diff options
Diffstat (limited to 'arch/x86_64')
52 files changed, 1598 insertions, 434 deletions
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 0f430d9d363..d09437b5c48 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -207,33 +207,6 @@ config SMP If you don't know what to do here, say N. -config PREEMPT - bool "Preemptible Kernel" - ---help--- - This option reduces the latency of the kernel when reacting to - real-time or interactive events by allowing a low priority process to - be preempted even if it is in kernel mode executing a system call. - This allows applications to run more reliably even when the system is - under load. On contrary it may also break your drivers and add - priority inheritance problems to your system. Don't select it if - you rely on a stable system or have slightly obscure hardware. - It's also not very well tested on x86-64 currently. - You have been warned. - - Say Y here if you are feeling brave and building a kernel for a - desktop, embedded or real-time system. Say N if you are unsure. - -config PREEMPT_BKL - bool "Preempt The Big Kernel Lock" - depends on PREEMPT - default y - help - This option reduces the latency of the kernel by making the - big kernel lock preemptible. - - Say Y here if you are building a kernel for a desktop system. - Say N if you are unsure. - config SCHED_SMT bool "SMT (Hyperthreading) scheduler support" depends on SMP @@ -244,6 +217,8 @@ config SCHED_SMT cost of slightly increased overhead in some places. If unsure say N here. +source "kernel/Kconfig.preempt" + config K8_NUMA bool "K8 NUMA support" select NUMA @@ -265,7 +240,7 @@ config NUMA_EMU into virtual nodes when booted with "numa=fake=N", where N is the number of nodes. This is only useful for debugging. -config DISCONTIGMEM +config ARCH_DISCONTIGMEM_ENABLE bool depends on NUMA default y @@ -274,6 +249,27 @@ config NUMA bool default n +config ARCH_DISCONTIGMEM_ENABLE + def_bool y + depends on NUMA + +config ARCH_DISCONTIGMEM_DEFAULT + def_bool y + depends on NUMA + +config ARCH_SPARSEMEM_ENABLE + def_bool y + depends on NUMA + +config ARCH_FLATMEM_ENABLE + def_bool y + depends on !NUMA + +source "mm/Kconfig" + +config HAVE_ARCH_EARLY_PFN_TO_NID + def_bool y + config HAVE_DEC_LOCK bool depends on SMP @@ -292,6 +288,15 @@ config NR_CPUS This is purely to save memory - each supported CPU requires memory in the static kernel configuration. +config HOTPLUG_CPU + bool "Support for hot-pluggable CPUs (EXPERIMENTAL)" + depends on SMP && HOTPLUG && EXPERIMENTAL + help + Say Y here to experiment with turning CPUs off and on. CPUs + can be controlled through /sys/devices/system/cpu/cpu#. + Say N if you want to disable CPU hotplug. + + config HPET_TIMER bool default y @@ -305,6 +310,7 @@ config HPET_TIMER config X86_PM_TIMER bool "PM timer" + depends on ACPI default y help Support the ACPI PM timer for time keeping. This is slow, @@ -363,6 +369,34 @@ config X86_MCE_INTEL Additional support for intel specific MCE features such as the thermal monitor. +config PHYSICAL_START + hex "Physical address where the kernel is loaded" if EMBEDDED + default "0x100000" + help + This gives the physical address where the kernel is loaded. + Primarily used in the case of kexec on panic where the + fail safe kernel needs to run at a different address than + the panic-ed kernel. + + Don't change this unless you know what you are doing. + +config KEXEC + bool "kexec system call (EXPERIMENTAL)" + depends on EXPERIMENTAL + help + kexec is a system call that implements the ability to shutdown your + current kernel, and to start another kernel. It is like a reboot + but it is indepedent of the system firmware. And like a reboot + you can start any kernel with it, not just Linux. + + The name comes from the similiarity to the exec system call. + + It is an ongoing process to be certain the hardware in a machine + is properly shutdown, so do not be surprised if this code does not + initially work for you. It may help to enable device hotplugging + support. As of this writing the exact hardware interface is + strongly in flux, so no good recommendation can be made. + config SECCOMP bool "Enable seccomp to safely compute untrusted bytecode" depends on PROC_FS @@ -380,6 +414,8 @@ config SECCOMP If unsure, say Y. Only embedded should say N here. +source kernel/Kconfig.hz + endmenu # diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index 6f90c246c41..8a73794f9b9 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -35,7 +35,7 @@ export IA32_CC IA32_LD IA32_AS IA32_OBJCOPY IA32_CPP LDFLAGS := -m elf_x86_64 OBJCOPYFLAGS := -O binary -R .note -R .comment -S -LDFLAGS_vmlinux := -e stext +LDFLAGS_vmlinux := CHECKFLAGS += -D__x86_64__ -m64 diff --git a/arch/x86_64/boot/compressed/head.S b/arch/x86_64/boot/compressed/head.S index 27264dbd575..6f55565e4d4 100644 --- a/arch/x86_64/boot/compressed/head.S +++ b/arch/x86_64/boot/compressed/head.S @@ -2,8 +2,6 @@ * linux/boot/head.S * * Copyright (C) 1991, 1992, 1993 Linus Torvalds - * - * $Id: head.S,v 1.3 2001/04/20 00:59:28 ak Exp $ */ /* @@ -21,13 +19,14 @@ */ /* - * High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996 + * High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996 */ .code32 .text #include <linux/linkage.h> #include <asm/segment.h> +#include <asm/page.h> .code32 .globl startup_32 @@ -77,7 +76,7 @@ startup_32: jnz 3f addl $8,%esp xorl %ebx,%ebx - ljmp $(__KERNEL_CS), $0x100000 + ljmp $(__KERNEL_CS), $__PHYSICAL_START /* * We come here, if we were loaded high. @@ -103,7 +102,7 @@ startup_32: popl %ecx # lcount popl %edx # high_buffer_start popl %eax # hcount - movl $0x100000,%edi + movl $__PHYSICAL_START,%edi cli # make sure we don't get interrupted ljmp $(__KERNEL_CS), $0x1000 # and jump to the move routine @@ -128,7 +127,7 @@ move_routine_start: movsl movl %ebx,%esi # Restore setup pointer xorl %ebx,%ebx - ljmp $(__KERNEL_CS), $0x100000 + ljmp $(__KERNEL_CS), $__PHYSICAL_START move_routine_end: diff --git a/arch/x86_64/boot/compressed/misc.c b/arch/x86_64/boot/compressed/misc.c index c8b9216f9e6..b38d5b8b5fb 100644 --- a/arch/x86_64/boot/compressed/misc.c +++ b/arch/x86_64/boot/compressed/misc.c @@ -11,6 +11,7 @@ #include "miscsetup.h" #include <asm/io.h> +#include <asm/page.h> /* * gzip declarations @@ -92,8 +93,11 @@ static unsigned long output_ptr = 0; static void *malloc(int size); static void free(void *where); +void* memset(void* s, int c, unsigned n); +void* memcpy(void* dest, const void* src, unsigned n); + static void putstr(const char *); - + extern int end; static long free_mem_ptr = (long)&end; static long free_mem_end_ptr; @@ -284,7 +288,7 @@ void setup_normal_output_buffer(void) #else if ((ALT_MEM_K > EXT_MEM_K ? ALT_MEM_K : EXT_MEM_K) < 1024) error("Less than 2MB of memory"); #endif - output_data = (char *)0x100000; /* Points to 1M */ + output_data = (char *)__PHYSICAL_START; /* Normally Points to 1M */ free_mem_end_ptr = (long)real_mode; } @@ -307,8 +311,8 @@ void setup_output_buffer_if_we_run_high(struct moveparams *mv) low_buffer_size = low_buffer_end - LOW_BUFFER_START; high_loaded = 1; free_mem_end_ptr = (long)high_buffer_start; - if ( (0x100000 + low_buffer_size) > ((ulg)high_buffer_start)) { - high_buffer_start = (uch *)(0x100000 + low_buffer_size); + if ( (__PHYSICAL_START + low_buffer_size) > ((ulg)high_buffer_start)) { + high_buffer_start = (uch *)(__PHYSICAL_START + low_buffer_size); mv->hcount = 0; /* say: we need not to move high_buffer */ } else mv->hcount = -1; diff --git a/arch/x86_64/boot/install.sh b/arch/x86_64/boot/install.sh index 90f2452b3b9..198af15a775 100644 --- a/arch/x86_64/boot/install.sh +++ b/arch/x86_64/boot/install.sh @@ -1,6 +1,6 @@ #!/bin/sh # -# arch/i386/boot/install.sh +# arch/x86_64/boot/install.sh # # 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 @@ -21,8 +21,8 @@ # User may have a custom install script -if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi -if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi +if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi +if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi # Default install - same as make zlilo diff --git a/arch/x86_64/boot/setup.S b/arch/x86_64/boot/setup.S index 75d4d2ad93b..ff58b2832b7 100644 --- a/arch/x86_64/boot/setup.S +++ b/arch/x86_64/boot/setup.S @@ -33,7 +33,7 @@ * Transcribed from Intel (as86) -> AT&T (gas) by Chris Noe, May 1999. * <stiker@northlink.com> * - * Fix to work around buggy BIOSes which dont use carry bit correctly + * Fix to work around buggy BIOSes which don't use carry bit correctly * and/or report extended memory in CX/DX for e801h memory size detection * call. As a result the kernel got wrong figures. The int15/e801h docs * from Ralf Brown interrupt list seem to indicate AX/BX should be used @@ -383,7 +383,7 @@ sse_ok: # a whole bunch of different types, and allows memory holes and # everything. We scan through this memory map and build a list # of the first 32 memory areas, which we return at [E820MAP]. -# This is documented at http://www.teleport.com/~acpi/acpihtml/topic245.htm +# This is documented at http://www.acpi.info/, in the ACPI 2.0 specification. #define SMAP 0x534d4150 @@ -436,7 +436,7 @@ bail820: meme801: stc # fix to work around buggy - xorw %cx,%cx # BIOSes which dont clear/set + xorw %cx,%cx # BIOSes which don't clear/set xorw %dx,%dx # carry on pass/error of # e801h memory size call # or merely pass cx,dx though @@ -733,7 +733,7 @@ flush_instr: # # but we yet haven't reloaded the CS register, so the default size # of the target offset still is 16 bit. -# However, using an operant prefix (0x66), the CPU will properly +# However, using an operand prefix (0x66), the CPU will properly # take our 48 bit far pointer. (INTeL 80386 Programmer's Reference # Manual, Mixing 16-bit and 32-bit code, page 16-6) diff --git a/arch/x86_64/boot/tools/build.c b/arch/x86_64/boot/tools/build.c index c2fa6631317..18b5bac1c42 100644 --- a/arch/x86_64/boot/tools/build.c +++ b/arch/x86_64/boot/tools/build.c @@ -1,6 +1,4 @@ /* - * $Id: build.c,v 1.3 2001/06/26 15:14:50 pavel Exp $ - * * Copyright (C) 1991, 1992 Linus Torvalds * Copyright (C) 1997 Martin Mares */ @@ -8,7 +6,8 @@ /* * This file builds a disk-image from three different files: * - * - bootsect: exactly 512 bytes of 8086 machine code, loads the rest + * - bootsect: compatibility mbr which prints an error message if + * someone tries to boot the kernel directly. * - setup: 8086 machine code, sets up system parm * - system: 80386 code for actual system * diff --git a/arch/x86_64/ia32/ia32_aout.c b/arch/x86_64/ia32/ia32_aout.c index 1965efc974d..c12edf5d97f 100644 --- a/arch/x86_64/ia32/ia32_aout.c +++ b/arch/x86_64/ia32/ia32_aout.c @@ -312,6 +312,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) current->mm->brk = ex.a_bss + (current->mm->start_brk = N_BSSADDR(ex)); current->mm->free_area_cache = TASK_UNMAPPED_BASE; + current->mm->cached_hole_size = 0; set_mm_counter(current->mm, rss, 0); current->mm->mmap = NULL; diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index 99b522052d1..c8131f342cf 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -46,7 +46,7 @@ struct elf_phdr; #define IA32_EMULATOR 1 -#define ELF_ET_DYN_BASE (TASK_UNMAPPED_32 + 0x1000000) +#define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x1000000) #undef ELF_ARCH #define ELF_ARCH EM_386 @@ -307,9 +307,6 @@ MODULE_AUTHOR("Eric Youngdale, Andi Kleen"); #define elf_addr_t __u32 -#undef TASK_SIZE -#define TASK_SIZE 0xffffffff - static void elf32_init(struct pt_regs *); #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c index fbd09b5126c..66e2821533d 100644 --- a/arch/x86_64/ia32/ia32_signal.c +++ b/arch/x86_64/ia32/ia32_signal.c @@ -428,8 +428,8 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size) return (void __user *)((rsp - frame_size) & -8UL); } -void ia32_setup_frame(int sig, struct k_sigaction *ka, - compat_sigset_t *set, struct pt_regs * regs) +int ia32_setup_frame(int sig, struct k_sigaction *ka, + compat_sigset_t *set, struct pt_regs * regs) { struct sigframe __user *frame; int err = 0; @@ -514,14 +514,15 @@ void ia32_setup_frame(int sig, struct k_sigaction *ka, current->comm, current->pid, frame, regs->rip, frame->pretcode); #endif - return; + return 1; give_sigsegv: force_sigsegv(sig, current); + return 0; } -void ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, - compat_sigset_t *set, struct pt_regs * regs) +int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, + compat_sigset_t *set, struct pt_regs * regs) { struct rt_sigframe __user *frame; int err = 0; @@ -613,9 +614,9 @@ void ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, current->comm, current->pid, frame, regs->rip, frame->pretcode); #endif - return; + return 1; give_sigsegv: force_sigsegv(sig, current); + return 0; } - diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S index f3ca0db85b5..cc935427d53 100644 --- a/arch/x86_64/ia32/ia32entry.S +++ b/arch/x86_64/ia32/ia32entry.S @@ -589,7 +589,7 @@ ia32_sys_call_table: .quad compat_sys_mq_timedreceive /* 280 */ .quad compat_sys_mq_notify .quad compat_sys_mq_getsetattr - .quad quiet_ni_syscall /* reserved for kexec */ + .quad compat_sys_kexec_load /* reserved for kexec */ .quad compat_sys_waitid .quad quiet_ni_syscall /* sys_altroot */ .quad sys_add_key diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile index 5ca4a4598fd..48f9e2c19cd 100644 --- a/arch/x86_64/kernel/Makefile +++ b/arch/x86_64/kernel/Makefile @@ -20,6 +20,7 @@ obj-$(CONFIG_SMP) += smp.o smpboot.o trampoline.o obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o obj-$(CONFIG_X86_IO_APIC) += io_apic.o mpparse.o \ genapic.o genapic_cluster.o genapic_flat.o +obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o crash.o obj-$(CONFIG_PM) += suspend.o obj-$(CONFIG_SOFTWARE_SUSPEND) += suspend_asm.o obj-$(CONFIG_CPU_FREQ) += cpufreq/ diff --git a/arch/x86_64/kernel/acpi/wakeup.S b/arch/x86_64/kernel/acpi/wakeup.S index a4c630034cd..185faa911db 100644 --- a/arch/x86_64/kernel/acpi/wakeup.S +++ b/arch/x86_64/kernel/acpi/wakeup.S @@ -67,7 +67,7 @@ wakeup_code: shll $4, %eax addl $(gdta - wakeup_code), %eax movl %eax, gdt_48a +2 - wakeup_code - lgdt %ds:gdt_48a - wakeup_code # load gdt with whatever is + lgdtl %ds:gdt_48a - wakeup_code # load gdt with whatever is # appropriate movl $1, %eax # protected mode (PE) bit diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c index a491f72cc96..c9a6b812e92 100644 --- a/arch/x86_64/kernel/aperture.c +++ b/arch/x86_64/kernel/aperture.c @@ -33,20 +33,14 @@ int fallback_aper_force __initdata = 0; int fix_aperture __initdata = 1; -#define NB_ID_3 (PCI_VENDOR_ID_AMD | (0x1103<<16)) +/* This code runs before the PCI subsystem is initialized, so just + access the northbridge directly. */ -static struct resource aper_res = { - .name = "Aperture", - .flags = IORESOURCE_MEM, -}; +#define NB_ID_3 (PCI_VENDOR_ID_AMD | (0x1103<<16)) static u32 __init allocate_aperture(void) { -#ifdef CONFIG_DISCONTIGMEM pg_data_t *nd0 = NODE_DATA(0); -#else - pg_data_t *nd0 = &contig_page_data; -#endif u32 aper_size; void *p; @@ -55,24 +49,11 @@ static u32 __init allocate_aperture(void) aper_size = (32 * 1024 * 1024) << fallback_aper_order; /* - * Aperture has to be naturally aligned. This means an 2GB - * aperture won't have much chances to find a place in the - * lower 4GB of memory. Unfortunately we cannot move it up - * because that would make the IOMMU useless. + * Aperture has to be naturally aligned. This means an 2GB aperture won't + * have much chances to find a place in the lower 4GB of memory. + * Unfortunately we cannot move it up because that would make the + * IOMMU useless. */ - - /* First try to find some free unused space */ - if (!allocate_resource(&iomem_resource, &aper_res, - aper_size, - 0, 0xffffffff, - aper_size, - NULL, NULL)) { - printk(KERN_INFO "Putting aperture at %lx-%lx\n", - aper_res.start, aper_res.end); - return aper_res.start; - } - - /* No free space found. Go on to waste some memory... */ p = __alloc_bootmem_node(nd0, aper_size, aper_size, 0); if (!p || __pa(p)+aper_size > 0xffffffff) { printk("Cannot allocate aperture memory hole (%p,%uK)\n", @@ -81,7 +62,7 @@ static u32 __init allocate_aperture(void) free_bootmem_node(nd0, (unsigned long)p, aper_size); return 0; } - printk("Mapping aperture over %d KB of precious RAM @ %lx\n", + printk("Mapping aperture over %d KB of RAM @ %lx\n", aper_size >> 10, __pa(p)); return (u32)__pa(p); } @@ -102,16 +83,10 @@ static int __init aperture_valid(char *name, u64 aper_base, u32 aper_size) printk("Aperture from %s pointing to e820 RAM. Ignoring.\n",name); return 0; } - /* Don't check the resource here because the aperture is usually - in an e820 reserved area, and we allocated these earlier. */ return 1; } -/* - * Find a PCI capability. - * This code runs before the PCI subsystem is initialized, so just - * access the northbridge directly. - */ +/* Find a PCI capability */ static __u32 __init find_cap(int num, int slot, int func, int cap) { u8 pos; @@ -276,6 +251,8 @@ void __init iommu_hole_init(void) fallback_aper_force) { printk("Your BIOS doesn't leave a aperture memory hole\n"); printk("Please enable the IOMMU option in the BIOS setup\n"); + printk("This costs you %d MB of RAM\n", + 32 << fallback_aper_order); aper_order = fallback_aper_order; aper_alloc = allocate_aperture(); diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c index f8e6cc4fecd..375d369570c 100644 --- a/arch/x86_64/kernel/apic.c +++ b/arch/x86_64/kernel/apic.c @@ -133,7 +133,7 @@ void __init connect_bsp_APIC(void) } } -void disconnect_bsp_APIC(void) +void disconnect_bsp_APIC(int virt_wire_setup) { if (pic_mode) { /* @@ -146,6 +146,42 @@ void disconnect_bsp_APIC(void) outb(0x70, 0x22); outb(0x00, 0x23); } + else { + /* Go back to Virtual Wire compatibility mode */ + unsigned long value; + + /* For the spurious interrupt use vector F, and enable it */ + value = apic_read(APIC_SPIV); + value &= ~APIC_VECTOR_MASK; + value |= APIC_SPIV_APIC_ENABLED; + value |= 0xf; + apic_write_around(APIC_SPIV, value); + + if (!virt_wire_setup) { + /* For LVT0 make it edge triggered, active high, external and enabled */ + value = apic_read(APIC_LVT0); + value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING | + APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR | + APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED ); + value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING; + value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT); + apic_write_around(APIC_LVT0, value); + } + else { + /* Disable LVT0 */ + apic_write_around(APIC_LVT0, APIC_LVT_MASKED); + } + + /* For LVT1 make it edge triggered, active high, nmi and enabled */ + value = apic_read(APIC_LVT1); + value &= ~( + APIC_MODE_MASK | APIC_SEND_PENDING | + APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR | + APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED); + value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING; + value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI); + apic_write_around(APIC_LVT1, value); + } } void disable_local_APIC(void) @@ -285,7 +321,7 @@ void __init init_bsp_APIC(void) apic_write_around(APIC_LVT1, value); } -void __init setup_local_APIC (void) +void __cpuinit setup_local_APIC (void) { unsigned int value, ver, maxlvt; @@ -534,7 +570,7 @@ static struct sys_device device_lapic = { .cls = &lapic_sysclass, }; -static void __init apic_pm_activate(void) +static void __cpuinit apic_pm_activate(void) { apic_pm_state.active = 1; } @@ -774,14 +810,14 @@ void __init setup_boot_APIC_clock (void) local_irq_enable(); } -void __init setup_secondary_APIC_clock(void) +void __cpuinit setup_secondary_APIC_clock(void) { local_irq_disable(); /* FIXME: Do we need this? --RR */ setup_APIC_timer(calibration_result); local_irq_enable(); } -void __init disable_APIC_timer(void) +void __cpuinit disable_APIC_timer(void) { if (using_apic_timer) { unsigned long v; diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c new file mode 100644 index 00000000000..d7fa4248501 --- /dev/null +++ b/arch/x86_64/kernel/crash.c @@ -0,0 +1,35 @@ +/* + * Architecture specific (x86_64) functions for kexec based crash dumps. + * + * Created by: Hariprasad Nellitheertha (hari@in.ibm.com) + * + * Copyright (C) IBM Corporation, 2004. All rights reserved. + * + */ + +#include <linux/init.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/smp.h> +#include <linux/irq.h> +#include <linux/reboot.h> +#include <linux/kexec.h> + +#include <asm/processor.h> +#include <asm/hardirq.h> +#include <asm/nmi.h> +#include <asm/hw_irq.h> + +note_buf_t crash_notes[NR_CPUS]; + +void machine_crash_shutdown(struct pt_regs *regs) +{ + /* This function is only called after the system + * has paniced or is otherwise in a critical state. + * The minimum amount of code to allow a kexec'd kernel + * to run successfully needs to happen here. + * + * In practice this means shooting down the other cpus in + * an SMP system. + */ +} diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index 7c154dfff64..6ded3a50dfe 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -16,6 +16,7 @@ #include <linux/bootmem.h> #include <linux/ioport.h> #include <linux/string.h> +#include <linux/kexec.h> #include <asm/page.h> #include <asm/e820.h> #include <asm/proto.h> @@ -191,8 +192,6 @@ void __init e820_reserve_resources(void) int i; for (i = 0; i < e820.nr_map; i++) { struct resource *res; - if (e820.map[i].addr + e820.map[i].size > 0x100000000ULL) - continue; res = alloc_bootmem_low(sizeof(struct resource)); switch (e820.map[i].type) { case E820_RAM: res->name = "System RAM"; break; @@ -212,6 +211,9 @@ void __init e820_reserve_resources(void) */ request_resource(res, &code_resource); request_resource(res, &data_resource); +#ifdef CONFIG_KEXEC + request_resource(res, &crashk_res); +#endif } } } diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c index e3a19e8ebbf..9631c747c5e 100644 --- a/arch/x86_64/kernel/early_printk.c +++ b/arch/x86_64/kernel/early_printk.c @@ -2,20 +2,24 @@ #include <linux/kernel.h> #include <linux/init.h> #include <linux/string.h> +#include <linux/tty.h> #include <asm/io.h> #include <asm/processor.h> /* Simple VGA output */ #ifdef __i386__ +#include <asm/setup.h> #define VGABASE (__ISA_IO_base + 0xb8000) #else +#include <asm/bootsetup.h> #define VGABASE ((void __iomem *)0xffffffff800b8000UL) #endif -#define MAX_YPOS 25 -#define MAX_XPOS 80 +#define MAX_YPOS max_ypos +#define MAX_XPOS max_xpos +static int max_ypos = 25, max_xpos = 80; static int current_ypos = 1, current_xpos = 0; static void early_vga_write(struct console *con, const char *str, unsigned n) @@ -196,7 +200,10 @@ int __init setup_early_printk(char *opt) } else if (!strncmp(buf, "ttyS", 4)) { early_serial_init(buf); early_console = &early_serial_console; - } else if (!strncmp(buf, "vga", 3)) { + } else if (!strncmp(buf, "vga", 3) + && SCREEN_INFO.orig_video_isVGA == 1) { + max_xpos = SCREEN_INFO.orig_video_cols; + max_ypos = SCREEN_INFO.orig_video_lines; early_console = &early_vga_console; } early_console_initialized = 1; diff --git a/arch/x86_64/kernel/genapic_flat.c b/arch/x86_64/kernel/genapic_flat.c index b4cbbad0422..28284696508 100644 --- a/arch/x86_64/kernel/genapic_flat.c +++ b/arch/x86_64/kernel/genapic_flat.c @@ -7,6 +7,8 @@ * Hacked for x86-64 by James Cleverdon from i386 architecture code by * Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and * James Cleverdon. + * Ashok Raj <ashok.raj@intel.com> + * Removed IPI broadcast shortcut to support CPU hotplug */ #include <linux/config.h> #include <linux/threads.h> @@ -18,6 +20,46 @@ #include <asm/smp.h> #include <asm/ipi.h> +/* + * The following permit choosing broadcast IPI shortcut v.s sending IPI only + * to online cpus via the send_IPI_mask varient. + * The mask version is my preferred option, since it eliminates a lot of + * other extra code that would need to be written to cleanup intrs sent + * to a CPU while offline. + * + * Sending broadcast introduces lots of trouble in CPU hotplug situations. + * These IPI's are delivered to cpu's irrespective of their offline status + * and could pickup stale intr data when these CPUS are turned online. + * + * Not using broadcast is a cleaner approach IMO, but Andi Kleen disagrees with + * the idea of not using broadcast IPI's anymore. Hence the run time check + * is introduced, on his request so we can choose an alternate mechanism. + * + * Initial wacky performance tests that collect cycle counts show + * no increase in using mask v.s broadcast version. In fact they seem + * identical in terms of cycle counts. + * + * if we need to use broadcast, we need to do the following. + * + * cli; + * hold call_lock; + * clear any pending IPI, just ack and clear all pending intr + * set cpu_online_map; + * release call_lock; + * sti; + * + * The complicated dummy irq processing shown above is not required if + * we didnt sent IPI's to wrong CPU's in the first place. + * + * - Ashok Raj <ashok.raj@intel.com> + */ +#ifdef CONFIG_HOTPLUG_CPU +#define DEFAULT_SEND_IPI (1) +#else +#define DEFAULT_SEND_IPI (0) +#endif + +static int no_broadcast=DEFAULT_SEND_IPI; static cpumask_t flat_target_cpus(void) { @@ -45,22 +87,6 @@ static void flat_init_apic_ldr(void) apic_write_around(APIC_LDR, val); } -static void flat_send_IPI_allbutself(int vector) -{ - /* - * if there are no other CPUs in the system then - * we get an APIC send error if we try to broadcast. - * thus we have to avoid sending IPIs in this case. - */ - if (num_online_cpus() > 1) - __send_IPI_shortcut(APIC_DEST_ALLBUT, vector, APIC_DEST_LOGICAL); -} - -static void flat_send_IPI_all(int vector) -{ - __send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL); -} - static void flat_send_IPI_mask(cpumask_t cpumask, int vector) { unsigned long mask = cpus_addr(cpumask)[0]; @@ -93,6 +119,39 @@ static void flat_send_IPI_mask(cpumask_t cpumask, int vector) local_irq_restore(flags); } +static inline void __local_flat_send_IPI_allbutself(int vector) +{ + if (no_broadcast) { + cpumask_t mask = cpu_online_map; + int this_cpu = get_cpu(); + + cpu_clear(this_cpu, mask); + flat_send_IPI_mask(mask, vector); + put_cpu(); + } + else + __send_IPI_shortcut(APIC_DEST_ALLBUT, vector, APIC_DEST_LOGICAL); +} + +static inline void __local_flat_send_IPI_all(int vector) +{ + if (no_broadcast) + flat_send_IPI_mask(cpu_online_map, vector); + else + __send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL); +} + +static void flat_send_IPI_allbutself(int vector) +{ + if (((num_online_cpus()) - 1) >= 1) + __local_flat_send_IPI_allbutself(vector); +} + +static void flat_send_IPI_all(int vector) +{ + __local_flat_send_IPI_all(vector); +} + static int flat_apic_id_registered(void) { return physid_isset(GET_APIC_ID(apic_read(APIC_ID)), phys_cpu_present_map); @@ -111,6 +170,16 @@ static unsigned int phys_pkg_id(int index_msb) return ((ebx >> 24) & 0xFF) >> index_msb; } +static __init int no_ipi_broadcast(char *str) +{ + get_option(&str, &no_broadcast); + printk ("Using %s mode\n", no_broadcast ? "No IPI Broadcast" : + "IPI Broadcast"); + return 1; +} + +__setup("no_ipi_broadcast", no_ipi_broadcast); + struct genapic apic_flat = { .name = "flat", .int_delivery_mode = dest_LowestPrio, @@ -125,3 +194,12 @@ struct genapic apic_flat = { .cpu_mask_to_apicid = flat_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, }; + +static int __init print_ipi_mode(void) +{ + printk ("Using IPI %s mode\n", no_broadcast ? "No-Shortcut" : + "Shortcut"); + return 0; +} + +late_initcall(print_ipi_mode); diff --git a/arch/x86_64/kernel/head.S b/arch/x86_64/kernel/head.S index 9bd2e7a4b81..8d765aa77a2 100644 --- a/arch/x86_64/kernel/head.S +++ b/arch/x86_64/kernel/head.S @@ -248,23 +248,23 @@ ENTRY(_stext) */ .org 0x1000 ENTRY(init_level4_pgt) - .quad 0x0000000000102007 /* -> level3_ident_pgt */ + .quad 0x0000000000002007 + __PHYSICAL_START /* -> level3_ident_pgt */ .fill 255,8,0 - .quad 0x000000000010a007 + .quad 0x000000000000a007 + __PHYSICAL_START .fill 254,8,0 /* (2^48-(2*1024*1024*1024))/(2^39) = 511 */ - .quad 0x0000000000103007 /* -> level3_kernel_pgt */ + .quad 0x0000000000003007 + __PHYSICAL_START /* -> level3_kernel_pgt */ .org 0x2000 ENTRY(level3_ident_pgt) - .quad 0x0000000000104007 + .quad 0x0000000000004007 + __PHYSICAL_START .fill 511,8,0 .org 0x3000 ENTRY(level3_kernel_pgt) .fill 510,8,0 /* (2^48-(2*1024*1024*1024)-((2^39)*511))/(2^30) = 510 */ - .quad 0x0000000000105007 /* -> level2_kernel_pgt */ + .quad 0x0000000000005007 + __PHYSICAL_START /* -> level2_kernel_pgt */ .fill 1,8,0 .org 0x4000 @@ -337,17 +337,17 @@ ENTRY(empty_bad_pmd_table) .org 0xa000 ENTRY(level3_physmem_pgt) - .quad 0x0000000000105007 /* -> level2_kernel_pgt (so that __va works even before pagetable_init) */ + .quad 0x0000000000005007 + __PHYSICAL_START /* -> level2_kernel_pgt (so that __va works even before pagetable_init) */ .org 0xb000 #ifdef CONFIG_ACPI_SLEEP ENTRY(wakeup_level4_pgt) - .quad 0x0000000000102007 /* -> level3_ident_pgt */ + .quad 0x0000000000002007 + __PHYSICAL_START /* -> level3_ident_pgt */ .fill 255,8,0 - .quad 0x000000000010a007 + .quad 0x000000000000a007 + __PHYSICAL_START .fill 254,8,0 /* (2^48-(2*1024*1024*1024))/(2^39) = 511 */ - .quad 0x0000000000103007 /* -> level3_kernel_pgt */ + .quad 0x0000000000003007 + __PHYSICAL_START /* -> level3_kernel_pgt */ #endif .data diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c index 0f8c78dcd38..cf6ab147a2a 100644 --- a/arch/x86_64/kernel/head64.c +++ b/arch/x86_64/kernel/head64.c @@ -94,7 +94,7 @@ void __init x86_64_start_kernel(char * real_mode_data) s = strstr(saved_command_line, "earlyprintk="); if (s != NULL) setup_early_printk(s); -#ifdef CONFIG_DISCONTIGMEM +#ifdef CONFIG_NUMA s = strstr(saved_command_line, "numa="); if (s != NULL) numa_setup(s+5); diff --git a/arch/x86_64/kernel/i387.c b/arch/x86_64/kernel/i387.c index ba139cac57c..d9b22b633e3 100644 --- a/arch/x86_64/kernel/i387.c +++ b/arch/x86_64/kernel/i387.c @@ -42,7 +42,7 @@ void mxcsr_feature_mask_init(void) * Called at bootup to set up the initial FPU state that is later cloned * into all processes. */ -void __init fpu_init(void) +void __cpuinit fpu_init(void) { unsigned long oldcr0 = read_cr0(); extern void __bad_fxsave_alignment(void); diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c index 7873d9ba881..a8916909512 100644 --- a/arch/x86_64/kernel/i8259.c +++ b/arch/x86_64/kernel/i8259.c @@ -157,14 +157,13 @@ static unsigned int startup_8259A_irq(unsigned int irq) } static struct hw_interrupt_type i8259A_irq_type = { - "XT-PIC", - startup_8259A_irq, - shutdown_8259A_irq, - enable_8259A_irq, - disable_8259A_irq, - mask_and_ack_8259A, - end_8259A_irq, - NULL + .typename = "XT-PIC", + .startup = startup_8259A_irq, + .shutdown = shutdown_8259A_irq, + .enable = enable_8259A_irq, + .disable = disable_8259A_irq, + .ack = mask_and_ack_8259A, + .end = end_8259A_irq, }; /* @@ -415,10 +414,22 @@ static int i8259A_suspend(struct sys_device *dev, pm_message_t state) return 0; } +static int i8259A_shutdown(struct sys_device *dev) +{ + /* Put the i8259A into a quiescent state that + * the kernel initialization code can get it + * out of. + */ + outb(0xff, 0x21); /* mask all of 8259A-1 */ + outb(0xff, 0xA1); /* mask all of 8259A-1 */ + return 0; +} + static struct sysdev_class i8259_sysdev_class = { set_kset_name("i8259"), .suspend = i8259A_suspend, .resume = i8259A_resume, + .shutdown = i8259A_shutdown, }; static struct sys_device device_i8259A = { diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 80e9b498c44..157190d986b 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -37,6 +37,7 @@ #include <asm/desc.h> #include <asm/proto.h> #include <asm/mach_apic.h> +#include <asm/acpi.h> #define __apicdebuginit __init @@ -329,7 +330,7 @@ static int find_irq_entry(int apic, int pin, int type) /* * Find the pin to which IRQ[irq] (ISA) is connected */ -static int __init find_isa_irq_pin(int irq, int type) +static int find_isa_irq_pin(int irq, int type) { int i; @@ -1131,12 +1132,44 @@ static void __init enable_IO_APIC(void) */ void disable_IO_APIC(void) { + int pin; /* * Clear the IO-APIC before rebooting: */ clear_IO_APIC(); - disconnect_bsp_APIC(); + /* + * If the i82559 is routed through an IOAPIC + * Put that IOAPIC in virtual wire mode + * so legacy interrups can be delivered. + */ + pin = find_isa_irq_pin(0, mp_ExtINT); + if (pin != -1) { + struct IO_APIC_route_entry entry; + unsigned long flags; + + memset(&entry, 0, sizeof(entry)); + entry.mask = 0; /* Enabled */ + entry.trigger = 0; /* Edge */ + entry.irr = 0; + entry.polarity = 0; /* High */ + entry.delivery_status = 0; + entry.dest_mode = 0; /* Physical */ + entry.delivery_mode = 7; /* ExtInt */ + entry.vector = 0; + entry.dest.physical.physical_dest = 0; + + + /* + * Add it to the IO-APIC irq-routing table: + */ + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(0, 0x11+2*pin, *(((int *)&entry)+1)); + io_apic_write(0, 0x10+2*pin, *(((int *)&entry)+0)); + spin_unlock_irqrestore(&ioapic_lock, flags); + } + + disconnect_bsp_APIC(pin != -1); } /* diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c index 62b112e4deb..cc3fb85f514 100644 --- a/arch/x86_64/kernel/irq.c +++ b/arch/x86_64/kernel/irq.c @@ -14,6 +14,7 @@ #include <linux/interrupt.h> #include <linux/seq_file.h> #include <linux/module.h> +#include <linux/delay.h> #include <asm/uaccess.h> #include <asm/io_apic.h> @@ -106,3 +107,31 @@ asmlinkage unsigned int do_IRQ(struct pt_regs *regs) return 1; } +#ifdef CONFIG_HOTPLUG_CPU +void fixup_irqs(cpumask_t map) +{ + unsigned int irq; + static int warned; + + for (irq = 0; irq < NR_IRQS; irq++) { + cpumask_t mask; + if (irq == 2) + continue; + + cpus_and(mask, irq_affinity[irq], map); + if (any_online_cpu(mask) == NR_CPUS) { + printk("Breaking affinity for irq %i\n", irq); + mask = map; + } + if (irq_desc[irq].handler->set_affinity) + irq_desc[irq].handler->set_affinity(irq, mask); + else if (irq_desc[irq].action && !(warned++)) + printk("Cannot set affinity for irq %i\n", irq); + } + + /* That doesn't seem sufficient. Give it 1ms. */ + local_irq_enable(); + mdelay(1); + local_irq_disable(); +} +#endif diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c index f77f8a0ff18..4e680f87a75 100644 --- a/arch/x86_64/kernel/kprobes.c +++ b/arch/x86_64/kernel/kprobes.c @@ -27,6 +27,8 @@ * <prasanna@in.ibm.com> adapted for x86_64 * 2005-Mar Roland McGrath <roland@redhat.com> * Fixed to handle %rip-relative addressing mode correctly. + * 2005-May Rusty Lynch <rusty.lynch@intel.com> + * Added function return probes functionality */ #include <linux/config.h> @@ -37,18 +39,16 @@ #include <linux/slab.h> #include <linux/preempt.h> #include <linux/moduleloader.h> - +#include <asm/cacheflush.h> #include <asm/pgtable.h> #include <asm/kdebug.h> static DECLARE_MUTEX(kprobe_mutex); -/* kprobe_status settings */ -#define KPROBE_HIT_ACTIVE 0x00000001 -#define KPROBE_HIT_SS 0x00000002 - static struct kprobe *current_kprobe; static unsigned long kprobe_status, kprobe_old_rflags, kprobe_saved_rflags; +static struct kprobe *kprobe_prev; +static unsigned long kprobe_status_prev, kprobe_old_rflags_prev, kprobe_saved_rflags_prev; static struct pt_regs jprobe_saved_regs; static long *jprobe_saved_rsp; static kprobe_opcode_t *get_insn_slot(void); @@ -214,6 +214,21 @@ void arch_copy_kprobe(struct kprobe *p) BUG_ON((s64) (s32) disp != disp); /* Sanity check. */ *ripdisp = disp; } + p->opcode = *p->addr; +} + +void arch_arm_kprobe(struct kprobe *p) +{ + *p->addr = BREAKPOINT_INSTRUCTION; + flush_icache_range((unsigned long) p->addr, + (unsigned long) p->addr + sizeof(kprobe_opcode_t)); +} + +void arch_disarm_kprobe(struct kprobe *p) +{ + *p->addr = p->opcode; + flush_icache_range((unsigned long) p->addr, + (unsigned long) p->addr + sizeof(kprobe_opcode_t)); } void arch_remove_kprobe(struct kprobe *p) @@ -223,10 +238,29 @@ void arch_remove_kprobe(struct kprobe *p) down(&kprobe_mutex); } -static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs) +static inline void save_previous_kprobe(void) { - *p->addr = p->opcode; - regs->rip = (unsigned long)p->addr; + kprobe_prev = current_kprobe; + kprobe_status_prev = kprobe_status; + kprobe_old_rflags_prev = kprobe_old_rflags; + kprobe_saved_rflags_prev = kprobe_saved_rflags; +} + +static inline void restore_previous_kprobe(void) +{ + current_kprobe = kprobe_prev; + kprobe_status = kprobe_status_prev; + kprobe_old_rflags = kprobe_old_rflags_prev; + kprobe_saved_rflags = kprobe_saved_rflags_prev; +} + +static inline void set_current_kprobe(struct kprobe *p, struct pt_regs *regs) +{ + current_kprobe = p; + kprobe_saved_rflags = kprobe_old_rflags + = (regs->eflags & (TF_MASK | IF_MASK)); + if (is_IF_modifier(p->ainsn.insn)) + kprobe_saved_rflags &= ~IF_MASK; } static void prepare_singlestep(struct kprobe *p, struct pt_regs *regs) @@ -240,6 +274,50 @@ static void prepare_singlestep(struct kprobe *p, struct pt_regs *regs) regs->rip = (unsigned long)p->ainsn.insn; } +struct task_struct *arch_get_kprobe_task(void *ptr) +{ + return ((struct thread_info *) (((unsigned long) ptr) & + (~(THREAD_SIZE -1))))->task; +} + +void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs) +{ + unsigned long *sara = (unsigned long *)regs->rsp; + struct kretprobe_instance *ri; + static void *orig_ret_addr; + + /* + * Save the return address when the return probe hits + * the first time, and use it to populate the (krprobe + * instance)->ret_addr for subsequent return probes at + * the same addrress since stack address would have + * the kretprobe_trampoline by then. + */ + if (((void*) *sara) != kretprobe_trampoline) + orig_ret_addr = (void*) *sara; + + if ((ri = get_free_rp_inst(rp)) != NULL) { + ri->rp = rp; + ri->stack_addr = sara; + ri->ret_addr = orig_ret_addr; + add_rp_inst(ri); + /* Replace the return addr with trampoline addr */ + *sara = (unsigned long) &kretprobe_trampoline; + } else { + rp->nmissed++; + } +} + +void arch_kprobe_flush_task(struct task_struct *tk) +{ + struct kretprobe_instance *ri; + while ((ri = get_rp_inst_tsk(tk)) != NULL) { + *((unsigned long *)(ri->stack_addr)) = + (unsigned long) ri->ret_addr; + recycle_rp_inst(ri); + } +} + /* * Interrupts are disabled on entry as trap3 is an interrupt gate and they * remain disabled thorough out this function. @@ -264,9 +342,30 @@ int kprobe_handler(struct pt_regs *regs) regs->eflags |= kprobe_saved_rflags; unlock_kprobes(); goto no_kprobe; + } else if (kprobe_status == KPROBE_HIT_SSDONE) { + /* TODO: Provide re-entrancy from + * post_kprobes_handler() and avoid exception + * stack corruption while single-stepping on + * the instruction of the new probe. + */ + arch_disarm_kprobe(p); + regs->rip = (unsigned long)p->addr; + ret = 1; + } else { + /* We have reentered the kprobe_handler(), since + * another probe was hit while within the + * handler. We here save the original kprobe + * variables and just single step on instruction + * of the new probe without calling any user + * handlers. + */ + save_previous_kprobe(); + set_current_kprobe(p, regs); + p->nmissed++; + prepare_singlestep(p, regs); + kprobe_status = KPROBE_REENTER; + return 1; } - disarm_kprobe(p, regs); - ret = 1; } else { p = current_kprobe; if (p->break_handler && p->break_handler(p, regs)) { @@ -296,11 +395,7 @@ int kprobe_handler(struct pt_regs *regs) } kprobe_status = KPROBE_HIT_ACTIVE; - current_kprobe = p; - kprobe_saved_rflags = kprobe_old_rflags - = (regs->eflags & (TF_MASK | IF_MASK)); - if (is_IF_modifier(p->ainsn.insn)) - kprobe_saved_rflags &= ~IF_MASK; + set_current_kprobe(p, regs); if (p->pre_handler && p->pre_handler(p, regs)) /* handler has already set things up, so skip ss setup */ @@ -317,6 +412,55 @@ no_kprobe: } /* + * For function-return probes, init_kprobes() establishes a probepoint + * here. When a retprobed function returns, this probe is hit and + * trampoline_probe_handler() runs, calling the kretprobe's handler. + */ + void kretprobe_trampoline_holder(void) + { + asm volatile ( ".global kretprobe_trampoline\n" + "kretprobe_trampoline: \n" + "nop\n"); + } + +/* + * Called when we hit the probe point at kretprobe_trampoline + */ +int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) +{ + struct task_struct *tsk; + struct kretprobe_instance *ri; + struct hlist_head *head; + struct hlist_node *node; + unsigned long *sara = (unsigned long *)regs->rsp - 1; + + tsk = arch_get_kprobe_task(sara); + head = kretprobe_inst_table_head(tsk); + + hlist_for_each_entry(ri, node, head, hlist) { + if (ri->stack_addr == sara && ri->rp) { + if (ri->rp->handler) + ri->rp->handler(ri, regs); + } + } + return 0; +} + +void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs, + unsigned long flags) +{ + struct kretprobe_instance *ri; + /* RA already popped */ + unsigned long *sara = ((unsigned long *)regs->rsp) - 1; + + while ((ri = get_rp_inst(sara))) { + regs->rip = (unsigned long)ri->ret_addr; + recycle_rp_inst(ri); + } + regs->eflags &= ~TF_MASK; +} + +/* * Called after single-stepping. p->addr is the address of the * instruction whose first byte has been replaced by the "int 3" * instruction. To avoid the SMP problems that can occur when we @@ -401,13 +545,23 @@ int post_kprobe_handler(struct pt_regs *regs) if (!kprobe_running()) return 0; - if (current_kprobe->post_handler) + if ((kprobe_status != KPROBE_REENTER) && current_kprobe->post_handler) { + kprobe_status = KPROBE_HIT_SSDONE; current_kprobe->post_handler(current_kprobe, regs, 0); + } - resume_execution(current_kprobe, regs); + if (current_kprobe->post_handler != trampoline_post_handler) + resume_execution(current_kprobe, regs); regs->eflags |= kprobe_saved_rflags; - unlock_kprobes(); + /* Restore the original saved kprobes variables and continue. */ + if (kprobe_status == KPROBE_REENTER) { + restore_previous_kprobe(); + goto out; + } else { + unlock_kprobes(); + } +out: preempt_enable_no_resched(); /* diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c new file mode 100644 index 00000000000..60d1eff4156 --- /dev/null +++ b/arch/x86_64/kernel/machine_kexec.c @@ -0,0 +1,250 @@ +/* + * machine_kexec.c - handle transition of Linux booting another kernel + * Copyright (C) 2002-2005 Eric Biederman <ebiederm@xmission.com> + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include <linux/mm.h> +#include <linux/kexec.h> +#include <linux/delay.h> +#include <linux/string.h> +#include <linux/reboot.h> +#include <asm/pda.h> +#include <asm/pgtable.h> +#include <asm/pgalloc.h> +#include <asm/tlbflush.h> +#include <asm/mmu_context.h> +#include <asm/io.h> +#include <asm/apic.h> +#include <asm/cpufeature.h> +#include <asm/hw_irq.h> + +#define LEVEL0_SIZE (1UL << 12UL) +#define LEVEL1_SIZE (1UL << 21UL) +#define LEVEL2_SIZE (1UL << 30UL) +#define LEVEL3_SIZE (1UL << 39UL) +#define LEVEL4_SIZE (1UL << 48UL) + +#define L0_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY) +#define L1_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_PSE) +#define L2_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY) +#define L3_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY) + +static void init_level2_page(u64 *level2p, unsigned long addr) +{ + unsigned long end_addr; + + addr &= PAGE_MASK; + end_addr = addr + LEVEL2_SIZE; + while (addr < end_addr) { + *(level2p++) = addr | L1_ATTR; + addr += LEVEL1_SIZE; + } +} + +static int init_level3_page(struct kimage *image, u64 *level3p, + unsigned long addr, unsigned long last_addr) +{ + unsigned long end_addr; + int result; + + result = 0; + addr &= PAGE_MASK; + end_addr = addr + LEVEL3_SIZE; + while ((addr < last_addr) && (addr < end_addr)) { + struct page *page; + u64 *level2p; + + page = kimage_alloc_control_pages(image, 0); + if (!page) { + result = -ENOMEM; + goto out; + } + level2p = (u64 *)page_address(page); + init_level2_page(level2p, addr); + *(level3p++) = __pa(level2p) | L2_ATTR; + addr += LEVEL2_SIZE; + } + /* clear the unused entries */ + while (addr < end_addr) { + *(level3p++) = 0; + addr += LEVEL2_SIZE; + } +out: + return result; +} + + +static int init_level4_page(struct kimage *image, u64 *level4p, + unsigned long addr, unsigned long last_addr) +{ + unsigned long end_addr; + int result; + + result = 0; + addr &= PAGE_MASK; + end_addr = addr + LEVEL4_SIZE; + while ((addr < last_addr) && (addr < end_addr)) { + struct page *page; + u64 *level3p; + + page = kimage_alloc_control_pages(image, 0); + if (!page) { + result = -ENOMEM; + goto out; + } + level3p = (u64 *)page_address(page); + result = init_level3_page(image, level3p, addr, last_addr); + if (result) { + goto out; + } + *(level4p++) = __pa(level3p) | L3_ATTR; + addr += LEVEL3_SIZE; + } + /* clear the unused entries */ + while (addr < end_addr) { + *(level4p++) = 0; + addr += LEVEL3_SIZE; + } +out: + return result; +} + + +static int init_pgtable(struct kimage *image, unsigned long start_pgtable) +{ + u64 *level4p; + level4p = (u64 *)__va(start_pgtable); + return init_level4_page(image, level4p, 0, end_pfn << PAGE_SHIFT); +} + +static void set_idt(void *newidt, u16 limit) +{ + unsigned char curidt[10]; + + /* x86-64 supports unaliged loads & stores */ + (*(u16 *)(curidt)) = limit; + (*(u64 *)(curidt +2)) = (unsigned long)(newidt); + + __asm__ __volatile__ ( + "lidt %0\n" + : "=m" (curidt) + ); +}; + + +static void set_gdt(void *newgdt, u16 limit) +{ + unsigned char curgdt[10]; + + /* x86-64 supports unaligned loads & stores */ + (*(u16 *)(curgdt)) = limit; + (*(u64 *)(curgdt +2)) = (unsigned long)(newgdt); + + __asm__ __volatile__ ( + "lgdt %0\n" + : "=m" (curgdt) + ); +}; + +static void load_segments(void) +{ + __asm__ __volatile__ ( + "\tmovl $"STR(__KERNEL_DS)",%eax\n" + "\tmovl %eax,%ds\n" + "\tmovl %eax,%es\n" + "\tmovl %eax,%ss\n" + "\tmovl %eax,%fs\n" + "\tmovl %eax,%gs\n" + ); +#undef STR +#undef __STR +} + +typedef NORET_TYPE void (*relocate_new_kernel_t)(unsigned long indirection_page, + unsigned long control_code_buffer, + unsigned long start_address, + unsigned long pgtable) ATTRIB_NORET; + +const extern unsigned char relocate_new_kernel[]; +const extern unsigned long relocate_new_kernel_size; + +int machine_kexec_prepare(struct kimage *image) +{ + unsigned long start_pgtable, control_code_buffer; + int result; + + /* Calculate the offsets */ + start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT; + control_code_buffer = start_pgtable + 4096UL; + + /* Setup the identity mapped 64bit page table */ + result = init_pgtable(image, start_pgtable); + if (result) + return result; + + /* Place the code in the reboot code buffer */ + memcpy(__va(control_code_buffer), relocate_new_kernel, + relocate_new_kernel_size); + + return 0; +} + +void machine_kexec_cleanup(struct kimage *image) +{ + return; +} + +/* + * Do not allocate memory (or fail in any way) in machine_kexec(). + * We are past the point of no return, committed to rebooting now. + */ +NORET_TYPE void machine_kexec(struct kimage *image) +{ + unsigned long page_list; + unsigned long control_code_buffer; + unsigned long start_pgtable; + relocate_new_kernel_t rnk; + + /* Interrupts aren't acceptable while we reboot */ + local_irq_disable(); + + /* Calculate the offsets */ + page_list = image->head; + start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT; + control_code_buffer = start_pgtable + 4096UL; + + /* Set the low half of the page table to my identity mapped + * page table for kexec. Leave the high half pointing at the + * kernel pages. Don't bother to flush the global pages + * as that will happen when I fully switch to my identity mapped + * page table anyway. + */ + memcpy(__va(read_cr3()), __va(start_pgtable), PAGE_SIZE/2); + __flush_tlb(); + + + /* The segment registers are funny things, they are + * automatically loaded from a table, in memory wherever you + * set them to a specific selector, but this table is never + * accessed again unless you set the segment to a different selector. + * + * The more common model are caches where the behide + * the scenes work is done, but is also dropped at arbitrary + * times. + * + * I take advantage of this here by force loading the + * segments, before I zap the gdt with an invalid value. + */ + load_segments(); + /* The gdt & idt are now invalid. + * If you want to load them you must set up your own idt & gdt. + */ + set_gdt(phys_to_virt(0),0); + set_idt(phys_to_virt(0),0); + /* now call it */ + rnk = (relocate_new_kernel_t) control_code_buffer; + (*rnk)(page_list, control_code_buffer, image->start, start_pgtable); +} diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index 3a89d735a4f..21e70625a49 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -327,7 +327,7 @@ static void mce_init(void *dummy) } /* Add per CPU specific workarounds here */ -static void __init mce_cpu_quirks(struct cpuinfo_x86 *c) +static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c) { /* This should be disabled by the BIOS, but isn't always */ if (c->x86_vendor == X86_VENDOR_AMD && c->x86 == 15) { @@ -337,7 +337,7 @@ static void __init mce_cpu_quirks(struct cpuinfo_x86 *c) } } -static void __init mce_cpu_features(struct cpuinfo_x86 *c) +static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c) { switch (c->x86_vendor) { case X86_VENDOR_INTEL: @@ -352,7 +352,7 @@ static void __init mce_cpu_features(struct cpuinfo_x86 *c) * Called for each booted CPU to set up machine checks. * Must be called with preempt off. */ -void __init mcheck_init(struct cpuinfo_x86 *c) +void __cpuinit mcheck_init(struct cpuinfo_x86 *c) { static cpumask_t mce_cpus __initdata = CPU_MASK_NONE; @@ -411,7 +411,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, loff memset(mcelog.entry, 0, next * sizeof(struct mce)); mcelog.next = 0; - synchronize_kernel(); + synchronize_sched(); /* Collect entries that were still getting written before the synchronize. */ @@ -542,7 +542,7 @@ ACCESSOR(bank4ctl,bank[4],mce_restart()) ACCESSOR(tolerant,tolerant,) ACCESSOR(check_interval,check_interval,mce_restart()) -static __init int mce_init_device(void) +static __cpuinit int mce_init_device(void) { int err; if (!mce_available(&boot_cpu_data)) diff --git a/arch/x86_64/kernel/mce_intel.c b/arch/x86_64/kernel/mce_intel.c index 4db9a640069..0be0a795981 100644 --- a/arch/x86_64/kernel/mce_intel.c +++ b/arch/x86_64/kernel/mce_intel.c @@ -42,7 +42,7 @@ done: irq_exit(); } -static void __init intel_init_thermal(struct cpuinfo_x86 *c) +static void __cpuinit intel_init_thermal(struct cpuinfo_x86 *c) { u32 l, h; int tm2 = 0; @@ -93,7 +93,7 @@ static void __init intel_init_thermal(struct cpuinfo_x86 *c) return; } -void __init mce_intel_feature_init(struct cpuinfo_x86 *c) +void __cpuinit mce_intel_feature_init(struct cpuinfo_x86 *c) { intel_init_thermal(c); } diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c index f86d9db94bf..9c5aa2a790c 100644 --- a/arch/x86_64/kernel/mpparse.c +++ b/arch/x86_64/kernel/mpparse.c @@ -23,6 +23,7 @@ #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> @@ -30,6 +31,7 @@ #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; @@ -44,7 +46,8 @@ int acpi_found_madt; int apic_version [MAX_APICS]; unsigned char mp_bus_id_to_type [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 }; int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 }; -cpumask_t pci_bus_to_cpumask [256] = { [0 ... 255] = CPU_MASK_ALL }; +unsigned char pci_bus_to_node [256]; +EXPORT_SYMBOL(pci_bus_to_node); static int mp_current_pci_id = 0; /* I/O APIC entries */ @@ -903,11 +906,20 @@ void __init mp_config_acpi_legacy_irqs (void) return; } +#define MAX_GSI_NUM 4096 + int mp_register_gsi(u32 gsi, int edge_level, int active_high_low) { int ioapic = -1; int ioapic_pin = 0; int idx, bit = 0; + static int pci_irq = 16; + /* + * Mapping between Global System Interrupts, which + * represent all possible interrupts, to the IRQs + * assigned to actual devices. + */ + static int gsi_to_irq[MAX_GSI_NUM]; if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC) return gsi; @@ -942,11 +954,21 @@ int mp_register_gsi(u32 gsi, int edge_level, int active_high_low) 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; + return gsi_to_irq[gsi]; } mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit); + if (edge_level) { + /* + * For PCI devices assign IRQs in order, avoiding gaps + * due to unused I/O APIC pins. + */ + int irq = gsi; + gsi = pci_irq++; + gsi_to_irq[irq] = gsi; + } + io_apic_set_pci_routing(ioapic, ioapic_pin, gsi, edge_level == ACPI_EDGE_SENSITIVE ? 0 : 1, active_high_low == ACPI_ACTIVE_HIGH ? 0 : 1); diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index 31c0f2e6ac9..4e44d6e6b7e 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -98,7 +98,7 @@ static unsigned int nmi_p4_cccr_val; (P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT| \ P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE) -static __init inline int nmi_known_cpu(void) +static __cpuinit inline int nmi_known_cpu(void) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: @@ -110,7 +110,7 @@ static __init inline int nmi_known_cpu(void) } /* Run after command line and cpu_init init, but before all other checks */ -void __init nmi_watchdog_default(void) +void __cpuinit nmi_watchdog_default(void) { if (nmi_watchdog != NMI_DEFAULT) return; diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 761b6d35e33..1d91271796e 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -8,7 +8,8 @@ * * X86-64 port * Andi Kleen. - * + * + * CPU hotplug support - ashok.raj@intel.com * $Id: process.c,v 1.38 2002/01/15 10:08:03 ak Exp $ */ @@ -18,6 +19,7 @@ #include <stdarg.h> +#include <linux/cpu.h> #include <linux/errno.h> #include <linux/sched.h> #include <linux/kernel.h> @@ -34,6 +36,7 @@ #include <linux/ptrace.h> #include <linux/utsname.h> #include <linux/random.h> +#include <linux/kprobes.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -153,6 +156,29 @@ void cpu_idle_wait(void) } EXPORT_SYMBOL_GPL(cpu_idle_wait); +#ifdef CONFIG_HOTPLUG_CPU +DECLARE_PER_CPU(int, cpu_state); + +#include <asm/nmi.h> +/* We don't actually take CPU down, just spin without interrupts. */ +static inline void play_dead(void) +{ + idle_task_exit(); + wbinvd(); + mb(); + /* Ack it */ + __get_cpu_var(cpu_state) = CPU_DEAD; + + while (1) + safe_halt(); +} +#else +static inline void play_dead(void) +{ + BUG(); +} +#endif /* CONFIG_HOTPLUG_CPU */ + /* * The idle thread. There's no useful work to be * done, so just try to conserve power and have a @@ -173,6 +199,8 @@ void cpu_idle (void) idle = pm_idle; if (!idle) idle = default_idle; + if (cpu_is_offline(smp_processor_id())) + play_dead(); idle(); } @@ -203,7 +231,7 @@ static void mwait_idle(void) } } -void __init select_idle_routine(const struct cpuinfo_x86 *c) +void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) { static int printed; if (cpu_has(c, X86_FEATURE_MWAIT)) { @@ -293,6 +321,14 @@ void exit_thread(void) { struct task_struct *me = current; struct thread_struct *t = &me->thread; + + /* + * Remove function-return probe instances associated with this task + * and put them back on the free list. Do not insert an exit probe for + * this function, it will be disabled by kprobe_flush_task if you do. + */ + kprobe_flush_task(me); + if (me->thread.io_bitmap_ptr) { struct tss_struct *tss = &per_cpu(init_tss, get_cpu()); @@ -312,6 +348,13 @@ void flush_thread(void) struct task_struct *tsk = current; struct thread_info *t = current_thread_info(); + /* + * Remove function-return probe instances associated with this task + * and put them back on the free list. Do not insert an exit probe for + * this function, it will be disabled by kprobe_flush_task if you do. + */ + kprobe_flush_task(tsk); + if (t->flags & _TIF_ABI_PENDING) t->flags ^= (_TIF_ABI_PENDING | _TIF_IA32); @@ -656,7 +699,7 @@ long do_arch_prctl(struct task_struct *task, int code, unsigned long addr) switch (code) { case ARCH_SET_GS: - if (addr >= TASK_SIZE) + if (addr >= TASK_SIZE_OF(task)) return -EPERM; cpu = get_cpu(); /* handle small bases via the GDT because that's faster to @@ -682,7 +725,7 @@ long do_arch_prctl(struct task_struct *task, int code, unsigned long addr) case ARCH_SET_FS: /* Not strictly needed for fs, but do it for symmetry with gs */ - if (addr >= TASK_SIZE) + if (addr >= TASK_SIZE_OF(task)) return -EPERM; cpu = get_cpu(); /* handle small bases via the GDT because that's faster to diff --git a/arch/x86_64/kernel/ptrace.c b/arch/x86_64/kernel/ptrace.c index 525f6a128a2..bbf64b59a21 100644 --- a/arch/x86_64/kernel/ptrace.c +++ b/arch/x86_64/kernel/ptrace.c @@ -257,12 +257,12 @@ static int putreg(struct task_struct *child, value &= 0xffff; return 0; case offsetof(struct user_regs_struct,fs_base): - if (value >= TASK_SIZE) + if (value >= TASK_SIZE_OF(child)) return -EIO; child->thread.fs = value; return 0; case offsetof(struct user_regs_struct,gs_base): - if (value >= TASK_SIZE) + if (value >= TASK_SIZE_OF(child)) return -EIO; child->thread.gs = value; return 0; @@ -279,7 +279,7 @@ static int putreg(struct task_struct *child, break; case offsetof(struct user_regs_struct, rip): /* Check if the new RIP address is canonical */ - if (value >= TASK_SIZE) + if (value >= TASK_SIZE_OF(child)) return -EIO; break; } @@ -419,6 +419,8 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data break; case PTRACE_POKEUSR: /* write the word at location addr in the USER area */ + { + int dsize = test_tsk_thread_flag(child, TIF_IA32) ? 3 : 7; ret = -EIO; if ((addr & 7) || addr > sizeof(struct user) - 7) @@ -430,22 +432,22 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data break; /* Disallows to set a breakpoint into the vsyscall */ case offsetof(struct user, u_debugreg[0]): - if (data >= TASK_SIZE-7) break; + if (data >= TASK_SIZE_OF(child) - dsize) break; child->thread.debugreg0 = data; ret = 0; break; case offsetof(struct user, u_debugreg[1]): - if (data >= TASK_SIZE-7) break; + if (data >= TASK_SIZE_OF(child) - dsize) break; child->thread.debugreg1 = data; ret = 0; break; case offsetof(struct user, u_debugreg[2]): - if (data >= TASK_SIZE-7) break; + if (data >= TASK_SIZE_OF(child) - dsize) break; child->thread.debugreg2 = data; ret = 0; break; case offsetof(struct user, u_debugreg[3]): - if (data >= TASK_SIZE-7) break; + if (data >= TASK_SIZE_OF(child) - dsize) break; child->thread.debugreg3 = data; ret = 0; break; @@ -469,6 +471,7 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data break; } break; + } case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: /* restart after signal. */ diff --git a/arch/x86_64/kernel/reboot.c b/arch/x86_64/kernel/reboot.c index be4b36f762c..57e71dbdfd6 100644 --- a/arch/x86_64/kernel/reboot.c +++ b/arch/x86_64/kernel/reboot.c @@ -66,41 +66,47 @@ static int __init reboot_setup(char *str) __setup("reboot=", reboot_setup); -#ifdef CONFIG_SMP -static void smp_halt(void) +static inline void kb_wait(void) { - int cpuid = safe_smp_processor_id(); - static int first_entry = 1; + int i; - if (reboot_force) - return; + for (i=0; i<0x10000; i++) + if ((inb_p(0x64) & 0x02) == 0) + break; +} - if (first_entry) { - first_entry = 0; - smp_call_function((void *)machine_restart, NULL, 1, 0); - } - - smp_stop_cpu(); +void machine_shutdown(void) +{ + /* Stop the cpus and apics */ +#ifdef CONFIG_SMP + int reboot_cpu_id; - /* AP calling this. Just halt */ - if (cpuid != boot_cpu_id) { - for (;;) - asm("hlt"); + /* The boot cpu is always logical cpu 0 */ + reboot_cpu_id = 0; + + /* Make certain the cpu I'm about to reboot on is online */ + if (!cpu_isset(reboot_cpu_id, cpu_online_map)) { + reboot_cpu_id = smp_processor_id(); } - /* Wait for all other CPUs to have run smp_stop_cpu */ - while (!cpus_empty(cpu_online_map)) - rep_nop(); -} + /* Make certain I only run on the appropriate processor */ + set_cpus_allowed(current, cpumask_of_cpu(reboot_cpu_id)); + + /* O.K Now that I'm on the appropriate processor, + * stop all of the others. + */ + smp_send_stop(); #endif -static inline void kb_wait(void) -{ - int i; + local_irq_disable(); - for (i=0; i<0x10000; i++) - if ((inb_p(0x64) & 0x02) == 0) - break; +#ifndef CONFIG_SMP + disable_local_APIC(); +#endif + + disable_IO_APIC(); + + local_irq_enable(); } void machine_restart(char * __unused) @@ -109,9 +115,7 @@ void machine_restart(char * __unused) printk("machine restart\n"); -#ifdef CONFIG_SMP - smp_halt(); -#endif + machine_shutdown(); if (!reboot_force) { local_irq_disable(); diff --git a/arch/x86_64/kernel/relocate_kernel.S b/arch/x86_64/kernel/relocate_kernel.S new file mode 100644 index 00000000000..d24fa9b72a2 --- /dev/null +++ b/arch/x86_64/kernel/relocate_kernel.S @@ -0,0 +1,143 @@ +/* + * relocate_kernel.S - put the kernel image in place to boot + * Copyright (C) 2002-2005 Eric Biederman <ebiederm@xmission.com> + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include <linux/linkage.h> + + /* + * Must be relocatable PIC code callable as a C function, that once + * it starts can not use the previous processes stack. + */ + .globl relocate_new_kernel + .code64 +relocate_new_kernel: + /* %rdi page_list + * %rsi reboot_code_buffer + * %rdx start address + * %rcx page_table + * %r8 arg5 + * %r9 arg6 + */ + + /* zero out flags, and disable interrupts */ + pushq $0 + popfq + + /* set a new stack at the bottom of our page... */ + lea 4096(%rsi), %rsp + + /* store the parameters back on the stack */ + pushq %rdx /* store the start address */ + + /* 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 + */ + movq %cr0, %rax + andq $~((1<<18)|(1<<16)|(1<<3)|(1<<2)), %rax + orl $((1<<31)|(1<<0)), %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 + */ + + movq $((1<<5)), %rax + movq %rax, %cr4 + + jmp 1f +1: + + /* Switch to the identity mapped page tables, + * and flush the TLB. + */ + movq %rcx, %cr3 + + /* Do the copies */ + movq %rdi, %rcx /* Put the page_list in %rcx */ + xorq %rdi, %rdi + xorq %rsi, %rsi + jmp 1f + +0: /* top, read another word for the indirection page */ + + movq (%rbx), %rcx + addq $8, %rbx +1: + testq $0x1, %rcx /* is it a destination page? */ + jz 2f + movq %rcx, %rdi + andq $0xfffffffffffff000, %rdi + jmp 0b +2: + testq $0x2, %rcx /* is it an indirection page? */ + jz 2f + movq %rcx, %rbx + andq $0xfffffffffffff000, %rbx + jmp 0b +2: + testq $0x4, %rcx /* is it the done indicator? */ + jz 2f + jmp 3f +2: + testq $0x8, %rcx /* is it the source indicator? */ + jz 0b /* Ignore it otherwise */ + movq %rcx, %rsi /* For ever source page do a copy */ + andq $0xfffffffffffff000, %rsi + + movq $512, %rcx + rep ; movsq + jmp 0b +3: + + /* To be certain of avoiding problems with self-modifying code + * I need to execute a serializing instruction here. + * So I flush the TLB by reloading %cr3 here, it's handy, + * and not processor dependent. + */ + movq %cr3, %rax + movq %rax, %cr3 + + /* set all of the registers to known values */ + /* leave %rsp alone */ + + xorq %rax, %rax + xorq %rbx, %rbx + xorq %rcx, %rcx + xorq %rdx, %rdx + xorq %rsi, %rsi + xorq %rdi, %rdi + xorq %rbp, %rbp + xorq %r8, %r8 + xorq %r9, %r9 + xorq %r10, %r9 + xorq %r11, %r11 + xorq %r12, %r12 + xorq %r13, %r13 + xorq %r14, %r14 + xorq %r15, %r15 + + ret +relocate_new_kernel_end: + + .globl relocate_new_kernel_size +relocate_new_kernel_size: + .quad relocate_new_kernel_end - relocate_new_kernel diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index 99f038ede23..b02d921da4f 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -40,6 +40,9 @@ #include <linux/acpi.h> #include <linux/kallsyms.h> #include <linux/edd.h> +#include <linux/mmzone.h> +#include <linux/kexec.h> + #include <asm/mtrr.h> #include <asm/uaccess.h> #include <asm/system.h> @@ -348,7 +351,7 @@ static __init void parse_cmdline_early (char ** cmdline_p) if (!memcmp(from, "mem=", 4)) parse_memopt(from+4, &from); -#ifdef CONFIG_DISCONTIGMEM +#ifdef CONFIG_NUMA if (!memcmp(from, "numa=", 5)) numa_setup(from+5); #endif @@ -365,6 +368,27 @@ static __init void parse_cmdline_early (char ** cmdline_p) if (!memcmp(from, "noexec=", 7)) nonx_setup(from + 7); +#ifdef CONFIG_KEXEC + /* crashkernel=size@addr specifies the location to reserve for + * a crash kernel. By reserving this memory we guarantee + * that linux never set's it up as a DMA target. + * Useful for holding code to do something appropriate + * after a kernel panic. + */ + else if (!memcmp(from, "crashkernel=", 12)) { + unsigned long size, base; + size = memparse(from+12, &from); + if (*from == '@') { + base = memparse(from+1, &from); + /* FIXME: Do I want a sanity check + * to validate the memory range? + */ + crashk_res.start = base; + crashk_res.end = base + size - 1; + } + } +#endif + next_char: c = *(from++); if (!c) @@ -377,17 +401,20 @@ static __init void parse_cmdline_early (char ** cmdline_p) *cmdline_p = command_line; } -#ifndef CONFIG_DISCONTIGMEM -static void __init contig_initmem_init(void) +#ifndef CONFIG_NUMA +static void __init +contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn) { - unsigned long bootmap_size, bootmap; - bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT; - bootmap = find_e820_area(0, end_pfn<<PAGE_SHIFT, bootmap_size); - if (bootmap == -1L) - panic("Cannot find bootmem map of size %ld\n",bootmap_size); - bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn); - e820_bootmem_free(&contig_page_data, 0, end_pfn << PAGE_SHIFT); - reserve_bootmem(bootmap, bootmap_size); + unsigned long bootmap_size, bootmap; + + memory_present(0, start_pfn, end_pfn); + bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT; + bootmap = find_e820_area(0, end_pfn<<PAGE_SHIFT, bootmap_size); + if (bootmap == -1L) + panic("Cannot find bootmem map of size %ld\n",bootmap_size); + bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn); + e820_bootmem_free(NODE_DATA(0), 0, end_pfn << PAGE_SHIFT); + reserve_bootmem(bootmap, bootmap_size); } #endif @@ -554,10 +581,10 @@ void __init setup_arch(char **cmdline_p) acpi_numa_init(); #endif -#ifdef CONFIG_DISCONTIGMEM +#ifdef CONFIG_NUMA numa_initmem_init(0, end_pfn); #else - contig_initmem_init(); + contig_initmem_init(0, end_pfn); #endif /* Reserve direct mapping */ @@ -618,6 +645,15 @@ void __init setup_arch(char **cmdline_p) } } #endif + + sparse_init(); + +#ifdef CONFIG_KEXEC + if (crashk_res.start != crashk_res.end) { + reserve_bootmem(crashk_res.start, + crashk_res.end - crashk_res.start + 1); + } +#endif paging_init(); check_ioapic(); @@ -669,7 +705,7 @@ void __init setup_arch(char **cmdline_p) #endif } -static int __init get_model_name(struct cpuinfo_x86 *c) +static int __cpuinit get_model_name(struct cpuinfo_x86 *c) { unsigned int *v; @@ -685,7 +721,7 @@ static int __init get_model_name(struct cpuinfo_x86 *c) } -static void __init display_cacheinfo(struct cpuinfo_x86 *c) +static void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) { unsigned int n, dummy, eax, ebx, ecx, edx; @@ -796,7 +832,7 @@ static int __init init_amd(struct cpuinfo_x86 *c) return r; } -static void __init detect_ht(struct cpuinfo_x86 *c) +static void __cpuinit detect_ht(struct cpuinfo_x86 *c) { #ifdef CONFIG_SMP u32 eax, ebx, ecx, edx; @@ -857,7 +893,7 @@ static void __init detect_ht(struct cpuinfo_x86 *c) /* * find out the number of processor cores on the die */ -static int __init intel_num_cpu_cores(struct cpuinfo_x86 *c) +static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c) { unsigned int eax; @@ -875,7 +911,7 @@ static int __init intel_num_cpu_cores(struct cpuinfo_x86 *c) return 1; } -static void __init init_intel(struct cpuinfo_x86 *c) +static void __cpuinit init_intel(struct cpuinfo_x86 *c) { /* Cache sizes */ unsigned n; @@ -895,7 +931,7 @@ static void __init init_intel(struct cpuinfo_x86 *c) c->x86_num_cores = intel_num_cpu_cores(c); } -void __init get_cpu_vendor(struct cpuinfo_x86 *c) +void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) { char *v = c->x86_vendor_id; @@ -916,7 +952,7 @@ struct cpu_model_info { /* Do some early cpuid on the boot CPU to get some parameter that are needed before check_bugs. Everything advanced is in identify_cpu below. */ -void __init early_identify_cpu(struct cpuinfo_x86 *c) +void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) { u32 tfms; @@ -970,7 +1006,7 @@ void __init early_identify_cpu(struct cpuinfo_x86 *c) /* * This does the hard work of actually picking apart the CPU stuff... */ -void __init identify_cpu(struct cpuinfo_x86 *c) +void __cpuinit identify_cpu(struct cpuinfo_x86 *c) { int i; u32 xlvl; @@ -1047,7 +1083,7 @@ void __init identify_cpu(struct cpuinfo_x86 *c) } -void __init print_cpu_info(struct cpuinfo_x86 *c) +void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) { if (c->x86_model_id[0]) printk("%s", c->x86_model_id); diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c index 678b7ac33b8..f2b9c6bc999 100644 --- a/arch/x86_64/kernel/setup64.c +++ b/arch/x86_64/kernel/setup64.c @@ -29,7 +29,7 @@ char x86_boot_params[BOOT_PARAM_SIZE] __initdata = {0,}; -cpumask_t cpu_initialized __initdata = CPU_MASK_NONE; +cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE; struct x8664_pda cpu_pda[NR_CPUS] __cacheline_aligned; @@ -171,7 +171,7 @@ void syscall_init(void) wrmsrl(MSR_SYSCALL_MASK, EF_TF|EF_DF|EF_IE|0x3000); } -void __init check_efer(void) +void __cpuinit check_efer(void) { unsigned long efer; @@ -188,7 +188,7 @@ void __init check_efer(void) * 'CPU state barrier', nothing should get across. * A lot of state is already set up in PDA init. */ -void __init cpu_init (void) +void __cpuinit cpu_init (void) { #ifdef CONFIG_SMP int cpu = stack_smp_processor_id(); diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c index 3fdcdba0fec..98590a989f3 100644 --- a/arch/x86_64/kernel/signal.c +++ b/arch/x86_64/kernel/signal.c @@ -28,14 +28,15 @@ #include <asm/uaccess.h> #include <asm/i387.h> #include <asm/proto.h> +#include <asm/ia32_unistd.h> /* #define DEBUG_SIG 1 */ #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) -void ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, +int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, struct pt_regs * regs); -void ia32_setup_frame(int sig, struct k_sigaction *ka, +int ia32_setup_frame(int sig, struct k_sigaction *ka, sigset_t *set, struct pt_regs * regs); asmlinkage long @@ -237,7 +238,7 @@ get_stack(struct k_sigaction *ka, struct pt_regs *regs, unsigned long size) return (void __user *)round_down(rsp - size, 16); } -static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, +static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, struct pt_regs * regs) { struct rt_sigframe __user *frame; @@ -326,20 +327,23 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, current->comm, current->pid, frame, regs->rip, frame->pretcode); #endif - return; + return 1; give_sigsegv: force_sigsegv(sig, current); + return 0; } /* * OK, we're invoking a handler */ -static void +static int handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, sigset_t *oldset, struct pt_regs *regs) { + int ret; + #ifdef DEBUG_SIG printk("handle_signal pid:%d sig:%lu rip:%lx rsp:%lx regs=%p\n", current->pid, sig, @@ -383,20 +387,22 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, #ifdef CONFIG_IA32_EMULATION if (test_thread_flag(TIF_IA32)) { if (ka->sa.sa_flags & SA_SIGINFO) - ia32_setup_rt_frame(sig, ka, info, oldset, regs); + ret = ia32_setup_rt_frame(sig, ka, info, oldset, regs); else - ia32_setup_frame(sig, ka, oldset, regs); + ret = ia32_setup_frame(sig, ka, oldset, regs); } else #endif - setup_rt_frame(sig, ka, info, oldset, regs); + ret = setup_rt_frame(sig, ka, info, oldset, regs); - if (!(ka->sa.sa_flags & SA_NODEFER)) { + if (ret && !(ka->sa.sa_flags & SA_NODEFER)) { spin_lock_irq(¤t->sighand->siglock); sigorsets(¤t->blocked,¤t->blocked,&ka->sa.sa_mask); sigaddset(¤t->blocked,sig); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); } + + return ret; } /* @@ -416,10 +422,10 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset) * kernel mode. Just return without doing anything * if so. */ - if ((regs->cs & 3) != 3) + if (!user_mode(regs)) return 1; - if (try_to_freeze(0)) + if (try_to_freeze()) goto no_signal; if (!oldset) @@ -433,11 +439,10 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset) * inside the kernel. */ if (current->thread.debugreg7) - asm volatile("movq %0,%%db7" : : "r" (current->thread.debugreg7)); + set_debugreg(current->thread.debugreg7, 7); /* Whee! Actually deliver the signal. */ - handle_signal(signr, &info, &ka, oldset, regs); - return 1; + return handle_signal(signr, &info, &ka, oldset, regs); } no_signal: diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c index 1e379ed17b1..ccae392886a 100644 --- a/arch/x86_64/kernel/smp.c +++ b/arch/x86_64/kernel/smp.c @@ -283,6 +283,16 @@ struct call_data_struct { 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 CPUs * in the system. diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index f1ec0f34594..b969ee12872 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -34,6 +34,7 @@ * Andi Kleen : Converted to new state machine. * Various cleanups. * Probably mostly hotplug CPU ready now. + * Ashok Raj : CPU hotplug support */ @@ -58,11 +59,6 @@ #include <asm/proto.h> #include <asm/nmi.h> -/* Change for real CPU hotplug. Note other files need to be fixed - first too. */ -#define __cpuinit __init -#define __cpuinitdata __initdata - /* Number of siblings per CPU package */ int smp_num_siblings = 1; /* Package ID of each logical CPU */ @@ -103,6 +99,37 @@ EXPORT_SYMBOL(cpu_core_map); extern unsigned char trampoline_data[]; extern 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. + */ +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)) + +/* + * 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. + * If cpu-hotplug is supported, then we need to preallocate for all + * those NR_CPUS, hence cpu_possible_map represents entire NR_CPUS range. + * - Ashok Raj + */ +#ifdef CONFIG_HOTPLUG_CPU +#define fixup_cpu_possible_map(x) cpu_set((x), cpu_possible_map) +#else +#define fixup_cpu_possible_map(x) +#endif + /* * Currently trivial. Write the real->protected mode * bootstrap into the page concerned. The caller @@ -418,6 +445,33 @@ void __cpuinit smp_callin(void) cpu_set(cpuid, cpu_callin_map); } +static inline void set_cpu_sibling_map(int cpu) +{ + int i; + + if (smp_num_siblings > 1) { + for_each_cpu(i) { + if (cpu_core_id[cpu] == cpu_core_id[i]) { + cpu_set(i, cpu_sibling_map[cpu]); + cpu_set(cpu, cpu_sibling_map[i]); + } + } + } else { + cpu_set(cpu, cpu_sibling_map[cpu]); + } + + if (current_cpu_data.x86_num_cores > 1) { + for_each_cpu(i) { + if (phys_proc_id[cpu] == phys_proc_id[i]) { + cpu_set(i, cpu_core_map[cpu]); + cpu_set(cpu, cpu_core_map[i]); + } + } + } else { + cpu_core_map[cpu] = cpu_sibling_map[cpu]; + } +} + /* * Setup code on secondary processor (after comming out of the trampoline) */ @@ -448,9 +502,28 @@ void __cpuinit start_secondary(void) enable_APIC_timer(); /* + * 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 receipients, 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(); + + /* * Allow the master to continue. */ cpu_set(smp_processor_id(), cpu_online_map); + per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE; + unlock_ipi_call_lock(); + mb(); /* Wait for TSC sync to not schedule things before. @@ -628,33 +701,77 @@ static int __cpuinit wakeup_secondary_via_INIT(int phys_apicid, unsigned int sta return (send_status | accept_status); } +struct create_idle { + struct task_struct *idle; + struct completion done; + int cpu; +}; + +void do_fork_idle(void *_c_idle) +{ + struct create_idle *c_idle = _c_idle; + + 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) { - struct task_struct *idle; unsigned long boot_error; int timeout; unsigned long start_rip; + struct create_idle c_idle = { + .cpu = cpu, + .done = COMPLETION_INITIALIZER(c_idle.done), + }; + DECLARE_WORK(work, do_fork_idle, &c_idle); + + c_idle.idle = get_idle_for_cpu(cpu); + + if (c_idle.idle) { + c_idle.idle->thread.rsp = (unsigned long) (((struct pt_regs *) + (THREAD_SIZE + (unsigned long) c_idle.idle->thread_info)) - 1); + init_idle(c_idle.idle, cpu); + goto do_rest; + } + /* - * We can't use kernel_thread since we must avoid to - * reschedule the child. + * 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. */ - idle = fork_idle(cpu); - if (IS_ERR(idle)) { + if (!keventd_up() || current_is_keventd()) + work.func(work.data); + else { + schedule_work(&work); + wait_for_completion(&c_idle.done); + } + + if (IS_ERR(c_idle.idle)) { printk("failed fork for CPU %d\n", cpu); - return PTR_ERR(idle); + return PTR_ERR(c_idle.idle); } - cpu_pda[cpu].pcurrent = idle; + set_idle_for_cpu(cpu, c_idle.idle); + +do_rest: + + cpu_pda[cpu].pcurrent = c_idle.idle; start_rip = setup_trampoline(); - init_rsp = idle->thread.rsp; + init_rsp = c_idle.idle->thread.rsp; per_cpu(init_tss,cpu).rsp0 = init_rsp; initial_code = start_secondary; - clear_ti_thread_flag(idle->thread_info, TIF_FORK); + clear_ti_thread_flag(c_idle.idle->thread_info, TIF_FORK); printk(KERN_INFO "Booting processor %d/%d rip %lx rsp %lx\n", cpu, apicid, start_rip, init_rsp); @@ -746,51 +863,6 @@ cycles_t cacheflush_time; unsigned long cache_decay_ticks; /* - * Construct cpu_sibling_map[], so that we can tell the sibling CPU - * on SMT systems efficiently. - */ -static __cpuinit void detect_siblings(void) -{ - int cpu; - - for (cpu = 0; cpu < NR_CPUS; cpu++) { - cpus_clear(cpu_sibling_map[cpu]); - cpus_clear(cpu_core_map[cpu]); - } - - for_each_online_cpu (cpu) { - struct cpuinfo_x86 *c = cpu_data + cpu; - int siblings = 0; - int i; - if (smp_num_siblings > 1) { - for_each_online_cpu (i) { - if (cpu_core_id[cpu] == cpu_core_id[i]) { - siblings++; - cpu_set(i, cpu_sibling_map[cpu]); - } - } - } else { - siblings++; - cpu_set(cpu, cpu_sibling_map[cpu]); - } - - if (siblings != smp_num_siblings) { - printk(KERN_WARNING - "WARNING: %d siblings found for CPU%d, should be %d\n", - siblings, cpu, smp_num_siblings); - smp_num_siblings = siblings; - } - if (c->x86_num_cores > 1) { - for_each_online_cpu(i) { - if (phys_proc_id[cpu] == phys_proc_id[i]) - cpu_set(i, cpu_core_map[cpu]); - } - } else - cpu_core_map[cpu] = cpu_sibling_map[cpu]; - } -} - -/* * Cleanup possible dangling ends... */ static __cpuinit void smp_cleanup_boot(void) @@ -823,7 +895,7 @@ static __cpuinit void smp_cleanup_boot(void) * * RED-PEN audit/test this more. I bet there is more state messed up here. */ -static __cpuinit void disable_smp(void) +static __init void disable_smp(void) { cpu_present_map = cpumask_of_cpu(0); cpu_possible_map = cpumask_of_cpu(0); @@ -838,7 +910,7 @@ static __cpuinit void disable_smp(void) /* * Handle user cpus=... parameter. */ -static __cpuinit void enforce_max_cpus(unsigned max_cpus) +static __init void enforce_max_cpus(unsigned max_cpus) { int i, k; k = 0; @@ -855,7 +927,7 @@ static __cpuinit void enforce_max_cpus(unsigned max_cpus) /* * Various sanity checks. */ -static int __cpuinit smp_sanity_check(unsigned max_cpus) +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", @@ -913,7 +985,7 @@ static int __cpuinit smp_sanity_check(unsigned max_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 __cpuinit smp_prepare_cpus(unsigned int max_cpus) +void __init smp_prepare_cpus(unsigned int max_cpus) { int i; @@ -930,10 +1002,9 @@ void __cpuinit smp_prepare_cpus(unsigned int max_cpus) int apicid = cpu_present_to_apicid(i); if (physid_isset(apicid, phys_cpu_present_map)) { cpu_set(i, cpu_present_map); - /* possible map would be different if we supported real - CPU hotplug. */ cpu_set(i, cpu_possible_map); } + fixup_cpu_possible_map(i); } if (smp_sanity_check(max_cpus) < 0) { @@ -978,13 +1049,13 @@ void __init smp_prepare_boot_cpu(void) int me = smp_processor_id(); cpu_set(me, cpu_online_map); cpu_set(me, cpu_callout_map); + cpu_set(0, cpu_sibling_map[0]); + cpu_set(0, cpu_core_map[0]); + per_cpu(cpu_state, me) = CPU_ONLINE; } /* * Entry point to boot a CPU. - * - * This is all __cpuinit, not __devinit for now because we don't support - * CPU hotplug (yet). */ int __cpuinit __cpu_up(unsigned int cpu) { @@ -1001,6 +1072,15 @@ int __cpuinit __cpu_up(unsigned int cpu) return -EINVAL; } + /* + * Already booted CPU? + */ + if (cpu_isset(cpu, cpu_callin_map)) { + Dprintk("do_boot_cpu %d Already started\n", cpu); + return -ENOSYS; + } + + per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; /* Boot it! */ err = do_boot_cpu(cpu, apicid); if (err < 0) { @@ -1013,23 +1093,118 @@ int __cpuinit __cpu_up(unsigned int cpu) while (!cpu_isset(cpu, cpu_online_map)) cpu_relax(); - return 0; + err = 0; + + return err; } /* * Finish the SMP boot. */ -void __cpuinit smp_cpus_done(unsigned int max_cpus) +void __init smp_cpus_done(unsigned int max_cpus) { +#ifndef CONFIG_HOTPLUG_CPU zap_low_mappings(); +#endif smp_cleanup_boot(); #ifdef CONFIG_X86_IO_APIC setup_ioapic_dest(); #endif - detect_siblings(); time_init_gtod(); check_nmi_watchdog(); } + +#ifdef CONFIG_HOTPLUG_CPU + +static void remove_siblinginfo(int cpu) +{ + int sibling; + + for_each_cpu_mask(sibling, cpu_sibling_map[cpu]) + cpu_clear(cpu, cpu_sibling_map[sibling]); + for_each_cpu_mask(sibling, cpu_core_map[cpu]) + cpu_clear(cpu, cpu_core_map[sibling]); + cpus_clear(cpu_sibling_map[cpu]); + cpus_clear(cpu_core_map[cpu]); + phys_proc_id[cpu] = BAD_APICID; + cpu_core_id[cpu] = BAD_APICID; +} + +void 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, &cpu_initialized); /* was set by cpu_init() */ +} + +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; + + disable_APIC_timer(); + + /* + * HACK: + * Allow any queued timer interrupts to get serviced + * This is only a temporary solution until we cleanup + * fixup_irqs as we do for IA64. + */ + local_irq_enable(); + mdelay(1); + + local_irq_disable(); + remove_siblinginfo(cpu); + + /* It's now safe to remove this processor from the online map */ + cpu_clear(cpu, cpu_online_map); + 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); + return; + } + current->state = TASK_UNINTERRUPTIBLE; + schedule_timeout(HZ/10); + } + printk(KERN_ERR "CPU %u didn't die...\n", cpu); +} + +#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_64/kernel/suspend.c b/arch/x86_64/kernel/suspend.c index ebaa1e37d65..6c0f402e3a8 100644 --- a/arch/x86_64/kernel/suspend.c +++ b/arch/x86_64/kernel/suspend.c @@ -44,7 +44,6 @@ void __save_processor_state(struct saved_context *ctxt) */ asm volatile ("sgdt %0" : "=m" (ctxt->gdt_limit)); asm volatile ("sidt %0" : "=m" (ctxt->idt_limit)); - asm volatile ("sldt %0" : "=m" (ctxt->ldt)); asm volatile ("str %0" : "=m" (ctxt->tr)); /* XMM0..XMM15 should be handled by kernel_fpu_begin(). */ @@ -69,6 +68,7 @@ void __save_processor_state(struct saved_context *ctxt) asm volatile ("movq %%cr2, %0" : "=r" (ctxt->cr2)); asm volatile ("movq %%cr3, %0" : "=r" (ctxt->cr3)); asm volatile ("movq %%cr4, %0" : "=r" (ctxt->cr4)); + asm volatile ("movq %%cr8, %0" : "=r" (ctxt->cr8)); } void save_processor_state(void) @@ -90,12 +90,20 @@ void __restore_processor_state(struct saved_context *ctxt) /* * control registers */ + asm volatile ("movq %0, %%cr8" :: "r" (ctxt->cr8)); asm volatile ("movq %0, %%cr4" :: "r" (ctxt->cr4)); asm volatile ("movq %0, %%cr3" :: "r" (ctxt->cr3)); asm volatile ("movq %0, %%cr2" :: "r" (ctxt->cr2)); asm volatile ("movq %0, %%cr0" :: "r" (ctxt->cr0)); /* + * now restore the descriptor tables to their proper values + * ltr is done i fix_processor_context(). + */ + asm volatile ("lgdt %0" :: "m" (ctxt->gdt_limit)); + asm volatile ("lidt %0" :: "m" (ctxt->idt_limit)); + + /* * segment registers */ asm volatile ("movw %0, %%ds" :: "r" (ctxt->ds)); @@ -108,14 +116,6 @@ void __restore_processor_state(struct saved_context *ctxt) wrmsrl(MSR_GS_BASE, ctxt->gs_base); wrmsrl(MSR_KERNEL_GS_BASE, ctxt->gs_kernel_base); - /* - * now restore the descriptor tables to their proper values - * ltr is done i fix_processor_context(). - */ - asm volatile ("lgdt %0" :: "m" (ctxt->gdt_limit)); - asm volatile ("lidt %0" :: "m" (ctxt->idt_limit)); - asm volatile ("lldt %0" :: "m" (ctxt->ldt)); - fix_processor_context(); do_fpu_end(); diff --git a/arch/x86_64/kernel/sys_x86_64.c b/arch/x86_64/kernel/sys_x86_64.c index dbebd5ccba6..cc7821c6885 100644 --- a/arch/x86_64/kernel/sys_x86_64.c +++ b/arch/x86_64/kernel/sys_x86_64.c @@ -68,13 +68,7 @@ out: static void find_start_end(unsigned long flags, unsigned long *begin, unsigned long *end) { -#ifdef CONFIG_IA32_EMULATION - if (test_thread_flag(TIF_IA32)) { - *begin = TASK_UNMAPPED_32; - *end = IA32_PAGE_OFFSET; - } else -#endif - if (flags & MAP_32BIT) { + if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) { /* This is usually used needed to map code in small model, so it needs to be in the first 31bit. Limit it to that. This means we need to move the @@ -84,10 +78,10 @@ static void find_start_end(unsigned long flags, unsigned long *begin, of playground for now. -AK */ *begin = 0x40000000; *end = 0x80000000; - } else { - *begin = TASK_UNMAPPED_64; + } else { + *begin = TASK_UNMAPPED_BASE; *end = TASK_SIZE; - } + } } unsigned long @@ -111,6 +105,11 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, (!vma || addr + len <= vma->vm_start)) return addr; } + if (((flags & MAP_32BIT) || test_thread_flag(TIF_IA32)) + && len <= mm->cached_hole_size) { + mm->cached_hole_size = 0; + mm->free_area_cache = begin; + } addr = mm->free_area_cache; if (addr < begin) addr = begin; @@ -126,6 +125,7 @@ full_search: */ if (start_addr != begin) { start_addr = addr = begin; + mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; @@ -137,6 +137,9 @@ full_search: mm->free_area_cache = addr + len; return addr; } + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; + addr = vma->vm_end; } } diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index 735b6767c8e..66bf6ddeb0c 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -27,7 +27,9 @@ #include <linux/bcd.h> #include <linux/kallsyms.h> #include <linux/acpi.h> +#ifdef CONFIG_ACPI #include <acpi/achware.h> /* for PM timer frequency */ +#endif #include <asm/8253pit.h> #include <asm/pgtable.h> #include <asm/vsyscall.h> @@ -62,6 +64,7 @@ static int notsc __initdata = 0; unsigned int cpu_khz; /* TSC clocks / usec, not used here */ static unsigned long hpet_period; /* fsecs / HPET clock */ unsigned long hpet_tick; /* HPET clocks / interrupt */ +static int hpet_use_timer; unsigned long vxtime_hz = PIT_TICK_RATE; int report_lost_ticks; /* command line option */ unsigned long long monotonic_base; @@ -103,7 +106,9 @@ static inline unsigned int do_gettimeoffset_tsc(void) static inline unsigned int do_gettimeoffset_hpet(void) { - return ((hpet_readl(HPET_COUNTER) - vxtime.last) * vxtime.quot) >> 32; + /* cap counter read to one tick to avoid inconsistencies */ + unsigned long counter = hpet_readl(HPET_COUNTER) - vxtime.last; + return (min(counter,hpet_tick) * vxtime.quot) >> 32; } unsigned int (*do_gettimeoffset)(void) = do_gettimeoffset_tsc; @@ -299,7 +304,7 @@ unsigned long long monotonic_clock(void) last_offset = vxtime.last; base = monotonic_base; - this_offset = hpet_readl(HPET_T0_CMP) - hpet_tick; + this_offset = hpet_readl(HPET_COUNTER); } while (read_seqretry(&xtime_lock, seq)); offset = (this_offset - last_offset); @@ -375,7 +380,14 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) write_seqlock(&xtime_lock); - if (vxtime.hpet_address) { + if (vxtime.hpet_address) + offset = hpet_readl(HPET_COUNTER); + + if (hpet_use_timer) { + /* if we're using the hpet timer functionality, + * we can more accurately know the counter value + * when the timer interrupt occured. + */ offset = hpet_readl(HPET_T0_CMP) - hpet_tick; delay = hpet_readl(HPET_COUNTER) - offset; } else { @@ -801,17 +813,18 @@ static int hpet_timer_stop_set_go(unsigned long tick) * Set up timer 0, as periodic with first interrupt to happen at hpet_tick, * and period also hpet_tick. */ - - hpet_writel(HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL | + if (hpet_use_timer) { + hpet_writel(HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL | HPET_TN_32BIT, HPET_T0_CFG); - hpet_writel(hpet_tick, HPET_T0_CMP); - hpet_writel(hpet_tick, HPET_T0_CMP); /* AK: why twice? */ - + hpet_writel(hpet_tick, HPET_T0_CMP); + hpet_writel(hpet_tick, HPET_T0_CMP); /* AK: why twice? */ + cfg |= HPET_CFG_LEGACY; + } /* * Go! */ - cfg |= HPET_CFG_ENABLE | HPET_CFG_LEGACY; + cfg |= HPET_CFG_ENABLE; hpet_writel(cfg, HPET_CFG); return 0; @@ -832,8 +845,7 @@ static int hpet_init(void) id = hpet_readl(HPET_ID); - if (!(id & HPET_ID_VENDOR) || !(id & HPET_ID_NUMBER) || - !(id & HPET_ID_LEGSUP)) + if (!(id & HPET_ID_VENDOR) || !(id & HPET_ID_NUMBER)) return -1; hpet_period = hpet_readl(HPET_PERIOD); @@ -843,6 +855,8 @@ static int hpet_init(void) hpet_tick = (1000000000L * (USEC_PER_SEC / HZ) + hpet_period / 2) / hpet_period; + hpet_use_timer = (id & HPET_ID_LEGSUP); + return hpet_timer_stop_set_go(hpet_tick); } @@ -899,9 +913,11 @@ void __init time_init(void) set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec); - if (!hpet_init()) { + if (!hpet_init()) vxtime_hz = (1000000000000000L + hpet_period / 2) / hpet_period; + + if (hpet_use_timer) { cpu_khz = hpet_calibrate_tsc(); timename = "HPET"; #ifdef CONFIG_X86_PM_TIMER @@ -966,7 +982,7 @@ void __init time_init_gtod(void) if (unsynchronized_tsc()) notsc = 1; if (vxtime.hpet_address && notsc) { - timetype = "HPET"; + timetype = hpet_use_timer ? "HPET" : "PIT/HPET"; vxtime.last = hpet_readl(HPET_T0_CMP) - hpet_tick; vxtime.mode = VXTIME_HPET; do_gettimeoffset = do_gettimeoffset_hpet; @@ -981,7 +997,7 @@ void __init time_init_gtod(void) printk(KERN_INFO "Disabling vsyscall due to use of PM timer\n"); #endif } else { - timetype = vxtime.hpet_address ? "HPET/TSC" : "PIT/TSC"; + timetype = hpet_use_timer ? "HPET/TSC" : "PIT/TSC"; vxtime.mode = VXTIME_TSC; } diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 3dfec8fdabc..10273663000 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -274,7 +274,7 @@ EXPORT_SYMBOL(dump_stack); void show_registers(struct pt_regs *regs) { int i; - int in_kernel = (regs->cs & 3) == 0; + int in_kernel = !user_mode(regs); unsigned long rsp; const int cpu = safe_smp_processor_id(); struct task_struct *cur = cpu_pda[cpu].pcurrent; @@ -318,7 +318,7 @@ void handle_BUG(struct pt_regs *regs) struct bug_frame f; char tmp; - if (regs->cs & 3) + if (user_mode(regs)) return; if (__copy_from_user(&f, (struct bug_frame *) regs->rip, sizeof(struct bug_frame))) @@ -437,7 +437,7 @@ static void do_trap(int trapnr, int signr, char *str, } #endif - if ((regs->cs & 3) != 0) { + if (user_mode(regs)) { struct task_struct *tsk = current; if (exception_trace && unhandled_signal(tsk, signr)) @@ -522,7 +522,7 @@ asmlinkage void do_general_protection(struct pt_regs * regs, long error_code) } #endif - if ((regs->cs & 3)!=0) { + if (user_mode(regs)) { struct task_struct *tsk = current; if (exception_trace && unhandled_signal(tsk, SIGSEGV)) @@ -586,11 +586,17 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs) asmlinkage void default_do_nmi(struct pt_regs *regs) { unsigned char reason = 0; + int cpu; + + cpu = smp_processor_id(); /* Only the BSP gets external NMIs from the system. */ - if (!smp_processor_id()) + if (!cpu) reason = get_nmi_reason(); + if (!cpu_online(cpu)) + return; + if (!(reason & 0xc0)) { if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 0, SIGINT) == NOTIFY_STOP) @@ -638,7 +644,7 @@ asmlinkage struct pt_regs *sync_regs(struct pt_regs *eregs) if (eregs == (struct pt_regs *)eregs->rsp) ; /* Exception from user space */ - else if (eregs->cs & 3) + else if (user_mode(eregs)) regs = ((struct pt_regs *)current->thread.rsp0) - 1; /* Exception from kernel and interrupts are enabled. Move to kernel process stack. */ @@ -669,7 +675,7 @@ asmlinkage void do_debug(struct pt_regs * regs, unsigned long error_code) } #endif - asm("movq %%db6,%0" : "=r" (condition)); + get_debugreg(condition, 6); if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code, SIGTRAP) == NOTIFY_STOP) @@ -697,7 +703,7 @@ asmlinkage void do_debug(struct pt_regs * regs, unsigned long error_code) * allowing programs to debug themselves without the ptrace() * interface. */ - if ((regs->cs & 3) == 0) + if (!user_mode(regs)) goto clear_TF_reenable; /* * Was the TF flag set by a debugger? If so, clear it now, @@ -715,13 +721,13 @@ asmlinkage void do_debug(struct pt_regs * regs, unsigned long error_code) info.si_signo = SIGTRAP; info.si_errno = 0; info.si_code = TRAP_BRKPT; - if ((regs->cs & 3) == 0) + if (!user_mode(regs)) goto clear_dr7; info.si_addr = (void __user *)regs->rip; force_sig_info(SIGTRAP, &info, tsk); clear_dr7: - asm volatile("movq %0,%%db7"::"r"(0UL)); + set_debugreg(0UL, 7); return; clear_TF_reenable: @@ -756,7 +762,7 @@ asmlinkage void do_coprocessor_error(struct pt_regs *regs) unsigned short cwd, swd; conditional_sti(regs); - if ((regs->cs & 3) == 0 && + if (!user_mode(regs) && kernel_math_error(regs, "kernel x87 math error")) return; @@ -822,7 +828,7 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs) unsigned short mxcsr; conditional_sti(regs); - if ((regs->cs & 3) == 0 && + if (!user_mode(regs) && kernel_math_error(regs, "kernel simd math error")) return; diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index 59ebd5beda8..73389f51c4e 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -2,7 +2,10 @@ * Written by Martin Mares <mj@atrey.karlin.mff.cuni.cz>; */ +#define LOAD_OFFSET __START_KERNEL_map + #include <asm-generic/vmlinux.lds.h> +#include <asm/page.h> #include <linux/config.h> OUTPUT_FORMAT("elf64-x86-64", "elf64-x86-64", "elf64-x86-64") @@ -11,28 +14,30 @@ ENTRY(phys_startup_64) jiffies_64 = jiffies; SECTIONS { - . = 0xffffffff80100000; + . = __START_KERNEL; phys_startup_64 = startup_64 - LOAD_OFFSET; _text = .; /* Text and read-only data */ - .text : { + .text : AT(ADDR(.text) - LOAD_OFFSET) { *(.text) SCHED_TEXT LOCK_TEXT *(.fixup) *(.gnu.warning) } = 0x9090 - .text.lock : { *(.text.lock) } /* out-of-line lock text */ + /* out-of-line lock text */ + .text.lock : AT(ADDR(.text.lock) - LOAD_OFFSET) { *(.text.lock) } _etext = .; /* End of text section */ . = ALIGN(16); /* Exception table */ __start___ex_table = .; - __ex_table : { *(__ex_table) } + __ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) { *(__ex_table) } __stop___ex_table = .; RODATA - .data : { /* Data */ + /* Data */ + .data : AT(ADDR(.data) - LOAD_OFFSET) { *(.data) CONSTRUCTORS } @@ -40,62 +45,95 @@ SECTIONS _edata = .; /* End of data section */ __bss_start = .; /* BSS */ - .bss : { + .bss : AT(ADDR(.bss) - LOAD_OFFSET) { *(.bss.page_aligned) *(.bss) } __bss_end = .; + . = ALIGN(PAGE_SIZE); . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); - .data.cacheline_aligned : { *(.data.cacheline_aligned) } + .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) { + *(.data.cacheline_aligned) + } + +#define VSYSCALL_ADDR (-10*1024*1024) +#define VSYSCALL_PHYS_ADDR ((LOADADDR(.data.cacheline_aligned) + SIZEOF(.data.cacheline_aligned) + 4095) & ~(4095)) +#define VSYSCALL_VIRT_ADDR ((ADDR(.data.cacheline_aligned) + SIZEOF(.data.cacheline_aligned) + 4095) & ~(4095)) + +#define VLOAD_OFFSET (VSYSCALL_ADDR - VSYSCALL_PHYS_ADDR) +#define VLOAD(x) (ADDR(x) - VLOAD_OFFSET) + +#define VVIRT_OFFSET (VSYSCALL_ADDR - VSYSCALL_VIRT_ADDR) +#define VVIRT(x) (ADDR(x) - VVIRT_OFFSET) -#define AFTER(x) BINALIGN(LOADADDR(x) + SIZEOF(x), 16) -#define BINALIGN(x,y) (((x) + (y) - 1) & ~((y) - 1)) -#define CACHE_ALIGN(x) BINALIGN(x, CONFIG_X86_L1_CACHE_BYTES) + . = VSYSCALL_ADDR; + .vsyscall_0 : AT(VSYSCALL_PHYS_ADDR) { *(.vsyscall_0) } + __vsyscall_0 = VSYSCALL_VIRT_ADDR; - .vsyscall_0 -10*1024*1024: AT ((LOADADDR(.data.cacheline_aligned) + SIZEOF(.data.cacheline_aligned) + 4095) & ~(4095)) { *(.vsyscall_0) } - __vsyscall_0 = LOADADDR(.vsyscall_0); . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); - .xtime_lock : AT CACHE_ALIGN(AFTER(.vsyscall_0)) { *(.xtime_lock) } - xtime_lock = LOADADDR(.xtime_lock); - .vxtime : AT AFTER(.xtime_lock) { *(.vxtime) } - vxtime = LOADADDR(.vxtime); - .wall_jiffies : AT AFTER(.vxtime) { *(.wall_jiffies) } - wall_jiffies = LOADADDR(.wall_jiffies); - .sys_tz : AT AFTER(.wall_jiffies) { *(.sys_tz) } - sys_tz = LOADADDR(.sys_tz); - .sysctl_vsyscall : AT AFTER(.sys_tz) { *(.sysctl_vsyscall) } - sysctl_vsyscall = LOADADDR(.sysctl_vsyscall); - .xtime : AT AFTER(.sysctl_vsyscall) { *(.xtime) } - xtime = LOADADDR(.xtime); + .xtime_lock : AT(VLOAD(.xtime_lock)) { *(.xtime_lock) } + xtime_lock = VVIRT(.xtime_lock); + + .vxtime : AT(VLOAD(.vxtime)) { *(.vxtime) } + vxtime = VVIRT(.vxtime); + + .wall_jiffies : AT(VLOAD(.wall_jiffies)) { *(.wall_jiffies) } + wall_jiffies = VVIRT(.wall_jiffies); + + .sys_tz : AT(VLOAD(.sys_tz)) { *(.sys_tz) } + sys_tz = VVIRT(.sys_tz); + + .sysctl_vsyscall : AT(VLOAD(.sysctl_vsyscall)) { *(.sysctl_vsyscall) } + sysctl_vsyscall = VVIRT(.sysctl_vsyscall); + + .xtime : AT(VLOAD(.xtime)) { *(.xtime) } + xtime = VVIRT(.xtime); + . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); - .jiffies : AT CACHE_ALIGN(AFTER(.xtime)) { *(.jiffies) } - jiffies = LOADADDR(.jiffies); - .vsyscall_1 ADDR(.vsyscall_0) + 1024: AT (LOADADDR(.vsyscall_0) + 1024) { *(.vsyscall_1) } - . = LOADADDR(.vsyscall_0) + 4096; + .jiffies : AT(VLOAD(.jiffies)) { *(.jiffies) } + jiffies = VVIRT(.jiffies); + + .vsyscall_1 ADDR(.vsyscall_0) + 1024: AT(VLOAD(.vsyscall_1)) { *(.vsyscall_1) } + .vsyscall_2 ADDR(.vsyscall_0) + 2048: AT(VLOAD(.vsyscall_2)) { *(.vsyscall_2) } + .vsyscall_3 ADDR(.vsyscall_0) + 3072: AT(VLOAD(.vsyscall_3)) { *(.vsyscall_3) } + + . = VSYSCALL_VIRT_ADDR + 4096; + +#undef VSYSCALL_ADDR +#undef VSYSCALL_PHYS_ADDR +#undef VSYSCALL_VIRT_ADDR +#undef VLOAD_OFFSET +#undef VLOAD +#undef VVIRT_OFFSET +#undef VVIRT . = ALIGN(8192); /* init_task */ - .data.init_task : { *(.data.init_task) } + .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) { + *(.data.init_task) + } . = ALIGN(4096); - .data.page_aligned : { *(.data.page_aligned) } + .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) { + *(.data.page_aligned) + } . = ALIGN(4096); /* Init code and data */ __init_begin = .; - .init.text : { + .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) { _sinittext = .; *(.init.text) _einittext = .; } __initdata_begin = .; - .init.data : { *(.init.data) } + .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { *(.init.data) } __initdata_end = .; . = ALIGN(16); __setup_start = .; - .init.setup : { *(.init.setup) } + .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { *(.init.setup) } __setup_end = .; __initcall_start = .; - .initcall.init : { + .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { *(.initcall1.init) *(.initcall2.init) *(.initcall3.init) @@ -106,32 +144,38 @@ SECTIONS } __initcall_end = .; __con_initcall_start = .; - .con_initcall.init : { *(.con_initcall.init) } + .con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) { + *(.con_initcall.init) + } __con_initcall_end = .; SECURITY_INIT . = ALIGN(8); __alt_instructions = .; - .altinstructions : { *(.altinstructions) } + .altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) { + *(.altinstructions) + } __alt_instructions_end = .; - .altinstr_replacement : { *(.altinstr_replacement) } + .altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) { + *(.altinstr_replacement) + } /* .exit.text is discard at runtime, not link time, to deal with references from .altinstructions and .eh_frame */ - .exit.text : { *(.exit.text) } - .exit.data : { *(.exit.data) } + .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { *(.exit.text) } + .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { *(.exit.data) } . = ALIGN(4096); __initramfs_start = .; - .init.ramfs : { *(.init.ramfs) } + .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) } __initramfs_end = .; . = ALIGN(32); __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } + .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) } __per_cpu_end = .; . = ALIGN(4096); __init_end = .; . = ALIGN(4096); __nosave_begin = .; - .data_nosave : { *(.data.nosave) } + .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { *(.data.nosave) } . = ALIGN(4096); __nosave_end = .; diff --git a/arch/x86_64/lib/delay.c b/arch/x86_64/lib/delay.c index 6e2d66472eb..33a873a3c22 100644 --- a/arch/x86_64/lib/delay.c +++ b/arch/x86_64/lib/delay.c @@ -12,6 +12,7 @@ #include <linux/sched.h> #include <linux/delay.h> #include <asm/delay.h> +#include <asm/msr.h> #ifdef CONFIG_SMP #include <asm/smp.h> @@ -19,6 +20,12 @@ int x86_udelay_tsc = 0; /* Delay via TSC */ +int read_current_timer(unsigned long *timer_value) +{ + rdtscll(*timer_value); + return 0; +} + void __delay(unsigned long loops) { unsigned bclock, now; @@ -34,7 +41,7 @@ void __delay(unsigned long loops) inline void __const_udelay(unsigned long xloops) { - __delay(((xloops * cpu_data[_smp_processor_id()].loops_per_jiffy) >> 32) * HZ); + __delay(((xloops * cpu_data[raw_smp_processor_id()].loops_per_jiffy) >> 32) * HZ); } void __udelay(unsigned long usecs) diff --git a/arch/x86_64/mm/Makefile b/arch/x86_64/mm/Makefile index 66c354ad80c..1d232a87f11 100644 --- a/arch/x86_64/mm/Makefile +++ b/arch/x86_64/mm/Makefile @@ -4,7 +4,7 @@ obj-y := init.o fault.o ioremap.o extable.o pageattr.o obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o -obj-$(CONFIG_DISCONTIGMEM) += numa.o +obj-$(CONFIG_NUMA) += numa.o obj-$(CONFIG_K8_NUMA) += k8topology.o obj-$(CONFIG_ACPI_NUMA) += srat.o diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c index 5d6b2114cc9..2f187986f94 100644 --- a/arch/x86_64/mm/fault.c +++ b/arch/x86_64/mm/fault.c @@ -74,7 +74,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr, instr = (unsigned char *)convert_rip_to_linear(current, regs); max_instr = instr + 15; - if ((regs->cs & 3) != 0 && instr >= (unsigned char *)TASK_SIZE) + if (user_mode(regs) && instr >= (unsigned char *)TASK_SIZE) return 0; while (scan_more && instr < max_instr) { @@ -106,7 +106,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr, /* Could check the LDT for lm, but for now it's good enough to assume that long mode only uses well known segments or kernel. */ - scan_more = ((regs->cs & 3) == 0) || (regs->cs == __USER_CS); + scan_more = (!user_mode(regs)) || (regs->cs == __USER_CS); break; case 0x60: @@ -350,7 +350,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code) * (error_code & 4) == 0, and that the fault was not a * protection error (error_code & 1) == 0. */ - if (unlikely(address >= TASK_SIZE)) { + if (unlikely(address >= TASK_SIZE64)) { if (!(error_code & 5) && ((address >= VMALLOC_START && address < VMALLOC_END) || (address >= MODULES_VADDR && address < MODULES_END))) { diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index dbe53b4c7e6..72e4b364ed7 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -318,7 +318,7 @@ void zap_low_mappings(void) flush_tlb_all(); } -#ifndef CONFIG_DISCONTIGMEM +#ifndef CONFIG_NUMA void __init paging_init(void) { { @@ -427,13 +427,16 @@ void __init mem_init(void) reservedpages = 0; /* this will put all low memory onto the freelists */ -#ifdef CONFIG_DISCONTIGMEM +#ifdef CONFIG_NUMA totalram_pages += numa_free_all_bootmem(); tmp = 0; /* should count reserved pages here for all nodes */ #else + +#ifdef CONFIG_FLATMEM max_mapnr = end_pfn; if (!mem_map) BUG(); +#endif totalram_pages += free_all_bootmem(); @@ -515,7 +518,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) void __init reserve_bootmem_generic(unsigned long phys, unsigned len) { /* Should check here against the e820 map to avoid double free */ -#ifdef CONFIG_DISCONTIGMEM +#ifdef CONFIG_NUMA int nid = phys_to_nid(phys); reserve_bootmem_node(NODE_DATA(nid), phys, len); #else diff --git a/arch/x86_64/mm/ioremap.c b/arch/x86_64/mm/ioremap.c index 58aac23760e..6972df480d2 100644 --- a/arch/x86_64/mm/ioremap.c +++ b/arch/x86_64/mm/ioremap.c @@ -178,7 +178,7 @@ void __iomem * __ioremap(unsigned long phys_addr, unsigned long size, unsigned l if (phys_addr >= ISA_START_ADDRESS && last_addr < ISA_END_ADDRESS) return (__force void __iomem *)phys_to_virt(phys_addr); -#ifndef CONFIG_DISCONTIGMEM +#ifdef CONFIG_FLATMEM /* * Don't allow anybody to remap normal RAM that we're using.. */ diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c index fd9f25d7a6c..ac61c186eb0 100644 --- a/arch/x86_64/mm/numa.c +++ b/arch/x86_64/mm/numa.c @@ -66,6 +66,13 @@ int __init compute_hash_shift(struct node *nodes, int numnodes) return -1; } +#ifdef CONFIG_SPARSEMEM +int early_pfn_to_nid(unsigned long pfn) +{ + return phys_to_nid(pfn << PAGE_SHIFT); +} +#endif + /* Initialize bootmem allocator for a node */ void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long end) { @@ -80,6 +87,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long en start_pfn = start >> PAGE_SHIFT; end_pfn = end >> PAGE_SHIFT; + memory_present(nodeid, start_pfn, end_pfn); nodedata_phys = find_e820_area(start, end, pgdat_size); if (nodedata_phys == -1L) panic("Cannot find memory pgdat in node %d\n", nodeid); @@ -243,7 +251,7 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn) setup_node_bootmem(0, start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT); } -__init void numa_add_cpu(int cpu) +__cpuinit void numa_add_cpu(int cpu) { /* BP is initialized elsewhere */ if (cpu) diff --git a/arch/x86_64/pci/k8-bus.c b/arch/x86_64/pci/k8-bus.c index 62349c78db5..7e7d0c2a002 100644 --- a/arch/x86_64/pci/k8-bus.c +++ b/arch/x86_64/pci/k8-bus.c @@ -53,25 +53,11 @@ fill_mp_bus_to_cpumask(void) for (j = SECONDARY_LDT_BUS_NUMBER(ldtbus); j <= SUBORDINATE_LDT_BUS_NUMBER(ldtbus); j++) - pci_bus_to_cpumask[j] = - node_to_cpumask(NODE_ID(nid)); + pci_bus_to_node[j] = NODE_ID(nid); } } } - /* quick sanity check */ - printed = 0; - for (i = 0; i < 256; i++) { - if (cpus_empty(pci_bus_to_cpumask[i])) { - pci_bus_to_cpumask[i] = CPU_MASK_ALL; - if (printed) - continue; - printk(KERN_ERR - "k8-bus.c: some busses have empty cpu mask\n"); - printed = 1; - } - } - return 0; } |