (!mce_available(¤t_cpu_data) || !cmci_supported(&banks)) + return; + local_irq_save(flags); + machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); + local_irq_restore(flags); +} + +/* + * Disable CMCI on this CPU for all banks it owns when it goes down. + * This allows other CPUs to claim the banks on rediscovery. + */ +void cmci_clear(void) +{ + int i; + int banks; + u64 val; + + if (!cmci_supported(&banks)) + return; + spin_lock(&cmci_discover_lock); + for (i = 0; i < banks; i++) { + if (!test_bit(i, __get_cpu_var(mce_banks_owned))) + continue; + /* Disable CMCI */ + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + val &= ~(CMCI_EN|CMCI_THRESHOLD_MASK); + wrmsrl(MSR_IA32_MC0_CTL2 + i, val); + __clear_bit(i, __get_cpu_var(mce_banks_owned)); + } + spin_unlock(&cmci_discover_lock); +} + +/* + * After a CPU went down cycle through all the others and rediscover + * Must run in process context. + */ +void cmci_rediscover(int dying) +{ + int banks; + int cpu; + cpumask_var_t old; + + if (!cmci_supported(&banks)) + return; + if (!alloc_cpumask_var(&old, GFP_KERNEL)) + return; + cpumask_copy(old, ¤t->cpus_allowed); + + for_each_online_cpu (cpu) { + if (cpu == dying) + continue; + if (set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu))) + continue; + /* Recheck banks in case CPUs don't all have the same */ + if (cmci_supported(&banks)) + cmci_discover(banks, 0); + } + + set_cpus_allowed_ptr(current, old); + free_cpumask_var(old); +} + +/* + * Reenable CMCI on this CPU in case a CPU down failed. + */ +void cmci_reenable(void) +{ + int banks; + if (cmci_supported(&banks)) + cmci_discover(banks, 0); +} + +static void intel_init_cmci(void) +{ + int banks; + + if (!cmci_supported(&banks)) + return; + + mce_threshold_vector = intel_threshold_interrupt; + cmci_discover(banks, 1); + /* + * For CPU #0 this runs with still disabled APIC, but that's + * ok because only the vector is set up. We still do another + * check for the banks later for CPU #0 just to make sure + * to not miss any events. + */ + apic_write(APIC_LVTCMCI, THRESHOLD_APIC_VECTOR|APIC_DM_FIXED); + cmci_recheck(); +} + void mce_intel_feature_init(struct cpuinfo_x86 *c) { intel_init_thermal(c); + intel_init_cmci(); } diff --git a/arch/x86/kernel/cpu/mcheck/p4.c b/arch/x86/kernel/cpu/mcheck/p4.c index 9b60fce09f7..f53bdcbaf38 100644 --- a/arch/x86/kernel/cpu/mcheck/p4.c +++ b/arch/x86/kernel/cpu/mcheck/p4.c @@ -85,7 +85,7 @@ static void intel_init_thermal(struct cpuinfo_x86 *c) */ rdmsr(MSR_IA32_MISC_ENABLE, l, h); h = apic_read(APIC_LVTTHMR); - if ((l & (1<<3)) && (h & APIC_DM_SMI)) { + if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { printk(KERN_DEBUG "CPU%d: Thermal monitoring handled by SMI\n", cpu); return; /* -EBUSY */ @@ -111,7 +111,7 @@ static void intel_init_thermal(struct cpuinfo_x86 *c) vendor_thermal_interrupt = intel_thermal_interrupt; rdmsr(MSR_IA32_MISC_ENABLE, l, h); - wrmsr(MSR_IA32_MISC_ENABLE, l | (1<<3), h); + wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h); l = apic_read(APIC_LVTTHMR); apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); diff --git a/arch/x86/kernel/cpu/mcheck/threshold.c b/arch/x86/kernel/cpu/mcheck/threshold.c new file mode 100644 index 00000000000..23ee9e730f7 --- /dev/null +++ b/arch/x86/kernel/cpu/mcheck/threshold.c @@ -0,0 +1,29 @@ +/* + * Common corrected MCE threshold handler code: + */ +#include <linux/interrupt.h> +#include <linux/kernel.h> + +#include <asm/irq_vectors.h> +#include <asm/apic.h> +#include <asm/idle.h> +#include <asm/mce.h> + +static void default_threshold_interrupt(void) +{ + printk(KERN_ERR "Unexpected threshold interrupt at vector %x\n", + THRESHOLD_APIC_VECTOR); +} + +void (*mce_threshold_vector)(void) = default_threshold_interrupt; + +asmlinkage void mce_threshold_interrupt(void) +{ + exit_idle(); + irq_enter(); + inc_irq_stat(irq_threshold_count); + mce_threshold_vector(); + irq_exit(); + /* Ack only at the end to avoid potential reentry */ + ack_APIC_irq(); +} diff --git a/arch/x86/kernel/cpu/mtrr/Makefile b/arch/x86/kernel/cpu/mtrr/Makefile index 191fc053364..f4361b56f8e 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 +obj-y := main.o if.o generic.o state.o cleanup.o obj-$(CONFIG_X86_32) += amd.o cyrix.o centaur.o diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c new file mode 100644 index 00000000000..ce0fe4b5c04 --- /dev/null +++ b/arch/x86/kernel/cpu/mtrr/cleanup.c @@ -0,0 +1,1101 @@ +/* MTRR (Memory Type Range Register) cleanup + + Copyright (C) 2009 Yinghai Lu + + This library is free software; you can redistribute it and/or + modify it under the terms of the GNU Library General Public + License as published by the Free Software Foundation; either + version 2 of the License, or (at your option) any later version. + + This library 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 + Library General Public License for more details. + + You should have received a copy of the GNU Library General Public + License along with this library; if not, write to the Free + Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. +*/ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/smp.h> +#include <linux/cpu.h> +#include <linux/mutex.h> +#include <linux/sort.h> + +#include <asm/e820.h> +#include <asm/mtrr.h> +#include <asm/uaccess.h> +#include <asm/processor.h> +#include <asm/msr.h> +#include <asm/kvm_para.h> +#include "mtrr.h" + +/* should be related to MTRR_VAR_RANGES nums */ +#define RANGE_NUM 256 + +struct res_range { + unsigned long start; + unsigned long end; +}; + +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 that */ + 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; +} + +struct var_mtrr_range_state { + unsigned long base_pfn; + unsigned long size_pfn; + mtrr_type type; +}; + +static struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; +static int __initdata debug_print; + +static int __init +x86_get_mtrr_mem_range(struct res_range *range, int nr_range, + unsigned long extra_remove_base, + unsigned long extra_remove_size) +{ + unsigned long base, size; + mtrr_type type; + int i; + + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type != MTRR_TYPE_WRBACK) + 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); + } + 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); + } + + /* take out UC ranges */ + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type != MTRR_TYPE_UNCACHABLE && + type != MTRR_TYPE_WRPROT) + continue; + size = range_state[i].size_pfn; + if (!size) + continue; + base = range_state[i].base_pfn; + if (base < (1<<(20-PAGE_SHIFT)) && mtrr_state.have_fixed && + (mtrr_state.enabled & 1)) { + /* Var MTRR contains UC entry below 1M? Skip it: */ + printk(KERN_WARNING "WARNING: BIOS bug: VAR MTRR %d " + "contains strange UC entry under 1M, check " + "with your system vendor!\n", i); + if (base + size <= (1<<(20-PAGE_SHIFT))) + continue; + size -= (1<<(20-PAGE_SHIFT)) - base; + base = 1<<(20-PAGE_SHIFT); + } + subtract_range(range, base, base + size - 1); + } + if (extra_remove_size) + subtract_range(range, extra_remove_base, + extra_remove_base + extra_remove_size - 1); + + /* 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); + } + + /* sort the ranges */ + sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); + 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); + } + + /* clear those is not used */ + for (i = nr_range; i < RANGE_NUM; i++) + memset(&range[i], 0, sizeof(range[i])); + + return nr_range; +} + +static struct res_range __initdata range[RANGE_NUM]; +static int __initdata nr_range; + +#ifdef CONFIG_MTRR_SANITIZER + +static unsigned long __init sum_ranges(struct res_range *range, int nr_range) +{ + unsigned long sum; + int i; + + sum = 0; + for (i = 0; i < nr_range; i++) + sum += range[i].end + 1 - range[i].start; + + return sum; +} + +static int enable_mtrr_cleanup __initdata = + CONFIG_MTRR_SANITIZER_ENABLE_DEFAULT; + +static int __init disable_mtrr_cleanup_setup(char *str) +{ + enable_mtrr_cleanup = 0; + return 0; +} +early_param("disable_mtrr_cleanup", disable_mtrr_cleanup_setup); + +static int __init enable_mtrr_cleanup_setup(char *str) +{ + enable_mtrr_cleanup = 1; + return 0; +} +early_param("enable_mtrr_cleanup", enable_mtrr_cleanup_setup); + +static int __init mtrr_cleanup_debug_setup(char *str) +{ + debug_print = 1; + return 0; +} +early_param("mtrr_cleanup_debug", mtrr_cleanup_debug_setup); + +struct var_mtrr_state { + unsigned long range_startk; + unsigned long range_sizek; + unsigned long chunk_sizek; + unsigned long gran_sizek; + unsigned int reg; +}; + +static void __init +set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, + unsigned char type, unsigned int address_bits) +{ + u32 base_lo, base_hi, mask_lo, mask_hi; + u64 base, mask; + + if (!sizek) { + fill_mtrr_var_range(reg, 0, 0, 0, 0); + return; + } + + mask = (1ULL << address_bits) - 1; + mask &= ~((((u64)sizek) << 10) - 1); + + base = ((u64)basek) << 10; + + base |= type; + mask |= 0x800; + + base_lo = base & ((1ULL<<32) - 1); + base_hi = base >> 32; + + mask_lo = mask & ((1ULL<<32) - 1); + mask_hi = mask >> 32; + + fill_mtrr_var_range(reg, base_lo, base_hi, mask_lo, mask_hi); +} + +static void __init +save_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, + unsigned char type) +{ + range_state[reg].base_pfn = basek >> (PAGE_SHIFT - 10); + range_state[reg].size_pfn = sizek >> (PAGE_SHIFT - 10); + range_state[reg].type = type; +} + +static void __init +set_var_mtrr_all(unsigned int address_bits) +{ + unsigned long basek, sizek; + unsigned char type; + unsigned int reg; + + for (reg = 0; reg < num_var_ranges; reg++) { + basek = range_state[reg].base_pfn << (PAGE_SHIFT - 10); + sizek = range_state[reg].size_pfn << (PAGE_SHIFT - 10); + type = range_state[reg].type; + + set_var_mtrr(reg, basek, sizek, type, address_bits); + } +} + +static unsigned long to_size_factor(unsigned long sizek, char *factorp) +{ + char factor; + unsigned long base = sizek; + + if (base & ((1<<10) - 1)) { + /* not MB alignment */ + factor = 'K'; + } else if (base & ((1<<20) - 1)) { + factor = 'M'; + base >>= 10; + } else { + factor = 'G'; + base >>= 20; + } + + *factorp = factor; + + return base; +} + +static unsigned int __init +range_to_mtrr(unsigned int reg, unsigned long range_startk, + unsigned long range_sizek, unsigned char type) +{ + if (!range_sizek || (reg >= num_var_ranges)) + return reg; + + while (range_sizek) { + unsigned long max_align, align; + unsigned long sizek; + + /* Compute the maximum size I can make a range */ + if (range_startk) + max_align = ffs(range_startk) - 1; + else + max_align = 32; + align = fls(range_sizek) - 1; + if (align > max_align) + align = max_align; + + sizek = 1 << align; + if (debug_print) { + char start_factor = 'K', size_factor = 'K'; + unsigned long start_base, size_base; + + start_base = to_size_factor(range_startk, + &start_factor), + size_base = to_size_factor(sizek, &size_factor), + + printk(KERN_DEBUG "Setting variable MTRR %d, " + "base: %ld%cB, range: %ld%cB, type %s\n", + reg, start_base, start_factor, + size_base, size_factor, + (type == MTRR_TYPE_UNCACHABLE) ? "UC" : + ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other") + ); + } + save_var_mtrr(reg++, range_startk, sizek, type); + range_startk += sizek; + range_sizek -= sizek; + if (reg >= num_var_ranges) + break; + } + return reg; +} + +static unsigned __init +range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, + unsigned long sizek) +{ + unsigned long hole_basek, hole_sizek; + unsigned long second_basek, second_sizek; + unsigned long range0_basek, range0_sizek; + unsigned long range_basek, range_sizek; + unsigned long chunk_sizek; + unsigned long gran_sizek; + + hole_basek = 0; + hole_sizek = 0; + second_basek = 0; + second_sizek = 0; + chunk_sizek = state->chunk_sizek; + gran_sizek = state->gran_sizek; + + /* align with gran size, prevent small block used up MTRRs */ + range_basek = ALIGN(state->range_startk, gran_sizek); + if ((range_basek > basek) && basek) + return second_sizek; + state->range_sizek -= (range_basek - state->range_startk); + range_sizek = ALIGN(state->range_sizek, gran_sizek); + + while (range_sizek > state->range_sizek) { + range_sizek -= gran_sizek; + if (!range_sizek) + return 0; + } + state->range_sizek = range_sizek; + + /* try to append some small hole */ + range0_basek = state->range_startk; + range0_sizek = ALIGN(state->range_sizek, chunk_sizek); + + /* no increase */ + if (range0_sizek == state->range_sizek) { + if (debug_print) + printk(KERN_DEBUG "rangeX: %016lx - %016lx\n", + range0_basek<<10, + (range0_basek + state->range_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range0_basek, + state->range_sizek, MTRR_TYPE_WRBACK); + return 0; + } + + /* only cut back, when it is not the last */ + if (sizek) { + while (range0_basek + range0_sizek > (basek + sizek)) { + if (range0_sizek >= chunk_sizek) + range0_sizek -= chunk_sizek; + else + range0_sizek = 0; + + if (!range0_sizek) + break; + } + } + +second_try: + range_basek = range0_basek + range0_sizek; + + /* one hole in the middle */ + if (range_basek > basek && range_basek <= (basek + sizek)) + second_sizek = range_basek - basek; + + if (range0_sizek > state->range_sizek) { + + /* one hole in middle or at end */ + hole_sizek = range0_sizek - state->range_sizek - second_sizek; + + /* hole size should be less than half of range0 size */ + if (hole_sizek >= (range0_sizek >> 1) && + range0_sizek >= chunk_sizek) { + range0_sizek -= chunk_sizek; + second_sizek = 0; + hole_sizek = 0; + + goto second_try; + } + } + + if (range0_sizek) { + if (debug_print) + printk(KERN_DEBUG "range0: %016lx - %016lx\n", + range0_basek<<10, + (range0_basek + range0_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range0_basek, + range0_sizek, MTRR_TYPE_WRBACK); + } + + if (range0_sizek < state->range_sizek) { + /* need to handle left over */ + range_sizek = state->range_sizek - range0_sizek; + + if (debug_print) + printk(KERN_DEBUG "range: %016lx - %016lx\n", + range_basek<<10, + (range_basek + range_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range_basek, + range_sizek, MTRR_TYPE_WRBACK); + } + + if (hole_sizek) { + hole_basek = range_basek - hole_sizek - second_sizek; + if (debug_print) + printk(KERN_DEBUG "hole: %016lx - %016lx\n", + hole_basek<<10, + (hole_basek + hole_sizek)<<10); + state->reg = range_to_mtrr(state->reg, hole_basek, + hole_sizek, MTRR_TYPE_UNCACHABLE); + } + + return second_sizek; +} + +static void __init +set_var_mtrr_range(struct var_mtrr_state *state, unsigned long base_pfn, + unsigned long size_pfn) +{ + unsigned long basek, sizek; + unsigned long second_sizek = 0; + + if (state->reg >= num_var_ranges) + return; + + basek = base_pfn << (PAGE_SHIFT - 10); + sizek = size_pfn << (PAGE_SHIFT - 10); + + /* See if I can merge with the last range */ + if ((basek <= 1024) || + (state->range_startk + state->range_sizek == basek)) { + unsigned long endk = basek + sizek; + state->range_sizek = endk - state->range_startk; + return; + } + /* Write the range mtrrs */ + if (state->range_sizek != 0) + second_sizek = range_to_mtrr_with_hole(state, basek, sizek); + + /* Allocate an msr */ + state->range_startk = basek + second_sizek; + state->range_sizek = sizek - second_sizek; +} + +/* mininum size of mtrr block that can take hole */ +static u64 mtrr_chunk_size __initdata = (256ULL<<20); + +static int __init parse_mtrr_chunk_size_opt(char *p) +{ + if (!p) + return -EINVAL; + mtrr_chunk_size = memparse(p, &p); + return 0; +} +early_param("mtrr_chunk_size", parse_mtrr_chunk_size_opt); + +/* granity of mtrr of block */ +static u64 mtrr_gran_size __initdata; + +static int __init parse_mtrr_gran_size_opt(char *p) +{ + if (!p) + return -EINVAL; + mtrr_gran_size = memparse(p, &p); + return 0; +} +early_param("mtrr_gran_size", parse_mtrr_gran_size_opt); + +static int nr_mtrr_spare_reg __initdata = + CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT; + +static int __init parse_mtrr_spare_reg(char *arg) +{ + if (arg) + nr_mtrr_spare_reg = simple_strtoul(arg, NULL, 0); + return 0; +} + +early_param("mtrr_spare_reg_nr", parse_mtrr_spare_reg); + +static int __init +x86_setup_var_mtrrs(struct res_range *range, int nr_range, + u64 chunk_size, u64 gran_size) +{ + struct var_mtrr_state var_state; + int i; + int num_reg; + + var_state.range_startk = 0; + var_state.range_sizek = 0; + var_state.reg = 0; + var_state.chunk_sizek = chunk_size >> 10; + var_state.gran_sizek = gran_size >> 10; + + memset(range_state, 0, sizeof(range_state)); + + /* Write the range etc */ + for (i = 0; i < nr_range; i++) + set_var_mtrr_range(&var_state, range[i].start, + range[i].end - range[i].start + 1); + + /* Write the last range */ + if (var_state.range_sizek != 0) + range_to_mtrr_with_hole(&var_state, 0, 0); + + num_reg = var_state.reg; + /* Clear out the extra MTRR's */ + while (var_state.reg < num_var_ranges) { + save_var_mtrr(var_state.reg, 0, 0, 0); + var_state.reg++; + } + + return num_reg; +} + +struct mtrr_cleanup_result { + unsigned long gran_sizek; + unsigned long chunk_sizek; + unsigned long lose_cover_sizek; + unsigned int num_reg; + int bad; +}; + +/* + * gran_size: 64K, 128K, 256K, 512K, 1M, 2M, ..., 2G + * chunk size: gran_size, ..., 2G + * so we need (1+16)*8 + */ +#define NUM_RESULT 136 +#define PSHIFT (PAGE_SHIFT - 10) + +static struct mtrr_cleanup_result __initdata result[NUM_RESULT]; +static unsigned long __initdata min_loss_pfn[RANGE_NUM]; + +static void __init print_out_mtrr_range_state(void) +{ + int i; + char start_factor = 'K', size_factor = 'K'; + unsigned long start_base, size_base; + mtrr_type type; + + for (i = 0; i < num_var_ranges; i++) { + + size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10); + if (!size_base) + continue; + + size_base = to_size_factor(size_base, &size_factor), + start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10); + start_base = to_size_factor(start_base, &start_factor), + type = range_state[i].type; + + printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n", + i, start_base, start_factor, + size_base, size_factor, + (type == MTRR_TYPE_UNCACHABLE) ? "UC" : + ((type == MTRR_TYPE_WRPROT) ? "WP" : + ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other")) + ); + } +} + +static int __init mtrr_need_cleanup(void) +{ + int i; + mtrr_type type; + unsigned long size; + /* extra one for all 0 */ + int num[MTRR_NUM_TYPES + 1]; + + /* check entries number */ + memset(num, 0, sizeof(num)); + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + size = range_state[i].size_pfn; + if (type >= MTRR_NUM_TYPES) + continue; + if (!size) + type = MTRR_NUM_TYPES; + if (type == MTRR_TYPE_WRPROT) + type = MTRR_TYPE_UNCACHABLE; + num[type]++; + } + + /* check if we got UC entries */ + if (!num[MTRR_TYPE_UNCACHABLE]) + return 0; + + /* check if we only had WB and UC */ + if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != + num_var_ranges - num[MTRR_NUM_TYPES]) + return 0; + + return 1; +} + +static unsigned long __initdata range_sums; +static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size, + unsigned long extra_remove_base, + unsigned long extra_remove_size, + int i) +{ + int num_reg; + static struct res_range range_new[RANGE_NUM]; + static int nr_range_new; + unsigned long range_sums_new; + + /* convert ranges to var ranges state */ + num_reg = x86_setup_var_mtrrs(range, nr_range, + chunk_size, gran_size); + + /* we got new setting in range_state, check it */ + memset(range_new, 0, sizeof(range_new)); + nr_range_new = x86_get_mtrr_mem_range(range_new, 0, + extra_remove_base, extra_remove_size); + range_sums_new = sum_ranges(range_new, nr_range_new); + + result[i].chunk_sizek = chunk_size >> 10; + result[i].gran_sizek = gran_size >> 10; + result[i].num_reg = num_reg; + if (range_sums < range_sums_new) { + result[i].lose_cover_sizek = + (range_sums_new - range_sums) << PSHIFT; + result[i].bad = 1; + } else + result[i].lose_cover_sizek = + (range_sums - range_sums_new) << PSHIFT; + + /* double check it */ + if (!result[i].bad && !result[i].lose_cover_sizek) { + if (nr_range_new != nr_range || + memcmp(range, range_new, sizeof(range))) + result[i].bad = 1; + } + + if (!result[i].bad && (range_sums - range_sums_new < + min_loss_pfn[num_reg])) { + min_loss_pfn[num_reg] = + range_sums - range_sums_new; + } +} + +static void __init mtrr_print_out_one_result(int i) +{ + char gran_factor, chunk_factor, lose_factor; + unsigned long gran_base, chunk_base, lose_base; + + gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), + chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), + lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), + printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", + result[i].bad ? "*BAD*" : " ", + gran_base, gran_factor, chunk_base, chunk_factor); + printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", + result[i].num_reg, result[i].bad ? "-" : "", + lose_base, lose_factor); +} + +static int __init mtrr_search_optimal_index(void) +{ + int i; + int num_reg_good; + int index_good; + + if (nr_mtrr_spare_reg >= num_var_ranges) + nr_mtrr_spare_reg = num_var_ranges - 1; + num_reg_good = -1; + for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) { + if (!min_loss_pfn[i]) + num_reg_good = i; + } + + index_good = -1; + if (num_reg_good != -1) { + for (i = 0; i < NUM_RESULT; i++) { + if (!result[i].bad && + result[i].num_reg == num_reg_good && + !result[i].lose_cover_sizek) { + index_good = i; + break; + } + } + } + + return index_good; +} + + +int __init mtrr_cleanup(unsigned address_bits) +{ + unsigned long extra_remove_base, extra_remove_size; + unsigned long base, size, def, dummy; + mtrr_type type; + u64 chunk_size, gran_size; + int index_good; + int i; + + if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1) + return 0; + rdmsr(MTRRdefType_MSR, def, dummy); + def &= 0xff; + if (def != MTRR_TYPE_UNCACHABLE) + return 0; + + /* get it and store it aside */ + memset(range_state, 0, sizeof(range_state)); + for (i = 0; i < num_var_ranges; i++) { + mtrr_if->get(i, &base, &size, &type); + range_state[i].base_pfn = base; + range_state[i].size_pfn = size; + range_state[i].type = type; + } + + /* check if we need handle it and can handle it */ + if (!mtrr_need_cleanup()) + return 0; + + /* print original var MTRRs at first, for debugging: */ + printk(KERN_DEBUG "original variable MTRRs\n"); + print_out_mtrr_range_state(); + + memset(range, 0, sizeof(range)); + extra_remove_size = 0; + extra_remove_base = 1 << (32 - PAGE_SHIFT); + if (mtrr_tom2) + extra_remove_size = + (mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base; + nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base, + extra_remove_size); + /* + * [0, 1M) should always be coverred by var mtrr with WB + * and fixed mtrrs should take effective before var mtrr for it + */ + nr_range = add_range_with_merge(range, nr_range, 0, + (1ULL<<(20 - PAGE_SHIFT)) - 1); + /* sort the ranges */ + sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); + + range_sums = sum_ranges(range, nr_range); + printk(KERN_INFO "total RAM coverred: %ldM\n", + range_sums >> (20 - PAGE_SHIFT)); + + if (mtrr_chunk_size && mtrr_gran_size) { + i = 0; + mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size, + extra_remove_base, extra_remove_size, i); + + mtrr_print_out_one_result(i); + + if (!result[i].bad) { + set_var_mtrr_all(address_bits); + printk(KERN_DEBUG "New variable MTRRs\n"); + print_out_mtrr_range_state(); + return 1; + } + printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, " + "will find optimal one\n"); + } + + i = 0; + memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn)); + memset(result, 0, sizeof(result)); + for (gran_size = (1ULL<<16); gran_size < (1ULL<<32); gran_size <<= 1) { + + for (chunk_size = gran_size; chunk_size < (1ULL<<32); + chunk_size <<= 1) { + + if (i >= NUM_RESULT) + continue; + + mtrr_calc_range_state(chunk_size, gran_size, + extra_remove_base, extra_remove_size, i); + if (debug_print) { + mtrr_print_out_one_result(i); + printk(KERN_INFO "\n"); + } + + i++; + } + } + + /* try to find the optimal index */ + index_good = mtrr_search_optimal_index(); + + if (index_good != -1) { + printk(KERN_INFO "Found optimal setting for mtrr clean up\n"); + i = index_good; + mtrr_print_out_one_result(i); + + /* convert ranges to var ranges state */ + chunk_size = result[i].chunk_sizek; + chunk_size <<= 10; + gran_size = result[i].gran_sizek; + gran_size <<= 10; + x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size); + set_var_mtrr_all(address_bits); + printk(KERN_DEBUG "New variable MTRRs\n"); + print_out_mtrr_range_state(); + return 1; + } else { + /* print out all */ + for (i = 0; i < NUM_RESULT; i++) + mtrr_print_out_one_result(i); + } + + printk(KERN_INFO "mtrr_cleanup: can not find optimal value\n"); + printk(KERN_INFO "please specify mtrr_gran_size/mtrr_chunk_size\n"); + + return 0; +} +#else +int __init mtrr_cleanup(unsigned address_bits) +{ + return 0; +} +#endif + +static int disable_mtrr_trim; + +static int __init disable_mtrr_trim_setup(char *str) +{ + disable_mtrr_trim = 1; + return 0; +} +early_param("disable_mtrr_trim", disable_mtrr_trim_setup); + +/* + * Newer AMD K8s and later CPUs have a special magic MSR way to force WB + * for memory >4GB. Check for that here. + * Note this won't check if the MTRRs < 4GB where the magic bit doesn't + * apply to are wrong, but so far we don't know of any such case in the wild. + */ +#define Tom2Enabled (1U << 21) +#define Tom2ForceMemTypeWB (1U << 22) + +int __init amd_special_default_mtrr(void) +{ + u32 l, h; + + if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) + return 0; + if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11) + return 0; + /* In case some hypervisor doesn't pass SYSCFG through */ + if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0) + return 0; + /* + * Memory between 4GB and top of mem is forced WB by this magic bit. + * Reserved before K8RevF, but should be zero there. + */ + if ((l & (Tom2Enabled | Tom2ForceMemTypeWB)) == + (Tom2Enabled | Tom2ForceMemTypeWB)) + return 1; + return 0; +} + +static u64 __init real_trim_memory(unsigned long start_pfn, + unsigned long limit_pfn) +{ + u64 trim_start, trim_size; + trim_start = start_pfn; + trim_start <<= PAGE_SHIFT; + trim_size = limit_pfn; + trim_size <<= PAGE_SHIFT; + trim_size -= trim_start; + + return e820_update_range(trim_start, trim_size, E820_RAM, + E820_RESERVED); +} +/** + * mtrr_trim_uncached_memory - trim RAM not covered by MTRRs + * @end_pfn: ending page frame number + * + * Some buggy BIOSes don't setup the MTRRs properly for systems with certain + * memory configurations. This routine checks that the highest MTRR matches + * the end of memory, to make sure the MTRRs having a write back type cover + * all of the memory the kernel is intending to use. If not, it'll trim any + * memory off the end by adjusting end_pfn, removing it from the kernel's + * allocation pools, warning the user with an obnoxious message. + */ +int __init mtrr_trim_uncached_memory(unsigned long end_pfn) +{ + unsigned long i, base, size, highest_pfn = 0, def, dummy; + mtrr_type type; + u64 total_trim_size; + + /* extra one for all 0 */ + int num[MTRR_NUM_TYPES + 1]; + /* + * Make sure we only trim uncachable memory on machines that + * support the Intel MTRR architecture: + */ + if (!is_cpu(INTEL) || disable_mtrr_trim) + return 0; + rdmsr(MTRRdefType_MSR, def, dummy); + def &= 0xff; + if (def != MTRR_TYPE_UNCACHABLE) + return 0; + + /* get it and store it aside */ + memset(range_state, 0, sizeof(range_state)); + for (i = 0; i < num_var_ranges; i++) { + mtrr_if->get(i, &base, &size, &type); + range_state[i].base_pfn = base; + range_state[i].size_pfn = size; + range_state[i].type = type; + } + + /* Find highest cached pfn */ + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type != MTRR_TYPE_WRBACK) + continue; + base = range_state[i].base_pfn; + size = range_state[i].size_pfn; + if (highest_pfn < base + size) + highest_pfn = base + size; + } + + /* kvm/qemu doesn't have mtrr set right, don't trim them all */ + if (!highest_pfn) { + printk(KERN_INFO "CPU MTRRs all blank - virtualized system.\n"); + return 0; + } + + /* check entries number */ + memset(num, 0, sizeof(num)); + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type >= MTRR_NUM_TYPES) + continue; + size = range_state[i].size_pfn; + if (!size) + type = MTRR_NUM_TYPES; + num[type]++; + } + + /* no entry for WB? */ + if (!num[MTRR_TYPE_WRBACK]) + return 0; + + /* check if we only had WB and UC */ + if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != + num_var_ranges - num[MTRR_NUM_TYPES]) + return 0; + + memset(range, 0, sizeof(range)); + 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; + nr_range++; + } + nr_range = x86_get_mtrr_mem_range(range, nr_range, 0, 0); + + total_trim_size = 0; + /* check the head */ + if (range[0].start) + total_trim_size += real_trim_memory(0, range[0].start); + /* 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, + 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, + end_pfn); + + if (total_trim_size) { + printk(KERN_WARNING "WARNING: BIOS bug: CPU MTRRs don't cover" + " all of memory, losing %lluMB of RAM.\n", + total_trim_size >> 20); + + if (!changed_by_mtrr_cleanup) + WARN_ON(1); + + printk(KERN_INFO "update e820 for mtrr\n"); + update_e820(); + + return 1; + } + + return 0; +} + diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 0c0a455fe95..37f28fc7cf9 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -33,13 +33,31 @@ u64 mtrr_tom2; struct mtrr_state_type mtrr_state = {}; EXPORT_SYMBOL_GPL(mtrr_state); -static int __initdata mtrr_show; -static int __init mtrr_debug(char *opt) +/** + * BIOS is expected to clear MtrrFixDramModEn bit, see for example + * "BIOS and Kernel Developer's Guide for the AMD Athlon 64 and AMD + * Opteron Processors" (26094 Rev. 3.30 February 2006), section + * "13.2.1.2 SYSCFG Register": "The MtrrFixDramModEn bit should be set + * to 1 during BIOS initalization of the fixed MTRRs, then cleared to + * 0 for operation." + */ +static inline void k8_check_syscfg_dram_mod_en(void) { - mtrr_show = 1; - return 0; + u32 lo, hi; + + if (!((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && + (boot_cpu_data.x86 >= 0x0f))) + return; + + rdmsr(MSR_K8_SYSCFG, lo, hi); + if (lo & K8_MTRRFIXRANGE_DRAM_MODIFY) { + printk(KERN_ERR FW_WARN "MTRR: CPU %u: SYSCFG[MtrrFixDramModEn]" + " not cleared by BIOS, clearing this bit\n", + smp_processor_id()); + lo &= ~K8_MTRRFIXRANGE_DRAM_MODIFY; + mtrr_wrmsr(MSR_K8_SYSCFG, lo, hi); + } } -early_param("mtrr.show", mtrr_debug); /* * Returns the effective MTRR type for the region @@ -174,6 +192,8 @@ get_fixed_ranges(mtrr_type * frs) unsigned int *p = (unsigned int *) frs; int i; + k8_check_syscfg_dram_mod_en(); + rdmsr(MTRRfix64K_00000_MSR, p[0], p[1]); for (i = 0; i < 2; i++) @@ -188,18 +208,94 @@ void mtrr_save_fixed_ranges(void *info) get_fixed_ranges(mtrr_state.fixed_ranges); } -static void print_fixed(unsigned base, unsigned step, const mtrr_type*types) +static unsigned __initdata last_fixed_start; +static unsigned __initdata last_fixed_end; +static mtrr_type __initdata last_fixed_type; + +static void __init print_fixed_last(void) +{ + if (!last_fixed_end) + return; + + printk(KERN_DEBUG " %05X-%05X %s\n", last_fixed_start, + last_fixed_end - 1, mtrr_attrib_to_str(last_fixed_type)); + + last_fixed_end = 0; +} + +static void __init update_fixed_last(unsigned base, unsigned end, + mtrr_type type) +{ + last_fixed_start = base; + last_fixed_end = end; + last_fixed_type = type; +} + +static void __init print_fixed(unsigned base, unsigned step, + const mtrr_type *types) { unsigned i; - for (i = 0; i < 8; ++i, ++types, base += step) - printk(KERN_INFO "MTRR %05X-%05X %s\n", - base, base + step - 1, mtrr_attrib_to_str(*types)); + for (i = 0; i < 8; ++i, ++types, base += step) { + if (last_fixed_end == 0) { + update_fixed_last(base, base + step, *types); + continue; + } + if (last_fixed_end == base && last_fixed_type == *types) { + last_fixed_end = base + step; + continue; + } + /* new segments: gap or different type */ + print_fixed_last(); + update_fixed_last(base, base + step, *types); + } } static void prepare_set(void); static void post_set(void); +static void __init print_mtrr_state(void) +{ + unsigned int i; + int high_width; + + printk(KERN_DEBUG "MTRR default type: %s\n", + mtrr_attrib_to_str(mtrr_state.def_type)); + if (mtrr_state.have_fixed) { + printk(KERN_DEBUG "MTRR fixed ranges %sabled:\n", + mtrr_state.enabled & 1 ? "en" : "dis"); + print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0); + for (i = 0; i < 2; ++i) + print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8); + for (i = 0; i < 8; ++i) + print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8); + + /* tail */ + print_fixed_last(); + } + printk(KERN_DEBUG "MTRR variable ranges %sabled:\n", + mtrr_state.enabled & 2 ? "en" : "dis"); + high_width = ((size_or_mask ? ffs(size_or_mask) - 1 : 32) - (32 - PAGE_SHIFT) + 3) / 4; + for (i = 0; i < num_var_ranges; ++i) { + if (mtrr_state.var_ranges[i].mask_lo & (1 << 11)) + printk(KERN_DEBUG " %u base %0*X%05X000 mask %0*X%05X000 %s\n", + i, + high_width, + mtrr_state.var_ranges[i].base_hi, + mtrr_state.var_ranges[i].base_lo >> 12, + high_width, + mtrr_state.var_ranges[i].mask_hi, + mtrr_state.var_ranges[i].mask_lo >> 12, + mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff)); + else + printk(KERN_DEBUG " %u disabled\n", i); + } + if (mtrr_tom2) { + printk(KERN_DEBUG "TOM2: %016llx aka %lldM\n", + mtrr_tom2, mtrr_tom2>>20); + } +} + /* Grab all of the MTRR state for this CPU into *state */ void __init get_mtrr_state(void) { @@ -231,41 +327,9 @@ void __init get_mtrr_state(void) mtrr_tom2 |= low; mtrr_tom2 &= 0xffffff800000ULL; } - if (mtrr_show) { - int high_width; - - printk(KERN_INFO "MTRR default type: %s\n", mtrr_attrib_to_str(mtrr_state.def_type)); - if (mtrr_state.have_fixed) { - printk(KERN_INFO "MTRR fixed ranges %sabled:\n", - mtrr_state.enabled & 1 ? "en" : "dis"); - print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0); - for (i = 0; i < 2; ++i) - print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8); - for (i = 0; i < 8; ++i) - print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8); - } - printk(KERN_INFO "MTRR variable ranges %sabled:\n", - mtrr_state.enabled & 2 ? "en" : "dis"); - high_width = ((size_or_mask ? ffs(size_or_mask) - 1 : 32) - (32 - PAGE_SHIFT) + 3) / 4; - for (i = 0; i < num_var_ranges; ++i) { - if (mtrr_state.var_ranges[i].mask_lo & (1 << 11)) - printk(KERN_INFO "MTRR %u base %0*X%05X000 mask %0*X%05X000 %s\n", - i, - high_width, - mtrr_state.var_ranges[i].base_hi, - mtrr_state.var_ranges[i].base_lo >> 12, - high_width, - mtrr_state.var_ranges[i].mask_hi, - mtrr_state.var_ranges[i].mask_lo >> 12, - mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff)); - else - printk(KERN_INFO "MTRR %u disabled\n", i); - } - if (mtrr_tom2) { - printk(KERN_INFO "TOM2: %016llx aka %lldM\n", - mtrr_tom2, mtrr_tom2>>20); - } - } + + print_mtrr_state(); + mtrr_state_set = 1; /* PAT setup for BP. We need to go through sync steps here */ @@ -308,27 +372,10 @@ void mtrr_wrmsr(unsigned msr, unsigned a, unsigned b) } /** - * Enable and allow read/write of extended fixed-range MTRR bits on K8 CPUs - * see AMD publication no. 24593, chapter 3.2.1 for more information - */ -static inline void k8_enable_fixed_iorrs(void) -{ - unsigned lo, hi; - - rdmsr(MSR_K8_SYSCFG, lo, hi); - mtrr_wrmsr(MSR_K8_SYSCFG, lo - | K8_MTRRFIXRANGE_DRAM_ENABLE - | K8_MTRRFIXRANGE_DRAM_MODIFY, hi); -} - -/** * set_fixed_range - checks & updates a fixed-range MTRR if it differs from the value it should have * @msr: MSR address of the MTTR which should be checked and updated * @changed: pointer which indicates whether the MTRR needed to be changed * @msrwords: pointer to the MSR values which the MSR should have - * - * If K8 extentions are wanted, update the K8 SYSCFG MSR also. - * See AMD publication no. 24593, chapter 7.8.1, page 233 for more information. */ static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords) { @@ -337,10 +384,6 @@ static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords) rdmsr(msr, lo, hi); if (lo != msrwords[0] || hi != msrwords[1]) { - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && - (boot_cpu_data.x86 >= 0x0f && boot_cpu_data.x86 <= 0x11) && - ((msrwords[0] | msrwords[1]) & K8_MTRR_RDMEM_WRMEM_MASK)) - k8_enable_fixed_iorrs(); mtrr_wrmsr(msr, msrwords[0], msrwords[1]); *changed = true; } @@ -376,22 +419,31 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base, { unsigned int mask_lo, mask_hi, base_lo, base_hi; unsigned int tmp, hi; + int cpu; + + /* + * get_mtrr doesn't need to update mtrr_state, also it could be called + * from any cpu, so try to print it out directly. + */ + cpu = get_cpu(); rdmsr(MTRRphysMask_MSR(reg), mask_lo, mask_hi); + if ((mask_lo & 0x800) == 0) { /* Invalid (i.e. free) range */ *base = 0; *size = 0; *type = 0; - return; + goto out_put_cpu; } rdmsr(MTRRphysBase_MSR(reg), base_lo, base_hi); - /* Work out the shifted address mask. */ + /* Work out the shifted address mask: */ tmp = mask_hi << (32 - PAGE_SHIFT) | mask_lo >> PAGE_SHIFT; mask_lo = size_or_mask | tmp; - /* Expand tmp with high bits to all 1s*/ + + /* Expand tmp with high bits to all 1s: */ hi = fls(tmp); if (hi > 0) { tmp |= ~((1<<(hi - 1)) - 1); @@ -402,11 +454,19 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base, } } - /* This works correctly if size is a power of two, i.e. a - contiguous range. */ + /* + * This works correctly if size is a power of two, i.e. a + * contiguous range: + */ *size = -mask_lo; *base = base_hi << (32 - PAGE_SHIFT) | base_lo >> PAGE_SHIFT; *type = base_lo & 0xff; + + printk(KERN_DEBUG " get_mtrr: cpu%d reg%02d base=%010lx size=%010lx %s\n", + cpu, reg, *base, *size, + mtrr_attrib_to_str(*type & 0xff)); +out_put_cpu: + put_cpu(); } /** @@ -419,6 +479,8 @@ static int set_fixed_ranges(mtrr_type * frs) bool changed = false; int block=-1, range; + k8_check_syscfg_dram_mod_en(); + while (fixed_range_blocks[++block].ranges) for (range=0; range < fixed_range_blocks[block].ranges; range++) set_fixed_range(fixed_range_blocks[block].base_msr + range, diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index 236a401b825..03cda01f57c 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -574,7 +574,7 @@ struct mtrr_value { unsigned long lsize; }; -static struct mtrr_value mtrr_state[MTRR_MAX_VAR_RANGES]; +static struct mtrr_value mtrr_value[MTRR_MAX_VAR_RANGES]; static int mtrr_save(struct sys_device * sysdev, pm_message_t state) { @@ -582,9 +582,9 @@ static int mtrr_save(struct sys_device * sysdev, pm_message_t state) for (i = 0; i < num_var_ranges; i++) { mtrr_if->get(i, - &mtrr_state[i].lbase, - &mtrr_state[i].lsize, - &mtrr_state[i].ltype); + &mtrr_value[i].lbase, + &mtrr_value[i].lsize, + &mtrr_value[i].ltype); } return 0; } @@ -594,11 +594,11 @@ static int mtrr_restore(struct sys_device * sysdev) int i; for (i = 0; i < num_var_ranges; i++) { - if (mtrr_state[i].lsize) + if (mtrr_value[i].lsize) set_mtrr(i, - mtrr_state[i].lbase, - mtrr_state[i].lsize, - mtrr_state[i].ltype); + mtrr_value[i].lbase, + mtrr_value[i].lsize, + mtrr_value[i].ltype); } return 0; } @@ -610,1058 +610,7 @@ static struct sysdev_driver mtrr_sysdev_driver = { .resume = mtrr_restore, }; -/* should be related to MTRR_VAR_RANGES nums */ -#define RANGE_NUM 256 - -struct res_range { - unsigned long start; - unsigned long end; -}; - -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 that */ - 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; -} - -struct var_mtrr_range_state { - unsigned long base_pfn; - unsigned long size_pfn; - mtrr_type type; -}; - -static struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; -static int __initdata debug_print; - -static int __init -x86_get_mtrr_mem_range(struct res_range *range, int nr_range, - unsigned long extra_remove_base, - unsigned long extra_remove_size) -{ - unsigned long i, base, size; - mtrr_type type; - - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - if (type != MTRR_TYPE_WRBACK) - 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); - } - 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); - } - - /* take out UC ranges */ - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - if (type != MTRR_TYPE_UNCACHABLE && - type != MTRR_TYPE_WRPROT) - continue; - size = range_state[i].size_pfn; - if (!size) - continue; - base = range_state[i].base_pfn; - subtract_range(range, base, base + size - 1); - } - if (extra_remove_size) - subtract_range(range, extra_remove_base, - extra_remove_base + extra_remove_size - 1); - - /* 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); - } - - /* sort the ranges */ - sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); - 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); - } - - /* clear those is not used */ - for (i = nr_range; i < RANGE_NUM; i++) - memset(&range[i], 0, sizeof(range[i])); - - return nr_range; -} - -static struct res_range __initdata range[RANGE_NUM]; -static int __initdata nr_range; - -#ifdef CONFIG_MTRR_SANITIZER - -static unsigned long __init sum_ranges(struct res_range *range, int nr_range) -{ - unsigned long sum; - int i; - - sum = 0; - for (i = 0; i < nr_range; i++) - sum += range[i].end + 1 - range[i].start; - - return sum; -} - -static int enable_mtrr_cleanup __initdata = - CONFIG_MTRR_SANITIZER_ENABLE_DEFAULT; - -static int __init disable_mtrr_cleanup_setup(char *str) -{ - enable_mtrr_cleanup = 0; - return 0; -} -early_param("disable_mtrr_cleanup", disable_mtrr_cleanup_setup); - -static int __init enable_mtrr_cleanup_setup(char *str) -{ - enable_mtrr_cleanup = 1; - return 0; -} -early_param("enable_mtrr_cleanup", enable_mtrr_cleanup_setup); - -static int __init mtrr_cleanup_debug_setup(char *str) -{ - debug_print = 1; - return 0; -} -early_param("mtrr_cleanup_debug", mtrr_cleanup_debug_setup); - -struct var_mtrr_state { - unsigned long range_startk; - unsigned long range_sizek; - unsigned long chunk_sizek; - unsigned long gran_sizek; - unsigned int reg; -}; - -static void __init -set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, - unsigned char type, unsigned int address_bits) -{ - u32 base_lo, base_hi, mask_lo, mask_hi; - u64 base, mask; - - if (!sizek) { - fill_mtrr_var_range(reg, 0, 0, 0, 0); - return; - } - - mask = (1ULL << address_bits) - 1; - mask &= ~((((u64)sizek) << 10) - 1); - - base = ((u64)basek) << 10; - - base |= type; - mask |= 0x800; - - base_lo = base & ((1ULL<<32) - 1); - base_hi = base >> 32; - - mask_lo = mask & ((1ULL<<32) - 1); - mask_hi = mask >> 32; - - fill_mtrr_var_range(reg, base_lo, base_hi, mask_lo, mask_hi); -} - -static void __init -save_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, - unsigned char type) -{ - range_state[reg].base_pfn = basek >> (PAGE_SHIFT - 10); - range_state[reg].size_pfn = sizek >> (PAGE_SHIFT - 10); - range_state[reg].type = type; -} - -static void __init -set_var_mtrr_all(unsigned int address_bits) -{ - unsigned long basek, sizek; - unsigned char type; - unsigned int reg; - - for (reg = 0; reg < num_var_ranges; reg++) { - basek = range_state[reg].base_pfn << (PAGE_SHIFT - 10); - sizek = range_state[reg].size_pfn << (PAGE_SHIFT - 10); - type = range_state[reg].type; - - set_var_mtrr(reg, basek, sizek, type, address_bits); - } -} - -static unsigned long to_size_factor(unsigned long sizek, char *factorp) -{ - char factor; - unsigned long base = sizek; - - if (base & ((1<<10) - 1)) { - /* not MB alignment */ - factor = 'K'; - } else if (base & ((1<<20) - 1)){ - factor = 'M'; - base >>= 10; - } else { - factor = 'G'; - base >>= 20; - } - - *factorp = factor; - - return base; -} - -static unsigned int __init -range_to_mtrr(unsigned int reg, unsigned long range_startk, - unsigned long range_sizek, unsigned char type) -{ - if (!range_sizek || (reg >= num_var_ranges)) - return reg; - - while (range_sizek) { - unsigned long max_align, align; - unsigned long sizek; - - /* Compute the maximum size I can make a range */ - if (range_startk) - max_align = ffs(range_startk) - 1; - else - max_align = 32; - align = fls(range_sizek) - 1; - if (align > max_align) - align = max_align; - - sizek = 1 << align; - if (debug_print) { - char start_factor = 'K', size_factor = 'K'; - unsigned long start_base, size_base; - - start_base = to_size_factor(range_startk, &start_factor), - size_base = to_size_factor(sizek, &size_factor), - - printk(KERN_DEBUG "Setting variable MTRR %d, " - "base: %ld%cB, range: %ld%cB, type %s\n", - reg, start_base, start_factor, - size_base, size_factor, - (type == MTRR_TYPE_UNCACHABLE)?"UC": - ((type == MTRR_TYPE_WRBACK)?"WB":"Other") - ); - } - save_var_mtrr(reg++, range_startk, sizek, type); - range_startk += sizek; - range_sizek -= sizek; - if (reg >= num_var_ranges) - break; - } - return reg; -} - -static unsigned __init -range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, - unsigned long sizek) -{ - unsigned long hole_basek, hole_sizek; - unsigned long second_basek, second_sizek; - unsigned long range0_basek, range0_sizek; - unsigned long range_basek, range_sizek; - unsigned long chunk_sizek; - unsigned long gran_sizek; - - hole_basek = 0; - hole_sizek = 0; - second_basek = 0; - second_sizek = 0; - chunk_sizek = state->chunk_sizek; - gran_sizek = state->gran_sizek; - - /* align with gran size, prevent small block used up MTRRs */ - range_basek = ALIGN(state->range_startk, gran_sizek); - if ((range_basek > basek) && basek) - return second_sizek; - state->range_sizek -= (range_basek - state->range_startk); - range_sizek = ALIGN(state->range_sizek, gran_sizek); - - while (range_sizek > state->range_sizek) { - range_sizek -= gran_sizek; - if (!range_sizek) - return 0; - } - state->range_sizek = range_sizek; - - /* try to append some small hole */ - range0_basek = state->range_startk; - range0_sizek = ALIGN(state->range_sizek, chunk_sizek); - - /* no increase */ - if (range0_sizek == state->range_sizek) { - if (debug_print) - printk(KERN_DEBUG "rangeX: %016lx - %016lx\n", - range0_basek<<10, - (range0_basek + state->range_sizek)<<10); - state->reg = range_to_mtrr(state->reg, range0_basek, - state->range_sizek, MTRR_TYPE_WRBACK); - return 0; - } - - /* only cut back, when it is not the last */ - if (sizek) { - while (range0_basek + range0_sizek > (basek + sizek)) { - if (range0_sizek >= chunk_sizek) - range0_sizek -= chunk_sizek; - else - range0_sizek = 0; - - if (!range0_sizek) - break; - } - } - -second_try: - range_basek = range0_basek + range0_sizek; - - /* one hole in the middle */ - if (range_basek > basek && range_basek <= (basek + sizek)) - second_sizek = range_basek - basek; - - if (range0_sizek > state->range_sizek) { - - /* one hole in middle or at end */ - hole_sizek = range0_sizek - state->range_sizek - second_sizek; - - /* hole size should be less than half of range0 size */ - if (hole_sizek >= (range0_sizek >> 1) && - range0_sizek >= chunk_sizek) { - range0_sizek -= chunk_sizek; - second_sizek = 0; - hole_sizek = 0; - - goto second_try; - } - } - - if (range0_sizek) { - if (debug_print) - printk(KERN_DEBUG "range0: %016lx - %016lx\n", - range0_basek<<10, - (range0_basek + range0_sizek)<<10); - state->reg = range_to_mtrr(state->reg, range0_basek, - range0_sizek, MTRR_TYPE_WRBACK); - } - - if (range0_sizek < state->range_sizek) { - /* need to handle left over */ - range_sizek = state->range_sizek - range0_sizek; - - if (debug_print) - printk(KERN_DEBUG "range: %016lx - %016lx\n", - range_basek<<10, - (range_basek + range_sizek)<<10); - state->reg = range_to_mtrr(state->reg, range_basek, - range_sizek, MTRR_TYPE_WRBACK); - } - - if (hole_sizek) { - hole_basek = range_basek - hole_sizek - second_sizek; - if (debug_print) - printk(KERN_DEBUG "hole: %016lx - %016lx\n", - hole_basek<<10, - (hole_basek + hole_sizek)<<10); - state->reg = range_to_mtrr(state->reg, hole_basek, - hole_sizek, MTRR_TYPE_UNCACHABLE); - } - - return second_sizek; -} - -static void __init -set_var_mtrr_range(struct var_mtrr_state *state, unsigned long base_pfn, - unsigned long size_pfn) -{ - unsigned long basek, sizek; - unsigned long second_sizek = 0; - - if (state->reg >= num_var_ranges) - return; - - basek = base_pfn << (PAGE_SHIFT - 10); - sizek = size_pfn << (PAGE_SHIFT - 10); - - /* See if I can merge with the last range */ - if ((basek <= 1024) || - (state->range_startk + state->range_sizek == basek)) { - unsigned long endk = basek + sizek; - state->range_sizek = endk - state->range_startk; - return; - } - /* Write the range mtrrs */ - if (state->range_sizek != 0) - second_sizek = range_to_mtrr_with_hole(state, basek, sizek); - - /* Allocate an msr */ - state->range_startk = basek + second_sizek; - state->range_sizek = sizek - second_sizek; -} - -/* mininum size of mtrr block that can take hole */ -static u64 mtrr_chunk_size __initdata = (256ULL<<20); - -static int __init parse_mtrr_chunk_size_opt(char *p) -{ - if (!p) - return -EINVAL; - mtrr_chunk_size = memparse(p, &p); - return 0; -} -early_param("mtrr_chunk_size", parse_mtrr_chunk_size_opt); - -/* granity of mtrr of block */ -static u64 mtrr_gran_size __initdata; - -static int __init parse_mtrr_gran_size_opt(char *p) -{ - if (!p) - return -EINVAL; - mtrr_gran_size = memparse(p, &p); - return 0; -} -early_param("mtrr_gran_size", parse_mtrr_gran_size_opt); - -static int nr_mtrr_spare_reg __initdata = - CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT; - -static int __init parse_mtrr_spare_reg(char *arg) -{ - if (arg) - nr_mtrr_spare_reg = simple_strtoul(arg, NULL, 0); - return 0; -} - -early_param("mtrr_spare_reg_nr", parse_mtrr_spare_reg); - -static int __init -x86_setup_var_mtrrs(struct res_range *range, int nr_range, - u64 chunk_size, u64 gran_size) -{ - struct var_mtrr_state var_state; - int i; - int num_reg; - - var_state.range_startk = 0; - var_state.range_sizek = 0; - var_state.reg = 0; - var_state.chunk_sizek = chunk_size >> 10; - var_state.gran_sizek = gran_size >> 10; - - memset(range_state, 0, sizeof(range_state)); - - /* Write the range etc */ - for (i = 0; i < nr_range; i++) - set_var_mtrr_range(&var_state, range[i].start, - range[i].end - range[i].start + 1); - - /* Write the last range */ - if (var_state.range_sizek != 0) - range_to_mtrr_with_hole(&var_state, 0, 0); - - num_reg = var_state.reg; - /* Clear out the extra MTRR's */ - while (var_state.reg < num_var_ranges) { - save_var_mtrr(var_state.reg, 0, 0, 0); - var_state.reg++; - } - - return num_reg; -} - -struct mtrr_cleanup_result { - unsigned long gran_sizek; - unsigned long chunk_sizek; - unsigned long lose_cover_sizek; - unsigned int num_reg; - int bad; -}; - -/* - * gran_size: 64K, 128K, 256K, 512K, 1M, 2M, ..., 2G - * chunk size: gran_size, ..., 2G - * so we need (1+16)*8 - */ -#define NUM_RESULT 136 -#define PSHIFT (PAGE_SHIFT - 10) - -static struct mtrr_cleanup_result __initdata result[NUM_RESULT]; -static unsigned long __initdata min_loss_pfn[RANGE_NUM]; - -static void __init print_out_mtrr_range_state(void) -{ - int i; - char start_factor = 'K', size_factor = 'K'; - unsigned long start_base, size_base; - mtrr_type type; - - for (i = 0; i < num_var_ranges; i++) { - - size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10); - if (!size_base) - continue; - - size_base = to_size_factor(size_base, &size_factor), - start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10); - start_base = to_size_factor(start_base, &start_factor), - type = range_state[i].type; - - printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n", - i, start_base, start_factor, - size_base, size_factor, - (type == MTRR_TYPE_UNCACHABLE) ? "UC" : - ((type == MTRR_TYPE_WRPROT) ? "WP" : - ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other")) - ); - } -} - -static int __init mtrr_need_cleanup(void) -{ - int i; - mtrr_type type; - unsigned long size; - /* extra one for all 0 */ - int num[MTRR_NUM_TYPES + 1]; - - /* check entries number */ - memset(num, 0, sizeof(num)); - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - size = range_state[i].size_pfn; - if (type >= MTRR_NUM_TYPES) - continue; - if (!size) - type = MTRR_NUM_TYPES; - if (type == MTRR_TYPE_WRPROT) - type = MTRR_TYPE_UNCACHABLE; - num[type]++; - } - - /* check if we got UC entries */ - if (!num[MTRR_TYPE_UNCACHABLE]) - return 0; - - /* check if we only had WB and UC */ - if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != - num_var_ranges - num[MTRR_NUM_TYPES]) - return 0; - - return 1; -} - -static unsigned long __initdata range_sums; -static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size, - unsigned long extra_remove_base, - unsigned long extra_remove_size, - int i) -{ - int num_reg; - static struct res_range range_new[RANGE_NUM]; - static int nr_range_new; - unsigned long range_sums_new; - - /* convert ranges to var ranges state */ - num_reg = x86_setup_var_mtrrs(range, nr_range, - chunk_size, gran_size); - - /* we got new setting in range_state, check it */ - memset(range_new, 0, sizeof(range_new)); - nr_range_new = x86_get_mtrr_mem_range(range_new, 0, - extra_remove_base, extra_remove_size); - range_sums_new = sum_ranges(range_new, nr_range_new); - - result[i].chunk_sizek = chunk_size >> 10; - result[i].gran_sizek = gran_size >> 10; - result[i].num_reg = num_reg; - if (range_sums < range_sums_new) { - result[i].lose_cover_sizek = - (range_sums_new - range_sums) << PSHIFT; - result[i].bad = 1; - } else - result[i].lose_cover_sizek = - (range_sums - range_sums_new) << PSHIFT; - - /* double check it */ - if (!result[i].bad && !result[i].lose_cover_sizek) { - if (nr_range_new != nr_range || - memcmp(range, range_new, sizeof(range))) - result[i].bad = 1; - } - - if (!result[i].bad && (range_sums - range_sums_new < - min_loss_pfn[num_reg])) { - min_loss_pfn[num_reg] = - range_sums - range_sums_new; - } -} - -static void __init mtrr_print_out_one_result(int i) -{ - char gran_factor, chunk_factor, lose_factor; - unsigned long gran_base, chunk_base, lose_base; - - gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), - chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), - lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), - printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", - result[i].bad ? "*BAD*" : " ", - gran_base, gran_factor, chunk_base, chunk_factor); - printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", - result[i].num_reg, result[i].bad ? "-" : "", - lose_base, lose_factor); -} - -static int __init mtrr_search_optimal_index(void) -{ - int i; - int num_reg_good; - int index_good; - - if (nr_mtrr_spare_reg >= num_var_ranges) - nr_mtrr_spare_reg = num_var_ranges - 1; - num_reg_good = -1; - for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) { - if (!min_loss_pfn[i]) - num_reg_good = i; - } - - index_good = -1; - if (num_reg_good != -1) { - for (i = 0; i < NUM_RESULT; i++) { - if (!result[i].bad && - result[i].num_reg == num_reg_good && - !result[i].lose_cover_sizek) { - index_good = i; - break; - } - } - } - - return index_good; -} - - -static int __init mtrr_cleanup(unsigned address_bits) -{ - unsigned long extra_remove_base, extra_remove_size; - unsigned long base, size, def, dummy; - mtrr_type type; - u64 chunk_size, gran_size; - int index_good; - int i; - - if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1) - return 0; - rdmsr(MTRRdefType_MSR, def, dummy); - def &= 0xff; - if (def != MTRR_TYPE_UNCACHABLE) - return 0; - - /* get it and store it aside */ - memset(range_state, 0, sizeof(range_state)); - for (i = 0; i < num_var_ranges; i++) { - mtrr_if->get(i, &base, &size, &type); - range_state[i].base_pfn = base; - range_state[i].size_pfn = size; - range_state[i].type = type; - } - - /* check if we need handle it and can handle it */ - if (!mtrr_need_cleanup()) - return 0; - - /* print original var MTRRs at first, for debugging: */ - printk(KERN_DEBUG "original variable MTRRs\n"); - print_out_mtrr_range_state(); - - memset(range, 0, sizeof(range)); - extra_remove_size = 0; - extra_remove_base = 1 << (32 - PAGE_SHIFT); - if (mtrr_tom2) - extra_remove_size = - (mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base; - nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base, - extra_remove_size); - /* - * [0, 1M) should always be coverred by var mtrr with WB - * and fixed mtrrs should take effective before var mtrr for it - */ - nr_range = add_range_with_merge(range, nr_range, 0, - (1ULL<<(20 - PAGE_SHIFT)) - 1); - /* sort the ranges */ - sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); - - range_sums = sum_ranges(range, nr_range); - printk(KERN_INFO "total RAM coverred: %ldM\n", - range_sums >> (20 - PAGE_SHIFT)); - - if (mtrr_chunk_size && mtrr_gran_size) { - i = 0; - mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size, - extra_remove_base, extra_remove_size, i); - - mtrr_print_out_one_result(i); - - if (!result[i].bad) { - set_var_mtrr_all(address_bits); - return 1; - } - printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, " - "will find optimal one\n"); - } - - i = 0; - memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn)); - memset(result, 0, sizeof(result)); - for (gran_size = (1ULL<<16); gran_size < (1ULL<<32); gran_size <<= 1) { - - for (chunk_size = gran_size; chunk_size < (1ULL<<32); - chunk_size <<= 1) { - - if (i >= NUM_RESULT) - continue; - - mtrr_calc_range_state(chunk_size, gran_size, - extra_remove_base, extra_remove_size, i); - if (debug_print) { - mtrr_print_out_one_result(i); - printk(KERN_INFO "\n"); - } - - i++; - } - } - - /* try to find the optimal index */ - index_good = mtrr_search_optimal_index(); - - if (index_good != -1) { - printk(KERN_INFO "Found optimal setting for mtrr clean up\n"); - i = index_good; - mtrr_print_out_one_result(i); - - /* convert ranges to var ranges state */ - chunk_size = result[i].chunk_sizek; - chunk_size <<= 10; - gran_size = result[i].gran_sizek; - gran_size <<= 10; - x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size); - set_var_mtrr_all(address_bits); - printk(KERN_DEBUG "New variable MTRRs\n"); - print_out_mtrr_range_state(); - return 1; - } else { - /* print out all */ - for (i = 0; i < NUM_RESULT; i++) - mtrr_print_out_one_result(i); - } - - printk(KERN_INFO "mtrr_cleanup: can not find optimal value\n"); - printk(KERN_INFO "please specify mtrr_gran_size/mtrr_chunk_size\n"); - - return 0; -} -#else -static int __init mtrr_cleanup(unsigned address_bits) -{ - return 0; -} -#endif - -static int __initdata changed_by_mtrr_cleanup; - -static int disable_mtrr_trim; - -static int __init disable_mtrr_trim_setup(char *str) -{ - disable_mtrr_trim = 1; - return 0; -} -early_param("disable_mtrr_trim", disable_mtrr_trim_setup); - -/* - * Newer AMD K8s and later CPUs have a special magic MSR way to force WB - * for memory >4GB. Check for that here. - * Note this won't check if the MTRRs < 4GB where the magic bit doesn't - * apply to are wrong, but so far we don't know of any such case in the wild. - */ -#define Tom2Enabled (1U << 21) -#define Tom2ForceMemTypeWB (1U << 22) - -int __init amd_special_default_mtrr(void) -{ - u32 l, h; - - if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) - return 0; - if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11) - return 0; - /* In case some hypervisor doesn't pass SYSCFG through */ - if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0) - return 0; - /* - * Memory between 4GB and top of mem is forced WB by this magic bit. - * Reserved before K8RevF, but should be zero there. - */ - if ((l & (Tom2Enabled | Tom2ForceMemTypeWB)) == - (Tom2Enabled | Tom2ForceMemTypeWB)) - return 1; - return 0; -} - -static u64 __init real_trim_memory(unsigned long start_pfn, - unsigned long limit_pfn) -{ - u64 trim_start, trim_size; - trim_start = start_pfn; - trim_start <<= PAGE_SHIFT; - trim_size = limit_pfn; - trim_size <<= PAGE_SHIFT; - trim_size -= trim_start; - - return e820_update_range(trim_start, trim_size, E820_RAM, - E820_RESERVED); -} -/** - * mtrr_trim_uncached_memory - trim RAM not covered by MTRRs - * @end_pfn: ending page frame number - * - * Some buggy BIOSes don't setup the MTRRs properly for systems with certain - * memory configurations. This routine checks that the highest MTRR matches - * the end of memory, to make sure the MTRRs having a write back type cover - * all of the memory the kernel is intending to use. If not, it'll trim any - * memory off the end by adjusting end_pfn, removing it from the kernel's - * allocation pools, warning the user with an obnoxious message. - */ -int __init mtrr_trim_uncached_memory(unsigned long end_pfn) -{ - unsigned long i, base, size, highest_pfn = 0, def, dummy; - mtrr_type type; - u64 total_trim_size; - - /* extra one for all 0 */ - int num[MTRR_NUM_TYPES + 1]; - /* - * Make sure we only trim uncachable memory on machines that - * support the Intel MTRR architecture: - */ - if (!is_cpu(INTEL) || disable_mtrr_trim) - return 0; - rdmsr(MTRRdefType_MSR, def, dummy); - def &= 0xff; - if (def != MTRR_TYPE_UNCACHABLE) - return 0; - - /* get it and store it aside */ - memset(range_state, 0, sizeof(range_state)); - for (i = 0; i < num_var_ranges; i++) { - mtrr_if->get(i, &base, &size, &type); - range_state[i].base_pfn = base; - range_state[i].size_pfn = size; - range_state[i].type = type; - } - - /* Find highest cached pfn */ - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - if (type != MTRR_TYPE_WRBACK) - continue; - base = range_state[i].base_pfn; - size = range_state[i].size_pfn; - if (highest_pfn < base + size) - highest_pfn = base + size; - } - - /* kvm/qemu doesn't have mtrr set right, don't trim them all */ - if (!highest_pfn) { - printk(KERN_INFO "CPU MTRRs all blank - virtualized system.\n"); - return 0; - } - - /* check entries number */ - memset(num, 0, sizeof(num)); - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - if (type >= MTRR_NUM_TYPES) - continue; - size = range_state[i].size_pfn; - if (!size) - type = MTRR_NUM_TYPES; - num[type]++; - } - - /* no entry for WB? */ - if (!num[MTRR_TYPE_WRBACK]) - return 0; - - /* check if we only had WB and UC */ - if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != - num_var_ranges - num[MTRR_NUM_TYPES]) - return 0; - - memset(range, 0, sizeof(range)); - 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; - nr_range++; - } - nr_range = x86_get_mtrr_mem_range(range, nr_range, 0, 0); - - total_trim_size = 0; - /* check the head */ - if (range[0].start) - total_trim_size += real_trim_memory(0, range[0].start); - /* 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, - 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, - end_pfn); - - if (total_trim_size) { - printk(KERN_WARNING "WARNING: BIOS bug: CPU MTRRs don't cover" - " all of memory, losing %lluMB of RAM.\n", - total_trim_size >> 20); - - if (!changed_by_mtrr_cleanup) - WARN_ON(1); - - printk(KERN_INFO "update e820 for mtrr\n"); - update_e820(); - - return 1; - } - - return 0; -} +int __initdata changed_by_mtrr_cleanup; /** * mtrr_bp_init - initialize mtrrs on the boot CPU diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h index ffd60409cc6..77f67f7b347 100644 --- a/arch/x86/kernel/cpu/mtrr/mtrr.h +++ b/arch/x86/kernel/cpu/mtrr/mtrr.h @@ -79,6 +79,7 @@ extern struct mtrr_ops * mtrr_if; extern unsigned int num_var_ranges; extern u64 mtrr_tom2; +extern struct mtrr_state_type mtrr_state; void mtrr_state_warn(void); const char *mtrr_attrib_to_str(int x); @@ -88,3 +89,6 @@ void mtrr_wrmsr(unsigned, unsigned, unsigned); int amd_init_mtrr(void); int cyrix_init_mtrr(void); int centaur_init_mtrr(void); + +extern int changed_by_mtrr_cleanup; +extern int mtrr_cleanup(unsigned address_bits); diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index 9abd48b2267..f6c70a164e3 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -19,7 +19,7 @@ #include <linux/nmi.h> #include <linux/kprobes.h> -#include <asm/apic.h> +#include <asm/genapic.h> #include <asm/intel_arch_perfmon.h> struct nmi_watchdog_ctlblk { diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c index 01b1244ef1c..d67e0e48bc2 100644 --- a/arch/x86/kernel/cpu/proc.c +++ b/arch/x86/kernel/cpu/proc.c @@ -7,11 +7,10 @@ /* * Get CPU information for use by the procfs. */ -#ifdef CONFIG_X86_32 static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c, unsigned int cpu) { -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP if (c->x86_max_cores * smp_num_siblings > 1) { seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); seq_printf(m, "siblings\t: %d\n", @@ -24,6 +23,7 @@ static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c, #endif } +#ifdef CONFIG_X86_32 static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c) { /* @@ -50,22 +50,6 @@ static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c) c->wp_works_ok ? "yes" : "no"); } #else -static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c, - unsigned int cpu) -{ -#ifdef CONFIG_SMP - if (c->x86_max_cores * smp_num_siblings > 1) { - seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); - seq_printf(m, "siblings\t: %d\n", - cpus_weight(per_cpu(cpu_core_map, cpu))); - seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id); - seq_printf(m, "cpu cores\t: %d\n", c->booted_cores); - seq_printf(m, "apicid\t\t: %d\n", c->apicid); - seq_printf(m, "initial apicid\t: %d\n", c->initial_apicid); - } -#endif -} - static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c) { seq_printf(m, diff --git a/arch/x86/kernel/cpu/transmeta.c b/arch/x86/kernel/cpu/transmeta.c index 52b3fefbd5a..bb62b3e5caa 100644 --- a/arch/x86/kernel/cpu/transmeta.c +++ b/arch/x86/kernel/cpu/transmeta.c @@ -98,7 +98,7 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c) #endif } -static struct cpu_dev transmeta_cpu_dev __cpuinitdata = { +static const struct cpu_dev __cpuinitconst transmeta_cpu_dev = { .c_vendor = "Transmeta", .c_ident = { "GenuineTMx86", "TransmetaCPU" }, .c_early_init = early_init_transmeta, diff --git a/arch/x86/kernel/cpu/umc.c b/arch/x86/kernel/cpu/umc.c index e777f79e096..fd2c37bf7ac 100644 --- a/arch/x86/kernel/cpu/umc.c +++ b/arch/x86/kernel/cpu/umc.c @@ -8,7 +8,7 @@ * so no special init takes place. */ -static struct cpu_dev umc_cpu_dev __cpuinitdata = { +static const struct cpu_dev __cpuinitconst umc_cpu_dev = { .c_vendor = "UMC", .c_ident = { "UMC UMC UMC" }, .c_models = { diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index c689d19e35a..ff958248e61 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -24,12 +24,10 @@ #include <asm/apic.h> #include <asm/hpet.h> #include <linux/kdebug.h> -#include <asm/smp.h> +#include <asm/cpu.h> #include <asm/reboot.h> #include <asm/virtext.h> -#include <mach_ipi.h> - #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index 6b1f6f6f866..87d103ded1c 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -99,7 +99,7 @@ print_context_stack(struct thread_info *tinfo, frame = frame->next_frame; bp = (unsigned long) frame; } else { - ops->address(data, addr, bp == 0); + ops->address(data, addr, 0); } print_ftrace_graph_addr(addr, data, ops, tinfo, graph); } diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c index c302d070704..d35db5993fd 100644 --- a/arch/x86/kernel/dumpstack_64.c +++ b/arch/x86/kernel/dumpstack_64.c @@ -106,7 +106,8 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, const struct stacktrace_ops *ops, void *data) { const unsigned cpu = get_cpu(); - unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr; + unsigned long *irq_stack_end = + (unsigned long *)per_cpu(irq_stack_ptr, cpu); unsigned used = 0; struct thread_info *tinfo; int graph = 0; @@ -160,23 +161,23 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, stack = (unsigned long *) estack_end[-2]; continue; } - if (irqstack_end) { - unsigned long *irqstack; - irqstack = irqstack_end - - (IRQSTACKSIZE - 64) / sizeof(*irqstack); + if (irq_stack_end) { + unsigned long *irq_stack; + irq_stack = irq_stack_end - + (IRQ_STACK_SIZE - 64) / sizeof(*irq_stack); - if (stack >= irqstack && stack < irqstack_end) { + if (stack >= irq_stack && stack < irq_stack_end) { if (ops->stack(data, "IRQ") < 0) break; bp = print_context_stack(tinfo, stack, bp, - ops, data, irqstack_end, &graph); + ops, data, irq_stack_end, &graph); /* * We link to the next stack (which would be * the process stack normally) the last * pointer (index -1 to end) in the IRQ stack: */ - stack = (unsigned long *) (irqstack_end[-1]); - irqstack_end = NULL; + stack = (unsigned long *) (irq_stack_end[-1]); + irq_stack_end = NULL; ops->stack(data, "EOI"); continue; } @@ -199,10 +200,10 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, unsigned long *stack; int i; const int cpu = smp_processor_id(); - unsigned long *irqstack_end = - (unsigned long *) (cpu_pda(cpu)->irqstackptr); - unsigned long *irqstack = - (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE); + 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); /* * debugging aid: "show_stack(NULL, NULL);" prints the @@ -218,9 +219,9 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, stack = sp; for (i = 0; i < kstack_depth_to_print; i++) { - if (stack >= irqstack && stack <= irqstack_end) { - if (stack == irqstack_end) { - stack = (unsigned long *) (irqstack_end[-1]); + if (stack >= irq_stack && stack <= irq_stack_end) { + if (stack == irq_stack_end) { + stack = (unsigned long *) (irq_stack_end[-1]); printk(" <EOI> "); } } else { @@ -241,7 +242,7 @@ void show_registers(struct pt_regs *regs) int i; unsigned long sp; const int cpu = smp_processor_id(); - struct task_struct *cur = cpu_pda(cpu)->pcurrent; + struct task_struct *cur = current; sp = regs->sp; printk("CPU %d ", cpu); diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index e85826829cf..fb638d9ce6d 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -110,19 +110,50 @@ int __init e820_all_mapped(u64 start, u64 end, unsigned type) /* * Add a memory region to the kernel e820 map. */ -void __init e820_add_region(u64 start, u64 size, int type) +static void __init __e820_add_region(struct e820map *e820x, u64 start, u64 size, + int type) { - int x = e820.nr_map; + int x = e820x->nr_map; - if (x == ARRAY_SIZE(e820.map)) { + if (x == ARRAY_SIZE(e820x->map)) { printk(KERN_ERR "Ooops! Too many entries in the memory map!\n"); return; } - e820.map[x].addr = start; - e820.map[x].size = size; - e820.map[x].type = type; - e820.nr_map++; + e820x->map[x].addr = start; + e820x->map[x].size = size; + e820x->map[x].type = type; + e820x->nr_map++; +} + +void __init e820_add_region(u64 start, u64 size, int type) +{ + __e820_add_region(&e820, start, size, type); +} + +static void __init e820_print_type(u32 type) +{ + switch (type) { + case E820_RAM: + case E820_RESERVED_KERN: + printk(KERN_CONT "(usable)"); + break; + case E820_RESERVED: + printk(KERN_CONT "(reserved)"); + break; + case E820_ACPI: + printk(KERN_CONT "(ACPI data)"); + break; + case E820_NVS: + printk(KERN_CONT "(ACPI NVS)"); + break; + case E820_UNUSABLE: + printk(KERN_CONT "(unusable)"); + break; + default: + printk(KERN_CONT "type %u", type); + break; + } } void __init e820_print_map(char *who) @@ -134,27 +165,8 @@ void __init e820_print_map(char *who) (unsigned long long) e820.map[i].addr, (unsigned long long) (e820.map[i].addr + e820.map[i].size)); - switch (e820.map[i].type) { - case E820_RAM: - case E820_RESERVED_KERN: - printk(KERN_CONT "(usable)\n"); - break; - case E820_RESERVED: - printk(KERN_CONT "(reserved)\n"); - break; - case E820_ACPI: - printk(KERN_CONT "(ACPI data)\n"); - break; - case E820_NVS: - printk(KERN_CONT "(ACPI NVS)\n"); - break; - case E820_UNUSABLE: - printk("(unusable)\n"); - break; - default: - printk(KERN_CONT "type %u\n", e820.map[i].type); - break; - } + e820_print_type(e820.map[i].type); + printk(KERN_CONT "\n"); } } @@ -417,11 +429,12 @@ static int __init append_e820_map(struct e820entry *biosmap, int nr_map) return __append_e820_map(biosmap, nr_map); } -static u64 __init e820_update_range_map(struct e820map *e820x, u64 start, +static u64 __init __e820_update_range(struct e820map *e820x, u64 start, u64 size, unsigned old_type, unsigned new_type) { - int i; + u64 end; + unsigned int i; u64 real_updated_size = 0; BUG_ON(old_type == new_type); @@ -429,27 +442,55 @@ static u64 __init e820_update_range_map(struct e820map *e820x, u64 start, if (size > (ULLONG_MAX - start)) size = ULLONG_MAX - start; - for (i = 0; i < e820.nr_map; i++) { + end = start + size; + printk(KERN_DEBUG "e820 update range: %016Lx - %016Lx ", + (unsigned long long) start, + (unsigned long long) end); + e820_print_type(old_type); + printk(KERN_CONT " ==> "); + e820_print_type(new_type); + printk(KERN_CONT "\n"); + + for (i = 0; i < e820x->nr_map; i++) { struct e820entry *ei = &e820x->map[i]; u64 final_start, final_end; + u64 ei_end; + if (ei->type != old_type) continue; - /* totally covered? */ - if (ei->addr >= start && - (ei->addr + ei->size) <= (start + size)) { + + ei_end = ei->addr + ei->size; + /* totally covered by new range? */ + if (ei->addr >= start && ei_end <= end) { ei->type = new_type; real_updated_size += ei->size; continue; } + + /* new range is totally covered? */ + if (ei->addr < start && ei_end > end) { + __e820_add_region(e820x, start, size, new_type); + __e820_add_region(e820x, end, ei_end - end, ei->type); + ei->size = start - ei->addr; + real_updated_size += size; + continue; + } + /* partially covered */ final_start = max(start, ei->addr); - final_end = min(start + size, ei->addr + ei->size); + final_end = min(end, ei_end); if (final_start >= final_end) continue; - e820_add_region(final_start, final_end - final_start, - new_type); + + __e820_add_region(e820x, final_start, final_end - final_start, + new_type); + real_updated_size += final_end - final_start; + /* + * left range could be head or tail, so need to update + * size at first. + */ ei->size -= final_end - final_start; if (ei->addr < final_start) continue; @@ -461,13 +502,13 @@ static u64 __init e820_update_range_map(struct e820map *e820x, u64 start, u64 __init e820_update_range(u64 start, u64 size, unsigned old_type, unsigned new_type) { - return e820_update_range_map(&e820, start, size, old_type, new_type); + return __e820_update_range(&e820, start, size, old_type, new_type); } static u64 __init e820_update_range_saved(u64 start, u64 size, unsigned old_type, unsigned new_type) { - return e820_update_range_map(&e820_saved, start, size, old_type, + return __e820_update_range(&e820_saved, start, size, old_type, new_type); } @@ -858,6 +899,9 @@ void __init reserve_early_overlap_ok(u64 start, u64 end, char *name) */ 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); } @@ -1017,8 +1061,8 @@ u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align) continue; return addr; } - return -1UL; + return -1ULL; } /* @@ -1031,13 +1075,22 @@ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) u64 start; start = startt; - while (size < sizet) + while (size < sizet && (start + 1)) start = find_e820_area_size(start, &size, align); if (size < sizet) return 0; +#ifdef CONFIG_X86_32 + if (start >= MAXMEM) + return 0; + if (start + size > MAXMEM) + size = MAXMEM - start; +#endif + addr = round_down(start + size - sizet, align); + if (addr < start) + return 0; e820_update_range(addr, sizet, E820_RAM, E820_RESERVED); e820_update_range_saved(addr, sizet, E820_RAM, E820_RESERVED); printk(KERN_INFO "update e820 for early_reserve_e820\n"); diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c index 504ad198e4a..335f049d110 100644 --- a/arch/x86/kernel/early_printk.c +++ b/arch/x86/kernel/early_printk.c @@ -13,8 +13,8 @@ #include <asm/setup.h> #include <xen/hvc-console.h> #include <asm/pci-direct.h> -#include <asm/pgtable.h> #include <asm/fixmap.h> +#include <asm/pgtable.h> #include <linux/usb/ehci_def.h> /* Simple VGA output */ @@ -250,7 +250,7 @@ static int dbgp_wait_until_complete(void) return (ctrl & DBGP_ERROR) ? -DBGP_ERRCODE(ctrl) : DBGP_LEN(ctrl); } -static void dbgp_mdelay(int ms) +static void __init dbgp_mdelay(int ms) { int i; @@ -311,7 +311,7 @@ static void dbgp_set_data(const void *buf, int size) writel(hi, &ehci_debug->data47); } -static void dbgp_get_data(void *buf, int size) +static void __init dbgp_get_data(void *buf, int size) { unsigned char *bytes = buf; u32 lo, hi; @@ -355,7 +355,7 @@ static int dbgp_bulk_write(unsigned devnum, unsigned endpoint, return ret; } -static int dbgp_bulk_read(unsigned devnum, unsigned endpoint, void *data, +static int __init dbgp_bulk_read(unsigned devnum, unsigned endpoint, void *data, int size) { u32 pids, addr, ctrl; @@ -386,8 +386,8 @@ static int dbgp_bulk_read(unsigned devnum, unsigned endpoint, void *data, return ret; } -static int dbgp_control_msg(unsigned devnum, int requesttype, int request, - int value, int index, void *data, int size) +static int __init dbgp_control_msg(unsigned devnum, int requesttype, + int request, int value, int index, void *data, int size) { u32 pids, addr, ctrl; struct usb_ctrlrequest req; @@ -489,7 +489,7 @@ static u32 __init find_dbgp(int ehci_num, u32 *rbus, u32 *rslot, u32 *rfunc) return 0; } -static int ehci_reset_port(int port) +static int __init ehci_reset_port(int port) { u32 portsc; u32 delay_time, delay; @@ -532,7 +532,7 @@ static int ehci_reset_port(int port) return -EBUSY; } -static int ehci_wait_for_port(int port) +static int __init ehci_wait_for_port(int port) { u32 status; int ret, reps; @@ -557,13 +557,13 @@ static inline void dbgp_printk(const char *fmt, ...) { } typedef void (*set_debug_port_t)(int port); -static void default_set_debug_port(int port) +static void __init default_set_debug_port(int port) { } -static set_debug_port_t set_debug_port = default_set_debug_port; +static set_debug_port_t __initdata set_debug_port = default_set_debug_port; -static void nvidia_set_debug_port(int port) +static void __init nvidia_set_debug_port(int port) { u32 dword; dword = read_pci_config(ehci_dev.bus, ehci_dev.slot, ehci_dev.func, diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index eb1ef3b67dd..1736acc4d7a 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -366,10 +366,12 @@ void __init efi_init(void) SMBIOS_TABLE_GUID)) { efi.smbios = config_tables[i].table; printk(" SMBIOS=0x%lx ", config_tables[i].table); +#ifdef CONFIG_X86_UV } else if (!efi_guidcmp(config_tables[i].guid, UV_SYSTEM_TABLE_GUID)) { efi.uv_systab = config_tables[i].table; printk(" UVsystab=0x%lx ", config_tables[i].table); +#endif } else if (!efi_guidcmp(config_tables[i].guid, HCDP_TABLE_GUID)) { efi.hcdp = config_tables[i].table; diff --git a/arch/x86/kernel/efi_64.c b/arch/x86/kernel/efi_64.c index cb783b92c50..22c3b7828c5 100644 --- a/arch/x86/kernel/efi_64.c +++ b/arch/x86/kernel/efi_64.c @@ -36,6 +36,7 @@ #include <asm/proto.h> #include <asm/efi.h> #include <asm/cacheflush.h> +#include <asm/fixmap.h> static pgd_t save_pgd __initdata; static unsigned long efi_flags __initdata; diff --git a/arch/x86/kernel/efi_stub_32.S b/arch/x86/kernel/efi_stub_32.S index ef00bb77d7e..fbe66e626c0 100644 --- a/arch/x86/kernel/efi_stub_32.S +++ b/arch/x86/kernel/efi_stub_32.S @@ -6,7 +6,7 @@ */ #include <linux/linkage.h> -#include <asm/page.h> +#include <asm/page_types.h> /* * efi_call_phys(void *, ...) is a function with variable parameters. @@ -113,6 +113,7 @@ ENTRY(efi_call_phys) movl (%edx), %ecx pushl %ecx ret +ENDPROC(efi_call_phys) .previous .data diff --git a/arch/x86/kernel/efi_stub_64.S b/arch/x86/kernel/efi_stub_64.S index 99b47d48c9f..4c07ccab814 100644 --- a/arch/x86/kernel/efi_stub_64.S +++ b/arch/x86/kernel/efi_stub_64.S @@ -41,6 +41,7 @@ ENTRY(efi_call0) addq $32, %rsp RESTORE_XMM ret +ENDPROC(efi_call0) ENTRY(efi_call1) SAVE_XMM @@ -50,6 +51,7 @@ ENTRY(efi_call1) addq $32, %rsp RESTORE_XMM ret +ENDPROC(efi_call1) ENTRY(efi_call2) SAVE_XMM @@ -59,6 +61,7 @@ ENTRY(efi_call2) addq $32, %rsp RESTORE_XMM ret +ENDPROC(efi_call2) ENTRY(efi_call3) SAVE_XMM @@ -69,6 +72,7 @@ ENTRY(efi_call3) addq $32, %rsp RESTORE_XMM ret +ENDPROC(efi_call3) ENTRY(efi_call4) SAVE_XMM @@ -80,6 +84,7 @@ ENTRY(efi_call4) addq $32, %rsp RESTORE_XMM ret +ENDPROC(efi_call4) ENTRY(efi_call5) SAVE_XMM @@ -92,6 +97,7 @@ ENTRY(efi_call5) addq $48, %rsp RESTORE_XMM ret +ENDPROC(efi_call5) ENTRY(efi_call6) SAVE_XMM @@ -107,3 +113,4 @@ ENTRY(efi_call6) addq $48, %rsp RESTORE_XMM ret +ENDPROC(efi_call6) diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index 46469029e9d..c929add475c 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -30,12 +30,13 @@ * 1C(%esp) - %ds * 20(%esp) - %es * 24(%esp) - %fs - * 28(%esp) - orig_eax - * 2C(%esp) - %eip - * 30(%esp) - %cs - * 34(%esp) - %eflags - * 38(%esp) - %oldesp - * 3C(%esp) - %oldss + * 28(%esp) - %gs saved iff !CONFIG_X86_32_LAZY_GS + * 2C(%esp) - orig_eax + * 30(%esp) - %eip + * 34(%esp) - %cs + * 38(%esp) - %eflags + * 3C(%esp) - %oldesp + * 40(%esp) - %oldss * * "current" is in register %ebx during any slow entries. */ @@ -46,7 +47,7 @@ #include <asm/errno.h> #include <asm/segment.h> #include <asm/smp.h> -#include <asm/page.h> +#include <asm/page_types.h> #include <asm/desc.h> #include <asm/percpu.h> #include <asm/dwarf2.h> @@ -101,121 +102,221 @@ #define resume_userspace_sig resume_userspace #endif -#define SAVE_ALL \ - cld; \ - pushl %fs; \ - CFI_ADJUST_CFA_OFFSET 4;\ - /*CFI_REL_OFFSET fs, 0;*/\ - pushl %es; \ - CFI_ADJUST_CFA_OFFSET 4;\ - /*CFI_REL_OFFSET es, 0;*/\ - pushl %ds; \ - CFI_ADJUST_CFA_OFFSET 4;\ - /*CFI_REL_OFFSET ds, 0;*/\ - pushl %eax; \ - CFI_ADJUST_CFA_OFFSET 4;\ - CFI_REL_OFFSET eax, 0;\ - pushl %ebp; \ - CFI_ADJUST_CFA_OFFSET 4;\ - CFI_REL_OFFSET ebp, 0;\ - pushl %edi; \ - CFI_ADJUST_CFA_OFFSET 4;\ - CFI_REL_OFFSET edi, 0;\ - pushl %esi; \ - CFI_ADJUST_CFA_OFFSET 4;\ - CFI_REL_OFFSET esi, 0;\ - pushl %edx; \ - CFI_ADJUST_CFA_OFFSET 4;\ - CFI_REL_OFFSET edx, 0;\ - pushl %ecx; \ - CFI_ADJUST_CFA_OFFSET 4;\ - CFI_REL_OFFSET ecx, 0;\ - pushl %ebx; \ - CFI_ADJUST_CFA_OFFSET 4;\ - CFI_REL_OFFSET ebx, 0;\ - movl $(__USER_DS), %edx; \ - movl %edx, %ds; \ - movl %edx, %es; \ - movl $(__KERNEL_PERCPU), %edx; \ +/* + * User gs save/restore + * + * %gs is used for userland TLS and kernel only uses it for stack + * canary which is required to be at %gs:20 by gcc. Read the comment + * at the top of stackprotector.h for more info. + * + * Local labels 98 and 99 are used. + */ +#ifdef CONFIG_X86_32_LAZY_GS + + /* unfortunately push/pop can't be no-op */ +.macro PUSH_GS + pushl $0 + CFI_ADJUST_CFA_OFFSET 4 +.endm +.macro POP_GS pop=0 + addl $(4 + \pop), %esp + CFI_ADJUST_CFA_OFFSET -(4 + \pop) +.endm +.macro POP_GS_EX +.endm + + /* all the rest are no-op */ +.macro PTGS_TO_GS +.endm +.macro PTGS_TO_GS_EX +.endm +.macro GS_TO_REG reg +.endm +.macro REG_TO_PTGS reg +.endm +.macro SET_KERNEL_GS reg +.endm + +#else /* CONFIG_X86_32_LAZY_GS */ + +.macro PUSH_GS + pushl %gs + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET gs, 0*/ +.endm + +.macro POP_GS pop=0 +98: popl %gs + CFI_ADJUST_CFA_OFFSET -4 + /*CFI_RESTORE gs*/ + .if \pop <> 0 + add $\pop, %esp + CFI_ADJUST_CFA_OFFSET -\pop + .endif +.endm +.macro POP_GS_EX +.pushsection .fixup, "ax" +99: movl $0, (%esp) + jmp 98b +.section __ex_table, "a" + .align 4 + .long 98b, 99b +.popsection +.endm + +.macro PTGS_TO_GS +98: mov PT_GS(%esp), %gs +.endm +.macro PTGS_TO_GS_EX +.pushsection .fixup, "ax" +99: movl $0, PT_GS(%esp) + jmp 98b +.section __ex_table, "a" + .align 4 + .long 98b, 99b +.popsection +.endm + +.macro GS_TO_REG reg + movl %gs, \reg + /*CFI_REGISTER gs, \reg*/ +.endm +.macro REG_TO_PTGS reg + movl \reg, PT_GS(%esp) + /*CFI_REL_OFFSET gs, PT_GS*/ +.endm +.macro SET_KERNEL_GS reg + movl $(__KERNEL_STACK_CANARY), \reg + movl \reg, %gs +.endm + +#endif /* CONFIG_X86_32_LAZY_GS */ + +.macro SAVE_ALL + cld + PUSH_GS + pushl %fs + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET fs, 0;*/ + pushl %es + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET es, 0;*/ + pushl %ds + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET ds, 0;*/ + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET eax, 0 + pushl %ebp + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ebp, 0 + pushl %edi + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET edi, 0 + pushl %esi + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET esi, 0 + pushl %edx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET edx, 0 + pushl %ecx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ecx, 0 + pushl %ebx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ebx, 0 + movl $(__USER_DS), %edx + movl %edx, %ds + movl %edx, %es + movl $(__KERNEL_PERCPU), %edx movl %edx, %fs + SET_KERNEL_GS %edx +.endm -#define RESTORE_INT_REGS \ - popl %ebx; \ - CFI_ADJUST_CFA_OFFSET -4;\ - CFI_RESTORE ebx;\ - popl %ecx; \ - CFI_ADJUST_CFA_OFFSET -4;\ - CFI_RESTORE ecx;\ - popl %edx; \ - CFI_ADJUST_CFA_OFFSET -4;\ - CFI_RESTORE edx;\ - popl %esi; \ - CFI_ADJUST_CFA_OFFSET -4;\ - CFI_RESTORE esi;\ - popl %edi; \ - CFI_ADJUST_CFA_OFFSET -4;\ - CFI_RESTORE edi;\ - popl %ebp; \ - CFI_ADJUST_CFA_OFFSET -4;\ - CFI_RESTORE ebp;\ - popl %eax; \ - CFI_ADJUST_CFA_OFFSET -4;\ +.macro RESTORE_INT_REGS + popl %ebx + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE ebx + popl %ecx + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE ecx + popl %edx + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE edx + popl %esi + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE esi + popl %edi + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE edi + popl %ebp + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE ebp + popl %eax + CFI_ADJUST_CFA_OFFSET -4 CFI_RESTORE eax +.endm -#define RESTORE_REGS \ - RESTORE_INT_REGS; \ -1: popl %ds; \ - CFI_ADJUST_CFA_OFFSET -4;\ - /*CFI_RESTORE ds;*/\ -2: popl %es; \ - CFI_ADJUST_CFA_OFFSET -4;\ - /*CFI_RESTORE es;*/\ -3: popl %fs; \ - CFI_ADJUST_CFA_OFFSET -4;\ - /*CFI_RESTORE fs;*/\ -.pushsection .fixup,"ax"; \ -4: movl $0,(%esp); \ - jmp 1b; \ -5: movl $0,(%esp); \ - jmp 2b; \ -6: movl $0,(%esp); \ - jmp 3b; \ -.section __ex_table,"a";\ - .align 4; \ - .long 1b,4b; \ - .long 2b,5b; \ - .long 3b,6b; \ +.macro RESTORE_REGS pop=0 + RESTORE_INT_REGS +1: popl %ds + CFI_ADJUST_CFA_OFFSET -4 + /*CFI_RESTORE ds;*/ +2: popl %es + CFI_ADJUST_CFA_OFFSET -4 + /*CFI_RESTORE es;*/ +3: popl %fs + CFI_ADJUST_CFA_OFFSET -4 + /*CFI_RESTORE fs;*/ + POP_GS \pop +.pushsection .fixup, "ax" +4: movl $0, (%esp) + jmp 1b +5: movl $0, (%esp) + jmp 2b +6: movl $0, (%esp) + jmp 3b +.section __ex_table, "a" + .align 4 + .long 1b, 4b + .long 2b, 5b + .long 3b, 6b .popsection + POP_GS_EX +.endm -#define RING0_INT_FRAME \ - CFI_STARTPROC simple;\ - CFI_SIGNAL_FRAME;\ - CFI_DEF_CFA esp, 3*4;\ - /*CFI_OFFSET cs, -2*4;*/\ +.macro RING0_INT_FRAME + CFI_STARTPROC simple + CFI_SIGNAL_FRAME + CFI_DEF_CFA esp, 3*4 + /*CFI_OFFSET cs, -2*4;*/ CFI_OFFSET eip, -3*4 +.endm -#define RING0_EC_FRAME \ - CFI_STARTPROC simple;\ - CFI_SIGNAL_FRAME;\ - CFI_DEF_CFA esp, 4*4;\ - /*CFI_OFFSET cs, -2*4;*/\ +.macro RING0_EC_FRAME + CFI_STARTPROC simple + CFI_SIGNAL_FRAME + CFI_DEF_CFA esp, 4*4 + /*CFI_OFFSET cs, -2*4;*/ CFI_OFFSET eip, -3*4 +.endm -#define RING0_PTREGS_FRAME \ - CFI_STARTPROC simple;\ - CFI_SIGNAL_FRAME;\ - CFI_DEF_CFA esp, PT_OLDESP-PT_EBX;\ - /*CFI_OFFSET cs, PT_CS-PT_OLDESP;*/\ - CFI_OFFSET eip, PT_EIP-PT_OLDESP;\ - /*CFI_OFFSET es, PT_ES-PT_OLDESP;*/\ - /*CFI_OFFSET ds, PT_DS-PT_OLDESP;*/\ - CFI_OFFSET eax, PT_EAX-PT_OLDESP;\ - CFI_OFFSET ebp, PT_EBP-PT_OLDESP;\ - CFI_OFFSET edi, PT_EDI-PT_OLDESP;\ - CFI_OFFSET esi, PT_ESI-PT_OLDESP;\ - CFI_OFFSET edx, PT_EDX-PT_OLDESP;\ - CFI_OFFSET ecx, PT_ECX-PT_OLDESP;\ +.macro RING0_PTREGS_FRAME + CFI_STARTPROC simple + CFI_SIGNAL_FRAME + CFI_DEF_CFA esp, PT_OLDESP-PT_EBX + /*CFI_OFFSET cs, PT_CS-PT_OLDESP;*/ + CFI_OFFSET eip, PT_EIP-PT_OLDESP + /*CFI_OFFSET es, PT_ES-PT_OLDESP;*/ + /*CFI_OFFSET ds, PT_DS-PT_OLDESP;*/ + CFI_OFFSET eax, PT_EAX-PT_OLDESP + CFI_OFFSET ebp, PT_EBP-PT_OLDESP + CFI_OFFSET edi, PT_EDI-PT_OLDESP + CFI_OFFSET esi, PT_ESI-PT_OLDESP + CFI_OFFSET edx, PT_EDX-PT_OLDESP + CFI_OFFSET ecx, PT_ECX-PT_OLDESP CFI_OFFSET ebx, PT_EBX-PT_OLDESP +.endm ENTRY(ret_from_fork) CFI_STARTPROC @@ -341,8 +442,7 @@ sysenter_past_esp: GET_THREAD_INFO(%ebp) - /* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */ - testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp) + testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp) jnz sysenter_audit sysenter_do_call: cmpl $(nr_syscalls), %eax @@ -353,7 +453,7 @@ sysenter_do_call: DISABLE_INTERRUPTS(CLBR_ANY) TRACE_IRQS_OFF movl TI_flags(%ebp), %ecx - testw $_TIF_ALLWORK_MASK, %cx + testl $_TIF_ALLWORK_MASK, %ecx jne sysexit_audit sysenter_exit: /* if something modifies registers it must also disable sysexit */ @@ -362,11 +462,12 @@ sysenter_exit: xorl %ebp,%ebp TRACE_IRQS_ON 1: mov PT_FS(%esp), %fs + PTGS_TO_GS ENABLE_INTERRUPTS_SYSEXIT #ifdef CONFIG_AUDITSYSCALL sysenter_audit: - testw $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%ebp) + testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%ebp) jnz syscall_trace_entry addl $4,%esp CFI_ADJUST_CFA_OFFSET -4 @@ -383,7 +484,7 @@ sysenter_audit: jmp sysenter_do_call sysexit_audit: - testw $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %cx + testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %ecx jne syscall_exit_work TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_ANY) @@ -396,7 +497,7 @@ sysexit_audit: DISABLE_INTERRUPTS(CLBR_ANY) TRACE_IRQS_OFF movl TI_flags(%ebp), %ecx - testw $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %cx + testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %ecx jne syscall_exit_work movl PT_EAX(%esp),%eax /* reload syscall return value */ jmp sysenter_exit @@ -410,6 +511,7 @@ sysexit_audit: .align 4 .long 1b,2b .popsection + PTGS_TO_GS_EX ENDPROC(ia32_sysenter_target) # system call handler stub @@ -420,8 +522,7 @@ ENTRY(system_call) SAVE_ALL GET_THREAD_INFO(%ebp) # system call tracing in operation / emulation - /* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */ - testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp) + testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp) jnz syscall_trace_entry cmpl $(nr_syscalls), %eax jae syscall_badsys @@ -435,7 +536,7 @@ syscall_exit: # between sampling and the iret TRACE_IRQS_OFF movl TI_flags(%ebp), %ecx - testw $_TIF_ALLWORK_MASK, %cx # current->work + testl $_TIF_ALLWORK_MASK, %ecx # current->work jne syscall_exit_work restore_all: @@ -452,8 +553,7 @@ restore_all: restore_nocheck: TRACE_IRQS_IRET restore_nocheck_notrace: - RESTORE_REGS - addl $4, %esp # skip orig_eax/error_code + RESTORE_REGS 4 # skip orig_eax/error_code CFI_ADJUST_CFA_OFFSET -4 irq_return: INTERRUPT_RETURN @@ -571,7 +671,7 @@ END(syscall_trace_entry) # perform syscall exit tracing ALIGN syscall_exit_work: - testb $_TIF_WORK_SYSCALL_EXIT, %cl + testl $_TIF_WORK_SYSCALL_EXIT, %ecx jz work_pending TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_ANY) # could let syscall_trace_leave() call @@ -595,28 +695,50 @@ syscall_badsys: END(syscall_badsys) CFI_ENDPROC -#define FIXUP_ESPFIX_STACK \ - /* since we are on a wrong stack, we cant make it a C code :( */ \ - PER_CPU(gdt_page, %ebx); \ - GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah); \ - addl %esp, %eax; \ - pushl $__KERNEL_DS; \ - CFI_ADJUST_CFA_OFFSET 4; \ - pushl %eax; \ - CFI_ADJUST_CFA_OFFSET 4; \ - lss (%esp), %esp; \ - CFI_ADJUST_CFA_OFFSET -8; -#define UNWIND_ESPFIX_STACK \ - movl %ss, %eax; \ - /* see if on espfix stack */ \ - cmpw $__ESPFIX_SS, %ax; \ - jne 27f; \ - movl $__KERNEL_DS, %eax; \ - movl %eax, %ds; \ - movl %eax, %es; \ - /* switch to normal stack */ \ - FIXUP_ESPFIX_STACK; \ -27:; +/* + * System calls that need a pt_regs pointer. + */ +#define PTREGSCALL(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) + +.macro FIXUP_ESPFIX_STACK + /* since we are on a wrong stack, we cant make it a C code :( */ + PER_CPU(gdt_page, %ebx) + GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah) + addl %esp, %eax + pushl $__KERNEL_DS + CFI_ADJUST_CFA_OFFSET 4 + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + lss (%esp), %esp + CFI_ADJUST_CFA_OFFSET -8 +.endm +.macro UNWIND_ESPFIX_STACK + movl %ss, %eax + /* see if on espfix stack */ + cmpw $__ESPFIX_SS, %ax + jne 27f + movl $__KERNEL_DS, %eax + movl %eax, %ds + movl %eax, %es + /* switch to normal stack */ + FIXUP_ESPFIX_STACK +27: +.endm /* * Build the entry stubs and pointer table with some assembler magic. @@ -672,7 +794,7 @@ common_interrupt: ENDPROC(common_interrupt) CFI_ENDPROC -#define BUILD_INTERRUPT(name, nr) \ +#define BUILD_INTERRUPT3(name, nr, fn) \ ENTRY(name) \ RING0_INT_FRAME; \ pushl $~(nr); \ @@ -680,13 +802,15 @@ ENTRY(name) \ SAVE_ALL; \ TRACE_IRQS_OFF \ movl %esp,%eax; \ - call smp_##name; \ + call fn; \ jmp ret_from_intr; \ CFI_ENDPROC; \ ENDPROC(name) +#define BUILD_INTERRUPT(name, nr) BUILD_INTERRUPT3(name, nr, smp_##name) + /* The include is where all of the SMP etc. interrupts come from */ -#include "entry_arch.h" +#include <asm/entry_arch.h> ENTRY(coprocessor_error) RING0_INT_FRAME @@ -1068,7 +1192,10 @@ ENTRY(page_fault) CFI_ADJUST_CFA_OFFSET 4 ALIGN error_code: - /* the function address is in %fs's slot on the stack */ + /* the function address is in %gs's slot on the stack */ + pushl %fs + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET fs, 0*/ pushl %es CFI_ADJUST_CFA_OFFSET 4 /*CFI_REL_OFFSET es, 0*/ @@ -1097,20 +1224,15 @@ error_code: CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET ebx, 0 cld - pushl %fs - CFI_ADJUST_CFA_OFFSET 4 - /*CFI_REL_OFFSET fs, 0*/ movl $(__KERNEL_PERCPU), %ecx movl %ecx, %fs UNWIND_ESPFIX_STACK - popl %ecx - CFI_ADJUST_CFA_OFFSET -4 - /*CFI_REGISTER es, ecx*/ - movl PT_FS(%esp), %edi # get the function address + GS_TO_REG %ecx + movl PT_GS(%esp), %edi # get the function address movl PT_ORIG_EAX(%esp), %edx # get the error code movl $-1, PT_ORIG_EAX(%esp) # no syscall to restart - mov %ecx, PT_FS(%esp) - /*CFI_REL_OFFSET fs, ES*/ + REG_TO_PTGS %ecx + SET_KERNEL_GS %ecx movl $(__USER_DS), %ecx movl %ecx, %ds movl %ecx, %es @@ -1134,26 +1256,27 @@ END(page_fault) * by hand onto the new stack - while updating the return eip past * the instruction that would have done it for sysenter. */ -#define FIX_STACK(offset, ok, label) \ - cmpw $__KERNEL_CS,4(%esp); \ - jne ok; \ -label: \ - movl TSS_sysenter_sp0+offset(%esp),%esp; \ - CFI_DEF_CFA esp, 0; \ - CFI_UNDEFINED eip; \ - pushfl; \ - CFI_ADJUST_CFA_OFFSET 4; \ - pushl $__KERNEL_CS; \ - CFI_ADJUST_CFA_OFFSET 4; \ - pushl $sysenter_past_esp; \ - CFI_ADJUST_CFA_OFFSET 4; \ +.macro FIX_STACK offset ok label + cmpw $__KERNEL_CS, 4(%esp) + jne \ok +\label: + movl TSS_sysenter_sp0 + \offset(%esp), %esp + CFI_DEF_CFA esp, 0 + CFI_UNDEFINED eip + pushfl + CFI_ADJUST_CFA_OFFSET 4 + pushl $__KERNEL_CS + CFI_ADJUST_CFA_OFFSET 4 + pushl $sysenter_past_esp + CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET eip, 0 +.endm ENTRY(debug) RING0_INT_FRAME cmpl $ia32_sysenter_target,(%esp) jne debug_stack_correct - FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn) + FIX_STACK 12, debug_stack_correct, debug_esp_fix_insn debug_stack_correct: pushl $-1 # mark this as an int CFI_ADJUST_CFA_OFFSET 4 @@ -1211,7 +1334,7 @@ nmi_stack_correct: nmi_stack_fixup: RING0_INT_FRAME - FIX_STACK(12,nmi_stack_correct, 1) + FIX_STACK 12, nmi_stack_correct, 1 jmp nmi_stack_correct nmi_debug_stack_check: @@ -1222,7 +1345,7 @@ nmi_debug_stack_check: jb nmi_stack_correct cmpl $debug_esp_fix_insn,(%esp) ja nmi_stack_correct - FIX_STACK(24,nmi_stack_correct, 1) + FIX_STACK 24, nmi_stack_correct, 1 jmp nmi_stack_correct nmi_espfix_stack: @@ -1234,7 +1357,7 @@ nmi_espfix_stack: CFI_ADJUST_CFA_OFFSET 4 pushl %esp CFI_ADJUST_CFA_OFFSET 4 - addw $4, (%esp) + addl $4, (%esp) /* copy the iret frame of 12 bytes */ .rept 3 pushl 16(%esp) diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index a1346217e43..a331ec38af9 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -48,10 +48,11 @@ #include <asm/unistd.h> #include <asm/thread_info.h> #include <asm/hw_irq.h> -#include <asm/page.h> +#include <asm/page_types.h> #include <asm/irqflags.h> #include <asm/paravirt.h> #include <asm/ftrace.h> +#include <asm/percpu.h> /* Avoid __ASSEMBLER__'ifying <linux/audit.h> just for this. */ #include <linux/elf-em.h> @@ -76,20 +77,17 @@ ENTRY(ftrace_caller) movq 8(%rbp), %rsi subq $MCOUNT_INSN_SIZE, %rdi -.globl ftrace_call -ftrace_call: +GLOBAL(ftrace_call) call ftrace_stub MCOUNT_RESTORE_FRAME #ifdef CONFIG_FUNCTION_GRAPH_TRACER -.globl ftrace_graph_call -ftrace_graph_call: +GLOBAL(ftrace_graph_call) jmp ftrace_stub #endif -.globl ftrace_stub -ftrace_stub: +GLOBAL(ftrace_stub) retq END(ftrace_caller) @@ -109,8 +107,7 @@ ENTRY(mcount) jnz ftrace_graph_caller #endif -.globl ftrace_stub -ftrace_stub: +GLOBAL(ftrace_stub) retq trace: @@ -147,9 +144,7 @@ ENTRY(ftrace_graph_caller) retq END(ftrace_graph_caller) - -.globl return_to_handler -return_to_handler: +GLOBAL(return_to_handler) subq $80, %rsp movq %rax, (%rsp) @@ -187,6 +182,7 @@ return_to_handler: ENTRY(native_usergs_sysret64) swapgs sysretq +ENDPROC(native_usergs_sysret64) #endif /* CONFIG_PARAVIRT */ @@ -209,7 +205,7 @@ ENTRY(native_usergs_sysret64) /* %rsp:at FRAMEEND */ .macro FIXUP_TOP_OF_STACK tmp offset=0 - movq %gs:pda_oldrsp,\tmp + movq PER_CPU_VAR(old_rsp),\tmp movq \tmp,RSP+\offset(%rsp) movq $__USER_DS,SS+\offset(%rsp) movq $__USER_CS,CS+\offset(%rsp) @@ -220,7 +216,7 @@ ENTRY(native_usergs_sysret64) .macro RESTORE_TOP_OF_STACK tmp offset=0 movq RSP+\offset(%rsp),\tmp - movq \tmp,%gs:pda_oldrsp + movq \tmp,PER_CPU_VAR(old_rsp) movq EFLAGS+\offset(%rsp),\tmp movq \tmp,R11+\offset(%rsp) .endm @@ -336,15 +332,15 @@ ENTRY(save_args) je 1f SWAPGS /* - * irqcount is used to check if a CPU is already on an interrupt stack + * irq_count is used to check if a CPU is already on an interrupt stack * or not. While this is essentially redundant with preempt_count it is * a little cheaper to use a separate counter in the PDA (short of * moving irq_enter into assembly, which would be too much work) */ -1: incl %gs:pda_irqcount +1: incl PER_CPU_VAR(irq_count) jne 2f popq_cfi %rax /* move return address... */ - mov %gs:pda_irqstackptr,%rsp + mov PER_CPU_VAR(irq_stack_ptr),%rsp EMPTY_FRAME 0 pushq_cfi %rbp /* backlink for unwinder */ pushq_cfi %rax /* ... to the new stack */ @@ -372,6 +368,7 @@ ENTRY(save_rest) END(save_rest) /* save complete stack frame */ + .pushsection .kprobes.text, "ax" ENTRY(save_paranoid) XCPT_FRAME 1 RDI+8 cld @@ -400,6 +397,7 @@ ENTRY(save_paranoid) 1: ret CFI_ENDPROC END(save_paranoid) + .popsection /* * A newly forked process directly context switches into this address. @@ -409,6 +407,8 @@ END(save_paranoid) ENTRY(ret_from_fork) DEFAULT_FRAME + LOCK ; btr $TIF_FORK,TI_flags(%r8) + push kernel_eflags(%rip) CFI_ADJUST_CFA_OFFSET 8 popf # reset kernel eflags @@ -418,7 +418,6 @@ ENTRY(ret_from_fork) GET_THREAD_INFO(%rcx) - CFI_REMEMBER_STATE RESTORE_REST testl $3, CS-ARGOFFSET(%rsp) # from kernel_thread? @@ -430,7 +429,6 @@ ENTRY(ret_from_fork) RESTORE_TOP_OF_STACK %rdi, -ARGOFFSET jmp ret_from_sys_call # go to the SYSRET fastpath - CFI_RESTORE_STATE CFI_ENDPROC END(ret_from_fork) @@ -468,7 +466,7 @@ END(ret_from_fork) ENTRY(system_call) CFI_STARTPROC simple CFI_SIGNAL_FRAME - CFI_DEF_CFA rsp,PDA_STACKOFFSET + CFI_DEF_CFA rsp,KERNEL_STACK_OFFSET CFI_REGISTER rip,rcx /*CFI_REGISTER rflags,r11*/ SWAPGS_UNSAFE_STACK @@ -479,8 +477,8 @@ ENTRY(system_call) */ ENTRY(system_call_after_swapgs) - movq %rsp,%gs:pda_oldrsp - movq %gs:pda_kernelstack,%rsp + movq %rsp,PER_CPU_VAR(old_rsp) + movq PER_CPU_VAR(kernel_stack),%rsp /* * No need to follow this irqs off/on section - it's straight * and short: @@ -523,7 +521,7 @@ sysret_check: CFI_REGISTER rip,rcx RESTORE_ARGS 0,-ARG_SKIP,1 /*CFI_REGISTER rflags,r11*/ - movq %gs:pda_oldrsp, %rsp + movq PER_CPU_VAR(old_rsp), %rsp USERGS_SYSRET64 CFI_RESTORE_STATE @@ -630,16 +628,14 @@ tracesys: * Syscall return path ending with IRET. * Has correct top of stack, but partial stack frame. */ - .globl int_ret_from_sys_call - .globl int_with_check -int_ret_from_sys_call: +GLOBAL(int_ret_from_sys_call) DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF testl $3,CS-ARGOFFSET(%rsp) je retint_restore_args movl $_TIF_ALLWORK_MASK,%edi /* edi: mask to check */ -int_with_check: +GLOBAL(int_with_check) LOCKDEP_SYS_EXIT_IRQ GET_THREAD_INFO(%rcx) movl TI_flags(%rcx),%edx @@ -833,11 +829,11 @@ common_interrupt: XCPT_FRAME addq $-0x80,(%rsp) /* Adjust vector to [-256,-1] range */ interrupt do_IRQ - /* 0(%rsp): oldrsp-ARGOFFSET */ + /* 0(%rsp): old_rsp-ARGOFFSET */ ret_from_intr: DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF - decl %gs:pda_irqcount + decl PER_CPU_VAR(irq_count) leaveq CFI_DEF_CFA_REGISTER rsp CFI_ADJUST_CFA_OFFSET -8 @@ -982,10 +978,14 @@ apicinterrupt IRQ_MOVE_CLEANUP_VECTOR \ irq_move_cleanup_interrupt smp_irq_move_cleanup_interrupt #endif +#ifdef CONFIG_X86_UV apicinterrupt UV_BAU_MESSAGE \ uv_bau_message_intr1 uv_bau_message_interrupt +#endif apicinterrupt LOCAL_TIMER_VECTOR \ apic_timer_interrupt smp_apic_timer_interrupt +apicinterrupt GENERIC_INTERRUPT_VECTOR \ + generic_interrupt smp_generic_interrupt #ifdef CONFIG_SMP apicinterrupt INVALIDATE_TLB_VECTOR_START+0 \ @@ -1073,10 +1073,10 @@ ENTRY(\sym) TRACE_IRQS_OFF movq %rsp,%rdi /* pt_regs pointer */ xorl %esi,%esi /* no error code */ - movq %gs:pda_data_offset, %rbp - subq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp) + PER_CPU(init_tss, %rbp) + subq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%rbp) call \do_sym - addq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp) + addq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%rbp) jmp paranoid_exit /* %ebx: no swapgs flag */ CFI_ENDPROC END(\sym) @@ -1138,7 +1138,7 @@ ENTRY(native_load_gs_index) CFI_STARTPROC pushf CFI_ADJUST_CFA_OFFSET 8 - DISABLE_INTERRUPTS(CLBR_ANY | ~(CLBR_RDI)) + DISABLE_INTERRUPTS(CLBR_ANY & ~CLBR_RDI) SWAPGS gs_change: movl %edi,%gs @@ -1260,14 +1260,14 @@ ENTRY(call_softirq) CFI_REL_OFFSET rbp,0 mov %rsp,%rbp CFI_DEF_CFA_REGISTER rbp - incl %gs:pda_irqcount - cmove %gs:pda_irqstackptr,%rsp + incl PER_CPU_VAR(irq_count) + cmove PER_CPU_VAR(irq_stack_ptr),%rsp push %rbp # backlink for old unwinder call __do_softirq leaveq CFI_DEF_CFA_REGISTER rsp CFI_ADJUST_CFA_OFFSET -8 - decl %gs:pda_irqcount + decl PER_CPU_VAR(irq_count) ret CFI_ENDPROC END(call_softirq) @@ -1297,15 +1297,15 @@ ENTRY(xen_do_hypervisor_callback) # do_hypervisor_callback(struct *pt_regs) movq %rdi, %rsp # we don't return, adjust the stack frame CFI_ENDPROC DEFAULT_FRAME -11: incl %gs:pda_irqcount +11: incl PER_CPU_VAR(irq_count) movq %rsp,%rbp CFI_DEF_CFA_REGISTER rbp - cmovzq %gs:pda_irqstackptr,%rsp + cmovzq PER_CPU_VAR(irq_stack_ptr),%rsp pushq %rbp # backlink for old unwinder call xen_evtchn_do_upcall popq %rsp CFI_DEF_CFA_REGISTER rsp - decl %gs:pda_irqcount + decl PER_CPU_VAR(irq_count) jmp error_exit CFI_ENDPROC END(do_hypervisor_callback) diff --git a/arch/x86/kernel/es7000_32.c b/arch/x86/kernel/es7000_32.c deleted file mode 100644 index 53699c931ad..00000000000 --- a/arch/x86/kernel/es7000_32.c +++ /dev/null @@ -1,378 +0,0 @@ -/* - * Written by: Garry Forsgren, Unisys Corporation - * Natalie Protasevich, Unisys Corporation - * This file contains the code to configure and interface - * with Unisys ES7000 series hardware system manager. - * - * Copyright (c) 2003 Unisys Corporation. All Rights Reserved. - * - * 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. - * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Unisys Corporation, Township Line & Union Meeting - * Roads-A, Unisys Way, Blue Bell, Pennsylvania, 19424, or: - * - * http://www.unisys.com - */ - -#include <linux/module.h> -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/smp.h> -#include <linux/string.h> -#include <linux/spinlock.h> -#include <linux/errno.h> -#include <linux/notifier.h> -#include <linux/reboot.h> -#include <linux/init.h> -#include <linux/acpi.h> -#include <asm/io.h> -#include <asm/nmi.h> -#include <asm/smp.h> -#include <asm/atomic.h> -#include <asm/apicdef.h> -#include <mach_mpparse.h> -#include <asm/genapic.h> -#include <asm/setup.h> - -/* - * ES7000 chipsets - */ - -#define NON_UNISYS 0 -#define ES7000_CLASSIC 1 -#define ES7000_ZORRO 2 - - -#define MIP_REG 1 -#define MIP_PSAI_REG 4 - -#define MIP_BUSY 1 -#define MIP_SPIN 0xf0000 -#define MIP_VALID 0x0100000000000000ULL -#define MIP_PORT(VALUE) ((VALUE >> 32) & 0xffff) - -#define MIP_RD_LO(VALUE) (VALUE & 0xffffffff) - -struct mip_reg_info { - unsigned long long mip_info; - unsigned long long delivery_info; - unsigned long long host_reg; - unsigned long long mip_reg; -}; - -struct part_info { - unsigned char type; - unsigned char length; - unsigned char part_id; - unsigned char apic_mode; - unsigned long snum; - char ptype[16]; - char sname[64]; - char pname[64]; -}; - -struct psai { - unsigned long long entry_type; - unsigned long long addr; - unsigned long long bep_addr; -}; - -struct es7000_mem_info { - unsigned char type; - unsigned char length; - unsigned char resv[6]; - unsigned long long start; - unsigned long long size; -}; - -struct es7000_oem_table { - unsigned long long hdr; - struct mip_reg_info mip; - struct part_info pif; - struct es7000_mem_info shm; - struct psai psai; -}; - -#ifdef CONFIG_ACPI - -struct oem_table { - struct acpi_table_header Header; - u32 OEMTableAddr; - u32 OEMTableSize; -}; - -extern int find_unisys_acpi_oem_table(unsigned long *oem_addr); -extern void unmap_unisys_acpi_oem_table(unsigned long oem_addr); -#endif - -struct mip_reg { - unsigned long long off_0; - unsigned long long off_8; - unsigned long long off_10; - unsigned long long off_18; - unsigned long long off_20; - unsigned long long off_28; - unsigned long long off_30; - unsigned long long off_38; -}; - -#define MIP_SW_APIC 0x1020b -#define MIP_FUNC(VALUE) (VALUE & 0xff) - -/* - * ES7000 Globals - */ - -static volatile unsigned long *psai = NULL; -static struct mip_reg *mip_reg; -static struct mip_reg *host_reg; -static int mip_port; -static unsigned long mip_addr, host_addr; - -int es7000_plat; - -/* - * GSI override for ES7000 platforms. - */ - -static unsigned int base; - -static int -es7000_rename_gsi(int ioapic, int gsi) -{ - if (es7000_plat == ES7000_ZORRO) - return gsi; - - if (!base) { - int i; - for (i = 0; i < nr_ioapics; i++) - base += nr_ioapic_registers[i]; - } - - if (!ioapic && (gsi < 16)) - gsi += base; - return gsi; -} - -static int wakeup_secondary_cpu_via_mip(int cpu, unsigned long eip) -{ - unsigned long vect = 0, psaival = 0; - - if (psai == NULL) - return -1; - - vect = ((unsigned long)__pa(eip)/0x1000) << 16; - psaival = (0x1000000 | vect | cpu); - - while (*psai & 0x1000000) - ; - - *psai = psaival; - - return 0; -} - -static void noop_wait_for_deassert(atomic_t *deassert_not_used) -{ -} - -static int __init es7000_update_genapic(void) -{ - genapic->wakeup_cpu = wakeup_secondary_cpu_via_mip; - - /* MPENTIUMIII */ - if (boot_cpu_data.x86 == 6 && - (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11)) { - es7000_update_genapic_to_cluster(); - genapic->wait_for_init_deassert = noop_wait_for_deassert; - genapic->wakeup_cpu = wakeup_secondary_cpu_via_mip; - } - - return 0; -} - -void __init -setup_unisys(void) -{ - /* - * Determine the generation of the ES7000 currently running. - * - * es7000_plat = 1 if the machine is a 5xx ES7000 box - * es7000_plat = 2 if the machine is a x86_64 ES7000 box - * - */ - if (!(boot_cpu_data.x86 <= 15 && boot_cpu_data.x86_model <= 2)) - es7000_plat = ES7000_ZORRO; - else - es7000_plat = ES7000_CLASSIC; - ioapic_renumber_irq = es7000_rename_gsi; - - x86_quirks->update_genapic = es7000_update_genapic; -} - -/* - * Parse the OEM Table - */ - -int __init -parse_unisys_oem (char *oemptr) -{ - int i; - int success = 0; - unsigned char type, size; - unsigned long val; - char *tp = NULL; - struct psai *psaip = NULL; - struct mip_reg_info *mi; - struct mip_reg *host, *mip; - - tp = oemptr; - - tp += 8; - - for (i=0; i <= 6; i++) { - type = *tp++; - size = *tp++; - tp -= 2; - switch (type) { - case MIP_REG: - mi = (struct mip_reg_info *)tp; - val = MIP_RD_LO(mi->host_reg); - host_addr = val; - host = (struct mip_reg *)val; - host_reg = __va(host); - val = MIP_RD_LO(mi->mip_reg); - mip_port = MIP_PORT(mi->mip_info); - mip_addr = val; - mip = (struct mip_reg *)val; - mip_reg = __va(mip); - pr_debug("es7000_mipcfg: host_reg = 0x%lx \n", - (unsigned long)host_reg); - pr_debug("es7000_mipcfg: mip_reg = 0x%lx \n", - (unsigned long)mip_reg); - success++; - break; - case MIP_PSAI_REG: - psaip = (struct psai *)tp; - if (tp != NULL) { - if (psaip->addr) - psai = __va(psaip->addr); - else - psai = NULL; - success++; - } - break; - default: - break; - } - tp += size; - } - - if (success < 2) { - es7000_plat = NON_UNISYS; - } else - setup_unisys(); - return es7000_plat; -} - -#ifdef CONFIG_ACPI -static unsigned long oem_addrX; -static unsigned long oem_size; -int __init find_unisys_acpi_oem_table(unsigned long *oem_addr) -{ - struct acpi_table_header *header = NULL; - int i = 0; - - while (ACPI_SUCCESS(acpi_get_table("OEM1", i++, &header))) { - if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) { - struct oem_table *t = (struct oem_table *)header; - - oem_addrX = t->OEMTableAddr; - oem_size = t->OEMTableSize; - - *oem_addr = (unsigned long)__acpi_map_table(oem_addrX, - oem_size); - return 0; - } - } - return -1; -} - -void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr) -{ -} -#endif - -static void -es7000_spin(int n) -{ - int i = 0; - - while (i++ < n) - rep_nop(); -} - -static int __init -es7000_mip_write(struct mip_reg *mip_reg) -{ - int status = 0; - int spin; - - spin = MIP_SPIN; - while (((unsigned long long)host_reg->off_38 & - (unsigned long long)MIP_VALID) != 0) { - if (--spin <= 0) { - printk("es7000_mip_write: Timeout waiting for Host Valid Flag"); - return -1; - } - es7000_spin(MIP_SPIN); - } - - memcpy(host_reg, mip_reg, sizeof(struct mip_reg)); - outb(1, mip_port); - - spin = MIP_SPIN; - - while (((unsigned long long)mip_reg->off_38 & - (unsigned long long)MIP_VALID) == 0) { - if (--spin <= 0) { - printk("es7000_mip_write: Timeout waiting for MIP Valid Flag"); - return -1; - } - es7000_spin(MIP_SPIN); - } - - status = ((unsigned long long)mip_reg->off_0 & - (unsigned long long)0xffff0000000000ULL) >> 48; - mip_reg->off_38 = ((unsigned long long)mip_reg->off_38 & - (unsigned long long)~MIP_VALID); - return status; -} - -void __init -es7000_sw_apic(void) -{ - if (es7000_plat) { - int mip_status; - struct mip_reg es7000_mip_reg; - - printk("ES7000: Enabling APIC mode.\n"); - memset(&es7000_mip_reg, 0, sizeof(struct mip_reg)); - es7000_mip_reg.off_0 = MIP_SW_APIC; - es7000_mip_reg.off_38 = (MIP_VALID); - while ((mip_status = es7000_mip_write(&es7000_mip_reg)) != 0) - printk("es7000_sw_apic: command failed, status = %x\n", - mip_status); - return; - } -} diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c deleted file mode 100644 index 2bced78b0b8..00000000000 --- a/arch/x86/kernel/genapic_64.c +++ /dev/null @@ -1,82 +0,0 @@ -/* - * Copyright 2004 James Cleverdon, IBM. - * Subject to the GNU Public License, v.2 - * - * Generic APIC sub-arch probe layer. - * - * Hacked for x86-64 by James Cleverdon from i386 architecture code by - * Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and - * James Cleverdon. - */ -#include <linux/threads.h> -#include <linux/cpumask.h> -#include <linux/string.h> -#include <linux/module.h> -#include <linux/kernel.h> -#include <linux/ctype.h> -#include <linux/init.h> -#include <linux/hardirq.h> -#include <linux/dmar.h> - -#include <asm/smp.h> -#include <asm/ipi.h> -#include <asm/genapic.h> -#include <asm/setup.h> - -extern struct genapic apic_flat; -extern struct genapic apic_physflat; -extern struct genapic apic_x2xpic_uv_x; -extern struct genapic apic_x2apic_phys; -extern struct genapic apic_x2apic_cluster; - -struct genapic __read_mostly *genapic = &apic_flat; - -static struct genapic *apic_probe[] __initdata = { - &apic_x2apic_uv_x, - &apic_x2apic_phys, - &apic_x2apic_cluster, - &apic_physflat, - NULL, -}; - -/* - * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode. - */ -void __init setup_apic_routing(void) -{ - if (genapic == &apic_x2apic_phys || genapic == &apic_x2apic_cluster) { - if (!intr_remapping_enabled) - genapic = &apic_flat; - } - - if (genapic == &apic_flat) { - if (max_physical_apicid >= 8) - genapic = &apic_physflat; - printk(KERN_INFO "Setting APIC routing to %s\n", genapic->name); - } - - if (x86_quirks->update_genapic) - x86_quirks->update_genapic(); -} - -/* Same for both flat and physical. */ - -void apic_send_IPI_self(int vector) -{ - __send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL); -} - -int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) -{ - int i; - - for (i = 0; apic_probe[i]; ++i) { - if (apic_probe[i]->acpi_madt_oem_check(oem_id, oem_table_id)) { - genapic = apic_probe[i]; - printk(KERN_INFO "Setting APIC routing to %s.\n", - genapic->name); - return 1; - } - } - return 0; -} diff --git a/arch/x86/kernel/genx2apic_cluster.c b/arch/x86/kernel/genx2apic_cluster.c deleted file mode 100644 index 6ce497cc372..00000000000 --- a/arch/x86/kernel/genx2apic_cluster.c +++ /dev/null @@ -1,198 +0,0 @@ -#include <linux/threads.h> -#include <linux/cpumask.h> -#include <linux/string.h> -#include <linux/kernel.h> -#include <linux/ctype.h> -#include <linux/init.h> -#include <linux/dmar.h> - -#include <asm/smp.h> -#include <asm/ipi.h> -#include <asm/genapic.h> - -DEFINE_PER_CPU(u32, x86_cpu_to_logical_apicid); - -static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) -{ - if (cpu_has_x2apic) - return 1; - - return 0; -} - -/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ - -static const struct cpumask *x2apic_target_cpus(void) -{ - return cpumask_of(0); -} - -/* - * for now each logical cpu is in its own vector allocation domain. - */ -static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask) -{ - cpumask_clear(retmask); - cpumask_set_cpu(cpu, retmask); -} - -static void __x2apic_send_IPI_dest(unsigned int apicid, int vector, - unsigned int dest) -{ - unsigned long cfg; - - cfg = __prepare_ICR(0, vector, dest); - - /* - * send the IPI. - */ - x2apic_icr_write(cfg, apicid); -} - -/* - * for now, we send the IPI's one by one in the cpumask. - * TBD: Based on the cpu mask, we can send the IPI's to the cluster group - * at once. We have 16 cpu's in a cluster. This will minimize IPI register - * writes. - */ -static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) -{ - unsigned long flags; - unsigned long query_cpu; - - local_irq_save(flags); - for_each_cpu(query_cpu, mask) - __x2apic_send_IPI_dest( - per_cpu(x86_cpu_to_logical_apicid, query_cpu), - vector, APIC_DEST_LOGICAL); - local_irq_restore(flags); -} - -static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, - int vector) -{ - unsigned long flags; - unsigned long query_cpu; - unsigned long this_cpu = smp_processor_id(); - - local_irq_save(flags); - for_each_cpu(query_cpu, mask) - if (query_cpu != this_cpu) - __x2apic_send_IPI_dest( - per_cpu(x86_cpu_to_logical_apicid, query_cpu), - vector, APIC_DEST_LOGICAL); - local_irq_restore(flags); -} - -static void x2apic_send_IPI_allbutself(int vector) -{ - unsigned long flags; - unsigned long query_cpu; - unsigned long this_cpu = smp_processor_id(); - - local_irq_save(flags); - for_each_online_cpu(query_cpu) - if (query_cpu != this_cpu) - __x2apic_send_IPI_dest( - per_cpu(x86_cpu_to_logical_apicid, query_cpu), - vector, APIC_DEST_LOGICAL); - local_irq_restore(flags); -} - -static void x2apic_send_IPI_all(int vector) -{ - x2apic_send_IPI_mask(cpu_online_mask, vector); -} - -static int x2apic_apic_id_registered(void) -{ - return 1; -} - -static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask) -{ - int cpu; - - /* - * We're using fixed IRQ delivery, can only return one logical APIC ID. - * May as well be the first. - */ - cpu = cpumask_first(cpumask); - if ((unsigned)cpu < nr_cpu_ids) - return per_cpu(x86_cpu_to_logical_apicid, cpu); - else - return BAD_APICID; -} - -static unsigned int x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, - const struct cpumask *andmask) -{ - int cpu; - - /* - * We're using fixed IRQ delivery, can only return one logical APIC ID. - * May as well be the first. - */ - for_each_cpu_and(cpu, cpumask, andmask) - if (cpumask_test_cpu(cpu, cpu_online_mask)) - break; - if (cpu < nr_cpu_ids) - return per_cpu(x86_cpu_to_logical_apicid, cpu); - return BAD_APICID; -} - -static unsigned int get_apic_id(unsigned long x) -{ - unsigned int id; - - id = x; - return id; -} - -static unsigned long set_apic_id(unsigned int id) -{ - unsigned long x; - - x = id; - return x; -} - -static unsigned int phys_pkg_id(int index_msb) -{ - return current_cpu_data.initial_apicid >> index_msb; -} - -static void x2apic_send_IPI_self(int vector) -{ - apic_write(APIC_SELF_IPI, vector); -} - -static void init_x2apic_ldr(void) -{ - int cpu = smp_processor_id(); - - per_cpu(x86_cpu_to_logical_apicid, cpu) = apic_read(APIC_LDR); - return; -} - -struct genapic apic_x2apic_cluster = { - .name = "cluster x2apic", - .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, - .int_delivery_mode = dest_LowestPrio, - .int_dest_mode = (APIC_DEST_LOGICAL != 0), - .target_cpus = x2apic_target_cpus, - .vector_allocation_domain = x2apic_vector_allocation_domain, - .apic_id_registered = x2apic_apic_id_registered, - .init_apic_ldr = init_x2apic_ldr, - .send_IPI_all = x2apic_send_IPI_all, - .send_IPI_allbutself = x2apic_send_IPI_allbutself, - .send_IPI_mask = x2apic_send_IPI_mask, - .send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself, - .send_IPI_self = x2apic_send_IPI_self, - .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, - .cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and, - .phys_pkg_id = phys_pkg_id, - .get_apic_id = get_apic_id, - .set_apic_id = set_apic_id, - .apic_id_mask = (0xFFFFFFFFu), -}; diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index ac108d1fe18..3f8579f8d42 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -18,7 +18,7 @@ void __init i386_start_kernel(void) { reserve_trampoline_memory(); - reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS"); + reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ @@ -29,9 +29,6 @@ void __init i386_start_kernel(void) reserve_early(ramdisk_image, ramdisk_end, "RAMDISK"); } #endif - reserve_early(init_pg_tables_start, init_pg_tables_end, - "INIT_PG_TABLE"); - reserve_ebda_region(); /* diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index b9a4d8c4b93..70eaa852c73 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -26,27 +26,6 @@ #include <asm/bios_ebda.h> #include <asm/trampoline.h> -/* boot cpu pda */ -static struct x8664_pda _boot_cpu_pda; - -#ifdef CONFIG_SMP -/* - * We install an empty cpu_pda pointer table to indicate to early users - * (numa_set_node) that the cpu_pda pointer table for cpus other than - * the boot cpu is not yet setup. - */ -static struct x8664_pda *__cpu_pda[NR_CPUS] __initdata; -#else -static struct x8664_pda *__cpu_pda[NR_CPUS] __read_mostly; -#endif - -void __init x86_64_init_pda(void) -{ - _cpu_pda = __cpu_pda; - cpu_pda(0) = &_boot_cpu_pda; - pda_init(0); -} - static void __init zap_identity_mappings(void) { pgd_t *pgd = pgd_offset_k(0UL); @@ -112,8 +91,6 @@ void __init x86_64_start_kernel(char * real_mode_data) if (console_loglevel == 10) early_printk("Kernel alive\n"); - x86_64_init_pda(); - x86_64_start_reservations(real_mode_data); } @@ -123,7 +100,7 @@ void __init x86_64_start_reservations(char *real_mode_data) reserve_trampoline_memory(); - reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS"); + reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index e835b4eea70..30683883e0c 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -11,14 +11,15 @@ #include <linux/init.h> #include <linux/linkage.h> #include <asm/segment.h> -#include <asm/page.h> -#include <asm/pgtable.h> +#include <asm/page_types.h> +#include <asm/pgtable_types.h> #include <asm/desc.h> #include <asm/cache.h> #include <asm/thread_info.h> #include <asm/asm-offsets.h> #include <asm/setup.h> #include <asm/processor-flags.h> +#include <asm/percpu.h> /* Physical address */ #define pa(X) ((X) - __PAGE_OFFSET) @@ -37,42 +38,40 @@ #define X86_VENDOR_ID new_cpu_data+CPUINFO_x86_vendor_id /* - * This is how much memory *in addition to the memory covered up to - * and including _end* we need mapped initially. + * This is how much memory in addition to the memory covered up to + * and including _end we need mapped initially. * We need: - * - one bit for each possible page, but only in low memory, which means - * 2^32/4096/8 = 128K worst case (4G/4G split.) - * - enough space to map all low memory, which means - * (2^32/4096) / 1024 pages (worst case, non PAE) - * (2^32/4096) / 512 + 4 pages (worst case for PAE) - * - a few pages for allocator use before the kernel pagetable has - * been set up + * (KERNEL_IMAGE_SIZE/4096) / 1024 pages (worst case, non PAE) + * (KERNEL_IMAGE_SIZE/4096) / 512 + 4 pages (worst case for PAE) * * Modulo rounding, each megabyte assigned here requires a kilobyte of * memory, which is currently unreclaimed. * * This should be a multiple of a page. + * + * KERNEL_IMAGE_SIZE should be greater than pa(_end) + * and small than max_low_pfn, otherwise will waste some page table entries */ -LOW_PAGES = 1<<(32-PAGE_SHIFT_asm) - -/* - * To preserve the DMA pool in PAGEALLOC kernels, we'll allocate - * pagetables from above the 16MB DMA limit, so we'll have to set - * up pagetables 16MB more (worst-case): - */ -#ifdef CONFIG_DEBUG_PAGEALLOC -LOW_PAGES = LOW_PAGES + 0x1000000 -#endif #if PTRS_PER_PMD > 1 -PAGE_TABLE_SIZE = (LOW_PAGES / PTRS_PER_PMD) + PTRS_PER_PGD +#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) + PTRS_PER_PGD) #else -PAGE_TABLE_SIZE = (LOW_PAGES / PTRS_PER_PGD) +#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD) #endif -BOOTBITMAP_SIZE = LOW_PAGES / 8 -ALLOCATOR_SLOP = 4 -INIT_MAP_BEYOND_END = BOOTBITMAP_SIZE + (PAGE_TABLE_SIZE + ALLOCATOR_SLOP)*PAGE_SIZE_asm +/* Enough space to fit pagetables for the low memory linear map */ +MAPPING_BEYOND_END = \ + PAGE_TABLE_SIZE(((1<<32) - __PAGE_OFFSET) >> PAGE_SHIFT) << PAGE_SHIFT + +/* + * Worst-case size of the kernel mapping we need to make: + * the worst-case size of the kernel itself, plus the extra we need + * to map for the linear map. + */ +KERNEL_PAGES = (KERNEL_IMAGE_SIZE + MAPPING_BEYOND_END)>>PAGE_SHIFT + +INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_PAGES) * PAGE_SIZE_asm +RESERVE_BRK(pagetables, INIT_MAP_SIZE) /* * 32-bit kernel entrypoint; only used by the boot CPU. On entry, @@ -165,10 +164,10 @@ num_subarch_entries = (. - subarch_entries) / 4 /* * Initialize page tables. This creates a PDE and a set of page - * tables, which are located immediately beyond _end. The variable - * init_pg_tables_end is set up to point to the first "safe" location. + * tables, which are located immediately beyond __brk_base. The variable + * _brk_end is set up to point to the first "safe" location. * Mappings are created both at virtual address 0 (identity mapping) - * and PAGE_OFFSET for up to _end+sizeof(page tables)+INIT_MAP_BEYOND_END. + * and PAGE_OFFSET for up to _end. * * Note that the stack is not yet set up! */ @@ -189,8 +188,7 @@ default_entry: xorl %ebx,%ebx /* %ebx is kept at zero */ - movl $pa(pg0), %edi - movl %edi, pa(init_pg_tables_start) + movl $pa(__brk_base), %edi movl $pa(swapper_pg_pmd), %edx movl $PTE_IDENT_ATTR, %eax 10: @@ -208,14 +206,14 @@ default_entry: loop 11b /* - * End condition: we must map up to and including INIT_MAP_BEYOND_END - * bytes beyond the end of our own page tables. + * End condition: we must map up to the end + MAPPING_BEYOND_END. */ - leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp + movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp cmpl %ebp,%eax jb 10b 1: - movl %edi,pa(init_pg_tables_end) + addl $__PAGE_OFFSET, %edi + movl %edi, pa(_brk_end) shrl $12, %eax movl %eax, pa(max_pfn_mapped) @@ -226,8 +224,7 @@ default_entry: page_pde_offset = (__PAGE_OFFSET >> 20); - movl $pa(pg0), %edi - movl %edi, pa(init_pg_tables_start) + movl $pa(__brk_base), %edi movl $pa(swapper_pg_dir), %edx movl $PTE_IDENT_ATTR, %eax 10: @@ -241,14 +238,13 @@ page_pde_offset = (__PAGE_OFFSET >> 20); addl $0x1000,%eax loop 11b /* - * End condition: we must map up to and including INIT_MAP_BEYOND_END - * bytes beyond the end of our own page tables; the +0x007 is - * the attribute bits + * End condition: we must map up to the end + MAPPING_BEYOND_END. */ - leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp + movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp cmpl %ebp,%eax jb 10b - movl %edi,pa(init_pg_tables_end) + addl $__PAGE_OFFSET, %edi + movl %edi, pa(_brk_end) shrl $12, %eax movl %eax, pa(max_pfn_mapped) @@ -429,14 +425,34 @@ is386: movl $2,%ecx # set MP ljmp $(__KERNEL_CS),$1f 1: movl $(__KERNEL_DS),%eax # reload all the segment registers movl %eax,%ss # after changing gdt. - movl %eax,%fs # gets reset once there's real percpu movl $(__USER_DS),%eax # DS/ES contains default USER segment movl %eax,%ds movl %eax,%es - xorl %eax,%eax # Clear GS and LDT + movl $(__KERNEL_PERCPU), %eax + movl %eax,%fs # set this cpu's percpu + +#ifdef CONFIG_CC_STACKPROTECTOR + /* + * The linker can't handle this by relocation. Manually set + * base address in stack canary segment descriptor. + */ + cmpb $0,ready + jne 1f + movl $per_cpu__gdt_page,%eax + movl $per_cpu__stack_canary,%ecx + subl $20, %ecx + movw %cx, 8 * GDT_ENTRY_STACK_CANARY + 2(%eax) + shrl $16, %ecx + movb %cl, 8 * GDT_ENTRY_STACK_CANARY + 4(%eax) + movb %ch, 8 * GDT_ENTRY_STACK_CANARY + 7(%eax) +1: +#endif + movl $(__KERNEL_STACK_CANARY),%eax movl %eax,%gs + + xorl %eax,%eax # Clear LDT lldt %ax cld # gcc2 wants the direction flag cleared at all times @@ -446,8 +462,6 @@ is386: movl $2,%ecx # set MP movb $1, ready cmpb $0,%cl # the first CPU calls start_kernel je 1f - movl $(__KERNEL_PERCPU), %eax - movl %eax,%fs # set this cpu's percpu movl (stack_start), %esp 1: #endif /* CONFIG_SMP */ @@ -548,12 +562,8 @@ early_fault: pushl %eax pushl %edx /* trapno */ pushl $fault_msg -#ifdef CONFIG_EARLY_PRINTK - call early_printk -#else call printk #endif -#endif call dump_stack hlt_loop: hlt @@ -580,11 +590,10 @@ ignore_int: pushl 32(%esp) pushl 40(%esp) pushl $int_msg -#ifdef CONFIG_EARLY_PRINTK - call early_printk -#else call printk -#endif + + call dump_stack + addl $(5*4),%esp popl %ds popl %es @@ -622,6 +631,7 @@ swapper_pg_fixmap: .fill 1024,4,0 ENTRY(empty_zero_page) .fill 4096,1,0 + /* * This starts the data section. */ @@ -660,7 +670,7 @@ early_recursion_flag: .long 0 int_msg: - .asciz "Unknown interrupt or fault at EIP %p %p %p\n" + .asciz "Unknown interrupt or fault at: %p %p %p\n" fault_msg: /* fault info: */ diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index 0e275d49556..54b29bb24e7 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -19,6 +19,7 @@ #include <asm/msr.h> #include <asm/cache.h> #include <asm/processor-flags.h> +#include <asm/percpu.h> #ifdef CONFIG_PARAVIRT #include <asm/asm-offsets.h> @@ -226,12 +227,15 @@ ENTRY(secondary_startup_64) movl %eax,%fs movl %eax,%gs - /* - * Setup up a dummy PDA. this is just for some early bootup code - * that does in_interrupt() - */ + /* Set up %gs. + * + * The base of %gs always points to the bottom of the irqstack + * union. If the stack protector canary is enabled, it is + * located at %gs:40. Note that, on SMP, the boot cpu uses + * init data section till per cpu areas are set up. + */ movl $MSR_GS_BASE,%ecx - movq $empty_zero_page,%rax + movq initial_gs(%rip),%rax movq %rax,%rdx shrq $32,%rdx wrmsr @@ -257,6 +261,8 @@ ENTRY(secondary_startup_64) .align 8 ENTRY(initial_code) .quad x86_64_start_kernel + ENTRY(initial_gs) + .quad INIT_PER_CPU_VAR(irq_stack_union) __FINITDATA ENTRY(stack_start) @@ -323,8 +329,6 @@ early_idt_ripmsg: #endif /* CONFIG_EARLY_PRINTK */ .previous -.balign PAGE_SIZE - #define NEXT_PAGE(name) \ .balign PAGE_SIZE; \ ENTRY(name) @@ -401,7 +405,8 @@ NEXT_PAGE(level2_spare_pgt) .globl early_gdt_descr early_gdt_descr: .word GDT_ENTRIES*8-1 - .quad per_cpu__gdt_page +early_gdt_descr_base: + .quad INIT_PER_CPU_VAR(gdt_page) ENTRY(phys_base) /* This must match the first entry in level2_kernel_pgt */ @@ -412,7 +417,7 @@ ENTRY(phys_base) .section .bss, "aw", @nobits .align L1_CACHE_BYTES ENTRY(idt_table) - .skip 256 * 16 + .skip IDT_ENTRIES * 16 .section .bss.page_aligned, "aw", @nobits .align PAGE_SIZE diff --git a/arch/x86/kernel/i8259.c b/arch/x86/kernel/i8259.c index 11d5093eb28..df89102bef8 100644 --- a/arch/x86/kernel/i8259.c +++ b/arch/x86/kernel/i8259.c @@ -22,7 +22,6 @@ #include <asm/pgtable.h> #include <asm/desc.h> #include <asm/apic.h> -#include <asm/arch_hooks.h> #include <asm/i8259.h> /* diff --git a/arch/x86/kernel/ioport.c b/arch/x86/kernel/ioport.c index b12208f4dfe..99c4d308f16 100644 --- a/arch/x86/kernel/ioport.c +++ b/arch/x86/kernel/ioport.c @@ -85,19 +85,8 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on) t->io_bitmap_max = bytes; -#ifdef CONFIG_X86_32 - /* - * Sets the lazy trigger so that the next I/O operation will - * reload the correct bitmap. - * Reset the owner so that a process switch will not set - * tss->io_bitmap_base to IO_BITMAP_OFFSET. - */ - tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET_LAZY; - tss->io_bitmap_owner = NULL; -#else /* Update the TSS: */ memcpy(tss->io_bitmap, t->io_bitmap_ptr, bytes_updated); -#endif put_cpu(); @@ -131,9 +120,8 @@ static int do_iopl(unsigned int level, struct pt_regs *regs) } #ifdef CONFIG_X86_32 -asmlinkage long sys_iopl(unsigned long regsp) +long sys_iopl(struct pt_regs *regs) { - struct pt_regs *regs = (struct pt_regs *)®sp; unsigned int level = regs->bx; struct thread_struct *t = ¤t->thread; int rc; diff --git a/arch/x86/kernel/ipi.c b/arch/x86/kernel/ipi.c deleted file mode 100644 index 285bbf8831f..00000000000 --- a/arch/x86/kernel/ipi.c +++ /dev/null @@ -1,190 +0,0 @@ -#include <linux/cpumask.h> -#include <linux/interrupt.h> -#include <linux/init.h> - -#include <linux/mm.h> -#include <linux/delay.h> -#include <linux/spinlock.h> -#include <linux/kernel_stat.h> -#include <linux/mc146818rtc.h> -#include <linux/cache.h> -#include <linux/cpu.h> -#include <linux/module.h> - -#include <asm/smp.h> -#include <asm/mtrr.h> -#include <asm/tlbflush.h> -#include <asm/mmu_context.h> -#include <asm/apic.h> -#include <asm/proto.h> - -#ifdef CONFIG_X86_32 -#include <mach_apic.h> -#include <mach_ipi.h> - -/* - * the following functions deal with sending IPIs between CPUs. - * - * We use 'broadcast', CPU->CPU IPIs and self-IPIs too. - */ - -static inline int __prepare_ICR(unsigned int shortcut, int vector) -{ - unsigned int icr = shortcut | APIC_DEST_LOGICAL; - - switch (vector) { - default: - icr |= APIC_DM_FIXED | vector; - break; - case NMI_VECTOR: - icr |= APIC_DM_NMI; - break; - } - return icr; -} - -static inline int __prepare_ICR2(unsigned int mask) -{ - return SET_APIC_DEST_FIELD(mask); -} - -void __send_IPI_shortcut(unsigned int shortcut, int vector) -{ - /* - * Subtle. In the case of the 'never do double writes' workaround - * we have to lock out interrupts to be safe. As we don't care - * of the value read we use an atomic rmw access to avoid costly - * cli/sti. Otherwise we use an even cheaper single atomic write - * to the APIC. - */ - unsigned int cfg; - - /* - * Wait for idle. - */ - apic_wait_icr_idle(); - - /* - * No need to touch the target chip field - */ - cfg = __prepare_ICR(shortcut, vector); - - /* - * Send the IPI. The write to APIC_ICR fires this off. - */ - apic_write(APIC_ICR, cfg); -} - -void send_IPI_self(int vector) -{ - __send_IPI_shortcut(APIC_DEST_SELF, vector); -} - -/* - * This is used to send an IPI with no shorthand notation (the destination is - * specified in bits 56 to 63 of the ICR). - */ -static inline void __send_IPI_dest_field(unsigned long mask, int vector) -{ - unsigned long cfg; - - /* - * Wait for idle. - */ - if (unlikely(vector == NMI_VECTOR)) - safe_apic_wait_icr_idle(); - else - apic_wait_icr_idle(); - - /* - * prepare target chip field - */ - cfg = __prepare_ICR2(mask); - apic_write(APIC_ICR2, cfg); - - /* - * program the ICR - */ - cfg = __prepare_ICR(0, vector); - - /* - * Send the IPI. The write to APIC_ICR fires this off. - */ - apic_write(APIC_ICR, cfg); -} - -/* - * This is only used on smaller machines. - */ -void send_IPI_mask_bitmask(const struct cpumask *cpumask, int vector) -{ - unsigned long mask = cpumask_bits(cpumask)[0]; - unsigned long flags; - - local_irq_save(flags); - WARN_ON(mask & ~cpumask_bits(cpu_online_mask)[0]); - __send_IPI_dest_field(mask, vector); - local_irq_restore(flags); -} - -void send_IPI_mask_sequence(const struct cpumask *mask, int vector) -{ - unsigned long flags; - unsigned int query_cpu; - - /* - * Hack. The clustered APIC addressing mode doesn't allow us to send - * to an arbitrary mask, so I do a unicasts to each CPU instead. This - * should be modified to do 1 message per cluster ID - mbligh - */ - - local_irq_save(flags); - for_each_cpu(query_cpu, mask) - __send_IPI_dest_field(cpu_to_logical_apicid(query_cpu), vector); - local_irq_restore(flags); -} - -void send_IPI_mask_allbutself(const struct cpumask *mask, int vector) -{ - unsigned long flags; - unsigned int query_cpu; - unsigned int this_cpu = smp_processor_id(); - - /* See Hack comment above */ - - local_irq_save(flags); - for_each_cpu(query_cpu, mask) - if (query_cpu != this_cpu) - __send_IPI_dest_field(cpu_to_logical_apicid(query_cpu), - vector); - local_irq_restore(flags); -} - -/* must come after the send_IPI functions above for inlining */ -static int convert_apicid_to_cpu(int apic_id) -{ - int i; - - for_each_possible_cpu(i) { - if (per_cpu(x86_cpu_to_apicid, i) == apic_id) - return i; - } - return -1; -} - -int safe_smp_processor_id(void) -{ - int apicid, cpuid; - - if (!boot_cpu_has(X86_FEATURE_APIC)) - return 0; - - apicid = hard_smp_processor_id(); - if (apicid == BAD_APICID) - return 0; - - cpuid = convert_apicid_to_cpu(apicid); - - return cpuid >= 0 ? cpuid : 0; -} -#endif diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 3973e2df7f8..b8ac3b6cf77 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -6,13 +6,18 @@ #include <linux/kernel_stat.h> #include <linux/seq_file.h> #include <linux/smp.h> +#include <linux/ftrace.h> #include <asm/apic.h> #include <asm/io_apic.h> #include <asm/irq.h> +#include <asm/idle.h> atomic_t irq_err_count; +/* Function pointer for generic interrupt vector handling */ +void (*generic_interrupt_extension)(void) = NULL; + /* * 'what should we do if we get a hw irq event on an illegal vector'. * each architecture has to answer this themselves. @@ -36,63 +41,65 @@ void ack_bad_irq(unsigned int irq) #endif } -#ifdef CONFIG_X86_32 -# define irq_stats(x) (&per_cpu(irq_stat, x)) -#else -# define irq_stats(x) cpu_pda(x) -#endif +#define irq_stats(x) (&per_cpu(irq_stat, x)) /* * /proc/interrupts printing: */ -static int show_other_interrupts(struct seq_file *p) +static int show_other_interrupts(struct seq_file *p, int prec) { int j; - seq_printf(p, "NMI: "); + seq_printf(p, "%*s: ", prec, "NMI"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->__nmi_count); seq_printf(p, " Non-maskable interrupts\n"); #ifdef CONFIG_X86_LOCAL_APIC - seq_printf(p, "LOC: "); + seq_printf(p, "%*s: ", prec, "LOC"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->apic_timer_irqs); seq_printf(p, " Local timer interrupts\n"); #endif + if (generic_interrupt_extension) { + seq_printf(p, "PLT: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->generic_irqs); + seq_printf(p, " Platform interrupts\n"); + } #ifdef CONFIG_SMP - seq_printf(p, "RES: "); + seq_printf(p, "%*s: ", prec, "RES"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->irq_resched_count); seq_printf(p, " Rescheduling interrupts\n"); - seq_printf(p, "CAL: "); + seq_printf(p, "%*s: ", prec, "CAL"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->irq_call_count); seq_printf(p, " Function call interrupts\n"); - seq_printf(p, "TLB: "); + seq_printf(p, "%*s: ", prec, "TLB"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->irq_tlb_count); seq_printf(p, " TLB shootdowns\n"); #endif #ifdef CONFIG_X86_MCE - seq_printf(p, "TRM: "); + 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_64 - seq_printf(p, "THR: "); + 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_LOCAL_APIC - seq_printf(p, "SPU: "); + seq_printf(p, "%*s: ", prec, "SPU"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->irq_spurious_count); seq_printf(p, " Spurious interrupts\n"); #endif - seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); + seq_printf(p, "%*s: %10u\n", prec, "ERR", atomic_read(&irq_err_count)); #if defined(CONFIG_X86_IO_APIC) - seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count)); + seq_printf(p, "%*s: %10u\n", prec, "MIS", atomic_read(&irq_mis_count)); #endif return 0; } @@ -100,19 +107,22 @@ static int show_other_interrupts(struct seq_file *p) int show_interrupts(struct seq_file *p, void *v) { unsigned long flags, any_count = 0; - int i = *(loff_t *) v, j; + int i = *(loff_t *) v, j, prec; struct irqaction *action; struct irq_desc *desc; if (i > nr_irqs) return 0; + for (prec = 3, j = 1000; prec < 10 && j <= nr_irqs; ++prec) + j *= 10; + if (i == nr_irqs) - return show_other_interrupts(p); + return show_other_interrupts(p, prec); /* print header */ if (i == 0) { - seq_printf(p, " "); + seq_printf(p, "%*s", prec + 8, ""); for_each_online_cpu(j) seq_printf(p, "CPU%-8d", j); seq_putc(p, '\n'); @@ -133,7 +143,7 @@ int show_interrupts(struct seq_file *p, void *v) if (!action && !any_count) goto out; - seq_printf(p, "%3d: ", i); + seq_printf(p, "%*d: ", prec, i); #ifndef CONFIG_SMP seq_printf(p, "%10u ", kstat_irqs(i)); #else @@ -165,6 +175,8 @@ u64 arch_irq_stat_cpu(unsigned int cpu) #ifdef CONFIG_X86_LOCAL_APIC sum += irq_stats(cpu)->apic_timer_irqs; #endif + if (generic_interrupt_extension) + sum += irq_stats(cpu)->generic_irqs; #ifdef CONFIG_SMP sum += irq_stats(cpu)->irq_resched_count; sum += irq_stats(cpu)->irq_call_count; @@ -192,4 +204,63 @@ u64 arch_irq_stat(void) return sum; } + +/* + * do_IRQ handles all normal device IRQ's (the special + * SMP cross-CPU interrupts have their own specific + * handlers). + */ +unsigned int __irq_entry do_IRQ(struct pt_regs *regs) +{ + struct pt_regs *old_regs = set_irq_regs(regs); + + /* high bit used in ret_from_ code */ + unsigned vector = ~regs->orig_ax; + unsigned irq; + + exit_idle(); + irq_enter(); + + irq = __get_cpu_var(vector_irq)[vector]; + + if (!handle_irq(irq, regs)) { +#ifdef CONFIG_X86_64 + if (!disable_apic) + ack_APIC_irq(); +#endif + + if (printk_ratelimit()) + printk(KERN_EMERG "%s: %d.%d No irq handler for vector (irq %d)\n", + __func__, smp_processor_id(), vector, irq); + } + + irq_exit(); + + set_irq_regs(old_regs); + return 1; +} + +/* + * Handler for GENERIC_INTERRUPT_VECTOR. + */ +void smp_generic_interrupt(struct pt_regs *regs) +{ + struct pt_regs *old_regs = set_irq_regs(regs); + + ack_APIC_irq(); + + exit_idle(); + + irq_enter(); + + inc_irq_stat(generic_irqs); + + if (generic_interrupt_extension) + generic_interrupt_extension(); + + irq_exit(); + + set_irq_regs(old_regs); +} + EXPORT_SYMBOL_GPL(vector_used_by_percpu_irq); diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index 74b9ff7341e..3b09634a515 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -16,6 +16,7 @@ #include <linux/cpu.h> #include <linux/delay.h> #include <linux/uaccess.h> +#include <linux/percpu.h> #include <asm/apic.h> @@ -55,13 +56,13 @@ static inline void print_stack_overflow(void) { } union irq_ctx { struct thread_info tinfo; u32 stack[THREAD_SIZE/sizeof(u32)]; -}; +} __attribute__((aligned(PAGE_SIZE))); -static union irq_ctx *hardirq_ctx[NR_CPUS] __read_mostly; -static union irq_ctx *softirq_ctx[NR_CPUS] __read_mostly; +static DEFINE_PER_CPU(union irq_ctx *, hardirq_ctx); +static DEFINE_PER_CPU(union irq_ctx *, softirq_ctx); -static char softirq_stack[NR_CPUS * THREAD_SIZE] __page_aligned_bss; -static char hardirq_stack[NR_CPUS * THREAD_SIZE] __page_aligned_bss; +static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, hardirq_stack); +static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, softirq_stack); static void call_on_stack(void *func, void *stack) { @@ -81,7 +82,7 @@ execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq) u32 *isp, arg1, arg2; curctx = (union irq_ctx *) current_thread_info(); - irqctx = hardirq_ctx[smp_processor_id()]; + irqctx = __get_cpu_var(hardirq_ctx); /* * this is where we switch to the IRQ stack. However, if we are @@ -125,34 +126,34 @@ void __cpuinit irq_ctx_init(int cpu) { union irq_ctx *irqctx; - if (hardirq_ctx[cpu]) + if (per_cpu(hardirq_ctx, cpu)) return; - irqctx = (union irq_ctx*) &hardirq_stack[cpu*THREAD_SIZE]; + irqctx = &per_cpu(hardirq_stack, cpu); irqctx->tinfo.task = NULL; irqctx->tinfo.exec_domain = NULL; irqctx->tinfo.cpu = cpu; irqctx->tinfo.preempt_count = HARDIRQ_OFFSET; irqctx->tinfo.addr_limit = MAKE_MM_SEG(0); - hardirq_ctx[cpu] = irqctx; + per_cpu(hardirq_ctx, cpu) = irqctx; - irqctx = (union irq_ctx *) &softirq_stack[cpu*THREAD_SIZE]; + irqctx = &per_cpu(softirq_stack, cpu); irqctx->tinfo.task = NULL; irqctx->tinfo.exec_domain = NULL; irqctx->tinfo.cpu = cpu; irqctx->tinfo.preempt_count = 0; irqctx->tinfo.addr_limit = MAKE_MM_SEG(0); - softirq_ctx[cpu] = irqctx; + per_cpu(softirq_ctx, cpu) = irqctx; printk(KERN_DEBUG "CPU %u irqstacks, hard=%p soft=%p\n", - cpu, hardirq_ctx[cpu], softirq_ctx[cpu]); + cpu, per_cpu(hardirq_ctx, cpu), per_cpu(softirq_ctx, cpu)); } void irq_ctx_exit(int cpu) { - hardirq_ctx[cpu] = NULL; + per_cpu(hardirq_ctx, cpu) = NULL; } asmlinkage void do_softirq(void) @@ -169,7 +170,7 @@ asmlinkage void do_softirq(void) if (local_softirq_pending()) { curctx = current_thread_info(); - irqctx = softirq_ctx[smp_processor_id()]; + irqctx = __get_cpu_var(softirq_ctx); irqctx->tinfo.task = curctx->task; irqctx->tinfo.previous_esp = current_stack_pointer; @@ -191,33 +192,16 @@ static inline int execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq) { return 0; } #endif -/* - * do_IRQ handles all normal device IRQ's (the special - * SMP cross-CPU interrupts have their own specific - * handlers). - */ -unsigned int do_IRQ(struct pt_regs *regs) +bool handle_irq(unsigned irq, struct pt_regs *regs) { - struct pt_regs *old_regs; - /* high bit used in ret_from_ code */ - int overflow; - unsigned vector = ~regs->orig_ax; struct irq_desc *desc; - unsigned irq; - - - old_regs = set_irq_regs(regs); - irq_enter(); - irq = __get_cpu_var(vector_irq)[vector]; + int overflow; overflow = check_stack_overflow(); desc = irq_to_desc(irq); - if (unlikely(!desc)) { - printk(KERN_EMERG "%s: cannot handle IRQ %d vector %#x cpu %d\n", - __func__, irq, vector, smp_processor_id()); - BUG(); - } + if (unlikely(!desc)) + return false; if (!execute_on_irq_stack(overflow, desc, irq)) { if (unlikely(overflow)) @@ -225,13 +209,10 @@ unsigned int do_IRQ(struct pt_regs *regs) desc->handle_irq(irq, desc); } - irq_exit(); - set_irq_regs(old_regs); - return 1; + return true; } #ifdef CONFIG_HOTPLUG_CPU -#include <mach_apic.h> /* A cpu has been removed from cpu_online_mask. Reset irq affinities. */ void fixup_irqs(void) @@ -248,7 +229,7 @@ void fixup_irqs(void) if (irq == 2) continue; - affinity = &desc->affinity; + 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; diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c index 63c88e6ec02..977d8b43a0d 100644 --- a/arch/x86/kernel/irq_64.c +++ b/arch/x86/kernel/irq_64.c @@ -18,6 +18,13 @@ #include <linux/smp.h> #include <asm/io_apic.h> #include <asm/idle.h> +#include <asm/apic.h> + +DEFINE_PER_CPU_SHARED_ALIGNED(irq_cpustat_t, irq_stat); +EXPORT_PER_CPU_SYMBOL(irq_stat); + +DEFINE_PER_CPU(struct pt_regs *, irq_regs); +EXPORT_PER_CPU_SYMBOL(irq_regs); /* * Probabilistic stack overflow check: @@ -41,42 +48,18 @@ static inline void stack_overflow_check(struct pt_regs *regs) #endif } -/* - * do_IRQ handles all normal device IRQ's (the special - * SMP cross-CPU interrupts have their own specific - * handlers). - */ -asmlinkage unsigned int __irq_entry do_IRQ(struct pt_regs *regs) +bool handle_irq(unsigned irq, struct pt_regs *regs) { - struct pt_regs *old_regs = set_irq_regs(regs); struct irq_desc *desc; - /* high bit used in ret_from_ code */ - unsigned vector = ~regs->orig_ax; - unsigned irq; - - exit_idle(); - irq_enter(); - irq = __get_cpu_var(vector_irq)[vector]; - stack_overflow_check(regs); desc = irq_to_desc(irq); - if (likely(desc)) - generic_handle_irq_desc(irq, desc); - else { - if (!disable_apic) - ack_APIC_irq(); - - if (printk_ratelimit()) - printk(KERN_EMERG "%s: %d.%d No irq handler for vector\n", - __func__, smp_processor_id(), vector); - } - - irq_exit(); + if (unlikely(!desc)) + return false; - set_irq_regs(old_regs); - return 1; + generic_handle_irq_desc(irq, desc); + return true; } #ifdef CONFIG_HOTPLUG_CPU @@ -100,7 +83,7 @@ void fixup_irqs(void) /* interrupt's are disabled at this point */ spin_lock(&desc->lock); - affinity = &desc->affinity; + affinity = desc->affinity; if (!irq_has_action(irq) || cpumask_equal(affinity, cpu_online_mask)) { spin_unlock(&desc->lock); diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c index 10a09c2f182..bc132610544 100644 --- a/arch/x86/kernel/irqinit_32.c +++ b/arch/x86/kernel/irqinit_32.c @@ -18,7 +18,7 @@ #include <asm/pgtable.h> #include <asm/desc.h> #include <asm/apic.h> -#include <asm/arch_hooks.h> +#include <asm/setup.h> #include <asm/i8259.h> #include <asm/traps.h> @@ -78,6 +78,15 @@ void __init init_ISA_irqs(void) } } +/* + * IRQ2 is cascade interrupt to second interrupt controller + */ +static struct irqaction irq2 = { + .handler = no_action, + .mask = CPU_MASK_NONE, + .name = "cascade", +}; + DEFINE_PER_CPU(vector_irq_t, vector_irq) = { [0 ... IRQ0_VECTOR - 1] = -1, [IRQ0_VECTOR] = 0, @@ -118,8 +127,8 @@ void __init native_init_IRQ(void) { int i; - /* all the set up before the call gates are initialised */ - pre_intr_init_hook(); + /* Execute any quirks before the call gates are initialised: */ + x86_quirk_pre_intr_init(); /* * Cover the whole vector space, no vector can escape @@ -140,8 +149,15 @@ void __init native_init_IRQ(void) */ alloc_intr_gate(RESCHEDULE_VECTOR, reschedule_interrupt); - /* IPI for invalidation */ - alloc_intr_gate(INVALIDATE_TLB_VECTOR, invalidate_interrupt); + /* IPIs for invalidation */ + alloc_intr_gate(INVALIDATE_TLB_VECTOR_START+0, invalidate_interrupt0); + alloc_intr_gate(INVALIDATE_TLB_VECTOR_START+1, invalidate_interrupt1); + alloc_intr_gate(INVALIDATE_TLB_VECTOR_START+2, invalidate_interrupt2); + alloc_intr_gate(INVALIDATE_TLB_VECTOR_START+3, invalidate_interrupt3); + alloc_intr_gate(INVALIDATE_TLB_VECTOR_START+4, invalidate_interrupt4); + alloc_intr_gate(INVALIDATE_TLB_VECTOR_START+5, invalidate_interrupt5); + alloc_intr_gate(INVALIDATE_TLB_VECTOR_START+6, invalidate_interrupt6); + alloc_intr_gate(INVALIDATE_TLB_VECTOR_START+7, invalidate_interrupt7); /* IPI for generic function call */ alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt); @@ -159,6 +175,9 @@ void __init native_init_IRQ(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 vectors for APIC spurious and error interrupts */ alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt); alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt); @@ -169,10 +188,14 @@ void __init native_init_IRQ(void) alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt); #endif - /* setup after call gates are initialised (usually add in - * the architecture specific gates) + if (!acpi_ioapic) + setup_irq(2, &irq2); + + /* + * Call quirks after call gates are initialised (usually add in + * the architecture specific gates): */ - intr_init_hook(); + x86_quirk_intr_init(); /* * External FPU? Set up irq13 if so, for diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c index da481a1e3f3..c7a49e0ffbf 100644 --- a/arch/x86/kernel/irqinit_64.c +++ b/arch/x86/kernel/irqinit_64.c @@ -147,6 +147,9 @@ 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 vectors for APIC spurious and error interrupts */ alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt); alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt); diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index 10435a120d2..eedfaebe106 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -46,7 +46,7 @@ #include <asm/apicdef.h> #include <asm/system.h> -#include <mach_ipi.h> +#include <asm/apic.h> /* * Put the error code here just in case the user cares: @@ -347,7 +347,7 @@ void kgdb_post_primary_code(struct pt_regs *regs, int e_vector, int err_code) */ void kgdb_roundup_cpus(unsigned long flags) { - send_IPI_allbutself(APIC_DM_NMI); + apic->send_IPI_allbutself(APIC_DM_NMI); } #endif diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index e948b28a5a9..4558dd3918c 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -193,6 +193,9 @@ static int __kprobes can_boost(kprobe_opcode_t *opcodes) kprobe_opcode_t opcode; kprobe_opcode_t *orig_opcodes = opcodes; + if (search_exception_tables(opcodes)) + return 0; /* Page fault may occur on this address. */ + retry: if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1) return 0; diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index 652fce6d2cc..137f2e8132d 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -19,7 +19,6 @@ #include <linux/clocksource.h> #include <linux/kvm_para.h> #include <asm/pvclock.h> -#include <asm/arch_hooks.h> #include <asm/msr.h> #include <asm/apic.h> #include <linux/percpu.h> diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c index 37f420018a4..e7368c1da01 100644 --- a/arch/x86/kernel/machine_kexec_32.c +++ b/arch/x86/kernel/machine_kexec_32.c @@ -14,12 +14,12 @@ #include <linux/ftrace.h> #include <linux/suspend.h> #include <linux/gfp.h> +#include <linux/io.h> #include <asm/pgtable.h> #include <asm/pgalloc.h> #include <asm/tlbflush.h> #include <asm/mmu_context.h> -#include <asm/io.h> #include <asm/apic.h> #include <asm/cpufeature.h> #include <asm/desc.h> @@ -63,7 +63,7 @@ static void load_segments(void) "\tmovl %%eax,%%fs\n" "\tmovl %%eax,%%gs\n" "\tmovl %%eax,%%ss\n" - ::: "eax", "memory"); + : : : "eax", "memory"); #undef STR #undef __STR } @@ -121,7 +121,7 @@ static void machine_kexec_page_table_set_one( static void machine_kexec_prepare_page_tables(struct kimage *image) { void *control_page; - pmd_t *pmd = 0; + pmd_t *pmd = NULL; control_page = page_address(image->control_code_page); #ifdef CONFIG_X86_PAE @@ -205,7 +205,8 @@ void machine_kexec(struct kimage *image) if (image->preserve_context) { #ifdef CONFIG_X86_IO_APIC - /* We need to put APICs in legacy mode so that we can + /* + * We need to put APICs in legacy mode so that we can * get timer interrupts in second kernel. kexec/kdump * paths already have calls to disable_IO_APIC() in * one form or other. kexec jump path also need @@ -227,7 +228,8 @@ void machine_kexec(struct kimage *image) page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page) << PAGE_SHIFT); - /* The segment registers are funny things, they have both a + /* + * The segment registers are funny things, they have both a * visible and an invisible part. Whenever the visible part is * set to a specific selector, the invisible part is loaded * with from a table in memory. At no other time is the @@ -237,11 +239,12 @@ void machine_kexec(struct kimage *image) * segments, before I zap the gdt with an invalid value. */ load_segments(); - /* The gdt & idt are now invalid. + /* + * The gdt & idt are now invalid. * If you want to load them you must set up your own idt & gdt. */ - set_gdt(phys_to_virt(0),0); - set_idt(phys_to_virt(0),0); + set_gdt(phys_to_virt(0), 0); + set_idt(phys_to_virt(0), 0); /* now call it */ image->start = relocate_kernel_ptr((unsigned long)image->head, diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c index c43caa3a91f..89cea4d4467 100644 --- a/arch/x86/kernel/machine_kexec_64.c +++ b/arch/x86/kernel/machine_kexec_64.c @@ -12,20 +12,47 @@ #include <linux/reboot.h> #include <linux/numa.h> #include <linux/ftrace.h> +#include <linux/io.h> +#include <linux/suspend.h> #include <asm/pgtable.h> #include <asm/tlbflush.h> #include <asm/mmu_context.h> -#include <asm/io.h> -#define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE))) -static u64 kexec_pgd[512] PAGE_ALIGNED; -static u64 kexec_pud0[512] PAGE_ALIGNED; -static u64 kexec_pmd0[512] PAGE_ALIGNED; -static u64 kexec_pte0[512] PAGE_ALIGNED; -static u64 kexec_pud1[512] PAGE_ALIGNED; -static u64 kexec_pmd1[512] PAGE_ALIGNED; -static u64 kexec_pte1[512] PAGE_ALIGNED; +static int init_one_level2_page(struct kimage *image, pgd_t *pgd, + unsigned long addr) +{ + pud_t *pud; + pmd_t *pmd; + struct page *page; + int result = -ENOMEM; + + addr &= PMD_MASK; + pgd += pgd_index(addr); + if (!pgd_present(*pgd)) { + page = kimage_alloc_control_pages(image, 0); + if (!page) + goto out; + pud = (pud_t *)page_address(page); + memset(pud, 0, PAGE_SIZE); + set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE)); + } + pud = pud_offset(pgd, addr); + if (!pud_present(*pud)) { + page = kimage_alloc_control_pages(image, 0); + if (!page) + goto out; + pmd = (pmd_t *)page_address(page); + memset(pmd, 0, PAGE_SIZE); + set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE)); + } + pmd = pmd_offset(pud, addr); + if (!pmd_present(*pmd)) + set_pmd(pmd, __pmd(addr | __PAGE_KERNEL_LARGE_EXEC)); + result = 0; +out: + return result; +} static void init_level2_page(pmd_t *level2p, unsigned long addr) { @@ -92,9 +119,8 @@ static int init_level4_page(struct kimage *image, pgd_t *level4p, } level3p = (pud_t *)page_address(page); result = init_level3_page(image, level3p, addr, last_addr); - if (result) { + if (result) goto out; - } set_pgd(level4p++, __pgd(__pa(level3p) | _KERNPG_TABLE)); addr += PGDIR_SIZE; } @@ -107,12 +133,72 @@ out: return result; } +static void free_transition_pgtable(struct kimage *image) +{ + free_page((unsigned long)image->arch.pud); + free_page((unsigned long)image->arch.pmd); + free_page((unsigned long)image->arch.pte); +} + +static int init_transition_pgtable(struct kimage *image, pgd_t *pgd) +{ + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long vaddr, paddr; + int result = -ENOMEM; + + vaddr = (unsigned long)relocate_kernel; + paddr = __pa(page_address(image->control_code_page)+PAGE_SIZE); + pgd += pgd_index(vaddr); + if (!pgd_present(*pgd)) { + pud = (pud_t *)get_zeroed_page(GFP_KERNEL); + if (!pud) + goto err; + image->arch.pud = pud; + set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE)); + } + pud = pud_offset(pgd, vaddr); + if (!pud_present(*pud)) { + pmd = (pmd_t *)get_zeroed_page(GFP_KERNEL); + if (!pmd) + goto err; + image->arch.pmd = pmd; + set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE)); + } + pmd = pmd_offset(pud, vaddr); + if (!pmd_present(*pmd)) { + pte = (pte_t *)get_zeroed_page(GFP_KERNEL); + if (!pte) + goto err; + image->arch.pte = pte; + set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE)); + } + pte = pte_offset_kernel(pmd, vaddr); + set_pte(pte, pfn_pte(paddr >> PAGE_SHIFT, PAGE_KERNEL_EXEC)); + return 0; +err: + free_transition_pgtable(image); + return result; +} + static int init_pgtable(struct kimage *image, unsigned long start_pgtable) { pgd_t *level4p; + int result; level4p = (pgd_t *)__va(start_pgtable); - return init_level4_page(image, level4p, 0, max_pfn << PAGE_SHIFT); + result = init_level4_page(image, level4p, 0, max_pfn << PAGE_SHIFT); + if (result) + return result; + /* + * image->start may be outside 0 ~ max_pfn, for example when + * jump back to original kernel from kexeced kernel + */ + result = init_one_level2_page(image, level4p, image->start); + if (result) + return result; + return init_transition_pgtable(image, level4p); } static void set_idt(void *newidt, u16 limit) @@ -174,7 +260,7 @@ int machine_kexec_prepare(struct kimage *image) void machine_kexec_cleanup(struct kimage *image) { - return; + free_transition_pgtable(image); } /* @@ -185,36 +271,45 @@ void machine_kexec(struct kimage *image) { unsigned long page_list[PAGES_NR]; void *control_page; + int save_ftrace_enabled; - tracer_disable(); +#ifdef CONFIG_KEXEC_JUMP + if (kexec_image->preserve_context) + save_processor_state(); +#endif + + save_ftrace_enabled = __ftrace_enabled_save(); /* Interrupts aren't acceptable while we reboot */ local_irq_disable(); + if (image->preserve_context) { +#ifdef CONFIG_X86_IO_APIC + /* + * We need to put APICs in legacy mode so that we can + * get timer interrupts in second kernel. kexec/kdump + * paths already have calls to disable_IO_APIC() in + * one form or other. kexec jump path also need + * one. + */ + disable_IO_APIC(); +#endif + } + control_page = page_address(image->control_code_page) + PAGE_SIZE; - memcpy(control_page, relocate_kernel, PAGE_SIZE); + memcpy(control_page, relocate_kernel, KEXEC_CONTROL_CODE_MAX_SIZE); page_list[PA_CONTROL_PAGE] = virt_to_phys(control_page); - page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel; - page_list[PA_PGD] = virt_to_phys(&kexec_pgd); - page_list[VA_PGD] = (unsigned long)kexec_pgd; - page_list[PA_PUD_0] = virt_to_phys(&kexec_pud0); - page_list[VA_PUD_0] = (unsigned long)kexec_pud0; - page_list[PA_PMD_0] = virt_to_phys(&kexec_pmd0); - page_list[VA_PMD_0] = (unsigned long)kexec_pmd0; - page_list[PA_PTE_0] = virt_to_phys(&kexec_pte0); - page_list[VA_PTE_0] = (unsigned long)kexec_pte0; - page_list[PA_PUD_1] = virt_to_phys(&kexec_pud1); - page_list[VA_PUD_1] = (unsigned long)kexec_pud1; - page_list[PA_PMD_1] = virt_to_phys(&kexec_pmd1); - page_list[VA_PMD_1] = (unsigned long)kexec_pmd1; - page_list[PA_PTE_1] = virt_to_phys(&kexec_pte1); - page_list[VA_PTE_1] = (unsigned long)kexec_pte1; - + page_list[VA_CONTROL_PAGE] = (unsigned long)control_page; page_list[PA_TABLE_PAGE] = (unsigned long)__pa(page_address(image->control_code_page)); - /* The segment registers are funny things, they have both a + if (image->type == KEXEC_TYPE_DEFAULT) + page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page) + << PAGE_SHIFT); + + /* + * The segment registers are funny things, they have both a * visible and an invisible part. Whenever the visible part is * set to a specific selector, the invisible part is loaded * with from a table in memory. At no other time is the @@ -224,15 +319,25 @@ void machine_kexec(struct kimage *image) * segments, before I zap the gdt with an invalid value. */ load_segments(); - /* The gdt & idt are now invalid. + /* + * The gdt & idt are now invalid. * If you want to load them you must set up your own idt & gdt. */ - set_gdt(phys_to_virt(0),0); - set_idt(phys_to_virt(0),0); + set_gdt(phys_to_virt(0), 0); + set_idt(phys_to_virt(0), 0); /* now call it */ - relocate_kernel((unsigned long)image->head, (unsigned long)page_list, - image->start); + image->start = relocate_kernel((unsigned long)image->head, + (unsigned long)page_list, + image->start, + image->preserve_context); + +#ifdef CONFIG_KEXEC_JUMP + if (kexec_image->preserve_context) + restore_processor_state(); +#endif + + __ftrace_enabled_restore(save_ftrace_enabled); } void arch_crash_save_vmcoreinfo(void) diff --git a/arch/x86/kernel/mca_32.c b/arch/x86/kernel/mca_32.c index 2dc183758be..845d80ce1ef 100644 --- a/arch/x86/kernel/mca_32.c +++ b/arch/x86/kernel/mca_32.c @@ -51,7 +51,6 @@ #include <linux/ioport.h> #include <asm/uaccess.h> #include <linux/init.h> -#include <asm/arch_hooks.h> static unsigned char which_scsi; @@ -474,6 +473,4 @@ void __kprobes mca_handle_nmi(void) * adapter was responsible for the error. */ bus_for_each_dev(&mca_bus_type, NULL, NULL, mca_handle_nmi_callback); - - mca_nmi_hook(); -} /* mca_handle_nmi */ +} diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index b7f4c929e61..5e9f4fc5138 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -87,9 +87,9 @@ #include <linux/cpu.h> #include <linux/firmware.h> #include <linux/platform_device.h> +#include <linux/uaccess.h> #include <asm/msr.h> -#include <asm/uaccess.h> #include <asm/processor.h> #include <asm/microcode.h> @@ -196,7 +196,7 @@ static inline int update_match_cpu(struct cpu_signature *csig, int sig, int pf) return (!sigmatch(sig, csig->sig, pf, csig->pf)) ? 0 : 1; } -static inline int +static inline int update_match_revision(struct microcode_header_intel *mc_header, int rev) { return (mc_header->rev <= rev) ? 0 : 1; @@ -442,8 +442,8 @@ static int request_microcode_fw(int cpu, struct device *device) return ret; } - ret = generic_load_microcode(cpu, (void*)firmware->data, firmware->size, - &get_ucode_fw); + ret = generic_load_microcode(cpu, (void *)firmware->data, + firmware->size, &get_ucode_fw); release_firmware(firmware); @@ -460,7 +460,7 @@ static int request_microcode_user(int cpu, const void __user *buf, size_t size) /* We should bind the task to the CPU */ BUG_ON(cpu != raw_smp_processor_id()); - return generic_load_microcode(cpu, (void*)buf, size, &get_ucode_user); + return generic_load_microcode(cpu, (void *)buf, size, &get_ucode_user); } static void microcode_fini_cpu(int cpu) diff --git a/arch/x86/kernel/mmconf-fam10h_64.c b/arch/x86/kernel/mmconf-fam10h_64.c index 666e43df51f..712d15fdc41 100644 --- a/arch/x86/kernel/mmconf-fam10h_64.c +++ b/arch/x86/kernel/mmconf-fam10h_64.c @@ -226,7 +226,7 @@ static int __devinit set_check_enable_amd_mmconf(const struct dmi_system_id *d) return 0; } -static struct dmi_system_id __devinitdata mmconf_dmi_table[] = { +static const struct dmi_system_id __cpuinitconst mmconf_dmi_table[] = { { .callback = set_check_enable_amd_mmconf, .ident = "Sun Microsystems Machine", diff --git a/arch/x86/kernel/module_32.c b/arch/x86/kernel/module_32.c index 3db0a5442eb..0edd819050e 100644 --- a/arch/x86/kernel/module_32.c +++ b/arch/x86/kernel/module_32.c @@ -42,7 +42,7 @@ void module_free(struct module *mod, void *module_region) { vfree(module_region); /* FIXME: If module_region == mod->init_region, trim exception - table entries. */ + table entries. */ } /* We don't need anything special. */ @@ -113,13 +113,13 @@ int module_finalize(const Elf_Ehdr *hdr, *para = NULL; char *secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; - for (s = sechdrs; s < sechdrs + hdr->e_shnum; s++) { + for (s = sechdrs; s < sechdrs + hdr->e_shnum; s++) { if (!strcmp(".text", secstrings + s->sh_name)) text = s; if (!strcmp(".altinstructions", secstrings + s->sh_name)) alt = s; if (!strcmp(".smp_locks", secstrings + s->sh_name)) - locks= s; + locks = s; if (!strcmp(".parainstructions", secstrings + s->sh_name)) para = s; } diff --git a/arch/x86/kernel/module_64.c b/arch/x86/kernel/module_64.c index 6ba87830d4b..c23880b90b5 100644 --- a/arch/x86/kernel/module_64.c +++ b/arch/x86/kernel/module_64.c @@ -30,14 +30,14 @@ #include <asm/page.h> #include <asm/pgtable.h> -#define DEBUGP(fmt...) +#define DEBUGP(fmt...) #ifndef CONFIG_UML void module_free(struct module *mod, void *module_region) { vfree(module_region); /* FIXME: If module_region == mod->init_region, trim exception - table entries. */ + table entries. */ } void *module_alloc(unsigned long size) @@ -77,7 +77,7 @@ int apply_relocate_add(Elf64_Shdr *sechdrs, Elf64_Rela *rel = (void *)sechdrs[relsec].sh_addr; Elf64_Sym *sym; void *loc; - u64 val; + u64 val; DEBUGP("Applying relocate section %u to %u\n", relsec, sechdrs[relsec].sh_info); @@ -91,11 +91,11 @@ int apply_relocate_add(Elf64_Shdr *sechdrs, sym = (Elf64_Sym *)sechdrs[symindex].sh_addr + ELF64_R_SYM(rel[i].r_info); - DEBUGP("type %d st_value %Lx r_addend %Lx loc %Lx\n", - (int)ELF64_R_TYPE(rel[i].r_info), - sym->st_value, rel[i].r_addend, (u64)loc); + DEBUGP("type %d st_value %Lx r_addend %Lx loc %Lx\n", + (int)ELF64_R_TYPE(rel[i].r_info), + sym->st_value, rel[i].r_addend, (u64)loc); - val = sym->st_value + rel[i].r_addend; + val = sym->st_value + rel[i].r_addend; switch (ELF64_R_TYPE(rel[i].r_info)) { case R_X86_64_NONE: @@ -113,16 +113,16 @@ int apply_relocate_add(Elf64_Shdr *sechdrs, if ((s64)val != *(s32 *)loc) goto overflow; break; - case R_X86_64_PC32: + case R_X86_64_PC32: val -= (u64)loc; *(u32 *)loc = val; #if 0 if ((s64)val != *(s32 *)loc) - goto overflow; + goto overflow; #endif break; default: - printk(KERN_ERR "module %s: Unknown rela relocation: %Lu\n", + printk(KERN_ERR "module %s: Unknown rela relocation: %llu\n", me->name, ELF64_R_TYPE(rel[i].r_info)); return -ENOEXEC; } @@ -130,7 +130,7 @@ int apply_relocate_add(Elf64_Shdr *sechdrs, return 0; overflow: - printk(KERN_ERR "overflow in relocation type %d val %Lx\n", + printk(KERN_ERR "overflow in relocation type %d val %Lx\n", (int)ELF64_R_TYPE(rel[i].r_info), val); printk(KERN_ERR "`%s' likely not compiled with -mcmodel=kernel\n", me->name); @@ -143,13 +143,13 @@ int apply_relocate(Elf_Shdr *sechdrs, unsigned int relsec, struct module *me) { - printk("non add relocation not supported\n"); + printk(KERN_ERR "non add relocation not supported\n"); return -ENOSYS; -} +} int module_finalize(const Elf_Ehdr *hdr, - const Elf_Shdr *sechdrs, - struct module *me) + const Elf_Shdr *sechdrs, + struct module *me) { const Elf_Shdr *s, *text = NULL, *alt = NULL, *locks = NULL, *para = NULL; @@ -161,7 +161,7 @@ int module_finalize(const Elf_Ehdr *hdr, if (!strcmp(".altinstructions", secstrings + s->sh_name)) alt = s; if (!strcmp(".smp_locks", secstrings + s->sh_name)) - locks= s; + locks = s; if (!strcmp(".parainstructions", secstrings + s->sh_name)) para = s; } diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index a649a4ccad4..47673e02ae5 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -3,7 +3,7 @@ * compliant MP-table parsing routines. * * (c) 1995 Alan Cox, Building #3 <alan@lxorguk.ukuu.org.uk> - * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com> + * (c) 1998, 1999, 2000, 2009 Ingo Molnar <mingo@redhat.com> * (c) 2008 Alexey Starikovskiy <astarikovskiy@suse.de> */ @@ -29,12 +29,7 @@ #include <asm/setup.h> #include <asm/smp.h> -#include <mach_apic.h> -#ifdef CONFIG_X86_32 -#include <mach_apicdef.h> -#include <mach_mpparse.h> -#endif - +#include <asm/apic.h> /* * Checksum an MP configuration block. */ @@ -144,11 +139,11 @@ static void __init MP_ioapic_info(struct mpc_ioapic *m) if (bad_ioapic(m->apicaddr)) return; - mp_ioapics[nr_ioapics].mp_apicaddr = m->apicaddr; - mp_ioapics[nr_ioapics].mp_apicid = m->apicid; - mp_ioapics[nr_ioapics].mp_type = m->type; - mp_ioapics[nr_ioapics].mp_apicver = m->apicver; - mp_ioapics[nr_ioapics].mp_flags = m->flags; + mp_ioapics[nr_ioapics].apicaddr = m->apicaddr; + mp_ioapics[nr_ioapics].apicid = m->apicid; + mp_ioapics[nr_ioapics].type = m->type; + mp_ioapics[nr_ioapics].apicver = m->apicver; + mp_ioapics[nr_ioapics].flags = m->flags; nr_ioapics++; } @@ -160,55 +155,55 @@ static void print_MP_intsrc_info(struct mpc_intsrc *m) m->srcbusirq, m->dstapic, m->dstirq); } -static void __init print_mp_irq_info(struct mp_config_intsrc *mp_irq) +static void __init print_mp_irq_info(struct mpc_intsrc *mp_irq) { apic_printk(APIC_VERBOSE, "Int: type %d, pol %d, trig %d, bus %02x," " IRQ %02x, APIC ID %x, APIC INT %02x\n", - mp_irq->mp_irqtype, mp_irq->mp_irqflag & 3, - (mp_irq->mp_irqflag >> 2) & 3, mp_irq->mp_srcbus, - mp_irq->mp_srcbusirq, mp_irq->mp_dstapic, mp_irq->mp_dstirq); + mp_irq->irqtype, mp_irq->irqflag & 3, + (mp_irq->irqflag >> 2) & 3, mp_irq->srcbus, + mp_irq->srcbusirq, mp_irq->dstapic, mp_irq->dstirq); } static void __init assign_to_mp_irq(struct mpc_intsrc *m, - struct mp_config_intsrc *mp_irq) + struct mpc_intsrc *mp_irq) { - mp_irq->mp_dstapic = m->dstapic; - mp_irq->mp_type = m->type; - mp_irq->mp_irqtype = m->irqtype; - mp_irq->mp_irqflag = m->irqflag; - mp_irq->mp_srcbus = m->srcbus; - mp_irq->mp_srcbusirq = m->srcbusirq; - mp_irq->mp_dstirq = m->dstirq; + mp_irq->dstapic = m->dstapic; + mp_irq->type = m->type; + mp_irq->irqtype = m->irqtype; + mp_irq->irqflag = m->irqflag; + mp_irq->srcbus = m->srcbus; + mp_irq->srcbusirq = m->srcbusirq; + mp_irq->dstirq = m->dstirq; } -static void __init assign_to_mpc_intsrc(struct mp_config_intsrc *mp_irq, +static void __init assign_to_mpc_intsrc(struct mpc_intsrc *mp_irq, struct mpc_intsrc *m) { - m->dstapic = mp_irq->mp_dstapic; - m->type = mp_irq->mp_type; - m->irqtype = mp_irq->mp_irqtype; - m->irqflag = mp_irq->mp_irqflag; - m->srcbus = mp_irq->mp_srcbus; - m->srcbusirq = mp_irq->mp_srcbusirq; - m->dstirq = mp_irq->mp_dstirq; + m->dstapic = mp_irq->dstapic; + m->type = mp_irq->type; + m->irqtype = mp_irq->irqtype; + m->irqflag = mp_irq->irqflag; + m->srcbus = mp_irq->srcbus; + m->srcbusirq = mp_irq->srcbusirq; + m->dstirq = mp_irq->dstirq; } -static int __init mp_irq_mpc_intsrc_cmp(struct mp_config_intsrc *mp_irq, +static int __init mp_irq_mpc_intsrc_cmp(struct mpc_intsrc *mp_irq, struct mpc_intsrc *m) { - if (mp_irq->mp_dstapic != m->dstapic) + if (mp_irq->dstapic != m->dstapic) return 1; - if (mp_irq->mp_type != m->type) + if (mp_irq->type != m->type) return 2; - if (mp_irq->mp_irqtype != m->irqtype) + if (mp_irq->irqtype != m->irqtype) return 3; - if (mp_irq->mp_irqflag != m->irqflag) + if (mp_irq->irqflag != m->irqflag) return 4; - if (mp_irq->mp_srcbus != m->srcbus) + if (mp_irq->srcbus != m->srcbus) return 5; - if (mp_irq->mp_srcbusirq != m->srcbusirq) + if (mp_irq->srcbusirq != m->srcbusirq) return 6; - if (mp_irq->mp_dstirq != m->dstirq) + if (mp_irq->dstirq != m->dstirq) return 7; return 0; @@ -292,16 +287,7 @@ static int __init smp_read_mpc(struct mpc_table *mpc, unsigned early) return 0; #ifdef CONFIG_X86_32 - /* - * need to make sure summit and es7000's mps_oem_check is safe to be - * called early via genericarch 's mps_oem_check - */ - if (early) { -#ifdef CONFIG_X86_NUMAQ - numaq_mps_oem_check(mpc, oem, str); -#endif - } else - mps_oem_check(mpc, oem, str); + generic_mps_oem_check(mpc, oem, str); #endif /* save the local APIC address, it might be non-default */ if (!acpi_lapic) @@ -386,13 +372,13 @@ static int __init smp_read_mpc(struct mpc_table *mpc, unsigned early) (*x86_quirks->mpc_record)++; } -#ifdef CONFIG_X86_GENERICARCH - generic_bigsmp_probe(); +#ifdef CONFIG_X86_BIGSMP + generic_bigsmp_probe(); #endif -#ifdef CONFIG_X86_32 - setup_apic_routing(); -#endif + if (apic->setup_apic_routing) + apic->setup_apic_routing(); + if (!num_processors) printk(KERN_ERR "MPTABLE: no processors registered!\n"); return num_processors; @@ -417,7 +403,7 @@ static void __init construct_default_ioirq_mptable(int mpc_default_type) intsrc.type = MP_INTSRC; intsrc.irqflag = 0; /* conforming */ intsrc.srcbus = 0; - intsrc.dstapic = mp_ioapics[0].mp_apicid; + intsrc.dstapic = mp_ioapics[0].apicid; intsrc.irqtype = mp_INT; @@ -570,14 +556,27 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type) } } -static struct intel_mp_floating *mpf_found; +static struct mpf_intel *mpf_found; + +static unsigned long __init get_mpc_size(unsigned long physptr) +{ + struct mpc_table *mpc; + unsigned long size; + + mpc = early_ioremap(physptr, PAGE_SIZE); + size = mpc->length; + early_iounmap(mpc, PAGE_SIZE); + apic_printk(APIC_VERBOSE, " mpc: %lx-%lx\n", physptr, physptr + size); + + return size; +} /* * Scan the memory blocks for an SMP configuration block. */ static void __init __get_smp_config(unsigned int early) { - struct intel_mp_floating *mpf = mpf_found; + struct mpf_intel *mpf = mpf_found; if (!mpf) return; @@ -598,9 +597,9 @@ static void __init __get_smp_config(unsigned int early) } printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n", - mpf->mpf_specification); + mpf->specification); #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86_32) - if (mpf->mpf_feature2 & (1 << 7)) { + if (mpf->feature2 & (1 << 7)) { printk(KERN_INFO " IMCR and PIC compatibility mode.\n"); pic_mode = 1; } else { @@ -611,7 +610,7 @@ static void __init __get_smp_config(unsigned int early) /* * Now see if we need to read further. */ - if (mpf->mpf_feature1 != 0) { + if (mpf->feature1 != 0) { if (early) { /* * local APIC has default address @@ -621,16 +620,20 @@ static void __init __get_smp_config(unsigned int early) } printk(KERN_INFO "Default MP configuration #%d\n", - mpf->mpf_feature1); - construct_default_ISA_mptable(mpf->mpf_feature1); + mpf->feature1); + construct_default_ISA_mptable(mpf->feature1); - } else if (mpf->mpf_physptr) { + } else if (mpf->physptr) { + struct mpc_table *mpc; + unsigned long size; + size = get_mpc_size(mpf->physptr); + mpc = early_ioremap(mpf->physptr, size); /* * Read the physical hardware table. Anything here will * override the defaults. */ - if (!smp_read_mpc(phys_to_virt(mpf->mpf_physptr), early)) { + if (!smp_read_mpc(mpc, early)) { #ifdef CONFIG_X86_LOCAL_APIC smp_found_config = 0; #endif @@ -638,8 +641,10 @@ static void __init __get_smp_config(unsigned int early) "BIOS bug, MP table errors detected!...\n"); printk(KERN_ERR "... disabling SMP support. " "(tell your hw vendor)\n"); + early_iounmap(mpc, size); return; } + early_iounmap(mpc, size); if (early) return; @@ -688,33 +693,33 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, unsigned reserve) { unsigned int *bp = phys_to_virt(base); - struct intel_mp_floating *mpf; + struct mpf_intel *mpf; apic_printk(APIC_VERBOSE, "Scan SMP from %p for %ld bytes.\n", bp, length); BUILD_BUG_ON(sizeof(*mpf) != 16); while (length > 0) { - mpf = (struct intel_mp_floating *)bp; + mpf = (struct mpf_intel *)bp; if ((*bp == SMP_MAGIC_IDENT) && - (mpf->mpf_length == 1) && + (mpf->length == 1) && !mpf_checksum((unsigned char *)bp, 16) && - ((mpf->mpf_specification == 1) - || (mpf->mpf_specification == 4))) { + ((mpf->specification == 1) + || (mpf->specification == 4))) { #ifdef CONFIG_X86_LOCAL_APIC smp_found_config = 1; #endif mpf_found = mpf; - printk(KERN_INFO "found SMP MP-table at [%p] %08lx\n", - mpf, virt_to_phys(mpf)); + 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), PAGE_SIZE, + reserve_bootmem_generic(virt_to_phys(mpf), sizeof(*mpf), BOOTMEM_DEFAULT); - if (mpf->mpf_physptr) { - unsigned long size = PAGE_SIZE; + if (mpf->physptr) { + unsigned long size = get_mpc_size(mpf->physptr); #ifdef CONFIG_X86_32 /* * We cannot access to MPC table to compute @@ -722,15 +727,24 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, * 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 mpg->mpf_physptr yields BUG() + * 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->mpf_physptr + size > end) - size = end - mpf->mpf_physptr; -#endif - reserve_bootmem_generic(mpf->mpf_physptr, 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 } return 1; @@ -809,15 +823,15 @@ static int __init get_MP_intsrc_index(struct mpc_intsrc *m) /* not legacy */ for (i = 0; i < mp_irq_entries; i++) { - if (mp_irqs[i].mp_irqtype != mp_INT) + if (mp_irqs[i].irqtype != mp_INT) continue; - if (mp_irqs[i].mp_irqflag != 0x0f) + if (mp_irqs[i].irqflag != 0x0f) continue; - if (mp_irqs[i].mp_srcbus != m->srcbus) + if (mp_irqs[i].srcbus != m->srcbus) continue; - if (mp_irqs[i].mp_srcbusirq != m->srcbusirq) + if (mp_irqs[i].srcbusirq != m->srcbusirq) continue; if (irq_used[i]) { /* already claimed */ @@ -876,12 +890,12 @@ static int __init replace_intsrc_all(struct mpc_table *mpc, #ifdef CONFIG_X86_IO_APIC struct mpc_intsrc *m = (struct mpc_intsrc *)mpt; - printk(KERN_INFO "OLD "); + apic_printk(APIC_VERBOSE, "OLD "); print_MP_intsrc_info(m); i = get_MP_intsrc_index(m); if (i > 0) { assign_to_mpc_intsrc(&mp_irqs[i], m); - printk(KERN_INFO "NEW "); + apic_printk(APIC_VERBOSE, "NEW "); print_mp_irq_info(&mp_irqs[i]); } else if (!i) { /* legacy, do nothing */ @@ -922,14 +936,14 @@ static int __init replace_intsrc_all(struct mpc_table *mpc, if (irq_used[i]) continue; - if (mp_irqs[i].mp_irqtype != mp_INT) + if (mp_irqs[i].irqtype != mp_INT) continue; - if (mp_irqs[i].mp_irqflag != 0x0f) + if (mp_irqs[i].irqflag != 0x0f) continue; if (nr_m_spare > 0) { - printk(KERN_INFO "*NEW* found "); + apic_printk(APIC_VERBOSE, "*NEW* found\n"); nr_m_spare--; assign_to_mpc_intsrc(&mp_irqs[i], m_spare[nr_m_spare]); m_spare[nr_m_spare] = NULL; @@ -1001,7 +1015,7 @@ static int __init update_mp_table(void) { char str[16]; char oem[10]; - struct intel_mp_floating *mpf; + struct mpf_intel *mpf; struct mpc_table *mpc, *mpc_new; if (!enable_update_mptable) @@ -1014,19 +1028,19 @@ static int __init update_mp_table(void) /* * Now see if we need to go further. */ - if (mpf->mpf_feature1 != 0) + if (mpf->feature1 != 0) return 0; - if (!mpf->mpf_physptr) + if (!mpf->physptr) return 0; - mpc = phys_to_virt(mpf->mpf_physptr); + mpc = phys_to_virt(mpf->physptr); if (!smp_check_mpc(mpc, oem, str)) return 0; - printk(KERN_INFO "mpf: %lx\n", virt_to_phys(mpf)); - printk(KERN_INFO "mpf_physptr: %x\n", mpf->mpf_physptr); + printk(KERN_INFO "mpf: %llx\n", (u64)virt_to_phys(mpf)); + printk(KERN_INFO "physptr: %x\n", mpf->physptr); if (mpc_new_phys && mpc->length > mpc_new_length) { mpc_new_phys = 0; @@ -1047,23 +1061,23 @@ static int __init update_mp_table(void) } printk(KERN_INFO "use in-positon replacing\n"); } else { - mpf->mpf_physptr = mpc_new_phys; + mpf->physptr = mpc_new_phys; mpc_new = phys_to_virt(mpc_new_phys); memcpy(mpc_new, mpc, mpc->length); mpc = mpc_new; /* check if we can modify that */ - if (mpc_new_phys - mpf->mpf_physptr) { - struct intel_mp_floating *mpf_new; + if (mpc_new_phys - mpf->physptr) { + struct mpf_intel *mpf_new; /* steal 16 bytes from [0, 1k) */ printk(KERN_INFO "mpf new: %x\n", 0x400 - 16); mpf_new = phys_to_virt(0x400 - 16); memcpy(mpf_new, mpf, 16); mpf = mpf_new; - mpf->mpf_physptr = mpc_new_phys; + mpf->physptr = mpc_new_phys; } - mpf->mpf_checksum = 0; - mpf->mpf_checksum -= mpf_checksum((unsigned char *)mpf, 16); - printk(KERN_INFO "mpf_physptr new: %x\n", mpf->mpf_physptr); + mpf->checksum = 0; + mpf->checksum -= mpf_checksum((unsigned char *)mpf, 16); + printk(KERN_INFO "physptr new: %x\n", mpf->physptr); } /* diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c index 726266695b2..3cf3413ec62 100644 --- a/arch/x86/kernel/msr.c +++ b/arch/x86/kernel/msr.c @@ -35,10 +35,10 @@ #include <linux/device.h> #include <linux/cpu.h> #include <linux/notifier.h> +#include <linux/uaccess.h> #include <asm/processor.h> #include <asm/msr.h> -#include <asm/uaccess.h> #include <asm/system.h> static struct class *msr_class; diff --git a/arch/x86/kernel/numaq_32.c b/arch/x86/kernel/numaq_32.c deleted file mode 100644 index f2191d4f271..00000000000 --- a/arch/x86/kernel/numaq_32.c +++ /dev/null @@ -1,293 +0,0 @@ -/* - * Written by: Patricia Gaughen, IBM Corporation - * - * Copyright (C) 2002, IBM Corp. - * - * All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or - * NON INFRINGEMENT. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to <gone@us.ibm.com> - */ - -#include <linux/mm.h> -#include <linux/bootmem.h> -#include <linux/mmzone.h> -#include <linux/module.h> -#include <linux/nodemask.h> -#include <asm/numaq.h> -#include <asm/topology.h> -#include <asm/processor.h> -#include <asm/genapic.h> -#include <asm/e820.h> -#include <asm/setup.h> - -#define MB_TO_PAGES(addr) ((addr) << (20 - PAGE_SHIFT)) - -/* - * Function: smp_dump_qct() - * - * Description: gets memory layout from the quad config table. This - * function also updates node_online_map with the nodes (quads) present. - */ -static void __init smp_dump_qct(void) -{ - int node; - struct eachquadmem *eq; - struct sys_cfg_data *scd = - (struct sys_cfg_data *)__va(SYS_CFG_DATA_PRIV_ADDR); - - nodes_clear(node_online_map); - for_each_node(node) { - if (scd->quads_present31_0 & (1 << node)) { - node_set_online(node); - eq = &scd->eq[node]; - /* Convert to pages */ - node_start_pfn[node] = MB_TO_PAGES( - eq->hi_shrd_mem_start - eq->priv_mem_size); - node_end_pfn[node] = MB_TO_PAGES( - eq->hi_shrd_mem_start + eq->hi_shrd_mem_size); - - e820_register_active_regions(node, node_start_pfn[node], - node_end_pfn[node]); - memory_present(node, - node_start_pfn[node], node_end_pfn[node]); - node_remap_size[node] = node_memmap_size_bytes(node, - node_start_pfn[node], - node_end_pfn[node]); - } - } -} - - -void __cpuinit numaq_tsc_disable(void) -{ - if (!found_numaq) - return; - - if (num_online_nodes() > 1) { - printk(KERN_DEBUG "NUMAQ: disabling TSC\n"); - setup_clear_cpu_cap(X86_FEATURE_TSC); - } -} - -static int __init numaq_pre_time_init(void) -{ - numaq_tsc_disable(); - return 0; -} - -int found_numaq; -/* - * Have to match translation table entries to main table entries by counter - * hence the mpc_record variable .... can't see a less disgusting way of - * doing this .... - */ -struct mpc_config_translation { - unsigned char mpc_type; - unsigned char trans_len; - unsigned char trans_type; - unsigned char trans_quad; - unsigned char trans_global; - unsigned char trans_local; - unsigned short trans_reserved; -}; - -/* x86_quirks member */ -static int mpc_record; -static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] - __cpuinitdata; - -static inline int generate_logical_apicid(int quad, int phys_apicid) -{ - return (quad << 4) + (phys_apicid ? phys_apicid << 1 : 1); -} - -/* x86_quirks member */ -static int mpc_apic_id(struct mpc_cpu *m) -{ - int quad = translation_table[mpc_record]->trans_quad; - int logical_apicid = generate_logical_apicid(quad, m->apicid); - - printk(KERN_DEBUG "Processor #%d %u:%u APIC version %d (quad %d, apic %d)\n", - m->apicid, (m->cpufeature & CPU_FAMILY_MASK) >> 8, - (m->cpufeature & CPU_MODEL_MASK) >> 4, - m->apicver, quad, logical_apicid); - return logical_apicid; -} - -int mp_bus_id_to_node[MAX_MP_BUSSES]; - -int mp_bus_id_to_local[MAX_MP_BUSSES]; - -/* x86_quirks member */ -static void mpc_oem_bus_info(struct mpc_bus *m, char *name) -{ - int quad = translation_table[mpc_record]->trans_quad; - int local = translation_table[mpc_record]->trans_local; - - mp_bus_id_to_node[m->busid] = quad; - mp_bus_id_to_local[m->busid] = local; - printk(KERN_INFO "Bus #%d is %s (node %d)\n", - m->busid, name, quad); -} - -int quad_local_to_mp_bus_id [NR_CPUS/4][4]; - -/* x86_quirks member */ -static void mpc_oem_pci_bus(struct mpc_bus *m) -{ - int quad = translation_table[mpc_record]->trans_quad; - int local = translation_table[mpc_record]->trans_local; - - quad_local_to_mp_bus_id[quad][local] = m->busid; -} - -static void __init MP_translation_info(struct mpc_config_translation *m) -{ - printk(KERN_INFO - "Translation: record %d, type %d, quad %d, global %d, local %d\n", - mpc_record, m->trans_type, m->trans_quad, m->trans_global, - m->trans_local); - - if (mpc_record >= MAX_MPC_ENTRY) - printk(KERN_ERR "MAX_MPC_ENTRY exceeded!\n"); - else - translation_table[mpc_record] = m; /* stash this for later */ - if (m->trans_quad < MAX_NUMNODES && !node_online(m->trans_quad)) - node_set_online(m->trans_quad); -} - -static int __init mpf_checksum(unsigned char *mp, int len) -{ - int sum = 0; - - while (len--) - sum += *mp++; - - return sum & 0xFF; -} - -/* - * Read/parse the MPC oem tables - */ - -static void __init smp_read_mpc_oem(struct mpc_oemtable *oemtable, - unsigned short oemsize) -{ - int count = sizeof(*oemtable); /* the header size */ - unsigned char *oemptr = ((unsigned char *)oemtable) + count; - - mpc_record = 0; - printk(KERN_INFO "Found an OEM MPC table at %8p - parsing it ... \n", - oemtable); - if (memcmp(oemtable->signature, MPC_OEM_SIGNATURE, 4)) { - printk(KERN_WARNING - "SMP mpc oemtable: bad signature [%c%c%c%c]!\n", - oemtable->signature[0], oemtable->signature[1], - oemtable->signature[2], oemtable->signature[3]); - return; - } - if (mpf_checksum((unsigned char *)oemtable, oemtable->length)) { - printk(KERN_WARNING "SMP oem mptable: checksum error!\n"); - return; - } - while (count < oemtable->length) { - switch (*oemptr) { - case MP_TRANSLATION: - { - struct mpc_config_translation *m = - (struct mpc_config_translation *)oemptr; - MP_translation_info(m); - oemptr += sizeof(*m); - count += sizeof(*m); - ++mpc_record; - break; - } - default: - { - printk(KERN_WARNING - "Unrecognised OEM table entry type! - %d\n", - (int)*oemptr); - return; - } - } - } -} - -static int __init numaq_setup_ioapic_ids(void) -{ - /* so can skip it */ - return 1; -} - -static int __init numaq_update_genapic(void) -{ - genapic->wakeup_cpu = wakeup_secondary_cpu_via_nmi; - - return 0; -} - -static struct x86_quirks numaq_x86_quirks __initdata = { - .arch_pre_time_init = numaq_pre_time_init, - .arch_time_init = NULL, - .arch_pre_intr_init = NULL, - .arch_memory_setup = NULL, - .arch_intr_init = NULL, - .arch_trap_init = NULL, - .mach_get_smp_config = NULL, - .mach_find_smp_config = NULL, - .mpc_record = &mpc_record, - .mpc_apic_id = mpc_apic_id, - .mpc_oem_bus_info = mpc_oem_bus_info, - .mpc_oem_pci_bus = mpc_oem_pci_bus, - .smp_read_mpc_oem = smp_read_mpc_oem, - .setup_ioapic_ids = numaq_setup_ioapic_ids, - .update_genapic = numaq_update_genapic, -}; - -void numaq_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid) -{ - if (strncmp(oem, "IBM NUMA", 8)) - printk("Warning! Not a NUMA-Q system!\n"); - else - found_numaq = 1; -} - -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) - early_get_smp_config(); - - if (found_numaq) - x86_quirks = &numaq_x86_quirks; -} - -int __init get_memcfg_numaq(void) -{ - early_check_numaq(); - if (!found_numaq) - return 0; - smp_dump_qct(); - return 1; -} diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c index 95777b0faa7..3a7c5a44082 100644 --- a/arch/x86/kernel/paravirt-spinlocks.c +++ b/arch/x86/kernel/paravirt-spinlocks.c @@ -26,13 +26,3 @@ struct pv_lock_ops pv_lock_ops = { }; EXPORT_SYMBOL(pv_lock_ops); -void __init paravirt_use_bytelocks(void) -{ -#ifdef CONFIG_SMP - pv_lock_ops.spin_is_locked = __byte_spin_is_locked; - pv_lock_ops.spin_is_contended = __byte_spin_is_contended; - pv_lock_ops.spin_lock = __byte_spin_lock; - pv_lock_ops.spin_trylock = __byte_spin_trylock; - pv_lock_ops.spin_unlock = __byte_spin_unlock; -#endif -} diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index c6520a4e85d..63dd358d8ee 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -28,7 +28,6 @@ #include <asm/paravirt.h> #include <asm/desc.h> #include <asm/setup.h> -#include <asm/arch_hooks.h> #include <asm/pgtable.h> #include <asm/time.h> #include <asm/pgalloc.h> @@ -44,6 +43,17 @@ void _paravirt_nop(void) { } +/* identity function, which can be inlined */ +u32 _paravirt_ident_32(u32 x) +{ + return x; +} + +u64 _paravirt_ident_64(u64 x) +{ + return x; +} + static void __init default_banner(void) { printk(KERN_INFO "Booting paravirtualized kernel on %s\n", @@ -138,9 +148,16 @@ unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf, if (opfunc == NULL) /* If there's no function, patch it with a ud2a (BUG) */ ret = paravirt_patch_insns(insnbuf, len, ud2a, ud2a+sizeof(ud2a)); - else if (opfunc == paravirt_nop) + else if (opfunc == _paravirt_nop) /* If the operation is a nop, then nop the callsite */ ret = paravirt_patch_nop(); + + /* identity functions just return their single argument */ + else if (opfunc == _paravirt_ident_32) + ret = paravirt_patch_ident_32(insnbuf, len); + else if (opfunc == _paravirt_ident_64) + ret = paravirt_patch_ident_64(insnbuf, len); + else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) || type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit) || type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret32) || @@ -318,10 +335,10 @@ struct pv_time_ops pv_time_ops = { struct pv_irq_ops pv_irq_ops = { .init_IRQ = native_init_IRQ, - .save_fl = native_save_fl, - .restore_fl = native_restore_fl, - .irq_disable = native_irq_disable, - .irq_enable = native_irq_enable, + .save_fl = __PV_IS_CALLEE_SAVE(native_save_fl), + .restore_fl = __PV_IS_CALLEE_SAVE(native_restore_fl), + .irq_disable = __PV_IS_CALLEE_SAVE(native_irq_disable), + .irq_enable = __PV_IS_CALLEE_SAVE(native_irq_enable), .safe_halt = native_safe_halt, .halt = native_halt, #ifdef CONFIG_X86_64 @@ -399,6 +416,14 @@ struct pv_apic_ops pv_apic_ops = { #endif }; +#if defined(CONFIG_X86_32) && !defined(CONFIG_X86_PAE) +/* 32-bit pagetable entries */ +#define PTE_IDENT __PV_IS_CALLEE_SAVE(_paravirt_ident_32) +#else +/* 64-bit pagetable entries */ +#define PTE_IDENT __PV_IS_CALLEE_SAVE(_paravirt_ident_64) +#endif + struct pv_mmu_ops pv_mmu_ops = { #ifndef CONFIG_X86_64 .pagetable_setup_start = native_pagetable_setup_start, @@ -450,22 +475,23 @@ struct pv_mmu_ops pv_mmu_ops = { .pmd_clear = native_pmd_clear, #endif .set_pud = native_set_pud, - .pmd_val = native_pmd_val, - .make_pmd = native_make_pmd, + + .pmd_val = PTE_IDENT, + .make_pmd = PTE_IDENT, #if PAGETABLE_LEVELS == 4 - .pud_val = native_pud_val, - .make_pud = native_make_pud, + .pud_val = PTE_IDENT, + .make_pud = PTE_IDENT, + .set_pgd = native_set_pgd, #endif #endif /* PAGETABLE_LEVELS >= 3 */ - .pte_val = native_pte_val, - .pte_flags = native_pte_flags, - .pgd_val = native_pgd_val, + .pte_val = PTE_IDENT, + .pgd_val = PTE_IDENT, - .make_pte = native_make_pte, - .make_pgd = native_make_pgd, + .make_pte = PTE_IDENT, + .make_pgd = PTE_IDENT, .dup_mmap = paravirt_nop, .exit_mmap = paravirt_nop, diff --git a/arch/x86/kernel/paravirt_patch_32.c b/arch/x86/kernel/paravirt_patch_32.c index 9fe644f4861..d9f32e6d6ab 100644 --- a/arch/x86/kernel/paravirt_patch_32.c +++ b/arch/x86/kernel/paravirt_patch_32.c @@ -12,6 +12,18 @@ DEF_NATIVE(pv_mmu_ops, read_cr3, "mov %cr3, %eax"); DEF_NATIVE(pv_cpu_ops, clts, "clts"); DEF_NATIVE(pv_cpu_ops, read_tsc, "rdtsc"); +unsigned paravirt_patch_ident_32(void *insnbuf, unsigned len) +{ + /* arg in %eax, return in %eax */ + return 0; +} + +unsigned paravirt_patch_ident_64(void *insnbuf, unsigned len) +{ + /* arg in %edx:%eax, return in %edx:%eax */ + return 0; +} + unsigned native_patch(u8 type, u16 clobbers, void *ibuf, unsigned long addr, unsigned len) { diff --git a/arch/x86/kernel/paravirt_patch_64.c b/arch/x86/kernel/paravirt_patch_64.c index 061d01df9ae..3f08f34f93e 100644 --- a/arch/x86/kernel/paravirt_patch_64.c +++ b/arch/x86/kernel/paravirt_patch_64.c @@ -19,6 +19,21 @@ DEF_NATIVE(pv_cpu_ops, usergs_sysret64, "swapgs; sysretq"); DEF_NATIVE(pv_cpu_ops, usergs_sysret32, "swapgs; sysretl"); DEF_NATIVE(pv_cpu_ops, swapgs, "swapgs"); +DEF_NATIVE(, mov32, "mov %edi, %eax"); +DEF_NATIVE(, mov64, "mov %rdi, %rax"); + +unsigned paravirt_patch_ident_32(void *insnbuf, unsigned len) +{ + return paravirt_patch_insns(insnbuf, len, + start__mov32, end__mov32); +} + +unsigned paravirt_patch_ident_64(void *insnbuf, unsigned len) +{ + return paravirt_patch_insns(insnbuf, len, + start__mov64, end__mov64); +} + unsigned native_patch(u8 type, u16 clobbers, void *ibuf, unsigned long addr, unsigned len) { diff --git a/arch/x86/kernel/probe_roms_32.c b/arch/x86/kernel/probe_roms_32.c index 675a48c404a..071e7fea42e 100644 --- a/arch/x86/kernel/probe_roms_32.c +++ b/arch/x86/kernel/probe_roms_32.c @@ -18,7 +18,7 @@ #include <asm/setup.h> #include <asm/sections.h> #include <asm/io.h> -#include <setup_arch.h> +#include <asm/setup_arch.h> static struct resource system_rom_resource = { .name = "System ROM", diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 6d12f7e37f8..156f87582c6 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -1,8 +1,8 @@ #include <linux/errno.h> #include <linux/kernel.h> #include <linux/mm.h> -#include <asm/idle.h> #include <linux/smp.h> +#include <linux/prctl.h> #include <linux/slab.h> #include <linux/sched.h> #include <linux/module.h> @@ -11,6 +11,9 @@ #include <linux/ftrace.h> #include <asm/system.h> #include <asm/apic.h> +#include <asm/idle.h> +#include <asm/uaccess.h> +#include <asm/i387.h> unsigned long idle_halt; EXPORT_SYMBOL(idle_halt); @@ -56,6 +59,193 @@ void arch_task_cache_init(void) } /* + * Free current thread data structures etc.. + */ +void exit_thread(void) +{ + struct task_struct *me = current; + struct thread_struct *t = &me->thread; + unsigned long *bp = t->io_bitmap_ptr; + + if (bp) { + struct tss_struct *tss = &per_cpu(init_tss, get_cpu()); + + t->io_bitmap_ptr = NULL; + clear_thread_flag(TIF_IO_BITMAP); + /* + * Careful, clear this in the TSS too: + */ + memset(tss->io_bitmap, 0xff, t->io_bitmap_max); + t->io_bitmap_max = 0; + put_cpu(); + kfree(bp); + } + + ds_exit_thread(current); +} + +void flush_thread(void) +{ + struct task_struct *tsk = current; + +#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 + + clear_tsk_thread_flag(tsk, TIF_DEBUG); + + tsk->thread.debugreg0 = 0; + tsk->thread.debugreg1 = 0; + tsk->thread.debugreg2 = 0; + tsk->thread.debugreg3 = 0; + tsk->thread.debugreg6 = 0; + tsk->thread.debugreg7 = 0; + memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array)); + /* + * Forget coprocessor state.. + */ + tsk->fpu_counter = 0; + clear_fpu(tsk); + clear_used_math(); +} + +static void hard_disable_TSC(void) +{ + write_cr4(read_cr4() | X86_CR4_TSD); +} + +void disable_TSC(void) +{ + preempt_disable(); + if (!test_and_set_thread_flag(TIF_NOTSC)) + /* + * Must flip the CPU state synchronously with + * TIF_NOTSC in the current running context. + */ + hard_disable_TSC(); + preempt_enable(); +} + +static void hard_enable_TSC(void) +{ + write_cr4(read_cr4() & ~X86_CR4_TSD); +} + +static void enable_TSC(void) +{ + preempt_disable(); + if (test_and_clear_thread_flag(TIF_NOTSC)) + /* + * Must flip the CPU state synchronously with + * TIF_NOTSC in the current running context. + */ + hard_enable_TSC(); + preempt_enable(); +} + +int get_tsc_mode(unsigned long adr) +{ + unsigned int val; + + if (test_thread_flag(TIF_NOTSC)) + val = PR_TSC_SIGSEGV; + else + val = PR_TSC_ENABLE; + + return put_user(val, (unsigned int __user *)adr); +} + +int set_tsc_mode(unsigned int val) +{ + if (val == PR_TSC_SIGSEGV) + disable_TSC(); + else if (val == PR_TSC_ENABLE) + enable_TSC(); + else + return -EINVAL; + + return 0; +} + +void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, + struct tss_struct *tss) +{ + struct thread_struct *prev, *next; + + prev = &prev_p->thread; + next = &next_p->thread; + + if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) || + test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR)) + ds_switch_to(prev_p, next_p); + else if (next->debugctlmsr != prev->debugctlmsr) + update_debugctlmsr(next->debugctlmsr); + + if (test_tsk_thread_flag(next_p, TIF_DEBUG)) { + 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 */ + if (test_tsk_thread_flag(next_p, TIF_NOTSC)) + hard_disable_TSC(); + else + hard_enable_TSC(); + } + + if (test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) { + /* + * Copy the relevant range of the IO bitmap. + * Normally this is 128 bytes or less: + */ + memcpy(tss->io_bitmap, next->io_bitmap_ptr, + max(prev->io_bitmap_max, next->io_bitmap_max)); + } else if (test_tsk_thread_flag(prev_p, TIF_IO_BITMAP)) { + /* + * Clear any possible leftover bits: + */ + memset(tss->io_bitmap, 0xff, prev->io_bitmap_max); + } +} + +int sys_fork(struct pt_regs *regs) +{ + return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL); +} + +/* + * This is trivial, and on the face of it looks like it + * could equally well be done in user mode. + * + * Not so, for quite unobvious reasons - register pressure. + * In user mode vfork() cannot have a stack frame, and if + * done by calling the "clone()" system call directly, you + * do not have enough call-clobbered registers to hold all + * the information you need. + */ +int sys_vfork(struct pt_regs *regs) +{ + return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->sp, regs, 0, + NULL, NULL); +} + + +/* * Idle related variables and functions */ unsigned long boot_option_idle_override = 0; @@ -350,7 +540,7 @@ static void c1e_idle(void) void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) { -#ifdef CONFIG_X86_SMP +#ifdef CONFIG_SMP if (pm_idle == poll_idle && smp_num_siblings > 1) { printk(KERN_WARNING "WARNING: polling idle and HT enabled," " performance may degrade.\n"); diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index bd4da2af08a..14014d766ca 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -11,6 +11,7 @@ #include <stdarg.h> +#include <linux/stackprotector.h> #include <linux/cpu.h> #include <linux/errno.h> #include <linux/sched.h> @@ -66,9 +67,6 @@ asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task; EXPORT_PER_CPU_SYMBOL(current_task); -DEFINE_PER_CPU(int, cpu_number); -EXPORT_PER_CPU_SYMBOL(cpu_number); - /* * Return saved PC of a blocked thread. */ @@ -94,6 +92,15 @@ void cpu_idle(void) { int cpu = smp_processor_id(); + /* + * If we're the non-boot CPU, nothing set the stack canary up + * for us. CPU0 already has it initialized but no harm in + * doing it again. This is a good place for updating it, as + * we wont ever return from this function (so the invalid + * canaries already on the stack wont ever trigger). + */ + boot_init_stack_canary(); + current_thread_info()->status |= TS_POLLING; /* endless idle loop with no priority at all */ @@ -108,7 +115,6 @@ void cpu_idle(void) play_dead(); local_irq_disable(); - __get_cpu_var(irq_stat).idle_timestamp = jiffies; /* Don't trace irqs off for idle */ stop_critical_timings(); pm_idle(); @@ -132,7 +138,7 @@ void __show_regs(struct pt_regs *regs, int all) if (user_mode_vm(regs)) { sp = regs->sp; ss = regs->ss & 0xffff; - savesegment(gs, gs); + gs = get_user_gs(regs); } else { sp = (unsigned long) (®s->sp); savesegment(ss, ss); @@ -213,6 +219,7 @@ int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags) 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(); @@ -223,55 +230,6 @@ int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags) } EXPORT_SYMBOL(kernel_thread); -/* - * Free current thread data structures etc.. - */ -void exit_thread(void) -{ - /* The process may have allocated an io port bitmap... nuke it. */ - if (unlikely(test_thread_flag(TIF_IO_BITMAP))) { - struct task_struct *tsk = current; - struct thread_struct *t = &tsk->thread; - int cpu = get_cpu(); - struct tss_struct *tss = &per_cpu(init_tss, cpu); - - kfree(t->io_bitmap_ptr); - t->io_bitmap_ptr = NULL; - clear_thread_flag(TIF_IO_BITMAP); - /* - * Careful, clear this in the TSS too: - */ - memset(tss->io_bitmap, 0xff, tss->io_bitmap_max); - t->io_bitmap_max = 0; - tss->io_bitmap_owner = NULL; - tss->io_bitmap_max = 0; - tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET; - put_cpu(); - } - - ds_exit_thread(current); -} - -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; - memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array)); - clear_tsk_thread_flag(tsk, TIF_DEBUG); - /* - * Forget coprocessor state.. - */ - tsk->fpu_counter = 0; - clear_fpu(tsk); - clear_used_math(); -} - void release_thread(struct task_struct *dead_task) { BUG_ON(dead_task->mm); @@ -305,7 +263,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, p->thread.ip = (unsigned long) ret_from_fork; - savesegment(gs, p->thread.gs); + task_user_gs(p) = get_user_gs(regs); tsk = current; if (unlikely(test_tsk_thread_flag(tsk, TIF_IO_BITMAP))) { @@ -343,7 +301,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, void start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) { - __asm__("movl %0, %%gs" : : "r"(0)); + set_user_gs(regs, 0); regs->fs = 0; set_fs(USER_DS); regs->ds = __USER_DS; @@ -359,127 +317,6 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) } EXPORT_SYMBOL_GPL(start_thread); -static void hard_disable_TSC(void) -{ - write_cr4(read_cr4() | X86_CR4_TSD); -} - -void disable_TSC(void) -{ - preempt_disable(); - if (!test_and_set_thread_flag(TIF_NOTSC)) - /* - * Must flip the CPU state synchronously with - * TIF_NOTSC in the current running context. - */ - hard_disable_TSC(); - preempt_enable(); -} - -static void hard_enable_TSC(void) -{ - write_cr4(read_cr4() & ~X86_CR4_TSD); -} - -static void enable_TSC(void) -{ - preempt_disable(); - if (test_and_clear_thread_flag(TIF_NOTSC)) - /* - * Must flip the CPU state synchronously with - * TIF_NOTSC in the current running context. - */ - hard_enable_TSC(); - preempt_enable(); -} - -int get_tsc_mode(unsigned long adr) -{ - unsigned int val; - - if (test_thread_flag(TIF_NOTSC)) - val = PR_TSC_SIGSEGV; - else - val = PR_TSC_ENABLE; - - return put_user(val, (unsigned int __user *)adr); -} - -int set_tsc_mode(unsigned int val) -{ - if (val == PR_TSC_SIGSEGV) - disable_TSC(); - else if (val == PR_TSC_ENABLE) - enable_TSC(); - else - return -EINVAL; - - return 0; -} - -static noinline void -__switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, - struct tss_struct *tss) -{ - struct thread_struct *prev, *next; - - prev = &prev_p->thread; - next = &next_p->thread; - - if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) || - test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR)) - ds_switch_to(prev_p, next_p); - else if (next->debugctlmsr != prev->debugctlmsr) - update_debugctlmsr(next->debugctlmsr); - - if (test_tsk_thread_flag(next_p, TIF_DEBUG)) { - 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 */ - if (test_tsk_thread_flag(next_p, TIF_NOTSC)) - hard_disable_TSC(); - else - hard_enable_TSC(); - } - - if (!test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) { - /* - * Disable the bitmap via an invalid offset. We still cache - * the previous bitmap owner and the IO bitmap contents: - */ - tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET; - return; - } - - if (likely(next == tss->io_bitmap_owner)) { - /* - * Previous owner of the bitmap (hence the bitmap content) - * matches the next task, we dont have to do anything but - * to set a valid offset in the TSS: - */ - tss->x86_tss.io_bitmap_base = IO_BITMAP_OFFSET; - return; - } - /* - * Lazy TSS's I/O bitmap copy. We set an invalid offset here - * and we let the task to get a GPF in case an I/O instruction - * is performed. The handler of the GPF will verify that the - * faulting task has a valid I/O bitmap and, it true, does the - * real copy and restart the instruction. This will save us - * redundant copies when the currently switched task does not - * perform any I/O during its timeslice. - */ - tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET_LAZY; -} /* * switch_to(x,yn) should switch tasks from x to y. @@ -540,7 +377,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) * used %fs or %gs (it does not today), or if the kernel is * running inside of a hypervisor layer. */ - savesegment(gs, prev->gs); + lazy_save_gs(prev->gs); /* * Load the per-thread Thread-Local Storage descriptor. @@ -586,64 +423,44 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) * Restore %gs if needed (which is common) */ if (prev->gs | next->gs) - loadsegment(gs, next->gs); + lazy_load_gs(next->gs); - x86_write_percpu(current_task, next_p); + percpu_write(current_task, next_p); return prev_p; } -asmlinkage int sys_fork(struct pt_regs regs) -{ - return do_fork(SIGCHLD, regs.sp, ®s, 0, NULL, NULL); -} - -asmlinkage int sys_clone(struct pt_regs regs) +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; + 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, ®s, 0, parent_tidptr, child_tidptr); -} - -/* - * This is trivial, and on the face of it looks like it - * could equally well be done in user mode. - * - * Not so, for quite unobvious reasons - register pressure. - * In user mode vfork() cannot have a stack frame, and if - * done by calling the "clone()" system call directly, you - * do not have enough call-clobbered registers to hold all - * the information you need. - */ -asmlinkage int sys_vfork(struct pt_regs regs) -{ - return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs.sp, ®s, 0, NULL, NULL); + newsp = regs->sp; + return do_fork(clone_flags, newsp, regs, 0, parent_tidptr, child_tidptr); } /* * sys_execve() executes a new program. */ -asmlinkage int sys_execve(struct pt_regs regs) +int sys_execve(struct pt_regs *regs) { int error; char *filename; - filename = getname((char __user *) regs.bx); + 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, - ®s); + (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); diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 85b4cb5c198..abb7e6a7f0c 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -16,6 +16,7 @@ #include <stdarg.h> +#include <linux/stackprotector.h> #include <linux/cpu.h> #include <linux/errno.h> #include <linux/sched.h> @@ -47,7 +48,6 @@ #include <asm/processor.h> #include <asm/i387.h> #include <asm/mmu_context.h> -#include <asm/pda.h> #include <asm/prctl.h> #include <asm/desc.h> #include <asm/proto.h> @@ -58,6 +58,12 @@ asmlinkage extern void ret_from_fork(void); +DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task; +EXPORT_PER_CPU_SYMBOL(current_task); + +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); @@ -76,13 +82,13 @@ EXPORT_SYMBOL_GPL(idle_notifier_unregister); void enter_idle(void) { - write_pda(isidle, 1); + percpu_write(is_idle, 1); atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL); } static void __exit_idle(void) { - if (test_and_clear_bit_pda(0, isidle) == 0) + if (x86_test_and_clear_bit_percpu(0, is_idle) == 0) return; atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL); } @@ -112,6 +118,16 @@ static inline void play_dead(void) void cpu_idle(void) { current_thread_info()->status |= TS_POLLING; + + /* + * If we're the non-boot CPU, nothing set the stack canary up + * for us. CPU0 already has it initialized but no harm in + * doing it again. This is a good place for updating it, as + * we wont ever return from this function (so the invalid + * canaries already on the stack wont ever trigger). + */ + boot_init_stack_canary(); + /* endless idle loop with no priority at all */ while (1) { tick_nohz_stop_sched_tick(1); @@ -221,61 +237,6 @@ void show_regs(struct pt_regs *regs) show_trace(NULL, regs, (void *)(regs + 1), regs->bp); } -/* - * Free current thread data structures etc.. - */ -void exit_thread(void) -{ - struct task_struct *me = current; - struct thread_struct *t = &me->thread; - - if (me->thread.io_bitmap_ptr) { - struct tss_struct *tss = &per_cpu(init_tss, get_cpu()); - - kfree(t->io_bitmap_ptr); - t->io_bitmap_ptr = NULL; - clear_thread_flag(TIF_IO_BITMAP); - /* - * Careful, clear this in the TSS too: - */ - memset(tss->io_bitmap, 0xff, t->io_bitmap_max); - t->io_bitmap_max = 0; - put_cpu(); - } - - ds_exit_thread(current); -} - -void flush_thread(void) -{ - struct task_struct *tsk = current; - - 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; - } - } - clear_tsk_thread_flag(tsk, TIF_DEBUG); - - tsk->thread.debugreg0 = 0; - tsk->thread.debugreg1 = 0; - tsk->thread.debugreg2 = 0; - tsk->thread.debugreg3 = 0; - tsk->thread.debugreg6 = 0; - tsk->thread.debugreg7 = 0; - memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array)); - /* - * Forget coprocessor state.. - */ - tsk->fpu_counter = 0; - clear_fpu(tsk); - clear_used_math(); -} - void release_thread(struct task_struct *dead_task) { if (dead_task->mm) { @@ -397,7 +358,7 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) load_gs_index(0); regs->ip = new_ip; regs->sp = new_sp; - write_pda(oldrsp, new_sp); + percpu_write(old_rsp, new_sp); regs->cs = __USER_CS; regs->ss = __USER_DS; regs->flags = 0x200; @@ -409,118 +370,6 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) } EXPORT_SYMBOL_GPL(start_thread); -static void hard_disable_TSC(void) -{ - write_cr4(read_cr4() | X86_CR4_TSD); -} - -void disable_TSC(void) -{ - preempt_disable(); - if (!test_and_set_thread_flag(TIF_NOTSC)) - /* - * Must flip the CPU state synchronously with - * TIF_NOTSC in the current running context. - */ - hard_disable_TSC(); - preempt_enable(); -} - -static void hard_enable_TSC(void) -{ - write_cr4(read_cr4() & ~X86_CR4_TSD); -} - -static void enable_TSC(void) -{ - preempt_disable(); - if (test_and_clear_thread_flag(TIF_NOTSC)) - /* - * Must flip the CPU state synchronously with - * TIF_NOTSC in the current running context. - */ - hard_enable_TSC(); - preempt_enable(); -} - -int get_tsc_mode(unsigned long adr) -{ - unsigned int val; - - if (test_thread_flag(TIF_NOTSC)) - val = PR_TSC_SIGSEGV; - else - val = PR_TSC_ENABLE; - - return put_user(val, (unsigned int __user *)adr); -} - -int set_tsc_mode(unsigned int val) -{ - if (val == PR_TSC_SIGSEGV) - disable_TSC(); - else if (val == PR_TSC_ENABLE) - enable_TSC(); - else - return -EINVAL; - - return 0; -} - -/* - * This special macro can be used to load a debugging register - */ -#define loaddebug(thread, r) set_debugreg(thread->debugreg ## r, r) - -static inline void __switch_to_xtra(struct task_struct *prev_p, - struct task_struct *next_p, - struct tss_struct *tss) -{ - struct thread_struct *prev, *next; - - prev = &prev_p->thread, - next = &next_p->thread; - - if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) || - test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR)) - ds_switch_to(prev_p, next_p); - else if (next->debugctlmsr != prev->debugctlmsr) - update_debugctlmsr(next->debugctlmsr); - - if (test_tsk_thread_flag(next_p, TIF_DEBUG)) { - loaddebug(next, 0); - loaddebug(next, 1); - loaddebug(next, 2); - loaddebug(next, 3); - /* no 4 and 5 */ - loaddebug(next, 6); - loaddebug(next, 7); - } - - if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^ - test_tsk_thread_flag(next_p, TIF_NOTSC)) { - /* prev and next are different */ - if (test_tsk_thread_flag(next_p, TIF_NOTSC)) - hard_disable_TSC(); - else - hard_enable_TSC(); - } - - if (test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) { - /* - * Copy the relevant range of the IO bitmap. - * Normally this is 128 bytes or less: - */ - memcpy(tss->io_bitmap, next->io_bitmap_ptr, - max(prev->io_bitmap_max, next->io_bitmap_max)); - } else if (test_tsk_thread_flag(prev_p, TIF_IO_BITMAP)) { - /* - * Clear any possible leftover bits: - */ - memset(tss->io_bitmap, 0xff, prev->io_bitmap_max); - } -} - /* * switch_to(x,y) should switch tasks from x to y. * @@ -618,21 +467,13 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) /* * Switch the PDA and FPU contexts. */ - prev->usersp = read_pda(oldrsp); - write_pda(oldrsp, next->usersp); - write_pda(pcurrent, next_p); + prev->usersp = percpu_read(old_rsp); + percpu_write(old_rsp, next->usersp); + percpu_write(current_task, next_p); - write_pda(kernelstack, + percpu_write(kernel_stack, (unsigned long)task_stack_page(next_p) + - THREAD_SIZE - PDA_STACKOFFSET); -#ifdef CONFIG_CC_STACKPROTECTOR - write_pda(stack_canary, next_p->stack_canary); - /* - * Build time only check to make sure the stack_canary is at - * offset 40 in the pda; this is a gcc ABI requirement - */ - BUILD_BUG_ON(offsetof(struct x8664_pda, stack_canary) != 40); -#endif + THREAD_SIZE - KERNEL_STACK_OFFSET); /* * Now maybe reload the debug registers and handle I/O bitmaps @@ -686,11 +527,6 @@ void set_personality_64bit(void) current->personality &= ~READ_IMPLIES_EXEC; } -asmlinkage long sys_fork(struct pt_regs *regs) -{ - return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL); -} - asmlinkage long sys_clone(unsigned long clone_flags, unsigned long newsp, void __user *parent_tid, void __user *child_tid, struct pt_regs *regs) @@ -700,22 +536,6 @@ sys_clone(unsigned long clone_flags, unsigned long newsp, return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid); } -/* - * This is trivial, and on the face of it looks like it - * could equally well be done in user mode. - * - * Not so, for quite unobvious reasons - register pressure. - * In user mode vfork() cannot have a stack frame, and if - * done by calling the "clone()" system call directly, you - * do not have enough call-clobbered registers to hold all - * the information you need. - */ -asmlinkage long sys_vfork(struct pt_regs *regs) -{ - return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->sp, regs, 0, - NULL, NULL); -} - unsigned long get_wchan(struct task_struct *p) { unsigned long stack; diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 06ca07f6ad8..19378715f41 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -75,10 +75,7 @@ static inline bool invalid_selector(u16 value) static unsigned long *pt_regs_access(struct pt_regs *regs, unsigned long regno) { BUILD_BUG_ON(offsetof(struct pt_regs, bx) != 0); - regno >>= 2; - if (regno > FS) - --regno; - return ®s->bx + regno; + return ®s->bx + (regno >> 2); } static u16 get_segment_reg(struct task_struct *task, unsigned long offset) @@ -90,9 +87,10 @@ static u16 get_segment_reg(struct task_struct *task, unsigned long offset) if (offset != offsetof(struct user_regs_struct, gs)) retval = *pt_regs_access(task_pt_regs(task), offset); else { - retval = task->thread.gs; if (task == current) - savesegment(gs, retval); + retval = get_user_gs(task_pt_regs(task)); + else + retval = task_user_gs(task); } return retval; } @@ -126,13 +124,10 @@ static int set_segment_reg(struct task_struct *task, break; case offsetof(struct user_regs_struct, gs): - task->thread.gs = value; if (task == current) - /* - * The user-mode %gs is not affected by - * kernel entry, so we must update the CPU. - */ - loadsegment(gs, value); + set_user_gs(task_pt_regs(task), value); + else + task_user_gs(task) = value; } return 0; @@ -273,7 +268,7 @@ static unsigned long debugreg_addr_limit(struct task_struct *task) if (test_tsk_thread_flag(task, TIF_IA32)) return IA32_PAGE_OFFSET - 3; #endif - return TASK_SIZE64 - 7; + return TASK_SIZE_MAX - 7; } #endif /* CONFIG_X86_32 */ @@ -690,9 +685,8 @@ static int ptrace_bts_config(struct task_struct *child, if (!cfg.signal) return -EINVAL; - return -EOPNOTSUPP; - child->thread.bts_ovfl_signal = cfg.signal; + return -EOPNOTSUPP; } if ((cfg.flags & PTRACE_BTS_O_ALLOC) && diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c index 309949e9e1c..6a5a2970f4c 100644 --- a/arch/x86/kernel/quirks.c +++ b/arch/x86/kernel/quirks.c @@ -74,8 +74,7 @@ static void ich_force_hpet_resume(void) if (!force_hpet_address) return; - if (rcba_base == NULL) - BUG(); + BUG_ON(rcba_base == NULL); /* read the Function Disable register, dword mode only */ val = readl(rcba_base + 0x3404); diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 4526b3a75ed..2aef36d8aca 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -14,6 +14,7 @@ #include <asm/reboot.h> #include <asm/pci_x86.h> #include <asm/virtext.h> +#include <asm/cpu.h> #ifdef CONFIG_X86_32 # include <linux/dmi.h> @@ -23,8 +24,6 @@ # include <asm/iommu.h> #endif -#include <mach_ipi.h> - /* * Power off function, if any */ @@ -658,7 +657,7 @@ static int crash_nmi_callback(struct notifier_block *self, static void smp_send_nmi_allbutself(void) { - send_IPI_allbutself(NMI_VECTOR); + apic->send_IPI_allbutself(NMI_VECTOR); } static struct notifier_block crash_nmi_nb = { diff --git a/arch/x86/kernel/relocate_kernel_32.S b/arch/x86/kernel/relocate_kernel_32.S index a160f311972..41235531b11 100644 --- a/arch/x86/kernel/relocate_kernel_32.S +++ b/arch/x86/kernel/relocate_kernel_32.S @@ -7,7 +7,7 @@ */ #include <linux/linkage.h> -#include <asm/page.h> +#include <asm/page_types.h> #include <asm/kexec.h> #include <asm/processor-flags.h> @@ -17,7 +17,8 @@ #define PTR(x) (x << 2) -/* control_page + KEXEC_CONTROL_CODE_MAX_SIZE +/* + * control_page + KEXEC_CONTROL_CODE_MAX_SIZE * ~ control_page + PAGE_SIZE are used as data storage and stack for * jumping back */ @@ -76,8 +77,10 @@ relocate_kernel: movl %eax, CP_PA_SWAP_PAGE(%edi) movl %ebx, CP_PA_BACKUP_PAGES_MAP(%edi) - /* get physical address of control page now */ - /* this is impossible after page table switch */ + /* + * get physical address of control page now + * this is impossible after page table switch + */ movl PTR(PA_CONTROL_PAGE)(%ebp), %edi /* switch to new set of page tables */ @@ -97,7 +100,8 @@ identity_mapped: /* store the start address on the stack */ pushl %edx - /* Set cr0 to a known state: + /* + * Set cr0 to a known state: * - Paging disabled * - Alignment check disabled * - Write protect disabled @@ -113,7 +117,8 @@ identity_mapped: /* clear cr4 if applicable */ testl %ecx, %ecx jz 1f - /* Set cr4 to a known state: + /* + * Set cr4 to a known state: * Setting everything to zero seems safe. */ xorl %eax, %eax @@ -132,15 +137,18 @@ identity_mapped: call swap_pages addl $8, %esp - /* To be certain of avoiding problems with self-modifying code + /* + * To be certain of avoiding problems with self-modifying code * I need to execute a serializing instruction here. * So I flush the TLB, it's handy, and not processor dependent. */ xorl %eax, %eax movl %eax, %cr3 - /* set all of the registers to known values */ - /* leave %esp alone */ + /* + * set all of the registers to known values + * leave %esp alone + */ testl %esi, %esi jnz 1f diff --git a/arch/x86/kernel/relocate_kernel_64.S b/arch/x86/kernel/relocate_kernel_64.S index f5afe665a82..4de8f5b3d47 100644 --- a/arch/x86/kernel/relocate_kernel_64.S +++ b/arch/x86/kernel/relocate_kernel_64.S @@ -7,10 +7,10 @@ */ #include <linux/linkage.h> -#include <asm/page.h> +#include <asm/page_types.h> #include <asm/kexec.h> #include <asm/processor-flags.h> -#include <asm/pgtable.h> +#include <asm/pgtable_types.h> /* * Must be relocatable PIC code callable as a C function @@ -19,145 +19,76 @@ #define PTR(x) (x << 3) #define PAGE_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY) +/* + * control_page + KEXEC_CONTROL_CODE_MAX_SIZE + * ~ control_page + PAGE_SIZE are used as data storage and stack for + * jumping back + */ +#define DATA(offset) (KEXEC_CONTROL_CODE_MAX_SIZE+(offset)) + +/* Minimal CPU state */ +#define RSP DATA(0x0) +#define CR0 DATA(0x8) +#define CR3 DATA(0x10) +#define CR4 DATA(0x18) + +/* other data */ +#define CP_PA_TABLE_PAGE DATA(0x20) +#define CP_PA_SWAP_PAGE DATA(0x28) +#define CP_PA_BACKUP_PAGES_MAP DATA(0x30) + .text .align PAGE_SIZE .code64 .globl relocate_kernel relocate_kernel: - /* %rdi indirection_page + /* + * %rdi indirection_page * %rsi page_list * %rdx start address + * %rcx preserve_context */ - /* map the control page at its virtual address */ - - movq $0x0000ff8000000000, %r10 /* mask */ - mov $(39 - 3), %cl /* bits to shift */ - movq PTR(VA_CONTROL_PAGE)(%rsi), %r11 /* address to map */ - - movq %r11, %r9 - andq %r10, %r9 - shrq %cl, %r9 - - movq PTR(VA_PGD)(%rsi), %r8 - addq %r8, %r9 - movq PTR(PA_PUD_0)(%rsi), %r8 - orq $PAGE_ATTR, %r8 - movq %r8, (%r9) - - shrq $9, %r10 - sub $9, %cl - - movq %r11, %r9 - andq %r10, %r9 - shrq %cl, %r9 - - movq PTR(VA_PUD_0)(%rsi), %r8 - addq %r8, %r9 - movq PTR(PA_PMD_0)(%rsi), %r8 - orq $PAGE_ATTR, %r8 - movq %r8, (%r9) - - shrq $9, %r10 - sub $9, %cl - - movq %r11, %r9 - andq %r10, %r9 - shrq %cl, %r9 - - movq PTR(VA_PMD_0)(%rsi), %r8 - addq %r8, %r9 - movq PTR(PA_PTE_0)(%rsi), %r8 - orq $PAGE_ATTR, %r8 - movq %r8, (%r9) - - shrq $9, %r10 - sub $9, %cl - - movq %r11, %r9 - andq %r10, %r9 - shrq %cl, %r9 - - movq PTR(VA_PTE_0)(%rsi), %r8 - addq %r8, %r9 - movq PTR(PA_CONTROL_PAGE)(%rsi), %r8 - orq $PAGE_ATTR, %r8 - movq %r8, (%r9) - - /* identity map the control page at its physical address */ - - movq $0x0000ff8000000000, %r10 /* mask */ - mov $(39 - 3), %cl /* bits to shift */ - movq PTR(PA_CONTROL_PAGE)(%rsi), %r11 /* address to map */ - - movq %r11, %r9 - andq %r10, %r9 - shrq %cl, %r9 - - movq PTR(VA_PGD)(%rsi), %r8 - addq %r8, %r9 - movq PTR(PA_PUD_1)(%rsi), %r8 - orq $PAGE_ATTR, %r8 - movq %r8, (%r9) - - shrq $9, %r10 - sub $9, %cl - - movq %r11, %r9 - andq %r10, %r9 - shrq %cl, %r9 - - movq PTR(VA_PUD_1)(%rsi), %r8 - addq %r8, %r9 - movq PTR(PA_PMD_1)(%rsi), %r8 - orq $PAGE_ATTR, %r8 - movq %r8, (%r9) - - shrq $9, %r10 - sub $9, %cl - - movq %r11, %r9 - andq %r10, %r9 - shrq %cl, %r9 - - movq PTR(VA_PMD_1)(%rsi), %r8 - addq %r8, %r9 - movq PTR(PA_PTE_1)(%rsi), %r8 - orq $PAGE_ATTR, %r8 - movq %r8, (%r9) - - shrq $9, %r10 - sub $9, %cl - - movq %r11, %r9 - andq %r10, %r9 - shrq %cl, %r9 - - movq PTR(VA_PTE_1)(%rsi), %r8 - addq %r8, %r9 - movq PTR(PA_CONTROL_PAGE)(%rsi), %r8 - orq $PAGE_ATTR, %r8 - movq %r8, (%r9) - -relocate_new_kernel: - /* %rdi indirection_page - * %rsi page_list - * %rdx start address - */ + /* Save the CPU context, used for jumping back */ + pushq %rbx + pushq %rbp + pushq %r12 + pushq %r13 + pushq %r14 + pushq %r15 + pushf + + movq PTR(VA_CONTROL_PAGE)(%rsi), %r11 + movq %rsp, RSP(%r11) + movq %cr0, %rax + movq %rax, CR0(%r11) + movq %cr3, %rax + movq %rax, CR3(%r11) + movq %cr4, %rax + movq %rax, CR4(%r11) /* zero out flags, and disable interrupts */ pushq $0 popfq - /* get physical address of control page now */ - /* this is impossible after page table switch */ + /* + * get physical address of control page now + * this is impossible after page table switch + */ movq PTR(PA_CONTROL_PAGE)(%rsi), %r8 /* get physical address of page table now too */ - movq PTR(PA_TABLE_PAGE)(%rsi), %rcx + movq PTR(PA_TABLE_PAGE)(%rsi), %r9 + + /* get physical address of swap page now */ + movq PTR(PA_SWAP_PAGE)(%rsi), %r10 - /* switch to new set of page tables */ - movq PTR(PA_PGD)(%rsi), %r9 + /* save some information for jumping back */ + movq %r9, CP_PA_TABLE_PAGE(%r11) + movq %r10, CP_PA_SWAP_PAGE(%r11) + movq %rdi, CP_PA_BACKUP_PAGES_MAP(%r11) + + /* Switch to the identity mapped page tables */ movq %r9, %cr3 /* setup a new stack at the end of the physical control page */ @@ -172,7 +103,8 @@ identity_mapped: /* store the start address on the stack */ pushq %rdx - /* Set cr0 to a known state: + /* + * Set cr0 to a known state: * - Paging enabled * - Alignment check disabled * - Write protect disabled @@ -185,7 +117,8 @@ identity_mapped: orl $(X86_CR0_PG | X86_CR0_PE), %eax movq %rax, %cr0 - /* Set cr4 to a known state: + /* + * Set cr4 to a known state: * - physical address extension enabled */ movq $X86_CR4_PAE, %rax @@ -194,12 +127,88 @@ identity_mapped: jmp 1f 1: - /* Switch to the identity mapped page tables, - * and flush the TLB. - */ - movq %rcx, %cr3 + /* Flush the TLB (needed?) */ + movq %r9, %cr3 + + movq %rcx, %r11 + call swap_pages + + /* + * To be certain of avoiding problems with self-modifying code + * I need to execute a serializing instruction here. + * So I flush the TLB by reloading %cr3 here, it's handy, + * and not processor dependent. + */ + movq %cr3, %rax + movq %rax, %cr3 + + /* + * set all of the registers to known values + * leave %rsp alone + */ + + testq %r11, %r11 + jnz 1f + xorq %rax, %rax + xorq %rbx, %rbx + xorq %rcx, %rcx + xorq %rdx, %rdx + xorq %rsi, %rsi + xorq %rdi, %rdi + xorq %rbp, %rbp + xorq %r8, %r8 + xorq %r9, %r9 + xorq %r10, %r9 + xorq %r11, %r11 + xorq %r12, %r12 + xorq %r13, %r13 + xorq %r14, %r14 + xorq %r15, %r15 + + ret + +1: + popq %rdx + leaq PAGE_SIZE(%r10), %rsp + call *%rdx + + /* get the re-entry point of the peer system */ + movq 0(%rsp), %rbp + call 1f +1: + popq %r8 + subq $(1b - relocate_kernel), %r8 + movq CP_PA_SWAP_PAGE(%r8), %r10 + movq CP_PA_BACKUP_PAGES_MAP(%r8), %rdi + movq CP_PA_TABLE_PAGE(%r8), %rax + movq %rax, %cr3 + lea PAGE_SIZE(%r8), %rsp + call swap_pages + movq $virtual_mapped, %rax + pushq %rax + ret + +virtual_mapped: + movq RSP(%r8), %rsp + movq CR4(%r8), %rax + movq %rax, %cr4 + movq CR3(%r8), %rax + movq CR0(%r8), %r8 + movq %rax, %cr3 + movq %r8, %cr0 + movq %rbp, %rax + + popf + popq %r15 + popq %r14 + popq %r13 + popq %r12 + popq %rbp + popq %rbx + ret /* Do the copies */ +swap_pages: movq %rdi, %rcx /* Put the page_list in %rcx */ xorq %rdi, %rdi xorq %rsi, %rsi @@ -231,36 +240,27 @@ identity_mapped: movq %rcx, %rsi /* For ever source page do a copy */ andq $0xfffffffffffff000, %rsi + movq %rdi, %rdx + movq %rsi, %rax + + movq %r10, %rdi movq $512, %rcx rep ; movsq - jmp 0b -3: - - /* To be certain of avoiding problems with self-modifying code - * I need to execute a serializing instruction here. - * So I flush the TLB by reloading %cr3 here, it's handy, - * and not processor dependent. - */ - movq %cr3, %rax - movq %rax, %cr3 - /* set all of the registers to known values */ - /* leave %rsp alone */ + movq %rax, %rdi + movq %rdx, %rsi + movq $512, %rcx + rep ; movsq - xorq %rax, %rax - xorq %rbx, %rbx - xorq %rcx, %rcx - xorq %rdx, %rdx - xorq %rsi, %rsi - xorq %rdi, %rdi - xorq %rbp, %rbp - xorq %r8, %r8 - xorq %r9, %r9 - xorq %r10, %r9 - xorq %r11, %r11 - xorq %r12, %r12 - xorq %r13, %r13 - xorq %r14, %r14 - xorq %r15, %r15 + movq %rdx, %rdi + movq %r10, %rsi + movq $512, %rcx + rep ; movsq + lea PAGE_SIZE(%rax), %rsi + jmp 0b +3: ret + + .globl kexec_control_code_size +.set kexec_control_code_size, . - relocate_kernel diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 6a8811a6932..a0d26237d7c 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -74,14 +74,15 @@ #include <asm/e820.h> #include <asm/mpspec.h> #include <asm/setup.h> -#include <asm/arch_hooks.h> #include <asm/efi.h> +#include <asm/timer.h> +#include <asm/i8259.h> #include <asm/sections.h> #include <asm/dmi.h> #include <asm/io_apic.h> #include <asm/ist.h> #include <asm/vmi.h> -#include <setup_arch.h> +#include <asm/setup_arch.h> #include <asm/bios_ebda.h> #include <asm/cacheflush.h> #include <asm/processor.h> @@ -89,7 +90,7 @@ #include <asm/system.h> #include <asm/vsyscall.h> -#include <asm/smp.h> +#include <asm/cpu.h> #include <asm/desc.h> #include <asm/dma.h> #include <asm/iommu.h> @@ -97,7 +98,6 @@ #include <asm/mmu_context.h> #include <asm/proto.h> -#include <mach_apic.h> #include <asm/paravirt.h> #include <asm/hypervisor.h> @@ -112,6 +112,25 @@ #define ARCH_SETUP #endif +RESERVE_BRK(dmi_alloc, 65536); + +unsigned int boot_cpu_id __read_mostly; + +static __initdata unsigned long _brk_start = (unsigned long)__brk_base; +unsigned long _brk_end = (unsigned long)__brk_base; + +#ifdef CONFIG_X86_64 +int default_cpu_present_to_apicid(int mps_cpu) +{ + return __default_cpu_present_to_apicid(mps_cpu); +} + +int default_check_phys_apicid_present(int boot_cpu_physical_apicid) +{ + return __default_check_phys_apicid_present(boot_cpu_physical_apicid); +} +#endif + #ifndef CONFIG_DEBUG_BOOT_PARAMS struct boot_params __initdata boot_params; #else @@ -144,12 +163,6 @@ static struct resource bss_resource = { #ifdef CONFIG_X86_32 -/* This value is set up by the early boot code to point to the value - immediately after the boot time page tables. It contains a *physical* - address, and must not be in the .bss segment! */ -unsigned long init_pg_tables_start __initdata = ~0UL; -unsigned long init_pg_tables_end __initdata = ~0UL; - static struct resource video_ram_resource = { .name = "Video RAM area", .start = 0xa0000, @@ -188,7 +201,9 @@ struct ist_info ist_info; #endif #else -struct cpuinfo_x86 boot_cpu_data __read_mostly; +struct cpuinfo_x86 boot_cpu_data __read_mostly = { + .x86_phys_bits = MAX_PHYSMEM_BITS, +}; EXPORT_SYMBOL(boot_cpu_data); #endif @@ -203,12 +218,6 @@ unsigned long mmu_cr4_features = X86_CR4_PAE; int bootloader_type; /* - * Early DMI memory - */ -int dmi_alloc_index; -char dmi_alloc_data[DMI_MAX_DATA]; - -/* * Setup options */ struct screen_info screen_info; @@ -253,6 +262,35 @@ static inline void copy_edd(void) } #endif +void * __init extend_brk(size_t size, size_t align) +{ + size_t mask = align - 1; + void *ret; + + BUG_ON(_brk_start == 0); + BUG_ON(align & mask); + + _brk_end = (_brk_end + mask) & ~mask; + BUG_ON((char *)(_brk_end + size) > __brk_limit); + + ret = (void *)_brk_end; + _brk_end += size; + + memset(ret, 0, size); + + return ret; +} + +static void __init reserve_brk(void) +{ + if (_brk_end > _brk_start) + reserve_early(__pa(_brk_start), __pa(_brk_end), "BRK"); + + /* Mark brk area as locked down and no longer taking any + new allocations */ + _brk_start = 0; +} + #ifdef CONFIG_BLK_DEV_INITRD #ifdef CONFIG_X86_32 @@ -586,20 +624,7 @@ static int __init setup_elfcorehdr(char *arg) early_param("elfcorehdr", setup_elfcorehdr); #endif -static int __init default_update_genapic(void) -{ -#ifdef CONFIG_X86_SMP -# if defined(CONFIG_X86_GENERICARCH) || defined(CONFIG_X86_64) - genapic->wakeup_cpu = wakeup_secondary_cpu_via_init; -# endif -#endif - - return 0; -} - -static struct x86_quirks default_x86_quirks __initdata = { - .update_genapic = default_update_genapic, -}; +static struct x86_quirks default_x86_quirks __initdata; struct x86_quirks *x86_quirks __initdata = &default_x86_quirks; @@ -656,7 +681,6 @@ void __init setup_arch(char **cmdline_p) #ifdef CONFIG_X86_32 memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); visws_early_detect(); - pre_setup_arch_hook(); #else printk(KERN_INFO "Command line: %s\n", boot_command_line); #endif @@ -715,11 +739,7 @@ void __init setup_arch(char **cmdline_p) init_mm.start_code = (unsigned long) _text; init_mm.end_code = (unsigned long) _etext; init_mm.end_data = (unsigned long) _edata; -#ifdef CONFIG_X86_32 - init_mm.brk = init_pg_tables_end + PAGE_OFFSET; -#else - init_mm.brk = (unsigned long) &_end; -#endif + init_mm.brk = _brk_end; code_resource.start = virt_to_phys(_text); code_resource.end = virt_to_phys(_etext)-1; @@ -824,8 +844,7 @@ void __init setup_arch(char **cmdline_p) #else num_physpages = max_pfn; - if (cpu_has_x2apic) - check_x2apic(); + check_x2apic(); /* How many end-of-memory variables you have, grandma! */ /* need this before calling reserve_initrd */ @@ -841,6 +860,8 @@ void __init setup_arch(char **cmdline_p) setup_bios_corruption_check(); #endif + reserve_brk(); + /* max_pfn_mapped is updated here */ max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT); max_pfn_mapped = max_low_pfn_mapped; @@ -865,9 +886,7 @@ void __init setup_arch(char **cmdline_p) reserve_initrd(); -#ifdef CONFIG_X86_64 vsmp_init(); -#endif io_delay_init(); @@ -893,12 +912,11 @@ void __init setup_arch(char **cmdline_p) */ acpi_reserve_bootmem(); #endif -#ifdef CONFIG_X86_FIND_SMP_CONFIG /* * Find and reserve possible boot-time SMP configuration: */ find_smp_config(); -#endif + reserve_crashkernel(); #ifdef CONFIG_X86_64 @@ -925,9 +943,7 @@ void __init setup_arch(char **cmdline_p) map_vsyscall(); #endif -#ifdef CONFIG_X86_GENERICARCH generic_apic_probe(); -#endif early_quirks(); @@ -978,4 +994,95 @@ void __init setup_arch(char **cmdline_p) #endif } +#ifdef CONFIG_X86_32 + +/** + * x86_quirk_pre_intr_init - initialisation prior to setting up interrupt vectors + * + * Description: + * Perform any necessary interrupt initialisation prior to setting up + * the "ordinary" interrupt call gates. For legacy reasons, the ISA + * interrupts should be initialised here if the machine emulates a PC + * in any way. + **/ +void __init x86_quirk_pre_intr_init(void) +{ + if (x86_quirks->arch_pre_intr_init) { + if (x86_quirks->arch_pre_intr_init()) + return; + } + init_ISA_irqs(); +} +/** + * x86_quirk_intr_init - post gate setup interrupt initialisation + * + * Description: + * Fill in any interrupts that may have been left out by the general + * init_IRQ() routine. interrupts having to do with the machine rather + * than the devices on the I/O bus (like APIC interrupts in intel MP + * systems) are started here. + **/ +void __init x86_quirk_intr_init(void) +{ + if (x86_quirks->arch_intr_init) { + if (x86_quirks->arch_intr_init()) + return; + } +} + +/** + * x86_quirk_trap_init - initialise system specific traps + * + * Description: + * Called as the final act of trap_init(). Used in VISWS to initialise + * the various board specific APIC traps. + **/ +void __init x86_quirk_trap_init(void) +{ + if (x86_quirks->arch_trap_init) { + if (x86_quirks->arch_trap_init()) + return; + } +} + +static struct irqaction irq0 = { + .handler = timer_interrupt, + .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_IRQPOLL | IRQF_TIMER, + .mask = CPU_MASK_NONE, + .name = "timer" +}; + +/** + * x86_quirk_pre_time_init - do any specific initialisations before. + * + **/ +void __init x86_quirk_pre_time_init(void) +{ + if (x86_quirks->arch_pre_time_init) + x86_quirks->arch_pre_time_init(); +} + +/** + * x86_quirk_time_init - do any specific initialisations for the system timer. + * + * Description: + * Must plug the system timer interrupt source at HZ into the IRQ listed + * in irq_vectors.h:TIMER_IRQ + **/ +void __init x86_quirk_time_init(void) +{ + if (x86_quirks->arch_time_init) { + /* + * A nonzero return code does not mean failure, it means + * that the architecture quirk does not want any + * generic (timer) setup to be performed after this: + */ + if (x86_quirks->arch_time_init()) + return; + } + + irq0.mask = cpumask_of_cpu(0); + setup_irq(0, &irq0); +} +#endif /* CONFIG_X86_32 */ diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 01161077a49..400331b50a5 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -7,402 +7,439 @@ #include <linux/crash_dump.h> #include <linux/smp.h> #include <linux/topology.h> +#include <linux/pfn.h> #include <asm/sections.h> #include <asm/processor.h> #include <asm/setup.h> #include <asm/mpspec.h> #include <asm/apicdef.h> #include <asm/highmem.h> +#include <asm/proto.h> +#include <asm/cpumask.h> +#include <asm/cpu.h> +#include <asm/stackprotector.h> -#ifdef CONFIG_X86_LOCAL_APIC -unsigned int num_processors; -unsigned disabled_cpus __cpuinitdata; -/* Processor that is doing the boot up */ -unsigned int boot_cpu_physical_apicid = -1U; -EXPORT_SYMBOL(boot_cpu_physical_apicid); -unsigned int max_physical_apicid; - -/* Bitmask of physically existing CPUs */ -physid_mask_t phys_cpu_present_map; +#ifdef CONFIG_DEBUG_PER_CPU_MAPS +# define DBG(x...) printk(KERN_DEBUG x) +#else +# define DBG(x...) #endif -/* map cpu index to physical APIC ID */ -DEFINE_EARLY_PER_CPU(u16, x86_cpu_to_apicid, BAD_APICID); -DEFINE_EARLY_PER_CPU(u16, x86_bios_cpu_apicid, BAD_APICID); -EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_apicid); -EXPORT_EARLY_PER_CPU_SYMBOL(x86_bios_cpu_apicid); - -#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64) -#define X86_64_NUMA 1 +DEFINE_PER_CPU(int, cpu_number); +EXPORT_PER_CPU_SYMBOL(cpu_number); -/* map cpu index to node index */ -DEFINE_EARLY_PER_CPU(int, x86_cpu_to_node_map, NUMA_NO_NODE); -EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_node_map); +#ifdef CONFIG_X86_64 +#define BOOT_PERCPU_OFFSET ((unsigned long)__per_cpu_load) +#else +#define BOOT_PERCPU_OFFSET 0 +#endif -/* which logical CPUs are on which nodes */ -cpumask_t *node_to_cpumask_map; -EXPORT_SYMBOL(node_to_cpumask_map); +DEFINE_PER_CPU(unsigned long, this_cpu_off) = BOOT_PERCPU_OFFSET; +EXPORT_PER_CPU_SYMBOL(this_cpu_off); -/* setup node_to_cpumask_map */ -static void __init setup_node_to_cpumask_map(void); +unsigned long __per_cpu_offset[NR_CPUS] __read_mostly = { + [0 ... NR_CPUS-1] = BOOT_PERCPU_OFFSET, +}; +EXPORT_SYMBOL(__per_cpu_offset); +/* + * On x86_64 symbols referenced from code should be reachable using + * 32bit relocations. Reserve space for static percpu variables in + * modules so that they are always served from the first chunk which + * is located at the percpu segment base. On x86_32, anything can + * address anywhere. No need to reserve space in the first chunk. + */ +#ifdef CONFIG_X86_64 +#define PERCPU_FIRST_CHUNK_RESERVE PERCPU_MODULE_RESERVE #else -static inline void setup_node_to_cpumask_map(void) { } +#define PERCPU_FIRST_CHUNK_RESERVE 0 #endif -#if defined(CONFIG_HAVE_SETUP_PER_CPU_AREA) && defined(CONFIG_X86_SMP) -/* - * Copy data used in early init routines from the initial arrays to the - * per cpu data areas. These arrays then become expendable and the - * *_early_ptr's are zeroed indicating that the static arrays are gone. +/** + * pcpu_need_numa - determine percpu allocation needs to consider NUMA + * + * If NUMA is not configured or there is only one NUMA node available, + * there is no reason to consider NUMA. This function determines + * whether percpu allocation should consider NUMA or not. + * + * RETURNS: + * true if NUMA should be considered; otherwise, false. */ -static void __init setup_per_cpu_maps(void) +static bool __init pcpu_need_numa(void) { - int cpu; +#ifdef CONFIG_NEED_MULTIPLE_NODES + pg_data_t *last = NULL; + unsigned int cpu; for_each_possible_cpu(cpu) { - per_cpu(x86_cpu_to_apicid, cpu) = - early_per_cpu_map(x86_cpu_to_apicid, cpu); - per_cpu(x86_bios_cpu_apicid, cpu) = - early_per_cpu_map(x86_bios_cpu_apicid, cpu); -#ifdef X86_64_NUMA - per_cpu(x86_cpu_to_node_map, cpu) = - early_per_cpu_map(x86_cpu_to_node_map, cpu); -#endif - } + int node = early_cpu_to_node(cpu); - /* indicate the early static arrays will soon be gone */ - early_per_cpu_ptr(x86_cpu_to_apicid) = NULL; - early_per_cpu_ptr(x86_bios_cpu_apicid) = NULL; -#ifdef X86_64_NUMA - early_per_cpu_ptr(x86_cpu_to_node_map) = NULL; + if (node_online(node) && NODE_DATA(node) && + last && last != NODE_DATA(node)) + return true; + + last = NODE_DATA(node); + } #endif + return false; } -#ifdef CONFIG_X86_32 -/* - * Great future not-so-futuristic plan: make i386 and x86_64 do it - * the same way - */ -unsigned long __per_cpu_offset[NR_CPUS] __read_mostly; -EXPORT_SYMBOL(__per_cpu_offset); -static inline void setup_cpu_pda_map(void) { } - -#elif !defined(CONFIG_SMP) -static inline void setup_cpu_pda_map(void) { } - -#else /* CONFIG_SMP && CONFIG_X86_64 */ - -/* - * Allocate cpu_pda pointer table and array via alloc_bootmem. +/** + * pcpu_alloc_bootmem - NUMA friendly alloc_bootmem wrapper for percpu + * @cpu: cpu to allocate for + * @size: size allocation in bytes + * @align: alignment + * + * Allocate @size bytes aligned at @align for cpu @cpu. This wrapper + * does the right thing for NUMA regardless of the current + * configuration. + * + * RETURNS: + * Pointer to the allocated area on success, NULL on failure. */ -static void __init setup_cpu_pda_map(void) +static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size, + unsigned long align) { - char *pda; - struct x8664_pda **new_cpu_pda; - unsigned long size; - int cpu; - - size = roundup(sizeof(struct x8664_pda), cache_line_size()); - - /* allocate cpu_pda array and pointer table */ - { - unsigned long tsize = nr_cpu_ids * sizeof(void *); - unsigned long asize = size * (nr_cpu_ids - 1); - - tsize = roundup(tsize, cache_line_size()); - new_cpu_pda = alloc_bootmem(tsize + asize); - pda = (char *)new_cpu_pda + tsize; - } - - /* initialize pointer table to static pda's */ - for_each_possible_cpu(cpu) { - if (cpu == 0) { - /* leave boot cpu pda in place */ - new_cpu_pda[0] = cpu_pda(0); - continue; - } - new_cpu_pda[cpu] = (struct x8664_pda *)pda; - new_cpu_pda[cpu]->in_bootmem = 1; - pda += size; + const unsigned long goal = __pa(MAX_DMA_ADDRESS); +#ifdef CONFIG_NEED_MULTIPLE_NODES + int node = early_cpu_to_node(cpu); + void *ptr; + + if (!node_online(node) || !NODE_DATA(node)) { + ptr = __alloc_bootmem_nopanic(size, align, goal); + pr_info("cpu %d has no node %d or node-local memory\n", + cpu, node); + pr_debug("per cpu data for cpu%d %lu bytes at %016lx\n", + cpu, size, __pa(ptr)); + } 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)); } - - /* point to new pointer table */ - _cpu_pda = new_cpu_pda; + return ptr; +#else + return __alloc_bootmem_nopanic(size, align, goal); +#endif } -#endif /* CONFIG_SMP && CONFIG_X86_64 */ - -#ifdef CONFIG_X86_64 +/* + * Remap allocator + * + * This allocator uses PMD page as unit. A PMD page is allocated for + * each cpu and each is remapped into vmalloc area using PMD mapping. + * As PMD page is quite large, only part of it is used for the first + * chunk. Unused part is returned to the bootmem allocator. + * + * So, the PMD pages are mapped twice - once to the physical mapping + * and to the vmalloc area for the first percpu chunk. The double + * mapping does add one more PMD TLB entry pressure but still is much + * better than only using 4k mappings while still being NUMA friendly. + */ +#ifdef CONFIG_NEED_MULTIPLE_NODES +static size_t pcpur_size __initdata; +static void **pcpur_ptrs __initdata; -/* correctly size the local cpu masks */ -static void __init setup_cpu_local_masks(void) +static struct page * __init pcpur_get_page(unsigned int cpu, int pageno) { - alloc_bootmem_cpumask_var(&cpu_initialized_mask); - alloc_bootmem_cpumask_var(&cpu_callin_mask); - alloc_bootmem_cpumask_var(&cpu_callout_mask); - alloc_bootmem_cpumask_var(&cpu_sibling_setup_mask); -} + size_t off = (size_t)pageno << PAGE_SHIFT; -#else /* CONFIG_X86_32 */ + if (off >= pcpur_size) + return NULL; -static inline void setup_cpu_local_masks(void) -{ + return virt_to_page(pcpur_ptrs[cpu] + off); } -#endif /* CONFIG_X86_32 */ - -/* - * Great future plan: - * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data. - * Always point %gs to its beginning - */ -void __init setup_per_cpu_areas(void) +static ssize_t __init setup_pcpu_remap(size_t static_size) { - ssize_t size, old_size; - char *ptr; - int cpu; - unsigned long align = 1; - - /* Setup cpu_pda map */ - setup_cpu_pda_map(); + static struct vm_struct vm; + pg_data_t *last; + size_t ptrs_size, dyn_size; + unsigned int cpu; + ssize_t ret; + + /* + * If large page isn't supported, there's no benefit in doing + * this. Also, on non-NUMA, embedding is better. + */ + if (!cpu_has_pse || pcpu_need_numa()) + return -EINVAL; + + last = NULL; + for_each_possible_cpu(cpu) { + int node = early_cpu_to_node(cpu); - /* Copy section for each CPU (we discard the original) */ - old_size = PERCPU_ENOUGH_ROOM; - align = max_t(unsigned long, PAGE_SIZE, align); - size = roundup(old_size, align); + if (node_online(node) && NODE_DATA(node) && + last && last != NODE_DATA(node)) + goto proceed; - pr_info("NR_CPUS:%d nr_cpumask_bits:%d nr_cpu_ids:%d nr_node_ids:%d\n", - NR_CPUS, nr_cpumask_bits, nr_cpu_ids, nr_node_ids); + last = NODE_DATA(node); + } + return -EINVAL; + +proceed: + /* + * Currently supports only single page. Supporting multiple + * pages won't be too difficult if it ever becomes necessary. + */ + pcpur_size = PFN_ALIGN(static_size + PERCPU_MODULE_RESERVE + + PERCPU_DYNAMIC_RESERVE); + if (pcpur_size > PMD_SIZE) { + pr_warning("PERCPU: static data is larger than large page, " + "can't use large page\n"); + return -EINVAL; + } + dyn_size = pcpur_size - static_size - PERCPU_FIRST_CHUNK_RESERVE; - pr_info("PERCPU: Allocating %zd bytes of per cpu data\n", size); + /* allocate pointer array and alloc large pages */ + ptrs_size = PFN_ALIGN(num_possible_cpus() * sizeof(pcpur_ptrs[0])); + pcpur_ptrs = alloc_bootmem(ptrs_size); for_each_possible_cpu(cpu) { -#ifndef CONFIG_NEED_MULTIPLE_NODES - ptr = __alloc_bootmem(size, align, - __pa(MAX_DMA_ADDRESS)); -#else - int node = early_cpu_to_node(cpu); - if (!node_online(node) || !NODE_DATA(node)) { - ptr = __alloc_bootmem(size, align, - __pa(MAX_DMA_ADDRESS)); - pr_info("cpu %d has no node %d or node-local memory\n", - cpu, node); - pr_debug("per cpu data for cpu%d at %016lx\n", - cpu, __pa(ptr)); - } else { - ptr = __alloc_bootmem_node(NODE_DATA(node), size, align, - __pa(MAX_DMA_ADDRESS)); - pr_debug("per cpu data for cpu%d on node%d at %016lx\n", - cpu, node, __pa(ptr)); - } -#endif - per_cpu_offset(cpu) = ptr - __per_cpu_start; - memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start); + pcpur_ptrs[cpu] = pcpu_alloc_bootmem(cpu, PMD_SIZE, PMD_SIZE); + if (!pcpur_ptrs[cpu]) + goto enomem; + + /* + * Only use pcpur_size bytes and give back the rest. + * + * Ingo: The 2MB up-rounding bootmem is needed to make + * sure the partial 2MB page is still fully RAM - it's + * not well-specified to have a PAT-incompatible area + * (unmapped RAM, device memory, etc.) in that hole. + */ + free_bootmem(__pa(pcpur_ptrs[cpu] + pcpur_size), + PMD_SIZE - pcpur_size); + + memcpy(pcpur_ptrs[cpu], __per_cpu_load, static_size); } - /* Setup percpu data maps */ - setup_per_cpu_maps(); - - /* Setup node to cpumask map */ - setup_node_to_cpumask_map(); - - /* Setup cpu initialized, callin, callout masks */ - setup_cpu_local_masks(); -} - -#endif + /* allocate address and map */ + vm.flags = VM_ALLOC; + vm.size = num_possible_cpus() * PMD_SIZE; + vm_area_register_early(&vm, PMD_SIZE); -#ifdef X86_64_NUMA + for_each_possible_cpu(cpu) { + pmd_t *pmd; -/* - * Allocate node_to_cpumask_map based on number of available nodes - * Requires node_possible_map to be valid. - * - * Note: node_to_cpumask() is not valid until after this is done. - */ -static void __init setup_node_to_cpumask_map(void) -{ - unsigned int node, num = 0; - cpumask_t *map; - - /* setup nr_node_ids if not done yet */ - if (nr_node_ids == MAX_NUMNODES) { - for_each_node_mask(node, node_possible_map) - num = node; - nr_node_ids = num + 1; + pmd = populate_extra_pmd((unsigned long)vm.addr + + cpu * PMD_SIZE); + set_pmd(pmd, pfn_pmd(page_to_pfn(virt_to_page(pcpur_ptrs[cpu])), + PAGE_KERNEL_LARGE)); } - /* allocate the map */ - map = alloc_bootmem_low(nr_node_ids * sizeof(cpumask_t)); - - pr_debug("Node to cpumask map at %p for %d nodes\n", - map, nr_node_ids); - - /* node_to_cpumask() will now work */ - node_to_cpumask_map = map; + /* we're ready, commit */ + pr_info("PERCPU: Remapped at %p with large pages, static data " + "%zu bytes\n", vm.addr, static_size); + + ret = pcpu_setup_first_chunk(pcpur_get_page, static_size, + PERCPU_FIRST_CHUNK_RESERVE, dyn_size, + PMD_SIZE, vm.addr, NULL); + goto out_free_ar; + +enomem: + for_each_possible_cpu(cpu) + if (pcpur_ptrs[cpu]) + free_bootmem(__pa(pcpur_ptrs[cpu]), PMD_SIZE); + ret = -ENOMEM; +out_free_ar: + free_bootmem(__pa(pcpur_ptrs), ptrs_size); + return ret; } - -void __cpuinit numa_set_node(int cpu, int node) +#else +static ssize_t __init setup_pcpu_remap(size_t static_size) { - int *cpu_to_node_map = early_per_cpu_ptr(x86_cpu_to_node_map); - - if (cpu_pda(cpu) && node != NUMA_NO_NODE) - cpu_pda(cpu)->nodenumber = node; - - if (cpu_to_node_map) - cpu_to_node_map[cpu] = node; - - else if (per_cpu_offset(cpu)) - per_cpu(x86_cpu_to_node_map, cpu) = node; - - else - pr_debug("Setting node for non-present cpu %d\n", cpu); + return -EINVAL; } +#endif -void __cpuinit numa_clear_node(int cpu) +/* + * Embedding allocator + * + * The first chunk is sized to just contain the static area plus + * module and dynamic reserves and embedded into linear physical + * mapping so that it can use PMD mapping without additional TLB + * pressure. + */ +static ssize_t __init setup_pcpu_embed(size_t static_size) { - numa_set_node(cpu, NUMA_NO_NODE); + size_t reserve = PERCPU_MODULE_RESERVE + PERCPU_DYNAMIC_RESERVE; + + /* + * If large page isn't supported, there's no benefit in doing + * this. Also, embedding allocation doesn't play well with + * NUMA. + */ + if (!cpu_has_pse || pcpu_need_numa()) + return -EINVAL; + + return pcpu_embed_first_chunk(static_size, PERCPU_FIRST_CHUNK_RESERVE, + reserve - PERCPU_FIRST_CHUNK_RESERVE, -1); } -#ifndef CONFIG_DEBUG_PER_CPU_MAPS +/* + * 4k page allocator + * + * This is the basic allocator. Static percpu area is allocated + * page-by-page and most of initialization is done by the generic + * setup function. + */ +static struct page **pcpu4k_pages __initdata; +static int pcpu4k_nr_static_pages __initdata; -void __cpuinit numa_add_cpu(int cpu) +static struct page * __init pcpu4k_get_page(unsigned int cpu, int pageno) { - cpu_set(cpu, node_to_cpumask_map[early_cpu_to_node(cpu)]); + if (pageno < pcpu4k_nr_static_pages) + return pcpu4k_pages[cpu * pcpu4k_nr_static_pages + pageno]; + return NULL; } -void __cpuinit numa_remove_cpu(int cpu) +static void __init pcpu4k_populate_pte(unsigned long addr) { - cpu_clear(cpu, node_to_cpumask_map[cpu_to_node(cpu)]); + populate_extra_pte(addr); } -#else /* CONFIG_DEBUG_PER_CPU_MAPS */ - -/* - * --------- debug versions of the numa functions --------- - */ -static void __cpuinit numa_set_cpumask(int cpu, int enable) +static ssize_t __init setup_pcpu_4k(size_t static_size) { - int node = cpu_to_node(cpu); - cpumask_t *mask; - char buf[64]; - - if (node_to_cpumask_map == NULL) { - printk(KERN_ERR "node_to_cpumask_map NULL\n"); - dump_stack(); - return; - } - - mask = &node_to_cpumask_map[node]; - if (enable) - cpu_set(cpu, *mask); - else - cpu_clear(cpu, *mask); + size_t pages_size; + unsigned int cpu; + int i, j; + ssize_t ret; + + pcpu4k_nr_static_pages = PFN_UP(static_size); + + /* unaligned allocations can't be freed, round up to page size */ + pages_size = PFN_ALIGN(pcpu4k_nr_static_pages * num_possible_cpus() + * sizeof(pcpu4k_pages[0])); + pcpu4k_pages = alloc_bootmem(pages_size); + + /* allocate and copy */ + j = 0; + for_each_possible_cpu(cpu) + for (i = 0; i < pcpu4k_nr_static_pages; i++) { + void *ptr; + + ptr = pcpu_alloc_bootmem(cpu, PAGE_SIZE, PAGE_SIZE); + if (!ptr) + goto enomem; + + memcpy(ptr, __per_cpu_load + i * PAGE_SIZE, PAGE_SIZE); + pcpu4k_pages[j++] = virt_to_page(ptr); + } - cpulist_scnprintf(buf, sizeof(buf), mask); - printk(KERN_DEBUG "%s cpu %d node %d: mask now %s\n", - enable ? "numa_add_cpu" : "numa_remove_cpu", cpu, node, buf); + /* we're ready, commit */ + pr_info("PERCPU: Allocated %d 4k pages, static data %zu bytes\n", + pcpu4k_nr_static_pages, static_size); + + ret = pcpu_setup_first_chunk(pcpu4k_get_page, static_size, + PERCPU_FIRST_CHUNK_RESERVE, -1, + -1, NULL, pcpu4k_populate_pte); + goto out_free_ar; + +enomem: + while (--j >= 0) + free_bootmem(__pa(page_address(pcpu4k_pages[j])), PAGE_SIZE); + ret = -ENOMEM; +out_free_ar: + free_bootmem(__pa(pcpu4k_pages), pages_size); + return ret; } -void __cpuinit numa_add_cpu(int cpu) +static inline void setup_percpu_segment(int cpu) { - numa_set_cpumask(cpu, 1); -} - -void __cpuinit numa_remove_cpu(int cpu) -{ - numa_set_cpumask(cpu, 0); -} +#ifdef CONFIG_X86_32 + struct desc_struct gdt; -int cpu_to_node(int cpu) -{ - if (early_per_cpu_ptr(x86_cpu_to_node_map)) { - printk(KERN_WARNING - "cpu_to_node(%d): usage too early!\n", cpu); - dump_stack(); - return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu]; - } - return per_cpu(x86_cpu_to_node_map, cpu); + pack_descriptor(&gdt, per_cpu_offset(cpu), 0xFFFFF, + 0x2 | DESCTYPE_S, 0x8); + gdt.s = 1; + write_gdt_entry(get_cpu_gdt_table(cpu), + GDT_ENTRY_PERCPU, &gdt, DESCTYPE_S); +#endif } -EXPORT_SYMBOL(cpu_to_node); /* - * Same function as cpu_to_node() but used if called before the - * per_cpu areas are setup. + * Great future plan: + * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data. + * Always point %gs to its beginning */ -int early_cpu_to_node(int cpu) +void __init setup_per_cpu_areas(void) { - if (early_per_cpu_ptr(x86_cpu_to_node_map)) - return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu]; - - if (!per_cpu_offset(cpu)) { - printk(KERN_WARNING - "early_cpu_to_node(%d): no per_cpu area!\n", cpu); - dump_stack(); - return NUMA_NO_NODE; - } - return per_cpu(x86_cpu_to_node_map, cpu); -} + size_t static_size = __per_cpu_end - __per_cpu_start; + unsigned int cpu; + unsigned long delta; + size_t pcpu_unit_size; + ssize_t ret; + pr_info("NR_CPUS:%d nr_cpumask_bits:%d nr_cpu_ids:%d nr_node_ids:%d\n", + NR_CPUS, nr_cpumask_bits, nr_cpu_ids, nr_node_ids); -/* empty cpumask */ -static const cpumask_t cpu_mask_none; - -/* - * Returns a pointer to the bitmask of CPUs on Node 'node'. - */ -const cpumask_t *cpumask_of_node(int node) -{ - if (node_to_cpumask_map == NULL) { - printk(KERN_WARNING - "cpumask_of_node(%d): no node_to_cpumask_map!\n", - node); - dump_stack(); - return (const cpumask_t *)&cpu_online_map; - } - if (node >= nr_node_ids) { - printk(KERN_WARNING - "cpumask_of_node(%d): node > nr_node_ids(%d)\n", - node, nr_node_ids); - dump_stack(); - return &cpu_mask_none; - } - return &node_to_cpumask_map[node]; -} -EXPORT_SYMBOL(cpumask_of_node); - -/* - * Returns a bitmask of CPUs on Node 'node'. - * - * Side note: this function creates the returned cpumask on the stack - * so with a high NR_CPUS count, excessive stack space is used. The - * node_to_cpumask_ptr function should be used whenever possible. - */ -cpumask_t node_to_cpumask(int node) -{ - if (node_to_cpumask_map == NULL) { - printk(KERN_WARNING - "node_to_cpumask(%d): no node_to_cpumask_map!\n", node); - dump_stack(); - return cpu_online_map; - } - if (node >= nr_node_ids) { - printk(KERN_WARNING - "node_to_cpumask(%d): node > nr_node_ids(%d)\n", - node, nr_node_ids); - dump_stack(); - return cpu_mask_none; + /* + * Allocate percpu area. If PSE is supported, try to make use + * of large page mappings. Please read comments on top of + * each allocator for details. + */ + ret = setup_pcpu_remap(static_size); + if (ret < 0) + ret = setup_pcpu_embed(static_size); + if (ret < 0) + ret = setup_pcpu_4k(static_size); + if (ret < 0) + panic("cannot allocate static percpu area (%zu bytes, err=%zd)", + static_size, ret); + + pcpu_unit_size = ret; + + /* alrighty, percpu areas up and running */ + delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start; + for_each_possible_cpu(cpu) { + per_cpu_offset(cpu) = delta + cpu * pcpu_unit_size; + per_cpu(this_cpu_off, cpu) = per_cpu_offset(cpu); + per_cpu(cpu_number, cpu) = cpu; + setup_percpu_segment(cpu); + setup_stack_canary_segment(cpu); + /* + * Copy data used in early init routines from the + * initial arrays to the per cpu data areas. These + * arrays then become expendable and the *_early_ptr's + * are zeroed indicating that the static arrays are + * gone. + */ +#ifdef CONFIG_X86_LOCAL_APIC + per_cpu(x86_cpu_to_apicid, cpu) = + early_per_cpu_map(x86_cpu_to_apicid, cpu); + per_cpu(x86_bios_cpu_apicid, cpu) = + early_per_cpu_map(x86_bios_cpu_apicid, cpu); +#endif +#ifdef CONFIG_X86_64 + per_cpu(irq_stack_ptr, cpu) = + per_cpu(irq_stack_union.irq_stack, cpu) + + IRQ_STACK_SIZE - 64; +#ifdef CONFIG_NUMA + per_cpu(x86_cpu_to_node_map, cpu) = + early_per_cpu_map(x86_cpu_to_node_map, cpu); +#endif +#endif + /* + * Up to this point, the boot CPU has been using .data.init + * area. Reload any changed state for the boot CPU. + */ + if (cpu == boot_cpu_id) + switch_to_new_gdt(cpu); } - return node_to_cpumask_map[node]; -} -EXPORT_SYMBOL(node_to_cpumask); -/* - * --------- end of debug versions of the numa functions --------- - */ - -#endif /* CONFIG_DEBUG_PER_CPU_MAPS */ + /* indicate the early static arrays will soon be gone */ +#ifdef CONFIG_X86_LOCAL_APIC + early_per_cpu_ptr(x86_cpu_to_apicid) = NULL; + early_per_cpu_ptr(x86_bios_cpu_apicid) = NULL; +#endif +#if defined(CONFIG_X86_64) && defined(CONFIG_NUMA) + early_per_cpu_ptr(x86_cpu_to_node_map) = NULL; +#endif -#endif /* X86_64_NUMA */ + /* Setup node to cpumask map */ + setup_node_to_cpumask_map(); + /* Setup cpu initialized, callin, callout masks */ + setup_cpu_local_masks(); +} diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c index df0587f24c5..d2cc6428c58 100644 --- a/arch/x86/kernel/signal.c +++ b/arch/x86/kernel/signal.c @@ -50,27 +50,23 @@ # define FIX_EFLAGS __FIX_EFLAGS #endif -#define COPY(x) { \ - err |= __get_user(regs->x, &sc->x); \ -} +#define COPY(x) do { \ + get_user_ex(regs->x, &sc->x); \ +} while (0) -#define COPY_SEG(seg) { \ - unsigned short tmp; \ - err |= __get_user(tmp, &sc->seg); \ - regs->seg = tmp; \ -} +#define GET_SEG(seg) ({ \ + unsigned short tmp; \ + get_user_ex(tmp, &sc->seg); \ + tmp; \ +}) -#define COPY_SEG_CPL3(seg) { \ - unsigned short tmp; \ - err |= __get_user(tmp, &sc->seg); \ - regs->seg = tmp | 3; \ -} +#define COPY_SEG(seg) do { \ + regs->seg = GET_SEG(seg); \ +} while (0) -#define GET_SEG(seg) { \ - unsigned short tmp; \ - err |= __get_user(tmp, &sc->seg); \ - loadsegment(seg, tmp); \ -} +#define COPY_SEG_CPL3(seg) do { \ + regs->seg = GET_SEG(seg) | 3; \ +} while (0) static int restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, @@ -83,45 +79,49 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, /* Always make any pending restarted system calls return -EINTR */ current_thread_info()->restart_block.fn = do_no_restart_syscall; + get_user_try { + #ifdef CONFIG_X86_32 - GET_SEG(gs); - COPY_SEG(fs); - COPY_SEG(es); - COPY_SEG(ds); + set_user_gs(regs, GET_SEG(gs)); + COPY_SEG(fs); + COPY_SEG(es); + COPY_SEG(ds); #endif /* CONFIG_X86_32 */ - COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx); - COPY(dx); COPY(cx); COPY(ip); + COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx); + COPY(dx); COPY(cx); COPY(ip); #ifdef CONFIG_X86_64 - COPY(r8); - COPY(r9); - COPY(r10); - COPY(r11); - COPY(r12); - COPY(r13); - COPY(r14); - COPY(r15); + COPY(r8); + COPY(r9); + COPY(r10); + COPY(r11); + COPY(r12); + COPY(r13); + COPY(r14); + COPY(r15); #endif /* CONFIG_X86_64 */ #ifdef CONFIG_X86_32 - COPY_SEG_CPL3(cs); - COPY_SEG_CPL3(ss); + COPY_SEG_CPL3(cs); + COPY_SEG_CPL3(ss); #else /* !CONFIG_X86_32 */ - /* Kernel saves and restores only the CS segment register on signals, - * which is the bare minimum needed to allow mixed 32/64-bit code. - * App's signal handler can save/restore other segments if needed. */ - COPY_SEG_CPL3(cs); + /* Kernel saves and restores only the CS segment register on signals, + * which is the bare minimum needed to allow mixed 32/64-bit code. + * App's signal handler can save/restore other segments if needed. */ + COPY_SEG_CPL3(cs); #endif /* CONFIG_X86_32 */ - err |= __get_user(tmpflags, &sc->flags); - regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS); - regs->orig_ax = -1; /* disable syscall checks */ + get_user_ex(tmpflags, &sc->flags); + regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS); + regs->orig_ax = -1; /* disable syscall checks */ - err |= __get_user(buf, &sc->fpstate); - err |= restore_i387_xstate(buf); + get_user_ex(buf, &sc->fpstate); + err |= restore_i387_xstate(buf); + + get_user_ex(*pax, &sc->ax); + } get_user_catch(err); - err |= __get_user(*pax, &sc->ax); return err; } @@ -131,57 +131,55 @@ setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate, { int err = 0; -#ifdef CONFIG_X86_32 - { - unsigned int tmp; + put_user_try { - savesegment(gs, tmp); - err |= __put_user(tmp, (unsigned int __user *)&sc->gs); - } - err |= __put_user(regs->fs, (unsigned int __user *)&sc->fs); - err |= __put_user(regs->es, (unsigned int __user *)&sc->es); - err |= __put_user(regs->ds, (unsigned int __user *)&sc->ds); +#ifdef CONFIG_X86_32 + put_user_ex(get_user_gs(regs), (unsigned int __user *)&sc->gs); + put_user_ex(regs->fs, (unsigned int __user *)&sc->fs); + put_user_ex(regs->es, (unsigned int __user *)&sc->es); + put_user_ex(regs->ds, (unsigned int __user *)&sc->ds); #endif /* CONFIG_X86_32 */ - err |= __put_user(regs->di, &sc->di); - err |= __put_user(regs->si, &sc->si); - err |= __put_user(regs->bp, &sc->bp); - err |= __put_user(regs->sp, &sc->sp); - err |= __put_user(regs->bx, &sc->bx); - err |= __put_user(regs->dx, &sc->dx); - err |= __put_user(regs->cx, &sc->cx); - err |= __put_user(regs->ax, &sc->ax); + put_user_ex(regs->di, &sc->di); + put_user_ex(regs->si, &sc->si); + put_user_ex(regs->bp, &sc->bp); + put_user_ex(regs->sp, &sc->sp); + put_user_ex(regs->bx, &sc->bx); + put_user_ex(regs->dx, &sc->dx); + put_user_ex(regs->cx, &sc->cx); + put_user_ex(regs->ax, &sc->ax); #ifdef CONFIG_X86_64 - err |= __put_user(regs->r8, &sc->r8); - err |= __put_user(regs->r9, &sc->r9); - err |= __put_user(regs->r10, &sc->r10); - err |= __put_user(regs->r11, &sc->r11); - err |= __put_user(regs->r12, &sc->r12); - err |= __put_user(regs->r13, &sc->r13); - err |= __put_user(regs->r14, &sc->r14); - err |= __put_user(regs->r15, &sc->r15); + put_user_ex(regs->r8, &sc->r8); + put_user_ex(regs->r9, &sc->r9); + put_user_ex(regs->r10, &sc->r10); + put_user_ex(regs->r11, &sc->r11); + put_user_ex(regs->r12, &sc->r12); + put_user_ex(regs->r13, &sc->r13); + put_user_ex(regs->r14, &sc->r14); + put_user_ex(regs->r15, &sc->r15); #endif /* CONFIG_X86_64 */ - err |= __put_user(current->thread.trap_no, &sc->trapno); - err |= __put_user(current->thread.error_code, &sc->err); - err |= __put_user(regs->ip, &sc->ip); + put_user_ex(current->thread.trap_no, &sc->trapno); + put_user_ex(current->thread.error_code, &sc->err); + put_user_ex(regs->ip, &sc->ip); #ifdef CONFIG_X86_32 - err |= __put_user(regs->cs, (unsigned int __user *)&sc->cs); - err |= __put_user(regs->flags, &sc->flags); - err |= __put_user(regs->sp, &sc->sp_at_signal); - err |= __put_user(regs->ss, (unsigned int __user *)&sc->ss); + put_user_ex(regs->cs, (unsigned int __user *)&sc->cs); + put_user_ex(regs->flags, &sc->flags); + put_user_ex(regs->sp, &sc->sp_at_signal); + put_user_ex(regs->ss, (unsigned int __user *)&sc->ss); #else /* !CONFIG_X86_32 */ - err |= __put_user(regs->flags, &sc->flags); - err |= __put_user(regs->cs, &sc->cs); - err |= __put_user(0, &sc->gs); - err |= __put_user(0, &sc->fs); + put_user_ex(regs->flags, &sc->flags); + put_user_ex(regs->cs, &sc->cs); + put_user_ex(0, &sc->gs); + put_user_ex(0, &sc->fs); #endif /* CONFIG_X86_32 */ - err |= __put_user(fpstate, &sc->fpstate); + put_user_ex(fpstate, &sc->fpstate); - /* non-iBCS2 extensions.. */ - err |= __put_user(mask, &sc->oldmask); - err |= __put_user(current->thread.cr2, &sc->cr2); + /* non-iBCS2 extensions.. */ + put_user_ex(mask, &sc->oldmask); + put_user_ex(current->thread.cr2, &sc->cr2); + } put_user_catch(err); return err; } @@ -189,40 +187,35 @@ setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate, /* * Set up a signal frame. */ -#ifdef CONFIG_X86_32 -static const struct { - u16 poplmovl; - u32 val; - u16 int80; -} __attribute__((packed)) retcode = { - 0xb858, /* popl %eax; movl $..., %eax */ - __NR_sigreturn, - 0x80cd, /* int $0x80 */ -}; - -static const struct { - u8 movl; - u32 val; - u16 int80; - u8 pad; -} __attribute__((packed)) rt_retcode = { - 0xb8, /* movl $..., %eax */ - __NR_rt_sigreturn, - 0x80cd, /* int $0x80 */ - 0 -}; /* * Determine which stack to use.. */ +static unsigned long align_sigframe(unsigned long sp) +{ +#ifdef CONFIG_X86_32 + /* + * Align the stack pointer according to the i386 ABI, + * i.e. so that on function entry ((sp + 4) & 15) == 0. + */ + sp = ((sp + 4) & -16ul) - 4; +#else /* !CONFIG_X86_32 */ + sp = round_down(sp, 16) - 8; +#endif + return sp; +} + static inline void __user * get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size, - void **fpstate) + void __user **fpstate) { - unsigned long sp; - /* Default to using normal stack */ - sp = regs->sp; + unsigned long sp = regs->sp; + +#ifdef CONFIG_X86_64 + /* redzone */ + sp -= 128; +#endif /* CONFIG_X86_64 */ /* * If we are on the alternate signal stack and would overflow it, don't. @@ -236,30 +229,52 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size, if (sas_ss_flags(sp) == 0) sp = current->sas_ss_sp + current->sas_ss_size; } else { +#ifdef CONFIG_X86_32 /* This is the legacy signal stack switching. */ if ((regs->ss & 0xffff) != __USER_DS && !(ka->sa.sa_flags & SA_RESTORER) && ka->sa.sa_restorer) sp = (unsigned long) ka->sa.sa_restorer; +#endif /* CONFIG_X86_32 */ } if (used_math()) { - sp = sp - sig_xstate_size; - *fpstate = (struct _fpstate *) sp; + sp -= sig_xstate_size; +#ifdef CONFIG_X86_64 + sp = round_down(sp, 64); +#endif /* CONFIG_X86_64 */ + *fpstate = (void __user *)sp; + if (save_i387_xstate(*fpstate) < 0) return (void __user *)-1L; } - sp -= frame_size; - /* - * Align the stack pointer according to the i386 ABI, - * i.e. so that on function entry ((sp + 4) & 15) == 0. - */ - sp = ((sp + 4) & -16ul) - 4; - - return (void __user *) sp; + return (void __user *)align_sigframe(sp - frame_size); } +#ifdef CONFIG_X86_32 +static const struct { + u16 poplmovl; + u32 val; + u16 int80; +} __attribute__((packed)) retcode = { + 0xb858, /* popl %eax; movl $..., %eax */ + __NR_sigreturn, + 0x80cd, /* int $0x80 */ +}; + +static const struct { + u8 movl; + u32 val; + u16 int80; + u8 pad; +} __attribute__((packed)) rt_retcode = { + 0xb8, /* movl $..., %eax */ + __NR_rt_sigreturn, + 0x80cd, /* int $0x80 */ + 0 +}; + static int __setup_frame(int sig, struct k_sigaction *ka, sigset_t *set, struct pt_regs *regs) @@ -336,43 +351,41 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) return -EFAULT; - err |= __put_user(sig, &frame->sig); - err |= __put_user(&frame->info, &frame->pinfo); - err |= __put_user(&frame->uc, &frame->puc); - err |= copy_siginfo_to_user(&frame->info, info); - if (err) - return -EFAULT; + put_user_try { + put_user_ex(sig, &frame->sig); + put_user_ex(&frame->info, &frame->pinfo); + put_user_ex(&frame->uc, &frame->puc); + err |= copy_siginfo_to_user(&frame->info, info); - /* Create the ucontext. */ - if (cpu_has_xsave) - err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags); - else - err |= __put_user(0, &frame->uc.uc_flags); - err |= __put_user(0, &frame->uc.uc_link); - err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp); - err |= __put_user(sas_ss_flags(regs->sp), - &frame->uc.uc_stack.ss_flags); - err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size); - err |= setup_sigcontext(&frame->uc.uc_mcontext, fpstate, - regs, set->sig[0]); - err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); - if (err) - return -EFAULT; - - /* Set up to return from userspace. */ - restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); - if (ka->sa.sa_flags & SA_RESTORER) - restorer = ka->sa.sa_restorer; - err |= __put_user(restorer, &frame->pretcode); + /* Create the ucontext. */ + if (cpu_has_xsave) + put_user_ex(UC_FP_XSTATE, &frame->uc.uc_flags); + else + put_user_ex(0, &frame->uc.uc_flags); + put_user_ex(0, &frame->uc.uc_link); + put_user_ex(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp); + put_user_ex(sas_ss_flags(regs->sp), + &frame->uc.uc_stack.ss_flags); + put_user_ex(current->sas_ss_size, &frame->uc.uc_stack.ss_size); + err |= setup_sigcontext(&frame->uc.uc_mcontext, fpstate, + regs, set->sig[0]); + err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); + + /* Set up to return from userspace. */ + restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); + if (ka->sa.sa_flags & SA_RESTORER) + restorer = ka->sa.sa_restorer; + put_user_ex(restorer, &frame->pretcode); - /* - * This is movl $__NR_rt_sigreturn, %ax ; int $0x80 - * - * WE DO NOT USE IT ANY MORE! It's only left here for historical - * reasons and because gdb uses it as a signature to notice - * signal handler stack frames. - */ - err |= __put_user(*((u64 *)&rt_retcode), (u64 *)frame->retcode); + /* + * This is movl $__NR_rt_sigreturn, %ax ; int $0x80 + * + * WE DO NOT USE IT ANY MORE! It's only left here for historical + * reasons and because gdb uses it as a signature to notice + * signal handler stack frames. + */ + put_user_ex(*((u64 *)&rt_retcode), (u64 *)frame->retcode); + } put_user_catch(err); if (err) return -EFAULT; @@ -392,24 +405,6 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, return 0; } #else /* !CONFIG_X86_32 */ -/* - * Determine which stack to use.. - */ -static void __user * -get_stack(struct k_sigaction *ka, unsigned long sp, unsigned long size) -{ - /* Default to using normal stack - redzone*/ - sp -= 128; - - /* This is the X/Open sanctioned signal stack switching. */ - if (ka->sa.sa_flags & SA_ONSTACK) { - if (sas_ss_flags(sp) == 0) - sp = current->sas_ss_sp + current->sas_ss_size; - } - - return (void __user *)round_down(sp - size, 64); -} - static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, struct pt_regs *regs) { @@ -418,15 +413,7 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, int err = 0; struct task_struct *me = current; - if (used_math()) { - fp = get_stack(ka, regs->sp, sig_xstate_size); - frame = (void __user *)round_down( - (unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8; - - if (save_i387_xstate(fp) < 0) - return -EFAULT; - } else - frame = get_stack(ka, regs->sp, sizeof(struct rt_sigframe)) - 8; + frame = get_sigframe(ka, regs, sizeof(struct rt_sigframe), &fp); if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) return -EFAULT; @@ -436,28 +423,30 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, return -EFAULT; } - /* Create the ucontext. */ - if (cpu_has_xsave) - err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags); - else - err |= __put_user(0, &frame->uc.uc_flags); - err |= __put_user(0, &frame->uc.uc_link); - err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp); - err |= __put_user(sas_ss_flags(regs->sp), - &frame->uc.uc_stack.ss_flags); - err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size); - err |= setup_sigcontext(&frame->uc.uc_mcontext, fp, regs, set->sig[0]); - err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); - - /* Set up to return from userspace. If provided, use a stub - already in userspace. */ - /* x86-64 should always use SA_RESTORER. */ - if (ka->sa.sa_flags & SA_RESTORER) { - err |= __put_user(ka->sa.sa_restorer, &frame->pretcode); - } else { - /* could use a vstub here */ - return -EFAULT; - } + put_user_try { + /* Create the ucontext. */ + if (cpu_has_xsave) + put_user_ex(UC_FP_XSTATE, &frame->uc.uc_flags); + else + put_user_ex(0, &frame->uc.uc_flags); + put_user_ex(0, &frame->uc.uc_link); + put_user_ex(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp); + put_user_ex(sas_ss_flags(regs->sp), + &frame->uc.uc_stack.ss_flags); + put_user_ex(me->sas_ss_size, &frame->uc.uc_stack.ss_size); + err |= setup_sigcontext(&frame->uc.uc_mcontext, fp, regs, set->sig[0]); + err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); + + /* Set up to return from userspace. If provided, use a stub + already in userspace. */ + /* x86-64 should always use SA_RESTORER. */ + if (ka->sa.sa_flags & SA_RESTORER) { + put_user_ex(ka->sa.sa_restorer, &frame->pretcode); + } else { + /* could use a vstub here */ + err |= -EFAULT; + } + } put_user_catch(err); if (err) return -EFAULT; @@ -509,31 +498,41 @@ sys_sigaction(int sig, const struct old_sigaction __user *act, struct old_sigaction __user *oact) { struct k_sigaction new_ka, old_ka; - int ret; + int ret = 0; if (act) { old_sigset_t mask; - if (!access_ok(VERIFY_READ, act, sizeof(*act)) || - __get_user(new_ka.sa.sa_handler, &act->sa_handler) || - __get_user(new_ka.sa.sa_restorer, &act->sa_restorer)) + if (!access_ok(VERIFY_READ, act, sizeof(*act))) return -EFAULT; - __get_user(new_ka.sa.sa_flags, &act->sa_flags); - __get_user(mask, &act->sa_mask); + get_user_try { + get_user_ex(new_ka.sa.sa_handler, &act->sa_handler); + get_user_ex(new_ka.sa.sa_flags, &act->sa_flags); + get_user_ex(mask, &act->sa_mask); + get_user_ex(new_ka.sa.sa_restorer, &act->sa_restorer); + } get_user_catch(ret); + + if (ret) + return -EFAULT; siginitset(&new_ka.sa.sa_mask, mask); } ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); if (!ret && oact) { - if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) || - __put_user(old_ka.sa.sa_handler, &oact->sa_handler) || - __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer)) + if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact))) return -EFAULT; - __put_user(old_ka.sa.sa_flags, &oact->sa_flags); - __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); + put_user_try { + put_user_ex(old_ka.sa.sa_handler, &oact->sa_handler); + put_user_ex(old_ka.sa.sa_flags, &oact->sa_flags); + put_user_ex(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); + put_user_ex(old_ka.sa.sa_restorer, &oact->sa_restorer); + } put_user_catch(ret); + + if (ret) + return -EFAULT; } return ret; @@ -541,14 +540,9 @@ sys_sigaction(int sig, const struct old_sigaction __user *act, #endif /* CONFIG_X86_32 */ #ifdef CONFIG_X86_32 -asmlinkage int sys_sigaltstack(unsigned long bx) +int sys_sigaltstack(struct pt_regs *regs) { - /* - * This is needed to make gcc realize it doesn't own the - * "struct pt_regs" - */ - struct pt_regs *regs = (struct pt_regs *)&bx; - const stack_t __user *uss = (const stack_t __user *)bx; + 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); @@ -566,14 +560,12 @@ sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, * Do a signal return; undo the signal stack. */ #ifdef CONFIG_X86_32 -asmlinkage unsigned long sys_sigreturn(unsigned long __unused) +unsigned long sys_sigreturn(struct pt_regs *regs) { struct sigframe __user *frame; - struct pt_regs *regs; unsigned long ax; sigset_t set; - regs = (struct pt_regs *) &__unused; frame = (struct sigframe __user *)(regs->sp - 8); if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) @@ -600,7 +592,7 @@ badframe: } #endif /* CONFIG_X86_32 */ -static long do_rt_sigreturn(struct pt_regs *regs) +long sys_rt_sigreturn(struct pt_regs *regs) { struct rt_sigframe __user *frame; unsigned long ax; @@ -631,25 +623,6 @@ badframe: return 0; } -#ifdef CONFIG_X86_32 -/* - * Note: do not pass in pt_regs directly as with tail-call optimization - * GCC will incorrectly stomp on the caller's frame and corrupt user-space - * register state: - */ -asmlinkage int sys_rt_sigreturn(unsigned long __unused) -{ - struct pt_regs *regs = (struct pt_regs *)&__unused; - - return do_rt_sigreturn(regs); -} -#else /* !CONFIG_X86_32 */ -asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) -{ - return do_rt_sigreturn(regs); -} -#endif /* CONFIG_X86_32 */ - /* * OK, we're invoking a handler: */ diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c index e6faa3316bd..13f33ea8cca 100644 --- a/arch/x86/kernel/smp.c +++ b/arch/x86/kernel/smp.c @@ -2,7 +2,7 @@ * Intel SMP support routines. * * (c) 1995 Alan Cox, Building #3 <alan@lxorguk.ukuu.org.uk> - * (c) 1998-99, 2000 Ingo Molnar <mingo@redhat.com> + * (c) 1998-99, 2000, 2009 Ingo Molnar <mingo@redhat.com> * (c) 2002,2003 Andi Kleen, SuSE Labs. * * i386 and x86_64 integration by Glauber Costa <gcosta@redhat.com> @@ -26,8 +26,7 @@ #include <asm/tlbflush.h> #include <asm/mmu_context.h> #include <asm/proto.h> -#include <mach_ipi.h> -#include <mach_apic.h> +#include <asm/apic.h> /* * Some notes on x86 processor bugs affecting SMP operation: * @@ -118,12 +117,12 @@ static void native_smp_send_reschedule(int cpu) WARN_ON(1); return; } - send_IPI_mask(cpumask_of(cpu), RESCHEDULE_VECTOR); + apic->send_IPI_mask(cpumask_of(cpu), RESCHEDULE_VECTOR); } void native_send_call_func_single_ipi(int cpu) { - send_IPI_mask(cpumask_of(cpu), CALL_FUNCTION_SINGLE_VECTOR); + apic->send_IPI_mask(cpumask_of(cpu), CALL_FUNCTION_SINGLE_VECTOR); } void native_send_call_func_ipi(const struct cpumask *mask) @@ -131,7 +130,7 @@ void native_send_call_func_ipi(const struct cpumask *mask) cpumask_var_t allbutself; if (!alloc_cpumask_var(&allbutself, GFP_ATOMIC)) { - send_IPI_mask(mask, CALL_FUNCTION_VECTOR); + apic->send_IPI_mask(mask, CALL_FUNCTION_VECTOR); return; } @@ -140,9 +139,9 @@ void native_send_call_func_ipi(const struct cpumask *mask) if (cpumask_equal(mask, allbutself) && cpumask_equal(cpu_online_mask, cpu_callout_mask)) - send_IPI_allbutself(CALL_FUNCTION_VECTOR); + apic->send_IPI_allbutself(CALL_FUNCTION_VECTOR); else - send_IPI_mask(mask, CALL_FUNCTION_VECTOR); + apic->send_IPI_mask(mask, CALL_FUNCTION_VECTOR); free_cpumask_var(allbutself); } diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index bb1a3b1fc87..ef7d10170c3 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -2,7 +2,7 @@ * x86 SMP booting functions * * (c) 1995 Alan Cox, Building #3 <alan@lxorguk.ukuu.org.uk> - * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com> + * (c) 1998, 1999, 2000, 2009 Ingo Molnar <mingo@redhat.com> * Copyright 2001 Andi Kleen, SuSE Labs. * * Much of the core SMP work is based on previous work by Thomas Radke, to @@ -53,7 +53,6 @@ #include <asm/nmi.h> #include <asm/irq.h> #include <asm/idle.h> -#include <asm/smp.h> #include <asm/trampoline.h> #include <asm/cpu.h> #include <asm/numa.h> @@ -61,13 +60,12 @@ #include <asm/tlbflush.h> #include <asm/mtrr.h> #include <asm/vmi.h> -#include <asm/genapic.h> +#include <asm/apic.h> #include <asm/setup.h> +#include <asm/uv/uv.h> #include <linux/mc146818rtc.h> -#include <mach_apic.h> -#include <mach_wakecpu.h> -#include <smpboot_hooks.h> +#include <asm/smpboot_hooks.h> #ifdef CONFIG_X86_32 u8 apicid_2_node[MAX_APICID]; @@ -114,11 +112,7 @@ EXPORT_PER_CPU_SYMBOL(cpu_core_map); DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info); EXPORT_PER_CPU_SYMBOL(cpu_info); -static atomic_t init_deasserted; - - -/* Set if we find a B stepping CPU */ -static int __cpuinitdata smp_b_stepping; +atomic_t init_deasserted; #if defined(CONFIG_NUMA) && defined(CONFIG_X86_32) @@ -163,7 +157,7 @@ static void map_cpu_to_logical_apicid(void) { int cpu = smp_processor_id(); int apicid = logical_smp_processor_id(); - int node = apicid_to_node(apicid); + int node = apic->apicid_to_node(apicid); if (!node_online(node)) node = first_online_node; @@ -196,7 +190,8 @@ static void __cpuinit smp_callin(void) * our local APIC. We have to wait for the IPI or we'll * lock up on an APIC access. */ - wait_for_init_deassert(&init_deasserted); + if (apic->wait_for_init_deassert) + apic->wait_for_init_deassert(&init_deasserted); /* * (This works even if the APIC is not enabled.) @@ -243,7 +238,8 @@ static void __cpuinit smp_callin(void) */ pr_debug("CALLIN, before setup_local_APIC().\n"); - smp_callin_clear_local_apic(); + if (apic->smp_callin_clear_local_apic) + apic->smp_callin_clear_local_apic(); setup_local_APIC(); end_local_APIC_setup(); map_cpu_to_logical_apicid(); @@ -271,8 +267,6 @@ static void __cpuinit smp_callin(void) cpumask_set_cpu(cpuid, cpu_callin_mask); } -static int __cpuinitdata unsafe_smp; - /* * Activate a secondary processor. */ @@ -340,76 +334,6 @@ notrace static void __cpuinit start_secondary(void *unused) cpu_idle(); } -static void __cpuinit smp_apply_quirks(struct cpuinfo_x86 *c) -{ - /* - * Mask B, Pentium, but not Pentium MMX - */ - if (c->x86_vendor == X86_VENDOR_INTEL && - c->x86 == 5 && - c->x86_mask >= 1 && c->x86_mask <= 4 && - c->x86_model <= 3) - /* - * Remember we have B step Pentia with bugs - */ - smp_b_stepping = 1; - - /* - * Certain Athlons might work (for various values of 'work') in SMP - * but they are not certified as MP capable. - */ - if ((c->x86_vendor == X86_VENDOR_AMD) && (c->x86 == 6)) { - - if (num_possible_cpus() == 1) - goto valid_k7; - - /* Athlon 660/661 is valid. */ - if ((c->x86_model == 6) && ((c->x86_mask == 0) || - (c->x86_mask == 1))) - goto valid_k7; - - /* Duron 670 is valid */ - if ((c->x86_model == 7) && (c->x86_mask == 0)) - goto valid_k7; - - /* - * Athlon 662, Duron 671, and Athlon >model 7 have capability - * bit. It's worth noting that the A5 stepping (662) of some - * Athlon XP's have the MP bit set. - * See http://www.heise.de/newsticker/data/jow-18.10.01-000 for - * more. - */ - if (((c->x86_model == 6) && (c->x86_mask >= 2)) || - ((c->x86_model == 7) && (c->x86_mask >= 1)) || - (c->x86_model > 7)) - if (cpu_has_mp) - goto valid_k7; - - /* If we get here, not a certified SMP capable AMD system. */ - unsafe_smp = 1; - } - -valid_k7: - ; -} - -static void __cpuinit smp_checks(void) -{ - if (smp_b_stepping) - printk(KERN_WARNING "WARNING: SMP operation may be unreliable" - "with B stepping processors.\n"); - - /* - * Don't taint if we are running SMP kernel on a single non-MP - * approved Athlon - */ - if (unsafe_smp && num_online_cpus() > 1) { - printk(KERN_INFO "WARNING: This combination of AMD" - "processors is not suitable for SMP.\n"); - add_taint(TAINT_UNSAFE_SMP); - } -} - /* * The bootstrap kernel entry code has set these up. Save them for * a given CPU @@ -423,7 +347,6 @@ void __cpuinit smp_store_cpu_info(int id) c->cpu_index = id; if (id != 0) identify_secondary_cpu(c); - smp_apply_quirks(c); } @@ -583,7 +506,7 @@ wakeup_secondary_cpu_via_nmi(int logical_apicid, unsigned long start_eip) /* Target chip */ /* Boot on the stack */ /* Kick the second */ - apic_icr_write(APIC_DM_NMI | APIC_DEST_LOGICAL, logical_apicid); + apic_icr_write(APIC_DM_NMI | apic->dest_logical, logical_apicid); pr_debug("Waiting for send to finish...\n"); send_status = safe_apic_wait_icr_idle(); @@ -614,12 +537,6 @@ wakeup_secondary_cpu_via_init(int phys_apicid, unsigned long start_eip) unsigned long send_status, accept_status = 0; int maxlvt, num_starts, j; - if (get_uv_system_type() == UV_NON_UNIQUE_APIC) { - send_status = uv_wakeup_secondary(phys_apicid, start_eip); - atomic_set(&init_deasserted, 1); - return send_status; - } - maxlvt = lapic_get_maxlvt(); /* @@ -745,78 +662,23 @@ static void __cpuinit do_fork_idle(struct work_struct *work) complete(&c_idle->done); } -#ifdef CONFIG_X86_64 - -/* __ref because it's safe to call free_bootmem when after_bootmem == 0. */ -static void __ref free_bootmem_pda(struct x8664_pda *oldpda) -{ - if (!after_bootmem) - free_bootmem((unsigned long)oldpda, sizeof(*oldpda)); -} - -/* - * Allocate node local memory for the AP pda. - * - * Must be called after the _cpu_pda pointer table is initialized. - */ -int __cpuinit get_local_pda(int cpu) -{ - struct x8664_pda *oldpda, *newpda; - unsigned long size = sizeof(struct x8664_pda); - int node = cpu_to_node(cpu); - - if (cpu_pda(cpu) && !cpu_pda(cpu)->in_bootmem) - return 0; - - oldpda = cpu_pda(cpu); - newpda = kmalloc_node(size, GFP_ATOMIC, node); - if (!newpda) { - printk(KERN_ERR "Could not allocate node local PDA " - "for CPU %d on node %d\n", cpu, node); - - if (oldpda) - return 0; /* have a usable pda */ - else - return -1; - } - - if (oldpda) { - memcpy(newpda, oldpda, size); - free_bootmem_pda(oldpda); - } - - newpda->in_bootmem = 0; - cpu_pda(cpu) = newpda; - return 0; -} -#endif /* CONFIG_X86_64 */ - -static int __cpuinit do_boot_cpu(int apicid, int cpu) /* * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad * (ie clustered apic addressing mode), this is a LOGICAL apic ID. - * Returns zero if CPU booted OK, else error code from wakeup_secondary_cpu. + * Returns zero if CPU booted OK, else error code from + * ->wakeup_secondary_cpu. */ +static int __cpuinit do_boot_cpu(int apicid, int cpu) { unsigned long boot_error = 0; - int timeout; unsigned long start_ip; - unsigned short nmi_high = 0, nmi_low = 0; + int timeout; struct create_idle c_idle = { - .cpu = cpu, - .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done), + .cpu = cpu, + .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done), }; - INIT_WORK(&c_idle.work, do_fork_idle); -#ifdef CONFIG_X86_64 - /* Allocate node local memory for AP pdas */ - if (cpu > 0) { - boot_error = get_local_pda(cpu); - if (boot_error) - goto restore_state; - /* if can't get pda memory, can't start cpu */ - } -#endif + INIT_WORK(&c_idle.work, do_fork_idle); alternatives_smp_switch(1); @@ -847,14 +709,16 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu) set_idle_for_cpu(cpu, c_idle.idle); do_rest: -#ifdef CONFIG_X86_32 per_cpu(current_task, cpu) = c_idle.idle; - init_gdt(cpu); +#ifdef CONFIG_X86_32 /* Stack for startup_32 can be just as for start_secondary onwards */ irq_ctx_init(cpu); #else - cpu_pda(cpu)->pcurrent = c_idle.idle; clear_tsk_thread_flag(c_idle.idle, TIF_FORK); + initial_gs = per_cpu_offset(cpu); + per_cpu(kernel_stack, cpu) = + (unsigned long)task_stack_page(c_idle.idle) - + KERNEL_STACK_OFFSET + THREAD_SIZE; #endif early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu); initial_code = (unsigned long)start_secondary; @@ -878,8 +742,6 @@ do_rest: pr_debug("Setting warm reset code and vector.\n"); - store_NMI_vector(&nmi_high, &nmi_low); - smpboot_setup_warm_reset_vector(start_ip); /* * Be paranoid about clearing APIC errors. @@ -891,9 +753,13 @@ do_rest: } /* - * Starting actual IPI sequence... + * Kick the secondary CPU. Use the method in the APIC driver + * if it's defined - or use an INIT boot APIC message otherwise: */ - boot_error = wakeup_secondary_cpu(apicid, start_ip); + if (apic->wakeup_secondary_cpu) + boot_error = apic->wakeup_secondary_cpu(apicid, start_ip); + else + boot_error = wakeup_secondary_cpu_via_init(apicid, start_ip); if (!boot_error) { /* @@ -927,13 +793,11 @@ do_rest: else /* trampoline code not run */ printk(KERN_ERR "Not responding.\n"); - if (get_uv_system_type() != UV_NON_UNIQUE_APIC) - inquire_remote_apic(apicid); + if (apic->inquire_remote_apic) + apic->inquire_remote_apic(apicid); } } -#ifdef CONFIG_X86_64 -restore_state: -#endif + if (boot_error) { /* Try to put things back the way they were before ... */ numa_remove_cpu(cpu); /* was set by numa_add_cpu */ @@ -961,7 +825,7 @@ restore_state: int __cpuinit native_cpu_up(unsigned int cpu) { - int apicid = cpu_present_to_apicid(cpu); + int apicid = apic->cpu_present_to_apicid(cpu); unsigned long flags; int err; @@ -1054,14 +918,14 @@ static int __init smp_sanity_check(unsigned max_cpus) { preempt_disable(); -#if defined(CONFIG_X86_PC) && defined(CONFIG_X86_32) +#if !defined(CONFIG_X86_BIGSMP) && defined(CONFIG_X86_32) if (def_to_bigsmp && nr_cpu_ids > 8) { unsigned int cpu; unsigned nr; printk(KERN_WARNING "More than 8 CPUs detected - skipping them.\n" - "Use CONFIG_X86_GENERICARCH and CONFIG_X86_BIGSMP.\n"); + "Use CONFIG_X86_BIGSMP.\n"); nr = 0; for_each_present_cpu(cpu) { @@ -1107,7 +971,7 @@ static int __init smp_sanity_check(unsigned max_cpus) * Should not be necessary because the MP table should list the boot * CPU too, but we do it for the sake of robustness anyway. */ - if (!check_phys_apicid_present(boot_cpu_physical_apicid)) { + if (!apic->check_phys_apicid_present(boot_cpu_physical_apicid)) { printk(KERN_NOTICE "weird, boot CPU (#%d) not listed by the BIOS.\n", boot_cpu_physical_apicid); @@ -1125,6 +989,7 @@ static int __init smp_sanity_check(unsigned max_cpus) printk(KERN_ERR "... forcing use of dummy APIC emulation." "(tell your hw vendor)\n"); smpboot_clear_io_apic(); + arch_disable_smp_support(); return -1; } @@ -1181,9 +1046,9 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) current_thread_info()->cpu = 0; /* needed? */ set_cpu_sibling_map(0); -#ifdef CONFIG_X86_64 enable_IR_x2apic(); - setup_apic_routing(); +#ifdef CONFIG_X86_64 + default_setup_apic_routing(); #endif if (smp_sanity_check(max_cpus) < 0) { @@ -1207,18 +1072,18 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) */ setup_local_APIC(); -#ifdef CONFIG_X86_64 /* * Enable IO APIC before setting up error vector */ if (!skip_ioapic_setup && nr_ioapics) enable_IO_APIC(); -#endif + end_local_APIC_setup(); map_cpu_to_logical_apicid(); - setup_portio_remap(); + if (apic->setup_portio_remap) + apic->setup_portio_remap(); smpboot_setup_io_apic(); /* @@ -1240,10 +1105,7 @@ out: void __init native_smp_prepare_boot_cpu(void) { int me = smp_processor_id(); -#ifdef CONFIG_X86_32 - init_gdt(me); -#endif - switch_to_new_gdt(); + switch_to_new_gdt(me); /* already set me in cpu_online_mask in boot_cpu_init() */ cpumask_set_cpu(me, cpu_callout_mask); per_cpu(cpu_state, me) = CPU_ONLINE; @@ -1254,7 +1116,6 @@ void __init native_smp_cpus_done(unsigned int max_cpus) pr_debug("Boot done.\n"); impress_friends(); - smp_checks(); #ifdef CONFIG_X86_IO_APIC setup_ioapic_dest(); #endif diff --git a/arch/x86/kernel/smpcommon.c b/arch/x86/kernel/smpcommon.c deleted file mode 100644 index 397e309839d..00000000000 --- a/arch/x86/kernel/smpcommon.c +++ /dev/null @@ -1,30 +0,0 @@ -/* - * SMP stuff which is common to all sub-architectures. - */ -#include <linux/module.h> -#include <asm/smp.h> - -#ifdef CONFIG_X86_32 -DEFINE_PER_CPU(unsigned long, this_cpu_off); -EXPORT_PER_CPU_SYMBOL(this_cpu_off); - -/* - * Initialize the CPU's GDT. This is either the boot CPU doing itself - * (still using the master per-cpu area), or a CPU doing it for a - * secondary which will soon come up. - */ -__cpuinit void init_gdt(int cpu) -{ - struct desc_struct gdt; - - pack_descriptor(&gdt, __per_cpu_offset[cpu], 0xFFFFF, - 0x2 | DESCTYPE_S, 0x8); - gdt.s = 1; - - write_gdt_entry(get_cpu_gdt_table(cpu), - GDT_ENTRY_PERCPU, &gdt, DESCTYPE_S); - - per_cpu(this_cpu_off, cpu) = __per_cpu_offset[cpu]; - per_cpu(cpu_number, cpu) = cpu; -} -#endif diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c index 10786af9554..f7bddc2e37d 100644 --- a/arch/x86/kernel/stacktrace.c +++ b/arch/x86/kernel/stacktrace.c @@ -1,7 +1,7 @@ /* * Stack trace management functions * - * Copyright (C) 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * Copyright (C) 2006-2009 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> */ #include <linux/sched.h> #include <linux/stacktrace.h> diff --git a/arch/x86/kernel/summit_32.c b/arch/x86/kernel/summit_32.c deleted file mode 100644 index 7b987852e87..00000000000 --- a/arch/x86/kernel/summit_32.c +++ /dev/null @@ -1,188 +0,0 @@ -/* - * IBM Summit-Specific Code - * - * Written By: Matthew Dobson, IBM Corporation - * - * Copyright (c) 2003 IBM Corp. - * - * All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or (at - * your option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or - * NON INFRINGEMENT. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to <colpatch@us.ibm.com> - * - */ - -#include <linux/mm.h> -#include <linux/init.h> -#include <asm/io.h> -#include <asm/bios_ebda.h> -#include <asm/summit/mpparse.h> - -static struct rio_table_hdr *rio_table_hdr __initdata; -static struct scal_detail *scal_devs[MAX_NUMNODES] __initdata; -static struct rio_detail *rio_devs[MAX_NUMNODES*4] __initdata; - -#ifndef CONFIG_X86_NUMAQ -static int mp_bus_id_to_node[MAX_MP_BUSSES] __initdata; -#endif - -static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus) -{ - int twister = 0, node = 0; - int i, bus, num_buses; - - for (i = 0; i < rio_table_hdr->num_rio_dev; i++) { - if (rio_devs[i]->node_id == rio_devs[wpeg_num]->owner_id) { - twister = rio_devs[i]->owner_id; - break; - } - } - if (i == rio_table_hdr->num_rio_dev) { - printk(KERN_ERR "%s: Couldn't find owner Cyclone for Winnipeg!\n", __func__); - return last_bus; - } - - for (i = 0; i < rio_table_hdr->num_scal_dev; i++) { - if (scal_devs[i]->node_id == twister) { - node = scal_devs[i]->node_id; - break; - } - } - if (i == rio_table_hdr->num_scal_dev) { - printk(KERN_ERR "%s: Couldn't find owner Twister for Cyclone!\n", __func__); - return last_bus; - } - - switch (rio_devs[wpeg_num]->type) { - case CompatWPEG: - /* - * The Compatibility Winnipeg controls the 2 legacy buses, - * the 66MHz PCI bus [2 slots] and the 2 "extra" buses in case - * a PCI-PCI bridge card is used in either slot: total 5 buses. - */ - num_buses = 5; - break; - case AltWPEG: - /* - * The Alternate Winnipeg controls the 2 133MHz buses [1 slot - * each], their 2 "extra" buses, the 100MHz bus [2 slots] and - * the "extra" buses for each of those slots: total 7 buses. - */ - num_buses = 7; - break; - case LookOutAWPEG: - case LookOutBWPEG: - /* - * A Lookout Winnipeg controls 3 100MHz buses [2 slots each] - * & the "extra" buses for each of those slots: total 9 buses. - */ - num_buses = 9; - break; - default: - printk(KERN_INFO "%s: Unsupported Winnipeg type!\n", __func__); - return last_bus; - } - - for (bus = last_bus; bus < last_bus + num_buses; bus++) - mp_bus_id_to_node[bus] = node; - return bus; -} - -static int __init build_detail_arrays(void) -{ - unsigned long ptr; - int i, scal_detail_size, rio_detail_size; - - if (rio_table_hdr->num_scal_dev > MAX_NUMNODES) { - printk(KERN_WARNING "%s: MAX_NUMNODES too low! Defined as %d, but system has %d nodes.\n", __func__, MAX_NUMNODES, rio_table_hdr->num_scal_dev); - return 0; - } - - switch (rio_table_hdr->version) { - default: - printk(KERN_WARNING "%s: Invalid Rio Grande Table Version: %d\n", __func__, rio_table_hdr->version); - return 0; - case 2: - scal_detail_size = 11; - rio_detail_size = 13; - break; - case 3: - scal_detail_size = 12; - rio_detail_size = 15; - break; - } - - ptr = (unsigned long)rio_table_hdr + 3; - for (i = 0; i < rio_table_hdr->num_scal_dev; i++, ptr += scal_detail_size) - scal_devs[i] = (struct scal_detail *)ptr; - - for (i = 0; i < rio_table_hdr->num_rio_dev; i++, ptr += rio_detail_size) - rio_devs[i] = (struct rio_detail *)ptr; - - return 1; -} - -void __init setup_summit(void) -{ - unsigned long ptr; - unsigned short offset; - int i, next_wpeg, next_bus = 0; - - /* The pointer to the EBDA is stored in the word @ phys 0x40E(40:0E) */ - ptr = get_bios_ebda(); - ptr = (unsigned long)phys_to_virt(ptr); - - rio_table_hdr = NULL; - offset = 0x180; - while (offset) { - /* The block id is stored in the 2nd word */ - if (*((unsigned short *)(ptr + offset + 2)) == 0x4752) { - /* set the pointer past the offset & block id */ - rio_table_hdr = (struct rio_table_hdr *)(ptr + offset + 4); - break; - } - /* The next offset is stored in the 1st word. 0 means no more */ - offset = *((unsigned short *)(ptr + offset)); - } - if (!rio_table_hdr) { - printk(KERN_ERR "%s: Unable to locate Rio Grande Table in EBDA - bailing!\n", __func__); - return; - } - - if (!build_detail_arrays()) - return; - - /* The first Winnipeg we're looking for has an index of 0 */ - next_wpeg = 0; - do { - for (i = 0; i < rio_table_hdr->num_rio_dev; i++) { - if (is_WPEG(rio_devs[i]) && rio_devs[i]->WP_index == next_wpeg) { - /* It's the Winnipeg we're looking for! */ - next_bus = setup_pci_node_map_for_wpeg(i, next_bus); - next_wpeg++; - break; - } - } - /* - * If we go through all Rio devices and don't find one with - * the next index, it means we've found all the Winnipegs, - * and thus all the PCI buses. - */ - if (i == rio_table_hdr->num_rio_dev) - next_wpeg = 0; - } while (next_wpeg != 0); -} diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index e2e86a08f31..3bdb64829b8 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -1,7 +1,7 @@ ENTRY(sys_call_table) .long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */ .long sys_exit - .long sys_fork + .long ptregs_fork .long sys_read .long sys_write .long sys_open /* 5 */ @@ -10,7 +10,7 @@ ENTRY(sys_call_table) .long sys_creat .long sys_link .long sys_unlink /* 10 */ - .long sys_execve + .long ptregs_execve .long sys_chdir .long sys_time .long sys_mknod @@ -109,17 +109,17 @@ ENTRY(sys_call_table) .long sys_newlstat .long sys_newfstat .long sys_uname - .long sys_iopl /* 110 */ + .long ptregs_iopl /* 110 */ .long sys_vhangup .long sys_ni_syscall /* old "idle" system call */ - .long sys_vm86old + .long ptregs_vm86old .long sys_wait4 .long sys_swapoff /* 115 */ .long sys_sysinfo .long sys_ipc .long sys_fsync - .long sys_sigreturn - .long sys_clone /* 120 */ + .long ptregs_sigreturn + .long ptregs_clone /* 120 */ .long sys_setdomainname .long sys_newuname .long sys_modify_ldt @@ -165,14 +165,14 @@ ENTRY(sys_call_table) .long sys_mremap .long sys_setresuid16 .long sys_getresuid16 /* 165 */ - .long sys_vm86 + .long ptregs_vm86 .long sys_ni_syscall /* Old sys_query_module */ .long sys_poll .long sys_nfsservctl .long sys_setresgid16 /* 170 */ .long sys_getresgid16 .long sys_prctl - .long sys_rt_sigreturn + .long ptregs_rt_sigreturn .long sys_rt_sigaction .long sys_rt_sigprocmask /* 175 */ .long sys_rt_sigpending @@ -185,11 +185,11 @@ ENTRY(sys_call_table) .long sys_getcwd .long sys_capget .long sys_capset /* 185 */ - .long sys_sigaltstack + .long ptregs_sigaltstack .long sys_sendfile .long sys_ni_syscall /* reserved for streams1 */ .long sys_ni_syscall /* reserved for streams2 */ - .long sys_vfork /* 190 */ + .long ptregs_vfork /* 190 */ .long sys_getrlimit .long sys_mmap2 .long sys_truncate64 diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c index 3985cac0ed4..5c5d87f0b2e 100644 --- a/arch/x86/kernel/time_32.c +++ b/arch/x86/kernel/time_32.c @@ -33,12 +33,12 @@ #include <linux/time.h> #include <linux/mca.h> -#include <asm/arch_hooks.h> +#include <asm/setup.h> #include <asm/hpet.h> #include <asm/time.h> #include <asm/timer.h> -#include "do_timer.h" +#include <asm/do_timer.h> int timer_ack; @@ -118,7 +118,7 @@ void __init hpet_time_init(void) { if (!hpet_enable()) setup_pit_timer(); - time_init_hook(); + x86_quirk_time_init(); } /* @@ -131,7 +131,7 @@ void __init hpet_time_init(void) */ void __init time_init(void) { - pre_time_init_hook(); + x86_quirk_pre_time_init(); tsc_init(); late_time_init = choose_time_init(); } diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c deleted file mode 100644 index ce505464224..00000000000 --- a/arch/x86/kernel/tlb_32.c +++ /dev/null @@ -1,256 +0,0 @@ -#include <linux/spinlock.h> -#include <linux/cpu.h> -#include <linux/interrupt.h> - -#include <asm/tlbflush.h> - -DEFINE_PER_CPU(struct tlb_state, cpu_tlbstate) - ____cacheline_aligned = { &init_mm, 0, }; - -/* must come after the send_IPI functions above for inlining */ -#include <mach_ipi.h> - -/* - * Smarter SMP flushing macros. - * c/o Linus Torvalds. - * - * These mean you can really definitely utterly forget about - * writing to user space from interrupts. (Its not allowed anyway). - * - * Optimizations Manfred Spraul <manfred@colorfullife.com> - */ - -static cpumask_t flush_cpumask; -static struct mm_struct *flush_mm; -static unsigned long flush_va; -static DEFINE_SPINLOCK(tlbstate_lock); - -/* - * We cannot call mmdrop() because we are in interrupt context, - * instead update mm->cpu_vm_mask. - * - * We need to reload %cr3 since the page tables may be going - * away from under us.. - */ -void leave_mm(int cpu) -{ - BUG_ON(x86_read_percpu(cpu_tlbstate.state) == TLBSTATE_OK); - cpu_clear(cpu, x86_read_percpu(cpu_tlbstate.active_mm)->cpu_vm_mask); - load_cr3(swapper_pg_dir); -} -EXPORT_SYMBOL_GPL(leave_mm); - -/* - * - * The flush IPI assumes that a thread switch happens in this order: - * [cpu0: the cpu that switches] - * 1) switch_mm() either 1a) or 1b) - * 1a) thread switch to a different mm - * 1a1) cpu_clear(cpu, old_mm->cpu_vm_mask); - * Stop ipi delivery for the old mm. This is not synchronized with - * the other cpus, but smp_invalidate_interrupt ignore flush ipis - * for the wrong mm, and in the worst case we perform a superfluous - * tlb flush. - * 1a2) set cpu_tlbstate to TLBSTATE_OK - * Now the smp_invalidate_interrupt won't call leave_mm if cpu0 - * was in lazy tlb mode. - * 1a3) update cpu_tlbstate[].active_mm - * Now cpu0 accepts tlb flushes for the new mm. - * 1a4) cpu_set(cpu, new_mm->cpu_vm_mask); - * Now the other cpus will send tlb flush ipis. - * 1a4) change cr3. - * 1b) thread switch without mm change - * cpu_tlbstate[].active_mm is correct, cpu0 already handles - * flush ipis. - * 1b1) set cpu_tlbstate to TLBSTATE_OK - * 1b2) test_and_set the cpu bit in cpu_vm_mask. - * Atomically set the bit [other cpus will start sending flush ipis], - * and test the bit. - * 1b3) if the bit was 0: leave_mm was called, flush the tlb. - * 2) switch %%esp, ie current - * - * The interrupt must handle 2 special cases: - * - cr3 is changed before %%esp, ie. it cannot use current->{active_,}mm. - * - the cpu performs speculative tlb reads, i.e. even if the cpu only - * runs in kernel space, the cpu could load tlb entries for user space - * pages. - * - * The good news is that cpu_tlbstate is local to each cpu, no - * write/read ordering problems. - */ - -/* - * TLB flush IPI: - * - * 1) Flush the tlb entries if the cpu uses the mm that's being flushed. - * 2) Leave the mm if we are in the lazy tlb mode. - */ - -void smp_invalidate_interrupt(struct pt_regs *regs) -{ - unsigned long cpu; - - cpu = get_cpu(); - - if (!cpu_isset(cpu, flush_cpumask)) - goto out; - /* - * This was a BUG() but until someone can quote me the - * line from the intel manual that guarantees an IPI to - * multiple CPUs is retried _only_ on the erroring CPUs - * its staying as a return - * - * BUG(); - */ - - if (flush_mm == x86_read_percpu(cpu_tlbstate.active_mm)) { - if (x86_read_percpu(cpu_tlbstate.state) == TLBSTATE_OK) { - if (flush_va == TLB_FLUSH_ALL) - local_flush_tlb(); - else - __flush_tlb_one(flush_va); - } else - leave_mm(cpu); - } - ack_APIC_irq(); - smp_mb__before_clear_bit(); - cpu_clear(cpu, flush_cpumask); - smp_mb__after_clear_bit(); -out: - put_cpu_no_resched(); - inc_irq_stat(irq_tlb_count); -} - -void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm, - unsigned long va) -{ - cpumask_t cpumask = *cpumaskp; - - /* - * A couple of (to be removed) sanity checks: - * - * - current CPU must not be in mask - * - mask must exist :) - */ - BUG_ON(cpus_empty(cpumask)); - BUG_ON(cpu_isset(smp_processor_id(), cpumask)); - BUG_ON(!mm); - -#ifdef CONFIG_HOTPLUG_CPU - /* If a CPU which we ran on has gone down, OK. */ - cpus_and(cpumask, cpumask, cpu_online_map); - if (unlikely(cpus_empty(cpumask))) - return; -#endif - - /* - * i'm not happy about this global shared spinlock in the - * MM hot path, but we'll see how contended it is. - * AK: x86-64 has a faster method that could be ported. - */ - spin_lock(&tlbstate_lock); - - flush_mm = mm; - flush_va = va; - cpus_or(flush_cpumask, cpumask, flush_cpumask); - - /* - * Make the above memory operations globally visible before - * sending the IPI. - */ - smp_mb(); - /* - * We have to send the IPI only to - * CPUs affected. - */ - send_IPI_mask(&cpumask, INVALIDATE_TLB_VECTOR); - - while (!cpus_empty(flush_cpumask)) - /* nothing. lockup detection does not belong here */ - cpu_relax(); - - flush_mm = NULL; - flush_va = 0; - spin_unlock(&tlbstate_lock); -} - -void flush_tlb_current_task(void) -{ - struct mm_struct *mm = current->mm; - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - local_flush_tlb(); - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL); - preempt_enable(); -} - -void flush_tlb_mm(struct mm_struct *mm) -{ - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - if (current->active_mm == mm) { - if (current->mm) - local_flush_tlb(); - else - leave_mm(smp_processor_id()); - } - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL); - - preempt_enable(); -} - -void flush_tlb_page(struct vm_area_struct *vma, unsigned long va) -{ - struct mm_struct *mm = vma->vm_mm; - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - if (current->active_mm == mm) { - if (current->mm) - __flush_tlb_one(va); - else - leave_mm(smp_processor_id()); - } - - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, va); - - preempt_enable(); -} -EXPORT_SYMBOL(flush_tlb_page); - -static void do_flush_tlb_all(void *info) -{ - unsigned long cpu = smp_processor_id(); - - __flush_tlb_all(); - if (x86_read_percpu(cpu_tlbstate.state) == TLBSTATE_LAZY) - leave_mm(cpu); -} - -void flush_tlb_all(void) -{ - on_each_cpu(do_flush_tlb_all, NULL, 1); -} - -void reset_lazy_tlbstate(void) -{ - int cpu = raw_smp_processor_id(); - - per_cpu(cpu_tlbstate, cpu).state = 0; - per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm; -} - diff --git a/arch/x86/kernel/tlb_64.c b/arch/x86/kernel/tlb_64.c deleted file mode 100644 index f8be6f1d2e4..00000000000 --- a/arch/x86/kernel/tlb_64.c +++ /dev/null @@ -1,284 +0,0 @@ -#include <linux/init.h> - -#include <linux/mm.h> -#include <linux/delay.h> -#include <linux/spinlock.h> -#include <linux/smp.h> -#include <linux/kernel_stat.h> -#include <linux/mc146818rtc.h> -#include <linux/interrupt.h> - -#include <asm/mtrr.h> -#include <asm/pgalloc.h> -#include <asm/tlbflush.h> -#include <asm/mmu_context.h> -#include <asm/proto.h> -#include <asm/apicdef.h> -#include <asm/idle.h> -#include <asm/uv/uv_hub.h> -#include <asm/uv/uv_bau.h> - -#include <mach_ipi.h> -/* - * Smarter SMP flushing macros. - * c/o Linus Torvalds. - * - * These mean you can really definitely utterly forget about - * writing to user space from interrupts. (Its not allowed anyway). - * - * Optimizations Manfred Spraul <manfred@colorfullife.com> - * - * More scalable flush, from Andi Kleen - * - * To avoid global state use 8 different call vectors. - * Each CPU uses a specific vector to trigger flushes on other - * CPUs. Depending on the received vector the target CPUs look into - * the right per cpu variable for the flush data. - * - * With more than 8 CPUs they are hashed to the 8 available - * vectors. The limited global vector space forces us to this right now. - * In future when interrupts are split into per CPU domains this could be - * fixed, at the cost of triggering multiple IPIs in some cases. - */ - -union smp_flush_state { - struct { - cpumask_t flush_cpumask; - struct mm_struct *flush_mm; - unsigned long flush_va; - spinlock_t tlbstate_lock; - }; - char pad[SMP_CACHE_BYTES]; -} ____cacheline_aligned; - -/* State is put into the per CPU data section, but padded - to a full cache line because other CPUs can access it and we don't - want false sharing in the per cpu data segment. */ -static DEFINE_PER_CPU(union smp_flush_state, flush_state); - -/* - * We cannot call mmdrop() because we are in interrupt context, - * instead update mm->cpu_vm_mask. - */ -void leave_mm(int cpu) -{ - if (read_pda(mmu_state) == TLBSTATE_OK) - BUG(); - cpu_clear(cpu, read_pda(active_mm)->cpu_vm_mask); - load_cr3(swapper_pg_dir); -} -EXPORT_SYMBOL_GPL(leave_mm); - -/* - * - * The flush IPI assumes that a thread switch happens in this order: - * [cpu0: the cpu that switches] - * 1) switch_mm() either 1a) or 1b) - * 1a) thread switch to a different mm - * 1a1) cpu_clear(cpu, old_mm->cpu_vm_mask); - * Stop ipi delivery for the old mm. This is not synchronized with - * the other cpus, but smp_invalidate_interrupt ignore flush ipis - * for the wrong mm, and in the worst case we perform a superfluous - * tlb flush. - * 1a2) set cpu mmu_state to TLBSTATE_OK - * Now the smp_invalidate_interrupt won't call leave_mm if cpu0 - * was in lazy tlb mode. - * 1a3) update cpu active_mm - * Now cpu0 accepts tlb flushes for the new mm. - * 1a4) cpu_set(cpu, new_mm->cpu_vm_mask); - * Now the other cpus will send tlb flush ipis. - * 1a4) change cr3. - * 1b) thread switch without mm change - * cpu active_mm is correct, cpu0 already handles - * flush ipis. - * 1b1) set cpu mmu_state to TLBSTATE_OK - * 1b2) test_and_set the cpu bit in cpu_vm_mask. - * Atomically set the bit [other cpus will start sending flush ipis], - * and test the bit. - * 1b3) if the bit was 0: leave_mm was called, flush the tlb. - * 2) switch %%esp, ie current - * - * The interrupt must handle 2 special cases: - * - cr3 is changed before %%esp, ie. it cannot use current->{active_,}mm. - * - the cpu performs speculative tlb reads, i.e. even if the cpu only - * runs in kernel space, the cpu could load tlb entries for user space - * pages. - * - * The good news is that cpu mmu_state is local to each cpu, no - * write/read ordering problems. - */ - -/* - * TLB flush IPI: - * - * 1) Flush the tlb entries if the cpu uses the mm that's being flushed. - * 2) Leave the mm if we are in the lazy tlb mode. - * - * Interrupts are disabled. - */ - -asmlinkage void smp_invalidate_interrupt(struct pt_regs *regs) -{ - int cpu; - int sender; - union smp_flush_state *f; - - cpu = smp_processor_id(); - /* - * orig_rax contains the negated interrupt vector. - * Use that to determine where the sender put the data. - */ - sender = ~regs->orig_ax - INVALIDATE_TLB_VECTOR_START; - f = &per_cpu(flush_state, sender); - - if (!cpu_isset(cpu, f->flush_cpumask)) - goto out; - /* - * This was a BUG() but until someone can quote me the - * line from the intel manual that guarantees an IPI to - * multiple CPUs is retried _only_ on the erroring CPUs - * its staying as a return - * - * BUG(); - */ - - if (f->flush_mm == read_pda(active_mm)) { - if (read_pda(mmu_state) == TLBSTATE_OK) { - if (f->flush_va == TLB_FLUSH_ALL) - local_flush_tlb(); - else - __flush_tlb_one(f->flush_va); - } else - leave_mm(cpu); - } -out: - ack_APIC_irq(); - cpu_clear(cpu, f->flush_cpumask); - inc_irq_stat(irq_tlb_count); -} - -void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm, - unsigned long va) -{ - int sender; - union smp_flush_state *f; - cpumask_t cpumask = *cpumaskp; - - if (is_uv_system() && uv_flush_tlb_others(&cpumask, mm, va)) - return; - - /* Caller has disabled preemption */ - sender = smp_processor_id() % NUM_INVALIDATE_TLB_VECTORS; - f = &per_cpu(flush_state, sender); - - /* - * Could avoid this lock when - * num_online_cpus() <= NUM_INVALIDATE_TLB_VECTORS, but it is - * probably not worth checking this for a cache-hot lock. - */ - spin_lock(&f->tlbstate_lock); - - f->flush_mm = mm; - f->flush_va = va; - cpus_or(f->flush_cpumask, cpumask, f->flush_cpumask); - - /* - * Make the above memory operations globally visible before - * sending the IPI. - */ - smp_mb(); - /* - * We have to send the IPI only to - * CPUs affected. - */ - send_IPI_mask(&cpumask, INVALIDATE_TLB_VECTOR_START + sender); - - while (!cpus_empty(f->flush_cpumask)) - cpu_relax(); - - f->flush_mm = NULL; - f->flush_va = 0; - spin_unlock(&f->tlbstate_lock); -} - -static int __cpuinit init_smp_flush(void) -{ - int i; - - for_each_possible_cpu(i) - spin_lock_init(&per_cpu(flush_state, i).tlbstate_lock); - - return 0; -} -core_initcall(init_smp_flush); - -void flush_tlb_current_task(void) -{ - struct mm_struct *mm = current->mm; - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - local_flush_tlb(); - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL); - preempt_enable(); -} - -void flush_tlb_mm(struct mm_struct *mm) -{ - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - if (current->active_mm == mm) { - if (current->mm) - local_flush_tlb(); - else - leave_mm(smp_processor_id()); - } - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL); - - preempt_enable(); -} - -void flush_tlb_page(struct vm_area_struct *vma, unsigned long va) -{ - struct mm_struct *mm = vma->vm_mm; - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - if (current->active_mm == mm) { - if (current->mm) - __flush_tlb_one(va); - else - leave_mm(smp_processor_id()); - } - - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, va); - - preempt_enable(); -} - -static void do_flush_tlb_all(void *info) -{ - unsigned long cpu = smp_processor_id(); - - __flush_tlb_all(); - if (read_pda(mmu_state) == TLBSTATE_LAZY) - leave_mm(cpu); -} - -void flush_tlb_all(void) -{ - on_each_cpu(do_flush_tlb_all, NULL, 1); -} diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 6812b829ed8..79c07324728 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -11,16 +11,15 @@ #include <linux/kernel.h> #include <asm/mmu_context.h> +#include <asm/uv/uv.h> #include <asm/uv/uv_mmrs.h> #include <asm/uv/uv_hub.h> #include <asm/uv/uv_bau.h> -#include <asm/genapic.h> +#include <asm/apic.h> #include <asm/idle.h> #include <asm/tsc.h> #include <asm/irq_vectors.h> -#include <mach_apic.h> - static struct bau_control **uv_bau_table_bases __read_mostly; static int uv_bau_retry_limit __read_mostly; @@ -210,14 +209,15 @@ static int uv_wait_completion(struct bau_desc *bau_desc, * * Send a broadcast and wait for a broadcast message to complete. * - * The cpumaskp mask contains the cpus the broadcast was sent to. + * The flush_mask contains the cpus the broadcast was sent to. * - * Returns 1 if all remote flushing was done. The mask is zeroed. - * Returns 0 if some remote flushing remains to be done. The mask is left - * unchanged. + * Returns NULL if all remote flushing was done. The mask is zeroed. + * Returns @flush_mask if some remote flushing remains to be done. The + * mask will have some bits still set. */ -int uv_flush_send_and_wait(int cpu, int this_blade, struct bau_desc *bau_desc, - cpumask_t *cpumaskp) +const struct cpumask *uv_flush_send_and_wait(int cpu, int this_blade, + struct bau_desc *bau_desc, + struct cpumask *flush_mask) { int completion_status = 0; int right_shift; @@ -257,66 +257,74 @@ int uv_flush_send_and_wait(int cpu, int this_blade, struct bau_desc *bau_desc, * the cpu's, all of which are still in the mask. */ __get_cpu_var(ptcstats).ptc_i++; - return 0; + return flush_mask; } /* * Success, so clear the remote cpu's from the mask so we don't * use the IPI method of shootdown on them. */ - for_each_cpu_mask(bit, *cpumaskp) { + for_each_cpu(bit, flush_mask) { blade = uv_cpu_to_blade_id(bit); if (blade == this_blade) continue; - cpu_clear(bit, *cpumaskp); + cpumask_clear_cpu(bit, flush_mask); } - if (!cpus_empty(*cpumaskp)) - return 0; - return 1; + if (!cpumask_empty(flush_mask)) + return flush_mask; + return NULL; } /** * uv_flush_tlb_others - globally purge translation cache of a virtual * address or all TLB's - * @cpumaskp: mask of all cpu's in which the address is to be removed + * @cpumask: mask of all cpu's in which the address is to be removed * @mm: mm_struct containing virtual address range * @va: virtual address to be removed (or TLB_FLUSH_ALL for all TLB's on cpu) + * @cpu: the current cpu * * This is the entry point for initiating any UV global TLB shootdown. * * Purges the translation caches of all specified processors of the given * virtual address, or purges all TLB's on specified processors. * - * The caller has derived the cpumaskp from the mm_struct and has subtracted - * the local cpu from the mask. This function is called only if there - * are bits set in the mask. (e.g. flush_tlb_page()) + * The caller has derived the cpumask from the mm_struct. This function + * is called only if there are bits set in the mask. (e.g. flush_tlb_page()) * - * The cpumaskp is converted into a nodemask of the nodes containing + * The cpumask is converted into a nodemask of the nodes containing * the cpus. * - * Returns 1 if all remote flushing was done. - * Returns 0 if some remote flushing remains to be done. + * Note that this function should be called with preemption disabled. + * + * Returns NULL if all remote flushing was done. + * Returns pointer to cpumask if some remote flushing remains to be + * done. The returned pointer is valid till preemption is re-enabled. */ -int uv_flush_tlb_others(cpumask_t *cpumaskp, struct mm_struct *mm, - unsigned long va) +const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, + struct mm_struct *mm, + unsigned long va, unsigned int cpu) { + static DEFINE_PER_CPU(cpumask_t, flush_tlb_mask); + struct cpumask *flush_mask = &__get_cpu_var(flush_tlb_mask); int i; int bit; int blade; - int cpu; + int uv_cpu; int this_blade; int locals = 0; struct bau_desc *bau_desc; - cpu = uv_blade_processor_id(); + cpumask_andnot(flush_mask, cpumask, cpumask_of(cpu)); + + uv_cpu = uv_blade_processor_id(); this_blade = uv_numa_blade_id(); bau_desc = __get_cpu_var(bau_control).descriptor_base; - bau_desc += UV_ITEMS_PER_DESCRIPTOR * cpu; + bau_desc += UV_ITEMS_PER_DESCRIPTOR * uv_cpu; bau_nodes_clear(&bau_desc->distribution, UV_DISTRIBUTION_SIZE); i = 0; - for_each_cpu_mask(bit, *cpumaskp) { + for_each_cpu(bit, flush_mask) { blade = uv_cpu_to_blade_id(bit); BUG_ON(blade > (UV_DISTRIBUTION_SIZE - 1)); if (blade == this_blade) { @@ -331,17 +339,17 @@ int uv_flush_tlb_others(cpumask_t *cpumaskp, struct mm_struct *mm, * no off_node flushing; return status for local node */ if (locals) - return 0; + return flush_mask; else - return 1; + return NULL; } __get_cpu_var(ptcstats).requestor++; __get_cpu_var(ptcstats).ntargeted += i; bau_desc->payload.address = va; - bau_desc->payload.sending_cpu = smp_processor_id(); + bau_desc->payload.sending_cpu = cpu; - return uv_flush_send_and_wait(cpu, this_blade, bau_desc, cpumaskp); + return uv_flush_send_and_wait(uv_cpu, this_blade, bau_desc, flush_mask); } /* @@ -742,7 +750,7 @@ static int __init uv_bau_init(void) int node; int nblades; int last_blade; - int cur_cpu = 0; + int cur_cpu; if (!is_uv_system()) return 0; @@ -752,6 +760,7 @@ static int __init uv_bau_init(void) uv_mmask = (1UL << uv_hub_info->n_val) - 1; nblades = 0; last_blade = -1; + cur_cpu = 0; for_each_online_node(node) { blade = uv_node_to_blade_id(node); if (blade == last_blade) diff --git a/arch/x86/kernel/trampoline_32.S b/arch/x86/kernel/trampoline_32.S index d8ccc3c6552..66d874e5404 100644 --- a/arch/x86/kernel/trampoline_32.S +++ b/arch/x86/kernel/trampoline_32.S @@ -29,7 +29,7 @@ #include <linux/linkage.h> #include <asm/segment.h> -#include <asm/page.h> +#include <asm/page_types.h> /* We can free up trampoline after bootup if cpu hotplug is not supported. */ #ifndef CONFIG_HOTPLUG_CPU diff --git a/arch/x86/kernel/trampoline_64.S b/arch/x86/kernel/trampoline_64.S index 894293c598d..cddfb8d386b 100644 --- a/arch/x86/kernel/trampoline_64.S +++ b/arch/x86/kernel/trampoline_64.S @@ -25,10 +25,11 @@ */ #include <linux/linkage.h> -#include <asm/pgtable.h> -#include <asm/page.h> +#include <asm/pgtable_types.h> +#include <asm/page_types.h> #include <asm/msr.h> #include <asm/segment.h> +#include <asm/processor-flags.h> .section .rodata, "a", @progbits @@ -37,7 +38,7 @@ ENTRY(trampoline_data) r_base = . cli # We should be safe anyway - wbinvd + wbinvd mov %cs, %ax # Code and data in the same place mov %ax, %ds mov %ax, %es @@ -73,9 +74,8 @@ r_base = . lidtl tidt - r_base # load idt with 0, 0 lgdtl tgdt - r_base # load gdt with whatever is appropriate - xor %ax, %ax - inc %ax # protected mode (PE) bit - lmsw %ax # into protected mode + mov $X86_CR0_PE, %ax # protected mode (PE) bit + lmsw %ax # into protected mode # flush prefetch and jump to startup_32 ljmpl *(startup_32_vector - r_base) @@ -86,9 +86,8 @@ startup_32: movl $__KERNEL_DS, %eax # Initialize the %ds segment register movl %eax, %ds - xorl %eax, %eax - btsl $5, %eax # Enable PAE mode - movl %eax, %cr4 + movl $X86_CR4_PAE, %eax + movl %eax, %cr4 # Enable PAE mode # Setup trampoline 4 level pagetables leal (trampoline_level4_pgt - r_base)(%esi), %eax @@ -99,9 +98,9 @@ startup_32: xorl %edx, %edx wrmsr - xorl %eax, %eax - btsl $31, %eax # Enable paging and in turn activate Long Mode - btsl $0, %eax # Enable protected mode + # Enable paging and in turn activate Long Mode + # Enable protected mode + movl $(X86_CR0_PG | X86_CR0_PE), %eax movl %eax, %cr0 /* diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index a9e7548e179..a1d288327ff 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -54,15 +54,14 @@ #include <asm/desc.h> #include <asm/i387.h> -#include <mach_traps.h> +#include <asm/mach_traps.h> #ifdef CONFIG_X86_64 #include <asm/pgalloc.h> #include <asm/proto.h> -#include <asm/pda.h> #else #include <asm/processor-flags.h> -#include <asm/arch_hooks.h> +#include <asm/setup.h> #include <asm/traps.h> #include "cpu/mcheck/mce.h" @@ -119,47 +118,6 @@ die_if_kernel(const char *str, struct pt_regs *regs, long err) if (!user_mode_vm(regs)) die(str, regs, err); } - -/* - * Perform the lazy TSS's I/O bitmap copy. If the TSS has an - * invalid offset set (the LAZY one) and the faulting thread has - * a valid I/O bitmap pointer, we copy the I/O bitmap in the TSS, - * we set the offset field correctly and return 1. - */ -static int lazy_iobitmap_copy(void) -{ - struct thread_struct *thread; - struct tss_struct *tss; - int cpu; - - cpu = get_cpu(); - tss = &per_cpu(init_tss, cpu); - thread = ¤t->thread; - - if (tss->x86_tss.io_bitmap_base == INVALID_IO_BITMAP_OFFSET_LAZY && - thread->io_bitmap_ptr) { - memcpy(tss->io_bitmap, thread->io_bitmap_ptr, - thread->io_bitmap_max); - /* - * If the previously set map was extending to higher ports - * than the current one, pad extra space with 0xff (no access). - */ - if (thread->io_bitmap_max < tss->io_bitmap_max) { - memset((char *) tss->io_bitmap + - thread->io_bitmap_max, 0xff, - tss->io_bitmap_max - thread->io_bitmap_max); - } - tss->io_bitmap_max = thread->io_bitmap_max; - tss->x86_tss.io_bitmap_base = IO_BITMAP_OFFSET; - tss->io_bitmap_owner = thread; - put_cpu(); - - return 1; - } - put_cpu(); - - return 0; -} #endif static void __kprobes @@ -310,11 +268,6 @@ do_general_protection(struct pt_regs *regs, long error_code) conditional_sti(regs); #ifdef CONFIG_X86_32 - if (lazy_iobitmap_copy()) { - /* restart the faulting instruction */ - return; - } - if (regs->flags & X86_VM_MASK) goto gp_in_vm86; #endif @@ -914,19 +867,20 @@ void math_emulate(struct math_emu_info *info) } #endif /* CONFIG_MATH_EMULATION */ -dotraplinkage void __kprobes do_device_not_available(struct pt_regs regs) +dotraplinkage void __kprobes +do_device_not_available(struct pt_regs *regs, long error_code) { #ifdef CONFIG_X86_32 if (read_cr0() & X86_CR0_EM) { struct math_emu_info info = { }; - conditional_sti(®s); + conditional_sti(regs); - info.regs = ®s; + info.regs = regs; math_emulate(&info); } else { math_state_restore(); /* interrupts still off */ - conditional_sti(®s); + conditional_sti(regs); } #else math_state_restore(); @@ -942,7 +896,7 @@ dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code) info.si_signo = SIGILL; info.si_errno = 0; info.si_code = ILL_BADSTK; - info.si_addr = 0; + info.si_addr = NULL; if (notify_die(DIE_TRAP, "iret exception", regs, error_code, 32, SIGILL) == NOTIFY_STOP) return; @@ -1026,6 +980,6 @@ void __init trap_init(void) cpu_init(); #ifdef CONFIG_X86_32 - trap_init_hook(); + x86_quirk_trap_init(); #endif } diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 599e5816863..462b9ba67e9 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -273,30 +273,43 @@ static unsigned long pit_calibrate_tsc(u32 latch, unsigned long ms, int loopmin) * use the TSC value at the transitions to calculate a pretty * good value for the TSC frequencty. */ -static inline int pit_expect_msb(unsigned char val) +static inline int pit_expect_msb(unsigned char val, u64 *tscp, unsigned long *deltap) { - int count = 0; + int count; + u64 tsc = 0; for (count = 0; count < 50000; count++) { /* Ignore LSB */ inb(0x42); if (inb(0x42) != val) break; + tsc = get_cycles(); } - return count > 50; + *deltap = get_cycles() - tsc; + *tscp = tsc; + + /* + * We require _some_ success, but the quality control + * will be based on the error terms on the TSC values. + */ + return count > 5; } /* - * How many MSB values do we want to see? We aim for a - * 15ms calibration, which assuming a 2us counter read - * error should give us roughly 150 ppm precision for - * the calibration. + * How many MSB values do we want to see? We aim for + * a maximum error rate of 500ppm (in practice the + * real error is much smaller), but refuse to spend + * more than 25ms on it. */ -#define QUICK_PIT_MS 15 -#define QUICK_PIT_ITERATIONS (QUICK_PIT_MS * PIT_TICK_RATE / 1000 / 256) +#define MAX_QUICK_PIT_MS 25 +#define MAX_QUICK_PIT_ITERATIONS (MAX_QUICK_PIT_MS * PIT_TICK_RATE / 1000 / 256) static unsigned long quick_pit_calibrate(void) { + int i; + u64 tsc, delta; + unsigned long d1, d2; + /* Set the Gate high, disable speaker */ outb((inb(0x61) & ~0x02) | 0x01, 0x61); @@ -315,45 +328,52 @@ static unsigned long quick_pit_calibrate(void) outb(0xff, 0x42); outb(0xff, 0x42); - if (pit_expect_msb(0xff)) { - int i; - u64 t1, t2, delta; - unsigned char expect = 0xfe; - - t1 = get_cycles(); - for (i = 0; i < QUICK_PIT_ITERATIONS; i++, expect--) { - if (!pit_expect_msb(expect)) - goto failed; + /* + * The PIT starts counting at the next edge, so we + * need to delay for a microsecond. The easiest way + * to do that is to just read back the 16-bit counter + * once from the PIT. + */ + inb(0x42); + inb(0x42); + + if (pit_expect_msb(0xff, &tsc, &d1)) { + for (i = 1; i <= MAX_QUICK_PIT_ITERATIONS; i++) { + if (!pit_expect_msb(0xff-i, &delta, &d2)) + break; + + /* + * Iterate until the error is less than 500 ppm + */ + delta -= tsc; + if (d1+d2 < delta >> 11) + goto success; } - t2 = get_cycles(); - - /* - * Make sure we can rely on the second TSC timestamp: - */ - if (!pit_expect_msb(expect)) - goto failed; - - /* - * Ok, if we get here, then we've seen the - * MSB of the PIT decrement QUICK_PIT_ITERATIONS - * times, and each MSB had many hits, so we never - * had any sudden jumps. - * - * As a result, we can depend on there not being - * any odd delays anywhere, and the TSC reads are - * reliable. - * - * kHz = ticks / time-in-seconds / 1000; - * kHz = (t2 - t1) / (QPI * 256 / PIT_TICK_RATE) / 1000 - * kHz = ((t2 - t1) * PIT_TICK_RATE) / (QPI * 256 * 1000) - */ - delta = (t2 - t1)*PIT_TICK_RATE; - do_div(delta, QUICK_PIT_ITERATIONS*256*1000); - printk("Fast TSC calibration using PIT\n"); - return delta; } -failed: + printk("Fast TSC calibration failed\n"); return 0; + +success: + /* + * Ok, if we get here, then we've seen the + * MSB of the PIT decrement 'i' times, and the + * error has shrunk to less than 500 ppm. + * + * As a result, we can depend on there not being + * any odd delays anywhere, and the TSC reads are + * reliable (within the error). We also adjust the + * delta to the middle of the error bars, just + * because it looks nicer. + * + * kHz = ticks / time-in-seconds / 1000; + * kHz = (t2 - t1) / (I * 256 / PIT_TICK_RATE) / 1000 + * kHz = ((t2 - t1) * PIT_TICK_RATE) / (I * 256 * 1000) + */ + delta += (long)(d2 - d1)/2; + delta *= PIT_TICK_RATE; + do_div(delta, i*256*1000); + printk("Fast TSC calibration using PIT\n"); + return delta; } /** @@ -773,7 +793,7 @@ __cpuinit int unsynchronized_tsc(void) if (!cpu_has_tsc || tsc_unstable) return 1; -#ifdef CONFIG_X86_SMP +#ifdef CONFIG_SMP if (apic_is_clustered_box()) return 1; #endif diff --git a/arch/x86/kernel/uv_time.c b/arch/x86/kernel/uv_time.c new file mode 100644 index 00000000000..2ffb6c53326 --- /dev/null +++ b/arch/x86/kernel/uv_time.c @@ -0,0 +1,393 @@ +/* + * SGI RTC clock/timer routines. + * + * 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) 2009 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) Dimitri Sivanich + */ +#include <linux/clockchips.h> + +#include <asm/uv/uv_mmrs.h> +#include <asm/uv/uv_hub.h> +#include <asm/uv/bios.h> +#include <asm/uv/uv.h> +#include <asm/apic.h> +#include <asm/cpu.h> + +#define RTC_NAME "sgi_rtc" + +static cycle_t uv_read_rtc(void); +static int uv_rtc_next_event(unsigned long, struct clock_event_device *); +static void uv_rtc_timer_setup(enum clock_event_mode, + struct clock_event_device *); + +static struct clocksource clocksource_uv = { + .name = RTC_NAME, + .rating = 400, + .read = uv_read_rtc, + .mask = (cycle_t)UVH_RTC_REAL_TIME_CLOCK_MASK, + .shift = 10, + .flags = CLOCK_SOURCE_IS_CONTINUOUS, +}; + +static struct clock_event_device clock_event_device_uv = { + .name = RTC_NAME, + .features = CLOCK_EVT_FEAT_ONESHOT, + .shift = 20, + .rating = 400, + .irq = -1, + .set_next_event = uv_rtc_next_event, + .set_mode = uv_rtc_timer_setup, + .event_handler = NULL, +}; + +static DEFINE_PER_CPU(struct clock_event_device, cpu_ced); + +/* There is one of these allocated per node */ +struct uv_rtc_timer_head { + spinlock_t lock; + /* next cpu waiting for timer, local node relative: */ + int next_cpu; + /* number of cpus on this node: */ + int ncpus; + struct { + int lcpu; /* systemwide logical cpu number */ + u64 expires; /* next timer expiration for this cpu */ + } cpu[1]; +}; + +/* + * Access to uv_rtc_timer_head via blade id. + */ +static struct uv_rtc_timer_head **blade_info __read_mostly; + +static int uv_rtc_enable; + +/* + * Hardware interface routines + */ + +/* Send IPIs to another node */ +static void uv_rtc_send_IPI(int cpu) +{ + unsigned long apicid, val; + int pnode; + + apicid = cpu_physical_id(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); + + uv_write_global_mmr64(pnode, UVH_IPI_INT, val); +} + +/* Check for an RTC interrupt pending */ +static int uv_intr_pending(int pnode) +{ + return uv_read_global_mmr64(pnode, UVH_EVENT_OCCURRED0) & + UVH_EVENT_OCCURRED0_RTC1_MASK; +} + +/* Setup interrupt and return non-zero if early expiration occurred. */ +static int uv_setup_intr(int cpu, u64 expires) +{ + u64 val; + int pnode = uv_cpu_to_pnode(cpu); + + uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG, + UVH_RTC1_INT_CONFIG_M_MASK); + uv_write_global_mmr64(pnode, UVH_INT_CMPB, -1L); + + uv_write_global_mmr64(pnode, UVH_EVENT_OCCURRED0_ALIAS, + UVH_EVENT_OCCURRED0_RTC1_MASK); + + val = (GENERIC_INTERRUPT_VECTOR << UVH_RTC1_INT_CONFIG_VECTOR_SHFT) | + ((u64)cpu_physical_id(cpu) << UVH_RTC1_INT_CONFIG_APIC_ID_SHFT); + + /* Set configuration */ + uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG, val); + /* Initialize comparator value */ + uv_write_global_mmr64(pnode, UVH_INT_CMPB, expires); + + return (expires < uv_read_rtc() && !uv_intr_pending(pnode)); +} + +/* + * Per-cpu timer tracking routines + */ + +static __init void uv_rtc_deallocate_timers(void) +{ + int bid; + + for_each_possible_blade(bid) { + kfree(blade_info[bid]); + } + kfree(blade_info); +} + +/* Allocate per-node list of cpu timer expiration times. */ +static __init int uv_rtc_allocate_timers(void) +{ + int cpu; + + blade_info = kmalloc(uv_possible_blades * sizeof(void *), GFP_KERNEL); + if (!blade_info) + return -ENOMEM; + memset(blade_info, 0, uv_possible_blades * sizeof(void *)); + + for_each_present_cpu(cpu) { + int nid = cpu_to_node(cpu); + int bid = uv_cpu_to_blade_id(cpu); + int bcpu = uv_cpu_hub_info(cpu)->blade_processor_id; + struct uv_rtc_timer_head *head = blade_info[bid]; + + if (!head) { + head = kmalloc_node(sizeof(struct uv_rtc_timer_head) + + (uv_blade_nr_possible_cpus(bid) * + 2 * sizeof(u64)), + GFP_KERNEL, nid); + if (!head) { + uv_rtc_deallocate_timers(); + return -ENOMEM; + } + spin_lock_init(&head->lock); + head->ncpus = uv_blade_nr_possible_cpus(bid); + head->next_cpu = -1; + blade_info[bid] = head; + } + + head->cpu[bcpu].lcpu = cpu; + head->cpu[bcpu].expires = ULLONG_MAX; + } + + return 0; +} + +/* Find and set the next expiring timer. */ +static void uv_rtc_find_next_timer(struct uv_rtc_timer_head *head, int pnode) +{ + u64 lowest = ULLONG_MAX; + int c, bcpu = -1; + + head->next_cpu = -1; + for (c = 0; c < head->ncpus; c++) { + u64 exp = head->cpu[c].expires; + if (exp < lowest) { + bcpu = c; + lowest = exp; + } + } + if (bcpu >= 0) { + head->next_cpu = bcpu; + c = head->cpu[bcpu].lcpu; + if (uv_setup_intr(c, lowest)) + /* If we didn't set it up in time, trigger */ + uv_rtc_send_IPI(c); + } else { + uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG, + UVH_RTC1_INT_CONFIG_M_MASK); + } +} + +/* + * Set expiration time for current cpu. + * + * Returns 1 if we missed the expiration time. + */ +static int uv_rtc_set_timer(int cpu, u64 expires) +{ + int pnode = uv_cpu_to_pnode(cpu); + int bid = uv_cpu_to_blade_id(cpu); + struct uv_rtc_timer_head *head = blade_info[bid]; + int bcpu = uv_cpu_hub_info(cpu)->blade_processor_id; + u64 *t = &head->cpu[bcpu].expires; + unsigned long flags; + int next_cpu; + + spin_lock_irqsave(&head->lock, flags); + + 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) { + head->next_cpu = bcpu; + if (uv_setup_intr(cpu, expires)) { + *t = ULLONG_MAX; + uv_rtc_find_next_timer(head, pnode); + spin_unlock_irqrestore(&head->lock, flags); + return 1; + } + } + + spin_unlock_irqrestore(&head->lock, flags); + return 0; +} + +/* + * Unset expiration time for current cpu. + * + * Returns 1 if this timer was pending. + */ +static int uv_rtc_unset_timer(int cpu) +{ + int pnode = uv_cpu_to_pnode(cpu); + int bid = uv_cpu_to_blade_id(cpu); + struct uv_rtc_timer_head *head = blade_info[bid]; + int bcpu = uv_cpu_hub_info(cpu)->blade_processor_id; + u64 *t = &head->cpu[bcpu].expires; + unsigned long flags; + int rc = 0; + + spin_lock_irqsave(&head->lock, flags); + + if (head->next_cpu == bcpu && uv_read_rtc() >= *t) + rc = 1; + + *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); + + return rc; +} + + +/* + * Kernel interface routines. + */ + +/* + * Read the RTC. + */ +static cycle_t uv_read_rtc(void) +{ + return (cycle_t)uv_read_local_mmr(UVH_RTC); +} + +/* + * Program the next event, relative to now + */ +static int uv_rtc_next_event(unsigned long delta, + struct clock_event_device *ced) +{ + int ced_cpu = cpumask_first(ced->cpumask); + + return uv_rtc_set_timer(ced_cpu, delta + uv_read_rtc()); +} + +/* + * Setup the RTC timer in oneshot mode + */ +static void uv_rtc_timer_setup(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + int ced_cpu = cpumask_first(evt->cpumask); + + switch (mode) { + case CLOCK_EVT_MODE_PERIODIC: + case CLOCK_EVT_MODE_ONESHOT: + case CLOCK_EVT_MODE_RESUME: + /* Nothing to do here yet */ + break; + case CLOCK_EVT_MODE_UNUSED: + case CLOCK_EVT_MODE_SHUTDOWN: + uv_rtc_unset_timer(ced_cpu); + break; + } +} + +static void uv_rtc_interrupt(void) +{ + struct clock_event_device *ced = &__get_cpu_var(cpu_ced); + int cpu = smp_processor_id(); + + if (!ced || !ced->event_handler) + return; + + if (uv_rtc_unset_timer(cpu) != 1) + return; + + ced->event_handler(ced); +} + +static int __init uv_enable_rtc(char *str) +{ + uv_rtc_enable = 1; + + return 1; +} +__setup("uvrtc", uv_enable_rtc); + +static __init void uv_rtc_register_clockevents(struct work_struct *dummy) +{ + struct clock_event_device *ced = &__get_cpu_var(cpu_ced); + + *ced = clock_event_device_uv; + ced->cpumask = cpumask_of(smp_processor_id()); + clockevents_register_device(ced); +} + +static __init int uv_rtc_setup_clock(void) +{ + int rc; + + if (!uv_rtc_enable || !is_uv_system() || generic_interrupt_extension) + return -ENODEV; + + generic_interrupt_extension = uv_rtc_interrupt; + + clocksource_uv.mult = clocksource_hz2mult(sn_rtc_cycles_per_second, + clocksource_uv.shift); + + rc = clocksource_register(&clocksource_uv); + if (rc) { + generic_interrupt_extension = NULL; + return rc; + } + + /* Setup and register clockevents */ + rc = uv_rtc_allocate_timers(); + if (rc) { + clocksource_unregister(&clocksource_uv); + generic_interrupt_extension = NULL; + return rc; + } + + clock_event_device_uv.mult = div_sc(sn_rtc_cycles_per_second, + NSEC_PER_SEC, clock_event_device_uv.shift); + + clock_event_device_uv.min_delta_ns = NSEC_PER_SEC / + sn_rtc_cycles_per_second; + + clock_event_device_uv.max_delta_ns = clocksource_uv.mask * + (NSEC_PER_SEC / sn_rtc_cycles_per_second); + + rc = schedule_on_each_cpu(uv_rtc_register_clockevents); + if (rc) { + clocksource_unregister(&clocksource_uv); + generic_interrupt_extension = NULL; + uv_rtc_deallocate_timers(); + } + + 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 d801d06af06..31ffc24eec4 100644 --- a/arch/x86/kernel/visws_quirks.c +++ b/arch/x86/kernel/visws_quirks.c @@ -24,18 +24,14 @@ #include <asm/visws/cobalt.h> #include <asm/visws/piix4.h> -#include <asm/arch_hooks.h> #include <asm/io_apic.h> #include <asm/fixmap.h> #include <asm/reboot.h> #include <asm/setup.h> +#include <asm/apic.h> #include <asm/e820.h> #include <asm/io.h> -#include <mach_ipi.h> - -#include "mach_apic.h" - #include <linux/kernel_stat.h> #include <asm/i8259.h> @@ -49,8 +45,6 @@ extern int no_broadcast; -#include <asm/apic.h> - char visws_board_type = -1; char visws_board_rev = -1; @@ -200,7 +194,7 @@ static void __init MP_processor_info(struct mpc_cpu *m) return; } - apic_cpus = apicid_to_cpu_present(m->apicid); + apic_cpus = apic->apicid_to_cpu_present(m->apicid); physids_or(phys_cpu_present_map, phys_cpu_present_map, apic_cpus); /* * Validate version @@ -584,7 +578,7 @@ static struct irq_chip piix4_virtual_irq_type = { static irqreturn_t piix4_master_intr(int irq, void *dev_id) { int realirq; - irq_desc_t *desc; + struct irq_desc *desc; unsigned long flags; spin_lock_irqsave(&i8259A_lock, flags); diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c index 4eeb5cf9720..d7ac84e7fc1 100644 --- a/arch/x86/kernel/vm86_32.c +++ b/arch/x86/kernel/vm86_32.c @@ -158,7 +158,7 @@ struct pt_regs *save_v86_state(struct kernel_vm86_regs *regs) ret = KVM86->regs32; ret->fs = current->thread.saved_fs; - loadsegment(gs, current->thread.saved_gs); + set_user_gs(ret, current->thread.saved_gs); return ret; } @@ -197,9 +197,9 @@ 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); -asmlinkage int sys_vm86old(struct pt_regs regs) +int sys_vm86old(struct pt_regs *regs) { - struct vm86_struct __user *v86 = (struct vm86_struct __user *)regs.bx; + 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 @@ -218,7 +218,7 @@ asmlinkage int sys_vm86old(struct pt_regs regs) if (tmp) goto out; memset(&info.vm86plus, 0, (int)&info.regs32 - (int)&info.vm86plus); - info.regs32 = ®s; + info.regs32 = regs; tsk->thread.vm86_info = v86; do_sys_vm86(&info, tsk); ret = 0; /* we never return here */ @@ -227,7 +227,7 @@ out: } -asmlinkage int sys_vm86(struct pt_regs regs) +int sys_vm86(struct pt_regs *regs) { struct kernel_vm86_struct info; /* declare this _on top_, * this avoids wasting of stack space. @@ -239,12 +239,12 @@ asmlinkage int sys_vm86(struct pt_regs regs) struct vm86plus_struct __user *v86; tsk = current; - switch (regs.bx) { + switch (regs->bx) { 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(regs->bx, (int)regs->cx); goto out; case VM86_PLUS_INSTALL_CHECK: /* @@ -261,14 +261,14 @@ asmlinkage 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 *)regs->cx; tmp = copy_vm86_regs_from_user(&info.regs, &v86->regs, offsetof(struct kernel_vm86_struct, regs32) - sizeof(info.regs)); ret = -EFAULT; if (tmp) goto out; - info.regs32 = ®s; + info.regs32 = regs; info.vm86plus.is_vm86pus = 1; tsk->thread.vm86_info = (struct vm86_struct __user *)v86; do_sys_vm86(&info, tsk); @@ -323,7 +323,7 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk info->regs32->ax = 0; tsk->thread.saved_sp0 = tsk->thread.sp0; tsk->thread.saved_fs = info->regs32->fs; - savesegment(gs, tsk->thread.saved_gs); + tsk->thread.saved_gs = get_user_gs(info->regs32); tss = &per_cpu(init_tss, get_cpu()); tsk->thread.sp0 = (unsigned long) &info->VM86_TSS_ESP0; diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index bef58b4982d..2cc4a90e2cb 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -680,10 +680,11 @@ static inline int __init activate_vmi(void) para_fill(pv_mmu_ops.write_cr2, SetCR2); para_fill(pv_mmu_ops.write_cr3, SetCR3); para_fill(pv_cpu_ops.write_cr4, SetCR4); - para_fill(pv_irq_ops.save_fl, GetInterruptMask); - para_fill(pv_irq_ops.restore_fl, SetInterruptMask); - para_fill(pv_irq_ops.irq_disable, DisableInterrupts); - para_fill(pv_irq_ops.irq_enable, EnableInterrupts); + + para_fill(pv_irq_ops.save_fl.func, GetInterruptMask); + para_fill(pv_irq_ops.restore_fl.func, SetInterruptMask); + para_fill(pv_irq_ops.irq_disable.func, DisableInterrupts); + para_fill(pv_irq_ops.irq_enable.func, EnableInterrupts); para_fill(pv_cpu_ops.wbinvd, WBINVD); para_fill(pv_cpu_ops.read_tsc, RDTSC); @@ -797,8 +798,8 @@ static inline int __init activate_vmi(void) #endif #ifdef CONFIG_X86_LOCAL_APIC - para_fill(apic_ops->read, APICRead); - para_fill(apic_ops->write, APICWrite); + para_fill(apic->read, APICRead); + para_fill(apic->write, APICWrite); #endif /* diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c index e5b088fffa4..33a788d5879 100644 --- a/arch/x86/kernel/vmiclock_32.c +++ b/arch/x86/kernel/vmiclock_32.c @@ -28,7 +28,6 @@ #include <asm/vmi.h> #include <asm/vmi_time.h> -#include <asm/arch_hooks.h> #include <asm/apicdef.h> #include <asm/apic.h> #include <asm/timer.h> @@ -256,7 +255,7 @@ void __devinit vmi_time_bsp_init(void) */ clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL); local_irq_disable(); -#ifdef CONFIG_X86_SMP +#ifdef CONFIG_SMP /* * XXX handle_percpu_irq only defined for SMP; we need to switch over * to using it, since this is a local interrupt, which each CPU must @@ -288,8 +287,7 @@ static struct clocksource clocksource_vmi; static cycle_t read_real_cycles(void) { cycle_t ret = (cycle_t)vmi_timer_ops.get_cycle_counter(VMI_CYCLES_REAL); - return ret >= clocksource_vmi.cycle_last ? - ret : clocksource_vmi.cycle_last; + return max(ret, clocksource_vmi.cycle_last); } static struct clocksource clocksource_vmi = { diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S index 82c67559dde..62ad500d55f 100644 --- a/arch/x86/kernel/vmlinux_32.lds.S +++ b/arch/x86/kernel/vmlinux_32.lds.S @@ -12,7 +12,7 @@ #include <asm-generic/vmlinux.lds.h> #include <asm/thread_info.h> -#include <asm/page.h> +#include <asm/page_types.h> #include <asm/cache.h> #include <asm/boot.h> @@ -178,14 +178,7 @@ SECTIONS __initramfs_end = .; } #endif - . = ALIGN(PAGE_SIZE); - .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { - __per_cpu_start = .; - *(.data.percpu.page_aligned) - *(.data.percpu) - *(.data.percpu.shared_aligned) - __per_cpu_end = .; - } + PERCPU(PAGE_SIZE) . = ALIGN(PAGE_SIZE); /* freed after init ends here */ @@ -196,15 +189,24 @@ SECTIONS *(.bss) . = ALIGN(4); __bss_stop = .; - _end = . ; - /* This is where the kernel creates the early boot page tables */ + } + + .brk : AT(ADDR(.brk) - LOAD_OFFSET) { . = ALIGN(PAGE_SIZE); - pg0 = . ; + __brk_base = . ; + . += 64 * 1024 ; /* 64k alignment slop space */ + *(.brk_reservation) /* areas brk users have reserved */ + __brk_limit = . ; + } + + .end : AT(ADDR(.end) - LOAD_OFFSET) { + _end = . ; } /* Sections to be discarded */ /DISCARD/ : { *(.exitcall.exit) + *(.discard) } STABS_DEBUG @@ -212,6 +214,12 @@ SECTIONS DWARF_DEBUG } +/* + * Build-time check on the image size: + */ +ASSERT((_end - LOAD_OFFSET <= KERNEL_IMAGE_SIZE), + "kernel image bigger than KERNEL_IMAGE_SIZE") + #ifdef CONFIG_KEXEC /* Link time checks */ #include <asm/kexec.h> diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index 1a614c0e6be..c8742507b03 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -5,7 +5,8 @@ #define LOAD_OFFSET __START_KERNEL_map #include <asm-generic/vmlinux.lds.h> -#include <asm/page.h> +#include <asm/asm-offsets.h> +#include <asm/page_types.h> #undef i386 /* in case the preprocessor is a 32bit one */ @@ -13,20 +14,23 @@ OUTPUT_FORMAT("elf64-x86-64", "elf64-x86-64", "elf64-x86-64") OUTPUT_ARCH(i386:x86-64) ENTRY(phys_startup_64) jiffies_64 = jiffies; -_proxy_pda = 1; PHDRS { text PT_LOAD FLAGS(5); /* R_E */ data PT_LOAD FLAGS(7); /* RWE */ user PT_LOAD FLAGS(7); /* RWE */ data.init PT_LOAD FLAGS(7); /* RWE */ +#ifdef CONFIG_SMP + percpu PT_LOAD FLAGS(7); /* RWE */ +#endif + data.init2 PT_LOAD FLAGS(7); /* RWE */ note PT_NOTE FLAGS(0); /* ___ */ } SECTIONS { . = __START_KERNEL; phys_startup_64 = startup_64 - LOAD_OFFSET; - _text = .; /* Text and read-only data */ .text : AT(ADDR(.text) - LOAD_OFFSET) { + _text = .; /* Text and read-only data */ /* First the code that has to be first for bootstrapping */ *(.text.head) _stext = .; @@ -57,13 +61,13 @@ SECTIONS .data : AT(ADDR(.data) - LOAD_OFFSET) { DATA_DATA CONSTRUCTORS + _edata = .; /* End of data section */ } :data - _edata = .; /* End of data section */ - . = ALIGN(PAGE_SIZE); - . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) { + . = ALIGN(PAGE_SIZE); + . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); *(.data.cacheline_aligned) } . = ALIGN(CONFIG_X86_INTERNODE_CACHE_BYTES); @@ -121,29 +125,29 @@ SECTIONS #undef VVIRT_OFFSET #undef VVIRT - . = ALIGN(THREAD_SIZE); /* init_task */ .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) { + . = ALIGN(THREAD_SIZE); /* init_task */ *(.data.init_task) }:data.init - . = ALIGN(PAGE_SIZE); .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) { + . = ALIGN(PAGE_SIZE); *(.data.page_aligned) } - /* might get freed after init */ - . = ALIGN(PAGE_SIZE); - __smp_alt_begin = .; - __smp_locks = .; .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) { + /* might get freed after init */ + . = ALIGN(PAGE_SIZE); + __smp_alt_begin = .; + __smp_locks = .; *(.smp_locks) + __smp_locks_end = .; + . = ALIGN(PAGE_SIZE); + __smp_alt_end = .; } - __smp_locks_end = .; - . = ALIGN(PAGE_SIZE); - __smp_alt_end = .; . = ALIGN(PAGE_SIZE); /* Init code and data */ - __init_begin = .; + __init_begin = .; /* paired with __init_end */ .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) { _sinittext = .; INIT_TEXT @@ -155,40 +159,42 @@ SECTIONS __initdata_end = .; } - . = ALIGN(16); - __setup_start = .; - .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { *(.init.setup) } - __setup_end = .; - __initcall_start = .; + .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { + . = ALIGN(16); + __setup_start = .; + *(.init.setup) + __setup_end = .; + } .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { + __initcall_start = .; INITCALLS + __initcall_end = .; } - __initcall_end = .; - __con_initcall_start = .; .con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) { + __con_initcall_start = .; *(.con_initcall.init) + __con_initcall_end = .; } - __con_initcall_end = .; - __x86_cpu_dev_start = .; .x86_cpu_dev.init : AT(ADDR(.x86_cpu_dev.init) - LOAD_OFFSET) { + __x86_cpu_dev_start = .; *(.x86_cpu_dev.init) + __x86_cpu_dev_end = .; } - __x86_cpu_dev_end = .; SECURITY_INIT . = ALIGN(8); .parainstructions : AT(ADDR(.parainstructions) - LOAD_OFFSET) { - __parainstructions = .; + __parainstructions = .; *(.parainstructions) - __parainstructions_end = .; + __parainstructions_end = .; } - . = ALIGN(8); - __alt_instructions = .; .altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) { + . = ALIGN(8); + __alt_instructions = .; *(.altinstructions) + __alt_instructions_end = .; } - __alt_instructions_end = .; .altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) { *(.altinstr_replacement) } @@ -203,28 +209,53 @@ SECTIONS #ifdef CONFIG_BLK_DEV_INITRD . = ALIGN(PAGE_SIZE); - __initramfs_start = .; - .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) } - __initramfs_end = .; + .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { + __initramfs_start = .; + *(.init.ramfs) + __initramfs_end = .; + } #endif +#ifdef CONFIG_SMP + /* + * percpu offsets are zero-based on SMP. PERCPU_VADDR() changes the + * output PHDR, so the next output section - __data_nosave - should + * start another section data.init2. Also, pda should be at the head of + * percpu area. Preallocate it and define the percpu offset symbol + * so that it can be accessed as a percpu variable. + */ + . = ALIGN(PAGE_SIZE); + PERCPU_VADDR(0, :percpu) +#else PERCPU(PAGE_SIZE) +#endif . = ALIGN(PAGE_SIZE); __init_end = .; - . = ALIGN(PAGE_SIZE); - __nosave_begin = .; - .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { *(.data.nosave) } - . = ALIGN(PAGE_SIZE); - __nosave_end = .; + .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { + . = ALIGN(PAGE_SIZE); + __nosave_begin = .; + *(.data.nosave) + . = ALIGN(PAGE_SIZE); + __nosave_end = .; + } :data.init2 /* use another section data.init2, see PERCPU_VADDR() above */ - __bss_start = .; /* BSS */ .bss : AT(ADDR(.bss) - LOAD_OFFSET) { + . = ALIGN(PAGE_SIZE); + __bss_start = .; /* BSS */ *(.bss.page_aligned) *(.bss) - } - __bss_stop = .; + __bss_stop = .; + } + + .brk : AT(ADDR(.brk) - LOAD_OFFSET) { + . = ALIGN(PAGE_SIZE); + __brk_base = . ; + . += 64 * 1024 ; /* 64k alignment slop space */ + *(.brk_reservation) /* areas brk users have reserved */ + __brk_limit = . ; + } _end = . ; @@ -232,6 +263,7 @@ SECTIONS /DISCARD/ : { *(.exitcall.exit) *(.eh_frame) + *(.discard) } STABS_DEBUG @@ -239,8 +271,28 @@ SECTIONS DWARF_DEBUG } + /* + * 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 +INIT_PER_CPU(gdt_page); +INIT_PER_CPU(irq_stack_union); + /* * Build-time check on the image size: */ ASSERT((_end - _text <= KERNEL_IMAGE_SIZE), "kernel image bigger than KERNEL_IMAGE_SIZE") + +#ifdef CONFIG_SMP +ASSERT((per_cpu__irq_stack_union == 0), + "irq_stack_union is not at start of per-cpu area"); +#endif + +#ifdef CONFIG_KEXEC +#include <asm/kexec.h> + +ASSERT(kexec_control_code_size <= KEXEC_CONTROL_CODE_MAX_SIZE, + "kexec control code size is too big") +#endif diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c index a688f3bfaec..74de562812c 100644 --- a/arch/x86/kernel/vsmp_64.c +++ b/arch/x86/kernel/vsmp_64.c @@ -22,7 +22,7 @@ #include <asm/paravirt.h> #include <asm/setup.h> -#if defined CONFIG_PCI && defined CONFIG_PARAVIRT +#ifdef CONFIG_PARAVIRT /* * Interrupt control on vSMPowered systems: * ~AC is a shadow of IF. If IF is 'on' AC should be 'off' @@ -37,6 +37,7 @@ static unsigned long vsmp_save_fl(void) flags &= ~X86_EFLAGS_IF; return flags; } +PV_CALLEE_SAVE_REGS_THUNK(vsmp_save_fl); static void vsmp_restore_fl(unsigned long flags) { @@ -46,6 +47,7 @@ static void vsmp_restore_fl(unsigned long flags) flags |= X86_EFLAGS_AC; native_restore_fl(flags); } +PV_CALLEE_SAVE_REGS_THUNK(vsmp_restore_fl); static void vsmp_irq_disable(void) { @@ -53,6 +55,7 @@ static void vsmp_irq_disable(void) native_restore_fl((flags & ~X86_EFLAGS_IF) | X86_EFLAGS_AC); } +PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_disable); static void vsmp_irq_enable(void) { @@ -60,6 +63,7 @@ static void vsmp_irq_enable(void) native_restore_fl((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC)); } +PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_enable); static unsigned __init_or_module vsmp_patch(u8 type, u16 clobbers, void *ibuf, unsigned long addr, unsigned len) @@ -90,10 +94,10 @@ static void __init set_vsmp_pv_ops(void) cap, ctl); if (cap & ctl & (1 << 4)) { /* Setup irq ops and turn on vSMP IRQ fastpath handling */ - pv_irq_ops.irq_disable = vsmp_irq_disable; - pv_irq_ops.irq_enable = vsmp_irq_enable; - pv_irq_ops.save_fl = vsmp_save_fl; - pv_irq_ops.restore_fl = vsmp_restore_fl; + pv_irq_ops.irq_disable = PV_CALLEE_SAVE(vsmp_irq_disable); + pv_irq_ops.irq_enable = PV_CALLEE_SAVE(vsmp_irq_enable); + pv_irq_ops.save_fl = PV_CALLEE_SAVE(vsmp_save_fl); + pv_irq_ops.restore_fl = PV_CALLEE_SAVE(vsmp_restore_fl); pv_init_ops.patch = vsmp_patch; ctl &= ~(1 << 4); @@ -110,7 +114,6 @@ static void __init set_vsmp_pv_ops(void) } #endif -#ifdef CONFIG_PCI static int is_vsmp = -1; static void __init detect_vsmp_box(void) @@ -135,15 +138,6 @@ int is_vsmp_box(void) return 0; } } -#else -static void __init detect_vsmp_box(void) -{ -} -int is_vsmp_box(void) -{ - return 0; -} -#endif void __init vsmp_init(void) { diff --git a/arch/x86/kernel/x8664_ksyms_64.c b/arch/x86/kernel/x8664_ksyms_64.c index 695e426aa35..3909e3ba5ce 100644 --- a/arch/x86/kernel/x8664_ksyms_64.c +++ b/arch/x86/kernel/x8664_ksyms_64.c @@ -58,5 +58,3 @@ EXPORT_SYMBOL(__memcpy); EXPORT_SYMBOL(empty_zero_page); EXPORT_SYMBOL(init_level4_pgt); EXPORT_SYMBOL(load_gs_index); - -EXPORT_SYMBOL(_proxy_pda); |