diff options
68 files changed, 1638 insertions, 825 deletions
diff --git a/Documentation/x86/earlyprintk.txt b/Documentation/x86/earlyprintk.txt new file mode 100644 index 000000000000..607b1a016064 --- /dev/null +++ b/Documentation/x86/earlyprintk.txt @@ -0,0 +1,101 @@ + +Mini-HOWTO for using the earlyprintk=dbgp boot option with a +USB2 Debug port key and a debug cable, on x86 systems. + +You need two computers, the 'USB debug key' special gadget and +and two USB cables, connected like this: + + [host/target] <-------> [USB debug key] <-------> [client/console] + +1. There are three specific hardware requirements: + + a.) Host/target system needs to have USB debug port capability. + + You can check this capability by looking at a 'Debug port' bit in + the lspci -vvv output: + + # lspci -vvv + ... + 00:1d.7 USB Controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #1 (rev 03) (prog-if 20 [EHCI]) + Subsystem: Lenovo ThinkPad T61 + Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx- + Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- + Latency: 0 + Interrupt: pin D routed to IRQ 19 + Region 0: Memory at fe227000 (32-bit, non-prefetchable) [size=1K] + Capabilities: [50] Power Management version 2 + Flags: PMEClk- DSI- D1- D2- AuxCurrent=375mA PME(D0+,D1-,D2-,D3hot+,D3cold+) + Status: D0 PME-Enable- DSel=0 DScale=0 PME+ + Capabilities: [58] Debug port: BAR=1 offset=00a0 + ^^^^^^^^^^^ <==================== [ HERE ] + Kernel driver in use: ehci_hcd + Kernel modules: ehci-hcd + ... + +( If your system does not list a debug port capability then you probably + wont be able to use the USB debug key. ) + + b.) You also need a Netchip USB debug cable/key: + + http://www.plxtech.com/products/NET2000/NET20DC/default.asp + + This is a small blue plastic connector with two USB connections, + it draws power from its USB connections. + + c.) Thirdly, you need a second client/console system with a regular USB port. + +2. Software requirements: + + a.) On the host/target system: + + You need to enable the following kernel config option: + + CONFIG_EARLY_PRINTK_DBGP=y + + And you need to add the boot command line: "earlyprintk=dbgp". + (If you are using Grub, append it to the 'kernel' line in + /etc/grub.conf) + + NOTE: normally earlyprintk console gets turned off once the + regular console is alive - use "earlyprintk=dbgp,keep" to keep + this channel open beyond early bootup. This can be useful for + debugging crashes under Xorg, etc. + + b.) On the client/console system: + + You should enable the following kernel config option: + + CONFIG_USB_SERIAL_DEBUG=y + + On the next bootup with the modified kernel you should + get a /dev/ttyUSBx device(s). + + Now this channel of kernel messages is ready to be used: start + your favorite terminal emulator (minicom, etc.) and set + it up to use /dev/ttyUSB0 - or use a raw 'cat /dev/ttyUSBx' to + see the raw output. + + c.) On Nvidia Southbridge based systems: the kernel will try to probe + and find out which port has debug device connected. + +3. Testing that it works fine: + + You can test the output by using earlyprintk=dbgp,keep and provoking + kernel messages on the host/target system. You can provoke a harmless + kernel message by for example doing: + + echo h > /proc/sysrq-trigger + + On the host/target system you should see this help line in "dmesg" output: + + SysRq : HELP : loglevel(0-9) reBoot Crashdump terminate-all-tasks(E) memory-full-oom-kill(F) kill-all-tasks(I) saK show-backtrace-all-active-cpus(L) show-memory-usage(M) nice-all-RT-tasks(N) powerOff show-registers(P) show-all-timers(Q) unRaw Sync show-task-states(T) Unmount show-blocked-tasks(W) dump-ftrace-buffer(Z) + + On the client/console system do: + + cat /dev/ttyUSB0 + + And you should see the help line above displayed shortly after you've + provoked it on the host system. + +If it does not work then please ask about it on the linux-kernel@vger.kernel.org +mailing list or contact the x86 maintainers. @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 29 -EXTRAVERSION = -rc6 +EXTRAVERSION = -rc7 NAME = Erotic Pickled Herring # *DOCUMENTATION* diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 7049815d66d5..68d6494c0389 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -233,12 +233,13 @@ static void __init cacheid_init(void) unsigned int cachetype = read_cpuid_cachetype(); unsigned int arch = cpu_architecture(); - if (arch >= CPU_ARCH_ARMv7) { - cacheid = CACHEID_VIPT_NONALIASING; - if ((cachetype & (3 << 14)) == 1 << 14) - cacheid |= CACHEID_ASID_TAGGED; - } else if (arch >= CPU_ARCH_ARMv6) { - if (cachetype & (1 << 23)) + if (arch >= CPU_ARCH_ARMv6) { + if ((cachetype & (7 << 29)) == 4 << 29) { + /* ARMv7 register format */ + cacheid = CACHEID_VIPT_NONALIASING; + if ((cachetype & (3 << 14)) == 1 << 14) + cacheid |= CACHEID_ASID_TAGGED; + } else if (cachetype & (1 << 23)) cacheid = CACHEID_VIPT_ALIASING; else cacheid = CACHEID_VIPT_NONALIASING; diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c index 9bb4f043aa22..7ac812dc055a 100644 --- a/arch/arm/mach-at91/pm.c +++ b/arch/arm/mach-at91/pm.c @@ -332,7 +332,6 @@ static int at91_pm_enter(suspend_state_t state) at91_sys_read(AT91_AIC_IPR) & at91_sys_read(AT91_AIC_IMR)); error: - sdram_selfrefresh_disable(); target_state = PM_SUSPEND_ON; at91_irq_resume(); at91_gpio_resume(); diff --git a/arch/arm/mm/abort-ev6.S b/arch/arm/mm/abort-ev6.S index 8a7f65ba14b7..94077fbd96b7 100644 --- a/arch/arm/mm/abort-ev6.S +++ b/arch/arm/mm/abort-ev6.S @@ -23,7 +23,8 @@ ENTRY(v6_early_abort) #ifdef CONFIG_CPU_32v6K clrex #else - strex r0, r1, [sp] @ Clear the exclusive monitor + sub r1, sp, #4 @ Get unused stack location + strex r0, r1, [r1] @ Clear the exclusive monitor #endif mrc p15, 0, r1, c5, c0, 0 @ get FSR mrc p15, 0, r0, c6, c0, 0 @ get FAR diff --git a/arch/arm/plat-s3c64xx/irq-eint.c b/arch/arm/plat-s3c64xx/irq-eint.c index 1f7cc0067f5c..ebb305ce7689 100644 --- a/arch/arm/plat-s3c64xx/irq-eint.c +++ b/arch/arm/plat-s3c64xx/irq-eint.c @@ -55,7 +55,7 @@ static void s3c_irq_eint_unmask(unsigned int irq) u32 mask; mask = __raw_readl(S3C64XX_EINT0MASK); - mask |= eint_irq_to_bit(irq); + mask &= ~eint_irq_to_bit(irq); __raw_writel(mask, S3C64XX_EINT0MASK); } diff --git a/arch/powerpc/platforms/86xx/gef_sbc610.c b/arch/powerpc/platforms/86xx/gef_sbc610.c index fb371f5ce132..d6b772ba3b8f 100644 --- a/arch/powerpc/platforms/86xx/gef_sbc610.c +++ b/arch/powerpc/platforms/86xx/gef_sbc610.c @@ -142,6 +142,10 @@ static void __init gef_sbc610_nec_fixup(struct pci_dev *pdev) { unsigned int val; + /* Do not do the fixup on other platforms! */ + if (!machine_is(gef_sbc610)) + return; + printk(KERN_INFO "Running NEC uPD720101 Fixup\n"); /* Ensure ports 1, 2, 3, 4 & 5 are enabled */ diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c index c42cd898f68b..6118890c946d 100644 --- a/arch/s390/crypto/aes_s390.c +++ b/arch/s390/crypto/aes_s390.c @@ -556,7 +556,7 @@ static void __exit aes_s390_fini(void) module_init(aes_s390_init); module_exit(aes_s390_fini); -MODULE_ALIAS("aes"); +MODULE_ALIAS("aes-all"); MODULE_DESCRIPTION("Rijndael (AES) Cipher Algorithm"); MODULE_LICENSE("GPL"); diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index f5cef3fbf9a5..31758378bcd2 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -783,6 +783,11 @@ config X86_MCE_AMD Additional support for AMD specific MCE features such as the DRAM Error Threshold. +config X86_MCE_THRESHOLD + depends on X86_MCE_AMD || X86_MCE_INTEL + bool + default y + config X86_MCE_NONFATAL tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4" depends on X86_32 && X86_MCE diff --git a/arch/x86/include/asm/apicdef.h b/arch/x86/include/asm/apicdef.h index 63134e31e8b9..bc9514fb3b13 100644 --- a/arch/x86/include/asm/apicdef.h +++ b/arch/x86/include/asm/apicdef.h @@ -53,6 +53,7 @@ #define APIC_ESR_SENDILL 0x00020 #define APIC_ESR_RECVILL 0x00040 #define APIC_ESR_ILLREGA 0x00080 +#define APIC_LVTCMCI 0x2f0 #define APIC_ICR 0x300 #define APIC_DEST_SELF 0x40000 #define APIC_DEST_ALLINC 0x80000 diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index ca5ffb2856b6..edc90f23e708 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -37,8 +37,6 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...); #else /* !CONFIG_X86_32 */ -#define MAX_EFI_IO_PAGES 100 - extern u64 efi_call0(void *fp); extern u64 efi_call1(void *fp, u64 arg1); extern u64 efi_call2(void *fp, u64 arg1, u64 arg2); diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h index dca8f03da5b2..63a79c77d220 100644 --- a/arch/x86/include/asm/fixmap.h +++ b/arch/x86/include/asm/fixmap.h @@ -24,9 +24,6 @@ #include <asm/kmap_types.h> #else #include <asm/vsyscall.h> -#ifdef CONFIG_EFI -#include <asm/efi.h> -#endif #endif /* @@ -92,13 +89,6 @@ enum fixed_addresses { FIX_IO_APIC_BASE_0, FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS - 1, #endif -#ifdef CONFIG_X86_64 -#ifdef CONFIG_EFI - FIX_EFI_IO_MAP_LAST_PAGE, - FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE - + MAX_EFI_IO_PAGES - 1, -#endif -#endif #ifdef CONFIG_X86_VISWS_APIC FIX_CO_CPU, /* Cobalt timer */ FIX_CO_APIC, /* Cobalt APIC Redirection Table */ diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index 48f0004db8c9..71c9e5183982 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -172,7 +172,13 @@ static inline void __save_init_fpu(struct task_struct *tsk) #else /* CONFIG_X86_32 */ -extern void finit(void); +#ifdef CONFIG_MATH_EMULATION +extern void finit_task(struct task_struct *tsk); +#else +static inline void finit_task(struct task_struct *tsk) +{ +} +#endif static inline void tolerant_fwait(void) { diff --git a/arch/x86/include/asm/init.h b/arch/x86/include/asm/init.h new file mode 100644 index 000000000000..36fb1a6a5109 --- /dev/null +++ b/arch/x86/include/asm/init.h @@ -0,0 +1,18 @@ +#ifndef _ASM_X86_INIT_32_H +#define _ASM_X86_INIT_32_H + +#ifdef CONFIG_X86_32 +extern void __init early_ioremap_page_table_range_init(void); +#endif + +extern unsigned long __init +kernel_physical_mapping_init(unsigned long start, + unsigned long end, + unsigned long page_size_mask); + + +extern unsigned long __initdata e820_table_start; +extern unsigned long __meminitdata e820_table_end; +extern unsigned long __meminitdata e820_table_top; + +#endif /* _ASM_X86_INIT_32_H */ diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 32c6e17b960b..563933e06a35 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -11,6 +11,8 @@ */ #define MCG_CTL_P (1UL<<8) /* MCG_CAP register available */ +#define MCG_EXT_P (1ULL<<9) /* Extended registers available */ +#define MCG_CMCI_P (1ULL<<10) /* CMCI supported */ #define MCG_STATUS_RIPV (1UL<<0) /* restart ip valid */ #define MCG_STATUS_EIPV (1UL<<1) /* ip points to correct instruction */ @@ -90,14 +92,29 @@ extern int mce_disabled; #include <asm/atomic.h> +void mce_setup(struct mce *m); void mce_log(struct mce *m); DECLARE_PER_CPU(struct sys_device, device_mce); extern void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu); +/* + * To support more than 128 would need to escape the predefined + * Linux defined extended banks first. + */ +#define MAX_NR_BANKS (MCE_EXTENDED_BANK - 1) + #ifdef CONFIG_X86_MCE_INTEL void mce_intel_feature_init(struct cpuinfo_x86 *c); +void cmci_clear(void); +void cmci_reenable(void); +void cmci_rediscover(int dying); +void cmci_recheck(void); #else static inline void mce_intel_feature_init(struct cpuinfo_x86 *c) { } +static inline void cmci_clear(void) {} +static inline void cmci_reenable(void) {} +static inline void cmci_rediscover(int dying) {} +static inline void cmci_recheck(void) {} #endif #ifdef CONFIG_X86_MCE_AMD @@ -106,11 +123,23 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c); static inline void mce_amd_feature_init(struct cpuinfo_x86 *c) { } #endif -void mce_log_therm_throt_event(unsigned int cpu, __u64 status); +extern int mce_available(struct cpuinfo_x86 *c); + +void mce_log_therm_throt_event(__u64 status); extern atomic_t mce_entry; extern void do_machine_check(struct pt_regs *, long); + +typedef DECLARE_BITMAP(mce_banks_t, MAX_NR_BANKS); +DECLARE_PER_CPU(mce_banks_t, mce_poll_banks); + +enum mcp_flags { + MCP_TIMESTAMP = (1 << 0), /* log time stamp */ + MCP_UC = (1 << 1), /* log uncorrected errors */ +}; +extern void machine_check_poll(enum mcp_flags flags, mce_banks_t *b); + extern int mce_notify_user(void); #endif /* !CONFIG_X86_32 */ @@ -120,8 +149,8 @@ extern void mcheck_init(struct cpuinfo_x86 *c); #else #define mcheck_init(c) do { } while (0) #endif -extern void stop_mce(void); -extern void restart_mce(void); + +extern void (*mce_threshold_vector)(void); #endif /* __KERNEL__ */ #endif /* _ASM_X86_MCE_H */ diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 358acc59ae04..2dbd2314139e 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -77,6 +77,11 @@ #define MSR_IA32_MC0_ADDR 0x00000402 #define MSR_IA32_MC0_MISC 0x00000403 +/* These are consecutive and not in the normal 4er MCE bank block */ +#define MSR_IA32_MC0_CTL2 0x00000280 +#define CMCI_EN (1ULL << 30) +#define CMCI_THRESHOLD_MASK 0xffffULL + #define MSR_P6_PERFCTR0 0x000000c1 #define MSR_P6_PERFCTR1 0x000000c2 #define MSR_P6_EVNTSEL0 0x00000186 diff --git a/arch/x86/include/asm/page_types.h b/arch/x86/include/asm/page_types.h index 2d625da6603c..826ad37006ab 100644 --- a/arch/x86/include/asm/page_types.h +++ b/arch/x86/include/asm/page_types.h @@ -40,14 +40,8 @@ #ifndef __ASSEMBLY__ -struct pgprot; - extern int page_is_ram(unsigned long pagenr); extern int devmem_is_allowed(unsigned long pagenr); -extern void map_devmem(unsigned long pfn, unsigned long size, - struct pgprot vma_prot); -extern void unmap_devmem(unsigned long pfn, unsigned long size, - struct pgprot vma_prot); extern unsigned long max_low_pfn_mapped; extern unsigned long max_pfn_mapped; diff --git a/arch/x86/include/asm/pat.h b/arch/x86/include/asm/pat.h index b0e70056838e..2cd07b9422f4 100644 --- a/arch/x86/include/asm/pat.h +++ b/arch/x86/include/asm/pat.h @@ -2,6 +2,7 @@ #define _ASM_X86_PAT_H #include <linux/types.h> +#include <asm/pgtable_types.h> #ifdef CONFIG_X86_PAT extern int pat_enabled; @@ -17,5 +18,9 @@ extern int free_memtype(u64 start, u64 end); extern int kernel_map_sync_memtype(u64 base, unsigned long size, unsigned long flag); +extern void map_devmem(unsigned long pfn, unsigned long size, + struct pgprot vma_prot); +extern void unmap_devmem(unsigned long pfn, unsigned long size, + struct pgprot vma_prot); #endif /* _ASM_X86_PAT_H */ diff --git a/arch/x86/include/asm/pgtable_32_types.h b/arch/x86/include/asm/pgtable_32_types.h index bd8df3b2fe04..2733fad45f98 100644 --- a/arch/x86/include/asm/pgtable_32_types.h +++ b/arch/x86/include/asm/pgtable_32_types.h @@ -25,6 +25,11 @@ * area for the same reason. ;) */ #define VMALLOC_OFFSET (8 * 1024 * 1024) + +#ifndef __ASSEMBLER__ +extern bool __vmalloc_start_set; /* set once high_memory is set */ +#endif + #define VMALLOC_START ((unsigned long)high_memory + VMALLOC_OFFSET) #ifdef CONFIG_X86_PAE #define LAST_PKMAP 512 diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h index 4d258ad76a0f..b8238dc8786d 100644 --- a/arch/x86/include/asm/pgtable_types.h +++ b/arch/x86/include/asm/pgtable_types.h @@ -273,6 +273,7 @@ typedef struct page *pgtable_t; extern pteval_t __supported_pte_mask; extern int nx_enabled; +extern void set_nx(void); #define pgprot_writecombine pgprot_writecombine extern pgprot_t pgprot_writecombine(pgprot_t prot); diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 6907b8e85d52..4c80f1557433 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -414,9 +414,17 @@ void __init alternative_instructions(void) that might execute the to be patched code. Other CPUs are not running. */ stop_nmi(); -#ifdef CONFIG_X86_MCE - stop_mce(); -#endif + + /* + * Don't stop machine check exceptions while patching. + * MCEs only happen when something got corrupted and in this + * case we must do something about the corruption. + * Ignoring it is worse than a unlikely patching race. + * Also machine checks tend to be broadcast and if one CPU + * goes into machine check the others follow quickly, so we don't + * expect a machine check to cause undue problems during to code + * patching. + */ apply_alternatives(__alt_instructions, __alt_instructions_end); @@ -456,9 +464,6 @@ void __init alternative_instructions(void) (unsigned long)__smp_locks_end); restart_nmi(); -#ifdef CONFIG_X86_MCE - restart_mce(); -#endif } /** diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index f9cecdfd05c5..30909a258d0f 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -46,6 +46,7 @@ #include <asm/idle.h> #include <asm/mtrr.h> #include <asm/smp.h> +#include <asm/mce.h> unsigned int num_processors; @@ -842,6 +843,14 @@ void clear_local_APIC(void) apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED); } #endif +#ifdef CONFIG_X86_MCE_INTEL + if (maxlvt >= 6) { + v = apic_read(APIC_LVTCMCI); + if (!(v & APIC_LVT_MASKED)) + apic_write(APIC_LVTCMCI, v | APIC_LVT_MASKED); + } +#endif + /* * Clean APIC state for other OSs: */ @@ -1241,6 +1250,12 @@ void __cpuinit setup_local_APIC(void) apic_write(APIC_LVT1, value); preempt_enable(); + +#ifdef CONFIG_X86_MCE_INTEL + /* Recheck CMCI information after local APIC is up on CPU #0 */ + if (smp_processor_id() == 0) + cmci_recheck(); +#endif } void __cpuinit end_local_APIC_setup(void) diff --git a/arch/x86/kernel/cpu/mcheck/Makefile b/arch/x86/kernel/cpu/mcheck/Makefile index d7d2323bbb69..b2f89829bbe8 100644 --- a/arch/x86/kernel/cpu/mcheck/Makefile +++ b/arch/x86/kernel/cpu/mcheck/Makefile @@ -4,3 +4,4 @@ obj-$(CONFIG_X86_32) += k7.o p4.o p5.o p6.o winchip.o obj-$(CONFIG_X86_MCE_INTEL) += mce_intel_64.o obj-$(CONFIG_X86_MCE_AMD) += mce_amd_64.o obj-$(CONFIG_X86_MCE_NONFATAL) += non-fatal.o +obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o diff --git a/arch/x86/kernel/cpu/mcheck/mce_32.c b/arch/x86/kernel/cpu/mcheck/mce_32.c index dfaebce3633e..3552119b091d 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_32.c +++ b/arch/x86/kernel/cpu/mcheck/mce_32.c @@ -60,20 +60,6 @@ void mcheck_init(struct cpuinfo_x86 *c) } } -static unsigned long old_cr4 __initdata; - -void __init stop_mce(void) -{ - old_cr4 = read_cr4(); - clear_in_cr4(X86_CR4_MCE); -} - -void __init restart_mce(void) -{ - if (old_cr4 & X86_CR4_MCE) - set_in_cr4(X86_CR4_MCE); -} - static int __init mcheck_disable(char *str) { mce_disabled = 1; diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c index fe79985ce0f2..bfbd5323a635 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_64.c @@ -3,6 +3,8 @@ * K8 parts Copyright 2002,2003 Andi Kleen, SuSE Labs. * Rest from unknown author(s). * 2004 Andi Kleen. Rewrote most of it. + * Copyright 2008 Intel Corporation + * Author: Andi Kleen */ #include <linux/init.h> @@ -24,6 +26,9 @@ #include <linux/ctype.h> #include <linux/kmod.h> #include <linux/kdebug.h> +#include <linux/kobject.h> +#include <linux/sysfs.h> +#include <linux/ratelimit.h> #include <asm/processor.h> #include <asm/msr.h> #include <asm/mce.h> @@ -32,7 +37,6 @@ #include <asm/idle.h> #define MISC_MCELOG_MINOR 227 -#define NR_SYSFS_BANKS 6 atomic_t mce_entry; @@ -47,7 +51,7 @@ static int mce_dont_init; */ static int tolerant = 1; static int banks; -static unsigned long bank[NR_SYSFS_BANKS] = { [0 ... NR_SYSFS_BANKS-1] = ~0UL }; +static u64 *bank; static unsigned long notify_user; static int rip_msr; static int mce_bootlog = -1; @@ -58,6 +62,19 @@ static char *trigger_argv[2] = { trigger, NULL }; static DECLARE_WAIT_QUEUE_HEAD(mce_wait); +/* MCA banks polled by the period polling timer for corrected events */ +DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = { + [0 ... BITS_TO_LONGS(MAX_NR_BANKS)-1] = ~0UL +}; + +/* Do initial initialization of a struct mce */ +void mce_setup(struct mce *m) +{ + memset(m, 0, sizeof(struct mce)); + m->cpu = smp_processor_id(); + rdtscll(m->tsc); +} + /* * Lockless MCE logging infrastructure. * This avoids deadlocks on printk locks without having to break locks. Also @@ -119,11 +136,11 @@ static void print_mce(struct mce *m) print_symbol("{%s}", m->ip); printk("\n"); } - printk(KERN_EMERG "TSC %Lx ", m->tsc); + printk(KERN_EMERG "TSC %llx ", m->tsc); if (m->addr) - printk("ADDR %Lx ", m->addr); + printk("ADDR %llx ", m->addr); if (m->misc) - printk("MISC %Lx ", m->misc); + printk("MISC %llx ", m->misc); printk("\n"); printk(KERN_EMERG "This is not a software problem!\n"); printk(KERN_EMERG "Run through mcelog --ascii to decode " @@ -149,8 +166,10 @@ static void mce_panic(char *msg, struct mce *backup, unsigned long start) panic(msg); } -static int mce_available(struct cpuinfo_x86 *c) +int mce_available(struct cpuinfo_x86 *c) { + if (mce_dont_init) + return 0; return cpu_has(c, X86_FEATURE_MCE) && cpu_has(c, X86_FEATURE_MCA); } @@ -172,7 +191,77 @@ static inline void mce_get_rip(struct mce *m, struct pt_regs *regs) } /* - * The actual machine check handler + * Poll for corrected events or events that happened before reset. + * Those are just logged through /dev/mcelog. + * + * This is executed in standard interrupt context. + */ +void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) +{ + struct mce m; + int i; + + mce_setup(&m); + + rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus); + for (i = 0; i < banks; i++) { + if (!bank[i] || !test_bit(i, *b)) + continue; + + m.misc = 0; + m.addr = 0; + m.bank = i; + m.tsc = 0; + + barrier(); + rdmsrl(MSR_IA32_MC0_STATUS + i*4, m.status); + if (!(m.status & MCI_STATUS_VAL)) + continue; + + /* + * Uncorrected events are handled by the exception handler + * when it is enabled. But when the exception is disabled log + * everything. + * + * TBD do the same check for MCI_STATUS_EN here? + */ + if ((m.status & MCI_STATUS_UC) && !(flags & MCP_UC)) + continue; + + if (m.status & MCI_STATUS_MISCV) + rdmsrl(MSR_IA32_MC0_MISC + i*4, m.misc); + if (m.status & MCI_STATUS_ADDRV) + rdmsrl(MSR_IA32_MC0_ADDR + i*4, m.addr); + + if (!(flags & MCP_TIMESTAMP)) + m.tsc = 0; + /* + * Don't get the IP here because it's unlikely to + * have anything to do with the actual error location. + */ + + mce_log(&m); + add_taint(TAINT_MACHINE_CHECK); + + /* + * Clear state for this bank. + */ + wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + } + + /* + * Don't clear MCG_STATUS here because it's only defined for + * exceptions. + */ +} + +/* + * The actual machine check handler. This only handles real + * exceptions when something got corrupted coming in through int 18. + * + * This is executed in NMI context not subject to normal locking rules. This + * implies that most kernel services cannot be safely used. Don't even + * think about putting a printk in there! */ void do_machine_check(struct pt_regs * regs, long error_code) { @@ -190,17 +279,18 @@ void do_machine_check(struct pt_regs * regs, long error_code) * error. */ int kill_it = 0; + DECLARE_BITMAP(toclear, MAX_NR_BANKS); atomic_inc(&mce_entry); - if ((regs - && notify_die(DIE_NMI, "machine check", regs, error_code, + if (notify_die(DIE_NMI, "machine check", regs, error_code, 18, SIGKILL) == NOTIFY_STOP) - || !banks) + goto out2; + if (!banks) goto out2; - memset(&m, 0, sizeof(struct mce)); - m.cpu = smp_processor_id(); + mce_setup(&m); + rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus); /* if the restart IP is not valid, we're done for */ if (!(m.mcgstatus & MCG_STATUS_RIPV)) @@ -210,18 +300,32 @@ void do_machine_check(struct pt_regs * regs, long error_code) barrier(); for (i = 0; i < banks; i++) { - if (i < NR_SYSFS_BANKS && !bank[i]) + __clear_bit(i, toclear); + if (!bank[i]) continue; m.misc = 0; m.addr = 0; m.bank = i; - m.tsc = 0; rdmsrl(MSR_IA32_MC0_STATUS + i*4, m.status); if ((m.status & MCI_STATUS_VAL) == 0) continue; + /* + * Non uncorrected errors are handled by machine_check_poll + * Leave them alone. + */ + if ((m.status & MCI_STATUS_UC) == 0) + continue; + + /* + * Set taint even when machine check was not enabled. + */ + add_taint(TAINT_MACHINE_CHECK); + + __set_bit(i, toclear); + if (m.status & MCI_STATUS_EN) { /* if PCC was set, there's no way out */ no_way_out |= !!(m.status & MCI_STATUS_PCC); @@ -235,6 +339,12 @@ void do_machine_check(struct pt_regs * regs, long error_code) no_way_out = 1; kill_it = 1; } + } else { + /* + * Machine check event was not enabled. Clear, but + * ignore. + */ + continue; } if (m.status & MCI_STATUS_MISCV) @@ -243,10 +353,7 @@ void do_machine_check(struct pt_regs * regs, long error_code) rdmsrl(MSR_IA32_MC0_ADDR + i*4, m.addr); mce_get_rip(&m, regs); - if (error_code >= 0) - rdtscll(m.tsc); - if (error_code != -2) - mce_log(&m); + mce_log(&m); /* Did this bank cause the exception? */ /* Assume that the bank with uncorrectable errors did it, @@ -255,14 +362,8 @@ void do_machine_check(struct pt_regs * regs, long error_code) panicm = m; panicm_found = 1; } - - add_taint(TAINT_MACHINE_CHECK); } - /* Never do anything final in the polling timer */ - if (!regs) - goto out; - /* If we didn't find an uncorrectable error, pick the last one (shouldn't happen, just being safe). */ if (!panicm_found) @@ -309,10 +410,11 @@ void do_machine_check(struct pt_regs * regs, long error_code) /* notify userspace ASAP */ set_thread_flag(TIF_MCE_NOTIFY); - out: /* the last thing we do is clear state */ - for (i = 0; i < banks; i++) - wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + for (i = 0; i < banks; i++) { + if (test_bit(i, toclear)) + wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + } wrmsrl(MSR_IA32_MCG_STATUS, 0); out2: atomic_dec(&mce_entry); @@ -332,15 +434,13 @@ void do_machine_check(struct pt_regs * regs, long error_code) * and historically has been the register value of the * MSR_IA32_THERMAL_STATUS (Intel) msr. */ -void mce_log_therm_throt_event(unsigned int cpu, __u64 status) +void mce_log_therm_throt_event(__u64 status) { struct mce m; - memset(&m, 0, sizeof(m)); - m.cpu = cpu; + mce_setup(&m); m.bank = MCE_THERMAL_BANK; m.status = status; - rdtscll(m.tsc); mce_log(&m); } #endif /* CONFIG_X86_MCE_INTEL */ @@ -353,18 +453,18 @@ void mce_log_therm_throt_event(unsigned int cpu, __u64 status) static int check_interval = 5 * 60; /* 5 minutes */ static int next_interval; /* in jiffies */ -static void mcheck_timer(struct work_struct *work); -static DECLARE_DELAYED_WORK(mcheck_work, mcheck_timer); +static void mcheck_timer(unsigned long); +static DEFINE_PER_CPU(struct timer_list, mce_timer); -static void mcheck_check_cpu(void *info) +static void mcheck_timer(unsigned long data) { - if (mce_available(¤t_cpu_data)) - do_machine_check(NULL, 0); -} + struct timer_list *t = &per_cpu(mce_timer, data); -static void mcheck_timer(struct work_struct *work) -{ - on_each_cpu(mcheck_check_cpu, NULL, 1); + WARN_ON(smp_processor_id() != data); + + if (mce_available(¤t_cpu_data)) + machine_check_poll(MCP_TIMESTAMP, + &__get_cpu_var(mce_poll_banks)); /* * Alert userspace if needed. If we logged an MCE, reduce the @@ -377,31 +477,41 @@ static void mcheck_timer(struct work_struct *work) (int)round_jiffies_relative(check_interval*HZ)); } - schedule_delayed_work(&mcheck_work, next_interval); + t->expires = jiffies + next_interval; + add_timer(t); +} + +static void mce_do_trigger(struct work_struct *work) +{ + call_usermodehelper(trigger, trigger_argv, NULL, UMH_NO_WAIT); } +static DECLARE_WORK(mce_trigger_work, mce_do_trigger); + /* - * This is only called from process context. This is where we do - * anything we need to alert userspace about new MCEs. This is called - * directly from the poller and also from entry.S and idle, thanks to - * TIF_MCE_NOTIFY. + * Notify the user(s) about new machine check events. + * Can be called from interrupt context, but not from machine check/NMI + * context. */ int mce_notify_user(void) { + /* Not more than two messages every minute */ + static DEFINE_RATELIMIT_STATE(ratelimit, 60*HZ, 2); + clear_thread_flag(TIF_MCE_NOTIFY); if (test_and_clear_bit(0, ¬ify_user)) { - static unsigned long last_print; - unsigned long now = jiffies; - wake_up_interruptible(&mce_wait); - if (trigger[0]) - call_usermodehelper(trigger, trigger_argv, NULL, - UMH_NO_WAIT); - if (time_after_eq(now, last_print + (check_interval*HZ))) { - last_print = now; + /* + * There is no risk of missing notifications because + * work_pending is always cleared before the function is + * executed. + */ + if (trigger[0] && !work_pending(&mce_trigger_work)) + schedule_work(&mce_trigger_work); + + if (__ratelimit(&ratelimit)) printk(KERN_INFO "Machine check events logged\n"); - } return 1; } @@ -425,63 +535,78 @@ static struct notifier_block mce_idle_notifier = { static __init int periodic_mcheck_init(void) { - next_interval = check_interval * HZ; - if (next_interval) - schedule_delayed_work(&mcheck_work, - round_jiffies_relative(next_interval)); - idle_notifier_register(&mce_idle_notifier); - return 0; + idle_notifier_register(&mce_idle_notifier); + return 0; } __initcall(periodic_mcheck_init); - /* * Initialize Machine Checks for a CPU. */ -static void mce_init(void *dummy) +static int mce_cap_init(void) { u64 cap; - int i; + unsigned b; rdmsrl(MSR_IA32_MCG_CAP, cap); - banks = cap & 0xff; - if (banks > MCE_EXTENDED_BANK) { - banks = MCE_EXTENDED_BANK; - printk(KERN_INFO "MCE: warning: using only %d banks\n", - MCE_EXTENDED_BANK); + b = cap & 0xff; + if (b > MAX_NR_BANKS) { + printk(KERN_WARNING + "MCE: Using only %u machine check banks out of %u\n", + MAX_NR_BANKS, b); + b = MAX_NR_BANKS; } + + /* Don't support asymmetric configurations today */ + WARN_ON(banks != 0 && b != banks); + banks = b; + if (!bank) { + bank = kmalloc(banks * sizeof(u64), GFP_KERNEL); + if (!bank) + return -ENOMEM; + memset(bank, 0xff, banks * sizeof(u64)); + } + /* Use accurate RIP reporting if available. */ if ((cap & (1<<9)) && ((cap >> 16) & 0xff) >= 9) rip_msr = MSR_IA32_MCG_EIP; - /* Log the machine checks left over from the previous reset. - This also clears all registers */ - do_machine_check(NULL, mce_bootlog ? -1 : -2); + return 0; +} + +static void mce_init(void *dummy) +{ + u64 cap; + int i; + mce_banks_t all_banks; + + /* + * Log the machine checks left over from the previous reset. + */ + bitmap_fill(all_banks, MAX_NR_BANKS); + machine_check_poll(MCP_UC, &all_banks); set_in_cr4(X86_CR4_MCE); + rdmsrl(MSR_IA32_MCG_CAP, cap); if (cap & MCG_CTL_P) wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); for (i = 0; i < banks; i++) { - if (i < NR_SYSFS_BANKS) - wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]); - else - wrmsrl(MSR_IA32_MC0_CTL+4*i, ~0UL); - + wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]); wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); } } /* Add per CPU specific workarounds here */ -static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c) +static void mce_cpu_quirks(struct cpuinfo_x86 *c) { /* This should be disabled by the BIOS, but isn't always */ if (c->x86_vendor == X86_VENDOR_AMD) { - if(c->x86 == 15) + if (c->x86 == 15 && banks > 4) /* disable GART TBL walk error reporting, which trips off incorrectly with the IOMMU & 3ware & Cerberus. */ - clear_bit(10, &bank[4]); + clear_bit(10, (unsigned long *)&bank[4]); if(c->x86 <= 17 && mce_bootlog < 0) /* Lots of broken BIOS around that don't clear them by default and leave crap in there. Don't log. */ @@ -504,20 +629,38 @@ static void mce_cpu_features(struct cpuinfo_x86 *c) } } +static void mce_init_timer(void) +{ + struct timer_list *t = &__get_cpu_var(mce_timer); + + /* data race harmless because everyone sets to the same value */ + if (!next_interval) + next_interval = check_interval * HZ; + if (!next_interval) + return; + setup_timer(t, mcheck_timer, smp_processor_id()); + t->expires = round_jiffies_relative(jiffies + next_interval); + add_timer(t); +} + /* * Called for each booted CPU to set up machine checks. * Must be called with preempt off. */ void __cpuinit mcheck_init(struct cpuinfo_x86 *c) { - mce_cpu_quirks(c); + if (!mce_available(c)) + return; - if (mce_dont_init || - !mce_available(c)) + if (mce_cap_init() < 0) { + mce_dont_init = 1; return; + } + mce_cpu_quirks(c); mce_init(NULL); mce_cpu_features(c); + mce_init_timer(); } /* @@ -573,7 +716,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, { unsigned long *cpu_tsc; static DEFINE_MUTEX(mce_read_mutex); - unsigned next; + unsigned prev, next; char __user *buf = ubuf; int i, err; @@ -592,25 +735,32 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, } err = 0; - for (i = 0; i < next; i++) { - unsigned long start = jiffies; - - while (!mcelog.entry[i].finished) { - if (time_after_eq(jiffies, start + 2)) { - memset(mcelog.entry + i,0, sizeof(struct mce)); - goto timeout; + prev = 0; + do { + for (i = prev; i < next; i++) { + unsigned long start = jiffies; + + while (!mcelog.entry[i].finished) { + if (time_after_eq(jiffies, start + 2)) { + memset(mcelog.entry + i, 0, + sizeof(struct mce)); + goto timeout; + } + cpu_relax(); } - cpu_relax(); + smp_rmb(); + err |= copy_to_user(buf, mcelog.entry + i, + sizeof(struct mce)); + buf += sizeof(struct mce); +timeout: + ; } - smp_rmb(); - err |= copy_to_user(buf, mcelog.entry + i, sizeof(struct mce)); - buf += sizeof(struct mce); - timeout: - ; - } - memset(mcelog.entry, 0, next * sizeof(struct mce)); - mcelog.next = 0; + memset(mcelog.entry + prev, 0, + (next - prev) * sizeof(struct mce)); + prev = next; + next = cmpxchg(&mcelog.next, prev, 0); + } while (next != prev); synchronize_sched(); @@ -680,20 +830,6 @@ static struct miscdevice mce_log_device = { &mce_chrdev_ops, }; -static unsigned long old_cr4 __initdata; - -void __init stop_mce(void) -{ - old_cr4 = read_cr4(); - clear_in_cr4(X86_CR4_MCE); -} - -void __init restart_mce(void) -{ - if (old_cr4 & X86_CR4_MCE) - set_in_cr4(X86_CR4_MCE); -} - /* * Old style boot options parsing. Only for compatibility. */ @@ -703,8 +839,7 @@ static int __init mcheck_disable(char *str) return 1; } -/* mce=off disables machine check. Note you can re-enable it later - using sysfs. +/* mce=off disables machine check. mce=TOLERANCELEVEL (number, see above) mce=bootlog Log MCEs from before booting. Disabled by default on AMD. mce=nobootlog Don't log MCEs from before booting. */ @@ -728,6 +863,29 @@ __setup("mce=", mcheck_enable); * Sysfs support */ +/* + * Disable machine checks on suspend and shutdown. We can't really handle + * them later. + */ +static int mce_disable(void) +{ + int i; + + for (i = 0; i < banks; i++) + wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); + return 0; +} + +static int mce_suspend(struct sys_device *dev, pm_message_t state) +{ + return mce_disable(); +} + +static int mce_shutdown(struct sys_device *dev) +{ + return mce_disable(); +} + /* On resume clear all MCE state. Don't want to see leftovers from the BIOS. Only one CPU is active at this time, the others get readded later using CPU hotplug. */ @@ -738,20 +896,24 @@ static int mce_resume(struct sys_device *dev) return 0; } +static void mce_cpu_restart(void *data) +{ + del_timer_sync(&__get_cpu_var(mce_timer)); + if (mce_available(¤t_cpu_data)) + mce_init(NULL); + mce_init_timer(); +} + /* Reinit MCEs after user configuration changes */ static void mce_restart(void) { - if (next_interval) - cancel_delayed_work(&mcheck_work); - /* Timer race is harmless here */ - on_each_cpu(mce_init, NULL, 1); next_interval = check_interval * HZ; - if (next_interval) - schedule_delayed_work(&mcheck_work, - round_jiffies_relative(next_interval)); + on_each_cpu(mce_cpu_restart, NULL, 1); } static struct sysdev_class mce_sysclass = { + .suspend = mce_suspend, + .shutdown = mce_shutdown, .resume = mce_resume, .name = "machinecheck", }; @@ -778,16 +940,26 @@ void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu) __cpuinit } \ static SYSDEV_ATTR(name, 0644, show_ ## name, set_ ## name); -/* - * TBD should generate these dynamically based on number of available banks. - * Have only 6 contol banks in /sysfs until then. - */ -ACCESSOR(bank0ctl,bank[0],mce_restart()) -ACCESSOR(bank1ctl,bank[1],mce_restart()) -ACCESSOR(bank2ctl,bank[2],mce_restart()) -ACCESSOR(bank3ctl,bank[3],mce_restart()) -ACCESSOR(bank4ctl,bank[4],mce_restart()) -ACCESSOR(bank5ctl,bank[5],mce_restart()) +static struct sysdev_attribute *bank_attrs; + +static ssize_t show_bank(struct sys_device *s, struct sysdev_attribute *attr, + char *buf) +{ + u64 b = bank[attr - bank_attrs]; + return sprintf(buf, "%llx\n", b); +} + +static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr, + const char *buf, size_t siz) +{ + char *end; + u64 new = simple_strtoull(buf, &end, 0); + if (end == buf) + return -EINVAL; + bank[attr - bank_attrs] = new; + mce_restart(); + return end-buf; +} static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr, char *buf) @@ -814,8 +986,6 @@ static SYSDEV_ATTR(trigger, 0644, show_trigger, set_trigger); static SYSDEV_INT_ATTR(tolerant, 0644, tolerant); ACCESSOR(check_interval,check_interval,mce_restart()) static struct sysdev_attribute *mce_attributes[] = { - &attr_bank0ctl, &attr_bank1ctl, &attr_bank2ctl, - &attr_bank3ctl, &attr_bank4ctl, &attr_bank5ctl, &attr_tolerant.attr, &attr_check_interval, &attr_trigger, NULL }; @@ -845,11 +1015,22 @@ static __cpuinit int mce_create_device(unsigned int cpu) if (err) goto error; } + for (i = 0; i < banks; i++) { + err = sysdev_create_file(&per_cpu(device_mce, cpu), + &bank_attrs[i]); + if (err) + goto error2; + } cpu_set(cpu, mce_device_initialized); return 0; +error2: + while (--i >= 0) { + sysdev_remove_file(&per_cpu(device_mce, cpu), + &bank_attrs[i]); + } error: - while (i--) { + while (--i >= 0) { sysdev_remove_file(&per_cpu(device_mce,cpu), mce_attributes[i]); } @@ -868,15 +1049,46 @@ static __cpuinit void mce_remove_device(unsigned int cpu) for (i = 0; mce_attributes[i]; i++) sysdev_remove_file(&per_cpu(device_mce,cpu), mce_attributes[i]); + for (i = 0; i < banks; i++) + sysdev_remove_file(&per_cpu(device_mce, cpu), + &bank_attrs[i]); sysdev_unregister(&per_cpu(device_mce,cpu)); cpu_clear(cpu, mce_device_initialized); } +/* Make sure there are no machine checks on offlined CPUs. */ +static void mce_disable_cpu(void *h) +{ + int i; + unsigned long action = *(unsigned long *)h; + + if (!mce_available(¤t_cpu_data)) + return; + if (!(action & CPU_TASKS_FROZEN)) + cmci_clear(); + for (i = 0; i < banks; i++) + wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); +} + +static void mce_reenable_cpu(void *h) +{ + int i; + unsigned long action = *(unsigned long *)h; + + if (!mce_available(¤t_cpu_data)) + return; + if (!(action & CPU_TASKS_FROZEN)) + cmci_reenable(); + for (i = 0; i < banks; i++) + wrmsrl(MSR_IA32_MC0_CTL + i*4, bank[i]); +} + /* Get notified when a cpu comes on/off. Be hotplug friendly. */ static int __cpuinit mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; + struct timer_list *t = &per_cpu(mce_timer, cpu); switch (action) { case CPU_ONLINE: @@ -891,6 +1103,21 @@ static int __cpuinit mce_cpu_callback(struct notifier_block *nfb, threshold_cpu_callback(action, cpu); mce_remove_device(cpu); break; + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + del_timer_sync(t); + smp_call_function_single(cpu, mce_disable_cpu, &action, 1); + break; + case CPU_DOWN_FAILED: + case CPU_DOWN_FAILED_FROZEN: + t->expires = round_jiffies_relative(jiffies + next_interval); + add_timer_on(t, cpu); + smp_call_function_single(cpu, mce_reenable_cpu, &action, 1); + break; + case CPU_POST_DEAD: + /* intentionally ignoring frozen here */ + cmci_rediscover(cpu); + break; } return NOTIFY_OK; } @@ -899,6 +1126,34 @@ static struct notifier_block mce_cpu_notifier __cpuinitdata = { .notifier_call = mce_cpu_callback, }; +static __init int mce_init_banks(void) +{ + int i; + + bank_attrs = kzalloc(sizeof(struct sysdev_attribute) * banks, + GFP_KERNEL); + if (!bank_attrs) + return -ENOMEM; + + for (i = 0; i < banks; i++) { + struct sysdev_attribute *a = &bank_attrs[i]; + a->attr.name = kasprintf(GFP_KERNEL, "bank%d", i); + if (!a->attr.name) + goto nomem; + a->attr.mode = 0644; + a->show = show_bank; + a->store = set_bank; + } + return 0; + +nomem: + while (--i >= 0) + kfree(bank_attrs[i].attr.name); + kfree(bank_attrs); + bank_attrs = NULL; + return -ENOMEM; +} + static __init int mce_init_device(void) { int err; @@ -906,6 +1161,11 @@ static __init int mce_init_device(void) if (!mce_available(&boot_cpu_data)) return -EIO; + + err = mce_init_banks(); + if (err) + return err; + err = sysdev_class_register(&mce_sysclass); if (err) return err; diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c index 9817506dd469..c5a32f92d07e 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c @@ -79,6 +79,8 @@ static unsigned char shared_bank[NR_BANKS] = { static DEFINE_PER_CPU(unsigned char, bank_map); /* see which banks are on */ +static void amd_threshold_interrupt(void); + /* * CPU Initialization */ @@ -174,6 +176,8 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) tr.reset = 0; tr.old_limit = 0; threshold_restart_bank(&tr); + + mce_threshold_vector = amd_threshold_interrupt; } } } @@ -187,19 +191,13 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) * the interrupt goes off when error_count reaches threshold_limit. * the handler will simply log mcelog w/ software defined bank number. */ -asmlinkage void mce_threshold_interrupt(void) +static void amd_threshold_interrupt(void) { unsigned int bank, block; struct mce m; u32 low = 0, high = 0, address = 0; - ack_APIC_irq(); - exit_idle(); - irq_enter(); - - memset(&m, 0, sizeof(m)); - rdtscll(m.tsc); - m.cpu = smp_processor_id(); + mce_setup(&m); /* assume first bank caused it */ for (bank = 0; bank < NR_BANKS; ++bank) { @@ -233,7 +231,8 @@ asmlinkage void mce_threshold_interrupt(void) /* Log the machine check that caused the threshold event. */ - do_machine_check(NULL, 0); + machine_check_poll(MCP_TIMESTAMP, + &__get_cpu_var(mce_poll_banks)); if (high & MASK_OVERFLOW_HI) { rdmsrl(address, m.misc); @@ -243,13 +242,10 @@ asmlinkage void mce_threshold_interrupt(void) + bank * NR_BLOCKS + block; mce_log(&m); - goto out; + return; } } } -out: - inc_irq_stat(irq_threshold_count); - irq_exit(); } /* diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c index aa5e287c98e0..aaa7d9730938 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c @@ -1,6 +1,8 @@ /* * Intel specific MCE features. * Copyright 2004 Zwane Mwaikambo <zwane@linuxpower.ca> + * Copyright (C) 2008, 2009 Intel Corporation + * Author: Andi Kleen */ #include <linux/init.h> @@ -13,6 +15,7 @@ #include <asm/hw_irq.h> #include <asm/idle.h> #include <asm/therm_throt.h> +#include <asm/apic.h> asmlinkage void smp_thermal_interrupt(void) { @@ -25,7 +28,7 @@ asmlinkage void smp_thermal_interrupt(void) rdmsrl(MSR_IA32_THERM_STATUS, msr_val); if (therm_throt_process(msr_val & 1)) - mce_log_therm_throt_event(smp_processor_id(), msr_val); + mce_log_therm_throt_event(msr_val); inc_irq_stat(irq_thermal_count); irq_exit(); @@ -85,7 +88,209 @@ static void intel_init_thermal(struct cpuinfo_x86 *c) return; } +/* + * Support for Intel Correct Machine Check Interrupts. This allows + * the CPU to raise an interrupt when a corrected machine check happened. + * Normally we pick those up using a regular polling timer. + * Also supports reliable discovery of shared banks. + */ + +static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned); + +/* + * cmci_discover_lock protects against parallel discovery attempts + * which could race against each other. + */ +static DEFINE_SPINLOCK(cmci_discover_lock); + +#define CMCI_THRESHOLD 1 + +static int cmci_supported(int *banks) +{ + u64 cap; + + /* + * Vendor check is not strictly needed, but the initial + * initialization is vendor keyed and this + * makes sure none of the backdoors are entered otherwise. + */ + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) + return 0; + if (!cpu_has_apic || lapic_get_maxlvt() < 6) + return 0; + rdmsrl(MSR_IA32_MCG_CAP, cap); + *banks = min_t(unsigned, MAX_NR_BANKS, cap & 0xff); + return !!(cap & MCG_CMCI_P); +} + +/* + * The interrupt handler. This is called on every event. + * Just call the poller directly to log any events. + * This could in theory increase the threshold under high load, + * but doesn't for now. + */ +static void intel_threshold_interrupt(void) +{ + machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); + mce_notify_user(); +} + +static void print_update(char *type, int *hdr, int num) +{ + if (*hdr == 0) + printk(KERN_INFO "CPU %d MCA banks", smp_processor_id()); + *hdr = 1; + printk(KERN_CONT " %s:%d", type, num); +} + +/* + * Enable CMCI (Corrected Machine Check Interrupt) for available MCE banks + * on this CPU. Use the algorithm recommended in the SDM to discover shared + * banks. + */ +static void cmci_discover(int banks, int boot) +{ + unsigned long *owned = (void *)&__get_cpu_var(mce_banks_owned); + int hdr = 0; + int i; + + spin_lock(&cmci_discover_lock); + for (i = 0; i < banks; i++) { + u64 val; + + if (test_bit(i, owned)) + continue; + + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + + /* Already owned by someone else? */ + if (val & CMCI_EN) { + if (test_and_clear_bit(i, owned) || boot) + print_update("SHD", &hdr, i); + __clear_bit(i, __get_cpu_var(mce_poll_banks)); + continue; + } + + val |= CMCI_EN | CMCI_THRESHOLD; + wrmsrl(MSR_IA32_MC0_CTL2 + i, val); + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + + /* Did the enable bit stick? -- the bank supports CMCI */ + if (val & CMCI_EN) { + if (!test_and_set_bit(i, owned) || boot) + print_update("CMCI", &hdr, i); + __clear_bit(i, __get_cpu_var(mce_poll_banks)); + } else { + WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks))); + } + } + spin_unlock(&cmci_discover_lock); + if (hdr) + printk(KERN_CONT "\n"); +} + +/* + * Just in case we missed an event during initialization check + * all the CMCI owned banks. + */ +void cmci_recheck(void) +{ + unsigned long flags; + int banks; + + if (!mce_available(¤t_cpu_data) || !cmci_supported(&banks)) + return; + local_irq_save(flags); + machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); + local_irq_restore(flags); +} + +/* + * Disable CMCI on this CPU for all banks it owns when it goes down. + * This allows other CPUs to claim the banks on rediscovery. + */ +void cmci_clear(void) +{ + int i; + int banks; + u64 val; + + if (!cmci_supported(&banks)) + return; + spin_lock(&cmci_discover_lock); + for (i = 0; i < banks; i++) { + if (!test_bit(i, __get_cpu_var(mce_banks_owned))) + continue; + /* Disable CMCI */ + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + val &= ~(CMCI_EN|CMCI_THRESHOLD_MASK); + wrmsrl(MSR_IA32_MC0_CTL2 + i, val); + __clear_bit(i, __get_cpu_var(mce_banks_owned)); + } + spin_unlock(&cmci_discover_lock); +} + +/* + * After a CPU went down cycle through all the others and rediscover + * Must run in process context. + */ +void cmci_rediscover(int dying) +{ + int banks; + int cpu; + cpumask_var_t old; + + if (!cmci_supported(&banks)) + return; + if (!alloc_cpumask_var(&old, GFP_KERNEL)) + return; + cpumask_copy(old, ¤t->cpus_allowed); + + for_each_online_cpu (cpu) { + if (cpu == dying) + continue; + if (set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu))) + continue; + /* Recheck banks in case CPUs don't all have the same */ + if (cmci_supported(&banks)) + cmci_discover(banks, 0); + } + + set_cpus_allowed_ptr(current, old); + free_cpumask_var(old); +} + +/* + * Reenable CMCI on this CPU in case a CPU down failed. + */ +void cmci_reenable(void) +{ + int banks; + if (cmci_supported(&banks)) + cmci_discover(banks, 0); +} + +static __cpuinit void intel_init_cmci(void) +{ + int banks; + + if (!cmci_supported(&banks)) + return; + + mce_threshold_vector = intel_threshold_interrupt; + cmci_discover(banks, 1); + /* + * For CPU #0 this runs with still disabled APIC, but that's + * ok because only the vector is set up. We still do another + * check for the banks later for CPU #0 just to make sure + * to not miss any events. + */ + apic_write(APIC_LVTCMCI, THRESHOLD_APIC_VECTOR|APIC_DM_FIXED); + cmci_recheck(); +} + void mce_intel_feature_init(struct cpuinfo_x86 *c) { intel_init_thermal(c); + intel_init_cmci(); } diff --git a/arch/x86/kernel/cpu/mcheck/threshold.c b/arch/x86/kernel/cpu/mcheck/threshold.c new file mode 100644 index 000000000000..23ee9e730f78 --- /dev/null +++ b/arch/x86/kernel/cpu/mcheck/threshold.c @@ -0,0 +1,29 @@ +/* + * Common corrected MCE threshold handler code: + */ +#include <linux/interrupt.h> +#include <linux/kernel.h> + +#include <asm/irq_vectors.h> +#include <asm/apic.h> +#include <asm/idle.h> +#include <asm/mce.h> + +static void default_threshold_interrupt(void) +{ + printk(KERN_ERR "Unexpected threshold interrupt at vector %x\n", + THRESHOLD_APIC_VECTOR); +} + +void (*mce_threshold_vector)(void) = default_threshold_interrupt; + +asmlinkage void mce_threshold_interrupt(void) +{ + exit_idle(); + irq_enter(); + inc_irq_stat(irq_threshold_count); + mce_threshold_vector(); + irq_exit(); + /* Ack only at the end to avoid potential reentry */ + ack_APIC_irq(); +} diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index b205272ad394..1736acc4d7aa 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -469,7 +469,7 @@ void __init efi_enter_virtual_mode(void) efi_memory_desc_t *md; efi_status_t status; unsigned long size; - u64 end, systab, addr, npages; + u64 end, systab, addr, npages, end_pfn; void *p, *va; efi.systab = NULL; @@ -481,7 +481,10 @@ void __init efi_enter_virtual_mode(void) size = md->num_pages << EFI_PAGE_SHIFT; end = md->phys_addr + size; - if (PFN_UP(end) <= max_low_pfn_mapped) + end_pfn = PFN_UP(end); + if (end_pfn <= max_low_pfn_mapped + || (end_pfn > (1UL << (32 - PAGE_SHIFT)) + && end_pfn <= max_pfn_mapped)) va = __va(md->phys_addr); else va = efi_ioremap(md->phys_addr, size); diff --git a/arch/x86/kernel/efi_64.c b/arch/x86/kernel/efi_64.c index a4ee29127fdf..22c3b7828c50 100644 --- a/arch/x86/kernel/efi_64.c +++ b/arch/x86/kernel/efi_64.c @@ -100,24 +100,11 @@ void __init efi_call_phys_epilog(void) void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size) { - static unsigned pages_mapped __initdata; - unsigned i, pages; - unsigned long offset; + unsigned long last_map_pfn; - pages = PFN_UP(phys_addr + size) - PFN_DOWN(phys_addr); - offset = phys_addr & ~PAGE_MASK; - phys_addr &= PAGE_MASK; - - if (pages_mapped + pages > MAX_EFI_IO_PAGES) + last_map_pfn = init_memory_mapping(phys_addr, phys_addr + size); + if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size) return NULL; - for (i = 0; i < pages; i++) { - __set_fixmap(FIX_EFI_IO_MAP_FIRST_PAGE - pages_mapped, - phys_addr, PAGE_KERNEL); - phys_addr += PAGE_SIZE; - pages_mapped++; - } - - return (void __iomem *)__fix_to_virt(FIX_EFI_IO_MAP_FIRST_PAGE - \ - (pages_mapped - pages)) + offset; + return (void __iomem *)__va(phys_addr); } diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index b0f61f0dcd0a..f2f8540a7f3d 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -136,7 +136,7 @@ int init_fpu(struct task_struct *tsk) #ifdef CONFIG_X86_32 if (!HAVE_HWFP) { memset(tsk->thread.xstate, 0, xstate_size); - finit(); + finit_task(tsk); set_stopped_child_used_math(tsk); return 0; } diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 37cb1bda1baf..e8192401da47 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -558,6 +558,19 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type) static struct mpf_intel *mpf_found; +static unsigned long __init get_mpc_size(unsigned long physptr) +{ + struct mpc_table *mpc; + unsigned long size; + + mpc = early_ioremap(physptr, PAGE_SIZE); + size = mpc->length; + early_iounmap(mpc, PAGE_SIZE); + apic_printk(APIC_VERBOSE, " mpc: %lx-%lx\n", physptr, physptr + size); + + return size; +} + /* * Scan the memory blocks for an SMP configuration block. */ @@ -611,12 +624,16 @@ static void __init __get_smp_config(unsigned int early) construct_default_ISA_mptable(mpf->feature1); } else if (mpf->physptr) { + struct mpc_table *mpc; + unsigned long size; + size = get_mpc_size(mpf->physptr); + mpc = early_ioremap(mpf->physptr, size); /* * Read the physical hardware table. Anything here will * override the defaults. */ - if (!smp_read_mpc(phys_to_virt(mpf->physptr), early)) { + if (!smp_read_mpc(mpc, early)) { #ifdef CONFIG_X86_LOCAL_APIC smp_found_config = 0; #endif @@ -624,8 +641,10 @@ static void __init __get_smp_config(unsigned int early) "BIOS bug, MP table errors detected!...\n"); printk(KERN_ERR "... disabling SMP support. " "(tell your hw vendor)\n"); + early_iounmap(mpc, size); return; } + early_iounmap(mpc, size); if (early) return; @@ -697,10 +716,10 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, if (!reserve) return 1; - reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE, + reserve_bootmem_generic(virt_to_phys(mpf), sizeof(*mpf), BOOTMEM_DEFAULT); if (mpf->physptr) { - unsigned long size = PAGE_SIZE; + unsigned long size = get_mpc_size(mpf->physptr); #ifdef CONFIG_X86_32 /* * We cannot access to MPC table to compute diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 1cc18d439bbb..2aef36d8aca2 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -216,6 +216,14 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq"), }, }, + { /* Handle problems with rebooting on Dell XPS710 */ + .callback = set_bios_reboot, + .ident = "Dell XPS710", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "Dell XPS710"), + }, + }, { } }; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 4c54bc0d8ff3..f28c56e6bf94 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -202,7 +202,9 @@ struct ist_info ist_info; #endif #else -struct cpuinfo_x86 boot_cpu_data __read_mostly; +struct cpuinfo_x86 boot_cpu_data __read_mostly = { + .x86_phys_bits = MAX_PHYSMEM_BITS, +}; EXPORT_SYMBOL(boot_cpu_data); #endif @@ -770,6 +772,9 @@ void __init setup_arch(char **cmdline_p) finish_e820_parsing(); + if (efi_enabled) + efi_init(); + dmi_scan_machine(); dmi_check_system(bad_bios_dmi_table); @@ -789,8 +794,6 @@ void __init setup_arch(char **cmdline_p) insert_resource(&iomem_resource, &data_resource); insert_resource(&iomem_resource, &bss_resource); - if (efi_enabled) - efi_init(); #ifdef CONFIG_X86_32 if (ppro_with_ram_bug()) { diff --git a/arch/x86/math-emu/fpu_aux.c b/arch/x86/math-emu/fpu_aux.c index 491e737ce547..aa0987088774 100644 --- a/arch/x86/math-emu/fpu_aux.c +++ b/arch/x86/math-emu/fpu_aux.c @@ -30,20 +30,29 @@ static void fclex(void) } /* Needs to be externally visible */ -void finit(void) +void finit_task(struct task_struct *tsk) { - control_word = 0x037f; - partial_status = 0; - top = 0; /* We don't keep top in the status word internally. */ - fpu_tag_word = 0xffff; + struct i387_soft_struct *soft = &tsk->thread.xstate->soft; + struct address *oaddr, *iaddr; + soft->cwd = 0x037f; + soft->swd = 0; + soft->ftop = 0; /* We don't keep top in the status word internally. */ + soft->twd = 0xffff; /* The behaviour is different from that detailed in Section 15.1.6 of the Intel manual */ - operand_address.offset = 0; - operand_address.selector = 0; - instruction_address.offset = 0; - instruction_address.selector = 0; - instruction_address.opcode = 0; - no_ip_update = 1; + oaddr = (struct address *)&soft->foo; + oaddr->offset = 0; + oaddr->selector = 0; + iaddr = (struct address *)&soft->fip; + iaddr->offset = 0; + iaddr->selector = 0; + iaddr->opcode = 0; + soft->no_update = 1; +} + +void finit(void) +{ + finit_task(current); } /* diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c index 00f127c80b0e..d11745334a67 100644 --- a/arch/x86/mm/highmem_32.c +++ b/arch/x86/mm/highmem_32.c @@ -158,7 +158,6 @@ EXPORT_SYMBOL(kunmap); EXPORT_SYMBOL(kmap_atomic); EXPORT_SYMBOL(kunmap_atomic); -#ifdef CONFIG_NUMA void __init set_highmem_pages_init(void) { struct zone *zone; @@ -182,11 +181,3 @@ void __init set_highmem_pages_init(void) } totalram_pages += totalhigh_pages; } -#else -void __init set_highmem_pages_init(void) -{ - add_highpages_with_active_regions(0, highstart_pfn, highend_pfn); - - totalram_pages += totalhigh_pages; -} -#endif /* CONFIG_NUMA */ diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index ce6a722587d8..6d63e3d1253d 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -1,8 +1,345 @@ +#include <linux/ioport.h> #include <linux/swap.h> + #include <asm/cacheflush.h> +#include <asm/e820.h> +#include <asm/init.h> #include <asm/page.h> +#include <asm/page_types.h> #include <asm/sections.h> #include <asm/system.h> +#include <asm/tlbflush.h> + +unsigned long __initdata e820_table_start; +unsigned long __meminitdata e820_table_end; +unsigned long __meminitdata e820_table_top; + +int after_bootmem; + +int direct_gbpages +#ifdef CONFIG_DIRECT_GBPAGES + = 1 +#endif +; + +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 = 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 += roundup(pmds * sizeof(pmd_t), PAGE_SIZE); + + if (use_pse) { + unsigned long extra; + + extra = end - ((end>>PMD_SHIFT) << PMD_SHIFT); +#ifdef CONFIG_X86_32 + extra += PMD_SIZE; +#endif + ptes = (extra + PAGE_SIZE - 1) >> PAGE_SHIFT; + } else + ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT; + + tables += roundup(ptes * sizeof(pte_t), PAGE_SIZE); + +#ifdef CONFIG_X86_32 + /* for fixmap */ + tables += roundup(__end_of_fixed_addresses * sizeof(pte_t), PAGE_SIZE); +#endif + + /* + * RED-PEN putting page tables only on node 0 could + * cause a hotspot and fill up ZONE_DMA. The page tables + * need roughly 0.5KB per GB. + */ +#ifdef CONFIG_X86_32 + start = 0x7000; + e820_table_start = find_e820_area(start, max_pfn_mapped<<PAGE_SHIFT, + tables, PAGE_SIZE); +#else /* CONFIG_X86_64 */ + start = 0x8000; + e820_table_start = find_e820_area(start, end, tables, PAGE_SIZE); +#endif + if (e820_table_start == -1UL) + panic("Cannot find space for the kernel page tables"); + + e820_table_start >>= PAGE_SHIFT; + e820_table_end = e820_table_start; + e820_table_top = e820_table_start + (tables >> PAGE_SHIFT); + + printk(KERN_DEBUG "kernel direct mapping tables up to %lx @ %lx-%lx\n", + end, e820_table_start << PAGE_SHIFT, e820_table_top << PAGE_SHIFT); +} + +struct map_range { + unsigned long start; + unsigned long end; + unsigned page_size_mask; +}; + +#ifdef CONFIG_X86_32 +#define NR_RANGE_MR 3 +#else /* CONFIG_X86_64 */ +#define NR_RANGE_MR 5 +#endif + +static int save_mr(struct map_range *mr, int nr_range, + unsigned long start_pfn, unsigned long end_pfn, + unsigned long page_size_mask) +{ + if (start_pfn < end_pfn) { + if (nr_range >= NR_RANGE_MR) + panic("run out of range for init_memory_mapping\n"); + mr[nr_range].start = start_pfn<<PAGE_SHIFT; + mr[nr_range].end = end_pfn<<PAGE_SHIFT; + mr[nr_range].page_size_mask = page_size_mask; + nr_range++; + } + + return nr_range; +} + +#ifdef CONFIG_X86_64 +static void __init init_gbpages(void) +{ + if (direct_gbpages && cpu_has_gbpages) + printk(KERN_INFO "Using GB pages for direct mapping\n"); + else + direct_gbpages = 0; +} +#else +static inline void init_gbpages(void) +{ +} +#endif + +/* + * Setup the direct mapping of the physical memory at PAGE_OFFSET. + * This runs before bootmem is initialized and gets pages directly from + * the physical memory. To access them they are temporarily mapped. + */ +unsigned long __init_refok init_memory_mapping(unsigned long start, + unsigned long end) +{ + unsigned long page_size_mask = 0; + unsigned long start_pfn, end_pfn; + unsigned long pos; + unsigned long ret; + + struct map_range mr[NR_RANGE_MR]; + int nr_range, i; + int use_pse, use_gbpages; + + printk(KERN_INFO "init_memory_mapping: %016lx-%016lx\n", start, end); + + if (!after_bootmem) + init_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 + +#ifdef CONFIG_X86_32 +#ifdef CONFIG_X86_PAE + set_nx(); + if (nx_enabled) + printk(KERN_INFO "NX (Execute Disable) protection: active\n"); +#endif + + /* Enable PSE if available */ + if (cpu_has_pse) + set_in_cr4(X86_CR4_PSE); + + /* Enable PGE if available */ + if (cpu_has_pge) { + set_in_cr4(X86_CR4_PGE); + __supported_pte_mask |= _PAGE_GLOBAL; + } +#endif + + if (use_gbpages) + page_size_mask |= 1 << PG_LEVEL_1G; + if (use_pse) + page_size_mask |= 1 << PG_LEVEL_2M; + + memset(mr, 0, sizeof(mr)); + nr_range = 0; + + /* head if not big page alignment ? */ + start_pfn = start >> PAGE_SHIFT; + pos = start_pfn << PAGE_SHIFT; +#ifdef CONFIG_X86_32 + /* + * Don't use a large page for the first 2/4MB of memory + * because there are often fixed size MTRRs in there + * and overlapping MTRRs into large pages can cause + * slowdowns. + */ + if (pos == 0) + end_pfn = 1<<(PMD_SHIFT - PAGE_SHIFT); + else + end_pfn = ((pos + (PMD_SIZE - 1))>>PMD_SHIFT) + << (PMD_SHIFT - PAGE_SHIFT); +#else /* CONFIG_X86_64 */ + end_pfn = ((pos + (PMD_SIZE - 1)) >> PMD_SHIFT) + << (PMD_SHIFT - PAGE_SHIFT); +#endif + if (end_pfn > (end >> PAGE_SHIFT)) + end_pfn = end >> PAGE_SHIFT; + if (start_pfn < end_pfn) { + nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0); + pos = end_pfn << PAGE_SHIFT; + } + + /* big page (2M) range */ + start_pfn = ((pos + (PMD_SIZE - 1))>>PMD_SHIFT) + << (PMD_SHIFT - PAGE_SHIFT); +#ifdef CONFIG_X86_32 + end_pfn = (end>>PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT); +#else /* CONFIG_X86_64 */ + end_pfn = ((pos + (PUD_SIZE - 1))>>PUD_SHIFT) + << (PUD_SHIFT - PAGE_SHIFT); + if (end_pfn > ((end>>PMD_SHIFT)<<(PMD_SHIFT - PAGE_SHIFT))) + end_pfn = ((end>>PMD_SHIFT)<<(PMD_SHIFT - PAGE_SHIFT)); +#endif + + if (start_pfn < end_pfn) { + nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, + page_size_mask & (1<<PG_LEVEL_2M)); + pos = end_pfn << PAGE_SHIFT; + } + +#ifdef CONFIG_X86_64 + /* big page (1G) range */ + start_pfn = ((pos + (PUD_SIZE - 1))>>PUD_SHIFT) + << (PUD_SHIFT - PAGE_SHIFT); + end_pfn = (end >> PUD_SHIFT) << (PUD_SHIFT - PAGE_SHIFT); + if (start_pfn < end_pfn) { + nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, + page_size_mask & + ((1<<PG_LEVEL_2M)|(1<<PG_LEVEL_1G))); + pos = end_pfn << PAGE_SHIFT; + } + + /* tail is not big page (1G) alignment */ + start_pfn = ((pos + (PMD_SIZE - 1))>>PMD_SHIFT) + << (PMD_SHIFT - PAGE_SHIFT); + end_pfn = (end >> PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT); + if (start_pfn < end_pfn) { + nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, + page_size_mask & (1<<PG_LEVEL_2M)); + pos = end_pfn << PAGE_SHIFT; + } +#endif + + /* tail is not big page (2M) alignment */ + start_pfn = pos>>PAGE_SHIFT; + end_pfn = end>>PAGE_SHIFT; + nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0); + + /* try to merge same page size and continuous */ + for (i = 0; nr_range > 1 && i < nr_range - 1; i++) { + unsigned long old_start; + if (mr[i].end != mr[i+1].start || + mr[i].page_size_mask != mr[i+1].page_size_mask) + continue; + /* move it */ + old_start = mr[i].start; + memmove(&mr[i], &mr[i+1], + (nr_range - 1 - i) * sizeof(struct map_range)); + mr[i--].start = old_start; + nr_range--; + } + + for (i = 0; i < nr_range; i++) + printk(KERN_DEBUG " %010lx - %010lx page %s\n", + mr[i].start, mr[i].end, + (mr[i].page_size_mask & (1<<PG_LEVEL_1G))?"1G":( + (mr[i].page_size_mask & (1<<PG_LEVEL_2M))?"2M":"4k")); + + /* + * Find space for the kernel direct mapping tables. + * + * Later we should allocate these tables in the local node of the + * memory mapped. Unfortunately this is done currently before the + * nodes are discovered. + */ + if (!after_bootmem) + find_early_table_space(end, use_pse, use_gbpages); + +#ifdef CONFIG_X86_32 + for (i = 0; i < nr_range; i++) + kernel_physical_mapping_init(mr[i].start, mr[i].end, + mr[i].page_size_mask); + ret = end; +#else /* CONFIG_X86_64 */ + for (i = 0; i < nr_range; i++) + ret = kernel_physical_mapping_init(mr[i].start, mr[i].end, + mr[i].page_size_mask); +#endif + +#ifdef CONFIG_X86_32 + early_ioremap_page_table_range_init(); + + load_cr3(swapper_pg_dir); +#endif + +#ifdef CONFIG_X86_64 + if (!after_bootmem) + mmu_cr4_features = read_cr4(); +#endif + __flush_tlb_all(); + + if (!after_bootmem && e820_table_end > e820_table_start) + reserve_early(e820_table_start << PAGE_SHIFT, + e820_table_end << PAGE_SHIFT, "PGTABLE"); + + if (!after_bootmem) + early_memtest(start, end); + + return ret >> PAGE_SHIFT; +} + + +/* + * devmem_is_allowed() checks to see if /dev/mem access to a certain address + * is valid. The argument is a physical page number. + * + * + * On x86, access has to be given to the first megabyte of ram because that area + * contains bios code and data regions used by X and dosemu and similar apps. + * Access has to be given to non-kernel-ram areas as well, these contain the PCI + * mmio resources as well as potential bios/acpi data regions. + */ +int devmem_is_allowed(unsigned long pagenr) +{ + if (pagenr <= 256) + return 1; + if (iomem_is_exclusive(pagenr << PAGE_SHIFT)) + return 0; + if (!page_is_ram(pagenr)) + return 1; + return 0; +} void free_init_pages(char *what, unsigned long begin, unsigned long end) { @@ -47,3 +384,10 @@ void free_initmem(void) (unsigned long)(&__init_begin), (unsigned long)(&__init_end)); } + +#ifdef CONFIG_BLK_DEV_INITRD +void free_initrd_mem(unsigned long start, unsigned long end) +{ + free_init_pages("initrd memory", start, end); +} +#endif diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 47df0e1bbeb9..2966c6b8d304 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -49,6 +49,7 @@ #include <asm/paravirt.h> #include <asm/setup.h> #include <asm/cacheflush.h> +#include <asm/init.h> unsigned long max_low_pfn_mapped; unsigned long max_pfn_mapped; @@ -58,19 +59,14 @@ unsigned long highstart_pfn, highend_pfn; static noinline int do_test_wp_bit(void); - -static unsigned long __initdata table_start; -static unsigned long __meminitdata table_end; -static unsigned long __meminitdata table_top; - -static int __initdata after_init_bootmem; +bool __read_mostly __vmalloc_start_set = false; static __init void *alloc_low_page(void) { - unsigned long pfn = table_end++; + unsigned long pfn = e820_table_end++; void *adr; - if (pfn >= table_top) + if (pfn >= e820_table_top) panic("alloc_low_page: ran out of memory"); adr = __va(pfn * PAGE_SIZE); @@ -90,7 +86,7 @@ static pmd_t * __init one_md_table_init(pgd_t *pgd) #ifdef CONFIG_X86_PAE if (!(pgd_val(*pgd) & _PAGE_PRESENT)) { - if (after_init_bootmem) + if (after_bootmem) pmd_table = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE); else pmd_table = (pmd_t *)alloc_low_page(); @@ -117,7 +113,7 @@ static pte_t * __init one_page_table_init(pmd_t *pmd) if (!(pmd_val(*pmd) & _PAGE_PRESENT)) { pte_t *page_table = NULL; - if (after_init_bootmem) { + if (after_bootmem) { #ifdef CONFIG_DEBUG_PAGEALLOC page_table = (pte_t *) alloc_bootmem_pages(PAGE_SIZE); #endif @@ -168,12 +164,12 @@ static pte_t *__init page_table_kmap_check(pte_t *pte, pmd_t *pmd, if (pmd_idx_kmap_begin != pmd_idx_kmap_end && (vaddr >> PMD_SHIFT) >= pmd_idx_kmap_begin && (vaddr >> PMD_SHIFT) <= pmd_idx_kmap_end - && ((__pa(pte) >> PAGE_SHIFT) < table_start - || (__pa(pte) >> PAGE_SHIFT) >= table_end)) { + && ((__pa(pte) >> PAGE_SHIFT) < e820_table_start + || (__pa(pte) >> PAGE_SHIFT) >= e820_table_end)) { pte_t *newpte; int i; - BUG_ON(after_init_bootmem); + BUG_ON(after_bootmem); newpte = alloc_low_page(); for (i = 0; i < PTRS_PER_PTE; i++) set_pte(newpte + i, pte[i]); @@ -242,11 +238,14 @@ static inline int is_kernel_text(unsigned long addr) * of max_low_pfn pages, by creating page tables starting from address * PAGE_OFFSET: */ -static void __init kernel_physical_mapping_init(pgd_t *pgd_base, - unsigned long start_pfn, - unsigned long end_pfn, - int use_pse) +unsigned long __init +kernel_physical_mapping_init(unsigned long start, + unsigned long end, + unsigned long page_size_mask) { + int use_pse = page_size_mask == (1<<PG_LEVEL_2M); + unsigned long start_pfn, end_pfn; + pgd_t *pgd_base = swapper_pg_dir; int pgd_idx, pmd_idx, pte_ofs; unsigned long pfn; pgd_t *pgd; @@ -255,6 +254,9 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base, unsigned pages_2m, pages_4k; int mapping_iter; + start_pfn = start >> PAGE_SHIFT; + end_pfn = end >> PAGE_SHIFT; + /* * First iteration will setup identity mapping using large/small pages * based on use_pse, with other attributes same as set by @@ -369,26 +371,6 @@ repeat: mapping_iter = 2; goto repeat; } -} - -/* - * devmem_is_allowed() checks to see if /dev/mem access to a certain address - * is valid. The argument is a physical page number. - * - * - * On x86, access has to be given to the first megabyte of ram because that area - * contains bios code and data regions used by X and dosemu and similar apps. - * Access has to be given to non-kernel-ram areas as well, these contain the PCI - * mmio resources as well as potential bios/acpi data regions. - */ -int devmem_is_allowed(unsigned long pagenr) -{ - if (pagenr <= 256) - return 1; - if (iomem_is_exclusive(pagenr << PAGE_SHIFT)) - return 0; - if (!page_is_ram(pagenr)) - return 1; return 0; } @@ -545,8 +527,9 @@ void __init native_pagetable_setup_done(pgd_t *base) * be partially populated, and so it avoids stomping on any existing * mappings. */ -static void __init early_ioremap_page_table_range_init(pgd_t *pgd_base) +void __init early_ioremap_page_table_range_init(void) { + pgd_t *pgd_base = swapper_pg_dir; unsigned long vaddr, end; /* @@ -641,7 +624,7 @@ static int __init noexec_setup(char *str) } early_param("noexec", noexec_setup); -static void __init set_nx(void) +void __init set_nx(void) { unsigned int v[4], l, h; @@ -793,6 +776,8 @@ void __init initmem_init(unsigned long start_pfn, #ifdef CONFIG_FLATMEM max_mapnr = num_physpages; #endif + __vmalloc_start_set = true; + printk(KERN_NOTICE "%ldMB LOWMEM available.\n", pages_to_mb(max_low_pfn)); @@ -814,176 +799,61 @@ static void __init zone_sizes_init(void) free_area_init_nodes(max_zone_pfns); } +static unsigned long __init setup_node_bootmem(int nodeid, + unsigned long start_pfn, + unsigned long end_pfn, + unsigned long bootmap) +{ + unsigned long bootmap_size; + + if (start_pfn > max_low_pfn) + return bootmap; + if (end_pfn > max_low_pfn) + end_pfn = max_low_pfn; + + /* don't touch min_low_pfn */ + bootmap_size = init_bootmem_node(NODE_DATA(nodeid), + bootmap >> PAGE_SHIFT, + start_pfn, end_pfn); + printk(KERN_INFO " node %d low ram: %08lx - %08lx\n", + nodeid, start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT); + printk(KERN_INFO " node %d bootmap %08lx - %08lx\n", + nodeid, bootmap, bootmap + bootmap_size); + free_bootmem_with_active_regions(nodeid, end_pfn); + early_res_to_bootmem(start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT); + + return bootmap + bootmap_size; +} + void __init setup_bootmem_allocator(void) { - int i; + int nodeid; unsigned long bootmap_size, bootmap; /* * Initialize the boot-time allocator (with low memory only): */ bootmap_size = bootmem_bootmap_pages(max_low_pfn)<<PAGE_SHIFT; - bootmap = find_e820_area(min_low_pfn<<PAGE_SHIFT, - max_pfn_mapped<<PAGE_SHIFT, bootmap_size, + bootmap = find_e820_area(0, max_pfn_mapped<<PAGE_SHIFT, bootmap_size, PAGE_SIZE); if (bootmap == -1L) panic("Cannot find bootmem map of size %ld\n", bootmap_size); reserve_early(bootmap, bootmap + bootmap_size, "BOOTMAP"); - /* don't touch min_low_pfn */ - bootmap_size = init_bootmem_node(NODE_DATA(0), bootmap >> PAGE_SHIFT, - min_low_pfn, max_low_pfn); printk(KERN_INFO " mapped low ram: 0 - %08lx\n", max_pfn_mapped<<PAGE_SHIFT); - printk(KERN_INFO " low ram: %08lx - %08lx\n", - min_low_pfn<<PAGE_SHIFT, max_low_pfn<<PAGE_SHIFT); - printk(KERN_INFO " bootmap %08lx - %08lx\n", - bootmap, bootmap + bootmap_size); - for_each_online_node(i) - free_bootmem_with_active_regions(i, max_low_pfn); - early_res_to_bootmem(0, max_low_pfn<<PAGE_SHIFT); - - after_init_bootmem = 1; -} - -static void __init find_early_table_space(unsigned long end, int use_pse) -{ - unsigned long puds, pmds, ptes, tables, start; - - puds = (end + PUD_SIZE - 1) >> PUD_SHIFT; - tables = roundup(puds * sizeof(pud_t), PAGE_SIZE); - - pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT; - tables += roundup(pmds * sizeof(pmd_t), PAGE_SIZE); - - if (use_pse) { - unsigned long extra; - - extra = end - ((end>>PMD_SHIFT) << PMD_SHIFT); - extra += PMD_SIZE; - ptes = (extra + PAGE_SIZE - 1) >> PAGE_SHIFT; - } else - ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT; + printk(KERN_INFO " low ram: 0 - %08lx\n", max_low_pfn<<PAGE_SHIFT); - tables += roundup(ptes * sizeof(pte_t), PAGE_SIZE); - - /* for fixmap */ - tables += roundup(__end_of_fixed_addresses * sizeof(pte_t), PAGE_SIZE); - - /* - * RED-PEN putting page tables only on node 0 could - * cause a hotspot and fill up ZONE_DMA. The page tables - * need roughly 0.5KB per GB. - */ - start = 0x7000; - table_start = find_e820_area(start, max_pfn_mapped<<PAGE_SHIFT, - tables, PAGE_SIZE); - if (table_start == -1UL) - panic("Cannot find space for the kernel page tables"); - - table_start >>= PAGE_SHIFT; - table_end = table_start; - table_top = table_start + (tables>>PAGE_SHIFT); - - printk(KERN_DEBUG "kernel direct mapping tables up to %lx @ %lx-%lx\n", - end, table_start << PAGE_SHIFT, - (table_start << PAGE_SHIFT) + tables); -} - -unsigned long __init_refok init_memory_mapping(unsigned long start, - unsigned long end) -{ - 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; +#ifdef CONFIG_NEED_MULTIPLE_NODES + for_each_online_node(nodeid) + bootmap = setup_node_bootmem(nodeid, node_start_pfn[nodeid], + node_end_pfn[nodeid], bootmap); #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, use_pse); - -#ifdef CONFIG_X86_PAE - set_nx(); - if (nx_enabled) - printk(KERN_INFO "NX (Execute Disable) protection: active\n"); + bootmap = setup_node_bootmem(0, 0, max_low_pfn, bootmap); #endif - /* Enable PSE if available */ - if (cpu_has_pse) - set_in_cr4(X86_CR4_PSE); - - /* Enable PGE if available */ - if (cpu_has_pge) { - set_in_cr4(X86_CR4_PGE); - __supported_pte_mask |= _PAGE_GLOBAL; - } - - /* - * Don't use a large page for the first 2/4MB of memory - * because there are often fixed size MTRRs in there - * and overlapping MTRRs into large pages can cause - * slowdowns. - */ - big_page_start = PMD_SIZE; - - if (start < big_page_start) { - start_pfn = start >> PAGE_SHIFT; - end_pfn = min(big_page_start>>PAGE_SHIFT, end>>PAGE_SHIFT); - } else { - /* head is not big page alignment ? */ - start_pfn = start >> PAGE_SHIFT; - end_pfn = ((start + (PMD_SIZE - 1))>>PMD_SHIFT) - << (PMD_SHIFT - PAGE_SHIFT); - } - if (start_pfn < end_pfn) - kernel_physical_mapping_init(pgd_base, start_pfn, end_pfn, 0); - - /* big page range */ - start_pfn = ((start + (PMD_SIZE - 1))>>PMD_SHIFT) - << (PMD_SHIFT - PAGE_SHIFT); - if (start_pfn < (big_page_start >> PAGE_SHIFT)) - start_pfn = big_page_start >> PAGE_SHIFT; - end_pfn = (end>>PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT); - if (start_pfn < end_pfn) - kernel_physical_mapping_init(pgd_base, start_pfn, end_pfn, - use_pse); - - /* tail is not big page alignment ? */ - start_pfn = end_pfn; - if (start_pfn > (big_page_start>>PAGE_SHIFT)) { - end_pfn = end >> PAGE_SHIFT; - if (start_pfn < end_pfn) - kernel_physical_mapping_init(pgd_base, start_pfn, - end_pfn, 0); - } - - early_ioremap_page_table_range_init(pgd_base); - - load_cr3(swapper_pg_dir); - - __flush_tlb_all(); - - if (!after_init_bootmem) - reserve_early(table_start << PAGE_SHIFT, - table_end << PAGE_SHIFT, "PGTABLE"); - - if (!after_init_bootmem) - early_memtest(start, end); - - return end >> PAGE_SHIFT; + after_bootmem = 1; } - /* * paging_init() sets up the page tables - note that the first 8MB are * already mapped by head.S. @@ -1217,13 +1087,6 @@ void mark_rodata_ro(void) } #endif -#ifdef CONFIG_BLK_DEV_INITRD -void free_initrd_mem(unsigned long start, unsigned long end) -{ - free_init_pages("initrd memory", start, end); -} -#endif - int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, int flags) { diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 07f44d491df1..8a853bc3b287 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -48,6 +48,7 @@ #include <asm/kdebug.h> #include <asm/numa.h> #include <asm/cacheflush.h> +#include <asm/init.h> /* * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. @@ -61,12 +62,6 @@ static unsigned long dma_reserve __initdata; DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); -int direct_gbpages -#ifdef CONFIG_DIRECT_GBPAGES - = 1 -#endif -; - static int __init parse_direct_gbpages_off(char *arg) { direct_gbpages = 0; @@ -87,8 +82,6 @@ early_param("gbpages", parse_direct_gbpages_on); * around without checking the pgd every time. */ -int after_bootmem; - pteval_t __supported_pte_mask __read_mostly = ~_PAGE_IOMAP; EXPORT_SYMBOL_GPL(__supported_pte_mask); @@ -325,13 +318,9 @@ void __init cleanup_highmap(void) } } -static unsigned long __initdata table_start; -static unsigned long __meminitdata table_end; -static unsigned long __meminitdata table_top; - static __ref void *alloc_low_page(unsigned long *phys) { - unsigned long pfn = table_end++; + unsigned long pfn = e820_table_end++; void *adr; if (after_bootmem) { @@ -341,7 +330,7 @@ static __ref void *alloc_low_page(unsigned long *phys) return adr; } - if (pfn >= table_top) + if (pfn >= e820_table_top) panic("alloc_low_page: ran out of memory"); adr = early_memremap(pfn * PAGE_SIZE, PAGE_SIZE); @@ -581,58 +570,10 @@ 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, int use_pse, - int use_gbpages) -{ - unsigned long puds, pmds, ptes, tables, start; - - puds = (end + PUD_SIZE - 1) >> PUD_SHIFT; - 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 += roundup(pmds * sizeof(pmd_t), PAGE_SIZE); - - 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 += roundup(ptes * sizeof(pte_t), PAGE_SIZE); - - /* - * RED-PEN putting page tables only on node 0 could - * cause a hotspot and fill up ZONE_DMA. The page tables - * need roughly 0.5KB per GB. - */ - start = 0x8000; - table_start = find_e820_area(start, end, tables, PAGE_SIZE); - if (table_start == -1UL) - panic("Cannot find space for the kernel page tables"); - - table_start >>= PAGE_SHIFT; - table_end = table_start; - table_top = table_start + (tables >> PAGE_SHIFT); - - printk(KERN_DEBUG "kernel direct mapping tables up to %lx @ %lx-%lx\n", - end, table_start << PAGE_SHIFT, table_top << PAGE_SHIFT); -} - -static void __init init_gbpages(void) -{ - if (direct_gbpages && cpu_has_gbpages) - printk(KERN_INFO "Using GB pages for direct mapping\n"); - else - direct_gbpages = 0; -} - -static unsigned long __meminit kernel_physical_mapping_init(unsigned long start, - unsigned long end, - unsigned long page_size_mask) +unsigned long __init +kernel_physical_mapping_init(unsigned long start, + unsigned long end, + unsigned long page_size_mask) { unsigned long next, last_map_addr = end; @@ -669,176 +610,6 @@ static unsigned long __meminit kernel_physical_mapping_init(unsigned long start, return last_map_addr; } -struct map_range { - unsigned long start; - unsigned long end; - unsigned page_size_mask; -}; - -#define NR_RANGE_MR 5 - -static int save_mr(struct map_range *mr, int nr_range, - unsigned long start_pfn, unsigned long end_pfn, - unsigned long page_size_mask) -{ - - if (start_pfn < end_pfn) { - if (nr_range >= NR_RANGE_MR) - panic("run out of range for init_memory_mapping\n"); - mr[nr_range].start = start_pfn<<PAGE_SHIFT; - mr[nr_range].end = end_pfn<<PAGE_SHIFT; - mr[nr_range].page_size_mask = page_size_mask; - nr_range++; - } - - return nr_range; -} - -/* - * Setup the direct mapping of the physical memory at PAGE_OFFSET. - * This runs before bootmem is initialized and gets pages directly from - * the physical memory. To access them they are temporarily mapped. - */ -unsigned long __init_refok init_memory_mapping(unsigned long start, - unsigned long end) -{ - unsigned long last_map_addr = 0; - unsigned long page_size_mask = 0; - unsigned long start_pfn, end_pfn; - unsigned long pos; - - struct map_range mr[NR_RANGE_MR]; - int nr_range, i; - int use_pse, use_gbpages; - - printk(KERN_INFO "init_memory_mapping: %016lx-%016lx\n", start, end); - - /* - * Find space for the kernel direct mapping tables. - * - * Later we should allocate these tables in the local node of the - * memory mapped. Unfortunately this is done currently before the - * nodes are discovered. - */ - if (!after_bootmem) - init_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 (use_pse) - page_size_mask |= 1 << PG_LEVEL_2M; - - memset(mr, 0, sizeof(mr)); - nr_range = 0; - - /* head if not big page alignment ?*/ - start_pfn = start >> PAGE_SHIFT; - pos = start_pfn << PAGE_SHIFT; - end_pfn = ((pos + (PMD_SIZE - 1)) >> PMD_SHIFT) - << (PMD_SHIFT - PAGE_SHIFT); - if (end_pfn > (end >> PAGE_SHIFT)) - end_pfn = end >> PAGE_SHIFT; - if (start_pfn < end_pfn) { - nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0); - pos = end_pfn << PAGE_SHIFT; - } - - /* big page (2M) range*/ - start_pfn = ((pos + (PMD_SIZE - 1))>>PMD_SHIFT) - << (PMD_SHIFT - PAGE_SHIFT); - end_pfn = ((pos + (PUD_SIZE - 1))>>PUD_SHIFT) - << (PUD_SHIFT - PAGE_SHIFT); - if (end_pfn > ((end>>PMD_SHIFT)<<(PMD_SHIFT - PAGE_SHIFT))) - end_pfn = ((end>>PMD_SHIFT)<<(PMD_SHIFT - PAGE_SHIFT)); - if (start_pfn < end_pfn) { - nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, - page_size_mask & (1<<PG_LEVEL_2M)); - pos = end_pfn << PAGE_SHIFT; - } - - /* big page (1G) range */ - start_pfn = ((pos + (PUD_SIZE - 1))>>PUD_SHIFT) - << (PUD_SHIFT - PAGE_SHIFT); - end_pfn = (end >> PUD_SHIFT) << (PUD_SHIFT - PAGE_SHIFT); - if (start_pfn < end_pfn) { - nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, - page_size_mask & - ((1<<PG_LEVEL_2M)|(1<<PG_LEVEL_1G))); - pos = end_pfn << PAGE_SHIFT; - } - - /* tail is not big page (1G) alignment */ - start_pfn = ((pos + (PMD_SIZE - 1))>>PMD_SHIFT) - << (PMD_SHIFT - PAGE_SHIFT); - end_pfn = (end >> PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT); - if (start_pfn < end_pfn) { - nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, - page_size_mask & (1<<PG_LEVEL_2M)); - pos = end_pfn << PAGE_SHIFT; - } - - /* tail is not big page (2M) alignment */ - start_pfn = pos>>PAGE_SHIFT; - end_pfn = end>>PAGE_SHIFT; - nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0); - - /* try to merge same page size and continuous */ - for (i = 0; nr_range > 1 && i < nr_range - 1; i++) { - unsigned long old_start; - if (mr[i].end != mr[i+1].start || - mr[i].page_size_mask != mr[i+1].page_size_mask) - continue; - /* move it */ - old_start = mr[i].start; - memmove(&mr[i], &mr[i+1], - (nr_range - 1 - i) * sizeof (struct map_range)); - mr[i--].start = old_start; - nr_range--; - } - - for (i = 0; i < nr_range; i++) - printk(KERN_DEBUG " %010lx - %010lx page %s\n", - mr[i].start, mr[i].end, - (mr[i].page_size_mask & (1<<PG_LEVEL_1G))?"1G":( - (mr[i].page_size_mask & (1<<PG_LEVEL_2M))?"2M":"4k")); - - if (!after_bootmem) - find_early_table_space(end, use_pse, use_gbpages); - - for (i = 0; i < nr_range; i++) - last_map_addr = kernel_physical_mapping_init( - mr[i].start, mr[i].end, - mr[i].page_size_mask); - - if (!after_bootmem) - mmu_cr4_features = read_cr4(); - __flush_tlb_all(); - - if (!after_bootmem && table_end > table_start) - reserve_early(table_start << PAGE_SHIFT, - table_end << PAGE_SHIFT, "PGTABLE"); - - printk(KERN_INFO "last_map_addr: %lx end: %lx\n", - last_map_addr, end); - - if (!after_bootmem) - early_memtest(start, end); - - return last_map_addr >> PAGE_SHIFT; -} - #ifndef CONFIG_NUMA void __init initmem_init(unsigned long start_pfn, unsigned long end_pfn) { @@ -910,28 +681,6 @@ EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid); #endif /* CONFIG_MEMORY_HOTPLUG */ -/* - * devmem_is_allowed() checks to see if /dev/mem access to a certain address - * is valid. The argument is a physical page number. - * - * - * On x86, access has to be given to the first megabyte of ram because that area - * contains bios code and data regions used by X and dosemu and similar apps. - * Access has to be given to non-kernel-ram areas as well, these contain the PCI - * mmio resources as well as potential bios/acpi data regions. - */ -int devmem_is_allowed(unsigned long pagenr) -{ - if (pagenr <= 256) - return 1; - if (iomem_is_exclusive(pagenr << PAGE_SHIFT)) - return 0; - if (!page_is_ram(pagenr)) - return 1; - return 0; -} - - static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel, kcore_modules, kcore_vsyscall; @@ -1019,13 +768,6 @@ void mark_rodata_ro(void) #endif -#ifdef CONFIG_BLK_DEV_INITRD -void free_initrd_mem(unsigned long start, unsigned long end) -{ - free_init_pages("initrd memory", start, end); -} -#endif - int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, int flags) { diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 433f7bd4648a..62773abdf088 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -38,8 +38,7 @@ unsigned long __phys_addr(unsigned long x) } else { VIRTUAL_BUG_ON(x < PAGE_OFFSET); x -= PAGE_OFFSET; - VIRTUAL_BUG_ON(system_state == SYSTEM_BOOTING ? x > MAXMEM : - !phys_addr_valid(x)); + VIRTUAL_BUG_ON(!phys_addr_valid(x)); } return x; } @@ -56,10 +55,8 @@ bool __virt_addr_valid(unsigned long x) if (x < PAGE_OFFSET) return false; x -= PAGE_OFFSET; - if (system_state == SYSTEM_BOOTING ? - x > MAXMEM : !phys_addr_valid(x)) { + if (!phys_addr_valid(x)) return false; - } } return pfn_valid(x >> PAGE_SHIFT); @@ -76,10 +73,9 @@ static inline int phys_addr_valid(unsigned long addr) #ifdef CONFIG_DEBUG_VIRTUAL unsigned long __phys_addr(unsigned long x) { - /* VMALLOC_* aren't constants; not available at the boot time */ + /* VMALLOC_* aren't constants */ VIRTUAL_BUG_ON(x < PAGE_OFFSET); - VIRTUAL_BUG_ON(system_state != SYSTEM_BOOTING && - is_vmalloc_addr((void *) x)); + VIRTUAL_BUG_ON(__vmalloc_start_set && is_vmalloc_addr((void *) x)); return x - PAGE_OFFSET; } EXPORT_SYMBOL(__phys_addr); @@ -89,7 +85,7 @@ bool __virt_addr_valid(unsigned long x) { if (x < PAGE_OFFSET) return false; - if (system_state != SYSTEM_BOOTING && is_vmalloc_addr((void *) x)) + if (__vmalloc_start_set && is_vmalloc_addr((void *) x)) return false; return pfn_valid((x - PAGE_OFFSET) >> PAGE_SHIFT); } diff --git a/arch/x86/mm/kmmio.c b/arch/x86/mm/kmmio.c index 93d82038af4b..9f205030d9aa 100644 --- a/arch/x86/mm/kmmio.c +++ b/arch/x86/mm/kmmio.c @@ -32,11 +32,14 @@ struct kmmio_fault_page { struct list_head list; struct kmmio_fault_page *release_next; unsigned long page; /* location of the fault page */ + bool old_presence; /* page presence prior to arming */ + bool armed; /* * Number of times this page has been registered as a part * of a probe. If zero, page is disarmed and this may be freed. - * Used only by writers (RCU). + * Used only by writers (RCU) and post_kmmio_handler(). + * Protected by kmmio_lock, when linked into kmmio_page_table. */ int count; }; @@ -105,57 +108,85 @@ static struct kmmio_fault_page *get_kmmio_fault_page(unsigned long page) return NULL; } -static void set_page_present(unsigned long addr, bool present, - unsigned int *pglevel) +static void set_pmd_presence(pmd_t *pmd, bool present, bool *old) +{ + pmdval_t v = pmd_val(*pmd); + *old = !!(v & _PAGE_PRESENT); + v &= ~_PAGE_PRESENT; + if (present) + v |= _PAGE_PRESENT; + set_pmd(pmd, __pmd(v)); +} + +static void set_pte_presence(pte_t *pte, bool present, bool *old) +{ + pteval_t v = pte_val(*pte); + *old = !!(v & _PAGE_PRESENT); + v &= ~_PAGE_PRESENT; + if (present) + v |= _PAGE_PRESENT; + set_pte_atomic(pte, __pte(v)); +} + +static int set_page_presence(unsigned long addr, bool present, bool *old) { - pteval_t pteval; - pmdval_t pmdval; unsigned int level; - pmd_t *pmd; pte_t *pte = lookup_address(addr, &level); if (!pte) { pr_err("kmmio: no pte for page 0x%08lx\n", addr); - return; + return -1; } - if (pglevel) - *pglevel = level; - switch (level) { case PG_LEVEL_2M: - pmd = (pmd_t *)pte; - pmdval = pmd_val(*pmd) & ~_PAGE_PRESENT; - if (present) - pmdval |= _PAGE_PRESENT; - set_pmd(pmd, __pmd(pmdval)); + set_pmd_presence((pmd_t *)pte, present, old); break; - case PG_LEVEL_4K: - pteval = pte_val(*pte) & ~_PAGE_PRESENT; - if (present) - pteval |= _PAGE_PRESENT; - set_pte_atomic(pte, __pte(pteval)); + set_pte_presence(pte, present, old); break; - default: pr_err("kmmio: unexpected page level 0x%x.\n", level); - return; + return -1; } __flush_tlb_one(addr); + return 0; } -/** Mark the given page as not present. Access to it will trigger a fault. */ -static void arm_kmmio_fault_page(unsigned long page, unsigned int *pglevel) +/* + * Mark the given page as not present. Access to it will trigger a fault. + * + * Struct kmmio_fault_page is protected by RCU and kmmio_lock, but the + * protection is ignored here. RCU read lock is assumed held, so the struct + * will not disappear unexpectedly. Furthermore, the caller must guarantee, + * that double arming the same virtual address (page) cannot occur. + * + * Double disarming on the other hand is allowed, and may occur when a fault + * and mmiotrace shutdown happen simultaneously. + */ +static int arm_kmmio_fault_page(struct kmmio_fault_page *f) { - set_page_present(page & PAGE_MASK, false, pglevel); + int ret; + WARN_ONCE(f->armed, KERN_ERR "kmmio page already armed.\n"); + if (f->armed) { + pr_warning("kmmio double-arm: page 0x%08lx, ref %d, old %d\n", + f->page, f->count, f->old_presence); + } + ret = set_page_presence(f->page, false, &f->old_presence); + WARN_ONCE(ret < 0, KERN_ERR "kmmio arming 0x%08lx failed.\n", f->page); + f->armed = true; + return ret; } -/** Mark the given page as present. */ -static void disarm_kmmio_fault_page(unsigned long page, unsigned int *pglevel) +/** Restore the given page to saved presence state. */ +static void disarm_kmmio_fault_page(struct kmmio_fault_page *f) { - set_page_present(page & PAGE_MASK, true, pglevel); + bool tmp; + int ret = set_page_presence(f->page, f->old_presence, &tmp); + WARN_ONCE(ret < 0, + KERN_ERR "kmmio disarming 0x%08lx failed.\n", f->page); + f->armed = false; } /* @@ -202,28 +233,32 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr) ctx = &get_cpu_var(kmmio_ctx); if (ctx->active) { - disarm_kmmio_fault_page(faultpage->page, NULL); if (addr == ctx->addr) { /* - * On SMP we sometimes get recursive probe hits on the - * same address. Context is already saved, fall out. + * A second fault on the same page means some other + * condition needs handling by do_page_fault(), the + * page really not being present is the most common. */ - pr_debug("kmmio: duplicate probe hit on CPU %d, for " - "address 0x%08lx.\n", - smp_processor_id(), addr); - ret = 1; - goto no_kmmio_ctx; - } - /* - * Prevent overwriting already in-flight context. - * This should not happen, let's hope disarming at least - * prevents a panic. - */ - pr_emerg("kmmio: recursive probe hit on CPU %d, " + pr_debug("kmmio: secondary hit for 0x%08lx CPU %d.\n", + addr, smp_processor_id()); + + if (!faultpage->old_presence) + pr_info("kmmio: unexpected secondary hit for " + "address 0x%08lx on CPU %d.\n", addr, + smp_processor_id()); + } else { + /* + * Prevent overwriting already in-flight context. + * This should not happen, let's hope disarming at + * least prevents a panic. + */ + pr_emerg("kmmio: recursive probe hit on CPU %d, " "for address 0x%08lx. Ignoring.\n", smp_processor_id(), addr); - pr_emerg("kmmio: previous hit was at 0x%08lx.\n", - ctx->addr); + pr_emerg("kmmio: previous hit was at 0x%08lx.\n", + ctx->addr); + disarm_kmmio_fault_page(faultpage); + } goto no_kmmio_ctx; } ctx->active++; @@ -244,7 +279,7 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr) regs->flags &= ~X86_EFLAGS_IF; /* Now we set present bit in PTE and single step. */ - disarm_kmmio_fault_page(ctx->fpage->page, NULL); + disarm_kmmio_fault_page(ctx->fpage); /* * If another cpu accesses the same page while we are stepping, @@ -275,7 +310,7 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs) struct kmmio_context *ctx = &get_cpu_var(kmmio_ctx); if (!ctx->active) { - pr_debug("kmmio: spurious debug trap on CPU %d.\n", + pr_warning("kmmio: spurious debug trap on CPU %d.\n", smp_processor_id()); goto out; } @@ -283,7 +318,11 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs) if (ctx->probe && ctx->probe->post_handler) ctx->probe->post_handler(ctx->probe, condition, regs); - arm_kmmio_fault_page(ctx->fpage->page, NULL); + /* Prevent racing against release_kmmio_fault_page(). */ + spin_lock(&kmmio_lock); + if (ctx->fpage->count) + arm_kmmio_fault_page(ctx->fpage); + spin_unlock(&kmmio_lock); regs->flags &= ~X86_EFLAGS_TF; regs->flags |= ctx->saved_flags; @@ -315,20 +354,24 @@ static int add_kmmio_fault_page(unsigned long page) f = get_kmmio_fault_page(page); if (f) { if (!f->count) - arm_kmmio_fault_page(f->page, NULL); + arm_kmmio_fault_page(f); f->count++; return 0; } - f = kmalloc(sizeof(*f), GFP_ATOMIC); + f = kzalloc(sizeof(*f), GFP_ATOMIC); if (!f) return -1; f->count = 1; f->page = page; - list_add_rcu(&f->list, kmmio_page_list(f->page)); - arm_kmmio_fault_page(f->page, NULL); + if (arm_kmmio_fault_page(f)) { + kfree(f); + return -1; + } + + list_add_rcu(&f->list, kmmio_page_list(f->page)); return 0; } @@ -347,7 +390,7 @@ static void release_kmmio_fault_page(unsigned long page, f->count--; BUG_ON(f->count < 0); if (!f->count) { - disarm_kmmio_fault_page(f->page, NULL); + disarm_kmmio_fault_page(f); f->release_next = *release_list; *release_list = f; } diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index 451fe95a0352..3daefa04ace5 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -416,10 +416,11 @@ void __init initmem_init(unsigned long start_pfn, for_each_online_node(nid) propagate_e820_map_node(nid); - for_each_online_node(nid) + for_each_online_node(nid) { memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); + NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; + } - NODE_DATA(0)->bdata = &bootmem_node_data[0]; setup_bootmem_allocator(); } diff --git a/arch/x86/mm/testmmiotrace.c b/arch/x86/mm/testmmiotrace.c index ab50a8d7402c..427fd1b56df5 100644 --- a/arch/x86/mm/testmmiotrace.c +++ b/arch/x86/mm/testmmiotrace.c @@ -1,5 +1,5 @@ /* - * Written by Pekka Paalanen, 2008 <pq@iki.fi> + * Written by Pekka Paalanen, 2008-2009 <pq@iki.fi> */ #include <linux/module.h> #include <linux/io.h> @@ -9,35 +9,74 @@ static unsigned long mmio_address; module_param(mmio_address, ulong, 0); -MODULE_PARM_DESC(mmio_address, "Start address of the mapping of 16 kB."); +MODULE_PARM_DESC(mmio_address, " Start address of the mapping of 16 kB " + "(or 8 MB if read_far is non-zero)."); + +static unsigned long read_far = 0x400100; +module_param(read_far, ulong, 0); +MODULE_PARM_DESC(read_far, " Offset of a 32-bit read within 8 MB " + "(default: 0x400100)."); + +static unsigned v16(unsigned i) +{ + return i * 12 + 7; +} + +static unsigned v32(unsigned i) +{ + return i * 212371 + 13; +} static void do_write_test(void __iomem *p) { unsigned int i; + pr_info(MODULE_NAME ": write test.\n"); mmiotrace_printk("Write test.\n"); + for (i = 0; i < 256; i++) iowrite8(i, p + i); + for (i = 1024; i < (5 * 1024); i += 2) - iowrite16(i * 12 + 7, p + i); + iowrite16(v16(i), p + i); + for (i = (5 * 1024); i < (16 * 1024); i += 4) - iowrite32(i * 212371 + 13, p + i); + iowrite32(v32(i), p + i); } static void do_read_test(void __iomem *p) { unsigned int i; + unsigned errs[3] = { 0 }; + pr_info(MODULE_NAME ": read test.\n"); mmiotrace_printk("Read test.\n"); + for (i = 0; i < 256; i++) - ioread8(p + i); + if (ioread8(p + i) != i) + ++errs[0]; + for (i = 1024; i < (5 * 1024); i += 2) - ioread16(p + i); + if (ioread16(p + i) != v16(i)) + ++errs[1]; + for (i = (5 * 1024); i < (16 * 1024); i += 4) - ioread32(p + i); + if (ioread32(p + i) != v32(i)) + ++errs[2]; + + mmiotrace_printk("Read errors: 8-bit %d, 16-bit %d, 32-bit %d.\n", + errs[0], errs[1], errs[2]); } -static void do_test(void) +static void do_read_far_test(void __iomem *p) { - void __iomem *p = ioremap_nocache(mmio_address, 0x4000); + pr_info(MODULE_NAME ": read far test.\n"); + mmiotrace_printk("Read far test.\n"); + + ioread32(p + read_far); +} + +static void do_test(unsigned long size) +{ + void __iomem *p = ioremap_nocache(mmio_address, size); if (!p) { pr_err(MODULE_NAME ": could not ioremap, aborting.\n"); return; @@ -45,11 +84,15 @@ static void do_test(void) mmiotrace_printk("ioremap returned %p.\n", p); do_write_test(p); do_read_test(p); + if (read_far && read_far < size - 4) + do_read_far_test(p); iounmap(p); } static int __init init(void) { + unsigned long size = (read_far) ? (8 << 20) : (16 << 10); + if (mmio_address == 0) { pr_err(MODULE_NAME ": you have to use the module argument " "mmio_address.\n"); @@ -58,10 +101,11 @@ static int __init init(void) return -ENXIO; } - pr_warning(MODULE_NAME ": WARNING: mapping 16 kB @ 0x%08lx " - "in PCI address space, and writing " - "rubbish in there.\n", mmio_address); - do_test(); + pr_warning(MODULE_NAME ": WARNING: mapping %lu kB @ 0x%08lx in PCI " + "address space, and writing 16 kB of rubbish in there.\n", + size >> 10, mmio_address); + do_test(size); + pr_info(MODULE_NAME ": All done.\n"); return 0; } diff --git a/crypto/api.c b/crypto/api.c index efe77df6863f..38a2bc02a98c 100644 --- a/crypto/api.c +++ b/crypto/api.c @@ -215,8 +215,19 @@ struct crypto_alg *crypto_larval_lookup(const char *name, u32 type, u32 mask) mask &= ~(CRYPTO_ALG_LARVAL | CRYPTO_ALG_DEAD); type &= mask; - alg = try_then_request_module(crypto_alg_lookup(name, type, mask), - name); + alg = crypto_alg_lookup(name, type, mask); + if (!alg) { + char tmp[CRYPTO_MAX_ALG_NAME]; + + request_module(name); + + if (!((type ^ CRYPTO_ALG_NEED_FALLBACK) & mask) && + snprintf(tmp, sizeof(tmp), "%s-all", name) < sizeof(tmp)) + request_module(tmp); + + alg = crypto_alg_lookup(name, type, mask); + } + if (alg) return crypto_is_larval(alg) ? crypto_larval_wait(alg) : alg; diff --git a/drivers/crypto/ixp4xx_crypto.c b/drivers/crypto/ixp4xx_crypto.c index 2d637e0fbc03..d9e751be8c5f 100644 --- a/drivers/crypto/ixp4xx_crypto.c +++ b/drivers/crypto/ixp4xx_crypto.c @@ -457,10 +457,12 @@ static int init_ixp_crypto(void) if (!ctx_pool) { goto err; } - ret = qmgr_request_queue(SEND_QID, NPE_QLEN_TOTAL, 0, 0); + ret = qmgr_request_queue(SEND_QID, NPE_QLEN_TOTAL, 0, 0, + "ixp_crypto:out", NULL); if (ret) goto err; - ret = qmgr_request_queue(RECV_QID, NPE_QLEN, 0, 0); + ret = qmgr_request_queue(RECV_QID, NPE_QLEN, 0, 0, + "ixp_crypto:in", NULL); if (ret) { qmgr_release_queue(SEND_QID); goto err; diff --git a/drivers/crypto/padlock-aes.c b/drivers/crypto/padlock-aes.c index 856b3cc25583..3f0fdd18255d 100644 --- a/drivers/crypto/padlock-aes.c +++ b/drivers/crypto/padlock-aes.c @@ -489,4 +489,4 @@ MODULE_DESCRIPTION("VIA PadLock AES algorithm support"); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Michal Ludvig"); -MODULE_ALIAS("aes"); +MODULE_ALIAS("aes-all"); diff --git a/drivers/crypto/padlock-sha.c b/drivers/crypto/padlock-sha.c index a7fbadebf623..a2c8e8514b63 100644 --- a/drivers/crypto/padlock-sha.c +++ b/drivers/crypto/padlock-sha.c @@ -304,7 +304,7 @@ MODULE_DESCRIPTION("VIA PadLock SHA1/SHA256 algorithms support."); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Michal Ludvig"); -MODULE_ALIAS("sha1"); -MODULE_ALIAS("sha256"); +MODULE_ALIAS("sha1-all"); +MODULE_ALIAS("sha256-all"); MODULE_ALIAS("sha1-padlock"); MODULE_ALIAS("sha256-padlock"); diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c index ea5440dd10dc..647374acba94 100644 --- a/drivers/dma/iop-adma.c +++ b/drivers/dma/iop-adma.c @@ -1401,7 +1401,7 @@ MODULE_ALIAS("platform:iop-adma"); static struct platform_driver iop_adma_driver = { .probe = iop_adma_probe, - .remove = iop_adma_remove, + .remove = __devexit_p(iop_adma_remove), .driver = { .owner = THIS_MODULE, .name = "iop-adma", diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c index d35cbd1ff0b3..5d5d5b31867f 100644 --- a/drivers/dma/mv_xor.c +++ b/drivers/dma/mv_xor.c @@ -1287,7 +1287,7 @@ mv_xor_conf_mbus_windows(struct mv_xor_shared_private *msp, static struct platform_driver mv_xor_driver = { .probe = mv_xor_probe, - .remove = mv_xor_remove, + .remove = __devexit_p(mv_xor_remove), .driver = { .owner = THIS_MODULE, .name = MV_XOR_NAME, diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c index 096e2a37446d..7c8b15b22bf2 100644 --- a/drivers/gpu/drm/drm_stub.c +++ b/drivers/gpu/drm/drm_stub.c @@ -168,7 +168,7 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data, file_priv->minor->master != file_priv->master) { mutex_lock(&dev->struct_mutex); file_priv->minor->master = drm_master_get(file_priv->master); - mutex_lock(&dev->struct_mutex); + mutex_unlock(&dev->struct_mutex); } return 0; diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c index eeda276f8f16..7f186bbcb99d 100644 --- a/drivers/i2c/busses/i2c-mv64xxx.c +++ b/drivers/i2c/busses/i2c-mv64xxx.c @@ -482,7 +482,7 @@ mv64xxx_i2c_map_regs(struct platform_device *pd, return 0; } -static void __devexit +static void mv64xxx_i2c_unmap_regs(struct mv64xxx_i2c_data *drv_data) { if (drv_data->reg_base) { @@ -577,7 +577,7 @@ mv64xxx_i2c_remove(struct platform_device *dev) static struct platform_driver mv64xxx_i2c_driver = { .probe = mv64xxx_i2c_probe, - .remove = mv64xxx_i2c_remove, + .remove = __devexit_p(mv64xxx_i2c_remove), .driver = { .owner = THIS_MODULE, .name = MV64XXX_I2C_CTLR_NAME, diff --git a/drivers/mtd/nand/orion_nand.c b/drivers/mtd/nand/orion_nand.c index 917cf8d3ae95..c2dfd3ea353d 100644 --- a/drivers/mtd/nand/orion_nand.c +++ b/drivers/mtd/nand/orion_nand.c @@ -149,7 +149,7 @@ static int __devexit orion_nand_remove(struct platform_device *pdev) static struct platform_driver orion_nand_driver = { .probe = orion_nand_probe, - .remove = orion_nand_remove, + .remove = __devexit_p(orion_nand_remove), .driver = { .name = "orion_nand", .owner = THIS_MODULE, diff --git a/drivers/net/arm/Makefile b/drivers/net/arm/Makefile index c69c0cdba4a2..811a3ccd14c1 100644 --- a/drivers/net/arm/Makefile +++ b/drivers/net/arm/Makefile @@ -4,7 +4,7 @@ # obj-$(CONFIG_ARM_AM79C961A) += am79c961a.o -obj-$(CONFIG_ARM_ETHERH) += etherh.o ../8390.o +obj-$(CONFIG_ARM_ETHERH) += etherh.o obj-$(CONFIG_ARM_ETHER3) += ether3.o obj-$(CONFIG_ARM_ETHER1) += ether1.o obj-$(CONFIG_ARM_AT91_ETHER) += at91_ether.o diff --git a/drivers/net/arm/etherh.c b/drivers/net/arm/etherh.c index 54b52e5b1821..f52f668c49bf 100644 --- a/drivers/net/arm/etherh.c +++ b/drivers/net/arm/etherh.c @@ -641,15 +641,15 @@ static const struct net_device_ops etherh_netdev_ops = { .ndo_open = etherh_open, .ndo_stop = etherh_close, .ndo_set_config = etherh_set_config, - .ndo_start_xmit = ei_start_xmit, - .ndo_tx_timeout = ei_tx_timeout, - .ndo_get_stats = ei_get_stats, - .ndo_set_multicast_list = ei_set_multicast_list, + .ndo_start_xmit = __ei_start_xmit, + .ndo_tx_timeout = __ei_tx_timeout, + .ndo_get_stats = __ei_get_stats, + .ndo_set_multicast_list = __ei_set_multicast_list, .ndo_validate_addr = eth_validate_addr, .ndo_set_mac_address = eth_mac_addr, .ndo_change_mtu = eth_change_mtu, #ifdef CONFIG_NET_POLL_CONTROLLER - .ndo_poll_controller = ei_poll, + .ndo_poll_controller = __ei_poll, #endif }; diff --git a/drivers/video/pxafb.c b/drivers/video/pxafb.c index 48ff701d3a72..2552b9f325ee 100644 --- a/drivers/video/pxafb.c +++ b/drivers/video/pxafb.c @@ -2230,7 +2230,7 @@ static int __devexit pxafb_remove(struct platform_device *dev) static struct platform_driver pxafb_driver = { .probe = pxafb_probe, - .remove = pxafb_remove, + .remove = __devexit_p(pxafb_remove), .suspend = pxafb_suspend, .resume = pxafb_resume, .driver = { diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h index f3f697df1d71..80044a4f3ab9 100644 --- a/include/linux/rcuclassic.h +++ b/include/linux/rcuclassic.h @@ -181,4 +181,10 @@ extern long rcu_batches_completed_bh(void); #define rcu_enter_nohz() do { } while (0) #define rcu_exit_nohz() do { } while (0) +/* A context switch is a grace period for rcuclassic. */ +static inline int rcu_blocking_is_gp(void) +{ + return num_online_cpus() == 1; +} + #endif /* __LINUX_RCUCLASSIC_H */ diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 921340a7b71c..528343e6da51 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -52,6 +52,9 @@ struct rcu_head { void (*func)(struct rcu_head *head); }; +/* Internal to kernel, but needed by rcupreempt.h. */ +extern int rcu_scheduler_active; + #if defined(CONFIG_CLASSIC_RCU) #include <linux/rcuclassic.h> #elif defined(CONFIG_TREE_RCU) @@ -265,6 +268,7 @@ extern void rcu_barrier_sched(void); /* Internal to kernel */ extern void rcu_init(void); +extern void rcu_scheduler_starting(void); extern int rcu_needs_cpu(int cpu); #endif /* __LINUX_RCUPDATE_H */ diff --git a/include/linux/rcupreempt.h b/include/linux/rcupreempt.h index 3e05c09b54a2..74304b4538d8 100644 --- a/include/linux/rcupreempt.h +++ b/include/linux/rcupreempt.h @@ -142,4 +142,19 @@ static inline void rcu_exit_nohz(void) #define rcu_exit_nohz() do { } while (0) #endif /* CONFIG_NO_HZ */ +/* + * A context switch is a grace period for rcupreempt synchronize_rcu() + * only during early boot, before the scheduler has been initialized. + * So, how the heck do we get a context switch? Well, if the caller + * invokes synchronize_rcu(), they are willing to accept a context + * switch, so we simply pretend that one happened. + * + * After boot, there might be a blocked or preempted task in an RCU + * read-side critical section, so we cannot then take the fastpath. + */ +static inline int rcu_blocking_is_gp(void) +{ + return num_online_cpus() == 1 && !rcu_scheduler_active; +} + #endif /* __LINUX_RCUPREEMPT_H */ diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h index d4368b7975c3..a722fb67bb2d 100644 --- a/include/linux/rcutree.h +++ b/include/linux/rcutree.h @@ -326,4 +326,10 @@ static inline void rcu_exit_nohz(void) } #endif /* CONFIG_NO_HZ */ +/* A context switch is a grace period for rcutree. */ +static inline int rcu_blocking_is_gp(void) +{ + return num_online_cpus() == 1; +} + #endif /* __LINUX_RCUTREE_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index f0a50b20e8a0..a7c7698583bb 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2303,9 +2303,13 @@ extern long sched_group_rt_runtime(struct task_group *tg); extern int sched_group_set_rt_period(struct task_group *tg, long rt_period_us); extern long sched_group_rt_period(struct task_group *tg); +extern int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk); #endif #endif +extern int task_can_switch_user(struct user_struct *up, + struct task_struct *tsk); + #ifdef CONFIG_TASK_XACCT static inline void add_rchar(struct task_struct *tsk, ssize_t amt) { diff --git a/init/main.c b/init/main.c index 6441083f8273..6bf83afd654d 100644 --- a/init/main.c +++ b/init/main.c @@ -98,7 +98,7 @@ static inline void mark_rodata_ro(void) { } extern void tc_init(void); #endif -enum system_states system_state; +enum system_states system_state __read_mostly; EXPORT_SYMBOL(system_state); /* @@ -464,6 +464,7 @@ static noinline void __init_refok rest_init(void) * at least once to get things moving: */ init_idle_bootup_task(current); + rcu_scheduler_starting(); preempt_enable_no_resched(); schedule(); preempt_disable(); diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c index bd5a9003497c..654c640a6b9c 100644 --- a/kernel/rcuclassic.c +++ b/kernel/rcuclassic.c @@ -679,8 +679,8 @@ int rcu_needs_cpu(int cpu) void rcu_check_callbacks(int cpu, int user) { if (user || - (idle_cpu(cpu) && !in_softirq() && - hardirq_count() <= (1 << HARDIRQ_SHIFT))) { + (idle_cpu(cpu) && rcu_scheduler_active && + !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) { /* * Get here if this CPU took its interrupt from user diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index d92a76a881aa..cae8a059cf47 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -44,6 +44,7 @@ #include <linux/cpu.h> #include <linux/mutex.h> #include <linux/module.h> +#include <linux/kernel_stat.h> enum rcu_barrier { RCU_BARRIER_STD, @@ -55,6 +56,7 @@ static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL}; static atomic_t rcu_barrier_cpu_count; static DEFINE_MUTEX(rcu_barrier_mutex); static struct completion rcu_barrier_completion; +int rcu_scheduler_active __read_mostly; /* * Awaken the corresponding synchronize_rcu() instance now that a @@ -80,6 +82,10 @@ void wakeme_after_rcu(struct rcu_head *head) void synchronize_rcu(void) { struct rcu_synchronize rcu; + + if (rcu_blocking_is_gp()) + return; + init_completion(&rcu.completion); /* Will wake me after RCU finished. */ call_rcu(&rcu.head, wakeme_after_rcu); @@ -175,3 +181,9 @@ void __init rcu_init(void) __rcu_init(); } +void rcu_scheduler_starting(void) +{ + WARN_ON(num_online_cpus() != 1); + WARN_ON(nr_context_switches() > 0); + rcu_scheduler_active = 1; +} diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c index 33cfc50781f9..5d59e850fb71 100644 --- a/kernel/rcupreempt.c +++ b/kernel/rcupreempt.c @@ -1181,6 +1181,9 @@ void __synchronize_sched(void) { struct rcu_synchronize rcu; + if (num_online_cpus() == 1) + return; /* blocking is gp if only one CPU! */ + init_completion(&rcu.completion); /* Will wake me after RCU finished. */ call_rcu_sched(&rcu.head, wakeme_after_rcu); diff --git a/kernel/rcutree.c b/kernel/rcutree.c index b2fd602a6f6f..97ce31579ec0 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -948,8 +948,8 @@ static void rcu_do_batch(struct rcu_data *rdp) void rcu_check_callbacks(int cpu, int user) { if (user || - (idle_cpu(cpu) && !in_softirq() && - hardirq_count() <= (1 << HARDIRQ_SHIFT))) { + (idle_cpu(cpu) && rcu_scheduler_active && + !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) { /* * Get here if this CPU took its interrupt from user diff --git a/kernel/sched.c b/kernel/sched.c index 0e5c38e1c8b5..0a76d0b6f215 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -223,7 +223,7 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) { ktime_t now; - if (rt_bandwidth_enabled() && rt_b->rt_runtime == RUNTIME_INF) + if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF) return; if (hrtimer_active(&rt_b->rt_period_timer)) @@ -9219,6 +9219,16 @@ static int sched_rt_global_constraints(void) return ret; } + +int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk) +{ + /* Don't accept realtime tasks when there is no way for them to run */ + if (rt_task(tsk) && tg->rt_bandwidth.rt_runtime == 0) + return 0; + + return 1; +} + #else /* !CONFIG_RT_GROUP_SCHED */ static int sched_rt_global_constraints(void) { @@ -9312,8 +9322,7 @@ cpu_cgroup_can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, struct task_struct *tsk) { #ifdef CONFIG_RT_GROUP_SCHED - /* Don't accept realtime tasks when there is no way for them to run */ - if (rt_task(tsk) && cgroup_tg(cgrp)->rt_bandwidth.rt_runtime == 0) + if (!sched_rt_can_attach(cgroup_tg(cgrp), tsk)) return -EINVAL; #else /* We don't support RT-tasks being in separate groups */ diff --git a/kernel/sys.c b/kernel/sys.c index f145c415bc16..37f458e6882a 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -559,7 +559,7 @@ error: abort_creds(new); return retval; } - + /* * change the user struct in a credentials set to match the new UID */ @@ -571,6 +571,11 @@ static int set_user(struct cred *new) if (!new_user) return -EAGAIN; + if (!task_can_switch_user(new_user, current)) { + free_uid(new_user); + return -EINVAL; + } + if (atomic_read(&new_user->processes) >= current->signal->rlim[RLIMIT_NPROC].rlim_cur && new_user != INIT_USER) { @@ -631,10 +636,11 @@ SYSCALL_DEFINE2(setreuid, uid_t, ruid, uid_t, euid) goto error; } - retval = -EAGAIN; - if (new->uid != old->uid && set_user(new) < 0) - goto error; - + if (new->uid != old->uid) { + retval = set_user(new); + if (retval < 0) + goto error; + } if (ruid != (uid_t) -1 || (euid != (uid_t) -1 && euid != old->uid)) new->suid = new->euid; @@ -680,9 +686,10 @@ SYSCALL_DEFINE1(setuid, uid_t, uid) retval = -EPERM; if (capable(CAP_SETUID)) { new->suid = new->uid = uid; - if (uid != old->uid && set_user(new) < 0) { - retval = -EAGAIN; - goto error; + if (uid != old->uid) { + retval = set_user(new); + if (retval < 0) + goto error; } } else if (uid != old->uid && uid != new->suid) { goto error; @@ -734,11 +741,13 @@ SYSCALL_DEFINE3(setresuid, uid_t, ruid, uid_t, euid, uid_t, suid) goto error; } - retval = -EAGAIN; if (ruid != (uid_t) -1) { new->uid = ruid; - if (ruid != old->uid && set_user(new) < 0) - goto error; + if (ruid != old->uid) { + retval = set_user(new); + if (retval < 0) + goto error; + } } if (euid != (uid_t) -1) new->euid = euid; diff --git a/kernel/user.c b/kernel/user.c index 3551ac742395..6a9b696128c8 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -362,6 +362,24 @@ static void free_user(struct user_struct *up, unsigned long flags) #endif +#if defined(CONFIG_RT_GROUP_SCHED) && defined(CONFIG_USER_SCHED) +/* + * We need to check if a setuid can take place. This function should be called + * before successfully completing the setuid. + */ +int task_can_switch_user(struct user_struct *up, struct task_struct *tsk) +{ + + return sched_rt_can_attach(up->tg, tsk); + +} +#else +int task_can_switch_user(struct user_struct *up, struct task_struct *tsk) +{ + return 1; +} +#endif + /* * Locate the user_struct for the passed UID. If found, take a ref on it. The * caller must undo that ref with free_uid(). |