diff options
Diffstat (limited to '')
84 files changed, 2803 insertions, 2862 deletions
diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile index 96fb8a020af2..1b452a1665c4 100644 --- a/arch/i386/kernel/Makefile +++ b/arch/i386/kernel/Makefile @@ -7,10 +7,10 @@ extra-y := head.o init_task.o vmlinux.lds obj-y := process.o semaphore.o signal.o entry.o traps.o irq.o \ ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_i386.o \ pci-dma.o i386_ksyms.o i387.o bootflag.o \ - quirks.o i8237.o topology.o alternative.o + quirks.o i8237.o topology.o alternative.o i8253.o tsc.o +obj-$(CONFIG_STACKTRACE) += stacktrace.o obj-y += cpu/ -obj-y += timers/ obj-y += acpi/ obj-$(CONFIG_X86_BIOS_REBOOT) += reboot.o obj-$(CONFIG_MCA) += mca.o @@ -37,6 +37,9 @@ obj-$(CONFIG_EFI) += efi.o efi_stub.o obj-$(CONFIG_DOUBLEFAULT) += doublefault.o obj-$(CONFIG_VM86) += vm86.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o +obj-$(CONFIG_HPET_TIMER) += hpet.o +obj-$(CONFIG_K8_NB) += k8.o +obj-$(CONFIG_AUDIT) += audit.o EXTRA_AFLAGS := -traditional @@ -76,3 +79,6 @@ SYSCFLAGS_vsyscall-syms.o = -r $(obj)/vsyscall-syms.o: $(src)/vsyscall.lds \ $(obj)/vsyscall-sysenter.o $(obj)/vsyscall-note.o FORCE $(call if_changed,syscall) + +k8-y += ../../x86_64/kernel/k8.o + diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c index daee69579b1c..0db6387025ca 100644 --- a/arch/i386/kernel/acpi/boot.c +++ b/arch/i386/kernel/acpi/boot.c @@ -24,7 +24,6 @@ */ #include <linux/init.h> -#include <linux/config.h> #include <linux/acpi.h> #include <linux/efi.h> #include <linux/module.h> @@ -202,6 +201,8 @@ int __init acpi_parse_mcfg(unsigned long phys_addr, unsigned long size) if (mcfg->config[i].base_reserved) { printk(KERN_ERR PREFIX "MMCONFIG not in low 4GB of memory\n"); + kfree(pci_mmcfg_config); + pci_mmcfg_config_num = 0; return -ENODEV; } } @@ -215,7 +216,7 @@ static int __init acpi_parse_madt(unsigned long phys_addr, unsigned long size) { struct acpi_table_madt *madt = NULL; - if (!phys_addr || !size) + if (!phys_addr || !size || !cpu_has_apic) return -EINVAL; madt = (struct acpi_table_madt *)__acpi_map_table(phys_addr, size); @@ -621,9 +622,9 @@ extern u32 pmtmr_ioport; static int __init acpi_parse_fadt(unsigned long phys, unsigned long size) { - struct fadt_descriptor_rev2 *fadt = NULL; + struct fadt_descriptor *fadt = NULL; - fadt = (struct fadt_descriptor_rev2 *)__acpi_map_table(phys, size); + fadt = (struct fadt_descriptor *)__acpi_map_table(phys, size); if (!fadt) { printk(KERN_WARNING PREFIX "Unable to map FADT\n"); return 0; @@ -754,7 +755,7 @@ static int __init acpi_parse_madt_ioapic_entries(void) return -ENODEV; } - if (!cpu_has_apic) + if (!cpu_has_apic) return -ENODEV; /* @@ -1066,14 +1067,6 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 360"), }, }, - { - .callback = disable_acpi_pci, - .ident = "HP xw9300", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), - DMI_MATCH(DMI_PRODUCT_NAME, "HP xw9300 Workstation"), - }, - }, {} }; diff --git a/arch/i386/kernel/acpi/earlyquirk.c b/arch/i386/kernel/acpi/earlyquirk.c index 2e3b643a4dc4..1649a175a206 100644 --- a/arch/i386/kernel/acpi/earlyquirk.c +++ b/arch/i386/kernel/acpi/earlyquirk.c @@ -5,17 +5,34 @@ #include <linux/init.h> #include <linux/kernel.h> #include <linux/pci.h> +#include <linux/acpi.h> + #include <asm/pci-direct.h> #include <asm/acpi.h> #include <asm/apic.h> +#ifdef CONFIG_ACPI + +static int nvidia_hpet_detected __initdata; + +static int __init nvidia_hpet_check(unsigned long phys, unsigned long size) +{ + nvidia_hpet_detected = 1; + return 0; +} +#endif + static int __init check_bridge(int vendor, int device) { #ifdef CONFIG_ACPI - /* According to Nvidia all timer overrides are bogus. Just ignore - them all. */ + /* According to Nvidia all timer overrides are bogus unless HPET + is enabled. */ if (vendor == PCI_VENDOR_ID_NVIDIA) { - acpi_skip_timer_override = 1; + nvidia_hpet_detected = 0; + acpi_table_parse(ACPI_HPET, nvidia_hpet_check); + if (nvidia_hpet_detected == 0) { + acpi_skip_timer_override = 1; + } } #endif if (vendor == PCI_VENDOR_ID_ATI && timer_over_8254 == 1) { diff --git a/arch/i386/kernel/acpi/processor.c b/arch/i386/kernel/acpi/processor.c index 9f4cc02717ec..b54fded49834 100644 --- a/arch/i386/kernel/acpi/processor.c +++ b/arch/i386/kernel/acpi/processor.c @@ -47,7 +47,7 @@ static void init_intel_pdc(struct acpi_processor *pr, struct cpuinfo_x86 *c) buf[2] = ACPI_PDC_C_CAPABILITY_SMP; if (cpu_has(c, X86_FEATURE_EST)) - buf[2] |= ACPI_PDC_EST_CAPABILITY_SMP; + buf[2] |= ACPI_PDC_EST_CAPABILITY_SWSMP; obj->type = ACPI_TYPE_BUFFER; obj->buffer.length = 12; diff --git a/arch/i386/kernel/acpi/sleep.c b/arch/i386/kernel/acpi/sleep.c index 1cb2b186a3af..4ee83577bf61 100644 --- a/arch/i386/kernel/acpi/sleep.c +++ b/arch/i386/kernel/acpi/sleep.c @@ -8,30 +8,17 @@ #include <linux/acpi.h> #include <linux/bootmem.h> #include <linux/dmi.h> +#include <linux/cpumask.h> + #include <asm/smp.h> -#include <asm/tlbflush.h> /* address in low memory of the wakeup routine. */ unsigned long acpi_wakeup_address = 0; unsigned long acpi_video_flags; extern char wakeup_start, wakeup_end; -extern void zap_low_mappings(void); - extern unsigned long FASTCALL(acpi_copy_wakeup_routine(unsigned long)); -static void init_low_mapping(pgd_t * pgd, int pgd_limit) -{ - int pgd_ofs = 0; - - while ((pgd_ofs < pgd_limit) - && (pgd_ofs + USER_PTRS_PER_PGD < PTRS_PER_PGD)) { - set_pgd(pgd, *(pgd + USER_PTRS_PER_PGD)); - pgd_ofs++, pgd++; - } - flush_tlb_all(); -} - /** * acpi_save_state_mem - save kernel state * @@ -42,7 +29,6 @@ int acpi_save_state_mem(void) { if (!acpi_wakeup_address) return 1; - init_low_mapping(swapper_pg_dir, USER_PTRS_PER_PGD); memcpy((void *)acpi_wakeup_address, &wakeup_start, &wakeup_end - &wakeup_start); acpi_copy_wakeup_routine(acpi_wakeup_address); @@ -55,7 +41,6 @@ int acpi_save_state_mem(void) */ void acpi_restore_state_mem(void) { - zap_low_mappings(); } /** diff --git a/arch/i386/kernel/acpi/wakeup.S b/arch/i386/kernel/acpi/wakeup.S index 7c74fe0dc93c..9f408eee4e6f 100644 --- a/arch/i386/kernel/acpi/wakeup.S +++ b/arch/i386/kernel/acpi/wakeup.S @@ -56,7 +56,7 @@ wakeup_code: 1: # set up page table - movl $swapper_pg_dir-__PAGE_OFFSET, %eax + movl $swsusp_pg_dir-__PAGE_OFFSET, %eax movl %eax, %cr3 testl $1, real_efer_save_restore - wakeup_code @@ -265,11 +265,6 @@ ENTRY(acpi_copy_wakeup_routine) movl $0x12345678, saved_magic ret -.data -ALIGN -ENTRY(saved_magic) .long 0 -ENTRY(saved_eip) .long 0 - save_registers: leal 4(%esp), %eax movl %eax, saved_context_esp @@ -304,7 +299,11 @@ ret_point: call restore_processor_state ret +.data ALIGN +ENTRY(saved_magic) .long 0 +ENTRY(saved_eip) .long 0 + # saved registers saved_gdt: .long 0,0 saved_idt: .long 0,0 diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c index 5cbd6f99fb2a..28ab80649764 100644 --- a/arch/i386/kernel/alternative.c +++ b/arch/i386/kernel/alternative.c @@ -4,27 +4,41 @@ #include <asm/alternative.h> #include <asm/sections.h> -#define DEBUG 0 -#if DEBUG -# define DPRINTK(fmt, args...) printk(fmt, args) -#else -# define DPRINTK(fmt, args...) -#endif +static int no_replacement = 0; +static int smp_alt_once = 0; +static int debug_alternative = 0; + +static int __init noreplacement_setup(char *s) +{ + no_replacement = 1; + return 1; +} +static int __init bootonly(char *str) +{ + smp_alt_once = 1; + return 1; +} +static int __init debug_alt(char *str) +{ + debug_alternative = 1; + return 1; +} + +__setup("noreplacement", noreplacement_setup); +__setup("smp-alt-boot", bootonly); +__setup("debug-alternative", debug_alt); + +#define DPRINTK(fmt, args...) if (debug_alternative) \ + printk(KERN_DEBUG fmt, args) +#ifdef GENERIC_NOP1 /* Use inline assembly to define this because the nops are defined as inline assembly strings in the include files and we cannot get them easily into strings. */ asm("\t.data\nintelnops: " GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6 GENERIC_NOP7 GENERIC_NOP8); -asm("\t.data\nk8nops: " - K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6 - K8_NOP7 K8_NOP8); -asm("\t.data\nk7nops: " - K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6 - K7_NOP7 K7_NOP8); - -extern unsigned char intelnops[], k8nops[], k7nops[]; +extern unsigned char intelnops[]; static unsigned char *intel_nops[ASM_NOP_MAX+1] = { NULL, intelnops, @@ -36,6 +50,13 @@ static unsigned char *intel_nops[ASM_NOP_MAX+1] = { intelnops + 1 + 2 + 3 + 4 + 5 + 6, intelnops + 1 + 2 + 3 + 4 + 5 + 6 + 7, }; +#endif + +#ifdef K8_NOP1 +asm("\t.data\nk8nops: " + K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6 + K8_NOP7 K8_NOP8); +extern unsigned char k8nops[]; static unsigned char *k8_nops[ASM_NOP_MAX+1] = { NULL, k8nops, @@ -47,6 +68,13 @@ static unsigned char *k8_nops[ASM_NOP_MAX+1] = { k8nops + 1 + 2 + 3 + 4 + 5 + 6, k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7, }; +#endif + +#ifdef K7_NOP1 +asm("\t.data\nk7nops: " + K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6 + K7_NOP7 K7_NOP8); +extern unsigned char k7nops[]; static unsigned char *k7_nops[ASM_NOP_MAX+1] = { NULL, k7nops, @@ -58,6 +86,18 @@ static unsigned char *k7_nops[ASM_NOP_MAX+1] = { k7nops + 1 + 2 + 3 + 4 + 5 + 6, k7nops + 1 + 2 + 3 + 4 + 5 + 6 + 7, }; +#endif + +#ifdef CONFIG_X86_64 + +extern char __vsyscall_0; +static inline unsigned char** find_nop_table(void) +{ + return k8_nops; +} + +#else /* CONFIG_X86_64 */ + static struct nop { int cpuid; unsigned char **noptable; @@ -67,14 +107,6 @@ static struct nop { { -1, NULL } }; - -extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; -extern struct alt_instr __smp_alt_instructions[], __smp_alt_instructions_end[]; -extern u8 *__smp_locks[], *__smp_locks_end[]; - -extern u8 __smp_alt_begin[], __smp_alt_end[]; - - static unsigned char** find_nop_table(void) { unsigned char **noptable = intel_nops; @@ -89,6 +121,14 @@ static unsigned char** find_nop_table(void) return noptable; } +#endif /* CONFIG_X86_64 */ + +extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; +extern struct alt_instr __smp_alt_instructions[], __smp_alt_instructions_end[]; +extern u8 *__smp_locks[], *__smp_locks_end[]; + +extern u8 __smp_alt_begin[], __smp_alt_end[]; + /* Replace instructions with better alternatives for this CPU type. This runs before SMP is initialized to avoid SMP problems with self modifying code. This implies that assymetric systems where @@ -99,6 +139,7 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end) { unsigned char **noptable = find_nop_table(); struct alt_instr *a; + u8 *instr; int diff, i, k; DPRINTK("%s: alt table %p -> %p\n", __FUNCTION__, start, end); @@ -106,7 +147,16 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end) BUG_ON(a->replacementlen > a->instrlen); if (!boot_cpu_has(a->cpuid)) continue; - memcpy(a->instr, a->replacement, a->replacementlen); + instr = a->instr; +#ifdef CONFIG_X86_64 + /* vsyscall code is not mapped yet. resolve it manually. */ + if (instr >= (u8 *)VSYSCALL_START && instr < (u8*)VSYSCALL_END) { + instr = __va(instr - (u8*)VSYSCALL_START + (u8*)__pa_symbol(&__vsyscall_0)); + DPRINTK("%s: vsyscall fixup: %p => %p\n", + __FUNCTION__, a->instr, instr); + } +#endif + memcpy(instr, a->replacement, a->replacementlen); diff = a->instrlen - a->replacementlen; /* Pad the rest with nops */ for (i = a->replacementlen; diff > 0; diff -= k, i += k) { @@ -118,6 +168,8 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end) } } +#ifdef CONFIG_SMP + static void alternatives_smp_save(struct alt_instr *start, struct alt_instr *end) { struct alt_instr *a; @@ -186,14 +238,6 @@ struct smp_alt_module { static LIST_HEAD(smp_alt_modules); static DEFINE_SPINLOCK(smp_alt); -static int smp_alt_once = 0; -static int __init bootonly(char *str) -{ - smp_alt_once = 1; - return 1; -} -__setup("smp-alt-boot", bootonly); - void alternatives_smp_module_add(struct module *mod, char *name, void *locks, void *locks_end, void *text, void *text_end) @@ -201,6 +245,9 @@ void alternatives_smp_module_add(struct module *mod, char *name, struct smp_alt_module *smp; unsigned long flags; + if (no_replacement) + return; + if (smp_alt_once) { if (boot_cpu_has(X86_FEATURE_UP)) alternatives_smp_unlock(locks, locks_end, @@ -235,7 +282,7 @@ void alternatives_smp_module_del(struct module *mod) struct smp_alt_module *item; unsigned long flags; - if (smp_alt_once) + if (no_replacement || smp_alt_once) return; spin_lock_irqsave(&smp_alt, flags); @@ -256,7 +303,17 @@ void alternatives_smp_switch(int smp) struct smp_alt_module *mod; unsigned long flags; - if (smp_alt_once) +#ifdef CONFIG_LOCKDEP + /* + * A not yet fixed binutils section handling bug prevents + * alternatives-replacement from working reliably, so turn + * it off: + */ + printk("lockdep: not fixing up alternatives.\n"); + return; +#endif + + if (no_replacement || smp_alt_once) return; BUG_ON(!smp && (num_online_cpus() > 1)); @@ -283,8 +340,17 @@ void alternatives_smp_switch(int smp) spin_unlock_irqrestore(&smp_alt, flags); } +#endif + void __init alternative_instructions(void) { + if (no_replacement) { + printk(KERN_INFO "(SMP-)alternatives turned off\n"); + free_init_pages("SMP alternatives", + (unsigned long)__smp_alt_begin, + (unsigned long)__smp_alt_end); + return; + } apply_alternatives(__alt_instructions, __alt_instructions_end); /* switch to patch-once-at-boottime-only mode and free the @@ -297,6 +363,7 @@ void __init alternative_instructions(void) smp_alt_once = 1; #endif +#ifdef CONFIG_SMP if (smp_alt_once) { if (1 == num_possible_cpus()) { printk(KERN_INFO "SMP alternatives: switching to UP code\n"); @@ -318,4 +385,5 @@ void __init alternative_instructions(void) _text, _etext); alternatives_smp_switch(0); } +#endif } diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c index 3d4b2f3d116a..8c844d07862f 100644 --- a/arch/i386/kernel/apic.c +++ b/arch/i386/kernel/apic.c @@ -14,7 +14,6 @@ * Mikael Pettersson : PM converted to driver model. */ -#include <linux/config.h> #include <linux/init.h> #include <linux/mm.h> @@ -36,6 +35,7 @@ #include <asm/arch_hooks.h> #include <asm/hpet.h> #include <asm/i8253.h> +#include <asm/nmi.h> #include <mach_apic.h> #include <mach_apicdef.h> @@ -62,7 +62,7 @@ int apic_verbosity; static void apic_pm_activate(void); -int modern_apic(void) +static int modern_apic(void) { unsigned int lvr, version; /* AMD systems use old APIC versions, so check the CPU */ @@ -113,7 +113,7 @@ void __init apic_intr_init(void) } /* Using APIC to generate smp_local_timer_interrupt? */ -int using_apic_timer = 0; +int using_apic_timer __read_mostly = 0; static int enabled_via_apicbase; @@ -156,7 +156,7 @@ void clear_local_APIC(void) maxlvt = get_maxlvt(); /* - * Masking an LVT entry on a P6 can trigger a local APIC error + * Masking an LVT entry can trigger a local APIC error * if the vector is zero. Mask LVTERR first to prevent this. */ if (maxlvt >= 3) { @@ -1117,7 +1117,18 @@ void disable_APIC_timer(void) unsigned long v; v = apic_read(APIC_LVTT); - apic_write_around(APIC_LVTT, v | APIC_LVT_MASKED); + /* + * When an illegal vector value (0-15) is written to an LVT + * entry and delivery mode is Fixed, the APIC may signal an + * illegal vector error, with out regard to whether the mask + * bit is set or whether an interrupt is actually seen on input. + * + * Boot sequence might call this function when the LVTT has + * '0' vector value. So make sure vector field is set to + * valid value. + */ + v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); + apic_write_around(APIC_LVTT, v); } } diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c index df0e1745f189..8591f2fa920c 100644 --- a/arch/i386/kernel/apm.c +++ b/arch/i386/kernel/apm.c @@ -201,7 +201,6 @@ * http://www.microsoft.com/hwdev/busbios/amp_12.htm] */ -#include <linux/config.h> #include <linux/module.h> #include <linux/poll.h> @@ -374,14 +373,14 @@ static struct { unsigned short segment; } apm_bios_entry; static int clock_slowed; -static int idle_threshold = DEFAULT_IDLE_THRESHOLD; -static int idle_period = DEFAULT_IDLE_PERIOD; +static int idle_threshold __read_mostly = DEFAULT_IDLE_THRESHOLD; +static int idle_period __read_mostly = DEFAULT_IDLE_PERIOD; static int set_pm_idle; static int suspends_pending; static int standbys_pending; static int ignore_sys_suspend; static int ignore_normal_resume; -static int bounce_interval = DEFAULT_BOUNCE_INTERVAL; +static int bounce_interval __read_mostly = DEFAULT_BOUNCE_INTERVAL; #ifdef CONFIG_APM_RTC_IS_GMT # define clock_cmos_diff 0 @@ -390,8 +389,8 @@ static int bounce_interval = DEFAULT_BOUNCE_INTERVAL; static long clock_cmos_diff; static int got_clock_diff; #endif -static int debug; -static int smp; +static int debug __read_mostly; +static int smp __read_mostly; static int apm_disabled = -1; #ifdef CONFIG_SMP static int power_off; @@ -403,8 +402,8 @@ static int realmode_power_off = 1; #else static int realmode_power_off; #endif -static int exit_kapmd; -static int kapmd_running; +static int exit_kapmd __read_mostly; +static int kapmd_running __read_mostly; #ifdef CONFIG_APM_ALLOW_INTS static int allow_ints = 1; #else @@ -416,15 +415,15 @@ static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue); static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue); static struct apm_user * user_list; static DEFINE_SPINLOCK(user_list_lock); -static struct desc_struct bad_bios_desc = { 0, 0x00409200 }; +static const struct desc_struct bad_bios_desc = { 0, 0x00409200 }; -static char driver_version[] = "1.16ac"; /* no spaces */ +static const char driver_version[] = "1.16ac"; /* no spaces */ /* * APM event names taken from the APM 1.2 specification. These are * the message codes that the BIOS uses to tell us about events */ -static char * apm_event_name[] = { +static const char * const apm_event_name[] = { "system standby", "system suspend", "normal resume", @@ -616,7 +615,7 @@ static u8 apm_bios_call(u32 func, u32 ebx_in, u32 ecx_in, * @ecx_in: ECX register value for BIOS call * @eax: EAX register on return from the BIOS call * - * Make a BIOS call that does only returns one value, or just status. + * Make a BIOS call that returns one value only, or just status. * If there is an error, then the error code is returned in AH * (bits 8-15 of eax) and this function returns non-zero. This is * used for simpler BIOS operations. This call may hold interrupts @@ -764,9 +763,9 @@ static int apm_do_idle(void) int idled = 0; int polling; - polling = test_thread_flag(TIF_POLLING_NRFLAG); + polling = !!(current_thread_info()->status & TS_POLLING); if (polling) { - clear_thread_flag(TIF_POLLING_NRFLAG); + current_thread_info()->status &= ~TS_POLLING; smp_mb__after_clear_bit(); } if (!need_resched()) { @@ -774,7 +773,7 @@ static int apm_do_idle(void) ret = apm_bios_call_simple(APM_FUNC_IDLE, 0, 0, &eax); } if (polling) - set_thread_flag(TIF_POLLING_NRFLAG); + current_thread_info()->status |= TS_POLLING; if (!idled) return 0; @@ -822,7 +821,7 @@ static void apm_do_busy(void) #define IDLE_CALC_LIMIT (HZ * 100) #define IDLE_LEAKY_MAX 16 -static void (*original_pm_idle)(void); +static void (*original_pm_idle)(void) __read_mostly; /** * apm_cpu_idle - cpu idling for APM capable Linux @@ -1063,7 +1062,8 @@ static int apm_engage_power_management(u_short device, int enable) static int apm_console_blank(int blank) { - int error, i; + int error = APM_NOT_ENGAGED; /* silence gcc */ + int i; u_short state; static const u_short dev[3] = { 0x100, 0x1FF, 0x101 }; @@ -1104,7 +1104,8 @@ static int queue_empty(struct apm_user *as) static apm_event_t get_queued_event(struct apm_user *as) { - as->event_tail = (as->event_tail + 1) % APM_MAX_EVENTS; + if (++as->event_tail >= APM_MAX_EVENTS) + as->event_tail = 0; return as->events[as->event_tail]; } @@ -1118,13 +1119,16 @@ static void queue_event(apm_event_t event, struct apm_user *sender) for (as = user_list; as != NULL; as = as->next) { if ((as == sender) || (!as->reader)) continue; - as->event_head = (as->event_head + 1) % APM_MAX_EVENTS; + if (++as->event_head >= APM_MAX_EVENTS) + as->event_head = 0; + if (as->event_head == as->event_tail) { static int notified; if (notified++ == 0) printk(KERN_ERR "apm: an event queue overflowed\n"); - as->event_tail = (as->event_tail + 1) % APM_MAX_EVENTS; + if (++as->event_tail >= APM_MAX_EVENTS) + as->event_tail = 0; } as->events[as->event_head] = event; if ((!as->suser) || (!as->writer)) @@ -1282,7 +1286,7 @@ static void standby(void) static apm_event_t get_event(void) { int error; - apm_event_t event; + apm_event_t event = APM_NO_EVENTS; /* silence gcc */ apm_eventinfo_t info; static int notified; diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index 36d66e2077d0..c80271f8f084 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -4,6 +4,7 @@ * to extract and format the required data. */ +#include <linux/crypto.h> #include <linux/sched.h> #include <linux/signal.h> #include <linux/personality.h> @@ -13,6 +14,7 @@ #include <asm/fixmap.h> #include <asm/processor.h> #include <asm/thread_info.h> +#include <asm/elf.h> #define DEFINE(sym, val) \ asm volatile("\n->" #sym " %0 " #val : : "i" (val)) @@ -53,6 +55,7 @@ void foo(void) OFFSET(TI_preempt_count, thread_info, preempt_count); OFFSET(TI_addr_limit, thread_info, addr_limit); OFFSET(TI_restart_block, thread_info, restart_block); + OFFSET(TI_sysenter_return, thread_info, sysenter_return); BLANK(); OFFSET(EXEC_DOMAIN_handler, exec_domain, handler); @@ -68,5 +71,7 @@ void foo(void) sizeof(struct tss_struct)); DEFINE(PAGE_SIZE_asm, PAGE_SIZE); - DEFINE(VSYSCALL_BASE, __fix_to_virt(FIX_VSYSCALL)); + DEFINE(VDSO_PRELINK, VDSO_PRELINK); + + OFFSET(crypto_tfm_ctx_offset, crypto_tfm, __crt_ctx); } diff --git a/arch/i386/kernel/audit.c b/arch/i386/kernel/audit.c new file mode 100644 index 000000000000..5a53c6f371ff --- /dev/null +++ b/arch/i386/kernel/audit.c @@ -0,0 +1,23 @@ +#include <linux/init.h> +#include <linux/types.h> +#include <linux/audit.h> +#include <asm/unistd.h> + +static unsigned dir_class[] = { +#include <asm-generic/audit_dir_write.h> +~0U +}; + +static unsigned chattr_class[] = { +#include <asm-generic/audit_change_attr.h> +~0U +}; + +static int __init audit_classes_init(void) +{ + audit_register_class(AUDIT_CLASS_DIR_WRITE, dir_class); + audit_register_class(AUDIT_CLASS_CHATTR, chattr_class); + return 0; +} + +__initcall(audit_classes_init); diff --git a/arch/i386/kernel/bootflag.c b/arch/i386/kernel/bootflag.c index 4c30ed01f4e1..0b9860530a6b 100644 --- a/arch/i386/kernel/bootflag.c +++ b/arch/i386/kernel/bootflag.c @@ -3,7 +3,6 @@ */ -#include <linux/config.h> #include <linux/types.h> #include <linux/kernel.h> #include <linux/init.h> diff --git a/arch/i386/kernel/cpu/amd.c b/arch/i386/kernel/cpu/amd.c index 786d1a57048b..e6a2d6b80cda 100644 --- a/arch/i386/kernel/cpu/amd.c +++ b/arch/i386/kernel/cpu/amd.c @@ -224,22 +224,26 @@ static void __init init_amd(struct cpuinfo_x86 *c) #ifdef CONFIG_X86_HT /* - * On a AMD dual core setup the lower bits of the APIC id - * distingush the cores. Assumes number of cores is a power - * of two. + * On a AMD multi core setup the lower bits of the APIC id + * distingush the cores. */ if (c->x86_max_cores > 1) { int cpu = smp_processor_id(); - unsigned bits = 0; - while ((1 << bits) < c->x86_max_cores) - bits++; - cpu_core_id[cpu] = phys_proc_id[cpu] & ((1<<bits)-1); - phys_proc_id[cpu] >>= bits; + unsigned bits = (cpuid_ecx(0x80000008) >> 12) & 0xf; + + if (bits == 0) { + while ((1 << bits) < c->x86_max_cores) + bits++; + } + c->cpu_core_id = c->phys_proc_id & ((1<<bits)-1); + c->phys_proc_id >>= bits; printk(KERN_INFO "CPU %d(%d) -> Core %d\n", - cpu, c->x86_max_cores, cpu_core_id[cpu]); + cpu, c->x86_max_cores, c->cpu_core_id); } #endif + if (cpuid_eax(0x80000000) >= 0x80000006) + num_cache_leaves = 3; } static unsigned int amd_size_cache(struct cpuinfo_x86 * c, unsigned int size) diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index a06a49075f10..70c87de582c7 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -11,6 +11,8 @@ #include <asm/msr.h> #include <asm/io.h> #include <asm/mmu_context.h> +#include <asm/mtrr.h> +#include <asm/mce.h> #ifdef CONFIG_X86_LOCAL_APIC #include <asm/mpspec.h> #include <asm/apic.h> @@ -292,7 +294,7 @@ void __cpuinit generic_identify(struct cpuinfo_x86 * c) if (c->x86 >= 0x6) c->x86_model += ((tfms >> 16) & 0xF) << 4; c->x86_mask = tfms & 15; -#ifdef CONFIG_SMP +#ifdef CONFIG_X86_HT c->apicid = phys_pkg_id((ebx >> 24) & 0xFF, 0); #else c->apicid = (ebx >> 24) & 0xFF; @@ -317,7 +319,7 @@ void __cpuinit generic_identify(struct cpuinfo_x86 * c) early_intel_workaround(c); #ifdef CONFIG_X86_HT - phys_proc_id[smp_processor_id()] = (cpuid_ebx(1) >> 24) & 0xff; + c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff; #endif } @@ -475,11 +477,9 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) { u32 eax, ebx, ecx, edx; int index_msb, core_bits; - int cpu = smp_processor_id(); cpuid(1, &eax, &ebx, &ecx, &edx); - if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY)) return; @@ -490,16 +490,17 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) } else if (smp_num_siblings > 1 ) { if (smp_num_siblings > NR_CPUS) { - printk(KERN_WARNING "CPU: Unsupported number of the siblings %d", smp_num_siblings); + printk(KERN_WARNING "CPU: Unsupported number of the " + "siblings %d", smp_num_siblings); smp_num_siblings = 1; return; } index_msb = get_count_order(smp_num_siblings); - phys_proc_id[cpu] = phys_pkg_id((ebx >> 24) & 0xFF, index_msb); + c->phys_proc_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb); printk(KERN_INFO "CPU: Physical Processor ID: %d\n", - phys_proc_id[cpu]); + c->phys_proc_id); smp_num_siblings = smp_num_siblings / c->x86_max_cores; @@ -507,12 +508,12 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) core_bits = get_count_order(c->x86_max_cores); - cpu_core_id[cpu] = phys_pkg_id((ebx >> 24) & 0xFF, index_msb) & + c->cpu_core_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb) & ((1 << core_bits) - 1); if (c->x86_max_cores > 1) printk(KERN_INFO "CPU: Processor Core ID: %d\n", - cpu_core_id[cpu]); + c->cpu_core_id); } } #endif @@ -611,6 +612,12 @@ void __cpuinit cpu_init(void) set_in_cr4(X86_CR4_TSD); } + /* The CPU hotplug case */ + if (cpu_gdt_descr->address) { + gdt = (struct desc_struct *)cpu_gdt_descr->address; + memset(gdt, 0, PAGE_SIZE); + goto old_gdt; + } /* * This is a horrible hack to allocate the GDT. The problem * is that cpu_init() is called really early for the boot CPU @@ -629,7 +636,7 @@ void __cpuinit cpu_init(void) local_irq_enable(); } } - +old_gdt: /* * Initialize the per-CPU GDT with the boot GDT, * and set up the GDT descriptor: diff --git a/arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.c index 3852d0a4c1b5..567b39bea07e 100644 --- a/arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -24,7 +24,6 @@ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> @@ -48,12 +47,13 @@ MODULE_LICENSE("GPL"); struct cpufreq_acpi_io { - struct acpi_processor_performance acpi_data; + struct acpi_processor_performance *acpi_data; struct cpufreq_frequency_table *freq_table; unsigned int resume; }; static struct cpufreq_acpi_io *acpi_io_data[NR_CPUS]; +static struct acpi_processor_performance *acpi_perf_data[NR_CPUS]; static struct cpufreq_driver acpi_cpufreq_driver; @@ -104,64 +104,43 @@ acpi_processor_set_performance ( { u16 port = 0; u8 bit_width = 0; + int i = 0; int ret = 0; u32 value = 0; - int i = 0; - struct cpufreq_freqs cpufreq_freqs; - cpumask_t saved_mask; int retval; + struct acpi_processor_performance *perf; dprintk("acpi_processor_set_performance\n"); - /* - * TBD: Use something other than set_cpus_allowed. - * As set_cpus_allowed is a bit racy, - * with any other set_cpus_allowed for this process. - */ - saved_mask = current->cpus_allowed; - set_cpus_allowed(current, cpumask_of_cpu(cpu)); - if (smp_processor_id() != cpu) { - return (-EAGAIN); - } - - if (state == data->acpi_data.state) { + retval = 0; + perf = data->acpi_data; + if (state == perf->state) { if (unlikely(data->resume)) { dprintk("Called after resume, resetting to P%d\n", state); data->resume = 0; } else { dprintk("Already at target state (P%d)\n", state); - retval = 0; - goto migrate_end; + return (retval); } } - dprintk("Transitioning from P%d to P%d\n", - data->acpi_data.state, state); - - /* cpufreq frequency struct */ - cpufreq_freqs.cpu = cpu; - cpufreq_freqs.old = data->freq_table[data->acpi_data.state].frequency; - cpufreq_freqs.new = data->freq_table[state].frequency; - - /* notify cpufreq */ - cpufreq_notify_transition(&cpufreq_freqs, CPUFREQ_PRECHANGE); + dprintk("Transitioning from P%d to P%d\n", perf->state, state); /* * First we write the target state's 'control' value to the * control_register. */ - port = data->acpi_data.control_register.address; - bit_width = data->acpi_data.control_register.bit_width; - value = (u32) data->acpi_data.states[state].control; + port = perf->control_register.address; + bit_width = perf->control_register.bit_width; + value = (u32) perf->states[state].control; dprintk("Writing 0x%08x to port 0x%04x\n", value, port); ret = acpi_processor_write_port(port, bit_width, value); if (ret) { dprintk("Invalid port width 0x%04x\n", bit_width); - retval = ret; - goto migrate_end; + return (ret); } /* @@ -177,49 +156,35 @@ acpi_processor_set_performance ( * before giving up. */ - port = data->acpi_data.status_register.address; - bit_width = data->acpi_data.status_register.bit_width; + port = perf->status_register.address; + bit_width = perf->status_register.bit_width; dprintk("Looking for 0x%08x from port 0x%04x\n", - (u32) data->acpi_data.states[state].status, port); + (u32) perf->states[state].status, port); - for (i=0; i<100; i++) { + for (i = 0; i < 100; i++) { ret = acpi_processor_read_port(port, bit_width, &value); if (ret) { dprintk("Invalid port width 0x%04x\n", bit_width); - retval = ret; - goto migrate_end; + return (ret); } - if (value == (u32) data->acpi_data.states[state].status) + if (value == (u32) perf->states[state].status) break; udelay(10); } } else { - i = 0; - value = (u32) data->acpi_data.states[state].status; + value = (u32) perf->states[state].status; } - /* notify cpufreq */ - cpufreq_notify_transition(&cpufreq_freqs, CPUFREQ_POSTCHANGE); - - if (unlikely(value != (u32) data->acpi_data.states[state].status)) { - unsigned int tmp = cpufreq_freqs.new; - cpufreq_freqs.new = cpufreq_freqs.old; - cpufreq_freqs.old = tmp; - cpufreq_notify_transition(&cpufreq_freqs, CPUFREQ_PRECHANGE); - cpufreq_notify_transition(&cpufreq_freqs, CPUFREQ_POSTCHANGE); + if (unlikely(value != (u32) perf->states[state].status)) { printk(KERN_WARNING "acpi-cpufreq: Transition failed\n"); retval = -ENODEV; - goto migrate_end; + return (retval); } dprintk("Transition successful after %d microseconds\n", i * 10); - data->acpi_data.state = state; - - retval = 0; -migrate_end: - set_cpus_allowed(current, saved_mask); + perf->state = state; return (retval); } @@ -231,8 +196,17 @@ acpi_cpufreq_target ( unsigned int relation) { struct cpufreq_acpi_io *data = acpi_io_data[policy->cpu]; + struct acpi_processor_performance *perf; + struct cpufreq_freqs freqs; + cpumask_t online_policy_cpus; + cpumask_t saved_mask; + cpumask_t set_mask; + cpumask_t covered_cpus; + unsigned int cur_state = 0; unsigned int next_state = 0; unsigned int result = 0; + unsigned int j; + unsigned int tmp; dprintk("acpi_cpufreq_setpolicy\n"); @@ -241,11 +215,95 @@ acpi_cpufreq_target ( target_freq, relation, &next_state); - if (result) + if (unlikely(result)) return (result); - result = acpi_processor_set_performance (data, policy->cpu, next_state); + perf = data->acpi_data; + cur_state = perf->state; + freqs.old = data->freq_table[cur_state].frequency; + freqs.new = data->freq_table[next_state].frequency; + +#ifdef CONFIG_HOTPLUG_CPU + /* cpufreq holds the hotplug lock, so we are safe from here on */ + cpus_and(online_policy_cpus, cpu_online_map, policy->cpus); +#else + online_policy_cpus = policy->cpus; +#endif + + for_each_cpu_mask(j, online_policy_cpus) { + freqs.cpu = j; + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + } + /* + * We need to call driver->target() on all or any CPU in + * policy->cpus, depending on policy->shared_type. + */ + saved_mask = current->cpus_allowed; + cpus_clear(covered_cpus); + for_each_cpu_mask(j, online_policy_cpus) { + /* + * Support for SMP systems. + * Make sure we are running on CPU that wants to change freq + */ + cpus_clear(set_mask); + if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) + cpus_or(set_mask, set_mask, online_policy_cpus); + else + cpu_set(j, set_mask); + + set_cpus_allowed(current, set_mask); + if (unlikely(!cpu_isset(smp_processor_id(), set_mask))) { + dprintk("couldn't limit to CPUs in this domain\n"); + result = -EAGAIN; + break; + } + + result = acpi_processor_set_performance (data, j, next_state); + if (result) { + result = -EAGAIN; + break; + } + + if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) + break; + + cpu_set(j, covered_cpus); + } + + for_each_cpu_mask(j, online_policy_cpus) { + freqs.cpu = j; + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + } + + if (unlikely(result)) { + /* + * We have failed halfway through the frequency change. + * We have sent callbacks to online_policy_cpus and + * acpi_processor_set_performance() has been called on + * coverd_cpus. Best effort undo.. + */ + + if (!cpus_empty(covered_cpus)) { + for_each_cpu_mask(j, covered_cpus) { + policy->cpu = j; + acpi_processor_set_performance (data, + j, + cur_state); + } + } + + tmp = freqs.new; + freqs.new = freqs.old; + freqs.old = tmp; + for_each_cpu_mask(j, online_policy_cpus) { + freqs.cpu = j; + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + } + } + + set_cpus_allowed(current, saved_mask); return (result); } @@ -271,30 +329,65 @@ acpi_cpufreq_guess_freq ( struct cpufreq_acpi_io *data, unsigned int cpu) { + struct acpi_processor_performance *perf = data->acpi_data; + if (cpu_khz) { /* search the closest match to cpu_khz */ unsigned int i; unsigned long freq; - unsigned long freqn = data->acpi_data.states[0].core_frequency * 1000; + unsigned long freqn = perf->states[0].core_frequency * 1000; - for (i=0; i < (data->acpi_data.state_count - 1); i++) { + for (i = 0; i < (perf->state_count - 1); i++) { freq = freqn; - freqn = data->acpi_data.states[i+1].core_frequency * 1000; + freqn = perf->states[i+1].core_frequency * 1000; if ((2 * cpu_khz) > (freqn + freq)) { - data->acpi_data.state = i; + perf->state = i; return (freq); } } - data->acpi_data.state = data->acpi_data.state_count - 1; + perf->state = perf->state_count - 1; return (freqn); - } else + } else { /* assume CPU is at P0... */ - data->acpi_data.state = 0; - return data->acpi_data.states[0].core_frequency * 1000; - + perf->state = 0; + return perf->states[0].core_frequency * 1000; + } } +/* + * acpi_cpufreq_early_init - initialize ACPI P-States library + * + * Initialize the ACPI P-States library (drivers/acpi/processor_perflib.c) + * in order to determine correct frequency and voltage pairings. We can + * do _PDC and _PSD and find out the processor dependency for the + * actual init that will happen later... + */ +static int acpi_cpufreq_early_init_acpi(void) +{ + struct acpi_processor_performance *data; + unsigned int i, j; + + dprintk("acpi_cpufreq_early_init\n"); + + for_each_possible_cpu(i) { + data = kzalloc(sizeof(struct acpi_processor_performance), + GFP_KERNEL); + if (!data) { + for_each_possible_cpu(j) { + kfree(acpi_perf_data[j]); + acpi_perf_data[j] = NULL; + } + return (-ENOMEM); + } + acpi_perf_data[i] = data; + } + + /* Do initialization in ACPI core */ + acpi_processor_preregister_performance(acpi_perf_data); + return 0; +} + static int acpi_cpufreq_cpu_init ( struct cpufreq_policy *policy) @@ -304,41 +397,57 @@ acpi_cpufreq_cpu_init ( struct cpufreq_acpi_io *data; unsigned int result = 0; struct cpuinfo_x86 *c = &cpu_data[policy->cpu]; + struct acpi_processor_performance *perf; dprintk("acpi_cpufreq_cpu_init\n"); + if (!acpi_perf_data[cpu]) + return (-ENODEV); + data = kzalloc(sizeof(struct cpufreq_acpi_io), GFP_KERNEL); if (!data) return (-ENOMEM); + data->acpi_data = acpi_perf_data[cpu]; acpi_io_data[cpu] = data; - result = acpi_processor_register_performance(&data->acpi_data, cpu); + result = acpi_processor_register_performance(data->acpi_data, cpu); if (result) goto err_free; + perf = data->acpi_data; + policy->shared_type = perf->shared_type; + /* + * Will let policy->cpus know about dependency only when software + * coordination is required. + */ + if (policy->shared_type == CPUFREQ_SHARED_TYPE_ALL || + policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) + policy->cpus = perf->shared_cpu_map; + if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) { acpi_cpufreq_driver.flags |= CPUFREQ_CONST_LOOPS; } /* capability check */ - if (data->acpi_data.state_count <= 1) { + if (perf->state_count <= 1) { dprintk("No P-States\n"); result = -ENODEV; goto err_unreg; } - if ((data->acpi_data.control_register.space_id != ACPI_ADR_SPACE_SYSTEM_IO) || - (data->acpi_data.status_register.space_id != ACPI_ADR_SPACE_SYSTEM_IO)) { + + if ((perf->control_register.space_id != ACPI_ADR_SPACE_SYSTEM_IO) || + (perf->status_register.space_id != ACPI_ADR_SPACE_SYSTEM_IO)) { dprintk("Unsupported address space [%d, %d]\n", - (u32) (data->acpi_data.control_register.space_id), - (u32) (data->acpi_data.status_register.space_id)); + (u32) (perf->control_register.space_id), + (u32) (perf->status_register.space_id)); result = -ENODEV; goto err_unreg; } /* alloc freq_table */ - data->freq_table = kmalloc(sizeof(struct cpufreq_frequency_table) * (data->acpi_data.state_count + 1), GFP_KERNEL); + data->freq_table = kmalloc(sizeof(struct cpufreq_frequency_table) * (perf->state_count + 1), GFP_KERNEL); if (!data->freq_table) { result = -ENOMEM; goto err_unreg; @@ -346,9 +455,9 @@ acpi_cpufreq_cpu_init ( /* detect transition latency */ policy->cpuinfo.transition_latency = 0; - for (i=0; i<data->acpi_data.state_count; i++) { - if ((data->acpi_data.states[i].transition_latency * 1000) > policy->cpuinfo.transition_latency) - policy->cpuinfo.transition_latency = data->acpi_data.states[i].transition_latency * 1000; + for (i=0; i<perf->state_count; i++) { + if ((perf->states[i].transition_latency * 1000) > policy->cpuinfo.transition_latency) + policy->cpuinfo.transition_latency = perf->states[i].transition_latency * 1000; } policy->governor = CPUFREQ_DEFAULT_GOVERNOR; @@ -356,11 +465,11 @@ acpi_cpufreq_cpu_init ( policy->cur = acpi_cpufreq_guess_freq(data, policy->cpu); /* table init */ - for (i=0; i<=data->acpi_data.state_count; i++) + for (i=0; i<=perf->state_count; i++) { data->freq_table[i].index = i; - if (i<data->acpi_data.state_count) - data->freq_table[i].frequency = data->acpi_data.states[i].core_frequency * 1000; + if (i<perf->state_count) + data->freq_table[i].frequency = perf->states[i].core_frequency * 1000; else data->freq_table[i].frequency = CPUFREQ_TABLE_END; } @@ -375,12 +484,12 @@ acpi_cpufreq_cpu_init ( printk(KERN_INFO "acpi-cpufreq: CPU%u - ACPI performance management activated.\n", cpu); - for (i = 0; i < data->acpi_data.state_count; i++) + for (i = 0; i < perf->state_count; i++) dprintk(" %cP%d: %d MHz, %d mW, %d uS\n", - (i == data->acpi_data.state?'*':' '), i, - (u32) data->acpi_data.states[i].core_frequency, - (u32) data->acpi_data.states[i].power, - (u32) data->acpi_data.states[i].transition_latency); + (i == perf->state?'*':' '), i, + (u32) perf->states[i].core_frequency, + (u32) perf->states[i].power, + (u32) perf->states[i].transition_latency); cpufreq_frequency_table_get_attr(data->freq_table, policy->cpu); @@ -395,7 +504,7 @@ acpi_cpufreq_cpu_init ( err_freqfree: kfree(data->freq_table); err_unreg: - acpi_processor_unregister_performance(&data->acpi_data, cpu); + acpi_processor_unregister_performance(perf, cpu); err_free: kfree(data); acpi_io_data[cpu] = NULL; @@ -416,7 +525,7 @@ acpi_cpufreq_cpu_exit ( if (data) { cpufreq_frequency_table_put_attr(policy->cpu); acpi_io_data[policy->cpu] = NULL; - acpi_processor_unregister_performance(&data->acpi_data, policy->cpu); + acpi_processor_unregister_performance(data->acpi_data, policy->cpu); kfree(data); } @@ -444,14 +553,15 @@ static struct freq_attr* acpi_cpufreq_attr[] = { }; static struct cpufreq_driver acpi_cpufreq_driver = { - .verify = acpi_cpufreq_verify, - .target = acpi_cpufreq_target, - .init = acpi_cpufreq_cpu_init, - .exit = acpi_cpufreq_cpu_exit, - .resume = acpi_cpufreq_resume, - .name = "acpi-cpufreq", - .owner = THIS_MODULE, - .attr = acpi_cpufreq_attr, + .verify = acpi_cpufreq_verify, + .target = acpi_cpufreq_target, + .init = acpi_cpufreq_cpu_init, + .exit = acpi_cpufreq_cpu_exit, + .resume = acpi_cpufreq_resume, + .name = "acpi-cpufreq", + .owner = THIS_MODULE, + .attr = acpi_cpufreq_attr, + .flags = CPUFREQ_STICKY, }; @@ -462,7 +572,10 @@ acpi_cpufreq_init (void) dprintk("acpi_cpufreq_init\n"); - result = cpufreq_register_driver(&acpi_cpufreq_driver); + result = acpi_cpufreq_early_init_acpi(); + + if (!result) + result = cpufreq_register_driver(&acpi_cpufreq_driver); return (result); } @@ -471,10 +584,15 @@ acpi_cpufreq_init (void) static void __exit acpi_cpufreq_exit (void) { + unsigned int i; dprintk("acpi_cpufreq_exit\n"); cpufreq_unregister_driver(&acpi_cpufreq_driver); + for_each_possible_cpu(i) { + kfree(acpi_perf_data[i]); + acpi_perf_data[i] = NULL; + } return; } diff --git a/arch/i386/kernel/cpu/cpufreq/cpufreq-nforce2.c b/arch/i386/kernel/cpu/cpufreq/cpufreq-nforce2.c index f275e0d4aee5..0d49d73d1b71 100644 --- a/arch/i386/kernel/cpu/cpufreq/cpufreq-nforce2.c +++ b/arch/i386/kernel/cpu/cpufreq/cpufreq-nforce2.c @@ -1,5 +1,5 @@ /* - * (C) 2004 Sebastian Witt <se.witt@gmx.net> + * (C) 2004-2006 Sebastian Witt <se.witt@gmx.net> * * Licensed under the terms of the GNU GPL License version 2. * Based upon reverse engineered information @@ -90,7 +90,7 @@ static int nforce2_calc_pll(unsigned int fsb) /* Try to calculate multiplier and divider up to 4 times */ while (((mul == 0) || (div == 0)) && (tried <= 3)) { - for (xdiv = 1; xdiv <= 0x80; xdiv++) + for (xdiv = 2; xdiv <= 0x80; xdiv++) for (xmul = 1; xmul <= 0xfe; xmul++) if (nforce2_calc_fsb(NFORCE2_PLL(xmul, xdiv)) == fsb + tried) { @@ -117,8 +117,7 @@ static void nforce2_write_pll(int pll) int temp; /* Set the pll addr. to 0x00 */ - temp = 0x00; - pci_write_config_dword(nforce2_chipset_dev, NFORCE2_PLLADR, temp); + pci_write_config_dword(nforce2_chipset_dev, NFORCE2_PLLADR, 0); /* Now write the value in all 64 registers */ for (temp = 0; temp <= 0x3f; temp++) @@ -266,7 +265,7 @@ static int nforce2_target(struct cpufreq_policy *policy, if (freqs.old == freqs.new) return 0; - dprintk(KERN_INFO "cpufreq: Old CPU frequency %d kHz, new %d kHz\n", + dprintk("Old CPU frequency %d kHz, new %d kHz\n", freqs.old, freqs.new); cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); @@ -278,7 +277,7 @@ static int nforce2_target(struct cpufreq_policy *policy, printk(KERN_ERR "cpufreq: Changing FSB to %d failed\n", target_fsb); else - dprintk(KERN_INFO "cpufreq: Changed FSB successfully to %d\n", + dprintk("Changed FSB successfully to %d\n", target_fsb); /* Enable IRQs */ diff --git a/arch/i386/kernel/cpu/cpufreq/longhaul.c b/arch/i386/kernel/cpu/cpufreq/longhaul.c index 8ef38544453c..146f607e9c44 100644 --- a/arch/i386/kernel/cpu/cpufreq/longhaul.c +++ b/arch/i386/kernel/cpu/cpufreq/longhaul.c @@ -77,13 +77,17 @@ static char speedbuffer[8]; static char *print_speed(int speed) { - if (speed > 1000) { - if (speed%1000 == 0) - sprintf (speedbuffer, "%dGHz", speed/1000); - else - sprintf (speedbuffer, "%d.%dGHz", speed/1000, (speed%1000)/100); - } else - sprintf (speedbuffer, "%dMHz", speed); + if (speed < 1000) { + snprintf(speedbuffer, sizeof(speedbuffer),"%dMHz", speed); + return speedbuffer; + } + + if (speed%1000 == 0) + snprintf(speedbuffer, sizeof(speedbuffer), + "%dGHz", speed/1000); + else + snprintf(speedbuffer, sizeof(speedbuffer), + "%d.%dGHz", speed/1000, (speed%1000)/100); return speedbuffer; } @@ -675,7 +679,7 @@ static int __init longhaul_init(void) static void __exit longhaul_exit(void) { - int i=0; + int i; for (i=0; i < numscales; i++) { if (clock_ratio[i] == maxmult) { diff --git a/arch/i386/kernel/cpu/cpufreq/longrun.c b/arch/i386/kernel/cpu/cpufreq/longrun.c index e3868de4dc2e..b2689514295a 100644 --- a/arch/i386/kernel/cpu/cpufreq/longrun.c +++ b/arch/i386/kernel/cpu/cpufreq/longrun.c @@ -223,7 +223,6 @@ static unsigned int __init longrun_determine_freqs(unsigned int *low_freq, /* set to 0 to try_hi perf_pctg */ msr_lo &= 0xFFFFFF80; msr_hi &= 0xFFFFFF80; - msr_lo |= 0; msr_hi |= try_hi; wrmsr(MSR_TMTA_LONGRUN_CTRL, msr_lo, msr_hi); diff --git a/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c b/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c index ab6504efd801..304d2eaa4a1b 100644 --- a/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c +++ b/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c @@ -20,7 +20,6 @@ * */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k7.c b/arch/i386/kernel/cpu/cpufreq/powernow-k7.c index 2bf4237cb94e..54382760983a 100644 --- a/arch/i386/kernel/cpu/cpufreq/powernow-k7.c +++ b/arch/i386/kernel/cpu/cpufreq/powernow-k7.c @@ -12,7 +12,6 @@ * - We disable half multipliers if ACPI is used on A0 stepping CPUs. */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/moduleparam.h> @@ -452,23 +451,23 @@ static int powernow_decode_bios (int maxfid, int startvid) pst = (struct pst_s *) p; - for (i = 0 ; i <psb->numpst; i++) { + for (j=0; j<psb->numpst; j++) { pst = (struct pst_s *) p; number_scales = pst->numpstates; if ((etuple == pst->cpuid) && check_fsb(pst->fsbspeed) && (maxfid==pst->maxfid) && (startvid==pst->startvid)) { - dprintk ("PST:%d (@%p)\n", i, pst); + dprintk ("PST:%d (@%p)\n", j, pst); dprintk (" cpuid: 0x%x fsb: %d maxFID: 0x%x startvid: 0x%x\n", pst->cpuid, pst->fsbspeed, pst->maxfid, pst->startvid); ret = get_ranges ((char *) pst + sizeof (struct pst_s)); return ret; - } else { + unsigned int k; p = (char *) pst + sizeof (struct pst_s); - for (j=0 ; j < number_scales; j++) + for (k=0; k<number_scales; k++) p+=2; } } @@ -581,10 +580,7 @@ static int __init powernow_cpu_init (struct cpufreq_policy *policy) rdmsrl (MSR_K7_FID_VID_STATUS, fidvidstatus.val); - /* recalibrate cpu_khz */ - result = recalibrate_cpu_khz(); - if (result) - return result; + recalibrate_cpu_khz(); fsb = (10 * cpu_khz) / fid_codes[fidvidstatus.bits.CFID]; if (!fsb) { diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c index 71fffa174425..2d6491672559 100644 --- a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c @@ -1,5 +1,5 @@ /* - * (c) 2003, 2004, 2005 Advanced Micro Devices, Inc. + * (c) 2003-2006 Advanced Micro Devices, Inc. * Your use of this code is subject to the terms and conditions of the * GNU general public license version 2. See "COPYING" or * http://www.gnu.org/licenses/gpl.html @@ -14,13 +14,13 @@ * Based upon datasheets & sample CPUs kindly provided by AMD. * * Valuable input gratefully received from Dave Jones, Pavel Machek, - * Dominik Brodowski, and others. + * Dominik Brodowski, Jacob Shin, and others. * Originally developed by Paul Devriendt. * Processor information obtained from Chapter 9 (Power and Thermal Management) * of the "BIOS and Kernel Developer's Guide for the AMD Athlon 64 and AMD * Opteron Processors" available for download from www.amd.com * - * Tables for specific CPUs can be infrerred from + * Tables for specific CPUs can be inferred from * http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/30430.pdf */ @@ -46,7 +46,7 @@ #define PFX "powernow-k8: " #define BFX PFX "BIOS error: " -#define VERSION "version 1.60.2" +#define VERSION "version 2.00.00" #include "powernow-k8.h" /* serialize freq changes */ @@ -54,6 +54,8 @@ static DEFINE_MUTEX(fidvid_mutex); static struct powernow_k8_data *powernow_data[NR_CPUS]; +static int cpu_family = CPU_OPTERON; + #ifndef CONFIG_SMP static cpumask_t cpu_core_map[1]; #endif @@ -64,16 +66,36 @@ static u32 find_freq_from_fid(u32 fid) return 800 + (fid * 100); } + /* Return a frequency in KHz, given an input fid */ static u32 find_khz_freq_from_fid(u32 fid) { return 1000 * find_freq_from_fid(fid); } -/* Return a voltage in miliVolts, given an input vid */ -static u32 find_millivolts_from_vid(struct powernow_k8_data *data, u32 vid) +/* Return a frequency in MHz, given an input fid and did */ +static u32 find_freq_from_fiddid(u32 fid, u32 did) +{ + return 100 * (fid + 0x10) >> did; +} + +static u32 find_khz_freq_from_fiddid(u32 fid, u32 did) { - return 1550-vid*25; + return 1000 * find_freq_from_fiddid(fid, did); +} + +static u32 find_fid_from_pstate(u32 pstate) +{ + u32 hi, lo; + rdmsr(MSR_PSTATE_DEF_BASE + pstate, lo, hi); + return lo & HW_PSTATE_FID_MASK; +} + +static u32 find_did_from_pstate(u32 pstate) +{ + u32 hi, lo; + rdmsr(MSR_PSTATE_DEF_BASE + pstate, lo, hi); + return (lo & HW_PSTATE_DID_MASK) >> HW_PSTATE_DID_SHIFT; } /* Return the vco fid for an input fid @@ -98,6 +120,9 @@ static int pending_bit_stuck(void) { u32 lo, hi; + if (cpu_family == CPU_HW_PSTATE) + return 0; + rdmsr(MSR_FIDVID_STATUS, lo, hi); return lo & MSR_S_LO_CHANGE_PENDING ? 1 : 0; } @@ -111,6 +136,14 @@ static int query_current_values_with_pending_wait(struct powernow_k8_data *data) u32 lo, hi; u32 i = 0; + if (cpu_family == CPU_HW_PSTATE) { + rdmsr(MSR_PSTATE_STATUS, lo, hi); + i = lo & HW_PSTATE_MASK; + rdmsr(MSR_PSTATE_DEF_BASE + i, lo, hi); + data->currfid = lo & HW_PSTATE_FID_MASK; + data->currdid = (lo & HW_PSTATE_DID_MASK) >> HW_PSTATE_DID_SHIFT; + return 0; + } do { if (i++ > 10000) { dprintk("detected change pending stuck\n"); @@ -175,7 +208,7 @@ static int write_new_fid(struct powernow_k8_data *data, u32 fid) do { wrmsr(MSR_FIDVID_CTL, lo, data->plllock * PLL_LOCK_CONVERSION); if (i++ > 100) { - printk(KERN_ERR PFX "internal error - pending bit very stuck - no further pstate changes possible\n"); + printk(KERN_ERR PFX "Hardware error - pending bit very stuck - no further pstate changes possible\n"); return 1; } } while (query_current_values_with_pending_wait(data)); @@ -255,7 +288,15 @@ static int decrease_vid_code_by_step(struct powernow_k8_data *data, u32 reqvid, return 0; } -/* Change the fid and vid, by the 3 phases. */ +/* Change hardware pstate by single MSR write */ +static int transition_pstate(struct powernow_k8_data *data, u32 pstate) +{ + wrmsr(MSR_PSTATE_CTRL, pstate, 0); + data->currfid = find_fid_from_pstate(pstate); + return 0; +} + +/* Change Opteron/Athlon64 fid and vid, by the 3 phases. */ static int transition_fid_vid(struct powernow_k8_data *data, u32 reqfid, u32 reqvid) { if (core_voltage_pre_transition(data, reqvid)) @@ -474,26 +515,35 @@ static int check_supported_cpu(unsigned int cpu) goto out; eax = cpuid_eax(CPUID_PROCESSOR_SIGNATURE); - if ((eax & CPUID_XFAM) != CPUID_XFAM_K8) + if (((eax & CPUID_XFAM) != CPUID_XFAM_K8) && + ((eax & CPUID_XFAM) < CPUID_XFAM_10H)) goto out; - if (((eax & CPUID_USE_XFAM_XMOD) != CPUID_USE_XFAM_XMOD) || - ((eax & CPUID_XMOD) > CPUID_XMOD_REV_G)) { - printk(KERN_INFO PFX "Processor cpuid %x not supported\n", eax); - goto out; - } + if ((eax & CPUID_XFAM) == CPUID_XFAM_K8) { + if (((eax & CPUID_USE_XFAM_XMOD) != CPUID_USE_XFAM_XMOD) || + ((eax & CPUID_XMOD) > CPUID_XMOD_REV_G)) { + printk(KERN_INFO PFX "Processor cpuid %x not supported\n", eax); + goto out; + } - eax = cpuid_eax(CPUID_GET_MAX_CAPABILITIES); - if (eax < CPUID_FREQ_VOLT_CAPABILITIES) { - printk(KERN_INFO PFX - "No frequency change capabilities detected\n"); - goto out; - } + eax = cpuid_eax(CPUID_GET_MAX_CAPABILITIES); + if (eax < CPUID_FREQ_VOLT_CAPABILITIES) { + printk(KERN_INFO PFX + "No frequency change capabilities detected\n"); + goto out; + } - cpuid(CPUID_FREQ_VOLT_CAPABILITIES, &eax, &ebx, &ecx, &edx); - if ((edx & P_STATE_TRANSITION_CAPABLE) != P_STATE_TRANSITION_CAPABLE) { - printk(KERN_INFO PFX "Power state transitions not supported\n"); - goto out; + cpuid(CPUID_FREQ_VOLT_CAPABILITIES, &eax, &ebx, &ecx, &edx); + if ((edx & P_STATE_TRANSITION_CAPABLE) != P_STATE_TRANSITION_CAPABLE) { + printk(KERN_INFO PFX "Power state transitions not supported\n"); + goto out; + } + } else { /* must be a HW Pstate capable processor */ + cpuid(CPUID_FREQ_VOLT_CAPABILITIES, &eax, &ebx, &ecx, &edx); + if ((edx & USE_HW_PSTATE) == USE_HW_PSTATE) + cpu_family = CPU_HW_PSTATE; + else + goto out; } rc = 1; @@ -547,12 +597,18 @@ static void print_basics(struct powernow_k8_data *data) { int j; for (j = 0; j < data->numps; j++) { - if (data->powernow_table[j].frequency != CPUFREQ_ENTRY_INVALID) - printk(KERN_INFO PFX " %d : fid 0x%x (%d MHz), vid 0x%x (%d mV)\n", j, + if (data->powernow_table[j].frequency != CPUFREQ_ENTRY_INVALID) { + if (cpu_family == CPU_HW_PSTATE) { + printk(KERN_INFO PFX " %d : fid 0x%x gid 0x%x (%d MHz)\n", j, (data->powernow_table[j].index & 0xff00) >> 8, + (data->powernow_table[j].index & 0xff0000) >> 16, + data->powernow_table[j].frequency/1000); + } else { + printk(KERN_INFO PFX " %d : fid 0x%x (%d MHz), vid 0x%x\n", j, data->powernow_table[j].index & 0xff, data->powernow_table[j].frequency/1000, - data->powernow_table[j].index >> 8, - find_millivolts_from_vid(data, data->powernow_table[j].index >> 8)); + data->powernow_table[j].index >> 8); + } + } } if (data->batps) printk(KERN_INFO PFX "Only %d pstates on battery\n", data->batps); @@ -702,7 +758,7 @@ static int find_psb_table(struct powernow_k8_data *data) #ifdef CONFIG_X86_POWERNOW_K8_ACPI static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index) { - if (!data->acpi_data.state_count) + if (!data->acpi_data.state_count || (cpu_family == CPU_HW_PSTATE)) return; data->irt = (data->acpi_data.states[index].control >> IRT_SHIFT) & IRT_MASK; @@ -715,9 +771,8 @@ static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) { - int i; - int cntlofreq = 0; struct cpufreq_frequency_table *powernow_table; + int ret_val; if (acpi_processor_register_performance(&data->acpi_data, data->cpu)) { dprintk("register performance failed: bad ACPI data\n"); @@ -746,13 +801,92 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) goto err_out; } + if (cpu_family == CPU_HW_PSTATE) + ret_val = fill_powernow_table_pstate(data, powernow_table); + else + ret_val = fill_powernow_table_fidvid(data, powernow_table); + if (ret_val) + goto err_out_mem; + + powernow_table[data->acpi_data.state_count].frequency = CPUFREQ_TABLE_END; + powernow_table[data->acpi_data.state_count].index = 0; + data->powernow_table = powernow_table; + + /* fill in data */ + data->numps = data->acpi_data.state_count; + print_basics(data); + powernow_k8_acpi_pst_values(data, 0); + + /* notify BIOS that we exist */ + acpi_processor_notify_smm(THIS_MODULE); + + return 0; + +err_out_mem: + kfree(powernow_table); + +err_out: + acpi_processor_unregister_performance(&data->acpi_data, data->cpu); + + /* data->acpi_data.state_count informs us at ->exit() whether ACPI was used */ + data->acpi_data.state_count = 0; + + return -ENODEV; +} + +static int fill_powernow_table_pstate(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table) +{ + int i; + + for (i = 0; i < data->acpi_data.state_count; i++) { + u32 index; + u32 hi = 0, lo = 0; + u32 fid; + u32 did; + + index = data->acpi_data.states[i].control & HW_PSTATE_MASK; + if (index > MAX_HW_PSTATE) { + printk(KERN_ERR PFX "invalid pstate %d - bad value %d.\n", i, index); + printk(KERN_ERR PFX "Please report to BIOS manufacturer\n"); + } + rdmsr(MSR_PSTATE_DEF_BASE + index, lo, hi); + if (!(hi & HW_PSTATE_VALID_MASK)) { + dprintk("invalid pstate %d, ignoring\n", index); + powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID; + continue; + } + + fid = lo & HW_PSTATE_FID_MASK; + did = (lo & HW_PSTATE_DID_MASK) >> HW_PSTATE_DID_SHIFT; + + dprintk(" %d : fid 0x%x, did 0x%x\n", index, fid, did); + + powernow_table[i].index = index | (fid << HW_FID_INDEX_SHIFT) | (did << HW_DID_INDEX_SHIFT); + + powernow_table[i].frequency = find_khz_freq_from_fiddid(fid, did); + + if (powernow_table[i].frequency != (data->acpi_data.states[i].core_frequency * 1000)) { + printk(KERN_INFO PFX "invalid freq entries %u kHz vs. %u kHz\n", + powernow_table[i].frequency, + (unsigned int) (data->acpi_data.states[i].core_frequency * 1000)); + powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID; + continue; + } + } + return 0; +} + +static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table) +{ + int i; + int cntlofreq = 0; for (i = 0; i < data->acpi_data.state_count; i++) { u32 fid; u32 vid; if (data->exttype) { - fid = data->acpi_data.states[i].status & FID_MASK; - vid = (data->acpi_data.states[i].status >> VID_SHIFT) & VID_MASK; + fid = data->acpi_data.states[i].status & EXT_FID_MASK; + vid = (data->acpi_data.states[i].status >> VID_SHIFT) & EXT_VID_MASK; } else { fid = data->acpi_data.states[i].control & FID_MASK; vid = (data->acpi_data.states[i].control >> VID_SHIFT) & VID_MASK; @@ -786,7 +920,7 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) if ((powernow_table[i].frequency != powernow_table[cntlofreq].frequency) || (powernow_table[i].index != powernow_table[cntlofreq].index)) { printk(KERN_ERR PFX "Too many lo freq table entries\n"); - goto err_out_mem; + return 1; } dprintk("double low frequency table entry, ignoring it.\n"); @@ -804,31 +938,7 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) continue; } } - - powernow_table[data->acpi_data.state_count].frequency = CPUFREQ_TABLE_END; - powernow_table[data->acpi_data.state_count].index = 0; - data->powernow_table = powernow_table; - - /* fill in data */ - data->numps = data->acpi_data.state_count; - print_basics(data); - powernow_k8_acpi_pst_values(data, 0); - - /* notify BIOS that we exist */ - acpi_processor_notify_smm(THIS_MODULE); - return 0; - -err_out_mem: - kfree(powernow_table); - -err_out: - acpi_processor_unregister_performance(&data->acpi_data, data->cpu); - - /* data->acpi_data.state_count informs us at ->exit() whether ACPI was used */ - data->acpi_data.state_count = 0; - - return -ENODEV; } static void powernow_k8_cpu_exit_acpi(struct powernow_k8_data *data) @@ -844,20 +954,20 @@ static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned #endif /* CONFIG_X86_POWERNOW_K8_ACPI */ /* Take a frequency, and issue the fid/vid transition command */ -static int transition_frequency(struct powernow_k8_data *data, unsigned int index) +static int transition_frequency_fidvid(struct powernow_k8_data *data, unsigned int index) { - u32 fid; - u32 vid; + u32 fid = 0; + u32 vid = 0; int res, i; struct cpufreq_freqs freqs; dprintk("cpu %d transition to index %u\n", smp_processor_id(), index); + /* fid/vid correctness check for k8 */ /* fid are the lower 8 bits of the index we stored into - * the cpufreq frequency table in find_psb_table, vid are - * the upper 8 bits. + * the cpufreq frequency table in find_psb_table, vid + * are the upper 8 bits. */ - fid = data->powernow_table[index].index & 0xFF; vid = (data->powernow_table[index].index & 0xFF00) >> 8; @@ -881,22 +991,58 @@ static int transition_frequency(struct powernow_k8_data *data, unsigned int inde dprintk("cpu %d, changing to fid 0x%x, vid 0x%x\n", smp_processor_id(), fid, vid); - - freqs.cpu = data->cpu; freqs.old = find_khz_freq_from_fid(data->currfid); freqs.new = find_khz_freq_from_fid(fid); - for_each_cpu_mask(i, cpu_core_map[data->cpu]) { + + for_each_cpu_mask(i, *(data->available_cores)) { freqs.cpu = i; cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); } res = transition_fid_vid(data, fid, vid); - freqs.new = find_khz_freq_from_fid(data->currfid); - for_each_cpu_mask(i, cpu_core_map[data->cpu]) { + + for_each_cpu_mask(i, *(data->available_cores)) { freqs.cpu = i; cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - } + } + return res; +} + +/* Take a frequency, and issue the hardware pstate transition command */ +static int transition_frequency_pstate(struct powernow_k8_data *data, unsigned int index) +{ + u32 fid = 0; + u32 did = 0; + u32 pstate = 0; + int res, i; + struct cpufreq_freqs freqs; + + dprintk("cpu %d transition to index %u\n", smp_processor_id(), index); + + /* get fid did for hardware pstate transition */ + pstate = index & HW_PSTATE_MASK; + if (pstate > MAX_HW_PSTATE) + return 0; + fid = (index & HW_FID_INDEX_MASK) >> HW_FID_INDEX_SHIFT; + did = (index & HW_DID_INDEX_MASK) >> HW_DID_INDEX_SHIFT; + freqs.old = find_khz_freq_from_fiddid(data->currfid, data->currdid); + freqs.new = find_khz_freq_from_fiddid(fid, did); + + for_each_cpu_mask(i, *(data->available_cores)) { + freqs.cpu = i; + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + } + + res = transition_pstate(data, pstate); + data->currfid = find_fid_from_pstate(pstate); + data->currdid = find_did_from_pstate(pstate); + freqs.new = find_khz_freq_from_fiddid(data->currfid, data->currdid); + + for_each_cpu_mask(i, *(data->available_cores)) { + freqs.cpu = i; + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + } return res; } @@ -933,18 +1079,21 @@ static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsi dprintk("targ: cpu %d, %d kHz, min %d, max %d, relation %d\n", pol->cpu, targfreq, pol->min, pol->max, relation); - if (query_current_values_with_pending_wait(data)) { - ret = -EIO; + if (query_current_values_with_pending_wait(data)) goto err_out; - } - dprintk("targ: curr fid 0x%x, vid 0x%x\n", + if (cpu_family == CPU_HW_PSTATE) + dprintk("targ: curr fid 0x%x, did 0x%x\n", + data->currfid, data->currvid); + else { + dprintk("targ: curr fid 0x%x, vid 0x%x\n", data->currfid, data->currvid); - if ((checkvid != data->currvid) || (checkfid != data->currfid)) { - printk(KERN_INFO PFX - "error - out of sync, fix 0x%x 0x%x, vid 0x%x 0x%x\n", - checkfid, data->currfid, checkvid, data->currvid); + if ((checkvid != data->currvid) || (checkfid != data->currfid)) { + printk(KERN_INFO PFX + "error - out of sync, fix 0x%x 0x%x, vid 0x%x 0x%x\n", + checkfid, data->currfid, checkvid, data->currvid); + } } if (cpufreq_frequency_table_target(pol, data->powernow_table, targfreq, relation, &newstate)) @@ -954,7 +1103,11 @@ static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsi powernow_k8_acpi_pst_values(data, newstate); - if (transition_frequency(data, newstate)) { + if (cpu_family == CPU_HW_PSTATE) + ret = transition_frequency_pstate(data, newstate); + else + ret = transition_frequency_fidvid(data, newstate); + if (ret) { printk(KERN_ERR PFX "transition frequency failed\n"); ret = 1; mutex_unlock(&fidvid_mutex); @@ -962,7 +1115,10 @@ static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsi } mutex_unlock(&fidvid_mutex); - pol->cur = find_khz_freq_from_fid(data->currfid); + if (cpu_family == CPU_HW_PSTATE) + pol->cur = find_khz_freq_from_fiddid(data->currfid, data->currdid); + else + pol->cur = find_khz_freq_from_fid(data->currfid); ret = 0; err_out: @@ -1007,14 +1163,13 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) * Use the PSB BIOS structure. This is only availabe on * an UP version, and is deprecated by AMD. */ - - if ((num_online_cpus() != 1) || (num_possible_cpus() != 1)) { + if (num_online_cpus() != 1) { printk(KERN_ERR PFX "MP systems not supported by PSB BIOS structure\n"); kfree(data); return -ENODEV; } if (pol->cpu != 0) { - printk(KERN_ERR PFX "init not cpu 0\n"); + printk(KERN_ERR PFX "No _PSS objects for CPU other than CPU0\n"); kfree(data); return -ENODEV; } @@ -1042,20 +1197,28 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) if (query_current_values_with_pending_wait(data)) goto err_out; - fidvid_msr_init(); + if (cpu_family == CPU_OPTERON) + fidvid_msr_init(); /* run on any CPU again */ set_cpus_allowed(current, oldmask); pol->governor = CPUFREQ_DEFAULT_GOVERNOR; - pol->cpus = cpu_core_map[pol->cpu]; + if (cpu_family == CPU_HW_PSTATE) + pol->cpus = cpumask_of_cpu(pol->cpu); + else + pol->cpus = cpu_core_map[pol->cpu]; + data->available_cores = &(pol->cpus); /* Take a crude guess here. * That guess was in microseconds, so multiply with 1000 */ pol->cpuinfo.transition_latency = (((data->rvo + 8) * data->vstable * VST_UNITS_20US) + (3 * (1 << data->irt) * 10)) * 1000; - pol->cur = find_khz_freq_from_fid(data->currfid); + if (cpu_family == CPU_HW_PSTATE) + pol->cur = find_khz_freq_from_fiddid(data->currfid, data->currdid); + else + pol->cur = find_khz_freq_from_fid(data->currfid); dprintk("policy current frequency %d kHz\n", pol->cur); /* min/max the cpu is capable of */ @@ -1069,8 +1232,12 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) cpufreq_frequency_table_get_attr(data->powernow_table, pol->cpu); - printk("cpu_init done, current fid 0x%x, vid 0x%x\n", - data->currfid, data->currvid); + if (cpu_family == CPU_HW_PSTATE) + dprintk("cpu_init done, current fid 0x%x, did 0x%x\n", + data->currfid, data->currdid); + else + dprintk("cpu_init done, current fid 0x%x, vid 0x%x\n", + data->currfid, data->currvid); powernow_data[pol->cpu] = data; @@ -1156,8 +1323,9 @@ static int __cpuinit powernowk8_init(void) } if (supported_cpus == num_online_cpus()) { - printk(KERN_INFO PFX "Found %d AMD Athlon 64 / Opteron " - "processors (" VERSION ")\n", supported_cpus); + printk(KERN_INFO PFX "Found %d %s " + "processors (" VERSION ")\n", supported_cpus, + boot_cpu_data.x86_model_id); return cpufreq_register_driver(&cpufreq_amd64_driver); } diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k8.h b/arch/i386/kernel/cpu/cpufreq/powernow-k8.h index 79a7c5c87edc..0fb2a3001ba5 100644 --- a/arch/i386/kernel/cpu/cpufreq/powernow-k8.h +++ b/arch/i386/kernel/cpu/cpufreq/powernow-k8.h @@ -1,5 +1,5 @@ /* - * (c) 2003, 2004, 2005 Advanced Micro Devices, Inc. + * (c) 2003-2006 Advanced Micro Devices, Inc. * Your use of this code is subject to the terms and conditions of the * GNU general public license version 2. See "COPYING" or * http://www.gnu.org/licenses/gpl.html @@ -21,8 +21,8 @@ struct powernow_k8_data { u32 plllock; /* pll lock time, units 1 us */ u32 exttype; /* extended interface = 1 */ - /* keep track of the current fid / vid */ - u32 currvid, currfid; + /* keep track of the current fid / vid or did */ + u32 currvid, currfid, currdid; /* the powernow_table includes all frequency and vid/fid pairings: * fid are the lower 8 bits of the index, vid are the upper 8 bits. @@ -34,6 +34,10 @@ struct powernow_k8_data { * used to determine valid frequency/vid/fid states */ struct acpi_processor_performance acpi_data; #endif + /* we need to keep track of associated cores, but let cpufreq + * handle hotplug events - so just point at cpufreq pol->cpus + * structure */ + cpumask_t *available_cores; }; @@ -43,6 +47,7 @@ struct powernow_k8_data { #define CPUID_XFAM_K8 0 #define CPUID_XMOD 0x000f0000 /* extended model */ #define CPUID_XMOD_REV_G 0x00060000 +#define CPUID_XFAM_10H 0x00100000 /* family 0x10 */ #define CPUID_USE_XFAM_XMOD 0x00000f00 #define CPUID_GET_MAX_CAPABILITIES 0x80000000 #define CPUID_FREQ_VOLT_CAPABILITIES 0x80000007 @@ -79,6 +84,32 @@ struct powernow_k8_data { #define MSR_S_HI_CURRENT_VID 0x0000003f #define MSR_C_HI_STP_GNT_BENIGN 0x00000001 + +/* Hardware Pstate _PSS and MSR definitions */ +#define USE_HW_PSTATE 0x00000080 +#define HW_PSTATE_FID_MASK 0x0000003f +#define HW_PSTATE_DID_MASK 0x000001c0 +#define HW_PSTATE_DID_SHIFT 6 +#define HW_PSTATE_MASK 0x00000007 +#define HW_PSTATE_VALID_MASK 0x80000000 +#define HW_FID_INDEX_SHIFT 8 +#define HW_FID_INDEX_MASK 0x0000ff00 +#define HW_DID_INDEX_SHIFT 16 +#define HW_DID_INDEX_MASK 0x00ff0000 +#define HW_WATTS_MASK 0xff +#define HW_PWR_DVR_MASK 0x300 +#define HW_PWR_DVR_SHIFT 8 +#define HW_PWR_MAX_MULT 3 +#define MAX_HW_PSTATE 8 /* hw pstate supports up to 8 */ +#define MSR_PSTATE_DEF_BASE 0xc0010064 /* base of Pstate MSRs */ +#define MSR_PSTATE_STATUS 0xc0010063 /* Pstate Status MSR */ +#define MSR_PSTATE_CTRL 0xc0010062 /* Pstate control MSR */ + +/* define the two driver architectures */ +#define CPU_OPTERON 0 +#define CPU_HW_PSTATE 1 + + /* * There are restrictions frequencies have to follow: * - only 1 entry in the low fid table ( <=1.4GHz ) @@ -138,7 +169,9 @@ struct powernow_k8_data { #define MVS_MASK 3 #define VST_MASK 0x7f #define VID_MASK 0x1f -#define FID_MASK 0x3f +#define FID_MASK 0x1f +#define EXT_VID_MASK 0x3f +#define EXT_FID_MASK 0x3f /* @@ -182,6 +215,9 @@ static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid); static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index); +static int fill_powernow_table_pstate(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table); +static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table); + #ifdef CONFIG_SMP static inline void define_siblings(int cpu, cpumask_t cpu_sharedcore_mask[]) { diff --git a/arch/i386/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/i386/kernel/cpu/cpufreq/speedstep-centrino.c index b0ff9075708c..b77f1358bd79 100644 --- a/arch/i386/kernel/cpu/cpufreq/speedstep-centrino.c +++ b/arch/i386/kernel/cpu/cpufreq/speedstep-centrino.c @@ -2,26 +2,21 @@ * cpufreq driver for Enhanced SpeedStep, as found in Intel's Pentium * M (part of the Centrino chipset). * + * Since the original Pentium M, most new Intel CPUs support Enhanced + * SpeedStep. + * * Despite the "SpeedStep" in the name, this is almost entirely unlike * traditional SpeedStep. * * Modelled on speedstep.c * * Copyright (C) 2003 Jeremy Fitzhardinge <jeremy@goop.org> - * - * WARNING WARNING WARNING - * - * This driver manipulates the PERF_CTL MSR, which is only somewhat - * documented. While it seems to work on my laptop, it has not been - * tested anywhere else, and it may not work for you, do strange - * things or simply crash. */ #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> #include <linux/cpufreq.h> -#include <linux/config.h> #include <linux/sched.h> /* current */ #include <linux/delay.h> #include <linux/compiler.h> @@ -36,7 +31,7 @@ #include <asm/cpufeature.h> #define PFX "speedstep-centrino: " -#define MAINTAINER "Jeremy Fitzhardinge <jeremy@goop.org>" +#define MAINTAINER "cpufreq@lists.linux.org.uk" #define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "speedstep-centrino", msg) @@ -250,7 +245,7 @@ static int centrino_cpu_init_table(struct cpufreq_policy *policy) if (model->cpu_id == NULL) { /* No match at all */ - dprintk(KERN_INFO PFX "no support for CPU model \"%s\": " + dprintk("no support for CPU model \"%s\": " "send /proc/cpuinfo to " MAINTAINER "\n", cpu->x86_model_id); return -ENOENT; @@ -258,10 +253,10 @@ static int centrino_cpu_init_table(struct cpufreq_policy *policy) if (model->op_points == NULL) { /* Matched a non-match */ - dprintk(KERN_INFO PFX "no table support for CPU model \"%s\"\n", + dprintk("no table support for CPU model \"%s\"\n", cpu->x86_model_id); #ifndef CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI - dprintk(KERN_INFO PFX "try compiling with CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI enabled\n"); + dprintk("try compiling with CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI enabled\n"); #endif return -ENOENT; } @@ -351,7 +346,36 @@ static unsigned int get_cur_freq(unsigned int cpu) #ifdef CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI -static struct acpi_processor_performance p; +static struct acpi_processor_performance *acpi_perf_data[NR_CPUS]; + +/* + * centrino_cpu_early_init_acpi - Do the preregistering with ACPI P-States + * library + * + * Before doing the actual init, we need to do _PSD related setup whenever + * supported by the BIOS. These are handled by this early_init routine. + */ +static int centrino_cpu_early_init_acpi(void) +{ + unsigned int i, j; + struct acpi_processor_performance *data; + + for_each_possible_cpu(i) { + data = kzalloc(sizeof(struct acpi_processor_performance), + GFP_KERNEL); + if (!data) { + for_each_possible_cpu(j) { + kfree(acpi_perf_data[j]); + acpi_perf_data[j] = NULL; + } + return (-ENOMEM); + } + acpi_perf_data[i] = data; + } + + acpi_processor_preregister_performance(acpi_perf_data); + return 0; +} /* * centrino_cpu_init_acpi - register with ACPI P-States library @@ -365,46 +389,57 @@ static int centrino_cpu_init_acpi(struct cpufreq_policy *policy) unsigned long cur_freq; int result = 0, i; unsigned int cpu = policy->cpu; + struct acpi_processor_performance *p; + + p = acpi_perf_data[cpu]; /* register with ACPI core */ - if (acpi_processor_register_performance(&p, cpu)) { - dprintk(KERN_INFO PFX "obtaining ACPI data failed\n"); + if (acpi_processor_register_performance(p, cpu)) { + dprintk(PFX "obtaining ACPI data failed\n"); return -EIO; } + policy->shared_type = p->shared_type; + /* + * Will let policy->cpus know about dependency only when software + * coordination is required. + */ + if (policy->shared_type == CPUFREQ_SHARED_TYPE_ALL || + policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) + policy->cpus = p->shared_cpu_map; /* verify the acpi_data */ - if (p.state_count <= 1) { + if (p->state_count <= 1) { dprintk("No P-States\n"); result = -ENODEV; goto err_unreg; } - if ((p.control_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE) || - (p.status_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE)) { + if ((p->control_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE) || + (p->status_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE)) { dprintk("Invalid control/status registers (%x - %x)\n", - p.control_register.space_id, p.status_register.space_id); + p->control_register.space_id, p->status_register.space_id); result = -EIO; goto err_unreg; } - for (i=0; i<p.state_count; i++) { - if (p.states[i].control != p.states[i].status) { + for (i=0; i<p->state_count; i++) { + if (p->states[i].control != p->states[i].status) { dprintk("Different control (%llu) and status values (%llu)\n", - p.states[i].control, p.states[i].status); + p->states[i].control, p->states[i].status); result = -EINVAL; goto err_unreg; } - if (!p.states[i].core_frequency) { + if (!p->states[i].core_frequency) { dprintk("Zero core frequency for state %u\n", i); result = -EINVAL; goto err_unreg; } - if (p.states[i].core_frequency > p.states[0].core_frequency) { + if (p->states[i].core_frequency > p->states[0].core_frequency) { dprintk("P%u has larger frequency (%llu) than P0 (%llu), skipping\n", i, - p.states[i].core_frequency, p.states[0].core_frequency); - p.states[i].core_frequency = 0; + p->states[i].core_frequency, p->states[0].core_frequency); + p->states[i].core_frequency = 0; continue; } } @@ -416,26 +451,26 @@ static int centrino_cpu_init_acpi(struct cpufreq_policy *policy) } centrino_model[cpu]->model_name=NULL; - centrino_model[cpu]->max_freq = p.states[0].core_frequency * 1000; + centrino_model[cpu]->max_freq = p->states[0].core_frequency * 1000; centrino_model[cpu]->op_points = kmalloc(sizeof(struct cpufreq_frequency_table) * - (p.state_count + 1), GFP_KERNEL); + (p->state_count + 1), GFP_KERNEL); if (!centrino_model[cpu]->op_points) { result = -ENOMEM; goto err_kfree; } - for (i=0; i<p.state_count; i++) { - centrino_model[cpu]->op_points[i].index = p.states[i].control; - centrino_model[cpu]->op_points[i].frequency = p.states[i].core_frequency * 1000; + for (i=0; i<p->state_count; i++) { + centrino_model[cpu]->op_points[i].index = p->states[i].control; + centrino_model[cpu]->op_points[i].frequency = p->states[i].core_frequency * 1000; dprintk("adding state %i with frequency %u and control value %04x\n", i, centrino_model[cpu]->op_points[i].frequency, centrino_model[cpu]->op_points[i].index); } - centrino_model[cpu]->op_points[p.state_count].frequency = CPUFREQ_TABLE_END; + centrino_model[cpu]->op_points[p->state_count].frequency = CPUFREQ_TABLE_END; cur_freq = get_cur_freq(cpu); - for (i=0; i<p.state_count; i++) { - if (!p.states[i].core_frequency) { + for (i=0; i<p->state_count; i++) { + if (!p->states[i].core_frequency) { dprintk("skipping state %u\n", i); centrino_model[cpu]->op_points[i].frequency = CPUFREQ_ENTRY_INVALID; continue; @@ -451,7 +486,7 @@ static int centrino_cpu_init_acpi(struct cpufreq_policy *policy) } if (cur_freq == centrino_model[cpu]->op_points[i].frequency) - p.state = i; + p->state = i; } /* notify BIOS that we exist */ @@ -464,12 +499,13 @@ static int centrino_cpu_init_acpi(struct cpufreq_policy *policy) err_kfree: kfree(centrino_model[cpu]); err_unreg: - acpi_processor_unregister_performance(&p, cpu); - dprintk(KERN_INFO PFX "invalid ACPI data\n"); + acpi_processor_unregister_performance(p, cpu); + dprintk(PFX "invalid ACPI data\n"); return (result); } #else static inline int centrino_cpu_init_acpi(struct cpufreq_policy *policy) { return -ENODEV; } +static inline int centrino_cpu_early_init_acpi(void) { return 0; } #endif static int centrino_cpu_init(struct cpufreq_policy *policy) @@ -499,7 +535,7 @@ static int centrino_cpu_init(struct cpufreq_policy *policy) centrino_cpu[policy->cpu] = &cpu_ids[i]; if (!centrino_cpu[policy->cpu]) { - dprintk(KERN_INFO PFX "found unsupported CPU with " + dprintk("found unsupported CPU with " "Enhanced SpeedStep: send /proc/cpuinfo to " MAINTAINER "\n"); return -ENODEV; @@ -555,10 +591,15 @@ static int centrino_cpu_exit(struct cpufreq_policy *policy) #ifdef CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI if (!centrino_model[cpu]->model_name) { - dprintk("unregistering and freeing ACPI data\n"); - acpi_processor_unregister_performance(&p, cpu); - kfree(centrino_model[cpu]->op_points); - kfree(centrino_model[cpu]); + static struct acpi_processor_performance *p; + + if (acpi_perf_data[cpu]) { + p = acpi_perf_data[cpu]; + dprintk("unregistering and freeing ACPI data\n"); + acpi_processor_unregister_performance(p, cpu); + kfree(centrino_model[cpu]->op_points); + kfree(centrino_model[cpu]); + } } #endif @@ -592,63 +633,128 @@ static int centrino_target (struct cpufreq_policy *policy, unsigned int relation) { unsigned int newstate = 0; - unsigned int msr, oldmsr, h, cpu = policy->cpu; + unsigned int msr, oldmsr = 0, h = 0, cpu = policy->cpu; struct cpufreq_freqs freqs; + cpumask_t online_policy_cpus; cpumask_t saved_mask; - int retval; + cpumask_t set_mask; + cpumask_t covered_cpus; + int retval = 0; + unsigned int j, k, first_cpu, tmp; - if (centrino_model[cpu] == NULL) + if (unlikely(centrino_model[cpu] == NULL)) return -ENODEV; - /* - * Support for SMP systems. - * Make sure we are running on the CPU that wants to change frequency - */ - saved_mask = current->cpus_allowed; - set_cpus_allowed(current, policy->cpus); - if (!cpu_isset(smp_processor_id(), policy->cpus)) { - dprintk("couldn't limit to CPUs in this domain\n"); - return(-EAGAIN); + if (unlikely(cpufreq_frequency_table_target(policy, + centrino_model[cpu]->op_points, + target_freq, + relation, + &newstate))) { + return -EINVAL; } - if (cpufreq_frequency_table_target(policy, centrino_model[cpu]->op_points, target_freq, - relation, &newstate)) { - retval = -EINVAL; - goto migrate_end; - } +#ifdef CONFIG_HOTPLUG_CPU + /* cpufreq holds the hotplug lock, so we are safe from here on */ + cpus_and(online_policy_cpus, cpu_online_map, policy->cpus); +#else + online_policy_cpus = policy->cpus; +#endif - msr = centrino_model[cpu]->op_points[newstate].index; - rdmsr(MSR_IA32_PERF_CTL, oldmsr, h); + saved_mask = current->cpus_allowed; + first_cpu = 1; + cpus_clear(covered_cpus); + for_each_cpu_mask(j, online_policy_cpus) { + /* + * Support for SMP systems. + * Make sure we are running on CPU that wants to change freq + */ + cpus_clear(set_mask); + if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) + cpus_or(set_mask, set_mask, online_policy_cpus); + else + cpu_set(j, set_mask); + + set_cpus_allowed(current, set_mask); + if (unlikely(!cpu_isset(smp_processor_id(), set_mask))) { + dprintk("couldn't limit to CPUs in this domain\n"); + retval = -EAGAIN; + if (first_cpu) { + /* We haven't started the transition yet. */ + goto migrate_end; + } + break; + } - if (msr == (oldmsr & 0xffff)) { - retval = 0; - dprintk("no change needed - msr was and needs to be %x\n", oldmsr); - goto migrate_end; - } + msr = centrino_model[cpu]->op_points[newstate].index; + + if (first_cpu) { + rdmsr(MSR_IA32_PERF_CTL, oldmsr, h); + if (msr == (oldmsr & 0xffff)) { + dprintk("no change needed - msr was and needs " + "to be %x\n", oldmsr); + retval = 0; + goto migrate_end; + } + + freqs.old = extract_clock(oldmsr, cpu, 0); + freqs.new = extract_clock(msr, cpu, 0); + + dprintk("target=%dkHz old=%d new=%d msr=%04x\n", + target_freq, freqs.old, freqs.new, msr); + + for_each_cpu_mask(k, online_policy_cpus) { + freqs.cpu = k; + cpufreq_notify_transition(&freqs, + CPUFREQ_PRECHANGE); + } + + first_cpu = 0; + /* all but 16 LSB are reserved, treat them with care */ + oldmsr &= ~0xffff; + msr &= 0xffff; + oldmsr |= msr; + } - freqs.cpu = cpu; - freqs.old = extract_clock(oldmsr, cpu, 0); - freqs.new = extract_clock(msr, cpu, 0); + wrmsr(MSR_IA32_PERF_CTL, oldmsr, h); + if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) + break; - dprintk("target=%dkHz old=%d new=%d msr=%04x\n", - target_freq, freqs.old, freqs.new, msr); + cpu_set(j, covered_cpus); + } - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + for_each_cpu_mask(k, online_policy_cpus) { + freqs.cpu = k; + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + } - /* all but 16 LSB are "reserved", so treat them with - care */ - oldmsr &= ~0xffff; - msr &= 0xffff; - oldmsr |= msr; + if (unlikely(retval)) { + /* + * We have failed halfway through the frequency change. + * We have sent callbacks to policy->cpus and + * MSRs have already been written on coverd_cpus. + * Best effort undo.. + */ - wrmsr(MSR_IA32_PERF_CTL, oldmsr, h); + if (!cpus_empty(covered_cpus)) { + for_each_cpu_mask(j, covered_cpus) { + set_cpus_allowed(current, cpumask_of_cpu(j)); + wrmsr(MSR_IA32_PERF_CTL, oldmsr, h); + } + } - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + tmp = freqs.new; + freqs.new = freqs.old; + freqs.old = tmp; + for_each_cpu_mask(j, online_policy_cpus) { + freqs.cpu = j; + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + } + } - retval = 0; migrate_end: set_cpus_allowed(current, saved_mask); - return (retval); + return 0; } static struct freq_attr* centrino_attr[] = { @@ -690,12 +796,25 @@ static int __init centrino_init(void) if (!cpu_has(cpu, X86_FEATURE_EST)) return -ENODEV; + centrino_cpu_early_init_acpi(); + return cpufreq_register_driver(¢rino_driver); } static void __exit centrino_exit(void) { +#ifdef CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI + unsigned int j; +#endif + cpufreq_unregister_driver(¢rino_driver); + +#ifdef CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI + for_each_possible_cpu(j) { + kfree(acpi_perf_data[j]); + acpi_perf_data[j] = NULL; + } +#endif } MODULE_AUTHOR ("Jeremy Fitzhardinge <jeremy@goop.org>"); diff --git a/arch/i386/kernel/cpu/cyrix.c b/arch/i386/kernel/cpu/cyrix.c index 00f2e058797c..f03b7f94c304 100644 --- a/arch/i386/kernel/cpu/cyrix.c +++ b/arch/i386/kernel/cpu/cyrix.c @@ -184,7 +184,7 @@ static void __init geode_configure(void) #ifdef CONFIG_PCI -static struct pci_device_id cyrix_55x0[] = { +static struct pci_device_id __initdata cyrix_55x0[] = { { PCI_DEVICE(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5510) }, { PCI_DEVICE(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5520) }, { }, @@ -272,14 +272,15 @@ static void __init init_cyrix(struct cpuinfo_x86 *c) printk(KERN_INFO "Working around Cyrix MediaGX virtual DMA bugs.\n"); isa_dma_bridge_buggy = 2; -#endif - c->x86_cache_size=16; /* Yep 16K integrated cache thats it */ - + + /* * The 5510/5520 companion chips have a funky PIT. */ if (pci_dev_present(cyrix_55x0)) pit_latch_buggy = 1; +#endif + c->x86_cache_size=16; /* Yep 16K integrated cache thats it */ /* GXm supports extended cpuid levels 'ala' AMD */ if (c->cpuid_level == 2) { @@ -353,7 +354,7 @@ static void __init init_nsc(struct cpuinfo_x86 *c) * This function only handles the GX processor, and kicks every * thing else to the Cyrix init function above - that should * cover any processors that might have been branded differently - * after NSC aquired Cyrix. + * after NSC acquired Cyrix. * * If this breaks your GX1 horribly, please e-mail * info-linux@ldcmail.amd.com to tell us. diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c index 5386b29bb5a5..5a2e270924b1 100644 --- a/arch/i386/kernel/cpu/intel.c +++ b/arch/i386/kernel/cpu/intel.c @@ -1,4 +1,3 @@ -#include <linux/config.h> #include <linux/init.h> #include <linux/kernel.h> @@ -122,6 +121,12 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) select_idle_routine(c); l2 = init_intel_cacheinfo(c); + if (c->cpuid_level > 9 ) { + unsigned eax = cpuid_eax(10); + /* Check for version and the number of counters */ + if ((eax & 0xff) && (((eax>>8) & 0xff) > 1)) + set_bit(X86_FEATURE_ARCH_PERFMON, c->x86_capability); + } /* SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until model 3 mask 3 */ if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633) diff --git a/arch/i386/kernel/cpu/intel_cacheinfo.c b/arch/i386/kernel/cpu/intel_cacheinfo.c index c8547a6fa7e6..e9f0b928b0a9 100644 --- a/arch/i386/kernel/cpu/intel_cacheinfo.c +++ b/arch/i386/kernel/cpu/intel_cacheinfo.c @@ -4,6 +4,7 @@ * Changes: * Venkatesh Pallipadi : Adding cache identification through cpuid(4) * Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure. + * Andi Kleen : CPUID4 emulation on AMD. */ #include <linux/init.h> @@ -130,25 +131,111 @@ struct _cpuid4_info { cpumask_t shared_cpu_map; }; -static unsigned short num_cache_leaves; +unsigned short num_cache_leaves; + +/* AMD doesn't have CPUID4. Emulate it here to report the same + information to the user. This makes some assumptions about the machine: + No L3, L2 not shared, no SMT etc. that is currently true on AMD CPUs. + + In theory the TLBs could be reported as fake type (they are in "dummy"). + Maybe later */ +union l1_cache { + struct { + unsigned line_size : 8; + unsigned lines_per_tag : 8; + unsigned assoc : 8; + unsigned size_in_kb : 8; + }; + unsigned val; +}; + +union l2_cache { + struct { + unsigned line_size : 8; + unsigned lines_per_tag : 4; + unsigned assoc : 4; + unsigned size_in_kb : 16; + }; + unsigned val; +}; + +static const unsigned short assocs[] = { + [1] = 1, [2] = 2, [4] = 4, [6] = 8, + [8] = 16, + [0xf] = 0xffff // ?? + }; +static const unsigned char levels[] = { 1, 1, 2 }; +static const unsigned char types[] = { 1, 2, 3 }; + +static void __cpuinit amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax, + union _cpuid4_leaf_ebx *ebx, + union _cpuid4_leaf_ecx *ecx) +{ + unsigned dummy; + unsigned line_size, lines_per_tag, assoc, size_in_kb; + union l1_cache l1i, l1d; + union l2_cache l2; + + eax->full = 0; + ebx->full = 0; + ecx->full = 0; + + cpuid(0x80000005, &dummy, &dummy, &l1d.val, &l1i.val); + cpuid(0x80000006, &dummy, &dummy, &l2.val, &dummy); + + if (leaf > 2 || !l1d.val || !l1i.val || !l2.val) + return; + + eax->split.is_self_initializing = 1; + eax->split.type = types[leaf]; + eax->split.level = levels[leaf]; + eax->split.num_threads_sharing = 0; + eax->split.num_cores_on_die = current_cpu_data.x86_max_cores - 1; + + if (leaf <= 1) { + union l1_cache *l1 = leaf == 0 ? &l1d : &l1i; + assoc = l1->assoc; + line_size = l1->line_size; + lines_per_tag = l1->lines_per_tag; + size_in_kb = l1->size_in_kb; + } else { + assoc = l2.assoc; + line_size = l2.line_size; + lines_per_tag = l2.lines_per_tag; + /* cpu_data has errata corrections for K7 applied */ + size_in_kb = current_cpu_data.x86_cache_size; + } + + if (assoc == 0xf) + eax->split.is_fully_associative = 1; + ebx->split.coherency_line_size = line_size - 1; + ebx->split.ways_of_associativity = assocs[assoc] - 1; + ebx->split.physical_line_partition = lines_per_tag - 1; + ecx->split.number_of_sets = (size_in_kb * 1024) / line_size / + (ebx->split.ways_of_associativity + 1) - 1; +} static int __cpuinit cpuid4_cache_lookup(int index, struct _cpuid4_info *this_leaf) { - unsigned int eax, ebx, ecx, edx; - union _cpuid4_leaf_eax cache_eax; + union _cpuid4_leaf_eax eax; + union _cpuid4_leaf_ebx ebx; + union _cpuid4_leaf_ecx ecx; + unsigned edx; - cpuid_count(4, index, &eax, &ebx, &ecx, &edx); - cache_eax.full = eax; - if (cache_eax.split.type == CACHE_TYPE_NULL) + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) + amd_cpuid4(index, &eax, &ebx, &ecx); + else + cpuid_count(4, index, &eax.full, &ebx.full, &ecx.full, &edx); + if (eax.split.type == CACHE_TYPE_NULL) return -EIO; /* better error ? */ - this_leaf->eax.full = eax; - this_leaf->ebx.full = ebx; - this_leaf->ecx.full = ecx; - this_leaf->size = (this_leaf->ecx.split.number_of_sets + 1) * - (this_leaf->ebx.split.coherency_line_size + 1) * - (this_leaf->ebx.split.physical_line_partition + 1) * - (this_leaf->ebx.split.ways_of_associativity + 1); + this_leaf->eax = eax; + this_leaf->ebx = ebx; + this_leaf->ecx = ecx; + this_leaf->size = (ecx.split.number_of_sets + 1) * + (ebx.split.coherency_line_size + 1) * + (ebx.split.physical_line_partition + 1) * + (ebx.split.ways_of_associativity + 1); return 0; } @@ -174,7 +261,7 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */ unsigned int new_l2 = 0, new_l3 = 0, i; /* Cache sizes from cpuid(4) */ unsigned int l2_id = 0, l3_id = 0, num_threads_sharing, index_msb; -#ifdef CONFIG_SMP +#ifdef CONFIG_X86_HT unsigned int cpu = (c == &boot_cpu_data) ? 0 : (c - cpu_data); #endif @@ -296,14 +383,14 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) if (new_l2) { l2 = new_l2; -#ifdef CONFIG_SMP +#ifdef CONFIG_X86_HT cpu_llc_id[cpu] = l2_id; #endif } if (new_l3) { l3 = new_l3; -#ifdef CONFIG_SMP +#ifdef CONFIG_X86_HT cpu_llc_id[cpu] = l3_id; #endif } @@ -642,7 +729,7 @@ static void __cpuexit cache_remove_dev(struct sys_device * sys_dev) return; } -static int cacheinfo_cpu_callback(struct notifier_block *nfb, +static int __cpuinit cacheinfo_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -660,7 +747,7 @@ static int cacheinfo_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -static struct notifier_block cacheinfo_cpu_notifier = +static struct notifier_block __cpuinitdata cacheinfo_cpu_notifier = { .notifier_call = cacheinfo_cpu_callback, }; diff --git a/arch/i386/kernel/cpu/mcheck/k7.c b/arch/i386/kernel/cpu/mcheck/k7.c index fc5d5215e23d..b0862af595aa 100644 --- a/arch/i386/kernel/cpu/mcheck/k7.c +++ b/arch/i386/kernel/cpu/mcheck/k7.c @@ -6,7 +6,6 @@ #include <linux/init.h> #include <linux/types.h> #include <linux/kernel.h> -#include <linux/config.h> #include <linux/interrupt.h> #include <linux/smp.h> diff --git a/arch/i386/kernel/cpu/mcheck/mce.c b/arch/i386/kernel/cpu/mcheck/mce.c index afa0888f9a1e..d555bec0db99 100644 --- a/arch/i386/kernel/cpu/mcheck/mce.c +++ b/arch/i386/kernel/cpu/mcheck/mce.c @@ -6,7 +6,6 @@ #include <linux/init.h> #include <linux/types.h> #include <linux/kernel.h> -#include <linux/config.h> #include <linux/module.h> #include <linux/smp.h> #include <linux/thread_info.h> diff --git a/arch/i386/kernel/cpu/mcheck/non-fatal.c b/arch/i386/kernel/cpu/mcheck/non-fatal.c index 82dffe0d4954..1f9153ae5b03 100644 --- a/arch/i386/kernel/cpu/mcheck/non-fatal.c +++ b/arch/i386/kernel/cpu/mcheck/non-fatal.c @@ -11,7 +11,6 @@ #include <linux/types.h> #include <linux/kernel.h> #include <linux/jiffies.h> -#include <linux/config.h> #include <linux/workqueue.h> #include <linux/interrupt.h> #include <linux/smp.h> diff --git a/arch/i386/kernel/cpu/mcheck/p4.c b/arch/i386/kernel/cpu/mcheck/p4.c index fd2c459a31ef..b95f1b3d53aa 100644 --- a/arch/i386/kernel/cpu/mcheck/p4.c +++ b/arch/i386/kernel/cpu/mcheck/p4.c @@ -5,7 +5,6 @@ #include <linux/init.h> #include <linux/types.h> #include <linux/kernel.h> -#include <linux/config.h> #include <linux/interrupt.h> #include <linux/smp.h> diff --git a/arch/i386/kernel/cpu/proc.c b/arch/i386/kernel/cpu/proc.c index f94cdb7aca50..f54a15268ed7 100644 --- a/arch/i386/kernel/cpu/proc.c +++ b/arch/i386/kernel/cpu/proc.c @@ -18,7 +18,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) * applications want to get the raw CPUID data, they should access * /dev/cpu/<cpu_nr>/cpuid instead. */ - static char *x86_cap_flags[] = { + static const char * const x86_cap_flags[] = { /* Intel-defined */ "fpu", "vme", "de", "pse", "tsc", "msr", "pae", "mce", "cx8", "apic", NULL, "sep", "mtrr", "pge", "mca", "cmov", @@ -52,7 +52,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) /* VIA/Cyrix/Centaur-defined */ NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en", - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + "ace2", "ace2_en", "phe", "phe_en", "pmm", "pmm_en", NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, @@ -62,7 +62,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, }; - static char *x86_power_flags[] = { + static const char * const x86_power_flags[] = { "ts", /* temperature sensor */ "fid", /* frequency id control */ "vid", /* voltage id control */ @@ -109,9 +109,9 @@ static int show_cpuinfo(struct seq_file *m, void *v) seq_printf(m, "cache size\t: %d KB\n", c->x86_cache_size); #ifdef CONFIG_X86_HT if (c->x86_max_cores * smp_num_siblings > 1) { - seq_printf(m, "physical id\t: %d\n", phys_proc_id[n]); + seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); seq_printf(m, "siblings\t: %d\n", cpus_weight(cpu_core_map[n])); - seq_printf(m, "core id\t\t: %d\n", cpu_core_id[n]); + seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id); seq_printf(m, "cpu cores\t: %d\n", c->booted_cores); } #endif diff --git a/arch/i386/kernel/cpuid.c b/arch/i386/kernel/cpuid.c index 1d9a4abcdfc7..fde8bea85cee 100644 --- a/arch/i386/kernel/cpuid.c +++ b/arch/i386/kernel/cpuid.c @@ -24,7 +24,6 @@ */ #include <linux/module.h> -#include <linux/config.h> #include <linux/types.h> #include <linux/errno.h> @@ -168,6 +167,7 @@ static int cpuid_class_device_create(int i) return err; } +#ifdef CONFIG_HOTPLUG_CPU static int cpuid_class_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -183,10 +183,11 @@ static int cpuid_class_cpu_callback(struct notifier_block *nfb, unsigned long ac return NOTIFY_OK; } -static struct notifier_block cpuid_class_cpu_notifier = +static struct notifier_block __cpuinitdata cpuid_class_cpu_notifier = { .notifier_call = cpuid_class_cpu_callback, }; +#endif /* !CONFIG_HOTPLUG_CPU */ static int __init cpuid_init(void) { @@ -209,7 +210,7 @@ static int __init cpuid_init(void) if (err != 0) goto out_class; } - register_cpu_notifier(&cpuid_class_cpu_notifier); + register_hotcpu_notifier(&cpuid_class_cpu_notifier); err = 0; goto out; @@ -234,7 +235,7 @@ static void __exit cpuid_exit(void) class_device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, cpu)); class_destroy(cpuid_class); unregister_chrdev(CPUID_MAJOR, "cpu/cpuid"); - unregister_cpu_notifier(&cpuid_class_cpu_notifier); + unregister_hotcpu_notifier(&cpuid_class_cpu_notifier); } module_init(cpuid_init); diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c index 2b0cfce24a61..48f0f62f781c 100644 --- a/arch/i386/kernel/crash.c +++ b/arch/i386/kernel/crash.c @@ -114,19 +114,15 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu) atomic_dec(&waiting_for_crash_ipi); /* Assume hlt works */ halt(); - for(;;); + for (;;) + cpu_relax(); return 1; } -/* - * By using the NMI code instead of a vector we just sneak thru the - * word generator coming out with just what we want. AND it does - * not matter if clustered_apic_mode is set or not. - */ static void smp_send_nmi_allbutself(void) { - send_IPI_allbutself(APIC_DM_NMI); + send_IPI_allbutself(NMI_VECTOR); } static void nmi_shootdown_cpus(void) @@ -162,7 +158,7 @@ static void nmi_shootdown_cpus(void) void machine_crash_shutdown(struct pt_regs *regs) { /* This function is only called after the system - * has paniced or is otherwise in a critical state. + * has panicked or is otherwise in a critical state. * The minimum amount of code to allow a kexec'd kernel * to run successfully needs to happen here. * diff --git a/arch/i386/kernel/doublefault.c b/arch/i386/kernel/doublefault.c index 5edb1d379add..b4d14c2eb345 100644 --- a/arch/i386/kernel/doublefault.c +++ b/arch/i386/kernel/doublefault.c @@ -44,7 +44,8 @@ static void doublefault_fn(void) } } - for (;;) /* nothing */; + for (;;) + cpu_relax(); } struct tss_struct doublefault_tss __cacheline_aligned = { diff --git a/arch/i386/kernel/efi.c b/arch/i386/kernel/efi.c index 9202b67c4b2e..fe158042110b 100644 --- a/arch/i386/kernel/efi.c +++ b/arch/i386/kernel/efi.c @@ -19,7 +19,6 @@ * Skip non-WB memory and ignore empty memory ranges. */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/init.h> #include <linux/mm.h> @@ -601,8 +600,10 @@ efi_initialize_iomem_resources(struct resource *code_resource, res->end = res->start + ((md->num_pages << EFI_PAGE_SHIFT) - 1); res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; if (request_resource(&iomem_resource, res) < 0) - printk(KERN_ERR PFX "Failed to allocate res %s : 0x%lx-0x%lx\n", - res->name, res->start, res->end); + printk(KERN_ERR PFX "Failed to allocate res %s : " + "0x%llx-0x%llx\n", res->name, + (unsigned long long)res->start, + (unsigned long long)res->end); /* * We don't know which region contains kernel data so we try * it repeatedly and let the resource manager test it. diff --git a/arch/i386/kernel/efi_stub.S b/arch/i386/kernel/efi_stub.S index 08c0312d9b6c..d3ee73a3eee3 100644 --- a/arch/i386/kernel/efi_stub.S +++ b/arch/i386/kernel/efi_stub.S @@ -5,7 +5,6 @@ * turned off. */ -#include <linux/config.h> #include <linux/linkage.h> #include <asm/page.h> #include <asm/pgtable.h> diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index cfc683f153b9..d9a260f2efb4 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -40,14 +40,15 @@ * "current" is in register %ebx during any slow entries. */ -#include <linux/config.h> #include <linux/linkage.h> #include <asm/thread_info.h> +#include <asm/irqflags.h> #include <asm/errno.h> #include <asm/segment.h> #include <asm/smp.h> #include <asm/page.h> #include <asm/desc.h> +#include <asm/dwarf2.h> #include "irq_vectors.h" #define nr_syscalls ((syscall_table_size)/4) @@ -76,40 +77,91 @@ NT_MASK = 0x00004000 VM_MASK = 0x00020000 #ifdef CONFIG_PREEMPT -#define preempt_stop cli +#define preempt_stop cli; TRACE_IRQS_OFF #else #define preempt_stop #define resume_kernel restore_nocheck #endif +.macro TRACE_IRQS_IRET +#ifdef CONFIG_TRACE_IRQFLAGS + testl $IF_MASK,EFLAGS(%esp) # interrupts off? + jz 1f + TRACE_IRQS_ON +1: +#endif +.endm + +#ifdef CONFIG_VM86 +#define resume_userspace_sig check_userspace +#else +#define resume_userspace_sig resume_userspace +#endif + #define SAVE_ALL \ cld; \ pushl %es; \ + CFI_ADJUST_CFA_OFFSET 4;\ + /*CFI_REL_OFFSET es, 0;*/\ pushl %ds; \ + CFI_ADJUST_CFA_OFFSET 4;\ + /*CFI_REL_OFFSET ds, 0;*/\ pushl %eax; \ + CFI_ADJUST_CFA_OFFSET 4;\ + CFI_REL_OFFSET eax, 0;\ pushl %ebp; \ + CFI_ADJUST_CFA_OFFSET 4;\ + CFI_REL_OFFSET ebp, 0;\ pushl %edi; \ + CFI_ADJUST_CFA_OFFSET 4;\ + CFI_REL_OFFSET edi, 0;\ pushl %esi; \ + CFI_ADJUST_CFA_OFFSET 4;\ + CFI_REL_OFFSET esi, 0;\ pushl %edx; \ + CFI_ADJUST_CFA_OFFSET 4;\ + CFI_REL_OFFSET edx, 0;\ pushl %ecx; \ + CFI_ADJUST_CFA_OFFSET 4;\ + CFI_REL_OFFSET ecx, 0;\ pushl %ebx; \ + CFI_ADJUST_CFA_OFFSET 4;\ + CFI_REL_OFFSET ebx, 0;\ movl $(__USER_DS), %edx; \ movl %edx, %ds; \ movl %edx, %es; #define RESTORE_INT_REGS \ popl %ebx; \ + CFI_ADJUST_CFA_OFFSET -4;\ + CFI_RESTORE ebx;\ popl %ecx; \ + CFI_ADJUST_CFA_OFFSET -4;\ + CFI_RESTORE ecx;\ popl %edx; \ + CFI_ADJUST_CFA_OFFSET -4;\ + CFI_RESTORE edx;\ popl %esi; \ + CFI_ADJUST_CFA_OFFSET -4;\ + CFI_RESTORE esi;\ popl %edi; \ + CFI_ADJUST_CFA_OFFSET -4;\ + CFI_RESTORE edi;\ popl %ebp; \ - popl %eax + CFI_ADJUST_CFA_OFFSET -4;\ + CFI_RESTORE ebp;\ + popl %eax; \ + CFI_ADJUST_CFA_OFFSET -4;\ + CFI_RESTORE eax #define RESTORE_REGS \ RESTORE_INT_REGS; \ 1: popl %ds; \ + CFI_ADJUST_CFA_OFFSET -4;\ + /*CFI_RESTORE ds;*/\ 2: popl %es; \ + CFI_ADJUST_CFA_OFFSET -4;\ + /*CFI_RESTORE es;*/\ .section .fixup,"ax"; \ 3: movl $0,(%esp); \ jmp 1b; \ @@ -122,13 +174,43 @@ VM_MASK = 0x00020000 .long 2b,4b; \ .previous +#define RING0_INT_FRAME \ + CFI_STARTPROC simple;\ + CFI_DEF_CFA esp, 3*4;\ + /*CFI_OFFSET cs, -2*4;*/\ + CFI_OFFSET eip, -3*4 + +#define RING0_EC_FRAME \ + CFI_STARTPROC simple;\ + CFI_DEF_CFA esp, 4*4;\ + /*CFI_OFFSET cs, -2*4;*/\ + CFI_OFFSET eip, -3*4 + +#define RING0_PTREGS_FRAME \ + CFI_STARTPROC simple;\ + CFI_DEF_CFA esp, OLDESP-EBX;\ + /*CFI_OFFSET cs, CS-OLDESP;*/\ + CFI_OFFSET eip, EIP-OLDESP;\ + /*CFI_OFFSET es, ES-OLDESP;*/\ + /*CFI_OFFSET ds, DS-OLDESP;*/\ + CFI_OFFSET eax, EAX-OLDESP;\ + CFI_OFFSET ebp, EBP-OLDESP;\ + CFI_OFFSET edi, EDI-OLDESP;\ + CFI_OFFSET esi, ESI-OLDESP;\ + CFI_OFFSET edx, EDX-OLDESP;\ + CFI_OFFSET ecx, ECX-OLDESP;\ + CFI_OFFSET ebx, EBX-OLDESP ENTRY(ret_from_fork) + CFI_STARTPROC pushl %eax + CFI_ADJUST_CFA_OFFSET -4 call schedule_tail GET_THREAD_INFO(%ebp) popl %eax + CFI_ADJUST_CFA_OFFSET -4 jmp syscall_exit + CFI_ENDPROC /* * Return to user mode is not as complex as all this looks, @@ -139,10 +221,12 @@ ENTRY(ret_from_fork) # userspace resumption stub bypassing syscall exit tracing ALIGN + RING0_PTREGS_FRAME ret_from_exception: preempt_stop ret_from_intr: GET_THREAD_INFO(%ebp) +check_userspace: movl EFLAGS(%esp), %eax # mix EFLAGS and CS movb CS(%esp), %al testl $(VM_MASK | 3), %eax @@ -171,20 +255,42 @@ need_resched: call preempt_schedule_irq jmp need_resched #endif + CFI_ENDPROC /* SYSENTER_RETURN points to after the "sysenter" instruction in the vsyscall page. See vsyscall-sysentry.S, which defines the symbol. */ # sysenter call handler stub ENTRY(sysenter_entry) + CFI_STARTPROC simple + CFI_DEF_CFA esp, 0 + CFI_REGISTER esp, ebp movl TSS_sysenter_esp0(%esp),%esp sysenter_past_esp: + /* + * No need to follow this irqs on/off section: the syscall + * disabled irqs and here we enable it straight after entry: + */ sti pushl $(__USER_DS) + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET ss, 0*/ pushl %ebp + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET esp, 0 pushfl + CFI_ADJUST_CFA_OFFSET 4 pushl $(__USER_CS) - pushl $SYSENTER_RETURN + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET cs, 0*/ + /* + * Push current_thread_info()->sysenter_return to the stack. + * A tiny bit of offset fixup is necessary - 4*4 means the 4 words + * pushed above; +8 corresponds to copy_thread's esp0 setting. + */ + pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp) + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET eip, 0 /* * Load the potential sixth argument from user stack. @@ -199,6 +305,7 @@ sysenter_past_esp: .previous pushl %eax + CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL GET_THREAD_INFO(%ebp) @@ -210,6 +317,7 @@ sysenter_past_esp: call *sys_call_table(,%eax,4) movl %eax,EAX(%esp) cli + TRACE_IRQS_OFF movl TI_flags(%ebp), %ecx testw $_TIF_ALLWORK_MASK, %cx jne syscall_exit_work @@ -217,13 +325,17 @@ sysenter_past_esp: movl EIP(%esp), %edx movl OLDESP(%esp), %ecx xorl %ebp,%ebp + TRACE_IRQS_ON sti sysexit + CFI_ENDPROC # system call handler stub ENTRY(system_call) + RING0_INT_FRAME # can't unwind into user space anyway pushl %eax # save orig_eax + CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL GET_THREAD_INFO(%ebp) testl $TF_MASK,EFLAGS(%esp) @@ -243,6 +355,7 @@ syscall_exit: cli # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret + TRACE_IRQS_OFF movl TI_flags(%ebp), %ecx testw $_TIF_ALLWORK_MASK, %cx # current->work jne syscall_exit_work @@ -256,13 +369,18 @@ restore_all: movb CS(%esp), %al andl $(VM_MASK | (4 << 8) | 3), %eax cmpl $((4 << 8) | 3), %eax + CFI_REMEMBER_STATE je ldt_ss # returning to user-space with LDT SS restore_nocheck: + TRACE_IRQS_IRET +restore_nocheck_notrace: RESTORE_REGS addl $4, %esp + CFI_ADJUST_CFA_OFFSET -4 1: iret .section .fixup,"ax" iret_exc: + TRACE_IRQS_ON sti pushl $0 # no error code pushl $do_iret_error @@ -273,6 +391,7 @@ iret_exc: .long 1b,iret_exc .previous + CFI_RESTORE_STATE ldt_ss: larl OLDSS(%esp), %eax jnz restore_nocheck @@ -285,11 +404,15 @@ ldt_ss: * CPUs, which we can try to work around to make * dosemu and wine happy. */ subl $8, %esp # reserve space for switch16 pointer + CFI_ADJUST_CFA_OFFSET 8 cli + TRACE_IRQS_OFF movl %esp, %eax /* Set up the 16bit stack frame with switch32 pointer on top, * and a switch16 pointer on top of the current frame. */ call setup_x86_bogus_stack + CFI_ADJUST_CFA_OFFSET -8 # frame has moved + TRACE_IRQS_IRET RESTORE_REGS lss 20+4(%esp), %esp # switch to 16bit stack 1: iret @@ -297,9 +420,11 @@ ldt_ss: .align 4 .long 1b,iret_exc .previous + CFI_ENDPROC # perform work that needs to be done immediately before resumption ALIGN + RING0_PTREGS_FRAME # can't unwind into user space anyway work_pending: testb $_TIF_NEED_RESCHED, %cl jz work_notifysig @@ -308,6 +433,7 @@ work_resched: cli # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret + TRACE_IRQS_OFF movl TI_flags(%ebp), %ecx andl $_TIF_WORK_MASK, %ecx # is there any work to be done other # than syscall tracing? @@ -323,18 +449,20 @@ work_notifysig: # deal with pending signals and # vm86-space xorl %edx, %edx call do_notify_resume - jmp resume_userspace + jmp resume_userspace_sig ALIGN work_notifysig_v86: #ifdef CONFIG_VM86 pushl %ecx # save ti_flags for do_notify_resume + CFI_ADJUST_CFA_OFFSET 4 call save_v86_state # %eax contains pt_regs pointer popl %ecx + CFI_ADJUST_CFA_OFFSET -4 movl %eax, %esp xorl %edx, %edx call do_notify_resume - jmp resume_userspace + jmp resume_userspace_sig #endif # perform syscall exit tracing @@ -357,25 +485,28 @@ syscall_trace_entry: syscall_exit_work: testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP), %cl jz work_pending + TRACE_IRQS_ON sti # could let do_syscall_trace() call # schedule() instead movl %esp, %eax movl $1, %edx call do_syscall_trace jmp resume_userspace + CFI_ENDPROC - ALIGN + RING0_INT_FRAME # can't unwind into user space anyway syscall_fault: pushl %eax # save orig_eax + CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL GET_THREAD_INFO(%ebp) movl $-EFAULT,EAX(%esp) jmp resume_userspace - ALIGN syscall_badsys: movl $-ENOSYS,EAX(%esp) jmp resume_userspace + CFI_ENDPROC #define FIXUP_ESPFIX_STACK \ movl %esp, %eax; \ @@ -387,16 +518,21 @@ syscall_badsys: movl %eax, %esp; #define UNWIND_ESPFIX_STACK \ pushl %eax; \ + CFI_ADJUST_CFA_OFFSET 4; \ movl %ss, %eax; \ /* see if on 16bit stack */ \ cmpw $__ESPFIX_SS, %ax; \ - jne 28f; \ - movl $__KERNEL_DS, %edx; \ - movl %edx, %ds; \ - movl %edx, %es; \ + je 28f; \ +27: popl %eax; \ + CFI_ADJUST_CFA_OFFSET -4; \ +.section .fixup,"ax"; \ +28: movl $__KERNEL_DS, %eax; \ + movl %eax, %ds; \ + movl %eax, %es; \ /* switch to 32bit stack */ \ - FIXUP_ESPFIX_STACK \ -28: popl %eax; + FIXUP_ESPFIX_STACK; \ + jmp 27b; \ +.previous /* * Build the entry stubs and pointer table with @@ -408,9 +544,14 @@ ENTRY(interrupt) vector=0 ENTRY(irq_entries_start) + RING0_INT_FRAME .rept NR_IRQS ALIGN -1: pushl $vector-256 + .if vector + CFI_ADJUST_CFA_OFFSET -4 + .endif +1: pushl $~(vector) + CFI_ADJUST_CFA_OFFSET 4 jmp common_interrupt .data .long 1b @@ -418,66 +559,111 @@ ENTRY(irq_entries_start) vector=vector+1 .endr +/* + * the CPU automatically disables interrupts when executing an IRQ vector, + * so IRQ-flags tracing has to follow that: + */ ALIGN common_interrupt: SAVE_ALL + TRACE_IRQS_OFF movl %esp,%eax call do_IRQ jmp ret_from_intr + CFI_ENDPROC #define BUILD_INTERRUPT(name, nr) \ ENTRY(name) \ - pushl $nr-256; \ - SAVE_ALL \ + RING0_INT_FRAME; \ + pushl $~(nr); \ + CFI_ADJUST_CFA_OFFSET 4; \ + SAVE_ALL; \ + TRACE_IRQS_OFF \ movl %esp,%eax; \ call smp_/**/name; \ - jmp ret_from_intr; + jmp ret_from_intr; \ + CFI_ENDPROC /* The include is where all of the SMP etc. interrupts come from */ #include "entry_arch.h" ENTRY(divide_error) + RING0_INT_FRAME pushl $0 # no error code + CFI_ADJUST_CFA_OFFSET 4 pushl $do_divide_error + CFI_ADJUST_CFA_OFFSET 4 ALIGN error_code: pushl %ds + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET ds, 0*/ pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET eax, 0 xorl %eax, %eax pushl %ebp + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ebp, 0 pushl %edi + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET edi, 0 pushl %esi + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET esi, 0 pushl %edx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET edx, 0 decl %eax # eax = -1 pushl %ecx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ecx, 0 pushl %ebx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ebx, 0 cld pushl %es + CFI_ADJUST_CFA_OFFSET 4 + /*CFI_REL_OFFSET es, 0*/ UNWIND_ESPFIX_STACK popl %ecx + CFI_ADJUST_CFA_OFFSET -4 + /*CFI_REGISTER es, ecx*/ movl ES(%esp), %edi # get the function address movl ORIG_EAX(%esp), %edx # get the error code movl %eax, ORIG_EAX(%esp) movl %ecx, ES(%esp) + /*CFI_REL_OFFSET es, ES*/ movl $(__USER_DS), %ecx movl %ecx, %ds movl %ecx, %es movl %esp,%eax # pt_regs pointer call *%edi jmp ret_from_exception + CFI_ENDPROC ENTRY(coprocessor_error) + RING0_INT_FRAME pushl $0 + CFI_ADJUST_CFA_OFFSET 4 pushl $do_coprocessor_error + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC ENTRY(simd_coprocessor_error) + RING0_INT_FRAME pushl $0 + CFI_ADJUST_CFA_OFFSET 4 pushl $do_simd_coprocessor_error + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC ENTRY(device_not_available) + RING0_INT_FRAME pushl $-1 # mark this as an int + CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL movl %cr0, %eax testl $0x4, %eax # EM (math emulation bit) @@ -487,9 +673,12 @@ ENTRY(device_not_available) jmp ret_from_exception device_not_available_emulate: pushl $0 # temporary storage for ORIG_EIP + CFI_ADJUST_CFA_OFFSET 4 call math_emulate addl $4, %esp + CFI_ADJUST_CFA_OFFSET -4 jmp ret_from_exception + CFI_ENDPROC /* * Debug traps and NMI can happen at the one SYSENTER instruction @@ -514,16 +703,19 @@ label: \ pushl $sysenter_past_esp KPROBE_ENTRY(debug) + RING0_INT_FRAME cmpl $sysenter_entry,(%esp) jne debug_stack_correct FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn) debug_stack_correct: pushl $-1 # mark this as an int + CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL xorl %edx,%edx # error code 0 movl %esp,%eax # pt_regs pointer call do_debug jmp ret_from_exception + CFI_ENDPROC .previous .text /* * NMI is doubly nasty. It can happen _while_ we're handling @@ -534,14 +726,18 @@ debug_stack_correct: * fault happened on the sysenter path. */ ENTRY(nmi) + RING0_INT_FRAME pushl %eax + CFI_ADJUST_CFA_OFFSET 4 movl %ss, %eax cmpw $__ESPFIX_SS, %ax popl %eax + CFI_ADJUST_CFA_OFFSET -4 je nmi_16bit_stack cmpl $sysenter_entry,(%esp) je nmi_stack_fixup pushl %eax + CFI_ADJUST_CFA_OFFSET 4 movl %esp,%eax /* Do not access memory above the end of our stack page, * it might not exist. @@ -549,16 +745,19 @@ ENTRY(nmi) andl $(THREAD_SIZE-1),%eax cmpl $(THREAD_SIZE-20),%eax popl %eax + CFI_ADJUST_CFA_OFFSET -4 jae nmi_stack_correct cmpl $sysenter_entry,12(%esp) je nmi_debug_stack_check nmi_stack_correct: pushl %eax + CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_nmi - jmp restore_all + jmp restore_nocheck_notrace + CFI_ENDPROC nmi_stack_fixup: FIX_STACK(12,nmi_stack_correct, 1) @@ -574,94 +773,177 @@ nmi_debug_stack_check: jmp nmi_stack_correct nmi_16bit_stack: + RING0_INT_FRAME /* create the pointer to lss back */ pushl %ss + CFI_ADJUST_CFA_OFFSET 4 pushl %esp + CFI_ADJUST_CFA_OFFSET 4 movzwl %sp, %esp addw $4, (%esp) /* copy the iret frame of 12 bytes */ .rept 3 pushl 16(%esp) + CFI_ADJUST_CFA_OFFSET 4 .endr pushl %eax + CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL FIXUP_ESPFIX_STACK # %eax == %esp + CFI_ADJUST_CFA_OFFSET -20 # the frame has now moved xorl %edx,%edx # zero error code call do_nmi RESTORE_REGS lss 12+4(%esp), %esp # back to 16bit stack 1: iret + CFI_ENDPROC .section __ex_table,"a" .align 4 .long 1b,iret_exc .previous KPROBE_ENTRY(int3) + RING0_INT_FRAME pushl $-1 # mark this as an int + CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_int3 jmp ret_from_exception + CFI_ENDPROC .previous .text ENTRY(overflow) + RING0_INT_FRAME pushl $0 + CFI_ADJUST_CFA_OFFSET 4 pushl $do_overflow + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC ENTRY(bounds) + RING0_INT_FRAME pushl $0 + CFI_ADJUST_CFA_OFFSET 4 pushl $do_bounds + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC ENTRY(invalid_op) + RING0_INT_FRAME pushl $0 + CFI_ADJUST_CFA_OFFSET 4 pushl $do_invalid_op + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC ENTRY(coprocessor_segment_overrun) + RING0_INT_FRAME pushl $0 + CFI_ADJUST_CFA_OFFSET 4 pushl $do_coprocessor_segment_overrun + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC ENTRY(invalid_TSS) + RING0_EC_FRAME pushl $do_invalid_TSS + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC ENTRY(segment_not_present) + RING0_EC_FRAME pushl $do_segment_not_present + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC ENTRY(stack_segment) + RING0_EC_FRAME pushl $do_stack_segment + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC KPROBE_ENTRY(general_protection) + RING0_EC_FRAME pushl $do_general_protection + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC .previous .text ENTRY(alignment_check) + RING0_EC_FRAME pushl $do_alignment_check + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC KPROBE_ENTRY(page_fault) + RING0_EC_FRAME pushl $do_page_fault + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC .previous .text #ifdef CONFIG_X86_MCE ENTRY(machine_check) + RING0_INT_FRAME pushl $0 + CFI_ADJUST_CFA_OFFSET 4 pushl machine_check_vector + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC #endif ENTRY(spurious_interrupt_bug) + RING0_INT_FRAME pushl $0 + CFI_ADJUST_CFA_OFFSET 4 pushl $do_spurious_interrupt_bug + CFI_ADJUST_CFA_OFFSET 4 jmp error_code + CFI_ENDPROC + +#ifdef CONFIG_STACK_UNWIND +ENTRY(arch_unwind_init_running) + CFI_STARTPROC + movl 4(%esp), %edx + movl (%esp), %ecx + leal 4(%esp), %eax + movl %ebx, EBX(%edx) + xorl %ebx, %ebx + movl %ebx, ECX(%edx) + movl %ebx, EDX(%edx) + movl %esi, ESI(%edx) + movl %edi, EDI(%edx) + movl %ebp, EBP(%edx) + movl %ebx, EAX(%edx) + movl $__USER_DS, DS(%edx) + movl $__USER_DS, ES(%edx) + movl %ebx, ORIG_EAX(%edx) + movl %ecx, EIP(%edx) + movl 12(%esp), %ecx + movl $__KERNEL_CS, CS(%edx) + movl %ebx, EFLAGS(%edx) + movl %eax, OLDESP(%edx) + movl 8(%esp), %eax + movl %ecx, 8(%esp) + movl EBX(%edx), %ebx + movl $__KERNEL_DS, OLDSS(%edx) + jmpl *%eax + CFI_ENDPROC +ENDPROC(arch_unwind_init_running) +#endif .section .rodata,"a" #include "syscall_table.S" diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index 3debc2e26542..eb79aa2fa8bb 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -8,7 +8,6 @@ */ .text -#include <linux/config.h> #include <linux/threads.h> #include <linux/linkage.h> #include <asm/segment.h> diff --git a/arch/i386/kernel/hpet.c b/arch/i386/kernel/hpet.c new file mode 100644 index 000000000000..c6737c35815d --- /dev/null +++ b/arch/i386/kernel/hpet.c @@ -0,0 +1,67 @@ +#include <linux/clocksource.h> +#include <linux/errno.h> +#include <linux/hpet.h> +#include <linux/init.h> + +#include <asm/hpet.h> +#include <asm/io.h> + +#define HPET_MASK CLOCKSOURCE_MASK(32) +#define HPET_SHIFT 22 + +/* FSEC = 10^-15 NSEC = 10^-9 */ +#define FSEC_PER_NSEC 1000000 + +static void *hpet_ptr; + +static cycle_t read_hpet(void) +{ + return (cycle_t)readl(hpet_ptr); +} + +static struct clocksource clocksource_hpet = { + .name = "hpet", + .rating = 250, + .read = read_hpet, + .mask = HPET_MASK, + .mult = 0, /* set below */ + .shift = HPET_SHIFT, + .is_continuous = 1, +}; + +static int __init init_hpet_clocksource(void) +{ + unsigned long hpet_period; + void __iomem* hpet_base; + u64 tmp; + + if (!hpet_address) + return -ENODEV; + + /* calculate the hpet address: */ + hpet_base = + (void __iomem*)ioremap_nocache(hpet_address, HPET_MMAP_SIZE); + hpet_ptr = hpet_base + HPET_COUNTER; + + /* calculate the frequency: */ + hpet_period = readl(hpet_base + HPET_PERIOD); + + /* + * hpet period is in femto seconds per cycle + * so we need to convert this to ns/cyc units + * aproximated by mult/2^shift + * + * fsec/cyc * 1nsec/1000000fsec = nsec/cyc = mult/2^shift + * fsec/cyc * 1ns/1000000fsec * 2^shift = mult + * fsec/cyc * 2^shift * 1nsec/1000000fsec = mult + * (fsec/cyc << shift)/1000000 = mult + * (hpet_period << shift)/FSEC_PER_NSEC = mult + */ + tmp = (u64)hpet_period << HPET_SHIFT; + do_div(tmp, FSEC_PER_NSEC); + clocksource_hpet.mult = (u32)tmp; + + return clocksource_register(&clocksource_hpet); +} + +module_init(init_hpet_clocksource); diff --git a/arch/i386/kernel/i386_ksyms.c b/arch/i386/kernel/i386_ksyms.c index 036a9857936f..e3d4b73bfdb0 100644 --- a/arch/i386/kernel/i386_ksyms.c +++ b/arch/i386/kernel/i386_ksyms.c @@ -1,4 +1,3 @@ -#include <linux/config.h> #include <linux/module.h> #include <asm/checksum.h> #include <asm/desc.h> diff --git a/arch/i386/kernel/i387.c b/arch/i386/kernel/i387.c index d75524758daf..665847281ed2 100644 --- a/arch/i386/kernel/i387.c +++ b/arch/i386/kernel/i387.c @@ -8,7 +8,6 @@ * Gareth Hughes <gareth@valinux.com>, May 2000 */ -#include <linux/config.h> #include <linux/sched.h> #include <linux/module.h> #include <asm/processor.h> @@ -25,7 +24,7 @@ #define HAVE_HWFP 1 #endif -static unsigned long mxcsr_feature_mask = 0xffffffff; +static unsigned long mxcsr_feature_mask __read_mostly = 0xffffffff; void mxcsr_feature_mask_init(void) { diff --git a/arch/i386/kernel/i8253.c b/arch/i386/kernel/i8253.c new file mode 100644 index 000000000000..477b24daff53 --- /dev/null +++ b/arch/i386/kernel/i8253.c @@ -0,0 +1,118 @@ +/* + * i8253.c 8253/PIT functions + * + */ +#include <linux/clocksource.h> +#include <linux/spinlock.h> +#include <linux/jiffies.h> +#include <linux/sysdev.h> +#include <linux/module.h> +#include <linux/init.h> + +#include <asm/smp.h> +#include <asm/delay.h> +#include <asm/i8253.h> +#include <asm/io.h> + +#include "io_ports.h" + +DEFINE_SPINLOCK(i8253_lock); +EXPORT_SYMBOL(i8253_lock); + +void setup_pit_timer(void) +{ + unsigned long flags; + + spin_lock_irqsave(&i8253_lock, flags); + outb_p(0x34,PIT_MODE); /* binary, mode 2, LSB/MSB, ch 0 */ + udelay(10); + outb_p(LATCH & 0xff , PIT_CH0); /* LSB */ + udelay(10); + outb(LATCH >> 8 , PIT_CH0); /* MSB */ + spin_unlock_irqrestore(&i8253_lock, flags); +} + +/* + * Since the PIT overflows every tick, its not very useful + * to just read by itself. So use jiffies to emulate a free + * running counter: + */ +static cycle_t pit_read(void) +{ + unsigned long flags; + int count; + u32 jifs; + static int old_count; + static u32 old_jifs; + + spin_lock_irqsave(&i8253_lock, flags); + /* + * Although our caller may have the read side of xtime_lock, + * this is now a seqlock, and we are cheating in this routine + * by having side effects on state that we cannot undo if + * there is a collision on the seqlock and our caller has to + * retry. (Namely, old_jifs and old_count.) So we must treat + * jiffies as volatile despite the lock. We read jiffies + * before latching the timer count to guarantee that although + * the jiffies value might be older than the count (that is, + * the counter may underflow between the last point where + * jiffies was incremented and the point where we latch the + * count), it cannot be newer. + */ + jifs = jiffies; + outb_p(0x00, PIT_MODE); /* latch the count ASAP */ + count = inb_p(PIT_CH0); /* read the latched count */ + count |= inb_p(PIT_CH0) << 8; + + /* VIA686a test code... reset the latch if count > max + 1 */ + if (count > LATCH) { + outb_p(0x34, PIT_MODE); + outb_p(LATCH & 0xff, PIT_CH0); + outb(LATCH >> 8, PIT_CH0); + count = LATCH - 1; + } + + /* + * It's possible for count to appear to go the wrong way for a + * couple of reasons: + * + * 1. The timer counter underflows, but we haven't handled the + * resulting interrupt and incremented jiffies yet. + * 2. Hardware problem with the timer, not giving us continuous time, + * the counter does small "jumps" upwards on some Pentium systems, + * (see c't 95/10 page 335 for Neptun bug.) + * + * Previous attempts to handle these cases intelligently were + * buggy, so we just do the simple thing now. + */ + if (count > old_count && jifs == old_jifs) { + count = old_count; + } + old_count = count; + old_jifs = jifs; + + spin_unlock_irqrestore(&i8253_lock, flags); + + count = (LATCH - 1) - count; + + return (cycle_t)(jifs * LATCH) + count; +} + +static struct clocksource clocksource_pit = { + .name = "pit", + .rating = 110, + .read = pit_read, + .mask = CLOCKSOURCE_MASK(32), + .mult = 0, + .shift = 20, +}; + +static int __init init_pit_clocksource(void) +{ + if (num_possible_cpus() > 4) /* PIT does not scale! */ + return 0; + + clocksource_pit.mult = clocksource_hz2mult(CLOCK_TICK_RATE, 20); + return clocksource_register(&clocksource_pit); +} +module_init(init_pit_clocksource); diff --git a/arch/i386/kernel/i8259.c b/arch/i386/kernel/i8259.c index 323ef8ab3244..d4756d154f47 100644 --- a/arch/i386/kernel/i8259.c +++ b/arch/i386/kernel/i8259.c @@ -1,4 +1,3 @@ -#include <linux/config.h> #include <linux/errno.h> #include <linux/signal.h> #include <linux/sched.h> @@ -132,7 +131,7 @@ void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); io_apic_irqs &= ~(1<<irq); - irq_desc[irq].handler = &i8259A_irq_type; + irq_desc[irq].chip = &i8259A_irq_type; enable_irq(irq); } @@ -175,7 +174,7 @@ static void mask_and_ack_8259A(unsigned int irq) * Lightweight spurious IRQ detection. We do not want * to overdo spurious IRQ handling - it's usually a sign * of hardware problems, so we only do the checks we can - * do without slowing down good hardware unnecesserily. + * do without slowing down good hardware unnecessarily. * * Note that IRQ7 and IRQ15 (the two spurious IRQs * usually resulting from the 8259A-1|2 PICs) occur @@ -271,8 +270,8 @@ static int i8259A_shutdown(struct sys_device *dev) * the kernel initialization code can get it * out of. */ - outb(0xff, 0x21); /* mask all of 8259A-1 */ - outb(0xff, 0xA1); /* mask all of 8259A-1 */ + outb(0xff, PIC_MASTER_IMR); /* mask all of 8259A-1 */ + outb(0xff, PIC_SLAVE_IMR); /* mask all of 8259A-1 */ return 0; } @@ -386,12 +385,12 @@ void __init init_ISA_irqs (void) /* * 16 old-style INTA-cycle interrupts: */ - irq_desc[i].handler = &i8259A_irq_type; + irq_desc[i].chip = &i8259A_irq_type; } else { /* * 'high' PCI IRQs filled in on demand */ - irq_desc[i].handler = &no_irq_type; + irq_desc[i].chip = &no_irq_type; } } } diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index d70f2ade5cde..4fb32c551fe0 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -25,7 +25,6 @@ #include <linux/init.h> #include <linux/delay.h> #include <linux/sched.h> -#include <linux/config.h> #include <linux/smp_lock.h> #include <linux/mc146818rtc.h> #include <linux/compiler.h> @@ -38,6 +37,7 @@ #include <asm/desc.h> #include <asm/timer.h> #include <asm/i8259.h> +#include <asm/nmi.h> #include <mach_apic.h> @@ -50,6 +50,7 @@ atomic_t irq_mis_count; static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; static DEFINE_SPINLOCK(ioapic_lock); +static DEFINE_SPINLOCK(vector_lock); int timer_over_8254 __initdata = 1; @@ -267,7 +268,7 @@ static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t cpumask) # include <linux/slab.h> /* kmalloc() */ # include <linux/timer.h> /* time_after() */ -# ifdef CONFIG_BALANCED_IRQ_DEBUG +#ifdef CONFIG_BALANCED_IRQ_DEBUG # define TDprintk(x...) do { printk("<%ld:%s:%d>: ", jiffies, __FILE__, __LINE__); printk(x); } while (0) # define Dprintk(x...) do { TDprintk(x); } while (0) # else @@ -275,10 +276,15 @@ static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t cpumask) # define Dprintk(x...) # endif - #define IRQBALANCE_CHECK_ARCH -999 -static int irqbalance_disabled = IRQBALANCE_CHECK_ARCH; -static int physical_balance = 0; +#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; @@ -297,12 +303,14 @@ static struct irq_cpu_info { #define CPU_TO_PACKAGEINDEX(i) (first_cpu(cpu_sibling_map[i])) -#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 cpumask_t balance_irq_affinity[NR_IRQS] = { + [0 ... NR_IRQS-1] = CPU_MASK_ALL +}; -static long balanced_irq_interval = MAX_BALANCED_IRQ_INTERVAL; +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) @@ -340,7 +348,7 @@ static inline void balance_irq(int cpu, int irq) if (irqbalance_disabled) return; - cpus_and(allowed_mask, cpu_online_map, irq_affinity[irq]); + 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)); @@ -529,7 +537,9 @@ tryanotherirq: } } - cpus_and(allowed_mask, cpu_online_map, irq_affinity[selected_irq]); + 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); @@ -570,7 +580,7 @@ static int balanced_irq(void *unused) /* push everything to CPU 0 to give us a starting point. */ for (i = 0 ; i < NR_IRQS ; i++) { - pending_irq_cpumask[i] = cpumask_of_cpu(0); + irq_desc[i].pending_mask = cpumask_of_cpu(0); set_pending_irq(i, cpumask_of_cpu(0)); } @@ -1152,10 +1162,17 @@ u8 irq_vector[NR_IRQ_VECTORS] __read_mostly = { FIRST_DEVICE_VECTOR , 0 }; int assign_irq_vector(int irq) { static int current_vector = FIRST_DEVICE_VECTOR, offset = 0; + unsigned long flags; + int vector; + + BUG_ON(irq != AUTO_ASSIGN && (unsigned)irq >= NR_IRQ_VECTORS); - BUG_ON(irq >= NR_IRQ_VECTORS); - if (irq != AUTO_ASSIGN && IO_APIC_VECTOR(irq) > 0) + spin_lock_irqsave(&vector_lock, flags); + + if (irq != AUTO_ASSIGN && IO_APIC_VECTOR(irq) > 0) { + spin_unlock_irqrestore(&vector_lock, flags); return IO_APIC_VECTOR(irq); + } next: current_vector += 8; if (current_vector == SYSCALL_VECTOR) @@ -1163,16 +1180,21 @@ next: if (current_vector >= FIRST_SYSTEM_VECTOR) { offset++; - if (!(offset%8)) + if (!(offset%8)) { + spin_unlock_irqrestore(&vector_lock, flags); return -ENOSPC; + } current_vector = FIRST_DEVICE_VECTOR + offset; } - vector_irq[current_vector] = irq; + vector = current_vector; + vector_irq[vector] = irq; if (irq != AUTO_ASSIGN) - IO_APIC_VECTOR(irq) = current_vector; + IO_APIC_VECTOR(irq) = vector; + + spin_unlock_irqrestore(&vector_lock, flags); - return current_vector; + return vector; } static struct hw_interrupt_type ioapic_level_type; @@ -1182,23 +1204,18 @@ static struct hw_interrupt_type ioapic_edge_type; #define IOAPIC_EDGE 0 #define IOAPIC_LEVEL 1 -static inline void ioapic_register_intr(int irq, int vector, unsigned long trigger) +static void ioapic_register_intr(int irq, int vector, unsigned long trigger) { - if (use_pci_vector() && !platform_legacy_irq(irq)) { - if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || - trigger == IOAPIC_LEVEL) - irq_desc[vector].handler = &ioapic_level_type; - else - irq_desc[vector].handler = &ioapic_edge_type; - set_intr_gate(vector, interrupt[vector]); - } else { - if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || - trigger == IOAPIC_LEVEL) - irq_desc[irq].handler = &ioapic_level_type; - else - irq_desc[irq].handler = &ioapic_edge_type; - set_intr_gate(vector, interrupt[irq]); - } + unsigned idx; + + idx = use_pci_vector() && !platform_legacy_irq(irq) ? vector : irq; + + if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || + trigger == IOAPIC_LEVEL) + irq_desc[idx].chip = &ioapic_level_type; + else + irq_desc[idx].chip = &ioapic_edge_type; + set_intr_gate(vector, interrupt[idx]); } static void __init setup_IO_APIC_irqs(void) @@ -1309,7 +1326,7 @@ static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, in * The timer IRQ doesn't have to know that behind the * scene we have a 8259A-master in AEOI mode ... */ - irq_desc[0].handler = &ioapic_edge_type; + irq_desc[0].chip = &ioapic_edge_type; /* * Add it to the IO-APIC irq-routing table: @@ -2053,6 +2070,13 @@ static void set_ioapic_affinity_vector (unsigned int vector, #endif #endif +static int ioapic_retrigger(unsigned int irq) +{ + send_IPI_self(IO_APIC_VECTOR(irq)); + + return 1; +} + /* * Level and edge triggered IO-APIC interrupts need different handling, * so we use two separate IRQ descriptors. Edge triggered IRQs can be @@ -2072,6 +2096,7 @@ static struct hw_interrupt_type ioapic_edge_type __read_mostly = { #ifdef CONFIG_SMP .set_affinity = set_ioapic_affinity, #endif + .retrigger = ioapic_retrigger, }; static struct hw_interrupt_type ioapic_level_type __read_mostly = { @@ -2085,6 +2110,7 @@ static struct hw_interrupt_type ioapic_level_type __read_mostly = { #ifdef CONFIG_SMP .set_affinity = set_ioapic_affinity, #endif + .retrigger = ioapic_retrigger, }; static inline void init_IO_APIC_traps(void) @@ -2119,7 +2145,7 @@ static inline void init_IO_APIC_traps(void) make_8259A_irq(irq); else /* Strange. Oh, well.. */ - irq_desc[irq].handler = &no_irq_type; + irq_desc[irq].chip = &no_irq_type; } } } @@ -2335,7 +2361,7 @@ static inline void check_timer(void) printk(KERN_INFO "...trying to set up timer as Virtual Wire IRQ..."); disable_8259A_irq(0); - irq_desc[0].handler = &lapic_irq_type; + irq_desc[0].chip = &lapic_irq_type; apic_write_around(APIC_LVT0, APIC_DM_FIXED | vector); /* Fixed mode */ enable_8259A_irq(0); diff --git a/arch/i386/kernel/irq.c b/arch/i386/kernel/irq.c index f3a9c78c4a24..6cb529f60dcc 100644 --- a/arch/i386/kernel/irq.c +++ b/arch/i386/kernel/irq.c @@ -42,8 +42,8 @@ union irq_ctx { u32 stack[THREAD_SIZE/sizeof(u32)]; }; -static union irq_ctx *hardirq_ctx[NR_CPUS]; -static union irq_ctx *softirq_ctx[NR_CPUS]; +static union irq_ctx *hardirq_ctx[NR_CPUS] __read_mostly; +static union irq_ctx *softirq_ctx[NR_CPUS] __read_mostly; #endif /* @@ -53,13 +53,19 @@ static union irq_ctx *softirq_ctx[NR_CPUS]; */ fastcall unsigned int do_IRQ(struct pt_regs *regs) { - /* high bits used in ret_from_ code */ - int irq = regs->orig_eax & 0xff; + /* high bit used in ret_from_ code */ + int irq = ~regs->orig_eax; #ifdef CONFIG_4KSTACKS union irq_ctx *curctx, *irqctx; u32 *isp; #endif + if (unlikely((unsigned)irq >= NR_IRQS)) { + printk(KERN_EMERG "%s: cannot handle IRQ %d\n", + __FUNCTION__, irq); + BUG(); + } + irq_enter(); #ifdef CONFIG_DEBUG_STACKOVERFLOW /* Debugging check for stack overflow: is there less than 1KB free? */ @@ -76,6 +82,10 @@ fastcall unsigned int do_IRQ(struct pt_regs *regs) } #endif + if (!irq_desc[irq].handle_irq) { + __do_IRQ(irq, regs); + goto out_exit; + } #ifdef CONFIG_4KSTACKS curctx = (union irq_ctx *) current_thread_info(); @@ -95,6 +105,14 @@ fastcall unsigned int do_IRQ(struct pt_regs *regs) irqctx->tinfo.task = curctx->tinfo.task; irqctx->tinfo.previous_esp = current_stack_pointer; + /* + * Copy the softirq bits in preempt_count so that the + * softirq checks work in the hardirq context. + */ + irqctx->tinfo.preempt_count = + (irqctx->tinfo.preempt_count & ~SOFTIRQ_MASK) | + (curctx->tinfo.preempt_count & SOFTIRQ_MASK); + asm volatile( " xchgl %%ebx,%%esp \n" " call __do_IRQ \n" @@ -107,6 +125,7 @@ fastcall unsigned int do_IRQ(struct pt_regs *regs) #endif __do_IRQ(irq, regs); +out_exit: irq_exit(); return 1; @@ -147,7 +166,7 @@ void irq_ctx_init(int cpu) irqctx->tinfo.task = NULL; irqctx->tinfo.exec_domain = NULL; irqctx->tinfo.cpu = cpu; - irqctx->tinfo.preempt_count = SOFTIRQ_OFFSET; + irqctx->tinfo.preempt_count = 0; irqctx->tinfo.addr_limit = MAKE_MM_SEG(0); softirq_ctx[cpu] = irqctx; @@ -192,6 +211,10 @@ asmlinkage void do_softirq(void) : "0"(isp) : "memory", "cc", "edx", "ecx", "eax" ); + /* + * Shouldnt happen, we returned above if in_interrupt(): + */ + WARN_ON_ONCE(softirq_count()); } local_irq_restore(flags); @@ -219,7 +242,7 @@ int show_interrupts(struct seq_file *p, void *v) if (i == 0) { seq_printf(p, " "); for_each_online_cpu(j) - seq_printf(p, "CPU%d ",j); + seq_printf(p, "CPU%-8d",j); seq_putc(p, '\n'); } @@ -235,7 +258,7 @@ int show_interrupts(struct seq_file *p, void *v) for_each_online_cpu(j) seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); #endif - seq_printf(p, " %14s", irq_desc[i].handler->typename); + seq_printf(p, " %14s", irq_desc[i].chip->typename); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) @@ -277,13 +300,13 @@ void fixup_irqs(cpumask_t map) if (irq == 2) continue; - cpus_and(mask, irq_affinity[irq], map); + cpus_and(mask, irq_desc[irq].affinity, map); if (any_online_cpu(mask) == NR_CPUS) { printk("Breaking affinity for irq %i\n", irq); mask = map; } - if (irq_desc[irq].handler->set_affinity) - irq_desc[irq].handler->set_affinity(irq, mask); + if (irq_desc[irq].chip->set_affinity) + irq_desc[irq].chip->set_affinity(irq, mask); else if (irq_desc[irq].action && !(warned++)) printk("Cannot set affinity for irq %i\n", irq); } diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index 38806f427849..de2e16e561c0 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -28,7 +28,6 @@ * <prasanna@in.ibm.com> added function-return probes. */ -#include <linux/config.h> #include <linux/kprobes.h> #include <linux/ptrace.h> #include <linux/preempt.h> @@ -57,34 +56,85 @@ static __always_inline void set_jmp_op(void *from, void *to) /* * returns non-zero if opcodes can be boosted. */ -static __always_inline int can_boost(kprobe_opcode_t opcode) +static __always_inline int can_boost(kprobe_opcode_t *opcodes) { - switch (opcode & 0xf0 ) { +#define W(row,b0,b1,b2,b3,b4,b5,b6,b7,b8,b9,ba,bb,bc,bd,be,bf) \ + (((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) | \ + (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) | \ + (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) | \ + (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf)) \ + << (row % 32)) + /* + * Undefined/reserved opcodes, conditional jump, Opcode Extension + * Groups, and some special opcodes can not be boost. + */ + static const unsigned long twobyte_is_boostable[256 / 32] = { + /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */ + /* ------------------------------- */ + W(0x00, 0,0,1,1,0,0,1,0,1,1,0,0,0,0,0,0)| /* 00 */ + W(0x10, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 10 */ + W(0x20, 1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0)| /* 20 */ + W(0x30, 0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 30 */ + W(0x40, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* 40 */ + W(0x50, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 50 */ + W(0x60, 1,1,1,1,1,1,1,1,1,1,1,1,0,0,1,1)| /* 60 */ + W(0x70, 0,0,0,0,1,1,1,1,0,0,0,0,0,0,1,1), /* 70 */ + W(0x80, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)| /* 80 */ + W(0x90, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1), /* 90 */ + W(0xa0, 1,1,0,1,1,1,0,0,1,1,0,1,1,1,0,1)| /* a0 */ + W(0xb0, 1,1,1,1,1,1,1,1,0,0,0,1,1,1,1,1), /* b0 */ + W(0xc0, 1,1,0,0,0,0,0,0,1,1,1,1,1,1,1,1)| /* c0 */ + W(0xd0, 0,1,1,1,0,1,0,0,1,1,0,1,1,1,0,1), /* d0 */ + W(0xe0, 0,1,1,0,0,1,0,0,1,1,0,1,1,1,0,1)| /* e0 */ + W(0xf0, 0,1,1,1,0,1,0,0,1,1,1,0,1,1,1,0) /* f0 */ + /* ------------------------------- */ + /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */ + }; +#undef W + kprobe_opcode_t opcode; + kprobe_opcode_t *orig_opcodes = opcodes; +retry: + if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1) + return 0; + opcode = *(opcodes++); + + /* 2nd-byte opcode */ + if (opcode == 0x0f) { + if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1) + return 0; + return test_bit(*opcodes, twobyte_is_boostable); + } + + switch (opcode & 0xf0) { + case 0x60: + if (0x63 < opcode && opcode < 0x67) + goto retry; /* prefixes */ + /* can't boost Address-size override and bound */ + return (opcode != 0x62 && opcode != 0x67); case 0x70: return 0; /* can't boost conditional jump */ - case 0x90: - /* can't boost call and pushf */ - return opcode != 0x9a && opcode != 0x9c; case 0xc0: - /* can't boost undefined opcodes and soft-interruptions */ - return (0xc1 < opcode && opcode < 0xc6) || - (0xc7 < opcode && opcode < 0xcc) || opcode == 0xcf; + /* can't boost software-interruptions */ + return (0xc1 < opcode && opcode < 0xcc) || opcode == 0xcf; case 0xd0: /* can boost AA* and XLAT */ return (opcode == 0xd4 || opcode == 0xd5 || opcode == 0xd7); case 0xe0: - /* can boost in/out and (may be) jmps */ - return (0xe3 < opcode && opcode != 0xe8); + /* can boost in/out and absolute jmps */ + return ((opcode & 0x04) || opcode == 0xea); case 0xf0: + if ((opcode & 0x0c) == 0 && opcode != 0xf1) + goto retry; /* lock/rep(ne) prefix */ /* clear and set flags can be boost */ return (opcode == 0xf5 || (0xf7 < opcode && opcode < 0xfe)); default: - /* currently, can't boost 2 bytes opcodes */ - return opcode != 0x0f; + if (opcode == 0x26 || opcode == 0x36 || opcode == 0x3e) + goto retry; /* prefixes */ + /* can't boost CS override and call */ + return (opcode != 0x2e && opcode != 0x9a); } } - /* * returns non-zero if opcode modifies the interrupt flag. */ @@ -109,7 +159,7 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p) memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); p->opcode = *p->addr; - if (can_boost(p->opcode)) { + if (can_boost(p->addr)) { p->ainsn.boostable = 0; } else { p->ainsn.boostable = -1; @@ -208,7 +258,9 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) struct kprobe_ctlblk *kcb; #ifdef CONFIG_PREEMPT unsigned pre_preempt_count = preempt_count(); -#endif /* CONFIG_PREEMPT */ +#else + unsigned pre_preempt_count = 1; +#endif addr = (kprobe_opcode_t *)(regs->eip - sizeof(kprobe_opcode_t)); @@ -285,22 +337,14 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) /* handler has already set things up, so skip ss setup */ return 1; - if (p->ainsn.boostable == 1 && -#ifdef CONFIG_PREEMPT - !(pre_preempt_count) && /* - * This enables booster when the direct - * execution path aren't preempted. - */ -#endif /* CONFIG_PREEMPT */ - !p->post_handler && !p->break_handler ) { +ss_probe: + if (pre_preempt_count && p->ainsn.boostable == 1 && !p->post_handler){ /* Boost up -- we can execute copied instructions directly */ reset_current_kprobe(); regs->eip = (unsigned long)p->ainsn.insn; preempt_enable_no_resched(); return 1; } - -ss_probe: prepare_singlestep(p, regs); kcb->kprobe_status = KPROBE_HIT_SS; return 1; @@ -607,7 +651,7 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, struct die_args *args = (struct die_args *)data; int ret = NOTIFY_DONE; - if (args->regs && user_mode(args->regs)) + if (args->regs && user_mode_vm(args->regs)) return ret; switch (val) { diff --git a/arch/i386/kernel/machine_kexec.c b/arch/i386/kernel/machine_kexec.c index f73d7374a2ba..511abe52a94e 100644 --- a/arch/i386/kernel/machine_kexec.c +++ b/arch/i386/kernel/machine_kexec.c @@ -133,9 +133,9 @@ typedef asmlinkage NORET_TYPE void (*relocate_new_kernel_t)( unsigned long start_address, unsigned int has_pae) ATTRIB_NORET; -const extern unsigned char relocate_new_kernel[]; +extern const unsigned char relocate_new_kernel[]; extern void relocate_new_kernel_end(void); -const extern unsigned int relocate_new_kernel_size; +extern const unsigned int relocate_new_kernel_size; /* * A architecture hook called to validate the diff --git a/arch/i386/kernel/mca.c b/arch/i386/kernel/mca.c index 558bb207720f..cd5456f14af4 100644 --- a/arch/i386/kernel/mca.c +++ b/arch/i386/kernel/mca.c @@ -46,7 +46,6 @@ #include <asm/io.h> #include <linux/proc_fs.h> #include <linux/mman.h> -#include <linux/config.h> #include <linux/mm.h> #include <linux/pagemap.h> #include <linux/ioport.h> diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c index e7c138f66c5a..40b44cc0d14b 100644 --- a/arch/i386/kernel/microcode.c +++ b/arch/i386/kernel/microcode.c @@ -91,7 +91,10 @@ MODULE_DESCRIPTION("Intel CPU (IA-32) Microcode Update Driver"); MODULE_AUTHOR("Tigran Aivazian <tigran@veritas.com>"); MODULE_LICENSE("GPL"); -#define MICROCODE_VERSION "1.14" +static int verbose; +module_param(verbose, int, 0644); + +#define MICROCODE_VERSION "1.14a" #define DEFAULT_UCODE_DATASIZE (2000) /* 2000 bytes */ #define MC_HEADER_SIZE (sizeof (microcode_header_t)) /* 48 bytes */ @@ -122,14 +125,15 @@ static unsigned int user_buffer_size; /* it's size */ typedef enum mc_error_code { MC_SUCCESS = 0, - MC_NOTFOUND = 1, - MC_MARKED = 2, - MC_ALLOCATED = 3, + MC_IGNORED = 1, + MC_NOTFOUND = 2, + MC_MARKED = 3, + MC_ALLOCATED = 4, } mc_error_code_t; static struct ucode_cpu_info { unsigned int sig; - unsigned int pf; + unsigned int pf, orig_pf; unsigned int rev; unsigned int cksum; mc_error_code_t err; @@ -164,6 +168,7 @@ static void collect_cpu_info (void *unused) rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]); uci->pf = 1 << ((val[1] >> 18) & 7); } + uci->orig_pf = uci->pf; } wrmsr(MSR_IA32_UCODE_REV, 0, 0); @@ -197,21 +202,34 @@ static inline void mark_microcode_update (int cpu_num, microcode_header_t *mc_he pr_debug(" Checksum 0x%x\n", cksum); if (mc_header->rev < uci->rev) { - printk(KERN_ERR "microcode: CPU%d not 'upgrading' to earlier revision" - " 0x%x (current=0x%x)\n", cpu_num, mc_header->rev, uci->rev); - goto out; + if (uci->err == MC_NOTFOUND) { + uci->err = MC_IGNORED; + uci->cksum = mc_header->rev; + } else if (uci->err == MC_IGNORED && uci->cksum < mc_header->rev) + uci->cksum = mc_header->rev; } else if (mc_header->rev == uci->rev) { - /* notify the caller of success on this cpu */ - uci->err = MC_SUCCESS; - goto out; + if (uci->err < MC_MARKED) { + /* notify the caller of success on this cpu */ + uci->err = MC_SUCCESS; + } + } else if (uci->err != MC_ALLOCATED || mc_header->rev > uci->mc->hdr.rev) { + pr_debug("microcode: CPU%d found a matching microcode update with " + " revision 0x%x (current=0x%x)\n", cpu_num, mc_header->rev, uci->rev); + uci->cksum = cksum; + uci->pf = pf; /* keep the original mc pf for cksum calculation */ + uci->err = MC_MARKED; /* found the match */ + for_each_online_cpu(cpu_num) { + if (ucode_cpu_info + cpu_num != uci + && ucode_cpu_info[cpu_num].mc == uci->mc) { + uci->mc = NULL; + break; + } + } + if (uci->mc != NULL) { + vfree(uci->mc); + uci->mc = NULL; + } } - - pr_debug("microcode: CPU%d found a matching microcode update with " - " revision 0x%x (current=0x%x)\n", cpu_num, mc_header->rev, uci->rev); - uci->cksum = cksum; - uci->pf = pf; /* keep the original mc pf for cksum calculation */ - uci->err = MC_MARKED; /* found the match */ -out: return; } @@ -253,10 +271,8 @@ static int find_matching_ucodes (void) for_each_online_cpu(cpu_num) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; - if (uci->err != MC_NOTFOUND) /* already found a match or not an online cpu*/ - continue; - if (sigmatch(mc_header.sig, uci->sig, mc_header.pf, uci->pf)) + if (sigmatch(mc_header.sig, uci->sig, mc_header.pf, uci->orig_pf)) mark_microcode_update(cpu_num, &mc_header, mc_header.sig, mc_header.pf, mc_header.cksum); } @@ -295,9 +311,8 @@ static int find_matching_ucodes (void) } for_each_online_cpu(cpu_num) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; - if (uci->err != MC_NOTFOUND) /* already found a match or not an online cpu*/ - continue; - if (sigmatch(ext_sig.sig, uci->sig, ext_sig.pf, uci->pf)) { + + if (sigmatch(ext_sig.sig, uci->sig, ext_sig.pf, uci->orig_pf)) { mark_microcode_update(cpu_num, &mc_header, ext_sig.sig, ext_sig.pf, ext_sig.cksum); } } @@ -368,6 +383,13 @@ static void do_update_one (void * unused) struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; if (uci->mc == NULL) { + if (verbose) { + if (uci->err == MC_SUCCESS) + printk(KERN_INFO "microcode: CPU%d already at revision 0x%x\n", + cpu_num, uci->rev); + else + printk(KERN_INFO "microcode: No new microcode data for CPU%d\n", cpu_num); + } return; } @@ -426,6 +448,9 @@ out_free: ucode_cpu_info[j].mc = NULL; } } + if (ucode_cpu_info[i].err == MC_IGNORED && verbose) + printk(KERN_WARNING "microcode: CPU%d not 'upgrading' to earlier revision" + " 0x%x (current=0x%x)\n", i, ucode_cpu_info[i].cksum, ucode_cpu_info[i].rev); } out: return error; @@ -468,7 +493,6 @@ static struct file_operations microcode_fops = { static struct miscdevice microcode_dev = { .minor = MICROCODE_MINOR, .name = "microcode", - .devfs_name = "cpu/microcode", .fops = µcode_fops, }; diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c index 6b1392d33ed5..a70b5fa0ef06 100644 --- a/arch/i386/kernel/mpparse.c +++ b/arch/i386/kernel/mpparse.c @@ -17,7 +17,6 @@ #include <linux/init.h> #include <linux/acpi.h> #include <linux/delay.h> -#include <linux/config.h> #include <linux/bootmem.h> #include <linux/smp_lock.h> #include <linux/kernel_stat.h> diff --git a/arch/i386/kernel/msr.c b/arch/i386/kernel/msr.c index 7a328230e540..d535cdbbfd25 100644 --- a/arch/i386/kernel/msr.c +++ b/arch/i386/kernel/msr.c @@ -24,7 +24,6 @@ */ #include <linux/module.h> -#include <linux/config.h> #include <linux/types.h> #include <linux/errno.h> @@ -251,7 +250,9 @@ static int msr_class_device_create(int i) return err; } -static int msr_class_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) +#ifdef CONFIG_HOTPLUG_CPU +static int msr_class_cpu_callback(struct notifier_block *nfb, + unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -266,10 +267,11 @@ static int msr_class_cpu_callback(struct notifier_block *nfb, unsigned long acti return NOTIFY_OK; } -static struct notifier_block msr_class_cpu_notifier = +static struct notifier_block __cpuinitdata msr_class_cpu_notifier = { .notifier_call = msr_class_cpu_callback, }; +#endif static int __init msr_init(void) { @@ -292,7 +294,7 @@ static int __init msr_init(void) if (err != 0) goto out_class; } - register_cpu_notifier(&msr_class_cpu_notifier); + register_hotcpu_notifier(&msr_class_cpu_notifier); err = 0; goto out; @@ -315,7 +317,7 @@ static void __exit msr_exit(void) class_device_destroy(msr_class, MKDEV(MSR_MAJOR, cpu)); class_destroy(msr_class); unregister_chrdev(MSR_MAJOR, "cpu/msr"); - unregister_cpu_notifier(&msr_class_cpu_notifier); + unregister_hotcpu_notifier(&msr_class_cpu_notifier); } module_init(msr_init); diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index d43b498ec745..2dd928a84645 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -14,21 +14,17 @@ */ #include <linux/config.h> -#include <linux/mm.h> #include <linux/delay.h> -#include <linux/bootmem.h> -#include <linux/smp_lock.h> #include <linux/interrupt.h> -#include <linux/mc146818rtc.h> -#include <linux/kernel_stat.h> #include <linux/module.h> #include <linux/nmi.h> #include <linux/sysdev.h> #include <linux/sysctl.h> +#include <linux/percpu.h> #include <asm/smp.h> -#include <asm/div64.h> #include <asm/nmi.h> +#include <asm/intel_arch_perfmon.h> #include "mach_traps.h" @@ -100,6 +96,9 @@ int nmi_active; (P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT| \ P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE) +#define ARCH_PERFMON_NMI_EVENT_SEL ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL +#define ARCH_PERFMON_NMI_EVENT_UMASK ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK + #ifdef CONFIG_SMP /* The performance counters used by NMI_LOCAL_APIC don't trigger when * the CPU is idle. To make sure the NMI watchdog really ticks on all @@ -108,7 +107,7 @@ int nmi_active; static __init void nmi_cpu_busy(void *data) { volatile int *endflag = data; - local_irq_enable(); + local_irq_enable_in_hardirq(); /* Intentionally don't use cpu_relax here. This is to make sure that the performance counter really ticks, even if there is a simulator or similar that catches the @@ -212,6 +211,8 @@ static int __init setup_nmi_watchdog(char *str) __setup("nmi_watchdog=", setup_nmi_watchdog); +static void disable_intel_arch_watchdog(void); + static void disable_lapic_nmi_watchdog(void) { if (nmi_active <= 0) @@ -221,6 +222,10 @@ static void disable_lapic_nmi_watchdog(void) wrmsr(MSR_K7_EVNTSEL0, 0, 0); break; case X86_VENDOR_INTEL: + if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) { + disable_intel_arch_watchdog(); + break; + } switch (boot_cpu_data.x86) { case 6: if (boot_cpu_data.x86_model > 0xd) @@ -449,6 +454,53 @@ static int setup_p4_watchdog(void) return 1; } +static void disable_intel_arch_watchdog(void) +{ + unsigned ebx; + + /* + * Check whether the Architectural PerfMon supports + * Unhalted Core Cycles Event or not. + * NOTE: Corresponding bit = 0 in ebp indicates event present. + */ + ebx = cpuid_ebx(10); + if (!(ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT)) + wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, 0, 0); +} + +static int setup_intel_arch_watchdog(void) +{ + unsigned int evntsel; + unsigned ebx; + + /* + * Check whether the Architectural PerfMon supports + * Unhalted Core Cycles Event or not. + * NOTE: Corresponding bit = 0 in ebp indicates event present. + */ + ebx = cpuid_ebx(10); + if ((ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT)) + return 0; + + nmi_perfctr_msr = MSR_ARCH_PERFMON_PERFCTR0; + + clear_msr_range(MSR_ARCH_PERFMON_EVENTSEL0, 2); + clear_msr_range(MSR_ARCH_PERFMON_PERFCTR0, 2); + + evntsel = ARCH_PERFMON_EVENTSEL_INT + | ARCH_PERFMON_EVENTSEL_OS + | ARCH_PERFMON_EVENTSEL_USR + | ARCH_PERFMON_NMI_EVENT_SEL + | ARCH_PERFMON_NMI_EVENT_UMASK; + + wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0); + write_watchdog_counter("INTEL_ARCH_PERFCTR0"); + apic_write(APIC_LVTPC, APIC_DM_NMI); + evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE; + wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0); + return 1; +} + void setup_apic_nmi_watchdog (void) { switch (boot_cpu_data.x86_vendor) { @@ -458,6 +510,11 @@ void setup_apic_nmi_watchdog (void) setup_k7_watchdog(); break; case X86_VENDOR_INTEL: + if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) { + if (!setup_intel_arch_watchdog()) + return; + break; + } switch (boot_cpu_data.x86) { case 6: if (boot_cpu_data.x86_model > 0xd) @@ -561,7 +618,8 @@ void nmi_watchdog_tick (struct pt_regs * regs) wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0); apic_write(APIC_LVTPC, APIC_DM_NMI); } - else if (nmi_perfctr_msr == MSR_P6_PERFCTR0) { + else if (nmi_perfctr_msr == MSR_P6_PERFCTR0 || + nmi_perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0) { /* Only P6 based Pentium M need to re-unmask * the apic vector but it doesn't hurt * other P6 variant */ diff --git a/arch/i386/kernel/numaq.c b/arch/i386/kernel/numaq.c index 5f5b075f860a..9000d82c6dc0 100644 --- a/arch/i386/kernel/numaq.c +++ b/arch/i386/kernel/numaq.c @@ -23,7 +23,6 @@ * Send feedback to <gone@us.ibm.com> */ -#include <linux/config.h> #include <linux/mm.h> #include <linux/bootmem.h> #include <linux/mmzone.h> @@ -79,10 +78,12 @@ int __init get_memcfg_numaq(void) return 1; } -static int __init numaq_dsc_disable(void) +static int __init numaq_tsc_disable(void) { - printk(KERN_DEBUG "NUMAQ: disabling TSC\n"); - tsc_disable = 1; + if (num_online_nodes() > 1) { + printk(KERN_DEBUG "NUMAQ: disabling TSC\n"); + tsc_disable = 1; + } return 0; } -core_initcall(numaq_dsc_disable); +arch_initcall(numaq_tsc_disable); diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 6259afea46d1..94e2c87edeaa 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -28,7 +28,6 @@ #include <linux/user.h> #include <linux/a.out.h> #include <linux/interrupt.h> -#include <linux/config.h> #include <linux/utsname.h> #include <linux/delay.h> #include <linux/reboot.h> @@ -102,7 +101,7 @@ void default_idle(void) local_irq_enable(); if (!hlt_counter && boot_cpu_data.hlt_works_ok) { - clear_thread_flag(TIF_POLLING_NRFLAG); + current_thread_info()->status &= ~TS_POLLING; smp_mb__after_clear_bit(); while (!need_resched()) { local_irq_disable(); @@ -111,7 +110,7 @@ void default_idle(void) else local_irq_enable(); } - set_thread_flag(TIF_POLLING_NRFLAG); + current_thread_info()->status |= TS_POLLING; } else { while (!need_resched()) cpu_relax(); @@ -174,7 +173,7 @@ void cpu_idle(void) { int cpu = smp_processor_id(); - set_thread_flag(TIF_POLLING_NRFLAG); + current_thread_info()->status |= TS_POLLING; /* endless idle loop with no priority at all */ while (1) { @@ -312,7 +311,7 @@ void show_regs(struct pt_regs * regs) cr3 = read_cr3(); cr4 = read_cr4_safe(); printk("CR0: %08lx CR2: %08lx CR3: %08lx CR4: %08lx\n", cr0, cr2, cr3, cr4); - show_trace(NULL, ®s->esp); + show_trace(NULL, regs, ®s->esp); } /* diff --git a/arch/i386/kernel/quirks.c b/arch/i386/kernel/quirks.c index 87ccdac84928..9f6ab1789bb0 100644 --- a/arch/i386/kernel/quirks.c +++ b/arch/i386/kernel/quirks.c @@ -1,7 +1,6 @@ /* * This file contains work-arounds for x86 and x86_64 platform bugs. */ -#include <linux/config.h> #include <linux/pci.h> #include <linux/irq.h> diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c index d207242976d3..54cfeabbc5e4 100644 --- a/arch/i386/kernel/reboot.c +++ b/arch/i386/kernel/reboot.c @@ -2,7 +2,6 @@ * linux/arch/i386/kernel/reboot.c */ -#include <linux/config.h> #include <linux/mm.h> #include <linux/module.h> #include <linux/delay.h> diff --git a/arch/i386/kernel/scx200.c b/arch/i386/kernel/scx200.c index 321f5fd26e75..c7d3df23f589 100644 --- a/arch/i386/kernel/scx200.c +++ b/arch/i386/kernel/scx200.c @@ -4,11 +4,11 @@ National Semiconductor SCx200 support. */ -#include <linux/config.h> #include <linux/module.h> #include <linux/errno.h> #include <linux/kernel.h> #include <linux/init.h> +#include <linux/mutex.h> #include <linux/pci.h> #include <linux/scx200.h> @@ -45,11 +45,19 @@ static struct pci_driver scx200_pci_driver = { .probe = scx200_probe, }; -static DEFINE_SPINLOCK(scx200_gpio_config_lock); +static DEFINE_MUTEX(scx200_gpio_config_lock); -static int __devinit scx200_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +static void __devinit scx200_init_shadow(void) { int bank; + + /* read the current values driven on the GPIO signals */ + for (bank = 0; bank < 2; ++bank) + scx200_gpio_shadow[bank] = inl(scx200_gpio_base + 0x10 * bank); +} + +static int __devinit scx200_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ unsigned base; if (pdev->device == PCI_DEVICE_ID_NS_SCx200_BRIDGE || @@ -63,10 +71,7 @@ static int __devinit scx200_probe(struct pci_dev *pdev, const struct pci_device_ } scx200_gpio_base = base; - - /* read the current values driven on the GPIO signals */ - for (bank = 0; bank < 2; ++bank) - scx200_gpio_shadow[bank] = inl(scx200_gpio_base + 0x10 * bank); + scx200_init_shadow(); } else { /* find the base of the Configuration Block */ @@ -87,12 +92,11 @@ static int __devinit scx200_probe(struct pci_dev *pdev, const struct pci_device_ return 0; } -u32 scx200_gpio_configure(int index, u32 mask, u32 bits) +u32 scx200_gpio_configure(unsigned index, u32 mask, u32 bits) { u32 config, new_config; - unsigned long flags; - spin_lock_irqsave(&scx200_gpio_config_lock, flags); + mutex_lock(&scx200_gpio_config_lock); outl(index, scx200_gpio_base + 0x20); config = inl(scx200_gpio_base + 0x24); @@ -100,45 +104,11 @@ u32 scx200_gpio_configure(int index, u32 mask, u32 bits) new_config = (config & mask) | bits; outl(new_config, scx200_gpio_base + 0x24); - spin_unlock_irqrestore(&scx200_gpio_config_lock, flags); + mutex_unlock(&scx200_gpio_config_lock); return config; } -#if 0 -void scx200_gpio_dump(unsigned index) -{ - u32 config = scx200_gpio_configure(index, ~0, 0); - printk(KERN_DEBUG "GPIO%02u: 0x%08lx", index, (unsigned long)config); - - if (config & 1) - printk(" OE"); /* output enabled */ - else - printk(" TS"); /* tristate */ - if (config & 2) - printk(" PP"); /* push pull */ - else - printk(" OD"); /* open drain */ - if (config & 4) - printk(" PUE"); /* pull up enabled */ - else - printk(" PUD"); /* pull up disabled */ - if (config & 8) - printk(" LOCKED"); /* locked */ - if (config & 16) - printk(" LEVEL"); /* level input */ - else - printk(" EDGE"); /* edge input */ - if (config & 32) - printk(" HI"); /* trigger on rising edge */ - else - printk(" LO"); /* trigger on falling edge */ - if (config & 64) - printk(" DEBOUNCE"); /* debounce */ - printk("\n"); -} -#endif /* 0 */ - static int __init scx200_init(void) { printk(KERN_INFO NAME ": NatSemi SCx200 Driver\n"); @@ -159,10 +129,3 @@ EXPORT_SYMBOL(scx200_gpio_base); EXPORT_SYMBOL(scx200_gpio_shadow); EXPORT_SYMBOL(scx200_gpio_configure); EXPORT_SYMBOL(scx200_cb_base); - -/* - Local variables: - compile-command: "make -k -C ../../.. SUBDIRS=arch/i386/kernel modules" - c-basic-offset: 8 - End: -*/ diff --git a/arch/i386/kernel/semaphore.c b/arch/i386/kernel/semaphore.c index 967dc74df9ee..98352c374c76 100644 --- a/arch/i386/kernel/semaphore.c +++ b/arch/i386/kernel/semaphore.c @@ -12,7 +12,6 @@ * * rw semaphores implemented November 1999 by Benjamin LaHaise <bcrl@kvack.org> */ -#include <linux/config.h> #include <asm/semaphore.h> /* diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 846e1639ef7c..08c00d20f162 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -23,7 +23,6 @@ * This file handles the architecture-dependent parts of initialization */ -#include <linux/config.h> #include <linux/sched.h> #include <linux/mm.h> #include <linux/mmzone.h> @@ -60,7 +59,7 @@ #include <asm/io_apic.h> #include <asm/ist.h> #include <asm/io.h> -#include "setup_arch_pre.h" +#include <setup_arch.h> #include <bios_ebda.h> /* Forward Declaration. */ @@ -410,8 +409,8 @@ static void __init limit_regions(unsigned long long size) } } -static void __init add_memory_region(unsigned long long start, - unsigned long long size, int type) +void __init add_memory_region(unsigned long long start, + unsigned long long size, int type) { int x; @@ -474,7 +473,7 @@ static struct change_member *change_point[2*E820MAX] __initdata; static struct e820entry *overlap_list[E820MAX] __initdata; static struct e820entry new_bios[E820MAX] __initdata; -static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) +int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) { struct change_member *change_tmp; unsigned long current_type, last_type; @@ -643,7 +642,7 @@ static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) * thinkpad 560x, for example, does not cooperate with the memory * detection code.) */ -static int __init copy_e820_map(struct e820entry * biosmap, int nr_map) +int __init copy_e820_map(struct e820entry * biosmap, int nr_map) { /* Only one memory region (or negative)? Ignore it */ if (nr_map < 2) @@ -701,12 +700,6 @@ static inline void copy_edd(void) } #endif -/* - * Do NOT EVER look at the BIOS memory size location. - * It does not work on many machines. - */ -#define LOWMEMSIZE() (0x9f000) - static void __init parse_cmdline_early (char ** cmdline_p) { char c = ' ', *to = command_line, *from = saved_command_line; @@ -1320,8 +1313,10 @@ legacy_init_iomem_resources(struct resource *code_resource, struct resource *dat probe_roms(); for (i = 0; i < e820.nr_map; i++) { struct resource *res; +#ifndef CONFIG_RESOURCES_64BIT if (e820.map[i].addr + e820.map[i].size > 0x100000000ULL) continue; +#endif res = kzalloc(sizeof(struct resource), GFP_ATOMIC); switch (e820.map[i].type) { case E820_RAM: res->name = "System RAM"; break; @@ -1423,8 +1418,6 @@ static void __init register_memory(void) pci_mem_start, gapstart, gapsize); } -static char * __init machine_specific_memory_setup(void); - #ifdef CONFIG_MCA static void set_mca_bus(int x) { @@ -1547,15 +1540,18 @@ void __init setup_arch(char **cmdline_p) if (efi_enabled) efi_map_memmap(); -#ifdef CONFIG_X86_IO_APIC - check_acpi_pci(); /* Checks more than just ACPI actually */ -#endif - #ifdef CONFIG_ACPI /* * Parse the ACPI tables for possible boot-time SMP configuration. */ acpi_boot_table_init(); +#endif + +#ifdef CONFIG_X86_IO_APIC + check_acpi_pci(); /* Checks more than just ACPI actually */ +#endif + +#ifdef CONFIG_ACPI acpi_boot_init(); #if defined(CONFIG_SMP) && defined(CONFIG_X86_PC) @@ -1580,6 +1576,7 @@ void __init setup_arch(char **cmdline_p) conswitchp = &dummy_con; #endif #endif + tsc_init(); } static __init int add_pcspkr(void) @@ -1599,7 +1596,6 @@ static __init int add_pcspkr(void) } device_initcall(add_pcspkr); -#include "setup_arch_post.h" /* * Local Variables: * mode:c diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c index 5c352c3a9e7f..43002cfb40c4 100644 --- a/arch/i386/kernel/signal.c +++ b/arch/i386/kernel/signal.c @@ -351,7 +351,7 @@ static int setup_frame(int sig, struct k_sigaction *ka, goto give_sigsegv; } - restorer = &__kernel_sigreturn; + restorer = (void *)VDSO_SYM(&__kernel_sigreturn); if (ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; @@ -447,7 +447,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, goto give_sigsegv; /* Set up to return from userspace. */ - restorer = &__kernel_rt_sigreturn; + restorer = (void *)VDSO_SYM(&__kernel_rt_sigreturn); if (ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; err |= __put_user(restorer, &frame->pretcode); diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c index d134e9643a58..c10789d7a9d3 100644 --- a/arch/i386/kernel/smp.c +++ b/arch/i386/kernel/smp.c @@ -114,7 +114,17 @@ DEFINE_PER_CPU(struct tlb_state, cpu_tlbstate) ____cacheline_aligned = { &init_m static inline int __prepare_ICR (unsigned int shortcut, int vector) { - return APIC_DM_FIXED | shortcut | vector | APIC_DEST_LOGICAL; + unsigned int icr = shortcut | APIC_DEST_LOGICAL; + + switch (vector) { + default: + icr |= APIC_DM_FIXED | vector; + break; + case NMI_VECTOR: + icr |= APIC_DM_NMI; + break; + } + return icr; } static inline int __prepare_ICR2 (unsigned int mask) diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 825b2b4ca721..6f5fea05f1d7 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -34,7 +34,6 @@ * Rusty Russell : Hacked into shape for new "hotplug" boot process. */ #include <linux/module.h> -#include <linux/config.h> #include <linux/init.h> #include <linux/kernel.h> @@ -52,6 +51,7 @@ #include <asm/tlbflush.h> #include <asm/desc.h> #include <asm/arch_hooks.h> +#include <asm/nmi.h> #include <mach_apic.h> #include <mach_wakecpu.h> @@ -66,12 +66,6 @@ int smp_num_siblings = 1; EXPORT_SYMBOL(smp_num_siblings); #endif -/* Package ID of each logical CPU */ -int phys_proc_id[NR_CPUS] __read_mostly = {[0 ... NR_CPUS-1] = BAD_APICID}; - -/* Core ID of each logical CPU */ -int cpu_core_id[NR_CPUS] __read_mostly = {[0 ... NR_CPUS-1] = BAD_APICID}; - /* Last level cache ID of each logical CPU */ int cpu_llc_id[NR_CPUS] __cpuinitdata = {[0 ... NR_CPUS-1] = BAD_APICID}; @@ -257,7 +251,7 @@ static void __init synchronize_tsc_bp (void) * all APs synchronize but they loop on '== num_cpus' */ while (atomic_read(&tsc_count_start) != num_booting_cpus()-1) - mb(); + cpu_relax(); atomic_set(&tsc_count_stop, 0); wmb(); /* @@ -276,7 +270,7 @@ static void __init synchronize_tsc_bp (void) * Wait for all APs to leave the synchronization point: */ while (atomic_read(&tsc_count_stop) != num_booting_cpus()-1) - mb(); + cpu_relax(); atomic_set(&tsc_count_start, 0); wmb(); atomic_inc(&tsc_count_stop); @@ -333,19 +327,21 @@ static void __init synchronize_tsc_ap (void) * this gets called, so we first wait for the BP to * finish SMP initialization: */ - while (!atomic_read(&tsc_start_flag)) mb(); + while (!atomic_read(&tsc_start_flag)) + cpu_relax(); for (i = 0; i < NR_LOOPS; i++) { atomic_inc(&tsc_count_start); while (atomic_read(&tsc_count_start) != num_booting_cpus()) - mb(); + cpu_relax(); rdtscll(tsc_values[smp_processor_id()]); if (i == NR_LOOPS-1) write_tsc(0, 0); atomic_inc(&tsc_count_stop); - while (atomic_read(&tsc_count_stop) != num_booting_cpus()) mb(); + while (atomic_read(&tsc_count_stop) != num_booting_cpus()) + cpu_relax(); } } #undef NR_LOOPS @@ -451,10 +447,12 @@ cpumask_t cpu_coregroup_map(int cpu) struct cpuinfo_x86 *c = cpu_data + cpu; /* * For perf, we return last level cache shared map. - * TBD: when power saving sched policy is added, we will return - * cpu_core_map when power saving policy is enabled + * And for power savings, we return cpu_core_map */ - return c->llc_shared_map; + if (sched_mc_power_savings || sched_smt_power_savings) + return cpu_core_map[cpu]; + else + return c->llc_shared_map; } /* representing cpus for which sibling maps can be computed */ @@ -470,8 +468,8 @@ set_cpu_sibling_map(int cpu) if (smp_num_siblings > 1) { for_each_cpu_mask(i, cpu_sibling_setup_map) { - if (phys_proc_id[cpu] == phys_proc_id[i] && - cpu_core_id[cpu] == cpu_core_id[i]) { + if (c[cpu].phys_proc_id == c[i].phys_proc_id && + c[cpu].cpu_core_id == c[i].cpu_core_id) { cpu_set(i, cpu_sibling_map[cpu]); cpu_set(cpu, cpu_sibling_map[i]); cpu_set(i, cpu_core_map[cpu]); @@ -498,7 +496,7 @@ set_cpu_sibling_map(int cpu) cpu_set(i, c[cpu].llc_shared_map); cpu_set(cpu, c[i].llc_shared_map); } - if (phys_proc_id[cpu] == phys_proc_id[i]) { + if (c[cpu].phys_proc_id == c[i].phys_proc_id) { cpu_set(i, cpu_core_map[cpu]); cpu_set(cpu, cpu_core_map[i]); /* @@ -1053,6 +1051,7 @@ static int __cpuinit __smp_prepare_cpu(int cpu) struct warm_boot_cpu_info info; struct work_struct task; int apicid, ret; + struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); apicid = x86_cpu_to_apicid[cpu]; if (apicid == BAD_APICID) { @@ -1060,6 +1059,18 @@ static int __cpuinit __smp_prepare_cpu(int cpu) goto exit; } + /* + * the CPU isn't initialized at boot time, allocate gdt table here. + * cpu_init will initialize it + */ + if (!cpu_gdt_descr->address) { + cpu_gdt_descr->address = get_zeroed_page(GFP_KERNEL); + if (!cpu_gdt_descr->address) + printk(KERN_CRIT "CPU%d failed to allocate GDT\n", cpu); + ret = -ENOMEM; + goto exit; + } + info.complete = &done; info.apicid = apicid; info.cpu = cpu; @@ -1337,8 +1348,8 @@ remove_siblinginfo(int cpu) cpu_clear(cpu, cpu_sibling_map[sibling]); cpus_clear(cpu_sibling_map[cpu]); cpus_clear(cpu_core_map[cpu]); - phys_proc_id[cpu] = BAD_APICID; - cpu_core_id[cpu] = BAD_APICID; + c[cpu].phys_proc_id = 0; + c[cpu].cpu_core_id = 0; cpu_clear(cpu, cpu_sibling_setup_map); } @@ -1433,7 +1444,7 @@ int __devinit __cpu_up(unsigned int cpu) /* Unleash the CPU! */ cpu_set(cpu, smp_commenced_mask); while (!cpu_isset(cpu, cpu_online_map)) - mb(); + cpu_relax(); return 0; } diff --git a/arch/i386/kernel/srat.c b/arch/i386/kernel/srat.c index 52b3ed5d2cb5..b1809c9a0899 100644 --- a/arch/i386/kernel/srat.c +++ b/arch/i386/kernel/srat.c @@ -23,7 +23,6 @@ * * Send feedback to Pat Gaughen <gone@us.ibm.com> */ -#include <linux/config.h> #include <linux/mm.h> #include <linux/bootmem.h> #include <linux/mmzone.h> @@ -39,7 +38,6 @@ #define NODE_ARRAY_OFFSET(x) ((x) % 8) /* 8 bits/char */ #define BMAP_SET(bmap, bit) ((bmap)[NODE_ARRAY_INDEX(bit)] |= 1 << NODE_ARRAY_OFFSET(bit)) #define BMAP_TEST(bmap, bit) ((bmap)[NODE_ARRAY_INDEX(bit)] & (1 << NODE_ARRAY_OFFSET(bit))) -#define MAX_PXM_DOMAINS 256 /* 1 byte and no promises about values */ /* bitmap length; _PXM is at most 255 */ #define PXM_BITMAP_LEN (MAX_PXM_DOMAINS / 8) static u8 pxm_bitmap[PXM_BITMAP_LEN]; /* bitmap of proximity domains */ @@ -213,19 +211,11 @@ static __init void node_read_chunk(int nid, struct node_memory_chunk_s *memory_c node_end_pfn[nid] = memory_chunk->end_pfn; } -static u8 pxm_to_nid_map[MAX_PXM_DOMAINS];/* _PXM to logical node ID map */ - -int pxm_to_node(int pxm) -{ - return pxm_to_nid_map[pxm]; -} - /* Parse the ACPI Static Resource Affinity Table */ static int __init acpi20_parse_srat(struct acpi_table_srat *sratp) { u8 *start, *end, *p; int i, j, nid; - u8 nid_to_pxm_map[MAX_NUMNODES];/* logical node ID to _PXM map */ start = (u8 *)(&(sratp->reserved) + 1); /* skip header */ p = start; @@ -235,10 +225,6 @@ static int __init acpi20_parse_srat(struct acpi_table_srat *sratp) memset(node_memory_chunk, 0, sizeof(node_memory_chunk)); memset(zholes_size, 0, sizeof(zholes_size)); - /* -1 in these maps means not available */ - memset(pxm_to_nid_map, -1, sizeof(pxm_to_nid_map)); - memset(nid_to_pxm_map, -1, sizeof(nid_to_pxm_map)); - num_memory_chunks = 0; while (p < end) { switch (*p) { @@ -278,9 +264,7 @@ static int __init acpi20_parse_srat(struct acpi_table_srat *sratp) nodes_clear(node_online_map); for (i = 0; i < MAX_PXM_DOMAINS; i++) { if (BMAP_TEST(pxm_bitmap, i)) { - nid = num_online_nodes(); - pxm_to_nid_map[i] = nid; - nid_to_pxm_map[nid] = i; + int nid = acpi_map_pxm_to_node(i); node_set_online(nid); } } @@ -288,7 +272,7 @@ static int __init acpi20_parse_srat(struct acpi_table_srat *sratp) /* set cnode id in memory chunk structure */ for (i = 0; i < num_memory_chunks; i++) - node_memory_chunk[i].nid = pxm_to_nid_map[node_memory_chunk[i].pxm]; + node_memory_chunk[i].nid = pxm_to_node(node_memory_chunk[i].pxm); printk("pxm bitmap: "); for (i = 0; i < sizeof(pxm_bitmap); i++) { diff --git a/arch/i386/kernel/stacktrace.c b/arch/i386/kernel/stacktrace.c new file mode 100644 index 000000000000..e62a037ab399 --- /dev/null +++ b/arch/i386/kernel/stacktrace.c @@ -0,0 +1,98 @@ +/* + * arch/i386/kernel/stacktrace.c + * + * Stack trace management functions + * + * Copyright (C) 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + */ +#include <linux/sched.h> +#include <linux/stacktrace.h> + +static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) +{ + return p > (void *)tinfo && + p < (void *)tinfo + THREAD_SIZE - 3; +} + +/* + * Save stack-backtrace addresses into a stack_trace buffer: + */ +static inline unsigned long +save_context_stack(struct stack_trace *trace, unsigned int skip, + struct thread_info *tinfo, unsigned long *stack, + unsigned long ebp) +{ + unsigned long addr; + +#ifdef CONFIG_FRAME_POINTER + while (valid_stack_ptr(tinfo, (void *)ebp)) { + addr = *(unsigned long *)(ebp + 4); + if (!skip) + trace->entries[trace->nr_entries++] = addr; + else + skip--; + if (trace->nr_entries >= trace->max_entries) + break; + /* + * break out of recursive entries (such as + * end_of_stack_stop_unwind_function): + */ + if (ebp == *(unsigned long *)ebp) + break; + + ebp = *(unsigned long *)ebp; + } +#else + while (valid_stack_ptr(tinfo, stack)) { + addr = *stack++; + if (__kernel_text_address(addr)) { + if (!skip) + trace->entries[trace->nr_entries++] = addr; + else + skip--; + if (trace->nr_entries >= trace->max_entries) + break; + } + } +#endif + + return ebp; +} + +/* + * Save stack-backtrace addresses into a stack_trace buffer. + * If all_contexts is set, all contexts (hardirq, softirq and process) + * are saved. If not set then only the current context is saved. + */ +void save_stack_trace(struct stack_trace *trace, + struct task_struct *task, int all_contexts, + unsigned int skip) +{ + unsigned long ebp; + unsigned long *stack = &ebp; + + WARN_ON(trace->nr_entries || !trace->max_entries); + + if (!task || task == current) { + /* Grab ebp right from our regs: */ + asm ("movl %%ebp, %0" : "=r" (ebp)); + } else { + /* ebp is the last reg pushed by switch_to(): */ + ebp = *(unsigned long *) task->thread.esp; + } + + while (1) { + struct thread_info *context = (struct thread_info *) + ((unsigned long)stack & (~(THREAD_SIZE - 1))); + + ebp = save_context_stack(trace, skip, context, stack, ebp); + stack = (unsigned long *)context->previous_esp; + if (!all_contexts || !stack || + trace->nr_entries >= trace->max_entries) + break; + trace->entries[trace->nr_entries++] = ULONG_MAX; + if (trace->nr_entries >= trace->max_entries) + break; + } +} + diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S index f48bef15b4f0..dd63d4775398 100644 --- a/arch/i386/kernel/syscall_table.S +++ b/arch/i386/kernel/syscall_table.S @@ -315,3 +315,5 @@ ENTRY(sys_call_table) .long sys_splice .long sys_sync_file_range .long sys_tee /* 315 */ + .long sys_vmsplice + .long sys_move_pages diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c index 0bada1870bdf..713ba39d32c6 100644 --- a/arch/i386/kernel/sysenter.c +++ b/arch/i386/kernel/sysenter.c @@ -2,6 +2,8 @@ * linux/arch/i386/kernel/sysenter.c * * (C) Copyright 2002 Linus Torvalds + * Portions based on the vdso-randomization code from exec-shield: + * Copyright(C) 2005-2006, Red Hat, Inc., Ingo Molnar * * This file contains the needed initializations to support sysenter. */ @@ -13,12 +15,31 @@ #include <linux/gfp.h> #include <linux/string.h> #include <linux/elf.h> +#include <linux/mm.h> +#include <linux/module.h> #include <asm/cpufeature.h> #include <asm/msr.h> #include <asm/pgtable.h> #include <asm/unistd.h> +/* + * Should the kernel map a VDSO page into processes and pass its + * address down to glibc upon exec()? + */ +unsigned int __read_mostly vdso_enabled = 1; + +EXPORT_SYMBOL_GPL(vdso_enabled); + +static int __init vdso_setup(char *s) +{ + vdso_enabled = simple_strtoul(s, NULL, 0); + + return 1; +} + +__setup("vdso=", vdso_setup); + extern asmlinkage void sysenter_entry(void); void enable_sep_cpu(void) @@ -45,23 +66,120 @@ void enable_sep_cpu(void) */ extern const char vsyscall_int80_start, vsyscall_int80_end; extern const char vsyscall_sysenter_start, vsyscall_sysenter_end; +static void *syscall_page; int __init sysenter_setup(void) { - void *page = (void *)get_zeroed_page(GFP_ATOMIC); + syscall_page = (void *)get_zeroed_page(GFP_ATOMIC); - __set_fixmap(FIX_VSYSCALL, __pa(page), PAGE_READONLY_EXEC); +#ifdef CONFIG_COMPAT_VDSO + __set_fixmap(FIX_VDSO, __pa(syscall_page), PAGE_READONLY); + printk("Compat vDSO mapped to %08lx.\n", __fix_to_virt(FIX_VDSO)); +#else + /* + * In the non-compat case the ELF coredumping code needs the fixmap: + */ + __set_fixmap(FIX_VDSO, __pa(syscall_page), PAGE_KERNEL_RO); +#endif if (!boot_cpu_has(X86_FEATURE_SEP)) { - memcpy(page, + memcpy(syscall_page, &vsyscall_int80_start, &vsyscall_int80_end - &vsyscall_int80_start); return 0; } - memcpy(page, + memcpy(syscall_page, &vsyscall_sysenter_start, &vsyscall_sysenter_end - &vsyscall_sysenter_start); return 0; } + +static struct page *syscall_nopage(struct vm_area_struct *vma, + unsigned long adr, int *type) +{ + struct page *p = virt_to_page(adr - vma->vm_start + syscall_page); + get_page(p); + return p; +} + +/* Prevent VMA merging */ +static void syscall_vma_close(struct vm_area_struct *vma) +{ +} + +static struct vm_operations_struct syscall_vm_ops = { + .close = syscall_vma_close, + .nopage = syscall_nopage, +}; + +/* Defined in vsyscall-sysenter.S */ +extern void SYSENTER_RETURN; + +/* Setup a VMA at program startup for the vsyscall page */ +int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack) +{ + struct vm_area_struct *vma; + struct mm_struct *mm = current->mm; + unsigned long addr; + int ret; + + down_write(&mm->mmap_sem); + addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0); + if (IS_ERR_VALUE(addr)) { + ret = addr; + goto up_fail; + } + + vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL); + if (!vma) { + ret = -ENOMEM; + goto up_fail; + } + + vma->vm_start = addr; + vma->vm_end = addr + PAGE_SIZE; + /* MAYWRITE to allow gdb to COW and set breakpoints */ + vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE; + vma->vm_flags |= mm->def_flags; + vma->vm_page_prot = protection_map[vma->vm_flags & 7]; + vma->vm_ops = &syscall_vm_ops; + vma->vm_mm = mm; + + ret = insert_vm_struct(mm, vma); + if (unlikely(ret)) { + kmem_cache_free(vm_area_cachep, vma); + goto up_fail; + } + + current->mm->context.vdso = (void *)addr; + current_thread_info()->sysenter_return = + (void *)VDSO_SYM(&SYSENTER_RETURN); + mm->total_vm++; +up_fail: + up_write(&mm->mmap_sem); + return ret; +} + +const char *arch_vma_name(struct vm_area_struct *vma) +{ + if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso) + return "[vdso]"; + return NULL; +} + +struct vm_area_struct *get_gate_vma(struct task_struct *tsk) +{ + return NULL; +} + +int in_gate_area(struct task_struct *task, unsigned long addr) +{ + return 0; +} + +int in_gate_area_no_task(unsigned long addr) +{ + return 0; +} diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c index 9d3074759856..316421a7f56f 100644 --- a/arch/i386/kernel/time.c +++ b/arch/i386/kernel/time.c @@ -60,7 +60,6 @@ #include "mach_time.h" #include <linux/timex.h> -#include <linux/config.h> #include <asm/hpet.h> @@ -82,13 +81,6 @@ extern unsigned long wall_jiffies; DEFINE_SPINLOCK(rtc_lock); EXPORT_SYMBOL(rtc_lock); -#include <asm/i8253.h> - -DEFINE_SPINLOCK(i8253_lock); -EXPORT_SYMBOL(i8253_lock); - -struct timer_opts *cur_timer __read_mostly = &timer_none; - /* * This is a special lock that is owned by the CPU and holds the index * register we are working with. It is required for NMI access to the @@ -118,99 +110,19 @@ void rtc_cmos_write(unsigned char val, unsigned char addr) } EXPORT_SYMBOL(rtc_cmos_write); -/* - * This version of gettimeofday has microsecond resolution - * and better than microsecond precision on fast x86 machines with TSC. - */ -void do_gettimeofday(struct timeval *tv) -{ - unsigned long seq; - unsigned long usec, sec; - unsigned long max_ntp_tick; - - do { - unsigned long lost; - - seq = read_seqbegin(&xtime_lock); - - usec = cur_timer->get_offset(); - lost = jiffies - wall_jiffies; - - /* - * If time_adjust is negative then NTP is slowing the clock - * so make sure not to go into next possible interval. - * Better to lose some accuracy than have time go backwards.. - */ - if (unlikely(time_adjust < 0)) { - max_ntp_tick = (USEC_PER_SEC / HZ) - tickadj; - usec = min(usec, max_ntp_tick); - - if (lost) - usec += lost * max_ntp_tick; - } - else if (unlikely(lost)) - usec += lost * (USEC_PER_SEC / HZ); - - sec = xtime.tv_sec; - usec += (xtime.tv_nsec / 1000); - } while (read_seqretry(&xtime_lock, seq)); - - while (usec >= 1000000) { - usec -= 1000000; - sec++; - } - - tv->tv_sec = sec; - tv->tv_usec = usec; -} - -EXPORT_SYMBOL(do_gettimeofday); - -int do_settimeofday(struct timespec *tv) -{ - time_t wtm_sec, sec = tv->tv_sec; - long wtm_nsec, nsec = tv->tv_nsec; - - if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) - return -EINVAL; - - write_seqlock_irq(&xtime_lock); - /* - * This is revolting. We need to set "xtime" correctly. However, the - * value in this location is the value at the most recent update of - * wall time. Discover what correction gettimeofday() would have - * made, and then undo it! - */ - nsec -= cur_timer->get_offset() * NSEC_PER_USEC; - nsec -= (jiffies - wall_jiffies) * TICK_NSEC; - - wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec); - wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec); - - set_normalized_timespec(&xtime, sec, nsec); - set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec); - - ntp_clear(); - write_sequnlock_irq(&xtime_lock); - clock_was_set(); - return 0; -} - -EXPORT_SYMBOL(do_settimeofday); - static int set_rtc_mmss(unsigned long nowtime) { int retval; - - WARN_ON(irqs_disabled()); + unsigned long flags; /* gets recalled with irq locally disabled */ - spin_lock_irq(&rtc_lock); + /* XXX - does irqsave resolve this? -johnstul */ + spin_lock_irqsave(&rtc_lock, flags); if (efi_enabled) retval = efi_set_rtc_mmss(nowtime); else retval = mach_set_rtc_mmss(nowtime); - spin_unlock_irq(&rtc_lock); + spin_unlock_irqrestore(&rtc_lock, flags); return retval; } @@ -218,16 +130,6 @@ static int set_rtc_mmss(unsigned long nowtime) int timer_ack; -/* monotonic_clock(): returns # of nanoseconds passed since time_init() - * Note: This function is required to return accurate - * time even in the absence of multiple timer ticks. - */ -unsigned long long monotonic_clock(void) -{ - return cur_timer->monotonic_clock(); -} -EXPORT_SYMBOL(monotonic_clock); - #if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER) unsigned long profile_pc(struct pt_regs *regs) { @@ -242,11 +144,21 @@ EXPORT_SYMBOL(profile_pc); #endif /* - * timer_interrupt() needs to keep up the real-time clock, - * as well as call the "do_timer()" routine every clocktick + * This is the same as the above, except we _also_ save the current + * Time Stamp Counter value at the time of the timer interrupt, so that + * we later on can estimate the time of day more exactly. */ -static inline void do_timer_interrupt(int irq, struct pt_regs *regs) +irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) { + /* + * Here we are in the timer irq handler. We just have irqs locally + * disabled but we don't know if the timer_bh is running on the other + * CPU. We need to avoid to SMP race with it. NOTE: we don' t need + * the irq version of write_lock because as just said we have irq + * locally disabled. -arca + */ + write_seqlock(&xtime_lock); + #ifdef CONFIG_X86_IO_APIC if (timer_ack) { /* @@ -279,27 +191,6 @@ static inline void do_timer_interrupt(int irq, struct pt_regs *regs) irq = inb_p( 0x61 ); /* read the current state */ outb_p( irq|0x80, 0x61 ); /* reset the IRQ */ } -} - -/* - * This is the same as the above, except we _also_ save the current - * Time Stamp Counter value at the time of the timer interrupt, so that - * we later on can estimate the time of day more exactly. - */ -irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) -{ - /* - * Here we are in the timer irq handler. We just have irqs locally - * disabled but we don't know if the timer_bh is running on the other - * CPU. We need to avoid to SMP race with it. NOTE: we don' t need - * the irq version of write_lock because as just said we have irq - * locally disabled. -arca - */ - write_seqlock(&xtime_lock); - - cur_timer->mark_offset(); - - do_timer_interrupt(irq, regs); write_sequnlock(&xtime_lock); @@ -380,7 +271,6 @@ void notify_arch_cmos_timer(void) static long clock_cmos_diff, sleep_start; -static struct timer_opts *last_timer; static int timer_suspend(struct sys_device *dev, pm_message_t state) { /* @@ -389,10 +279,6 @@ static int timer_suspend(struct sys_device *dev, pm_message_t state) clock_cmos_diff = -get_cmos_time(); clock_cmos_diff += get_seconds(); sleep_start = get_cmos_time(); - last_timer = cur_timer; - cur_timer = &timer_none; - if (last_timer->suspend) - last_timer->suspend(state); return 0; } @@ -415,10 +301,6 @@ static int timer_resume(struct sys_device *dev) jiffies_64 += sleep_length; wall_jiffies += sleep_length; write_sequnlock_irqrestore(&xtime_lock, flags); - if (last_timer->resume) - last_timer->resume(); - cur_timer = last_timer; - last_timer = NULL; touch_softlockup_watchdog(); return 0; } @@ -460,9 +342,6 @@ static void __init hpet_time_init(void) printk("Using HPET for base-timer\n"); } - cur_timer = select_timer(); - printk(KERN_INFO "Using %s for high-res timesource\n",cur_timer->name); - time_init_hook(); } #endif @@ -484,8 +363,5 @@ void __init time_init(void) set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec); - cur_timer = select_timer(); - printk(KERN_INFO "Using %s for high-res timesource\n",cur_timer->name); - time_init_hook(); } diff --git a/arch/i386/kernel/time_hpet.c b/arch/i386/kernel/time_hpet.c index a529f0cdce17..14a1376fedd1 100644 --- a/arch/i386/kernel/time_hpet.c +++ b/arch/i386/kernel/time_hpet.c @@ -18,7 +18,6 @@ #include <asm/apic.h> #include <linux/timex.h> -#include <linux/config.h> #include <asm/hpet.h> #include <linux/hpet.h> diff --git a/arch/i386/kernel/timers/Makefile b/arch/i386/kernel/timers/Makefile deleted file mode 100644 index 8fa12be658dd..000000000000 --- a/arch/i386/kernel/timers/Makefile +++ /dev/null @@ -1,9 +0,0 @@ -# -# Makefile for x86 timers -# - -obj-y := timer.o timer_none.o timer_tsc.o timer_pit.o common.o - -obj-$(CONFIG_X86_CYCLONE_TIMER) += timer_cyclone.o -obj-$(CONFIG_HPET_TIMER) += timer_hpet.o -obj-$(CONFIG_X86_PM_TIMER) += timer_pm.o diff --git a/arch/i386/kernel/timers/common.c b/arch/i386/kernel/timers/common.c deleted file mode 100644 index 8163fe0cf1f0..000000000000 --- a/arch/i386/kernel/timers/common.c +++ /dev/null @@ -1,172 +0,0 @@ -/* - * Common functions used across the timers go here - */ - -#include <linux/init.h> -#include <linux/timex.h> -#include <linux/errno.h> -#include <linux/jiffies.h> -#include <linux/module.h> - -#include <asm/io.h> -#include <asm/timer.h> -#include <asm/hpet.h> - -#include "mach_timer.h" - -/* ------ Calibrate the TSC ------- - * Return 2^32 * (1 / (TSC clocks per usec)) for do_fast_gettimeoffset(). - * Too much 64-bit arithmetic here to do this cleanly in C, and for - * accuracy's sake we want to keep the overhead on the CTC speaker (channel 2) - * output busy loop as low as possible. We avoid reading the CTC registers - * directly because of the awkward 8-bit access mechanism of the 82C54 - * device. - */ - -#define CALIBRATE_TIME (5 * 1000020/HZ) - -unsigned long calibrate_tsc(void) -{ - mach_prepare_counter(); - - { - unsigned long startlow, starthigh; - unsigned long endlow, endhigh; - unsigned long count; - - rdtsc(startlow,starthigh); - mach_countup(&count); - rdtsc(endlow,endhigh); - - - /* Error: ECTCNEVERSET */ - if (count <= 1) - goto bad_ctc; - - /* 64-bit subtract - gcc just messes up with long longs */ - __asm__("subl %2,%0\n\t" - "sbbl %3,%1" - :"=a" (endlow), "=d" (endhigh) - :"g" (startlow), "g" (starthigh), - "0" (endlow), "1" (endhigh)); - - /* Error: ECPUTOOFAST */ - if (endhigh) - goto bad_ctc; - - /* Error: ECPUTOOSLOW */ - if (endlow <= CALIBRATE_TIME) - goto bad_ctc; - - __asm__("divl %2" - :"=a" (endlow), "=d" (endhigh) - :"r" (endlow), "0" (0), "1" (CALIBRATE_TIME)); - - return endlow; - } - - /* - * The CTC wasn't reliable: we got a hit on the very first read, - * or the CPU was so fast/slow that the quotient wouldn't fit in - * 32 bits.. - */ -bad_ctc: - return 0; -} - -#ifdef CONFIG_HPET_TIMER -/* ------ Calibrate the TSC using HPET ------- - * Return 2^32 * (1 / (TSC clocks per usec)) for getting the CPU freq. - * Second output is parameter 1 (when non NULL) - * Set 2^32 * (1 / (tsc per HPET clk)) for delay_hpet(). - * calibrate_tsc() calibrates the processor TSC by comparing - * it to the HPET timer of known frequency. - * Too much 64-bit arithmetic here to do this cleanly in C - */ -#define CALIBRATE_CNT_HPET (5 * hpet_tick) -#define CALIBRATE_TIME_HPET (5 * KERNEL_TICK_USEC) - -unsigned long __devinit calibrate_tsc_hpet(unsigned long *tsc_hpet_quotient_ptr) -{ - unsigned long tsc_startlow, tsc_starthigh; - unsigned long tsc_endlow, tsc_endhigh; - unsigned long hpet_start, hpet_end; - unsigned long result, remain; - - hpet_start = hpet_readl(HPET_COUNTER); - rdtsc(tsc_startlow, tsc_starthigh); - do { - hpet_end = hpet_readl(HPET_COUNTER); - } while ((hpet_end - hpet_start) < CALIBRATE_CNT_HPET); - rdtsc(tsc_endlow, tsc_endhigh); - - /* 64-bit subtract - gcc just messes up with long longs */ - __asm__("subl %2,%0\n\t" - "sbbl %3,%1" - :"=a" (tsc_endlow), "=d" (tsc_endhigh) - :"g" (tsc_startlow), "g" (tsc_starthigh), - "0" (tsc_endlow), "1" (tsc_endhigh)); - - /* Error: ECPUTOOFAST */ - if (tsc_endhigh) - goto bad_calibration; - - /* Error: ECPUTOOSLOW */ - if (tsc_endlow <= CALIBRATE_TIME_HPET) - goto bad_calibration; - - ASM_DIV64_REG(result, remain, tsc_endlow, 0, CALIBRATE_TIME_HPET); - if (remain > (tsc_endlow >> 1)) - result++; /* rounding the result */ - - if (tsc_hpet_quotient_ptr) { - unsigned long tsc_hpet_quotient; - - ASM_DIV64_REG(tsc_hpet_quotient, remain, tsc_endlow, 0, - CALIBRATE_CNT_HPET); - if (remain > (tsc_endlow >> 1)) - tsc_hpet_quotient++; /* rounding the result */ - *tsc_hpet_quotient_ptr = tsc_hpet_quotient; - } - - return result; -bad_calibration: - /* - * the CPU was so fast/slow that the quotient wouldn't fit in - * 32 bits.. - */ - return 0; -} -#endif - - -unsigned long read_timer_tsc(void) -{ - unsigned long retval; - rdtscl(retval); - return retval; -} - - -/* calculate cpu_khz */ -void init_cpu_khz(void) -{ - if (cpu_has_tsc) { - unsigned long tsc_quotient = calibrate_tsc(); - if (tsc_quotient) { - /* report CPU clock rate in Hz. - * The formula is (10^6 * 2^32) / (2^32 * 1 / (clocks/us)) = - * clock/second. Our precision is about 100 ppm. - */ - { unsigned long eax=0, edx=1000; - __asm__("divl %2" - :"=a" (cpu_khz), "=d" (edx) - :"r" (tsc_quotient), - "0" (eax), "1" (edx)); - printk("Detected %u.%03u MHz processor.\n", - cpu_khz / 1000, cpu_khz % 1000); - } - } - } -} - diff --git a/arch/i386/kernel/timers/timer.c b/arch/i386/kernel/timers/timer.c deleted file mode 100644 index 7e39ed8e33f8..000000000000 --- a/arch/i386/kernel/timers/timer.c +++ /dev/null @@ -1,75 +0,0 @@ -#include <linux/init.h> -#include <linux/kernel.h> -#include <linux/string.h> -#include <asm/timer.h> - -#ifdef CONFIG_HPET_TIMER -/* - * HPET memory read is slower than tsc reads, but is more dependable as it - * always runs at constant frequency and reduces complexity due to - * cpufreq. So, we prefer HPET timer to tsc based one. Also, we cannot use - * timer_pit when HPET is active. So, we default to timer_tsc. - */ -#endif -/* list of timers, ordered by preference, NULL terminated */ -static struct init_timer_opts* __initdata timers[] = { -#ifdef CONFIG_X86_CYCLONE_TIMER - &timer_cyclone_init, -#endif -#ifdef CONFIG_HPET_TIMER - &timer_hpet_init, -#endif -#ifdef CONFIG_X86_PM_TIMER - &timer_pmtmr_init, -#endif - &timer_tsc_init, - &timer_pit_init, - NULL, -}; - -static char clock_override[10] __initdata; - -static int __init clock_setup(char* str) -{ - if (str) - strlcpy(clock_override, str, sizeof(clock_override)); - return 1; -} -__setup("clock=", clock_setup); - - -/* The chosen timesource has been found to be bad. - * Fall back to a known good timesource (the PIT) - */ -void clock_fallback(void) -{ - cur_timer = &timer_pit; -} - -/* iterates through the list of timers, returning the first - * one that initializes successfully. - */ -struct timer_opts* __init select_timer(void) -{ - int i = 0; - - /* find most preferred working timer */ - while (timers[i]) { - if (timers[i]->init) - if (timers[i]->init(clock_override) == 0) - return timers[i]->opts; - ++i; - } - - panic("select_timer: Cannot find a suitable timer\n"); - return NULL; -} - -int read_current_timer(unsigned long *timer_val) -{ - if (cur_timer->read_timer) { - *timer_val = cur_timer->read_timer(); - return 0; - } - return -1; -} diff --git a/arch/i386/kernel/timers/timer_cyclone.c b/arch/i386/kernel/timers/timer_cyclone.c deleted file mode 100644 index 13892a65c941..000000000000 --- a/arch/i386/kernel/timers/timer_cyclone.c +++ /dev/null @@ -1,259 +0,0 @@ -/* Cyclone-timer: - * This code implements timer_ops for the cyclone counter found - * on IBM x440, x360, and other Summit based systems. - * - * Copyright (C) 2002 IBM, John Stultz (johnstul@us.ibm.com) - */ - - -#include <linux/spinlock.h> -#include <linux/init.h> -#include <linux/timex.h> -#include <linux/errno.h> -#include <linux/string.h> -#include <linux/jiffies.h> - -#include <asm/timer.h> -#include <asm/io.h> -#include <asm/pgtable.h> -#include <asm/fixmap.h> -#include <asm/i8253.h> - -#include "io_ports.h" - -/* Number of usecs that the last interrupt was delayed */ -static int delay_at_last_interrupt; - -#define CYCLONE_CBAR_ADDR 0xFEB00CD0 -#define CYCLONE_PMCC_OFFSET 0x51A0 -#define CYCLONE_MPMC_OFFSET 0x51D0 -#define CYCLONE_MPCS_OFFSET 0x51A8 -#define CYCLONE_TIMER_FREQ 100000000 -#define CYCLONE_TIMER_MASK (((u64)1<<40)-1) /* 40 bit mask */ -int use_cyclone = 0; - -static u32* volatile cyclone_timer; /* Cyclone MPMC0 register */ -static u32 last_cyclone_low; -static u32 last_cyclone_high; -static unsigned long long monotonic_base; -static seqlock_t monotonic_lock = SEQLOCK_UNLOCKED; - -/* helper macro to atomically read both cyclone counter registers */ -#define read_cyclone_counter(low,high) \ - do{ \ - high = cyclone_timer[1]; low = cyclone_timer[0]; \ - } while (high != cyclone_timer[1]); - - -static void mark_offset_cyclone(void) -{ - unsigned long lost, delay; - unsigned long delta = last_cyclone_low; - int count; - unsigned long long this_offset, last_offset; - - write_seqlock(&monotonic_lock); - last_offset = ((unsigned long long)last_cyclone_high<<32)|last_cyclone_low; - - spin_lock(&i8253_lock); - read_cyclone_counter(last_cyclone_low,last_cyclone_high); - - /* read values for delay_at_last_interrupt */ - outb_p(0x00, 0x43); /* latch the count ASAP */ - - count = inb_p(0x40); /* read the latched count */ - count |= inb(0x40) << 8; - - /* - * VIA686a test code... reset the latch if count > max + 1 - * from timer_pit.c - cjb - */ - if (count > LATCH) { - outb_p(0x34, PIT_MODE); - outb_p(LATCH & 0xff, PIT_CH0); - outb(LATCH >> 8, PIT_CH0); - count = LATCH - 1; - } - spin_unlock(&i8253_lock); - - /* lost tick compensation */ - delta = last_cyclone_low - delta; - delta /= (CYCLONE_TIMER_FREQ/1000000); - delta += delay_at_last_interrupt; - lost = delta/(1000000/HZ); - delay = delta%(1000000/HZ); - if (lost >= 2) - jiffies_64 += lost-1; - - /* update the monotonic base value */ - this_offset = ((unsigned long long)last_cyclone_high<<32)|last_cyclone_low; - monotonic_base += (this_offset - last_offset) & CYCLONE_TIMER_MASK; - write_sequnlock(&monotonic_lock); - - /* calculate delay_at_last_interrupt */ - count = ((LATCH-1) - count) * TICK_SIZE; - delay_at_last_interrupt = (count + LATCH/2) / LATCH; - - - /* catch corner case where tick rollover occured - * between cyclone and pit reads (as noted when - * usec delta is > 90% # of usecs/tick) - */ - if (lost && abs(delay - delay_at_last_interrupt) > (900000/HZ)) - jiffies_64++; -} - -static unsigned long get_offset_cyclone(void) -{ - u32 offset; - - if(!cyclone_timer) - return delay_at_last_interrupt; - - /* Read the cyclone timer */ - offset = cyclone_timer[0]; - - /* .. relative to previous jiffy */ - offset = offset - last_cyclone_low; - - /* convert cyclone ticks to microseconds */ - /* XXX slow, can we speed this up? */ - offset = offset/(CYCLONE_TIMER_FREQ/1000000); - - /* our adjusted time offset in microseconds */ - return delay_at_last_interrupt + offset; -} - -static unsigned long long monotonic_clock_cyclone(void) -{ - u32 now_low, now_high; - unsigned long long last_offset, this_offset, base; - unsigned long long ret; - unsigned seq; - - /* atomically read monotonic base & last_offset */ - do { - seq = read_seqbegin(&monotonic_lock); - last_offset = ((unsigned long long)last_cyclone_high<<32)|last_cyclone_low; - base = monotonic_base; - } while (read_seqretry(&monotonic_lock, seq)); - - - /* Read the cyclone counter */ - read_cyclone_counter(now_low,now_high); - this_offset = ((unsigned long long)now_high<<32)|now_low; - - /* convert to nanoseconds */ - ret = base + ((this_offset - last_offset)&CYCLONE_TIMER_MASK); - return ret * (1000000000 / CYCLONE_TIMER_FREQ); -} - -static int __init init_cyclone(char* override) -{ - u32* reg; - u32 base; /* saved cyclone base address */ - u32 pageaddr; /* page that contains cyclone_timer register */ - u32 offset; /* offset from pageaddr to cyclone_timer register */ - int i; - - /* check clock override */ - if (override[0] && strncmp(override,"cyclone",7)) - return -ENODEV; - - /*make sure we're on a summit box*/ - if(!use_cyclone) return -ENODEV; - - printk(KERN_INFO "Summit chipset: Starting Cyclone Counter.\n"); - - /* find base address */ - pageaddr = (CYCLONE_CBAR_ADDR)&PAGE_MASK; - offset = (CYCLONE_CBAR_ADDR)&(~PAGE_MASK); - set_fixmap_nocache(FIX_CYCLONE_TIMER, pageaddr); - reg = (u32*)(fix_to_virt(FIX_CYCLONE_TIMER) + offset); - if(!reg){ - printk(KERN_ERR "Summit chipset: Could not find valid CBAR register.\n"); - return -ENODEV; - } - base = *reg; - if(!base){ - printk(KERN_ERR "Summit chipset: Could not find valid CBAR value.\n"); - return -ENODEV; - } - - /* setup PMCC */ - pageaddr = (base + CYCLONE_PMCC_OFFSET)&PAGE_MASK; - offset = (base + CYCLONE_PMCC_OFFSET)&(~PAGE_MASK); - set_fixmap_nocache(FIX_CYCLONE_TIMER, pageaddr); - reg = (u32*)(fix_to_virt(FIX_CYCLONE_TIMER) + offset); - if(!reg){ - printk(KERN_ERR "Summit chipset: Could not find valid PMCC register.\n"); - return -ENODEV; - } - reg[0] = 0x00000001; - - /* setup MPCS */ - pageaddr = (base + CYCLONE_MPCS_OFFSET)&PAGE_MASK; - offset = (base + CYCLONE_MPCS_OFFSET)&(~PAGE_MASK); - set_fixmap_nocache(FIX_CYCLONE_TIMER, pageaddr); - reg = (u32*)(fix_to_virt(FIX_CYCLONE_TIMER) + offset); - if(!reg){ - printk(KERN_ERR "Summit chipset: Could not find valid MPCS register.\n"); - return -ENODEV; - } - reg[0] = 0x00000001; - - /* map in cyclone_timer */ - pageaddr = (base + CYCLONE_MPMC_OFFSET)&PAGE_MASK; - offset = (base + CYCLONE_MPMC_OFFSET)&(~PAGE_MASK); - set_fixmap_nocache(FIX_CYCLONE_TIMER, pageaddr); - cyclone_timer = (u32*)(fix_to_virt(FIX_CYCLONE_TIMER) + offset); - if(!cyclone_timer){ - printk(KERN_ERR "Summit chipset: Could not find valid MPMC register.\n"); - return -ENODEV; - } - - /*quick test to make sure its ticking*/ - for(i=0; i<3; i++){ - u32 old = cyclone_timer[0]; - int stall = 100; - while(stall--) barrier(); - if(cyclone_timer[0] == old){ - printk(KERN_ERR "Summit chipset: Counter not counting! DISABLED\n"); - cyclone_timer = 0; - return -ENODEV; - } - } - - init_cpu_khz(); - - /* Everything looks good! */ - return 0; -} - - -static void delay_cyclone(unsigned long loops) -{ - unsigned long bclock, now; - if(!cyclone_timer) - return; - bclock = cyclone_timer[0]; - do { - rep_nop(); - now = cyclone_timer[0]; - } while ((now-bclock) < loops); -} -/************************************************************/ - -/* cyclone timer_opts struct */ -static struct timer_opts timer_cyclone = { - .name = "cyclone", - .mark_offset = mark_offset_cyclone, - .get_offset = get_offset_cyclone, - .monotonic_clock = monotonic_clock_cyclone, - .delay = delay_cyclone, -}; - -struct init_timer_opts __initdata timer_cyclone_init = { - .init = init_cyclone, - .opts = &timer_cyclone, -}; diff --git a/arch/i386/kernel/timers/timer_hpet.c b/arch/i386/kernel/timers/timer_hpet.c deleted file mode 100644 index 17a6fe7166e7..000000000000 --- a/arch/i386/kernel/timers/timer_hpet.c +++ /dev/null @@ -1,217 +0,0 @@ -/* - * This code largely moved from arch/i386/kernel/time.c. - * See comments there for proper credits. - */ - -#include <linux/spinlock.h> -#include <linux/init.h> -#include <linux/timex.h> -#include <linux/errno.h> -#include <linux/string.h> -#include <linux/jiffies.h> - -#include <asm/timer.h> -#include <asm/io.h> -#include <asm/processor.h> - -#include "io_ports.h" -#include "mach_timer.h" -#include <asm/hpet.h> - -static unsigned long hpet_usec_quotient __read_mostly; /* convert hpet clks to usec */ -static unsigned long tsc_hpet_quotient __read_mostly; /* convert tsc to hpet clks */ -static unsigned long hpet_last; /* hpet counter value at last tick*/ -static unsigned long last_tsc_low; /* lsb 32 bits of Time Stamp Counter */ -static unsigned long last_tsc_high; /* msb 32 bits of Time Stamp Counter */ -static unsigned long long monotonic_base; -static seqlock_t monotonic_lock = SEQLOCK_UNLOCKED; - -/* convert from cycles(64bits) => nanoseconds (64bits) - * basic equation: - * ns = cycles / (freq / ns_per_sec) - * ns = cycles * (ns_per_sec / freq) - * ns = cycles * (10^9 / (cpu_khz * 10^3)) - * ns = cycles * (10^6 / cpu_khz) - * - * Then we use scaling math (suggested by george@mvista.com) to get: - * ns = cycles * (10^6 * SC / cpu_khz) / SC - * ns = cycles * cyc2ns_scale / SC - * - * And since SC is a constant power of two, we can convert the div - * into a shift. - * - * We can use khz divisor instead of mhz to keep a better percision, since - * cyc2ns_scale is limited to 10^6 * 2^10, which fits in 32 bits. - * (mathieu.desnoyers@polymtl.ca) - * - * -johnstul@us.ibm.com "math is hard, lets go shopping!" - */ -static unsigned long cyc2ns_scale __read_mostly; -#define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ - -static inline void set_cyc2ns_scale(unsigned long cpu_khz) -{ - cyc2ns_scale = (1000000 << CYC2NS_SCALE_FACTOR)/cpu_khz; -} - -static inline unsigned long long cycles_2_ns(unsigned long long cyc) -{ - return (cyc * cyc2ns_scale) >> CYC2NS_SCALE_FACTOR; -} - -static unsigned long long monotonic_clock_hpet(void) -{ - unsigned long long last_offset, this_offset, base; - unsigned seq; - - /* atomically read monotonic base & last_offset */ - do { - seq = read_seqbegin(&monotonic_lock); - last_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; - base = monotonic_base; - } while (read_seqretry(&monotonic_lock, seq)); - - /* Read the Time Stamp Counter */ - rdtscll(this_offset); - - /* return the value in ns */ - return base + cycles_2_ns(this_offset - last_offset); -} - -static unsigned long get_offset_hpet(void) -{ - register unsigned long eax, edx; - - eax = hpet_readl(HPET_COUNTER); - eax -= hpet_last; /* hpet delta */ - eax = min(hpet_tick, eax); - /* - * Time offset = (hpet delta) * ( usecs per HPET clock ) - * = (hpet delta) * ( usecs per tick / HPET clocks per tick) - * = (hpet delta) * ( hpet_usec_quotient ) / (2^32) - * - * Where, - * hpet_usec_quotient = (2^32 * usecs per tick)/HPET clocks per tick - * - * Using a mull instead of a divl saves some cycles in critical path. - */ - ASM_MUL64_REG(eax, edx, hpet_usec_quotient, eax); - - /* our adjusted time offset in microseconds */ - return edx; -} - -static void mark_offset_hpet(void) -{ - unsigned long long this_offset, last_offset; - unsigned long offset; - - write_seqlock(&monotonic_lock); - last_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; - rdtsc(last_tsc_low, last_tsc_high); - - if (hpet_use_timer) - offset = hpet_readl(HPET_T0_CMP) - hpet_tick; - else - offset = hpet_readl(HPET_COUNTER); - if (unlikely(((offset - hpet_last) >= (2*hpet_tick)) && (hpet_last != 0))) { - int lost_ticks = ((offset - hpet_last) / hpet_tick) - 1; - jiffies_64 += lost_ticks; - } - hpet_last = offset; - - /* update the monotonic base value */ - this_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; - monotonic_base += cycles_2_ns(this_offset - last_offset); - write_sequnlock(&monotonic_lock); -} - -static void delay_hpet(unsigned long loops) -{ - unsigned long hpet_start, hpet_end; - unsigned long eax; - - /* loops is the number of cpu cycles. Convert it to hpet clocks */ - ASM_MUL64_REG(eax, loops, tsc_hpet_quotient, loops); - - hpet_start = hpet_readl(HPET_COUNTER); - do { - rep_nop(); - hpet_end = hpet_readl(HPET_COUNTER); - } while ((hpet_end - hpet_start) < (loops)); -} - -static struct timer_opts timer_hpet; - -static int __init init_hpet(char* override) -{ - unsigned long result, remain; - - /* check clock override */ - if (override[0] && strncmp(override,"hpet",4)) - return -ENODEV; - - if (!is_hpet_enabled()) - return -ENODEV; - - printk("Using HPET for gettimeofday\n"); - if (cpu_has_tsc) { - unsigned long tsc_quotient = calibrate_tsc_hpet(&tsc_hpet_quotient); - if (tsc_quotient) { - /* report CPU clock rate in Hz. - * The formula is (10^6 * 2^32) / (2^32 * 1 / (clocks/us)) = - * clock/second. Our precision is about 100 ppm. - */ - { unsigned long eax=0, edx=1000; - ASM_DIV64_REG(cpu_khz, edx, tsc_quotient, - eax, edx); - printk("Detected %u.%03u MHz processor.\n", - cpu_khz / 1000, cpu_khz % 1000); - } - set_cyc2ns_scale(cpu_khz); - } - /* set this only when cpu_has_tsc */ - timer_hpet.read_timer = read_timer_tsc; - } - - /* - * Math to calculate hpet to usec multiplier - * Look for the comments at get_offset_hpet() - */ - ASM_DIV64_REG(result, remain, hpet_tick, 0, KERNEL_TICK_USEC); - if (remain > (hpet_tick >> 1)) - result++; /* rounding the result */ - hpet_usec_quotient = result; - - return 0; -} - -static int hpet_resume(void) -{ - write_seqlock(&monotonic_lock); - /* Assume this is the last mark offset time */ - rdtsc(last_tsc_low, last_tsc_high); - - if (hpet_use_timer) - hpet_last = hpet_readl(HPET_T0_CMP) - hpet_tick; - else - hpet_last = hpet_readl(HPET_COUNTER); - write_sequnlock(&monotonic_lock); - return 0; -} -/************************************************************/ - -/* tsc timer_opts struct */ -static struct timer_opts timer_hpet __read_mostly = { - .name = "hpet", - .mark_offset = mark_offset_hpet, - .get_offset = get_offset_hpet, - .monotonic_clock = monotonic_clock_hpet, - .delay = delay_hpet, - .resume = hpet_resume, -}; - -struct init_timer_opts __initdata timer_hpet_init = { - .init = init_hpet, - .opts = &timer_hpet, -}; diff --git a/arch/i386/kernel/timers/timer_none.c b/arch/i386/kernel/timers/timer_none.c deleted file mode 100644 index 4ea2f414dbbd..000000000000 --- a/arch/i386/kernel/timers/timer_none.c +++ /dev/null @@ -1,39 +0,0 @@ -#include <linux/init.h> -#include <asm/timer.h> - -static void mark_offset_none(void) -{ - /* nothing needed */ -} - -static unsigned long get_offset_none(void) -{ - return 0; -} - -static unsigned long long monotonic_clock_none(void) -{ - return 0; -} - -static void delay_none(unsigned long loops) -{ - int d0; - __asm__ __volatile__( - "\tjmp 1f\n" - ".align 16\n" - "1:\tjmp 2f\n" - ".align 16\n" - "2:\tdecl %0\n\tjns 2b" - :"=&a" (d0) - :"0" (loops)); -} - -/* none timer_opts struct */ -struct timer_opts timer_none = { - .name = "none", - .mark_offset = mark_offset_none, - .get_offset = get_offset_none, - .monotonic_clock = monotonic_clock_none, - .delay = delay_none, -}; diff --git a/arch/i386/kernel/timers/timer_pit.c b/arch/i386/kernel/timers/timer_pit.c deleted file mode 100644 index b9b6bd56b9ba..000000000000 --- a/arch/i386/kernel/timers/timer_pit.c +++ /dev/null @@ -1,177 +0,0 @@ -/* - * This code largely moved from arch/i386/kernel/time.c. - * See comments there for proper credits. - */ - -#include <linux/spinlock.h> -#include <linux/module.h> -#include <linux/device.h> -#include <linux/sysdev.h> -#include <linux/timex.h> -#include <asm/delay.h> -#include <asm/mpspec.h> -#include <asm/timer.h> -#include <asm/smp.h> -#include <asm/io.h> -#include <asm/arch_hooks.h> -#include <asm/i8253.h> - -#include "do_timer.h" -#include "io_ports.h" - -static int count_p; /* counter in get_offset_pit() */ - -static int __init init_pit(char* override) -{ - /* check clock override */ - if (override[0] && strncmp(override,"pit",3)) - printk(KERN_ERR "Warning: clock= override failed. Defaulting " - "to PIT\n"); - init_cpu_khz(); - count_p = LATCH; - return 0; -} - -static void mark_offset_pit(void) -{ - /* nothing needed */ -} - -static unsigned long long monotonic_clock_pit(void) -{ - return 0; -} - -static void delay_pit(unsigned long loops) -{ - int d0; - __asm__ __volatile__( - "\tjmp 1f\n" - ".align 16\n" - "1:\tjmp 2f\n" - ".align 16\n" - "2:\tdecl %0\n\tjns 2b" - :"=&a" (d0) - :"0" (loops)); -} - - -/* This function must be called with xtime_lock held. - * It was inspired by Steve McCanne's microtime-i386 for BSD. -- jrs - * - * However, the pc-audio speaker driver changes the divisor so that - * it gets interrupted rather more often - it loads 64 into the - * counter rather than 11932! This has an adverse impact on - * do_gettimeoffset() -- it stops working! What is also not - * good is that the interval that our timer function gets called - * is no longer 10.0002 ms, but 9.9767 ms. To get around this - * would require using a different timing source. Maybe someone - * could use the RTC - I know that this can interrupt at frequencies - * ranging from 8192Hz to 2Hz. If I had the energy, I'd somehow fix - * it so that at startup, the timer code in sched.c would select - * using either the RTC or the 8253 timer. The decision would be - * based on whether there was any other device around that needed - * to trample on the 8253. I'd set up the RTC to interrupt at 1024 Hz, - * and then do some jiggery to have a version of do_timer that - * advanced the clock by 1/1024 s. Every time that reached over 1/100 - * of a second, then do all the old code. If the time was kept correct - * then do_gettimeoffset could just return 0 - there is no low order - * divider that can be accessed. - * - * Ideally, you would be able to use the RTC for the speaker driver, - * but it appears that the speaker driver really needs interrupt more - * often than every 120 us or so. - * - * Anyway, this needs more thought.... pjsg (1993-08-28) - * - * If you are really that interested, you should be reading - * comp.protocols.time.ntp! - */ - -static unsigned long get_offset_pit(void) -{ - int count; - unsigned long flags; - static unsigned long jiffies_p = 0; - - /* - * cache volatile jiffies temporarily; we have xtime_lock. - */ - unsigned long jiffies_t; - - spin_lock_irqsave(&i8253_lock, flags); - /* timer count may underflow right here */ - outb_p(0x00, PIT_MODE); /* latch the count ASAP */ - - count = inb_p(PIT_CH0); /* read the latched count */ - - /* - * We do this guaranteed double memory access instead of a _p - * postfix in the previous port access. Wheee, hackady hack - */ - jiffies_t = jiffies; - - count |= inb_p(PIT_CH0) << 8; - - /* VIA686a test code... reset the latch if count > max + 1 */ - if (count > LATCH) { - outb_p(0x34, PIT_MODE); - outb_p(LATCH & 0xff, PIT_CH0); - outb(LATCH >> 8, PIT_CH0); - count = LATCH - 1; - } - - /* - * avoiding timer inconsistencies (they are rare, but they happen)... - * there are two kinds of problems that must be avoided here: - * 1. the timer counter underflows - * 2. hardware problem with the timer, not giving us continuous time, - * the counter does small "jumps" upwards on some Pentium systems, - * (see c't 95/10 page 335 for Neptun bug.) - */ - - if( jiffies_t == jiffies_p ) { - if( count > count_p ) { - /* the nutcase */ - count = do_timer_overflow(count); - } - } else - jiffies_p = jiffies_t; - - count_p = count; - - spin_unlock_irqrestore(&i8253_lock, flags); - - count = ((LATCH-1) - count) * TICK_SIZE; - count = (count + LATCH/2) / LATCH; - - return count; -} - - -/* tsc timer_opts struct */ -struct timer_opts timer_pit = { - .name = "pit", - .mark_offset = mark_offset_pit, - .get_offset = get_offset_pit, - .monotonic_clock = monotonic_clock_pit, - .delay = delay_pit, -}; - -struct init_timer_opts __initdata timer_pit_init = { - .init = init_pit, - .opts = &timer_pit, -}; - -void setup_pit_timer(void) -{ - unsigned long flags; - - spin_lock_irqsave(&i8253_lock, flags); - outb_p(0x34,PIT_MODE); /* binary, mode 2, LSB/MSB, ch 0 */ - udelay(10); - outb_p(LATCH & 0xff , PIT_CH0); /* LSB */ - udelay(10); - outb(LATCH >> 8 , PIT_CH0); /* MSB */ - spin_unlock_irqrestore(&i8253_lock, flags); -} diff --git a/arch/i386/kernel/timers/timer_pm.c b/arch/i386/kernel/timers/timer_pm.c deleted file mode 100644 index 144e94a04933..000000000000 --- a/arch/i386/kernel/timers/timer_pm.c +++ /dev/null @@ -1,342 +0,0 @@ -/* - * (C) Dominik Brodowski <linux@brodo.de> 2003 - * - * Driver to use the Power Management Timer (PMTMR) available in some - * southbridges as primary timing source for the Linux kernel. - * - * Based on parts of linux/drivers/acpi/hardware/hwtimer.c, timer_pit.c, - * timer_hpet.c, and on Arjan van de Ven's implementation for 2.4. - * - * This file is licensed under the GPL v2. - */ - - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/device.h> -#include <linux/init.h> -#include <linux/pci.h> -#include <asm/types.h> -#include <asm/timer.h> -#include <asm/smp.h> -#include <asm/io.h> -#include <asm/arch_hooks.h> - -#include <linux/timex.h> -#include "mach_timer.h" - -/* Number of PMTMR ticks expected during calibration run */ -#define PMTMR_TICKS_PER_SEC 3579545 -#define PMTMR_EXPECTED_RATE \ - ((CALIBRATE_LATCH * (PMTMR_TICKS_PER_SEC >> 10)) / (CLOCK_TICK_RATE>>10)) - - -/* The I/O port the PMTMR resides at. - * The location is detected during setup_arch(), - * in arch/i386/acpi/boot.c */ -u32 pmtmr_ioport = 0; - - -/* value of the Power timer at last timer interrupt */ -static u32 offset_tick; -static u32 offset_delay; - -static unsigned long long monotonic_base; -static seqlock_t monotonic_lock = SEQLOCK_UNLOCKED; - -#define ACPI_PM_MASK 0xFFFFFF /* limit it to 24 bits */ - -static int pmtmr_need_workaround __read_mostly = 1; - -/*helper function to safely read acpi pm timesource*/ -static inline u32 read_pmtmr(void) -{ - if (pmtmr_need_workaround) { - u32 v1, v2, v3; - - /* It has been reported that because of various broken - * chipsets (ICH4, PIIX4 and PIIX4E) where the ACPI PM time - * source is not latched, so you must read it multiple - * times to insure a safe value is read. - */ - do { - v1 = inl(pmtmr_ioport); - v2 = inl(pmtmr_ioport); - v3 = inl(pmtmr_ioport); - } while ((v1 > v2 && v1 < v3) || (v2 > v3 && v2 < v1) - || (v3 > v1 && v3 < v2)); - - /* mask the output to 24 bits */ - return v2 & ACPI_PM_MASK; - } - - return inl(pmtmr_ioport) & ACPI_PM_MASK; -} - - -/* - * Some boards have the PMTMR running way too fast. We check - * the PMTMR rate against PIT channel 2 to catch these cases. - */ -static int verify_pmtmr_rate(void) -{ - u32 value1, value2; - unsigned long count, delta; - - mach_prepare_counter(); - value1 = read_pmtmr(); - mach_countup(&count); - value2 = read_pmtmr(); - delta = (value2 - value1) & ACPI_PM_MASK; - - /* Check that the PMTMR delta is within 5% of what we expect */ - if (delta < (PMTMR_EXPECTED_RATE * 19) / 20 || - delta > (PMTMR_EXPECTED_RATE * 21) / 20) { - printk(KERN_INFO "PM-Timer running at invalid rate: %lu%% of normal - aborting.\n", 100UL * delta / PMTMR_EXPECTED_RATE); - return -1; - } - - return 0; -} - - -static int init_pmtmr(char* override) -{ - u32 value1, value2; - unsigned int i; - - if (override[0] && strncmp(override,"pmtmr",5)) - return -ENODEV; - - if (!pmtmr_ioport) - return -ENODEV; - - /* we use the TSC for delay_pmtmr, so make sure it exists */ - if (!cpu_has_tsc) - return -ENODEV; - - /* "verify" this timing source */ - value1 = read_pmtmr(); - for (i = 0; i < 10000; i++) { - value2 = read_pmtmr(); - if (value2 == value1) - continue; - if (value2 > value1) - goto pm_good; - if ((value2 < value1) && ((value2) < 0xFFF)) - goto pm_good; - printk(KERN_INFO "PM-Timer had inconsistent results: 0x%#x, 0x%#x - aborting.\n", value1, value2); - return -EINVAL; - } - printk(KERN_INFO "PM-Timer had no reasonable result: 0x%#x - aborting.\n", value1); - return -ENODEV; - -pm_good: - if (verify_pmtmr_rate() != 0) - return -ENODEV; - - init_cpu_khz(); - return 0; -} - -static inline u32 cyc2us(u32 cycles) -{ - /* The Power Management Timer ticks at 3.579545 ticks per microsecond. - * 1 / PM_TIMER_FREQUENCY == 0.27936511 =~ 286/1024 [error: 0.024%] - * - * Even with HZ = 100, delta is at maximum 35796 ticks, so it can - * easily be multiplied with 286 (=0x11E) without having to fear - * u32 overflows. - */ - cycles *= 286; - return (cycles >> 10); -} - -/* - * this gets called during each timer interrupt - * - Called while holding the writer xtime_lock - */ -static void mark_offset_pmtmr(void) -{ - u32 lost, delta, last_offset; - static int first_run = 1; - last_offset = offset_tick; - - write_seqlock(&monotonic_lock); - - offset_tick = read_pmtmr(); - - /* calculate tick interval */ - delta = (offset_tick - last_offset) & ACPI_PM_MASK; - - /* convert to usecs */ - delta = cyc2us(delta); - - /* update the monotonic base value */ - monotonic_base += delta * NSEC_PER_USEC; - write_sequnlock(&monotonic_lock); - - /* convert to ticks */ - delta += offset_delay; - lost = delta / (USEC_PER_SEC / HZ); - offset_delay = delta % (USEC_PER_SEC / HZ); - - - /* compensate for lost ticks */ - if (lost >= 2) - jiffies_64 += lost - 1; - - /* don't calculate delay for first run, - or if we've got less then a tick */ - if (first_run || (lost < 1)) { - first_run = 0; - offset_delay = 0; - } -} - -static int pmtmr_resume(void) -{ - write_seqlock(&monotonic_lock); - /* Assume this is the last mark offset time */ - offset_tick = read_pmtmr(); - write_sequnlock(&monotonic_lock); - return 0; -} - -static unsigned long long monotonic_clock_pmtmr(void) -{ - u32 last_offset, this_offset; - unsigned long long base, ret; - unsigned seq; - - - /* atomically read monotonic base & last_offset */ - do { - seq = read_seqbegin(&monotonic_lock); - last_offset = offset_tick; - base = monotonic_base; - } while (read_seqretry(&monotonic_lock, seq)); - - /* Read the pmtmr */ - this_offset = read_pmtmr(); - - /* convert to nanoseconds */ - ret = (this_offset - last_offset) & ACPI_PM_MASK; - ret = base + (cyc2us(ret) * NSEC_PER_USEC); - return ret; -} - -static void delay_pmtmr(unsigned long loops) -{ - unsigned long bclock, now; - - rdtscl(bclock); - do - { - rep_nop(); - rdtscl(now); - } while ((now-bclock) < loops); -} - - -/* - * get the offset (in microseconds) from the last call to mark_offset() - * - Called holding a reader xtime_lock - */ -static unsigned long get_offset_pmtmr(void) -{ - u32 now, offset, delta = 0; - - offset = offset_tick; - now = read_pmtmr(); - delta = (now - offset)&ACPI_PM_MASK; - - return (unsigned long) offset_delay + cyc2us(delta); -} - - -/* acpi timer_opts struct */ -static struct timer_opts timer_pmtmr = { - .name = "pmtmr", - .mark_offset = mark_offset_pmtmr, - .get_offset = get_offset_pmtmr, - .monotonic_clock = monotonic_clock_pmtmr, - .delay = delay_pmtmr, - .read_timer = read_timer_tsc, - .resume = pmtmr_resume, -}; - -struct init_timer_opts __initdata timer_pmtmr_init = { - .init = init_pmtmr, - .opts = &timer_pmtmr, -}; - -#ifdef CONFIG_PCI -/* - * PIIX4 Errata: - * - * The power management timer may return improper results when read. - * Although the timer value settles properly after incrementing, - * while incrementing there is a 3 ns window every 69.8 ns where the - * timer value is indeterminate (a 4.2% chance that the data will be - * incorrect when read). As a result, the ACPI free running count up - * timer specification is violated due to erroneous reads. - */ -static int __init pmtmr_bug_check(void) -{ - static struct pci_device_id gray_list[] __initdata = { - /* these chipsets may have bug. */ - { PCI_DEVICE(PCI_VENDOR_ID_INTEL, - PCI_DEVICE_ID_INTEL_82801DB_0) }, - { }, - }; - struct pci_dev *dev; - int pmtmr_has_bug = 0; - u8 rev; - - if (cur_timer != &timer_pmtmr || !pmtmr_need_workaround) - return 0; - - dev = pci_get_device(PCI_VENDOR_ID_INTEL, - PCI_DEVICE_ID_INTEL_82371AB_3, NULL); - if (dev) { - pci_read_config_byte(dev, PCI_REVISION_ID, &rev); - /* the bug has been fixed in PIIX4M */ - if (rev < 3) { - printk(KERN_WARNING "* Found PM-Timer Bug on this " - "chipset. Due to workarounds for a bug,\n" - "* this time source is slow. Consider trying " - "other time sources (clock=)\n"); - pmtmr_has_bug = 1; - } - pci_dev_put(dev); - } - - if (pci_dev_present(gray_list)) { - printk(KERN_WARNING "* This chipset may have PM-Timer Bug. Due" - " to workarounds for a bug,\n" - "* this time source is slow. If you are sure your timer" - " does not have\n" - "* this bug, please use \"pmtmr_good\" to disable the " - "workaround\n"); - pmtmr_has_bug = 1; - } - - if (!pmtmr_has_bug) - pmtmr_need_workaround = 0; - - return 0; -} -device_initcall(pmtmr_bug_check); -#endif - -static int __init pmtr_good_setup(char *__str) -{ - pmtmr_need_workaround = 0; - return 1; -} -__setup("pmtmr_good", pmtr_good_setup); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Dominik Brodowski <linux@brodo.de>"); -MODULE_DESCRIPTION("Power Management Timer (PMTMR) as primary timing source for x86"); diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c deleted file mode 100644 index f1187ddb0d0f..000000000000 --- a/arch/i386/kernel/timers/timer_tsc.c +++ /dev/null @@ -1,617 +0,0 @@ -/* - * This code largely moved from arch/i386/kernel/time.c. - * See comments there for proper credits. - * - * 2004-06-25 Jesper Juhl - * moved mark_offset_tsc below cpufreq_delayed_get to avoid gcc 3.4 - * failing to inline. - */ - -#include <linux/spinlock.h> -#include <linux/init.h> -#include <linux/timex.h> -#include <linux/errno.h> -#include <linux/cpufreq.h> -#include <linux/string.h> -#include <linux/jiffies.h> - -#include <asm/timer.h> -#include <asm/io.h> -/* processor.h for distable_tsc flag */ -#include <asm/processor.h> - -#include "io_ports.h" -#include "mach_timer.h" - -#include <asm/hpet.h> -#include <asm/i8253.h> - -#ifdef CONFIG_HPET_TIMER -static unsigned long hpet_usec_quotient; -static unsigned long hpet_last; -static struct timer_opts timer_tsc; -#endif - -static inline void cpufreq_delayed_get(void); - -int tsc_disable __devinitdata = 0; - -static int use_tsc; -/* Number of usecs that the last interrupt was delayed */ -static int delay_at_last_interrupt; - -static unsigned long last_tsc_low; /* lsb 32 bits of Time Stamp Counter */ -static unsigned long last_tsc_high; /* msb 32 bits of Time Stamp Counter */ -static unsigned long long monotonic_base; -static seqlock_t monotonic_lock = SEQLOCK_UNLOCKED; - -/* Avoid compensating for lost ticks before TSCs are synched */ -static int detect_lost_ticks; -static int __init start_lost_tick_compensation(void) -{ - detect_lost_ticks = 1; - return 0; -} -late_initcall(start_lost_tick_compensation); - -/* convert from cycles(64bits) => nanoseconds (64bits) - * basic equation: - * ns = cycles / (freq / ns_per_sec) - * ns = cycles * (ns_per_sec / freq) - * ns = cycles * (10^9 / (cpu_khz * 10^3)) - * ns = cycles * (10^6 / cpu_khz) - * - * Then we use scaling math (suggested by george@mvista.com) to get: - * ns = cycles * (10^6 * SC / cpu_khz) / SC - * ns = cycles * cyc2ns_scale / SC - * - * And since SC is a constant power of two, we can convert the div - * into a shift. - * - * We can use khz divisor instead of mhz to keep a better percision, since - * cyc2ns_scale is limited to 10^6 * 2^10, which fits in 32 bits. - * (mathieu.desnoyers@polymtl.ca) - * - * -johnstul@us.ibm.com "math is hard, lets go shopping!" - */ -static unsigned long cyc2ns_scale __read_mostly; -#define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ - -static inline void set_cyc2ns_scale(unsigned long cpu_khz) -{ - cyc2ns_scale = (1000000 << CYC2NS_SCALE_FACTOR)/cpu_khz; -} - -static inline unsigned long long cycles_2_ns(unsigned long long cyc) -{ - return (cyc * cyc2ns_scale) >> CYC2NS_SCALE_FACTOR; -} - -static int count2; /* counter for mark_offset_tsc() */ - -/* Cached *multiplier* to convert TSC counts to microseconds. - * (see the equation below). - * Equal to 2^32 * (1 / (clocks per usec) ). - * Initialized in time_init. - */ -static unsigned long fast_gettimeoffset_quotient; - -static unsigned long get_offset_tsc(void) -{ - register unsigned long eax, edx; - - /* Read the Time Stamp Counter */ - - rdtsc(eax,edx); - - /* .. relative to previous jiffy (32 bits is enough) */ - eax -= last_tsc_low; /* tsc_low delta */ - - /* - * Time offset = (tsc_low delta) * fast_gettimeoffset_quotient - * = (tsc_low delta) * (usecs_per_clock) - * = (tsc_low delta) * (usecs_per_jiffy / clocks_per_jiffy) - * - * Using a mull instead of a divl saves up to 31 clock cycles - * in the critical path. - */ - - __asm__("mull %2" - :"=a" (eax), "=d" (edx) - :"rm" (fast_gettimeoffset_quotient), - "0" (eax)); - - /* our adjusted time offset in microseconds */ - return delay_at_last_interrupt + edx; -} - -static unsigned long long monotonic_clock_tsc(void) -{ - unsigned long long last_offset, this_offset, base; - unsigned seq; - - /* atomically read monotonic base & last_offset */ - do { - seq = read_seqbegin(&monotonic_lock); - last_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; - base = monotonic_base; - } while (read_seqretry(&monotonic_lock, seq)); - - /* Read the Time Stamp Counter */ - rdtscll(this_offset); - - /* return the value in ns */ - return base + cycles_2_ns(this_offset - last_offset); -} - -/* - * Scheduler clock - returns current time in nanosec units. - */ -unsigned long long sched_clock(void) -{ - unsigned long long this_offset; - - /* - * In the NUMA case we dont use the TSC as they are not - * synchronized across all CPUs. - */ -#ifndef CONFIG_NUMA - if (!use_tsc) -#endif - /* no locking but a rare wrong value is not a big deal */ - return jiffies_64 * (1000000000 / HZ); - - /* Read the Time Stamp Counter */ - rdtscll(this_offset); - - /* return the value in ns */ - return cycles_2_ns(this_offset); -} - -static void delay_tsc(unsigned long loops) -{ - unsigned long bclock, now; - - rdtscl(bclock); - do - { - rep_nop(); - rdtscl(now); - } while ((now-bclock) < loops); -} - -#ifdef CONFIG_HPET_TIMER -static void mark_offset_tsc_hpet(void) -{ - unsigned long long this_offset, last_offset; - unsigned long offset, temp, hpet_current; - - write_seqlock(&monotonic_lock); - last_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; - /* - * It is important that these two operations happen almost at - * the same time. We do the RDTSC stuff first, since it's - * faster. To avoid any inconsistencies, we need interrupts - * disabled locally. - */ - /* - * Interrupts are just disabled locally since the timer irq - * has the SA_INTERRUPT flag set. -arca - */ - /* read Pentium cycle counter */ - - hpet_current = hpet_readl(HPET_COUNTER); - rdtsc(last_tsc_low, last_tsc_high); - - /* lost tick compensation */ - offset = hpet_readl(HPET_T0_CMP) - hpet_tick; - if (unlikely(((offset - hpet_last) > hpet_tick) && (hpet_last != 0)) - && detect_lost_ticks) { - int lost_ticks = (offset - hpet_last) / hpet_tick; - jiffies_64 += lost_ticks; - } - hpet_last = hpet_current; - - /* update the monotonic base value */ - this_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; - monotonic_base += cycles_2_ns(this_offset - last_offset); - write_sequnlock(&monotonic_lock); - - /* calculate delay_at_last_interrupt */ - /* - * Time offset = (hpet delta) * ( usecs per HPET clock ) - * = (hpet delta) * ( usecs per tick / HPET clocks per tick) - * = (hpet delta) * ( hpet_usec_quotient ) / (2^32) - * Where, - * hpet_usec_quotient = (2^32 * usecs per tick)/HPET clocks per tick - */ - delay_at_last_interrupt = hpet_current - offset; - ASM_MUL64_REG(temp, delay_at_last_interrupt, - hpet_usec_quotient, delay_at_last_interrupt); -} -#endif - - -#ifdef CONFIG_CPU_FREQ -#include <linux/workqueue.h> - -static unsigned int cpufreq_delayed_issched = 0; -static unsigned int cpufreq_init = 0; -static struct work_struct cpufreq_delayed_get_work; - -static void handle_cpufreq_delayed_get(void *v) -{ - unsigned int cpu; - for_each_online_cpu(cpu) { - cpufreq_get(cpu); - } - cpufreq_delayed_issched = 0; -} - -/* if we notice lost ticks, schedule a call to cpufreq_get() as it tries - * to verify the CPU frequency the timing core thinks the CPU is running - * at is still correct. - */ -static inline void cpufreq_delayed_get(void) -{ - if (cpufreq_init && !cpufreq_delayed_issched) { - cpufreq_delayed_issched = 1; - printk(KERN_DEBUG "Losing some ticks... checking if CPU frequency changed.\n"); - schedule_work(&cpufreq_delayed_get_work); - } -} - -/* If the CPU frequency is scaled, TSC-based delays will need a different - * loops_per_jiffy value to function properly. - */ - -static unsigned int ref_freq = 0; -static unsigned long loops_per_jiffy_ref = 0; - -#ifndef CONFIG_SMP -static unsigned long fast_gettimeoffset_ref = 0; -static unsigned int cpu_khz_ref = 0; -#endif - -static int -time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, - void *data) -{ - struct cpufreq_freqs *freq = data; - - if (val != CPUFREQ_RESUMECHANGE && val != CPUFREQ_SUSPENDCHANGE) - write_seqlock_irq(&xtime_lock); - if (!ref_freq) { - if (!freq->old){ - ref_freq = freq->new; - goto end; - } - ref_freq = freq->old; - loops_per_jiffy_ref = cpu_data[freq->cpu].loops_per_jiffy; -#ifndef CONFIG_SMP - fast_gettimeoffset_ref = fast_gettimeoffset_quotient; - cpu_khz_ref = cpu_khz; -#endif - } - - if ((val == CPUFREQ_PRECHANGE && freq->old < freq->new) || - (val == CPUFREQ_POSTCHANGE && freq->old > freq->new) || - (val == CPUFREQ_RESUMECHANGE)) { - if (!(freq->flags & CPUFREQ_CONST_LOOPS)) - cpu_data[freq->cpu].loops_per_jiffy = cpufreq_scale(loops_per_jiffy_ref, ref_freq, freq->new); -#ifndef CONFIG_SMP - if (cpu_khz) - cpu_khz = cpufreq_scale(cpu_khz_ref, ref_freq, freq->new); - if (use_tsc) { - if (!(freq->flags & CPUFREQ_CONST_LOOPS)) { - fast_gettimeoffset_quotient = cpufreq_scale(fast_gettimeoffset_ref, freq->new, ref_freq); - set_cyc2ns_scale(cpu_khz); - } - } -#endif - } - -end: - if (val != CPUFREQ_RESUMECHANGE && val != CPUFREQ_SUSPENDCHANGE) - write_sequnlock_irq(&xtime_lock); - - return 0; -} - -static struct notifier_block time_cpufreq_notifier_block = { - .notifier_call = time_cpufreq_notifier -}; - - -static int __init cpufreq_tsc(void) -{ - int ret; - INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get, NULL); - ret = cpufreq_register_notifier(&time_cpufreq_notifier_block, - CPUFREQ_TRANSITION_NOTIFIER); - if (!ret) - cpufreq_init = 1; - return ret; -} -core_initcall(cpufreq_tsc); - -#else /* CONFIG_CPU_FREQ */ -static inline void cpufreq_delayed_get(void) { return; } -#endif - -int recalibrate_cpu_khz(void) -{ -#ifndef CONFIG_SMP - unsigned int cpu_khz_old = cpu_khz; - - if (cpu_has_tsc) { - local_irq_disable(); - init_cpu_khz(); - local_irq_enable(); - cpu_data[0].loops_per_jiffy = - cpufreq_scale(cpu_data[0].loops_per_jiffy, - cpu_khz_old, - cpu_khz); - return 0; - } else - return -ENODEV; -#else - return -ENODEV; -#endif -} -EXPORT_SYMBOL(recalibrate_cpu_khz); - -static void mark_offset_tsc(void) -{ - unsigned long lost,delay; - unsigned long delta = last_tsc_low; - int count; - int countmp; - static int count1 = 0; - unsigned long long this_offset, last_offset; - static int lost_count = 0; - - write_seqlock(&monotonic_lock); - last_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; - /* - * It is important that these two operations happen almost at - * the same time. We do the RDTSC stuff first, since it's - * faster. To avoid any inconsistencies, we need interrupts - * disabled locally. - */ - - /* - * Interrupts are just disabled locally since the timer irq - * has the SA_INTERRUPT flag set. -arca - */ - - /* read Pentium cycle counter */ - - rdtsc(last_tsc_low, last_tsc_high); - - spin_lock(&i8253_lock); - outb_p(0x00, PIT_MODE); /* latch the count ASAP */ - - count = inb_p(PIT_CH0); /* read the latched count */ - count |= inb(PIT_CH0) << 8; - - /* - * VIA686a test code... reset the latch if count > max + 1 - * from timer_pit.c - cjb - */ - if (count > LATCH) { - outb_p(0x34, PIT_MODE); - outb_p(LATCH & 0xff, PIT_CH0); - outb(LATCH >> 8, PIT_CH0); - count = LATCH - 1; - } - - spin_unlock(&i8253_lock); - - if (pit_latch_buggy) { - /* get center value of last 3 time lutch */ - if ((count2 >= count && count >= count1) - || (count1 >= count && count >= count2)) { - count2 = count1; count1 = count; - } else if ((count1 >= count2 && count2 >= count) - || (count >= count2 && count2 >= count1)) { - countmp = count;count = count2; - count2 = count1;count1 = countmp; - } else { - count2 = count1; count1 = count; count = count1; - } - } - - /* lost tick compensation */ - delta = last_tsc_low - delta; - { - register unsigned long eax, edx; - eax = delta; - __asm__("mull %2" - :"=a" (eax), "=d" (edx) - :"rm" (fast_gettimeoffset_quotient), - "0" (eax)); - delta = edx; - } - delta += delay_at_last_interrupt; - lost = delta/(1000000/HZ); - delay = delta%(1000000/HZ); - if (lost >= 2 && detect_lost_ticks) { - jiffies_64 += lost-1; - - /* sanity check to ensure we're not always losing ticks */ - if (lost_count++ > 100) { - printk(KERN_WARNING "Losing too many ticks!\n"); - printk(KERN_WARNING "TSC cannot be used as a timesource. \n"); - printk(KERN_WARNING "Possible reasons for this are:\n"); - printk(KERN_WARNING " You're running with Speedstep,\n"); - printk(KERN_WARNING " You don't have DMA enabled for your hard disk (see hdparm),\n"); - printk(KERN_WARNING " Incorrect TSC synchronization on an SMP system (see dmesg).\n"); - printk(KERN_WARNING "Falling back to a sane timesource now.\n"); - - clock_fallback(); - } - /* ... but give the TSC a fair chance */ - if (lost_count > 25) - cpufreq_delayed_get(); - } else - lost_count = 0; - /* update the monotonic base value */ - this_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; - monotonic_base += cycles_2_ns(this_offset - last_offset); - write_sequnlock(&monotonic_lock); - - /* calculate delay_at_last_interrupt */ - count = ((LATCH-1) - count) * TICK_SIZE; - delay_at_last_interrupt = (count + LATCH/2) / LATCH; - - /* catch corner case where tick rollover occured - * between tsc and pit reads (as noted when - * usec delta is > 90% # of usecs/tick) - */ - if (lost && abs(delay - delay_at_last_interrupt) > (900000/HZ)) - jiffies_64++; -} - -static int __init init_tsc(char* override) -{ - - /* check clock override */ - if (override[0] && strncmp(override,"tsc",3)) { -#ifdef CONFIG_HPET_TIMER - if (is_hpet_enabled()) { - printk(KERN_ERR "Warning: clock= override failed. Defaulting to tsc\n"); - } else -#endif - { - return -ENODEV; - } - } - - /* - * If we have APM enabled or the CPU clock speed is variable - * (CPU stops clock on HLT or slows clock to save power) - * then the TSC timestamps may diverge by up to 1 jiffy from - * 'real time' but nothing will break. - * The most frequent case is that the CPU is "woken" from a halt - * state by the timer interrupt itself, so we get 0 error. In the - * rare cases where a driver would "wake" the CPU and request a - * timestamp, the maximum error is < 1 jiffy. But timestamps are - * still perfectly ordered. - * Note that the TSC counter will be reset if APM suspends - * to disk; this won't break the kernel, though, 'cuz we're - * smart. See arch/i386/kernel/apm.c. - */ - /* - * Firstly we have to do a CPU check for chips with - * a potentially buggy TSC. At this point we haven't run - * the ident/bugs checks so we must run this hook as it - * may turn off the TSC flag. - * - * NOTE: this doesn't yet handle SMP 486 machines where only - * some CPU's have a TSC. Thats never worked and nobody has - * moaned if you have the only one in the world - you fix it! - */ - - count2 = LATCH; /* initialize counter for mark_offset_tsc() */ - - if (cpu_has_tsc) { - unsigned long tsc_quotient; -#ifdef CONFIG_HPET_TIMER - if (is_hpet_enabled() && hpet_use_timer) { - unsigned long result, remain; - printk("Using TSC for gettimeofday\n"); - tsc_quotient = calibrate_tsc_hpet(NULL); - timer_tsc.mark_offset = &mark_offset_tsc_hpet; - /* - * Math to calculate hpet to usec multiplier - * Look for the comments at get_offset_tsc_hpet() - */ - ASM_DIV64_REG(result, remain, hpet_tick, - 0, KERNEL_TICK_USEC); - if (remain > (hpet_tick >> 1)) - result++; /* rounding the result */ - - hpet_usec_quotient = result; - } else -#endif - { - tsc_quotient = calibrate_tsc(); - } - - if (tsc_quotient) { - fast_gettimeoffset_quotient = tsc_quotient; - use_tsc = 1; - /* - * We could be more selective here I suspect - * and just enable this for the next intel chips ? - */ - /* report CPU clock rate in Hz. - * The formula is (10^6 * 2^32) / (2^32 * 1 / (clocks/us)) = - * clock/second. Our precision is about 100 ppm. - */ - { unsigned long eax=0, edx=1000; - __asm__("divl %2" - :"=a" (cpu_khz), "=d" (edx) - :"r" (tsc_quotient), - "0" (eax), "1" (edx)); - printk("Detected %u.%03u MHz processor.\n", - cpu_khz / 1000, cpu_khz % 1000); - } - set_cyc2ns_scale(cpu_khz); - return 0; - } - } - return -ENODEV; -} - -static int tsc_resume(void) -{ - write_seqlock(&monotonic_lock); - /* Assume this is the last mark offset time */ - rdtsc(last_tsc_low, last_tsc_high); -#ifdef CONFIG_HPET_TIMER - if (is_hpet_enabled() && hpet_use_timer) - hpet_last = hpet_readl(HPET_COUNTER); -#endif - write_sequnlock(&monotonic_lock); - return 0; -} - -#ifndef CONFIG_X86_TSC -/* disable flag for tsc. Takes effect by clearing the TSC cpu flag - * in cpu/common.c */ -static int __init tsc_setup(char *str) -{ - tsc_disable = 1; - return 1; -} -#else -static int __init tsc_setup(char *str) -{ - printk(KERN_WARNING "notsc: Kernel compiled with CONFIG_X86_TSC, " - "cannot disable TSC.\n"); - return 1; -} -#endif -__setup("notsc", tsc_setup); - - - -/************************************************************/ - -/* tsc timer_opts struct */ -static struct timer_opts timer_tsc = { - .name = "tsc", - .mark_offset = mark_offset_tsc, - .get_offset = get_offset_tsc, - .monotonic_clock = monotonic_clock_tsc, - .delay = delay_tsc, - .read_timer = read_timer_tsc, - .resume = tsc_resume, -}; - -struct init_timer_opts __initdata timer_tsc_init = { - .init = init_tsc, - .opts = &timer_tsc, -}; diff --git a/arch/i386/kernel/topology.c b/arch/i386/kernel/topology.c index 296355292c7c..e2e281d4bcc8 100644 --- a/arch/i386/kernel/topology.c +++ b/arch/i386/kernel/topology.c @@ -32,15 +32,8 @@ static struct i386_cpu cpu_devices[NR_CPUS]; -int arch_register_cpu(int num){ - struct node *parent = NULL; - -#ifdef CONFIG_NUMA - int node = cpu_to_node(num); - if (node_online(node)) - parent = &node_devices[node].node; -#endif /* CONFIG_NUMA */ - +int arch_register_cpu(int num) +{ /* * CPU0 cannot be offlined due to several * restrictions and assumptions in kernel. This basically @@ -50,21 +43,13 @@ int arch_register_cpu(int num){ if (!num) cpu_devices[num].cpu.no_control = 1; - return register_cpu(&cpu_devices[num].cpu, num, parent); + return register_cpu(&cpu_devices[num].cpu, num); } #ifdef CONFIG_HOTPLUG_CPU void arch_unregister_cpu(int num) { - struct node *parent = NULL; - -#ifdef CONFIG_NUMA - int node = cpu_to_node(num); - if (node_online(node)) - parent = &node_devices[node].node; -#endif /* CONFIG_NUMA */ - - return unregister_cpu(&cpu_devices[num].cpu, parent); + return unregister_cpu(&cpu_devices[num].cpu); } EXPORT_SYMBOL(arch_register_cpu); EXPORT_SYMBOL(arch_unregister_cpu); @@ -74,16 +59,13 @@ EXPORT_SYMBOL(arch_unregister_cpu); #ifdef CONFIG_NUMA #include <linux/mmzone.h> -#include <asm/node.h> - -struct i386_node node_devices[MAX_NUMNODES]; static int __init topology_init(void) { int i; for_each_online_node(i) - arch_register_node(i); + register_one_node(i); for_each_present_cpu(i) arch_register_cpu(i); diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 0e498369f35e..2bf8b55b91f8 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -11,7 +11,6 @@ * 'Traps.c' handles hardware traps and faults after we have saved some * state in 'asm.s'. */ -#include <linux/config.h> #include <linux/sched.h> #include <linux/kernel.h> #include <linux/string.h> @@ -28,6 +27,7 @@ #include <linux/utsname.h> #include <linux/kprobes.h> #include <linux/kexec.h> +#include <linux/unwind.h> #ifdef CONFIG_EISA #include <linux/ioport.h> @@ -47,7 +47,7 @@ #include <asm/desc.h> #include <asm/i387.h> #include <asm/nmi.h> - +#include <asm/unwind.h> #include <asm/smp.h> #include <asm/arch_hooks.h> #include <asm/kdebug.h> @@ -92,6 +92,7 @@ asmlinkage void spurious_interrupt_bug(void); asmlinkage void machine_check(void); static int kstack_depth_to_print = 24; +static int call_trace = 1; ATOMIC_NOTIFIER_HEAD(i386die_chain); int register_die_notifier(struct notifier_block *nb) @@ -114,28 +115,13 @@ static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) } /* - * Print CONFIG_STACK_BACKTRACE_COLS address/symbol entries per line. + * Print one address/symbol entries per line. */ -static inline int print_addr_and_symbol(unsigned long addr, char *log_lvl, - int printed) +static inline void print_addr_and_symbol(unsigned long addr, char *log_lvl) { - if (!printed) - printk(log_lvl); - -#if CONFIG_STACK_BACKTRACE_COLS == 1 printk(" [<%08lx>] ", addr); -#else - printk(" <%08lx> ", addr); -#endif - print_symbol("%s", addr); - - printed = (printed + 1) % CONFIG_STACK_BACKTRACE_COLS; - if (printed) - printk(" "); - else - printk("\n"); - return printed; + print_symbol("%s\n", addr); } static inline unsigned long print_context_stack(struct thread_info *tinfo, @@ -143,28 +129,44 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo, char *log_lvl) { unsigned long addr; - int printed = 0; /* nr of entries already printed on current line */ #ifdef CONFIG_FRAME_POINTER while (valid_stack_ptr(tinfo, (void *)ebp)) { addr = *(unsigned long *)(ebp + 4); - printed = print_addr_and_symbol(addr, log_lvl, printed); + print_addr_and_symbol(addr, log_lvl); + /* + * break out of recursive entries (such as + * end_of_stack_stop_unwind_function): + */ + if (ebp == *(unsigned long *)ebp) + break; ebp = *(unsigned long *)ebp; } #else while (valid_stack_ptr(tinfo, stack)) { addr = *stack++; if (__kernel_text_address(addr)) - printed = print_addr_and_symbol(addr, log_lvl, printed); + print_addr_and_symbol(addr, log_lvl); } #endif - if (printed) - printk("\n"); - return ebp; } -static void show_trace_log_lvl(struct task_struct *task, +static asmlinkage int +show_trace_unwind(struct unwind_frame_info *info, void *log_lvl) +{ + int n = 0; + + while (unwind(info) == 0 && UNW_PC(info)) { + n++; + print_addr_and_symbol(UNW_PC(info), log_lvl); + if (arch_unw_user_mode(info)) + break; + } + return n; +} + +static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, char *log_lvl) { unsigned long ebp; @@ -172,6 +174,26 @@ static void show_trace_log_lvl(struct task_struct *task, if (!task) task = current; + if (call_trace >= 0) { + int unw_ret = 0; + struct unwind_frame_info info; + + if (regs) { + if (unwind_init_frame_info(&info, task, regs) == 0) + unw_ret = show_trace_unwind(&info, log_lvl); + } else if (task == current) + unw_ret = unwind_init_running(&info, show_trace_unwind, log_lvl); + else { + if (unwind_init_blocked(&info, task) == 0) + unw_ret = show_trace_unwind(&info, log_lvl); + } + if (unw_ret > 0) { + if (call_trace > 0) + return; + printk("%sLegacy call trace:\n", log_lvl); + } + } + if (task == current) { /* Grab ebp right from our regs */ asm ("movl %%ebp, %0" : "=r" (ebp) : ); @@ -192,13 +214,13 @@ static void show_trace_log_lvl(struct task_struct *task, } } -void show_trace(struct task_struct *task, unsigned long * stack) +void show_trace(struct task_struct *task, struct pt_regs *regs, unsigned long * stack) { - show_trace_log_lvl(task, stack, ""); + show_trace_log_lvl(task, regs, stack, ""); } -static void show_stack_log_lvl(struct task_struct *task, unsigned long *esp, - char *log_lvl) +static void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, + unsigned long *esp, char *log_lvl) { unsigned long *stack; int i; @@ -219,13 +241,13 @@ static void show_stack_log_lvl(struct task_struct *task, unsigned long *esp, printk("%08lx ", *stack++); } printk("\n%sCall Trace:\n", log_lvl); - show_trace_log_lvl(task, esp, log_lvl); + show_trace_log_lvl(task, regs, esp, log_lvl); } void show_stack(struct task_struct *task, unsigned long *esp) { printk(" "); - show_stack_log_lvl(task, esp, ""); + show_stack_log_lvl(task, NULL, esp, ""); } /* @@ -235,7 +257,7 @@ void dump_stack(void) { unsigned long stack; - show_trace(current, &stack); + show_trace(current, NULL, &stack); } EXPORT_SYMBOL(dump_stack); @@ -268,8 +290,9 @@ void show_registers(struct pt_regs *regs) regs->esi, regs->edi, regs->ebp, esp); printk(KERN_EMERG "ds: %04x es: %04x ss: %04x\n", regs->xds & 0xffff, regs->xes & 0xffff, ss); - printk(KERN_EMERG "Process %s (pid: %d, threadinfo=%p task=%p)", - current->comm, current->pid, current_thread_info(), current); + printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)", + TASK_COMM_LEN, current->comm, current->pid, + current_thread_info(), current, current->thread_info); /* * When in-kernel, we also print out the stack and code at the * time of the fault.. @@ -278,7 +301,7 @@ void show_registers(struct pt_regs *regs) u8 __user *eip; printk("\n" KERN_EMERG "Stack: "); - show_stack_log_lvl(NULL, (unsigned long *)esp, KERN_EMERG); + show_stack_log_lvl(NULL, regs, (unsigned long *)esp, KERN_EMERG); printk(KERN_EMERG "Code: "); @@ -1208,3 +1231,15 @@ static int __init kstack_setup(char *s) return 1; } __setup("kstack=", kstack_setup); + +static int __init call_trace_setup(char *s) +{ + if (strcmp(s, "old") == 0) + call_trace = -1; + else if (strcmp(s, "both") == 0) + call_trace = 0; + else if (strcmp(s, "new") == 0) + call_trace = 1; + return 1; +} +__setup("call_trace=", call_trace_setup); diff --git a/arch/i386/kernel/tsc.c b/arch/i386/kernel/tsc.c new file mode 100644 index 000000000000..7e0d8dab2075 --- /dev/null +++ b/arch/i386/kernel/tsc.c @@ -0,0 +1,478 @@ +/* + * This code largely moved from arch/i386/kernel/timer/timer_tsc.c + * which was originally moved from arch/i386/kernel/time.c. + * See comments there for proper credits. + */ + +#include <linux/clocksource.h> +#include <linux/workqueue.h> +#include <linux/cpufreq.h> +#include <linux/jiffies.h> +#include <linux/init.h> +#include <linux/dmi.h> + +#include <asm/delay.h> +#include <asm/tsc.h> +#include <asm/delay.h> +#include <asm/io.h> + +#include "mach_timer.h" + +/* + * On some systems the TSC frequency does not + * change with the cpu frequency. So we need + * an extra value to store the TSC freq + */ +unsigned int tsc_khz; + +int tsc_disable __cpuinitdata = 0; + +#ifdef CONFIG_X86_TSC +static int __init tsc_setup(char *str) +{ + printk(KERN_WARNING "notsc: Kernel compiled with CONFIG_X86_TSC, " + "cannot disable TSC.\n"); + return 1; +} +#else +/* + * disable flag for tsc. Takes effect by clearing the TSC cpu flag + * in cpu/common.c + */ +static int __init tsc_setup(char *str) +{ + tsc_disable = 1; + + return 1; +} +#endif + +__setup("notsc", tsc_setup); + +/* + * code to mark and check if the TSC is unstable + * due to cpufreq or due to unsynced TSCs + */ +static int tsc_unstable; + +static inline int check_tsc_unstable(void) +{ + return tsc_unstable; +} + +void mark_tsc_unstable(void) +{ + tsc_unstable = 1; +} +EXPORT_SYMBOL_GPL(mark_tsc_unstable); + +/* Accellerators for sched_clock() + * convert from cycles(64bits) => nanoseconds (64bits) + * basic equation: + * ns = cycles / (freq / ns_per_sec) + * ns = cycles * (ns_per_sec / freq) + * ns = cycles * (10^9 / (cpu_khz * 10^3)) + * ns = cycles * (10^6 / cpu_khz) + * + * Then we use scaling math (suggested by george@mvista.com) to get: + * ns = cycles * (10^6 * SC / cpu_khz) / SC + * ns = cycles * cyc2ns_scale / SC + * + * And since SC is a constant power of two, we can convert the div + * into a shift. + * + * We can use khz divisor instead of mhz to keep a better percision, since + * cyc2ns_scale is limited to 10^6 * 2^10, which fits in 32 bits. + * (mathieu.desnoyers@polymtl.ca) + * + * -johnstul@us.ibm.com "math is hard, lets go shopping!" + */ +static unsigned long cyc2ns_scale __read_mostly; + +#define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ + +static inline void set_cyc2ns_scale(unsigned long cpu_khz) +{ + cyc2ns_scale = (1000000 << CYC2NS_SCALE_FACTOR)/cpu_khz; +} + +static inline unsigned long long cycles_2_ns(unsigned long long cyc) +{ + return (cyc * cyc2ns_scale) >> CYC2NS_SCALE_FACTOR; +} + +/* + * Scheduler clock - returns current time in nanosec units. + */ +unsigned long long sched_clock(void) +{ + unsigned long long this_offset; + + /* + * in the NUMA case we dont use the TSC as they are not + * synchronized across all CPUs. + */ +#ifndef CONFIG_NUMA + if (!cpu_khz || check_tsc_unstable()) +#endif + /* no locking but a rare wrong value is not a big deal */ + return (jiffies_64 - INITIAL_JIFFIES) * (1000000000 / HZ); + + /* read the Time Stamp Counter: */ + rdtscll(this_offset); + + /* return the value in ns */ + return cycles_2_ns(this_offset); +} + +static unsigned long calculate_cpu_khz(void) +{ + unsigned long long start, end; + unsigned long count; + u64 delta64; + int i; + unsigned long flags; + + local_irq_save(flags); + + /* run 3 times to ensure the cache is warm */ + for (i = 0; i < 3; i++) { + mach_prepare_counter(); + rdtscll(start); + mach_countup(&count); + rdtscll(end); + } + /* + * Error: ECTCNEVERSET + * The CTC wasn't reliable: we got a hit on the very first read, + * or the CPU was so fast/slow that the quotient wouldn't fit in + * 32 bits.. + */ + if (count <= 1) + goto err; + + delta64 = end - start; + + /* cpu freq too fast: */ + if (delta64 > (1ULL<<32)) + goto err; + + /* cpu freq too slow: */ + if (delta64 <= CALIBRATE_TIME_MSEC) + goto err; + + delta64 += CALIBRATE_TIME_MSEC/2; /* round for do_div */ + do_div(delta64,CALIBRATE_TIME_MSEC); + + local_irq_restore(flags); + return (unsigned long)delta64; +err: + local_irq_restore(flags); + return 0; +} + +int recalibrate_cpu_khz(void) +{ +#ifndef CONFIG_SMP + unsigned long cpu_khz_old = cpu_khz; + + if (cpu_has_tsc) { + cpu_khz = calculate_cpu_khz(); + tsc_khz = cpu_khz; + cpu_data[0].loops_per_jiffy = + cpufreq_scale(cpu_data[0].loops_per_jiffy, + cpu_khz_old, cpu_khz); + return 0; + } else + return -ENODEV; +#else + return -ENODEV; +#endif +} + +EXPORT_SYMBOL(recalibrate_cpu_khz); + +void tsc_init(void) +{ + if (!cpu_has_tsc || tsc_disable) + return; + + cpu_khz = calculate_cpu_khz(); + tsc_khz = cpu_khz; + + if (!cpu_khz) + return; + + printk("Detected %lu.%03lu MHz processor.\n", + (unsigned long)cpu_khz / 1000, + (unsigned long)cpu_khz % 1000); + + set_cyc2ns_scale(cpu_khz); + use_tsc_delay(); +} + +#ifdef CONFIG_CPU_FREQ + +static unsigned int cpufreq_delayed_issched = 0; +static unsigned int cpufreq_init = 0; +static struct work_struct cpufreq_delayed_get_work; + +static void handle_cpufreq_delayed_get(void *v) +{ + unsigned int cpu; + + for_each_online_cpu(cpu) + cpufreq_get(cpu); + + cpufreq_delayed_issched = 0; +} + +/* + * if we notice cpufreq oddness, schedule a call to cpufreq_get() as it tries + * to verify the CPU frequency the timing core thinks the CPU is running + * at is still correct. + */ +static inline void cpufreq_delayed_get(void) +{ + if (cpufreq_init && !cpufreq_delayed_issched) { + cpufreq_delayed_issched = 1; + printk(KERN_DEBUG "Checking if CPU frequency changed.\n"); + schedule_work(&cpufreq_delayed_get_work); + } +} + +/* + * if the CPU frequency is scaled, TSC-based delays will need a different + * loops_per_jiffy value to function properly. + */ +static unsigned int ref_freq = 0; +static unsigned long loops_per_jiffy_ref = 0; +static unsigned long cpu_khz_ref = 0; + +static int +time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data) +{ + struct cpufreq_freqs *freq = data; + + if (val != CPUFREQ_RESUMECHANGE && val != CPUFREQ_SUSPENDCHANGE) + write_seqlock_irq(&xtime_lock); + + if (!ref_freq) { + if (!freq->old){ + ref_freq = freq->new; + goto end; + } + ref_freq = freq->old; + loops_per_jiffy_ref = cpu_data[freq->cpu].loops_per_jiffy; + cpu_khz_ref = cpu_khz; + } + + if ((val == CPUFREQ_PRECHANGE && freq->old < freq->new) || + (val == CPUFREQ_POSTCHANGE && freq->old > freq->new) || + (val == CPUFREQ_RESUMECHANGE)) { + if (!(freq->flags & CPUFREQ_CONST_LOOPS)) + cpu_data[freq->cpu].loops_per_jiffy = + cpufreq_scale(loops_per_jiffy_ref, + ref_freq, freq->new); + + if (cpu_khz) { + + if (num_online_cpus() == 1) + cpu_khz = cpufreq_scale(cpu_khz_ref, + ref_freq, freq->new); + if (!(freq->flags & CPUFREQ_CONST_LOOPS)) { + tsc_khz = cpu_khz; + set_cyc2ns_scale(cpu_khz); + /* + * TSC based sched_clock turns + * to junk w/ cpufreq + */ + mark_tsc_unstable(); + } + } + } +end: + if (val != CPUFREQ_RESUMECHANGE && val != CPUFREQ_SUSPENDCHANGE) + write_sequnlock_irq(&xtime_lock); + + return 0; +} + +static struct notifier_block time_cpufreq_notifier_block = { + .notifier_call = time_cpufreq_notifier +}; + +static int __init cpufreq_tsc(void) +{ + int ret; + + INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get, NULL); + ret = cpufreq_register_notifier(&time_cpufreq_notifier_block, + CPUFREQ_TRANSITION_NOTIFIER); + if (!ret) + cpufreq_init = 1; + + return ret; +} + +core_initcall(cpufreq_tsc); + +#endif + +/* clock source code */ + +static unsigned long current_tsc_khz = 0; +static int tsc_update_callback(void); + +static cycle_t read_tsc(void) +{ + cycle_t ret; + + rdtscll(ret); + + return ret; +} + +static struct clocksource clocksource_tsc = { + .name = "tsc", + .rating = 300, + .read = read_tsc, + .mask = CLOCKSOURCE_MASK(64), + .mult = 0, /* to be set */ + .shift = 22, + .update_callback = tsc_update_callback, + .is_continuous = 1, +}; + +static int tsc_update_callback(void) +{ + int change = 0; + + /* check to see if we should switch to the safe clocksource: */ + if (clocksource_tsc.rating != 50 && check_tsc_unstable()) { + clocksource_tsc.rating = 50; + clocksource_reselect(); + change = 1; + } + + /* only update if tsc_khz has changed: */ + if (current_tsc_khz != tsc_khz) { + current_tsc_khz = tsc_khz; + clocksource_tsc.mult = clocksource_khz2mult(current_tsc_khz, + clocksource_tsc.shift); + change = 1; + } + + return change; +} + +static int __init dmi_mark_tsc_unstable(struct dmi_system_id *d) +{ + printk(KERN_NOTICE "%s detected: marking TSC unstable.\n", + d->ident); + mark_tsc_unstable(); + return 0; +} + +/* List of systems that have known TSC problems */ +static struct dmi_system_id __initdata bad_tsc_dmi_table[] = { + { + .callback = dmi_mark_tsc_unstable, + .ident = "IBM Thinkpad 380XD", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "IBM"), + DMI_MATCH(DMI_BOARD_NAME, "2635FA0"), + }, + }, + {} +}; + +#define TSC_FREQ_CHECK_INTERVAL (10*MSEC_PER_SEC) /* 10sec in MS */ +static struct timer_list verify_tsc_freq_timer; + +/* XXX - Probably should add locking */ +static void verify_tsc_freq(unsigned long unused) +{ + static u64 last_tsc; + static unsigned long last_jiffies; + + u64 now_tsc, interval_tsc; + unsigned long now_jiffies, interval_jiffies; + + + if (check_tsc_unstable()) + return; + + rdtscll(now_tsc); + now_jiffies = jiffies; + + if (!last_jiffies) { + goto out; + } + + interval_jiffies = now_jiffies - last_jiffies; + interval_tsc = now_tsc - last_tsc; + interval_tsc *= HZ; + do_div(interval_tsc, cpu_khz*1000); + + if (interval_tsc < (interval_jiffies * 3 / 4)) { + printk("TSC appears to be running slowly. " + "Marking it as unstable\n"); + mark_tsc_unstable(); + return; + } + +out: + last_tsc = now_tsc; + last_jiffies = now_jiffies; + /* set us up to go off on the next interval: */ + mod_timer(&verify_tsc_freq_timer, + jiffies + msecs_to_jiffies(TSC_FREQ_CHECK_INTERVAL)); +} + +/* + * Make an educated guess if the TSC is trustworthy and synchronized + * over all CPUs. + */ +static __init int unsynchronized_tsc(void) +{ + /* + * Intel systems are normally all synchronized. + * Exceptions must mark TSC as unstable: + */ + if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + return 0; + + /* assume multi socket systems are not synchronized: */ + return num_possible_cpus() > 1; +} + +static int __init init_tsc_clocksource(void) +{ + + if (cpu_has_tsc && tsc_khz && !tsc_disable) { + /* check blacklist */ + dmi_check_system(bad_tsc_dmi_table); + + if (unsynchronized_tsc()) /* mark unstable if unsynced */ + mark_tsc_unstable(); + current_tsc_khz = tsc_khz; + clocksource_tsc.mult = clocksource_khz2mult(current_tsc_khz, + clocksource_tsc.shift); + /* lower the rating if we already know its unstable: */ + if (check_tsc_unstable()) + clocksource_tsc.rating = 50; + + init_timer(&verify_tsc_freq_timer); + verify_tsc_freq_timer.function = verify_tsc_freq; + verify_tsc_freq_timer.expires = + jiffies + msecs_to_jiffies(TSC_FREQ_CHECK_INTERVAL); + add_timer(&verify_tsc_freq_timer); + + return clocksource_register(&clocksource_tsc); + } + + return 0; +} + +module_init(init_tsc_clocksource); diff --git a/arch/i386/kernel/vm86.c b/arch/i386/kernel/vm86.c index 00e0118e717c..8355d8d87d18 100644 --- a/arch/i386/kernel/vm86.c +++ b/arch/i386/kernel/vm86.c @@ -31,7 +31,6 @@ */ #include <linux/capability.h> -#include <linux/config.h> #include <linux/errno.h> #include <linux/interrupt.h> #include <linux/sched.h> diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 8831303a473f..2d4f1386e2b1 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -37,6 +37,13 @@ SECTIONS RODATA + . = ALIGN(4); + __tracedata_start = .; + .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) { + *(.tracedata) + } + __tracedata_end = .; + /* writeable */ .data : AT(ADDR(.data) - LOAD_OFFSET) { /* Data */ *(.data) @@ -64,6 +71,15 @@ SECTIONS .data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET) { *(.data.read_mostly) } _edata = .; /* End of data section */ +#ifdef CONFIG_STACK_UNWIND + . = ALIGN(4); + .eh_frame : AT(ADDR(.eh_frame) - LOAD_OFFSET) { + __start_unwind = .; + *(.eh_frame) + __end_unwind = .; + } +#endif + . = ALIGN(THREAD_SIZE); /* init_task */ .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) { *(.data.init_task) diff --git a/arch/i386/kernel/vsyscall-sysenter.S b/arch/i386/kernel/vsyscall-sysenter.S index 3b62baa6a371..1a36d26e15eb 100644 --- a/arch/i386/kernel/vsyscall-sysenter.S +++ b/arch/i386/kernel/vsyscall-sysenter.S @@ -42,10 +42,10 @@ __kernel_vsyscall: /* 7: align return point with nop's to make disassembly easier */ .space 7,0x90 - /* 14: System call restart point is here! (SYSENTER_RETURN - 2) */ + /* 14: System call restart point is here! (SYSENTER_RETURN-2) */ jmp .Lenter_kernel /* 16: System call normal return point is here! */ - .globl SYSENTER_RETURN /* Symbol used by entry.S. */ + .globl SYSENTER_RETURN /* Symbol used by sysenter.c */ SYSENTER_RETURN: pop %ebp .Lpop_ebp: diff --git a/arch/i386/kernel/vsyscall.lds.S b/arch/i386/kernel/vsyscall.lds.S index 98699ca6e52d..e26975fc68b6 100644 --- a/arch/i386/kernel/vsyscall.lds.S +++ b/arch/i386/kernel/vsyscall.lds.S @@ -7,7 +7,7 @@ SECTIONS { - . = VSYSCALL_BASE + SIZEOF_HEADERS; + . = VDSO_PRELINK + SIZEOF_HEADERS; .hash : { *(.hash) } :text .dynsym : { *(.dynsym) } @@ -20,7 +20,7 @@ SECTIONS For the layouts to match, we need to skip more than enough space for the dynamic symbol table et al. If this amount is insufficient, ld -shared will barf. Just increase it here. */ - . = VSYSCALL_BASE + 0x400; + . = VDSO_PRELINK + 0x400; .text : { *(.text) } :text =0x90909090 .note : { *(.note.*) } :text :note |