diff options
Diffstat (limited to 'arch/x86')
96 files changed, 5114 insertions, 7007 deletions
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index f65c2744d57..350bee1d54d 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -26,6 +26,7 @@ config X86 select HAVE_KPROBES select ARCH_WANT_OPTIONAL_GPIOLIB select HAVE_KRETPROBES + select HAVE_FTRACE_MCOUNT_RECORD select HAVE_DYNAMIC_FTRACE select HAVE_FTRACE select HAVE_KVM if ((X86_32 && !X86_VOYAGER && !X86_VISWS && !X86_NUMAQ) || X86_64) @@ -39,10 +40,6 @@ config ARCH_DEFCONFIG default "arch/x86/configs/i386_defconfig" if X86_32 default "arch/x86/configs/x86_64_defconfig" if X86_64 - -config GENERIC_LOCKBREAK - def_bool n - config GENERIC_TIME def_bool y @@ -95,7 +92,7 @@ config GENERIC_HWEIGHT def_bool y config GENERIC_GPIO - def_bool n + bool config ARCH_MAY_HAVE_PC_FDC def_bool y @@ -106,12 +103,6 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM def_bool X86_XADD -config ARCH_HAS_ILOG2_U32 - def_bool n - -config ARCH_HAS_ILOG2_U64 - def_bool n - config ARCH_HAS_CPU_IDLE_WAIT def_bool y @@ -125,6 +116,9 @@ config GENERIC_TIME_VSYSCALL config ARCH_HAS_CPU_RELAX def_bool y +config ARCH_HAS_DEFAULT_IDLE + def_bool y + config ARCH_HAS_CACHE_LINE_SIZE def_bool y @@ -203,6 +197,7 @@ config X86_TRAMPOLINE config KTIME_SCALAR def_bool X86_32 source "init/Kconfig" +source "kernel/Kconfig.freezer" menu "Processor type and features" @@ -758,9 +753,8 @@ config I8K Say N otherwise. config X86_REBOOTFIXUPS - def_bool n - prompt "Enable X86 board specific fixups for reboot" - depends on X86_32 && X86 + bool "Enable X86 board specific fixups for reboot" + depends on X86_32 ---help--- This enables chipset and/or board specific fixups to be done in order to get reboot to work correctly. This is only needed on @@ -944,16 +938,17 @@ config HIGHMEM depends on X86_32 && (HIGHMEM64G || HIGHMEM4G) config X86_PAE - def_bool n - prompt "PAE (Physical Address Extension) Support" + bool "PAE (Physical Address Extension) Support" depends on X86_32 && !HIGHMEM4G - select RESOURCES_64BIT help PAE is required for NX support, and furthermore enables larger swapspace support for non-overcommit purposes. It has the cost of more pagetable lookup overhead, and also consumes more pagetable space per process. +config ARCH_PHYS_ADDR_T_64BIT + def_bool X86_64 || X86_PAE + # Common NUMA Features config NUMA bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)" @@ -1238,8 +1233,7 @@ config X86_PAT If unsure, say Y. config EFI - def_bool n - prompt "EFI runtime service support" + bool "EFI runtime service support" depends on ACPI ---help--- This enables the kernel to use EFI runtime services that are @@ -1252,14 +1246,6 @@ config EFI resultant kernel should continue to boot on existing non-EFI platforms. -config IRQBALANCE - def_bool y - prompt "Enable kernel irq balancing" - depends on X86_32 && SMP && X86_IO_APIC - help - The default yes will allow the kernel to do irq load balancing. - Saying no will keep the kernel from doing irq load balancing. - config SECCOMP def_bool y prompt "Enable seccomp to safely compute untrusted bytecode" @@ -1652,6 +1638,8 @@ source "arch/x86/kernel/cpu/cpufreq/Kconfig" source "drivers/cpuidle/Kconfig" +source "drivers/idle/Kconfig" + endmenu diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c index 1e6fe0214c8..99b3079dc6a 100644 --- a/arch/x86/boot/video-vesa.c +++ b/arch/x86/boot/video-vesa.c @@ -88,14 +88,11 @@ static int vesa_probe(void) (vminfo.memory_layout == 4 || vminfo.memory_layout == 6) && vminfo.memory_planes == 1) { -#ifdef CONFIG_FB +#ifdef CONFIG_FB_BOOT_VESA_SUPPORT /* Graphics mode, color, linear frame buffer supported. Only register the mode if if framebuffer is configured, however, - otherwise the user will be left without a screen. - We don't require CONFIG_FB_VESA, however, since - some of the other framebuffer drivers can use - this mode-setting, too. */ + otherwise the user will be left without a screen. */ mi = GET_HEAP(struct mode_info, 1); mi->mode = mode + VIDEO_FIRST_VESA; mi->depth = vminfo.bpp; @@ -133,10 +130,12 @@ static int vesa_set_mode(struct mode_info *mode) if ((vminfo.mode_attr & 0x15) == 0x05) { /* It's a supported text mode */ is_graphic = 0; +#ifdef CONFIG_FB_BOOT_VESA_SUPPORT } else if ((vminfo.mode_attr & 0x99) == 0x99) { /* It's a graphics mode with linear frame buffer */ is_graphic = 1; vesa_mode |= 0x4000; /* Request linear frame buffer */ +#endif } else { return -1; /* Invalid mode */ } diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig index 52d0359719d..13b8c86ae98 100644 --- a/arch/x86/configs/i386_defconfig +++ b/arch/x86/configs/i386_defconfig @@ -287,7 +287,6 @@ CONFIG_MTRR=y # CONFIG_MTRR_SANITIZER is not set CONFIG_X86_PAT=y CONFIG_EFI=y -# CONFIG_IRQBALANCE is not set CONFIG_SECCOMP=y # CONFIG_HZ_100 is not set # CONFIG_HZ_250 is not set diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index eb4314768bf..256b00b6189 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -571,8 +571,8 @@ ia32_sys_call_table: .quad compat_sys_setrlimit /* 75 */ .quad compat_sys_old_getrlimit /* old_getrlimit */ .quad compat_sys_getrusage - .quad sys32_gettimeofday - .quad sys32_settimeofday + .quad compat_sys_gettimeofday + .quad compat_sys_settimeofday .quad sys_getgroups16 /* 80 */ .quad sys_setgroups16 .quad sys32_old_select diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c index beda4232ce6..2e09dcd3c0a 100644 --- a/arch/x86/ia32/sys_ia32.c +++ b/arch/x86/ia32/sys_ia32.c @@ -49,41 +49,6 @@ #define AA(__x) ((unsigned long)(__x)) -int cp_compat_stat(struct kstat *kbuf, struct compat_stat __user *ubuf) -{ - compat_ino_t ino; - - typeof(ubuf->st_uid) uid = 0; - typeof(ubuf->st_gid) gid = 0; - SET_UID(uid, kbuf->uid); - SET_GID(gid, kbuf->gid); - if (!old_valid_dev(kbuf->dev) || !old_valid_dev(kbuf->rdev)) - return -EOVERFLOW; - if (kbuf->size >= 0x7fffffff) - return -EOVERFLOW; - ino = kbuf->ino; - if (sizeof(ino) < sizeof(kbuf->ino) && ino != kbuf->ino) - return -EOVERFLOW; - if (!access_ok(VERIFY_WRITE, ubuf, sizeof(struct compat_stat)) || - __put_user(old_encode_dev(kbuf->dev), &ubuf->st_dev) || - __put_user(ino, &ubuf->st_ino) || - __put_user(kbuf->mode, &ubuf->st_mode) || - __put_user(kbuf->nlink, &ubuf->st_nlink) || - __put_user(uid, &ubuf->st_uid) || - __put_user(gid, &ubuf->st_gid) || - __put_user(old_encode_dev(kbuf->rdev), &ubuf->st_rdev) || - __put_user(kbuf->size, &ubuf->st_size) || - __put_user(kbuf->atime.tv_sec, &ubuf->st_atime) || - __put_user(kbuf->atime.tv_nsec, &ubuf->st_atime_nsec) || - __put_user(kbuf->mtime.tv_sec, &ubuf->st_mtime) || - __put_user(kbuf->mtime.tv_nsec, &ubuf->st_mtime_nsec) || - __put_user(kbuf->ctime.tv_sec, &ubuf->st_ctime) || - __put_user(kbuf->ctime.tv_nsec, &ubuf->st_ctime_nsec) || - __put_user(kbuf->blksize, &ubuf->st_blksize) || - __put_user(kbuf->blocks, &ubuf->st_blocks)) - return -EFAULT; - return 0; -} asmlinkage long sys32_truncate64(char __user *filename, unsigned long offset_low, @@ -402,75 +367,11 @@ asmlinkage long sys32_rt_sigprocmask(int how, compat_sigset_t __user *set, return 0; } -static inline long get_tv32(struct timeval *o, struct compat_timeval __user *i) -{ - int err = -EFAULT; - - if (access_ok(VERIFY_READ, i, sizeof(*i))) { - err = __get_user(o->tv_sec, &i->tv_sec); - err |= __get_user(o->tv_usec, &i->tv_usec); - } - return err; -} - -static inline long put_tv32(struct compat_timeval __user *o, struct timeval *i) -{ - int err = -EFAULT; - - if (access_ok(VERIFY_WRITE, o, sizeof(*o))) { - err = __put_user(i->tv_sec, &o->tv_sec); - err |= __put_user(i->tv_usec, &o->tv_usec); - } - return err; -} - asmlinkage long sys32_alarm(unsigned int seconds) { return alarm_setitimer(seconds); } -/* - * Translations due to time_t size differences. Which affects all - * sorts of things, like timeval and itimerval. - */ -asmlinkage long sys32_gettimeofday(struct compat_timeval __user *tv, - struct timezone __user *tz) -{ - if (tv) { - struct timeval ktv; - - do_gettimeofday(&ktv); - if (put_tv32(tv, &ktv)) - return -EFAULT; - } - if (tz) { - if (copy_to_user(tz, &sys_tz, sizeof(sys_tz))) - return -EFAULT; - } - return 0; -} - -asmlinkage long sys32_settimeofday(struct compat_timeval __user *tv, - struct timezone __user *tz) -{ - struct timeval ktv; - struct timespec kts; - struct timezone ktz; - - if (tv) { - if (get_tv32(&ktv, tv)) - return -EFAULT; - kts.tv_sec = ktv.tv_sec; - kts.tv_nsec = ktv.tv_usec * NSEC_PER_USEC; - } - if (tz) { - if (copy_from_user(&ktz, tz, sizeof(ktz))) - return -EFAULT; - } - - return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL); -} - struct sel_arg_struct { unsigned int n; unsigned int inp; diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 0d41f0343dc..d7e5a58ee22 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -23,7 +23,7 @@ CFLAGS_hpet.o := $(nostackp) CFLAGS_tsc.o := $(nostackp) obj-y := process_$(BITS).o signal_$(BITS).o entry_$(BITS).o -obj-y += traps.o irq_$(BITS).o dumpstack_$(BITS).o +obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o obj-y += time_$(BITS).o ioport.o ldt.o obj-y += setup.o i8259.o irqinit_$(BITS).o setup_percpu.o obj-$(CONFIG_X86_VISWS) += visws_quirks.o @@ -60,8 +60,8 @@ obj-$(CONFIG_X86_32_SMP) += smpcommon.o obj-$(CONFIG_X86_64_SMP) += tsc_sync.o smpcommon.o obj-$(CONFIG_X86_TRAMPOLINE) += trampoline_$(BITS).o obj-$(CONFIG_X86_MPPARSE) += mpparse.o -obj-$(CONFIG_X86_LOCAL_APIC) += apic_$(BITS).o nmi.o -obj-$(CONFIG_X86_IO_APIC) += io_apic_$(BITS).o +obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o +obj-$(CONFIG_X86_IO_APIC) += io_apic.o obj-$(CONFIG_X86_REBOOTFIXUPS) += reboot_fixups_32.o obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o obj-$(CONFIG_KEXEC) += machine_kexec_$(BITS).o @@ -108,7 +108,7 @@ obj-$(CONFIG_MICROCODE) += microcode.o # 64 bit specific files ifeq ($(CONFIG_X86_64),y) obj-y += genapic_64.o genapic_flat_64.o genx2apic_uv_x.o tlb_uv.o - obj-y += bios_uv.o + obj-y += bios_uv.o uv_irq.o uv_sysfs.o obj-y += genx2apic_cluster.o obj-y += genx2apic_phys.o obj-$(CONFIG_X86_PM_TIMER) += pmtimer_64.o diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index eb875cdc736..53b01a1ae10 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -153,12 +153,13 @@ char *__init __acpi_map_table(unsigned long phys, unsigned long size) } #ifdef CONFIG_PCI_MMCONFIG + +static int acpi_mcfg_64bit_base_addr __initdata = FALSE; + /* The physical address of the MMCONFIG aperture. Set from ACPI tables. */ struct acpi_mcfg_allocation *pci_mmcfg_config; int pci_mmcfg_config_num; -static int acpi_mcfg_64bit_base_addr __initdata = FALSE; - static int __init acpi_mcfg_oem_check(struct acpi_table_mcfg *mcfg) { if (!strcmp(mcfg->header.oem_id, "SGI")) @@ -1256,7 +1257,7 @@ static int __init acpi_parse_madt_ioapic_entries(void) count = acpi_table_parse_madt(ACPI_MADT_TYPE_INTERRUPT_OVERRIDE, acpi_parse_int_src_ovr, - NR_IRQ_VECTORS); + nr_irqs); if (count < 0) { printk(KERN_ERR PREFIX "Error parsing interrupt source overrides entry\n"); @@ -1276,7 +1277,7 @@ static int __init acpi_parse_madt_ioapic_entries(void) count = acpi_table_parse_madt(ACPI_MADT_TYPE_NMI_SOURCE, acpi_parse_nmi_src, - NR_IRQ_VECTORS); + nr_irqs); if (count < 0) { printk(KERN_ERR PREFIX "Error parsing NMI SRC entry\n"); /* TBD: Cleanup to allow fallback to MPS */ diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 426e5d91b63..806b4e9051b 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -10,6 +10,7 @@ #include <linux/dmi.h> #include <linux/cpumask.h> #include <asm/segment.h> +#include <asm/desc.h> #include "realmode/wakeup.h" #include "sleep.h" @@ -21,7 +22,7 @@ unsigned long acpi_realmode_flags; static unsigned long acpi_realmode; #if defined(CONFIG_SMP) && defined(CONFIG_64BIT) -static char temp_stack[10240]; +static char temp_stack[4096]; #endif /** @@ -97,7 +98,9 @@ int acpi_save_state_mem(void) #else /* CONFIG_64BIT */ header->trampoline_segment = setup_trampoline() >> 4; #ifdef CONFIG_SMP - stack_start.sp = temp_stack + 4096; + stack_start.sp = temp_stack + sizeof(temp_stack); + early_gdt_descr.address = + (unsigned long)get_cpu_gdt_table(smp_processor_id()); #endif initial_code = (unsigned long)wakeup_long64; saved_magic = 0x123456789abcdef0; diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 34e4d112b1e..a8fd9ebdc8e 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -295,7 +295,7 @@ static int iommu_flush_pages(struct amd_iommu *iommu, u16 domid, u64 address, size_t size) { int s = 0; - unsigned pages = iommu_num_pages(address, size); + unsigned pages = iommu_num_pages(address, size, PAGE_SIZE); address &= PAGE_MASK; @@ -680,7 +680,8 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu, iommu->exclusion_start < dma_dom->aperture_size) { unsigned long startpage = iommu->exclusion_start >> PAGE_SHIFT; int pages = iommu_num_pages(iommu->exclusion_start, - iommu->exclusion_length); + iommu->exclusion_length, + PAGE_SIZE); dma_ops_reserve_addresses(dma_dom, startpage, pages); } @@ -935,7 +936,7 @@ static dma_addr_t __map_single(struct device *dev, unsigned long align_mask = 0; int i; - pages = iommu_num_pages(paddr, size); + pages = iommu_num_pages(paddr, size, PAGE_SIZE); paddr &= PAGE_MASK; if (align) @@ -980,7 +981,7 @@ static void __unmap_single(struct amd_iommu *iommu, if ((dma_addr == 0) || (dma_addr + size > dma_dom->aperture_size)) return; - pages = iommu_num_pages(dma_addr, size); + pages = iommu_num_pages(dma_addr, size, PAGE_SIZE); dma_addr &= PAGE_MASK; start = dma_addr; diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic.c index 21c831d96af..04a7f960bbc 100644 --- a/arch/x86/kernel/apic_32.c +++ b/arch/x86/kernel/apic.c @@ -23,11 +23,13 @@ #include <linux/mc146818rtc.h> #include <linux/kernel_stat.h> #include <linux/sysdev.h> +#include <linux/ioport.h> #include <linux/cpu.h> #include <linux/clockchips.h> #include <linux/acpi_pmtmr.h> #include <linux/module.h> #include <linux/dmi.h> +#include <linux/dmar.h> #include <asm/atomic.h> #include <asm/smp.h> @@ -36,8 +38,14 @@ #include <asm/desc.h> #include <asm/arch_hooks.h> #include <asm/hpet.h> +#include <asm/pgalloc.h> #include <asm/i8253.h> #include <asm/nmi.h> +#include <asm/idle.h> +#include <asm/proto.h> +#include <asm/timex.h> +#include <asm/apic.h> +#include <asm/i8259.h> #include <mach_apic.h> #include <mach_apicdef.h> @@ -50,16 +58,58 @@ # error SPURIOUS_APIC_VECTOR definition error #endif -unsigned long mp_lapic_addr; - +#ifdef CONFIG_X86_32 /* * Knob to control our willingness to enable the local APIC. * * +1=force-enable */ static int force_enable_local_apic; -int disable_apic; +/* + * APIC command line parameters + */ +static int __init parse_lapic(char *arg) +{ + force_enable_local_apic = 1; + return 0; +} +early_param("lapic", parse_lapic); +/* Local APIC was disabled by the BIOS and enabled by the kernel */ +static int enabled_via_apicbase; + +#endif + +#ifdef CONFIG_X86_64 +static int apic_calibrate_pmtmr __initdata; +static __init int setup_apicpmtimer(char *s) +{ + apic_calibrate_pmtmr = 1; + notsc_setup(NULL); + return 0; +} +__setup("apicpmtimer", setup_apicpmtimer); +#endif + +#ifdef CONFIG_X86_64 +#define HAVE_X2APIC +#endif + +#ifdef HAVE_X2APIC +int x2apic; +/* x2apic enabled before OS handover */ +int x2apic_preenabled; +int disable_x2apic; +static __init int setup_nox2apic(char *str) +{ + disable_x2apic = 1; + setup_clear_cpu_cap(X86_FEATURE_X2APIC); + return 0; +} +early_param("nox2apic", setup_nox2apic); +#endif +unsigned long mp_lapic_addr; +int disable_apic; /* Disable local APIC timer from the kernel commandline or via dmi quirk */ static int disable_apic_timer __cpuinitdata; /* Local APIC timer works in C2 */ @@ -110,9 +160,6 @@ static struct clock_event_device lapic_clockevent = { }; static DEFINE_PER_CPU(struct clock_event_device, lapic_events); -/* Local APIC was disabled by the BIOS and enabled by the kernel */ -static int enabled_via_apicbase; - static unsigned long apic_phys; /* @@ -202,6 +249,42 @@ static struct apic_ops xapic_ops = { struct apic_ops __read_mostly *apic_ops = &xapic_ops; EXPORT_SYMBOL_GPL(apic_ops); +#ifdef HAVE_X2APIC +static void x2apic_wait_icr_idle(void) +{ + /* no need to wait for icr idle in x2apic */ + return; +} + +static u32 safe_x2apic_wait_icr_idle(void) +{ + /* no need to wait for icr idle in x2apic */ + return 0; +} + +void x2apic_icr_write(u32 low, u32 id) +{ + wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low); +} + +u64 x2apic_icr_read(void) +{ + unsigned long val; + + rdmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), val); + return val; +} + +static struct apic_ops x2apic_ops = { + .read = native_apic_msr_read, + .write = native_apic_msr_write, + .icr_read = x2apic_icr_read, + .icr_write = x2apic_icr_write, + .wait_icr_idle = x2apic_wait_icr_idle, + .safe_wait_icr_idle = safe_x2apic_wait_icr_idle, +}; +#endif + /** * enable_NMI_through_LVT0 - enable NMI through local vector table 0 */ @@ -219,6 +302,7 @@ void __cpuinit enable_NMI_through_LVT0(void) apic_write(APIC_LVT0, v); } +#ifdef CONFIG_X86_32 /** * get_physical_broadcast - Get number of physical broadcast IDs */ @@ -226,6 +310,7 @@ int get_physical_broadcast(void) { return modern_apic() ? 0xff : 0xf; } +#endif /** * lapic_get_maxlvt - get the maximum number of local vector table entries @@ -247,11 +332,7 @@ int lapic_get_maxlvt(void) */ /* Clock divisor */ -#ifdef CONFG_X86_64 -#define APIC_DIVISOR 1 -#else #define APIC_DIVISOR 16 -#endif /* * This function sets up the local APIC timer, with a timeout of @@ -383,7 +464,7 @@ static void lapic_timer_broadcast(cpumask_t mask) * Setup the local APIC timer for this CPU. Copy the initilized values * of the boot CPU and register the clock event in the framework. */ -static void __devinit setup_APIC_timer(void) +static void __cpuinit setup_APIC_timer(void) { struct clock_event_device *levt = &__get_cpu_var(lapic_events); @@ -453,14 +534,51 @@ static void __init lapic_cal_handler(struct clock_event_device *dev) } } +static int __init calibrate_by_pmtimer(long deltapm, long *delta) +{ + const long pm_100ms = PMTMR_TICKS_PER_SEC / 10; + const long pm_thresh = pm_100ms / 100; + unsigned long mult; + u64 res; + +#ifndef CONFIG_X86_PM_TIMER + return -1; +#endif + + apic_printk(APIC_VERBOSE, "... PM timer delta = %ld\n", deltapm); + + /* Check, if the PM timer is available */ + if (!deltapm) + return -1; + + mult = clocksource_hz2mult(PMTMR_TICKS_PER_SEC, 22); + + if (deltapm > (pm_100ms - pm_thresh) && + deltapm < (pm_100ms + pm_thresh)) { + apic_printk(APIC_VERBOSE, "... PM timer result ok\n"); + } else { + res = (((u64)deltapm) * mult) >> 22; + do_div(res, 1000000); + printk(KERN_WARNING "APIC calibration not consistent " + "with PM Timer: %ldms instead of 100ms\n", + (long)res); + /* Correct the lapic counter value */ + res = (((u64)(*delta)) * pm_100ms); + do_div(res, deltapm); + printk(KERN_INFO "APIC delta adjusted to PM-Timer: " + "%lu (%ld)\n", (unsigned long)res, *delta); + *delta = (long)res; + } + + return 0; +} + static int __init calibrate_APIC_clock(void) { struct clock_event_device *levt = &__get_cpu_var(lapic_events); - const long pm_100ms = PMTMR_TICKS_PER_SEC/10; - const long pm_thresh = pm_100ms/100; void (*real_handler)(struct clock_event_device *dev); unsigned long deltaj; - long delta, deltapm; + long delta; int pm_referenced = 0; local_irq_disable(); @@ -470,10 +588,10 @@ static int __init calibrate_APIC_clock(void) global_clock_event->event_handler = lapic_cal_handler; /* - * Setup the APIC counter to 1e9. There is no way the lapic + * Setup the APIC counter to maximum. There is no way the lapic * can underflow in the 100ms detection time frame */ - __setup_APIC_LVTT(1000000000, 0, 0); + __setup_APIC_LVTT(0xffffffff, 0, 0); /* Let the interrupts run */ local_irq_enable(); @@ -490,34 +608,9 @@ static int __init calibrate_APIC_clock(void) delta = lapic_cal_t1 - lapic_cal_t2; apic_printk(APIC_VERBOSE, "... lapic delta = %ld\n", delta); - /* Check, if the PM timer is available */ - deltapm = lapic_cal_pm2 - lapic_cal_pm1; - apic_printk(APIC_VERBOSE, "... PM timer delta = %ld\n", deltapm); - - if (deltapm) { - unsigned long mult; - u64 res; - - mult = clocksource_hz2mult(PMTMR_TICKS_PER_SEC, 22); - - if (deltapm > (pm_100ms - pm_thresh) && - deltapm < (pm_100ms + pm_thresh)) { - apic_printk(APIC_VERBOSE, "... PM timer result ok\n"); - } else { - res = (((u64) deltapm) * mult) >> 22; - do_div(res, 1000000); - printk(KERN_WARNING "APIC calibration not consistent " - "with PM Timer: %ldms instead of 100ms\n", - (long)res); - /* Correct the lapic counter value */ - res = (((u64) delta) * pm_100ms); - do_div(res, deltapm); - printk(KERN_INFO "APIC delta adjusted to PM-Timer: " - "%lu (%ld)\n", (unsigned long) res, delta); - delta = (long) res; - } - pm_referenced = 1; - } + /* we trust the PM based calibration if possible */ + pm_referenced = !calibrate_by_pmtimer(lapic_cal_pm2 - lapic_cal_pm1, + &delta); /* Calculate the scaled math multiplication factor */ lapic_clockevent.mult = div_sc(delta, TICK_NSEC * LAPIC_CAL_LOOPS, @@ -559,7 +652,10 @@ static int __init calibrate_APIC_clock(void) levt->features &= ~CLOCK_EVT_FEAT_DUMMY; - /* We trust the pm timer based calibration */ + /* + * PM timer calibration failed or not turned on + * so lets try APIC timer based calibration + */ if (!pm_referenced) { apic_printk(APIC_VERBOSE, "... verify APIC timer\n"); @@ -652,7 +748,7 @@ void __init setup_boot_APIC_clock(void) setup_APIC_timer(); } -void __devinit setup_secondary_APIC_clock(void) +void __cpuinit setup_secondary_APIC_clock(void) { setup_APIC_timer(); } @@ -718,6 +814,9 @@ void smp_apic_timer_interrupt(struct pt_regs *regs) * Besides, if we don't timer interrupts ignore the global * interrupt lock, which is the WrongThing (tm) to do. */ +#ifdef CONFIG_X86_64 + exit_idle(); +#endif irq_enter(); local_apic_timer_interrupt(); irq_exit(); @@ -991,40 +1090,43 @@ void __init init_bsp_APIC(void) static void __cpuinit lapic_setup_esr(void) { - unsigned long oldvalue, value, maxlvt; - if (lapic_is_integrated() && !esr_disable) { - if (esr_disable) { - /* - * Something untraceable is creating bad interrupts on - * secondary quads ... for the moment, just leave the - * ESR disabled - we can't do anything useful with the - * errors anyway - mbligh - */ - printk(KERN_INFO "Leaving ESR disabled.\n"); - return; - } - /* !82489DX */ - maxlvt = lapic_get_maxlvt(); - if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ - apic_write(APIC_ESR, 0); - oldvalue = apic_read(APIC_ESR); + unsigned int oldvalue, value, maxlvt; + + if (!lapic_is_integrated()) { + printk(KERN_INFO "No ESR for 82489DX.\n"); + return; + } - /* enables sending errors */ - value = ERROR_APIC_VECTOR; - apic_write(APIC_LVTERR, value); + if (esr_disable) { /* - * spec says clear errors after enabling vector. + * Something untraceable is creating bad interrupts on + * secondary quads ... for the moment, just leave the + * ESR disabled - we can't do anything useful with the + * errors anyway - mbligh */ - if (maxlvt > 3) - apic_write(APIC_ESR, 0); - value = apic_read(APIC_ESR); - if (value != oldvalue) - apic_printk(APIC_VERBOSE, "ESR value before enabling " - "vector: 0x%08lx after: 0x%08lx\n", - oldvalue, value); - } else { - printk(KERN_INFO "No ESR for 82489DX.\n"); + printk(KERN_INFO "Leaving ESR disabled.\n"); + return; } + + maxlvt = lapic_get_maxlvt(); + if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ + apic_write(APIC_ESR, 0); + oldvalue = apic_read(APIC_ESR); + + /* enables sending errors */ + value = ERROR_APIC_VECTOR; + apic_write(APIC_LVTERR, value); + + /* + * spec says clear errors after enabling vector. + */ + if (maxlvt > 3) + apic_write(APIC_ESR, 0); + value = apic_read(APIC_ESR); + if (value != oldvalue) + apic_printk(APIC_VERBOSE, "ESR value before enabling " + "vector: 0x%08x after: 0x%08x\n", + oldvalue, value); } @@ -1033,24 +1135,27 @@ static void __cpuinit lapic_setup_esr(void) */ void __cpuinit setup_local_APIC(void) { - unsigned long value, integrated; + unsigned int value; int i, j; +#ifdef CONFIG_X86_32 /* Pound the ESR really hard over the head with a big hammer - mbligh */ - if (esr_disable) { + if (lapic_is_integrated() && esr_disable) { apic_write(APIC_ESR, 0); apic_write(APIC_ESR, 0); apic_write(APIC_ESR, 0); apic_write(APIC_ESR, 0); } +#endif - integrated = lapic_is_integrated(); + preempt_disable(); /* * Double-check whether this APIC is really registered. + * This is meaningless in clustered apic mode, so we skip it. */ if (!apic_id_registered()) - WARN_ON_ONCE(1); + BUG(); /* * Intel recommends to set DFR, LDR and TPR before enabling @@ -1096,6 +1201,7 @@ void __cpuinit setup_local_APIC(void) */ value |= APIC_SPIV_APIC_ENABLED; +#ifdef CONFIG_X86_32 /* * Some unknown Intel IO/APIC (or APIC) errata is biting us with * certain networking cards. If high frequency interrupts are @@ -1116,8 +1222,13 @@ void __cpuinit setup_local_APIC(void) * See also the comment in end_level_ioapic_irq(). --macro */ - /* Enable focus processor (bit==0) */ + /* + * - enable focus processor (bit==0) + * - 64bit mode always use processor focus + * so no need to set it + */ value &= ~APIC_SPIV_FOCUS_DISABLED; +#endif /* * Set spurious IRQ vector @@ -1154,9 +1265,11 @@ void __cpuinit setup_local_APIC(void) value = APIC_DM_NMI; else value = APIC_DM_NMI | APIC_LVT_MASKED; - if (!integrated) /* 82489DX */ + if (!lapic_is_integrated()) /* 82489DX */ value |= APIC_LVT_LEVEL_TRIGGER; apic_write(APIC_LVT1, value); + + preempt_enable(); } void __cpuinit end_local_APIC_setup(void) @@ -1177,6 +1290,153 @@ void __cpuinit end_local_APIC_setup(void) apic_pm_activate(); } +#ifdef HAVE_X2APIC +void check_x2apic(void) +{ + int msr, msr2; + + rdmsr(MSR_IA32_APICBASE, msr, msr2); + + if (msr & X2APIC_ENABLE) { + printk("x2apic enabled by BIOS, switching to x2apic ops\n"); + x2apic_preenabled = x2apic = 1; + apic_ops = &x2apic_ops; + } +} + +void enable_x2apic(void) +{ + int msr, msr2; + + rdmsr(MSR_IA32_APICBASE, msr, msr2); + if (!(msr & X2APIC_ENABLE)) { + printk("Enabling x2apic\n"); + wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0); + } +} + +void enable_IR_x2apic(void) +{ +#ifdef CONFIG_INTR_REMAP + int ret; + unsigned long flags; + + if (!cpu_has_x2apic) + return; + + if (!x2apic_preenabled && disable_x2apic) { + printk(KERN_INFO + "Skipped enabling x2apic and Interrupt-remapping " + "because of nox2apic\n"); + return; + } + + if (x2apic_preenabled && disable_x2apic) + panic("Bios already enabled x2apic, can't enforce nox2apic"); + + if (!x2apic_preenabled && skip_ioapic_setup) { + printk(KERN_INFO + "Skipped enabling x2apic and Interrupt-remapping " + "because of skipping io-apic setup\n"); + return; + } + + ret = dmar_table_init(); + if (ret) { + printk(KERN_INFO + "dmar_table_init() failed with %d:\n", ret); + + if (x2apic_preenabled) + panic("x2apic enabled by bios. But IR enabling failed"); + else + printk(KERN_INFO + "Not enabling x2apic,Intr-remapping\n"); + return; + } + + local_irq_save(flags); + mask_8259A(); + + ret = save_mask_IO_APIC_setup(); + if (ret) { + printk(KERN_INFO "Saving IO-APIC state failed: %d\n", ret); + goto end; + } + + ret = enable_intr_remapping(1); + + if (ret && x2apic_preenabled) { + local_irq_restore(flags); + panic("x2apic enabled by bios. But IR enabling failed"); + } + + if (ret) + goto end_restore; + + if (!x2apic) { + x2apic = 1; + apic_ops = &x2apic_ops; + enable_x2apic(); + } + +end_restore: + if (ret) + /* + * IR enabling failed + */ + restore_IO_APIC_setup(); + else + reinit_intr_remapped_IO_APIC(x2apic_preenabled); + +end: + unmask_8259A(); + local_irq_restore(flags); + + if (!ret) { + if (!x2apic_preenabled) + printk(KERN_INFO + "Enabled x2apic and interrupt-remapping\n"); + else + printk(KERN_INFO + "Enabled Interrupt-remapping\n"); + } else + printk(KERN_ERR + "Failed to enable Interrupt-remapping and x2apic\n"); +#else + if (!cpu_has_x2apic) + return; + + if (x2apic_preenabled) + panic("x2apic enabled prior OS handover," + " enable CONFIG_INTR_REMAP"); + + printk(KERN_INFO "Enable CONFIG_INTR_REMAP for enabling intr-remapping " + " and x2apic\n"); +#endif + + return; +} +#endif /* HAVE_X2APIC */ + +#ifdef CONFIG_X86_64 +/* + * Detect and enable local APICs on non-SMP boards. + * Original code written by Keir Fraser. + * On AMD64 we trust the BIOS - if it says no APIC it is likely + * not correctly set up (usually the APIC timer won't work etc.) + */ +static int __init detect_init_APIC(void) +{ + if (!cpu_has_apic) { + printk(KERN_INFO "No local APIC present\n"); + return -1; + } + + mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; + boot_cpu_physical_apicid = 0; + return 0; +} +#else /* * Detect and initialize APIC */ @@ -1255,12 +1515,46 @@ no_apic: printk(KERN_INFO "No local APIC present or hardware disabled\n"); return -1; } +#endif + +#ifdef CONFIG_X86_64 +void __init early_init_lapic_mapping(void) +{ + unsigned long phys_addr; + + /* + * If no local APIC can be found then go out + * : it means there is no mpatable and MADT + */ + if (!smp_found_config) + return; + + phys_addr = mp_lapic_addr; + + set_fixmap_nocache(FIX_APIC_BASE, phys_addr); + apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", + APIC_BASE, phys_addr); + + /* + * Fetch the APIC ID of the BSP in case we have a + * default configuration (or the MP table is broken). + */ + boot_cpu_physical_apicid = read_apic_id(); +} +#endif /** * init_apic_mappings - initialize APIC mappings */ void __init init_apic_mappings(void) { +#ifdef HAVE_X2APIC + if (x2apic) { + boot_cpu_physical_apicid = read_apic_id(); + return; + } +#endif + /* * If no local APIC can be found then set up a fake all * zeroes page to simulate the local APIC and another @@ -1273,8 +1567,8 @@ void __init init_apic_mappings(void) apic_phys = mp_lapic_addr; set_fixmap_nocache(FIX_APIC_BASE, apic_phys); - printk(KERN_DEBUG "mapped APIC to %08lx (%08lx)\n", APIC_BASE, - apic_phys); + apic_printk(APIC_VERBOSE, "mapped APIC to %08lx (%08lx)\n", + APIC_BASE, apic_phys); /* * Fetch the APIC ID of the BSP in case we have a @@ -1282,18 +1576,27 @@ void __init init_apic_mappings(void) */ if (boot_cpu_physical_apicid == -1U) boot_cpu_physical_apicid = read_apic_id(); - } /* * This initializes the IO-APIC and APIC hardware if this is * a UP kernel. */ - int apic_version[MAX_APICS]; int __init APIC_init_uniprocessor(void) { +#ifdef CONFIG_X86_64 + if (disable_apic) { + printk(KERN_INFO "Apic disabled\n"); + return -1; + } + if (!cpu_has_apic) { + disable_apic = 1; + printk(KERN_INFO "Apic disabled by BIOS\n"); + return -1; + } +#else if (!smp_found_config && !cpu_has_apic) return -1; @@ -1302,39 +1605,68 @@ int __init APIC_init_uniprocessor(void) */ if (!cpu_has_apic && APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) { - printk(KERN_ERR "BIOS bug, local APIC #%d not detected!...\n", + printk(KERN_ERR "BIOS bug, local APIC 0x%x not detected!...\n", boot_cpu_physical_apicid); clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC); return -1; } +#endif - verify_local_APIC(); +#ifdef HAVE_X2APIC + enable_IR_x2apic(); +#endif +#ifdef CONFIG_X86_64 + setup_apic_routing(); +#endif + verify_local_APIC(); connect_bsp_APIC(); +#ifdef CONFIG_X86_64 + apic_write(APIC_ID, SET_APIC_ID(boot_cpu_physical_apicid)); +#else /* * Hack: In case of kdump, after a crash, kernel might be booting * on a cpu with non-zero lapic id. But boot_cpu_physical_apicid * might be zero if read from MP tables. Get it from LAPIC. */ -#ifdef CONFIG_CRASH_DUMP +# ifdef CONFIG_CRASH_DUMP boot_cpu_physical_apicid = read_apic_id(); +# endif #endif physid_set_mask_of_physid(boot_cpu_physical_apicid, &phys_cpu_present_map); - setup_local_APIC(); +#ifdef CONFIG_X86_64 + /* + * Now enable IO-APICs, actually call clear_IO_APIC + * We need clear_IO_APIC before enabling vector on BP + */ + if (!skip_ioapic_setup && nr_ioapics) + enable_IO_APIC(); +#endif + #ifdef CONFIG_X86_IO_APIC if (!smp_found_config || skip_ioapic_setup || !nr_ioapics) #endif localise_nmi_watchdog(); end_local_APIC_setup(); + #ifdef CONFIG_X86_IO_APIC - if (smp_found_config) - if (!skip_ioapic_setup && nr_ioapics) - setup_IO_APIC(); + if (smp_found_config && !skip_ioapic_setup && nr_ioapics) + setup_IO_APIC(); +# ifdef CONFIG_X86_64 + else + nr_ioapics = 0; +# endif #endif + +#ifdef CONFIG_X86_64 + setup_boot_APIC_clock(); + check_nmi_watchdog(); +#else setup_boot_clock(); +#endif return 0; } @@ -1348,8 +1680,11 @@ int __init APIC_init_uniprocessor(void) */ void smp_spurious_interrupt(struct pt_regs *regs) { - unsigned long v; + u32 v; +#ifdef CONFIG_X86_64 + exit_idle(); +#endif irq_enter(); /* * Check if this really is a spurious interrupt and ACK it @@ -1360,10 +1695,14 @@ void smp_spurious_interrupt(struct pt_regs *regs) if (v & (1 << (SPURIOUS_APIC_VECTOR & 0x1f))) ack_APIC_irq(); +#ifdef CONFIG_X86_64 + add_pda(irq_spurious_count, 1); +#else /* see sw-dev-man vol 3, chapter 7.4.13.5 */ printk(KERN_INFO "spurious APIC interrupt on CPU#%d, " "should never happen.\n", smp_processor_id()); __get_cpu_var(irq_stat).irq_spurious_count++; +#endif irq_exit(); } @@ -1372,8 +1711,11 @@ void smp_spurious_interrupt(struct pt_regs *regs) */ void smp_error_interrupt(struct pt_regs *regs) { - unsigned long v, v1; + u32 v, v1; +#ifdef CONFIG_X86_64 + exit_idle(); +#endif irq_enter(); /* First tickle the hardware, only then report what went on. -- REW */ v = apic_read(APIC_ESR); @@ -1392,7 +1734,7 @@ void smp_error_interrupt(struct pt_regs *regs) 6: Received illegal vector 7: Illegal register address */ - printk(KERN_DEBUG "APIC error on CPU%d: %02lx(%02lx)\n", + printk(KERN_DEBUG "APIC error on CPU%d: %02x(%02x)\n", smp_processor_id(), v , v1); irq_exit(); } @@ -1565,6 +1907,13 @@ void __cpuinit generic_processor_info(int apicid, int version) cpu_set(cpu, cpu_present_map); } +#ifdef CONFIG_X86_64 +int hard_smp_processor_id(void) +{ + return read_apic_id(); +} +#endif + /* * Power management */ @@ -1640,7 +1989,7 @@ static int lapic_resume(struct sys_device *dev) local_irq_save(flags); -#ifdef CONFIG_X86_64 +#ifdef HAVE_X2APIC if (x2apic) enable_x2apic(); else @@ -1702,7 +2051,7 @@ static struct sys_device device_lapic = { .cls = &lapic_sysclass, }; -static void __devinit apic_pm_activate(void) +static void __cpuinit apic_pm_activate(void) { apic_pm_state.active = 1; } @@ -1728,16 +2077,87 @@ static void apic_pm_activate(void) { } #endif /* CONFIG_PM */ +#ifdef CONFIG_X86_64 /* - * APIC command line parameters + * apic_is_clustered_box() -- Check if we can expect good TSC + * + * Thus far, the major user of this is IBM's Summit2 series: + * + * Clustered boxes may have unsynced TSC problems if they are + * multi-chassis. Use available data to take a good guess. + * If in doubt, go HPET. */ -static int __init parse_lapic(char *arg) +__cpuinit int apic_is_clustered_box(void) { - force_enable_local_apic = 1; - return 0; + int i, clusters, zeros; + unsigned id; + u16 *bios_cpu_apicid; + DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS); + + /* + * there is not this kind of box with AMD CPU yet. + * Some AMD box with quadcore cpu and 8 sockets apicid + * will be [4, 0x23] or [8, 0x27] could be thought to + * vsmp box still need checking... + */ + if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && !is_vsmp_box()) + return 0; + + bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); + bitmap_zero(clustermap, NUM_APIC_CLUSTERS); + + for (i = 0; i < NR_CPUS; i++) { + /* are we being called early in kernel startup? */ + if (bios_cpu_apicid) { + id = bios_cpu_apicid[i]; + } + else if (i < nr_cpu_ids) { + if (cpu_present(i)) + id = per_cpu(x86_bios_cpu_apicid, i); + else + continue; + } + else + break; + + if (id != BAD_APICID) + __set_bit(APIC_CLUSTERID(id), clustermap); + } + + /* Problem: Partially populated chassis may not have CPUs in some of + * the APIC clusters they have been allocated. Only present CPUs have + * x86_bios_cpu_apicid entries, thus causing zeroes in the bitmap. + * Since clusters are allocated sequentially, count zeros only if + * they are bounded by ones. + */ + clusters = 0; + zeros = 0; + for (i = 0; i < NUM_APIC_CLUSTERS; i++) { + if (test_bit(i, clustermap)) { + clusters += 1 + zeros; + zeros = 0; + } else + ++zeros; + } + + /* ScaleMP vSMPowered boxes have one cluster per board and TSCs are + * not guaranteed to be synced between boards + */ + if (is_vsmp_box() && clusters > 1) + return 1; + + /* + * If clusters > 2, then should be multi-chassis. + * May have to revisit this when multi-core + hyperthreaded CPUs come + * out, but AFAIK this will work even for them. + */ + return (clusters > 2); } -early_param("lapic", parse_lapic); +#endif +/* + * APIC command line parameters + */ static int __init setup_disableapic(char *arg) { disable_apic = 1; @@ -1779,7 +2199,6 @@ static int __init apic_set_verbosity(char *arg) if (!arg) { #ifdef CONFIG_X86_64 skip_ioapic_setup = 0; - ioapic_force = 1; return 0; #endif return -EINVAL; diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c deleted file mode 100644 index 94ddb69ae15..00000000000 --- a/arch/x86/kernel/apic_64.c +++ /dev/null @@ -1,1848 +0,0 @@ -/* - * Local APIC handling, local APIC timers - * - * (c) 1999, 2000 Ingo Molnar <mingo@redhat.com> - * - * Fixes - * Maciej W. Rozycki : Bits for genuine 82489DX APICs; - * thanks to Eric Gilmore - * and Rolf G. Tews - * for testing these extensively. - * Maciej W. Rozycki : Various updates and fixes. - * Mikael Pettersson : Power Management for UP-APIC. - * Pavel Machek and - * Mikael Pettersson : PM converted to driver model. - */ - -#include <linux/init.h> - -#include <linux/mm.h> -#include <linux/delay.h> -#include <linux/bootmem.h> -#include <linux/interrupt.h> -#include <linux/mc146818rtc.h> -#include <linux/kernel_stat.h> -#include <linux/sysdev.h> -#include <linux/ioport.h> -#include <linux/clockchips.h> -#include <linux/acpi_pmtmr.h> -#include <linux/module.h> -#include <linux/dmar.h> - -#include <asm/atomic.h> -#include <asm/smp.h> -#include <asm/mtrr.h> -#include <asm/mpspec.h> -#include <asm/hpet.h> -#include <asm/pgalloc.h> -#include <asm/nmi.h> -#include <asm/idle.h> -#include <asm/proto.h> -#include <asm/timex.h> -#include <asm/apic.h> -#include <asm/i8259.h> - -#include <mach_ipi.h> -#include <mach_apic.h> - -/* Disable local APIC timer from the kernel commandline or via dmi quirk */ -static int disable_apic_timer __cpuinitdata; -static int apic_calibrate_pmtmr __initdata; -int disable_apic; -int disable_x2apic; -int x2apic; - -/* x2apic enabled before OS handover */ -int x2apic_preenabled; - -/* Local APIC timer works in C2 */ -int local_apic_timer_c2_ok; -EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok); - -/* - * Debug level, exported for io_apic.c - */ -unsigned int apic_verbosity; - -/* Have we found an MP table */ -int smp_found_config; - -static struct resource lapic_resource = { - .name = "Local APIC", - .flags = IORESOURCE_MEM | IORESOURCE_BUSY, -}; - -static unsigned int calibration_result; - -static int lapic_next_event(unsigned long delta, - struct clock_event_device *evt); -static void lapic_timer_setup(enum clock_event_mode mode, - struct clock_event_device *evt); -static void lapic_timer_broadcast(cpumask_t mask); -static void apic_pm_activate(void); - -/* - * The local apic timer can be used for any function which is CPU local. - */ -static struct clock_event_device lapic_clockevent = { - .name = "lapic", - .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT - | CLOCK_EVT_FEAT_C3STOP | CLOCK_EVT_FEAT_DUMMY, - .shift = 32, - .set_mode = lapic_timer_setup, - .set_next_event = lapic_next_event, - .broadcast = lapic_timer_broadcast, - .rating = 100, - .irq = -1, -}; -static DEFINE_PER_CPU(struct clock_event_device, lapic_events); - -static unsigned long apic_phys; - -unsigned long mp_lapic_addr; - -/* - * Get the LAPIC version - */ -static inline int lapic_get_version(void) -{ - return GET_APIC_VERSION(apic_read(APIC_LVR)); -} - -/* - * Check, if the APIC is integrated or a separate chip - */ -static inline int lapic_is_integrated(void) -{ -#ifdef CONFIG_X86_64 - return 1; -#else - return APIC_INTEGRATED(lapic_get_version()); -#endif -} - -/* - * Check, whether this is a modern or a first generation APIC - */ -static int modern_apic(void) -{ - /* AMD systems use old APIC versions, so check the CPU */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && - boot_cpu_data.x86 >= 0xf) - return 1; - return lapic_get_version() >= 0x14; -} - -/* - * Paravirt kernels also might be using these below ops. So we still - * use generic apic_read()/apic_write(), which might be pointing to different - * ops in PARAVIRT case. - */ -void xapic_wait_icr_idle(void) -{ - while (apic_read(APIC_ICR) & APIC_ICR_BUSY) - cpu_relax(); -} - -u32 safe_xapic_wait_icr_idle(void) -{ - u32 send_status; - int timeout; - - timeout = 0; - do { - send_status = apic_read(APIC_ICR) & APIC_ICR_BUSY; - if (!send_status) - break; - udelay(100); - } while (timeout++ < 1000); - - return send_status; -} - -void xapic_icr_write(u32 low, u32 id) -{ - apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(id)); - apic_write(APIC_ICR, low); -} - -u64 xapic_icr_read(void) -{ - u32 icr1, icr2; - - icr2 = apic_read(APIC_ICR2); - icr1 = apic_read(APIC_ICR); - - return icr1 | ((u64)icr2 << 32); -} - -static struct apic_ops xapic_ops = { - .read = native_apic_mem_read, - .write = native_apic_mem_write, - .icr_read = xapic_icr_read, - .icr_write = xapic_icr_write, - .wait_icr_idle = xapic_wait_icr_idle, - .safe_wait_icr_idle = safe_xapic_wait_icr_idle, -}; - -struct apic_ops __read_mostly *apic_ops = &xapic_ops; -EXPORT_SYMBOL_GPL(apic_ops); - -static void x2apic_wait_icr_idle(void) -{ - /* no need to wait for icr idle in x2apic */ - return; -} - -static u32 safe_x2apic_wait_icr_idle(void) -{ - /* no need to wait for icr idle in x2apic */ - return 0; -} - -void x2apic_icr_write(u32 low, u32 id) -{ - wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low); -} - -u64 x2apic_icr_read(void) -{ - unsigned long val; - - rdmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), val); - return val; -} - -static struct apic_ops x2apic_ops = { - .read = native_apic_msr_read, - .write = native_apic_msr_write, - .icr_read = x2apic_icr_read, - .icr_write = x2apic_icr_write, - .wait_icr_idle = x2apic_wait_icr_idle, - .safe_wait_icr_idle = safe_x2apic_wait_icr_idle, -}; - -/** - * enable_NMI_through_LVT0 - enable NMI through local vector table 0 - */ -void __cpuinit enable_NMI_through_LVT0(void) -{ - unsigned int v; - - /* unmask and set to NMI */ - v = APIC_DM_NMI; - - /* Level triggered for 82489DX (32bit mode) */ - if (!lapic_is_integrated()) - v |= APIC_LVT_LEVEL_TRIGGER; - - apic_write(APIC_LVT0, v); -} - -/** - * lapic_get_maxlvt - get the maximum number of local vector table entries - */ -int lapic_get_maxlvt(void) -{ - unsigned int v; - - v = apic_read(APIC_LVR); - /* - * - we always have APIC integrated on 64bit mode - * - 82489DXs do not report # of LVT entries - */ - return APIC_INTEGRATED(GET_APIC_VERSION(v)) ? GET_APIC_MAXLVT(v) : 2; -} - -/* - * Local APIC timer - */ - -/* Clock divisor */ -#ifdef CONFG_X86_64 -#define APIC_DIVISOR 1 -#else -#define APIC_DIVISOR 16 -#endif - -/* - * This function sets up the local APIC timer, with a timeout of - * 'clocks' APIC bus clock. During calibration we actually call - * this function twice on the boot CPU, once with a bogus timeout - * value, second time for real. The other (noncalibrating) CPUs - * call this function only once, with the real, calibrated value. - * - * We do reads before writes even if unnecessary, to get around the - * P5 APIC double write bug. - */ -static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen) -{ - unsigned int lvtt_value, tmp_value; - - lvtt_value = LOCAL_TIMER_VECTOR; - if (!oneshot) - lvtt_value |= APIC_LVT_TIMER_PERIODIC; - if (!lapic_is_integrated()) - lvtt_value |= SET_APIC_TIMER_BASE(APIC_TIMER_BASE_DIV); - - if (!irqen) - lvtt_value |= APIC_LVT_MASKED; - - apic_write(APIC_LVTT, lvtt_value); - - /* - * Divide PICLK by 16 - */ - tmp_value = apic_read(APIC_TDCR); - apic_write(APIC_TDCR, - (tmp_value & ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE)) | - APIC_TDR_DIV_16); - - if (!oneshot) - apic_write(APIC_TMICT, clocks / APIC_DIVISOR); -} - -/* - * Setup extended LVT, AMD specific (K8, family 10h) - * - * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and - * MCE interrupts are supported. Thus MCE offset must be set to 0. - * - * If mask=1, the LVT entry does not generate interrupts while mask=0 - * enables the vector. See also the BKDGs. - */ - -#define APIC_EILVT_LVTOFF_MCE 0 -#define APIC_EILVT_LVTOFF_IBS 1 - -static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask) -{ - unsigned long reg = (lvt_off << 4) + APIC_EILVT0; - unsigned int v = (mask << 16) | (msg_type << 8) | vector; - - apic_write(reg, v); -} - -u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask) -{ - setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask); - return APIC_EILVT_LVTOFF_MCE; -} - -u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask) -{ - setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask); - return APIC_EILVT_LVTOFF_IBS; -} -EXPORT_SYMBOL_GPL(setup_APIC_eilvt_ibs); - -/* - * Program the next event, relative to now - */ -static int lapic_next_event(unsigned long delta, - struct clock_event_device *evt) -{ - apic_write(APIC_TMICT, delta); - return 0; -} - -/* - * Setup the lapic timer in periodic or oneshot mode - */ -static void lapic_timer_setup(enum clock_event_mode mode, - struct clock_event_device *evt) -{ - unsigned long flags; - unsigned int v; - - /* Lapic used as dummy for broadcast ? */ - if (evt->features & CLOCK_EVT_FEAT_DUMMY) - return; - - local_irq_save(flags); - - switch (mode) { - case CLOCK_EVT_MODE_PERIODIC: - case CLOCK_EVT_MODE_ONESHOT: - __setup_APIC_LVTT(calibration_result, - mode != CLOCK_EVT_MODE_PERIODIC, 1); - break; - case CLOCK_EVT_MODE_UNUSED: - case CLOCK_EVT_MODE_SHUTDOWN: - v = apic_read(APIC_LVTT); - v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); - apic_write(APIC_LVTT, v); - break; - case CLOCK_EVT_MODE_RESUME: - /* Nothing to do here */ - break; - } - - local_irq_restore(flags); -} - -/* - * Local APIC timer broadcast function - */ -static void lapic_timer_broadcast(cpumask_t mask) -{ -#ifdef CONFIG_SMP - send_IPI_mask(mask, LOCAL_TIMER_VECTOR); -#endif -} - -/* - * Setup the local APIC timer for this CPU. Copy the initilized values - * of the boot CPU and register the clock event in the framework. - */ -static void setup_APIC_timer(void) -{ - struct clock_event_device *levt = &__get_cpu_var(lapic_events); - - memcpy(levt, &lapic_clockevent, sizeof(*levt)); - levt->cpumask = cpumask_of_cpu(smp_processor_id()); - - clockevents_register_device(levt); -} - -/* - * In this function we calibrate APIC bus clocks to the external - * timer. Unfortunately we cannot use jiffies and the timer irq - * to calibrate, since some later bootup code depends on getting - * the first irq? Ugh. - * - * We want to do the calibration only once since we - * want to have local timer irqs syncron. CPUs connected - * by the same APIC bus have the very same bus frequency. - * And we want to have irqs off anyways, no accidental - * APIC irq that way. - */ - -#define TICK_COUNT 100000000 - -static int __init calibrate_APIC_clock(void) -{ - unsigned apic, apic_start; - unsigned long tsc, tsc_start; - int result; - - local_irq_disable(); - - /* - * Put whatever arbitrary (but long enough) timeout - * value into the APIC clock, we just want to get the - * counter running for calibration. - * - * No interrupt enable ! - */ - __setup_APIC_LVTT(250000000, 0, 0); - - apic_start = apic_read(APIC_TMCCT); -#ifdef CONFIG_X86_PM_TIMER - if (apic_calibrate_pmtmr && pmtmr_ioport) { - pmtimer_wait(5000); /* 5ms wait */ - apic = apic_read(APIC_TMCCT); - result = (apic_start - apic) * 1000L / 5; - } else -#endif - { - rdtscll(tsc_start); - - do { - apic = apic_read(APIC_TMCCT); - rdtscll(tsc); - } while ((tsc - tsc_start) < TICK_COUNT && - (apic_start - apic) < TICK_COUNT); - - result = (apic_start - apic) * 1000L * tsc_khz / - (tsc - tsc_start); - } - - local_irq_enable(); - - printk(KERN_DEBUG "APIC timer calibration result %d\n", result); - - printk(KERN_INFO "Detected %d.%03d MHz APIC timer.\n", - result / 1000 / 1000, result / 1000 % 1000); - - /* Calculate the scaled math multiplication factor */ - lapic_clockevent.mult = div_sc(result, NSEC_PER_SEC, - lapic_clockevent.shift); - lapic_clockevent.max_delta_ns = - clockevent_delta2ns(0x7FFFFF, &lapic_clockevent); - lapic_clockevent.min_delta_ns = - clockevent_delta2ns(0xF, &lapic_clockevent); - - calibration_result = (result * APIC_DIVISOR) / HZ; - - /* - * Do a sanity check on the APIC calibration result - */ - if (calibration_result < (1000000 / HZ)) { - printk(KERN_WARNING - "APIC frequency too slow, disabling apic timer\n"); - return -1; - } - - return 0; -} - -/* - * Setup the boot APIC - * - * Calibrate and verify the result. - */ -void __init setup_boot_APIC_clock(void) -{ - /* - * The local apic timer can be disabled via the kernel - * commandline or from the CPU detection code. Register the lapic - * timer as a dummy clock event source on SMP systems, so the - * broadcast mechanism is used. On UP systems simply ignore it. - */ - if (disable_apic_timer) { - printk(KERN_INFO "Disabling APIC timer\n"); - /* No broadcast on UP ! */ - if (num_possible_cpus() > 1) { - lapic_clockevent.mult = 1; - setup_APIC_timer(); - } - return; - } - - apic_printk(APIC_VERBOSE, "Using local APIC timer interrupts.\n" - "calibrating APIC timer ...\n"); - - if (calibrate_APIC_clock()) { - /* No broadcast on UP ! */ - if (num_possible_cpus() > 1) - setup_APIC_timer(); - return; - } - - /* - * If nmi_watchdog is set to IO_APIC, we need the - * PIT/HPET going. Otherwise register lapic as a dummy - * device. - */ - if (nmi_watchdog != NMI_IO_APIC) - lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY; - else - printk(KERN_WARNING "APIC timer registered as dummy," - " due to nmi_watchdog=%d!\n", nmi_watchdog); - - /* Setup the lapic or request the broadcast */ - setup_APIC_timer(); -} - -void __cpuinit setup_secondary_APIC_clock(void) -{ - setup_APIC_timer(); -} - -/* - * The guts of the apic timer interrupt - */ -static void local_apic_timer_interrupt(void) -{ - int cpu = smp_processor_id(); - struct clock_event_device *evt = &per_cpu(lapic_events, cpu); - - /* - * Normally we should not be here till LAPIC has been initialized but - * in some cases like kdump, its possible that there is a pending LAPIC - * timer interrupt from previous kernel's context and is delivered in - * new kernel the moment interrupts are enabled. - * - * Interrupts are enabled early and LAPIC is setup much later, hence - * its possible that when we get here evt->event_handler is NULL. - * Check for event_handler being NULL and discard the interrupt as - * spurious. - */ - if (!evt->event_handler) { - printk(KERN_WARNING - "Spurious LAPIC timer interrupt on cpu %d\n", cpu); - /* Switch it off */ - lapic_timer_setup(CLOCK_EVT_MODE_SHUTDOWN, evt); - return; - } - - /* - * the NMI deadlock-detector uses this. - */ -#ifdef CONFIG_X86_64 - add_pda(apic_timer_irqs, 1); -#else - per_cpu(irq_stat, cpu).apic_timer_irqs++; -#endif - - evt->event_handler(evt); -} - -/* - * Local APIC timer interrupt. This is the most natural way for doing - * local interrupts, but local timer interrupts can be emulated by - * broadcast interrupts too. [in case the hw doesn't support APIC timers] - * - * [ if a single-CPU system runs an SMP kernel then we call the local - * interrupt as well. Thus we cannot inline the local irq ... ] - */ -void smp_apic_timer_interrupt(struct pt_regs *regs) -{ - struct pt_regs *old_regs = set_irq_regs(regs); - - /* - * NOTE! We'd better ACK the irq immediately, - * because timer handling can be slow. - */ - ack_APIC_irq(); - /* - * update_process_times() expects us to have done irq_enter(). - * Besides, if we don't timer interrupts ignore the global - * interrupt lock, which is the WrongThing (tm) to do. - */ - exit_idle(); - irq_enter(); - local_apic_timer_interrupt(); - irq_exit(); - - set_irq_regs(old_regs); -} - -int setup_profiling_timer(unsigned int multiplier) -{ - return -EINVAL; -} - - -/* - * Local APIC start and shutdown - */ - -/** - * clear_local_APIC - shutdown the local APIC - * - * This is called, when a CPU is disabled and before rebooting, so the state of - * the local APIC has no dangling leftovers. Also used to cleanout any BIOS - * leftovers during boot. - */ -void clear_local_APIC(void) -{ - int maxlvt; - u32 v; - - /* APIC hasn't been mapped yet */ - if (!apic_phys) - return; - - maxlvt = lapic_get_maxlvt(); - /* - * Masking an LVT entry can trigger a local APIC error - * if the vector is zero. Mask LVTERR first to prevent this. - */ - if (maxlvt >= 3) { - v = ERROR_APIC_VECTOR; /* any non-zero vector will do */ - apic_write(APIC_LVTERR, v | APIC_LVT_MASKED); - } - /* - * Careful: we have to set masks only first to deassert - * any level-triggered sources. - */ - v = apic_read(APIC_LVTT); - apic_write(APIC_LVTT, v | APIC_LVT_MASKED); - v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v | APIC_LVT_MASKED); - v = apic_read(APIC_LVT1); - apic_write(APIC_LVT1, v | APIC_LVT_MASKED); - if (maxlvt >= 4) { - v = apic_read(APIC_LVTPC); - apic_write(APIC_LVTPC, v | APIC_LVT_MASKED); - } - - /* lets not touch this if we didn't frob it */ -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(X86_MCE_INTEL) - if (maxlvt >= 5) { - v = apic_read(APIC_LVTTHMR); - apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED); - } -#endif - /* - * Clean APIC state for other OSs: - */ - apic_write(APIC_LVTT, APIC_LVT_MASKED); - apic_write(APIC_LVT0, APIC_LVT_MASKED); - apic_write(APIC_LVT1, APIC_LVT_MASKED); - if (maxlvt >= 3) - apic_write(APIC_LVTERR, APIC_LVT_MASKED); - if (maxlvt >= 4) - apic_write(APIC_LVTPC, APIC_LVT_MASKED); - - /* Integrated APIC (!82489DX) ? */ - if (lapic_is_integrated()) { - if (maxlvt > 3) - /* Clear ESR due to Pentium errata 3AP and 11AP */ - apic_write(APIC_ESR, 0); - apic_read(APIC_ESR); - } -} - -/** - * disable_local_APIC - clear and disable the local APIC - */ -void disable_local_APIC(void) -{ - unsigned int value; - - clear_local_APIC(); - - /* - * Disable APIC (implies clearing of registers - * for 82489DX!). - */ - value = apic_read(APIC_SPIV); - value &= ~APIC_SPIV_APIC_ENABLED; - apic_write(APIC_SPIV, value); - -#ifdef CONFIG_X86_32 - /* - * When LAPIC was disabled by the BIOS and enabled by the kernel, - * restore the disabled state. - */ - if (enabled_via_apicbase) { - unsigned int l, h; - - rdmsr(MSR_IA32_APICBASE, l, h); - l &= ~MSR_IA32_APICBASE_ENABLE; - wrmsr(MSR_IA32_APICBASE, l, h); - } -#endif -} - -/* - * If Linux enabled the LAPIC against the BIOS default disable it down before - * re-entering the BIOS on shutdown. Otherwise the BIOS may get confused and - * not power-off. Additionally clear all LVT entries before disable_local_APIC - * for the case where Linux didn't enable the LAPIC. - */ -void lapic_shutdown(void) -{ - unsigned long flags; - - if (!cpu_has_apic) - return; - - local_irq_save(flags); - -#ifdef CONFIG_X86_32 - if (!enabled_via_apicbase) - clear_local_APIC(); - else -#endif - disable_local_APIC(); - - - local_irq_restore(flags); -} - -/* - * This is to verify that we're looking at a real local APIC. - * Check these against your board if the CPUs aren't getting - * started for no apparent reason. - */ -int __init verify_local_APIC(void) -{ - unsigned int reg0, reg1; - - /* - * The version register is read-only in a real APIC. - */ - reg0 = apic_read(APIC_LVR); - apic_printk(APIC_DEBUG, "Getting VERSION: %x\n", reg0); - apic_write(APIC_LVR, reg0 ^ APIC_LVR_MASK); - reg1 = apic_read(APIC_LVR); - apic_printk(APIC_DEBUG, "Getting VERSION: %x\n", reg1); - - /* - * The two version reads above should print the same - * numbers. If the second one is different, then we - * poke at a non-APIC. - */ - if (reg1 != reg0) - return 0; - - /* - * Check if the version looks reasonably. - */ - reg1 = GET_APIC_VERSION(reg0); - if (reg1 == 0x00 || reg1 == 0xff) - return 0; - reg1 = lapic_get_maxlvt(); - if (reg1 < 0x02 || reg1 == 0xff) - return 0; - - /* - * The ID register is read/write in a real APIC. - */ - reg0 = apic_read(APIC_ID); - apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg0); - apic_write(APIC_ID, reg0 ^ APIC_ID_MASK); - reg1 = apic_read(APIC_ID); - apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg1); - apic_write(APIC_ID, reg0); - if (reg1 != (reg0 ^ APIC_ID_MASK)) - return 0; - - /* - * The next two are just to see if we have sane values. - * They're only really relevant if we're in Virtual Wire - * compatibility mode, but most boxes are anymore. - */ - reg0 = apic_read(APIC_LVT0); - apic_printk(APIC_DEBUG, "Getting LVT0: %x\n", reg0); - reg1 = apic_read(APIC_LVT1); - apic_printk(APIC_DEBUG, "Getting LVT1: %x\n", reg1); - - return 1; -} - -/** - * sync_Arb_IDs - synchronize APIC bus arbitration IDs - */ -void __init sync_Arb_IDs(void) -{ - /* - * Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 And not - * needed on AMD. - */ - if (modern_apic() || boot_cpu_data.x86_vendor == X86_VENDOR_AMD) - return; - - /* - * Wait for idle. - */ - apic_wait_icr_idle(); - - apic_printk(APIC_DEBUG, "Synchronizing Arb IDs.\n"); - apic_write(APIC_ICR, APIC_DEST_ALLINC | - APIC_INT_LEVELTRIG | APIC_DM_INIT); -} - -/* - * An initial setup of the virtual wire mode. - */ -void __init init_bsp_APIC(void) -{ - unsigned int value; - - /* - * Don't do the setup now if we have a SMP BIOS as the - * through-I/O-APIC virtual wire mode might be active. - */ - if (smp_found_config || !cpu_has_apic) - return; - - /* - * Do not trust the local APIC being empty at bootup. - */ - clear_local_APIC(); - - /* - * Enable APIC. - */ - value = apic_read(APIC_SPIV); - value &= ~APIC_VECTOR_MASK; - value |= APIC_SPIV_APIC_ENABLED; - -#ifdef CONFIG_X86_32 - /* This bit is reserved on P4/Xeon and should be cleared */ - if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && - (boot_cpu_data.x86 == 15)) - value &= ~APIC_SPIV_FOCUS_DISABLED; - else -#endif - value |= APIC_SPIV_FOCUS_DISABLED; - value |= SPURIOUS_APIC_VECTOR; - apic_write(APIC_SPIV, value); - - /* - * Set up the virtual wire mode. - */ - apic_write(APIC_LVT0, APIC_DM_EXTINT); - value = APIC_DM_NMI; - if (!lapic_is_integrated()) /* 82489DX */ - value |= APIC_LVT_LEVEL_TRIGGER; - apic_write(APIC_LVT1, value); -} - -static void __cpuinit lapic_setup_esr(void) -{ - unsigned long oldvalue, value, maxlvt; - if (lapic_is_integrated() && !esr_disable) { - if (esr_disable) { - /* - * Something untraceable is creating bad interrupts on - * secondary quads ... for the moment, just leave the - * ESR disabled - we can't do anything useful with the - * errors anyway - mbligh - */ - printk(KERN_INFO "Leaving ESR disabled.\n"); - return; - } - /* !82489DX */ - maxlvt = lapic_get_maxlvt(); - if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ - apic_write(APIC_ESR, 0); - oldvalue = apic_read(APIC_ESR); - - /* enables sending errors */ - value = ERROR_APIC_VECTOR; - apic_write(APIC_LVTERR, value); - /* - * spec says clear errors after enabling vector. - */ - if (maxlvt > 3) - apic_write(APIC_ESR, 0); - value = apic_read(APIC_ESR); - if (value != oldvalue) - apic_printk(APIC_VERBOSE, "ESR value before enabling " - "vector: 0x%08lx after: 0x%08lx\n", - oldvalue, value); - } else { - printk(KERN_INFO "No ESR for 82489DX.\n"); - } -} - - -/** - * setup_local_APIC - setup the local APIC - */ -void __cpuinit setup_local_APIC(void) -{ - unsigned int value; - int i, j; - - preempt_disable(); - value = apic_read(APIC_LVR); - - BUILD_BUG_ON((SPURIOUS_APIC_VECTOR & 0x0f) != 0x0f); - - /* - * Double-check whether this APIC is really registered. - * This is meaningless in clustered apic mode, so we skip it. - */ - if (!apic_id_registered()) - BUG(); - - /* - * Intel recommends to set DFR, LDR and TPR before enabling - * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel - * document number 292116). So here it goes... - */ - init_apic_ldr(); - - /* - * Set Task Priority to 'accept all'. We never change this - * later on. - */ - value = apic_read(APIC_TASKPRI); - value &= ~APIC_TPRI_MASK; - apic_write(APIC_TASKPRI, value); - - /* - * After a crash, we no longer service the interrupts and a pending - * interrupt from previous kernel might still have ISR bit set. - * - * Most probably by now CPU has serviced that pending interrupt and - * it might not have done the ack_APIC_irq() because it thought, - * interrupt came from i8259 as ExtInt. LAPIC did not get EOI so it - * does not clear the ISR bit and cpu thinks it has already serivced - * the interrupt. Hence a vector might get locked. It was noticed - * for timer irq (vector 0x31). Issue an extra EOI to clear ISR. - */ - for (i = APIC_ISR_NR - 1; i >= 0; i--) { - value = apic_read(APIC_ISR + i*0x10); - for (j = 31; j >= 0; j--) { - if (value & (1<<j)) - ack_APIC_irq(); - } - } - - /* - * Now that we are all set up, enable the APIC - */ - value = apic_read(APIC_SPIV); - value &= ~APIC_VECTOR_MASK; - /* - * Enable APIC - */ - value |= APIC_SPIV_APIC_ENABLED; - - /* We always use processor focus */ - - /* - * Set spurious IRQ vector - */ - value |= SPURIOUS_APIC_VECTOR; - apic_write(APIC_SPIV, value); - - /* - * Set up LVT0, LVT1: - * - * set up through-local-APIC on the BP's LINT0. This is not - * strictly necessary in pure symmetric-IO mode, but sometimes - * we delegate interrupts to the 8259A. - */ - /* - * TODO: set up through-local-APIC from through-I/O-APIC? --macro - */ - value = apic_read(APIC_LVT0) & APIC_LVT_MASKED; - if (!smp_processor_id() && !value) { - value = APIC_DM_EXTINT; - apic_printk(APIC_VERBOSE, "enabled ExtINT on CPU#%d\n", - smp_processor_id()); - } else { - value = APIC_DM_EXTINT | APIC_LVT_MASKED; - apic_printk(APIC_VERBOSE, "masked ExtINT on CPU#%d\n", - smp_processor_id()); - } - apic_write(APIC_LVT0, value); - - /* - * only the BP should see the LINT1 NMI signal, obviously. - */ - if (!smp_processor_id()) - value = APIC_DM_NMI; - else - value = APIC_DM_NMI | APIC_LVT_MASKED; - apic_write(APIC_LVT1, value); - preempt_enable(); -} - -void __cpuinit end_local_APIC_setup(void) -{ - lapic_setup_esr(); - -#ifdef CONFIG_X86_32 - { - unsigned int value; - /* Disable the local apic timer */ - value = apic_read(APIC_LVTT); - value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); - apic_write(APIC_LVTT, value); - } -#endif - - setup_apic_nmi_watchdog(NULL); - apic_pm_activate(); -} - -void check_x2apic(void) -{ - int msr, msr2; - - rdmsr(MSR_IA32_APICBASE, msr, msr2); - - if (msr & X2APIC_ENABLE) { - printk("x2apic enabled by BIOS, switching to x2apic ops\n"); - x2apic_preenabled = x2apic = 1; - apic_ops = &x2apic_ops; - } -} - -void enable_x2apic(void) -{ - int msr, msr2; - - rdmsr(MSR_IA32_APICBASE, msr, msr2); - if (!(msr & X2APIC_ENABLE)) { - printk("Enabling x2apic\n"); - wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0); - } -} - -void enable_IR_x2apic(void) -{ -#ifdef CONFIG_INTR_REMAP - int ret; - unsigned long flags; - - if (!cpu_has_x2apic) - return; - - if (!x2apic_preenabled && disable_x2apic) { - printk(KERN_INFO - "Skipped enabling x2apic and Interrupt-remapping " - "because of nox2apic\n"); - return; - } - - if (x2apic_preenabled && disable_x2apic) - panic("Bios already enabled x2apic, can't enforce nox2apic"); - - if (!x2apic_preenabled && skip_ioapic_setup) { - printk(KERN_INFO - "Skipped enabling x2apic and Interrupt-remapping " - "because of skipping io-apic setup\n"); - return; - } - - ret = dmar_table_init(); - if (ret) { - printk(KERN_INFO - "dmar_table_init() failed with %d:\n", ret); - - if (x2apic_preenabled) - panic("x2apic enabled by bios. But IR enabling failed"); - else - printk(KERN_INFO - "Not enabling x2apic,Intr-remapping\n"); - return; - } - - local_irq_save(flags); - mask_8259A(); - save_mask_IO_APIC_setup(); - - ret = enable_intr_remapping(1); - - if (ret && x2apic_preenabled) { - local_irq_restore(flags); - panic("x2apic enabled by bios. But IR enabling failed"); - } - - if (ret) - goto end; - - if (!x2apic) { - x2apic = 1; - apic_ops = &x2apic_ops; - enable_x2apic(); - } -end: - if (ret) - /* - * IR enabling failed - */ - restore_IO_APIC_setup(); - else - reinit_intr_remapped_IO_APIC(x2apic_preenabled); - - unmask_8259A(); - local_irq_restore(flags); - - if (!ret) { - if (!x2apic_preenabled) - printk(KERN_INFO - "Enabled x2apic and interrupt-remapping\n"); - else - printk(KERN_INFO - "Enabled Interrupt-remapping\n"); - } else - printk(KERN_ERR - "Failed to enable Interrupt-remapping and x2apic\n"); -#else - if (!cpu_has_x2apic) - return; - - if (x2apic_preenabled) - panic("x2apic enabled prior OS handover," - " enable CONFIG_INTR_REMAP"); - - printk(KERN_INFO "Enable CONFIG_INTR_REMAP for enabling intr-remapping " - " and x2apic\n"); -#endif - - return; -} - -/* - * Detect and enable local APICs on non-SMP boards. - * Original code written by Keir Fraser. - * On AMD64 we trust the BIOS - if it says no APIC it is likely - * not correctly set up (usually the APIC timer won't work etc.) - */ -static int __init detect_init_APIC(void) -{ - if (!cpu_has_apic) { - printk(KERN_INFO "No local APIC present\n"); - return -1; - } - - mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; - boot_cpu_physical_apicid = 0; - return 0; -} - -void __init early_init_lapic_mapping(void) -{ - unsigned long phys_addr; - - /* - * If no local APIC can be found then go out - * : it means there is no mpatable and MADT - */ - if (!smp_found_config) - return; - - phys_addr = mp_lapic_addr; - - set_fixmap_nocache(FIX_APIC_BASE, phys_addr); - apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", - APIC_BASE, phys_addr); - - /* - * Fetch the APIC ID of the BSP in case we have a - * default configuration (or the MP table is broken). - */ - boot_cpu_physical_apicid = read_apic_id(); -} - -/** - * init_apic_mappings - initialize APIC mappings - */ -void __init init_apic_mappings(void) -{ - if (x2apic) { - boot_cpu_physical_apicid = read_apic_id(); - return; - } - - /* - * If no local APIC can be found then set up a fake all - * zeroes page to simulate the local APIC and another - * one for the IO-APIC. - */ - if (!smp_found_config && detect_init_APIC()) { - apic_phys = (unsigned long) alloc_bootmem_pages(PAGE_SIZE); - apic_phys = __pa(apic_phys); - } else - apic_phys = mp_lapic_addr; - - set_fixmap_nocache(FIX_APIC_BASE, apic_phys); - apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", - APIC_BASE, apic_phys); - - /* - * Fetch the APIC ID of the BSP in case we have a - * default configuration (or the MP table is broken). - */ - boot_cpu_physical_apicid = read_apic_id(); -} - -/* - * This initializes the IO-APIC and APIC hardware if this is - * a UP kernel. - */ -int apic_version[MAX_APICS]; - -int __init APIC_init_uniprocessor(void) -{ - if (disable_apic) { - printk(KERN_INFO "Apic disabled\n"); - return -1; - } - if (!cpu_has_apic) { - disable_apic = 1; - printk(KERN_INFO "Apic disabled by BIOS\n"); - return -1; - } - - enable_IR_x2apic(); - setup_apic_routing(); - - verify_local_APIC(); - - connect_bsp_APIC(); - - physid_set_mask_of_physid(boot_cpu_physical_apicid, &phys_cpu_present_map); - apic_write(APIC_ID, SET_APIC_ID(boot_cpu_physical_apicid)); - - setup_local_APIC(); - - /* - * Now enable IO-APICs, actually call clear_IO_APIC - * We need clear_IO_APIC before enabling vector on BP - */ - if (!skip_ioapic_setup && nr_ioapics) - enable_IO_APIC(); - - if (!smp_found_config || skip_ioapic_setup || !nr_ioapics) - localise_nmi_watchdog(); - end_local_APIC_setup(); - - if (smp_found_config && !skip_ioapic_setup && nr_ioapics) - setup_IO_APIC(); - else - nr_ioapics = 0; - setup_boot_APIC_clock(); - check_nmi_watchdog(); - return 0; -} - -/* - * Local APIC interrupts - */ - -/* - * This interrupt should _never_ happen with our APIC/SMP architecture - */ -asmlinkage void smp_spurious_interrupt(void) -{ - unsigned int v; - exit_idle(); - irq_enter(); - /* - * Check if this really is a spurious interrupt and ACK it - * if it is a vectored one. Just in case... - * Spurious interrupts should not be ACKed. - */ - v = apic_read(APIC_ISR + ((SPURIOUS_APIC_VECTOR & ~0x1f) >> 1)); - if (v & (1 << (SPURIOUS_APIC_VECTOR & 0x1f))) - ack_APIC_irq(); - - add_pda(irq_spurious_count, 1); - irq_exit(); -} - -/* - * This interrupt should never happen with our APIC/SMP architecture - */ -asmlinkage void smp_error_interrupt(void) -{ - unsigned int v, v1; - - exit_idle(); - irq_enter(); - /* First tickle the hardware, only then report what went on. -- REW */ - v = apic_read(APIC_ESR); - apic_write(APIC_ESR, 0); - v1 = apic_read(APIC_ESR); - ack_APIC_irq(); - atomic_inc(&irq_err_count); - - /* Here is what the APIC error bits mean: - 0: Send CS error - 1: Receive CS error - 2: Send accept error - 3: Receive accept error - 4: Reserved - 5: Send illegal vector - 6: Received illegal vector - 7: Illegal register address - */ - printk(KERN_DEBUG "APIC error on CPU%d: %02x(%02x)\n", - smp_processor_id(), v , v1); - irq_exit(); -} - -/** - * connect_bsp_APIC - attach the APIC to the interrupt system - */ -void __init connect_bsp_APIC(void) -{ -#ifdef CONFIG_X86_32 - if (pic_mode) { - /* - * Do not trust the local APIC being empty at bootup. - */ - clear_local_APIC(); - /* - * PIC mode, enable APIC mode in the IMCR, i.e. connect BSP's - * local APIC to INT and NMI lines. - */ - apic_printk(APIC_VERBOSE, "leaving PIC mode, " - "enabling APIC mode.\n"); - outb(0x70, 0x22); - outb(0x01, 0x23); - } -#endif - enable_apic_mode(); -} - -/** - * disconnect_bsp_APIC - detach the APIC from the interrupt system - * @virt_wire_setup: indicates, whether virtual wire mode is selected - * - * Virtual wire mode is necessary to deliver legacy interrupts even when the - * APIC is disabled. - */ -void disconnect_bsp_APIC(int virt_wire_setup) -{ - unsigned int value; - -#ifdef CONFIG_X86_32 - if (pic_mode) { - /* - * Put the board back into PIC mode (has an effect only on - * certain older boards). Note that APIC interrupts, including - * IPIs, won't work beyond this point! The only exception are - * INIT IPIs. - */ - apic_printk(APIC_VERBOSE, "disabling APIC mode, " - "entering PIC mode.\n"); - outb(0x70, 0x22); - outb(0x00, 0x23); - return; - } -#endif - - /* Go back to Virtual Wire compatibility mode */ - - /* For the spurious interrupt use vector F, and enable it */ - value = apic_read(APIC_SPIV); - value &= ~APIC_VECTOR_MASK; - value |= APIC_SPIV_APIC_ENABLED; - value |= 0xf; - apic_write(APIC_SPIV, value); - - if (!virt_wire_setup) { - /* - * For LVT0 make it edge triggered, active high, - * external and enabled - */ - value = apic_read(APIC_LVT0); - value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING | - APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR | - APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED); - value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING; - value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT); - apic_write(APIC_LVT0, value); - } else { - /* Disable LVT0 */ - apic_write(APIC_LVT0, APIC_LVT_MASKED); - } - - /* - * For LVT1 make it edge triggered, active high, - * nmi and enabled - */ - value = apic_read(APIC_LVT1); - value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING | - APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR | - APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED); - value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING; - value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI); - apic_write(APIC_LVT1, value); -} - -void __cpuinit generic_processor_info(int apicid, int version) -{ - int cpu; - cpumask_t tmp_map; - - /* - * Validate version - */ - if (version == 0x0) { - printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! " - "fixing up to 0x10. (tell your hw vendor)\n", - version); - version = 0x10; - } - apic_version[apicid] = version; - - if (num_processors >= NR_CPUS) { - printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached." - " Processor ignored.\n", NR_CPUS); - return; - } - - num_processors++; - cpus_complement(tmp_map, cpu_present_map); - cpu = first_cpu(tmp_map); - - physid_set(apicid, phys_cpu_present_map); - if (apicid == boot_cpu_physical_apicid) { - /* - * x86_bios_cpu_apicid is required to have processors listed - * in same order as logical cpu numbers. Hence the first - * entry is BSP, and so on. - */ - cpu = 0; - } - if (apicid > max_physical_apicid) - max_physical_apicid = apicid; - -#ifdef CONFIG_X86_32 - /* - * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y - * but we need to work other dependencies like SMP_SUSPEND etc - * before this can be done without some confusion. - * if (CPU_HOTPLUG_ENABLED || num_processors > 8) - * - Ashok Raj <ashok.raj@intel.com> - */ - if (max_physical_apicid >= 8) { - switch (boot_cpu_data.x86_vendor) { - case X86_VENDOR_INTEL: - if (!APIC_XAPIC(version)) { - def_to_bigsmp = 0; - break; - } - /* If P4 and above fall through */ - case X86_VENDOR_AMD: - def_to_bigsmp = 1; - } - } -#endif - -#if defined(CONFIG_X86_SMP) || defined(CONFIG_X86_64) - /* are we being called early in kernel startup? */ - if (early_per_cpu_ptr(x86_cpu_to_apicid)) { - u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid); - u16 *bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); - - cpu_to_apicid[cpu] = apicid; - bios_cpu_apicid[cpu] = apicid; - } else { - per_cpu(x86_cpu_to_apicid, cpu) = apicid; - per_cpu(x86_bios_cpu_apicid, cpu) = apicid; - } -#endif - - cpu_set(cpu, cpu_possible_map); - cpu_set(cpu, cpu_present_map); -} - -int hard_smp_processor_id(void) -{ - return read_apic_id(); -} - -/* - * Power management - */ -#ifdef CONFIG_PM - -static struct { - /* - * 'active' is true if the local APIC was enabled by us and - * not the BIOS; this signifies that we are also responsible - * for disabling it before entering apm/acpi suspend - */ - int active; - /* r/w apic fields */ - unsigned int apic_id; - unsigned int apic_taskpri; - unsigned int apic_ldr; - unsigned int apic_dfr; - unsigned int apic_spiv; - unsigned int apic_lvtt; - unsigned int apic_lvtpc; - unsigned int apic_lvt0; - unsigned int apic_lvt1; - unsigned int apic_lvterr; - unsigned int apic_tmict; - unsigned int apic_tdcr; - unsigned int apic_thmr; -} apic_pm_state; - -static int lapic_suspend(struct sys_device *dev, pm_message_t state) -{ - unsigned long flags; - int maxlvt; - - if (!apic_pm_state.active) - return 0; - - maxlvt = lapic_get_maxlvt(); - - apic_pm_state.apic_id = apic_read(APIC_ID); - apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI); - apic_pm_state.apic_ldr = apic_read(APIC_LDR); - apic_pm_state.apic_dfr = apic_read(APIC_DFR); - apic_pm_state.apic_spiv = apic_read(APIC_SPIV); - apic_pm_state.apic_lvtt = apic_read(APIC_LVTT); - if (maxlvt >= 4) - apic_pm_state.apic_lvtpc = apic_read(APIC_LVTPC); - apic_pm_state.apic_lvt0 = apic_read(APIC_LVT0); - apic_pm_state.apic_lvt1 = apic_read(APIC_LVT1); - apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR); - apic_pm_state.apic_tmict = apic_read(APIC_TMICT); - apic_pm_state.apic_tdcr = apic_read(APIC_TDCR); -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) - if (maxlvt >= 5) - apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); -#endif - - local_irq_save(flags); - disable_local_APIC(); - local_irq_restore(flags); - return 0; -} - -static int lapic_resume(struct sys_device *dev) -{ - unsigned int l, h; - unsigned long flags; - int maxlvt; - - if (!apic_pm_state.active) - return 0; - - maxlvt = lapic_get_maxlvt(); - - local_irq_save(flags); - -#ifdef CONFIG_X86_64 - if (x2apic) - enable_x2apic(); - else -#endif - { - /* - * Make sure the APICBASE points to the right address - * - * FIXME! This will be wrong if we ever support suspend on - * SMP! We'll need to do this as part of the CPU restore! - */ - rdmsr(MSR_IA32_APICBASE, l, h); - l &= ~MSR_IA32_APICBASE_BASE; - l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr; - wrmsr(MSR_IA32_APICBASE, l, h); - } - - apic_write(APIC_LVTERR, ERROR_APIC_VECTOR | APIC_LVT_MASKED); - apic_write(APIC_ID, apic_pm_state.apic_id); - apic_write(APIC_DFR, apic_pm_state.apic_dfr); - apic_write(APIC_LDR, apic_pm_state.apic_ldr); - apic_write(APIC_TASKPRI, apic_pm_state.apic_taskpri); - apic_write(APIC_SPIV, apic_pm_state.apic_spiv); - apic_write(APIC_LVT0, apic_pm_state.apic_lvt0); - apic_write(APIC_LVT1, apic_pm_state.apic_lvt1); -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) - if (maxlvt >= 5) - apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr); -#endif - if (maxlvt >= 4) - apic_write(APIC_LVTPC, apic_pm_state.apic_lvtpc); - apic_write(APIC_LVTT, apic_pm_state.apic_lvtt); - apic_write(APIC_TDCR, apic_pm_state.apic_tdcr); - apic_write(APIC_TMICT, apic_pm_state.apic_tmict); - apic_write(APIC_ESR, 0); - apic_read(APIC_ESR); - apic_write(APIC_LVTERR, apic_pm_state.apic_lvterr); - apic_write(APIC_ESR, 0); - apic_read(APIC_ESR); - - local_irq_restore(flags); - - return 0; -} - -/* - * This device has no shutdown method - fully functioning local APICs - * are needed on every CPU up until machine_halt/restart/poweroff. - */ - -static struct sysdev_class lapic_sysclass = { - .name = "lapic", - .resume = lapic_resume, - .suspend = lapic_suspend, -}; - -static struct sys_device device_lapic = { - .id = 0, - .cls = &lapic_sysclass, -}; - -static void __cpuinit apic_pm_activate(void) -{ - apic_pm_state.active = 1; -} - -static int __init init_lapic_sysfs(void) -{ - int error; - - if (!cpu_has_apic) - return 0; - /* XXX: remove suspend/resume procs if !apic_pm_state.active? */ - - error = sysdev_class_register(&lapic_sysclass); - if (!error) - error = sysdev_register(&device_lapic); - return error; -} -device_initcall(init_lapic_sysfs); - -#else /* CONFIG_PM */ - -static void apic_pm_activate(void) { } - -#endif /* CONFIG_PM */ - -/* - * apic_is_clustered_box() -- Check if we can expect good TSC - * - * Thus far, the major user of this is IBM's Summit2 series: - * - * Clustered boxes may have unsynced TSC problems if they are - * multi-chassis. Use available data to take a good guess. - * If in doubt, go HPET. - */ -__cpuinit int apic_is_clustered_box(void) -{ - int i, clusters, zeros; - unsigned id; - u16 *bios_cpu_apicid; - DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS); - - /* - * there is not this kind of box with AMD CPU yet. - * Some AMD box with quadcore cpu and 8 sockets apicid - * will be [4, 0x23] or [8, 0x27] could be thought to - * vsmp box still need checking... - */ - if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && !is_vsmp_box()) - return 0; - - bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); - bitmap_zero(clustermap, NUM_APIC_CLUSTERS); - - for (i = 0; i < NR_CPUS; i++) { - /* are we being called early in kernel startup? */ - if (bios_cpu_apicid) { - id = bios_cpu_apicid[i]; - } - else if (i < nr_cpu_ids) { - if (cpu_present(i)) - id = per_cpu(x86_bios_cpu_apicid, i); - else - continue; - } - else - break; - - if (id != BAD_APICID) - __set_bit(APIC_CLUSTERID(id), clustermap); - } - - /* Problem: Partially populated chassis may not have CPUs in some of - * the APIC clusters they have been allocated. Only present CPUs have - * x86_bios_cpu_apicid entries, thus causing zeroes in the bitmap. - * Since clusters are allocated sequentially, count zeros only if - * they are bounded by ones. - */ - clusters = 0; - zeros = 0; - for (i = 0; i < NUM_APIC_CLUSTERS; i++) { - if (test_bit(i, clustermap)) { - clusters += 1 + zeros; - zeros = 0; - } else - ++zeros; - } - - /* ScaleMP vSMPowered boxes have one cluster per board and TSCs are - * not guaranteed to be synced between boards - */ - if (is_vsmp_box() && clusters > 1) - return 1; - - /* - * If clusters > 2, then should be multi-chassis. - * May have to revisit this when multi-core + hyperthreaded CPUs come - * out, but AFAIK this will work even for them. - */ - return (clusters > 2); -} - -static __init int setup_nox2apic(char *str) -{ - disable_x2apic = 1; - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_X2APIC); - return 0; -} -early_param("nox2apic", setup_nox2apic); - - -/* - * APIC command line parameters - */ -static int __init setup_disableapic(char *arg) -{ - disable_apic = 1; - setup_clear_cpu_cap(X86_FEATURE_APIC); - return 0; -} -early_param("disableapic", setup_disableapic); - -/* same as disableapic, for compatibility */ -static int __init setup_nolapic(char *arg) -{ - return setup_disableapic(arg); -} -early_param("nolapic", setup_nolapic); - -static int __init parse_lapic_timer_c2_ok(char *arg) -{ - local_apic_timer_c2_ok = 1; - return 0; -} -early_param("lapic_timer_c2_ok", parse_lapic_timer_c2_ok); - -static int __init parse_disable_apic_timer(char *arg) -{ - disable_apic_timer = 1; - return 0; -} -early_param("noapictimer", parse_disable_apic_timer); - -static int __init parse_nolapic_timer(char *arg) -{ - disable_apic_timer = 1; - return 0; -} -early_param("nolapic_timer", parse_nolapic_timer); - -static __init int setup_apicpmtimer(char *s) -{ - apic_calibrate_pmtmr = 1; - notsc_setup(NULL); - return 0; -} -__setup("apicpmtimer", setup_apicpmtimer); - -static int __init apic_set_verbosity(char *arg) -{ - if (!arg) { -#ifdef CONFIG_X86_64 - skip_ioapic_setup = 0; - ioapic_force = 1; - return 0; -#endif - return -EINVAL; - } - - if (strcmp("debug", arg) == 0) - apic_verbosity = APIC_DEBUG; - else if (strcmp("verbose", arg) == 0) - apic_verbosity = APIC_VERBOSE; - else { - printk(KERN_WARNING "APIC Verbosity level %s not recognised" - " use apic=verbose or apic=debug\n", arg); - return -EINVAL; - } - - return 0; -} -early_param("apic", apic_set_verbosity); - -static int __init lapic_insert_resource(void) -{ - if (!apic_phys) - return -1; - - /* Put local APIC into the resource map. */ - lapic_resource.start = apic_phys; - lapic_resource.end = lapic_resource.start + PAGE_SIZE - 1; - insert_resource(&iomem_resource, &lapic_resource); - - return 0; -} - -/* - * need call insert after e820_reserve_resources() - * that is using request_resource - */ -late_initcall(lapic_insert_resource); diff --git a/arch/x86/kernel/bios_uv.c b/arch/x86/kernel/bios_uv.c index fdd585f9c53..f0dfe6f17e7 100644 --- a/arch/x86/kernel/bios_uv.c +++ b/arch/x86/kernel/bios_uv.c @@ -1,8 +1,6 @@ /* * BIOS run time interface routines. * - * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved. - * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or @@ -16,33 +14,128 @@ * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) Russ Anderson */ +#include <linux/efi.h> +#include <asm/efi.h> +#include <linux/io.h> #include <asm/uv/bios.h> +#include <asm/uv/uv_hub.h> + +struct uv_systab uv_systab; -const char * -x86_bios_strerror(long status) +s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5) { - const char *str; - switch (status) { - case 0: str = "Call completed without error"; break; - case -1: str = "Not implemented"; break; - case -2: str = "Invalid argument"; break; - case -3: str = "Call completed with error"; break; - default: str = "Unknown BIOS status code"; break; - } - return str; + struct uv_systab *tab = &uv_systab; + + if (!tab->function) + /* + * BIOS does not support UV systab + */ + return BIOS_STATUS_UNIMPLEMENTED; + + return efi_call6((void *)__va(tab->function), + (u64)which, a1, a2, a3, a4, a5); } -long -x86_bios_freq_base(unsigned long which, unsigned long *ticks_per_second, - unsigned long *drift_info) +s64 uv_bios_call_irqsave(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, + u64 a4, u64 a5) { - struct uv_bios_retval isrv; + unsigned long bios_flags; + s64 ret; - BIOS_CALL(isrv, BIOS_FREQ_BASE, which, 0, 0, 0, 0, 0, 0); - *ticks_per_second = isrv.v0; - *drift_info = isrv.v1; - return isrv.status; + local_irq_save(bios_flags); + ret = uv_bios_call(which, a1, a2, a3, a4, a5); + local_irq_restore(bios_flags); + + return ret; } -EXPORT_SYMBOL_GPL(x86_bios_freq_base); + +s64 uv_bios_call_reentrant(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, + u64 a4, u64 a5) +{ + s64 ret; + + preempt_disable(); + ret = uv_bios_call(which, a1, a2, a3, a4, a5); + preempt_enable(); + + return ret; +} + + +long sn_partition_id; +EXPORT_SYMBOL_GPL(sn_partition_id); +long uv_coherency_id; +EXPORT_SYMBOL_GPL(uv_coherency_id); +long uv_region_size; +EXPORT_SYMBOL_GPL(uv_region_size); +int uv_type; + + +s64 uv_bios_get_sn_info(int fc, int *uvtype, long *partid, long *coher, + long *region) +{ + s64 ret; + u64 v0, v1; + union partition_info_u part; + + ret = uv_bios_call_irqsave(UV_BIOS_GET_SN_INFO, fc, + (u64)(&v0), (u64)(&v1), 0, 0); + if (ret != BIOS_STATUS_SUCCESS) + return ret; + + part.val = v0; + if (uvtype) + *uvtype = part.hub_version; + if (partid) + *partid = part.partition_id; + if (coher) + *coher = part.coherence_id; + if (region) + *region = part.region_size; + return ret; +} + + +s64 uv_bios_freq_base(u64 clock_type, u64 *ticks_per_second) +{ + return uv_bios_call(UV_BIOS_FREQ_BASE, clock_type, + (u64)ticks_per_second, 0, 0, 0); +} +EXPORT_SYMBOL_GPL(uv_bios_freq_base); + + +#ifdef CONFIG_EFI +void uv_bios_init(void) +{ + struct uv_systab *tab; + + if ((efi.uv_systab == EFI_INVALID_TABLE_ADDR) || + (efi.uv_systab == (unsigned long)NULL)) { + printk(KERN_CRIT "No EFI UV System Table.\n"); + uv_systab.function = (unsigned long)NULL; + return; + } + + tab = (struct uv_systab *)ioremap(efi.uv_systab, + sizeof(struct uv_systab)); + if (strncmp(tab->signature, "UVST", 4) != 0) + printk(KERN_ERR "bad signature in UV system table!"); + + /* + * Copy table to permanent spot for later use. + */ + memcpy(&uv_systab, tab, sizeof(struct uv_systab)); + iounmap(tab); + + printk(KERN_INFO "EFI UV System Table Revision %d\n", tab->revision); +} +#else /* !CONFIG_EFI */ + +void uv_bios_init(void) { } +#endif + diff --git a/arch/x86/kernel/cpu/.gitignore b/arch/x86/kernel/cpu/.gitignore new file mode 100644 index 00000000000..667df55a439 --- /dev/null +++ b/arch/x86/kernel/cpu/.gitignore @@ -0,0 +1 @@ +capflags.c diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 32e73520adf..8f1e31db2ad 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -249,7 +249,7 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) } numa_set_node(cpu, node); - printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node); + printk(KERN_INFO "CPU %d/0x%x -> Node %d\n", cpu, apicid, node); #endif } diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index c24c4a487b7..8e48c5d4467 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -780,6 +780,9 @@ static int __init acpi_cpufreq_init(void) { int ret; + if (acpi_disabled) + return 0; + dprintk("acpi_cpufreq_init\n"); ret = acpi_cpufreq_early_init(); diff --git a/arch/x86/kernel/cpu/cpufreq/longhaul.c b/arch/x86/kernel/cpu/cpufreq/longhaul.c index 06fcce516d5..b0461856acf 100644 --- a/arch/x86/kernel/cpu/cpufreq/longhaul.c +++ b/arch/x86/kernel/cpu/cpufreq/longhaul.c @@ -1,5 +1,5 @@ /* - * (C) 2001-2004 Dave Jones. <davej@codemonkey.org.uk> + * (C) 2001-2004 Dave Jones. <davej@redhat.com> * (C) 2002 Padraig Brady. <padraig@antefacto.com> * * Licensed under the terms of the GNU GPL License version 2. @@ -1019,7 +1019,7 @@ MODULE_PARM_DESC(scale_voltage, "Scale voltage of processor"); module_param(revid_errata, int, 0644); MODULE_PARM_DESC(revid_errata, "Ignore CPU Revision ID"); -MODULE_AUTHOR ("Dave Jones <davej@codemonkey.org.uk>"); +MODULE_AUTHOR ("Dave Jones <davej@redhat.com>"); MODULE_DESCRIPTION ("Longhaul driver for VIA Cyrix processors."); MODULE_LICENSE ("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c index b5ced806a31..c1ac5790c63 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c @@ -246,7 +246,7 @@ static void __exit powernow_k6_exit(void) } -MODULE_AUTHOR("Arjan van de Ven <arjanv@redhat.com>, Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>"); +MODULE_AUTHOR("Arjan van de Ven, Dave Jones <davej@redhat.com>, Dominik Brodowski <linux@brodo.de>"); MODULE_DESCRIPTION("PowerNow! driver for AMD K6-2+ / K6-3+ processors."); MODULE_LICENSE("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c index 0a61159d7b7..7c7d56b4313 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c @@ -1,6 +1,6 @@ /* * AMD K7 Powernow driver. - * (C) 2003 Dave Jones <davej@codemonkey.org.uk> on behalf of SuSE Labs. + * (C) 2003 Dave Jones on behalf of SuSE Labs. * (C) 2003-2004 Dave Jones <davej@redhat.com> * * Licensed under the terms of the GNU GPL License version 2. @@ -692,7 +692,7 @@ static void __exit powernow_exit (void) module_param(acpi_force, int, 0444); MODULE_PARM_DESC(acpi_force, "Force ACPI to be used."); -MODULE_AUTHOR ("Dave Jones <davej@codemonkey.org.uk>"); +MODULE_AUTHOR ("Dave Jones <davej@redhat.com>"); MODULE_DESCRIPTION ("Powernow driver for AMD K7 processors."); MODULE_LICENSE ("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index 84bb395038d..d3dcd58b87c 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -7,7 +7,7 @@ * Support : mark.langsdorf@amd.com * * Based on the powernow-k7.c module written by Dave Jones. - * (C) 2003 Dave Jones <davej@codemonkey.org.uk> on behalf of SuSE Labs + * (C) 2003 Dave Jones on behalf of SuSE Labs * (C) 2004 Dominik Brodowski <linux@brodo.de> * (C) 2004 Pavel Machek <pavel@suse.cz> * Licensed under the terms of the GNU GPL License version 2. @@ -45,7 +45,6 @@ #endif #define PFX "powernow-k8: " -#define BFX PFX "BIOS error: " #define VERSION "version 2.20.00" #include "powernow-k8.h" @@ -536,35 +535,40 @@ static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst, u8 for (j = 0; j < data->numps; j++) { if (pst[j].vid > LEAST_VID) { - printk(KERN_ERR PFX "vid %d invalid : 0x%x\n", j, pst[j].vid); + printk(KERN_ERR FW_BUG PFX "vid %d invalid : 0x%x\n", + j, pst[j].vid); return -EINVAL; } if (pst[j].vid < data->rvo) { /* vid + rvo >= 0 */ - printk(KERN_ERR BFX "0 vid exceeded with pstate %d\n", j); + printk(KERN_ERR FW_BUG PFX "0 vid exceeded with pstate" + " %d\n", j); return -ENODEV; } if (pst[j].vid < maxvid + data->rvo) { /* vid + rvo >= maxvid */ - printk(KERN_ERR BFX "maxvid exceeded with pstate %d\n", j); + printk(KERN_ERR FW_BUG PFX "maxvid exceeded with pstate" + " %d\n", j); return -ENODEV; } if (pst[j].fid > MAX_FID) { - printk(KERN_ERR BFX "maxfid exceeded with pstate %d\n", j); + printk(KERN_ERR FW_BUG PFX "maxfid exceeded with pstate" + " %d\n", j); return -ENODEV; } if (j && (pst[j].fid < HI_FID_TABLE_BOTTOM)) { /* Only first fid is allowed to be in "low" range */ - printk(KERN_ERR BFX "two low fids - %d : 0x%x\n", j, pst[j].fid); + printk(KERN_ERR FW_BUG PFX "two low fids - %d : " + "0x%x\n", j, pst[j].fid); return -EINVAL; } if (pst[j].fid < lastfid) lastfid = pst[j].fid; } if (lastfid & 1) { - printk(KERN_ERR BFX "lastfid invalid\n"); + printk(KERN_ERR FW_BUG PFX "lastfid invalid\n"); return -EINVAL; } if (lastfid > LO_FID_TABLE_TOP) - printk(KERN_INFO BFX "first fid not from lo freq table\n"); + printk(KERN_INFO FW_BUG PFX "first fid not from lo freq table\n"); return 0; } @@ -672,13 +676,13 @@ static int find_psb_table(struct powernow_k8_data *data) dprintk("table vers: 0x%x\n", psb->tableversion); if (psb->tableversion != PSB_VERSION_1_4) { - printk(KERN_ERR BFX "PSB table is not v1.4\n"); + printk(KERN_ERR FW_BUG PFX "PSB table is not v1.4\n"); return -ENODEV; } dprintk("flags: 0x%x\n", psb->flags1); if (psb->flags1) { - printk(KERN_ERR BFX "unknown flags\n"); + printk(KERN_ERR FW_BUG PFX "unknown flags\n"); return -ENODEV; } @@ -705,7 +709,7 @@ static int find_psb_table(struct powernow_k8_data *data) } } if (cpst != 1) { - printk(KERN_ERR BFX "numpst must be 1\n"); + printk(KERN_ERR FW_BUG PFX "numpst must be 1\n"); return -ENODEV; } @@ -1130,17 +1134,19 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) "ACPI Processor module before starting this " "driver.\n"); #else - printk(KERN_ERR PFX "Your BIOS does not provide ACPI " - "_PSS objects in a way that Linux understands. " - "Please report this to the Linux ACPI maintainers" - " and complain to your BIOS vendor.\n"); + printk(KERN_ERR FW_BUG PFX "Your BIOS does not provide" + " ACPI _PSS objects in a way that Linux " + "understands. Please report this to the Linux " + "ACPI maintainers and complain to your BIOS " + "vendor.\n"); #endif kfree(data); return -ENODEV; } if (pol->cpu != 0) { - printk(KERN_ERR PFX "No ACPI _PSS objects for CPU other than " - "CPU0. Complain to your BIOS vendor.\n"); + printk(KERN_ERR FW_BUG PFX "No ACPI _PSS objects for " + "CPU other than CPU0. Complain to your BIOS " + "vendor.\n"); kfree(data); return -ENODEV; } @@ -1193,7 +1199,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) /* min/max the cpu is capable of */ if (cpufreq_frequency_table_cpuinfo(pol, data->powernow_table)) { - printk(KERN_ERR PFX "invalid powernow_table\n"); + printk(KERN_ERR FW_BUG PFX "invalid powernow_table\n"); powernow_k8_cpu_exit_acpi(data); kfree(data->powernow_table); kfree(data); diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c index 191f7263c61..04d0376b64b 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c @@ -431,7 +431,7 @@ static void __exit speedstep_exit(void) } -MODULE_AUTHOR ("Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>"); +MODULE_AUTHOR ("Dave Jones <davej@redhat.com>, Dominik Brodowski <linux@brodo.de>"); MODULE_DESCRIPTION ("Speedstep driver for Intel mobile processors on chipsets with ICH-M southbridges."); MODULE_LICENSE ("GPL"); diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 99468dbd08d..cce0b6118d5 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -174,7 +174,7 @@ static void __cpuinit srat_detect_node(void) node = first_node(node_online_map); numa_set_node(cpu, node); - printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node); + printk(KERN_INFO "CPU %d/0x%x -> Node %d\n", cpu, apicid, node); #endif } diff --git a/arch/x86/kernel/cpu/mcheck/k7.c b/arch/x86/kernel/cpu/mcheck/k7.c index f390c9f6635..dd3af6e7b39 100644 --- a/arch/x86/kernel/cpu/mcheck/k7.c +++ b/arch/x86/kernel/cpu/mcheck/k7.c @@ -1,6 +1,6 @@ /* - * Athlon/Hammer specific Machine Check Exception Reporting - * (C) Copyright 2002 Dave Jones <davej@codemonkey.org.uk> + * Athlon specific Machine Check Exception Reporting + * (C) Copyright 2002 Dave Jones <davej@redhat.com> */ #include <linux/init.h> diff --git a/arch/x86/kernel/cpu/mcheck/mce_32.c b/arch/x86/kernel/cpu/mcheck/mce_32.c index 774d87cfd8c..0ebf3fc6a61 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_32.c +++ b/arch/x86/kernel/cpu/mcheck/mce_32.c @@ -1,6 +1,6 @@ /* * mce.c - x86 Machine Check Exception Reporting - * (c) 2002 Alan Cox <alan@redhat.com>, Dave Jones <davej@codemonkey.org.uk> + * (c) 2002 Alan Cox <alan@redhat.com>, Dave Jones <davej@redhat.com> */ #include <linux/init.h> diff --git a/arch/x86/kernel/cpu/mcheck/non-fatal.c b/arch/x86/kernel/cpu/mcheck/non-fatal.c index cc1fccdd31e..a74af128efc 100644 --- a/arch/x86/kernel/cpu/mcheck/non-fatal.c +++ b/arch/x86/kernel/cpu/mcheck/non-fatal.c @@ -1,7 +1,7 @@ /* * Non Fatal Machine Check Exception Reporting * - * (C) Copyright 2002 Dave Jones. <davej@codemonkey.org.uk> + * (C) Copyright 2002 Dave Jones. <davej@redhat.com> * * This file contains routines to check for non-fatal MCEs every 15s * diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index 6bff382094f..9abd48b2267 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -17,6 +17,8 @@ #include <linux/bitops.h> #include <linux/smp.h> #include <linux/nmi.h> +#include <linux/kprobes.h> + #include <asm/apic.h> #include <asm/intel_arch_perfmon.h> @@ -336,7 +338,8 @@ static void single_msr_unreserve(void) release_perfctr_nmi(wd_ops->perfctr); } -static void single_msr_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) +static void __kprobes +single_msr_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) { /* start the cycle over again */ write_watchdog_counter(wd->perfctr_msr, NULL, nmi_hz); @@ -401,7 +404,7 @@ static int setup_p6_watchdog(unsigned nmi_hz) return 1; } -static void p6_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) +static void __kprobes p6_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) { /* * P6 based Pentium M need to re-unmask @@ -605,7 +608,7 @@ static void p4_unreserve(void) release_perfctr_nmi(MSR_P4_IQ_PERFCTR0); } -static void p4_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) +static void __kprobes p4_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) { unsigned dummy; /* @@ -784,7 +787,7 @@ unsigned lapic_adjust_nmi_hz(unsigned hz) return hz; } -int lapic_wd_event(unsigned nmi_hz) +int __kprobes lapic_wd_event(unsigned nmi_hz) { struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); u64 ctr; diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c index 6a44d646599..72cefd1e649 100644 --- a/arch/x86/kernel/cpuid.c +++ b/arch/x86/kernel/cpuid.c @@ -147,8 +147,8 @@ static __cpuinit int cpuid_device_create(int cpu) { struct device *dev; - dev = device_create_drvdata(cpuid_class, NULL, MKDEV(CPUID_MAJOR, cpu), - NULL, "cpu%d", cpu); + dev = device_create(cpuid_class, NULL, MKDEV(CPUID_MAJOR, cpu), NULL, + "cpu%d", cpu); return IS_ERR(dev) ? PTR_ERR(dev) : 0; } diff --git a/arch/x86/kernel/crash_dump_32.c b/arch/x86/kernel/crash_dump_32.c index 72d0c56c1b4..f7cdb3b457a 100644 --- a/arch/x86/kernel/crash_dump_32.c +++ b/arch/x86/kernel/crash_dump_32.c @@ -13,6 +13,9 @@ static void *kdump_buf_page; +/* Stores the physical address of elf header of crash image. */ +unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX; + /** * copy_oldmem_page - copy one page from "oldmem" * @pfn: page frame number to be copied diff --git a/arch/x86/kernel/crash_dump_64.c b/arch/x86/kernel/crash_dump_64.c index e90a60ef10c..045b36cada6 100644 --- a/arch/x86/kernel/crash_dump_64.c +++ b/arch/x86/kernel/crash_dump_64.c @@ -10,6 +10,9 @@ #include <linux/uaccess.h> #include <linux/io.h> +/* Stores the physical address of elf header of crash image. */ +unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX; + /** * copy_oldmem_page - copy one page from "oldmem" * @pfn: page frame number to be copied diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c index 201ee359a1a..1a78180f08d 100644 --- a/arch/x86/kernel/dumpstack_32.c +++ b/arch/x86/kernel/dumpstack_32.c @@ -13,6 +13,7 @@ #include <linux/kexec.h> #include <linux/bug.h> #include <linux/nmi.h> +#include <linux/sysfs.h> #include <asm/stacktrace.h> @@ -343,6 +344,7 @@ int __kprobes __die(const char *str, struct pt_regs *regs, long err) printk("DEBUG_PAGEALLOC"); #endif printk("\n"); + sysfs_printk_last_file(); if (notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) return 1; diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c index 086cc8118e3..96a5db7da8a 100644 --- a/arch/x86/kernel/dumpstack_64.c +++ b/arch/x86/kernel/dumpstack_64.c @@ -13,6 +13,7 @@ #include <linux/kexec.h> #include <linux/bug.h> #include <linux/nmi.h> +#include <linux/sysfs.h> #include <asm/stacktrace.h> @@ -489,6 +490,7 @@ int __kprobes __die(const char *str, struct pt_regs *regs, long err) printk("DEBUG_PAGEALLOC"); #endif printk("\n"); + sysfs_printk_last_file(); if (notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) return 1; diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 78e642feac3..ce97bf3bed1 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -1282,12 +1282,10 @@ void __init e820_reserve_resources(void) e820_res = res; for (i = 0; i < e820.nr_map; i++) { end = e820.map[i].addr + e820.map[i].size - 1; -#ifndef CONFIG_RESOURCES_64BIT - if (end > 0x100000000ULL) { + if (end != (resource_size_t)end) { res++; continue; } -#endif res->name = e820_type_to_string(e820.map[i].type); res->start = e820.map[i].addr; res->end = end; diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index 945a31cdd81..1119d247fe1 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -367,6 +367,10 @@ void __init efi_init(void) efi.smbios = config_tables[i].table; printk(" SMBIOS=0x%lx ", config_tables[i].table); } else if (!efi_guidcmp(config_tables[i].guid, + UV_SYSTEM_TABLE_GUID)) { + efi.uv_systab = config_tables[i].table; + printk(" UVsystab=0x%lx ", config_tables[i].table); + } else if (!efi_guidcmp(config_tables[i].guid, HCDP_TABLE_GUID)) { efi.hcdp = config_tables[i].table; printk(" HCDP=0x%lx ", config_tables[i].table); diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index b21fbfaffe3..c356423a602 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -629,7 +629,7 @@ ENTRY(interrupt) ENTRY(irq_entries_start) RING0_INT_FRAME vector=0 -.rept NR_IRQS +.rept NR_VECTORS ALIGN .if vector CFI_ADJUST_CFA_OFFSET -4 @@ -1153,20 +1153,6 @@ ENDPROC(xen_failsafe_callback) #ifdef CONFIG_DYNAMIC_FTRACE ENTRY(mcount) - pushl %eax - pushl %ecx - pushl %edx - movl 0xc(%esp), %eax - subl $MCOUNT_INSN_SIZE, %eax - -.globl mcount_call -mcount_call: - call ftrace_stub - - popl %edx - popl %ecx - popl %eax - ret END(mcount) diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 1db6ce4314e..09e7145484c 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -64,32 +64,6 @@ #ifdef CONFIG_FTRACE #ifdef CONFIG_DYNAMIC_FTRACE ENTRY(mcount) - - subq $0x38, %rsp - movq %rax, (%rsp) - movq %rcx, 8(%rsp) - movq %rdx, 16(%rsp) - movq %rsi, 24(%rsp) - movq %rdi, 32(%rsp) - movq %r8, 40(%rsp) - movq %r9, 48(%rsp) - - movq 0x38(%rsp), %rdi - subq $MCOUNT_INSN_SIZE, %rdi - -.globl mcount_call -mcount_call: - call ftrace_stub - - movq 48(%rsp), %r9 - movq 40(%rsp), %r8 - movq 32(%rsp), %rdi - movq 24(%rsp), %rsi - movq 16(%rsp), %rdx - movq 8(%rsp), %rcx - movq (%rsp), %rax - addq $0x38, %rsp - retq END(mcount) diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index ab115cd15fd..d073d981a73 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -11,17 +11,18 @@ #include <linux/spinlock.h> #include <linux/hardirq.h> +#include <linux/uaccess.h> #include <linux/ftrace.h> #include <linux/percpu.h> #include <linux/init.h> #include <linux/list.h> -#include <asm/alternative.h> #include <asm/ftrace.h> +#include <asm/nops.h> /* Long is fine, even if it is only 4 bytes ;-) */ -static long *ftrace_nop; +static unsigned long *ftrace_nop; union ftrace_code_union { char code[MCOUNT_INSN_SIZE]; @@ -60,11 +61,7 @@ notrace int ftrace_modify_code(unsigned long ip, unsigned char *old_code, unsigned char *new_code) { - unsigned replaced; - unsigned old = *(unsigned *)old_code; /* 4 bytes */ - unsigned new = *(unsigned *)new_code; /* 4 bytes */ - unsigned char newch = new_code[4]; - int faulted = 0; + unsigned char replaced[MCOUNT_INSN_SIZE]; /* * Note: Due to modules and __init, code can @@ -72,29 +69,20 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code, * as well as code changing. * * No real locking needed, this code is run through - * kstop_machine. + * kstop_machine, or before SMP starts. */ - asm volatile ( - "1: lock\n" - " cmpxchg %3, (%2)\n" - " jnz 2f\n" - " movb %b4, 4(%2)\n" - "2:\n" - ".section .fixup, \"ax\"\n" - "3: movl $1, %0\n" - " jmp 2b\n" - ".previous\n" - _ASM_EXTABLE(1b, 3b) - : "=r"(faulted), "=a"(replaced) - : "r"(ip), "r"(new), "c"(newch), - "0"(faulted), "a"(old) - : "memory"); - sync_core(); + if (__copy_from_user_inatomic(replaced, (char __user *)ip, MCOUNT_INSN_SIZE)) + return 1; + + if (memcmp(replaced, old_code, MCOUNT_INSN_SIZE) != 0) + return 2; - if (replaced != old && replaced != new) - faulted = 2; + WARN_ON_ONCE(__copy_to_user_inatomic((char __user *)ip, new_code, + MCOUNT_INSN_SIZE)); - return faulted; + sync_core(); + + return 0; } notrace int ftrace_update_ftrace_func(ftrace_func_t func) @@ -112,30 +100,76 @@ notrace int ftrace_update_ftrace_func(ftrace_func_t func) notrace int ftrace_mcount_set(unsigned long *data) { - unsigned long ip = (long)(&mcount_call); - unsigned long *addr = data; - unsigned char old[MCOUNT_INSN_SIZE], *new; - - /* - * Replace the mcount stub with a pointer to the - * ip recorder function. - */ - memcpy(old, &mcount_call, MCOUNT_INSN_SIZE); - new = ftrace_call_replace(ip, *addr); - *addr = ftrace_modify_code(ip, old, new); - + /* mcount is initialized as a nop */ + *data = 0; return 0; } int __init ftrace_dyn_arch_init(void *data) { - const unsigned char *const *noptable = find_nop_table(); - - /* This is running in kstop_machine */ - - ftrace_mcount_set(data); + extern const unsigned char ftrace_test_p6nop[]; + extern const unsigned char ftrace_test_nop5[]; + extern const unsigned char ftrace_test_jmp[]; + int faulted = 0; - ftrace_nop = (unsigned long *)noptable[MCOUNT_INSN_SIZE]; + /* + * There is no good nop for all x86 archs. + * We will default to using the P6_NOP5, but first we + * will test to make sure that the nop will actually + * work on this CPU. If it faults, we will then + * go to a lesser efficient 5 byte nop. If that fails + * we then just use a jmp as our nop. This isn't the most + * efficient nop, but we can not use a multi part nop + * since we would then risk being preempted in the middle + * of that nop, and if we enabled tracing then, it might + * cause a system crash. + * + * TODO: check the cpuid to determine the best nop. + */ + asm volatile ( + "jmp ftrace_test_jmp\n" + /* This code needs to stay around */ + ".section .text, \"ax\"\n" + "ftrace_test_jmp:" + "jmp ftrace_test_p6nop\n" + "nop\n" + "nop\n" + "nop\n" /* 2 byte jmp + 3 bytes */ + "ftrace_test_p6nop:" + P6_NOP5 + "jmp 1f\n" + "ftrace_test_nop5:" + ".byte 0x66,0x66,0x66,0x66,0x90\n" + "jmp 1f\n" + ".previous\n" + "1:" + ".section .fixup, \"ax\"\n" + "2: movl $1, %0\n" + " jmp ftrace_test_nop5\n" + "3: movl $2, %0\n" + " jmp 1b\n" + ".previous\n" + _ASM_EXTABLE(ftrace_test_p6nop, 2b) + _ASM_EXTABLE(ftrace_test_nop5, 3b) + : "=r"(faulted) : "0" (faulted)); + + switch (faulted) { + case 0: + pr_info("ftrace: converting mcount calls to 0f 1f 44 00 00\n"); + ftrace_nop = (unsigned long *)ftrace_test_p6nop; + break; + case 1: + pr_info("ftrace: converting mcount calls to 66 66 66 66 90\n"); + ftrace_nop = (unsigned long *)ftrace_test_nop5; + break; + case 2: + pr_info("ftrace: converting mcount calls to jmp . + 5\n"); + ftrace_nop = (unsigned long *)ftrace_test_jmp; + break; + } + + /* The return code is retured via data */ + *(unsigned long *)data = 0; return 0; } diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c index 9eca5ba7a6b..2ec2de8d8c4 100644 --- a/arch/x86/kernel/genapic_flat_64.c +++ b/arch/x86/kernel/genapic_flat_64.c @@ -179,8 +179,10 @@ static int __init physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) * is an example). */ if (acpi_gbl_FADT.header.revision > FADT2_REVISION_ID && - (acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL)) + (acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL)) { + printk(KERN_DEBUG "system APIC only can use physical flat"); return 1; + } #endif return 0; diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c index 33581d94a90..bfd532843df 100644 --- a/arch/x86/kernel/genx2apic_uv_x.c +++ b/arch/x86/kernel/genx2apic_uv_x.c @@ -341,12 +341,12 @@ static __init void map_mmioh_high(int max_pnode) static __init void uv_rtc_init(void) { - long status, ticks_per_sec, drift; + long status; + u64 ticks_per_sec; - status = - x86_bios_freq_base(BIOS_FREQ_BASE_REALTIME_CLOCK, &ticks_per_sec, - &drift); - if (status != 0 || ticks_per_sec < 100000) { + status = uv_bios_freq_base(BIOS_FREQ_BASE_REALTIME_CLOCK, + &ticks_per_sec); + if (status != BIOS_STATUS_SUCCESS || ticks_per_sec < 100000) { printk(KERN_WARNING "unable to determine platform RTC clock frequency, " "guessing.\n"); @@ -356,7 +356,22 @@ static __init void uv_rtc_init(void) sn_rtc_cycles_per_second = ticks_per_sec; } -static bool uv_system_inited; +/* + * Called on each cpu to initialize the per_cpu UV data area. + * ZZZ hotplug not supported yet + */ +void __cpuinit uv_cpu_init(void) +{ + /* CPU 0 initilization will be done via uv_system_init. */ + if (!uv_blade_info) + return; + + uv_blade_info[uv_numa_blade_id()].nr_online_cpus++; + + if (get_uv_system_type() == UV_NON_UNIQUE_APIC) + set_x2apic_extra_bits(uv_hub_info->pnode); +} + void __init uv_system_init(void) { @@ -412,6 +427,9 @@ void __init uv_system_init(void) gnode_upper = (((unsigned long)node_id.s.node_id) & ~((1 << n_val) - 1)) << m_val; + uv_bios_init(); + uv_bios_get_sn_info(0, &uv_type, &sn_partition_id, + &uv_coherency_id, &uv_region_size); uv_rtc_init(); for_each_present_cpu(cpu) { @@ -433,7 +451,7 @@ void __init uv_system_init(void) uv_cpu_hub_info(cpu)->gpa_mask = (1 << (m_val + n_val)) - 1; uv_cpu_hub_info(cpu)->gnode_upper = gnode_upper; uv_cpu_hub_info(cpu)->global_mmr_base = mmr_base; - uv_cpu_hub_info(cpu)->coherency_domain_number = 0;/* ZZZ */ + uv_cpu_hub_info(cpu)->coherency_domain_number = uv_coherency_id; uv_node_to_blade[nid] = blade; uv_cpu_to_blade[cpu] = blade; max_pnode = max(pnode, max_pnode); @@ -448,21 +466,6 @@ void __init uv_system_init(void) map_mmr_high(max_pnode); map_config_high(max_pnode); map_mmioh_high(max_pnode); - uv_system_inited = true; -} -/* - * Called on each cpu to initialize the per_cpu UV data area. - * ZZZ hotplug not supported yet - */ -void __cpuinit uv_cpu_init(void) -{ - BUG_ON(!uv_system_inited); - - uv_blade_info[uv_numa_blade_id()].nr_online_cpus++; - - if (get_uv_system_type() == UV_NON_UNIQUE_APIC) - set_x2apic_extra_bits(uv_hub_info->pnode); + uv_cpu_init(); } - - diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index acf62fc233d..77017e834cf 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -1,29 +1,49 @@ #include <linux/clocksource.h> #include <linux/clockchips.h> +#include <linux/interrupt.h> +#include <linux/sysdev.h> #include <linux/delay.h> #include <linux/errno.h> #include <linux/hpet.h> #include <linux/init.h> -#include <linux/sysdev.h> +#include <linux/cpu.h> #include <linux/pm.h> +#include <linux/io.h> #include <asm/fixmap.h> -#include <asm/hpet.h> #include <asm/i8253.h> -#include <asm/io.h> +#include <asm/hpet.h> -#define HPET_MASK CLOCKSOURCE_MASK(32) -#define HPET_SHIFT 22 +#define HPET_MASK CLOCKSOURCE_MASK(32) +#define HPET_SHIFT 22 /* FSEC = 10^-15 NSEC = 10^-9 */ -#define FSEC_PER_NSEC 1000000L +#define FSEC_PER_NSEC 1000000L + +#define HPET_DEV_USED_BIT 2 +#define HPET_DEV_USED (1 << HPET_DEV_USED_BIT) +#define HPET_DEV_VALID 0x8 +#define HPET_DEV_FSB_CAP 0x1000 +#define HPET_DEV_PERI_CAP 0x2000 + +#define EVT_TO_HPET_DEV(evt) container_of(evt, struct hpet_dev, evt) /* * HPET address is set in acpi/boot.c, when an ACPI entry exists */ -unsigned long hpet_address; -static void __iomem *hpet_virt_address; +unsigned long hpet_address; +unsigned long hpet_num_timers; +static void __iomem *hpet_virt_address; + +struct hpet_dev { + struct clock_event_device evt; + unsigned int num; + int cpu; + unsigned int irq; + unsigned int flags; + char name[10]; +}; unsigned long hpet_readl(unsigned long a) { @@ -59,7 +79,7 @@ static inline void hpet_clear_mapping(void) static int boot_hpet_disable; int hpet_force_user; -static int __init hpet_setup(char* str) +static int __init hpet_setup(char *str) { if (str) { if (!strncmp("disable", str, 7)) @@ -80,7 +100,7 @@ __setup("nohpet", disable_hpet); static inline int is_hpet_capable(void) { - return (!boot_hpet_disable && hpet_address); + return !boot_hpet_disable && hpet_address; } /* @@ -102,6 +122,9 @@ EXPORT_SYMBOL_GPL(is_hpet_enabled); * timer 0 and timer 1 in case of RTC emulation. */ #ifdef CONFIG_HPET + +static void hpet_reserve_msi_timers(struct hpet_data *hd); + static void hpet_reserve_platform_timers(unsigned long id) { struct hpet __iomem *hpet = hpet_virt_address; @@ -111,10 +134,10 @@ static void hpet_reserve_platform_timers(unsigned long id) nrtimers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT) + 1; - memset(&hd, 0, sizeof (hd)); - hd.hd_phys_address = hpet_address; - hd.hd_address = hpet; - hd.hd_nirqs = nrtimers; + memset(&hd, 0, sizeof(hd)); + hd.hd_phys_address = hpet_address; + hd.hd_address = hpet; + hd.hd_nirqs = nrtimers; hpet_reserve_timer(&hd, 0); #ifdef CONFIG_HPET_EMULATE_RTC @@ -130,10 +153,12 @@ static void hpet_reserve_platform_timers(unsigned long id) hd.hd_irq[1] = HPET_LEGACY_RTC; for (i = 2; i < nrtimers; timer++, i++) { - hd.hd_irq[i] = (readl(&timer->hpet_config) & Tn_INT_ROUTE_CNF_MASK) >> - Tn_INT_ROUTE_CNF_SHIFT; + hd.hd_irq[i] = (readl(&timer->hpet_config) & + Tn_INT_ROUTE_CNF_MASK) >> Tn_INT_ROUTE_CNF_SHIFT; } + hpet_reserve_msi_timers(&hd); + hpet_alloc(&hd); } @@ -227,60 +252,70 @@ static void hpet_legacy_clockevent_register(void) printk(KERN_DEBUG "hpet clockevent registered\n"); } -static void hpet_legacy_set_mode(enum clock_event_mode mode, - struct clock_event_device *evt) +static int hpet_setup_msi_irq(unsigned int irq); + +static void hpet_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt, int timer) { unsigned long cfg, cmp, now; uint64_t delta; - switch(mode) { + switch (mode) { case CLOCK_EVT_MODE_PERIODIC: - delta = ((uint64_t)(NSEC_PER_SEC/HZ)) * hpet_clockevent.mult; - delta >>= hpet_clockevent.shift; + delta = ((uint64_t)(NSEC_PER_SEC/HZ)) * evt->mult; + delta >>= evt->shift; now = hpet_readl(HPET_COUNTER); cmp = now + (unsigned long) delta; - cfg = hpet_readl(HPET_T0_CFG); + cfg = hpet_readl(HPET_Tn_CFG(timer)); cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL | HPET_TN_32BIT; - hpet_writel(cfg, HPET_T0_CFG); + hpet_writel(cfg, HPET_Tn_CFG(timer)); /* * The first write after writing TN_SETVAL to the * config register sets the counter value, the second * write sets the period. */ - hpet_writel(cmp, HPET_T0_CMP); + hpet_writel(cmp, HPET_Tn_CMP(timer)); udelay(1); - hpet_writel((unsigned long) delta, HPET_T0_CMP); + hpet_writel((unsigned long) delta, HPET_Tn_CMP(timer)); break; case CLOCK_EVT_MODE_ONESHOT: - cfg = hpet_readl(HPET_T0_CFG); + cfg = hpet_readl(HPET_Tn_CFG(timer)); cfg &= ~HPET_TN_PERIODIC; cfg |= HPET_TN_ENABLE | HPET_TN_32BIT; - hpet_writel(cfg, HPET_T0_CFG); + hpet_writel(cfg, HPET_Tn_CFG(timer)); break; case CLOCK_EVT_MODE_UNUSED: case CLOCK_EVT_MODE_SHUTDOWN: - cfg = hpet_readl(HPET_T0_CFG); + cfg = hpet_readl(HPET_Tn_CFG(timer)); cfg &= ~HPET_TN_ENABLE; - hpet_writel(cfg, HPET_T0_CFG); + hpet_writel(cfg, HPET_Tn_CFG(timer)); break; case CLOCK_EVT_MODE_RESUME: - hpet_enable_legacy_int(); + if (timer == 0) { + hpet_enable_legacy_int(); + } else { + struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt); + hpet_setup_msi_irq(hdev->irq); + disable_irq(hdev->irq); + irq_set_affinity(hdev->irq, cpumask_of_cpu(hdev->cpu)); + enable_irq(hdev->irq); + } break; } } -static int hpet_legacy_next_event(unsigned long delta, - struct clock_event_device *evt) +static int hpet_next_event(unsigned long delta, + struct clock_event_device *evt, int timer) { u32 cnt; cnt = hpet_readl(HPET_COUNTER); cnt += (u32) delta; - hpet_writel(cnt, HPET_T0_CMP); + hpet_writel(cnt, HPET_Tn_CMP(timer)); /* * We need to read back the CMP register to make sure that @@ -292,6 +327,347 @@ static int hpet_legacy_next_event(unsigned long delta, return (s32)((u32)hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0; } +static void hpet_legacy_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + hpet_set_mode(mode, evt, 0); +} + +static int hpet_legacy_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + return hpet_next_event(delta, evt, 0); +} + +/* + * HPET MSI Support + */ +#ifdef CONFIG_PCI_MSI + +static DEFINE_PER_CPU(struct hpet_dev *, cpu_hpet_dev); +static struct hpet_dev *hpet_devs; + +void hpet_msi_unmask(unsigned int irq) +{ + struct hpet_dev *hdev = get_irq_data(irq); + unsigned long cfg; + + /* unmask it */ + cfg = hpet_readl(HPET_Tn_CFG(hdev->num)); + cfg |= HPET_TN_FSB; + hpet_writel(cfg, HPET_Tn_CFG(hdev->num)); +} + +void hpet_msi_mask(unsigned int irq) +{ + unsigned long cfg; + struct hpet_dev *hdev = get_irq_data(irq); + + /* mask it */ + cfg = hpet_readl(HPET_Tn_CFG(hdev->num)); + cfg &= ~HPET_TN_FSB; + hpet_writel(cfg, HPET_Tn_CFG(hdev->num)); +} + +void hpet_msi_write(unsigned int irq, struct msi_msg *msg) +{ + struct hpet_dev *hdev = get_irq_data(irq); + + hpet_writel(msg->data, HPET_Tn_ROUTE(hdev->num)); + hpet_writel(msg->address_lo, HPET_Tn_ROUTE(hdev->num) + 4); +} + +void hpet_msi_read(unsigned int irq, struct msi_msg *msg) +{ + struct hpet_dev *hdev = get_irq_data(irq); + + msg->data = hpet_readl(HPET_Tn_ROUTE(hdev->num)); + msg->address_lo = hpet_readl(HPET_Tn_ROUTE(hdev->num) + 4); + msg->address_hi = 0; +} + +static void hpet_msi_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt); + hpet_set_mode(mode, evt, hdev->num); +} + +static int hpet_msi_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt); + return hpet_next_event(delta, evt, hdev->num); +} + +static int hpet_setup_msi_irq(unsigned int irq) +{ + if (arch_setup_hpet_msi(irq)) { + destroy_irq(irq); + return -EINVAL; + } + return 0; +} + +static int hpet_assign_irq(struct hpet_dev *dev) +{ + unsigned int irq; + + irq = create_irq(); + if (!irq) + return -EINVAL; + + set_irq_data(irq, dev); + + if (hpet_setup_msi_irq(irq)) + return -EINVAL; + + dev->irq = irq; + return 0; +} + +static irqreturn_t hpet_interrupt_handler(int irq, void *data) +{ + struct hpet_dev *dev = (struct hpet_dev *)data; + struct clock_event_device *hevt = &dev->evt; + + if (!hevt->event_handler) { + printk(KERN_INFO "Spurious HPET timer interrupt on HPET timer %d\n", + dev->num); + return IRQ_HANDLED; + } + + hevt->event_handler(hevt); + return IRQ_HANDLED; +} + +static int hpet_setup_irq(struct hpet_dev *dev) +{ + + if (request_irq(dev->irq, hpet_interrupt_handler, + IRQF_SHARED|IRQF_NOBALANCING, dev->name, dev)) + return -1; + + disable_irq(dev->irq); + irq_set_affinity(dev->irq, cpumask_of_cpu(dev->cpu)); + enable_irq(dev->irq); + + printk(KERN_DEBUG "hpet: %s irq %d for MSI\n", + dev->name, dev->irq); + + return 0; +} + +/* This should be called in specific @cpu */ +static void init_one_hpet_msi_clockevent(struct hpet_dev *hdev, int cpu) +{ + struct clock_event_device *evt = &hdev->evt; + uint64_t hpet_freq; + + WARN_ON(cpu != smp_processor_id()); + if (!(hdev->flags & HPET_DEV_VALID)) + return; + + if (hpet_setup_msi_irq(hdev->irq)) + return; + + hdev->cpu = cpu; + per_cpu(cpu_hpet_dev, cpu) = hdev; + evt->name = hdev->name; + hpet_setup_irq(hdev); + evt->irq = hdev->irq; + + evt->rating = 110; + evt->features = CLOCK_EVT_FEAT_ONESHOT; + if (hdev->flags & HPET_DEV_PERI_CAP) + evt->features |= CLOCK_EVT_FEAT_PERIODIC; + + evt->set_mode = hpet_msi_set_mode; + evt->set_next_event = hpet_msi_next_event; + evt->shift = 32; + + /* + * The period is a femto seconds value. We need to calculate the + * scaled math multiplication factor for nanosecond to hpet tick + * conversion. + */ + hpet_freq = 1000000000000000ULL; + do_div(hpet_freq, hpet_period); + evt->mult = div_sc((unsigned long) hpet_freq, + NSEC_PER_SEC, evt->shift); + /* Calculate the max delta */ + evt->max_delta_ns = clockevent_delta2ns(0x7FFFFFFF, evt); + /* 5 usec minimum reprogramming delta. */ + evt->min_delta_ns = 5000; + + evt->cpumask = cpumask_of_cpu(hdev->cpu); + clockevents_register_device(evt); +} + +#ifdef CONFIG_HPET +/* Reserve at least one timer for userspace (/dev/hpet) */ +#define RESERVE_TIMERS 1 +#else +#define RESERVE_TIMERS 0 +#endif + +static void hpet_msi_capability_lookup(unsigned int start_timer) +{ + unsigned int id; + unsigned int num_timers; + unsigned int num_timers_used = 0; + int i; + + id = hpet_readl(HPET_ID); + + num_timers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT); + num_timers++; /* Value read out starts from 0 */ + + hpet_devs = kzalloc(sizeof(struct hpet_dev) * num_timers, GFP_KERNEL); + if (!hpet_devs) + return; + + hpet_num_timers = num_timers; + + for (i = start_timer; i < num_timers - RESERVE_TIMERS; i++) { + struct hpet_dev *hdev = &hpet_devs[num_timers_used]; + unsigned long cfg = hpet_readl(HPET_Tn_CFG(i)); + + /* Only consider HPET timer with MSI support */ + if (!(cfg & HPET_TN_FSB_CAP)) + continue; + + hdev->flags = 0; + if (cfg & HPET_TN_PERIODIC_CAP) + hdev->flags |= HPET_DEV_PERI_CAP; + hdev->num = i; + + sprintf(hdev->name, "hpet%d", i); + if (hpet_assign_irq(hdev)) + continue; + + hdev->flags |= HPET_DEV_FSB_CAP; + hdev->flags |= HPET_DEV_VALID; + num_timers_used++; + if (num_timers_used == num_possible_cpus()) + break; + } + + printk(KERN_INFO "HPET: %d timers in total, %d timers will be used for per-cpu timer\n", + num_timers, num_timers_used); +} + +#ifdef CONFIG_HPET +static void hpet_reserve_msi_timers(struct hpet_data *hd) +{ + int i; + + if (!hpet_devs) + return; + + for (i = 0; i < hpet_num_timers; i++) { + struct hpet_dev *hdev = &hpet_devs[i]; + + if (!(hdev->flags & HPET_DEV_VALID)) + continue; + + hd->hd_irq[hdev->num] = hdev->irq; + hpet_reserve_timer(hd, hdev->num); + } +} +#endif + +static struct hpet_dev *hpet_get_unused_timer(void) +{ + int i; + + if (!hpet_devs) + return NULL; + + for (i = 0; i < hpet_num_timers; i++) { + struct hpet_dev *hdev = &hpet_devs[i]; + + if (!(hdev->flags & HPET_DEV_VALID)) + continue; + if (test_and_set_bit(HPET_DEV_USED_BIT, + (unsigned long *)&hdev->flags)) + continue; + return hdev; + } + return NULL; +} + +struct hpet_work_struct { + struct delayed_work work; + struct completion complete; +}; + +static void hpet_work(struct work_struct *w) +{ + struct hpet_dev *hdev; + int cpu = smp_processor_id(); + struct hpet_work_struct *hpet_work; + + hpet_work = container_of(w, struct hpet_work_struct, work.work); + + hdev = hpet_get_unused_timer(); + if (hdev) + init_one_hpet_msi_clockevent(hdev, cpu); + + complete(&hpet_work->complete); +} + +static int hpet_cpuhp_notify(struct notifier_block *n, + unsigned long action, void *hcpu) +{ + unsigned long cpu = (unsigned long)hcpu; + struct hpet_work_struct work; + struct hpet_dev *hdev = per_cpu(cpu_hpet_dev, cpu); + + switch (action & 0xf) { + case CPU_ONLINE: + INIT_DELAYED_WORK(&work.work, hpet_work); + init_completion(&work.complete); + /* FIXME: add schedule_work_on() */ + schedule_delayed_work_on(cpu, &work.work, 0); + wait_for_completion(&work.complete); + break; + case CPU_DEAD: + if (hdev) { + free_irq(hdev->irq, hdev); + hdev->flags &= ~HPET_DEV_USED; + per_cpu(cpu_hpet_dev, cpu) = NULL; + } + break; + } + return NOTIFY_OK; +} +#else + +static int hpet_setup_msi_irq(unsigned int irq) +{ + return 0; +} +static void hpet_msi_capability_lookup(unsigned int start_timer) +{ + return; +} + +#ifdef CONFIG_HPET +static void hpet_reserve_msi_timers(struct hpet_data *hd) +{ + return; +} +#endif + +static int hpet_cpuhp_notify(struct notifier_block *n, + unsigned long action, void *hcpu) +{ + return NOTIFY_OK; +} + +#endif + /* * Clock source related code */ @@ -427,8 +803,10 @@ int __init hpet_enable(void) if (id & HPET_ID_LEGSUP) { hpet_legacy_clockevent_register(); + hpet_msi_capability_lookup(2); return 1; } + hpet_msi_capability_lookup(0); return 0; out_nohpet: @@ -445,6 +823,8 @@ out_nohpet: */ static __init int hpet_late_init(void) { + int cpu; + if (boot_hpet_disable) return -ENODEV; @@ -460,6 +840,13 @@ static __init int hpet_late_init(void) hpet_reserve_platform_timers(hpet_readl(HPET_ID)); + for_each_online_cpu(cpu) { + hpet_cpuhp_notify(NULL, CPU_ONLINE, (void *)(long)cpu); + } + + /* This notifier should be called after workqueue is ready */ + hotcpu_notifier(hpet_cpuhp_notify, -20); + return 0; } fs_initcall(hpet_late_init); diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic.c index 02063ae042f..b764d7429c6 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic.c @@ -27,17 +27,21 @@ #include <linux/sched.h> #include <linux/pci.h> #include <linux/mc146818rtc.h> +#include <linux/compiler.h> #include <linux/acpi.h> +#include <linux/module.h> #include <linux/sysdev.h> #include <linux/msi.h> #include <linux/htirq.h> -#include <linux/dmar.h> -#include <linux/jiffies.h> +#include <linux/freezer.h> +#include <linux/kthread.h> +#include <linux/jiffies.h> /* time_after() */ #ifdef CONFIG_ACPI #include <acpi/acpi_bus.h> #endif #include <linux/bootmem.h> #include <linux/dmar.h> +#include <linux/hpet.h> #include <asm/idle.h> #include <asm/io.h> @@ -46,61 +50,28 @@ #include <asm/proto.h> #include <asm/acpi.h> #include <asm/dma.h> +#include <asm/timer.h> #include <asm/i8259.h> #include <asm/nmi.h> #include <asm/msidef.h> #include <asm/hypertransport.h> +#include <asm/setup.h> #include <asm/irq_remapping.h> +#include <asm/hpet.h> +#include <asm/uv/uv_hub.h> +#include <asm/uv/uv_irq.h> #include <mach_ipi.h> #include <mach_apic.h> +#include <mach_apicdef.h> #define __apicdebuginit(type) static type __init -struct irq_cfg { - cpumask_t domain; - cpumask_t old_domain; - unsigned move_cleanup_count; - u8 vector; - u8 move_in_progress : 1; -}; - -/* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */ -static struct irq_cfg irq_cfg[NR_IRQS] __read_mostly = { - [0] = { .domain = CPU_MASK_ALL, .vector = IRQ0_VECTOR, }, - [1] = { .domain = CPU_MASK_ALL, .vector = IRQ1_VECTOR, }, - [2] = { .domain = CPU_MASK_ALL, .vector = IRQ2_VECTOR, }, - [3] = { .domain = CPU_MASK_ALL, .vector = IRQ3_VECTOR, }, - [4] = { .domain = CPU_MASK_ALL, .vector = IRQ4_VECTOR, }, - [5] = { .domain = CPU_MASK_ALL, .vector = IRQ5_VECTOR, }, - [6] = { .domain = CPU_MASK_ALL, .vector = IRQ6_VECTOR, }, - [7] = { .domain = CPU_MASK_ALL, .vector = IRQ7_VECTOR, }, - [8] = { .domain = CPU_MASK_ALL, .vector = IRQ8_VECTOR, }, - [9] = { .domain = CPU_MASK_ALL, .vector = IRQ9_VECTOR, }, - [10] = { .domain = CPU_MASK_ALL, .vector = IRQ10_VECTOR, }, - [11] = { .domain = CPU_MASK_ALL, .vector = IRQ11_VECTOR, }, - [12] = { .domain = CPU_MASK_ALL, .vector = IRQ12_VECTOR, }, - [13] = { .domain = CPU_MASK_ALL, .vector = IRQ13_VECTOR, }, - [14] = { .domain = CPU_MASK_ALL, .vector = IRQ14_VECTOR, }, - [15] = { .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, }, -}; - -static int assign_irq_vector(int irq, cpumask_t mask); - -int first_system_vector = 0xfe; - -char system_vectors[NR_VECTORS] = { [0 ... NR_VECTORS-1] = SYS_VECTOR_FREE}; - -int sis_apic_bug; /* not actually supported, dummy for compile */ - -static int no_timer_check; - -static int disable_timer_pin_1 __initdata; - -int timer_through_8259 __initdata; - -/* Where if anywhere is the i8259 connect in external int mode */ -static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; +/* + * Is the SiS APIC rmw bug present ? + * -1 = don't know, 0 = no, 1 = yes + */ +int sis_apic_bug = -1; static DEFINE_SPINLOCK(ioapic_lock); static DEFINE_SPINLOCK(vector_lock); @@ -110,9 +81,6 @@ static DEFINE_SPINLOCK(vector_lock); */ int nr_ioapic_registers[MAX_IO_APICS]; -/* I/O APIC RTE contents at the OS boot up */ -struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS]; - /* I/O APIC entries */ struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; int nr_ioapics; @@ -123,11 +91,69 @@ struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* # of MP IRQ source entries */ int mp_irq_entries; +#if defined (CONFIG_MCA) || defined (CONFIG_EISA) +int mp_bus_id_to_type[MAX_MP_BUSSES]; +#endif + DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); +int skip_ioapic_setup; + +static int __init parse_noapic(char *str) +{ + /* disable IO-APIC */ + disable_ioapic_setup(); + return 0; +} +early_param("noapic", parse_noapic); + +struct irq_pin_list; +struct irq_cfg { + unsigned int irq; + struct irq_pin_list *irq_2_pin; + cpumask_t domain; + cpumask_t old_domain; + unsigned move_cleanup_count; + u8 vector; + u8 move_in_progress : 1; +}; + +/* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */ +static struct irq_cfg irq_cfgx[NR_IRQS] = { + [0] = { .irq = 0, .domain = CPU_MASK_ALL, .vector = IRQ0_VECTOR, }, + [1] = { .irq = 1, .domain = CPU_MASK_ALL, .vector = IRQ1_VECTOR, }, + [2] = { .irq = 2, .domain = CPU_MASK_ALL, .vector = IRQ2_VECTOR, }, + [3] = { .irq = 3, .domain = CPU_MASK_ALL, .vector = IRQ3_VECTOR, }, + [4] = { .irq = 4, .domain = CPU_MASK_ALL, .vector = IRQ4_VECTOR, }, + [5] = { .irq = 5, .domain = CPU_MASK_ALL, .vector = IRQ5_VECTOR, }, + [6] = { .irq = 6, .domain = CPU_MASK_ALL, .vector = IRQ6_VECTOR, }, + [7] = { .irq = 7, .domain = CPU_MASK_ALL, .vector = IRQ7_VECTOR, }, + [8] = { .irq = 8, .domain = CPU_MASK_ALL, .vector = IRQ8_VECTOR, }, + [9] = { .irq = 9, .domain = CPU_MASK_ALL, .vector = IRQ9_VECTOR, }, + [10] = { .irq = 10, .domain = CPU_MASK_ALL, .vector = IRQ10_VECTOR, }, + [11] = { .irq = 11, .domain = CPU_MASK_ALL, .vector = IRQ11_VECTOR, }, + [12] = { .irq = 12, .domain = CPU_MASK_ALL, .vector = IRQ12_VECTOR, }, + [13] = { .irq = 13, .domain = CPU_MASK_ALL, .vector = IRQ13_VECTOR, }, + [14] = { .irq = 14, .domain = CPU_MASK_ALL, .vector = IRQ14_VECTOR, }, + [15] = { .irq = 15, .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, }, +}; + +#define for_each_irq_cfg(irq, cfg) \ + for (irq = 0, cfg = irq_cfgx; irq < nr_irqs; irq++, cfg++) + +static struct irq_cfg *irq_cfg(unsigned int irq) +{ + return irq < nr_irqs ? irq_cfgx + irq : NULL; +} + +static struct irq_cfg *irq_cfg_alloc(unsigned int irq) +{ + return irq_cfg(irq); +} + /* - * Rough estimation of how many shared IRQs there are, can - * be changed anytime. + * Rough estimation of how many shared IRQs there are, can be changed + * anytime. */ #define MAX_PLUS_SHARED_IRQS NR_IRQS #define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS) @@ -139,9 +165,36 @@ DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); * between pins and IRQs. */ -static struct irq_pin_list { - short apic, pin, next; -} irq_2_pin[PIN_MAP_SIZE]; +struct irq_pin_list { + int apic, pin; + struct irq_pin_list *next; +}; + +static struct irq_pin_list irq_2_pin_head[PIN_MAP_SIZE]; +static struct irq_pin_list *irq_2_pin_ptr; + +static void __init irq_2_pin_init(void) +{ + struct irq_pin_list *pin = irq_2_pin_head; + int i; + + for (i = 1; i < PIN_MAP_SIZE; i++) + pin[i-1].next = &pin[i]; + + irq_2_pin_ptr = &pin[0]; +} + +static struct irq_pin_list *get_one_free_irq_2_pin(void) +{ + struct irq_pin_list *pin = irq_2_pin_ptr; + + if (!pin) + panic("can not get more irq_2_pin\n"); + + irq_2_pin_ptr = pin->next; + pin->next = NULL; + return pin; +} struct io_apic { unsigned int index; @@ -172,10 +225,15 @@ static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned i /* * Re-write a value: to be used for read-modify-write * cycles where the read already set up the index register. + * + * Older SiS APIC requires we rewrite the index register */ -static inline void io_apic_modify(unsigned int apic, unsigned int value) +static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value) { struct io_apic __iomem *io_apic = io_apic_base(apic); + + if (sis_apic_bug) + writel(reg, &io_apic->index); writel(value, &io_apic->data); } @@ -183,16 +241,17 @@ static bool io_apic_level_ack_pending(unsigned int irq) { struct irq_pin_list *entry; unsigned long flags; + struct irq_cfg *cfg = irq_cfg(irq); spin_lock_irqsave(&ioapic_lock, flags); - entry = irq_2_pin + irq; + entry = cfg->irq_2_pin; for (;;) { unsigned int reg; int pin; - pin = entry->pin; - if (pin == -1) + if (!entry) break; + pin = entry->pin; reg = io_apic_read(entry->apic, 0x10 + pin*2); /* Is the remote IRR bit set? */ if (reg & IO_APIC_REDIR_REMOTE_IRR) { @@ -201,45 +260,13 @@ static bool io_apic_level_ack_pending(unsigned int irq) } if (!entry->next) break; - entry = irq_2_pin + entry->next; + entry = entry->next; } spin_unlock_irqrestore(&ioapic_lock, flags); return false; } -/* - * Synchronize the IO-APIC and the CPU by doing - * a dummy read from the IO-APIC - */ -static inline void io_apic_sync(unsigned int apic) -{ - struct io_apic __iomem *io_apic = io_apic_base(apic); - readl(&io_apic->data); -} - -#define __DO_ACTION(R, ACTION, FINAL) \ - \ -{ \ - int pin; \ - struct irq_pin_list *entry = irq_2_pin + irq; \ - \ - BUG_ON(irq >= NR_IRQS); \ - for (;;) { \ - unsigned int reg; \ - pin = entry->pin; \ - if (pin == -1) \ - break; \ - reg = io_apic_read(entry->apic, 0x10 + R + pin*2); \ - reg ACTION; \ - io_apic_modify(entry->apic, reg); \ - FINAL; \ - if (!entry->next) \ - break; \ - entry = irq_2_pin + entry->next; \ - } \ -} - union entry_union { struct { u32 w1, w2; }; struct IO_APIC_route_entry entry; @@ -299,59 +326,71 @@ static void ioapic_mask_entry(int apic, int pin) static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector) { int apic, pin; - struct irq_pin_list *entry = irq_2_pin + irq; + struct irq_cfg *cfg; + struct irq_pin_list *entry; - BUG_ON(irq >= NR_IRQS); + cfg = irq_cfg(irq); + entry = cfg->irq_2_pin; for (;;) { unsigned int reg; + + if (!entry) + break; + apic = entry->apic; pin = entry->pin; - if (pin == -1) - break; +#ifdef CONFIG_INTR_REMAP /* * With interrupt-remapping, destination information comes * from interrupt-remapping table entry. */ if (!irq_remapped(irq)) io_apic_write(apic, 0x11 + pin*2, dest); +#else + io_apic_write(apic, 0x11 + pin*2, dest); +#endif reg = io_apic_read(apic, 0x10 + pin*2); reg &= ~IO_APIC_REDIR_VECTOR_MASK; reg |= vector; - io_apic_modify(apic, reg); + io_apic_modify(apic, 0x10 + pin*2, reg); if (!entry->next) break; - entry = irq_2_pin + entry->next; + entry = entry->next; } } +static int assign_irq_vector(int irq, cpumask_t mask); + static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; unsigned long flags; unsigned int dest; cpumask_t tmp; + struct irq_desc *desc; cpus_and(tmp, mask, cpu_online_map); if (cpus_empty(tmp)) return; + cfg = irq_cfg(irq); if (assign_irq_vector(irq, mask)) return; cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); - /* * Only the high 8 bits are valid. */ dest = SET_APIC_LOGICAL_ID(dest); + desc = irq_to_desc(irq); spin_lock_irqsave(&ioapic_lock, flags); __target_IO_APIC_irq(irq, dest, cfg->vector); - irq_desc[irq].affinity = mask; + desc->affinity = mask; spin_unlock_irqrestore(&ioapic_lock, flags); } -#endif +#endif /* CONFIG_SMP */ /* * The common case is 1:1 IRQ<->pin mappings. Sometimes there are @@ -360,19 +399,30 @@ static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) */ static void add_pin_to_irq(unsigned int irq, int apic, int pin) { - static int first_free_entry = NR_IRQS; - struct irq_pin_list *entry = irq_2_pin + irq; + struct irq_cfg *cfg; + struct irq_pin_list *entry; + + /* first time to refer irq_cfg, so with new */ + cfg = irq_cfg_alloc(irq); + entry = cfg->irq_2_pin; + if (!entry) { + entry = get_one_free_irq_2_pin(); + cfg->irq_2_pin = entry; + entry->apic = apic; + entry->pin = pin; + return; + } - BUG_ON(irq >= NR_IRQS); - while (entry->next) - entry = irq_2_pin + entry->next; + while (entry->next) { + /* not again, please */ + if (entry->apic == apic && entry->pin == pin) + return; - if (entry->pin != -1) { - entry->next = first_free_entry; - entry = irq_2_pin + entry->next; - if (++first_free_entry >= PIN_MAP_SIZE) - panic("io_apic.c: ran out of irq_2_pin entries!"); + entry = entry->next; } + + entry->next = get_one_free_irq_2_pin(); + entry = entry->next; entry->apic = apic; entry->pin = pin; } @@ -384,30 +434,86 @@ static void __init replace_pin_at_irq(unsigned int irq, int oldapic, int oldpin, int newapic, int newpin) { - struct irq_pin_list *entry = irq_2_pin + irq; + struct irq_cfg *cfg = irq_cfg(irq); + struct irq_pin_list *entry = cfg->irq_2_pin; + int replaced = 0; - while (1) { + while (entry) { if (entry->apic == oldapic && entry->pin == oldpin) { entry->apic = newapic; entry->pin = newpin; - } - if (!entry->next) + replaced = 1; + /* every one is different, right? */ break; - entry = irq_2_pin + entry->next; + } + entry = entry->next; + } + + /* why? call replace before add? */ + if (!replaced) + add_pin_to_irq(irq, newapic, newpin); +} + +static inline void io_apic_modify_irq(unsigned int irq, + int mask_and, int mask_or, + void (*final)(struct irq_pin_list *entry)) +{ + int pin; + struct irq_cfg *cfg; + struct irq_pin_list *entry; + + cfg = irq_cfg(irq); + for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) { + unsigned int reg; + pin = entry->pin; + reg = io_apic_read(entry->apic, 0x10 + pin * 2); + reg &= mask_and; + reg |= mask_or; + io_apic_modify(entry->apic, 0x10 + pin * 2, reg); + if (final) + final(entry); } } +static void __unmask_IO_APIC_irq(unsigned int irq) +{ + io_apic_modify_irq(irq, ~IO_APIC_REDIR_MASKED, 0, NULL); +} -#define DO_ACTION(name,R,ACTION, FINAL) \ - \ - static void name##_IO_APIC_irq (unsigned int irq) \ - __DO_ACTION(R, ACTION, FINAL) +#ifdef CONFIG_X86_64 +void io_apic_sync(struct irq_pin_list *entry) +{ + /* + * Synchronize the IO-APIC and the CPU by doing + * a dummy read from the IO-APIC + */ + struct io_apic __iomem *io_apic; + io_apic = io_apic_base(entry->apic); + readl(&io_apic->data); +} -/* mask = 1 */ -DO_ACTION(__mask, 0, |= IO_APIC_REDIR_MASKED, io_apic_sync(entry->apic)) +static void __mask_IO_APIC_irq(unsigned int irq) +{ + io_apic_modify_irq(irq, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync); +} +#else /* CONFIG_X86_32 */ +static void __mask_IO_APIC_irq(unsigned int irq) +{ + io_apic_modify_irq(irq, ~0, IO_APIC_REDIR_MASKED, NULL); +} -/* mask = 0 */ -DO_ACTION(__unmask, 0, &= ~IO_APIC_REDIR_MASKED, ) +static void __mask_and_edge_IO_APIC_irq(unsigned int irq) +{ + io_apic_modify_irq(irq, ~IO_APIC_REDIR_LEVEL_TRIGGER, + IO_APIC_REDIR_MASKED, NULL); +} + +static void __unmask_and_level_IO_APIC_irq(unsigned int irq) +{ + io_apic_modify_irq(irq, ~IO_APIC_REDIR_MASKED, + IO_APIC_REDIR_LEVEL_TRIGGER, NULL); +} +#endif /* CONFIG_X86_32 */ static void mask_IO_APIC_irq (unsigned int irq) { @@ -450,6 +556,68 @@ static void clear_IO_APIC (void) clear_IO_APIC_pin(apic, pin); } +#if !defined(CONFIG_SMP) && defined(CONFIG_X86_32) +void send_IPI_self(int vector) +{ + unsigned int cfg; + + /* + * Wait for idle. + */ + apic_wait_icr_idle(); + cfg = APIC_DM_FIXED | APIC_DEST_SELF | vector | APIC_DEST_LOGICAL; + /* + * Send the IPI. The write to APIC_ICR fires this off. + */ + apic_write(APIC_ICR, cfg); +} +#endif /* !CONFIG_SMP && CONFIG_X86_32*/ + +#ifdef CONFIG_X86_32 +/* + * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to + * specific CPU-side IRQs. + */ + +#define MAX_PIRQS 8 +static int pirq_entries [MAX_PIRQS]; +static int pirqs_enabled; + +static int __init ioapic_pirq_setup(char *str) +{ + int i, max; + int ints[MAX_PIRQS+1]; + + get_options(str, ARRAY_SIZE(ints), ints); + + for (i = 0; i < MAX_PIRQS; i++) + pirq_entries[i] = -1; + + pirqs_enabled = 1; + apic_printk(APIC_VERBOSE, KERN_INFO + "PIRQ redirection, working around broken MP-BIOS.\n"); + max = MAX_PIRQS; + if (ints[0] < MAX_PIRQS) + max = ints[0]; + + for (i = 0; i < max; i++) { + apic_printk(APIC_VERBOSE, KERN_DEBUG + "... PIRQ%d -> IRQ %d\n", i, ints[i+1]); + /* + * PIRQs are mapped upside down, usually. + */ + pirq_entries[MAX_PIRQS-i-1] = ints[i+1]; + } + return 1; +} + +__setup("pirq=", ioapic_pirq_setup); +#endif /* CONFIG_X86_32 */ + +#ifdef CONFIG_INTR_REMAP +/* I/O APIC RTE contents at the OS boot up */ +static struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS]; + /* * Saves and masks all the unmasked IO-APIC RTE's */ @@ -474,7 +642,7 @@ int save_mask_IO_APIC_setup(void) kzalloc(sizeof(struct IO_APIC_route_entry) * nr_ioapic_registers[apic], GFP_KERNEL); if (!early_ioapic_entries[apic]) - return -ENOMEM; + goto nomem; } for (apic = 0; apic < nr_ioapics; apic++) @@ -488,17 +656,31 @@ int save_mask_IO_APIC_setup(void) ioapic_write_entry(apic, pin, entry); } } + return 0; + +nomem: + while (apic >= 0) + kfree(early_ioapic_entries[apic--]); + memset(early_ioapic_entries, 0, + ARRAY_SIZE(early_ioapic_entries)); + + return -ENOMEM; } void restore_IO_APIC_setup(void) { int apic, pin; - for (apic = 0; apic < nr_ioapics; apic++) + for (apic = 0; apic < nr_ioapics; apic++) { + if (!early_ioapic_entries[apic]) + break; for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) ioapic_write_entry(apic, pin, early_ioapic_entries[apic][pin]); + kfree(early_ioapic_entries[apic]); + early_ioapic_entries[apic] = NULL; + } } void reinit_intr_remapped_IO_APIC(int intr_remapping) @@ -512,25 +694,7 @@ void reinit_intr_remapped_IO_APIC(int intr_remapping) */ restore_IO_APIC_setup(); } - -int skip_ioapic_setup; -int ioapic_force; - -static int __init parse_noapic(char *str) -{ - disable_ioapic_setup(); - return 0; -} -early_param("noapic", parse_noapic); - -/* Actually the next is obsolete, but keep it for paranoid reasons -AK */ -static int __init disable_timer_pin_setup(char *arg) -{ - disable_timer_pin_1 = 1; - return 1; -} -__setup("disable_timer_pin_1", disable_timer_pin_setup); - +#endif /* * Find the IRQ entry number of a certain pin. @@ -634,22 +798,54 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) best_guess = irq; } } - BUG_ON(best_guess >= NR_IRQS); return best_guess; } +EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector); + +#if defined(CONFIG_EISA) || defined(CONFIG_MCA) +/* + * EISA Edge/Level control register, ELCR + */ +static int EISA_ELCR(unsigned int irq) +{ + if (irq < 16) { + unsigned int port = 0x4d0 + (irq >> 3); + return (inb(port) >> (irq & 7)) & 1; + } + apic_printk(APIC_VERBOSE, KERN_INFO + "Broken MPtable reports ISA irq %d\n", irq); + return 0; +} + +#endif + /* ISA interrupts are always polarity zero edge triggered, * when listed as conforming in the MP table. */ #define default_ISA_trigger(idx) (0) #define default_ISA_polarity(idx) (0) +/* EISA interrupts are always polarity zero and can be edge or level + * trigger depending on the ELCR value. If an interrupt is listed as + * EISA conforming in the MP table, that means its trigger type must + * be read in from the ELCR */ + +#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mp_srcbusirq)) +#define default_EISA_polarity(idx) default_ISA_polarity(idx) + /* PCI interrupts are always polarity one level triggered, * when listed as conforming in the MP table. */ #define default_PCI_trigger(idx) (1) #define default_PCI_polarity(idx) (1) +/* MCA interrupts are always polarity zero level triggered, + * when listed as conforming in the MP table. */ + +#define default_MCA_trigger(idx) (1) +#define default_MCA_polarity(idx) default_ISA_polarity(idx) + static int MPBIOS_polarity(int idx) { int bus = mp_irqs[idx].mp_srcbus; @@ -707,6 +903,36 @@ static int MPBIOS_trigger(int idx) trigger = default_ISA_trigger(idx); else trigger = default_PCI_trigger(idx); +#if defined(CONFIG_EISA) || defined(CONFIG_MCA) + switch (mp_bus_id_to_type[bus]) { + case MP_BUS_ISA: /* ISA pin */ + { + /* set before the switch */ + break; + } + case MP_BUS_EISA: /* EISA pin */ + { + trigger = default_EISA_trigger(idx); + break; + } + case MP_BUS_PCI: /* PCI pin */ + { + /* set before the switch */ + break; + } + case MP_BUS_MCA: /* MCA pin */ + { + trigger = default_MCA_trigger(idx); + break; + } + default: + { + printk(KERN_WARNING "broken BIOS!!\n"); + trigger = 1; + break; + } + } +#endif break; case 1: /* edge */ { @@ -744,6 +970,7 @@ static inline int irq_trigger(int idx) return MPBIOS_trigger(idx); } +int (*ioapic_renumber_irq)(int ioapic, int irq); static int pin_2_irq(int idx, int apic, int pin) { int irq, i; @@ -765,8 +992,32 @@ static int pin_2_irq(int idx, int apic, int pin) while (i < apic) irq += nr_ioapic_registers[i++]; irq += pin; + /* + * For MPS mode, so far only needed by ES7000 platform + */ + if (ioapic_renumber_irq) + irq = ioapic_renumber_irq(apic, irq); } - BUG_ON(irq >= NR_IRQS); + +#ifdef CONFIG_X86_32 + /* + * PCI IRQ command line redirection. Yes, limits are hardcoded. + */ + if ((pin >= 16) && (pin <= 23)) { + if (pirq_entries[pin-16] != -1) { + if (!pirq_entries[pin-16]) { + apic_printk(APIC_VERBOSE, KERN_DEBUG + "disabling PIRQ%d\n", pin-16); + } else { + irq = pirq_entries[pin-16]; + apic_printk(APIC_VERBOSE, KERN_DEBUG + "using PIRQ%d -> IRQ %d\n", + pin-16, irq); + } + } + } +#endif + return irq; } @@ -801,8 +1052,7 @@ static int __assign_irq_vector(int irq, cpumask_t mask) int cpu; struct irq_cfg *cfg; - BUG_ON((unsigned)irq >= NR_IRQS); - cfg = &irq_cfg[irq]; + cfg = irq_cfg(irq); /* Only try and allocate irqs on cpus that are present */ cpus_and(mask, mask, cpu_online_map); @@ -837,8 +1087,13 @@ next: } if (unlikely(current_vector == vector)) continue; +#ifdef CONFIG_X86_64 if (vector == IA32_SYSCALL_VECTOR) goto next; +#else + if (vector == SYSCALL_VECTOR) + goto next; +#endif for_each_cpu_mask_nr(new_cpu, new_mask) if (per_cpu(vector_irq, new_cpu)[vector] != -1) goto next; @@ -875,8 +1130,7 @@ static void __clear_irq_vector(int irq) cpumask_t mask; int cpu, vector; - BUG_ON((unsigned)irq >= NR_IRQS); - cfg = &irq_cfg[irq]; + cfg = irq_cfg(irq); BUG_ON(!cfg->vector); vector = cfg->vector; @@ -893,12 +1147,13 @@ void __setup_vector_irq(int cpu) /* Initialize vector_irq on a new cpu */ /* This function must be called with vector_lock held */ int irq, vector; + struct irq_cfg *cfg; /* Mark the inuse vectors */ - for (irq = 0; irq < NR_IRQS; ++irq) { - if (!cpu_isset(cpu, irq_cfg[irq].domain)) + for_each_irq_cfg(irq, cfg) { + if (!cpu_isset(cpu, cfg->domain)) continue; - vector = irq_cfg[irq].vector; + vector = cfg->vector; per_cpu(vector_irq, cpu)[vector] = irq; } /* Mark the free vectors */ @@ -906,7 +1161,9 @@ void __setup_vector_irq(int cpu) irq = per_cpu(vector_irq, cpu)[vector]; if (irq < 0) continue; - if (!cpu_isset(cpu, irq_cfg[irq].domain)) + + cfg = irq_cfg(irq); + if (!cpu_isset(cpu, cfg->domain)) per_cpu(vector_irq, cpu)[vector] = -1; } } @@ -916,16 +1173,49 @@ static struct irq_chip ioapic_chip; static struct irq_chip ir_ioapic_chip; #endif +#define IOAPIC_AUTO -1 +#define IOAPIC_EDGE 0 +#define IOAPIC_LEVEL 1 + +#ifdef CONFIG_X86_32 +static inline int IO_APIC_irq_trigger(int irq) +{ + int apic, idx, pin; + + for (apic = 0; apic < nr_ioapics; apic++) { + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + idx = find_irq_entry(apic, pin, mp_INT); + if ((idx != -1) && (irq == pin_2_irq(idx, apic, pin))) + return irq_trigger(idx); + } + } + /* + * nonexistent IRQs are edge default + */ + return 0; +} +#else +static inline int IO_APIC_irq_trigger(int irq) +{ + return 1; +} +#endif + static void ioapic_register_intr(int irq, unsigned long trigger) { - if (trigger) - irq_desc[irq].status |= IRQ_LEVEL; + struct irq_desc *desc; + + desc = irq_to_desc(irq); + + if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || + trigger == IOAPIC_LEVEL) + desc->status |= IRQ_LEVEL; else - irq_desc[irq].status &= ~IRQ_LEVEL; + desc->status &= ~IRQ_LEVEL; #ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { - irq_desc[irq].status |= IRQ_MOVE_PCNTXT; + desc->status |= IRQ_MOVE_PCNTXT; if (trigger) set_irq_chip_and_handler_name(irq, &ir_ioapic_chip, handle_fasteoi_irq, @@ -936,7 +1226,8 @@ static void ioapic_register_intr(int irq, unsigned long trigger) return; } #endif - if (trigger) + if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || + trigger == IOAPIC_LEVEL) set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_fasteoi_irq, "fasteoi"); @@ -1009,13 +1300,15 @@ static int setup_ioapic_entry(int apic, int irq, static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, int trigger, int polarity) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; struct IO_APIC_route_entry entry; cpumask_t mask; if (!IO_APIC_IRQ(irq)) return; + cfg = irq_cfg(irq); + mask = TARGET_CPUS; if (assign_irq_vector(irq, mask)) return; @@ -1047,37 +1340,49 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, static void __init setup_IO_APIC_irqs(void) { - int apic, pin, idx, irq, first_notcon = 1; + int apic, pin, idx, irq; + int notcon = 0; apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); for (apic = 0; apic < nr_ioapics; apic++) { - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - - idx = find_irq_entry(apic,pin,mp_INT); - if (idx == -1) { - if (first_notcon) { - apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", mp_ioapics[apic].mp_apicid, pin); - first_notcon = 0; - } else - apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mp_apicid, pin); - continue; - } - if (!first_notcon) { - apic_printk(APIC_VERBOSE, " not connected.\n"); - first_notcon = 1; - } + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - irq = pin_2_irq(idx, apic, pin); - add_pin_to_irq(irq, apic, pin); + idx = find_irq_entry(apic, pin, mp_INT); + if (idx == -1) { + if (!notcon) { + notcon = 1; + apic_printk(APIC_VERBOSE, + KERN_DEBUG " %d-%d", + mp_ioapics[apic].mp_apicid, + pin); + } else + apic_printk(APIC_VERBOSE, " %d-%d", + mp_ioapics[apic].mp_apicid, + pin); + continue; + } + if (notcon) { + apic_printk(APIC_VERBOSE, + " (apicid-pin) not connected\n"); + notcon = 0; + } - setup_IO_APIC_irq(apic, pin, irq, - irq_trigger(idx), irq_polarity(idx)); - } + irq = pin_2_irq(idx, apic, pin); +#ifdef CONFIG_X86_32 + if (multi_timer_check(apic, irq)) + continue; +#endif + add_pin_to_irq(irq, apic, pin); + + setup_IO_APIC_irq(apic, pin, irq, + irq_trigger(idx), irq_polarity(idx)); + } } - if (!first_notcon) - apic_printk(APIC_VERBOSE, " not connected.\n"); + if (notcon) + apic_printk(APIC_VERBOSE, + " (apicid-pin) not connected\n"); } /* @@ -1088,8 +1393,10 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, { struct IO_APIC_route_entry entry; +#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) return; +#endif memset(&entry, 0, sizeof(entry)); @@ -1124,7 +1431,10 @@ __apicdebuginit(void) print_IO_APIC(void) union IO_APIC_reg_00 reg_00; union IO_APIC_reg_01 reg_01; union IO_APIC_reg_02 reg_02; + union IO_APIC_reg_03 reg_03; unsigned long flags; + struct irq_cfg *cfg; + unsigned int irq; if (apic_verbosity == APIC_QUIET) return; @@ -1147,12 +1457,16 @@ __apicdebuginit(void) print_IO_APIC(void) reg_01.raw = io_apic_read(apic, 1); if (reg_01.bits.version >= 0x10) reg_02.raw = io_apic_read(apic, 2); + if (reg_01.bits.version >= 0x20) + reg_03.raw = io_apic_read(apic, 3); spin_unlock_irqrestore(&ioapic_lock, flags); printk("\n"); printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid); printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); + printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type); + printk(KERN_DEBUG "....... : LTS : %X\n", reg_00.bits.LTS); printk(KERN_DEBUG ".... register #01: %08X\n", *(int *)®_01); printk(KERN_DEBUG "....... : max redirection entries: %04X\n", reg_01.bits.entries); @@ -1160,11 +1474,27 @@ __apicdebuginit(void) print_IO_APIC(void) printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ); printk(KERN_DEBUG "....... : IO APIC version: %04X\n", reg_01.bits.version); - if (reg_01.bits.version >= 0x10) { + /* + * Some Intel chipsets with IO APIC VERSION of 0x1? don't have reg_02, + * but the value of reg_02 is read as the previous read register + * value, so ignore it if reg_02 == reg_01. + */ + if (reg_01.bits.version >= 0x10 && reg_02.raw != reg_01.raw) { printk(KERN_DEBUG ".... register #02: %08X\n", reg_02.raw); printk(KERN_DEBUG "....... : arbitration: %02X\n", reg_02.bits.arbitration); } + /* + * Some Intel chipsets with IO APIC VERSION of 0x2? don't have reg_02 + * or reg_03, but the value of reg_0[23] is read as the previous read + * register value, so ignore it if reg_03 == reg_0[12]. + */ + if (reg_01.bits.version >= 0x20 && reg_03.raw != reg_02.raw && + reg_03.raw != reg_01.raw) { + printk(KERN_DEBUG ".... register #03: %08X\n", reg_03.raw); + printk(KERN_DEBUG "....... : Boot DT : %X\n", reg_03.bits.boot_DT); + } + printk(KERN_DEBUG ".... IRQ redirection table:\n"); printk(KERN_DEBUG " NR Dst Mask Trig IRR Pol" @@ -1193,16 +1523,16 @@ __apicdebuginit(void) print_IO_APIC(void) } } printk(KERN_DEBUG "IRQ to pin mappings:\n"); - for (i = 0; i < NR_IRQS; i++) { - struct irq_pin_list *entry = irq_2_pin + i; - if (entry->pin < 0) + for_each_irq_cfg(irq, cfg) { + struct irq_pin_list *entry = cfg->irq_2_pin; + if (!entry) continue; - printk(KERN_DEBUG "IRQ%d ", i); + printk(KERN_DEBUG "IRQ%d ", irq); for (;;) { printk("-> %d:%d", entry->apic, entry->pin); if (!entry->next) break; - entry = irq_2_pin + entry->next; + entry = entry->next; } printk("\n"); } @@ -1236,7 +1566,7 @@ __apicdebuginit(void) print_APIC_bitfield(int base) __apicdebuginit(void) print_local_APIC(void *dummy) { unsigned int v, ver, maxlvt; - unsigned long icr; + u64 icr; if (apic_verbosity == APIC_QUIET) return; @@ -1253,20 +1583,31 @@ __apicdebuginit(void) print_local_APIC(void *dummy) v = apic_read(APIC_TASKPRI); printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK); - v = apic_read(APIC_ARBPRI); - printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v, - v & APIC_ARBPRI_MASK); - v = apic_read(APIC_PROCPRI); - printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v); + if (APIC_INTEGRATED(ver)) { /* !82489DX */ + if (!APIC_XAPIC(ver)) { + v = apic_read(APIC_ARBPRI); + printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v, + v & APIC_ARBPRI_MASK); + } + v = apic_read(APIC_PROCPRI); + printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v); + } + + /* + * Remote read supported only in the 82489DX and local APIC for + * Pentium processors. + */ + if (!APIC_INTEGRATED(ver) || maxlvt == 3) { + v = apic_read(APIC_RRR); + printk(KERN_DEBUG "... APIC RRR: %08x\n", v); + } - v = apic_read(APIC_EOI); - printk(KERN_DEBUG "... APIC EOI: %08x\n", v); - v = apic_read(APIC_RRR); - printk(KERN_DEBUG "... APIC RRR: %08x\n", v); v = apic_read(APIC_LDR); printk(KERN_DEBUG "... APIC LDR: %08x\n", v); - v = apic_read(APIC_DFR); - printk(KERN_DEBUG "... APIC DFR: %08x\n", v); + if (!x2apic_enabled()) { + v = apic_read(APIC_DFR); + printk(KERN_DEBUG "... APIC DFR: %08x\n", v); + } v = apic_read(APIC_SPIV); printk(KERN_DEBUG "... APIC SPIV: %08x\n", v); @@ -1277,8 +1618,13 @@ __apicdebuginit(void) print_local_APIC(void *dummy) printk(KERN_DEBUG "... APIC IRR field:\n"); print_APIC_bitfield(APIC_IRR); - v = apic_read(APIC_ESR); - printk(KERN_DEBUG "... APIC ESR: %08x\n", v); + if (APIC_INTEGRATED(ver)) { /* !82489DX */ + if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ + apic_write(APIC_ESR, 0); + + v = apic_read(APIC_ESR); + printk(KERN_DEBUG "... APIC ESR: %08x\n", v); + } icr = apic_icr_read(); printk(KERN_DEBUG "... APIC ICR: %08x\n", (u32)icr); @@ -1312,7 +1658,12 @@ __apicdebuginit(void) print_local_APIC(void *dummy) __apicdebuginit(void) print_all_local_APICs(void) { - on_each_cpu(print_local_APIC, NULL, 1); + int cpu; + + preempt_disable(); + for_each_online_cpu(cpu) + smp_call_function_single(cpu, print_local_APIC, NULL, 1); + preempt_enable(); } __apicdebuginit(void) print_PIC(void) @@ -1359,17 +1710,22 @@ __apicdebuginit(int) print_all_ICs(void) fs_initcall(print_all_ICs); +/* Where if anywhere is the i8259 connect in external int mode */ +static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; + void __init enable_IO_APIC(void) { union IO_APIC_reg_01 reg_01; int i8259_apic, i8259_pin; - int i, apic; + int apic; unsigned long flags; - for (i = 0; i < PIN_MAP_SIZE; i++) { - irq_2_pin[i].pin = -1; - irq_2_pin[i].next = 0; - } +#ifdef CONFIG_X86_32 + int i; + if (!pirqs_enabled) + for (i = 0; i < MAX_PIRQS; i++) + pirq_entries[i] = -1; +#endif /* * The number of IO-APIC IRQ registers (== #pins): @@ -1399,6 +1755,10 @@ void __init enable_IO_APIC(void) } found_i8259: /* Look to see what if the MP table has reported the ExtINT */ + /* If we could not find the appropriate pin by looking at the ioapic + * the i8259 probably is not connected the ioapic but give the + * mptable a chance anyway. + */ i8259_pin = find_isa_irq_pin(0, mp_ExtINT); i8259_apic = find_isa_irq_apic(0, mp_ExtINT); /* Trust the MP table if nothing is setup in the hardware */ @@ -1458,6 +1818,133 @@ void disable_IO_APIC(void) disconnect_bsp_APIC(ioapic_i8259.pin != -1); } +#ifdef CONFIG_X86_32 +/* + * function to set the IO-APIC physical IDs based on the + * values stored in the MPC table. + * + * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999 + */ + +static void __init setup_ioapic_ids_from_mpc(void) +{ + union IO_APIC_reg_00 reg_00; + physid_mask_t phys_id_present_map; + int apic; + int i; + unsigned char old_id; + unsigned long flags; + + if (x86_quirks->setup_ioapic_ids && x86_quirks->setup_ioapic_ids()) + return; + + /* + * Don't check I/O APIC IDs for xAPIC systems. They have + * no meaning without the serial APIC bus. + */ + if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + || APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) + return; + /* + * This is broken; anything with a real cpu count has to + * circumvent this idiocy regardless. + */ + phys_id_present_map = ioapic_phys_id_map(phys_cpu_present_map); + + /* + * Set the IOAPIC ID to the value stored in the MPC table. + */ + for (apic = 0; apic < nr_ioapics; apic++) { + + /* Read the register 0 value */ + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(apic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + + old_id = mp_ioapics[apic].mp_apicid; + + if (mp_ioapics[apic].mp_apicid >= get_physical_broadcast()) { + printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n", + apic, mp_ioapics[apic].mp_apicid); + printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", + reg_00.bits.ID); + mp_ioapics[apic].mp_apicid = reg_00.bits.ID; + } + + /* + * Sanity check, is the ID really free? Every APIC in a + * system must have a unique ID or we get lots of nice + * 'stuck on smp_invalidate_needed IPI wait' messages. + */ + if (check_apicid_used(phys_id_present_map, + mp_ioapics[apic].mp_apicid)) { + printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n", + apic, mp_ioapics[apic].mp_apicid); + for (i = 0; i < get_physical_broadcast(); i++) + if (!physid_isset(i, phys_id_present_map)) + break; + if (i >= get_physical_broadcast()) + panic("Max APIC ID exceeded!\n"); + printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", + i); + physid_set(i, phys_id_present_map); + mp_ioapics[apic].mp_apicid = i; + } else { + physid_mask_t tmp; + tmp = apicid_to_cpu_present(mp_ioapics[apic].mp_apicid); + apic_printk(APIC_VERBOSE, "Setting %d in the " + "phys_id_present_map\n", + mp_ioapics[apic].mp_apicid); + physids_or(phys_id_present_map, phys_id_present_map, tmp); + } + + + /* + * We need to adjust the IRQ routing table + * if the ID changed. + */ + if (old_id != mp_ioapics[apic].mp_apicid) + for (i = 0; i < mp_irq_entries; i++) + if (mp_irqs[i].mp_dstapic == old_id) + mp_irqs[i].mp_dstapic + = mp_ioapics[apic].mp_apicid; + + /* + * Read the right value from the MPC table and + * write it into the ID register. + */ + apic_printk(APIC_VERBOSE, KERN_INFO + "...changing IO-APIC physical APIC ID to %d ...", + mp_ioapics[apic].mp_apicid); + + reg_00.bits.ID = mp_ioapics[apic].mp_apicid; + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(apic, 0, reg_00.raw); + spin_unlock_irqrestore(&ioapic_lock, flags); + + /* + * Sanity check + */ + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(apic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + if (reg_00.bits.ID != mp_ioapics[apic].mp_apicid) + printk("could not set ID!\n"); + else + apic_printk(APIC_VERBOSE, " ok.\n"); + } +} +#endif + +int no_timer_check __initdata; + +static int __init notimercheck(char *s) +{ + no_timer_check = 1; + return 1; +} +__setup("no_timer_check", notimercheck); + /* * There is a nasty bug in some older SMP boards, their mptable lies * about the timer IRQ. We do the following to work around the situation: @@ -1471,6 +1958,9 @@ static int __init timer_irq_works(void) unsigned long t1 = jiffies; unsigned long flags; + if (no_timer_check) + return 1; + local_save_flags(flags); local_irq_enable(); /* Let ten ticks pass... */ @@ -1531,9 +2021,11 @@ static unsigned int startup_ioapic_irq(unsigned int irq) return was_pending; } +#ifdef CONFIG_X86_64 static int ioapic_retrigger_irq(unsigned int irq) { - struct irq_cfg *cfg = &irq_cfg[irq]; + + struct irq_cfg *cfg = irq_cfg(irq); unsigned long flags; spin_lock_irqsave(&vector_lock, flags); @@ -1542,6 +2034,14 @@ static int ioapic_retrigger_irq(unsigned int irq) return 1; } +#else +static int ioapic_retrigger_irq(unsigned int irq) +{ + send_IPI_self(irq_cfg(irq)->vector); + + return 1; +} +#endif /* * Level and edge triggered IO-APIC interrupts need different handling, @@ -1580,11 +2080,11 @@ static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration); */ static void migrate_ioapic_irq(int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; - struct irq_desc *desc = irq_desc + irq; + struct irq_cfg *cfg; + struct irq_desc *desc; cpumask_t tmp, cleanup_mask; struct irte irte; - int modify_ioapic_rte = desc->status & IRQ_LEVEL; + int modify_ioapic_rte; unsigned int dest; unsigned long flags; @@ -1598,9 +2098,12 @@ static void migrate_ioapic_irq(int irq, cpumask_t mask) if (assign_irq_vector(irq, mask)) return; + cfg = irq_cfg(irq); cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); + desc = irq_to_desc(irq); + modify_ioapic_rte = desc->status & IRQ_LEVEL; if (modify_ioapic_rte) { spin_lock_irqsave(&ioapic_lock, flags); __target_IO_APIC_irq(irq, dest, cfg->vector); @@ -1622,18 +2125,19 @@ static void migrate_ioapic_irq(int irq, cpumask_t mask) cfg->move_in_progress = 0; } - irq_desc[irq].affinity = mask; + desc->affinity = mask; } static int migrate_irq_remapped_level(int irq) { int ret = -1; + struct irq_desc *desc = irq_to_desc(irq); mask_IO_APIC_irq(irq); if (io_apic_level_ack_pending(irq)) { /* - * Interrupt in progress. Migrating irq now will change the + * Interrupt in progress. Migrating irq now will change the * vector information in the IO-APIC RTE and that will confuse * the EOI broadcast performed by cpu. * So, delay the irq migration to the next instance. @@ -1643,11 +2147,11 @@ static int migrate_irq_remapped_level(int irq) } /* everthing is clear. we have right of way */ - migrate_ioapic_irq(irq, irq_desc[irq].pending_mask); + migrate_ioapic_irq(irq, desc->pending_mask); ret = 0; - irq_desc[irq].status &= ~IRQ_MOVE_PENDING; - cpus_clear(irq_desc[irq].pending_mask); + desc->status &= ~IRQ_MOVE_PENDING; + cpus_clear(desc->pending_mask); unmask: unmask_IO_APIC_irq(irq); @@ -1656,10 +2160,10 @@ unmask: static void ir_irq_migration(struct work_struct *work) { - int irq; + unsigned int irq; + struct irq_desc *desc; - for (irq = 0; irq < NR_IRQS; irq++) { - struct irq_desc *desc = irq_desc + irq; + for_each_irq_desc(irq, desc) { if (desc->status & IRQ_MOVE_PENDING) { unsigned long flags; @@ -1671,8 +2175,7 @@ static void ir_irq_migration(struct work_struct *work) continue; } - desc->chip->set_affinity(irq, - irq_desc[irq].pending_mask); + desc->chip->set_affinity(irq, desc->pending_mask); spin_unlock_irqrestore(&desc->lock, flags); } } @@ -1683,9 +2186,11 @@ static void ir_irq_migration(struct work_struct *work) */ static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) { - if (irq_desc[irq].status & IRQ_LEVEL) { - irq_desc[irq].status |= IRQ_MOVE_PENDING; - irq_desc[irq].pending_mask = mask; + struct irq_desc *desc = irq_to_desc(irq); + + if (desc->status & IRQ_LEVEL) { + desc->status |= IRQ_MOVE_PENDING; + desc->pending_mask = mask; migrate_irq_remapped_level(irq); return; } @@ -1698,7 +2203,9 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) { unsigned vector, me; ack_APIC_irq(); +#ifdef CONFIG_X86_64 exit_idle(); +#endif irq_enter(); me = smp_processor_id(); @@ -1707,11 +2214,12 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) struct irq_desc *desc; struct irq_cfg *cfg; irq = __get_cpu_var(vector_irq)[vector]; - if (irq >= NR_IRQS) + + desc = irq_to_desc(irq); + if (!desc) continue; - desc = irq_desc + irq; - cfg = irq_cfg + irq; + cfg = irq_cfg(irq); spin_lock(&desc->lock); if (!cfg->move_cleanup_count) goto unlock; @@ -1730,7 +2238,7 @@ unlock: static void irq_complete_move(unsigned int irq) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg = irq_cfg(irq); unsigned vector, me; if (likely(!cfg->move_in_progress)) @@ -1769,19 +2277,50 @@ static void ack_apic_edge(unsigned int irq) ack_APIC_irq(); } +atomic_t irq_mis_count; + static void ack_apic_level(unsigned int irq) { +#ifdef CONFIG_X86_32 + unsigned long v; + int i; +#endif int do_unmask_irq = 0; irq_complete_move(irq); #ifdef CONFIG_GENERIC_PENDING_IRQ /* If we are moving the irq we need to mask it */ - if (unlikely(irq_desc[irq].status & IRQ_MOVE_PENDING)) { + if (unlikely(irq_to_desc(irq)->status & IRQ_MOVE_PENDING)) { do_unmask_irq = 1; mask_IO_APIC_irq(irq); } #endif +#ifdef CONFIG_X86_32 + /* + * It appears there is an erratum which affects at least version 0x11 + * of I/O APIC (that's the 82093AA and cores integrated into various + * chipsets). Under certain conditions a level-triggered interrupt is + * erroneously delivered as edge-triggered one but the respective IRR + * bit gets set nevertheless. As a result the I/O unit expects an EOI + * message but it will never arrive and further interrupts are blocked + * from the source. The exact reason is so far unknown, but the + * phenomenon was observed when two consecutive interrupt requests + * from a given source get delivered to the same CPU and the source is + * temporarily disabled in between. + * + * A workaround is to simulate an EOI message manually. We achieve it + * by setting the trigger mode to edge and then to level when the edge + * trigger mode gets detected in the TMR of a local APIC for a + * level-triggered interrupt. We mask the source for the time of the + * operation to prevent an edge-triggered interrupt escaping meanwhile. + * The idea is from Manfred Spraul. --macro + */ + i = irq_cfg(irq)->vector; + + v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1)); +#endif + /* * We must acknowledge the irq before we move it or the acknowledge will * not propagate properly. @@ -1820,31 +2359,41 @@ static void ack_apic_level(unsigned int irq) move_masked_irq(irq); unmask_IO_APIC_irq(irq); } + +#ifdef CONFIG_X86_32 + if (!(v & (1 << (i & 0x1f)))) { + atomic_inc(&irq_mis_count); + spin_lock(&ioapic_lock); + __mask_and_edge_IO_APIC_irq(irq); + __unmask_and_level_IO_APIC_irq(irq); + spin_unlock(&ioapic_lock); + } +#endif } static struct irq_chip ioapic_chip __read_mostly = { - .name = "IO-APIC", - .startup = startup_ioapic_irq, - .mask = mask_IO_APIC_irq, - .unmask = unmask_IO_APIC_irq, - .ack = ack_apic_edge, - .eoi = ack_apic_level, + .name = "IO-APIC", + .startup = startup_ioapic_irq, + .mask = mask_IO_APIC_irq, + .unmask = unmask_IO_APIC_irq, + .ack = ack_apic_edge, + .eoi = ack_apic_level, #ifdef CONFIG_SMP - .set_affinity = set_ioapic_affinity_irq, + .set_affinity = set_ioapic_affinity_irq, #endif .retrigger = ioapic_retrigger_irq, }; #ifdef CONFIG_INTR_REMAP static struct irq_chip ir_ioapic_chip __read_mostly = { - .name = "IR-IO-APIC", - .startup = startup_ioapic_irq, - .mask = mask_IO_APIC_irq, - .unmask = unmask_IO_APIC_irq, - .ack = ack_x2apic_edge, - .eoi = ack_x2apic_level, + .name = "IR-IO-APIC", + .startup = startup_ioapic_irq, + .mask = mask_IO_APIC_irq, + .unmask = unmask_IO_APIC_irq, + .ack = ack_x2apic_edge, + .eoi = ack_x2apic_level, #ifdef CONFIG_SMP - .set_affinity = set_ir_ioapic_affinity_irq, + .set_affinity = set_ir_ioapic_affinity_irq, #endif .retrigger = ioapic_retrigger_irq, }; @@ -1853,6 +2402,8 @@ static struct irq_chip ir_ioapic_chip __read_mostly = { static inline void init_IO_APIC_traps(void) { int irq; + struct irq_desc *desc; + struct irq_cfg *cfg; /* * NOTE! The local APIC isn't very good at handling @@ -1865,8 +2416,8 @@ static inline void init_IO_APIC_traps(void) * Also, we've got to be careful not to trash gate * 0x80, because int 0x80 is hm, kind of importantish. ;) */ - for (irq = 0; irq < NR_IRQS ; irq++) { - if (IO_APIC_IRQ(irq) && !irq_cfg[irq].vector) { + for_each_irq_cfg(irq, cfg) { + if (IO_APIC_IRQ(irq) && !cfg->vector) { /* * Hmm.. We don't have an entry for this, * so default to an old-fashioned 8259 @@ -1874,27 +2425,33 @@ static inline void init_IO_APIC_traps(void) */ if (irq < 16) make_8259A_irq(irq); - else + else { + desc = irq_to_desc(irq); /* Strange. Oh, well.. */ - irq_desc[irq].chip = &no_irq_chip; + desc->chip = &no_irq_chip; + } } } } -static void unmask_lapic_irq(unsigned int irq) +/* + * The local APIC irq-chip implementation: + */ + +static void mask_lapic_irq(unsigned int irq) { unsigned long v; v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); + apic_write(APIC_LVT0, v | APIC_LVT_MASKED); } -static void mask_lapic_irq(unsigned int irq) +static void unmask_lapic_irq(unsigned int irq) { unsigned long v; v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v | APIC_LVT_MASKED); + apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); } static void ack_lapic_irq (unsigned int irq) @@ -1911,7 +2468,10 @@ static struct irq_chip lapic_chip __read_mostly = { static void lapic_register_intr(int irq) { - irq_desc[irq].status &= ~IRQ_LEVEL; + struct irq_desc *desc; + + desc = irq_to_desc(irq); + desc->status &= ~IRQ_LEVEL; set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq, "edge"); } @@ -1919,19 +2479,19 @@ static void lapic_register_intr(int irq) static void __init setup_nmi(void) { /* - * Dirty trick to enable the NMI watchdog ... + * Dirty trick to enable the NMI watchdog ... * We put the 8259A master into AEOI mode and * unmask on all local APICs LVT0 as NMI. * * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire') * is from Maciej W. Rozycki - so we do not have to EOI from * the NMI handler or the timer interrupt. - */ - printk(KERN_INFO "activating NMI Watchdog ..."); + */ + apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ..."); enable_NMI_through_LVT0(); - printk(" done.\n"); + apic_printk(APIC_VERBOSE, " done.\n"); } /* @@ -1948,12 +2508,17 @@ static inline void __init unlock_ExtINT_logic(void) unsigned char save_control, save_freq_select; pin = find_isa_irq_pin(8, mp_INT); + if (pin == -1) { + WARN_ON_ONCE(1); + return; + } apic = find_isa_irq_apic(8, mp_INT); - if (pin == -1) + if (apic == -1) { + WARN_ON_ONCE(1); return; + } entry0 = ioapic_read_entry(apic, pin); - clear_IO_APIC_pin(apic, pin); memset(&entry1, 0, sizeof(entry1)); @@ -1988,23 +2553,38 @@ static inline void __init unlock_ExtINT_logic(void) ioapic_write_entry(apic, pin, entry0); } +static int disable_timer_pin_1 __initdata; +/* Actually the next is obsolete, but keep it for paranoid reasons -AK */ +static int __init disable_timer_pin_setup(char *arg) +{ + disable_timer_pin_1 = 1; + return 0; +} +early_param("disable_timer_pin_1", disable_timer_pin_setup); + +int timer_through_8259 __initdata; + /* * This code may look a bit paranoid, but it's supposed to cooperate with * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ * is so screwy. Thanks to Brian Perkins for testing/hacking this beast * fanatically on his truly buggy board. * - * FIXME: really need to revamp this for modern platforms only. + * FIXME: really need to revamp this for all platforms. */ static inline void __init check_timer(void) { - struct irq_cfg *cfg = irq_cfg + 0; + struct irq_cfg *cfg = irq_cfg(0); 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: */ @@ -2013,10 +2593,18 @@ static inline void __init check_timer(void) /* * As IRQ0 is to be enabled in the 8259A, the virtual - * wire has to be disabled in the local APIC. + * wire has to be disabled in the local APIC. Also + * timer interrupts need to be acknowledged manually in + * the 8259A for the i82489DX when using the NMI + * watchdog as that APIC treats NMIs as level-triggered. + * The AEOI mode will finish them in the 8259A + * automatically. */ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); init_8259A(1); +#ifdef CONFIG_X86_32 + timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); +#endif pin1 = find_isa_irq_pin(0, mp_INT); apic1 = find_isa_irq_apic(0, mp_INT); @@ -2035,8 +2623,10 @@ static inline void __init check_timer(void) * 8259A. */ if (pin1 == -1) { +#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) panic("BIOS bug: timer not connected to IO-APIC"); +#endif pin1 = pin2; apic1 = apic2; no_pin1 = 1; @@ -2054,7 +2644,7 @@ static inline void __init check_timer(void) setup_timer_IRQ0_pin(apic1, pin1, cfg->vector); } unmask_IO_APIC_irq(0); - if (!no_timer_check && timer_irq_works()) { + if (timer_irq_works()) { if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); enable_8259A_irq(0); @@ -2063,8 +2653,10 @@ static inline void __init check_timer(void) clear_IO_APIC_pin(0, pin1); goto out; } +#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) panic("timer doesn't work through Interrupt-remapped IO-APIC"); +#endif clear_IO_APIC_pin(apic1, pin1); if (!no_pin1) apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: " @@ -2104,6 +2696,9 @@ static inline void __init check_timer(void) "through the IO-APIC - disabling NMI Watchdog!\n"); nmi_watchdog = NMI_NONE; } +#ifdef CONFIG_X86_32 + timer_ack = 0; +#endif apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer as Virtual Wire IRQ...\n"); @@ -2140,13 +2735,6 @@ out: local_irq_restore(flags); } -static int __init notimercheck(char *s) -{ - no_timer_check = 1; - return 1; -} -__setup("no_timer_check", notimercheck); - /* * Traditionally ISA IRQ2 is the cascade IRQ, and is not available * to devices. However there may be an I/O APIC pin available for @@ -2164,25 +2752,49 @@ __setup("no_timer_check", notimercheck); * the I/O APIC in all cases now. No actual device should request * it anyway. --macro */ -#define PIC_IRQS (1<<2) +#define PIC_IRQS (1 << PIC_CASCADE_IR) void __init setup_IO_APIC(void) { +#ifdef CONFIG_X86_32 + enable_IO_APIC(); +#else /* * calling enable_IO_APIC() is moved to setup_local_APIC for BP */ +#endif io_apic_irqs = ~PIC_IRQS; apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n"); - + /* + * Set up IO-APIC IRQ routing. + */ +#ifdef CONFIG_X86_32 + if (!acpi_ioapic) + setup_ioapic_ids_from_mpc(); +#endif sync_Arb_IDs(); setup_IO_APIC_irqs(); init_IO_APIC_traps(); check_timer(); } +/* + * Called after all the initialization is done. If we didnt find any + * APIC bugs then we can allow the modify fast path + */ + +static int __init io_apic_bug_finalize(void) +{ + if (sis_apic_bug == -1) + sis_apic_bug = 0; + return 0; +} + +late_initcall(io_apic_bug_finalize); + struct sysfs_ioapic_data { struct sys_device dev; struct IO_APIC_route_entry entry[0]; @@ -2270,32 +2882,51 @@ device_initcall(ioapic_init_sysfs); /* * Dynamic irq allocate and deallocation */ -int create_irq(void) +unsigned int create_irq_nr(unsigned int irq_want) { /* Allocate an unused irq */ - int irq; - int new; + unsigned int irq; + unsigned int new; unsigned long flags; + struct irq_cfg *cfg_new; + + irq_want = nr_irqs - 1; - irq = -ENOSPC; + irq = 0; spin_lock_irqsave(&vector_lock, flags); - for (new = (NR_IRQS - 1); new >= 0; new--) { + for (new = irq_want; new > 0; new--) { if (platform_legacy_irq(new)) continue; - if (irq_cfg[new].vector != 0) + cfg_new = irq_cfg(new); + if (cfg_new && cfg_new->vector != 0) continue; + /* check if need to create one */ + if (!cfg_new) + cfg_new = irq_cfg_alloc(new); if (__assign_irq_vector(new, TARGET_CPUS) == 0) irq = new; break; } spin_unlock_irqrestore(&vector_lock, flags); - if (irq >= 0) { + if (irq > 0) { dynamic_irq_init(irq); } return irq; } +int create_irq(void) +{ + int irq; + + irq = create_irq_nr(nr_irqs - 1); + + if (irq == 0) + irq = -1; + + return irq; +} + void destroy_irq(unsigned int irq) { unsigned long flags; @@ -2316,7 +2947,7 @@ void destroy_irq(unsigned int irq) #ifdef CONFIG_PCI_MSI static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; int err; unsigned dest; cpumask_t tmp; @@ -2326,6 +2957,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms if (err) return err; + cfg = irq_cfg(irq); cpus_and(tmp, cfg->domain, tmp); dest = cpu_mask_to_apicid(tmp); @@ -2383,10 +3015,11 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms #ifdef CONFIG_SMP static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; struct msi_msg msg; unsigned int dest; cpumask_t tmp; + struct irq_desc *desc; cpus_and(tmp, mask, cpu_online_map); if (cpus_empty(tmp)) @@ -2395,6 +3028,7 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) if (assign_irq_vector(irq, mask)) return; + cfg = irq_cfg(irq); cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); @@ -2406,7 +3040,8 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) msg.address_lo |= MSI_ADDR_DEST_ID(dest); write_msi_msg(irq, &msg); - irq_desc[irq].affinity = mask; + desc = irq_to_desc(irq); + desc->affinity = mask; } #ifdef CONFIG_INTR_REMAP @@ -2416,10 +3051,11 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) */ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; unsigned int dest; cpumask_t tmp, cleanup_mask; struct irte irte; + struct irq_desc *desc; cpus_and(tmp, mask, cpu_online_map); if (cpus_empty(tmp)) @@ -2431,6 +3067,7 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) if (assign_irq_vector(irq, mask)) return; + cfg = irq_cfg(irq); cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); @@ -2454,7 +3091,8 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) cfg->move_in_progress = 0; } - irq_desc[irq].affinity = mask; + desc = irq_to_desc(irq); + desc->affinity = mask; } #endif #endif /* CONFIG_SMP */ @@ -2507,7 +3145,7 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec) if (index < 0) { printk(KERN_ERR "Unable to allocate %d IRTE for PCI %s\n", nvec, - pci_name(dev)); + pci_name(dev)); return -ENOSPC; } return index; @@ -2528,7 +3166,7 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq) #ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { - struct irq_desc *desc = irq_desc + irq; + struct irq_desc *desc = irq_to_desc(irq); /* * irq migration in process context */ @@ -2538,16 +3176,34 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq) #endif set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge"); + dev_printk(KERN_DEBUG, &dev->dev, "irq %d for MSI/MSI-X\n", irq); + return 0; } +static unsigned int build_irq_for_pci_dev(struct pci_dev *dev) +{ + unsigned int irq; + + irq = dev->bus->number; + irq <<= 8; + irq |= dev->devfn; + irq <<= 12; + + return irq; +} + int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc) { - int irq, ret; + unsigned int irq; + int ret; + unsigned int irq_want; - irq = create_irq(); - if (irq < 0) - return irq; + irq_want = build_irq_for_pci_dev(dev) + 0x100; + + irq = create_irq_nr(irq_want); + if (irq == 0) + return -1; #ifdef CONFIG_INTR_REMAP if (!intr_remapping_enabled) @@ -2574,18 +3230,22 @@ error: int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) { - int irq, ret, sub_handle; + unsigned int irq; + int ret, sub_handle; struct msi_desc *desc; + unsigned int irq_want; + #ifdef CONFIG_INTR_REMAP struct intel_iommu *iommu = 0; int index = 0; #endif + irq_want = build_irq_for_pci_dev(dev) + 0x100; sub_handle = 0; list_for_each_entry(desc, &dev->msi_list, list) { - irq = create_irq(); - if (irq < 0) - return irq; + irq = create_irq_nr(irq_want--); + if (irq == 0) + return -1; #ifdef CONFIG_INTR_REMAP if (!intr_remapping_enabled) goto no_ir; @@ -2636,10 +3296,11 @@ void arch_teardown_msi_irq(unsigned int irq) #ifdef CONFIG_SMP static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; struct msi_msg msg; unsigned int dest; cpumask_t tmp; + struct irq_desc *desc; cpus_and(tmp, mask, cpu_online_map); if (cpus_empty(tmp)) @@ -2648,6 +3309,7 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) if (assign_irq_vector(irq, mask)) return; + cfg = irq_cfg(irq); cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); @@ -2659,7 +3321,8 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) msg.address_lo |= MSI_ADDR_DEST_ID(dest); dmar_msi_write(irq, &msg); - irq_desc[irq].affinity = mask; + desc = irq_to_desc(irq); + desc->affinity = mask; } #endif /* CONFIG_SMP */ @@ -2689,6 +3352,69 @@ int arch_setup_dmar_msi(unsigned int irq) } #endif +#ifdef CONFIG_HPET_TIMER + +#ifdef CONFIG_SMP +static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask) +{ + struct irq_cfg *cfg; + struct irq_desc *desc; + struct msi_msg msg; + unsigned int dest; + cpumask_t tmp; + + cpus_and(tmp, mask, cpu_online_map); + if (cpus_empty(tmp)) + return; + + if (assign_irq_vector(irq, mask)) + return; + + cfg = irq_cfg(irq); + cpus_and(tmp, cfg->domain, mask); + dest = cpu_mask_to_apicid(tmp); + + hpet_msi_read(irq, &msg); + + msg.data &= ~MSI_DATA_VECTOR_MASK; + msg.data |= MSI_DATA_VECTOR(cfg->vector); + msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; + msg.address_lo |= MSI_ADDR_DEST_ID(dest); + + hpet_msi_write(irq, &msg); + desc = irq_to_desc(irq); + desc->affinity = mask; +} +#endif /* CONFIG_SMP */ + +struct irq_chip hpet_msi_type = { + .name = "HPET_MSI", + .unmask = hpet_msi_unmask, + .mask = hpet_msi_mask, + .ack = ack_apic_edge, +#ifdef CONFIG_SMP + .set_affinity = hpet_msi_set_affinity, +#endif + .retrigger = ioapic_retrigger_irq, +}; + +int arch_setup_hpet_msi(unsigned int irq) +{ + int ret; + struct msi_msg msg; + + ret = msi_compose_msg(NULL, irq, &msg); + if (ret < 0) + return ret; + + hpet_msi_write(irq, &msg); + set_irq_chip_and_handler_name(irq, &hpet_msi_type, handle_edge_irq, + "edge"); + + return 0; +} +#endif + #endif /* CONFIG_PCI_MSI */ /* * Hypertransport interrupt support @@ -2713,9 +3439,10 @@ static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; unsigned int dest; cpumask_t tmp; + struct irq_desc *desc; cpus_and(tmp, mask, cpu_online_map); if (cpus_empty(tmp)) @@ -2724,11 +3451,13 @@ static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) if (assign_irq_vector(irq, mask)) return; + cfg = irq_cfg(irq); cpus_and(tmp, cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); target_ht_irq(irq, dest, cfg->vector); - irq_desc[irq].affinity = mask; + desc = irq_to_desc(irq); + desc->affinity = mask; } #endif @@ -2745,7 +3474,7 @@ static struct irq_chip ht_irq_chip = { int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) { - struct irq_cfg *cfg = irq_cfg + irq; + struct irq_cfg *cfg; int err; cpumask_t tmp; @@ -2755,6 +3484,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) struct ht_irq_msg msg; unsigned dest; + cfg = irq_cfg(irq); cpus_and(tmp, cfg->domain, tmp); dest = cpu_mask_to_apicid(tmp); @@ -2777,20 +3507,196 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) set_irq_chip_and_handler_name(irq, &ht_irq_chip, handle_edge_irq, "edge"); + + dev_printk(KERN_DEBUG, &dev->dev, "irq %d for HT\n", irq); } return err; } #endif /* CONFIG_HT_IRQ */ +#ifdef CONFIG_X86_64 +/* + * Re-target the irq to the specified CPU and enable the specified MMR located + * on the specified blade to allow the sending of MSIs to the specified CPU. + */ +int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, + unsigned long mmr_offset) +{ + const cpumask_t *eligible_cpu = get_cpu_mask(cpu); + struct irq_cfg *cfg; + int mmr_pnode; + unsigned long mmr_value; + struct uv_IO_APIC_route_entry *entry; + unsigned long flags; + int err; + + err = assign_irq_vector(irq, *eligible_cpu); + if (err != 0) + return err; + + spin_lock_irqsave(&vector_lock, flags); + set_irq_chip_and_handler_name(irq, &uv_irq_chip, handle_percpu_irq, + irq_name); + spin_unlock_irqrestore(&vector_lock, flags); + + cfg = irq_cfg(irq); + + mmr_value = 0; + entry = (struct uv_IO_APIC_route_entry *)&mmr_value; + BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); + + entry->vector = cfg->vector; + entry->delivery_mode = INT_DELIVERY_MODE; + entry->dest_mode = INT_DEST_MODE; + entry->polarity = 0; + entry->trigger = 0; + entry->mask = 0; + entry->dest = cpu_mask_to_apicid(*eligible_cpu); + + mmr_pnode = uv_blade_to_pnode(mmr_blade); + uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); + + return irq; +} + +/* + * Disable the specified MMR located on the specified blade so that MSIs are + * longer allowed to be sent. + */ +void arch_disable_uv_irq(int mmr_blade, unsigned long mmr_offset) +{ + unsigned long mmr_value; + struct uv_IO_APIC_route_entry *entry; + int mmr_pnode; + + mmr_value = 0; + entry = (struct uv_IO_APIC_route_entry *)&mmr_value; + BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); + + entry->mask = 1; + + mmr_pnode = uv_blade_to_pnode(mmr_blade); + uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); +} +#endif /* CONFIG_X86_64 */ + +int __init io_apic_get_redir_entries (int ioapic) +{ + union IO_APIC_reg_01 reg_01; + unsigned long flags; + + spin_lock_irqsave(&ioapic_lock, flags); + reg_01.raw = io_apic_read(ioapic, 1); + spin_unlock_irqrestore(&ioapic_lock, flags); + + return reg_01.bits.entries; +} + +int __init probe_nr_irqs(void) +{ + int idx; + int nr = 0; +#ifndef CONFIG_XEN + int nr_min = 32; +#else + int nr_min = NR_IRQS; +#endif + + for (idx = 0; idx < nr_ioapics; idx++) + nr += io_apic_get_redir_entries(idx) + 1; + + /* double it for hotplug and msi and nmi */ + nr <<= 1; + + /* something wrong ? */ + if (nr < nr_min) + nr = nr_min; + + return nr; +} + /* -------------------------------------------------------------------------- ACPI-based IOAPIC Configuration -------------------------------------------------------------------------- */ #ifdef CONFIG_ACPI -#define IO_APIC_MAX_ID 0xFE +#ifdef CONFIG_X86_32 +int __init io_apic_get_unique_id(int ioapic, int apic_id) +{ + union IO_APIC_reg_00 reg_00; + static physid_mask_t apic_id_map = PHYSID_MASK_NONE; + physid_mask_t tmp; + unsigned long flags; + int i = 0; -int __init io_apic_get_redir_entries (int ioapic) + /* + * The P4 platform supports up to 256 APIC IDs on two separate APIC + * buses (one for LAPICs, one for IOAPICs), where predecessors only + * supports up to 16 on one shared APIC bus. + * + * TBD: Expand LAPIC/IOAPIC support on P4-class systems to take full + * advantage of new APIC bus architecture. + */ + + if (physids_empty(apic_id_map)) + apic_id_map = ioapic_phys_id_map(phys_cpu_present_map); + + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(ioapic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + + if (apic_id >= get_physical_broadcast()) { + printk(KERN_WARNING "IOAPIC[%d]: Invalid apic_id %d, trying " + "%d\n", ioapic, apic_id, reg_00.bits.ID); + apic_id = reg_00.bits.ID; + } + + /* + * Every APIC in a system must have a unique ID or we get lots of nice + * 'stuck on smp_invalidate_needed IPI wait' messages. + */ + if (check_apicid_used(apic_id_map, apic_id)) { + + for (i = 0; i < get_physical_broadcast(); i++) { + if (!check_apicid_used(apic_id_map, i)) + break; + } + + if (i == get_physical_broadcast()) + panic("Max apic_id exceeded!\n"); + + printk(KERN_WARNING "IOAPIC[%d]: apic_id %d already used, " + "trying %d\n", ioapic, apic_id, i); + + apic_id = i; + } + + tmp = apicid_to_cpu_present(apic_id); + physids_or(apic_id_map, apic_id_map, tmp); + + if (reg_00.bits.ID != apic_id) { + reg_00.bits.ID = apic_id; + + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(ioapic, 0, reg_00.raw); + reg_00.raw = io_apic_read(ioapic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + + /* Sanity check */ + if (reg_00.bits.ID != apic_id) { + printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic); + return -1; + } + } + + apic_printk(APIC_VERBOSE, KERN_INFO + "IOAPIC[%d]: Assigned apic_id %d\n", ioapic, apic_id); + + return apic_id; +} + +int __init io_apic_get_version(int ioapic) { union IO_APIC_reg_01 reg_01; unsigned long flags; @@ -2799,9 +3705,9 @@ int __init io_apic_get_redir_entries (int ioapic) reg_01.raw = io_apic_read(ioapic, 1); spin_unlock_irqrestore(&ioapic_lock, flags); - return reg_01.bits.entries; + return reg_01.bits.version; } - +#endif int io_apic_set_pci_routing (int ioapic, int pin, int irq, int triggering, int polarity) { @@ -2853,6 +3759,7 @@ int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) void __init setup_ioapic_dest(void) { int pin, ioapic, irq, irq_entry; + struct irq_cfg *cfg; if (skip_ioapic_setup == 1) return; @@ -2868,7 +3775,8 @@ void __init setup_ioapic_dest(void) * when you have too many devices, because at that time only boot * cpu is online. */ - if (!irq_cfg[irq].vector) + cfg = irq_cfg(irq); + if (!cfg->vector) setup_IO_APIC_irq(ioapic, pin, irq, irq_trigger(irq_entry), irq_polarity(irq_entry)); @@ -2926,18 +3834,33 @@ void __init ioapic_init_mappings(void) struct resource *ioapic_res; int i; + irq_2_pin_init(); ioapic_res = ioapic_setup_resources(); for (i = 0; i < nr_ioapics; i++) { if (smp_found_config) { ioapic_phys = mp_ioapics[i].mp_apicaddr; +#ifdef CONFIG_X86_32 + if (!ioapic_phys) { + printk(KERN_ERR + "WARNING: bogus zero IO-APIC " + "address found in MPTABLE, " + "disabling IO/APIC support!\n"); + smp_found_config = 0; + skip_ioapic_setup = 1; + goto fake_ioapic_page; + } +#endif } else { +#ifdef CONFIG_X86_32 +fake_ioapic_page: +#endif ioapic_phys = (unsigned long) alloc_bootmem_pages(PAGE_SIZE); ioapic_phys = __pa(ioapic_phys); } set_fixmap_nocache(idx, ioapic_phys); apic_printk(APIC_VERBOSE, - "mapped IOAPIC to %016lx (%016lx)\n", + "mapped IOAPIC to %08lx (%08lx)\n", __fix_to_virt(idx), ioapic_phys); idx++; @@ -2971,4 +3894,3 @@ static int __init ioapic_insert_resources(void) /* Insert the IO APIC resources after PCI initialization has occured to handle * IO APICS that are mapped in on a BAR in PCI space. */ late_initcall(ioapic_insert_resources); - diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c deleted file mode 100644 index e710289f673..00000000000 --- a/arch/x86/kernel/io_apic_32.c +++ /dev/null @@ -1,2908 +0,0 @@ -/* - * Intel IO-APIC support for multi-Pentium hosts. - * - * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar, Hajnalka Szabo - * - * Many thanks to Stig Venaas for trying out countless experimental - * patches and reporting/debugging problems patiently! - * - * (c) 1999, Multiple IO-APIC support, developed by - * Ken-ichi Yaku <yaku@css1.kbnes.nec.co.jp> and - * Hidemi Kishimoto <kisimoto@css1.kbnes.nec.co.jp>, - * further tested and cleaned up by Zach Brown <zab@redhat.com> - * and Ingo Molnar <mingo@redhat.com> - * - * Fixes - * Maciej W. Rozycki : Bits for genuine 82489DX APICs; - * thanks to Eric Gilmore - * and Rolf G. Tews - * for testing these extensively - * Paul Diefenbaugh : Added full ACPI support - */ - -#include <linux/mm.h> -#include <linux/interrupt.h> -#include <linux/init.h> -#include <linux/delay.h> -#include <linux/sched.h> -#include <linux/bootmem.h> -#include <linux/mc146818rtc.h> -#include <linux/compiler.h> -#include <linux/acpi.h> -#include <linux/module.h> -#include <linux/sysdev.h> -#include <linux/pci.h> -#include <linux/msi.h> -#include <linux/htirq.h> -#include <linux/freezer.h> -#include <linux/kthread.h> -#include <linux/jiffies.h> /* time_after() */ - -#include <asm/io.h> -#include <asm/smp.h> -#include <asm/desc.h> -#include <asm/timer.h> -#include <asm/i8259.h> -#include <asm/nmi.h> -#include <asm/msidef.h> -#include <asm/hypertransport.h> -#include <asm/setup.h> - -#include <mach_apic.h> -#include <mach_apicdef.h> - -#define __apicdebuginit(type) static type __init - -int (*ioapic_renumber_irq)(int ioapic, int irq); -atomic_t irq_mis_count; - -/* Where if anywhere is the i8259 connect in external int mode */ -static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; - -static DEFINE_SPINLOCK(ioapic_lock); -DEFINE_SPINLOCK(vector_lock); - -int timer_through_8259 __initdata; - -/* - * Is the SiS APIC rmw bug present ? - * -1 = don't know, 0 = no, 1 = yes - */ -int sis_apic_bug = -1; - -/* - * # of IRQ routing registers - */ -int nr_ioapic_registers[MAX_IO_APICS]; - -/* I/O APIC entries */ -struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; -int nr_ioapics; - -/* MP IRQ source entries */ -struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; - -/* # of MP IRQ source entries */ -int mp_irq_entries; - -#if defined (CONFIG_MCA) || defined (CONFIG_EISA) -int mp_bus_id_to_type[MAX_MP_BUSSES]; -#endif - -DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); - -static int disable_timer_pin_1 __initdata; - -/* - * Rough estimation of how many shared IRQs there are, can - * be changed anytime. - */ -#define MAX_PLUS_SHARED_IRQS NR_IRQS -#define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS) - -/* - * This is performance-critical, we want to do it O(1) - * - * the indexing order of this array favors 1:1 mappings - * between pins and IRQs. - */ - -static struct irq_pin_list { - int apic, pin, next; -} irq_2_pin[PIN_MAP_SIZE]; - -struct io_apic { - unsigned int index; - unsigned int unused[3]; - unsigned int data; -}; - -static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) -{ - return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) - + (mp_ioapics[idx].mp_apicaddr & ~PAGE_MASK); -} - -static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) -{ - struct io_apic __iomem *io_apic = io_apic_base(apic); - writel(reg, &io_apic->index); - return readl(&io_apic->data); -} - -static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) -{ - struct io_apic __iomem *io_apic = io_apic_base(apic); - writel(reg, &io_apic->index); - writel(value, &io_apic->data); -} - -/* - * Re-write a value: to be used for read-modify-write - * cycles where the read already set up the index register. - * - * Older SiS APIC requires we rewrite the index register - */ -static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value) -{ - volatile struct io_apic __iomem *io_apic = io_apic_base(apic); - if (sis_apic_bug) - writel(reg, &io_apic->index); - writel(value, &io_apic->data); -} - -union entry_union { - struct { u32 w1, w2; }; - struct IO_APIC_route_entry entry; -}; - -static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) -{ - union entry_union eu; - unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); - eu.w1 = io_apic_read(apic, 0x10 + 2 * pin); - eu.w2 = io_apic_read(apic, 0x11 + 2 * pin); - spin_unlock_irqrestore(&ioapic_lock, flags); - return eu.entry; -} - -/* - * When we write a new IO APIC routing entry, we need to write the high - * word first! If the mask bit in the low word is clear, we will enable - * the interrupt, and we need to make sure the entry is fully populated - * before that happens. - */ -static void -__ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) -{ - union entry_union eu; - eu.entry = e; - io_apic_write(apic, 0x11 + 2*pin, eu.w2); - io_apic_write(apic, 0x10 + 2*pin, eu.w1); -} - -static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) -{ - unsigned long flags; - spin_lock_irqsave(&ioapic_lock, flags); - __ioapic_write_entry(apic, pin, e); - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -/* - * When we mask an IO APIC routing entry, we need to write the low - * word first, in order to set the mask bit before we change the - * high bits! - */ -static void ioapic_mask_entry(int apic, int pin) -{ - unsigned long flags; - union entry_union eu = { .entry.mask = 1 }; - - spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(apic, 0x10 + 2*pin, eu.w1); - io_apic_write(apic, 0x11 + 2*pin, eu.w2); - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -/* - * The common case is 1:1 IRQ<->pin mappings. Sometimes there are - * shared ISA-space IRQs, so we have to support them. We are super - * fast in the common case, and fast for shared ISA-space IRQs. - */ -static void add_pin_to_irq(unsigned int irq, int apic, int pin) -{ - static int first_free_entry = NR_IRQS; - struct irq_pin_list *entry = irq_2_pin + irq; - - while (entry->next) - entry = irq_2_pin + entry->next; - - if (entry->pin != -1) { - entry->next = first_free_entry; - entry = irq_2_pin + entry->next; - if (++first_free_entry >= PIN_MAP_SIZE) - panic("io_apic.c: whoops"); - } - entry->apic = apic; - entry->pin = pin; -} - -/* - * Reroute an IRQ to a different pin. - */ -static void __init replace_pin_at_irq(unsigned int irq, - int oldapic, int oldpin, - int newapic, int newpin) -{ - struct irq_pin_list *entry = irq_2_pin + irq; - - while (1) { - if (entry->apic == oldapic && entry->pin == oldpin) { - entry->apic = newapic; - entry->pin = newpin; - } - if (!entry->next) - break; - entry = irq_2_pin + entry->next; - } -} - -static void __modify_IO_APIC_irq(unsigned int irq, unsigned long enable, unsigned long disable) -{ - struct irq_pin_list *entry = irq_2_pin + irq; - unsigned int pin, reg; - - for (;;) { - pin = entry->pin; - if (pin == -1) - break; - reg = io_apic_read(entry->apic, 0x10 + pin*2); - reg &= ~disable; - reg |= enable; - io_apic_modify(entry->apic, 0x10 + pin*2, reg); - if (!entry->next) - break; - entry = irq_2_pin + entry->next; - } -} - -/* mask = 1 */ -static void __mask_IO_APIC_irq(unsigned int irq) -{ - __modify_IO_APIC_irq(irq, IO_APIC_REDIR_MASKED, 0); -} - -/* mask = 0 */ -static void __unmask_IO_APIC_irq(unsigned int irq) -{ - __modify_IO_APIC_irq(irq, 0, IO_APIC_REDIR_MASKED); -} - -/* mask = 1, trigger = 0 */ -static void __mask_and_edge_IO_APIC_irq(unsigned int irq) -{ - __modify_IO_APIC_irq(irq, IO_APIC_REDIR_MASKED, - IO_APIC_REDIR_LEVEL_TRIGGER); -} - -/* mask = 0, trigger = 1 */ -static void __unmask_and_level_IO_APIC_irq(unsigned int irq) -{ - __modify_IO_APIC_irq(irq, IO_APIC_REDIR_LEVEL_TRIGGER, - IO_APIC_REDIR_MASKED); -} - -static void mask_IO_APIC_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - __mask_IO_APIC_irq(irq); - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -static void unmask_IO_APIC_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - __unmask_IO_APIC_irq(irq); - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) -{ - struct IO_APIC_route_entry entry; - - /* Check delivery_mode to be sure we're not clearing an SMI pin */ - entry = ioapic_read_entry(apic, pin); - if (entry.delivery_mode == dest_SMI) - return; - - /* - * Disable it in the IO-APIC irq-routing table: - */ - ioapic_mask_entry(apic, pin); -} - -static void clear_IO_APIC(void) -{ - int apic, pin; - - for (apic = 0; apic < nr_ioapics; apic++) - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) - clear_IO_APIC_pin(apic, pin); -} - -#ifdef CONFIG_SMP -static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t cpumask) -{ - unsigned long flags; - int pin; - struct irq_pin_list *entry = irq_2_pin + irq; - unsigned int apicid_value; - cpumask_t tmp; - - cpus_and(tmp, cpumask, cpu_online_map); - if (cpus_empty(tmp)) - tmp = TARGET_CPUS; - - cpus_and(cpumask, tmp, CPU_MASK_ALL); - - apicid_value = cpu_mask_to_apicid(cpumask); - /* Prepare to do the io_apic_write */ - apicid_value = apicid_value << 24; - spin_lock_irqsave(&ioapic_lock, flags); - for (;;) { - pin = entry->pin; - if (pin == -1) - break; - io_apic_write(entry->apic, 0x10 + 1 + pin*2, apicid_value); - if (!entry->next) - break; - entry = irq_2_pin + entry->next; - } - irq_desc[irq].affinity = cpumask; - spin_unlock_irqrestore(&ioapic_lock, flags); -} - -#if defined(CONFIG_IRQBALANCE) -# include <asm/processor.h> /* kernel_thread() */ -# include <linux/kernel_stat.h> /* kstat */ -# include <linux/slab.h> /* kmalloc() */ -# include <linux/timer.h> - -#define IRQBALANCE_CHECK_ARCH -999 -#define MAX_BALANCED_IRQ_INTERVAL (5*HZ) -#define MIN_BALANCED_IRQ_INTERVAL (HZ/2) -#define BALANCED_IRQ_MORE_DELTA (HZ/10) -#define BALANCED_IRQ_LESS_DELTA (HZ) - -static int irqbalance_disabled __read_mostly = IRQBALANCE_CHECK_ARCH; -static int physical_balance __read_mostly; -static long balanced_irq_interval __read_mostly = MAX_BALANCED_IRQ_INTERVAL; - -static struct irq_cpu_info { - unsigned long *last_irq; - unsigned long *irq_delta; - unsigned long irq; -} irq_cpu_data[NR_CPUS]; - -#define CPU_IRQ(cpu) (irq_cpu_data[cpu].irq) -#define LAST_CPU_IRQ(cpu, irq) (irq_cpu_data[cpu].last_irq[irq]) -#define IRQ_DELTA(cpu, irq) (irq_cpu_data[cpu].irq_delta[irq]) - -#define IDLE_ENOUGH(cpu,now) \ - (idle_cpu(cpu) && ((now) - per_cpu(irq_stat, (cpu)).idle_timestamp > 1)) - -#define IRQ_ALLOWED(cpu, allowed_mask) cpu_isset(cpu, allowed_mask) - -#define CPU_TO_PACKAGEINDEX(i) (first_cpu(per_cpu(cpu_sibling_map, i))) - -static cpumask_t balance_irq_affinity[NR_IRQS] = { - [0 ... NR_IRQS-1] = CPU_MASK_ALL -}; - -void set_balance_irq_affinity(unsigned int irq, cpumask_t mask) -{ - balance_irq_affinity[irq] = mask; -} - -static unsigned long move(int curr_cpu, cpumask_t allowed_mask, - unsigned long now, int direction) -{ - int search_idle = 1; - int cpu = curr_cpu; - - goto inside; - - do { - if (unlikely(cpu == curr_cpu)) - search_idle = 0; -inside: - if (direction == 1) { - cpu++; - if (cpu >= NR_CPUS) - cpu = 0; - } else { - cpu--; - if (cpu == -1) - cpu = NR_CPUS-1; - } - } while (!cpu_online(cpu) || !IRQ_ALLOWED(cpu, allowed_mask) || - (search_idle && !IDLE_ENOUGH(cpu, now))); - - return cpu; -} - -static inline void balance_irq(int cpu, int irq) -{ - unsigned long now = jiffies; - cpumask_t allowed_mask; - unsigned int new_cpu; - - if (irqbalance_disabled) - return; - - cpus_and(allowed_mask, cpu_online_map, balance_irq_affinity[irq]); - new_cpu = move(cpu, allowed_mask, now, 1); - if (cpu != new_cpu) - set_pending_irq(irq, cpumask_of_cpu(new_cpu)); -} - -static inline void rotate_irqs_among_cpus(unsigned long useful_load_threshold) -{ - int i, j; - - for_each_online_cpu(i) { - for (j = 0; j < NR_IRQS; j++) { - if (!irq_desc[j].action) - continue; - /* Is it a significant load ? */ - if (IRQ_DELTA(CPU_TO_PACKAGEINDEX(i), j) < - useful_load_threshold) - continue; - balance_irq(i, j); - } - } - balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL, - balanced_irq_interval - BALANCED_IRQ_LESS_DELTA); - return; -} - -static void do_irq_balance(void) -{ - int i, j; - unsigned long max_cpu_irq = 0, min_cpu_irq = (~0); - unsigned long move_this_load = 0; - int max_loaded = 0, min_loaded = 0; - int load; - unsigned long useful_load_threshold = balanced_irq_interval + 10; - int selected_irq; - int tmp_loaded, first_attempt = 1; - unsigned long tmp_cpu_irq; - unsigned long imbalance = 0; - cpumask_t allowed_mask, target_cpu_mask, tmp; - - for_each_possible_cpu(i) { - int package_index; - CPU_IRQ(i) = 0; - if (!cpu_online(i)) - continue; - package_index = CPU_TO_PACKAGEINDEX(i); - for (j = 0; j < NR_IRQS; j++) { - unsigned long value_now, delta; - /* Is this an active IRQ or balancing disabled ? */ - if (!irq_desc[j].action || irq_balancing_disabled(j)) - continue; - if (package_index == i) - IRQ_DELTA(package_index, j) = 0; - /* Determine the total count per processor per IRQ */ - value_now = (unsigned long) kstat_cpu(i).irqs[j]; - - /* Determine the activity per processor per IRQ */ - delta = value_now - LAST_CPU_IRQ(i, j); - - /* Update last_cpu_irq[][] for the next time */ - LAST_CPU_IRQ(i, j) = value_now; - - /* Ignore IRQs whose rate is less than the clock */ - if (delta < useful_load_threshold) - continue; - /* update the load for the processor or package total */ - IRQ_DELTA(package_index, j) += delta; - - /* Keep track of the higher numbered sibling as well */ - if (i != package_index) - CPU_IRQ(i) += delta; - /* - * We have sibling A and sibling B in the package - * - * cpu_irq[A] = load for cpu A + load for cpu B - * cpu_irq[B] = load for cpu B - */ - CPU_IRQ(package_index) += delta; - } - } - /* Find the least loaded processor package */ - for_each_online_cpu(i) { - if (i != CPU_TO_PACKAGEINDEX(i)) - continue; - if (min_cpu_irq > CPU_IRQ(i)) { - min_cpu_irq = CPU_IRQ(i); - min_loaded = i; - } - } - max_cpu_irq = ULONG_MAX; - -tryanothercpu: - /* - * Look for heaviest loaded processor. - * We may come back to get the next heaviest loaded processor. - * Skip processors with trivial loads. - */ - tmp_cpu_irq = 0; - tmp_loaded = -1; - for_each_online_cpu(i) { - if (i != CPU_TO_PACKAGEINDEX(i)) - continue; - if (max_cpu_irq <= CPU_IRQ(i)) - continue; - if (tmp_cpu_irq < CPU_IRQ(i)) { - tmp_cpu_irq = CPU_IRQ(i); - tmp_loaded = i; - } - } - - if (tmp_loaded == -1) { - /* - * In the case of small number of heavy interrupt sources, - * loading some of the cpus too much. We use Ingo's original - * approach to rotate them around. - */ - if (!first_attempt && imbalance >= useful_load_threshold) { - rotate_irqs_among_cpus(useful_load_threshold); - return; - } - goto not_worth_the_effort; - } - - first_attempt = 0; /* heaviest search */ - max_cpu_irq = tmp_cpu_irq; /* load */ - max_loaded = tmp_loaded; /* processor */ - imbalance = (max_cpu_irq - min_cpu_irq) / 2; - - /* - * if imbalance is less than approx 10% of max load, then - * observe diminishing returns action. - quit - */ - if (imbalance < (max_cpu_irq >> 3)) - goto not_worth_the_effort; - -tryanotherirq: - /* if we select an IRQ to move that can't go where we want, then - * see if there is another one to try. - */ - move_this_load = 0; - selected_irq = -1; - for (j = 0; j < NR_IRQS; j++) { - /* Is this an active IRQ? */ - if (!irq_desc[j].action) - continue; - if (imbalance <= IRQ_DELTA(max_loaded, j)) - continue; - /* Try to find the IRQ that is closest to the imbalance - * without going over. - */ - if (move_this_load < IRQ_DELTA(max_loaded, j)) { - move_this_load = IRQ_DELTA(max_loaded, j); - selected_irq = j; - } - } - if (selected_irq == -1) - goto tryanothercpu; - - imbalance = move_this_load; - - /* For physical_balance case, we accumulated both load - * values in the one of the siblings cpu_irq[], - * to use the same code for physical and logical processors - * as much as possible. - * - * NOTE: the cpu_irq[] array holds the sum of the load for - * sibling A and sibling B in the slot for the lowest numbered - * sibling (A), _AND_ the load for sibling B in the slot for - * the higher numbered sibling. - * - * We seek the least loaded sibling by making the comparison - * (A+B)/2 vs B - */ - load = CPU_IRQ(min_loaded) >> 1; - for_each_cpu_mask(j, per_cpu(cpu_sibling_map, min_loaded)) { - if (load > CPU_IRQ(j)) { - /* This won't change cpu_sibling_map[min_loaded] */ - load = CPU_IRQ(j); - min_loaded = j; - } - } - - cpus_and(allowed_mask, - cpu_online_map, - balance_irq_affinity[selected_irq]); - target_cpu_mask = cpumask_of_cpu(min_loaded); - cpus_and(tmp, target_cpu_mask, allowed_mask); - - if (!cpus_empty(tmp)) { - /* mark for change destination */ - set_pending_irq(selected_irq, cpumask_of_cpu(min_loaded)); - - /* Since we made a change, come back sooner to - * check for more variation. - */ - balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL, - balanced_irq_interval - BALANCED_IRQ_LESS_DELTA); - return; - } - goto tryanotherirq; - -not_worth_the_effort: - /* - * if we did not find an IRQ to move, then adjust the time interval - * upward - */ - balanced_irq_interval = min((long)MAX_BALANCED_IRQ_INTERVAL, - balanced_irq_interval + BALANCED_IRQ_MORE_DELTA); - return; -} - -static int balanced_irq(void *unused) -{ - int i; - unsigned long prev_balance_time = jiffies; - long time_remaining = balanced_irq_interval; - - /* push everything to CPU 0 to give us a starting point. */ - for (i = 0 ; i < NR_IRQS ; i++) { - irq_desc[i].pending_mask = cpumask_of_cpu(0); - set_pending_irq(i, cpumask_of_cpu(0)); - } - - set_freezable(); - for ( ; ; ) { - time_remaining = schedule_timeout_interruptible(time_remaining); - try_to_freeze(); - if (time_after(jiffies, - prev_balance_time+balanced_irq_interval)) { - preempt_disable(); - do_irq_balance(); - prev_balance_time = jiffies; - time_remaining = balanced_irq_interval; - preempt_enable(); - } - } - return 0; -} - -static int __init balanced_irq_init(void) -{ - int i; - struct cpuinfo_x86 *c; - cpumask_t tmp; - - cpus_shift_right(tmp, cpu_online_map, 2); - c = &boot_cpu_data; - /* When not overwritten by the command line ask subarchitecture. */ - if (irqbalance_disabled == IRQBALANCE_CHECK_ARCH) - irqbalance_disabled = NO_BALANCE_IRQ; - if (irqbalance_disabled) - return 0; - - /* disable irqbalance completely if there is only one processor online */ - if (num_online_cpus() < 2) { - irqbalance_disabled = 1; - return 0; - } - /* - * Enable physical balance only if more than 1 physical processor - * is present - */ - if (smp_num_siblings > 1 && !cpus_empty(tmp)) - physical_balance = 1; - - for_each_online_cpu(i) { - irq_cpu_data[i].irq_delta = kzalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL); - irq_cpu_data[i].last_irq = kzalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL); - if (irq_cpu_data[i].irq_delta == NULL || irq_cpu_data[i].last_irq == NULL) { - printk(KERN_ERR "balanced_irq_init: out of memory"); - goto failed; - } - } - - printk(KERN_INFO "Starting balanced_irq\n"); - if (!IS_ERR(kthread_run(balanced_irq, NULL, "kirqd"))) - return 0; - printk(KERN_ERR "balanced_irq_init: failed to spawn balanced_irq"); -failed: - for_each_possible_cpu(i) { - kfree(irq_cpu_data[i].irq_delta); - irq_cpu_data[i].irq_delta = NULL; - kfree(irq_cpu_data[i].last_irq); - irq_cpu_data[i].last_irq = NULL; - } - return 0; -} - -int __devinit irqbalance_disable(char *str) -{ - irqbalance_disabled = 1; - return 1; -} - -__setup("noirqbalance", irqbalance_disable); - -late_initcall(balanced_irq_init); -#endif /* CONFIG_IRQBALANCE */ -#endif /* CONFIG_SMP */ - -#ifndef CONFIG_SMP -void send_IPI_self(int vector) -{ - unsigned int cfg; - - /* - * Wait for idle. - */ - apic_wait_icr_idle(); - cfg = APIC_DM_FIXED | APIC_DEST_SELF | vector | APIC_DEST_LOGICAL; - /* - * Send the IPI. The write to APIC_ICR fires this off. - */ - apic_write(APIC_ICR, cfg); -} -#endif /* !CONFIG_SMP */ - - -/* - * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to - * specific CPU-side IRQs. - */ - -#define MAX_PIRQS 8 -static int pirq_entries [MAX_PIRQS]; -static int pirqs_enabled; -int skip_ioapic_setup; - -static int __init ioapic_pirq_setup(char *str) -{ - int i, max; - int ints[MAX_PIRQS+1]; - - get_options(str, ARRAY_SIZE(ints), ints); - - for (i = 0; i < MAX_PIRQS; i++) - pirq_entries[i] = -1; - - pirqs_enabled = 1; - apic_printk(APIC_VERBOSE, KERN_INFO - "PIRQ redirection, working around broken MP-BIOS.\n"); - max = MAX_PIRQS; - if (ints[0] < MAX_PIRQS) - max = ints[0]; - - for (i = 0; i < max; i++) { - apic_printk(APIC_VERBOSE, KERN_DEBUG - "... PIRQ%d -> IRQ %d\n", i, ints[i+1]); - /* - * PIRQs are mapped upside down, usually. - */ - pirq_entries[MAX_PIRQS-i-1] = ints[i+1]; - } - return 1; -} - -__setup("pirq=", ioapic_pirq_setup); - -/* - * Find the IRQ entry number of a certain pin. - */ -static int find_irq_entry(int apic, int pin, int type) -{ - int i; - - for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mp_irqtype == type && - (mp_irqs[i].mp_dstapic == mp_ioapics[apic].mp_apicid || - mp_irqs[i].mp_dstapic == MP_APIC_ALL) && - mp_irqs[i].mp_dstirq == pin) - return i; - - return -1; -} - -/* - * Find the pin to which IRQ[irq] (ISA) is connected - */ -static int __init find_isa_irq_pin(int irq, int type) -{ - int i; - - for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mp_srcbus; - - if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mp_irqtype == type) && - (mp_irqs[i].mp_srcbusirq == irq)) - - return mp_irqs[i].mp_dstirq; - } - return -1; -} - -static int __init find_isa_irq_apic(int irq, int type) -{ - int i; - - for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mp_srcbus; - - if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mp_irqtype == type) && - (mp_irqs[i].mp_srcbusirq == irq)) - break; - } - if (i < mp_irq_entries) { - int apic; - for (apic = 0; apic < nr_ioapics; apic++) { - if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic) - return apic; - } - } - - return -1; -} - -/* - * Find a specific PCI IRQ entry. - * Not an __init, possibly needed by modules - */ -static int pin_2_irq(int idx, int apic, int pin); - -int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) -{ - int apic, i, best_guess = -1; - - apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, " - "slot:%d, pin:%d.\n", bus, slot, pin); - if (test_bit(bus, mp_bus_not_pci)) { - printk(KERN_WARNING "PCI BIOS passed nonexistent PCI bus %d!\n", bus); - return -1; - } - for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mp_srcbus; - - for (apic = 0; apic < nr_ioapics; apic++) - if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic || - mp_irqs[i].mp_dstapic == MP_APIC_ALL) - break; - - if (!test_bit(lbus, mp_bus_not_pci) && - !mp_irqs[i].mp_irqtype && - (bus == lbus) && - (slot == ((mp_irqs[i].mp_srcbusirq >> 2) & 0x1f))) { - int irq = pin_2_irq(i, apic, mp_irqs[i].mp_dstirq); - - if (!(apic || IO_APIC_IRQ(irq))) - continue; - - if (pin == (mp_irqs[i].mp_srcbusirq & 3)) - return irq; - /* - * Use the first all-but-pin matching entry as a - * best-guess fuzzy result for broken mptables. - */ - if (best_guess < 0) - best_guess = irq; - } - } - return best_guess; -} -EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector); - -/* - * This function currently is only a helper for the i386 smp boot process where - * we need to reprogram the ioredtbls to cater for the cpus which have come online - * so mask in all cases should simply be TARGET_CPUS - */ -#ifdef CONFIG_SMP -void __init setup_ioapic_dest(void) -{ - int pin, ioapic, irq, irq_entry; - - if (skip_ioapic_setup == 1) - return; - - for (ioapic = 0; ioapic < nr_ioapics; ioapic++) { - for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) { - irq_entry = find_irq_entry(ioapic, pin, mp_INT); - if (irq_entry == -1) - continue; - irq = pin_2_irq(irq_entry, ioapic, pin); - set_ioapic_affinity_irq(irq, TARGET_CPUS); - } - - } -} -#endif - -#if defined(CONFIG_EISA) || defined(CONFIG_MCA) -/* - * EISA Edge/Level control register, ELCR - */ -static int EISA_ELCR(unsigned int irq) -{ - if (irq < 16) { - unsigned int port = 0x4d0 + (irq >> 3); - return (inb(port) >> (irq & 7)) & 1; - } - apic_printk(APIC_VERBOSE, KERN_INFO - "Broken MPtable reports ISA irq %d\n", irq); - return 0; -} -#endif - -/* ISA interrupts are always polarity zero edge triggered, - * when listed as conforming in the MP table. */ - -#define default_ISA_trigger(idx) (0) -#define default_ISA_polarity(idx) (0) - -/* EISA interrupts are always polarity zero and can be edge or level - * trigger depending on the ELCR value. If an interrupt is listed as - * EISA conforming in the MP table, that means its trigger type must - * be read in from the ELCR */ - -#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mp_srcbusirq)) -#define default_EISA_polarity(idx) default_ISA_polarity(idx) - -/* PCI interrupts are always polarity one level triggered, - * when listed as conforming in the MP table. */ - -#define default_PCI_trigger(idx) (1) -#define default_PCI_polarity(idx) (1) - -/* MCA interrupts are always polarity zero level triggered, - * when listed as conforming in the MP table. */ - -#define default_MCA_trigger(idx) (1) -#define default_MCA_polarity(idx) default_ISA_polarity(idx) - -static int MPBIOS_polarity(int idx) -{ - int bus = mp_irqs[idx].mp_srcbus; - int polarity; - - /* - * Determine IRQ line polarity (high active or low active): - */ - switch (mp_irqs[idx].mp_irqflag & 3) { - case 0: /* conforms, ie. bus-type dependent polarity */ - { - polarity = test_bit(bus, mp_bus_not_pci)? - default_ISA_polarity(idx): - default_PCI_polarity(idx); - break; - } - case 1: /* high active */ - { - polarity = 0; - break; - } - case 2: /* reserved */ - { - printk(KERN_WARNING "broken BIOS!!\n"); - polarity = 1; - break; - } - case 3: /* low active */ - { - polarity = 1; - break; - } - default: /* invalid */ - { - printk(KERN_WARNING "broken BIOS!!\n"); - polarity = 1; - break; - } - } - return polarity; -} - -static int MPBIOS_trigger(int idx) -{ - int bus = mp_irqs[idx].mp_srcbus; - int trigger; - - /* - * Determine IRQ trigger mode (edge or level sensitive): - */ - switch ((mp_irqs[idx].mp_irqflag>>2) & 3) { - case 0: /* conforms, ie. bus-type dependent */ - { - trigger = test_bit(bus, mp_bus_not_pci)? - default_ISA_trigger(idx): - default_PCI_trigger(idx); -#if defined(CONFIG_EISA) || defined(CONFIG_MCA) - switch (mp_bus_id_to_type[bus]) { - case MP_BUS_ISA: /* ISA pin */ - { - /* set before the switch */ - break; - } - case MP_BUS_EISA: /* EISA pin */ - { - trigger = default_EISA_trigger(idx); - break; - } - case MP_BUS_PCI: /* PCI pin */ - { - /* set before the switch */ - break; - } - case MP_BUS_MCA: /* MCA pin */ - { - trigger = default_MCA_trigger(idx); - break; - } - default: - { - printk(KERN_WARNING "broken BIOS!!\n"); - trigger = 1; - break; - } - } -#endif - break; - } - case 1: /* edge */ - { - trigger = 0; - break; - } - case 2: /* reserved */ - { - printk(KERN_WARNING "broken BIOS!!\n"); - trigger = 1; - break; - } - case 3: /* level */ - { - trigger = 1; - break; - } - default: /* invalid */ - { - printk(KERN_WARNING "broken BIOS!!\n"); - trigger = 0; - break; - } - } - return trigger; -} - -static inline int irq_polarity(int idx) -{ - return MPBIOS_polarity(idx); -} - -static inline int irq_trigger(int idx) -{ - return MPBIOS_trigger(idx); -} - -static int pin_2_irq(int idx, int apic, int pin) -{ - int irq, i; - int bus = mp_irqs[idx].mp_srcbus; - - /* - * Debugging check, we are in big trouble if this message pops up! - */ - if (mp_irqs[idx].mp_dstirq != pin) - printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n"); - - if (test_bit(bus, mp_bus_not_pci)) - irq = mp_irqs[idx].mp_srcbusirq; - else { - /* - * PCI IRQs are mapped in order - */ - i = irq = 0; - while (i < apic) - irq += nr_ioapic_registers[i++]; - irq += pin; - - /* - * For MPS mode, so far only needed by ES7000 platform - */ - if (ioapic_renumber_irq) - irq = ioapic_renumber_irq(apic, irq); - } - - /* - * PCI IRQ command line redirection. Yes, limits are hardcoded. - */ - if ((pin >= 16) && (pin <= 23)) { - if (pirq_entries[pin-16] != -1) { - if (!pirq_entries[pin-16]) { - apic_printk(APIC_VERBOSE, KERN_DEBUG - "disabling PIRQ%d\n", pin-16); - } else { - irq = pirq_entries[pin-16]; - apic_printk(APIC_VERBOSE, KERN_DEBUG - "using PIRQ%d -> IRQ %d\n", - pin-16, irq); - } - } - } - return irq; -} - -static inline int IO_APIC_irq_trigger(int irq) -{ - int apic, idx, pin; - - for (apic = 0; apic < nr_ioapics; apic++) { - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - idx = find_irq_entry(apic, pin, mp_INT); - if ((idx != -1) && (irq == pin_2_irq(idx, apic, pin))) - return irq_trigger(idx); - } - } - /* - * nonexistent IRQs are edge default - */ - return 0; -} - -/* irq_vectors is indexed by the sum of all RTEs in all I/O APICs. */ -static u8 irq_vector[NR_IRQ_VECTORS] __read_mostly = { FIRST_DEVICE_VECTOR , 0 }; - -static int __assign_irq_vector(int irq) -{ - static int current_vector = FIRST_DEVICE_VECTOR, current_offset; - int vector, offset; - - BUG_ON((unsigned)irq >= NR_IRQ_VECTORS); - - if (irq_vector[irq] > 0) - return irq_vector[irq]; - - vector = current_vector; - offset = current_offset; -next: - vector += 8; - if (vector >= first_system_vector) { - offset = (offset + 1) % 8; - vector = FIRST_DEVICE_VECTOR + offset; - } - if (vector == current_vector) - return -ENOSPC; - if (test_and_set_bit(vector, used_vectors)) - goto next; - - current_vector = vector; - current_offset = offset; - irq_vector[irq] = vector; - - return vector; -} - -static int assign_irq_vector(int irq) -{ - unsigned long flags; - int vector; - - spin_lock_irqsave(&vector_lock, flags); - vector = __assign_irq_vector(irq); - spin_unlock_irqrestore(&vector_lock, flags); - - return vector; -} - -static struct irq_chip ioapic_chip; - -#define IOAPIC_AUTO -1 -#define IOAPIC_EDGE 0 -#define IOAPIC_LEVEL 1 - -static void ioapic_register_intr(int irq, int vector, unsigned long trigger) -{ - if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || - trigger == IOAPIC_LEVEL) { - irq_desc[irq].status |= IRQ_LEVEL; - set_irq_chip_and_handler_name(irq, &ioapic_chip, - handle_fasteoi_irq, "fasteoi"); - } else { - irq_desc[irq].status &= ~IRQ_LEVEL; - set_irq_chip_and_handler_name(irq, &ioapic_chip, - handle_edge_irq, "edge"); - } - set_intr_gate(vector, interrupt[irq]); -} - -static void __init setup_IO_APIC_irqs(void) -{ - struct IO_APIC_route_entry entry; - int apic, pin, idx, irq, first_notcon = 1, vector; - - apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); - - for (apic = 0; apic < nr_ioapics; apic++) { - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - - /* - * add it to the IO-APIC irq-routing table: - */ - memset(&entry, 0, sizeof(entry)); - - entry.delivery_mode = INT_DELIVERY_MODE; - entry.dest_mode = INT_DEST_MODE; - entry.mask = 0; /* enable IRQ */ - entry.dest.logical.logical_dest = - cpu_mask_to_apicid(TARGET_CPUS); - - idx = find_irq_entry(apic, pin, mp_INT); - if (idx == -1) { - if (first_notcon) { - apic_printk(APIC_VERBOSE, KERN_DEBUG - " IO-APIC (apicid-pin) %d-%d", - mp_ioapics[apic].mp_apicid, - pin); - first_notcon = 0; - } else - apic_printk(APIC_VERBOSE, ", %d-%d", - mp_ioapics[apic].mp_apicid, pin); - continue; - } - - if (!first_notcon) { - apic_printk(APIC_VERBOSE, " not connected.\n"); - first_notcon = 1; - } - - entry.trigger = irq_trigger(idx); - entry.polarity = irq_polarity(idx); - - if (irq_trigger(idx)) { - entry.trigger = 1; - entry.mask = 1; - } - - irq = pin_2_irq(idx, apic, pin); - /* - * skip adding the timer int on secondary nodes, which causes - * a small but painful rift in the time-space continuum - */ - if (multi_timer_check(apic, irq)) - continue; - else - add_pin_to_irq(irq, apic, pin); - - if (!apic && !IO_APIC_IRQ(irq)) - continue; - - if (IO_APIC_IRQ(irq)) { - vector = assign_irq_vector(irq); - entry.vector = vector; - ioapic_register_intr(irq, vector, IOAPIC_AUTO); - - if (!apic && (irq < 16)) - disable_8259A_irq(irq); - } - ioapic_write_entry(apic, pin, entry); - } - } - - if (!first_notcon) - apic_printk(APIC_VERBOSE, " not connected.\n"); -} - -/* - * Set up the timer pin, possibly with the 8259A-master behind. - */ -static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, - int vector) -{ - struct IO_APIC_route_entry entry; - - memset(&entry, 0, sizeof(entry)); - - /* - * We use logical delivery to get the timer IRQ - * to the first CPU. - */ - entry.dest_mode = INT_DEST_MODE; - entry.mask = 1; /* mask IRQ now */ - entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS); - entry.delivery_mode = INT_DELIVERY_MODE; - entry.polarity = 0; - entry.trigger = 0; - entry.vector = vector; - - /* - * The timer IRQ doesn't have to know that behind the - * scene we may have a 8259A-master in AEOI mode ... - */ - ioapic_register_intr(0, vector, IOAPIC_EDGE); - - /* - * Add it to the IO-APIC irq-routing table: - */ - ioapic_write_entry(apic, pin, entry); -} - - -__apicdebuginit(void) print_IO_APIC(void) -{ - int apic, i; - union IO_APIC_reg_00 reg_00; - union IO_APIC_reg_01 reg_01; - union IO_APIC_reg_02 reg_02; - union IO_APIC_reg_03 reg_03; - unsigned long flags; - - if (apic_verbosity == APIC_QUIET) - return; - - printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); - for (i = 0; i < nr_ioapics; i++) - printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n", - mp_ioapics[i].mp_apicid, nr_ioapic_registers[i]); - - /* - * We are a bit conservative about what we expect. We have to - * know about every hardware change ASAP. - */ - printk(KERN_INFO "testing the IO APIC.......................\n"); - - for (apic = 0; apic < nr_ioapics; apic++) { - - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(apic, 0); - reg_01.raw = io_apic_read(apic, 1); - if (reg_01.bits.version >= 0x10) - reg_02.raw = io_apic_read(apic, 2); - if (reg_01.bits.version >= 0x20) - reg_03.raw = io_apic_read(apic, 3); - spin_unlock_irqrestore(&ioapic_lock, flags); - - printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid); - printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); - printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); - printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type); - printk(KERN_DEBUG "....... : LTS : %X\n", reg_00.bits.LTS); - - printk(KERN_DEBUG ".... register #01: %08X\n", reg_01.raw); - printk(KERN_DEBUG "....... : max redirection entries: %04X\n", reg_01.bits.entries); - - printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ); - printk(KERN_DEBUG "....... : IO APIC version: %04X\n", reg_01.bits.version); - - /* - * Some Intel chipsets with IO APIC VERSION of 0x1? don't have reg_02, - * but the value of reg_02 is read as the previous read register - * value, so ignore it if reg_02 == reg_01. - */ - if (reg_01.bits.version >= 0x10 && reg_02.raw != reg_01.raw) { - printk(KERN_DEBUG ".... register #02: %08X\n", reg_02.raw); - printk(KERN_DEBUG "....... : arbitration: %02X\n", reg_02.bits.arbitration); - } - - /* - * Some Intel chipsets with IO APIC VERSION of 0x2? don't have reg_02 - * or reg_03, but the value of reg_0[23] is read as the previous read - * register value, so ignore it if reg_03 == reg_0[12]. - */ - if (reg_01.bits.version >= 0x20 && reg_03.raw != reg_02.raw && - reg_03.raw != reg_01.raw) { - printk(KERN_DEBUG ".... register #03: %08X\n", reg_03.raw); - printk(KERN_DEBUG "....... : Boot DT : %X\n", reg_03.bits.boot_DT); - } - - printk(KERN_DEBUG ".... IRQ redirection table:\n"); - - printk(KERN_DEBUG " NR Log Phy Mask Trig IRR Pol" - " Stat Dest Deli Vect: \n"); - - for (i = 0; i <= reg_01.bits.entries; i++) { - struct IO_APIC_route_entry entry; - - entry = ioapic_read_entry(apic, i); - - printk(KERN_DEBUG " %02x %03X %02X ", - i, - entry.dest.logical.logical_dest, - entry.dest.physical.physical_dest - ); - - printk("%1d %1d %1d %1d %1d %1d %1d %02X\n", - entry.mask, - entry.trigger, - entry.irr, - entry.polarity, - entry.delivery_status, - entry.dest_mode, - entry.delivery_mode, - entry.vector - ); - } - } - printk(KERN_DEBUG "IRQ to pin mappings:\n"); - for (i = 0; i < NR_IRQS; i++) { - struct irq_pin_list *entry = irq_2_pin + i; - if (entry->pin < 0) - continue; - printk(KERN_DEBUG "IRQ%d ", i); - for (;;) { - printk("-> %d:%d", entry->apic, entry->pin); - if (!entry->next) - break; - entry = irq_2_pin + entry->next; - } - printk("\n"); - } - - printk(KERN_INFO ".................................... done.\n"); - - return; -} - -__apicdebuginit(void) print_APIC_bitfield(int base) -{ - unsigned int v; - int i, j; - - if (apic_verbosity == APIC_QUIET) - return; - - printk(KERN_DEBUG "0123456789abcdef0123456789abcdef\n" KERN_DEBUG); - for (i = 0; i < 8; i++) { - v = apic_read(base + i*0x10); - for (j = 0; j < 32; j++) { - if (v & (1<<j)) - printk("1"); - else - printk("0"); - } - printk("\n"); - } -} - -__apicdebuginit(void) print_local_APIC(void *dummy) -{ - unsigned int v, ver, maxlvt; - u64 icr; - - if (apic_verbosity == APIC_QUIET) - return; - - printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n", - smp_processor_id(), hard_smp_processor_id()); - v = apic_read(APIC_ID); - printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, - GET_APIC_ID(v)); - v = apic_read(APIC_LVR); - printk(KERN_INFO "... APIC VERSION: %08x\n", v); - ver = GET_APIC_VERSION(v); - maxlvt = lapic_get_maxlvt(); - - v = apic_read(APIC_TASKPRI); - printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK); - - if (APIC_INTEGRATED(ver)) { /* !82489DX */ - v = apic_read(APIC_ARBPRI); - printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v, - v & APIC_ARBPRI_MASK); - v = apic_read(APIC_PROCPRI); - printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v); - } - - v = apic_read(APIC_EOI); - printk(KERN_DEBUG "... APIC EOI: %08x\n", v); - v = apic_read(APIC_RRR); - printk(KERN_DEBUG "... APIC RRR: %08x\n", v); - v = apic_read(APIC_LDR); - printk(KERN_DEBUG "... APIC LDR: %08x\n", v); - v = apic_read(APIC_DFR); - printk(KERN_DEBUG "... APIC DFR: %08x\n", v); - v = apic_read(APIC_SPIV); - printk(KERN_DEBUG "... APIC SPIV: %08x\n", v); - - printk(KERN_DEBUG "... APIC ISR field:\n"); - print_APIC_bitfield(APIC_ISR); - printk(KERN_DEBUG "... APIC TMR field:\n"); - print_APIC_bitfield(APIC_TMR); - printk(KERN_DEBUG "... APIC IRR field:\n"); - print_APIC_bitfield(APIC_IRR); - - if (APIC_INTEGRATED(ver)) { /* !82489DX */ - if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ - apic_write(APIC_ESR, 0); - v = apic_read(APIC_ESR); - printk(KERN_DEBUG "... APIC ESR: %08x\n", v); - } - - icr = apic_icr_read(); - printk(KERN_DEBUG "... APIC ICR: %08x\n", icr); - printk(KERN_DEBUG "... APIC ICR2: %08x\n", icr >> 32); - - v = apic_read(APIC_LVTT); - printk(KERN_DEBUG "... APIC LVTT: %08x\n", v); - - if (maxlvt > 3) { /* PC is LVT#4. */ - v = apic_read(APIC_LVTPC); - printk(KERN_DEBUG "... APIC LVTPC: %08x\n", v); - } - v = apic_read(APIC_LVT0); - printk(KERN_DEBUG "... APIC LVT0: %08x\n", v); - v = apic_read(APIC_LVT1); - printk(KERN_DEBUG "... APIC LVT1: %08x\n", v); - - if (maxlvt > 2) { /* ERR is LVT#3. */ - v = apic_read(APIC_LVTERR); - printk(KERN_DEBUG "... APIC LVTERR: %08x\n", v); - } - - v = apic_read(APIC_TMICT); - printk(KERN_DEBUG "... APIC TMICT: %08x\n", v); - v = apic_read(APIC_TMCCT); - printk(KERN_DEBUG "... APIC TMCCT: %08x\n", v); - v = apic_read(APIC_TDCR); - printk(KERN_DEBUG "... APIC TDCR: %08x\n", v); - printk("\n"); -} - -__apicdebuginit(void) print_all_local_APICs(void) -{ - on_each_cpu(print_local_APIC, NULL, 1); -} - -__apicdebuginit(void) print_PIC(void) -{ - unsigned int v; - unsigned long flags; - - if (apic_verbosity == APIC_QUIET) - return; - - printk(KERN_DEBUG "\nprinting PIC contents\n"); - - spin_lock_irqsave(&i8259A_lock, flags); - - v = inb(0xa1) << 8 | inb(0x21); - printk(KERN_DEBUG "... PIC IMR: %04x\n", v); - - v = inb(0xa0) << 8 | inb(0x20); - printk(KERN_DEBUG "... PIC IRR: %04x\n", v); - - outb(0x0b, 0xa0); - outb(0x0b, 0x20); - v = inb(0xa0) << 8 | inb(0x20); - outb(0x0a, 0xa0); - outb(0x0a, 0x20); - - spin_unlock_irqrestore(&i8259A_lock, flags); - - printk(KERN_DEBUG "... PIC ISR: %04x\n", v); - - v = inb(0x4d1) << 8 | inb(0x4d0); - printk(KERN_DEBUG "... PIC ELCR: %04x\n", v); -} - -__apicdebuginit(int) print_all_ICs(void) -{ - print_PIC(); - print_all_local_APICs(); - print_IO_APIC(); - - return 0; -} - -fs_initcall(print_all_ICs); - - -static void __init enable_IO_APIC(void) -{ - union IO_APIC_reg_01 reg_01; - int i8259_apic, i8259_pin; - int i, apic; - unsigned long flags; - - for (i = 0; i < PIN_MAP_SIZE; i++) { - irq_2_pin[i].pin = -1; - irq_2_pin[i].next = 0; - } - if (!pirqs_enabled) - for (i = 0; i < MAX_PIRQS; i++) - pirq_entries[i] = -1; - - /* - * The number of IO-APIC IRQ registers (== #pins): - */ - for (apic = 0; apic < nr_ioapics; apic++) { - spin_lock_irqsave(&ioapic_lock, flags); - reg_01.raw = io_apic_read(apic, 1); - spin_unlock_irqrestore(&ioapic_lock, flags); - nr_ioapic_registers[apic] = reg_01.bits.entries+1; - } - for (apic = 0; apic < nr_ioapics; apic++) { - int pin; - /* See if any of the pins is in ExtINT mode */ - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - struct IO_APIC_route_entry entry; - entry = ioapic_read_entry(apic, pin); - - - /* If the interrupt line is enabled and in ExtInt mode - * I have found the pin where the i8259 is connected. - */ - if ((entry.mask == 0) && (entry.delivery_mode == dest_ExtINT)) { - ioapic_i8259.apic = apic; - ioapic_i8259.pin = pin; - goto found_i8259; - } - } - } - found_i8259: - /* Look to see what if the MP table has reported the ExtINT */ - /* If we could not find the appropriate pin by looking at the ioapic - * the i8259 probably is not connected the ioapic but give the - * mptable a chance anyway. - */ - i8259_pin = find_isa_irq_pin(0, mp_ExtINT); - i8259_apic = find_isa_irq_apic(0, mp_ExtINT); - /* Trust the MP table if nothing is setup in the hardware */ - if ((ioapic_i8259.pin == -1) && (i8259_pin >= 0)) { - printk(KERN_WARNING "ExtINT not setup in hardware but reported by MP table\n"); - ioapic_i8259.pin = i8259_pin; - ioapic_i8259.apic = i8259_apic; - } - /* Complain if the MP table and the hardware disagree */ - if (((ioapic_i8259.apic != i8259_apic) || (ioapic_i8259.pin != i8259_pin)) && - (i8259_pin >= 0) && (ioapic_i8259.pin >= 0)) - { - printk(KERN_WARNING "ExtINT in hardware and MP table differ\n"); - } - - /* - * Do not trust the IO-APIC being empty at bootup - */ - clear_IO_APIC(); -} - -/* - * Not an __init, needed by the reboot code - */ -void disable_IO_APIC(void) -{ - /* - * Clear the IO-APIC before rebooting: - */ - clear_IO_APIC(); - - /* - * If the i8259 is routed through an IOAPIC - * Put that IOAPIC in virtual wire mode - * so legacy interrupts can be delivered. - */ - if (ioapic_i8259.pin != -1) { - struct IO_APIC_route_entry entry; - - memset(&entry, 0, sizeof(entry)); - entry.mask = 0; /* Enabled */ - entry.trigger = 0; /* Edge */ - entry.irr = 0; - entry.polarity = 0; /* High */ - entry.delivery_status = 0; - entry.dest_mode = 0; /* Physical */ - entry.delivery_mode = dest_ExtINT; /* ExtInt */ - entry.vector = 0; - entry.dest.physical.physical_dest = read_apic_id(); - - /* - * Add it to the IO-APIC irq-routing table: - */ - ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry); - } - disconnect_bsp_APIC(ioapic_i8259.pin != -1); -} - -/* - * function to set the IO-APIC physical IDs based on the - * values stored in the MPC table. - * - * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999 - */ - -static void __init setup_ioapic_ids_from_mpc(void) -{ - union IO_APIC_reg_00 reg_00; - physid_mask_t phys_id_present_map; - int apic; - int i; - unsigned char old_id; - unsigned long flags; - - if (x86_quirks->setup_ioapic_ids && x86_quirks->setup_ioapic_ids()) - return; - - /* - * Don't check I/O APIC IDs for xAPIC systems. They have - * no meaning without the serial APIC bus. - */ - if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) - || APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) - return; - /* - * This is broken; anything with a real cpu count has to - * circumvent this idiocy regardless. - */ - phys_id_present_map = ioapic_phys_id_map(phys_cpu_present_map); - - /* - * Set the IOAPIC ID to the value stored in the MPC table. - */ - for (apic = 0; apic < nr_ioapics; apic++) { - - /* Read the register 0 value */ - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(apic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); - - old_id = mp_ioapics[apic].mp_apicid; - - if (mp_ioapics[apic].mp_apicid >= get_physical_broadcast()) { - printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n", - apic, mp_ioapics[apic].mp_apicid); - printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", - reg_00.bits.ID); - mp_ioapics[apic].mp_apicid = reg_00.bits.ID; - } - - /* - * Sanity check, is the ID really free? Every APIC in a - * system must have a unique ID or we get lots of nice - * 'stuck on smp_invalidate_needed IPI wait' messages. - */ - if (check_apicid_used(phys_id_present_map, - mp_ioapics[apic].mp_apicid)) { - printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n", - apic, mp_ioapics[apic].mp_apicid); - for (i = 0; i < get_physical_broadcast(); i++) - if (!physid_isset(i, phys_id_present_map)) - break; - if (i >= get_physical_broadcast()) - panic("Max APIC ID exceeded!\n"); - printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", - i); - physid_set(i, phys_id_present_map); - mp_ioapics[apic].mp_apicid = i; - } else { - physid_mask_t tmp; - tmp = apicid_to_cpu_present(mp_ioapics[apic].mp_apicid); - apic_printk(APIC_VERBOSE, "Setting %d in the " - "phys_id_present_map\n", - mp_ioapics[apic].mp_apicid); - physids_or(phys_id_present_map, phys_id_present_map, tmp); - } - - - /* - * We need to adjust the IRQ routing table - * if the ID changed. - */ - if (old_id != mp_ioapics[apic].mp_apicid) - for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mp_dstapic == old_id) - mp_irqs[i].mp_dstapic - = mp_ioapics[apic].mp_apicid; - - /* - * Read the right value from the MPC table and - * write it into the ID register. - */ - apic_printk(APIC_VERBOSE, KERN_INFO - "...changing IO-APIC physical APIC ID to %d ...", - mp_ioapics[apic].mp_apicid); - - reg_00.bits.ID = mp_ioapics[apic].mp_apicid; - spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(apic, 0, reg_00.raw); - spin_unlock_irqrestore(&ioapic_lock, flags); - - /* - * Sanity check - */ - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(apic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); - if (reg_00.bits.ID != mp_ioapics[apic].mp_apicid) - printk("could not set ID!\n"); - else - apic_printk(APIC_VERBOSE, " ok.\n"); - } -} - -int no_timer_check __initdata; - -static int __init notimercheck(char *s) -{ - no_timer_check = 1; - return 1; -} -__setup("no_timer_check", notimercheck); - -/* - * There is a nasty bug in some older SMP boards, their mptable lies - * about the timer IRQ. We do the following to work around the situation: - * - * - timer IRQ defaults to IO-APIC IRQ - * - if this function detects that timer IRQs are defunct, then we fall - * back to ISA timer IRQs - */ -static int __init timer_irq_works(void) -{ - unsigned long t1 = jiffies; - unsigned long flags; - - if (no_timer_check) - return 1; - - local_save_flags(flags); - local_irq_enable(); - /* Let ten ticks pass... */ - mdelay((10 * 1000) / HZ); - local_irq_restore(flags); - - /* - * Expect a few ticks at least, to be sure some possible - * glue logic does not lock up after one or two first - * ticks in a non-ExtINT mode. Also the local APIC - * might have cached one ExtINT interrupt. Finally, at - * least one tick may be lost due to delays. - */ - if (time_after(jiffies, t1 + 4)) - return 1; - - return 0; -} - -/* - * In the SMP+IOAPIC case it might happen that there are an unspecified - * number of pending IRQ events unhandled. These cases are very rare, - * so we 'resend' these IRQs via IPIs, to the same CPU. It's much - * better to do it this way as thus we do not have to be aware of - * 'pending' interrupts in the IRQ path, except at this point. - */ -/* - * Edge triggered needs to resend any interrupt - * that was delayed but this is now handled in the device - * independent code. - */ - -/* - * Startup quirk: - * - * Starting up a edge-triggered IO-APIC interrupt is - * nasty - we need to make sure that we get the edge. - * If it is already asserted for some reason, we need - * return 1 to indicate that is was pending. - * - * This is not complete - we should be able to fake - * an edge even if it isn't on the 8259A... - * - * (We do this for level-triggered IRQs too - it cannot hurt.) - */ -static unsigned int startup_ioapic_irq(unsigned int irq) -{ - int was_pending = 0; - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - if (irq < 16) { - disable_8259A_irq(irq); - if (i8259A_irq_pending(irq)) - was_pending = 1; - } - __unmask_IO_APIC_irq(irq); - spin_unlock_irqrestore(&ioapic_lock, flags); - - return was_pending; -} - -static void ack_ioapic_irq(unsigned int irq) -{ - move_native_irq(irq); - ack_APIC_irq(); -} - -static void ack_ioapic_quirk_irq(unsigned int irq) -{ - unsigned long v; - int i; - - move_native_irq(irq); -/* - * It appears there is an erratum which affects at least version 0x11 - * of I/O APIC (that's the 82093AA and cores integrated into various - * chipsets). Under certain conditions a level-triggered interrupt is - * erroneously delivered as edge-triggered one but the respective IRR - * bit gets set nevertheless. As a result the I/O unit expects an EOI - * message but it will never arrive and further interrupts are blocked - * from the source. The exact reason is so far unknown, but the - * phenomenon was observed when two consecutive interrupt requests - * from a given source get delivered to the same CPU and the source is - * temporarily disabled in between. - * - * A workaround is to simulate an EOI message manually. We achieve it - * by setting the trigger mode to edge and then to level when the edge - * trigger mode gets detected in the TMR of a local APIC for a - * level-triggered interrupt. We mask the source for the time of the - * operation to prevent an edge-triggered interrupt escaping meanwhile. - * The idea is from Manfred Spraul. --macro - */ - i = irq_vector[irq]; - - v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1)); - - ack_APIC_irq(); - - if (!(v & (1 << (i & 0x1f)))) { - atomic_inc(&irq_mis_count); - spin_lock(&ioapic_lock); - __mask_and_edge_IO_APIC_irq(irq); - __unmask_and_level_IO_APIC_irq(irq); - spin_unlock(&ioapic_lock); - } -} - -static int ioapic_retrigger_irq(unsigned int irq) -{ - send_IPI_self(irq_vector[irq]); - - return 1; -} - -static struct irq_chip ioapic_chip __read_mostly = { - .name = "IO-APIC", - .startup = startup_ioapic_irq, - .mask = mask_IO_APIC_irq, - .unmask = unmask_IO_APIC_irq, - .ack = ack_ioapic_irq, - .eoi = ack_ioapic_quirk_irq, -#ifdef CONFIG_SMP - .set_affinity = set_ioapic_affinity_irq, -#endif - .retrigger = ioapic_retrigger_irq, -}; - - -static inline void init_IO_APIC_traps(void) -{ - int irq; - - /* - * NOTE! The local APIC isn't very good at handling - * multiple interrupts at the same interrupt level. - * As the interrupt level is determined by taking the - * vector number and shifting that right by 4, we - * want to spread these out a bit so that they don't - * all fall in the same interrupt level. - * - * Also, we've got to be careful not to trash gate - * 0x80, because int 0x80 is hm, kind of importantish. ;) - */ - for (irq = 0; irq < NR_IRQS ; irq++) { - if (IO_APIC_IRQ(irq) && !irq_vector[irq]) { - /* - * Hmm.. We don't have an entry for this, - * so default to an old-fashioned 8259 - * interrupt if we can.. - */ - if (irq < 16) - make_8259A_irq(irq); - else - /* Strange. Oh, well.. */ - irq_desc[irq].chip = &no_irq_chip; - } - } -} - -/* - * The local APIC irq-chip implementation: - */ - -static void ack_lapic_irq(unsigned int irq) -{ - ack_APIC_irq(); -} - -static void mask_lapic_irq(unsigned int irq) -{ - unsigned long v; - - v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v | APIC_LVT_MASKED); -} - -static void unmask_lapic_irq(unsigned int irq) -{ - unsigned long v; - - v = apic_read(APIC_LVT0); - apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); -} - -static struct irq_chip lapic_chip __read_mostly = { - .name = "local-APIC", - .mask = mask_lapic_irq, - .unmask = unmask_lapic_irq, - .ack = ack_lapic_irq, -}; - -static void lapic_register_intr(int irq, int vector) -{ - irq_desc[irq].status &= ~IRQ_LEVEL; - set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq, - "edge"); - set_intr_gate(vector, interrupt[irq]); -} - -static void __init setup_nmi(void) -{ - /* - * Dirty trick to enable the NMI watchdog ... - * We put the 8259A master into AEOI mode and - * unmask on all local APICs LVT0 as NMI. - * - * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire') - * is from Maciej W. Rozycki - so we do not have to EOI from - * the NMI handler or the timer interrupt. - */ - apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ..."); - - enable_NMI_through_LVT0(); - - apic_printk(APIC_VERBOSE, " done.\n"); -} - -/* - * This looks a bit hackish but it's about the only one way of sending - * a few INTA cycles to 8259As and any associated glue logic. ICR does - * not support the ExtINT mode, unfortunately. We need to send these - * cycles as some i82489DX-based boards have glue logic that keeps the - * 8259A interrupt line asserted until INTA. --macro - */ -static inline void __init unlock_ExtINT_logic(void) -{ - int apic, pin, i; - struct IO_APIC_route_entry entry0, entry1; - unsigned char save_control, save_freq_select; - - pin = find_isa_irq_pin(8, mp_INT); - if (pin == -1) { - WARN_ON_ONCE(1); - return; - } - apic = find_isa_irq_apic(8, mp_INT); - if (apic == -1) { - WARN_ON_ONCE(1); - return; - } - - entry0 = ioapic_read_entry(apic, pin); - clear_IO_APIC_pin(apic, pin); - - memset(&entry1, 0, sizeof(entry1)); - - entry1.dest_mode = 0; /* physical delivery */ - entry1.mask = 0; /* unmask IRQ now */ - entry1.dest.physical.physical_dest = hard_smp_processor_id(); - entry1.delivery_mode = dest_ExtINT; - entry1.polarity = entry0.polarity; - entry1.trigger = 0; - entry1.vector = 0; - - ioapic_write_entry(apic, pin, entry1); - - save_control = CMOS_READ(RTC_CONTROL); - save_freq_select = CMOS_READ(RTC_FREQ_SELECT); - CMOS_WRITE((save_freq_select & ~RTC_RATE_SELECT) | 0x6, - RTC_FREQ_SELECT); - CMOS_WRITE(save_control | RTC_PIE, RTC_CONTROL); - - i = 100; - while (i-- > 0) { - mdelay(10); - if ((CMOS_READ(RTC_INTR_FLAGS) & RTC_PF) == RTC_PF) - i -= 10; - } - - CMOS_WRITE(save_control, RTC_CONTROL); - CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT); - clear_IO_APIC_pin(apic, pin); - - ioapic_write_entry(apic, pin, entry0); -} - -/* - * This code may look a bit paranoid, but it's supposed to cooperate with - * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ - * is so screwy. Thanks to Brian Perkins for testing/hacking this beast - * fanatically on his truly buggy board. - */ -static inline void __init check_timer(void) -{ - int apic1, pin1, apic2, pin2; - int no_pin1 = 0; - int vector; - unsigned int ver; - unsigned long flags; - - local_irq_save(flags); - - ver = apic_read(APIC_LVR); - ver = GET_APIC_VERSION(ver); - - /* - * get/set the timer IRQ vector: - */ - disable_8259A_irq(0); - vector = assign_irq_vector(0); - set_intr_gate(vector, interrupt[0]); - - /* - * As IRQ0 is to be enabled in the 8259A, the virtual - * wire has to be disabled in the local APIC. Also - * timer interrupts need to be acknowledged manually in - * the 8259A for the i82489DX when using the NMI - * watchdog as that APIC treats NMIs as level-triggered. - * The AEOI mode will finish them in the 8259A - * automatically. - */ - apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); - init_8259A(1); - timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); - - pin1 = find_isa_irq_pin(0, mp_INT); - apic1 = find_isa_irq_apic(0, mp_INT); - pin2 = ioapic_i8259.pin; - apic2 = ioapic_i8259.apic; - - apic_printk(APIC_QUIET, KERN_INFO "..TIMER: vector=0x%02X " - "apic1=%d pin1=%d apic2=%d pin2=%d\n", - vector, apic1, pin1, apic2, pin2); - - /* - * Some BIOS writers are clueless and report the ExtINTA - * I/O APIC input from the cascaded 8259A as the timer - * interrupt input. So just in case, if only one pin - * was found above, try it both directly and through the - * 8259A. - */ - if (pin1 == -1) { - pin1 = pin2; - apic1 = apic2; - no_pin1 = 1; - } else if (pin2 == -1) { - pin2 = pin1; - apic2 = apic1; - } - - if (pin1 != -1) { - /* - * Ok, does IRQ0 through the IOAPIC work? - */ - if (no_pin1) { - add_pin_to_irq(0, apic1, pin1); - setup_timer_IRQ0_pin(apic1, pin1, vector); - } - unmask_IO_APIC_irq(0); - if (timer_irq_works()) { - if (nmi_watchdog == NMI_IO_APIC) { - setup_nmi(); - enable_8259A_irq(0); - } - if (disable_timer_pin_1 > 0) - clear_IO_APIC_pin(0, pin1); - goto out; - } - clear_IO_APIC_pin(apic1, pin1); - if (!no_pin1) - apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: " - "8254 timer not connected to IO-APIC\n"); - - apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer " - "(IRQ0) through the 8259A ...\n"); - apic_printk(APIC_QUIET, KERN_INFO - "..... (found apic %d pin %d) ...\n", apic2, pin2); - /* - * legacy devices should be connected to IO APIC #0 - */ - replace_pin_at_irq(0, apic1, pin1, apic2, pin2); - setup_timer_IRQ0_pin(apic2, pin2, vector); - unmask_IO_APIC_irq(0); - enable_8259A_irq(0); - if (timer_irq_works()) { - apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); - timer_through_8259 = 1; - if (nmi_watchdog == NMI_IO_APIC) { - disable_8259A_irq(0); - setup_nmi(); - enable_8259A_irq(0); - } - goto out; - } - /* - * Cleanup, just in case ... - */ - disable_8259A_irq(0); - clear_IO_APIC_pin(apic2, pin2); - apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); - } - - if (nmi_watchdog == NMI_IO_APIC) { - apic_printk(APIC_QUIET, KERN_WARNING "timer doesn't work " - "through the IO-APIC - disabling NMI Watchdog!\n"); - nmi_watchdog = NMI_NONE; - } - timer_ack = 0; - - apic_printk(APIC_QUIET, KERN_INFO - "...trying to set up timer as Virtual Wire IRQ...\n"); - - lapic_register_intr(0, vector); - apic_write(APIC_LVT0, APIC_DM_FIXED | vector); /* Fixed mode */ - enable_8259A_irq(0); - - if (timer_irq_works()) { - apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); - goto out; - } - disable_8259A_irq(0); - apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector); - apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n"); - - apic_printk(APIC_QUIET, KERN_INFO - "...trying to set up timer as ExtINT IRQ...\n"); - - init_8259A(0); - make_8259A_irq(0); - apic_write(APIC_LVT0, APIC_DM_EXTINT); - - unlock_ExtINT_logic(); - - if (timer_irq_works()) { - apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); - goto out; - } - apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n"); - panic("IO-APIC + timer doesn't work! Boot with apic=debug and send a " - "report. Then try booting with the 'noapic' option.\n"); -out: - local_irq_restore(flags); -} - -/* - * Traditionally ISA IRQ2 is the cascade IRQ, and is not available - * to devices. However there may be an I/O APIC pin available for - * this interrupt regardless. The pin may be left unconnected, but - * typically it will be reused as an ExtINT cascade interrupt for - * the master 8259A. In the MPS case such a pin will normally be - * reported as an ExtINT interrupt in the MP table. With ACPI - * there is no provision for ExtINT interrupts, and in the absence - * of an override it would be treated as an ordinary ISA I/O APIC - * interrupt, that is edge-triggered and unmasked by default. We - * used to do this, but it caused problems on some systems because - * of the NMI watchdog and sometimes IRQ0 of the 8254 timer using - * the same ExtINT cascade interrupt to drive the local APIC of the - * bootstrap processor. Therefore we refrain from routing IRQ2 to - * the I/O APIC in all cases now. No actual device should request - * it anyway. --macro - */ -#define PIC_IRQS (1 << PIC_CASCADE_IR) - -void __init setup_IO_APIC(void) -{ - int i; - - /* Reserve all the system vectors. */ - for (i = first_system_vector; i < NR_VECTORS; i++) - set_bit(i, used_vectors); - - enable_IO_APIC(); - - io_apic_irqs = ~PIC_IRQS; - - printk("ENABLING IO-APIC IRQs\n"); - - /* - * Set up IO-APIC IRQ routing. - */ - if (!acpi_ioapic) - setup_ioapic_ids_from_mpc(); - sync_Arb_IDs(); - setup_IO_APIC_irqs(); - init_IO_APIC_traps(); - check_timer(); -} - -/* - * Called after all the initialization is done. If we didnt find any - * APIC bugs then we can allow the modify fast path - */ - -static int __init io_apic_bug_finalize(void) -{ - if (sis_apic_bug == -1) - sis_apic_bug = 0; - return 0; -} - -late_initcall(io_apic_bug_finalize); - -struct sysfs_ioapic_data { - struct sys_device dev; - struct IO_APIC_route_entry entry[0]; -}; -static struct sysfs_ioapic_data *mp_ioapic_data[MAX_IO_APICS]; - -static int ioapic_suspend(struct sys_device *dev, pm_message_t state) -{ - struct IO_APIC_route_entry *entry; - struct sysfs_ioapic_data *data; - int i; - - data = container_of(dev, struct sysfs_ioapic_data, dev); - entry = data->entry; - for (i = 0; i < nr_ioapic_registers[dev->id]; i++) - entry[i] = ioapic_read_entry(dev->id, i); - - return 0; -} - -static int ioapic_resume(struct sys_device *dev) -{ - struct IO_APIC_route_entry *entry; - struct sysfs_ioapic_data *data; - unsigned long flags; - union IO_APIC_reg_00 reg_00; - int i; - - data = container_of(dev, struct sysfs_ioapic_data, dev); - entry = data->entry; - - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(dev->id, 0); - if (reg_00.bits.ID != mp_ioapics[dev->id].mp_apicid) { - reg_00.bits.ID = mp_ioapics[dev->id].mp_apicid; - io_apic_write(dev->id, 0, reg_00.raw); - } - spin_unlock_irqrestore(&ioapic_lock, flags); - for (i = 0; i < nr_ioapic_registers[dev->id]; i++) - ioapic_write_entry(dev->id, i, entry[i]); - - return 0; -} - -static struct sysdev_class ioapic_sysdev_class = { - .name = "ioapic", - .suspend = ioapic_suspend, - .resume = ioapic_resume, -}; - -static int __init ioapic_init_sysfs(void) -{ - struct sys_device *dev; - int i, size, error = 0; - - error = sysdev_class_register(&ioapic_sysdev_class); - if (error) - return error; - - for (i = 0; i < nr_ioapics; i++) { - size = sizeof(struct sys_device) + nr_ioapic_registers[i] - * sizeof(struct IO_APIC_route_entry); - mp_ioapic_data[i] = kzalloc(size, GFP_KERNEL); - if (!mp_ioapic_data[i]) { - printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i); - continue; - } - dev = &mp_ioapic_data[i]->dev; - dev->id = i; - dev->cls = &ioapic_sysdev_class; - error = sysdev_register(dev); - if (error) { - kfree(mp_ioapic_data[i]); - mp_ioapic_data[i] = NULL; - printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i); - continue; - } - } - - return 0; -} - -device_initcall(ioapic_init_sysfs); - -/* - * Dynamic irq allocate and deallocation - */ -int create_irq(void) -{ - /* Allocate an unused irq */ - int irq, new, vector = 0; - unsigned long flags; - - irq = -ENOSPC; - spin_lock_irqsave(&vector_lock, flags); - for (new = (NR_IRQS - 1); new >= 0; new--) { - if (platform_legacy_irq(new)) - continue; - if (irq_vector[new] != 0) - continue; - vector = __assign_irq_vector(new); - if (likely(vector > 0)) - irq = new; - break; - } - spin_unlock_irqrestore(&vector_lock, flags); - - if (irq >= 0) { - set_intr_gate(vector, interrupt[irq]); - dynamic_irq_init(irq); - } - return irq; -} - -void destroy_irq(unsigned int irq) -{ - unsigned long flags; - - dynamic_irq_cleanup(irq); - - spin_lock_irqsave(&vector_lock, flags); - clear_bit(irq_vector[irq], used_vectors); - irq_vector[irq] = 0; - spin_unlock_irqrestore(&vector_lock, flags); -} - -/* - * MSI message composition - */ -#ifdef CONFIG_PCI_MSI -static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg) -{ - int vector; - unsigned dest; - - vector = assign_irq_vector(irq); - if (vector >= 0) { - dest = cpu_mask_to_apicid(TARGET_CPUS); - - msg->address_hi = MSI_ADDR_BASE_HI; - msg->address_lo = - MSI_ADDR_BASE_LO | - ((INT_DEST_MODE == 0) ? -MSI_ADDR_DEST_MODE_PHYSICAL: - MSI_ADDR_DEST_MODE_LOGICAL) | - ((INT_DELIVERY_MODE != dest_LowestPrio) ? - MSI_ADDR_REDIRECTION_CPU: - MSI_ADDR_REDIRECTION_LOWPRI) | - MSI_ADDR_DEST_ID(dest); - - msg->data = - MSI_DATA_TRIGGER_EDGE | - MSI_DATA_LEVEL_ASSERT | - ((INT_DELIVERY_MODE != dest_LowestPrio) ? -MSI_DATA_DELIVERY_FIXED: - MSI_DATA_DELIVERY_LOWPRI) | - MSI_DATA_VECTOR(vector); - } - return vector; -} - -#ifdef CONFIG_SMP -static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) -{ - struct msi_msg msg; - unsigned int dest; - cpumask_t tmp; - int vector; - - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) - tmp = TARGET_CPUS; - - vector = assign_irq_vector(irq); - if (vector < 0) - return; - - dest = cpu_mask_to_apicid(mask); - - read_msi_msg(irq, &msg); - - msg.data &= ~MSI_DATA_VECTOR_MASK; - msg.data |= MSI_DATA_VECTOR(vector); - msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; - msg.address_lo |= MSI_ADDR_DEST_ID(dest); - - write_msi_msg(irq, &msg); - irq_desc[irq].affinity = mask; -} -#endif /* CONFIG_SMP */ - -/* - * IRQ Chip for MSI PCI/PCI-X/PCI-Express Devices, - * which implement the MSI or MSI-X Capability Structure. - */ -static struct irq_chip msi_chip = { - .name = "PCI-MSI", - .unmask = unmask_msi_irq, - .mask = mask_msi_irq, - .ack = ack_ioapic_irq, -#ifdef CONFIG_SMP - .set_affinity = set_msi_irq_affinity, -#endif - .retrigger = ioapic_retrigger_irq, -}; - -int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc) -{ - struct msi_msg msg; - int irq, ret; - irq = create_irq(); - if (irq < 0) - return irq; - - ret = msi_compose_msg(dev, irq, &msg); - if (ret < 0) { - destroy_irq(irq); - return ret; - } - - set_irq_msi(irq, desc); - write_msi_msg(irq, &msg); - - set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, - "edge"); - - return 0; -} - -void arch_teardown_msi_irq(unsigned int irq) -{ - destroy_irq(irq); -} - -#endif /* CONFIG_PCI_MSI */ - -/* - * Hypertransport interrupt support - */ -#ifdef CONFIG_HT_IRQ - -#ifdef CONFIG_SMP - -static void target_ht_irq(unsigned int irq, unsigned int dest) -{ - struct ht_irq_msg msg; - fetch_ht_irq_msg(irq, &msg); - - msg.address_lo &= ~(HT_IRQ_LOW_DEST_ID_MASK); - msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); - - msg.address_lo |= HT_IRQ_LOW_DEST_ID(dest); - msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); - - write_ht_irq_msg(irq, &msg); -} - -static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) -{ - unsigned int dest; - cpumask_t tmp; - - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) - tmp = TARGET_CPUS; - - cpus_and(mask, tmp, CPU_MASK_ALL); - - dest = cpu_mask_to_apicid(mask); - - target_ht_irq(irq, dest); - irq_desc[irq].affinity = mask; -} -#endif - -static struct irq_chip ht_irq_chip = { - .name = "PCI-HT", - .mask = mask_ht_irq, - .unmask = unmask_ht_irq, - .ack = ack_ioapic_irq, -#ifdef CONFIG_SMP - .set_affinity = set_ht_irq_affinity, -#endif - .retrigger = ioapic_retrigger_irq, -}; - -int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) -{ - int vector; - - vector = assign_irq_vector(irq); - if (vector >= 0) { - struct ht_irq_msg msg; - unsigned dest; - cpumask_t tmp; - - cpus_clear(tmp); - cpu_set(vector >> 8, tmp); - dest = cpu_mask_to_apicid(tmp); - - msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); - - msg.address_lo = - HT_IRQ_LOW_BASE | - HT_IRQ_LOW_DEST_ID(dest) | - HT_IRQ_LOW_VECTOR(vector) | - ((INT_DEST_MODE == 0) ? - HT_IRQ_LOW_DM_PHYSICAL : - HT_IRQ_LOW_DM_LOGICAL) | - HT_IRQ_LOW_RQEOI_EDGE | - ((INT_DELIVERY_MODE != dest_LowestPrio) ? - HT_IRQ_LOW_MT_FIXED : - HT_IRQ_LOW_MT_ARBITRATED) | - HT_IRQ_LOW_IRQ_MASKED; - - write_ht_irq_msg(irq, &msg); - - set_irq_chip_and_handler_name(irq, &ht_irq_chip, - handle_edge_irq, "edge"); - } - return vector; -} -#endif /* CONFIG_HT_IRQ */ - -/* -------------------------------------------------------------------------- - ACPI-based IOAPIC Configuration - -------------------------------------------------------------------------- */ - -#ifdef CONFIG_ACPI - -int __init io_apic_get_unique_id(int ioapic, int apic_id) -{ - union IO_APIC_reg_00 reg_00; - static physid_mask_t apic_id_map = PHYSID_MASK_NONE; - physid_mask_t tmp; - unsigned long flags; - int i = 0; - - /* - * The P4 platform supports up to 256 APIC IDs on two separate APIC - * buses (one for LAPICs, one for IOAPICs), where predecessors only - * supports up to 16 on one shared APIC bus. - * - * TBD: Expand LAPIC/IOAPIC support on P4-class systems to take full - * advantage of new APIC bus architecture. - */ - - if (physids_empty(apic_id_map)) - apic_id_map = ioapic_phys_id_map(phys_cpu_present_map); - - spin_lock_irqsave(&ioapic_lock, flags); - reg_00.raw = io_apic_read(ioapic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); - - if (apic_id >= get_physical_broadcast()) { - printk(KERN_WARNING "IOAPIC[%d]: Invalid apic_id %d, trying " - "%d\n", ioapic, apic_id, reg_00.bits.ID); - apic_id = reg_00.bits.ID; - } - - /* - * Every APIC in a system must have a unique ID or we get lots of nice - * 'stuck on smp_invalidate_needed IPI wait' messages. - */ - if (check_apicid_used(apic_id_map, apic_id)) { - - for (i = 0; i < get_physical_broadcast(); i++) { - if (!check_apicid_used(apic_id_map, i)) - break; - } - - if (i == get_physical_broadcast()) - panic("Max apic_id exceeded!\n"); - - printk(KERN_WARNING "IOAPIC[%d]: apic_id %d already used, " - "trying %d\n", ioapic, apic_id, i); - - apic_id = i; - } - - tmp = apicid_to_cpu_present(apic_id); - physids_or(apic_id_map, apic_id_map, tmp); - - if (reg_00.bits.ID != apic_id) { - reg_00.bits.ID = apic_id; - - spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(ioapic, 0, reg_00.raw); - reg_00.raw = io_apic_read(ioapic, 0); - spin_unlock_irqrestore(&ioapic_lock, flags); - - /* Sanity check */ - if (reg_00.bits.ID != apic_id) { - printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic); - return -1; - } - } - - apic_printk(APIC_VERBOSE, KERN_INFO - "IOAPIC[%d]: Assigned apic_id %d\n", ioapic, apic_id); - - return apic_id; -} - - -int __init io_apic_get_version(int ioapic) -{ - union IO_APIC_reg_01 reg_01; - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - reg_01.raw = io_apic_read(ioapic, 1); - spin_unlock_irqrestore(&ioapic_lock, flags); - - return reg_01.bits.version; -} - - -int __init io_apic_get_redir_entries(int ioapic) -{ - union IO_APIC_reg_01 reg_01; - unsigned long flags; - - spin_lock_irqsave(&ioapic_lock, flags); - reg_01.raw = io_apic_read(ioapic, 1); - spin_unlock_irqrestore(&ioapic_lock, flags); - - return reg_01.bits.entries; -} - - -int io_apic_set_pci_routing(int ioapic, int pin, int irq, int edge_level, int active_high_low) -{ - struct IO_APIC_route_entry entry; - - if (!IO_APIC_IRQ(irq)) { - printk(KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n", - ioapic); - return -EINVAL; - } - - /* - * Generate a PCI IRQ routing entry and program the IOAPIC accordingly. - * Note that we mask (disable) IRQs now -- these get enabled when the - * corresponding device driver registers for this IRQ. - */ - - memset(&entry, 0, sizeof(entry)); - - entry.delivery_mode = INT_DELIVERY_MODE; - entry.dest_mode = INT_DEST_MODE; - entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS); - entry.trigger = edge_level; - entry.polarity = active_high_low; - entry.mask = 1; - - /* - * IRQs < 16 are already in the irq_2_pin[] map - */ - if (irq >= 16) - add_pin_to_irq(irq, ioapic, pin); - - entry.vector = assign_irq_vector(irq); - - apic_printk(APIC_DEBUG, KERN_DEBUG "IOAPIC[%d]: Set PCI routing entry " - "(%d-%d -> 0x%x -> IRQ %d Mode:%i Active:%i)\n", ioapic, - mp_ioapics[ioapic].mp_apicid, pin, entry.vector, irq, - edge_level, active_high_low); - - ioapic_register_intr(irq, entry.vector, edge_level); - - if (!ioapic && (irq < 16)) - disable_8259A_irq(irq); - - ioapic_write_entry(ioapic, pin, entry); - - return 0; -} - -int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) -{ - int i; - - if (skip_ioapic_setup) - return -1; - - for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mp_irqtype == mp_INT && - mp_irqs[i].mp_srcbusirq == bus_irq) - break; - if (i >= mp_irq_entries) - return -1; - - *trigger = irq_trigger(i); - *polarity = irq_polarity(i); - return 0; -} - -#endif /* CONFIG_ACPI */ - -static int __init parse_disable_timer_pin_1(char *arg) -{ - disable_timer_pin_1 = 1; - return 0; -} -early_param("disable_timer_pin_1", parse_disable_timer_pin_1); - -static int __init parse_enable_timer_pin_1(char *arg) -{ - disable_timer_pin_1 = -1; - return 0; -} -early_param("enable_timer_pin_1", parse_enable_timer_pin_1); - -static int __init parse_noapic(char *arg) -{ - /* disable IO-APIC */ - disable_ioapic_setup(); - return 0; -} -early_param("noapic", parse_noapic); - -void __init ioapic_init_mappings(void) -{ - unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0; - int i; - - for (i = 0; i < nr_ioapics; i++) { - if (smp_found_config) { - ioapic_phys = mp_ioapics[i].mp_apicaddr; - if (!ioapic_phys) { - printk(KERN_ERR - "WARNING: bogus zero IO-APIC " - "address found in MPTABLE, " - "disabling IO/APIC support!\n"); - smp_found_config = 0; - skip_ioapic_setup = 1; - goto fake_ioapic_page; - } - } else { -fake_ioapic_page: - ioapic_phys = (unsigned long) - alloc_bootmem_pages(PAGE_SIZE); - ioapic_phys = __pa(ioapic_phys); - } - set_fixmap_nocache(idx, ioapic_phys); - printk(KERN_DEBUG "mapped IOAPIC to %08lx (%08lx)\n", - __fix_to_virt(idx), ioapic_phys); - idx++; - } -} - diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c new file mode 100644 index 00000000000..ccf6c503fc3 --- /dev/null +++ b/arch/x86/kernel/irq.c @@ -0,0 +1,189 @@ +/* + * Common interrupt code for 32 and 64 bit + */ +#include <linux/cpu.h> +#include <linux/interrupt.h> +#include <linux/kernel_stat.h> +#include <linux/seq_file.h> + +#include <asm/apic.h> +#include <asm/io_apic.h> +#include <asm/smp.h> + +atomic_t irq_err_count; + +/* + * 'what should we do if we get a hw irq event on an illegal vector'. + * each architecture has to answer this themselves. + */ +void ack_bad_irq(unsigned int irq) +{ + printk(KERN_ERR "unexpected IRQ trap at vector %02x\n", irq); + +#ifdef CONFIG_X86_LOCAL_APIC + /* + * Currently unexpected vectors happen only on SMP and APIC. + * We _must_ ack these because every local APIC has only N + * irq slots per priority level, and a 'hanging, unacked' IRQ + * holds up an irq slot - in excessive cases (when multiple + * unexpected vectors occur) that might lock up the APIC + * completely. + * But only ack when the APIC is enabled -AK + */ + if (cpu_has_apic) + ack_APIC_irq(); +#endif +} + +#ifdef CONFIG_X86_32 +# define irq_stats(x) (&per_cpu(irq_stat,x)) +#else +# define irq_stats(x) cpu_pda(x) +#endif +/* + * /proc/interrupts printing: + */ +static int show_other_interrupts(struct seq_file *p) +{ + int j; + + seq_printf(p, "NMI: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->__nmi_count); + seq_printf(p, " Non-maskable interrupts\n"); +#ifdef CONFIG_X86_LOCAL_APIC + seq_printf(p, "LOC: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->apic_timer_irqs); + seq_printf(p, " Local timer interrupts\n"); +#endif +#ifdef CONFIG_SMP + seq_printf(p, "RES: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_resched_count); + seq_printf(p, " Rescheduling interrupts\n"); + seq_printf(p, "CAL: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_call_count); + seq_printf(p, " Function call interrupts\n"); + seq_printf(p, "TLB: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_tlb_count); + seq_printf(p, " TLB shootdowns\n"); +#endif +#ifdef CONFIG_X86_MCE + seq_printf(p, "TRM: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_thermal_count); + seq_printf(p, " Thermal event interrupts\n"); +# ifdef CONFIG_X86_64 + seq_printf(p, "THR: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_threshold_count); + seq_printf(p, " Threshold APIC interrupts\n"); +# endif +#endif +#ifdef CONFIG_X86_LOCAL_APIC + seq_printf(p, "SPU: "); + for_each_online_cpu(j) + seq_printf(p, "%10u ", irq_stats(j)->irq_spurious_count); + seq_printf(p, " Spurious interrupts\n"); +#endif + seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); +#if defined(CONFIG_X86_IO_APIC) + seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count)); +#endif + return 0; +} + +int show_interrupts(struct seq_file *p, void *v) +{ + unsigned long flags, any_count = 0; + int i = *(loff_t *) v, j; + struct irqaction *action; + struct irq_desc *desc; + + if (i > nr_irqs) + return 0; + + if (i == nr_irqs) + return show_other_interrupts(p); + + /* print header */ + if (i == 0) { + seq_printf(p, " "); + for_each_online_cpu(j) + seq_printf(p, "CPU%-8d",j); + seq_putc(p, '\n'); + } + + desc = irq_to_desc(i); + spin_lock_irqsave(&desc->lock, flags); +#ifndef CONFIG_SMP + any_count = kstat_irqs(i); +#else + for_each_online_cpu(j) + any_count |= kstat_irqs_cpu(i, j); +#endif + action = desc->action; + if (!action && !any_count) + goto out; + + seq_printf(p, "%3d: ", i); +#ifndef CONFIG_SMP + seq_printf(p, "%10u ", kstat_irqs(i)); +#else + for_each_online_cpu(j) + seq_printf(p, "%10u ", kstat_irqs_cpu(i, j)); +#endif + seq_printf(p, " %8s", desc->chip->name); + seq_printf(p, "-%-8s", desc->name); + + if (action) { + seq_printf(p, " %s", action->name); + while ((action = action->next) != NULL) + seq_printf(p, ", %s", action->name); + } + + seq_putc(p, '\n'); +out: + spin_unlock_irqrestore(&desc->lock, flags); + return 0; +} + +/* + * /proc/stat helpers + */ +u64 arch_irq_stat_cpu(unsigned int cpu) +{ + u64 sum = irq_stats(cpu)->__nmi_count; + +#ifdef CONFIG_X86_LOCAL_APIC + sum += irq_stats(cpu)->apic_timer_irqs; +#endif +#ifdef CONFIG_SMP + sum += irq_stats(cpu)->irq_resched_count; + sum += irq_stats(cpu)->irq_call_count; + sum += irq_stats(cpu)->irq_tlb_count; +#endif +#ifdef CONFIG_X86_MCE + sum += irq_stats(cpu)->irq_thermal_count; +# ifdef CONFIG_X86_64 + sum += irq_stats(cpu)->irq_threshold_count; +#endif +#endif +#ifdef CONFIG_X86_LOCAL_APIC + sum += irq_stats(cpu)->irq_spurious_count; +#endif + return sum; +} + +u64 arch_irq_stat(void) +{ + u64 sum = atomic_read(&irq_err_count); + +#ifdef CONFIG_X86_IO_APIC + sum += atomic_read(&irq_mis_count); +#endif + return sum; +} diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index b71e02d42f4..a51382672de 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -25,29 +25,6 @@ EXPORT_PER_CPU_SYMBOL(irq_stat); DEFINE_PER_CPU(struct pt_regs *, irq_regs); EXPORT_PER_CPU_SYMBOL(irq_regs); -/* - * 'what should we do if we get a hw irq event on an illegal vector'. - * each architecture has to answer this themselves. - */ -void ack_bad_irq(unsigned int irq) -{ - printk(KERN_ERR "unexpected IRQ trap at vector %02x\n", irq); - -#ifdef CONFIG_X86_LOCAL_APIC - /* - * Currently unexpected vectors happen only on SMP and APIC. - * We _must_ ack these because every local APIC has only N - * irq slots per priority level, and a 'hanging, unacked' IRQ - * holds up an irq slot - in excessive cases (when multiple - * unexpected vectors occur) that might lock up the APIC - * completely. - * But only ack when the APIC is enabled -AK - */ - if (cpu_has_apic) - ack_APIC_irq(); -#endif -} - #ifdef CONFIG_DEBUG_STACKOVERFLOW /* Debugging check for stack overflow: is there less than 1KB free? */ static int check_stack_overflow(void) @@ -223,20 +200,25 @@ unsigned int do_IRQ(struct pt_regs *regs) { struct pt_regs *old_regs; /* high bit used in ret_from_ code */ - int overflow, irq = ~regs->orig_ax; - struct irq_desc *desc = irq_desc + irq; + int overflow; + unsigned vector = ~regs->orig_ax; + struct irq_desc *desc; + unsigned irq; - if (unlikely((unsigned)irq >= NR_IRQS)) { - printk(KERN_EMERG "%s: cannot handle IRQ %d\n", - __func__, irq); - BUG(); - } old_regs = set_irq_regs(regs); irq_enter(); + irq = __get_cpu_var(vector_irq)[vector]; overflow = check_stack_overflow(); + desc = irq_to_desc(irq); + if (unlikely(!desc)) { + printk(KERN_EMERG "%s: cannot handle IRQ %d vector %#x cpu %d\n", + __func__, irq, vector, smp_processor_id()); + BUG(); + } + if (!execute_on_irq_stack(overflow, desc, irq)) { if (unlikely(overflow)) print_stack_overflow(); @@ -248,146 +230,6 @@ unsigned int do_IRQ(struct pt_regs *regs) return 1; } -/* - * Interrupt statistics: - */ - -atomic_t irq_err_count; - -/* - * /proc/interrupts printing: - */ - -int show_interrupts(struct seq_file *p, void *v) -{ - int i = *(loff_t *) v, j; - struct irqaction * action; - unsigned long flags; - - if (i == 0) { - seq_printf(p, " "); - for_each_online_cpu(j) - seq_printf(p, "CPU%-8d",j); - seq_putc(p, '\n'); - } - - if (i < NR_IRQS) { - unsigned any_count = 0; - - spin_lock_irqsave(&irq_desc[i].lock, flags); -#ifndef CONFIG_SMP - any_count = kstat_irqs(i); -#else - for_each_online_cpu(j) - any_count |= kstat_cpu(j).irqs[i]; -#endif - action = irq_desc[i].action; - if (!action && !any_count) - goto skip; - seq_printf(p, "%3d: ",i); -#ifndef CONFIG_SMP - seq_printf(p, "%10u ", kstat_irqs(i)); -#else - for_each_online_cpu(j) - seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); -#endif - seq_printf(p, " %8s", irq_desc[i].chip->name); - seq_printf(p, "-%-8s", irq_desc[i].name); - - if (action) { - seq_printf(p, " %s", action->name); - while ((action = action->next) != NULL) - seq_printf(p, ", %s", action->name); - } - - seq_putc(p, '\n'); -skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); - } else if (i == NR_IRQS) { - seq_printf(p, "NMI: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", nmi_count(j)); - seq_printf(p, " Non-maskable interrupts\n"); -#ifdef CONFIG_X86_LOCAL_APIC - seq_printf(p, "LOC: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).apic_timer_irqs); - seq_printf(p, " Local timer interrupts\n"); -#endif -#ifdef CONFIG_SMP - seq_printf(p, "RES: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_resched_count); - seq_printf(p, " Rescheduling interrupts\n"); - seq_printf(p, "CAL: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_call_count); - seq_printf(p, " Function call interrupts\n"); - seq_printf(p, "TLB: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_tlb_count); - seq_printf(p, " TLB shootdowns\n"); -#endif -#ifdef CONFIG_X86_MCE - seq_printf(p, "TRM: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_thermal_count); - seq_printf(p, " Thermal event interrupts\n"); -#endif -#ifdef CONFIG_X86_LOCAL_APIC - seq_printf(p, "SPU: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", - per_cpu(irq_stat,j).irq_spurious_count); - seq_printf(p, " Spurious interrupts\n"); -#endif - seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); -#if defined(CONFIG_X86_IO_APIC) - seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count)); -#endif - } - return 0; -} - -/* - * /proc/stat helpers - */ -u64 arch_irq_stat_cpu(unsigned int cpu) -{ - u64 sum = nmi_count(cpu); - -#ifdef CONFIG_X86_LOCAL_APIC - sum += per_cpu(irq_stat, cpu).apic_timer_irqs; -#endif -#ifdef CONFIG_SMP - sum += per_cpu(irq_stat, cpu).irq_resched_count; - sum += per_cpu(irq_stat, cpu).irq_call_count; - sum += per_cpu(irq_stat, cpu).irq_tlb_count; -#endif -#ifdef CONFIG_X86_MCE - sum += per_cpu(irq_stat, cpu).irq_thermal_count; -#endif -#ifdef CONFIG_X86_LOCAL_APIC - sum += per_cpu(irq_stat, cpu).irq_spurious_count; -#endif - return sum; -} - -u64 arch_irq_stat(void) -{ - u64 sum = atomic_read(&irq_err_count); - -#ifdef CONFIG_X86_IO_APIC - sum += atomic_read(&irq_mis_count); -#endif - return sum; -} - #ifdef CONFIG_HOTPLUG_CPU #include <mach_apic.h> @@ -395,20 +237,22 @@ void fixup_irqs(cpumask_t map) { unsigned int irq; static int warned; + struct irq_desc *desc; - for (irq = 0; irq < NR_IRQS; irq++) { + for_each_irq_desc(irq, desc) { cpumask_t mask; + if (irq == 2) continue; - cpus_and(mask, irq_desc[irq].affinity, map); + cpus_and(mask, desc->affinity, map); if (any_online_cpu(mask) == NR_CPUS) { printk("Breaking affinity for irq %i\n", irq); mask = map; } - if (irq_desc[irq].chip->set_affinity) - irq_desc[irq].chip->set_affinity(irq, mask); - else if (irq_desc[irq].action && !(warned++)) + if (desc->chip->set_affinity) + desc->chip->set_affinity(irq, mask); + else if (desc->action && !(warned++)) printk("Cannot set affinity for irq %i\n", irq); } diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c index f065fe9071b..60eb84eb77a 100644 --- a/arch/x86/kernel/irq_64.c +++ b/arch/x86/kernel/irq_64.c @@ -18,28 +18,6 @@ #include <asm/idle.h> #include <asm/smp.h> -atomic_t irq_err_count; - -/* - * 'what should we do if we get a hw irq event on an illegal vector'. - * each architecture has to answer this themselves. - */ -void ack_bad_irq(unsigned int irq) -{ - printk(KERN_WARNING "unexpected IRQ trap at vector %02x\n", irq); - /* - * Currently unexpected vectors happen only on SMP and APIC. - * We _must_ ack these because every local APIC has only N - * irq slots per priority level, and a 'hanging, unacked' IRQ - * holds up an irq slot - in excessive cases (when multiple - * unexpected vectors occur) that might lock up the APIC - * completely. - * But don't ack when the APIC is disabled. -AK - */ - if (!disable_apic) - ack_APIC_irq(); -} - #ifdef CONFIG_DEBUG_STACKOVERFLOW /* * Probabilistic stack overflow check: @@ -65,122 +43,6 @@ static inline void stack_overflow_check(struct pt_regs *regs) #endif /* - * Generic, controller-independent functions: - */ - -int show_interrupts(struct seq_file *p, void *v) -{ - int i = *(loff_t *) v, j; - struct irqaction * action; - unsigned long flags; - - if (i == 0) { - seq_printf(p, " "); - for_each_online_cpu(j) - seq_printf(p, "CPU%-8d",j); - seq_putc(p, '\n'); - } - - if (i < NR_IRQS) { - unsigned any_count = 0; - - spin_lock_irqsave(&irq_desc[i].lock, flags); -#ifndef CONFIG_SMP - any_count = kstat_irqs(i); -#else - for_each_online_cpu(j) - any_count |= kstat_cpu(j).irqs[i]; -#endif - action = irq_desc[i].action; - if (!action && !any_count) - goto skip; - seq_printf(p, "%3d: ",i); -#ifndef CONFIG_SMP - seq_printf(p, "%10u ", kstat_irqs(i)); -#else - for_each_online_cpu(j) - seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); -#endif - seq_printf(p, " %8s", irq_desc[i].chip->name); - seq_printf(p, "-%-8s", irq_desc[i].name); - - if (action) { - seq_printf(p, " %s", action->name); - while ((action = action->next) != NULL) - seq_printf(p, ", %s", action->name); - } - seq_putc(p, '\n'); -skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); - } else if (i == NR_IRQS) { - seq_printf(p, "NMI: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count); - seq_printf(p, " Non-maskable interrupts\n"); - seq_printf(p, "LOC: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs); - seq_printf(p, " Local timer interrupts\n"); -#ifdef CONFIG_SMP - seq_printf(p, "RES: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_resched_count); - seq_printf(p, " Rescheduling interrupts\n"); - seq_printf(p, "CAL: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_call_count); - seq_printf(p, " Function call interrupts\n"); - seq_printf(p, "TLB: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_tlb_count); - seq_printf(p, " TLB shootdowns\n"); -#endif -#ifdef CONFIG_X86_MCE - seq_printf(p, "TRM: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_thermal_count); - seq_printf(p, " Thermal event interrupts\n"); - seq_printf(p, "THR: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_threshold_count); - seq_printf(p, " Threshold APIC interrupts\n"); -#endif - seq_printf(p, "SPU: "); - for_each_online_cpu(j) - seq_printf(p, "%10u ", cpu_pda(j)->irq_spurious_count); - seq_printf(p, " Spurious interrupts\n"); - seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); - } - return 0; -} - -/* - * /proc/stat helpers - */ -u64 arch_irq_stat_cpu(unsigned int cpu) -{ - u64 sum = cpu_pda(cpu)->__nmi_count; - - sum += cpu_pda(cpu)->apic_timer_irqs; -#ifdef CONFIG_SMP - sum += cpu_pda(cpu)->irq_resched_count; - sum += cpu_pda(cpu)->irq_call_count; - sum += cpu_pda(cpu)->irq_tlb_count; -#endif -#ifdef CONFIG_X86_MCE - sum += cpu_pda(cpu)->irq_thermal_count; - sum += cpu_pda(cpu)->irq_threshold_count; -#endif - sum += cpu_pda(cpu)->irq_spurious_count; - return sum; -} - -u64 arch_irq_stat(void) -{ - return atomic_read(&irq_err_count); -} - -/* * do_IRQ handles all normal device IRQ's (the special * SMP cross-CPU interrupts have their own specific * handlers). @@ -188,6 +50,7 @@ u64 arch_irq_stat(void) asmlinkage unsigned int do_IRQ(struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); + struct irq_desc *desc; /* high bit used in ret_from_ code */ unsigned vector = ~regs->orig_ax; @@ -201,8 +64,9 @@ asmlinkage unsigned int do_IRQ(struct pt_regs *regs) stack_overflow_check(regs); #endif - if (likely(irq < NR_IRQS)) - generic_handle_irq(irq); + desc = irq_to_desc(irq); + if (likely(desc)) + generic_handle_irq_desc(irq, desc); else { if (!disable_apic) ack_APIC_irq(); @@ -223,8 +87,9 @@ void fixup_irqs(cpumask_t map) { unsigned int irq; static int warned; + struct irq_desc *desc; - for (irq = 0; irq < NR_IRQS; irq++) { + for_each_irq_desc(irq, desc) { cpumask_t mask; int break_affinity = 0; int set_affinity = 1; @@ -233,32 +98,32 @@ void fixup_irqs(cpumask_t map) continue; /* interrupt's are disabled at this point */ - spin_lock(&irq_desc[irq].lock); + spin_lock(&desc->lock); if (!irq_has_action(irq) || - cpus_equal(irq_desc[irq].affinity, map)) { - spin_unlock(&irq_desc[irq].lock); + cpus_equal(desc->affinity, map)) { + spin_unlock(&desc->lock); continue; } - cpus_and(mask, irq_desc[irq].affinity, map); + cpus_and(mask, desc->affinity, map); if (cpus_empty(mask)) { break_affinity = 1; mask = map; } - if (irq_desc[irq].chip->mask) - irq_desc[irq].chip->mask(irq); + if (desc->chip->mask) + desc->chip->mask(irq); - if (irq_desc[irq].chip->set_affinity) - irq_desc[irq].chip->set_affinity(irq, mask); + if (desc->chip->set_affinity) + desc->chip->set_affinity(irq, mask); else if (!(warned++)) set_affinity = 0; - if (irq_desc[irq].chip->unmask) - irq_desc[irq].chip->unmask(irq); + if (desc->chip->unmask) + desc->chip->unmask(irq); - spin_unlock(&irq_desc[irq].lock); + spin_unlock(&desc->lock); if (break_affinity && set_affinity) printk("Broke affinity for irq %i\n", irq); diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c index 9200a1e2752..845aa9803e8 100644 --- a/arch/x86/kernel/irqinit_32.c +++ b/arch/x86/kernel/irqinit_32.c @@ -69,6 +69,13 @@ void __init init_ISA_irqs (void) * 16 old-style INTA-cycle interrupts: */ for (i = 0; i < 16; i++) { + /* first time call this irq_desc */ + struct irq_desc *desc = irq_to_desc(i); + + desc->status = IRQ_DISABLED; + desc->action = NULL; + desc->depth = 1; + set_irq_chip_and_handler_name(i, &i8259A_chip, handle_level_irq, "XT"); } @@ -83,6 +90,27 @@ static struct irqaction irq2 = { .name = "cascade", }; +DEFINE_PER_CPU(vector_irq_t, vector_irq) = { + [0 ... IRQ0_VECTOR - 1] = -1, + [IRQ0_VECTOR] = 0, + [IRQ1_VECTOR] = 1, + [IRQ2_VECTOR] = 2, + [IRQ3_VECTOR] = 3, + [IRQ4_VECTOR] = 4, + [IRQ5_VECTOR] = 5, + [IRQ6_VECTOR] = 6, + [IRQ7_VECTOR] = 7, + [IRQ8_VECTOR] = 8, + [IRQ9_VECTOR] = 9, + [IRQ10_VECTOR] = 10, + [IRQ11_VECTOR] = 11, + [IRQ12_VECTOR] = 12, + [IRQ13_VECTOR] = 13, + [IRQ14_VECTOR] = 14, + [IRQ15_VECTOR] = 15, + [IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1 +}; + /* Overridden in paravirt.c */ void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ"))); @@ -98,22 +126,14 @@ void __init native_init_IRQ(void) * us. (some of these will be overridden and become * 'special' SMP interrupts) */ - for (i = 0; i < (NR_VECTORS - FIRST_EXTERNAL_VECTOR); i++) { - int vector = FIRST_EXTERNAL_VECTOR + i; - if (i >= NR_IRQS) - break; + for (i = FIRST_EXTERNAL_VECTOR; i < NR_VECTORS; i++) { /* SYSCALL_VECTOR was reserved in trap_init. */ - if (!test_bit(vector, used_vectors)) - set_intr_gate(vector, interrupt[i]); + if (i != SYSCALL_VECTOR) + set_intr_gate(i, interrupt[i]); } -#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_SMP) - /* - * IRQ0 must be given a fixed assignment and initialized, - * because it's used before the IO-APIC is set up. - */ - set_intr_gate(FIRST_DEVICE_VECTOR, interrupt[0]); +#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_SMP) /* * The reschedule interrupt is a CPU-to-CPU reschedule-helper * IPI, driven by wakeup. @@ -128,6 +148,9 @@ void __init native_init_IRQ(void) /* IPI for single call function */ set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR, call_function_single_interrupt); + + /* Low priority IPI to cleanup after moving an irq */ + set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt); #endif #ifdef CONFIG_X86_LOCAL_APIC diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c index 5b5be9d43c2..ff023539128 100644 --- a/arch/x86/kernel/irqinit_64.c +++ b/arch/x86/kernel/irqinit_64.c @@ -142,23 +142,19 @@ void __init init_ISA_irqs(void) init_bsp_APIC(); init_8259A(0); - for (i = 0; i < NR_IRQS; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; - - if (i < 16) { - /* - * 16 old-style INTA-cycle interrupts: - */ - set_irq_chip_and_handler_name(i, &i8259A_chip, + for (i = 0; i < 16; i++) { + /* first time call this irq_desc */ + struct irq_desc *desc = irq_to_desc(i); + + desc->status = IRQ_DISABLED; + desc->action = NULL; + desc->depth = 1; + + /* + * 16 old-style INTA-cycle interrupts: + */ + set_irq_chip_and_handler_name(i, &i8259A_chip, handle_level_irq, "XT"); - } else { - /* - * 'high' PCI IRQs filled in on demand - */ - irq_desc[i].chip = &no_irq_chip; - } } } diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index d02def06ca9..774ac499156 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -78,6 +78,34 @@ static cycle_t kvm_clock_read(void) return ret; } +/* + * If we don't do that, there is the possibility that the guest + * will calibrate under heavy load - thus, getting a lower lpj - + * and execute the delays themselves without load. This is wrong, + * because no delay loop can finish beforehand. + * Any heuristics is subject to fail, because ultimately, a large + * poll of guests can be running and trouble each other. So we preset + * lpj here + */ +static unsigned long kvm_get_tsc_khz(void) +{ + return preset_lpj; +} + +static void kvm_get_preset_lpj(void) +{ + struct pvclock_vcpu_time_info *src; + unsigned long khz; + u64 lpj; + + src = &per_cpu(hv_clock, 0); + khz = pvclock_tsc_khz(src); + + lpj = ((u64)khz * 1000); + do_div(lpj, HZ); + preset_lpj = lpj; +} + static struct clocksource kvm_clock = { .name = "kvm-clock", .read = kvm_clock_read, @@ -153,6 +181,7 @@ void __init kvmclock_init(void) pv_time_ops.get_wallclock = kvm_get_wallclock; pv_time_ops.set_wallclock = kvm_set_wallclock; pv_time_ops.sched_clock = kvm_clock_read; + pv_time_ops.get_tsc_khz = kvm_get_tsc_khz; #ifdef CONFIG_X86_LOCAL_APIC pv_apic_ops.setup_secondary_clock = kvm_setup_secondary_clock; #endif @@ -163,6 +192,7 @@ void __init kvmclock_init(void) #ifdef CONFIG_KEXEC machine_ops.crash_shutdown = kvm_crash_shutdown; #endif + kvm_get_preset_lpj(); clocksource_register(&kvm_clock); } } diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c index 2e2af5d1819..82a7c7ed6d4 100644 --- a/arch/x86/kernel/msr.c +++ b/arch/x86/kernel/msr.c @@ -163,8 +163,8 @@ static int __cpuinit msr_device_create(int cpu) { struct device *dev; - dev = device_create_drvdata(msr_class, NULL, MKDEV(MSR_MAJOR, cpu), - NULL, "msr%d", cpu); + dev = device_create(msr_class, NULL, MKDEV(MSR_MAJOR, cpu), NULL, + "msr%d", cpu); return IS_ERR(dev) ? PTR_ERR(dev) : 0; } diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c index 080d1d27f37..e1e731d78f3 100644 --- a/arch/x86/kernel/pci-calgary_64.c +++ b/arch/x86/kernel/pci-calgary_64.c @@ -217,16 +217,6 @@ static inline unsigned long verify_bit_range(unsigned long* bitmap, #endif /* CONFIG_IOMMU_DEBUG */ -static inline unsigned int num_dma_pages(unsigned long dma, unsigned int dmalen) -{ - unsigned int npages; - - npages = PAGE_ALIGN(dma + dmalen) - (dma & PAGE_MASK); - npages >>= PAGE_SHIFT; - - return npages; -} - static inline int translation_enabled(struct iommu_table *tbl) { /* only PHBs with translation enabled have an IOMMU table */ @@ -408,7 +398,7 @@ static void calgary_unmap_sg(struct device *dev, if (dmalen == 0) break; - npages = num_dma_pages(dma, dmalen); + npages = iommu_num_pages(dma, dmalen, PAGE_SIZE); iommu_free(tbl, dma, npages); } } @@ -427,7 +417,7 @@ static int calgary_map_sg(struct device *dev, struct scatterlist *sg, BUG_ON(!sg_page(s)); vaddr = (unsigned long) sg_virt(s); - npages = num_dma_pages(vaddr, s->length); + npages = iommu_num_pages(vaddr, s->length, PAGE_SIZE); entry = iommu_range_alloc(dev, tbl, npages); if (entry == bad_dma_address) { @@ -464,7 +454,7 @@ static dma_addr_t calgary_map_single(struct device *dev, phys_addr_t paddr, struct iommu_table *tbl = find_iommu_table(dev); uaddr = (unsigned long)vaddr; - npages = num_dma_pages(uaddr, size); + npages = iommu_num_pages(uaddr, size, PAGE_SIZE); return iommu_alloc(dev, tbl, vaddr, npages, direction); } @@ -475,7 +465,7 @@ static void calgary_unmap_single(struct device *dev, dma_addr_t dma_handle, struct iommu_table *tbl = find_iommu_table(dev); unsigned int npages; - npages = num_dma_pages(dma_handle, size); + npages = iommu_num_pages(dma_handle, size, PAGE_SIZE); iommu_free(tbl, dma_handle, npages); } diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 0a3824e837b..19262482021 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -125,13 +125,13 @@ void __init pci_iommu_alloc(void) pci_swiotlb_init(); } -unsigned long iommu_num_pages(unsigned long addr, unsigned long len) +unsigned long iommu_nr_pages(unsigned long addr, unsigned long len) { unsigned long size = roundup((addr & ~PAGE_MASK) + len, PAGE_SIZE); return size >> PAGE_SHIFT; } -EXPORT_SYMBOL(iommu_num_pages); +EXPORT_SYMBOL(iommu_nr_pages); #endif void *dma_generic_alloc_coherent(struct device *dev, size_t size, diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 145f1c83369..e3f75bbcede 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -231,7 +231,7 @@ nonforced_iommu(struct device *dev, unsigned long addr, size_t size) static dma_addr_t dma_map_area(struct device *dev, dma_addr_t phys_mem, size_t size, int dir, unsigned long align_mask) { - unsigned long npages = iommu_num_pages(phys_mem, size); + unsigned long npages = iommu_num_pages(phys_mem, size, PAGE_SIZE); unsigned long iommu_page = alloc_iommu(dev, npages, align_mask); int i; @@ -285,7 +285,7 @@ static void gart_unmap_single(struct device *dev, dma_addr_t dma_addr, return; iommu_page = (dma_addr - iommu_bus_base)>>PAGE_SHIFT; - npages = iommu_num_pages(dma_addr, size); + npages = iommu_num_pages(dma_addr, size, PAGE_SIZE); for (i = 0; i < npages; i++) { iommu_gatt_base[iommu_page + i] = gart_unmapped_entry; CLEAR_LEAK(iommu_page + i); @@ -368,7 +368,7 @@ static int __dma_map_cont(struct device *dev, struct scatterlist *start, } addr = phys_addr; - pages = iommu_num_pages(s->offset, s->length); + pages = iommu_num_pages(s->offset, s->length, PAGE_SIZE); while (pages--) { iommu_gatt_base[iommu_page] = GPTE_ENCODE(addr); SET_LEAK(iommu_page); @@ -451,7 +451,7 @@ gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, int dir) seg_size += s->length; need = nextneed; - pages += iommu_num_pages(s->offset, s->length); + pages += iommu_num_pages(s->offset, s->length, PAGE_SIZE); ps = s; } if (dma_map_cont(dev, start_sg, i - start, sgmap, pages, need) < 0) diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index cd8c0ed02b7..c958120fb1b 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -63,6 +63,13 @@ void idle_notifier_register(struct notifier_block *n) { atomic_notifier_chain_register(&idle_notifier, n); } +EXPORT_SYMBOL_GPL(idle_notifier_register); + +void idle_notifier_unregister(struct notifier_block *n) +{ + atomic_notifier_chain_unregister(&idle_notifier, n); +} +EXPORT_SYMBOL_GPL(idle_notifier_unregister); void enter_idle(void) { diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index 05fbe9a0325..4f9c55f3a7c 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -97,6 +97,18 @@ static unsigned pvclock_get_time_values(struct pvclock_shadow_time *dst, return dst->version; } +unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src) +{ + u64 pv_tsc_khz = 1000000ULL << 32; + + do_div(pv_tsc_khz, src->tsc_to_system_mul); + if (src->tsc_shift < 0) + pv_tsc_khz <<= -src->tsc_shift; + else + pv_tsc_khz >>= src->tsc_shift; + return pv_tsc_khz; +} + cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src) { struct pvclock_shadow_time shadow; diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c index f6a11b9b1f9..67465ed8931 100644 --- a/arch/x86/kernel/quirks.c +++ b/arch/x86/kernel/quirks.c @@ -35,9 +35,6 @@ static void __devinit quirk_intel_irqbalance(struct pci_dev *dev) if (!(word & (1 << 13))) { dev_info(&dev->dev, "Intel E7520/7320/7525 detected; " "disabling irq balancing and affinity\n"); -#ifdef CONFIG_IRQBALANCE - irqbalance_disable(""); -#endif noirqdebug_setup(""); #ifdef CONFIG_PROC_FS no_irq_affinity = 1; diff --git a/arch/x86/kernel/rtc.c b/arch/x86/kernel/rtc.c index 0a23b5795b2..dd6f2b71561 100644 --- a/arch/x86/kernel/rtc.c +++ b/arch/x86/kernel/rtc.c @@ -52,7 +52,7 @@ int mach_set_rtc_mmss(unsigned long nowtime) cmos_minutes = CMOS_READ(RTC_MINUTES); if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD) - BCD_TO_BIN(cmos_minutes); + cmos_minutes = bcd2bin(cmos_minutes); /* * since we're only adjusting minutes and seconds, @@ -69,8 +69,8 @@ int mach_set_rtc_mmss(unsigned long nowtime) if (abs(real_minutes - cmos_minutes) < 30) { if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD) { - BIN_TO_BCD(real_seconds); - BIN_TO_BCD(real_minutes); + real_seconds = bin2bcd(real_seconds); + real_minutes = bin2bcd(real_minutes); } CMOS_WRITE(real_seconds,RTC_SECONDS); CMOS_WRITE(real_minutes,RTC_MINUTES); @@ -124,16 +124,16 @@ unsigned long mach_get_cmos_time(void) WARN_ON_ONCE(RTC_ALWAYS_BCD && (status & RTC_DM_BINARY)); if (RTC_ALWAYS_BCD || !(status & RTC_DM_BINARY)) { - BCD_TO_BIN(sec); - BCD_TO_BIN(min); - BCD_TO_BIN(hour); - BCD_TO_BIN(day); - BCD_TO_BIN(mon); - BCD_TO_BIN(year); + sec = bcd2bin(sec); + min = bcd2bin(min); + hour = bcd2bin(hour); + day = bcd2bin(day); + mon = bcd2bin(mon); + year = bcd2bin(year); } if (century) { - BCD_TO_BIN(century); + century = bcd2bin(century); year += century * 100; printk(KERN_INFO "Extended CMOS year: %d\n", century * 100); } else diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 2255782e8d4..0fa6790c1dd 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -561,7 +561,13 @@ static void __init reserve_standard_io_resources(void) } -#ifdef CONFIG_PROC_VMCORE +/* + * Note: elfcorehdr_addr is not just limited to vmcore. It is also used by + * is_kdump_kernel() to determine if we are booting after a panic. Hence + * ifdef it under CONFIG_CRASH_DUMP and not CONFIG_PROC_VMCORE. + */ + +#ifdef CONFIG_CRASH_DUMP /* elfcorehdr= specifies the location of elf core header * stored by the crashed kernel. This option will be passed * by kexec loader to the capture kernel. @@ -1067,6 +1073,7 @@ void __init setup_arch(char **cmdline_p) #endif prefill_possible_map(); + #ifdef CONFIG_X86_64 init_cpu_to_node(); #endif @@ -1074,6 +1081,9 @@ void __init setup_arch(char **cmdline_p) init_apic_mappings(); ioapic_init_mappings(); + /* need to wait for io_apic is mapped */ + nr_irqs = probe_nr_irqs(); + kvm_guest_init(); e820_reserve_resources(); diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 0e67f72d931..410c88f0bfe 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -140,25 +140,30 @@ static void __init setup_cpu_pda_map(void) */ void __init setup_per_cpu_areas(void) { - ssize_t size = PERCPU_ENOUGH_ROOM; + ssize_t size, old_size; char *ptr; int cpu; + unsigned long align = 1; /* Setup cpu_pda map */ setup_cpu_pda_map(); /* Copy section for each CPU (we discard the original) */ - size = PERCPU_ENOUGH_ROOM; + old_size = PERCPU_ENOUGH_ROOM; + align = max_t(unsigned long, PAGE_SIZE, align); + size = roundup(old_size, align); printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n", size); for_each_possible_cpu(cpu) { #ifndef CONFIG_NEED_MULTIPLE_NODES - ptr = alloc_bootmem_pages(size); + ptr = __alloc_bootmem(size, align, + __pa(MAX_DMA_ADDRESS)); #else int node = early_cpu_to_node(cpu); if (!node_online(node) || !NODE_DATA(node)) { - ptr = alloc_bootmem_pages(size); + ptr = __alloc_bootmem(size, align, + __pa(MAX_DMA_ADDRESS)); printk(KERN_INFO "cpu %d has no node %d or node-local memory\n", cpu, node); @@ -167,7 +172,8 @@ void __init setup_per_cpu_areas(void) cpu, __pa(ptr)); } else { - ptr = alloc_bootmem_pages_node(NODE_DATA(node), size); + ptr = __alloc_bootmem_node(NODE_DATA(node), size, align, + __pa(MAX_DMA_ADDRESS)); if (ptr) printk(KERN_DEBUG "per cpu data for cpu%d on node%d at %016lx\n", cpu, node, __pa(ptr)); @@ -175,7 +181,6 @@ void __init setup_per_cpu_areas(void) #endif per_cpu_offset(cpu) = ptr - __per_cpu_start; memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start); - } printk(KERN_DEBUG "NR_CPUS: %d, nr_cpu_ids: %d, nr_node_ids %d\n", diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 8c3aca7cb34..7ece815ea63 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -282,6 +282,8 @@ static void __cpuinit smp_callin(void) cpu_set(cpuid, cpu_callin_map); } +static int __cpuinitdata unsafe_smp; + /* * Activate a secondary processor. */ @@ -397,7 +399,7 @@ static void __cpuinit smp_apply_quirks(struct cpuinfo_x86 *c) goto valid_k7; /* If we get here, not a certified SMP capable AMD system. */ - add_taint(TAINT_UNSAFE_SMP); + unsafe_smp = 1; } valid_k7: @@ -414,12 +416,10 @@ static void __cpuinit smp_checks(void) * Don't taint if we are running SMP kernel on a single non-MP * approved Athlon */ - if (tainted & TAINT_UNSAFE_SMP) { - if (num_online_cpus()) - printk(KERN_INFO "WARNING: This combination of AMD" - "processors is not suitable for SMP.\n"); - else - tainted &= ~TAINT_UNSAFE_SMP; + if (unsafe_smp && num_online_cpus() > 1) { + printk(KERN_INFO "WARNING: This combination of AMD" + "processors is not suitable for SMP.\n"); + add_taint(TAINT_UNSAFE_SMP); } } @@ -543,10 +543,10 @@ static inline void __inquire_remote_apic(int apicid) int timeout; u32 status; - printk(KERN_INFO "Inquiring remote APIC #%d...\n", apicid); + printk(KERN_INFO "Inquiring remote APIC 0x%x...\n", apicid); for (i = 0; i < ARRAY_SIZE(regs); i++) { - printk(KERN_INFO "... APIC #%d %s: ", apicid, names[i]); + printk(KERN_INFO "... APIC 0x%x %s: ", apicid, names[i]); /* * Wait for idle. @@ -874,7 +874,7 @@ do_rest: start_ip = setup_trampoline(); /* So we see what's up */ - printk(KERN_INFO "Booting processor %d/%d ip %lx\n", + printk(KERN_INFO "Booting processor %d APIC 0x%x ip 0x%lx\n", cpu, apicid, start_ip); /* diff --git a/arch/x86/kernel/uv_irq.c b/arch/x86/kernel/uv_irq.c new file mode 100644 index 00000000000..aeef529917e --- /dev/null +++ b/arch/x86/kernel/uv_irq.c @@ -0,0 +1,79 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * SGI UV IRQ functions + * + * Copyright (C) 2008 Silicon Graphics, Inc. All rights reserved. + */ + +#include <linux/module.h> +#include <linux/irq.h> + +#include <asm/apic.h> +#include <asm/uv/uv_irq.h> + +static void uv_noop(unsigned int irq) +{ +} + +static unsigned int uv_noop_ret(unsigned int irq) +{ + return 0; +} + +static void uv_ack_apic(unsigned int irq) +{ + ack_APIC_irq(); +} + +struct irq_chip uv_irq_chip = { + .name = "UV-CORE", + .startup = uv_noop_ret, + .shutdown = uv_noop, + .enable = uv_noop, + .disable = uv_noop, + .ack = uv_noop, + .mask = uv_noop, + .unmask = uv_noop, + .eoi = uv_ack_apic, + .end = uv_noop, +}; + +/* + * Set up a mapping of an available irq and vector, and enable the specified + * MMR that defines the MSI that is to be sent to the specified CPU when an + * interrupt is raised. + */ +int uv_setup_irq(char *irq_name, int cpu, int mmr_blade, + unsigned long mmr_offset) +{ + int irq; + int ret; + + irq = create_irq(); + if (irq <= 0) + return -EBUSY; + + ret = arch_enable_uv_irq(irq_name, irq, cpu, mmr_blade, mmr_offset); + if (ret != irq) + destroy_irq(irq); + + return ret; +} +EXPORT_SYMBOL_GPL(uv_setup_irq); + +/* + * Tear down a mapping of an irq and vector, and disable the specified MMR that + * defined the MSI that was to be sent to the specified CPU when an interrupt + * was raised. + * + * Set mmr_blade and mmr_offset to what was passed in on uv_setup_irq(). + */ +void uv_teardown_irq(unsigned int irq, int mmr_blade, unsigned long mmr_offset) +{ + arch_disable_uv_irq(mmr_blade, mmr_offset); + destroy_irq(irq); +} +EXPORT_SYMBOL_GPL(uv_teardown_irq); diff --git a/arch/x86/kernel/uv_sysfs.c b/arch/x86/kernel/uv_sysfs.c new file mode 100644 index 00000000000..67f9b9dbf80 --- /dev/null +++ b/arch/x86/kernel/uv_sysfs.c @@ -0,0 +1,72 @@ +/* + * This file supports the /sys/firmware/sgi_uv interfaces for SGI UV. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) Russ Anderson + */ + +#include <linux/sysdev.h> +#include <asm/uv/bios.h> + +struct kobject *sgi_uv_kobj; + +static ssize_t partition_id_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%ld\n", sn_partition_id); +} + +static ssize_t coherence_id_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%ld\n", partition_coherence_id()); +} + +static struct kobj_attribute partition_id_attr = + __ATTR(partition_id, S_IRUGO, partition_id_show, NULL); + +static struct kobj_attribute coherence_id_attr = + __ATTR(coherence_id, S_IRUGO, coherence_id_show, NULL); + + +static int __init sgi_uv_sysfs_init(void) +{ + unsigned long ret; + + if (!sgi_uv_kobj) + sgi_uv_kobj = kobject_create_and_add("sgi_uv", firmware_kobj); + if (!sgi_uv_kobj) { + printk(KERN_WARNING "kobject_create_and_add sgi_uv failed \n"); + return -EINVAL; + } + + ret = sysfs_create_file(sgi_uv_kobj, &partition_id_attr.attr); + if (ret) { + printk(KERN_WARNING "sysfs_create_file partition_id failed \n"); + return ret; + } + + ret = sysfs_create_file(sgi_uv_kobj, &coherence_id_attr.attr); + if (ret) { + printk(KERN_WARNING "sysfs_create_file coherence_id failed \n"); + return ret; + } + + return 0; +} + +device_initcall(sgi_uv_sysfs_init); diff --git a/arch/x86/kernel/visws_quirks.c b/arch/x86/kernel/visws_quirks.c index 61a97e616f7..0c9667f0752 100644 --- a/arch/x86/kernel/visws_quirks.c +++ b/arch/x86/kernel/visws_quirks.c @@ -484,10 +484,11 @@ static void disable_cobalt_irq(unsigned int irq) static unsigned int startup_cobalt_irq(unsigned int irq) { unsigned long flags; + struct irq_desc *desc = irq_to_desc(irq); spin_lock_irqsave(&cobalt_lock, flags); - if ((irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING))) - irq_desc[irq].status &= ~(IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING); + if ((desc->status & (IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING))) + desc->status &= ~(IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING); enable_cobalt_irq(irq); spin_unlock_irqrestore(&cobalt_lock, flags); return 0; @@ -506,9 +507,10 @@ static void ack_cobalt_irq(unsigned int irq) static void end_cobalt_irq(unsigned int irq) { unsigned long flags; + struct irq_desc *desc = irq_to_desc(irq); spin_lock_irqsave(&cobalt_lock, flags); - if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) + if (!(desc->status & (IRQ_DISABLED | IRQ_INPROGRESS))) enable_cobalt_irq(irq); spin_unlock_irqrestore(&cobalt_lock, flags); } @@ -626,12 +628,12 @@ static irqreturn_t piix4_master_intr(int irq, void *dev_id) spin_unlock_irqrestore(&i8259A_lock, flags); - desc = irq_desc + realirq; + desc = irq_to_desc(realirq); /* * handle this 'virtual interrupt' as a Cobalt one now. */ - kstat_cpu(smp_processor_id()).irqs[realirq]++; + kstat_incr_irqs_this_cpu(realirq, desc); if (likely(desc->action != NULL)) handle_IRQ_event(realirq, desc->action); @@ -662,27 +664,29 @@ void init_VISWS_APIC_irqs(void) int i; for (i = 0; i < CO_IRQ_APIC0 + CO_APIC_LAST + 1; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; + struct irq_desc *desc = irq_to_desc(i); + + desc->status = IRQ_DISABLED; + desc->action = 0; + desc->depth = 1; if (i == 0) { - irq_desc[i].chip = &cobalt_irq_type; + desc->chip = &cobalt_irq_type; } else if (i == CO_IRQ_IDE0) { - irq_desc[i].chip = &cobalt_irq_type; + desc->chip = &cobalt_irq_type; } else if (i == CO_IRQ_IDE1) { - irq_desc[i].chip = &cobalt_irq_type; + desc->chip = &cobalt_irq_type; } else if (i == CO_IRQ_8259) { - irq_desc[i].chip = &piix4_master_irq_type; + desc->chip = &piix4_master_irq_type; } else if (i < CO_IRQ_APIC0) { - irq_desc[i].chip = &piix4_virtual_irq_type; + desc->chip = &piix4_virtual_irq_type; } else if (IS_CO_APIC(i)) { - irq_desc[i].chip = &cobalt_irq_type; + desc->chip = &cobalt_irq_type; } } diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c index 6953859fe28..254ee07f863 100644 --- a/arch/x86/kernel/vmiclock_32.c +++ b/arch/x86/kernel/vmiclock_32.c @@ -235,11 +235,14 @@ static void __devinit vmi_time_init_clockevent(void) void __init vmi_time_init(void) { + unsigned int cpu; /* Disable PIT: BIOSes start PIT CH0 with 18.2hz peridic. */ outb_pit(0x3a, PIT_MODE); /* binary, mode 5, LSB/MSB, ch 0 */ vmi_time_init_clockevent(); setup_irq(0, &vmi_clock_action); + for_each_possible_cpu(cpu) + per_cpu(vector_irq, cpu)[vmi_get_timer_vector()] = 0; } #ifdef CONFIG_X86_LOCAL_APIC diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile index d0e940bb6f4..c02343594b4 100644 --- a/arch/x86/kvm/Makefile +++ b/arch/x86/kvm/Makefile @@ -3,10 +3,13 @@ # common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \ - coalesced_mmio.o) + coalesced_mmio.o irq_comm.o) ifeq ($(CONFIG_KVM_TRACE),y) common-objs += $(addprefix ../../../virt/kvm/, kvm_trace.o) endif +ifeq ($(CONFIG_DMAR),y) +common-objs += $(addprefix ../../../virt/kvm/, vtd.o) +endif EXTRA_CFLAGS += -Ivirt/kvm -Iarch/x86/kvm diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index c0f7872a912..634132a9a51 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -200,13 +200,14 @@ static int __pit_timer_fn(struct kvm_kpit_state *ps) if (!atomic_inc_and_test(&pt->pending)) set_bit(KVM_REQ_PENDING_TIMER, &vcpu0->requests); - if (vcpu0 && waitqueue_active(&vcpu0->wq)) { - vcpu0->arch.mp_state = KVM_MP_STATE_RUNNABLE; + + if (vcpu0 && waitqueue_active(&vcpu0->wq)) wake_up_interruptible(&vcpu0->wq); - } pt->timer.expires = ktime_add_ns(pt->timer.expires, pt->period); pt->scheduled = ktime_to_ns(pt->timer.expires); + if (pt->period) + ps->channels[0].count_load_time = pt->timer.expires; return (pt->period == 0 ? 0 : 1); } @@ -215,12 +216,22 @@ int pit_has_pending_timer(struct kvm_vcpu *vcpu) { struct kvm_pit *pit = vcpu->kvm->arch.vpit; - if (pit && vcpu->vcpu_id == 0 && pit->pit_state.inject_pending) + if (pit && vcpu->vcpu_id == 0 && pit->pit_state.irq_ack) return atomic_read(&pit->pit_state.pit_timer.pending); - return 0; } +static void kvm_pit_ack_irq(struct kvm_irq_ack_notifier *kian) +{ + struct kvm_kpit_state *ps = container_of(kian, struct kvm_kpit_state, + irq_ack_notifier); + spin_lock(&ps->inject_lock); + if (atomic_dec_return(&ps->pit_timer.pending) < 0) + atomic_inc(&ps->pit_timer.pending); + ps->irq_ack = 1; + spin_unlock(&ps->inject_lock); +} + static enum hrtimer_restart pit_timer_fn(struct hrtimer *data) { struct kvm_kpit_state *ps; @@ -255,8 +266,9 @@ static void destroy_pit_timer(struct kvm_kpit_timer *pt) hrtimer_cancel(&pt->timer); } -static void create_pit_timer(struct kvm_kpit_timer *pt, u32 val, int is_period) +static void create_pit_timer(struct kvm_kpit_state *ps, u32 val, int is_period) { + struct kvm_kpit_timer *pt = &ps->pit_timer; s64 interval; interval = muldiv64(val, NSEC_PER_SEC, KVM_PIT_FREQ); @@ -268,6 +280,7 @@ static void create_pit_timer(struct kvm_kpit_timer *pt, u32 val, int is_period) pt->period = (is_period == 0) ? 0 : interval; pt->timer.function = pit_timer_fn; atomic_set(&pt->pending, 0); + ps->irq_ack = 1; hrtimer_start(&pt->timer, ktime_add_ns(ktime_get(), interval), HRTIMER_MODE_ABS); @@ -302,11 +315,11 @@ static void pit_load_count(struct kvm *kvm, int channel, u32 val) case 1: /* FIXME: enhance mode 4 precision */ case 4: - create_pit_timer(&ps->pit_timer, val, 0); + create_pit_timer(ps, val, 0); break; case 2: case 3: - create_pit_timer(&ps->pit_timer, val, 1); + create_pit_timer(ps, val, 1); break; default: destroy_pit_timer(&ps->pit_timer); @@ -520,7 +533,7 @@ void kvm_pit_reset(struct kvm_pit *pit) mutex_unlock(&pit->pit_state.lock); atomic_set(&pit->pit_state.pit_timer.pending, 0); - pit->pit_state.inject_pending = 1; + pit->pit_state.irq_ack = 1; } struct kvm_pit *kvm_create_pit(struct kvm *kvm) @@ -534,6 +547,7 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm) mutex_init(&pit->pit_state.lock); mutex_lock(&pit->pit_state.lock); + spin_lock_init(&pit->pit_state.inject_lock); /* Initialize PIO device */ pit->dev.read = pit_ioport_read; @@ -555,6 +569,9 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm) pit_state->pit = pit; hrtimer_init(&pit_state->pit_timer.timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS); + pit_state->irq_ack_notifier.gsi = 0; + pit_state->irq_ack_notifier.irq_acked = kvm_pit_ack_irq; + kvm_register_irq_ack_notifier(kvm, &pit_state->irq_ack_notifier); mutex_unlock(&pit->pit_state.lock); kvm_pit_reset(pit); @@ -578,10 +595,8 @@ void kvm_free_pit(struct kvm *kvm) static void __inject_pit_timer_intr(struct kvm *kvm) { mutex_lock(&kvm->lock); - kvm_ioapic_set_irq(kvm->arch.vioapic, 0, 1); - kvm_ioapic_set_irq(kvm->arch.vioapic, 0, 0); - kvm_pic_set_irq(pic_irqchip(kvm), 0, 1); - kvm_pic_set_irq(pic_irqchip(kvm), 0, 0); + kvm_set_irq(kvm, 0, 1); + kvm_set_irq(kvm, 0, 0); mutex_unlock(&kvm->lock); } @@ -592,37 +607,19 @@ void kvm_inject_pit_timer_irqs(struct kvm_vcpu *vcpu) struct kvm_kpit_state *ps; if (vcpu && pit) { + int inject = 0; ps = &pit->pit_state; - /* Try to inject pending interrupts when: - * 1. Pending exists - * 2. Last interrupt was accepted or waited for too long time*/ - if (atomic_read(&ps->pit_timer.pending) && - (ps->inject_pending || - (jiffies - ps->last_injected_time - >= KVM_MAX_PIT_INTR_INTERVAL))) { - ps->inject_pending = 0; - __inject_pit_timer_intr(kvm); - ps->last_injected_time = jiffies; - } - } -} - -void kvm_pit_timer_intr_post(struct kvm_vcpu *vcpu, int vec) -{ - struct kvm_arch *arch = &vcpu->kvm->arch; - struct kvm_kpit_state *ps; - - if (vcpu && arch->vpit) { - ps = &arch->vpit->pit_state; - if (atomic_read(&ps->pit_timer.pending) && - (((arch->vpic->pics[0].imr & 1) == 0 && - arch->vpic->pics[0].irq_base == vec) || - (arch->vioapic->redirtbl[0].fields.vector == vec && - arch->vioapic->redirtbl[0].fields.mask != 1))) { - ps->inject_pending = 1; - atomic_dec(&ps->pit_timer.pending); - ps->channels[0].count_load_time = ktime_get(); + /* Try to inject pending interrupts when + * last one has been acked. + */ + spin_lock(&ps->inject_lock); + if (atomic_read(&ps->pit_timer.pending) && ps->irq_ack) { + ps->irq_ack = 0; + inject = 1; } + spin_unlock(&ps->inject_lock); + if (inject) + __inject_pit_timer_intr(kvm); } } diff --git a/arch/x86/kvm/i8254.h b/arch/x86/kvm/i8254.h index db25c2a6c8c..e436d4983aa 100644 --- a/arch/x86/kvm/i8254.h +++ b/arch/x86/kvm/i8254.h @@ -8,7 +8,6 @@ struct kvm_kpit_timer { int irq; s64 period; /* unit: ns */ s64 scheduled; - ktime_t last_update; atomic_t pending; }; @@ -34,8 +33,9 @@ struct kvm_kpit_state { u32 speaker_data_on; struct mutex lock; struct kvm_pit *pit; - bool inject_pending; /* if inject pending interrupts */ - unsigned long last_injected_time; + spinlock_t inject_lock; + unsigned long irq_ack; + struct kvm_irq_ack_notifier irq_ack_notifier; }; struct kvm_pit { @@ -54,7 +54,6 @@ struct kvm_pit { #define KVM_PIT_CHANNEL_MASK 0x3 void kvm_inject_pit_timer_irqs(struct kvm_vcpu *vcpu); -void kvm_pit_timer_intr_post(struct kvm_vcpu *vcpu, int vec); void kvm_pit_load_count(struct kvm *kvm, int channel, u32 val); struct kvm_pit *kvm_create_pit(struct kvm *kvm); void kvm_free_pit(struct kvm *kvm); diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c index c31164e8aa4..17e41e165f1 100644 --- a/arch/x86/kvm/i8259.c +++ b/arch/x86/kvm/i8259.c @@ -30,6 +30,19 @@ #include <linux/kvm_host.h> +static void pic_clear_isr(struct kvm_kpic_state *s, int irq) +{ + s->isr &= ~(1 << irq); + s->isr_ack |= (1 << irq); +} + +void kvm_pic_clear_isr_ack(struct kvm *kvm) +{ + struct kvm_pic *s = pic_irqchip(kvm); + s->pics[0].isr_ack = 0xff; + s->pics[1].isr_ack = 0xff; +} + /* * set irq level. If an edge is detected, then the IRR is set to 1 */ @@ -141,11 +154,12 @@ void kvm_pic_set_irq(void *opaque, int irq, int level) */ static inline void pic_intack(struct kvm_kpic_state *s, int irq) { + s->isr |= 1 << irq; if (s->auto_eoi) { if (s->rotate_on_auto_eoi) s->priority_add = (irq + 1) & 7; - } else - s->isr |= (1 << irq); + pic_clear_isr(s, irq); + } /* * We don't clear a level sensitive interrupt here */ @@ -153,9 +167,10 @@ static inline void pic_intack(struct kvm_kpic_state *s, int irq) s->irr &= ~(1 << irq); } -int kvm_pic_read_irq(struct kvm_pic *s) +int kvm_pic_read_irq(struct kvm *kvm) { int irq, irq2, intno; + struct kvm_pic *s = pic_irqchip(kvm); irq = pic_get_irq(&s->pics[0]); if (irq >= 0) { @@ -181,16 +196,32 @@ int kvm_pic_read_irq(struct kvm_pic *s) intno = s->pics[0].irq_base + irq; } pic_update_irq(s); + kvm_notify_acked_irq(kvm, irq); return intno; } void kvm_pic_reset(struct kvm_kpic_state *s) { + int irq, irqbase; + struct kvm *kvm = s->pics_state->irq_request_opaque; + struct kvm_vcpu *vcpu0 = kvm->vcpus[0]; + + if (s == &s->pics_state->pics[0]) + irqbase = 0; + else + irqbase = 8; + + for (irq = 0; irq < PIC_NUM_PINS/2; irq++) { + if (vcpu0 && kvm_apic_accept_pic_intr(vcpu0)) + if (s->irr & (1 << irq) || s->isr & (1 << irq)) + kvm_notify_acked_irq(kvm, irq+irqbase); + } s->last_irr = 0; s->irr = 0; s->imr = 0; s->isr = 0; + s->isr_ack = 0xff; s->priority_add = 0; s->irq_base = 0; s->read_reg_select = 0; @@ -243,7 +274,7 @@ static void pic_ioport_write(void *opaque, u32 addr, u32 val) priority = get_priority(s, s->isr); if (priority != 8) { irq = (priority + s->priority_add) & 7; - s->isr &= ~(1 << irq); + pic_clear_isr(s, irq); if (cmd == 5) s->priority_add = (irq + 1) & 7; pic_update_irq(s->pics_state); @@ -251,7 +282,7 @@ static void pic_ioport_write(void *opaque, u32 addr, u32 val) break; case 3: irq = val & 7; - s->isr &= ~(1 << irq); + pic_clear_isr(s, irq); pic_update_irq(s->pics_state); break; case 6: @@ -260,8 +291,8 @@ static void pic_ioport_write(void *opaque, u32 addr, u32 val) break; case 7: irq = val & 7; - s->isr &= ~(1 << irq); s->priority_add = (irq + 1) & 7; + pic_clear_isr(s, irq); pic_update_irq(s->pics_state); break; default: @@ -303,7 +334,7 @@ static u32 pic_poll_read(struct kvm_kpic_state *s, u32 addr1) s->pics_state->pics[0].irr &= ~(1 << 2); } s->irr &= ~(1 << ret); - s->isr &= ~(1 << ret); + pic_clear_isr(s, ret); if (addr1 >> 7 || ret != 2) pic_update_irq(s->pics_state); } else { @@ -422,10 +453,14 @@ static void pic_irq_request(void *opaque, int level) { struct kvm *kvm = opaque; struct kvm_vcpu *vcpu = kvm->vcpus[0]; + struct kvm_pic *s = pic_irqchip(kvm); + int irq = pic_get_irq(&s->pics[0]); - pic_irqchip(kvm)->output = level; - if (vcpu) + s->output = level; + if (vcpu && level && (s->pics[0].isr_ack & (1 << irq))) { + s->pics[0].isr_ack &= ~(1 << irq); kvm_vcpu_kick(vcpu); + } } struct kvm_pic *kvm_create_pic(struct kvm *kvm) diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c index 76d736b5f66..c019b8edcdb 100644 --- a/arch/x86/kvm/irq.c +++ b/arch/x86/kvm/irq.c @@ -72,7 +72,7 @@ int kvm_cpu_get_interrupt(struct kvm_vcpu *v) if (kvm_apic_accept_pic_intr(v)) { s = pic_irqchip(v->kvm); s->output = 0; /* PIC */ - vector = kvm_pic_read_irq(s); + vector = kvm_pic_read_irq(v->kvm); } } return vector; @@ -90,7 +90,6 @@ EXPORT_SYMBOL_GPL(kvm_inject_pending_timer_irqs); void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec) { kvm_apic_timer_intr_post(vcpu, vec); - kvm_pit_timer_intr_post(vcpu, vec); /* TODO: PIT, RTC etc. */ } EXPORT_SYMBOL_GPL(kvm_timer_intr_post); diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index 7ca47cbb48b..f17c8f5bbf3 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -42,6 +42,7 @@ struct kvm_kpic_state { u8 irr; /* interrupt request register */ u8 imr; /* interrupt mask register */ u8 isr; /* interrupt service register */ + u8 isr_ack; /* interrupt ack detection */ u8 priority_add; /* highest irq priority */ u8 irq_base; u8 read_reg_select; @@ -63,12 +64,13 @@ struct kvm_pic { void *irq_request_opaque; int output; /* intr from master PIC */ struct kvm_io_device dev; + void (*ack_notifier)(void *opaque, int irq); }; struct kvm_pic *kvm_create_pic(struct kvm *kvm); -void kvm_pic_set_irq(void *opaque, int irq, int level); -int kvm_pic_read_irq(struct kvm_pic *s); +int kvm_pic_read_irq(struct kvm *kvm); void kvm_pic_update_irq(struct kvm_pic *s); +void kvm_pic_clear_isr_ack(struct kvm *kvm); static inline struct kvm_pic *pic_irqchip(struct kvm *kvm) { diff --git a/arch/x86/kvm/kvm_cache_regs.h b/arch/x86/kvm/kvm_cache_regs.h new file mode 100644 index 00000000000..1ff819dce7d --- /dev/null +++ b/arch/x86/kvm/kvm_cache_regs.h @@ -0,0 +1,32 @@ +#ifndef ASM_KVM_CACHE_REGS_H +#define ASM_KVM_CACHE_REGS_H + +static inline unsigned long kvm_register_read(struct kvm_vcpu *vcpu, + enum kvm_reg reg) +{ + if (!test_bit(reg, (unsigned long *)&vcpu->arch.regs_avail)) + kvm_x86_ops->cache_reg(vcpu, reg); + + return vcpu->arch.regs[reg]; +} + +static inline void kvm_register_write(struct kvm_vcpu *vcpu, + enum kvm_reg reg, + unsigned long val) +{ + vcpu->arch.regs[reg] = val; + __set_bit(reg, (unsigned long *)&vcpu->arch.regs_dirty); + __set_bit(reg, (unsigned long *)&vcpu->arch.regs_avail); +} + +static inline unsigned long kvm_rip_read(struct kvm_vcpu *vcpu) +{ + return kvm_register_read(vcpu, VCPU_REGS_RIP); +} + +static inline void kvm_rip_write(struct kvm_vcpu *vcpu, unsigned long val) +{ + kvm_register_write(vcpu, VCPU_REGS_RIP, val); +} + +#endif diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 73f43de69f6..6571926bfd3 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -32,6 +32,7 @@ #include <asm/current.h> #include <asm/apicdef.h> #include <asm/atomic.h> +#include "kvm_cache_regs.h" #include "irq.h" #define PRId64 "d" @@ -338,13 +339,7 @@ static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode, } else apic_clear_vector(vector, apic->regs + APIC_TMR); - if (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE) - kvm_vcpu_kick(vcpu); - else if (vcpu->arch.mp_state == KVM_MP_STATE_HALTED) { - vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; - if (waitqueue_active(&vcpu->wq)) - wake_up_interruptible(&vcpu->wq); - } + kvm_vcpu_kick(vcpu); result = (orig_irr == 0); break; @@ -370,21 +365,18 @@ static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode, vcpu->arch.mp_state = KVM_MP_STATE_INIT_RECEIVED; kvm_vcpu_kick(vcpu); } else { - printk(KERN_DEBUG - "Ignoring de-assert INIT to vcpu %d\n", - vcpu->vcpu_id); + apic_debug("Ignoring de-assert INIT to vcpu %d\n", + vcpu->vcpu_id); } - break; case APIC_DM_STARTUP: - printk(KERN_DEBUG "SIPI to vcpu %d vector 0x%02x\n", - vcpu->vcpu_id, vector); + apic_debug("SIPI to vcpu %d vector 0x%02x\n", + vcpu->vcpu_id, vector); if (vcpu->arch.mp_state == KVM_MP_STATE_INIT_RECEIVED) { vcpu->arch.sipi_vector = vector; vcpu->arch.mp_state = KVM_MP_STATE_SIPI_RECEIVED; - if (waitqueue_active(&vcpu->wq)) - wake_up_interruptible(&vcpu->wq); + kvm_vcpu_kick(vcpu); } break; @@ -438,7 +430,7 @@ struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector, static void apic_set_eoi(struct kvm_lapic *apic) { int vector = apic_find_highest_isr(apic); - + int trigger_mode; /* * Not every write EOI will has corresponding ISR, * one example is when Kernel check timer on setup_IO_APIC @@ -450,7 +442,10 @@ static void apic_set_eoi(struct kvm_lapic *apic) apic_update_ppr(apic); if (apic_test_and_clear_vector(vector, apic->regs + APIC_TMR)) - kvm_ioapic_update_eoi(apic->vcpu->kvm, vector); + trigger_mode = IOAPIC_LEVEL_TRIG; + else + trigger_mode = IOAPIC_EDGE_TRIG; + kvm_ioapic_update_eoi(apic->vcpu->kvm, vector, trigger_mode); } static void apic_send_ipi(struct kvm_lapic *apic) @@ -558,8 +553,7 @@ static void __report_tpr_access(struct kvm_lapic *apic, bool write) struct kvm_run *run = vcpu->run; set_bit(KVM_REQ_REPORT_TPR_ACCESS, &vcpu->requests); - kvm_x86_ops->cache_regs(vcpu); - run->tpr_access.rip = vcpu->arch.rip; + run->tpr_access.rip = kvm_rip_read(vcpu); run->tpr_access.is_write = write; } @@ -683,9 +677,9 @@ static void apic_mmio_write(struct kvm_io_device *this, * Refer SDM 8.4.1 */ if (len != 4 || alignment) { - if (printk_ratelimit()) - printk(KERN_ERR "apic write: bad size=%d %lx\n", - len, (long)address); + /* Don't shout loud, $infamous_os would cause only noise. */ + apic_debug("apic write: bad size=%d %lx\n", + len, (long)address); return; } @@ -947,10 +941,9 @@ static int __apic_timer_fn(struct kvm_lapic *apic) if(!atomic_inc_and_test(&apic->timer.pending)) set_bit(KVM_REQ_PENDING_TIMER, &apic->vcpu->requests); - if (waitqueue_active(q)) { - apic->vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; + if (waitqueue_active(q)) wake_up_interruptible(q); - } + if (apic_lvtt_period(apic)) { result = 1; apic->timer.dev.expires = ktime_add_ns( diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 3da2508eb22..99c239c5c0a 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -70,6 +70,9 @@ static int dbg = 0; module_param(dbg, bool, 0644); #endif +static int oos_shadow = 1; +module_param(oos_shadow, bool, 0644); + #ifndef MMU_DEBUG #define ASSERT(x) do { } while (0) #else @@ -135,18 +138,24 @@ module_param(dbg, bool, 0644); #define ACC_USER_MASK PT_USER_MASK #define ACC_ALL (ACC_EXEC_MASK | ACC_WRITE_MASK | ACC_USER_MASK) -struct kvm_pv_mmu_op_buffer { - void *ptr; - unsigned len; - unsigned processed; - char buf[512] __aligned(sizeof(long)); -}; +#define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level) struct kvm_rmap_desc { u64 *shadow_ptes[RMAP_EXT]; struct kvm_rmap_desc *more; }; +struct kvm_shadow_walk { + int (*entry)(struct kvm_shadow_walk *walk, struct kvm_vcpu *vcpu, + u64 addr, u64 *spte, int level); +}; + +struct kvm_unsync_walk { + int (*entry) (struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk); +}; + +typedef int (*mmu_parent_walk_fn) (struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp); + static struct kmem_cache *pte_chain_cache; static struct kmem_cache *rmap_desc_cache; static struct kmem_cache *mmu_page_header_cache; @@ -405,16 +414,19 @@ static int host_largepage_backed(struct kvm *kvm, gfn_t gfn) { struct vm_area_struct *vma; unsigned long addr; + int ret = 0; addr = gfn_to_hva(kvm, gfn); if (kvm_is_error_hva(addr)) - return 0; + return ret; + down_read(¤t->mm->mmap_sem); vma = find_vma(current->mm, addr); if (vma && is_vm_hugetlb_page(vma)) - return 1; + ret = 1; + up_read(¤t->mm->mmap_sem); - return 0; + return ret; } static int is_largepage_backed(struct kvm_vcpu *vcpu, gfn_t large_gfn) @@ -649,8 +661,6 @@ static void rmap_write_protect(struct kvm *kvm, u64 gfn) if (write_protected) kvm_flush_remote_tlbs(kvm); - - account_shadowed(kvm, gfn); } static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp) @@ -859,6 +869,77 @@ static void mmu_page_remove_parent_pte(struct kvm_mmu_page *sp, BUG(); } + +static void mmu_parent_walk(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, + mmu_parent_walk_fn fn) +{ + struct kvm_pte_chain *pte_chain; + struct hlist_node *node; + struct kvm_mmu_page *parent_sp; + int i; + + if (!sp->multimapped && sp->parent_pte) { + parent_sp = page_header(__pa(sp->parent_pte)); + fn(vcpu, parent_sp); + mmu_parent_walk(vcpu, parent_sp, fn); + return; + } + hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link) + for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i) { + if (!pte_chain->parent_ptes[i]) + break; + parent_sp = page_header(__pa(pte_chain->parent_ptes[i])); + fn(vcpu, parent_sp); + mmu_parent_walk(vcpu, parent_sp, fn); + } +} + +static void kvm_mmu_update_unsync_bitmap(u64 *spte) +{ + unsigned int index; + struct kvm_mmu_page *sp = page_header(__pa(spte)); + + index = spte - sp->spt; + __set_bit(index, sp->unsync_child_bitmap); + sp->unsync_children = 1; +} + +static void kvm_mmu_update_parents_unsync(struct kvm_mmu_page *sp) +{ + struct kvm_pte_chain *pte_chain; + struct hlist_node *node; + int i; + + if (!sp->parent_pte) + return; + + if (!sp->multimapped) { + kvm_mmu_update_unsync_bitmap(sp->parent_pte); + return; + } + + hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link) + for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i) { + if (!pte_chain->parent_ptes[i]) + break; + kvm_mmu_update_unsync_bitmap(pte_chain->parent_ptes[i]); + } +} + +static int unsync_walk_fn(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + sp->unsync_children = 1; + kvm_mmu_update_parents_unsync(sp); + return 1; +} + +static void kvm_mmu_mark_parents_unsync(struct kvm_vcpu *vcpu, + struct kvm_mmu_page *sp) +{ + mmu_parent_walk(vcpu, sp, unsync_walk_fn); + kvm_mmu_update_parents_unsync(sp); +} + static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) { @@ -868,6 +949,58 @@ static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu, sp->spt[i] = shadow_trap_nonpresent_pte; } +static int nonpaging_sync_page(struct kvm_vcpu *vcpu, + struct kvm_mmu_page *sp) +{ + return 1; +} + +static void nonpaging_invlpg(struct kvm_vcpu *vcpu, gva_t gva) +{ +} + +#define for_each_unsync_children(bitmap, idx) \ + for (idx = find_first_bit(bitmap, 512); \ + idx < 512; \ + idx = find_next_bit(bitmap, 512, idx+1)) + +static int mmu_unsync_walk(struct kvm_mmu_page *sp, + struct kvm_unsync_walk *walker) +{ + int i, ret; + + if (!sp->unsync_children) + return 0; + + for_each_unsync_children(sp->unsync_child_bitmap, i) { + u64 ent = sp->spt[i]; + + if (is_shadow_present_pte(ent)) { + struct kvm_mmu_page *child; + child = page_header(ent & PT64_BASE_ADDR_MASK); + + if (child->unsync_children) { + ret = mmu_unsync_walk(child, walker); + if (ret) + return ret; + __clear_bit(i, sp->unsync_child_bitmap); + } + + if (child->unsync) { + ret = walker->entry(child, walker); + __clear_bit(i, sp->unsync_child_bitmap); + if (ret) + return ret; + } + } + } + + if (find_first_bit(sp->unsync_child_bitmap, 512) == 512) + sp->unsync_children = 0; + + return 0; +} + static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) { unsigned index; @@ -888,6 +1021,59 @@ static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) return NULL; } +static void kvm_unlink_unsync_page(struct kvm *kvm, struct kvm_mmu_page *sp) +{ + WARN_ON(!sp->unsync); + sp->unsync = 0; + --kvm->stat.mmu_unsync; +} + +static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp); + +static int kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + if (sp->role.glevels != vcpu->arch.mmu.root_level) { + kvm_mmu_zap_page(vcpu->kvm, sp); + return 1; + } + + rmap_write_protect(vcpu->kvm, sp->gfn); + if (vcpu->arch.mmu.sync_page(vcpu, sp)) { + kvm_mmu_zap_page(vcpu->kvm, sp); + return 1; + } + + kvm_mmu_flush_tlb(vcpu); + kvm_unlink_unsync_page(vcpu->kvm, sp); + return 0; +} + +struct sync_walker { + struct kvm_vcpu *vcpu; + struct kvm_unsync_walk walker; +}; + +static int mmu_sync_fn(struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk) +{ + struct sync_walker *sync_walk = container_of(walk, struct sync_walker, + walker); + struct kvm_vcpu *vcpu = sync_walk->vcpu; + + kvm_sync_page(vcpu, sp); + return (need_resched() || spin_needbreak(&vcpu->kvm->mmu_lock)); +} + +static void mmu_sync_children(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + struct sync_walker walker = { + .walker = { .entry = mmu_sync_fn, }, + .vcpu = vcpu, + }; + + while (mmu_unsync_walk(sp, &walker.walker)) + cond_resched_lock(&vcpu->kvm->mmu_lock); +} + static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gaddr, @@ -901,7 +1087,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, unsigned quadrant; struct hlist_head *bucket; struct kvm_mmu_page *sp; - struct hlist_node *node; + struct hlist_node *node, *tmp; role.word = 0; role.glevels = vcpu->arch.mmu.root_level; @@ -917,9 +1103,20 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, gfn, role.word); index = kvm_page_table_hashfn(gfn); bucket = &vcpu->kvm->arch.mmu_page_hash[index]; - hlist_for_each_entry(sp, node, bucket, hash_link) - if (sp->gfn == gfn && sp->role.word == role.word) { + hlist_for_each_entry_safe(sp, node, tmp, bucket, hash_link) + if (sp->gfn == gfn) { + if (sp->unsync) + if (kvm_sync_page(vcpu, sp)) + continue; + + if (sp->role.word != role.word) + continue; + mmu_page_add_parent_pte(vcpu, sp, parent_pte); + if (sp->unsync_children) { + set_bit(KVM_REQ_MMU_SYNC, &vcpu->requests); + kvm_mmu_mark_parents_unsync(vcpu, sp); + } pgprintk("%s: found\n", __func__); return sp; } @@ -931,8 +1128,10 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, sp->gfn = gfn; sp->role = role; hlist_add_head(&sp->hash_link, bucket); - if (!metaphysical) + if (!metaphysical) { rmap_write_protect(vcpu->kvm, gfn); + account_shadowed(vcpu->kvm, gfn); + } if (shadow_trap_nonpresent_pte != shadow_notrap_nonpresent_pte) vcpu->arch.mmu.prefetch_page(vcpu, sp); else @@ -940,6 +1139,35 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, return sp; } +static int walk_shadow(struct kvm_shadow_walk *walker, + struct kvm_vcpu *vcpu, u64 addr) +{ + hpa_t shadow_addr; + int level; + int r; + u64 *sptep; + unsigned index; + + shadow_addr = vcpu->arch.mmu.root_hpa; + level = vcpu->arch.mmu.shadow_root_level; + if (level == PT32E_ROOT_LEVEL) { + shadow_addr = vcpu->arch.mmu.pae_root[(addr >> 30) & 3]; + shadow_addr &= PT64_BASE_ADDR_MASK; + --level; + } + + while (level >= PT_PAGE_TABLE_LEVEL) { + index = SHADOW_PT_INDEX(addr, level); + sptep = ((u64 *)__va(shadow_addr)) + index; + r = walker->entry(walker, vcpu, addr, sptep, level); + if (r) + return r; + shadow_addr = *sptep & PT64_BASE_ADDR_MASK; + --level; + } + return 0; +} + static void kvm_mmu_page_unlink_children(struct kvm *kvm, struct kvm_mmu_page *sp) { @@ -955,7 +1183,6 @@ static void kvm_mmu_page_unlink_children(struct kvm *kvm, rmap_remove(kvm, &pt[i]); pt[i] = shadow_trap_nonpresent_pte; } - kvm_flush_remote_tlbs(kvm); return; } @@ -974,7 +1201,6 @@ static void kvm_mmu_page_unlink_children(struct kvm *kvm, } pt[i] = shadow_trap_nonpresent_pte; } - kvm_flush_remote_tlbs(kvm); } static void kvm_mmu_put_page(struct kvm_mmu_page *sp, u64 *parent_pte) @@ -991,11 +1217,10 @@ static void kvm_mmu_reset_last_pte_updated(struct kvm *kvm) kvm->vcpus[i]->arch.last_pte_updated = NULL; } -static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) +static void kvm_mmu_unlink_parents(struct kvm *kvm, struct kvm_mmu_page *sp) { u64 *parent_pte; - ++kvm->stat.mmu_shadow_zapped; while (sp->multimapped || sp->parent_pte) { if (!sp->multimapped) parent_pte = sp->parent_pte; @@ -1010,21 +1235,59 @@ static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) kvm_mmu_put_page(sp, parent_pte); set_shadow_pte(parent_pte, shadow_trap_nonpresent_pte); } +} + +struct zap_walker { + struct kvm_unsync_walk walker; + struct kvm *kvm; + int zapped; +}; + +static int mmu_zap_fn(struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk) +{ + struct zap_walker *zap_walk = container_of(walk, struct zap_walker, + walker); + kvm_mmu_zap_page(zap_walk->kvm, sp); + zap_walk->zapped = 1; + return 0; +} + +static int mmu_zap_unsync_children(struct kvm *kvm, struct kvm_mmu_page *sp) +{ + struct zap_walker walker = { + .walker = { .entry = mmu_zap_fn, }, + .kvm = kvm, + .zapped = 0, + }; + + if (sp->role.level == PT_PAGE_TABLE_LEVEL) + return 0; + mmu_unsync_walk(sp, &walker.walker); + return walker.zapped; +} + +static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) +{ + int ret; + ++kvm->stat.mmu_shadow_zapped; + ret = mmu_zap_unsync_children(kvm, sp); kvm_mmu_page_unlink_children(kvm, sp); + kvm_mmu_unlink_parents(kvm, sp); + kvm_flush_remote_tlbs(kvm); + if (!sp->role.invalid && !sp->role.metaphysical) + unaccount_shadowed(kvm, sp->gfn); + if (sp->unsync) + kvm_unlink_unsync_page(kvm, sp); if (!sp->root_count) { - if (!sp->role.metaphysical && !sp->role.invalid) - unaccount_shadowed(kvm, sp->gfn); hlist_del(&sp->hash_link); kvm_mmu_free_page(kvm, sp); } else { - int invalid = sp->role.invalid; - list_move(&sp->link, &kvm->arch.active_mmu_pages); sp->role.invalid = 1; + list_move(&sp->link, &kvm->arch.active_mmu_pages); kvm_reload_remote_mmus(kvm); - if (!sp->role.metaphysical && !invalid) - unaccount_shadowed(kvm, sp->gfn); } kvm_mmu_reset_last_pte_updated(kvm); + return ret; } /* @@ -1077,8 +1340,9 @@ static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn) if (sp->gfn == gfn && !sp->role.metaphysical) { pgprintk("%s: gfn %lx role %x\n", __func__, gfn, sp->role.word); - kvm_mmu_zap_page(kvm, sp); r = 1; + if (kvm_mmu_zap_page(kvm, sp)) + n = bucket->first; } return r; } @@ -1101,6 +1365,20 @@ static void page_header_update_slot(struct kvm *kvm, void *pte, gfn_t gfn) __set_bit(slot, &sp->slot_bitmap); } +static void mmu_convert_notrap(struct kvm_mmu_page *sp) +{ + int i; + u64 *pt = sp->spt; + + if (shadow_trap_nonpresent_pte == shadow_notrap_nonpresent_pte) + return; + + for (i = 0; i < PT64_ENT_PER_PAGE; ++i) { + if (pt[i] == shadow_notrap_nonpresent_pte) + set_shadow_pte(&pt[i], shadow_trap_nonpresent_pte); + } +} + struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva) { struct page *page; @@ -1110,51 +1388,60 @@ struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva) if (gpa == UNMAPPED_GVA) return NULL; - down_read(¤t->mm->mmap_sem); page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT); - up_read(¤t->mm->mmap_sem); return page; } -static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, - unsigned pt_access, unsigned pte_access, - int user_fault, int write_fault, int dirty, - int *ptwrite, int largepage, gfn_t gfn, - pfn_t pfn, bool speculative) +static int kvm_unsync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) { - u64 spte; - int was_rmapped = 0; - int was_writeble = is_writeble_pte(*shadow_pte); + unsigned index; + struct hlist_head *bucket; + struct kvm_mmu_page *s; + struct hlist_node *node, *n; - pgprintk("%s: spte %llx access %x write_fault %d" - " user_fault %d gfn %lx\n", - __func__, *shadow_pte, pt_access, - write_fault, user_fault, gfn); + index = kvm_page_table_hashfn(sp->gfn); + bucket = &vcpu->kvm->arch.mmu_page_hash[index]; + /* don't unsync if pagetable is shadowed with multiple roles */ + hlist_for_each_entry_safe(s, node, n, bucket, hash_link) { + if (s->gfn != sp->gfn || s->role.metaphysical) + continue; + if (s->role.word != sp->role.word) + return 1; + } + kvm_mmu_mark_parents_unsync(vcpu, sp); + ++vcpu->kvm->stat.mmu_unsync; + sp->unsync = 1; + mmu_convert_notrap(sp); + return 0; +} - if (is_rmap_pte(*shadow_pte)) { - /* - * If we overwrite a PTE page pointer with a 2MB PMD, unlink - * the parent of the now unreachable PTE. - */ - if (largepage && !is_large_pte(*shadow_pte)) { - struct kvm_mmu_page *child; - u64 pte = *shadow_pte; +static int mmu_need_write_protect(struct kvm_vcpu *vcpu, gfn_t gfn, + bool can_unsync) +{ + struct kvm_mmu_page *shadow; - child = page_header(pte & PT64_BASE_ADDR_MASK); - mmu_page_remove_parent_pte(child, shadow_pte); - } else if (pfn != spte_to_pfn(*shadow_pte)) { - pgprintk("hfn old %lx new %lx\n", - spte_to_pfn(*shadow_pte), pfn); - rmap_remove(vcpu->kvm, shadow_pte); - } else { - if (largepage) - was_rmapped = is_large_pte(*shadow_pte); - else - was_rmapped = 1; - } + shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn); + if (shadow) { + if (shadow->role.level != PT_PAGE_TABLE_LEVEL) + return 1; + if (shadow->unsync) + return 0; + if (can_unsync && oos_shadow) + return kvm_unsync_page(vcpu, shadow); + return 1; } + return 0; +} +static int set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, + unsigned pte_access, int user_fault, + int write_fault, int dirty, int largepage, + gfn_t gfn, pfn_t pfn, bool speculative, + bool can_unsync) +{ + u64 spte; + int ret = 0; /* * We don't set the accessed bit, since we sometimes want to see * whether the guest actually used the pte (in order to detect @@ -1162,7 +1449,7 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, */ spte = shadow_base_present_pte | shadow_dirty_mask; if (!speculative) - pte_access |= PT_ACCESSED_MASK; + spte |= shadow_accessed_mask; if (!dirty) pte_access &= ~ACC_WRITE_MASK; if (pte_access & ACC_EXEC_MASK) @@ -1178,35 +1465,82 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, if ((pte_access & ACC_WRITE_MASK) || (write_fault && !is_write_protection(vcpu) && !user_fault)) { - struct kvm_mmu_page *shadow; + + if (largepage && has_wrprotected_page(vcpu->kvm, gfn)) { + ret = 1; + spte = shadow_trap_nonpresent_pte; + goto set_pte; + } spte |= PT_WRITABLE_MASK; - shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn); - if (shadow || - (largepage && has_wrprotected_page(vcpu->kvm, gfn))) { + if (mmu_need_write_protect(vcpu, gfn, can_unsync)) { pgprintk("%s: found shadow page for %lx, marking ro\n", __func__, gfn); + ret = 1; pte_access &= ~ACC_WRITE_MASK; - if (is_writeble_pte(spte)) { + if (is_writeble_pte(spte)) spte &= ~PT_WRITABLE_MASK; - kvm_x86_ops->tlb_flush(vcpu); - } - if (write_fault) - *ptwrite = 1; } } if (pte_access & ACC_WRITE_MASK) mark_page_dirty(vcpu->kvm, gfn); - pgprintk("%s: setting spte %llx\n", __func__, spte); - pgprintk("instantiating %s PTE (%s) at %ld (%llx) addr %p\n", - (spte&PT_PAGE_SIZE_MASK)? "2MB" : "4kB", - (spte&PT_WRITABLE_MASK)?"RW":"R", gfn, spte, shadow_pte); +set_pte: set_shadow_pte(shadow_pte, spte); - if (!was_rmapped && (spte & PT_PAGE_SIZE_MASK) - && (spte & PT_PRESENT_MASK)) + return ret; +} + +static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, + unsigned pt_access, unsigned pte_access, + int user_fault, int write_fault, int dirty, + int *ptwrite, int largepage, gfn_t gfn, + pfn_t pfn, bool speculative) +{ + int was_rmapped = 0; + int was_writeble = is_writeble_pte(*shadow_pte); + + pgprintk("%s: spte %llx access %x write_fault %d" + " user_fault %d gfn %lx\n", + __func__, *shadow_pte, pt_access, + write_fault, user_fault, gfn); + + if (is_rmap_pte(*shadow_pte)) { + /* + * If we overwrite a PTE page pointer with a 2MB PMD, unlink + * the parent of the now unreachable PTE. + */ + if (largepage && !is_large_pte(*shadow_pte)) { + struct kvm_mmu_page *child; + u64 pte = *shadow_pte; + + child = page_header(pte & PT64_BASE_ADDR_MASK); + mmu_page_remove_parent_pte(child, shadow_pte); + } else if (pfn != spte_to_pfn(*shadow_pte)) { + pgprintk("hfn old %lx new %lx\n", + spte_to_pfn(*shadow_pte), pfn); + rmap_remove(vcpu->kvm, shadow_pte); + } else { + if (largepage) + was_rmapped = is_large_pte(*shadow_pte); + else + was_rmapped = 1; + } + } + if (set_spte(vcpu, shadow_pte, pte_access, user_fault, write_fault, + dirty, largepage, gfn, pfn, speculative, true)) { + if (write_fault) + *ptwrite = 1; + kvm_x86_ops->tlb_flush(vcpu); + } + + pgprintk("%s: setting spte %llx\n", __func__, *shadow_pte); + pgprintk("instantiating %s PTE (%s) at %ld (%llx) addr %p\n", + is_large_pte(*shadow_pte)? "2MB" : "4kB", + is_present_pte(*shadow_pte)?"RW":"R", gfn, + *shadow_pte, shadow_pte); + if (!was_rmapped && is_large_pte(*shadow_pte)) ++vcpu->kvm->stat.lpages; page_header_update_slot(vcpu->kvm, shadow_pte, gfn); @@ -1230,54 +1564,67 @@ static void nonpaging_new_cr3(struct kvm_vcpu *vcpu) { } -static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, - int largepage, gfn_t gfn, pfn_t pfn, - int level) -{ - hpa_t table_addr = vcpu->arch.mmu.root_hpa; - int pt_write = 0; - - for (; ; level--) { - u32 index = PT64_INDEX(v, level); - u64 *table; - - ASSERT(VALID_PAGE(table_addr)); - table = __va(table_addr); +struct direct_shadow_walk { + struct kvm_shadow_walk walker; + pfn_t pfn; + int write; + int largepage; + int pt_write; +}; - if (level == 1) { - mmu_set_spte(vcpu, &table[index], ACC_ALL, ACC_ALL, - 0, write, 1, &pt_write, 0, gfn, pfn, false); - return pt_write; - } +static int direct_map_entry(struct kvm_shadow_walk *_walk, + struct kvm_vcpu *vcpu, + u64 addr, u64 *sptep, int level) +{ + struct direct_shadow_walk *walk = + container_of(_walk, struct direct_shadow_walk, walker); + struct kvm_mmu_page *sp; + gfn_t pseudo_gfn; + gfn_t gfn = addr >> PAGE_SHIFT; + + if (level == PT_PAGE_TABLE_LEVEL + || (walk->largepage && level == PT_DIRECTORY_LEVEL)) { + mmu_set_spte(vcpu, sptep, ACC_ALL, ACC_ALL, + 0, walk->write, 1, &walk->pt_write, + walk->largepage, gfn, walk->pfn, false); + ++vcpu->stat.pf_fixed; + return 1; + } - if (largepage && level == 2) { - mmu_set_spte(vcpu, &table[index], ACC_ALL, ACC_ALL, - 0, write, 1, &pt_write, 1, gfn, pfn, false); - return pt_write; + if (*sptep == shadow_trap_nonpresent_pte) { + pseudo_gfn = (addr & PT64_DIR_BASE_ADDR_MASK) >> PAGE_SHIFT; + sp = kvm_mmu_get_page(vcpu, pseudo_gfn, (gva_t)addr, level - 1, + 1, ACC_ALL, sptep); + if (!sp) { + pgprintk("nonpaging_map: ENOMEM\n"); + kvm_release_pfn_clean(walk->pfn); + return -ENOMEM; } - if (table[index] == shadow_trap_nonpresent_pte) { - struct kvm_mmu_page *new_table; - gfn_t pseudo_gfn; - - pseudo_gfn = (v & PT64_DIR_BASE_ADDR_MASK) - >> PAGE_SHIFT; - new_table = kvm_mmu_get_page(vcpu, pseudo_gfn, - v, level - 1, - 1, ACC_ALL, &table[index]); - if (!new_table) { - pgprintk("nonpaging_map: ENOMEM\n"); - kvm_release_pfn_clean(pfn); - return -ENOMEM; - } - - set_shadow_pte(&table[index], - __pa(new_table->spt) - | PT_PRESENT_MASK | PT_WRITABLE_MASK - | shadow_user_mask | shadow_x_mask); - } - table_addr = table[index] & PT64_BASE_ADDR_MASK; + set_shadow_pte(sptep, + __pa(sp->spt) + | PT_PRESENT_MASK | PT_WRITABLE_MASK + | shadow_user_mask | shadow_x_mask); } + return 0; +} + +static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, + int largepage, gfn_t gfn, pfn_t pfn) +{ + int r; + struct direct_shadow_walk walker = { + .walker = { .entry = direct_map_entry, }, + .pfn = pfn, + .largepage = largepage, + .write = write, + .pt_write = 0, + }; + + r = walk_shadow(&walker.walker, vcpu, gfn << PAGE_SHIFT); + if (r < 0) + return r; + return walker.pt_write; } static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn) @@ -1287,16 +1634,14 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn) pfn_t pfn; unsigned long mmu_seq; - down_read(¤t->mm->mmap_sem); if (is_largepage_backed(vcpu, gfn & ~(KVM_PAGES_PER_HPAGE-1))) { gfn &= ~(KVM_PAGES_PER_HPAGE-1); largepage = 1; } mmu_seq = vcpu->kvm->mmu_notifier_seq; - /* implicit mb(), we'll read before PT lock is unlocked */ + smp_rmb(); pfn = gfn_to_pfn(vcpu->kvm, gfn); - up_read(¤t->mm->mmap_sem); /* mmio */ if (is_error_pfn(pfn)) { @@ -1308,8 +1653,7 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn) if (mmu_notifier_retry(vcpu, mmu_seq)) goto out_unlock; kvm_mmu_free_some_pages(vcpu); - r = __direct_map(vcpu, v, write, largepage, gfn, pfn, - PT32E_ROOT_LEVEL); + r = __direct_map(vcpu, v, write, largepage, gfn, pfn); spin_unlock(&vcpu->kvm->mmu_lock); @@ -1405,6 +1749,37 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu) vcpu->arch.mmu.root_hpa = __pa(vcpu->arch.mmu.pae_root); } +static void mmu_sync_roots(struct kvm_vcpu *vcpu) +{ + int i; + struct kvm_mmu_page *sp; + + if (!VALID_PAGE(vcpu->arch.mmu.root_hpa)) + return; + if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) { + hpa_t root = vcpu->arch.mmu.root_hpa; + sp = page_header(root); + mmu_sync_children(vcpu, sp); + return; + } + for (i = 0; i < 4; ++i) { + hpa_t root = vcpu->arch.mmu.pae_root[i]; + + if (root) { + root &= PT64_BASE_ADDR_MASK; + sp = page_header(root); + mmu_sync_children(vcpu, sp); + } + } +} + +void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu) +{ + spin_lock(&vcpu->kvm->mmu_lock); + mmu_sync_roots(vcpu); + spin_unlock(&vcpu->kvm->mmu_lock); +} + static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t vaddr) { return vaddr; @@ -1446,15 +1821,13 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, if (r) return r; - down_read(¤t->mm->mmap_sem); if (is_largepage_backed(vcpu, gfn & ~(KVM_PAGES_PER_HPAGE-1))) { gfn &= ~(KVM_PAGES_PER_HPAGE-1); largepage = 1; } mmu_seq = vcpu->kvm->mmu_notifier_seq; - /* implicit mb(), we'll read before PT lock is unlocked */ + smp_rmb(); pfn = gfn_to_pfn(vcpu->kvm, gfn); - up_read(¤t->mm->mmap_sem); if (is_error_pfn(pfn)) { kvm_release_pfn_clean(pfn); return 1; @@ -1464,7 +1837,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, goto out_unlock; kvm_mmu_free_some_pages(vcpu); r = __direct_map(vcpu, gpa, error_code & PFERR_WRITE_MASK, - largepage, gfn, pfn, kvm_x86_ops->get_tdp_level()); + largepage, gfn, pfn); spin_unlock(&vcpu->kvm->mmu_lock); return r; @@ -1489,6 +1862,8 @@ static int nonpaging_init_context(struct kvm_vcpu *vcpu) context->gva_to_gpa = nonpaging_gva_to_gpa; context->free = nonpaging_free; context->prefetch_page = nonpaging_prefetch_page; + context->sync_page = nonpaging_sync_page; + context->invlpg = nonpaging_invlpg; context->root_level = 0; context->shadow_root_level = PT32E_ROOT_LEVEL; context->root_hpa = INVALID_PAGE; @@ -1536,6 +1911,8 @@ static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level) context->page_fault = paging64_page_fault; context->gva_to_gpa = paging64_gva_to_gpa; context->prefetch_page = paging64_prefetch_page; + context->sync_page = paging64_sync_page; + context->invlpg = paging64_invlpg; context->free = paging_free; context->root_level = level; context->shadow_root_level = level; @@ -1557,6 +1934,8 @@ static int paging32_init_context(struct kvm_vcpu *vcpu) context->gva_to_gpa = paging32_gva_to_gpa; context->free = paging_free; context->prefetch_page = paging32_prefetch_page; + context->sync_page = paging32_sync_page; + context->invlpg = paging32_invlpg; context->root_level = PT32_ROOT_LEVEL; context->shadow_root_level = PT32E_ROOT_LEVEL; context->root_hpa = INVALID_PAGE; @@ -1576,6 +1955,8 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) context->page_fault = tdp_page_fault; context->free = nonpaging_free; context->prefetch_page = nonpaging_prefetch_page; + context->sync_page = nonpaging_sync_page; + context->invlpg = nonpaging_invlpg; context->shadow_root_level = kvm_x86_ops->get_tdp_level(); context->root_hpa = INVALID_PAGE; @@ -1647,6 +2028,7 @@ int kvm_mmu_load(struct kvm_vcpu *vcpu) spin_lock(&vcpu->kvm->mmu_lock); kvm_mmu_free_some_pages(vcpu); mmu_alloc_roots(vcpu); + mmu_sync_roots(vcpu); spin_unlock(&vcpu->kvm->mmu_lock); kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa); kvm_mmu_flush_tlb(vcpu); @@ -1767,15 +2149,13 @@ static void mmu_guess_page_from_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, return; gfn = (gpte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT; - down_read(¤t->mm->mmap_sem); if (is_large_pte(gpte) && is_largepage_backed(vcpu, gfn)) { gfn &= ~(KVM_PAGES_PER_HPAGE-1); vcpu->arch.update_pte.largepage = 1; } vcpu->arch.update_pte.mmu_seq = vcpu->kvm->mmu_notifier_seq; - /* implicit mb(), we'll read before PT lock is unlocked */ + smp_rmb(); pfn = gfn_to_pfn(vcpu->kvm, gfn); - up_read(¤t->mm->mmap_sem); if (is_error_pfn(pfn)) { kvm_release_pfn_clean(pfn); @@ -1837,7 +2217,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, index = kvm_page_table_hashfn(gfn); bucket = &vcpu->kvm->arch.mmu_page_hash[index]; hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) { - if (sp->gfn != gfn || sp->role.metaphysical) + if (sp->gfn != gfn || sp->role.metaphysical || sp->role.invalid) continue; pte_size = sp->role.glevels == PT32_ROOT_LEVEL ? 4 : 8; misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1); @@ -1855,7 +2235,8 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, */ pgprintk("misaligned: gpa %llx bytes %d role %x\n", gpa, bytes, sp->role.word); - kvm_mmu_zap_page(vcpu->kvm, sp); + if (kvm_mmu_zap_page(vcpu->kvm, sp)) + n = bucket->first; ++vcpu->kvm->stat.mmu_flooded; continue; } @@ -1969,6 +2350,16 @@ out: } EXPORT_SYMBOL_GPL(kvm_mmu_page_fault); +void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva) +{ + spin_lock(&vcpu->kvm->mmu_lock); + vcpu->arch.mmu.invlpg(vcpu, gva); + spin_unlock(&vcpu->kvm->mmu_lock); + kvm_mmu_flush_tlb(vcpu); + ++vcpu->stat.invlpg; +} +EXPORT_SYMBOL_GPL(kvm_mmu_invlpg); + void kvm_enable_tdp(void) { tdp_enabled = true; @@ -2055,6 +2446,7 @@ void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot) { struct kvm_mmu_page *sp; + spin_lock(&kvm->mmu_lock); list_for_each_entry(sp, &kvm->arch.active_mmu_pages, link) { int i; u64 *pt; @@ -2068,6 +2460,8 @@ void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot) if (pt[i] & PT_WRITABLE_MASK) pt[i] &= ~PT_WRITABLE_MASK; } + kvm_flush_remote_tlbs(kvm); + spin_unlock(&kvm->mmu_lock); } void kvm_mmu_zap_all(struct kvm *kvm) @@ -2076,7 +2470,9 @@ void kvm_mmu_zap_all(struct kvm *kvm) spin_lock(&kvm->mmu_lock); list_for_each_entry_safe(sp, node, &kvm->arch.active_mmu_pages, link) - kvm_mmu_zap_page(kvm, sp); + if (kvm_mmu_zap_page(kvm, sp)) + node = container_of(kvm->arch.active_mmu_pages.next, + struct kvm_mmu_page, link); spin_unlock(&kvm->mmu_lock); kvm_flush_remote_tlbs(kvm); @@ -2291,18 +2687,18 @@ int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes, gpa_t addr, unsigned long *ret) { int r; - struct kvm_pv_mmu_op_buffer buffer; + struct kvm_pv_mmu_op_buffer *buffer = &vcpu->arch.mmu_op_buffer; - buffer.ptr = buffer.buf; - buffer.len = min_t(unsigned long, bytes, sizeof buffer.buf); - buffer.processed = 0; + buffer->ptr = buffer->buf; + buffer->len = min_t(unsigned long, bytes, sizeof buffer->buf); + buffer->processed = 0; - r = kvm_read_guest(vcpu->kvm, addr, buffer.buf, buffer.len); + r = kvm_read_guest(vcpu->kvm, addr, buffer->buf, buffer->len); if (r) goto out; - while (buffer.len) { - r = kvm_pv_mmu_op_one(vcpu, &buffer); + while (buffer->len) { + r = kvm_pv_mmu_op_one(vcpu, buffer); if (r < 0) goto out; if (r == 0) @@ -2311,7 +2707,7 @@ int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes, r = 1; out: - *ret = buffer.processed; + *ret = buffer->processed; return r; } diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index 4a814bff21f..613ec9aa674 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -25,11 +25,11 @@ #if PTTYPE == 64 #define pt_element_t u64 #define guest_walker guest_walker64 + #define shadow_walker shadow_walker64 #define FNAME(name) paging##64_##name #define PT_BASE_ADDR_MASK PT64_BASE_ADDR_MASK #define PT_DIR_BASE_ADDR_MASK PT64_DIR_BASE_ADDR_MASK #define PT_INDEX(addr, level) PT64_INDEX(addr, level) - #define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level) #define PT_LEVEL_MASK(level) PT64_LEVEL_MASK(level) #define PT_LEVEL_BITS PT64_LEVEL_BITS #ifdef CONFIG_X86_64 @@ -42,11 +42,11 @@ #elif PTTYPE == 32 #define pt_element_t u32 #define guest_walker guest_walker32 + #define shadow_walker shadow_walker32 #define FNAME(name) paging##32_##name #define PT_BASE_ADDR_MASK PT32_BASE_ADDR_MASK #define PT_DIR_BASE_ADDR_MASK PT32_DIR_BASE_ADDR_MASK #define PT_INDEX(addr, level) PT32_INDEX(addr, level) - #define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level) #define PT_LEVEL_MASK(level) PT32_LEVEL_MASK(level) #define PT_LEVEL_BITS PT32_LEVEL_BITS #define PT_MAX_FULL_LEVELS 2 @@ -73,6 +73,17 @@ struct guest_walker { u32 error_code; }; +struct shadow_walker { + struct kvm_shadow_walk walker; + struct guest_walker *guest_walker; + int user_fault; + int write_fault; + int largepage; + int *ptwrite; + pfn_t pfn; + u64 *sptep; +}; + static gfn_t gpte_to_gfn(pt_element_t gpte) { return (gpte & PT_BASE_ADDR_MASK) >> PAGE_SHIFT; @@ -91,14 +102,10 @@ static bool FNAME(cmpxchg_gpte)(struct kvm *kvm, pt_element_t *table; struct page *page; - down_read(¤t->mm->mmap_sem); page = gfn_to_page(kvm, table_gfn); - up_read(¤t->mm->mmap_sem); table = kmap_atomic(page, KM_USER0); - ret = CMPXCHG(&table[index], orig_pte, new_pte); - kunmap_atomic(table, KM_USER0); kvm_release_page_dirty(page); @@ -274,86 +281,89 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page, /* * Fetch a shadow pte for a specific level in the paging hierarchy. */ -static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, - struct guest_walker *walker, - int user_fault, int write_fault, int largepage, - int *ptwrite, pfn_t pfn) +static int FNAME(shadow_walk_entry)(struct kvm_shadow_walk *_sw, + struct kvm_vcpu *vcpu, u64 addr, + u64 *sptep, int level) { - hpa_t shadow_addr; - int level; - u64 *shadow_ent; - unsigned access = walker->pt_access; - - if (!is_present_pte(walker->ptes[walker->level - 1])) - return NULL; - - shadow_addr = vcpu->arch.mmu.root_hpa; - level = vcpu->arch.mmu.shadow_root_level; - if (level == PT32E_ROOT_LEVEL) { - shadow_addr = vcpu->arch.mmu.pae_root[(addr >> 30) & 3]; - shadow_addr &= PT64_BASE_ADDR_MASK; - --level; + struct shadow_walker *sw = + container_of(_sw, struct shadow_walker, walker); + struct guest_walker *gw = sw->guest_walker; + unsigned access = gw->pt_access; + struct kvm_mmu_page *shadow_page; + u64 spte; + int metaphysical; + gfn_t table_gfn; + int r; + pt_element_t curr_pte; + + if (level == PT_PAGE_TABLE_LEVEL + || (sw->largepage && level == PT_DIRECTORY_LEVEL)) { + mmu_set_spte(vcpu, sptep, access, gw->pte_access & access, + sw->user_fault, sw->write_fault, + gw->ptes[gw->level-1] & PT_DIRTY_MASK, + sw->ptwrite, sw->largepage, gw->gfn, sw->pfn, + false); + sw->sptep = sptep; + return 1; } - for (; ; level--) { - u32 index = SHADOW_PT_INDEX(addr, level); - struct kvm_mmu_page *shadow_page; - u64 shadow_pte; - int metaphysical; - gfn_t table_gfn; - - shadow_ent = ((u64 *)__va(shadow_addr)) + index; - if (level == PT_PAGE_TABLE_LEVEL) - break; - - if (largepage && level == PT_DIRECTORY_LEVEL) - break; + if (is_shadow_present_pte(*sptep) && !is_large_pte(*sptep)) + return 0; - if (is_shadow_present_pte(*shadow_ent) - && !is_large_pte(*shadow_ent)) { - shadow_addr = *shadow_ent & PT64_BASE_ADDR_MASK; - continue; - } + if (is_large_pte(*sptep)) { + set_shadow_pte(sptep, shadow_trap_nonpresent_pte); + kvm_flush_remote_tlbs(vcpu->kvm); + rmap_remove(vcpu->kvm, sptep); + } - if (is_large_pte(*shadow_ent)) - rmap_remove(vcpu->kvm, shadow_ent); - - if (level - 1 == PT_PAGE_TABLE_LEVEL - && walker->level == PT_DIRECTORY_LEVEL) { - metaphysical = 1; - if (!is_dirty_pte(walker->ptes[level - 1])) - access &= ~ACC_WRITE_MASK; - table_gfn = gpte_to_gfn(walker->ptes[level - 1]); - } else { - metaphysical = 0; - table_gfn = walker->table_gfn[level - 2]; - } - shadow_page = kvm_mmu_get_page(vcpu, table_gfn, addr, level-1, - metaphysical, access, - shadow_ent); - if (!metaphysical) { - int r; - pt_element_t curr_pte; - r = kvm_read_guest_atomic(vcpu->kvm, - walker->pte_gpa[level - 2], - &curr_pte, sizeof(curr_pte)); - if (r || curr_pte != walker->ptes[level - 2]) { - kvm_release_pfn_clean(pfn); - return NULL; - } + if (level == PT_DIRECTORY_LEVEL && gw->level == PT_DIRECTORY_LEVEL) { + metaphysical = 1; + if (!is_dirty_pte(gw->ptes[level - 1])) + access &= ~ACC_WRITE_MASK; + table_gfn = gpte_to_gfn(gw->ptes[level - 1]); + } else { + metaphysical = 0; + table_gfn = gw->table_gfn[level - 2]; + } + shadow_page = kvm_mmu_get_page(vcpu, table_gfn, (gva_t)addr, level-1, + metaphysical, access, sptep); + if (!metaphysical) { + r = kvm_read_guest_atomic(vcpu->kvm, gw->pte_gpa[level - 2], + &curr_pte, sizeof(curr_pte)); + if (r || curr_pte != gw->ptes[level - 2]) { + kvm_release_pfn_clean(sw->pfn); + sw->sptep = NULL; + return 1; } - shadow_addr = __pa(shadow_page->spt); - shadow_pte = shadow_addr | PT_PRESENT_MASK | PT_ACCESSED_MASK - | PT_WRITABLE_MASK | PT_USER_MASK; - set_shadow_pte(shadow_ent, shadow_pte); } - mmu_set_spte(vcpu, shadow_ent, access, walker->pte_access & access, - user_fault, write_fault, - walker->ptes[walker->level-1] & PT_DIRTY_MASK, - ptwrite, largepage, walker->gfn, pfn, false); + spte = __pa(shadow_page->spt) | PT_PRESENT_MASK | PT_ACCESSED_MASK + | PT_WRITABLE_MASK | PT_USER_MASK; + *sptep = spte; + return 0; +} + +static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, + struct guest_walker *guest_walker, + int user_fault, int write_fault, int largepage, + int *ptwrite, pfn_t pfn) +{ + struct shadow_walker walker = { + .walker = { .entry = FNAME(shadow_walk_entry), }, + .guest_walker = guest_walker, + .user_fault = user_fault, + .write_fault = write_fault, + .largepage = largepage, + .ptwrite = ptwrite, + .pfn = pfn, + }; + + if (!is_present_pte(guest_walker->ptes[guest_walker->level - 1])) + return NULL; + + walk_shadow(&walker.walker, vcpu, addr); - return shadow_ent; + return walker.sptep; } /* @@ -407,7 +417,6 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, return 0; } - down_read(¤t->mm->mmap_sem); if (walker.level == PT_DIRECTORY_LEVEL) { gfn_t large_gfn; large_gfn = walker.gfn & ~(KVM_PAGES_PER_HPAGE-1); @@ -417,9 +426,8 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, } } mmu_seq = vcpu->kvm->mmu_notifier_seq; - /* implicit mb(), we'll read before PT lock is unlocked */ + smp_rmb(); pfn = gfn_to_pfn(vcpu->kvm, walker.gfn); - up_read(¤t->mm->mmap_sem); /* mmio */ if (is_error_pfn(pfn)) { @@ -453,6 +461,31 @@ out_unlock: return 0; } +static int FNAME(shadow_invlpg_entry)(struct kvm_shadow_walk *_sw, + struct kvm_vcpu *vcpu, u64 addr, + u64 *sptep, int level) +{ + + if (level == PT_PAGE_TABLE_LEVEL) { + if (is_shadow_present_pte(*sptep)) + rmap_remove(vcpu->kvm, sptep); + set_shadow_pte(sptep, shadow_trap_nonpresent_pte); + return 1; + } + if (!is_shadow_present_pte(*sptep)) + return 1; + return 0; +} + +static void FNAME(invlpg)(struct kvm_vcpu *vcpu, gva_t gva) +{ + struct shadow_walker walker = { + .walker = { .entry = FNAME(shadow_invlpg_entry), }, + }; + + walk_shadow(&walker.walker, vcpu, gva); +} + static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr) { struct guest_walker walker; @@ -499,12 +532,66 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu, } } +/* + * Using the cached information from sp->gfns is safe because: + * - The spte has a reference to the struct page, so the pfn for a given gfn + * can't change unless all sptes pointing to it are nuked first. + * - Alias changes zap the entire shadow cache. + */ +static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + int i, offset, nr_present; + + offset = nr_present = 0; + + if (PTTYPE == 32) + offset = sp->role.quadrant << PT64_LEVEL_BITS; + + for (i = 0; i < PT64_ENT_PER_PAGE; i++) { + unsigned pte_access; + pt_element_t gpte; + gpa_t pte_gpa; + gfn_t gfn = sp->gfns[i]; + + if (!is_shadow_present_pte(sp->spt[i])) + continue; + + pte_gpa = gfn_to_gpa(sp->gfn); + pte_gpa += (i+offset) * sizeof(pt_element_t); + + if (kvm_read_guest_atomic(vcpu->kvm, pte_gpa, &gpte, + sizeof(pt_element_t))) + return -EINVAL; + + if (gpte_to_gfn(gpte) != gfn || !is_present_pte(gpte) || + !(gpte & PT_ACCESSED_MASK)) { + u64 nonpresent; + + rmap_remove(vcpu->kvm, &sp->spt[i]); + if (is_present_pte(gpte)) + nonpresent = shadow_trap_nonpresent_pte; + else + nonpresent = shadow_notrap_nonpresent_pte; + set_shadow_pte(&sp->spt[i], nonpresent); + continue; + } + + nr_present++; + pte_access = sp->role.access & FNAME(gpte_access)(vcpu, gpte); + set_spte(vcpu, &sp->spt[i], pte_access, 0, 0, + is_dirty_pte(gpte), 0, gfn, + spte_to_pfn(sp->spt[i]), true, false); + } + + return !nr_present; +} + #undef pt_element_t #undef guest_walker +#undef shadow_walker #undef FNAME #undef PT_BASE_ADDR_MASK #undef PT_INDEX -#undef SHADOW_PT_INDEX #undef PT_LEVEL_MASK #undef PT_DIR_BASE_ADDR_MASK #undef PT_LEVEL_BITS diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 8233b86c778..9c4ce657d96 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -18,6 +18,7 @@ #include "kvm_svm.h" #include "irq.h" #include "mmu.h" +#include "kvm_cache_regs.h" #include <linux/module.h> #include <linux/kernel.h> @@ -35,10 +36,6 @@ MODULE_LICENSE("GPL"); #define IOPM_ALLOC_ORDER 2 #define MSRPM_ALLOC_ORDER 1 -#define DB_VECTOR 1 -#define UD_VECTOR 6 -#define GP_VECTOR 13 - #define DR7_GD_MASK (1 << 13) #define DR6_BD_MASK (1 << 13) @@ -47,7 +44,7 @@ MODULE_LICENSE("GPL"); #define SVM_FEATURE_NPT (1 << 0) #define SVM_FEATURE_LBRV (1 << 1) -#define SVM_DEATURE_SVML (1 << 2) +#define SVM_FEATURE_SVML (1 << 2) #define DEBUGCTL_RESERVED_BITS (~(0x3fULL)) @@ -236,13 +233,11 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu) printk(KERN_DEBUG "%s: NOP\n", __func__); return; } - if (svm->next_rip - svm->vmcb->save.rip > MAX_INST_SIZE) - printk(KERN_ERR "%s: ip 0x%llx next 0x%llx\n", - __func__, - svm->vmcb->save.rip, - svm->next_rip); + if (svm->next_rip - kvm_rip_read(vcpu) > MAX_INST_SIZE) + printk(KERN_ERR "%s: ip 0x%lx next 0x%llx\n", + __func__, kvm_rip_read(vcpu), svm->next_rip); - vcpu->arch.rip = svm->vmcb->save.rip = svm->next_rip; + kvm_rip_write(vcpu, svm->next_rip); svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK; vcpu->arch.interrupt_window_open = 1; @@ -530,6 +525,7 @@ static void init_vmcb(struct vcpu_svm *svm) (1ULL << INTERCEPT_CPUID) | (1ULL << INTERCEPT_INVD) | (1ULL << INTERCEPT_HLT) | + (1ULL << INTERCEPT_INVLPG) | (1ULL << INTERCEPT_INVLPGA) | (1ULL << INTERCEPT_IOIO_PROT) | (1ULL << INTERCEPT_MSR_PROT) | @@ -581,6 +577,7 @@ static void init_vmcb(struct vcpu_svm *svm) save->dr7 = 0x400; save->rflags = 2; save->rip = 0x0000fff0; + svm->vcpu.arch.regs[VCPU_REGS_RIP] = save->rip; /* * cr0 val on cpu init should be 0x60000010, we enable cpu @@ -593,7 +590,8 @@ static void init_vmcb(struct vcpu_svm *svm) if (npt_enabled) { /* Setup VMCB for Nested Paging */ control->nested_ctl = 1; - control->intercept &= ~(1ULL << INTERCEPT_TASK_SWITCH); + control->intercept &= ~((1ULL << INTERCEPT_TASK_SWITCH) | + (1ULL << INTERCEPT_INVLPG)); control->intercept_exceptions &= ~(1 << PF_VECTOR); control->intercept_cr_read &= ~(INTERCEPT_CR0_MASK| INTERCEPT_CR3_MASK); @@ -615,10 +613,12 @@ static int svm_vcpu_reset(struct kvm_vcpu *vcpu) init_vmcb(svm); if (vcpu->vcpu_id != 0) { - svm->vmcb->save.rip = 0; + kvm_rip_write(vcpu, 0); svm->vmcb->save.cs.base = svm->vcpu.arch.sipi_vector << 12; svm->vmcb->save.cs.selector = svm->vcpu.arch.sipi_vector << 8; } + vcpu->arch.regs_avail = ~0; + vcpu->arch.regs_dirty = ~0; return 0; } @@ -721,23 +721,6 @@ static void svm_vcpu_put(struct kvm_vcpu *vcpu) rdtscll(vcpu->arch.host_tsc); } -static void svm_cache_regs(struct kvm_vcpu *vcpu) -{ - struct vcpu_svm *svm = to_svm(vcpu); - - vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax; - vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; - vcpu->arch.rip = svm->vmcb->save.rip; -} - -static void svm_decache_regs(struct kvm_vcpu *vcpu) -{ - struct vcpu_svm *svm = to_svm(vcpu); - svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; - svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; - svm->vmcb->save.rip = vcpu->arch.rip; -} - static unsigned long svm_get_rflags(struct kvm_vcpu *vcpu) { return to_svm(vcpu)->vmcb->save.rflags; @@ -1040,7 +1023,7 @@ static int pf_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) if (npt_enabled) svm_flush_tlb(&svm->vcpu); - if (event_injection) + if (!npt_enabled && event_injection) kvm_mmu_unprotect_page_virt(&svm->vcpu, fault_address); return kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code); } @@ -1139,14 +1122,14 @@ static int nop_on_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) static int halt_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { - svm->next_rip = svm->vmcb->save.rip + 1; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 1; skip_emulated_instruction(&svm->vcpu); return kvm_emulate_halt(&svm->vcpu); } static int vmmcall_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { - svm->next_rip = svm->vmcb->save.rip + 3; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; skip_emulated_instruction(&svm->vcpu); kvm_emulate_hypercall(&svm->vcpu); return 1; @@ -1178,11 +1161,18 @@ static int task_switch_interception(struct vcpu_svm *svm, static int cpuid_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { - svm->next_rip = svm->vmcb->save.rip + 2; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 2; kvm_emulate_cpuid(&svm->vcpu); return 1; } +static int invlpg_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) +{ + if (emulate_instruction(&svm->vcpu, kvm_run, 0, 0, 0) != EMULATE_DONE) + pr_unimpl(&svm->vcpu, "%s: failed\n", __func__); + return 1; +} + static int emulate_on_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { @@ -1273,9 +1263,9 @@ static int rdmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) KVMTRACE_3D(MSR_READ, &svm->vcpu, ecx, (u32)data, (u32)(data >> 32), handler); - svm->vmcb->save.rax = data & 0xffffffff; + svm->vcpu.arch.regs[VCPU_REGS_RAX] = data & 0xffffffff; svm->vcpu.arch.regs[VCPU_REGS_RDX] = data >> 32; - svm->next_rip = svm->vmcb->save.rip + 2; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 2; skip_emulated_instruction(&svm->vcpu); } return 1; @@ -1359,13 +1349,13 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data) static int wrmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX]; - u64 data = (svm->vmcb->save.rax & -1u) + u64 data = (svm->vcpu.arch.regs[VCPU_REGS_RAX] & -1u) | ((u64)(svm->vcpu.arch.regs[VCPU_REGS_RDX] & -1u) << 32); KVMTRACE_3D(MSR_WRITE, &svm->vcpu, ecx, (u32)data, (u32)(data >> 32), handler); - svm->next_rip = svm->vmcb->save.rip + 2; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 2; if (svm_set_msr(&svm->vcpu, ecx, data)) kvm_inject_gp(&svm->vcpu, 0); else @@ -1436,7 +1426,7 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm, [SVM_EXIT_CPUID] = cpuid_interception, [SVM_EXIT_INVD] = emulate_on_interception, [SVM_EXIT_HLT] = halt_interception, - [SVM_EXIT_INVLPG] = emulate_on_interception, + [SVM_EXIT_INVLPG] = invlpg_interception, [SVM_EXIT_INVLPGA] = invalid_op_interception, [SVM_EXIT_IOIO] = io_interception, [SVM_EXIT_MSR] = msr_interception, @@ -1538,6 +1528,7 @@ static inline void svm_inject_irq(struct vcpu_svm *svm, int irq) KVMTRACE_1D(INJ_VIRQ, &svm->vcpu, (u32)irq, handler); + ++svm->vcpu.stat.irq_injections; control = &svm->vmcb->control; control->int_vector = irq; control->int_ctl &= ~V_INTR_PRIO_MASK; @@ -1716,6 +1707,12 @@ static inline void sync_lapic_to_cr8(struct kvm_vcpu *vcpu) svm->vmcb->control.int_ctl |= cr8 & V_TPR_MASK; } +#ifdef CONFIG_X86_64 +#define R "r" +#else +#define R "e" +#endif + static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { struct vcpu_svm *svm = to_svm(vcpu); @@ -1723,6 +1720,10 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) u16 gs_selector; u16 ldt_selector; + svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; + svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; + svm->vmcb->save.rip = vcpu->arch.regs[VCPU_REGS_RIP]; + pre_svm_run(svm); sync_lapic_to_cr8(vcpu); @@ -1750,19 +1751,14 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) local_irq_enable(); asm volatile ( + "push %%"R"bp; \n\t" + "mov %c[rbx](%[svm]), %%"R"bx \n\t" + "mov %c[rcx](%[svm]), %%"R"cx \n\t" + "mov %c[rdx](%[svm]), %%"R"dx \n\t" + "mov %c[rsi](%[svm]), %%"R"si \n\t" + "mov %c[rdi](%[svm]), %%"R"di \n\t" + "mov %c[rbp](%[svm]), %%"R"bp \n\t" #ifdef CONFIG_X86_64 - "push %%rbp; \n\t" -#else - "push %%ebp; \n\t" -#endif - -#ifdef CONFIG_X86_64 - "mov %c[rbx](%[svm]), %%rbx \n\t" - "mov %c[rcx](%[svm]), %%rcx \n\t" - "mov %c[rdx](%[svm]), %%rdx \n\t" - "mov %c[rsi](%[svm]), %%rsi \n\t" - "mov %c[rdi](%[svm]), %%rdi \n\t" - "mov %c[rbp](%[svm]), %%rbp \n\t" "mov %c[r8](%[svm]), %%r8 \n\t" "mov %c[r9](%[svm]), %%r9 \n\t" "mov %c[r10](%[svm]), %%r10 \n\t" @@ -1771,41 +1767,24 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) "mov %c[r13](%[svm]), %%r13 \n\t" "mov %c[r14](%[svm]), %%r14 \n\t" "mov %c[r15](%[svm]), %%r15 \n\t" -#else - "mov %c[rbx](%[svm]), %%ebx \n\t" - "mov %c[rcx](%[svm]), %%ecx \n\t" - "mov %c[rdx](%[svm]), %%edx \n\t" - "mov %c[rsi](%[svm]), %%esi \n\t" - "mov %c[rdi](%[svm]), %%edi \n\t" - "mov %c[rbp](%[svm]), %%ebp \n\t" #endif -#ifdef CONFIG_X86_64 - /* Enter guest mode */ - "push %%rax \n\t" - "mov %c[vmcb](%[svm]), %%rax \n\t" - __ex(SVM_VMLOAD) "\n\t" - __ex(SVM_VMRUN) "\n\t" - __ex(SVM_VMSAVE) "\n\t" - "pop %%rax \n\t" -#else /* Enter guest mode */ - "push %%eax \n\t" - "mov %c[vmcb](%[svm]), %%eax \n\t" + "push %%"R"ax \n\t" + "mov %c[vmcb](%[svm]), %%"R"ax \n\t" __ex(SVM_VMLOAD) "\n\t" __ex(SVM_VMRUN) "\n\t" __ex(SVM_VMSAVE) "\n\t" - "pop %%eax \n\t" -#endif + "pop %%"R"ax \n\t" /* Save guest registers, load host registers */ + "mov %%"R"bx, %c[rbx](%[svm]) \n\t" + "mov %%"R"cx, %c[rcx](%[svm]) \n\t" + "mov %%"R"dx, %c[rdx](%[svm]) \n\t" + "mov %%"R"si, %c[rsi](%[svm]) \n\t" + "mov %%"R"di, %c[rdi](%[svm]) \n\t" + "mov %%"R"bp, %c[rbp](%[svm]) \n\t" #ifdef CONFIG_X86_64 - "mov %%rbx, %c[rbx](%[svm]) \n\t" - "mov %%rcx, %c[rcx](%[svm]) \n\t" - "mov %%rdx, %c[rdx](%[svm]) \n\t" - "mov %%rsi, %c[rsi](%[svm]) \n\t" - "mov %%rdi, %c[rdi](%[svm]) \n\t" - "mov %%rbp, %c[rbp](%[svm]) \n\t" "mov %%r8, %c[r8](%[svm]) \n\t" "mov %%r9, %c[r9](%[svm]) \n\t" "mov %%r10, %c[r10](%[svm]) \n\t" @@ -1814,18 +1793,8 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) "mov %%r13, %c[r13](%[svm]) \n\t" "mov %%r14, %c[r14](%[svm]) \n\t" "mov %%r15, %c[r15](%[svm]) \n\t" - - "pop %%rbp; \n\t" -#else - "mov %%ebx, %c[rbx](%[svm]) \n\t" - "mov %%ecx, %c[rcx](%[svm]) \n\t" - "mov %%edx, %c[rdx](%[svm]) \n\t" - "mov %%esi, %c[rsi](%[svm]) \n\t" - "mov %%edi, %c[rdi](%[svm]) \n\t" - "mov %%ebp, %c[rbp](%[svm]) \n\t" - - "pop %%ebp; \n\t" #endif + "pop %%"R"bp" : : [svm]"a"(svm), [vmcb]"i"(offsetof(struct vcpu_svm, vmcb_pa)), @@ -1846,11 +1815,9 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) [r15]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R15])) #endif : "cc", "memory" + , R"bx", R"cx", R"dx", R"si", R"di" #ifdef CONFIG_X86_64 - , "rbx", "rcx", "rdx", "rsi", "rdi" , "r8", "r9", "r10", "r11" , "r12", "r13", "r14", "r15" -#else - , "ebx", "ecx", "edx" , "esi", "edi" #endif ); @@ -1858,6 +1825,9 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) load_db_regs(svm->host_db_regs); vcpu->arch.cr2 = svm->vmcb->save.cr2; + vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax; + vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; + vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip; write_dr6(svm->host_dr6); write_dr7(svm->host_dr7); @@ -1879,6 +1849,8 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) svm->next_rip = 0; } +#undef R + static void svm_set_cr3(struct kvm_vcpu *vcpu, unsigned long root) { struct vcpu_svm *svm = to_svm(vcpu); @@ -1977,8 +1949,6 @@ static struct kvm_x86_ops svm_x86_ops = { .set_gdt = svm_set_gdt, .get_dr = svm_get_dr, .set_dr = svm_set_dr, - .cache_regs = svm_cache_regs, - .decache_regs = svm_decache_regs, .get_rflags = svm_get_rflags, .set_rflags = svm_set_rflags, diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 7041cc52b56..2643b430d83 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -26,6 +26,8 @@ #include <linux/highmem.h> #include <linux/sched.h> #include <linux/moduleparam.h> +#include "kvm_cache_regs.h" +#include "x86.h" #include <asm/io.h> #include <asm/desc.h> @@ -47,6 +49,9 @@ module_param(flexpriority_enabled, bool, 0); static int enable_ept = 1; module_param(enable_ept, bool, 0); +static int emulate_invalid_guest_state = 0; +module_param(emulate_invalid_guest_state, bool, 0); + struct vmcs { u32 revision_id; u32 abort; @@ -56,6 +61,7 @@ struct vmcs { struct vcpu_vmx { struct kvm_vcpu vcpu; struct list_head local_vcpus_link; + unsigned long host_rsp; int launched; u8 fail; u32 idt_vectoring_info; @@ -83,6 +89,7 @@ struct vcpu_vmx { } irq; } rmode; int vpid; + bool emulation_required; }; static inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu) @@ -468,7 +475,7 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu) if (!vcpu->fpu_active) eb |= 1u << NM_VECTOR; if (vcpu->guest_debug.enabled) - eb |= 1u << 1; + eb |= 1u << DB_VECTOR; if (vcpu->arch.rmode.active) eb = ~0; if (vm_need_ept()) @@ -715,9 +722,9 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu) unsigned long rip; u32 interruptibility; - rip = vmcs_readl(GUEST_RIP); + rip = kvm_rip_read(vcpu); rip += vmcs_read32(VM_EXIT_INSTRUCTION_LEN); - vmcs_writel(GUEST_RIP, rip); + kvm_rip_write(vcpu, rip); /* * We emulated an instruction, so temporary interrupt blocking @@ -733,19 +740,35 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu) static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr, bool has_error_code, u32 error_code) { + struct vcpu_vmx *vmx = to_vmx(vcpu); + + if (has_error_code) + vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code); + + if (vcpu->arch.rmode.active) { + vmx->rmode.irq.pending = true; + vmx->rmode.irq.vector = nr; + vmx->rmode.irq.rip = kvm_rip_read(vcpu); + if (nr == BP_VECTOR) + vmx->rmode.irq.rip++; + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, + nr | INTR_TYPE_SOFT_INTR + | (has_error_code ? INTR_INFO_DELIVER_CODE_MASK : 0) + | INTR_INFO_VALID_MASK); + vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1); + kvm_rip_write(vcpu, vmx->rmode.irq.rip - 1); + return; + } + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, nr | INTR_TYPE_EXCEPTION | (has_error_code ? INTR_INFO_DELIVER_CODE_MASK : 0) | INTR_INFO_VALID_MASK); - if (has_error_code) - vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code); } static bool vmx_exception_injected(struct kvm_vcpu *vcpu) { - struct vcpu_vmx *vmx = to_vmx(vcpu); - - return !(vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK); + return false; } /* @@ -947,24 +970,19 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) return ret; } -/* - * Sync the rsp and rip registers into the vcpu structure. This allows - * registers to be accessed by indexing vcpu->arch.regs. - */ -static void vcpu_load_rsp_rip(struct kvm_vcpu *vcpu) -{ - vcpu->arch.regs[VCPU_REGS_RSP] = vmcs_readl(GUEST_RSP); - vcpu->arch.rip = vmcs_readl(GUEST_RIP); -} - -/* - * Syncs rsp and rip back into the vmcs. Should be called after possible - * modification. - */ -static void vcpu_put_rsp_rip(struct kvm_vcpu *vcpu) +static void vmx_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg) { - vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]); - vmcs_writel(GUEST_RIP, vcpu->arch.rip); + __set_bit(reg, (unsigned long *)&vcpu->arch.regs_avail); + switch (reg) { + case VCPU_REGS_RSP: + vcpu->arch.regs[VCPU_REGS_RSP] = vmcs_readl(GUEST_RSP); + break; + case VCPU_REGS_RIP: + vcpu->arch.regs[VCPU_REGS_RIP] = vmcs_readl(GUEST_RIP); + break; + default: + break; + } } static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg) @@ -1007,17 +1025,9 @@ static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg) static int vmx_get_irq(struct kvm_vcpu *vcpu) { - struct vcpu_vmx *vmx = to_vmx(vcpu); - u32 idtv_info_field; - - idtv_info_field = vmx->idt_vectoring_info; - if (idtv_info_field & INTR_INFO_VALID_MASK) { - if (is_external_interrupt(idtv_info_field)) - return idtv_info_field & VECTORING_INFO_VECTOR_MASK; - else - printk(KERN_DEBUG "pending exception: not handled yet\n"); - } - return -1; + if (!vcpu->arch.interrupt.pending) + return -1; + return vcpu->arch.interrupt.nr; } static __init int cpu_has_kvm_support(void) @@ -1031,9 +1041,9 @@ static __init int vmx_disabled_by_bios(void) u64 msr; rdmsrl(MSR_IA32_FEATURE_CONTROL, msr); - return (msr & (MSR_IA32_FEATURE_CONTROL_LOCKED | - MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED)) - == MSR_IA32_FEATURE_CONTROL_LOCKED; + return (msr & (FEATURE_CONTROL_LOCKED | + FEATURE_CONTROL_VMXON_ENABLED)) + == FEATURE_CONTROL_LOCKED; /* locked but not enabled */ } @@ -1045,14 +1055,14 @@ static void hardware_enable(void *garbage) INIT_LIST_HEAD(&per_cpu(vcpus_on_cpu, cpu)); rdmsrl(MSR_IA32_FEATURE_CONTROL, old); - if ((old & (MSR_IA32_FEATURE_CONTROL_LOCKED | - MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED)) - != (MSR_IA32_FEATURE_CONTROL_LOCKED | - MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED)) + if ((old & (FEATURE_CONTROL_LOCKED | + FEATURE_CONTROL_VMXON_ENABLED)) + != (FEATURE_CONTROL_LOCKED | + FEATURE_CONTROL_VMXON_ENABLED)) /* enable and lock */ wrmsrl(MSR_IA32_FEATURE_CONTROL, old | - MSR_IA32_FEATURE_CONTROL_LOCKED | - MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED); + FEATURE_CONTROL_LOCKED | + FEATURE_CONTROL_VMXON_ENABLED); write_cr4(read_cr4() | X86_CR4_VMXE); /* FIXME: not cpu hotplug safe */ asm volatile (ASM_VMX_VMXON_RAX : : "a"(&phys_addr), "m"(phys_addr) @@ -1120,7 +1130,8 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) CPU_BASED_CR3_STORE_EXITING | CPU_BASED_USE_IO_BITMAPS | CPU_BASED_MOV_DR_EXITING | - CPU_BASED_USE_TSC_OFFSETING; + CPU_BASED_USE_TSC_OFFSETING | + CPU_BASED_INVLPG_EXITING; opt = CPU_BASED_TPR_SHADOW | CPU_BASED_USE_MSR_BITMAPS | CPU_BASED_ACTIVATE_SECONDARY_CONTROLS; @@ -1149,9 +1160,11 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) _cpu_based_exec_control &= ~CPU_BASED_TPR_SHADOW; #endif if (_cpu_based_2nd_exec_control & SECONDARY_EXEC_ENABLE_EPT) { - /* CR3 accesses don't need to cause VM Exits when EPT enabled */ + /* CR3 accesses and invlpg don't need to cause VM Exits when EPT + enabled */ min &= ~(CPU_BASED_CR3_LOAD_EXITING | - CPU_BASED_CR3_STORE_EXITING); + CPU_BASED_CR3_STORE_EXITING | + CPU_BASED_INVLPG_EXITING); if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PROCBASED_CTLS, &_cpu_based_exec_control) < 0) return -EIO; @@ -1288,7 +1301,9 @@ static void fix_pmode_dataseg(int seg, struct kvm_save_segment *save) static void enter_pmode(struct kvm_vcpu *vcpu) { unsigned long flags; + struct vcpu_vmx *vmx = to_vmx(vcpu); + vmx->emulation_required = 1; vcpu->arch.rmode.active = 0; vmcs_writel(GUEST_TR_BASE, vcpu->arch.rmode.tr.base); @@ -1305,6 +1320,9 @@ static void enter_pmode(struct kvm_vcpu *vcpu) update_exception_bitmap(vcpu); + if (emulate_invalid_guest_state) + return; + fix_pmode_dataseg(VCPU_SREG_ES, &vcpu->arch.rmode.es); fix_pmode_dataseg(VCPU_SREG_DS, &vcpu->arch.rmode.ds); fix_pmode_dataseg(VCPU_SREG_GS, &vcpu->arch.rmode.gs); @@ -1345,7 +1363,9 @@ static void fix_rmode_seg(int seg, struct kvm_save_segment *save) static void enter_rmode(struct kvm_vcpu *vcpu) { unsigned long flags; + struct vcpu_vmx *vmx = to_vmx(vcpu); + vmx->emulation_required = 1; vcpu->arch.rmode.active = 1; vcpu->arch.rmode.tr.base = vmcs_readl(GUEST_TR_BASE); @@ -1367,6 +1387,9 @@ static void enter_rmode(struct kvm_vcpu *vcpu) vmcs_writel(GUEST_CR4, vmcs_readl(GUEST_CR4) | X86_CR4_VME); update_exception_bitmap(vcpu); + if (emulate_invalid_guest_state) + goto continue_rmode; + vmcs_write16(GUEST_SS_SELECTOR, vmcs_readl(GUEST_SS_BASE) >> 4); vmcs_write32(GUEST_SS_LIMIT, 0xffff); vmcs_write32(GUEST_SS_AR_BYTES, 0xf3); @@ -1382,6 +1405,7 @@ static void enter_rmode(struct kvm_vcpu *vcpu) fix_rmode_seg(VCPU_SREG_GS, &vcpu->arch.rmode.gs); fix_rmode_seg(VCPU_SREG_FS, &vcpu->arch.rmode.fs); +continue_rmode: kvm_mmu_reset_context(vcpu); init_rmode(vcpu->kvm); } @@ -1715,6 +1739,186 @@ static void vmx_set_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) vmcs_writel(GUEST_GDTR_BASE, dt->base); } +static bool rmode_segment_valid(struct kvm_vcpu *vcpu, int seg) +{ + struct kvm_segment var; + u32 ar; + + vmx_get_segment(vcpu, &var, seg); + ar = vmx_segment_access_rights(&var); + + if (var.base != (var.selector << 4)) + return false; + if (var.limit != 0xffff) + return false; + if (ar != 0xf3) + return false; + + return true; +} + +static bool code_segment_valid(struct kvm_vcpu *vcpu) +{ + struct kvm_segment cs; + unsigned int cs_rpl; + + vmx_get_segment(vcpu, &cs, VCPU_SREG_CS); + cs_rpl = cs.selector & SELECTOR_RPL_MASK; + + if (~cs.type & (AR_TYPE_CODE_MASK|AR_TYPE_ACCESSES_MASK)) + return false; + if (!cs.s) + return false; + if (!(~cs.type & (AR_TYPE_CODE_MASK|AR_TYPE_WRITEABLE_MASK))) { + if (cs.dpl > cs_rpl) + return false; + } else if (cs.type & AR_TYPE_CODE_MASK) { + if (cs.dpl != cs_rpl) + return false; + } + if (!cs.present) + return false; + + /* TODO: Add Reserved field check, this'll require a new member in the kvm_segment_field structure */ + return true; +} + +static bool stack_segment_valid(struct kvm_vcpu *vcpu) +{ + struct kvm_segment ss; + unsigned int ss_rpl; + + vmx_get_segment(vcpu, &ss, VCPU_SREG_SS); + ss_rpl = ss.selector & SELECTOR_RPL_MASK; + + if ((ss.type != 3) || (ss.type != 7)) + return false; + if (!ss.s) + return false; + if (ss.dpl != ss_rpl) /* DPL != RPL */ + return false; + if (!ss.present) + return false; + + return true; +} + +static bool data_segment_valid(struct kvm_vcpu *vcpu, int seg) +{ + struct kvm_segment var; + unsigned int rpl; + + vmx_get_segment(vcpu, &var, seg); + rpl = var.selector & SELECTOR_RPL_MASK; + + if (!var.s) + return false; + if (!var.present) + return false; + if (~var.type & (AR_TYPE_CODE_MASK|AR_TYPE_WRITEABLE_MASK)) { + if (var.dpl < rpl) /* DPL < RPL */ + return false; + } + + /* TODO: Add other members to kvm_segment_field to allow checking for other access + * rights flags + */ + return true; +} + +static bool tr_valid(struct kvm_vcpu *vcpu) +{ + struct kvm_segment tr; + + vmx_get_segment(vcpu, &tr, VCPU_SREG_TR); + + if (tr.selector & SELECTOR_TI_MASK) /* TI = 1 */ + return false; + if ((tr.type != 3) || (tr.type != 11)) /* TODO: Check if guest is in IA32e mode */ + return false; + if (!tr.present) + return false; + + return true; +} + +static bool ldtr_valid(struct kvm_vcpu *vcpu) +{ + struct kvm_segment ldtr; + + vmx_get_segment(vcpu, &ldtr, VCPU_SREG_LDTR); + + if (ldtr.selector & SELECTOR_TI_MASK) /* TI = 1 */ + return false; + if (ldtr.type != 2) + return false; + if (!ldtr.present) + return false; + + return true; +} + +static bool cs_ss_rpl_check(struct kvm_vcpu *vcpu) +{ + struct kvm_segment cs, ss; + + vmx_get_segment(vcpu, &cs, VCPU_SREG_CS); + vmx_get_segment(vcpu, &ss, VCPU_SREG_SS); + + return ((cs.selector & SELECTOR_RPL_MASK) == + (ss.selector & SELECTOR_RPL_MASK)); +} + +/* + * Check if guest state is valid. Returns true if valid, false if + * not. + * We assume that registers are always usable + */ +static bool guest_state_valid(struct kvm_vcpu *vcpu) +{ + /* real mode guest state checks */ + if (!(vcpu->arch.cr0 & X86_CR0_PE)) { + if (!rmode_segment_valid(vcpu, VCPU_SREG_CS)) + return false; + if (!rmode_segment_valid(vcpu, VCPU_SREG_SS)) + return false; + if (!rmode_segment_valid(vcpu, VCPU_SREG_DS)) + return false; + if (!rmode_segment_valid(vcpu, VCPU_SREG_ES)) + return false; + if (!rmode_segment_valid(vcpu, VCPU_SREG_FS)) + return false; + if (!rmode_segment_valid(vcpu, VCPU_SREG_GS)) + return false; + } else { + /* protected mode guest state checks */ + if (!cs_ss_rpl_check(vcpu)) + return false; + if (!code_segment_valid(vcpu)) + return false; + if (!stack_segment_valid(vcpu)) + return false; + if (!data_segment_valid(vcpu, VCPU_SREG_DS)) + return false; + if (!data_segment_valid(vcpu, VCPU_SREG_ES)) + return false; + if (!data_segment_valid(vcpu, VCPU_SREG_FS)) + return false; + if (!data_segment_valid(vcpu, VCPU_SREG_GS)) + return false; + if (!tr_valid(vcpu)) + return false; + if (!ldtr_valid(vcpu)) + return false; + } + /* TODO: + * - Add checks on RIP + * - Add checks on RFLAGS + */ + + return true; +} + static int init_rmode_tss(struct kvm *kvm) { gfn_t fn = rmode_tss_base(kvm) >> PAGE_SHIFT; @@ -1726,7 +1930,8 @@ static int init_rmode_tss(struct kvm *kvm) if (r < 0) goto out; data = TSS_BASE_SIZE + TSS_REDIRECTION_SIZE; - r = kvm_write_guest_page(kvm, fn++, &data, 0x66, sizeof(u16)); + r = kvm_write_guest_page(kvm, fn++, &data, + TSS_IOPB_BASE_OFFSET, sizeof(u16)); if (r < 0) goto out; r = kvm_clear_guest_page(kvm, fn++, 0, PAGE_SIZE); @@ -1789,7 +1994,7 @@ static void seg_setup(int seg) vmcs_write16(sf->selector, 0); vmcs_writel(sf->base, 0); vmcs_write32(sf->limit, 0xffff); - vmcs_write32(sf->ar_bytes, 0x93); + vmcs_write32(sf->ar_bytes, 0xf3); } static int alloc_apic_access_page(struct kvm *kvm) @@ -1808,9 +2013,7 @@ static int alloc_apic_access_page(struct kvm *kvm) if (r) goto out; - down_read(¤t->mm->mmap_sem); kvm->arch.apic_access_page = gfn_to_page(kvm, 0xfee00); - up_read(¤t->mm->mmap_sem); out: up_write(&kvm->slots_lock); return r; @@ -1832,10 +2035,8 @@ static int alloc_identity_pagetable(struct kvm *kvm) if (r) goto out; - down_read(¤t->mm->mmap_sem); kvm->arch.ept_identity_pagetable = gfn_to_page(kvm, VMX_EPT_IDENTITY_PAGETABLE_ADDR >> PAGE_SHIFT); - up_read(¤t->mm->mmap_sem); out: up_write(&kvm->slots_lock); return r; @@ -1917,7 +2118,8 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) } if (!vm_need_ept()) exec_control |= CPU_BASED_CR3_STORE_EXITING | - CPU_BASED_CR3_LOAD_EXITING; + CPU_BASED_CR3_LOAD_EXITING | + CPU_BASED_INVLPG_EXITING; vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, exec_control); if (cpu_has_secondary_exec_ctrls()) { @@ -2019,6 +2221,7 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) u64 msr; int ret; + vcpu->arch.regs_avail = ~((1 << VCPU_REGS_RIP) | (1 << VCPU_REGS_RSP)); down_read(&vcpu->kvm->slots_lock); if (!init_rmode(vmx->vcpu.kvm)) { ret = -ENOMEM; @@ -2036,6 +2239,7 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) fx_init(&vmx->vcpu); + seg_setup(VCPU_SREG_CS); /* * GUEST_CS_BASE should really be 0xffff0000, but VT vm86 mode * insists on having GUEST_CS_BASE == GUEST_CS_SELECTOR << 4. Sigh. @@ -2047,8 +2251,6 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) vmcs_write16(GUEST_CS_SELECTOR, vmx->vcpu.arch.sipi_vector << 8); vmcs_writel(GUEST_CS_BASE, vmx->vcpu.arch.sipi_vector << 12); } - vmcs_write32(GUEST_CS_LIMIT, 0xffff); - vmcs_write32(GUEST_CS_AR_BYTES, 0x9b); seg_setup(VCPU_SREG_DS); seg_setup(VCPU_SREG_ES); @@ -2072,10 +2274,10 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) vmcs_writel(GUEST_RFLAGS, 0x02); if (vmx->vcpu.vcpu_id == 0) - vmcs_writel(GUEST_RIP, 0xfff0); + kvm_rip_write(vcpu, 0xfff0); else - vmcs_writel(GUEST_RIP, 0); - vmcs_writel(GUEST_RSP, 0); + kvm_rip_write(vcpu, 0); + kvm_register_write(vcpu, VCPU_REGS_RSP, 0); /* todo: dr0 = dr1 = dr2 = dr3 = 0; dr6 = 0xffff0ff0 */ vmcs_writel(GUEST_DR7, 0x400); @@ -2125,6 +2327,9 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) ret = 0; + /* HACK: Don't enable emulation on guest boot/reset */ + vmx->emulation_required = 0; + out: up_read(&vcpu->kvm->slots_lock); return ret; @@ -2136,14 +2341,15 @@ static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq) KVMTRACE_1D(INJ_VIRQ, vcpu, (u32)irq, handler); + ++vcpu->stat.irq_injections; if (vcpu->arch.rmode.active) { vmx->rmode.irq.pending = true; vmx->rmode.irq.vector = irq; - vmx->rmode.irq.rip = vmcs_readl(GUEST_RIP); + vmx->rmode.irq.rip = kvm_rip_read(vcpu); vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, irq | INTR_TYPE_SOFT_INTR | INTR_INFO_VALID_MASK); vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1); - vmcs_writel(GUEST_RIP, vmx->rmode.irq.rip - 1); + kvm_rip_write(vcpu, vmx->rmode.irq.rip - 1); return; } vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, @@ -2154,7 +2360,6 @@ static void vmx_inject_nmi(struct kvm_vcpu *vcpu) { vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, INTR_TYPE_NMI_INTR | INTR_INFO_VALID_MASK | NMI_VECTOR); - vcpu->arch.nmi_pending = 0; } static void kvm_do_inject_irq(struct kvm_vcpu *vcpu) @@ -2166,7 +2371,7 @@ static void kvm_do_inject_irq(struct kvm_vcpu *vcpu) clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]); if (!vcpu->arch.irq_pending[word_index]) clear_bit(word_index, &vcpu->arch.irq_summary); - vmx_inject_irq(vcpu, irq); + kvm_queue_interrupt(vcpu, irq); } @@ -2180,13 +2385,12 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu, (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0); if (vcpu->arch.interrupt_window_open && - vcpu->arch.irq_summary && - !(vmcs_read32(VM_ENTRY_INTR_INFO_FIELD) & INTR_INFO_VALID_MASK)) - /* - * If interrupts enabled, and not blocked by sti or mov ss. Good. - */ + vcpu->arch.irq_summary && !vcpu->arch.interrupt.pending) kvm_do_inject_irq(vcpu); + if (vcpu->arch.interrupt_window_open && vcpu->arch.interrupt.pending) + vmx_inject_irq(vcpu, vcpu->arch.interrupt.nr); + cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); if (!vcpu->arch.interrupt_window_open && (vcpu->arch.irq_summary || kvm_run->request_interrupt_window)) @@ -2237,9 +2441,6 @@ static void kvm_guest_debug_pre(struct kvm_vcpu *vcpu) static int handle_rmode_exception(struct kvm_vcpu *vcpu, int vec, u32 err_code) { - if (!vcpu->arch.rmode.active) - return 0; - /* * Instruction with address size override prefix opcode 0x67 * Cause the #SS fault with 0 error code in VM86 mode. @@ -2247,6 +2448,25 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu, if (((vec == GP_VECTOR) || (vec == SS_VECTOR)) && err_code == 0) if (emulate_instruction(vcpu, NULL, 0, 0, 0) == EMULATE_DONE) return 1; + /* + * Forward all other exceptions that are valid in real mode. + * FIXME: Breaks guest debugging in real mode, needs to be fixed with + * the required debugging infrastructure rework. + */ + switch (vec) { + case DE_VECTOR: + case DB_VECTOR: + case BP_VECTOR: + case OF_VECTOR: + case BR_VECTOR: + case UD_VECTOR: + case DF_VECTOR: + case SS_VECTOR: + case GP_VECTOR: + case MF_VECTOR: + kvm_queue_exception(vcpu, vec); + return 1; + } return 0; } @@ -2288,7 +2508,7 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } error_code = 0; - rip = vmcs_readl(GUEST_RIP); + rip = kvm_rip_read(vcpu); if (intr_info & INTR_INFO_DELIVER_CODE_MASK) error_code = vmcs_read32(VM_EXIT_INTR_ERROR_CODE); if (is_page_fault(intr_info)) { @@ -2298,7 +2518,7 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) cr2 = vmcs_readl(EXIT_QUALIFICATION); KVMTRACE_3D(PAGE_FAULT, vcpu, error_code, (u32)cr2, (u32)((u64)cr2 >> 32), handler); - if (vect_info & VECTORING_INFO_VALID_MASK) + if (vcpu->arch.interrupt.pending || vcpu->arch.exception.pending) kvm_mmu_unprotect_page_virt(vcpu, cr2); return kvm_mmu_page_fault(vcpu, cr2, error_code); } @@ -2386,27 +2606,25 @@ static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) reg = (exit_qualification >> 8) & 15; switch ((exit_qualification >> 4) & 3) { case 0: /* mov to cr */ - KVMTRACE_3D(CR_WRITE, vcpu, (u32)cr, (u32)vcpu->arch.regs[reg], - (u32)((u64)vcpu->arch.regs[reg] >> 32), handler); + KVMTRACE_3D(CR_WRITE, vcpu, (u32)cr, + (u32)kvm_register_read(vcpu, reg), + (u32)((u64)kvm_register_read(vcpu, reg) >> 32), + handler); switch (cr) { case 0: - vcpu_load_rsp_rip(vcpu); - kvm_set_cr0(vcpu, vcpu->arch.regs[reg]); + kvm_set_cr0(vcpu, kvm_register_read(vcpu, reg)); skip_emulated_instruction(vcpu); return 1; case 3: - vcpu_load_rsp_rip(vcpu); - kvm_set_cr3(vcpu, vcpu->arch.regs[reg]); + kvm_set_cr3(vcpu, kvm_register_read(vcpu, reg)); skip_emulated_instruction(vcpu); return 1; case 4: - vcpu_load_rsp_rip(vcpu); - kvm_set_cr4(vcpu, vcpu->arch.regs[reg]); + kvm_set_cr4(vcpu, kvm_register_read(vcpu, reg)); skip_emulated_instruction(vcpu); return 1; case 8: - vcpu_load_rsp_rip(vcpu); - kvm_set_cr8(vcpu, vcpu->arch.regs[reg]); + kvm_set_cr8(vcpu, kvm_register_read(vcpu, reg)); skip_emulated_instruction(vcpu); if (irqchip_in_kernel(vcpu->kvm)) return 1; @@ -2415,7 +2633,6 @@ static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) }; break; case 2: /* clts */ - vcpu_load_rsp_rip(vcpu); vmx_fpu_deactivate(vcpu); vcpu->arch.cr0 &= ~X86_CR0_TS; vmcs_writel(CR0_READ_SHADOW, vcpu->arch.cr0); @@ -2426,21 +2643,17 @@ static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) case 1: /*mov from cr*/ switch (cr) { case 3: - vcpu_load_rsp_rip(vcpu); - vcpu->arch.regs[reg] = vcpu->arch.cr3; - vcpu_put_rsp_rip(vcpu); + kvm_register_write(vcpu, reg, vcpu->arch.cr3); KVMTRACE_3D(CR_READ, vcpu, (u32)cr, - (u32)vcpu->arch.regs[reg], - (u32)((u64)vcpu->arch.regs[reg] >> 32), + (u32)kvm_register_read(vcpu, reg), + (u32)((u64)kvm_register_read(vcpu, reg) >> 32), handler); skip_emulated_instruction(vcpu); return 1; case 8: - vcpu_load_rsp_rip(vcpu); - vcpu->arch.regs[reg] = kvm_get_cr8(vcpu); - vcpu_put_rsp_rip(vcpu); + kvm_register_write(vcpu, reg, kvm_get_cr8(vcpu)); KVMTRACE_2D(CR_READ, vcpu, (u32)cr, - (u32)vcpu->arch.regs[reg], handler); + (u32)kvm_register_read(vcpu, reg), handler); skip_emulated_instruction(vcpu); return 1; } @@ -2472,7 +2685,6 @@ static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) exit_qualification = vmcs_readl(EXIT_QUALIFICATION); dr = exit_qualification & 7; reg = (exit_qualification >> 8) & 15; - vcpu_load_rsp_rip(vcpu); if (exit_qualification & 16) { /* mov from dr */ switch (dr) { @@ -2485,12 +2697,11 @@ static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) default: val = 0; } - vcpu->arch.regs[reg] = val; + kvm_register_write(vcpu, reg, val); KVMTRACE_2D(DR_READ, vcpu, (u32)dr, (u32)val, handler); } else { /* mov to dr */ } - vcpu_put_rsp_rip(vcpu); skip_emulated_instruction(vcpu); return 1; } @@ -2583,6 +2794,15 @@ static int handle_vmcall(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return 1; } +static int handle_invlpg(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u64 exit_qualification = vmcs_read64(EXIT_QUALIFICATION); + + kvm_mmu_invlpg(vcpu, exit_qualification); + skip_emulated_instruction(vcpu); + return 1; +} + static int handle_wbinvd(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { skip_emulated_instruction(vcpu); @@ -2695,6 +2915,43 @@ static int handle_nmi_window(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return 1; } +static void handle_invalid_guest_state(struct kvm_vcpu *vcpu, + struct kvm_run *kvm_run) +{ + struct vcpu_vmx *vmx = to_vmx(vcpu); + int err; + + preempt_enable(); + local_irq_enable(); + + while (!guest_state_valid(vcpu)) { + err = emulate_instruction(vcpu, kvm_run, 0, 0, 0); + + switch (err) { + case EMULATE_DONE: + break; + case EMULATE_DO_MMIO: + kvm_report_emulation_failure(vcpu, "mmio"); + /* TODO: Handle MMIO */ + return; + default: + kvm_report_emulation_failure(vcpu, "emulation failure"); + return; + } + + if (signal_pending(current)) + break; + if (need_resched()) + schedule(); + } + + local_irq_disable(); + preempt_disable(); + + /* Guest state should be valid now, no more emulation should be needed */ + vmx->emulation_required = 0; +} + /* * The exit handlers return 1 if the exit was handled fully and guest execution * may resume. Otherwise they set the kvm_run parameter to indicate what needs @@ -2714,6 +2971,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu, [EXIT_REASON_MSR_WRITE] = handle_wrmsr, [EXIT_REASON_PENDING_INTERRUPT] = handle_interrupt_window, [EXIT_REASON_HLT] = handle_halt, + [EXIT_REASON_INVLPG] = handle_invlpg, [EXIT_REASON_VMCALL] = handle_vmcall, [EXIT_REASON_TPR_BELOW_THRESHOLD] = handle_tpr_below_threshold, [EXIT_REASON_APIC_ACCESS] = handle_apic_access, @@ -2735,8 +2993,8 @@ static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) struct vcpu_vmx *vmx = to_vmx(vcpu); u32 vectoring_info = vmx->idt_vectoring_info; - KVMTRACE_3D(VMEXIT, vcpu, exit_reason, (u32)vmcs_readl(GUEST_RIP), - (u32)((u64)vmcs_readl(GUEST_RIP) >> 32), entryexit); + KVMTRACE_3D(VMEXIT, vcpu, exit_reason, (u32)kvm_rip_read(vcpu), + (u32)((u64)kvm_rip_read(vcpu) >> 32), entryexit); /* Access CR3 don't cause VMExit in paging mode, so we need * to sync with guest real CR3. */ @@ -2829,88 +3087,92 @@ static void enable_intr_window(struct kvm_vcpu *vcpu) enable_irq_window(vcpu); } -static void vmx_intr_assist(struct kvm_vcpu *vcpu) +static void vmx_complete_interrupts(struct vcpu_vmx *vmx) { - struct vcpu_vmx *vmx = to_vmx(vcpu); - u32 idtv_info_field, intr_info_field, exit_intr_info_field; - int vector; + u32 exit_intr_info; + u32 idt_vectoring_info; + bool unblock_nmi; + u8 vector; + int type; + bool idtv_info_valid; + u32 error; - update_tpr_threshold(vcpu); - - intr_info_field = vmcs_read32(VM_ENTRY_INTR_INFO_FIELD); - exit_intr_info_field = vmcs_read32(VM_EXIT_INTR_INFO); - idtv_info_field = vmx->idt_vectoring_info; - if (intr_info_field & INTR_INFO_VALID_MASK) { - if (idtv_info_field & INTR_INFO_VALID_MASK) { - /* TODO: fault when IDT_Vectoring */ - if (printk_ratelimit()) - printk(KERN_ERR "Fault when IDT_Vectoring\n"); - } - enable_intr_window(vcpu); - return; + exit_intr_info = vmcs_read32(VM_EXIT_INTR_INFO); + if (cpu_has_virtual_nmis()) { + unblock_nmi = (exit_intr_info & INTR_INFO_UNBLOCK_NMI) != 0; + vector = exit_intr_info & INTR_INFO_VECTOR_MASK; + /* + * SDM 3: 25.7.1.2 + * Re-set bit "block by NMI" before VM entry if vmexit caused by + * a guest IRET fault. + */ + if (unblock_nmi && vector != DF_VECTOR) + vmcs_set_bits(GUEST_INTERRUPTIBILITY_INFO, + GUEST_INTR_STATE_NMI); } - if (unlikely(idtv_info_field & INTR_INFO_VALID_MASK)) { - if ((idtv_info_field & VECTORING_INFO_TYPE_MASK) - == INTR_TYPE_EXT_INTR - && vcpu->arch.rmode.active) { - u8 vect = idtv_info_field & VECTORING_INFO_VECTOR_MASK; - - vmx_inject_irq(vcpu, vect); - enable_intr_window(vcpu); - return; - } - - KVMTRACE_1D(REDELIVER_EVT, vcpu, idtv_info_field, handler); + idt_vectoring_info = vmx->idt_vectoring_info; + idtv_info_valid = idt_vectoring_info & VECTORING_INFO_VALID_MASK; + vector = idt_vectoring_info & VECTORING_INFO_VECTOR_MASK; + type = idt_vectoring_info & VECTORING_INFO_TYPE_MASK; + if (vmx->vcpu.arch.nmi_injected) { /* * SDM 3: 25.7.1.2 * Clear bit "block by NMI" before VM entry if a NMI delivery * faulted. */ - if ((idtv_info_field & VECTORING_INFO_TYPE_MASK) - == INTR_TYPE_NMI_INTR && cpu_has_virtual_nmis()) - vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, - vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & - ~GUEST_INTR_STATE_NMI); - - vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field - & ~INTR_INFO_RESVD_BITS_MASK); - vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, - vmcs_read32(VM_EXIT_INSTRUCTION_LEN)); - - if (unlikely(idtv_info_field & INTR_INFO_DELIVER_CODE_MASK)) - vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, - vmcs_read32(IDT_VECTORING_ERROR_CODE)); - enable_intr_window(vcpu); - return; + if (idtv_info_valid && type == INTR_TYPE_NMI_INTR) + vmcs_clear_bits(GUEST_INTERRUPTIBILITY_INFO, + GUEST_INTR_STATE_NMI); + else + vmx->vcpu.arch.nmi_injected = false; + } + kvm_clear_exception_queue(&vmx->vcpu); + if (idtv_info_valid && type == INTR_TYPE_EXCEPTION) { + if (idt_vectoring_info & VECTORING_INFO_DELIVER_CODE_MASK) { + error = vmcs_read32(IDT_VECTORING_ERROR_CODE); + kvm_queue_exception_e(&vmx->vcpu, vector, error); + } else + kvm_queue_exception(&vmx->vcpu, vector); + vmx->idt_vectoring_info = 0; } + kvm_clear_interrupt_queue(&vmx->vcpu); + if (idtv_info_valid && type == INTR_TYPE_EXT_INTR) { + kvm_queue_interrupt(&vmx->vcpu, vector); + vmx->idt_vectoring_info = 0; + } +} + +static void vmx_intr_assist(struct kvm_vcpu *vcpu) +{ + update_tpr_threshold(vcpu); + if (cpu_has_virtual_nmis()) { - /* - * SDM 3: 25.7.1.2 - * Re-set bit "block by NMI" before VM entry if vmexit caused by - * a guest IRET fault. - */ - if ((exit_intr_info_field & INTR_INFO_UNBLOCK_NMI) && - (exit_intr_info_field & INTR_INFO_VECTOR_MASK) != 8) - vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, - vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) | - GUEST_INTR_STATE_NMI); - else if (vcpu->arch.nmi_pending) { - if (vmx_nmi_enabled(vcpu)) - vmx_inject_nmi(vcpu); + if (vcpu->arch.nmi_pending && !vcpu->arch.nmi_injected) { + if (vmx_nmi_enabled(vcpu)) { + vcpu->arch.nmi_pending = false; + vcpu->arch.nmi_injected = true; + } else { + enable_intr_window(vcpu); + return; + } + } + if (vcpu->arch.nmi_injected) { + vmx_inject_nmi(vcpu); enable_intr_window(vcpu); return; } - } - if (!kvm_cpu_has_interrupt(vcpu)) - return; - if (vmx_irq_enabled(vcpu)) { - vector = kvm_cpu_get_interrupt(vcpu); - vmx_inject_irq(vcpu, vector); - kvm_timer_intr_post(vcpu, vector); - } else - enable_irq_window(vcpu); + if (!vcpu->arch.interrupt.pending && kvm_cpu_has_interrupt(vcpu)) { + if (vmx_irq_enabled(vcpu)) + kvm_queue_interrupt(vcpu, kvm_cpu_get_interrupt(vcpu)); + else + enable_irq_window(vcpu); + } + if (vcpu->arch.interrupt.pending) { + vmx_inject_irq(vcpu, vcpu->arch.interrupt.nr); + kvm_timer_intr_post(vcpu, vcpu->arch.interrupt.nr); + } } /* @@ -2922,9 +3184,9 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) static void fixup_rmode_irq(struct vcpu_vmx *vmx) { vmx->rmode.irq.pending = 0; - if (vmcs_readl(GUEST_RIP) + 1 != vmx->rmode.irq.rip) + if (kvm_rip_read(&vmx->vcpu) + 1 != vmx->rmode.irq.rip) return; - vmcs_writel(GUEST_RIP, vmx->rmode.irq.rip); + kvm_rip_write(&vmx->vcpu, vmx->rmode.irq.rip); if (vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK) { vmx->idt_vectoring_info &= ~VECTORING_INFO_TYPE_MASK; vmx->idt_vectoring_info |= INTR_TYPE_EXT_INTR; @@ -2936,11 +3198,30 @@ static void fixup_rmode_irq(struct vcpu_vmx *vmx) | vmx->rmode.irq.vector; } +#ifdef CONFIG_X86_64 +#define R "r" +#define Q "q" +#else +#define R "e" +#define Q "l" +#endif + static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { struct vcpu_vmx *vmx = to_vmx(vcpu); u32 intr_info; + /* Handle invalid guest state instead of entering VMX */ + if (vmx->emulation_required && emulate_invalid_guest_state) { + handle_invalid_guest_state(vcpu, kvm_run); + return; + } + + if (test_bit(VCPU_REGS_RSP, (unsigned long *)&vcpu->arch.regs_dirty)) + vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]); + if (test_bit(VCPU_REGS_RIP, (unsigned long *)&vcpu->arch.regs_dirty)) + vmcs_writel(GUEST_RIP, vcpu->arch.regs[VCPU_REGS_RIP]); + /* * Loading guest fpu may have cleared host cr0.ts */ @@ -2948,26 +3229,25 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) asm( /* Store host registers */ -#ifdef CONFIG_X86_64 - "push %%rdx; push %%rbp;" - "push %%rcx \n\t" -#else - "push %%edx; push %%ebp;" - "push %%ecx \n\t" -#endif + "push %%"R"dx; push %%"R"bp;" + "push %%"R"cx \n\t" + "cmp %%"R"sp, %c[host_rsp](%0) \n\t" + "je 1f \n\t" + "mov %%"R"sp, %c[host_rsp](%0) \n\t" __ex(ASM_VMX_VMWRITE_RSP_RDX) "\n\t" + "1: \n\t" /* Check if vmlaunch of vmresume is needed */ "cmpl $0, %c[launched](%0) \n\t" /* Load guest registers. Don't clobber flags. */ + "mov %c[cr2](%0), %%"R"ax \n\t" + "mov %%"R"ax, %%cr2 \n\t" + "mov %c[rax](%0), %%"R"ax \n\t" + "mov %c[rbx](%0), %%"R"bx \n\t" + "mov %c[rdx](%0), %%"R"dx \n\t" + "mov %c[rsi](%0), %%"R"si \n\t" + "mov %c[rdi](%0), %%"R"di \n\t" + "mov %c[rbp](%0), %%"R"bp \n\t" #ifdef CONFIG_X86_64 - "mov %c[cr2](%0), %%rax \n\t" - "mov %%rax, %%cr2 \n\t" - "mov %c[rax](%0), %%rax \n\t" - "mov %c[rbx](%0), %%rbx \n\t" - "mov %c[rdx](%0), %%rdx \n\t" - "mov %c[rsi](%0), %%rsi \n\t" - "mov %c[rdi](%0), %%rdi \n\t" - "mov %c[rbp](%0), %%rbp \n\t" "mov %c[r8](%0), %%r8 \n\t" "mov %c[r9](%0), %%r9 \n\t" "mov %c[r10](%0), %%r10 \n\t" @@ -2976,18 +3256,9 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) "mov %c[r13](%0), %%r13 \n\t" "mov %c[r14](%0), %%r14 \n\t" "mov %c[r15](%0), %%r15 \n\t" - "mov %c[rcx](%0), %%rcx \n\t" /* kills %0 (rcx) */ -#else - "mov %c[cr2](%0), %%eax \n\t" - "mov %%eax, %%cr2 \n\t" - "mov %c[rax](%0), %%eax \n\t" - "mov %c[rbx](%0), %%ebx \n\t" - "mov %c[rdx](%0), %%edx \n\t" - "mov %c[rsi](%0), %%esi \n\t" - "mov %c[rdi](%0), %%edi \n\t" - "mov %c[rbp](%0), %%ebp \n\t" - "mov %c[rcx](%0), %%ecx \n\t" /* kills %0 (ecx) */ #endif + "mov %c[rcx](%0), %%"R"cx \n\t" /* kills %0 (ecx) */ + /* Enter guest mode */ "jne .Llaunched \n\t" __ex(ASM_VMX_VMLAUNCH) "\n\t" @@ -2995,15 +3266,15 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) ".Llaunched: " __ex(ASM_VMX_VMRESUME) "\n\t" ".Lkvm_vmx_return: " /* Save guest registers, load host registers, keep flags */ + "xchg %0, (%%"R"sp) \n\t" + "mov %%"R"ax, %c[rax](%0) \n\t" + "mov %%"R"bx, %c[rbx](%0) \n\t" + "push"Q" (%%"R"sp); pop"Q" %c[rcx](%0) \n\t" + "mov %%"R"dx, %c[rdx](%0) \n\t" + "mov %%"R"si, %c[rsi](%0) \n\t" + "mov %%"R"di, %c[rdi](%0) \n\t" + "mov %%"R"bp, %c[rbp](%0) \n\t" #ifdef CONFIG_X86_64 - "xchg %0, (%%rsp) \n\t" - "mov %%rax, %c[rax](%0) \n\t" - "mov %%rbx, %c[rbx](%0) \n\t" - "pushq (%%rsp); popq %c[rcx](%0) \n\t" - "mov %%rdx, %c[rdx](%0) \n\t" - "mov %%rsi, %c[rsi](%0) \n\t" - "mov %%rdi, %c[rdi](%0) \n\t" - "mov %%rbp, %c[rbp](%0) \n\t" "mov %%r8, %c[r8](%0) \n\t" "mov %%r9, %c[r9](%0) \n\t" "mov %%r10, %c[r10](%0) \n\t" @@ -3012,28 +3283,16 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) "mov %%r13, %c[r13](%0) \n\t" "mov %%r14, %c[r14](%0) \n\t" "mov %%r15, %c[r15](%0) \n\t" - "mov %%cr2, %%rax \n\t" - "mov %%rax, %c[cr2](%0) \n\t" - - "pop %%rbp; pop %%rbp; pop %%rdx \n\t" -#else - "xchg %0, (%%esp) \n\t" - "mov %%eax, %c[rax](%0) \n\t" - "mov %%ebx, %c[rbx](%0) \n\t" - "pushl (%%esp); popl %c[rcx](%0) \n\t" - "mov %%edx, %c[rdx](%0) \n\t" - "mov %%esi, %c[rsi](%0) \n\t" - "mov %%edi, %c[rdi](%0) \n\t" - "mov %%ebp, %c[rbp](%0) \n\t" - "mov %%cr2, %%eax \n\t" - "mov %%eax, %c[cr2](%0) \n\t" - - "pop %%ebp; pop %%ebp; pop %%edx \n\t" #endif + "mov %%cr2, %%"R"ax \n\t" + "mov %%"R"ax, %c[cr2](%0) \n\t" + + "pop %%"R"bp; pop %%"R"bp; pop %%"R"dx \n\t" "setbe %c[fail](%0) \n\t" : : "c"(vmx), "d"((unsigned long)HOST_RSP), [launched]"i"(offsetof(struct vcpu_vmx, launched)), [fail]"i"(offsetof(struct vcpu_vmx, fail)), + [host_rsp]"i"(offsetof(struct vcpu_vmx, host_rsp)), [rax]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RAX])), [rbx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBX])), [rcx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RCX])), @@ -3053,14 +3312,15 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) #endif [cr2]"i"(offsetof(struct vcpu_vmx, vcpu.arch.cr2)) : "cc", "memory" + , R"bx", R"di", R"si" #ifdef CONFIG_X86_64 - , "rbx", "rdi", "rsi" , "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15" -#else - , "ebx", "edi", "rsi" #endif ); + vcpu->arch.regs_avail = ~((1 << VCPU_REGS_RIP) | (1 << VCPU_REGS_RSP)); + vcpu->arch.regs_dirty = 0; + vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD); if (vmx->rmode.irq.pending) fixup_rmode_irq(vmx); @@ -3080,8 +3340,13 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) KVMTRACE_0D(NMI, vcpu, handler); asm("int $2"); } + + vmx_complete_interrupts(vmx); } +#undef R +#undef Q + static void vmx_free_vmcs(struct kvm_vcpu *vcpu) { struct vcpu_vmx *vmx = to_vmx(vcpu); @@ -3224,8 +3489,7 @@ static struct kvm_x86_ops vmx_x86_ops = { .set_idt = vmx_set_idt, .get_gdt = vmx_get_gdt, .set_gdt = vmx_set_gdt, - .cache_regs = vcpu_load_rsp_rip, - .decache_regs = vcpu_put_rsp_rip, + .cache_reg = vmx_cache_reg, .get_rflags = vmx_get_rflags, .set_rflags = vmx_set_rflags, diff --git a/arch/x86/kvm/vmx.h b/arch/x86/kvm/vmx.h index 17e25995b65..3e010d21fdd 100644 --- a/arch/x86/kvm/vmx.h +++ b/arch/x86/kvm/vmx.h @@ -331,9 +331,6 @@ enum vmcs_field { #define AR_RESERVD_MASK 0xfffe0f00 -#define MSR_IA32_FEATURE_CONTROL_LOCKED 0x1 -#define MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED 0x4 - #define APIC_ACCESS_PAGE_PRIVATE_MEMSLOT 9 #define IDENTITY_PAGETABLE_PRIVATE_MEMSLOT 10 diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 0d682fc6aeb..4f0677d1eae 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4,10 +4,14 @@ * derived from drivers/kvm/kvm_main.c * * Copyright (C) 2006 Qumranet, Inc. + * Copyright (C) 2008 Qumranet, Inc. + * Copyright IBM Corporation, 2008 * * Authors: * Avi Kivity <avi@qumranet.com> * Yaniv Kamay <yaniv@qumranet.com> + * Amit Shah <amit.shah@qumranet.com> + * Ben-Ami Yassour <benami@il.ibm.com> * * This work is licensed under the terms of the GNU GPL, version 2. See * the COPYING file in the top-level directory. @@ -19,14 +23,18 @@ #include "mmu.h" #include "i8254.h" #include "tss.h" +#include "kvm_cache_regs.h" +#include "x86.h" #include <linux/clocksource.h> +#include <linux/interrupt.h> #include <linux/kvm.h> #include <linux/fs.h> #include <linux/vmalloc.h> #include <linux/module.h> #include <linux/mman.h> #include <linux/highmem.h> +#include <linux/intel-iommu.h> #include <asm/uaccess.h> #include <asm/msr.h> @@ -61,6 +69,7 @@ static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, struct kvm_cpuid_entry2 __user *entries); struct kvm_x86_ops *kvm_x86_ops; +EXPORT_SYMBOL_GPL(kvm_x86_ops); struct kvm_stats_debugfs_item debugfs_entries[] = { { "pf_fixed", VCPU_STAT(pf_fixed) }, @@ -83,6 +92,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "fpu_reload", VCPU_STAT(fpu_reload) }, { "insn_emulation", VCPU_STAT(insn_emulation) }, { "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) }, + { "irq_injections", VCPU_STAT(irq_injections) }, { "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) }, { "mmu_pte_write", VM_STAT(mmu_pte_write) }, { "mmu_pte_updated", VM_STAT(mmu_pte_updated) }, @@ -90,12 +100,12 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "mmu_flooded", VM_STAT(mmu_flooded) }, { "mmu_recycled", VM_STAT(mmu_recycled) }, { "mmu_cache_miss", VM_STAT(mmu_cache_miss) }, + { "mmu_unsync", VM_STAT(mmu_unsync) }, { "remote_tlb_flush", VM_STAT(remote_tlb_flush) }, { "largepages", VM_STAT(lpages) }, { NULL } }; - unsigned long segment_base(u16 selector) { struct descriptor_table gdt; @@ -352,6 +362,7 @@ EXPORT_SYMBOL_GPL(kvm_set_cr4); void kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) { if (cr3 == vcpu->arch.cr3 && !pdptrs_changed(vcpu)) { + kvm_mmu_sync_roots(vcpu); kvm_mmu_flush_tlb(vcpu); return; } @@ -564,7 +575,7 @@ static void kvm_set_time_scale(uint32_t tsc_khz, struct pvclock_vcpu_time_info * hv_clock->tsc_to_system_mul = div_frac(nsecs, tps32); pr_debug("%s: tsc_khz %u, tsc_shift %d, tsc_mul %u\n", - __FUNCTION__, tsc_khz, hv_clock->tsc_shift, + __func__, tsc_khz, hv_clock->tsc_shift, hv_clock->tsc_to_system_mul); } @@ -662,6 +673,18 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) pr_unimpl(vcpu, "%s: MSR_IA32_MCG_CTL 0x%llx, nop\n", __func__, data); break; + case MSR_IA32_DEBUGCTLMSR: + if (!data) { + /* We support the non-activated case already */ + break; + } else if (data & ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_BTF)) { + /* Values other than LBR and BTF are vendor-specific, + thus reserved and should throw a #GP */ + return 1; + } + pr_unimpl(vcpu, "%s: MSR_IA32_DEBUGCTLMSR 0x%llx, nop\n", + __func__, data); + break; case MSR_IA32_UCODE_REV: case MSR_IA32_UCODE_WRITE: break; @@ -692,10 +715,8 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) /* ...but clean it before doing the actual write */ vcpu->arch.time_offset = data & ~(PAGE_MASK | 1); - down_read(¤t->mm->mmap_sem); vcpu->arch.time_page = gfn_to_page(vcpu->kvm, data >> PAGE_SHIFT); - up_read(¤t->mm->mmap_sem); if (is_error_page(vcpu->arch.time_page)) { kvm_release_page_clean(vcpu->arch.time_page); @@ -752,8 +773,14 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) case MSR_IA32_MC0_MISC+8: case MSR_IA32_MC0_MISC+12: case MSR_IA32_MC0_MISC+16: + case MSR_IA32_MC0_MISC+20: case MSR_IA32_UCODE_REV: case MSR_IA32_EBL_CR_POWERON: + case MSR_IA32_DEBUGCTLMSR: + case MSR_IA32_LASTBRANCHFROMIP: + case MSR_IA32_LASTBRANCHTOIP: + case MSR_IA32_LASTINTFROMIP: + case MSR_IA32_LASTINTTOIP: data = 0; break; case MSR_MTRRcap: @@ -901,6 +928,9 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_PV_MMU: r = !tdp_enabled; break; + case KVM_CAP_IOMMU: + r = intel_iommu_found(); + break; default: r = 0; break; @@ -1303,28 +1333,33 @@ long kvm_arch_vcpu_ioctl(struct file *filp, struct kvm_vcpu *vcpu = filp->private_data; void __user *argp = (void __user *)arg; int r; + struct kvm_lapic_state *lapic = NULL; switch (ioctl) { case KVM_GET_LAPIC: { - struct kvm_lapic_state lapic; + lapic = kzalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL); - memset(&lapic, 0, sizeof lapic); - r = kvm_vcpu_ioctl_get_lapic(vcpu, &lapic); + r = -ENOMEM; + if (!lapic) + goto out; + r = kvm_vcpu_ioctl_get_lapic(vcpu, lapic); if (r) goto out; r = -EFAULT; - if (copy_to_user(argp, &lapic, sizeof lapic)) + if (copy_to_user(argp, lapic, sizeof(struct kvm_lapic_state))) goto out; r = 0; break; } case KVM_SET_LAPIC: { - struct kvm_lapic_state lapic; - + lapic = kmalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL); + r = -ENOMEM; + if (!lapic) + goto out; r = -EFAULT; - if (copy_from_user(&lapic, argp, sizeof lapic)) + if (copy_from_user(lapic, argp, sizeof(struct kvm_lapic_state))) goto out; - r = kvm_vcpu_ioctl_set_lapic(vcpu, &lapic);; + r = kvm_vcpu_ioctl_set_lapic(vcpu, lapic); if (r) goto out; r = 0; @@ -1422,6 +1457,8 @@ long kvm_arch_vcpu_ioctl(struct file *filp, r = -EINVAL; } out: + if (lapic) + kfree(lapic); return r; } @@ -1630,6 +1667,15 @@ long kvm_arch_vm_ioctl(struct file *filp, struct kvm *kvm = filp->private_data; void __user *argp = (void __user *)arg; int r = -EINVAL; + /* + * This union makes it completely explicit to gcc-3.x + * that these two variables' stack usage should be + * combined, not added together. + */ + union { + struct kvm_pit_state ps; + struct kvm_memory_alias alias; + } u; switch (ioctl) { case KVM_SET_TSS_ADDR: @@ -1661,17 +1707,14 @@ long kvm_arch_vm_ioctl(struct file *filp, case KVM_GET_NR_MMU_PAGES: r = kvm_vm_ioctl_get_nr_mmu_pages(kvm); break; - case KVM_SET_MEMORY_ALIAS: { - struct kvm_memory_alias alias; - + case KVM_SET_MEMORY_ALIAS: r = -EFAULT; - if (copy_from_user(&alias, argp, sizeof alias)) + if (copy_from_user(&u.alias, argp, sizeof(struct kvm_memory_alias))) goto out; - r = kvm_vm_ioctl_set_memory_alias(kvm, &alias); + r = kvm_vm_ioctl_set_memory_alias(kvm, &u.alias); if (r) goto out; break; - } case KVM_CREATE_IRQCHIP: r = -ENOMEM; kvm->arch.vpic = kvm_create_pic(kvm); @@ -1699,13 +1742,7 @@ long kvm_arch_vm_ioctl(struct file *filp, goto out; if (irqchip_in_kernel(kvm)) { mutex_lock(&kvm->lock); - if (irq_event.irq < 16) - kvm_pic_set_irq(pic_irqchip(kvm), - irq_event.irq, - irq_event.level); - kvm_ioapic_set_irq(kvm->arch.vioapic, - irq_event.irq, - irq_event.level); + kvm_set_irq(kvm, irq_event.irq, irq_event.level); mutex_unlock(&kvm->lock); r = 0; } @@ -1713,65 +1750,77 @@ long kvm_arch_vm_ioctl(struct file *filp, } case KVM_GET_IRQCHIP: { /* 0: PIC master, 1: PIC slave, 2: IOAPIC */ - struct kvm_irqchip chip; + struct kvm_irqchip *chip = kmalloc(sizeof(*chip), GFP_KERNEL); - r = -EFAULT; - if (copy_from_user(&chip, argp, sizeof chip)) + r = -ENOMEM; + if (!chip) goto out; + r = -EFAULT; + if (copy_from_user(chip, argp, sizeof *chip)) + goto get_irqchip_out; r = -ENXIO; if (!irqchip_in_kernel(kvm)) - goto out; - r = kvm_vm_ioctl_get_irqchip(kvm, &chip); + goto get_irqchip_out; + r = kvm_vm_ioctl_get_irqchip(kvm, chip); if (r) - goto out; + goto get_irqchip_out; r = -EFAULT; - if (copy_to_user(argp, &chip, sizeof chip)) - goto out; + if (copy_to_user(argp, chip, sizeof *chip)) + goto get_irqchip_out; r = 0; + get_irqchip_out: + kfree(chip); + if (r) + goto out; break; } case KVM_SET_IRQCHIP: { /* 0: PIC master, 1: PIC slave, 2: IOAPIC */ - struct kvm_irqchip chip; + struct kvm_irqchip *chip = kmalloc(sizeof(*chip), GFP_KERNEL); - r = -EFAULT; - if (copy_from_user(&chip, argp, sizeof chip)) + r = -ENOMEM; + if (!chip) goto out; + r = -EFAULT; + if (copy_from_user(chip, argp, sizeof *chip)) + goto set_irqchip_out; r = -ENXIO; if (!irqchip_in_kernel(kvm)) - goto out; - r = kvm_vm_ioctl_set_irqchip(kvm, &chip); + goto set_irqchip_out; + r = kvm_vm_ioctl_set_irqchip(kvm, chip); if (r) - goto out; + goto set_irqchip_out; r = 0; + set_irqchip_out: + kfree(chip); + if (r) + goto out; break; } case KVM_GET_PIT: { - struct kvm_pit_state ps; r = -EFAULT; - if (copy_from_user(&ps, argp, sizeof ps)) + if (copy_from_user(&u.ps, argp, sizeof(struct kvm_pit_state))) goto out; r = -ENXIO; if (!kvm->arch.vpit) goto out; - r = kvm_vm_ioctl_get_pit(kvm, &ps); + r = kvm_vm_ioctl_get_pit(kvm, &u.ps); if (r) goto out; r = -EFAULT; - if (copy_to_user(argp, &ps, sizeof ps)) + if (copy_to_user(argp, &u.ps, sizeof(struct kvm_pit_state))) goto out; r = 0; break; } case KVM_SET_PIT: { - struct kvm_pit_state ps; r = -EFAULT; - if (copy_from_user(&ps, argp, sizeof ps)) + if (copy_from_user(&u.ps, argp, sizeof u.ps)) goto out; r = -ENXIO; if (!kvm->arch.vpit) goto out; - r = kvm_vm_ioctl_set_pit(kvm, &ps); + r = kvm_vm_ioctl_set_pit(kvm, &u.ps); if (r) goto out; r = 0; @@ -2018,9 +2067,7 @@ static int emulator_cmpxchg_emulated(unsigned long addr, val = *(u64 *)new; - down_read(¤t->mm->mmap_sem); page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT); - up_read(¤t->mm->mmap_sem); kaddr = kmap_atomic(page, KM_USER0); set_64bit((u64 *)(kaddr + offset_in_page(gpa)), val); @@ -2040,6 +2087,7 @@ static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg) int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address) { + kvm_mmu_invlpg(vcpu, address); return X86EMUL_CONTINUE; } @@ -2080,7 +2128,7 @@ int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value) void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context) { u8 opcodes[4]; - unsigned long rip = vcpu->arch.rip; + unsigned long rip = kvm_rip_read(vcpu); unsigned long rip_linear; if (!printk_ratelimit()) @@ -2102,6 +2150,14 @@ static struct x86_emulate_ops emulate_ops = { .cmpxchg_emulated = emulator_cmpxchg_emulated, }; +static void cache_all_regs(struct kvm_vcpu *vcpu) +{ + kvm_register_read(vcpu, VCPU_REGS_RAX); + kvm_register_read(vcpu, VCPU_REGS_RSP); + kvm_register_read(vcpu, VCPU_REGS_RIP); + vcpu->arch.regs_dirty = ~0; +} + int emulate_instruction(struct kvm_vcpu *vcpu, struct kvm_run *run, unsigned long cr2, @@ -2111,8 +2167,15 @@ int emulate_instruction(struct kvm_vcpu *vcpu, int r; struct decode_cache *c; + kvm_clear_exception_queue(vcpu); vcpu->arch.mmio_fault_cr2 = cr2; - kvm_x86_ops->cache_regs(vcpu); + /* + * TODO: fix x86_emulate.c to use guest_read/write_register + * instead of direct ->regs accesses, can save hundred cycles + * on Intel for instructions that don't read/change RSP, for + * for example. + */ + cache_all_regs(vcpu); vcpu->mmio_is_write = 0; vcpu->arch.pio.string = 0; @@ -2172,7 +2235,6 @@ int emulate_instruction(struct kvm_vcpu *vcpu, return EMULATE_DO_MMIO; } - kvm_x86_ops->decache_regs(vcpu); kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags); if (vcpu->mmio_is_write) { @@ -2225,20 +2287,19 @@ int complete_pio(struct kvm_vcpu *vcpu) struct kvm_pio_request *io = &vcpu->arch.pio; long delta; int r; - - kvm_x86_ops->cache_regs(vcpu); + unsigned long val; if (!io->string) { - if (io->in) - memcpy(&vcpu->arch.regs[VCPU_REGS_RAX], vcpu->arch.pio_data, - io->size); + if (io->in) { + val = kvm_register_read(vcpu, VCPU_REGS_RAX); + memcpy(&val, vcpu->arch.pio_data, io->size); + kvm_register_write(vcpu, VCPU_REGS_RAX, val); + } } else { if (io->in) { r = pio_copy_data(vcpu); - if (r) { - kvm_x86_ops->cache_regs(vcpu); + if (r) return r; - } } delta = 1; @@ -2248,19 +2309,24 @@ int complete_pio(struct kvm_vcpu *vcpu) * The size of the register should really depend on * current address size. */ - vcpu->arch.regs[VCPU_REGS_RCX] -= delta; + val = kvm_register_read(vcpu, VCPU_REGS_RCX); + val -= delta; + kvm_register_write(vcpu, VCPU_REGS_RCX, val); } if (io->down) delta = -delta; delta *= io->size; - if (io->in) - vcpu->arch.regs[VCPU_REGS_RDI] += delta; - else - vcpu->arch.regs[VCPU_REGS_RSI] += delta; + if (io->in) { + val = kvm_register_read(vcpu, VCPU_REGS_RDI); + val += delta; + kvm_register_write(vcpu, VCPU_REGS_RDI, val); + } else { + val = kvm_register_read(vcpu, VCPU_REGS_RSI); + val += delta; + kvm_register_write(vcpu, VCPU_REGS_RSI, val); + } } - kvm_x86_ops->decache_regs(vcpu); - io->count -= io->cur_count; io->cur_count = 0; @@ -2313,6 +2379,7 @@ int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, int size, unsigned port) { struct kvm_io_device *pio_dev; + unsigned long val; vcpu->run->exit_reason = KVM_EXIT_IO; vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT; @@ -2333,8 +2400,8 @@ int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, KVMTRACE_2D(IO_WRITE, vcpu, vcpu->run->io.port, (u32)size, handler); - kvm_x86_ops->cache_regs(vcpu); - memcpy(vcpu->arch.pio_data, &vcpu->arch.regs[VCPU_REGS_RAX], 4); + val = kvm_register_read(vcpu, VCPU_REGS_RAX); + memcpy(vcpu->arch.pio_data, &val, 4); kvm_x86_ops->skip_emulated_instruction(vcpu); @@ -2492,11 +2559,6 @@ int kvm_emulate_halt(struct kvm_vcpu *vcpu) KVMTRACE_0D(HLT, vcpu, handler); if (irqchip_in_kernel(vcpu->kvm)) { vcpu->arch.mp_state = KVM_MP_STATE_HALTED; - up_read(&vcpu->kvm->slots_lock); - kvm_vcpu_block(vcpu); - down_read(&vcpu->kvm->slots_lock); - if (vcpu->arch.mp_state != KVM_MP_STATE_RUNNABLE) - return -EINTR; return 1; } else { vcpu->run->exit_reason = KVM_EXIT_HLT; @@ -2519,13 +2581,11 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu) unsigned long nr, a0, a1, a2, a3, ret; int r = 1; - kvm_x86_ops->cache_regs(vcpu); - - nr = vcpu->arch.regs[VCPU_REGS_RAX]; - a0 = vcpu->arch.regs[VCPU_REGS_RBX]; - a1 = vcpu->arch.regs[VCPU_REGS_RCX]; - a2 = vcpu->arch.regs[VCPU_REGS_RDX]; - a3 = vcpu->arch.regs[VCPU_REGS_RSI]; + nr = kvm_register_read(vcpu, VCPU_REGS_RAX); + a0 = kvm_register_read(vcpu, VCPU_REGS_RBX); + a1 = kvm_register_read(vcpu, VCPU_REGS_RCX); + a2 = kvm_register_read(vcpu, VCPU_REGS_RDX); + a3 = kvm_register_read(vcpu, VCPU_REGS_RSI); KVMTRACE_1D(VMMCALL, vcpu, (u32)nr, handler); @@ -2548,8 +2608,7 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu) ret = -KVM_ENOSYS; break; } - vcpu->arch.regs[VCPU_REGS_RAX] = ret; - kvm_x86_ops->decache_regs(vcpu); + kvm_register_write(vcpu, VCPU_REGS_RAX, ret); ++vcpu->stat.hypercalls; return r; } @@ -2559,6 +2618,7 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu) { char instruction[3]; int ret = 0; + unsigned long rip = kvm_rip_read(vcpu); /* @@ -2568,9 +2628,8 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu) */ kvm_mmu_zap_all(vcpu->kvm); - kvm_x86_ops->cache_regs(vcpu); kvm_x86_ops->patch_hypercall(vcpu, instruction); - if (emulator_write_emulated(vcpu->arch.rip, instruction, 3, vcpu) + if (emulator_write_emulated(rip, instruction, 3, vcpu) != X86EMUL_CONTINUE) ret = -EFAULT; @@ -2700,13 +2759,12 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu) u32 function, index; struct kvm_cpuid_entry2 *e, *best; - kvm_x86_ops->cache_regs(vcpu); - function = vcpu->arch.regs[VCPU_REGS_RAX]; - index = vcpu->arch.regs[VCPU_REGS_RCX]; - vcpu->arch.regs[VCPU_REGS_RAX] = 0; - vcpu->arch.regs[VCPU_REGS_RBX] = 0; - vcpu->arch.regs[VCPU_REGS_RCX] = 0; - vcpu->arch.regs[VCPU_REGS_RDX] = 0; + function = kvm_register_read(vcpu, VCPU_REGS_RAX); + index = kvm_register_read(vcpu, VCPU_REGS_RCX); + kvm_register_write(vcpu, VCPU_REGS_RAX, 0); + kvm_register_write(vcpu, VCPU_REGS_RBX, 0); + kvm_register_write(vcpu, VCPU_REGS_RCX, 0); + kvm_register_write(vcpu, VCPU_REGS_RDX, 0); best = NULL; for (i = 0; i < vcpu->arch.cpuid_nent; ++i) { e = &vcpu->arch.cpuid_entries[i]; @@ -2724,18 +2782,17 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu) best = e; } if (best) { - vcpu->arch.regs[VCPU_REGS_RAX] = best->eax; - vcpu->arch.regs[VCPU_REGS_RBX] = best->ebx; - vcpu->arch.regs[VCPU_REGS_RCX] = best->ecx; - vcpu->arch.regs[VCPU_REGS_RDX] = best->edx; + kvm_register_write(vcpu, VCPU_REGS_RAX, best->eax); + kvm_register_write(vcpu, VCPU_REGS_RBX, best->ebx); + kvm_register_write(vcpu, VCPU_REGS_RCX, best->ecx); + kvm_register_write(vcpu, VCPU_REGS_RDX, best->edx); } - kvm_x86_ops->decache_regs(vcpu); kvm_x86_ops->skip_emulated_instruction(vcpu); KVMTRACE_5D(CPUID, vcpu, function, - (u32)vcpu->arch.regs[VCPU_REGS_RAX], - (u32)vcpu->arch.regs[VCPU_REGS_RBX], - (u32)vcpu->arch.regs[VCPU_REGS_RCX], - (u32)vcpu->arch.regs[VCPU_REGS_RDX], handler); + (u32)kvm_register_read(vcpu, VCPU_REGS_RAX), + (u32)kvm_register_read(vcpu, VCPU_REGS_RBX), + (u32)kvm_register_read(vcpu, VCPU_REGS_RCX), + (u32)kvm_register_read(vcpu, VCPU_REGS_RDX), handler); } EXPORT_SYMBOL_GPL(kvm_emulate_cpuid); @@ -2776,9 +2833,7 @@ static void vapic_enter(struct kvm_vcpu *vcpu) if (!apic || !apic->vapic_addr) return; - down_read(¤t->mm->mmap_sem); page = gfn_to_page(vcpu->kvm, apic->vapic_addr >> PAGE_SHIFT); - up_read(¤t->mm->mmap_sem); vcpu->arch.apic->vapic_page = page; } @@ -2796,28 +2851,10 @@ static void vapic_exit(struct kvm_vcpu *vcpu) up_read(&vcpu->kvm->slots_lock); } -static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +static int vcpu_enter_guest(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { int r; - if (unlikely(vcpu->arch.mp_state == KVM_MP_STATE_SIPI_RECEIVED)) { - pr_debug("vcpu %d received sipi with vector # %x\n", - vcpu->vcpu_id, vcpu->arch.sipi_vector); - kvm_lapic_reset(vcpu); - r = kvm_x86_ops->vcpu_reset(vcpu); - if (r) - return r; - vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; - } - - down_read(&vcpu->kvm->slots_lock); - vapic_enter(vcpu); - -preempted: - if (vcpu->guest_debug.enabled) - kvm_x86_ops->guest_debug_pre(vcpu); - -again: if (vcpu->requests) if (test_and_clear_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests)) kvm_mmu_unload(vcpu); @@ -2829,6 +2866,8 @@ again: if (vcpu->requests) { if (test_and_clear_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests)) __kvm_migrate_timers(vcpu); + if (test_and_clear_bit(KVM_REQ_MMU_SYNC, &vcpu->requests)) + kvm_mmu_sync_roots(vcpu); if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests)) kvm_x86_ops->tlb_flush(vcpu); if (test_and_clear_bit(KVM_REQ_REPORT_TPR_ACCESS, @@ -2854,21 +2893,15 @@ again: local_irq_disable(); - if (vcpu->requests || need_resched()) { + if (vcpu->requests || need_resched() || signal_pending(current)) { local_irq_enable(); preempt_enable(); r = 1; goto out; } - if (signal_pending(current)) { - local_irq_enable(); - preempt_enable(); - r = -EINTR; - kvm_run->exit_reason = KVM_EXIT_INTR; - ++vcpu->stat.signal_exits; - goto out; - } + if (vcpu->guest_debug.enabled) + kvm_x86_ops->guest_debug_pre(vcpu); vcpu->guest_mode = 1; /* @@ -2917,8 +2950,8 @@ again: * Profile KVM exit RIPs: */ if (unlikely(prof_on == KVM_PROFILING)) { - kvm_x86_ops->cache_regs(vcpu); - profile_hit(KVM_PROFILING, (void *)vcpu->arch.rip); + unsigned long rip = kvm_rip_read(vcpu); + profile_hit(KVM_PROFILING, (void *)rip); } if (vcpu->arch.exception.pending && kvm_x86_ops->exception_injected(vcpu)) @@ -2927,26 +2960,63 @@ again: kvm_lapic_sync_from_vapic(vcpu); r = kvm_x86_ops->handle_exit(kvm_run, vcpu); +out: + return r; +} - if (r > 0) { - if (dm_request_for_irq_injection(vcpu, kvm_run)) { - r = -EINTR; - kvm_run->exit_reason = KVM_EXIT_INTR; - ++vcpu->stat.request_irq_exits; - goto out; - } - if (!need_resched()) - goto again; +static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + int r; + + if (unlikely(vcpu->arch.mp_state == KVM_MP_STATE_SIPI_RECEIVED)) { + pr_debug("vcpu %d received sipi with vector # %x\n", + vcpu->vcpu_id, vcpu->arch.sipi_vector); + kvm_lapic_reset(vcpu); + r = kvm_x86_ops->vcpu_reset(vcpu); + if (r) + return r; + vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; } -out: - up_read(&vcpu->kvm->slots_lock); - if (r > 0) { - kvm_resched(vcpu); - down_read(&vcpu->kvm->slots_lock); - goto preempted; + down_read(&vcpu->kvm->slots_lock); + vapic_enter(vcpu); + + r = 1; + while (r > 0) { + if (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE) + r = vcpu_enter_guest(vcpu, kvm_run); + else { + up_read(&vcpu->kvm->slots_lock); + kvm_vcpu_block(vcpu); + down_read(&vcpu->kvm->slots_lock); + if (test_and_clear_bit(KVM_REQ_UNHALT, &vcpu->requests)) + if (vcpu->arch.mp_state == KVM_MP_STATE_HALTED) + vcpu->arch.mp_state = + KVM_MP_STATE_RUNNABLE; + if (vcpu->arch.mp_state != KVM_MP_STATE_RUNNABLE) + r = -EINTR; + } + + if (r > 0) { + if (dm_request_for_irq_injection(vcpu, kvm_run)) { + r = -EINTR; + kvm_run->exit_reason = KVM_EXIT_INTR; + ++vcpu->stat.request_irq_exits; + } + if (signal_pending(current)) { + r = -EINTR; + kvm_run->exit_reason = KVM_EXIT_INTR; + ++vcpu->stat.signal_exits; + } + if (need_resched()) { + up_read(&vcpu->kvm->slots_lock); + kvm_resched(vcpu); + down_read(&vcpu->kvm->slots_lock); + } + } } + up_read(&vcpu->kvm->slots_lock); post_kvm_run_save(vcpu, kvm_run); vapic_exit(vcpu); @@ -2966,6 +3036,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) if (unlikely(vcpu->arch.mp_state == KVM_MP_STATE_UNINITIALIZED)) { kvm_vcpu_block(vcpu); + clear_bit(KVM_REQ_UNHALT, &vcpu->requests); r = -EAGAIN; goto out; } @@ -2999,11 +3070,9 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } } #endif - if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) { - kvm_x86_ops->cache_regs(vcpu); - vcpu->arch.regs[VCPU_REGS_RAX] = kvm_run->hypercall.ret; - kvm_x86_ops->decache_regs(vcpu); - } + if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) + kvm_register_write(vcpu, VCPU_REGS_RAX, + kvm_run->hypercall.ret); r = __vcpu_run(vcpu, kvm_run); @@ -3019,28 +3088,26 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) { vcpu_load(vcpu); - kvm_x86_ops->cache_regs(vcpu); - - regs->rax = vcpu->arch.regs[VCPU_REGS_RAX]; - regs->rbx = vcpu->arch.regs[VCPU_REGS_RBX]; - regs->rcx = vcpu->arch.regs[VCPU_REGS_RCX]; - regs->rdx = vcpu->arch.regs[VCPU_REGS_RDX]; - regs->rsi = vcpu->arch.regs[VCPU_REGS_RSI]; - regs->rdi = vcpu->arch.regs[VCPU_REGS_RDI]; - regs->rsp = vcpu->arch.regs[VCPU_REGS_RSP]; - regs->rbp = vcpu->arch.regs[VCPU_REGS_RBP]; + regs->rax = kvm_register_read(vcpu, VCPU_REGS_RAX); + regs->rbx = kvm_register_read(vcpu, VCPU_REGS_RBX); + regs->rcx = kvm_register_read(vcpu, VCPU_REGS_RCX); + regs->rdx = kvm_register_read(vcpu, VCPU_REGS_RDX); + regs->rsi = kvm_register_read(vcpu, VCPU_REGS_RSI); + regs->rdi = kvm_register_read(vcpu, VCPU_REGS_RDI); + regs->rsp = kvm_register_read(vcpu, VCPU_REGS_RSP); + regs->rbp = kvm_register_read(vcpu, VCPU_REGS_RBP); #ifdef CONFIG_X86_64 - regs->r8 = vcpu->arch.regs[VCPU_REGS_R8]; - regs->r9 = vcpu->arch.regs[VCPU_REGS_R9]; - regs->r10 = vcpu->arch.regs[VCPU_REGS_R10]; - regs->r11 = vcpu->arch.regs[VCPU_REGS_R11]; - regs->r12 = vcpu->arch.regs[VCPU_REGS_R12]; - regs->r13 = vcpu->arch.regs[VCPU_REGS_R13]; - regs->r14 = vcpu->arch.regs[VCPU_REGS_R14]; - regs->r15 = vcpu->arch.regs[VCPU_REGS_R15]; + regs->r8 = kvm_register_read(vcpu, VCPU_REGS_R8); + regs->r9 = kvm_register_read(vcpu, VCPU_REGS_R9); + regs->r10 = kvm_register_read(vcpu, VCPU_REGS_R10); + regs->r11 = kvm_register_read(vcpu, VCPU_REGS_R11); + regs->r12 = kvm_register_read(vcpu, VCPU_REGS_R12); + regs->r13 = kvm_register_read(vcpu, VCPU_REGS_R13); + regs->r14 = kvm_register_read(vcpu, VCPU_REGS_R14); + regs->r15 = kvm_register_read(vcpu, VCPU_REGS_R15); #endif - regs->rip = vcpu->arch.rip; + regs->rip = kvm_rip_read(vcpu); regs->rflags = kvm_x86_ops->get_rflags(vcpu); /* @@ -3058,29 +3125,29 @@ int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) { vcpu_load(vcpu); - vcpu->arch.regs[VCPU_REGS_RAX] = regs->rax; - vcpu->arch.regs[VCPU_REGS_RBX] = regs->rbx; - vcpu->arch.regs[VCPU_REGS_RCX] = regs->rcx; - vcpu->arch.regs[VCPU_REGS_RDX] = regs->rdx; - vcpu->arch.regs[VCPU_REGS_RSI] = regs->rsi; - vcpu->arch.regs[VCPU_REGS_RDI] = regs->rdi; - vcpu->arch.regs[VCPU_REGS_RSP] = regs->rsp; - vcpu->arch.regs[VCPU_REGS_RBP] = regs->rbp; + kvm_register_write(vcpu, VCPU_REGS_RAX, regs->rax); + kvm_register_write(vcpu, VCPU_REGS_RBX, regs->rbx); + kvm_register_write(vcpu, VCPU_REGS_RCX, regs->rcx); + kvm_register_write(vcpu, VCPU_REGS_RDX, regs->rdx); + kvm_register_write(vcpu, VCPU_REGS_RSI, regs->rsi); + kvm_register_write(vcpu, VCPU_REGS_RDI, regs->rdi); + kvm_register_write(vcpu, VCPU_REGS_RSP, regs->rsp); + kvm_register_write(vcpu, VCPU_REGS_RBP, regs->rbp); #ifdef CONFIG_X86_64 - vcpu->arch.regs[VCPU_REGS_R8] = regs->r8; - vcpu->arch.regs[VCPU_REGS_R9] = regs->r9; - vcpu->arch.regs[VCPU_REGS_R10] = regs->r10; - vcpu->arch.regs[VCPU_REGS_R11] = regs->r11; - vcpu->arch.regs[VCPU_REGS_R12] = regs->r12; - vcpu->arch.regs[VCPU_REGS_R13] = regs->r13; - vcpu->arch.regs[VCPU_REGS_R14] = regs->r14; - vcpu->arch.regs[VCPU_REGS_R15] = regs->r15; + kvm_register_write(vcpu, VCPU_REGS_R8, regs->r8); + kvm_register_write(vcpu, VCPU_REGS_R9, regs->r9); + kvm_register_write(vcpu, VCPU_REGS_R10, regs->r10); + kvm_register_write(vcpu, VCPU_REGS_R11, regs->r11); + kvm_register_write(vcpu, VCPU_REGS_R12, regs->r12); + kvm_register_write(vcpu, VCPU_REGS_R13, regs->r13); + kvm_register_write(vcpu, VCPU_REGS_R14, regs->r14); + kvm_register_write(vcpu, VCPU_REGS_R15, regs->r15); + #endif - vcpu->arch.rip = regs->rip; + kvm_rip_write(vcpu, regs->rip); kvm_x86_ops->set_rflags(vcpu, regs->rflags); - kvm_x86_ops->decache_regs(vcpu); vcpu->arch.exception.pending = false; @@ -3294,11 +3361,33 @@ static int load_segment_descriptor_to_kvm_desct(struct kvm_vcpu *vcpu, return 0; } +static int kvm_load_realmode_segment(struct kvm_vcpu *vcpu, u16 selector, int seg) +{ + struct kvm_segment segvar = { + .base = selector << 4, + .limit = 0xffff, + .selector = selector, + .type = 3, + .present = 1, + .dpl = 3, + .db = 0, + .s = 1, + .l = 0, + .g = 0, + .avl = 0, + .unusable = 0, + }; + kvm_x86_ops->set_segment(vcpu, &segvar, seg); + return 0; +} + int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, int type_bits, int seg) { struct kvm_segment kvm_seg; + if (!(vcpu->arch.cr0 & X86_CR0_PE)) + return kvm_load_realmode_segment(vcpu, selector, seg); if (load_segment_descriptor_to_kvm_desct(vcpu, selector, &kvm_seg)) return 1; kvm_seg.type |= type_bits; @@ -3316,17 +3405,16 @@ static void save_state_to_tss32(struct kvm_vcpu *vcpu, struct tss_segment_32 *tss) { tss->cr3 = vcpu->arch.cr3; - tss->eip = vcpu->arch.rip; + tss->eip = kvm_rip_read(vcpu); tss->eflags = kvm_x86_ops->get_rflags(vcpu); - tss->eax = vcpu->arch.regs[VCPU_REGS_RAX]; - tss->ecx = vcpu->arch.regs[VCPU_REGS_RCX]; - tss->edx = vcpu->arch.regs[VCPU_REGS_RDX]; - tss->ebx = vcpu->arch.regs[VCPU_REGS_RBX]; - tss->esp = vcpu->arch.regs[VCPU_REGS_RSP]; - tss->ebp = vcpu->arch.regs[VCPU_REGS_RBP]; - tss->esi = vcpu->arch.regs[VCPU_REGS_RSI]; - tss->edi = vcpu->arch.regs[VCPU_REGS_RDI]; - + tss->eax = kvm_register_read(vcpu, VCPU_REGS_RAX); + tss->ecx = kvm_register_read(vcpu, VCPU_REGS_RCX); + tss->edx = kvm_register_read(vcpu, VCPU_REGS_RDX); + tss->ebx = kvm_register_read(vcpu, VCPU_REGS_RBX); + tss->esp = kvm_register_read(vcpu, VCPU_REGS_RSP); + tss->ebp = kvm_register_read(vcpu, VCPU_REGS_RBP); + tss->esi = kvm_register_read(vcpu, VCPU_REGS_RSI); + tss->edi = kvm_register_read(vcpu, VCPU_REGS_RDI); tss->es = get_segment_selector(vcpu, VCPU_SREG_ES); tss->cs = get_segment_selector(vcpu, VCPU_SREG_CS); tss->ss = get_segment_selector(vcpu, VCPU_SREG_SS); @@ -3342,17 +3430,17 @@ static int load_state_from_tss32(struct kvm_vcpu *vcpu, { kvm_set_cr3(vcpu, tss->cr3); - vcpu->arch.rip = tss->eip; + kvm_rip_write(vcpu, tss->eip); kvm_x86_ops->set_rflags(vcpu, tss->eflags | 2); - vcpu->arch.regs[VCPU_REGS_RAX] = tss->eax; - vcpu->arch.regs[VCPU_REGS_RCX] = tss->ecx; - vcpu->arch.regs[VCPU_REGS_RDX] = tss->edx; - vcpu->arch.regs[VCPU_REGS_RBX] = tss->ebx; - vcpu->arch.regs[VCPU_REGS_RSP] = tss->esp; - vcpu->arch.regs[VCPU_REGS_RBP] = tss->ebp; - vcpu->arch.regs[VCPU_REGS_RSI] = tss->esi; - vcpu->arch.regs[VCPU_REGS_RDI] = tss->edi; + kvm_register_write(vcpu, VCPU_REGS_RAX, tss->eax); + kvm_register_write(vcpu, VCPU_REGS_RCX, tss->ecx); + kvm_register_write(vcpu, VCPU_REGS_RDX, tss->edx); + kvm_register_write(vcpu, VCPU_REGS_RBX, tss->ebx); + kvm_register_write(vcpu, VCPU_REGS_RSP, tss->esp); + kvm_register_write(vcpu, VCPU_REGS_RBP, tss->ebp); + kvm_register_write(vcpu, VCPU_REGS_RSI, tss->esi); + kvm_register_write(vcpu, VCPU_REGS_RDI, tss->edi); if (kvm_load_segment_descriptor(vcpu, tss->ldt_selector, 0, VCPU_SREG_LDTR)) return 1; @@ -3380,16 +3468,16 @@ static int load_state_from_tss32(struct kvm_vcpu *vcpu, static void save_state_to_tss16(struct kvm_vcpu *vcpu, struct tss_segment_16 *tss) { - tss->ip = vcpu->arch.rip; + tss->ip = kvm_rip_read(vcpu); tss->flag = kvm_x86_ops->get_rflags(vcpu); - tss->ax = vcpu->arch.regs[VCPU_REGS_RAX]; - tss->cx = vcpu->arch.regs[VCPU_REGS_RCX]; - tss->dx = vcpu->arch.regs[VCPU_REGS_RDX]; - tss->bx = vcpu->arch.regs[VCPU_REGS_RBX]; - tss->sp = vcpu->arch.regs[VCPU_REGS_RSP]; - tss->bp = vcpu->arch.regs[VCPU_REGS_RBP]; - tss->si = vcpu->arch.regs[VCPU_REGS_RSI]; - tss->di = vcpu->arch.regs[VCPU_REGS_RDI]; + tss->ax = kvm_register_read(vcpu, VCPU_REGS_RAX); + tss->cx = kvm_register_read(vcpu, VCPU_REGS_RCX); + tss->dx = kvm_register_read(vcpu, VCPU_REGS_RDX); + tss->bx = kvm_register_read(vcpu, VCPU_REGS_RBX); + tss->sp = kvm_register_read(vcpu, VCPU_REGS_RSP); + tss->bp = kvm_register_read(vcpu, VCPU_REGS_RBP); + tss->si = kvm_register_read(vcpu, VCPU_REGS_RSI); + tss->di = kvm_register_read(vcpu, VCPU_REGS_RDI); tss->es = get_segment_selector(vcpu, VCPU_SREG_ES); tss->cs = get_segment_selector(vcpu, VCPU_SREG_CS); @@ -3402,16 +3490,16 @@ static void save_state_to_tss16(struct kvm_vcpu *vcpu, static int load_state_from_tss16(struct kvm_vcpu *vcpu, struct tss_segment_16 *tss) { - vcpu->arch.rip = tss->ip; + kvm_rip_write(vcpu, tss->ip); kvm_x86_ops->set_rflags(vcpu, tss->flag | 2); - vcpu->arch.regs[VCPU_REGS_RAX] = tss->ax; - vcpu->arch.regs[VCPU_REGS_RCX] = tss->cx; - vcpu->arch.regs[VCPU_REGS_RDX] = tss->dx; - vcpu->arch.regs[VCPU_REGS_RBX] = tss->bx; - vcpu->arch.regs[VCPU_REGS_RSP] = tss->sp; - vcpu->arch.regs[VCPU_REGS_RBP] = tss->bp; - vcpu->arch.regs[VCPU_REGS_RSI] = tss->si; - vcpu->arch.regs[VCPU_REGS_RDI] = tss->di; + kvm_register_write(vcpu, VCPU_REGS_RAX, tss->ax); + kvm_register_write(vcpu, VCPU_REGS_RCX, tss->cx); + kvm_register_write(vcpu, VCPU_REGS_RDX, tss->dx); + kvm_register_write(vcpu, VCPU_REGS_RBX, tss->bx); + kvm_register_write(vcpu, VCPU_REGS_RSP, tss->sp); + kvm_register_write(vcpu, VCPU_REGS_RBP, tss->bp); + kvm_register_write(vcpu, VCPU_REGS_RSI, tss->si); + kvm_register_write(vcpu, VCPU_REGS_RDI, tss->di); if (kvm_load_segment_descriptor(vcpu, tss->ldt, 0, VCPU_SREG_LDTR)) return 1; @@ -3534,7 +3622,6 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason) } kvm_x86_ops->skip_emulated_instruction(vcpu); - kvm_x86_ops->cache_regs(vcpu); if (nseg_desc.type & 8) ret = kvm_task_switch_32(vcpu, tss_selector, old_tss_base, @@ -3559,7 +3646,6 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason) tr_seg.type = 11; kvm_set_segment(vcpu, &tr_seg, VCPU_SREG_TR); out: - kvm_x86_ops->decache_regs(vcpu); return ret; } EXPORT_SYMBOL_GPL(kvm_task_switch); @@ -3622,6 +3708,7 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, pr_debug("Set back pending irq %d\n", pending_vec); } + kvm_pic_clear_isr_ack(vcpu->kvm); } kvm_set_segment(vcpu, &sregs->cs, VCPU_SREG_CS); @@ -3634,6 +3721,12 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, kvm_set_segment(vcpu, &sregs->tr, VCPU_SREG_TR); kvm_set_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR); + /* Older userspace won't unhalt the vcpu on reset. */ + if (vcpu->vcpu_id == 0 && kvm_rip_read(vcpu) == 0xfff0 && + sregs->cs.selector == 0xf000 && sregs->cs.base == 0xffff0000 && + !(vcpu->arch.cr0 & X86_CR0_PE)) + vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; + vcpu_put(vcpu); return 0; @@ -3918,6 +4011,7 @@ struct kvm *kvm_arch_create_vm(void) return ERR_PTR(-ENOMEM); INIT_LIST_HEAD(&kvm->arch.active_mmu_pages); + INIT_LIST_HEAD(&kvm->arch.assigned_dev_head); return kvm; } @@ -3950,6 +4044,8 @@ static void kvm_free_vcpus(struct kvm *kvm) void kvm_arch_destroy_vm(struct kvm *kvm) { + kvm_iommu_unmap_guest(kvm); + kvm_free_all_assigned_devices(kvm); kvm_free_pit(kvm); kfree(kvm->arch.vpic); kfree(kvm->arch.vioapic); @@ -3981,7 +4077,7 @@ int kvm_arch_set_memory_region(struct kvm *kvm, userspace_addr = do_mmap(NULL, 0, npages * PAGE_SIZE, PROT_READ | PROT_WRITE, - MAP_SHARED | MAP_ANONYMOUS, + MAP_PRIVATE | MAP_ANONYMOUS, 0); up_write(¤t->mm->mmap_sem); diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h new file mode 100644 index 00000000000..6a4be78a738 --- /dev/null +++ b/arch/x86/kvm/x86.h @@ -0,0 +1,22 @@ +#ifndef ARCH_X86_KVM_X86_H +#define ARCH_X86_KVM_X86_H + +#include <linux/kvm_host.h> + +static inline void kvm_clear_exception_queue(struct kvm_vcpu *vcpu) +{ + vcpu->arch.exception.pending = false; +} + +static inline void kvm_queue_interrupt(struct kvm_vcpu *vcpu, u8 vector) +{ + vcpu->arch.interrupt.pending = true; + vcpu->arch.interrupt.nr = vector; +} + +static inline void kvm_clear_interrupt_queue(struct kvm_vcpu *vcpu) +{ + vcpu->arch.interrupt.pending = false; +} + +#endif diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c index f2f90468f8b..ea051173b0d 100644 --- a/arch/x86/kvm/x86_emulate.c +++ b/arch/x86/kvm/x86_emulate.c @@ -26,6 +26,7 @@ #define DPRINTF(_f, _a ...) printf(_f , ## _a) #else #include <linux/kvm_host.h> +#include "kvm_cache_regs.h" #define DPRINTF(x...) do {} while (0) #endif #include <linux/module.h> @@ -46,25 +47,26 @@ #define ImplicitOps (1<<1) /* Implicit in opcode. No generic decode. */ #define DstReg (2<<1) /* Register operand. */ #define DstMem (3<<1) /* Memory operand. */ -#define DstMask (3<<1) +#define DstAcc (4<<1) /* Destination Accumulator */ +#define DstMask (7<<1) /* Source operand type. */ -#define SrcNone (0<<3) /* No source operand. */ -#define SrcImplicit (0<<3) /* Source operand is implicit in the opcode. */ -#define SrcReg (1<<3) /* Register operand. */ -#define SrcMem (2<<3) /* Memory operand. */ -#define SrcMem16 (3<<3) /* Memory operand (16-bit). */ -#define SrcMem32 (4<<3) /* Memory operand (32-bit). */ -#define SrcImm (5<<3) /* Immediate operand. */ -#define SrcImmByte (6<<3) /* 8-bit sign-extended immediate operand. */ -#define SrcMask (7<<3) +#define SrcNone (0<<4) /* No source operand. */ +#define SrcImplicit (0<<4) /* Source operand is implicit in the opcode. */ +#define SrcReg (1<<4) /* Register operand. */ +#define SrcMem (2<<4) /* Memory operand. */ +#define SrcMem16 (3<<4) /* Memory operand (16-bit). */ +#define SrcMem32 (4<<4) /* Memory operand (32-bit). */ +#define SrcImm (5<<4) /* Immediate operand. */ +#define SrcImmByte (6<<4) /* 8-bit sign-extended immediate operand. */ +#define SrcMask (7<<4) /* Generic ModRM decode. */ -#define ModRM (1<<6) +#define ModRM (1<<7) /* Destination is only written; never read. */ -#define Mov (1<<7) -#define BitOp (1<<8) -#define MemAbs (1<<9) /* Memory operand is absolute displacement */ -#define String (1<<10) /* String instruction (rep capable) */ -#define Stack (1<<11) /* Stack instruction (push/pop) */ +#define Mov (1<<8) +#define BitOp (1<<9) +#define MemAbs (1<<10) /* Memory operand is absolute displacement */ +#define String (1<<12) /* String instruction (rep capable) */ +#define Stack (1<<13) /* Stack instruction (push/pop) */ #define Group (1<<14) /* Bits 3:5 of modrm byte extend opcode */ #define GroupDual (1<<15) /* Alternate decoding of mod == 3 */ #define GroupMask 0xff /* Group number stored in bits 0:7 */ @@ -94,7 +96,7 @@ static u16 opcode_table[256] = { /* 0x20 - 0x27 */ ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, - SrcImmByte, SrcImm, 0, 0, + DstAcc | SrcImmByte, DstAcc | SrcImm, 0, 0, /* 0x28 - 0x2F */ ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, @@ -106,7 +108,8 @@ static u16 opcode_table[256] = { /* 0x38 - 0x3F */ ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, - 0, 0, 0, 0, + ByteOp | DstAcc | SrcImm, DstAcc | SrcImm, + 0, 0, /* 0x40 - 0x47 */ DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, /* 0x48 - 0x4F */ @@ -153,9 +156,16 @@ static u16 opcode_table[256] = { 0, 0, ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String, ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String, ByteOp | ImplicitOps | String, ImplicitOps | String, - /* 0xB0 - 0xBF */ - 0, 0, 0, 0, 0, 0, 0, 0, - DstReg | SrcImm | Mov, 0, 0, 0, 0, 0, 0, 0, + /* 0xB0 - 0xB7 */ + ByteOp | DstReg | SrcImm | Mov, ByteOp | DstReg | SrcImm | Mov, + ByteOp | DstReg | SrcImm | Mov, ByteOp | DstReg | SrcImm | Mov, + ByteOp | DstReg | SrcImm | Mov, ByteOp | DstReg | SrcImm | Mov, + ByteOp | DstReg | SrcImm | Mov, ByteOp | DstReg | SrcImm | Mov, + /* 0xB8 - 0xBF */ + DstReg | SrcImm | Mov, DstReg | SrcImm | Mov, + DstReg | SrcImm | Mov, DstReg | SrcImm | Mov, + DstReg | SrcImm | Mov, DstReg | SrcImm | Mov, + DstReg | SrcImm | Mov, DstReg | SrcImm | Mov, /* 0xC0 - 0xC7 */ ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImmByte | ModRM, 0, ImplicitOps | Stack, 0, 0, @@ -169,17 +179,20 @@ static u16 opcode_table[256] = { /* 0xD8 - 0xDF */ 0, 0, 0, 0, 0, 0, 0, 0, /* 0xE0 - 0xE7 */ - 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, + SrcNone | ByteOp | ImplicitOps, SrcNone | ImplicitOps, + SrcNone | ByteOp | ImplicitOps, SrcNone | ImplicitOps, /* 0xE8 - 0xEF */ ImplicitOps | Stack, SrcImm | ImplicitOps, ImplicitOps, SrcImmByte | ImplicitOps, - 0, 0, 0, 0, + SrcNone | ByteOp | ImplicitOps, SrcNone | ImplicitOps, + SrcNone | ByteOp | ImplicitOps, SrcNone | ImplicitOps, /* 0xF0 - 0xF7 */ 0, 0, 0, 0, ImplicitOps, ImplicitOps, Group | Group3_Byte, Group | Group3, /* 0xF8 - 0xFF */ ImplicitOps, 0, ImplicitOps, ImplicitOps, - 0, 0, Group | Group4, Group | Group5, + ImplicitOps, ImplicitOps, Group | Group4, Group | Group5, }; static u16 twobyte_table[256] = { @@ -268,15 +281,16 @@ static u16 group_table[] = { ByteOp | DstMem | SrcNone | ModRM, ByteOp | DstMem | SrcNone | ModRM, 0, 0, 0, 0, [Group3*8] = - DstMem | SrcImm | ModRM | SrcImm, 0, - DstMem | SrcNone | ModRM, ByteOp | DstMem | SrcNone | ModRM, + DstMem | SrcImm | ModRM, 0, + DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM, 0, 0, 0, 0, [Group4*8] = ByteOp | DstMem | SrcNone | ModRM, ByteOp | DstMem | SrcNone | ModRM, 0, 0, 0, 0, 0, 0, [Group5*8] = - DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM, 0, 0, - SrcMem | ModRM, 0, SrcMem | ModRM | Stack, 0, + DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM, + SrcMem | ModRM | Stack, 0, + SrcMem | ModRM | Stack, 0, SrcMem | ModRM | Stack, 0, [Group7*8] = 0, 0, ModRM | SrcMem, ModRM | SrcMem, SrcNone | ModRM | DstMem | Mov, 0, @@ -839,7 +853,7 @@ x86_decode_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) /* Shadow copy of register state. Committed on successful emulation. */ memset(c, 0, sizeof(struct decode_cache)); - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); ctxt->cs_base = seg_base(ctxt, VCPU_SREG_CS); memcpy(c->regs, ctxt->vcpu->arch.regs, sizeof c->regs); @@ -1048,6 +1062,23 @@ done_prefixes: } c->dst.type = OP_MEM; break; + case DstAcc: + c->dst.type = OP_REG; + c->dst.bytes = c->op_bytes; + c->dst.ptr = &c->regs[VCPU_REGS_RAX]; + switch (c->op_bytes) { + case 1: + c->dst.val = *(u8 *)c->dst.ptr; + break; + case 2: + c->dst.val = *(u16 *)c->dst.ptr; + break; + case 4: + c->dst.val = *(u32 *)c->dst.ptr; + break; + } + c->dst.orig_val = c->dst.val; + break; } if (c->rip_relative) @@ -1151,6 +1182,14 @@ static inline int emulate_grp45(struct x86_emulate_ctxt *ctxt, case 1: /* dec */ emulate_1op("dec", c->dst, ctxt->eflags); break; + case 2: /* call near abs */ { + long int old_eip; + old_eip = c->eip; + c->eip = c->src.val; + c->src.val = old_eip; + emulate_push(ctxt); + break; + } case 4: /* jmp abs */ c->eip = c->src.val; break; @@ -1251,6 +1290,8 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) u64 msr_data; unsigned long saved_eip = 0; struct decode_cache *c = &ctxt->decode; + unsigned int port; + int io_dir_in; int rc = 0; /* Shadow copy of register state. Committed on successful emulation. @@ -1267,7 +1308,7 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) if (c->rep_prefix && (c->d & String)) { /* All REP prefixes have the same first termination condition */ if (c->regs[VCPU_REGS_RCX] == 0) { - ctxt->vcpu->arch.rip = c->eip; + kvm_rip_write(ctxt->vcpu, c->eip); goto done; } /* The second termination condition only applies for REPE @@ -1281,17 +1322,17 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) (c->b == 0xae) || (c->b == 0xaf)) { if ((c->rep_prefix == REPE_PREFIX) && ((ctxt->eflags & EFLG_ZF) == 0)) { - ctxt->vcpu->arch.rip = c->eip; + kvm_rip_write(ctxt->vcpu, c->eip); goto done; } if ((c->rep_prefix == REPNE_PREFIX) && ((ctxt->eflags & EFLG_ZF) == EFLG_ZF)) { - ctxt->vcpu->arch.rip = c->eip; + kvm_rip_write(ctxt->vcpu, c->eip); goto done; } } c->regs[VCPU_REGS_RCX]--; - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); } if (c->src.type == OP_MEM) { @@ -1351,27 +1392,10 @@ special_insn: sbb: /* sbb */ emulate_2op_SrcV("sbb", c->src, c->dst, ctxt->eflags); break; - case 0x20 ... 0x23: + case 0x20 ... 0x25: and: /* and */ emulate_2op_SrcV("and", c->src, c->dst, ctxt->eflags); break; - case 0x24: /* and al imm8 */ - c->dst.type = OP_REG; - c->dst.ptr = &c->regs[VCPU_REGS_RAX]; - c->dst.val = *(u8 *)c->dst.ptr; - c->dst.bytes = 1; - c->dst.orig_val = c->dst.val; - goto and; - case 0x25: /* and ax imm16, or eax imm32 */ - c->dst.type = OP_REG; - c->dst.bytes = c->op_bytes; - c->dst.ptr = &c->regs[VCPU_REGS_RAX]; - if (c->op_bytes == 2) - c->dst.val = *(u16 *)c->dst.ptr; - else - c->dst.val = *(u32 *)c->dst.ptr; - c->dst.orig_val = c->dst.val; - goto and; case 0x28 ... 0x2d: sub: /* sub */ emulate_2op_SrcV("sub", c->src, c->dst, ctxt->eflags); @@ -1659,7 +1683,7 @@ special_insn: case 0xae ... 0xaf: /* scas */ DPRINTF("Urk! I don't handle SCAS.\n"); goto cannot_emulate; - case 0xb8: /* mov r, imm */ + case 0xb0 ... 0xbf: /* mov r, imm */ goto mov; case 0xc0 ... 0xc1: emulate_grp2(ctxt); @@ -1679,6 +1703,16 @@ special_insn: c->src.val = c->regs[VCPU_REGS_RCX]; emulate_grp2(ctxt); break; + case 0xe4: /* inb */ + case 0xe5: /* in */ + port = insn_fetch(u8, 1, c->eip); + io_dir_in = 1; + goto do_io; + case 0xe6: /* outb */ + case 0xe7: /* out */ + port = insn_fetch(u8, 1, c->eip); + io_dir_in = 0; + goto do_io; case 0xe8: /* call (near) */ { long int rel; switch (c->op_bytes) { @@ -1729,6 +1763,22 @@ special_insn: jmp_rel(c, c->src.val); c->dst.type = OP_NONE; /* Disable writeback. */ break; + case 0xec: /* in al,dx */ + case 0xed: /* in (e/r)ax,dx */ + port = c->regs[VCPU_REGS_RDX]; + io_dir_in = 1; + goto do_io; + case 0xee: /* out al,dx */ + case 0xef: /* out (e/r)ax,dx */ + port = c->regs[VCPU_REGS_RDX]; + io_dir_in = 0; + do_io: if (kvm_emulate_pio(ctxt->vcpu, NULL, io_dir_in, + (c->d & ByteOp) ? 1 : c->op_bytes, + port) != 0) { + c->eip = saved_eip; + goto cannot_emulate; + } + return 0; case 0xf4: /* hlt */ ctxt->vcpu->arch.halt_request = 1; break; @@ -1754,6 +1804,14 @@ special_insn: ctxt->eflags |= X86_EFLAGS_IF; c->dst.type = OP_NONE; /* Disable writeback. */ break; + case 0xfc: /* cld */ + ctxt->eflags &= ~EFLG_DF; + c->dst.type = OP_NONE; /* Disable writeback. */ + break; + case 0xfd: /* std */ + ctxt->eflags |= EFLG_DF; + c->dst.type = OP_NONE; /* Disable writeback. */ + break; case 0xfe ... 0xff: /* Grp4/Grp5 */ rc = emulate_grp45(ctxt, ops); if (rc != 0) @@ -1768,7 +1826,7 @@ writeback: /* Commit shadow register state. */ memcpy(ctxt->vcpu->arch.regs, c->regs, sizeof c->regs); - ctxt->vcpu->arch.rip = c->eip; + kvm_rip_write(ctxt->vcpu, c->eip); done: if (rc == X86EMUL_UNHANDLEABLE) { @@ -1793,7 +1851,7 @@ twobyte_insn: goto done; /* Let the processor re-execute the fixed hypercall */ - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); /* Disable writeback. */ c->dst.type = OP_NONE; break; @@ -1889,7 +1947,7 @@ twobyte_insn: rc = kvm_set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data); if (rc) { kvm_inject_gp(ctxt->vcpu, 0); - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); } rc = X86EMUL_CONTINUE; c->dst.type = OP_NONE; @@ -1899,7 +1957,7 @@ twobyte_insn: rc = kvm_get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data); if (rc) { kvm_inject_gp(ctxt->vcpu, 0); - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); } else { c->regs[VCPU_REGS_RAX] = (u32)msr_data; c->regs[VCPU_REGS_RDX] = msr_data >> 32; diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index 65f0b8a47be..48ee4f9435f 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -582,7 +582,7 @@ static void __init lguest_init_IRQ(void) for (i = 0; i < LGUEST_IRQS; i++) { int vector = FIRST_EXTERNAL_VECTOR + i; if (vector != SYSCALL_VECTOR) { - set_intr_gate(vector, interrupt[i]); + set_intr_gate(vector, interrupt[vector]); set_irq_chip_and_handler_name(i, &lguest_irq_controller, handle_level_irq, "level"); diff --git a/arch/x86/mach-generic/bigsmp.c b/arch/x86/mach-generic/bigsmp.c index df37fc9d6a2..3c3b471ea49 100644 --- a/arch/x86/mach-generic/bigsmp.c +++ b/arch/x86/mach-generic/bigsmp.c @@ -41,6 +41,10 @@ static const struct dmi_system_id bigsmp_dmi_table[] = { { } }; +static cpumask_t vector_allocation_domain(int cpu) +{ + return cpumask_of_cpu(cpu); +} static int probe_bigsmp(void) { diff --git a/arch/x86/mach-generic/es7000.c b/arch/x86/mach-generic/es7000.c index 6513d41ea21..28459cab3dd 100644 --- a/arch/x86/mach-generic/es7000.c +++ b/arch/x86/mach-generic/es7000.c @@ -75,4 +75,18 @@ static int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) } #endif +static cpumask_t vector_allocation_domain(int cpu) +{ + /* 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. + */ + cpumask_t domain = { { [0] = APIC_ALL_CPUS, } }; + return domain; +} + struct genapic __initdata_refok apic_es7000 = APIC_INIT("es7000", probe_es7000); diff --git a/arch/x86/mach-generic/numaq.c b/arch/x86/mach-generic/numaq.c index 8cf58394975..71a309b122e 100644 --- a/arch/x86/mach-generic/numaq.c +++ b/arch/x86/mach-generic/numaq.c @@ -38,4 +38,18 @@ static int acpi_madt_oem_check(char *oem_id, char *oem_table_id) return 0; } +static cpumask_t vector_allocation_domain(int cpu) +{ + /* 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. + */ + cpumask_t domain = { { [0] = APIC_ALL_CPUS, } }; + return domain; +} + struct genapic apic_numaq = APIC_INIT("NUMAQ", probe_numaq); diff --git a/arch/x86/mach-generic/summit.c b/arch/x86/mach-generic/summit.c index 6ad6b67a723..6272b5e69da 100644 --- a/arch/x86/mach-generic/summit.c +++ b/arch/x86/mach-generic/summit.c @@ -23,4 +23,18 @@ static int probe_summit(void) return 0; } +static cpumask_t vector_allocation_domain(int cpu) +{ + /* 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. + */ + cpumask_t domain = { { [0] = APIC_ALL_CPUS, } }; + return domain; +} + struct genapic apic_summit = APIC_INIT("summit", probe_summit); diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c index 199a5f4a873..0f6e8a6523a 100644 --- a/arch/x86/mach-voyager/voyager_smp.c +++ b/arch/x86/mach-voyager/voyager_smp.c @@ -1483,7 +1483,7 @@ static void disable_local_vic_irq(unsigned int irq) * the interrupt off to another CPU */ static void before_handle_vic_irq(unsigned int irq) { - irq_desc_t *desc = irq_desc + irq; + irq_desc_t *desc = irq_to_desc(irq); __u8 cpu = smp_processor_id(); _raw_spin_lock(&vic_irq_lock); @@ -1518,7 +1518,7 @@ static void before_handle_vic_irq(unsigned int irq) /* Finish the VIC interrupt: basically mask */ static void after_handle_vic_irq(unsigned int irq) { - irq_desc_t *desc = irq_desc + irq; + irq_desc_t *desc = irq_to_desc(irq); _raw_spin_lock(&vic_irq_lock); { diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 3f2b8962cbd..31e8730fa24 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -640,24 +640,23 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) } -#ifdef CONFIG_X86_32 - /* It's safe to allow irq's after cr2 has been saved and the vmalloc - fault has been handled. */ - if (regs->flags & (X86_EFLAGS_IF | X86_VM_MASK)) - local_irq_enable(); - /* - * If we're in an interrupt, have no user context or are running in an - * atomic region then we must not take the fault. + * It's safe to allow irq's after cr2 has been saved and the + * vmalloc fault has been handled. + * + * User-mode registers count as a user access even for any + * potential system fault or CPU buglet. */ - if (in_atomic() || !mm) - goto bad_area_nosemaphore; -#else /* CONFIG_X86_64 */ - if (likely(regs->flags & X86_EFLAGS_IF)) + if (user_mode_vm(regs)) { + local_irq_enable(); + error_code |= PF_USER; + } else if (regs->flags & X86_EFLAGS_IF) local_irq_enable(); +#ifdef CONFIG_X86_64 if (unlikely(error_code & PF_RSVD)) pgtable_bad(address, regs, error_code); +#endif /* * If we're in an interrupt, have no user context or are running in an @@ -666,15 +665,9 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) if (unlikely(in_atomic() || !mm)) goto bad_area_nosemaphore; - /* - * User-mode registers count as a user access even for any - * potential system fault or CPU buglet. - */ - if (user_mode_vm(regs)) - error_code |= PF_USER; again: -#endif - /* When running in the kernel we expect faults to occur only to + /* + * When running in the kernel we expect faults to occur only to * addresses in user space. All other faults represent errors in the * kernel and should generate an OOPS. Unfortunately, in the case of an * erroneous fault occurring in a code path which already holds mmap_sem @@ -737,9 +730,6 @@ good_area: goto bad_area; } -#ifdef CONFIG_X86_32 -survive: -#endif /* * If for any reason at all we couldn't handle the fault, * make sure we exit gracefully rather than endlessly redo @@ -874,12 +864,11 @@ out_of_memory: up_read(&mm->mmap_sem); if (is_global_init(tsk)) { yield(); -#ifdef CONFIG_X86_32 - down_read(&mm->mmap_sem); - goto survive; -#else + /* + * Re-lookup the vma - in theory the vma tree might + * have changed: + */ goto again; -#endif } printk("VM: killing process %s\n", tsk->comm); diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c index 165c871ba9a..bcc079c282d 100644 --- a/arch/x86/mm/highmem_32.c +++ b/arch/x86/mm/highmem_32.c @@ -137,6 +137,7 @@ void *kmap_atomic_pfn(unsigned long pfn, enum km_type type) return (void*) vaddr; } +EXPORT_SYMBOL_GPL(kmap_atomic_pfn); /* temporarily in use by i915 GEM until vmap */ struct page *kmap_atomic_to_page(void *ptr) { diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index e4c43ec71b2..ae71e11eb3e 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -220,6 +220,12 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr, return (__force void __iomem *)phys_to_virt(phys_addr); /* + * Check if the request spans more than any BAR in the iomem resource + * tree. + */ + WARN_ON(iomem_map_sanity_check(phys_addr, size)); + + /* * Don't allow anybody to remap normal RAM that we're using.. */ for (pfn = phys_addr >> PAGE_SHIFT; diff --git a/arch/x86/mm/mmio-mod.c b/arch/x86/mm/mmio-mod.c index 635b50e8558..2c4baa88f2c 100644 --- a/arch/x86/mm/mmio-mod.c +++ b/arch/x86/mm/mmio-mod.c @@ -56,13 +56,6 @@ struct remap_trace { static DEFINE_PER_CPU(struct trap_reason, pf_reason); static DEFINE_PER_CPU(struct mmiotrace_rw, cpu_trace); -#if 0 /* XXX: no way gather this info anymore */ -/* Access to this is not per-cpu. */ -static DEFINE_PER_CPU(atomic_t, dropped); -#endif - -static struct dentry *marker_file; - static DEFINE_MUTEX(mmiotrace_mutex); static DEFINE_SPINLOCK(trace_lock); static atomic_t mmiotrace_enabled; @@ -75,7 +68,7 @@ static LIST_HEAD(trace_list); /* struct remap_trace */ * and trace_lock. * - Routines depending on is_enabled() must take trace_lock. * - trace_list users must hold trace_lock. - * - is_enabled() guarantees that mmio_trace_record is allowed. + * - is_enabled() guarantees that mmio_trace_{rw,mapping} are allowed. * - pre/post callbacks assume the effect of is_enabled() being true. */ @@ -97,44 +90,6 @@ static bool is_enabled(void) return atomic_read(&mmiotrace_enabled); } -#if 0 /* XXX: needs rewrite */ -/* - * Write callback for the debugfs entry: - * Read a marker and write it to the mmio trace log - */ -static ssize_t write_marker(struct file *file, const char __user *buffer, - size_t count, loff_t *ppos) -{ - char *event = NULL; - struct mm_io_header *headp; - ssize_t len = (count > 65535) ? 65535 : count; - - event = kzalloc(sizeof(*headp) + len, GFP_KERNEL); - if (!event) - return -ENOMEM; - - headp = (struct mm_io_header *)event; - headp->type = MMIO_MAGIC | (MMIO_MARKER << MMIO_OPCODE_SHIFT); - headp->data_len = len; - - if (copy_from_user(event + sizeof(*headp), buffer, len)) { - kfree(event); - return -EFAULT; - } - - spin_lock_irq(&trace_lock); -#if 0 /* XXX: convert this to use tracing */ - if (is_enabled()) - relay_write(chan, event, sizeof(*headp) + len); - else -#endif - len = -EINVAL; - spin_unlock_irq(&trace_lock); - kfree(event); - return len; -} -#endif - static void print_pte(unsigned long address) { unsigned int level; @@ -307,8 +262,10 @@ static void ioremap_trace_core(resource_size_t offset, unsigned long size, map.map_id = trace->id; spin_lock_irq(&trace_lock); - if (!is_enabled()) + if (!is_enabled()) { + kfree(trace); goto not_enabled; + } mmio_trace_mapping(&map); list_add_tail(&trace->list, &trace_list); @@ -377,6 +334,23 @@ void mmiotrace_iounmap(volatile void __iomem *addr) iounmap_trace_core(addr); } +int mmiotrace_printk(const char *fmt, ...) +{ + int ret = 0; + va_list args; + unsigned long flags; + va_start(args, fmt); + + spin_lock_irqsave(&trace_lock, flags); + if (is_enabled()) + ret = mmio_trace_printk(fmt, args); + spin_unlock_irqrestore(&trace_lock, flags); + + va_end(args); + return ret; +} +EXPORT_SYMBOL(mmiotrace_printk); + static void clear_trace_list(void) { struct remap_trace *trace; @@ -462,26 +436,12 @@ static void leave_uniprocessor(void) } #endif -#if 0 /* XXX: out of order */ -static struct file_operations fops_marker = { - .owner = THIS_MODULE, - .write = write_marker -}; -#endif - void enable_mmiotrace(void) { mutex_lock(&mmiotrace_mutex); if (is_enabled()) goto out; -#if 0 /* XXX: tracing does not support text entries */ - marker_file = debugfs_create_file("marker", 0660, dir, NULL, - &fops_marker); - if (!marker_file) - pr_err(NAME "marker file creation failed.\n"); -#endif - if (nommiotrace) pr_info(NAME "MMIO tracing disabled.\n"); enter_uniprocessor(); @@ -506,11 +466,6 @@ void disable_mmiotrace(void) clear_trace_list(); /* guarantees: no more kmmio callbacks */ leave_uniprocessor(); - if (marker_file) { - debugfs_remove(marker_file); - marker_file = NULL; - } - pr_info(NAME "disabled.\n"); out: mutex_unlock(&mmiotrace_mutex); diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index a9ec89c3fbc..407d8784f66 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -792,6 +792,8 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, /* Must avoid aliasing mappings in the highmem code */ kmap_flush_unused(); + vm_unmap_aliases(); + cpa.vaddr = addr; cpa.numpages = numpages; cpa.mask_set = mask_set; diff --git a/arch/x86/mm/pf_in.c b/arch/x86/mm/pf_in.c index efa1911e20c..df3d5c861cd 100644 --- a/arch/x86/mm/pf_in.c +++ b/arch/x86/mm/pf_in.c @@ -79,25 +79,34 @@ static unsigned int mw32[] = { 0xC7 }; static unsigned int mw64[] = { 0x89, 0x8B }; #endif /* not __i386__ */ -static int skip_prefix(unsigned char *addr, int *shorted, int *enlarged, - int *rexr) +struct prefix_bits { + unsigned shorted:1; + unsigned enlarged:1; + unsigned rexr:1; + unsigned rex:1; +}; + +static int skip_prefix(unsigned char *addr, struct prefix_bits *prf) { int i; unsigned char *p = addr; - *shorted = 0; - *enlarged = 0; - *rexr = 0; + prf->shorted = 0; + prf->enlarged = 0; + prf->rexr = 0; + prf->rex = 0; restart: for (i = 0; i < ARRAY_SIZE(prefix_codes); i++) { if (*p == prefix_codes[i]) { if (*p == 0x66) - *shorted = 1; + prf->shorted = 1; #ifdef __amd64__ if ((*p & 0xf8) == 0x48) - *enlarged = 1; + prf->enlarged = 1; if ((*p & 0xf4) == 0x44) - *rexr = 1; + prf->rexr = 1; + if ((*p & 0xf0) == 0x40) + prf->rex = 1; #endif p++; goto restart; @@ -135,12 +144,12 @@ enum reason_type get_ins_type(unsigned long ins_addr) { unsigned int opcode; unsigned char *p; - int shorted, enlarged, rexr; + struct prefix_bits prf; int i; enum reason_type rv = OTHERS; p = (unsigned char *)ins_addr; - p += skip_prefix(p, &shorted, &enlarged, &rexr); + p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); CHECK_OP_TYPE(opcode, reg_rop, REG_READ); @@ -156,10 +165,11 @@ static unsigned int get_ins_reg_width(unsigned long ins_addr) { unsigned int opcode; unsigned char *p; - int i, shorted, enlarged, rexr; + struct prefix_bits prf; + int i; p = (unsigned char *)ins_addr; - p += skip_prefix(p, &shorted, &enlarged, &rexr); + p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); for (i = 0; i < ARRAY_SIZE(rw8); i++) @@ -168,7 +178,7 @@ static unsigned int get_ins_reg_width(unsigned long ins_addr) for (i = 0; i < ARRAY_SIZE(rw32); i++) if (rw32[i] == opcode) - return (shorted ? 2 : (enlarged ? 8 : 4)); + return prf.shorted ? 2 : (prf.enlarged ? 8 : 4); printk(KERN_ERR "mmiotrace: Unknown opcode 0x%02x\n", opcode); return 0; @@ -178,10 +188,11 @@ unsigned int get_ins_mem_width(unsigned long ins_addr) { unsigned int opcode; unsigned char *p; - int i, shorted, enlarged, rexr; + struct prefix_bits prf; + int i; p = (unsigned char *)ins_addr; - p += skip_prefix(p, &shorted, &enlarged, &rexr); + p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); for (i = 0; i < ARRAY_SIZE(mw8); i++) @@ -194,11 +205,11 @@ unsigned int get_ins_mem_width(unsigned long ins_addr) for (i = 0; i < ARRAY_SIZE(mw32); i++) if (mw32[i] == opcode) - return shorted ? 2 : 4; + return prf.shorted ? 2 : 4; for (i = 0; i < ARRAY_SIZE(mw64); i++) if (mw64[i] == opcode) - return shorted ? 2 : (enlarged ? 8 : 4); + return prf.shorted ? 2 : (prf.enlarged ? 8 : 4); printk(KERN_ERR "mmiotrace: Unknown opcode 0x%02x\n", opcode); return 0; @@ -238,7 +249,7 @@ enum { #endif }; -static unsigned char *get_reg_w8(int no, struct pt_regs *regs) +static unsigned char *get_reg_w8(int no, int rex, struct pt_regs *regs) { unsigned char *rv = NULL; @@ -255,18 +266,6 @@ static unsigned char *get_reg_w8(int no, struct pt_regs *regs) case arg_DL: rv = (unsigned char *)®s->dx; break; - case arg_AH: - rv = 1 + (unsigned char *)®s->ax; - break; - case arg_BH: - rv = 1 + (unsigned char *)®s->bx; - break; - case arg_CH: - rv = 1 + (unsigned char *)®s->cx; - break; - case arg_DH: - rv = 1 + (unsigned char *)®s->dx; - break; #ifdef __amd64__ case arg_R8: rv = (unsigned char *)®s->r8; @@ -294,9 +293,55 @@ static unsigned char *get_reg_w8(int no, struct pt_regs *regs) break; #endif default: - printk(KERN_ERR "mmiotrace: Error reg no# %d\n", no); break; } + + if (rv) + return rv; + + if (rex) { + /* + * If REX prefix exists, access low bytes of SI etc. + * instead of AH etc. + */ + switch (no) { + case arg_SI: + rv = (unsigned char *)®s->si; + break; + case arg_DI: + rv = (unsigned char *)®s->di; + break; + case arg_BP: + rv = (unsigned char *)®s->bp; + break; + case arg_SP: + rv = (unsigned char *)®s->sp; + break; + default: + break; + } + } else { + switch (no) { + case arg_AH: + rv = 1 + (unsigned char *)®s->ax; + break; + case arg_BH: + rv = 1 + (unsigned char *)®s->bx; + break; + case arg_CH: + rv = 1 + (unsigned char *)®s->cx; + break; + case arg_DH: + rv = 1 + (unsigned char *)®s->dx; + break; + default: + break; + } + } + + if (!rv) + printk(KERN_ERR "mmiotrace: Error reg no# %d\n", no); + return rv; } @@ -368,11 +413,12 @@ unsigned long get_ins_reg_val(unsigned long ins_addr, struct pt_regs *regs) unsigned char mod_rm; int reg; unsigned char *p; - int i, shorted, enlarged, rexr; + struct prefix_bits prf; + int i; unsigned long rv; p = (unsigned char *)ins_addr; - p += skip_prefix(p, &shorted, &enlarged, &rexr); + p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); for (i = 0; i < ARRAY_SIZE(reg_rop); i++) if (reg_rop[i] == opcode) { @@ -392,10 +438,10 @@ unsigned long get_ins_reg_val(unsigned long ins_addr, struct pt_regs *regs) do_work: mod_rm = *p; - reg = ((mod_rm >> 3) & 0x7) | (rexr << 3); + reg = ((mod_rm >> 3) & 0x7) | (prf.rexr << 3); switch (get_ins_reg_width(ins_addr)) { case 1: - return *get_reg_w8(reg, regs); + return *get_reg_w8(reg, prf.rex, regs); case 2: return *(unsigned short *)get_reg_w32(reg, regs); @@ -422,11 +468,12 @@ unsigned long get_ins_imm_val(unsigned long ins_addr) unsigned char mod_rm; unsigned char mod; unsigned char *p; - int i, shorted, enlarged, rexr; + struct prefix_bits prf; + int i; unsigned long rv; p = (unsigned char *)ins_addr; - p += skip_prefix(p, &shorted, &enlarged, &rexr); + p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); for (i = 0; i < ARRAY_SIZE(imm_wop); i++) if (imm_wop[i] == opcode) { diff --git a/arch/x86/mm/testmmiotrace.c b/arch/x86/mm/testmmiotrace.c index d877c5b423e..ab50a8d7402 100644 --- a/arch/x86/mm/testmmiotrace.c +++ b/arch/x86/mm/testmmiotrace.c @@ -3,6 +3,7 @@ */ #include <linux/module.h> #include <linux/io.h> +#include <linux/mmiotrace.h> #define MODULE_NAME "testmmiotrace" @@ -13,6 +14,7 @@ MODULE_PARM_DESC(mmio_address, "Start address of the mapping of 16 kB."); static void do_write_test(void __iomem *p) { unsigned int i; + mmiotrace_printk("Write test.\n"); for (i = 0; i < 256; i++) iowrite8(i, p + i); for (i = 1024; i < (5 * 1024); i += 2) @@ -24,6 +26,7 @@ static void do_write_test(void __iomem *p) static void do_read_test(void __iomem *p) { unsigned int i; + mmiotrace_printk("Read test.\n"); for (i = 0; i < 256; i++) ioread8(p + i); for (i = 1024; i < (5 * 1024); i += 2) @@ -39,6 +42,7 @@ static void do_test(void) pr_err(MODULE_NAME ": could not ioremap, aborting.\n"); return; } + mmiotrace_printk("ioremap returned %p.\n", p); do_write_test(p); do_read_test(p); iounmap(p); diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c index 006599db0dc..bf69dbe08bf 100644 --- a/arch/x86/pci/irq.c +++ b/arch/x86/pci/irq.c @@ -493,7 +493,7 @@ static int pirq_amd756_get(struct pci_dev *router, struct pci_dev *dev, int pirq if (pirq <= 4) irq = read_config_nybble(router, 0x56, pirq - 1); dev_info(&dev->dev, - "AMD756: dev [%04x/%04x], router PIRQ %d get IRQ %d\n", + "AMD756: dev [%04x:%04x], router PIRQ %d get IRQ %d\n", dev->vendor, dev->device, pirq, irq); return irq; } @@ -501,7 +501,7 @@ static int pirq_amd756_get(struct pci_dev *router, struct pci_dev *dev, int pirq static int pirq_amd756_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq) { dev_info(&dev->dev, - "AMD756: dev [%04x/%04x], router PIRQ %d set IRQ %d\n", + "AMD756: dev [%04x:%04x], router PIRQ %d set IRQ %d\n", dev->vendor, dev->device, pirq, irq); if (pirq <= 4) write_config_nybble(router, 0x56, pirq - 1, irq); @@ -590,13 +590,20 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH10_1: case PCI_DEVICE_ID_INTEL_ICH10_2: case PCI_DEVICE_ID_INTEL_ICH10_3: - case PCI_DEVICE_ID_INTEL_PCH_0: - case PCI_DEVICE_ID_INTEL_PCH_1: r->name = "PIIX/ICH"; r->get = pirq_piix_get; r->set = pirq_piix_set; return 1; } + + if ((device >= PCI_DEVICE_ID_INTEL_PCH_LPC_MIN) && + (device <= PCI_DEVICE_ID_INTEL_PCH_LPC_MAX)) { + r->name = "PIIX/ICH"; + r->get = pirq_piix_get; + r->set = pirq_piix_set; + return 1; + } + return 0; } @@ -823,7 +830,7 @@ static void __init pirq_find_router(struct irq_router *r) r->get = NULL; r->set = NULL; - DBG(KERN_DEBUG "PCI: Attempting to find IRQ router for %04x:%04x\n", + DBG(KERN_DEBUG "PCI: Attempting to find IRQ router for [%04x:%04x]\n", rt->rtr_vendor, rt->rtr_device); pirq_router_dev = pci_get_bus_and_slot(rt->rtr_bus, rt->rtr_devfn); @@ -843,7 +850,7 @@ static void __init pirq_find_router(struct irq_router *r) h->probe(r, pirq_router_dev, pirq_router_dev->device)) break; } - dev_info(&pirq_router_dev->dev, "%s IRQ router [%04x/%04x]\n", + dev_info(&pirq_router_dev->dev, "%s IRQ router [%04x:%04x]\n", pirq_router.name, pirq_router_dev->vendor, pirq_router_dev->device); diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 0013a729b41..b61534c7a4c 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -871,6 +871,7 @@ static void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, unsigned l /* make sure there are no stray mappings of this page */ kmap_flush_unused(); + vm_unmap_aliases(); } } diff --git a/arch/x86/xen/irq.c b/arch/x86/xen/irq.c index 28b85ab8422..bb042608c60 100644 --- a/arch/x86/xen/irq.c +++ b/arch/x86/xen/irq.c @@ -21,7 +21,6 @@ void xen_force_evtchn_callback(void) static void __init __xen_init_IRQ(void) { -#ifdef CONFIG_X86_64 int i; /* Create identity vector->irq map */ @@ -31,7 +30,6 @@ static void __init __xen_init_IRQ(void) for_each_possible_cpu(cpu) per_cpu(vector_irq, cpu)[i] = i; } -#endif /* CONFIG_X86_64 */ xen_init_IRQ(); } diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index ae173f6edd8..d4d52f5a1cf 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -846,6 +846,7 @@ static void __xen_pgd_pin(struct mm_struct *mm, pgd_t *pgd) /* re-enable interrupts for kmap_flush_unused */ xen_mc_issue(0); kmap_flush_unused(); + vm_unmap_aliases(); xen_mc_batch(); } diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c index dd71e3a021c..5601506f2dd 100644 --- a/arch/x86/xen/spinlock.c +++ b/arch/x86/xen/spinlock.c @@ -241,7 +241,7 @@ static noinline int xen_spin_lock_slow(struct raw_spinlock *lock, bool irq_enabl ADD_STATS(taken_slow_spurious, !xen_test_irq_pending(irq)); } while (!xen_test_irq_pending(irq)); /* check for spurious wakeups */ - kstat_this_cpu.irqs[irq]++; + kstat_incr_irqs_this_cpu(irq, irq_to_desc(irq)); out: raw_local_irq_restore(flags); diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index 004ba86326a..c9f7cda48ed 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -198,17 +198,10 @@ unsigned long long xen_sched_clock(void) /* Get the TSC speed from Xen */ unsigned long xen_tsc_khz(void) { - u64 xen_khz = 1000000ULL << 32; - const struct pvclock_vcpu_time_info *info = + struct pvclock_vcpu_time_info *info = &HYPERVISOR_shared_info->vcpu_info[0].time; - do_div(xen_khz, info->tsc_to_system_mul); - if (info->tsc_shift < 0) - xen_khz <<= -info->tsc_shift; - else - xen_khz >>= info->tsc_shift; - - return xen_khz; + return pvclock_tsc_khz(info); } cycle_t xen_clocksource_read(void) |