diff options
Diffstat (limited to 'arch/x86/kernel')
118 files changed, 6461 insertions, 7049 deletions
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 4eb5ce84110..c3920ea8ac5 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -2,8 +2,7 @@ # Makefile for the linux kernel. # -extra-y := head_$(BITS).o init_task.o vmlinux.lds -extra-$(CONFIG_X86_64) += head64.o +extra-y := head_$(BITS).o head$(BITS).o init_task.o vmlinux.lds CPPFLAGS_vmlinux.lds += -U$(UTS_MACHINE) @@ -19,7 +18,7 @@ CFLAGS_tsc_64.o := $(nostackp) obj-y := process_$(BITS).o signal_$(BITS).o entry_$(BITS).o obj-y += traps_$(BITS).o irq_$(BITS).o obj-y += time_$(BITS).o ioport.o ldt.o -obj-y += setup_$(BITS).o i8259_$(BITS).o +obj-y += setup_$(BITS).o i8259_$(BITS).o setup.o obj-$(CONFIG_X86_32) += sys_i386_32.o i386_ksyms_32.o obj-$(CONFIG_X86_64) += sys_x86_64.o x8664_ksyms_64.o obj-$(CONFIG_X86_64) += syscall_64.o vsyscall_64.o setup64.o @@ -29,6 +28,7 @@ obj-y += alternative.o i8253.o obj-$(CONFIG_X86_64) += pci-nommu_64.o bugs_64.o obj-y += tsc_$(BITS).o io_delay.o rtc.o +obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o obj-y += i387.o obj-y += ptrace.o obj-y += ds.o @@ -47,11 +47,12 @@ obj-$(CONFIG_MICROCODE) += microcode.o obj-$(CONFIG_PCI) += early-quirks.o apm-y := apm_32.o obj-$(CONFIG_APM) += apm.o -obj-$(CONFIG_X86_SMP) += smp_$(BITS).o smpboot_$(BITS).o tsc_sync.o -obj-$(CONFIG_X86_32_SMP) += smpcommon_32.o -obj-$(CONFIG_X86_64_SMP) += smp_64.o smpboot_64.o tsc_sync.o +obj-$(CONFIG_X86_SMP) += smp.o +obj-$(CONFIG_X86_SMP) += smpboot.o tsc_sync.o ipi.o tlb_$(BITS).o +obj-$(CONFIG_X86_32_SMP) += smpcommon.o +obj-$(CONFIG_X86_64_SMP) += tsc_sync.o smpcommon.o obj-$(CONFIG_X86_TRAMPOLINE) += trampoline_$(BITS).o -obj-$(CONFIG_X86_MPPARSE) += mpparse_$(BITS).o +obj-$(CONFIG_X86_MPPARSE) += mpparse.o obj-$(CONFIG_X86_LOCAL_APIC) += apic_$(BITS).o nmi_$(BITS).o obj-$(CONFIG_X86_IO_APIC) += io_apic_$(BITS).o obj-$(CONFIG_X86_REBOOTFIXUPS) += reboot_fixups_32.o @@ -60,12 +61,13 @@ obj-$(CONFIG_KEXEC) += relocate_kernel_$(BITS).o crash.o obj-$(CONFIG_CRASH_DUMP) += crash_dump_$(BITS).o obj-$(CONFIG_X86_NUMAQ) += numaq_32.o obj-$(CONFIG_X86_SUMMIT_NUMA) += summit_32.o -obj-$(CONFIG_X86_VSMP) += vsmp_64.o +obj-y += vsmp_64.o obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_MODULES) += module_$(BITS).o obj-$(CONFIG_ACPI_SRAT) += srat_32.o obj-$(CONFIG_EFI) += efi.o efi_$(BITS).o efi_stub_$(BITS).o obj-$(CONFIG_DOUBLEFAULT) += doublefault_32.o +obj-$(CONFIG_KGDB) += kgdb.o obj-$(CONFIG_VM86) += vm86_32.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o @@ -89,7 +91,7 @@ scx200-y += scx200_32.o ### # 64 bit specific files ifeq ($(CONFIG_X86_64),y) - obj-y += genapic_64.o genapic_flat_64.o + obj-y += genapic_64.o genapic_flat_64.o genx2apic_uv_x.o obj-$(CONFIG_X86_PM_TIMER) += pmtimer_64.o obj-$(CONFIG_AUDIT) += audit_64.o diff --git a/arch/x86/kernel/acpi/Makefile b/arch/x86/kernel/acpi/Makefile index 19d3d6e9d09..7335959b6af 100644 --- a/arch/x86/kernel/acpi/Makefile +++ b/arch/x86/kernel/acpi/Makefile @@ -1,7 +1,14 @@ +subdir- := realmode + obj-$(CONFIG_ACPI) += boot.o -obj-$(CONFIG_ACPI_SLEEP) += sleep.o wakeup_$(BITS).o +obj-$(CONFIG_ACPI_SLEEP) += sleep.o wakeup_rm.o wakeup_$(BITS).o ifneq ($(CONFIG_ACPI_PROCESSOR),) obj-y += cstate.o processor.o endif +$(obj)/wakeup_rm.o: $(obj)/realmode/wakeup.bin + +$(obj)/realmode/wakeup.bin: FORCE + $(Q)$(MAKE) $(build)=$(obj)/realmode $@ + diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 2cdc9de9371..057ccf1d5ad 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -39,6 +39,11 @@ #include <asm/apic.h> #include <asm/io.h> #include <asm/mpspec.h> +#include <asm/smp.h> + +#ifdef CONFIG_X86_LOCAL_APIC +# include <mach_apic.h> +#endif static int __initdata acpi_force = 0; @@ -52,9 +57,7 @@ EXPORT_SYMBOL(acpi_disabled); #ifdef CONFIG_X86_64 #include <asm/proto.h> - -static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 0; } - +#include <asm/genapic.h> #else /* X86 */ @@ -111,7 +114,7 @@ char *__init __acpi_map_table(unsigned long phys_addr, unsigned long size) if (!phys_addr || !size) return NULL; - if (phys_addr+size <= (end_pfn_map << PAGE_SHIFT) + PAGE_SIZE) + if (phys_addr+size <= (max_pfn_mapped << PAGE_SHIFT) + PAGE_SIZE) return __va(phys_addr); return NULL; @@ -237,6 +240,16 @@ static int __init acpi_parse_madt(struct acpi_table_header *table) return 0; } +static void __cpuinit acpi_register_lapic(int id, u8 enabled) +{ + if (!enabled) { + ++disabled_cpus; + return; + } + + generic_processor_info(id, 0); +} + static int __init acpi_parse_lapic(struct acpi_subtable_header * header, const unsigned long end) { @@ -256,8 +269,26 @@ acpi_parse_lapic(struct acpi_subtable_header * header, const unsigned long end) * to not preallocating memory for all NR_CPUS * when we use CPU hotplug. */ - mp_register_lapic(processor->id, /* APIC ID */ - processor->lapic_flags & ACPI_MADT_ENABLED); /* Enabled? */ + acpi_register_lapic(processor->id, /* APIC ID */ + processor->lapic_flags & ACPI_MADT_ENABLED); + + return 0; +} + +static int __init +acpi_parse_sapic(struct acpi_subtable_header *header, const unsigned long end) +{ + struct acpi_madt_local_sapic *processor = NULL; + + processor = (struct acpi_madt_local_sapic *)header; + + if (BAD_MADT_ENTRY(processor, end)) + return -EINVAL; + + acpi_table_print_madt_entry(header); + + acpi_register_lapic((processor->id << 8) | processor->eid,/* APIC ID */ + processor->lapic_flags & ACPI_MADT_ENABLED); return 0; } @@ -300,6 +331,8 @@ acpi_parse_lapic_nmi(struct acpi_subtable_header * header, const unsigned long e #ifdef CONFIG_X86_IO_APIC +struct mp_ioapic_routing mp_ioapic_routing[MAX_IO_APICS]; + static int __init acpi_parse_ioapic(struct acpi_subtable_header * header, const unsigned long end) { @@ -532,7 +565,7 @@ static int __cpuinit _acpi_map_lsapic(acpi_handle handle, int *pcpu) buffer.pointer = NULL; tmp_map = cpu_present_map; - mp_register_lapic(physid, lapic->lapic_flags & ACPI_MADT_ENABLED); + acpi_register_lapic(physid, lapic->lapic_flags & ACPI_MADT_ENABLED); /* * If mp_register_lapic successfully generates a new logical cpu @@ -732,6 +765,16 @@ static int __init acpi_parse_fadt(struct acpi_table_header *table) * Parse LAPIC entries in MADT * returns 0 on success, < 0 on error */ + +static void __init acpi_register_lapic_address(unsigned long address) +{ + mp_lapic_addr = address; + + set_fixmap_nocache(FIX_APIC_BASE, address); + if (boot_cpu_physical_apicid == -1U) + boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); +} + static int __init acpi_parse_madt_lapic_entries(void) { int count; @@ -753,10 +796,14 @@ static int __init acpi_parse_madt_lapic_entries(void) return count; } - mp_register_lapic_address(acpi_lapic_addr); + acpi_register_lapic_address(acpi_lapic_addr); + + count = acpi_table_parse_madt(ACPI_MADT_TYPE_LOCAL_SAPIC, + acpi_parse_sapic, MAX_APICS); - count = acpi_table_parse_madt(ACPI_MADT_TYPE_LOCAL_APIC, acpi_parse_lapic, - MAX_APICS); + if (!count) + count = acpi_table_parse_madt(ACPI_MADT_TYPE_LOCAL_APIC, + acpi_parse_lapic, MAX_APICS); if (!count) { printk(KERN_ERR PREFIX "No LAPIC entries present\n"); /* TBD: Cleanup to allow fallback to MPS */ diff --git a/arch/x86/kernel/acpi/realmode/Makefile b/arch/x86/kernel/acpi/realmode/Makefile new file mode 100644 index 00000000000..092900854ac --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/Makefile @@ -0,0 +1,57 @@ +# +# arch/x86/kernel/acpi/realmode/Makefile +# +# This file is subject to the terms and conditions of the GNU General Public +# License. See the file "COPYING" in the main directory of this archive +# for more details. +# + +targets := wakeup.bin wakeup.elf + +wakeup-y += wakeup.o wakemain.o video-mode.o copy.o + +# The link order of the video-*.o modules can matter. In particular, +# video-vga.o *must* be listed first, followed by video-vesa.o. +# Hardware-specific drivers should follow in the order they should be +# probed, and video-bios.o should typically be last. +wakeup-y += video-vga.o +wakeup-y += video-vesa.o +wakeup-y += video-bios.o + +targets += $(wakeup-y) + +bootsrc := $(src)/../../../boot + +# --------------------------------------------------------------------------- + +# How to compile the 16-bit code. Note we always compile for -march=i386, +# that way we can complain to the user if the CPU is insufficient. +# Compile with _SETUP since this is similar to the boot-time setup code. +KBUILD_CFLAGS := $(LINUXINCLUDE) -g -Os -D_SETUP -D_WAKEUP -D__KERNEL__ \ + -I$(srctree)/$(bootsrc) \ + $(cflags-y) \ + -Wall -Wstrict-prototypes \ + -march=i386 -mregparm=3 \ + -include $(srctree)/$(bootsrc)/code16gcc.h \ + -fno-strict-aliasing -fomit-frame-pointer \ + $(call cc-option, -ffreestanding) \ + $(call cc-option, -fno-toplevel-reorder,\ + $(call cc-option, -fno-unit-at-a-time)) \ + $(call cc-option, -fno-stack-protector) \ + $(call cc-option, -mpreferred-stack-boundary=2) +KBUILD_CFLAGS += $(call cc-option, -m32) +KBUILD_AFLAGS := $(KBUILD_CFLAGS) -D__ASSEMBLY__ + +WAKEUP_OBJS = $(addprefix $(obj)/,$(wakeup-y)) + +LDFLAGS_wakeup.elf := -T + +CPPFLAGS_wakeup.lds += -P -C + +$(obj)/wakeup.elf: $(src)/wakeup.lds $(WAKEUP_OBJS) FORCE + $(call if_changed,ld) + +OBJCOPYFLAGS_wakeup.bin := -O binary + +$(obj)/wakeup.bin: $(obj)/wakeup.elf FORCE + $(call if_changed,objcopy) diff --git a/arch/x86/kernel/acpi/realmode/copy.S b/arch/x86/kernel/acpi/realmode/copy.S new file mode 100644 index 00000000000..dc59ebee69d --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/copy.S @@ -0,0 +1 @@ +#include "../../../boot/copy.S" diff --git a/arch/x86/kernel/acpi/realmode/video-bios.c b/arch/x86/kernel/acpi/realmode/video-bios.c new file mode 100644 index 00000000000..7deabc144a2 --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/video-bios.c @@ -0,0 +1 @@ +#include "../../../boot/video-bios.c" diff --git a/arch/x86/kernel/acpi/realmode/video-mode.c b/arch/x86/kernel/acpi/realmode/video-mode.c new file mode 100644 index 00000000000..328ad209f11 --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/video-mode.c @@ -0,0 +1 @@ +#include "../../../boot/video-mode.c" diff --git a/arch/x86/kernel/acpi/realmode/video-vesa.c b/arch/x86/kernel/acpi/realmode/video-vesa.c new file mode 100644 index 00000000000..9dbb9672226 --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/video-vesa.c @@ -0,0 +1 @@ +#include "../../../boot/video-vesa.c" diff --git a/arch/x86/kernel/acpi/realmode/video-vga.c b/arch/x86/kernel/acpi/realmode/video-vga.c new file mode 100644 index 00000000000..bcc81255f37 --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/video-vga.c @@ -0,0 +1 @@ +#include "../../../boot/video-vga.c" diff --git a/arch/x86/kernel/acpi/realmode/wakemain.c b/arch/x86/kernel/acpi/realmode/wakemain.c new file mode 100644 index 00000000000..883962d9eef --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/wakemain.c @@ -0,0 +1,81 @@ +#include "wakeup.h" +#include "boot.h" + +static void udelay(int loops) +{ + while (loops--) + io_delay(); /* Approximately 1 us */ +} + +static void beep(unsigned int hz) +{ + u8 enable; + + if (!hz) { + enable = 0x00; /* Turn off speaker */ + } else { + u16 div = 1193181/hz; + + outb(0xb6, 0x43); /* Ctr 2, squarewave, load, binary */ + io_delay(); + outb(div, 0x42); /* LSB of counter */ + io_delay(); + outb(div >> 8, 0x42); /* MSB of counter */ + io_delay(); + + enable = 0x03; /* Turn on speaker */ + } + inb(0x61); /* Dummy read of System Control Port B */ + io_delay(); + outb(enable, 0x61); /* Enable timer 2 output to speaker */ + io_delay(); +} + +#define DOT_HZ 880 +#define DASH_HZ 587 +#define US_PER_DOT 125000 + +/* Okay, this is totally silly, but it's kind of fun. */ +static void send_morse(const char *pattern) +{ + char s; + + while ((s = *pattern++)) { + switch (s) { + case '.': + beep(DOT_HZ); + udelay(US_PER_DOT); + beep(0); + udelay(US_PER_DOT); + break; + case '-': + beep(DASH_HZ); + udelay(US_PER_DOT * 3); + beep(0); + udelay(US_PER_DOT); + break; + default: /* Assume it's a space */ + udelay(US_PER_DOT * 3); + break; + } + } +} + +void main(void) +{ + /* Kill machine if structures are wrong */ + if (wakeup_header.real_magic != 0x12345678) + while (1); + + if (wakeup_header.realmode_flags & 4) + send_morse("...-"); + + if (wakeup_header.realmode_flags & 1) + asm volatile("lcallw $0xc000,$3"); + + if (wakeup_header.realmode_flags & 2) { + /* Need to call BIOS */ + probe_cards(0); + set_mode(wakeup_header.video_mode); + } +} diff --git a/arch/x86/kernel/acpi/realmode/wakeup.S b/arch/x86/kernel/acpi/realmode/wakeup.S new file mode 100644 index 00000000000..f9b77fb37e5 --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/wakeup.S @@ -0,0 +1,113 @@ +/* + * ACPI wakeup real mode startup stub + */ +#include <asm/segment.h> +#include <asm/msr-index.h> +#include <asm/page.h> +#include <asm/pgtable.h> + + .code16 + .section ".header", "a" + +/* This should match the structure in wakeup.h */ + .globl wakeup_header +wakeup_header: +video_mode: .short 0 /* Video mode number */ +pmode_return: .byte 0x66, 0xea /* ljmpl */ + .long 0 /* offset goes here */ + .short __KERNEL_CS +pmode_cr0: .long 0 /* Saved %cr0 */ +pmode_cr3: .long 0 /* Saved %cr3 */ +pmode_cr4: .long 0 /* Saved %cr4 */ +pmode_efer: .quad 0 /* Saved EFER */ +pmode_gdt: .quad 0 +realmode_flags: .long 0 +real_magic: .long 0 +trampoline_segment: .word 0 +signature: .long 0x51ee1111 + + .text + .globl _start + .code16 +wakeup_code: +_start: + cli + cld + + /* Set up segments */ + movw %cs, %ax + movw %ax, %ds + movw %ax, %es + movw %ax, %ss + + movl $wakeup_stack_end, %esp + + /* Clear the EFLAGS */ + pushl $0 + popfl + + /* Check header signature... */ + movl signature, %eax + cmpl $0x51ee1111, %eax + jne bogus_real_magic + + /* Check we really have everything... */ + movl end_signature, %eax + cmpl $0x65a22c82, %eax + jne bogus_real_magic + + /* Call the C code */ + calll main + + /* Do any other stuff... */ + +#ifndef CONFIG_64BIT + /* This could also be done in C code... */ + movl pmode_cr3, %eax + movl %eax, %cr3 + + movl pmode_cr4, %ecx + jecxz 1f + movl %ecx, %cr4 +1: + movl pmode_efer, %eax + movl pmode_efer + 4, %edx + movl %eax, %ecx + orl %edx, %ecx + jz 1f + movl $0xc0000080, %ecx + wrmsr +1: + + lgdtl pmode_gdt + + /* This really couldn't... */ + movl pmode_cr0, %eax + movl %eax, %cr0 + jmp pmode_return +#else + pushw $0 + pushw trampoline_segment + pushw $0 + lret +#endif + +bogus_real_magic: +1: + hlt + jmp 1b + + .data + .balign 4 + .globl HEAP, heap_end +HEAP: + .long wakeup_heap +heap_end: + .long wakeup_stack + + .bss +wakeup_heap: + .space 2048 +wakeup_stack: + .space 2048 +wakeup_stack_end: diff --git a/arch/x86/kernel/acpi/realmode/wakeup.h b/arch/x86/kernel/acpi/realmode/wakeup.h new file mode 100644 index 00000000000..ef8166fe802 --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/wakeup.h @@ -0,0 +1,36 @@ +/* + * Definitions for the wakeup data structure at the head of the + * wakeup code. + */ + +#ifndef ARCH_X86_KERNEL_ACPI_RM_WAKEUP_H +#define ARCH_X86_KERNEL_ACPI_RM_WAKEUP_H + +#ifndef __ASSEMBLY__ +#include <linux/types.h> + +/* This must match data at wakeup.S */ +struct wakeup_header { + u16 video_mode; /* Video mode number */ + u16 _jmp1; /* ljmpl opcode, 32-bit only */ + u32 pmode_entry; /* Protected mode resume point, 32-bit only */ + u16 _jmp2; /* CS value, 32-bit only */ + u32 pmode_cr0; /* Protected mode cr0 */ + u32 pmode_cr3; /* Protected mode cr3 */ + u32 pmode_cr4; /* Protected mode cr4 */ + u32 pmode_efer_low; /* Protected mode EFER */ + u32 pmode_efer_high; + u64 pmode_gdt; + u32 realmode_flags; + u32 real_magic; + u16 trampoline_segment; /* segment with trampoline code, 64-bit only */ + u32 signature; /* To check we have correct structure */ +} __attribute__((__packed__)); + +extern struct wakeup_header wakeup_header; +#endif + +#define HEADER_OFFSET 0x3f00 +#define WAKEUP_SIZE 0x4000 + +#endif /* ARCH_X86_KERNEL_ACPI_RM_WAKEUP_H */ diff --git a/arch/x86/kernel/acpi/realmode/wakeup.lds.S b/arch/x86/kernel/acpi/realmode/wakeup.lds.S new file mode 100644 index 00000000000..22fab6c4be1 --- /dev/null +++ b/arch/x86/kernel/acpi/realmode/wakeup.lds.S @@ -0,0 +1,61 @@ +/* + * wakeup.ld + * + * Linker script for the real-mode wakeup code + */ +#undef i386 +#include "wakeup.h" + +OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386") +OUTPUT_ARCH(i386) +ENTRY(_start) + +SECTIONS +{ + . = HEADER_OFFSET; + .header : { + *(.header) + } + + . = 0; + .text : { + *(.text*) + } + + . = ALIGN(16); + .rodata : { + *(.rodata*) + } + + .videocards : { + video_cards = .; + *(.videocards) + video_cards_end = .; + } + + . = ALIGN(16); + .data : { + *(.data*) + } + + .signature : { + end_signature = .; + LONG(0x65a22c82) + } + + . = ALIGN(16); + .bss : { + __bss_start = .; + *(.bss) + __bss_end = .; + } + + . = ALIGN(16); + _end = .; + + /DISCARD/ : { + *(.note*) + } + + . = ASSERT(_end <= WAKEUP_SIZE, "Wakeup too big!"); +} diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 6bc815cd8cb..afc25ee9964 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -10,30 +10,72 @@ #include <linux/dmi.h> #include <linux/cpumask.h> -#include <asm/smp.h> +#include "realmode/wakeup.h" +#include "sleep.h" -/* address in low memory of the wakeup routine. */ -unsigned long acpi_wakeup_address = 0; +unsigned long acpi_wakeup_address; unsigned long acpi_realmode_flags; -extern char wakeup_start, wakeup_end; -extern unsigned long acpi_copy_wakeup_routine(unsigned long); +/* address in low memory of the wakeup routine. */ +static unsigned long acpi_realmode; + +#ifdef CONFIG_64BIT +static char temp_stack[10240]; +#endif /** * acpi_save_state_mem - save kernel state * * Create an identity mapped page table and copy the wakeup routine to * low memory. + * + * Note that this is too late to change acpi_wakeup_address. */ int acpi_save_state_mem(void) { - if (!acpi_wakeup_address) { - printk(KERN_ERR "Could not allocate memory during boot, S3 disabled\n"); + struct wakeup_header *header; + + if (!acpi_realmode) { + printk(KERN_ERR "Could not allocate memory during boot, " + "S3 disabled\n"); return -ENOMEM; } - memcpy((void *)acpi_wakeup_address, &wakeup_start, - &wakeup_end - &wakeup_start); - acpi_copy_wakeup_routine(acpi_wakeup_address); + memcpy((void *)acpi_realmode, &wakeup_code_start, WAKEUP_SIZE); + + header = (struct wakeup_header *)(acpi_realmode + HEADER_OFFSET); + if (header->signature != 0x51ee1111) { + printk(KERN_ERR "wakeup header does not match\n"); + return -EINVAL; + } + + header->video_mode = saved_video_mode; + +#ifndef CONFIG_64BIT + store_gdt((struct desc_ptr *)&header->pmode_gdt); + + header->pmode_efer_low = nx_enabled; + if (header->pmode_efer_low & 1) { + /* This is strange, why not save efer, always? */ + rdmsr(MSR_EFER, header->pmode_efer_low, + header->pmode_efer_high); + } +#endif /* !CONFIG_64BIT */ + + header->pmode_cr0 = read_cr0(); + header->pmode_cr4 = read_cr4(); + header->realmode_flags = acpi_realmode_flags; + header->real_magic = 0x12345678; + +#ifndef CONFIG_64BIT + header->pmode_entry = (u32)&wakeup_pmode_return; + header->pmode_cr3 = (u32)(swsusp_pg_dir - __PAGE_OFFSET); + saved_magic = 0x12345678; +#else /* CONFIG_64BIT */ + header->trampoline_segment = setup_trampoline() >> 4; + init_rsp = (unsigned long)temp_stack + 4096; + initial_code = (unsigned long)wakeup_long64; + saved_magic = 0x123456789abcdef0; +#endif /* CONFIG_64BIT */ return 0; } @@ -56,15 +98,20 @@ void acpi_restore_state_mem(void) */ void __init acpi_reserve_bootmem(void) { - if ((&wakeup_end - &wakeup_start) > PAGE_SIZE*2) { + if ((&wakeup_code_end - &wakeup_code_start) > WAKEUP_SIZE) { printk(KERN_ERR "ACPI: Wakeup code way too big, S3 disabled.\n"); return; } - acpi_wakeup_address = (unsigned long)alloc_bootmem_low(PAGE_SIZE*2); - if (!acpi_wakeup_address) + acpi_realmode = (unsigned long)alloc_bootmem_low(WAKEUP_SIZE); + + if (!acpi_realmode) { printk(KERN_ERR "ACPI: Cannot allocate lowmem, S3 disabled.\n"); + return; + } + + acpi_wakeup_address = acpi_realmode; } diff --git a/arch/x86/kernel/acpi/sleep.h b/arch/x86/kernel/acpi/sleep.h new file mode 100644 index 00000000000..adbcbaa6f1d --- /dev/null +++ b/arch/x86/kernel/acpi/sleep.h @@ -0,0 +1,16 @@ +/* + * Variables and functions used by the code in sleep.c + */ + +#include <asm/trampoline.h> + +extern char wakeup_code_start, wakeup_code_end; + +extern unsigned long saved_video_mode; +extern long saved_magic; + +extern int wakeup_pmode_return; +extern char swsusp_pg_dir[PAGE_SIZE]; + +extern unsigned long acpi_copy_wakeup_routine(unsigned long); +extern void wakeup_long64(void); diff --git a/arch/x86/kernel/acpi/sleep_32.c b/arch/x86/kernel/acpi/sleep_32.c deleted file mode 100644 index 63fe5525e02..00000000000 --- a/arch/x86/kernel/acpi/sleep_32.c +++ /dev/null @@ -1,40 +0,0 @@ -/* - * sleep.c - x86-specific ACPI sleep support. - * - * Copyright (C) 2001-2003 Patrick Mochel - * Copyright (C) 2001-2003 Pavel Machek <pavel@suse.cz> - */ - -#include <linux/acpi.h> -#include <linux/bootmem.h> -#include <linux/dmi.h> -#include <linux/cpumask.h> - -#include <asm/smp.h> - -/* Ouch, we want to delete this. We already have better version in userspace, in - s2ram from suspend.sf.net project */ -static __init int reset_videomode_after_s3(const struct dmi_system_id *d) -{ - acpi_realmode_flags |= 2; - return 0; -} - -static __initdata struct dmi_system_id acpisleep_dmi_table[] = { - { /* Reset video mode after returning from ACPI S3 sleep */ - .callback = reset_videomode_after_s3, - .ident = "Toshiba Satellite 4030cdt", - .matches = { - DMI_MATCH(DMI_PRODUCT_NAME, "S4030CDT/4.3"), - }, - }, - {} -}; - -static int __init acpisleep_dmi_init(void) -{ - dmi_check_system(acpisleep_dmi_table); - return 0; -} - -core_initcall(acpisleep_dmi_init); diff --git a/arch/x86/kernel/acpi/wakeup_32.S b/arch/x86/kernel/acpi/wakeup_32.S index f53e3277f8e..a12e6a9fb65 100644 --- a/arch/x86/kernel/acpi/wakeup_32.S +++ b/arch/x86/kernel/acpi/wakeup_32.S @@ -3,178 +3,12 @@ #include <asm/segment.h> #include <asm/page.h> -# -# wakeup_code runs in real mode, and at unknown address (determined at run-time). -# Therefore it must only use relative jumps/calls. -# -# Do we need to deal with A20? It is okay: ACPI specs says A20 must be enabled -# -# If physical address of wakeup_code is 0x12345, BIOS should call us with -# cs = 0x1234, eip = 0x05 -# - -#define BEEP \ - inb $97, %al; \ - outb %al, $0x80; \ - movb $3, %al; \ - outb %al, $97; \ - outb %al, $0x80; \ - movb $-74, %al; \ - outb %al, $67; \ - outb %al, $0x80; \ - movb $-119, %al; \ - outb %al, $66; \ - outb %al, $0x80; \ - movb $15, %al; \ - outb %al, $66; - -ALIGN - .align 4096 -ENTRY(wakeup_start) -wakeup_code: - wakeup_code_start = . - .code16 - - cli - cld - - # setup data segment - movw %cs, %ax - movw %ax, %ds # Make ds:0 point to wakeup_start - movw %ax, %ss - - testl $4, realmode_flags - wakeup_code - jz 1f - BEEP -1: - mov $(wakeup_stack - wakeup_code), %sp # Private stack is needed for ASUS board - - pushl $0 # Kill any dangerous flags - popfl - - movl real_magic - wakeup_code, %eax - cmpl $0x12345678, %eax - jne bogus_real_magic - - testl $1, realmode_flags - wakeup_code - jz 1f - lcall $0xc000,$3 - movw %cs, %ax - movw %ax, %ds # Bios might have played with that - movw %ax, %ss -1: - - testl $2, realmode_flags - wakeup_code - jz 1f - mov video_mode - wakeup_code, %ax - call mode_set -1: - - # set up page table - movl $swsusp_pg_dir-__PAGE_OFFSET, %eax - movl %eax, %cr3 - - testl $1, real_efer_save_restore - wakeup_code - jz 4f - # restore efer setting - movl real_save_efer_edx - wakeup_code, %edx - movl real_save_efer_eax - wakeup_code, %eax - mov $0xc0000080, %ecx - wrmsr -4: - # make sure %cr4 is set correctly (features, etc) - movl real_save_cr4 - wakeup_code, %eax - movl %eax, %cr4 - - # need a gdt -- use lgdtl to force 32-bit operands, in case - # the GDT is located past 16 megabytes. - lgdtl real_save_gdt - wakeup_code - - movl real_save_cr0 - wakeup_code, %eax - movl %eax, %cr0 - jmp 1f -1: - movl real_magic - wakeup_code, %eax - cmpl $0x12345678, %eax - jne bogus_real_magic - - testl $8, realmode_flags - wakeup_code - jz 1f - BEEP -1: - ljmpl $__KERNEL_CS, $wakeup_pmode_return - -real_save_gdt: .word 0 - .long 0 -real_save_cr0: .long 0 -real_save_cr3: .long 0 -real_save_cr4: .long 0 -real_magic: .long 0 -video_mode: .long 0 -realmode_flags: .long 0 -real_efer_save_restore: .long 0 -real_save_efer_edx: .long 0 -real_save_efer_eax: .long 0 - -bogus_real_magic: - jmp bogus_real_magic - -/* This code uses an extended set of video mode numbers. These include: - * Aliases for standard modes - * NORMAL_VGA (-1) - * EXTENDED_VGA (-2) - * ASK_VGA (-3) - * Video modes numbered by menu position -- NOT RECOMMENDED because of lack - * of compatibility when extending the table. These are between 0x00 and 0xff. - */ -#define VIDEO_FIRST_MENU 0x0000 - -/* Standard BIOS video modes (BIOS number + 0x0100) */ -#define VIDEO_FIRST_BIOS 0x0100 - -/* VESA BIOS video modes (VESA number + 0x0200) */ -#define VIDEO_FIRST_VESA 0x0200 - -/* Video7 special modes (BIOS number + 0x0900) */ -#define VIDEO_FIRST_V7 0x0900 - -# Setting of user mode (AX=mode ID) => CF=success - -# For now, we only handle VESA modes (0x0200..0x03ff). To handle other -# modes, we should probably compile in the video code from the boot -# directory. -mode_set: - movw %ax, %bx - subb $VIDEO_FIRST_VESA>>8, %bh - cmpb $2, %bh - jb check_vesa - -setbad: - clc - ret - -check_vesa: - orw $0x4000, %bx # Use linear frame buffer - movw $0x4f02, %ax # VESA BIOS mode set call - int $0x10 - cmpw $0x004f, %ax # AL=4f if implemented - jnz setbad # AH=0 if OK - - stc - ret +# Copyright 2003, 2008 Pavel Machek <pavel@suse.cz>, distribute under GPLv2 .code32 ALIGN -.org 0x800 -wakeup_stack_begin: # Stack grows down - -.org 0xff0 # Just below end of page -wakeup_stack: -ENTRY(wakeup_end) - -.org 0x1000 - +ENTRY(wakeup_pmode_return) wakeup_pmode_return: movw $__KERNEL_DS, %ax movw %ax, %ss @@ -187,7 +21,7 @@ wakeup_pmode_return: lgdt saved_gdt lidt saved_idt lldt saved_ldt - ljmp $(__KERNEL_CS),$1f + ljmp $(__KERNEL_CS), $1f 1: movl %cr3, %eax movl %eax, %cr3 @@ -201,82 +35,41 @@ wakeup_pmode_return: jne bogus_magic # jump to place where we left off - movl saved_eip,%eax + movl saved_eip, %eax jmp *%eax bogus_magic: jmp bogus_magic -## -# acpi_copy_wakeup_routine -# -# Copy the above routine to low memory. -# -# Parameters: -# %eax: place to copy wakeup routine to -# -# Returned address is location of code in low memory (past data and stack) -# -ENTRY(acpi_copy_wakeup_routine) - pushl %ebx +save_registers: sgdt saved_gdt sidt saved_idt sldt saved_ldt str saved_tss - movl nx_enabled, %edx - movl %edx, real_efer_save_restore - wakeup_start (%eax) - testl $1, real_efer_save_restore - wakeup_start (%eax) - jz 2f - # save efer setting - pushl %eax - movl %eax, %ebx - mov $0xc0000080, %ecx - rdmsr - movl %edx, real_save_efer_edx - wakeup_start (%ebx) - movl %eax, real_save_efer_eax - wakeup_start (%ebx) - popl %eax -2: - - movl %cr3, %edx - movl %edx, real_save_cr3 - wakeup_start (%eax) - movl %cr4, %edx - movl %edx, real_save_cr4 - wakeup_start (%eax) - movl %cr0, %edx - movl %edx, real_save_cr0 - wakeup_start (%eax) - sgdt real_save_gdt - wakeup_start (%eax) - - movl saved_videomode, %edx - movl %edx, video_mode - wakeup_start (%eax) - movl acpi_realmode_flags, %edx - movl %edx, realmode_flags - wakeup_start (%eax) - movl $0x12345678, real_magic - wakeup_start (%eax) - movl $0x12345678, saved_magic - popl %ebx - ret - -save_registers: leal 4(%esp), %eax movl %eax, saved_context_esp - movl %ebx, saved_context_ebx - movl %ebp, saved_context_ebp - movl %esi, saved_context_esi - movl %edi, saved_context_edi - pushfl ; popl saved_context_eflags - - movl $ret_point, saved_eip + movl %ebx, saved_context_ebx + movl %ebp, saved_context_ebp + movl %esi, saved_context_esi + movl %edi, saved_context_edi + pushfl + popl saved_context_eflags + + movl $ret_point, saved_eip ret restore_registers: - movl saved_context_ebp, %ebp - movl saved_context_ebx, %ebx - movl saved_context_esi, %esi - movl saved_context_edi, %edi - pushl saved_context_eflags ; popfl - ret + movl saved_context_ebp, %ebp + movl saved_context_ebx, %ebx + movl saved_context_esi, %esi + movl saved_context_edi, %edi + pushl saved_context_eflags + popfl + ret ENTRY(do_suspend_lowlevel) call save_processor_state diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S index 2e1b9e0d076..bcc293423a7 100644 --- a/arch/x86/kernel/acpi/wakeup_64.S +++ b/arch/x86/kernel/acpi/wakeup_64.S @@ -7,191 +7,18 @@ #include <asm/asm-offsets.h> # Copyright 2003 Pavel Machek <pavel@suse.cz>, distribute under GPLv2 -# -# wakeup_code runs in real mode, and at unknown address (determined at run-time). -# Therefore it must only use relative jumps/calls. -# -# Do we need to deal with A20? It is okay: ACPI specs says A20 must be enabled -# -# If physical address of wakeup_code is 0x12345, BIOS should call us with -# cs = 0x1234, eip = 0x05 -# - -#define BEEP \ - inb $97, %al; \ - outb %al, $0x80; \ - movb $3, %al; \ - outb %al, $97; \ - outb %al, $0x80; \ - movb $-74, %al; \ - outb %al, $67; \ - outb %al, $0x80; \ - movb $-119, %al; \ - outb %al, $66; \ - outb %al, $0x80; \ - movb $15, %al; \ - outb %al, $66; - - -ALIGN - .align 16 -ENTRY(wakeup_start) -wakeup_code: - wakeup_code_start = . - .code16 - -# Running in *copy* of this code, somewhere in low 1MB. - - cli - cld - # setup data segment - movw %cs, %ax - movw %ax, %ds # Make ds:0 point to wakeup_start - movw %ax, %ss - - # Data segment must be set up before we can see whether to beep. - testl $4, realmode_flags - wakeup_code - jz 1f - BEEP -1: - - # Private stack is needed for ASUS board - mov $(wakeup_stack - wakeup_code), %sp - - pushl $0 # Kill any dangerous flags - popfl - - movl real_magic - wakeup_code, %eax - cmpl $0x12345678, %eax - jne bogus_real_magic - - testl $1, realmode_flags - wakeup_code - jz 1f - lcall $0xc000,$3 - movw %cs, %ax - movw %ax, %ds # Bios might have played with that - movw %ax, %ss -1: - - testl $2, realmode_flags - wakeup_code - jz 1f - mov video_mode - wakeup_code, %ax - call mode_set -1: - - mov %ds, %ax # Find 32bit wakeup_code addr - movzx %ax, %esi # (Convert %ds:gdt to a liner ptr) - shll $4, %esi - # Fix up the vectors - addl %esi, wakeup_32_vector - wakeup_code - addl %esi, wakeup_long64_vector - wakeup_code - addl %esi, gdt_48a + 2 - wakeup_code # Fixup the gdt pointer - - lidtl %ds:idt_48a - wakeup_code - lgdtl %ds:gdt_48a - wakeup_code # load gdt with whatever is - # appropriate - - movl $1, %eax # protected mode (PE) bit - lmsw %ax # This is it! - jmp 1f -1: - - ljmpl *(wakeup_32_vector - wakeup_code) - - .balign 4 -wakeup_32_vector: - .long wakeup_32 - wakeup_code - .word __KERNEL32_CS, 0 - - .code32 -wakeup_32: -# Running in this code, but at low address; paging is not yet turned on. - - movl $__KERNEL_DS, %eax - movl %eax, %ds - - /* - * Prepare for entering 64bits mode - */ - - /* Enable PAE */ - xorl %eax, %eax - btsl $5, %eax - movl %eax, %cr4 - - /* Setup early boot stage 4 level pagetables */ - leal (wakeup_level4_pgt - wakeup_code)(%esi), %eax - movl %eax, %cr3 - - /* Check if nx is implemented */ - movl $0x80000001, %eax - cpuid - movl %edx,%edi - - /* Enable Long Mode */ - xorl %eax, %eax - btsl $_EFER_LME, %eax - - /* No Execute supported? */ - btl $20,%edi - jnc 1f - btsl $_EFER_NX, %eax - - /* Make changes effective */ -1: movl $MSR_EFER, %ecx - xorl %edx, %edx - wrmsr - - xorl %eax, %eax - btsl $31, %eax /* Enable paging and in turn activate Long Mode */ - btsl $0, %eax /* Enable protected mode */ - - /* Make changes effective */ - movl %eax, %cr0 - - /* At this point: - CR4.PAE must be 1 - CS.L must be 0 - CR3 must point to PML4 - Next instruction must be a branch - This must be on identity-mapped page - */ - /* - * At this point we're in long mode but in 32bit compatibility mode - * with EFER.LME = 1, CS.L = 0, CS.D = 1 (and in turn - * EFER.LMA = 1). Now we want to jump in 64bit mode, to do that we load - * the new gdt/idt that has __KERNEL_CS with CS.L = 1. - */ - - /* Finally jump in 64bit mode */ - ljmp *(wakeup_long64_vector - wakeup_code)(%esi) - - .balign 4 -wakeup_long64_vector: - .long wakeup_long64 - wakeup_code - .word __KERNEL_CS, 0 .code64 - - /* Hooray, we are in Long 64-bit mode (but still running in - * low memory) - */ -wakeup_long64: /* - * We must switch to a new descriptor in kernel space for the GDT - * because soon the kernel won't have access anymore to the userspace - * addresses where we're currently running on. We have to do that here - * because in 32bit we couldn't load a 64bit linear address. + * Hooray, we are in Long 64-bit mode (but still running in low memory) */ - lgdt cpu_gdt_descr - - movq saved_magic, %rax - movq $0x123456789abcdef0, %rdx - cmpq %rdx, %rax - jne bogus_64_magic +ENTRY(wakeup_long64) +wakeup_long64: + movq saved_magic, %rax + movq $0x123456789abcdef0, %rdx + cmpq %rdx, %rax + jne bogus_64_magic - nop - nop movw $__KERNEL_DS, %ax movw %ax, %ss movw %ax, %ds @@ -208,130 +35,8 @@ wakeup_long64: movq saved_rip, %rax jmp *%rax -.code32 - - .align 64 -gdta: - /* Its good to keep gdt in sync with one in trampoline.S */ - .word 0, 0, 0, 0 # dummy - /* ??? Why I need the accessed bit set in order for this to work? */ - .quad 0x00cf9b000000ffff # __KERNEL32_CS - .quad 0x00af9b000000ffff # __KERNEL_CS - .quad 0x00cf93000000ffff # __KERNEL_DS - -idt_48a: - .word 0 # idt limit = 0 - .word 0, 0 # idt base = 0L - -gdt_48a: - .word 0x800 # gdt limit=2048, - # 256 GDT entries - .long gdta - wakeup_code # gdt base (relocated in later) - -real_magic: .quad 0 -video_mode: .quad 0 -realmode_flags: .quad 0 - -.code16 -bogus_real_magic: - jmp bogus_real_magic - -.code64 bogus_64_magic: - jmp bogus_64_magic - -/* This code uses an extended set of video mode numbers. These include: - * Aliases for standard modes - * NORMAL_VGA (-1) - * EXTENDED_VGA (-2) - * ASK_VGA (-3) - * Video modes numbered by menu position -- NOT RECOMMENDED because of lack - * of compatibility when extending the table. These are between 0x00 and 0xff. - */ -#define VIDEO_FIRST_MENU 0x0000 - -/* Standard BIOS video modes (BIOS number + 0x0100) */ -#define VIDEO_FIRST_BIOS 0x0100 - -/* VESA BIOS video modes (VESA number + 0x0200) */ -#define VIDEO_FIRST_VESA 0x0200 - -/* Video7 special modes (BIOS number + 0x0900) */ -#define VIDEO_FIRST_V7 0x0900 - -# Setting of user mode (AX=mode ID) => CF=success - -# For now, we only handle VESA modes (0x0200..0x03ff). To handle other -# modes, we should probably compile in the video code from the boot -# directory. -.code16 -mode_set: - movw %ax, %bx - subb $VIDEO_FIRST_VESA>>8, %bh - cmpb $2, %bh - jb check_vesa - -setbad: - clc - ret - -check_vesa: - orw $0x4000, %bx # Use linear frame buffer - movw $0x4f02, %ax # VESA BIOS mode set call - int $0x10 - cmpw $0x004f, %ax # AL=4f if implemented - jnz setbad # AH=0 if OK - - stc - ret - -wakeup_stack_begin: # Stack grows down - -.org 0xff0 -wakeup_stack: # Just below end of page - -.org 0x1000 -ENTRY(wakeup_level4_pgt) - .quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE - .fill 510,8,0 - /* (2^48-(2*1024*1024*1024))/(2^39) = 511 */ - .quad level3_kernel_pgt - __START_KERNEL_map + _KERNPG_TABLE - -ENTRY(wakeup_end) - -## -# acpi_copy_wakeup_routine -# -# Copy the above routine to low memory. -# -# Parameters: -# %rdi: place to copy wakeup routine to -# -# Returned address is location of code in low memory (past data and stack) -# - .code64 -ENTRY(acpi_copy_wakeup_routine) - pushq %rax - pushq %rdx - - movl saved_video_mode, %edx - movl %edx, video_mode - wakeup_start (,%rdi) - movl acpi_realmode_flags, %edx - movl %edx, realmode_flags - wakeup_start (,%rdi) - movq $0x12345678, real_magic - wakeup_start (,%rdi) - movq $0x123456789abcdef0, %rdx - movq %rdx, saved_magic - - movq saved_magic, %rax - movq $0x123456789abcdef0, %rdx - cmpq %rdx, %rax - jne bogus_64_magic - - # restore the regs we used - popq %rdx - popq %rax -ENTRY(do_suspend_lowlevel_s4bios) - ret + jmp bogus_64_magic .align 2 .p2align 4,,15 @@ -414,7 +119,7 @@ do_suspend_lowlevel: jmp restore_processor_state .LFE5: .Lfe5: - .size do_suspend_lowlevel,.Lfe5-do_suspend_lowlevel + .size do_suspend_lowlevel, .Lfe5-do_suspend_lowlevel .data ALIGN diff --git a/arch/x86/kernel/acpi/wakeup_rm.S b/arch/x86/kernel/acpi/wakeup_rm.S new file mode 100644 index 00000000000..6ff3b573057 --- /dev/null +++ b/arch/x86/kernel/acpi/wakeup_rm.S @@ -0,0 +1,10 @@ +/* + * Wrapper script for the realmode binary as a transport object + * before copying to low memory. + */ + .section ".rodata","a" + .globl wakeup_code_start, wakeup_code_end +wakeup_code_start: + .incbin "arch/x86/kernel/acpi/realmode/wakeup.bin" +wakeup_code_end: + .size wakeup_code_start, .-wakeup_code_start diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 45d79ea890a..df4099dc1c6 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -11,6 +11,8 @@ #include <asm/mce.h> #include <asm/nmi.h> #include <asm/vsyscall.h> +#include <asm/cacheflush.h> +#include <asm/io.h> #define MAX_PATCH_LEN (255-1) @@ -65,7 +67,8 @@ __setup("noreplace-paravirt", setup_noreplace_paravirt); get them easily into strings. */ asm("\t.section .rodata, \"a\"\nintelnops: " GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6 - GENERIC_NOP7 GENERIC_NOP8); + GENERIC_NOP7 GENERIC_NOP8 + "\t.previous"); extern const unsigned char intelnops[]; static const unsigned char *const intel_nops[ASM_NOP_MAX+1] = { NULL, @@ -83,7 +86,8 @@ static const unsigned char *const intel_nops[ASM_NOP_MAX+1] = { #ifdef K8_NOP1 asm("\t.section .rodata, \"a\"\nk8nops: " K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6 - K8_NOP7 K8_NOP8); + K8_NOP7 K8_NOP8 + "\t.previous"); extern const unsigned char k8nops[]; static const unsigned char *const k8_nops[ASM_NOP_MAX+1] = { NULL, @@ -101,7 +105,8 @@ static const unsigned char *const k8_nops[ASM_NOP_MAX+1] = { #ifdef K7_NOP1 asm("\t.section .rodata, \"a\"\nk7nops: " K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6 - K7_NOP7 K7_NOP8); + K7_NOP7 K7_NOP8 + "\t.previous"); extern const unsigned char k7nops[]; static const unsigned char *const k7_nops[ASM_NOP_MAX+1] = { NULL, @@ -119,7 +124,8 @@ static const unsigned char *const k7_nops[ASM_NOP_MAX+1] = { #ifdef P6_NOP1 asm("\t.section .rodata, \"a\"\np6nops: " P6_NOP1 P6_NOP2 P6_NOP3 P6_NOP4 P6_NOP5 P6_NOP6 - P6_NOP7 P6_NOP8); + P6_NOP7 P6_NOP8 + "\t.previous"); extern const unsigned char p6nops[]; static const unsigned char *const p6_nops[ASM_NOP_MAX+1] = { NULL, @@ -173,7 +179,7 @@ static const unsigned char*const * find_nop_table(void) #endif /* CONFIG_X86_64 */ /* Use this to add nops to a buffer, then text_poke the whole buffer. */ -static void add_nops(void *insns, unsigned int len) +void add_nops(void *insns, unsigned int len) { const unsigned char *const *noptable = find_nop_table(); @@ -186,6 +192,7 @@ static void add_nops(void *insns, unsigned int len) len -= noplen; } } +EXPORT_SYMBOL_GPL(add_nops); extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; extern u8 *__smp_locks[], *__smp_locks_end[]; @@ -201,7 +208,7 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end) struct alt_instr *a; char insnbuf[MAX_PATCH_LEN]; - DPRINTK("%s: alt table %p -> %p\n", __FUNCTION__, start, end); + DPRINTK("%s: alt table %p -> %p\n", __func__, start, end); for (a = start; a < end; a++) { u8 *instr = a->instr; BUG_ON(a->replacementlen > a->instrlen); @@ -213,13 +220,13 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end) 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); + __func__, a->instr, instr); } #endif memcpy(insnbuf, a->replacement, a->replacementlen); add_nops(insnbuf + a->replacementlen, a->instrlen - a->replacementlen); - text_poke(instr, insnbuf, a->instrlen); + text_poke_early(instr, insnbuf, a->instrlen); } } @@ -280,7 +287,6 @@ void alternatives_smp_module_add(struct module *mod, char *name, void *text, void *text_end) { struct smp_alt_module *smp; - unsigned long flags; if (noreplace_smp) return; @@ -303,42 +309,40 @@ void alternatives_smp_module_add(struct module *mod, char *name, smp->text = text; smp->text_end = text_end; DPRINTK("%s: locks %p -> %p, text %p -> %p, name %s\n", - __FUNCTION__, smp->locks, smp->locks_end, + __func__, smp->locks, smp->locks_end, smp->text, smp->text_end, smp->name); - spin_lock_irqsave(&smp_alt, flags); + spin_lock(&smp_alt); list_add_tail(&smp->next, &smp_alt_modules); if (boot_cpu_has(X86_FEATURE_UP)) alternatives_smp_unlock(smp->locks, smp->locks_end, smp->text, smp->text_end); - spin_unlock_irqrestore(&smp_alt, flags); + spin_unlock(&smp_alt); } void alternatives_smp_module_del(struct module *mod) { struct smp_alt_module *item; - unsigned long flags; if (smp_alt_once || noreplace_smp) return; - spin_lock_irqsave(&smp_alt, flags); + spin_lock(&smp_alt); list_for_each_entry(item, &smp_alt_modules, next) { if (mod != item->mod) continue; list_del(&item->next); - spin_unlock_irqrestore(&smp_alt, flags); - DPRINTK("%s: %s\n", __FUNCTION__, item->name); + spin_unlock(&smp_alt); + DPRINTK("%s: %s\n", __func__, item->name); kfree(item); return; } - spin_unlock_irqrestore(&smp_alt, flags); + spin_unlock(&smp_alt); } void alternatives_smp_switch(int smp) { struct smp_alt_module *mod; - unsigned long flags; #ifdef CONFIG_LOCKDEP /* @@ -355,7 +359,7 @@ void alternatives_smp_switch(int smp) return; BUG_ON(!smp && (num_online_cpus() > 1)); - spin_lock_irqsave(&smp_alt, flags); + spin_lock(&smp_alt); /* * Avoid unnecessary switches because it forces JIT based VMs to @@ -379,7 +383,7 @@ void alternatives_smp_switch(int smp) mod->text, mod->text_end); } smp_mode = smp; - spin_unlock_irqrestore(&smp_alt, flags); + spin_unlock(&smp_alt); } #endif @@ -407,7 +411,7 @@ void apply_paravirt(struct paravirt_patch_site *start, /* Pad the rest with nops */ add_nops(insnbuf + used, p->len - used); - text_poke(p->instr, insnbuf, p->len); + text_poke_early(p->instr, insnbuf, p->len); } } extern struct paravirt_patch_site __start_parainstructions[], @@ -416,8 +420,6 @@ extern struct paravirt_patch_site __start_parainstructions[], void __init alternative_instructions(void) { - unsigned long flags; - /* The patching is not fully atomic, so try to avoid local interruptions that might execute the to be patched code. Other CPUs are not running. */ @@ -426,7 +428,6 @@ void __init alternative_instructions(void) stop_mce(); #endif - local_irq_save(flags); apply_alternatives(__alt_instructions, __alt_instructions_end); /* switch to patch-once-at-boottime-only mode and free the @@ -458,7 +459,6 @@ void __init alternative_instructions(void) } #endif apply_paravirt(__parainstructions, __parainstructions_end); - local_irq_restore(flags); if (smp_alt_once) free_init_pages("SMP alternatives", @@ -471,18 +471,71 @@ void __init alternative_instructions(void) #endif } -/* - * Warning: +/** + * text_poke_early - Update instructions on a live kernel at boot time + * @addr: address to modify + * @opcode: source of the copy + * @len: length to copy + * * When you use this code to patch more than one byte of an instruction * you need to make sure that other CPUs cannot execute this code in parallel. - * Also no thread must be currently preempted in the middle of these instructions. - * And on the local CPU you need to be protected again NMI or MCE handlers - * seeing an inconsistent instruction while you patch. + * Also no thread must be currently preempted in the middle of these + * instructions. And on the local CPU you need to be protected again NMI or MCE + * handlers seeing an inconsistent instruction while you patch. */ -void __kprobes text_poke(void *addr, unsigned char *opcode, int len) +void *text_poke_early(void *addr, const void *opcode, size_t len) { + unsigned long flags; + local_irq_save(flags); memcpy(addr, opcode, len); + local_irq_restore(flags); + sync_core(); + /* Could also do a CLFLUSH here to speed up CPU recovery; but + that causes hangs on some VIA CPUs. */ + return addr; +} + +/** + * text_poke - Update instructions on a live kernel + * @addr: address to modify + * @opcode: source of the copy + * @len: length to copy + * + * Only atomic text poke/set should be allowed when not doing early patching. + * It means the size must be writable atomically and the address must be aligned + * in a way that permits an atomic write. It also makes sure we fit on a single + * page. + */ +void *__kprobes text_poke(void *addr, const void *opcode, size_t len) +{ + unsigned long flags; + char *vaddr; + int nr_pages = 2; + + BUG_ON(len > sizeof(long)); + BUG_ON((((long)addr + len - 1) & ~(sizeof(long) - 1)) + - ((long)addr & ~(sizeof(long) - 1))); + if (kernel_text_address((unsigned long)addr)) { + struct page *pages[2] = { virt_to_page(addr), + virt_to_page(addr + PAGE_SIZE) }; + if (!pages[1]) + nr_pages = 1; + vaddr = vmap(pages, nr_pages, VM_MAP, PAGE_KERNEL); + BUG_ON(!vaddr); + local_irq_save(flags); + memcpy(&vaddr[(unsigned long)addr & ~PAGE_MASK], opcode, len); + local_irq_restore(flags); + vunmap(vaddr); + } else { + /* + * modules are in vmalloc'ed memory, always writable. + */ + local_irq_save(flags); + memcpy(addr, opcode, len); + local_irq_restore(flags); + } sync_core(); /* Could also do a CLFLUSH here to speed up CPU recovery; but that causes hangs on some VIA CPUs. */ + return addr; } diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 00df126169b..479926d9e00 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -27,11 +27,11 @@ #include <asm/k8.h> int gart_iommu_aperture; -int gart_iommu_aperture_disabled __initdata = 0; -int gart_iommu_aperture_allowed __initdata = 0; +int gart_iommu_aperture_disabled __initdata; +int gart_iommu_aperture_allowed __initdata; int fallback_aper_order __initdata = 1; /* 64MB */ -int fallback_aper_force __initdata = 0; +int fallback_aper_force __initdata; int fix_aperture __initdata = 1; diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c index 35a568ea840..687208190b0 100644 --- a/arch/x86/kernel/apic_32.c +++ b/arch/x86/kernel/apic_32.c @@ -50,6 +50,11 @@ # error SPURIOUS_APIC_VECTOR definition error #endif +unsigned long mp_lapic_addr; + +DEFINE_PER_CPU(u16, x86_bios_cpu_apicid) = BAD_APICID; +EXPORT_PER_CPU_SYMBOL(x86_bios_cpu_apicid); + /* * Knob to control our willingness to enable the local APIC. * @@ -621,6 +626,35 @@ int setup_profiling_timer(unsigned int multiplier) } /* + * Setup extended LVT, AMD specific (K8, family 10h) + * + * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and + * MCE interrupts are supported. Thus MCE offset must be set to 0. + */ + +#define APIC_EILVT_LVTOFF_MCE 0 +#define APIC_EILVT_LVTOFF_IBS 1 + +static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask) +{ + unsigned long reg = (lvt_off << 4) + APIC_EILVT0; + unsigned int v = (mask << 16) | (msg_type << 8) | vector; + apic_write(reg, v); +} + +u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask) +{ + setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask); + return APIC_EILVT_LVTOFF_MCE; +} + +u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask) +{ + setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask); + return APIC_EILVT_LVTOFF_IBS; +} + +/* * Local APIC start and shutdown */ @@ -868,12 +902,50 @@ void __init init_bsp_APIC(void) apic_write_around(APIC_LVT1, value); } +void __cpuinit lapic_setup_esr(void) +{ + unsigned long oldvalue, value, maxlvt; + if (lapic_is_integrated() && !esr_disable) { + /* !82489DX */ + maxlvt = lapic_get_maxlvt(); + if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ + apic_write(APIC_ESR, 0); + oldvalue = apic_read(APIC_ESR); + + /* enables sending errors */ + value = ERROR_APIC_VECTOR; + apic_write_around(APIC_LVTERR, value); + /* + * spec says clear errors after enabling vector. + */ + if (maxlvt > 3) + apic_write(APIC_ESR, 0); + value = apic_read(APIC_ESR); + if (value != oldvalue) + apic_printk(APIC_VERBOSE, "ESR value before enabling " + "vector: 0x%08lx after: 0x%08lx\n", + oldvalue, value); + } else { + if (esr_disable) + /* + * Something untraceable is creating bad interrupts on + * secondary quads ... for the moment, just leave the + * ESR disabled - we can't do anything useful with the + * errors anyway - mbligh + */ + printk(KERN_INFO "Leaving ESR disabled.\n"); + else + printk(KERN_INFO "No ESR for 82489DX.\n"); + } +} + + /** * setup_local_APIC - setup the local APIC */ void __cpuinit setup_local_APIC(void) { - unsigned long oldvalue, value, maxlvt, integrated; + unsigned long value, integrated; int i, j; /* Pound the ESR really hard over the head with a big hammer - mbligh */ @@ -997,40 +1069,13 @@ void __cpuinit setup_local_APIC(void) if (!integrated) /* 82489DX */ value |= APIC_LVT_LEVEL_TRIGGER; apic_write_around(APIC_LVT1, value); +} - if (integrated && !esr_disable) { - /* !82489DX */ - maxlvt = lapic_get_maxlvt(); - if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ - apic_write(APIC_ESR, 0); - oldvalue = apic_read(APIC_ESR); - - /* enables sending errors */ - value = ERROR_APIC_VECTOR; - apic_write_around(APIC_LVTERR, value); - /* - * spec says clear errors after enabling vector. - */ - if (maxlvt > 3) - apic_write(APIC_ESR, 0); - value = apic_read(APIC_ESR); - if (value != oldvalue) - apic_printk(APIC_VERBOSE, "ESR value before enabling " - "vector: 0x%08lx after: 0x%08lx\n", - oldvalue, value); - } else { - if (esr_disable) - /* - * Something untraceable is creating bad interrupts on - * secondary quads ... for the moment, just leave the - * ESR disabled - we can't do anything useful with the - * errors anyway - mbligh - */ - printk(KERN_INFO "Leaving ESR disabled.\n"); - else - printk(KERN_INFO "No ESR for 82489DX.\n"); - } +void __cpuinit end_local_APIC_setup(void) +{ + unsigned long value; + lapic_setup_esr(); /* Disable the local apic timer */ value = apic_read(APIC_LVTT); value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); @@ -1147,7 +1192,7 @@ void __init init_apic_mappings(void) * default configuration (or the MP table is broken). */ if (boot_cpu_physical_apicid == -1U) - boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID)); + boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); #ifdef CONFIG_X86_IO_APIC { @@ -1185,6 +1230,9 @@ fake_ioapic_page: * This initializes the IO-APIC and APIC hardware if this is * a UP kernel. */ + +int apic_version[MAX_APICS]; + int __init APIC_init_uniprocessor(void) { if (enable_local_apic < 0) @@ -1214,12 +1262,13 @@ int __init APIC_init_uniprocessor(void) * might be zero if read from MP tables. Get it from LAPIC. */ #ifdef CONFIG_CRASH_DUMP - boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID)); + boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); #endif phys_cpu_present_map = physid_mask_of_physid(boot_cpu_physical_apicid); setup_local_APIC(); + end_local_APIC_setup(); #ifdef CONFIG_X86_IO_APIC if (smp_found_config) if (!skip_ioapic_setup && nr_ioapics) @@ -1288,6 +1337,29 @@ void smp_error_interrupt(struct pt_regs *regs) irq_exit(); } +#ifdef CONFIG_SMP +void __init smp_intr_init(void) +{ + /* + * IRQ0 must be given a fixed assignment and initialized, + * because it's used before the IO-APIC is set up. + */ + set_intr_gate(FIRST_DEVICE_VECTOR, interrupt[0]); + + /* + * The reschedule interrupt is a CPU-to-CPU reschedule-helper + * IPI, driven by wakeup. + */ + set_intr_gate(RESCHEDULE_VECTOR, reschedule_interrupt); + + /* IPI for invalidation */ + set_intr_gate(INVALIDATE_TLB_VECTOR, invalidate_interrupt); + + /* IPI for generic function call */ + set_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt); +} +#endif + /* * Initialize APIC interrupts */ @@ -1394,6 +1466,88 @@ void disconnect_bsp_APIC(int virt_wire_setup) } } +unsigned int __cpuinitdata maxcpus = NR_CPUS; + +void __cpuinit generic_processor_info(int apicid, int version) +{ + int cpu; + cpumask_t tmp_map; + physid_mask_t phys_cpu; + + /* + * Validate version + */ + if (version == 0x0) { + printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! " + "fixing up to 0x10. (tell your hw vendor)\n", + version); + version = 0x10; + } + apic_version[apicid] = version; + + phys_cpu = apicid_to_cpu_present(apicid); + physids_or(phys_cpu_present_map, phys_cpu_present_map, phys_cpu); + + if (num_processors >= NR_CPUS) { + printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached." + " Processor ignored.\n", NR_CPUS); + return; + } + + if (num_processors >= maxcpus) { + printk(KERN_WARNING "WARNING: maxcpus limit of %i reached." + " Processor ignored.\n", maxcpus); + return; + } + + num_processors++; + cpus_complement(tmp_map, cpu_present_map); + cpu = first_cpu(tmp_map); + + if (apicid == boot_cpu_physical_apicid) + /* + * x86_bios_cpu_apicid is required to have processors listed + * in same order as logical cpu numbers. Hence the first + * entry is BSP, and so on. + */ + cpu = 0; + + /* + * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y + * but we need to work other dependencies like SMP_SUSPEND etc + * before this can be done without some confusion. + * if (CPU_HOTPLUG_ENABLED || num_processors > 8) + * - Ashok Raj <ashok.raj@intel.com> + */ + if (num_processors > 8) { + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_INTEL: + if (!APIC_XAPIC(version)) { + def_to_bigsmp = 0; + break; + } + /* If P4 and above fall through */ + case X86_VENDOR_AMD: + def_to_bigsmp = 1; + } + } +#ifdef CONFIG_SMP + /* are we being called early in kernel startup? */ + if (x86_cpu_to_apicid_early_ptr) { + u16 *cpu_to_apicid = x86_cpu_to_apicid_early_ptr; + u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr; + + cpu_to_apicid[cpu] = apicid; + bios_cpu_apicid[cpu] = apicid; + } else { + per_cpu(x86_cpu_to_apicid, cpu) = apicid; + per_cpu(x86_bios_cpu_apicid, cpu) = apicid; + } +#endif + cpu_set(cpu, cpu_possible_map); + cpu_set(cpu, cpu_present_map); +} + /* * Power management */ diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c index d8d03e09dea..9e8e5c050c5 100644 --- a/arch/x86/kernel/apic_64.c +++ b/arch/x86/kernel/apic_64.c @@ -34,13 +34,15 @@ #include <asm/mpspec.h> #include <asm/hpet.h> #include <asm/pgalloc.h> -#include <asm/mach_apic.h> #include <asm/nmi.h> #include <asm/idle.h> #include <asm/proto.h> #include <asm/timex.h> #include <asm/apic.h> +#include <mach_ipi.h> +#include <mach_apic.h> + int disable_apic_timer __cpuinitdata; static int apic_calibrate_pmtmr __initdata; int disable_apic; @@ -83,6 +85,12 @@ static DEFINE_PER_CPU(struct clock_event_device, lapic_events); static unsigned long apic_phys; +unsigned long mp_lapic_addr; + +DEFINE_PER_CPU(u16, x86_bios_cpu_apicid) = BAD_APICID; +EXPORT_PER_CPU_SYMBOL(x86_bios_cpu_apicid); + +unsigned int __cpuinitdata maxcpus = NR_CPUS; /* * Get the LAPIC version */ @@ -431,7 +439,8 @@ void __cpuinit check_boot_apic_timer_broadcast(void) lapic_clockevent.features |= CLOCK_EVT_FEAT_DUMMY; local_irq_enable(); - clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_FORCE, &boot_cpu_id); + clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_FORCE, + &boot_cpu_physical_apicid); local_irq_disable(); } @@ -640,10 +649,10 @@ int __init verify_local_APIC(void) /* * The ID register is read/write in a real APIC. */ - reg0 = apic_read(APIC_ID); + reg0 = read_apic_id(); apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg0); apic_write(APIC_ID, reg0 ^ APIC_ID_MASK); - reg1 = apic_read(APIC_ID); + reg1 = read_apic_id(); apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg1); apic_write(APIC_ID, reg0); if (reg1 != (reg0 ^ APIC_ID_MASK)) @@ -728,6 +737,7 @@ void __cpuinit setup_local_APIC(void) unsigned int value; int i, j; + preempt_disable(); value = apic_read(APIC_LVR); BUILD_BUG_ON((SPURIOUS_APIC_VECTOR & 0x0f) != 0x0f); @@ -821,6 +831,7 @@ void __cpuinit setup_local_APIC(void) else value = APIC_DM_NMI | APIC_LVT_MASKED; apic_write(APIC_LVT1, value); + preempt_enable(); } void __cpuinit lapic_setup_esr(void) @@ -857,10 +868,34 @@ static int __init detect_init_APIC(void) } mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; - boot_cpu_id = 0; + boot_cpu_physical_apicid = 0; return 0; } +void __init early_init_lapic_mapping(void) +{ + unsigned long apic_phys; + + /* + * If no local APIC can be found then go out + * : it means there is no mpatable and MADT + */ + if (!smp_found_config) + return; + + apic_phys = mp_lapic_addr; + + set_fixmap_nocache(FIX_APIC_BASE, apic_phys); + apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", + APIC_BASE, apic_phys); + + /* + * Fetch the APIC ID of the BSP in case we have a + * default configuration (or the MP table is broken). + */ + boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); +} + /** * init_apic_mappings - initialize APIC mappings */ @@ -881,16 +916,11 @@ void __init init_apic_mappings(void) apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", APIC_BASE, apic_phys); - /* Put local APIC into the resource map. */ - lapic_resource.start = apic_phys; - lapic_resource.end = lapic_resource.start + PAGE_SIZE - 1; - insert_resource(&iomem_resource, &lapic_resource); - /* * Fetch the APIC ID of the BSP in case we have a * default configuration (or the MP table is broken). */ - boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID)); + boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); } /* @@ -911,8 +941,8 @@ int __init APIC_init_uniprocessor(void) verify_local_APIC(); - phys_cpu_present_map = physid_mask_of_physid(boot_cpu_id); - apic_write(APIC_ID, SET_APIC_ID(boot_cpu_id)); + phys_cpu_present_map = physid_mask_of_physid(boot_cpu_physical_apicid); + apic_write(APIC_ID, SET_APIC_ID(boot_cpu_physical_apicid)); setup_local_APIC(); @@ -1029,6 +1059,52 @@ void disconnect_bsp_APIC(int virt_wire_setup) apic_write(APIC_LVT1, value); } +void __cpuinit generic_processor_info(int apicid, int version) +{ + int cpu; + cpumask_t tmp_map; + + if (num_processors >= NR_CPUS) { + printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached." + " Processor ignored.\n", NR_CPUS); + return; + } + + if (num_processors >= maxcpus) { + printk(KERN_WARNING "WARNING: maxcpus limit of %i reached." + " Processor ignored.\n", maxcpus); + return; + } + + num_processors++; + cpus_complement(tmp_map, cpu_present_map); + cpu = first_cpu(tmp_map); + + physid_set(apicid, phys_cpu_present_map); + if (apicid == boot_cpu_physical_apicid) { + /* + * x86_bios_cpu_apicid is required to have processors listed + * in same order as logical cpu numbers. Hence the first + * entry is BSP, and so on. + */ + cpu = 0; + } + /* are we being called early in kernel startup? */ + if (x86_cpu_to_apicid_early_ptr) { + u16 *cpu_to_apicid = x86_cpu_to_apicid_early_ptr; + u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr; + + cpu_to_apicid[cpu] = apicid; + bios_cpu_apicid[cpu] = apicid; + } else { + per_cpu(x86_cpu_to_apicid, cpu) = apicid; + per_cpu(x86_bios_cpu_apicid, cpu) = apicid; + } + + cpu_set(cpu, cpu_possible_map); + cpu_set(cpu, cpu_present_map); +} + /* * Power management */ @@ -1065,7 +1141,7 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state) maxlvt = lapic_get_maxlvt(); - apic_pm_state.apic_id = apic_read(APIC_ID); + apic_pm_state.apic_id = read_apic_id(); apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI); apic_pm_state.apic_ldr = apic_read(APIC_LDR); apic_pm_state.apic_dfr = apic_read(APIC_DFR); @@ -1180,9 +1256,19 @@ __cpuinit int apic_is_clustered_box(void) { int i, clusters, zeros; unsigned id; - u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr; + u16 *bios_cpu_apicid; DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS); + /* + * there is not this kind of box with AMD CPU yet. + * Some AMD box with quadcore cpu and 8 sockets apicid + * will be [4, 0x23] or [8, 0x27] could be thought to + * vsmp box still need checking... + */ + if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && !is_vsmp_box()) + return 0; + + bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr; bitmap_zero(clustermap, NUM_APIC_CLUSTERS); for (i = 0; i < NR_CPUS; i++) { @@ -1219,6 +1305,12 @@ __cpuinit int apic_is_clustered_box(void) ++zeros; } + /* ScaleMP vSMPowered boxes have one cluster per board and TSCs are + * not guaranteed to be synced between boards + */ + if (is_vsmp_box() && clusters > 1) + return 1; + /* * If clusters > 2, then should be multi-chassis. * May have to revisit this when multi-core + hyperthreaded CPUs come @@ -1290,3 +1382,21 @@ static __init int setup_apicpmtimer(char *s) } __setup("apicpmtimer", setup_apicpmtimer); +static int __init lapic_insert_resource(void) +{ + if (!apic_phys) + return -1; + + /* Put local APIC into the resource map. */ + lapic_resource.start = apic_phys; + lapic_resource.end = lapic_resource.start + PAGE_SIZE - 1; + insert_resource(&iomem_resource, &lapic_resource); + + return 0; +} + +/* + * need call insert after e820_reserve_resources() + * that is using request_resource + */ +late_initcall(lapic_insert_resource); diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index d4438ef296d..f0030a0999c 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -2217,7 +2217,6 @@ static struct dmi_system_id __initdata apm_dmi_table[] = { */ static int __init apm_init(void) { - struct proc_dir_entry *apm_proc; struct desc_struct *gdt; int err; @@ -2322,9 +2321,7 @@ static int __init apm_init(void) set_base(gdt[APM_DS >> 3], __va((unsigned long)apm_info.bios.dseg << 4)); - apm_proc = create_proc_entry("apm", 0, NULL); - if (apm_proc) - apm_proc->proc_fops = &apm_file_ops; + proc_create("apm", 0, NULL, &apm_file_ops); kapmd_task = kthread_create(apm, NULL, "kapmd"); if (IS_ERR(kapmd_task)) { diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c index 8ea040124f7..670c3c31128 100644 --- a/arch/x86/kernel/asm-offsets_32.c +++ b/arch/x86/kernel/asm-offsets_32.c @@ -10,7 +10,7 @@ #include <linux/personality.h> #include <linux/suspend.h> #include <asm/ucontext.h> -#include "sigframe_32.h" +#include "sigframe.h" #include <asm/pgtable.h> #include <asm/fixmap.h> #include <asm/processor.h> diff --git a/arch/x86/kernel/bugs_64.c b/arch/x86/kernel/bugs_64.c index 8f520f93ffd..9a3ed0649d4 100644 --- a/arch/x86/kernel/bugs_64.c +++ b/arch/x86/kernel/bugs_64.c @@ -9,13 +9,25 @@ #include <asm/bugs.h> #include <asm/processor.h> #include <asm/mtrr.h> +#include <asm/cacheflush.h> void __init check_bugs(void) { - identify_cpu(&boot_cpu_data); + identify_boot_cpu(); #if !defined(CONFIG_SMP) printk("CPU: "); print_cpu_info(&boot_cpu_data); #endif alternative_instructions(); + + /* + * Make sure the first 2MB area is not mapped by huge pages + * There are typically fixed size MTRRs in there and overlapping + * MTRRs into large pages causes slow downs. + * + * Right now we don't do that with gbpages because there seems + * very little benefit for that case. + */ + if (!direct_gbpages) + set_memory_4k((unsigned long)__va(0), 1); } diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index a0c4d7c5dbd..ee7c45235e5 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -3,9 +3,9 @@ # obj-y := intel_cacheinfo.o addon_cpuid_features.o -obj-y += feature_names.o +obj-y += proc.o feature_names.o -obj-$(CONFIG_X86_32) += common.o proc.o bugs.o +obj-$(CONFIG_X86_32) += common.o bugs.o obj-$(CONFIG_X86_32) += amd.o obj-$(CONFIG_X86_32) += cyrix.o obj-$(CONFIG_X86_32) += centaur.o diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 693e353999c..0173065dc3b 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -4,8 +4,8 @@ #include <asm/io.h> #include <asm/processor.h> #include <asm/apic.h> -#include <asm/mach_apic.h> +#include <mach_apic.h> #include "cpu.h" /* @@ -20,7 +20,7 @@ * the chip setting when fixing the bug but they also tweaked some * performance at the same time.. */ - + extern void vide(void); __asm__(".align 4\nvide: ret"); @@ -63,12 +63,12 @@ static __cpuinit int amd_apic_timer_broken(void) int force_mwait __cpuinitdata; -void __cpuinit early_init_amd(struct cpuinfo_x86 *c) +static void __cpuinit early_init_amd(struct cpuinfo_x86 *c) { if (cpuid_eax(0x80000000) >= 0x80000007) { c->x86_power = cpuid_edx(0x80000007); if (c->x86_power & (1<<8)) - set_bit(X86_FEATURE_CONSTANT_TSC, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); } } @@ -81,7 +81,8 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) #ifdef CONFIG_SMP unsigned long long value; - /* Disable TLB flush filter by setting HWCR.FFDIS on K8 + /* + * Disable TLB flush filter by setting HWCR.FFDIS on K8 * bit 6 of msr C001_0015 * * Errata 63 for SH-B3 steppings @@ -102,15 +103,16 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) * no bus pipeline) */ - /* Bit 31 in normal CPUID used for nonstandard 3DNow ID; - 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */ - clear_bit(0*32+31, c->x86_capability); - + /* + * Bit 31 in normal CPUID used for nonstandard 3DNow ID; + * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway + */ + clear_cpu_cap(c, 0*32+31); + r = get_model_name(c); - switch(c->x86) - { - case 4: + switch (c->x86) { + case 4: /* * General Systems BIOSen alias the cpu frequency registers * of the Elan at 0x000df000. Unfortuantly, one of the Linux @@ -120,61 +122,60 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) #define CBAR (0xfffc) /* Configuration Base Address (32-bit) */ #define CBAR_ENB (0x80000000) #define CBAR_KEY (0X000000CB) - if (c->x86_model==9 || c->x86_model == 10) { + if (c->x86_model == 9 || c->x86_model == 10) { if (inl (CBAR) & CBAR_ENB) outl (0 | CBAR_KEY, CBAR); } break; - case 5: - if( c->x86_model < 6 ) - { + case 5: + if (c->x86_model < 6) { /* Based on AMD doc 20734R - June 2000 */ - if ( c->x86_model == 0 ) { - clear_bit(X86_FEATURE_APIC, c->x86_capability); - set_bit(X86_FEATURE_PGE, c->x86_capability); + if (c->x86_model == 0) { + clear_cpu_cap(c, X86_FEATURE_APIC); + set_cpu_cap(c, X86_FEATURE_PGE); } break; } - - if ( c->x86_model == 6 && c->x86_mask == 1 ) { + + if (c->x86_model == 6 && c->x86_mask == 1) { const int K6_BUG_LOOP = 1000000; int n; void (*f_vide)(void); unsigned long d, d2; - + printk(KERN_INFO "AMD K6 stepping B detected - "); - + /* - * It looks like AMD fixed the 2.6.2 bug and improved indirect + * It looks like AMD fixed the 2.6.2 bug and improved indirect * calls at the same time. */ n = K6_BUG_LOOP; f_vide = vide; rdtscl(d); - while (n--) + while (n--) f_vide(); rdtscl(d2); d = d2-d; - if (d > 20*K6_BUG_LOOP) + if (d > 20*K6_BUG_LOOP) printk("system stability may be impaired when more than 32 MB are used.\n"); - else + else printk("probably OK (after B9730xxxx).\n"); printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n"); } /* K6 with old style WHCR */ if (c->x86_model < 8 || - (c->x86_model== 8 && c->x86_mask < 8)) { + (c->x86_model == 8 && c->x86_mask < 8)) { /* We can only write allocate on the low 508Mb */ - if(mbytes>508) - mbytes=508; + if (mbytes > 508) + mbytes = 508; rdmsr(MSR_K6_WHCR, l, h); - if ((l&0x0000FFFF)==0) { + if ((l&0x0000FFFF) == 0) { unsigned long flags; - l=(1<<0)|((mbytes/4)<<1); + l = (1<<0)|((mbytes/4)<<1); local_irq_save(flags); wbinvd(); wrmsr(MSR_K6_WHCR, l, h); @@ -185,17 +186,17 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) break; } - if ((c->x86_model == 8 && c->x86_mask >7) || + if ((c->x86_model == 8 && c->x86_mask > 7) || c->x86_model == 9 || c->x86_model == 13) { /* The more serious chips .. */ - if(mbytes>4092) - mbytes=4092; + if (mbytes > 4092) + mbytes = 4092; rdmsr(MSR_K6_WHCR, l, h); - if ((l&0xFFFF0000)==0) { + if ((l&0xFFFF0000) == 0) { unsigned long flags; - l=((mbytes>>2)<<22)|(1<<16); + l = ((mbytes>>2)<<22)|(1<<16); local_irq_save(flags); wbinvd(); wrmsr(MSR_K6_WHCR, l, h); @@ -207,7 +208,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) /* Set MTRR capability flag if appropriate */ if (c->x86_model == 13 || c->x86_model == 9 || (c->x86_model == 8 && c->x86_mask >= 8)) - set_bit(X86_FEATURE_K6_MTRR, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_K6_MTRR); break; } @@ -217,10 +218,11 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) break; } break; - case 6: /* An Athlon/Duron */ - - /* Bit 15 of Athlon specific MSR 15, needs to be 0 - * to enable SSE on Palomino/Morgan/Barton CPU's. + case 6: /* An Athlon/Duron */ + + /* + * Bit 15 of Athlon specific MSR 15, needs to be 0 + * to enable SSE on Palomino/Morgan/Barton CPU's. * If the BIOS didn't enable it already, enable it here. */ if (c->x86_model >= 6 && c->x86_model <= 10) { @@ -229,15 +231,16 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) rdmsr(MSR_K7_HWCR, l, h); l &= ~0x00008000; wrmsr(MSR_K7_HWCR, l, h); - set_bit(X86_FEATURE_XMM, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_XMM); } } - /* It's been determined by AMD that Athlons since model 8 stepping 1 + /* + * It's been determined by AMD that Athlons since model 8 stepping 1 * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx * As per AMD technical note 27212 0.2 */ - if ((c->x86_model == 8 && c->x86_mask>=1) || (c->x86_model > 8)) { + if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) { rdmsr(MSR_K7_CLK_CTL, l, h); if ((l & 0xfff00000) != 0x20000000) { printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l, @@ -253,20 +256,19 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) /* Use K8 tuning for Fam10h and Fam11h */ case 0x10: case 0x11: - set_bit(X86_FEATURE_K8, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_K8); break; case 6: - set_bit(X86_FEATURE_K7, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_K7); break; } if (c->x86 >= 6) - set_bit(X86_FEATURE_FXSAVE_LEAK, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_FXSAVE_LEAK); display_cacheinfo(c); - if (cpuid_eax(0x80000000) >= 0x80000008) { + if (cpuid_eax(0x80000000) >= 0x80000008) c->x86_max_cores = (cpuid_ecx(0x80000008) & 0xff) + 1; - } #ifdef CONFIG_X86_HT /* @@ -302,20 +304,20 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) /* K6s reports MCEs but don't actually have all the MSRs */ if (c->x86 < 6) - clear_bit(X86_FEATURE_MCE, c->x86_capability); + clear_cpu_cap(c, X86_FEATURE_MCE); if (cpu_has_xmm2) - set_bit(X86_FEATURE_MFENCE_RDTSC, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC); } -static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 * c, unsigned int size) +static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int size) { /* AMD errata T13 (order #21922) */ if ((c->x86 == 6)) { if (c->x86_model == 3 && c->x86_mask == 0) /* Duron Rev A0 */ size = 64; if (c->x86_model == 4 && - (c->x86_mask==0 || c->x86_mask==1)) /* Tbird rev A1/A2 */ + (c->x86_mask == 0 || c->x86_mask == 1)) /* Tbird rev A1/A2 */ size = 256; } return size; @@ -323,19 +325,20 @@ static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 * c, unsigned in static struct cpu_dev amd_cpu_dev __cpuinitdata = { .c_vendor = "AMD", - .c_ident = { "AuthenticAMD" }, + .c_ident = { "AuthenticAMD" }, .c_models = { { .vendor = X86_VENDOR_AMD, .family = 4, .model_names = { [3] = "486 DX/2", [7] = "486 DX/2-WB", - [8] = "486 DX/4", - [9] = "486 DX/4-WB", + [8] = "486 DX/4", + [9] = "486 DX/4-WB", [14] = "Am5x86-WT", - [15] = "Am5x86-WB" + [15] = "Am5x86-WB" } }, }, + .c_early_init = early_init_amd, .c_init = init_amd, .c_size_cache = amd_size_cache, }; @@ -345,3 +348,5 @@ int __init amd_init_cpu(void) cpu_devs[X86_VENDOR_AMD] = &amd_cpu_dev; return 0; } + +cpu_vendor_dev_register(X86_VENDOR_AMD, &amd_cpu_dev); diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 027e5c003b1..170d2f5523b 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -143,14 +143,6 @@ static void __init check_config(void) #endif /* - * If we configured ourselves for a TSC, we'd better have one! - */ -#ifdef CONFIG_X86_TSC - if (!cpu_has_tsc) - panic("Kernel compiled for Pentium+, requires TSC feature!"); -#endif - -/* * If we were told we had a good local APIC, check for buggy Pentia, * i.e. all B steppings and the C2 stepping of P54C when using their * integrated APIC (see 11AP erratum in "Pentium Processor diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c index 9681fa15ddf..e0f45edd6a5 100644 --- a/arch/x86/kernel/cpu/centaur.c +++ b/arch/x86/kernel/cpu/centaur.c @@ -1,31 +1,34 @@ #include <linux/kernel.h> #include <linux/init.h> #include <linux/bitops.h> + #include <asm/processor.h> #include <asm/msr.h> #include <asm/e820.h> #include <asm/mtrr.h> + #include "cpu.h" #ifdef CONFIG_X86_OOSTORE static u32 __cpuinit power2(u32 x) { - u32 s=1; - while(s<=x) - s<<=1; - return s>>=1; + u32 s = 1; + + while (s <= x) + s <<= 1; + + return s >>= 1; } /* - * Set up an actual MCR + * Set up an actual MCR */ - static void __cpuinit centaur_mcr_insert(int reg, u32 base, u32 size, int key) { u32 lo, hi; - + hi = base & ~0xFFF; lo = ~(size-1); /* Size is a power of 2 so this makes a mask */ lo &= ~0xFFF; /* Remove the ctrl value bits */ @@ -35,30 +38,28 @@ static void __cpuinit centaur_mcr_insert(int reg, u32 base, u32 size, int key) } /* - * Figure what we can cover with MCR's + * Figure what we can cover with MCR's * - * Shortcut: We know you can't put 4Gig of RAM on a winchip + * Shortcut: We know you can't put 4Gig of RAM on a winchip */ - -static u32 __cpuinit ramtop(void) /* 16388 */ +static u32 __cpuinit ramtop(void) { - int i; - u32 top = 0; u32 clip = 0xFFFFFFFFUL; - + u32 top = 0; + int i; + for (i = 0; i < e820.nr_map; i++) { unsigned long start, end; if (e820.map[i].addr > 0xFFFFFFFFUL) continue; /* - * Don't MCR over reserved space. Ignore the ISA hole - * we frob around that catastrophe already + * Don't MCR over reserved space. Ignore the ISA hole + * we frob around that catastrophe already */ - - if (e820.map[i].type == E820_RESERVED) - { - if(e820.map[i].addr >= 0x100000UL && e820.map[i].addr < clip) + if (e820.map[i].type == E820_RESERVED) { + if (e820.map[i].addr >= 0x100000UL && + e820.map[i].addr < clip) clip = e820.map[i].addr; continue; } @@ -69,28 +70,27 @@ static u32 __cpuinit ramtop(void) /* 16388 */ if (end > top) top = end; } - /* Everything below 'top' should be RAM except for the ISA hole. - Because of the limited MCR's we want to map NV/ACPI into our - MCR range for gunk in RAM - - Clip might cause us to MCR insufficient RAM but that is an - acceptable failure mode and should only bite obscure boxes with - a VESA hole at 15Mb - - The second case Clip sometimes kicks in is when the EBDA is marked - as reserved. Again we fail safe with reasonable results - */ - - if(top>clip) - top=clip; - + /* + * Everything below 'top' should be RAM except for the ISA hole. + * Because of the limited MCR's we want to map NV/ACPI into our + * MCR range for gunk in RAM + * + * Clip might cause us to MCR insufficient RAM but that is an + * acceptable failure mode and should only bite obscure boxes with + * a VESA hole at 15Mb + * + * The second case Clip sometimes kicks in is when the EBDA is marked + * as reserved. Again we fail safe with reasonable results + */ + if (top > clip) + top = clip; + return top; } /* - * Compute a set of MCR's to give maximum coverage + * Compute a set of MCR's to give maximum coverage */ - static int __cpuinit centaur_mcr_compute(int nr, int key) { u32 mem = ramtop(); @@ -99,141 +99,131 @@ static int __cpuinit centaur_mcr_compute(int nr, int key) u32 top = root; u32 floor = 0; int ct = 0; - - while(ct<nr) - { + + while (ct < nr) { u32 fspace = 0; + u32 high; + u32 low; /* - * Find the largest block we will fill going upwards + * Find the largest block we will fill going upwards */ - - u32 high = power2(mem-top); + high = power2(mem-top); /* - * Find the largest block we will fill going downwards + * Find the largest block we will fill going downwards */ - - u32 low = base/2; + low = base/2; /* - * Don't fill below 1Mb going downwards as there - * is an ISA hole in the way. - */ - - if(base <= 1024*1024) + * Don't fill below 1Mb going downwards as there + * is an ISA hole in the way. + */ + if (base <= 1024*1024) low = 0; - + /* - * See how much space we could cover by filling below - * the ISA hole + * See how much space we could cover by filling below + * the ISA hole */ - - if(floor == 0) + + if (floor == 0) fspace = 512*1024; - else if(floor ==512*1024) + else if (floor == 512*1024) fspace = 128*1024; /* And forget ROM space */ - + /* - * Now install the largest coverage we get + * Now install the largest coverage we get */ - - if(fspace > high && fspace > low) - { + if (fspace > high && fspace > low) { centaur_mcr_insert(ct, floor, fspace, key); floor += fspace; - } - else if(high > low) - { + } else if (high > low) { centaur_mcr_insert(ct, top, high, key); top += high; - } - else if(low > 0) - { + } else if (low > 0) { base -= low; centaur_mcr_insert(ct, base, low, key); - } - else break; + } else + break; ct++; } /* - * We loaded ct values. We now need to set the mask. The caller - * must do this bit. + * We loaded ct values. We now need to set the mask. The caller + * must do this bit. */ - return ct; } static void __cpuinit centaur_create_optimal_mcr(void) { + int used; int i; + /* - * Allocate up to 6 mcrs to mark as much of ram as possible - * as write combining and weak write ordered. + * Allocate up to 6 mcrs to mark as much of ram as possible + * as write combining and weak write ordered. * - * To experiment with: Linux never uses stack operations for - * mmio spaces so we could globally enable stack operation wc + * To experiment with: Linux never uses stack operations for + * mmio spaces so we could globally enable stack operation wc * - * Load the registers with type 31 - full write combining, all - * writes weakly ordered. + * Load the registers with type 31 - full write combining, all + * writes weakly ordered. */ - int used = centaur_mcr_compute(6, 31); + used = centaur_mcr_compute(6, 31); /* - * Wipe unused MCRs + * Wipe unused MCRs */ - - for(i=used;i<8;i++) + for (i = used; i < 8; i++) wrmsr(MSR_IDT_MCR0+i, 0, 0); } static void __cpuinit winchip2_create_optimal_mcr(void) { u32 lo, hi; + int used; int i; /* - * Allocate up to 6 mcrs to mark as much of ram as possible - * as write combining, weak store ordered. + * Allocate up to 6 mcrs to mark as much of ram as possible + * as write combining, weak store ordered. * - * Load the registers with type 25 - * 8 - weak write ordering - * 16 - weak read ordering - * 1 - write combining + * Load the registers with type 25 + * 8 - weak write ordering + * 16 - weak read ordering + * 1 - write combining */ + used = centaur_mcr_compute(6, 25); - int used = centaur_mcr_compute(6, 25); - /* - * Mark the registers we are using. + * Mark the registers we are using. */ - rdmsr(MSR_IDT_MCR_CTRL, lo, hi); - for(i=0;i<used;i++) - lo|=1<<(9+i); + for (i = 0; i < used; i++) + lo |= 1<<(9+i); wrmsr(MSR_IDT_MCR_CTRL, lo, hi); - + /* - * Wipe unused MCRs + * Wipe unused MCRs */ - - for(i=used;i<8;i++) + + for (i = used; i < 8; i++) wrmsr(MSR_IDT_MCR0+i, 0, 0); } /* - * Handle the MCR key on the Winchip 2. + * Handle the MCR key on the Winchip 2. */ - static void __cpuinit winchip2_unprotect_mcr(void) { u32 lo, hi; u32 key; - + rdmsr(MSR_IDT_MCR_CTRL, lo, hi); - lo&=~0x1C0; /* blank bits 8-6 */ + lo &= ~0x1C0; /* blank bits 8-6 */ key = (lo>>17) & 7; lo |= key<<6; /* replace with unlock key */ wrmsr(MSR_IDT_MCR_CTRL, lo, hi); @@ -242,9 +232,9 @@ static void __cpuinit winchip2_unprotect_mcr(void) static void __cpuinit winchip2_protect_mcr(void) { u32 lo, hi; - + rdmsr(MSR_IDT_MCR_CTRL, lo, hi); - lo&=~0x1C0; /* blank bits 8-6 */ + lo &= ~0x1C0; /* blank bits 8-6 */ wrmsr(MSR_IDT_MCR_CTRL, lo, hi); } #endif /* CONFIG_X86_OOSTORE */ @@ -267,17 +257,17 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c) /* enable ACE unit, if present and disabled */ if ((tmp & (ACE_PRESENT | ACE_ENABLED)) == ACE_PRESENT) { - rdmsr (MSR_VIA_FCR, lo, hi); + rdmsr(MSR_VIA_FCR, lo, hi); lo |= ACE_FCR; /* enable ACE unit */ - wrmsr (MSR_VIA_FCR, lo, hi); + wrmsr(MSR_VIA_FCR, lo, hi); printk(KERN_INFO "CPU: Enabled ACE h/w crypto\n"); } /* enable RNG unit, if present and disabled */ if ((tmp & (RNG_PRESENT | RNG_ENABLED)) == RNG_PRESENT) { - rdmsr (MSR_VIA_RNG, lo, hi); + rdmsr(MSR_VIA_RNG, lo, hi); lo |= RNG_ENABLE; /* enable RNG unit */ - wrmsr (MSR_VIA_RNG, lo, hi); + wrmsr(MSR_VIA_RNG, lo, hi); printk(KERN_INFO "CPU: Enabled h/w RNG\n"); } @@ -288,171 +278,183 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c) } /* Cyrix III family needs CX8 & PGE explicitly enabled. */ - if (c->x86_model >=6 && c->x86_model <= 9) { - rdmsr (MSR_VIA_FCR, lo, hi); + if (c->x86_model >= 6 && c->x86_model <= 9) { + rdmsr(MSR_VIA_FCR, lo, hi); lo |= (1<<1 | 1<<7); - wrmsr (MSR_VIA_FCR, lo, hi); - set_bit(X86_FEATURE_CX8, c->x86_capability); + wrmsr(MSR_VIA_FCR, lo, hi); + set_cpu_cap(c, X86_FEATURE_CX8); } /* Before Nehemiah, the C3's had 3dNOW! */ - if (c->x86_model >=6 && c->x86_model <9) - set_bit(X86_FEATURE_3DNOW, c->x86_capability); + if (c->x86_model >= 6 && c->x86_model < 9) + set_cpu_cap(c, X86_FEATURE_3DNOW); get_model_name(c); display_cacheinfo(c); } +enum { + ECX8 = 1<<1, + EIERRINT = 1<<2, + DPM = 1<<3, + DMCE = 1<<4, + DSTPCLK = 1<<5, + ELINEAR = 1<<6, + DSMC = 1<<7, + DTLOCK = 1<<8, + EDCTLB = 1<<8, + EMMX = 1<<9, + DPDC = 1<<11, + EBRPRED = 1<<12, + DIC = 1<<13, + DDC = 1<<14, + DNA = 1<<15, + ERETSTK = 1<<16, + E2MMX = 1<<19, + EAMD3D = 1<<20, +}; + static void __cpuinit init_centaur(struct cpuinfo_x86 *c) { - enum { - ECX8=1<<1, - EIERRINT=1<<2, - DPM=1<<3, - DMCE=1<<4, - DSTPCLK=1<<5, - ELINEAR=1<<6, - DSMC=1<<7, - DTLOCK=1<<8, - EDCTLB=1<<8, - EMMX=1<<9, - DPDC=1<<11, - EBRPRED=1<<12, - DIC=1<<13, - DDC=1<<14, - DNA=1<<15, - ERETSTK=1<<16, - E2MMX=1<<19, - EAMD3D=1<<20, - }; char *name; - u32 fcr_set=0; - u32 fcr_clr=0; - u32 lo,hi,newlo; - u32 aa,bb,cc,dd; + u32 fcr_set = 0; + u32 fcr_clr = 0; + u32 lo, hi, newlo; + u32 aa, bb, cc, dd; - /* Bit 31 in normal CPUID used for nonstandard 3DNow ID; - 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */ - clear_bit(0*32+31, c->x86_capability); + /* + * Bit 31 in normal CPUID used for nonstandard 3DNow ID; + * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway + */ + clear_cpu_cap(c, 0*32+31); switch (c->x86) { - - case 5: - switch(c->x86_model) { - case 4: - name="C6"; - fcr_set=ECX8|DSMC|EDCTLB|EMMX|ERETSTK; - fcr_clr=DPDC; - printk(KERN_NOTICE "Disabling bugged TSC.\n"); - clear_bit(X86_FEATURE_TSC, c->x86_capability); + case 5: + switch (c->x86_model) { + case 4: + name = "C6"; + fcr_set = ECX8|DSMC|EDCTLB|EMMX|ERETSTK; + fcr_clr = DPDC; + printk(KERN_NOTICE "Disabling bugged TSC.\n"); + clear_cpu_cap(c, X86_FEATURE_TSC); #ifdef CONFIG_X86_OOSTORE - centaur_create_optimal_mcr(); - /* Enable - write combining on non-stack, non-string - write combining on string, all types - weak write ordering - - The C6 original lacks weak read order - - Note 0x120 is write only on Winchip 1 */ - - wrmsr(MSR_IDT_MCR_CTRL, 0x01F0001F, 0); -#endif + centaur_create_optimal_mcr(); + /* + * Enable: + * write combining on non-stack, non-string + * write combining on string, all types + * weak write ordering + * + * The C6 original lacks weak read order + * + * Note 0x120 is write only on Winchip 1 + */ + wrmsr(MSR_IDT_MCR_CTRL, 0x01F0001F, 0); +#endif + break; + case 8: + switch (c->x86_mask) { + default: + name = "2"; + break; + case 7 ... 9: + name = "2A"; break; - case 8: - switch(c->x86_mask) { - default: - name="2"; - break; - case 7 ... 9: - name="2A"; - break; - case 10 ... 15: - name="2B"; - break; - } - fcr_set=ECX8|DSMC|DTLOCK|EMMX|EBRPRED|ERETSTK|E2MMX|EAMD3D; - fcr_clr=DPDC; + case 10 ... 15: + name = "2B"; + break; + } + fcr_set = ECX8|DSMC|DTLOCK|EMMX|EBRPRED|ERETSTK| + E2MMX|EAMD3D; + fcr_clr = DPDC; #ifdef CONFIG_X86_OOSTORE - winchip2_unprotect_mcr(); - winchip2_create_optimal_mcr(); - rdmsr(MSR_IDT_MCR_CTRL, lo, hi); - /* Enable - write combining on non-stack, non-string - write combining on string, all types - weak write ordering - */ - lo|=31; - wrmsr(MSR_IDT_MCR_CTRL, lo, hi); - winchip2_protect_mcr(); + winchip2_unprotect_mcr(); + winchip2_create_optimal_mcr(); + rdmsr(MSR_IDT_MCR_CTRL, lo, hi); + /* + * Enable: + * write combining on non-stack, non-string + * write combining on string, all types + * weak write ordering + */ + lo |= 31; + wrmsr(MSR_IDT_MCR_CTRL, lo, hi); + winchip2_protect_mcr(); #endif - break; - case 9: - name="3"; - fcr_set=ECX8|DSMC|DTLOCK|EMMX|EBRPRED|ERETSTK|E2MMX|EAMD3D; - fcr_clr=DPDC; + break; + case 9: + name = "3"; + fcr_set = ECX8|DSMC|DTLOCK|EMMX|EBRPRED|ERETSTK| + E2MMX|EAMD3D; + fcr_clr = DPDC; #ifdef CONFIG_X86_OOSTORE - winchip2_unprotect_mcr(); - winchip2_create_optimal_mcr(); - rdmsr(MSR_IDT_MCR_CTRL, lo, hi); - /* Enable - write combining on non-stack, non-string - write combining on string, all types - weak write ordering - */ - lo|=31; - wrmsr(MSR_IDT_MCR_CTRL, lo, hi); - winchip2_protect_mcr(); + winchip2_unprotect_mcr(); + winchip2_create_optimal_mcr(); + rdmsr(MSR_IDT_MCR_CTRL, lo, hi); + /* + * Enable: + * write combining on non-stack, non-string + * write combining on string, all types + * weak write ordering + */ + lo |= 31; + wrmsr(MSR_IDT_MCR_CTRL, lo, hi); + winchip2_protect_mcr(); #endif - break; - default: - name="??"; - } + break; + default: + name = "??"; + } - rdmsr(MSR_IDT_FCR1, lo, hi); - newlo=(lo|fcr_set) & (~fcr_clr); + rdmsr(MSR_IDT_FCR1, lo, hi); + newlo = (lo|fcr_set) & (~fcr_clr); - if (newlo!=lo) { - printk(KERN_INFO "Centaur FCR was 0x%X now 0x%X\n", lo, newlo ); - wrmsr(MSR_IDT_FCR1, newlo, hi ); - } else { - printk(KERN_INFO "Centaur FCR is 0x%X\n",lo); - } - /* Emulate MTRRs using Centaur's MCR. */ - set_bit(X86_FEATURE_CENTAUR_MCR, c->x86_capability); - /* Report CX8 */ - set_bit(X86_FEATURE_CX8, c->x86_capability); - /* Set 3DNow! on Winchip 2 and above. */ - if (c->x86_model >=8) - set_bit(X86_FEATURE_3DNOW, c->x86_capability); - /* See if we can find out some more. */ - if ( cpuid_eax(0x80000000) >= 0x80000005 ) { - /* Yes, we can. */ - cpuid(0x80000005,&aa,&bb,&cc,&dd); - /* Add L1 data and code cache sizes. */ - c->x86_cache_size = (cc>>24)+(dd>>24); - } - sprintf( c->x86_model_id, "WinChip %s", name ); - break; + if (newlo != lo) { + printk(KERN_INFO "Centaur FCR was 0x%X now 0x%X\n", + lo, newlo); + wrmsr(MSR_IDT_FCR1, newlo, hi); + } else { + printk(KERN_INFO "Centaur FCR is 0x%X\n", lo); + } + /* Emulate MTRRs using Centaur's MCR. */ + set_cpu_cap(c, X86_FEATURE_CENTAUR_MCR); + /* Report CX8 */ + set_cpu_cap(c, X86_FEATURE_CX8); + /* Set 3DNow! on Winchip 2 and above. */ + if (c->x86_model >= 8) + set_cpu_cap(c, X86_FEATURE_3DNOW); + /* See if we can find out some more. */ + if (cpuid_eax(0x80000000) >= 0x80000005) { + /* Yes, we can. */ + cpuid(0x80000005, &aa, &bb, &cc, &dd); + /* Add L1 data and code cache sizes. */ + c->x86_cache_size = (cc>>24)+(dd>>24); + } + sprintf(c->x86_model_id, "WinChip %s", name); + break; - case 6: - init_c3(c); - break; + case 6: + init_c3(c); + break; } } -static unsigned int __cpuinit centaur_size_cache(struct cpuinfo_x86 * c, unsigned int size) +static unsigned int __cpuinit +centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size) { /* VIA C3 CPUs (670-68F) need further shifting. */ if ((c->x86 == 6) && ((c->x86_model == 7) || (c->x86_model == 8))) size >>= 8; - /* VIA also screwed up Nehemiah stepping 1, and made - it return '65KB' instead of '64KB' - - Note, it seems this may only be in engineering samples. */ - if ((c->x86==6) && (c->x86_model==9) && (c->x86_mask==1) && (size==65)) - size -=1; + /* + * There's also an erratum in Nehemiah stepping 1, which + * returns '65KB' instead of '64KB' + * - Note, it seems this may only be in engineering samples. + */ + if ((c->x86 == 6) && (c->x86_model == 9) && + (c->x86_mask == 1) && (size == 65)) + size -= 1; return size; } @@ -464,8 +466,4 @@ static struct cpu_dev centaur_cpu_dev __cpuinitdata = { .c_size_cache = centaur_size_cache, }; -int __init centaur_init_cpu(void) -{ - cpu_devs[X86_VENDOR_CENTAUR] = ¢aur_cpu_dev; - return 0; -} +cpu_vendor_dev_register(X86_VENDOR_CENTAUR, ¢aur_cpu_dev); diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index a38aafaefc2..d999d7833bc 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -62,9 +62,9 @@ __u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata; static int cachesize_override __cpuinitdata = -1; static int disable_x86_serial_nr __cpuinitdata = 1; -struct cpu_dev * cpu_devs[X86_VENDOR_NUM] = {}; +struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {}; -static void __cpuinit default_init(struct cpuinfo_x86 * c) +static void __cpuinit default_init(struct cpuinfo_x86 *c) { /* Not much we can do here... */ /* Check if at least it has cpuid */ @@ -81,11 +81,11 @@ static struct cpu_dev __cpuinitdata default_cpu = { .c_init = default_init, .c_vendor = "Unknown", }; -static struct cpu_dev * this_cpu __cpuinitdata = &default_cpu; +static struct cpu_dev *this_cpu __cpuinitdata = &default_cpu; static int __init cachesize_setup(char *str) { - get_option (&str, &cachesize_override); + get_option(&str, &cachesize_override); return 1; } __setup("cachesize=", cachesize_setup); @@ -107,12 +107,12 @@ int __cpuinit get_model_name(struct cpuinfo_x86 *c) /* Intel chips right-justify this string for some dumb reason; undo that brain damage */ p = q = &c->x86_model_id[0]; - while ( *p == ' ' ) + while (*p == ' ') p++; - if ( p != q ) { - while ( *p ) + if (p != q) { + while (*p) *q++ = *p++; - while ( q <= &c->x86_model_id[48] ) + while (q <= &c->x86_model_id[48]) *q++ = '\0'; /* Zero-pad the rest */ } @@ -130,7 +130,7 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) cpuid(0x80000005, &dummy, &dummy, &ecx, &edx); printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n", edx>>24, edx&0xFF, ecx>>24, ecx&0xFF); - c->x86_cache_size=(ecx>>24)+(edx>>24); + c->x86_cache_size = (ecx>>24)+(edx>>24); } if (n < 0x80000006) /* Some chips just has a large L1. */ @@ -138,16 +138,16 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) ecx = cpuid_ecx(0x80000006); l2size = ecx >> 16; - + /* do processor-specific cache resizing */ if (this_cpu->c_size_cache) - l2size = this_cpu->c_size_cache(c,l2size); + l2size = this_cpu->c_size_cache(c, l2size); /* Allow user to override all this if necessary. */ if (cachesize_override != -1) l2size = cachesize_override; - if ( l2size == 0 ) + if (l2size == 0) return; /* Again, no L2 cache is possible */ c->x86_cache_size = l2size; @@ -156,16 +156,19 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) l2size, ecx & 0xFF); } -/* Naming convention should be: <Name> [(<Codename>)] */ -/* This table only is used unless init_<vendor>() below doesn't set it; */ -/* in particular, if CPUID levels 0x80000002..4 are supported, this isn't used */ +/* + * Naming convention should be: <Name> [(<Codename>)] + * This table only is used unless init_<vendor>() below doesn't set it; + * in particular, if CPUID levels 0x80000002..4 are supported, this isn't used + * + */ /* Look up CPU names by table lookup. */ static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c) { struct cpu_model_info *info; - if ( c->x86_model >= 16 ) + if (c->x86_model >= 16) return NULL; /* Range check */ if (!this_cpu) @@ -190,9 +193,9 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early) for (i = 0; i < X86_VENDOR_NUM; i++) { if (cpu_devs[i]) { - if (!strcmp(v,cpu_devs[i]->c_ident[0]) || - (cpu_devs[i]->c_ident[1] && - !strcmp(v,cpu_devs[i]->c_ident[1]))) { + if (!strcmp(v, cpu_devs[i]->c_ident[0]) || + (cpu_devs[i]->c_ident[1] && + !strcmp(v, cpu_devs[i]->c_ident[1]))) { c->x86_vendor = i; if (!early) this_cpu = cpu_devs[i]; @@ -210,7 +213,7 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early) } -static int __init x86_fxsr_setup(char * s) +static int __init x86_fxsr_setup(char *s) { setup_clear_cpu_cap(X86_FEATURE_FXSR); setup_clear_cpu_cap(X86_FEATURE_XMM); @@ -219,7 +222,7 @@ static int __init x86_fxsr_setup(char * s) __setup("nofxsr", x86_fxsr_setup); -static int __init x86_sep_setup(char * s) +static int __init x86_sep_setup(char *s) { setup_clear_cpu_cap(X86_FEATURE_SEP); return 1; @@ -306,14 +309,30 @@ static void __cpuinit early_get_cap(struct cpuinfo_x86 *c) } -} + clear_cpu_cap(c, X86_FEATURE_PAT); + + switch (c->x86_vendor) { + case X86_VENDOR_AMD: + if (c->x86 >= 0xf && c->x86 <= 0x11) + set_cpu_cap(c, X86_FEATURE_PAT); + break; + case X86_VENDOR_INTEL: + if (c->x86 == 0xF || (c->x86 == 6 && c->x86_model >= 15)) + set_cpu_cap(c, X86_FEATURE_PAT); + break; + } -/* Do minimum CPU detection early. - Fields really needed: vendor, cpuid_level, family, model, mask, cache alignment. - The others are not touched to avoid unwanted side effects. +} - WARNING: this function is only called on the BP. Don't add code here - that is supposed to run on all CPUs. */ +/* + * Do minimum CPU detection early. + * Fields really needed: vendor, cpuid_level, family, model, mask, + * cache alignment. + * The others are not touched to avoid unwanted side effects. + * + * WARNING: this function is only called on the BP. Don't add code here + * that is supposed to run on all CPUs. + */ static void __init early_cpu_detect(void) { struct cpuinfo_x86 *c = &boot_cpu_data; @@ -328,19 +347,14 @@ static void __init early_cpu_detect(void) get_cpu_vendor(c, 1); - switch (c->x86_vendor) { - case X86_VENDOR_AMD: - early_init_amd(c); - break; - case X86_VENDOR_INTEL: - early_init_intel(c); - break; - } + if (c->x86_vendor != X86_VENDOR_UNKNOWN && + cpu_devs[c->x86_vendor]->c_early_init) + cpu_devs[c->x86_vendor]->c_early_init(c); early_get_cap(c); } -static void __cpuinit generic_identify(struct cpuinfo_x86 * c) +static void __cpuinit generic_identify(struct cpuinfo_x86 *c) { u32 tfms, xlvl; unsigned int ebx; @@ -351,13 +365,12 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 * c) (unsigned int *)&c->x86_vendor_id[0], (unsigned int *)&c->x86_vendor_id[8], (unsigned int *)&c->x86_vendor_id[4]); - + get_cpu_vendor(c, 0); /* Initialize the standard set of capabilities */ /* Note that the vendor-specific code below might override */ - /* Intel-defined flags: level 0x00000001 */ - if ( c->cpuid_level >= 0x00000001 ) { + if (c->cpuid_level >= 0x00000001) { u32 capability, excap; cpuid(0x00000001, &tfms, &ebx, &excap, &capability); c->x86_capability[0] = capability; @@ -369,12 +382,14 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 * c) if (c->x86 >= 0x6) c->x86_model += ((tfms >> 16) & 0xF) << 4; c->x86_mask = tfms & 15; + c->initial_apicid = (ebx >> 24) & 0xFF; #ifdef CONFIG_X86_HT - c->apicid = phys_pkg_id((ebx >> 24) & 0xFF, 0); + c->apicid = phys_pkg_id(c->initial_apicid, 0); + c->phys_proc_id = c->initial_apicid; #else - c->apicid = (ebx >> 24) & 0xFF; + c->apicid = c->initial_apicid; #endif - if (c->x86_capability[0] & (1<<19)) + if (test_cpu_cap(c, X86_FEATURE_CLFLSH)) c->x86_clflush_size = ((ebx >> 8) & 0xff) * 8; } else { /* Have CPUID level 0 only - unheard of */ @@ -383,33 +398,42 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 * c) /* AMD-defined flags: level 0x80000001 */ xlvl = cpuid_eax(0x80000000); - if ( (xlvl & 0xffff0000) == 0x80000000 ) { - if ( xlvl >= 0x80000001 ) { + if ((xlvl & 0xffff0000) == 0x80000000) { + if (xlvl >= 0x80000001) { c->x86_capability[1] = cpuid_edx(0x80000001); c->x86_capability[6] = cpuid_ecx(0x80000001); } - if ( xlvl >= 0x80000004 ) + if (xlvl >= 0x80000004) get_model_name(c); /* Default name */ } init_scattered_cpuid_features(c); } -#ifdef CONFIG_X86_HT - c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff; -#endif + clear_cpu_cap(c, X86_FEATURE_PAT); + + switch (c->x86_vendor) { + case X86_VENDOR_AMD: + if (c->x86 >= 0xf && c->x86 <= 0x11) + set_cpu_cap(c, X86_FEATURE_PAT); + break; + case X86_VENDOR_INTEL: + if (c->x86 == 0xF || (c->x86 == 6 && c->x86_model >= 15)) + set_cpu_cap(c, X86_FEATURE_PAT); + break; + } } static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c) { - if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr ) { + if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr) { /* Disable processor serial number */ - unsigned long lo,hi; - rdmsr(MSR_IA32_BBL_CR_CTL,lo,hi); + unsigned long lo, hi; + rdmsr(MSR_IA32_BBL_CR_CTL, lo, hi); lo |= 0x200000; - wrmsr(MSR_IA32_BBL_CR_CTL,lo,hi); + wrmsr(MSR_IA32_BBL_CR_CTL, lo, hi); printk(KERN_NOTICE "CPU serial number disabled.\n"); - clear_bit(X86_FEATURE_PN, c->x86_capability); + clear_cpu_cap(c, X86_FEATURE_PN); /* Disabling the serial number may affect the cpuid level */ c->cpuid_level = cpuid_eax(0); @@ -444,9 +468,11 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c) memset(&c->x86_capability, 0, sizeof c->x86_capability); if (!have_cpuid_p()) { - /* First of all, decide if this is a 486 or higher */ - /* It's a 486 if we can modify the AC flag */ - if ( flag_is_changeable_p(X86_EFLAGS_AC) ) + /* + * First of all, decide if this is a 486 or higher + * It's a 486 if we can modify the AC flag + */ + if (flag_is_changeable_p(X86_EFLAGS_AC)) c->x86 = 4; else c->x86 = 3; @@ -479,10 +505,10 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c) */ /* If the model name is still unset, do table lookup. */ - if ( !c->x86_model_id[0] ) { + if (!c->x86_model_id[0]) { char *p; p = table_lookup_model(c); - if ( p ) + if (p) strcpy(c->x86_model_id, p); else /* Last resort... */ @@ -496,9 +522,9 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c) * common between the CPUs. The first time this routine gets * executed, c == &boot_cpu_data. */ - if ( c != &boot_cpu_data ) { + if (c != &boot_cpu_data) { /* AND the already accumulated flags with these */ - for ( i = 0 ; i < NCAPINTS ; i++ ) + for (i = 0 ; i < NCAPINTS ; i++) boot_cpu_data.x86_capability[i] &= c->x86_capability[i]; } @@ -542,7 +568,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) if (smp_num_siblings == 1) { printk(KERN_INFO "CPU: Hyper-Threading is disabled\n"); - } else if (smp_num_siblings > 1 ) { + } else if (smp_num_siblings > 1) { if (smp_num_siblings > NR_CPUS) { printk(KERN_WARNING "CPU: Unsupported number of the " @@ -552,7 +578,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) } index_msb = get_count_order(smp_num_siblings); - c->phys_proc_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb); + c->phys_proc_id = phys_pkg_id(c->initial_apicid, index_msb); printk(KERN_INFO "CPU: Physical Processor ID: %d\n", c->phys_proc_id); @@ -563,7 +589,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) core_bits = get_count_order(c->x86_max_cores); - c->cpu_core_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb) & + c->cpu_core_id = phys_pkg_id(c->initial_apicid, index_msb) & ((1 << core_bits) - 1); if (c->x86_max_cores > 1) @@ -597,7 +623,7 @@ void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) else printk("%s", c->x86_model_id); - if (c->x86_mask || c->cpuid_level >= 0) + if (c->x86_mask || c->cpuid_level >= 0) printk(" stepping %02x\n", c->x86_mask); else printk("\n"); @@ -616,23 +642,15 @@ __setup("clearcpuid=", setup_disablecpuid); cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE; -/* This is hacky. :) - * We're emulating future behavior. - * In the future, the cpu-specific init functions will be called implicitly - * via the magic of initcalls. - * They will insert themselves into the cpu_devs structure. - * Then, when cpu_init() is called, we can just iterate over that array. - */ void __init early_cpu_init(void) { - intel_cpu_init(); - cyrix_init_cpu(); - nsc_init_cpu(); - amd_init_cpu(); - centaur_init_cpu(); - transmeta_init_cpu(); - nexgen_init_cpu(); - umc_init_cpu(); + struct cpu_vendor_dev *cvdev; + + for (cvdev = __x86cpuvendor_start ; + cvdev < __x86cpuvendor_end ; + cvdev++) + cpu_devs[cvdev->vendor] = cvdev->cpu_dev; + early_cpu_detect(); } @@ -666,7 +684,7 @@ void __cpuinit cpu_init(void) { int cpu = smp_processor_id(); struct task_struct *curr = current; - struct tss_struct * t = &per_cpu(init_tss, cpu); + struct tss_struct *t = &per_cpu(init_tss, cpu); struct thread_struct *thread = &curr->thread; if (cpu_test_and_set(cpu, cpu_initialized)) { @@ -692,7 +710,7 @@ void __cpuinit cpu_init(void) enter_lazy_tlb(&init_mm, curr); load_sp0(t, thread); - set_tss_desc(cpu,t); + set_tss_desc(cpu, t); load_TR_desc(); load_LDT(&init_mm.context); diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h index e0b38c33d84..783691b2a73 100644 --- a/arch/x86/kernel/cpu/cpu.h +++ b/arch/x86/kernel/cpu/cpu.h @@ -14,6 +14,7 @@ struct cpu_dev { struct cpu_model_info c_models[4]; + void (*c_early_init)(struct cpuinfo_x86 *c); void (*c_init)(struct cpuinfo_x86 * c); void (*c_identify)(struct cpuinfo_x86 * c); unsigned int (*c_size_cache)(struct cpuinfo_x86 * c, unsigned int size); @@ -21,18 +22,17 @@ struct cpu_dev { extern struct cpu_dev * cpu_devs [X86_VENDOR_NUM]; +struct cpu_vendor_dev { + int vendor; + struct cpu_dev *cpu_dev; +}; + +#define cpu_vendor_dev_register(cpu_vendor_id, cpu_dev) \ + static struct cpu_vendor_dev __cpu_vendor_dev_##cpu_vendor_id __used \ + __attribute__((__section__(".x86cpuvendor.init"))) = \ + { cpu_vendor_id, cpu_dev } + +extern struct cpu_vendor_dev __x86cpuvendor_start[], __x86cpuvendor_end[]; + extern int get_model_name(struct cpuinfo_x86 *c); extern void display_cacheinfo(struct cpuinfo_x86 *c); - -extern void early_init_intel(struct cpuinfo_x86 *c); -extern void early_init_amd(struct cpuinfo_x86 *c); - -/* Specific CPU type init functions */ -int intel_cpu_init(void); -int amd_init_cpu(void); -int cyrix_init_cpu(void); -int nsc_init_cpu(void); -int centaur_init_cpu(void); -int transmeta_init_cpu(void); -int nexgen_init_cpu(void); -int umc_init_cpu(void); diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c index 7139b026270..3fd7a67bb06 100644 --- a/arch/x86/kernel/cpu/cyrix.c +++ b/arch/x86/kernel/cpu/cyrix.c @@ -19,7 +19,7 @@ static void __cpuinit do_cyrix_devid(unsigned char *dir0, unsigned char *dir1) { unsigned char ccr2, ccr3; unsigned long flags; - + /* we test for DEVID by checking whether CCR3 is writable */ local_irq_save(flags); ccr3 = getCx86(CX86_CCR3); @@ -37,8 +37,7 @@ static void __cpuinit do_cyrix_devid(unsigned char *dir0, unsigned char *dir1) setCx86(CX86_CCR2, ccr2); *dir0 = 0xfe; } - } - else { + } else { setCx86(CX86_CCR3, ccr3); /* restore CCR3 */ /* read DIR0 and DIR1 CPU registers */ @@ -86,7 +85,7 @@ static char cyrix_model_mult2[] __cpuinitdata = "12233445"; static void __cpuinit check_cx686_slop(struct cpuinfo_x86 *c) { unsigned long flags; - + if (Cx86_dir0_msb == 3) { unsigned char ccr3, ccr5; @@ -132,7 +131,7 @@ static void __cpuinit set_cx86_memwb(void) /* set 'Not Write-through' */ write_cr0(read_cr0() | X86_CR0_NW); /* CCR2 bit 2: lock NW bit and set WT1 */ - setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x14 ); + setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x14); } static void __cpuinit set_cx86_inc(void) @@ -148,7 +147,7 @@ static void __cpuinit set_cx86_inc(void) setCx86(CX86_PCR1, getCx86(CX86_PCR1) | 0x02); /* PCR0 -- Performance Control */ /* Incrementor Margin 10 */ - setCx86(CX86_PCR0, getCx86(CX86_PCR0) | 0x04); + setCx86(CX86_PCR0, getCx86(CX86_PCR0) | 0x04); setCx86(CX86_CCR3, ccr3); /* disable MAPEN */ } @@ -167,16 +166,16 @@ static void __cpuinit geode_configure(void) ccr3 = getCx86(CX86_CCR3); setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */ - + /* FPU fast, DTE cache, Mem bypass */ setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x38); setCx86(CX86_CCR3, ccr3); /* disable MAPEN */ - + set_cx86_memwb(); - set_cx86_reorder(); + set_cx86_reorder(); set_cx86_inc(); - + local_irq_restore(flags); } @@ -187,14 +186,16 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) char *buf = c->x86_model_id; const char *p = NULL; - /* Bit 31 in normal CPUID used for nonstandard 3DNow ID; - 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */ - clear_bit(0*32+31, c->x86_capability); + /* + * Bit 31 in normal CPUID used for nonstandard 3DNow ID; + * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway + */ + clear_cpu_cap(c, 0*32+31); /* Cyrix used bit 24 in extended (AMD) CPUID for Cyrix MMX extensions */ - if ( test_bit(1*32+24, c->x86_capability) ) { - clear_bit(1*32+24, c->x86_capability); - set_bit(X86_FEATURE_CXMMX, c->x86_capability); + if (test_cpu_cap(c, 1*32+24)) { + clear_cpu_cap(c, 1*32+24); + set_cpu_cap(c, X86_FEATURE_CXMMX); } do_cyrix_devid(&dir0, &dir1); @@ -213,7 +214,7 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) * the model, multiplier and stepping. Black magic included, * to make the silicon step/rev numbers match the printed ones. */ - + switch (dir0_msn) { unsigned char tmp; @@ -241,7 +242,7 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) } else /* 686 */ p = Cx86_cb+1; /* Emulate MTRRs using Cyrix's ARRs. */ - set_bit(X86_FEATURE_CYRIX_ARR, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_CYRIX_ARR); /* 6x86's contain this bug */ c->coma_bug = 1; break; @@ -250,17 +251,18 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) #ifdef CONFIG_PCI { u32 vendor, device; - /* It isn't really a PCI quirk directly, but the cure is the - same. The MediaGX has deep magic SMM stuff that handles the - SB emulation. It throws away the fifo on disable_dma() which - is wrong and ruins the audio. - - Bug2: VSA1 has a wrap bug so that using maximum sized DMA - causes bad things. According to NatSemi VSA2 has another - bug to do with 'hlt'. I've not seen any boards using VSA2 - and X doesn't seem to support it either so who cares 8). - VSA1 we work around however. - */ + /* + * It isn't really a PCI quirk directly, but the cure is the + * same. The MediaGX has deep magic SMM stuff that handles the + * SB emulation. It throws away the fifo on disable_dma() which + * is wrong and ruins the audio. + * + * Bug2: VSA1 has a wrap bug so that using maximum sized DMA + * causes bad things. According to NatSemi VSA2 has another + * bug to do with 'hlt'. I've not seen any boards using VSA2 + * and X doesn't seem to support it either so who cares 8). + * VSA1 we work around however. + */ printk(KERN_INFO "Working around Cyrix MediaGX virtual DMA bugs.\n"); isa_dma_bridge_buggy = 2; @@ -273,55 +275,51 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) /* * The 5510/5520 companion chips have a funky PIT. - */ + */ if (vendor == PCI_VENDOR_ID_CYRIX && (device == PCI_DEVICE_ID_CYRIX_5510 || device == PCI_DEVICE_ID_CYRIX_5520)) mark_tsc_unstable("cyrix 5510/5520 detected"); } #endif - c->x86_cache_size=16; /* Yep 16K integrated cache thats it */ + c->x86_cache_size = 16; /* Yep 16K integrated cache thats it */ /* GXm supports extended cpuid levels 'ala' AMD */ if (c->cpuid_level == 2) { /* Enable cxMMX extensions (GX1 Datasheet 54) */ setCx86(CX86_CCR7, getCx86(CX86_CCR7) | 1); - + /* * GXm : 0x30 ... 0x5f GXm datasheet 51 * GXlv: 0x6x GXlv datasheet 54 * ? : 0x7x * GX1 : 0x8x GX1 datasheet 56 */ - if((0x30 <= dir1 && dir1 <= 0x6f) || (0x80 <=dir1 && dir1 <= 0x8f)) + if ((0x30 <= dir1 && dir1 <= 0x6f) || (0x80 <= dir1 && dir1 <= 0x8f)) geode_configure(); get_model_name(c); /* get CPU marketing name */ return; - } - else { /* MediaGX */ + } else { /* MediaGX */ Cx86_cb[2] = (dir0_lsn & 1) ? '3' : '4'; p = Cx86_cb+2; c->x86_model = (dir1 & 0x20) ? 1 : 2; } break; - case 5: /* 6x86MX/M II */ - if (dir1 > 7) - { + case 5: /* 6x86MX/M II */ + if (dir1 > 7) { dir0_msn++; /* M II */ /* Enable MMX extensions (App note 108) */ setCx86(CX86_CCR7, getCx86(CX86_CCR7)|1); - } - else - { + } else { c->coma_bug = 1; /* 6x86MX, it has the bug. */ } tmp = (!(dir0_lsn & 7) || dir0_lsn & 1) ? 2 : 0; Cx86_cb[tmp] = cyrix_model_mult2[dir0_lsn & 7]; p = Cx86_cb+tmp; - if (((dir1 & 0x0f) > 4) || ((dir1 & 0xf0) == 0x20)) + if (((dir1 & 0x0f) > 4) || ((dir1 & 0xf0) == 0x20)) (c->x86_model)++; /* Emulate MTRRs using Cyrix's ARRs. */ - set_bit(X86_FEATURE_CYRIX_ARR, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_CYRIX_ARR); break; case 0xf: /* Cyrix 486 without DEVID registers */ @@ -343,7 +341,8 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) break; } strcpy(buf, Cx86_model[dir0_msn & 7]); - if (p) strcat(buf, p); + if (p) + strcat(buf, p); return; } @@ -352,7 +351,8 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) */ static void __cpuinit init_nsc(struct cpuinfo_x86 *c) { - /* There may be GX1 processors in the wild that are branded + /* + * There may be GX1 processors in the wild that are branded * NSC and not Cyrix. * * This function only handles the GX processor, and kicks every @@ -377,7 +377,7 @@ static void __cpuinit init_nsc(struct cpuinfo_x86 *c) * by the fact that they preserve the flags across the division of 5/2. * PII and PPro exhibit this behavior too, but they have cpuid available. */ - + /* * Perform the Cyrix 5/2 test. A Cyrix won't change * the flags, while other 486 chips will. @@ -398,27 +398,26 @@ static inline int test_cyrix_52div(void) return (unsigned char) (test >> 8) == 0x02; } -static void __cpuinit cyrix_identify(struct cpuinfo_x86 * c) +static void __cpuinit cyrix_identify(struct cpuinfo_x86 *c) { /* Detect Cyrix with disabled CPUID */ - if ( c->x86 == 4 && test_cyrix_52div() ) { + if (c->x86 == 4 && test_cyrix_52div()) { unsigned char dir0, dir1; - + strcpy(c->x86_vendor_id, "CyrixInstead"); - c->x86_vendor = X86_VENDOR_CYRIX; - - /* Actually enable cpuid on the older cyrix */ - - /* Retrieve CPU revisions */ - + c->x86_vendor = X86_VENDOR_CYRIX; + + /* Actually enable cpuid on the older cyrix */ + + /* Retrieve CPU revisions */ + do_cyrix_devid(&dir0, &dir1); - dir0>>=4; - + dir0 >>= 4; + /* Check it is an affected model */ - - if (dir0 == 5 || dir0 == 3) - { + + if (dir0 == 5 || dir0 == 3) { unsigned char ccr3; unsigned long flags; printk(KERN_INFO "Enabling CPUID on Cyrix processor.\n"); @@ -434,26 +433,17 @@ static void __cpuinit cyrix_identify(struct cpuinfo_x86 * c) static struct cpu_dev cyrix_cpu_dev __cpuinitdata = { .c_vendor = "Cyrix", - .c_ident = { "CyrixInstead" }, + .c_ident = { "CyrixInstead" }, .c_init = init_cyrix, .c_identify = cyrix_identify, }; -int __init cyrix_init_cpu(void) -{ - cpu_devs[X86_VENDOR_CYRIX] = &cyrix_cpu_dev; - return 0; -} +cpu_vendor_dev_register(X86_VENDOR_CYRIX, &cyrix_cpu_dev); static struct cpu_dev nsc_cpu_dev __cpuinitdata = { .c_vendor = "NSC", - .c_ident = { "Geode by NSC" }, + .c_ident = { "Geode by NSC" }, .c_init = init_nsc, }; -int __init nsc_init_cpu(void) -{ - cpu_devs[X86_VENDOR_NSC] = &nsc_cpu_dev; - return 0; -} - +cpu_vendor_dev_register(X86_VENDOR_NSC, &nsc_cpu_dev); diff --git a/arch/x86/kernel/cpu/feature_names.c b/arch/x86/kernel/cpu/feature_names.c index ee975ac6bbc..e43ad4ad4cb 100644 --- a/arch/x86/kernel/cpu/feature_names.c +++ b/arch/x86/kernel/cpu/feature_names.c @@ -4,7 +4,7 @@ * This file must not contain any executable code. */ -#include "asm/cpufeature.h" +#include <asm/cpufeature.h> /* * These flag bits must match the definitions in <asm/cpufeature.h>. diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index fae31ce747b..fe9224c51d3 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -30,7 +30,7 @@ struct movsl_mask movsl_mask __read_mostly; #endif -void __cpuinit early_init_intel(struct cpuinfo_x86 *c) +static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) { /* Netburst reports 64 bytes clflush size, but does IO in 128 bytes */ if (c->x86 == 15 && c->x86_cache_alignment == 64) @@ -45,7 +45,7 @@ void __cpuinit early_init_intel(struct cpuinfo_x86 *c) * * This is called before we do cpu ident work */ - + int __cpuinit ppro_with_ram_bug(void) { /* Uses data from early_cpu_detect now */ @@ -58,7 +58,7 @@ int __cpuinit ppro_with_ram_bug(void) } return 0; } - + /* * P4 Xeon errata 037 workaround. @@ -69,7 +69,7 @@ static void __cpuinit Intel_errata_workarounds(struct cpuinfo_x86 *c) unsigned long lo, hi; if ((c->x86 == 15) && (c->x86_model == 1) && (c->x86_mask == 1)) { - rdmsr (MSR_IA32_MISC_ENABLE, lo, hi); + rdmsr(MSR_IA32_MISC_ENABLE, lo, hi); if ((lo & (1<<9)) == 0) { printk (KERN_INFO "CPU: C0 stepping P4 Xeon detected.\n"); printk (KERN_INFO "CPU: Disabling hardware prefetching (Errata 037)\n"); @@ -127,10 +127,10 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) */ c->f00f_bug = 0; if (!paravirt_enabled() && c->x86 == 5) { - static int f00f_workaround_enabled = 0; + static int f00f_workaround_enabled; c->f00f_bug = 1; - if ( !f00f_workaround_enabled ) { + if (!f00f_workaround_enabled) { trap_init_f00f_bug(); printk(KERN_NOTICE "Intel Pentium with F0 0F bug - workaround enabled.\n"); f00f_workaround_enabled = 1; @@ -139,20 +139,22 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) #endif l2 = init_intel_cacheinfo(c); - if (c->cpuid_level > 9 ) { + 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); + set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON); } /* 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) - clear_bit(X86_FEATURE_SEP, c->x86_capability); + clear_cpu_cap(c, X86_FEATURE_SEP); - /* Names for the Pentium II/Celeron processors - detectable only by also checking the cache size. - Dixon is NOT a Celeron. */ + /* + * Names for the Pentium II/Celeron processors + * detectable only by also checking the cache size. + * Dixon is NOT a Celeron. + */ if (c->x86 == 6) { switch (c->x86_model) { case 5: @@ -163,14 +165,14 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) p = "Mobile Pentium II (Dixon)"; } break; - + case 6: if (l2 == 128) p = "Celeron (Mendocino)"; else if (c->x86_mask == 0 || c->x86_mask == 5) p = "Celeron-A"; break; - + case 8: if (l2 == 128) p = "Celeron (Coppermine)"; @@ -178,9 +180,9 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) } } - if ( p ) + if (p) strcpy(c->x86_model_id, p); - + c->x86_max_cores = num_cpu_cores(c); detect_ht(c); @@ -207,28 +209,29 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) #endif if (cpu_has_xmm2) - set_bit(X86_FEATURE_LFENCE_RDTSC, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); if (c->x86 == 15) { - set_bit(X86_FEATURE_P4, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_P4); } - if (c->x86 == 6) - set_bit(X86_FEATURE_P3, c->x86_capability); + if (c->x86 == 6) + set_cpu_cap(c, X86_FEATURE_P3); if (cpu_has_ds) { unsigned int l1; rdmsr(MSR_IA32_MISC_ENABLE, l1, l2); if (!(l1 & (1<<11))) - set_bit(X86_FEATURE_BTS, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_BTS); if (!(l1 & (1<<12))) - set_bit(X86_FEATURE_PEBS, c->x86_capability); + set_cpu_cap(c, X86_FEATURE_PEBS); } if (cpu_has_bts) ds_init_intel(c); } -static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 * c, unsigned int size) +static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 *c, unsigned int size) { - /* Intel PIII Tualatin. This comes in two flavours. + /* + * Intel PIII Tualatin. This comes in two flavours. * One has 256kb of cache, the other 512. We have no way * to determine which, so we use a boottime override * for the 512kb model, and assume 256 otherwise. @@ -240,42 +243,42 @@ static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 * c, unsigned static struct cpu_dev intel_cpu_dev __cpuinitdata = { .c_vendor = "Intel", - .c_ident = { "GenuineIntel" }, + .c_ident = { "GenuineIntel" }, .c_models = { - { .vendor = X86_VENDOR_INTEL, .family = 4, .model_names = - { - [0] = "486 DX-25/33", - [1] = "486 DX-50", - [2] = "486 SX", - [3] = "486 DX/2", - [4] = "486 SL", - [5] = "486 SX/2", - [7] = "486 DX/2-WB", - [8] = "486 DX/4", + { .vendor = X86_VENDOR_INTEL, .family = 4, .model_names = + { + [0] = "486 DX-25/33", + [1] = "486 DX-50", + [2] = "486 SX", + [3] = "486 DX/2", + [4] = "486 SL", + [5] = "486 SX/2", + [7] = "486 DX/2-WB", + [8] = "486 DX/4", [9] = "486 DX/4-WB" } }, { .vendor = X86_VENDOR_INTEL, .family = 5, .model_names = - { - [0] = "Pentium 60/66 A-step", - [1] = "Pentium 60/66", + { + [0] = "Pentium 60/66 A-step", + [1] = "Pentium 60/66", [2] = "Pentium 75 - 200", - [3] = "OverDrive PODP5V83", + [3] = "OverDrive PODP5V83", [4] = "Pentium MMX", - [7] = "Mobile Pentium 75 - 200", + [7] = "Mobile Pentium 75 - 200", [8] = "Mobile Pentium MMX" } }, { .vendor = X86_VENDOR_INTEL, .family = 6, .model_names = - { + { [0] = "Pentium Pro A-step", - [1] = "Pentium Pro", - [3] = "Pentium II (Klamath)", - [4] = "Pentium II (Deschutes)", - [5] = "Pentium II (Deschutes)", + [1] = "Pentium Pro", + [3] = "Pentium II (Klamath)", + [4] = "Pentium II (Deschutes)", + [5] = "Pentium II (Deschutes)", [6] = "Mobile Pentium II", - [7] = "Pentium III (Katmai)", - [8] = "Pentium III (Coppermine)", + [7] = "Pentium III (Katmai)", + [8] = "Pentium III (Coppermine)", [10] = "Pentium III (Cascades)", [11] = "Pentium III (Tualatin)", } @@ -290,15 +293,12 @@ static struct cpu_dev intel_cpu_dev __cpuinitdata = { } }, }, + .c_early_init = early_init_intel, .c_init = init_intel, .c_size_cache = intel_size_cache, }; -__init int intel_cpu_init(void) -{ - cpu_devs[X86_VENDOR_INTEL] = &intel_cpu_dev; - return 0; -} +cpu_vendor_dev_register(X86_VENDOR_INTEL, &intel_cpu_dev); #ifndef CONFIG_X86_CMPXCHG unsigned long cmpxchg_386_u8(volatile void *ptr, u8 old, u8 new) @@ -364,5 +364,5 @@ unsigned long long cmpxchg_486_u64(volatile void *ptr, u64 old, u64 new) EXPORT_SYMBOL(cmpxchg_486_u64); #endif -// arch_initcall(intel_cpu_init); +/* arch_initcall(intel_cpu_init); */ diff --git a/arch/x86/kernel/cpu/mcheck/mce_32.c b/arch/x86/kernel/cpu/mcheck/mce_32.c index a5182dcd94a..774d87cfd8c 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_32.c +++ b/arch/x86/kernel/cpu/mcheck/mce_32.c @@ -10,20 +10,20 @@ #include <linux/smp.h> #include <linux/thread_info.h> -#include <asm/processor.h> +#include <asm/processor.h> #include <asm/system.h> #include <asm/mce.h> #include "mce.h" -int mce_disabled = 0; +int mce_disabled; int nr_mce_banks; EXPORT_SYMBOL_GPL(nr_mce_banks); /* non-fatal.o */ /* Handle unconfigured int18 (should never happen) */ -static void unexpected_machine_check(struct pt_regs * regs, long error_code) -{ +static void unexpected_machine_check(struct pt_regs *regs, long error_code) +{ printk(KERN_ERR "CPU#%d: Unexpected int18 (Machine Check).\n", smp_processor_id()); } @@ -33,30 +33,30 @@ void (*machine_check_vector)(struct pt_regs *, long error_code) = unexpected_mac /* This has to be run for each processor */ void mcheck_init(struct cpuinfo_x86 *c) { - if (mce_disabled==1) + if (mce_disabled == 1) return; switch (c->x86_vendor) { - case X86_VENDOR_AMD: - amd_mcheck_init(c); - break; - - case X86_VENDOR_INTEL: - if (c->x86==5) - intel_p5_mcheck_init(c); - if (c->x86==6) - intel_p6_mcheck_init(c); - if (c->x86==15) - intel_p4_mcheck_init(c); - break; - - case X86_VENDOR_CENTAUR: - if (c->x86==5) - winchip_mcheck_init(c); - break; - - default: - break; + case X86_VENDOR_AMD: + amd_mcheck_init(c); + break; + + case X86_VENDOR_INTEL: + if (c->x86 == 5) + intel_p5_mcheck_init(c); + if (c->x86 == 6) + intel_p6_mcheck_init(c); + if (c->x86 == 15) + intel_p4_mcheck_init(c); + break; + + case X86_VENDOR_CENTAUR: + if (c->x86 == 5) + winchip_mcheck_init(c); + break; + + default: + break; } } diff --git a/arch/x86/kernel/cpu/mcheck/non-fatal.c b/arch/x86/kernel/cpu/mcheck/non-fatal.c index bf39409b383..00ccb6c14ec 100644 --- a/arch/x86/kernel/cpu/mcheck/non-fatal.c +++ b/arch/x86/kernel/cpu/mcheck/non-fatal.c @@ -16,7 +16,7 @@ #include <linux/smp.h> #include <linux/module.h> -#include <asm/processor.h> +#include <asm/processor.h> #include <asm/system.h> #include <asm/msr.h> @@ -26,23 +26,26 @@ static int firstbank; #define MCE_RATE 15*HZ /* timer rate is 15s */ -static void mce_checkregs (void *info) +static void mce_checkregs(void *info) { u32 low, high; int i; - for (i=firstbank; i<nr_mce_banks; i++) { - rdmsr (MSR_IA32_MC0_STATUS+i*4, low, high); + for (i = firstbank; i < nr_mce_banks; i++) { + rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high); if (high & (1<<31)) { printk(KERN_INFO "MCE: The hardware reports a non " "fatal, correctable incident occurred on " "CPU %d.\n", smp_processor_id()); - printk (KERN_INFO "Bank %d: %08x%08x\n", i, high, low); + printk(KERN_INFO "Bank %d: %08x%08x\n", i, high, low); - /* Scrub the error so we don't pick it up in MCE_RATE seconds time. */ - wrmsr (MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL); + /* + * Scrub the error so we don't pick it up in MCE_RATE + * seconds time. + */ + wrmsr(MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL); /* Serialize */ wmb(); @@ -55,10 +58,10 @@ static void mce_work_fn(struct work_struct *work); static DECLARE_DELAYED_WORK(mce_work, mce_work_fn); static void mce_work_fn(struct work_struct *work) -{ +{ on_each_cpu(mce_checkregs, NULL, 1, 1); schedule_delayed_work(&mce_work, round_jiffies_relative(MCE_RATE)); -} +} static int __init init_nonfatal_mce_checker(void) { diff --git a/arch/x86/kernel/cpu/mcheck/p5.c b/arch/x86/kernel/cpu/mcheck/p5.c index a18310aaae0..bfa5817afdd 100644 --- a/arch/x86/kernel/cpu/mcheck/p5.c +++ b/arch/x86/kernel/cpu/mcheck/p5.c @@ -9,20 +9,20 @@ #include <linux/interrupt.h> #include <linux/smp.h> -#include <asm/processor.h> +#include <asm/processor.h> #include <asm/system.h> #include <asm/msr.h> #include "mce.h" /* Machine check handler for Pentium class Intel */ -static void pentium_machine_check(struct pt_regs * regs, long error_code) +static void pentium_machine_check(struct pt_regs *regs, long error_code) { u32 loaddr, hi, lotype; rdmsr(MSR_IA32_P5_MC_ADDR, loaddr, hi); rdmsr(MSR_IA32_P5_MC_TYPE, lotype, hi); printk(KERN_EMERG "CPU#%d: Machine Check Exception: 0x%8X (type 0x%8X).\n", smp_processor_id(), loaddr, lotype); - if(lotype&(1<<5)) + if (lotype&(1<<5)) printk(KERN_EMERG "CPU#%d: Possible thermal failure (CPU on fire ?).\n", smp_processor_id()); add_taint(TAINT_MACHINE_CHECK); } @@ -31,13 +31,13 @@ static void pentium_machine_check(struct pt_regs * regs, long error_code) void intel_p5_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; - + /*Check for MCE support */ - if( !cpu_has(c, X86_FEATURE_MCE) ) - return; + if (!cpu_has(c, X86_FEATURE_MCE)) + return; /* Default P5 to off as its often misconnected */ - if(mce_disabled != -1) + if (mce_disabled != -1) return; machine_check_vector = pentium_machine_check; wmb(); @@ -47,7 +47,7 @@ void intel_p5_mcheck_init(struct cpuinfo_x86 *c) rdmsr(MSR_IA32_P5_MC_TYPE, l, h); printk(KERN_INFO "Intel old style machine check architecture supported.\n"); - /* Enable MCE */ + /* Enable MCE */ set_in_cr4(X86_CR4_MCE); printk(KERN_INFO "Intel old style machine check reporting enabled on CPU#%d.\n", smp_processor_id()); } diff --git a/arch/x86/kernel/cpu/mcheck/p6.c b/arch/x86/kernel/cpu/mcheck/p6.c index 74342604d30..62efc9c2b3a 100644 --- a/arch/x86/kernel/cpu/mcheck/p6.c +++ b/arch/x86/kernel/cpu/mcheck/p6.c @@ -9,23 +9,23 @@ #include <linux/interrupt.h> #include <linux/smp.h> -#include <asm/processor.h> +#include <asm/processor.h> #include <asm/system.h> #include <asm/msr.h> #include "mce.h" /* Machine Check Handler For PII/PIII */ -static void intel_machine_check(struct pt_regs * regs, long error_code) +static void intel_machine_check(struct pt_regs *regs, long error_code) { - int recover=1; + int recover = 1; u32 alow, ahigh, high, low; u32 mcgstl, mcgsth; int i; - rdmsr (MSR_IA32_MCG_STATUS, mcgstl, mcgsth); + rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); if (mcgstl & (1<<0)) /* Recoverable ? */ - recover=0; + recover = 0; printk(KERN_EMERG "CPU %d: Machine Check Exception: %08x%08x\n", smp_processor_id(), mcgsth, mcgstl); @@ -55,30 +55,30 @@ static void intel_machine_check(struct pt_regs * regs, long error_code) } if (recover & 2) - panic ("CPU context corrupt"); + panic("CPU context corrupt"); if (recover & 1) - panic ("Unable to continue"); + panic("Unable to continue"); - printk (KERN_EMERG "Attempting to continue.\n"); - /* - * Do not clear the MSR_IA32_MCi_STATUS if the error is not + printk(KERN_EMERG "Attempting to continue.\n"); + /* + * Do not clear the MSR_IA32_MCi_STATUS if the error is not * recoverable/continuable.This will allow BIOS to look at the MSRs * for errors if the OS could not log the error. */ - for (i=0; i<nr_mce_banks; i++) { + for (i = 0; i < nr_mce_banks; i++) { unsigned int msr; msr = MSR_IA32_MC0_STATUS+i*4; - rdmsr (msr,low, high); + rdmsr(msr, low, high); if (high & (1<<31)) { /* Clear it */ - wrmsr (msr, 0UL, 0UL); + wrmsr(msr, 0UL, 0UL); /* Serialize */ wmb(); add_taint(TAINT_MACHINE_CHECK); } } mcgstl &= ~(1<<2); - wrmsr (MSR_IA32_MCG_STATUS,mcgstl, mcgsth); + wrmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); } /* Set up machine check reporting for processors with Intel style MCE */ @@ -86,21 +86,21 @@ void intel_p6_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; int i; - + /* Check for MCE support */ if (!cpu_has(c, X86_FEATURE_MCE)) return; /* Check for PPro style MCA */ - if (!cpu_has(c, X86_FEATURE_MCA)) + if (!cpu_has(c, X86_FEATURE_MCA)) return; /* Ok machine check is available */ machine_check_vector = intel_machine_check; wmb(); - printk (KERN_INFO "Intel machine check architecture supported.\n"); - rdmsr (MSR_IA32_MCG_CAP, l, h); + printk(KERN_INFO "Intel machine check architecture supported.\n"); + rdmsr(MSR_IA32_MCG_CAP, l, h); if (l & (1<<8)) /* Control register present ? */ wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); nr_mce_banks = l & 0xff; @@ -110,13 +110,13 @@ void intel_p6_mcheck_init(struct cpuinfo_x86 *c) * - MC0_CTL should not be written * - Status registers on all banks should be cleared on reset */ - for (i=1; i<nr_mce_banks; i++) - wrmsr (MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff); + for (i = 1; i < nr_mce_banks; i++) + wrmsr(MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff); - for (i=0; i<nr_mce_banks; i++) - wrmsr (MSR_IA32_MC0_STATUS+4*i, 0x0, 0x0); + for (i = 0; i < nr_mce_banks; i++) + wrmsr(MSR_IA32_MC0_STATUS+4*i, 0x0, 0x0); - set_in_cr4 (X86_CR4_MCE); - printk (KERN_INFO "Intel machine check reporting enabled on CPU#%d.\n", + set_in_cr4(X86_CR4_MCE); + printk(KERN_INFO "Intel machine check reporting enabled on CPU#%d.\n", smp_processor_id()); } diff --git a/arch/x86/kernel/cpu/mcheck/winchip.c b/arch/x86/kernel/cpu/mcheck/winchip.c index 3d428d5afc5..f2be3e190c6 100644 --- a/arch/x86/kernel/cpu/mcheck/winchip.c +++ b/arch/x86/kernel/cpu/mcheck/winchip.c @@ -8,14 +8,14 @@ #include <linux/kernel.h> #include <linux/interrupt.h> -#include <asm/processor.h> +#include <asm/processor.h> #include <asm/system.h> #include <asm/msr.h> #include "mce.h" /* Machine check handler for WinChip C6 */ -static void winchip_machine_check(struct pt_regs * regs, long error_code) +static void winchip_machine_check(struct pt_regs *regs, long error_code) { printk(KERN_EMERG "CPU0: Machine Check Exception.\n"); add_taint(TAINT_MACHINE_CHECK); @@ -28,8 +28,8 @@ void winchip_mcheck_init(struct cpuinfo_x86 *c) machine_check_vector = winchip_machine_check; wmb(); rdmsr(MSR_IDT_FCR1, lo, hi); - lo|= (1<<2); /* Enable EIERRINT (int 18 MCE) */ - lo&= ~(1<<4); /* Enable MCE */ + lo |= (1<<2); /* Enable EIERRINT (int 18 MCE) */ + lo &= ~(1<<4); /* Enable MCE */ wrmsr(MSR_IDT_FCR1, lo, hi); set_in_cr4(X86_CR4_MCE); printk(KERN_INFO "Winchip machine check reporting enabled on CPU#0.\n"); diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 3e18db4cefe..353efe4f501 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -11,6 +11,7 @@ #include <asm/cpufeature.h> #include <asm/processor-flags.h> #include <asm/tlbflush.h> +#include <asm/pat.h> #include "mtrr.h" struct mtrr_state { @@ -35,6 +36,8 @@ static struct fixed_range_block fixed_range_blocks[] = { static unsigned long smp_changes_mask; static struct mtrr_state mtrr_state = {}; +static int mtrr_state_set; +static u64 tom2; #undef MODULE_PARAM_PREFIX #define MODULE_PARAM_PREFIX "mtrr." @@ -42,6 +45,111 @@ static struct mtrr_state mtrr_state = {}; static int mtrr_show; module_param_named(show, mtrr_show, bool, 0); +/* + * Returns the effective MTRR type for the region + * Error returns: + * - 0xFE - when the range is "not entirely covered" by _any_ var range MTRR + * - 0xFF - when MTRR is not enabled + */ +u8 mtrr_type_lookup(u64 start, u64 end) +{ + int i; + u64 base, mask; + u8 prev_match, curr_match; + + if (!mtrr_state_set) + return 0xFF; + + if (!mtrr_state.enabled) + return 0xFF; + + /* Make end inclusive end, instead of exclusive */ + end--; + + /* Look in fixed ranges. Just return the type as per start */ + if (mtrr_state.have_fixed && (start < 0x100000)) { + int idx; + + if (start < 0x80000) { + idx = 0; + idx += (start >> 16); + return mtrr_state.fixed_ranges[idx]; + } else if (start < 0xC0000) { + idx = 1 * 8; + idx += ((start - 0x80000) >> 14); + return mtrr_state.fixed_ranges[idx]; + } else if (start < 0x1000000) { + idx = 3 * 8; + idx += ((start - 0xC0000) >> 12); + return mtrr_state.fixed_ranges[idx]; + } + } + + /* + * Look in variable ranges + * Look of multiple ranges matching this address and pick type + * as per MTRR precedence + */ + if (!mtrr_state.enabled & 2) { + return mtrr_state.def_type; + } + + prev_match = 0xFF; + for (i = 0; i < num_var_ranges; ++i) { + unsigned short start_state, end_state; + + if (!(mtrr_state.var_ranges[i].mask_lo & (1 << 11))) + continue; + + base = (((u64)mtrr_state.var_ranges[i].base_hi) << 32) + + (mtrr_state.var_ranges[i].base_lo & PAGE_MASK); + mask = (((u64)mtrr_state.var_ranges[i].mask_hi) << 32) + + (mtrr_state.var_ranges[i].mask_lo & PAGE_MASK); + + start_state = ((start & mask) == (base & mask)); + end_state = ((end & mask) == (base & mask)); + if (start_state != end_state) + return 0xFE; + + if ((start & mask) != (base & mask)) { + continue; + } + + curr_match = mtrr_state.var_ranges[i].base_lo & 0xff; + if (prev_match == 0xFF) { + prev_match = curr_match; + continue; + } + + if (prev_match == MTRR_TYPE_UNCACHABLE || + curr_match == MTRR_TYPE_UNCACHABLE) { + return MTRR_TYPE_UNCACHABLE; + } + + if ((prev_match == MTRR_TYPE_WRBACK && + curr_match == MTRR_TYPE_WRTHROUGH) || + (prev_match == MTRR_TYPE_WRTHROUGH && + curr_match == MTRR_TYPE_WRBACK)) { + prev_match = MTRR_TYPE_WRTHROUGH; + curr_match = MTRR_TYPE_WRTHROUGH; + } + + if (prev_match != curr_match) { + return MTRR_TYPE_UNCACHABLE; + } + } + + if (tom2) { + if (start >= (1ULL<<32) && (end < tom2)) + return MTRR_TYPE_WRBACK; + } + + if (prev_match != 0xFF) + return prev_match; + + return mtrr_state.def_type; +} + /* Get the MSR pair relating to a var range */ static void get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr) @@ -79,12 +187,16 @@ static void print_fixed(unsigned base, unsigned step, const mtrr_type*types) base, base + step - 1, mtrr_attrib_to_str(*types)); } +static void prepare_set(void); +static void post_set(void); + /* Grab all of the MTRR state for this CPU into *state */ void __init get_mtrr_state(void) { unsigned int i; struct mtrr_var_range *vrs; unsigned lo, dummy; + unsigned long flags; vrs = mtrr_state.var_ranges; @@ -100,6 +212,15 @@ void __init get_mtrr_state(void) mtrr_state.def_type = (lo & 0xff); mtrr_state.enabled = (lo & 0xc00) >> 10; + if (amd_special_default_mtrr()) { + unsigned lo, hi; + /* TOP_MEM2 */ + rdmsr(MSR_K8_TOP_MEM2, lo, hi); + tom2 = hi; + tom2 <<= 32; + tom2 |= lo; + tom2 &= 0xffffff8000000ULL; + } if (mtrr_show) { int high_width; @@ -130,7 +251,22 @@ void __init get_mtrr_state(void) else printk(KERN_INFO "MTRR %u disabled\n", i); } + if (tom2) { + printk(KERN_INFO "TOM2: %016llx aka %lldM\n", + tom2, tom2>>20); + } } + mtrr_state_set = 1; + + /* PAT setup for BP. We need to go through sync steps here */ + local_irq_save(flags); + prepare_set(); + + pat_init(); + + post_set(); + local_irq_restore(flags); + } /* Some BIOS's are fucked and don't set all MTRRs the same! */ @@ -397,6 +533,9 @@ static void generic_set_all(void) /* Actually set the state */ mask = set_mtrr_state(); + /* also set PAT */ + pat_init(); + post_set(); local_irq_restore(flags); diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c index 91e150acb46..1960f1985e5 100644 --- a/arch/x86/kernel/cpu/mtrr/if.c +++ b/arch/x86/kernel/cpu/mtrr/if.c @@ -424,11 +424,10 @@ static int __init mtrr_if_init(void) return -ENODEV; proc_root_mtrr = - create_proc_entry("mtrr", S_IWUSR | S_IRUGO, &proc_root); - if (proc_root_mtrr) { + proc_create("mtrr", S_IWUSR | S_IRUGO, &proc_root, &mtrr_fops); + + if (proc_root_mtrr) proc_root_mtrr->owner = THIS_MODULE; - proc_root_mtrr->proc_fops = &mtrr_fops; - } return 0; } diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index a6450b3ae75..6a1e278d932 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -627,7 +627,7 @@ early_param("disable_mtrr_trim", disable_mtrr_trim_setup); #define Tom2Enabled (1U << 21) #define Tom2ForceMemTypeWB (1U << 22) -static __init int amd_special_default_mtrr(void) +int __init amd_special_default_mtrr(void) { u32 l, h; diff --git a/arch/x86/kernel/cpu/mtrr/state.c b/arch/x86/kernel/cpu/mtrr/state.c index 9f8ba923d1c..7f7e2753685 100644 --- a/arch/x86/kernel/cpu/mtrr/state.c +++ b/arch/x86/kernel/cpu/mtrr/state.c @@ -19,13 +19,15 @@ void set_mtrr_prepare_save(struct set_mtrr_context *ctxt) if (use_intel() || is_cpu(CYRIX)) { /* Save value of CR4 and clear Page Global Enable (bit 7) */ - if ( cpu_has_pge ) { + if (cpu_has_pge) { ctxt->cr4val = read_cr4(); write_cr4(ctxt->cr4val & ~X86_CR4_PGE); } - /* Disable and flush caches. Note that wbinvd flushes the TLBs as - a side-effect */ + /* + * Disable and flush caches. Note that wbinvd flushes the TLBs + * as a side-effect + */ cr0 = read_cr0() | X86_CR0_CD; wbinvd(); write_cr0(cr0); @@ -42,7 +44,7 @@ void set_mtrr_prepare_save(struct set_mtrr_context *ctxt) void set_mtrr_cache_disable(struct set_mtrr_context *ctxt) { - if (use_intel()) + if (use_intel()) /* Disable MTRRs, and set the default type to uncached */ mtrr_wrmsr(MTRRdefType_MSR, ctxt->deftype_lo & 0xf300UL, ctxt->deftype_hi); @@ -66,12 +68,12 @@ void set_mtrr_done(struct set_mtrr_context *ctxt) else /* Cyrix ARRs - everything else was excluded at the top */ setCx86(CX86_CCR3, ctxt->ccr3); - + /* Enable caches */ write_cr0(read_cr0() & 0xbfffffff); /* Restore value of CR4 */ - if ( cpu_has_pge ) + if (cpu_has_pge) write_cr4(ctxt->cr4val); } /* Re-enable interrupts locally (if enabled previously) */ diff --git a/arch/x86/kernel/cpu/nexgen.c b/arch/x86/kernel/cpu/nexgen.c index 961fbe1a748..5d5e1c13412 100644 --- a/arch/x86/kernel/cpu/nexgen.c +++ b/arch/x86/kernel/cpu/nexgen.c @@ -9,11 +9,11 @@ * Detect a NexGen CPU running without BIOS hypercode new enough * to have CPUID. (Thanks to Herbert Oppmann) */ - + static int __cpuinit deep_magic_nexgen_probe(void) { int ret; - + __asm__ __volatile__ ( " movw $0x5555, %%ax\n" " xorw %%dx,%%dx\n" @@ -22,22 +22,21 @@ static int __cpuinit deep_magic_nexgen_probe(void) " movl $0, %%eax\n" " jnz 1f\n" " movl $1, %%eax\n" - "1:\n" - : "=a" (ret) : : "cx", "dx" ); + "1:\n" + : "=a" (ret) : : "cx", "dx"); return ret; } -static void __cpuinit init_nexgen(struct cpuinfo_x86 * c) +static void __cpuinit init_nexgen(struct cpuinfo_x86 *c) { c->x86_cache_size = 256; /* A few had 1 MB... */ } -static void __cpuinit nexgen_identify(struct cpuinfo_x86 * c) +static void __cpuinit nexgen_identify(struct cpuinfo_x86 *c) { /* Detect NexGen with old hypercode */ - if ( deep_magic_nexgen_probe() ) { + if (deep_magic_nexgen_probe()) strcpy(c->x86_vendor_id, "NexGenDriven"); - } } static struct cpu_dev nexgen_cpu_dev __cpuinitdata = { diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c index af11d31dce0..0978a4a3941 100644 --- a/arch/x86/kernel/cpu/proc.c +++ b/arch/x86/kernel/cpu/proc.c @@ -8,78 +8,139 @@ /* * Get CPU information for use by the procfs. */ +#ifdef CONFIG_X86_32 +static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c, + unsigned int cpu) +{ +#ifdef CONFIG_X86_HT + if (c->x86_max_cores * smp_num_siblings > 1) { + seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); + seq_printf(m, "siblings\t: %d\n", + cpus_weight(per_cpu(cpu_core_map, cpu))); + seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id); + seq_printf(m, "cpu cores\t: %d\n", c->booted_cores); + seq_printf(m, "apicid\t\t: %d\n", c->apicid); + seq_printf(m, "initial apicid\t: %d\n", c->initial_apicid); + } +#endif +} + +static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c) +{ + /* + * We use exception 16 if we have hardware math and we've either seen + * it or the CPU claims it is internal + */ + int fpu_exception = c->hard_math && (ignore_fpu_irq || cpu_has_fpu); + seq_printf(m, + "fdiv_bug\t: %s\n" + "hlt_bug\t\t: %s\n" + "f00f_bug\t: %s\n" + "coma_bug\t: %s\n" + "fpu\t\t: %s\n" + "fpu_exception\t: %s\n" + "cpuid level\t: %d\n" + "wp\t\t: %s\n", + c->fdiv_bug ? "yes" : "no", + c->hlt_works_ok ? "no" : "yes", + c->f00f_bug ? "yes" : "no", + c->coma_bug ? "yes" : "no", + c->hard_math ? "yes" : "no", + fpu_exception ? "yes" : "no", + c->cpuid_level, + c->wp_works_ok ? "yes" : "no"); +} +#else +static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c, + unsigned int cpu) +{ +#ifdef CONFIG_SMP + if (c->x86_max_cores * smp_num_siblings > 1) { + seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); + seq_printf(m, "siblings\t: %d\n", + cpus_weight(per_cpu(cpu_core_map, cpu))); + seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id); + seq_printf(m, "cpu cores\t: %d\n", c->booted_cores); + seq_printf(m, "apicid\t\t: %d\n", c->apicid); + seq_printf(m, "initial apicid\t: %d\n", c->initial_apicid); + } +#endif +} + +static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c) +{ + seq_printf(m, + "fpu\t\t: yes\n" + "fpu_exception\t: yes\n" + "cpuid level\t: %d\n" + "wp\t\t: yes\n", + c->cpuid_level); +} +#endif + static int show_cpuinfo(struct seq_file *m, void *v) { struct cpuinfo_x86 *c = v; - int i, n = 0; - int fpu_exception; + unsigned int cpu = 0; + int i; #ifdef CONFIG_SMP - n = c->cpu_index; + cpu = c->cpu_index; #endif - seq_printf(m, "processor\t: %d\n" - "vendor_id\t: %s\n" - "cpu family\t: %d\n" - "model\t\t: %d\n" - "model name\t: %s\n", - n, - c->x86_vendor_id[0] ? c->x86_vendor_id : "unknown", - c->x86, - c->x86_model, - c->x86_model_id[0] ? c->x86_model_id : "unknown"); + seq_printf(m, "processor\t: %u\n" + "vendor_id\t: %s\n" + "cpu family\t: %d\n" + "model\t\t: %u\n" + "model name\t: %s\n", + cpu, + c->x86_vendor_id[0] ? c->x86_vendor_id : "unknown", + c->x86, + c->x86_model, + c->x86_model_id[0] ? c->x86_model_id : "unknown"); if (c->x86_mask || c->cpuid_level >= 0) seq_printf(m, "stepping\t: %d\n", c->x86_mask); else seq_printf(m, "stepping\t: unknown\n"); - if ( cpu_has(c, X86_FEATURE_TSC) ) { - unsigned int freq = cpufreq_quick_get(n); + if (cpu_has(c, X86_FEATURE_TSC)) { + unsigned int freq = cpufreq_quick_get(cpu); + if (!freq) freq = cpu_khz; seq_printf(m, "cpu MHz\t\t: %u.%03u\n", - freq / 1000, (freq % 1000)); + freq / 1000, (freq % 1000)); } /* Cache size */ if (c->x86_cache_size >= 0) 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", c->phys_proc_id); - seq_printf(m, "siblings\t: %d\n", - cpus_weight(per_cpu(cpu_core_map, 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 - - /* We use exception 16 if we have hardware math and we've either seen it or the CPU claims it is internal */ - fpu_exception = c->hard_math && (ignore_fpu_irq || cpu_has_fpu); - seq_printf(m, "fdiv_bug\t: %s\n" - "hlt_bug\t\t: %s\n" - "f00f_bug\t: %s\n" - "coma_bug\t: %s\n" - "fpu\t\t: %s\n" - "fpu_exception\t: %s\n" - "cpuid level\t: %d\n" - "wp\t\t: %s\n" - "flags\t\t:", - c->fdiv_bug ? "yes" : "no", - c->hlt_works_ok ? "no" : "yes", - c->f00f_bug ? "yes" : "no", - c->coma_bug ? "yes" : "no", - c->hard_math ? "yes" : "no", - fpu_exception ? "yes" : "no", - c->cpuid_level, - c->wp_works_ok ? "yes" : "no"); - - for ( i = 0 ; i < 32*NCAPINTS ; i++ ) - if ( test_bit(i, c->x86_capability) && - x86_cap_flags[i] != NULL ) + + show_cpuinfo_core(m, c, cpu); + show_cpuinfo_misc(m, c); + + seq_printf(m, "flags\t\t:"); + for (i = 0; i < 32*NCAPINTS; i++) + if (cpu_has(c, i) && x86_cap_flags[i] != NULL) seq_printf(m, " %s", x86_cap_flags[i]); - for (i = 0; i < 32; i++) + seq_printf(m, "\nbogomips\t: %lu.%02lu\n", + c->loops_per_jiffy/(500000/HZ), + (c->loops_per_jiffy/(5000/HZ)) % 100); + +#ifdef CONFIG_X86_64 + if (c->x86_tlbsize > 0) + seq_printf(m, "TLB size\t: %d 4K pages\n", c->x86_tlbsize); +#endif + seq_printf(m, "clflush size\t: %u\n", c->x86_clflush_size); +#ifdef CONFIG_X86_64 + seq_printf(m, "cache_alignment\t: %d\n", c->x86_cache_alignment); + seq_printf(m, "address sizes\t: %u bits physical, %u bits virtual\n", + c->x86_phys_bits, c->x86_virt_bits); +#endif + + seq_printf(m, "power management:"); + for (i = 0; i < 32; i++) { if (c->x86_power & (1 << i)) { if (i < ARRAY_SIZE(x86_power_flags) && x86_power_flags[i]) @@ -89,11 +150,9 @@ static int show_cpuinfo(struct seq_file *m, void *v) else seq_printf(m, " [%d]", i); } + } - seq_printf(m, "\nbogomips\t: %lu.%02lu\n", - c->loops_per_jiffy/(500000/HZ), - (c->loops_per_jiffy/(5000/HZ)) % 100); - seq_printf(m, "clflush size\t: %u\n\n", c->x86_clflush_size); + seq_printf(m, "\n\n"); return 0; } @@ -106,14 +165,17 @@ static void *c_start(struct seq_file *m, loff_t *pos) return &cpu_data(*pos); return NULL; } + static void *c_next(struct seq_file *m, void *v, loff_t *pos) { *pos = next_cpu(*pos, cpu_online_map); return c_start(m, pos); } + static void c_stop(struct seq_file *m, void *v) { } + const struct seq_operations cpuinfo_op = { .start = c_start, .next = c_next, diff --git a/arch/x86/kernel/cpu/transmeta.c b/arch/x86/kernel/cpu/transmeta.c index e8b422c1c51..b911a2c61b8 100644 --- a/arch/x86/kernel/cpu/transmeta.c +++ b/arch/x86/kernel/cpu/transmeta.c @@ -18,8 +18,8 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c) /* Print CMS and CPU revision */ max = cpuid_eax(0x80860000); cpu_rev = 0; - if ( max >= 0x80860001 ) { - cpuid(0x80860001, &dummy, &cpu_rev, &cpu_freq, &cpu_flags); + if (max >= 0x80860001) { + cpuid(0x80860001, &dummy, &cpu_rev, &cpu_freq, &cpu_flags); if (cpu_rev != 0x02000000) { printk(KERN_INFO "CPU: Processor revision %u.%u.%u.%u, %u MHz\n", (cpu_rev >> 24) & 0xff, @@ -29,7 +29,7 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c) cpu_freq); } } - if ( max >= 0x80860002 ) { + if (max >= 0x80860002) { cpuid(0x80860002, &new_cpu_rev, &cms_rev1, &cms_rev2, &dummy); if (cpu_rev == 0x02000000) { printk(KERN_INFO "CPU: Processor revision %08X, %u MHz\n", @@ -42,7 +42,7 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c) cms_rev1 & 0xff, cms_rev2); } - if ( max >= 0x80860006 ) { + if (max >= 0x80860006) { cpuid(0x80860003, (void *)&cpu_info[0], (void *)&cpu_info[4], @@ -74,23 +74,25 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c) wrmsr(0x80860004, cap_mask, uk); /* All Transmeta CPUs have a constant TSC */ - set_bit(X86_FEATURE_CONSTANT_TSC, c->x86_capability); - + set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); + #ifdef CONFIG_SYSCTL - /* randomize_va_space slows us down enormously; - it probably triggers retranslation of x86->native bytecode */ + /* + * randomize_va_space slows us down enormously; + * it probably triggers retranslation of x86->native bytecode + */ randomize_va_space = 0; #endif } -static void __cpuinit transmeta_identify(struct cpuinfo_x86 * c) +static void __cpuinit transmeta_identify(struct cpuinfo_x86 *c) { u32 xlvl; /* Transmeta-defined flags: level 0x80860001 */ xlvl = cpuid_eax(0x80860000); - if ( (xlvl & 0xffff0000) == 0x80860000 ) { - if ( xlvl >= 0x80860001 ) + if ((xlvl & 0xffff0000) == 0x80860000) { + if (xlvl >= 0x80860001) c->x86_capability[2] = cpuid_edx(0x80860001); } } @@ -102,8 +104,4 @@ static struct cpu_dev transmeta_cpu_dev __cpuinitdata = { .c_identify = transmeta_identify, }; -int __init transmeta_init_cpu(void) -{ - cpu_devs[X86_VENDOR_TRANSMETA] = &transmeta_cpu_dev; - return 0; -} +cpu_vendor_dev_register(X86_VENDOR_TRANSMETA, &transmeta_cpu_dev); diff --git a/arch/x86/kernel/cpu/umc.c b/arch/x86/kernel/cpu/umc.c index a7a4e75bdcd..b1fc90989d7 100644 --- a/arch/x86/kernel/cpu/umc.c +++ b/arch/x86/kernel/cpu/umc.c @@ -3,24 +3,23 @@ #include <asm/processor.h> #include "cpu.h" -/* UMC chips appear to be only either 386 or 486, so no special init takes place. +/* + * UMC chips appear to be only either 386 or 486, + * so no special init takes place. */ static struct cpu_dev umc_cpu_dev __cpuinitdata = { .c_vendor = "UMC", - .c_ident = { "UMC UMC UMC" }, + .c_ident = { "UMC UMC UMC" }, .c_models = { { .vendor = X86_VENDOR_UMC, .family = 4, .model_names = - { - [1] = "U5D", - [2] = "U5S", + { + [1] = "U5D", + [2] = "U5S", } }, }, }; -int __init umc_init_cpu(void) -{ - cpu_devs[X86_VENDOR_UMC] = &umc_cpu_dev; - return 0; -} +cpu_vendor_dev_register(X86_VENDOR_UMC, &umc_cpu_dev); + diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index 9a5fa0abfcc..2251d0ae957 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -26,11 +26,7 @@ #include <linux/kdebug.h> #include <asm/smp.h> -#ifdef CONFIG_X86_32 #include <mach_ipi.h> -#else -#include <asm/mach_apic.h> -#endif /* This keeps a track of which one is crashing cpu. */ static int crashing_cpu; diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index dcd918c1580..11c11b8ec48 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -220,11 +220,11 @@ int ds_allocate(void **dsp, size_t bts_size_in_bytes) int ds_free(void **dsp) { - if (*dsp) + if (*dsp) { kfree((void *)get_bts_buffer_base(*dsp)); - kfree(*dsp); - *dsp = NULL; - + kfree(*dsp); + *dsp = NULL; + } return 0; } diff --git a/arch/x86/kernel/e820_32.c b/arch/x86/kernel/e820_32.c index 80444c5c9b1..0240cd77836 100644 --- a/arch/x86/kernel/e820_32.c +++ b/arch/x86/kernel/e820_32.c @@ -450,38 +450,25 @@ int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) * thinkpad 560x, for example, does not cooperate with the memory * detection code.) */ -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) return -1; do { - unsigned long long start = biosmap->addr; - unsigned long long size = biosmap->size; - unsigned long long end = start + size; - unsigned long type = biosmap->type; + u64 start = biosmap->addr; + u64 size = biosmap->size; + u64 end = start + size; + u32 type = biosmap->type; /* Overflow in 64 bits? Ignore the memory map. */ if (start > end) return -1; - /* - * Some BIOSes claim RAM in the 640k - 1M region. - * Not right. Fix it up. - */ - if (type == E820_RAM) { - if (start < 0x100000ULL && end > 0xA0000ULL) { - if (start < 0xA0000ULL) - add_memory_region(start, 0xA0000ULL-start, type); - if (end <= 0x100000ULL) - continue; - start = 0x100000ULL; - size = end - start; - } - } add_memory_region(start, size, type); - } while (biosmap++,--nr_map); + } while (biosmap++, --nr_map); + return 0; } diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c index 9be69712601..7f6c0c85c8f 100644 --- a/arch/x86/kernel/e820_64.c +++ b/arch/x86/kernel/e820_64.c @@ -27,6 +27,7 @@ #include <asm/setup.h> #include <asm/sections.h> #include <asm/kdebug.h> +#include <asm/trampoline.h> struct e820map e820; @@ -36,11 +37,11 @@ struct e820map e820; unsigned long end_pfn; /* - * end_pfn only includes RAM, while end_pfn_map includes all e820 entries. - * The direct mapping extends to end_pfn_map, so that we can directly access + * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. + * The direct mapping extends to max_pfn_mapped, so that we can directly access * apertures, ACPI and other tables without having to play with fixmaps. */ -unsigned long end_pfn_map; +unsigned long max_pfn_mapped; /* * Last pfn which the user wants to use. @@ -58,8 +59,8 @@ struct early_res { }; static struct early_res early_res[MAX_EARLY_RES] __initdata = { { 0, PAGE_SIZE, "BIOS data page" }, /* BIOS data page */ -#ifdef CONFIG_SMP - { SMP_TRAMPOLINE_BASE, SMP_TRAMPOLINE_BASE + 2*PAGE_SIZE, "SMP_TRAMPOLINE" }, +#ifdef CONFIG_X86_TRAMPOLINE + { TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" }, #endif {} }; @@ -95,7 +96,8 @@ void __init early_res_to_bootmem(void) } /* Check for already reserved areas */ -static inline int bad_addr(unsigned long *addrp, unsigned long size) +static inline int +bad_addr(unsigned long *addrp, unsigned long size, unsigned long align) { int i; unsigned long addr = *addrp, last; @@ -105,7 +107,7 @@ again: for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { struct early_res *r = &early_res[i]; if (last >= r->start && addr < r->end) { - *addrp = addr = r->end; + *addrp = addr = round_up(r->end, align); changed = 1; goto again; } @@ -113,6 +115,40 @@ again: return changed; } +/* Check for already reserved areas */ +static inline int +bad_addr_size(unsigned long *addrp, unsigned long *sizep, unsigned long align) +{ + int i; + unsigned long addr = *addrp, last; + unsigned long size = *sizep; + int changed = 0; +again: + last = addr + size; + for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { + struct early_res *r = &early_res[i]; + if (last > r->start && addr < r->start) { + size = r->start - addr; + changed = 1; + goto again; + } + if (last > r->end && addr < r->end) { + addr = round_up(r->end, align); + size = last - addr; + changed = 1; + goto again; + } + if (last <= r->end && addr >= r->start) { + (*sizep)++; + return 0; + } + } + if (changed) { + *addrp = addr; + *sizep = size; + } + return changed; +} /* * This function checks if any part of the range <start,end> is mapped * with type. @@ -174,26 +210,27 @@ int __init e820_all_mapped(unsigned long start, unsigned long end, * Find a free area with specified alignment in a specific range. */ unsigned long __init find_e820_area(unsigned long start, unsigned long end, - unsigned size, unsigned long align) + unsigned long size, unsigned long align) { int i; - unsigned long mask = ~(align - 1); for (i = 0; i < e820.nr_map; i++) { struct e820entry *ei = &e820.map[i]; - unsigned long addr = ei->addr, last; + unsigned long addr, last; + unsigned long ei_last; if (ei->type != E820_RAM) continue; + addr = round_up(ei->addr, align); + ei_last = ei->addr + ei->size; if (addr < start) - addr = start; - if (addr > ei->addr + ei->size) + addr = round_up(start, align); + if (addr >= ei_last) continue; - while (bad_addr(&addr, size) && addr+size <= ei->addr+ei->size) + while (bad_addr(&addr, size, align) && addr+size <= ei_last) ; - addr = (addr + align - 1) & mask; last = addr + size; - if (last > ei->addr + ei->size) + if (last > ei_last) continue; if (last > end) continue; @@ -203,6 +240,40 @@ unsigned long __init find_e820_area(unsigned long start, unsigned long end, } /* + * Find next free range after *start + */ +unsigned long __init find_e820_area_size(unsigned long start, + unsigned long *sizep, + unsigned long align) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + unsigned long addr, last; + unsigned long ei_last; + + if (ei->type != E820_RAM) + continue; + addr = round_up(ei->addr, align); + ei_last = ei->addr + ei->size; + if (addr < start) + addr = round_up(start, align); + if (addr >= ei_last) + continue; + *sizep = ei_last - addr; + while (bad_addr_size(&addr, sizep, align) && + addr + *sizep <= ei_last) + ; + last = addr + *sizep; + if (last > ei_last) + continue; + return addr; + } + return -1UL; + +} +/* * Find the highest page frame number we have available */ unsigned long __init e820_end_of_ram(void) @@ -211,29 +282,29 @@ unsigned long __init e820_end_of_ram(void) end_pfn = find_max_pfn_with_active_regions(); - if (end_pfn > end_pfn_map) - end_pfn_map = end_pfn; - if (end_pfn_map > MAXMEM>>PAGE_SHIFT) - end_pfn_map = MAXMEM>>PAGE_SHIFT; + if (end_pfn > max_pfn_mapped) + max_pfn_mapped = end_pfn; + if (max_pfn_mapped > MAXMEM>>PAGE_SHIFT) + max_pfn_mapped = MAXMEM>>PAGE_SHIFT; if (end_pfn > end_user_pfn) end_pfn = end_user_pfn; - if (end_pfn > end_pfn_map) - end_pfn = end_pfn_map; + if (end_pfn > max_pfn_mapped) + end_pfn = max_pfn_mapped; - printk(KERN_INFO "end_pfn_map = %lu\n", end_pfn_map); + printk(KERN_INFO "max_pfn_mapped = %lu\n", max_pfn_mapped); return end_pfn; } /* * Mark e820 reserved areas as busy for the resource manager. */ -void __init e820_reserve_resources(struct resource *code_resource, - struct resource *data_resource, struct resource *bss_resource) +void __init e820_reserve_resources(void) { int i; + struct resource *res; + + res = alloc_bootmem_low(sizeof(struct resource) * e820.nr_map); for (i = 0; i < e820.nr_map; i++) { - struct resource *res; - res = alloc_bootmem_low(sizeof(struct resource)); switch (e820.map[i].type) { case E820_RAM: res->name = "System RAM"; break; case E820_ACPI: res->name = "ACPI Tables"; break; @@ -243,21 +314,8 @@ void __init e820_reserve_resources(struct resource *code_resource, res->start = e820.map[i].addr; res->end = res->start + e820.map[i].size - 1; res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; - request_resource(&iomem_resource, res); - if (e820.map[i].type == E820_RAM) { - /* - * We don't know which RAM region contains kernel data, - * so we try it repeatedly and let the resource manager - * test it. - */ - request_resource(res, code_resource); - request_resource(res, data_resource); - request_resource(res, bss_resource); -#ifdef CONFIG_KEXEC - if (crashk_res.start != crashk_res.end) - request_resource(res, &crashk_res); -#endif - } + insert_resource(&iomem_resource, res); + res++; } } @@ -309,9 +367,9 @@ static int __init e820_find_active_region(const struct e820entry *ei, if (*ei_startpfn >= *ei_endpfn) return 0; - /* Check if end_pfn_map should be updated */ - if (ei->type != E820_RAM && *ei_endpfn > end_pfn_map) - end_pfn_map = *ei_endpfn; + /* Check if max_pfn_mapped should be updated */ + if (ei->type != E820_RAM && *ei_endpfn > max_pfn_mapped) + max_pfn_mapped = *ei_endpfn; /* Skip if map is outside the node */ if (ei->type != E820_RAM || *ei_endpfn <= start_pfn || @@ -634,10 +692,10 @@ static int __init copy_e820_map(struct e820entry *biosmap, int nr_map) return -1; do { - unsigned long start = biosmap->addr; - unsigned long size = biosmap->size; - unsigned long end = start + size; - unsigned long type = biosmap->type; + u64 start = biosmap->addr; + u64 size = biosmap->size; + u64 end = start + size; + u32 type = biosmap->type; /* Overflow in 64 bits? Ignore the memory map. */ if (start > end) @@ -702,7 +760,7 @@ static int __init parse_memmap_opt(char *p) saved_max_pfn = e820_end_of_ram(); remove_all_active_ranges(); #endif - end_pfn_map = 0; + max_pfn_mapped = 0; e820.nr_map = 0; userdef = 1; return 0; diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c index cff84cd9987..643fd861b72 100644 --- a/arch/x86/kernel/early_printk.c +++ b/arch/x86/kernel/early_printk.c @@ -13,7 +13,7 @@ #define VGABASE (__ISA_IO_base + 0xb8000) static int max_ypos = 25, max_xpos = 80; -static int current_ypos = 25, current_xpos = 0; +static int current_ypos = 25, current_xpos; static void early_vga_write(struct console *con, const char *str, unsigned n) { @@ -108,12 +108,12 @@ static __init void early_serial_init(char *s) if (*s) { unsigned port; - if (!strncmp(s,"0x",2)) { + if (!strncmp(s, "0x", 2)) { early_serial_base = simple_strtoul(s, &e, 16); } else { static int bases[] = { 0x3f8, 0x2f8 }; - if (!strncmp(s,"ttyS",4)) + if (!strncmp(s, "ttyS", 4)) s += 4; port = simple_strtoul(s, &e, 10); if (port > 1 || s == e) @@ -194,7 +194,7 @@ static struct console simnow_console = { /* Direct interface for emergencies */ static struct console *early_console = &early_vga_console; -static int early_console_initialized = 0; +static int early_console_initialized; void early_printk(const char *fmt, ...) { @@ -202,9 +202,9 @@ void early_printk(const char *fmt, ...) int n; va_list ap; - va_start(ap,fmt); - n = vscnprintf(buf,512,fmt,ap); - early_console->write(early_console,buf,n); + va_start(ap, fmt); + n = vscnprintf(buf, 512, fmt, ap); + early_console->write(early_console, buf, n); va_end(ap); } @@ -229,15 +229,15 @@ static int __init setup_early_printk(char *buf) early_serial_init(buf); early_console = &early_serial_console; } else if (!strncmp(buf, "vga", 3) - && boot_params.screen_info.orig_video_isVGA == 1) { + && boot_params.screen_info.orig_video_isVGA == 1) { max_xpos = boot_params.screen_info.orig_video_cols; max_ypos = boot_params.screen_info.orig_video_lines; current_ypos = boot_params.screen_info.orig_y; early_console = &early_vga_console; - } else if (!strncmp(buf, "simnow", 6)) { - simnow_init(buf + 6); - early_console = &simnow_console; - keep_early = 1; + } else if (!strncmp(buf, "simnow", 6)) { + simnow_init(buf + 6); + early_console = &simnow_console; + keep_early = 1; #ifdef CONFIG_HVC_XEN } else if (!strncmp(buf, "xen", 3)) { early_console = &xenboot_console; diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index 4b87c32b639..9ba49a26dff 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -51,6 +51,7 @@ #include <asm/desc.h> #include <asm/percpu.h> #include <asm/dwarf2.h> +#include <asm/processor-flags.h> #include "irq_vectors.h" /* @@ -68,13 +69,6 @@ #define nr_syscalls ((syscall_table_size)/4) -CF_MASK = 0x00000001 -TF_MASK = 0x00000100 -IF_MASK = 0x00000200 -DF_MASK = 0x00000400 -NT_MASK = 0x00004000 -VM_MASK = 0x00020000 - #ifdef CONFIG_PREEMPT #define preempt_stop(clobbers) DISABLE_INTERRUPTS(clobbers); TRACE_IRQS_OFF #else @@ -84,7 +78,7 @@ VM_MASK = 0x00020000 .macro TRACE_IRQS_IRET #ifdef CONFIG_TRACE_IRQFLAGS - testl $IF_MASK,PT_EFLAGS(%esp) # interrupts off? + testl $X86_EFLAGS_IF,PT_EFLAGS(%esp) # interrupts off? jz 1f TRACE_IRQS_ON 1: @@ -246,7 +240,7 @@ ret_from_intr: check_userspace: movl PT_EFLAGS(%esp), %eax # mix EFLAGS and CS movb PT_CS(%esp), %al - andl $(VM_MASK | SEGMENT_RPL_MASK), %eax + andl $(X86_EFLAGS_VM | SEGMENT_RPL_MASK), %eax cmpl $USER_RPL, %eax jb resume_kernel # not returning to v8086 or userspace @@ -271,7 +265,7 @@ need_resched: movl TI_flags(%ebp), %ecx # need_resched set ? testb $_TIF_NEED_RESCHED, %cl jz restore_all - testl $IF_MASK,PT_EFLAGS(%esp) # interrupts off (exception path) ? + testl $X86_EFLAGS_IF,PT_EFLAGS(%esp) # interrupts off (exception path) ? jz restore_all call preempt_schedule_irq jmp need_resched @@ -291,10 +285,10 @@ ENTRY(ia32_sysenter_target) movl TSS_sysenter_sp0(%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: + * Interrupts are disabled here, but we can't trace it until + * enough kernel state to call TRACE_IRQS_OFF can be called - but + * we immediately enable interrupts at that point anyway. */ - ENABLE_INTERRUPTS(CLBR_NONE) pushl $(__USER_DS) CFI_ADJUST_CFA_OFFSET 4 /*CFI_REL_OFFSET ss, 0*/ @@ -302,6 +296,7 @@ sysenter_past_esp: CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET esp, 0 pushfl + orl $X86_EFLAGS_IF, (%esp) CFI_ADJUST_CFA_OFFSET 4 pushl $(__USER_CS) CFI_ADJUST_CFA_OFFSET 4 @@ -315,6 +310,11 @@ sysenter_past_esp: CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET eip, 0 + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + SAVE_ALL + ENABLE_INTERRUPTS(CLBR_NONE) + /* * Load the potential sixth argument from user stack. * Careful about security. @@ -322,14 +322,12 @@ sysenter_past_esp: cmpl $__PAGE_OFFSET-3,%ebp jae syscall_fault 1: movl (%ebp),%ebp + movl %ebp,PT_EBP(%esp) .section __ex_table,"a" .align 4 .long 1b,syscall_fault .previous - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 - SAVE_ALL GET_THREAD_INFO(%ebp) /* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */ @@ -384,7 +382,7 @@ syscall_exit: # setting need_resched or sigpending # between sampling and the iret TRACE_IRQS_OFF - testl $TF_MASK,PT_EFLAGS(%esp) # If tracing set singlestep flag on exit + testl $X86_EFLAGS_TF,PT_EFLAGS(%esp) # If tracing set singlestep flag on exit jz no_singlestep orl $_TIF_SINGLESTEP,TI_flags(%ebp) no_singlestep: @@ -399,7 +397,7 @@ restore_all: # See comments in process.c:copy_thread() for details. movb PT_OLDSS(%esp), %ah movb PT_CS(%esp), %al - andl $(VM_MASK | (SEGMENT_TI_MASK << 8) | SEGMENT_RPL_MASK), %eax + andl $(X86_EFLAGS_VM | (SEGMENT_TI_MASK << 8) | SEGMENT_RPL_MASK), %eax cmpl $((SEGMENT_LDT << 8) | USER_RPL), %eax CFI_REMEMBER_STATE je ldt_ss # returning to user-space with LDT SS @@ -486,7 +484,7 @@ work_resched: work_notifysig: # deal with pending signals and # notify-resume requests #ifdef CONFIG_VM86 - testl $VM_MASK, PT_EFLAGS(%esp) + testl $X86_EFLAGS_VM, PT_EFLAGS(%esp) movl %esp, %eax jne work_notifysig_v86 # returning to kernel-space or # vm86-space @@ -543,9 +541,6 @@ END(syscall_exit_work) 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,PT_EAX(%esp) jmp resume_userspace diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index c20c9e7e08d..556a8df522a 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -319,19 +319,17 @@ badsys: /* Do syscall tracing */ tracesys: SAVE_REST - movq $-ENOSYS,RAX(%rsp) + movq $-ENOSYS,RAX(%rsp) /* ptrace can change this for a bad syscall */ FIXUP_TOP_OF_STACK %rdi movq %rsp,%rdi call syscall_trace_enter LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */ RESTORE_REST cmpq $__NR_syscall_max,%rax - movq $-ENOSYS,%rcx - cmova %rcx,%rax - ja 1f + ja int_ret_from_sys_call /* RAX(%rsp) set to -ENOSYS above */ movq %r10,%rcx /* fixup for C */ call *sys_call_table(,%rax,8) -1: movq %rax,RAX-ARGOFFSET(%rsp) + movq %rax,RAX-ARGOFFSET(%rsp) /* Use IRET because user could have changed frame */ /* diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c index 4ae7b644026..9546ef408b9 100644 --- a/arch/x86/kernel/genapic_64.c +++ b/arch/x86/kernel/genapic_64.c @@ -15,6 +15,7 @@ #include <linux/kernel.h> #include <linux/ctype.h> #include <linux/init.h> +#include <linux/hardirq.h> #include <asm/smp.h> #include <asm/ipi.h> @@ -24,20 +25,20 @@ #include <acpi/acpi_bus.h> #endif -/* which logical CPU number maps to which CPU (physical APIC ID) */ -u16 x86_cpu_to_apicid_init[NR_CPUS] __initdata - = { [0 ... NR_CPUS-1] = BAD_APICID }; -void *x86_cpu_to_apicid_early_ptr; -DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID; -EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid); +DEFINE_PER_CPU(int, x2apic_extra_bits); struct genapic __read_mostly *genapic = &apic_flat; +static enum uv_system_type uv_system_type; + /* * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode. */ void __init setup_apic_routing(void) { + if (uv_system_type == UV_NON_UNIQUE_APIC) + genapic = &apic_x2apic_uv_x; + else #ifdef CONFIG_ACPI /* * Quirk: some x86_64 machines can only use physical APIC mode @@ -64,3 +65,37 @@ void send_IPI_self(int vector) { __send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL); } + +int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + if (!strcmp(oem_id, "SGI")) { + if (!strcmp(oem_table_id, "UVL")) + uv_system_type = UV_LEGACY_APIC; + else if (!strcmp(oem_table_id, "UVX")) + uv_system_type = UV_X2APIC; + else if (!strcmp(oem_table_id, "UVH")) + uv_system_type = UV_NON_UNIQUE_APIC; + } + return 0; +} + +unsigned int read_apic_id(void) +{ + unsigned int id; + + WARN_ON(preemptible()); + id = apic_read(APIC_ID); + if (uv_system_type >= UV_X2APIC) + id |= __get_cpu_var(x2apic_extra_bits); + return id; +} + +enum uv_system_type get_uv_system_type(void) +{ + return uv_system_type; +} + +int is_uv_system(void) +{ + return uv_system_type != UV_NONE; +} diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c index 07352b74bda..1a9c68845ee 100644 --- a/arch/x86/kernel/genapic_flat_64.c +++ b/arch/x86/kernel/genapic_flat_64.c @@ -97,7 +97,7 @@ static void flat_send_IPI_all(int vector) static int flat_apic_id_registered(void) { - return physid_isset(GET_APIC_ID(apic_read(APIC_ID)), phys_cpu_present_map); + return physid_isset(GET_APIC_ID(read_apic_id()), phys_cpu_present_map); } static unsigned int flat_cpu_mask_to_apicid(cpumask_t cpumask) @@ -138,12 +138,9 @@ static cpumask_t physflat_target_cpus(void) static cpumask_t physflat_vector_allocation_domain(int cpu) { - cpumask_t domain = CPU_MASK_NONE; - cpu_set(cpu, domain); - return domain; + return cpumask_of_cpu(cpu); } - static void physflat_send_IPI_mask(cpumask_t cpumask, int vector) { send_IPI_mask_sequence(cpumask, vector); diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c new file mode 100644 index 00000000000..5d77c9cd8e1 --- /dev/null +++ b/arch/x86/kernel/genx2apic_uv_x.c @@ -0,0 +1,245 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * SGI UV APIC functions (note: not an Intel compatible APIC) + * + * Copyright (C) 2007 Silicon Graphics, Inc. All rights reserved. + */ + +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/string.h> +#include <linux/kernel.h> +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/sched.h> +#include <linux/bootmem.h> +#include <linux/module.h> +#include <asm/smp.h> +#include <asm/ipi.h> +#include <asm/genapic.h> +#include <asm/uv/uv_mmrs.h> +#include <asm/uv/uv_hub.h> + +DEFINE_PER_CPU(struct uv_hub_info_s, __uv_hub_info); +EXPORT_PER_CPU_SYMBOL_GPL(__uv_hub_info); + +struct uv_blade_info *uv_blade_info; +EXPORT_SYMBOL_GPL(uv_blade_info); + +short *uv_node_to_blade; +EXPORT_SYMBOL_GPL(uv_node_to_blade); + +short *uv_cpu_to_blade; +EXPORT_SYMBOL_GPL(uv_cpu_to_blade); + +short uv_possible_blades; +EXPORT_SYMBOL_GPL(uv_possible_blades); + +/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ + +static cpumask_t uv_target_cpus(void) +{ + return cpumask_of_cpu(0); +} + +static cpumask_t uv_vector_allocation_domain(int cpu) +{ + cpumask_t domain = CPU_MASK_NONE; + cpu_set(cpu, domain); + return domain; +} + +int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip) +{ + unsigned long val; + int nasid; + + nasid = uv_apicid_to_nasid(phys_apicid); + val = (1UL << UVH_IPI_INT_SEND_SHFT) | + (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) | + (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) | + (6 << UVH_IPI_INT_DELIVERY_MODE_SHFT); + uv_write_global_mmr64(nasid, UVH_IPI_INT, val); + return 0; +} + +static void uv_send_IPI_one(int cpu, int vector) +{ + unsigned long val, apicid; + int nasid; + + apicid = per_cpu(x86_cpu_to_apicid, cpu); /* ZZZ - cache node-local ? */ + nasid = uv_apicid_to_nasid(apicid); + val = + (1UL << UVH_IPI_INT_SEND_SHFT) | (apicid << + UVH_IPI_INT_APIC_ID_SHFT) | + (vector << UVH_IPI_INT_VECTOR_SHFT); + uv_write_global_mmr64(nasid, UVH_IPI_INT, val); + printk(KERN_DEBUG + "UV: IPI to cpu %d, apicid 0x%lx, vec %d, nasid%d, val 0x%lx\n", + cpu, apicid, vector, nasid, val); +} + +static void uv_send_IPI_mask(cpumask_t mask, int vector) +{ + unsigned int cpu; + + for (cpu = 0; cpu < NR_CPUS; ++cpu) + if (cpu_isset(cpu, mask)) + uv_send_IPI_one(cpu, vector); +} + +static void uv_send_IPI_allbutself(int vector) +{ + cpumask_t mask = cpu_online_map; + + cpu_clear(smp_processor_id(), mask); + + if (!cpus_empty(mask)) + uv_send_IPI_mask(mask, vector); +} + +static void uv_send_IPI_all(int vector) +{ + uv_send_IPI_mask(cpu_online_map, vector); +} + +static int uv_apic_id_registered(void) +{ + return 1; +} + +static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + cpu = first_cpu(cpumask); + if ((unsigned)cpu < NR_CPUS) + return per_cpu(x86_cpu_to_apicid, cpu); + else + return BAD_APICID; +} + +static unsigned int phys_pkg_id(int index_msb) +{ + return GET_APIC_ID(read_apic_id()) >> index_msb; +} + +#ifdef ZZZ /* Needs x2apic patch */ +static void uv_send_IPI_self(int vector) +{ + apic_write(APIC_SELF_IPI, vector); +} +#endif + +struct genapic apic_x2apic_uv_x = { + .name = "UV large system", + .int_delivery_mode = dest_Fixed, + .int_dest_mode = (APIC_DEST_PHYSICAL != 0), + .target_cpus = uv_target_cpus, + .vector_allocation_domain = uv_vector_allocation_domain,/* Fixme ZZZ */ + .apic_id_registered = uv_apic_id_registered, + .send_IPI_all = uv_send_IPI_all, + .send_IPI_allbutself = uv_send_IPI_allbutself, + .send_IPI_mask = uv_send_IPI_mask, + /* ZZZ.send_IPI_self = uv_send_IPI_self, */ + .cpu_mask_to_apicid = uv_cpu_mask_to_apicid, + .phys_pkg_id = phys_pkg_id, /* Fixme ZZZ */ +}; + +static __cpuinit void set_x2apic_extra_bits(int nasid) +{ + __get_cpu_var(x2apic_extra_bits) = ((nasid >> 1) << 6); +} + +/* + * Called on boot cpu. + */ +static __init void uv_system_init(void) +{ + union uvh_si_addr_map_config_u m_n_config; + int bytes, nid, cpu, lcpu, nasid, last_nasid, blade; + unsigned long mmr_base; + + m_n_config.v = uv_read_local_mmr(UVH_SI_ADDR_MAP_CONFIG); + mmr_base = + uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR) & + ~UV_MMR_ENABLE; + printk(KERN_DEBUG "UV: global MMR base 0x%lx\n", mmr_base); + + last_nasid = -1; + for_each_possible_cpu(cpu) { + nid = cpu_to_node(cpu); + nasid = uv_apicid_to_nasid(per_cpu(x86_cpu_to_apicid, cpu)); + if (nasid != last_nasid) + uv_possible_blades++; + last_nasid = nasid; + } + printk(KERN_DEBUG "UV: Found %d blades\n", uv_num_possible_blades()); + + bytes = sizeof(struct uv_blade_info) * uv_num_possible_blades(); + uv_blade_info = alloc_bootmem_pages(bytes); + + bytes = sizeof(uv_node_to_blade[0]) * num_possible_nodes(); + uv_node_to_blade = alloc_bootmem_pages(bytes); + memset(uv_node_to_blade, 255, bytes); + + bytes = sizeof(uv_cpu_to_blade[0]) * num_possible_cpus(); + uv_cpu_to_blade = alloc_bootmem_pages(bytes); + memset(uv_cpu_to_blade, 255, bytes); + + last_nasid = -1; + blade = -1; + lcpu = -1; + for_each_possible_cpu(cpu) { + nid = cpu_to_node(cpu); + nasid = uv_apicid_to_nasid(per_cpu(x86_cpu_to_apicid, cpu)); + if (nasid != last_nasid) { + blade++; + lcpu = -1; + uv_blade_info[blade].nr_posible_cpus = 0; + uv_blade_info[blade].nr_online_cpus = 0; + } + last_nasid = nasid; + lcpu++; + + uv_cpu_hub_info(cpu)->m_val = m_n_config.s.m_skt; + uv_cpu_hub_info(cpu)->n_val = m_n_config.s.n_skt; + uv_cpu_hub_info(cpu)->numa_blade_id = blade; + uv_cpu_hub_info(cpu)->blade_processor_id = lcpu; + uv_cpu_hub_info(cpu)->local_nasid = nasid; + uv_cpu_hub_info(cpu)->gnode_upper = + nasid & ~((1 << uv_hub_info->n_val) - 1); + uv_cpu_hub_info(cpu)->global_mmr_base = mmr_base; + uv_cpu_hub_info(cpu)->coherency_domain_number = 0;/* ZZZ */ + uv_blade_info[blade].nasid = nasid; + uv_blade_info[blade].nr_posible_cpus++; + uv_node_to_blade[nid] = blade; + uv_cpu_to_blade[cpu] = blade; + + printk(KERN_DEBUG "UV cpu %d, apicid 0x%x, nasid %d, nid %d\n", + cpu, per_cpu(x86_cpu_to_apicid, cpu), nasid, nid); + printk(KERN_DEBUG "UV lcpu %d, blade %d\n", lcpu, blade); + } +} + +/* + * Called on each cpu to initialize the per_cpu UV data area. + */ +void __cpuinit uv_cpu_init(void) +{ + if (!uv_node_to_blade) + uv_system_init(); + + uv_blade_info[uv_numa_blade_id()].nr_online_cpus++; + + if (get_uv_system_type() == UV_NON_UNIQUE_APIC) + set_x2apic_extra_bits(uv_hub_info->local_nasid); +} diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c new file mode 100644 index 00000000000..3db05905892 --- /dev/null +++ b/arch/x86/kernel/head32.c @@ -0,0 +1,14 @@ +/* + * linux/arch/i386/kernel/head32.c -- prepare to run common code + * + * Copyright (C) 2000 Andrea Arcangeli <andrea@suse.de> SuSE + * Copyright (C) 2007 Eric Biederman <ebiederm@xmission.com> + */ + +#include <linux/init.h> +#include <linux/start_kernel.h> + +void __init i386_start_kernel(void) +{ + start_kernel(); +} diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index ad2440832de..d6d54faa84d 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -49,39 +49,75 @@ static void __init copy_bootdata(char *real_mode_data) } } -#define EBDA_ADDR_POINTER 0x40E +#define BIOS_EBDA_SEGMENT 0x40E +#define BIOS_LOWMEM_KILOBYTES 0x413 -static __init void reserve_ebda(void) +/* + * The BIOS places the EBDA/XBDA at the top of conventional + * memory, and usually decreases the reported amount of + * conventional memory (int 0x12) too. This also contains a + * workaround for Dell systems that neglect to reserve EBDA. + * The same workaround also avoids a problem with the AMD768MPX + * chipset: reserve a page before VGA to prevent PCI prefetch + * into it (errata #56). Usually the page is reserved anyways, + * unless you have no PS/2 mouse plugged in. + */ +static void __init reserve_ebda_region(void) { - unsigned ebda_addr, ebda_size; + unsigned int lowmem, ebda_addr; + + /* To determine the position of the EBDA and the */ + /* end of conventional memory, we need to look at */ + /* the BIOS data area. In a paravirtual environment */ + /* that area is absent. We'll just have to assume */ + /* that the paravirt case can handle memory setup */ + /* correctly, without our help. */ + if (paravirt_enabled()) + return; - /* - * there is a real-mode segmented pointer pointing to the - * 4K EBDA area at 0x40E - */ - ebda_addr = *(unsigned short *)__va(EBDA_ADDR_POINTER); + /* end of low (conventional) memory */ + lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES); + lowmem <<= 10; + + /* start of EBDA area */ + ebda_addr = *(unsigned short *)__va(BIOS_EBDA_SEGMENT); ebda_addr <<= 4; - if (!ebda_addr) - return; + /* Fixup: bios puts an EBDA in the top 64K segment */ + /* of conventional memory, but does not adjust lowmem. */ + if ((lowmem - ebda_addr) <= 0x10000) + lowmem = ebda_addr; - ebda_size = *(unsigned short *)__va(ebda_addr); + /* Fixup: bios does not report an EBDA at all. */ + /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */ + if ((ebda_addr == 0) && (lowmem >= 0x9f000)) + lowmem = 0x9f000; - /* Round EBDA up to pages */ - if (ebda_size == 0) - ebda_size = 1; - ebda_size <<= 10; - ebda_size = round_up(ebda_size + (ebda_addr & ~PAGE_MASK), PAGE_SIZE); - if (ebda_size > 64*1024) - ebda_size = 64*1024; + /* Paranoia: should never happen, but... */ + if ((lowmem == 0) || (lowmem >= 0x100000)) + lowmem = 0x9f000; - reserve_early(ebda_addr, ebda_addr + ebda_size, "EBDA"); + /* reserve all memory between lowmem and the 1MB mark */ + reserve_early(lowmem, 0x100000, "BIOS reserved"); } void __init x86_64_start_kernel(char * real_mode_data) { int i; + /* + * Build-time sanity checks on the kernel image and module + * area mappings. (these are purely build-time and produce no code) + */ + BUILD_BUG_ON(MODULES_VADDR < KERNEL_IMAGE_START); + BUILD_BUG_ON(MODULES_VADDR-KERNEL_IMAGE_START < KERNEL_IMAGE_SIZE); + BUILD_BUG_ON(MODULES_LEN + KERNEL_IMAGE_SIZE > 2*PUD_SIZE); + BUILD_BUG_ON((KERNEL_IMAGE_START & ~PMD_MASK) != 0); + BUILD_BUG_ON((MODULES_VADDR & ~PMD_MASK) != 0); + BUILD_BUG_ON(!(MODULES_VADDR > __START_KERNEL)); + BUILD_BUG_ON(!(((MODULES_END - 1) & PGDIR_MASK) == + (__START_KERNEL & PGDIR_MASK))); + /* clear bss before set_intr_gate with early_idt_handler */ clear_bss(); @@ -91,7 +127,7 @@ void __init x86_64_start_kernel(char * real_mode_data) /* Cleanup the over mapped high alias */ cleanup_highmap(); - for (i = 0; i < IDT_ENTRIES; i++) { + for (i = 0; i < NUM_EXCEPTION_VECTORS; i++) { #ifdef CONFIG_EARLY_PRINTK set_intr_gate(i, &early_idt_handlers[i]); #else @@ -118,7 +154,7 @@ void __init x86_64_start_kernel(char * real_mode_data) reserve_early(ramdisk_image, ramdisk_end, "RAMDISK"); } - reserve_ebda(); + reserve_ebda_region(); /* * At this point everything still needed from the boot loader diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index 74d87ea85b5..826988a6e96 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -450,7 +450,7 @@ is386: movl $2,%ecx # set MP jmp initialize_secondary # all other CPUs call initialize_secondary 1: #endif /* CONFIG_SMP */ - jmp start_kernel + jmp i386_start_kernel /* * We depend on ET to be correct. This checks for 287/387. diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index a007454133a..10a1955bb1d 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -132,10 +132,6 @@ ident_complete: addq %rbp, trampoline_level4_pgt + 0(%rip) addq %rbp, trampoline_level4_pgt + (511*8)(%rip) #endif -#ifdef CONFIG_ACPI_SLEEP - addq %rbp, wakeup_level4_pgt + 0(%rip) - addq %rbp, wakeup_level4_pgt + (511*8)(%rip) -#endif /* Due to ENTRY(), sometimes the empty space gets filled with * zeros. Better take a jmp than relying on empty space being @@ -267,21 +263,16 @@ ENTRY(secondary_startup_64) bad_address: jmp bad_address + .section ".init.text","ax" #ifdef CONFIG_EARLY_PRINTK -.macro early_idt_tramp first, last - .ifgt \last-\first - early_idt_tramp \first, \last-1 - .endif - movl $\last,%esi - jmp early_idt_handler -.endm - .globl early_idt_handlers early_idt_handlers: - early_idt_tramp 0, 63 - early_idt_tramp 64, 127 - early_idt_tramp 128, 191 - early_idt_tramp 192, 255 + i = 0 + .rept NUM_EXCEPTION_VECTORS + movl $i, %esi + jmp early_idt_handler + i = i + 1 + .endr #endif ENTRY(early_idt_handler) @@ -327,6 +318,7 @@ early_idt_msg: early_idt_ripmsg: .asciz "RIP %s\n" #endif /* CONFIG_EARLY_PRINTK */ + .previous .balign PAGE_SIZE @@ -383,12 +375,12 @@ NEXT_PAGE(level2_ident_pgt) NEXT_PAGE(level2_kernel_pgt) /* - * 128 MB kernel mapping. We spend a full page on this pagetable + * 512 MB kernel mapping. We spend a full page on this pagetable * anyway. * * The kernel code+data+bss must not be bigger than that. * - * (NOTE: at +128MB starts the module area, see MODULES_VADDR. + * (NOTE: at +512MB starts the module area, see MODULES_VADDR. * If you want to increase this then increase MODULES_VADDR * too.) */ diff --git a/arch/x86/kernel/i386_ksyms_32.c b/arch/x86/kernel/i386_ksyms_32.c index 061627806a2..deb43785e92 100644 --- a/arch/x86/kernel/i386_ksyms_32.c +++ b/arch/x86/kernel/i386_ksyms_32.c @@ -1,13 +1,8 @@ #include <linux/module.h> -#include <asm/semaphore.h> #include <asm/checksum.h> #include <asm/desc.h> #include <asm/pgtable.h> -EXPORT_SYMBOL(__down_failed); -EXPORT_SYMBOL(__down_failed_interruptible); -EXPORT_SYMBOL(__down_failed_trylock); -EXPORT_SYMBOL(__up_wakeup); /* Networking helper routines. */ EXPORT_SYMBOL(csum_partial_copy_generic); diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index d2e39e69aaf..8f8102d967b 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -5,45 +5,41 @@ * General FPU state handling cleanups * Gareth Hughes <gareth@valinux.com>, May 2000 */ - -#include <linux/sched.h> #include <linux/module.h> #include <linux/regset.h> +#include <linux/sched.h> + +#include <asm/sigcontext.h> #include <asm/processor.h> -#include <asm/i387.h> #include <asm/math_emu.h> -#include <asm/sigcontext.h> -#include <asm/user.h> -#include <asm/ptrace.h> #include <asm/uaccess.h> +#include <asm/ptrace.h> +#include <asm/i387.h> +#include <asm/user.h> #ifdef CONFIG_X86_64 - -#include <asm/sigcontext32.h> -#include <asm/user32.h> - +# include <asm/sigcontext32.h> +# include <asm/user32.h> #else - -#define save_i387_ia32 save_i387 -#define restore_i387_ia32 restore_i387 - -#define _fpstate_ia32 _fpstate -#define user_i387_ia32_struct user_i387_struct -#define user32_fxsr_struct user_fxsr_struct - +# define save_i387_ia32 save_i387 +# define restore_i387_ia32 restore_i387 +# define _fpstate_ia32 _fpstate +# define user_i387_ia32_struct user_i387_struct +# define user32_fxsr_struct user_fxsr_struct #endif #ifdef CONFIG_MATH_EMULATION -#define HAVE_HWFP (boot_cpu_data.hard_math) +# define HAVE_HWFP (boot_cpu_data.hard_math) #else -#define HAVE_HWFP 1 +# define HAVE_HWFP 1 #endif -static unsigned int mxcsr_feature_mask __read_mostly = 0xffffffffu; +static unsigned int mxcsr_feature_mask __read_mostly = 0xffffffffu; void mxcsr_feature_mask_init(void) { unsigned long mask = 0; + clts(); if (cpu_has_fxsr) { memset(¤t->thread.i387.fxsave, 0, @@ -69,10 +65,11 @@ void __cpuinit fpu_init(void) if (offsetof(struct task_struct, thread.i387.fxsave) & 15) __bad_fxsave_alignment(); + set_in_cr4(X86_CR4_OSFXSR); set_in_cr4(X86_CR4_OSXMMEXCPT); - write_cr0(oldcr0 & ~((1UL<<3)|(1UL<<2))); /* clear TS and EM */ + write_cr0(oldcr0 & ~(X86_CR0_TS|X86_CR0_EM)); /* clear TS and EM */ mxcsr_feature_mask_init(); /* clean state in init */ @@ -178,6 +175,7 @@ static inline unsigned short twd_i387_to_fxsr(unsigned short twd) tmp = (tmp | (tmp >> 1)) & 0x3333; /* 00VV00VV00VV00VV */ tmp = (tmp | (tmp >> 2)) & 0x0f0f; /* 0000VVVV0000VVVV */ tmp = (tmp | (tmp >> 4)) & 0x00ff; /* 00000000VVVVVVVV */ + return tmp; } @@ -232,8 +230,8 @@ static inline u32 twd_fxsr_to_i387(struct i387_fxsave_struct *fxsave) * FXSR floating point environment conversions. */ -static void convert_from_fxsr(struct user_i387_ia32_struct *env, - struct task_struct *tsk) +static void +convert_from_fxsr(struct user_i387_ia32_struct *env, struct task_struct *tsk) { struct i387_fxsave_struct *fxsave = &tsk->thread.i387.fxsave; struct _fpreg *to = (struct _fpreg *) &env->st_space[0]; @@ -252,10 +250,11 @@ static void convert_from_fxsr(struct user_i387_ia32_struct *env, * should be actually ds/cs at fpu exception time, but * that information is not available in 64bit mode. */ - asm("mov %%ds,%0" : "=r" (env->fos)); - asm("mov %%cs,%0" : "=r" (env->fcs)); + asm("mov %%ds, %[fos]" : [fos] "=r" (env->fos)); + asm("mov %%cs, %[fcs]" : [fcs] "=r" (env->fcs)); } else { struct pt_regs *regs = task_pt_regs(tsk); + env->fos = 0xffff0000 | tsk->thread.ds; env->fcs = regs->cs; } @@ -309,9 +308,10 @@ int fpregs_get(struct task_struct *target, const struct user_regset *regset, init_fpu(target); - if (!cpu_has_fxsr) + if (!cpu_has_fxsr) { return user_regset_copyout(&pos, &count, &kbuf, &ubuf, &target->thread.i387.fsave, 0, -1); + } if (kbuf && pos == 0 && count == sizeof(env)) { convert_from_fxsr(kbuf, target); @@ -319,6 +319,7 @@ int fpregs_get(struct task_struct *target, const struct user_regset *regset, } convert_from_fxsr(&env, target); + return user_regset_copyout(&pos, &count, &kbuf, &ubuf, &env, 0, -1); } @@ -335,9 +336,10 @@ int fpregs_set(struct task_struct *target, const struct user_regset *regset, init_fpu(target); set_stopped_child_used_math(target); - if (!cpu_has_fxsr) + if (!cpu_has_fxsr) { return user_regset_copyin(&pos, &count, &kbuf, &ubuf, &target->thread.i387.fsave, 0, -1); + } if (pos > 0 || count < sizeof(env)) convert_from_fxsr(&env, target); @@ -392,28 +394,28 @@ int save_i387_ia32(struct _fpstate_ia32 __user *buf) { if (!used_math()) return 0; - - /* This will cause a "finit" to be triggered by the next + /* + * This will cause a "finit" to be triggered by the next * attempted FPU operation by the 'current' process. */ clear_used_math(); - if (HAVE_HWFP) { - if (cpu_has_fxsr) { - return save_i387_fxsave(buf); - } else { - return save_i387_fsave(buf); - } - } else { + if (!HAVE_HWFP) { return fpregs_soft_get(current, NULL, 0, sizeof(struct user_i387_ia32_struct), NULL, buf) ? -1 : 1; } + + if (cpu_has_fxsr) + return save_i387_fxsave(buf); + else + return save_i387_fsave(buf); } static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf) { struct task_struct *tsk = current; + clear_fpu(tsk); return __copy_from_user(&tsk->thread.i387.fsave, buf, sizeof(struct i387_fsave_struct)); @@ -421,9 +423,10 @@ static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf) static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf) { - int err; struct task_struct *tsk = current; struct user_i387_ia32_struct env; + int err; + clear_fpu(tsk); err = __copy_from_user(&tsk->thread.i387.fxsave, &buf->_fxsr_env[0], sizeof(struct i387_fxsave_struct)); @@ -432,6 +435,7 @@ static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf) if (err || __copy_from_user(&env, buf, sizeof(env))) return 1; convert_to_fxsr(tsk, &env); + return 0; } @@ -440,17 +444,17 @@ int restore_i387_ia32(struct _fpstate_ia32 __user *buf) int err; if (HAVE_HWFP) { - if (cpu_has_fxsr) { + if (cpu_has_fxsr) err = restore_i387_fxsave(buf); - } else { + else err = restore_i387_fsave(buf); - } } else { err = fpregs_soft_set(current, NULL, 0, sizeof(struct user_i387_ia32_struct), NULL, buf) != 0; } set_used_math(); + return err; } @@ -463,8 +467,8 @@ int restore_i387_ia32(struct _fpstate_ia32 __user *buf) */ int dump_fpu(struct pt_regs *regs, struct user_i387_struct *fpu) { - int fpvalid; struct task_struct *tsk = current; + int fpvalid; fpvalid = !!used_math(); if (fpvalid) diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c index 4ca548632c8..2e2f42074e1 100644 --- a/arch/x86/kernel/io_apic_32.c +++ b/arch/x86/kernel/io_apic_32.c @@ -71,6 +71,16 @@ int sis_apic_bug = -1; */ int nr_ioapic_registers[MAX_IO_APICS]; +/* I/O APIC entries */ +struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS]; +int nr_ioapics; + +/* MP IRQ source entries */ +struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; + +/* # of MP IRQ source entries */ +int mp_irq_entries; + static int disable_timer_pin_1 __initdata; /* @@ -810,10 +820,7 @@ static int __init find_isa_irq_pin(int irq, int type) for (i = 0; i < mp_irq_entries; i++) { int lbus = mp_irqs[i].mpc_srcbus; - if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA || - mp_bus_id_to_type[lbus] == MP_BUS_EISA || - mp_bus_id_to_type[lbus] == MP_BUS_MCA - ) && + if (test_bit(lbus, mp_bus_not_pci) && (mp_irqs[i].mpc_irqtype == type) && (mp_irqs[i].mpc_srcbusirq == irq)) @@ -829,10 +836,7 @@ static int __init find_isa_irq_apic(int irq, int type) for (i = 0; i < mp_irq_entries; i++) { int lbus = mp_irqs[i].mpc_srcbus; - if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA || - mp_bus_id_to_type[lbus] == MP_BUS_EISA || - mp_bus_id_to_type[lbus] == MP_BUS_MCA - ) && + if (test_bit(lbus, mp_bus_not_pci) && (mp_irqs[i].mpc_irqtype == type) && (mp_irqs[i].mpc_srcbusirq == irq)) break; @@ -872,7 +876,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) mp_irqs[i].mpc_dstapic == MP_APIC_ALL) break; - if ((mp_bus_id_to_type[lbus] == MP_BUS_PCI) && + if (!test_bit(lbus, mp_bus_not_pci) && !mp_irqs[i].mpc_irqtype && (bus == lbus) && (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) { @@ -921,6 +925,7 @@ void __init setup_ioapic_dest(void) } #endif +#if defined(CONFIG_EISA) || defined(CONFIG_MCA) /* * EISA Edge/Level control register, ELCR */ @@ -934,6 +939,13 @@ static int EISA_ELCR(unsigned int irq) "Broken MPtable reports ISA irq %d\n", irq); return 0; } +#endif + +/* ISA interrupts are always polarity zero edge triggered, + * when listed as conforming in the MP table. */ + +#define default_ISA_trigger(idx) (0) +#define default_ISA_polarity(idx) (0) /* EISA interrupts are always polarity zero and can be edge or level * trigger depending on the ELCR value. If an interrupt is listed as @@ -941,13 +953,7 @@ static int EISA_ELCR(unsigned int irq) * be read in from the ELCR */ #define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mpc_srcbusirq)) -#define default_EISA_polarity(idx) (0) - -/* ISA interrupts are always polarity zero edge triggered, - * when listed as conforming in the MP table. */ - -#define default_ISA_trigger(idx) (0) -#define default_ISA_polarity(idx) (0) +#define default_EISA_polarity(idx) default_ISA_polarity(idx) /* PCI interrupts are always polarity one level triggered, * when listed as conforming in the MP table. */ @@ -959,7 +965,7 @@ static int EISA_ELCR(unsigned int irq) * when listed as conforming in the MP table. */ #define default_MCA_trigger(idx) (1) -#define default_MCA_polarity(idx) (0) +#define default_MCA_polarity(idx) default_ISA_polarity(idx) static int MPBIOS_polarity(int idx) { @@ -973,35 +979,9 @@ static int MPBIOS_polarity(int idx) { case 0: /* conforms, ie. bus-type dependent polarity */ { - switch (mp_bus_id_to_type[bus]) - { - case MP_BUS_ISA: /* ISA pin */ - { - polarity = default_ISA_polarity(idx); - break; - } - case MP_BUS_EISA: /* EISA pin */ - { - polarity = default_EISA_polarity(idx); - break; - } - case MP_BUS_PCI: /* PCI pin */ - { - polarity = default_PCI_polarity(idx); - break; - } - case MP_BUS_MCA: /* MCA pin */ - { - polarity = default_MCA_polarity(idx); - break; - } - default: - { - printk(KERN_WARNING "broken BIOS!!\n"); - polarity = 1; - break; - } - } + polarity = test_bit(bus, mp_bus_not_pci)? + default_ISA_polarity(idx): + default_PCI_polarity(idx); break; } case 1: /* high active */ @@ -1042,11 +1022,15 @@ static int MPBIOS_trigger(int idx) { case 0: /* conforms, ie. bus-type dependent */ { + trigger = test_bit(bus, mp_bus_not_pci)? + default_ISA_trigger(idx): + default_PCI_trigger(idx); +#if defined(CONFIG_EISA) || defined(CONFIG_MCA) switch (mp_bus_id_to_type[bus]) { case MP_BUS_ISA: /* ISA pin */ { - trigger = default_ISA_trigger(idx); + /* set before the switch */ break; } case MP_BUS_EISA: /* EISA pin */ @@ -1056,7 +1040,7 @@ static int MPBIOS_trigger(int idx) } case MP_BUS_PCI: /* PCI pin */ { - trigger = default_PCI_trigger(idx); + /* set before the switch */ break; } case MP_BUS_MCA: /* MCA pin */ @@ -1071,6 +1055,7 @@ static int MPBIOS_trigger(int idx) break; } } +#endif break; } case 1: /* edge */ @@ -1120,39 +1105,22 @@ static int pin_2_irq(int idx, int apic, int pin) if (mp_irqs[idx].mpc_dstirq != pin) printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n"); - switch (mp_bus_id_to_type[bus]) - { - case MP_BUS_ISA: /* ISA pin */ - case MP_BUS_EISA: - case MP_BUS_MCA: - { - irq = mp_irqs[idx].mpc_srcbusirq; - break; - } - case MP_BUS_PCI: /* PCI pin */ - { - /* - * PCI IRQs are mapped in order - */ - i = irq = 0; - while (i < apic) - irq += nr_ioapic_registers[i++]; - irq += pin; - - /* - * For MPS mode, so far only needed by ES7000 platform - */ - if (ioapic_renumber_irq) - irq = ioapic_renumber_irq(apic, irq); + if (test_bit(bus, mp_bus_not_pci)) + irq = mp_irqs[idx].mpc_srcbusirq; + else { + /* + * PCI IRQs are mapped in order + */ + i = irq = 0; + while (i < apic) + irq += nr_ioapic_registers[i++]; + irq += pin; - break; - } - default: - { - printk(KERN_ERR "unknown bus type %d.\n",bus); - irq = 0; - break; - } + /* + * For MPS mode, so far only needed by ES7000 platform + */ + if (ioapic_renumber_irq) + irq = ioapic_renumber_irq(apic, irq); } /* @@ -1260,7 +1228,6 @@ static void __init setup_IO_APIC_irqs(void) { struct IO_APIC_route_entry entry; int apic, pin, idx, irq, first_notcon = 1, vector; - unsigned long flags; apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); @@ -1326,9 +1293,7 @@ static void __init setup_IO_APIC_irqs(void) if (!apic && (irq < 16)) disable_8259A_irq(irq); } - spin_lock_irqsave(&ioapic_lock, flags); - __ioapic_write_entry(apic, pin, entry); - spin_unlock_irqrestore(&ioapic_lock, flags); + ioapic_write_entry(apic, pin, entry); } } @@ -1524,8 +1489,8 @@ void /*__init*/ print_local_APIC(void * dummy) printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n", smp_processor_id(), hard_smp_processor_id()); - v = apic_read(APIC_ID); - printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, GET_APIC_ID(v)); + printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, + GET_APIC_ID(read_apic_id())); v = apic_read(APIC_LVR); printk(KERN_INFO "... APIC VERSION: %08x\n", v); ver = GET_APIC_VERSION(v); @@ -1734,7 +1699,7 @@ void disable_IO_APIC(void) entry.delivery_mode = dest_ExtINT; /* ExtInt */ entry.vector = 0; entry.dest.physical.physical_dest = - GET_APIC_ID(apic_read(APIC_ID)); + GET_APIC_ID(read_apic_id()); /* * Add it to the IO-APIC irq-routing table: @@ -2031,8 +1996,7 @@ static inline void init_IO_APIC_traps(void) * 0x80, because int 0x80 is hm, kind of importantish. ;) */ for (irq = 0; irq < NR_IRQS ; irq++) { - int tmp = irq; - if (IO_APIC_IRQ(tmp) && !irq_vector[tmp]) { + if (IO_APIC_IRQ(irq) && !irq_vector[irq]) { /* * Hmm.. We don't have an entry for this, * so default to an old-fashioned 8259 @@ -2156,8 +2120,6 @@ static inline void unlock_ExtINT_logic(void) ioapic_write_entry(apic, pin, entry0); } -int timer_uses_ioapic_pin_0; - /* * This code may look a bit paranoid, but it's supposed to cooperate with * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ @@ -2168,10 +2130,14 @@ static inline void __init check_timer(void) { int apic1, pin1, apic2, pin2; int vector; + unsigned int ver; unsigned long flags; local_irq_save(flags); + ver = apic_read(APIC_LVR); + ver = GET_APIC_VERSION(ver); + /* * get/set the timer IRQ vector: */ @@ -2184,11 +2150,15 @@ static inline void __init check_timer(void) * mode for the 8259A whenever interrupts are routed * through I/O APICs. Also IRQ0 has to be enabled in * the 8259A which implies the virtual wire has to be - * disabled in the local APIC. + * disabled in the local APIC. Finally timer interrupts + * need to be acknowledged manually in the 8259A for + * timer_interrupt() and for the i82489DX when using + * the NMI watchdog. */ apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); init_8259A(1); - timer_ack = 1; + timer_ack = !cpu_has_tsc; + timer_ack |= (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); if (timer_over_8254 > 0) enable_8259A_irq(0); @@ -2197,9 +2167,6 @@ static inline void __init check_timer(void) pin2 = ioapic_i8259.pin; apic2 = ioapic_i8259.apic; - if (pin1 == 0) - timer_uses_ioapic_pin_0 = 1; - printk(KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n", vector, apic1, pin1, apic2, pin2); @@ -2789,7 +2756,6 @@ int __init io_apic_get_redir_entries (int ioapic) int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int active_high_low) { struct IO_APIC_route_entry entry; - unsigned long flags; if (!IO_APIC_IRQ(irq)) { printk(KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n", @@ -2830,9 +2796,7 @@ int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int a if (!ioapic && (irq < 16)) disable_8259A_irq(irq); - spin_lock_irqsave(&ioapic_lock, flags); - __ioapic_write_entry(ioapic, pin, entry); - spin_unlock_irqrestore(&ioapic_lock, flags); + ioapic_write_entry(ioapic, pin, entry); return 0; } diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index 1627c0d53e0..b54464b2665 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic_64.c @@ -43,13 +43,15 @@ #include <asm/smp.h> #include <asm/desc.h> #include <asm/proto.h> -#include <asm/mach_apic.h> #include <asm/acpi.h> #include <asm/dma.h> #include <asm/nmi.h> #include <asm/msidef.h> #include <asm/hypertransport.h> +#include <mach_ipi.h> +#include <mach_apic.h> + struct irq_cfg { cpumask_t domain; cpumask_t old_domain; @@ -101,6 +103,16 @@ DEFINE_SPINLOCK(vector_lock); */ int nr_ioapic_registers[MAX_IO_APICS]; +/* I/O APIC entries */ +struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS]; +int nr_ioapics; + +/* MP IRQ source entries */ +struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; + +/* # of MP IRQ source entries */ +int mp_irq_entries; + /* * Rough estimation of how many shared IRQs there are, can * be changed anytime. @@ -155,11 +167,10 @@ static inline void io_apic_modify(unsigned int apic, unsigned int value) writel(value, &io_apic->data); } -static int io_apic_level_ack_pending(unsigned int irq) +static bool io_apic_level_ack_pending(unsigned int irq) { struct irq_pin_list *entry; unsigned long flags; - int pending = 0; spin_lock_irqsave(&ioapic_lock, flags); entry = irq_2_pin + irq; @@ -172,13 +183,17 @@ static int io_apic_level_ack_pending(unsigned int irq) break; reg = io_apic_read(entry->apic, 0x10 + pin*2); /* Is the remote IRR bit set? */ - pending |= (reg >> 14) & 1; + if ((reg >> 14) & 1) { + spin_unlock_irqrestore(&ioapic_lock, flags); + return true; + } if (!entry->next) break; entry = irq_2_pin + entry->next; } spin_unlock_irqrestore(&ioapic_lock, flags); - return pending; + + return false; } /* @@ -902,9 +917,8 @@ static void __init setup_IO_APIC_irqs(void) static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector) { struct IO_APIC_route_entry entry; - unsigned long flags; - memset(&entry,0,sizeof(entry)); + memset(&entry, 0, sizeof(entry)); disable_8259A_irq(0); @@ -932,10 +946,7 @@ static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, in /* * Add it to the IO-APIC irq-routing table: */ - spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1)); - io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0)); - spin_unlock_irqrestore(&ioapic_lock, flags); + ioapic_write_entry(apic, pin, entry); enable_8259A_irq(0); } @@ -1066,8 +1077,7 @@ void __apicdebuginit print_local_APIC(void * dummy) printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n", smp_processor_id(), hard_smp_processor_id()); - v = apic_read(APIC_ID); - printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, GET_APIC_ID(v)); + printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, GET_APIC_ID(read_apic_id())); v = apic_read(APIC_LVR); printk(KERN_INFO "... APIC VERSION: %08x\n", v); ver = GET_APIC_VERSION(v); @@ -1261,7 +1271,7 @@ void disable_IO_APIC(void) entry.dest_mode = 0; /* Physical */ entry.delivery_mode = dest_ExtINT; /* ExtInt */ entry.vector = 0; - entry.dest = GET_APIC_ID(apic_read(APIC_ID)); + entry.dest = GET_APIC_ID(read_apic_id()); /* * Add it to the IO-APIC irq-routing table: @@ -1352,9 +1362,7 @@ static int ioapic_retrigger_irq(unsigned int irq) unsigned long flags; spin_lock_irqsave(&vector_lock, flags); - cpus_clear(mask); - cpu_set(first_cpu(cfg->domain), mask); - + mask = cpumask_of_cpu(first_cpu(cfg->domain)); send_IPI_mask(mask, cfg->vector); spin_unlock_irqrestore(&vector_lock, flags); @@ -1517,8 +1525,7 @@ static inline void init_IO_APIC_traps(void) * 0x80, because int 0x80 is hm, kind of importantish. ;) */ for (irq = 0; irq < NR_IRQS ; irq++) { - int tmp = irq; - if (IO_APIC_IRQ(tmp) && !irq_cfg[tmp].vector) { + if (IO_APIC_IRQ(irq) && !irq_cfg[irq].vector) { /* * Hmm.. We don't have an entry for this, * so default to an old-fashioned 8259 @@ -1597,17 +1604,14 @@ static inline void unlock_ExtINT_logic(void) int apic, pin, i; struct IO_APIC_route_entry entry0, entry1; unsigned char save_control, save_freq_select; - unsigned long flags; pin = find_isa_irq_pin(8, mp_INT); apic = find_isa_irq_apic(8, mp_INT); if (pin == -1) return; - spin_lock_irqsave(&ioapic_lock, flags); - *(((int *)&entry0) + 1) = io_apic_read(apic, 0x11 + 2 * pin); - *(((int *)&entry0) + 0) = io_apic_read(apic, 0x10 + 2 * pin); - spin_unlock_irqrestore(&ioapic_lock, flags); + entry0 = ioapic_read_entry(apic, pin); + clear_IO_APIC_pin(apic, pin); memset(&entry1, 0, sizeof(entry1)); @@ -1620,10 +1624,7 @@ static inline void unlock_ExtINT_logic(void) entry1.trigger = 0; entry1.vector = 0; - spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry1) + 1)); - io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry1) + 0)); - spin_unlock_irqrestore(&ioapic_lock, flags); + ioapic_write_entry(apic, pin, entry1); save_control = CMOS_READ(RTC_CONTROL); save_freq_select = CMOS_READ(RTC_FREQ_SELECT); @@ -1642,10 +1643,7 @@ static inline void unlock_ExtINT_logic(void) CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT); clear_IO_APIC_pin(apic, pin); - spin_lock_irqsave(&ioapic_lock, flags); - io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry0) + 1)); - io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry0) + 0)); - spin_unlock_irqrestore(&ioapic_lock, flags); + ioapic_write_entry(apic, pin, entry0); } /* @@ -2314,7 +2312,6 @@ static struct resource * __init ioapic_setup_resources(void) res = (void *)mem; if (mem != NULL) { - memset(mem, 0, n); mem += sizeof(struct resource) * nr_ioapics; for (i = 0; i < nr_ioapics; i++) { diff --git a/arch/x86/kernel/ipi.c b/arch/x86/kernel/ipi.c new file mode 100644 index 00000000000..c0df7b89ca2 --- /dev/null +++ b/arch/x86/kernel/ipi.c @@ -0,0 +1,178 @@ +#include <linux/cpumask.h> +#include <linux/interrupt.h> +#include <linux/init.h> + +#include <linux/mm.h> +#include <linux/delay.h> +#include <linux/spinlock.h> +#include <linux/kernel_stat.h> +#include <linux/mc146818rtc.h> +#include <linux/cache.h> +#include <linux/interrupt.h> +#include <linux/cpu.h> +#include <linux/module.h> + +#include <asm/smp.h> +#include <asm/mtrr.h> +#include <asm/tlbflush.h> +#include <asm/mmu_context.h> +#include <asm/apic.h> +#include <asm/proto.h> + +#ifdef CONFIG_X86_32 +#include <mach_apic.h> +/* + * the following functions deal with sending IPIs between CPUs. + * + * We use 'broadcast', CPU->CPU IPIs and self-IPIs too. + */ + +static inline int __prepare_ICR(unsigned int shortcut, int vector) +{ + unsigned int icr = shortcut | APIC_DEST_LOGICAL; + + switch (vector) { + default: + icr |= APIC_DM_FIXED | vector; + break; + case NMI_VECTOR: + icr |= APIC_DM_NMI; + break; + } + return icr; +} + +static inline int __prepare_ICR2(unsigned int mask) +{ + return SET_APIC_DEST_FIELD(mask); +} + +void __send_IPI_shortcut(unsigned int shortcut, int vector) +{ + /* + * Subtle. In the case of the 'never do double writes' workaround + * we have to lock out interrupts to be safe. As we don't care + * of the value read we use an atomic rmw access to avoid costly + * cli/sti. Otherwise we use an even cheaper single atomic write + * to the APIC. + */ + unsigned int cfg; + + /* + * Wait for idle. + */ + apic_wait_icr_idle(); + + /* + * No need to touch the target chip field + */ + cfg = __prepare_ICR(shortcut, vector); + + /* + * Send the IPI. The write to APIC_ICR fires this off. + */ + apic_write_around(APIC_ICR, cfg); +} + +void send_IPI_self(int vector) +{ + __send_IPI_shortcut(APIC_DEST_SELF, vector); +} + +/* + * This is used to send an IPI with no shorthand notation (the destination is + * specified in bits 56 to 63 of the ICR). + */ +static inline void __send_IPI_dest_field(unsigned long mask, int vector) +{ + unsigned long cfg; + + /* + * Wait for idle. + */ + if (unlikely(vector == NMI_VECTOR)) + safe_apic_wait_icr_idle(); + else + apic_wait_icr_idle(); + + /* + * prepare target chip field + */ + cfg = __prepare_ICR2(mask); + apic_write_around(APIC_ICR2, cfg); + + /* + * program the ICR + */ + cfg = __prepare_ICR(0, vector); + + /* + * Send the IPI. The write to APIC_ICR fires this off. + */ + apic_write_around(APIC_ICR, cfg); +} + +/* + * This is only used on smaller machines. + */ +void send_IPI_mask_bitmask(cpumask_t cpumask, int vector) +{ + unsigned long mask = cpus_addr(cpumask)[0]; + unsigned long flags; + + local_irq_save(flags); + WARN_ON(mask & ~cpus_addr(cpu_online_map)[0]); + __send_IPI_dest_field(mask, vector); + local_irq_restore(flags); +} + +void send_IPI_mask_sequence(cpumask_t mask, int vector) +{ + unsigned long flags; + unsigned int query_cpu; + + /* + * Hack. The clustered APIC addressing mode doesn't allow us to send + * to an arbitrary mask, so I do a unicasts to each CPU instead. This + * should be modified to do 1 message per cluster ID - mbligh + */ + + local_irq_save(flags); + for_each_possible_cpu(query_cpu) { + if (cpu_isset(query_cpu, mask)) { + __send_IPI_dest_field(cpu_to_logical_apicid(query_cpu), + vector); + } + } + local_irq_restore(flags); +} + +/* must come after the send_IPI functions above for inlining */ +#include <mach_ipi.h> +static int convert_apicid_to_cpu(int apic_id) +{ + int i; + + for_each_possible_cpu(i) { + if (per_cpu(x86_cpu_to_apicid, i) == apic_id) + return i; + } + return -1; +} + +int safe_smp_processor_id(void) +{ + int apicid, cpuid; + + if (!boot_cpu_has(X86_FEATURE_APIC)) + return 0; + + apicid = hard_smp_processor_id(); + if (apicid == BAD_APICID) + return 0; + + cpuid = convert_apicid_to_cpu(apicid); + + return cpuid >= 0 ? cpuid : 0; +} +#endif diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index cef054b09d2..6ea67b76a21 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -79,7 +79,7 @@ unsigned int do_IRQ(struct pt_regs *regs) if (unlikely((unsigned)irq >= NR_IRQS)) { printk(KERN_EMERG "%s: cannot handle IRQ %d\n", - __FUNCTION__, irq); + __func__, irq); BUG(); } diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c new file mode 100644 index 00000000000..24362ecf5f9 --- /dev/null +++ b/arch/x86/kernel/kgdb.c @@ -0,0 +1,571 @@ +/* + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2, or (at your option) any + * later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + */ + +/* + * Copyright (C) 2004 Amit S. Kale <amitkale@linsyssoft.com> + * Copyright (C) 2000-2001 VERITAS Software Corporation. + * Copyright (C) 2002 Andi Kleen, SuSE Labs + * Copyright (C) 2004 LinSysSoft Technologies Pvt. Ltd. + * Copyright (C) 2007 MontaVista Software, Inc. + * Copyright (C) 2007-2008 Jason Wessel, Wind River Systems, Inc. + */ +/**************************************************************************** + * Contributor: Lake Stevens Instrument Division$ + * Written by: Glenn Engel $ + * Updated by: Amit Kale<akale@veritas.com> + * Updated by: Tom Rini <trini@kernel.crashing.org> + * Updated by: Jason Wessel <jason.wessel@windriver.com> + * Modified for 386 by Jim Kingdon, Cygnus Support. + * Origianl kgdb, compatibility with 2.1.xx kernel by + * David Grothe <dave@gcom.com> + * Integrated into 2.2.5 kernel by Tigran Aivazian <tigran@sco.com> + * X86_64 changes from Andi Kleen's patch merged by Jim Houston + */ +#include <linux/spinlock.h> +#include <linux/kdebug.h> +#include <linux/string.h> +#include <linux/kernel.h> +#include <linux/ptrace.h> +#include <linux/sched.h> +#include <linux/delay.h> +#include <linux/kgdb.h> +#include <linux/init.h> +#include <linux/smp.h> +#include <linux/nmi.h> + +#include <asm/apicdef.h> +#include <asm/system.h> + +#ifdef CONFIG_X86_32 +# include <mach_ipi.h> +#else +# include <asm/mach_apic.h> +#endif + +/* + * Put the error code here just in case the user cares: + */ +static int gdb_x86errcode; + +/* + * Likewise, the vector number here (since GDB only gets the signal + * number through the usual means, and that's not very specific): + */ +static int gdb_x86vector = -1; + +/** + * pt_regs_to_gdb_regs - Convert ptrace regs to GDB regs + * @gdb_regs: A pointer to hold the registers in the order GDB wants. + * @regs: The &struct pt_regs of the current process. + * + * Convert the pt_regs in @regs into the format for registers that + * GDB expects, stored in @gdb_regs. + */ +void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) +{ + gdb_regs[GDB_AX] = regs->ax; + gdb_regs[GDB_BX] = regs->bx; + gdb_regs[GDB_CX] = regs->cx; + gdb_regs[GDB_DX] = regs->dx; + gdb_regs[GDB_SI] = regs->si; + gdb_regs[GDB_DI] = regs->di; + gdb_regs[GDB_BP] = regs->bp; + gdb_regs[GDB_PS] = regs->flags; + gdb_regs[GDB_PC] = regs->ip; +#ifdef CONFIG_X86_32 + gdb_regs[GDB_DS] = regs->ds; + gdb_regs[GDB_ES] = regs->es; + gdb_regs[GDB_CS] = regs->cs; + gdb_regs[GDB_SS] = __KERNEL_DS; + gdb_regs[GDB_FS] = 0xFFFF; + gdb_regs[GDB_GS] = 0xFFFF; +#else + gdb_regs[GDB_R8] = regs->r8; + gdb_regs[GDB_R9] = regs->r9; + gdb_regs[GDB_R10] = regs->r10; + gdb_regs[GDB_R11] = regs->r11; + gdb_regs[GDB_R12] = regs->r12; + gdb_regs[GDB_R13] = regs->r13; + gdb_regs[GDB_R14] = regs->r14; + gdb_regs[GDB_R15] = regs->r15; +#endif + gdb_regs[GDB_SP] = regs->sp; +} + +/** + * sleeping_thread_to_gdb_regs - Convert ptrace regs to GDB regs + * @gdb_regs: A pointer to hold the registers in the order GDB wants. + * @p: The &struct task_struct of the desired process. + * + * Convert the register values of the sleeping process in @p to + * the format that GDB expects. + * This function is called when kgdb does not have access to the + * &struct pt_regs and therefore it should fill the gdb registers + * @gdb_regs with what has been saved in &struct thread_struct + * thread field during switch_to. + */ +void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) +{ + gdb_regs[GDB_AX] = 0; + gdb_regs[GDB_BX] = 0; + gdb_regs[GDB_CX] = 0; + gdb_regs[GDB_DX] = 0; + gdb_regs[GDB_SI] = 0; + gdb_regs[GDB_DI] = 0; + gdb_regs[GDB_BP] = *(unsigned long *)p->thread.sp; +#ifdef CONFIG_X86_32 + gdb_regs[GDB_DS] = __KERNEL_DS; + gdb_regs[GDB_ES] = __KERNEL_DS; + gdb_regs[GDB_PS] = 0; + gdb_regs[GDB_CS] = __KERNEL_CS; + gdb_regs[GDB_PC] = p->thread.ip; + gdb_regs[GDB_SS] = __KERNEL_DS; + gdb_regs[GDB_FS] = 0xFFFF; + gdb_regs[GDB_GS] = 0xFFFF; +#else + gdb_regs[GDB_PS] = *(unsigned long *)(p->thread.sp + 8); + gdb_regs[GDB_PC] = 0; + gdb_regs[GDB_R8] = 0; + gdb_regs[GDB_R9] = 0; + gdb_regs[GDB_R10] = 0; + gdb_regs[GDB_R11] = 0; + gdb_regs[GDB_R12] = 0; + gdb_regs[GDB_R13] = 0; + gdb_regs[GDB_R14] = 0; + gdb_regs[GDB_R15] = 0; +#endif + gdb_regs[GDB_SP] = p->thread.sp; +} + +/** + * gdb_regs_to_pt_regs - Convert GDB regs to ptrace regs. + * @gdb_regs: A pointer to hold the registers we've received from GDB. + * @regs: A pointer to a &struct pt_regs to hold these values in. + * + * Convert the GDB regs in @gdb_regs into the pt_regs, and store them + * in @regs. + */ +void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) +{ + regs->ax = gdb_regs[GDB_AX]; + regs->bx = gdb_regs[GDB_BX]; + regs->cx = gdb_regs[GDB_CX]; + regs->dx = gdb_regs[GDB_DX]; + regs->si = gdb_regs[GDB_SI]; + regs->di = gdb_regs[GDB_DI]; + regs->bp = gdb_regs[GDB_BP]; + regs->flags = gdb_regs[GDB_PS]; + regs->ip = gdb_regs[GDB_PC]; +#ifdef CONFIG_X86_32 + regs->ds = gdb_regs[GDB_DS]; + regs->es = gdb_regs[GDB_ES]; + regs->cs = gdb_regs[GDB_CS]; +#else + regs->r8 = gdb_regs[GDB_R8]; + regs->r9 = gdb_regs[GDB_R9]; + regs->r10 = gdb_regs[GDB_R10]; + regs->r11 = gdb_regs[GDB_R11]; + regs->r12 = gdb_regs[GDB_R12]; + regs->r13 = gdb_regs[GDB_R13]; + regs->r14 = gdb_regs[GDB_R14]; + regs->r15 = gdb_regs[GDB_R15]; +#endif +} + +static struct hw_breakpoint { + unsigned enabled; + unsigned type; + unsigned len; + unsigned long addr; +} breakinfo[4]; + +static void kgdb_correct_hw_break(void) +{ + unsigned long dr7; + int correctit = 0; + int breakbit; + int breakno; + + get_debugreg(dr7, 7); + for (breakno = 0; breakno < 4; breakno++) { + breakbit = 2 << (breakno << 1); + if (!(dr7 & breakbit) && breakinfo[breakno].enabled) { + correctit = 1; + dr7 |= breakbit; + dr7 &= ~(0xf0000 << (breakno << 2)); + dr7 |= ((breakinfo[breakno].len << 2) | + breakinfo[breakno].type) << + ((breakno << 2) + 16); + if (breakno >= 0 && breakno <= 3) + set_debugreg(breakinfo[breakno].addr, breakno); + + } else { + if ((dr7 & breakbit) && !breakinfo[breakno].enabled) { + correctit = 1; + dr7 &= ~breakbit; + dr7 &= ~(0xf0000 << (breakno << 2)); + } + } + } + if (correctit) + set_debugreg(dr7, 7); +} + +static int +kgdb_remove_hw_break(unsigned long addr, int len, enum kgdb_bptype bptype) +{ + int i; + + for (i = 0; i < 4; i++) + if (breakinfo[i].addr == addr && breakinfo[i].enabled) + break; + if (i == 4) + return -1; + + breakinfo[i].enabled = 0; + + return 0; +} + +static void kgdb_remove_all_hw_break(void) +{ + int i; + + for (i = 0; i < 4; i++) + memset(&breakinfo[i], 0, sizeof(struct hw_breakpoint)); +} + +static int +kgdb_set_hw_break(unsigned long addr, int len, enum kgdb_bptype bptype) +{ + unsigned type; + int i; + + for (i = 0; i < 4; i++) + if (!breakinfo[i].enabled) + break; + if (i == 4) + return -1; + + switch (bptype) { + case BP_HARDWARE_BREAKPOINT: + type = 0; + len = 1; + break; + case BP_WRITE_WATCHPOINT: + type = 1; + break; + case BP_ACCESS_WATCHPOINT: + type = 3; + break; + default: + return -1; + } + + if (len == 1 || len == 2 || len == 4) + breakinfo[i].len = len - 1; + else + return -1; + + breakinfo[i].enabled = 1; + breakinfo[i].addr = addr; + breakinfo[i].type = type; + + return 0; +} + +/** + * kgdb_disable_hw_debug - Disable hardware debugging while we in kgdb. + * @regs: Current &struct pt_regs. + * + * This function will be called if the particular architecture must + * disable hardware debugging while it is processing gdb packets or + * handling exception. + */ +void kgdb_disable_hw_debug(struct pt_regs *regs) +{ + /* Disable hardware debugging while we are in kgdb: */ + set_debugreg(0UL, 7); +} + +/** + * kgdb_post_primary_code - Save error vector/code numbers. + * @regs: Original pt_regs. + * @e_vector: Original error vector. + * @err_code: Original error code. + * + * This is needed on architectures which support SMP and KGDB. + * This function is called after all the slave cpus have been put + * to a know spin state and the primary CPU has control over KGDB. + */ +void kgdb_post_primary_code(struct pt_regs *regs, int e_vector, int err_code) +{ + /* primary processor is completely in the debugger */ + gdb_x86vector = e_vector; + gdb_x86errcode = err_code; +} + +#ifdef CONFIG_SMP +/** + * kgdb_roundup_cpus - Get other CPUs into a holding pattern + * @flags: Current IRQ state + * + * On SMP systems, we need to get the attention of the other CPUs + * and get them be in a known state. This should do what is needed + * to get the other CPUs to call kgdb_wait(). Note that on some arches, + * the NMI approach is not used for rounding up all the CPUs. For example, + * in case of MIPS, smp_call_function() is used to roundup CPUs. In + * this case, we have to make sure that interrupts are enabled before + * calling smp_call_function(). The argument to this function is + * the flags that will be used when restoring the interrupts. There is + * local_irq_save() call before kgdb_roundup_cpus(). + * + * On non-SMP systems, this is not called. + */ +void kgdb_roundup_cpus(unsigned long flags) +{ + send_IPI_allbutself(APIC_DM_NMI); +} +#endif + +/** + * kgdb_arch_handle_exception - Handle architecture specific GDB packets. + * @vector: The error vector of the exception that happened. + * @signo: The signal number of the exception that happened. + * @err_code: The error code of the exception that happened. + * @remcom_in_buffer: The buffer of the packet we have read. + * @remcom_out_buffer: The buffer of %BUFMAX bytes to write a packet into. + * @regs: The &struct pt_regs of the current process. + * + * This function MUST handle the 'c' and 's' command packets, + * as well packets to set / remove a hardware breakpoint, if used. + * If there are additional packets which the hardware needs to handle, + * they are handled here. The code should return -1 if it wants to + * process more packets, and a %0 or %1 if it wants to exit from the + * kgdb callback. + */ +int kgdb_arch_handle_exception(int e_vector, int signo, int err_code, + char *remcomInBuffer, char *remcomOutBuffer, + struct pt_regs *linux_regs) +{ + unsigned long addr; + unsigned long dr6; + char *ptr; + int newPC; + + switch (remcomInBuffer[0]) { + case 'c': + case 's': + /* try to read optional parameter, pc unchanged if no parm */ + ptr = &remcomInBuffer[1]; + if (kgdb_hex2long(&ptr, &addr)) + linux_regs->ip = addr; + case 'D': + case 'k': + newPC = linux_regs->ip; + + /* clear the trace bit */ + linux_regs->flags &= ~X86_EFLAGS_TF; + atomic_set(&kgdb_cpu_doing_single_step, -1); + + /* set the trace bit if we're stepping */ + if (remcomInBuffer[0] == 's') { + linux_regs->flags |= X86_EFLAGS_TF; + kgdb_single_step = 1; + if (kgdb_contthread) { + atomic_set(&kgdb_cpu_doing_single_step, + raw_smp_processor_id()); + } + } + + get_debugreg(dr6, 6); + if (!(dr6 & 0x4000)) { + int breakno; + + for (breakno = 0; breakno < 4; breakno++) { + if (dr6 & (1 << breakno) && + breakinfo[breakno].type == 0) { + /* Set restore flag: */ + linux_regs->flags |= X86_EFLAGS_RF; + break; + } + } + } + set_debugreg(0UL, 6); + kgdb_correct_hw_break(); + + return 0; + } + + /* this means that we do not want to exit from the handler: */ + return -1; +} + +static inline int +single_step_cont(struct pt_regs *regs, struct die_args *args) +{ + /* + * Single step exception from kernel space to user space so + * eat the exception and continue the process: + */ + printk(KERN_ERR "KGDB: trap/step from kernel to user space, " + "resuming...\n"); + kgdb_arch_handle_exception(args->trapnr, args->signr, + args->err, "c", "", regs); + + return NOTIFY_STOP; +} + +static int was_in_debug_nmi[NR_CPUS]; + +static int __kgdb_notify(struct die_args *args, unsigned long cmd) +{ + struct pt_regs *regs = args->regs; + + switch (cmd) { + case DIE_NMI: + if (atomic_read(&kgdb_active) != -1) { + /* KGDB CPU roundup */ + kgdb_nmicallback(raw_smp_processor_id(), regs); + was_in_debug_nmi[raw_smp_processor_id()] = 1; + touch_nmi_watchdog(); + return NOTIFY_STOP; + } + return NOTIFY_DONE; + + case DIE_NMI_IPI: + if (atomic_read(&kgdb_active) != -1) { + /* KGDB CPU roundup */ + kgdb_nmicallback(raw_smp_processor_id(), regs); + was_in_debug_nmi[raw_smp_processor_id()] = 1; + touch_nmi_watchdog(); + } + return NOTIFY_DONE; + + case DIE_NMIUNKNOWN: + if (was_in_debug_nmi[raw_smp_processor_id()]) { + was_in_debug_nmi[raw_smp_processor_id()] = 0; + return NOTIFY_STOP; + } + return NOTIFY_DONE; + + case DIE_NMIWATCHDOG: + if (atomic_read(&kgdb_active) != -1) { + /* KGDB CPU roundup: */ + kgdb_nmicallback(raw_smp_processor_id(), regs); + return NOTIFY_STOP; + } + /* Enter debugger: */ + break; + + case DIE_DEBUG: + if (atomic_read(&kgdb_cpu_doing_single_step) == + raw_smp_processor_id() && + user_mode(regs)) + return single_step_cont(regs, args); + /* fall through */ + default: + if (user_mode(regs)) + return NOTIFY_DONE; + } + + if (kgdb_handle_exception(args->trapnr, args->signr, args->err, regs)) + return NOTIFY_DONE; + + /* Must touch watchdog before return to normal operation */ + touch_nmi_watchdog(); + return NOTIFY_STOP; +} + +static int +kgdb_notify(struct notifier_block *self, unsigned long cmd, void *ptr) +{ + unsigned long flags; + int ret; + + local_irq_save(flags); + ret = __kgdb_notify(ptr, cmd); + local_irq_restore(flags); + + return ret; +} + +static struct notifier_block kgdb_notifier = { + .notifier_call = kgdb_notify, + + /* + * Lowest-prio notifier priority, we want to be notified last: + */ + .priority = -INT_MAX, +}; + +/** + * kgdb_arch_init - Perform any architecture specific initalization. + * + * This function will handle the initalization of any architecture + * specific callbacks. + */ +int kgdb_arch_init(void) +{ + return register_die_notifier(&kgdb_notifier); +} + +/** + * kgdb_arch_exit - Perform any architecture specific uninitalization. + * + * This function will handle the uninitalization of any architecture + * specific callbacks, for dynamic registration and unregistration. + */ +void kgdb_arch_exit(void) +{ + unregister_die_notifier(&kgdb_notifier); +} + +/** + * + * kgdb_skipexception - Bail out of KGDB when we've been triggered. + * @exception: Exception vector number + * @regs: Current &struct pt_regs. + * + * On some architectures we need to skip a breakpoint exception when + * it occurs after a breakpoint has been removed. + * + * Skip an int3 exception when it occurs after a breakpoint has been + * removed. Backtrack eip by 1 since the int3 would have caused it to + * increment by 1. + */ +int kgdb_skipexception(int exception, struct pt_regs *regs) +{ + if (exception == 3 && kgdb_isremovedbreak(regs->ip - 1)) { + regs->ip -= 1; + return 1; + } + return 0; +} + +unsigned long kgdb_arch_pc(int exception, struct pt_regs *regs) +{ + if (exception == 3) + return instruction_pointer(regs) - 1; + return instruction_pointer(regs); +} + +struct kgdb_arch arch_kgdb_ops = { + /* Breakpoint instruction: */ + .gdb_bpt_instr = { 0xcc }, + .flags = KGDB_HW_BREAKPOINT, + .set_hw_breakpoint = kgdb_set_hw_break, + .remove_hw_breakpoint = kgdb_remove_hw_break, + .remove_all_hw_break = kgdb_remove_all_hw_break, + .correct_hw_break = kgdb_correct_hw_break, +}; diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 34a591283f5..b8c6743a13d 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -410,13 +410,13 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs, static void __kprobes clear_btf(void) { if (test_thread_flag(TIF_DEBUGCTLMSR)) - wrmsrl(MSR_IA32_DEBUGCTLMSR, 0); + update_debugctlmsr(0); } static void __kprobes restore_btf(void) { if (test_thread_flag(TIF_DEBUGCTLMSR)) - wrmsrl(MSR_IA32_DEBUGCTLMSR, current->thread.debugctlmsr); + update_debugctlmsr(current->thread.debugctlmsr); } static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs) @@ -489,7 +489,7 @@ static int __kprobes reenter_kprobe(struct kprobe *p, struct pt_regs *regs, break; case KPROBE_HIT_SS: if (p == kprobe_running()) { - regs->flags &= ~TF_MASK; + regs->flags &= ~X86_EFLAGS_TF; regs->flags |= kcb->kprobe_saved_flags; return 0; } else { @@ -858,15 +858,15 @@ static int __kprobes post_kprobe_handler(struct pt_regs *regs) if (!cur) return 0; + resume_execution(cur, regs, kcb); + regs->flags |= kcb->kprobe_saved_flags; + trace_hardirqs_fixup_flags(regs->flags); + if ((kcb->kprobe_status != KPROBE_REENTER) && cur->post_handler) { kcb->kprobe_status = KPROBE_HIT_SSDONE; cur->post_handler(cur, regs, 0); } - resume_execution(cur, regs, kcb); - regs->flags |= kcb->kprobe_saved_flags; - trace_hardirqs_fixup_flags(regs->flags); - /* Restore back the original saved kprobes variables and continue. */ if (kcb->kprobe_status == KPROBE_REENTER) { restore_previous_kprobe(kcb); diff --git a/arch/x86/kernel/mca_32.c b/arch/x86/kernel/mca_32.c index 9482033ed0f..2dc183758be 100644 --- a/arch/x86/kernel/mca_32.c +++ b/arch/x86/kernel/mca_32.c @@ -53,9 +53,9 @@ #include <linux/init.h> #include <asm/arch_hooks.h> -static unsigned char which_scsi = 0; +static unsigned char which_scsi; -int MCA_bus = 0; +int MCA_bus; EXPORT_SYMBOL(MCA_bus); /* @@ -68,15 +68,17 @@ static DEFINE_SPINLOCK(mca_lock); /* Build the status info for the adapter */ -static void mca_configure_adapter_status(struct mca_device *mca_dev) { +static void mca_configure_adapter_status(struct mca_device *mca_dev) +{ mca_dev->status = MCA_ADAPTER_NONE; mca_dev->pos_id = mca_dev->pos[0] + (mca_dev->pos[1] << 8); - if(!mca_dev->pos_id && mca_dev->slot < MCA_MAX_SLOT_NR) { + if (!mca_dev->pos_id && mca_dev->slot < MCA_MAX_SLOT_NR) { - /* id = 0x0000 usually indicates hardware failure, + /* + * id = 0x0000 usually indicates hardware failure, * however, ZP Gu (zpg@castle.net> reports that his 9556 * has 0x0000 as id and everything still works. There * also seem to be an adapter with id = 0x0000; the @@ -87,9 +89,10 @@ static void mca_configure_adapter_status(struct mca_device *mca_dev) { mca_dev->status = MCA_ADAPTER_ERROR; return; - } else if(mca_dev->pos_id != 0xffff) { + } else if (mca_dev->pos_id != 0xffff) { - /* 0xffff usually indicates that there's no adapter, + /* + * 0xffff usually indicates that there's no adapter, * however, some integrated adapters may have 0xffff as * their id and still be valid. Examples are on-board * VGA of the 55sx, the integrated SCSI of the 56 & 57, @@ -99,19 +102,19 @@ static void mca_configure_adapter_status(struct mca_device *mca_dev) { mca_dev->status = MCA_ADAPTER_NORMAL; } - if((mca_dev->pos_id == 0xffff || + if ((mca_dev->pos_id == 0xffff || mca_dev->pos_id == 0x0000) && mca_dev->slot >= MCA_MAX_SLOT_NR) { int j; - for(j = 2; j < 8; j++) { - if(mca_dev->pos[j] != 0xff) { + for (j = 2; j < 8; j++) { + if (mca_dev->pos[j] != 0xff) { mca_dev->status = MCA_ADAPTER_NORMAL; break; } } } - if(!(mca_dev->pos[2] & MCA_ENABLED)) { + if (!(mca_dev->pos[2] & MCA_ENABLED)) { /* enabled bit is in POS 2 */ @@ -133,7 +136,7 @@ static struct resource mca_standard_resources[] = { #define MCA_STANDARD_RESOURCES ARRAY_SIZE(mca_standard_resources) -/** +/* * mca_read_and_store_pos - read the POS registers into a memory buffer * @pos: a char pointer to 8 bytes, contains the POS register value on * successful return @@ -141,12 +144,14 @@ static struct resource mca_standard_resources[] = { * Returns 1 if a card actually exists (i.e. the pos isn't * all 0xff) or 0 otherwise */ -static int mca_read_and_store_pos(unsigned char *pos) { +static int mca_read_and_store_pos(unsigned char *pos) +{ int j; int found = 0; - for(j=0; j<8; j++) { - if((pos[j] = inb_p(MCA_POS_REG(j))) != 0xff) { + for (j = 0; j < 8; j++) { + pos[j] = inb_p(MCA_POS_REG(j)); + if (pos[j] != 0xff) { /* 0xff all across means no device. 0x00 means * something's broken, but a device is * probably there. However, if you get 0x00 @@ -167,11 +172,11 @@ static unsigned char mca_pc_read_pos(struct mca_device *mca_dev, int reg) unsigned char byte; unsigned long flags; - if(reg < 0 || reg >= 8) + if (reg < 0 || reg >= 8) return 0; spin_lock_irqsave(&mca_lock, flags); - if(mca_dev->pos_register) { + if (mca_dev->pos_register) { /* Disable adapter setup, enable motherboard setup */ outb_p(0, MCA_ADAPTER_SETUP_REG); @@ -203,7 +208,7 @@ static void mca_pc_write_pos(struct mca_device *mca_dev, int reg, { unsigned long flags; - if(reg < 0 || reg >= 8) + if (reg < 0 || reg >= 8) return; spin_lock_irqsave(&mca_lock, flags); @@ -227,17 +232,17 @@ static void mca_pc_write_pos(struct mca_device *mca_dev, int reg, } /* for the primary MCA bus, we have identity transforms */ -static int mca_dummy_transform_irq(struct mca_device * mca_dev, int irq) +static int mca_dummy_transform_irq(struct mca_device *mca_dev, int irq) { return irq; } -static int mca_dummy_transform_ioport(struct mca_device * mca_dev, int port) +static int mca_dummy_transform_ioport(struct mca_device *mca_dev, int port) { return port; } -static void *mca_dummy_transform_memory(struct mca_device * mca_dev, void *mem) +static void *mca_dummy_transform_memory(struct mca_device *mca_dev, void *mem) { return mem; } @@ -251,7 +256,8 @@ static int __init mca_init(void) short mca_builtin_scsi_ports[] = {0xf7, 0xfd, 0x00}; struct mca_bus *bus; - /* WARNING: Be careful when making changes here. Putting an adapter + /* + * WARNING: Be careful when making changes here. Putting an adapter * and the motherboard simultaneously into setup mode may result in * damage to chips (according to The Indispensible PC Hardware Book * by Hans-Peter Messmer). Also, we disable system interrupts (so @@ -283,7 +289,7 @@ static int __init mca_init(void) /* get the motherboard device */ mca_dev = kzalloc(sizeof(struct mca_device), GFP_KERNEL); - if(unlikely(!mca_dev)) + if (unlikely(!mca_dev)) goto out_nomem; /* @@ -309,7 +315,7 @@ static int __init mca_init(void) mca_register_device(MCA_PRIMARY_BUS, mca_dev); mca_dev = kzalloc(sizeof(struct mca_device), GFP_ATOMIC); - if(unlikely(!mca_dev)) + if (unlikely(!mca_dev)) goto out_unlock_nomem; /* Put motherboard into video setup mode, read integrated video @@ -326,7 +332,8 @@ static int __init mca_init(void) mca_dev->slot = MCA_INTEGVIDEO; mca_register_device(MCA_PRIMARY_BUS, mca_dev); - /* Put motherboard into scsi setup mode, read integrated scsi + /* + * Put motherboard into scsi setup mode, read integrated scsi * POS registers, and turn motherboard setup off. * * It seems there are two possible SCSI registers. Martin says that @@ -338,18 +345,18 @@ static int __init mca_init(void) * machine. */ - for(i = 0; (which_scsi = mca_builtin_scsi_ports[i]) != 0; i++) { + for (i = 0; (which_scsi = mca_builtin_scsi_ports[i]) != 0; i++) { outb_p(which_scsi, MCA_MOTHERBOARD_SETUP_REG); - if(mca_read_and_store_pos(pos)) + if (mca_read_and_store_pos(pos)) break; } - if(which_scsi) { + if (which_scsi) { /* found a scsi card */ mca_dev = kzalloc(sizeof(struct mca_device), GFP_ATOMIC); - if(unlikely(!mca_dev)) + if (unlikely(!mca_dev)) goto out_unlock_nomem; - for(j = 0; j < 8; j++) + for (j = 0; j < 8; j++) mca_dev->pos[j] = pos[j]; mca_configure_adapter_status(mca_dev); @@ -364,21 +371,22 @@ static int __init mca_init(void) outb_p(0xff, MCA_MOTHERBOARD_SETUP_REG); - /* Now loop over MCA slots: put each adapter into setup mode, and + /* + * Now loop over MCA slots: put each adapter into setup mode, and * read its POS registers. Then put adapter setup off. */ - for(i=0; i<MCA_MAX_SLOT_NR; i++) { + for (i = 0; i < MCA_MAX_SLOT_NR; i++) { outb_p(0x8|(i&0xf), MCA_ADAPTER_SETUP_REG); - if(!mca_read_and_store_pos(pos)) + if (!mca_read_and_store_pos(pos)) continue; mca_dev = kzalloc(sizeof(struct mca_device), GFP_ATOMIC); - if(unlikely(!mca_dev)) + if (unlikely(!mca_dev)) goto out_unlock_nomem; - for(j=0; j<8; j++) - mca_dev->pos[j]=pos[j]; + for (j = 0; j < 8; j++) + mca_dev->pos[j] = pos[j]; mca_dev->driver_loaded = 0; mca_dev->slot = i; @@ -414,20 +422,20 @@ mca_handle_nmi_device(struct mca_device *mca_dev, int check_flag) { int slot = mca_dev->slot; - if(slot == MCA_INTEGSCSI) { + if (slot == MCA_INTEGSCSI) { printk(KERN_CRIT "NMI: caused by MCA integrated SCSI adapter (%s)\n", mca_dev->name); - } else if(slot == MCA_INTEGVIDEO) { + } else if (slot == MCA_INTEGVIDEO) { printk(KERN_CRIT "NMI: caused by MCA integrated video adapter (%s)\n", mca_dev->name); - } else if(slot == MCA_MOTHERBOARD) { + } else if (slot == MCA_MOTHERBOARD) { printk(KERN_CRIT "NMI: caused by motherboard (%s)\n", mca_dev->name); } /* More info available in POS 6 and 7? */ - if(check_flag) { + if (check_flag) { unsigned char pos6, pos7; pos6 = mca_device_read_pos(mca_dev, 6); @@ -447,8 +455,9 @@ static int __kprobes mca_handle_nmi_callback(struct device *dev, void *data) pos5 = mca_device_read_pos(mca_dev, 5); - if(!(pos5 & 0x80)) { - /* Bit 7 of POS 5 is reset when this adapter has a hardware + if (!(pos5 & 0x80)) { + /* + * Bit 7 of POS 5 is reset when this adapter has a hardware * error. Bit 7 it reset if there's error information * available in POS 6 and 7. */ @@ -460,7 +469,8 @@ static int __kprobes mca_handle_nmi_callback(struct device *dev, void *data) void __kprobes mca_handle_nmi(void) { - /* First try - scan the various adapters and see if a specific + /* + * First try - scan the various adapters and see if a specific * adapter was responsible for the error. */ bus_for_each_dev(&mca_bus_type, NULL, NULL, mca_handle_nmi_callback); diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c index f2702d01b8a..25cf6dee4e5 100644 --- a/arch/x86/kernel/microcode.c +++ b/arch/x86/kernel/microcode.c @@ -290,7 +290,7 @@ static int get_maching_microcode(void *mc, int cpu) } return 0; find: - pr_debug("microcode: CPU %d found a matching microcode update with" + pr_debug("microcode: CPU%d found a matching microcode update with" " version 0x%x (current=0x%x)\n", cpu, mc_header->rev,uci->rev); new_mc = vmalloc(total_size); if (!new_mc) { @@ -336,11 +336,11 @@ static void apply_microcode(int cpu) spin_unlock_irqrestore(µcode_update_lock, flags); if (val[1] != uci->mc->hdr.rev) { - printk(KERN_ERR "microcode: CPU%d updated from revision " + printk(KERN_ERR "microcode: CPU%d update from revision " "0x%x to 0x%x failed\n", cpu_num, uci->rev, val[1]); return; } - pr_debug("microcode: CPU%d updated from revision " + printk(KERN_INFO "microcode: CPU%d updated from revision " "0x%x to 0x%x, date = %08x \n", cpu_num, uci->rev, val[1], uci->mc->hdr.date); uci->rev = val[1]; @@ -534,7 +534,7 @@ static int cpu_request_microcode(int cpu) c->x86, c->x86_model, c->x86_mask); error = request_firmware(&firmware, name, µcode_pdev->dev); if (error) { - pr_debug("ucode data file %s load failed\n", name); + pr_debug("microcode: ucode data file %s load failed\n", name); return error; } buf = firmware->data; @@ -709,7 +709,7 @@ static int __mc_sysdev_add(struct sys_device *sys_dev, int resume) if (!cpu_online(cpu)) return 0; - pr_debug("Microcode:CPU %d added\n", cpu); + pr_debug("microcode: CPU%d added\n", cpu); memset(uci, 0, sizeof(*uci)); err = sysfs_create_group(&sys_dev->kobj, &mc_attr_group); @@ -733,7 +733,7 @@ static int mc_sysdev_remove(struct sys_device *sys_dev) if (!cpu_online(cpu)) return 0; - pr_debug("Microcode:CPU %d removed\n", cpu); + pr_debug("microcode: CPU%d removed\n", cpu); microcode_fini_cpu(cpu); sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); return 0; @@ -745,7 +745,7 @@ static int mc_sysdev_resume(struct sys_device *dev) if (!cpu_online(cpu)) return 0; - pr_debug("Microcode:CPU %d resumed\n", cpu); + pr_debug("microcode: CPU%d resumed\n", cpu); /* only CPU 0 will apply ucode here */ apply_microcode(0); return 0; @@ -783,7 +783,7 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) } case CPU_DOWN_FAILED_FROZEN: if (sysfs_create_group(&sys_dev->kobj, &mc_attr_group)) - printk(KERN_ERR "Microcode: Failed to create the sysfs " + printk(KERN_ERR "microcode: Failed to create the sysfs " "group for CPU%d\n", cpu); break; case CPU_DOWN_PREPARE: diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse.c index f349e68e45a..70744e344fa 100644 --- a/arch/x86/kernel/mpparse_32.c +++ b/arch/x86/kernel/mpparse.c @@ -4,82 +4,56 @@ * * (c) 1995 Alan Cox, Building #3 <alan@redhat.com> * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com> - * - * Fixes - * Erich Boleyn : MP v1.4 and additional changes. - * Alan Cox : Added EBDA scanning - * Ingo Molnar : various cleanups and rewrites - * Maciej W. Rozycki: Bits for default MP configurations - * Paul Diefenbaugh: Added full ACPI support + * (c) 2008 Alexey Starikovskiy <astarikovskiy@suse.de> */ #include <linux/mm.h> #include <linux/init.h> -#include <linux/acpi.h> #include <linux/delay.h> #include <linux/bootmem.h> #include <linux/kernel_stat.h> #include <linux/mc146818rtc.h> #include <linux/bitops.h> +#include <linux/acpi.h> +#include <linux/module.h> #include <asm/smp.h> -#include <asm/acpi.h> #include <asm/mtrr.h> #include <asm/mpspec.h> +#include <asm/pgalloc.h> #include <asm/io_apic.h> +#include <asm/proto.h> +#include <asm/acpi.h> +#include <asm/bios_ebda.h> #include <mach_apic.h> +#ifdef CONFIG_X86_32 #include <mach_apicdef.h> #include <mach_mpparse.h> -#include <bios_ebda.h> +#endif /* Have we found an MP table */ int smp_found_config; -unsigned int __cpuinitdata maxcpus = NR_CPUS; /* * Various Linux-internal data structures created from the * MP-table. */ -int apic_version [MAX_APICS]; -int mp_bus_id_to_type [MAX_MP_BUSSES]; -int mp_bus_id_to_node [MAX_MP_BUSSES]; -int mp_bus_id_to_local [MAX_MP_BUSSES]; -int quad_local_to_mp_bus_id [NR_CPUS/4][4]; -int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 }; -static int mp_current_pci_id; - -/* I/O APIC entries */ -struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS]; - -/* # of MP IRQ source entries */ -struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; +#if defined (CONFIG_MCA) || defined (CONFIG_EISA) +int mp_bus_id_to_type[MAX_MP_BUSSES]; +#endif -/* MP IRQ source entries */ -int mp_irq_entries; +DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); +int mp_bus_id_to_pci_bus[MAX_MP_BUSSES] = {[0 ... MAX_MP_BUSSES - 1] = -1 }; -int nr_ioapics; +static int mp_current_pci_id; int pic_mode; -unsigned long mp_lapic_addr; - -unsigned int def_to_bigsmp = 0; - -/* Processor that is doing the boot up */ -unsigned int boot_cpu_physical_apicid = -1U; -/* Internal processor count */ -unsigned int num_processors; - -/* Bitmask of physically existing CPUs */ -physid_mask_t phys_cpu_present_map; - -u8 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID }; /* * Intel MP BIOS table parsing routines: */ - /* * Checksum an MP configuration block. */ @@ -94,216 +68,153 @@ static int __init mpf_checksum(unsigned char *mp, int len) return sum & 0xFF; } +#ifdef CONFIG_X86_NUMAQ /* * Have to match translation table entries to main table entries by counter * hence the mpc_record variable .... can't see a less disgusting way of * doing this .... */ -static int mpc_record; -static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __cpuinitdata; +static int mpc_record; +static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] + __cpuinitdata; +#endif -static void __cpuinit MP_processor_info (struct mpc_config_processor *m) +static void __cpuinit MP_processor_info(struct mpc_config_processor *m) { - int ver, apicid; - physid_mask_t phys_cpu; - - if (!(m->mpc_cpuflag & CPU_ENABLED)) - return; + int apicid; + char *bootup_cpu = ""; + if (!(m->mpc_cpuflag & CPU_ENABLED)) { + disabled_cpus++; + return; + } +#ifdef CONFIG_X86_NUMAQ apicid = mpc_apic_id(m, translation_table[mpc_record]); - - if (m->mpc_featureflag&(1<<0)) - Dprintk(" Floating point unit present.\n"); - if (m->mpc_featureflag&(1<<7)) - Dprintk(" Machine Exception supported.\n"); - if (m->mpc_featureflag&(1<<8)) - Dprintk(" 64 bit compare & exchange supported.\n"); - if (m->mpc_featureflag&(1<<9)) - Dprintk(" Internal APIC present.\n"); - if (m->mpc_featureflag&(1<<11)) - Dprintk(" SEP present.\n"); - if (m->mpc_featureflag&(1<<12)) - Dprintk(" MTRR present.\n"); - if (m->mpc_featureflag&(1<<13)) - Dprintk(" PGE present.\n"); - if (m->mpc_featureflag&(1<<14)) - Dprintk(" MCA present.\n"); - if (m->mpc_featureflag&(1<<15)) - Dprintk(" CMOV present.\n"); - if (m->mpc_featureflag&(1<<16)) - Dprintk(" PAT present.\n"); - if (m->mpc_featureflag&(1<<17)) - Dprintk(" PSE present.\n"); - if (m->mpc_featureflag&(1<<18)) - Dprintk(" PSN present.\n"); - if (m->mpc_featureflag&(1<<19)) - Dprintk(" Cache Line Flush Instruction present.\n"); - /* 20 Reserved */ - if (m->mpc_featureflag&(1<<21)) - Dprintk(" Debug Trace and EMON Store present.\n"); - if (m->mpc_featureflag&(1<<22)) - Dprintk(" ACPI Thermal Throttle Registers present.\n"); - if (m->mpc_featureflag&(1<<23)) - Dprintk(" MMX present.\n"); - if (m->mpc_featureflag&(1<<24)) - Dprintk(" FXSR present.\n"); - if (m->mpc_featureflag&(1<<25)) - Dprintk(" XMM present.\n"); - if (m->mpc_featureflag&(1<<26)) - Dprintk(" Willamette New Instructions present.\n"); - if (m->mpc_featureflag&(1<<27)) - Dprintk(" Self Snoop present.\n"); - if (m->mpc_featureflag&(1<<28)) - Dprintk(" HT present.\n"); - if (m->mpc_featureflag&(1<<29)) - Dprintk(" Thermal Monitor present.\n"); - /* 30, 31 Reserved */ - - +#else + apicid = m->mpc_apicid; +#endif if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) { - Dprintk(" Bootup CPU\n"); + bootup_cpu = " (Bootup-CPU)"; boot_cpu_physical_apicid = m->mpc_apicid; } - ver = m->mpc_apicver; - - /* - * Validate version - */ - if (ver == 0x0) { - printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! " - "fixing up to 0x10. (tell your hw vendor)\n", - m->mpc_apicid); - ver = 0x10; - } - apic_version[m->mpc_apicid] = ver; - - phys_cpu = apicid_to_cpu_present(apicid); - physids_or(phys_cpu_present_map, phys_cpu_present_map, phys_cpu); - - if (num_processors >= NR_CPUS) { - printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached." - " Processor ignored.\n", NR_CPUS); - return; - } - - if (num_processors >= maxcpus) { - printk(KERN_WARNING "WARNING: maxcpus limit of %i reached." - " Processor ignored.\n", maxcpus); - return; - } - - cpu_set(num_processors, cpu_possible_map); - num_processors++; - - /* - * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y - * but we need to work other dependencies like SMP_SUSPEND etc - * before this can be done without some confusion. - * if (CPU_HOTPLUG_ENABLED || num_processors > 8) - * - Ashok Raj <ashok.raj@intel.com> - */ - if (num_processors > 8) { - switch (boot_cpu_data.x86_vendor) { - case X86_VENDOR_INTEL: - if (!APIC_XAPIC(ver)) { - def_to_bigsmp = 0; - break; - } - /* If P4 and above fall through */ - case X86_VENDOR_AMD: - def_to_bigsmp = 1; - } - } - bios_cpu_apicid[num_processors - 1] = m->mpc_apicid; + printk(KERN_INFO "Processor #%d%s\n", m->mpc_apicid, bootup_cpu); + generic_processor_info(apicid, m->mpc_apicver); } -static void __init MP_bus_info (struct mpc_config_bus *m) +static void __init MP_bus_info(struct mpc_config_bus *m) { char str[7]; memcpy(str, m->mpc_bustype, 6); str[6] = 0; +#ifdef CONFIG_X86_NUMAQ mpc_oem_bus_info(m, str, translation_table[mpc_record]); +#else + Dprintk("Bus #%d is %s\n", m->mpc_busid, str); +#endif #if MAX_MP_BUSSES < 256 if (m->mpc_busid >= MAX_MP_BUSSES) { printk(KERN_WARNING "MP table busid value (%d) for bustype %s " - " is too large, max. supported is %d\n", - m->mpc_busid, str, MAX_MP_BUSSES - 1); + " is too large, max. supported is %d\n", + m->mpc_busid, str, MAX_MP_BUSSES - 1); return; } #endif - if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA)-1) == 0) { + if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA) - 1) == 0) { + set_bit(m->mpc_busid, mp_bus_not_pci); +#if defined(CONFIG_EISA) || defined (CONFIG_MCA) mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA; - } else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA)-1) == 0) { - mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA; - } else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI)-1) == 0) { +#endif + } else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI) - 1) == 0) { +#ifdef CONFIG_X86_NUMAQ mpc_oem_pci_bus(m, translation_table[mpc_record]); - mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI; +#endif + clear_bit(m->mpc_busid, mp_bus_not_pci); mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id; mp_current_pci_id++; - } else if (strncmp(str, BUSTYPE_MCA, sizeof(BUSTYPE_MCA)-1) == 0) { +#if defined(CONFIG_EISA) || defined (CONFIG_MCA) + mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI; + } else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA) - 1) == 0) { + mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA; + } else if (strncmp(str, BUSTYPE_MCA, sizeof(BUSTYPE_MCA) - 1) == 0) { mp_bus_id_to_type[m->mpc_busid] = MP_BUS_MCA; - } else { +#endif + } else printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str); +} + +#ifdef CONFIG_X86_IO_APIC + +static int bad_ioapic(unsigned long address) +{ + if (nr_ioapics >= MAX_IO_APICS) { + printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded " + "(found %d)\n", MAX_IO_APICS, nr_ioapics); + panic("Recompile kernel with bigger MAX_IO_APICS!\n"); + } + if (!address) { + printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address" + " found in table, skipping!\n"); + return 1; } + return 0; } -static void __init MP_ioapic_info (struct mpc_config_ioapic *m) +static void __init MP_ioapic_info(struct mpc_config_ioapic *m) { if (!(m->mpc_flags & MPC_APIC_USABLE)) return; printk(KERN_INFO "I/O APIC #%d Version %d at 0x%X.\n", - m->mpc_apicid, m->mpc_apicver, m->mpc_apicaddr); - if (nr_ioapics >= MAX_IO_APICS) { - printk(KERN_CRIT "Max # of I/O APICs (%d) exceeded (found %d).\n", - MAX_IO_APICS, nr_ioapics); - panic("Recompile kernel with bigger MAX_IO_APICS!.\n"); - } - if (!m->mpc_apicaddr) { - printk(KERN_ERR "WARNING: bogus zero I/O APIC address" - " found in MP table, skipping!\n"); + m->mpc_apicid, m->mpc_apicver, m->mpc_apicaddr); + + if (bad_ioapic(m->mpc_apicaddr)) return; - } + mp_ioapics[nr_ioapics] = *m; nr_ioapics++; } -static void __init MP_intsrc_info (struct mpc_config_intsrc *m) +static void __init MP_intsrc_info(struct mpc_config_intsrc *m) { - mp_irqs [mp_irq_entries] = *m; + mp_irqs[mp_irq_entries] = *m; Dprintk("Int: type %d, pol %d, trig %d, bus %d," " IRQ %02x, APIC ID %x, APIC INT %02x\n", - m->mpc_irqtype, m->mpc_irqflag & 3, - (m->mpc_irqflag >> 2) & 3, m->mpc_srcbus, - m->mpc_srcbusirq, m->mpc_dstapic, m->mpc_dstirq); + m->mpc_irqtype, m->mpc_irqflag & 3, + (m->mpc_irqflag >> 2) & 3, m->mpc_srcbus, + m->mpc_srcbusirq, m->mpc_dstapic, m->mpc_dstirq); if (++mp_irq_entries == MAX_IRQ_SOURCES) panic("Max # of irq sources exceeded!!\n"); } -static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m) +#endif + +static void __init MP_lintsrc_info(struct mpc_config_lintsrc *m) { Dprintk("Lint: type %d, pol %d, trig %d, bus %d," " IRQ %02x, APIC ID %x, APIC LINT %02x\n", - m->mpc_irqtype, m->mpc_irqflag & 3, - (m->mpc_irqflag >> 2) &3, m->mpc_srcbusid, - m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint); + m->mpc_irqtype, m->mpc_irqflag & 3, + (m->mpc_irqflag >> 2) & 3, m->mpc_srcbusid, + m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint); } #ifdef CONFIG_X86_NUMAQ -static void __init MP_translation_info (struct mpc_config_translation *m) +static void __init MP_translation_info(struct mpc_config_translation *m) { - printk(KERN_INFO "Translation: record %d, type %d, quad %d, global %d, local %d\n", mpc_record, m->trans_type, m->trans_quad, m->trans_global, m->trans_local); + printk(KERN_INFO + "Translation: record %d, type %d, quad %d, global %d, local %d\n", + mpc_record, m->trans_type, m->trans_quad, m->trans_global, + m->trans_local); - if (mpc_record >= MAX_MPC_ENTRY) + if (mpc_record >= MAX_MPC_ENTRY) printk(KERN_ERR "MAX_MPC_ENTRY exceeded!\n"); else - translation_table[mpc_record] = m; /* stash this for later */ + translation_table[mpc_record] = m; /* stash this for later */ if (m->trans_quad < MAX_NUMNODES && !node_online(m->trans_quad)) node_set_online(m->trans_quad); } @@ -312,118 +223,124 @@ static void __init MP_translation_info (struct mpc_config_translation *m) * Read/parse the MPC oem tables */ -static void __init smp_read_mpc_oem(struct mp_config_oemtable *oemtable, \ - unsigned short oemsize) +static void __init smp_read_mpc_oem(struct mp_config_oemtable *oemtable, + unsigned short oemsize) { - int count = sizeof (*oemtable); /* the header size */ - unsigned char *oemptr = ((unsigned char *)oemtable)+count; - + int count = sizeof(*oemtable); /* the header size */ + unsigned char *oemptr = ((unsigned char *)oemtable) + count; + mpc_record = 0; - printk(KERN_INFO "Found an OEM MPC table at %8p - parsing it ... \n", oemtable); - if (memcmp(oemtable->oem_signature,MPC_OEM_SIGNATURE,4)) - { - printk(KERN_WARNING "SMP mpc oemtable: bad signature [%c%c%c%c]!\n", - oemtable->oem_signature[0], - oemtable->oem_signature[1], - oemtable->oem_signature[2], - oemtable->oem_signature[3]); + printk(KERN_INFO "Found an OEM MPC table at %8p - parsing it ... \n", + oemtable); + if (memcmp(oemtable->oem_signature, MPC_OEM_SIGNATURE, 4)) { + printk(KERN_WARNING + "SMP mpc oemtable: bad signature [%c%c%c%c]!\n", + oemtable->oem_signature[0], oemtable->oem_signature[1], + oemtable->oem_signature[2], oemtable->oem_signature[3]); return; } - if (mpf_checksum((unsigned char *)oemtable,oemtable->oem_length)) - { + if (mpf_checksum((unsigned char *)oemtable, oemtable->oem_length)) { printk(KERN_WARNING "SMP oem mptable: checksum error!\n"); return; } while (count < oemtable->oem_length) { switch (*oemptr) { - case MP_TRANSLATION: + case MP_TRANSLATION: { - struct mpc_config_translation *m= - (struct mpc_config_translation *)oemptr; + struct mpc_config_translation *m = + (struct mpc_config_translation *)oemptr; MP_translation_info(m); oemptr += sizeof(*m); count += sizeof(*m); ++mpc_record; break; } - default: + default: { - printk(KERN_WARNING "Unrecognised OEM table entry type! - %d\n", (int) *oemptr); + printk(KERN_WARNING + "Unrecognised OEM table entry type! - %d\n", + (int)*oemptr); return; } } - } + } } static inline void mps_oem_check(struct mp_config_table *mpc, char *oem, - char *productid) + char *productid) { if (strncmp(oem, "IBM NUMA", 8)) printk("Warning! May not be a NUMA-Q system!\n"); if (mpc->mpc_oemptr) - smp_read_mpc_oem((struct mp_config_oemtable *) mpc->mpc_oemptr, - mpc->mpc_oemsize); + smp_read_mpc_oem((struct mp_config_oemtable *)mpc->mpc_oemptr, + mpc->mpc_oemsize); } -#endif /* CONFIG_X86_NUMAQ */ +#endif /* CONFIG_X86_NUMAQ */ /* * Read/parse the MPC */ -static int __init smp_read_mpc(struct mp_config_table *mpc) +static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early) { char str[16]; char oem[10]; - int count=sizeof(*mpc); - unsigned char *mpt=((unsigned char *)mpc)+count; + int count = sizeof(*mpc); + unsigned char *mpt = ((unsigned char *)mpc) + count; - if (memcmp(mpc->mpc_signature,MPC_SIGNATURE,4)) { - printk(KERN_ERR "SMP mptable: bad signature [0x%x]!\n", - *(u32 *)mpc->mpc_signature); + if (memcmp(mpc->mpc_signature, MPC_SIGNATURE, 4)) { + printk(KERN_ERR "MPTABLE: bad signature [%c%c%c%c]!\n", + mpc->mpc_signature[0], mpc->mpc_signature[1], + mpc->mpc_signature[2], mpc->mpc_signature[3]); return 0; } - if (mpf_checksum((unsigned char *)mpc,mpc->mpc_length)) { - printk(KERN_ERR "SMP mptable: checksum error!\n"); + if (mpf_checksum((unsigned char *)mpc, mpc->mpc_length)) { + printk(KERN_ERR "MPTABLE: checksum error!\n"); return 0; } - if (mpc->mpc_spec!=0x01 && mpc->mpc_spec!=0x04) { - printk(KERN_ERR "SMP mptable: bad table version (%d)!!\n", - mpc->mpc_spec); + if (mpc->mpc_spec != 0x01 && mpc->mpc_spec != 0x04) { + printk(KERN_ERR "MPTABLE: bad table version (%d)!!\n", + mpc->mpc_spec); return 0; } if (!mpc->mpc_lapic) { - printk(KERN_ERR "SMP mptable: null local APIC address!\n"); + printk(KERN_ERR "MPTABLE: null local APIC address!\n"); return 0; } - memcpy(oem,mpc->mpc_oem,8); - oem[8]=0; - printk(KERN_INFO "OEM ID: %s ",oem); + memcpy(oem, mpc->mpc_oem, 8); + oem[8] = 0; + printk(KERN_INFO "MPTABLE: OEM ID: %s ", oem); - memcpy(str,mpc->mpc_productid,12); - str[12]=0; - printk("Product ID: %s ",str); + memcpy(str, mpc->mpc_productid, 12); + str[12] = 0; + printk("Product ID: %s ", str); +#ifdef CONFIG_X86_32 mps_oem_check(mpc, oem, str); +#endif + printk(KERN_INFO "MPTABLE: Product ID: %s ", str); - printk("APIC at: 0x%X\n", mpc->mpc_lapic); + printk(KERN_INFO "MPTABLE: APIC at: 0x%X\n", mpc->mpc_lapic); - /* - * Save the local APIC address (it might be non-default) -- but only - * if we're not using ACPI. - */ + /* save the local APIC address, it might be non-default */ if (!acpi_lapic) mp_lapic_addr = mpc->mpc_lapic; + if (early) + return 1; + /* - * Now process the configuration blocks. + * Now process the configuration blocks. */ +#ifdef CONFIG_X86_NUMAQ mpc_record = 0; +#endif while (count < mpc->mpc_length) { - switch(*mpt) { - case MP_PROCESSOR: + switch (*mpt) { + case MP_PROCESSOR: { - struct mpc_config_processor *m= - (struct mpc_config_processor *)mpt; + struct mpc_config_processor *m = + (struct mpc_config_processor *)mpt; /* ACPI may have already provided this data */ if (!acpi_lapic) MP_processor_info(m); @@ -431,57 +348,68 @@ static int __init smp_read_mpc(struct mp_config_table *mpc) count += sizeof(*m); break; } - case MP_BUS: + case MP_BUS: { - struct mpc_config_bus *m= - (struct mpc_config_bus *)mpt; + struct mpc_config_bus *m = + (struct mpc_config_bus *)mpt; MP_bus_info(m); mpt += sizeof(*m); count += sizeof(*m); break; } - case MP_IOAPIC: + case MP_IOAPIC: { - struct mpc_config_ioapic *m= - (struct mpc_config_ioapic *)mpt; +#ifdef CONFIG_X86_IO_APIC + struct mpc_config_ioapic *m = + (struct mpc_config_ioapic *)mpt; MP_ioapic_info(m); - mpt+=sizeof(*m); - count+=sizeof(*m); +#endif + mpt += sizeof(struct mpc_config_ioapic); + count += sizeof(struct mpc_config_ioapic); break; } - case MP_INTSRC: + case MP_INTSRC: { - struct mpc_config_intsrc *m= - (struct mpc_config_intsrc *)mpt; +#ifdef CONFIG_X86_IO_APIC + struct mpc_config_intsrc *m = + (struct mpc_config_intsrc *)mpt; MP_intsrc_info(m); - mpt+=sizeof(*m); - count+=sizeof(*m); +#endif + mpt += sizeof(struct mpc_config_intsrc); + count += sizeof(struct mpc_config_intsrc); break; } - case MP_LINTSRC: + case MP_LINTSRC: { - struct mpc_config_lintsrc *m= - (struct mpc_config_lintsrc *)mpt; + struct mpc_config_lintsrc *m = + (struct mpc_config_lintsrc *)mpt; MP_lintsrc_info(m); - mpt+=sizeof(*m); - count+=sizeof(*m); - break; - } - default: - { - count = mpc->mpc_length; + mpt += sizeof(*m); + count += sizeof(*m); break; } + default: + /* wrong mptable */ + printk(KERN_ERR "Your mptable is wrong, contact your HW vendor!\n"); + printk(KERN_ERR "type %x\n", *mpt); + print_hex_dump(KERN_ERR, " ", DUMP_PREFIX_ADDRESS, 16, + 1, mpc, mpc->mpc_length, 1); + count = mpc->mpc_length; + break; } +#ifdef CONFIG_X86_NUMAQ ++mpc_record; +#endif } setup_apic_routing(); if (!num_processors) - printk(KERN_ERR "SMP mptable: no processors registered!\n"); + printk(KERN_ERR "MPTABLE: no processors registered!\n"); return num_processors; } +#ifdef CONFIG_X86_IO_APIC + static int __init ELCR_trigger(unsigned int irq) { unsigned int port; @@ -497,7 +425,7 @@ static void __init construct_default_ioirq_mptable(int mpc_default_type) int ELCR_fallback = 0; intsrc.mpc_type = MP_INTSRC; - intsrc.mpc_irqflag = 0; /* conforming */ + intsrc.mpc_irqflag = 0; /* conforming */ intsrc.mpc_srcbus = 0; intsrc.mpc_dstapic = mp_ioapics[0].mpc_apicid; @@ -512,12 +440,16 @@ static void __init construct_default_ioirq_mptable(int mpc_default_type) * If it does, we assume it's valid. */ if (mpc_default_type == 5) { - printk(KERN_INFO "ISA/PCI bus type with no IRQ information... falling back to ELCR\n"); + printk(KERN_INFO "ISA/PCI bus type with no IRQ information... " + "falling back to ELCR\n"); - if (ELCR_trigger(0) || ELCR_trigger(1) || ELCR_trigger(2) || ELCR_trigger(13)) - printk(KERN_WARNING "ELCR contains invalid data... not using ELCR\n"); + if (ELCR_trigger(0) || ELCR_trigger(1) || ELCR_trigger(2) || + ELCR_trigger(13)) + printk(KERN_ERR "ELCR contains invalid data... " + "not using ELCR\n"); else { - printk(KERN_INFO "Using ELCR to identify PCI interrupts\n"); + printk(KERN_INFO + "Using ELCR to identify PCI interrupts\n"); ELCR_fallback = 1; } } @@ -546,21 +478,25 @@ static void __init construct_default_ioirq_mptable(int mpc_default_type) } intsrc.mpc_srcbusirq = i; - intsrc.mpc_dstirq = i ? i : 2; /* IRQ0 to INTIN2 */ + intsrc.mpc_dstirq = i ? i : 2; /* IRQ0 to INTIN2 */ MP_intsrc_info(&intsrc); } intsrc.mpc_irqtype = mp_ExtINT; intsrc.mpc_srcbusirq = 0; - intsrc.mpc_dstirq = 0; /* 8259A to INTIN0 */ + intsrc.mpc_dstirq = 0; /* 8259A to INTIN0 */ MP_intsrc_info(&intsrc); } +#endif + static inline void __init construct_default_ISA_mptable(int mpc_default_type) { struct mpc_config_processor processor; struct mpc_config_bus bus; +#ifdef CONFIG_X86_IO_APIC struct mpc_config_ioapic ioapic; +#endif struct mpc_config_lintsrc lintsrc; int linttypes[2] = { mp_ExtINT, mp_NMI }; int i; @@ -578,8 +514,7 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type) processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01; processor.mpc_cpuflag = CPU_ENABLED; processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) | - (boot_cpu_data.x86_model << 4) | - boot_cpu_data.x86_mask; + (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask; processor.mpc_featureflag = boot_cpu_data.x86_capability[0]; processor.mpc_reserved[0] = 0; processor.mpc_reserved[1] = 0; @@ -591,23 +526,22 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type) bus.mpc_type = MP_BUS; bus.mpc_busid = 0; switch (mpc_default_type) { - default: - printk("???\n"); - printk(KERN_ERR "Unknown standard configuration %d\n", - mpc_default_type); - /* fall through */ - case 1: - case 5: - memcpy(bus.mpc_bustype, "ISA ", 6); - break; - case 2: - case 6: - case 3: - memcpy(bus.mpc_bustype, "EISA ", 6); - break; - case 4: - case 7: - memcpy(bus.mpc_bustype, "MCA ", 6); + default: + printk(KERN_ERR "???\nUnknown standard configuration %d\n", + mpc_default_type); + /* fall through */ + case 1: + case 5: + memcpy(bus.mpc_bustype, "ISA ", 6); + break; + case 2: + case 6: + case 3: + memcpy(bus.mpc_bustype, "EISA ", 6); + break; + case 4: + case 7: + memcpy(bus.mpc_bustype, "MCA ", 6); } MP_bus_info(&bus); if (mpc_default_type > 4) { @@ -616,6 +550,7 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type) MP_bus_info(&bus); } +#ifdef CONFIG_X86_IO_APIC ioapic.mpc_type = MP_IOAPIC; ioapic.mpc_apicid = 2; ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01; @@ -627,9 +562,9 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type) * We set up most of the low 16 IO-APIC pins according to MPS rules. */ construct_default_ioirq_mptable(mpc_default_type); - +#endif lintsrc.mpc_type = MP_LINTSRC; - lintsrc.mpc_irqflag = 0; /* conforming */ + lintsrc.mpc_irqflag = 0; /* conforming */ lintsrc.mpc_srcbusid = 0; lintsrc.mpc_srcbusirq = 0; lintsrc.mpc_destapic = MP_APIC_ALL; @@ -645,36 +580,49 @@ static struct intel_mp_floating *mpf_found; /* * Scan the memory blocks for an SMP configuration block. */ -void __init get_smp_config (void) +static void __init __get_smp_config(unsigned early) { struct intel_mp_floating *mpf = mpf_found; + if (acpi_lapic && early) + return; /* - * ACPI supports both logical (e.g. Hyper-Threading) and physical + * ACPI supports both logical (e.g. Hyper-Threading) and physical * processors, where MPS only supports physical. */ if (acpi_lapic && acpi_ioapic) { - printk(KERN_INFO "Using ACPI (MADT) for SMP configuration information\n"); + printk(KERN_INFO "Using ACPI (MADT) for SMP configuration " + "information\n"); return; - } - else if (acpi_lapic) - printk(KERN_INFO "Using ACPI for processor (LAPIC) configuration information\n"); - - printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification); - if (mpf->mpf_feature2 & (1<<7)) { + } else if (acpi_lapic) + printk(KERN_INFO "Using ACPI for processor (LAPIC) " + "configuration information\n"); + + printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n", + mpf->mpf_specification); +#ifdef CONFIG_X86_32 + if (mpf->mpf_feature2 & (1 << 7)) { printk(KERN_INFO " IMCR and PIC compatibility mode.\n"); pic_mode = 1; } else { printk(KERN_INFO " Virtual Wire compatibility mode.\n"); pic_mode = 0; } - +#endif /* * Now see if we need to read further. */ if (mpf->mpf_feature1 != 0) { + if (early) { + /* + * local APIC has default address + */ + mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; + return; + } - printk(KERN_INFO "Default MP configuration #%d\n", mpf->mpf_feature1); + printk(KERN_INFO "Default MP configuration #%d\n", + mpf->mpf_feature1); construct_default_ISA_mptable(mpf->mpf_feature1); } else if (mpf->mpf_physptr) { @@ -683,12 +631,18 @@ void __init get_smp_config (void) * Read the physical hardware table. Anything here will * override the defaults. */ - if (!smp_read_mpc(phys_to_virt(mpf->mpf_physptr))) { + if (!smp_read_mpc(phys_to_virt(mpf->mpf_physptr), early)) { smp_found_config = 0; - printk(KERN_ERR "BIOS bug, MP table errors detected!...\n"); - printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n"); + printk(KERN_ERR + "BIOS bug, MP table errors detected!...\n"); + printk(KERN_ERR "... disabling SMP support. " + "(tell your hw vendor)\n"); return; } + + if (early) + return; +#ifdef CONFIG_X86_IO_APIC /* * If there are no explicit MP IRQ entries, then we are * broken. We set up most of the low 16 IO-APIC pins to @@ -697,7 +651,9 @@ void __init get_smp_config (void) if (!mp_irq_entries) { struct mpc_config_bus bus; - printk(KERN_ERR "BIOS bug, no explicit IRQ entries, using default mptable. (tell your hw vendor)\n"); + printk(KERN_ERR "BIOS bug, no explicit IRQ entries, " + "using default mptable. " + "(tell your hw vendor)\n"); bus.mpc_type = MP_BUS; bus.mpc_busid = 0; @@ -706,36 +662,51 @@ void __init get_smp_config (void) construct_default_ioirq_mptable(0); } - +#endif } else BUG(); - printk(KERN_INFO "Processors: %d\n", num_processors); + if (!early) + printk(KERN_INFO "Processors: %d\n", num_processors); /* * Only use the first configuration found. */ } -static int __init smp_scan_config (unsigned long base, unsigned long length) +void __init early_get_smp_config(void) +{ + __get_smp_config(1); +} + +void __init get_smp_config(void) { - unsigned long *bp = phys_to_virt(base); + __get_smp_config(0); +} + +static int __init smp_scan_config(unsigned long base, unsigned long length, + unsigned reserve) +{ + extern void __bad_mpf_size(void); + unsigned int *bp = phys_to_virt(base); struct intel_mp_floating *mpf; - printk(KERN_INFO "Scan SMP from %p for %ld bytes.\n", bp,length); + Dprintk("Scan SMP from %p for %ld bytes.\n", bp, length); if (sizeof(*mpf) != 16) - printk("Error: MPF size\n"); + __bad_mpf_size(); while (length > 0) { mpf = (struct intel_mp_floating *)bp; if ((*bp == SMP_MAGIC_IDENT) && - (mpf->mpf_length == 1) && - !mpf_checksum((unsigned char *)bp, 16) && - ((mpf->mpf_specification == 1) - || (mpf->mpf_specification == 4)) ) { + (mpf->mpf_length == 1) && + !mpf_checksum((unsigned char *)bp, 16) && + ((mpf->mpf_specification == 1) + || (mpf->mpf_specification == 4))) { smp_found_config = 1; + mpf_found = mpf; +#ifdef CONFIG_X86_32 printk(KERN_INFO "found SMP MP-table at [%p] %08lx\n", - mpf, virt_to_phys(mpf)); + mpf, virt_to_phys(mpf)); reserve_bootmem(virt_to_phys(mpf), PAGE_SIZE, BOOTMEM_DEFAULT); if (mpf->mpf_physptr) { @@ -756,8 +727,16 @@ static int __init smp_scan_config (unsigned long base, unsigned long length) BOOTMEM_DEFAULT); } - mpf_found = mpf; - return 1; +#else + if (!reserve) + return 1; + + reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE); + if (mpf->mpf_physptr) + reserve_bootmem_generic(mpf->mpf_physptr, + PAGE_SIZE); +#endif + return 1; } bp += 4; length -= 16; @@ -765,7 +744,7 @@ static int __init smp_scan_config (unsigned long base, unsigned long length) return 0; } -void __init find_smp_config (void) +static void __init __find_smp_config(unsigned reserve) { unsigned int address; @@ -777,9 +756,9 @@ void __init find_smp_config (void) * 2) Scan the top 1K of base RAM * 3) Scan the 64K of bios */ - if (smp_scan_config(0x0,0x400) || - smp_scan_config(639*0x400,0x400) || - smp_scan_config(0xF0000,0x10000)) + if (smp_scan_config(0x0, 0x400, reserve) || + smp_scan_config(639 * 0x400, 0x400, reserve) || + smp_scan_config(0xF0000, 0x10000, reserve)) return; /* * If it is an SMP machine we should know now, unless the @@ -800,144 +779,113 @@ void __init find_smp_config (void) address = get_bios_ebda(); if (address) - smp_scan_config(address, 0x400); + smp_scan_config(address, 0x400, reserve); } -int es7000_plat; - -/* -------------------------------------------------------------------------- - ACPI-based MP Configuration - -------------------------------------------------------------------------- */ - -#ifdef CONFIG_ACPI - -void __init mp_register_lapic_address(u64 address) +void __init early_find_smp_config(void) { - mp_lapic_addr = (unsigned long) address; - - set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr); - - if (boot_cpu_physical_apicid == -1U) - boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID)); - - Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid); + __find_smp_config(0); } -void __cpuinit mp_register_lapic (u8 id, u8 enabled) +void __init find_smp_config(void) { - struct mpc_config_processor processor; - int boot_cpu = 0; - - if (MAX_APICS - id <= 0) { - printk(KERN_WARNING "Processor #%d invalid (max %d)\n", - id, MAX_APICS); - return; - } - - if (id == boot_cpu_physical_apicid) - boot_cpu = 1; + __find_smp_config(1); +} - processor.mpc_type = MP_PROCESSOR; - processor.mpc_apicid = id; - processor.mpc_apicver = GET_APIC_VERSION(apic_read(APIC_LVR)); - processor.mpc_cpuflag = (enabled ? CPU_ENABLED : 0); - processor.mpc_cpuflag |= (boot_cpu ? CPU_BOOTPROCESSOR : 0); - processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) | - (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask; - processor.mpc_featureflag = boot_cpu_data.x86_capability[0]; - processor.mpc_reserved[0] = 0; - processor.mpc_reserved[1] = 0; +/* -------------------------------------------------------------------------- + ACPI-based MP Configuration + -------------------------------------------------------------------------- */ - MP_processor_info(&processor); -} +#ifdef CONFIG_ACPI #ifdef CONFIG_X86_IO_APIC #define MP_ISA_BUS 0 #define MP_MAX_IOAPIC_PIN 127 -static struct mp_ioapic_routing { - int apic_id; - int gsi_base; - int gsi_end; - u32 pin_programmed[4]; -} mp_ioapic_routing[MAX_IO_APICS]; +extern struct mp_ioapic_routing mp_ioapic_routing[MAX_IO_APICS]; -static int mp_find_ioapic (int gsi) +static int mp_find_ioapic(int gsi) { int i = 0; /* Find the IOAPIC that manages this GSI. */ for (i = 0; i < nr_ioapics; i++) { if ((gsi >= mp_ioapic_routing[i].gsi_base) - && (gsi <= mp_ioapic_routing[i].gsi_end)) + && (gsi <= mp_ioapic_routing[i].gsi_end)) return i; } printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi); - return -1; } -void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base) +static u8 uniq_ioapic_id(u8 id) +{ +#ifdef CONFIG_X86_32 + if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && + !APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) + return io_apic_get_unique_id(nr_ioapics, id); + else + return id; +#else + int i; + DECLARE_BITMAP(used, 256); + bitmap_zero(used, 256); + for (i = 0; i < nr_ioapics; i++) { + struct mpc_config_ioapic *ia = &mp_ioapics[i]; + __set_bit(ia->mpc_apicid, used); + } + if (!test_bit(id, used)) + return id; + return find_first_zero_bit(used, 256); +#endif +} + +void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) { int idx = 0; - int tmpid; - if (nr_ioapics >= MAX_IO_APICS) { - printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded " - "(found %d)\n", MAX_IO_APICS, nr_ioapics); - panic("Recompile kernel with bigger MAX_IO_APICS!\n"); - } - if (!address) { - printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address" - " found in MADT table, skipping!\n"); + if (bad_ioapic(address)) return; - } - idx = nr_ioapics++; + idx = nr_ioapics; mp_ioapics[idx].mpc_type = MP_IOAPIC; mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE; mp_ioapics[idx].mpc_apicaddr = address; set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address); - if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) - && !APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) - tmpid = io_apic_get_unique_id(idx, id); - else - tmpid = id; - if (tmpid == -1) { - nr_ioapics--; - return; - } - mp_ioapics[idx].mpc_apicid = tmpid; + mp_ioapics[idx].mpc_apicid = uniq_ioapic_id(id); +#ifdef CONFIG_X86_32 mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx); - - /* +#else + mp_ioapics[idx].mpc_apicver = 0; +#endif + /* * Build basic GSI lookup table to facilitate gsi->io_apic lookups * and to prevent reprogramming of IOAPIC pins (PCI GSIs). */ mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mpc_apicid; mp_ioapic_routing[idx].gsi_base = gsi_base; mp_ioapic_routing[idx].gsi_end = gsi_base + - io_apic_get_redir_entries(idx); + io_apic_get_redir_entries(idx); - printk("IOAPIC[%d]: apic_id %d, version %d, address 0x%x, " + printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, " "GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid, mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr, - mp_ioapic_routing[idx].gsi_base, - mp_ioapic_routing[idx].gsi_end); + mp_ioapic_routing[idx].gsi_base, mp_ioapic_routing[idx].gsi_end); + + nr_ioapics++; } -void __init -mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) +void __init mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) { struct mpc_config_intsrc intsrc; - int ioapic = -1; - int pin = -1; + int ioapic = -1; + int pin = -1; - /* + /* * Convert 'gsi' to 'ioapic.pin'. */ ioapic = mp_find_ioapic(gsi); @@ -947,7 +895,7 @@ mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) /* * TBD: This check is for faulty timer entries, where the override - * erroneously sets the trigger to level, resulting in a HUGE + * erroneously sets the trigger to level, resulting in a HUGE * increase of timer interrupts! */ if ((bus_irq == 0) && (trigger == 3)) @@ -957,13 +905,13 @@ mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) intsrc.mpc_irqtype = mp_INT; intsrc.mpc_irqflag = (trigger << 2) | polarity; intsrc.mpc_srcbus = MP_ISA_BUS; - intsrc.mpc_srcbusirq = bus_irq; /* IRQ */ - intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; /* APIC ID */ - intsrc.mpc_dstirq = pin; /* INTIN# */ + intsrc.mpc_srcbusirq = bus_irq; /* IRQ */ + intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; /* APIC ID */ + intsrc.mpc_dstirq = pin; /* INTIN# */ Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, %d-%d\n", - intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3, - (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus, + intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3, + (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus, intsrc.mpc_srcbusirq, intsrc.mpc_dstapic, intsrc.mpc_dstirq); mp_irqs[mp_irq_entries] = intsrc; @@ -971,16 +919,21 @@ mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) panic("Max # of irq sources exceeded!\n"); } -void __init mp_config_acpi_legacy_irqs (void) +int es7000_plat; + +void __init mp_config_acpi_legacy_irqs(void) { struct mpc_config_intsrc intsrc; int i = 0; int ioapic = -1; - /* +#if defined (CONFIG_MCA) || defined (CONFIG_EISA) + /* * Fabricate the legacy ISA bus (bus #31). */ mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA; +#endif + set_bit(MP_ISA_BUS, mp_bus_not_pci); Dprintk("Bus #%d is ISA\n", MP_ISA_BUS); /* @@ -989,19 +942,20 @@ void __init mp_config_acpi_legacy_irqs (void) if (es7000_plat == 1) return; - /* - * Locate the IOAPIC that manages the ISA IRQs (0-15). + /* + * Locate the IOAPIC that manages the ISA IRQs (0-15). */ ioapic = mp_find_ioapic(0); if (ioapic < 0) return; intsrc.mpc_type = MP_INTSRC; - intsrc.mpc_irqflag = 0; /* Conforming */ + intsrc.mpc_irqflag = 0; /* Conforming */ intsrc.mpc_srcbus = MP_ISA_BUS; +#ifdef CONFIG_X86_IO_APIC intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; - - /* +#endif + /* * Use the default configuration for the IRQs 0-15. Unless * overridden by (MADT) interrupt source override entries. */ @@ -1012,28 +966,29 @@ void __init mp_config_acpi_legacy_irqs (void) struct mpc_config_intsrc *irq = mp_irqs + idx; /* Do we already have a mapping for this ISA IRQ? */ - if (irq->mpc_srcbus == MP_ISA_BUS && irq->mpc_srcbusirq == i) + if (irq->mpc_srcbus == MP_ISA_BUS + && irq->mpc_srcbusirq == i) break; /* Do we already have a mapping for this IOAPIC pin */ if ((irq->mpc_dstapic == intsrc.mpc_dstapic) && - (irq->mpc_dstirq == i)) + (irq->mpc_dstirq == i)) break; } if (idx != mp_irq_entries) { printk(KERN_DEBUG "ACPI: IRQ%d used by override.\n", i); - continue; /* IRQ already used */ + continue; /* IRQ already used */ } intsrc.mpc_irqtype = mp_INT; - intsrc.mpc_srcbusirq = i; /* Identity mapped */ + intsrc.mpc_srcbusirq = i; /* Identity mapped */ intsrc.mpc_dstirq = i; Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, " - "%d-%d\n", intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3, - (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus, - intsrc.mpc_srcbusirq, intsrc.mpc_dstapic, + "%d-%d\n", intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3, + (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus, + intsrc.mpc_srcbusirq, intsrc.mpc_dstapic, intsrc.mpc_dstirq); mp_irqs[mp_irq_entries] = intsrc; @@ -1042,21 +997,27 @@ void __init mp_config_acpi_legacy_irqs (void) } } -#define MAX_GSI_NUM 4096 -#define IRQ_COMPRESSION_START 64 - int mp_register_gsi(u32 gsi, int triggering, int polarity) { int ioapic = -1; int ioapic_pin = 0; int idx, bit = 0; +#ifdef CONFIG_X86_32 +#define MAX_GSI_NUM 4096 +#define IRQ_COMPRESSION_START 64 + static int pci_irq = IRQ_COMPRESSION_START; /* * Mapping between Global System Interrupts, which * represent all possible interrupts, and IRQs * assigned to actual devices. */ - static int gsi_to_irq[MAX_GSI_NUM]; + static int gsi_to_irq[MAX_GSI_NUM]; +#else + + if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC) + return gsi; +#endif /* Don't set up the ACPI SCI because it's already set up */ if (acpi_gbl_FADT.sci_interrupt == gsi) @@ -1070,11 +1031,13 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity) ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_base; +#ifdef CONFIG_X86_32 if (ioapic_renumber_irq) gsi = ioapic_renumber_irq(ioapic, gsi); +#endif - /* - * Avoid pin reprogramming. PRTs typically include entries + /* + * Avoid pin reprogramming. PRTs typically include entries * with redundant pin->gsi mappings (but unique PCI devices); * we only program the IOAPIC on the first. */ @@ -1082,23 +1045,27 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity) idx = (ioapic_pin < 32) ? 0 : (ioapic_pin / 32); if (idx > 3) { printk(KERN_ERR "Invalid reference to IOAPIC pin " - "%d-%d\n", mp_ioapic_routing[ioapic].apic_id, - ioapic_pin); + "%d-%d\n", mp_ioapic_routing[ioapic].apic_id, + ioapic_pin); return gsi; } - if ((1<<bit) & mp_ioapic_routing[ioapic].pin_programmed[idx]) { + if ((1 << bit) & mp_ioapic_routing[ioapic].pin_programmed[idx]) { Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n", mp_ioapic_routing[ioapic].apic_id, ioapic_pin); +#ifdef CONFIG_X86_32 return (gsi < IRQ_COMPRESSION_START ? gsi : gsi_to_irq[gsi]); +#else + return gsi; +#endif } - mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit); - + mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1 << bit); +#ifdef CONFIG_X86_32 /* * For GSI >= 64, use IRQ compression */ if ((gsi >= IRQ_COMPRESSION_START) - && (triggering == ACPI_LEVEL_SENSITIVE)) { + && (triggering == ACPI_LEVEL_SENSITIVE)) { /* * For PCI devices assign IRQs in order, avoiding gaps * due to unused I/O APIC pins. @@ -1115,8 +1082,7 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity) * So test for this condition, and if necessary, avoid * the pin collision. */ - if (gsi > 15 || (gsi == 0 && !timer_uses_ioapic_pin_0)) - gsi = pci_irq++; + gsi = pci_irq++; /* * Don't assign IRQ used by ACPI SCI */ @@ -1128,10 +1094,10 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity) return gsi; } } - +#endif io_apic_set_pci_routing(ioapic, ioapic_pin, gsi, - triggering == ACPI_EDGE_SENSITIVE ? 0 : 1, - polarity == ACPI_ACTIVE_HIGH ? 0 : 1); + triggering == ACPI_EDGE_SENSITIVE ? 0 : 1, + polarity == ACPI_ACTIVE_HIGH ? 0 : 1); return gsi; } diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c deleted file mode 100644 index 72ab1403fed..00000000000 --- a/arch/x86/kernel/mpparse_64.c +++ /dev/null @@ -1,867 +0,0 @@ -/* - * Intel Multiprocessor Specification 1.1 and 1.4 - * compliant MP-table parsing routines. - * - * (c) 1995 Alan Cox, Building #3 <alan@redhat.com> - * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com> - * - * Fixes - * Erich Boleyn : MP v1.4 and additional changes. - * Alan Cox : Added EBDA scanning - * Ingo Molnar : various cleanups and rewrites - * Maciej W. Rozycki: Bits for default MP configurations - * Paul Diefenbaugh: Added full ACPI support - */ - -#include <linux/mm.h> -#include <linux/init.h> -#include <linux/delay.h> -#include <linux/bootmem.h> -#include <linux/kernel_stat.h> -#include <linux/mc146818rtc.h> -#include <linux/acpi.h> -#include <linux/module.h> - -#include <asm/smp.h> -#include <asm/mtrr.h> -#include <asm/mpspec.h> -#include <asm/pgalloc.h> -#include <asm/io_apic.h> -#include <asm/proto.h> -#include <asm/acpi.h> - -/* Have we found an MP table */ -int smp_found_config; - -/* - * Various Linux-internal data structures created from the - * MP-table. - */ -DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); -int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 }; - -static int mp_current_pci_id = 0; -/* I/O APIC entries */ -struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS]; - -/* # of MP IRQ source entries */ -struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; - -/* MP IRQ source entries */ -int mp_irq_entries; - -int nr_ioapics; -unsigned long mp_lapic_addr = 0; - - - -/* Processor that is doing the boot up */ -unsigned int boot_cpu_id = -1U; -EXPORT_SYMBOL(boot_cpu_id); - -/* Internal processor count */ -unsigned int num_processors; - -unsigned disabled_cpus __cpuinitdata; - -/* Bitmask of physically existing CPUs */ -physid_mask_t phys_cpu_present_map = PHYSID_MASK_NONE; - -u16 x86_bios_cpu_apicid_init[NR_CPUS] __initdata - = { [0 ... NR_CPUS-1] = BAD_APICID }; -void *x86_bios_cpu_apicid_early_ptr; -DEFINE_PER_CPU(u16, x86_bios_cpu_apicid) = BAD_APICID; -EXPORT_PER_CPU_SYMBOL(x86_bios_cpu_apicid); - - -/* - * Intel MP BIOS table parsing routines: - */ - -/* - * Checksum an MP configuration block. - */ - -static int __init mpf_checksum(unsigned char *mp, int len) -{ - int sum = 0; - - while (len--) - sum += *mp++; - - return sum & 0xFF; -} - -static void __cpuinit MP_processor_info(struct mpc_config_processor *m) -{ - int cpu; - cpumask_t tmp_map; - char *bootup_cpu = ""; - - if (!(m->mpc_cpuflag & CPU_ENABLED)) { - disabled_cpus++; - return; - } - if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) { - bootup_cpu = " (Bootup-CPU)"; - boot_cpu_id = m->mpc_apicid; - } - - printk(KERN_INFO "Processor #%d%s\n", m->mpc_apicid, bootup_cpu); - - if (num_processors >= NR_CPUS) { - printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached." - " Processor ignored.\n", NR_CPUS); - return; - } - - num_processors++; - cpus_complement(tmp_map, cpu_present_map); - cpu = first_cpu(tmp_map); - - physid_set(m->mpc_apicid, phys_cpu_present_map); - if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) { - /* - * x86_bios_cpu_apicid is required to have processors listed - * in same order as logical cpu numbers. Hence the first - * entry is BSP, and so on. - */ - cpu = 0; - } - /* are we being called early in kernel startup? */ - if (x86_cpu_to_apicid_early_ptr) { - u16 *cpu_to_apicid = x86_cpu_to_apicid_early_ptr; - u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr; - - cpu_to_apicid[cpu] = m->mpc_apicid; - bios_cpu_apicid[cpu] = m->mpc_apicid; - } else { - per_cpu(x86_cpu_to_apicid, cpu) = m->mpc_apicid; - per_cpu(x86_bios_cpu_apicid, cpu) = m->mpc_apicid; - } - - cpu_set(cpu, cpu_possible_map); - cpu_set(cpu, cpu_present_map); -} - -static void __init MP_bus_info (struct mpc_config_bus *m) -{ - char str[7]; - - memcpy(str, m->mpc_bustype, 6); - str[6] = 0; - Dprintk("Bus #%d is %s\n", m->mpc_busid, str); - - if (strncmp(str, "ISA", 3) == 0) { - set_bit(m->mpc_busid, mp_bus_not_pci); - } else if (strncmp(str, "PCI", 3) == 0) { - clear_bit(m->mpc_busid, mp_bus_not_pci); - mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id; - mp_current_pci_id++; - } else { - printk(KERN_ERR "Unknown bustype %s\n", str); - } -} - -static int bad_ioapic(unsigned long address) -{ - if (nr_ioapics >= MAX_IO_APICS) { - printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded " - "(found %d)\n", MAX_IO_APICS, nr_ioapics); - panic("Recompile kernel with bigger MAX_IO_APICS!\n"); - } - if (!address) { - printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address" - " found in table, skipping!\n"); - return 1; - } - return 0; -} - -static void __init MP_ioapic_info (struct mpc_config_ioapic *m) -{ - if (!(m->mpc_flags & MPC_APIC_USABLE)) - return; - - printk("I/O APIC #%d at 0x%X.\n", - m->mpc_apicid, m->mpc_apicaddr); - - if (bad_ioapic(m->mpc_apicaddr)) - return; - - mp_ioapics[nr_ioapics] = *m; - nr_ioapics++; -} - -static void __init MP_intsrc_info (struct mpc_config_intsrc *m) -{ - mp_irqs [mp_irq_entries] = *m; - Dprintk("Int: type %d, pol %d, trig %d, bus %d," - " IRQ %02x, APIC ID %x, APIC INT %02x\n", - m->mpc_irqtype, m->mpc_irqflag & 3, - (m->mpc_irqflag >> 2) & 3, m->mpc_srcbus, - m->mpc_srcbusirq, m->mpc_dstapic, m->mpc_dstirq); - if (++mp_irq_entries >= MAX_IRQ_SOURCES) - panic("Max # of irq sources exceeded!!\n"); -} - -static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m) -{ - Dprintk("Lint: type %d, pol %d, trig %d, bus %d," - " IRQ %02x, APIC ID %x, APIC LINT %02x\n", - m->mpc_irqtype, m->mpc_irqflag & 3, - (m->mpc_irqflag >> 2) &3, m->mpc_srcbusid, - m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint); -} - -/* - * Read/parse the MPC - */ - -static int __init smp_read_mpc(struct mp_config_table *mpc) -{ - char str[16]; - int count=sizeof(*mpc); - unsigned char *mpt=((unsigned char *)mpc)+count; - - if (memcmp(mpc->mpc_signature,MPC_SIGNATURE,4)) { - printk("MPTABLE: bad signature [%c%c%c%c]!\n", - mpc->mpc_signature[0], - mpc->mpc_signature[1], - mpc->mpc_signature[2], - mpc->mpc_signature[3]); - return 0; - } - if (mpf_checksum((unsigned char *)mpc,mpc->mpc_length)) { - printk("MPTABLE: checksum error!\n"); - return 0; - } - if (mpc->mpc_spec!=0x01 && mpc->mpc_spec!=0x04) { - printk(KERN_ERR "MPTABLE: bad table version (%d)!!\n", - mpc->mpc_spec); - return 0; - } - if (!mpc->mpc_lapic) { - printk(KERN_ERR "MPTABLE: null local APIC address!\n"); - return 0; - } - memcpy(str,mpc->mpc_oem,8); - str[8] = 0; - printk(KERN_INFO "MPTABLE: OEM ID: %s ",str); - - memcpy(str,mpc->mpc_productid,12); - str[12] = 0; - printk("MPTABLE: Product ID: %s ",str); - - printk("MPTABLE: APIC at: 0x%X\n",mpc->mpc_lapic); - - /* save the local APIC address, it might be non-default */ - if (!acpi_lapic) - mp_lapic_addr = mpc->mpc_lapic; - - /* - * Now process the configuration blocks. - */ - while (count < mpc->mpc_length) { - switch(*mpt) { - case MP_PROCESSOR: - { - struct mpc_config_processor *m= - (struct mpc_config_processor *)mpt; - if (!acpi_lapic) - MP_processor_info(m); - mpt += sizeof(*m); - count += sizeof(*m); - break; - } - case MP_BUS: - { - struct mpc_config_bus *m= - (struct mpc_config_bus *)mpt; - MP_bus_info(m); - mpt += sizeof(*m); - count += sizeof(*m); - break; - } - case MP_IOAPIC: - { - struct mpc_config_ioapic *m= - (struct mpc_config_ioapic *)mpt; - MP_ioapic_info(m); - mpt += sizeof(*m); - count += sizeof(*m); - break; - } - case MP_INTSRC: - { - struct mpc_config_intsrc *m= - (struct mpc_config_intsrc *)mpt; - - MP_intsrc_info(m); - mpt += sizeof(*m); - count += sizeof(*m); - break; - } - case MP_LINTSRC: - { - struct mpc_config_lintsrc *m= - (struct mpc_config_lintsrc *)mpt; - MP_lintsrc_info(m); - mpt += sizeof(*m); - count += sizeof(*m); - break; - } - } - } - setup_apic_routing(); - if (!num_processors) - printk(KERN_ERR "MPTABLE: no processors registered!\n"); - return num_processors; -} - -static int __init ELCR_trigger(unsigned int irq) -{ - unsigned int port; - - port = 0x4d0 + (irq >> 3); - return (inb(port) >> (irq & 7)) & 1; -} - -static void __init construct_default_ioirq_mptable(int mpc_default_type) -{ - struct mpc_config_intsrc intsrc; - int i; - int ELCR_fallback = 0; - - intsrc.mpc_type = MP_INTSRC; - intsrc.mpc_irqflag = 0; /* conforming */ - intsrc.mpc_srcbus = 0; - intsrc.mpc_dstapic = mp_ioapics[0].mpc_apicid; - - intsrc.mpc_irqtype = mp_INT; - - /* - * If true, we have an ISA/PCI system with no IRQ entries - * in the MP table. To prevent the PCI interrupts from being set up - * incorrectly, we try to use the ELCR. The sanity check to see if - * there is good ELCR data is very simple - IRQ0, 1, 2 and 13 can - * never be level sensitive, so we simply see if the ELCR agrees. - * If it does, we assume it's valid. - */ - if (mpc_default_type == 5) { - printk(KERN_INFO "ISA/PCI bus type with no IRQ information... falling back to ELCR\n"); - - if (ELCR_trigger(0) || ELCR_trigger(1) || ELCR_trigger(2) || ELCR_trigger(13)) - printk(KERN_ERR "ELCR contains invalid data... not using ELCR\n"); - else { - printk(KERN_INFO "Using ELCR to identify PCI interrupts\n"); - ELCR_fallback = 1; - } - } - - for (i = 0; i < 16; i++) { - switch (mpc_default_type) { - case 2: - if (i == 0 || i == 13) - continue; /* IRQ0 & IRQ13 not connected */ - /* fall through */ - default: - if (i == 2) - continue; /* IRQ2 is never connected */ - } - - if (ELCR_fallback) { - /* - * If the ELCR indicates a level-sensitive interrupt, we - * copy that information over to the MP table in the - * irqflag field (level sensitive, active high polarity). - */ - if (ELCR_trigger(i)) - intsrc.mpc_irqflag = 13; - else - intsrc.mpc_irqflag = 0; - } - - intsrc.mpc_srcbusirq = i; - intsrc.mpc_dstirq = i ? i : 2; /* IRQ0 to INTIN2 */ - MP_intsrc_info(&intsrc); - } - - intsrc.mpc_irqtype = mp_ExtINT; - intsrc.mpc_srcbusirq = 0; - intsrc.mpc_dstirq = 0; /* 8259A to INTIN0 */ - MP_intsrc_info(&intsrc); -} - -static inline void __init construct_default_ISA_mptable(int mpc_default_type) -{ - struct mpc_config_processor processor; - struct mpc_config_bus bus; - struct mpc_config_ioapic ioapic; - struct mpc_config_lintsrc lintsrc; - int linttypes[2] = { mp_ExtINT, mp_NMI }; - int i; - - /* - * local APIC has default address - */ - mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; - - /* - * 2 CPUs, numbered 0 & 1. - */ - processor.mpc_type = MP_PROCESSOR; - processor.mpc_apicver = 0; - processor.mpc_cpuflag = CPU_ENABLED; - processor.mpc_cpufeature = 0; - processor.mpc_featureflag = 0; - processor.mpc_reserved[0] = 0; - processor.mpc_reserved[1] = 0; - for (i = 0; i < 2; i++) { - processor.mpc_apicid = i; - MP_processor_info(&processor); - } - - bus.mpc_type = MP_BUS; - bus.mpc_busid = 0; - switch (mpc_default_type) { - default: - printk(KERN_ERR "???\nUnknown standard configuration %d\n", - mpc_default_type); - /* fall through */ - case 1: - case 5: - memcpy(bus.mpc_bustype, "ISA ", 6); - break; - } - MP_bus_info(&bus); - if (mpc_default_type > 4) { - bus.mpc_busid = 1; - memcpy(bus.mpc_bustype, "PCI ", 6); - MP_bus_info(&bus); - } - - ioapic.mpc_type = MP_IOAPIC; - ioapic.mpc_apicid = 2; - ioapic.mpc_apicver = 0; - ioapic.mpc_flags = MPC_APIC_USABLE; - ioapic.mpc_apicaddr = 0xFEC00000; - MP_ioapic_info(&ioapic); - - /* - * We set up most of the low 16 IO-APIC pins according to MPS rules. - */ - construct_default_ioirq_mptable(mpc_default_type); - - lintsrc.mpc_type = MP_LINTSRC; - lintsrc.mpc_irqflag = 0; /* conforming */ - lintsrc.mpc_srcbusid = 0; - lintsrc.mpc_srcbusirq = 0; - lintsrc.mpc_destapic = MP_APIC_ALL; - for (i = 0; i < 2; i++) { - lintsrc.mpc_irqtype = linttypes[i]; - lintsrc.mpc_destapiclint = i; - MP_lintsrc_info(&lintsrc); - } -} - -static struct intel_mp_floating *mpf_found; - -/* - * Scan the memory blocks for an SMP configuration block. - */ -void __init get_smp_config (void) -{ - struct intel_mp_floating *mpf = mpf_found; - - /* - * ACPI supports both logical (e.g. Hyper-Threading) and physical - * processors, where MPS only supports physical. - */ - if (acpi_lapic && acpi_ioapic) { - printk(KERN_INFO "Using ACPI (MADT) for SMP configuration information\n"); - return; - } - else if (acpi_lapic) - printk(KERN_INFO "Using ACPI for processor (LAPIC) configuration information\n"); - - printk("Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification); - - /* - * Now see if we need to read further. - */ - if (mpf->mpf_feature1 != 0) { - - printk(KERN_INFO "Default MP configuration #%d\n", mpf->mpf_feature1); - construct_default_ISA_mptable(mpf->mpf_feature1); - - } else if (mpf->mpf_physptr) { - - /* - * Read the physical hardware table. Anything here will - * override the defaults. - */ - if (!smp_read_mpc(phys_to_virt(mpf->mpf_physptr))) { - smp_found_config = 0; - printk(KERN_ERR "BIOS bug, MP table errors detected!...\n"); - printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n"); - return; - } - /* - * If there are no explicit MP IRQ entries, then we are - * broken. We set up most of the low 16 IO-APIC pins to - * ISA defaults and hope it will work. - */ - if (!mp_irq_entries) { - struct mpc_config_bus bus; - - printk(KERN_ERR "BIOS bug, no explicit IRQ entries, using default mptable. (tell your hw vendor)\n"); - - bus.mpc_type = MP_BUS; - bus.mpc_busid = 0; - memcpy(bus.mpc_bustype, "ISA ", 6); - MP_bus_info(&bus); - - construct_default_ioirq_mptable(0); - } - - } else - BUG(); - - printk(KERN_INFO "Processors: %d\n", num_processors); - /* - * Only use the first configuration found. - */ -} - -static int __init smp_scan_config (unsigned long base, unsigned long length) -{ - extern void __bad_mpf_size(void); - unsigned int *bp = phys_to_virt(base); - struct intel_mp_floating *mpf; - - Dprintk("Scan SMP from %p for %ld bytes.\n", bp,length); - if (sizeof(*mpf) != 16) - __bad_mpf_size(); - - while (length > 0) { - mpf = (struct intel_mp_floating *)bp; - if ((*bp == SMP_MAGIC_IDENT) && - (mpf->mpf_length == 1) && - !mpf_checksum((unsigned char *)bp, 16) && - ((mpf->mpf_specification == 1) - || (mpf->mpf_specification == 4)) ) { - - smp_found_config = 1; - reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE); - if (mpf->mpf_physptr) - reserve_bootmem_generic(mpf->mpf_physptr, PAGE_SIZE); - mpf_found = mpf; - return 1; - } - bp += 4; - length -= 16; - } - return 0; -} - -void __init find_smp_config(void) -{ - unsigned int address; - - /* - * FIXME: Linux assumes you have 640K of base ram.. - * this continues the error... - * - * 1) Scan the bottom 1K for a signature - * 2) Scan the top 1K of base RAM - * 3) Scan the 64K of bios - */ - if (smp_scan_config(0x0,0x400) || - smp_scan_config(639*0x400,0x400) || - smp_scan_config(0xF0000,0x10000)) - return; - /* - * If it is an SMP machine we should know now. - * - * there is a real-mode segmented pointer pointing to the - * 4K EBDA area at 0x40E, calculate and scan it here. - * - * NOTE! There are Linux loaders that will corrupt the EBDA - * area, and as such this kind of SMP config may be less - * trustworthy, simply because the SMP table may have been - * stomped on during early boot. These loaders are buggy and - * should be fixed. - */ - - address = *(unsigned short *)phys_to_virt(0x40E); - address <<= 4; - if (smp_scan_config(address, 0x1000)) - return; - - /* If we have come this far, we did not find an MP table */ - printk(KERN_INFO "No mptable found.\n"); -} - -/* -------------------------------------------------------------------------- - ACPI-based MP Configuration - -------------------------------------------------------------------------- */ - -#ifdef CONFIG_ACPI - -void __init mp_register_lapic_address(u64 address) -{ - mp_lapic_addr = (unsigned long) address; - set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr); - if (boot_cpu_id == -1U) - boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID)); -} - -void __cpuinit mp_register_lapic (u8 id, u8 enabled) -{ - struct mpc_config_processor processor; - int boot_cpu = 0; - - if (id == boot_cpu_id) - boot_cpu = 1; - - processor.mpc_type = MP_PROCESSOR; - processor.mpc_apicid = id; - processor.mpc_apicver = 0; - processor.mpc_cpuflag = (enabled ? CPU_ENABLED : 0); - processor.mpc_cpuflag |= (boot_cpu ? CPU_BOOTPROCESSOR : 0); - processor.mpc_cpufeature = 0; - processor.mpc_featureflag = 0; - processor.mpc_reserved[0] = 0; - processor.mpc_reserved[1] = 0; - - MP_processor_info(&processor); -} - -#define MP_ISA_BUS 0 -#define MP_MAX_IOAPIC_PIN 127 - -static struct mp_ioapic_routing { - int apic_id; - int gsi_start; - int gsi_end; - u32 pin_programmed[4]; -} mp_ioapic_routing[MAX_IO_APICS]; - -static int mp_find_ioapic(int gsi) -{ - int i = 0; - - /* Find the IOAPIC that manages this GSI. */ - for (i = 0; i < nr_ioapics; i++) { - if ((gsi >= mp_ioapic_routing[i].gsi_start) - && (gsi <= mp_ioapic_routing[i].gsi_end)) - return i; - } - - printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi); - return -1; -} - -static u8 uniq_ioapic_id(u8 id) -{ - int i; - DECLARE_BITMAP(used, 256); - bitmap_zero(used, 256); - for (i = 0; i < nr_ioapics; i++) { - struct mpc_config_ioapic *ia = &mp_ioapics[i]; - __set_bit(ia->mpc_apicid, used); - } - if (!test_bit(id, used)) - return id; - return find_first_zero_bit(used, 256); -} - -void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base) -{ - int idx = 0; - - if (bad_ioapic(address)) - return; - - idx = nr_ioapics; - - mp_ioapics[idx].mpc_type = MP_IOAPIC; - mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE; - mp_ioapics[idx].mpc_apicaddr = address; - - set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address); - mp_ioapics[idx].mpc_apicid = uniq_ioapic_id(id); - mp_ioapics[idx].mpc_apicver = 0; - - /* - * Build basic IRQ lookup table to facilitate gsi->io_apic lookups - * and to prevent reprogramming of IOAPIC pins (PCI IRQs). - */ - mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mpc_apicid; - mp_ioapic_routing[idx].gsi_start = gsi_base; - mp_ioapic_routing[idx].gsi_end = gsi_base + - io_apic_get_redir_entries(idx); - - printk(KERN_INFO "IOAPIC[%d]: apic_id %d, address 0x%x, " - "GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid, - mp_ioapics[idx].mpc_apicaddr, - mp_ioapic_routing[idx].gsi_start, - mp_ioapic_routing[idx].gsi_end); - - nr_ioapics++; -} - -void __init -mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) -{ - struct mpc_config_intsrc intsrc; - int ioapic = -1; - int pin = -1; - - /* - * Convert 'gsi' to 'ioapic.pin'. - */ - ioapic = mp_find_ioapic(gsi); - if (ioapic < 0) - return; - pin = gsi - mp_ioapic_routing[ioapic].gsi_start; - - /* - * TBD: This check is for faulty timer entries, where the override - * erroneously sets the trigger to level, resulting in a HUGE - * increase of timer interrupts! - */ - if ((bus_irq == 0) && (trigger == 3)) - trigger = 1; - - intsrc.mpc_type = MP_INTSRC; - intsrc.mpc_irqtype = mp_INT; - intsrc.mpc_irqflag = (trigger << 2) | polarity; - intsrc.mpc_srcbus = MP_ISA_BUS; - intsrc.mpc_srcbusirq = bus_irq; /* IRQ */ - intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; /* APIC ID */ - intsrc.mpc_dstirq = pin; /* INTIN# */ - - Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, %d-%d\n", - intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3, - (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus, - intsrc.mpc_srcbusirq, intsrc.mpc_dstapic, intsrc.mpc_dstirq); - - mp_irqs[mp_irq_entries] = intsrc; - if (++mp_irq_entries == MAX_IRQ_SOURCES) - panic("Max # of irq sources exceeded!\n"); -} - -void __init mp_config_acpi_legacy_irqs(void) -{ - struct mpc_config_intsrc intsrc; - int i = 0; - int ioapic = -1; - - /* - * Fabricate the legacy ISA bus (bus #31). - */ - set_bit(MP_ISA_BUS, mp_bus_not_pci); - - /* - * Locate the IOAPIC that manages the ISA IRQs (0-15). - */ - ioapic = mp_find_ioapic(0); - if (ioapic < 0) - return; - - intsrc.mpc_type = MP_INTSRC; - intsrc.mpc_irqflag = 0; /* Conforming */ - intsrc.mpc_srcbus = MP_ISA_BUS; - intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; - - /* - * Use the default configuration for the IRQs 0-15. Unless - * overridden by (MADT) interrupt source override entries. - */ - for (i = 0; i < 16; i++) { - int idx; - - for (idx = 0; idx < mp_irq_entries; idx++) { - struct mpc_config_intsrc *irq = mp_irqs + idx; - - /* Do we already have a mapping for this ISA IRQ? */ - if (irq->mpc_srcbus == MP_ISA_BUS && irq->mpc_srcbusirq == i) - break; - - /* Do we already have a mapping for this IOAPIC pin */ - if ((irq->mpc_dstapic == intsrc.mpc_dstapic) && - (irq->mpc_dstirq == i)) - break; - } - - if (idx != mp_irq_entries) { - printk(KERN_DEBUG "ACPI: IRQ%d used by override.\n", i); - continue; /* IRQ already used */ - } - - intsrc.mpc_irqtype = mp_INT; - intsrc.mpc_srcbusirq = i; /* Identity mapped */ - intsrc.mpc_dstirq = i; - - Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, " - "%d-%d\n", intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3, - (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus, - intsrc.mpc_srcbusirq, intsrc.mpc_dstapic, - intsrc.mpc_dstirq); - - mp_irqs[mp_irq_entries] = intsrc; - if (++mp_irq_entries == MAX_IRQ_SOURCES) - panic("Max # of irq sources exceeded!\n"); - } -} - -int mp_register_gsi(u32 gsi, int triggering, int polarity) -{ - int ioapic = -1; - int ioapic_pin = 0; - int idx, bit = 0; - - if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC) - return gsi; - - /* Don't set up the ACPI SCI because it's already set up */ - if (acpi_gbl_FADT.sci_interrupt == gsi) - return gsi; - - ioapic = mp_find_ioapic(gsi); - if (ioapic < 0) { - printk(KERN_WARNING "No IOAPIC for GSI %u\n", gsi); - return gsi; - } - - ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_start; - - /* - * Avoid pin reprogramming. PRTs typically include entries - * with redundant pin->gsi mappings (but unique PCI devices); - * we only program the IOAPIC on the first. - */ - bit = ioapic_pin % 32; - idx = (ioapic_pin < 32) ? 0 : (ioapic_pin / 32); - if (idx > 3) { - printk(KERN_ERR "Invalid reference to IOAPIC pin " - "%d-%d\n", mp_ioapic_routing[ioapic].apic_id, - ioapic_pin); - return gsi; - } - if ((1<<bit) & mp_ioapic_routing[ioapic].pin_programmed[idx]) { - Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n", - mp_ioapic_routing[ioapic].apic_id, ioapic_pin); - return gsi; - } - - mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit); - - io_apic_set_pci_routing(ioapic, ioapic_pin, gsi, - triggering == ACPI_EDGE_SENSITIVE ? 0 : 1, - polarity == ACPI_ACTIVE_HIGH ? 0 : 1); - return gsi; -} -#endif /*CONFIG_ACPI*/ diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c index af51ea8400b..4dfb4053005 100644 --- a/arch/x86/kernel/msr.c +++ b/arch/x86/kernel/msr.c @@ -65,8 +65,8 @@ static loff_t msr_seek(struct file *file, loff_t offset, int orig) return ret; } -static ssize_t msr_read(struct file *file, char __user * buf, - size_t count, loff_t * ppos) +static ssize_t msr_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) { u32 __user *tmp = (u32 __user *) buf; u32 data[2]; diff --git a/arch/x86/kernel/nmi_32.c b/arch/x86/kernel/nmi_32.c index 6a0aa703868..8421d0ac6f2 100644 --- a/arch/x86/kernel/nmi_32.c +++ b/arch/x86/kernel/nmi_32.c @@ -22,9 +22,11 @@ #include <linux/cpumask.h> #include <linux/kernel_stat.h> #include <linux/kdebug.h> +#include <linux/slab.h> #include <asm/smp.h> #include <asm/nmi.h> +#include <asm/timer.h> #include "mach_traps.h" @@ -67,7 +69,7 @@ static __init void nmi_cpu_busy(void *data) } #endif -static int __init check_nmi_watchdog(void) +int __init check_nmi_watchdog(void) { unsigned int *prev_nmi_count; int cpu; @@ -80,7 +82,7 @@ static int __init check_nmi_watchdog(void) prev_nmi_count = kmalloc(NR_CPUS * sizeof(int), GFP_KERNEL); if (!prev_nmi_count) - return -1; + goto error; printk(KERN_INFO "Testing NMI watchdog ... "); @@ -117,7 +119,7 @@ static int __init check_nmi_watchdog(void) if (!atomic_read(&nmi_active)) { kfree(prev_nmi_count); atomic_set(&nmi_active, -1); - return -1; + goto error; } printk("OK.\n"); @@ -128,9 +130,11 @@ static int __init check_nmi_watchdog(void) kfree(prev_nmi_count); return 0; +error: + timer_ack = !cpu_has_tsc; + + return -1; } -/* This needs to happen later in boot so counters are working */ -late_initcall(check_nmi_watchdog); static int __init setup_nmi_watchdog(char *str) { diff --git a/arch/x86/kernel/nmi_64.c b/arch/x86/kernel/nmi_64.c index 9a4fde74bee..11f9130ac51 100644 --- a/arch/x86/kernel/nmi_64.c +++ b/arch/x86/kernel/nmi_64.c @@ -26,6 +26,8 @@ #include <asm/proto.h> #include <asm/mce.h> +#include <mach_traps.h> + int unknown_nmi_panic; int nmi_watchdog_enabled; int panic_on_unrecovered_nmi; diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 075962cc75a..3733412d135 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -206,13 +206,6 @@ static struct resource reserve_ioports = { .flags = IORESOURCE_IO | IORESOURCE_BUSY, }; -static struct resource reserve_iomem = { - .start = 0, - .end = -1, - .name = "paravirt-iomem", - .flags = IORESOURCE_MEM | IORESOURCE_BUSY, -}; - /* * Reserve the whole legacy IO space to prevent any legacy drivers * from wasting time probing for their hardware. This is a fairly @@ -222,16 +215,7 @@ static struct resource reserve_iomem = { */ int paravirt_disable_iospace(void) { - int ret; - - ret = request_resource(&ioport_resource, &reserve_ioports); - if (ret == 0) { - ret = request_resource(&iomem_resource, &reserve_iomem); - if (ret) - release_resource(&reserve_ioports); - } - - return ret; + return request_resource(&ioport_resource, &reserve_ioports); } static DEFINE_PER_CPU(enum paravirt_lazy_mode, paravirt_lazy_mode) = PARAVIRT_LAZY_NONE; diff --git a/arch/x86/kernel/pci-dma_64.c b/arch/x86/kernel/pci-dma_64.c index 375cb2bc45b..ada5a060499 100644 --- a/arch/x86/kernel/pci-dma_64.c +++ b/arch/x86/kernel/pci-dma_64.c @@ -232,32 +232,32 @@ static __init int iommu_setup(char *p) return -EINVAL; while (*p) { - if (!strncmp(p,"off",3)) + if (!strncmp(p, "off", 3)) no_iommu = 1; /* gart_parse_options has more force support */ - if (!strncmp(p,"force",5)) + if (!strncmp(p, "force", 5)) force_iommu = 1; - if (!strncmp(p,"noforce",7)) { + if (!strncmp(p, "noforce", 7)) { iommu_merge = 0; force_iommu = 0; } - if (!strncmp(p, "biomerge",8)) { + if (!strncmp(p, "biomerge", 8)) { iommu_bio_merge = 4096; iommu_merge = 1; force_iommu = 1; } - if (!strncmp(p, "panic",5)) + if (!strncmp(p, "panic", 5)) panic_on_overflow = 1; - if (!strncmp(p, "nopanic",7)) + if (!strncmp(p, "nopanic", 7)) panic_on_overflow = 0; - if (!strncmp(p, "merge",5)) { + if (!strncmp(p, "merge", 5)) { iommu_merge = 1; force_iommu = 1; } - if (!strncmp(p, "nomerge",7)) + if (!strncmp(p, "nomerge", 7)) iommu_merge = 0; - if (!strncmp(p, "forcesac",8)) + if (!strncmp(p, "forcesac", 8)) iommu_sac_force = 1; if (!strncmp(p, "allowdac", 8)) forbid_dac = 0; @@ -265,7 +265,7 @@ static __init int iommu_setup(char *p) forbid_dac = -1; #ifdef CONFIG_SWIOTLB - if (!strncmp(p, "soft",4)) + if (!strncmp(p, "soft", 4)) swiotlb = 1; #endif diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index be3c7a299f0..3903a8f2eb9 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -82,7 +82,6 @@ unsigned long thread_saved_pc(struct task_struct *tsk) */ void (*pm_idle)(void); EXPORT_SYMBOL(pm_idle); -static DEFINE_PER_CPU(unsigned int, cpu_idle_state); void disable_hlt(void) { @@ -114,20 +113,13 @@ void default_idle(void) local_irq_disable(); if (!need_resched()) { - ktime_t t0, t1; - u64 t0n, t1n; - - t0 = ktime_get(); - t0n = ktime_to_ns(t0); safe_halt(); /* enables interrupts racelessly */ local_irq_disable(); - t1 = ktime_get(); - t1n = ktime_to_ns(t1); - sched_clock_idle_wakeup_event(t1n - t0n); } local_irq_enable(); current_thread_info()->status |= TS_POLLING; } else { + local_irq_enable(); /* loop is done by the caller */ cpu_relax(); } @@ -143,6 +135,7 @@ EXPORT_SYMBOL(default_idle); */ static void poll_idle(void) { + local_irq_enable(); cpu_relax(); } @@ -190,9 +183,6 @@ void cpu_idle(void) while (!need_resched()) { void (*idle)(void); - if (__get_cpu_var(cpu_idle_state)) - __get_cpu_var(cpu_idle_state) = 0; - check_pgt_cache(); rmb(); idle = pm_idle; @@ -220,40 +210,19 @@ static void do_nothing(void *unused) { } +/* + * cpu_idle_wait - Used to ensure that all the CPUs discard old value of + * pm_idle and update to new pm_idle value. Required while changing pm_idle + * handler on SMP systems. + * + * Caller must have changed pm_idle to the new value before the call. Old + * pm_idle value will not be used by any CPU after the return of this function. + */ void cpu_idle_wait(void) { - unsigned int cpu, this_cpu = get_cpu(); - cpumask_t map, tmp = current->cpus_allowed; - - set_cpus_allowed(current, cpumask_of_cpu(this_cpu)); - put_cpu(); - - cpus_clear(map); - for_each_online_cpu(cpu) { - per_cpu(cpu_idle_state, cpu) = 1; - cpu_set(cpu, map); - } - - __get_cpu_var(cpu_idle_state) = 0; - - wmb(); - do { - ssleep(1); - for_each_online_cpu(cpu) { - if (cpu_isset(cpu, map) && !per_cpu(cpu_idle_state, cpu)) - cpu_clear(cpu, map); - } - cpus_and(map, map, cpu_online_map); - /* - * We waited 1 sec, if a CPU still did not call idle - * it may be because it is in idle and not waking up - * because it has nothing to do. - * Give all the remaining CPUS a kick. - */ - smp_call_function_mask(map, do_nothing, NULL, 0); - } while (!cpus_empty(map)); - - set_cpus_allowed(current, tmp); + smp_mb(); + /* kick all the CPUs so that they exit out of pm_idle */ + smp_call_function(do_nothing, NULL, 0, 1); } EXPORT_SYMBOL_GPL(cpu_idle_wait); @@ -273,8 +242,11 @@ void mwait_idle_with_hints(unsigned long ax, unsigned long cx) __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); if (!need_resched()) - __mwait(ax, cx); - } + __sti_mwait(ax, cx); + else + local_irq_enable(); + } else + local_irq_enable(); } /* Default MONITOR/MWAIT with no hints, used for default C1 state */ @@ -357,7 +329,7 @@ void __show_registers(struct pt_regs *regs, int all) init_utsname()->version); printk("EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n", - 0xffff & regs->cs, regs->ip, regs->flags, + (u16)regs->cs, regs->ip, regs->flags, smp_processor_id()); print_symbol("EIP is at %s\n", regs->ip); @@ -366,8 +338,7 @@ void __show_registers(struct pt_regs *regs, int all) printk("ESI: %08lx EDI: %08lx EBP: %08lx ESP: %08lx\n", regs->si, regs->di, regs->bp, sp); printk(" DS: %04x ES: %04x FS: %04x GS: %04x SS: %04x\n", - regs->ds & 0xffff, regs->es & 0xffff, - regs->fs & 0xffff, gs, ss); + (u16)regs->ds, (u16)regs->es, (u16)regs->fs, gs, ss); if (!all) return; @@ -538,6 +509,21 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, return err; } +void +start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) +{ + __asm__("movl %0, %%gs" :: "r"(0)); + regs->fs = 0; + set_fs(USER_DS); + regs->ds = __USER_DS; + regs->es = __USER_DS; + regs->ss = __USER_DS; + regs->cs = __USER_CS; + regs->ip = new_ip; + regs->sp = new_sp; +} +EXPORT_SYMBOL_GPL(start_thread); + #ifdef CONFIG_SECCOMP static void hard_disable_TSC(void) { @@ -575,12 +561,12 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, /* we clear debugctl to make sure DS * is not in use when we change it */ debugctl = 0; - wrmsrl(MSR_IA32_DEBUGCTLMSR, 0); + update_debugctlmsr(0); wrmsr(MSR_IA32_DS_AREA, next->ds_area_msr, 0); } if (next->debugctlmsr != debugctl) - wrmsr(MSR_IA32_DEBUGCTLMSR, next->debugctlmsr, 0); + update_debugctlmsr(next->debugctlmsr); if (test_tsk_thread_flag(next_p, TIF_DEBUG)) { set_debugreg(next->debugreg0, 0); diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 3baf9b9f4c8..e75ccc8a2b8 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -63,7 +63,6 @@ EXPORT_SYMBOL(boot_option_idle_override); */ void (*pm_idle)(void); EXPORT_SYMBOL(pm_idle); -static DEFINE_PER_CPU(unsigned int, cpu_idle_state); static ATOMIC_NOTIFIER_HEAD(idle_notifier); @@ -108,16 +107,8 @@ void default_idle(void) smp_mb(); local_irq_disable(); if (!need_resched()) { - ktime_t t0, t1; - u64 t0n, t1n; - - t0 = ktime_get(); - t0n = ktime_to_ns(t0); safe_halt(); /* enables interrupts racelessly */ local_irq_disable(); - t1 = ktime_get(); - t1n = ktime_to_ns(t1); - sched_clock_idle_wakeup_event(t1n - t0n); } local_irq_enable(); current_thread_info()->status |= TS_POLLING; @@ -173,9 +164,6 @@ void cpu_idle(void) while (!need_resched()) { void (*idle)(void); - if (__get_cpu_var(cpu_idle_state)) - __get_cpu_var(cpu_idle_state) = 0; - rmb(); idle = pm_idle; if (!idle) @@ -207,40 +195,19 @@ static void do_nothing(void *unused) { } +/* + * cpu_idle_wait - Used to ensure that all the CPUs discard old value of + * pm_idle and update to new pm_idle value. Required while changing pm_idle + * handler on SMP systems. + * + * Caller must have changed pm_idle to the new value before the call. Old + * pm_idle value will not be used by any CPU after the return of this function. + */ void cpu_idle_wait(void) { - unsigned int cpu, this_cpu = get_cpu(); - cpumask_t map, tmp = current->cpus_allowed; - - set_cpus_allowed(current, cpumask_of_cpu(this_cpu)); - put_cpu(); - - cpus_clear(map); - for_each_online_cpu(cpu) { - per_cpu(cpu_idle_state, cpu) = 1; - cpu_set(cpu, map); - } - - __get_cpu_var(cpu_idle_state) = 0; - - wmb(); - do { - ssleep(1); - for_each_online_cpu(cpu) { - if (cpu_isset(cpu, map) && !per_cpu(cpu_idle_state, cpu)) - cpu_clear(cpu, map); - } - cpus_and(map, map, cpu_online_map); - /* - * We waited 1 sec, if a CPU still did not call idle - * it may be because it is in idle and not waking up - * because it has nothing to do. - * Give all the remaining CPUS a kick. - */ - smp_call_function_mask(map, do_nothing, 0, 0); - } while (!cpus_empty(map)); - - set_cpus_allowed(current, tmp); + smp_mb(); + /* kick all the CPUs so that they exit out of pm_idle */ + smp_call_function(do_nothing, NULL, 0, 1); } EXPORT_SYMBOL_GPL(cpu_idle_wait); @@ -553,6 +520,21 @@ out: return err; } +void +start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) +{ + asm volatile("movl %0, %%fs; movl %0, %%es; movl %0, %%ds" :: "r"(0)); + load_gs_index(0); + regs->ip = new_ip; + regs->sp = new_sp; + write_pda(oldrsp, new_sp); + regs->cs = __USER_CS; + regs->ss = __USER_DS; + regs->flags = 0x200; + set_fs(USER_DS); +} +EXPORT_SYMBOL_GPL(start_thread); + /* * This special macro can be used to load a debugging register */ @@ -573,12 +555,12 @@ static inline void __switch_to_xtra(struct task_struct *prev_p, /* we clear debugctl to make sure DS * is not in use when we change it */ debugctl = 0; - wrmsrl(MSR_IA32_DEBUGCTLMSR, 0); + update_debugctlmsr(0); wrmsrl(MSR_IA32_DS_AREA, next->ds_area_msr); } if (next->debugctlmsr != debugctl) - wrmsrl(MSR_IA32_DEBUGCTLMSR, next->debugctlmsr); + update_debugctlmsr(next->debugctlmsr); if (test_tsk_thread_flag(next_p, TIF_DEBUG)) { loaddebug(next, 0); diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index eb92ccbb350..559c1b02741 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -1456,7 +1456,6 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code) /* notification of system call entry/exit * - triggered by current->work.syscall_trace */ -__attribute__((regparm(3))) int do_syscall_trace(struct pt_regs *regs, int entryexit) { int is_sysemu = test_thread_flag(TIF_SYSCALL_EMU); diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 484c4a80d38..9692202d3bf 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -1,5 +1,4 @@ #include <linux/module.h> -#include <linux/init.h> #include <linux/reboot.h> #include <linux/init.h> #include <linux/pm.h> @@ -412,12 +411,12 @@ static void native_machine_shutdown(void) #ifdef CONFIG_X86_32 /* See if there has been given a command line override */ if ((reboot_cpu != -1) && (reboot_cpu < NR_CPUS) && - cpu_isset(reboot_cpu, cpu_online_map)) + cpu_online(reboot_cpu)) reboot_cpu_id = reboot_cpu; #endif /* Make certain the cpu I'm about to reboot on is online */ - if (!cpu_isset(reboot_cpu_id, cpu_online_map)) + if (!cpu_online(reboot_cpu_id)) reboot_cpu_id = smp_processor_id(); /* Make certain I only run on the appropriate processor */ diff --git a/arch/x86/kernel/relocate_kernel_32.S b/arch/x86/kernel/relocate_kernel_32.S index f151d6fae46..c30fe25d470 100644 --- a/arch/x86/kernel/relocate_kernel_32.S +++ b/arch/x86/kernel/relocate_kernel_32.S @@ -9,18 +9,19 @@ #include <linux/linkage.h> #include <asm/page.h> #include <asm/kexec.h> +#include <asm/processor-flags.h> +#include <asm/pgtable.h> /* * Must be relocatable PIC code callable as a C function */ #define PTR(x) (x << 2) -#define PAGE_ALIGNED (1 << PAGE_SHIFT) -#define PAGE_ATTR 0x63 /* _PAGE_PRESENT|_PAGE_RW|_PAGE_ACCESSED|_PAGE_DIRTY */ -#define PAE_PGD_ATTR 0x01 /* _PAGE_PRESENT */ +#define PAGE_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY) +#define PAE_PGD_ATTR (_PAGE_PRESENT) .text - .align PAGE_ALIGNED + .align PAGE_SIZE .globl relocate_kernel relocate_kernel: movl 8(%esp), %ebp /* list of pages */ @@ -155,7 +156,7 @@ relocate_new_kernel: movl %eax, %cr3 /* setup a new stack at the end of the physical control page */ - lea 4096(%edi), %esp + lea PAGE_SIZE(%edi), %esp /* jump to identity mapped page */ movl %edi, %eax @@ -168,16 +169,16 @@ identity_mapped: pushl %edx /* Set cr0 to a known state: - * 31 0 == Paging disabled - * 18 0 == Alignment check disabled - * 16 0 == Write protect disabled - * 3 0 == No task switch - * 2 0 == Don't do FP software emulation. - * 0 1 == Proctected mode enabled + * - Paging disabled + * - Alignment check disabled + * - Write protect disabled + * - No task switch + * - Don't do FP software emulation. + * - Proctected mode enabled */ movl %cr0, %eax - andl $~((1<<31)|(1<<18)|(1<<16)|(1<<3)|(1<<2)), %eax - orl $(1<<0), %eax + andl $~(X86_CR0_PG | X86_CR0_AM | X86_CR0_WP | X86_CR0_TS | X86_CR0_EM), %eax + orl $(X86_CR0_PE), %eax movl %eax, %cr0 /* clear cr4 if applicable */ @@ -186,8 +187,7 @@ identity_mapped: /* Set cr4 to a known state: * Setting everything to zero seems safe. */ - movl %cr4, %eax - andl $0, %eax + xorl %eax, %eax movl %eax, %cr4 jmp 1f diff --git a/arch/x86/kernel/relocate_kernel_64.S b/arch/x86/kernel/relocate_kernel_64.S index 14e95872c6a..f5afe665a82 100644 --- a/arch/x86/kernel/relocate_kernel_64.S +++ b/arch/x86/kernel/relocate_kernel_64.S @@ -9,17 +9,18 @@ #include <linux/linkage.h> #include <asm/page.h> #include <asm/kexec.h> +#include <asm/processor-flags.h> +#include <asm/pgtable.h> /* * Must be relocatable PIC code callable as a C function */ #define PTR(x) (x << 3) -#define PAGE_ALIGNED (1 << PAGE_SHIFT) -#define PAGE_ATTR 0x63 /* _PAGE_PRESENT|_PAGE_RW|_PAGE_ACCESSED|_PAGE_DIRTY */ +#define PAGE_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY) .text - .align PAGE_ALIGNED + .align PAGE_SIZE .code64 .globl relocate_kernel relocate_kernel: @@ -160,7 +161,7 @@ relocate_new_kernel: movq %r9, %cr3 /* setup a new stack at the end of the physical control page */ - lea 4096(%r8), %rsp + lea PAGE_SIZE(%r8), %rsp /* jump to identity mapped page */ addq $(identity_mapped - relocate_kernel), %r8 @@ -172,33 +173,22 @@ identity_mapped: pushq %rdx /* Set cr0 to a known state: - * 31 1 == Paging enabled - * 18 0 == Alignment check disabled - * 16 0 == Write protect disabled - * 3 0 == No task switch - * 2 0 == Don't do FP software emulation. - * 0 1 == Proctected mode enabled + * - Paging enabled + * - Alignment check disabled + * - Write protect disabled + * - No task switch + * - Don't do FP software emulation. + * - Proctected mode enabled */ movq %cr0, %rax - andq $~((1<<18)|(1<<16)|(1<<3)|(1<<2)), %rax - orl $((1<<31)|(1<<0)), %eax + andq $~(X86_CR0_AM | X86_CR0_WP | X86_CR0_TS | X86_CR0_EM), %rax + orl $(X86_CR0_PG | X86_CR0_PE), %eax movq %rax, %cr0 /* Set cr4 to a known state: - * 10 0 == xmm exceptions disabled - * 9 0 == xmm registers instructions disabled - * 8 0 == performance monitoring counter disabled - * 7 0 == page global disabled - * 6 0 == machine check exceptions disabled - * 5 1 == physical address extension enabled - * 4 0 == page size extensions disabled - * 3 0 == Debug extensions disabled - * 2 0 == Time stamp disable (disabled) - * 1 0 == Protected mode virtual interrupts disabled - * 0 0 == VME disabled + * - physical address extension enabled */ - - movq $((1<<5)), %rax + movq $X86_CR4_PAE, %rax movq %rax, %cr4 jmp 1f diff --git a/arch/x86/kernel/rtc.c b/arch/x86/kernel/rtc.c index eb9b1a198f5..9615eee9b77 100644 --- a/arch/x86/kernel/rtc.c +++ b/arch/x86/kernel/rtc.c @@ -9,7 +9,6 @@ #include <asm/vsyscall.h> #ifdef CONFIG_X86_32 -# define CMOS_YEARS_OFFS 1900 /* * 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 @@ -17,14 +16,11 @@ */ volatile unsigned long cmos_lock = 0; EXPORT_SYMBOL(cmos_lock); -#else -/* - * x86-64 systems only exists since 2002. - * This will work up to Dec 31, 2100 - */ -# define CMOS_YEARS_OFFS 2000 #endif +/* For two digit years assume time is always after that */ +#define CMOS_YEARS_OFFS 2000 + DEFINE_SPINLOCK(rtc_lock); EXPORT_SYMBOL(rtc_lock); @@ -98,7 +94,7 @@ int mach_set_rtc_mmss(unsigned long nowtime) unsigned long mach_get_cmos_time(void) { - unsigned int year, mon, day, hour, min, sec, century = 0; + unsigned int status, year, mon, day, hour, min, sec, century = 0; /* * If UIP is clear, then we have >= 244 microseconds before @@ -116,14 +112,16 @@ unsigned long mach_get_cmos_time(void) mon = CMOS_READ(RTC_MONTH); year = CMOS_READ(RTC_YEAR); -#if defined(CONFIG_ACPI) && defined(CONFIG_X86_64) - /* CHECKME: Is this really 64bit only ??? */ +#ifdef CONFIG_ACPI if (acpi_gbl_FADT.header.revision >= FADT2_REVISION_ID && acpi_gbl_FADT.century) century = CMOS_READ(acpi_gbl_FADT.century); #endif - if (RTC_ALWAYS_BCD || !(CMOS_READ(RTC_CONTROL) & RTC_DM_BINARY)) { + status = CMOS_READ(RTC_CONTROL); + WARN_ON_ONCE(RTC_ALWAYS_BCD && (status & RTC_DM_BINARY)); + + if (RTC_ALWAYS_BCD || !(status & RTC_DM_BINARY)) { BCD_TO_BIN(sec); BCD_TO_BIN(min); BCD_TO_BIN(hour); @@ -136,11 +134,8 @@ unsigned long mach_get_cmos_time(void) BCD_TO_BIN(century); year += century * 100; printk(KERN_INFO "Extended CMOS year: %d\n", century * 100); - } else { + } else year += CMOS_YEARS_OFFS; - if (year < 1970) - year += 100; - } return mktime(year, mon, day, hour, min, sec); } @@ -151,8 +146,8 @@ unsigned char rtc_cmos_read(unsigned char addr) unsigned char val; lock_cmos_prefix(addr); - outb_p(addr, RTC_PORT(0)); - val = inb_p(RTC_PORT(1)); + outb(addr, RTC_PORT(0)); + val = inb(RTC_PORT(1)); lock_cmos_suffix(addr); return val; } @@ -161,8 +156,8 @@ EXPORT_SYMBOL(rtc_cmos_read); void rtc_cmos_write(unsigned char val, unsigned char addr) { lock_cmos_prefix(addr); - outb_p(addr, RTC_PORT(0)); - outb_p(val, RTC_PORT(1)); + outb(addr, RTC_PORT(0)); + outb(val, RTC_PORT(1)); lock_cmos_suffix(addr); } EXPORT_SYMBOL(rtc_cmos_write); diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c new file mode 100644 index 00000000000..ed157c90412 --- /dev/null +++ b/arch/x86/kernel/setup.c @@ -0,0 +1,113 @@ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/bootmem.h> +#include <linux/percpu.h> +#include <asm/smp.h> +#include <asm/percpu.h> +#include <asm/sections.h> +#include <asm/processor.h> +#include <asm/setup.h> +#include <asm/topology.h> +#include <asm/mpspec.h> +#include <asm/apicdef.h> + +unsigned int num_processors; +unsigned disabled_cpus __cpuinitdata; +/* Processor that is doing the boot up */ +unsigned int boot_cpu_physical_apicid = -1U; +EXPORT_SYMBOL(boot_cpu_physical_apicid); + +physid_mask_t phys_cpu_present_map; + +DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID; +EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid); + +/* Bitmask of physically existing CPUs */ +physid_mask_t phys_cpu_present_map; + +#if defined(CONFIG_HAVE_SETUP_PER_CPU_AREA) && defined(CONFIG_SMP) +/* + * Copy data used in early init routines from the initial arrays to the + * per cpu data areas. These arrays then become expendable and the + * *_early_ptr's are zeroed indicating that the static arrays are gone. + */ +static void __init setup_per_cpu_maps(void) +{ + int cpu; + + for_each_possible_cpu(cpu) { + per_cpu(x86_cpu_to_apicid, cpu) = x86_cpu_to_apicid_init[cpu]; + per_cpu(x86_bios_cpu_apicid, cpu) = + x86_bios_cpu_apicid_init[cpu]; +#ifdef CONFIG_NUMA + per_cpu(x86_cpu_to_node_map, cpu) = + x86_cpu_to_node_map_init[cpu]; +#endif + } + + /* indicate the early static arrays will soon be gone */ + x86_cpu_to_apicid_early_ptr = NULL; + x86_bios_cpu_apicid_early_ptr = NULL; +#ifdef CONFIG_NUMA + x86_cpu_to_node_map_early_ptr = NULL; +#endif +} + +#ifdef CONFIG_X86_32 +/* + * Great future not-so-futuristic plan: make i386 and x86_64 do it + * the same way + */ +unsigned long __per_cpu_offset[NR_CPUS] __read_mostly; +EXPORT_SYMBOL(__per_cpu_offset); +#endif + +/* + * Great future plan: + * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data. + * Always point %gs to its beginning + */ +void __init setup_per_cpu_areas(void) +{ + int i; + unsigned long size; + +#ifdef CONFIG_HOTPLUG_CPU + prefill_possible_map(); +#endif + + /* Copy section for each CPU (we discard the original) */ + size = PERCPU_ENOUGH_ROOM; + printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n", + size); + + for_each_possible_cpu(i) { + char *ptr; +#ifndef CONFIG_NEED_MULTIPLE_NODES + ptr = alloc_bootmem_pages(size); +#else + int node = early_cpu_to_node(i); + if (!node_online(node) || !NODE_DATA(node)) { + ptr = alloc_bootmem_pages(size); + printk(KERN_INFO + "cpu %d has no node or node-local memory\n", i); + } + else + ptr = alloc_bootmem_pages_node(NODE_DATA(node), size); +#endif + if (!ptr) + panic("Cannot allocate cpu data for CPU %d\n", i); +#ifdef CONFIG_X86_64 + cpu_pda(i)->data_offset = ptr - __per_cpu_start; +#else + __per_cpu_offset[i] = ptr - __per_cpu_start; +#endif + memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start); + } + + /* Setup percpu data maps */ + setup_per_cpu_maps(); +} + +#endif diff --git a/arch/x86/kernel/setup64.c b/arch/x86/kernel/setup64.c index e24c4567709..9042fb0e36f 100644 --- a/arch/x86/kernel/setup64.c +++ b/arch/x86/kernel/setup64.c @@ -11,6 +11,7 @@ #include <linux/bootmem.h> #include <linux/bitops.h> #include <linux/module.h> +#include <linux/kgdb.h> #include <asm/pda.h> #include <asm/pgtable.h> #include <asm/processor.h> @@ -23,6 +24,7 @@ #include <asm/proto.h> #include <asm/sections.h> #include <asm/setup.h> +#include <asm/genapic.h> #ifndef CONFIG_DEBUG_BOOT_PARAMS struct boot_params __initdata boot_params; @@ -85,83 +87,6 @@ static int __init nonx32_setup(char *str) } __setup("noexec32=", nonx32_setup); -/* - * Copy data used in early init routines from the initial arrays to the - * per cpu data areas. These arrays then become expendable and the - * *_early_ptr's are zeroed indicating that the static arrays are gone. - */ -static void __init setup_per_cpu_maps(void) -{ - int cpu; - - for_each_possible_cpu(cpu) { -#ifdef CONFIG_SMP - if (per_cpu_offset(cpu)) { -#endif - per_cpu(x86_cpu_to_apicid, cpu) = - x86_cpu_to_apicid_init[cpu]; - per_cpu(x86_bios_cpu_apicid, cpu) = - x86_bios_cpu_apicid_init[cpu]; -#ifdef CONFIG_NUMA - per_cpu(x86_cpu_to_node_map, cpu) = - x86_cpu_to_node_map_init[cpu]; -#endif -#ifdef CONFIG_SMP - } - else - printk(KERN_NOTICE "per_cpu_offset zero for cpu %d\n", - cpu); -#endif - } - - /* indicate the early static arrays will soon be gone */ - x86_cpu_to_apicid_early_ptr = NULL; - x86_bios_cpu_apicid_early_ptr = NULL; -#ifdef CONFIG_NUMA - x86_cpu_to_node_map_early_ptr = NULL; -#endif -} - -/* - * Great future plan: - * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data. - * Always point %gs to its beginning - */ -void __init setup_per_cpu_areas(void) -{ - int i; - unsigned long size; - -#ifdef CONFIG_HOTPLUG_CPU - prefill_possible_map(); -#endif - - /* Copy section for each CPU (we discard the original) */ - size = PERCPU_ENOUGH_ROOM; - - printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n", size); - for_each_cpu_mask (i, cpu_possible_map) { - char *ptr; -#ifndef CONFIG_NEED_MULTIPLE_NODES - ptr = alloc_bootmem_pages(size); -#else - int node = early_cpu_to_node(i); - - if (!node_online(node) || !NODE_DATA(node)) - ptr = alloc_bootmem_pages(size); - else - ptr = alloc_bootmem_pages_node(NODE_DATA(node), size); -#endif - if (!ptr) - panic("Cannot allocate cpu data for CPU %d\n", i); - cpu_pda(i)->data_offset = ptr - __per_cpu_start; - memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start); - } - - /* setup percpu data maps early */ - setup_per_cpu_maps(); -} - void pda_init(int cpu) { struct x8664_pda *pda = cpu_pda(cpu); @@ -327,6 +252,17 @@ void __cpuinit cpu_init (void) load_TR_desc(); load_LDT(&init_mm.context); +#ifdef CONFIG_KGDB + /* + * If the kgdb is connected no debug regs should be altered. This + * is only applicable when KGDB and a KGDB I/O module are built + * into the kernel and you are using early debugging with + * kgdbwait. KGDB will control the kernel HW breakpoint registers. + */ + if (kgdb_connected && arch_kgdb_ops.correct_hw_break) + arch_kgdb_ops.correct_hw_break(); + else { +#endif /* * Clear all 6 debug registers: */ @@ -337,8 +273,15 @@ void __cpuinit cpu_init (void) set_debugreg(0UL, 3); set_debugreg(0UL, 6); set_debugreg(0UL, 7); +#ifdef CONFIG_KGDB + /* If the kgdb is connected no debug regs should be altered. */ + } +#endif fpu_init(); raw_local_save_flags(kernel_eflags); + + if (is_uv_system()) + uv_cpu_init(); } diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c index 2b3e5d45176..5b0bffb7fcc 100644 --- a/arch/x86/kernel/setup_32.c +++ b/arch/x86/kernel/setup_32.c @@ -62,8 +62,9 @@ #include <asm/io.h> #include <asm/vmi.h> #include <setup_arch.h> -#include <bios_ebda.h> +#include <asm/bios_ebda.h> #include <asm/cacheflush.h> +#include <asm/processor.h> /* This value is set up by the early boot code to point to the value immediately after the boot time page tables. It contains a *physical* @@ -154,6 +155,8 @@ struct cpuinfo_x86 new_cpu_data __cpuinitdata = { 0, 0, 0, 0, -1, 1, 0, 0, -1 }; struct cpuinfo_x86 boot_cpu_data __read_mostly = { 0, 0, 0, 0, -1, 1, 0, 0, -1 }; EXPORT_SYMBOL(boot_cpu_data); +unsigned int def_to_bigsmp; + #ifndef CONFIG_X86_PAE unsigned long mmu_cr4_features; #else @@ -189,7 +192,7 @@ EXPORT_SYMBOL(ist_info); extern void early_cpu_init(void); extern int root_mountflags; -unsigned long saved_videomode; +unsigned long saved_video_mode; #define RAMDISK_IMAGE_START_MASK 0x07FF #define RAMDISK_PROMPT_FLAG 0x8000 @@ -227,7 +230,7 @@ static inline void copy_edd(void) } #endif -int __initdata user_defined_memmap = 0; +int __initdata user_defined_memmap; /* * "mem=nopentium" disables the 4MB page tables. @@ -385,15 +388,56 @@ unsigned long __init find_max_low_pfn(void) return max_low_pfn; } +#define BIOS_EBDA_SEGMENT 0x40E +#define BIOS_LOWMEM_KILOBYTES 0x413 + /* - * workaround for Dell systems that neglect to reserve EBDA + * The BIOS places the EBDA/XBDA at the top of conventional + * memory, and usually decreases the reported amount of + * conventional memory (int 0x12) too. This also contains a + * workaround for Dell systems that neglect to reserve EBDA. + * The same workaround also avoids a problem with the AMD768MPX + * chipset: reserve a page before VGA to prevent PCI prefetch + * into it (errata #56). Usually the page is reserved anyways, + * unless you have no PS/2 mouse plugged in. */ static void __init reserve_ebda_region(void) { - unsigned int addr; - addr = get_bios_ebda(); - if (addr) - reserve_bootmem(addr, PAGE_SIZE, BOOTMEM_DEFAULT); + unsigned int lowmem, ebda_addr; + + /* To determine the position of the EBDA and the */ + /* end of conventional memory, we need to look at */ + /* the BIOS data area. In a paravirtual environment */ + /* that area is absent. We'll just have to assume */ + /* that the paravirt case can handle memory setup */ + /* correctly, without our help. */ + if (paravirt_enabled()) + return; + + /* end of low (conventional) memory */ + lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES); + lowmem <<= 10; + + /* start of EBDA area */ + ebda_addr = *(unsigned short *)__va(BIOS_EBDA_SEGMENT); + ebda_addr <<= 4; + + /* Fixup: bios puts an EBDA in the top 64K segment */ + /* of conventional memory, but does not adjust lowmem. */ + if ((lowmem - ebda_addr) <= 0x10000) + lowmem = ebda_addr; + + /* Fixup: bios does not report an EBDA at all. */ + /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */ + if ((ebda_addr == 0) && (lowmem >= 0x9f000)) + lowmem = 0x9f000; + + /* Paranoia: should never happen, but... */ + if ((lowmem == 0) || (lowmem >= 0x100000)) + lowmem = 0x9f000; + + /* reserve all memory between lowmem and the 1MB mark */ + reserve_bootmem(lowmem, 0x100000 - lowmem, BOOTMEM_DEFAULT); } #ifndef CONFIG_NEED_MULTIPLE_NODES @@ -617,16 +661,9 @@ void __init setup_bootmem_allocator(void) */ reserve_bootmem(0, PAGE_SIZE, BOOTMEM_DEFAULT); - /* reserve EBDA region, it's a 4K region */ + /* reserve EBDA region */ reserve_ebda_region(); - /* could be an AMD 768MPX chipset. Reserve a page before VGA to prevent - PCI prefetch into it (errata #56). Usually the page is reserved anyways, - unless you have no PS/2 mouse plugged in. */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && - boot_cpu_data.x86 == 6) - reserve_bootmem(0xa0000 - 4096, 4096, BOOTMEM_DEFAULT); - #ifdef CONFIG_SMP /* * But first pinch a few for the stack/trampoline stuff @@ -687,6 +724,18 @@ char * __init __attribute__((weak)) memory_setup(void) return machine_specific_memory_setup(); } +#ifdef CONFIG_NUMA +/* + * In the golden day, when everything among i386 and x86_64 will be + * integrated, this will not live here + */ +void *x86_cpu_to_node_map_early_ptr; +int x86_cpu_to_node_map_init[NR_CPUS] = { + [0 ... NR_CPUS-1] = NUMA_NO_NODE +}; +DEFINE_PER_CPU(int, x86_cpu_to_node_map) = NUMA_NO_NODE; +#endif + /* * Determine if we were loaded by an EFI loader. If so, then we have also been * passed the efi memmap, systab, etc., so we should use these data structures @@ -714,7 +763,7 @@ void __init setup_arch(char **cmdline_p) edid_info = boot_params.edid_info; apm_info.bios = boot_params.apm_bios_info; ist_info = boot_params.ist_info; - saved_videomode = boot_params.hdr.vid_mode; + saved_video_mode = boot_params.hdr.vid_mode; if( boot_params.sys_desc_table.length != 0 ) { set_mca_bus(boot_params.sys_desc_table.table[3] & 0x2); machine_id = boot_params.sys_desc_table.table[0]; @@ -820,6 +869,18 @@ void __init setup_arch(char **cmdline_p) io_delay_init(); +#ifdef CONFIG_X86_SMP + /* + * setup to use the early static init tables during kernel startup + * X86_SMP will exclude sub-arches that don't deal well with it. + */ + x86_cpu_to_apicid_early_ptr = (void *)x86_cpu_to_apicid_init; + x86_bios_cpu_apicid_early_ptr = (void *)x86_bios_cpu_apicid_init; +#ifdef CONFIG_NUMA + x86_cpu_to_node_map_early_ptr = (void *)x86_cpu_to_node_map_init; +#endif +#endif + #ifdef CONFIG_X86_GENERICARCH generic_apic_probe(); #endif diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c index f4f7ecfb898..674ef3510cd 100644 --- a/arch/x86/kernel/setup_64.c +++ b/arch/x86/kernel/setup_64.c @@ -58,7 +58,6 @@ #include <asm/mmu_context.h> #include <asm/proto.h> #include <asm/setup.h> -#include <asm/mach_apic.h> #include <asm/numa.h> #include <asm/sections.h> #include <asm/dmi.h> @@ -66,7 +65,9 @@ #include <asm/mce.h> #include <asm/ds.h> #include <asm/topology.h> +#include <asm/trampoline.h> +#include <mach_apic.h> #ifdef CONFIG_PARAVIRT #include <asm/paravirt.h> #else @@ -248,6 +249,7 @@ static void __init reserve_crashkernel(void) (unsigned long)(total_mem >> 20)); crashk_res.start = crash_base; crashk_res.end = crash_base + crash_size - 1; + insert_resource(&iomem_resource, &crashk_res); } } #else @@ -322,6 +324,11 @@ void __init setup_arch(char **cmdline_p) finish_e820_parsing(); + /* after parse_early_param, so could debug it */ + insert_resource(&iomem_resource, &code_resource); + insert_resource(&iomem_resource, &data_resource); + insert_resource(&iomem_resource, &bss_resource); + early_gart_iommu_check(); e820_register_active_regions(0, 0, -1UL); @@ -341,10 +348,12 @@ void __init setup_arch(char **cmdline_p) check_efer(); - init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT)); + max_pfn_mapped = init_memory_mapping(0, (max_pfn_mapped << PAGE_SHIFT)); if (efi_enabled) efi_init(); + vsmp_init(); + dmi_scan_machine(); io_delay_init(); @@ -450,7 +459,7 @@ void __init setup_arch(char **cmdline_p) /* * We trust e820 completely. No explicit ROM probing in memory. */ - e820_reserve_resources(&code_resource, &data_resource, &bss_resource); + e820_reserve_resources(); e820_mark_nosave_regions(); /* request I/O space for devices used on all i[345]86 PCs */ @@ -552,9 +561,9 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c) bits = c->x86_coreid_bits; /* Low order bits define the core id (index of core in socket) */ - c->cpu_core_id = c->phys_proc_id & ((1 << bits)-1); - /* Convert the APIC ID into the socket ID */ - c->phys_proc_id = phys_pkg_id(bits); + c->cpu_core_id = c->initial_apicid & ((1 << bits)-1); + /* Convert the initial APIC ID into the socket ID */ + c->phys_proc_id = c->initial_apicid >> bits; #ifdef CONFIG_NUMA node = c->phys_proc_id; @@ -571,7 +580,7 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c) If that doesn't result in a usable node fall back to the path for the previous case. */ - int ht_nodeid = apicid - (cpu_data(0).phys_proc_id << bits); + int ht_nodeid = c->initial_apicid; if (ht_nodeid >= 0 && apicid_to_node[ht_nodeid] != NUMA_NO_NODE) @@ -677,7 +686,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) /* Bit 31 in normal CPUID used for nonstandard 3DNow ID; 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */ - clear_bit(0*32+31, (unsigned long *)&c->x86_capability); + clear_cpu_cap(c, 0*32+31); /* On C+ stepping K8 rep microcode works well for copy/memset */ level = cpuid_eax(1); @@ -721,6 +730,19 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) if (amd_apic_timer_broken()) disable_apic_timer = 1; + + if (c == &boot_cpu_data && c->x86 >= 0xf && c->x86 <= 0x11) { + unsigned long long tseg; + + /* + * Split up direct mapping around the TSEG SMM area. + * Don't do it for gbpages because there seems very little + * benefit in doing so. + */ + if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg) && + (tseg >> PMD_SHIFT) < (max_pfn_mapped >> (PMD_SHIFT-PAGE_SHIFT))) + set_memory_4k((unsigned long)__va(tseg), 1); + } } void __cpuinit detect_ht(struct cpuinfo_x86 *c) @@ -813,7 +835,7 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) { if ((c->x86 == 0xf && c->x86_model >= 0x03) || (c->x86 == 0x6 && c->x86_model >= 0x0e)) - set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability); + set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); } static void __cpuinit init_intel(struct cpuinfo_x86 *c) @@ -856,9 +878,6 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) if (c->x86 == 15) c->x86_cache_alignment = c->x86_clflush_size * 2; - if ((c->x86 == 0xf && c->x86_model >= 0x03) || - (c->x86 == 0x6 && c->x86_model >= 0x0e)) - set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); if (c->x86 == 6) set_cpu_cap(c, X86_FEATURE_REP_GOOD); set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); @@ -867,6 +886,32 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) srat_detect_node(); } +static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c) +{ + if (c->x86 == 0x6 && c->x86_model >= 0xf) + set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability); +} + +static void __cpuinit init_centaur(struct cpuinfo_x86 *c) +{ + /* Cache sizes */ + unsigned n; + + n = c->extended_cpuid_level; + if (n >= 0x80000008) { + unsigned eax = cpuid_eax(0x80000008); + c->x86_virt_bits = (eax >> 8) & 0xff; + c->x86_phys_bits = eax & 0xff; + } + + if (c->x86 == 0x6 && c->x86_model >= 0xf) { + c->x86_cache_alignment = c->x86_clflush_size * 2; + set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); + set_cpu_cap(c, X86_FEATURE_REP_GOOD); + } + set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); +} + static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) { char *v = c->x86_vendor_id; @@ -875,6 +920,8 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) c->x86_vendor = X86_VENDOR_AMD; else if (!strcmp(v, "GenuineIntel")) c->x86_vendor = X86_VENDOR_INTEL; + else if (!strcmp(v, "CentaurHauls")) + c->x86_vendor = X86_VENDOR_CENTAUR; else c->x86_vendor = X86_VENDOR_UNKNOWN; } @@ -922,15 +969,16 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) c->x86 += (tfms >> 20) & 0xff; if (c->x86 >= 0x6) c->x86_model += ((tfms >> 16) & 0xF) << 4; - if (c->x86_capability[0] & (1<<19)) + if (test_cpu_cap(c, X86_FEATURE_CLFLSH)) c->x86_clflush_size = ((misc >> 8) & 0xff) * 8; } else { /* Have CPUID level 0 only - unheard of */ c->x86 = 4; } + c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xff; #ifdef CONFIG_SMP - c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff; + c->phys_proc_id = c->initial_apicid; #endif /* AMD-defined flags: level 0x80000001 */ xlvl = cpuid_eax(0x80000000); @@ -956,12 +1004,22 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) if (c->extended_cpuid_level >= 0x80000007) c->x86_power = cpuid_edx(0x80000007); + + clear_cpu_cap(c, X86_FEATURE_PAT); + switch (c->x86_vendor) { case X86_VENDOR_AMD: early_init_amd(c); + if (c->x86 >= 0xf && c->x86 <= 0x11) + set_cpu_cap(c, X86_FEATURE_PAT); break; case X86_VENDOR_INTEL: early_init_intel(c); + if (c->x86 == 0xF || (c->x86 == 6 && c->x86_model >= 15)) + set_cpu_cap(c, X86_FEATURE_PAT); + break; + case X86_VENDOR_CENTAUR: + early_init_centaur(c); break; } @@ -999,6 +1057,10 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c) init_intel(c); break; + case X86_VENDOR_CENTAUR: + init_centaur(c); + break; + case X86_VENDOR_UNKNOWN: default: display_cacheinfo(c); @@ -1028,14 +1090,24 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c) #endif select_idle_routine(c); - if (c != &boot_cpu_data) - mtrr_ap_init(); #ifdef CONFIG_NUMA numa_add_cpu(smp_processor_id()); #endif } +void __cpuinit identify_boot_cpu(void) +{ + identify_cpu(&boot_cpu_data); +} + +void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c) +{ + BUG_ON(c == &boot_cpu_data); + identify_cpu(c); + mtrr_ap_init(); +} + static __init int setup_noclflush(char *arg) { setup_clear_cpu_cap(X86_FEATURE_CLFLSH); @@ -1064,123 +1136,3 @@ static __init int setup_disablecpuid(char *arg) return 1; } __setup("clearcpuid=", setup_disablecpuid); - -/* - * Get CPU information for use by the procfs. - */ - -static int show_cpuinfo(struct seq_file *m, void *v) -{ - struct cpuinfo_x86 *c = v; - int cpu = 0, i; - -#ifdef CONFIG_SMP - cpu = c->cpu_index; -#endif - - seq_printf(m, "processor\t: %u\n" - "vendor_id\t: %s\n" - "cpu family\t: %d\n" - "model\t\t: %d\n" - "model name\t: %s\n", - (unsigned)cpu, - c->x86_vendor_id[0] ? c->x86_vendor_id : "unknown", - c->x86, - (int)c->x86_model, - c->x86_model_id[0] ? c->x86_model_id : "unknown"); - - if (c->x86_mask || c->cpuid_level >= 0) - seq_printf(m, "stepping\t: %d\n", c->x86_mask); - else - seq_printf(m, "stepping\t: unknown\n"); - - if (cpu_has(c, X86_FEATURE_TSC)) { - unsigned int freq = cpufreq_quick_get((unsigned)cpu); - - if (!freq) - freq = cpu_khz; - seq_printf(m, "cpu MHz\t\t: %u.%03u\n", - freq / 1000, (freq % 1000)); - } - - /* Cache size */ - if (c->x86_cache_size >= 0) - seq_printf(m, "cache size\t: %d KB\n", c->x86_cache_size); - -#ifdef CONFIG_SMP - if (smp_num_siblings * c->x86_max_cores > 1) { - seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); - seq_printf(m, "siblings\t: %d\n", - cpus_weight(per_cpu(cpu_core_map, cpu))); - seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id); - seq_printf(m, "cpu cores\t: %d\n", c->booted_cores); - } -#endif - - seq_printf(m, - "fpu\t\t: yes\n" - "fpu_exception\t: yes\n" - "cpuid level\t: %d\n" - "wp\t\t: yes\n" - "flags\t\t:", - c->cpuid_level); - - for (i = 0; i < 32*NCAPINTS; i++) - if (cpu_has(c, i) && x86_cap_flags[i] != NULL) - seq_printf(m, " %s", x86_cap_flags[i]); - - seq_printf(m, "\nbogomips\t: %lu.%02lu\n", - c->loops_per_jiffy/(500000/HZ), - (c->loops_per_jiffy/(5000/HZ)) % 100); - - if (c->x86_tlbsize > 0) - seq_printf(m, "TLB size\t: %d 4K pages\n", c->x86_tlbsize); - seq_printf(m, "clflush size\t: %d\n", c->x86_clflush_size); - seq_printf(m, "cache_alignment\t: %d\n", c->x86_cache_alignment); - - seq_printf(m, "address sizes\t: %u bits physical, %u bits virtual\n", - c->x86_phys_bits, c->x86_virt_bits); - - seq_printf(m, "power management:"); - for (i = 0; i < 32; i++) { - if (c->x86_power & (1 << i)) { - if (i < ARRAY_SIZE(x86_power_flags) && - x86_power_flags[i]) - seq_printf(m, "%s%s", - x86_power_flags[i][0]?" ":"", - x86_power_flags[i]); - else - seq_printf(m, " [%d]", i); - } - } - - seq_printf(m, "\n\n"); - - return 0; -} - -static void *c_start(struct seq_file *m, loff_t *pos) -{ - if (*pos == 0) /* just in case, cpu 0 is not the first */ - *pos = first_cpu(cpu_online_map); - if ((*pos) < NR_CPUS && cpu_online(*pos)) - return &cpu_data(*pos); - return NULL; -} - -static void *c_next(struct seq_file *m, void *v, loff_t *pos) -{ - *pos = next_cpu(*pos, cpu_online_map); - return c_start(m, pos); -} - -static void c_stop(struct seq_file *m, void *v) -{ -} - -const struct seq_operations cpuinfo_op = { - .start = c_start, - .next = c_next, - .stop = c_stop, - .show = show_cpuinfo, -}; diff --git a/arch/x86/kernel/sigframe_32.h b/arch/x86/kernel/sigframe.h index 0b2221711da..72bbb519d2d 100644 --- a/arch/x86/kernel/sigframe_32.h +++ b/arch/x86/kernel/sigframe.h @@ -1,5 +1,5 @@ -struct sigframe -{ +#ifdef CONFIG_X86_32 +struct sigframe { char __user *pretcode; int sig; struct sigcontext sc; @@ -8,8 +8,7 @@ struct sigframe char retcode[8]; }; -struct rt_sigframe -{ +struct rt_sigframe { char __user *pretcode; int sig; struct siginfo __user *pinfo; @@ -19,3 +18,10 @@ struct rt_sigframe struct _fpstate fpstate; char retcode[8]; }; +#else +struct rt_sigframe { + char __user *pretcode; + struct ucontext uc; + struct siginfo info; +}; +#endif diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c index 0157a6f0f41..f1b11793083 100644 --- a/arch/x86/kernel/signal_32.c +++ b/arch/x86/kernel/signal_32.c @@ -4,32 +4,44 @@ * 1997-11-28 Modified for POSIX.1b signals by Richard Henderson * 2000-06-20 Pentium III FXSR, SSE support by Gareth Hughes */ +#include <linux/list.h> -#include <linux/sched.h> -#include <linux/mm.h> -#include <linux/smp.h> +#include <linux/personality.h> +#include <linux/binfmts.h> +#include <linux/suspend.h> #include <linux/kernel.h> +#include <linux/ptrace.h> #include <linux/signal.h> +#include <linux/stddef.h> +#include <linux/unistd.h> #include <linux/errno.h> +#include <linux/sched.h> #include <linux/wait.h> -#include <linux/unistd.h> -#include <linux/stddef.h> -#include <linux/personality.h> -#include <linux/suspend.h> -#include <linux/ptrace.h> #include <linux/elf.h> -#include <linux/binfmts.h> +#include <linux/smp.h> +#include <linux/mm.h> + #include <asm/processor.h> #include <asm/ucontext.h> #include <asm/uaccess.h> #include <asm/i387.h> #include <asm/vdso.h> -#include "sigframe_32.h" -#define DEBUG_SIG 0 +#include "sigframe.h" #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) +#define __FIX_EFLAGS (X86_EFLAGS_AC | X86_EFLAGS_OF | \ + X86_EFLAGS_DF | X86_EFLAGS_TF | X86_EFLAGS_SF | \ + X86_EFLAGS_ZF | X86_EFLAGS_AF | X86_EFLAGS_PF | \ + X86_EFLAGS_CF) + +#ifdef CONFIG_X86_32 +# define FIX_EFLAGS (__FIX_EFLAGS | X86_EFLAGS_RF) +#else +# define FIX_EFLAGS __FIX_EFLAGS +#endif + /* * Atomically swap in the new signal mask, and wait for a signal. */ @@ -46,10 +58,11 @@ sys_sigsuspend(int history0, int history1, old_sigset_t mask) current->state = TASK_INTERRUPTIBLE; schedule(); set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; } -asmlinkage int +asmlinkage int sys_sigaction(int sig, const struct old_sigaction __user *act, struct old_sigaction __user *oact) { @@ -58,10 +71,12 @@ sys_sigaction(int sig, const struct old_sigaction __user *act, if (act) { old_sigset_t mask; + if (!access_ok(VERIFY_READ, act, sizeof(*act)) || __get_user(new_ka.sa.sa_handler, &act->sa_handler) || __get_user(new_ka.sa.sa_restorer, &act->sa_restorer)) return -EFAULT; + __get_user(new_ka.sa.sa_flags, &act->sa_flags); __get_user(mask, &act->sa_mask); siginitset(&new_ka.sa.sa_mask, mask); @@ -74,6 +89,7 @@ sys_sigaction(int sig, const struct old_sigaction __user *act, __put_user(old_ka.sa.sa_handler, &oact->sa_handler) || __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer)) return -EFAULT; + __put_user(old_ka.sa.sa_flags, &oact->sa_flags); __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); } @@ -81,10 +97,12 @@ sys_sigaction(int sig, const struct old_sigaction __user *act, return ret; } -asmlinkage int -sys_sigaltstack(unsigned long bx) +asmlinkage int sys_sigaltstack(unsigned long bx) { - /* This is needed to make gcc realize it doesn't own the "struct pt_regs" */ + /* + * This is needed to make gcc realize it doesn't own the + * "struct pt_regs" + */ struct pt_regs *regs = (struct pt_regs *)&bx; const stack_t __user *uss = (const stack_t __user *)bx; stack_t __user *uoss = (stack_t __user *)regs->cx; @@ -96,9 +114,9 @@ sys_sigaltstack(unsigned long bx) /* * Do a signal return; undo the signal stack. */ - static int -restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax) +restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, + unsigned long *pax) { unsigned int err = 0; @@ -120,37 +138,29 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax #define GET_SEG(seg) \ { unsigned short tmp; \ err |= __get_user(tmp, &sc->seg); \ - loadsegment(seg,tmp); } - -#define FIX_EFLAGS (X86_EFLAGS_AC | X86_EFLAGS_RF | \ - X86_EFLAGS_OF | X86_EFLAGS_DF | \ - X86_EFLAGS_TF | X86_EFLAGS_SF | X86_EFLAGS_ZF | \ - X86_EFLAGS_AF | X86_EFLAGS_PF | X86_EFLAGS_CF) + loadsegment(seg, tmp); } GET_SEG(gs); COPY_SEG(fs); COPY_SEG(es); COPY_SEG(ds); - COPY(di); - COPY(si); - COPY(bp); - COPY(sp); - COPY(bx); - COPY(dx); - COPY(cx); - COPY(ip); + COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx); + COPY(dx); COPY(cx); COPY(ip); COPY_SEG_STRICT(cs); COPY_SEG_STRICT(ss); - + { unsigned int tmpflags; + err |= __get_user(tmpflags, &sc->flags); - regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS); + regs->flags = (regs->flags & ~FIX_EFLAGS) | + (tmpflags & FIX_EFLAGS); regs->orig_ax = -1; /* disable syscall checks */ } { - struct _fpstate __user * buf; + struct _fpstate __user *buf; + err |= __get_user(buf, &sc->fpstate); if (buf) { if (!access_ok(VERIFY_READ, buf, sizeof(*buf))) @@ -158,6 +168,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax err |= restore_i387(buf); } else { struct task_struct *me = current; + if (used_math()) { clear_fpu(me); clear_used_math(); @@ -165,24 +176,26 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax } } - err |= __get_user(*peax, &sc->ax); + err |= __get_user(*pax, &sc->ax); return err; badframe: return 1; } -asmlinkage int sys_sigreturn(unsigned long __unused) +asmlinkage unsigned long sys_sigreturn(unsigned long __unused) { - struct pt_regs *regs = (struct pt_regs *) &__unused; - struct sigframe __user *frame = (struct sigframe __user *)(regs->sp - 8); + struct sigframe __user *frame; + struct pt_regs *regs; + unsigned long ax; sigset_t set; - int ax; + + regs = (struct pt_regs *) &__unused; + frame = (struct sigframe __user *)(regs->sp - 8); if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) goto badframe; - if (__get_user(set.sig[0], &frame->sc.oldmask) - || (_NSIG_WORDS > 1 + if (__get_user(set.sig[0], &frame->sc.oldmask) || (_NSIG_WORDS > 1 && __copy_from_user(&set.sig[1], &frame->extramask, sizeof(frame->extramask)))) goto badframe; @@ -192,33 +205,35 @@ asmlinkage int sys_sigreturn(unsigned long __unused) current->blocked = set; recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - + if (restore_sigcontext(regs, &frame->sc, &ax)) goto badframe; return ax; badframe: if (show_unhandled_signals && printk_ratelimit()) { - printk("%s%s[%d] bad frame in sigreturn frame:%p ip:%lx" - " sp:%lx oeax:%lx", + printk(KERN_INFO "%s%s[%d] bad frame in sigreturn frame:" + "%p ip:%lx sp:%lx oeax:%lx", task_pid_nr(current) > 1 ? KERN_INFO : KERN_EMERG, current->comm, task_pid_nr(current), frame, regs->ip, regs->sp, regs->orig_ax); print_vma_addr(" in ", regs->ip); - printk("\n"); + printk(KERN_CONT "\n"); } force_sig(SIGSEGV, current); + return 0; -} +} asmlinkage int sys_rt_sigreturn(unsigned long __unused) { - struct pt_regs *regs = (struct pt_regs *) &__unused; - struct rt_sigframe __user *frame = (struct rt_sigframe __user *)(regs->sp - 4); + struct pt_regs *regs = (struct pt_regs *)&__unused; + struct rt_sigframe __user *frame; + unsigned long ax; sigset_t set; - int ax; + frame = (struct rt_sigframe __user *)(regs->sp - sizeof(long)); if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) goto badframe; if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) @@ -229,7 +244,7 @@ asmlinkage int sys_rt_sigreturn(unsigned long __unused) current->blocked = set; recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - + if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax)) goto badframe; @@ -241,12 +256,11 @@ asmlinkage int sys_rt_sigreturn(unsigned long __unused) badframe: force_sig(SIGSEGV, current); return 0; -} +} /* * Set up a signal frame. */ - static int setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate, struct pt_regs *regs, unsigned long mask) @@ -277,9 +291,9 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate, tmp = save_i387(fpstate); if (tmp < 0) - err = 1; + err = 1; else - err |= __put_user(tmp ? fpstate : NULL, &sc->fpstate); + err |= __put_user(tmp ? fpstate : NULL, &sc->fpstate); /* non-iBCS2 extensions.. */ err |= __put_user(mask, &sc->oldmask); @@ -292,7 +306,7 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate, * Determine which stack to use.. */ static inline void __user * -get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size) +get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size) { unsigned long sp; @@ -310,32 +324,30 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size) if (ka->sa.sa_flags & SA_ONSTACK) { if (sas_ss_flags(sp) == 0) sp = current->sas_ss_sp + current->sas_ss_size; - } - - /* This is the legacy signal stack switching. */ - else if ((regs->ss & 0xffff) != __USER_DS && - !(ka->sa.sa_flags & SA_RESTORER) && - ka->sa.sa_restorer) { - sp = (unsigned long) ka->sa.sa_restorer; + } else { + /* This is the legacy signal stack switching. */ + if ((regs->ss & 0xffff) != __USER_DS && + !(ka->sa.sa_flags & SA_RESTORER) && + ka->sa.sa_restorer) + sp = (unsigned long) ka->sa.sa_restorer; } sp -= frame_size; - /* Align the stack pointer according to the i386 ABI, - * i.e. so that on function entry ((sp + 4) & 15) == 0. */ + /* + * Align the stack pointer according to the i386 ABI, + * i.e. so that on function entry ((sp + 4) & 15) == 0. + */ sp = ((sp + 4) & -16ul) - 4; + return (void __user *) sp; } -/* These symbols are defined with the addresses in the vsyscall page. - See vsyscall-sigreturn.S. */ -extern void __user __kernel_sigreturn; -extern void __user __kernel_rt_sigreturn; - -static int setup_frame(int sig, struct k_sigaction *ka, - sigset_t *set, struct pt_regs * regs) +static int +setup_frame(int sig, struct k_sigaction *ka, sigset_t *set, + struct pt_regs *regs) { - void __user *restorer; struct sigframe __user *frame; + void __user *restorer; int err = 0; int usig; @@ -365,7 +377,7 @@ static int setup_frame(int sig, struct k_sigaction *ka, goto give_sigsegv; } - if (current->binfmt->hasvdso) + if (current->mm->context.vdso) restorer = VDSO32_SYMBOL(current->mm->context.vdso, sigreturn); else restorer = &frame->retcode; @@ -374,9 +386,9 @@ static int setup_frame(int sig, struct k_sigaction *ka, /* Set up to return from userspace. */ err |= __put_user(restorer, &frame->pretcode); - + /* - * This is popl %eax ; movl $,%eax ; int $0x80 + * This is popl %eax ; movl $__NR_sigreturn, %eax ; int $0x80 * * WE DO NOT USE IT ANY MORE! It's only left here for historical * reasons and because gdb uses it as a signature to notice @@ -390,11 +402,11 @@ static int setup_frame(int sig, struct k_sigaction *ka, goto give_sigsegv; /* Set up registers for signal handler */ - regs->sp = (unsigned long) frame; - regs->ip = (unsigned long) ka->sa.sa_handler; - regs->ax = (unsigned long) sig; - regs->dx = (unsigned long) 0; - regs->cx = (unsigned long) 0; + regs->sp = (unsigned long)frame; + regs->ip = (unsigned long)ka->sa.sa_handler; + regs->ax = (unsigned long)sig; + regs->dx = 0; + regs->cx = 0; regs->ds = __USER_DS; regs->es = __USER_DS; @@ -407,15 +419,10 @@ static int setup_frame(int sig, struct k_sigaction *ka, * The tracer may want to single-step inside the * handler too. */ - regs->flags &= ~(TF_MASK | X86_EFLAGS_DF); + regs->flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_DF); if (test_thread_flag(TIF_SINGLESTEP)) ptrace_notify(SIGTRAP); -#if DEBUG_SIG - printk("SIG deliver (%s:%d): sp=%p pc=%p ra=%p\n", - current->comm, current->pid, frame, regs->ip, frame->pretcode); -#endif - return 0; give_sigsegv: @@ -424,10 +431,10 @@ give_sigsegv: } static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, - sigset_t *set, struct pt_regs * regs) + sigset_t *set, struct pt_regs *regs) { - void __user *restorer; struct rt_sigframe __user *frame; + void __user *restorer; int err = 0; int usig; @@ -457,7 +464,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, &frame->uc.uc_stack.ss_flags); err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size); err |= setup_sigcontext(&frame->uc.uc_mcontext, &frame->fpstate, - regs, set->sig[0]); + regs, set->sig[0]); err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); if (err) goto give_sigsegv; @@ -467,9 +474,9 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, if (ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; err |= __put_user(restorer, &frame->pretcode); - + /* - * This is movl $,%ax ; int $0x80 + * This is movl $__NR_rt_sigreturn, %ax ; int $0x80 * * WE DO NOT USE IT ANY MORE! It's only left here for historical * reasons and because gdb uses it as a signature to notice @@ -483,11 +490,11 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, goto give_sigsegv; /* Set up registers for signal handler */ - regs->sp = (unsigned long) frame; - regs->ip = (unsigned long) ka->sa.sa_handler; - regs->ax = (unsigned long) usig; - regs->dx = (unsigned long) &frame->info; - regs->cx = (unsigned long) &frame->uc; + regs->sp = (unsigned long)frame; + regs->ip = (unsigned long)ka->sa.sa_handler; + regs->ax = (unsigned long)usig; + regs->dx = (unsigned long)&frame->info; + regs->cx = (unsigned long)&frame->uc; regs->ds = __USER_DS; regs->es = __USER_DS; @@ -500,15 +507,10 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, * The tracer may want to single-step inside the * handler too. */ - regs->flags &= ~(TF_MASK | X86_EFLAGS_DF); + regs->flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_DF); if (test_thread_flag(TIF_SINGLESTEP)) ptrace_notify(SIGTRAP); -#if DEBUG_SIG - printk("SIG deliver (%s:%d): sp=%p pc=%p ra=%p\n", - current->comm, current->pid, frame, regs->ip, frame->pretcode); -#endif - return 0; give_sigsegv: @@ -517,33 +519,33 @@ give_sigsegv: } /* - * OK, we're invoking a handler - */ - + * OK, we're invoking a handler: + */ static int handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, - sigset_t *oldset, struct pt_regs * regs) + sigset_t *oldset, struct pt_regs *regs) { int ret; /* Are we from a system call? */ - if (regs->orig_ax >= 0) { + if ((long)regs->orig_ax >= 0) { /* If so, check system call restarting.. */ switch (regs->ax) { - case -ERESTART_RESTARTBLOCK: - case -ERESTARTNOHAND: + case -ERESTART_RESTARTBLOCK: + case -ERESTARTNOHAND: + regs->ax = -EINTR; + break; + + case -ERESTARTSYS: + if (!(ka->sa.sa_flags & SA_RESTART)) { regs->ax = -EINTR; break; - - case -ERESTARTSYS: - if (!(ka->sa.sa_flags & SA_RESTART)) { - regs->ax = -EINTR; - break; - } - /* fallthrough */ - case -ERESTARTNOINTR: - regs->ax = regs->orig_ax; - regs->ip -= 2; + } + /* fallthrough */ + case -ERESTARTNOINTR: + regs->ax = regs->orig_ax; + regs->ip -= 2; + break; } } @@ -561,16 +563,17 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, else ret = setup_frame(sig, ka, oldset, regs); - if (ret == 0) { - spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked,¤t->blocked,&ka->sa.sa_mask); - if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked,sig); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - } + if (ret) + return ret; - return ret; + spin_lock_irq(¤t->sighand->siglock); + sigorsets(¤t->blocked, ¤t->blocked, &ka->sa.sa_mask); + if (!(ka->sa.sa_flags & SA_NODEFER)) + sigaddset(¤t->blocked, sig); + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + return 0; } /* @@ -580,18 +583,17 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, */ static void do_signal(struct pt_regs *regs) { + struct k_sigaction ka; siginfo_t info; int signr; - struct k_sigaction ka; sigset_t *oldset; /* - * We want the common case to go fast, which - * is why we may in certain cases get here from - * kernel mode. Just return without doing anything - * if so. vm86 regs switched out by assembly code - * before reaching here, so testing against kernel - * CS suffices. + * We want the common case to go fast, which is why we may in certain + * cases get here from kernel mode. Just return without doing anything + * if so. + * X86_32: vm86 regs switched out by assembly code before reaching + * here, so testing against kernel CS suffices. */ if (!user_mode(regs)) return; @@ -603,29 +605,31 @@ static void do_signal(struct pt_regs *regs) signr = get_signal_to_deliver(&info, &ka, regs, NULL); if (signr > 0) { - /* Re-enable any watchpoints before delivering the + /* + * Re-enable any watchpoints before delivering the * signal to user space. The processor register will * have been cleared if the watchpoint triggered * inside the kernel. */ - if (unlikely(current->thread.debugreg7)) + if (current->thread.debugreg7) set_debugreg(current->thread.debugreg7, 7); - /* Whee! Actually deliver the signal. */ + /* Whee! Actually deliver the signal. */ if (handle_signal(signr, &info, &ka, oldset, regs) == 0) { - /* a signal was successfully delivered; the saved + /* + * a signal was successfully delivered; the saved * sigmask will have been stored in the signal frame, * and will be restored by sigreturn, so we can simply - * clear the TIF_RESTORE_SIGMASK flag */ + * clear the TIF_RESTORE_SIGMASK flag + */ if (test_thread_flag(TIF_RESTORE_SIGMASK)) clear_thread_flag(TIF_RESTORE_SIGMASK); } - return; } /* Did we come from a system call? */ - if (regs->orig_ax >= 0) { + if ((long)regs->orig_ax >= 0) { /* Restart the system call - no handlers present */ switch (regs->ax) { case -ERESTARTNOHAND: @@ -642,8 +646,10 @@ static void do_signal(struct pt_regs *regs) } } - /* if there's no signal to deliver, we just put the saved sigmask - * back */ + /* + * If there's no signal to deliver, we just put the saved sigmask + * back. + */ if (test_thread_flag(TIF_RESTORE_SIGMASK)) { clear_thread_flag(TIF_RESTORE_SIGMASK); sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); @@ -654,13 +660,12 @@ static void do_signal(struct pt_regs *regs) * notification of userspace execution resumption * - triggered by the TIF_WORK_MASK flags */ -__attribute__((regparm(3))) -void do_notify_resume(struct pt_regs *regs, void *_unused, - __u32 thread_info_flags) +void +do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) { /* Pending single-step? */ if (thread_info_flags & _TIF_SINGLESTEP) { - regs->flags |= TF_MASK; + regs->flags |= X86_EFLAGS_TF; clear_thread_flag(TIF_SINGLESTEP); } @@ -670,6 +675,6 @@ void do_notify_resume(struct pt_regs *regs, void *_unused, if (thread_info_flags & _TIF_HRTICK_RESCHED) hrtick_resched(); - + clear_thread_flag(TIF_IRET); } diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c index 1c83e5124c6..827179c5b32 100644 --- a/arch/x86/kernel/signal_64.c +++ b/arch/x86/kernel/signal_64.c @@ -19,17 +19,28 @@ #include <linux/stddef.h> #include <linux/personality.h> #include <linux/compiler.h> +#include <asm/processor.h> #include <asm/ucontext.h> #include <asm/uaccess.h> #include <asm/i387.h> #include <asm/proto.h> #include <asm/ia32_unistd.h> #include <asm/mce.h> - -/* #define DEBUG_SIG 1 */ +#include "sigframe.h" #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) +#define __FIX_EFLAGS (X86_EFLAGS_AC | X86_EFLAGS_OF | \ + X86_EFLAGS_DF | X86_EFLAGS_TF | X86_EFLAGS_SF | \ + X86_EFLAGS_ZF | X86_EFLAGS_AF | X86_EFLAGS_PF | \ + X86_EFLAGS_CF) + +#ifdef CONFIG_X86_32 +# define FIX_EFLAGS (__FIX_EFLAGS | X86_EFLAGS_RF) +#else +# define FIX_EFLAGS __FIX_EFLAGS +#endif + int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, struct pt_regs * regs); int ia32_setup_frame(int sig, struct k_sigaction *ka, @@ -46,16 +57,9 @@ sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, /* * Do a signal return; undo the signal stack. */ - -struct rt_sigframe -{ - char __user *pretcode; - struct ucontext uc; - struct siginfo info; -}; - static int -restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, unsigned long *prax) +restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, + unsigned long *pax) { unsigned int err = 0; @@ -87,7 +91,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, unsigned { unsigned int tmpflags; err |= __get_user(tmpflags, &sc->flags); - regs->flags = (regs->flags & ~0x40DD5) | (tmpflags & 0x40DD5); + regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS); regs->orig_ax = -1; /* disable syscall checks */ } @@ -108,7 +112,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, unsigned } } - err |= __get_user(*prax, &sc->ax); + err |= __get_user(*pax, &sc->ax); return err; badframe: @@ -121,13 +125,11 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) sigset_t set; unsigned long ax; - frame = (struct rt_sigframe __user *)(regs->sp - 8); - if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) { + frame = (struct rt_sigframe __user *)(regs->sp - sizeof(long)); + if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) goto badframe; - } - if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) { + if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) goto badframe; - } sigdelsetmask(&set, ~_BLOCKABLE); spin_lock_irq(¤t->sighand->siglock); @@ -138,10 +140,6 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax)) goto badframe; -#ifdef DEBUG_SIG - printk("%d sigreturn ip:%lx sp:%lx frame:%p ax:%lx\n",current->pid,regs->ip,regs->sp,frame,ax); -#endif - if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->sp) == -EFAULT) goto badframe; @@ -270,10 +268,6 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, if (err) goto give_sigsegv; -#ifdef DEBUG_SIG - printk("%d old ip %lx old sp %lx old ax %lx\n", current->pid,regs->ip,regs->sp,regs->ax); -#endif - /* Set up registers for signal handler */ regs->di = sig; /* In case the signal handler was declared without prototypes */ @@ -298,10 +292,6 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, regs->flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_DF); if (test_thread_flag(TIF_SINGLESTEP)) ptrace_notify(SIGTRAP); -#ifdef DEBUG_SIG - printk("SIG deliver (%s:%d): sp=%p pc=%lx ra=%p\n", - current->comm, current->pid, frame, regs->ip, frame->pretcode); -#endif return 0; @@ -345,35 +335,29 @@ static long current_syscall_ret(struct pt_regs *regs) static int handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, - sigset_t *oldset, struct pt_regs *regs) + sigset_t *oldset, struct pt_regs *regs) { int ret; -#ifdef DEBUG_SIG - printk("handle_signal pid:%d sig:%lu ip:%lx sp:%lx regs=%p\n", - current->pid, sig, - regs->ip, regs->sp, regs); -#endif - /* Are we from a system call? */ if (current_syscall(regs) >= 0) { /* If so, check system call restarting.. */ switch (current_syscall_ret(regs)) { - case -ERESTART_RESTARTBLOCK: - case -ERESTARTNOHAND: - regs->ax = -EINTR; - break; + case -ERESTART_RESTARTBLOCK: + case -ERESTARTNOHAND: + regs->ax = -EINTR; + break; - case -ERESTARTSYS: - if (!(ka->sa.sa_flags & SA_RESTART)) { - regs->ax = -EINTR; - break; - } - /* fallthrough */ - case -ERESTARTNOINTR: - regs->ax = regs->orig_ax; - regs->ip -= 2; + case -ERESTARTSYS: + if (!(ka->sa.sa_flags & SA_RESTART)) { + regs->ax = -EINTR; break; + } + /* fallthrough */ + case -ERESTARTNOINTR: + regs->ax = regs->orig_ax; + regs->ip -= 2; + break; } } @@ -420,10 +404,11 @@ static void do_signal(struct pt_regs *regs) sigset_t *oldset; /* - * We want the common case to go fast, which - * is why we may in certain cases get here from - * kernel mode. Just return without doing anything + * We want the common case to go fast, which is why we may in certain + * cases get here from kernel mode. Just return without doing anything * if so. + * X86_32: vm86 regs switched out by assembly code before reaching + * here, so testing against kernel CS suffices. */ if (!user_mode(regs)) return; @@ -473,22 +458,19 @@ static void do_signal(struct pt_regs *regs) } } - /* if there's no signal to deliver, we just put the saved sigmask - back. */ + /* + * If there's no signal to deliver, we just put the saved sigmask + * back. + */ if (test_thread_flag(TIF_RESTORE_SIGMASK)) { clear_thread_flag(TIF_RESTORE_SIGMASK); sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); } } -void -do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) +void do_notify_resume(struct pt_regs *regs, void *unused, + __u32 thread_info_flags) { -#ifdef DEBUG_SIG - printk("do_notify_resume flags:%x ip:%lx sp:%lx caller:%p pending:%x\n", - thread_info_flags, regs->ip, regs->sp, __builtin_return_address(0),signal_pending(current)); -#endif - /* Pending single-step? */ if (thread_info_flags & _TIF_SINGLESTEP) { regs->flags |= X86_EFLAGS_TF; @@ -502,7 +484,7 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) #endif /* CONFIG_X86_MCE */ /* deal with pending signal delivery */ - if (thread_info_flags & (_TIF_SIGPENDING|_TIF_RESTORE_SIGMASK)) + if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK)) do_signal(regs); if (thread_info_flags & _TIF_HRTICK_RESCHED) diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c new file mode 100644 index 00000000000..8f75893a646 --- /dev/null +++ b/arch/x86/kernel/smp.c @@ -0,0 +1,343 @@ +/* + * Intel SMP support routines. + * + * (c) 1995 Alan Cox, Building #3 <alan@redhat.com> + * (c) 1998-99, 2000 Ingo Molnar <mingo@redhat.com> + * (c) 2002,2003 Andi Kleen, SuSE Labs. + * + * i386 and x86_64 integration by Glauber Costa <gcosta@redhat.com> + * + * This code is released under the GNU General Public License version 2 or + * later. + */ + +#include <linux/init.h> + +#include <linux/mm.h> +#include <linux/delay.h> +#include <linux/spinlock.h> +#include <linux/kernel_stat.h> +#include <linux/mc146818rtc.h> +#include <linux/cache.h> +#include <linux/interrupt.h> +#include <linux/cpu.h> + +#include <asm/mtrr.h> +#include <asm/tlbflush.h> +#include <asm/mmu_context.h> +#include <asm/proto.h> +#include <mach_ipi.h> +#include <mach_apic.h> +/* + * Some notes on x86 processor bugs affecting SMP operation: + * + * Pentium, Pentium Pro, II, III (and all CPUs) have bugs. + * The Linux implications for SMP are handled as follows: + * + * Pentium III / [Xeon] + * None of the E1AP-E3AP errata are visible to the user. + * + * E1AP. see PII A1AP + * E2AP. see PII A2AP + * E3AP. see PII A3AP + * + * Pentium II / [Xeon] + * None of the A1AP-A3AP errata are visible to the user. + * + * A1AP. see PPro 1AP + * A2AP. see PPro 2AP + * A3AP. see PPro 7AP + * + * Pentium Pro + * None of 1AP-9AP errata are visible to the normal user, + * except occasional delivery of 'spurious interrupt' as trap #15. + * This is very rare and a non-problem. + * + * 1AP. Linux maps APIC as non-cacheable + * 2AP. worked around in hardware + * 3AP. fixed in C0 and above steppings microcode update. + * Linux does not use excessive STARTUP_IPIs. + * 4AP. worked around in hardware + * 5AP. symmetric IO mode (normal Linux operation) not affected. + * 'noapic' mode has vector 0xf filled out properly. + * 6AP. 'noapic' mode might be affected - fixed in later steppings + * 7AP. We do not assume writes to the LVT deassering IRQs + * 8AP. We do not enable low power mode (deep sleep) during MP bootup + * 9AP. We do not use mixed mode + * + * Pentium + * There is a marginal case where REP MOVS on 100MHz SMP + * machines with B stepping processors can fail. XXX should provide + * an L1cache=Writethrough or L1cache=off option. + * + * B stepping CPUs may hang. There are hardware work arounds + * for this. We warn about it in case your board doesn't have the work + * arounds. Basically that's so I can tell anyone with a B stepping + * CPU and SMP problems "tough". + * + * Specific items [From Pentium Processor Specification Update] + * + * 1AP. Linux doesn't use remote read + * 2AP. Linux doesn't trust APIC errors + * 3AP. We work around this + * 4AP. Linux never generated 3 interrupts of the same priority + * to cause a lost local interrupt. + * 5AP. Remote read is never used + * 6AP. not affected - worked around in hardware + * 7AP. not affected - worked around in hardware + * 8AP. worked around in hardware - we get explicit CS errors if not + * 9AP. only 'noapic' mode affected. Might generate spurious + * interrupts, we log only the first one and count the + * rest silently. + * 10AP. not affected - worked around in hardware + * 11AP. Linux reads the APIC between writes to avoid this, as per + * the documentation. Make sure you preserve this as it affects + * the C stepping chips too. + * 12AP. not affected - worked around in hardware + * 13AP. not affected - worked around in hardware + * 14AP. we always deassert INIT during bootup + * 15AP. not affected - worked around in hardware + * 16AP. not affected - worked around in hardware + * 17AP. not affected - worked around in hardware + * 18AP. not affected - worked around in hardware + * 19AP. not affected - worked around in BIOS + * + * If this sounds worrying believe me these bugs are either ___RARE___, + * or are signal timing bugs worked around in hardware and there's + * about nothing of note with C stepping upwards. + */ + +/* + * this function sends a 'reschedule' IPI to another CPU. + * it goes straight through and wastes no time serializing + * anything. Worst case is that we lose a reschedule ... + */ +static void native_smp_send_reschedule(int cpu) +{ + if (unlikely(cpu_is_offline(cpu))) { + WARN_ON(1); + return; + } + send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR); +} + +/* + * Structure and data for smp_call_function(). This is designed to minimise + * static memory requirements. It also looks cleaner. + */ +static DEFINE_SPINLOCK(call_lock); + +struct call_data_struct { + void (*func) (void *info); + void *info; + atomic_t started; + atomic_t finished; + int wait; +}; + +void lock_ipi_call_lock(void) +{ + spin_lock_irq(&call_lock); +} + +void unlock_ipi_call_lock(void) +{ + spin_unlock_irq(&call_lock); +} + +static struct call_data_struct *call_data; + +static void __smp_call_function(void (*func) (void *info), void *info, + int nonatomic, int wait) +{ + struct call_data_struct data; + int cpus = num_online_cpus() - 1; + + if (!cpus) + return; + + data.func = func; + data.info = info; + atomic_set(&data.started, 0); + data.wait = wait; + if (wait) + atomic_set(&data.finished, 0); + + call_data = &data; + mb(); + + /* Send a message to all other CPUs and wait for them to respond */ + send_IPI_allbutself(CALL_FUNCTION_VECTOR); + + /* Wait for response */ + while (atomic_read(&data.started) != cpus) + cpu_relax(); + + if (wait) + while (atomic_read(&data.finished) != cpus) + cpu_relax(); +} + + +/** + * smp_call_function_mask(): Run a function on a set of other CPUs. + * @mask: The set of cpus to run on. Must not include the current cpu. + * @func: The function to run. This must be fast and non-blocking. + * @info: An arbitrary pointer to pass to the function. + * @wait: If true, wait (atomically) until function has completed on other CPUs. + * + * Returns 0 on success, else a negative status code. + * + * If @wait is true, then returns once @func has returned; otherwise + * it returns just before the target cpu calls @func. + * + * You must not call this function with disabled interrupts or from a + * hardware interrupt handler or from a bottom half handler. + */ +static int +native_smp_call_function_mask(cpumask_t mask, + void (*func)(void *), void *info, + int wait) +{ + struct call_data_struct data; + cpumask_t allbutself; + int cpus; + + /* Can deadlock when called with interrupts disabled */ + WARN_ON(irqs_disabled()); + + /* Holding any lock stops cpus from going down. */ + spin_lock(&call_lock); + + allbutself = cpu_online_map; + cpu_clear(smp_processor_id(), allbutself); + + cpus_and(mask, mask, allbutself); + cpus = cpus_weight(mask); + + if (!cpus) { + spin_unlock(&call_lock); + return 0; + } + + data.func = func; + data.info = info; + atomic_set(&data.started, 0); + data.wait = wait; + if (wait) + atomic_set(&data.finished, 0); + + call_data = &data; + wmb(); + + /* Send a message to other CPUs */ + if (cpus_equal(mask, allbutself)) + send_IPI_allbutself(CALL_FUNCTION_VECTOR); + else + send_IPI_mask(mask, CALL_FUNCTION_VECTOR); + + /* Wait for response */ + while (atomic_read(&data.started) != cpus) + cpu_relax(); + + if (wait) + while (atomic_read(&data.finished) != cpus) + cpu_relax(); + spin_unlock(&call_lock); + + return 0; +} + +static void stop_this_cpu(void *dummy) +{ + local_irq_disable(); + /* + * Remove this CPU: + */ + cpu_clear(smp_processor_id(), cpu_online_map); + disable_local_APIC(); + if (hlt_works(smp_processor_id())) + for (;;) halt(); + for (;;); +} + +/* + * this function calls the 'stop' function on all other CPUs in the system. + */ + +static void native_smp_send_stop(void) +{ + int nolock; + unsigned long flags; + + if (reboot_force) + return; + + /* Don't deadlock on the call lock in panic */ + nolock = !spin_trylock(&call_lock); + local_irq_save(flags); + __smp_call_function(stop_this_cpu, NULL, 0, 0); + if (!nolock) + spin_unlock(&call_lock); + disable_local_APIC(); + local_irq_restore(flags); +} + +/* + * Reschedule call back. Nothing to do, + * all the work is done automatically when + * we return from the interrupt. + */ +void smp_reschedule_interrupt(struct pt_regs *regs) +{ + ack_APIC_irq(); +#ifdef CONFIG_X86_32 + __get_cpu_var(irq_stat).irq_resched_count++; +#else + add_pda(irq_resched_count, 1); +#endif +} + +void smp_call_function_interrupt(struct pt_regs *regs) +{ + void (*func) (void *info) = call_data->func; + void *info = call_data->info; + int wait = call_data->wait; + + ack_APIC_irq(); + /* + * Notify initiating CPU that I've grabbed the data and am + * about to execute the function + */ + mb(); + atomic_inc(&call_data->started); + /* + * At this point the info structure may be out of scope unless wait==1 + */ + irq_enter(); + (*func)(info); +#ifdef CONFIG_X86_32 + __get_cpu_var(irq_stat).irq_call_count++; +#else + add_pda(irq_call_count, 1); +#endif + irq_exit(); + + if (wait) { + mb(); + atomic_inc(&call_data->finished); + } +} + +struct smp_ops smp_ops = { + .smp_prepare_boot_cpu = native_smp_prepare_boot_cpu, + .smp_prepare_cpus = native_smp_prepare_cpus, + .cpu_up = native_cpu_up, + .smp_cpus_done = native_smp_cpus_done, + + .smp_send_stop = native_smp_send_stop, + .smp_send_reschedule = native_smp_send_reschedule, + .smp_call_function_mask = native_smp_call_function_mask, +}; +EXPORT_SYMBOL_GPL(smp_ops); + diff --git a/arch/x86/kernel/smp_32.c b/arch/x86/kernel/smp_32.c deleted file mode 100644 index dc0cde9d16f..00000000000 --- a/arch/x86/kernel/smp_32.c +++ /dev/null @@ -1,712 +0,0 @@ -/* - * Intel SMP support routines. - * - * (c) 1995 Alan Cox, Building #3 <alan@redhat.com> - * (c) 1998-99, 2000 Ingo Molnar <mingo@redhat.com> - * - * This code is released under the GNU General Public License version 2 or - * later. - */ - -#include <linux/init.h> - -#include <linux/mm.h> -#include <linux/delay.h> -#include <linux/spinlock.h> -#include <linux/kernel_stat.h> -#include <linux/mc146818rtc.h> -#include <linux/cache.h> -#include <linux/interrupt.h> -#include <linux/cpu.h> -#include <linux/module.h> - -#include <asm/mtrr.h> -#include <asm/tlbflush.h> -#include <asm/mmu_context.h> -#include <mach_apic.h> - -/* - * Some notes on x86 processor bugs affecting SMP operation: - * - * Pentium, Pentium Pro, II, III (and all CPUs) have bugs. - * The Linux implications for SMP are handled as follows: - * - * Pentium III / [Xeon] - * None of the E1AP-E3AP errata are visible to the user. - * - * E1AP. see PII A1AP - * E2AP. see PII A2AP - * E3AP. see PII A3AP - * - * Pentium II / [Xeon] - * None of the A1AP-A3AP errata are visible to the user. - * - * A1AP. see PPro 1AP - * A2AP. see PPro 2AP - * A3AP. see PPro 7AP - * - * Pentium Pro - * None of 1AP-9AP errata are visible to the normal user, - * except occasional delivery of 'spurious interrupt' as trap #15. - * This is very rare and a non-problem. - * - * 1AP. Linux maps APIC as non-cacheable - * 2AP. worked around in hardware - * 3AP. fixed in C0 and above steppings microcode update. - * Linux does not use excessive STARTUP_IPIs. - * 4AP. worked around in hardware - * 5AP. symmetric IO mode (normal Linux operation) not affected. - * 'noapic' mode has vector 0xf filled out properly. - * 6AP. 'noapic' mode might be affected - fixed in later steppings - * 7AP. We do not assume writes to the LVT deassering IRQs - * 8AP. We do not enable low power mode (deep sleep) during MP bootup - * 9AP. We do not use mixed mode - * - * Pentium - * There is a marginal case where REP MOVS on 100MHz SMP - * machines with B stepping processors can fail. XXX should provide - * an L1cache=Writethrough or L1cache=off option. - * - * B stepping CPUs may hang. There are hardware work arounds - * for this. We warn about it in case your board doesn't have the work - * arounds. Basically that's so I can tell anyone with a B stepping - * CPU and SMP problems "tough". - * - * Specific items [From Pentium Processor Specification Update] - * - * 1AP. Linux doesn't use remote read - * 2AP. Linux doesn't trust APIC errors - * 3AP. We work around this - * 4AP. Linux never generated 3 interrupts of the same priority - * to cause a lost local interrupt. - * 5AP. Remote read is never used - * 6AP. not affected - worked around in hardware - * 7AP. not affected - worked around in hardware - * 8AP. worked around in hardware - we get explicit CS errors if not - * 9AP. only 'noapic' mode affected. Might generate spurious - * interrupts, we log only the first one and count the - * rest silently. - * 10AP. not affected - worked around in hardware - * 11AP. Linux reads the APIC between writes to avoid this, as per - * the documentation. Make sure you preserve this as it affects - * the C stepping chips too. - * 12AP. not affected - worked around in hardware - * 13AP. not affected - worked around in hardware - * 14AP. we always deassert INIT during bootup - * 15AP. not affected - worked around in hardware - * 16AP. not affected - worked around in hardware - * 17AP. not affected - worked around in hardware - * 18AP. not affected - worked around in hardware - * 19AP. not affected - worked around in BIOS - * - * If this sounds worrying believe me these bugs are either ___RARE___, - * or are signal timing bugs worked around in hardware and there's - * about nothing of note with C stepping upwards. - */ - -DEFINE_PER_CPU(struct tlb_state, cpu_tlbstate) ____cacheline_aligned = { &init_mm, 0, }; - -/* - * the following functions deal with sending IPIs between CPUs. - * - * We use 'broadcast', CPU->CPU IPIs and self-IPIs too. - */ - -static inline int __prepare_ICR (unsigned int shortcut, int vector) -{ - unsigned int icr = shortcut | APIC_DEST_LOGICAL; - - switch (vector) { - default: - icr |= APIC_DM_FIXED | vector; - break; - case NMI_VECTOR: - icr |= APIC_DM_NMI; - break; - } - return icr; -} - -static inline int __prepare_ICR2 (unsigned int mask) -{ - return SET_APIC_DEST_FIELD(mask); -} - -void __send_IPI_shortcut(unsigned int shortcut, int vector) -{ - /* - * Subtle. In the case of the 'never do double writes' workaround - * we have to lock out interrupts to be safe. As we don't care - * of the value read we use an atomic rmw access to avoid costly - * cli/sti. Otherwise we use an even cheaper single atomic write - * to the APIC. - */ - unsigned int cfg; - - /* - * Wait for idle. - */ - apic_wait_icr_idle(); - - /* - * No need to touch the target chip field - */ - cfg = __prepare_ICR(shortcut, vector); - - /* - * Send the IPI. The write to APIC_ICR fires this off. - */ - apic_write_around(APIC_ICR, cfg); -} - -void send_IPI_self(int vector) -{ - __send_IPI_shortcut(APIC_DEST_SELF, vector); -} - -/* - * This is used to send an IPI with no shorthand notation (the destination is - * specified in bits 56 to 63 of the ICR). - */ -static inline void __send_IPI_dest_field(unsigned long mask, int vector) -{ - unsigned long cfg; - - /* - * Wait for idle. - */ - if (unlikely(vector == NMI_VECTOR)) - safe_apic_wait_icr_idle(); - else - apic_wait_icr_idle(); - - /* - * prepare target chip field - */ - cfg = __prepare_ICR2(mask); - apic_write_around(APIC_ICR2, cfg); - - /* - * program the ICR - */ - cfg = __prepare_ICR(0, vector); - - /* - * Send the IPI. The write to APIC_ICR fires this off. - */ - apic_write_around(APIC_ICR, cfg); -} - -/* - * This is only used on smaller machines. - */ -void send_IPI_mask_bitmask(cpumask_t cpumask, int vector) -{ - unsigned long mask = cpus_addr(cpumask)[0]; - unsigned long flags; - - local_irq_save(flags); - WARN_ON(mask & ~cpus_addr(cpu_online_map)[0]); - __send_IPI_dest_field(mask, vector); - local_irq_restore(flags); -} - -void send_IPI_mask_sequence(cpumask_t mask, int vector) -{ - unsigned long flags; - unsigned int query_cpu; - - /* - * Hack. The clustered APIC addressing mode doesn't allow us to send - * to an arbitrary mask, so I do a unicasts to each CPU instead. This - * should be modified to do 1 message per cluster ID - mbligh - */ - - local_irq_save(flags); - for_each_possible_cpu(query_cpu) { - if (cpu_isset(query_cpu, mask)) { - __send_IPI_dest_field(cpu_to_logical_apicid(query_cpu), - vector); - } - } - local_irq_restore(flags); -} - -#include <mach_ipi.h> /* must come after the send_IPI functions above for inlining */ - -/* - * Smarter SMP flushing macros. - * c/o Linus Torvalds. - * - * These mean you can really definitely utterly forget about - * writing to user space from interrupts. (Its not allowed anyway). - * - * Optimizations Manfred Spraul <manfred@colorfullife.com> - */ - -static cpumask_t flush_cpumask; -static struct mm_struct * flush_mm; -static unsigned long flush_va; -static DEFINE_SPINLOCK(tlbstate_lock); - -/* - * We cannot call mmdrop() because we are in interrupt context, - * instead update mm->cpu_vm_mask. - * - * We need to reload %cr3 since the page tables may be going - * away from under us.. - */ -void leave_mm(int cpu) -{ - if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) - BUG(); - cpu_clear(cpu, per_cpu(cpu_tlbstate, cpu).active_mm->cpu_vm_mask); - load_cr3(swapper_pg_dir); -} -EXPORT_SYMBOL_GPL(leave_mm); - -/* - * - * The flush IPI assumes that a thread switch happens in this order: - * [cpu0: the cpu that switches] - * 1) switch_mm() either 1a) or 1b) - * 1a) thread switch to a different mm - * 1a1) cpu_clear(cpu, old_mm->cpu_vm_mask); - * Stop ipi delivery for the old mm. This is not synchronized with - * the other cpus, but smp_invalidate_interrupt ignore flush ipis - * for the wrong mm, and in the worst case we perform a superfluous - * tlb flush. - * 1a2) set cpu_tlbstate to TLBSTATE_OK - * Now the smp_invalidate_interrupt won't call leave_mm if cpu0 - * was in lazy tlb mode. - * 1a3) update cpu_tlbstate[].active_mm - * Now cpu0 accepts tlb flushes for the new mm. - * 1a4) cpu_set(cpu, new_mm->cpu_vm_mask); - * Now the other cpus will send tlb flush ipis. - * 1a4) change cr3. - * 1b) thread switch without mm change - * cpu_tlbstate[].active_mm is correct, cpu0 already handles - * flush ipis. - * 1b1) set cpu_tlbstate to TLBSTATE_OK - * 1b2) test_and_set the cpu bit in cpu_vm_mask. - * Atomically set the bit [other cpus will start sending flush ipis], - * and test the bit. - * 1b3) if the bit was 0: leave_mm was called, flush the tlb. - * 2) switch %%esp, ie current - * - * The interrupt must handle 2 special cases: - * - cr3 is changed before %%esp, ie. it cannot use current->{active_,}mm. - * - the cpu performs speculative tlb reads, i.e. even if the cpu only - * runs in kernel space, the cpu could load tlb entries for user space - * pages. - * - * The good news is that cpu_tlbstate is local to each cpu, no - * write/read ordering problems. - */ - -/* - * TLB flush IPI: - * - * 1) Flush the tlb entries if the cpu uses the mm that's being flushed. - * 2) Leave the mm if we are in the lazy tlb mode. - */ - -void smp_invalidate_interrupt(struct pt_regs *regs) -{ - unsigned long cpu; - - cpu = get_cpu(); - - if (!cpu_isset(cpu, flush_cpumask)) - goto out; - /* - * This was a BUG() but until someone can quote me the - * line from the intel manual that guarantees an IPI to - * multiple CPUs is retried _only_ on the erroring CPUs - * its staying as a return - * - * BUG(); - */ - - if (flush_mm == per_cpu(cpu_tlbstate, cpu).active_mm) { - if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) { - if (flush_va == TLB_FLUSH_ALL) - local_flush_tlb(); - else - __flush_tlb_one(flush_va); - } else - leave_mm(cpu); - } - ack_APIC_irq(); - smp_mb__before_clear_bit(); - cpu_clear(cpu, flush_cpumask); - smp_mb__after_clear_bit(); -out: - put_cpu_no_resched(); - __get_cpu_var(irq_stat).irq_tlb_count++; -} - -void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm, - unsigned long va) -{ - cpumask_t cpumask = *cpumaskp; - - /* - * A couple of (to be removed) sanity checks: - * - * - current CPU must not be in mask - * - mask must exist :) - */ - BUG_ON(cpus_empty(cpumask)); - BUG_ON(cpu_isset(smp_processor_id(), cpumask)); - BUG_ON(!mm); - -#ifdef CONFIG_HOTPLUG_CPU - /* If a CPU which we ran on has gone down, OK. */ - cpus_and(cpumask, cpumask, cpu_online_map); - if (unlikely(cpus_empty(cpumask))) - return; -#endif - - /* - * i'm not happy about this global shared spinlock in the - * MM hot path, but we'll see how contended it is. - * AK: x86-64 has a faster method that could be ported. - */ - spin_lock(&tlbstate_lock); - - flush_mm = mm; - flush_va = va; - cpus_or(flush_cpumask, cpumask, flush_cpumask); - /* - * We have to send the IPI only to - * CPUs affected. - */ - send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR); - - while (!cpus_empty(flush_cpumask)) - /* nothing. lockup detection does not belong here */ - cpu_relax(); - - flush_mm = NULL; - flush_va = 0; - spin_unlock(&tlbstate_lock); -} - -void flush_tlb_current_task(void) -{ - struct mm_struct *mm = current->mm; - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - local_flush_tlb(); - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL); - preempt_enable(); -} - -void flush_tlb_mm (struct mm_struct * mm) -{ - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - if (current->active_mm == mm) { - if (current->mm) - local_flush_tlb(); - else - leave_mm(smp_processor_id()); - } - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL); - - preempt_enable(); -} - -void flush_tlb_page(struct vm_area_struct * vma, unsigned long va) -{ - struct mm_struct *mm = vma->vm_mm; - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - if (current->active_mm == mm) { - if(current->mm) - __flush_tlb_one(va); - else - leave_mm(smp_processor_id()); - } - - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, va); - - preempt_enable(); -} -EXPORT_SYMBOL(flush_tlb_page); - -static void do_flush_tlb_all(void* info) -{ - unsigned long cpu = smp_processor_id(); - - __flush_tlb_all(); - if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_LAZY) - leave_mm(cpu); -} - -void flush_tlb_all(void) -{ - on_each_cpu(do_flush_tlb_all, NULL, 1, 1); -} - -/* - * this function sends a 'reschedule' IPI to another CPU. - * it goes straight through and wastes no time serializing - * anything. Worst case is that we lose a reschedule ... - */ -static void native_smp_send_reschedule(int cpu) -{ - WARN_ON(cpu_is_offline(cpu)); - send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR); -} - -/* - * Structure and data for smp_call_function(). This is designed to minimise - * static memory requirements. It also looks cleaner. - */ -static DEFINE_SPINLOCK(call_lock); - -struct call_data_struct { - void (*func) (void *info); - void *info; - atomic_t started; - atomic_t finished; - int wait; -}; - -void lock_ipi_call_lock(void) -{ - spin_lock_irq(&call_lock); -} - -void unlock_ipi_call_lock(void) -{ - spin_unlock_irq(&call_lock); -} - -static struct call_data_struct *call_data; - -static void __smp_call_function(void (*func) (void *info), void *info, - int nonatomic, int wait) -{ - struct call_data_struct data; - int cpus = num_online_cpus() - 1; - - if (!cpus) - return; - - data.func = func; - data.info = info; - atomic_set(&data.started, 0); - data.wait = wait; - if (wait) - atomic_set(&data.finished, 0); - - call_data = &data; - mb(); - - /* Send a message to all other CPUs and wait for them to respond */ - send_IPI_allbutself(CALL_FUNCTION_VECTOR); - - /* Wait for response */ - while (atomic_read(&data.started) != cpus) - cpu_relax(); - - if (wait) - while (atomic_read(&data.finished) != cpus) - cpu_relax(); -} - - -/** - * smp_call_function_mask(): Run a function on a set of other CPUs. - * @mask: The set of cpus to run on. Must not include the current cpu. - * @func: The function to run. This must be fast and non-blocking. - * @info: An arbitrary pointer to pass to the function. - * @wait: If true, wait (atomically) until function has completed on other CPUs. - * - * Returns 0 on success, else a negative status code. - * - * If @wait is true, then returns once @func has returned; otherwise - * it returns just before the target cpu calls @func. - * - * You must not call this function with disabled interrupts or from a - * hardware interrupt handler or from a bottom half handler. - */ -static int -native_smp_call_function_mask(cpumask_t mask, - void (*func)(void *), void *info, - int wait) -{ - struct call_data_struct data; - cpumask_t allbutself; - int cpus; - - /* Can deadlock when called with interrupts disabled */ - WARN_ON(irqs_disabled()); - - /* Holding any lock stops cpus from going down. */ - spin_lock(&call_lock); - - allbutself = cpu_online_map; - cpu_clear(smp_processor_id(), allbutself); - - cpus_and(mask, mask, allbutself); - cpus = cpus_weight(mask); - - if (!cpus) { - spin_unlock(&call_lock); - return 0; - } - - data.func = func; - data.info = info; - atomic_set(&data.started, 0); - data.wait = wait; - if (wait) - atomic_set(&data.finished, 0); - - call_data = &data; - mb(); - - /* Send a message to other CPUs */ - if (cpus_equal(mask, allbutself)) - send_IPI_allbutself(CALL_FUNCTION_VECTOR); - else - send_IPI_mask(mask, CALL_FUNCTION_VECTOR); - - /* Wait for response */ - while (atomic_read(&data.started) != cpus) - cpu_relax(); - - if (wait) - while (atomic_read(&data.finished) != cpus) - cpu_relax(); - spin_unlock(&call_lock); - - return 0; -} - -static void stop_this_cpu (void * dummy) -{ - local_irq_disable(); - /* - * Remove this CPU: - */ - cpu_clear(smp_processor_id(), cpu_online_map); - disable_local_APIC(); - if (cpu_data(smp_processor_id()).hlt_works_ok) - for(;;) halt(); - for (;;); -} - -/* - * this function calls the 'stop' function on all other CPUs in the system. - */ - -static void native_smp_send_stop(void) -{ - /* Don't deadlock on the call lock in panic */ - int nolock = !spin_trylock(&call_lock); - unsigned long flags; - - local_irq_save(flags); - __smp_call_function(stop_this_cpu, NULL, 0, 0); - if (!nolock) - spin_unlock(&call_lock); - disable_local_APIC(); - local_irq_restore(flags); -} - -/* - * Reschedule call back. Nothing to do, - * all the work is done automatically when - * we return from the interrupt. - */ -void smp_reschedule_interrupt(struct pt_regs *regs) -{ - ack_APIC_irq(); - __get_cpu_var(irq_stat).irq_resched_count++; -} - -void smp_call_function_interrupt(struct pt_regs *regs) -{ - void (*func) (void *info) = call_data->func; - void *info = call_data->info; - int wait = call_data->wait; - - ack_APIC_irq(); - /* - * Notify initiating CPU that I've grabbed the data and am - * about to execute the function - */ - mb(); - atomic_inc(&call_data->started); - /* - * At this point the info structure may be out of scope unless wait==1 - */ - irq_enter(); - (*func)(info); - __get_cpu_var(irq_stat).irq_call_count++; - irq_exit(); - - if (wait) { - mb(); - atomic_inc(&call_data->finished); - } -} - -static int convert_apicid_to_cpu(int apic_id) -{ - int i; - - for_each_possible_cpu(i) { - if (per_cpu(x86_cpu_to_apicid, i) == apic_id) - return i; - } - return -1; -} - -int safe_smp_processor_id(void) -{ - int apicid, cpuid; - - if (!boot_cpu_has(X86_FEATURE_APIC)) - return 0; - - apicid = hard_smp_processor_id(); - if (apicid == BAD_APICID) - return 0; - - cpuid = convert_apicid_to_cpu(apicid); - - return cpuid >= 0 ? cpuid : 0; -} - -struct smp_ops smp_ops = { - .smp_prepare_boot_cpu = native_smp_prepare_boot_cpu, - .smp_prepare_cpus = native_smp_prepare_cpus, - .cpu_up = native_cpu_up, - .smp_cpus_done = native_smp_cpus_done, - - .smp_send_stop = native_smp_send_stop, - .smp_send_reschedule = native_smp_send_reschedule, - .smp_call_function_mask = native_smp_call_function_mask, -}; -EXPORT_SYMBOL_GPL(smp_ops); diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot.c index 579b9b740c7..e6abe8a49b1 100644 --- a/arch/x86/kernel/smpboot_32.c +++ b/arch/x86/kernel/smpboot.c @@ -3,6 +3,7 @@ * * (c) 1995 Alan Cox, Building #3 <alan@redhat.com> * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com> + * Copyright 2001 Andi Kleen, SuSE Labs. * * Much of the core SMP work is based on previous work by Thomas Radke, to * whom a great many thanks are extended. @@ -29,53 +30,90 @@ * Ingo Molnar : various cleanups and rewrites * Tigran Aivazian : fixed "0.00 in /proc/uptime on SMP" bug. * Maciej W. Rozycki : Bits for genuine 82489DX APICs + * Andi Kleen : Changed for SMP boot into long mode. * Martin J. Bligh : Added support for multi-quad systems * Dave Jones : Report invalid combinations of Athlon CPUs. -* Rusty Russell : Hacked into shape for new "hotplug" boot process. */ + * Rusty Russell : Hacked into shape for new "hotplug" boot process. + * Andi Kleen : Converted to new state machine. + * Ashok Raj : CPU hotplug support + * Glauber Costa : i386 and x86_64 integration + */ -#include <linux/module.h> #include <linux/init.h> -#include <linux/kernel.h> - -#include <linux/mm.h> +#include <linux/smp.h> +#include <linux/module.h> #include <linux/sched.h> -#include <linux/kernel_stat.h> -#include <linux/bootmem.h> -#include <linux/notifier.h> -#include <linux/cpu.h> #include <linux/percpu.h> +#include <linux/bootmem.h> +#include <linux/err.h> #include <linux/nmi.h> -#include <linux/delay.h> -#include <linux/mc146818rtc.h> -#include <asm/tlbflush.h> +#include <asm/acpi.h> #include <asm/desc.h> -#include <asm/arch_hooks.h> #include <asm/nmi.h> +#include <asm/irq.h> +#include <asm/smp.h> +#include <asm/trampoline.h> +#include <asm/cpu.h> +#include <asm/numa.h> +#include <asm/pgtable.h> +#include <asm/tlbflush.h> +#include <asm/mtrr.h> +#include <asm/nmi.h> +#include <asm/vmi.h> +#include <linux/mc146818rtc.h> #include <mach_apic.h> #include <mach_wakecpu.h> #include <smpboot_hooks.h> -#include <asm/vmi.h> -#include <asm/mtrr.h> -/* Set if we find a B stepping CPU */ -static int __cpuinitdata smp_b_stepping; +/* + * FIXME: For x86_64, those are defined in other files. But moving them here, + * would make the setup areas dependent on smp, which is a loss. When we + * integrate apic between arches, we can probably do a better job, but + * right now, they'll stay here -- glommer + */ + +/* which logical CPU number maps to which CPU (physical APIC ID) */ +u16 x86_cpu_to_apicid_init[NR_CPUS] __initdata = + { [0 ... NR_CPUS-1] = BAD_APICID }; +void *x86_cpu_to_apicid_early_ptr; + +u16 x86_bios_cpu_apicid_init[NR_CPUS] __initdata + = { [0 ... NR_CPUS-1] = BAD_APICID }; +void *x86_bios_cpu_apicid_early_ptr; + +#ifdef CONFIG_X86_32 +u8 apicid_2_node[MAX_APICID]; +#endif + +/* State of each CPU */ +DEFINE_PER_CPU(int, cpu_state) = { 0 }; + +/* Store all idle threads, this can be reused instead of creating +* a new thread. Also avoids complicated thread destroy functionality +* for idle threads. +*/ +#ifdef CONFIG_HOTPLUG_CPU +/* + * Needed only for CONFIG_HOTPLUG_CPU because __cpuinitdata is + * removed after init for !CONFIG_HOTPLUG_CPU. + */ +static DEFINE_PER_CPU(struct task_struct *, idle_thread_array); +#define get_idle_for_cpu(x) (per_cpu(idle_thread_array, x)) +#define set_idle_for_cpu(x, p) (per_cpu(idle_thread_array, x) = (p)) +#else +struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ; +#define get_idle_for_cpu(x) (idle_thread_array[(x)]) +#define set_idle_for_cpu(x, p) (idle_thread_array[(x)] = (p)) +#endif /* Number of siblings per CPU package */ int smp_num_siblings = 1; EXPORT_SYMBOL(smp_num_siblings); /* Last level cache ID of each logical CPU */ -DEFINE_PER_CPU(u8, cpu_llc_id) = BAD_APICID; - -/* representing HT siblings of each logical CPU */ -DEFINE_PER_CPU(cpumask_t, cpu_sibling_map); -EXPORT_PER_CPU_SYMBOL(cpu_sibling_map); - -/* representing HT and core siblings of each logical CPU */ -DEFINE_PER_CPU(cpumask_t, cpu_core_map); -EXPORT_PER_CPU_SYMBOL(cpu_core_map); +DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID; /* bitmap of online cpus */ cpumask_t cpu_online_map __read_mostly; @@ -85,126 +123,94 @@ cpumask_t cpu_callin_map; cpumask_t cpu_callout_map; cpumask_t cpu_possible_map; EXPORT_SYMBOL(cpu_possible_map); -static cpumask_t smp_commenced_mask; + +/* representing HT siblings of each logical CPU */ +DEFINE_PER_CPU(cpumask_t, cpu_sibling_map); +EXPORT_PER_CPU_SYMBOL(cpu_sibling_map); + +/* representing HT and core siblings of each logical CPU */ +DEFINE_PER_CPU(cpumask_t, cpu_core_map); +EXPORT_PER_CPU_SYMBOL(cpu_core_map); /* Per CPU bogomips and other parameters */ DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info); EXPORT_PER_CPU_SYMBOL(cpu_info); -/* which logical CPU number maps to which CPU (physical APIC ID) */ -u8 x86_cpu_to_apicid_init[NR_CPUS] __initdata = - { [0 ... NR_CPUS-1] = BAD_APICID }; -void *x86_cpu_to_apicid_early_ptr; -DEFINE_PER_CPU(u8, x86_cpu_to_apicid) = BAD_APICID; -EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid); - -u8 apicid_2_node[MAX_APICID]; +static atomic_t init_deasserted; -/* - * Trampoline 80x86 program as an array. - */ +static int boot_cpu_logical_apicid; -extern const unsigned char trampoline_data []; -extern const unsigned char trampoline_end []; -static unsigned char *trampoline_base; +/* representing cpus for which sibling maps can be computed */ +static cpumask_t cpu_sibling_setup_map; -static void map_cpu_to_logical_apicid(void); +/* Set if we find a B stepping CPU */ +int __cpuinitdata smp_b_stepping; -/* State of each CPU. */ -DEFINE_PER_CPU(int, cpu_state) = { 0 }; +#if defined(CONFIG_NUMA) && defined(CONFIG_X86_32) -/* - * Currently trivial. Write the real->protected mode - * bootstrap into the page concerned. The caller - * has made sure it's suitably aligned. - */ +/* which logical CPUs are on which nodes */ +cpumask_t node_to_cpumask_map[MAX_NUMNODES] __read_mostly = + { [0 ... MAX_NUMNODES-1] = CPU_MASK_NONE }; +EXPORT_SYMBOL(node_to_cpumask_map); +/* which node each logical CPU is on */ +int cpu_to_node_map[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = 0 }; +EXPORT_SYMBOL(cpu_to_node_map); -static unsigned long __cpuinit setup_trampoline(void) +/* set up a mapping between cpu and node. */ +static void map_cpu_to_node(int cpu, int node) { - memcpy(trampoline_base, trampoline_data, trampoline_end - trampoline_data); - return virt_to_phys(trampoline_base); + printk(KERN_INFO "Mapping cpu %d to node %d\n", cpu, node); + cpu_set(cpu, node_to_cpumask_map[node]); + cpu_to_node_map[cpu] = node; } -/* - * We are called very early to get the low memory for the - * SMP bootup trampoline page. - */ -void __init smp_alloc_memory(void) +/* undo a mapping between cpu and node. */ +static void unmap_cpu_to_node(int cpu) { - trampoline_base = alloc_bootmem_low_pages(PAGE_SIZE); - /* - * Has to be in very low memory so we can execute - * real-mode AP code. - */ - if (__pa(trampoline_base) >= 0x9F000) - BUG(); + int node; + + printk(KERN_INFO "Unmapping cpu %d from all nodes\n", cpu); + for (node = 0; node < MAX_NUMNODES; node++) + cpu_clear(cpu, node_to_cpumask_map[node]); + cpu_to_node_map[cpu] = 0; } +#else /* !(CONFIG_NUMA && CONFIG_X86_32) */ +#define map_cpu_to_node(cpu, node) ({}) +#define unmap_cpu_to_node(cpu) ({}) +#endif -/* - * The bootstrap kernel entry code has set these up. Save them for - * a given CPU - */ +#ifdef CONFIG_X86_32 +u8 cpu_2_logical_apicid[NR_CPUS] __read_mostly = + { [0 ... NR_CPUS-1] = BAD_APICID }; -void __cpuinit smp_store_cpu_info(int id) +void map_cpu_to_logical_apicid(void) { - struct cpuinfo_x86 *c = &cpu_data(id); - - *c = boot_cpu_data; - c->cpu_index = id; - if (id!=0) - identify_secondary_cpu(c); - /* - * Mask B, Pentium, but not Pentium MMX - */ - if (c->x86_vendor == X86_VENDOR_INTEL && - c->x86 == 5 && - c->x86_mask >= 1 && c->x86_mask <= 4 && - c->x86_model <= 3) - /* - * Remember we have B step Pentia with bugs - */ - smp_b_stepping = 1; - - /* - * Certain Athlons might work (for various values of 'work') in SMP - * but they are not certified as MP capable. - */ - if ((c->x86_vendor == X86_VENDOR_AMD) && (c->x86 == 6)) { - - if (num_possible_cpus() == 1) - goto valid_k7; - - /* Athlon 660/661 is valid. */ - if ((c->x86_model==6) && ((c->x86_mask==0) || (c->x86_mask==1))) - goto valid_k7; - - /* Duron 670 is valid */ - if ((c->x86_model==7) && (c->x86_mask==0)) - goto valid_k7; - - /* - * Athlon 662, Duron 671, and Athlon >model 7 have capability bit. - * It's worth noting that the A5 stepping (662) of some Athlon XP's - * have the MP bit set. - * See http://www.heise.de/newsticker/data/jow-18.10.01-000 for more. - */ - if (((c->x86_model==6) && (c->x86_mask>=2)) || - ((c->x86_model==7) && (c->x86_mask>=1)) || - (c->x86_model> 7)) - if (cpu_has_mp) - goto valid_k7; + int cpu = smp_processor_id(); + int apicid = logical_smp_processor_id(); + int node = apicid_to_node(apicid); - /* If we get here, it's not a certified SMP capable AMD system. */ - add_taint(TAINT_UNSAFE_SMP); - } + if (!node_online(node)) + node = first_online_node; -valid_k7: - ; + cpu_2_logical_apicid[cpu] = apicid; + map_cpu_to_node(cpu, node); } -static atomic_t init_deasserted; +void unmap_cpu_to_logical_apicid(int cpu) +{ + cpu_2_logical_apicid[cpu] = BAD_APICID; + unmap_cpu_to_node(cpu); +} +#else +#define unmap_cpu_to_logical_apicid(cpu) do {} while (0) +#define map_cpu_to_logical_apicid() do {} while (0) +#endif -static void __cpuinit smp_callin(void) +/* + * Report back to the Boot Processor. + * Running on AP. + */ +void __cpuinit smp_callin(void) { int cpuid, phys_id; unsigned long timeout; @@ -220,12 +226,11 @@ static void __cpuinit smp_callin(void) /* * (This works even if the APIC is not enabled.) */ - phys_id = GET_APIC_ID(apic_read(APIC_ID)); + phys_id = GET_APIC_ID(read_apic_id()); cpuid = smp_processor_id(); if (cpu_isset(cpuid, cpu_callin_map)) { - printk("huh, phys CPU#%d, CPU#%d already present??\n", + panic("%s: phys CPU#%d, CPU#%d already present??\n", __func__, phys_id, cpuid); - BUG(); } Dprintk("CPU#%d (phys ID: %d) waiting for CALLOUT\n", cpuid, phys_id); @@ -247,13 +252,12 @@ static void __cpuinit smp_callin(void) */ if (cpu_isset(cpuid, cpu_callout_map)) break; - rep_nop(); + cpu_relax(); } if (!time_before(jiffies, timeout)) { - printk("BUG: CPU%d started up but did not get a callout!\n", - cpuid); - BUG(); + panic("%s: CPU%d started up but did not get a callout!\n", + __func__, cpuid); } /* @@ -266,13 +270,19 @@ static void __cpuinit smp_callin(void) Dprintk("CALLIN, before setup_local_APIC().\n"); smp_callin_clear_local_apic(); setup_local_APIC(); + end_local_APIC_setup(); map_cpu_to_logical_apicid(); /* * Get our bogomips. + * + * Need to enable IRQs because it can take longer and then + * the NMI watchdog might kill us. */ + local_irq_enable(); calibrate_delay(); - Dprintk("Stack at about %p\n",&cpuid); + local_irq_disable(); + Dprintk("Stack at about %p\n", &cpuid); /* * Save our processor parameters @@ -285,91 +295,10 @@ static void __cpuinit smp_callin(void) cpu_set(cpuid, cpu_callin_map); } -static int cpucount; - -/* maps the cpu to the sched domain representing multi-core */ -cpumask_t cpu_coregroup_map(int cpu) -{ - struct cpuinfo_x86 *c = &cpu_data(cpu); - /* - * For perf, we return last level cache shared map. - * And for power savings, we return cpu_core_map - */ - if (sched_mc_power_savings || sched_smt_power_savings) - return per_cpu(cpu_core_map, cpu); - else - return c->llc_shared_map; -} - -/* representing cpus for which sibling maps can be computed */ -static cpumask_t cpu_sibling_setup_map; - -void __cpuinit set_cpu_sibling_map(int cpu) -{ - int i; - struct cpuinfo_x86 *c = &cpu_data(cpu); - - cpu_set(cpu, cpu_sibling_setup_map); - - if (smp_num_siblings > 1) { - for_each_cpu_mask(i, cpu_sibling_setup_map) { - if (c->phys_proc_id == cpu_data(i).phys_proc_id && - c->cpu_core_id == cpu_data(i).cpu_core_id) { - cpu_set(i, per_cpu(cpu_sibling_map, cpu)); - cpu_set(cpu, per_cpu(cpu_sibling_map, i)); - cpu_set(i, per_cpu(cpu_core_map, cpu)); - cpu_set(cpu, per_cpu(cpu_core_map, i)); - cpu_set(i, c->llc_shared_map); - cpu_set(cpu, cpu_data(i).llc_shared_map); - } - } - } else { - cpu_set(cpu, per_cpu(cpu_sibling_map, cpu)); - } - - cpu_set(cpu, c->llc_shared_map); - - if (current_cpu_data.x86_max_cores == 1) { - per_cpu(cpu_core_map, cpu) = per_cpu(cpu_sibling_map, cpu); - c->booted_cores = 1; - return; - } - - for_each_cpu_mask(i, cpu_sibling_setup_map) { - if (per_cpu(cpu_llc_id, cpu) != BAD_APICID && - per_cpu(cpu_llc_id, cpu) == per_cpu(cpu_llc_id, i)) { - cpu_set(i, c->llc_shared_map); - cpu_set(cpu, cpu_data(i).llc_shared_map); - } - if (c->phys_proc_id == cpu_data(i).phys_proc_id) { - cpu_set(i, per_cpu(cpu_core_map, cpu)); - cpu_set(cpu, per_cpu(cpu_core_map, i)); - /* - * Does this new cpu bringup a new core? - */ - if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1) { - /* - * for each core in package, increment - * the booted_cores for this new cpu - */ - if (first_cpu(per_cpu(cpu_sibling_map, i)) == i) - c->booted_cores++; - /* - * increment the core count for all - * the other cpus in this package - */ - if (i != cpu) - cpu_data(i).booted_cores++; - } else if (i != cpu && !c->booted_cores) - c->booted_cores = cpu_data(i).booted_cores; - } - } -} - /* * Activate a secondary processor. */ -static void __cpuinit start_secondary(void *unused) +void __cpuinit start_secondary(void *unused) { /* * Don't put *anything* before cpu_init(), SMP booting is too @@ -382,24 +311,19 @@ static void __cpuinit start_secondary(void *unused) cpu_init(); preempt_disable(); smp_callin(); - while (!cpu_isset(smp_processor_id(), smp_commenced_mask)) - rep_nop(); + + /* otherwise gcc will move up smp_processor_id before the cpu_init */ + barrier(); /* * Check TSC synchronization with the BP: */ check_tsc_sync_target(); - setup_secondary_clock(); if (nmi_watchdog == NMI_IO_APIC) { disable_8259A_irq(0); enable_NMI_through_LVT0(); enable_8259A_irq(0); } - /* - * low-memory mappings have been cleared, flush them from - * the local TLBs too. - */ - local_flush_tlb(); /* This must be done before setting cpu_online_map */ set_cpu_sibling_map(raw_smp_processor_id()); @@ -414,17 +338,27 @@ static void __cpuinit start_secondary(void *unused) * smp_call_function(). */ lock_ipi_call_lock(); +#ifdef CONFIG_X86_64 + spin_lock(&vector_lock); + + /* Setup the per cpu irq handling data structures */ + __setup_vector_irq(smp_processor_id()); + /* + * Allow the master to continue. + */ + spin_unlock(&vector_lock); +#endif cpu_set(smp_processor_id(), cpu_online_map); unlock_ipi_call_lock(); per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE; - /* We can take interrupts now: we're officially "up". */ - local_irq_enable(); + setup_secondary_clock(); wmb(); cpu_idle(); } +#ifdef CONFIG_X86_32 /* * Everything has been set up for the secondary * CPUs - they just need to reload everything @@ -442,89 +376,233 @@ void __devinit initialize_secondary(void) "movl %0,%%esp\n\t" "jmp *%1" : - :"m" (current->thread.sp),"m" (current->thread.ip)); + :"m" (current->thread.sp), "m" (current->thread.ip)); } +#endif -/* Static state in head.S used to set up a CPU */ -extern struct { - void * sp; - unsigned short ss; -} stack_start; +static void __cpuinit smp_apply_quirks(struct cpuinfo_x86 *c) +{ +#ifdef CONFIG_X86_32 + /* + * Mask B, Pentium, but not Pentium MMX + */ + if (c->x86_vendor == X86_VENDOR_INTEL && + c->x86 == 5 && + c->x86_mask >= 1 && c->x86_mask <= 4 && + c->x86_model <= 3) + /* + * Remember we have B step Pentia with bugs + */ + smp_b_stepping = 1; -#ifdef CONFIG_NUMA + /* + * Certain Athlons might work (for various values of 'work') in SMP + * but they are not certified as MP capable. + */ + if ((c->x86_vendor == X86_VENDOR_AMD) && (c->x86 == 6)) { -/* which logical CPUs are on which nodes */ -cpumask_t node_to_cpumask_map[MAX_NUMNODES] __read_mostly = - { [0 ... MAX_NUMNODES-1] = CPU_MASK_NONE }; -EXPORT_SYMBOL(node_to_cpumask_map); -/* which node each logical CPU is on */ -int cpu_to_node_map[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = 0 }; -EXPORT_SYMBOL(cpu_to_node_map); + if (num_possible_cpus() == 1) + goto valid_k7; -/* set up a mapping between cpu and node. */ -static inline void map_cpu_to_node(int cpu, int node) -{ - printk("Mapping cpu %d to node %d\n", cpu, node); - cpu_set(cpu, node_to_cpumask_map[node]); - cpu_to_node_map[cpu] = node; + /* Athlon 660/661 is valid. */ + if ((c->x86_model == 6) && ((c->x86_mask == 0) || + (c->x86_mask == 1))) + goto valid_k7; + + /* Duron 670 is valid */ + if ((c->x86_model == 7) && (c->x86_mask == 0)) + goto valid_k7; + + /* + * Athlon 662, Duron 671, and Athlon >model 7 have capability + * bit. It's worth noting that the A5 stepping (662) of some + * Athlon XP's have the MP bit set. + * See http://www.heise.de/newsticker/data/jow-18.10.01-000 for + * more. + */ + if (((c->x86_model == 6) && (c->x86_mask >= 2)) || + ((c->x86_model == 7) && (c->x86_mask >= 1)) || + (c->x86_model > 7)) + if (cpu_has_mp) + goto valid_k7; + + /* If we get here, not a certified SMP capable AMD system. */ + add_taint(TAINT_UNSAFE_SMP); + } + +valid_k7: + ; +#endif } -/* undo a mapping between cpu and node. */ -static inline void unmap_cpu_to_node(int cpu) +void __cpuinit smp_checks(void) { - int node; + if (smp_b_stepping) + printk(KERN_WARNING "WARNING: SMP operation may be unreliable" + "with B stepping processors.\n"); - printk("Unmapping cpu %d from all nodes\n", cpu); - for (node = 0; node < MAX_NUMNODES; node ++) - cpu_clear(cpu, node_to_cpumask_map[node]); - cpu_to_node_map[cpu] = 0; + /* + * Don't taint if we are running SMP kernel on a single non-MP + * approved Athlon + */ + if (tainted & TAINT_UNSAFE_SMP) { + if (num_online_cpus()) + printk(KERN_INFO "WARNING: This combination of AMD" + "processors is not suitable for SMP.\n"); + else + tainted &= ~TAINT_UNSAFE_SMP; + } } -#else /* !CONFIG_NUMA */ -#define map_cpu_to_node(cpu, node) ({}) -#define unmap_cpu_to_node(cpu) ({}) +/* + * The bootstrap kernel entry code has set these up. Save them for + * a given CPU + */ -#endif /* CONFIG_NUMA */ +void __cpuinit smp_store_cpu_info(int id) +{ + struct cpuinfo_x86 *c = &cpu_data(id); + + *c = boot_cpu_data; + c->cpu_index = id; + if (id != 0) + identify_secondary_cpu(c); + smp_apply_quirks(c); +} -u8 cpu_2_logical_apicid[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = BAD_APICID }; -static void map_cpu_to_logical_apicid(void) +void __cpuinit set_cpu_sibling_map(int cpu) { - int cpu = smp_processor_id(); - int apicid = logical_smp_processor_id(); - int node = apicid_to_node(apicid); + int i; + struct cpuinfo_x86 *c = &cpu_data(cpu); - if (!node_online(node)) - node = first_online_node; + cpu_set(cpu, cpu_sibling_setup_map); - cpu_2_logical_apicid[cpu] = apicid; - map_cpu_to_node(cpu, node); + if (smp_num_siblings > 1) { + for_each_cpu_mask(i, cpu_sibling_setup_map) { + if (c->phys_proc_id == cpu_data(i).phys_proc_id && + c->cpu_core_id == cpu_data(i).cpu_core_id) { + cpu_set(i, per_cpu(cpu_sibling_map, cpu)); + cpu_set(cpu, per_cpu(cpu_sibling_map, i)); + cpu_set(i, per_cpu(cpu_core_map, cpu)); + cpu_set(cpu, per_cpu(cpu_core_map, i)); + cpu_set(i, c->llc_shared_map); + cpu_set(cpu, cpu_data(i).llc_shared_map); + } + } + } else { + cpu_set(cpu, per_cpu(cpu_sibling_map, cpu)); + } + + cpu_set(cpu, c->llc_shared_map); + + if (current_cpu_data.x86_max_cores == 1) { + per_cpu(cpu_core_map, cpu) = per_cpu(cpu_sibling_map, cpu); + c->booted_cores = 1; + return; + } + + for_each_cpu_mask(i, cpu_sibling_setup_map) { + if (per_cpu(cpu_llc_id, cpu) != BAD_APICID && + per_cpu(cpu_llc_id, cpu) == per_cpu(cpu_llc_id, i)) { + cpu_set(i, c->llc_shared_map); + cpu_set(cpu, cpu_data(i).llc_shared_map); + } + if (c->phys_proc_id == cpu_data(i).phys_proc_id) { + cpu_set(i, per_cpu(cpu_core_map, cpu)); + cpu_set(cpu, per_cpu(cpu_core_map, i)); + /* + * Does this new cpu bringup a new core? + */ + if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1) { + /* + * for each core in package, increment + * the booted_cores for this new cpu + */ + if (first_cpu(per_cpu(cpu_sibling_map, i)) == i) + c->booted_cores++; + /* + * increment the core count for all + * the other cpus in this package + */ + if (i != cpu) + cpu_data(i).booted_cores++; + } else if (i != cpu && !c->booted_cores) + c->booted_cores = cpu_data(i).booted_cores; + } + } } -static void unmap_cpu_to_logical_apicid(int cpu) +/* maps the cpu to the sched domain representing multi-core */ +cpumask_t cpu_coregroup_map(int cpu) { - cpu_2_logical_apicid[cpu] = BAD_APICID; - unmap_cpu_to_node(cpu); + struct cpuinfo_x86 *c = &cpu_data(cpu); + /* + * For perf, we return last level cache shared map. + * And for power savings, we return cpu_core_map + */ + if (sched_mc_power_savings || sched_smt_power_savings) + return per_cpu(cpu_core_map, cpu); + else + return c->llc_shared_map; +} + +#ifdef CONFIG_X86_32 +/* + * We are called very early to get the low memory for the + * SMP bootup trampoline page. + */ +void __init smp_alloc_memory(void) +{ + trampoline_base = alloc_bootmem_low_pages(PAGE_SIZE); + /* + * Has to be in very low memory so we can execute + * real-mode AP code. + */ + if (__pa(trampoline_base) >= 0x9F000) + BUG(); +} +#endif + +void impress_friends(void) +{ + int cpu; + unsigned long bogosum = 0; + /* + * Allow the user to impress friends. + */ + Dprintk("Before bogomips.\n"); + for_each_possible_cpu(cpu) + if (cpu_isset(cpu, cpu_callout_map)) + bogosum += cpu_data(cpu).loops_per_jiffy; + printk(KERN_INFO + "Total of %d processors activated (%lu.%02lu BogoMIPS).\n", + num_online_cpus(), + bogosum/(500000/HZ), + (bogosum/(5000/HZ))%100); + + Dprintk("Before bogocount - setting activated=1.\n"); } static inline void __inquire_remote_apic(int apicid) { - int i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 }; + unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 }; char *names[] = { "ID", "VERSION", "SPIV" }; int timeout; - unsigned long status; + u32 status; - printk("Inquiring remote APIC #%d...\n", apicid); + printk(KERN_INFO "Inquiring remote APIC #%d...\n", apicid); for (i = 0; i < ARRAY_SIZE(regs); i++) { - printk("... APIC #%d %s: ", apicid, names[i]); + printk(KERN_INFO "... APIC #%d %s: ", apicid, names[i]); /* * Wait for idle. */ status = safe_apic_wait_icr_idle(); if (status) - printk("a previous APIC delivery may have failed\n"); + printk(KERN_CONT + "a previous APIC delivery may have failed\n"); apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(apicid)); apic_write_around(APIC_ICR, APIC_DM_REMRD | regs[i]); @@ -538,16 +616,16 @@ static inline void __inquire_remote_apic(int apicid) switch (status) { case APIC_ICR_RR_VALID: status = apic_read(APIC_RRR); - printk("%lx\n", status); + printk(KERN_CONT "%08x\n", status); break; default: - printk("failed\n"); + printk(KERN_CONT "failed\n"); } } } #ifdef WAKE_SECONDARY_VIA_NMI -/* +/* * Poke the other CPU in the eye via NMI to wake it up. Remember that the normal * INIT, INIT, STARTUP sequence will reset the chip hard for us, and this * won't ... remember to clear down the APIC, etc later. @@ -584,9 +662,9 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip) Dprintk("NMI sent.\n"); if (send_status) - printk("APIC never delivered???\n"); + printk(KERN_ERR "APIC never delivered???\n"); if (accept_status) - printk("APIC delivery error (%lx).\n", accept_status); + printk(KERN_ERR "APIC delivery error (%lx).\n", accept_status); return (send_status | accept_status); } @@ -637,6 +715,7 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) Dprintk("Waiting for send to finish...\n"); send_status = safe_apic_wait_icr_idle(); + mb(); atomic_set(&init_deasserted, 1); /* @@ -655,7 +734,11 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) * target processor state. */ startup_ipi_hook(phys_apicid, (unsigned long) start_secondary, - (unsigned long) stack_start.sp); +#ifdef CONFIG_X86_64 + (unsigned long)init_rsp); +#else + (unsigned long)stack_start.sp); +#endif /* * Run STARTUP IPI loop. @@ -665,7 +748,7 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) maxlvt = lapic_get_maxlvt(); for (j = 1; j <= num_starts; j++) { - Dprintk("Sending STARTUP #%d.\n",j); + Dprintk("Sending STARTUP #%d.\n", j); apic_read_around(APIC_SPIV); apic_write(APIC_ESR, 0); apic_read(APIC_ESR); @@ -711,49 +794,29 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) Dprintk("After Startup.\n"); if (send_status) - printk("APIC never delivered???\n"); + printk(KERN_ERR "APIC never delivered???\n"); if (accept_status) - printk("APIC delivery error (%lx).\n", accept_status); + printk(KERN_ERR "APIC delivery error (%lx).\n", accept_status); return (send_status | accept_status); } #endif /* WAKE_SECONDARY_VIA_INIT */ -extern cpumask_t cpu_initialized; -static inline int alloc_cpu_id(void) -{ - cpumask_t tmp_map; +struct create_idle { + struct work_struct work; + struct task_struct *idle; + struct completion done; int cpu; - cpus_complement(tmp_map, cpu_present_map); - cpu = first_cpu(tmp_map); - if (cpu >= NR_CPUS) - return -ENODEV; - return cpu; -} +}; -#ifdef CONFIG_HOTPLUG_CPU -static struct task_struct * __cpuinitdata cpu_idle_tasks[NR_CPUS]; -static inline struct task_struct * __cpuinit alloc_idle_task(int cpu) +static void __cpuinit do_fork_idle(struct work_struct *work) { - struct task_struct *idle; + struct create_idle *c_idle = + container_of(work, struct create_idle, work); - if ((idle = cpu_idle_tasks[cpu]) != NULL) { - /* initialize thread_struct. we really want to avoid destroy - * idle tread - */ - idle->thread.sp = (unsigned long)task_pt_regs(idle); - init_idle(idle, cpu); - return idle; - } - idle = fork_idle(cpu); - - if (!IS_ERR(idle)) - cpu_idle_tasks[cpu] = idle; - return idle; + c_idle->idle = fork_idle(c_idle->cpu); + complete(&c_idle->done); } -#else -#define alloc_idle_task(cpu) fork_idle(cpu) -#endif static int __cpuinit do_boot_cpu(int apicid, int cpu) /* @@ -762,45 +825,92 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu) * Returns zero if CPU booted OK, else error code from wakeup_secondary_cpu. */ { - struct task_struct *idle; - unsigned long boot_error; + unsigned long boot_error = 0; int timeout; - unsigned long start_eip; + unsigned long start_ip; unsigned short nmi_high = 0, nmi_low = 0; + struct create_idle c_idle = { + .cpu = cpu, + .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done), + }; + INIT_WORK(&c_idle.work, do_fork_idle); +#ifdef CONFIG_X86_64 + /* allocate memory for gdts of secondary cpus. Hotplug is considered */ + if (!cpu_gdt_descr[cpu].address && + !(cpu_gdt_descr[cpu].address = get_zeroed_page(GFP_KERNEL))) { + printk(KERN_ERR "Failed to allocate GDT for CPU %d\n", cpu); + return -1; + } - /* - * Save current MTRR state in case it was changed since early boot - * (e.g. by the ACPI SMI) to initialize new CPUs with MTRRs in sync: - */ - mtrr_save_state(); + /* Allocate node local memory for AP pdas */ + if (cpu_pda(cpu) == &boot_cpu_pda[cpu]) { + struct x8664_pda *newpda, *pda; + int node = cpu_to_node(cpu); + pda = cpu_pda(cpu); + newpda = kmalloc_node(sizeof(struct x8664_pda), GFP_ATOMIC, + node); + if (newpda) { + memcpy(newpda, pda, sizeof(struct x8664_pda)); + cpu_pda(cpu) = newpda; + } else + printk(KERN_ERR + "Could not allocate node local PDA for CPU %d on node %d\n", + cpu, node); + } +#endif + + alternatives_smp_switch(1); + + c_idle.idle = get_idle_for_cpu(cpu); /* * We can't use kernel_thread since we must avoid to * reschedule the child. */ - idle = alloc_idle_task(cpu); - if (IS_ERR(idle)) - panic("failed fork for CPU %d", cpu); + if (c_idle.idle) { + c_idle.idle->thread.sp = (unsigned long) (((struct pt_regs *) + (THREAD_SIZE + task_stack_page(c_idle.idle))) - 1); + init_idle(c_idle.idle, cpu); + goto do_rest; + } + if (!keventd_up() || current_is_keventd()) + c_idle.work.func(&c_idle.work); + else { + schedule_work(&c_idle.work); + wait_for_completion(&c_idle.done); + } + + if (IS_ERR(c_idle.idle)) { + printk("failed fork for CPU %d\n", cpu); + return PTR_ERR(c_idle.idle); + } + + set_idle_for_cpu(cpu, c_idle.idle); +do_rest: +#ifdef CONFIG_X86_32 + per_cpu(current_task, cpu) = c_idle.idle; init_gdt(cpu); - per_cpu(current_task, cpu) = idle; early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu); + c_idle.idle->thread.ip = (unsigned long) start_secondary; + /* Stack for startup_32 can be just as for start_secondary onwards */ + stack_start.sp = (void *) c_idle.idle->thread.sp; + irq_ctx_init(cpu); +#else + cpu_pda(cpu)->pcurrent = c_idle.idle; + init_rsp = c_idle.idle->thread.sp; + load_sp0(&per_cpu(init_tss, cpu), &c_idle.idle->thread); + initial_code = (unsigned long)start_secondary; + clear_tsk_thread_flag(c_idle.idle, TIF_FORK); +#endif - idle->thread.ip = (unsigned long) start_secondary; - /* start_eip had better be page-aligned! */ - start_eip = setup_trampoline(); - - ++cpucount; - alternatives_smp_switch(1); + /* start_ip had better be page-aligned! */ + start_ip = setup_trampoline(); /* So we see what's up */ - printk("Booting processor %d/%d ip %lx\n", cpu, apicid, start_eip); - /* Stack for startup_32 can be just as for start_secondary onwards */ - stack_start.sp = (void *) idle->thread.sp; + printk(KERN_INFO "Booting processor %d/%d ip %lx\n", + cpu, apicid, start_ip); - irq_ctx_init(cpu); - - per_cpu(x86_cpu_to_apicid, cpu) = apicid; /* * This grunge runs the startup process for * the targeted processor. @@ -812,12 +922,17 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu) store_NMI_vector(&nmi_high, &nmi_low); - smpboot_setup_warm_reset_vector(start_eip); + smpboot_setup_warm_reset_vector(start_ip); + /* + * Be paranoid about clearing APIC errors. + */ + apic_write(APIC_ESR, 0); + apic_read(APIC_ESR); /* * Starting actual IPI sequence... */ - boot_error = wakeup_secondary_cpu(apicid, start_eip); + boot_error = wakeup_secondary_cpu(apicid, start_ip); if (!boot_error) { /* @@ -839,18 +954,18 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu) if (cpu_isset(cpu, cpu_callin_map)) { /* number CPUs logically, starting from 1 (BSP is 0) */ Dprintk("OK.\n"); - printk("CPU%d: ", cpu); + printk(KERN_INFO "CPU%d: ", cpu); print_cpu_info(&cpu_data(cpu)); Dprintk("CPU has booted.\n"); } else { - boot_error= 1; + boot_error = 1; if (*((volatile unsigned char *)trampoline_base) == 0xA5) /* trampoline started but...? */ - printk("Stuck ??\n"); + printk(KERN_ERR "Stuck ??\n"); else /* trampoline code not run */ - printk("Not responding.\n"); + printk(KERN_ERR "Not responding.\n"); inquire_remote_apic(apicid); } } @@ -858,156 +973,159 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu) if (boot_error) { /* Try to put things back the way they were before ... */ unmap_cpu_to_logical_apicid(cpu); - cpu_clear(cpu, cpu_callout_map); /* was set here (do_boot_cpu()) */ +#ifdef CONFIG_X86_64 + clear_node_cpumask(cpu); /* was set by numa_add_cpu */ +#endif + cpu_clear(cpu, cpu_callout_map); /* was set by do_boot_cpu() */ cpu_clear(cpu, cpu_initialized); /* was set by cpu_init() */ - cpucount--; - } else { - per_cpu(x86_cpu_to_apicid, cpu) = apicid; - cpu_set(cpu, cpu_present_map); + cpu_clear(cpu, cpu_possible_map); + cpu_clear(cpu, cpu_present_map); + per_cpu(x86_cpu_to_apicid, cpu) = BAD_APICID; } /* mark "stuck" area as not stuck */ *((volatile unsigned long *)trampoline_base) = 0; + /* + * Cleanup possible dangling ends... + */ + smpboot_restore_warm_reset_vector(); + return boot_error; } -#ifdef CONFIG_HOTPLUG_CPU -void cpu_exit_clear(void) +int __cpuinit native_cpu_up(unsigned int cpu) { - int cpu = raw_smp_processor_id(); - - idle_task_exit(); - - cpucount --; - cpu_uninit(); - irq_ctx_exit(cpu); - - cpu_clear(cpu, cpu_callout_map); - cpu_clear(cpu, cpu_callin_map); + int apicid = cpu_present_to_apicid(cpu); + unsigned long flags; + int err; - cpu_clear(cpu, smp_commenced_mask); - unmap_cpu_to_logical_apicid(cpu); -} + WARN_ON(irqs_disabled()); -struct warm_boot_cpu_info { - struct completion *complete; - struct work_struct task; - int apicid; - int cpu; -}; + Dprintk("++++++++++++++++++++=_---CPU UP %u\n", cpu); -static void __cpuinit do_warm_boot_cpu(struct work_struct *work) -{ - struct warm_boot_cpu_info *info = - container_of(work, struct warm_boot_cpu_info, task); - do_boot_cpu(info->apicid, info->cpu); - complete(info->complete); -} + if (apicid == BAD_APICID || apicid == boot_cpu_physical_apicid || + !physid_isset(apicid, phys_cpu_present_map)) { + printk(KERN_ERR "%s: bad cpu %d\n", __func__, cpu); + return -EINVAL; + } -static int __cpuinit __smp_prepare_cpu(int cpu) -{ - DECLARE_COMPLETION_ONSTACK(done); - struct warm_boot_cpu_info info; - int apicid, ret; - - apicid = per_cpu(x86_cpu_to_apicid, cpu); - if (apicid == BAD_APICID) { - ret = -ENODEV; - goto exit; + /* + * Already booted CPU? + */ + if (cpu_isset(cpu, cpu_callin_map)) { + Dprintk("do_boot_cpu %d Already started\n", cpu); + return -ENOSYS; } - info.complete = &done; - info.apicid = apicid; - info.cpu = cpu; - INIT_WORK(&info.task, do_warm_boot_cpu); + /* + * Save current MTRR state in case it was changed since early boot + * (e.g. by the ACPI SMI) to initialize new CPUs with MTRRs in sync: + */ + mtrr_save_state(); + + per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; +#ifdef CONFIG_X86_32 /* init low mem mapping */ clone_pgd_range(swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS, min_t(unsigned long, KERNEL_PGD_PTRS, USER_PGD_PTRS)); flush_tlb_all(); - schedule_work(&info.task); - wait_for_completion(&done); - - zap_low_mappings(); - ret = 0; -exit: - return ret; -} -#endif - -/* - * Cycle through the processors sending APIC IPIs to boot each. - */ - -static int boot_cpu_logical_apicid; -/* Where the IO area was mapped on multiquad, always 0 otherwise */ -void *xquad_portio; -#ifdef CONFIG_X86_NUMAQ -EXPORT_SYMBOL(xquad_portio); #endif -static void __init smp_boot_cpus(unsigned int max_cpus) -{ - int apicid, cpu, bit, kicked; - unsigned long bogosum = 0; + err = do_boot_cpu(apicid, cpu); + if (err < 0) { + Dprintk("do_boot_cpu failed %d\n", err); + return err; + } /* - * Setup boot CPU information + * Check TSC synchronization with the AP (keep irqs disabled + * while doing so): */ - smp_store_cpu_info(0); /* Final full version of the data */ - printk("CPU%d: ", 0); - print_cpu_info(&cpu_data(0)); + local_irq_save(flags); + check_tsc_sync_source(cpu); + local_irq_restore(flags); - boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID)); - boot_cpu_logical_apicid = logical_smp_processor_id(); - per_cpu(x86_cpu_to_apicid, 0) = boot_cpu_physical_apicid; + while (!cpu_isset(cpu, cpu_online_map)) { + cpu_relax(); + touch_nmi_watchdog(); + } - current_thread_info()->cpu = 0; + return 0; +} - set_cpu_sibling_map(0); +/* + * Fall back to non SMP mode after errors. + * + * RED-PEN audit/test this more. I bet there is more state messed up here. + */ +static __init void disable_smp(void) +{ + cpu_present_map = cpumask_of_cpu(0); + cpu_possible_map = cpumask_of_cpu(0); +#ifdef CONFIG_X86_32 + smpboot_clear_io_apic_irqs(); +#endif + if (smp_found_config) + phys_cpu_present_map = + physid_mask_of_physid(boot_cpu_physical_apicid); + else + phys_cpu_present_map = physid_mask_of_physid(0); + map_cpu_to_logical_apicid(); + cpu_set(0, per_cpu(cpu_sibling_map, 0)); + cpu_set(0, per_cpu(cpu_core_map, 0)); +} + +/* + * Various sanity checks. + */ +static int __init smp_sanity_check(unsigned max_cpus) +{ + preempt_disable(); + if (!physid_isset(hard_smp_processor_id(), phys_cpu_present_map)) { + printk(KERN_WARNING "weird, boot CPU (#%d) not listed" + "by the BIOS.\n", hard_smp_processor_id()); + physid_set(hard_smp_processor_id(), phys_cpu_present_map); + } /* * If we couldn't find an SMP configuration at boot time, * get out of here now! */ if (!smp_found_config && !acpi_lapic) { + preempt_enable(); printk(KERN_NOTICE "SMP motherboard not detected.\n"); - smpboot_clear_io_apic_irqs(); - phys_cpu_present_map = physid_mask_of_physid(0); + disable_smp(); if (APIC_init_uniprocessor()) printk(KERN_NOTICE "Local APIC not detected." " Using dummy APIC emulation.\n"); - map_cpu_to_logical_apicid(); - cpu_set(0, per_cpu(cpu_sibling_map, 0)); - cpu_set(0, per_cpu(cpu_core_map, 0)); - return; + return -1; } /* * Should not be necessary because the MP table should list the boot * CPU too, but we do it for the sake of robustness anyway. - * Makes no sense to do this check in clustered apic mode, so skip it */ if (!check_phys_apicid_present(boot_cpu_physical_apicid)) { - printk("weird, boot CPU (#%d) not listed by the BIOS.\n", - boot_cpu_physical_apicid); + printk(KERN_NOTICE + "weird, boot CPU (#%d) not listed by the BIOS.\n", + boot_cpu_physical_apicid); physid_set(hard_smp_processor_id(), phys_cpu_present_map); } + preempt_enable(); /* * If we couldn't find a local APIC, then get out of here now! */ - if (APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid]) && !cpu_has_apic) { + if (APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid]) && + !cpu_has_apic) { printk(KERN_ERR "BIOS bug, local APIC #%d not detected!...\n", boot_cpu_physical_apicid); - printk(KERN_ERR "... forcing use of dummy APIC emulation. (tell your hw vendor)\n"); - smpboot_clear_io_apic_irqs(); - phys_cpu_present_map = physid_mask_of_physid(0); - map_cpu_to_logical_apicid(); - cpu_set(0, per_cpu(cpu_sibling_map, 0)); - cpu_set(0, per_cpu(cpu_core_map, 0)); - return; + printk(KERN_ERR "... forcing use of dummy APIC emulation." + "(tell your hw vendor)\n"); + smpboot_clear_io_apic(); + return -1; } verify_local_APIC(); @@ -1016,137 +1134,148 @@ static void __init smp_boot_cpus(unsigned int max_cpus) * If SMP should be disabled, then really disable it! */ if (!max_cpus) { - smp_found_config = 0; - printk(KERN_INFO "SMP mode deactivated, forcing use of dummy APIC emulation.\n"); - + printk(KERN_INFO "SMP mode deactivated," + "forcing use of dummy APIC emulation.\n"); + smpboot_clear_io_apic(); +#ifdef CONFIG_X86_32 if (nmi_watchdog == NMI_LOCAL_APIC) { - printk(KERN_INFO "activating minimal APIC for NMI watchdog use.\n"); + printk(KERN_INFO "activating minimal APIC for" + "NMI watchdog use.\n"); connect_bsp_APIC(); setup_local_APIC(); + end_local_APIC_setup(); } - smpboot_clear_io_apic_irqs(); - phys_cpu_present_map = physid_mask_of_physid(0); - map_cpu_to_logical_apicid(); - cpu_set(0, per_cpu(cpu_sibling_map, 0)); - cpu_set(0, per_cpu(cpu_core_map, 0)); - return; +#endif + return -1; } - connect_bsp_APIC(); - setup_local_APIC(); - map_cpu_to_logical_apicid(); + return 0; +} +static void __init smp_cpu_index_default(void) +{ + int i; + struct cpuinfo_x86 *c; - setup_portio_remap(); + for_each_cpu_mask(i, cpu_possible_map) { + c = &cpu_data(i); + /* mark all to hotplug */ + c->cpu_index = NR_CPUS; + } +} +/* + * Prepare for SMP bootup. The MP table or ACPI has been read + * earlier. Just do some sanity checking here and enable APIC mode. + */ +void __init native_smp_prepare_cpus(unsigned int max_cpus) +{ + nmi_watchdog_default(); + smp_cpu_index_default(); + current_cpu_data = boot_cpu_data; + cpu_callin_map = cpumask_of_cpu(0); + mb(); /* - * Scan the CPU present map and fire up the other CPUs via do_boot_cpu - * - * In clustered apic mode, phys_cpu_present_map is a constructed thus: - * bits 0-3 are quad0, 4-7 are quad1, etc. A perverse twist on the - * clustered apic ID. + * Setup boot CPU information */ - Dprintk("CPU present map: %lx\n", physids_coerce(phys_cpu_present_map)); - - kicked = 1; - for (bit = 0; kicked < NR_CPUS && bit < MAX_APICS; bit++) { - apicid = cpu_present_to_apicid(bit); - /* - * Don't even attempt to start the boot CPU! - */ - if ((apicid == boot_cpu_apicid) || (apicid == BAD_APICID)) - continue; + smp_store_cpu_info(0); /* Final full version of the data */ + boot_cpu_logical_apicid = logical_smp_processor_id(); + current_thread_info()->cpu = 0; /* needed? */ + set_cpu_sibling_map(0); - if (!check_apicid_present(bit)) - continue; - if (max_cpus <= cpucount+1) - continue; + if (smp_sanity_check(max_cpus) < 0) { + printk(KERN_INFO "SMP disabled\n"); + disable_smp(); + return; + } - if (((cpu = alloc_cpu_id()) <= 0) || do_boot_cpu(apicid, cpu)) - printk("CPU #%d not responding - cannot use it.\n", - apicid); - else - ++kicked; + preempt_disable(); + if (GET_APIC_ID(read_apic_id()) != boot_cpu_physical_apicid) { + panic("Boot APIC ID in local APIC unexpected (%d vs %d)", + GET_APIC_ID(read_apic_id()), boot_cpu_physical_apicid); + /* Or can we switch back to PIC here? */ } + preempt_enable(); +#ifdef CONFIG_X86_32 + connect_bsp_APIC(); +#endif /* - * Cleanup possible dangling ends... + * Switch from PIC to APIC mode. */ - smpboot_restore_warm_reset_vector(); + setup_local_APIC(); +#ifdef CONFIG_X86_64 /* - * Allow the user to impress friends. + * Enable IO APIC before setting up error vector */ - Dprintk("Before bogomips.\n"); - for_each_possible_cpu(cpu) - if (cpu_isset(cpu, cpu_callout_map)) - bogosum += cpu_data(cpu).loops_per_jiffy; - printk(KERN_INFO - "Total of %d processors activated (%lu.%02lu BogoMIPS).\n", - cpucount+1, - bogosum/(500000/HZ), - (bogosum/(5000/HZ))%100); - - Dprintk("Before bogocount - setting activated=1.\n"); - - if (smp_b_stepping) - printk(KERN_WARNING "WARNING: SMP operation may be unreliable with B stepping processors.\n"); + if (!skip_ioapic_setup && nr_ioapics) + enable_IO_APIC(); +#endif + end_local_APIC_setup(); - /* - * Don't taint if we are running SMP kernel on a single non-MP - * approved Athlon - */ - if (tainted & TAINT_UNSAFE_SMP) { - if (cpucount) - printk (KERN_INFO "WARNING: This combination of AMD processors is not suitable for SMP.\n"); - else - tainted &= ~TAINT_UNSAFE_SMP; - } + map_cpu_to_logical_apicid(); - Dprintk("Boot done.\n"); + setup_portio_remap(); + smpboot_setup_io_apic(); /* - * construct cpu_sibling_map, so that we can tell sibling CPUs - * efficiently. + * Set up local APIC timer on boot CPU. */ - for_each_possible_cpu(cpu) { - cpus_clear(per_cpu(cpu_sibling_map, cpu)); - cpus_clear(per_cpu(cpu_core_map, cpu)); - } - - cpu_set(0, per_cpu(cpu_sibling_map, 0)); - cpu_set(0, per_cpu(cpu_core_map, 0)); - - smpboot_setup_io_apic(); + printk(KERN_INFO "CPU%d: ", 0); + print_cpu_info(&cpu_data(0)); setup_boot_clock(); } +/* + * Early setup to make printk work. + */ +void __init native_smp_prepare_boot_cpu(void) +{ + int me = smp_processor_id(); +#ifdef CONFIG_X86_32 + init_gdt(me); + switch_to_new_gdt(); +#endif + /* already set me in cpu_online_map in boot_cpu_init() */ + cpu_set(me, cpu_callout_map); + per_cpu(cpu_state, me) = CPU_ONLINE; +} -/* These are wrappers to interface to the new boot process. Someone - who understands all this stuff should rewrite it properly. --RR 15/Jul/02 */ -void __init native_smp_prepare_cpus(unsigned int max_cpus) +void __init native_smp_cpus_done(unsigned int max_cpus) { - smp_commenced_mask = cpumask_of_cpu(0); - cpu_callin_map = cpumask_of_cpu(0); - mb(); - smp_boot_cpus(max_cpus); + Dprintk("Boot done.\n"); + + impress_friends(); + smp_checks(); +#ifdef CONFIG_X86_IO_APIC + setup_ioapic_dest(); +#endif + check_nmi_watchdog(); +#ifdef CONFIG_X86_32 + zap_low_mappings(); +#endif } -void __init native_smp_prepare_boot_cpu(void) +#ifdef CONFIG_HOTPLUG_CPU + +# ifdef CONFIG_X86_32 +void cpu_exit_clear(void) { - unsigned int cpu = smp_processor_id(); + int cpu = raw_smp_processor_id(); - init_gdt(cpu); - switch_to_new_gdt(); + idle_task_exit(); + + cpu_uninit(); + irq_ctx_exit(cpu); + + cpu_clear(cpu, cpu_callout_map); + cpu_clear(cpu, cpu_callin_map); - cpu_set(cpu, cpu_online_map); - cpu_set(cpu, cpu_callout_map); - cpu_set(cpu, cpu_present_map); - cpu_set(cpu, cpu_possible_map); - __get_cpu_var(cpu_state) = CPU_ONLINE; + unmap_cpu_to_logical_apicid(cpu); } +# endif /* CONFIG_X86_32 */ -#ifdef CONFIG_HOTPLUG_CPU void remove_siblinginfo(int cpu) { int sibling; @@ -1160,7 +1289,7 @@ void remove_siblinginfo(int cpu) if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1) cpu_data(sibling).booted_cores--; } - + for_each_cpu_mask(sibling, per_cpu(cpu_sibling_map, cpu)) cpu_clear(cpu, per_cpu(cpu_sibling_map, sibling)); cpus_clear(per_cpu(cpu_sibling_map, cpu)); @@ -1170,35 +1299,99 @@ void remove_siblinginfo(int cpu) cpu_clear(cpu, cpu_sibling_setup_map); } +int additional_cpus __initdata = -1; + +static __init int setup_additional_cpus(char *s) +{ + return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL; +} +early_param("additional_cpus", setup_additional_cpus); + +/* + * cpu_possible_map should be static, it cannot change as cpu's + * are onlined, or offlined. The reason is per-cpu data-structures + * are allocated by some modules at init time, and dont expect to + * do this dynamically on cpu arrival/departure. + * cpu_present_map on the other hand can change dynamically. + * In case when cpu_hotplug is not compiled, then we resort to current + * behaviour, which is cpu_possible == cpu_present. + * - Ashok Raj + * + * Three ways to find out the number of additional hotplug CPUs: + * - If the BIOS specified disabled CPUs in ACPI/mptables use that. + * - The user can overwrite it with additional_cpus=NUM + * - Otherwise don't reserve additional CPUs. + * We do this because additional CPUs waste a lot of memory. + * -AK + */ +__init void prefill_possible_map(void) +{ + int i; + int possible; + + if (additional_cpus == -1) { + if (disabled_cpus > 0) + additional_cpus = disabled_cpus; + else + additional_cpus = 0; + } + possible = num_processors + additional_cpus; + if (possible > NR_CPUS) + possible = NR_CPUS; + + printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n", + possible, max_t(int, possible - num_processors, 0)); + + for (i = 0; i < possible; i++) + cpu_set(i, cpu_possible_map); +} + +static void __ref remove_cpu_from_maps(int cpu) +{ + cpu_clear(cpu, cpu_online_map); +#ifdef CONFIG_X86_64 + cpu_clear(cpu, cpu_callout_map); + cpu_clear(cpu, cpu_callin_map); + /* was set by cpu_init() */ + clear_bit(cpu, (unsigned long *)&cpu_initialized); + clear_node_cpumask(cpu); +#endif +} + int __cpu_disable(void) { - cpumask_t map = cpu_online_map; int cpu = smp_processor_id(); /* * Perhaps use cpufreq to drop frequency, but that could go * into generic code. - * + * * We won't take down the boot processor on i386 due to some * interrupts only being able to be serviced by the BSP. * Especially so if we're not using an IOAPIC -zwane */ if (cpu == 0) return -EBUSY; + if (nmi_watchdog == NMI_LOCAL_APIC) stop_apic_nmi_watchdog(NULL); clear_local_APIC(); - /* Allow any queued timer interrupts to get serviced */ + + /* + * HACK: + * Allow any queued timer interrupts to get serviced + * This is only a temporary solution until we cleanup + * fixup_irqs as we do for IA64. + */ local_irq_enable(); mdelay(1); - local_irq_disable(); + local_irq_disable(); remove_siblinginfo(cpu); - cpu_clear(cpu, map); - fixup_irqs(map); /* It's now safe to remove this processor from the online map */ - cpu_clear(cpu, cpu_online_map); + remove_cpu_from_maps(cpu); + fixup_irqs(cpu_online_map); return 0; } @@ -1210,14 +1403,14 @@ void __cpu_die(unsigned int cpu) for (i = 0; i < 10; i++) { /* They ack this in play_dead by setting CPU_DEAD */ if (per_cpu(cpu_state, cpu) == CPU_DEAD) { - printk ("CPU %d is now offline\n", cpu); + printk(KERN_INFO "CPU %d is now offline\n", cpu); if (1 == num_online_cpus()) alternatives_smp_switch(0); return; } msleep(100); } - printk(KERN_ERR "CPU %u didn't die...\n", cpu); + printk(KERN_ERR "CPU %u didn't die...\n", cpu); } #else /* ... !CONFIG_HOTPLUG_CPU */ int __cpu_disable(void) @@ -1230,81 +1423,7 @@ void __cpu_die(unsigned int cpu) /* We said "no" in __cpu_disable */ BUG(); } -#endif /* CONFIG_HOTPLUG_CPU */ - -int __cpuinit native_cpu_up(unsigned int cpu) -{ - unsigned long flags; -#ifdef CONFIG_HOTPLUG_CPU - int ret = 0; - - /* - * We do warm boot only on cpus that had booted earlier - * Otherwise cold boot is all handled from smp_boot_cpus(). - * cpu_callin_map is set during AP kickstart process. Its reset - * when a cpu is taken offline from cpu_exit_clear(). - */ - if (!cpu_isset(cpu, cpu_callin_map)) - ret = __smp_prepare_cpu(cpu); - - if (ret) - return -EIO; -#endif - - /* In case one didn't come up */ - if (!cpu_isset(cpu, cpu_callin_map)) { - printk(KERN_DEBUG "skipping cpu%d, didn't come online\n", cpu); - return -EIO; - } - - per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; - /* Unleash the CPU! */ - cpu_set(cpu, smp_commenced_mask); - - /* - * Check TSC synchronization with the AP (keep irqs disabled - * while doing so): - */ - local_irq_save(flags); - check_tsc_sync_source(cpu); - local_irq_restore(flags); - - while (!cpu_isset(cpu, cpu_online_map)) { - cpu_relax(); - touch_nmi_watchdog(); - } - - return 0; -} - -void __init native_smp_cpus_done(unsigned int max_cpus) -{ -#ifdef CONFIG_X86_IO_APIC - setup_ioapic_dest(); #endif - zap_low_mappings(); -} - -void __init smp_intr_init(void) -{ - /* - * IRQ0 must be given a fixed assignment and initialized, - * because it's used before the IO-APIC is set up. - */ - set_intr_gate(FIRST_DEVICE_VECTOR, interrupt[0]); - - /* - * The reschedule interrupt is a CPU-to-CPU reschedule-helper - * IPI, driven by wakeup. - */ - set_intr_gate(RESCHEDULE_VECTOR, reschedule_interrupt); - - /* IPI for invalidation */ - set_intr_gate(INVALIDATE_TLB_VECTOR, invalidate_interrupt); - - /* IPI for generic function call */ - set_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt); -} /* * If the BIOS enumerates physical processors before logical, diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c deleted file mode 100644 index 0880f2c388a..00000000000 --- a/arch/x86/kernel/smpboot_64.c +++ /dev/null @@ -1,1108 +0,0 @@ -/* - * x86 SMP booting functions - * - * (c) 1995 Alan Cox, Building #3 <alan@redhat.com> - * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com> - * Copyright 2001 Andi Kleen, SuSE Labs. - * - * Much of the core SMP work is based on previous work by Thomas Radke, to - * whom a great many thanks are extended. - * - * Thanks to Intel for making available several different Pentium, - * Pentium Pro and Pentium-II/Xeon MP machines. - * Original development of Linux SMP code supported by Caldera. - * - * This code is released under the GNU General Public License version 2 - * - * Fixes - * Felix Koop : NR_CPUS used properly - * Jose Renau : Handle single CPU case. - * Alan Cox : By repeated request 8) - Total BogoMIP report. - * Greg Wright : Fix for kernel stacks panic. - * Erich Boleyn : MP v1.4 and additional changes. - * Matthias Sattler : Changes for 2.1 kernel map. - * Michel Lespinasse : Changes for 2.1 kernel map. - * Michael Chastain : Change trampoline.S to gnu as. - * Alan Cox : Dumb bug: 'B' step PPro's are fine - * Ingo Molnar : Added APIC timers, based on code - * from Jose Renau - * Ingo Molnar : various cleanups and rewrites - * Tigran Aivazian : fixed "0.00 in /proc/uptime on SMP" bug. - * Maciej W. Rozycki : Bits for genuine 82489DX APICs - * Andi Kleen : Changed for SMP boot into long mode. - * Rusty Russell : Hacked into shape for new "hotplug" boot process. - * Andi Kleen : Converted to new state machine. - * Various cleanups. - * Probably mostly hotplug CPU ready now. - * Ashok Raj : CPU hotplug support - */ - - -#include <linux/init.h> - -#include <linux/mm.h> -#include <linux/kernel_stat.h> -#include <linux/bootmem.h> -#include <linux/thread_info.h> -#include <linux/module.h> -#include <linux/delay.h> -#include <linux/mc146818rtc.h> -#include <linux/smp.h> -#include <linux/kdebug.h> - -#include <asm/mtrr.h> -#include <asm/pgalloc.h> -#include <asm/desc.h> -#include <asm/tlbflush.h> -#include <asm/proto.h> -#include <asm/nmi.h> -#include <asm/irq.h> -#include <asm/hw_irq.h> -#include <asm/numa.h> - -/* Number of siblings per CPU package */ -int smp_num_siblings = 1; -EXPORT_SYMBOL(smp_num_siblings); - -/* Last level cache ID of each logical CPU */ -DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID; - -/* Bitmask of currently online CPUs */ -cpumask_t cpu_online_map __read_mostly; - -EXPORT_SYMBOL(cpu_online_map); - -/* - * Private maps to synchronize booting between AP and BP. - * Probably not needed anymore, but it makes for easier debugging. -AK - */ -cpumask_t cpu_callin_map; -cpumask_t cpu_callout_map; -cpumask_t cpu_possible_map; -EXPORT_SYMBOL(cpu_possible_map); - -/* Per CPU bogomips and other parameters */ -DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info); -EXPORT_PER_CPU_SYMBOL(cpu_info); - -/* Set when the idlers are all forked */ -int smp_threads_ready; - -/* representing HT siblings of each logical CPU */ -DEFINE_PER_CPU(cpumask_t, cpu_sibling_map); -EXPORT_PER_CPU_SYMBOL(cpu_sibling_map); - -/* representing HT and core siblings of each logical CPU */ -DEFINE_PER_CPU(cpumask_t, cpu_core_map); -EXPORT_PER_CPU_SYMBOL(cpu_core_map); - -/* - * Trampoline 80x86 program as an array. - */ - -extern const unsigned char trampoline_data[]; -extern const unsigned char trampoline_end[]; - -/* State of each CPU */ -DEFINE_PER_CPU(int, cpu_state) = { 0 }; - -/* - * Store all idle threads, this can be reused instead of creating - * a new thread. Also avoids complicated thread destroy functionality - * for idle threads. - */ -#ifdef CONFIG_HOTPLUG_CPU -/* - * Needed only for CONFIG_HOTPLUG_CPU because __cpuinitdata is - * removed after init for !CONFIG_HOTPLUG_CPU. - */ -static DEFINE_PER_CPU(struct task_struct *, idle_thread_array); -#define get_idle_for_cpu(x) (per_cpu(idle_thread_array, x)) -#define set_idle_for_cpu(x,p) (per_cpu(idle_thread_array, x) = (p)) -#else -struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ; -#define get_idle_for_cpu(x) (idle_thread_array[(x)]) -#define set_idle_for_cpu(x,p) (idle_thread_array[(x)] = (p)) -#endif - - -/* - * Currently trivial. Write the real->protected mode - * bootstrap into the page concerned. The caller - * has made sure it's suitably aligned. - */ - -static unsigned long __cpuinit setup_trampoline(void) -{ - void *tramp = __va(SMP_TRAMPOLINE_BASE); - memcpy(tramp, trampoline_data, trampoline_end - trampoline_data); - return virt_to_phys(tramp); -} - -/* - * The bootstrap kernel entry code has set these up. Save them for - * a given CPU - */ - -static void __cpuinit smp_store_cpu_info(int id) -{ - struct cpuinfo_x86 *c = &cpu_data(id); - - *c = boot_cpu_data; - c->cpu_index = id; - identify_cpu(c); - print_cpu_info(c); -} - -static atomic_t init_deasserted __cpuinitdata; - -/* - * Report back to the Boot Processor. - * Running on AP. - */ -void __cpuinit smp_callin(void) -{ - int cpuid, phys_id; - unsigned long timeout; - - /* - * If waken up by an INIT in an 82489DX configuration - * we may get here before an INIT-deassert IPI reaches - * our local APIC. We have to wait for the IPI or we'll - * lock up on an APIC access. - */ - while (!atomic_read(&init_deasserted)) - cpu_relax(); - - /* - * (This works even if the APIC is not enabled.) - */ - phys_id = GET_APIC_ID(apic_read(APIC_ID)); - cpuid = smp_processor_id(); - if (cpu_isset(cpuid, cpu_callin_map)) { - panic("smp_callin: phys CPU#%d, CPU#%d already present??\n", - phys_id, cpuid); - } - Dprintk("CPU#%d (phys ID: %d) waiting for CALLOUT\n", cpuid, phys_id); - - /* - * STARTUP IPIs are fragile beasts as they might sometimes - * trigger some glue motherboard logic. Complete APIC bus - * silence for 1 second, this overestimates the time the - * boot CPU is spending to send the up to 2 STARTUP IPIs - * by a factor of two. This should be enough. - */ - - /* - * Waiting 2s total for startup (udelay is not yet working) - */ - timeout = jiffies + 2*HZ; - while (time_before(jiffies, timeout)) { - /* - * Has the boot CPU finished it's STARTUP sequence? - */ - if (cpu_isset(cpuid, cpu_callout_map)) - break; - cpu_relax(); - } - - if (!time_before(jiffies, timeout)) { - panic("smp_callin: CPU%d started up but did not get a callout!\n", - cpuid); - } - - /* - * the boot CPU has finished the init stage and is spinning - * on callin_map until we finish. We are free to set up this - * CPU, first the APIC. (this is probably redundant on most - * boards) - */ - - Dprintk("CALLIN, before setup_local_APIC().\n"); - setup_local_APIC(); - end_local_APIC_setup(); - - /* - * Get our bogomips. - * - * Need to enable IRQs because it can take longer and then - * the NMI watchdog might kill us. - */ - local_irq_enable(); - calibrate_delay(); - local_irq_disable(); - Dprintk("Stack at about %p\n",&cpuid); - - /* - * Save our processor parameters - */ - smp_store_cpu_info(cpuid); - - /* - * Allow the master to continue. - */ - cpu_set(cpuid, cpu_callin_map); -} - -/* maps the cpu to the sched domain representing multi-core */ -cpumask_t cpu_coregroup_map(int cpu) -{ - struct cpuinfo_x86 *c = &cpu_data(cpu); - /* - * For perf, we return last level cache shared map. - * And for power savings, we return cpu_core_map - */ - if (sched_mc_power_savings || sched_smt_power_savings) - return per_cpu(cpu_core_map, cpu); - else - return c->llc_shared_map; -} - -/* representing cpus for which sibling maps can be computed */ -static cpumask_t cpu_sibling_setup_map; - -static inline void set_cpu_sibling_map(int cpu) -{ - int i; - struct cpuinfo_x86 *c = &cpu_data(cpu); - - cpu_set(cpu, cpu_sibling_setup_map); - - if (smp_num_siblings > 1) { - for_each_cpu_mask(i, cpu_sibling_setup_map) { - if (c->phys_proc_id == cpu_data(i).phys_proc_id && - c->cpu_core_id == cpu_data(i).cpu_core_id) { - cpu_set(i, per_cpu(cpu_sibling_map, cpu)); - cpu_set(cpu, per_cpu(cpu_sibling_map, i)); - cpu_set(i, per_cpu(cpu_core_map, cpu)); - cpu_set(cpu, per_cpu(cpu_core_map, i)); - cpu_set(i, c->llc_shared_map); - cpu_set(cpu, cpu_data(i).llc_shared_map); - } - } - } else { - cpu_set(cpu, per_cpu(cpu_sibling_map, cpu)); - } - - cpu_set(cpu, c->llc_shared_map); - - if (current_cpu_data.x86_max_cores == 1) { - per_cpu(cpu_core_map, cpu) = per_cpu(cpu_sibling_map, cpu); - c->booted_cores = 1; - return; - } - - for_each_cpu_mask(i, cpu_sibling_setup_map) { - if (per_cpu(cpu_llc_id, cpu) != BAD_APICID && - per_cpu(cpu_llc_id, cpu) == per_cpu(cpu_llc_id, i)) { - cpu_set(i, c->llc_shared_map); - cpu_set(cpu, cpu_data(i).llc_shared_map); - } - if (c->phys_proc_id == cpu_data(i).phys_proc_id) { - cpu_set(i, per_cpu(cpu_core_map, cpu)); - cpu_set(cpu, per_cpu(cpu_core_map, i)); - /* - * Does this new cpu bringup a new core? - */ - if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1) { - /* - * for each core in package, increment - * the booted_cores for this new cpu - */ - if (first_cpu(per_cpu(cpu_sibling_map, i)) == i) - c->booted_cores++; - /* - * increment the core count for all - * the other cpus in this package - */ - if (i != cpu) - cpu_data(i).booted_cores++; - } else if (i != cpu && !c->booted_cores) - c->booted_cores = cpu_data(i).booted_cores; - } - } -} - -/* - * Setup code on secondary processor (after comming out of the trampoline) - */ -void __cpuinit start_secondary(void) -{ - /* - * Dont put anything before smp_callin(), SMP - * booting is too fragile that we want to limit the - * things done here to the most necessary things. - */ - cpu_init(); - preempt_disable(); - smp_callin(); - - /* otherwise gcc will move up the smp_processor_id before the cpu_init */ - barrier(); - - /* - * Check TSC sync first: - */ - check_tsc_sync_target(); - - if (nmi_watchdog == NMI_IO_APIC) { - disable_8259A_irq(0); - enable_NMI_through_LVT0(); - enable_8259A_irq(0); - } - - /* - * The sibling maps must be set before turing the online map on for - * this cpu - */ - set_cpu_sibling_map(smp_processor_id()); - - /* - * We need to hold call_lock, so there is no inconsistency - * between the time smp_call_function() determines number of - * IPI recipients, and the time when the determination is made - * for which cpus receive the IPI in genapic_flat.c. Holding this - * lock helps us to not include this cpu in a currently in progress - * smp_call_function(). - */ - lock_ipi_call_lock(); - spin_lock(&vector_lock); - - /* Setup the per cpu irq handling data structures */ - __setup_vector_irq(smp_processor_id()); - /* - * Allow the master to continue. - */ - cpu_set(smp_processor_id(), cpu_online_map); - per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE; - spin_unlock(&vector_lock); - - unlock_ipi_call_lock(); - - setup_secondary_clock(); - - cpu_idle(); -} - -extern volatile unsigned long init_rsp; -extern void (*initial_code)(void); - -#ifdef APIC_DEBUG -static void inquire_remote_apic(int apicid) -{ - unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 }; - char *names[] = { "ID", "VERSION", "SPIV" }; - int timeout; - u32 status; - - printk(KERN_INFO "Inquiring remote APIC #%d...\n", apicid); - - for (i = 0; i < ARRAY_SIZE(regs); i++) { - printk(KERN_INFO "... APIC #%d %s: ", apicid, names[i]); - - /* - * Wait for idle. - */ - status = safe_apic_wait_icr_idle(); - if (status) - printk(KERN_CONT - "a previous APIC delivery may have failed\n"); - - apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(apicid)); - apic_write(APIC_ICR, APIC_DM_REMRD | regs[i]); - - timeout = 0; - do { - udelay(100); - status = apic_read(APIC_ICR) & APIC_ICR_RR_MASK; - } while (status == APIC_ICR_RR_INPROG && timeout++ < 1000); - - switch (status) { - case APIC_ICR_RR_VALID: - status = apic_read(APIC_RRR); - printk(KERN_CONT "%08x\n", status); - break; - default: - printk(KERN_CONT "failed\n"); - } - } -} -#endif - -/* - * Kick the secondary to wake up. - */ -static int __cpuinit wakeup_secondary_via_INIT(int phys_apicid, unsigned int start_rip) -{ - unsigned long send_status, accept_status = 0; - int maxlvt, num_starts, j; - - Dprintk("Asserting INIT.\n"); - - /* - * Turn INIT on target chip - */ - apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid)); - - /* - * Send IPI - */ - apic_write(APIC_ICR, APIC_INT_LEVELTRIG | APIC_INT_ASSERT - | APIC_DM_INIT); - - Dprintk("Waiting for send to finish...\n"); - send_status = safe_apic_wait_icr_idle(); - - mdelay(10); - - Dprintk("Deasserting INIT.\n"); - - /* Target chip */ - apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid)); - - /* Send IPI */ - apic_write(APIC_ICR, APIC_INT_LEVELTRIG | APIC_DM_INIT); - - Dprintk("Waiting for send to finish...\n"); - send_status = safe_apic_wait_icr_idle(); - - mb(); - atomic_set(&init_deasserted, 1); - - num_starts = 2; - - /* - * Run STARTUP IPI loop. - */ - Dprintk("#startup loops: %d.\n", num_starts); - - maxlvt = lapic_get_maxlvt(); - - for (j = 1; j <= num_starts; j++) { - Dprintk("Sending STARTUP #%d.\n",j); - apic_write(APIC_ESR, 0); - apic_read(APIC_ESR); - Dprintk("After apic_write.\n"); - - /* - * STARTUP IPI - */ - - /* Target chip */ - apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid)); - - /* Boot on the stack */ - /* Kick the second */ - apic_write(APIC_ICR, APIC_DM_STARTUP | (start_rip >> 12)); - - /* - * Give the other CPU some time to accept the IPI. - */ - udelay(300); - - Dprintk("Startup point 1.\n"); - - Dprintk("Waiting for send to finish...\n"); - send_status = safe_apic_wait_icr_idle(); - - /* - * Give the other CPU some time to accept the IPI. - */ - udelay(200); - /* - * Due to the Pentium erratum 3AP. - */ - if (maxlvt > 3) { - apic_write(APIC_ESR, 0); - } - accept_status = (apic_read(APIC_ESR) & 0xEF); - if (send_status || accept_status) - break; - } - Dprintk("After Startup.\n"); - - if (send_status) - printk(KERN_ERR "APIC never delivered???\n"); - if (accept_status) - printk(KERN_ERR "APIC delivery error (%lx).\n", accept_status); - - return (send_status | accept_status); -} - -struct create_idle { - struct work_struct work; - struct task_struct *idle; - struct completion done; - int cpu; -}; - -static void __cpuinit do_fork_idle(struct work_struct *work) -{ - struct create_idle *c_idle = - container_of(work, struct create_idle, work); - - c_idle->idle = fork_idle(c_idle->cpu); - complete(&c_idle->done); -} - -/* - * Boot one CPU. - */ -static int __cpuinit do_boot_cpu(int cpu, int apicid) -{ - unsigned long boot_error; - int timeout; - unsigned long start_rip; - struct create_idle c_idle = { - .cpu = cpu, - .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done), - }; - INIT_WORK(&c_idle.work, do_fork_idle); - - /* allocate memory for gdts of secondary cpus. Hotplug is considered */ - if (!cpu_gdt_descr[cpu].address && - !(cpu_gdt_descr[cpu].address = get_zeroed_page(GFP_KERNEL))) { - printk(KERN_ERR "Failed to allocate GDT for CPU %d\n", cpu); - return -1; - } - - /* Allocate node local memory for AP pdas */ - if (cpu_pda(cpu) == &boot_cpu_pda[cpu]) { - struct x8664_pda *newpda, *pda; - int node = cpu_to_node(cpu); - pda = cpu_pda(cpu); - newpda = kmalloc_node(sizeof (struct x8664_pda), GFP_ATOMIC, - node); - if (newpda) { - memcpy(newpda, pda, sizeof (struct x8664_pda)); - cpu_pda(cpu) = newpda; - } else - printk(KERN_ERR - "Could not allocate node local PDA for CPU %d on node %d\n", - cpu, node); - } - - alternatives_smp_switch(1); - - c_idle.idle = get_idle_for_cpu(cpu); - - if (c_idle.idle) { - c_idle.idle->thread.sp = (unsigned long) (((struct pt_regs *) - (THREAD_SIZE + task_stack_page(c_idle.idle))) - 1); - init_idle(c_idle.idle, cpu); - goto do_rest; - } - - /* - * During cold boot process, keventd thread is not spun up yet. - * When we do cpu hot-add, we create idle threads on the fly, we should - * not acquire any attributes from the calling context. Hence the clean - * way to create kernel_threads() is to do that from keventd(). - * We do the current_is_keventd() due to the fact that ACPI notifier - * was also queuing to keventd() and when the caller is already running - * in context of keventd(), we would end up with locking up the keventd - * thread. - */ - if (!keventd_up() || current_is_keventd()) - c_idle.work.func(&c_idle.work); - else { - schedule_work(&c_idle.work); - wait_for_completion(&c_idle.done); - } - - if (IS_ERR(c_idle.idle)) { - printk("failed fork for CPU %d\n", cpu); - return PTR_ERR(c_idle.idle); - } - - set_idle_for_cpu(cpu, c_idle.idle); - -do_rest: - - cpu_pda(cpu)->pcurrent = c_idle.idle; - - start_rip = setup_trampoline(); - - init_rsp = c_idle.idle->thread.sp; - load_sp0(&per_cpu(init_tss, cpu), &c_idle.idle->thread); - initial_code = start_secondary; - clear_tsk_thread_flag(c_idle.idle, TIF_FORK); - - printk(KERN_INFO "Booting processor %d/%d APIC 0x%x\n", cpu, - cpus_weight(cpu_present_map), - apicid); - - /* - * This grunge runs the startup process for - * the targeted processor. - */ - - atomic_set(&init_deasserted, 0); - - Dprintk("Setting warm reset code and vector.\n"); - - CMOS_WRITE(0xa, 0xf); - local_flush_tlb(); - Dprintk("1.\n"); - *((volatile unsigned short *) phys_to_virt(0x469)) = start_rip >> 4; - Dprintk("2.\n"); - *((volatile unsigned short *) phys_to_virt(0x467)) = start_rip & 0xf; - Dprintk("3.\n"); - - /* - * Be paranoid about clearing APIC errors. - */ - apic_write(APIC_ESR, 0); - apic_read(APIC_ESR); - - /* - * Status is now clean - */ - boot_error = 0; - - /* - * Starting actual IPI sequence... - */ - boot_error = wakeup_secondary_via_INIT(apicid, start_rip); - - if (!boot_error) { - /* - * allow APs to start initializing. - */ - Dprintk("Before Callout %d.\n", cpu); - cpu_set(cpu, cpu_callout_map); - Dprintk("After Callout %d.\n", cpu); - - /* - * Wait 5s total for a response - */ - for (timeout = 0; timeout < 50000; timeout++) { - if (cpu_isset(cpu, cpu_callin_map)) - break; /* It has booted */ - udelay(100); - } - - if (cpu_isset(cpu, cpu_callin_map)) { - /* number CPUs logically, starting from 1 (BSP is 0) */ - Dprintk("CPU has booted.\n"); - } else { - boot_error = 1; - if (*((volatile unsigned char *)phys_to_virt(SMP_TRAMPOLINE_BASE)) - == 0xA5) - /* trampoline started but...? */ - printk("Stuck ??\n"); - else - /* trampoline code not run */ - printk("Not responding.\n"); -#ifdef APIC_DEBUG - inquire_remote_apic(apicid); -#endif - } - } - if (boot_error) { - cpu_clear(cpu, cpu_callout_map); /* was set here (do_boot_cpu()) */ - clear_bit(cpu, (unsigned long *)&cpu_initialized); /* was set by cpu_init() */ - clear_node_cpumask(cpu); /* was set by numa_add_cpu */ - cpu_clear(cpu, cpu_present_map); - cpu_clear(cpu, cpu_possible_map); - per_cpu(x86_cpu_to_apicid, cpu) = BAD_APICID; - return -EIO; - } - - return 0; -} - -cycles_t cacheflush_time; -unsigned long cache_decay_ticks; - -/* - * Cleanup possible dangling ends... - */ -static __cpuinit void smp_cleanup_boot(void) -{ - /* - * Paranoid: Set warm reset code and vector here back - * to default values. - */ - CMOS_WRITE(0, 0xf); - - /* - * Reset trampoline flag - */ - *((volatile int *) phys_to_virt(0x467)) = 0; -} - -/* - * Fall back to non SMP mode after errors. - * - * RED-PEN audit/test this more. I bet there is more state messed up here. - */ -static __init void disable_smp(void) -{ - cpu_present_map = cpumask_of_cpu(0); - cpu_possible_map = cpumask_of_cpu(0); - if (smp_found_config) - phys_cpu_present_map = physid_mask_of_physid(boot_cpu_id); - else - phys_cpu_present_map = physid_mask_of_physid(0); - cpu_set(0, per_cpu(cpu_sibling_map, 0)); - cpu_set(0, per_cpu(cpu_core_map, 0)); -} - -#ifdef CONFIG_HOTPLUG_CPU - -int additional_cpus __initdata = -1; - -/* - * cpu_possible_map should be static, it cannot change as cpu's - * are onlined, or offlined. The reason is per-cpu data-structures - * are allocated by some modules at init time, and dont expect to - * do this dynamically on cpu arrival/departure. - * cpu_present_map on the other hand can change dynamically. - * In case when cpu_hotplug is not compiled, then we resort to current - * behaviour, which is cpu_possible == cpu_present. - * - Ashok Raj - * - * Three ways to find out the number of additional hotplug CPUs: - * - If the BIOS specified disabled CPUs in ACPI/mptables use that. - * - The user can overwrite it with additional_cpus=NUM - * - Otherwise don't reserve additional CPUs. - * We do this because additional CPUs waste a lot of memory. - * -AK - */ -__init void prefill_possible_map(void) -{ - int i; - int possible; - - if (additional_cpus == -1) { - if (disabled_cpus > 0) - additional_cpus = disabled_cpus; - else - additional_cpus = 0; - } - possible = num_processors + additional_cpus; - if (possible > NR_CPUS) - possible = NR_CPUS; - - printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n", - possible, - max_t(int, possible - num_processors, 0)); - - for (i = 0; i < possible; i++) - cpu_set(i, cpu_possible_map); -} -#endif - -/* - * Various sanity checks. - */ -static int __init smp_sanity_check(unsigned max_cpus) -{ - if (!physid_isset(hard_smp_processor_id(), phys_cpu_present_map)) { - printk("weird, boot CPU (#%d) not listed by the BIOS.\n", - hard_smp_processor_id()); - physid_set(hard_smp_processor_id(), phys_cpu_present_map); - } - - /* - * If we couldn't find an SMP configuration at boot time, - * get out of here now! - */ - if (!smp_found_config) { - printk(KERN_NOTICE "SMP motherboard not detected.\n"); - disable_smp(); - if (APIC_init_uniprocessor()) - printk(KERN_NOTICE "Local APIC not detected." - " Using dummy APIC emulation.\n"); - return -1; - } - - /* - * Should not be necessary because the MP table should list the boot - * CPU too, but we do it for the sake of robustness anyway. - */ - if (!physid_isset(boot_cpu_id, phys_cpu_present_map)) { - printk(KERN_NOTICE "weird, boot CPU (#%d) not listed by the BIOS.\n", - boot_cpu_id); - physid_set(hard_smp_processor_id(), phys_cpu_present_map); - } - - /* - * If we couldn't find a local APIC, then get out of here now! - */ - if (!cpu_has_apic) { - printk(KERN_ERR "BIOS bug, local APIC #%d not detected!...\n", - boot_cpu_id); - printk(KERN_ERR "... forcing use of dummy APIC emulation. (tell your hw vendor)\n"); - nr_ioapics = 0; - return -1; - } - - /* - * If SMP should be disabled, then really disable it! - */ - if (!max_cpus) { - printk(KERN_INFO "SMP mode deactivated, forcing use of dummy APIC emulation.\n"); - nr_ioapics = 0; - return -1; - } - - return 0; -} - -static void __init smp_cpu_index_default(void) -{ - int i; - struct cpuinfo_x86 *c; - - for_each_cpu_mask(i, cpu_possible_map) { - c = &cpu_data(i); - /* mark all to hotplug */ - c->cpu_index = NR_CPUS; - } -} - -/* - * Prepare for SMP bootup. The MP table or ACPI has been read - * earlier. Just do some sanity checking here and enable APIC mode. - */ -void __init smp_prepare_cpus(unsigned int max_cpus) -{ - nmi_watchdog_default(); - smp_cpu_index_default(); - current_cpu_data = boot_cpu_data; - current_thread_info()->cpu = 0; /* needed? */ - set_cpu_sibling_map(0); - - if (smp_sanity_check(max_cpus) < 0) { - printk(KERN_INFO "SMP disabled\n"); - disable_smp(); - return; - } - - - /* - * Switch from PIC to APIC mode. - */ - setup_local_APIC(); - - /* - * Enable IO APIC before setting up error vector - */ - if (!skip_ioapic_setup && nr_ioapics) - enable_IO_APIC(); - end_local_APIC_setup(); - - if (GET_APIC_ID(apic_read(APIC_ID)) != boot_cpu_id) { - panic("Boot APIC ID in local APIC unexpected (%d vs %d)", - GET_APIC_ID(apic_read(APIC_ID)), boot_cpu_id); - /* Or can we switch back to PIC here? */ - } - - /* - * Now start the IO-APICs - */ - if (!skip_ioapic_setup && nr_ioapics) - setup_IO_APIC(); - else - nr_ioapics = 0; - - /* - * Set up local APIC timer on boot CPU. - */ - - setup_boot_clock(); -} - -/* - * Early setup to make printk work. - */ -void __init smp_prepare_boot_cpu(void) -{ - int me = smp_processor_id(); - /* already set me in cpu_online_map in boot_cpu_init() */ - cpu_set(me, cpu_callout_map); - per_cpu(cpu_state, me) = CPU_ONLINE; -} - -/* - * Entry point to boot a CPU. - */ -int __cpuinit __cpu_up(unsigned int cpu) -{ - int apicid = cpu_present_to_apicid(cpu); - unsigned long flags; - int err; - - WARN_ON(irqs_disabled()); - - Dprintk("++++++++++++++++++++=_---CPU UP %u\n", cpu); - - if (apicid == BAD_APICID || apicid == boot_cpu_id || - !physid_isset(apicid, phys_cpu_present_map)) { - printk("__cpu_up: bad cpu %d\n", cpu); - return -EINVAL; - } - - /* - * Already booted CPU? - */ - if (cpu_isset(cpu, cpu_callin_map)) { - Dprintk("do_boot_cpu %d Already started\n", cpu); - return -ENOSYS; - } - - /* - * Save current MTRR state in case it was changed since early boot - * (e.g. by the ACPI SMI) to initialize new CPUs with MTRRs in sync: - */ - mtrr_save_state(); - - per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; - /* Boot it! */ - err = do_boot_cpu(cpu, apicid); - if (err < 0) { - Dprintk("do_boot_cpu failed %d\n", err); - return err; - } - - /* Unleash the CPU! */ - Dprintk("waiting for cpu %d\n", cpu); - - /* - * Make sure and check TSC sync: - */ - local_irq_save(flags); - check_tsc_sync_source(cpu); - local_irq_restore(flags); - - while (!cpu_isset(cpu, cpu_online_map)) - cpu_relax(); - err = 0; - - return err; -} - -/* - * Finish the SMP boot. - */ -void __init smp_cpus_done(unsigned int max_cpus) -{ - smp_cleanup_boot(); - setup_ioapic_dest(); - check_nmi_watchdog(); -} - -#ifdef CONFIG_HOTPLUG_CPU - -static void remove_siblinginfo(int cpu) -{ - int sibling; - struct cpuinfo_x86 *c = &cpu_data(cpu); - - for_each_cpu_mask(sibling, per_cpu(cpu_core_map, cpu)) { - cpu_clear(cpu, per_cpu(cpu_core_map, sibling)); - /* - * last thread sibling in this cpu core going down - */ - if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1) - cpu_data(sibling).booted_cores--; - } - - for_each_cpu_mask(sibling, per_cpu(cpu_sibling_map, cpu)) - cpu_clear(cpu, per_cpu(cpu_sibling_map, sibling)); - cpus_clear(per_cpu(cpu_sibling_map, cpu)); - cpus_clear(per_cpu(cpu_core_map, cpu)); - c->phys_proc_id = 0; - c->cpu_core_id = 0; - cpu_clear(cpu, cpu_sibling_setup_map); -} - -static void __ref remove_cpu_from_maps(void) -{ - int cpu = smp_processor_id(); - - cpu_clear(cpu, cpu_callout_map); - cpu_clear(cpu, cpu_callin_map); - clear_bit(cpu, (unsigned long *)&cpu_initialized); /* was set by cpu_init() */ - clear_node_cpumask(cpu); -} - -int __cpu_disable(void) -{ - int cpu = smp_processor_id(); - - /* - * Perhaps use cpufreq to drop frequency, but that could go - * into generic code. - * - * We won't take down the boot processor on i386 due to some - * interrupts only being able to be serviced by the BSP. - * Especially so if we're not using an IOAPIC -zwane - */ - if (cpu == 0) - return -EBUSY; - - if (nmi_watchdog == NMI_LOCAL_APIC) - stop_apic_nmi_watchdog(NULL); - clear_local_APIC(); - - /* - * HACK: - * Allow any queued timer interrupts to get serviced - * This is only a temporary solution until we cleanup - * fixup_irqs as we do for IA64. - */ - local_irq_enable(); - mdelay(1); - - local_irq_disable(); - remove_siblinginfo(cpu); - - spin_lock(&vector_lock); - /* It's now safe to remove this processor from the online map */ - cpu_clear(cpu, cpu_online_map); - spin_unlock(&vector_lock); - remove_cpu_from_maps(); - fixup_irqs(cpu_online_map); - return 0; -} - -void __cpu_die(unsigned int cpu) -{ - /* We don't do anything here: idle task is faking death itself. */ - unsigned int i; - - for (i = 0; i < 10; i++) { - /* They ack this in play_dead by setting CPU_DEAD */ - if (per_cpu(cpu_state, cpu) == CPU_DEAD) { - printk ("CPU %d is now offline\n", cpu); - if (1 == num_online_cpus()) - alternatives_smp_switch(0); - return; - } - msleep(100); - } - printk(KERN_ERR "CPU %u didn't die...\n", cpu); -} - -static __init int setup_additional_cpus(char *s) -{ - return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL; -} -early_param("additional_cpus", setup_additional_cpus); - -#else /* ... !CONFIG_HOTPLUG_CPU */ - -int __cpu_disable(void) -{ - return -ENOSYS; -} - -void __cpu_die(unsigned int cpu) -{ - /* We said "no" in __cpu_disable */ - BUG(); -} -#endif /* CONFIG_HOTPLUG_CPU */ diff --git a/arch/x86/kernel/smpcommon.c b/arch/x86/kernel/smpcommon.c new file mode 100644 index 00000000000..3449064d141 --- /dev/null +++ b/arch/x86/kernel/smpcommon.c @@ -0,0 +1,83 @@ +/* + * SMP stuff which is common to all sub-architectures. + */ +#include <linux/module.h> +#include <asm/smp.h> + +#ifdef CONFIG_X86_32 +DEFINE_PER_CPU(unsigned long, this_cpu_off); +EXPORT_PER_CPU_SYMBOL(this_cpu_off); + +/* Initialize the CPU's GDT. This is either the boot CPU doing itself + (still using the master per-cpu area), or a CPU doing it for a + secondary which will soon come up. */ +__cpuinit void init_gdt(int cpu) +{ + struct desc_struct *gdt = get_cpu_gdt_table(cpu); + + pack_descriptor(&gdt[GDT_ENTRY_PERCPU], + __per_cpu_offset[cpu], 0xFFFFF, + 0x2 | DESCTYPE_S, 0x8); + + gdt[GDT_ENTRY_PERCPU].s = 1; + + per_cpu(this_cpu_off, cpu) = __per_cpu_offset[cpu]; + per_cpu(cpu_number, cpu) = cpu; +} +#endif + +/** + * smp_call_function(): Run a function on all other CPUs. + * @func: The function to run. This must be fast and non-blocking. + * @info: An arbitrary pointer to pass to the function. + * @nonatomic: Unused. + * @wait: If true, wait (atomically) until function has completed on other CPUs. + * + * Returns 0 on success, else a negative status code. + * + * If @wait is true, then returns once @func has returned; otherwise + * it returns just before the target cpu calls @func. + * + * You must not call this function with disabled interrupts or from a + * hardware interrupt handler or from a bottom half handler. + */ +int smp_call_function(void (*func) (void *info), void *info, int nonatomic, + int wait) +{ + return smp_call_function_mask(cpu_online_map, func, info, wait); +} +EXPORT_SYMBOL(smp_call_function); + +/** + * smp_call_function_single - Run a function on a specific CPU + * @cpu: The target CPU. Cannot be the calling CPU. + * @func: The function to run. This must be fast and non-blocking. + * @info: An arbitrary pointer to pass to the function. + * @nonatomic: Unused. + * @wait: If true, wait until function has completed on other CPUs. + * + * Returns 0 on success, else a negative status code. + * + * If @wait is true, then returns once @func has returned; otherwise + * it returns just before the target cpu calls @func. + */ +int smp_call_function_single(int cpu, void (*func) (void *info), void *info, + int nonatomic, int wait) +{ + /* prevent preemption and reschedule on another processor */ + int ret; + int me = get_cpu(); + if (cpu == me) { + local_irq_disable(); + func(info); + local_irq_enable(); + put_cpu(); + return 0; + } + + ret = smp_call_function_mask(cpumask_of_cpu(cpu), func, info, wait); + + put_cpu(); + return ret; +} +EXPORT_SYMBOL(smp_call_function_single); diff --git a/arch/x86/kernel/smpcommon_32.c b/arch/x86/kernel/smpcommon_32.c index 8bc38af29ae..8b137891791 100644 --- a/arch/x86/kernel/smpcommon_32.c +++ b/arch/x86/kernel/smpcommon_32.c @@ -1,82 +1 @@ -/* - * SMP stuff which is common to all sub-architectures. - */ -#include <linux/module.h> -#include <asm/smp.h> -DEFINE_PER_CPU(unsigned long, this_cpu_off); -EXPORT_PER_CPU_SYMBOL(this_cpu_off); - -/* Initialize the CPU's GDT. This is either the boot CPU doing itself - (still using the master per-cpu area), or a CPU doing it for a - secondary which will soon come up. */ -__cpuinit void init_gdt(int cpu) -{ - struct desc_struct *gdt = get_cpu_gdt_table(cpu); - - pack_descriptor(&gdt[GDT_ENTRY_PERCPU], - __per_cpu_offset[cpu], 0xFFFFF, - 0x2 | DESCTYPE_S, 0x8); - - gdt[GDT_ENTRY_PERCPU].s = 1; - - per_cpu(this_cpu_off, cpu) = __per_cpu_offset[cpu]; - per_cpu(cpu_number, cpu) = cpu; -} - - -/** - * smp_call_function(): Run a function on all other CPUs. - * @func: The function to run. This must be fast and non-blocking. - * @info: An arbitrary pointer to pass to the function. - * @nonatomic: Unused. - * @wait: If true, wait (atomically) until function has completed on other CPUs. - * - * Returns 0 on success, else a negative status code. - * - * If @wait is true, then returns once @func has returned; otherwise - * it returns just before the target cpu calls @func. - * - * You must not call this function with disabled interrupts or from a - * hardware interrupt handler or from a bottom half handler. - */ -int smp_call_function(void (*func) (void *info), void *info, int nonatomic, - int wait) -{ - return smp_call_function_mask(cpu_online_map, func, info, wait); -} -EXPORT_SYMBOL(smp_call_function); - -/** - * smp_call_function_single - Run a function on a specific CPU - * @cpu: The target CPU. Cannot be the calling CPU. - * @func: The function to run. This must be fast and non-blocking. - * @info: An arbitrary pointer to pass to the function. - * @nonatomic: Unused. - * @wait: If true, wait until function has completed on other CPUs. - * - * Returns 0 on success, else a negative status code. - * - * If @wait is true, then returns once @func has returned; otherwise - * it returns just before the target cpu calls @func. - */ -int smp_call_function_single(int cpu, void (*func) (void *info), void *info, - int nonatomic, int wait) -{ - /* prevent preemption and reschedule on another processor */ - int ret; - int me = get_cpu(); - if (cpu == me) { - local_irq_disable(); - func(info); - local_irq_enable(); - put_cpu(); - return 0; - } - - ret = smp_call_function_mask(cpumask_of_cpu(cpu), func, info, wait); - - put_cpu(); - return ret; -} -EXPORT_SYMBOL(smp_call_function_single); diff --git a/arch/x86/kernel/srat_32.c b/arch/x86/kernel/srat_32.c index b72e61359c3..70e4a374b4e 100644 --- a/arch/x86/kernel/srat_32.c +++ b/arch/x86/kernel/srat_32.c @@ -277,14 +277,14 @@ int __init get_memcfg_from_srat(void) rsdp_address = acpi_os_get_root_pointer(); if (!rsdp_address) { printk("%s: System description tables not found\n", - __FUNCTION__); + __func__); goto out_err; } - printk("%s: assigning address to rsdp\n", __FUNCTION__); + printk("%s: assigning address to rsdp\n", __func__); rsdp = (struct acpi_table_rsdp *)(u32)rsdp_address; if (!rsdp) { - printk("%s: Didn't find ACPI root!\n", __FUNCTION__); + printk("%s: Didn't find ACPI root!\n", __func__); goto out_err; } @@ -292,7 +292,7 @@ int __init get_memcfg_from_srat(void) rsdp->oem_id); if (strncmp(rsdp->signature, ACPI_SIG_RSDP,strlen(ACPI_SIG_RSDP))) { - printk(KERN_WARNING "%s: RSDP table signature incorrect\n", __FUNCTION__); + printk(KERN_WARNING "%s: RSDP table signature incorrect\n", __func__); goto out_err; } @@ -302,7 +302,7 @@ int __init get_memcfg_from_srat(void) if (!rsdt) { printk(KERN_WARNING "%s: ACPI: Invalid root system description tables (RSDT)\n", - __FUNCTION__); + __func__); goto out_err; } diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c index 071ff479823..92c20fee678 100644 --- a/arch/x86/kernel/step.c +++ b/arch/x86/kernel/step.c @@ -148,7 +148,7 @@ static void write_debugctlmsr(struct task_struct *child, unsigned long val) if (child != current) return; - wrmsrl(MSR_IA32_DEBUGCTLMSR, val); + update_debugctlmsr(val); } /* diff --git a/arch/x86/kernel/summit_32.c b/arch/x86/kernel/summit_32.c index 72f46340159..6878a9c2df5 100644 --- a/arch/x86/kernel/summit_32.c +++ b/arch/x86/kernel/summit_32.c @@ -35,43 +35,47 @@ static struct rio_table_hdr *rio_table_hdr __initdata; static struct scal_detail *scal_devs[MAX_NUMNODES] __initdata; static struct rio_detail *rio_devs[MAX_NUMNODES*4] __initdata; +static int mp_bus_id_to_node[MAX_MP_BUSSES] __initdata; + static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus) { int twister = 0, node = 0; int i, bus, num_buses; - for(i = 0; i < rio_table_hdr->num_rio_dev; i++){ - if (rio_devs[i]->node_id == rio_devs[wpeg_num]->owner_id){ + for (i = 0; i < rio_table_hdr->num_rio_dev; i++) { + if (rio_devs[i]->node_id == rio_devs[wpeg_num]->owner_id) { twister = rio_devs[i]->owner_id; break; } } - if (i == rio_table_hdr->num_rio_dev){ - printk(KERN_ERR "%s: Couldn't find owner Cyclone for Winnipeg!\n", __FUNCTION__); + if (i == rio_table_hdr->num_rio_dev) { + printk(KERN_ERR "%s: Couldn't find owner Cyclone for Winnipeg!\n", __func__); return last_bus; } - for(i = 0; i < rio_table_hdr->num_scal_dev; i++){ - if (scal_devs[i]->node_id == twister){ + for (i = 0; i < rio_table_hdr->num_scal_dev; i++) { + if (scal_devs[i]->node_id == twister) { node = scal_devs[i]->node_id; break; } } - if (i == rio_table_hdr->num_scal_dev){ - printk(KERN_ERR "%s: Couldn't find owner Twister for Cyclone!\n", __FUNCTION__); + if (i == rio_table_hdr->num_scal_dev) { + printk(KERN_ERR "%s: Couldn't find owner Twister for Cyclone!\n", __func__); return last_bus; } - switch (rio_devs[wpeg_num]->type){ + switch (rio_devs[wpeg_num]->type) { case CompatWPEG: - /* The Compatibility Winnipeg controls the 2 legacy buses, + /* + * The Compatibility Winnipeg controls the 2 legacy buses, * the 66MHz PCI bus [2 slots] and the 2 "extra" buses in case * a PCI-PCI bridge card is used in either slot: total 5 buses. */ num_buses = 5; break; case AltWPEG: - /* The Alternate Winnipeg controls the 2 133MHz buses [1 slot + /* + * The Alternate Winnipeg controls the 2 133MHz buses [1 slot * each], their 2 "extra" buses, the 100MHz bus [2 slots] and * the "extra" buses for each of those slots: total 7 buses. */ @@ -79,17 +83,18 @@ static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus) break; case LookOutAWPEG: case LookOutBWPEG: - /* A Lookout Winnipeg controls 3 100MHz buses [2 slots each] + /* + * A Lookout Winnipeg controls 3 100MHz buses [2 slots each] * & the "extra" buses for each of those slots: total 9 buses. */ num_buses = 9; break; default: - printk(KERN_INFO "%s: Unsupported Winnipeg type!\n", __FUNCTION__); + printk(KERN_INFO "%s: Unsupported Winnipeg type!\n", __func__); return last_bus; } - for(bus = last_bus; bus < last_bus + num_buses; bus++) + for (bus = last_bus; bus < last_bus + num_buses; bus++) mp_bus_id_to_node[bus] = node; return bus; } @@ -99,14 +104,14 @@ static int __init build_detail_arrays(void) unsigned long ptr; int i, scal_detail_size, rio_detail_size; - if (rio_table_hdr->num_scal_dev > MAX_NUMNODES){ - printk(KERN_WARNING "%s: MAX_NUMNODES too low! Defined as %d, but system has %d nodes.\n", __FUNCTION__, MAX_NUMNODES, rio_table_hdr->num_scal_dev); + if (rio_table_hdr->num_scal_dev > MAX_NUMNODES) { + printk(KERN_WARNING "%s: MAX_NUMNODES too low! Defined as %d, but system has %d nodes.\n", __func__, MAX_NUMNODES, rio_table_hdr->num_scal_dev); return 0; } - switch (rio_table_hdr->version){ + switch (rio_table_hdr->version) { default: - printk(KERN_WARNING "%s: Invalid Rio Grande Table Version: %d\n", __FUNCTION__, rio_table_hdr->version); + printk(KERN_WARNING "%s: Invalid Rio Grande Table Version: %d\n", __func__, rio_table_hdr->version); return 0; case 2: scal_detail_size = 11; @@ -119,10 +124,10 @@ static int __init build_detail_arrays(void) } ptr = (unsigned long)rio_table_hdr + 3; - for(i = 0; i < rio_table_hdr->num_scal_dev; i++, ptr += scal_detail_size) + for (i = 0; i < rio_table_hdr->num_scal_dev; i++, ptr += scal_detail_size) scal_devs[i] = (struct scal_detail *)ptr; - for(i = 0; i < rio_table_hdr->num_rio_dev; i++, ptr += rio_detail_size) + for (i = 0; i < rio_table_hdr->num_rio_dev; i++, ptr += rio_detail_size) rio_devs[i] = (struct rio_detail *)ptr; return 1; @@ -140,9 +145,9 @@ void __init setup_summit(void) rio_table_hdr = NULL; offset = 0x180; - while (offset){ + while (offset) { /* The block id is stored in the 2nd word */ - if (*((unsigned short *)(ptr + offset + 2)) == 0x4752){ + if (*((unsigned short *)(ptr + offset + 2)) == 0x4752) { /* set the pointer past the offset & block id */ rio_table_hdr = (struct rio_table_hdr *)(ptr + offset + 4); break; @@ -150,8 +155,8 @@ void __init setup_summit(void) /* The next offset is stored in the 1st word. 0 means no more */ offset = *((unsigned short *)(ptr + offset)); } - if (!rio_table_hdr){ - printk(KERN_ERR "%s: Unable to locate Rio Grande Table in EBDA - bailing!\n", __FUNCTION__); + if (!rio_table_hdr) { + printk(KERN_ERR "%s: Unable to locate Rio Grande Table in EBDA - bailing!\n", __func__); return; } @@ -161,8 +166,8 @@ void __init setup_summit(void) /* The first Winnipeg we're looking for has an index of 0 */ next_wpeg = 0; do { - for(i = 0; i < rio_table_hdr->num_rio_dev; i++){ - if (is_WPEG(rio_devs[i]) && rio_devs[i]->WP_index == next_wpeg){ + for (i = 0; i < rio_table_hdr->num_rio_dev; i++) { + if (is_WPEG(rio_devs[i]) && rio_devs[i]->WP_index == next_wpeg) { /* It's the Winnipeg we're looking for! */ next_bus = setup_pci_node_map_for_wpeg(i, next_bus); next_wpeg++; diff --git a/arch/x86/kernel/syscall_64.c b/arch/x86/kernel/syscall_64.c index 9d498c2f8ee..170d43c1748 100644 --- a/arch/x86/kernel/syscall_64.c +++ b/arch/x86/kernel/syscall_64.c @@ -1,4 +1,4 @@ -/* System call table for x86-64. */ +/* System call table for x86-64. */ #include <linux/linkage.h> #include <linux/sys.h> @@ -7,20 +7,23 @@ #define __NO_STUBS -#define __SYSCALL(nr, sym) extern asmlinkage void sym(void) ; +#define __SYSCALL(nr, sym) extern asmlinkage void sym(void) ; #undef _ASM_X86_64_UNISTD_H_ #include <asm/unistd_64.h> #undef __SYSCALL -#define __SYSCALL(nr, sym) [ nr ] = sym, +#define __SYSCALL(nr, sym) [nr] = sym, #undef _ASM_X86_64_UNISTD_H_ -typedef void (*sys_call_ptr_t)(void); +typedef void (*sys_call_ptr_t)(void); extern void sys_ni_syscall(void); const sys_call_ptr_t sys_call_table[__NR_syscall_max+1] = { - /* Smells like a like a compiler bug -- it doesn't work when the & below is removed. */ + /* + *Smells like a like a compiler bug -- it doesn't work + *when the & below is removed. + */ [0 ... __NR_syscall_max] = &sys_ni_syscall, #include <asm/unistd_64.h> }; diff --git a/arch/x86/kernel/test_nx.c b/arch/x86/kernel/test_nx.c index 10b8a6f69f8..787a5e499dd 100644 --- a/arch/x86/kernel/test_nx.c +++ b/arch/x86/kernel/test_nx.c @@ -11,6 +11,8 @@ */ #include <linux/module.h> #include <linux/sort.h> +#include <linux/slab.h> + #include <asm/uaccess.h> #include <asm/asm.h> diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c new file mode 100644 index 00000000000..9bb2363851a --- /dev/null +++ b/arch/x86/kernel/tlb_32.c @@ -0,0 +1,243 @@ +#include <linux/spinlock.h> +#include <linux/cpu.h> +#include <linux/interrupt.h> + +#include <asm/tlbflush.h> + +DEFINE_PER_CPU(struct tlb_state, cpu_tlbstate) + ____cacheline_aligned = { &init_mm, 0, }; + +/* must come after the send_IPI functions above for inlining */ +#include <mach_ipi.h> + +/* + * Smarter SMP flushing macros. + * c/o Linus Torvalds. + * + * These mean you can really definitely utterly forget about + * writing to user space from interrupts. (Its not allowed anyway). + * + * Optimizations Manfred Spraul <manfred@colorfullife.com> + */ + +static cpumask_t flush_cpumask; +static struct mm_struct *flush_mm; +static unsigned long flush_va; +static DEFINE_SPINLOCK(tlbstate_lock); + +/* + * We cannot call mmdrop() because we are in interrupt context, + * instead update mm->cpu_vm_mask. + * + * We need to reload %cr3 since the page tables may be going + * away from under us.. + */ +void leave_mm(int cpu) +{ + if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) + BUG(); + cpu_clear(cpu, per_cpu(cpu_tlbstate, cpu).active_mm->cpu_vm_mask); + load_cr3(swapper_pg_dir); +} +EXPORT_SYMBOL_GPL(leave_mm); + +/* + * + * The flush IPI assumes that a thread switch happens in this order: + * [cpu0: the cpu that switches] + * 1) switch_mm() either 1a) or 1b) + * 1a) thread switch to a different mm + * 1a1) cpu_clear(cpu, old_mm->cpu_vm_mask); + * Stop ipi delivery for the old mm. This is not synchronized with + * the other cpus, but smp_invalidate_interrupt ignore flush ipis + * for the wrong mm, and in the worst case we perform a superfluous + * tlb flush. + * 1a2) set cpu_tlbstate to TLBSTATE_OK + * Now the smp_invalidate_interrupt won't call leave_mm if cpu0 + * was in lazy tlb mode. + * 1a3) update cpu_tlbstate[].active_mm + * Now cpu0 accepts tlb flushes for the new mm. + * 1a4) cpu_set(cpu, new_mm->cpu_vm_mask); + * Now the other cpus will send tlb flush ipis. + * 1a4) change cr3. + * 1b) thread switch without mm change + * cpu_tlbstate[].active_mm is correct, cpu0 already handles + * flush ipis. + * 1b1) set cpu_tlbstate to TLBSTATE_OK + * 1b2) test_and_set the cpu bit in cpu_vm_mask. + * Atomically set the bit [other cpus will start sending flush ipis], + * and test the bit. + * 1b3) if the bit was 0: leave_mm was called, flush the tlb. + * 2) switch %%esp, ie current + * + * The interrupt must handle 2 special cases: + * - cr3 is changed before %%esp, ie. it cannot use current->{active_,}mm. + * - the cpu performs speculative tlb reads, i.e. even if the cpu only + * runs in kernel space, the cpu could load tlb entries for user space + * pages. + * + * The good news is that cpu_tlbstate is local to each cpu, no + * write/read ordering problems. + */ + +/* + * TLB flush IPI: + * + * 1) Flush the tlb entries if the cpu uses the mm that's being flushed. + * 2) Leave the mm if we are in the lazy tlb mode. + */ + +void smp_invalidate_interrupt(struct pt_regs *regs) +{ + unsigned long cpu; + + cpu = get_cpu(); + + if (!cpu_isset(cpu, flush_cpumask)) + goto out; + /* + * This was a BUG() but until someone can quote me the + * line from the intel manual that guarantees an IPI to + * multiple CPUs is retried _only_ on the erroring CPUs + * its staying as a return + * + * BUG(); + */ + + if (flush_mm == per_cpu(cpu_tlbstate, cpu).active_mm) { + if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) { + if (flush_va == TLB_FLUSH_ALL) + local_flush_tlb(); + else + __flush_tlb_one(flush_va); + } else + leave_mm(cpu); + } + ack_APIC_irq(); + smp_mb__before_clear_bit(); + cpu_clear(cpu, flush_cpumask); + smp_mb__after_clear_bit(); +out: + put_cpu_no_resched(); + __get_cpu_var(irq_stat).irq_tlb_count++; +} + +void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm, + unsigned long va) +{ + cpumask_t cpumask = *cpumaskp; + + /* + * A couple of (to be removed) sanity checks: + * + * - current CPU must not be in mask + * - mask must exist :) + */ + BUG_ON(cpus_empty(cpumask)); + BUG_ON(cpu_isset(smp_processor_id(), cpumask)); + BUG_ON(!mm); + +#ifdef CONFIG_HOTPLUG_CPU + /* If a CPU which we ran on has gone down, OK. */ + cpus_and(cpumask, cpumask, cpu_online_map); + if (unlikely(cpus_empty(cpumask))) + return; +#endif + + /* + * i'm not happy about this global shared spinlock in the + * MM hot path, but we'll see how contended it is. + * AK: x86-64 has a faster method that could be ported. + */ + spin_lock(&tlbstate_lock); + + flush_mm = mm; + flush_va = va; + cpus_or(flush_cpumask, cpumask, flush_cpumask); + /* + * We have to send the IPI only to + * CPUs affected. + */ + send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR); + + while (!cpus_empty(flush_cpumask)) + /* nothing. lockup detection does not belong here */ + cpu_relax(); + + flush_mm = NULL; + flush_va = 0; + spin_unlock(&tlbstate_lock); +} + +void flush_tlb_current_task(void) +{ + struct mm_struct *mm = current->mm; + cpumask_t cpu_mask; + + preempt_disable(); + cpu_mask = mm->cpu_vm_mask; + cpu_clear(smp_processor_id(), cpu_mask); + + local_flush_tlb(); + if (!cpus_empty(cpu_mask)) + flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL); + preempt_enable(); +} + +void flush_tlb_mm(struct mm_struct *mm) +{ + cpumask_t cpu_mask; + + preempt_disable(); + cpu_mask = mm->cpu_vm_mask; + cpu_clear(smp_processor_id(), cpu_mask); + + if (current->active_mm == mm) { + if (current->mm) + local_flush_tlb(); + else + leave_mm(smp_processor_id()); + } + if (!cpus_empty(cpu_mask)) + flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL); + + preempt_enable(); +} + +void flush_tlb_page(struct vm_area_struct *vma, unsigned long va) +{ + struct mm_struct *mm = vma->vm_mm; + cpumask_t cpu_mask; + + preempt_disable(); + cpu_mask = mm->cpu_vm_mask; + cpu_clear(smp_processor_id(), cpu_mask); + + if (current->active_mm == mm) { + if (current->mm) + __flush_tlb_one(va); + else + leave_mm(smp_processor_id()); + } + + if (!cpus_empty(cpu_mask)) + flush_tlb_others(cpu_mask, mm, va); + + preempt_enable(); +} +EXPORT_SYMBOL(flush_tlb_page); + +static void do_flush_tlb_all(void *info) +{ + unsigned long cpu = smp_processor_id(); + + __flush_tlb_all(); + if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_LAZY) + leave_mm(cpu); +} + +void flush_tlb_all(void) +{ + on_each_cpu(do_flush_tlb_all, NULL, 1, 1); +} + diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/tlb_64.c index 2fd74b06db6..1558e513757 100644 --- a/arch/x86/kernel/smp_64.c +++ b/arch/x86/kernel/tlb_64.c @@ -1,14 +1,3 @@ -/* - * Intel SMP support routines. - * - * (c) 1995 Alan Cox, Building #3 <alan@redhat.com> - * (c) 1998-99, 2000 Ingo Molnar <mingo@redhat.com> - * (c) 2002,2003 Andi Kleen, SuSE Labs. - * - * This code is released under the GNU General Public License version 2 or - * later. - */ - #include <linux/init.h> #include <linux/mm.h> @@ -22,12 +11,12 @@ #include <asm/mtrr.h> #include <asm/pgalloc.h> #include <asm/tlbflush.h> -#include <asm/mach_apic.h> #include <asm/mmu_context.h> #include <asm/proto.h> #include <asm/apicdef.h> #include <asm/idle.h> +#include <mach_ipi.h> /* * Smarter SMP flushing macros. * c/o Linus Torvalds. @@ -228,7 +217,7 @@ void flush_tlb_current_task(void) preempt_enable(); } -void flush_tlb_mm (struct mm_struct * mm) +void flush_tlb_mm(struct mm_struct *mm) { cpumask_t cpu_mask; @@ -248,7 +237,7 @@ void flush_tlb_mm (struct mm_struct * mm) preempt_enable(); } -void flush_tlb_page(struct vm_area_struct * vma, unsigned long va) +void flush_tlb_page(struct vm_area_struct *vma, unsigned long va) { struct mm_struct *mm = vma->vm_mm; cpumask_t cpu_mask; @@ -258,7 +247,7 @@ void flush_tlb_page(struct vm_area_struct * vma, unsigned long va) cpu_clear(smp_processor_id(), cpu_mask); if (current->active_mm == mm) { - if(current->mm) + if (current->mm) __flush_tlb_one(va); else leave_mm(smp_processor_id()); @@ -270,7 +259,7 @@ void flush_tlb_page(struct vm_area_struct * vma, unsigned long va) preempt_enable(); } -static void do_flush_tlb_all(void* info) +static void do_flush_tlb_all(void *info) { unsigned long cpu = smp_processor_id(); @@ -283,248 +272,3 @@ void flush_tlb_all(void) { on_each_cpu(do_flush_tlb_all, NULL, 1, 1); } - -/* - * this function sends a 'reschedule' IPI to another CPU. - * it goes straight through and wastes no time serializing - * anything. Worst case is that we lose a reschedule ... - */ - -void smp_send_reschedule(int cpu) -{ - send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR); -} - -/* - * Structure and data for smp_call_function(). This is designed to minimise - * static memory requirements. It also looks cleaner. - */ -static DEFINE_SPINLOCK(call_lock); - -struct call_data_struct { - void (*func) (void *info); - void *info; - atomic_t started; - atomic_t finished; - int wait; -}; - -static struct call_data_struct * call_data; - -void lock_ipi_call_lock(void) -{ - spin_lock_irq(&call_lock); -} - -void unlock_ipi_call_lock(void) -{ - spin_unlock_irq(&call_lock); -} - -/* - * this function sends a 'generic call function' IPI to all other CPU - * of the system defined in the mask. - */ -static int __smp_call_function_mask(cpumask_t mask, - void (*func)(void *), void *info, - int wait) -{ - struct call_data_struct data; - cpumask_t allbutself; - int cpus; - - allbutself = cpu_online_map; - cpu_clear(smp_processor_id(), allbutself); - - cpus_and(mask, mask, allbutself); - cpus = cpus_weight(mask); - - if (!cpus) - return 0; - - data.func = func; - data.info = info; - atomic_set(&data.started, 0); - data.wait = wait; - if (wait) - atomic_set(&data.finished, 0); - - call_data = &data; - wmb(); - - /* Send a message to other CPUs */ - if (cpus_equal(mask, allbutself)) - send_IPI_allbutself(CALL_FUNCTION_VECTOR); - else - send_IPI_mask(mask, CALL_FUNCTION_VECTOR); - - /* Wait for response */ - while (atomic_read(&data.started) != cpus) - cpu_relax(); - - if (!wait) - return 0; - - while (atomic_read(&data.finished) != cpus) - cpu_relax(); - - return 0; -} -/** - * smp_call_function_mask(): Run a function on a set of other CPUs. - * @mask: The set of cpus to run on. Must not include the current cpu. - * @func: The function to run. This must be fast and non-blocking. - * @info: An arbitrary pointer to pass to the function. - * @wait: If true, wait (atomically) until function has completed on other CPUs. - * - * Returns 0 on success, else a negative status code. - * - * If @wait is true, then returns once @func has returned; otherwise - * it returns just before the target cpu calls @func. - * - * You must not call this function with disabled interrupts or from a - * hardware interrupt handler or from a bottom half handler. - */ -int smp_call_function_mask(cpumask_t mask, - void (*func)(void *), void *info, - int wait) -{ - int ret; - - /* Can deadlock when called with interrupts disabled */ - WARN_ON(irqs_disabled()); - - spin_lock(&call_lock); - ret = __smp_call_function_mask(mask, func, info, wait); - spin_unlock(&call_lock); - return ret; -} -EXPORT_SYMBOL(smp_call_function_mask); - -/* - * smp_call_function_single - Run a function on a specific CPU - * @func: The function to run. This must be fast and non-blocking. - * @info: An arbitrary pointer to pass to the function. - * @nonatomic: Currently unused. - * @wait: If true, wait until function has completed on other CPUs. - * - * Retrurns 0 on success, else a negative status code. - * - * Does not return until the remote CPU is nearly ready to execute <func> - * or is or has executed. - */ - -int smp_call_function_single (int cpu, void (*func) (void *info), void *info, - int nonatomic, int wait) -{ - /* prevent preemption and reschedule on another processor */ - int ret, me = get_cpu(); - - /* Can deadlock when called with interrupts disabled */ - WARN_ON(irqs_disabled()); - - if (cpu == me) { - local_irq_disable(); - func(info); - local_irq_enable(); - put_cpu(); - return 0; - } - - ret = smp_call_function_mask(cpumask_of_cpu(cpu), func, info, wait); - - put_cpu(); - return ret; -} -EXPORT_SYMBOL(smp_call_function_single); - -/* - * smp_call_function - run a function on all other CPUs. - * @func: The function to run. This must be fast and non-blocking. - * @info: An arbitrary pointer to pass to the function. - * @nonatomic: currently unused. - * @wait: If true, wait (atomically) until function has completed on other - * CPUs. - * - * Returns 0 on success, else a negative status code. Does not return until - * remote CPUs are nearly ready to execute func or are or have executed. - * - * You must not call this function with disabled interrupts or from a - * hardware interrupt handler or from a bottom half handler. - * Actually there are a few legal cases, like panic. - */ -int smp_call_function (void (*func) (void *info), void *info, int nonatomic, - int wait) -{ - return smp_call_function_mask(cpu_online_map, func, info, wait); -} -EXPORT_SYMBOL(smp_call_function); - -static void stop_this_cpu(void *dummy) -{ - local_irq_disable(); - /* - * Remove this CPU: - */ - cpu_clear(smp_processor_id(), cpu_online_map); - disable_local_APIC(); - for (;;) - halt(); -} - -void smp_send_stop(void) -{ - int nolock; - unsigned long flags; - - if (reboot_force) - return; - - /* Don't deadlock on the call lock in panic */ - nolock = !spin_trylock(&call_lock); - local_irq_save(flags); - __smp_call_function_mask(cpu_online_map, stop_this_cpu, NULL, 0); - if (!nolock) - spin_unlock(&call_lock); - disable_local_APIC(); - local_irq_restore(flags); -} - -/* - * Reschedule call back. Nothing to do, - * all the work is done automatically when - * we return from the interrupt. - */ -asmlinkage void smp_reschedule_interrupt(void) -{ - ack_APIC_irq(); - add_pda(irq_resched_count, 1); -} - -asmlinkage void smp_call_function_interrupt(void) -{ - void (*func) (void *info) = call_data->func; - void *info = call_data->info; - int wait = call_data->wait; - - ack_APIC_irq(); - /* - * Notify initiating CPU that I've grabbed the data and am - * about to execute the function - */ - mb(); - atomic_inc(&call_data->started); - /* - * At this point the info structure may be out of scope unless wait==1 - */ - exit_idle(); - irq_enter(); - (*func)(info); - add_pda(irq_call_count, 1); - irq_exit(); - if (wait) { - mb(); - atomic_inc(&call_data->finished); - } -} - diff --git a/arch/x86/kernel/tls.c b/arch/x86/kernel/tls.c index 022bcaa3b42..ab6bf375a30 100644 --- a/arch/x86/kernel/tls.c +++ b/arch/x86/kernel/tls.c @@ -92,7 +92,7 @@ int do_set_thread_area(struct task_struct *p, int idx, asmlinkage int sys_set_thread_area(struct user_desc __user *u_info) { int ret = do_set_thread_area(current, -1, u_info, 1); - prevent_tail_call(ret); + asmlinkage_protect(1, ret, u_info); return ret; } @@ -142,7 +142,7 @@ int do_get_thread_area(struct task_struct *p, int idx, asmlinkage int sys_get_thread_area(struct user_desc __user *u_info) { int ret = do_get_thread_area(current, -1, u_info); - prevent_tail_call(ret); + asmlinkage_protect(1, ret, u_info); return ret; } diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c new file mode 100644 index 00000000000..abbf199adeb --- /dev/null +++ b/arch/x86/kernel/trampoline.c @@ -0,0 +1,18 @@ +#include <linux/io.h> + +#include <asm/trampoline.h> + +/* ready for x86_64, no harm for x86, since it will overwrite after alloc */ +unsigned char *trampoline_base = __va(TRAMPOLINE_BASE); + +/* + * Currently trivial. Write the real->protected mode + * bootstrap into the page concerned. The caller + * has made sure it's suitably aligned. + */ +unsigned long setup_trampoline(void) +{ + memcpy(trampoline_base, trampoline_data, + trampoline_end - trampoline_data); + return virt_to_phys(trampoline_base); +} diff --git a/arch/x86/kernel/trampoline_64.S b/arch/x86/kernel/trampoline_64.S index 4aedd0bcee4..894293c598d 100644 --- a/arch/x86/kernel/trampoline_64.S +++ b/arch/x86/kernel/trampoline_64.S @@ -30,12 +30,7 @@ #include <asm/msr.h> #include <asm/segment.h> -/* We can free up trampoline after bootup if cpu hotplug is not supported. */ -#ifndef CONFIG_HOTPLUG_CPU -.section .init.data, "aw", @progbits -#else .section .rodata, "a", @progbits -#endif .code16 diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index b22c01e05a1..65791ca2824 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -9,26 +9,28 @@ * 'Traps.c' handles hardware traps and faults after we have saved some * state in 'asm.s'. */ -#include <linux/sched.h> +#include <linux/interrupt.h> +#include <linux/kallsyms.h> +#include <linux/spinlock.h> +#include <linux/highmem.h> +#include <linux/kprobes.h> +#include <linux/uaccess.h> +#include <linux/utsname.h> +#include <linux/kdebug.h> #include <linux/kernel.h> +#include <linux/module.h> +#include <linux/ptrace.h> #include <linux/string.h> +#include <linux/unwind.h> +#include <linux/delay.h> #include <linux/errno.h> +#include <linux/kexec.h> +#include <linux/sched.h> #include <linux/timer.h> -#include <linux/mm.h> #include <linux/init.h> -#include <linux/delay.h> -#include <linux/spinlock.h> -#include <linux/interrupt.h> -#include <linux/highmem.h> -#include <linux/kallsyms.h> -#include <linux/ptrace.h> -#include <linux/utsname.h> -#include <linux/kprobes.h> -#include <linux/kexec.h> -#include <linux/unwind.h> -#include <linux/uaccess.h> -#include <linux/nmi.h> #include <linux/bug.h> +#include <linux/nmi.h> +#include <linux/mm.h> #ifdef CONFIG_EISA #include <linux/ioport.h> @@ -43,21 +45,18 @@ #include <linux/edac.h> #endif +#include <asm/arch_hooks.h> +#include <asm/stacktrace.h> #include <asm/processor.h> -#include <asm/system.h> -#include <asm/io.h> -#include <asm/atomic.h> #include <asm/debugreg.h> +#include <asm/atomic.h> +#include <asm/system.h> +#include <asm/unwind.h> #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 <linux/kdebug.h> -#include <asm/stacktrace.h> - -#include <linux/module.h> +#include <asm/io.h> #include "mach_traps.h" @@ -69,7 +68,7 @@ EXPORT_SYMBOL_GPL(used_vectors); asmlinkage int system_call(void); /* Do we ignore FPU interrupts ? */ -char ignore_fpu_irq = 0; +char ignore_fpu_irq; /* * The IDT has to be page-aligned to simplify the Pentium @@ -105,12 +104,13 @@ static unsigned int code_bytes = 64; void printk_address(unsigned long address, int reliable) { #ifdef CONFIG_KALLSYMS - unsigned long offset = 0, symsize; + char namebuf[KSYM_NAME_LEN]; + unsigned long offset = 0; + unsigned long symsize; const char *symname; - char *modname; - char *delim = ":"; - char namebuf[128]; char reliab[4] = ""; + char *delim = ":"; + char *modname; symname = kallsyms_lookup(address, &symsize, &offset, &modname, namebuf); @@ -138,13 +138,14 @@ static inline int valid_stack_ptr(struct thread_info *tinfo, void *p, unsigned s /* The form of the top of the frame on the stack */ struct stack_frame { - struct stack_frame *next_frame; - unsigned long return_address; + struct stack_frame *next_frame; + unsigned long return_address; }; -static inline unsigned long print_context_stack(struct thread_info *tinfo, - unsigned long *stack, unsigned long bp, - const struct stacktrace_ops *ops, void *data) +static inline unsigned long +print_context_stack(struct thread_info *tinfo, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data) { struct stack_frame *frame = (struct stack_frame *)bp; @@ -166,7 +167,7 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo, return bp; } -#define MSG(msg) ops->warning(data, msg) +#define MSG(msg) ops->warning(data, msg) void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, @@ -177,6 +178,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, if (!stack) { unsigned long dummy; + stack = &dummy; if (task != current) stack = (unsigned long *)task->thread.sp; @@ -186,7 +188,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, if (!bp) { if (task == current) { /* Grab bp right from our regs */ - asm ("movl %%ebp, %0" : "=r" (bp) : ); + asm("movl %%ebp, %0" : "=r" (bp) :); } else { /* bp is the last reg pushed by switch_to */ bp = *(unsigned long *) task->thread.sp; @@ -196,15 +198,18 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, while (1) { struct thread_info *context; + context = (struct thread_info *) ((unsigned long)stack & (~(THREAD_SIZE - 1))); bp = print_context_stack(context, stack, bp, ops, data); - /* Should be after the line below, but somewhere - in early boot context comes out corrupted and we - can't reference it -AK */ + /* + * Should be after the line below, but somewhere + * in early boot context comes out corrupted and we + * can't reference it: + */ if (ops->stack(data, "IRQ") < 0) break; - stack = (unsigned long*)context->previous_esp; + stack = (unsigned long *)context->previous_esp; if (!stack) break; touch_nmi_watchdog(); @@ -243,15 +248,15 @@ static void print_trace_address(void *data, unsigned long addr, int reliable) } static const struct stacktrace_ops print_trace_ops = { - .warning = print_trace_warning, - .warning_symbol = print_trace_warning_symbol, - .stack = print_trace_stack, - .address = print_trace_address, + .warning = print_trace_warning, + .warning_symbol = print_trace_warning_symbol, + .stack = print_trace_stack, + .address = print_trace_address, }; static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, char *log_lvl) + unsigned long *stack, unsigned long bp, char *log_lvl) { dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); printk("%s =======================\n", log_lvl); @@ -263,21 +268,22 @@ void show_trace(struct task_struct *task, struct pt_regs *regs, show_trace_log_lvl(task, regs, stack, bp, ""); } -static void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *sp, unsigned long bp, char *log_lvl) +static void +show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, + unsigned long *sp, unsigned long bp, char *log_lvl) { unsigned long *stack; int i; if (sp == NULL) { if (task) - sp = (unsigned long*)task->thread.sp; + sp = (unsigned long *)task->thread.sp; else sp = (unsigned long *)&sp; } stack = sp; - for(i = 0; i < kstack_depth_to_print; i++) { + for (i = 0; i < kstack_depth_to_print; i++) { if (kstack_end(stack)) break; if (i && ((i % 8) == 0)) @@ -285,6 +291,7 @@ static void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, printk("%08lx ", *stack++); } printk("\n%sCall Trace:\n", log_lvl); + show_trace_log_lvl(task, regs, sp, bp, log_lvl); } @@ -299,8 +306,8 @@ void show_stack(struct task_struct *task, unsigned long *sp) */ void dump_stack(void) { - unsigned long stack; unsigned long bp = 0; + unsigned long stack; #ifdef CONFIG_FRAME_POINTER if (!bp) @@ -312,6 +319,7 @@ void dump_stack(void) init_utsname()->release, (int)strcspn(init_utsname()->version, " "), init_utsname()->version); + show_trace(current, NULL, &stack, bp); } @@ -323,6 +331,7 @@ void show_registers(struct pt_regs *regs) print_modules(); __show_registers(regs, 0); + printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)", TASK_COMM_LEN, current->comm, task_pid_nr(current), current_thread_info(), current, task_thread_info(current)); @@ -331,10 +340,10 @@ void show_registers(struct pt_regs *regs) * time of the fault.. */ if (!user_mode_vm(regs)) { - u8 *ip; unsigned int code_prologue = code_bytes * 43 / 64; unsigned int code_len = code_bytes; unsigned char c; + u8 *ip; printk("\n" KERN_EMERG "Stack: "); show_stack_log_lvl(NULL, regs, ®s->sp, 0, KERN_EMERG); @@ -361,7 +370,7 @@ void show_registers(struct pt_regs *regs) } } printk("\n"); -} +} int is_valid_bugaddr(unsigned long ip) { @@ -377,10 +386,10 @@ int is_valid_bugaddr(unsigned long ip) static int die_counter; -int __kprobes __die(const char * str, struct pt_regs * regs, long err) +int __kprobes __die(const char *str, struct pt_regs *regs, long err) { - unsigned long sp; unsigned short ss; + unsigned long sp; printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter); #ifdef CONFIG_PREEMPT @@ -395,8 +404,8 @@ int __kprobes __die(const char * str, struct pt_regs * regs, long err) printk("\n"); if (notify_die(DIE_OOPS, str, regs, err, - current->thread.trap_no, SIGSEGV) != - NOTIFY_STOP) { + current->thread.trap_no, SIGSEGV) != NOTIFY_STOP) { + show_registers(regs); /* Executive summary in case the oops scrolled away */ sp = (unsigned long) (®s->sp); @@ -408,17 +417,18 @@ int __kprobes __die(const char * str, struct pt_regs * regs, long err) printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip); print_symbol("%s", regs->ip); printk(" SS:ESP %04x:%08lx\n", ss, sp); + return 0; - } else { - return 1; } + + return 1; } /* - * This is gone through when something in the kernel has done something bad and - * is about to be terminated. + * This is gone through when something in the kernel has done something bad + * and is about to be terminated: */ -void die(const char * str, struct pt_regs * regs, long err) +void die(const char *str, struct pt_regs *regs, long err) { static struct { raw_spinlock_t lock; @@ -440,8 +450,9 @@ void die(const char * str, struct pt_regs * regs, long err) die.lock_owner = smp_processor_id(); die.lock_owner_depth = 0; bust_spinlocks(1); - } else + } else { raw_local_irq_save(flags); + } if (++die.lock_owner_depth < 3) { report_bug(regs->ip, regs); @@ -474,19 +485,20 @@ void die(const char * str, struct pt_regs * regs, long err) do_exit(SIGSEGV); } -static inline void die_if_kernel(const char * str, struct pt_regs * regs, long err) +static inline void +die_if_kernel(const char *str, struct pt_regs *regs, long err) { if (!user_mode_vm(regs)) die(str, regs, err); } -static void __kprobes do_trap(int trapnr, int signr, char *str, int vm86, - struct pt_regs * regs, long error_code, - siginfo_t *info) +static void __kprobes +do_trap(int trapnr, int signr, char *str, int vm86, struct pt_regs *regs, + long error_code, siginfo_t *info) { struct task_struct *tsk = current; - if (regs->flags & VM_MASK) { + if (regs->flags & X86_VM_MASK) { if (vm86) goto vm86_trap; goto trap_signal; @@ -495,111 +507,112 @@ static void __kprobes do_trap(int trapnr, int signr, char *str, int vm86, if (!user_mode(regs)) goto kernel_trap; - trap_signal: { - /* - * We want error_code and trap_no set for userspace faults and - * kernelspace faults which result in die(), but not - * kernelspace faults which are fixed up. die() gives the - * process no chance to handle the signal and notice the - * kernel fault information, so that won't result in polluting - * the information about previously queued, but not yet - * delivered, faults. See also do_general_protection below. - */ - tsk->thread.error_code = error_code; - tsk->thread.trap_no = trapnr; +trap_signal: + /* + * We want error_code and trap_no set for userspace faults and + * kernelspace faults which result in die(), but not + * kernelspace faults which are fixed up. die() gives the + * process no chance to handle the signal and notice the + * kernel fault information, so that won't result in polluting + * the information about previously queued, but not yet + * delivered, faults. See also do_general_protection below. + */ + tsk->thread.error_code = error_code; + tsk->thread.trap_no = trapnr; - if (info) - force_sig_info(signr, info, tsk); - else - force_sig(signr, tsk); - return; - } + if (info) + force_sig_info(signr, info, tsk); + else + force_sig(signr, tsk); + return; - kernel_trap: { - if (!fixup_exception(regs)) { - tsk->thread.error_code = error_code; - tsk->thread.trap_no = trapnr; - die(str, regs, error_code); - } - return; +kernel_trap: + if (!fixup_exception(regs)) { + tsk->thread.error_code = error_code; + tsk->thread.trap_no = trapnr; + die(str, regs, error_code); } + return; - vm86_trap: { - int ret = handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, trapnr); - if (ret) goto trap_signal; - return; - } +vm86_trap: + if (handle_vm86_trap((struct kernel_vm86_regs *) regs, + error_code, trapnr)) + goto trap_signal; + return; } -#define DO_ERROR(trapnr, signr, str, name) \ -void do_##name(struct pt_regs * regs, long error_code) \ -{ \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ - do_trap(trapnr, signr, str, 0, regs, error_code, NULL); \ +#define DO_ERROR(trapnr, signr, str, name) \ +void do_##name(struct pt_regs *regs, long error_code) \ +{ \ + if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ + == NOTIFY_STOP) \ + return; \ + do_trap(trapnr, signr, str, 0, regs, error_code, NULL); \ } -#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr, irq) \ -void do_##name(struct pt_regs * regs, long error_code) \ -{ \ - siginfo_t info; \ - if (irq) \ - local_irq_enable(); \ - info.si_signo = signr; \ - info.si_errno = 0; \ - info.si_code = sicode; \ - info.si_addr = (void __user *)siaddr; \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ - do_trap(trapnr, signr, str, 0, regs, error_code, &info); \ +#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr, irq) \ +void do_##name(struct pt_regs *regs, long error_code) \ +{ \ + siginfo_t info; \ + if (irq) \ + local_irq_enable(); \ + info.si_signo = signr; \ + info.si_errno = 0; \ + info.si_code = sicode; \ + info.si_addr = (void __user *)siaddr; \ + if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ + == NOTIFY_STOP) \ + return; \ + do_trap(trapnr, signr, str, 0, regs, error_code, &info); \ } -#define DO_VM86_ERROR(trapnr, signr, str, name) \ -void do_##name(struct pt_regs * regs, long error_code) \ -{ \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ - do_trap(trapnr, signr, str, 1, regs, error_code, NULL); \ +#define DO_VM86_ERROR(trapnr, signr, str, name) \ +void do_##name(struct pt_regs *regs, long error_code) \ +{ \ + if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ + == NOTIFY_STOP) \ + return; \ + do_trap(trapnr, signr, str, 1, regs, error_code, NULL); \ } -#define DO_VM86_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ -void do_##name(struct pt_regs * regs, long error_code) \ -{ \ - siginfo_t info; \ - info.si_signo = signr; \ - info.si_errno = 0; \ - info.si_code = sicode; \ - info.si_addr = (void __user *)siaddr; \ - trace_hardirqs_fixup(); \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ - do_trap(trapnr, signr, str, 1, regs, error_code, &info); \ +#define DO_VM86_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ +void do_##name(struct pt_regs *regs, long error_code) \ +{ \ + siginfo_t info; \ + info.si_signo = signr; \ + info.si_errno = 0; \ + info.si_code = sicode; \ + info.si_addr = (void __user *)siaddr; \ + trace_hardirqs_fixup(); \ + if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ + == NOTIFY_STOP) \ + return; \ + do_trap(trapnr, signr, str, 1, regs, error_code, &info); \ } -DO_VM86_ERROR_INFO( 0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip) +DO_VM86_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip) #ifndef CONFIG_KPROBES -DO_VM86_ERROR( 3, SIGTRAP, "int3", int3) +DO_VM86_ERROR(3, SIGTRAP, "int3", int3) #endif -DO_VM86_ERROR( 4, SIGSEGV, "overflow", overflow) -DO_VM86_ERROR( 5, SIGSEGV, "bounds", bounds) -DO_ERROR_INFO( 6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip, 0) -DO_ERROR( 9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun) +DO_VM86_ERROR(4, SIGSEGV, "overflow", overflow) +DO_VM86_ERROR(5, SIGSEGV, "bounds", bounds) +DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip, 0) +DO_ERROR(9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun) DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS) DO_ERROR(11, SIGBUS, "segment not present", segment_not_present) DO_ERROR(12, SIGBUS, "stack segment", stack_segment) DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0, 0) DO_ERROR_INFO(32, SIGSEGV, "iret exception", iret_error, ILL_BADSTK, 0, 1) -void __kprobes do_general_protection(struct pt_regs * regs, - long error_code) +void __kprobes do_general_protection(struct pt_regs *regs, long error_code) { - int cpu = get_cpu(); - struct tss_struct *tss = &per_cpu(init_tss, cpu); - struct thread_struct *thread = ¤t->thread; + struct thread_struct *thread; + struct tss_struct *tss; + int cpu; + + cpu = get_cpu(); + tss = &per_cpu(init_tss, cpu); + thread = ¤t->thread; /* * Perform the lazy TSS's I/O bitmap copy. If the TSS has an @@ -616,19 +629,21 @@ void __kprobes do_general_protection(struct pt_regs * regs, * If the previously set map was extending to higher ports * than the current one, pad extra space with 0xff (no access). */ - if (thread->io_bitmap_max < tss->io_bitmap_max) + if (thread->io_bitmap_max < tss->io_bitmap_max) { memset((char *) tss->io_bitmap + thread->io_bitmap_max, 0xff, tss->io_bitmap_max - thread->io_bitmap_max); + } tss->io_bitmap_max = thread->io_bitmap_max; tss->x86_tss.io_bitmap_base = IO_BITMAP_OFFSET; tss->io_bitmap_owner = thread; put_cpu(); + return; } put_cpu(); - if (regs->flags & VM_MASK) + if (regs->flags & X86_VM_MASK) goto gp_in_vm86; if (!user_mode(regs)) @@ -636,6 +651,7 @@ void __kprobes do_general_protection(struct pt_regs * regs, current->thread.error_code = error_code; current->thread.trap_no = 13; + if (show_unhandled_signals && unhandled_signal(current, SIGSEGV) && printk_ratelimit()) { printk(KERN_INFO @@ -666,21 +682,24 @@ gp_in_kernel: } static __kprobes void -mem_parity_error(unsigned char reason, struct pt_regs * regs) +mem_parity_error(unsigned char reason, struct pt_regs *regs) { - printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x on " - "CPU %d.\n", reason, smp_processor_id()); - printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n"); + printk(KERN_EMERG + "Uhhuh. NMI received for unknown reason %02x on CPU %d.\n", + reason, smp_processor_id()); + + printk(KERN_EMERG + "You have some hardware problem, likely on the PCI bus.\n"); #if defined(CONFIG_EDAC) - if(edac_handler_set()) { + if (edac_handler_set()) { edac_atomic_assert_error(); return; } #endif if (panic_on_unrecovered_nmi) - panic("NMI: Not continuing"); + panic("NMI: Not continuing"); printk(KERN_EMERG "Dazed and confused, but trying to continue\n"); @@ -689,7 +708,7 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs) } static __kprobes void -io_check_error(unsigned char reason, struct pt_regs * regs) +io_check_error(unsigned char reason, struct pt_regs *regs) { unsigned long i; @@ -699,28 +718,37 @@ io_check_error(unsigned char reason, struct pt_regs * regs) /* Re-enable the IOCK line, wait for a few seconds */ reason = (reason & 0xf) | 8; outb(reason, 0x61); + i = 2000; - while (--i) udelay(1000); + while (--i) + udelay(1000); + reason &= ~8; outb(reason, 0x61); } static __kprobes void -unknown_nmi_error(unsigned char reason, struct pt_regs * regs) +unknown_nmi_error(unsigned char reason, struct pt_regs *regs) { + if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) + return; #ifdef CONFIG_MCA - /* Might actually be able to figure out what the guilty party - * is. */ - if( MCA_bus ) { + /* + * Might actually be able to figure out what the guilty party + * is: + */ + if (MCA_bus) { mca_handle_nmi(); return; } #endif - printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x on " - "CPU %d.\n", reason, smp_processor_id()); + printk(KERN_EMERG + "Uhhuh. NMI received for unknown reason %02x on CPU %d.\n", + reason, smp_processor_id()); + printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n"); if (panic_on_unrecovered_nmi) - panic("NMI: Not continuing"); + panic("NMI: Not continuing"); printk(KERN_EMERG "Dazed and confused, but trying to continue\n"); } @@ -729,14 +757,13 @@ static DEFINE_SPINLOCK(nmi_print_lock); void __kprobes die_nmi(struct pt_regs *regs, const char *msg) { - if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) == - NOTIFY_STOP) + if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) == NOTIFY_STOP) return; spin_lock(&nmi_print_lock); /* * We are in trouble anyway, lets at least try - * to get a message out. + * to get a message out: */ bust_spinlocks(1); printk(KERN_EMERG "%s", msg); @@ -747,9 +774,10 @@ void __kprobes die_nmi(struct pt_regs *regs, const char *msg) spin_unlock(&nmi_print_lock); bust_spinlocks(0); - /* If we are in kernel we are probably nested up pretty bad - * and might aswell get out now while we still can. - */ + /* + * If we are in kernel we are probably nested up pretty bad + * and might aswell get out now while we still can: + */ if (!user_mode_vm(regs)) { current->thread.trap_no = 2; crash_kexec(regs); @@ -758,14 +786,14 @@ void __kprobes die_nmi(struct pt_regs *regs, const char *msg) do_exit(SIGSEGV); } -static __kprobes void default_do_nmi(struct pt_regs * regs) +static __kprobes void default_do_nmi(struct pt_regs *regs) { unsigned char reason = 0; - /* Only the BSP gets external NMIs from the system. */ + /* Only the BSP gets external NMIs from the system: */ if (!smp_processor_id()) reason = get_nmi_reason(); - + if (!(reason & 0xc0)) { if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT) == NOTIFY_STOP) @@ -778,8 +806,10 @@ static __kprobes void default_do_nmi(struct pt_regs * regs) if (nmi_watchdog_tick(regs, reason)) return; if (!do_nmi_callback(regs, smp_processor_id())) -#endif unknown_nmi_error(reason, regs); +#else + unknown_nmi_error(reason, regs); +#endif return; } @@ -791,14 +821,14 @@ static __kprobes void default_do_nmi(struct pt_regs * regs) io_check_error(reason, regs); /* * Reassert NMI in case it became active meanwhile - * as it's edge-triggered. + * as it's edge-triggered: */ reassert_nmi(); } static int ignore_nmis; -__kprobes void do_nmi(struct pt_regs * regs, long error_code) +__kprobes void do_nmi(struct pt_regs *regs, long error_code) { int cpu; @@ -834,9 +864,12 @@ void __kprobes do_int3(struct pt_regs *regs, long error_code) if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) return; - /* This is an interrupt gate, because kprobes wants interrupts - disabled. Normal trap handlers don't. */ + /* + * This is an interrupt gate, because kprobes wants interrupts + * disabled. Normal trap handlers don't. + */ restore_interrupts(regs); + do_trap(3, SIGTRAP, "int3", 1, regs, error_code, NULL); } #endif @@ -851,7 +884,7 @@ void __kprobes do_int3(struct pt_regs *regs, long error_code) * from user space. Such code must not hold kernel locks (since it * can equally take a page fault), therefore it is safe to call * force_sig_info even though that claims and releases locks. - * + * * Code in ./signal.c ensures that the debug control register * is restored before we deliver any signal, and therefore that * user code runs with the correct debug control register even though @@ -863,10 +896,10 @@ void __kprobes do_int3(struct pt_regs *regs, long error_code) * find every occurrence of the TF bit that could be saved away even * by user code) */ -void __kprobes do_debug(struct pt_regs * regs, long error_code) +void __kprobes do_debug(struct pt_regs *regs, long error_code) { - unsigned int condition; struct task_struct *tsk = current; + unsigned int condition; trace_hardirqs_fixup(); @@ -891,7 +924,7 @@ void __kprobes do_debug(struct pt_regs * regs, long error_code) goto clear_dr7; } - if (regs->flags & VM_MASK) + if (regs->flags & X86_VM_MASK) goto debug_vm86; /* Save debug status register where ptrace can see it */ @@ -914,7 +947,8 @@ void __kprobes do_debug(struct pt_regs * regs, long error_code) /* Ok, finally something we can handle */ send_sigtrap(tsk, regs, error_code); - /* Disable additional traps. They'll be re-enabled when + /* + * Disable additional traps. They'll be re-enabled when * the signal is delivered. */ clear_dr7: @@ -927,7 +961,7 @@ debug_vm86: clear_TF_reenable: set_tsk_thread_flag(tsk, TIF_SINGLESTEP); - regs->flags &= ~TF_MASK; + regs->flags &= ~X86_EFLAGS_TF; return; } @@ -938,9 +972,10 @@ clear_TF_reenable: */ void math_error(void __user *ip) { - struct task_struct * task; + struct task_struct *task; + unsigned short cwd; + unsigned short swd; siginfo_t info; - unsigned short cwd, swd; /* * Save the info for the exception handler and clear the error. @@ -966,36 +1001,36 @@ void math_error(void __user *ip) cwd = get_fpu_cwd(task); swd = get_fpu_swd(task); switch (swd & ~cwd & 0x3f) { - case 0x000: /* No unmasked exception */ - return; - default: /* Multiple exceptions */ - break; - case 0x001: /* Invalid Op */ - /* - * swd & 0x240 == 0x040: Stack Underflow - * swd & 0x240 == 0x240: Stack Overflow - * User must clear the SF bit (0x40) if set - */ - info.si_code = FPE_FLTINV; - break; - case 0x002: /* Denormalize */ - case 0x010: /* Underflow */ - info.si_code = FPE_FLTUND; - break; - case 0x004: /* Zero Divide */ - info.si_code = FPE_FLTDIV; - break; - case 0x008: /* Overflow */ - info.si_code = FPE_FLTOVF; - break; - case 0x020: /* Precision */ - info.si_code = FPE_FLTRES; - break; + case 0x000: /* No unmasked exception */ + return; + default: /* Multiple exceptions */ + break; + case 0x001: /* Invalid Op */ + /* + * swd & 0x240 == 0x040: Stack Underflow + * swd & 0x240 == 0x240: Stack Overflow + * User must clear the SF bit (0x40) if set + */ + info.si_code = FPE_FLTINV; + break; + case 0x002: /* Denormalize */ + case 0x010: /* Underflow */ + info.si_code = FPE_FLTUND; + break; + case 0x004: /* Zero Divide */ + info.si_code = FPE_FLTDIV; + break; + case 0x008: /* Overflow */ + info.si_code = FPE_FLTOVF; + break; + case 0x020: /* Precision */ + info.si_code = FPE_FLTRES; + break; } force_sig_info(SIGFPE, &info, task); } -void do_coprocessor_error(struct pt_regs * regs, long error_code) +void do_coprocessor_error(struct pt_regs *regs, long error_code) { ignore_fpu_irq = 1; math_error((void __user *)regs->ip); @@ -1003,9 +1038,9 @@ void do_coprocessor_error(struct pt_regs * regs, long error_code) static void simd_math_error(void __user *ip) { - struct task_struct * task; - siginfo_t info; + struct task_struct *task; unsigned short mxcsr; + siginfo_t info; /* * Save the info for the exception handler and clear the error. @@ -1026,82 +1061,80 @@ static void simd_math_error(void __user *ip) */ mxcsr = get_fpu_mxcsr(task); switch (~((mxcsr & 0x1f80) >> 7) & (mxcsr & 0x3f)) { - case 0x000: - default: - break; - case 0x001: /* Invalid Op */ - info.si_code = FPE_FLTINV; - break; - case 0x002: /* Denormalize */ - case 0x010: /* Underflow */ - info.si_code = FPE_FLTUND; - break; - case 0x004: /* Zero Divide */ - info.si_code = FPE_FLTDIV; - break; - case 0x008: /* Overflow */ - info.si_code = FPE_FLTOVF; - break; - case 0x020: /* Precision */ - info.si_code = FPE_FLTRES; - break; + case 0x000: + default: + break; + case 0x001: /* Invalid Op */ + info.si_code = FPE_FLTINV; + break; + case 0x002: /* Denormalize */ + case 0x010: /* Underflow */ + info.si_code = FPE_FLTUND; + break; + case 0x004: /* Zero Divide */ + info.si_code = FPE_FLTDIV; + break; + case 0x008: /* Overflow */ + info.si_code = FPE_FLTOVF; + break; + case 0x020: /* Precision */ + info.si_code = FPE_FLTRES; + break; } force_sig_info(SIGFPE, &info, task); } -void do_simd_coprocessor_error(struct pt_regs * regs, - long error_code) +void do_simd_coprocessor_error(struct pt_regs *regs, long error_code) { if (cpu_has_xmm) { /* Handle SIMD FPU exceptions on PIII+ processors. */ ignore_fpu_irq = 1; simd_math_error((void __user *)regs->ip); - } else { - /* - * Handle strange cache flush from user space exception - * in all other cases. This is undocumented behaviour. - */ - if (regs->flags & VM_MASK) { - handle_vm86_fault((struct kernel_vm86_regs *)regs, - error_code); - return; - } - current->thread.trap_no = 19; - current->thread.error_code = error_code; - die_if_kernel("cache flush denied", regs, error_code); - force_sig(SIGSEGV, current); + return; + } + /* + * Handle strange cache flush from user space exception + * in all other cases. This is undocumented behaviour. + */ + if (regs->flags & X86_VM_MASK) { + handle_vm86_fault((struct kernel_vm86_regs *)regs, error_code); + return; } + current->thread.trap_no = 19; + current->thread.error_code = error_code; + die_if_kernel("cache flush denied", regs, error_code); + force_sig(SIGSEGV, current); } -void do_spurious_interrupt_bug(struct pt_regs * regs, - long error_code) +void do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) { #if 0 /* No need to warn about this any longer. */ - printk("Ignoring P6 Local APIC Spurious Interrupt Bug...\n"); + printk(KERN_INFO "Ignoring P6 Local APIC Spurious Interrupt Bug...\n"); #endif } -unsigned long patch_espfix_desc(unsigned long uesp, - unsigned long kesp) +unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp) { struct desc_struct *gdt = __get_cpu_var(gdt_page).gdt; unsigned long base = (kesp - uesp) & -THREAD_SIZE; unsigned long new_kesp = kesp - base; unsigned long lim_pages = (new_kesp | (THREAD_SIZE - 1)) >> PAGE_SHIFT; __u64 desc = *(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS]; + /* Set up base for espfix segment */ - desc &= 0x00f0ff0000000000ULL; - desc |= ((((__u64)base) << 16) & 0x000000ffffff0000ULL) | + desc &= 0x00f0ff0000000000ULL; + desc |= ((((__u64)base) << 16) & 0x000000ffffff0000ULL) | ((((__u64)base) << 32) & 0xff00000000000000ULL) | ((((__u64)lim_pages) << 32) & 0x000f000000000000ULL) | (lim_pages & 0xffff); *(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS] = desc; + return new_kesp; } /* - * 'math_state_restore()' saves the current math information in the + * 'math_state_restore()' saves the current math information in the * old math state array, and gets the new ones from the current task * * Careful.. There are problems with IBM-designed IRQ13 behaviour. @@ -1115,7 +1148,7 @@ asmlinkage void math_state_restore(void) struct thread_info *thread = current_thread_info(); struct task_struct *tsk = thread->task; - clts(); /* Allow maths ops (or we recurse) */ + clts(); /* Allow maths ops (or we recurse) */ if (!tsk_used_math(tsk)) init_fpu(tsk); restore_fpu(tsk); @@ -1128,53 +1161,52 @@ EXPORT_SYMBOL_GPL(math_state_restore); asmlinkage void math_emulate(long arg) { - printk(KERN_EMERG "math-emulation not enabled and no coprocessor found.\n"); - printk(KERN_EMERG "killing %s.\n",current->comm); - force_sig(SIGFPE,current); + printk(KERN_EMERG + "math-emulation not enabled and no coprocessor found.\n"); + printk(KERN_EMERG "killing %s.\n", current->comm); + force_sig(SIGFPE, current); schedule(); } #endif /* CONFIG_MATH_EMULATION */ - void __init trap_init(void) { int i; #ifdef CONFIG_EISA void __iomem *p = early_ioremap(0x0FFFD9, 4); - if (readl(p) == 'E'+('I'<<8)+('S'<<16)+('A'<<24)) { + + if (readl(p) == 'E' + ('I'<<8) + ('S'<<16) + ('A'<<24)) EISA_bus = 1; - } early_iounmap(p, 4); #endif #ifdef CONFIG_X86_LOCAL_APIC init_apic_mappings(); #endif - - set_trap_gate(0,÷_error); - set_intr_gate(1,&debug); - set_intr_gate(2,&nmi); + set_trap_gate(0, ÷_error); + set_intr_gate(1, &debug); + set_intr_gate(2, &nmi); set_system_intr_gate(3, &int3); /* int3/4 can be called from all */ - set_system_gate(4,&overflow); - set_trap_gate(5,&bounds); - set_trap_gate(6,&invalid_op); - set_trap_gate(7,&device_not_available); - set_task_gate(8,GDT_ENTRY_DOUBLEFAULT_TSS); - set_trap_gate(9,&coprocessor_segment_overrun); - set_trap_gate(10,&invalid_TSS); - set_trap_gate(11,&segment_not_present); - set_trap_gate(12,&stack_segment); - set_trap_gate(13,&general_protection); - set_intr_gate(14,&page_fault); - set_trap_gate(15,&spurious_interrupt_bug); - set_trap_gate(16,&coprocessor_error); - set_trap_gate(17,&alignment_check); + set_system_gate(4, &overflow); + set_trap_gate(5, &bounds); + set_trap_gate(6, &invalid_op); + set_trap_gate(7, &device_not_available); + set_task_gate(8, GDT_ENTRY_DOUBLEFAULT_TSS); + set_trap_gate(9, &coprocessor_segment_overrun); + set_trap_gate(10, &invalid_TSS); + set_trap_gate(11, &segment_not_present); + set_trap_gate(12, &stack_segment); + set_trap_gate(13, &general_protection); + set_intr_gate(14, &page_fault); + set_trap_gate(15, &spurious_interrupt_bug); + set_trap_gate(16, &coprocessor_error); + set_trap_gate(17, &alignment_check); #ifdef CONFIG_X86_MCE - set_trap_gate(18,&machine_check); + set_trap_gate(18, &machine_check); #endif - set_trap_gate(19,&simd_coprocessor_error); + set_trap_gate(19, &simd_coprocessor_error); /* * Verify that the FXSAVE/FXRSTOR data will be 16-byte aligned. @@ -1187,21 +1219,22 @@ void __init trap_init(void) printk("done.\n"); } if (cpu_has_xmm) { - printk(KERN_INFO "Enabling unmasked SIMD FPU exception " - "support... "); + printk(KERN_INFO + "Enabling unmasked SIMD FPU exception support... "); set_in_cr4(X86_CR4_OSXMMEXCPT); printk("done.\n"); } - set_system_gate(SYSCALL_VECTOR,&system_call); + set_system_gate(SYSCALL_VECTOR, &system_call); - /* Reserve all the builtin and the syscall vector. */ + /* Reserve all the builtin and the syscall vector: */ for (i = 0; i < FIRST_EXTERNAL_VECTOR; i++) set_bit(i, used_vectors); + set_bit(SYSCALL_VECTOR, used_vectors); /* - * Should be a barrier for any external CPU state. + * Should be a barrier for any external CPU state: */ cpu_init(); @@ -1211,6 +1244,7 @@ void __init trap_init(void) static int __init kstack_setup(char *s) { kstack_depth_to_print = simple_strtoul(s, NULL, 0); + return 1; } __setup("kstack=", kstack_setup); diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index 04546668191..79aa6fc0815 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -33,6 +33,8 @@ #include <linux/kdebug.h> #include <linux/utsname.h> +#include <mach_traps.h> + #if defined(CONFIG_EDAC) #include <linux/edac.h> #endif @@ -600,8 +602,13 @@ void die(const char * str, struct pt_regs * regs, long err) void __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic) { - unsigned long flags = oops_begin(); + unsigned long flags; + if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == + NOTIFY_STOP) + return; + + flags = oops_begin(); /* * We are in trouble anyway, lets at least try * to get a message out. @@ -806,6 +813,8 @@ io_check_error(unsigned char reason, struct pt_regs * regs) static __kprobes void unknown_nmi_error(unsigned char reason, struct pt_regs * regs) { + if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) + return; printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n", reason); printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n"); diff --git a/arch/x86/kernel/tsc_32.c b/arch/x86/kernel/tsc_32.c index d7498b34c8e..3d7e6e9fa6c 100644 --- a/arch/x86/kernel/tsc_32.c +++ b/arch/x86/kernel/tsc_32.c @@ -84,8 +84,8 @@ DEFINE_PER_CPU(unsigned long, cyc2ns); static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu) { - unsigned long flags, prev_scale, *scale; unsigned long long tsc_now, ns_now; + unsigned long flags, *scale; local_irq_save(flags); sched_clock_idle_sleep_event(); @@ -95,7 +95,6 @@ static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu) rdtscll(tsc_now); ns_now = __cycles_2_ns(tsc_now); - prev_scale = *scale; if (cpu_khz) *scale = (NSEC_PER_MSEC << CYC2NS_SCALE_FACTOR)/cpu_khz; @@ -256,9 +255,7 @@ time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data) ref_freq, freq->new); if (!(freq->flags & CPUFREQ_CONST_LOOPS)) { tsc_khz = cpu_khz; - preempt_disable(); - set_cyc2ns_scale(cpu_khz, smp_processor_id()); - preempt_enable(); + set_cyc2ns_scale(cpu_khz, freq->cpu); /* * TSC based sched_clock turns * to junk w/ cpufreq @@ -287,27 +284,14 @@ core_initcall(cpufreq_tsc); /* clock source code */ static unsigned long current_tsc_khz = 0; -static struct clocksource clocksource_tsc; -/* - * We compare the TSC to the cycle_last value in the clocksource - * structure to avoid a nasty time-warp issue. This can be observed in - * a very small window right after one CPU updated cycle_last under - * xtime lock and the other CPU reads a TSC value which is smaller - * than the cycle_last reference value due to a TSC which is slighty - * behind. This delta is nowhere else observable, but in that case it - * results in a forward time jump in the range of hours due to the - * unsigned delta calculation of the time keeping core code, which is - * necessary to support wrapping clocksources like pm timer. - */ static cycle_t read_tsc(void) { cycle_t ret; rdtscll(ret); - return ret >= clocksource_tsc.cycle_last ? - ret : clocksource_tsc.cycle_last; + return ret; } static struct clocksource clocksource_tsc = { @@ -407,13 +391,15 @@ void __init tsc_init(void) int cpu; if (!cpu_has_tsc) - goto out_no_tsc; + return; cpu_khz = calculate_cpu_khz(); tsc_khz = cpu_khz; - if (!cpu_khz) - goto out_no_tsc; + if (!cpu_khz) { + mark_tsc_unstable("could not calculate TSC khz"); + return; + } printk("Detected %lu.%03lu MHz processor.\n", (unsigned long)cpu_khz / 1000, @@ -446,9 +432,4 @@ void __init tsc_init(void) tsc_enabled = 1; clocksource_register(&clocksource_tsc); - - return; - -out_no_tsc: - setup_clear_cpu_cap(X86_FEATURE_TSC); } diff --git a/arch/x86/kernel/tsc_64.c b/arch/x86/kernel/tsc_64.c index 01fc9f0c39e..ceeba01e7f4 100644 --- a/arch/x86/kernel/tsc_64.c +++ b/arch/x86/kernel/tsc_64.c @@ -11,7 +11,6 @@ #include <asm/hpet.h> #include <asm/timex.h> #include <asm/timer.h> -#include <asm/vgtod.h> static int notsc __initdata = 0; @@ -45,8 +44,8 @@ DEFINE_PER_CPU(unsigned long, cyc2ns); static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu) { - unsigned long flags, prev_scale, *scale; unsigned long long tsc_now, ns_now; + unsigned long flags, *scale; local_irq_save(flags); sched_clock_idle_sleep_event(); @@ -56,7 +55,6 @@ static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu) rdtscll(tsc_now); ns_now = __cycles_2_ns(tsc_now); - prev_scale = *scale; if (cpu_khz) *scale = (NSEC_PER_MSEC << CYC2NS_SCALE_FACTOR)/cpu_khz; @@ -149,9 +147,7 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, mark_tsc_unstable("cpufreq changes"); } - preempt_disable(); - set_cyc2ns_scale(tsc_khz_ref, smp_processor_id()); - preempt_enable(); + set_cyc2ns_scale(tsc_khz_ref, freq->cpu); return 0; } @@ -291,34 +287,18 @@ int __init notsc_setup(char *s) __setup("notsc", notsc_setup); -static struct clocksource clocksource_tsc; -/* - * We compare the TSC to the cycle_last value in the clocksource - * structure to avoid a nasty time-warp. This can be observed in a - * very small window right after one CPU updated cycle_last under - * xtime/vsyscall_gtod lock and the other CPU reads a TSC value which - * is smaller than the cycle_last reference value due to a TSC which - * is slighty behind. This delta is nowhere else observable, but in - * that case it results in a forward time jump in the range of hours - * due to the unsigned delta calculation of the time keeping core - * code, which is necessary to support wrapping clocksources like pm - * timer. - */ +/* clock source code: */ static cycle_t read_tsc(void) { cycle_t ret = (cycle_t)get_cycles(); - - return ret >= clocksource_tsc.cycle_last ? - ret : clocksource_tsc.cycle_last; + return ret; } static cycle_t __vsyscall_fn vread_tsc(void) { cycle_t ret = (cycle_t)vget_cycles(); - - return ret >= __vsyscall_gtod_data.clock.cycle_last ? - ret : __vsyscall_gtod_data.clock.cycle_last; + return ret; } static struct clocksource clocksource_tsc = { diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c index 738c2104df3..38f566fa27d 100644 --- a/arch/x86/kernel/vm86_32.c +++ b/arch/x86/kernel/vm86_32.c @@ -64,7 +64,7 @@ #define KVM86 ((struct kernel_vm86_struct *)regs) -#define VMPI KVM86->vm86plus +#define VMPI KVM86->vm86plus /* @@ -81,7 +81,7 @@ #define VFLAGS (*(unsigned short *)&(current->thread.v86flags)) #define VEFLAGS (current->thread.v86flags) -#define set_flags(X,new,mask) \ +#define set_flags(X, new, mask) \ ((X) = ((X) & ~(mask)) | ((new) & (mask))) #define SAFE_MASK (0xDD5) @@ -93,8 +93,10 @@ static int copy_vm86_regs_to_user(struct vm86_regs __user *user, { int ret = 0; - /* kernel_vm86_regs is missing gs, so copy everything up to - (but not including) orig_eax, and then rest including orig_eax. */ + /* + * kernel_vm86_regs is missing gs, so copy everything up to + * (but not including) orig_eax, and then rest including orig_eax. + */ ret += copy_to_user(user, regs, offsetof(struct kernel_vm86_regs, pt.orig_ax)); ret += copy_to_user(&user->orig_eax, ®s->pt.orig_ax, sizeof(struct kernel_vm86_regs) - @@ -120,7 +122,7 @@ static int copy_vm86_regs_from_user(struct kernel_vm86_regs *regs, return ret; } -struct pt_regs * save_v86_state(struct kernel_vm86_regs * regs) +struct pt_regs *save_v86_state(struct kernel_vm86_regs *regs) { struct tss_struct *tss; struct pt_regs *ret; @@ -137,9 +139,9 @@ struct pt_regs * save_v86_state(struct kernel_vm86_regs * regs) printk("no vm86_info: BAD\n"); do_exit(SIGSEGV); } - set_flags(regs->pt.flags, VEFLAGS, VIF_MASK | current->thread.v86mask); - tmp = copy_vm86_regs_to_user(¤t->thread.vm86_info->regs,regs); - tmp += put_user(current->thread.screen_bitmap,¤t->thread.vm86_info->screen_bitmap); + set_flags(regs->pt.flags, VEFLAGS, X86_EFLAGS_VIF | current->thread.v86mask); + tmp = copy_vm86_regs_to_user(¤t->thread.vm86_info->regs, regs); + tmp += put_user(current->thread.screen_bitmap, ¤t->thread.vm86_info->screen_bitmap); if (tmp) { printk("vm86: could not access userspace vm86_info\n"); do_exit(SIGSEGV); @@ -237,20 +239,21 @@ asmlinkage int sys_vm86(struct pt_regs regs) tsk = current; switch (regs.bx) { - case VM86_REQUEST_IRQ: - case VM86_FREE_IRQ: - case VM86_GET_IRQ_BITS: - case VM86_GET_AND_RESET_IRQ: - ret = do_vm86_irq_handling(regs.bx, (int)regs.cx); - goto out; - case VM86_PLUS_INSTALL_CHECK: - /* NOTE: on old vm86 stuff this will return the error - from access_ok(), because the subfunction is - interpreted as (invalid) address to vm86_struct. - So the installation check works. - */ - ret = 0; - goto out; + case VM86_REQUEST_IRQ: + case VM86_FREE_IRQ: + case VM86_GET_IRQ_BITS: + case VM86_GET_AND_RESET_IRQ: + ret = do_vm86_irq_handling(regs.bx, (int)regs.cx); + goto out; + case VM86_PLUS_INSTALL_CHECK: + /* + * NOTE: on old vm86 stuff this will return the error + * from access_ok(), because the subfunction is + * interpreted as (invalid) address to vm86_struct. + * So the installation check works. + */ + ret = 0; + goto out; } /* we come here only for functions VM86_ENTER, VM86_ENTER_NO_BYPASS */ @@ -296,21 +299,21 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk VEFLAGS = info->regs.pt.flags; info->regs.pt.flags &= SAFE_MASK; info->regs.pt.flags |= info->regs32->flags & ~SAFE_MASK; - info->regs.pt.flags |= VM_MASK; + info->regs.pt.flags |= X86_VM_MASK; switch (info->cpu_type) { - case CPU_286: - tsk->thread.v86mask = 0; - break; - case CPU_386: - tsk->thread.v86mask = NT_MASK | IOPL_MASK; - break; - case CPU_486: - tsk->thread.v86mask = AC_MASK | NT_MASK | IOPL_MASK; - break; - default: - tsk->thread.v86mask = ID_MASK | AC_MASK | NT_MASK | IOPL_MASK; - break; + case CPU_286: + tsk->thread.v86mask = 0; + break; + case CPU_386: + tsk->thread.v86mask = X86_EFLAGS_NT | X86_EFLAGS_IOPL; + break; + case CPU_486: + tsk->thread.v86mask = X86_EFLAGS_AC | X86_EFLAGS_NT | X86_EFLAGS_IOPL; + break; + default: + tsk->thread.v86mask = X86_EFLAGS_ID | X86_EFLAGS_AC | X86_EFLAGS_NT | X86_EFLAGS_IOPL; + break; } /* @@ -346,9 +349,9 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk /* we never return here */ } -static inline void return_to_32bit(struct kernel_vm86_regs * regs16, int retval) +static inline void return_to_32bit(struct kernel_vm86_regs *regs16, int retval) { - struct pt_regs * regs32; + struct pt_regs *regs32; regs32 = save_v86_state(regs16); regs32->ax = retval; @@ -358,29 +361,30 @@ static inline void return_to_32bit(struct kernel_vm86_regs * regs16, int retval) : : "r" (regs32), "r" (current_thread_info())); } -static inline void set_IF(struct kernel_vm86_regs * regs) +static inline void set_IF(struct kernel_vm86_regs *regs) { - VEFLAGS |= VIF_MASK; - if (VEFLAGS & VIP_MASK) + VEFLAGS |= X86_EFLAGS_VIF; + if (VEFLAGS & X86_EFLAGS_VIP) return_to_32bit(regs, VM86_STI); } -static inline void clear_IF(struct kernel_vm86_regs * regs) +static inline void clear_IF(struct kernel_vm86_regs *regs) { - VEFLAGS &= ~VIF_MASK; + VEFLAGS &= ~X86_EFLAGS_VIF; } -static inline void clear_TF(struct kernel_vm86_regs * regs) +static inline void clear_TF(struct kernel_vm86_regs *regs) { - regs->pt.flags &= ~TF_MASK; + regs->pt.flags &= ~X86_EFLAGS_TF; } -static inline void clear_AC(struct kernel_vm86_regs * regs) +static inline void clear_AC(struct kernel_vm86_regs *regs) { - regs->pt.flags &= ~AC_MASK; + regs->pt.flags &= ~X86_EFLAGS_AC; } -/* It is correct to call set_IF(regs) from the set_vflags_* +/* + * It is correct to call set_IF(regs) from the set_vflags_* * functions. However someone forgot to call clear_IF(regs) * in the opposite case. * After the command sequence CLI PUSHF STI POPF you should @@ -391,41 +395,41 @@ static inline void clear_AC(struct kernel_vm86_regs * regs) * [KD] */ -static inline void set_vflags_long(unsigned long flags, struct kernel_vm86_regs * regs) +static inline void set_vflags_long(unsigned long flags, struct kernel_vm86_regs *regs) { set_flags(VEFLAGS, flags, current->thread.v86mask); set_flags(regs->pt.flags, flags, SAFE_MASK); - if (flags & IF_MASK) + if (flags & X86_EFLAGS_IF) set_IF(regs); else clear_IF(regs); } -static inline void set_vflags_short(unsigned short flags, struct kernel_vm86_regs * regs) +static inline void set_vflags_short(unsigned short flags, struct kernel_vm86_regs *regs) { set_flags(VFLAGS, flags, current->thread.v86mask); set_flags(regs->pt.flags, flags, SAFE_MASK); - if (flags & IF_MASK) + if (flags & X86_EFLAGS_IF) set_IF(regs); else clear_IF(regs); } -static inline unsigned long get_vflags(struct kernel_vm86_regs * regs) +static inline unsigned long get_vflags(struct kernel_vm86_regs *regs) { unsigned long flags = regs->pt.flags & RETURN_MASK; - if (VEFLAGS & VIF_MASK) - flags |= IF_MASK; - flags |= IOPL_MASK; + if (VEFLAGS & X86_EFLAGS_VIF) + flags |= X86_EFLAGS_IF; + flags |= X86_EFLAGS_IOPL; return flags | (VEFLAGS & current->thread.v86mask); } -static inline int is_revectored(int nr, struct revectored_struct * bitmap) +static inline int is_revectored(int nr, struct revectored_struct *bitmap) { __asm__ __volatile__("btl %2,%1\n\tsbbl %0,%0" :"=r" (nr) - :"m" (*bitmap),"r" (nr)); + :"m" (*bitmap), "r" (nr)); return nr; } @@ -437,7 +441,7 @@ static inline int is_revectored(int nr, struct revectored_struct * bitmap) ptr--; \ if (put_user(__val, base + ptr) < 0) \ goto err_label; \ - } while(0) + } while (0) #define pushw(base, ptr, val, err_label) \ do { \ @@ -448,7 +452,7 @@ static inline int is_revectored(int nr, struct revectored_struct * bitmap) ptr--; \ if (put_user(val_byte(__val, 0), base + ptr) < 0) \ goto err_label; \ - } while(0) + } while (0) #define pushl(base, ptr, val, err_label) \ do { \ @@ -465,7 +469,7 @@ static inline int is_revectored(int nr, struct revectored_struct * bitmap) ptr--; \ if (put_user(val_byte(__val, 0), base + ptr) < 0) \ goto err_label; \ - } while(0) + } while (0) #define popb(base, ptr, err_label) \ ({ \ @@ -512,7 +516,7 @@ static inline int is_revectored(int nr, struct revectored_struct * bitmap) * in userspace is always better than an Oops anyway.) [KD] */ static void do_int(struct kernel_vm86_regs *regs, int i, - unsigned char __user * ssp, unsigned short sp) + unsigned char __user *ssp, unsigned short sp) { unsigned long __user *intr_ptr; unsigned long segoffs; @@ -521,7 +525,7 @@ static void do_int(struct kernel_vm86_regs *regs, int i, goto cannot_handle; if (is_revectored(i, &KVM86->int_revectored)) goto cannot_handle; - if (i==0x21 && is_revectored(AH(regs),&KVM86->int21_revectored)) + if (i == 0x21 && is_revectored(AH(regs), &KVM86->int21_revectored)) goto cannot_handle; intr_ptr = (unsigned long __user *) (i << 2); if (get_user(segoffs, intr_ptr)) @@ -543,30 +547,23 @@ cannot_handle: return_to_32bit(regs, VM86_INTx + (i << 8)); } -int handle_vm86_trap(struct kernel_vm86_regs * regs, long error_code, int trapno) +int handle_vm86_trap(struct kernel_vm86_regs *regs, long error_code, int trapno) { if (VMPI.is_vm86pus) { - if ( (trapno==3) || (trapno==1) ) + if ((trapno == 3) || (trapno == 1)) return_to_32bit(regs, VM86_TRAP + (trapno << 8)); do_int(regs, trapno, (unsigned char __user *) (regs->pt.ss << 4), SP(regs)); return 0; } - if (trapno !=1) + if (trapno != 1) return 1; /* we let this handle by the calling routine */ - if (current->ptrace & PT_PTRACED) { - unsigned long flags; - spin_lock_irqsave(¤t->sighand->siglock, flags); - sigdelset(¤t->blocked, SIGTRAP); - recalc_sigpending(); - spin_unlock_irqrestore(¤t->sighand->siglock, flags); - } - send_sig(SIGTRAP, current, 1); current->thread.trap_no = trapno; current->thread.error_code = error_code; + force_sig(SIGTRAP, current); return 0; } -void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code) +void handle_vm86_fault(struct kernel_vm86_regs *regs, long error_code) { unsigned char opcode; unsigned char __user *csp; @@ -576,11 +573,11 @@ void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code) #define CHECK_IF_IN_TRAP \ if (VMPI.vm86dbg_active && VMPI.vm86dbg_TFpendig) \ - newflags |= TF_MASK + newflags |= X86_EFLAGS_TF #define VM86_FAULT_RETURN do { \ - if (VMPI.force_return_for_pic && (VEFLAGS & (IF_MASK | VIF_MASK))) \ + if (VMPI.force_return_for_pic && (VEFLAGS & (X86_EFLAGS_IF | X86_EFLAGS_VIF))) \ return_to_32bit(regs, VM86_PICRETURN); \ - if (orig_flags & TF_MASK) \ + if (orig_flags & X86_EFLAGS_TF) \ handle_vm86_trap(regs, 0, 1); \ return; } while (0) @@ -595,17 +592,17 @@ void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code) pref_done = 0; do { switch (opcode = popb(csp, ip, simulate_sigsegv)) { - case 0x66: /* 32-bit data */ data32=1; break; - case 0x67: /* 32-bit address */ break; - case 0x2e: /* CS */ break; - case 0x3e: /* DS */ break; - case 0x26: /* ES */ break; - case 0x36: /* SS */ break; - case 0x65: /* GS */ break; - case 0x64: /* FS */ break; - case 0xf2: /* repnz */ break; - case 0xf3: /* rep */ break; - default: pref_done = 1; + case 0x66: /* 32-bit data */ data32 = 1; break; + case 0x67: /* 32-bit address */ break; + case 0x2e: /* CS */ break; + case 0x3e: /* DS */ break; + case 0x26: /* ES */ break; + case 0x36: /* SS */ break; + case 0x65: /* GS */ break; + case 0x64: /* FS */ break; + case 0xf2: /* repnz */ break; + case 0xf3: /* rep */ break; + default: pref_done = 1; } } while (!pref_done); @@ -628,7 +625,7 @@ void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code) { unsigned long newflags; if (data32) { - newflags=popl(ssp, sp, simulate_sigsegv); + newflags = popl(ssp, sp, simulate_sigsegv); SP(regs) += 4; } else { newflags = popw(ssp, sp, simulate_sigsegv); @@ -636,20 +633,20 @@ void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code) } IP(regs) = ip; CHECK_IF_IN_TRAP; - if (data32) { + if (data32) set_vflags_long(newflags, regs); - } else { + else set_vflags_short(newflags, regs); - } + VM86_FAULT_RETURN; } /* int xx */ case 0xcd: { - int intno=popb(csp, ip, simulate_sigsegv); + int intno = popb(csp, ip, simulate_sigsegv); IP(regs) = ip; if (VMPI.vm86dbg_active) { - if ( (1 << (intno &7)) & VMPI.vm86dbg_intxxtab[intno >> 3] ) + if ((1 << (intno & 7)) & VMPI.vm86dbg_intxxtab[intno >> 3]) return_to_32bit(regs, VM86_INTx + (intno << 8)); } do_int(regs, intno, ssp, sp); @@ -663,9 +660,9 @@ void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code) unsigned long newcs; unsigned long newflags; if (data32) { - newip=popl(ssp, sp, simulate_sigsegv); - newcs=popl(ssp, sp, simulate_sigsegv); - newflags=popl(ssp, sp, simulate_sigsegv); + newip = popl(ssp, sp, simulate_sigsegv); + newcs = popl(ssp, sp, simulate_sigsegv); + newflags = popl(ssp, sp, simulate_sigsegv); SP(regs) += 12; } else { newip = popw(ssp, sp, simulate_sigsegv); @@ -734,18 +731,18 @@ static struct vm86_irqs { static DEFINE_SPINLOCK(irqbits_lock); static int irqbits; -#define ALLOWED_SIGS ( 1 /* 0 = don't send a signal */ \ +#define ALLOWED_SIGS (1 /* 0 = don't send a signal */ \ | (1 << SIGUSR1) | (1 << SIGUSR2) | (1 << SIGIO) | (1 << SIGURG) \ - | (1 << SIGUNUSED) ) - + | (1 << SIGUNUSED)) + static irqreturn_t irq_handler(int intno, void *dev_id) { int irq_bit; unsigned long flags; - spin_lock_irqsave(&irqbits_lock, flags); + spin_lock_irqsave(&irqbits_lock, flags); irq_bit = 1 << intno; - if ((irqbits & irq_bit) || ! vm86_irqs[intno].tsk) + if ((irqbits & irq_bit) || !vm86_irqs[intno].tsk) goto out; irqbits |= irq_bit; if (vm86_irqs[intno].sig) @@ -759,7 +756,7 @@ static irqreturn_t irq_handler(int intno, void *dev_id) return IRQ_HANDLED; out: - spin_unlock_irqrestore(&irqbits_lock, flags); + spin_unlock_irqrestore(&irqbits_lock, flags); return IRQ_NONE; } @@ -770,9 +767,9 @@ static inline void free_vm86_irq(int irqnumber) free_irq(irqnumber, NULL); vm86_irqs[irqnumber].tsk = NULL; - spin_lock_irqsave(&irqbits_lock, flags); + spin_lock_irqsave(&irqbits_lock, flags); irqbits &= ~(1 << irqnumber); - spin_unlock_irqrestore(&irqbits_lock, flags); + spin_unlock_irqrestore(&irqbits_lock, flags); } void release_vm86_irqs(struct task_struct *task) @@ -788,10 +785,10 @@ static inline int get_and_reset_irq(int irqnumber) int bit; unsigned long flags; int ret = 0; - + if (invalid_vm86_irq(irqnumber)) return 0; if (vm86_irqs[irqnumber].tsk != current) return 0; - spin_lock_irqsave(&irqbits_lock, flags); + spin_lock_irqsave(&irqbits_lock, flags); bit = irqbits & (1 << irqnumber); irqbits &= ~bit; if (bit) { @@ -799,7 +796,7 @@ static inline int get_and_reset_irq(int irqnumber) ret = 1; } - spin_unlock_irqrestore(&irqbits_lock, flags); + spin_unlock_irqrestore(&irqbits_lock, flags); return ret; } diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S index 2ffa9656fe7..ce5ed083a1e 100644 --- a/arch/x86/kernel/vmlinux_32.lds.S +++ b/arch/x86/kernel/vmlinux_32.lds.S @@ -149,6 +149,11 @@ SECTIONS *(.con_initcall.init) __con_initcall_end = .; } + .x86cpuvendor.init : AT(ADDR(.x86cpuvendor.init) - LOAD_OFFSET) { + __x86cpuvendor_start = .; + *(.x86cpuvendor.init) + __x86cpuvendor_end = .; + } SECURITY_INIT . = ALIGN(4); .altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) { diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index fab13229973..b7ab3c335fa 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -177,6 +177,11 @@ SECTIONS *(.con_initcall.init) } __con_initcall_end = .; + __x86cpuvendor_start = .; + .x86cpuvendor.init : AT(ADDR(.x86cpuvendor.init) - LOAD_OFFSET) { + *(.x86cpuvendor.init) + } + __x86cpuvendor_end = .; SECURITY_INIT . = ALIGN(8); @@ -247,3 +252,9 @@ SECTIONS DWARF_DEBUG } + +/* + * Build-time check on the image size: + */ +ASSERT((_end - _text <= KERNEL_IMAGE_SIZE), + "kernel image bigger than KERNEL_IMAGE_SIZE") diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c index d971210a6d3..caf2a26f5cf 100644 --- a/arch/x86/kernel/vsmp_64.c +++ b/arch/x86/kernel/vsmp_64.c @@ -8,6 +8,8 @@ * * Ravikiran Thirumalai <kiran@scalemp.com>, * Shai Fultheim <shai@scalemp.com> + * Paravirt ops integration: Glauber de Oliveira Costa <gcosta@redhat.com>, + * Ravikiran Thirumalai <kiran@scalemp.com> */ #include <linux/init.h> @@ -15,38 +17,137 @@ #include <linux/pci_regs.h> #include <asm/pci-direct.h> #include <asm/io.h> +#include <asm/paravirt.h> -static int __init vsmp_init(void) +#if defined CONFIG_PCI && defined CONFIG_PARAVIRT +/* + * Interrupt control on vSMPowered systems: + * ~AC is a shadow of IF. If IF is 'on' AC should be 'off' + * and vice versa. + */ + +static unsigned long vsmp_save_fl(void) { - void *address; - unsigned int cap, ctl; + unsigned long flags = native_save_fl(); - if (!early_pci_allowed()) - return 0; + if (!(flags & X86_EFLAGS_IF) || (flags & X86_EFLAGS_AC)) + flags &= ~X86_EFLAGS_IF; + return flags; +} - /* Check if we are running on a ScaleMP vSMP box */ - if ((read_pci_config_16(0, 0x1f, 0, PCI_VENDOR_ID) != - PCI_VENDOR_ID_SCALEMP) || - (read_pci_config_16(0, 0x1f, 0, PCI_DEVICE_ID) != - PCI_DEVICE_ID_SCALEMP_VSMP_CTL)) - return 0; +static void vsmp_restore_fl(unsigned long flags) +{ + if (flags & X86_EFLAGS_IF) + flags &= ~X86_EFLAGS_AC; + else + flags |= X86_EFLAGS_AC; + native_restore_fl(flags); +} + +static void vsmp_irq_disable(void) +{ + unsigned long flags = native_save_fl(); + + native_restore_fl((flags & ~X86_EFLAGS_IF) | X86_EFLAGS_AC); +} + +static void vsmp_irq_enable(void) +{ + unsigned long flags = native_save_fl(); + + native_restore_fl((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC)); +} + +static unsigned __init vsmp_patch(u8 type, u16 clobbers, void *ibuf, + unsigned long addr, unsigned len) +{ + switch (type) { + case PARAVIRT_PATCH(pv_irq_ops.irq_enable): + case PARAVIRT_PATCH(pv_irq_ops.irq_disable): + case PARAVIRT_PATCH(pv_irq_ops.save_fl): + case PARAVIRT_PATCH(pv_irq_ops.restore_fl): + return paravirt_patch_default(type, clobbers, ibuf, addr, len); + default: + return native_patch(type, clobbers, ibuf, addr, len); + } + +} + +static void __init set_vsmp_pv_ops(void) +{ + void *address; + unsigned int cap, ctl, cfg; /* set vSMP magic bits to indicate vSMP capable kernel */ - address = ioremap(read_pci_config(0, 0x1f, 0, PCI_BASE_ADDRESS_0), 8); + cfg = read_pci_config(0, 0x1f, 0, PCI_BASE_ADDRESS_0); + address = early_ioremap(cfg, 8); cap = readl(address); ctl = readl(address + 4); printk(KERN_INFO "vSMP CTL: capabilities:0x%08x control:0x%08x\n", cap, ctl); if (cap & ctl & (1 << 4)) { - /* Turn on vSMP IRQ fastpath handling (see system.h) */ + /* Setup irq ops and turn on vSMP IRQ fastpath handling */ + pv_irq_ops.irq_disable = vsmp_irq_disable; + pv_irq_ops.irq_enable = vsmp_irq_enable; + pv_irq_ops.save_fl = vsmp_save_fl; + pv_irq_ops.restore_fl = vsmp_restore_fl; + pv_init_ops.patch = vsmp_patch; + ctl &= ~(1 << 4); writel(ctl, address + 4); ctl = readl(address + 4); printk(KERN_INFO "vSMP CTL: control set to:0x%08x\n", ctl); } - iounmap(address); + early_iounmap(address, 8); +} +#else +static void __init set_vsmp_pv_ops(void) +{ +} +#endif + +#ifdef CONFIG_PCI +static int is_vsmp = -1; + +static void __init detect_vsmp_box(void) +{ + is_vsmp = 0; + + if (!early_pci_allowed()) + return; + + /* Check if we are running on a ScaleMP vSMPowered box */ + if (read_pci_config(0, 0x1f, 0, PCI_VENDOR_ID) == + (PCI_VENDOR_ID_SCALEMP | (PCI_DEVICE_ID_SCALEMP_VSMP_CTL << 16))) + is_vsmp = 1; +} + +int is_vsmp_box(void) +{ + if (is_vsmp != -1) + return is_vsmp; + else { + WARN_ON_ONCE(1); + return 0; + } +} +#else +static int __init detect_vsmp_box(void) +{ +} +int is_vsmp_box(void) +{ return 0; } +#endif -core_initcall(vsmp_init); +void __init vsmp_init(void) +{ + detect_vsmp_box(); + if (!is_vsmp_box()) + return; + + set_vsmp_pv_ops(); + return; +} diff --git a/arch/x86/kernel/x8664_ksyms_64.c b/arch/x86/kernel/x8664_ksyms_64.c index a66e9c1a053..58882f9f263 100644 --- a/arch/x86/kernel/x8664_ksyms_64.c +++ b/arch/x86/kernel/x8664_ksyms_64.c @@ -4,7 +4,6 @@ #include <linux/module.h> #include <linux/smp.h> -#include <asm/semaphore.h> #include <asm/processor.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -12,11 +11,6 @@ EXPORT_SYMBOL(kernel_thread); -EXPORT_SYMBOL(__down_failed); -EXPORT_SYMBOL(__down_failed_interruptible); -EXPORT_SYMBOL(__down_failed_trylock); -EXPORT_SYMBOL(__up_wakeup); - EXPORT_SYMBOL(__get_user_1); EXPORT_SYMBOL(__get_user_2); EXPORT_SYMBOL(__get_user_4); @@ -35,15 +29,17 @@ EXPORT_SYMBOL(__copy_from_user_inatomic); EXPORT_SYMBOL(copy_page); EXPORT_SYMBOL(clear_page); -/* Export string functions. We normally rely on gcc builtin for most of these, - but gcc sometimes decides not to inline them. */ +/* + * Export string functions. We normally rely on gcc builtin for most of these, + * but gcc sometimes decides not to inline them. + */ #undef memcpy #undef memset #undef memmove -extern void * memset(void *,int,__kernel_size_t); -extern void * memcpy(void *,const void *,__kernel_size_t); -extern void * __memcpy(void *,const void *,__kernel_size_t); +extern void *memset(void *, int, __kernel_size_t); +extern void *memcpy(void *, const void *, __kernel_size_t); +extern void *__memcpy(void *, const void *, __kernel_size_t); EXPORT_SYMBOL(memset); EXPORT_SYMBOL(memcpy); |