diff options
Diffstat (limited to 'arch')
175 files changed, 4078 insertions, 2298 deletions
diff --git a/arch/arm/include/asm/cnt32_to_63.h b/arch/arm/include/asm/cnt32_to_63.h deleted file mode 100644 index 480c873..0000000 --- a/arch/arm/include/asm/cnt32_to_63.h +++ /dev/null @@ -1,78 +0,0 @@ -/* - * include/asm/cnt32_to_63.h -- extend a 32-bit counter to 63 bits - * - * Author: Nicolas Pitre - * Created: December 3, 2006 - * Copyright: MontaVista Software, Inc. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 - * as published by the Free Software Foundation. - */ - -#ifndef __INCLUDE_CNT32_TO_63_H__ -#define __INCLUDE_CNT32_TO_63_H__ - -#include <linux/compiler.h> -#include <asm/types.h> -#include <asm/byteorder.h> - -/* - * Prototype: u64 cnt32_to_63(u32 cnt) - * Many hardware clock counters are only 32 bits wide and therefore have - * a relatively short period making wrap-arounds rather frequent. This - * is a problem when implementing sched_clock() for example, where a 64-bit - * non-wrapping monotonic value is expected to be returned. - * - * To overcome that limitation, let's extend a 32-bit counter to 63 bits - * in a completely lock free fashion. Bits 0 to 31 of the clock are provided - * by the hardware while bits 32 to 62 are stored in memory. The top bit in - * memory is used to synchronize with the hardware clock half-period. When - * the top bit of both counters (hardware and in memory) differ then the - * memory is updated with a new value, incrementing it when the hardware - * counter wraps around. - * - * Because a word store in memory is atomic then the incremented value will - * always be in synch with the top bit indicating to any potential concurrent - * reader if the value in memory is up to date or not with regards to the - * needed increment. And any race in updating the value in memory is harmless - * as the same value would simply be stored more than once. - * - * The only restriction for the algorithm to work properly is that this - * code must be executed at least once per each half period of the 32-bit - * counter to properly update the state bit in memory. This is usually not a - * problem in practice, but if it is then a kernel timer could be scheduled - * to manage for this code to be executed often enough. - * - * Note that the top bit (bit 63) in the returned value should be considered - * as garbage. It is not cleared here because callers are likely to use a - * multiplier on the returned value which can get rid of the top bit - * implicitly by making the multiplier even, therefore saving on a runtime - * clear-bit instruction. Otherwise caller must remember to clear the top - * bit explicitly. - */ - -/* this is used only to give gcc a clue about good code generation */ -typedef union { - struct { -#if defined(__LITTLE_ENDIAN) - u32 lo, hi; -#elif defined(__BIG_ENDIAN) - u32 hi, lo; -#endif - }; - u64 val; -} cnt32_to_63_t; - -#define cnt32_to_63(cnt_lo) \ -({ \ - static volatile u32 __m_cnt_hi = 0; \ - cnt32_to_63_t __x; \ - __x.hi = __m_cnt_hi; \ - __x.lo = (cnt_lo); \ - if (unlikely((s32)(__x.hi ^ __x.lo) < 0)) \ - __m_cnt_hi = __x.hi = (__x.hi ^ 0x80000000) + (__x.hi >> 31); \ - __x.val; \ -}) - -#endif diff --git a/arch/arm/kernel/kgdb.c b/arch/arm/kernel/kgdb.c index aaffaec..ba8ccfe 100644 --- a/arch/arm/kernel/kgdb.c +++ b/arch/arm/kernel/kgdb.c @@ -111,8 +111,6 @@ int kgdb_arch_handle_exception(int exception_vector, int signo, case 'D': case 'k': case 'c': - kgdb_contthread = NULL; - /* * Try to read optional parameter, pc unchanged if no parm. * If this was a compiled breakpoint, we need to move diff --git a/arch/arm/mach-davinci/psc.c b/arch/arm/mach-davinci/psc.c index 720c48b..aa2fc37 100644 --- a/arch/arm/mach-davinci/psc.c +++ b/arch/arm/mach-davinci/psc.c @@ -70,9 +70,6 @@ void davinci_psc_config(unsigned int domain, unsigned int id, char enable) { u32 epcpr, ptcmd, ptstat, pdstat, pdctl1, mdstat, mdctl, mdstat_mask; - if (id < 0) - return; - mdctl = davinci_readl(DAVINCI_PWR_SLEEP_CNTRL_BASE + MDCTL + 4 * id); if (enable) mdctl |= 0x00000003; /* Enable Module */ diff --git a/arch/arm/mach-pxa/time.c b/arch/arm/mach-pxa/time.c index 67e1850..b0d6b32 100644 --- a/arch/arm/mach-pxa/time.c +++ b/arch/arm/mach-pxa/time.c @@ -17,9 +17,9 @@ #include <linux/interrupt.h> #include <linux/clockchips.h> #include <linux/sched.h> +#include <linux/cnt32_to_63.h> #include <asm/div64.h> -#include <asm/cnt32_to_63.h> #include <asm/mach/irq.h> #include <asm/mach/time.h> #include <mach/pxa-regs.h> diff --git a/arch/arm/mach-sa1100/generic.c b/arch/arm/mach-sa1100/generic.c index 1362994..b422526 100644 --- a/arch/arm/mach-sa1100/generic.c +++ b/arch/arm/mach-sa1100/generic.c @@ -18,9 +18,9 @@ #include <linux/ioport.h> #include <linux/sched.h> /* just for sched_clock() - funny that */ #include <linux/platform_device.h> +#include <linux/cnt32_to_63.h> #include <asm/div64.h> -#include <asm/cnt32_to_63.h> #include <mach/hardware.h> #include <asm/system.h> #include <asm/pgtable.h> diff --git a/arch/arm/mach-sa1100/include/mach/jornada720.h b/arch/arm/mach-sa1100/include/mach/jornada720.h index bc12085..cc6b4bf 100644 --- a/arch/arm/mach-sa1100/include/mach/jornada720.h +++ b/arch/arm/mach-sa1100/include/mach/jornada720.h @@ -1,10 +1,10 @@ /* * arch/arm/mach-sa1100/include/mach/jornada720.h * - * This file contains SSP/MCU communication definitions for HP Jornada 710/720/728 + * SSP/MCU communication definitions for HP Jornada 710/720/728 * - * Copyright (C) 2007 Kristoffer Ericson <Kristoffer.Ericson@gmail.com> - * Copyright (C) 2000 John Ankcorn <jca@lcs.mit.edu> + * Copyright 2007,2008 Kristoffer Ericson <Kristoffer.Ericson@gmail.com> + * Copyright 2000 John Ankcorn <jca@lcs.mit.edu> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -25,3 +25,8 @@ #define PWMOFF 0xDF #define TXDUMMY 0x11 #define ERRORCODE 0x00 + +extern void jornada_ssp_start(void); +extern void jornada_ssp_end(void); +extern int jornada_ssp_inout(u8 byte); +extern int jornada_ssp_byte(u8 byte); diff --git a/arch/arm/mach-sa1100/jornada720_ssp.c b/arch/arm/mach-sa1100/jornada720_ssp.c index 06ea7ab..28cf369 100644 --- a/arch/arm/mach-sa1100/jornada720_ssp.c +++ b/arch/arm/mach-sa1100/jornada720_ssp.c @@ -21,8 +21,8 @@ #include <linux/slab.h> #include <mach/hardware.h> -#include <asm/hardware/ssp.h> #include <mach/jornada720.h> +#include <asm/hardware/ssp.h> static DEFINE_SPINLOCK(jornada_ssp_lock); static unsigned long jornada_ssp_flags; @@ -109,12 +109,12 @@ EXPORT_SYMBOL(jornada_ssp_inout); * jornada_ssp_start - enable mcu * */ -int jornada_ssp_start() +void jornada_ssp_start(void) { spin_lock_irqsave(&jornada_ssp_lock, jornada_ssp_flags); GPCR = GPIO_GPIO25; udelay(50); - return 0; + return; }; EXPORT_SYMBOL(jornada_ssp_start); @@ -122,11 +122,11 @@ EXPORT_SYMBOL(jornada_ssp_start); * jornada_ssp_end - disable mcu and turn off lock * */ -int jornada_ssp_end() +void jornada_ssp_end(void) { GPSR = GPIO_GPIO25; spin_unlock_irqrestore(&jornada_ssp_lock, jornada_ssp_flags); - return 0; + return; }; EXPORT_SYMBOL(jornada_ssp_end); diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c index d75e795..b638f10 100644 --- a/arch/arm/mach-versatile/core.c +++ b/arch/arm/mach-versatile/core.c @@ -28,8 +28,8 @@ #include <linux/amba/clcd.h> #include <linux/clocksource.h> #include <linux/clockchips.h> +#include <linux/cnt32_to_63.h> -#include <asm/cnt32_to_63.h> #include <asm/system.h> #include <mach/hardware.h> #include <asm/io.h> diff --git a/arch/arm/plat-omap/devices.c b/arch/arm/plat-omap/devices.c index bc1cf30..01da719 100644 --- a/arch/arm/plat-omap/devices.c +++ b/arch/arm/plat-omap/devices.c @@ -316,19 +316,6 @@ static inline void omap_init_mmc_conf(const struct omap_mmc_config *mmc_conf) omap_cfg_reg(MMC_DAT3); } } -#if defined(CONFIG_ARCH_OMAP2420) - if (mmc_conf->mmc[0].internal_clock) { - /* - * Use internal loop-back in MMC/SDIO - * Module Input Clock selection - */ - if (cpu_is_omap24xx()) { - u32 v = omap_ctrl_readl(OMAP2_CONTROL_DEVCONF0); - v |= (1 << 24); /* not used in 243x */ - omap_ctrl_writel(v, OMAP2_CONTROL_DEVCONF0); - } - } -#endif } #ifdef CONFIG_ARCH_OMAP16XX diff --git a/arch/avr32/boards/atstk1000/atstk1002.c b/arch/avr32/boards/atstk1000/atstk1002.c index ee4c292..dfc3443 100644 --- a/arch/avr32/boards/atstk1000/atstk1002.c +++ b/arch/avr32/boards/atstk1000/atstk1002.c @@ -325,7 +325,7 @@ static int __init atstk1002_init(void) #ifdef CONFIG_BOARD_ATSTK100X_SPI1 at32_add_device_spi(1, spi1_board_info, ARRAY_SIZE(spi1_board_info)); #endif -#ifndef CONFIG_BOARD_ATSTK1002_SW2_CUSTOM +#ifndef CONFIG_BOARD_ATSTK100X_SW2_CUSTOM at32_add_device_mci(0, MCI_PDATA); #endif #ifdef CONFIG_BOARD_ATSTK1002_SW5_CUSTOM diff --git a/arch/avr32/boot/images/.gitignore b/arch/avr32/boot/images/.gitignore new file mode 100644 index 0000000..64ea9d0 --- /dev/null +++ b/arch/avr32/boot/images/.gitignore @@ -0,0 +1,4 @@ +uImage +uImage.srec +vmlinux.cso +sfdwarf.log diff --git a/arch/avr32/kernel/.gitignore b/arch/avr32/kernel/.gitignore new file mode 100644 index 0000000..c5f676c --- /dev/null +++ b/arch/avr32/kernel/.gitignore @@ -0,0 +1 @@ +vmlinux.lds diff --git a/arch/avr32/kernel/avr32_ksyms.c b/arch/avr32/kernel/avr32_ksyms.c index 84a7d44..11e310c 100644 --- a/arch/avr32/kernel/avr32_ksyms.c +++ b/arch/avr32/kernel/avr32_ksyms.c @@ -58,6 +58,7 @@ EXPORT_SYMBOL(find_first_zero_bit); EXPORT_SYMBOL(find_next_zero_bit); EXPORT_SYMBOL(find_first_bit); EXPORT_SYMBOL(find_next_bit); +EXPORT_SYMBOL(generic_find_next_le_bit); EXPORT_SYMBOL(generic_find_next_zero_le_bit); /* I/O primitives (lib/io-*.S) */ diff --git a/arch/avr32/kernel/syscall-stubs.S b/arch/avr32/kernel/syscall-stubs.S index 890286a..673178e 100644 --- a/arch/avr32/kernel/syscall-stubs.S +++ b/arch/avr32/kernel/syscall-stubs.S @@ -109,3 +109,12 @@ __sys_epoll_pwait: rcall sys_epoll_pwait sub sp, -4 popm pc + + .global __sys_sync_file_range + .type __sys_sync_file_range,@function +__sys_sync_file_range: + pushm lr + st.w --sp, ARG6 + rcall sys_sync_file_range + sub sp, -4 + popm pc diff --git a/arch/avr32/kernel/syscall_table.S b/arch/avr32/kernel/syscall_table.S index 478bda4..7ee0057 100644 --- a/arch/avr32/kernel/syscall_table.S +++ b/arch/avr32/kernel/syscall_table.S @@ -275,7 +275,7 @@ sys_call_table: .long sys_set_robust_list .long sys_get_robust_list /* 260 */ .long __sys_splice - .long sys_sync_file_range + .long __sys_sync_file_range .long sys_tee .long sys_vmsplice .long __sys_epoll_pwait /* 265 */ diff --git a/arch/avr32/kernel/traps.c b/arch/avr32/kernel/traps.c index b835c4c..0d98737 100644 --- a/arch/avr32/kernel/traps.c +++ b/arch/avr32/kernel/traps.c @@ -116,15 +116,15 @@ asmlinkage void do_nmi(unsigned long ecr, struct pt_regs *regs) switch (ret) { case NOTIFY_OK: case NOTIFY_STOP: - return; + break; case NOTIFY_BAD: die("Fatal Non-Maskable Interrupt", regs, SIGINT); default: + printk(KERN_ALERT "Got NMI, but nobody cared. Disabling...\n"); + nmi_disable(); break; } - - printk(KERN_ALERT "Got NMI, but nobody cared. Disabling...\n"); - nmi_disable(); + nmi_exit(); } asmlinkage void do_critical_exception(unsigned long ecr, struct pt_regs *regs) diff --git a/arch/avr32/lib/findbit.S b/arch/avr32/lib/findbit.S index c6b91de..997b33b 100644 --- a/arch/avr32/lib/findbit.S +++ b/arch/avr32/lib/findbit.S @@ -123,6 +123,36 @@ ENTRY(find_next_bit) brgt 1b retal r11 +ENTRY(generic_find_next_le_bit) + lsr r8, r10, 5 + sub r9, r11, r10 + retle r11 + + lsl r8, 2 + add r12, r8 + andl r10, 31, COH + breq 1f + + /* offset is not word-aligned. Handle the first (32 - r10) bits */ + ldswp.w r8, r12[0] + sub r12, -4 + lsr r8, r8, r10 + brne .L_found + + /* r9 = r9 - (32 - r10) = r9 + r10 - 32 */ + add r9, r10 + sub r9, 32 + retle r11 + + /* Main loop. offset must be word-aligned */ +1: ldswp.w r8, r12[0] + cp.w r8, 0 + brne .L_found + sub r12, -4 + sub r9, 32 + brgt 1b + retal r11 + ENTRY(generic_find_next_zero_le_bit) lsr r8, r10, 5 sub r9, r11, r10 diff --git a/arch/ia64/include/asm/sections.h b/arch/ia64/include/asm/sections.h index f667998..1a873b3 100644 --- a/arch/ia64/include/asm/sections.h +++ b/arch/ia64/include/asm/sections.h @@ -11,6 +11,9 @@ #include <asm-generic/sections.h> extern char __per_cpu_start[], __per_cpu_end[], __phys_per_cpu_start[]; +#ifdef CONFIG_SMP +extern char __cpu0_per_cpu[]; +#endif extern char __start___vtop_patchlist[], __end___vtop_patchlist[]; extern char __start___rse_patchlist[], __end___rse_patchlist[]; extern char __start___mckinley_e9_bundles[], __end___mckinley_e9_bundles[]; diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c index d45f215..51b75ce 100644 --- a/arch/ia64/kernel/efi.c +++ b/arch/ia64/kernel/efi.c @@ -1232,9 +1232,10 @@ efi_initialize_iomem_resources(struct resource *code_resource, if (md->attribute & EFI_MEMORY_WP) { name = "System ROM"; flags |= IORESOURCE_READONLY; - } else { + } else if (md->attribute == EFI_MEMORY_UC) + name = "Uncached RAM"; + else name = "System RAM"; - } break; case EFI_ACPI_MEMORY_NVS: diff --git a/arch/ia64/kernel/head.S b/arch/ia64/kernel/head.S index 8bdea8e..66e491d 100644 --- a/arch/ia64/kernel/head.S +++ b/arch/ia64/kernel/head.S @@ -367,16 +367,17 @@ start_ap: ;; #else (isAP) br.few 2f - mov r20=r19 - sub r19=r19,r18 + movl r20=__cpu0_per_cpu ;; shr.u r18=r18,3 1: - ld8 r21=[r20],8;; - st8[r19]=r21,8 + ld8 r21=[r19],8;; + st8[r20]=r21,8 adds r18=-1,r18;; cmp4.lt p7,p6=0,r18 (p7) br.cond.dptk.few 1b + mov r19=r20 + ;; 2: #endif tpa r19=r19 diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index c27d5b2..de636b2 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -616,7 +616,9 @@ setup_arch (char **cmdline_p) ia64_mca_init(); platform_setup(cmdline_p); +#ifndef CONFIG_IA64_HP_SIM check_sal_cache_flush(); +#endif paging_init(); } diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S index de71da8..10a7d47e 100644 --- a/arch/ia64/kernel/vmlinux.lds.S +++ b/arch/ia64/kernel/vmlinux.lds.S @@ -215,9 +215,6 @@ SECTIONS /* Per-cpu data: */ percpu : { } :percpu . = ALIGN(PERCPU_PAGE_SIZE); -#ifdef CONFIG_SMP - . = . + PERCPU_PAGE_SIZE; /* cpu0 per-cpu space */ -#endif __phys_per_cpu_start = .; .data.percpu PERCPU_ADDR : AT(__phys_per_cpu_start - LOAD_OFFSET) { @@ -233,6 +230,11 @@ SECTIONS data : { } :data .data : AT(ADDR(.data) - LOAD_OFFSET) { +#ifdef CONFIG_SMP + . = ALIGN(PERCPU_PAGE_SIZE); + __cpu0_per_cpu = .; + . = . + PERCPU_PAGE_SIZE; /* cpu0 per-cpu space */ +#endif DATA_DATA *(.data1) *(.gnu.linkonce.d*) diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 7a37d06..cd0d1a7 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -38,6 +38,7 @@ #include <asm/cacheflush.h> #include <asm/div64.h> #include <asm/tlb.h> +#include <asm/elf.h> #include "misc.h" #include "vti.h" @@ -61,12 +62,6 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { NULL } }; - -struct fdesc{ - unsigned long ip; - unsigned long gp; -}; - static void kvm_flush_icache(unsigned long start, unsigned long len) { int l; diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c index e566ff4..0ee085e 100644 --- a/arch/ia64/mm/contig.c +++ b/arch/ia64/mm/contig.c @@ -163,7 +163,7 @@ per_cpu_init (void) * get_zeroed_page(). */ if (first_time) { - void *cpu0_data = __phys_per_cpu_start - PERCPU_PAGE_SIZE; + void *cpu0_data = __cpu0_per_cpu; first_time=0; diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 78026aa..d8c5fcd 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -144,7 +144,7 @@ static void *per_cpu_node_setup(void *cpu_data, int node) for_each_possible_early_cpu(cpu) { if (cpu == 0) { - void *cpu0_data = __phys_per_cpu_start - PERCPU_PAGE_SIZE; + void *cpu0_data = __cpu0_per_cpu; __per_cpu_offset[cpu] = (char*)cpu0_data - __per_cpu_start; } else if (node == node_cpuid[cpu].nid) { diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig index a5f864c..f57113f 100644 --- a/arch/m32r/Kconfig +++ b/arch/m32r/Kconfig @@ -216,10 +216,6 @@ config MEMORY_SIZE default "01000000" if PLAT_M32104UT default "00800000" if PLAT_OAKS32R -config NOHIGHMEM - bool - default y - config ARCH_DISCONTIGMEM_ENABLE bool "Internal RAM Support" depends on CHIP_M32700 || CHIP_M32102 || CHIP_VDEC2 || CHIP_OPSP || CHIP_M32104 @@ -410,11 +406,7 @@ config PCI_DIRECT source "drivers/pci/Kconfig" config ISA - bool "ISA support" - help - Find out whether you have ISA slots on your motherboard. ISA is the - name of a bus system, i.e. the way the CPU talks to the other stuff - inside your box. If you have ISA, say Y, otherwise N. + bool source "drivers/pcmcia/Kconfig" diff --git a/arch/m32r/kernel/entry.S b/arch/m32r/kernel/entry.S index d4eaa2f..612d35b 100644 --- a/arch/m32r/kernel/entry.S +++ b/arch/m32r/kernel/entry.S @@ -143,7 +143,7 @@ ret_from_intr: and3 r4, r4, #0x8000 ; check BSM bit #endif beqz r4, resume_kernel -ENTRY(resume_userspace) +resume_userspace: DISABLE_INTERRUPTS(r4) ; make sure we don't miss an interrupt ; setting need_resched or sigpending ; between sampling and the iret diff --git a/arch/m32r/kernel/head.S b/arch/m32r/kernel/head.S index dab7436..4018077 100644 --- a/arch/m32r/kernel/head.S +++ b/arch/m32r/kernel/head.S @@ -29,7 +29,6 @@ __INITDATA .global _end ENTRY(stext) ENTRY(_stext) -ENTRY(startup_32) /* Setup up the stack pointer */ LDIMM (r0, spi_stack_top) LDIMM (r1, spu_stack_top) diff --git a/arch/m32r/kernel/irq.c b/arch/m32r/kernel/irq.c index d0c5b0b..2aeae46 100644 --- a/arch/m32r/kernel/irq.c +++ b/arch/m32r/kernel/irq.c @@ -22,9 +22,6 @@ #include <linux/module.h> #include <asm/uaccess.h> -atomic_t irq_err_count; -atomic_t irq_mis_count; - /* * Generic, controller-independent functions: */ @@ -63,9 +60,6 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: spin_unlock_irqrestore(&irq_desc[i].lock, flags); - } else if (i == NR_IRQS) { - seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); - seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count)); } return 0; } diff --git a/arch/m32r/kernel/m32r_ksyms.c b/arch/m32r/kernel/m32r_ksyms.c index 16bcb18..22624b5 100644 --- a/arch/m32r/kernel/m32r_ksyms.c +++ b/arch/m32r/kernel/m32r_ksyms.c @@ -14,6 +14,7 @@ #include <asm/delay.h> #include <asm/irq.h> #include <asm/tlbflush.h> +#include <asm/pgtable.h> /* platform dependent support */ EXPORT_SYMBOL(boot_cpu_data); @@ -65,6 +66,7 @@ EXPORT_SYMBOL(memset); EXPORT_SYMBOL(copy_page); EXPORT_SYMBOL(clear_page); EXPORT_SYMBOL(strlen); +EXPORT_SYMBOL(empty_zero_page); EXPORT_SYMBOL(_inb); EXPORT_SYMBOL(_inw); diff --git a/arch/m32r/kernel/process.c b/arch/m32r/kernel/process.c index a689e29..5be4faa 100644 --- a/arch/m32r/kernel/process.c +++ b/arch/m32r/kernel/process.c @@ -35,8 +35,6 @@ #include <linux/err.h> -static int hlt_counter=0; - /* * Return saved PC of a blocked thread. */ @@ -48,31 +46,16 @@ unsigned long thread_saved_pc(struct task_struct *tsk) /* * Powermanagement idle function, if any.. */ -void (*pm_idle)(void) = NULL; -EXPORT_SYMBOL(pm_idle); +static void (*pm_idle)(void) = NULL; void (*pm_power_off)(void) = NULL; EXPORT_SYMBOL(pm_power_off); -void disable_hlt(void) -{ - hlt_counter++; -} - -EXPORT_SYMBOL(disable_hlt); - -void enable_hlt(void) -{ - hlt_counter--; -} - -EXPORT_SYMBOL(enable_hlt); - /* * We use this is we don't have any better * idle routine.. */ -void default_idle(void) +static void default_idle(void) { /* M32R_FIXME: Please use "cpu_sleep" mode. */ cpu_relax(); @@ -260,15 +243,6 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long spu, return 0; } -/* - * Capture the user space registers if the task is not running (in user space) - */ -int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs) -{ - /* M32R_FIXME */ - return 1; -} - asmlinkage int sys_fork(unsigned long r0, unsigned long r1, unsigned long r2, unsigned long r3, unsigned long r4, unsigned long r5, unsigned long r6, struct pt_regs regs) diff --git a/arch/m32r/kernel/smp.c b/arch/m32r/kernel/smp.c index 7577f97..929e5c9d 100644 --- a/arch/m32r/kernel/smp.c +++ b/arch/m32r/kernel/smp.c @@ -84,7 +84,7 @@ void smp_send_timer(void); void smp_ipi_timer_interrupt(struct pt_regs *); void smp_local_timer_interrupt(void); -void send_IPI_allbutself(int, int); +static void send_IPI_allbutself(int, int); static void send_IPI_mask(cpumask_t, int, int); unsigned long send_IPI_mask_phys(cpumask_t, int, int); @@ -722,7 +722,7 @@ void smp_local_timer_interrupt(void) * ---------- --- -------------------------------------------------------- * *==========================================================================*/ -void send_IPI_allbutself(int ipi_num, int try) +static void send_IPI_allbutself(int ipi_num, int try) { cpumask_t cpumask; diff --git a/arch/m32r/kernel/time.c b/arch/m32r/kernel/time.c index 994cc15..6ea0177 100644 --- a/arch/m32r/kernel/time.c +++ b/arch/m32r/kernel/time.c @@ -34,7 +34,6 @@ #include <asm/hw_irq.h> #ifdef CONFIG_SMP -extern void send_IPI_allbutself(int, int); extern void smp_local_timer_interrupt(void); #endif @@ -188,7 +187,7 @@ static long last_rtc_update = 0; * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick */ -irqreturn_t timer_interrupt(int irq, void *dev_id) +static irqreturn_t timer_interrupt(int irq, void *dev_id) { #ifndef CONFIG_SMP profile_tick(CPU_PROFILING); @@ -228,7 +227,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -struct irqaction irq0 = { +static struct irqaction irq0 = { .handler = timer_interrupt, .flags = IRQF_DISABLED, .mask = CPU_MASK_NONE, diff --git a/arch/m32r/kernel/traps.c b/arch/m32r/kernel/traps.c index 46159a4..03b14e5 100644 --- a/arch/m32r/kernel/traps.c +++ b/arch/m32r/kernel/traps.c @@ -61,7 +61,7 @@ extern unsigned long eit_vector[]; ((unsigned long)func - (unsigned long)eit_vector - entry*4)/4 \ + 0xff000000UL -void set_eit_vector_entries(void) +static void set_eit_vector_entries(void) { extern void default_eit_handler(void); extern void system_call(void); @@ -121,9 +121,9 @@ void __init trap_init(void) cpu_init(); } -int kstack_depth_to_print = 24; +static int kstack_depth_to_print = 24; -void show_trace(struct task_struct *task, unsigned long *stack) +static void show_trace(struct task_struct *task, unsigned long *stack) { unsigned long addr; @@ -224,7 +224,7 @@ bad: printk("\n"); } -DEFINE_SPINLOCK(die_lock); +static DEFINE_SPINLOCK(die_lock); void die(const char * str, struct pt_regs * regs, long err) { diff --git a/arch/m32r/lib/delay.c b/arch/m32r/lib/delay.c index 59bfc34..ced549b 100644 --- a/arch/m32r/lib/delay.c +++ b/arch/m32r/lib/delay.c @@ -6,6 +6,7 @@ */ #include <linux/param.h> +#include <linux/module.h> #ifdef CONFIG_SMP #include <linux/sched.h> #include <asm/current.h> @@ -121,3 +122,4 @@ void __ndelay(unsigned long nsecs) { __const_udelay(nsecs * 0x00005); /* 2**32 / 1000000000 (rounded up) */ } +EXPORT_SYMBOL(__ndelay); diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 49896a2..1e06d23 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -211,6 +211,7 @@ config MIPS_MALTA select SYS_SUPPORTS_64BIT_KERNEL select SYS_SUPPORTS_BIG_ENDIAN select SYS_SUPPORTS_LITTLE_ENDIAN + select SYS_SUPPORTS_MIPS_CMP if BROKEN # because SYNC_R4K is broken select SYS_SUPPORTS_MULTITHREADING select SYS_SUPPORTS_SMARTMIPS help @@ -1403,7 +1404,6 @@ config MIPS_MT_SMTC depends on CPU_MIPS32_R2 #depends on CPU_MIPS64_R2 # once there is hardware ... depends on SYS_SUPPORTS_MULTITHREADING - select GENERIC_CLOCKEVENTS_BROADCAST select CPU_MIPSR2_IRQ_VI select CPU_MIPSR2_IRQ_EI select MIPS_MT @@ -1451,32 +1451,17 @@ config MIPS_VPE_LOADER Includes a loader for loading an elf relocatable object onto another VPE and running it. -config MIPS_MT_SMTC_INSTANT_REPLAY - bool "Low-latency Dispatch of Deferred SMTC IPIs" - depends on MIPS_MT_SMTC && !PREEMPT - default y - help - SMTC pseudo-interrupts between TCs are deferred and queued - if the target TC is interrupt-inhibited (IXMT). In the first - SMTC prototypes, these queued IPIs were serviced on return - to user mode, or on entry into the kernel idle loop. The - INSTANT_REPLAY option dispatches them as part of local_irq_restore() - processing, which adds runtime overhead (hence the option to turn - it off), but ensures that IPIs are handled promptly even under - heavy I/O interrupt load. - config MIPS_MT_SMTC_IM_BACKSTOP bool "Use per-TC register bits as backstop for inhibited IM bits" depends on MIPS_MT_SMTC - default y + default n help To support multiple TC microthreads acting as "CPUs" within a VPE, VPE-wide interrupt mask bits must be specially manipulated during interrupt handling. To support legacy drivers and interrupt controller management code, SMTC has a "backstop" to track and if necessary restore the interrupt mask. This has some performance - impact on interrupt service overhead. Disable it only if you know - what you are doing. + impact on interrupt service overhead. config MIPS_MT_SMTC_IRQAFF bool "Support IRQ affinity API" @@ -1486,10 +1471,8 @@ config MIPS_MT_SMTC_IRQAFF Enables SMP IRQ affinity API (/proc/irq/*/smp_affinity, etc.) for SMTC Linux kernel. Requires platform support, of which an example can be found in the MIPS kernel i8259 and Malta - platform code. It is recommended that MIPS_MT_SMTC_INSTANT_REPLAY - be enabled if MIPS_MT_SMTC_IRQAFF is used. Adds overhead to - interrupt dispatch, and should be used only if you know what - you are doing. + platform code. Adds some overhead to interrupt dispatch, and + should be used only if you know what you are doing. config MIPS_VPE_LOADER_TOM bool "Load VPE program into memory hidden from linux" @@ -1517,6 +1500,18 @@ config MIPS_APSP_KSPD "exit" syscall notifying other kernel modules the SP program is exiting. You probably want to say yes here. +config MIPS_CMP + bool "MIPS CMP framework support" + depends on SYS_SUPPORTS_MIPS_CMP + select SYNC_R4K if BROKEN + select SYS_SUPPORTS_SMP + select SYS_SUPPORTS_SCHED_SMT if SMP + select WEAK_ORDERING + default n + help + This is a placeholder option for the GCMP work. It will need to + be handled differently... + config SB1_PASS_1_WORKAROUNDS bool depends on CPU_SB1_PASS_1 @@ -1693,6 +1688,9 @@ config SMP config SMP_UP bool +config SYS_SUPPORTS_MIPS_CMP + bool + config SYS_SUPPORTS_SMP bool @@ -1740,17 +1738,6 @@ config NR_CPUS performance should round up your number of processors to the next power of two. -config MIPS_CMP - bool "MIPS CMP framework support" - depends on SMP - select SYNC_R4K - select SYS_SUPPORTS_SCHED_SMT - select WEAK_ORDERING - default n - help - This is a placeholder option for the GCMP work. It will need to - be handled differently... - source "kernel/time/Kconfig" # diff --git a/arch/mips/au1000/common/gpio.c b/arch/mips/au1000/common/gpio.c index b485d94..e660ddd 100644 --- a/arch/mips/au1000/common/gpio.c +++ b/arch/mips/au1000/common/gpio.c @@ -48,7 +48,7 @@ static void au1xxx_gpio2_write(unsigned gpio, int value) { gpio -= AU1XXX_GPIO_BASE; - gpio2->output = (GPIO2_OUTPUT_ENABLE_MASK << gpio) | (value << gpio); + gpio2->output = (GPIO2_OUTPUT_ENABLE_MASK << gpio) | ((!!value) << gpio); } static int au1xxx_gpio2_direction_input(unsigned gpio) @@ -61,7 +61,8 @@ static int au1xxx_gpio2_direction_input(unsigned gpio) static int au1xxx_gpio2_direction_output(unsigned gpio, int value) { gpio -= AU1XXX_GPIO_BASE; - gpio2->dir = (0x01 << gpio) | (value << gpio); + gpio2->dir |= 0x01 << gpio; + gpio2->output = (GPIO2_OUTPUT_ENABLE_MASK << gpio) | ((!!value) << gpio); return 0; } @@ -90,6 +91,7 @@ static int au1xxx_gpio1_direction_input(unsigned gpio) static int au1xxx_gpio1_direction_output(unsigned gpio, int value) { gpio1->trioutclr = (0x01 & gpio); + au1xxx_gpio1_write(gpio, value); return 0; } diff --git a/arch/mips/kernel/Makefile b/arch/mips/kernel/Makefile index 706f939..25775cb 100644 --- a/arch/mips/kernel/Makefile +++ b/arch/mips/kernel/Makefile @@ -10,6 +10,7 @@ obj-y += cpu-probe.o branch.o entry.o genex.o irq.o process.o \ obj-$(CONFIG_CEVT_BCM1480) += cevt-bcm1480.o obj-$(CONFIG_CEVT_R4K) += cevt-r4k.o +obj-$(CONFIG_MIPS_MT_SMTC) += cevt-smtc.o obj-$(CONFIG_CEVT_DS1287) += cevt-ds1287.o obj-$(CONFIG_CEVT_GT641XX) += cevt-gt641xx.o obj-$(CONFIG_CEVT_SB1250) += cevt-sb1250.o diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c index 24a2d90..4a4c59f 100644 --- a/arch/mips/kernel/cevt-r4k.c +++ b/arch/mips/kernel/cevt-r4k.c @@ -12,6 +12,14 @@ #include <asm/smtc_ipi.h> #include <asm/time.h> +#include <asm/cevt-r4k.h> + +/* + * The SMTC Kernel for the 34K, 1004K, et. al. replaces several + * of these routines with SMTC-specific variants. + */ + +#ifndef CONFIG_MIPS_MT_SMTC static int mips_next_event(unsigned long delta, struct clock_event_device *evt) @@ -19,60 +27,27 @@ static int mips_next_event(unsigned long delta, unsigned int cnt; int res; -#ifdef CONFIG_MIPS_MT_SMTC - { - unsigned long flags, vpflags; - local_irq_save(flags); - vpflags = dvpe(); -#endif cnt = read_c0_count(); cnt += delta; write_c0_compare(cnt); res = ((int)(read_c0_count() - cnt) > 0) ? -ETIME : 0; -#ifdef CONFIG_MIPS_MT_SMTC - evpe(vpflags); - local_irq_restore(flags); - } -#endif return res; } -static void mips_set_mode(enum clock_event_mode mode, - struct clock_event_device *evt) +#endif /* CONFIG_MIPS_MT_SMTC */ + +void mips_set_clock_mode(enum clock_event_mode mode, + struct clock_event_device *evt) { /* Nothing to do ... */ } -static DEFINE_PER_CPU(struct clock_event_device, mips_clockevent_device); -static int cp0_timer_irq_installed; +DEFINE_PER_CPU(struct clock_event_device, mips_clockevent_device); +int cp0_timer_irq_installed; -/* - * Timer ack for an R4k-compatible timer of a known frequency. - */ -static void c0_timer_ack(void) -{ - write_c0_compare(read_c0_compare()); -} +#ifndef CONFIG_MIPS_MT_SMTC -/* - * Possibly handle a performance counter interrupt. - * Return true if the timer interrupt should not be checked - */ -static inline int handle_perf_irq(int r2) -{ - /* - * The performance counter overflow interrupt may be shared with the - * timer interrupt (cp0_perfcount_irq < 0). If it is and a - * performance counter has overflowed (perf_irq() == IRQ_HANDLED) - * and we can't reliably determine if a counter interrupt has also - * happened (!r2) then don't check for a timer interrupt. - */ - return (cp0_perfcount_irq < 0) && - perf_irq() == IRQ_HANDLED && - !r2; -} - -static irqreturn_t c0_compare_interrupt(int irq, void *dev_id) +irqreturn_t c0_compare_interrupt(int irq, void *dev_id) { const int r2 = cpu_has_mips_r2; struct clock_event_device *cd; @@ -93,12 +68,8 @@ static irqreturn_t c0_compare_interrupt(int irq, void *dev_id) * interrupt. Being the paranoiacs we are we check anyway. */ if (!r2 || (read_c0_cause() & (1 << 30))) { - c0_timer_ack(); -#ifdef CONFIG_MIPS_MT_SMTC - if (cpu_data[cpu].vpe_id) - goto out; - cpu = 0; -#endif + /* Clear Count/Compare Interrupt */ + write_c0_compare(read_c0_compare()); cd = &per_cpu(mips_clockevent_device, cpu); cd->event_handler(cd); } @@ -107,65 +78,16 @@ out: return IRQ_HANDLED; } -static struct irqaction c0_compare_irqaction = { +#endif /* Not CONFIG_MIPS_MT_SMTC */ + +struct irqaction c0_compare_irqaction = { .handler = c0_compare_interrupt, -#ifdef CONFIG_MIPS_MT_SMTC - .flags = IRQF_DISABLED, -#else .flags = IRQF_DISABLED | IRQF_PERCPU, -#endif .name = "timer", }; -#ifdef CONFIG_MIPS_MT_SMTC -DEFINE_PER_CPU(struct clock_event_device, smtc_dummy_clockevent_device); - -static void smtc_set_mode(enum clock_event_mode mode, - struct clock_event_device *evt) -{ -} - -static void mips_broadcast(cpumask_t mask) -{ - unsigned int cpu; - - for_each_cpu_mask(cpu, mask) - smtc_send_ipi(cpu, SMTC_CLOCK_TICK, 0); -} - -static void setup_smtc_dummy_clockevent_device(void) -{ - //uint64_t mips_freq = mips_hpt_^frequency; - unsigned int cpu = smp_processor_id(); - struct clock_event_device *cd; - cd = &per_cpu(smtc_dummy_clockevent_device, cpu); - - cd->name = "SMTC"; - cd->features = CLOCK_EVT_FEAT_DUMMY; - - /* Calculate the min / max delta */ - cd->mult = 0; //div_sc((unsigned long) mips_freq, NSEC_PER_SEC, 32); - cd->shift = 0; //32; - cd->max_delta_ns = 0; //clockevent_delta2ns(0x7fffffff, cd); - cd->min_delta_ns = 0; //clockevent_delta2ns(0x30, cd); - - cd->rating = 200; - cd->irq = 17; //-1; -// if (cpu) -// cd->cpumask = CPU_MASK_ALL; // cpumask_of_cpu(cpu); -// else - cd->cpumask = cpumask_of_cpu(cpu); - - cd->set_mode = smtc_set_mode; - - cd->broadcast = mips_broadcast; - - clockevents_register_device(cd); -} -#endif - -static void mips_event_handler(struct clock_event_device *dev) +void mips_event_handler(struct clock_event_device *dev) { } @@ -177,7 +99,23 @@ static int c0_compare_int_pending(void) return (read_c0_cause() >> cp0_compare_irq) & 0x100; } -static int c0_compare_int_usable(void) +/* + * Compare interrupt can be routed and latched outside the core, + * so a single execution hazard barrier may not be enough to give + * it time to clear as seen in the Cause register. 4 time the + * pipeline depth seems reasonably conservative, and empirically + * works better in configurations with high CPU/bus clock ratios. + */ + +#define compare_change_hazard() \ + do { \ + irq_disable_hazard(); \ + irq_disable_hazard(); \ + irq_disable_hazard(); \ + irq_disable_hazard(); \ + } while (0) + +int c0_compare_int_usable(void) { unsigned int delta; unsigned int cnt; @@ -187,7 +125,7 @@ static int c0_compare_int_usable(void) */ if (c0_compare_int_pending()) { write_c0_compare(read_c0_count()); - irq_disable_hazard(); + compare_change_hazard(); if (c0_compare_int_pending()) return 0; } @@ -196,7 +134,7 @@ static int c0_compare_int_usable(void) cnt = read_c0_count(); cnt += delta; write_c0_compare(cnt); - irq_disable_hazard(); + compare_change_hazard(); if ((int)(read_c0_count() - cnt) < 0) break; /* increase delta if the timer was already expired */ @@ -205,11 +143,12 @@ static int c0_compare_int_usable(void) while ((int)(read_c0_count() - cnt) <= 0) ; /* Wait for expiry */ + compare_change_hazard(); if (!c0_compare_int_pending()) return 0; write_c0_compare(read_c0_count()); - irq_disable_hazard(); + compare_change_hazard(); if (c0_compare_int_pending()) return 0; @@ -219,6 +158,8 @@ static int c0_compare_int_usable(void) return 1; } +#ifndef CONFIG_MIPS_MT_SMTC + int __cpuinit mips_clockevent_init(void) { uint64_t mips_freq = mips_hpt_frequency; @@ -229,17 +170,6 @@ int __cpuinit mips_clockevent_init(void) if (!cpu_has_counter || !mips_hpt_frequency) return -ENXIO; -#ifdef CONFIG_MIPS_MT_SMTC - setup_smtc_dummy_clockevent_device(); - - /* - * On SMTC we only register VPE0's compare interrupt as clockevent - * device. - */ - if (cpu) - return 0; -#endif - if (!c0_compare_int_usable()) return -ENXIO; @@ -265,13 +195,9 @@ int __cpuinit mips_clockevent_init(void) cd->rating = 300; cd->irq = irq; -#ifdef CONFIG_MIPS_MT_SMTC - cd->cpumask = CPU_MASK_ALL; -#else cd->cpumask = cpumask_of_cpu(cpu); -#endif cd->set_next_event = mips_next_event; - cd->set_mode = mips_set_mode; + cd->set_mode = mips_set_clock_mode; cd->event_handler = mips_event_handler; clockevents_register_device(cd); @@ -281,12 +207,9 @@ int __cpuinit mips_clockevent_init(void) cp0_timer_irq_installed = 1; -#ifdef CONFIG_MIPS_MT_SMTC -#define CPUCTR_IMASKBIT (0x100 << cp0_compare_irq) - setup_irq_smtc(irq, &c0_compare_irqaction, CPUCTR_IMASKBIT); -#else setup_irq(irq, &c0_compare_irqaction); -#endif return 0; } + +#endif /* Not CONFIG_MIPS_MT_SMTC */ diff --git a/arch/mips/kernel/cevt-smtc.c b/arch/mips/kernel/cevt-smtc.c new file mode 100644 index 0000000..5162fe4 --- /dev/null +++ b/arch/mips/kernel/cevt-smtc.c @@ -0,0 +1,321 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2007 MIPS Technologies, Inc. + * Copyright (C) 2007 Ralf Baechle <ralf@linux-mips.org> + * Copyright (C) 2008 Kevin D. Kissell, Paralogos sarl + */ +#include <linux/clockchips.h> +#include <linux/interrupt.h> +#include <linux/percpu.h> + +#include <asm/smtc_ipi.h> +#include <asm/time.h> +#include <asm/cevt-r4k.h> + +/* + * Variant clock event timer support for SMTC on MIPS 34K, 1004K + * or other MIPS MT cores. + * + * Notes on SMTC Support: + * + * SMTC has multiple microthread TCs pretending to be Linux CPUs. + * But there's only one Count/Compare pair per VPE, and Compare + * interrupts are taken opportunisitically by available TCs + * bound to the VPE with the Count register. The new timer + * framework provides for global broadcasts, but we really + * want VPE-level multicasts for best behavior. So instead + * of invoking the high-level clock-event broadcast code, + * this version of SMTC support uses the historical SMTC + * multicast mechanisms "under the hood", appearing to the + * generic clock layer as if the interrupts are per-CPU. + * + * The approach taken here is to maintain a set of NR_CPUS + * virtual timers, and track which "CPU" needs to be alerted + * at each event. + * + * It's unlikely that we'll see a MIPS MT core with more than + * 2 VPEs, but we *know* that we won't need to handle more + * VPEs than we have "CPUs". So NCPUs arrays of NCPUs elements + * is always going to be overkill, but always going to be enough. + */ + +unsigned long smtc_nexttime[NR_CPUS][NR_CPUS]; +static int smtc_nextinvpe[NR_CPUS]; + +/* + * Timestamps stored are absolute values to be programmed + * into Count register. Valid timestamps will never be zero. + * If a Zero Count value is actually calculated, it is converted + * to be a 1, which will introduce 1 or two CPU cycles of error + * roughly once every four billion events, which at 1000 HZ means + * about once every 50 days. If that's actually a problem, one + * could alternate squashing 0 to 1 and to -1. + */ + +#define MAKEVALID(x) (((x) == 0L) ? 1L : (x)) +#define ISVALID(x) ((x) != 0L) + +/* + * Time comparison is subtle, as it's really truncated + * modular arithmetic. + */ + +#define IS_SOONER(a, b, reference) \ + (((a) - (unsigned long)(reference)) < ((b) - (unsigned long)(reference))) + +/* + * CATCHUP_INCREMENT, used when the function falls behind the counter. + * Could be an increasing function instead of a constant; + */ + +#define CATCHUP_INCREMENT 64 + +static int mips_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + unsigned long flags; + unsigned int mtflags; + unsigned long timestamp, reference, previous; + unsigned long nextcomp = 0L; + int vpe = current_cpu_data.vpe_id; + int cpu = smp_processor_id(); + local_irq_save(flags); + mtflags = dmt(); + + /* + * Maintain the per-TC virtual timer + * and program the per-VPE shared Count register + * as appropriate here... + */ + reference = (unsigned long)read_c0_count(); + timestamp = MAKEVALID(reference + delta); + /* + * To really model the clock, we have to catch the case + * where the current next-in-VPE timestamp is the old + * timestamp for the calling CPE, but the new value is + * in fact later. In that case, we have to do a full + * scan and discover the new next-in-VPE CPU id and + * timestamp. + */ + previous = smtc_nexttime[vpe][cpu]; + if (cpu == smtc_nextinvpe[vpe] && ISVALID(previous) + && IS_SOONER(previous, timestamp, reference)) { + int i; + int soonest = cpu; + + /* + * Update timestamp array here, so that new + * value gets considered along with those of + * other virtual CPUs on the VPE. + */ + smtc_nexttime[vpe][cpu] = timestamp; + for_each_online_cpu(i) { + if (ISVALID(smtc_nexttime[vpe][i]) + && IS_SOONER(smtc_nexttime[vpe][i], + smtc_nexttime[vpe][soonest], reference)) { + soonest = i; + } + } + smtc_nextinvpe[vpe] = soonest; + nextcomp = smtc_nexttime[vpe][soonest]; + /* + * Otherwise, we don't have to process the whole array rank, + * we just have to see if the event horizon has gotten closer. + */ + } else { + if (!ISVALID(smtc_nexttime[vpe][smtc_nextinvpe[vpe]]) || + IS_SOONER(timestamp, + smtc_nexttime[vpe][smtc_nextinvpe[vpe]], reference)) { + smtc_nextinvpe[vpe] = cpu; + nextcomp = timestamp; + } + /* + * Since next-in-VPE may me the same as the executing + * virtual CPU, we update the array *after* checking + * its value. + */ + smtc_nexttime[vpe][cpu] = timestamp; + } + + /* + * It may be that, in fact, we don't need to update Compare, + * but if we do, we want to make sure we didn't fall into + * a crack just behind Count. + */ + if (ISVALID(nextcomp)) { + write_c0_compare(nextcomp); + ehb(); + /* + * We never return an error, we just make sure + * that we trigger the handlers as quickly as + * we can if we fell behind. + */ + while ((nextcomp - (unsigned long)read_c0_count()) + > (unsigned long)LONG_MAX) { + nextcomp += CATCHUP_INCREMENT; + write_c0_compare(nextcomp); + ehb(); + } + } + emt(mtflags); + local_irq_restore(flags); + return 0; +} + + +void smtc_distribute_timer(int vpe) +{ + unsigned long flags; + unsigned int mtflags; + int cpu; + struct clock_event_device *cd; + unsigned long nextstamp = 0L; + unsigned long reference; + + +repeat: + for_each_online_cpu(cpu) { + /* + * Find virtual CPUs within the current VPE who have + * unserviced timer requests whose time is now past. + */ + local_irq_save(flags); + mtflags = dmt(); + if (cpu_data[cpu].vpe_id == vpe && + ISVALID(smtc_nexttime[vpe][cpu])) { + reference = (unsigned long)read_c0_count(); + if ((smtc_nexttime[vpe][cpu] - reference) + > (unsigned long)LONG_MAX) { + smtc_nexttime[vpe][cpu] = 0L; + emt(mtflags); + local_irq_restore(flags); + /* + * We don't send IPIs to ourself. + */ + if (cpu != smp_processor_id()) { + smtc_send_ipi(cpu, SMTC_CLOCK_TICK, 0); + } else { + cd = &per_cpu(mips_clockevent_device, cpu); + cd->event_handler(cd); + } + } else { + /* Local to VPE but Valid Time not yet reached. */ + if (!ISVALID(nextstamp) || + IS_SOONER(smtc_nexttime[vpe][cpu], nextstamp, + reference)) { + smtc_nextinvpe[vpe] = cpu; + nextstamp = smtc_nexttime[vpe][cpu]; + } + emt(mtflags); + local_irq_restore(flags); + } + } else { + emt(mtflags); + local_irq_restore(flags); + + } + } + /* Reprogram for interrupt at next soonest timestamp for VPE */ + if (ISVALID(nextstamp)) { + write_c0_compare(nextstamp); + ehb(); + if ((nextstamp - (unsigned long)read_c0_count()) + > (unsigned long)LONG_MAX) + goto repeat; + } +} + + +irqreturn_t c0_compare_interrupt(int irq, void *dev_id) +{ + int cpu = smp_processor_id(); + + /* If we're running SMTC, we've got MIPS MT and therefore MIPS32R2 */ + handle_perf_irq(1); + + if (read_c0_cause() & (1 << 30)) { + /* Clear Count/Compare Interrupt */ + write_c0_compare(read_c0_compare()); + smtc_distribute_timer(cpu_data[cpu].vpe_id); + } + return IRQ_HANDLED; +} + + +int __cpuinit mips_clockevent_init(void) +{ + uint64_t mips_freq = mips_hpt_frequency; + unsigned int cpu = smp_processor_id(); + struct clock_event_device *cd; + unsigned int irq; + int i; + int j; + + if (!cpu_has_counter || !mips_hpt_frequency) + return -ENXIO; + if (cpu == 0) { + for (i = 0; i < num_possible_cpus(); i++) { + smtc_nextinvpe[i] = 0; + for (j = 0; j < num_possible_cpus(); j++) + smtc_nexttime[i][j] = 0L; + } + /* + * SMTC also can't have the usablility test + * run by secondary TCs once Compare is in use. + */ + if (!c0_compare_int_usable()) + return -ENXIO; + } + + /* + * With vectored interrupts things are getting platform specific. + * get_c0_compare_int is a hook to allow a platform to return the + * interrupt number of it's liking. + */ + irq = MIPS_CPU_IRQ_BASE + cp0_compare_irq; + if (get_c0_compare_int) + irq = get_c0_compare_int(); + + cd = &per_cpu(mips_clockevent_device, cpu); + + cd->name = "MIPS"; + cd->features = CLOCK_EVT_FEAT_ONESHOT; + + /* Calculate the min / max delta */ + cd->mult = div_sc((unsigned long) mips_freq, NSEC_PER_SEC, 32); + cd->shift = 32; + cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd); + cd->min_delta_ns = clockevent_delta2ns(0x300, cd); + + cd->rating = 300; + cd->irq = irq; + cd->cpumask = cpumask_of_cpu(cpu); + cd->set_next_event = mips_next_event; + cd->set_mode = mips_set_clock_mode; + cd->event_handler = mips_event_handler; + + clockevents_register_device(cd); + + /* + * On SMTC we only want to do the data structure + * initialization and IRQ setup once. + */ + if (cpu) + return 0; + /* + * And we need the hwmask associated with the c0_compare + * vector to be initialized. + */ + irq_hwmask[irq] = (0x100 << cp0_compare_irq); + if (cp0_timer_irq_installed) + return 0; + + cp0_timer_irq_installed = 1; + + setup_irq(irq, &c0_compare_irqaction); + + return 0; +} diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c index 335a6ae..e621fda 100644 --- a/arch/mips/kernel/cpu-probe.c +++ b/arch/mips/kernel/cpu-probe.c @@ -45,18 +45,7 @@ static void r39xx_wait(void) local_irq_enable(); } -/* - * There is a race when WAIT instruction executed with interrupt - * enabled. - * But it is implementation-dependent wheter the pipelie restarts when - * a non-enabled interrupt is requested. - */ -static void r4k_wait(void) -{ - __asm__(" .set mips3 \n" - " wait \n" - " .set mips0 \n"); -} +extern void r4k_wait(void); /* * This variant is preferable as it allows testing need_resched and going to @@ -65,14 +54,18 @@ static void r4k_wait(void) * interrupt is requested" restriction in the MIPS32/MIPS64 architecture makes * using this version a gamble. */ -static void r4k_wait_irqoff(void) +void r4k_wait_irqoff(void) { local_irq_disable(); if (!need_resched()) - __asm__(" .set mips3 \n" + __asm__(" .set push \n" + " .set mips3 \n" " wait \n" - " .set mips0 \n"); + " .set pop \n"); local_irq_enable(); + __asm__(" .globl __pastwait \n" + "__pastwait: \n"); + return; } /* @@ -128,7 +121,7 @@ static int __init wait_disable(char *s) __setup("nowait", wait_disable); -static inline void check_wait(void) +void __init check_wait(void) { struct cpuinfo_mips *c = ¤t_cpu_data; @@ -242,7 +235,6 @@ static inline void check_errata(void) void __init check_bugs32(void) { - check_wait(); check_errata(); } diff --git a/arch/mips/kernel/entry.S b/arch/mips/kernel/entry.S index e29598a..ffa3310 100644 --- a/arch/mips/kernel/entry.S +++ b/arch/mips/kernel/entry.S @@ -79,11 +79,6 @@ FEXPORT(syscall_exit) FEXPORT(restore_all) # restore full frame #ifdef CONFIG_MIPS_MT_SMTC -/* Detect and execute deferred IPI "interrupts" */ - LONG_L s0, TI_REGS($28) - LONG_S sp, TI_REGS($28) - jal deferred_smtc_ipi - LONG_S s0, TI_REGS($28) #ifdef CONFIG_MIPS_MT_SMTC_IM_BACKSTOP /* Re-arm any temporarily masked interrupts not explicitly "acked" */ mfc0 v0, CP0_TCSTATUS @@ -112,6 +107,11 @@ FEXPORT(restore_all) # restore full frame xor t0, t0, t3 mtc0 t0, CP0_TCCONTEXT #endif /* CONFIG_MIPS_MT_SMTC_IM_BACKSTOP */ +/* Detect and execute deferred IPI "interrupts" */ + LONG_L s0, TI_REGS($28) + LONG_S sp, TI_REGS($28) + jal deferred_smtc_ipi + LONG_S s0, TI_REGS($28) #endif /* CONFIG_MIPS_MT_SMTC */ .set noat RESTORE_TEMP diff --git a/arch/mips/kernel/genex.S b/arch/mips/kernel/genex.S index c6ada98..01dcbe3 100644 --- a/arch/mips/kernel/genex.S +++ b/arch/mips/kernel/genex.S @@ -20,6 +20,7 @@ #include <asm/stackframe.h> #include <asm/war.h> #include <asm/page.h> +#include <asm/thread_info.h> #define PANIC_PIC(msg) \ .set push; \ @@ -126,7 +127,42 @@ handle_vcei: __FINIT + .align 5 /* 32 byte rollback region */ +LEAF(r4k_wait) + .set push + .set noreorder + /* start of rollback region */ + LONG_L t0, TI_FLAGS($28) + nop + andi t0, _TIF_NEED_RESCHED + bnez t0, 1f + nop + nop + nop + .set mips3 + wait + /* end of rollback region (the region size must be power of two) */ + .set pop +1: + jr ra + END(r4k_wait) + + .macro BUILD_ROLLBACK_PROLOGUE handler + FEXPORT(rollback_\handler) + .set push + .set noat + MFC0 k0, CP0_EPC + PTR_LA k1, r4k_wait + ori k0, 0x1f /* 32 byte rollback region */ + xori k0, 0x1f + bne k0, k1, 9f + MTC0 k0, CP0_EPC +9: + .set pop + .endm + .align 5 +BUILD_ROLLBACK_PROLOGUE handle_int NESTED(handle_int, PT_SIZE, sp) #ifdef CONFIG_TRACE_IRQFLAGS /* @@ -201,6 +237,7 @@ NESTED(except_vec_ejtag_debug, 0, sp) * This prototype is copied to ebase + n*IntCtl.VS and patched * to invoke the handler */ +BUILD_ROLLBACK_PROLOGUE except_vec_vi NESTED(except_vec_vi, 0, sp) SAVE_SOME SAVE_AT @@ -245,8 +282,8 @@ NESTED(except_vec_vi_handler, 0, sp) and t0, a0, t1 #ifdef CONFIG_MIPS_MT_SMTC_IM_BACKSTOP mfc0 t2, CP0_TCCONTEXT - or t0, t0, t2 - mtc0 t0, CP0_TCCONTEXT + or t2, t0, t2 + mtc0 t2, CP0_TCCONTEXT #endif /* CONFIG_MIPS_MT_SMTC_IM_BACKSTOP */ xor t1, t1, t0 mtc0 t1, CP0_STATUS diff --git a/arch/mips/kernel/head.S b/arch/mips/kernel/head.S index 3613645..492a0a8 100644 --- a/arch/mips/kernel/head.S +++ b/arch/mips/kernel/head.S @@ -22,6 +22,7 @@ #include <asm/irqflags.h> #include <asm/regdef.h> #include <asm/page.h> +#include <asm/pgtable-bits.h> #include <asm/mipsregs.h> #include <asm/stackframe.h> diff --git a/arch/mips/kernel/kgdb.c b/arch/mips/kernel/kgdb.c index 8f6d58e..6e152c8 100644 --- a/arch/mips/kernel/kgdb.c +++ b/arch/mips/kernel/kgdb.c @@ -236,8 +236,7 @@ int kgdb_arch_handle_exception(int vector, int signo, int err_code, atomic_set(&kgdb_cpu_doing_single_step, -1); if (remcom_in_buffer[0] == 's') - if (kgdb_contthread) - atomic_set(&kgdb_cpu_doing_single_step, cpu); + atomic_set(&kgdb_cpu_doing_single_step, cpu); return 0; } diff --git a/arch/mips/kernel/mips-mt-fpaff.c b/arch/mips/kernel/mips-mt-fpaff.c index df4d3f2..dc9eb72 100644 --- a/arch/mips/kernel/mips-mt-fpaff.c +++ b/arch/mips/kernel/mips-mt-fpaff.c @@ -159,7 +159,7 @@ __setup("fpaff=", fpaff_thresh); /* * FPU Use Factor empirically derived from experiments on 34K */ -#define FPUSEFACTOR 333 +#define FPUSEFACTOR 2000 static __init int mt_fp_affinity_init(void) { diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c index b16facd..22fc19b 100644 --- a/arch/mips/kernel/process.c +++ b/arch/mips/kernel/process.c @@ -55,7 +55,7 @@ void __noreturn cpu_idle(void) while (1) { tick_nohz_stop_sched_tick(1); while (!need_resched()) { -#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG +#ifdef CONFIG_MIPS_MT_SMTC extern void smtc_idle_loop_hook(void); smtc_idle_loop_hook(); @@ -145,17 +145,18 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp, */ p->thread.cp0_status = read_c0_status() & ~(ST0_CU2|ST0_CU1); childregs->cp0_status &= ~(ST0_CU2|ST0_CU1); - clear_tsk_thread_flag(p, TIF_USEDFPU); -#ifdef CONFIG_MIPS_MT_FPAFF +#ifdef CONFIG_MIPS_MT_SMTC /* - * FPU affinity support is cleaner if we track the - * user-visible CPU affinity from the very beginning. - * The generic cpus_allowed mask will already have - * been copied from the parent before copy_thread - * is invoked. + * SMTC restores TCStatus after Status, and the CU bits + * are aliased there. */ - p->thread.user_cpus_allowed = p->cpus_allowed; + childregs->cp0_tcstatus &= ~(ST0_CU2|ST0_CU1); +#endif + clear_tsk_thread_flag(p, TIF_USEDFPU); + +#ifdef CONFIG_MIPS_MT_FPAFF + clear_tsk_thread_flag(p, TIF_FPUBOUND); #endif /* CONFIG_MIPS_MT_FPAFF */ if (clone_flags & CLONE_SETTLS) diff --git a/arch/mips/kernel/ptrace.c b/arch/mips/kernel/ptrace.c index 35234b9..96ffc9c 100644 --- a/arch/mips/kernel/ptrace.c +++ b/arch/mips/kernel/ptrace.c @@ -238,7 +238,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) case FPC_EIR: { /* implementation / version register */ unsigned int flags; #ifdef CONFIG_MIPS_MT_SMTC - unsigned int irqflags; + unsigned long irqflags; unsigned int mtflags; #endif /* CONFIG_MIPS_MT_SMTC */ diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c index a516286..897fb2b 100644 --- a/arch/mips/kernel/smtc.c +++ b/arch/mips/kernel/smtc.c @@ -1,4 +1,21 @@ -/* Copyright (C) 2004 Mips Technologies, Inc */ +/* + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) 2004 Mips Technologies, Inc + * Copyright (C) 2008 Kevin D. Kissell + */ #include <linux/clockchips.h> #include <linux/kernel.h> @@ -21,7 +38,6 @@ #include <asm/time.h> #include <asm/addrspace.h> #include <asm/smtc.h> -#include <asm/smtc_ipi.h> #include <asm/smtc_proc.h> /* @@ -58,11 +74,6 @@ unsigned long irq_hwmask[NR_IRQS]; asiduse smtc_live_asid[MAX_SMTC_TLBS][MAX_SMTC_ASIDS]; -/* - * Clock interrupt "latch" buffers, per "CPU" - */ - -static atomic_t ipi_timer_latch[NR_CPUS]; /* * Number of InterProcessor Interrupt (IPI) message buffers to allocate @@ -70,7 +81,7 @@ static atomic_t ipi_timer_latch[NR_CPUS]; #define IPIBUF_PER_CPU 4 -static struct smtc_ipi_q IPIQ[NR_CPUS]; +struct smtc_ipi_q IPIQ[NR_CPUS]; static struct smtc_ipi_q freeIPIq; @@ -282,7 +293,7 @@ static void smtc_configure_tlb(void) * phys_cpu_present_map and the logical/physical mappings. */ -int __init mipsmt_build_cpu_map(int start_cpu_slot) +int __init smtc_build_cpu_map(int start_cpu_slot) { int i, ntcs; @@ -325,7 +336,12 @@ static void smtc_tc_setup(int vpe, int tc, int cpu) write_tc_c0_tcstatus((read_tc_c0_tcstatus() & ~(TCSTATUS_TKSU | TCSTATUS_DA | TCSTATUS_IXMT)) | TCSTATUS_A); - write_tc_c0_tccontext(0); + /* + * TCContext gets an offset from the base of the IPIQ array + * to be used in low-level code to detect the presence of + * an active IPI queue + */ + write_tc_c0_tccontext((sizeof(struct smtc_ipi_q) * cpu) << 16); /* Bind tc to vpe */ write_tc_c0_tcbind(vpe); /* In general, all TCs should have the same cpu_data indications */ @@ -336,10 +352,18 @@ static void smtc_tc_setup(int vpe, int tc, int cpu) cpu_data[cpu].options &= ~MIPS_CPU_FPU; cpu_data[cpu].vpe_id = vpe; cpu_data[cpu].tc_id = tc; + /* Multi-core SMTC hasn't been tested, but be prepared */ + cpu_data[cpu].core = (read_vpe_c0_ebase() >> 1) & 0xff; } +/* + * Tweak to get Count registes in as close a sync as possible. + * Value seems good for 34K-class cores. + */ + +#define CP0_SKEW 8 -void mipsmt_prepare_cpus(void) +void smtc_prepare_cpus(int cpus) { int i, vpe, tc, ntc, nvpe, tcpervpe[NR_CPUS], slop, cpu; unsigned long flags; @@ -363,13 +387,13 @@ void mipsmt_prepare_cpus(void) IPIQ[i].head = IPIQ[i].tail = NULL; spin_lock_init(&IPIQ[i].lock); IPIQ[i].depth = 0; - atomic_set(&ipi_timer_latch[i], 0); } /* cpu_data index starts at zero */ cpu = 0; cpu_data[cpu].vpe_id = 0; cpu_data[cpu].tc_id = 0; + cpu_data[cpu].core = (read_c0_ebase() >> 1) & 0xff; cpu++; /* Report on boot-time options */ @@ -484,7 +508,8 @@ void mipsmt_prepare_cpus(void) write_vpe_c0_compare(0); /* Propagate Config7 */ write_vpe_c0_config7(read_c0_config7()); - write_vpe_c0_count(read_c0_count()); + write_vpe_c0_count(read_c0_count() + CP0_SKEW); + ehb(); } /* enable multi-threading within VPE */ write_vpe_c0_vpecontrol(read_vpe_c0_vpecontrol() | VPECONTROL_TE); @@ -556,7 +581,7 @@ void mipsmt_prepare_cpus(void) void __cpuinit smtc_boot_secondary(int cpu, struct task_struct *idle) { extern u32 kernelsp[NR_CPUS]; - long flags; + unsigned long flags; int mtflags; LOCK_MT_PRA(); @@ -585,24 +610,22 @@ void __cpuinit smtc_boot_secondary(int cpu, struct task_struct *idle) void smtc_init_secondary(void) { - /* - * Start timer on secondary VPEs if necessary. - * plat_timer_setup has already have been invoked by init/main - * on "boot" TC. Like per_cpu_trap_init() hack, this assumes that - * SMTC init code assigns TCs consdecutively and in ascending order - * to across available VPEs. - */ - if (((read_c0_tcbind() & TCBIND_CURTC) != 0) && - ((read_c0_tcbind() & TCBIND_CURVPE) - != cpu_data[smp_processor_id() - 1].vpe_id)){ - write_c0_compare(read_c0_count() + mips_hpt_frequency/HZ); - } - local_irq_enable(); } void smtc_smp_finish(void) { + int cpu = smp_processor_id(); + + /* + * Lowest-numbered CPU per VPE starts a clock tick. + * Like per_cpu_trap_init() hack, this assumes that + * SMTC init code assigns TCs consdecutively and + * in ascending order across available VPEs. + */ + if (cpu > 0 && (cpu_data[cpu].vpe_id != cpu_data[cpu - 1].vpe_id)) + write_c0_compare(read_c0_count() + mips_hpt_frequency/HZ); + printk("TC %d going on-line as CPU %d\n", cpu_data[smp_processor_id()].tc_id, smp_processor_id()); } @@ -753,8 +776,10 @@ void smtc_send_ipi(int cpu, int type, unsigned int action) { int tcstatus; struct smtc_ipi *pipi; - long flags; + unsigned long flags; int mtflags; + unsigned long tcrestart; + extern void r4k_wait_irqoff(void), __pastwait(void); if (cpu == smp_processor_id()) { printk("Cannot Send IPI to self!\n"); @@ -771,8 +796,6 @@ void smtc_send_ipi(int cpu, int type, unsigned int action) pipi->arg = (void *)action; pipi->dest = cpu; if (cpu_data[cpu].vpe_id != cpu_data[smp_processor_id()].vpe_id) { - if (type == SMTC_CLOCK_TICK) - atomic_inc(&ipi_timer_latch[cpu]); /* If not on same VPE, enqueue and send cross-VPE interrupt */ smtc_ipi_nq(&IPIQ[cpu], pipi); LOCK_CORE_PRA(); @@ -800,22 +823,29 @@ void smtc_send_ipi(int cpu, int type, unsigned int action) if ((tcstatus & TCSTATUS_IXMT) != 0) { /* - * Spin-waiting here can deadlock, - * so we queue the message for the target TC. + * If we're in the the irq-off version of the wait + * loop, we need to force exit from the wait and + * do a direct post of the IPI. + */ + if (cpu_wait == r4k_wait_irqoff) { + tcrestart = read_tc_c0_tcrestart(); + if (tcrestart >= (unsigned long)r4k_wait_irqoff + && tcrestart < (unsigned long)__pastwait) { + write_tc_c0_tcrestart(__pastwait); + tcstatus &= ~TCSTATUS_IXMT; + write_tc_c0_tcstatus(tcstatus); + goto postdirect; + } + } + /* + * Otherwise we queue the message for the target TC + * to pick up when he does a local_irq_restore() */ write_tc_c0_tchalt(0); UNLOCK_CORE_PRA(); - /* Try to reduce redundant timer interrupt messages */ - if (type == SMTC_CLOCK_TICK) { - if (atomic_postincrement(&ipi_timer_latch[cpu])!=0){ - smtc_ipi_nq(&freeIPIq, pipi); - return; - } - } smtc_ipi_nq(&IPIQ[cpu], pipi); } else { - if (type == SMTC_CLOCK_TICK) - atomic_inc(&ipi_timer_latch[cpu]); +postdirect: post_direct_ipi(cpu, pipi); write_tc_c0_tchalt(0); UNLOCK_CORE_PRA(); @@ -883,7 +913,7 @@ static void ipi_call_interrupt(void) smp_call_function_interrupt(); } -DECLARE_PER_CPU(struct clock_event_device, smtc_dummy_clockevent_device); +DECLARE_PER_CPU(struct clock_event_device, mips_clockevent_device); void ipi_decode(struct smtc_ipi *pipi) { @@ -891,20 +921,13 @@ void ipi_decode(struct smtc_ipi *pipi) struct clock_event_device *cd; void *arg_copy = pipi->arg; int type_copy = pipi->type; - int ticks; - smtc_ipi_nq(&freeIPIq, pipi); switch (type_copy) { case SMTC_CLOCK_TICK: irq_enter(); kstat_this_cpu.irqs[MIPS_CPU_IRQ_BASE + 1]++; - cd = &per_cpu(smtc_dummy_clockevent_device, cpu); - ticks = atomic_read(&ipi_timer_latch[cpu]); - atomic_sub(ticks, &ipi_timer_latch[cpu]); - while (ticks) { - cd->event_handler(cd); - ticks--; - } + cd = &per_cpu(mips_clockevent_device, cpu); + cd->event_handler(cd); irq_exit(); break; @@ -937,24 +960,48 @@ void ipi_decode(struct smtc_ipi *pipi) } } +/* + * Similar to smtc_ipi_replay(), but invoked from context restore, + * so it reuses the current exception frame rather than set up a + * new one with self_ipi. + */ + void deferred_smtc_ipi(void) { - struct smtc_ipi *pipi; - unsigned long flags; -/* DEBUG */ - int q = smp_processor_id(); + int cpu = smp_processor_id(); /* * Test is not atomic, but much faster than a dequeue, * and the vast majority of invocations will have a null queue. + * If irq_disabled when this was called, then any IPIs queued + * after we test last will be taken on the next irq_enable/restore. + * If interrupts were enabled, then any IPIs added after the + * last test will be taken directly. */ - if (IPIQ[q].head != NULL) { - while((pipi = smtc_ipi_dq(&IPIQ[q])) != NULL) { - /* ipi_decode() should be called with interrupts off */ - local_irq_save(flags); + + while (IPIQ[cpu].head != NULL) { + struct smtc_ipi_q *q = &IPIQ[cpu]; + struct smtc_ipi *pipi; + unsigned long flags; + + /* + * It may be possible we'll come in with interrupts + * already enabled. + */ + local_irq_save(flags); + + spin_lock(&q->lock); + pipi = __smtc_ipi_dq(q); + spin_unlock(&q->lock); + if (pipi != NULL) ipi_decode(pipi); - local_irq_restore(flags); - } + /* + * The use of the __raw_local restore isn't + * as obviously necessary here as in smtc_ipi_replay(), + * but it's more efficient, given that we're already + * running down the IPI queue. + */ + __raw_local_irq_restore(flags); } } @@ -975,7 +1022,7 @@ static irqreturn_t ipi_interrupt(int irq, void *dev_idm) struct smtc_ipi *pipi; unsigned long tcstatus; int sent; - long flags; + unsigned long flags; unsigned int mtflags; unsigned int vpflags; @@ -1066,55 +1113,53 @@ static void setup_cross_vpe_interrupts(unsigned int nvpe) /* * SMTC-specific hacks invoked from elsewhere in the kernel. - * - * smtc_ipi_replay is called from raw_local_irq_restore which is only ever - * called with interrupts disabled. We do rely on interrupts being disabled - * here because using spin_lock_irqsave()/spin_unlock_irqrestore() would - * result in a recursive call to raw_local_irq_restore(). */ -static void __smtc_ipi_replay(void) + /* + * smtc_ipi_replay is called from raw_local_irq_restore + */ + +void smtc_ipi_replay(void) { unsigned int cpu = smp_processor_id(); /* * To the extent that we've ever turned interrupts off, * we may have accumulated deferred IPIs. This is subtle. - * If we use the smtc_ipi_qdepth() macro, we'll get an - * exact number - but we'll also disable interrupts - * and create a window of failure where a new IPI gets - * queued after we test the depth but before we re-enable - * interrupts. So long as IXMT never gets set, however, * we should be OK: If we pick up something and dispatch * it here, that's great. If we see nothing, but concurrent * with this operation, another TC sends us an IPI, IXMT * is clear, and we'll handle it as a real pseudo-interrupt - * and not a pseudo-pseudo interrupt. + * and not a pseudo-pseudo interrupt. The important thing + * is to do the last check for queued message *after* the + * re-enabling of interrupts. */ - if (IPIQ[cpu].depth > 0) { - while (1) { - struct smtc_ipi_q *q = &IPIQ[cpu]; - struct smtc_ipi *pipi; - extern void self_ipi(struct smtc_ipi *); - - spin_lock(&q->lock); - pipi = __smtc_ipi_dq(q); - spin_unlock(&q->lock); - if (!pipi) - break; + while (IPIQ[cpu].head != NULL) { + struct smtc_ipi_q *q = &IPIQ[cpu]; + struct smtc_ipi *pipi; + unsigned long flags; + + /* + * It's just possible we'll come in with interrupts + * already enabled. + */ + local_irq_save(flags); + + spin_lock(&q->lock); + pipi = __smtc_ipi_dq(q); + spin_unlock(&q->lock); + /* + ** But use a raw restore here to avoid recursion. + */ + __raw_local_irq_restore(flags); + if (pipi) { self_ipi(pipi); smtc_cpu_stats[cpu].selfipis++; } } } -void smtc_ipi_replay(void) -{ - raw_local_irq_disable(); - __smtc_ipi_replay(); -} - EXPORT_SYMBOL(smtc_ipi_replay); void smtc_idle_loop_hook(void) @@ -1193,40 +1238,13 @@ void smtc_idle_loop_hook(void) } } - /* - * Now that we limit outstanding timer IPIs, check for hung TC - */ - for (tc = 0; tc < NR_CPUS; tc++) { - /* Don't check ourself - we'll dequeue IPIs just below */ - if ((tc != smp_processor_id()) && - atomic_read(&ipi_timer_latch[tc]) > timerq_limit) { - if (clock_hang_reported[tc] == 0) { - pdb_msg += sprintf(pdb_msg, - "TC %d looks hung with timer latch at %d\n", - tc, atomic_read(&ipi_timer_latch[tc])); - clock_hang_reported[tc]++; - } - } - } emt(mtflags); local_irq_restore(flags); if (pdb_msg != &id_ho_db_msg[0]) printk("CPU%d: %s", smp_processor_id(), id_ho_db_msg); #endif /* CONFIG_SMTC_IDLE_HOOK_DEBUG */ - /* - * Replay any accumulated deferred IPIs. If "Instant Replay" - * is in use, there should never be any. - */ -#ifndef CONFIG_MIPS_MT_SMTC_INSTANT_REPLAY - { - unsigned long flags; - - local_irq_save(flags); - __smtc_ipi_replay(); - local_irq_restore(flags); - } -#endif /* CONFIG_MIPS_MT_SMTC_INSTANT_REPLAY */ + smtc_ipi_replay(); } void smtc_soft_dump(void) @@ -1242,10 +1260,6 @@ void smtc_soft_dump(void) printk("%d: %ld\n", i, smtc_cpu_stats[i].selfipis); } smtc_ipi_qdump(); - printk("Timer IPI Backlogs:\n"); - for (i=0; i < NR_CPUS; i++) { - printk("%d: %d\n", i, atomic_read(&ipi_timer_latch[i])); - } printk("%d Recoveries of \"stolen\" FPU\n", atomic_read(&smtc_fpu_recoveries)); } diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c index 6bee290..b602ac6 100644 --- a/arch/mips/kernel/traps.c +++ b/arch/mips/kernel/traps.c @@ -46,6 +46,9 @@ #include <asm/types.h> #include <asm/stacktrace.h> +extern void check_wait(void); +extern asmlinkage void r4k_wait(void); +extern asmlinkage void rollback_handle_int(void); extern asmlinkage void handle_int(void); extern asmlinkage void handle_tlbm(void); extern asmlinkage void handle_tlbl(void); @@ -822,8 +825,10 @@ static void mt_ase_fp_affinity(void) if (cpus_intersects(current->cpus_allowed, mt_fpu_cpumask)) { cpumask_t tmask; - cpus_and(tmask, current->thread.user_cpus_allowed, - mt_fpu_cpumask); + current->thread.user_cpus_allowed + = current->cpus_allowed; + cpus_and(tmask, current->cpus_allowed, + mt_fpu_cpumask); set_cpus_allowed(current, tmask); set_thread_flag(TIF_FPUBOUND); } @@ -1251,6 +1256,9 @@ static void *set_vi_srs_handler(int n, vi_handler_t addr, int srs) extern char except_vec_vi, except_vec_vi_lui; extern char except_vec_vi_ori, except_vec_vi_end; + extern char rollback_except_vec_vi; + char *vec_start = (cpu_wait == r4k_wait) ? + &rollback_except_vec_vi : &except_vec_vi; #ifdef CONFIG_MIPS_MT_SMTC /* * We need to provide the SMTC vectored interrupt handler @@ -1258,11 +1266,11 @@ static void *set_vi_srs_handler(int n, vi_handler_t addr, int srs) * Status.IM bit to be masked before going there. */ extern char except_vec_vi_mori; - const int mori_offset = &except_vec_vi_mori - &except_vec_vi; + const int mori_offset = &except_vec_vi_mori - vec_start; #endif /* CONFIG_MIPS_MT_SMTC */ - const int handler_len = &except_vec_vi_end - &except_vec_vi; - const int lui_offset = &except_vec_vi_lui - &except_vec_vi; - const int ori_offset = &except_vec_vi_ori - &except_vec_vi; + const int handler_len = &except_vec_vi_end - vec_start; + const int lui_offset = &except_vec_vi_lui - vec_start; + const int ori_offset = &except_vec_vi_ori - vec_start; if (handler_len > VECTORSPACING) { /* @@ -1272,7 +1280,7 @@ static void *set_vi_srs_handler(int n, vi_handler_t addr, int srs) panic("VECTORSPACING too small"); } - memcpy(b, &except_vec_vi, handler_len); + memcpy(b, vec_start, handler_len); #ifdef CONFIG_MIPS_MT_SMTC BUG_ON(n > 7); /* Vector index %d exceeds SMTC maximum. */ @@ -1554,6 +1562,10 @@ void __init trap_init(void) extern char except_vec3_generic, except_vec3_r4000; extern char except_vec4; unsigned long i; + int rollback; + + check_wait(); + rollback = (cpu_wait == r4k_wait); #if defined(CONFIG_KGDB) if (kgdb_early_setup) @@ -1618,7 +1630,7 @@ void __init trap_init(void) if (board_be_init) board_be_init(); - set_except_vector(0, handle_int); + set_except_vector(0, rollback ? rollback_handle_int : handle_int); set_except_vector(1, handle_tlbm); set_except_vector(2, handle_tlbl); set_except_vector(3, handle_tlbs); diff --git a/arch/mips/kernel/vmlinux.lds.S b/arch/mips/kernel/vmlinux.lds.S index b5470ce..afb119f 100644 --- a/arch/mips/kernel/vmlinux.lds.S +++ b/arch/mips/kernel/vmlinux.lds.S @@ -36,6 +36,7 @@ SECTIONS SCHED_TEXT LOCK_TEXT KPROBES_TEXT + *(.text.*) *(.fixup) *(.gnu.warning) } :text = 0 diff --git a/arch/mips/lib/csum_partial.S b/arch/mips/lib/csum_partial.S index 8d77841..edac989 100644 --- a/arch/mips/lib/csum_partial.S +++ b/arch/mips/lib/csum_partial.S @@ -39,12 +39,14 @@ #ifdef USE_DOUBLE #define LOAD ld +#define LOAD32 lwu #define ADD daddu #define NBYTES 8 #else #define LOAD lw +#define LOAD32 lw #define ADD addu #define NBYTES 4 @@ -60,6 +62,14 @@ ADD sum, v1; \ .set pop +#define ADDC32(sum,reg) \ + .set push; \ + .set noat; \ + addu sum, reg; \ + sltu v1, sum, reg; \ + addu sum, v1; \ + .set pop + #define CSUM_BIGCHUNK1(src, offset, sum, _t0, _t1, _t2, _t3) \ LOAD _t0, (offset + UNIT(0))(src); \ LOAD _t1, (offset + UNIT(1))(src); \ @@ -132,7 +142,7 @@ LEAF(csum_partial) beqz t8, .Lqword_align andi t8, src, 0x8 - lw t0, 0x00(src) + LOAD32 t0, 0x00(src) LONG_SUBU a1, a1, 0x4 ADDC(sum, t0) PTR_ADDU src, src, 0x4 @@ -211,7 +221,7 @@ LEAF(csum_partial) LONG_SRL t8, t8, 0x2 .Lend_words: - lw t0, (src) + LOAD32 t0, (src) LONG_SUBU t8, t8, 0x1 ADDC(sum, t0) .set reorder /* DADDI_WAR */ @@ -230,6 +240,9 @@ LEAF(csum_partial) /* Still a full word to go */ ulw t1, (src) PTR_ADDIU src, 4 +#ifdef USE_DOUBLE + dsll t1, t1, 32 /* clear lower 32bit */ +#endif ADDC(sum, t1) 1: move t1, zero @@ -280,7 +293,7 @@ LEAF(csum_partial) 1: .set reorder /* Add the passed partial csum. */ - ADDC(sum, a2) + ADDC32(sum, a2) jr ra .set noreorder END(csum_partial) @@ -681,7 +694,7 @@ EXC( sb t0, NBYTES-2(dst), .Ls_exc) .set pop 1: .set reorder - ADDC(sum, psum) + ADDC32(sum, psum) jr ra .set noreorder diff --git a/arch/mips/mti-malta/Makefile b/arch/mips/mti-malta/Makefile index 3b7dd72..cef2db8 100644 --- a/arch/mips/mti-malta/Makefile +++ b/arch/mips/mti-malta/Makefile @@ -15,6 +15,6 @@ obj-$(CONFIG_EARLY_PRINTK) += malta-console.o obj-$(CONFIG_PCI) += malta-pci.o # FIXME FIXME FIXME -obj-$(CONFIG_MIPS_MT_SMTC) += malta_smtc.o +obj-$(CONFIG_MIPS_MT_SMTC) += malta-smtc.o EXTRA_CFLAGS += -Werror diff --git a/arch/mips/mti-malta/malta-smtc.c b/arch/mips/mti-malta/malta-smtc.c index 5ea705e..f84a46a 100644 --- a/arch/mips/mti-malta/malta-smtc.c +++ b/arch/mips/mti-malta/malta-smtc.c @@ -84,12 +84,17 @@ static void msmtc_cpus_done(void) static void __init msmtc_smp_setup(void) { - mipsmt_build_cpu_map(0); + /* + * we won't get the definitive value until + * we've run smtc_prepare_cpus later, but + * we would appear to need an upper bound now. + */ + smp_num_siblings = smtc_build_cpu_map(0); } static void __init msmtc_prepare_cpus(unsigned int max_cpus) { - mipsmt_prepare_cpus(); + smtc_prepare_cpus(max_cpus); } struct plat_smp_ops msmtc_smp_ops = { diff --git a/arch/mips/pci/Makefile b/arch/mips/pci/Makefile index 15e01ae..c8c32f4 100644 --- a/arch/mips/pci/Makefile +++ b/arch/mips/pci/Makefile @@ -15,6 +15,7 @@ obj-$(CONFIG_SOC_TX3927) += ops-tx3927.o obj-$(CONFIG_PCI_VR41XX) += ops-vr41xx.o pci-vr41xx.o obj-$(CONFIG_MARKEINS) += ops-emma2rh.o pci-emma2rh.o fixup-emma2rh.o obj-$(CONFIG_PCI_TX4927) += ops-tx4927.o +obj-$(CONFIG_BCM47XX) += pci-bcm47xx.o # # These are still pretty much in the old state, watch, go blind. diff --git a/arch/mips/pci/pci-bcm47xx.c b/arch/mips/pci/pci-bcm47xx.c new file mode 100644 index 0000000..bea9b6c --- /dev/null +++ b/arch/mips/pci/pci-bcm47xx.c @@ -0,0 +1,60 @@ +/* + * Copyright (C) 2008 Aurelien Jarno <aurelien@aurel32.net> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN + * NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF + * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON + * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF + * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#include <linux/types.h> +#include <linux/pci.h> +#include <linux/ssb/ssb.h> + +int __init pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) +{ + return 0; +} + +int pcibios_plat_dev_init(struct pci_dev *dev) +{ + int res; + u8 slot, pin; + + res = ssb_pcibios_plat_dev_init(dev); + if (res < 0) { + printk(KERN_ALERT "PCI: Failed to init device %s\n", + pci_name(dev)); + return res; + } + + pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin); + slot = PCI_SLOT(dev->devfn); + res = ssb_pcibios_map_irq(dev, slot, pin); + + /* IRQ-0 and IRQ-1 are software interrupts. */ + if (res < 2) { + printk(KERN_ALERT "PCI: Failed to map IRQ of device %s\n", + pci_name(dev)); + return res; + } + + dev->irq = res; + return 0; +} + diff --git a/arch/mips/pci/pci-ip27.c b/arch/mips/pci/pci-ip27.c index bd78368..f97ab14 100644 --- a/arch/mips/pci/pci-ip27.c +++ b/arch/mips/pci/pci-ip27.c @@ -143,25 +143,47 @@ int __cpuinit bridge_probe(nasid_t nasid, int widget_id, int masterwid) */ int __devinit pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) { + return 0; +} + +/* Most MIPS systems have straight-forward swizzling needs. */ +static inline u8 bridge_swizzle(u8 pin, u8 slot) +{ + return (((pin - 1) + slot) % 4) + 1; +} + +static inline struct pci_dev *bridge_root_dev(struct pci_dev *dev) +{ + while (dev->bus->parent) { + /* Move up the chain of bridges. */ + dev = dev->bus->self; + } + + return dev; +} + +/* Do platform specific device initialization at pci_enable_device() time */ +int pcibios_plat_dev_init(struct pci_dev *dev) +{ struct bridge_controller *bc = BRIDGE_CONTROLLER(dev->bus); - int irq = bc->pci_int[slot]; + struct pci_dev *rdev = bridge_root_dev(dev); + int slot = PCI_SLOT(rdev->devfn); + int irq; + irq = bc->pci_int[slot]; if (irq == -1) { - irq = bc->pci_int[slot] = request_bridge_irq(bc); + irq = request_bridge_irq(bc); if (irq < 0) - panic("Can't allocate interrupt for PCI device %s\n", - pci_name(dev)); + return irq; + + bc->pci_int[slot] = irq; } irq_to_bridge[irq] = bc; irq_to_slot[irq] = slot; - return irq; -} + dev->irq = irq; -/* Do platform specific device initialization at pci_enable_device() time */ -int pcibios_plat_dev_init(struct pci_dev *dev) -{ return 0; } diff --git a/arch/mips/sibyte/swarm/Makefile b/arch/mips/sibyte/swarm/Makefile index f18ba92..7b45f19 100644 --- a/arch/mips/sibyte/swarm/Makefile +++ b/arch/mips/sibyte/swarm/Makefile @@ -1,3 +1,4 @@ -obj-y := setup.o rtc_xicor1241.o rtc_m41t81.o +obj-y := platform.o setup.o rtc_xicor1241.o \ + rtc_m41t81.o obj-$(CONFIG_I2C_BOARDINFO) += swarm-i2c.o diff --git a/arch/mips/sibyte/swarm/platform.c b/arch/mips/sibyte/swarm/platform.c new file mode 100644 index 0000000..54847fe --- /dev/null +++ b/arch/mips/sibyte/swarm/platform.c @@ -0,0 +1,85 @@ +#include <linux/err.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/io.h> +#include <linux/platform_device.h> +#include <linux/ata_platform.h> + +#include <asm/sibyte/board.h> +#include <asm/sibyte/sb1250_genbus.h> +#include <asm/sibyte/sb1250_regs.h> + +#if defined(CONFIG_SIBYTE_SWARM) || defined(CONFIG_SIBYTE_LITTLESUR) + +#define DRV_NAME "pata-swarm" + +#define SWARM_IDE_SHIFT 5 +#define SWARM_IDE_BASE 0x1f0 +#define SWARM_IDE_CTRL 0x3f6 + +static struct resource swarm_pata_resource[] = { + { + .name = "Swarm GenBus IDE", + .flags = IORESOURCE_MEM, + }, { + .name = "Swarm GenBus IDE", + .flags = IORESOURCE_MEM, + }, { + .name = "Swarm GenBus IDE", + .flags = IORESOURCE_IRQ, + .start = K_INT_GB_IDE, + .end = K_INT_GB_IDE, + }, +}; + +static struct pata_platform_info pata_platform_data = { + .ioport_shift = SWARM_IDE_SHIFT, +}; + +static struct platform_device swarm_pata_device = { + .name = "pata_platform", + .id = -1, + .resource = swarm_pata_resource, + .num_resources = ARRAY_SIZE(swarm_pata_resource), + .dev = { + .platform_data = &pata_platform_data, + .coherent_dma_mask = ~0, /* grumble */ + }, +}; + +static int __init swarm_pata_init(void) +{ + u8 __iomem *base; + phys_t offset, size; + struct resource *r; + + if (!SIBYTE_HAVE_IDE) + return -ENODEV; + + base = ioremap(A_IO_EXT_BASE, 0x800); + offset = __raw_readq(base + R_IO_EXT_REG(R_IO_EXT_START_ADDR, IDE_CS)); + size = __raw_readq(base + R_IO_EXT_REG(R_IO_EXT_MULT_SIZE, IDE_CS)); + iounmap(base); + + offset = G_IO_START_ADDR(offset) << S_IO_ADDRBASE; + size = (G_IO_MULT_SIZE(size) + 1) << S_IO_REGSIZE; + if (offset < A_PHYS_GENBUS || offset >= A_PHYS_GENBUS_END) { + pr_info(DRV_NAME ": PATA interface at GenBus disabled\n"); + + return -EBUSY; + } + + pr_info(DRV_NAME ": PATA interface at GenBus slot %i\n", IDE_CS); + + r = swarm_pata_resource; + r[0].start = offset + (SWARM_IDE_BASE << SWARM_IDE_SHIFT); + r[0].end = offset + ((SWARM_IDE_BASE + 8) << SWARM_IDE_SHIFT) - 1; + r[1].start = offset + (SWARM_IDE_CTRL << SWARM_IDE_SHIFT); + r[1].end = offset + ((SWARM_IDE_CTRL + 1) << SWARM_IDE_SHIFT) - 1; + + return platform_device_register(&swarm_pata_device); +} + +device_initcall(swarm_pata_init); + +#endif /* defined(CONFIG_SIBYTE_SWARM) || defined(CONFIG_SIBYTE_LITTLESUR) */ diff --git a/arch/mips/vr41xx/common/irq.c b/arch/mips/vr41xx/common/irq.c index cba36a2..92dd1a0 100644 --- a/arch/mips/vr41xx/common/irq.c +++ b/arch/mips/vr41xx/common/irq.c @@ -72,6 +72,7 @@ static void irq_dispatch(unsigned int irq) cascade = irq_cascade + irq; if (cascade->get_irq != NULL) { unsigned int source_irq = irq; + int ret; desc = irq_desc + source_irq; if (desc->chip->mask_ack) desc->chip->mask_ack(source_irq); @@ -79,8 +80,9 @@ static void irq_dispatch(unsigned int irq) desc->chip->mask(source_irq); desc->chip->ack(source_irq); } - irq = cascade->get_irq(irq); - if (irq < 0) + ret = cascade->get_irq(irq); + irq = ret; + if (ret < 0) atomic_inc(&irq_err_count); else irq_dispatch(irq); diff --git a/arch/mn10300/kernel/irq.c b/arch/mn10300/kernel/irq.c index 761c434..56c64cc 100644 --- a/arch/mn10300/kernel/irq.c +++ b/arch/mn10300/kernel/irq.c @@ -20,22 +20,8 @@ EXPORT_SYMBOL(__mn10300_irq_enabled_epsw); atomic_t irq_err_count; /* - * MN10300 INTC controller operations + * MN10300 interrupt controller operations */ -static void mn10300_cpupic_disable(unsigned int irq) -{ - u16 tmp = GxICR(irq); - GxICR(irq) = (tmp & GxICR_LEVEL) | GxICR_DETECT; - tmp = GxICR(irq); -} - -static void mn10300_cpupic_enable(unsigned int irq) -{ - u16 tmp = GxICR(irq); - GxICR(irq) = (tmp & GxICR_LEVEL) | GxICR_ENABLE; - tmp = GxICR(irq); -} - static void mn10300_cpupic_ack(unsigned int irq) { u16 tmp; @@ -60,26 +46,54 @@ static void mn10300_cpupic_mask_ack(unsigned int irq) static void mn10300_cpupic_unmask(unsigned int irq) { u16 tmp = GxICR(irq); - GxICR(irq) = (tmp & GxICR_LEVEL) | GxICR_ENABLE | GxICR_DETECT; + GxICR(irq) = (tmp & GxICR_LEVEL) | GxICR_ENABLE; tmp = GxICR(irq); } -static void mn10300_cpupic_end(unsigned int irq) +static void mn10300_cpupic_unmask_clear(unsigned int irq) { + /* the MN10300 PIC latches its interrupt request bit, even after the + * device has ceased to assert its interrupt line and the interrupt + * channel has been disabled in the PIC, so for level-triggered + * interrupts we need to clear the request bit when we re-enable */ u16 tmp = GxICR(irq); - GxICR(irq) = (tmp & GxICR_LEVEL) | GxICR_ENABLE; + GxICR(irq) = (tmp & GxICR_LEVEL) | GxICR_ENABLE | GxICR_DETECT; tmp = GxICR(irq); } -static struct irq_chip mn10300_cpu_pic = { - .name = "cpu", - .disable = mn10300_cpupic_disable, - .enable = mn10300_cpupic_enable, +/* + * MN10300 PIC level-triggered IRQ handling. + * + * The PIC has no 'ACK' function per se. It is possible to clear individual + * channel latches, but each latch relatches whether or not the channel is + * masked, so we need to clear the latch when we unmask the channel. + * + * Also for this reason, we don't supply an ack() op (it's unused anyway if + * mask_ack() is provided), and mask_ack() just masks. + */ +static struct irq_chip mn10300_cpu_pic_level = { + .name = "cpu_l", + .disable = mn10300_cpupic_mask, + .enable = mn10300_cpupic_unmask_clear, + .ack = NULL, + .mask = mn10300_cpupic_mask, + .mask_ack = mn10300_cpupic_mask, + .unmask = mn10300_cpupic_unmask_clear, +}; + +/* + * MN10300 PIC edge-triggered IRQ handling. + * + * We use the latch clearing function of the PIC as the 'ACK' function. + */ +static struct irq_chip mn10300_cpu_pic_edge = { + .name = "cpu_e", + .disable = mn10300_cpupic_mask, + .enable = mn10300_cpupic_unmask, .ack = mn10300_cpupic_ack, .mask = mn10300_cpupic_mask, .mask_ack = mn10300_cpupic_mask_ack, .unmask = mn10300_cpupic_unmask, - .end = mn10300_cpupic_end, }; /* @@ -114,7 +128,8 @@ void set_intr_level(int irq, u16 level) */ void set_intr_postackable(int irq) { - set_irq_handler(irq, handle_level_irq); + set_irq_chip_and_handler(irq, &mn10300_cpu_pic_level, + handle_level_irq); } /* @@ -126,8 +141,12 @@ void __init init_IRQ(void) for (irq = 0; irq < NR_IRQS; irq++) if (irq_desc[irq].chip == &no_irq_type) - set_irq_chip_and_handler(irq, &mn10300_cpu_pic, - handle_edge_irq); + /* due to the PIC latching interrupt requests, even + * when the IRQ is disabled, IRQ_PENDING is superfluous + * and we can use handle_level_irq() for edge-triggered + * interrupts */ + set_irq_chip_and_handler(irq, &mn10300_cpu_pic_edge, + handle_level_irq); unit_init_IRQ(); } diff --git a/arch/mn10300/kernel/time.c b/arch/mn10300/kernel/time.c index babb7c2..e460658 100644 --- a/arch/mn10300/kernel/time.c +++ b/arch/mn10300/kernel/time.c @@ -1,6 +1,6 @@ /* MN10300 Low level time management * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2007-2008 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * - Derived from arch/i386/kernel/time.c * @@ -16,6 +16,7 @@ #include <linux/init.h> #include <linux/smp.h> #include <linux/profile.h> +#include <linux/cnt32_to_63.h> #include <asm/irq.h> #include <asm/div64.h> #include <asm/processor.h> @@ -40,27 +41,54 @@ static struct irqaction timer_irq = { .name = "timer", }; +static unsigned long sched_clock_multiplier; + /* * scheduler clock - returns current time in nanosec units. */ unsigned long long sched_clock(void) { union { - unsigned long long l; - u32 w[2]; - } quot; + unsigned long long ll; + unsigned l[2]; + } tsc64, result; + unsigned long tsc, tmp; + unsigned product[3]; /* 96-bit intermediate value */ + + /* read the TSC value + */ + tsc = 0 - get_cycles(); /* get_cycles() counts down */ - quot.w[0] = mn10300_last_tsc - get_cycles(); - quot.w[1] = 1000000000; + /* expand to 64-bits. + * - sched_clock() must be called once a minute or better or the + * following will go horribly wrong - see cnt32_to_63() + */ + tsc64.ll = cnt32_to_63(tsc) & 0x7fffffffffffffffULL; - asm("mulu %2,%3,%0,%1" - : "=r"(quot.w[1]), "=r"(quot.w[0]) - : "0"(quot.w[1]), "1"(quot.w[0]) + /* scale the 64-bit TSC value to a nanosecond value via a 96-bit + * intermediate + */ + asm("mulu %2,%0,%3,%0 \n" /* LSW * mult -> 0:%3:%0 */ + "mulu %2,%1,%2,%1 \n" /* MSW * mult -> %2:%1:0 */ + "add %3,%1 \n" + "addc 0,%2 \n" /* result in %2:%1:%0 */ + : "=r"(product[0]), "=r"(product[1]), "=r"(product[2]), "=r"(tmp) + : "0"(tsc64.l[0]), "1"(tsc64.l[1]), "2"(sched_clock_multiplier) : "cc"); - do_div(quot.l, MN10300_TSCCLK); + result.l[0] = product[1] << 16 | product[0] >> 16; + result.l[1] = product[2] << 16 | product[1] >> 16; - return quot.l; + return result.ll; +} + +/* + * initialise the scheduler clock + */ +static void __init mn10300_sched_clock_init(void) +{ + sched_clock_multiplier = + __muldiv64u(NSEC_PER_SEC, 1 << 16, MN10300_TSCCLK); } /* @@ -128,4 +156,6 @@ void __init time_init(void) /* start the watchdog timer */ watchdog_go(); #endif + + mn10300_sched_clock_init(); } diff --git a/arch/mn10300/unit-asb2303/unit-init.c b/arch/mn10300/unit-asb2303/unit-init.c index 14b2c81..70e8cb4 100644 --- a/arch/mn10300/unit-asb2303/unit-init.c +++ b/arch/mn10300/unit-asb2303/unit-init.c @@ -51,7 +51,7 @@ void __init unit_init_IRQ(void) switch (GET_XIRQ_TRIGGER(extnum)) { case XIRQ_TRIGGER_HILEVEL: case XIRQ_TRIGGER_LOWLEVEL: - set_irq_handler(XIRQ2IRQ(extnum), handle_level_irq); + set_intr_postackable(XIRQ2IRQ(extnum)); break; default: break; diff --git a/arch/mn10300/unit-asb2305/unit-init.c b/arch/mn10300/unit-asb2305/unit-init.c index 6a35241..72812a9 100644 --- a/arch/mn10300/unit-asb2305/unit-init.c +++ b/arch/mn10300/unit-asb2305/unit-init.c @@ -52,7 +52,7 @@ void __init unit_init_IRQ(void) switch (GET_XIRQ_TRIGGER(extnum)) { case XIRQ_TRIGGER_HILEVEL: case XIRQ_TRIGGER_LOWLEVEL: - set_irq_handler(XIRQ2IRQ(extnum), handle_level_irq); + set_intr_postackable(XIRQ2IRQ(extnum)); break; default: break; diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile index 717a3bc..65d1a84 100644 --- a/arch/powerpc/boot/Makefile +++ b/arch/powerpc/boot/Makefile @@ -195,7 +195,7 @@ image-$(CONFIG_PPC_CELLEB) += zImage.pseries image-$(CONFIG_PPC_CHRP) += zImage.chrp image-$(CONFIG_PPC_EFIKA) += zImage.chrp image-$(CONFIG_PPC_PMAC) += zImage.pmac -image-$(CONFIG_PPC_HOLLY) += zImage.holly +image-$(CONFIG_PPC_HOLLY) += dtbImage.holly image-$(CONFIG_PPC_PRPMC2800) += dtbImage.prpmc2800 image-$(CONFIG_PPC_ISERIES) += zImage.iseries image-$(CONFIG_DEFAULT_UIMAGE) += uImage diff --git a/arch/powerpc/boot/dts/holly.dts b/arch/powerpc/boot/dts/holly.dts index f87fe7b..c6e11eb 100644 --- a/arch/powerpc/boot/dts/holly.dts +++ b/arch/powerpc/boot/dts/holly.dts @@ -133,61 +133,61 @@ reg = <0x00007400 0x00000400>; big-endian; }; + }; - pci@1000 { - device_type = "pci"; - compatible = "tsi109-pci", "tsi108-pci"; - #interrupt-cells = <1>; - #size-cells = <2>; - #address-cells = <3>; - reg = <0x00001000 0x00001000>; - bus-range = <0x0 0x0>; - /*----------------------------------------------------+ - | PCI memory range. - | 01 denotes I/O space - | 02 denotes 32-bit memory space - +----------------------------------------------------*/ - ranges = <0x02000000 0x00000000 0x40000000 0x40000000 0x00000000 0x10000000 - 0x01000000 0x00000000 0x00000000 0x7e000000 0x00000000 0x00010000>; - clock-frequency = <133333332>; - interrupt-parent = <&MPIC>; + pci@c0001000 { + device_type = "pci"; + compatible = "tsi109-pci", "tsi108-pci"; + #interrupt-cells = <1>; + #size-cells = <2>; + #address-cells = <3>; + reg = <0xc0001000 0x00001000>; + bus-range = <0x0 0x0>; + /*----------------------------------------------------+ + | PCI memory range. + | 01 denotes I/O space + | 02 denotes 32-bit memory space + +----------------------------------------------------*/ + ranges = <0x02000000 0x00000000 0x40000000 0x40000000 0x00000000 0x10000000 + 0x01000000 0x00000000 0x00000000 0x7e000000 0x00000000 0x00010000>; + clock-frequency = <133333332>; + interrupt-parent = <&MPIC>; + interrupts = <0x17 0x2>; + interrupt-map-mask = <0xf800 0x0 0x0 0x7>; + /*----------------------------------------------------+ + | The INTA, INTB, INTC, INTD are shared. + +----------------------------------------------------*/ + interrupt-map = < + 0x800 0x0 0x0 0x1 &RT0 0x24 0x0 + 0x800 0x0 0x0 0x2 &RT0 0x25 0x0 + 0x800 0x0 0x0 0x3 &RT0 0x26 0x0 + 0x800 0x0 0x0 0x4 &RT0 0x27 0x0 + + 0x1000 0x0 0x0 0x1 &RT0 0x25 0x0 + 0x1000 0x0 0x0 0x2 &RT0 0x26 0x0 + 0x1000 0x0 0x0 0x3 &RT0 0x27 0x0 + 0x1000 0x0 0x0 0x4 &RT0 0x24 0x0 + + 0x1800 0x0 0x0 0x1 &RT0 0x26 0x0 + 0x1800 0x0 0x0 0x2 &RT0 0x27 0x0 + 0x1800 0x0 0x0 0x3 &RT0 0x24 0x0 + 0x1800 0x0 0x0 0x4 &RT0 0x25 0x0 + + 0x2000 0x0 0x0 0x1 &RT0 0x27 0x0 + 0x2000 0x0 0x0 0x2 &RT0 0x24 0x0 + 0x2000 0x0 0x0 0x3 &RT0 0x25 0x0 + 0x2000 0x0 0x0 0x4 &RT0 0x26 0x0 + >; + + RT0: router@1180 { + device_type = "pic-router"; + interrupt-controller; + big-endian; + clock-frequency = <0>; + #address-cells = <0>; + #interrupt-cells = <2>; interrupts = <0x17 0x2>; - interrupt-map-mask = <0xf800 0x0 0x0 0x7>; - /*----------------------------------------------------+ - | The INTA, INTB, INTC, INTD are shared. - +----------------------------------------------------*/ - interrupt-map = < - 0x800 0x0 0x0 0x1 &RT0 0x24 0x0 - 0x800 0x0 0x0 0x2 &RT0 0x25 0x0 - 0x800 0x0 0x0 0x3 &RT0 0x26 0x0 - 0x800 0x0 0x0 0x4 &RT0 0x27 0x0 - - 0x1000 0x0 0x0 0x1 &RT0 0x25 0x0 - 0x1000 0x0 0x0 0x2 &RT0 0x26 0x0 - 0x1000 0x0 0x0 0x3 &RT0 0x27 0x0 - 0x1000 0x0 0x0 0x4 &RT0 0x24 0x0 - - 0x1800 0x0 0x0 0x1 &RT0 0x26 0x0 - 0x1800 0x0 0x0 0x2 &RT0 0x27 0x0 - 0x1800 0x0 0x0 0x3 &RT0 0x24 0x0 - 0x1800 0x0 0x0 0x4 &RT0 0x25 0x0 - - 0x2000 0x0 0x0 0x1 &RT0 0x27 0x0 - 0x2000 0x0 0x0 0x2 &RT0 0x24 0x0 - 0x2000 0x0 0x0 0x3 &RT0 0x25 0x0 - 0x2000 0x0 0x0 0x4 &RT0 0x26 0x0 - >; - - RT0: router@1180 { - device_type = "pic-router"; - interrupt-controller; - big-endian; - clock-frequency = <0>; - #address-cells = <0>; - #interrupt-cells = <2>; - interrupts = <0x17 0x2>; - interrupt-parent = <&MPIC>; - }; + interrupt-parent = <&MPIC>; }; }; diff --git a/arch/powerpc/boot/dts/mpc8610_hpcd.dts b/arch/powerpc/boot/dts/mpc8610_hpcd.dts index 3b3a106..584a4f1 100644 --- a/arch/powerpc/boot/dts/mpc8610_hpcd.dts +++ b/arch/powerpc/boot/dts/mpc8610_hpcd.dts @@ -281,7 +281,7 @@ cell-index = <0>; reg = <0x0 0x80>; interrupt-parent = <&mpic>; - interrupts = <60 2>; + interrupts = <76 2>; }; dma-channel@1 { compatible = "fsl,mpc8610-dma-channel", @@ -289,7 +289,7 @@ cell-index = <1>; reg = <0x80 0x80>; interrupt-parent = <&mpic>; - interrupts = <61 2>; + interrupts = <77 2>; }; dma-channel@2 { compatible = "fsl,mpc8610-dma-channel", @@ -297,7 +297,7 @@ cell-index = <2>; reg = <0x100 0x80>; interrupt-parent = <&mpic>; - interrupts = <62 2>; + interrupts = <78 2>; }; dma-channel@3 { compatible = "fsl,mpc8610-dma-channel", @@ -305,7 +305,7 @@ cell-index = <3>; reg = <0x180 0x80>; interrupt-parent = <&mpic>; - interrupts = <63 2>; + interrupts = <79 2>; }; }; diff --git a/arch/powerpc/include/asm/elf.h b/arch/powerpc/include/asm/elf.h index 80d1f39..64c6ee2 100644 --- a/arch/powerpc/include/asm/elf.h +++ b/arch/powerpc/include/asm/elf.h @@ -409,6 +409,13 @@ do { \ /* Keep this the last entry. */ #define R_PPC64_NUM 107 +/* There's actually a third entry here, but it's unused */ +struct ppc64_opd_entry +{ + unsigned long funcaddr; + unsigned long r2; +}; + #ifdef __KERNEL__ #ifdef CONFIG_SPU_BASE diff --git a/arch/powerpc/include/asm/sections.h b/arch/powerpc/include/asm/sections.h index 7710e9e..07956f3 100644 --- a/arch/powerpc/include/asm/sections.h +++ b/arch/powerpc/include/asm/sections.h @@ -2,6 +2,8 @@ #define _ASM_POWERPC_SECTIONS_H #ifdef __KERNEL__ +#include <linux/elf.h> +#include <linux/uaccess.h> #include <asm-generic/sections.h> #ifdef __powerpc64__ @@ -17,7 +19,15 @@ static inline int in_kernel_text(unsigned long addr) } #undef dereference_function_descriptor -void *dereference_function_descriptor(void *); +static inline void *dereference_function_descriptor(void *ptr) +{ + struct ppc64_opd_entry *desc = ptr; + void *p; + + if (!probe_kernel_address(&desc->funcaddr, p)) + ptr = p; + return ptr; +} #endif diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c index d308a9f..31982d0 100644 --- a/arch/powerpc/kernel/idle.c +++ b/arch/powerpc/kernel/idle.c @@ -34,11 +34,7 @@ #include <asm/smp.h> #ifdef CONFIG_HOTPLUG_CPU -/* this is used for software suspend, and that shuts down - * CPUs even while the system is still booting... */ -#define cpu_should_die() (cpu_is_offline(smp_processor_id()) && \ - (system_state == SYSTEM_RUNNING \ - || system_state == SYSTEM_BOOTING)) +#define cpu_should_die() cpu_is_offline(smp_processor_id()) #else #define cpu_should_die() 0 #endif diff --git a/arch/powerpc/kernel/kgdb.c b/arch/powerpc/kernel/kgdb.c index b4fdf2f..fe8f71d 100644 --- a/arch/powerpc/kernel/kgdb.c +++ b/arch/powerpc/kernel/kgdb.c @@ -347,9 +347,8 @@ int kgdb_arch_handle_exception(int vector, int signo, int err_code, linux_regs->msr |= MSR_SE; #endif kgdb_single_step = 1; - if (kgdb_contthread) - atomic_set(&kgdb_cpu_doing_single_step, - raw_smp_processor_id()); + atomic_set(&kgdb_cpu_doing_single_step, + raw_smp_processor_id()); } return 0; } diff --git a/arch/powerpc/kernel/module_64.c b/arch/powerpc/kernel/module_64.c index ad79de2..1af2377 100644 --- a/arch/powerpc/kernel/module_64.c +++ b/arch/powerpc/kernel/module_64.c @@ -21,9 +21,7 @@ #include <linux/err.h> #include <linux/vmalloc.h> #include <linux/bug.h> -#include <linux/uaccess.h> #include <asm/module.h> -#include <asm/sections.h> #include <asm/firmware.h> #include <asm/code-patching.h> #include <linux/sort.h> @@ -43,13 +41,6 @@ #define DEBUGP(fmt , ...) #endif -/* There's actually a third entry here, but it's unused */ -struct ppc64_opd_entry -{ - unsigned long funcaddr; - unsigned long r2; -}; - /* Like PPC32, we need little trampolines to do > 24-bit jumps (into the kernel itself). But on PPC64, these need to be used for every jump, actually, to reset r2 (TOC+0x8000). */ @@ -452,13 +443,3 @@ int apply_relocate_add(Elf64_Shdr *sechdrs, return 0; } - -void *dereference_function_descriptor(void *ptr) -{ - struct ppc64_opd_entry *desc = ptr; - void *p; - - if (!probe_kernel_address(&desc->funcaddr, p)) - ptr = p; - return ptr; -} diff --git a/arch/powerpc/platforms/fsl_uli1575.c b/arch/powerpc/platforms/fsl_uli1575.c index ef74a07..8c61996 100644 --- a/arch/powerpc/platforms/fsl_uli1575.c +++ b/arch/powerpc/platforms/fsl_uli1575.c @@ -219,11 +219,21 @@ static void __devinit quirk_final_uli5249(struct pci_dev *dev) int i; u8 *dummy; struct pci_bus *bus = dev->bus; + resource_size_t end = 0; + + for (i = PCI_BRIDGE_RESOURCES; i < PCI_BRIDGE_RESOURCES+3; i++) { + unsigned long flags = pci_resource_flags(dev, i); + if ((flags & (IORESOURCE_MEM|IORESOURCE_PREFETCH)) == IORESOURCE_MEM) + end = pci_resource_end(dev, i); + } for (i = 0; i < PCI_BUS_NUM_RESOURCES; i++) { if ((bus->resource[i]) && (bus->resource[i]->flags & IORESOURCE_MEM)) { - dummy = ioremap(bus->resource[i]->end - 3, 0x4); + if (bus->resource[i]->end == end) + dummy = ioremap(bus->resource[i]->start, 0x4); + else + dummy = ioremap(bus->resource[i]->end - 3, 0x4); if (dummy) { in_8(dummy); iounmap(dummy); diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index ca114fe..06acb1a 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -169,6 +169,8 @@ void init_cpu_timer(void) static void clock_comparator_interrupt(__u16 code) { + if (S390_lowcore.clock_comparator == -1ULL) + set_clock_comparator(S390_lowcore.clock_comparator); } static void etr_timing_alert(struct etr_irq_parm *); diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c index fc6ab60..0953cee 100644 --- a/arch/s390/lib/delay.c +++ b/arch/s390/lib/delay.c @@ -1,14 +1,9 @@ /* - * arch/s390/lib/delay.c * Precise Delay Loops for S390 * - * S390 version - * Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation - * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), - * - * Derived from "arch/i386/lib/delay.c" - * Copyright (C) 1993 Linus Torvalds - * Copyright (C) 1997 Martin Mares <mj@atrey.karlin.mff.cuni.cz> + * Copyright IBM Corp. 1999,2008 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Heiko Carstens <heiko.carstens@de.ibm.com>, */ #include <linux/sched.h> @@ -29,30 +24,31 @@ void __delay(unsigned long loops) asm volatile("0: brct %0,0b" : : "d" ((loops/2) + 1)); } -/* - * Waits for 'usecs' microseconds using the TOD clock comparator. - */ -void __udelay(unsigned long usecs) +static void __udelay_disabled(unsigned long usecs) { - u64 end, time, old_cc = 0; - unsigned long flags, cr0, mask, dummy; - int irq_context; + unsigned long mask, cr0, cr0_saved; + u64 clock_saved; - irq_context = in_interrupt(); - if (!irq_context) - local_bh_disable(); - local_irq_save(flags); - if (raw_irqs_disabled_flags(flags)) { - old_cc = local_tick_disable(); - S390_lowcore.clock_comparator = -1ULL; - __ctl_store(cr0, 0, 0); - dummy = (cr0 & 0xffff00e0) | 0x00000800; - __ctl_load(dummy , 0, 0); - mask = psw_kernel_bits | PSW_MASK_WAIT | PSW_MASK_EXT; - } else - mask = psw_kernel_bits | PSW_MASK_WAIT | - PSW_MASK_EXT | PSW_MASK_IO; + clock_saved = local_tick_disable(); + set_clock_comparator(get_clock() + ((u64) usecs << 12)); + __ctl_store(cr0_saved, 0, 0); + cr0 = (cr0_saved & 0xffff00e0) | 0x00000800; + __ctl_load(cr0 , 0, 0); + mask = psw_kernel_bits | PSW_MASK_WAIT | PSW_MASK_EXT; + trace_hardirqs_on(); + __load_psw_mask(mask); + local_irq_disable(); + __ctl_load(cr0_saved, 0, 0); + local_tick_enable(clock_saved); + set_clock_comparator(S390_lowcore.clock_comparator); +} +static void __udelay_enabled(unsigned long usecs) +{ + unsigned long mask; + u64 end, time; + + mask = psw_kernel_bits | PSW_MASK_WAIT | PSW_MASK_EXT | PSW_MASK_IO; end = get_clock() + ((u64) usecs << 12); do { time = end < S390_lowcore.clock_comparator ? @@ -62,13 +58,37 @@ void __udelay(unsigned long usecs) __load_psw_mask(mask); local_irq_disable(); } while (get_clock() < end); + set_clock_comparator(S390_lowcore.clock_comparator); +} - if (raw_irqs_disabled_flags(flags)) { - __ctl_load(cr0, 0, 0); - local_tick_enable(old_cc); +/* + * Waits for 'usecs' microseconds using the TOD clock comparator. + */ +void __udelay(unsigned long usecs) +{ + unsigned long flags; + + preempt_disable(); + local_irq_save(flags); + if (in_irq()) { + __udelay_disabled(usecs); + goto out; + } + if (in_softirq()) { + if (raw_irqs_disabled_flags(flags)) + __udelay_disabled(usecs); + else + __udelay_enabled(usecs); + goto out; } - if (!irq_context) + if (raw_irqs_disabled_flags(flags)) { + local_bh_disable(); + __udelay_disabled(usecs); _local_bh_enable(); - set_clock_comparator(S390_lowcore.clock_comparator); + goto out; + } + __udelay_enabled(usecs); +out: local_irq_restore(flags); + preempt_enable(); } diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c index 2396388..7495bc7 100644 --- a/arch/sparc64/kernel/irq.c +++ b/arch/sparc64/kernel/irq.c @@ -7,6 +7,7 @@ #include <linux/module.h> #include <linux/sched.h> +#include <linux/linkage.h> #include <linux/ptrace.h> #include <linux/errno.h> #include <linux/kernel_stat.h> @@ -866,7 +867,7 @@ static void kill_prom_timer(void) : "g1", "g2"); } -void init_irqwork_curcpu(void) +void notrace init_irqwork_curcpu(void) { int cpu = hard_smp_processor_id(); @@ -897,7 +898,7 @@ static void __cpuinit register_one_mondo(unsigned long paddr, unsigned long type } } -void __cpuinit sun4v_register_mondo_queues(int this_cpu) +void __cpuinit notrace sun4v_register_mondo_queues(int this_cpu) { struct trap_per_cpu *tb = &trap_block[this_cpu]; diff --git a/arch/sparc64/kernel/of_device.c b/arch/sparc64/kernel/of_device.c index f845f15..100ebd5 100644 --- a/arch/sparc64/kernel/of_device.c +++ b/arch/sparc64/kernel/of_device.c @@ -169,7 +169,7 @@ static unsigned long of_bus_default_get_flags(const u32 *addr, unsigned long fla static int of_bus_pci_match(struct device_node *np) { - if (!strcmp(np->type, "pci") || !strcmp(np->type, "pciex")) { + if (!strcmp(np->name, "pci")) { const char *model = of_get_property(np, "model", NULL); if (model && !strcmp(model, "SUNW,simba")) @@ -200,7 +200,7 @@ static int of_bus_simba_match(struct device_node *np) /* Treat PCI busses lacking ranges property just like * simba. */ - if (!strcmp(np->type, "pci") || !strcmp(np->type, "pciex")) { + if (!strcmp(np->name, "pci")) { if (!of_find_property(np, "ranges", NULL)) return 1; } @@ -429,7 +429,7 @@ static int __init use_1to1_mapping(struct device_node *pp) * it lacks a ranges property, and this will include * cases like Simba. */ - if (!strcmp(pp->type, "pci") || !strcmp(pp->type, "pciex")) + if (!strcmp(pp->name, "pci")) return 0; return 1; @@ -714,8 +714,7 @@ static unsigned int __init build_one_device_irq(struct of_device *op, break; } } else { - if (!strcmp(pp->type, "pci") || - !strcmp(pp->type, "pciex")) { + if (!strcmp(pp->name, "pci")) { unsigned int this_orig_irq = irq; irq = pci_irq_swizzle(dp, pp, irq); diff --git a/arch/sparc64/kernel/pci.c b/arch/sparc64/kernel/pci.c index 5509619..80dad76 100644 --- a/arch/sparc64/kernel/pci.c +++ b/arch/sparc64/kernel/pci.c @@ -425,7 +425,7 @@ struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, dev->current_state = 4; /* unknown power state */ dev->error_state = pci_channel_io_normal; - if (!strcmp(type, "pci") || !strcmp(type, "pciex")) { + if (!strcmp(node->name, "pci")) { /* a PCI-PCI bridge */ dev->hdr_type = PCI_HEADER_TYPE_BRIDGE; dev->rom_base_reg = PCI_ROM_ADDRESS1; diff --git a/arch/sparc64/kernel/pci_psycho.c b/arch/sparc64/kernel/pci_psycho.c index e205ade..f85b6be 100644 --- a/arch/sparc64/kernel/pci_psycho.c +++ b/arch/sparc64/kernel/pci_psycho.c @@ -575,7 +575,7 @@ static irqreturn_t psycho_pcierr_intr_other(struct pci_pbm_info *pbm, int is_pbm { unsigned long csr_reg, csr, csr_error_bits; irqreturn_t ret = IRQ_NONE; - u16 stat; + u16 stat, *addr; if (is_pbm_a) { csr_reg = pbm->controller_regs + PSYCHO_PCIA_CTRL; @@ -597,7 +597,9 @@ static irqreturn_t psycho_pcierr_intr_other(struct pci_pbm_info *pbm, int is_pbm printk("%s: PCI SERR signal asserted.\n", pbm->name); ret = IRQ_HANDLED; } - pci_read_config_word(pbm->pci_bus->self, PCI_STATUS, &stat); + addr = psycho_pci_config_mkaddr(pbm, pbm->pci_first_busno, + 0, PCI_STATUS); + pci_config_read16(addr, &stat); if (stat & (PCI_STATUS_PARITY | PCI_STATUS_SIG_TARGET_ABORT | PCI_STATUS_REC_TARGET_ABORT | @@ -605,7 +607,7 @@ static irqreturn_t psycho_pcierr_intr_other(struct pci_pbm_info *pbm, int is_pbm PCI_STATUS_SIG_SYSTEM_ERROR)) { printk("%s: PCI bus error, PCI_STATUS[%04x]\n", pbm->name, stat); - pci_write_config_word(pbm->pci_bus->self, PCI_STATUS, 0xffff); + pci_config_write16(addr, 0xffff); ret = IRQ_HANDLED; } return ret; diff --git a/arch/sparc64/kernel/traps.c b/arch/sparc64/kernel/traps.c index 3d92412..c824df1 100644 --- a/arch/sparc64/kernel/traps.c +++ b/arch/sparc64/kernel/traps.c @@ -10,6 +10,7 @@ #include <linux/module.h> #include <linux/sched.h> +#include <linux/linkage.h> #include <linux/kernel.h> #include <linux/signal.h> #include <linux/smp.h> @@ -2453,7 +2454,7 @@ struct trap_per_cpu trap_block[NR_CPUS]; /* This can get invoked before sched_init() so play it super safe * and use hard_smp_processor_id(). */ -void init_cur_cpu_trap(struct thread_info *t) +void notrace init_cur_cpu_trap(struct thread_info *t) { int cpu = hard_smp_processor_id(); struct trap_per_cpu *p = &trap_block[cpu]; diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 39fd3f4..0d7cdbb 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -29,6 +29,7 @@ config X86 select HAVE_FTRACE select HAVE_KVM if ((X86_32 && !X86_VOYAGER && !X86_VISWS && !X86_NUMAQ) || X86_64) select HAVE_ARCH_KGDB if !X86_VOYAGER + select HAVE_ARCH_TRACEHOOK select HAVE_GENERIC_DMA_COHERENT if X86_32 select HAVE_EFFICIENT_UNALIGNED_ACCESS @@ -1021,7 +1022,7 @@ config HAVE_ARCH_ALLOC_REMAP config ARCH_FLATMEM_ENABLE def_bool y - depends on X86_32 && ARCH_SELECT_MEMORY_MODEL && X86_PC && !NUMA + depends on X86_32 && ARCH_SELECT_MEMORY_MODEL && !NUMA config ARCH_DISCONTIGMEM_ENABLE def_bool y @@ -1037,7 +1038,7 @@ config ARCH_SPARSEMEM_DEFAULT config ARCH_SPARSEMEM_ENABLE def_bool y - depends on X86_64 || NUMA || (EXPERIMENTAL && X86_PC) + depends on X86_64 || NUMA || (EXPERIMENTAL && X86_PC) || X86_GENERICARCH select SPARSEMEM_STATIC if X86_32 select SPARSEMEM_VMEMMAP_ENABLE if X86_64 @@ -1118,10 +1119,10 @@ config MTRR You can safely say Y even if your machine doesn't have MTRRs, you'll just add about 9 KB to your kernel. - See <file:Documentation/mtrr.txt> for more information. + See <file:Documentation/x86/mtrr.txt> for more information. config MTRR_SANITIZER - bool + def_bool y prompt "MTRR cleanup support" depends on MTRR help @@ -1132,7 +1133,7 @@ config MTRR_SANITIZER The largest mtrr entry size for a continous block can be set with mtrr_chunk_size. - If unsure, say N. + If unsure, say Y. config MTRR_SANITIZER_ENABLE_DEFAULT int "MTRR cleanup enable value (0-1)" @@ -1192,7 +1193,6 @@ config IRQBALANCE config SECCOMP def_bool y prompt "Enable seccomp to safely compute untrusted bytecode" - depends on PROC_FS help This kernel feature is useful for number crunching applications that may need to compute untrusted bytecode during their @@ -1200,7 +1200,7 @@ config SECCOMP the process as file descriptors supporting the read/write syscalls, it's possible to isolate those applications in their own address space using seccomp. Once seccomp is - enabled via /proc/<pid>/seccomp, it cannot be disabled + enabled via prctl(PR_SET_SECCOMP), it cannot be disabled and the task is only allowed to execute a few safe syscalls defined by each seccomp mode. @@ -1357,14 +1357,14 @@ config PHYSICAL_ALIGN Don't change this unless you know what you are doing. config HOTPLUG_CPU - bool "Support for suspend on SMP and hot-pluggable CPUs (EXPERIMENTAL)" - depends on SMP && HOTPLUG && EXPERIMENTAL && !X86_VOYAGER + bool "Support for hot-pluggable CPUs" + depends on SMP && HOTPLUG && !X86_VOYAGER ---help--- - Say Y here to experiment with turning CPUs off and on, and to - enable suspend on SMP systems. CPUs can be controlled through - /sys/devices/system/cpu. - Say N if you want to disable CPU hotplug and don't need to - suspend. + Say Y here to allow turning CPUs off and on. CPUs can be + controlled through /sys/devices/system/cpu. + ( Note: power management support will enable this option + automatically on SMP systems. ) + Say N if you want to disable CPU hotplug. config COMPAT_VDSO def_bool y @@ -1379,6 +1379,51 @@ config COMPAT_VDSO If unsure, say Y. +config CMDLINE_BOOL + bool "Built-in kernel command line" + default n + help + Allow for specifying boot arguments to the kernel at + build time. On some systems (e.g. embedded ones), it is + necessary or convenient to provide some or all of the + kernel boot arguments with the kernel itself (that is, + to not rely on the boot loader to provide them.) + + To compile command line arguments into the kernel, + set this option to 'Y', then fill in the + the boot arguments in CONFIG_CMDLINE. + + Systems with fully functional boot loaders (i.e. non-embedded) + should leave this option set to 'N'. + +config CMDLINE + string "Built-in kernel command string" + depends on CMDLINE_BOOL + default "" + help + Enter arguments here that should be compiled into the kernel + image and used at boot time. If the boot loader provides a + command line at boot time, it is appended to this string to + form the full kernel command line, when the system boots. + + However, you can use the CONFIG_CMDLINE_OVERRIDE option to + change this behavior. + + In most cases, the command line (whether built-in or provided + by the boot loader) should specify the device for the root + file system. + +config CMDLINE_OVERRIDE + bool "Built-in command line overrides boot loader arguments" + default n + depends on CMDLINE_BOOL + help + Set this option to 'Y' to have the kernel ignore the boot loader + command line, and use ONLY the built-in command line. + + This is used to work around broken boot loaders. This should + be set to 'N' under normal conditions. + endmenu config ARCH_ENABLE_MEMORY_HOTPLUG @@ -1774,7 +1819,7 @@ config COMPAT_FOR_U64_ALIGNMENT config SYSVIPC_COMPAT def_bool y - depends on X86_64 && COMPAT && SYSVIPC + depends on COMPAT && SYSVIPC endmenu diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index b225219..60a8576 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -418,3 +418,21 @@ config X86_MINIMUM_CPU_FAMILY config X86_DEBUGCTLMSR def_bool y depends on !(MK6 || MWINCHIPC6 || MWINCHIP2 || MWINCHIP3D || MCYRIXIII || M586MMX || M586TSC || M586 || M486 || M386) + +config X86_DS + bool "Debug Store support" + default y + help + Add support for Debug Store. + This allows the kernel to provide a memory buffer to the hardware + to store various profiling and tracing events. + +config X86_PTRACE_BTS + bool "ptrace interface to Branch Trace Store" + default y + depends on (X86_DS && X86_DEBUGCTLMSR) + help + Add a ptrace interface to allow collecting an execution trace + of the traced task. + This collects control flow changes in a (cyclic) buffer and allows + debuggers to fill in the gaps and show an execution trace of the debuggee. diff --git a/arch/x86/boot/compressed/head_32.S b/arch/x86/boot/compressed/head_32.S index ba7736c..29c5fbf 100644 --- a/arch/x86/boot/compressed/head_32.S +++ b/arch/x86/boot/compressed/head_32.S @@ -137,14 +137,15 @@ relocated: */ movl output_len(%ebx), %eax pushl %eax + # push arguments for decompress_kernel: pushl %ebp # output address movl input_len(%ebx), %eax pushl %eax # input_len leal input_data(%ebx), %eax pushl %eax # input_data leal boot_heap(%ebx), %eax - pushl %eax # heap area as third argument - pushl %esi # real mode pointer as second arg + pushl %eax # heap area + pushl %esi # real mode pointer call decompress_kernel addl $20, %esp popl %ecx diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c index 9fea737..5780d36 100644 --- a/arch/x86/boot/compressed/misc.c +++ b/arch/x86/boot/compressed/misc.c @@ -16,7 +16,7 @@ */ #undef CONFIG_PARAVIRT #ifdef CONFIG_X86_32 -#define _ASM_DESC_H_ 1 +#define ASM_X86__DESC_H 1 #endif #ifdef CONFIG_X86_64 @@ -27,7 +27,7 @@ #include <linux/linkage.h> #include <linux/screen_info.h> #include <linux/elf.h> -#include <asm/io.h> +#include <linux/io.h> #include <asm/page.h> #include <asm/boot.h> #include <asm/bootparam.h> @@ -251,7 +251,7 @@ static void __putstr(int error, const char *s) y--; } } else { - vidmem [(x + cols * y) * 2] = c; + vidmem[(x + cols * y) * 2] = c; if (++x >= cols) { x = 0; if (++y >= lines) { @@ -277,7 +277,8 @@ static void *memset(void *s, int c, unsigned n) int i; char *ss = s; - for (i = 0; i < n; i++) ss[i] = c; + for (i = 0; i < n; i++) + ss[i] = c; return s; } @@ -287,7 +288,8 @@ static void *memcpy(void *dest, const void *src, unsigned n) const char *s = src; char *d = dest; - for (i = 0; i < n; i++) d[i] = s[i]; + for (i = 0; i < n; i++) + d[i] = s[i]; return dest; } diff --git a/arch/x86/boot/compressed/relocs.c b/arch/x86/boot/compressed/relocs.c index a1310c5..857e492 100644 --- a/arch/x86/boot/compressed/relocs.c +++ b/arch/x86/boot/compressed/relocs.c @@ -492,7 +492,7 @@ static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym)) continue; } sh_symtab = sec_symtab->symtab; - sym_strtab = sec->link->strtab; + sym_strtab = sec_symtab->link->strtab; for (j = 0; j < sec->shdr.sh_size/sizeof(Elf32_Rel); j++) { Elf32_Rel *rel; Elf32_Sym *sym; diff --git a/arch/x86/boot/header.S b/arch/x86/boot/header.S index af86e43..b993062 100644 --- a/arch/x86/boot/header.S +++ b/arch/x86/boot/header.S @@ -30,7 +30,6 @@ SYSSEG = DEF_SYSSEG /* system loaded at 0x10000 (65536) */ SYSSIZE = DEF_SYSSIZE /* system size: # of 16-byte clicks */ /* to be loaded */ ROOT_DEV = 0 /* ROOT_DEV is now written by "build" */ -SWAP_DEV = 0 /* SWAP_DEV is now written by "build" */ #ifndef SVGA_MODE #define SVGA_MODE ASK_VGA diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig index 104275e..ef9a520 100644 --- a/arch/x86/configs/i386_defconfig +++ b/arch/x86/configs/i386_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.27-rc4 -# Mon Aug 25 15:04:00 2008 +# Linux kernel version: 2.6.27-rc5 +# Wed Sep 3 17:23:09 2008 # # CONFIG_64BIT is not set CONFIG_X86_32=y @@ -202,7 +202,7 @@ CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y # CONFIG_M586 is not set # CONFIG_M586TSC is not set # CONFIG_M586MMX is not set -# CONFIG_M686 is not set +CONFIG_M686=y # CONFIG_MPENTIUMII is not set # CONFIG_MPENTIUMIII is not set # CONFIG_MPENTIUMM is not set @@ -221,13 +221,14 @@ CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y # CONFIG_MVIAC3_2 is not set # CONFIG_MVIAC7 is not set # CONFIG_MPSC is not set -CONFIG_MCORE2=y +# CONFIG_MCORE2 is not set # CONFIG_GENERIC_CPU is not set CONFIG_X86_GENERIC=y CONFIG_X86_CPU=y CONFIG_X86_CMPXCHG=y CONFIG_X86_L1_CACHE_SHIFT=7 CONFIG_X86_XADD=y +# CONFIG_X86_PPRO_FENCE is not set CONFIG_X86_WP_WORKS_OK=y CONFIG_X86_INVLPG=y CONFIG_X86_BSWAP=y @@ -235,14 +236,15 @@ CONFIG_X86_POPAD_OK=y CONFIG_X86_INTEL_USERCOPY=y CONFIG_X86_USE_PPRO_CHECKSUM=y CONFIG_X86_TSC=y +CONFIG_X86_CMOV=y CONFIG_X86_MINIMUM_CPU_FAMILY=4 CONFIG_X86_DEBUGCTLMSR=y CONFIG_HPET_TIMER=y CONFIG_HPET_EMULATE_RTC=y CONFIG_DMI=y # CONFIG_IOMMU_HELPER is not set -CONFIG_NR_CPUS=4 -# CONFIG_SCHED_SMT is not set +CONFIG_NR_CPUS=64 +CONFIG_SCHED_SMT=y CONFIG_SCHED_MC=y # CONFIG_PREEMPT_NONE is not set CONFIG_PREEMPT_VOLUNTARY=y @@ -254,7 +256,8 @@ CONFIG_VM86=y # CONFIG_TOSHIBA is not set # CONFIG_I8K is not set CONFIG_X86_REBOOTFIXUPS=y -# CONFIG_MICROCODE is not set +CONFIG_MICROCODE=y +CONFIG_MICROCODE_OLD_INTERFACE=y CONFIG_X86_MSR=y CONFIG_X86_CPUID=y # CONFIG_NOHIGHMEM is not set @@ -2115,7 +2118,7 @@ CONFIG_IO_DELAY_0X80=y CONFIG_DEFAULT_IO_DELAY_TYPE=0 CONFIG_DEBUG_BOOT_PARAMS=y # CONFIG_CPA_DEBUG is not set -# CONFIG_OPTIMIZE_INLINING is not set +CONFIG_OPTIMIZE_INLINING=y # # Security options diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig index 678c8ac..e620ea6 100644 --- a/arch/x86/configs/x86_64_defconfig +++ b/arch/x86/configs/x86_64_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.27-rc4 -# Mon Aug 25 14:40:46 2008 +# Linux kernel version: 2.6.27-rc5 +# Wed Sep 3 17:13:39 2008 # CONFIG_64BIT=y # CONFIG_X86_32 is not set @@ -218,17 +218,14 @@ CONFIG_X86_PC=y # CONFIG_MVIAC3_2 is not set # CONFIG_MVIAC7 is not set # CONFIG_MPSC is not set -CONFIG_MCORE2=y -# CONFIG_GENERIC_CPU is not set +# CONFIG_MCORE2 is not set +CONFIG_GENERIC_CPU=y CONFIG_X86_CPU=y -CONFIG_X86_L1_CACHE_BYTES=64 -CONFIG_X86_INTERNODE_CACHE_BYTES=64 +CONFIG_X86_L1_CACHE_BYTES=128 +CONFIG_X86_INTERNODE_CACHE_BYTES=128 CONFIG_X86_CMPXCHG=y -CONFIG_X86_L1_CACHE_SHIFT=6 +CONFIG_X86_L1_CACHE_SHIFT=7 CONFIG_X86_WP_WORKS_OK=y -CONFIG_X86_INTEL_USERCOPY=y -CONFIG_X86_USE_PPRO_CHECKSUM=y -CONFIG_X86_P6_NOP=y CONFIG_X86_TSC=y CONFIG_X86_CMPXCHG64=y CONFIG_X86_CMOV=y @@ -243,9 +240,8 @@ CONFIG_CALGARY_IOMMU_ENABLED_BY_DEFAULT=y CONFIG_AMD_IOMMU=y CONFIG_SWIOTLB=y CONFIG_IOMMU_HELPER=y -# CONFIG_MAXSMP is not set -CONFIG_NR_CPUS=4 -# CONFIG_SCHED_SMT is not set +CONFIG_NR_CPUS=64 +CONFIG_SCHED_SMT=y CONFIG_SCHED_MC=y # CONFIG_PREEMPT_NONE is not set CONFIG_PREEMPT_VOLUNTARY=y @@ -254,7 +250,8 @@ CONFIG_X86_LOCAL_APIC=y CONFIG_X86_IO_APIC=y # CONFIG_X86_MCE is not set # CONFIG_I8K is not set -# CONFIG_MICROCODE is not set +CONFIG_MICROCODE=y +CONFIG_MICROCODE_OLD_INTERFACE=y CONFIG_X86_MSR=y CONFIG_X86_CPUID=y CONFIG_NUMA=y @@ -290,7 +287,7 @@ CONFIG_BOUNCE=y CONFIG_VIRT_TO_BUS=y CONFIG_MTRR=y # CONFIG_MTRR_SANITIZER is not set -# CONFIG_X86_PAT is not set +CONFIG_X86_PAT=y CONFIG_EFI=y CONFIG_SECCOMP=y # CONFIG_HZ_100 is not set @@ -2089,7 +2086,7 @@ CONFIG_IO_DELAY_0X80=y CONFIG_DEFAULT_IO_DELAY_TYPE=0 CONFIG_DEBUG_BOOT_PARAMS=y # CONFIG_CPA_DEBUG is not set -# CONFIG_OPTIMIZE_INLINING is not set +CONFIG_OPTIMIZE_INLINING=y # # Security options diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c index a0e1dbe..127ec3f 100644 --- a/arch/x86/ia32/ia32_aout.c +++ b/arch/x86/ia32/ia32_aout.c @@ -85,8 +85,10 @@ static void dump_thread32(struct pt_regs *regs, struct user32 *dump) dump->regs.ax = regs->ax; dump->regs.ds = current->thread.ds; dump->regs.es = current->thread.es; - asm("movl %%fs,%0" : "=r" (fs)); dump->regs.fs = fs; - asm("movl %%gs,%0" : "=r" (gs)); dump->regs.gs = gs; + savesegment(fs, fs); + dump->regs.fs = fs; + savesegment(gs, gs); + dump->regs.gs = gs; dump->regs.orig_ax = regs->orig_ax; dump->regs.ip = regs->ip; dump->regs.cs = regs->cs; @@ -430,8 +432,9 @@ beyond_if: current->mm->start_stack = (unsigned long)create_aout_tables((char __user *)bprm->p, bprm); /* start thread */ - asm volatile("movl %0,%%fs" :: "r" (0)); \ - asm volatile("movl %0,%%es; movl %0,%%ds": :"r" (__USER32_DS)); + loadsegment(fs, 0); + loadsegment(ds, __USER32_DS); + loadsegment(es, __USER32_DS); load_gs_index(0); (regs)->ip = ex.a_entry; (regs)->sp = current->mm->start_stack; diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c index 20af4c7..f1a2ac7 100644 --- a/arch/x86/ia32/ia32_signal.c +++ b/arch/x86/ia32/ia32_signal.c @@ -206,7 +206,7 @@ struct rt_sigframe { unsigned int cur; \ unsigned short pre; \ err |= __get_user(pre, &sc->seg); \ - asm volatile("movl %%" #seg ",%0" : "=r" (cur)); \ + savesegment(seg, cur); \ pre |= mask; \ if (pre != cur) loadsegment(seg, pre); } @@ -235,7 +235,7 @@ static int ia32_restore_sigcontext(struct pt_regs *regs, */ err |= __get_user(gs, &sc->gs); gs |= 3; - asm("movl %%gs,%0" : "=r" (oldgs)); + savesegment(gs, oldgs); if (gs != oldgs) load_gs_index(gs); @@ -355,14 +355,13 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc, { int tmp, err = 0; - tmp = 0; - __asm__("movl %%gs,%0" : "=r"(tmp): "0"(tmp)); + savesegment(gs, tmp); err |= __put_user(tmp, (unsigned int __user *)&sc->gs); - __asm__("movl %%fs,%0" : "=r"(tmp): "0"(tmp)); + savesegment(fs, tmp); err |= __put_user(tmp, (unsigned int __user *)&sc->fs); - __asm__("movl %%ds,%0" : "=r"(tmp): "0"(tmp)); + savesegment(ds, tmp); err |= __put_user(tmp, (unsigned int __user *)&sc->ds); - __asm__("movl %%es,%0" : "=r"(tmp): "0"(tmp)); + savesegment(es, tmp); err |= __put_user(tmp, (unsigned int __user *)&sc->es); err |= __put_user((u32)regs->di, &sc->di); @@ -498,8 +497,8 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka, regs->dx = 0; regs->cx = 0; - asm volatile("movl %0,%%ds" :: "r" (__USER32_DS)); - asm volatile("movl %0,%%es" :: "r" (__USER32_DS)); + loadsegment(ds, __USER32_DS); + loadsegment(es, __USER32_DS); regs->cs = __USER32_CS; regs->ss = __USER32_DS; @@ -591,8 +590,8 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, regs->dx = (unsigned long) &frame->info; regs->cx = (unsigned long) &frame->uc; - asm volatile("movl %0,%%ds" :: "r" (__USER32_DS)); - asm volatile("movl %0,%%es" :: "r" (__USER32_DS)); + loadsegment(ds, __USER32_DS); + loadsegment(es, __USER32_DS); regs->cs = __USER32_CS; regs->ss = __USER32_DS; diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c index d3c6408..beda423 100644 --- a/arch/x86/ia32/sys_ia32.c +++ b/arch/x86/ia32/sys_ia32.c @@ -556,15 +556,6 @@ asmlinkage long sys32_rt_sigqueueinfo(int pid, int sig, return ret; } -/* These are here just in case some old ia32 binary calls it. */ -asmlinkage long sys32_pause(void) -{ - current->state = TASK_INTERRUPTIBLE; - schedule(); - return -ERESTARTNOHAND; -} - - #ifdef CONFIG_SYSCTL_SYSCALL struct sysctl_ia32 { unsigned int name; diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index bfd10fd..7d40ef7 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -58,7 +58,6 @@ EXPORT_SYMBOL(acpi_disabled); #ifdef CONFIG_X86_64 #include <asm/proto.h> -#include <asm/genapic.h> #else /* X86 */ @@ -97,8 +96,6 @@ static u64 acpi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE; #warning ACPI uses CMPXCHG, i486 and later hardware #endif -static int acpi_mcfg_64bit_base_addr __initdata = FALSE; - /* -------------------------------------------------------------------------- Boot-time Configuration -------------------------------------------------------------------------- */ @@ -160,6 +157,8 @@ char *__init __acpi_map_table(unsigned long phys, unsigned long size) struct acpi_mcfg_allocation *pci_mmcfg_config; int pci_mmcfg_config_num; +static int acpi_mcfg_64bit_base_addr __initdata = FALSE; + static int __init acpi_mcfg_oem_check(struct acpi_table_mcfg *mcfg) { if (!strcmp(mcfg->header.oem_id, "SGI")) @@ -1605,6 +1604,14 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = { */ { .callback = dmi_ignore_irq0_timer_override, + .ident = "HP nx6115 laptop", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq nx6115"), + }, + }, + { + .callback = dmi_ignore_irq0_timer_override, .ident = "HP NX6125 laptop", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), @@ -1619,6 +1626,14 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq nx6325"), }, }, + { + .callback = dmi_ignore_irq0_timer_override, + .ident = "HP 6715b laptop", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq 6715b"), + }, + }, {} }; diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 65a0c1b..fb04e49 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -231,25 +231,25 @@ static void alternatives_smp_lock(u8 **start, u8 **end, u8 *text, u8 *text_end) continue; if (*ptr > text_end) continue; - text_poke(*ptr, ((unsigned char []){0xf0}), 1); /* add lock prefix */ + /* turn DS segment override prefix into lock prefix */ + text_poke(*ptr, ((unsigned char []){0xf0}), 1); }; } static void alternatives_smp_unlock(u8 **start, u8 **end, u8 *text, u8 *text_end) { u8 **ptr; - char insn[1]; if (noreplace_smp) return; - add_nops(insn, 1); for (ptr = start; ptr < end; ptr++) { if (*ptr < text) continue; if (*ptr > text_end) continue; - text_poke(*ptr, insn, 1); + /* turn lock prefix into DS segment override prefix */ + text_poke(*ptr, ((unsigned char []){0x3E}), 1); }; } diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 3b346c6..34e4d11 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -201,10 +201,10 @@ static int iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd) */ static int iommu_completion_wait(struct amd_iommu *iommu) { - int ret, ready = 0; + int ret = 0, ready = 0; unsigned status = 0; struct iommu_cmd cmd; - unsigned long i = 0; + unsigned long flags, i = 0; memset(&cmd, 0, sizeof(cmd)); cmd.data[0] = CMD_COMPL_WAIT_INT_MASK; @@ -212,10 +212,12 @@ static int iommu_completion_wait(struct amd_iommu *iommu) iommu->need_sync = 0; - ret = iommu_queue_command(iommu, &cmd); + spin_lock_irqsave(&iommu->lock, flags); + + ret = __iommu_queue_command(iommu, &cmd); if (ret) - return ret; + goto out; while (!ready && (i < EXIT_LOOP_COUNT)) { ++i; @@ -230,6 +232,8 @@ static int iommu_completion_wait(struct amd_iommu *iommu) if (unlikely((i == EXIT_LOOP_COUNT) && printk_ratelimit())) printk(KERN_WARNING "AMD IOMMU: Completion wait loop failed\n"); +out: + spin_unlock_irqrestore(&iommu->lock, flags); return 0; } @@ -240,6 +244,7 @@ static int iommu_completion_wait(struct amd_iommu *iommu) static int iommu_queue_inv_dev_entry(struct amd_iommu *iommu, u16 devid) { struct iommu_cmd cmd; + int ret; BUG_ON(iommu == NULL); @@ -247,9 +252,11 @@ static int iommu_queue_inv_dev_entry(struct amd_iommu *iommu, u16 devid) CMD_SET_TYPE(&cmd, CMD_INV_DEV_ENTRY); cmd.data[0] = devid; + ret = iommu_queue_command(iommu, &cmd); + iommu->need_sync = 1; - return iommu_queue_command(iommu, &cmd); + return ret; } /* @@ -259,6 +266,7 @@ static int iommu_queue_inv_iommu_pages(struct amd_iommu *iommu, u64 address, u16 domid, int pde, int s) { struct iommu_cmd cmd; + int ret; memset(&cmd, 0, sizeof(cmd)); address &= PAGE_MASK; @@ -271,9 +279,11 @@ static int iommu_queue_inv_iommu_pages(struct amd_iommu *iommu, if (pde) /* PDE bit - we wan't flush everything not only the PTEs */ cmd.data[2] |= CMD_INV_IOMMU_PAGES_PDE_MASK; + ret = iommu_queue_command(iommu, &cmd); + iommu->need_sync = 1; - return iommu_queue_command(iommu, &cmd); + return ret; } /* diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 44e2182..9a32b37 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -455,11 +455,11 @@ out: force_iommu || valid_agp || fallback_aper_force) { - printk(KERN_ERR + printk(KERN_INFO "Your BIOS doesn't leave a aperture memory hole\n"); - printk(KERN_ERR + printk(KERN_INFO "Please enable the IOMMU option in the BIOS setup\n"); - printk(KERN_ERR + printk(KERN_INFO "This costs you %d MB of RAM\n", 32 << fallback_aper_order); diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 9ee24e6..5145a6e 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -228,12 +228,12 @@ #include <linux/suspend.h> #include <linux/kthread.h> #include <linux/jiffies.h> -#include <linux/smp_lock.h> #include <asm/system.h> #include <asm/uaccess.h> #include <asm/desc.h> #include <asm/i8253.h> +#include <asm/olpc.h> #include <asm/paravirt.h> #include <asm/reboot.h> @@ -2217,7 +2217,7 @@ static int __init apm_init(void) dmi_check_system(apm_dmi_table); - if (apm_info.bios.version == 0 || paravirt_enabled()) { + if (apm_info.bios.version == 0 || paravirt_enabled() || machine_is_olpc()) { printk(KERN_INFO "apm: BIOS not found.\n"); return -ENODEV; } diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c index aa89387..505543a 100644 --- a/arch/x86/kernel/asm-offsets_64.c +++ b/arch/x86/kernel/asm-offsets_64.c @@ -22,7 +22,7 @@ #define __NO_STUBS 1 #undef __SYSCALL -#undef _ASM_X86_64_UNISTD_H_ +#undef ASM_X86__UNISTD_64_H #define __SYSCALL(nr, sym) [nr] = 1, static char syscalls[] = { #include <asm/unistd.h> diff --git a/arch/x86/kernel/bios_uv.c b/arch/x86/kernel/bios_uv.c index c639bd5..fdd585f 100644 --- a/arch/x86/kernel/bios_uv.c +++ b/arch/x86/kernel/bios_uv.c @@ -25,11 +25,11 @@ x86_bios_strerror(long status) { const char *str; switch (status) { - case 0: str = "Call completed without error"; break; - case -1: str = "Not implemented"; break; - case -2: str = "Invalid argument"; break; - case -3: str = "Call completed with error"; break; - default: str = "Unknown BIOS status code"; break; + case 0: str = "Call completed without error"; break; + case -1: str = "Not implemented"; break; + case -2: str = "Invalid argument"; break; + case -3: str = "Call completed with error"; break; + default: str = "Unknown BIOS status code"; break; } return str; } diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 8aab851..4e456bd 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -344,31 +344,15 @@ static void __init early_cpu_detect(void) /* * The NOPL instruction is supposed to exist on all CPUs with - * family >= 6, unfortunately, that's not true in practice because + * family >= 6; unfortunately, that's not true in practice because * of early VIA chips and (more importantly) broken virtualizers that - * are not easy to detect. Hence, probe for it based on first - * principles. + * are not easy to detect. In the latter case it doesn't even *fail* + * reliably, so probing for it doesn't even work. Disable it completely + * unless we can find a reliable way to detect all the broken cases. */ static void __cpuinit detect_nopl(struct cpuinfo_x86 *c) { - const u32 nopl_signature = 0x888c53b1; /* Random number */ - u32 has_nopl = nopl_signature; - clear_cpu_cap(c, X86_FEATURE_NOPL); - if (c->x86 >= 6) { - asm volatile("\n" - "1: .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */ - "2:\n" - " .section .fixup,\"ax\"\n" - "3: xor %0,%0\n" - " jmp 2b\n" - " .previous\n" - _ASM_EXTABLE(1b,3b) - : "+a" (has_nopl)); - - if (has_nopl == nopl_signature) - set_cpu_cap(c, X86_FEATURE_NOPL); - } } static void __cpuinit generic_identify(struct cpuinfo_x86 *c) diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c index a11f5d4..305b465 100644 --- a/arch/x86/kernel/cpu/common_64.c +++ b/arch/x86/kernel/cpu/common_64.c @@ -430,6 +430,49 @@ static __init int setup_noclflush(char *arg) } __setup("noclflush", setup_noclflush); +struct msr_range { + unsigned min; + unsigned max; +}; + +static struct msr_range msr_range_array[] __cpuinitdata = { + { 0x00000000, 0x00000418}, + { 0xc0000000, 0xc000040b}, + { 0xc0010000, 0xc0010142}, + { 0xc0011000, 0xc001103b}, +}; + +static void __cpuinit print_cpu_msr(void) +{ + unsigned index; + u64 val; + int i; + unsigned index_min, index_max; + + for (i = 0; i < ARRAY_SIZE(msr_range_array); i++) { + index_min = msr_range_array[i].min; + index_max = msr_range_array[i].max; + for (index = index_min; index < index_max; index++) { + if (rdmsrl_amd_safe(index, &val)) + continue; + printk(KERN_INFO " MSR%08x: %016llx\n", index, val); + } + } +} + +static int show_msr __cpuinitdata; +static __init int setup_show_msr(char *arg) +{ + int num; + + get_option(&arg, &num); + + if (num > 0) + show_msr = num; + return 1; +} +__setup("show_msr=", setup_show_msr); + void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) { if (c->x86_model_id[0]) @@ -439,6 +482,14 @@ void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) printk(KERN_CONT " stepping %02x\n", c->x86_mask); else printk(KERN_CONT "\n"); + +#ifdef CONFIG_SMP + if (c->cpu_index < show_msr) + print_cpu_msr(); +#else + if (show_msr) + print_cpu_msr(); +#endif } static __init int setup_disablecpuid(char *arg) diff --git a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c index f1685fb..b8e05ee 100644 --- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c +++ b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c @@ -171,7 +171,7 @@ static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c) } if (c->x86 != 0xF) { - printk(KERN_WARNING PFX "Unknown p4-clockmod-capable CPU. Please send an e-mail to <cpufreq@lists.linux.org.uk>\n"); + printk(KERN_WARNING PFX "Unknown p4-clockmod-capable CPU. Please send an e-mail to <cpufreq@vger.kernel.org>\n"); return 0; } diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c index 15e13c0..3b5f064 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c @@ -26,7 +26,7 @@ #include <asm/cpufeature.h> #define PFX "speedstep-centrino: " -#define MAINTAINER "cpufreq@lists.linux.org.uk" +#define MAINTAINER "cpufreq@vger.kernel.org" #define dprintk(msg...) \ cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "speedstep-centrino", msg) diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index b75f256..f113ef4 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -222,10 +222,11 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_BTS); if (!(l1 & (1<<12))) set_cpu_cap(c, X86_FEATURE_PEBS); + ds_init_intel(c); } if (cpu_has_bts) - ds_init_intel(c); + ptrace_bts_init_intel(c); /* * See if we have a good local APIC by checking for buggy Pentia, diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index cb7d3b6..4e8d77f 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -401,12 +401,7 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base, tmp |= ~((1<<(hi - 1)) - 1); if (tmp != mask_lo) { - static int once = 1; - - if (once) { - printk(KERN_INFO "mtrr: your BIOS has set up an incorrect mask, fixing it up.\n"); - once = 0; - } + WARN_ONCE(1, KERN_INFO "mtrr: your BIOS has set up an incorrect mask, fixing it up.\n"); mask_lo = tmp; } } diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c index 84c480b..4c42146 100644 --- a/arch/x86/kernel/cpu/mtrr/if.c +++ b/arch/x86/kernel/cpu/mtrr/if.c @@ -405,9 +405,9 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset) } /* RED-PEN: base can be > 32bit */ len += seq_printf(seq, - "reg%02i: base=0x%05lx000 (%4luMB), size=%4lu%cB: %s, count=%d\n", + "reg%02i: base=0x%06lx000 (%5luMB), size=%5lu%cB, count=%d: %s\n", i, base, base >> (20 - PAGE_SHIFT), size, factor, - mtrr_attrib_to_str(type), mtrr_usage_table[i]); + mtrr_usage_table[i], mtrr_attrib_to_str(type)); } } return 0; diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index b117d7f..c78c048 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -729,7 +729,7 @@ struct var_mtrr_range_state { mtrr_type type; }; -struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; +static struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; static int __initdata debug_print; static int __init @@ -759,7 +759,8 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range, /* take out UC ranges */ for (i = 0; i < num_var_ranges; i++) { type = range_state[i].type; - if (type != MTRR_TYPE_UNCACHABLE) + if (type != MTRR_TYPE_UNCACHABLE && + type != MTRR_TYPE_WRPROT) continue; size = range_state[i].size_pfn; if (!size) @@ -834,7 +835,14 @@ static int __init enable_mtrr_cleanup_setup(char *str) enable_mtrr_cleanup = 1; return 0; } -early_param("enble_mtrr_cleanup", enable_mtrr_cleanup_setup); +early_param("enable_mtrr_cleanup", enable_mtrr_cleanup_setup); + +static int __init mtrr_cleanup_debug_setup(char *str) +{ + debug_print = 1; + return 0; +} +early_param("mtrr_cleanup_debug", mtrr_cleanup_debug_setup); struct var_mtrr_state { unsigned long range_startk; @@ -898,6 +906,27 @@ set_var_mtrr_all(unsigned int address_bits) } } +static unsigned long to_size_factor(unsigned long sizek, char *factorp) +{ + char factor; + unsigned long base = sizek; + + if (base & ((1<<10) - 1)) { + /* not MB alignment */ + factor = 'K'; + } else if (base & ((1<<20) - 1)){ + factor = 'M'; + base >>= 10; + } else { + factor = 'G'; + base >>= 20; + } + + *factorp = factor; + + return base; +} + static unsigned int __init range_to_mtrr(unsigned int reg, unsigned long range_startk, unsigned long range_sizek, unsigned char type) @@ -919,13 +948,21 @@ range_to_mtrr(unsigned int reg, unsigned long range_startk, align = max_align; sizek = 1 << align; - if (debug_print) + if (debug_print) { + char start_factor = 'K', size_factor = 'K'; + unsigned long start_base, size_base; + + start_base = to_size_factor(range_startk, &start_factor), + size_base = to_size_factor(sizek, &size_factor), + printk(KERN_DEBUG "Setting variable MTRR %d, " - "base: %ldMB, range: %ldMB, type %s\n", - reg, range_startk >> 10, sizek >> 10, + "base: %ld%cB, range: %ld%cB, type %s\n", + reg, start_base, start_factor, + size_base, size_factor, (type == MTRR_TYPE_UNCACHABLE)?"UC": ((type == MTRR_TYPE_WRBACK)?"WB":"Other") ); + } save_var_mtrr(reg++, range_startk, sizek, type); range_startk += sizek; range_sizek -= sizek; @@ -970,6 +1007,8 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, /* try to append some small hole */ range0_basek = state->range_startk; range0_sizek = ALIGN(state->range_sizek, chunk_sizek); + + /* no increase */ if (range0_sizek == state->range_sizek) { if (debug_print) printk(KERN_DEBUG "rangeX: %016lx - %016lx\n", @@ -980,13 +1019,40 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, return 0; } - range0_sizek -= chunk_sizek; - if (range0_sizek && sizek) { - while (range0_basek + range0_sizek > (basek + sizek)) { - range0_sizek -= chunk_sizek; - if (!range0_sizek) - break; - } + /* only cut back, when it is not the last */ + if (sizek) { + while (range0_basek + range0_sizek > (basek + sizek)) { + if (range0_sizek >= chunk_sizek) + range0_sizek -= chunk_sizek; + else + range0_sizek = 0; + + if (!range0_sizek) + break; + } + } + +second_try: + range_basek = range0_basek + range0_sizek; + + /* one hole in the middle */ + if (range_basek > basek && range_basek <= (basek + sizek)) + second_sizek = range_basek - basek; + + if (range0_sizek > state->range_sizek) { + + /* one hole in middle or at end */ + hole_sizek = range0_sizek - state->range_sizek - second_sizek; + + /* hole size should be less than half of range0 size */ + if (hole_sizek >= (range0_sizek >> 1) && + range0_sizek >= chunk_sizek) { + range0_sizek -= chunk_sizek; + second_sizek = 0; + hole_sizek = 0; + + goto second_try; + } } if (range0_sizek) { @@ -996,50 +1062,28 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, (range0_basek + range0_sizek)<<10); state->reg = range_to_mtrr(state->reg, range0_basek, range0_sizek, MTRR_TYPE_WRBACK); - - } - - range_basek = range0_basek + range0_sizek; - range_sizek = chunk_sizek; - - if (range_basek + range_sizek > basek && - range_basek + range_sizek <= (basek + sizek)) { - /* one hole */ - second_basek = basek; - second_sizek = range_basek + range_sizek - basek; } - /* if last piece, only could one hole near end */ - if ((second_basek || !basek) && - range_sizek - (state->range_sizek - range0_sizek) - second_sizek < - (chunk_sizek >> 1)) { - /* - * one hole in middle (second_sizek is 0) or at end - * (second_sizek is 0 ) - */ - hole_sizek = range_sizek - (state->range_sizek - range0_sizek) - - second_sizek; - hole_basek = range_basek + range_sizek - hole_sizek - - second_sizek; - } else { - /* fallback for big hole, or several holes */ + if (range0_sizek < state->range_sizek) { + /* need to handle left over */ range_sizek = state->range_sizek - range0_sizek; - second_basek = 0; - second_sizek = 0; + + if (debug_print) + printk(KERN_DEBUG "range: %016lx - %016lx\n", + range_basek<<10, + (range_basek + range_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range_basek, + range_sizek, MTRR_TYPE_WRBACK); } - if (debug_print) - printk(KERN_DEBUG "range: %016lx - %016lx\n", range_basek<<10, - (range_basek + range_sizek)<<10); - state->reg = range_to_mtrr(state->reg, range_basek, range_sizek, - MTRR_TYPE_WRBACK); if (hole_sizek) { + hole_basek = range_basek - hole_sizek - second_sizek; if (debug_print) printk(KERN_DEBUG "hole: %016lx - %016lx\n", - hole_basek<<10, (hole_basek + hole_sizek)<<10); - state->reg = range_to_mtrr(state->reg, hole_basek, hole_sizek, - MTRR_TYPE_UNCACHABLE); - + hole_basek<<10, + (hole_basek + hole_sizek)<<10); + state->reg = range_to_mtrr(state->reg, hole_basek, + hole_sizek, MTRR_TYPE_UNCACHABLE); } return second_sizek; @@ -1154,11 +1198,11 @@ struct mtrr_cleanup_result { }; /* - * gran_size: 1M, 2M, ..., 2G - * chunk size: gran_size, ..., 4G - * so we need (2+13)*6 + * gran_size: 64K, 128K, 256K, 512K, 1M, 2M, ..., 2G + * chunk size: gran_size, ..., 2G + * so we need (1+16)*8 */ -#define NUM_RESULT 90 +#define NUM_RESULT 136 #define PSHIFT (PAGE_SHIFT - 10) static struct mtrr_cleanup_result __initdata result[NUM_RESULT]; @@ -1168,13 +1212,14 @@ static unsigned long __initdata min_loss_pfn[RANGE_NUM]; static int __init mtrr_cleanup(unsigned address_bits) { unsigned long extra_remove_base, extra_remove_size; - unsigned long i, base, size, def, dummy; + unsigned long base, size, def, dummy; mtrr_type type; int nr_range, nr_range_new; u64 chunk_size, gran_size; unsigned long range_sums, range_sums_new; int index_good; int num_reg_good; + int i; /* extra one for all 0 */ int num[MTRR_NUM_TYPES + 1]; @@ -1204,6 +1249,8 @@ static int __init mtrr_cleanup(unsigned address_bits) continue; if (!size) type = MTRR_NUM_TYPES; + if (type == MTRR_TYPE_WRPROT) + type = MTRR_TYPE_UNCACHABLE; num[type]++; } @@ -1216,23 +1263,57 @@ static int __init mtrr_cleanup(unsigned address_bits) num_var_ranges - num[MTRR_NUM_TYPES]) return 0; + /* print original var MTRRs at first, for debugging: */ + printk(KERN_DEBUG "original variable MTRRs\n"); + for (i = 0; i < num_var_ranges; i++) { + char start_factor = 'K', size_factor = 'K'; + unsigned long start_base, size_base; + + size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10); + if (!size_base) + continue; + + size_base = to_size_factor(size_base, &size_factor), + start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10); + start_base = to_size_factor(start_base, &start_factor), + type = range_state[i].type; + + printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n", + i, start_base, start_factor, + size_base, size_factor, + (type == MTRR_TYPE_UNCACHABLE) ? "UC" : + ((type == MTRR_TYPE_WRPROT) ? "WP" : + ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other")) + ); + } + memset(range, 0, sizeof(range)); extra_remove_size = 0; - if (mtrr_tom2) { - extra_remove_base = 1 << (32 - PAGE_SHIFT); + extra_remove_base = 1 << (32 - PAGE_SHIFT); + if (mtrr_tom2) extra_remove_size = (mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base; - } nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base, extra_remove_size); + /* + * [0, 1M) should always be coverred by var mtrr with WB + * and fixed mtrrs should take effective before var mtrr for it + */ + nr_range = add_range_with_merge(range, nr_range, 0, + (1ULL<<(20 - PAGE_SHIFT)) - 1); + /* sort the ranges */ + sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); + range_sums = sum_ranges(range, nr_range); printk(KERN_INFO "total RAM coverred: %ldM\n", range_sums >> (20 - PAGE_SHIFT)); if (mtrr_chunk_size && mtrr_gran_size) { int num_reg; + char gran_factor, chunk_factor, lose_factor; + unsigned long gran_base, chunk_base, lose_base; - debug_print = 1; + debug_print++; /* convert ranges to var ranges state */ num_reg = x86_setup_var_mtrrs(range, nr_range, mtrr_chunk_size, mtrr_gran_size); @@ -1256,34 +1337,48 @@ static int __init mtrr_cleanup(unsigned address_bits) result[i].lose_cover_sizek = (range_sums - range_sums_new) << PSHIFT; - printk(KERN_INFO "%sgran_size: %ldM \tchunk_size: %ldM \t", - result[i].bad?"*BAD*":" ", result[i].gran_sizek >> 10, - result[i].chunk_sizek >> 10); - printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ldM \n", + gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), + chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), + lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), + printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", + result[i].bad?"*BAD*":" ", + gran_base, gran_factor, chunk_base, chunk_factor); + printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", result[i].num_reg, result[i].bad?"-":"", - result[i].lose_cover_sizek >> 10); + lose_base, lose_factor); if (!result[i].bad) { set_var_mtrr_all(address_bits); return 1; } printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, " "will find optimal one\n"); - debug_print = 0; + debug_print--; memset(result, 0, sizeof(result[0])); } i = 0; memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn)); memset(result, 0, sizeof(result)); - for (gran_size = (1ULL<<20); gran_size < (1ULL<<32); gran_size <<= 1) { - for (chunk_size = gran_size; chunk_size < (1ULL<<33); + for (gran_size = (1ULL<<16); gran_size < (1ULL<<32); gran_size <<= 1) { + char gran_factor; + unsigned long gran_base; + + if (debug_print) + gran_base = to_size_factor(gran_size >> 10, &gran_factor); + + for (chunk_size = gran_size; chunk_size < (1ULL<<32); chunk_size <<= 1) { int num_reg; - if (debug_print) - printk(KERN_INFO - "\ngran_size: %lldM chunk_size_size: %lldM\n", - gran_size >> 20, chunk_size >> 20); + if (debug_print) { + char chunk_factor; + unsigned long chunk_base; + + chunk_base = to_size_factor(chunk_size>>10, &chunk_factor), + printk(KERN_INFO "\n"); + printk(KERN_INFO "gran_size: %ld%c chunk_size: %ld%c \n", + gran_base, gran_factor, chunk_base, chunk_factor); + } if (i >= NUM_RESULT) continue; @@ -1326,12 +1421,18 @@ static int __init mtrr_cleanup(unsigned address_bits) /* print out all */ for (i = 0; i < NUM_RESULT; i++) { - printk(KERN_INFO "%sgran_size: %ldM \tchunk_size: %ldM \t", - result[i].bad?"*BAD* ":" ", result[i].gran_sizek >> 10, - result[i].chunk_sizek >> 10); - printk(KERN_CONT "num_reg: %d \tlose RAM: %s%ldM\n", - result[i].num_reg, result[i].bad?"-":"", - result[i].lose_cover_sizek >> 10); + char gran_factor, chunk_factor, lose_factor; + unsigned long gran_base, chunk_base, lose_base; + + gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), + chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), + lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), + printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", + result[i].bad?"*BAD*":" ", + gran_base, gran_factor, chunk_base, chunk_factor); + printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", + result[i].num_reg, result[i].bad?"-":"", + lose_base, lose_factor); } /* try to find the optimal index */ @@ -1339,10 +1440,8 @@ static int __init mtrr_cleanup(unsigned address_bits) nr_mtrr_spare_reg = num_var_ranges - 1; num_reg_good = -1; for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) { - if (!min_loss_pfn[i]) { + if (!min_loss_pfn[i]) num_reg_good = i; - break; - } } index_good = -1; @@ -1358,21 +1457,26 @@ static int __init mtrr_cleanup(unsigned address_bits) } if (index_good != -1) { + char gran_factor, chunk_factor, lose_factor; + unsigned long gran_base, chunk_base, lose_base; + printk(KERN_INFO "Found optimal setting for mtrr clean up\n"); i = index_good; - printk(KERN_INFO "gran_size: %ldM \tchunk_size: %ldM \t", - result[i].gran_sizek >> 10, - result[i].chunk_sizek >> 10); - printk(KERN_CONT "num_reg: %d \tlose RAM: %ldM\n", - result[i].num_reg, - result[i].lose_cover_sizek >> 10); + gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), + chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), + lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), + printk(KERN_INFO "gran_size: %ld%c \tchunk_size: %ld%c \t", + gran_base, gran_factor, chunk_base, chunk_factor); + printk(KERN_CONT "num_reg: %d \tlose RAM: %ld%c\n", + result[i].num_reg, lose_base, lose_factor); /* convert ranges to var ranges state */ chunk_size = result[i].chunk_sizek; chunk_size <<= 10; gran_size = result[i].gran_sizek; gran_size <<= 10; - debug_print = 1; + debug_print++; x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size); + debug_print--; set_var_mtrr_all(address_bits); return 1; } diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index 05cc22d..6bff382 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -295,13 +295,19 @@ static int setup_k7_watchdog(unsigned nmi_hz) /* setup the timer */ wrmsr(evntsel_msr, evntsel, 0); write_watchdog_counter(perfctr_msr, "K7_PERFCTR0",nmi_hz); - apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= K7_EVNTSEL_ENABLE; - wrmsr(evntsel_msr, evntsel, 0); + /* initialize the wd struct before enabling */ wd->perfctr_msr = perfctr_msr; wd->evntsel_msr = evntsel_msr; wd->cccr_msr = 0; /* unused */ + + /* ok, everything is initialized, announce that we're set */ + cpu_nmi_set_wd_enabled(); + + apic_write(APIC_LVTPC, APIC_DM_NMI); + evntsel |= K7_EVNTSEL_ENABLE; + wrmsr(evntsel_msr, evntsel, 0); + return 1; } @@ -379,13 +385,19 @@ static int setup_p6_watchdog(unsigned nmi_hz) wrmsr(evntsel_msr, evntsel, 0); nmi_hz = adjust_for_32bit_ctr(nmi_hz); write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0",nmi_hz); - apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= P6_EVNTSEL0_ENABLE; - wrmsr(evntsel_msr, evntsel, 0); + /* initialize the wd struct before enabling */ wd->perfctr_msr = perfctr_msr; wd->evntsel_msr = evntsel_msr; wd->cccr_msr = 0; /* unused */ + + /* ok, everything is initialized, announce that we're set */ + cpu_nmi_set_wd_enabled(); + + apic_write(APIC_LVTPC, APIC_DM_NMI); + evntsel |= P6_EVNTSEL0_ENABLE; + wrmsr(evntsel_msr, evntsel, 0); + return 1; } @@ -432,6 +444,27 @@ static const struct wd_ops p6_wd_ops = { #define P4_CCCR_ENABLE (1 << 12) #define P4_CCCR_OVF (1 << 31) +#define P4_CONTROLS 18 +static unsigned int p4_controls[18] = { + MSR_P4_BPU_CCCR0, + MSR_P4_BPU_CCCR1, + MSR_P4_BPU_CCCR2, + MSR_P4_BPU_CCCR3, + MSR_P4_MS_CCCR0, + MSR_P4_MS_CCCR1, + MSR_P4_MS_CCCR2, + MSR_P4_MS_CCCR3, + MSR_P4_FLAME_CCCR0, + MSR_P4_FLAME_CCCR1, + MSR_P4_FLAME_CCCR2, + MSR_P4_FLAME_CCCR3, + MSR_P4_IQ_CCCR0, + MSR_P4_IQ_CCCR1, + MSR_P4_IQ_CCCR2, + MSR_P4_IQ_CCCR3, + MSR_P4_IQ_CCCR4, + MSR_P4_IQ_CCCR5, +}; /* * Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter * CRU_ESCR0 (with any non-null event selector) through a complemented @@ -473,6 +506,26 @@ static int setup_p4_watchdog(unsigned nmi_hz) evntsel_msr = MSR_P4_CRU_ESCR0; cccr_msr = MSR_P4_IQ_CCCR0; cccr_val = P4_CCCR_OVF_PMI0 | P4_CCCR_ESCR_SELECT(4); + + /* + * If we're on the kdump kernel or other situation, we may + * still have other performance counter registers set to + * interrupt and they'll keep interrupting forever because + * of the P4_CCCR_OVF quirk. So we need to ACK all the + * pending interrupts and disable all the registers here, + * before reenabling the NMI delivery. Refer to p4_rearm() + * about the P4_CCCR_OVF quirk. + */ + if (reset_devices) { + unsigned int low, high; + int i; + + for (i = 0; i < P4_CONTROLS; i++) { + rdmsr(p4_controls[i], low, high); + low &= ~(P4_CCCR_ENABLE | P4_CCCR_OVF); + wrmsr(p4_controls[i], low, high); + } + } } else { /* logical cpu 1 */ perfctr_msr = MSR_P4_IQ_PERFCTR1; @@ -499,12 +552,17 @@ static int setup_p4_watchdog(unsigned nmi_hz) wrmsr(evntsel_msr, evntsel, 0); wrmsr(cccr_msr, cccr_val, 0); write_watchdog_counter(perfctr_msr, "P4_IQ_COUNTER0", nmi_hz); - apic_write(APIC_LVTPC, APIC_DM_NMI); - cccr_val |= P4_CCCR_ENABLE; - wrmsr(cccr_msr, cccr_val, 0); + wd->perfctr_msr = perfctr_msr; wd->evntsel_msr = evntsel_msr; wd->cccr_msr = cccr_msr; + + /* ok, everything is initialized, announce that we're set */ + cpu_nmi_set_wd_enabled(); + + apic_write(APIC_LVTPC, APIC_DM_NMI); + cccr_val |= P4_CCCR_ENABLE; + wrmsr(cccr_msr, cccr_val, 0); return 1; } @@ -620,13 +678,17 @@ static int setup_intel_arch_watchdog(unsigned nmi_hz) wrmsr(evntsel_msr, evntsel, 0); nmi_hz = adjust_for_32bit_ctr(nmi_hz); write_watchdog_counter32(perfctr_msr, "INTEL_ARCH_PERFCTR0", nmi_hz); - apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE; - wrmsr(evntsel_msr, evntsel, 0); wd->perfctr_msr = perfctr_msr; wd->evntsel_msr = evntsel_msr; wd->cccr_msr = 0; /* unused */ + + /* ok, everything is initialized, announce that we're set */ + cpu_nmi_set_wd_enabled(); + + apic_write(APIC_LVTPC, APIC_DM_NMI); + evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE; + wrmsr(evntsel_msr, evntsel, 0); intel_arch_wd_ops.checkbit = 1ULL << (eax.split.bit_width - 1); return 1; } diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c index 8e9cd6a..6a44d64 100644 --- a/arch/x86/kernel/cpuid.c +++ b/arch/x86/kernel/cpuid.c @@ -36,7 +36,6 @@ #include <linux/smp_lock.h> #include <linux/major.h> #include <linux/fs.h> -#include <linux/smp_lock.h> #include <linux/device.h> #include <linux/cpu.h> #include <linux/notifier.h> diff --git a/arch/x86/kernel/crash_dump_64.c b/arch/x86/kernel/crash_dump_64.c index 15e6c6b..e90a60e 100644 --- a/arch/x86/kernel/crash_dump_64.c +++ b/arch/x86/kernel/crash_dump_64.c @@ -7,9 +7,8 @@ #include <linux/errno.h> #include <linux/crash_dump.h> - -#include <asm/uaccess.h> -#include <asm/io.h> +#include <linux/uaccess.h> +#include <linux/io.h> /** * copy_oldmem_page - copy one page from "oldmem" @@ -25,7 +24,7 @@ * in the current kernel. We stitch up a pte, similar to kmap_atomic. */ ssize_t copy_oldmem_page(unsigned long pfn, char *buf, - size_t csize, unsigned long offset, int userbuf) + size_t csize, unsigned long offset, int userbuf) { void *vaddr; @@ -33,14 +32,16 @@ ssize_t copy_oldmem_page(unsigned long pfn, char *buf, return 0; vaddr = ioremap(pfn << PAGE_SHIFT, PAGE_SIZE); + if (!vaddr) + return -ENOMEM; if (userbuf) { - if (copy_to_user(buf, (vaddr + offset), csize)) { + if (copy_to_user(buf, vaddr + offset, csize)) { iounmap(vaddr); return -EFAULT; } } else - memcpy(buf, (vaddr + offset), csize); + memcpy(buf, vaddr + offset, csize); iounmap(vaddr); return csize; diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index 11c11b8..2b69994 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -2,26 +2,49 @@ * Debug Store support * * This provides a low-level interface to the hardware's Debug Store - * feature that is used for last branch recording (LBR) and + * feature that is used for branch trace store (BTS) and * precise-event based sampling (PEBS). * - * Different architectures use a different DS layout/pointer size. - * The below functions therefore work on a void*. + * It manages: + * - per-thread and per-cpu allocation of BTS and PEBS + * - buffer memory allocation (optional) + * - buffer overflow handling + * - buffer access * + * It assumes: + * - get_task_struct on all parameter tasks + * - current is allowed to trace parameter tasks * - * Since there is no user for PEBS, yet, only LBR (or branch - * trace store, BTS) is supported. * - * - * Copyright (C) 2007 Intel Corporation. - * Markus Metzger <markus.t.metzger@intel.com>, Dec 2007 + * Copyright (C) 2007-2008 Intel Corporation. + * Markus Metzger <markus.t.metzger@intel.com>, 2007-2008 */ + +#ifdef CONFIG_X86_DS + #include <asm/ds.h> #include <linux/errno.h> #include <linux/string.h> #include <linux/slab.h> +#include <linux/sched.h> +#include <linux/mm.h> + + +/* + * The configuration for a particular DS hardware implementation. + */ +struct ds_configuration { + /* the size of the DS structure in bytes */ + unsigned char sizeof_ds; + /* the size of one pointer-typed field in the DS structure in bytes; + this covers the first 8 fields related to buffer management. */ + unsigned char sizeof_field; + /* the size of a BTS/PEBS record in bytes */ + unsigned char sizeof_rec[2]; +}; +static struct ds_configuration ds_cfg; /* @@ -44,378 +67,747 @@ * (interrupt occurs when write pointer passes interrupt pointer) * - value to which counter is reset following counter overflow * - * On later architectures, the last branch recording hardware uses - * 64bit pointers even in 32bit mode. - * - * - * Branch Trace Store (BTS) records store information about control - * flow changes. They at least provide the following information: - * - source linear address - * - destination linear address + * Later architectures use 64bit pointers throughout, whereas earlier + * architectures use 32bit pointers in 32bit mode. * - * Netburst supported a predicated bit that had been dropped in later - * architectures. We do not suppor it. * + * We compute the base address for the first 8 fields based on: + * - the field size stored in the DS configuration + * - the relative field position + * - an offset giving the start of the respective region * - * In order to abstract from the actual DS and BTS layout, we describe - * the access to the relevant fields. - * Thanks to Andi Kleen for proposing this design. + * This offset is further used to index various arrays holding + * information for BTS and PEBS at the respective index. * - * The implementation, however, is not as general as it might seem. In - * order to stay somewhat simple and efficient, we assume an - * underlying unsigned type (mostly a pointer type) and we expect the - * field to be at least as big as that type. + * On later 32bit processors, we only access the lower 32bit of the + * 64bit pointer fields. The upper halves will be zeroed out. */ -/* - * A special from_ip address to indicate that the BTS record is an - * info record that needs to be interpreted or skipped. - */ -#define BTS_ESCAPE_ADDRESS (-1) +enum ds_field { + ds_buffer_base = 0, + ds_index, + ds_absolute_maximum, + ds_interrupt_threshold, +}; -/* - * A field access descriptor - */ -struct access_desc { - unsigned char offset; - unsigned char size; +enum ds_qualifier { + ds_bts = 0, + ds_pebs }; +static inline unsigned long ds_get(const unsigned char *base, + enum ds_qualifier qual, enum ds_field field) +{ + base += (ds_cfg.sizeof_field * (field + (4 * qual))); + return *(unsigned long *)base; +} + +static inline void ds_set(unsigned char *base, enum ds_qualifier qual, + enum ds_field field, unsigned long value) +{ + base += (ds_cfg.sizeof_field * (field + (4 * qual))); + (*(unsigned long *)base) = value; +} + + /* - * The configuration for a particular DS/BTS hardware implementation. + * Locking is done only for allocating BTS or PEBS resources and for + * guarding context and buffer memory allocation. + * + * Most functions require the current task to own the ds context part + * they are going to access. All the locking is done when validating + * access to the context. */ -struct ds_configuration { - /* the DS configuration */ - unsigned char sizeof_ds; - struct access_desc bts_buffer_base; - struct access_desc bts_index; - struct access_desc bts_absolute_maximum; - struct access_desc bts_interrupt_threshold; - /* the BTS configuration */ - unsigned char sizeof_bts; - struct access_desc from_ip; - struct access_desc to_ip; - /* BTS variants used to store additional information like - timestamps */ - struct access_desc info_type; - struct access_desc info_data; - unsigned long debugctl_mask; -}; +static spinlock_t ds_lock = __SPIN_LOCK_UNLOCKED(ds_lock); /* - * The global configuration used by the below accessor functions + * Validate that the current task is allowed to access the BTS/PEBS + * buffer of the parameter task. + * + * Returns 0, if access is granted; -Eerrno, otherwise. */ -static struct ds_configuration ds_cfg; +static inline int ds_validate_access(struct ds_context *context, + enum ds_qualifier qual) +{ + if (!context) + return -EPERM; + + if (context->owner[qual] == current) + return 0; + + return -EPERM; +} + /* - * Accessor functions for some DS and BTS fields using the above - * global ptrace_bts_cfg. + * We either support (system-wide) per-cpu or per-thread allocation. + * We distinguish the two based on the task_struct pointer, where a + * NULL pointer indicates per-cpu allocation for the current cpu. + * + * Allocations are use-counted. As soon as resources are allocated, + * further allocations must be of the same type (per-cpu or + * per-thread). We model this by counting allocations (i.e. the number + * of tracers of a certain type) for one type negatively: + * =0 no tracers + * >0 number of per-thread tracers + * <0 number of per-cpu tracers + * + * The below functions to get and put tracers and to check the + * allocation type require the ds_lock to be held by the caller. + * + * Tracers essentially gives the number of ds contexts for a certain + * type of allocation. */ -static inline unsigned long get_bts_buffer_base(char *base) +static long tracers; + +static inline void get_tracer(struct task_struct *task) { - return *(unsigned long *)(base + ds_cfg.bts_buffer_base.offset); + tracers += (task ? 1 : -1); } -static inline void set_bts_buffer_base(char *base, unsigned long value) + +static inline void put_tracer(struct task_struct *task) { - (*(unsigned long *)(base + ds_cfg.bts_buffer_base.offset)) = value; + tracers -= (task ? 1 : -1); } -static inline unsigned long get_bts_index(char *base) + +static inline int check_tracer(struct task_struct *task) { - return *(unsigned long *)(base + ds_cfg.bts_index.offset); + return (task ? (tracers >= 0) : (tracers <= 0)); } -static inline void set_bts_index(char *base, unsigned long value) + + +/* + * The DS context is either attached to a thread or to a cpu: + * - in the former case, the thread_struct contains a pointer to the + * attached context. + * - in the latter case, we use a static array of per-cpu context + * pointers. + * + * Contexts are use-counted. They are allocated on first access and + * deallocated when the last user puts the context. + * + * We distinguish between an allocating and a non-allocating get of a + * context: + * - the allocating get is used for requesting BTS/PEBS resources. It + * requires the caller to hold the global ds_lock. + * - the non-allocating get is used for all other cases. A + * non-existing context indicates an error. It acquires and releases + * the ds_lock itself for obtaining the context. + * + * A context and its DS configuration are allocated and deallocated + * together. A context always has a DS configuration of the + * appropriate size. + */ +static DEFINE_PER_CPU(struct ds_context *, system_context); + +#define this_system_context per_cpu(system_context, smp_processor_id()) + +/* + * Returns the pointer to the parameter task's context or to the + * system-wide context, if task is NULL. + * + * Increases the use count of the returned context, if not NULL. + */ +static inline struct ds_context *ds_get_context(struct task_struct *task) { - (*(unsigned long *)(base + ds_cfg.bts_index.offset)) = value; + struct ds_context *context; + + spin_lock(&ds_lock); + + context = (task ? task->thread.ds_ctx : this_system_context); + if (context) + context->count++; + + spin_unlock(&ds_lock); + + return context; } -static inline unsigned long get_bts_absolute_maximum(char *base) + +/* + * Same as ds_get_context, but allocates the context and it's DS + * structure, if necessary; returns NULL; if out of memory. + * + * pre: requires ds_lock to be held + */ +static inline struct ds_context *ds_alloc_context(struct task_struct *task) { - return *(unsigned long *)(base + ds_cfg.bts_absolute_maximum.offset); + struct ds_context **p_context = + (task ? &task->thread.ds_ctx : &this_system_context); + struct ds_context *context = *p_context; + + if (!context) { + context = kzalloc(sizeof(*context), GFP_KERNEL); + + if (!context) + return NULL; + + context->ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL); + if (!context->ds) { + kfree(context); + return NULL; + } + + *p_context = context; + + context->this = p_context; + context->task = task; + + if (task) + set_tsk_thread_flag(task, TIF_DS_AREA_MSR); + + if (!task || (task == current)) + wrmsr(MSR_IA32_DS_AREA, (unsigned long)context->ds, 0); + + get_tracer(task); + } + + context->count++; + + return context; } -static inline void set_bts_absolute_maximum(char *base, unsigned long value) + +/* + * Decreases the use count of the parameter context, if not NULL. + * Deallocates the context, if the use count reaches zero. + */ +static inline void ds_put_context(struct ds_context *context) { - (*(unsigned long *)(base + ds_cfg.bts_absolute_maximum.offset)) = value; + if (!context) + return; + + spin_lock(&ds_lock); + + if (--context->count) + goto out; + + *(context->this) = NULL; + + if (context->task) + clear_tsk_thread_flag(context->task, TIF_DS_AREA_MSR); + + if (!context->task || (context->task == current)) + wrmsrl(MSR_IA32_DS_AREA, 0); + + put_tracer(context->task); + + /* free any leftover buffers from tracers that did not + * deallocate them properly. */ + kfree(context->buffer[ds_bts]); + kfree(context->buffer[ds_pebs]); + kfree(context->ds); + kfree(context); + out: + spin_unlock(&ds_lock); } -static inline unsigned long get_bts_interrupt_threshold(char *base) + + +/* + * Handle a buffer overflow + * + * task: the task whose buffers are overflowing; + * NULL for a buffer overflow on the current cpu + * context: the ds context + * qual: the buffer type + */ +static void ds_overflow(struct task_struct *task, struct ds_context *context, + enum ds_qualifier qual) { - return *(unsigned long *)(base + ds_cfg.bts_interrupt_threshold.offset); + if (!context) + return; + + if (context->callback[qual]) + (*context->callback[qual])(task); + + /* todo: do some more overflow handling */ } -static inline void set_bts_interrupt_threshold(char *base, unsigned long value) + + +/* + * Allocate a non-pageable buffer of the parameter size. + * Checks the memory and the locked memory rlimit. + * + * Returns the buffer, if successful; + * NULL, if out of memory or rlimit exceeded. + * + * size: the requested buffer size in bytes + * pages (out): if not NULL, contains the number of pages reserved + */ +static inline void *ds_allocate_buffer(size_t size, unsigned int *pages) { - (*(unsigned long *)(base + ds_cfg.bts_interrupt_threshold.offset)) = value; + unsigned long rlim, vm, pgsz; + void *buffer; + + pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT; + + rlim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT; + vm = current->mm->total_vm + pgsz; + if (rlim < vm) + return NULL; + + rlim = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT; + vm = current->mm->locked_vm + pgsz; + if (rlim < vm) + return NULL; + + buffer = kzalloc(size, GFP_KERNEL); + if (!buffer) + return NULL; + + current->mm->total_vm += pgsz; + current->mm->locked_vm += pgsz; + + if (pages) + *pages = pgsz; + + return buffer; } -static inline unsigned long get_from_ip(char *base) + +static int ds_request(struct task_struct *task, void *base, size_t size, + ds_ovfl_callback_t ovfl, enum ds_qualifier qual) { - return *(unsigned long *)(base + ds_cfg.from_ip.offset); + struct ds_context *context; + unsigned long buffer, adj; + const unsigned long alignment = (1 << 3); + int error = 0; + + if (!ds_cfg.sizeof_ds) + return -EOPNOTSUPP; + + /* we require some space to do alignment adjustments below */ + if (size < (alignment + ds_cfg.sizeof_rec[qual])) + return -EINVAL; + + /* buffer overflow notification is not yet implemented */ + if (ovfl) + return -EOPNOTSUPP; + + + spin_lock(&ds_lock); + + if (!check_tracer(task)) + return -EPERM; + + error = -ENOMEM; + context = ds_alloc_context(task); + if (!context) + goto out_unlock; + + error = -EALREADY; + if (context->owner[qual] == current) + goto out_unlock; + error = -EPERM; + if (context->owner[qual] != NULL) + goto out_unlock; + context->owner[qual] = current; + + spin_unlock(&ds_lock); + + + error = -ENOMEM; + if (!base) { + base = ds_allocate_buffer(size, &context->pages[qual]); + if (!base) + goto out_release; + + context->buffer[qual] = base; + } + error = 0; + + context->callback[qual] = ovfl; + + /* adjust the buffer address and size to meet alignment + * constraints: + * - buffer is double-word aligned + * - size is multiple of record size + * + * We checked the size at the very beginning; we have enough + * space to do the adjustment. + */ + buffer = (unsigned long)base; + + adj = ALIGN(buffer, alignment) - buffer; + buffer += adj; + size -= adj; + + size /= ds_cfg.sizeof_rec[qual]; + size *= ds_cfg.sizeof_rec[qual]; + + ds_set(context->ds, qual, ds_buffer_base, buffer); + ds_set(context->ds, qual, ds_index, buffer); + ds_set(context->ds, qual, ds_absolute_maximum, buffer + size); + + if (ovfl) { + /* todo: select a suitable interrupt threshold */ + } else + ds_set(context->ds, qual, + ds_interrupt_threshold, buffer + size + 1); + + /* we keep the context until ds_release */ + return error; + + out_release: + context->owner[qual] = NULL; + ds_put_context(context); + return error; + + out_unlock: + spin_unlock(&ds_lock); + ds_put_context(context); + return error; } -static inline void set_from_ip(char *base, unsigned long value) + +int ds_request_bts(struct task_struct *task, void *base, size_t size, + ds_ovfl_callback_t ovfl) { - (*(unsigned long *)(base + ds_cfg.from_ip.offset)) = value; + return ds_request(task, base, size, ovfl, ds_bts); } -static inline unsigned long get_to_ip(char *base) + +int ds_request_pebs(struct task_struct *task, void *base, size_t size, + ds_ovfl_callback_t ovfl) { - return *(unsigned long *)(base + ds_cfg.to_ip.offset); + return ds_request(task, base, size, ovfl, ds_pebs); } -static inline void set_to_ip(char *base, unsigned long value) + +static int ds_release(struct task_struct *task, enum ds_qualifier qual) { - (*(unsigned long *)(base + ds_cfg.to_ip.offset)) = value; + struct ds_context *context; + int error; + + context = ds_get_context(task); + error = ds_validate_access(context, qual); + if (error < 0) + goto out; + + kfree(context->buffer[qual]); + context->buffer[qual] = NULL; + + current->mm->total_vm -= context->pages[qual]; + current->mm->locked_vm -= context->pages[qual]; + context->pages[qual] = 0; + context->owner[qual] = NULL; + + /* + * we put the context twice: + * once for the ds_get_context + * once for the corresponding ds_request + */ + ds_put_context(context); + out: + ds_put_context(context); + return error; } -static inline unsigned char get_info_type(char *base) + +int ds_release_bts(struct task_struct *task) { - return *(unsigned char *)(base + ds_cfg.info_type.offset); + return ds_release(task, ds_bts); } -static inline void set_info_type(char *base, unsigned char value) + +int ds_release_pebs(struct task_struct *task) { - (*(unsigned char *)(base + ds_cfg.info_type.offset)) = value; + return ds_release(task, ds_pebs); } -static inline unsigned long get_info_data(char *base) + +static int ds_get_index(struct task_struct *task, size_t *pos, + enum ds_qualifier qual) { - return *(unsigned long *)(base + ds_cfg.info_data.offset); + struct ds_context *context; + unsigned long base, index; + int error; + + context = ds_get_context(task); + error = ds_validate_access(context, qual); + if (error < 0) + goto out; + + base = ds_get(context->ds, qual, ds_buffer_base); + index = ds_get(context->ds, qual, ds_index); + + error = ((index - base) / ds_cfg.sizeof_rec[qual]); + if (pos) + *pos = error; + out: + ds_put_context(context); + return error; } -static inline void set_info_data(char *base, unsigned long value) + +int ds_get_bts_index(struct task_struct *task, size_t *pos) { - (*(unsigned long *)(base + ds_cfg.info_data.offset)) = value; + return ds_get_index(task, pos, ds_bts); } +int ds_get_pebs_index(struct task_struct *task, size_t *pos) +{ + return ds_get_index(task, pos, ds_pebs); +} -int ds_allocate(void **dsp, size_t bts_size_in_bytes) +static int ds_get_end(struct task_struct *task, size_t *pos, + enum ds_qualifier qual) { - size_t bts_size_in_records; - unsigned long bts; - void *ds; + struct ds_context *context; + unsigned long base, end; + int error; + + context = ds_get_context(task); + error = ds_validate_access(context, qual); + if (error < 0) + goto out; + + base = ds_get(context->ds, qual, ds_buffer_base); + end = ds_get(context->ds, qual, ds_absolute_maximum); + + error = ((end - base) / ds_cfg.sizeof_rec[qual]); + if (pos) + *pos = error; + out: + ds_put_context(context); + return error; +} - if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts) - return -EOPNOTSUPP; +int ds_get_bts_end(struct task_struct *task, size_t *pos) +{ + return ds_get_end(task, pos, ds_bts); +} - if (bts_size_in_bytes < 0) - return -EINVAL; +int ds_get_pebs_end(struct task_struct *task, size_t *pos) +{ + return ds_get_end(task, pos, ds_pebs); +} - bts_size_in_records = - bts_size_in_bytes / ds_cfg.sizeof_bts; - bts_size_in_bytes = - bts_size_in_records * ds_cfg.sizeof_bts; +static int ds_access(struct task_struct *task, size_t index, + const void **record, enum ds_qualifier qual) +{ + struct ds_context *context; + unsigned long base, idx; + int error; - if (bts_size_in_bytes <= 0) + if (!record) return -EINVAL; - bts = (unsigned long)kzalloc(bts_size_in_bytes, GFP_KERNEL); - - if (!bts) - return -ENOMEM; + context = ds_get_context(task); + error = ds_validate_access(context, qual); + if (error < 0) + goto out; - ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL); + base = ds_get(context->ds, qual, ds_buffer_base); + idx = base + (index * ds_cfg.sizeof_rec[qual]); - if (!ds) { - kfree((void *)bts); - return -ENOMEM; - } - - set_bts_buffer_base(ds, bts); - set_bts_index(ds, bts); - set_bts_absolute_maximum(ds, bts + bts_size_in_bytes); - set_bts_interrupt_threshold(ds, bts + bts_size_in_bytes + 1); + error = -EINVAL; + if (idx > ds_get(context->ds, qual, ds_absolute_maximum)) + goto out; - *dsp = ds; - return 0; + *record = (const void *)idx; + error = ds_cfg.sizeof_rec[qual]; + out: + ds_put_context(context); + return error; } -int ds_free(void **dsp) +int ds_access_bts(struct task_struct *task, size_t index, const void **record) { - if (*dsp) { - kfree((void *)get_bts_buffer_base(*dsp)); - kfree(*dsp); - *dsp = NULL; - } - return 0; + return ds_access(task, index, record, ds_bts); } -int ds_get_bts_size(void *ds) +int ds_access_pebs(struct task_struct *task, size_t index, const void **record) { - int size_in_bytes; - - if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts) - return -EOPNOTSUPP; - - if (!ds) - return 0; - - size_in_bytes = - get_bts_absolute_maximum(ds) - - get_bts_buffer_base(ds); - return size_in_bytes; + return ds_access(task, index, record, ds_pebs); } -int ds_get_bts_end(void *ds) +static int ds_write(struct task_struct *task, const void *record, size_t size, + enum ds_qualifier qual, int force) { - int size_in_bytes = ds_get_bts_size(ds); - - if (size_in_bytes <= 0) - return size_in_bytes; + struct ds_context *context; + int error; - return size_in_bytes / ds_cfg.sizeof_bts; -} + if (!record) + return -EINVAL; -int ds_get_bts_index(void *ds) -{ - int index_offset_in_bytes; + error = -EPERM; + context = ds_get_context(task); + if (!context) + goto out; - if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts) - return -EOPNOTSUPP; + if (!force) { + error = ds_validate_access(context, qual); + if (error < 0) + goto out; + } - index_offset_in_bytes = - get_bts_index(ds) - - get_bts_buffer_base(ds); + error = 0; + while (size) { + unsigned long base, index, end, write_end, int_th; + unsigned long write_size, adj_write_size; + + /* + * write as much as possible without producing an + * overflow interrupt. + * + * interrupt_threshold must either be + * - bigger than absolute_maximum or + * - point to a record between buffer_base and absolute_maximum + * + * index points to a valid record. + */ + base = ds_get(context->ds, qual, ds_buffer_base); + index = ds_get(context->ds, qual, ds_index); + end = ds_get(context->ds, qual, ds_absolute_maximum); + int_th = ds_get(context->ds, qual, ds_interrupt_threshold); + + write_end = min(end, int_th); + + /* if we are already beyond the interrupt threshold, + * we fill the entire buffer */ + if (write_end <= index) + write_end = end; + + if (write_end <= index) + goto out; + + write_size = min((unsigned long) size, write_end - index); + memcpy((void *)index, record, write_size); + + record = (const char *)record + write_size; + size -= write_size; + error += write_size; + + adj_write_size = write_size / ds_cfg.sizeof_rec[qual]; + adj_write_size *= ds_cfg.sizeof_rec[qual]; + + /* zero out trailing bytes */ + memset((char *)index + write_size, 0, + adj_write_size - write_size); + index += adj_write_size; + + if (index >= end) + index = base; + ds_set(context->ds, qual, ds_index, index); + + if (index >= int_th) + ds_overflow(task, context, qual); + } - return index_offset_in_bytes / ds_cfg.sizeof_bts; + out: + ds_put_context(context); + return error; } -int ds_set_overflow(void *ds, int method) +int ds_write_bts(struct task_struct *task, const void *record, size_t size) { - switch (method) { - case DS_O_SIGNAL: - return -EOPNOTSUPP; - case DS_O_WRAP: - return 0; - default: - return -EINVAL; - } + return ds_write(task, record, size, ds_bts, /* force = */ 0); } -int ds_get_overflow(void *ds) +int ds_write_pebs(struct task_struct *task, const void *record, size_t size) { - return DS_O_WRAP; + return ds_write(task, record, size, ds_pebs, /* force = */ 0); } -int ds_clear(void *ds) +int ds_unchecked_write_bts(struct task_struct *task, + const void *record, size_t size) { - int bts_size = ds_get_bts_size(ds); - unsigned long bts_base; - - if (bts_size <= 0) - return bts_size; - - bts_base = get_bts_buffer_base(ds); - memset((void *)bts_base, 0, bts_size); - - set_bts_index(ds, bts_base); - return 0; + return ds_write(task, record, size, ds_bts, /* force = */ 1); } -int ds_read_bts(void *ds, int index, struct bts_struct *out) +int ds_unchecked_write_pebs(struct task_struct *task, + const void *record, size_t size) { - void *bts; + return ds_write(task, record, size, ds_pebs, /* force = */ 1); +} - if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts) - return -EOPNOTSUPP; +static int ds_reset_or_clear(struct task_struct *task, + enum ds_qualifier qual, int clear) +{ + struct ds_context *context; + unsigned long base, end; + int error; - if (index < 0) - return -EINVAL; + context = ds_get_context(task); + error = ds_validate_access(context, qual); + if (error < 0) + goto out; - if (index >= ds_get_bts_size(ds)) - return -EINVAL; + base = ds_get(context->ds, qual, ds_buffer_base); + end = ds_get(context->ds, qual, ds_absolute_maximum); - bts = (void *)(get_bts_buffer_base(ds) + (index * ds_cfg.sizeof_bts)); + if (clear) + memset((void *)base, 0, end - base); - memset(out, 0, sizeof(*out)); - if (get_from_ip(bts) == BTS_ESCAPE_ADDRESS) { - out->qualifier = get_info_type(bts); - out->variant.jiffies = get_info_data(bts); - } else { - out->qualifier = BTS_BRANCH; - out->variant.lbr.from_ip = get_from_ip(bts); - out->variant.lbr.to_ip = get_to_ip(bts); - } + ds_set(context->ds, qual, ds_index, base); - return sizeof(*out);; + error = 0; + out: + ds_put_context(context); + return error; } -int ds_write_bts(void *ds, const struct bts_struct *in) +int ds_reset_bts(struct task_struct *task) { - unsigned long bts; - - if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts) - return -EOPNOTSUPP; - - if (ds_get_bts_size(ds) <= 0) - return -ENXIO; + return ds_reset_or_clear(task, ds_bts, /* clear = */ 0); +} - bts = get_bts_index(ds); +int ds_reset_pebs(struct task_struct *task) +{ + return ds_reset_or_clear(task, ds_pebs, /* clear = */ 0); +} - memset((void *)bts, 0, ds_cfg.sizeof_bts); - switch (in->qualifier) { - case BTS_INVALID: - break; +int ds_clear_bts(struct task_struct *task) +{ + return ds_reset_or_clear(task, ds_bts, /* clear = */ 1); +} - case BTS_BRANCH: - set_from_ip((void *)bts, in->variant.lbr.from_ip); - set_to_ip((void *)bts, in->variant.lbr.to_ip); - break; +int ds_clear_pebs(struct task_struct *task) +{ + return ds_reset_or_clear(task, ds_pebs, /* clear = */ 1); +} - case BTS_TASK_ARRIVES: - case BTS_TASK_DEPARTS: - set_from_ip((void *)bts, BTS_ESCAPE_ADDRESS); - set_info_type((void *)bts, in->qualifier); - set_info_data((void *)bts, in->variant.jiffies); - break; +int ds_get_pebs_reset(struct task_struct *task, u64 *value) +{ + struct ds_context *context; + int error; - default: + if (!value) return -EINVAL; - } - bts = bts + ds_cfg.sizeof_bts; - if (bts >= get_bts_absolute_maximum(ds)) - bts = get_bts_buffer_base(ds); - set_bts_index(ds, bts); + context = ds_get_context(task); + error = ds_validate_access(context, ds_pebs); + if (error < 0) + goto out; - return ds_cfg.sizeof_bts; + *value = *(u64 *)(context->ds + (ds_cfg.sizeof_field * 8)); + + error = 0; + out: + ds_put_context(context); + return error; } -unsigned long ds_debugctl_mask(void) +int ds_set_pebs_reset(struct task_struct *task, u64 value) { - return ds_cfg.debugctl_mask; -} + struct ds_context *context; + int error; -#ifdef __i386__ -static const struct ds_configuration ds_cfg_netburst = { - .sizeof_ds = 9 * 4, - .bts_buffer_base = { 0, 4 }, - .bts_index = { 4, 4 }, - .bts_absolute_maximum = { 8, 4 }, - .bts_interrupt_threshold = { 12, 4 }, - .sizeof_bts = 3 * 4, - .from_ip = { 0, 4 }, - .to_ip = { 4, 4 }, - .info_type = { 4, 1 }, - .info_data = { 8, 4 }, - .debugctl_mask = (1<<2)|(1<<3) -}; + context = ds_get_context(task); + error = ds_validate_access(context, ds_pebs); + if (error < 0) + goto out; -static const struct ds_configuration ds_cfg_pentium_m = { - .sizeof_ds = 9 * 4, - .bts_buffer_base = { 0, 4 }, - .bts_index = { 4, 4 }, - .bts_absolute_maximum = { 8, 4 }, - .bts_interrupt_threshold = { 12, 4 }, - .sizeof_bts = 3 * 4, - .from_ip = { 0, 4 }, - .to_ip = { 4, 4 }, - .info_type = { 4, 1 }, - .info_data = { 8, 4 }, - .debugctl_mask = (1<<6)|(1<<7) + *(u64 *)(context->ds + (ds_cfg.sizeof_field * 8)) = value; + + error = 0; + out: + ds_put_context(context); + return error; +} + +static const struct ds_configuration ds_cfg_var = { + .sizeof_ds = sizeof(long) * 12, + .sizeof_field = sizeof(long), + .sizeof_rec[ds_bts] = sizeof(long) * 3, + .sizeof_rec[ds_pebs] = sizeof(long) * 10 }; -#endif /* _i386_ */ - -static const struct ds_configuration ds_cfg_core2 = { - .sizeof_ds = 9 * 8, - .bts_buffer_base = { 0, 8 }, - .bts_index = { 8, 8 }, - .bts_absolute_maximum = { 16, 8 }, - .bts_interrupt_threshold = { 24, 8 }, - .sizeof_bts = 3 * 8, - .from_ip = { 0, 8 }, - .to_ip = { 8, 8 }, - .info_type = { 8, 1 }, - .info_data = { 16, 8 }, - .debugctl_mask = (1<<6)|(1<<7)|(1<<9) +static const struct ds_configuration ds_cfg_64 = { + .sizeof_ds = 8 * 12, + .sizeof_field = 8, + .sizeof_rec[ds_bts] = 8 * 3, + .sizeof_rec[ds_pebs] = 8 * 10 }; static inline void @@ -429,14 +821,13 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) switch (c->x86) { case 0x6: switch (c->x86_model) { -#ifdef __i386__ case 0xD: case 0xE: /* Pentium M */ - ds_configure(&ds_cfg_pentium_m); + ds_configure(&ds_cfg_var); break; -#endif /* _i386_ */ case 0xF: /* Core2 */ - ds_configure(&ds_cfg_core2); + case 0x1C: /* Atom */ + ds_configure(&ds_cfg_64); break; default: /* sorry, don't know about them */ @@ -445,13 +836,11 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) break; case 0xF: switch (c->x86_model) { -#ifdef __i386__ case 0x0: case 0x1: case 0x2: /* Netburst */ - ds_configure(&ds_cfg_netburst); + ds_configure(&ds_cfg_var); break; -#endif /* _i386_ */ default: /* sorry, don't know about them */ break; @@ -462,3 +851,14 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) break; } } + +void ds_free(struct ds_context *context) +{ + /* This is called when the task owning the parameter context + * is dying. There should not be any user of that context left + * to disturb us, anymore. */ + unsigned long leftovers = context->count; + while (leftovers--) + ds_put_context(context); +} +#endif /* CONFIG_X86_DS */ diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index 06cc8d4..945a31c 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -414,9 +414,11 @@ void __init efi_init(void) if (memmap.map == NULL) printk(KERN_ERR "Could not map the EFI memory map!\n"); memmap.map_end = memmap.map + (memmap.nr_map * memmap.desc_size); + if (memmap.desc_size != sizeof(efi_memory_desc_t)) - printk(KERN_WARNING "Kernel-defined memdesc" - "doesn't match the one from EFI!\n"); + printk(KERN_WARNING + "Kernel-defined memdesc doesn't match the one from EFI!\n"); + if (add_efi_memmap) do_add_efi_memmap(); diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 89434d4..cf3a0b2 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -275,9 +275,9 @@ ENTRY(native_usergs_sysret64) ENTRY(ret_from_fork) CFI_DEFAULT_STACK push kernel_eflags(%rip) - CFI_ADJUST_CFA_OFFSET 4 + CFI_ADJUST_CFA_OFFSET 8 popf # reset kernel eflags - CFI_ADJUST_CFA_OFFSET -4 + CFI_ADJUST_CFA_OFFSET -8 call schedule_tail GET_THREAD_INFO(%rcx) testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),TI_flags(%rcx) diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 9bfc4d7..d16084f 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -108,12 +108,11 @@ void __init x86_64_start_kernel(char * real_mode_data) } load_idt((const struct desc_ptr *)&idt_descr); - early_printk("Kernel alive\n"); + if (console_loglevel == 10) + early_printk("Kernel alive\n"); x86_64_init_pda(); - early_printk("Kernel really alive\n"); - x86_64_start_reservations(real_mode_data); } diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index a7010c3..e835b4e 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -172,10 +172,6 @@ num_subarch_entries = (. - subarch_entries) / 4 * * Note that the stack is not yet set up! */ -#define PTE_ATTR 0x007 /* PRESENT+RW+USER */ -#define PDE_ATTR 0x067 /* PRESENT+RW+USER+DIRTY+ACCESSED */ -#define PGD_ATTR 0x001 /* PRESENT (no other attributes) */ - default_entry: #ifdef CONFIG_X86_PAE @@ -196,9 +192,9 @@ default_entry: movl $pa(pg0), %edi movl %edi, pa(init_pg_tables_start) movl $pa(swapper_pg_pmd), %edx - movl $PTE_ATTR, %eax + movl $PTE_IDENT_ATTR, %eax 10: - leal PDE_ATTR(%edi),%ecx /* Create PMD entry */ + leal PDE_IDENT_ATTR(%edi),%ecx /* Create PMD entry */ movl %ecx,(%edx) /* Store PMD entry */ /* Upper half already zero */ addl $8,%edx @@ -215,7 +211,7 @@ default_entry: * End condition: we must map up to and including INIT_MAP_BEYOND_END * bytes beyond the end of our own page tables. */ - leal (INIT_MAP_BEYOND_END+PTE_ATTR)(%edi),%ebp + leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp cmpl %ebp,%eax jb 10b 1: @@ -224,7 +220,7 @@ default_entry: movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_ATTR,%eax + movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax movl %eax,pa(swapper_pg_pmd+0x1000*KPMDS-8) #else /* Not PAE */ @@ -233,9 +229,9 @@ page_pde_offset = (__PAGE_OFFSET >> 20); movl $pa(pg0), %edi movl %edi, pa(init_pg_tables_start) movl $pa(swapper_pg_dir), %edx - movl $PTE_ATTR, %eax + movl $PTE_IDENT_ATTR, %eax 10: - leal PDE_ATTR(%edi),%ecx /* Create PDE entry */ + leal PDE_IDENT_ATTR(%edi),%ecx /* Create PDE entry */ movl %ecx,(%edx) /* Store identity PDE entry */ movl %ecx,page_pde_offset(%edx) /* Store kernel PDE entry */ addl $4,%edx @@ -249,7 +245,7 @@ page_pde_offset = (__PAGE_OFFSET >> 20); * bytes beyond the end of our own page tables; the +0x007 is * the attribute bits */ - leal (INIT_MAP_BEYOND_END+PTE_ATTR)(%edi),%ebp + leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp cmpl %ebp,%eax jb 10b movl %edi,pa(init_pg_tables_end) @@ -257,7 +253,7 @@ page_pde_offset = (__PAGE_OFFSET >> 20); movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_ATTR,%eax + movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax movl %eax,pa(swapper_pg_dir+0xffc) #endif jmp 3f @@ -634,19 +630,19 @@ ENTRY(empty_zero_page) /* Page-aligned for the benefit of paravirt? */ .align PAGE_SIZE_asm ENTRY(swapper_pg_dir) - .long pa(swapper_pg_pmd+PGD_ATTR),0 /* low identity map */ + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 /* low identity map */ # if KPMDS == 3 - .long pa(swapper_pg_pmd+PGD_ATTR),0 - .long pa(swapper_pg_pmd+PGD_ATTR+0x1000),0 - .long pa(swapper_pg_pmd+PGD_ATTR+0x2000),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x2000),0 # elif KPMDS == 2 .long 0,0 - .long pa(swapper_pg_pmd+PGD_ATTR),0 - .long pa(swapper_pg_pmd+PGD_ATTR+0x1000),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0 # elif KPMDS == 1 .long 0,0 .long 0,0 - .long pa(swapper_pg_pmd+PGD_ATTR),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 # else # error "Kernel PMDs should be 1, 2 or 3" # endif diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index db3280a..26cfdc1 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -110,7 +110,7 @@ startup_64: movq %rdi, %rax shrq $PMD_SHIFT, %rax andq $(PTRS_PER_PMD - 1), %rax - leaq __PAGE_KERNEL_LARGE_EXEC(%rdi), %rdx + leaq __PAGE_KERNEL_IDENT_LARGE_EXEC(%rdi), %rdx leaq level2_spare_pgt(%rip), %rbx movq %rdx, 0(%rbx, %rax, 8) ident_complete: @@ -374,7 +374,7 @@ NEXT_PAGE(level2_ident_pgt) /* Since I easily can, map the first 1G. * Don't set NX because code runs from these pages. */ - PMDS(0, __PAGE_KERNEL_LARGE_EXEC, PTRS_PER_PMD) + PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD) NEXT_PAGE(level2_kernel_pgt) /* diff --git a/arch/x86/kernel/ioport.c b/arch/x86/kernel/ioport.c index 50e5e4a..1919143 100644 --- a/arch/x86/kernel/ioport.c +++ b/arch/x86/kernel/ioport.c @@ -14,6 +14,7 @@ #include <linux/slab.h> #include <linux/thread_info.h> #include <linux/syscalls.h> +#include <asm/syscalls.h> /* Set EXTENT bits starting at BASE in BITMAP to value TURN_ON. */ static void set_bitmap(unsigned long *bitmap, unsigned int base, diff --git a/arch/x86/kernel/ipi.c b/arch/x86/kernel/ipi.c index 3f7537b..f1c688e 100644 --- a/arch/x86/kernel/ipi.c +++ b/arch/x86/kernel/ipi.c @@ -20,6 +20,8 @@ #ifdef CONFIG_X86_32 #include <mach_apic.h> +#include <mach_ipi.h> + /* * the following functions deal with sending IPIs between CPUs. * @@ -147,7 +149,6 @@ void send_IPI_mask_sequence(cpumask_t mask, int vector) } /* must come after the send_IPI functions above for inlining */ -#include <mach_ipi.h> static int convert_apicid_to_cpu(int apic_id) { int i; diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index 1cf8c1f..b71e02d 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -325,7 +325,7 @@ skip: for_each_online_cpu(j) seq_printf(p, "%10u ", per_cpu(irq_stat,j).irq_call_count); - seq_printf(p, " function call interrupts\n"); + seq_printf(p, " Function call interrupts\n"); seq_printf(p, "TLB: "); for_each_online_cpu(j) seq_printf(p, "%10u ", diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c index 1f78b23..f065fe9 100644 --- a/arch/x86/kernel/irq_64.c +++ b/arch/x86/kernel/irq_64.c @@ -129,7 +129,7 @@ skip: seq_printf(p, "CAL: "); for_each_online_cpu(j) seq_printf(p, "%10u ", cpu_pda(j)->irq_call_count); - seq_printf(p, " function call interrupts\n"); + seq_printf(p, " Function call interrupts\n"); seq_printf(p, "TLB: "); for_each_online_cpu(j) seq_printf(p, "%10u ", cpu_pda(j)->irq_tlb_count); diff --git a/arch/x86/kernel/kdebugfs.c b/arch/x86/kernel/kdebugfs.c index f2d43bc..ff7d3b0 100644 --- a/arch/x86/kernel/kdebugfs.c +++ b/arch/x86/kernel/kdebugfs.c @@ -139,6 +139,7 @@ static int __init create_setup_data_nodes(struct dentry *parent) if (PageHighMem(pg)) { data = ioremap_cache(pa_data, sizeof(*data)); if (!data) { + kfree(node); error = -ENXIO; goto err_dir; } diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index f47f0eb..10435a1 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -69,6 +69,9 @@ static int gdb_x86vector = -1; */ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) { +#ifndef CONFIG_X86_32 + u32 *gdb_regs32 = (u32 *)gdb_regs; +#endif gdb_regs[GDB_AX] = regs->ax; gdb_regs[GDB_BX] = regs->bx; gdb_regs[GDB_CX] = regs->cx; @@ -76,9 +79,9 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) gdb_regs[GDB_SI] = regs->si; gdb_regs[GDB_DI] = regs->di; gdb_regs[GDB_BP] = regs->bp; - gdb_regs[GDB_PS] = regs->flags; gdb_regs[GDB_PC] = regs->ip; #ifdef CONFIG_X86_32 + gdb_regs[GDB_PS] = regs->flags; gdb_regs[GDB_DS] = regs->ds; gdb_regs[GDB_ES] = regs->es; gdb_regs[GDB_CS] = regs->cs; @@ -94,6 +97,9 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) gdb_regs[GDB_R13] = regs->r13; gdb_regs[GDB_R14] = regs->r14; gdb_regs[GDB_R15] = regs->r15; + gdb_regs32[GDB_PS] = regs->flags; + gdb_regs32[GDB_CS] = regs->cs; + gdb_regs32[GDB_SS] = regs->ss; #endif gdb_regs[GDB_SP] = regs->sp; } @@ -112,6 +118,9 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) */ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) { +#ifndef CONFIG_X86_32 + u32 *gdb_regs32 = (u32 *)gdb_regs; +#endif gdb_regs[GDB_AX] = 0; gdb_regs[GDB_BX] = 0; gdb_regs[GDB_CX] = 0; @@ -129,8 +138,10 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) gdb_regs[GDB_FS] = 0xFFFF; gdb_regs[GDB_GS] = 0xFFFF; #else - gdb_regs[GDB_PS] = *(unsigned long *)(p->thread.sp + 8); - gdb_regs[GDB_PC] = 0; + gdb_regs32[GDB_PS] = *(unsigned long *)(p->thread.sp + 8); + gdb_regs32[GDB_CS] = __KERNEL_CS; + gdb_regs32[GDB_SS] = __KERNEL_DS; + gdb_regs[GDB_PC] = p->thread.ip; gdb_regs[GDB_R8] = 0; gdb_regs[GDB_R9] = 0; gdb_regs[GDB_R10] = 0; @@ -153,6 +164,9 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) */ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) { +#ifndef CONFIG_X86_32 + u32 *gdb_regs32 = (u32 *)gdb_regs; +#endif regs->ax = gdb_regs[GDB_AX]; regs->bx = gdb_regs[GDB_BX]; regs->cx = gdb_regs[GDB_CX]; @@ -160,9 +174,9 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) regs->si = gdb_regs[GDB_SI]; regs->di = gdb_regs[GDB_DI]; regs->bp = gdb_regs[GDB_BP]; - regs->flags = gdb_regs[GDB_PS]; regs->ip = gdb_regs[GDB_PC]; #ifdef CONFIG_X86_32 + regs->flags = gdb_regs[GDB_PS]; regs->ds = gdb_regs[GDB_DS]; regs->es = gdb_regs[GDB_ES]; regs->cs = gdb_regs[GDB_CS]; @@ -175,6 +189,9 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) regs->r13 = gdb_regs[GDB_R13]; regs->r14 = gdb_regs[GDB_R14]; regs->r15 = gdb_regs[GDB_R15]; + regs->flags = gdb_regs32[GDB_PS]; + regs->cs = gdb_regs32[GDB_CS]; + regs->ss = gdb_regs32[GDB_SS]; #endif } @@ -378,10 +395,8 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code, if (remcomInBuffer[0] == 's') { linux_regs->flags |= X86_EFLAGS_TF; kgdb_single_step = 1; - if (kgdb_contthread) { - atomic_set(&kgdb_cpu_doing_single_step, - raw_smp_processor_id()); - } + atomic_set(&kgdb_cpu_doing_single_step, + raw_smp_processor_id()); } get_debugreg(dr6, 6); @@ -440,12 +455,7 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd) return NOTIFY_DONE; case DIE_NMI_IPI: - if (atomic_read(&kgdb_active) != -1) { - /* KGDB CPU roundup */ - kgdb_nmicallback(raw_smp_processor_id(), regs); - was_in_debug_nmi[raw_smp_processor_id()] = 1; - touch_nmi_watchdog(); - } + /* Just ignore, we will handle the roundup on DIE_NMI. */ return NOTIFY_DONE; case DIE_NMIUNKNOWN: @@ -466,9 +476,15 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd) case DIE_DEBUG: if (atomic_read(&kgdb_cpu_doing_single_step) == - raw_smp_processor_id() && - user_mode(regs)) - return single_step_cont(regs, args); + raw_smp_processor_id()) { + if (user_mode(regs)) + return single_step_cont(regs, args); + break; + } else if (test_thread_flag(TIF_SINGLESTEP)) + /* This means a user thread is single stepping + * a system call which should be ignored + */ + return NOTIFY_DONE; /* fall through */ default: if (user_mode(regs)) diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index 8b7a3cf..478bca9 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -178,7 +178,7 @@ static void kvm_flush_tlb(void) kvm_deferred_mmu_op(&ftlb, sizeof ftlb); } -static void kvm_release_pt(u32 pfn) +static void kvm_release_pt(unsigned long pfn) { struct kvm_mmu_op_release_pt rpt = { .header.op = KVM_MMU_OP_RELEASE_PT, diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c index b68e21f..0ed5f93 100644 --- a/arch/x86/kernel/ldt.c +++ b/arch/x86/kernel/ldt.c @@ -18,6 +18,7 @@ #include <asm/ldt.h> #include <asm/desc.h> #include <asm/mmu_context.h> +#include <asm/syscalls.h> #ifdef CONFIG_SMP static void flush_ldt(void *current_mm) diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c index abb78a2..2c97f07 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/nmi.c @@ -299,6 +299,15 @@ void acpi_nmi_disable(void) on_each_cpu(__acpi_nmi_disable, NULL, 1); } +/* + * This function is called as soon the LAPIC NMI watchdog driver has everything + * in place and it's ready to check if the NMIs belong to the NMI watchdog + */ +void cpu_nmi_set_wd_enabled(void) +{ + __get_cpu_var(wd_enabled) = 1; +} + void setup_apic_nmi_watchdog(void *unused) { if (__get_cpu_var(wd_enabled)) @@ -311,8 +320,6 @@ void setup_apic_nmi_watchdog(void *unused) switch (nmi_watchdog) { case NMI_LOCAL_APIC: - /* enable it before to avoid race with handler */ - __get_cpu_var(wd_enabled) = 1; if (lapic_watchdog_init(nmi_hz) < 0) { __get_cpu_var(wd_enabled) = 0; return; diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c index 3e66722..7a13fac 100644 --- a/arch/x86/kernel/olpc.c +++ b/arch/x86/kernel/olpc.c @@ -190,12 +190,12 @@ EXPORT_SYMBOL_GPL(olpc_ec_cmd); static void __init platform_detect(void) { size_t propsize; - u32 rev; + __be32 rev; if (ofw("getprop", 4, 1, NULL, "board-revision-int", &rev, 4, &propsize) || propsize != 4) { printk(KERN_ERR "ofw: getprop call failed!\n"); - rev = 0; + rev = cpu_to_be32(0); } olpc_platform_info.boardrev = be32_to_cpu(rev); } @@ -203,7 +203,7 @@ static void __init platform_detect(void) static void __init platform_detect(void) { /* stopgap until OFW support is added to the kernel */ - olpc_platform_info.boardrev = be32_to_cpu(0xc2); + olpc_platform_info.boardrev = 0xc2; } #endif diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 300da17..e2f4376 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -330,6 +330,7 @@ struct pv_cpu_ops pv_cpu_ops = { #endif .wbinvd = native_wbinvd, .read_msr = native_read_msr_safe, + .read_msr_amd = native_read_msr_amd_safe, .write_msr = native_write_msr_safe, .read_tsc = native_read_tsc, .read_pmc = native_read_pmc, diff --git a/arch/x86/kernel/paravirt_patch_32.c b/arch/x86/kernel/paravirt_patch_32.c index 5826221..9fe644f 100644 --- a/arch/x86/kernel/paravirt_patch_32.c +++ b/arch/x86/kernel/paravirt_patch_32.c @@ -23,7 +23,7 @@ unsigned native_patch(u8 type, u16 clobbers, void *ibuf, start = start_##ops##_##x; \ end = end_##ops##_##x; \ goto patch_site - switch(type) { + switch (type) { PATCH_SITE(pv_irq_ops, irq_disable); PATCH_SITE(pv_irq_ops, irq_enable); PATCH_SITE(pv_irq_ops, restore_fl); diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 4e612d2..0a3824e 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -83,7 +83,7 @@ void __init dma32_reserve_bootmem(void) * using 512M as goal */ align = 64ULL<<20; - size = round_up(dma32_bootmem_size, align); + size = roundup(dma32_bootmem_size, align); dma32_bootmem_ptr = __alloc_bootmem_nopanic(size, align, 512ULL<<20); if (dma32_bootmem_ptr) diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index d077116..145f1c8 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -650,7 +650,6 @@ static __init int init_k8_gatt(struct agp_kern_info *info) struct pci_dev *dev; void *gatt; int i, error; - unsigned long start_pfn, end_pfn; printk(KERN_INFO "PCI-DMA: Disabling AGP.\n"); aper_size = aper_base = info->aper_size = 0; @@ -697,12 +696,6 @@ static __init int init_k8_gatt(struct agp_kern_info *info) printk(KERN_INFO "PCI-DMA: aperture base @ %x size %u KB\n", aper_base, aper_size>>10); - /* need to map that range */ - end_pfn = (aper_base>>PAGE_SHIFT) + (aper_size>>PAGE_SHIFT); - if (end_pfn > max_low_pfn_mapped) { - start_pfn = (aper_base>>PAGE_SHIFT); - init_memory_mapping(start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT); - } return 0; nommu: @@ -745,7 +738,8 @@ void __init gart_iommu_init(void) { struct agp_kern_info info; unsigned long iommu_start; - unsigned long aper_size; + unsigned long aper_base, aper_size; + unsigned long start_pfn, end_pfn; unsigned long scratch; long i; @@ -783,8 +777,16 @@ void __init gart_iommu_init(void) return; } + /* need to map that range */ + aper_size = info.aper_size << 20; + aper_base = info.aper_base; + end_pfn = (aper_base>>PAGE_SHIFT) + (aper_size>>PAGE_SHIFT); + if (end_pfn > max_low_pfn_mapped) { + start_pfn = (aper_base>>PAGE_SHIFT); + init_memory_mapping(start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT); + } + printk(KERN_INFO "PCI-DMA: using GART IOMMU.\n"); - aper_size = info.aper_size * 1024 * 1024; iommu_size = check_iommu_size(info.aper_base, aper_size); iommu_pages = iommu_size >> PAGE_SHIFT; diff --git a/arch/x86/kernel/pcspeaker.c b/arch/x86/kernel/pcspeaker.c index bc1f2d3..a311ffc 100644 --- a/arch/x86/kernel/pcspeaker.c +++ b/arch/x86/kernel/pcspeaker.c @@ -1,20 +1,13 @@ #include <linux/platform_device.h> -#include <linux/errno.h> +#include <linux/err.h> #include <linux/init.h> static __init int add_pcspkr(void) { struct platform_device *pd; - int ret; - pd = platform_device_alloc("pcspkr", -1); - if (!pd) - return -ENOMEM; + pd = platform_device_register_simple("pcspkr", -1, NULL, 0); - ret = platform_device_add(pd); - if (ret) - platform_device_put(pd); - - return ret; + return IS_ERR(pd) ? PTR_ERR(pd) : 0; } device_initcall(add_pcspkr); diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 7fc4d5b..ec7a2ba 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -185,7 +185,8 @@ static void mwait_idle(void) static void poll_idle(void) { local_irq_enable(); - cpu_relax(); + while (!need_resched()) + cpu_relax(); } /* @@ -246,6 +247,14 @@ static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c) return 1; } +static cpumask_t c1e_mask = CPU_MASK_NONE; +static int c1e_detected; + +void c1e_remove_cpu(int cpu) +{ + cpu_clear(cpu, c1e_mask); +} + /* * C1E aware idle routine. We check for C1E active in the interrupt * pending message MSR. If we detect C1E, then we handle it the same @@ -253,9 +262,6 @@ static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c) */ static void c1e_idle(void) { - static cpumask_t c1e_mask = CPU_MASK_NONE; - static int c1e_detected; - if (need_resched()) return; @@ -265,8 +271,10 @@ static void c1e_idle(void) rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi); if (lo & K8_INTP_C1E_ACTIVE_MASK) { c1e_detected = 1; - mark_tsc_unstable("TSC halt in C1E"); - printk(KERN_INFO "System has C1E enabled\n"); + if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) + mark_tsc_unstable("TSC halt in AMD C1E"); + printk(KERN_INFO "System has AMD C1E enabled\n"); + set_cpu_cap(&boot_cpu_data, X86_FEATURE_AMDC1E); } } diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 3b7a1dd..205188d 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -37,6 +37,7 @@ #include <linux/tick.h> #include <linux/percpu.h> #include <linux/prctl.h> +#include <linux/dmi.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -55,6 +56,9 @@ #include <asm/tlbflush.h> #include <asm/cpu.h> #include <asm/kdebug.h> +#include <asm/idle.h> +#include <asm/syscalls.h> +#include <asm/smp.h> asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); @@ -88,6 +92,7 @@ static void cpu_exit_clear(void) cpu_clear(cpu, cpu_callin_map); numa_remove_cpu(cpu); + c1e_remove_cpu(cpu); } /* We don't actually take CPU down, just spin without interrupts. */ @@ -159,6 +164,7 @@ void __show_registers(struct pt_regs *regs, int all) unsigned long d0, d1, d2, d3, d6, d7; unsigned long sp; unsigned short ss, gs; + const char *board; if (user_mode_vm(regs)) { sp = regs->sp; @@ -171,11 +177,15 @@ void __show_registers(struct pt_regs *regs, int all) } printk("\n"); - printk("Pid: %d, comm: %s %s (%s %.*s)\n", + + board = dmi_get_system_info(DMI_PRODUCT_NAME); + if (!board) + board = ""; + printk("Pid: %d, comm: %s %s (%s %.*s) %s\n", task_pid_nr(current), current->comm, print_tainted(), init_utsname()->release, (int)strcspn(init_utsname()->version, " "), - init_utsname()->version); + init_utsname()->version, board); printk("EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n", (u16)regs->cs, regs->ip, regs->flags, @@ -275,6 +285,14 @@ void exit_thread(void) tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET; put_cpu(); } +#ifdef CONFIG_X86_DS + /* Free any DS contexts that have not been properly released. */ + if (unlikely(current->thread.ds_ctx)) { + /* we clear debugctl to make sure DS is not used. */ + update_debugctlmsr(0); + ds_free(current->thread.ds_ctx); + } +#endif /* CONFIG_X86_DS */ } void flush_thread(void) @@ -436,6 +454,35 @@ int set_tsc_mode(unsigned int val) return 0; } +#ifdef CONFIG_X86_DS +static int update_debugctl(struct thread_struct *prev, + struct thread_struct *next, unsigned long debugctl) +{ + unsigned long ds_prev = 0; + unsigned long ds_next = 0; + + if (prev->ds_ctx) + ds_prev = (unsigned long)prev->ds_ctx->ds; + if (next->ds_ctx) + ds_next = (unsigned long)next->ds_ctx->ds; + + if (ds_next != ds_prev) { + /* we clear debugctl to make sure DS + * is not in use when we change it */ + debugctl = 0; + update_debugctlmsr(0); + wrmsr(MSR_IA32_DS_AREA, ds_next, 0); + } + return debugctl; +} +#else +static int update_debugctl(struct thread_struct *prev, + struct thread_struct *next, unsigned long debugctl) +{ + return debugctl; +} +#endif /* CONFIG_X86_DS */ + static noinline void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, struct tss_struct *tss) @@ -446,14 +493,7 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, prev = &prev_p->thread; next = &next_p->thread; - debugctl = prev->debugctlmsr; - if (next->ds_area_msr != prev->ds_area_msr) { - /* we clear debugctl to make sure DS - * is not in use when we change it */ - debugctl = 0; - update_debugctlmsr(0); - wrmsr(MSR_IA32_DS_AREA, next->ds_area_msr, 0); - } + debugctl = update_debugctl(prev, next, prev->debugctlmsr); if (next->debugctlmsr != debugctl) update_debugctlmsr(next->debugctlmsr); @@ -477,13 +517,13 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, hard_enable_TSC(); } -#ifdef X86_BTS +#ifdef CONFIG_X86_PTRACE_BTS if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS)) ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS); if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS)) ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES); -#endif +#endif /* CONFIG_X86_PTRACE_BTS */ if (!test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) { diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 71553b6..2a8ccb9 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -37,11 +37,11 @@ #include <linux/kdebug.h> #include <linux/tick.h> #include <linux/prctl.h> +#include <linux/uaccess.h> +#include <linux/io.h> -#include <asm/uaccess.h> #include <asm/pgtable.h> #include <asm/system.h> -#include <asm/io.h> #include <asm/processor.h> #include <asm/i387.h> #include <asm/mmu_context.h> @@ -51,6 +51,7 @@ #include <asm/proto.h> #include <asm/ia32.h> #include <asm/idle.h> +#include <asm/syscalls.h> asmlinkage extern void ret_from_fork(void); @@ -88,11 +89,13 @@ void exit_idle(void) #ifdef CONFIG_HOTPLUG_CPU DECLARE_PER_CPU(int, cpu_state); -#include <asm/nmi.h> +#include <linux/nmi.h> /* We halt the CPU with physical CPU hotplug */ static inline void play_dead(void) { idle_task_exit(); + c1e_remove_cpu(raw_smp_processor_id()); + mb(); /* Ack it */ __get_cpu_var(cpu_state) = CPU_DEAD; @@ -151,7 +154,7 @@ void cpu_idle(void) } /* Prints also some state that isn't saved in the pt_regs */ -void __show_regs(struct pt_regs * regs) +void __show_regs(struct pt_regs *regs) { unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L, fs, gs, shadowgs; unsigned long d0, d1, d2, d3, d6, d7; @@ -160,59 +163,61 @@ void __show_regs(struct pt_regs * regs) printk("\n"); print_modules(); - printk("Pid: %d, comm: %.20s %s %s %.*s\n", + printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s\n", current->pid, current->comm, print_tainted(), init_utsname()->release, (int)strcspn(init_utsname()->version, " "), init_utsname()->version); - printk("RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip); + printk(KERN_INFO "RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip); printk_address(regs->ip, 1); - printk("RSP: %04lx:%016lx EFLAGS: %08lx\n", regs->ss, regs->sp, - regs->flags); - printk("RAX: %016lx RBX: %016lx RCX: %016lx\n", + printk(KERN_INFO "RSP: %04lx:%016lx EFLAGS: %08lx\n", regs->ss, + regs->sp, regs->flags); + printk(KERN_INFO "RAX: %016lx RBX: %016lx RCX: %016lx\n", regs->ax, regs->bx, regs->cx); - printk("RDX: %016lx RSI: %016lx RDI: %016lx\n", + printk(KERN_INFO "RDX: %016lx RSI: %016lx RDI: %016lx\n", regs->dx, regs->si, regs->di); - printk("RBP: %016lx R08: %016lx R09: %016lx\n", + printk(KERN_INFO "RBP: %016lx R08: %016lx R09: %016lx\n", regs->bp, regs->r8, regs->r9); - printk("R10: %016lx R11: %016lx R12: %016lx\n", - regs->r10, regs->r11, regs->r12); - printk("R13: %016lx R14: %016lx R15: %016lx\n", - regs->r13, regs->r14, regs->r15); - - asm("movl %%ds,%0" : "=r" (ds)); - asm("movl %%cs,%0" : "=r" (cs)); - asm("movl %%es,%0" : "=r" (es)); + printk(KERN_INFO "R10: %016lx R11: %016lx R12: %016lx\n", + regs->r10, regs->r11, regs->r12); + printk(KERN_INFO "R13: %016lx R14: %016lx R15: %016lx\n", + regs->r13, regs->r14, regs->r15); + + asm("movl %%ds,%0" : "=r" (ds)); + asm("movl %%cs,%0" : "=r" (cs)); + asm("movl %%es,%0" : "=r" (es)); asm("movl %%fs,%0" : "=r" (fsindex)); asm("movl %%gs,%0" : "=r" (gsindex)); rdmsrl(MSR_FS_BASE, fs); - rdmsrl(MSR_GS_BASE, gs); - rdmsrl(MSR_KERNEL_GS_BASE, shadowgs); + rdmsrl(MSR_GS_BASE, gs); + rdmsrl(MSR_KERNEL_GS_BASE, shadowgs); cr0 = read_cr0(); cr2 = read_cr2(); cr3 = read_cr3(); cr4 = read_cr4(); - printk("FS: %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n", - fs,fsindex,gs,gsindex,shadowgs); - printk("CS: %04x DS: %04x ES: %04x CR0: %016lx\n", cs, ds, es, cr0); - printk("CR2: %016lx CR3: %016lx CR4: %016lx\n", cr2, cr3, cr4); + printk(KERN_INFO "FS: %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n", + fs, fsindex, gs, gsindex, shadowgs); + printk(KERN_INFO "CS: %04x DS: %04x ES: %04x CR0: %016lx\n", cs, ds, + es, cr0); + printk(KERN_INFO "CR2: %016lx CR3: %016lx CR4: %016lx\n", cr2, cr3, + cr4); get_debugreg(d0, 0); get_debugreg(d1, 1); get_debugreg(d2, 2); - printk("DR0: %016lx DR1: %016lx DR2: %016lx\n", d0, d1, d2); + printk(KERN_INFO "DR0: %016lx DR1: %016lx DR2: %016lx\n", d0, d1, d2); get_debugreg(d3, 3); get_debugreg(d6, 6); get_debugreg(d7, 7); - printk("DR3: %016lx DR6: %016lx DR7: %016lx\n", d3, d6, d7); + printk(KERN_INFO "DR3: %016lx DR6: %016lx DR7: %016lx\n", d3, d6, d7); } void show_regs(struct pt_regs *regs) { - printk("CPU %d:", smp_processor_id()); + printk(KERN_INFO "CPU %d:", smp_processor_id()); __show_regs(regs); show_trace(NULL, regs, (void *)(regs + 1), regs->bp); } @@ -238,6 +243,14 @@ void exit_thread(void) t->io_bitmap_max = 0; put_cpu(); } +#ifdef CONFIG_X86_DS + /* Free any DS contexts that have not been properly released. */ + if (unlikely(t->ds_ctx)) { + /* we clear debugctl to make sure DS is not used. */ + update_debugctlmsr(0); + ds_free(t->ds_ctx); + } +#endif /* CONFIG_X86_DS */ } void flush_thread(void) @@ -313,10 +326,10 @@ void prepare_to_copy(struct task_struct *tsk) int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, unsigned long unused, - struct task_struct * p, struct pt_regs * regs) + struct task_struct *p, struct pt_regs *regs) { int err; - struct pt_regs * childregs; + struct pt_regs *childregs; struct task_struct *me = current; childregs = ((struct pt_regs *) @@ -361,10 +374,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, if (test_thread_flag(TIF_IA32)) err = do_set_thread_area(p, -1, (struct user_desc __user *)childregs->si, 0); - else -#endif - err = do_arch_prctl(p, ARCH_SET_FS, childregs->r8); - if (err) + else +#endif + err = do_arch_prctl(p, ARCH_SET_FS, childregs->r8); + if (err) goto out; } err = 0; @@ -471,13 +484,27 @@ static inline void __switch_to_xtra(struct task_struct *prev_p, next = &next_p->thread; debugctl = prev->debugctlmsr; - if (next->ds_area_msr != prev->ds_area_msr) { - /* we clear debugctl to make sure DS - * is not in use when we change it */ - debugctl = 0; - update_debugctlmsr(0); - wrmsrl(MSR_IA32_DS_AREA, next->ds_area_msr); + +#ifdef CONFIG_X86_DS + { + unsigned long ds_prev = 0, ds_next = 0; + + if (prev->ds_ctx) + ds_prev = (unsigned long)prev->ds_ctx->ds; + if (next->ds_ctx) + ds_next = (unsigned long)next->ds_ctx->ds; + + if (ds_next != ds_prev) { + /* + * We clear debugctl to make sure DS + * is not in use when we change it: + */ + debugctl = 0; + update_debugctlmsr(0); + wrmsrl(MSR_IA32_DS_AREA, ds_next); + } } +#endif /* CONFIG_X86_DS */ if (next->debugctlmsr != debugctl) update_debugctlmsr(next->debugctlmsr); @@ -515,13 +542,13 @@ static inline void __switch_to_xtra(struct task_struct *prev_p, memset(tss->io_bitmap, 0xff, prev->io_bitmap_max); } -#ifdef X86_BTS +#ifdef CONFIG_X86_PTRACE_BTS if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS)) ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS); if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS)) ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES); -#endif +#endif /* CONFIG_X86_PTRACE_BTS */ } /* @@ -543,7 +570,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) unsigned fsindex, gsindex; /* we're going to use this soon, after a few expensive things */ - if (next_p->fpu_counter>5) + if (next_p->fpu_counter > 5) prefetch(next->xstate); /* @@ -551,13 +578,13 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) */ load_sp0(tss, next); - /* + /* * Switch DS and ES. * This won't pick up thread selector changes, but I guess that is ok. */ savesegment(es, prev->es); if (unlikely(next->es | prev->es)) - loadsegment(es, next->es); + loadsegment(es, next->es); savesegment(ds, prev->ds); if (unlikely(next->ds | prev->ds)) @@ -583,7 +610,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) */ arch_leave_lazy_cpu_mode(); - /* + /* * Switch FS and GS. * * Segment register != 0 always requires a reload. Also @@ -592,13 +619,13 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) */ if (unlikely(fsindex | next->fsindex | prev->fs)) { loadsegment(fs, next->fsindex); - /* + /* * Check if the user used a selector != 0; if yes * clear 64bit base, since overloaded base is always * mapped to the Null selector */ if (fsindex) - prev->fs = 0; + prev->fs = 0; } /* when next process has a 64bit base use it */ if (next->fs) @@ -608,7 +635,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) if (unlikely(gsindex | next->gsindex | prev->gs)) { load_gs_index(next->gsindex); if (gsindex) - prev->gs = 0; + prev->gs = 0; } if (next->gs) wrmsrl(MSR_KERNEL_GS_BASE, next->gs); @@ -617,12 +644,12 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) /* Must be after DS reload */ unlazy_fpu(prev_p); - /* + /* * Switch the PDA and FPU contexts. */ prev->usersp = read_pda(oldrsp); write_pda(oldrsp, next->usersp); - write_pda(pcurrent, next_p); + write_pda(pcurrent, next_p); write_pda(kernelstack, (unsigned long)task_stack_page(next_p) + @@ -663,7 +690,7 @@ long sys_execve(char __user *name, char __user * __user *argv, char __user * __user *envp, struct pt_regs *regs) { long error; - char * filename; + char *filename; filename = getname(name); error = PTR_ERR(filename); @@ -721,55 +748,55 @@ asmlinkage long sys_vfork(struct pt_regs *regs) unsigned long get_wchan(struct task_struct *p) { unsigned long stack; - u64 fp,ip; + u64 fp, ip; int count = 0; - if (!p || p == current || p->state==TASK_RUNNING) - return 0; + if (!p || p == current || p->state == TASK_RUNNING) + return 0; stack = (unsigned long)task_stack_page(p); if (p->thread.sp < stack || p->thread.sp > stack+THREAD_SIZE) return 0; fp = *(u64 *)(p->thread.sp); - do { + do { if (fp < (unsigned long)stack || fp > (unsigned long)stack+THREAD_SIZE) - return 0; + return 0; ip = *(u64 *)(fp+8); if (!in_sched_functions(ip)) return ip; - fp = *(u64 *)fp; - } while (count++ < 16); + fp = *(u64 *)fp; + } while (count++ < 16); return 0; } long do_arch_prctl(struct task_struct *task, int code, unsigned long addr) -{ - int ret = 0; +{ + int ret = 0; int doit = task == current; int cpu; - switch (code) { + switch (code) { case ARCH_SET_GS: if (addr >= TASK_SIZE_OF(task)) - return -EPERM; + return -EPERM; cpu = get_cpu(); - /* handle small bases via the GDT because that's faster to + /* handle small bases via the GDT because that's faster to switch. */ - if (addr <= 0xffffffff) { - set_32bit_tls(task, GS_TLS, addr); - if (doit) { + if (addr <= 0xffffffff) { + set_32bit_tls(task, GS_TLS, addr); + if (doit) { load_TLS(&task->thread, cpu); - load_gs_index(GS_TLS_SEL); + load_gs_index(GS_TLS_SEL); } - task->thread.gsindex = GS_TLS_SEL; + task->thread.gsindex = GS_TLS_SEL; task->thread.gs = 0; - } else { + } else { task->thread.gsindex = 0; task->thread.gs = addr; if (doit) { load_gs_index(0); ret = checking_wrmsrl(MSR_KERNEL_GS_BASE, addr); - } + } } put_cpu(); break; @@ -823,8 +850,7 @@ long do_arch_prctl(struct task_struct *task, int code, unsigned long addr) rdmsrl(MSR_KERNEL_GS_BASE, base); else base = task->thread.gs; - } - else + } else base = task->thread.gs; ret = put_user(base, (unsigned long __user *)addr); break; diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index e37dccc..e375b65 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -14,6 +14,7 @@ #include <linux/errno.h> #include <linux/ptrace.h> #include <linux/regset.h> +#include <linux/tracehook.h> #include <linux/user.h> #include <linux/elf.h> #include <linux/security.h> @@ -69,7 +70,7 @@ static inline bool invalid_selector(u16 value) #define FLAG_MASK FLAG_MASK_32 -static long *pt_regs_access(struct pt_regs *regs, unsigned long regno) +static unsigned long *pt_regs_access(struct pt_regs *regs, unsigned long regno) { BUILD_BUG_ON(offsetof(struct pt_regs, bx) != 0); regno >>= 2; @@ -554,45 +555,115 @@ static int ptrace_set_debugreg(struct task_struct *child, return 0; } -#ifdef X86_BTS +#ifdef CONFIG_X86_PTRACE_BTS +/* + * The configuration for a particular BTS hardware implementation. + */ +struct bts_configuration { + /* the size of a BTS record in bytes; at most BTS_MAX_RECORD_SIZE */ + unsigned char sizeof_bts; + /* the size of a field in the BTS record in bytes */ + unsigned char sizeof_field; + /* a bitmask to enable/disable BTS in DEBUGCTL MSR */ + unsigned long debugctl_mask; +}; +static struct bts_configuration bts_cfg; + +#define BTS_MAX_RECORD_SIZE (8 * 3) + + +/* + * Branch Trace Store (BTS) uses the following format. Different + * architectures vary in the size of those fields. + * - source linear address + * - destination linear address + * - flags + * + * Later architectures use 64bit pointers throughout, whereas earlier + * architectures use 32bit pointers in 32bit mode. + * + * We compute the base address for the first 8 fields based on: + * - the field size stored in the DS configuration + * - the relative field position + * + * In order to store additional information in the BTS buffer, we use + * a special source address to indicate that the record requires + * special interpretation. + * + * Netburst indicated via a bit in the flags field whether the branch + * was predicted; this is ignored. + */ + +enum bts_field { + bts_from = 0, + bts_to, + bts_flags, + + bts_escape = (unsigned long)-1, + bts_qual = bts_to, + bts_jiffies = bts_flags +}; + +static inline unsigned long bts_get(const char *base, enum bts_field field) +{ + base += (bts_cfg.sizeof_field * field); + return *(unsigned long *)base; +} -static int ptrace_bts_get_size(struct task_struct *child) +static inline void bts_set(char *base, enum bts_field field, unsigned long val) { - if (!child->thread.ds_area_msr) - return -ENXIO; + base += (bts_cfg.sizeof_field * field);; + (*(unsigned long *)base) = val; +} - return ds_get_bts_index((void *)child->thread.ds_area_msr); +/* + * Translate a BTS record from the raw format into the bts_struct format + * + * out (out): bts_struct interpretation + * raw: raw BTS record + */ +static void ptrace_bts_translate_record(struct bts_struct *out, const void *raw) +{ + memset(out, 0, sizeof(*out)); + if (bts_get(raw, bts_from) == bts_escape) { + out->qualifier = bts_get(raw, bts_qual); + out->variant.jiffies = bts_get(raw, bts_jiffies); + } else { + out->qualifier = BTS_BRANCH; + out->variant.lbr.from_ip = bts_get(raw, bts_from); + out->variant.lbr.to_ip = bts_get(raw, bts_to); + } } -static int ptrace_bts_read_record(struct task_struct *child, - long index, +static int ptrace_bts_read_record(struct task_struct *child, size_t index, struct bts_struct __user *out) { struct bts_struct ret; - int retval; - int bts_end; - int bts_index; - - if (!child->thread.ds_area_msr) - return -ENXIO; + const void *bts_record; + size_t bts_index, bts_end; + int error; - if (index < 0) - return -EINVAL; + error = ds_get_bts_end(child, &bts_end); + if (error < 0) + return error; - bts_end = ds_get_bts_end((void *)child->thread.ds_area_msr); if (bts_end <= index) return -EINVAL; + error = ds_get_bts_index(child, &bts_index); + if (error < 0) + return error; + /* translate the ptrace bts index into the ds bts index */ - bts_index = ds_get_bts_index((void *)child->thread.ds_area_msr); - bts_index -= (index + 1); - if (bts_index < 0) - bts_index += bts_end; + bts_index += bts_end - (index + 1); + if (bts_end <= bts_index) + bts_index -= bts_end; - retval = ds_read_bts((void *)child->thread.ds_area_msr, - bts_index, &ret); - if (retval < 0) - return retval; + error = ds_access_bts(child, bts_index, &bts_record); + if (error < 0) + return error; + + ptrace_bts_translate_record(&ret, bts_record); if (copy_to_user(out, &ret, sizeof(ret))) return -EFAULT; @@ -600,101 +671,106 @@ static int ptrace_bts_read_record(struct task_struct *child, return sizeof(ret); } -static int ptrace_bts_clear(struct task_struct *child) -{ - if (!child->thread.ds_area_msr) - return -ENXIO; - - return ds_clear((void *)child->thread.ds_area_msr); -} - static int ptrace_bts_drain(struct task_struct *child, long size, struct bts_struct __user *out) { - int end, i; - void *ds = (void *)child->thread.ds_area_msr; - - if (!ds) - return -ENXIO; + struct bts_struct ret; + const unsigned char *raw; + size_t end, i; + int error; - end = ds_get_bts_index(ds); - if (end <= 0) - return end; + error = ds_get_bts_index(child, &end); + if (error < 0) + return error; if (size < (end * sizeof(struct bts_struct))) return -EIO; - for (i = 0; i < end; i++, out++) { - struct bts_struct ret; - int retval; + error = ds_access_bts(child, 0, (const void **)&raw); + if (error < 0) + return error; - retval = ds_read_bts(ds, i, &ret); - if (retval < 0) - return retval; + for (i = 0; i < end; i++, out++, raw += bts_cfg.sizeof_bts) { + ptrace_bts_translate_record(&ret, raw); if (copy_to_user(out, &ret, sizeof(ret))) return -EFAULT; } - ds_clear(ds); + error = ds_clear_bts(child); + if (error < 0) + return error; return end; } +static void ptrace_bts_ovfl(struct task_struct *child) +{ + send_sig(child->thread.bts_ovfl_signal, child, 0); +} + static int ptrace_bts_config(struct task_struct *child, long cfg_size, const struct ptrace_bts_config __user *ucfg) { struct ptrace_bts_config cfg; - int bts_size, ret = 0; - void *ds; + int error = 0; + + error = -EOPNOTSUPP; + if (!bts_cfg.sizeof_bts) + goto errout; + error = -EIO; if (cfg_size < sizeof(cfg)) - return -EIO; + goto errout; + error = -EFAULT; if (copy_from_user(&cfg, ucfg, sizeof(cfg))) - return -EFAULT; + goto errout; - if ((int)cfg.size < 0) - return -EINVAL; + error = -EINVAL; + if ((cfg.flags & PTRACE_BTS_O_SIGNAL) && + !(cfg.flags & PTRACE_BTS_O_ALLOC)) + goto errout; - bts_size = 0; - ds = (void *)child->thread.ds_area_msr; - if (ds) { - bts_size = ds_get_bts_size(ds); - if (bts_size < 0) - return bts_size; - } - cfg.size = PAGE_ALIGN(cfg.size); + if (cfg.flags & PTRACE_BTS_O_ALLOC) { + ds_ovfl_callback_t ovfl = NULL; + unsigned int sig = 0; + + /* we ignore the error in case we were not tracing child */ + (void)ds_release_bts(child); - if (bts_size != cfg.size) { - ret = ptrace_bts_realloc(child, cfg.size, - cfg.flags & PTRACE_BTS_O_CUT_SIZE); - if (ret < 0) + if (cfg.flags & PTRACE_BTS_O_SIGNAL) { + if (!cfg.signal) + goto errout; + + sig = cfg.signal; + ovfl = ptrace_bts_ovfl; + } + + error = ds_request_bts(child, /* base = */ NULL, cfg.size, ovfl); + if (error < 0) goto errout; - ds = (void *)child->thread.ds_area_msr; + child->thread.bts_ovfl_signal = sig; } - if (cfg.flags & PTRACE_BTS_O_SIGNAL) - ret = ds_set_overflow(ds, DS_O_SIGNAL); - else - ret = ds_set_overflow(ds, DS_O_WRAP); - if (ret < 0) + error = -EINVAL; + if (!child->thread.ds_ctx && cfg.flags) goto errout; if (cfg.flags & PTRACE_BTS_O_TRACE) - child->thread.debugctlmsr |= ds_debugctl_mask(); + child->thread.debugctlmsr |= bts_cfg.debugctl_mask; else - child->thread.debugctlmsr &= ~ds_debugctl_mask(); + child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask; if (cfg.flags & PTRACE_BTS_O_SCHED) set_tsk_thread_flag(child, TIF_BTS_TRACE_TS); else clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS); - ret = sizeof(cfg); + error = sizeof(cfg); out: if (child->thread.debugctlmsr) @@ -702,10 +778,10 @@ out: else clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR); - return ret; + return error; errout: - child->thread.debugctlmsr &= ~ds_debugctl_mask(); + child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask; clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS); goto out; } @@ -714,29 +790,40 @@ static int ptrace_bts_status(struct task_struct *child, long cfg_size, struct ptrace_bts_config __user *ucfg) { - void *ds = (void *)child->thread.ds_area_msr; struct ptrace_bts_config cfg; + size_t end; + const void *base, *max; + int error; if (cfg_size < sizeof(cfg)) return -EIO; - memset(&cfg, 0, sizeof(cfg)); + error = ds_get_bts_end(child, &end); + if (error < 0) + return error; - if (ds) { - cfg.size = ds_get_bts_size(ds); + error = ds_access_bts(child, /* index = */ 0, &base); + if (error < 0) + return error; - if (ds_get_overflow(ds) == DS_O_SIGNAL) - cfg.flags |= PTRACE_BTS_O_SIGNAL; + error = ds_access_bts(child, /* index = */ end, &max); + if (error < 0) + return error; - if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) && - child->thread.debugctlmsr & ds_debugctl_mask()) - cfg.flags |= PTRACE_BTS_O_TRACE; + memset(&cfg, 0, sizeof(cfg)); + cfg.size = (max - base); + cfg.signal = child->thread.bts_ovfl_signal; + cfg.bts_size = sizeof(struct bts_struct); - if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS)) - cfg.flags |= PTRACE_BTS_O_SCHED; - } + if (cfg.signal) + cfg.flags |= PTRACE_BTS_O_SIGNAL; - cfg.bts_size = sizeof(struct bts_struct); + if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) && + child->thread.debugctlmsr & bts_cfg.debugctl_mask) + cfg.flags |= PTRACE_BTS_O_TRACE; + + if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS)) + cfg.flags |= PTRACE_BTS_O_SCHED; if (copy_to_user(ucfg, &cfg, sizeof(cfg))) return -EFAULT; @@ -744,89 +831,38 @@ static int ptrace_bts_status(struct task_struct *child, return sizeof(cfg); } - static int ptrace_bts_write_record(struct task_struct *child, const struct bts_struct *in) { - int retval; + unsigned char bts_record[BTS_MAX_RECORD_SIZE]; - if (!child->thread.ds_area_msr) - return -ENXIO; + BUG_ON(BTS_MAX_RECORD_SIZE < bts_cfg.sizeof_bts); - retval = ds_write_bts((void *)child->thread.ds_area_msr, in); - if (retval) - return retval; + memset(bts_record, 0, bts_cfg.sizeof_bts); + switch (in->qualifier) { + case BTS_INVALID: + break; - return sizeof(*in); -} + case BTS_BRANCH: + bts_set(bts_record, bts_from, in->variant.lbr.from_ip); + bts_set(bts_record, bts_to, in->variant.lbr.to_ip); + break; -static int ptrace_bts_realloc(struct task_struct *child, - int size, int reduce_size) -{ - unsigned long rlim, vm; - int ret, old_size; + case BTS_TASK_ARRIVES: + case BTS_TASK_DEPARTS: + bts_set(bts_record, bts_from, bts_escape); + bts_set(bts_record, bts_qual, in->qualifier); + bts_set(bts_record, bts_jiffies, in->variant.jiffies); + break; - if (size < 0) + default: return -EINVAL; - - old_size = ds_get_bts_size((void *)child->thread.ds_area_msr); - if (old_size < 0) - return old_size; - - ret = ds_free((void **)&child->thread.ds_area_msr); - if (ret < 0) - goto out; - - size >>= PAGE_SHIFT; - old_size >>= PAGE_SHIFT; - - current->mm->total_vm -= old_size; - current->mm->locked_vm -= old_size; - - if (size == 0) - goto out; - - rlim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT; - vm = current->mm->total_vm + size; - if (rlim < vm) { - ret = -ENOMEM; - - if (!reduce_size) - goto out; - - size = rlim - current->mm->total_vm; - if (size <= 0) - goto out; - } - - rlim = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT; - vm = current->mm->locked_vm + size; - if (rlim < vm) { - ret = -ENOMEM; - - if (!reduce_size) - goto out; - - size = rlim - current->mm->locked_vm; - if (size <= 0) - goto out; } - ret = ds_allocate((void **)&child->thread.ds_area_msr, - size << PAGE_SHIFT); - if (ret < 0) - goto out; - - current->mm->total_vm += size; - current->mm->locked_vm += size; - -out: - if (child->thread.ds_area_msr) - set_tsk_thread_flag(child, TIF_DS_AREA_MSR); - else - clear_tsk_thread_flag(child, TIF_DS_AREA_MSR); - - return ret; + /* The writing task will be the switched-to task on a context + * switch. It needs to write into the switched-from task's BTS + * buffer. */ + return ds_unchecked_write_bts(child, bts_record, bts_cfg.sizeof_bts); } void ptrace_bts_take_timestamp(struct task_struct *tsk, @@ -839,7 +875,66 @@ void ptrace_bts_take_timestamp(struct task_struct *tsk, ptrace_bts_write_record(tsk, &rec); } -#endif /* X86_BTS */ + +static const struct bts_configuration bts_cfg_netburst = { + .sizeof_bts = sizeof(long) * 3, + .sizeof_field = sizeof(long), + .debugctl_mask = (1<<2)|(1<<3)|(1<<5) +}; + +static const struct bts_configuration bts_cfg_pentium_m = { + .sizeof_bts = sizeof(long) * 3, + .sizeof_field = sizeof(long), + .debugctl_mask = (1<<6)|(1<<7) +}; + +static const struct bts_configuration bts_cfg_core2 = { + .sizeof_bts = 8 * 3, + .sizeof_field = 8, + .debugctl_mask = (1<<6)|(1<<7)|(1<<9) +}; + +static inline void bts_configure(const struct bts_configuration *cfg) +{ + bts_cfg = *cfg; +} + +void __cpuinit ptrace_bts_init_intel(struct cpuinfo_x86 *c) +{ + switch (c->x86) { + case 0x6: + switch (c->x86_model) { + case 0xD: + case 0xE: /* Pentium M */ + bts_configure(&bts_cfg_pentium_m); + break; + case 0xF: /* Core2 */ + case 0x1C: /* Atom */ + bts_configure(&bts_cfg_core2); + break; + default: + /* sorry, don't know about them */ + break; + } + break; + case 0xF: + switch (c->x86_model) { + case 0x0: + case 0x1: + case 0x2: /* Netburst */ + bts_configure(&bts_cfg_netburst); + break; + default: + /* sorry, don't know about them */ + break; + } + break; + default: + /* sorry, don't know about them */ + break; + } +} +#endif /* CONFIG_X86_PTRACE_BTS */ /* * Called by kernel/ptrace.c when detaching.. @@ -852,15 +947,15 @@ void ptrace_disable(struct task_struct *child) #ifdef TIF_SYSCALL_EMU clear_tsk_thread_flag(child, TIF_SYSCALL_EMU); #endif - if (child->thread.ds_area_msr) { -#ifdef X86_BTS - ptrace_bts_realloc(child, 0, 0); -#endif - child->thread.debugctlmsr &= ~ds_debugctl_mask(); - if (!child->thread.debugctlmsr) - clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR); - clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS); - } +#ifdef CONFIG_X86_PTRACE_BTS + (void)ds_release_bts(child); + + child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask; + if (!child->thread.debugctlmsr) + clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR); + + clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS); +#endif /* CONFIG_X86_PTRACE_BTS */ } #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION @@ -980,7 +1075,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) /* * These bits need more cooking - not enabled yet: */ -#ifdef X86_BTS +#ifdef CONFIG_X86_PTRACE_BTS case PTRACE_BTS_CONFIG: ret = ptrace_bts_config (child, data, (struct ptrace_bts_config __user *)addr); @@ -992,7 +1087,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) break; case PTRACE_BTS_SIZE: - ret = ptrace_bts_get_size(child); + ret = ds_get_bts_index(child, /* pos = */ NULL); break; case PTRACE_BTS_GET: @@ -1001,14 +1096,14 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) break; case PTRACE_BTS_CLEAR: - ret = ptrace_bts_clear(child); + ret = ds_clear_bts(child); break; case PTRACE_BTS_DRAIN: ret = ptrace_bts_drain (child, data, (struct bts_struct __user *) addr); break; -#endif +#endif /* CONFIG_X86_PTRACE_BTS */ default: ret = ptrace_request(child, request, addr, data); @@ -1375,30 +1470,6 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code) force_sig_info(SIGTRAP, &info, tsk); } -static void syscall_trace(struct pt_regs *regs) -{ - if (!(current->ptrace & PT_PTRACED)) - return; - -#if 0 - printk("trace %s ip %lx sp %lx ax %d origrax %d caller %lx tiflags %x ptrace %x\n", - current->comm, - regs->ip, regs->sp, regs->ax, regs->orig_ax, __builtin_return_address(0), - current_thread_info()->flags, current->ptrace); -#endif - - ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD) - ? 0x80 : 0)); - /* - * this isn't the same as continuing with a signal, but it will do - * for normal use. strace only continues with a signal if the - * stopping signal is not SIGTRAP. -brl - */ - if (current->exit_code) { - send_sig(current->exit_code, current, 1); - current->exit_code = 0; - } -} #ifdef CONFIG_X86_32 # define IS_IA32 1 @@ -1432,8 +1503,9 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs) if (unlikely(test_thread_flag(TIF_SYSCALL_EMU))) ret = -1L; - if (ret || test_thread_flag(TIF_SYSCALL_TRACE)) - syscall_trace(regs); + if ((ret || test_thread_flag(TIF_SYSCALL_TRACE)) && + tracehook_report_syscall_entry(regs)) + ret = -1L; if (unlikely(current->audit_context)) { if (IS_IA32) @@ -1459,7 +1531,7 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs) audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax); if (test_thread_flag(TIF_SYSCALL_TRACE)) - syscall_trace(regs); + tracehook_report_syscall_exit(regs, 0); /* * If TIF_SYSCALL_EMU is set, we only get here because of @@ -1475,6 +1547,6 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs) * system call instruction. */ if (test_thread_flag(TIF_SINGLESTEP) && - (current->ptrace & PT_PTRACED)) + tracehook_consider_fatal_signal(current, SIGTRAP, SIG_DFL)) send_sigtrap(current, regs, 0); } diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 724adfc..f4c93f1 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -29,7 +29,11 @@ EXPORT_SYMBOL(pm_power_off); static const struct desc_ptr no_idt = {}; static int reboot_mode; -enum reboot_type reboot_type = BOOT_KBD; +/* + * Keyboard reset and triple fault may result in INIT, not RESET, which + * doesn't work when we're in vmx root mode. Try ACPI first. + */ +enum reboot_type reboot_type = BOOT_ACPI; int reboot_force; #if defined(CONFIG_X86_32) && defined(CONFIG_SMP) diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 362d4e7..141efab 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -223,6 +223,9 @@ unsigned long saved_video_mode; #define RAMDISK_LOAD_FLAG 0x4000 static char __initdata command_line[COMMAND_LINE_SIZE]; +#ifdef CONFIG_CMDLINE_BOOL +static char __initdata builtin_cmdline[COMMAND_LINE_SIZE] = CONFIG_CMDLINE; +#endif #if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE) struct edd edd; @@ -665,11 +668,28 @@ void __init setup_arch(char **cmdline_p) bss_resource.start = virt_to_phys(&__bss_start); bss_resource.end = virt_to_phys(&__bss_stop)-1; +#ifdef CONFIG_CMDLINE_BOOL +#ifdef CONFIG_CMDLINE_OVERRIDE + strlcpy(boot_command_line, builtin_cmdline, COMMAND_LINE_SIZE); +#else + if (builtin_cmdline[0]) { + /* append boot loader cmdline to builtin */ + strlcat(builtin_cmdline, " ", COMMAND_LINE_SIZE); + strlcat(builtin_cmdline, boot_command_line, COMMAND_LINE_SIZE); + strlcpy(boot_command_line, builtin_cmdline, COMMAND_LINE_SIZE); + } +#endif +#endif + strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE); *cmdline_p = command_line; parse_early_param(); +#ifdef CONFIG_X86_64 + check_efer(); +#endif + #if defined(CONFIG_VMI) && defined(CONFIG_X86_32) /* * Must be before kernel pagetables are setup @@ -738,7 +758,6 @@ void __init setup_arch(char **cmdline_p) #else num_physpages = max_pfn; - check_efer(); /* How many end-of-memory variables you have, grandma! */ /* need this before calling reserve_initrd */ diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 76e305e..0e67f72 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -162,9 +162,16 @@ void __init setup_per_cpu_areas(void) printk(KERN_INFO "cpu %d has no node %d or node-local memory\n", cpu, node); + if (ptr) + printk(KERN_DEBUG "per cpu data for cpu%d at %016lx\n", + cpu, __pa(ptr)); } - else + else { ptr = alloc_bootmem_pages_node(NODE_DATA(node), size); + if (ptr) + printk(KERN_DEBUG "per cpu data for cpu%d on node%d at %016lx\n", + cpu, node, __pa(ptr)); + } #endif per_cpu_offset(cpu) = ptr - __per_cpu_start; memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start); diff --git a/arch/x86/kernel/sigframe.h b/arch/x86/kernel/sigframe.h index 72bbb51..8b4956e 100644 --- a/arch/x86/kernel/sigframe.h +++ b/arch/x86/kernel/sigframe.h @@ -24,4 +24,9 @@ struct rt_sigframe { struct ucontext uc; struct siginfo info; }; + +int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, + sigset_t *set, struct pt_regs *regs); +int ia32_setup_frame(int sig, struct k_sigaction *ka, + sigset_t *set, struct pt_regs *regs); #endif diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c index 6fb5bcd..2a2435d 100644 --- a/arch/x86/kernel/signal_32.c +++ b/arch/x86/kernel/signal_32.c @@ -17,6 +17,7 @@ #include <linux/errno.h> #include <linux/sched.h> #include <linux/wait.h> +#include <linux/tracehook.h> #include <linux/elf.h> #include <linux/smp.h> #include <linux/mm.h> @@ -26,6 +27,7 @@ #include <asm/uaccess.h> #include <asm/i387.h> #include <asm/vdso.h> +#include <asm/syscalls.h> #include "sigframe.h" @@ -558,8 +560,6 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, * handler too. */ regs->flags &= ~X86_EFLAGS_TF; - if (test_thread_flag(TIF_SINGLESTEP)) - ptrace_notify(SIGTRAP); spin_lock_irq(¤t->sighand->siglock); sigorsets(¤t->blocked, ¤t->blocked, &ka->sa.sa_mask); @@ -568,6 +568,9 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); + tracehook_signal_handler(sig, info, ka, regs, + test_thread_flag(TIF_SINGLESTEP)); + return 0; } @@ -661,5 +664,10 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) if (thread_info_flags & _TIF_SIGPENDING) do_signal(regs); + if (thread_info_flags & _TIF_NOTIFY_RESUME) { + clear_thread_flag(TIF_NOTIFY_RESUME); + tracehook_notify_resume(regs); + } + clear_thread_flag(TIF_IRET); } diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c index ca316b5..694aa88 100644 --- a/arch/x86/kernel/signal_64.c +++ b/arch/x86/kernel/signal_64.c @@ -15,17 +15,21 @@ #include <linux/errno.h> #include <linux/wait.h> #include <linux/ptrace.h> +#include <linux/tracehook.h> #include <linux/unistd.h> #include <linux/stddef.h> #include <linux/personality.h> #include <linux/compiler.h> +#include <linux/uaccess.h> + #include <asm/processor.h> #include <asm/ucontext.h> -#include <asm/uaccess.h> #include <asm/i387.h> #include <asm/proto.h> #include <asm/ia32_unistd.h> #include <asm/mce.h> +#include <asm/syscall.h> +#include <asm/syscalls.h> #include "sigframe.h" #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) @@ -41,11 +45,6 @@ # define FIX_EFLAGS __FIX_EFLAGS #endif -int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, - sigset_t *set, struct pt_regs * regs); -int ia32_setup_frame(int sig, struct k_sigaction *ka, - sigset_t *set, struct pt_regs * regs); - asmlinkage long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, struct pt_regs *regs) @@ -128,7 +127,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, /* Always make any pending restarted system calls return -EINTR */ current_thread_info()->restart_block.fn = do_no_restart_syscall; -#define COPY(x) err |= __get_user(regs->x, &sc->x) +#define COPY(x) (err |= __get_user(regs->x, &sc->x)) COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx); COPY(dx); COPY(cx); COPY(ip); @@ -158,7 +157,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, } { - struct _fpstate __user * buf; + struct _fpstate __user *buf; err |= __get_user(buf, &sc->fpstate); if (buf) { @@ -198,7 +197,7 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) current->blocked = set; recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - + if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax)) goto badframe; @@ -208,16 +207,17 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) return ax; badframe: - signal_fault(regs,frame,"sigreturn"); + signal_fault(regs, frame, "sigreturn"); return 0; -} +} /* * Set up a signal frame. */ static inline int -setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, unsigned long mask, struct task_struct *me) +setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, + unsigned long mask, struct task_struct *me) { int err = 0; @@ -273,35 +273,35 @@ get_stack(struct k_sigaction *ka, struct pt_regs *regs, unsigned long size) } static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, - sigset_t *set, struct pt_regs * regs) + sigset_t *set, struct pt_regs *regs) { struct rt_sigframe __user *frame; - struct _fpstate __user *fp = NULL; + struct _fpstate __user *fp = NULL; int err = 0; struct task_struct *me = current; if (used_math()) { - fp = get_stack(ka, regs, sizeof(struct _fpstate)); + fp = get_stack(ka, regs, sizeof(struct _fpstate)); frame = (void __user *)round_down( (unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8; if (!access_ok(VERIFY_WRITE, fp, sizeof(struct _fpstate))) goto give_sigsegv; - if (save_i387(fp) < 0) - err |= -1; + if (save_i387(fp) < 0) + err |= -1; } else frame = get_stack(ka, regs, sizeof(struct rt_sigframe)) - 8; if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) goto give_sigsegv; - if (ka->sa.sa_flags & SA_SIGINFO) { + if (ka->sa.sa_flags & SA_SIGINFO) { err |= copy_siginfo_to_user(&frame->info, info); if (err) goto give_sigsegv; } - + /* Create the ucontext. */ err |= __put_user(0, &frame->uc.uc_flags); err |= __put_user(0, &frame->uc.uc_link); @@ -311,9 +311,9 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size); err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, set->sig[0], me); err |= __put_user(fp, &frame->uc.uc_mcontext.fpstate); - if (sizeof(*set) == 16) { + if (sizeof(*set) == 16) { __put_user(set->sig[0], &frame->uc.uc_sigmask.sig[0]); - __put_user(set->sig[1], &frame->uc.uc_sigmask.sig[1]); + __put_user(set->sig[1], &frame->uc.uc_sigmask.sig[1]); } else err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); @@ -324,7 +324,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, err |= __put_user(ka->sa.sa_restorer, &frame->pretcode); } else { /* could use a vstub here */ - goto give_sigsegv; + goto give_sigsegv; } if (err) @@ -332,7 +332,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, /* Set up registers for signal handler */ regs->di = sig; - /* In case the signal handler was declared without prototypes */ + /* In case the signal handler was declared without prototypes */ regs->ax = 0; /* This also works for non SA_SIGINFO handlers because they expect the @@ -355,37 +355,8 @@ give_sigsegv: } /* - * Return -1L or the syscall number that @regs is executing. - */ -static long current_syscall(struct pt_regs *regs) -{ - /* - * We always sign-extend a -1 value being set here, - * so this is always either -1L or a syscall number. - */ - return regs->orig_ax; -} - -/* - * Return a value that is -EFOO if the system call in @regs->orig_ax - * returned an error. This only works for @regs from @current. - */ -static long current_syscall_ret(struct pt_regs *regs) -{ -#ifdef CONFIG_IA32_EMULATION - if (test_thread_flag(TIF_IA32)) - /* - * Sign-extend the value so (int)-EFOO becomes (long)-EFOO - * and will match correctly in comparisons. - */ - return (int) regs->ax; -#endif - return regs->ax; -} - -/* * OK, we're invoking a handler - */ + */ static int handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, @@ -394,9 +365,9 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, int ret; /* Are we from a system call? */ - if (current_syscall(regs) >= 0) { + if (syscall_get_nr(current, regs) >= 0) { /* If so, check system call restarting.. */ - switch (current_syscall_ret(regs)) { + switch (syscall_get_error(current, regs)) { case -ERESTART_RESTARTBLOCK: case -ERESTARTNOHAND: regs->ax = -EINTR; @@ -429,7 +400,7 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, ret = ia32_setup_rt_frame(sig, ka, info, oldset, regs); else ret = ia32_setup_frame(sig, ka, oldset, regs); - } else + } else #endif ret = setup_rt_frame(sig, ka, info, oldset, regs); @@ -453,15 +424,16 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, * handler too. */ regs->flags &= ~X86_EFLAGS_TF; - if (test_thread_flag(TIF_SINGLESTEP)) - ptrace_notify(SIGTRAP); spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked,¤t->blocked,&ka->sa.sa_mask); + sigorsets(¤t->blocked, ¤t->blocked, &ka->sa.sa_mask); if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked,sig); + sigaddset(¤t->blocked, sig); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); + + tracehook_signal_handler(sig, info, ka, regs, + test_thread_flag(TIF_SINGLESTEP)); } return ret; @@ -518,9 +490,9 @@ static void do_signal(struct pt_regs *regs) } /* Did we come from a system call? */ - if (current_syscall(regs) >= 0) { + if (syscall_get_nr(current, regs) >= 0) { /* Restart the system call - no handlers present */ - switch (current_syscall_ret(regs)) { + switch (syscall_get_error(current, regs)) { case -ERESTARTNOHAND: case -ERESTARTSYS: case -ERESTARTNOINTR: @@ -558,17 +530,23 @@ void do_notify_resume(struct pt_regs *regs, void *unused, /* deal with pending signal delivery */ if (thread_info_flags & _TIF_SIGPENDING) do_signal(regs); + + if (thread_info_flags & _TIF_NOTIFY_RESUME) { + clear_thread_flag(TIF_NOTIFY_RESUME); + tracehook_notify_resume(regs); + } } void signal_fault(struct pt_regs *regs, void __user *frame, char *where) -{ - struct task_struct *me = current; +{ + struct task_struct *me = current; if (show_unhandled_signals && printk_ratelimit()) { printk("%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx", - me->comm,me->pid,where,frame,regs->ip,regs->sp,regs->orig_ax); + me->comm, me->pid, where, frame, regs->ip, + regs->sp, regs->orig_ax); print_vma_addr(" in ", regs->ip); printk("\n"); } - force_sig(SIGSEGV, me); -} + force_sig(SIGSEGV, me); +} diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 7985c5b..45531e3 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -88,7 +88,7 @@ static DEFINE_PER_CPU(struct task_struct *, idle_thread_array); #define get_idle_for_cpu(x) (per_cpu(idle_thread_array, x)) #define set_idle_for_cpu(x, p) (per_cpu(idle_thread_array, x) = (p)) #else -struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ; +static struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ; #define get_idle_for_cpu(x) (idle_thread_array[(x)]) #define set_idle_for_cpu(x, p) (idle_thread_array[(x)] = (p)) #endif @@ -129,7 +129,7 @@ static int boot_cpu_logical_apicid; static cpumask_t cpu_sibling_setup_map; /* Set if we find a B stepping CPU */ -int __cpuinitdata smp_b_stepping; +static int __cpuinitdata smp_b_stepping; #if defined(CONFIG_NUMA) && defined(CONFIG_X86_32) @@ -1313,16 +1313,13 @@ __init void prefill_possible_map(void) if (!num_processors) num_processors = 1; -#ifdef CONFIG_HOTPLUG_CPU if (additional_cpus == -1) { if (disabled_cpus > 0) additional_cpus = disabled_cpus; else additional_cpus = 0; } -#else - additional_cpus = 0; -#endif + possible = num_processors + additional_cpus; if (possible > NR_CPUS) possible = NR_CPUS; diff --git a/arch/x86/kernel/sys_i386_32.c b/arch/x86/kernel/sys_i386_32.c index 7066cb8..1884a8d 100644 --- a/arch/x86/kernel/sys_i386_32.c +++ b/arch/x86/kernel/sys_i386_32.c @@ -22,6 +22,8 @@ #include <linux/uaccess.h> #include <linux/unistd.h> +#include <asm/syscalls.h> + asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c index 3b360ef..6bc211a 100644 --- a/arch/x86/kernel/sys_x86_64.c +++ b/arch/x86/kernel/sys_x86_64.c @@ -13,15 +13,17 @@ #include <linux/utsname.h> #include <linux/personality.h> #include <linux/random.h> +#include <linux/uaccess.h> -#include <asm/uaccess.h> #include <asm/ia32.h> +#include <asm/syscalls.h> -asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, - unsigned long fd, unsigned long off) +asmlinkage long sys_mmap(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long off) { long error; - struct file * file; + struct file *file; error = -EINVAL; if (off & ~PAGE_MASK) @@ -56,9 +58,9 @@ static void find_start_end(unsigned long flags, unsigned long *begin, unmapped base down for this case. This can give conflicts with the heap, but we assume that glibc malloc knows how to fall back to mmap. Give it 1GB - of playground for now. -AK */ - *begin = 0x40000000; - *end = 0x80000000; + of playground for now. -AK */ + *begin = 0x40000000; + *end = 0x80000000; if (current->flags & PF_RANDOMIZE) { new_begin = randomize_range(*begin, *begin + 0x02000000, 0); if (new_begin) @@ -66,9 +68,9 @@ static void find_start_end(unsigned long flags, unsigned long *begin, } } else { *begin = TASK_UNMAPPED_BASE; - *end = TASK_SIZE; + *end = TASK_SIZE; } -} +} unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, @@ -78,11 +80,11 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, struct vm_area_struct *vma; unsigned long start_addr; unsigned long begin, end; - + if (flags & MAP_FIXED) return addr; - find_start_end(flags, &begin, &end); + find_start_end(flags, &begin, &end); if (len > end) return -ENOMEM; @@ -96,12 +98,12 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, } if (((flags & MAP_32BIT) || test_thread_flag(TIF_IA32)) && len <= mm->cached_hole_size) { - mm->cached_hole_size = 0; + mm->cached_hole_size = 0; mm->free_area_cache = begin; } addr = mm->free_area_cache; - if (addr < begin) - addr = begin; + if (addr < begin) + addr = begin; start_addr = addr; full_search: @@ -127,7 +129,7 @@ full_search: return addr; } if (addr + mm->cached_hole_size < vma->vm_start) - mm->cached_hole_size = vma->vm_start - addr; + mm->cached_hole_size = vma->vm_start - addr; addr = vma->vm_end; } @@ -177,7 +179,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, vma = find_vma(mm, addr-len); if (!vma || addr <= vma->vm_start) /* remember the address as a hint for next time */ - return (mm->free_area_cache = addr-len); + return mm->free_area_cache = addr-len; } if (mm->mmap_base < len) @@ -194,7 +196,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, vma = find_vma(mm, addr); if (!vma || addr+len <= vma->vm_start) /* remember the address as a hint for next time */ - return (mm->free_area_cache = addr); + return mm->free_area_cache = addr; /* remember the largest hole we saw so far */ if (addr + mm->cached_hole_size < vma->vm_start) @@ -224,13 +226,13 @@ bottomup: } -asmlinkage long sys_uname(struct new_utsname __user * name) +asmlinkage long sys_uname(struct new_utsname __user *name) { int err; down_read(&uts_sem); - err = copy_to_user(name, utsname(), sizeof (*name)); + err = copy_to_user(name, utsname(), sizeof(*name)); up_read(&uts_sem); - if (personality(current->personality) == PER_LINUX32) - err |= copy_to_user(&name->machine, "i686", 5); + if (personality(current->personality) == PER_LINUX32) + err |= copy_to_user(&name->machine, "i686", 5); return err ? -EFAULT : 0; } diff --git a/arch/x86/kernel/syscall_64.c b/arch/x86/kernel/syscall_64.c index 170d43c..3d1be4f 100644 --- a/arch/x86/kernel/syscall_64.c +++ b/arch/x86/kernel/syscall_64.c @@ -8,12 +8,12 @@ #define __NO_STUBS #define __SYSCALL(nr, sym) extern asmlinkage void sym(void) ; -#undef _ASM_X86_64_UNISTD_H_ +#undef ASM_X86__UNISTD_64_H #include <asm/unistd_64.h> #undef __SYSCALL #define __SYSCALL(nr, sym) [nr] = sym, -#undef _ASM_X86_64_UNISTD_H_ +#undef ASM_X86__UNISTD_64_H typedef void (*sys_call_ptr_t)(void); diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c index ffe3c66..bbecf8b 100644 --- a/arch/x86/kernel/time_32.c +++ b/arch/x86/kernel/time_32.c @@ -36,6 +36,7 @@ #include <asm/arch_hooks.h> #include <asm/hpet.h> #include <asm/time.h> +#include <asm/timer.h> #include "do_timer.h" diff --git a/arch/x86/kernel/tls.c b/arch/x86/kernel/tls.c index ab6bf37..6bb7b85 100644 --- a/arch/x86/kernel/tls.c +++ b/arch/x86/kernel/tls.c @@ -10,6 +10,7 @@ #include <asm/ldt.h> #include <asm/processor.h> #include <asm/proto.h> +#include <asm/syscalls.h> #include "tls.h" diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index 513caac..7a31f10 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -32,6 +32,8 @@ #include <linux/bug.h> #include <linux/nmi.h> #include <linux/mm.h> +#include <linux/smp.h> +#include <linux/io.h> #if defined(CONFIG_EDAC) #include <linux/edac.h> @@ -45,9 +47,6 @@ #include <asm/unwind.h> #include <asm/desc.h> #include <asm/i387.h> -#include <asm/nmi.h> -#include <asm/smp.h> -#include <asm/io.h> #include <asm/pgalloc.h> #include <asm/proto.h> #include <asm/pda.h> @@ -85,7 +84,8 @@ static inline void preempt_conditional_cli(struct pt_regs *regs) void printk_address(unsigned long address, int reliable) { - printk(" [<%016lx>] %s%pS\n", address, reliable ? "": "? ", (void *) address); + printk(" [<%016lx>] %s%pS\n", + address, reliable ? "" : "? ", (void *) address); } static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, @@ -98,7 +98,8 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, [STACKFAULT_STACK - 1] = "#SS", [MCE_STACK - 1] = "#MC", #if DEBUG_STKSZ > EXCEPTION_STKSZ - [N_EXCEPTION_STACKS ... N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]" + [N_EXCEPTION_STACKS ... + N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]" #endif }; unsigned k; @@ -163,7 +164,7 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, } /* - * x86-64 can have up to three kernel stacks: + * x86-64 can have up to three kernel stacks: * process stack * interrupt stack * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack @@ -219,7 +220,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, const struct stacktrace_ops *ops, void *data) { const unsigned cpu = get_cpu(); - unsigned long *irqstack_end = (unsigned long*)cpu_pda(cpu)->irqstackptr; + unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr; unsigned used = 0; struct thread_info *tinfo; @@ -237,7 +238,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, if (!bp) { if (task == current) { /* Grab bp right from our regs */ - asm("movq %%rbp, %0" : "=r" (bp) :); + asm("movq %%rbp, %0" : "=r" (bp) : ); } else { /* bp is the last reg pushed by switch_to */ bp = *(unsigned long *) task->thread.sp; @@ -339,9 +340,8 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, char *log_lvl) { - printk("\nCall Trace:\n"); + printk("Call Trace:\n"); dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); - printk("\n"); } void show_trace(struct task_struct *task, struct pt_regs *regs, @@ -357,11 +357,15 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, unsigned long *stack; int i; const int cpu = smp_processor_id(); - unsigned long *irqstack_end = (unsigned long *) (cpu_pda(cpu)->irqstackptr); - unsigned long *irqstack = (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE); + unsigned long *irqstack_end = + (unsigned long *) (cpu_pda(cpu)->irqstackptr); + unsigned long *irqstack = + (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE); - // debugging aid: "show_stack(NULL, NULL);" prints the - // back trace for this cpu. + /* + * debugging aid: "show_stack(NULL, NULL);" prints the + * back trace for this cpu. + */ if (sp == NULL) { if (task) @@ -386,6 +390,7 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, printk(" %016lx", *stack++); touch_nmi_watchdog(); } + printk("\n"); show_trace_log_lvl(task, regs, sp, bp, log_lvl); } @@ -404,7 +409,7 @@ void dump_stack(void) #ifdef CONFIG_FRAME_POINTER if (!bp) - asm("movq %%rbp, %0" : "=r" (bp):); + asm("movq %%rbp, %0" : "=r" (bp) : ); #endif printk("Pid: %d, comm: %.20s %s %s %.*s\n", @@ -414,7 +419,6 @@ void dump_stack(void) init_utsname()->version); show_trace(NULL, NULL, &stack, bp); } - EXPORT_SYMBOL(dump_stack); void show_registers(struct pt_regs *regs) @@ -443,7 +447,6 @@ void show_registers(struct pt_regs *regs) printk("Stack: "); show_stack_log_lvl(NULL, regs, (unsigned long *)sp, regs->bp, ""); - printk("\n"); printk(KERN_EMERG "Code: "); @@ -493,7 +496,7 @@ unsigned __kprobes long oops_begin(void) raw_local_irq_save(flags); cpu = smp_processor_id(); if (!__raw_spin_trylock(&die_lock)) { - if (cpu == die_owner) + if (cpu == die_owner) /* nested oops. should stop eventually */; else __raw_spin_lock(&die_lock); @@ -638,7 +641,7 @@ kernel_trap: } #define DO_ERROR(trapnr, signr, str, name) \ -asmlinkage void do_##name(struct pt_regs * regs, long error_code) \ +asmlinkage void do_##name(struct pt_regs *regs, long error_code) \ { \ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ == NOTIFY_STOP) \ @@ -648,7 +651,7 @@ asmlinkage void do_##name(struct pt_regs * regs, long error_code) \ } #define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ -asmlinkage void do_##name(struct pt_regs * regs, long error_code) \ +asmlinkage void do_##name(struct pt_regs *regs, long error_code) \ { \ siginfo_t info; \ info.si_signo = signr; \ @@ -683,7 +686,7 @@ asmlinkage void do_stack_segment(struct pt_regs *regs, long error_code) preempt_conditional_cli(regs); } -asmlinkage void do_double_fault(struct pt_regs * regs, long error_code) +asmlinkage void do_double_fault(struct pt_regs *regs, long error_code) { static const char str[] = "double fault"; struct task_struct *tsk = current; @@ -778,9 +781,10 @@ io_check_error(unsigned char reason, struct pt_regs *regs) } static notrace __kprobes void -unknown_nmi_error(unsigned char reason, struct pt_regs * regs) +unknown_nmi_error(unsigned char reason, struct pt_regs *regs) { - if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) + if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == + NOTIFY_STOP) return; printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n", reason); @@ -882,7 +886,7 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) else if (user_mode(eregs)) regs = task_pt_regs(current); /* Exception from kernel and interrupts are enabled. Move to - kernel process stack. */ + kernel process stack. */ else if (eregs->flags & X86_EFLAGS_IF) regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs)); if (eregs != regs) @@ -891,7 +895,7 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) } /* runs on IST stack. */ -asmlinkage void __kprobes do_debug(struct pt_regs * regs, +asmlinkage void __kprobes do_debug(struct pt_regs *regs, unsigned long error_code) { struct task_struct *tsk = current; @@ -1035,7 +1039,7 @@ asmlinkage void do_coprocessor_error(struct pt_regs *regs) asmlinkage void bad_intr(void) { - printk("bad interrupt"); + printk("bad interrupt"); } asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs) @@ -1047,7 +1051,7 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs) conditional_sti(regs); if (!user_mode(regs) && - kernel_math_error(regs, "kernel simd math error", 19)) + kernel_math_error(regs, "kernel simd math error", 19)) return; /* @@ -1092,7 +1096,7 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs) force_sig_info(SIGFPE, &info, task); } -asmlinkage void do_spurious_interrupt_bug(struct pt_regs * regs) +asmlinkage void do_spurious_interrupt_bug(struct pt_regs *regs) { } @@ -1149,8 +1153,10 @@ void __init trap_init(void) set_intr_gate(0, ÷_error); set_intr_gate_ist(1, &debug, DEBUG_STACK); set_intr_gate_ist(2, &nmi, NMI_STACK); - set_system_gate_ist(3, &int3, DEBUG_STACK); /* int3 can be called from all */ - set_system_gate(4, &overflow); /* int4 can be called from all */ + /* int3 can be called from all */ + set_system_gate_ist(3, &int3, DEBUG_STACK); + /* int4 can be called from all */ + set_system_gate(4, &overflow); set_intr_gate(5, &bounds); set_intr_gate(6, &invalid_op); set_intr_gate(7, &device_not_available); diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 8f98e9d..161bb85 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -104,7 +104,7 @@ __setup("notsc", notsc_setup); /* * Read TSC and the reference counters. Take care of SMI disturbance */ -static u64 tsc_read_refs(u64 *pm, u64 *hpet) +static u64 tsc_read_refs(u64 *p, int hpet) { u64 t1, t2; int i; @@ -112,9 +112,9 @@ static u64 tsc_read_refs(u64 *pm, u64 *hpet) for (i = 0; i < MAX_RETRIES; i++) { t1 = get_cycles(); if (hpet) - *hpet = hpet_readl(HPET_COUNTER) & 0xFFFFFFFF; + *p = hpet_readl(HPET_COUNTER) & 0xFFFFFFFF; else - *pm = acpi_pm_read_early(); + *p = acpi_pm_read_early(); t2 = get_cycles(); if ((t2 - t1) < SMI_TRESHOLD) return t2; @@ -123,13 +123,59 @@ static u64 tsc_read_refs(u64 *pm, u64 *hpet) } /* + * Calculate the TSC frequency from HPET reference + */ +static unsigned long calc_hpet_ref(u64 deltatsc, u64 hpet1, u64 hpet2) +{ + u64 tmp; + + if (hpet2 < hpet1) + hpet2 += 0x100000000ULL; + hpet2 -= hpet1; + tmp = ((u64)hpet2 * hpet_readl(HPET_PERIOD)); + do_div(tmp, 1000000); + do_div(deltatsc, tmp); + + return (unsigned long) deltatsc; +} + +/* + * Calculate the TSC frequency from PMTimer reference + */ +static unsigned long calc_pmtimer_ref(u64 deltatsc, u64 pm1, u64 pm2) +{ + u64 tmp; + + if (!pm1 && !pm2) + return ULONG_MAX; + + if (pm2 < pm1) + pm2 += (u64)ACPI_PM_OVRRUN; + pm2 -= pm1; + tmp = pm2 * 1000000000LL; + do_div(tmp, PMTMR_TICKS_PER_SEC); + do_div(deltatsc, tmp); + + return (unsigned long) deltatsc; +} + +#define CAL_MS 10 +#define CAL_LATCH (CLOCK_TICK_RATE / (1000 / CAL_MS)) +#define CAL_PIT_LOOPS 1000 + +#define CAL2_MS 50 +#define CAL2_LATCH (CLOCK_TICK_RATE / (1000 / CAL2_MS)) +#define CAL2_PIT_LOOPS 5000 + + +/* * Try to calibrate the TSC against the Programmable * Interrupt Timer and return the frequency of the TSC * in kHz. * * Return ULONG_MAX on failure to calibrate. */ -static unsigned long pit_calibrate_tsc(void) +static unsigned long pit_calibrate_tsc(u32 latch, unsigned long ms, int loopmin) { u64 tsc, t1, t2, delta; unsigned long tscmin, tscmax; @@ -144,8 +190,8 @@ static unsigned long pit_calibrate_tsc(void) * (LSB then MSB) to begin countdown. */ outb(0xb0, 0x43); - outb((CLOCK_TICK_RATE / (1000 / 50)) & 0xff, 0x42); - outb((CLOCK_TICK_RATE / (1000 / 50)) >> 8, 0x42); + outb(latch & 0xff, 0x42); + outb(latch >> 8, 0x42); tsc = t1 = t2 = get_cycles(); @@ -166,31 +212,154 @@ static unsigned long pit_calibrate_tsc(void) /* * Sanity checks: * - * If we were not able to read the PIT more than 5000 + * If we were not able to read the PIT more than loopmin * times, then we have been hit by a massive SMI * * If the maximum is 10 times larger than the minimum, * then we got hit by an SMI as well. */ - if (pitcnt < 5000 || tscmax > 10 * tscmin) + if (pitcnt < loopmin || tscmax > 10 * tscmin) return ULONG_MAX; /* Calculate the PIT value */ delta = t2 - t1; - do_div(delta, 50); + do_div(delta, ms); return delta; } +/* + * This reads the current MSB of the PIT counter, and + * checks if we are running on sufficiently fast and + * non-virtualized hardware. + * + * Our expectations are: + * + * - the PIT is running at roughly 1.19MHz + * + * - each IO is going to take about 1us on real hardware, + * but we allow it to be much faster (by a factor of 10) or + * _slightly_ slower (ie we allow up to a 2us read+counter + * update - anything else implies a unacceptably slow CPU + * or PIT for the fast calibration to work. + * + * - with 256 PIT ticks to read the value, we have 214us to + * see the same MSB (and overhead like doing a single TSC + * read per MSB value etc). + * + * - We're doing 2 reads per loop (LSB, MSB), and we expect + * them each to take about a microsecond on real hardware. + * So we expect a count value of around 100. But we'll be + * generous, and accept anything over 50. + * + * - if the PIT is stuck, and we see *many* more reads, we + * return early (and the next caller of pit_expect_msb() + * then consider it a failure when they don't see the + * next expected value). + * + * These expectations mean that we know that we have seen the + * transition from one expected value to another with a fairly + * high accuracy, and we didn't miss any events. We can thus + * use the TSC value at the transitions to calculate a pretty + * good value for the TSC frequencty. + */ +static inline int pit_expect_msb(unsigned char val) +{ + int count = 0; + + for (count = 0; count < 50000; count++) { + /* Ignore LSB */ + inb(0x42); + if (inb(0x42) != val) + break; + } + return count > 50; +} + +/* + * How many MSB values do we want to see? We aim for a + * 15ms calibration, which assuming a 2us counter read + * error should give us roughly 150 ppm precision for + * the calibration. + */ +#define QUICK_PIT_MS 15 +#define QUICK_PIT_ITERATIONS (QUICK_PIT_MS * PIT_TICK_RATE / 1000 / 256) + +static unsigned long quick_pit_calibrate(void) +{ + /* Set the Gate high, disable speaker */ + outb((inb(0x61) & ~0x02) | 0x01, 0x61); + + /* + * Counter 2, mode 0 (one-shot), binary count + * + * NOTE! Mode 2 decrements by two (and then the + * output is flipped each time, giving the same + * final output frequency as a decrement-by-one), + * so mode 0 is much better when looking at the + * individual counts. + */ + outb(0xb0, 0x43); + + /* Start at 0xffff */ + outb(0xff, 0x42); + outb(0xff, 0x42); + + if (pit_expect_msb(0xff)) { + int i; + u64 t1, t2, delta; + unsigned char expect = 0xfe; + + t1 = get_cycles(); + for (i = 0; i < QUICK_PIT_ITERATIONS; i++, expect--) { + if (!pit_expect_msb(expect)) + goto failed; + } + t2 = get_cycles(); + + /* + * Make sure we can rely on the second TSC timestamp: + */ + if (!pit_expect_msb(expect)) + goto failed; + + /* + * Ok, if we get here, then we've seen the + * MSB of the PIT decrement QUICK_PIT_ITERATIONS + * times, and each MSB had many hits, so we never + * had any sudden jumps. + * + * As a result, we can depend on there not being + * any odd delays anywhere, and the TSC reads are + * reliable. + * + * kHz = ticks / time-in-seconds / 1000; + * kHz = (t2 - t1) / (QPI * 256 / PIT_TICK_RATE) / 1000 + * kHz = ((t2 - t1) * PIT_TICK_RATE) / (QPI * 256 * 1000) + */ + delta = (t2 - t1)*PIT_TICK_RATE; + do_div(delta, QUICK_PIT_ITERATIONS*256*1000); + printk("Fast TSC calibration using PIT\n"); + return delta; + } +failed: + return 0; +} /** * native_calibrate_tsc - calibrate the tsc on boot */ unsigned long native_calibrate_tsc(void) { - u64 tsc1, tsc2, delta, pm1, pm2, hpet1, hpet2; + u64 tsc1, tsc2, delta, ref1, ref2; unsigned long tsc_pit_min = ULONG_MAX, tsc_ref_min = ULONG_MAX; - unsigned long flags; - int hpet = is_hpet_enabled(), i; + unsigned long flags, latch, ms, fast_calibrate; + int hpet = is_hpet_enabled(), i, loopmin; + + local_irq_save(flags); + fast_calibrate = quick_pit_calibrate(); + local_irq_restore(flags); + if (fast_calibrate) + return fast_calibrate; /* * Run 5 calibration loops to get the lowest frequency value @@ -216,7 +385,13 @@ unsigned long native_calibrate_tsc(void) * calibration delay loop as we have to wait for a certain * amount of time anyway. */ - for (i = 0; i < 5; i++) { + + /* Preset PIT loop values */ + latch = CAL_LATCH; + ms = CAL_MS; + loopmin = CAL_PIT_LOOPS; + + for (i = 0; i < 3; i++) { unsigned long tsc_pit_khz; /* @@ -226,16 +401,16 @@ unsigned long native_calibrate_tsc(void) * read the end value. */ local_irq_save(flags); - tsc1 = tsc_read_refs(&pm1, hpet ? &hpet1 : NULL); - tsc_pit_khz = pit_calibrate_tsc(); - tsc2 = tsc_read_refs(&pm2, hpet ? &hpet2 : NULL); + tsc1 = tsc_read_refs(&ref1, hpet); + tsc_pit_khz = pit_calibrate_tsc(latch, ms, loopmin); + tsc2 = tsc_read_refs(&ref2, hpet); local_irq_restore(flags); /* Pick the lowest PIT TSC calibration so far */ tsc_pit_min = min(tsc_pit_min, tsc_pit_khz); /* hpet or pmtimer available ? */ - if (!hpet && !pm1 && !pm2) + if (!hpet && !ref1 && !ref2) continue; /* Check, whether the sampling was disturbed by an SMI */ @@ -243,23 +418,41 @@ unsigned long native_calibrate_tsc(void) continue; tsc2 = (tsc2 - tsc1) * 1000000LL; + if (hpet) + tsc2 = calc_hpet_ref(tsc2, ref1, ref2); + else + tsc2 = calc_pmtimer_ref(tsc2, ref1, ref2); - if (hpet) { - if (hpet2 < hpet1) - hpet2 += 0x100000000ULL; - hpet2 -= hpet1; - tsc1 = ((u64)hpet2 * hpet_readl(HPET_PERIOD)); - do_div(tsc1, 1000000); - } else { - if (pm2 < pm1) - pm2 += (u64)ACPI_PM_OVRRUN; - pm2 -= pm1; - tsc1 = pm2 * 1000000000LL; - do_div(tsc1, PMTMR_TICKS_PER_SEC); + tsc_ref_min = min(tsc_ref_min, (unsigned long) tsc2); + + /* Check the reference deviation */ + delta = ((u64) tsc_pit_min) * 100; + do_div(delta, tsc_ref_min); + + /* + * If both calibration results are inside a 10% window + * then we can be sure, that the calibration + * succeeded. We break out of the loop right away. We + * use the reference value, as it is more precise. + */ + if (delta >= 90 && delta <= 110) { + printk(KERN_INFO + "TSC: PIT calibration matches %s. %d loops\n", + hpet ? "HPET" : "PMTIMER", i + 1); + return tsc_ref_min; } - do_div(tsc2, tsc1); - tsc_ref_min = min(tsc_ref_min, (unsigned long) tsc2); + /* + * Check whether PIT failed more than once. This + * happens in virtualized environments. We need to + * give the virtual PC a slightly longer timeframe for + * the HPET/PMTIMER to make the result precise. + */ + if (i == 1 && tsc_pit_min == ULONG_MAX) { + latch = CAL2_LATCH; + ms = CAL2_MS; + loopmin = CAL2_PIT_LOOPS; + } } /* @@ -270,7 +463,7 @@ unsigned long native_calibrate_tsc(void) printk(KERN_WARNING "TSC: Unable to calibrate against PIT\n"); /* We don't have an alternative source, disable TSC */ - if (!hpet && !pm1 && !pm2) { + if (!hpet && !ref1 && !ref2) { printk("TSC: No reference (HPET/PMTIMER) available\n"); return 0; } @@ -278,7 +471,7 @@ unsigned long native_calibrate_tsc(void) /* The alternative source failed as well, disable TSC */ if (tsc_ref_min == ULONG_MAX) { printk(KERN_WARNING "TSC: HPET/PMTIMER calibration " - "failed due to SMI disturbance.\n"); + "failed.\n"); return 0; } @@ -290,44 +483,25 @@ unsigned long native_calibrate_tsc(void) } /* We don't have an alternative source, use the PIT calibration value */ - if (!hpet && !pm1 && !pm2) { + if (!hpet && !ref1 && !ref2) { printk(KERN_INFO "TSC: Using PIT calibration value\n"); return tsc_pit_min; } /* The alternative source failed, use the PIT calibration value */ if (tsc_ref_min == ULONG_MAX) { - printk(KERN_WARNING "TSC: HPET/PMTIMER calibration failed due " - "to SMI disturbance. Using PIT calibration\n"); + printk(KERN_WARNING "TSC: HPET/PMTIMER calibration failed. " + "Using PIT calibration\n"); return tsc_pit_min; } - /* Check the reference deviation */ - delta = ((u64) tsc_pit_min) * 100; - do_div(delta, tsc_ref_min); - - /* - * If both calibration results are inside a 5% window, the we - * use the lower frequency of those as it is probably the - * closest estimate. - */ - if (delta >= 95 && delta <= 105) { - printk(KERN_INFO "TSC: PIT calibration confirmed by %s.\n", - hpet ? "HPET" : "PMTIMER"); - printk(KERN_INFO "TSC: using %s calibration value\n", - tsc_pit_min <= tsc_ref_min ? "PIT" : - hpet ? "HPET" : "PMTIMER"); - return tsc_pit_min <= tsc_ref_min ? tsc_pit_min : tsc_ref_min; - } - - printk(KERN_WARNING "TSC: PIT calibration deviates from %s: %lu %lu.\n", - hpet ? "HPET" : "PMTIMER", tsc_pit_min, tsc_ref_min); - /* * The calibration values differ too much. In doubt, we use * the PIT value as we know that there are PMTIMERs around - * running at double speed. + * running at double speed. At least we let the user know: */ + printk(KERN_WARNING "TSC: PIT calibration deviates from %s: %lu %lu.\n", + hpet ? "HPET" : "PMTIMER", tsc_pit_min, tsc_ref_min); printk(KERN_INFO "TSC: Using PIT calibration value\n"); return tsc_pit_min; } diff --git a/arch/x86/kernel/visws_quirks.c b/arch/x86/kernel/visws_quirks.c index 594ef47..61a97e6 100644 --- a/arch/x86/kernel/visws_quirks.c +++ b/arch/x86/kernel/visws_quirks.c @@ -25,45 +25,31 @@ #include <asm/visws/cobalt.h> #include <asm/visws/piix4.h> #include <asm/arch_hooks.h> +#include <asm/io_apic.h> #include <asm/fixmap.h> #include <asm/reboot.h> #include <asm/setup.h> #include <asm/e820.h> -#include <asm/smp.h> #include <asm/io.h> #include <mach_ipi.h> #include "mach_apic.h" -#include <linux/init.h> -#include <linux/smp.h> - #include <linux/kernel_stat.h> -#include <linux/interrupt.h> -#include <linux/init.h> -#include <asm/io.h> -#include <asm/apic.h> #include <asm/i8259.h> #include <asm/irq_vectors.h> -#include <asm/visws/cobalt.h> #include <asm/visws/lithium.h> -#include <asm/visws/piix4.h> #include <linux/sched.h> #include <linux/kernel.h> -#include <linux/init.h> #include <linux/pci.h> #include <linux/pci_ids.h> extern int no_broadcast; -#include <asm/io.h> #include <asm/apic.h> -#include <asm/arch_hooks.h> -#include <asm/visws/cobalt.h> -#include <asm/visws/lithium.h> char visws_board_type = -1; char visws_board_rev = -1; diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c index 38f566f..4eeb5cf 100644 --- a/arch/x86/kernel/vm86_32.c +++ b/arch/x86/kernel/vm86_32.c @@ -46,6 +46,7 @@ #include <asm/io.h> #include <asm/tlbflush.h> #include <asm/irq.h> +#include <asm/syscalls.h> /* * Known problems: diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index 6ca515d..8c9ad02 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -235,7 +235,7 @@ static void vmi_write_ldt_entry(struct desc_struct *dt, int entry, const void *desc) { u32 *ldt_entry = (u32 *)desc; - vmi_ops.write_idt_entry(dt, entry, ldt_entry[0], ldt_entry[1]); + vmi_ops.write_ldt_entry(dt, entry, ldt_entry[0], ldt_entry[1]); } static void vmi_load_sp0(struct tss_struct *tss, @@ -393,13 +393,13 @@ static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type) } #endif -static void vmi_allocate_pte(struct mm_struct *mm, u32 pfn) +static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn) { vmi_set_page_type(pfn, VMI_PAGE_L1); vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0); } -static void vmi_allocate_pmd(struct mm_struct *mm, u32 pfn) +static void vmi_allocate_pmd(struct mm_struct *mm, unsigned long pfn) { /* * This call comes in very early, before mem_map is setup. @@ -410,20 +410,20 @@ static void vmi_allocate_pmd(struct mm_struct *mm, u32 pfn) vmi_ops.allocate_page(pfn, VMI_PAGE_L2, 0, 0, 0); } -static void vmi_allocate_pmd_clone(u32 pfn, u32 clonepfn, u32 start, u32 count) +static void vmi_allocate_pmd_clone(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count) { vmi_set_page_type(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE); vmi_check_page_type(clonepfn, VMI_PAGE_L2); vmi_ops.allocate_page(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE, clonepfn, start, count); } -static void vmi_release_pte(u32 pfn) +static void vmi_release_pte(unsigned long pfn) { vmi_ops.release_page(pfn, VMI_PAGE_L1); vmi_set_page_type(pfn, VMI_PAGE_NORMAL); } -static void vmi_release_pmd(u32 pfn) +static void vmi_release_pmd(unsigned long pfn) { vmi_ops.release_page(pfn, VMI_PAGE_L2); vmi_set_page_type(pfn, VMI_PAGE_NORMAL); diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c index 0c029e8..7766d36 100644 --- a/arch/x86/kernel/vsmp_64.c +++ b/arch/x86/kernel/vsmp_64.c @@ -61,7 +61,7 @@ static void vsmp_irq_enable(void) native_restore_fl((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC)); } -static unsigned __init vsmp_patch(u8 type, u16 clobbers, void *ibuf, +static unsigned __init_or_module vsmp_patch(u8 type, u16 clobbers, void *ibuf, unsigned long addr, unsigned len) { switch (type) { diff --git a/arch/x86/lib/msr-on-cpu.c b/arch/x86/lib/msr-on-cpu.c index 01b868b..321cf72 100644 --- a/arch/x86/lib/msr-on-cpu.c +++ b/arch/x86/lib/msr-on-cpu.c @@ -16,37 +16,46 @@ static void __rdmsr_on_cpu(void *info) rdmsr(rv->msr_no, rv->l, rv->h); } -static void __rdmsr_safe_on_cpu(void *info) +static void __wrmsr_on_cpu(void *info) { struct msr_info *rv = info; - rv->err = rdmsr_safe(rv->msr_no, &rv->l, &rv->h); + wrmsr(rv->msr_no, rv->l, rv->h); } -static int _rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h, int safe) +int rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h) { - int err = 0; + int err; struct msr_info rv; rv.msr_no = msr_no; - if (safe) { - err = smp_call_function_single(cpu, __rdmsr_safe_on_cpu, - &rv, 1); - err = err ? err : rv.err; - } else { - err = smp_call_function_single(cpu, __rdmsr_on_cpu, &rv, 1); - } + err = smp_call_function_single(cpu, __rdmsr_on_cpu, &rv, 1); *l = rv.l; *h = rv.h; return err; } -static void __wrmsr_on_cpu(void *info) +int wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h) +{ + int err; + struct msr_info rv; + + rv.msr_no = msr_no; + rv.l = l; + rv.h = h; + err = smp_call_function_single(cpu, __wrmsr_on_cpu, &rv, 1); + + return err; +} + +/* These "safe" variants are slower and should be used when the target MSR + may not actually exist. */ +static void __rdmsr_safe_on_cpu(void *info) { struct msr_info *rv = info; - wrmsr(rv->msr_no, rv->l, rv->h); + rv->err = rdmsr_safe(rv->msr_no, &rv->l, &rv->h); } static void __wrmsr_safe_on_cpu(void *info) @@ -56,45 +65,30 @@ static void __wrmsr_safe_on_cpu(void *info) rv->err = wrmsr_safe(rv->msr_no, rv->l, rv->h); } -static int _wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h, int safe) +int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h) { - int err = 0; + int err; struct msr_info rv; rv.msr_no = msr_no; - rv.l = l; - rv.h = h; - if (safe) { - err = smp_call_function_single(cpu, __wrmsr_safe_on_cpu, - &rv, 1); - err = err ? err : rv.err; - } else { - err = smp_call_function_single(cpu, __wrmsr_on_cpu, &rv, 1); - } - - return err; -} + err = smp_call_function_single(cpu, __rdmsr_safe_on_cpu, &rv, 1); + *l = rv.l; + *h = rv.h; -int wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h) -{ - return _wrmsr_on_cpu(cpu, msr_no, l, h, 0); + return err ? err : rv.err; } -int rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h) -{ - return _rdmsr_on_cpu(cpu, msr_no, l, h, 0); -} - -/* These "safe" variants are slower and should be used when the target MSR - may not actually exist. */ int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h) { - return _wrmsr_on_cpu(cpu, msr_no, l, h, 1); -} + int err; + struct msr_info rv; -int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h) -{ - return _rdmsr_on_cpu(cpu, msr_no, l, h, 1); + rv.msr_no = msr_no; + rv.l = l; + rv.h = h; + err = smp_call_function_single(cpu, __wrmsr_safe_on_cpu, &rv, 1); + + return err ? err : rv.err; } EXPORT_SYMBOL(rdmsr_on_cpu); diff --git a/arch/x86/lib/string_32.c b/arch/x86/lib/string_32.c index 94972e7..82004d2 100644 --- a/arch/x86/lib/string_32.c +++ b/arch/x86/lib/string_32.c @@ -22,7 +22,7 @@ char *strcpy(char *dest, const char *src) "testb %%al,%%al\n\t" "jne 1b" : "=&S" (d0), "=&D" (d1), "=&a" (d2) - :"0" (src), "1" (dest) : "memory"); + : "0" (src), "1" (dest) : "memory"); return dest; } EXPORT_SYMBOL(strcpy); @@ -42,7 +42,7 @@ char *strncpy(char *dest, const char *src, size_t count) "stosb\n" "2:" : "=&S" (d0), "=&D" (d1), "=&c" (d2), "=&a" (d3) - :"0" (src), "1" (dest), "2" (count) : "memory"); + : "0" (src), "1" (dest), "2" (count) : "memory"); return dest; } EXPORT_SYMBOL(strncpy); @@ -60,7 +60,7 @@ char *strcat(char *dest, const char *src) "testb %%al,%%al\n\t" "jne 1b" : "=&S" (d0), "=&D" (d1), "=&a" (d2), "=&c" (d3) - : "0" (src), "1" (dest), "2" (0), "3" (0xffffffffu): "memory"); + : "0" (src), "1" (dest), "2" (0), "3" (0xffffffffu) : "memory"); return dest; } EXPORT_SYMBOL(strcat); @@ -105,9 +105,9 @@ int strcmp(const char *cs, const char *ct) "2:\tsbbl %%eax,%%eax\n\t" "orb $1,%%al\n" "3:" - :"=a" (res), "=&S" (d0), "=&D" (d1) - :"1" (cs), "2" (ct) - :"memory"); + : "=a" (res), "=&S" (d0), "=&D" (d1) + : "1" (cs), "2" (ct) + : "memory"); return res; } EXPORT_SYMBOL(strcmp); @@ -130,9 +130,9 @@ int strncmp(const char *cs, const char *ct, size_t count) "3:\tsbbl %%eax,%%eax\n\t" "orb $1,%%al\n" "4:" - :"=a" (res), "=&S" (d0), "=&D" (d1), "=&c" (d2) - :"1" (cs), "2" (ct), "3" (count) - :"memory"); + : "=a" (res), "=&S" (d0), "=&D" (d1), "=&c" (d2) + : "1" (cs), "2" (ct), "3" (count) + : "memory"); return res; } EXPORT_SYMBOL(strncmp); @@ -152,9 +152,9 @@ char *strchr(const char *s, int c) "movl $1,%1\n" "2:\tmovl %1,%0\n\t" "decl %0" - :"=a" (res), "=&S" (d0) - :"1" (s), "0" (c) - :"memory"); + : "=a" (res), "=&S" (d0) + : "1" (s), "0" (c) + : "memory"); return res; } EXPORT_SYMBOL(strchr); @@ -169,9 +169,9 @@ size_t strlen(const char *s) "scasb\n\t" "notl %0\n\t" "decl %0" - :"=c" (res), "=&D" (d0) - :"1" (s), "a" (0), "0" (0xffffffffu) - :"memory"); + : "=c" (res), "=&D" (d0) + : "1" (s), "a" (0), "0" (0xffffffffu) + : "memory"); return res; } EXPORT_SYMBOL(strlen); @@ -189,9 +189,9 @@ void *memchr(const void *cs, int c, size_t count) "je 1f\n\t" "movl $1,%0\n" "1:\tdecl %0" - :"=D" (res), "=&c" (d0) - :"a" (c), "0" (cs), "1" (count) - :"memory"); + : "=D" (res), "=&c" (d0) + : "a" (c), "0" (cs), "1" (count) + : "memory"); return res; } EXPORT_SYMBOL(memchr); @@ -228,9 +228,9 @@ size_t strnlen(const char *s, size_t count) "cmpl $-1,%1\n\t" "jne 1b\n" "3:\tsubl %2,%0" - :"=a" (res), "=&d" (d0) - :"c" (s), "1" (count) - :"memory"); + : "=a" (res), "=&d" (d0) + : "c" (s), "1" (count) + : "memory"); return res; } EXPORT_SYMBOL(strnlen); diff --git a/arch/x86/lib/strstr_32.c b/arch/x86/lib/strstr_32.c index 42e8a50..8e2d55f 100644 --- a/arch/x86/lib/strstr_32.c +++ b/arch/x86/lib/strstr_32.c @@ -23,9 +23,9 @@ __asm__ __volatile__( "jne 1b\n\t" "xorl %%eax,%%eax\n\t" "2:" - :"=a" (__res), "=&c" (d0), "=&S" (d1) - :"0" (0), "1" (0xffffffff), "2" (cs), "g" (ct) - :"dx", "di"); + : "=a" (__res), "=&c" (d0), "=&S" (d1) + : "0" (0), "1" (0xffffffff), "2" (cs), "g" (ct) + : "dx", "di"); return __res; } diff --git a/arch/x86/mach-default/setup.c b/arch/x86/mach-default/setup.c index 3d31783..3f2cf11 100644 --- a/arch/x86/mach-default/setup.c +++ b/arch/x86/mach-default/setup.c @@ -10,13 +10,15 @@ #include <asm/e820.h> #include <asm/setup.h> +#include <mach_ipi.h> + #ifdef CONFIG_HOTPLUG_CPU #define DEFAULT_SEND_IPI (1) #else #define DEFAULT_SEND_IPI (0) #endif -int no_broadcast=DEFAULT_SEND_IPI; +int no_broadcast = DEFAULT_SEND_IPI; /** * pre_intr_init_hook - initialisation prior to setting up interrupt vectors diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c index 62fa440..847c164 100644 --- a/arch/x86/mm/discontig_32.c +++ b/arch/x86/mm/discontig_32.c @@ -328,7 +328,7 @@ void __init initmem_init(unsigned long start_pfn, get_memcfg_numa(); - kva_pages = round_up(calculate_numa_remap_pages(), PTRS_PER_PTE); + kva_pages = roundup(calculate_numa_remap_pages(), PTRS_PER_PTE); kva_target_pfn = round_down(max_low_pfn - kva_pages, PTRS_PER_PTE); do { diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c index a20d1fa..e7277cb 100644 --- a/arch/x86/mm/dump_pagetables.c +++ b/arch/x86/mm/dump_pagetables.c @@ -148,8 +148,8 @@ static void note_page(struct seq_file *m, struct pg_state *st, * we have now. "break" is either changing perms, levels or * address space marker. */ - prot = pgprot_val(new_prot) & ~(PTE_PFN_MASK); - cur = pgprot_val(st->current_prot) & ~(PTE_PFN_MASK); + prot = pgprot_val(new_prot) & PTE_FLAGS_MASK; + cur = pgprot_val(st->current_prot) & PTE_FLAGS_MASK; if (!st->level) { /* First entry */ diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 455f3fe..8f92cac 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -35,6 +35,7 @@ #include <asm/tlbflush.h> #include <asm/proto.h> #include <asm-generic/sections.h> +#include <asm/traps.h> /* * Page fault error code bits @@ -357,8 +358,6 @@ static int is_errata100(struct pt_regs *regs, unsigned long address) return 0; } -void do_invalid_op(struct pt_regs *, unsigned long); - static int is_f00f_bug(struct pt_regs *regs, unsigned long address) { #ifdef CONFIG_X86_F00F_BUG diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index d37f293..c3789bb 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -47,6 +47,7 @@ #include <asm/paravirt.h> #include <asm/setup.h> #include <asm/cacheflush.h> +#include <asm/smp.h> unsigned int __VMALLOC_RESERVE = 128 << 20; @@ -194,11 +195,30 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base, pgd_t *pgd; pmd_t *pmd; pte_t *pte; - unsigned pages_2m = 0, pages_4k = 0; + unsigned pages_2m, pages_4k; + int mapping_iter; + + /* + * First iteration will setup identity mapping using large/small pages + * based on use_pse, with other attributes same as set by + * the early code in head_32.S + * + * Second iteration will setup the appropriate attributes (NX, GLOBAL..) + * as desired for the kernel identity mapping. + * + * This two pass mechanism conforms to the TLB app note which says: + * + * "Software should not write to a paging-structure entry in a way + * that would change, for any linear address, both the page size + * and either the page frame or attributes." + */ + mapping_iter = 1; if (!cpu_has_pse) use_pse = 0; +repeat: + pages_2m = pages_4k = 0; pfn = start_pfn; pgd_idx = pgd_index((pfn<<PAGE_SHIFT) + PAGE_OFFSET); pgd = pgd_base + pgd_idx; @@ -224,6 +244,13 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base, if (use_pse) { unsigned int addr2; pgprot_t prot = PAGE_KERNEL_LARGE; + /* + * first pass will use the same initial + * identity mapping attribute + _PAGE_PSE. + */ + pgprot_t init_prot = + __pgprot(PTE_IDENT_ATTR | + _PAGE_PSE); addr2 = (pfn + PTRS_PER_PTE-1) * PAGE_SIZE + PAGE_OFFSET + PAGE_SIZE-1; @@ -233,7 +260,10 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base, prot = PAGE_KERNEL_LARGE_EXEC; pages_2m++; - set_pmd(pmd, pfn_pmd(pfn, prot)); + if (mapping_iter == 1) + set_pmd(pmd, pfn_pmd(pfn, init_prot)); + else + set_pmd(pmd, pfn_pmd(pfn, prot)); pfn += PTRS_PER_PTE; continue; @@ -245,17 +275,43 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base, for (; pte_ofs < PTRS_PER_PTE && pfn < end_pfn; pte++, pfn++, pte_ofs++, addr += PAGE_SIZE) { pgprot_t prot = PAGE_KERNEL; + /* + * first pass will use the same initial + * identity mapping attribute. + */ + pgprot_t init_prot = __pgprot(PTE_IDENT_ATTR); if (is_kernel_text(addr)) prot = PAGE_KERNEL_EXEC; pages_4k++; - set_pte(pte, pfn_pte(pfn, prot)); + if (mapping_iter == 1) + set_pte(pte, pfn_pte(pfn, init_prot)); + else + set_pte(pte, pfn_pte(pfn, prot)); } } } - update_page_count(PG_LEVEL_2M, pages_2m); - update_page_count(PG_LEVEL_4K, pages_4k); + if (mapping_iter == 1) { + /* + * update direct mapping page count only in the first + * iteration. + */ + update_page_count(PG_LEVEL_2M, pages_2m); + update_page_count(PG_LEVEL_4K, pages_4k); + + /* + * local global flush tlb, which will flush the previous + * mappings present in both small and large page TLB's. + */ + __flush_tlb_all(); + + /* + * Second iteration will set the actual desired PTE attributes. + */ + mapping_iter = 2; + goto repeat; + } } /* @@ -458,11 +514,7 @@ static void __init pagetable_init(void) { pgd_t *pgd_base = swapper_pg_dir; - paravirt_pagetable_setup_start(pgd_base); - permanent_kmaps_init(pgd_base); - - paravirt_pagetable_setup_done(pgd_base); } #ifdef CONFIG_ACPI_SLEEP @@ -722,7 +774,7 @@ void __init setup_bootmem_allocator(void) after_init_bootmem = 1; } -static void __init find_early_table_space(unsigned long end) +static void __init find_early_table_space(unsigned long end, int use_pse) { unsigned long puds, pmds, ptes, tables, start; @@ -732,7 +784,7 @@ static void __init find_early_table_space(unsigned long end) pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT; tables += PAGE_ALIGN(pmds * sizeof(pmd_t)); - if (cpu_has_pse) { + if (use_pse) { unsigned long extra; extra = end - ((end>>PMD_SHIFT) << PMD_SHIFT); @@ -772,12 +824,22 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, pgd_t *pgd_base = swapper_pg_dir; unsigned long start_pfn, end_pfn; unsigned long big_page_start; +#ifdef CONFIG_DEBUG_PAGEALLOC + /* + * For CONFIG_DEBUG_PAGEALLOC, identity mapping will use small pages. + * This will simplify cpa(), which otherwise needs to support splitting + * large pages into small in interrupt context, etc. + */ + int use_pse = 0; +#else + int use_pse = cpu_has_pse; +#endif /* * Find space for the kernel direct mapping tables. */ if (!after_init_bootmem) - find_early_table_space(end); + find_early_table_space(end, use_pse); #ifdef CONFIG_X86_PAE set_nx(); @@ -823,7 +885,7 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, end_pfn = (end>>PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT); if (start_pfn < end_pfn) kernel_physical_mapping_init(pgd_base, start_pfn, end_pfn, - cpu_has_pse); + use_pse); /* tail is not big page alignment ? */ start_pfn = end_pfn; @@ -986,7 +1048,6 @@ void __init mem_init(void) if (boot_cpu_data.wp_works_ok < 0) test_wp_bit(); - cpa_init(); save_pg_dir(); zap_low_mappings(); } diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index d3746ef..fb30486 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -225,7 +225,7 @@ void __init init_extra_mapping_uc(unsigned long phys, unsigned long size) void __init cleanup_highmap(void) { unsigned long vaddr = __START_KERNEL_map; - unsigned long end = round_up((unsigned long)_end, PMD_SIZE) - 1; + unsigned long end = roundup((unsigned long)_end, PMD_SIZE) - 1; pmd_t *pmd = level2_kernel_pgt; pmd_t *last_pmd = pmd + PTRS_PER_PMD; @@ -271,7 +271,8 @@ static __ref void unmap_low_page(void *adr) } static unsigned long __meminit -phys_pte_init(pte_t *pte_page, unsigned long addr, unsigned long end) +phys_pte_init(pte_t *pte_page, unsigned long addr, unsigned long end, + pgprot_t prot) { unsigned pages = 0; unsigned long last_map_addr = end; @@ -289,36 +290,43 @@ phys_pte_init(pte_t *pte_page, unsigned long addr, unsigned long end) break; } + /* + * We will re-use the existing mapping. + * Xen for example has some special requirements, like mapping + * pagetable pages as RO. So assume someone who pre-setup + * these mappings are more intelligent. + */ if (pte_val(*pte)) continue; if (0) printk(" pte=%p addr=%lx pte=%016lx\n", pte, addr, pfn_pte(addr >> PAGE_SHIFT, PAGE_KERNEL).pte); - set_pte(pte, pfn_pte(addr >> PAGE_SHIFT, PAGE_KERNEL)); - last_map_addr = (addr & PAGE_MASK) + PAGE_SIZE; pages++; + set_pte(pte, pfn_pte(addr >> PAGE_SHIFT, prot)); + last_map_addr = (addr & PAGE_MASK) + PAGE_SIZE; } + update_page_count(PG_LEVEL_4K, pages); return last_map_addr; } static unsigned long __meminit -phys_pte_update(pmd_t *pmd, unsigned long address, unsigned long end) +phys_pte_update(pmd_t *pmd, unsigned long address, unsigned long end, + pgprot_t prot) { pte_t *pte = (pte_t *)pmd_page_vaddr(*pmd); - return phys_pte_init(pte, address, end); + return phys_pte_init(pte, address, end, prot); } static unsigned long __meminit phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end, - unsigned long page_size_mask) + unsigned long page_size_mask, pgprot_t prot) { unsigned long pages = 0; unsigned long last_map_addr = end; - unsigned long start = address; int i = pmd_index(address); @@ -326,6 +334,7 @@ phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end, unsigned long pte_phys; pmd_t *pmd = pmd_page + pmd_index(address); pte_t *pte; + pgprot_t new_prot = prot; if (address >= end) { if (!after_bootmem) { @@ -339,27 +348,40 @@ phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end, if (!pmd_large(*pmd)) { spin_lock(&init_mm.page_table_lock); last_map_addr = phys_pte_update(pmd, address, - end); + end, prot); spin_unlock(&init_mm.page_table_lock); + continue; } - /* Count entries we're using from level2_ident_pgt */ - if (start == 0) - pages++; - continue; + /* + * If we are ok with PG_LEVEL_2M mapping, then we will + * use the existing mapping, + * + * Otherwise, we will split the large page mapping but + * use the same existing protection bits except for + * large page, so that we don't violate Intel's TLB + * Application note (317080) which says, while changing + * the page sizes, new and old translations should + * not differ with respect to page frame and + * attributes. + */ + if (page_size_mask & (1 << PG_LEVEL_2M)) + continue; + new_prot = pte_pgprot(pte_clrhuge(*(pte_t *)pmd)); } if (page_size_mask & (1<<PG_LEVEL_2M)) { pages++; spin_lock(&init_mm.page_table_lock); set_pte((pte_t *)pmd, - pfn_pte(address >> PAGE_SHIFT, PAGE_KERNEL_LARGE)); + pfn_pte(address >> PAGE_SHIFT, + __pgprot(pgprot_val(prot) | _PAGE_PSE))); spin_unlock(&init_mm.page_table_lock); last_map_addr = (address & PMD_MASK) + PMD_SIZE; continue; } pte = alloc_low_page(&pte_phys); - last_map_addr = phys_pte_init(pte, address, end); + last_map_addr = phys_pte_init(pte, address, end, new_prot); unmap_low_page(pte); spin_lock(&init_mm.page_table_lock); @@ -372,12 +394,12 @@ phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end, static unsigned long __meminit phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end, - unsigned long page_size_mask) + unsigned long page_size_mask, pgprot_t prot) { pmd_t *pmd = pmd_offset(pud, 0); unsigned long last_map_addr; - last_map_addr = phys_pmd_init(pmd, address, end, page_size_mask); + last_map_addr = phys_pmd_init(pmd, address, end, page_size_mask, prot); __flush_tlb_all(); return last_map_addr; } @@ -394,6 +416,7 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end, unsigned long pmd_phys; pud_t *pud = pud_page + pud_index(addr); pmd_t *pmd; + pgprot_t prot = PAGE_KERNEL; if (addr >= end) break; @@ -405,10 +428,26 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end, } if (pud_val(*pud)) { - if (!pud_large(*pud)) + if (!pud_large(*pud)) { last_map_addr = phys_pmd_update(pud, addr, end, - page_size_mask); - continue; + page_size_mask, prot); + continue; + } + /* + * If we are ok with PG_LEVEL_1G mapping, then we will + * use the existing mapping. + * + * Otherwise, we will split the gbpage mapping but use + * the same existing protection bits except for large + * page, so that we don't violate Intel's TLB + * Application note (317080) which says, while changing + * the page sizes, new and old translations should + * not differ with respect to page frame and + * attributes. + */ + if (page_size_mask & (1 << PG_LEVEL_1G)) + continue; + prot = pte_pgprot(pte_clrhuge(*(pte_t *)pud)); } if (page_size_mask & (1<<PG_LEVEL_1G)) { @@ -422,7 +461,8 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end, } pmd = alloc_low_page(&pmd_phys); - last_map_addr = phys_pmd_init(pmd, addr, end, page_size_mask); + last_map_addr = phys_pmd_init(pmd, addr, end, page_size_mask, + prot); unmap_low_page(pmd); spin_lock(&init_mm.page_table_lock); @@ -430,6 +470,7 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end, spin_unlock(&init_mm.page_table_lock); } __flush_tlb_all(); + update_page_count(PG_LEVEL_1G, pages); return last_map_addr; @@ -446,27 +487,28 @@ phys_pud_update(pgd_t *pgd, unsigned long addr, unsigned long end, return phys_pud_init(pud, addr, end, page_size_mask); } -static void __init find_early_table_space(unsigned long end) +static void __init find_early_table_space(unsigned long end, int use_pse, + int use_gbpages) { unsigned long puds, pmds, ptes, tables, start; puds = (end + PUD_SIZE - 1) >> PUD_SHIFT; - tables = round_up(puds * sizeof(pud_t), PAGE_SIZE); - if (direct_gbpages) { + tables = roundup(puds * sizeof(pud_t), PAGE_SIZE); + if (use_gbpages) { unsigned long extra; extra = end - ((end>>PUD_SHIFT) << PUD_SHIFT); pmds = (extra + PMD_SIZE - 1) >> PMD_SHIFT; } else pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT; - tables += round_up(pmds * sizeof(pmd_t), PAGE_SIZE); + tables += roundup(pmds * sizeof(pmd_t), PAGE_SIZE); - if (cpu_has_pse) { + if (use_pse) { unsigned long extra; extra = end - ((end>>PMD_SHIFT) << PMD_SHIFT); ptes = (extra + PAGE_SIZE - 1) >> PAGE_SHIFT; } else ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT; - tables += round_up(ptes * sizeof(pte_t), PAGE_SIZE); + tables += roundup(ptes * sizeof(pte_t), PAGE_SIZE); /* * RED-PEN putting page tables only on node 0 could @@ -528,6 +570,7 @@ static unsigned long __init kernel_physical_mapping_init(unsigned long start, pgd_populate(&init_mm, pgd, __va(pud_phys)); spin_unlock(&init_mm.page_table_lock); } + __flush_tlb_all(); return last_map_addr; } @@ -571,6 +614,7 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, struct map_range mr[NR_RANGE_MR]; int nr_range, i; + int use_pse, use_gbpages; printk(KERN_INFO "init_memory_mapping\n"); @@ -584,9 +628,21 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, if (!after_bootmem) init_gbpages(); - if (direct_gbpages) +#ifdef CONFIG_DEBUG_PAGEALLOC + /* + * For CONFIG_DEBUG_PAGEALLOC, identity mapping will use small pages. + * This will simplify cpa(), which otherwise needs to support splitting + * large pages into small in interrupt context, etc. + */ + use_pse = use_gbpages = 0; +#else + use_pse = cpu_has_pse; + use_gbpages = direct_gbpages; +#endif + + if (use_gbpages) page_size_mask |= 1 << PG_LEVEL_1G; - if (cpu_has_pse) + if (use_pse) page_size_mask |= 1 << PG_LEVEL_2M; memset(mr, 0, sizeof(mr)); @@ -647,7 +703,7 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, (mr[i].page_size_mask & (1<<PG_LEVEL_2M))?"2M":"4k")); if (!after_bootmem) - find_early_table_space(end); + find_early_table_space(end, use_pse, use_gbpages); for (i = 0; i < nr_range; i++) last_map_addr = kernel_physical_mapping_init( @@ -806,8 +862,6 @@ void __init mem_init(void) reservedpages << (PAGE_SHIFT-10), datasize >> 10, initsize >> 10); - - cpa_init(); } void free_init_pages(char *what, unsigned long begin, unsigned long end) diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index d4b6e6a..6ab3196 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -83,6 +83,25 @@ int page_is_ram(unsigned long pagenr) return 0; } +int pagerange_is_ram(unsigned long start, unsigned long end) +{ + int ram_page = 0, not_rampage = 0; + unsigned long page_nr; + + for (page_nr = (start >> PAGE_SHIFT); page_nr < (end >> PAGE_SHIFT); + ++page_nr) { + if (page_is_ram(page_nr)) + ram_page = 1; + else + not_rampage = 1; + + if (ram_page == not_rampage) + return -1; + } + + return ram_page; +} + /* * Fix up the linear direct mapping of the kernel to avoid cache attribute * conflicts. @@ -421,7 +440,7 @@ void unxlate_dev_mem_ptr(unsigned long phys, void *addr) return; } -int __initdata early_ioremap_debug; +static int __initdata early_ioremap_debug; static int __init early_ioremap_debug_setup(char *str) { @@ -547,7 +566,7 @@ static inline void __init early_clear_fixmap(enum fixed_addresses idx) } -int __initdata early_ioremap_nested; +static int __initdata early_ioremap_nested; static int __init check_early_ioremap_leak(void) { diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index a4dd793..cebcbf1 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -79,7 +79,7 @@ static int __init allocate_cachealigned_memnodemap(void) return 0; addr = 0x8000; - nodemap_size = round_up(sizeof(s16) * memnodemapsize, L1_CACHE_BYTES); + nodemap_size = roundup(sizeof(s16) * memnodemapsize, L1_CACHE_BYTES); nodemap_addr = find_e820_area(addr, max_pfn<<PAGE_SHIFT, nodemap_size, L1_CACHE_BYTES); if (nodemap_addr == -1UL) { @@ -176,10 +176,10 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long start_pfn, last_pfn, bootmap_pages, bootmap_size; unsigned long bootmap_start, nodedata_phys; void *bootmap; - const int pgdat_size = round_up(sizeof(pg_data_t), PAGE_SIZE); + const int pgdat_size = roundup(sizeof(pg_data_t), PAGE_SIZE); int nid; - start = round_up(start, ZONE_ALIGN); + start = roundup(start, ZONE_ALIGN); printk(KERN_INFO "Bootmem setup node %d %016lx-%016lx\n", nodeid, start, end); @@ -210,9 +210,9 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, bootmap_pages = bootmem_bootmap_pages(last_pfn - start_pfn); nid = phys_to_nid(nodedata_phys); if (nid == nodeid) - bootmap_start = round_up(nodedata_phys + pgdat_size, PAGE_SIZE); + bootmap_start = roundup(nodedata_phys + pgdat_size, PAGE_SIZE); else - bootmap_start = round_up(start, PAGE_SIZE); + bootmap_start = roundup(start, PAGE_SIZE); /* * SMP_CACHE_BYTES could be enough, but init_bootmem_node like * to use that to align to PAGE_SIZE diff --git a/arch/x86/mm/pageattr-test.c b/arch/x86/mm/pageattr-test.c index d4aa503..e1d1069 100644 --- a/arch/x86/mm/pageattr-test.c +++ b/arch/x86/mm/pageattr-test.c @@ -32,7 +32,7 @@ enum { GPS = (1<<30) }; -#define PAGE_TESTBIT __pgprot(_PAGE_UNUSED1) +#define PAGE_CPA_TEST __pgprot(_PAGE_CPA_TEST) static int pte_testbit(pte_t pte) { @@ -118,6 +118,7 @@ static int pageattr_test(void) unsigned int level; int i, k; int err; + unsigned long test_addr; if (print) printk(KERN_INFO "CPA self-test:\n"); @@ -172,7 +173,8 @@ static int pageattr_test(void) continue; } - err = change_page_attr_set(addr[i], len[i], PAGE_TESTBIT); + test_addr = addr[i]; + err = change_page_attr_set(&test_addr, len[i], PAGE_CPA_TEST, 0); if (err < 0) { printk(KERN_ERR "CPA %d failed %d\n", i, err); failed++; @@ -204,7 +206,8 @@ static int pageattr_test(void) failed++; continue; } - err = change_page_attr_clear(addr[i], len[i], PAGE_TESTBIT); + test_addr = addr[i]; + err = change_page_attr_clear(&test_addr, len[i], PAGE_CPA_TEST, 0); if (err < 0) { printk(KERN_ERR "CPA reverting failed: %d\n", err); failed++; diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 43e2f84..a9ec89c 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -25,15 +25,27 @@ * The current flushing context - we pass it instead of 5 arguments: */ struct cpa_data { - unsigned long vaddr; + unsigned long *vaddr; pgprot_t mask_set; pgprot_t mask_clr; int numpages; - int flushtlb; + int flags; unsigned long pfn; unsigned force_split : 1; + int curpage; }; +/* + * Serialize cpa() (for !DEBUG_PAGEALLOC which uses large identity mappings) + * using cpa_lock. So that we don't allow any other cpu, with stale large tlb + * entries change the page attribute in parallel to some other cpu + * splitting a large page entry along with changing the attribute. + */ +static DEFINE_SPINLOCK(cpa_lock); + +#define CPA_FLUSHTLB 1 +#define CPA_ARRAY 2 + #ifdef CONFIG_PROC_FS static unsigned long direct_pages_count[PG_LEVEL_NUM]; @@ -84,7 +96,7 @@ static inline unsigned long highmap_start_pfn(void) static inline unsigned long highmap_end_pfn(void) { - return __pa(round_up((unsigned long)_end, PMD_SIZE)) >> PAGE_SHIFT; + return __pa(roundup((unsigned long)_end, PMD_SIZE)) >> PAGE_SHIFT; } #endif @@ -190,6 +202,41 @@ static void cpa_flush_range(unsigned long start, int numpages, int cache) } } +static void cpa_flush_array(unsigned long *start, int numpages, int cache) +{ + unsigned int i, level; + unsigned long *addr; + + BUG_ON(irqs_disabled()); + + on_each_cpu(__cpa_flush_range, NULL, 1); + + if (!cache) + return; + + /* 4M threshold */ + if (numpages >= 1024) { + if (boot_cpu_data.x86_model >= 4) + wbinvd(); + return; + } + /* + * We only need to flush on one CPU, + * clflush is a MESI-coherent instruction that + * will cause all other CPUs to flush the same + * cachelines: + */ + for (i = 0, addr = start; i < numpages; i++, addr++) { + pte_t *pte = lookup_address(*addr, &level); + + /* + * Only flush present addresses: + */ + if (pte && (pte_val(*pte) & _PAGE_PRESENT)) + clflush_cache_range((void *) *addr, PAGE_SIZE); + } +} + /* * Certain areas of memory on x86 require very specific protection flags, * for example the BIOS area or kernel text. Callers don't always get this @@ -398,7 +445,7 @@ try_preserve_large_page(pte_t *kpte, unsigned long address, */ new_pte = pfn_pte(pte_pfn(old_pte), canon_pgprot(new_prot)); __set_pmd_pte(kpte, address, new_pte); - cpa->flushtlb = 1; + cpa->flags |= CPA_FLUSHTLB; do_split = 0; } @@ -408,84 +455,6 @@ out_unlock: return do_split; } -static LIST_HEAD(page_pool); -static unsigned long pool_size, pool_pages, pool_low; -static unsigned long pool_used, pool_failed; - -static void cpa_fill_pool(struct page **ret) -{ - gfp_t gfp = GFP_KERNEL; - unsigned long flags; - struct page *p; - - /* - * Avoid recursion (on debug-pagealloc) and also signal - * our priority to get to these pagetables: - */ - if (current->flags & PF_MEMALLOC) - return; - current->flags |= PF_MEMALLOC; - - /* - * Allocate atomically from atomic contexts: - */ - if (in_atomic() || irqs_disabled() || debug_pagealloc) - gfp = GFP_ATOMIC | __GFP_NORETRY | __GFP_NOWARN; - - while (pool_pages < pool_size || (ret && !*ret)) { - p = alloc_pages(gfp, 0); - if (!p) { - pool_failed++; - break; - } - /* - * If the call site needs a page right now, provide it: - */ - if (ret && !*ret) { - *ret = p; - continue; - } - spin_lock_irqsave(&pgd_lock, flags); - list_add(&p->lru, &page_pool); - pool_pages++; - spin_unlock_irqrestore(&pgd_lock, flags); - } - - current->flags &= ~PF_MEMALLOC; -} - -#define SHIFT_MB (20 - PAGE_SHIFT) -#define ROUND_MB_GB ((1 << 10) - 1) -#define SHIFT_MB_GB 10 -#define POOL_PAGES_PER_GB 16 - -void __init cpa_init(void) -{ - struct sysinfo si; - unsigned long gb; - - si_meminfo(&si); - /* - * Calculate the number of pool pages: - * - * Convert totalram (nr of pages) to MiB and round to the next - * GiB. Shift MiB to Gib and multiply the result by - * POOL_PAGES_PER_GB: - */ - if (debug_pagealloc) { - gb = ((si.totalram >> SHIFT_MB) + ROUND_MB_GB) >> SHIFT_MB_GB; - pool_size = POOL_PAGES_PER_GB * gb; - } else { - pool_size = 1; - } - pool_low = pool_size; - - cpa_fill_pool(NULL); - printk(KERN_DEBUG - "CPA: page pool initialized %lu of %lu pages preallocated\n", - pool_pages, pool_size); -} - static int split_large_page(pte_t *kpte, unsigned long address) { unsigned long flags, pfn, pfninc = 1; @@ -494,28 +463,15 @@ static int split_large_page(pte_t *kpte, unsigned long address) pgprot_t ref_prot; struct page *base; - /* - * Get a page from the pool. The pool list is protected by the - * pgd_lock, which we have to take anyway for the split - * operation: - */ - spin_lock_irqsave(&pgd_lock, flags); - if (list_empty(&page_pool)) { - spin_unlock_irqrestore(&pgd_lock, flags); - base = NULL; - cpa_fill_pool(&base); - if (!base) - return -ENOMEM; - spin_lock_irqsave(&pgd_lock, flags); - } else { - base = list_first_entry(&page_pool, struct page, lru); - list_del(&base->lru); - pool_pages--; - - if (pool_pages < pool_low) - pool_low = pool_pages; - } + if (!debug_pagealloc) + spin_unlock(&cpa_lock); + base = alloc_pages(GFP_KERNEL, 0); + if (!debug_pagealloc) + spin_lock(&cpa_lock); + if (!base) + return -ENOMEM; + spin_lock_irqsave(&pgd_lock, flags); /* * Check for races, another CPU might have split this page * up for us already: @@ -572,11 +528,8 @@ out_unlock: * If we dropped out via the lookup_address check under * pgd_lock then stick the page back into the pool: */ - if (base) { - list_add(&base->lru, &page_pool); - pool_pages++; - } else - pool_used++; + if (base) + __free_page(base); spin_unlock_irqrestore(&pgd_lock, flags); return 0; @@ -584,11 +537,16 @@ out_unlock: static int __change_page_attr(struct cpa_data *cpa, int primary) { - unsigned long address = cpa->vaddr; + unsigned long address; int do_split, err; unsigned int level; pte_t *kpte, old_pte; + if (cpa->flags & CPA_ARRAY) + address = cpa->vaddr[cpa->curpage]; + else + address = *cpa->vaddr; + repeat: kpte = lookup_address(address, &level); if (!kpte) @@ -600,7 +558,7 @@ repeat: return 0; WARN(1, KERN_WARNING "CPA: called for zero pte. " "vaddr = %lx cpa->vaddr = %lx\n", address, - cpa->vaddr); + *cpa->vaddr); return -EINVAL; } @@ -626,7 +584,7 @@ repeat: */ if (pte_val(old_pte) != pte_val(new_pte)) { set_pte_atomic(kpte, new_pte); - cpa->flushtlb = 1; + cpa->flags |= CPA_FLUSHTLB; } cpa->numpages = 1; return 0; @@ -650,7 +608,25 @@ repeat: */ err = split_large_page(kpte, address); if (!err) { - cpa->flushtlb = 1; + /* + * Do a global flush tlb after splitting the large page + * and before we do the actual change page attribute in the PTE. + * + * With out this, we violate the TLB application note, that says + * "The TLBs may contain both ordinary and large-page + * translations for a 4-KByte range of linear addresses. This + * may occur if software modifies the paging structures so that + * the page size used for the address range changes. If the two + * translations differ with respect to page frame or attributes + * (e.g., permissions), processor behavior is undefined and may + * be implementation-specific." + * + * We do this global tlb flush inside the cpa_lock, so that we + * don't allow any other cpu, with stale tlb entries change the + * page attribute in parallel, that also falls into the + * just split large page entry. + */ + flush_tlb_all(); goto repeat; } @@ -663,6 +639,7 @@ static int cpa_process_alias(struct cpa_data *cpa) { struct cpa_data alias_cpa; int ret = 0; + unsigned long temp_cpa_vaddr, vaddr; if (cpa->pfn >= max_pfn_mapped) return 0; @@ -675,16 +652,24 @@ static int cpa_process_alias(struct cpa_data *cpa) * No need to redo, when the primary call touched the direct * mapping already: */ - if (!(within(cpa->vaddr, PAGE_OFFSET, + if (cpa->flags & CPA_ARRAY) + vaddr = cpa->vaddr[cpa->curpage]; + else + vaddr = *cpa->vaddr; + + if (!(within(vaddr, PAGE_OFFSET, PAGE_OFFSET + (max_low_pfn_mapped << PAGE_SHIFT)) #ifdef CONFIG_X86_64 - || within(cpa->vaddr, PAGE_OFFSET + (1UL<<32), + || within(vaddr, PAGE_OFFSET + (1UL<<32), PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT)) #endif )) { alias_cpa = *cpa; - alias_cpa.vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT); + temp_cpa_vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT); + alias_cpa.vaddr = &temp_cpa_vaddr; + alias_cpa.flags &= ~CPA_ARRAY; + ret = __change_page_attr_set_clr(&alias_cpa, 0); } @@ -696,7 +681,7 @@ static int cpa_process_alias(struct cpa_data *cpa) * No need to redo, when the primary call touched the high * mapping already: */ - if (within(cpa->vaddr, (unsigned long) _text, (unsigned long) _end)) + if (within(vaddr, (unsigned long) _text, (unsigned long) _end)) return 0; /* @@ -707,8 +692,9 @@ static int cpa_process_alias(struct cpa_data *cpa) return 0; alias_cpa = *cpa; - alias_cpa.vaddr = - (cpa->pfn << PAGE_SHIFT) + __START_KERNEL_map - phys_base; + temp_cpa_vaddr = (cpa->pfn << PAGE_SHIFT) + __START_KERNEL_map - phys_base; + alias_cpa.vaddr = &temp_cpa_vaddr; + alias_cpa.flags &= ~CPA_ARRAY; /* * The high mapping range is imprecise, so ignore the return value. @@ -728,8 +714,15 @@ static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias) * preservation check. */ cpa->numpages = numpages; + /* for array changes, we can't use large page */ + if (cpa->flags & CPA_ARRAY) + cpa->numpages = 1; + if (!debug_pagealloc) + spin_lock(&cpa_lock); ret = __change_page_attr(cpa, checkalias); + if (!debug_pagealloc) + spin_unlock(&cpa_lock); if (ret) return ret; @@ -746,7 +739,11 @@ static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias) */ BUG_ON(cpa->numpages > numpages); numpages -= cpa->numpages; - cpa->vaddr += cpa->numpages * PAGE_SIZE; + if (cpa->flags & CPA_ARRAY) + cpa->curpage++; + else + *cpa->vaddr += cpa->numpages * PAGE_SIZE; + } return 0; } @@ -757,9 +754,9 @@ static inline int cache_attr(pgprot_t attr) (_PAGE_PAT | _PAGE_PAT_LARGE | _PAGE_PWT | _PAGE_PCD); } -static int change_page_attr_set_clr(unsigned long addr, int numpages, +static int change_page_attr_set_clr(unsigned long *addr, int numpages, pgprot_t mask_set, pgprot_t mask_clr, - int force_split) + int force_split, int array) { struct cpa_data cpa; int ret, cache, checkalias; @@ -774,21 +771,38 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, return 0; /* Ensure we are PAGE_SIZE aligned */ - if (addr & ~PAGE_MASK) { - addr &= PAGE_MASK; - /* - * People should not be passing in unaligned addresses: - */ - WARN_ON_ONCE(1); + if (!array) { + if (*addr & ~PAGE_MASK) { + *addr &= PAGE_MASK; + /* + * People should not be passing in unaligned addresses: + */ + WARN_ON_ONCE(1); + } + } else { + int i; + for (i = 0; i < numpages; i++) { + if (addr[i] & ~PAGE_MASK) { + addr[i] &= PAGE_MASK; + WARN_ON_ONCE(1); + } + } } + /* Must avoid aliasing mappings in the highmem code */ + kmap_flush_unused(); + cpa.vaddr = addr; cpa.numpages = numpages; cpa.mask_set = mask_set; cpa.mask_clr = mask_clr; - cpa.flushtlb = 0; + cpa.flags = 0; + cpa.curpage = 0; cpa.force_split = force_split; + if (array) + cpa.flags |= CPA_ARRAY; + /* No alias checking for _NX bit modifications */ checkalias = (pgprot_val(mask_set) | pgprot_val(mask_clr)) != _PAGE_NX; @@ -797,7 +811,7 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, /* * Check whether we really changed something: */ - if (!cpa.flushtlb) + if (!(cpa.flags & CPA_FLUSHTLB)) goto out; /* @@ -812,27 +826,30 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, * error case we fall back to cpa_flush_all (which uses * wbindv): */ - if (!ret && cpu_has_clflush) - cpa_flush_range(addr, numpages, cache); - else + if (!ret && cpu_has_clflush) { + if (cpa.flags & CPA_ARRAY) + cpa_flush_array(addr, numpages, cache); + else + cpa_flush_range(*addr, numpages, cache); + } else cpa_flush_all(cache); out: - cpa_fill_pool(NULL); - return ret; } -static inline int change_page_attr_set(unsigned long addr, int numpages, - pgprot_t mask) +static inline int change_page_attr_set(unsigned long *addr, int numpages, + pgprot_t mask, int array) { - return change_page_attr_set_clr(addr, numpages, mask, __pgprot(0), 0); + return change_page_attr_set_clr(addr, numpages, mask, __pgprot(0), 0, + array); } -static inline int change_page_attr_clear(unsigned long addr, int numpages, - pgprot_t mask) +static inline int change_page_attr_clear(unsigned long *addr, int numpages, + pgprot_t mask, int array) { - return change_page_attr_set_clr(addr, numpages, __pgprot(0), mask, 0); + return change_page_attr_set_clr(addr, numpages, __pgprot(0), mask, 0, + array); } int _set_memory_uc(unsigned long addr, int numpages) @@ -840,8 +857,8 @@ int _set_memory_uc(unsigned long addr, int numpages) /* * for now UC MINUS. see comments in ioremap_nocache() */ - return change_page_attr_set(addr, numpages, - __pgprot(_PAGE_CACHE_UC_MINUS)); + return change_page_attr_set(&addr, numpages, + __pgprot(_PAGE_CACHE_UC_MINUS), 0); } int set_memory_uc(unsigned long addr, int numpages) @@ -857,10 +874,48 @@ int set_memory_uc(unsigned long addr, int numpages) } EXPORT_SYMBOL(set_memory_uc); +int set_memory_array_uc(unsigned long *addr, int addrinarray) +{ + unsigned long start; + unsigned long end; + int i; + /* + * for now UC MINUS. see comments in ioremap_nocache() + */ + for (i = 0; i < addrinarray; i++) { + start = __pa(addr[i]); + for (end = start + PAGE_SIZE; i < addrinarray - 1; end += PAGE_SIZE) { + if (end != __pa(addr[i + 1])) + break; + i++; + } + if (reserve_memtype(start, end, _PAGE_CACHE_UC_MINUS, NULL)) + goto out; + } + + return change_page_attr_set(addr, addrinarray, + __pgprot(_PAGE_CACHE_UC_MINUS), 1); +out: + for (i = 0; i < addrinarray; i++) { + unsigned long tmp = __pa(addr[i]); + + if (tmp == start) + break; + for (end = tmp + PAGE_SIZE; i < addrinarray - 1; end += PAGE_SIZE) { + if (end != __pa(addr[i + 1])) + break; + i++; + } + free_memtype(tmp, end); + } + return -EINVAL; +} +EXPORT_SYMBOL(set_memory_array_uc); + int _set_memory_wc(unsigned long addr, int numpages) { - return change_page_attr_set(addr, numpages, - __pgprot(_PAGE_CACHE_WC)); + return change_page_attr_set(&addr, numpages, + __pgprot(_PAGE_CACHE_WC), 0); } int set_memory_wc(unsigned long addr, int numpages) @@ -878,8 +933,8 @@ EXPORT_SYMBOL(set_memory_wc); int _set_memory_wb(unsigned long addr, int numpages) { - return change_page_attr_clear(addr, numpages, - __pgprot(_PAGE_CACHE_MASK)); + return change_page_attr_clear(&addr, numpages, + __pgprot(_PAGE_CACHE_MASK), 0); } int set_memory_wb(unsigned long addr, int numpages) @@ -890,37 +945,59 @@ int set_memory_wb(unsigned long addr, int numpages) } EXPORT_SYMBOL(set_memory_wb); +int set_memory_array_wb(unsigned long *addr, int addrinarray) +{ + int i; + + for (i = 0; i < addrinarray; i++) { + unsigned long start = __pa(addr[i]); + unsigned long end; + + for (end = start + PAGE_SIZE; i < addrinarray - 1; end += PAGE_SIZE) { + if (end != __pa(addr[i + 1])) + break; + i++; + } + free_memtype(start, end); + } + return change_page_attr_clear(addr, addrinarray, + __pgprot(_PAGE_CACHE_MASK), 1); +} +EXPORT_SYMBOL(set_memory_array_wb); + int set_memory_x(unsigned long addr, int numpages) { - return change_page_attr_clear(addr, numpages, __pgprot(_PAGE_NX)); + return change_page_attr_clear(&addr, numpages, __pgprot(_PAGE_NX), 0); } EXPORT_SYMBOL(set_memory_x); int set_memory_nx(unsigned long addr, int numpages) { - return change_page_attr_set(addr, numpages, __pgprot(_PAGE_NX)); + return change_page_attr_set(&addr, numpages, __pgprot(_PAGE_NX), 0); } EXPORT_SYMBOL(set_memory_nx); int set_memory_ro(unsigned long addr, int numpages) { - return change_page_attr_clear(addr, numpages, __pgprot(_PAGE_RW)); + return change_page_attr_clear(&addr, numpages, __pgprot(_PAGE_RW), 0); } +EXPORT_SYMBOL_GPL(set_memory_ro); int set_memory_rw(unsigned long addr, int numpages) { - return change_page_attr_set(addr, numpages, __pgprot(_PAGE_RW)); + return change_page_attr_set(&addr, numpages, __pgprot(_PAGE_RW), 0); } +EXPORT_SYMBOL_GPL(set_memory_rw); int set_memory_np(unsigned long addr, int numpages) { - return change_page_attr_clear(addr, numpages, __pgprot(_PAGE_PRESENT)); + return change_page_attr_clear(&addr, numpages, __pgprot(_PAGE_PRESENT), 0); } int set_memory_4k(unsigned long addr, int numpages) { - return change_page_attr_set_clr(addr, numpages, __pgprot(0), - __pgprot(0), 1); + return change_page_attr_set_clr(&addr, numpages, __pgprot(0), + __pgprot(0), 1, 0); } int set_pages_uc(struct page *page, int numpages) @@ -973,22 +1050,38 @@ int set_pages_rw(struct page *page, int numpages) static int __set_pages_p(struct page *page, int numpages) { - struct cpa_data cpa = { .vaddr = (unsigned long) page_address(page), + unsigned long tempaddr = (unsigned long) page_address(page); + struct cpa_data cpa = { .vaddr = &tempaddr, .numpages = numpages, .mask_set = __pgprot(_PAGE_PRESENT | _PAGE_RW), - .mask_clr = __pgprot(0)}; + .mask_clr = __pgprot(0), + .flags = 0}; - return __change_page_attr_set_clr(&cpa, 1); + /* + * No alias checking needed for setting present flag. otherwise, + * we may need to break large pages for 64-bit kernel text + * mappings (this adds to complexity if we want to do this from + * atomic context especially). Let's keep it simple! + */ + return __change_page_attr_set_clr(&cpa, 0); } static int __set_pages_np(struct page *page, int numpages) { - struct cpa_data cpa = { .vaddr = (unsigned long) page_address(page), + unsigned long tempaddr = (unsigned long) page_address(page); + struct cpa_data cpa = { .vaddr = &tempaddr, .numpages = numpages, .mask_set = __pgprot(0), - .mask_clr = __pgprot(_PAGE_PRESENT | _PAGE_RW)}; + .mask_clr = __pgprot(_PAGE_PRESENT | _PAGE_RW), + .flags = 0}; - return __change_page_attr_set_clr(&cpa, 1); + /* + * No alias checking needed for setting not present flag. otherwise, + * we may need to break large pages for 64-bit kernel text + * mappings (this adds to complexity if we want to do this from + * atomic context especially). Let's keep it simple! + */ + return __change_page_attr_set_clr(&cpa, 0); } void kernel_map_pages(struct page *page, int numpages, int enable) @@ -1008,11 +1101,8 @@ void kernel_map_pages(struct page *page, int numpages, int enable) /* * The return value is ignored as the calls cannot fail. - * Large pages are kept enabled at boot time, and are - * split up quickly with DEBUG_PAGEALLOC. If a splitup - * fails here (due to temporary memory shortage) no damage - * is done because we just keep the largepage intact up - * to the next attempt when it will likely be split up: + * Large pages for identity mappings are not used at boot time + * and hence no memory allocations during large page split. */ if (enable) __set_pages_p(page, numpages); @@ -1024,53 +1114,8 @@ void kernel_map_pages(struct page *page, int numpages, int enable) * but that can deadlock->flush only current cpu: */ __flush_tlb_all(); - - /* - * Try to refill the page pool here. We can do this only after - * the tlb flush. - */ - cpa_fill_pool(NULL); } -#ifdef CONFIG_DEBUG_FS -static int dpa_show(struct seq_file *m, void *v) -{ - seq_puts(m, "DEBUG_PAGEALLOC\n"); - seq_printf(m, "pool_size : %lu\n", pool_size); - seq_printf(m, "pool_pages : %lu\n", pool_pages); - seq_printf(m, "pool_low : %lu\n", pool_low); - seq_printf(m, "pool_used : %lu\n", pool_used); - seq_printf(m, "pool_failed : %lu\n", pool_failed); - - return 0; -} - -static int dpa_open(struct inode *inode, struct file *filp) -{ - return single_open(filp, dpa_show, NULL); -} - -static const struct file_operations dpa_fops = { - .open = dpa_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static int __init debug_pagealloc_proc_init(void) -{ - struct dentry *de; - - de = debugfs_create_file("debug_pagealloc", 0600, NULL, NULL, - &dpa_fops); - if (!de) - return -ENOMEM; - - return 0; -} -__initcall(debug_pagealloc_proc_init); -#endif - #ifdef CONFIG_HIBERNATION bool kernel_page_present(struct page *page) diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 2a50e0f..738fd0f2 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -7,24 +7,24 @@ * Loosely based on earlier PAT patchset from Eric Biederman and Andi Kleen. */ -#include <linux/mm.h> +#include <linux/seq_file.h> +#include <linux/bootmem.h> +#include <linux/debugfs.h> #include <linux/kernel.h> #include <linux/gfp.h> +#include <linux/mm.h> #include <linux/fs.h> -#include <linux/bootmem.h> -#include <linux/debugfs.h> -#include <linux/seq_file.h> -#include <asm/msr.h> -#include <asm/tlbflush.h> +#include <asm/cacheflush.h> #include <asm/processor.h> -#include <asm/page.h> +#include <asm/tlbflush.h> #include <asm/pgtable.h> -#include <asm/pat.h> -#include <asm/e820.h> -#include <asm/cacheflush.h> #include <asm/fcntl.h> +#include <asm/e820.h> #include <asm/mtrr.h> +#include <asm/page.h> +#include <asm/msr.h> +#include <asm/pat.h> #include <asm/io.h> #ifdef CONFIG_X86_PAT @@ -46,6 +46,7 @@ early_param("nopat", nopat); static int debug_enable; + static int __init pat_debug_setup(char *str) { debug_enable = 1; @@ -145,14 +146,14 @@ static char *cattr_name(unsigned long flags) */ struct memtype { - u64 start; - u64 end; - unsigned long type; - struct list_head nd; + u64 start; + u64 end; + unsigned long type; + struct list_head nd; }; static LIST_HEAD(memtype_list); -static DEFINE_SPINLOCK(memtype_lock); /* protects memtype list */ +static DEFINE_SPINLOCK(memtype_lock); /* protects memtype list */ /* * Does intersection of PAT memory type and MTRR memory type and returns @@ -180,8 +181,8 @@ static unsigned long pat_x_mtrr_type(u64 start, u64 end, unsigned long req_type) return req_type; } -static int chk_conflict(struct memtype *new, struct memtype *entry, - unsigned long *type) +static int +chk_conflict(struct memtype *new, struct memtype *entry, unsigned long *type) { if (new->type != entry->type) { if (type) { @@ -211,6 +212,66 @@ static struct memtype *cached_entry; static u64 cached_start; /* + * For RAM pages, mark the pages as non WB memory type using + * PageNonWB (PG_arch_1). We allow only one set_memory_uc() or + * set_memory_wc() on a RAM page at a time before marking it as WB again. + * This is ok, because only one driver will be owning the page and + * doing set_memory_*() calls. + * + * For now, we use PageNonWB to track that the RAM page is being mapped + * as non WB. In future, we will have to use one more flag + * (or some other mechanism in page_struct) to distinguish between + * UC and WC mapping. + */ +static int reserve_ram_pages_type(u64 start, u64 end, unsigned long req_type, + unsigned long *new_type) +{ + struct page *page; + u64 pfn, end_pfn; + + for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) { + page = pfn_to_page(pfn); + if (page_mapped(page) || PageNonWB(page)) + goto out; + + SetPageNonWB(page); + } + return 0; + +out: + end_pfn = pfn; + for (pfn = (start >> PAGE_SHIFT); pfn < end_pfn; ++pfn) { + page = pfn_to_page(pfn); + ClearPageNonWB(page); + } + + return -EINVAL; +} + +static int free_ram_pages_type(u64 start, u64 end) +{ + struct page *page; + u64 pfn, end_pfn; + + for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) { + page = pfn_to_page(pfn); + if (page_mapped(page) || !PageNonWB(page)) + goto out; + + ClearPageNonWB(page); + } + return 0; + +out: + end_pfn = pfn; + for (pfn = (start >> PAGE_SHIFT); pfn < end_pfn; ++pfn) { + page = pfn_to_page(pfn); + SetPageNonWB(page); + } + return -EINVAL; +} + +/* * req_type typically has one of the: * - _PAGE_CACHE_WB * - _PAGE_CACHE_WC @@ -226,14 +287,15 @@ static u64 cached_start; * it will return a negative return value. */ int reserve_memtype(u64 start, u64 end, unsigned long req_type, - unsigned long *new_type) + unsigned long *new_type) { struct memtype *new, *entry; unsigned long actual_type; struct list_head *where; + int is_range_ram; int err = 0; - BUG_ON(start >= end); /* end is exclusive */ + BUG_ON(start >= end); /* end is exclusive */ if (!pat_enabled) { /* This is identical to page table setting without PAT */ @@ -266,17 +328,24 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, actual_type = _PAGE_CACHE_WB; else actual_type = _PAGE_CACHE_UC_MINUS; - } else + } else { actual_type = pat_x_mtrr_type(start, end, req_type & _PAGE_CACHE_MASK); + } + + is_range_ram = pagerange_is_ram(start, end); + if (is_range_ram == 1) + return reserve_ram_pages_type(start, end, req_type, new_type); + else if (is_range_ram < 0) + return -EINVAL; new = kmalloc(sizeof(struct memtype), GFP_KERNEL); if (!new) return -ENOMEM; - new->start = start; - new->end = end; - new->type = actual_type; + new->start = start; + new->end = end; + new->type = actual_type; if (new_type) *new_type = actual_type; @@ -335,6 +404,7 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, start, end, cattr_name(new->type), cattr_name(req_type)); kfree(new); spin_unlock(&memtype_lock); + return err; } @@ -358,6 +428,7 @@ int free_memtype(u64 start, u64 end) { struct memtype *entry; int err = -EINVAL; + int is_range_ram; if (!pat_enabled) return 0; @@ -366,6 +437,12 @@ int free_memtype(u64 start, u64 end) if (is_ISA_range(start, end - 1)) return 0; + is_range_ram = pagerange_is_ram(start, end); + if (is_range_ram == 1) + return free_ram_pages_type(start, end); + else if (is_range_ram < 0) + return -EINVAL; + spin_lock(&memtype_lock); list_for_each_entry(entry, &memtype_list, nd) { if (entry->start == start && entry->end == end) { @@ -386,6 +463,7 @@ int free_memtype(u64 start, u64 end) } dprintk("free_memtype request 0x%Lx-0x%Lx\n", start, end); + return err; } @@ -492,9 +570,9 @@ int phys_mem_access_prot_allowed(struct file *file, unsigned long pfn, void map_devmem(unsigned long pfn, unsigned long size, pgprot_t vma_prot) { + unsigned long want_flags = (pgprot_val(vma_prot) & _PAGE_CACHE_MASK); u64 addr = (u64)pfn << PAGE_SHIFT; unsigned long flags; - unsigned long want_flags = (pgprot_val(vma_prot) & _PAGE_CACHE_MASK); reserve_memtype(addr, addr + size, want_flags, &flags); if (flags != want_flags) { @@ -514,7 +592,7 @@ void unmap_devmem(unsigned long pfn, unsigned long size, pgprot_t vma_prot) free_memtype(addr, addr + size); } -#if defined(CONFIG_DEBUG_FS) +#if defined(CONFIG_DEBUG_FS) && defined(CONFIG_X86_PAT) /* get Nth element of the linked list */ static struct memtype *memtype_get_idx(loff_t pos) @@ -537,6 +615,7 @@ static struct memtype *memtype_get_idx(loff_t pos) } spin_unlock(&memtype_lock); kfree(print_entry); + return NULL; } @@ -567,6 +646,7 @@ static int memtype_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "%s @ 0x%Lx-0x%Lx\n", cattr_name(print_entry->type), print_entry->start, print_entry->end); kfree(print_entry); + return 0; } @@ -598,4 +678,4 @@ static int __init pat_memtype_list_init(void) late_initcall(pat_memtype_list_init); -#endif /* CONFIG_DEBUG_FS */ +#endif /* CONFIG_DEBUG_FS && CONFIG_X86_PAT */ diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index d503027..86f2ffc 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -63,10 +63,8 @@ static inline void pgd_list_del(pgd_t *pgd) #define UNSHARED_PTRS_PER_PGD \ (SHARED_KERNEL_PMD ? KERNEL_PGD_BOUNDARY : PTRS_PER_PGD) -static void pgd_ctor(void *p) +static void pgd_ctor(pgd_t *pgd) { - pgd_t *pgd = p; - /* If the pgd points to a shared pagetable level (either the ptes in non-PAE, or shared PMD in PAE), then just copy the references from swapper_pg_dir. */ @@ -87,7 +85,7 @@ static void pgd_ctor(void *p) pgd_list_add(pgd); } -static void pgd_dtor(void *pgd) +static void pgd_dtor(pgd_t *pgd) { unsigned long flags; /* can be called from interrupt context */ diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c index cab0abb..0951db9 100644 --- a/arch/x86/mm/pgtable_32.c +++ b/arch/x86/mm/pgtable_32.c @@ -123,7 +123,8 @@ static int __init parse_vmalloc(char *arg) if (!arg) return -EINVAL; - __VMALLOC_RESERVE = memparse(arg, &arg); + /* Add VMALLOC_OFFSET to the parsed value due to vm area guard hole*/ + __VMALLOC_RESERVE = memparse(arg, &arg) + VMALLOC_OFFSET; return 0; } early_param("vmalloc", parse_vmalloc); diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index 0227694..8a5f161 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -295,10 +295,12 @@ static void nmi_cpu_shutdown(void *dummy) static void nmi_shutdown(void) { - struct op_msrs *msrs = &get_cpu_var(cpu_msrs); + struct op_msrs *msrs; + nmi_enabled = 0; on_each_cpu(nmi_cpu_shutdown, NULL, 1); unregister_die_notifier(&profile_exceptions_nb); + msrs = &get_cpu_var(cpu_msrs); model->shutdown(msrs); free_msrs(); put_cpu_var(cpu_msrs); diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c index 56b4757..43ac5af 100644 --- a/arch/x86/oprofile/op_model_p4.c +++ b/arch/x86/oprofile/op_model_p4.c @@ -10,11 +10,12 @@ #include <linux/oprofile.h> #include <linux/smp.h> +#include <linux/ptrace.h> +#include <linux/nmi.h> #include <asm/msr.h> -#include <asm/ptrace.h> #include <asm/fixmap.h> #include <asm/apic.h> -#include <asm/nmi.h> + #include "op_x86_model.h" #include "op_counter.h" @@ -40,7 +41,7 @@ static unsigned int num_controls = NUM_CONTROLS_NON_HT; static inline void setup_num_counters(void) { #ifdef CONFIG_SMP - if (smp_num_siblings == 2){ + if (smp_num_siblings == 2) { num_counters = NUM_COUNTERS_HT2; num_controls = NUM_CONTROLS_HT2; } @@ -86,7 +87,7 @@ struct p4_event_binding { #define CTR_FLAME_2 (1 << 6) #define CTR_IQ_5 (1 << 7) -static struct p4_counter_binding p4_counters [NUM_COUNTERS_NON_HT] = { +static struct p4_counter_binding p4_counters[NUM_COUNTERS_NON_HT] = { { CTR_BPU_0, MSR_P4_BPU_PERFCTR0, MSR_P4_BPU_CCCR0 }, { CTR_MS_0, MSR_P4_MS_PERFCTR0, MSR_P4_MS_CCCR0 }, { CTR_FLAME_0, MSR_P4_FLAME_PERFCTR0, MSR_P4_FLAME_CCCR0 }, @@ -97,32 +98,32 @@ static struct p4_counter_binding p4_counters [NUM_COUNTERS_NON_HT] = { { CTR_IQ_5, MSR_P4_IQ_PERFCTR5, MSR_P4_IQ_CCCR5 } }; -#define NUM_UNUSED_CCCRS NUM_CCCRS_NON_HT - NUM_COUNTERS_NON_HT +#define NUM_UNUSED_CCCRS (NUM_CCCRS_NON_HT - NUM_COUNTERS_NON_HT) /* p4 event codes in libop/op_event.h are indices into this table. */ static struct p4_event_binding p4_events[NUM_EVENTS] = { - + { /* BRANCH_RETIRED */ - 0x05, 0x06, + 0x05, 0x06, { {CTR_IQ_4, MSR_P4_CRU_ESCR2}, {CTR_IQ_5, MSR_P4_CRU_ESCR3} } }, - + { /* MISPRED_BRANCH_RETIRED */ - 0x04, 0x03, + 0x04, 0x03, { { CTR_IQ_4, MSR_P4_CRU_ESCR0}, { CTR_IQ_5, MSR_P4_CRU_ESCR1} } }, - + { /* TC_DELIVER_MODE */ 0x01, 0x01, - { { CTR_MS_0, MSR_P4_TC_ESCR0}, + { { CTR_MS_0, MSR_P4_TC_ESCR0}, { CTR_MS_2, MSR_P4_TC_ESCR1} } }, - + { /* BPU_FETCH_REQUEST */ - 0x00, 0x03, + 0x00, 0x03, { { CTR_BPU_0, MSR_P4_BPU_ESCR0}, { CTR_BPU_2, MSR_P4_BPU_ESCR1} } }, @@ -146,7 +147,7 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { }, { /* LOAD_PORT_REPLAY */ - 0x02, 0x04, + 0x02, 0x04, { { CTR_FLAME_0, MSR_P4_SAAT_ESCR0}, { CTR_FLAME_2, MSR_P4_SAAT_ESCR1} } }, @@ -170,43 +171,43 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { }, { /* BSQ_CACHE_REFERENCE */ - 0x07, 0x0c, + 0x07, 0x0c, { { CTR_BPU_0, MSR_P4_BSU_ESCR0}, { CTR_BPU_2, MSR_P4_BSU_ESCR1} } }, { /* IOQ_ALLOCATION */ - 0x06, 0x03, + 0x06, 0x03, { { CTR_BPU_0, MSR_P4_FSB_ESCR0}, { 0, 0 } } }, { /* IOQ_ACTIVE_ENTRIES */ - 0x06, 0x1a, + 0x06, 0x1a, { { CTR_BPU_2, MSR_P4_FSB_ESCR1}, { 0, 0 } } }, { /* FSB_DATA_ACTIVITY */ - 0x06, 0x17, + 0x06, 0x17, { { CTR_BPU_0, MSR_P4_FSB_ESCR0}, { CTR_BPU_2, MSR_P4_FSB_ESCR1} } }, { /* BSQ_ALLOCATION */ - 0x07, 0x05, + 0x07, 0x05, { { CTR_BPU_0, MSR_P4_BSU_ESCR0}, { 0, 0 } } }, { /* BSQ_ACTIVE_ENTRIES */ 0x07, 0x06, - { { CTR_BPU_2, MSR_P4_BSU_ESCR1 /* guess */}, + { { CTR_BPU_2, MSR_P4_BSU_ESCR1 /* guess */}, { 0, 0 } } }, { /* X87_ASSIST */ - 0x05, 0x03, + 0x05, 0x03, { { CTR_IQ_4, MSR_P4_CRU_ESCR2}, { CTR_IQ_5, MSR_P4_CRU_ESCR3} } }, @@ -216,21 +217,21 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { { { CTR_FLAME_0, MSR_P4_FIRM_ESCR0}, { CTR_FLAME_2, MSR_P4_FIRM_ESCR1} } }, - + { /* PACKED_SP_UOP */ - 0x01, 0x08, + 0x01, 0x08, { { CTR_FLAME_0, MSR_P4_FIRM_ESCR0}, { CTR_FLAME_2, MSR_P4_FIRM_ESCR1} } }, - + { /* PACKED_DP_UOP */ - 0x01, 0x0c, + 0x01, 0x0c, { { CTR_FLAME_0, MSR_P4_FIRM_ESCR0}, { CTR_FLAME_2, MSR_P4_FIRM_ESCR1} } }, { /* SCALAR_SP_UOP */ - 0x01, 0x0a, + 0x01, 0x0a, { { CTR_FLAME_0, MSR_P4_FIRM_ESCR0}, { CTR_FLAME_2, MSR_P4_FIRM_ESCR1} } }, @@ -242,31 +243,31 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { }, { /* 64BIT_MMX_UOP */ - 0x01, 0x02, + 0x01, 0x02, { { CTR_FLAME_0, MSR_P4_FIRM_ESCR0}, { CTR_FLAME_2, MSR_P4_FIRM_ESCR1} } }, - + { /* 128BIT_MMX_UOP */ - 0x01, 0x1a, + 0x01, 0x1a, { { CTR_FLAME_0, MSR_P4_FIRM_ESCR0}, { CTR_FLAME_2, MSR_P4_FIRM_ESCR1} } }, { /* X87_FP_UOP */ - 0x01, 0x04, + 0x01, 0x04, { { CTR_FLAME_0, MSR_P4_FIRM_ESCR0}, { CTR_FLAME_2, MSR_P4_FIRM_ESCR1} } }, - + { /* X87_SIMD_MOVES_UOP */ - 0x01, 0x2e, + 0x01, 0x2e, { { CTR_FLAME_0, MSR_P4_FIRM_ESCR0}, { CTR_FLAME_2, MSR_P4_FIRM_ESCR1} } }, - + { /* MACHINE_CLEAR */ - 0x05, 0x02, + 0x05, 0x02, { { CTR_IQ_4, MSR_P4_CRU_ESCR2}, { CTR_IQ_5, MSR_P4_CRU_ESCR3} } }, @@ -276,9 +277,9 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { { { CTR_BPU_0, MSR_P4_FSB_ESCR0}, { CTR_BPU_2, MSR_P4_FSB_ESCR1} } }, - + { /* TC_MS_XFER */ - 0x00, 0x05, + 0x00, 0x05, { { CTR_MS_0, MSR_P4_MS_ESCR0}, { CTR_MS_2, MSR_P4_MS_ESCR1} } }, @@ -308,7 +309,7 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { }, { /* INSTR_RETIRED */ - 0x04, 0x02, + 0x04, 0x02, { { CTR_IQ_4, MSR_P4_CRU_ESCR0}, { CTR_IQ_5, MSR_P4_CRU_ESCR1} } }, @@ -319,14 +320,14 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { { CTR_IQ_5, MSR_P4_CRU_ESCR1} } }, - { /* UOP_TYPE */ - 0x02, 0x02, + { /* UOP_TYPE */ + 0x02, 0x02, { { CTR_IQ_4, MSR_P4_RAT_ESCR0}, { CTR_IQ_5, MSR_P4_RAT_ESCR1} } }, { /* RETIRED_MISPRED_BRANCH_TYPE */ - 0x02, 0x05, + 0x02, 0x05, { { CTR_MS_0, MSR_P4_TBPU_ESCR0}, { CTR_MS_2, MSR_P4_TBPU_ESCR1} } }, @@ -349,8 +350,8 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { #define ESCR_SET_OS_1(escr, os) ((escr) |= (((os) & 1) << 1)) #define ESCR_SET_EVENT_SELECT(escr, sel) ((escr) |= (((sel) & 0x3f) << 25)) #define ESCR_SET_EVENT_MASK(escr, mask) ((escr) |= (((mask) & 0xffff) << 9)) -#define ESCR_READ(escr,high,ev,i) do {rdmsr(ev->bindings[(i)].escr_address, (escr), (high));} while (0) -#define ESCR_WRITE(escr,high,ev,i) do {wrmsr(ev->bindings[(i)].escr_address, (escr), (high));} while (0) +#define ESCR_READ(escr, high, ev, i) do {rdmsr(ev->bindings[(i)].escr_address, (escr), (high)); } while (0) +#define ESCR_WRITE(escr, high, ev, i) do {wrmsr(ev->bindings[(i)].escr_address, (escr), (high)); } while (0) #define CCCR_RESERVED_BITS 0x38030FFF #define CCCR_CLEAR(cccr) ((cccr) &= CCCR_RESERVED_BITS) @@ -360,15 +361,15 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { #define CCCR_SET_PMI_OVF_1(cccr) ((cccr) |= (1<<27)) #define CCCR_SET_ENABLE(cccr) ((cccr) |= (1<<12)) #define CCCR_SET_DISABLE(cccr) ((cccr) &= ~(1<<12)) -#define CCCR_READ(low, high, i) do {rdmsr(p4_counters[(i)].cccr_address, (low), (high));} while (0) -#define CCCR_WRITE(low, high, i) do {wrmsr(p4_counters[(i)].cccr_address, (low), (high));} while (0) +#define CCCR_READ(low, high, i) do {rdmsr(p4_counters[(i)].cccr_address, (low), (high)); } while (0) +#define CCCR_WRITE(low, high, i) do {wrmsr(p4_counters[(i)].cccr_address, (low), (high)); } while (0) #define CCCR_OVF_P(cccr) ((cccr) & (1U<<31)) #define CCCR_CLEAR_OVF(cccr) ((cccr) &= (~(1U<<31))) -#define CTRL_IS_RESERVED(msrs,c) (msrs->controls[(c)].addr ? 1 : 0) -#define CTR_IS_RESERVED(msrs,c) (msrs->counters[(c)].addr ? 1 : 0) -#define CTR_READ(l,h,i) do {rdmsr(p4_counters[(i)].counter_address, (l), (h));} while (0) -#define CTR_WRITE(l,i) do {wrmsr(p4_counters[(i)].counter_address, -(u32)(l), -1);} while (0) +#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0) +#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0) +#define CTR_READ(l, h, i) do {rdmsr(p4_counters[(i)].counter_address, (l), (h)); } while (0) +#define CTR_WRITE(l, i) do {wrmsr(p4_counters[(i)].counter_address, -(u32)(l), -1); } while (0) #define CTR_OVERFLOW_P(ctr) (!((ctr) & 0x80000000)) @@ -380,7 +381,7 @@ static unsigned int get_stagger(void) #ifdef CONFIG_SMP int cpu = smp_processor_id(); return (cpu != first_cpu(per_cpu(cpu_sibling_map, cpu))); -#endif +#endif return 0; } @@ -395,25 +396,23 @@ static unsigned long reset_value[NUM_COUNTERS_NON_HT]; static void p4_fill_in_addresses(struct op_msrs * const msrs) { - unsigned int i; + unsigned int i; unsigned int addr, cccraddr, stag; setup_num_counters(); stag = get_stagger(); /* initialize some registers */ - for (i = 0; i < num_counters; ++i) { + for (i = 0; i < num_counters; ++i) msrs->counters[i].addr = 0; - } - for (i = 0; i < num_controls; ++i) { + for (i = 0; i < num_controls; ++i) msrs->controls[i].addr = 0; - } - + /* the counter & cccr registers we pay attention to */ for (i = 0; i < num_counters; ++i) { addr = p4_counters[VIRT_CTR(stag, i)].counter_address; cccraddr = p4_counters[VIRT_CTR(stag, i)].cccr_address; - if (reserve_perfctr_nmi(addr)){ + if (reserve_perfctr_nmi(addr)) { msrs->counters[i].addr = addr; msrs->controls[i].addr = cccraddr; } @@ -447,22 +446,22 @@ static void p4_fill_in_addresses(struct op_msrs * const msrs) if (reserve_evntsel_nmi(addr)) msrs->controls[i].addr = addr; } - + for (addr = MSR_P4_MS_ESCR0 + stag; - addr <= MSR_P4_TC_ESCR1; ++i, addr += addr_increment()) { + addr <= MSR_P4_TC_ESCR1; ++i, addr += addr_increment()) { if (reserve_evntsel_nmi(addr)) msrs->controls[i].addr = addr; } - + for (addr = MSR_P4_IX_ESCR0 + stag; - addr <= MSR_P4_CRU_ESCR3; ++i, addr += addr_increment()) { + addr <= MSR_P4_CRU_ESCR3; ++i, addr += addr_increment()) { if (reserve_evntsel_nmi(addr)) msrs->controls[i].addr = addr; } /* there are 2 remaining non-contiguously located ESCRs */ - if (num_counters == NUM_COUNTERS_NON_HT) { + if (num_counters == NUM_COUNTERS_NON_HT) { /* standard non-HT CPUs handle both remaining ESCRs*/ if (reserve_evntsel_nmi(MSR_P4_CRU_ESCR5)) msrs->controls[i++].addr = MSR_P4_CRU_ESCR5; @@ -498,20 +497,20 @@ static void pmc_setup_one_p4_counter(unsigned int ctr) unsigned int stag; stag = get_stagger(); - + /* convert from counter *number* to counter *bit* */ counter_bit = 1 << VIRT_CTR(stag, ctr); - + /* find our event binding structure. */ if (counter_config[ctr].event <= 0 || counter_config[ctr].event > NUM_EVENTS) { - printk(KERN_ERR - "oprofile: P4 event code 0x%lx out of range\n", + printk(KERN_ERR + "oprofile: P4 event code 0x%lx out of range\n", counter_config[ctr].event); return; } - + ev = &(p4_events[counter_config[ctr].event - 1]); - + for (i = 0; i < maxbind; i++) { if (ev->bindings[i].virt_counter & counter_bit) { @@ -526,25 +525,24 @@ static void pmc_setup_one_p4_counter(unsigned int ctr) ESCR_SET_OS_1(escr, counter_config[ctr].kernel); } ESCR_SET_EVENT_SELECT(escr, ev->event_select); - ESCR_SET_EVENT_MASK(escr, counter_config[ctr].unit_mask); + ESCR_SET_EVENT_MASK(escr, counter_config[ctr].unit_mask); ESCR_WRITE(escr, high, ev, i); - + /* modify CCCR */ CCCR_READ(cccr, high, VIRT_CTR(stag, ctr)); CCCR_CLEAR(cccr); CCCR_SET_REQUIRED_BITS(cccr); CCCR_SET_ESCR_SELECT(cccr, ev->escr_select); - if (stag == 0) { + if (stag == 0) CCCR_SET_PMI_OVF_0(cccr); - } else { + else CCCR_SET_PMI_OVF_1(cccr); - } CCCR_WRITE(cccr, high, VIRT_CTR(stag, ctr)); return; } } - printk(KERN_ERR + printk(KERN_ERR "oprofile: P4 event code 0x%lx no binding, stag %d ctr %d\n", counter_config[ctr].event, stag, ctr); } @@ -559,14 +557,14 @@ static void p4_setup_ctrs(struct op_msrs const * const msrs) stag = get_stagger(); rdmsr(MSR_IA32_MISC_ENABLE, low, high); - if (! MISC_PMC_ENABLED_P(low)) { + if (!MISC_PMC_ENABLED_P(low)) { printk(KERN_ERR "oprofile: P4 PMC not available\n"); return; } /* clear the cccrs we will use */ for (i = 0 ; i < num_counters ; i++) { - if (unlikely(!CTRL_IS_RESERVED(msrs,i))) + if (unlikely(!CTRL_IS_RESERVED(msrs, i))) continue; rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high); CCCR_CLEAR(low); @@ -576,14 +574,14 @@ static void p4_setup_ctrs(struct op_msrs const * const msrs) /* clear all escrs (including those outside our concern) */ for (i = num_counters; i < num_controls; i++) { - if (unlikely(!CTRL_IS_RESERVED(msrs,i))) + if (unlikely(!CTRL_IS_RESERVED(msrs, i))) continue; wrmsr(msrs->controls[i].addr, 0, 0); } /* setup all counters */ for (i = 0 ; i < num_counters ; ++i) { - if ((counter_config[i].enabled) && (CTRL_IS_RESERVED(msrs,i))) { + if ((counter_config[i].enabled) && (CTRL_IS_RESERVED(msrs, i))) { reset_value[i] = counter_config[i].count; pmc_setup_one_p4_counter(i); CTR_WRITE(counter_config[i].count, VIRT_CTR(stag, i)); @@ -603,11 +601,11 @@ static int p4_check_ctrs(struct pt_regs * const regs, stag = get_stagger(); for (i = 0; i < num_counters; ++i) { - - if (!reset_value[i]) + + if (!reset_value[i]) continue; - /* + /* * there is some eccentricity in the hardware which * requires that we perform 2 extra corrections: * @@ -616,24 +614,24 @@ static int p4_check_ctrs(struct pt_regs * const regs, * * - write the counter back twice to ensure it gets * updated properly. - * + * * the former seems to be related to extra NMIs happening * during the current NMI; the latter is reported as errata * N15 in intel doc 249199-029, pentium 4 specification * update, though their suggested work-around does not * appear to solve the problem. */ - + real = VIRT_CTR(stag, i); CCCR_READ(low, high, real); - CTR_READ(ctr, high, real); + CTR_READ(ctr, high, real); if (CCCR_OVF_P(low) || CTR_OVERFLOW_P(ctr)) { oprofile_add_sample(regs, i); - CTR_WRITE(reset_value[i], real); + CTR_WRITE(reset_value[i], real); CCCR_CLEAR_OVF(low); CCCR_WRITE(low, high, real); - CTR_WRITE(reset_value[i], real); + CTR_WRITE(reset_value[i], real); } } @@ -683,15 +681,16 @@ static void p4_shutdown(struct op_msrs const * const msrs) int i; for (i = 0 ; i < num_counters ; ++i) { - if (CTR_IS_RESERVED(msrs,i)) + if (CTR_IS_RESERVED(msrs, i)) release_perfctr_nmi(msrs->counters[i].addr); } - /* some of the control registers are specially reserved in + /* + * some of the control registers are specially reserved in * conjunction with the counter registers (hence the starting offset). * This saves a few bits. */ for (i = num_counters ; i < num_controls ; ++i) { - if (CTRL_IS_RESERVED(msrs,i)) + if (CTRL_IS_RESERVED(msrs, i)) release_evntsel_nmi(msrs->controls[i].addr); } } diff --git a/arch/x86/pci/amd_bus.c b/arch/x86/pci/amd_bus.c index 6a0fca7..22e0576 100644 --- a/arch/x86/pci/amd_bus.c +++ b/arch/x86/pci/amd_bus.c @@ -580,7 +580,7 @@ static int __cpuinit amd_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { int cpu = (long)hcpu; - switch(action) { + switch (action) { case CPU_ONLINE: case CPU_ONLINE_FROZEN: smp_call_function_single(cpu, enable_pci_io_ecs, NULL, 0); diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c index 8e07718..006599d 100644 --- a/arch/x86/pci/irq.c +++ b/arch/x86/pci/irq.c @@ -1043,35 +1043,44 @@ static void __init pcibios_fixup_irqs(void) if (io_apic_assign_pci_irqs) { int irq; - if (pin) { - /* - * interrupt pins are numbered starting - * from 1 - */ - pin--; - irq = IO_APIC_get_PCI_irq_vector(dev->bus->number, - PCI_SLOT(dev->devfn), pin); - /* - * Busses behind bridges are typically not listed in the MP-table. - * In this case we have to look up the IRQ based on the parent bus, - * parent slot, and pin number. The SMP code detects such bridged - * busses itself so we should get into this branch reliably. - */ - if (irq < 0 && dev->bus->parent) { /* go back to the bridge */ - struct pci_dev *bridge = dev->bus->self; - - pin = (pin + PCI_SLOT(dev->devfn)) % 4; - irq = IO_APIC_get_PCI_irq_vector(bridge->bus->number, - PCI_SLOT(bridge->devfn), pin); - if (irq >= 0) - dev_warn(&dev->dev, "using bridge %s INT %c to get IRQ %d\n", - pci_name(bridge), - 'A' + pin, irq); - } - if (irq >= 0) { - dev_info(&dev->dev, "PCI->APIC IRQ transform: INT %c -> IRQ %d\n", 'A' + pin, irq); - dev->irq = irq; - } + if (!pin) + continue; + + /* + * interrupt pins are numbered starting from 1 + */ + pin--; + irq = IO_APIC_get_PCI_irq_vector(dev->bus->number, + PCI_SLOT(dev->devfn), pin); + /* + * Busses behind bridges are typically not listed in the + * MP-table. In this case we have to look up the IRQ + * based on the parent bus, parent slot, and pin number. + * The SMP code detects such bridged busses itself so we + * should get into this branch reliably. + */ + if (irq < 0 && dev->bus->parent) { + /* go back to the bridge */ + struct pci_dev *bridge = dev->bus->self; + int bus; + + pin = (pin + PCI_SLOT(dev->devfn)) % 4; + bus = bridge->bus->number; + irq = IO_APIC_get_PCI_irq_vector(bus, + PCI_SLOT(bridge->devfn), pin); + if (irq >= 0) + dev_warn(&dev->dev, + "using bridge %s INT %c to " + "get IRQ %d\n", + pci_name(bridge), + 'A' + pin, irq); + } + if (irq >= 0) { + dev_info(&dev->dev, + "PCI->APIC IRQ transform: INT %c " + "-> IRQ %d\n", + 'A' + pin, irq); + dev->irq = irq; } } #endif diff --git a/arch/x86/power/hibernate_asm_32.S b/arch/x86/power/hibernate_asm_32.S index 4fc7e87..d1e9b53 100644 --- a/arch/x86/power/hibernate_asm_32.S +++ b/arch/x86/power/hibernate_asm_32.S @@ -1,5 +1,3 @@ -.text - /* * This may not use any stack, nor any variable that is not "NoSave": * @@ -12,17 +10,18 @@ #include <asm/segment.h> #include <asm/page.h> #include <asm/asm-offsets.h> +#include <asm/processor-flags.h> - .text +.text ENTRY(swsusp_arch_suspend) - movl %esp, saved_context_esp movl %ebx, saved_context_ebx movl %ebp, saved_context_ebp movl %esi, saved_context_esi movl %edi, saved_context_edi - pushfl ; popl saved_context_eflags + pushfl + popl saved_context_eflags call swsusp_save ret @@ -59,7 +58,7 @@ done: movl mmu_cr4_features, %ecx jecxz 1f # cr4 Pentium and higher, skip if zero movl %ecx, %edx - andl $~(1<<7), %edx; # PGE + andl $~(X86_CR4_PGE), %edx movl %edx, %cr4; # turn off PGE 1: movl %cr3, %eax; # flush TLB @@ -74,7 +73,8 @@ done: movl saved_context_esi, %esi movl saved_context_edi, %edi - pushl saved_context_eflags ; popfl + pushl saved_context_eflags + popfl xorl %eax, %eax diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index a4e201b..7dcd321 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -812,7 +812,7 @@ static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high) /* Early in boot, while setting up the initial pagetable, assume everything is pinned. */ -static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn) +static __init void xen_alloc_pte_init(struct mm_struct *mm, unsigned long pfn) { #ifdef CONFIG_FLATMEM BUG_ON(mem_map); /* should only be used early */ @@ -822,7 +822,7 @@ static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn) /* Early release_pte assumes that all pts are pinned, since there's only init_mm and anything attached to that is pinned. */ -static void xen_release_pte_init(u32 pfn) +static void xen_release_pte_init(unsigned long pfn) { make_lowmem_page_readwrite(__va(PFN_PHYS(pfn))); } @@ -838,7 +838,7 @@ static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn) /* This needs to make sure the new pte page is pinned iff its being attached to a pinned pagetable. */ -static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level) +static void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, unsigned level) { struct page *page = pfn_to_page(pfn); @@ -856,12 +856,12 @@ static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level) } } -static void xen_alloc_pte(struct mm_struct *mm, u32 pfn) +static void xen_alloc_pte(struct mm_struct *mm, unsigned long pfn) { xen_alloc_ptpage(mm, pfn, PT_PTE); } -static void xen_alloc_pmd(struct mm_struct *mm, u32 pfn) +static void xen_alloc_pmd(struct mm_struct *mm, unsigned long pfn) { xen_alloc_ptpage(mm, pfn, PT_PMD); } @@ -909,7 +909,7 @@ static void xen_pgd_free(struct mm_struct *mm, pgd_t *pgd) } /* This should never happen until we're OK to use struct page */ -static void xen_release_ptpage(u32 pfn, unsigned level) +static void xen_release_ptpage(unsigned long pfn, unsigned level) { struct page *page = pfn_to_page(pfn); @@ -923,23 +923,23 @@ static void xen_release_ptpage(u32 pfn, unsigned level) } } -static void xen_release_pte(u32 pfn) +static void xen_release_pte(unsigned long pfn) { xen_release_ptpage(pfn, PT_PTE); } -static void xen_release_pmd(u32 pfn) +static void xen_release_pmd(unsigned long pfn) { xen_release_ptpage(pfn, PT_PMD); } #if PAGETABLE_LEVELS == 4 -static void xen_alloc_pud(struct mm_struct *mm, u32 pfn) +static void xen_alloc_pud(struct mm_struct *mm, unsigned long pfn) { xen_alloc_ptpage(mm, pfn, PT_PUD); } -static void xen_release_pud(u32 pfn) +static void xen_release_pud(unsigned long pfn) { xen_release_ptpage(pfn, PT_PUD); } diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index b6acc3a..d679010 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -42,7 +42,7 @@ char * __init xen_memory_setup(void) e820.nr_map = 0; - e820_add_region(0, PFN_PHYS(max_pfn), E820_RAM); + e820_add_region(0, PFN_PHYS((u64)max_pfn), E820_RAM); /* * Even though this is normal, usable memory under Xen, reserve |