From 93fa7636dfdc059b25df148f230c0991096afdef Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Tue, 8 Apr 2008 11:01:58 +0200 Subject: x86, ptrace: PEBS support Polish the ds.h interface and add support for PEBS. Ds.c is meant to be the resource allocator for per-thread and per-cpu BTS and PEBS recording. It is used by ptrace/utrace to provide execution tracing of debugged tasks. It will be used by profilers (e.g. perfmon2). It may be used by kernel debuggers to provide a kernel execution trace. Changes in detail: - guard DS and ptrace by CONFIG macros - separate DS and BTS more clearly - simplify field accesses - add functions to manage PEBS buffers - add simple protection/allocation mechanism - added support for Atom Opens: - buffer overflow handling Currently, only circular buffers are supported. This is all we need for debugging. Profilers would want an overflow notification. This is planned to be added when perfmon2 is made to use the ds.h interface. - utrace intermediate layer Signed-off-by: Markus Metzger Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- include/asm-x86/ds.h | 258 +++++++++++++++++++++++++++++++++++-------- include/asm-x86/processor.h | 12 +- include/asm-x86/ptrace-abi.h | 14 ++- include/asm-x86/ptrace.h | 38 ++++++- 4 files changed, 265 insertions(+), 57 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/ds.h b/include/asm-x86/ds.h index 7881368142f..72c5a190bf4 100644 --- a/include/asm-x86/ds.h +++ b/include/asm-x86/ds.h @@ -2,71 +2,237 @@ * Debug Store (DS) support * * This provides a low-level interface to the hardware's Debug Store - * feature that is used for last branch recording (LBR) and + * feature that is used for branch trace store (BTS) and * precise-event based sampling (PEBS). * - * Different architectures use a different DS layout/pointer size. - * The below functions therefore work on a void*. + * It manages: + * - per-thread and per-cpu allocation of BTS and PEBS + * - buffer memory allocation (optional) + * - buffer overflow handling + * - buffer access * + * It assumes: + * - get_task_struct on all parameter tasks + * - current is allowed to trace parameter tasks * - * Since there is no user for PEBS, yet, only LBR (or branch - * trace store, BTS) is supported. * - * - * Copyright (C) 2007 Intel Corporation. - * Markus Metzger , Dec 2007 + * Copyright (C) 2007-2008 Intel Corporation. + * Markus Metzger , 2007-2008 */ #ifndef _ASM_X86_DS_H #define _ASM_X86_DS_H +#ifdef CONFIG_X86_DS + #include #include -struct cpuinfo_x86; +struct task_struct; -/* a branch trace record entry +/* + * Request BTS or PEBS + * + * Due to alignement constraints, the actual buffer may be slightly + * smaller than the requested or provided buffer. * - * In order to unify the interface between various processor versions, - * we use the below data structure for all processors. + * Returns 0 on success; -Eerrno otherwise + * + * task: the task to request recording for; + * NULL for per-cpu recording on the current cpu + * base: the base pointer for the (non-pageable) buffer; + * NULL if buffer allocation requested + * size: the size of the requested or provided buffer + * ovfl: pointer to a function to be called on buffer overflow; + * NULL if cyclic buffer requested */ -enum bts_qualifier { - BTS_INVALID = 0, - BTS_BRANCH, - BTS_TASK_ARRIVES, - BTS_TASK_DEPARTS -}; +typedef void (*ds_ovfl_callback_t)(struct task_struct *); +extern int ds_request_bts(struct task_struct *task, void *base, size_t size, + ds_ovfl_callback_t ovfl); +extern int ds_request_pebs(struct task_struct *task, void *base, size_t size, + ds_ovfl_callback_t ovfl); + +/* + * Release BTS or PEBS resources + * + * Frees buffers allocated on ds_request. + * + * Returns 0 on success; -Eerrno otherwise + * + * task: the task to release resources for; + * NULL to release resources for the current cpu + */ +extern int ds_release_bts(struct task_struct *task); +extern int ds_release_pebs(struct task_struct *task); + +/* + * Return the (array) index of the write pointer. + * (assuming an array of BTS/PEBS records) + * + * Returns -Eerrno on error + * + * task: the task to access; + * NULL to access the current cpu + * pos (out): if not NULL, will hold the result + */ +extern int ds_get_bts_index(struct task_struct *task, size_t *pos); +extern int ds_get_pebs_index(struct task_struct *task, size_t *pos); + +/* + * Return the (array) index one record beyond the end of the array. + * (assuming an array of BTS/PEBS records) + * + * Returns -Eerrno on error + * + * task: the task to access; + * NULL to access the current cpu + * pos (out): if not NULL, will hold the result + */ +extern int ds_get_bts_end(struct task_struct *task, size_t *pos); +extern int ds_get_pebs_end(struct task_struct *task, size_t *pos); + +/* + * Provide a pointer to the BTS/PEBS record at parameter index. + * (assuming an array of BTS/PEBS records) + * + * The pointer points directly into the buffer. The user is + * responsible for copying the record. + * + * Returns the size of a single record on success; -Eerrno on error + * + * task: the task to access; + * NULL to access the current cpu + * index: the index of the requested record + * record (out): pointer to the requested record + */ +extern int ds_access_bts(struct task_struct *task, + size_t index, const void **record); +extern int ds_access_pebs(struct task_struct *task, + size_t index, const void **record); + +/* + * Write one or more BTS/PEBS records at the write pointer index and + * advance the write pointer. + * + * If size is not a multiple of the record size, trailing bytes are + * zeroed out. + * + * May result in one or more overflow notifications. + * + * If called during overflow handling, that is, with index >= + * interrupt threshold, the write will wrap around. + * + * An overflow notification is given if and when the interrupt + * threshold is reached during or after the write. + * + * Returns the number of bytes written or -Eerrno. + * + * task: the task to access; + * NULL to access the current cpu + * buffer: the buffer to write + * size: the size of the buffer + */ +extern int ds_write_bts(struct task_struct *task, + const void *buffer, size_t size); +extern int ds_write_pebs(struct task_struct *task, + const void *buffer, size_t size); + +/* + * Same as ds_write_bts/pebs, but omit ownership checks. + * + * This is needed to have some other task than the owner of the + * BTS/PEBS buffer or the parameter task itself write into the + * respective buffer. + */ +extern int ds_unchecked_write_bts(struct task_struct *task, + const void *buffer, size_t size); +extern int ds_unchecked_write_pebs(struct task_struct *task, + const void *buffer, size_t size); + +/* + * Reset the write pointer of the BTS/PEBS buffer. + * + * Returns 0 on success; -Eerrno on error + * + * task: the task to access; + * NULL to access the current cpu + */ +extern int ds_reset_bts(struct task_struct *task); +extern int ds_reset_pebs(struct task_struct *task); + +/* + * Clear the BTS/PEBS buffer and reset the write pointer. + * The entire buffer will be zeroed out. + * + * Returns 0 on success; -Eerrno on error + * + * task: the task to access; + * NULL to access the current cpu + */ +extern int ds_clear_bts(struct task_struct *task); +extern int ds_clear_pebs(struct task_struct *task); + +/* + * Provide the PEBS counter reset value. + * + * Returns 0 on success; -Eerrno on error + * + * task: the task to access; + * NULL to access the current cpu + * value (out): the counter reset value + */ +extern int ds_get_pebs_reset(struct task_struct *task, u64 *value); + +/* + * Set the PEBS counter reset value. + * + * Returns 0 on success; -Eerrno on error + * + * task: the task to access; + * NULL to access the current cpu + * value: the new counter reset value + */ +extern int ds_set_pebs_reset(struct task_struct *task, u64 value); + +/* + * Initialization + */ +struct cpuinfo_x86; +extern void __cpuinit ds_init_intel(struct cpuinfo_x86 *); + + -struct bts_struct { - u64 qualifier; - union { - /* BTS_BRANCH */ - struct { - u64 from_ip; - u64 to_ip; - } lbr; - /* BTS_TASK_ARRIVES or - BTS_TASK_DEPARTS */ - u64 jiffies; - } variant; +/* + * The DS context - part of struct thread_struct. + */ +struct ds_context { + /* pointer to the DS configuration; goes into MSR_IA32_DS_AREA */ + unsigned char *ds; + /* the owner of the BTS and PEBS configuration, respectively */ + struct task_struct *owner[2]; + /* buffer overflow notification function for BTS and PEBS */ + ds_ovfl_callback_t callback[2]; + /* the original buffer address */ + void *buffer[2]; + /* the number of allocated pages for on-request allocated buffers */ + unsigned int pages[2]; + /* use count */ + unsigned long count; + /* a pointer to the context location inside the thread_struct + * or the per_cpu context array */ + struct ds_context **this; + /* a pointer to the task owning this context, or NULL, if the + * context is owned by a cpu */ + struct task_struct *task; }; -/* Overflow handling mechanisms */ -#define DS_O_SIGNAL 1 /* send overflow signal */ -#define DS_O_WRAP 2 /* wrap around */ - -extern int ds_allocate(void **, size_t); -extern int ds_free(void **); -extern int ds_get_bts_size(void *); -extern int ds_get_bts_end(void *); -extern int ds_get_bts_index(void *); -extern int ds_set_overflow(void *, int); -extern int ds_get_overflow(void *); -extern int ds_clear(void *); -extern int ds_read_bts(void *, int, struct bts_struct *); -extern int ds_write_bts(void *, const struct bts_struct *); -extern unsigned long ds_debugctl_mask(void); -extern void __cpuinit ds_init_intel(struct cpuinfo_x86 *c); +/* called by exit_thread() to free leftover contexts */ +extern void ds_free(struct ds_context *context); + +#else /* CONFIG_X86_DS */ + +#define ds_init_intel(config) do {} while (0) +#endif /* CONFIG_X86_DS */ #endif /* _ASM_X86_DS_H */ diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index 559105220a4..beaccb71628 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -20,6 +20,7 @@ struct mm_struct; #include #include #include +#include #include #include @@ -415,9 +416,14 @@ struct thread_struct { unsigned io_bitmap_max; /* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set. */ unsigned long debugctlmsr; -/* Debug Store - if not 0 points to a DS Save Area configuration; - * goes into MSR_IA32_DS_AREA */ - unsigned long ds_area_msr; +#ifdef CONFIG_X86_DS +/* Debug Store context; see include/asm-x86/ds.h; goes into MSR_IA32_DS_AREA */ + struct ds_context *ds_ctx; +#endif /* CONFIG_X86_DS */ +#ifdef CONFIG_X86_PTRACE_BTS +/* the signal to send on a bts buffer overflow */ + unsigned int bts_ovfl_signal; +#endif /* CONFIG_X86_PTRACE_BTS */ }; static inline unsigned long native_get_debugreg(int regno) diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h index f224eb3c315..9bcaa75cbca 100644 --- a/include/asm-x86/ptrace-abi.h +++ b/include/asm-x86/ptrace-abi.h @@ -80,8 +80,9 @@ #define PTRACE_SINGLEBLOCK 33 /* resume execution until next branch */ -#ifndef __ASSEMBLY__ +#ifdef CONFIG_X86_PTRACE_BTS +#ifndef __ASSEMBLY__ #include /* configuration/status structure used in PTRACE_BTS_CONFIG and @@ -97,20 +98,20 @@ struct ptrace_bts_config { /* actual size of bts_struct in bytes */ __u32 bts_size; }; -#endif +#endif /* __ASSEMBLY__ */ #define PTRACE_BTS_O_TRACE 0x1 /* branch trace */ #define PTRACE_BTS_O_SCHED 0x2 /* scheduling events w/ jiffies */ #define PTRACE_BTS_O_SIGNAL 0x4 /* send SIG on buffer overflow instead of wrapping around */ -#define PTRACE_BTS_O_CUT_SIZE 0x8 /* cut requested size to max available - instead of failing */ +#define PTRACE_BTS_O_ALLOC 0x8 /* (re)allocate buffer */ #define PTRACE_BTS_CONFIG 40 /* Configure branch trace recording. ADDR points to a struct ptrace_bts_config. DATA gives the size of that buffer. - A new buffer is allocated, iff the size changes. + A new buffer is allocated, if requested in the flags. + An overflow signal may only be requested for new buffers. Returns the number of bytes read. */ #define PTRACE_BTS_STATUS 41 @@ -119,7 +120,7 @@ struct ptrace_bts_config { Returns the number of bytes written. */ #define PTRACE_BTS_SIZE 42 -/* Return the number of available BTS records. +/* Return the number of available BTS records for draining. DATA and ADDR are ignored. */ #define PTRACE_BTS_GET 43 @@ -139,5 +140,6 @@ struct ptrace_bts_config { BTS records are read from oldest to newest. Returns number of BTS records drained. */ +#endif /* CONFIG_X86_PTRACE_BTS */ #endif diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h index 9f922b0b95d..6303701d18e 100644 --- a/include/asm-x86/ptrace.h +++ b/include/asm-x86/ptrace.h @@ -125,14 +125,48 @@ struct pt_regs { #endif /* __KERNEL__ */ #endif /* !__i386__ */ + +#ifdef CONFIG_X86_PTRACE_BTS +/* a branch trace record entry + * + * In order to unify the interface between various processor versions, + * we use the below data structure for all processors. + */ +enum bts_qualifier { + BTS_INVALID = 0, + BTS_BRANCH, + BTS_TASK_ARRIVES, + BTS_TASK_DEPARTS +}; + +struct bts_struct { + __u64 qualifier; + union { + /* BTS_BRANCH */ + struct { + __u64 from_ip; + __u64 to_ip; + } lbr; + /* BTS_TASK_ARRIVES or + BTS_TASK_DEPARTS */ + __u64 jiffies; + } variant; +}; +#endif /* CONFIG_X86_PTRACE_BTS */ + #ifdef __KERNEL__ -/* the DS BTS struct is used for ptrace as well */ -#include +#include +struct cpuinfo_x86; struct task_struct; +#ifdef CONFIG_X86_PTRACE_BTS +extern void __cpuinit ptrace_bts_init_intel(struct cpuinfo_x86 *); extern void ptrace_bts_take_timestamp(struct task_struct *, enum bts_qualifier); +#else +#define ptrace_bts_init_intel(config) do {} while (0) +#endif /* CONFIG_X86_PTRACE_BTS */ extern unsigned long profile_pc(struct pt_regs *regs); -- cgit v1.2.3-70-g09d2 From 59ea746337c69f6a5f1bc4d5e8544b3cbf12f801 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Thu, 12 Jun 2008 13:56:40 +0200 Subject: MM: virtual address debug Add some (configurable) expensive sanity checking to catch wrong address translations on x86. - create linux/mmdebug.h file to be able include this file in asm headers to not get unsolvable loops in header files - __phys_addr on x86_32 became a function in ioremap.c since PAGE_OFFSET, is_vmalloc_addr and VMALLOC_* non-constasts are undefined if declared in page_32.h - add __phys_addr_const for initializing doublefault_tss.__cr3 Tested on 386, 386pae, x86_64 and x86_64 numa=fake=2. Contains Andi's enable numa virtual address debug patch. Signed-off-by: Jiri Slaby Cc: Andi Kleen Signed-off-by: Ingo Molnar --- arch/x86/kernel/doublefault_32.c | 2 +- arch/x86/mm/ioremap.c | 31 ++++++++++++++++++++++++------- include/asm-x86/mmzone_64.h | 2 +- include/asm-x86/page_32.h | 3 ++- include/linux/mm.h | 7 +------ include/linux/mmdebug.h | 18 ++++++++++++++++++ lib/Kconfig.debug | 9 +++++++++ mm/vmalloc.c | 5 +++++ 8 files changed, 61 insertions(+), 16 deletions(-) create mode 100644 include/linux/mmdebug.h (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c index a47798b59f0..395acb12b0d 100644 --- a/arch/x86/kernel/doublefault_32.c +++ b/arch/x86/kernel/doublefault_32.c @@ -66,6 +66,6 @@ struct tss_struct doublefault_tss __cacheline_aligned = { .ds = __USER_DS, .fs = __KERNEL_PERCPU, - .__cr3 = __pa(swapper_pg_dir) + .__cr3 = __phys_addr_const((unsigned long)swapper_pg_dir) } }; diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 2b2bb3f9b68..a78ffef62a2 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -23,18 +23,26 @@ #ifdef CONFIG_X86_64 -unsigned long __phys_addr(unsigned long x) +static inline int phys_addr_valid(unsigned long addr) { - if (x >= __START_KERNEL_map) - return x - __START_KERNEL_map + phys_base; - return x - PAGE_OFFSET; + return addr < (1UL << boot_cpu_data.x86_phys_bits); } -EXPORT_SYMBOL(__phys_addr); -static inline int phys_addr_valid(unsigned long addr) +unsigned long __phys_addr(unsigned long x) { - return addr < (1UL << boot_cpu_data.x86_phys_bits); + if (x >= __START_KERNEL_map) { + x -= __START_KERNEL_map; + VIRTUAL_BUG_ON(x >= KERNEL_IMAGE_SIZE); + x += phys_base; + } else { + VIRTUAL_BUG_ON(x < PAGE_OFFSET); + x -= PAGE_OFFSET; + VIRTUAL_BUG_ON(system_state == SYSTEM_BOOTING ? x > MAXMEM : + !phys_addr_valid(x)); + } + return x; } +EXPORT_SYMBOL(__phys_addr); #else @@ -43,6 +51,15 @@ static inline int phys_addr_valid(unsigned long addr) return 1; } +unsigned long __phys_addr(unsigned long x) +{ + /* VMALLOC_* aren't constants; not available at the boot time */ + VIRTUAL_BUG_ON(x < PAGE_OFFSET || (system_state != SYSTEM_BOOTING && + is_vmalloc_addr((void *)x))); + return x - PAGE_OFFSET; +} +EXPORT_SYMBOL(__phys_addr); + #endif int page_is_ram(unsigned long pagenr) diff --git a/include/asm-x86/mmzone_64.h b/include/asm-x86/mmzone_64.h index 594bd0dc1d0..facde3e5314 100644 --- a/include/asm-x86/mmzone_64.h +++ b/include/asm-x86/mmzone_64.h @@ -7,7 +7,7 @@ #ifdef CONFIG_NUMA -#define VIRTUAL_BUG_ON(x) +#include #include diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index 424e82f8ae2..9159bfb9dcf 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -64,7 +64,8 @@ typedef struct page *pgtable_t; #endif #ifndef __ASSEMBLY__ -#define __phys_addr(x) ((x) - PAGE_OFFSET) +#define __phys_addr_const(x) ((x) - PAGE_OFFSET) +extern unsigned long __phys_addr(unsigned long); #define __phys_reloc_hide(x) RELOC_HIDE((x), 0) #ifdef CONFIG_FLATMEM diff --git a/include/linux/mm.h b/include/linux/mm.h index 586a943cab0..3414a8813e9 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -7,6 +7,7 @@ #include #include +#include #include #include #include @@ -210,12 +211,6 @@ struct inode; */ #include -#ifdef CONFIG_DEBUG_VM -#define VM_BUG_ON(cond) BUG_ON(cond) -#else -#define VM_BUG_ON(condition) do { } while(0) -#endif - /* * Methods to modify the page usage count. * diff --git a/include/linux/mmdebug.h b/include/linux/mmdebug.h new file mode 100644 index 00000000000..860ed1a71bb --- /dev/null +++ b/include/linux/mmdebug.h @@ -0,0 +1,18 @@ +#ifndef LINUX_MM_DEBUG_H +#define LINUX_MM_DEBUG_H 1 + +#include + +#ifdef CONFIG_DEBUG_VM +#define VM_BUG_ON(cond) BUG_ON(cond) +#else +#define VM_BUG_ON(cond) do { } while(0) +#endif + +#ifdef CONFIG_DEBUG_VIRTUAL +#define VIRTUAL_BUG_ON(cond) BUG_ON(cond) +#else +#define VIRTUAL_BUG_ON(cond) do { } while(0) +#endif + +#endif diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index d2099f41aa1..9d9dc0ddf13 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -469,6 +469,15 @@ config DEBUG_VM If unsure, say N. +config DEBUG_VIRTUAL + bool "Debug VM translations" + depends on DEBUG_KERNEL && X86 + help + Enable some costly sanity checks in virtual to page code. This can + catch mistakes with virt_to_page() and friends. + + If unsure, say N. + config DEBUG_WRITECOUNT bool "Debug filesystem writers count" depends on DEBUG_KERNEL diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 6e45b0f3d12..dc41e9c8ca6 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -180,6 +180,11 @@ struct page *vmalloc_to_page(const void *vmalloc_addr) pmd_t *pmd; pte_t *ptep, pte; + /* XXX we might need to change this if we add VIRTUAL_BUG_ON for + * architectures that do not vmalloc module space */ + VIRTUAL_BUG_ON(!is_vmalloc_addr(vmalloc_addr) && + !is_module_address(addr)); + if (!pgd_none(*pgd)) { pud = pud_offset(pgd, addr); if (!pud_none(*pud)) { -- cgit v1.2.3-70-g09d2 From a1bf9631be7332ce0641e299ddafad2d8223100f Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Thu, 12 Jun 2008 13:56:40 +0200 Subject: x86, MM: virtual address debug, v2 I've removed the test from phys_to_nid and made a function from __phys_addr only when the debugging is enabled (on x86_32). Signed-off-by: Jiri Slaby Cc: tglx@linutronix.de Cc: hpa@zytor.com Cc: Mike Travis Cc: Nick Piggin Cc: Cc: linux-mm@kvack.org Cc: Jiri Slaby Cc: Andi Kleen Signed-off-by: Ingo Molnar --- arch/x86/mm/ioremap.c | 2 ++ include/asm-x86/mmzone_64.h | 1 - include/asm-x86/page_32.h | 4 ++++ 3 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index a78ffef62a2..9dd3cb90597 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -51,6 +51,7 @@ static inline int phys_addr_valid(unsigned long addr) return 1; } +#ifdef CONFIG_DEBUG_VIRTUAL unsigned long __phys_addr(unsigned long x) { /* VMALLOC_* aren't constants; not available at the boot time */ @@ -59,6 +60,7 @@ unsigned long __phys_addr(unsigned long x) return x - PAGE_OFFSET; } EXPORT_SYMBOL(__phys_addr); +#endif #endif diff --git a/include/asm-x86/mmzone_64.h b/include/asm-x86/mmzone_64.h index facde3e5314..5e3a6cbddb4 100644 --- a/include/asm-x86/mmzone_64.h +++ b/include/asm-x86/mmzone_64.h @@ -29,7 +29,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) { unsigned nid; VIRTUAL_BUG_ON(!memnodemap); - VIRTUAL_BUG_ON((addr >> memnode_shift) >= memnodemapsize); nid = memnodemap[addr >> memnode_shift]; VIRTUAL_BUG_ON(nid >= MAX_NUMNODES || !node_data[nid]); return nid; diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index 9159bfb9dcf..71a2e424e58 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -65,7 +65,11 @@ typedef struct page *pgtable_t; #ifndef __ASSEMBLY__ #define __phys_addr_const(x) ((x) - PAGE_OFFSET) +#ifdef CONFIG_DEBUG_VIRTUAL extern unsigned long __phys_addr(unsigned long); +#else +#define __phys_addr(x) ((x) - PAGE_OFFSET) +#endif #define __phys_reloc_hide(x) RELOC_HIDE((x), 0) #ifdef CONFIG_FLATMEM -- cgit v1.2.3-70-g09d2 From d94d93ca5cc36cd78c532def62772c98fe8ba5d7 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:46 -0700 Subject: x64, x2apic/intr-remap: 8259 specific mask/unmask routines 8259 specific mask/unmask routines which be used later while enabling interrupt-remapping. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/i8259.c | 24 ++++++++++++++++++++++++ include/asm-x86/i8259.h | 3 +++ 2 files changed, 27 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/i8259.c b/arch/x86/kernel/i8259.c index dc92b49d920..4b8a53d841f 100644 --- a/arch/x86/kernel/i8259.c +++ b/arch/x86/kernel/i8259.c @@ -282,6 +282,30 @@ static int __init i8259A_init_sysfs(void) device_initcall(i8259A_init_sysfs); +void mask_8259A(void) +{ + unsigned long flags; + + spin_lock_irqsave(&i8259A_lock, flags); + + outb(0xff, PIC_MASTER_IMR); /* mask all of 8259A-1 */ + outb(0xff, PIC_SLAVE_IMR); /* mask all of 8259A-2 */ + + spin_unlock_irqrestore(&i8259A_lock, flags); +} + +void unmask_8259A(void) +{ + unsigned long flags; + + spin_lock_irqsave(&i8259A_lock, flags); + + outb(cached_master_mask, PIC_MASTER_IMR); /* restore master IRQ mask */ + outb(cached_slave_mask, PIC_SLAVE_IMR); /* restore slave IRQ mask */ + + spin_unlock_irqrestore(&i8259A_lock, flags); +} + void init_8259A(int auto_eoi) { unsigned long flags; diff --git a/include/asm-x86/i8259.h b/include/asm-x86/i8259.h index 2f98df91f1f..31112b6c595 100644 --- a/include/asm-x86/i8259.h +++ b/include/asm-x86/i8259.h @@ -57,4 +57,7 @@ static inline void outb_pic(unsigned char value, unsigned int port) extern struct irq_chip i8259A_chip; +extern void mask_8259A(void); +extern void unmask_8259A(void); + #endif /* __ASM_I8259_H__ */ -- cgit v1.2.3-70-g09d2 From 4dc2f96cacd1e74c688f94348a3bfd0a980817d5 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:47 -0700 Subject: x64, x2apic/intr-remap: ioapic routines which deal with initial io-apic RTE setup Generic ioapic specific routines which be used later during enabling interrupt-remapping. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/io_apic_64.c | 66 ++++++++++++++++++++++++++++++++++++++++++++ include/asm-x86/io_apic.h | 6 ++++ 2 files changed, 72 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index 9e645cba11c..84dd63c13d6 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic_64.c @@ -114,6 +114,9 @@ DEFINE_SPINLOCK(vector_lock); */ int nr_ioapic_registers[MAX_IO_APICS]; +/* I/O APIC RTE contents at the OS boot up */ +struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS]; + /* I/O APIC entries */ struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; int nr_ioapics; @@ -446,6 +449,69 @@ static void clear_IO_APIC (void) clear_IO_APIC_pin(apic, pin); } +/* + * Saves and masks all the unmasked IO-APIC RTE's + */ +int save_mask_IO_APIC_setup(void) +{ + union IO_APIC_reg_01 reg_01; + unsigned long flags; + int apic, pin; + + /* + * The number of IO-APIC IRQ registers (== #pins): + */ + for (apic = 0; apic < nr_ioapics; apic++) { + spin_lock_irqsave(&ioapic_lock, flags); + reg_01.raw = io_apic_read(apic, 1); + spin_unlock_irqrestore(&ioapic_lock, flags); + nr_ioapic_registers[apic] = reg_01.bits.entries+1; + } + + for (apic = 0; apic < nr_ioapics; apic++) { + early_ioapic_entries[apic] = + kzalloc(sizeof(struct IO_APIC_route_entry) * + nr_ioapic_registers[apic], GFP_KERNEL); + if (!early_ioapic_entries[apic]) + return -ENOMEM; + } + + for (apic = 0; apic < nr_ioapics; apic++) + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + struct IO_APIC_route_entry entry; + + entry = early_ioapic_entries[apic][pin] = + ioapic_read_entry(apic, pin); + if (!entry.mask) { + entry.mask = 1; + ioapic_write_entry(apic, pin, entry); + } + } + return 0; +} + +void restore_IO_APIC_setup(void) +{ + int apic, pin; + + for (apic = 0; apic < nr_ioapics; apic++) + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) + ioapic_write_entry(apic, pin, + early_ioapic_entries[apic][pin]); +} + +void reinit_intr_remapped_IO_APIC(int intr_remapping) +{ + /* + * for now plain restore of previous settings. + * TBD: In the case of OS enabling interrupt-remapping, + * IO-APIC RTE's need to be setup to point to interrupt-remapping + * table entries. for now, do a plain restore, and wait for + * the setup_IO_APIC_irqs() to do proper initialization. + */ + restore_IO_APIC_setup(); +} + int skip_ioapic_setup; int ioapic_force; diff --git a/include/asm-x86/io_apic.h b/include/asm-x86/io_apic.h index 14f82bbcb5f..1c4a99d882f 100644 --- a/include/asm-x86/io_apic.h +++ b/include/asm-x86/io_apic.h @@ -183,6 +183,12 @@ extern int io_apic_set_pci_routing(int ioapic, int pin, int irq, extern int (*ioapic_renumber_irq)(int ioapic, int irq); extern void ioapic_init_mappings(void); +#ifdef CONFIG_X86_64 +extern int save_mask_IO_APIC_setup(void); +extern void restore_IO_APIC_setup(void); +extern void reinit_intr_remapped_IO_APIC(int); +#endif + #else /* !CONFIG_X86_IO_APIC */ #define io_apic_assign_pci_irqs 0 static const int timer_through_8259 = 0; -- cgit v1.2.3-70-g09d2 From 0c81c746f9bdbfaafe64322d540c8b7b59c27314 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:48 -0700 Subject: x64, x2apic/intr-remap: introduce read_apic_id() to genapic routines Move the read_apic_id() to genapic routines. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic_64.c | 5 +++++ arch/x86/kernel/genapic_64.c | 11 ----------- arch/x86/kernel/genapic_flat_64.c | 14 +++++++++++++- arch/x86/kernel/genx2apic_uv_x.c | 14 +++++++++++++- include/asm-x86/genapic_64.h | 1 + include/asm-x86/mach-default/mach_apic.h | 1 + include/asm-x86/mach-default/mach_apicdef.h | 3 ++- include/asm-x86/smp.h | 4 +--- 8 files changed, 36 insertions(+), 17 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c index 1e3d32e27c1..9dd4ee4735f 100644 --- a/arch/x86/kernel/apic_64.c +++ b/arch/x86/kernel/apic_64.c @@ -1096,6 +1096,11 @@ void __cpuinit generic_processor_info(int apicid, int version) cpu_set(cpu, cpu_present_map); } +int hard_smp_processor_id(void) +{ + return read_apic_id(); +} + /* * Power management */ diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c index 1fa8be5bd21..7414871751a 100644 --- a/arch/x86/kernel/genapic_64.c +++ b/arch/x86/kernel/genapic_64.c @@ -79,17 +79,6 @@ int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) return 0; } -unsigned int read_apic_id(void) -{ - unsigned int id; - - WARN_ON(preemptible() && num_online_cpus() > 1); - id = apic_read(APIC_ID); - if (uv_system_type >= UV_X2APIC) - id |= __get_cpu_var(x2apic_extra_bits); - return id; -} - enum uv_system_type get_uv_system_type(void) { return uv_system_type; diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c index 1a9c68845ee..400ed8df8b4 100644 --- a/arch/x86/kernel/genapic_flat_64.c +++ b/arch/x86/kernel/genapic_flat_64.c @@ -15,9 +15,11 @@ #include #include #include +#include #include #include #include +#include static cpumask_t flat_target_cpus(void) { @@ -95,9 +97,17 @@ static void flat_send_IPI_all(int vector) __send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL); } +static unsigned int read_xapic_id(void) +{ + unsigned int id; + + id = GET_XAPIC_ID(apic_read(APIC_ID)); + return id; +} + static int flat_apic_id_registered(void) { - return physid_isset(GET_APIC_ID(read_apic_id()), phys_cpu_present_map); + return physid_isset(read_xapic_id(), phys_cpu_present_map); } static unsigned int flat_cpu_mask_to_apicid(cpumask_t cpumask) @@ -123,6 +133,7 @@ struct genapic apic_flat = { .send_IPI_mask = flat_send_IPI_mask, .cpu_mask_to_apicid = flat_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, + .read_apic_id = read_xapic_id, }; /* @@ -187,4 +198,5 @@ struct genapic apic_physflat = { .send_IPI_mask = physflat_send_IPI_mask, .cpu_mask_to_apicid = physflat_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, + .read_apic_id = read_xapic_id, }; diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c index 711f11c30b0..1ef99be1848 100644 --- a/arch/x86/kernel/genx2apic_uv_x.c +++ b/arch/x86/kernel/genx2apic_uv_x.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -134,9 +135,19 @@ static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask) return BAD_APICID; } +static unsigned int uv_read_apic_id(void) +{ + unsigned int id; + + WARN_ON(preemptible() && num_online_cpus() > 1); + id = apic_read(APIC_ID) | __get_cpu_var(x2apic_extra_bits); + + return id; +} + static unsigned int phys_pkg_id(int index_msb) { - return GET_APIC_ID(read_apic_id()) >> index_msb; + return uv_read_apic_id() >> index_msb; } #ifdef ZZZ /* Needs x2apic patch */ @@ -159,6 +170,7 @@ struct genapic apic_x2apic_uv_x = { /* ZZZ.send_IPI_self = uv_send_IPI_self, */ .cpu_mask_to_apicid = uv_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, /* Fixme ZZZ */ + .read_apic_id = uv_read_apic_id, }; static __cpuinit void set_x2apic_extra_bits(int pnode) diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h index 647e4e5c258..d567abc347a 100644 --- a/include/asm-x86/genapic_64.h +++ b/include/asm-x86/genapic_64.h @@ -27,6 +27,7 @@ struct genapic { /* */ unsigned int (*cpu_mask_to_apicid)(cpumask_t cpumask); unsigned int (*phys_pkg_id)(int index_msb); + unsigned int (*read_apic_id)(void); }; extern struct genapic *genapic; diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h index 0b2cde5e1b7..d172c554ab9 100644 --- a/include/asm-x86/mach-default/mach_apic.h +++ b/include/asm-x86/mach-default/mach_apic.h @@ -30,6 +30,7 @@ static inline cpumask_t target_cpus(void) #define cpu_mask_to_apicid (genapic->cpu_mask_to_apicid) #define phys_pkg_id (genapic->phys_pkg_id) #define vector_allocation_domain (genapic->vector_allocation_domain) +#define read_apic_id (genapic->read_apic_id) extern void setup_apic_routing(void); #else #define INT_DELIVERY_MODE dest_LowestPrio diff --git a/include/asm-x86/mach-default/mach_apicdef.h b/include/asm-x86/mach-default/mach_apicdef.h index e4b29ba37de..453b58a67e2 100644 --- a/include/asm-x86/mach-default/mach_apicdef.h +++ b/include/asm-x86/mach-default/mach_apicdef.h @@ -5,8 +5,9 @@ #ifdef CONFIG_X86_64 #define APIC_ID_MASK (0xFFu<<24) -#define GET_APIC_ID(x) (((x)>>24)&0xFFu) +#define GET_APIC_ID(x) (x) #define SET_APIC_ID(x) (((x)<<24)) +#define GET_XAPIC_ID(x) (((x) >> 24) & 0xFFu) #else #define APIC_ID_MASK (0xF<<24) static inline unsigned get_apic_id(unsigned long x) diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 2e221f1ce0b..9848715fbd9 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -169,12 +169,10 @@ static inline unsigned int read_apic_id(void) { return *(u32 *)(APIC_BASE + APIC_ID); } -#else -extern unsigned int read_apic_id(void); #endif -# ifdef APIC_DEFINITION +# if defined(APIC_DEFINITION) || defined(CONFIG_X86_64) extern int hard_smp_processor_id(void); # else # include -- cgit v1.2.3-70-g09d2 From 1b374e4d6f8b3eb2fcd034fcc24ea8ba1dfde7aa Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:49 -0700 Subject: x64, x2apic/intr-remap: basic apic ops support Introduce basic apic operations which handle the apic programming. This will be used later to introduce another specific operations for x2apic. For the perfomance critial accesses like IPI's, EOI etc, we use the native operations as they are already referenced by different indirections like genapic, irq_chip etc. 64bit Paravirt ops can also define their apic operations accordingly. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic_32.c | 6 ++++++ arch/x86/kernel/apic_64.c | 34 ++++++++++++++++++++++++++++++++-- arch/x86/kernel/io_apic_64.c | 8 ++++---- arch/x86/kernel/paravirt.c | 8 +++++--- arch/x86/kernel/smpboot.c | 28 +++++++++++----------------- include/asm-x86/apic.h | 43 +++++++++++++++++++++++++++++++++++++------ include/asm-x86/ipi.h | 16 +++++++++++----- include/asm-x86/paravirt.h | 2 ++ include/asm-x86/smp.h | 2 +- 9 files changed, 109 insertions(+), 38 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c index 3e58b676d23..2a83c07bd88 100644 --- a/arch/x86/kernel/apic_32.c +++ b/arch/x86/kernel/apic_32.c @@ -145,6 +145,12 @@ static int modern_apic(void) return lapic_get_version() >= 0x14; } +void apic_icr_write(u32 low, u32 id) +{ + apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(id)); + apic_write_around(APIC_ICR, low); +} + void apic_wait_icr_idle(void) { while (apic_read(APIC_ICR) & APIC_ICR_BUSY) diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c index 3963f590c3d..9bb040689b3 100644 --- a/arch/x86/kernel/apic_64.c +++ b/arch/x86/kernel/apic_64.c @@ -119,13 +119,13 @@ static int modern_apic(void) return lapic_get_version() >= 0x14; } -void apic_wait_icr_idle(void) +void xapic_wait_icr_idle(void) { while (apic_read(APIC_ICR) & APIC_ICR_BUSY) cpu_relax(); } -u32 safe_apic_wait_icr_idle(void) +u32 safe_xapic_wait_icr_idle(void) { u32 send_status; int timeout; @@ -141,6 +141,36 @@ u32 safe_apic_wait_icr_idle(void) return send_status; } +void xapic_icr_write(u32 low, u32 id) +{ + apic_write(APIC_ICR2, id << 24); + apic_write(APIC_ICR, low); +} + +u64 xapic_icr_read(void) +{ + u32 icr1, icr2; + + icr2 = apic_read(APIC_ICR2); + icr1 = apic_read(APIC_ICR); + + return (icr1 | ((u64)icr2 << 32)); +} + +static struct apic_ops xapic_ops = { + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .write_atomic = native_apic_mem_write_atomic, + .icr_read = xapic_icr_read, + .icr_write = xapic_icr_write, + .wait_icr_idle = xapic_wait_icr_idle, + .safe_wait_icr_idle = safe_xapic_wait_icr_idle, +}; + +struct apic_ops __read_mostly *apic_ops = &xapic_ops; + +EXPORT_SYMBOL_GPL(apic_ops); + /** * enable_NMI_through_LVT0 - enable NMI through local vector table 0 */ diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index 84dd63c13d6..b62d42ef928 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic_64.c @@ -1157,6 +1157,7 @@ static __apicdebuginit void print_APIC_bitfield (int base) void __apicdebuginit print_local_APIC(void * dummy) { unsigned int v, ver, maxlvt; + unsigned long icr; if (apic_verbosity == APIC_QUIET) return; @@ -1200,10 +1201,9 @@ void __apicdebuginit print_local_APIC(void * dummy) v = apic_read(APIC_ESR); printk(KERN_DEBUG "... APIC ESR: %08x\n", v); - v = apic_read(APIC_ICR); - printk(KERN_DEBUG "... APIC ICR: %08x\n", v); - v = apic_read(APIC_ICR2); - printk(KERN_DEBUG "... APIC ICR2: %08x\n", v); + icr = apic_icr_read(); + printk(KERN_DEBUG "... APIC ICR: %08x\n", icr); + printk(KERN_DEBUG "... APIC ICR2: %08x\n", icr >> 32); v = apic_read(APIC_LVTT); printk(KERN_DEBUG "... APIC LVTT: %08x\n", v); diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index e0f571d58c1..b80105a0f47 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -360,9 +360,11 @@ struct pv_cpu_ops pv_cpu_ops = { struct pv_apic_ops pv_apic_ops = { #ifdef CONFIG_X86_LOCAL_APIC - .apic_write = native_apic_write, - .apic_write_atomic = native_apic_write_atomic, - .apic_read = native_apic_read, +#ifnded CONFIG_X86_64 + .apic_write = native_apic_mem_write, + .apic_write_atomic = native_apic_mem_write_atomic, + .apic_read = native_apic_mem_read, +#endif .setup_boot_clock = setup_boot_APIC_clock, .setup_secondary_clock = setup_secondary_APIC_clock, .startup_ipi_hook = paravirt_nop, diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index f35c2d8016a..c55263b3df0 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -123,7 +123,6 @@ EXPORT_PER_CPU_SYMBOL(cpu_info); static atomic_t init_deasserted; -static int boot_cpu_logical_apicid; /* representing cpus for which sibling maps can be computed */ static cpumask_t cpu_sibling_setup_map; @@ -165,6 +164,8 @@ static void unmap_cpu_to_node(int cpu) #endif #ifdef CONFIG_X86_32 +static int boot_cpu_logical_apicid; + u8 cpu_2_logical_apicid[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = BAD_APICID }; @@ -546,8 +547,7 @@ static inline void __inquire_remote_apic(int apicid) printk(KERN_CONT "a previous APIC delivery may have failed\n"); - apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(apicid)); - apic_write_around(APIC_ICR, APIC_DM_REMRD | regs[i]); + apic_icr_write(APIC_DM_REMRD | regs[i], apicid); timeout = 0; do { @@ -579,11 +579,9 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip) int maxlvt; /* Target chip */ - apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(logical_apicid)); - /* Boot on the stack */ /* Kick the second */ - apic_write_around(APIC_ICR, APIC_DM_NMI | APIC_DEST_LOGICAL); + apic_icr_write(APIC_DM_NMI | APIC_DEST_LOGICAL, logical_apicid); Dprintk("Waiting for send to finish...\n"); send_status = safe_apic_wait_icr_idle(); @@ -639,13 +637,11 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) /* * Turn INIT on target chip */ - apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid)); - /* * Send IPI */ - apic_write_around(APIC_ICR, APIC_INT_LEVELTRIG | APIC_INT_ASSERT - | APIC_DM_INIT); + apic_icr_write(APIC_INT_LEVELTRIG | APIC_INT_ASSERT | APIC_DM_INIT, + phys_apicid); Dprintk("Waiting for send to finish...\n"); send_status = safe_apic_wait_icr_idle(); @@ -655,10 +651,8 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) Dprintk("Deasserting INIT.\n"); /* Target chip */ - apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid)); - /* Send IPI */ - apic_write_around(APIC_ICR, APIC_INT_LEVELTRIG | APIC_DM_INIT); + apic_icr_write(APIC_INT_LEVELTRIG | APIC_DM_INIT, phys_apicid); Dprintk("Waiting for send to finish...\n"); send_status = safe_apic_wait_icr_idle(); @@ -703,12 +697,10 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) */ /* Target chip */ - apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid)); - /* Boot on the stack */ /* Kick the second */ - apic_write_around(APIC_ICR, APIC_DM_STARTUP - | (start_eip >> 12)); + apic_icr_write(APIC_DM_STARTUP | (start_eip >> 12), + phys_apicid); /* * Give the other CPU some time to accept the IPI. @@ -1147,7 +1139,9 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) * Setup boot CPU information */ smp_store_cpu_info(0); /* Final full version of the data */ +#ifdef CONFIG_X86_32 boot_cpu_logical_apicid = logical_smp_processor_id(); +#endif current_thread_info()->cpu = 0; /* needed? */ set_cpu_sibling_map(0); diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h index 4e2c1e517f0..6fda195337c 100644 --- a/include/asm-x86/apic.h +++ b/include/asm-x86/apic.h @@ -47,32 +47,59 @@ extern int disable_apic; #ifdef CONFIG_PARAVIRT #include #else -#define apic_write native_apic_write -#define apic_write_atomic native_apic_write_atomic -#define apic_read native_apic_read +#ifndef CONFIG_X86_64 +#define apic_write native_apic_mem_write +#define apic_write_atomic native_apic_mem_write_atomic +#define apic_read native_apic_mem_read +#endif #define setup_boot_clock setup_boot_APIC_clock #define setup_secondary_clock setup_secondary_APIC_clock #endif extern int is_vsmp_box(void); -static inline void native_apic_write(unsigned long reg, u32 v) +static inline void native_apic_mem_write(u32 reg, u32 v) { *((volatile u32 *)(APIC_BASE + reg)) = v; } -static inline void native_apic_write_atomic(unsigned long reg, u32 v) +static inline void native_apic_mem_write_atomic(u32 reg, u32 v) { (void)xchg((u32 *)(APIC_BASE + reg), v); } -static inline u32 native_apic_read(unsigned long reg) +static inline u32 native_apic_mem_read(u32 reg) { return *((volatile u32 *)(APIC_BASE + reg)); } +#ifdef CONFIG_X86_32 extern void apic_wait_icr_idle(void); extern u32 safe_apic_wait_icr_idle(void); +extern void apic_icr_write(u32 low, u32 id); +#else + +struct apic_ops { + u32 (*read)(u32 reg); + void (*write)(u32 reg, u32 v); + void (*write_atomic)(u32 reg, u32 v); + u64 (*icr_read)(void); + void (*icr_write)(u32 low, u32 high); + void (*wait_icr_idle)(void); + u32 (*safe_wait_icr_idle)(void); +}; + +extern struct apic_ops *apic_ops; + +#define apic_read (apic_ops->read) +#define apic_write (apic_ops->write) +#define apic_write_atomic (apic_ops->write_atomic) +#define apic_icr_read (apic_ops->icr_read) +#define apic_icr_write (apic_ops->icr_write) +#define apic_wait_icr_idle (apic_ops->wait_icr_idle) +#define safe_apic_wait_icr_idle (apic_ops->safe_wait_icr_idle) +#endif + extern int get_physical_broadcast(void); #ifdef CONFIG_X86_GOOD_APIC @@ -95,7 +122,11 @@ static inline void ack_APIC_irq(void) */ /* Docs say use 0 for future compatibility */ +#ifdef CONFIG_X86_32 apic_write_around(APIC_EOI, 0); +#else + native_apic_mem_write(APIC_EOI, 0); +#endif } extern int lapic_get_maxlvt(void); diff --git a/include/asm-x86/ipi.h b/include/asm-x86/ipi.h index 196d63c28aa..3d8d6a6c1f8 100644 --- a/include/asm-x86/ipi.h +++ b/include/asm-x86/ipi.h @@ -49,6 +49,12 @@ static inline int __prepare_ICR2(unsigned int mask) return SET_APIC_DEST_FIELD(mask); } +static inline void __xapic_wait_icr_idle(void) +{ + while (native_apic_mem_read(APIC_ICR) & APIC_ICR_BUSY) + cpu_relax(); +} + static inline void __send_IPI_shortcut(unsigned int shortcut, int vector, unsigned int dest) { @@ -64,7 +70,7 @@ static inline void __send_IPI_shortcut(unsigned int shortcut, int vector, /* * Wait for idle. */ - apic_wait_icr_idle(); + __xapic_wait_icr_idle(); /* * No need to touch the target chip field @@ -74,7 +80,7 @@ static inline void __send_IPI_shortcut(unsigned int shortcut, int vector, /* * Send the IPI. The write to APIC_ICR fires this off. */ - apic_write(APIC_ICR, cfg); + native_apic_mem_write(APIC_ICR, cfg); } /* @@ -92,13 +98,13 @@ static inline void __send_IPI_dest_field(unsigned int mask, int vector, if (unlikely(vector == NMI_VECTOR)) safe_apic_wait_icr_idle(); else - apic_wait_icr_idle(); + __xapic_wait_icr_idle(); /* * prepare target chip field */ cfg = __prepare_ICR2(mask); - apic_write(APIC_ICR2, cfg); + native_apic_mem_write(APIC_ICR2, cfg); /* * program the ICR @@ -108,7 +114,7 @@ static inline void __send_IPI_dest_field(unsigned int mask, int vector, /* * Send the IPI. The write to APIC_ICR fires this off. */ - apic_write(APIC_ICR, cfg); + native_apic_mem_write(APIC_ICR, cfg); } static inline void send_IPI_mask_sequence(cpumask_t mask, int vector) diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index ef5e8ec6a6a..10adac02e6d 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -891,6 +891,7 @@ static inline void slow_down_io(void) /* * Basic functions accessing APICs. */ +#ifndef CONFIG_X86_64 static inline void apic_write(unsigned long reg, u32 v) { PVOP_VCALL2(pv_apic_ops.apic_write, reg, v); @@ -905,6 +906,7 @@ static inline u32 apic_read(unsigned long reg) { return PVOP_CALL1(unsigned long, pv_apic_ops.apic_read, reg); } +#endif static inline void setup_boot_clock(void) { diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 9848715fbd9..d9d007d2278 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -158,13 +158,13 @@ extern int safe_smp_processor_id(void); #ifdef CONFIG_X86_LOCAL_APIC +#ifndef CONFIG_X86_64 static inline int logical_smp_processor_id(void) { /* we don't want to mark this access volatile - bad code generation */ return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR)); } -#ifndef CONFIG_X86_64 static inline unsigned int read_apic_id(void) { return *(u32 *)(APIC_BASE + APIC_ID); -- cgit v1.2.3-70-g09d2 From 32e1d0a0651004f5fe47f85a2a5c725ad579a90c Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:50 -0700 Subject: x64, x2apic/intr-remap: cpuid bits for x2apic feature cpuid feature for x2apic. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/feature_names.c | 2 +- include/asm-x86/cpufeature.h | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/feature_names.c b/arch/x86/kernel/cpu/feature_names.c index e43ad4ad4cb..0bf4d37a048 100644 --- a/arch/x86/kernel/cpu/feature_names.c +++ b/arch/x86/kernel/cpu/feature_names.c @@ -45,7 +45,7 @@ const char * const x86_cap_flags[NCAPINTS*32] = { /* Intel-defined (#2) */ "pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est", "tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL, - NULL, NULL, "dca", "sse4_1", "sse4_2", NULL, NULL, "popcnt", + NULL, NULL, "dca", "sse4_1", "sse4_2", "x2apic", NULL, "popcnt", NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, /* VIA/Cyrix/Centaur-defined */ diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 75ef959db32..5be9510ee01 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -90,6 +90,7 @@ #define X86_FEATURE_CX16 (4*32+13) /* CMPXCHG16B */ #define X86_FEATURE_XTPR (4*32+14) /* Send Task Priority Messages */ #define X86_FEATURE_DCA (4*32+18) /* Direct Cache Access */ +#define X86_FEATURE_X2APIC (4*32+21) /* x2APIC */ /* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */ #define X86_FEATURE_XSTORE (5*32+ 2) /* on-CPU RNG present (xstore insn) */ @@ -188,6 +189,7 @@ extern const char * const x86_power_flags[32]; #define cpu_has_gbpages boot_cpu_has(X86_FEATURE_GBPAGES) #define cpu_has_arch_perfmon boot_cpu_has(X86_FEATURE_ARCH_PERFMON) #define cpu_has_pat boot_cpu_has(X86_FEATURE_PAT) +#define cpu_has_x2apic boot_cpu_has(X86_FEATURE_X2APIC) #if defined(CONFIG_X86_INVLPG) || defined(CONFIG_X86_64) # define cpu_has_invlpg 1 -- cgit v1.2.3-70-g09d2 From 13c88fb58d0112d47f7839f24a755715c6218822 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:52 -0700 Subject: x64, x2apic/intr-remap: x2apic ops for x2apic mode support x2apic ops for x2apic mode support. This uses MSR interface and differs slightly from the xapic register layout. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic_64.c | 35 +++++++++++++++++++++++++++++++++++ include/asm-x86/apic.h | 22 ++++++++++++++++++++++ include/asm-x86/apicdef.h | 3 +++ 3 files changed, 60 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c index 9bb040689b3..a969ef78e12 100644 --- a/arch/x86/kernel/apic_64.c +++ b/arch/x86/kernel/apic_64.c @@ -171,6 +171,41 @@ struct apic_ops __read_mostly *apic_ops = &xapic_ops; EXPORT_SYMBOL_GPL(apic_ops); +static void x2apic_wait_icr_idle(void) +{ + /* no need to wait for icr idle in x2apic */ + return; +} + +static u32 safe_x2apic_wait_icr_idle(void) +{ + /* no need to wait for icr idle in x2apic */ + return 0; +} + +void x2apic_icr_write(u32 low, u32 id) +{ + wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low); +} + +u64 x2apic_icr_read(void) +{ + unsigned long val; + + rdmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), val); + return val; +} + +static struct apic_ops x2apic_ops = { + .read = native_apic_msr_read, + .write = native_apic_msr_write, + .write_atomic = native_apic_msr_write, + .icr_read = x2apic_icr_read, + .icr_write = x2apic_icr_write, + .wait_icr_idle = x2apic_wait_icr_idle, + .safe_wait_icr_idle = safe_x2apic_wait_icr_idle, +}; + /** * enable_NMI_through_LVT0 - enable NMI through local vector table 0 */ diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h index 6fda195337c..bb54928373c 100644 --- a/include/asm-x86/apic.h +++ b/include/asm-x86/apic.h @@ -7,6 +7,8 @@ #include #include #include +#include +#include #define ARCH_APICTIMER_STOPS_ON_C3 1 @@ -73,6 +75,26 @@ static inline u32 native_apic_mem_read(u32 reg) return *((volatile u32 *)(APIC_BASE + reg)); } +static inline void native_apic_msr_write(u32 reg, u32 v) +{ + if (reg == APIC_DFR || reg == APIC_ID || reg == APIC_LDR || + reg == APIC_LVR) + return; + + wrmsr(APIC_BASE_MSR + (reg >> 4), v, 0); +} + +static inline u32 native_apic_msr_read(u32 reg) +{ + u32 low, high; + + if (reg == APIC_DFR) + return -1; + + rdmsr(APIC_BASE_MSR + (reg >> 4), low, high); + return low; +} + #ifdef CONFIG_X86_32 extern void apic_wait_icr_idle(void); extern u32 safe_apic_wait_icr_idle(void); diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h index 6b9008c7873..bcae297b30b 100644 --- a/include/asm-x86/apicdef.h +++ b/include/asm-x86/apicdef.h @@ -105,6 +105,7 @@ #define APIC_TMICT 0x380 #define APIC_TMCCT 0x390 #define APIC_TDCR 0x3E0 +#define APIC_SELF_IPI 0x3F0 #define APIC_TDR_DIV_TMBASE (1 << 2) #define APIC_TDR_DIV_1 0xB #define APIC_TDR_DIV_2 0x0 @@ -128,6 +129,8 @@ #define APIC_EILVT3 0x530 #define APIC_BASE (fix_to_virt(FIX_APIC_BASE)) +#define APIC_BASE_MSR 0x800 +#define X2APIC_ENABLE (1UL << 10) #ifdef CONFIG_X86_32 # define MAX_IO_APICS 64 -- cgit v1.2.3-70-g09d2 From cff73a6ffaed726780b001937d2a42efde553922 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:53 -0700 Subject: x64, x2apic/intr-remap: introcude self IPI to genapic routines Introduce self IPI op for genapic. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/genapic_64.c | 2 +- arch/x86/kernel/genapic_flat_64.c | 2 ++ include/asm-x86/genapic_64.h | 2 ++ include/asm-x86/hw_irq.h | 2 ++ include/asm-x86/mach-default/mach_apic.h | 1 + 5 files changed, 8 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c index 7414871751a..6657de609dc 100644 --- a/arch/x86/kernel/genapic_64.c +++ b/arch/x86/kernel/genapic_64.c @@ -61,7 +61,7 @@ void __init setup_apic_routing(void) /* Same for both flat and physical. */ -void send_IPI_self(int vector) +void apic_send_IPI_self(int vector) { __send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL); } diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c index 400ed8df8b4..73558682213 100644 --- a/arch/x86/kernel/genapic_flat_64.c +++ b/arch/x86/kernel/genapic_flat_64.c @@ -131,6 +131,7 @@ struct genapic apic_flat = { .send_IPI_all = flat_send_IPI_all, .send_IPI_allbutself = flat_send_IPI_allbutself, .send_IPI_mask = flat_send_IPI_mask, + .send_IPI_self = apic_send_IPI_self, .cpu_mask_to_apicid = flat_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, .read_apic_id = read_xapic_id, @@ -196,6 +197,7 @@ struct genapic apic_physflat = { .send_IPI_all = physflat_send_IPI_all, .send_IPI_allbutself = physflat_send_IPI_allbutself, .send_IPI_mask = physflat_send_IPI_mask, + .send_IPI_self = apic_send_IPI_self, .cpu_mask_to_apicid = physflat_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, .read_apic_id = read_xapic_id, diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h index d567abc347a..6777d71aabc 100644 --- a/include/asm-x86/genapic_64.h +++ b/include/asm-x86/genapic_64.h @@ -24,6 +24,7 @@ struct genapic { void (*send_IPI_mask)(cpumask_t mask, int vector); void (*send_IPI_allbutself)(int vector); void (*send_IPI_all)(int vector); + void (*send_IPI_self)(int vector); /* */ unsigned int (*cpu_mask_to_apicid)(cpumask_t cpumask); unsigned int (*phys_pkg_id)(int index_msb); @@ -36,6 +37,7 @@ extern struct genapic apic_flat; extern struct genapic apic_physflat; extern int acpi_madt_oem_check(char *, char *); +extern void apic_send_IPI_self(int vector); enum uv_system_type {UV_NONE, UV_LEGACY_APIC, UV_X2APIC, UV_NON_UNIQUE_APIC}; extern enum uv_system_type get_uv_system_type(void); extern int is_uv_system(void); diff --git a/include/asm-x86/hw_irq.h b/include/asm-x86/hw_irq.h index 18f067c310f..2ae47e7c106 100644 --- a/include/asm-x86/hw_irq.h +++ b/include/asm-x86/hw_irq.h @@ -72,7 +72,9 @@ extern void enable_IO_APIC(void); #endif /* IPI functions */ +#ifdef CONFIG_X86_32 extern void send_IPI_self(int vector); +#endif extern void send_IPI(int dest, int vector); /* Statistics */ diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h index d172c554ab9..e06d23975d6 100644 --- a/include/asm-x86/mach-default/mach_apic.h +++ b/include/asm-x86/mach-default/mach_apic.h @@ -31,6 +31,7 @@ static inline cpumask_t target_cpus(void) #define phys_pkg_id (genapic->phys_pkg_id) #define vector_allocation_domain (genapic->vector_allocation_domain) #define read_apic_id (genapic->read_apic_id) +#define send_IPI_self (genapic->send_IPI_self) extern void setup_apic_routing(void); #else #define INT_DELIVERY_MODE dest_LowestPrio -- cgit v1.2.3-70-g09d2 From 12a67cf6851871ca8df42025c94f140c303d0f7f Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:54 -0700 Subject: x64, x2apic/intr-remap: x2apic cluster mode support x2apic cluster mode support. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/Makefile | 1 + arch/x86/kernel/genapic_64.c | 2 + arch/x86/kernel/genx2apic_cluster.c | 135 ++++++++++++++++++++++++++++++++++++ include/asm-x86/genapic_64.h | 1 + 4 files changed, 139 insertions(+) create mode 100644 arch/x86/kernel/genx2apic_cluster.c (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 55ff016e9f6..bde3e9b6fec 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -95,6 +95,7 @@ obj-$(CONFIG_OLPC) += olpc.o # 64 bit specific files ifeq ($(CONFIG_X86_64),y) obj-y += genapic_64.o genapic_flat_64.o genx2apic_uv_x.o tlb_uv.o + obj-y += genx2apic_cluster.o obj-$(CONFIG_X86_PM_TIMER) += pmtimer_64.o obj-$(CONFIG_AUDIT) += audit_64.o diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c index 6657de609dc..1029e178cdf 100644 --- a/arch/x86/kernel/genapic_64.c +++ b/arch/x86/kernel/genapic_64.c @@ -38,6 +38,8 @@ void __init setup_apic_routing(void) { if (uv_system_type == UV_NON_UNIQUE_APIC) genapic = &apic_x2apic_uv_x; + else if (cpu_has_x2apic && intr_remapping_enabled) + genapic = &apic_x2apic_cluster; else #ifdef CONFIG_ACPI /* diff --git a/arch/x86/kernel/genx2apic_cluster.c b/arch/x86/kernel/genx2apic_cluster.c new file mode 100644 index 00000000000..ed0fdede800 --- /dev/null +++ b/arch/x86/kernel/genx2apic_cluster.c @@ -0,0 +1,135 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +DEFINE_PER_CPU(u32, x86_cpu_to_logical_apicid); + +/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ + +static cpumask_t x2apic_target_cpus(void) +{ + return cpumask_of_cpu(0); +} + +/* + * for now each logical cpu is in its own vector allocation domain. + */ +static cpumask_t x2apic_vector_allocation_domain(int cpu) +{ + cpumask_t domain = CPU_MASK_NONE; + cpu_set(cpu, domain); + return domain; +} + +static void __x2apic_send_IPI_dest(unsigned int apicid, int vector, + unsigned int dest) +{ + unsigned long cfg; + + cfg = __prepare_ICR(0, vector, dest); + + /* + * send the IPI. + */ + x2apic_icr_write(cfg, apicid); +} + +/* + * for now, we send the IPI's one by one in the cpumask. + * TBD: Based on the cpu mask, we can send the IPI's to the cluster group + * at once. We have 16 cpu's in a cluster. This will minimize IPI register + * writes. + */ +static void x2apic_send_IPI_mask(cpumask_t mask, int vector) +{ + unsigned long flags; + unsigned long query_cpu; + + local_irq_save(flags); + for_each_cpu_mask(query_cpu, mask) { + __x2apic_send_IPI_dest(per_cpu(x86_cpu_to_logical_apicid, query_cpu), + vector, APIC_DEST_LOGICAL); + } + local_irq_restore(flags); +} + +static void x2apic_send_IPI_allbutself(int vector) +{ + cpumask_t mask = cpu_online_map; + + cpu_clear(smp_processor_id(), mask); + + if (!cpus_empty(mask)) + x2apic_send_IPI_mask(mask, vector); +} + +static void x2apic_send_IPI_all(int vector) +{ + x2apic_send_IPI_mask(cpu_online_map, vector); +} + +static int x2apic_apic_id_registered(void) +{ + return 1; +} + +static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + cpu = first_cpu(cpumask); + if ((unsigned)cpu < NR_CPUS) + return per_cpu(x86_cpu_to_logical_apicid, cpu); + else + return BAD_APICID; +} + +static unsigned int x2apic_read_id(void) +{ + return apic_read(APIC_ID); +} + +static unsigned int phys_pkg_id(int index_msb) +{ + return x2apic_read_id() >> index_msb; +} + +static void x2apic_send_IPI_self(int vector) +{ + apic_write(APIC_SELF_IPI, vector); +} + +static void init_x2apic_ldr(void) +{ + int cpu = smp_processor_id(); + + per_cpu(x86_cpu_to_logical_apicid, cpu) = apic_read(APIC_LDR); + return; +} + +struct genapic apic_x2apic_cluster = { + .name = "cluster x2apic", + .int_delivery_mode = dest_LowestPrio, + .int_dest_mode = (APIC_DEST_LOGICAL != 0), + .target_cpus = x2apic_target_cpus, + .vector_allocation_domain = x2apic_vector_allocation_domain, + .apic_id_registered = x2apic_apic_id_registered, + .init_apic_ldr = init_x2apic_ldr, + .send_IPI_all = x2apic_send_IPI_all, + .send_IPI_allbutself = x2apic_send_IPI_allbutself, + .send_IPI_mask = x2apic_send_IPI_mask, + .send_IPI_self = x2apic_send_IPI_self, + .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, + .phys_pkg_id = phys_pkg_id, + .read_apic_id = x2apic_read_id, +}; diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h index 6777d71aabc..23246030587 100644 --- a/include/asm-x86/genapic_64.h +++ b/include/asm-x86/genapic_64.h @@ -35,6 +35,7 @@ extern struct genapic *genapic; extern struct genapic apic_flat; extern struct genapic apic_physflat; +extern struct genapic apic_x2apic_cluster; extern int acpi_madt_oem_check(char *, char *); extern void apic_send_IPI_self(int vector); -- cgit v1.2.3-70-g09d2 From 89027d35aa5b8f45ce0f7fa0911db85b46563da0 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:56 -0700 Subject: x64, x2apic/intr-remap: IO-APIC support for interrupt-remapping IO-APIC support in the presence of interrupt-remapping infrastructure. IO-APIC RTE will be programmed with interrupt-remapping table entry(IRTE) index and the IRTE will contain information about the vector, cpu destination, trigger mode etc, which traditionally was present in the IO-APIC RTE. Introduce a new irq_chip for cleaner irq migration (in the process context as opposed to the current irq migration in the context of an interrupt. interrupt-remapping infrastructure will help us achieve this cleanly). For edge triggered, irq migration is a simple atomic update(of vector and cpu destination) of IRTE and flush the hardware cache. For level triggered, we need to modify the io-apic RTE aswell with the update vector information, along with modifying IRTE with vector and cpu destination. So irq migration for level triggered is little bit more complex compared to edge triggered migration. But the good news is, we use the same algorithm for level triggered migration as we have today, only difference being, we now initiate the irq migration from process context instead of the interrupt context. In future, when we do a directed EOI (combined with cpu EOI broadcast suppression) to the IO-APIC, level triggered irq migration will also be as simple as edge triggered migration and we can do the irq migration with a simple atomic update to IO-APIC RTE. TBD: some tests/changes needed in the presence of fixup_irqs() for level triggered irq migration. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic_64.c | 1 + arch/x86/kernel/io_apic_64.c | 300 +++++++++++++++++++++++++++++++++++++--- drivers/pci/intr_remapping.c | 10 ++ include/asm-x86/apic.h | 9 ++ include/asm-x86/io_apic.h | 14 ++ include/asm-x86/irq_remapping.h | 8 ++ include/linux/dmar.h | 1 + 7 files changed, 321 insertions(+), 22 deletions(-) create mode 100644 include/asm-x86/irq_remapping.h (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c index a969ef78e12..d5c06917b5b 100644 --- a/arch/x86/kernel/apic_64.c +++ b/arch/x86/kernel/apic_64.c @@ -46,6 +46,7 @@ static int disable_apic_timer __cpuinitdata; static int apic_calibrate_pmtmr __initdata; int disable_apic; +int x2apic; /* Local APIC timer works in C2 */ int local_apic_timer_c2_ok; diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index b62d42ef928..9bd02ef049a 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic_64.c @@ -37,6 +37,7 @@ #include #endif #include +#include #include #include @@ -48,6 +49,7 @@ #include #include #include +#include #include #include @@ -312,7 +314,12 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector) pin = entry->pin; if (pin == -1) break; - io_apic_write(apic, 0x11 + pin*2, dest); + /* + * With interrupt-remapping, destination information comes + * from interrupt-remapping table entry. + */ + if (!irq_remapped(irq)) + io_apic_write(apic, 0x11 + pin*2, dest); reg = io_apic_read(apic, 0x10 + pin*2); reg &= ~IO_APIC_REDIR_VECTOR_MASK; reg |= vector; @@ -906,18 +913,98 @@ void setup_vector_irq(int cpu) static struct irq_chip ioapic_chip; +#ifdef CONFIG_INTR_REMAP +static struct irq_chip ir_ioapic_chip; +#endif static void ioapic_register_intr(int irq, unsigned long trigger) { - if (trigger) { + if (trigger) irq_desc[irq].status |= IRQ_LEVEL; - set_irq_chip_and_handler_name(irq, &ioapic_chip, - handle_fasteoi_irq, "fasteoi"); - } else { + else irq_desc[irq].status &= ~IRQ_LEVEL; + +#ifdef CONFIG_INTR_REMAP + if (irq_remapped(irq)) { + irq_desc[irq].status |= IRQ_MOVE_PCNTXT; + if (trigger) + set_irq_chip_and_handler_name(irq, &ir_ioapic_chip, + handle_fasteoi_irq, + "fasteoi"); + else + set_irq_chip_and_handler_name(irq, &ir_ioapic_chip, + handle_edge_irq, "edge"); + return; + } +#endif + if (trigger) + set_irq_chip_and_handler_name(irq, &ioapic_chip, + handle_fasteoi_irq, + "fasteoi"); + else set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_edge_irq, "edge"); +} + +static int setup_ioapic_entry(int apic, int irq, + struct IO_APIC_route_entry *entry, + unsigned int destination, int trigger, + int polarity, int vector) +{ + /* + * add it to the IO-APIC irq-routing table: + */ + memset(entry,0,sizeof(*entry)); + +#ifdef CONFIG_INTR_REMAP + if (intr_remapping_enabled) { + struct intel_iommu *iommu = map_ioapic_to_ir(apic); + struct irte irte; + struct IR_IO_APIC_route_entry *ir_entry = + (struct IR_IO_APIC_route_entry *) entry; + int index; + + if (!iommu) + panic("No mapping iommu for ioapic %d\n", apic); + + index = alloc_irte(iommu, irq, 1); + if (index < 0) + panic("Failed to allocate IRTE for ioapic %d\n", apic); + + memset(&irte, 0, sizeof(irte)); + + irte.present = 1; + irte.dst_mode = INT_DEST_MODE; + irte.trigger_mode = trigger; + irte.dlvry_mode = INT_DELIVERY_MODE; + irte.vector = vector; + irte.dest_id = IRTE_DEST(destination); + + modify_irte(irq, &irte); + + ir_entry->index2 = (index >> 15) & 0x1; + ir_entry->zero = 0; + ir_entry->format = 1; + ir_entry->index = (index & 0x7fff); + } else +#endif + { + entry->delivery_mode = INT_DELIVERY_MODE; + entry->dest_mode = INT_DEST_MODE; + entry->dest = destination; } + + entry->mask = 0; /* enable IRQ */ + entry->trigger = trigger; + entry->polarity = polarity; + entry->vector = vector; + + /* Mask level triggered irqs. + * Use IRQ_DELAYED_DISABLE for edge triggered irqs. + */ + if (trigger) + entry->mask = 1; + return 0; } static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, @@ -942,24 +1029,15 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, apic, mp_ioapics[apic].mp_apicid, pin, cfg->vector, irq, trigger, polarity); - /* - * add it to the IO-APIC irq-routing table: - */ - memset(&entry,0,sizeof(entry)); - - entry.delivery_mode = INT_DELIVERY_MODE; - entry.dest_mode = INT_DEST_MODE; - entry.dest = cpu_mask_to_apicid(mask); - entry.mask = 0; /* enable IRQ */ - entry.trigger = trigger; - entry.polarity = polarity; - entry.vector = cfg->vector; - /* Mask level triggered irqs. - * Use IRQ_DELAYED_DISABLE for edge triggered irqs. - */ - if (trigger) - entry.mask = 1; + if (setup_ioapic_entry(mp_ioapics[apic].mp_apicid, irq, &entry, + cpu_mask_to_apicid(mask), trigger, polarity, + cfg->vector)) { + printk("Failed to setup ioapic entry for ioapic %d, pin %d\n", + mp_ioapics[apic].mp_apicid, pin); + __clear_irq_vector(irq); + return; + } ioapic_register_intr(irq, trigger); if (irq < 16) @@ -1011,6 +1089,9 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, { struct IO_APIC_route_entry entry; + if (intr_remapping_enabled) + return; + memset(&entry, 0, sizeof(entry)); /* @@ -1466,6 +1547,147 @@ static int ioapic_retrigger_irq(unsigned int irq) */ #ifdef CONFIG_SMP + +#ifdef CONFIG_INTR_REMAP +static void ir_irq_migration(struct work_struct *work); + +static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration); + +/* + * Migrate the IO-APIC irq in the presence of intr-remapping. + * + * For edge triggered, irq migration is a simple atomic update(of vector + * and cpu destination) of IRTE and flush the hardware cache. + * + * For level triggered, we need to modify the io-apic RTE aswell with the update + * vector information, along with modifying IRTE with vector and destination. + * So irq migration for level triggered is little bit more complex compared to + * edge triggered migration. But the good news is, we use the same algorithm + * for level triggered migration as we have today, only difference being, + * we now initiate the irq migration from process context instead of the + * interrupt context. + * + * In future, when we do a directed EOI (combined with cpu EOI broadcast + * suppression) to the IO-APIC, level triggered irq migration will also be + * as simple as edge triggered migration and we can do the irq migration + * with a simple atomic update to IO-APIC RTE. + */ +static void migrate_ioapic_irq(int irq, cpumask_t mask) +{ + struct irq_cfg *cfg = irq_cfg + irq; + struct irq_desc *desc = irq_desc + irq; + cpumask_t tmp, cleanup_mask; + struct irte irte; + int modify_ioapic_rte = desc->status & IRQ_LEVEL; + unsigned int dest; + unsigned long flags; + + cpus_and(tmp, mask, cpu_online_map); + if (cpus_empty(tmp)) + return; + + if (get_irte(irq, &irte)) + return; + + if (assign_irq_vector(irq, mask)) + return; + + cpus_and(tmp, cfg->domain, mask); + dest = cpu_mask_to_apicid(tmp); + + if (modify_ioapic_rte) { + spin_lock_irqsave(&ioapic_lock, flags); + __target_IO_APIC_irq(irq, dest, cfg->vector); + spin_unlock_irqrestore(&ioapic_lock, flags); + } + + irte.vector = cfg->vector; + irte.dest_id = IRTE_DEST(dest); + + /* + * Modified the IRTE and flushes the Interrupt entry cache. + */ + modify_irte(irq, &irte); + + if (cfg->move_in_progress) { + cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map); + cfg->move_cleanup_count = cpus_weight(cleanup_mask); + send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR); + cfg->move_in_progress = 0; + } + + irq_desc[irq].affinity = mask; +} + +static int migrate_irq_remapped_level(int irq) +{ + int ret = -1; + + mask_IO_APIC_irq(irq); + + if (io_apic_level_ack_pending(irq)) { + /* + * Interrupt in progress. Migrating irq now will change the + * vector information in the IO-APIC RTE and that will confuse + * the EOI broadcast performed by cpu. + * So, delay the irq migration to the next instance. + */ + schedule_delayed_work(&ir_migration_work, 1); + goto unmask; + } + + /* everthing is clear. we have right of way */ + migrate_ioapic_irq(irq, irq_desc[irq].pending_mask); + + ret = 0; + irq_desc[irq].status &= ~IRQ_MOVE_PENDING; + cpus_clear(irq_desc[irq].pending_mask); + +unmask: + unmask_IO_APIC_irq(irq); + return ret; +} + +static void ir_irq_migration(struct work_struct *work) +{ + int irq; + + for (irq = 0; irq < NR_IRQS; irq++) { + struct irq_desc *desc = irq_desc + irq; + if (desc->status & IRQ_MOVE_PENDING) { + unsigned long flags; + + spin_lock_irqsave(&desc->lock, flags); + if (!desc->chip->set_affinity || + !(desc->status & IRQ_MOVE_PENDING)) { + desc->status &= ~IRQ_MOVE_PENDING; + spin_unlock_irqrestore(&desc->lock, flags); + continue; + } + + desc->chip->set_affinity(irq, + irq_desc[irq].pending_mask); + spin_unlock_irqrestore(&desc->lock, flags); + } + } +} + +/* + * Migrates the IRQ destination in the process context. + */ +static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) +{ + if (irq_desc[irq].status & IRQ_LEVEL) { + irq_desc[irq].status |= IRQ_MOVE_PENDING; + irq_desc[irq].pending_mask = mask; + migrate_irq_remapped_level(irq); + return; + } + + migrate_ioapic_irq(irq, mask); +} +#endif + asmlinkage void smp_irq_move_cleanup_interrupt(void) { unsigned vector, me; @@ -1522,6 +1744,17 @@ static void irq_complete_move(unsigned int irq) #else static inline void irq_complete_move(unsigned int irq) {} #endif +#ifdef CONFIG_INTR_REMAP +static void ack_x2apic_level(unsigned int irq) +{ + ack_x2APIC_irq(); +} + +static void ack_x2apic_edge(unsigned int irq) +{ + ack_x2APIC_irq(); +} +#endif static void ack_apic_edge(unsigned int irq) { @@ -1596,6 +1829,21 @@ static struct irq_chip ioapic_chip __read_mostly = { .retrigger = ioapic_retrigger_irq, }; +#ifdef CONFIG_INTR_REMAP +static struct irq_chip ir_ioapic_chip __read_mostly = { + .name = "IR-IO-APIC", + .startup = startup_ioapic_irq, + .mask = mask_IO_APIC_irq, + .unmask = unmask_IO_APIC_irq, + .ack = ack_x2apic_edge, + .eoi = ack_x2apic_level, +#ifdef CONFIG_SMP + .set_affinity = set_ir_ioapic_affinity_irq, +#endif + .retrigger = ioapic_retrigger_irq, +}; +#endif + static inline void init_IO_APIC_traps(void) { int irq; @@ -1783,6 +2031,8 @@ static inline void __init check_timer(void) * 8259A. */ if (pin1 == -1) { + if (intr_remapping_enabled) + panic("BIOS bug: timer not connected to IO-APIC"); pin1 = pin2; apic1 = apic2; no_pin1 = 1; @@ -1809,6 +2059,8 @@ static inline void __init check_timer(void) clear_IO_APIC_pin(0, pin1); goto out; } + if (intr_remapping_enabled) + panic("timer doesn't work through Interrupt-remapped IO-APIC"); clear_IO_APIC_pin(apic1, pin1); if (!no_pin1) apic_printk(APIC_QUIET,KERN_ERR "..MP-BIOS bug: " @@ -2401,6 +2653,10 @@ void __init setup_ioapic_dest(void) setup_IO_APIC_irq(ioapic, pin, irq, irq_trigger(irq_entry), irq_polarity(irq_entry)); +#ifdef CONFIG_INTR_REMAP + else if (intr_remapping_enabled) + set_ir_ioapic_affinity_irq(irq, TARGET_CPUS); +#endif else set_ioapic_affinity_irq(irq, TARGET_CPUS); } diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c index bddb4b19b6c..32e55c7a980 100644 --- a/drivers/pci/intr_remapping.c +++ b/drivers/pci/intr_remapping.c @@ -220,6 +220,16 @@ int flush_irte(int irq) return 0; } +struct intel_iommu *map_ioapic_to_ir(int apic) +{ + int i; + + for (i = 0; i < MAX_IO_APICS; i++) + if (ir_ioapic[i].id == apic) + return ir_ioapic[i].iommu; + return NULL; +} + int free_irte(int irq) { int index, i; diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h index bb54928373c..aa746704a5c 100644 --- a/include/asm-x86/apic.h +++ b/include/asm-x86/apic.h @@ -134,6 +134,15 @@ extern int get_physical_broadcast(void); # define apic_write_around(x, y) apic_write_atomic((x), (y)) #endif +#ifdef CONFIG_X86_64 +static inline void ack_x2APIC_irq(void) +{ + /* Docs say use 0 for future compatibility */ + native_apic_msr_write(APIC_EOI, 0); +} +#endif + + static inline void ack_APIC_irq(void) { /* diff --git a/include/asm-x86/io_apic.h b/include/asm-x86/io_apic.h index 1c4a99d882f..8dc2622714c 100644 --- a/include/asm-x86/io_apic.h +++ b/include/asm-x86/io_apic.h @@ -107,6 +107,20 @@ struct IO_APIC_route_entry { } __attribute__ ((packed)); +struct IR_IO_APIC_route_entry { + __u64 vector : 8, + zero : 3, + index2 : 1, + delivery_status : 1, + polarity : 1, + irr : 1, + trigger : 1, + mask : 1, + reserved : 31, + format : 1, + index : 15; +} __attribute__ ((packed)); + #ifdef CONFIG_X86_IO_APIC /* diff --git a/include/asm-x86/irq_remapping.h b/include/asm-x86/irq_remapping.h new file mode 100644 index 00000000000..78242c6ffa5 --- /dev/null +++ b/include/asm-x86/irq_remapping.h @@ -0,0 +1,8 @@ +#ifndef _ASM_IRQ_REMAPPING_H +#define _ASM_IRQ_REMAPPING_H + +extern int x2apic; + +#define IRTE_DEST(dest) ((x2apic) ? dest : dest << 8) + +#endif diff --git a/include/linux/dmar.h b/include/linux/dmar.h index 324bbca85a2..bf41ffa7470 100644 --- a/include/linux/dmar.h +++ b/include/linux/dmar.h @@ -109,6 +109,7 @@ extern int flush_irte(int irq); extern int free_irte(int irq); extern int irq_remapped(int irq); +extern struct intel_iommu *map_ioapic_to_ir(int apic); #else #define irq_remapped(irq) (0) #define enable_intr_remapping(mode) (-1) -- cgit v1.2.3-70-g09d2 From 75c46fa61bc5b4ccd20a168ff325c58771248fcd Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:57 -0700 Subject: x64, x2apic/intr-remap: MSI and MSI-X support for interrupt remapping infrastructure MSI and MSI-X support for interrupt remapping infrastructure. MSI address register will be programmed with interrupt-remapping table entry(IRTE) index and the IRTE will contain information about the vector, cpu destination, etc. For MSI-X, all the IRTE's will be consecutively allocated in the table, and the address registers will contain the starting index to the block and the data register will contain the subindex with in that block. This also introduces a new irq_chip for cleaner irq migration (in the process context as opposed to the current irq migration in the context of an interrupt. interrupt-remapping infrastructure will help us achieve this). As MSI is edge triggered, irq migration is a simple atomic update(of vector and cpu destination) of IRTE and flushing the hardware cache. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/io_apic_64.c | 230 +++++++++++++++++++++++++++++++++++++++++-- drivers/pci/intr_remapping.c | 11 +++ include/asm-x86/msidef.h | 4 + include/linux/dmar.h | 1 + 4 files changed, 238 insertions(+), 8 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index 9bd02ef049a..877aa2e9d7e 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic_64.c @@ -2297,6 +2297,9 @@ void destroy_irq(unsigned int irq) dynamic_irq_cleanup(irq); +#ifdef CONFIG_INTR_REMAP + free_irte(irq); +#endif spin_lock_irqsave(&vector_lock, flags); __clear_irq_vector(irq); spin_unlock_irqrestore(&vector_lock, flags); @@ -2315,10 +2318,41 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms tmp = TARGET_CPUS; err = assign_irq_vector(irq, tmp); - if (!err) { - cpus_and(tmp, cfg->domain, tmp); - dest = cpu_mask_to_apicid(tmp); + if (err) + return err; + + cpus_and(tmp, cfg->domain, tmp); + dest = cpu_mask_to_apicid(tmp); + +#ifdef CONFIG_INTR_REMAP + if (irq_remapped(irq)) { + struct irte irte; + int ir_index; + u16 sub_handle; + + ir_index = map_irq_to_irte_handle(irq, &sub_handle); + BUG_ON(ir_index == -1); + + memset (&irte, 0, sizeof(irte)); + + irte.present = 1; + irte.dst_mode = INT_DEST_MODE; + irte.trigger_mode = 0; /* edge */ + irte.dlvry_mode = INT_DELIVERY_MODE; + irte.vector = cfg->vector; + irte.dest_id = IRTE_DEST(dest); + + modify_irte(irq, &irte); + msg->address_hi = MSI_ADDR_BASE_HI; + msg->data = sub_handle; + msg->address_lo = MSI_ADDR_BASE_LO | MSI_ADDR_IR_EXT_INT | + MSI_ADDR_IR_SHV | + MSI_ADDR_IR_INDEX1(ir_index) | + MSI_ADDR_IR_INDEX2(ir_index); + } else +#endif + { msg->address_hi = MSI_ADDR_BASE_HI; msg->address_lo = MSI_ADDR_BASE_LO | @@ -2369,6 +2403,55 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) write_msi_msg(irq, &msg); irq_desc[irq].affinity = mask; } + +#ifdef CONFIG_INTR_REMAP +/* + * Migrate the MSI irq to another cpumask. This migration is + * done in the process context using interrupt-remapping hardware. + */ +static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) +{ + struct irq_cfg *cfg = irq_cfg + irq; + unsigned int dest; + cpumask_t tmp, cleanup_mask; + struct irte irte; + + cpus_and(tmp, mask, cpu_online_map); + if (cpus_empty(tmp)) + return; + + if (get_irte(irq, &irte)) + return; + + if (assign_irq_vector(irq, mask)) + return; + + cpus_and(tmp, cfg->domain, mask); + dest = cpu_mask_to_apicid(tmp); + + irte.vector = cfg->vector; + irte.dest_id = IRTE_DEST(dest); + + /* + * atomically update the IRTE with the new destination and vector. + */ + modify_irte(irq, &irte); + + /* + * After this point, all the interrupts will start arriving + * at the new destination. So, time to cleanup the previous + * vector allocation. + */ + if (cfg->move_in_progress) { + cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map); + cfg->move_cleanup_count = cpus_weight(cleanup_mask); + send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR); + cfg->move_in_progress = 0; + } + + irq_desc[irq].affinity = mask; +} +#endif #endif /* CONFIG_SMP */ /* @@ -2386,26 +2469,157 @@ static struct irq_chip msi_chip = { .retrigger = ioapic_retrigger_irq, }; -int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc) +#ifdef CONFIG_INTR_REMAP +static struct irq_chip msi_ir_chip = { + .name = "IR-PCI-MSI", + .unmask = unmask_msi_irq, + .mask = mask_msi_irq, + .ack = ack_x2apic_edge, +#ifdef CONFIG_SMP + .set_affinity = ir_set_msi_irq_affinity, +#endif + .retrigger = ioapic_retrigger_irq, +}; + +/* + * Map the PCI dev to the corresponding remapping hardware unit + * and allocate 'nvec' consecutive interrupt-remapping table entries + * in it. + */ +static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec) +{ + struct intel_iommu *iommu; + int index; + + iommu = map_dev_to_ir(dev); + if (!iommu) { + printk(KERN_ERR + "Unable to map PCI %s to iommu\n", pci_name(dev)); + return -ENOENT; + } + + index = alloc_irte(iommu, irq, nvec); + if (index < 0) { + printk(KERN_ERR + "Unable to allocate %d IRTE for PCI %s\n", nvec, + pci_name(dev)); + return -ENOSPC; + } + return index; +} +#endif + +static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq) { + int ret; struct msi_msg msg; + + ret = msi_compose_msg(dev, irq, &msg); + if (ret < 0) + return ret; + + set_irq_msi(irq, desc); + write_msi_msg(irq, &msg); + +#ifdef CONFIG_INTR_REMAP + if (irq_remapped(irq)) { + struct irq_desc *desc = irq_desc + irq; + /* + * irq migration in process context + */ + desc->status |= IRQ_MOVE_PCNTXT; + set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge"); + } else +#endif + set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge"); + + return 0; +} + +int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc) +{ int irq, ret; + irq = create_irq(); if (irq < 0) return irq; - ret = msi_compose_msg(dev, irq, &msg); +#ifdef CONFIG_INTR_REMAP + if (!intr_remapping_enabled) + goto no_ir; + + ret = msi_alloc_irte(dev, irq, 1); + if (ret < 0) + goto error; +no_ir: +#endif + ret = setup_msi_irq(dev, desc, irq); if (ret < 0) { destroy_irq(irq); return ret; } + return 0; - set_irq_msi(irq, desc); - write_msi_msg(irq, &msg); +#ifdef CONFIG_INTR_REMAP +error: + destroy_irq(irq); + return ret; +#endif +} - set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge"); +int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) +{ + int irq, ret, sub_handle; + struct msi_desc *desc; +#ifdef CONFIG_INTR_REMAP + struct intel_iommu *iommu = 0; + int index = 0; +#endif + sub_handle = 0; + list_for_each_entry(desc, &dev->msi_list, list) { + irq = create_irq(); + if (irq < 0) + return irq; +#ifdef CONFIG_INTR_REMAP + if (!intr_remapping_enabled) + goto no_ir; + + if (!sub_handle) { + /* + * allocate the consecutive block of IRTE's + * for 'nvec' + */ + index = msi_alloc_irte(dev, irq, nvec); + if (index < 0) { + ret = index; + goto error; + } + } else { + iommu = map_dev_to_ir(dev); + if (!iommu) { + ret = -ENOENT; + goto error; + } + /* + * setup the mapping between the irq and the IRTE + * base index, the sub_handle pointing to the + * appropriate interrupt remap table entry. + */ + set_irte_irq(irq, iommu, index, sub_handle); + } +no_ir: +#endif + ret = setup_msi_irq(dev, desc, irq); + if (ret < 0) + goto error; + sub_handle++; + } return 0; + +error: + destroy_irq(irq); + return ret; } void arch_teardown_msi_irq(unsigned int irq) diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c index 32e55c7a980..bb642cc5e18 100644 --- a/drivers/pci/intr_remapping.c +++ b/drivers/pci/intr_remapping.c @@ -230,6 +230,17 @@ struct intel_iommu *map_ioapic_to_ir(int apic) return NULL; } +struct intel_iommu *map_dev_to_ir(struct pci_dev *dev) +{ + struct dmar_drhd_unit *drhd; + + drhd = dmar_find_matched_drhd_unit(dev); + if (!drhd) + return NULL; + + return drhd->iommu; +} + int free_irte(int irq) { int index, i; diff --git a/include/asm-x86/msidef.h b/include/asm-x86/msidef.h index 296f29ce426..57fd85935e5 100644 --- a/include/asm-x86/msidef.h +++ b/include/asm-x86/msidef.h @@ -48,4 +48,8 @@ #define MSI_ADDR_DEST_ID(dest) (((dest) << MSI_ADDR_DEST_ID_SHIFT) & \ MSI_ADDR_DEST_ID_MASK) +#define MSI_ADDR_IR_EXT_INT (1 << 4) +#define MSI_ADDR_IR_SHV (1 << 3) +#define MSI_ADDR_IR_INDEX1(index) ((index & 0x8000) >> 13) +#define MSI_ADDR_IR_INDEX2(index) ((index & 0x7fff) << 5) #endif /* ASM_MSIDEF_H */ diff --git a/include/linux/dmar.h b/include/linux/dmar.h index bf41ffa7470..c360c558e59 100644 --- a/include/linux/dmar.h +++ b/include/linux/dmar.h @@ -109,6 +109,7 @@ extern int flush_irte(int irq); extern int free_irte(int irq); extern int irq_remapped(int irq); +extern struct intel_iommu *map_dev_to_ir(struct pci_dev *dev); extern struct intel_iommu *map_ioapic_to_ir(int apic); #else #define irq_remapped(irq) (0) -- cgit v1.2.3-70-g09d2 From 6e1cb38a2aef7680975e71f23de187859ee8b158 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:58 -0700 Subject: x64, x2apic/intr-remap: add x2apic support, including enabling interrupt-remapping x2apic support. Interrupt-remapping must be enabled before enabling x2apic, this is needed to ensure that IO interrupts continue to work properly after the cpu mode is changed to x2apic(which uses 32bit extended physical/cluster apic id). On systems where apicid's are > 255, BIOS can handover the control to OS in x2apic mode. Or if the OS handover was in legacy xapic mode, check if it is capable of x2apic mode. And if we succeed in enabling Interrupt-remapping, then we can enable x2apic mode in the CPU. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- Documentation/kernel-parameters.txt | 2 + arch/x86/kernel/acpi/boot.c | 2 + arch/x86/kernel/apic_64.c | 154 +++++++++++++++++++++++++++++++++++- arch/x86/kernel/cpu/common_64.c | 2 + arch/x86/kernel/genapic_64.c | 1 + arch/x86/kernel/mpparse.c | 2 + arch/x86/kernel/setup.c | 2 + arch/x86/kernel/smpboot.c | 5 ++ include/asm-x86/apic.h | 5 ++ 9 files changed, 171 insertions(+), 4 deletions(-) (limited to 'include/asm-x86') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 795c487af8e..56689ef1a15 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1377,6 +1377,8 @@ and is between 256 and 4096 characters. It is defined in the file nolapic_timer [X86-32,APIC] Do not use the local APIC timer. + nox2apic [X86-64,APIC] Do not enable x2APIC mode. + noltlbs [PPC] Do not use large page/tlb entries for kernel lowmem mapping on PPC40x. diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 785700a08e9..8705262ddcd 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -1337,7 +1337,9 @@ static void __init acpi_process_madt(void) acpi_ioapic = 1; smp_found_config = 1; +#ifdef CONFIG_X86_32 setup_apic_routing(); +#endif } } if (error == -EINVAL) { diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c index d5c06917b5b..dd0501039f0 100644 --- a/arch/x86/kernel/apic_64.c +++ b/arch/x86/kernel/apic_64.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include @@ -39,6 +40,7 @@ #include #include #include +#include #include #include @@ -46,8 +48,12 @@ static int disable_apic_timer __cpuinitdata; static int apic_calibrate_pmtmr __initdata; int disable_apic; +int disable_x2apic; int x2apic; +/* x2apic enabled before OS handover */ +int x2apic_preenabled; + /* Local APIC timer works in C2 */ int local_apic_timer_c2_ok; EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok); @@ -896,6 +902,125 @@ void __cpuinit end_local_APIC_setup(void) apic_pm_activate(); } +void check_x2apic(void) +{ + int msr, msr2; + + rdmsr(MSR_IA32_APICBASE, msr, msr2); + + if (msr & X2APIC_ENABLE) { + printk("x2apic enabled by BIOS, switching to x2apic ops\n"); + x2apic_preenabled = x2apic = 1; + apic_ops = &x2apic_ops; + } +} + +void enable_x2apic(void) +{ + int msr, msr2; + + rdmsr(MSR_IA32_APICBASE, msr, msr2); + if (!(msr & X2APIC_ENABLE)) { + printk("Enabling x2apic\n"); + wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0); + } +} + +void enable_IR_x2apic(void) +{ +#ifdef CONFIG_INTR_REMAP + int ret; + unsigned long flags; + + if (!cpu_has_x2apic) + return; + + if (!x2apic_preenabled && disable_x2apic) { + printk(KERN_INFO + "Skipped enabling x2apic and Interrupt-remapping " + "because of nox2apic\n"); + return; + } + + if (x2apic_preenabled && disable_x2apic) + panic("Bios already enabled x2apic, can't enforce nox2apic"); + + if (!x2apic_preenabled && skip_ioapic_setup) { + printk(KERN_INFO + "Skipped enabling x2apic and Interrupt-remapping " + "because of skipping io-apic setup\n"); + return; + } + + ret = dmar_table_init(); + if (ret) { + printk(KERN_INFO + "dmar_table_init() failed with %d:\n", ret); + + if (x2apic_preenabled) + panic("x2apic enabled by bios. But IR enabling failed"); + else + printk(KERN_INFO + "Not enabling x2apic,Intr-remapping\n"); + return; + } + + local_irq_save(flags); + mask_8259A(); + save_mask_IO_APIC_setup(); + + ret = enable_intr_remapping(1); + + if (ret && x2apic_preenabled) { + local_irq_restore(flags); + panic("x2apic enabled by bios. But IR enabling failed"); + } + + if (ret) + goto end; + + if (!x2apic) { + x2apic = 1; + apic_ops = &x2apic_ops; + enable_x2apic(); + } +end: + if (ret) + /* + * IR enabling failed + */ + restore_IO_APIC_setup(); + else + reinit_intr_remapped_IO_APIC(x2apic_preenabled); + + unmask_8259A(); + local_irq_restore(flags); + + if (!ret) { + if (!x2apic_preenabled) + printk(KERN_INFO + "Enabled x2apic and interrupt-remapping\n"); + else + printk(KERN_INFO + "Enabled Interrupt-remapping\n"); + } else + printk(KERN_ERR + "Failed to enable Interrupt-remapping and x2apic\n"); +#else + if (!cpu_has_x2apic) + return; + + if (x2apic_preenabled) + panic("x2apic enabled prior OS handover," + " enable CONFIG_INTR_REMAP"); + + printk(KERN_INFO "Enable CONFIG_INTR_REMAP for enabling intr-remapping " + " and x2apic\n"); +#endif + + return; +} + /* * Detect and enable local APICs on non-SMP boards. * Original code written by Keir Fraser. @@ -943,6 +1068,11 @@ void __init early_init_lapic_mapping(void) */ void __init init_apic_mappings(void) { + if (x2apic) { + boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); + return; + } + /* * If no local APIC can be found then set up a fake all * zeroes page to simulate the local APIC and another @@ -981,6 +1111,9 @@ int __init APIC_init_uniprocessor(void) return -1; } + enable_IR_x2apic(); + setup_apic_routing(); + verify_local_APIC(); connect_bsp_APIC(); @@ -1238,10 +1371,14 @@ static int lapic_resume(struct sys_device *dev) maxlvt = lapic_get_maxlvt(); local_irq_save(flags); - rdmsr(MSR_IA32_APICBASE, l, h); - l &= ~MSR_IA32_APICBASE_BASE; - l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr; - wrmsr(MSR_IA32_APICBASE, l, h); + if (!x2apic) { + rdmsr(MSR_IA32_APICBASE, l, h); + l &= ~MSR_IA32_APICBASE_BASE; + l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr; + wrmsr(MSR_IA32_APICBASE, l, h); + } else + enable_x2apic(); + apic_write(APIC_LVTERR, ERROR_APIC_VECTOR | APIC_LVT_MASKED); apic_write(APIC_ID, apic_pm_state.apic_id); apic_write(APIC_DFR, apic_pm_state.apic_dfr); @@ -1381,6 +1518,15 @@ __cpuinit int apic_is_clustered_box(void) return (clusters > 2); } +static __init int setup_nox2apic(char *str) +{ + disable_x2apic = 1; + clear_cpu_cap(&boot_cpu_data, X86_FEATURE_X2APIC); + return 0; +} +early_param("nox2apic", setup_nox2apic); + + /* * APIC command line parameters */ diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c index 36537ab9e56..e7bf3c2dc5f 100644 --- a/arch/x86/kernel/cpu/common_64.c +++ b/arch/x86/kernel/cpu/common_64.c @@ -606,6 +606,8 @@ void __cpuinit cpu_init(void) barrier(); check_efer(); + if (cpu != 0 && x2apic) + enable_x2apic(); /* * set up and load the per-CPU TSS diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c index 1029e178cdf..792e21ba1a8 100644 --- a/arch/x86/kernel/genapic_64.c +++ b/arch/x86/kernel/genapic_64.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 3b25e49380c..70e1f3e287f 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -545,7 +545,9 @@ static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early) generic_bigsmp_probe(); #endif +#ifdef CONFIG_X86_32 setup_apic_routing(); +#endif if (!num_processors) printk(KERN_ERR "MPTABLE: no processors registered!\n"); return num_processors; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 987b6fde3a9..2e78a143dec 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -730,6 +730,8 @@ void __init setup_arch(char **cmdline_p) num_physpages = max_pfn; check_efer(); + if (cpu_has_x2apic) + check_x2apic(); /* How many end-of-memory variables you have, grandma! */ /* need this before calling reserve_initrd */ diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index c55263b3df0..0c43e1f2e7d 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1145,6 +1145,11 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) current_thread_info()->cpu = 0; /* needed? */ set_cpu_sibling_map(0); +#ifdef CONFIG_X86_64 + enable_IR_x2apic(); + setup_apic_routing(); +#endif + if (smp_sanity_check(max_cpus) < 0) { printk(KERN_INFO "SMP disabled\n"); disable_smp(); diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h index aa746704a5c..129752dd252 100644 --- a/include/asm-x86/apic.h +++ b/include/asm-x86/apic.h @@ -100,6 +100,11 @@ extern void apic_wait_icr_idle(void); extern u32 safe_apic_wait_icr_idle(void); extern void apic_icr_write(u32 low, u32 id); #else +extern int x2apic, x2apic_preenabled; +extern void check_x2apic(void); +extern void enable_x2apic(void); +extern void enable_IR_x2apic(void); +extern void x2apic_icr_write(u32 low, u32 id); struct apic_ops { u32 (*read)(u32 reg); -- cgit v1.2.3-70-g09d2 From 2d9579a124d746a3e0e0ba45e57d80800ee80807 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Thu, 10 Jul 2008 11:16:59 -0700 Subject: x64, x2apic/intr-remap: support for x2apic physical mode support x2apic Physical mode support. By default we will use x2apic cluster mode. x2apic physical mode can be selected using "x2apic_phys" boot parameter. Signed-off-by: Suresh Siddha Cc: akpm@linux-foundation.org Cc: arjan@linux.intel.com Cc: andi@firstfloor.org Cc: ebiederm@xmission.com Cc: jbarnes@virtuousgeek.org Cc: steiner@sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/Makefile | 1 + arch/x86/kernel/genapic_64.c | 18 +++++- arch/x86/kernel/genx2apic_phys.c | 122 +++++++++++++++++++++++++++++++++++++++ include/asm-x86/genapic_64.h | 1 + 4 files changed, 139 insertions(+), 3 deletions(-) create mode 100644 arch/x86/kernel/genx2apic_phys.c (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index bde3e9b6fec..81280e93e79 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -96,6 +96,7 @@ obj-$(CONFIG_OLPC) += olpc.o ifeq ($(CONFIG_X86_64),y) obj-y += genapic_64.o genapic_flat_64.o genx2apic_uv_x.o tlb_uv.o obj-y += genx2apic_cluster.o + obj-y += genx2apic_phys.o obj-$(CONFIG_X86_PM_TIMER) += pmtimer_64.o obj-$(CONFIG_AUDIT) += audit_64.o diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c index 792e21ba1a8..3940d8161f8 100644 --- a/arch/x86/kernel/genapic_64.c +++ b/arch/x86/kernel/genapic_64.c @@ -30,6 +30,15 @@ DEFINE_PER_CPU(int, x2apic_extra_bits); struct genapic __read_mostly *genapic = &apic_flat; +static int x2apic_phys = 0; + +static int set_x2apic_phys_mode(char *arg) +{ + x2apic_phys = 1; + return 0; +} +early_param("x2apic_phys", set_x2apic_phys_mode); + static enum uv_system_type uv_system_type; /* @@ -39,9 +48,12 @@ void __init setup_apic_routing(void) { if (uv_system_type == UV_NON_UNIQUE_APIC) genapic = &apic_x2apic_uv_x; - else if (cpu_has_x2apic && intr_remapping_enabled) - genapic = &apic_x2apic_cluster; - else + else if (cpu_has_x2apic && intr_remapping_enabled) { + if (x2apic_phys) + genapic = &apic_x2apic_phys; + else + genapic = &apic_x2apic_cluster; + } else #ifdef CONFIG_ACPI /* * Quirk: some x86_64 machines can only use physical APIC mode diff --git a/arch/x86/kernel/genx2apic_phys.c b/arch/x86/kernel/genx2apic_phys.c new file mode 100644 index 00000000000..3c70b9d692b --- /dev/null +++ b/arch/x86/kernel/genx2apic_phys.c @@ -0,0 +1,122 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ + +static cpumask_t x2apic_target_cpus(void) +{ + return cpumask_of_cpu(0); +} + +static cpumask_t x2apic_vector_allocation_domain(int cpu) +{ + cpumask_t domain = CPU_MASK_NONE; + cpu_set(cpu, domain); + return domain; +} + +static void __x2apic_send_IPI_dest(unsigned int apicid, int vector, + unsigned int dest) +{ + unsigned long cfg; + + cfg = __prepare_ICR(0, vector, dest); + + /* + * send the IPI. + */ + x2apic_icr_write(cfg, apicid); +} + +static void x2apic_send_IPI_mask(cpumask_t mask, int vector) +{ + unsigned long flags; + unsigned long query_cpu; + + local_irq_save(flags); + for_each_cpu_mask(query_cpu, mask) { + __x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu), + vector, APIC_DEST_PHYSICAL); + } + local_irq_restore(flags); +} + +static void x2apic_send_IPI_allbutself(int vector) +{ + cpumask_t mask = cpu_online_map; + + cpu_clear(smp_processor_id(), mask); + + if (!cpus_empty(mask)) + x2apic_send_IPI_mask(mask, vector); +} + +static void x2apic_send_IPI_all(int vector) +{ + x2apic_send_IPI_mask(cpu_online_map, vector); +} + +static int x2apic_apic_id_registered(void) +{ + return 1; +} + +static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + cpu = first_cpu(cpumask); + if ((unsigned)cpu < NR_CPUS) + return per_cpu(x86_cpu_to_apicid, cpu); + else + return BAD_APICID; +} + +static unsigned int x2apic_read_id(void) +{ + return apic_read(APIC_ID); +} + +static unsigned int phys_pkg_id(int index_msb) +{ + return x2apic_read_id() >> index_msb; +} + +void x2apic_send_IPI_self(int vector) +{ + apic_write(APIC_SELF_IPI, vector); +} + +void init_x2apic_ldr(void) +{ + return; +} + +struct genapic apic_x2apic_phys = { + .name = "physical x2apic", + .int_delivery_mode = dest_Fixed, + .int_dest_mode = (APIC_DEST_PHYSICAL != 0), + .target_cpus = x2apic_target_cpus, + .vector_allocation_domain = x2apic_vector_allocation_domain, + .apic_id_registered = x2apic_apic_id_registered, + .init_apic_ldr = init_x2apic_ldr, + .send_IPI_all = x2apic_send_IPI_all, + .send_IPI_allbutself = x2apic_send_IPI_allbutself, + .send_IPI_mask = x2apic_send_IPI_mask, + .send_IPI_self = x2apic_send_IPI_self, + .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, + .phys_pkg_id = phys_pkg_id, + .read_apic_id = x2apic_read_id, +}; diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h index 23246030587..122b9242a40 100644 --- a/include/asm-x86/genapic_64.h +++ b/include/asm-x86/genapic_64.h @@ -36,6 +36,7 @@ extern struct genapic *genapic; extern struct genapic apic_flat; extern struct genapic apic_physflat; extern struct genapic apic_x2apic_cluster; +extern struct genapic apic_x2apic_phys; extern int acpi_madt_oem_check(char *, char *); extern void apic_send_IPI_self(int vector); -- cgit v1.2.3-70-g09d2 From ad66dd340f561bdde2285992314d9e4fd9b6191e Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Fri, 11 Jul 2008 13:11:56 -0700 Subject: x2apic: xen64 paravirt basic apic ops Define the Xen specific basic apic ops, in additon to paravirt apic ops, with some misc warning fixes. Signed-off-by: Suresh Siddha Cc: Jeremy Fitzhardinge Cc: akpm@linux-foundation.org Signed-off-by: Ingo Molnar --- arch/x86/lguest/boot.c | 4 ++-- arch/x86/xen/enlighten.c | 41 +++++++++++++++++++++++++++++++++++++++-- include/asm-x86/paravirt.h | 14 ++++++++------ 3 files changed, 49 insertions(+), 10 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index 50dad44fb54..0c45df20e2b 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -783,11 +783,11 @@ static void lguest_wbinvd(void) * code qualifies for Advanced. It will also never interrupt anything. It * does, however, allow us to get through the Linux boot code. */ #ifdef CONFIG_X86_LOCAL_APIC -static void lguest_apic_write(unsigned long reg, u32 v) +static void lguest_apic_write(u32 reg, u32 v) { } -static u32 lguest_apic_read(unsigned long reg) +static u32 lguest_apic_read(u32 reg) { return 0; } diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index dcd4e51f2f1..54e25566753 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -548,16 +548,45 @@ static void xen_io_delay(void) } #ifdef CONFIG_X86_LOCAL_APIC -static u32 xen_apic_read(unsigned long reg) +static u32 xen_apic_read(u32 reg) { return 0; } -static void xen_apic_write(unsigned long reg, u32 val) +static void xen_apic_write(u32 reg, u32 val) { /* Warn to see if there's any stray references */ WARN_ON(1); } + +#ifdef CONFIG_X86_64 +static u64 xen_apic_icr_read(void) +{ + return 0; +} + +static void xen_apic_icr_write(u32 low, u32 id) +{ + /* Warn to see if there's any stray references */ + WARN_ON(1); +} + +static void xen_apic_wait_icr_idle(void) +{ + return; +} + +static struct apic_ops xen_basic_apic_ops = { + .read = xen_apic_read, + .write = xen_apic_write, + .write_atomic = xen_apic_write, + .icr_read = xen_apic_icr_read, + .icr_write = xen_apic_icr_write, + .wait_icr_idle = xen_apic_wait_icr_idle, + .safe_wait_icr_idle = xen_apic_wait_icr_idle, +}; +#endif + #endif static void xen_flush_tlb(void) @@ -1130,9 +1159,11 @@ static const struct pv_irq_ops xen_irq_ops __initdata = { static const struct pv_apic_ops xen_apic_ops __initdata = { #ifdef CONFIG_X86_LOCAL_APIC +#ifndef CONFIG_X86_64 .apic_write = xen_apic_write, .apic_write_atomic = xen_apic_write, .apic_read = xen_apic_read, +#endif .setup_boot_clock = paravirt_nop, .setup_secondary_clock = paravirt_nop, .startup_ipi_hook = paravirt_nop, @@ -1291,6 +1322,12 @@ asmlinkage void __init xen_start_kernel(void) pv_irq_ops = xen_irq_ops; pv_apic_ops = xen_apic_ops; pv_mmu_ops = xen_mmu_ops; +#ifdef CONFIG_X86_64 + /* + * for 64bit, set up the basic apic ops aswell. + */ + apic_ops = &xen_basic_apic_ops; +#endif if (xen_feature(XENFEAT_mmu_pt_update_preserve_ad)) { pv_mmu_ops.ptep_modify_prot_start = xen_ptep_modify_prot_start; diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index 10adac02e6d..5e34d26aa3b 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -200,13 +200,15 @@ struct pv_irq_ops { struct pv_apic_ops { #ifdef CONFIG_X86_LOCAL_APIC +#ifndef CONFIG_X86_64 /* * Direct APIC operations, principally for VMI. Ideally * these shouldn't be in this interface. */ - void (*apic_write)(unsigned long reg, u32 v); - void (*apic_write_atomic)(unsigned long reg, u32 v); - u32 (*apic_read)(unsigned long reg); + void (*apic_write)(u32 reg, u32 v); + void (*apic_write_atomic)(u32 reg, u32 v); + u32 (*apic_read)(u32 reg); +#endif void (*setup_boot_clock)(void); void (*setup_secondary_clock)(void); @@ -892,17 +894,17 @@ static inline void slow_down_io(void) * Basic functions accessing APICs. */ #ifndef CONFIG_X86_64 -static inline void apic_write(unsigned long reg, u32 v) +static inline void apic_write(u32 reg, u32 v) { PVOP_VCALL2(pv_apic_ops.apic_write, reg, v); } -static inline void apic_write_atomic(unsigned long reg, u32 v) +static inline void apic_write_atomic(u32 reg, u32 v) { PVOP_VCALL2(pv_apic_ops.apic_write_atomic, reg, v); } -static inline u32 apic_read(unsigned long reg) +static inline u32 apic_read(u32 reg) { return PVOP_CALL1(unsigned long, pv_apic_ops.apic_read, reg); } -- cgit v1.2.3-70-g09d2 From c535b6a1a685eb23f96e2c221777d6c1e05080d5 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 11 Jul 2008 18:41:54 -0700 Subject: x86: let 32bit use apic_ops too Signed-off-by: Yinghai Lu Cc: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic_32.c | 39 +++++++++++++++++++++++++++++++-------- include/asm-x86/apic.h | 13 ++----------- 2 files changed, 33 insertions(+), 19 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c index 2a83c07bd88..7413354c9ff 100644 --- a/arch/x86/kernel/apic_32.c +++ b/arch/x86/kernel/apic_32.c @@ -145,19 +145,13 @@ static int modern_apic(void) return lapic_get_version() >= 0x14; } -void apic_icr_write(u32 low, u32 id) -{ - apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(id)); - apic_write_around(APIC_ICR, low); -} - -void apic_wait_icr_idle(void) +void xapic_wait_icr_idle(void) { while (apic_read(APIC_ICR) & APIC_ICR_BUSY) cpu_relax(); } -u32 safe_apic_wait_icr_idle(void) +u32 safe_xapic_wait_icr_idle(void) { u32 send_status; int timeout; @@ -173,6 +167,35 @@ u32 safe_apic_wait_icr_idle(void) return send_status; } +void xapic_icr_write(u32 low, u32 id) +{ + apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(id)); + apic_write_around(APIC_ICR, low); +} + +u64 xapic_icr_read(void) +{ + u32 icr1, icr2; + + icr2 = apic_read(APIC_ICR2); + icr1 = apic_read(APIC_ICR); + + return icr1 | ((u64)icr2 << 32); +} + +static struct apic_ops xapic_ops = { + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .write_atomic = native_apic_mem_write_atomic, + .icr_read = xapic_icr_read, + .icr_write = xapic_icr_write, + .wait_icr_idle = xapic_wait_icr_idle, + .safe_wait_icr_idle = safe_xapic_wait_icr_idle, +}; + +struct apic_ops __read_mostly *apic_ops = &xapic_ops; +EXPORT_SYMBOL_GPL(apic_ops); + /** * enable_NMI_through_LVT0 - enable NMI through local vector table 0 */ diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h index 129752dd252..fcd2f01277b 100644 --- a/include/asm-x86/apic.h +++ b/include/asm-x86/apic.h @@ -49,11 +49,6 @@ extern int disable_apic; #ifdef CONFIG_PARAVIRT #include #else -#ifndef CONFIG_X86_64 -#define apic_write native_apic_mem_write -#define apic_write_atomic native_apic_mem_write_atomic -#define apic_read native_apic_mem_read -#endif #define setup_boot_clock setup_boot_APIC_clock #define setup_secondary_clock setup_secondary_APIC_clock #endif @@ -95,16 +90,13 @@ static inline u32 native_apic_msr_read(u32 reg) return low; } -#ifdef CONFIG_X86_32 -extern void apic_wait_icr_idle(void); -extern u32 safe_apic_wait_icr_idle(void); -extern void apic_icr_write(u32 low, u32 id); -#else +#ifndef CONFIG_X86_32 extern int x2apic, x2apic_preenabled; extern void check_x2apic(void); extern void enable_x2apic(void); extern void enable_IR_x2apic(void); extern void x2apic_icr_write(u32 low, u32 id); +#endif struct apic_ops { u32 (*read)(u32 reg); @@ -125,7 +117,6 @@ extern struct apic_ops *apic_ops; #define apic_icr_write (apic_ops->icr_write) #define apic_wait_icr_idle (apic_ops->wait_icr_idle) #define safe_apic_wait_icr_idle (apic_ops->safe_wait_icr_idle) -#endif extern int get_physical_broadcast(void); -- cgit v1.2.3-70-g09d2 From 4c9961d56ec20c27ec5d02e49fd7427748312741 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 11 Jul 2008 18:44:16 -0700 Subject: x86: make read_apic_id return final apicid also remove GET_APIC_ID when read_apic_id is used. need to apply after [PATCH] x86: mach_apicdef.h need to include before smp.h Signed-off-by: Yinghai Lu Cc: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/kernel/acpi/boot.c | 2 +- arch/x86/kernel/apic_32.c | 4 ++-- arch/x86/kernel/apic_64.c | 6 +++--- arch/x86/kernel/genapic_flat_64.c | 2 +- arch/x86/kernel/io_apic_32.c | 5 ++--- arch/x86/kernel/io_apic_64.c | 4 ++-- arch/x86/kernel/smpboot.c | 6 +++--- include/asm-x86/mach-default/mach_apic.h | 2 +- include/asm-x86/mach-default/mach_apicdef.h | 3 +-- include/asm-x86/mach-es7000/mach_apic.h | 2 +- include/asm-x86/smp.h | 11 ++++++++--- 11 files changed, 25 insertions(+), 22 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 8705262ddcd..b314bcd0840 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -761,7 +761,7 @@ static void __init acpi_register_lapic_address(unsigned long address) set_fixmap_nocache(FIX_APIC_BASE, address); if (boot_cpu_physical_apicid == -1U) { - boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); + boot_cpu_physical_apicid = read_apic_id(); #ifdef CONFIG_X86_32 apic_version[boot_cpu_physical_apicid] = GET_APIC_VERSION(apic_read(APIC_LVR)); diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c index 7413354c9ff..47ff978aecf 100644 --- a/arch/x86/kernel/apic_32.c +++ b/arch/x86/kernel/apic_32.c @@ -1230,7 +1230,7 @@ void __init init_apic_mappings(void) * default configuration (or the MP table is broken). */ if (boot_cpu_physical_apicid == -1U) - boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); + boot_cpu_physical_apicid = read_apic_id(); } @@ -1270,7 +1270,7 @@ int __init APIC_init_uniprocessor(void) * might be zero if read from MP tables. Get it from LAPIC. */ #ifdef CONFIG_CRASH_DUMP - boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); + boot_cpu_physical_apicid = read_apic_id(); #endif physid_set_mask_of_physid(boot_cpu_physical_apicid, &phys_cpu_present_map); diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c index dd0501039f0..c75f58a66d8 100644 --- a/arch/x86/kernel/apic_64.c +++ b/arch/x86/kernel/apic_64.c @@ -1060,7 +1060,7 @@ void __init early_init_lapic_mapping(void) * Fetch the APIC ID of the BSP in case we have a * default configuration (or the MP table is broken). */ - boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); + boot_cpu_physical_apicid = read_apic_id(); } /** @@ -1069,7 +1069,7 @@ void __init early_init_lapic_mapping(void) void __init init_apic_mappings(void) { if (x2apic) { - boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); + boot_cpu_physical_apicid = read_apic_id(); return; } @@ -1092,7 +1092,7 @@ void __init init_apic_mappings(void) * Fetch the APIC ID of the BSP in case we have a * default configuration (or the MP table is broken). */ - boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); + boot_cpu_physical_apicid = read_apic_id(); } /* diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c index 73558682213..7dac2f275fa 100644 --- a/arch/x86/kernel/genapic_flat_64.c +++ b/arch/x86/kernel/genapic_flat_64.c @@ -101,7 +101,7 @@ static unsigned int read_xapic_id(void) { unsigned int id; - id = GET_XAPIC_ID(apic_read(APIC_ID)); + id = GET_APIC_ID(apic_read(APIC_ID)); return id; } diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c index c50adb84ea6..382208d11f8 100644 --- a/arch/x86/kernel/io_apic_32.c +++ b/arch/x86/kernel/io_apic_32.c @@ -1501,7 +1501,7 @@ void /*__init*/ print_local_APIC(void *dummy) smp_processor_id(), hard_smp_processor_id()); v = apic_read(APIC_ID); printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, - GET_APIC_ID(read_apic_id())); + GET_APIC_ID(v)); v = apic_read(APIC_LVR); printk(KERN_INFO "... APIC VERSION: %08x\n", v); ver = GET_APIC_VERSION(v); @@ -1709,8 +1709,7 @@ void disable_IO_APIC(void) entry.dest_mode = 0; /* Physical */ entry.delivery_mode = dest_ExtINT; /* ExtInt */ entry.vector = 0; - entry.dest.physical.physical_dest = - GET_APIC_ID(read_apic_id()); + entry.dest.physical.physical_dest = read_apic_id(); /* * Add it to the IO-APIC irq-routing table: diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index 877aa2e9d7e..2db0f98e2af 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic_64.c @@ -1246,7 +1246,7 @@ void __apicdebuginit print_local_APIC(void * dummy) printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n", smp_processor_id(), hard_smp_processor_id()); v = apic_read(APIC_ID); - printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, GET_APIC_ID(read_apic_id())); + printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, read_apic_id()); v = apic_read(APIC_LVR); printk(KERN_INFO "... APIC VERSION: %08x\n", v); ver = GET_APIC_VERSION(v); @@ -1439,7 +1439,7 @@ void disable_IO_APIC(void) entry.dest_mode = 0; /* Physical */ entry.delivery_mode = dest_ExtINT; /* ExtInt */ entry.vector = 0; - entry.dest = GET_APIC_ID(read_apic_id()); + entry.dest = read_apic_id(); /* * Add it to the IO-APIC irq-routing table: diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 0c43e1f2e7d..6cd002f3e20 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -211,7 +211,7 @@ static void __cpuinit smp_callin(void) /* * (This works even if the APIC is not enabled.) */ - phys_id = GET_APIC_ID(read_apic_id()); + phys_id = read_apic_id(); cpuid = smp_processor_id(); if (cpu_isset(cpuid, cpu_callin_map)) { panic("%s: phys CPU#%d, CPU#%d already present??\n", __func__, @@ -1157,9 +1157,9 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) } preempt_disable(); - if (GET_APIC_ID(read_apic_id()) != boot_cpu_physical_apicid) { + if (read_apic_id() != boot_cpu_physical_apicid) { panic("Boot APIC ID in local APIC unexpected (%d vs %d)", - GET_APIC_ID(read_apic_id()), boot_cpu_physical_apicid); + read_apic_id(), boot_cpu_physical_apicid); /* Or can we switch back to PIC here? */ } preempt_enable(); diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h index e06d23975d6..925b797e2a2 100644 --- a/include/asm-x86/mach-default/mach_apic.h +++ b/include/asm-x86/mach-default/mach_apic.h @@ -56,7 +56,7 @@ static inline void init_apic_ldr(void) static inline int apic_id_registered(void) { - return physid_isset(GET_APIC_ID(read_apic_id()), phys_cpu_present_map); + return physid_isset(read_apic_id(), phys_cpu_present_map); } static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) diff --git a/include/asm-x86/mach-default/mach_apicdef.h b/include/asm-x86/mach-default/mach_apicdef.h index 453b58a67e2..3e1be6c99b3 100644 --- a/include/asm-x86/mach-default/mach_apicdef.h +++ b/include/asm-x86/mach-default/mach_apicdef.h @@ -5,9 +5,8 @@ #ifdef CONFIG_X86_64 #define APIC_ID_MASK (0xFFu<<24) -#define GET_APIC_ID(x) (x) +#define GET_APIC_ID(x) (((x)>>24) & 0xFFu) #define SET_APIC_ID(x) (((x)<<24)) -#define GET_XAPIC_ID(x) (((x) >> 24) & 0xFFu) #else #define APIC_ID_MASK (0xF<<24) static inline unsigned get_apic_id(unsigned long x) diff --git a/include/asm-x86/mach-es7000/mach_apic.h b/include/asm-x86/mach-es7000/mach_apic.h index fbc8ad256f5..b3556ec3bca 100644 --- a/include/asm-x86/mach-es7000/mach_apic.h +++ b/include/asm-x86/mach-es7000/mach_apic.h @@ -141,7 +141,7 @@ static inline void setup_portio_remap(void) extern unsigned int boot_cpu_physical_apicid; static inline int check_phys_apicid_present(int cpu_physical_apicid) { - boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id()); + boot_cpu_physical_apicid = read_apic_id(); return (1); } diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index d9d007d2278..3b43ca202c3 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -165,9 +165,14 @@ static inline int logical_smp_processor_id(void) return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR)); } +#include static inline unsigned int read_apic_id(void) { - return *(u32 *)(APIC_BASE + APIC_ID); + unsigned int reg; + + reg = *(u32 *)(APIC_BASE + APIC_ID); + + return GET_APIC_ID(reg); } #endif @@ -175,11 +180,11 @@ static inline unsigned int read_apic_id(void) # if defined(APIC_DEFINITION) || defined(CONFIG_X86_64) extern int hard_smp_processor_id(void); # else -# include +#include static inline int hard_smp_processor_id(void) { /* we don't want to mark this access volatile - bad code generation */ - return GET_APIC_ID(read_apic_id()); + return read_apic_id(); } # endif /* APIC_DEFINITION */ -- cgit v1.2.3-70-g09d2 From f910a9dc7c865896815e2a95fe33363e9522f277 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sat, 12 Jul 2008 01:01:20 -0700 Subject: x86: make 64bit have get_apic_id generalize the x2apic code some more. let read_apic_id become a macro (later on a function/inline) GET_APIC_ID(apic_read(APIC_ID)) +#define read_apic_id() (GET_APIC_ID(apic_read(APIC_ID))) instead of this weird construct: -#define read_apic_id (genapic->read_apic_id) Signed-off-by: Yinghai Lu Cc: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/kernel/genapic_flat_64.c | 26 +++++++++++++++++++++++--- arch/x86/kernel/genx2apic_cluster.c | 20 +++++++++++++++++++- arch/x86/kernel/genx2apic_phys.c | 20 +++++++++++++++++++- arch/x86/kernel/genx2apic_uv_x.c | 23 ++++++++++++++++++++--- include/asm-x86/genapic_64.h | 4 +++- include/asm-x86/mach-default/mach_apic.h | 2 +- include/asm-x86/mach-default/mach_apicdef.h | 6 +++--- 7 files changed, 88 insertions(+), 13 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c index 7dac2f275fa..2c973cbf054 100644 --- a/arch/x86/kernel/genapic_flat_64.c +++ b/arch/x86/kernel/genapic_flat_64.c @@ -97,11 +97,27 @@ static void flat_send_IPI_all(int vector) __send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL); } +static unsigned int get_apic_id(unsigned long x) +{ + unsigned int id; + + id = (((x)>>24) & 0xFFu); + return id; +} + +static unsigned long set_apic_id(unsigned int id) +{ + unsigned long x; + + x = ((id & 0xFFu)<<24); + return x; +} + static unsigned int read_xapic_id(void) { unsigned int id; - id = GET_APIC_ID(apic_read(APIC_ID)); + id = get_apic_id(apic_read(APIC_ID)); return id; } @@ -134,7 +150,9 @@ struct genapic apic_flat = { .send_IPI_self = apic_send_IPI_self, .cpu_mask_to_apicid = flat_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, - .read_apic_id = read_xapic_id, + .get_apic_id = get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = (0xFFu<<24), }; /* @@ -200,5 +218,7 @@ struct genapic apic_physflat = { .send_IPI_self = apic_send_IPI_self, .cpu_mask_to_apicid = physflat_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, - .read_apic_id = read_xapic_id, + .get_apic_id = get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = (0xFFu<<24), }; diff --git a/arch/x86/kernel/genx2apic_cluster.c b/arch/x86/kernel/genx2apic_cluster.c index ed0fdede800..40bc0140d89 100644 --- a/arch/x86/kernel/genx2apic_cluster.c +++ b/arch/x86/kernel/genx2apic_cluster.c @@ -94,6 +94,22 @@ static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask) return BAD_APICID; } +static unsigned int get_apic_id(unsigned long x) +{ + unsigned int id; + + id = x; + return id; +} + +static unsigned long set_apic_id(unsigned int id) +{ + unsigned long x; + + x = id; + return x; +} + static unsigned int x2apic_read_id(void) { return apic_read(APIC_ID); @@ -131,5 +147,7 @@ struct genapic apic_x2apic_cluster = { .send_IPI_self = x2apic_send_IPI_self, .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, - .read_apic_id = x2apic_read_id, + .get_apic_id = get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = (0xFFFFFFFFu), }; diff --git a/arch/x86/kernel/genx2apic_phys.c b/arch/x86/kernel/genx2apic_phys.c index 3c70b9d692b..2f3c6ca19de 100644 --- a/arch/x86/kernel/genx2apic_phys.c +++ b/arch/x86/kernel/genx2apic_phys.c @@ -84,6 +84,22 @@ static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask) return BAD_APICID; } +static unsigned int get_apic_id(unsigned long x) +{ + unsigned int id; + + id = x; + return id; +} + +static unsigned long set_apic_id(unsigned int id) +{ + unsigned long x; + + x = id; + return x; +} + static unsigned int x2apic_read_id(void) { return apic_read(APIC_ID); @@ -118,5 +134,7 @@ struct genapic apic_x2apic_phys = { .send_IPI_self = x2apic_send_IPI_self, .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, - .read_apic_id = x2apic_read_id, + .get_apic_id = get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = (0xFFFFFFFFu), }; diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c index c915f750241..3ca29cd8c23 100644 --- a/arch/x86/kernel/genx2apic_uv_x.c +++ b/arch/x86/kernel/genx2apic_uv_x.c @@ -139,16 +139,31 @@ static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask) return BAD_APICID; } -static unsigned int uv_read_apic_id(void) +static unsigned int get_apic_id(unsigned long x) { unsigned int id; WARN_ON(preemptible() && num_online_cpus() > 1); - id = apic_read(APIC_ID) | __get_cpu_var(x2apic_extra_bits); + id = x | __get_cpu_var(x2apic_extra_bits); return id; } +static long set_apic_id(unsigned int id) +{ + unsigned long x; + + /* maskout x2apic_extra_bits ? */ + x = id; + return x; +} + +static unsigned int uv_read_apic_id(void) +{ + + return get_apic_id(apic_read(APIC_ID)); +} + static unsigned int phys_pkg_id(int index_msb) { return uv_read_apic_id() >> index_msb; @@ -175,7 +190,9 @@ struct genapic apic_x2apic_uv_x = { /* ZZZ.send_IPI_self = uv_send_IPI_self, */ .cpu_mask_to_apicid = uv_cpu_mask_to_apicid, .phys_pkg_id = phys_pkg_id, /* Fixme ZZZ */ - .read_apic_id = uv_read_apic_id, + .get_apic_id = get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = (0xFFFFFFFFu), }; static __cpuinit void set_x2apic_extra_bits(int pnode) diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h index 122b9242a40..8ff2589da93 100644 --- a/include/asm-x86/genapic_64.h +++ b/include/asm-x86/genapic_64.h @@ -28,7 +28,9 @@ struct genapic { /* */ unsigned int (*cpu_mask_to_apicid)(cpumask_t cpumask); unsigned int (*phys_pkg_id)(int index_msb); - unsigned int (*read_apic_id)(void); + unsigned int (*get_apic_id)(unsigned long x); + unsigned long (*set_apic_id)(unsigned int id); + unsigned long apic_id_mask; }; extern struct genapic *genapic; diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h index 925b797e2a2..3d2b455581e 100644 --- a/include/asm-x86/mach-default/mach_apic.h +++ b/include/asm-x86/mach-default/mach_apic.h @@ -30,7 +30,7 @@ static inline cpumask_t target_cpus(void) #define cpu_mask_to_apicid (genapic->cpu_mask_to_apicid) #define phys_pkg_id (genapic->phys_pkg_id) #define vector_allocation_domain (genapic->vector_allocation_domain) -#define read_apic_id (genapic->read_apic_id) +#define read_apic_id() (GET_APIC_ID(apic_read(APIC_ID))) #define send_IPI_self (genapic->send_IPI_self) extern void setup_apic_routing(void); #else diff --git a/include/asm-x86/mach-default/mach_apicdef.h b/include/asm-x86/mach-default/mach_apicdef.h index 3e1be6c99b3..a55518aa5a2 100644 --- a/include/asm-x86/mach-default/mach_apicdef.h +++ b/include/asm-x86/mach-default/mach_apicdef.h @@ -4,9 +4,9 @@ #include #ifdef CONFIG_X86_64 -#define APIC_ID_MASK (0xFFu<<24) -#define GET_APIC_ID(x) (((x)>>24) & 0xFFu) -#define SET_APIC_ID(x) (((x)<<24)) +#define APIC_ID_MASK (genapic->apic_id_mask) +#define GET_APIC_ID(x) (genapic->get_apic_id(x)) +#define SET_APIC_ID(x) (genapic->set_apic_id(x)) #else #define APIC_ID_MASK (0xF<<24) static inline unsigned get_apic_id(unsigned long x) -- cgit v1.2.3-70-g09d2 From 94a8c3c2437c8946f1b6c8e0b2c560a7db8ed3c6 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 13 Jul 2008 22:19:35 -0700 Subject: x86: let 32bit use apic_ops too - fix fix for pv - clean up the namespace there too. Signed-off-by: Yinghai Lu Cc: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/kernel/paravirt.c | 5 ----- arch/x86/kernel/vmi_32.c | 51 +++++++++++++++++++++++++++++++++++++++++++--- arch/x86/xen/enlighten.c | 19 ++++++++--------- include/asm-x86/paravirt.h | 29 -------------------------- 4 files changed, 57 insertions(+), 47 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 4f29ff847eb..e0f139106c7 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -360,11 +360,6 @@ struct pv_cpu_ops pv_cpu_ops = { struct pv_apic_ops pv_apic_ops = { #ifdef CONFIG_X86_LOCAL_APIC -#ifndef CONFIG_X86_64 - .apic_write = native_apic_mem_write, - .apic_write_atomic = native_apic_mem_write_atomic, - .apic_read = native_apic_mem_read, -#endif .setup_boot_clock = setup_boot_APIC_clock, .setup_secondary_clock = setup_secondary_APIC_clock, .startup_ipi_hook = paravirt_nop, diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index b15346092b7..cf307435455 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -676,6 +676,50 @@ static inline int __init probe_vmi_rom(void) return 0; } +#ifdef CONFIG_X86_LOCAL_APIC +static u32 vmi_apic_read(u32 reg) +{ + return 0; +} + +static void vmi_apic_write(u32 reg, u32 val) +{ + /* Warn to see if there's any stray references */ + WARN_ON(1); +} + +static u64 vmi_apic_icr_read(void) +{ + return 0; +} + +static void vmi_apic_icr_write(u32 low, u32 id) +{ + /* Warn to see if there's any stray references */ + WARN_ON(1); +} + +static void vmi_apic_wait_icr_idle(void) +{ + return; +} + +static u32 vmi_safe_apic_wait_icr_idle(void) +{ + return 0; +} + +static struct apic_ops vmi_basic_apic_ops = { + .read = vmi_apic_read, + .write = vmi_apic_write, + .write_atomic = vmi_apic_write, + .icr_read = vmi_apic_icr_read, + .icr_write = vmi_apic_icr_write, + .wait_icr_idle = vmi_apic_wait_icr_idle, + .safe_wait_icr_idle = vmi_safe_apic_wait_icr_idle, +}; +#endif + /* * VMI setup common to all processors */ @@ -904,9 +948,10 @@ static inline int __init activate_vmi(void) #endif #ifdef CONFIG_X86_LOCAL_APIC - para_fill(pv_apic_ops.apic_read, APICRead); - para_fill(pv_apic_ops.apic_write, APICWrite); - para_fill(pv_apic_ops.apic_write_atomic, APICWrite); + para_fill(vmi_basic_apic_ops.read, APICRead); + para_fill(vmi_basic_apic_ops.write, APICWrite); + para_fill(vmi_basic_apic_ops.write_atomic, APICWrite); + apic_ops = &vmi_basic_apic_ops; #endif /* diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 54e25566753..d11dda7ebd7 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -559,7 +559,6 @@ static void xen_apic_write(u32 reg, u32 val) WARN_ON(1); } -#ifdef CONFIG_X86_64 static u64 xen_apic_icr_read(void) { return 0; @@ -576,6 +575,11 @@ static void xen_apic_wait_icr_idle(void) return; } +static u32 xen_safe_apic_wait_icr_idle(void) +{ + return 0; +} + static struct apic_ops xen_basic_apic_ops = { .read = xen_apic_read, .write = xen_apic_write, @@ -583,9 +587,8 @@ static struct apic_ops xen_basic_apic_ops = { .icr_read = xen_apic_icr_read, .icr_write = xen_apic_icr_write, .wait_icr_idle = xen_apic_wait_icr_idle, - .safe_wait_icr_idle = xen_apic_wait_icr_idle, + .safe_wait_icr_idle = xen_safe_apic_wait_icr_idle, }; -#endif #endif @@ -1159,11 +1162,6 @@ static const struct pv_irq_ops xen_irq_ops __initdata = { static const struct pv_apic_ops xen_apic_ops __initdata = { #ifdef CONFIG_X86_LOCAL_APIC -#ifndef CONFIG_X86_64 - .apic_write = xen_apic_write, - .apic_write_atomic = xen_apic_write, - .apic_read = xen_apic_read, -#endif .setup_boot_clock = paravirt_nop, .setup_secondary_clock = paravirt_nop, .startup_ipi_hook = paravirt_nop, @@ -1322,9 +1320,10 @@ asmlinkage void __init xen_start_kernel(void) pv_irq_ops = xen_irq_ops; pv_apic_ops = xen_apic_ops; pv_mmu_ops = xen_mmu_ops; -#ifdef CONFIG_X86_64 + +#ifdef CONFIG_X86_LOCAL_APIC /* - * for 64bit, set up the basic apic ops aswell. + * set up the basic apic ops. */ apic_ops = &xen_basic_apic_ops; #endif diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index 5e34d26aa3b..08f89e385a9 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -200,15 +200,6 @@ struct pv_irq_ops { struct pv_apic_ops { #ifdef CONFIG_X86_LOCAL_APIC -#ifndef CONFIG_X86_64 - /* - * Direct APIC operations, principally for VMI. Ideally - * these shouldn't be in this interface. - */ - void (*apic_write)(u32 reg, u32 v); - void (*apic_write_atomic)(u32 reg, u32 v); - u32 (*apic_read)(u32 reg); -#endif void (*setup_boot_clock)(void); void (*setup_secondary_clock)(void); @@ -890,26 +881,6 @@ static inline void slow_down_io(void) } #ifdef CONFIG_X86_LOCAL_APIC -/* - * Basic functions accessing APICs. - */ -#ifndef CONFIG_X86_64 -static inline void apic_write(u32 reg, u32 v) -{ - PVOP_VCALL2(pv_apic_ops.apic_write, reg, v); -} - -static inline void apic_write_atomic(u32 reg, u32 v) -{ - PVOP_VCALL2(pv_apic_ops.apic_write_atomic, reg, v); -} - -static inline u32 apic_read(u32 reg) -{ - return PVOP_CALL1(unsigned long, pv_apic_ops.apic_read, reg); -} -#endif - static inline void setup_boot_clock(void) { PVOP_VCALL0(pv_apic_ops.setup_boot_clock); -- cgit v1.2.3-70-g09d2 From 1b9b89e7f163336ad84200b66a17284dbf26aced Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Mon, 21 Jul 2008 22:08:21 -0700 Subject: x86: add apic probe for genapic 64bit, v2 introducing an APIC handling probing abstraction: static struct genapic *apic_probe[] __initdata = { &apic_x2apic_uv_x, &apic_x2apic_phys, &apic_x2apic_cluster, &apic_physflat, NULL, }; This way we can remove UV, x2apic specific code from genapic_64.c and move them to their specific genapic files. [ v2: fix compiling when CONFIG_ACPI is not set ] Signed-off-by: Yinghai Lu Cc: Jack Steiner Cc: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/kernel/genapic_64.c | 85 ++++++++++++------------------------- arch/x86/kernel/genapic_flat_64.c | 26 ++++++++++++ arch/x86/kernel/genx2apic_cluster.c | 11 +++++ arch/x86/kernel/genx2apic_phys.c | 21 +++++++++ arch/x86/kernel/genx2apic_uv_x.c | 32 +++++++++++++- include/asm-x86/genapic_64.h | 1 + 6 files changed, 116 insertions(+), 60 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c index 3940d8161f8..b3ba969c50d 100644 --- a/arch/x86/kernel/genapic_64.c +++ b/arch/x86/kernel/genapic_64.c @@ -16,62 +16,37 @@ #include #include #include -#include #include #include #include -#ifdef CONFIG_ACPI -#include -#endif - -DEFINE_PER_CPU(int, x2apic_extra_bits); +extern struct genapic apic_flat; +extern struct genapic apic_physflat; +extern struct genapic apic_x2xpic_uv_x; +extern struct genapic apic_x2apic_phys; +extern struct genapic apic_x2apic_cluster; struct genapic __read_mostly *genapic = &apic_flat; -static int x2apic_phys = 0; - -static int set_x2apic_phys_mode(char *arg) -{ - x2apic_phys = 1; - return 0; -} -early_param("x2apic_phys", set_x2apic_phys_mode); - -static enum uv_system_type uv_system_type; +static struct genapic *apic_probe[] __initdata = { + &apic_x2apic_uv_x, + &apic_x2apic_phys, + &apic_x2apic_cluster, + &apic_physflat, + NULL, +}; /* * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode. */ void __init setup_apic_routing(void) { - if (uv_system_type == UV_NON_UNIQUE_APIC) - genapic = &apic_x2apic_uv_x; - else if (cpu_has_x2apic && intr_remapping_enabled) { - if (x2apic_phys) - genapic = &apic_x2apic_phys; - else - genapic = &apic_x2apic_cluster; - } else -#ifdef CONFIG_ACPI - /* - * Quirk: some x86_64 machines can only use physical APIC mode - * regardless of how many processors are present (x86_64 ES7000 - * is an example). - */ - if (acpi_gbl_FADT.header.revision > FADT2_REVISION_ID && - (acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL)) - genapic = &apic_physflat; - else -#endif - - if (max_physical_apicid < 8) - genapic = &apic_flat; - else - genapic = &apic_physflat; - - printk(KERN_INFO "Setting APIC routing to %s\n", genapic->name); + if (genapic == &apic_flat) { + if (max_physical_apicid >= 8) + genapic = &apic_physflat; + printk(KERN_INFO "Setting APIC routing to %s\n", genapic->name); + } } /* Same for both flat and physical. */ @@ -83,23 +58,15 @@ void apic_send_IPI_self(int vector) int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) { - if (!strcmp(oem_id, "SGI")) { - if (!strcmp(oem_table_id, "UVL")) - uv_system_type = UV_LEGACY_APIC; - else if (!strcmp(oem_table_id, "UVX")) - uv_system_type = UV_X2APIC; - else if (!strcmp(oem_table_id, "UVH")) - uv_system_type = UV_NON_UNIQUE_APIC; + int i; + + for (i = 0; apic_probe[i]; ++i) { + if (apic_probe[i]->acpi_madt_oem_check(oem_id, oem_table_id)) { + genapic = apic_probe[i]; + printk(KERN_INFO "Setting APIC routing to %s.\n", + genapic->name); + return 1; + } } return 0; } - -enum uv_system_type get_uv_system_type(void) -{ - return uv_system_type; -} - -int is_uv_system(void) -{ - return uv_system_type != UV_NONE; -} diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c index 2c973cbf054..1740b83329f 100644 --- a/arch/x86/kernel/genapic_flat_64.c +++ b/arch/x86/kernel/genapic_flat_64.c @@ -21,6 +21,15 @@ #include #include +#ifdef CONFIG_ACPI +#include +#endif + +static int __init flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + return 1; +} + static cpumask_t flat_target_cpus(void) { return cpu_online_map; @@ -138,6 +147,7 @@ static unsigned int phys_pkg_id(int index_msb) struct genapic apic_flat = { .name = "flat", + .acpi_madt_oem_check = flat_acpi_madt_oem_check, .int_delivery_mode = dest_LowestPrio, .int_dest_mode = (APIC_DEST_LOGICAL != 0), .target_cpus = flat_target_cpus, @@ -160,6 +170,21 @@ struct genapic apic_flat = { * We cannot use logical delivery in this case because the mask * overflows, so use physical mode. */ +static int __init physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ +#ifdef CONFIG_ACPI + /* + * Quirk: some x86_64 machines can only use physical APIC mode + * regardless of how many processors are present (x86_64 ES7000 + * is an example). + */ + if (acpi_gbl_FADT.header.revision > FADT2_REVISION_ID && + (acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL)) + return 1; +#endif + + return 0; +} static cpumask_t physflat_target_cpus(void) { @@ -206,6 +231,7 @@ static unsigned int physflat_cpu_mask_to_apicid(cpumask_t cpumask) struct genapic apic_physflat = { .name = "physical flat", + .acpi_madt_oem_check = physflat_acpi_madt_oem_check, .int_delivery_mode = dest_Fixed, .int_dest_mode = (APIC_DEST_PHYSICAL != 0), .target_cpus = physflat_target_cpus, diff --git a/arch/x86/kernel/genx2apic_cluster.c b/arch/x86/kernel/genx2apic_cluster.c index 40bc0140d89..ef3f3182d50 100644 --- a/arch/x86/kernel/genx2apic_cluster.c +++ b/arch/x86/kernel/genx2apic_cluster.c @@ -4,12 +4,22 @@ #include #include #include +#include + #include #include #include DEFINE_PER_CPU(u32, x86_cpu_to_logical_apicid); +static int __init x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + if (cpu_has_x2apic && intr_remapping_enabled) + return 1; + + return 0; +} + /* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ static cpumask_t x2apic_target_cpus(void) @@ -135,6 +145,7 @@ static void init_x2apic_ldr(void) struct genapic apic_x2apic_cluster = { .name = "cluster x2apic", + .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, .int_delivery_mode = dest_LowestPrio, .int_dest_mode = (APIC_DEST_LOGICAL != 0), .target_cpus = x2apic_target_cpus, diff --git a/arch/x86/kernel/genx2apic_phys.c b/arch/x86/kernel/genx2apic_phys.c index 2f3c6ca19de..f35a3bf3a9e 100644 --- a/arch/x86/kernel/genx2apic_phys.c +++ b/arch/x86/kernel/genx2apic_phys.c @@ -4,10 +4,30 @@ #include #include #include +#include + #include #include #include +DEFINE_PER_CPU(int, x2apic_extra_bits); + +static int x2apic_phys; + +static int set_x2apic_phys_mode(char *arg) +{ + x2apic_phys = 1; + return 0; +} +early_param("x2apic_phys", set_x2apic_phys_mode); + +static int __init x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + if (cpu_has_x2apic && intr_remapping_enabled && x2apic_phys) + return 1; + + return 0; +} /* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ @@ -122,6 +142,7 @@ void init_x2apic_ldr(void) struct genapic apic_x2apic_phys = { .name = "physical x2apic", + .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, .int_delivery_mode = dest_Fixed, .int_dest_mode = (APIC_DEST_PHYSICAL != 0), .target_cpus = x2apic_target_cpus, diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c index a8e5cb4c4d4..a882bc3a2ce 100644 --- a/arch/x86/kernel/genx2apic_uv_x.c +++ b/arch/x86/kernel/genx2apic_uv_x.c @@ -27,6 +27,33 @@ #include #include +static enum uv_system_type uv_system_type; + +static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + if (!strcmp(oem_id, "SGI")) { + if (!strcmp(oem_table_id, "UVL")) + uv_system_type = UV_LEGACY_APIC; + else if (!strcmp(oem_table_id, "UVX")) + uv_system_type = UV_X2APIC; + else if (!strcmp(oem_table_id, "UVH")) { + uv_system_type = UV_NON_UNIQUE_APIC; + return 1; + } + } + return 0; +} + +enum uv_system_type get_uv_system_type(void) +{ + return uv_system_type; +} + +int is_uv_system(void) +{ + return uv_system_type != UV_NONE; +} + DEFINE_PER_CPU(struct uv_hub_info_s, __uv_hub_info); EXPORT_PER_CPU_SYMBOL_GPL(__uv_hub_info); @@ -153,7 +180,7 @@ static unsigned int get_apic_id(unsigned long x) return id; } -static long set_apic_id(unsigned int id) +static unsigned long set_apic_id(unsigned int id) { unsigned long x; @@ -182,6 +209,7 @@ static void uv_send_IPI_self(int vector) struct genapic apic_x2apic_uv_x = { .name = "UV large system", + .acpi_madt_oem_check = uv_acpi_madt_oem_check, .int_delivery_mode = dest_Fixed, .int_dest_mode = (APIC_DEST_PHYSICAL != 0), .target_cpus = uv_target_cpus, @@ -433,3 +461,5 @@ void __cpuinit uv_cpu_init(void) if (get_uv_system_type() == UV_NON_UNIQUE_APIC) set_x2apic_extra_bits(uv_hub_info->pnode); } + + diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h index 2871b3fccb2..1e832e49f54 100644 --- a/include/asm-x86/genapic_64.h +++ b/include/asm-x86/genapic_64.h @@ -14,6 +14,7 @@ struct genapic { char *name; + int (*acpi_madt_oem_check)(char *oem_id, char *oem_table_id); u32 int_delivery_mode; u32 int_dest_mode; int (*apic_id_registered)(void); -- cgit v1.2.3-70-g09d2 From 026e2c05ef58ef413e2d52696f125d5ea1aa8bce Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 22 Jul 2008 11:58:14 +0200 Subject: x86, cyrix: debug Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/cyrix.c | 20 ++++++++++---------- include/asm-x86/processor-cyrix.h | 8 ++++++++ 2 files changed, 18 insertions(+), 10 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c index 3fd7a67bb06..db5868cd244 100644 --- a/arch/x86/kernel/cpu/cyrix.c +++ b/arch/x86/kernel/cpu/cyrix.c @@ -116,7 +116,7 @@ static void __cpuinit set_cx86_reorder(void) setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */ /* Load/Store Serialize to mem access disable (=reorder it) */ - setCx86(CX86_PCR0, getCx86(CX86_PCR0) & ~0x80); + setCx86_old(CX86_PCR0, getCx86_old(CX86_PCR0) & ~0x80); /* set load/store serialize from 1GB to 4GB */ ccr3 |= 0xe0; setCx86(CX86_CCR3, ccr3); @@ -127,11 +127,11 @@ static void __cpuinit set_cx86_memwb(void) printk(KERN_INFO "Enable Memory-Write-back mode on Cyrix/NSC processor.\n"); /* CCR2 bit 2: unlock NW bit */ - setCx86(CX86_CCR2, getCx86(CX86_CCR2) & ~0x04); + setCx86_old(CX86_CCR2, getCx86_old(CX86_CCR2) & ~0x04); /* set 'Not Write-through' */ write_cr0(read_cr0() | X86_CR0_NW); /* CCR2 bit 2: lock NW bit and set WT1 */ - setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x14); + setCx86_old(CX86_CCR2, getCx86_old(CX86_CCR2) | 0x14); } static void __cpuinit set_cx86_inc(void) @@ -144,10 +144,10 @@ static void __cpuinit set_cx86_inc(void) setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */ /* PCR1 -- Performance Control */ /* Incrementor on, whatever that is */ - setCx86(CX86_PCR1, getCx86(CX86_PCR1) | 0x02); + setCx86_old(CX86_PCR1, getCx86_old(CX86_PCR1) | 0x02); /* PCR0 -- Performance Control */ /* Incrementor Margin 10 */ - setCx86(CX86_PCR0, getCx86(CX86_PCR0) | 0x04); + setCx86_old(CX86_PCR0, getCx86_old(CX86_PCR0) | 0x04); setCx86(CX86_CCR3, ccr3); /* disable MAPEN */ } @@ -162,14 +162,14 @@ static void __cpuinit geode_configure(void) local_irq_save(flags); /* Suspend on halt power saving and enable #SUSP pin */ - setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x88); + setCx86_old(CX86_CCR2, getCx86_old(CX86_CCR2) | 0x88); ccr3 = getCx86(CX86_CCR3); setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */ /* FPU fast, DTE cache, Mem bypass */ - setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x38); + setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x38); setCx86(CX86_CCR3, ccr3); /* disable MAPEN */ set_cx86_memwb(); @@ -286,7 +286,7 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) /* GXm supports extended cpuid levels 'ala' AMD */ if (c->cpuid_level == 2) { /* Enable cxMMX extensions (GX1 Datasheet 54) */ - setCx86(CX86_CCR7, getCx86(CX86_CCR7) | 1); + setCx86_old(CX86_CCR7, getCx86_old(CX86_CCR7) | 1); /* * GXm : 0x30 ... 0x5f GXm datasheet 51 @@ -309,7 +309,7 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) if (dir1 > 7) { dir0_msn++; /* M II */ /* Enable MMX extensions (App note 108) */ - setCx86(CX86_CCR7, getCx86(CX86_CCR7)|1); + setCx86_old(CX86_CCR7, getCx86_old(CX86_CCR7)|1); } else { c->coma_bug = 1; /* 6x86MX, it has the bug. */ } @@ -424,7 +424,7 @@ static void __cpuinit cyrix_identify(struct cpuinfo_x86 *c) local_irq_save(flags); ccr3 = getCx86(CX86_CCR3); setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */ - setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x80); /* enable cpuid */ + setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x80); /* enable cpuid */ setCx86(CX86_CCR3, ccr3); /* disable MAPEN */ local_irq_restore(flags); } diff --git a/include/asm-x86/processor-cyrix.h b/include/asm-x86/processor-cyrix.h index 97568ada1f9..1198f2a0e42 100644 --- a/include/asm-x86/processor-cyrix.h +++ b/include/asm-x86/processor-cyrix.h @@ -28,3 +28,11 @@ static inline void setCx86(u8 reg, u8 data) outb(reg, 0x22); outb(data, 0x23); } + +#define getCx86_old(reg) ({ outb((reg), 0x22); inb(0x23); }) + +#define setCx86_old(reg, data) do { \ + outb((reg), 0x22); \ + outb((data), 0x23); \ +} while (0) + -- cgit v1.2.3-70-g09d2 From bbc1f698a508927d21324b57500e863f9bd562b9 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 21:34:13 +0530 Subject: x86: Introducing asm/syscalls.h Declaring arch-dependent syscalls for x86 architecture Signed-off-by: Jaswinder Singh --- arch/x86/kernel/ioport.c | 1 + arch/x86/kernel/ldt.c | 1 + arch/x86/kernel/process_32.c | 1 + arch/x86/kernel/process_64.c | 1 + arch/x86/kernel/signal_32.c | 1 + arch/x86/kernel/signal_64.c | 1 + arch/x86/kernel/sys_i386_32.c | 2 + arch/x86/kernel/sys_x86_64.c | 1 + arch/x86/kernel/tls.c | 1 + arch/x86/kernel/vm86_32.c | 1 + include/asm-x86/syscalls.h | 90 +++++++++++++++++++++++++++++++++++++++++++ 11 files changed, 101 insertions(+) create mode 100644 include/asm-x86/syscalls.h (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/ioport.c b/arch/x86/kernel/ioport.c index 50e5e4a31c8..19191430274 100644 --- a/arch/x86/kernel/ioport.c +++ b/arch/x86/kernel/ioport.c @@ -14,6 +14,7 @@ #include #include #include +#include /* Set EXTENT bits starting at BASE in BITMAP to value TURN_ON. */ static void set_bitmap(unsigned long *bitmap, unsigned int base, diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c index a8449571858..c49ff2dc8f8 100644 --- a/arch/x86/kernel/ldt.c +++ b/arch/x86/kernel/ldt.c @@ -18,6 +18,7 @@ #include #include #include +#include #ifdef CONFIG_SMP static void flush_ldt(void *current_mm) diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 0c3927accb0..6490e498126 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -55,6 +55,7 @@ #include #include #include +#include asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index e8a8e1b9981..c78090dd0c5 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -51,6 +51,7 @@ #include #include #include +#include asmlinkage extern void ret_from_fork(void); diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c index 07faaa5109c..5cede1045ce 100644 --- a/arch/x86/kernel/signal_32.c +++ b/arch/x86/kernel/signal_32.c @@ -26,6 +26,7 @@ #include #include #include +#include #include "sigframe.h" diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c index bf87684474f..b95a0a60905 100644 --- a/arch/x86/kernel/signal_64.c +++ b/arch/x86/kernel/signal_64.c @@ -26,6 +26,7 @@ #include #include #include +#include #include "sigframe.h" #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) diff --git a/arch/x86/kernel/sys_i386_32.c b/arch/x86/kernel/sys_i386_32.c index 7066cb855a6..1884a8d12bf 100644 --- a/arch/x86/kernel/sys_i386_32.c +++ b/arch/x86/kernel/sys_i386_32.c @@ -22,6 +22,8 @@ #include #include +#include + asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c index 3b360ef3381..c9288c883e2 100644 --- a/arch/x86/kernel/sys_x86_64.c +++ b/arch/x86/kernel/sys_x86_64.c @@ -16,6 +16,7 @@ #include #include +#include asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long off) diff --git a/arch/x86/kernel/tls.c b/arch/x86/kernel/tls.c index ab6bf375a30..6bb7b8579e7 100644 --- a/arch/x86/kernel/tls.c +++ b/arch/x86/kernel/tls.c @@ -10,6 +10,7 @@ #include #include #include +#include #include "tls.h" diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c index 38f566fa27d..4eeb5cf9720 100644 --- a/arch/x86/kernel/vm86_32.c +++ b/arch/x86/kernel/vm86_32.c @@ -46,6 +46,7 @@ #include #include #include +#include /* * Known problems: diff --git a/include/asm-x86/syscalls.h b/include/asm-x86/syscalls.h new file mode 100644 index 00000000000..170fcb132f5 --- /dev/null +++ b/include/asm-x86/syscalls.h @@ -0,0 +1,90 @@ +/* + * syscalls.h - Linux syscall interfaces (arch-specific) + * + * Copyright (c) 2008 Jaswinder Singh + * + * This file is released under the GPLv2. + * See the file COPYING for more details. + * + * Please do not call me directly, include linux/syscalls.h + */ + +#ifndef _ASM_X86_SYSCALLS_H +#define _ASM_X86_SYSCALLS_H + +#include +#include +#include +#include + +/* Common in X86_32 and X86_64 */ +/* kernel/ioport.c */ +asmlinkage long sys_ioperm(unsigned long, unsigned long, int); + +/* X86_32 only */ +#ifdef CONFIG_X86_32 +/* kernel/process_32.c */ +asmlinkage int sys_fork(struct pt_regs); +asmlinkage int sys_clone(struct pt_regs); +asmlinkage int sys_vfork(struct pt_regs); +asmlinkage int sys_execve(struct pt_regs); + +/* kernel/signal_32.c */ +asmlinkage int sys_sigsuspend(int, int, old_sigset_t); +asmlinkage int sys_sigaction(int, const struct old_sigaction __user *, + struct old_sigaction __user *); +asmlinkage int sys_sigaltstack(unsigned long); +asmlinkage unsigned long sys_sigreturn(unsigned long); +asmlinkage int sys_rt_sigreturn(unsigned long); + +/* kernel/ioport.c */ +asmlinkage long sys_iopl(unsigned long); + +/* kernel/ldt.c */ +asmlinkage int sys_modify_ldt(int, void __user *, unsigned long); + +/* kernel/sys_i386_32.c */ +asmlinkage long sys_mmap2(unsigned long, unsigned long, unsigned long, + unsigned long, unsigned long, unsigned long); +struct mmap_arg_struct; +asmlinkage int old_mmap(struct mmap_arg_struct __user *); +struct sel_arg_struct; +asmlinkage int old_select(struct sel_arg_struct __user *); +asmlinkage int sys_ipc(uint, int, int, int, void __user *, long); +struct old_utsname; +asmlinkage int sys_uname(struct old_utsname __user *); +struct oldold_utsname; +asmlinkage int sys_olduname(struct oldold_utsname __user *); + +/* kernel/tls.c */ +asmlinkage int sys_set_thread_area(struct user_desc __user *); +asmlinkage int sys_get_thread_area(struct user_desc __user *); + +#else /* CONFIG_X86_32 */ + +/* X86_64 only */ +/* kernel/process_64.c */ +asmlinkage long sys_fork(struct pt_regs *); +asmlinkage long sys_clone(unsigned long, unsigned long, + void __user *, void __user *, + struct pt_regs *); +asmlinkage long sys_vfork(struct pt_regs *); +asmlinkage long sys_execve(char __user *, char __user * __user *, + char __user * __user *, + struct pt_regs *); + +/* kernel/ioport.c */ +asmlinkage long sys_iopl(unsigned int, struct pt_regs *); + +/* kernel/signal_64.c */ +asmlinkage long sys_sigaltstack(const stack_t __user *, stack_t __user *, + struct pt_regs *); +asmlinkage long sys_rt_sigreturn(struct pt_regs *); + +/* kernel/sys_x86_64.c */ +asmlinkage long sys_mmap(unsigned long, unsigned long, unsigned long, + unsigned long, unsigned long, unsigned long); +asmlinkage long sys_uname(struct new_utsname __user *); + +#endif /* CONFIG_X86_32 */ +#endif /* _ASM_X86_SYSCALLS_H */ -- cgit v1.2.3-70-g09d2 From fb26132b441e75d6ba9996efc29b42081aee0abd Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 21:36:40 +0530 Subject: x86: process_32.c declare cpu_number before they get used Moved DECLARE_PER_CPU(int, cpu_number) from CONFIG_X86_32_SMP to CONFIG_X86_32 because cpu_number is required for both. And include asm/smp.h in process_32.c Signed-off-by: Jaswinder Singh --- arch/x86/kernel/process_32.c | 1 + include/asm-x86/smp.h | 4 +++- 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 6490e498126..7218bccd176 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -56,6 +56,7 @@ #include #include #include +#include asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 3c877f74f27..3a1480648f6 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -34,6 +34,9 @@ extern cpumask_t cpu_initialized; DECLARE_PER_CPU(cpumask_t, cpu_sibling_map); DECLARE_PER_CPU(cpumask_t, cpu_core_map); DECLARE_PER_CPU(u16, cpu_llc_id); +#ifdef CONFIG_X86_32 +DECLARE_PER_CPU(int, cpu_number); +#endif DECLARE_EARLY_PER_CPU(u16, x86_cpu_to_apicid); DECLARE_EARLY_PER_CPU(u16, x86_bios_cpu_apicid); @@ -142,7 +145,6 @@ extern unsigned disabled_cpus __cpuinitdata; * from the initial startup. We map APIC_BASE very early in page_setup(), * so this is correct in the x86 case. */ -DECLARE_PER_CPU(int, cpu_number); #define raw_smp_processor_id() (x86_read_percpu(cpu_number)) extern int safe_smp_processor_id(void); -- cgit v1.2.3-70-g09d2 From b994b6c0332a5499b33880855dadad04d74cde54 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 21:37:52 +0530 Subject: x86: signal_XX.c declare do_notify_resume before they get used Signed-off-by: Jaswinder Singh --- include/asm-x86/signal.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/signal.h b/include/asm-x86/signal.h index 6dac49364e9..5e4a1b0d8dc 100644 --- a/include/asm-x86/signal.h +++ b/include/asm-x86/signal.h @@ -140,6 +140,9 @@ struct sigaction { struct k_sigaction { struct sigaction sa; }; + +extern void do_notify_resume(struct pt_regs *, void *, __u32); + # else /* __KERNEL__ */ /* Here we must cater to libcs that poke about in kernel headers. */ -- cgit v1.2.3-70-g09d2 From cc0384917bf69079088701a0725c5fc6b554bf35 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 21:52:51 +0530 Subject: x86: time_XX.c declare functions before they get used Declare time_init() in asm-x86/time.h Also did cleanup in asm-x86/timer.h : timer_ack is only required for X86_32 int recalibrate_cpu_khz(void) is for X86_32 Signed-off-by: Jaswinder Singh --- arch/x86/kernel/time_32.c | 1 + include/asm-x86/time.h | 2 ++ include/asm-x86/timer.h | 5 ++++- 3 files changed, 7 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c index ffe3c664afc..bbecf8b6bf9 100644 --- a/arch/x86/kernel/time_32.c +++ b/arch/x86/kernel/time_32.c @@ -36,6 +36,7 @@ #include #include #include +#include #include "do_timer.h" diff --git a/include/asm-x86/time.h b/include/asm-x86/time.h index a17fa473e91..5f4fc3e0238 100644 --- a/include/asm-x86/time.h +++ b/include/asm-x86/time.h @@ -46,6 +46,8 @@ static inline int native_set_wallclock(unsigned long nowtime) #endif +extern void time_init(void); + #ifdef CONFIG_PARAVIRT #include #else /* !CONFIG_PARAVIRT */ diff --git a/include/asm-x86/timer.h b/include/asm-x86/timer.h index fb2a4ddddf3..2a8a92d5787 100644 --- a/include/asm-x86/timer.h +++ b/include/asm-x86/timer.h @@ -9,9 +9,12 @@ unsigned long long native_sched_clock(void); unsigned long native_calibrate_tsc(void); +#ifdef CONFIG_X86_32 extern int timer_ack; -extern int no_timer_check; extern int recalibrate_cpu_khz(void); +#endif /* CONFIG_X86_32 */ + +extern int no_timer_check; #ifndef CONFIG_PARAVIRT #define calibrate_tsc() native_calibrate_tsc() -- cgit v1.2.3-70-g09d2 From 5314d48ed54c1a0111c597d1510f77850a1b3232 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 22:12:23 +0530 Subject: x86: setup.c declare saved_video_mode before they get used Signed-off-by: Jaswinder Singh --- include/asm-x86/setup.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/setup.h b/include/asm-x86/setup.h index a07c6f1c01e..946f8f20e91 100644 --- a/include/asm-x86/setup.h +++ b/include/asm-x86/setup.h @@ -41,6 +41,7 @@ struct x86_quirks { }; extern struct x86_quirks *x86_quirks; +extern unsigned long saved_video_mode; #ifndef CONFIG_PARAVIRT #define paravirt_post_allocator_init() do {} while (0) -- cgit v1.2.3-70-g09d2 From a7b7511ac1404eaf0e7b6c445a7c61b48ccfcf0b Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 22:19:29 +0530 Subject: x86: e820.c declare pci_mem_start before they get used Signed-off-by: Jaswinder Singh --- include/asm-x86/e820.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/e820.h b/include/asm-x86/e820.h index 16a31e2c7c5..213d1ed3d41 100644 --- a/include/asm-x86/e820.h +++ b/include/asm-x86/e820.h @@ -64,6 +64,7 @@ struct e820map { extern struct e820map e820; extern struct e820map e820_saved; +extern unsigned long pci_mem_start; extern int e820_any_mapped(u64 start, u64 end, unsigned type); extern int e820_all_mapped(u64 start, u64 end, unsigned type); extern void e820_add_region(u64 start, u64 size, int type); -- cgit v1.2.3-70-g09d2 From 9321b8cbbbf3a8dbd4748e3722facaeb8401bd13 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 22:24:29 +0530 Subject: x86: pci-dma.c declare iommu_bio_merge before they get used moved iommu_bio_merge from io_64.h to io.h because it is required for both. Signed-off-by: Jaswinder Singh --- include/asm-x86/io.h | 2 ++ include/asm-x86/io_64.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/io.h b/include/asm-x86/io.h index bf5d629b3a3..a145210ac16 100644 --- a/include/asm-x86/io.h +++ b/include/asm-x86/io.h @@ -73,6 +73,8 @@ build_mmio_write(__writeq, "q", unsigned long, "r", ) #define writeq writeq #endif +extern int iommu_bio_merge; + #ifdef CONFIG_X86_32 # include "io_32.h" #else diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h index ddd8058a502..1e271378afa 100644 --- a/include/asm-x86/io_64.h +++ b/include/asm-x86/io_64.h @@ -233,7 +233,6 @@ void memset_io(volatile void __iomem *a, int b, size_t c); #define flush_write_buffers() -extern int iommu_bio_merge; #define BIO_VMERGE_BOUNDARY iommu_bio_merge /* -- cgit v1.2.3-70-g09d2 From 791b897ccd1a2c6c184b88ca6d1aaf053499c3df Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 22:28:22 +0530 Subject: x86: pci-nommu.c declare nommu_dma_ops before they get used Signed-off-by: Jaswinder Singh --- include/asm-x86/dma-mapping.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index c2ddd3d1b88..f14b04d8bae 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -57,6 +57,7 @@ struct dma_mapping_ops { }; extern const struct dma_mapping_ops *dma_ops; +extern const struct dma_mapping_ops nommu_dma_ops; static inline int dma_mapping_error(dma_addr_t dma_addr) { -- cgit v1.2.3-70-g09d2 From 36454936c00c700ae86b5ff376d3c1c1a862c4f5 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 22:31:57 +0530 Subject: x86: i387.c declare dump_fpu() before they get used Signed-off-by: Jaswinder Singh --- include/asm-x86/i387.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h index 37672f79dcc..48789892383 100644 --- a/include/asm-x86/i387.h +++ b/include/asm-x86/i387.h @@ -24,6 +24,7 @@ extern void mxcsr_feature_mask_init(void); extern int init_fpu(struct task_struct *child); extern asmlinkage void math_state_restore(void); extern void init_thread_xstate(void); +extern int dump_fpu(struct pt_regs *, struct user_i387_struct *); extern user_regset_active_fn fpregs_active, xfpregs_active; extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get; -- cgit v1.2.3-70-g09d2 From c1686aeaf0780055ffcd4b224b73d5ada77630e8 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 22:35:38 +0530 Subject: x86: ptrace.c declare functions before they get used Signed-off-by: Jaswinder Singh --- include/asm-x86/ptrace.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h index 8a71db803da..48c1d5a890d 100644 --- a/include/asm-x86/ptrace.h +++ b/include/asm-x86/ptrace.h @@ -148,6 +148,9 @@ extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, void signal_fault(struct pt_regs *regs, void __user *frame, char *where); #endif +extern long syscall_trace_enter(struct pt_regs *); +extern void syscall_trace_leave(struct pt_regs *); + static inline unsigned long regs_return_value(struct pt_regs *regs) { return regs->ax; -- cgit v1.2.3-70-g09d2 From 1c6c727d9c12c84a612abe31b60948f06fc2ab2d Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 22:40:37 +0530 Subject: x86: proc.c declare cpuinfo_op before they get used Signed-off-by: Jaswinder Singh --- include/asm-x86/processor.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index 15cb82a44e8..93b6adc72ad 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -140,6 +140,8 @@ DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info); #define current_cpu_data boot_cpu_data #endif +extern const struct seq_operations cpuinfo_op; + static inline int hlt_works(int cpu) { #ifdef CONFIG_X86_32 -- cgit v1.2.3-70-g09d2 From 8fd329a1ac696973ba5467c510302ae1248cc11a Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Mon, 21 Jul 2008 22:54:56 +0530 Subject: x86: common.c declare idle_regs before they get used Signed-off-by: Jaswinder Singh --- include/asm-x86/processor.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index 93b6adc72ad..a2fb5a43641 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -155,6 +155,8 @@ static inline int hlt_works(int cpu) extern void cpu_detect(struct cpuinfo_x86 *c); +extern struct pt_regs *idle_regs(struct pt_regs *); + extern void early_cpu_init(void); extern void identify_boot_cpu(void); extern void identify_secondary_cpu(struct cpuinfo_x86 *); -- cgit v1.2.3-70-g09d2 From df1be4372eae5f104b7fb4991bc4b35f00b11a11 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 22 Jul 2008 10:13:11 -0400 Subject: x64, apic: use generic apic_write() for ack_APIC_irq() I tested tip/master and found an issue (patch attached) for x2apic support. This is not because of the recent merges we had, but because of something(where we still access memory based interface after enabling x2apic mode) that slipped through my earlier tests. Probably it is a good idea to unmap the memory mapped interface, once we switch to x2apic mode. That will catch the issues much earlier. I will post another patch for this. ack_APIC_irq() is used at too many generic places (and not just during irq_chip handling!) to use the native_apic_mem_write(). For ex, this will break x2apic based systems. Fix ack_APIC_irq() to use the generic apic_write() even for 64-bit. Signed-off-by: Suresh Siddha Cc: suresh.b.siddha@intel.com Cc: yong.y.wang@linux.intel.com Signed-off-by: Ingo Molnar --- include/asm-x86/apic.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h index 51339910fdc..502ca6594b1 100644 --- a/include/asm-x86/apic.h +++ b/include/asm-x86/apic.h @@ -136,11 +136,7 @@ static inline void ack_APIC_irq(void) */ /* Docs say use 0 for future compatibility */ -#ifdef CONFIG_X86_32 apic_write(APIC_EOI, 0); -#else - native_apic_mem_write(APIC_EOI, 0); -#endif } extern int lapic_get_maxlvt(void); -- cgit v1.2.3-70-g09d2 From a656c8efb40a8700046df20da2195f8aa39ce38a Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Tue, 22 Jul 2008 21:27:11 +0200 Subject: x86: fix spurious '#' in kvm header Signed-off-by: Vegard Nossum --- include/asm-x86/kvm_host.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index fdde0bedaa9..924e1bfe3a6 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -1,4 +1,4 @@ -#/* +/* * Kernel-based Virtual Machine driver for Linux * * This header defines architecture specific interfaces, x86 version -- cgit v1.2.3-70-g09d2 From 77ef50a522717fa040636ee1017179ceba12ff62 Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Wed, 18 Jun 2008 17:08:48 +0200 Subject: x86: consolidate header guards This patch is the result of an automatic script that consolidates the format of all the headers in include/asm-x86/. The format: 1. No leading underscore. Names with leading underscores are reserved. 2. Pathname components are separated by two underscores. So we can distinguish between mm_types.h and mm/types.h. 3. Everything except letters and numbers are turned into single underscores. Signed-off-by: Vegard Nossum --- include/asm-x86/a.out-core.h | 6 +++--- include/asm-x86/a.out.h | 6 +++--- include/asm-x86/acpi.h | 6 +++--- include/asm-x86/agp.h | 6 +++--- include/asm-x86/alternative.h | 6 +++--- include/asm-x86/amd_iommu.h | 6 +++--- include/asm-x86/amd_iommu_types.h | 6 +++--- include/asm-x86/apic.h | 6 +++--- include/asm-x86/apicdef.h | 6 +++--- include/asm-x86/arch_hooks.h | 6 +++--- include/asm-x86/asm.h | 6 +++--- include/asm-x86/atomic_32.h | 6 +++--- include/asm-x86/atomic_64.h | 6 +++--- include/asm-x86/auxvec.h | 6 +++--- include/asm-x86/bios_ebda.h | 6 +++--- include/asm-x86/bitops.h | 6 +++--- include/asm-x86/boot.h | 6 +++--- include/asm-x86/bootparam.h | 6 +++--- include/asm-x86/bug.h | 6 +++--- include/asm-x86/bugs.h | 6 +++--- include/asm-x86/byteorder.h | 6 +++--- include/asm-x86/cache.h | 6 +++--- include/asm-x86/cacheflush.h | 6 +++--- include/asm-x86/calgary.h | 6 +++--- include/asm-x86/checksum_32.h | 6 +++--- include/asm-x86/checksum_64.h | 6 +++--- include/asm-x86/cmpxchg_32.h | 6 +++--- include/asm-x86/cmpxchg_64.h | 6 +++--- include/asm-x86/compat.h | 6 +++--- include/asm-x86/cpu.h | 6 +++--- include/asm-x86/cpufeature.h | 6 +++--- include/asm-x86/current.h | 6 +++--- include/asm-x86/debugreg.h | 6 +++--- include/asm-x86/delay.h | 6 +++--- include/asm-x86/desc_defs.h | 6 +++--- include/asm-x86/device.h | 6 +++--- include/asm-x86/div64.h | 6 +++--- include/asm-x86/dma-mapping.h | 6 +++--- include/asm-x86/dma.h | 6 +++--- include/asm-x86/dmi.h | 6 +++--- include/asm-x86/ds.h | 6 +++--- include/asm-x86/dwarf2.h | 6 +++--- include/asm-x86/e820.h | 6 +++--- include/asm-x86/edac.h | 6 +++--- include/asm-x86/efi.h | 6 +++--- include/asm-x86/elf.h | 6 +++--- include/asm-x86/emergency-restart.h | 6 +++--- include/asm-x86/fb.h | 6 +++--- include/asm-x86/fixmap.h | 6 +++--- include/asm-x86/fixmap_32.h | 6 +++--- include/asm-x86/fixmap_64.h | 6 +++--- include/asm-x86/floppy.h | 6 +++--- include/asm-x86/ftrace.h | 6 +++--- include/asm-x86/futex.h | 6 +++--- include/asm-x86/gart.h | 6 +++--- include/asm-x86/genapic_32.h | 6 +++--- include/asm-x86/genapic_64.h | 6 +++--- include/asm-x86/geode.h | 6 +++--- include/asm-x86/gpio.h | 6 +++--- include/asm-x86/hardirq_32.h | 6 +++--- include/asm-x86/hardirq_64.h | 6 +++--- include/asm-x86/highmem.h | 6 +++--- include/asm-x86/hpet.h | 6 +++--- include/asm-x86/hugetlb.h | 6 +++--- include/asm-x86/hw_irq.h | 6 +++--- include/asm-x86/hypertransport.h | 6 +++--- include/asm-x86/i387.h | 6 +++--- include/asm-x86/i8253.h | 6 +++--- include/asm-x86/i8259.h | 6 +++--- include/asm-x86/ia32.h | 6 +++--- include/asm-x86/ia32_unistd.h | 6 +++--- include/asm-x86/ide.h | 6 +++--- include/asm-x86/idle.h | 6 +++--- include/asm-x86/intel_arch_perfmon.h | 6 +++--- include/asm-x86/io.h | 6 +++--- include/asm-x86/io_32.h | 6 +++--- include/asm-x86/io_64.h | 6 +++--- include/asm-x86/io_apic.h | 6 +++--- include/asm-x86/ioctls.h | 6 +++--- include/asm-x86/iommu.h | 6 +++--- include/asm-x86/ipcbuf.h | 6 +++--- include/asm-x86/ipi.h | 6 +++--- include/asm-x86/irq.h | 6 +++--- include/asm-x86/irq_regs_32.h | 6 +++--- include/asm-x86/irq_vectors.h | 6 +++--- include/asm-x86/ist.h | 6 +++--- include/asm-x86/k8.h | 6 +++--- include/asm-x86/kdebug.h | 6 +++--- include/asm-x86/kexec.h | 6 +++--- include/asm-x86/kgdb.h | 6 +++--- include/asm-x86/kmap_types.h | 6 +++--- include/asm-x86/kprobes.h | 6 +++--- include/asm-x86/kvm.h | 6 +++--- include/asm-x86/kvm_host.h | 6 +++--- include/asm-x86/kvm_para.h | 6 +++--- include/asm-x86/kvm_x86_emulate.h | 6 +++--- include/asm-x86/ldt.h | 6 +++--- include/asm-x86/lguest.h | 6 +++--- include/asm-x86/lguest_hcall.h | 6 +++--- include/asm-x86/linkage.h | 6 +++--- include/asm-x86/local.h | 6 +++--- include/asm-x86/mach-bigsmp/mach_apic.h | 6 +++--- include/asm-x86/mach-bigsmp/mach_apicdef.h | 6 +++--- include/asm-x86/mach-bigsmp/mach_ipi.h | 6 +++--- include/asm-x86/mach-default/apm.h | 6 +++--- include/asm-x86/mach-default/mach_apic.h | 6 +++--- include/asm-x86/mach-default/mach_apicdef.h | 6 +++--- include/asm-x86/mach-default/mach_ipi.h | 6 +++--- include/asm-x86/mach-default/mach_mpparse.h | 6 +++--- include/asm-x86/mach-default/mach_mpspec.h | 6 +++--- include/asm-x86/mach-default/mach_timer.h | 6 +++--- include/asm-x86/mach-default/mach_traps.h | 6 +++--- include/asm-x86/mach-default/mach_wakecpu.h | 6 +++--- include/asm-x86/mach-es7000/mach_apic.h | 6 +++--- include/asm-x86/mach-es7000/mach_apicdef.h | 6 +++--- include/asm-x86/mach-es7000/mach_ipi.h | 6 +++--- include/asm-x86/mach-es7000/mach_mpparse.h | 6 +++--- include/asm-x86/mach-es7000/mach_wakecpu.h | 6 +++--- include/asm-x86/mach-generic/gpio.h | 6 +++--- include/asm-x86/mach-generic/irq_vectors_limits.h | 6 +++--- include/asm-x86/mach-generic/mach_apic.h | 6 +++--- include/asm-x86/mach-generic/mach_apicdef.h | 6 +++--- include/asm-x86/mach-generic/mach_ipi.h | 6 +++--- include/asm-x86/mach-generic/mach_mpparse.h | 6 +++--- include/asm-x86/mach-generic/mach_mpspec.h | 6 +++--- include/asm-x86/mach-numaq/mach_apic.h | 6 +++--- include/asm-x86/mach-numaq/mach_apicdef.h | 6 +++--- include/asm-x86/mach-numaq/mach_ipi.h | 6 +++--- include/asm-x86/mach-numaq/mach_mpparse.h | 6 +++--- include/asm-x86/mach-numaq/mach_wakecpu.h | 6 +++--- include/asm-x86/mach-rdc321x/gpio.h | 6 +++--- include/asm-x86/mach-summit/irq_vectors_limits.h | 6 +++--- include/asm-x86/mach-summit/mach_apic.h | 6 +++--- include/asm-x86/mach-summit/mach_apicdef.h | 6 +++--- include/asm-x86/mach-summit/mach_ipi.h | 6 +++--- include/asm-x86/mach-summit/mach_mpparse.h | 6 +++--- include/asm-x86/math_emu.h | 6 +++--- include/asm-x86/mc146818rtc.h | 6 +++--- include/asm-x86/mca.h | 6 +++--- include/asm-x86/mca_dma.h | 6 +++--- include/asm-x86/mce.h | 6 +++--- include/asm-x86/mman.h | 6 +++--- include/asm-x86/mmconfig.h | 6 +++--- include/asm-x86/mmu.h | 6 +++--- include/asm-x86/mmu_context.h | 6 +++--- include/asm-x86/mmu_context_32.h | 6 +++--- include/asm-x86/mmu_context_64.h | 6 +++--- include/asm-x86/mmx.h | 6 +++--- include/asm-x86/mmzone_32.h | 6 +++--- include/asm-x86/mmzone_64.h | 6 +++--- include/asm-x86/module.h | 6 +++--- include/asm-x86/mpspec.h | 6 +++--- include/asm-x86/mpspec_def.h | 6 +++--- include/asm-x86/msgbuf.h | 6 +++--- include/asm-x86/msidef.h | 6 +++--- include/asm-x86/msr-index.h | 6 +++--- include/asm-x86/msr.h | 6 +++--- include/asm-x86/mtrr.h | 6 +++--- include/asm-x86/mutex_32.h | 6 +++--- include/asm-x86/mutex_64.h | 6 +++--- include/asm-x86/namei.h | 6 +++--- include/asm-x86/nmi.h | 6 +++--- include/asm-x86/nops.h | 6 +++--- include/asm-x86/numa_32.h | 6 +++--- include/asm-x86/numa_64.h | 6 +++--- include/asm-x86/numaq.h | 6 +++--- include/asm-x86/olpc.h | 6 +++--- include/asm-x86/page.h | 6 +++--- include/asm-x86/page_32.h | 6 +++--- include/asm-x86/page_64.h | 6 +++--- include/asm-x86/param.h | 6 +++--- include/asm-x86/paravirt.h | 6 +++--- include/asm-x86/parport.h | 6 +++--- include/asm-x86/pat.h | 6 +++--- include/asm-x86/pci-direct.h | 6 +++--- include/asm-x86/pci.h | 6 +++--- include/asm-x86/pci_32.h | 6 +++--- include/asm-x86/pci_64.h | 6 +++--- include/asm-x86/pda.h | 6 +++--- include/asm-x86/percpu.h | 6 +++--- include/asm-x86/pgalloc.h | 6 +++--- include/asm-x86/pgtable-2level-defs.h | 6 +++--- include/asm-x86/pgtable-2level.h | 6 +++--- include/asm-x86/pgtable-3level-defs.h | 6 +++--- include/asm-x86/pgtable-3level.h | 6 +++--- include/asm-x86/pgtable.h | 6 +++--- include/asm-x86/pgtable_32.h | 6 +++--- include/asm-x86/pgtable_64.h | 6 +++--- include/asm-x86/posix_types_32.h | 6 +++--- include/asm-x86/posix_types_64.h | 6 +++--- include/asm-x86/prctl.h | 6 +++--- include/asm-x86/processor-flags.h | 6 +++--- include/asm-x86/processor.h | 6 +++--- include/asm-x86/proto.h | 6 +++--- include/asm-x86/ptrace-abi.h | 6 +++--- include/asm-x86/ptrace.h | 6 +++--- include/asm-x86/pvclock-abi.h | 6 +++--- include/asm-x86/pvclock.h | 6 +++--- include/asm-x86/reboot.h | 6 +++--- include/asm-x86/reboot_fixups.h | 6 +++--- include/asm-x86/required-features.h | 6 +++--- include/asm-x86/resume-trace.h | 6 +++--- include/asm-x86/rio.h | 6 +++--- include/asm-x86/rwlock.h | 6 +++--- include/asm-x86/rwsem.h | 6 +++--- include/asm-x86/scatterlist.h | 6 +++--- include/asm-x86/seccomp_32.h | 6 +++--- include/asm-x86/seccomp_64.h | 6 +++--- include/asm-x86/segment.h | 6 +++--- include/asm-x86/sembuf.h | 6 +++--- include/asm-x86/serial.h | 6 +++--- include/asm-x86/setup.h | 6 +++--- include/asm-x86/shmbuf.h | 6 +++--- include/asm-x86/shmparam.h | 6 +++--- include/asm-x86/sigcontext.h | 6 +++--- include/asm-x86/sigcontext32.h | 6 +++--- include/asm-x86/siginfo.h | 6 +++--- include/asm-x86/signal.h | 6 +++--- include/asm-x86/smp.h | 6 +++--- include/asm-x86/socket.h | 6 +++--- include/asm-x86/sockios.h | 6 +++--- include/asm-x86/sparsemem.h | 6 +++--- include/asm-x86/spinlock.h | 6 +++--- include/asm-x86/spinlock_types.h | 6 +++--- include/asm-x86/srat.h | 6 +++--- include/asm-x86/stacktrace.h | 6 +++--- include/asm-x86/stat.h | 6 +++--- include/asm-x86/statfs.h | 6 +++--- include/asm-x86/string_32.h | 6 +++--- include/asm-x86/string_64.h | 6 +++--- include/asm-x86/suspend_32.h | 6 +++--- include/asm-x86/suspend_64.h | 6 +++--- include/asm-x86/swiotlb.h | 6 +++--- include/asm-x86/sync_bitops.h | 6 +++--- include/asm-x86/system.h | 6 +++--- include/asm-x86/system_64.h | 6 +++--- include/asm-x86/tce.h | 6 +++--- include/asm-x86/termbits.h | 6 +++--- include/asm-x86/termios.h | 6 +++--- include/asm-x86/therm_throt.h | 6 +++--- include/asm-x86/thread_info.h | 6 +++--- include/asm-x86/time.h | 6 +++--- include/asm-x86/timer.h | 6 +++--- include/asm-x86/timex.h | 6 +++--- include/asm-x86/tlb.h | 6 +++--- include/asm-x86/tlbflush.h | 6 +++--- include/asm-x86/topology.h | 6 +++--- include/asm-x86/trampoline.h | 6 +++--- include/asm-x86/traps.h | 6 +++--- include/asm-x86/tsc.h | 6 +++--- include/asm-x86/types.h | 6 +++--- include/asm-x86/uaccess.h | 6 +++--- include/asm-x86/uaccess_32.h | 6 +++--- include/asm-x86/uaccess_64.h | 6 +++--- include/asm-x86/ucontext.h | 6 +++--- include/asm-x86/unaligned.h | 6 +++--- include/asm-x86/unistd_32.h | 6 +++--- include/asm-x86/unwind.h | 6 +++--- include/asm-x86/user32.h | 6 +++--- include/asm-x86/user_32.h | 6 +++--- include/asm-x86/user_64.h | 6 +++--- include/asm-x86/uv/bios.h | 6 +++--- include/asm-x86/uv/uv_bau.h | 6 +++--- include/asm-x86/uv/uv_hub.h | 6 +++--- include/asm-x86/uv/uv_mmrs.h | 6 +++--- include/asm-x86/vdso.h | 6 +++--- include/asm-x86/vga.h | 6 +++--- include/asm-x86/vgtod.h | 6 +++--- include/asm-x86/visws/cobalt.h | 6 +++--- include/asm-x86/visws/lithium.h | 6 +++--- include/asm-x86/visws/piix4.h | 6 +++--- include/asm-x86/vm86.h | 6 +++--- include/asm-x86/vmi_time.h | 6 +++--- include/asm-x86/vsyscall.h | 6 +++--- include/asm-x86/xen/events.h | 6 +++--- include/asm-x86/xen/grant_table.h | 6 +++--- include/asm-x86/xen/hypercall.h | 6 +++--- include/asm-x86/xen/hypervisor.h | 6 +++--- include/asm-x86/xen/interface.h | 6 +++--- include/asm-x86/xen/interface_32.h | 6 +++--- include/asm-x86/xen/interface_64.h | 6 +++--- include/asm-x86/xen/page.h | 6 +++--- 282 files changed, 846 insertions(+), 846 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/a.out-core.h b/include/asm-x86/a.out-core.h index 714207a1c38..f5705761a37 100644 --- a/include/asm-x86/a.out-core.h +++ b/include/asm-x86/a.out-core.h @@ -9,8 +9,8 @@ * 2 of the Licence, or (at your option) any later version. */ -#ifndef _ASM_A_OUT_CORE_H -#define _ASM_A_OUT_CORE_H +#ifndef ASM_X86__A_OUT_CORE_H +#define ASM_X86__A_OUT_CORE_H #ifdef __KERNEL__ #ifdef CONFIG_X86_32 @@ -70,4 +70,4 @@ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump) #endif /* CONFIG_X86_32 */ #endif /* __KERNEL__ */ -#endif /* _ASM_A_OUT_CORE_H */ +#endif /* ASM_X86__A_OUT_CORE_H */ diff --git a/include/asm-x86/a.out.h b/include/asm-x86/a.out.h index 4684f97a5bb..0948748bc69 100644 --- a/include/asm-x86/a.out.h +++ b/include/asm-x86/a.out.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_A_OUT_H -#define _ASM_X86_A_OUT_H +#ifndef ASM_X86__A_OUT_H +#define ASM_X86__A_OUT_H struct exec { @@ -17,4 +17,4 @@ struct exec #define N_DRSIZE(a) ((a).a_drsize) #define N_SYMSIZE(a) ((a).a_syms) -#endif /* _ASM_X86_A_OUT_H */ +#endif /* ASM_X86__A_OUT_H */ diff --git a/include/asm-x86/acpi.h b/include/asm-x86/acpi.h index 635d764dc13..bd76299586b 100644 --- a/include/asm-x86/acpi.h +++ b/include/asm-x86/acpi.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_ACPI_H -#define _ASM_X86_ACPI_H +#ifndef ASM_X86__ACPI_H +#define ASM_X86__ACPI_H /* * Copyright (C) 2001 Paul Diefenbaugh @@ -173,4 +173,4 @@ static inline void acpi_fake_nodes(const struct bootnode *fake_nodes, #define acpi_unlazy_tlb(x) leave_mm(x) -#endif /*__X86_ASM_ACPI_H*/ +#endif /* ASM_X86__ACPI_H */ diff --git a/include/asm-x86/agp.h b/include/asm-x86/agp.h index e4004a9f6a9..3617fd4fcdf 100644 --- a/include/asm-x86/agp.h +++ b/include/asm-x86/agp.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_AGP_H -#define _ASM_X86_AGP_H +#ifndef ASM_X86__AGP_H +#define ASM_X86__AGP_H #include #include @@ -32,4 +32,4 @@ #define free_gatt_pages(table, order) \ free_pages((unsigned long)(table), (order)) -#endif +#endif /* ASM_X86__AGP_H */ diff --git a/include/asm-x86/alternative.h b/include/asm-x86/alternative.h index f6aa18eadf7..22d3c9862bf 100644 --- a/include/asm-x86/alternative.h +++ b/include/asm-x86/alternative.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_ALTERNATIVE_H -#define _ASM_X86_ALTERNATIVE_H +#ifndef ASM_X86__ALTERNATIVE_H +#define ASM_X86__ALTERNATIVE_H #include #include @@ -180,4 +180,4 @@ extern void add_nops(void *insns, unsigned int len); extern void *text_poke(void *addr, const void *opcode, size_t len); extern void *text_poke_early(void *addr, const void *opcode, size_t len); -#endif /* _ASM_X86_ALTERNATIVE_H */ +#endif /* ASM_X86__ALTERNATIVE_H */ diff --git a/include/asm-x86/amd_iommu.h b/include/asm-x86/amd_iommu.h index 30a12049353..783f43e5805 100644 --- a/include/asm-x86/amd_iommu.h +++ b/include/asm-x86/amd_iommu.h @@ -17,8 +17,8 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#ifndef _ASM_X86_AMD_IOMMU_H -#define _ASM_X86_AMD_IOMMU_H +#ifndef ASM_X86__AMD_IOMMU_H +#define ASM_X86__AMD_IOMMU_H #ifdef CONFIG_AMD_IOMMU extern int amd_iommu_init(void); @@ -29,4 +29,4 @@ static inline int amd_iommu_init(void) { return -ENODEV; } static inline void amd_iommu_detect(void) { } #endif -#endif +#endif /* ASM_X86__AMD_IOMMU_H */ diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index 22aa58ca199..e6b4d5b0837 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -17,8 +17,8 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#ifndef __AMD_IOMMU_TYPES_H__ -#define __AMD_IOMMU_TYPES_H__ +#ifndef ASM_X86__AMD_IOMMU_TYPES_H +#define ASM_X86__AMD_IOMMU_TYPES_H #include #include @@ -339,4 +339,4 @@ static inline u16 calc_devid(u8 bus, u8 devfn) return (((u16)bus) << 8) | devfn; } -#endif +#endif /* ASM_X86__AMD_IOMMU_TYPES_H */ diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h index 133c998161c..d0ace3d63a2 100644 --- a/include/asm-x86/apic.h +++ b/include/asm-x86/apic.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_APIC_H -#define _ASM_X86_APIC_H +#ifndef ASM_X86__APIC_H +#define ASM_X86__APIC_H #include #include @@ -128,4 +128,4 @@ static inline void init_apic_mappings(void) { } #endif /* !CONFIG_X86_LOCAL_APIC */ -#endif /* __ASM_APIC_H */ +#endif /* ASM_X86__APIC_H */ diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h index 6b9008c7873..c40687da20f 100644 --- a/include/asm-x86/apicdef.h +++ b/include/asm-x86/apicdef.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_APICDEF_H -#define _ASM_X86_APICDEF_H +#ifndef ASM_X86__APICDEF_H +#define ASM_X86__APICDEF_H /* * Constants for various Intel APICs. (local APIC, IOAPIC, etc.) @@ -411,4 +411,4 @@ struct local_apic { #else #define BAD_APICID 0xFFFFu #endif -#endif +#endif /* ASM_X86__APICDEF_H */ diff --git a/include/asm-x86/arch_hooks.h b/include/asm-x86/arch_hooks.h index 8411750ceb6..72adc3a109c 100644 --- a/include/asm-x86/arch_hooks.h +++ b/include/asm-x86/arch_hooks.h @@ -1,5 +1,5 @@ -#ifndef _ASM_ARCH_HOOKS_H -#define _ASM_ARCH_HOOKS_H +#ifndef ASM_X86__ARCH_HOOKS_H +#define ASM_X86__ARCH_HOOKS_H #include @@ -25,4 +25,4 @@ extern void pre_time_init_hook(void); extern void time_init_hook(void); extern void mca_nmi_hook(void); -#endif +#endif /* ASM_X86__ARCH_HOOKS_H */ diff --git a/include/asm-x86/asm.h b/include/asm-x86/asm.h index 97220321f39..2439ae49e8a 100644 --- a/include/asm-x86/asm.h +++ b/include/asm-x86/asm.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_ASM_H -#define _ASM_X86_ASM_H +#ifndef ASM_X86__ASM_H +#define ASM_X86__ASM_H #ifdef __ASSEMBLY__ # define __ASM_FORM(x) x @@ -39,4 +39,4 @@ _ASM_PTR #from "," #to "\n" \ " .previous\n" -#endif /* _ASM_X86_ASM_H */ +#endif /* ASM_X86__ASM_H */ diff --git a/include/asm-x86/atomic_32.h b/include/asm-x86/atomic_32.h index 21a4825148c..14d3f0beb88 100644 --- a/include/asm-x86/atomic_32.h +++ b/include/asm-x86/atomic_32.h @@ -1,5 +1,5 @@ -#ifndef __ARCH_I386_ATOMIC__ -#define __ARCH_I386_ATOMIC__ +#ifndef ASM_X86__ATOMIC_32_H +#define ASM_X86__ATOMIC_32_H #include #include @@ -256,4 +256,4 @@ static inline int atomic_add_unless(atomic_t *v, int a, int u) #define smp_mb__after_atomic_inc() barrier() #include -#endif +#endif /* ASM_X86__ATOMIC_32_H */ diff --git a/include/asm-x86/atomic_64.h b/include/asm-x86/atomic_64.h index a0095191c02..ebbc753af6a 100644 --- a/include/asm-x86/atomic_64.h +++ b/include/asm-x86/atomic_64.h @@ -1,5 +1,5 @@ -#ifndef __ARCH_X86_64_ATOMIC__ -#define __ARCH_X86_64_ATOMIC__ +#ifndef ASM_X86__ATOMIC_64_H +#define ASM_X86__ATOMIC_64_H #include #include @@ -470,4 +470,4 @@ static inline void atomic_or_long(unsigned long *v1, unsigned long v2) #define smp_mb__after_atomic_inc() barrier() #include -#endif +#endif /* ASM_X86__ATOMIC_64_H */ diff --git a/include/asm-x86/auxvec.h b/include/asm-x86/auxvec.h index 87f5e6d5a02..12c7cac7420 100644 --- a/include/asm-x86/auxvec.h +++ b/include/asm-x86/auxvec.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_AUXVEC_H -#define _ASM_X86_AUXVEC_H +#ifndef ASM_X86__AUXVEC_H +#define ASM_X86__AUXVEC_H /* * Architecture-neutral AT_ values in 0-17, leave some room * for more of them, start the x86-specific ones at 32. @@ -9,4 +9,4 @@ #endif #define AT_SYSINFO_EHDR 33 -#endif +#endif /* ASM_X86__AUXVEC_H */ diff --git a/include/asm-x86/bios_ebda.h b/include/asm-x86/bios_ebda.h index 0033e50c13b..ec42ed87459 100644 --- a/include/asm-x86/bios_ebda.h +++ b/include/asm-x86/bios_ebda.h @@ -1,5 +1,5 @@ -#ifndef _MACH_BIOS_EBDA_H -#define _MACH_BIOS_EBDA_H +#ifndef ASM_X86__BIOS_EBDA_H +#define ASM_X86__BIOS_EBDA_H #include @@ -16,4 +16,4 @@ static inline unsigned int get_bios_ebda(void) void reserve_ebda_region(void); -#endif /* _MACH_BIOS_EBDA_H */ +#endif /* ASM_X86__BIOS_EBDA_H */ diff --git a/include/asm-x86/bitops.h b/include/asm-x86/bitops.h index cfb2b64f76e..61989b93b47 100644 --- a/include/asm-x86/bitops.h +++ b/include/asm-x86/bitops.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_BITOPS_H -#define _ASM_X86_BITOPS_H +#ifndef ASM_X86__BITOPS_H +#define ASM_X86__BITOPS_H /* * Copyright 1992, Linus Torvalds. @@ -458,4 +458,4 @@ static inline void set_bit_string(unsigned long *bitmap, #include #endif /* __KERNEL__ */ -#endif /* _ASM_X86_BITOPS_H */ +#endif /* ASM_X86__BITOPS_H */ diff --git a/include/asm-x86/boot.h b/include/asm-x86/boot.h index 2faed7ecb09..825de5dc867 100644 --- a/include/asm-x86/boot.h +++ b/include/asm-x86/boot.h @@ -1,5 +1,5 @@ -#ifndef _ASM_BOOT_H -#define _ASM_BOOT_H +#ifndef ASM_X86__BOOT_H +#define ASM_X86__BOOT_H /* Don't touch these, unless you really know what you're doing. */ #define DEF_INITSEG 0x9000 @@ -25,4 +25,4 @@ #define BOOT_STACK_SIZE 0x1000 #endif -#endif /* _ASM_BOOT_H */ +#endif /* ASM_X86__BOOT_H */ diff --git a/include/asm-x86/bootparam.h b/include/asm-x86/bootparam.h index ae22bdf0ab1..ccf027e2d97 100644 --- a/include/asm-x86/bootparam.h +++ b/include/asm-x86/bootparam.h @@ -1,5 +1,5 @@ -#ifndef _ASM_BOOTPARAM_H -#define _ASM_BOOTPARAM_H +#ifndef ASM_X86__BOOTPARAM_H +#define ASM_X86__BOOTPARAM_H #include #include @@ -108,4 +108,4 @@ struct boot_params { __u8 _pad9[276]; /* 0xeec */ } __attribute__((packed)); -#endif /* _ASM_BOOTPARAM_H */ +#endif /* ASM_X86__BOOTPARAM_H */ diff --git a/include/asm-x86/bug.h b/include/asm-x86/bug.h index b69aa64b82a..91ad43a54c4 100644 --- a/include/asm-x86/bug.h +++ b/include/asm-x86/bug.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_BUG_H -#define _ASM_X86_BUG_H +#ifndef ASM_X86__BUG_H +#define ASM_X86__BUG_H #ifdef CONFIG_BUG #define HAVE_ARCH_BUG @@ -36,4 +36,4 @@ do { \ #endif /* !CONFIG_BUG */ #include -#endif +#endif /* ASM_X86__BUG_H */ diff --git a/include/asm-x86/bugs.h b/include/asm-x86/bugs.h index 021cbdd5f25..4761c461d23 100644 --- a/include/asm-x86/bugs.h +++ b/include/asm-x86/bugs.h @@ -1,7 +1,7 @@ -#ifndef _ASM_X86_BUGS_H -#define _ASM_X86_BUGS_H +#ifndef ASM_X86__BUGS_H +#define ASM_X86__BUGS_H extern void check_bugs(void); int ppro_with_ram_bug(void); -#endif /* _ASM_X86_BUGS_H */ +#endif /* ASM_X86__BUGS_H */ diff --git a/include/asm-x86/byteorder.h b/include/asm-x86/byteorder.h index e02ae2d89ac..722f27d6810 100644 --- a/include/asm-x86/byteorder.h +++ b/include/asm-x86/byteorder.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_BYTEORDER_H -#define _ASM_X86_BYTEORDER_H +#ifndef ASM_X86__BYTEORDER_H +#define ASM_X86__BYTEORDER_H #include #include @@ -78,4 +78,4 @@ static inline __attribute_const__ __u32 ___arch__swab32(__u32 x) #include -#endif /* _ASM_X86_BYTEORDER_H */ +#endif /* ASM_X86__BYTEORDER_H */ diff --git a/include/asm-x86/cache.h b/include/asm-x86/cache.h index 1e0bac86f38..ea3f1cc06a9 100644 --- a/include/asm-x86/cache.h +++ b/include/asm-x86/cache.h @@ -1,5 +1,5 @@ -#ifndef _ARCH_X86_CACHE_H -#define _ARCH_X86_CACHE_H +#ifndef ASM_X86__CACHE_H +#define ASM_X86__CACHE_H /* L1 cache line size */ #define L1_CACHE_SHIFT (CONFIG_X86_L1_CACHE_SHIFT) @@ -17,4 +17,4 @@ #endif #endif -#endif +#endif /* ASM_X86__CACHE_H */ diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h index f4c0ab50d2c..59859cb28a3 100644 --- a/include/asm-x86/cacheflush.h +++ b/include/asm-x86/cacheflush.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_CACHEFLUSH_H -#define _ASM_X86_CACHEFLUSH_H +#ifndef ASM_X86__CACHEFLUSH_H +#define ASM_X86__CACHEFLUSH_H /* Keep includes the same across arches. */ #include @@ -112,4 +112,4 @@ static inline int rodata_test(void) } #endif -#endif +#endif /* ASM_X86__CACHEFLUSH_H */ diff --git a/include/asm-x86/calgary.h b/include/asm-x86/calgary.h index 67f60406e2d..933fd272f82 100644 --- a/include/asm-x86/calgary.h +++ b/include/asm-x86/calgary.h @@ -21,8 +21,8 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#ifndef _ASM_X86_64_CALGARY_H -#define _ASM_X86_64_CALGARY_H +#ifndef ASM_X86__CALGARY_H +#define ASM_X86__CALGARY_H #include #include @@ -69,4 +69,4 @@ static inline int calgary_iommu_init(void) { return 1; } static inline void detect_calgary(void) { return; } #endif -#endif /* _ASM_X86_64_CALGARY_H */ +#endif /* ASM_X86__CALGARY_H */ diff --git a/include/asm-x86/checksum_32.h b/include/asm-x86/checksum_32.h index 52bbb0d8c4c..d041e8cda22 100644 --- a/include/asm-x86/checksum_32.h +++ b/include/asm-x86/checksum_32.h @@ -1,5 +1,5 @@ -#ifndef _I386_CHECKSUM_H -#define _I386_CHECKSUM_H +#ifndef ASM_X86__CHECKSUM_32_H +#define ASM_X86__CHECKSUM_32_H #include @@ -186,4 +186,4 @@ static inline __wsum csum_and_copy_to_user(const void *src, return (__force __wsum)-1; /* invalid checksum */ } -#endif +#endif /* ASM_X86__CHECKSUM_32_H */ diff --git a/include/asm-x86/checksum_64.h b/include/asm-x86/checksum_64.h index 8bd861cc526..110f403beb8 100644 --- a/include/asm-x86/checksum_64.h +++ b/include/asm-x86/checksum_64.h @@ -1,5 +1,5 @@ -#ifndef _X86_64_CHECKSUM_H -#define _X86_64_CHECKSUM_H +#ifndef ASM_X86__CHECKSUM_64_H +#define ASM_X86__CHECKSUM_64_H /* * Checksums for x86-64 @@ -188,4 +188,4 @@ static inline unsigned add32_with_carry(unsigned a, unsigned b) return a; } -#endif +#endif /* ASM_X86__CHECKSUM_64_H */ diff --git a/include/asm-x86/cmpxchg_32.h b/include/asm-x86/cmpxchg_32.h index bf5a69d1329..0622e45cdf7 100644 --- a/include/asm-x86/cmpxchg_32.h +++ b/include/asm-x86/cmpxchg_32.h @@ -1,5 +1,5 @@ -#ifndef __ASM_CMPXCHG_H -#define __ASM_CMPXCHG_H +#ifndef ASM_X86__CMPXCHG_32_H +#define ASM_X86__CMPXCHG_32_H #include /* for LOCK_PREFIX */ @@ -341,4 +341,4 @@ extern unsigned long long cmpxchg_486_u64(volatile void *, u64, u64); #endif -#endif +#endif /* ASM_X86__CMPXCHG_32_H */ diff --git a/include/asm-x86/cmpxchg_64.h b/include/asm-x86/cmpxchg_64.h index 17463ccf816..63c1a5e61b9 100644 --- a/include/asm-x86/cmpxchg_64.h +++ b/include/asm-x86/cmpxchg_64.h @@ -1,5 +1,5 @@ -#ifndef __ASM_CMPXCHG_H -#define __ASM_CMPXCHG_H +#ifndef ASM_X86__CMPXCHG_64_H +#define ASM_X86__CMPXCHG_64_H #include /* Provides LOCK_PREFIX */ @@ -182,4 +182,4 @@ static inline unsigned long __cmpxchg_local(volatile void *ptr, cmpxchg_local((ptr), (o), (n)); \ }) -#endif +#endif /* ASM_X86__CMPXCHG_64_H */ diff --git a/include/asm-x86/compat.h b/include/asm-x86/compat.h index 1793ac317a3..6732b150949 100644 --- a/include/asm-x86/compat.h +++ b/include/asm-x86/compat.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_64_COMPAT_H -#define _ASM_X86_64_COMPAT_H +#ifndef ASM_X86__COMPAT_H +#define ASM_X86__COMPAT_H /* * Architecture specific compatibility types @@ -215,4 +215,4 @@ static inline int is_compat_task(void) return current_thread_info()->status & TS_COMPAT; } -#endif /* _ASM_X86_64_COMPAT_H */ +#endif /* ASM_X86__COMPAT_H */ diff --git a/include/asm-x86/cpu.h b/include/asm-x86/cpu.h index 73f2ea84fd7..83a115083f0 100644 --- a/include/asm-x86/cpu.h +++ b/include/asm-x86/cpu.h @@ -1,5 +1,5 @@ -#ifndef _ASM_I386_CPU_H_ -#define _ASM_I386_CPU_H_ +#ifndef ASM_X86__CPU_H +#define ASM_X86__CPU_H #include #include @@ -17,4 +17,4 @@ extern void arch_unregister_cpu(int); #endif DECLARE_PER_CPU(int, cpu_state); -#endif /* _ASM_I386_CPU_H_ */ +#endif /* ASM_X86__CPU_H */ diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 2f5a792b0ac..2f3143e4514 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -1,8 +1,8 @@ /* * Defines x86 CPU feature bits */ -#ifndef _ASM_X86_CPUFEATURE_H -#define _ASM_X86_CPUFEATURE_H +#ifndef ASM_X86__CPUFEATURE_H +#define ASM_X86__CPUFEATURE_H #include @@ -220,4 +220,4 @@ extern const char * const x86_power_flags[32]; #endif /* defined(__KERNEL__) && !defined(__ASSEMBLY__) */ -#endif /* _ASM_X86_CPUFEATURE_H */ +#endif /* ASM_X86__CPUFEATURE_H */ diff --git a/include/asm-x86/current.h b/include/asm-x86/current.h index 7515c19d498..a863ead856f 100644 --- a/include/asm-x86/current.h +++ b/include/asm-x86/current.h @@ -1,5 +1,5 @@ -#ifndef _X86_CURRENT_H -#define _X86_CURRENT_H +#ifndef ASM_X86__CURRENT_H +#define ASM_X86__CURRENT_H #ifdef CONFIG_X86_32 #include @@ -36,4 +36,4 @@ static __always_inline struct task_struct *get_current(void) #define current get_current() -#endif /* X86_CURRENT_H */ +#endif /* ASM_X86__CURRENT_H */ diff --git a/include/asm-x86/debugreg.h b/include/asm-x86/debugreg.h index c6344d572b0..ecb6907c3ea 100644 --- a/include/asm-x86/debugreg.h +++ b/include/asm-x86/debugreg.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_DEBUGREG_H -#define _ASM_X86_DEBUGREG_H +#ifndef ASM_X86__DEBUGREG_H +#define ASM_X86__DEBUGREG_H /* Indicate the register numbers for a number of the specific @@ -67,4 +67,4 @@ #define DR_LOCAL_SLOWDOWN (0x100) /* Local slow the pipeline */ #define DR_GLOBAL_SLOWDOWN (0x200) /* Global slow the pipeline */ -#endif +#endif /* ASM_X86__DEBUGREG_H */ diff --git a/include/asm-x86/delay.h b/include/asm-x86/delay.h index 409a649204a..8a0da95b4fc 100644 --- a/include/asm-x86/delay.h +++ b/include/asm-x86/delay.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_DELAY_H -#define _ASM_X86_DELAY_H +#ifndef ASM_X86__DELAY_H +#define ASM_X86__DELAY_H /* * Copyright (C) 1993 Linus Torvalds @@ -28,4 +28,4 @@ extern void __delay(unsigned long loops); void use_tsc_delay(void); -#endif /* _ASM_X86_DELAY_H */ +#endif /* ASM_X86__DELAY_H */ diff --git a/include/asm-x86/desc_defs.h b/include/asm-x86/desc_defs.h index f7bacf357da..b881db664b4 100644 --- a/include/asm-x86/desc_defs.h +++ b/include/asm-x86/desc_defs.h @@ -1,6 +1,6 @@ /* Written 2000 by Andi Kleen */ -#ifndef __ARCH_DESC_DEFS_H -#define __ARCH_DESC_DEFS_H +#ifndef ASM_X86__DESC_DEFS_H +#define ASM_X86__DESC_DEFS_H /* * Segment descriptor structure definitions, usable from both x86_64 and i386 @@ -92,4 +92,4 @@ struct desc_ptr { #endif /* !__ASSEMBLY__ */ -#endif +#endif /* ASM_X86__DESC_DEFS_H */ diff --git a/include/asm-x86/device.h b/include/asm-x86/device.h index 87a715367a1..dbe88554389 100644 --- a/include/asm-x86/device.h +++ b/include/asm-x86/device.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_DEVICE_H -#define _ASM_X86_DEVICE_H +#ifndef ASM_X86__DEVICE_H +#define ASM_X86__DEVICE_H struct dev_archdata { #ifdef CONFIG_ACPI @@ -10,4 +10,4 @@ struct dev_archdata { #endif }; -#endif /* _ASM_X86_DEVICE_H */ +#endif /* ASM_X86__DEVICE_H */ diff --git a/include/asm-x86/div64.h b/include/asm-x86/div64.h index 9a2d644c08e..f9530f23f1d 100644 --- a/include/asm-x86/div64.h +++ b/include/asm-x86/div64.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_DIV64_H -#define _ASM_X86_DIV64_H +#ifndef ASM_X86__DIV64_H +#define ASM_X86__DIV64_H #ifdef CONFIG_X86_32 @@ -57,4 +57,4 @@ static inline u64 div_u64_rem(u64 dividend, u32 divisor, u32 *remainder) # include #endif /* CONFIG_X86_32 */ -#endif /* _ASM_X86_DIV64_H */ +#endif /* ASM_X86__DIV64_H */ diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index c2ddd3d1b88..71b6f7d22e9 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -1,5 +1,5 @@ -#ifndef _ASM_DMA_MAPPING_H_ -#define _ASM_DMA_MAPPING_H_ +#ifndef ASM_X86__DMA_MAPPING_H +#define ASM_X86__DMA_MAPPING_H /* * IOMMU interface. See Documentation/DMA-mapping.txt and DMA-API.txt for @@ -233,4 +233,4 @@ extern void * dma_mark_declared_memory_occupied(struct device *dev, dma_addr_t device_addr, size_t size); #endif /* CONFIG_X86_32 */ -#endif +#endif /* ASM_X86__DMA_MAPPING_H */ diff --git a/include/asm-x86/dma.h b/include/asm-x86/dma.h index ca1098a7e58..c9f7a4eec55 100644 --- a/include/asm-x86/dma.h +++ b/include/asm-x86/dma.h @@ -5,8 +5,8 @@ * and John Boyd, Nov. 1992. */ -#ifndef _ASM_X86_DMA_H -#define _ASM_X86_DMA_H +#ifndef ASM_X86__DMA_H +#define ASM_X86__DMA_H #include /* And spinlocks */ #include /* need byte IO */ @@ -315,4 +315,4 @@ extern int isa_dma_bridge_buggy; #define isa_dma_bridge_buggy (0) #endif -#endif /* _ASM_X86_DMA_H */ +#endif /* ASM_X86__DMA_H */ diff --git a/include/asm-x86/dmi.h b/include/asm-x86/dmi.h index 58a86571fe0..1cff6fe81fa 100644 --- a/include/asm-x86/dmi.h +++ b/include/asm-x86/dmi.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_DMI_H -#define _ASM_X86_DMI_H +#ifndef ASM_X86__DMI_H +#define ASM_X86__DMI_H #include @@ -23,4 +23,4 @@ static inline void *dmi_alloc(unsigned len) #define dmi_ioremap early_ioremap #define dmi_iounmap early_iounmap -#endif +#endif /* ASM_X86__DMI_H */ diff --git a/include/asm-x86/ds.h b/include/asm-x86/ds.h index 7881368142f..6b27c686fa1 100644 --- a/include/asm-x86/ds.h +++ b/include/asm-x86/ds.h @@ -17,8 +17,8 @@ * Markus Metzger , Dec 2007 */ -#ifndef _ASM_X86_DS_H -#define _ASM_X86_DS_H +#ifndef ASM_X86__DS_H +#define ASM_X86__DS_H #include #include @@ -69,4 +69,4 @@ extern int ds_write_bts(void *, const struct bts_struct *); extern unsigned long ds_debugctl_mask(void); extern void __cpuinit ds_init_intel(struct cpuinfo_x86 *c); -#endif /* _ASM_X86_DS_H */ +#endif /* ASM_X86__DS_H */ diff --git a/include/asm-x86/dwarf2.h b/include/asm-x86/dwarf2.h index 738bb9fb3e5..21d1bc32ad7 100644 --- a/include/asm-x86/dwarf2.h +++ b/include/asm-x86/dwarf2.h @@ -1,5 +1,5 @@ -#ifndef _DWARF2_H -#define _DWARF2_H +#ifndef ASM_X86__DWARF2_H +#define ASM_X86__DWARF2_H #ifndef __ASSEMBLY__ #warning "asm/dwarf2.h should be only included in pure assembly files" @@ -58,4 +58,4 @@ #endif -#endif +#endif /* ASM_X86__DWARF2_H */ diff --git a/include/asm-x86/e820.h b/include/asm-x86/e820.h index 16a31e2c7c5..38741ac7690 100644 --- a/include/asm-x86/e820.h +++ b/include/asm-x86/e820.h @@ -1,5 +1,5 @@ -#ifndef __ASM_E820_H -#define __ASM_E820_H +#ifndef ASM_X86__E820_H +#define ASM_X86__E820_H #define E820MAP 0x2d0 /* our map */ #define E820MAX 128 /* number of entries in E820MAP */ @@ -140,4 +140,4 @@ extern char *memory_setup(void); #define HIGH_MEMORY (1024*1024) #endif /* __KERNEL__ */ -#endif /* __ASM_E820_H */ +#endif /* ASM_X86__E820_H */ diff --git a/include/asm-x86/edac.h b/include/asm-x86/edac.h index a8088f63a30..9493c5b27bb 100644 --- a/include/asm-x86/edac.h +++ b/include/asm-x86/edac.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_EDAC_H -#define _ASM_X86_EDAC_H +#ifndef ASM_X86__EDAC_H +#define ASM_X86__EDAC_H /* ECC atomic, DMA, SMP and interrupt safe scrub function */ @@ -15,4 +15,4 @@ static inline void atomic_scrub(void *va, u32 size) asm volatile("lock; addl $0, %0"::"m" (*virt_addr)); } -#endif +#endif /* ASM_X86__EDAC_H */ diff --git a/include/asm-x86/efi.h b/include/asm-x86/efi.h index 7ed2bd7a7f5..69c7b7ab43e 100644 --- a/include/asm-x86/efi.h +++ b/include/asm-x86/efi.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_EFI_H -#define _ASM_X86_EFI_H +#ifndef ASM_X86__EFI_H +#define ASM_X86__EFI_H #ifdef CONFIG_X86_32 @@ -94,4 +94,4 @@ extern void efi_reserve_early(void); extern void efi_call_phys_prelog(void); extern void efi_call_phys_epilog(void); -#endif +#endif /* ASM_X86__EFI_H */ diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h index 7be4733c793..cd678b2d6a7 100644 --- a/include/asm-x86/elf.h +++ b/include/asm-x86/elf.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_ELF_H -#define _ASM_X86_ELF_H +#ifndef ASM_X86__ELF_H +#define ASM_X86__ELF_H /* * ELF register definitions.. @@ -332,4 +332,4 @@ extern int syscall32_setup_pages(struct linux_binprm *, int exstack); extern unsigned long arch_randomize_brk(struct mm_struct *mm); #define arch_randomize_brk arch_randomize_brk -#endif +#endif /* ASM_X86__ELF_H */ diff --git a/include/asm-x86/emergency-restart.h b/include/asm-x86/emergency-restart.h index 8e6aef19f8f..190d0d8b71e 100644 --- a/include/asm-x86/emergency-restart.h +++ b/include/asm-x86/emergency-restart.h @@ -1,5 +1,5 @@ -#ifndef _ASM_EMERGENCY_RESTART_H -#define _ASM_EMERGENCY_RESTART_H +#ifndef ASM_X86__EMERGENCY_RESTART_H +#define ASM_X86__EMERGENCY_RESTART_H enum reboot_type { BOOT_TRIPLE = 't', @@ -15,4 +15,4 @@ extern enum reboot_type reboot_type; extern void machine_emergency_restart(void); -#endif /* _ASM_EMERGENCY_RESTART_H */ +#endif /* ASM_X86__EMERGENCY_RESTART_H */ diff --git a/include/asm-x86/fb.h b/include/asm-x86/fb.h index 53018464aea..aca38dbd9a6 100644 --- a/include/asm-x86/fb.h +++ b/include/asm-x86/fb.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_FB_H -#define _ASM_X86_FB_H +#ifndef ASM_X86__FB_H +#define ASM_X86__FB_H #include #include @@ -18,4 +18,4 @@ extern int fb_is_primary_device(struct fb_info *info); static inline int fb_is_primary_device(struct fb_info *info) { return 0; } #endif -#endif /* _ASM_X86_FB_H */ +#endif /* ASM_X86__FB_H */ diff --git a/include/asm-x86/fixmap.h b/include/asm-x86/fixmap.h index 44d4f821734..78e33a1bc59 100644 --- a/include/asm-x86/fixmap.h +++ b/include/asm-x86/fixmap.h @@ -1,5 +1,5 @@ -#ifndef _ASM_FIXMAP_H -#define _ASM_FIXMAP_H +#ifndef ASM_X86__FIXMAP_H +#define ASM_X86__FIXMAP_H #ifdef CONFIG_X86_32 # include "fixmap_32.h" @@ -65,4 +65,4 @@ static inline unsigned long virt_to_fix(const unsigned long vaddr) BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START); return __virt_to_fix(vaddr); } -#endif +#endif /* ASM_X86__FIXMAP_H */ diff --git a/include/asm-x86/fixmap_32.h b/include/asm-x86/fixmap_32.h index f1ac2b2167d..784e3e75986 100644 --- a/include/asm-x86/fixmap_32.h +++ b/include/asm-x86/fixmap_32.h @@ -10,8 +10,8 @@ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999 */ -#ifndef _ASM_FIXMAP_32_H -#define _ASM_FIXMAP_32_H +#ifndef ASM_X86__FIXMAP_32_H +#define ASM_X86__FIXMAP_32_H /* used by vmalloc.c, vsyscall.lds.S. @@ -120,4 +120,4 @@ extern void reserve_top_address(unsigned long reserve); #define FIXADDR_BOOT_START (FIXADDR_TOP - __FIXADDR_BOOT_SIZE) #endif /* !__ASSEMBLY__ */ -#endif +#endif /* ASM_X86__FIXMAP_32_H */ diff --git a/include/asm-x86/fixmap_64.h b/include/asm-x86/fixmap_64.h index 00f3d74a052..dafb24bc042 100644 --- a/include/asm-x86/fixmap_64.h +++ b/include/asm-x86/fixmap_64.h @@ -8,8 +8,8 @@ * Copyright (C) 1998 Ingo Molnar */ -#ifndef _ASM_FIXMAP_64_H -#define _ASM_FIXMAP_64_H +#ifndef ASM_X86__FIXMAP_64_H +#define ASM_X86__FIXMAP_64_H #include #include @@ -80,4 +80,4 @@ enum fixed_addresses { #define FIXADDR_USER_START ((unsigned long)VSYSCALL32_VSYSCALL) #define FIXADDR_USER_END (FIXADDR_USER_START + PAGE_SIZE) -#endif +#endif /* ASM_X86__FIXMAP_64_H */ diff --git a/include/asm-x86/floppy.h b/include/asm-x86/floppy.h index dbe82a5c5ea..7d83a3a83e3 100644 --- a/include/asm-x86/floppy.h +++ b/include/asm-x86/floppy.h @@ -7,8 +7,8 @@ * * Copyright (C) 1995 */ -#ifndef _ASM_X86_FLOPPY_H -#define _ASM_X86_FLOPPY_H +#ifndef ASM_X86__FLOPPY_H +#define ASM_X86__FLOPPY_H #include @@ -278,4 +278,4 @@ static int FDC2 = -1; #define EXTRA_FLOPPY_PARAMS -#endif /* _ASM_X86_FLOPPY_H */ +#endif /* ASM_X86__FLOPPY_H */ diff --git a/include/asm-x86/ftrace.h b/include/asm-x86/ftrace.h index 5c68b32ee1c..be0e004ad14 100644 --- a/include/asm-x86/ftrace.h +++ b/include/asm-x86/ftrace.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_FTRACE -#define _ASM_X86_FTRACE +#ifndef ASM_X86__FTRACE_H +#define ASM_X86__FTRACE_H #ifdef CONFIG_FTRACE #define MCOUNT_ADDR ((long)(mcount)) @@ -11,4 +11,4 @@ extern void mcount(void); #endif /* CONFIG_FTRACE */ -#endif /* _ASM_X86_FTRACE */ +#endif /* ASM_X86__FTRACE_H */ diff --git a/include/asm-x86/futex.h b/include/asm-x86/futex.h index e7a76b37b33..45dc24d8418 100644 --- a/include/asm-x86/futex.h +++ b/include/asm-x86/futex.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_FUTEX_H -#define _ASM_X86_FUTEX_H +#ifndef ASM_X86__FUTEX_H +#define ASM_X86__FUTEX_H #ifdef __KERNEL__ @@ -137,4 +137,4 @@ static inline int futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, } #endif -#endif +#endif /* ASM_X86__FUTEX_H */ diff --git a/include/asm-x86/gart.h b/include/asm-x86/gart.h index 3f62a83887f..07f44584414 100644 --- a/include/asm-x86/gart.h +++ b/include/asm-x86/gart.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X8664_GART_H -#define _ASM_X8664_GART_H 1 +#ifndef ASM_X86__GART_H +#define ASM_X86__GART_H #include @@ -68,4 +68,4 @@ static inline int aperture_valid(u64 aper_base, u32 aper_size, u32 min_size) return 1; } -#endif +#endif /* ASM_X86__GART_H */ diff --git a/include/asm-x86/genapic_32.h b/include/asm-x86/genapic_32.h index b02ea6e17de..4904c672e4f 100644 --- a/include/asm-x86/genapic_32.h +++ b/include/asm-x86/genapic_32.h @@ -1,5 +1,5 @@ -#ifndef _ASM_GENAPIC_H -#define _ASM_GENAPIC_H 1 +#ifndef ASM_X86__GENAPIC_32_H +#define ASM_X86__GENAPIC_32_H #include @@ -120,4 +120,4 @@ enum uv_system_type {UV_NONE, UV_LEGACY_APIC, UV_X2APIC, UV_NON_UNIQUE_APIC}; #define uv_wakeup_secondary(a, b) 1 -#endif +#endif /* ASM_X86__GENAPIC_32_H */ diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h index 0f8504627c4..381a09d19d5 100644 --- a/include/asm-x86/genapic_64.h +++ b/include/asm-x86/genapic_64.h @@ -1,5 +1,5 @@ -#ifndef _ASM_GENAPIC_H -#define _ASM_GENAPIC_H 1 +#ifndef ASM_X86__GENAPIC_64_H +#define ASM_X86__GENAPIC_64_H /* * Copyright 2004 James Cleverdon, IBM. @@ -46,4 +46,4 @@ extern int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip); extern void setup_apic_routing(void); -#endif +#endif /* ASM_X86__GENAPIC_64_H */ diff --git a/include/asm-x86/geode.h b/include/asm-x86/geode.h index bb06027fc83..1ef738e01a0 100644 --- a/include/asm-x86/geode.h +++ b/include/asm-x86/geode.h @@ -7,8 +7,8 @@ * as published by the Free Software Foundation. */ -#ifndef _ASM_GEODE_H_ -#define _ASM_GEODE_H_ +#ifndef ASM_X86__GEODE_H +#define ASM_X86__GEODE_H #include #include @@ -249,4 +249,4 @@ extern int __init mfgpt_timer_setup(void); static inline int mfgpt_timer_setup(void) { return 0; } #endif -#endif +#endif /* ASM_X86__GEODE_H */ diff --git a/include/asm-x86/gpio.h b/include/asm-x86/gpio.h index ff87fca0caf..f269ff9cd40 100644 --- a/include/asm-x86/gpio.h +++ b/include/asm-x86/gpio.h @@ -1,6 +1,6 @@ -#ifndef _ASM_I386_GPIO_H -#define _ASM_I386_GPIO_H +#ifndef ASM_X86__GPIO_H +#define ASM_X86__GPIO_H #include -#endif /* _ASM_I386_GPIO_H */ +#endif /* ASM_X86__GPIO_H */ diff --git a/include/asm-x86/hardirq_32.h b/include/asm-x86/hardirq_32.h index 4f85f0f4b56..700fe230d91 100644 --- a/include/asm-x86/hardirq_32.h +++ b/include/asm-x86/hardirq_32.h @@ -1,5 +1,5 @@ -#ifndef __ASM_HARDIRQ_H -#define __ASM_HARDIRQ_H +#ifndef ASM_X86__HARDIRQ_32_H +#define ASM_X86__HARDIRQ_32_H #include #include @@ -25,4 +25,4 @@ DECLARE_PER_CPU(irq_cpustat_t, irq_stat); void ack_bad_irq(unsigned int irq); #include -#endif /* __ASM_HARDIRQ_H */ +#endif /* ASM_X86__HARDIRQ_32_H */ diff --git a/include/asm-x86/hardirq_64.h b/include/asm-x86/hardirq_64.h index 95d5e090ed8..f8bd2919a8c 100644 --- a/include/asm-x86/hardirq_64.h +++ b/include/asm-x86/hardirq_64.h @@ -1,5 +1,5 @@ -#ifndef __ASM_HARDIRQ_H -#define __ASM_HARDIRQ_H +#ifndef ASM_X86__HARDIRQ_64_H +#define ASM_X86__HARDIRQ_64_H #include #include @@ -20,4 +20,4 @@ extern void ack_bad_irq(unsigned int irq); -#endif /* __ASM_HARDIRQ_H */ +#endif /* ASM_X86__HARDIRQ_64_H */ diff --git a/include/asm-x86/highmem.h b/include/asm-x86/highmem.h index 4514b16cc72..bc3f6a28031 100644 --- a/include/asm-x86/highmem.h +++ b/include/asm-x86/highmem.h @@ -15,8 +15,8 @@ * Copyright (C) 1999 Ingo Molnar */ -#ifndef _ASM_HIGHMEM_H -#define _ASM_HIGHMEM_H +#ifndef ASM_X86__HIGHMEM_H +#define ASM_X86__HIGHMEM_H #ifdef __KERNEL__ @@ -79,4 +79,4 @@ extern void add_highpages_with_active_regions(int nid, unsigned long start_pfn, #endif /* __KERNEL__ */ -#endif /* _ASM_HIGHMEM_H */ +#endif /* ASM_X86__HIGHMEM_H */ diff --git a/include/asm-x86/hpet.h b/include/asm-x86/hpet.h index 82f1ac641bd..cbbbb6d4dd3 100644 --- a/include/asm-x86/hpet.h +++ b/include/asm-x86/hpet.h @@ -1,5 +1,5 @@ -#ifndef ASM_X86_HPET_H -#define ASM_X86_HPET_H +#ifndef ASM_X86__HPET_H +#define ASM_X86__HPET_H #ifdef CONFIG_HPET_TIMER @@ -90,4 +90,4 @@ static inline int is_hpet_enabled(void) { return 0; } #define hpet_readl(a) 0 #endif -#endif /* ASM_X86_HPET_H */ +#endif /* ASM_X86__HPET_H */ diff --git a/include/asm-x86/hugetlb.h b/include/asm-x86/hugetlb.h index 14171a4924f..0e0a0828182 100644 --- a/include/asm-x86/hugetlb.h +++ b/include/asm-x86/hugetlb.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_HUGETLB_H -#define _ASM_X86_HUGETLB_H +#ifndef ASM_X86__HUGETLB_H +#define ASM_X86__HUGETLB_H #include @@ -88,4 +88,4 @@ static inline void arch_release_hugepage(struct page *page) { } -#endif /* _ASM_X86_HUGETLB_H */ +#endif /* ASM_X86__HUGETLB_H */ diff --git a/include/asm-x86/hw_irq.h b/include/asm-x86/hw_irq.h index 77ba51df566..3f3c465b915 100644 --- a/include/asm-x86/hw_irq.h +++ b/include/asm-x86/hw_irq.h @@ -1,5 +1,5 @@ -#ifndef _ASM_HW_IRQ_H -#define _ASM_HW_IRQ_H +#ifndef ASM_X86__HW_IRQ_H +#define ASM_X86__HW_IRQ_H /* * (C) 1992, 1993 Linus Torvalds, (C) 1997 Ingo Molnar @@ -104,4 +104,4 @@ extern void setup_vector_irq(int cpu); #endif /* !ASSEMBLY_ */ -#endif +#endif /* ASM_X86__HW_IRQ_H */ diff --git a/include/asm-x86/hypertransport.h b/include/asm-x86/hypertransport.h index d2bbd238b3e..cc011a3bc1c 100644 --- a/include/asm-x86/hypertransport.h +++ b/include/asm-x86/hypertransport.h @@ -1,5 +1,5 @@ -#ifndef ASM_HYPERTRANSPORT_H -#define ASM_HYPERTRANSPORT_H +#ifndef ASM_X86__HYPERTRANSPORT_H +#define ASM_X86__HYPERTRANSPORT_H /* * Constants for x86 Hypertransport Interrupts. @@ -42,4 +42,4 @@ #define HT_IRQ_HIGH_DEST_ID(v) \ ((((v) >> 8) << HT_IRQ_HIGH_DEST_ID_SHIFT) & HT_IRQ_HIGH_DEST_ID_MASK) -#endif /* ASM_HYPERTRANSPORT_H */ +#endif /* ASM_X86__HYPERTRANSPORT_H */ diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h index 37672f79dcc..3e83cbb728a 100644 --- a/include/asm-x86/i387.h +++ b/include/asm-x86/i387.h @@ -7,8 +7,8 @@ * x86-64 work by Andi Kleen 2002 */ -#ifndef _ASM_X86_I387_H -#define _ASM_X86_I387_H +#ifndef ASM_X86__I387_H +#define ASM_X86__I387_H #include #include @@ -360,4 +360,4 @@ static inline unsigned short get_fpu_mxcsr(struct task_struct *tsk) } } -#endif /* _ASM_X86_I387_H */ +#endif /* ASM_X86__I387_H */ diff --git a/include/asm-x86/i8253.h b/include/asm-x86/i8253.h index b51c0487fc4..15a5b530044 100644 --- a/include/asm-x86/i8253.h +++ b/include/asm-x86/i8253.h @@ -1,5 +1,5 @@ -#ifndef __ASM_I8253_H__ -#define __ASM_I8253_H__ +#ifndef ASM_X86__I8253_H +#define ASM_X86__I8253_H /* i8253A PIT registers */ #define PIT_MODE 0x43 @@ -15,4 +15,4 @@ extern void setup_pit_timer(void); #define inb_pit inb_p #define outb_pit outb_p -#endif /* __ASM_I8253_H__ */ +#endif /* ASM_X86__I8253_H */ diff --git a/include/asm-x86/i8259.h b/include/asm-x86/i8259.h index 2f98df91f1f..c586559a695 100644 --- a/include/asm-x86/i8259.h +++ b/include/asm-x86/i8259.h @@ -1,5 +1,5 @@ -#ifndef __ASM_I8259_H__ -#define __ASM_I8259_H__ +#ifndef ASM_X86__I8259_H +#define ASM_X86__I8259_H #include @@ -57,4 +57,4 @@ static inline void outb_pic(unsigned char value, unsigned int port) extern struct irq_chip i8259A_chip; -#endif /* __ASM_I8259_H__ */ +#endif /* ASM_X86__I8259_H */ diff --git a/include/asm-x86/ia32.h b/include/asm-x86/ia32.h index 55d3abe5276..f932f7ad51d 100644 --- a/include/asm-x86/ia32.h +++ b/include/asm-x86/ia32.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_64_IA32_H -#define _ASM_X86_64_IA32_H +#ifndef ASM_X86__IA32_H +#define ASM_X86__IA32_H #ifdef CONFIG_IA32_EMULATION @@ -167,4 +167,4 @@ extern void ia32_pick_mmap_layout(struct mm_struct *mm); #endif /* !CONFIG_IA32_SUPPORT */ -#endif +#endif /* ASM_X86__IA32_H */ diff --git a/include/asm-x86/ia32_unistd.h b/include/asm-x86/ia32_unistd.h index 61cea9e7c5c..dbd887d8a5a 100644 --- a/include/asm-x86/ia32_unistd.h +++ b/include/asm-x86/ia32_unistd.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_64_IA32_UNISTD_H_ -#define _ASM_X86_64_IA32_UNISTD_H_ +#ifndef ASM_X86__IA32_UNISTD_H +#define ASM_X86__IA32_UNISTD_H /* * This file contains the system call numbers of the ia32 port, @@ -15,4 +15,4 @@ #define __NR_ia32_sigreturn 119 #define __NR_ia32_rt_sigreturn 173 -#endif /* _ASM_X86_64_IA32_UNISTD_H_ */ +#endif /* ASM_X86__IA32_UNISTD_H */ diff --git a/include/asm-x86/ide.h b/include/asm-x86/ide.h index cf9c98e5bdb..8bf3cdc6bb4 100644 --- a/include/asm-x86/ide.h +++ b/include/asm-x86/ide.h @@ -6,8 +6,8 @@ * This file contains the i386 architecture specific IDE code. */ -#ifndef __ASMi386_IDE_H -#define __ASMi386_IDE_H +#ifndef ASM_X86__IDE_H +#define ASM_X86__IDE_H #ifdef __KERNEL__ @@ -62,4 +62,4 @@ static __inline__ unsigned long ide_default_io_base(int index) #endif /* __KERNEL__ */ -#endif /* __ASMi386_IDE_H */ +#endif /* ASM_X86__IDE_H */ diff --git a/include/asm-x86/idle.h b/include/asm-x86/idle.h index d240e5b30a4..dc9c7944847 100644 --- a/include/asm-x86/idle.h +++ b/include/asm-x86/idle.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_64_IDLE_H -#define _ASM_X86_64_IDLE_H 1 +#ifndef ASM_X86__IDLE_H +#define ASM_X86__IDLE_H #define IDLE_START 1 #define IDLE_END 2 @@ -10,4 +10,4 @@ void idle_notifier_register(struct notifier_block *n); void enter_idle(void); void exit_idle(void); -#endif +#endif /* ASM_X86__IDLE_H */ diff --git a/include/asm-x86/intel_arch_perfmon.h b/include/asm-x86/intel_arch_perfmon.h index fa0fd068bc2..07c03c6c9a1 100644 --- a/include/asm-x86/intel_arch_perfmon.h +++ b/include/asm-x86/intel_arch_perfmon.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_INTEL_ARCH_PERFMON_H -#define _ASM_X86_INTEL_ARCH_PERFMON_H +#ifndef ASM_X86__INTEL_ARCH_PERFMON_H +#define ASM_X86__INTEL_ARCH_PERFMON_H #define MSR_ARCH_PERFMON_PERFCTR0 0xc1 #define MSR_ARCH_PERFMON_PERFCTR1 0xc2 @@ -28,4 +28,4 @@ union cpuid10_eax { unsigned int full; }; -#endif /* _ASM_X86_INTEL_ARCH_PERFMON_H */ +#endif /* ASM_X86__INTEL_ARCH_PERFMON_H */ diff --git a/include/asm-x86/io.h b/include/asm-x86/io.h index bf5d629b3a3..1b75a43bb6c 100644 --- a/include/asm-x86/io.h +++ b/include/asm-x86/io.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_IO_H -#define _ASM_X86_IO_H +#ifndef ASM_X86__IO_H +#define ASM_X86__IO_H #define ARCH_HAS_IOREMAP_WC @@ -99,4 +99,4 @@ extern void early_iounmap(void *addr, unsigned long size); extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys); -#endif /* _ASM_X86_IO_H */ +#endif /* ASM_X86__IO_H */ diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h index 4df44ed5407..d9d1e3f269a 100644 --- a/include/asm-x86/io_32.h +++ b/include/asm-x86/io_32.h @@ -1,5 +1,5 @@ -#ifndef _ASM_IO_H -#define _ASM_IO_H +#ifndef ASM_X86__IO_32_H +#define ASM_X86__IO_32_H #include #include @@ -279,4 +279,4 @@ BUILDIO(b, b, char) BUILDIO(w, w, short) BUILDIO(l, , int) -#endif +#endif /* ASM_X86__IO_32_H */ diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h index ddd8058a502..00ac4729e52 100644 --- a/include/asm-x86/io_64.h +++ b/include/asm-x86/io_64.h @@ -1,5 +1,5 @@ -#ifndef _ASM_IO_H -#define _ASM_IO_H +#ifndef ASM_X86__IO_64_H +#define ASM_X86__IO_64_H /* @@ -243,4 +243,4 @@ extern int iommu_bio_merge; #endif /* __KERNEL__ */ -#endif +#endif /* ASM_X86__IO_64_H */ diff --git a/include/asm-x86/io_apic.h b/include/asm-x86/io_apic.h index 14f82bbcb5f..be62847ab07 100644 --- a/include/asm-x86/io_apic.h +++ b/include/asm-x86/io_apic.h @@ -1,5 +1,5 @@ -#ifndef __ASM_IO_APIC_H -#define __ASM_IO_APIC_H +#ifndef ASM_X86__IO_APIC_H +#define ASM_X86__IO_APIC_H #include #include @@ -189,4 +189,4 @@ static const int timer_through_8259 = 0; static inline void ioapic_init_mappings(void) { } #endif -#endif +#endif /* ASM_X86__IO_APIC_H */ diff --git a/include/asm-x86/ioctls.h b/include/asm-x86/ioctls.h index c0c338bd406..33660351239 100644 --- a/include/asm-x86/ioctls.h +++ b/include/asm-x86/ioctls.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_IOCTLS_H -#define _ASM_X86_IOCTLS_H +#ifndef ASM_X86__IOCTLS_H +#define ASM_X86__IOCTLS_H #include @@ -85,4 +85,4 @@ #define TIOCSER_TEMT 0x01 /* Transmitter physically empty */ -#endif +#endif /* ASM_X86__IOCTLS_H */ diff --git a/include/asm-x86/iommu.h b/include/asm-x86/iommu.h index d63166fb3ab..4f3d212d827 100644 --- a/include/asm-x86/iommu.h +++ b/include/asm-x86/iommu.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X8664_IOMMU_H -#define _ASM_X8664_IOMMU_H 1 +#ifndef ASM_X86__IOMMU_H +#define ASM_X86__IOMMU_H extern void pci_iommu_shutdown(void); extern void no_iommu_init(void); @@ -39,4 +39,4 @@ static inline void gart_iommu_hole_init(void) } #endif -#endif +#endif /* ASM_X86__IOMMU_H */ diff --git a/include/asm-x86/ipcbuf.h b/include/asm-x86/ipcbuf.h index ee678fd5159..910304fbdc8 100644 --- a/include/asm-x86/ipcbuf.h +++ b/include/asm-x86/ipcbuf.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_IPCBUF_H -#define _ASM_X86_IPCBUF_H +#ifndef ASM_X86__IPCBUF_H +#define ASM_X86__IPCBUF_H /* * The ipc64_perm structure for x86 architecture. @@ -25,4 +25,4 @@ struct ipc64_perm { unsigned long __unused2; }; -#endif /* _ASM_X86_IPCBUF_H */ +#endif /* ASM_X86__IPCBUF_H */ diff --git a/include/asm-x86/ipi.h b/include/asm-x86/ipi.h index 196d63c28aa..63390ea87d1 100644 --- a/include/asm-x86/ipi.h +++ b/include/asm-x86/ipi.h @@ -1,5 +1,5 @@ -#ifndef __ASM_IPI_H -#define __ASM_IPI_H +#ifndef ASM_X86__IPI_H +#define ASM_X86__IPI_H /* * Copyright 2004 James Cleverdon, IBM. @@ -129,4 +129,4 @@ static inline void send_IPI_mask_sequence(cpumask_t mask, int vector) local_irq_restore(flags); } -#endif /* __ASM_IPI_H */ +#endif /* ASM_X86__IPI_H */ diff --git a/include/asm-x86/irq.h b/include/asm-x86/irq.h index 1a292575731..1e5f2909c1d 100644 --- a/include/asm-x86/irq.h +++ b/include/asm-x86/irq.h @@ -1,5 +1,5 @@ -#ifndef _ASM_IRQ_H -#define _ASM_IRQ_H +#ifndef ASM_X86__IRQ_H +#define ASM_X86__IRQ_H /* * (C) 1992, 1993 Linus Torvalds, (C) 1997 Ingo Molnar * @@ -47,4 +47,4 @@ extern void native_init_IRQ(void); /* Interrupt vector management */ extern DECLARE_BITMAP(used_vectors, NR_VECTORS); -#endif /* _ASM_IRQ_H */ +#endif /* ASM_X86__IRQ_H */ diff --git a/include/asm-x86/irq_regs_32.h b/include/asm-x86/irq_regs_32.h index 3368b20c0b4..316a3b25887 100644 --- a/include/asm-x86/irq_regs_32.h +++ b/include/asm-x86/irq_regs_32.h @@ -4,8 +4,8 @@ * * Jeremy Fitzhardinge */ -#ifndef _ASM_I386_IRQ_REGS_H -#define _ASM_I386_IRQ_REGS_H +#ifndef ASM_X86__IRQ_REGS_32_H +#define ASM_X86__IRQ_REGS_32_H #include @@ -26,4 +26,4 @@ static inline struct pt_regs *set_irq_regs(struct pt_regs *new_regs) return old_regs; } -#endif /* _ASM_I386_IRQ_REGS_H */ +#endif /* ASM_X86__IRQ_REGS_32_H */ diff --git a/include/asm-x86/irq_vectors.h b/include/asm-x86/irq_vectors.h index 90b1d1f12f0..646d59f5ebf 100644 --- a/include/asm-x86/irq_vectors.h +++ b/include/asm-x86/irq_vectors.h @@ -1,5 +1,5 @@ -#ifndef _ASM_IRQ_VECTORS_H -#define _ASM_IRQ_VECTORS_H +#ifndef ASM_X86__IRQ_VECTORS_H +#define ASM_X86__IRQ_VECTORS_H #include @@ -170,4 +170,4 @@ #define VIC_CPU_BOOT_ERRATA_CPI (VIC_CPI_LEVEL0 + 8) -#endif /* _ASM_IRQ_VECTORS_H */ +#endif /* ASM_X86__IRQ_VECTORS_H */ diff --git a/include/asm-x86/ist.h b/include/asm-x86/ist.h index 6ec6ceed95a..35a2fe9bc92 100644 --- a/include/asm-x86/ist.h +++ b/include/asm-x86/ist.h @@ -1,5 +1,5 @@ -#ifndef _ASM_IST_H -#define _ASM_IST_H +#ifndef ASM_X86__IST_H +#define ASM_X86__IST_H /* * Include file for the interface to IST BIOS @@ -31,4 +31,4 @@ struct ist_info { extern struct ist_info ist_info; #endif /* __KERNEL__ */ -#endif /* _ASM_IST_H */ +#endif /* ASM_X86__IST_H */ diff --git a/include/asm-x86/k8.h b/include/asm-x86/k8.h index 452e2b696ff..2bbaf4370a5 100644 --- a/include/asm-x86/k8.h +++ b/include/asm-x86/k8.h @@ -1,5 +1,5 @@ -#ifndef _ASM_K8_H -#define _ASM_K8_H 1 +#ifndef ASM_X86__K8_H +#define ASM_X86__K8_H #include @@ -12,4 +12,4 @@ extern int cache_k8_northbridges(void); extern void k8_flush_garts(void); extern int k8_scan_nodes(unsigned long start, unsigned long end); -#endif +#endif /* ASM_X86__K8_H */ diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h index 96651bb59ba..5ec3ad3e825 100644 --- a/include/asm-x86/kdebug.h +++ b/include/asm-x86/kdebug.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_KDEBUG_H -#define _ASM_X86_KDEBUG_H +#ifndef ASM_X86__KDEBUG_H +#define ASM_X86__KDEBUG_H #include @@ -35,4 +35,4 @@ extern void show_regs(struct pt_regs *regs); extern unsigned long oops_begin(void); extern void oops_end(unsigned long, struct pt_regs *, int signr); -#endif +#endif /* ASM_X86__KDEBUG_H */ diff --git a/include/asm-x86/kexec.h b/include/asm-x86/kexec.h index 8f855a15f64..262b63ec911 100644 --- a/include/asm-x86/kexec.h +++ b/include/asm-x86/kexec.h @@ -1,5 +1,5 @@ -#ifndef _KEXEC_H -#define _KEXEC_H +#ifndef ASM_X86__KEXEC_H +#define ASM_X86__KEXEC_H #ifdef CONFIG_X86_32 # define PA_CONTROL_PAGE 0 @@ -166,4 +166,4 @@ relocate_kernel(unsigned long indirection_page, #endif /* __ASSEMBLY__ */ -#endif /* _KEXEC_H */ +#endif /* ASM_X86__KEXEC_H */ diff --git a/include/asm-x86/kgdb.h b/include/asm-x86/kgdb.h index 484c47554f3..83a7ee228ab 100644 --- a/include/asm-x86/kgdb.h +++ b/include/asm-x86/kgdb.h @@ -1,5 +1,5 @@ -#ifndef _ASM_KGDB_H_ -#define _ASM_KGDB_H_ +#ifndef ASM_X86__KGDB_H +#define ASM_X86__KGDB_H /* * Copyright (C) 2001-2004 Amit S. Kale @@ -78,4 +78,4 @@ static inline void arch_kgdb_breakpoint(void) #define BREAK_INSTR_SIZE 1 #define CACHE_FLUSH_IS_SAFE 1 -#endif /* _ASM_KGDB_H_ */ +#endif /* ASM_X86__KGDB_H */ diff --git a/include/asm-x86/kmap_types.h b/include/asm-x86/kmap_types.h index 5f4174132a2..89f44493e64 100644 --- a/include/asm-x86/kmap_types.h +++ b/include/asm-x86/kmap_types.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_KMAP_TYPES_H -#define _ASM_X86_KMAP_TYPES_H +#ifndef ASM_X86__KMAP_TYPES_H +#define ASM_X86__KMAP_TYPES_H #if defined(CONFIG_X86_32) && defined(CONFIG_DEBUG_HIGHMEM) # define D(n) __KM_FENCE_##n , @@ -26,4 +26,4 @@ D(13) KM_TYPE_NR #undef D -#endif +#endif /* ASM_X86__KMAP_TYPES_H */ diff --git a/include/asm-x86/kprobes.h b/include/asm-x86/kprobes.h index 54980b0b389..bd8407863c1 100644 --- a/include/asm-x86/kprobes.h +++ b/include/asm-x86/kprobes.h @@ -1,5 +1,5 @@ -#ifndef _ASM_KPROBES_H -#define _ASM_KPROBES_H +#ifndef ASM_X86__KPROBES_H +#define ASM_X86__KPROBES_H /* * Kernel Probes (KProbes) * @@ -94,4 +94,4 @@ static inline void restore_interrupts(struct pt_regs *regs) extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr); extern int kprobe_exceptions_notify(struct notifier_block *self, unsigned long val, void *data); -#endif /* _ASM_KPROBES_H */ +#endif /* ASM_X86__KPROBES_H */ diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h index 6f1840812e5..78e954db1e7 100644 --- a/include/asm-x86/kvm.h +++ b/include/asm-x86/kvm.h @@ -1,5 +1,5 @@ -#ifndef __LINUX_KVM_X86_H -#define __LINUX_KVM_X86_H +#ifndef ASM_X86__KVM_H +#define ASM_X86__KVM_H /* * KVM x86 specific structures and definitions @@ -230,4 +230,4 @@ struct kvm_pit_state { #define KVM_TRC_APIC_ACCESS (KVM_TRC_HANDLER + 0x14) #define KVM_TRC_TDP_FAULT (KVM_TRC_HANDLER + 0x15) -#endif +#endif /* ASM_X86__KVM_H */ diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 924e1bfe3a6..920823d53d6 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -8,8 +8,8 @@ * */ -#ifndef ASM_KVM_HOST_H -#define ASM_KVM_HOST_H +#ifndef ASM_X86__KVM_HOST_H +#define ASM_X86__KVM_HOST_H #include #include @@ -728,4 +728,4 @@ asmlinkage void kvm_handle_fault_on_reboot(void); KVM_EX_ENTRY " 666b, 667b \n\t" \ ".popsection" -#endif +#endif /* ASM_X86__KVM_HOST_H */ diff --git a/include/asm-x86/kvm_para.h b/include/asm-x86/kvm_para.h index 76f392146da..30054fded4f 100644 --- a/include/asm-x86/kvm_para.h +++ b/include/asm-x86/kvm_para.h @@ -1,5 +1,5 @@ -#ifndef __X86_KVM_PARA_H -#define __X86_KVM_PARA_H +#ifndef ASM_X86__KVM_PARA_H +#define ASM_X86__KVM_PARA_H /* This CPUID returns the signature 'KVMKVMKVM' in ebx, ecx, and edx. It * should be used to determine that a VM is running under KVM. @@ -144,4 +144,4 @@ static inline unsigned int kvm_arch_para_features(void) #endif -#endif +#endif /* ASM_X86__KVM_PARA_H */ diff --git a/include/asm-x86/kvm_x86_emulate.h b/include/asm-x86/kvm_x86_emulate.h index 4e8c1e48d91..e2d9b030c1a 100644 --- a/include/asm-x86/kvm_x86_emulate.h +++ b/include/asm-x86/kvm_x86_emulate.h @@ -8,8 +8,8 @@ * From: xen-unstable 10676:af9809f51f81a3c43f276f00c81a52ef558afda4 */ -#ifndef __X86_EMULATE_H__ -#define __X86_EMULATE_H__ +#ifndef ASM_X86__KVM_X86_EMULATE_H +#define ASM_X86__KVM_X86_EMULATE_H struct x86_emulate_ctxt; @@ -181,4 +181,4 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, int x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops); -#endif /* __X86_EMULATE_H__ */ +#endif /* ASM_X86__KVM_X86_EMULATE_H */ diff --git a/include/asm-x86/ldt.h b/include/asm-x86/ldt.h index 20c597242b5..a5228504d86 100644 --- a/include/asm-x86/ldt.h +++ b/include/asm-x86/ldt.h @@ -3,8 +3,8 @@ * * Definitions of structures used with the modify_ldt system call. */ -#ifndef _ASM_X86_LDT_H -#define _ASM_X86_LDT_H +#ifndef ASM_X86__LDT_H +#define ASM_X86__LDT_H /* Maximum number of LDT entries supported. */ #define LDT_ENTRIES 8192 @@ -37,4 +37,4 @@ struct user_desc { #define MODIFY_LDT_CONTENTS_CODE 2 #endif /* !__ASSEMBLY__ */ -#endif +#endif /* ASM_X86__LDT_H */ diff --git a/include/asm-x86/lguest.h b/include/asm-x86/lguest.h index be4a7247fa2..7505e947ed2 100644 --- a/include/asm-x86/lguest.h +++ b/include/asm-x86/lguest.h @@ -1,5 +1,5 @@ -#ifndef _X86_LGUEST_H -#define _X86_LGUEST_H +#ifndef ASM_X86__LGUEST_H +#define ASM_X86__LGUEST_H #define GDT_ENTRY_LGUEST_CS 10 #define GDT_ENTRY_LGUEST_DS 11 @@ -91,4 +91,4 @@ static inline void lguest_set_ts(void) #endif /* __ASSEMBLY__ */ -#endif +#endif /* ASM_X86__LGUEST_H */ diff --git a/include/asm-x86/lguest_hcall.h b/include/asm-x86/lguest_hcall.h index a3241f28e34..8f034ba4b53 100644 --- a/include/asm-x86/lguest_hcall.h +++ b/include/asm-x86/lguest_hcall.h @@ -1,6 +1,6 @@ /* Architecture specific portion of the lguest hypercalls */ -#ifndef _X86_LGUEST_HCALL_H -#define _X86_LGUEST_HCALL_H +#ifndef ASM_X86__LGUEST_HCALL_H +#define ASM_X86__LGUEST_HCALL_H #define LHCALL_FLUSH_ASYNC 0 #define LHCALL_LGUEST_INIT 1 @@ -68,4 +68,4 @@ struct hcall_args { }; #endif /* !__ASSEMBLY__ */ -#endif /* _I386_LGUEST_HCALL_H */ +#endif /* ASM_X86__LGUEST_HCALL_H */ diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h index 64e444f8e85..42d8b62ee8a 100644 --- a/include/asm-x86/linkage.h +++ b/include/asm-x86/linkage.h @@ -1,5 +1,5 @@ -#ifndef __ASM_LINKAGE_H -#define __ASM_LINKAGE_H +#ifndef ASM_X86__LINKAGE_H +#define ASM_X86__LINKAGE_H #undef notrace #define notrace __attribute__((no_instrument_function)) @@ -57,5 +57,5 @@ #define __ALIGN_STR ".align 16,0x90" #endif -#endif +#endif /* ASM_X86__LINKAGE_H */ diff --git a/include/asm-x86/local.h b/include/asm-x86/local.h index 330a72496ab..ae91994fd6c 100644 --- a/include/asm-x86/local.h +++ b/include/asm-x86/local.h @@ -1,5 +1,5 @@ -#ifndef _ARCH_LOCAL_H -#define _ARCH_LOCAL_H +#ifndef ASM_X86__LOCAL_H +#define ASM_X86__LOCAL_H #include @@ -232,4 +232,4 @@ static inline long local_sub_return(long i, local_t *l) #define __cpu_local_add(i, l) cpu_local_add((i), (l)) #define __cpu_local_sub(i, l) cpu_local_sub((i), (l)) -#endif /* _ARCH_LOCAL_H */ +#endif /* ASM_X86__LOCAL_H */ diff --git a/include/asm-x86/mach-bigsmp/mach_apic.h b/include/asm-x86/mach-bigsmp/mach_apic.h index c3b9dc6970c..05362d44a3e 100644 --- a/include/asm-x86/mach-bigsmp/mach_apic.h +++ b/include/asm-x86/mach-bigsmp/mach_apic.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APIC_H -#define __ASM_MACH_APIC_H +#ifndef ASM_X86__MACH_BIGSMP__MACH_APIC_H +#define ASM_X86__MACH_BIGSMP__MACH_APIC_H #define xapic_phys_to_log_apicid(cpu) (per_cpu(x86_bios_cpu_apicid, cpu)) #define esr_disable (1) @@ -141,4 +141,4 @@ static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) return cpuid_apic >> index_msb; } -#endif /* __ASM_MACH_APIC_H */ +#endif /* ASM_X86__MACH_BIGSMP__MACH_APIC_H */ diff --git a/include/asm-x86/mach-bigsmp/mach_apicdef.h b/include/asm-x86/mach-bigsmp/mach_apicdef.h index a58ab5a75c8..811935d9d49 100644 --- a/include/asm-x86/mach-bigsmp/mach_apicdef.h +++ b/include/asm-x86/mach-bigsmp/mach_apicdef.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APICDEF_H -#define __ASM_MACH_APICDEF_H +#ifndef ASM_X86__MACH_BIGSMP__MACH_APICDEF_H +#define ASM_X86__MACH_BIGSMP__MACH_APICDEF_H #define APIC_ID_MASK (0xFF<<24) @@ -10,4 +10,4 @@ static inline unsigned get_apic_id(unsigned long x) #define GET_APIC_ID(x) get_apic_id(x) -#endif +#endif /* ASM_X86__MACH_BIGSMP__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-bigsmp/mach_ipi.h b/include/asm-x86/mach-bigsmp/mach_ipi.h index 9404c535b7e..b1b0f966a00 100644 --- a/include/asm-x86/mach-bigsmp/mach_ipi.h +++ b/include/asm-x86/mach-bigsmp/mach_ipi.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_IPI_H -#define __ASM_MACH_IPI_H +#ifndef ASM_X86__MACH_BIGSMP__MACH_IPI_H +#define ASM_X86__MACH_BIGSMP__MACH_IPI_H void send_IPI_mask_sequence(cpumask_t mask, int vector); @@ -22,4 +22,4 @@ static inline void send_IPI_all(int vector) send_IPI_mask(cpu_online_map, vector); } -#endif /* __ASM_MACH_IPI_H */ +#endif /* ASM_X86__MACH_BIGSMP__MACH_IPI_H */ diff --git a/include/asm-x86/mach-default/apm.h b/include/asm-x86/mach-default/apm.h index 989f34c37d3..2aa61b54fbd 100644 --- a/include/asm-x86/mach-default/apm.h +++ b/include/asm-x86/mach-default/apm.h @@ -3,8 +3,8 @@ * Split out from apm.c by Osamu Tomita */ -#ifndef _ASM_APM_H -#define _ASM_APM_H +#ifndef ASM_X86__MACH_DEFAULT__APM_H +#define ASM_X86__MACH_DEFAULT__APM_H #ifdef APM_ZERO_SEGS # define APM_DO_ZERO_SEGS \ @@ -70,4 +70,4 @@ static inline u8 apm_bios_call_simple_asm(u32 func, u32 ebx_in, return error; } -#endif /* _ASM_APM_H */ +#endif /* ASM_X86__MACH_DEFAULT__APM_H */ diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h index f3226b9a6b8..b615f40736b 100644 --- a/include/asm-x86/mach-default/mach_apic.h +++ b/include/asm-x86/mach-default/mach_apic.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APIC_H -#define __ASM_MACH_APIC_H +#ifndef ASM_X86__MACH_DEFAULT__MACH_APIC_H +#define ASM_X86__MACH_DEFAULT__MACH_APIC_H #ifdef CONFIG_X86_LOCAL_APIC @@ -138,4 +138,4 @@ static inline void enable_apic_mode(void) } #endif /* CONFIG_X86_LOCAL_APIC */ -#endif /* __ASM_MACH_APIC_H */ +#endif /* ASM_X86__MACH_DEFAULT__MACH_APIC_H */ diff --git a/include/asm-x86/mach-default/mach_apicdef.h b/include/asm-x86/mach-default/mach_apicdef.h index e4b29ba37de..936704f816d 100644 --- a/include/asm-x86/mach-default/mach_apicdef.h +++ b/include/asm-x86/mach-default/mach_apicdef.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APICDEF_H -#define __ASM_MACH_APICDEF_H +#ifndef ASM_X86__MACH_DEFAULT__MACH_APICDEF_H +#define ASM_X86__MACH_DEFAULT__MACH_APICDEF_H #include @@ -21,4 +21,4 @@ static inline unsigned get_apic_id(unsigned long x) #define GET_APIC_ID(x) get_apic_id(x) #endif -#endif +#endif /* ASM_X86__MACH_DEFAULT__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-default/mach_ipi.h b/include/asm-x86/mach-default/mach_ipi.h index be323364e68..674bc7e50c3 100644 --- a/include/asm-x86/mach-default/mach_ipi.h +++ b/include/asm-x86/mach-default/mach_ipi.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_IPI_H -#define __ASM_MACH_IPI_H +#ifndef ASM_X86__MACH_DEFAULT__MACH_IPI_H +#define ASM_X86__MACH_DEFAULT__MACH_IPI_H /* Avoid include hell */ #define NMI_VECTOR 0x02 @@ -61,4 +61,4 @@ static inline void send_IPI_all(int vector) } #endif -#endif /* __ASM_MACH_IPI_H */ +#endif /* ASM_X86__MACH_DEFAULT__MACH_IPI_H */ diff --git a/include/asm-x86/mach-default/mach_mpparse.h b/include/asm-x86/mach-default/mach_mpparse.h index d14108505bb..9c381f2815a 100644 --- a/include/asm-x86/mach-default/mach_mpparse.h +++ b/include/asm-x86/mach-default/mach_mpparse.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_MPPARSE_H -#define __ASM_MACH_MPPARSE_H +#ifndef ASM_X86__MACH_DEFAULT__MACH_MPPARSE_H +#define ASM_X86__MACH_DEFAULT__MACH_MPPARSE_H static inline int mps_oem_check(struct mp_config_table *mpc, char *oem, char *productid) @@ -14,4 +14,4 @@ static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) } -#endif /* __ASM_MACH_MPPARSE_H */ +#endif /* ASM_X86__MACH_DEFAULT__MACH_MPPARSE_H */ diff --git a/include/asm-x86/mach-default/mach_mpspec.h b/include/asm-x86/mach-default/mach_mpspec.h index 51c9a977593..d77646f011f 100644 --- a/include/asm-x86/mach-default/mach_mpspec.h +++ b/include/asm-x86/mach-default/mach_mpspec.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_MPSPEC_H -#define __ASM_MACH_MPSPEC_H +#ifndef ASM_X86__MACH_DEFAULT__MACH_MPSPEC_H +#define ASM_X86__MACH_DEFAULT__MACH_MPSPEC_H #define MAX_IRQ_SOURCES 256 @@ -9,4 +9,4 @@ #define MAX_MP_BUSSES 32 #endif -#endif /* __ASM_MACH_MPSPEC_H */ +#endif /* ASM_X86__MACH_DEFAULT__MACH_MPSPEC_H */ diff --git a/include/asm-x86/mach-default/mach_timer.h b/include/asm-x86/mach-default/mach_timer.h index 4b76e536cd9..990b1583383 100644 --- a/include/asm-x86/mach-default/mach_timer.h +++ b/include/asm-x86/mach-default/mach_timer.h @@ -10,8 +10,8 @@ * directly because of the awkward 8-bit access mechanism of the 82C54 * device. */ -#ifndef _MACH_TIMER_H -#define _MACH_TIMER_H +#ifndef ASM_X86__MACH_DEFAULT__MACH_TIMER_H +#define ASM_X86__MACH_DEFAULT__MACH_TIMER_H #define CALIBRATE_TIME_MSEC 30 /* 30 msecs */ #define CALIBRATE_LATCH \ @@ -45,4 +45,4 @@ static inline void mach_countup(unsigned long *count_p) *count_p = count; } -#endif /* !_MACH_TIMER_H */ +#endif /* ASM_X86__MACH_DEFAULT__MACH_TIMER_H */ diff --git a/include/asm-x86/mach-default/mach_traps.h b/include/asm-x86/mach-default/mach_traps.h index 2fe7705c048..de9ac3f5c4c 100644 --- a/include/asm-x86/mach-default/mach_traps.h +++ b/include/asm-x86/mach-default/mach_traps.h @@ -2,8 +2,8 @@ * Machine specific NMI handling for generic. * Split out from traps.c by Osamu Tomita */ -#ifndef _MACH_TRAPS_H -#define _MACH_TRAPS_H +#ifndef ASM_X86__MACH_DEFAULT__MACH_TRAPS_H +#define ASM_X86__MACH_DEFAULT__MACH_TRAPS_H #include @@ -36,4 +36,4 @@ static inline void reassert_nmi(void) unlock_cmos(); } -#endif /* !_MACH_TRAPS_H */ +#endif /* ASM_X86__MACH_DEFAULT__MACH_TRAPS_H */ diff --git a/include/asm-x86/mach-default/mach_wakecpu.h b/include/asm-x86/mach-default/mach_wakecpu.h index 3ebb17893aa..361b810f516 100644 --- a/include/asm-x86/mach-default/mach_wakecpu.h +++ b/include/asm-x86/mach-default/mach_wakecpu.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_WAKECPU_H -#define __ASM_MACH_WAKECPU_H +#ifndef ASM_X86__MACH_DEFAULT__MACH_WAKECPU_H +#define ASM_X86__MACH_DEFAULT__MACH_WAKECPU_H /* * This file copes with machines that wakeup secondary CPUs by the @@ -39,4 +39,4 @@ static inline void restore_NMI_vector(unsigned short *high, unsigned short *low) #define inquire_remote_apic(apicid) {} #endif -#endif /* __ASM_MACH_WAKECPU_H */ +#endif /* ASM_X86__MACH_DEFAULT__MACH_WAKECPU_H */ diff --git a/include/asm-x86/mach-es7000/mach_apic.h b/include/asm-x86/mach-es7000/mach_apic.h index 0a3fdf93067..c1f6f682d61 100644 --- a/include/asm-x86/mach-es7000/mach_apic.h +++ b/include/asm-x86/mach-es7000/mach_apic.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APIC_H -#define __ASM_MACH_APIC_H +#ifndef ASM_X86__MACH_ES7000__MACH_APIC_H +#define ASM_X86__MACH_ES7000__MACH_APIC_H #define xapic_phys_to_log_apicid(cpu) per_cpu(x86_bios_cpu_apicid, cpu) #define esr_disable (1) @@ -191,4 +191,4 @@ static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) return cpuid_apic >> index_msb; } -#endif /* __ASM_MACH_APIC_H */ +#endif /* ASM_X86__MACH_ES7000__MACH_APIC_H */ diff --git a/include/asm-x86/mach-es7000/mach_apicdef.h b/include/asm-x86/mach-es7000/mach_apicdef.h index a58ab5a75c8..a07e5674402 100644 --- a/include/asm-x86/mach-es7000/mach_apicdef.h +++ b/include/asm-x86/mach-es7000/mach_apicdef.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APICDEF_H -#define __ASM_MACH_APICDEF_H +#ifndef ASM_X86__MACH_ES7000__MACH_APICDEF_H +#define ASM_X86__MACH_ES7000__MACH_APICDEF_H #define APIC_ID_MASK (0xFF<<24) @@ -10,4 +10,4 @@ static inline unsigned get_apic_id(unsigned long x) #define GET_APIC_ID(x) get_apic_id(x) -#endif +#endif /* ASM_X86__MACH_ES7000__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-es7000/mach_ipi.h b/include/asm-x86/mach-es7000/mach_ipi.h index 5e61bd220b0..3a21240e03d 100644 --- a/include/asm-x86/mach-es7000/mach_ipi.h +++ b/include/asm-x86/mach-es7000/mach_ipi.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_IPI_H -#define __ASM_MACH_IPI_H +#ifndef ASM_X86__MACH_ES7000__MACH_IPI_H +#define ASM_X86__MACH_ES7000__MACH_IPI_H void send_IPI_mask_sequence(cpumask_t mask, int vector); @@ -21,4 +21,4 @@ static inline void send_IPI_all(int vector) send_IPI_mask(cpu_online_map, vector); } -#endif /* __ASM_MACH_IPI_H */ +#endif /* ASM_X86__MACH_ES7000__MACH_IPI_H */ diff --git a/include/asm-x86/mach-es7000/mach_mpparse.h b/include/asm-x86/mach-es7000/mach_mpparse.h index ef26d352362..befde24705b 100644 --- a/include/asm-x86/mach-es7000/mach_mpparse.h +++ b/include/asm-x86/mach-es7000/mach_mpparse.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_MPPARSE_H -#define __ASM_MACH_MPPARSE_H +#ifndef ASM_X86__MACH_ES7000__MACH_MPPARSE_H +#define ASM_X86__MACH_ES7000__MACH_MPPARSE_H #include @@ -26,4 +26,4 @@ static inline int es7000_check_dsdt(void) } #endif -#endif /* __ASM_MACH_MPPARSE_H */ +#endif /* ASM_X86__MACH_ES7000__MACH_MPPARSE_H */ diff --git a/include/asm-x86/mach-es7000/mach_wakecpu.h b/include/asm-x86/mach-es7000/mach_wakecpu.h index 84ff5831450..97c776ce13f 100644 --- a/include/asm-x86/mach-es7000/mach_wakecpu.h +++ b/include/asm-x86/mach-es7000/mach_wakecpu.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_WAKECPU_H -#define __ASM_MACH_WAKECPU_H +#ifndef ASM_X86__MACH_ES7000__MACH_WAKECPU_H +#define ASM_X86__MACH_ES7000__MACH_WAKECPU_H /* * This file copes with machines that wakeup secondary CPUs by the @@ -56,4 +56,4 @@ static inline void restore_NMI_vector(unsigned short *high, unsigned short *low) #define inquire_remote_apic(apicid) {} #endif -#endif /* __ASM_MACH_WAKECPU_H */ +#endif /* ASM_X86__MACH_ES7000__MACH_WAKECPU_H */ diff --git a/include/asm-x86/mach-generic/gpio.h b/include/asm-x86/mach-generic/gpio.h index 5305dcb96df..6ce0f7786ef 100644 --- a/include/asm-x86/mach-generic/gpio.h +++ b/include/asm-x86/mach-generic/gpio.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_GENERIC_GPIO_H -#define __ASM_MACH_GENERIC_GPIO_H +#ifndef ASM_X86__MACH_GENERIC__GPIO_H +#define ASM_X86__MACH_GENERIC__GPIO_H int gpio_request(unsigned gpio, const char *label); void gpio_free(unsigned gpio); @@ -12,4 +12,4 @@ int irq_to_gpio(unsigned irq); #include /* cansleep wrappers */ -#endif /* __ASM_MACH_GENERIC_GPIO_H */ +#endif /* ASM_X86__MACH_GENERIC__GPIO_H */ diff --git a/include/asm-x86/mach-generic/irq_vectors_limits.h b/include/asm-x86/mach-generic/irq_vectors_limits.h index 890ce3f5e09..f7870e1a220 100644 --- a/include/asm-x86/mach-generic/irq_vectors_limits.h +++ b/include/asm-x86/mach-generic/irq_vectors_limits.h @@ -1,5 +1,5 @@ -#ifndef _ASM_IRQ_VECTORS_LIMITS_H -#define _ASM_IRQ_VECTORS_LIMITS_H +#ifndef ASM_X86__MACH_GENERIC__IRQ_VECTORS_LIMITS_H +#define ASM_X86__MACH_GENERIC__IRQ_VECTORS_LIMITS_H /* * For Summit or generic (i.e. installer) kernels, we have lots of I/O APICs, @@ -11,4 +11,4 @@ #define NR_IRQS 224 #define NR_IRQ_VECTORS 1024 -#endif /* _ASM_IRQ_VECTORS_LIMITS_H */ +#endif /* ASM_X86__MACH_GENERIC__IRQ_VECTORS_LIMITS_H */ diff --git a/include/asm-x86/mach-generic/mach_apic.h b/include/asm-x86/mach-generic/mach_apic.h index 6eff343e123..5d010c6881d 100644 --- a/include/asm-x86/mach-generic/mach_apic.h +++ b/include/asm-x86/mach-generic/mach_apic.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APIC_H -#define __ASM_MACH_APIC_H +#ifndef ASM_X86__MACH_GENERIC__MACH_APIC_H +#define ASM_X86__MACH_GENERIC__MACH_APIC_H #include @@ -29,4 +29,4 @@ extern void generic_bigsmp_probe(void); -#endif /* __ASM_MACH_APIC_H */ +#endif /* ASM_X86__MACH_GENERIC__MACH_APIC_H */ diff --git a/include/asm-x86/mach-generic/mach_apicdef.h b/include/asm-x86/mach-generic/mach_apicdef.h index 28ed98972ca..1657f38b8f2 100644 --- a/include/asm-x86/mach-generic/mach_apicdef.h +++ b/include/asm-x86/mach-generic/mach_apicdef.h @@ -1,5 +1,5 @@ -#ifndef _GENAPIC_MACH_APICDEF_H -#define _GENAPIC_MACH_APICDEF_H 1 +#ifndef ASM_X86__MACH_GENERIC__MACH_APICDEF_H +#define ASM_X86__MACH_GENERIC__MACH_APICDEF_H #ifndef APIC_DEFINITION #include @@ -8,4 +8,4 @@ #define APIC_ID_MASK (genapic->apic_id_mask) #endif -#endif +#endif /* ASM_X86__MACH_GENERIC__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-generic/mach_ipi.h b/include/asm-x86/mach-generic/mach_ipi.h index 441b0fe3ed1..f67433dbd65 100644 --- a/include/asm-x86/mach-generic/mach_ipi.h +++ b/include/asm-x86/mach-generic/mach_ipi.h @@ -1,5 +1,5 @@ -#ifndef _MACH_IPI_H -#define _MACH_IPI_H 1 +#ifndef ASM_X86__MACH_GENERIC__MACH_IPI_H +#define ASM_X86__MACH_GENERIC__MACH_IPI_H #include @@ -7,4 +7,4 @@ #define send_IPI_allbutself (genapic->send_IPI_allbutself) #define send_IPI_all (genapic->send_IPI_all) -#endif +#endif /* ASM_X86__MACH_GENERIC__MACH_IPI_H */ diff --git a/include/asm-x86/mach-generic/mach_mpparse.h b/include/asm-x86/mach-generic/mach_mpparse.h index 586cadbf378..3115564e557 100644 --- a/include/asm-x86/mach-generic/mach_mpparse.h +++ b/include/asm-x86/mach-generic/mach_mpparse.h @@ -1,5 +1,5 @@ -#ifndef _MACH_MPPARSE_H -#define _MACH_MPPARSE_H 1 +#ifndef ASM_X86__MACH_GENERIC__MACH_MPPARSE_H +#define ASM_X86__MACH_GENERIC__MACH_MPPARSE_H extern int mps_oem_check(struct mp_config_table *mpc, char *oem, @@ -7,4 +7,4 @@ extern int mps_oem_check(struct mp_config_table *mpc, char *oem, extern int acpi_madt_oem_check(char *oem_id, char *oem_table_id); -#endif +#endif /* ASM_X86__MACH_GENERIC__MACH_MPPARSE_H */ diff --git a/include/asm-x86/mach-generic/mach_mpspec.h b/include/asm-x86/mach-generic/mach_mpspec.h index c83c120be53..6061b153613 100644 --- a/include/asm-x86/mach-generic/mach_mpspec.h +++ b/include/asm-x86/mach-generic/mach_mpspec.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_MPSPEC_H -#define __ASM_MACH_MPSPEC_H +#ifndef ASM_X86__MACH_GENERIC__MACH_MPSPEC_H +#define ASM_X86__MACH_GENERIC__MACH_MPSPEC_H #define MAX_IRQ_SOURCES 256 @@ -9,4 +9,4 @@ extern void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem, char *productid); -#endif /* __ASM_MACH_MPSPEC_H */ +#endif /* ASM_X86__MACH_GENERIC__MACH_MPSPEC_H */ diff --git a/include/asm-x86/mach-numaq/mach_apic.h b/include/asm-x86/mach-numaq/mach_apic.h index d802465e026..7a0d39edfcf 100644 --- a/include/asm-x86/mach-numaq/mach_apic.h +++ b/include/asm-x86/mach-numaq/mach_apic.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APIC_H -#define __ASM_MACH_APIC_H +#ifndef ASM_X86__MACH_NUMAQ__MACH_APIC_H +#define ASM_X86__MACH_NUMAQ__MACH_APIC_H #include #include @@ -135,4 +135,4 @@ static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) return cpuid_apic >> index_msb; } -#endif /* __ASM_MACH_APIC_H */ +#endif /* ASM_X86__MACH_NUMAQ__MACH_APIC_H */ diff --git a/include/asm-x86/mach-numaq/mach_apicdef.h b/include/asm-x86/mach-numaq/mach_apicdef.h index bf439d0690f..f870ec5f778 100644 --- a/include/asm-x86/mach-numaq/mach_apicdef.h +++ b/include/asm-x86/mach-numaq/mach_apicdef.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APICDEF_H -#define __ASM_MACH_APICDEF_H +#ifndef ASM_X86__MACH_NUMAQ__MACH_APICDEF_H +#define ASM_X86__MACH_NUMAQ__MACH_APICDEF_H #define APIC_ID_MASK (0xF<<24) @@ -11,4 +11,4 @@ static inline unsigned get_apic_id(unsigned long x) #define GET_APIC_ID(x) get_apic_id(x) -#endif +#endif /* ASM_X86__MACH_NUMAQ__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-numaq/mach_ipi.h b/include/asm-x86/mach-numaq/mach_ipi.h index c6044488e9e..1e835823f4b 100644 --- a/include/asm-x86/mach-numaq/mach_ipi.h +++ b/include/asm-x86/mach-numaq/mach_ipi.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_IPI_H -#define __ASM_MACH_IPI_H +#ifndef ASM_X86__MACH_NUMAQ__MACH_IPI_H +#define ASM_X86__MACH_NUMAQ__MACH_IPI_H void send_IPI_mask_sequence(cpumask_t, int vector); @@ -22,4 +22,4 @@ static inline void send_IPI_all(int vector) send_IPI_mask(cpu_online_map, vector); } -#endif /* __ASM_MACH_IPI_H */ +#endif /* ASM_X86__MACH_NUMAQ__MACH_IPI_H */ diff --git a/include/asm-x86/mach-numaq/mach_mpparse.h b/include/asm-x86/mach-numaq/mach_mpparse.h index 626aef6b155..74ade184920 100644 --- a/include/asm-x86/mach-numaq/mach_mpparse.h +++ b/include/asm-x86/mach-numaq/mach_mpparse.h @@ -1,7 +1,7 @@ -#ifndef __ASM_MACH_MPPARSE_H -#define __ASM_MACH_MPPARSE_H +#ifndef ASM_X86__MACH_NUMAQ__MACH_MPPARSE_H +#define ASM_X86__MACH_NUMAQ__MACH_MPPARSE_H extern void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem, char *productid); -#endif /* __ASM_MACH_MPPARSE_H */ +#endif /* ASM_X86__MACH_NUMAQ__MACH_MPPARSE_H */ diff --git a/include/asm-x86/mach-numaq/mach_wakecpu.h b/include/asm-x86/mach-numaq/mach_wakecpu.h index 00530041a99..0db8cea643c 100644 --- a/include/asm-x86/mach-numaq/mach_wakecpu.h +++ b/include/asm-x86/mach-numaq/mach_wakecpu.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_WAKECPU_H -#define __ASM_MACH_WAKECPU_H +#ifndef ASM_X86__MACH_NUMAQ__MACH_WAKECPU_H +#define ASM_X86__MACH_NUMAQ__MACH_WAKECPU_H /* This file copes with machines that wakeup secondary CPUs by NMIs */ @@ -40,4 +40,4 @@ static inline void restore_NMI_vector(unsigned short *high, unsigned short *low) #define inquire_remote_apic(apicid) {} -#endif /* __ASM_MACH_WAKECPU_H */ +#endif /* ASM_X86__MACH_NUMAQ__MACH_WAKECPU_H */ diff --git a/include/asm-x86/mach-rdc321x/gpio.h b/include/asm-x86/mach-rdc321x/gpio.h index acce0b7d397..6184561980f 100644 --- a/include/asm-x86/mach-rdc321x/gpio.h +++ b/include/asm-x86/mach-rdc321x/gpio.h @@ -1,5 +1,5 @@ -#ifndef _RDC321X_GPIO_H -#define _RDC321X_GPIO_H +#ifndef ASM_X86__MACH_RDC321X__GPIO_H +#define ASM_X86__MACH_RDC321X__GPIO_H extern int rdc_gpio_get_value(unsigned gpio); extern void rdc_gpio_set_value(unsigned gpio, int value); @@ -54,4 +54,4 @@ static inline int irq_to_gpio(unsigned irq) /* For cansleep */ #include -#endif /* _RDC321X_GPIO_H_ */ +#endif /* ASM_X86__MACH_RDC321X__GPIO_H */ diff --git a/include/asm-x86/mach-summit/irq_vectors_limits.h b/include/asm-x86/mach-summit/irq_vectors_limits.h index 890ce3f5e09..22f376ad68e 100644 --- a/include/asm-x86/mach-summit/irq_vectors_limits.h +++ b/include/asm-x86/mach-summit/irq_vectors_limits.h @@ -1,5 +1,5 @@ -#ifndef _ASM_IRQ_VECTORS_LIMITS_H -#define _ASM_IRQ_VECTORS_LIMITS_H +#ifndef ASM_X86__MACH_SUMMIT__IRQ_VECTORS_LIMITS_H +#define ASM_X86__MACH_SUMMIT__IRQ_VECTORS_LIMITS_H /* * For Summit or generic (i.e. installer) kernels, we have lots of I/O APICs, @@ -11,4 +11,4 @@ #define NR_IRQS 224 #define NR_IRQ_VECTORS 1024 -#endif /* _ASM_IRQ_VECTORS_LIMITS_H */ +#endif /* ASM_X86__MACH_SUMMIT__IRQ_VECTORS_LIMITS_H */ diff --git a/include/asm-x86/mach-summit/mach_apic.h b/include/asm-x86/mach-summit/mach_apic.h index 75d2c95005d..ef77af36cc9 100644 --- a/include/asm-x86/mach-summit/mach_apic.h +++ b/include/asm-x86/mach-summit/mach_apic.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APIC_H -#define __ASM_MACH_APIC_H +#ifndef ASM_X86__MACH_SUMMIT__MACH_APIC_H +#define ASM_X86__MACH_SUMMIT__MACH_APIC_H #include @@ -182,4 +182,4 @@ static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) return hard_smp_processor_id() >> index_msb; } -#endif /* __ASM_MACH_APIC_H */ +#endif /* ASM_X86__MACH_SUMMIT__MACH_APIC_H */ diff --git a/include/asm-x86/mach-summit/mach_apicdef.h b/include/asm-x86/mach-summit/mach_apicdef.h index a58ab5a75c8..d4bc8590c4f 100644 --- a/include/asm-x86/mach-summit/mach_apicdef.h +++ b/include/asm-x86/mach-summit/mach_apicdef.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_APICDEF_H -#define __ASM_MACH_APICDEF_H +#ifndef ASM_X86__MACH_SUMMIT__MACH_APICDEF_H +#define ASM_X86__MACH_SUMMIT__MACH_APICDEF_H #define APIC_ID_MASK (0xFF<<24) @@ -10,4 +10,4 @@ static inline unsigned get_apic_id(unsigned long x) #define GET_APIC_ID(x) get_apic_id(x) -#endif +#endif /* ASM_X86__MACH_SUMMIT__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-summit/mach_ipi.h b/include/asm-x86/mach-summit/mach_ipi.h index 9404c535b7e..a3b31c528d9 100644 --- a/include/asm-x86/mach-summit/mach_ipi.h +++ b/include/asm-x86/mach-summit/mach_ipi.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_IPI_H -#define __ASM_MACH_IPI_H +#ifndef ASM_X86__MACH_SUMMIT__MACH_IPI_H +#define ASM_X86__MACH_SUMMIT__MACH_IPI_H void send_IPI_mask_sequence(cpumask_t mask, int vector); @@ -22,4 +22,4 @@ static inline void send_IPI_all(int vector) send_IPI_mask(cpu_online_map, vector); } -#endif /* __ASM_MACH_IPI_H */ +#endif /* ASM_X86__MACH_SUMMIT__MACH_IPI_H */ diff --git a/include/asm-x86/mach-summit/mach_mpparse.h b/include/asm-x86/mach-summit/mach_mpparse.h index fdf59170133..92396f28772 100644 --- a/include/asm-x86/mach-summit/mach_mpparse.h +++ b/include/asm-x86/mach-summit/mach_mpparse.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MACH_MPPARSE_H -#define __ASM_MACH_MPPARSE_H +#ifndef ASM_X86__MACH_SUMMIT__MACH_MPPARSE_H +#define ASM_X86__MACH_SUMMIT__MACH_MPPARSE_H #include #include @@ -107,4 +107,4 @@ static inline int is_WPEG(struct rio_detail *rio){ rio->type == LookOutAWPEG || rio->type == LookOutBWPEG); } -#endif /* __ASM_MACH_MPPARSE_H */ +#endif /* ASM_X86__MACH_SUMMIT__MACH_MPPARSE_H */ diff --git a/include/asm-x86/math_emu.h b/include/asm-x86/math_emu.h index 9bf4ae93ab1..5768d8e95c8 100644 --- a/include/asm-x86/math_emu.h +++ b/include/asm-x86/math_emu.h @@ -1,5 +1,5 @@ -#ifndef _I386_MATH_EMU_H -#define _I386_MATH_EMU_H +#ifndef ASM_X86__MATH_EMU_H +#define ASM_X86__MATH_EMU_H /* This structure matches the layout of the data saved to the stack following a device-not-present interrupt, part of it saved @@ -28,4 +28,4 @@ struct info { long ___vm86_fs; long ___vm86_gs; }; -#endif +#endif /* ASM_X86__MATH_EMU_H */ diff --git a/include/asm-x86/mc146818rtc.h b/include/asm-x86/mc146818rtc.h index daf1ccde77a..a995f33176c 100644 --- a/include/asm-x86/mc146818rtc.h +++ b/include/asm-x86/mc146818rtc.h @@ -1,8 +1,8 @@ /* * Machine dependent access functions for RTC registers. */ -#ifndef _ASM_MC146818RTC_H -#define _ASM_MC146818RTC_H +#ifndef ASM_X86__MC146818RTC_H +#define ASM_X86__MC146818RTC_H #include #include @@ -101,4 +101,4 @@ extern unsigned long mach_get_cmos_time(void); #define RTC_IRQ 8 -#endif /* _ASM_MC146818RTC_H */ +#endif /* ASM_X86__MC146818RTC_H */ diff --git a/include/asm-x86/mca.h b/include/asm-x86/mca.h index 09adf2eac4d..60d1ed287b1 100644 --- a/include/asm-x86/mca.h +++ b/include/asm-x86/mca.h @@ -1,8 +1,8 @@ /* -*- mode: c; c-basic-offset: 8 -*- */ /* Platform specific MCA defines */ -#ifndef _ASM_MCA_H -#define _ASM_MCA_H +#ifndef ASM_X86__MCA_H +#define ASM_X86__MCA_H /* Maximal number of MCA slots - actually, some machines have less, but * they all have sufficient number of POS registers to cover 8. @@ -40,4 +40,4 @@ */ #define MCA_NUMADAPTERS (MCA_MAX_SLOT_NR+3) -#endif +#endif /* ASM_X86__MCA_H */ diff --git a/include/asm-x86/mca_dma.h b/include/asm-x86/mca_dma.h index c3dca6edc6b..49f22be237d 100644 --- a/include/asm-x86/mca_dma.h +++ b/include/asm-x86/mca_dma.h @@ -1,5 +1,5 @@ -#ifndef MCA_DMA_H -#define MCA_DMA_H +#ifndef ASM_X86__MCA_DMA_H +#define ASM_X86__MCA_DMA_H #include #include @@ -198,4 +198,4 @@ static inline void mca_set_dma_mode(unsigned int dmanr, unsigned int mode) outb(mode, MCA_DMA_REG_EXE); } -#endif /* MCA_DMA_H */ +#endif /* ASM_X86__MCA_DMA_H */ diff --git a/include/asm-x86/mce.h b/include/asm-x86/mce.h index 94f1fd79e22..6a580f24d4a 100644 --- a/include/asm-x86/mce.h +++ b/include/asm-x86/mce.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_MCE_H -#define _ASM_X86_MCE_H +#ifndef ASM_X86__MCE_H +#define ASM_X86__MCE_H #ifdef __x86_64__ @@ -126,4 +126,4 @@ extern void restart_mce(void); #endif /* __KERNEL__ */ -#endif +#endif /* ASM_X86__MCE_H */ diff --git a/include/asm-x86/mman.h b/include/asm-x86/mman.h index c1682b542da..b6b41aa1cbc 100644 --- a/include/asm-x86/mman.h +++ b/include/asm-x86/mman.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_MMAN_H -#define _ASM_X86_MMAN_H +#ifndef ASM_X86__MMAN_H +#define ASM_X86__MMAN_H #include @@ -16,4 +16,4 @@ #define MCL_CURRENT 1 /* lock all current mappings */ #define MCL_FUTURE 2 /* lock all future mappings */ -#endif /* _ASM_X86_MMAN_H */ +#endif /* ASM_X86__MMAN_H */ diff --git a/include/asm-x86/mmconfig.h b/include/asm-x86/mmconfig.h index 95beda07c6f..8689f1e7bc0 100644 --- a/include/asm-x86/mmconfig.h +++ b/include/asm-x86/mmconfig.h @@ -1,5 +1,5 @@ -#ifndef _ASM_MMCONFIG_H -#define _ASM_MMCONFIG_H +#ifndef ASM_X86__MMCONFIG_H +#define ASM_X86__MMCONFIG_H #ifdef CONFIG_PCI_MMCONFIG extern void __cpuinit fam10h_check_enable_mmcfg(void); @@ -9,4 +9,4 @@ static inline void fam10h_check_enable_mmcfg(void) { } static inline void check_enable_amd_mmconf_dmi(void) { } #endif -#endif +#endif /* ASM_X86__MMCONFIG_H */ diff --git a/include/asm-x86/mmu.h b/include/asm-x86/mmu.h index 00e88679e11..a30d7a9c829 100644 --- a/include/asm-x86/mmu.h +++ b/include/asm-x86/mmu.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_MMU_H -#define _ASM_X86_MMU_H +#ifndef ASM_X86__MMU_H +#define ASM_X86__MMU_H #include #include @@ -28,4 +28,4 @@ static inline void leave_mm(int cpu) } #endif -#endif /* _ASM_X86_MMU_H */ +#endif /* ASM_X86__MMU_H */ diff --git a/include/asm-x86/mmu_context.h b/include/asm-x86/mmu_context.h index fac57014e7c..8ec940bfd07 100644 --- a/include/asm-x86/mmu_context.h +++ b/include/asm-x86/mmu_context.h @@ -1,5 +1,5 @@ -#ifndef __ASM_X86_MMU_CONTEXT_H -#define __ASM_X86_MMU_CONTEXT_H +#ifndef ASM_X86__MMU_CONTEXT_H +#define ASM_X86__MMU_CONTEXT_H #include #include @@ -34,4 +34,4 @@ do { \ } while (0); -#endif /* __ASM_X86_MMU_CONTEXT_H */ +#endif /* ASM_X86__MMU_CONTEXT_H */ diff --git a/include/asm-x86/mmu_context_32.h b/include/asm-x86/mmu_context_32.h index 824fc575c6d..cce6f6e4afd 100644 --- a/include/asm-x86/mmu_context_32.h +++ b/include/asm-x86/mmu_context_32.h @@ -1,5 +1,5 @@ -#ifndef __I386_SCHED_H -#define __I386_SCHED_H +#ifndef ASM_X86__MMU_CONTEXT_32_H +#define ASM_X86__MMU_CONTEXT_32_H static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk) { @@ -53,4 +53,4 @@ static inline void switch_mm(struct mm_struct *prev, #define deactivate_mm(tsk, mm) \ asm("movl %0,%%gs": :"r" (0)); -#endif +#endif /* ASM_X86__MMU_CONTEXT_32_H */ diff --git a/include/asm-x86/mmu_context_64.h b/include/asm-x86/mmu_context_64.h index c7000634cca..26758673c82 100644 --- a/include/asm-x86/mmu_context_64.h +++ b/include/asm-x86/mmu_context_64.h @@ -1,5 +1,5 @@ -#ifndef __X86_64_MMU_CONTEXT_H -#define __X86_64_MMU_CONTEXT_H +#ifndef ASM_X86__MMU_CONTEXT_64_H +#define ASM_X86__MMU_CONTEXT_64_H #include @@ -51,4 +51,4 @@ do { \ asm volatile("movl %0,%%fs"::"r"(0)); \ } while (0) -#endif +#endif /* ASM_X86__MMU_CONTEXT_64_H */ diff --git a/include/asm-x86/mmx.h b/include/asm-x86/mmx.h index 940881218ff..2e7299bb365 100644 --- a/include/asm-x86/mmx.h +++ b/include/asm-x86/mmx.h @@ -1,5 +1,5 @@ -#ifndef _ASM_MMX_H -#define _ASM_MMX_H +#ifndef ASM_X86__MMX_H +#define ASM_X86__MMX_H /* * MMX 3Dnow! helper operations @@ -11,4 +11,4 @@ extern void *_mmx_memcpy(void *to, const void *from, size_t size); extern void mmx_clear_page(void *page); extern void mmx_copy_page(void *to, void *from); -#endif +#endif /* ASM_X86__MMX_H */ diff --git a/include/asm-x86/mmzone_32.h b/include/asm-x86/mmzone_32.h index b2298a22756..b98590fdc9e 100644 --- a/include/asm-x86/mmzone_32.h +++ b/include/asm-x86/mmzone_32.h @@ -3,8 +3,8 @@ * */ -#ifndef _ASM_MMZONE_H_ -#define _ASM_MMZONE_H_ +#ifndef ASM_X86__MMZONE_32_H +#define ASM_X86__MMZONE_32_H #include @@ -125,4 +125,4 @@ static inline int pfn_valid(int pfn) }) #endif /* CONFIG_NEED_MULTIPLE_NODES */ -#endif /* _ASM_MMZONE_H_ */ +#endif /* ASM_X86__MMZONE_32_H */ diff --git a/include/asm-x86/mmzone_64.h b/include/asm-x86/mmzone_64.h index 594bd0dc1d0..626b03a1487 100644 --- a/include/asm-x86/mmzone_64.h +++ b/include/asm-x86/mmzone_64.h @@ -1,8 +1,8 @@ /* K8 NUMA support */ /* Copyright 2002,2003 by Andi Kleen, SuSE Labs */ /* 2.5 Version loosely based on the NUMAQ Code by Pat Gaughen. */ -#ifndef _ASM_X86_64_MMZONE_H -#define _ASM_X86_64_MMZONE_H 1 +#ifndef ASM_X86__MMZONE_64_H +#define ASM_X86__MMZONE_64_H #ifdef CONFIG_NUMA @@ -49,4 +49,4 @@ extern int early_pfn_to_nid(unsigned long pfn); #endif #endif -#endif +#endif /* ASM_X86__MMZONE_64_H */ diff --git a/include/asm-x86/module.h b/include/asm-x86/module.h index bfedb247871..48dc3e0c07d 100644 --- a/include/asm-x86/module.h +++ b/include/asm-x86/module.h @@ -1,5 +1,5 @@ -#ifndef _ASM_MODULE_H -#define _ASM_MODULE_H +#ifndef ASM_X86__MODULE_H +#define ASM_X86__MODULE_H /* x86_32/64 are simple */ struct mod_arch_specific {}; @@ -79,4 +79,4 @@ struct mod_arch_specific {}; # define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE #endif -#endif /* _ASM_MODULE_H */ +#endif /* ASM_X86__MODULE_H */ diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h index b6995e567fc..118da365e37 100644 --- a/include/asm-x86/mpspec.h +++ b/include/asm-x86/mpspec.h @@ -1,5 +1,5 @@ -#ifndef _AM_X86_MPSPEC_H -#define _AM_X86_MPSPEC_H +#ifndef ASM_X86__MPSPEC_H +#define ASM_X86__MPSPEC_H #include @@ -141,4 +141,4 @@ static inline void physid_set_mask_of_physid(int physid, physid_mask_t *map) extern physid_mask_t phys_cpu_present_map; -#endif +#endif /* ASM_X86__MPSPEC_H */ diff --git a/include/asm-x86/mpspec_def.h b/include/asm-x86/mpspec_def.h index 38d1e73b49e..79166b04801 100644 --- a/include/asm-x86/mpspec_def.h +++ b/include/asm-x86/mpspec_def.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MPSPEC_DEF_H -#define __ASM_MPSPEC_DEF_H +#ifndef ASM_X86__MPSPEC_DEF_H +#define ASM_X86__MPSPEC_DEF_H /* * Structure definitions for SMP machines following the @@ -177,4 +177,4 @@ enum mp_bustype { MP_BUS_PCI, MP_BUS_MCA, }; -#endif +#endif /* ASM_X86__MPSPEC_DEF_H */ diff --git a/include/asm-x86/msgbuf.h b/include/asm-x86/msgbuf.h index 7e4e9481f51..1b538c907a3 100644 --- a/include/asm-x86/msgbuf.h +++ b/include/asm-x86/msgbuf.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_MSGBUF_H -#define _ASM_X86_MSGBUF_H +#ifndef ASM_X86__MSGBUF_H +#define ASM_X86__MSGBUF_H /* * The msqid64_ds structure for i386 architecture. @@ -36,4 +36,4 @@ struct msqid64_ds { unsigned long __unused5; }; -#endif /* _ASM_X86_MSGBUF_H */ +#endif /* ASM_X86__MSGBUF_H */ diff --git a/include/asm-x86/msidef.h b/include/asm-x86/msidef.h index 296f29ce426..3139666a94f 100644 --- a/include/asm-x86/msidef.h +++ b/include/asm-x86/msidef.h @@ -1,5 +1,5 @@ -#ifndef ASM_MSIDEF_H -#define ASM_MSIDEF_H +#ifndef ASM_X86__MSIDEF_H +#define ASM_X86__MSIDEF_H /* * Constants for Intel APIC based MSI messages. @@ -48,4 +48,4 @@ #define MSI_ADDR_DEST_ID(dest) (((dest) << MSI_ADDR_DEST_ID_SHIFT) & \ MSI_ADDR_DEST_ID_MASK) -#endif /* ASM_MSIDEF_H */ +#endif /* ASM_X86__MSIDEF_H */ diff --git a/include/asm-x86/msr-index.h b/include/asm-x86/msr-index.h index 44bce773012..3052f058ab0 100644 --- a/include/asm-x86/msr-index.h +++ b/include/asm-x86/msr-index.h @@ -1,5 +1,5 @@ -#ifndef __ASM_MSR_INDEX_H -#define __ASM_MSR_INDEX_H +#ifndef ASM_X86__MSR_INDEX_H +#define ASM_X86__MSR_INDEX_H /* CPU model specific register (MSR) numbers */ @@ -310,4 +310,4 @@ /* Geode defined MSRs */ #define MSR_GEODE_BUSCONT_CONF0 0x00001900 -#endif /* __ASM_MSR_INDEX_H */ +#endif /* ASM_X86__MSR_INDEX_H */ diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h index ca110ee73f0..032992035bd 100644 --- a/include/asm-x86/msr.h +++ b/include/asm-x86/msr.h @@ -1,5 +1,5 @@ -#ifndef __ASM_X86_MSR_H_ -#define __ASM_X86_MSR_H_ +#ifndef ASM_X86__MSR_H +#define ASM_X86__MSR_H #include @@ -220,4 +220,4 @@ static inline int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h) #endif /* __KERNEL__ */ -#endif +#endif /* ASM_X86__MSR_H */ diff --git a/include/asm-x86/mtrr.h b/include/asm-x86/mtrr.h index a69a01a5172..23a7f83da95 100644 --- a/include/asm-x86/mtrr.h +++ b/include/asm-x86/mtrr.h @@ -20,8 +20,8 @@ The postal address is: Richard Gooch, c/o ATNF, P. O. Box 76, Epping, N.S.W., 2121, Australia. */ -#ifndef _ASM_X86_MTRR_H -#define _ASM_X86_MTRR_H +#ifndef ASM_X86__MTRR_H +#define ASM_X86__MTRR_H #include #include @@ -170,4 +170,4 @@ struct mtrr_gentry32 { #endif /* __KERNEL__ */ -#endif /* _ASM_X86_MTRR_H */ +#endif /* ASM_X86__MTRR_H */ diff --git a/include/asm-x86/mutex_32.h b/include/asm-x86/mutex_32.h index 73e928ef5f0..25c16d8ba3c 100644 --- a/include/asm-x86/mutex_32.h +++ b/include/asm-x86/mutex_32.h @@ -6,8 +6,8 @@ * * Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar */ -#ifndef _ASM_MUTEX_H -#define _ASM_MUTEX_H +#ifndef ASM_X86__MUTEX_32_H +#define ASM_X86__MUTEX_32_H #include @@ -122,4 +122,4 @@ static inline int __mutex_fastpath_trylock(atomic_t *count, #endif } -#endif +#endif /* ASM_X86__MUTEX_32_H */ diff --git a/include/asm-x86/mutex_64.h b/include/asm-x86/mutex_64.h index f3fae9becb3..918ba21ab9d 100644 --- a/include/asm-x86/mutex_64.h +++ b/include/asm-x86/mutex_64.h @@ -6,8 +6,8 @@ * * Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar */ -#ifndef _ASM_MUTEX_H -#define _ASM_MUTEX_H +#ifndef ASM_X86__MUTEX_64_H +#define ASM_X86__MUTEX_64_H /** * __mutex_fastpath_lock - decrement and call function if negative @@ -97,4 +97,4 @@ static inline int __mutex_fastpath_trylock(atomic_t *count, return 0; } -#endif +#endif /* ASM_X86__MUTEX_64_H */ diff --git a/include/asm-x86/namei.h b/include/asm-x86/namei.h index 415ef5d9550..6e5de58c10c 100644 --- a/include/asm-x86/namei.h +++ b/include/asm-x86/namei.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_NAMEI_H -#define _ASM_X86_NAMEI_H +#ifndef ASM_X86__NAMEI_H +#define ASM_X86__NAMEI_H /* This dummy routine maybe changed to something useful * for /usr/gnemul/ emulation stuff. @@ -8,4 +8,4 @@ #define __emul_prefix() NULL -#endif /* _ASM_X86_NAMEI_H */ +#endif /* ASM_X86__NAMEI_H */ diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h index 21f8d0202a8..f8b76f38390 100644 --- a/include/asm-x86/nmi.h +++ b/include/asm-x86/nmi.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_NMI_H_ -#define _ASM_X86_NMI_H_ +#ifndef ASM_X86__NMI_H +#define ASM_X86__NMI_H #include #include @@ -81,4 +81,4 @@ void enable_lapic_nmi_watchdog(void); void stop_nmi(void); void restart_nmi(void); -#endif +#endif /* ASM_X86__NMI_H */ diff --git a/include/asm-x86/nops.h b/include/asm-x86/nops.h index ad0bedd10b8..ae742721ae7 100644 --- a/include/asm-x86/nops.h +++ b/include/asm-x86/nops.h @@ -1,5 +1,5 @@ -#ifndef _ASM_NOPS_H -#define _ASM_NOPS_H 1 +#ifndef ASM_X86__NOPS_H +#define ASM_X86__NOPS_H /* Define nops for use with alternative() */ @@ -115,4 +115,4 @@ #define ASM_NOP_MAX 8 -#endif +#endif /* ASM_X86__NOPS_H */ diff --git a/include/asm-x86/numa_32.h b/include/asm-x86/numa_32.h index 220d7b7707a..44cb07855c5 100644 --- a/include/asm-x86/numa_32.h +++ b/include/asm-x86/numa_32.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_32_NUMA_H -#define _ASM_X86_32_NUMA_H 1 +#ifndef ASM_X86__NUMA_32_H +#define ASM_X86__NUMA_32_H extern int pxm_to_nid(int pxm); extern void numa_remove_cpu(int cpu); @@ -8,4 +8,4 @@ extern void numa_remove_cpu(int cpu); extern void set_highmem_pages_init(void); #endif -#endif /* _ASM_X86_32_NUMA_H */ +#endif /* ASM_X86__NUMA_32_H */ diff --git a/include/asm-x86/numa_64.h b/include/asm-x86/numa_64.h index 3830094434a..15c990395b0 100644 --- a/include/asm-x86/numa_64.h +++ b/include/asm-x86/numa_64.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X8664_NUMA_H -#define _ASM_X8664_NUMA_H 1 +#ifndef ASM_X86__NUMA_64_H +#define ASM_X86__NUMA_64_H #include #include @@ -40,4 +40,4 @@ static inline void numa_add_cpu(int cpu, int node) { } static inline void numa_remove_cpu(int cpu) { } #endif -#endif +#endif /* ASM_X86__NUMA_64_H */ diff --git a/include/asm-x86/numaq.h b/include/asm-x86/numaq.h index 34b92d581fa..124bf7d4b70 100644 --- a/include/asm-x86/numaq.h +++ b/include/asm-x86/numaq.h @@ -23,8 +23,8 @@ * Send feedback to */ -#ifndef NUMAQ_H -#define NUMAQ_H +#ifndef ASM_X86__NUMAQ_H +#define ASM_X86__NUMAQ_H #ifdef CONFIG_X86_NUMAQ @@ -165,5 +165,5 @@ static inline int get_memcfg_numaq(void) return 0; } #endif /* CONFIG_X86_NUMAQ */ -#endif /* NUMAQ_H */ +#endif /* ASM_X86__NUMAQ_H */ diff --git a/include/asm-x86/olpc.h b/include/asm-x86/olpc.h index 97d47133486..d7328b1a05c 100644 --- a/include/asm-x86/olpc.h +++ b/include/asm-x86/olpc.h @@ -1,7 +1,7 @@ /* OLPC machine specific definitions */ -#ifndef ASM_OLPC_H_ -#define ASM_OLPC_H_ +#ifndef ASM_X86__OLPC_H +#define ASM_X86__OLPC_H #include @@ -129,4 +129,4 @@ extern int olpc_ec_mask_unset(uint8_t bits); #define OLPC_GPIO_LID geode_gpio(26) #define OLPC_GPIO_ECSCI geode_gpio(27) -#endif +#endif /* ASM_X86__OLPC_H */ diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h index 28d7b4533b1..068a636e0bb 100644 --- a/include/asm-x86/page.h +++ b/include/asm-x86/page.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PAGE_H -#define _ASM_X86_PAGE_H +#ifndef ASM_X86__PAGE_H +#define ASM_X86__PAGE_H #include @@ -192,4 +192,4 @@ static inline pteval_t native_pte_val(pte_t pte) #define __HAVE_ARCH_GATE_AREA 1 #endif /* __KERNEL__ */ -#endif /* _ASM_X86_PAGE_H */ +#endif /* ASM_X86__PAGE_H */ diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index ab8528793f0..85ac2b8ec7c 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PAGE_32_H -#define _ASM_X86_PAGE_32_H +#ifndef ASM_X86__PAGE_32_H +#define ASM_X86__PAGE_32_H /* * This handles the memory map. @@ -126,4 +126,4 @@ static inline void copy_page(void *to, void *from) #endif /* CONFIG_X86_3DNOW */ #endif /* !__ASSEMBLY__ */ -#endif /* _ASM_X86_PAGE_32_H */ +#endif /* ASM_X86__PAGE_32_H */ diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h index c6916c83e6b..87aef3c48b5 100644 --- a/include/asm-x86/page_64.h +++ b/include/asm-x86/page_64.h @@ -1,5 +1,5 @@ -#ifndef _X86_64_PAGE_H -#define _X86_64_PAGE_H +#ifndef ASM_X86__PAGE_64_H +#define ASM_X86__PAGE_64_H #define PAGETABLE_LEVELS 4 @@ -102,4 +102,4 @@ extern void init_extra_mapping_wb(unsigned long phys, unsigned long size); #endif -#endif /* _X86_64_PAGE_H */ +#endif /* ASM_X86__PAGE_64_H */ diff --git a/include/asm-x86/param.h b/include/asm-x86/param.h index 6f0d0422f4c..0009cfb11a5 100644 --- a/include/asm-x86/param.h +++ b/include/asm-x86/param.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PARAM_H -#define _ASM_X86_PARAM_H +#ifndef ASM_X86__PARAM_H +#define ASM_X86__PARAM_H #ifdef __KERNEL__ # define HZ CONFIG_HZ /* Internal kernel timer frequency */ @@ -19,4 +19,4 @@ #define MAXHOSTNAMELEN 64 /* max length of hostname */ -#endif /* _ASM_X86_PARAM_H */ +#endif /* ASM_X86__PARAM_H */ diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index aec9767836b..1b7eff0b41b 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -1,5 +1,5 @@ -#ifndef __ASM_PARAVIRT_H -#define __ASM_PARAVIRT_H +#ifndef ASM_X86__PARAVIRT_H +#define ASM_X86__PARAVIRT_H /* Various instructions on x86 need to be replaced for * para-virtualization: those hooks are defined here. */ @@ -1631,4 +1631,4 @@ static inline unsigned long __raw_local_irq_save(void) #endif /* __ASSEMBLY__ */ #endif /* CONFIG_PARAVIRT */ -#endif /* __ASM_PARAVIRT_H */ +#endif /* ASM_X86__PARAVIRT_H */ diff --git a/include/asm-x86/parport.h b/include/asm-x86/parport.h index 3c4ffeb467e..2e3dda4dc3d 100644 --- a/include/asm-x86/parport.h +++ b/include/asm-x86/parport.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PARPORT_H -#define _ASM_X86_PARPORT_H +#ifndef ASM_X86__PARPORT_H +#define ASM_X86__PARPORT_H static int __devinit parport_pc_find_isa_ports(int autoirq, int autodma); static int __devinit parport_pc_find_nonpci_ports(int autoirq, int autodma) @@ -7,4 +7,4 @@ static int __devinit parport_pc_find_nonpci_ports(int autoirq, int autodma) return parport_pc_find_isa_ports(autoirq, autodma); } -#endif /* _ASM_X86_PARPORT_H */ +#endif /* ASM_X86__PARPORT_H */ diff --git a/include/asm-x86/pat.h b/include/asm-x86/pat.h index 7edc4730721..482c3e3f987 100644 --- a/include/asm-x86/pat.h +++ b/include/asm-x86/pat.h @@ -1,5 +1,5 @@ -#ifndef _ASM_PAT_H -#define _ASM_PAT_H +#ifndef ASM_X86__PAT_H +#define ASM_X86__PAT_H #include @@ -19,4 +19,4 @@ extern int free_memtype(u64 start, u64 end); extern void pat_disable(char *reason); -#endif +#endif /* ASM_X86__PAT_H */ diff --git a/include/asm-x86/pci-direct.h b/include/asm-x86/pci-direct.h index 80c775d9fe2..da42be07b69 100644 --- a/include/asm-x86/pci-direct.h +++ b/include/asm-x86/pci-direct.h @@ -1,5 +1,5 @@ -#ifndef ASM_PCI_DIRECT_H -#define ASM_PCI_DIRECT_H 1 +#ifndef ASM_X86__PCI_DIRECT_H +#define ASM_X86__PCI_DIRECT_H #include @@ -18,4 +18,4 @@ extern int early_pci_allowed(void); extern unsigned int pci_early_dump_regs; extern void early_dump_pci_device(u8 bus, u8 slot, u8 func); extern void early_dump_pci_devices(void); -#endif +#endif /* ASM_X86__PCI_DIRECT_H */ diff --git a/include/asm-x86/pci.h b/include/asm-x86/pci.h index 2db14cf17db..60258319299 100644 --- a/include/asm-x86/pci.h +++ b/include/asm-x86/pci.h @@ -1,5 +1,5 @@ -#ifndef __x86_PCI_H -#define __x86_PCI_H +#ifndef ASM_X86__PCI_H +#define ASM_X86__PCI_H #include /* for struct page */ #include @@ -111,4 +111,4 @@ static inline cpumask_t __pcibus_to_cpumask(struct pci_bus *bus) } #endif -#endif +#endif /* ASM_X86__PCI_H */ diff --git a/include/asm-x86/pci_32.h b/include/asm-x86/pci_32.h index a50d4685128..3f2288207c0 100644 --- a/include/asm-x86/pci_32.h +++ b/include/asm-x86/pci_32.h @@ -1,5 +1,5 @@ -#ifndef __i386_PCI_H -#define __i386_PCI_H +#ifndef ASM_X86__PCI_32_H +#define ASM_X86__PCI_32_H #ifdef __KERNEL__ @@ -31,4 +31,4 @@ struct pci_dev; #endif /* __KERNEL__ */ -#endif /* __i386_PCI_H */ +#endif /* ASM_X86__PCI_32_H */ diff --git a/include/asm-x86/pci_64.h b/include/asm-x86/pci_64.h index f330234ffa5..f72e12d5770 100644 --- a/include/asm-x86/pci_64.h +++ b/include/asm-x86/pci_64.h @@ -1,5 +1,5 @@ -#ifndef __x8664_PCI_H -#define __x8664_PCI_H +#ifndef ASM_X86__PCI_64_H +#define ASM_X86__PCI_64_H #ifdef __KERNEL__ @@ -63,4 +63,4 @@ extern void pci_iommu_alloc(void); #endif /* __KERNEL__ */ -#endif /* __x8664_PCI_H */ +#endif /* ASM_X86__PCI_64_H */ diff --git a/include/asm-x86/pda.h b/include/asm-x86/pda.h index b34e9a7cc80..80860afffbd 100644 --- a/include/asm-x86/pda.h +++ b/include/asm-x86/pda.h @@ -1,5 +1,5 @@ -#ifndef X86_64_PDA_H -#define X86_64_PDA_H +#ifndef ASM_X86__PDA_H +#define ASM_X86__PDA_H #ifndef __ASSEMBLY__ #include @@ -134,4 +134,4 @@ do { \ #define PDA_STACKOFFSET (5*8) -#endif +#endif /* ASM_X86__PDA_H */ diff --git a/include/asm-x86/percpu.h b/include/asm-x86/percpu.h index 4e91ee1e37a..0afc8324807 100644 --- a/include/asm-x86/percpu.h +++ b/include/asm-x86/percpu.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PERCPU_H_ -#define _ASM_X86_PERCPU_H_ +#ifndef ASM_X86__PERCPU_H +#define ASM_X86__PERCPU_H #ifdef CONFIG_X86_64 #include @@ -215,4 +215,4 @@ do { \ #endif /* !CONFIG_SMP */ -#endif /* _ASM_X86_PERCPU_H_ */ +#endif /* ASM_X86__PERCPU_H */ diff --git a/include/asm-x86/pgalloc.h b/include/asm-x86/pgalloc.h index d63ea431cb3..3cd23adedae 100644 --- a/include/asm-x86/pgalloc.h +++ b/include/asm-x86/pgalloc.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PGALLOC_H -#define _ASM_X86_PGALLOC_H +#ifndef ASM_X86__PGALLOC_H +#define ASM_X86__PGALLOC_H #include #include /* for struct page */ @@ -111,4 +111,4 @@ extern void __pud_free_tlb(struct mmu_gather *tlb, pud_t *pud); #endif /* PAGETABLE_LEVELS > 3 */ #endif /* PAGETABLE_LEVELS > 2 */ -#endif /* _ASM_X86_PGALLOC_H */ +#endif /* ASM_X86__PGALLOC_H */ diff --git a/include/asm-x86/pgtable-2level-defs.h b/include/asm-x86/pgtable-2level-defs.h index 0f71c9f13da..7ec48f4e534 100644 --- a/include/asm-x86/pgtable-2level-defs.h +++ b/include/asm-x86/pgtable-2level-defs.h @@ -1,5 +1,5 @@ -#ifndef _I386_PGTABLE_2LEVEL_DEFS_H -#define _I386_PGTABLE_2LEVEL_DEFS_H +#ifndef ASM_X86__PGTABLE_2LEVEL_DEFS_H +#define ASM_X86__PGTABLE_2LEVEL_DEFS_H #define SHARED_KERNEL_PMD 0 @@ -17,4 +17,4 @@ #define PTRS_PER_PTE 1024 -#endif /* _I386_PGTABLE_2LEVEL_DEFS_H */ +#endif /* ASM_X86__PGTABLE_2LEVEL_DEFS_H */ diff --git a/include/asm-x86/pgtable-2level.h b/include/asm-x86/pgtable-2level.h index 46bc52c0eae..60440b19162 100644 --- a/include/asm-x86/pgtable-2level.h +++ b/include/asm-x86/pgtable-2level.h @@ -1,5 +1,5 @@ -#ifndef _I386_PGTABLE_2LEVEL_H -#define _I386_PGTABLE_2LEVEL_H +#ifndef ASM_X86__PGTABLE_2LEVEL_H +#define ASM_X86__PGTABLE_2LEVEL_H #define pte_ERROR(e) \ printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, (e).pte_low) @@ -78,4 +78,4 @@ static inline pte_t native_ptep_get_and_clear(pte_t *xp) #define __pte_to_swp_entry(pte) ((swp_entry_t) { (pte).pte_low }) #define __swp_entry_to_pte(x) ((pte_t) { .pte = (x).val }) -#endif /* _I386_PGTABLE_2LEVEL_H */ +#endif /* ASM_X86__PGTABLE_2LEVEL_H */ diff --git a/include/asm-x86/pgtable-3level-defs.h b/include/asm-x86/pgtable-3level-defs.h index 448ac951631..c05fe6ff372 100644 --- a/include/asm-x86/pgtable-3level-defs.h +++ b/include/asm-x86/pgtable-3level-defs.h @@ -1,5 +1,5 @@ -#ifndef _I386_PGTABLE_3LEVEL_DEFS_H -#define _I386_PGTABLE_3LEVEL_DEFS_H +#ifndef ASM_X86__PGTABLE_3LEVEL_DEFS_H +#define ASM_X86__PGTABLE_3LEVEL_DEFS_H #ifdef CONFIG_PARAVIRT #define SHARED_KERNEL_PMD (pv_info.shared_kernel_pmd) @@ -25,4 +25,4 @@ */ #define PTRS_PER_PTE 512 -#endif /* _I386_PGTABLE_3LEVEL_DEFS_H */ +#endif /* ASM_X86__PGTABLE_3LEVEL_DEFS_H */ diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h index c93dbb6c262..a9ad971a733 100644 --- a/include/asm-x86/pgtable-3level.h +++ b/include/asm-x86/pgtable-3level.h @@ -1,5 +1,5 @@ -#ifndef _I386_PGTABLE_3LEVEL_H -#define _I386_PGTABLE_3LEVEL_H +#ifndef ASM_X86__PGTABLE_3LEVEL_H +#define ASM_X86__PGTABLE_3LEVEL_H /* * Intel Physical Address Extension (PAE) Mode - three-level page @@ -179,4 +179,4 @@ static inline unsigned long pte_pfn(pte_t pte) #define __pte_to_swp_entry(pte) ((swp_entry_t){ (pte).pte_high }) #define __swp_entry_to_pte(x) ((pte_t){ { .pte_high = (x).val } }) -#endif /* _I386_PGTABLE_3LEVEL_H */ +#endif /* ASM_X86__PGTABLE_3LEVEL_H */ diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index 96aa76e691d..b8d7c530054 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PGTABLE_H -#define _ASM_X86_PGTABLE_H +#ifndef ASM_X86__PGTABLE_H +#define ASM_X86__PGTABLE_H #define FIRST_USER_ADDRESS 0 @@ -518,4 +518,4 @@ static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count) #include #endif /* __ASSEMBLY__ */ -#endif /* _ASM_X86_PGTABLE_H */ +#endif /* ASM_X86__PGTABLE_H */ diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h index 0611abf96a5..4fa3b046e1b 100644 --- a/include/asm-x86/pgtable_32.h +++ b/include/asm-x86/pgtable_32.h @@ -1,5 +1,5 @@ -#ifndef _I386_PGTABLE_H -#define _I386_PGTABLE_H +#ifndef ASM_X86__PGTABLE_32_H +#define ASM_X86__PGTABLE_32_H /* @@ -186,4 +186,4 @@ do { \ #define io_remap_pfn_range(vma, vaddr, pfn, size, prot) \ remap_pfn_range(vma, vaddr, pfn, size, prot) -#endif /* _I386_PGTABLE_H */ +#endif /* ASM_X86__PGTABLE_32_H */ diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h index 805d3128bfc..b80c0d7658c 100644 --- a/include/asm-x86/pgtable_64.h +++ b/include/asm-x86/pgtable_64.h @@ -1,5 +1,5 @@ -#ifndef _X86_64_PGTABLE_H -#define _X86_64_PGTABLE_H +#ifndef ASM_X86__PGTABLE_64_H +#define ASM_X86__PGTABLE_64_H #include #ifndef __ASSEMBLY__ @@ -284,4 +284,4 @@ extern void cleanup_highmap(void); #define __HAVE_ARCH_PTE_SAME #endif /* !__ASSEMBLY__ */ -#endif /* _X86_64_PGTABLE_H */ +#endif /* ASM_X86__PGTABLE_64_H */ diff --git a/include/asm-x86/posix_types_32.h b/include/asm-x86/posix_types_32.h index b031efda37e..70cf2bb0593 100644 --- a/include/asm-x86/posix_types_32.h +++ b/include/asm-x86/posix_types_32.h @@ -1,5 +1,5 @@ -#ifndef __ARCH_I386_POSIX_TYPES_H -#define __ARCH_I386_POSIX_TYPES_H +#ifndef ASM_X86__POSIX_TYPES_32_H +#define ASM_X86__POSIX_TYPES_32_H /* * This file is generally used by user-level software, so you need to @@ -82,4 +82,4 @@ do { \ #endif /* defined(__KERNEL__) */ -#endif +#endif /* ASM_X86__POSIX_TYPES_32_H */ diff --git a/include/asm-x86/posix_types_64.h b/include/asm-x86/posix_types_64.h index d6624c95854..388b4e7f4a4 100644 --- a/include/asm-x86/posix_types_64.h +++ b/include/asm-x86/posix_types_64.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_64_POSIX_TYPES_H -#define _ASM_X86_64_POSIX_TYPES_H +#ifndef ASM_X86__POSIX_TYPES_64_H +#define ASM_X86__POSIX_TYPES_64_H /* * This file is generally used by user-level software, so you need to @@ -116,4 +116,4 @@ static inline void __FD_ZERO(__kernel_fd_set *p) #endif /* defined(__KERNEL__) */ -#endif +#endif /* ASM_X86__POSIX_TYPES_64_H */ diff --git a/include/asm-x86/prctl.h b/include/asm-x86/prctl.h index 52952adef1c..e7ae34eb410 100644 --- a/include/asm-x86/prctl.h +++ b/include/asm-x86/prctl.h @@ -1,5 +1,5 @@ -#ifndef X86_64_PRCTL_H -#define X86_64_PRCTL_H 1 +#ifndef ASM_X86__PRCTL_H +#define ASM_X86__PRCTL_H #define ARCH_SET_GS 0x1001 #define ARCH_SET_FS 0x1002 @@ -7,4 +7,4 @@ #define ARCH_GET_GS 0x1004 -#endif +#endif /* ASM_X86__PRCTL_H */ diff --git a/include/asm-x86/processor-flags.h b/include/asm-x86/processor-flags.h index 092b39b3a7e..ae1d434f197 100644 --- a/include/asm-x86/processor-flags.h +++ b/include/asm-x86/processor-flags.h @@ -1,5 +1,5 @@ -#ifndef __ASM_I386_PROCESSOR_FLAGS_H -#define __ASM_I386_PROCESSOR_FLAGS_H +#ifndef ASM_X86__PROCESSOR_FLAGS_H +#define ASM_X86__PROCESSOR_FLAGS_H /* Various flags defined: can be included from assembler. */ /* @@ -94,4 +94,4 @@ #define X86_VM_MASK 0 /* No VM86 support */ #endif -#endif /* __ASM_I386_PROCESSOR_FLAGS_H */ +#endif /* ASM_X86__PROCESSOR_FLAGS_H */ diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index 15cb82a44e8..a9a28baef69 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -1,5 +1,5 @@ -#ifndef __ASM_X86_PROCESSOR_H -#define __ASM_X86_PROCESSOR_H +#ifndef ASM_X86__PROCESSOR_H +#define ASM_X86__PROCESSOR_H #include @@ -920,4 +920,4 @@ extern void start_thread(struct pt_regs *regs, unsigned long new_ip, extern int get_tsc_mode(unsigned long adr); extern int set_tsc_mode(unsigned int val); -#endif +#endif /* ASM_X86__PROCESSOR_H */ diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h index 3dd458c385c..6e89e8b4de0 100644 --- a/include/asm-x86/proto.h +++ b/include/asm-x86/proto.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X8664_PROTO_H -#define _ASM_X8664_PROTO_H 1 +#ifndef ASM_X86__PROTO_H +#define ASM_X86__PROTO_H #include @@ -29,4 +29,4 @@ long do_arch_prctl(struct task_struct *task, int code, unsigned long addr); #define round_up(x, y) (((x) + (y) - 1) & ~((y) - 1)) #define round_down(x, y) ((x) & ~((y) - 1)) -#endif +#endif /* ASM_X86__PROTO_H */ diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h index 72e7b9db29b..d0cf3344a58 100644 --- a/include/asm-x86/ptrace-abi.h +++ b/include/asm-x86/ptrace-abi.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PTRACE_ABI_H -#define _ASM_X86_PTRACE_ABI_H +#ifndef ASM_X86__PTRACE_ABI_H +#define ASM_X86__PTRACE_ABI_H #ifdef __i386__ @@ -140,4 +140,4 @@ struct ptrace_bts_config { Returns number of BTS records drained. */ -#endif +#endif /* ASM_X86__PTRACE_ABI_H */ diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h index 8a71db803da..d464f252edc 100644 --- a/include/asm-x86/ptrace.h +++ b/include/asm-x86/ptrace.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PTRACE_H -#define _ASM_X86_PTRACE_H +#ifndef ASM_X86__PTRACE_H +#define ASM_X86__PTRACE_H #include /* For __user */ #include @@ -239,4 +239,4 @@ extern int do_set_thread_area(struct task_struct *p, int idx, #endif /* !__ASSEMBLY__ */ -#endif +#endif /* ASM_X86__PTRACE_H */ diff --git a/include/asm-x86/pvclock-abi.h b/include/asm-x86/pvclock-abi.h index 6857f840b24..edb3b4ecfc8 100644 --- a/include/asm-x86/pvclock-abi.h +++ b/include/asm-x86/pvclock-abi.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PVCLOCK_ABI_H_ -#define _ASM_X86_PVCLOCK_ABI_H_ +#ifndef ASM_X86__PVCLOCK_ABI_H +#define ASM_X86__PVCLOCK_ABI_H #ifndef __ASSEMBLY__ /* @@ -39,4 +39,4 @@ struct pvclock_wall_clock { } __attribute__((__packed__)); #endif /* __ASSEMBLY__ */ -#endif /* _ASM_X86_PVCLOCK_ABI_H_ */ +#endif /* ASM_X86__PVCLOCK_ABI_H */ diff --git a/include/asm-x86/pvclock.h b/include/asm-x86/pvclock.h index 85b1bba8e0a..1a38f683480 100644 --- a/include/asm-x86/pvclock.h +++ b/include/asm-x86/pvclock.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_PVCLOCK_H_ -#define _ASM_X86_PVCLOCK_H_ +#ifndef ASM_X86__PVCLOCK_H +#define ASM_X86__PVCLOCK_H #include #include @@ -10,4 +10,4 @@ void pvclock_read_wallclock(struct pvclock_wall_clock *wall, struct pvclock_vcpu_time_info *vcpu, struct timespec *ts); -#endif /* _ASM_X86_PVCLOCK_H_ */ +#endif /* ASM_X86__PVCLOCK_H */ diff --git a/include/asm-x86/reboot.h b/include/asm-x86/reboot.h index 206f355786d..1c2f0ce9e31 100644 --- a/include/asm-x86/reboot.h +++ b/include/asm-x86/reboot.h @@ -1,5 +1,5 @@ -#ifndef _ASM_REBOOT_H -#define _ASM_REBOOT_H +#ifndef ASM_X86__REBOOT_H +#define ASM_X86__REBOOT_H struct pt_regs; @@ -18,4 +18,4 @@ void native_machine_crash_shutdown(struct pt_regs *regs); void native_machine_shutdown(void); void machine_real_restart(const unsigned char *code, int length); -#endif /* _ASM_REBOOT_H */ +#endif /* ASM_X86__REBOOT_H */ diff --git a/include/asm-x86/reboot_fixups.h b/include/asm-x86/reboot_fixups.h index 0cb7d87c2b6..2c2987d9757 100644 --- a/include/asm-x86/reboot_fixups.h +++ b/include/asm-x86/reboot_fixups.h @@ -1,6 +1,6 @@ -#ifndef _LINUX_REBOOT_FIXUPS_H -#define _LINUX_REBOOT_FIXUPS_H +#ifndef ASM_X86__REBOOT_FIXUPS_H +#define ASM_X86__REBOOT_FIXUPS_H extern void mach_reboot_fixups(void); -#endif /* _LINUX_REBOOT_FIXUPS_H */ +#endif /* ASM_X86__REBOOT_FIXUPS_H */ diff --git a/include/asm-x86/required-features.h b/include/asm-x86/required-features.h index adec887dd7c..d6822e099c5 100644 --- a/include/asm-x86/required-features.h +++ b/include/asm-x86/required-features.h @@ -1,5 +1,5 @@ -#ifndef _ASM_REQUIRED_FEATURES_H -#define _ASM_REQUIRED_FEATURES_H 1 +#ifndef ASM_X86__REQUIRED_FEATURES_H +#define ASM_X86__REQUIRED_FEATURES_H /* Define minimum CPUID feature set for kernel These bits are checked really early to actually display a visible error message before the @@ -73,4 +73,4 @@ #define REQUIRED_MASK6 0 #define REQUIRED_MASK7 0 -#endif +#endif /* ASM_X86__REQUIRED_FEATURES_H */ diff --git a/include/asm-x86/resume-trace.h b/include/asm-x86/resume-trace.h index 8d9f0b41ee8..519a8ecbfc9 100644 --- a/include/asm-x86/resume-trace.h +++ b/include/asm-x86/resume-trace.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_RESUME_TRACE_H -#define _ASM_X86_RESUME_TRACE_H +#ifndef ASM_X86__RESUME_TRACE_H +#define ASM_X86__RESUME_TRACE_H #include @@ -18,4 +18,4 @@ do { \ } \ } while (0) -#endif +#endif /* ASM_X86__RESUME_TRACE_H */ diff --git a/include/asm-x86/rio.h b/include/asm-x86/rio.h index c9448bd8968..5e1256bdee8 100644 --- a/include/asm-x86/rio.h +++ b/include/asm-x86/rio.h @@ -5,8 +5,8 @@ * Author: Laurent Vivier */ -#ifndef __ASM_RIO_H -#define __ASM_RIO_H +#ifndef ASM_X86__RIO_H +#define ASM_X86__RIO_H #define RIO_TABLE_VERSION 3 @@ -60,4 +60,4 @@ enum { ALT_CALGARY = 5, /* Second Planar Calgary */ }; -#endif /* __ASM_RIO_H */ +#endif /* ASM_X86__RIO_H */ diff --git a/include/asm-x86/rwlock.h b/include/asm-x86/rwlock.h index 6a8c0d64510..48a3109e1a7 100644 --- a/include/asm-x86/rwlock.h +++ b/include/asm-x86/rwlock.h @@ -1,8 +1,8 @@ -#ifndef _ASM_X86_RWLOCK_H -#define _ASM_X86_RWLOCK_H +#ifndef ASM_X86__RWLOCK_H +#define ASM_X86__RWLOCK_H #define RW_LOCK_BIAS 0x01000000 /* Actual code is in asm/spinlock.h or in arch/x86/lib/rwlock.S */ -#endif /* _ASM_X86_RWLOCK_H */ +#endif /* ASM_X86__RWLOCK_H */ diff --git a/include/asm-x86/rwsem.h b/include/asm-x86/rwsem.h index 750f2a3542b..3ff3015b71a 100644 --- a/include/asm-x86/rwsem.h +++ b/include/asm-x86/rwsem.h @@ -29,8 +29,8 @@ * front, then they'll all be woken up, but no other readers will be. */ -#ifndef _I386_RWSEM_H -#define _I386_RWSEM_H +#ifndef ASM_X86__RWSEM_H +#define ASM_X86__RWSEM_H #ifndef _LINUX_RWSEM_H #error "please don't include asm/rwsem.h directly, use linux/rwsem.h instead" @@ -262,4 +262,4 @@ static inline int rwsem_is_locked(struct rw_semaphore *sem) } #endif /* __KERNEL__ */ -#endif /* _I386_RWSEM_H */ +#endif /* ASM_X86__RWSEM_H */ diff --git a/include/asm-x86/scatterlist.h b/include/asm-x86/scatterlist.h index c0432061f81..ee48f880005 100644 --- a/include/asm-x86/scatterlist.h +++ b/include/asm-x86/scatterlist.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SCATTERLIST_H -#define _ASM_X86_SCATTERLIST_H +#ifndef ASM_X86__SCATTERLIST_H +#define ASM_X86__SCATTERLIST_H #include @@ -30,4 +30,4 @@ struct scatterlist { # define sg_dma_len(sg) ((sg)->dma_length) #endif -#endif +#endif /* ASM_X86__SCATTERLIST_H */ diff --git a/include/asm-x86/seccomp_32.h b/include/asm-x86/seccomp_32.h index 36e71c5f306..cf9ab2dbcef 100644 --- a/include/asm-x86/seccomp_32.h +++ b/include/asm-x86/seccomp_32.h @@ -1,5 +1,5 @@ -#ifndef _ASM_SECCOMP_H -#define _ASM_SECCOMP_H +#ifndef ASM_X86__SECCOMP_32_H +#define ASM_X86__SECCOMP_32_H #include @@ -14,4 +14,4 @@ #define __NR_seccomp_exit __NR_exit #define __NR_seccomp_sigreturn __NR_sigreturn -#endif /* _ASM_SECCOMP_H */ +#endif /* ASM_X86__SECCOMP_32_H */ diff --git a/include/asm-x86/seccomp_64.h b/include/asm-x86/seccomp_64.h index 76cfe69aa63..03274cea751 100644 --- a/include/asm-x86/seccomp_64.h +++ b/include/asm-x86/seccomp_64.h @@ -1,5 +1,5 @@ -#ifndef _ASM_SECCOMP_H -#define _ASM_SECCOMP_H +#ifndef ASM_X86__SECCOMP_64_H +#define ASM_X86__SECCOMP_64_H #include @@ -22,4 +22,4 @@ #define __NR_seccomp_exit_32 __NR_ia32_exit #define __NR_seccomp_sigreturn_32 __NR_ia32_sigreturn -#endif /* _ASM_SECCOMP_H */ +#endif /* ASM_X86__SECCOMP_64_H */ diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h index 646452ea9ea..ea5f0a8686f 100644 --- a/include/asm-x86/segment.h +++ b/include/asm-x86/segment.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SEGMENT_H_ -#define _ASM_X86_SEGMENT_H_ +#ifndef ASM_X86__SEGMENT_H +#define ASM_X86__SEGMENT_H /* Constructor for a conventional segment GDT (or LDT) entry */ /* This is a macro so it can be used in initializers */ @@ -212,4 +212,4 @@ extern const char early_idt_handlers[NUM_EXCEPTION_VECTORS][10]; #endif #endif -#endif +#endif /* ASM_X86__SEGMENT_H */ diff --git a/include/asm-x86/sembuf.h b/include/asm-x86/sembuf.h index ee50c801f7b..81f06b7e5a3 100644 --- a/include/asm-x86/sembuf.h +++ b/include/asm-x86/sembuf.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SEMBUF_H -#define _ASM_X86_SEMBUF_H +#ifndef ASM_X86__SEMBUF_H +#define ASM_X86__SEMBUF_H /* * The semid64_ds structure for x86 architecture. @@ -21,4 +21,4 @@ struct semid64_ds { unsigned long __unused4; }; -#endif /* _ASM_X86_SEMBUF_H */ +#endif /* ASM_X86__SEMBUF_H */ diff --git a/include/asm-x86/serial.h b/include/asm-x86/serial.h index 628c801535e..303660b671e 100644 --- a/include/asm-x86/serial.h +++ b/include/asm-x86/serial.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SERIAL_H -#define _ASM_X86_SERIAL_H +#ifndef ASM_X86__SERIAL_H +#define ASM_X86__SERIAL_H /* * This assumes you have a 1.8432 MHz clock for your UART. @@ -26,4 +26,4 @@ { 0, BASE_BAUD, 0x3E8, 4, STD_COM_FLAGS }, /* ttyS2 */ \ { 0, BASE_BAUD, 0x2E8, 3, STD_COM4_FLAGS }, /* ttyS3 */ -#endif /* _ASM_X86_SERIAL_H */ +#endif /* ASM_X86__SERIAL_H */ diff --git a/include/asm-x86/setup.h b/include/asm-x86/setup.h index a07c6f1c01e..b7aeba94e43 100644 --- a/include/asm-x86/setup.h +++ b/include/asm-x86/setup.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SETUP_H -#define _ASM_X86_SETUP_H +#ifndef ASM_X86__SETUP_H +#define ASM_X86__SETUP_H #define COMMAND_LINE_SIZE 2048 @@ -100,4 +100,4 @@ void __init x86_64_start_reservations(char *real_mode_data); #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ -#endif /* _ASM_X86_SETUP_H */ +#endif /* ASM_X86__SETUP_H */ diff --git a/include/asm-x86/shmbuf.h b/include/asm-x86/shmbuf.h index b51413b7497..f51aec2298e 100644 --- a/include/asm-x86/shmbuf.h +++ b/include/asm-x86/shmbuf.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SHMBUF_H -#define _ASM_X86_SHMBUF_H +#ifndef ASM_X86__SHMBUF_H +#define ASM_X86__SHMBUF_H /* * The shmid64_ds structure for x86 architecture. @@ -48,4 +48,4 @@ struct shminfo64 { unsigned long __unused4; }; -#endif /* _ASM_X86_SHMBUF_H */ +#endif /* ASM_X86__SHMBUF_H */ diff --git a/include/asm-x86/shmparam.h b/include/asm-x86/shmparam.h index 0880cf0917b..a83a1fd96a0 100644 --- a/include/asm-x86/shmparam.h +++ b/include/asm-x86/shmparam.h @@ -1,6 +1,6 @@ -#ifndef _ASM_X86_SHMPARAM_H -#define _ASM_X86_SHMPARAM_H +#ifndef ASM_X86__SHMPARAM_H +#define ASM_X86__SHMPARAM_H #define SHMLBA PAGE_SIZE /* attach addr a multiple of this */ -#endif /* _ASM_X86_SHMPARAM_H */ +#endif /* ASM_X86__SHMPARAM_H */ diff --git a/include/asm-x86/sigcontext.h b/include/asm-x86/sigcontext.h index 2f9c884d2c0..24879c85b29 100644 --- a/include/asm-x86/sigcontext.h +++ b/include/asm-x86/sigcontext.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SIGCONTEXT_H -#define _ASM_X86_SIGCONTEXT_H +#ifndef ASM_X86__SIGCONTEXT_H +#define ASM_X86__SIGCONTEXT_H #include #include @@ -202,4 +202,4 @@ struct sigcontext { #endif /* !__i386__ */ -#endif +#endif /* ASM_X86__SIGCONTEXT_H */ diff --git a/include/asm-x86/sigcontext32.h b/include/asm-x86/sigcontext32.h index 57a9686fb49..4e2ec732dd0 100644 --- a/include/asm-x86/sigcontext32.h +++ b/include/asm-x86/sigcontext32.h @@ -1,5 +1,5 @@ -#ifndef _SIGCONTEXT32_H -#define _SIGCONTEXT32_H 1 +#ifndef ASM_X86__SIGCONTEXT32_H +#define ASM_X86__SIGCONTEXT32_H /* signal context for 32bit programs. */ @@ -68,4 +68,4 @@ struct sigcontext_ia32 { unsigned int cr2; }; -#endif +#endif /* ASM_X86__SIGCONTEXT32_H */ diff --git a/include/asm-x86/siginfo.h b/include/asm-x86/siginfo.h index a477bea0c2a..808bdfb2958 100644 --- a/include/asm-x86/siginfo.h +++ b/include/asm-x86/siginfo.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SIGINFO_H -#define _ASM_X86_SIGINFO_H +#ifndef ASM_X86__SIGINFO_H +#define ASM_X86__SIGINFO_H #ifdef __x86_64__ # define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) @@ -7,4 +7,4 @@ #include -#endif +#endif /* ASM_X86__SIGINFO_H */ diff --git a/include/asm-x86/signal.h b/include/asm-x86/signal.h index 6dac49364e9..6ee32822779 100644 --- a/include/asm-x86/signal.h +++ b/include/asm-x86/signal.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SIGNAL_H -#define _ASM_X86_SIGNAL_H +#ifndef ASM_X86__SIGNAL_H +#define ASM_X86__SIGNAL_H #ifndef __ASSEMBLY__ #include @@ -256,4 +256,4 @@ struct pt_regs; #endif /* __KERNEL__ */ #endif /* __ASSEMBLY__ */ -#endif +#endif /* ASM_X86__SIGNAL_H */ diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 3c877f74f27..cd3c5f03fca 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SMP_H_ -#define _ASM_X86_SMP_H_ +#ifndef ASM_X86__SMP_H +#define ASM_X86__SMP_H #ifndef __ASSEMBLY__ #include #include @@ -205,4 +205,4 @@ extern void cpu_uninit(void); #endif #endif /* __ASSEMBLY__ */ -#endif +#endif /* ASM_X86__SMP_H */ diff --git a/include/asm-x86/socket.h b/include/asm-x86/socket.h index 80af9c4ccad..db73274c83c 100644 --- a/include/asm-x86/socket.h +++ b/include/asm-x86/socket.h @@ -1,5 +1,5 @@ -#ifndef _ASM_SOCKET_H -#define _ASM_SOCKET_H +#ifndef ASM_X86__SOCKET_H +#define ASM_X86__SOCKET_H #include @@ -54,4 +54,4 @@ #define SO_MARK 36 -#endif /* _ASM_SOCKET_H */ +#endif /* ASM_X86__SOCKET_H */ diff --git a/include/asm-x86/sockios.h b/include/asm-x86/sockios.h index 49cc72b5d3c..a006704fdc8 100644 --- a/include/asm-x86/sockios.h +++ b/include/asm-x86/sockios.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SOCKIOS_H -#define _ASM_X86_SOCKIOS_H +#ifndef ASM_X86__SOCKIOS_H +#define ASM_X86__SOCKIOS_H /* Socket-level I/O control calls. */ #define FIOSETOWN 0x8901 @@ -10,4 +10,4 @@ #define SIOCGSTAMP 0x8906 /* Get stamp (timeval) */ #define SIOCGSTAMPNS 0x8907 /* Get stamp (timespec) */ -#endif /* _ASM_X86_SOCKIOS_H */ +#endif /* ASM_X86__SOCKIOS_H */ diff --git a/include/asm-x86/sparsemem.h b/include/asm-x86/sparsemem.h index 9bd48b0a534..38f8e6bc318 100644 --- a/include/asm-x86/sparsemem.h +++ b/include/asm-x86/sparsemem.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SPARSEMEM_H -#define _ASM_X86_SPARSEMEM_H +#ifndef ASM_X86__SPARSEMEM_H +#define ASM_X86__SPARSEMEM_H #ifdef CONFIG_SPARSEMEM /* @@ -31,4 +31,4 @@ #endif #endif /* CONFIG_SPARSEMEM */ -#endif +#endif /* ASM_X86__SPARSEMEM_H */ diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h index 4f9a9861799..cbe01086ba6 100644 --- a/include/asm-x86/spinlock.h +++ b/include/asm-x86/spinlock.h @@ -1,5 +1,5 @@ -#ifndef _X86_SPINLOCK_H_ -#define _X86_SPINLOCK_H_ +#ifndef ASM_X86__SPINLOCK_H +#define ASM_X86__SPINLOCK_H #include #include @@ -366,4 +366,4 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw) #define _raw_read_relax(lock) cpu_relax() #define _raw_write_relax(lock) cpu_relax() -#endif +#endif /* ASM_X86__SPINLOCK_H */ diff --git a/include/asm-x86/spinlock_types.h b/include/asm-x86/spinlock_types.h index 06c071c9eee..6aa9b562c50 100644 --- a/include/asm-x86/spinlock_types.h +++ b/include/asm-x86/spinlock_types.h @@ -1,5 +1,5 @@ -#ifndef __ASM_SPINLOCK_TYPES_H -#define __ASM_SPINLOCK_TYPES_H +#ifndef ASM_X86__SPINLOCK_TYPES_H +#define ASM_X86__SPINLOCK_TYPES_H #ifndef __LINUX_SPINLOCK_TYPES_H # error "please don't include this file directly" @@ -17,4 +17,4 @@ typedef struct { #define __RAW_RW_LOCK_UNLOCKED { RW_LOCK_BIAS } -#endif +#endif /* ASM_X86__SPINLOCK_TYPES_H */ diff --git a/include/asm-x86/srat.h b/include/asm-x86/srat.h index 774c919dc23..5363e4f7e1c 100644 --- a/include/asm-x86/srat.h +++ b/include/asm-x86/srat.h @@ -24,8 +24,8 @@ * Send feedback to Pat Gaughen */ -#ifndef _ASM_SRAT_H_ -#define _ASM_SRAT_H_ +#ifndef ASM_X86__SRAT_H +#define ASM_X86__SRAT_H #ifdef CONFIG_ACPI_NUMA extern int get_memcfg_from_srat(void); @@ -36,4 +36,4 @@ static inline int get_memcfg_from_srat(void) } #endif -#endif /* _ASM_SRAT_H_ */ +#endif /* ASM_X86__SRAT_H */ diff --git a/include/asm-x86/stacktrace.h b/include/asm-x86/stacktrace.h index 30f82526a8e..f43517e2853 100644 --- a/include/asm-x86/stacktrace.h +++ b/include/asm-x86/stacktrace.h @@ -1,5 +1,5 @@ -#ifndef _ASM_STACKTRACE_H -#define _ASM_STACKTRACE_H 1 +#ifndef ASM_X86__STACKTRACE_H +#define ASM_X86__STACKTRACE_H extern int kstack_depth_to_print; @@ -18,4 +18,4 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack, unsigned long bp, const struct stacktrace_ops *ops, void *data); -#endif +#endif /* ASM_X86__STACKTRACE_H */ diff --git a/include/asm-x86/stat.h b/include/asm-x86/stat.h index 5c22dcb5d17..1e120f62890 100644 --- a/include/asm-x86/stat.h +++ b/include/asm-x86/stat.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_STAT_H -#define _ASM_X86_STAT_H +#ifndef ASM_X86__STAT_H +#define ASM_X86__STAT_H #define STAT_HAVE_NSEC 1 @@ -111,4 +111,4 @@ struct __old_kernel_stat { #endif }; -#endif +#endif /* ASM_X86__STAT_H */ diff --git a/include/asm-x86/statfs.h b/include/asm-x86/statfs.h index 7c651aa9725..3f005bc3aa5 100644 --- a/include/asm-x86/statfs.h +++ b/include/asm-x86/statfs.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_STATFS_H -#define _ASM_X86_STATFS_H +#ifndef ASM_X86__STATFS_H +#define ASM_X86__STATFS_H #ifdef __i386__ #include @@ -60,4 +60,4 @@ struct compat_statfs64 { } __attribute__((packed)); #endif /* !__i386__ */ -#endif +#endif /* ASM_X86__STATFS_H */ diff --git a/include/asm-x86/string_32.h b/include/asm-x86/string_32.h index 193578cd1fd..487843ed245 100644 --- a/include/asm-x86/string_32.h +++ b/include/asm-x86/string_32.h @@ -1,5 +1,5 @@ -#ifndef _I386_STRING_H_ -#define _I386_STRING_H_ +#ifndef ASM_X86__STRING_32_H +#define ASM_X86__STRING_32_H #ifdef __KERNEL__ @@ -323,4 +323,4 @@ extern void *memscan(void *addr, int c, size_t size); #endif /* __KERNEL__ */ -#endif +#endif /* ASM_X86__STRING_32_H */ diff --git a/include/asm-x86/string_64.h b/include/asm-x86/string_64.h index 52b5ab38339..a2add11d3b6 100644 --- a/include/asm-x86/string_64.h +++ b/include/asm-x86/string_64.h @@ -1,5 +1,5 @@ -#ifndef _X86_64_STRING_H_ -#define _X86_64_STRING_H_ +#ifndef ASM_X86__STRING_64_H +#define ASM_X86__STRING_64_H #ifdef __KERNEL__ @@ -57,4 +57,4 @@ int strcmp(const char *cs, const char *ct); #endif /* __KERNEL__ */ -#endif +#endif /* ASM_X86__STRING_64_H */ diff --git a/include/asm-x86/suspend_32.h b/include/asm-x86/suspend_32.h index 8675c6782a7..acb6d4d491f 100644 --- a/include/asm-x86/suspend_32.h +++ b/include/asm-x86/suspend_32.h @@ -3,8 +3,8 @@ * Based on code * Copyright 2001 Patrick Mochel */ -#ifndef __ASM_X86_32_SUSPEND_H -#define __ASM_X86_32_SUSPEND_H +#ifndef ASM_X86__SUSPEND_32_H +#define ASM_X86__SUSPEND_32_H #include #include @@ -48,4 +48,4 @@ static inline void acpi_save_register_state(unsigned long return_point) extern int acpi_save_state_mem(void); #endif -#endif /* __ASM_X86_32_SUSPEND_H */ +#endif /* ASM_X86__SUSPEND_32_H */ diff --git a/include/asm-x86/suspend_64.h b/include/asm-x86/suspend_64.h index dc3262b4307..cf821dd310e 100644 --- a/include/asm-x86/suspend_64.h +++ b/include/asm-x86/suspend_64.h @@ -3,8 +3,8 @@ * Based on code * Copyright 2001 Patrick Mochel */ -#ifndef __ASM_X86_64_SUSPEND_H -#define __ASM_X86_64_SUSPEND_H +#ifndef ASM_X86__SUSPEND_64_H +#define ASM_X86__SUSPEND_64_H #include #include @@ -49,4 +49,4 @@ extern int acpi_save_state_mem(void); extern char core_restore_code; extern char restore_registers; -#endif /* __ASM_X86_64_SUSPEND_H */ +#endif /* ASM_X86__SUSPEND_64_H */ diff --git a/include/asm-x86/swiotlb.h b/include/asm-x86/swiotlb.h index c706a744263..9486c400a71 100644 --- a/include/asm-x86/swiotlb.h +++ b/include/asm-x86/swiotlb.h @@ -1,5 +1,5 @@ -#ifndef _ASM_SWIOTLB_H -#define _ASM_SWIOTLB_H 1 +#ifndef ASM_X86__SWIOTLB_H +#define ASM_X86__SWIOTLB_H #include @@ -55,4 +55,4 @@ static inline void pci_swiotlb_init(void) static inline void dma_mark_clean(void *addr, size_t size) {} -#endif /* _ASM_SWIOTLB_H */ +#endif /* ASM_X86__SWIOTLB_H */ diff --git a/include/asm-x86/sync_bitops.h b/include/asm-x86/sync_bitops.h index b47a1d0b8a8..b689bee7110 100644 --- a/include/asm-x86/sync_bitops.h +++ b/include/asm-x86/sync_bitops.h @@ -1,5 +1,5 @@ -#ifndef _I386_SYNC_BITOPS_H -#define _I386_SYNC_BITOPS_H +#ifndef ASM_X86__SYNC_BITOPS_H +#define ASM_X86__SYNC_BITOPS_H /* * Copyright 1992, Linus Torvalds. @@ -127,4 +127,4 @@ static inline int sync_test_and_change_bit(int nr, volatile unsigned long *addr) #undef ADDR -#endif /* _I386_SYNC_BITOPS_H */ +#endif /* ASM_X86__SYNC_BITOPS_H */ diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h index 983ce37c491..34505dd7b24 100644 --- a/include/asm-x86/system.h +++ b/include/asm-x86/system.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_SYSTEM_H_ -#define _ASM_X86_SYSTEM_H_ +#ifndef ASM_X86__SYSTEM_H +#define ASM_X86__SYSTEM_H #include #include @@ -419,4 +419,4 @@ static inline void rdtsc_barrier(void) alternative(ASM_NOP3, "lfence", X86_FEATURE_LFENCE_RDTSC); } -#endif +#endif /* ASM_X86__SYSTEM_H */ diff --git a/include/asm-x86/system_64.h b/include/asm-x86/system_64.h index 97fa251ccb2..5aedb8bffc5 100644 --- a/include/asm-x86/system_64.h +++ b/include/asm-x86/system_64.h @@ -1,5 +1,5 @@ -#ifndef __ASM_SYSTEM_H -#define __ASM_SYSTEM_H +#ifndef ASM_X86__SYSTEM_64_H +#define ASM_X86__SYSTEM_64_H #include #include @@ -19,4 +19,4 @@ static inline void write_cr8(unsigned long val) #include -#endif +#endif /* ASM_X86__SYSTEM_64_H */ diff --git a/include/asm-x86/tce.h b/include/asm-x86/tce.h index b1a4ea00df7..e7932d7fbba 100644 --- a/include/asm-x86/tce.h +++ b/include/asm-x86/tce.h @@ -21,8 +21,8 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#ifndef _ASM_X86_64_TCE_H -#define _ASM_X86_64_TCE_H +#ifndef ASM_X86__TCE_H +#define ASM_X86__TCE_H extern unsigned int specified_table_size; struct iommu_table; @@ -45,4 +45,4 @@ extern void * __init alloc_tce_table(void); extern void __init free_tce_table(void *tbl); extern int __init build_tce_table(struct pci_dev *dev, void __iomem *bbar); -#endif /* _ASM_X86_64_TCE_H */ +#endif /* ASM_X86__TCE_H */ diff --git a/include/asm-x86/termbits.h b/include/asm-x86/termbits.h index af1b70ea440..3d00dc5e0c7 100644 --- a/include/asm-x86/termbits.h +++ b/include/asm-x86/termbits.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_TERMBITS_H -#define _ASM_X86_TERMBITS_H +#ifndef ASM_X86__TERMBITS_H +#define ASM_X86__TERMBITS_H #include @@ -195,4 +195,4 @@ struct ktermios { #define TCSADRAIN 1 #define TCSAFLUSH 2 -#endif /* _ASM_X86_TERMBITS_H */ +#endif /* ASM_X86__TERMBITS_H */ diff --git a/include/asm-x86/termios.h b/include/asm-x86/termios.h index f72956331c4..e235db24807 100644 --- a/include/asm-x86/termios.h +++ b/include/asm-x86/termios.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_TERMIOS_H -#define _ASM_X86_TERMIOS_H +#ifndef ASM_X86__TERMIOS_H +#define ASM_X86__TERMIOS_H #include #include @@ -110,4 +110,4 @@ static inline int kernel_termios_to_user_termios_1(struct termios __user *u, #endif /* __KERNEL__ */ -#endif /* _ASM_X86_TERMIOS_H */ +#endif /* ASM_X86__TERMIOS_H */ diff --git a/include/asm-x86/therm_throt.h b/include/asm-x86/therm_throt.h index 399bf6026b1..1c7f57b6b66 100644 --- a/include/asm-x86/therm_throt.h +++ b/include/asm-x86/therm_throt.h @@ -1,9 +1,9 @@ -#ifndef __ASM_I386_THERM_THROT_H__ -#define __ASM_I386_THERM_THROT_H__ 1 +#ifndef ASM_X86__THERM_THROT_H +#define ASM_X86__THERM_THROT_H #include extern atomic_t therm_throt_en; int therm_throt_process(int curr); -#endif /* __ASM_I386_THERM_THROT_H__ */ +#endif /* ASM_X86__THERM_THROT_H */ diff --git a/include/asm-x86/thread_info.h b/include/asm-x86/thread_info.h index 0a8f27d31d0..50315a5dd18 100644 --- a/include/asm-x86/thread_info.h +++ b/include/asm-x86/thread_info.h @@ -4,8 +4,8 @@ * - Incorporating suggestions made by Linus Torvalds and Dave Miller */ -#ifndef _ASM_X86_THREAD_INFO_H -#define _ASM_X86_THREAD_INFO_H +#ifndef ASM_X86__THREAD_INFO_H +#define ASM_X86__THREAD_INFO_H #include #include @@ -258,4 +258,4 @@ extern void free_thread_info(struct thread_info *ti); extern int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src); #define arch_task_cache_init arch_task_cache_init #endif -#endif /* _ASM_X86_THREAD_INFO_H */ +#endif /* ASM_X86__THREAD_INFO_H */ diff --git a/include/asm-x86/time.h b/include/asm-x86/time.h index a17fa473e91..7457a5d7b4a 100644 --- a/include/asm-x86/time.h +++ b/include/asm-x86/time.h @@ -1,5 +1,5 @@ -#ifndef _ASMX86_TIME_H -#define _ASMX86_TIME_H +#ifndef ASM_X86__TIME_H +#define ASM_X86__TIME_H extern void hpet_time_init(void); @@ -58,4 +58,4 @@ static inline int native_set_wallclock(unsigned long nowtime) extern unsigned long __init calibrate_cpu(void); -#endif +#endif /* ASM_X86__TIME_H */ diff --git a/include/asm-x86/timer.h b/include/asm-x86/timer.h index fb2a4ddddf3..b87123dec3d 100644 --- a/include/asm-x86/timer.h +++ b/include/asm-x86/timer.h @@ -1,5 +1,5 @@ -#ifndef _ASMi386_TIMER_H -#define _ASMi386_TIMER_H +#ifndef ASM_X86__TIMER_H +#define ASM_X86__TIMER_H #include #include #include @@ -60,4 +60,4 @@ static inline unsigned long long cycles_2_ns(unsigned long long cyc) return ns; } -#endif +#endif /* ASM_X86__TIMER_H */ diff --git a/include/asm-x86/timex.h b/include/asm-x86/timex.h index 43e5a78500c..d1ce2416a5d 100644 --- a/include/asm-x86/timex.h +++ b/include/asm-x86/timex.h @@ -1,6 +1,6 @@ /* x86 architecture timex specifications */ -#ifndef _ASM_X86_TIMEX_H -#define _ASM_X86_TIMEX_H +#ifndef ASM_X86__TIMEX_H +#define ASM_X86__TIMEX_H #include #include @@ -16,4 +16,4 @@ #define ARCH_HAS_READ_CURRENT_TIMER -#endif +#endif /* ASM_X86__TIMEX_H */ diff --git a/include/asm-x86/tlb.h b/include/asm-x86/tlb.h index e4e9e2d07a9..db36e9e89e8 100644 --- a/include/asm-x86/tlb.h +++ b/include/asm-x86/tlb.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_TLB_H -#define _ASM_X86_TLB_H +#ifndef ASM_X86__TLB_H +#define ASM_X86__TLB_H #define tlb_start_vma(tlb, vma) do { } while (0) #define tlb_end_vma(tlb, vma) do { } while (0) @@ -8,4 +8,4 @@ #include -#endif +#endif /* ASM_X86__TLB_H */ diff --git a/include/asm-x86/tlbflush.h b/include/asm-x86/tlbflush.h index 35c76ceb9f4..ef68b76dc3c 100644 --- a/include/asm-x86/tlbflush.h +++ b/include/asm-x86/tlbflush.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_TLBFLUSH_H -#define _ASM_X86_TLBFLUSH_H +#ifndef ASM_X86__TLBFLUSH_H +#define ASM_X86__TLBFLUSH_H #include #include @@ -165,4 +165,4 @@ static inline void flush_tlb_kernel_range(unsigned long start, flush_tlb_all(); } -#endif /* _ASM_X86_TLBFLUSH_H */ +#endif /* ASM_X86__TLBFLUSH_H */ diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h index 90ac7718469..7eca9bc022b 100644 --- a/include/asm-x86/topology.h +++ b/include/asm-x86/topology.h @@ -22,8 +22,8 @@ * * Send feedback to */ -#ifndef _ASM_X86_TOPOLOGY_H -#define _ASM_X86_TOPOLOGY_H +#ifndef ASM_X86__TOPOLOGY_H +#define ASM_X86__TOPOLOGY_H #ifdef CONFIG_X86_32 # ifdef CONFIG_X86_HT @@ -255,4 +255,4 @@ static inline void set_mp_bus_to_node(int busnum, int node) } #endif -#endif /* _ASM_X86_TOPOLOGY_H */ +#endif /* ASM_X86__TOPOLOGY_H */ diff --git a/include/asm-x86/trampoline.h b/include/asm-x86/trampoline.h index b156b08d013..0406bbd898a 100644 --- a/include/asm-x86/trampoline.h +++ b/include/asm-x86/trampoline.h @@ -1,5 +1,5 @@ -#ifndef __TRAMPOLINE_HEADER -#define __TRAMPOLINE_HEADER +#ifndef ASM_X86__TRAMPOLINE_H +#define ASM_X86__TRAMPOLINE_H #ifndef __ASSEMBLY__ @@ -18,4 +18,4 @@ extern unsigned long setup_trampoline(void); #endif /* __ASSEMBLY__ */ -#endif /* __TRAMPOLINE_HEADER */ +#endif /* ASM_X86__TRAMPOLINE_H */ diff --git a/include/asm-x86/traps.h b/include/asm-x86/traps.h index a4b65a71bd6..2a891a70471 100644 --- a/include/asm-x86/traps.h +++ b/include/asm-x86/traps.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_TRAPS_H -#define _ASM_X86_TRAPS_H +#ifndef ASM_X86__TRAPS_H +#define ASM_X86__TRAPS_H /* Common in X86_32 and X86_64 */ asmlinkage void divide_error(void); @@ -63,4 +63,4 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *); asmlinkage void do_spurious_interrupt_bug(struct pt_regs *); #endif /* CONFIG_X86_32 */ -#endif /* _ASM_X86_TRAPS_H */ +#endif /* ASM_X86__TRAPS_H */ diff --git a/include/asm-x86/tsc.h b/include/asm-x86/tsc.h index cb6f6ee45b8..ad0f5c41e78 100644 --- a/include/asm-x86/tsc.h +++ b/include/asm-x86/tsc.h @@ -1,8 +1,8 @@ /* * x86 TSC related functions */ -#ifndef _ASM_X86_TSC_H -#define _ASM_X86_TSC_H +#ifndef ASM_X86__TSC_H +#define ASM_X86__TSC_H #include @@ -59,4 +59,4 @@ extern void check_tsc_sync_target(void); extern int notsc_setup(char *); -#endif +#endif /* ASM_X86__TSC_H */ diff --git a/include/asm-x86/types.h b/include/asm-x86/types.h index 1ac80cd9acf..e78b52e1744 100644 --- a/include/asm-x86/types.h +++ b/include/asm-x86/types.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_TYPES_H -#define _ASM_X86_TYPES_H +#ifndef ASM_X86__TYPES_H +#define ASM_X86__TYPES_H #include @@ -33,4 +33,4 @@ typedef u32 dma_addr_t; #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ -#endif +#endif /* ASM_X86__TYPES_H */ diff --git a/include/asm-x86/uaccess.h b/include/asm-x86/uaccess.h index f6fa4d841bb..1838f3959a5 100644 --- a/include/asm-x86/uaccess.h +++ b/include/asm-x86/uaccess.h @@ -1,5 +1,5 @@ -#ifndef _ASM_UACCES_H_ -#define _ASM_UACCES_H_ +#ifndef ASM_X86__UACCESS_H +#define ASM_X86__UACCESS_H /* * User space memory access functions */ @@ -450,4 +450,4 @@ extern struct movsl_mask { # include "uaccess_64.h" #endif -#endif +#endif /* ASM_X86__UACCESS_H */ diff --git a/include/asm-x86/uaccess_32.h b/include/asm-x86/uaccess_32.h index 6fdef39a0bc..6b5b57d9c6d 100644 --- a/include/asm-x86/uaccess_32.h +++ b/include/asm-x86/uaccess_32.h @@ -1,5 +1,5 @@ -#ifndef __i386_UACCESS_H -#define __i386_UACCESS_H +#ifndef ASM_X86__UACCESS_32_H +#define ASM_X86__UACCESS_32_H /* * User space memory access functions @@ -215,4 +215,4 @@ long strnlen_user(const char __user *str, long n); unsigned long __must_check clear_user(void __user *mem, unsigned long len); unsigned long __must_check __clear_user(void __user *mem, unsigned long len); -#endif /* __i386_UACCESS_H */ +#endif /* ASM_X86__UACCESS_32_H */ diff --git a/include/asm-x86/uaccess_64.h b/include/asm-x86/uaccess_64.h index 515d4dce96b..5cfd2951c9e 100644 --- a/include/asm-x86/uaccess_64.h +++ b/include/asm-x86/uaccess_64.h @@ -1,5 +1,5 @@ -#ifndef __X86_64_UACCESS_H -#define __X86_64_UACCESS_H +#ifndef ASM_X86__UACCESS_64_H +#define ASM_X86__UACCESS_64_H /* * User space memory access functions @@ -198,4 +198,4 @@ static inline int __copy_from_user_inatomic_nocache(void *dst, unsigned long copy_user_handle_tail(char *to, char *from, unsigned len, unsigned zerorest); -#endif /* __X86_64_UACCESS_H */ +#endif /* ASM_X86__UACCESS_64_H */ diff --git a/include/asm-x86/ucontext.h b/include/asm-x86/ucontext.h index 50a79f7fcde..9948dd32808 100644 --- a/include/asm-x86/ucontext.h +++ b/include/asm-x86/ucontext.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_UCONTEXT_H -#define _ASM_X86_UCONTEXT_H +#ifndef ASM_X86__UCONTEXT_H +#define ASM_X86__UCONTEXT_H struct ucontext { unsigned long uc_flags; @@ -9,4 +9,4 @@ struct ucontext { sigset_t uc_sigmask; /* mask last for extensibility */ }; -#endif /* _ASM_X86_UCONTEXT_H */ +#endif /* ASM_X86__UCONTEXT_H */ diff --git a/include/asm-x86/unaligned.h b/include/asm-x86/unaligned.h index a7bd416b476..59dcdec3716 100644 --- a/include/asm-x86/unaligned.h +++ b/include/asm-x86/unaligned.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_UNALIGNED_H -#define _ASM_X86_UNALIGNED_H +#ifndef ASM_X86__UNALIGNED_H +#define ASM_X86__UNALIGNED_H /* * The x86 can do unaligned accesses itself. @@ -11,4 +11,4 @@ #define get_unaligned __get_unaligned_le #define put_unaligned __put_unaligned_le -#endif /* _ASM_X86_UNALIGNED_H */ +#endif /* ASM_X86__UNALIGNED_H */ diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h index 8317d94771d..602ddd88c68 100644 --- a/include/asm-x86/unistd_32.h +++ b/include/asm-x86/unistd_32.h @@ -1,5 +1,5 @@ -#ifndef _ASM_I386_UNISTD_H_ -#define _ASM_I386_UNISTD_H_ +#ifndef ASM_X86__UNISTD_32_H +#define ASM_X86__UNISTD_32_H /* * This file contains the system call numbers. @@ -370,4 +370,4 @@ #endif #endif /* __KERNEL__ */ -#endif /* _ASM_I386_UNISTD_H_ */ +#endif /* ASM_X86__UNISTD_32_H */ diff --git a/include/asm-x86/unwind.h b/include/asm-x86/unwind.h index 8b064bd9c55..a2151567db4 100644 --- a/include/asm-x86/unwind.h +++ b/include/asm-x86/unwind.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_UNWIND_H -#define _ASM_X86_UNWIND_H +#ifndef ASM_X86__UNWIND_H +#define ASM_X86__UNWIND_H #define UNW_PC(frame) ((void)(frame), 0UL) #define UNW_SP(frame) ((void)(frame), 0UL) @@ -10,4 +10,4 @@ static inline int arch_unw_user_mode(const void *info) return 0; } -#endif /* _ASM_X86_UNWIND_H */ +#endif /* ASM_X86__UNWIND_H */ diff --git a/include/asm-x86/user32.h b/include/asm-x86/user32.h index a3d91004787..aa66c1857f0 100644 --- a/include/asm-x86/user32.h +++ b/include/asm-x86/user32.h @@ -1,5 +1,5 @@ -#ifndef USER32_H -#define USER32_H 1 +#ifndef ASM_X86__USER32_H +#define ASM_X86__USER32_H /* IA32 compatible user structures for ptrace. * These should be used for 32bit coredumps too. */ @@ -67,4 +67,4 @@ struct user32 { }; -#endif +#endif /* ASM_X86__USER32_H */ diff --git a/include/asm-x86/user_32.h b/include/asm-x86/user_32.h index d6e51edc259..e0fe2f55f1a 100644 --- a/include/asm-x86/user_32.h +++ b/include/asm-x86/user_32.h @@ -1,5 +1,5 @@ -#ifndef _I386_USER_H -#define _I386_USER_H +#ifndef ASM_X86__USER_32_H +#define ASM_X86__USER_32_H #include /* Core file format: The core file is written in such a way that gdb @@ -128,4 +128,4 @@ struct user{ #define HOST_TEXT_START_ADDR (u.start_code) #define HOST_STACK_END_ADDR (u.start_stack + u.u_ssize * NBPG) -#endif /* _I386_USER_H */ +#endif /* ASM_X86__USER_32_H */ diff --git a/include/asm-x86/user_64.h b/include/asm-x86/user_64.h index 6037b634c77..38b5799863b 100644 --- a/include/asm-x86/user_64.h +++ b/include/asm-x86/user_64.h @@ -1,5 +1,5 @@ -#ifndef _X86_64_USER_H -#define _X86_64_USER_H +#ifndef ASM_X86__USER_64_H +#define ASM_X86__USER_64_H #include #include @@ -134,4 +134,4 @@ struct user { #define HOST_TEXT_START_ADDR (u.start_code) #define HOST_STACK_END_ADDR (u.start_stack + u.u_ssize * NBPG) -#endif /* _X86_64_USER_H */ +#endif /* ASM_X86__USER_64_H */ diff --git a/include/asm-x86/uv/bios.h b/include/asm-x86/uv/bios.h index aa73362ff5d..7cd6d7ec130 100644 --- a/include/asm-x86/uv/bios.h +++ b/include/asm-x86/uv/bios.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_BIOS_H -#define _ASM_X86_BIOS_H +#ifndef ASM_X86__UV__BIOS_H +#define ASM_X86__UV__BIOS_H /* * BIOS layer definitions. @@ -65,4 +65,4 @@ x86_bios_freq_base(unsigned long which, unsigned long *ticks_per_second, unsigned long *drift_info); extern const char *x86_bios_strerror(long status); -#endif /* _ASM_X86_BIOS_H */ +#endif /* ASM_X86__UV__BIOS_H */ diff --git a/include/asm-x86/uv/uv_bau.h b/include/asm-x86/uv/uv_bau.h index 91ac0dfb758..0950239acaf 100644 --- a/include/asm-x86/uv/uv_bau.h +++ b/include/asm-x86/uv/uv_bau.h @@ -8,8 +8,8 @@ * Copyright (C) 2008 Silicon Graphics, Inc. All rights reserved. */ -#ifndef __ASM_X86_UV_BAU__ -#define __ASM_X86_UV_BAU__ +#ifndef ASM_X86__UV__UV_BAU_H +#define ASM_X86__UV__UV_BAU_H #include #define BITSPERBYTE 8 @@ -334,4 +334,4 @@ extern int uv_flush_tlb_others(cpumask_t *, struct mm_struct *, unsigned long); extern void uv_bau_message_intr1(void); extern void uv_bau_timeout_intr1(void); -#endif /* __ASM_X86_UV_BAU__ */ +#endif /* ASM_X86__UV__UV_BAU_H */ diff --git a/include/asm-x86/uv/uv_hub.h b/include/asm-x86/uv/uv_hub.h index a4ef26e5850..bdb5b01afbf 100644 --- a/include/asm-x86/uv/uv_hub.h +++ b/include/asm-x86/uv/uv_hub.h @@ -8,8 +8,8 @@ * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved. */ -#ifndef __ASM_X86_UV_HUB_H__ -#define __ASM_X86_UV_HUB_H__ +#ifndef ASM_X86__UV__UV_HUB_H +#define ASM_X86__UV__UV_HUB_H #include #include @@ -350,5 +350,5 @@ static inline int uv_num_possible_blades(void) return uv_possible_blades; } -#endif /* __ASM_X86_UV_HUB__ */ +#endif /* ASM_X86__UV__UV_HUB_H */ diff --git a/include/asm-x86/uv/uv_mmrs.h b/include/asm-x86/uv/uv_mmrs.h index 151fd7fcb80..8b03d89d245 100644 --- a/include/asm-x86/uv/uv_mmrs.h +++ b/include/asm-x86/uv/uv_mmrs.h @@ -8,8 +8,8 @@ * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved. */ -#ifndef __ASM_X86_UV_MMRS__ -#define __ASM_X86_UV_MMRS__ +#ifndef ASM_X86__UV__UV_MMRS_H +#define ASM_X86__UV__UV_MMRS_H #define UV_MMR_ENABLE (1UL << 63) @@ -1292,4 +1292,4 @@ union uvh_si_alias2_overlay_config_u { }; -#endif /* __ASM_X86_UV_MMRS__ */ +#endif /* ASM_X86__UV__UV_MMRS_H */ diff --git a/include/asm-x86/vdso.h b/include/asm-x86/vdso.h index 8e18fb80f5e..4ab320913ea 100644 --- a/include/asm-x86/vdso.h +++ b/include/asm-x86/vdso.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_VDSO_H -#define _ASM_X86_VDSO_H 1 +#ifndef ASM_X86__VDSO_H +#define ASM_X86__VDSO_H #ifdef CONFIG_X86_64 extern const char VDSO64_PRELINK[]; @@ -44,4 +44,4 @@ extern const char vdso32_int80_start, vdso32_int80_end; extern const char vdso32_syscall_start, vdso32_syscall_end; extern const char vdso32_sysenter_start, vdso32_sysenter_end; -#endif /* asm-x86/vdso.h */ +#endif /* ASM_X86__VDSO_H */ diff --git a/include/asm-x86/vga.h b/include/asm-x86/vga.h index 0ccf804377e..b9e493d07d0 100644 --- a/include/asm-x86/vga.h +++ b/include/asm-x86/vga.h @@ -4,8 +4,8 @@ * (c) 1998 Martin Mares */ -#ifndef _LINUX_ASM_VGA_H_ -#define _LINUX_ASM_VGA_H_ +#ifndef ASM_X86__VGA_H +#define ASM_X86__VGA_H /* * On the PC, we can just recalculate addresses and then @@ -17,4 +17,4 @@ #define vga_readb(x) (*(x)) #define vga_writeb(x, y) (*(y) = (x)) -#endif +#endif /* ASM_X86__VGA_H */ diff --git a/include/asm-x86/vgtod.h b/include/asm-x86/vgtod.h index 3301f092934..38fd1336402 100644 --- a/include/asm-x86/vgtod.h +++ b/include/asm-x86/vgtod.h @@ -1,5 +1,5 @@ -#ifndef _ASM_VGTOD_H -#define _ASM_VGTOD_H 1 +#ifndef ASM_X86__VGTOD_H +#define ASM_X86__VGTOD_H #include #include @@ -26,4 +26,4 @@ extern struct vsyscall_gtod_data __vsyscall_gtod_data __section_vsyscall_gtod_data; extern struct vsyscall_gtod_data vsyscall_gtod_data; -#endif +#endif /* ASM_X86__VGTOD_H */ diff --git a/include/asm-x86/visws/cobalt.h b/include/asm-x86/visws/cobalt.h index 995258831b7..9627a8fe84e 100644 --- a/include/asm-x86/visws/cobalt.h +++ b/include/asm-x86/visws/cobalt.h @@ -1,5 +1,5 @@ -#ifndef __I386_SGI_COBALT_H -#define __I386_SGI_COBALT_H +#ifndef ASM_X86__VISWS__COBALT_H +#define ASM_X86__VISWS__COBALT_H #include @@ -122,4 +122,4 @@ extern char visws_board_type; extern char visws_board_rev; -#endif /* __I386_SGI_COBALT_H */ +#endif /* ASM_X86__VISWS__COBALT_H */ diff --git a/include/asm-x86/visws/lithium.h b/include/asm-x86/visws/lithium.h index dfcd4f07ab8..b36d3b378c6 100644 --- a/include/asm-x86/visws/lithium.h +++ b/include/asm-x86/visws/lithium.h @@ -1,5 +1,5 @@ -#ifndef __I386_SGI_LITHIUM_H -#define __I386_SGI_LITHIUM_H +#ifndef ASM_X86__VISWS__LITHIUM_H +#define ASM_X86__VISWS__LITHIUM_H #include @@ -49,5 +49,5 @@ static inline unsigned short li_pcib_read16(unsigned long reg) return *((volatile unsigned short *)(LI_PCIB_VADDR+reg)); } -#endif +#endif /* ASM_X86__VISWS__LITHIUM_H */ diff --git a/include/asm-x86/visws/piix4.h b/include/asm-x86/visws/piix4.h index 83ea4f46e41..61c938045ec 100644 --- a/include/asm-x86/visws/piix4.h +++ b/include/asm-x86/visws/piix4.h @@ -1,5 +1,5 @@ -#ifndef __I386_SGI_PIIX_H -#define __I386_SGI_PIIX_H +#ifndef ASM_X86__VISWS__PIIX4_H +#define ASM_X86__VISWS__PIIX4_H /* * PIIX4 as used on SGI Visual Workstations @@ -104,4 +104,4 @@ */ #define PIIX_GPI_STPCLK 0x4 // STPCLK signal routed back in -#endif +#endif /* ASM_X86__VISWS__PIIX4_H */ diff --git a/include/asm-x86/vm86.h b/include/asm-x86/vm86.h index 5ce351325e0..998bd18eb73 100644 --- a/include/asm-x86/vm86.h +++ b/include/asm-x86/vm86.h @@ -1,5 +1,5 @@ -#ifndef _LINUX_VM86_H -#define _LINUX_VM86_H +#ifndef ASM_X86__VM86_H +#define ASM_X86__VM86_H /* * I'm guessing at the VIF/VIP flag usage, but hope that this is how @@ -205,4 +205,4 @@ static inline int handle_vm86_trap(struct kernel_vm86_regs *a, long b, int c) #endif /* __KERNEL__ */ -#endif +#endif /* ASM_X86__VM86_H */ diff --git a/include/asm-x86/vmi_time.h b/include/asm-x86/vmi_time.h index c3118c38515..b2d39e6a08b 100644 --- a/include/asm-x86/vmi_time.h +++ b/include/asm-x86/vmi_time.h @@ -22,8 +22,8 @@ * */ -#ifndef __VMI_TIME_H -#define __VMI_TIME_H +#ifndef ASM_X86__VMI_TIME_H +#define ASM_X86__VMI_TIME_H /* * Raw VMI call indices for timer functions @@ -95,4 +95,4 @@ extern void __devinit vmi_time_ap_init(void); #define CONFIG_VMI_ALARM_HZ 100 -#endif +#endif /* ASM_X86__VMI_TIME_H */ diff --git a/include/asm-x86/vsyscall.h b/include/asm-x86/vsyscall.h index 6b66ff905af..dcd4682413d 100644 --- a/include/asm-x86/vsyscall.h +++ b/include/asm-x86/vsyscall.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_64_VSYSCALL_H_ -#define _ASM_X86_64_VSYSCALL_H_ +#ifndef ASM_X86__VSYSCALL_H +#define ASM_X86__VSYSCALL_H enum vsyscall_num { __NR_vgettimeofday, @@ -41,4 +41,4 @@ extern void map_vsyscall(void); #endif /* __KERNEL__ */ -#endif /* _ASM_X86_64_VSYSCALL_H_ */ +#endif /* ASM_X86__VSYSCALL_H */ diff --git a/include/asm-x86/xen/events.h b/include/asm-x86/xen/events.h index 8ded7472002..8151f5b8b6c 100644 --- a/include/asm-x86/xen/events.h +++ b/include/asm-x86/xen/events.h @@ -1,5 +1,5 @@ -#ifndef __XEN_EVENTS_H -#define __XEN_EVENTS_H +#ifndef ASM_X86__XEN__EVENTS_H +#define ASM_X86__XEN__EVENTS_H enum ipi_vector { XEN_RESCHEDULE_VECTOR, @@ -21,4 +21,4 @@ static inline void xen_do_IRQ(int irq, struct pt_regs *regs) do_IRQ(regs); } -#endif /* __XEN_EVENTS_H */ +#endif /* ASM_X86__XEN__EVENTS_H */ diff --git a/include/asm-x86/xen/grant_table.h b/include/asm-x86/xen/grant_table.h index 2444d4593a3..c4baab4d2b6 100644 --- a/include/asm-x86/xen/grant_table.h +++ b/include/asm-x86/xen/grant_table.h @@ -1,7 +1,7 @@ -#ifndef __XEN_GRANT_TABLE_H -#define __XEN_GRANT_TABLE_H +#ifndef ASM_X86__XEN__GRANT_TABLE_H +#define ASM_X86__XEN__GRANT_TABLE_H #define xen_alloc_vm_area(size) alloc_vm_area(size) #define xen_free_vm_area(area) free_vm_area(area) -#endif /* __XEN_GRANT_TABLE_H */ +#endif /* ASM_X86__XEN__GRANT_TABLE_H */ diff --git a/include/asm-x86/xen/hypercall.h b/include/asm-x86/xen/hypercall.h index 91cb7fd5c12..44f4259bee3 100644 --- a/include/asm-x86/xen/hypercall.h +++ b/include/asm-x86/xen/hypercall.h @@ -30,8 +30,8 @@ * IN THE SOFTWARE. */ -#ifndef __HYPERCALL_H__ -#define __HYPERCALL_H__ +#ifndef ASM_X86__XEN__HYPERCALL_H +#define ASM_X86__XEN__HYPERCALL_H #include #include @@ -524,4 +524,4 @@ MULTI_stack_switch(struct multicall_entry *mcl, mcl->args[1] = esp; } -#endif /* __HYPERCALL_H__ */ +#endif /* ASM_X86__XEN__HYPERCALL_H */ diff --git a/include/asm-x86/xen/hypervisor.h b/include/asm-x86/xen/hypervisor.h index 8e15dd28c91..06c350452c5 100644 --- a/include/asm-x86/xen/hypervisor.h +++ b/include/asm-x86/xen/hypervisor.h @@ -30,8 +30,8 @@ * IN THE SOFTWARE. */ -#ifndef __HYPERVISOR_H__ -#define __HYPERVISOR_H__ +#ifndef ASM_X86__XEN__HYPERVISOR_H +#define ASM_X86__XEN__HYPERVISOR_H #include #include @@ -70,4 +70,4 @@ u64 jiffies_to_st(unsigned long jiffies); #define is_running_on_xen() (xen_start_info ? 1 : 0) -#endif /* __HYPERVISOR_H__ */ +#endif /* ASM_X86__XEN__HYPERVISOR_H */ diff --git a/include/asm-x86/xen/interface.h b/include/asm-x86/xen/interface.h index 9d810f2538a..d077bba96da 100644 --- a/include/asm-x86/xen/interface.h +++ b/include/asm-x86/xen/interface.h @@ -6,8 +6,8 @@ * Copyright (c) 2004, K A Fraser */ -#ifndef __ASM_X86_XEN_INTERFACE_H -#define __ASM_X86_XEN_INTERFACE_H +#ifndef ASM_X86__XEN__INTERFACE_H +#define ASM_X86__XEN__INTERFACE_H #ifdef __XEN__ #define __DEFINE_GUEST_HANDLE(name, type) \ @@ -172,4 +172,4 @@ DEFINE_GUEST_HANDLE_STRUCT(vcpu_guest_context); #define XEN_CPUID XEN_EMULATE_PREFIX "cpuid" #endif -#endif /* __ASM_X86_XEN_INTERFACE_H */ +#endif /* ASM_X86__XEN__INTERFACE_H */ diff --git a/include/asm-x86/xen/interface_32.h b/include/asm-x86/xen/interface_32.h index d8ac41d5db8..08167e19fc6 100644 --- a/include/asm-x86/xen/interface_32.h +++ b/include/asm-x86/xen/interface_32.h @@ -6,8 +6,8 @@ * Copyright (c) 2004, K A Fraser */ -#ifndef __ASM_X86_XEN_INTERFACE_32_H -#define __ASM_X86_XEN_INTERFACE_32_H +#ifndef ASM_X86__XEN__INTERFACE_32_H +#define ASM_X86__XEN__INTERFACE_32_H /* @@ -94,4 +94,4 @@ typedef struct xen_callback xen_callback_t; #define xen_pfn_to_cr3(pfn) (((unsigned)(pfn) << 12) | ((unsigned)(pfn) >> 20)) #define xen_cr3_to_pfn(cr3) (((unsigned)(cr3) >> 12) | ((unsigned)(cr3) << 20)) -#endif /* __ASM_X86_XEN_INTERFACE_32_H */ +#endif /* ASM_X86__XEN__INTERFACE_32_H */ diff --git a/include/asm-x86/xen/interface_64.h b/include/asm-x86/xen/interface_64.h index 842266ce96e..046c0f1e01d 100644 --- a/include/asm-x86/xen/interface_64.h +++ b/include/asm-x86/xen/interface_64.h @@ -1,5 +1,5 @@ -#ifndef __ASM_X86_XEN_INTERFACE_64_H -#define __ASM_X86_XEN_INTERFACE_64_H +#ifndef ASM_X86__XEN__INTERFACE_64_H +#define ASM_X86__XEN__INTERFACE_64_H /* * 64-bit segment selectors @@ -156,4 +156,4 @@ typedef unsigned long xen_callback_t; #endif /* !__ASSEMBLY__ */ -#endif /* __ASM_X86_XEN_INTERFACE_64_H */ +#endif /* ASM_X86__XEN__INTERFACE_64_H */ diff --git a/include/asm-x86/xen/page.h b/include/asm-x86/xen/page.h index 05e678a8662..a17a86433c7 100644 --- a/include/asm-x86/xen/page.h +++ b/include/asm-x86/xen/page.h @@ -1,5 +1,5 @@ -#ifndef __XEN_PAGE_H -#define __XEN_PAGE_H +#ifndef ASM_X86__XEN__PAGE_H +#define ASM_X86__XEN__PAGE_H #include @@ -162,4 +162,4 @@ xmaddr_t arbitrary_virt_to_machine(void *address); void make_lowmem_page_readonly(void *vaddr); void make_lowmem_page_readwrite(void *vaddr); -#endif /* __XEN_PAGE_H */ +#endif /* ASM_X86__XEN__PAGE_H */ -- cgit v1.2.3-70-g09d2 From a31863168660c6b6f6c7ffe05bb6a38e97803326 Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Tue, 22 Jul 2008 21:53:53 +0200 Subject: x86: consolidate header guards This patch consolidates the header guard names which are also used externally, i.e. in .c files. Signed-off-by: Vegard Nossum --- arch/x86/boot/compressed/misc.c | 2 +- arch/x86/kernel/asm-offsets_64.c | 2 +- arch/x86/kernel/syscall_64.c | 4 ++-- include/asm-x86/desc.h | 6 +++--- include/asm-x86/unistd_64.h | 6 +++--- 5 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c index bc5553b496f..3ee338c7d1d 100644 --- a/arch/x86/boot/compressed/misc.c +++ b/arch/x86/boot/compressed/misc.c @@ -16,7 +16,7 @@ */ #undef CONFIG_PARAVIRT #ifdef CONFIG_X86_32 -#define _ASM_DESC_H_ 1 +#define ASM_X86__DESC_H 1 #endif #ifdef CONFIG_X86_64 diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c index aa89387006f..505543a75a5 100644 --- a/arch/x86/kernel/asm-offsets_64.c +++ b/arch/x86/kernel/asm-offsets_64.c @@ -22,7 +22,7 @@ #define __NO_STUBS 1 #undef __SYSCALL -#undef _ASM_X86_64_UNISTD_H_ +#undef ASM_X86__UNISTD_64_H #define __SYSCALL(nr, sym) [nr] = 1, static char syscalls[] = { #include diff --git a/arch/x86/kernel/syscall_64.c b/arch/x86/kernel/syscall_64.c index 170d43c1748..3d1be4f0fac 100644 --- a/arch/x86/kernel/syscall_64.c +++ b/arch/x86/kernel/syscall_64.c @@ -8,12 +8,12 @@ #define __NO_STUBS #define __SYSCALL(nr, sym) extern asmlinkage void sym(void) ; -#undef _ASM_X86_64_UNISTD_H_ +#undef ASM_X86__UNISTD_64_H #include #undef __SYSCALL #define __SYSCALL(nr, sym) [nr] = sym, -#undef _ASM_X86_64_UNISTD_H_ +#undef ASM_X86__UNISTD_64_H typedef void (*sys_call_ptr_t)(void); diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h index a44c4dc7059..b73fea54def 100644 --- a/include/asm-x86/desc.h +++ b/include/asm-x86/desc.h @@ -1,5 +1,5 @@ -#ifndef _ASM_DESC_H_ -#define _ASM_DESC_H_ +#ifndef ASM_X86__DESC_H +#define ASM_X86__DESC_H #ifndef __ASSEMBLY__ #include @@ -397,4 +397,4 @@ static inline void set_system_gate_ist(int n, void *addr, unsigned ist) #endif /* __ASSEMBLY__ */ -#endif +#endif /* ASM_X86__DESC_H */ diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h index 9c1a4a3470d..1e10189ed6b 100644 --- a/include/asm-x86/unistd_64.h +++ b/include/asm-x86/unistd_64.h @@ -1,5 +1,5 @@ -#ifndef _ASM_X86_64_UNISTD_H_ -#define _ASM_X86_64_UNISTD_H_ +#ifndef ASM_X86__UNISTD_64_H +#define ASM_X86__UNISTD_64_H #ifndef __SYSCALL #define __SYSCALL(a, b) @@ -676,4 +676,4 @@ __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime) #define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall") #endif /* __KERNEL__ */ -#endif /* _ASM_X86_64_UNISTD_H_ */ +#endif /* ASM_X86__UNISTD_64_H */ -- cgit v1.2.3-70-g09d2 From 2b97df06ce44b1d145bd1299f50765803c2fabee Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Wed, 23 Jul 2008 17:13:14 +0530 Subject: x86: apic_XX.c declare functions before they get used declared following smp interrupts in asm-x86/hw_irq.h: smp_apic_timer_interrupt, smp_spurious_interrupt, smp_error_interrupt Signed-off-by: Jaswinder Singh --- include/asm-x86/apic.h | 5 +++++ include/asm-x86/hw_irq.h | 10 ++++++++++ 2 files changed, 15 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h index 133c998161c..519ad65708e 100644 --- a/include/asm-x86/apic.h +++ b/include/asm-x86/apic.h @@ -54,6 +54,11 @@ extern int disable_apic; #endif extern int is_vsmp_box(void); +extern void xapic_wait_icr_idle(void); +extern u32 safe_xapic_wait_icr_idle(void); +extern u64 xapic_icr_read(void); +extern void xapic_icr_write(u32, u32); +extern int setup_profiling_timer(unsigned int); static inline void native_apic_write(unsigned long reg, u32 v) { diff --git a/include/asm-x86/hw_irq.h b/include/asm-x86/hw_irq.h index 77ba51df566..40b941ac81c 100644 --- a/include/asm-x86/hw_irq.h +++ b/include/asm-x86/hw_irq.h @@ -93,6 +93,16 @@ extern asmlinkage void qic_reschedule_interrupt(void); extern asmlinkage void qic_enable_irq_interrupt(void); extern asmlinkage void qic_call_function_interrupt(void); +/* SMP */ +extern void smp_apic_timer_interrupt(struct pt_regs *); +#ifdef CONFIG_X86_32 +extern void smp_spurious_interrupt(struct pt_regs *); +extern void smp_error_interrupt(struct pt_regs *); +#else +extern asmlinkage void smp_spurious_interrupt(void); +extern asmlinkage void smp_error_interrupt(void); +#endif + #ifdef CONFIG_X86_32 extern void (*const interrupt[NR_IRQS])(void); #else -- cgit v1.2.3-70-g09d2 From 8f7db5186cf126b56035d9a9735774d751090d66 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Wed, 23 Jul 2008 17:31:02 +0530 Subject: x86: vm86_32.c declare functions before they get used declared following syscalls in asm-x86/syscalls.h: sys_vm86old, sys_vm86 Signed-off-by: Jaswinder Singh --- include/asm-x86/syscalls.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/syscalls.h b/include/asm-x86/syscalls.h index 170fcb132f5..3d6b159d348 100644 --- a/include/asm-x86/syscalls.h +++ b/include/asm-x86/syscalls.h @@ -6,7 +6,6 @@ * This file is released under the GPLv2. * See the file COPYING for more details. * - * Please do not call me directly, include linux/syscalls.h */ #ifndef _ASM_X86_SYSCALLS_H @@ -60,6 +59,10 @@ asmlinkage int sys_olduname(struct oldold_utsname __user *); asmlinkage int sys_set_thread_area(struct user_desc __user *); asmlinkage int sys_get_thread_area(struct user_desc __user *); +/* kernel/vm86_32.c */ +asmlinkage int sys_vm86old(struct pt_regs); +asmlinkage int sys_vm86(struct pt_regs); + #else /* CONFIG_X86_32 */ /* X86_64 only */ -- cgit v1.2.3-70-g09d2 From a80495ec927e8ec2b1ff085592bbe9bed77ffb3b Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Wed, 23 Jul 2008 17:33:57 +0530 Subject: x86: mm/init_XX.c declare functions before they get used included in mm/init_32.c for zap_low_mappings() declared free_initmem() in asm-x86/page_XX.h Signed-off-by: Jaswinder Singh --- arch/x86/mm/init_32.c | 1 + include/asm-x86/page_32.h | 1 + include/asm-x86/page_64.h | 1 + 3 files changed, 3 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index d37f29376b0..4974e97dedf 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -47,6 +47,7 @@ #include #include #include +#include unsigned int __VMALLOC_RESERVE = 128 << 20; diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index ab8528793f0..c9dd069d4ef 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -96,6 +96,7 @@ extern void find_low_pfn_range(void); extern unsigned long init_memory_mapping(unsigned long start, unsigned long end); extern void initmem_init(unsigned long, unsigned long); +extern void free_initmem(void); extern void setup_bootmem_allocator(void); diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h index c6916c83e6b..e5a754e97f2 100644 --- a/include/asm-x86/page_64.h +++ b/include/asm-x86/page_64.h @@ -91,6 +91,7 @@ extern unsigned long init_memory_mapping(unsigned long start, unsigned long end); extern void initmem_init(unsigned long start_pfn, unsigned long end_pfn); +extern void free_initmem(void); extern void init_extra_mapping_uc(unsigned long phys, unsigned long size); extern void init_extra_mapping_wb(unsigned long phys, unsigned long size); -- cgit v1.2.3-70-g09d2 From 70ef56414ec7e01d787c8e959bb259845df4ee4f Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Wed, 23 Jul 2008 17:36:37 +0530 Subject: x86: mm/fault.c declare do_page_fault before they get used declared do_page_fault() in asm-x86/trap.h for both X86_32 and X86_64 removed do_invalid_op declaration from mm/fault.c as it is already declared in asm-x86/trap.h Signed-off-by: Jaswinder Singh --- arch/x86/mm/fault.c | 3 +-- include/asm-x86/traps.h | 4 ++++ 2 files changed, 5 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 455f3fe67b4..8f92cac4e6d 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -35,6 +35,7 @@ #include #include #include +#include /* * Page fault error code bits @@ -357,8 +358,6 @@ static int is_errata100(struct pt_regs *regs, unsigned long address) return 0; } -void do_invalid_op(struct pt_regs *, unsigned long); - static int is_f00f_bug(struct pt_regs *regs, unsigned long address) { #ifdef CONFIG_X86_F00F_BUG diff --git a/include/asm-x86/traps.h b/include/asm-x86/traps.h index a4b65a71bd6..b60fab546da 100644 --- a/include/asm-x86/traps.h +++ b/include/asm-x86/traps.h @@ -51,6 +51,8 @@ void do_spurious_interrupt_bug(struct pt_regs *, long); unsigned long patch_espfix_desc(unsigned long, unsigned long); asmlinkage void math_emulate(long); +void do_page_fault(struct pt_regs *regs, unsigned long error_code); + #else /* CONFIG_X86_32 */ asmlinkage void double_fault(void); @@ -62,5 +64,7 @@ asmlinkage void do_coprocessor_error(struct pt_regs *); asmlinkage void do_simd_coprocessor_error(struct pt_regs *); asmlinkage void do_spurious_interrupt_bug(struct pt_regs *); +asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code); + #endif /* CONFIG_X86_32 */ #endif /* _ASM_X86_TRAPS_H */ -- cgit v1.2.3-70-g09d2 From e0b7c8192ded4c2096388008d3ca6708caa8b601 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Wed, 23 Jul 2008 17:40:34 +0530 Subject: x86: mm/pageattr.c declare arch_report_meminfo before they get used declared arch_report_meminfo() in asm-x86/pgtable.h as it will be also accessible by fs/proc/proc_misc.c Signed-off-by: Jaswinder Singh --- include/asm-x86/pgtable.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index 96aa76e691d..4fb22b9c528 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -310,6 +310,8 @@ static inline void native_pagetable_setup_start(pgd_t *base) {} static inline void native_pagetable_setup_done(pgd_t *base) {} #endif +extern int arch_report_meminfo(char *page); + #ifdef CONFIG_PARAVIRT #include #else /* !CONFIG_PARAVIRT */ -- cgit v1.2.3-70-g09d2 From 01eb7858c017b1c63b962f8c2ad37133383ca560 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Wed, 23 Jul 2008 17:41:59 +0530 Subject: x86: mm/pgtable_32.c declare set_pmd_pfn before they get used Signed-off-by: Jaswinder Singh --- include/asm-x86/pgtable_32.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h index 0611abf96a5..015fb4e19de 100644 --- a/include/asm-x86/pgtable_32.h +++ b/include/asm-x86/pgtable_32.h @@ -31,6 +31,7 @@ static inline void pgtable_cache_init(void) { } static inline void check_pgt_cache(void) { } void paging_init(void); +extern void set_pmd_pfn(unsigned long, unsigned long, pgprot_t); /* * The Linux x86 paging architecture is 'compile-time dual-mode', it -- cgit v1.2.3-70-g09d2 From 38ffbe66d59051fd9cfcfc8545f164700e2fa3bc Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 23 Jul 2008 14:21:18 -0700 Subject: x86/paravirt/xen: properly fill out the ldt ops LTP testing showed that Xen does not properly implement sys_modify_ldt(). This patch does the final little bits needed to make the ldt work properly. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/kernel/ldt.c | 9 ++++++++- arch/x86/kernel/paravirt.c | 4 ++++ arch/x86/xen/enlighten.c | 23 +++++++++++++++++++++++ include/asm-x86/desc.h | 10 +++++++++- include/asm-x86/paravirt.h | 13 +++++++++++++ 5 files changed, 57 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c index 3fee2aa50f3..4895e0634d2 100644 --- a/arch/x86/kernel/ldt.c +++ b/arch/x86/kernel/ldt.c @@ -51,6 +51,8 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload) memset(newldt + oldsize * LDT_ENTRY_SIZE, 0, (mincount - oldsize) * LDT_ENTRY_SIZE); + paravirt_alloc_ldt(newldt, mincount); + #ifdef CONFIG_X86_64 /* CHECKME: Do we really need this ? */ wmb(); @@ -75,6 +77,7 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload) #endif } if (oldsize) { + paravirt_free_ldt(oldldt, oldsize); if (oldsize * LDT_ENTRY_SIZE > PAGE_SIZE) vfree(oldldt); else @@ -86,10 +89,13 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload) static inline int copy_ldt(mm_context_t *new, mm_context_t *old) { int err = alloc_ldt(new, old->size, 0); + int i; if (err < 0) return err; - memcpy(new->ldt, old->ldt, old->size * LDT_ENTRY_SIZE); + + for(i = 0; i < old->size; i++) + write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE); return 0; } @@ -126,6 +132,7 @@ void destroy_context(struct mm_struct *mm) if (mm == current->active_mm) clear_LDT(); #endif + paravirt_free_ldt(mm->context.ldt, mm->context.size); if (mm->context.size * LDT_ENTRY_SIZE > PAGE_SIZE) vfree(mm->context.ldt); else diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 94da4d52d79..d8f2277be5a 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -348,6 +348,10 @@ struct pv_cpu_ops pv_cpu_ops = { .write_ldt_entry = native_write_ldt_entry, .write_gdt_entry = native_write_gdt_entry, .write_idt_entry = native_write_idt_entry, + + .alloc_ldt = paravirt_nop, + .free_ldt = paravirt_nop, + .load_sp0 = native_load_sp0, #if defined(CONFIG_X86_32) || defined(CONFIG_IA32_EMULATION) diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 9ff6e3cbf08..06219e60e9c 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -325,6 +325,26 @@ static unsigned long xen_store_tr(void) return 0; } +static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries) +{ + unsigned pages = roundup(entries * LDT_ENTRY_SIZE, PAGE_SIZE); + void *v = ldt; + int i; + + for(i = 0; i < pages; i += PAGE_SIZE) + make_lowmem_page_readonly(v + i); +} + +static void xen_free_ldt(struct desc_struct *ldt, unsigned entries) +{ + unsigned pages = roundup(entries * LDT_ENTRY_SIZE, PAGE_SIZE); + void *v = ldt; + int i; + + for(i = 0; i < pages; i += PAGE_SIZE) + make_lowmem_page_readwrite(v + i); +} + static void xen_set_ldt(const void *addr, unsigned entries) { struct mmuext_op *op; @@ -1220,6 +1240,9 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = { .load_gs_index = xen_load_gs_index, #endif + .alloc_ldt = xen_alloc_ldt, + .free_ldt = xen_free_ldt, + .store_gdt = native_store_gdt, .store_idt = native_store_idt, .store_tr = xen_store_tr, diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h index a44c4dc7059..24a524f5e1a 100644 --- a/include/asm-x86/desc.h +++ b/include/asm-x86/desc.h @@ -97,7 +97,15 @@ static inline int desc_empty(const void *ptr) native_write_gdt_entry(dt, entry, desc, type) #define write_idt_entry(dt, entry, g) \ native_write_idt_entry(dt, entry, g) -#endif + +static inline void paravirt_alloc_ldt(struct desc_struct *ldt, unsigned entries) +{ +} + +static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries) +{ +} +#endif /* CONFIG_PARAVIRT */ static inline void native_write_idt_entry(gate_desc *idt, int entry, const gate_desc *gate) diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index fbbde93f12d..db9b0647b34 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -124,6 +124,9 @@ struct pv_cpu_ops { int entrynum, const void *desc, int size); void (*write_idt_entry)(gate_desc *, int entrynum, const gate_desc *gate); + void (*alloc_ldt)(struct desc_struct *ldt, unsigned entries); + void (*free_ldt)(struct desc_struct *ldt, unsigned entries); + void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t); void (*set_iopl_mask)(unsigned mask); @@ -824,6 +827,16 @@ do { \ (aux) = __aux; \ } while (0) +static inline void paravirt_alloc_ldt(struct desc_struct *ldt, unsigned entries) +{ + PVOP_VCALL2(pv_cpu_ops.alloc_ldt, ldt, entries); +} + +static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries) +{ + PVOP_VCALL2(pv_cpu_ops.free_ldt, ldt, entries); +} + static inline void load_TR_desc(void) { PVOP_VCALL0(pv_cpu_ops.load_tr_desc); -- cgit v1.2.3-70-g09d2 From 32f71aff77b6470d272f80ac28f43f9601c4d140 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Mon, 21 Jul 2008 00:52:49 +0100 Subject: x86: PIC, L-APIC and I/O APIC debug information Dump all the PIC, local APIC and I/O APIC information at the fs_initcall() level, which is after ACPI (if used) has initialised PCI information, making the point of invocation consistent across MP-table and ACPI platforms. Remove explicit calls to print_IO_APIC() from elsewhere. Make the interface of all the functions involved consistent between 32-bit and 64-bit versions and make them all static by default by the means of a New-and-Improved(TM) __apicdebuginit() macro. Note that like print_IO_APIC() all these only output anything if "apic=debug" has been passed to the kernel through the command line. Signed-off-by: Maciej W. Rozycki Cc: Chuck Ebbert Signed-off-by: Ingo Molnar --- arch/x86/kernel/io_apic_32.c | 29 +++++++++++++++++++---------- arch/x86/kernel/io_apic_64.c | 31 +++++++++++++++++++------------ arch/x86/pci/acpi.c | 5 ----- include/asm-x86/hw_irq.h | 1 - 4 files changed, 38 insertions(+), 28 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c index de9aa0e3a9c..d54455ec985 100644 --- a/arch/x86/kernel/io_apic_32.c +++ b/arch/x86/kernel/io_apic_32.c @@ -50,6 +50,8 @@ #include #include +#define __apicdebuginit(type) static type __init + int (*ioapic_renumber_irq)(int ioapic, int irq); atomic_t irq_mis_count; @@ -1345,7 +1347,8 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, ioapic_write_entry(apic, pin, entry); } -void __init print_IO_APIC(void) + +__apicdebuginit(void) print_IO_APIC(void) { int apic, i; union IO_APIC_reg_00 reg_00; @@ -1460,9 +1463,7 @@ void __init print_IO_APIC(void) return; } -#if 0 - -static void print_APIC_bitfield(int base) +__apicdebuginit(void) print_APIC_bitfield(int base) { unsigned int v; int i, j; @@ -1483,7 +1484,7 @@ static void print_APIC_bitfield(int base) } } -void /*__init*/ print_local_APIC(void *dummy) +__apicdebuginit(void) print_local_APIC(void *dummy) { unsigned int v, ver, maxlvt; @@ -1567,12 +1568,12 @@ void /*__init*/ print_local_APIC(void *dummy) printk("\n"); } -void print_all_local_APICs(void) +__apicdebuginit(void) print_all_local_APICs(void) { on_each_cpu(print_local_APIC, NULL, 1); } -void /*__init*/ print_PIC(void) +__apicdebuginit(void) print_PIC(void) { unsigned int v; unsigned long flags; @@ -1604,7 +1605,17 @@ void /*__init*/ print_PIC(void) printk(KERN_DEBUG "... PIC ELCR: %04x\n", v); } -#endif /* 0 */ +__apicdebuginit(int) print_all_ICs(void) +{ + print_PIC(); + print_all_local_APICs(); + print_IO_APIC(); + + return 0; +} + +fs_initcall(print_all_ICs); + static void __init enable_IO_APIC(void) { @@ -2333,8 +2344,6 @@ void __init setup_IO_APIC(void) setup_IO_APIC_irqs(); init_IO_APIC_traps(); check_timer(); - if (!acpi_ioapic) - print_IO_APIC(); } /* diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index 8269434d170..8cdcc4f287c 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic_64.c @@ -53,6 +53,8 @@ #include #include +#define __apicdebuginit(type) static type __init + struct irq_cfg { cpumask_t domain; cpumask_t old_domain; @@ -87,8 +89,6 @@ int first_system_vector = 0xfe; char system_vectors[NR_VECTORS] = { [0 ... NR_VECTORS-1] = SYS_VECTOR_FREE}; -#define __apicdebuginit __init - int sis_apic_bug; /* not actually supported, dummy for compile */ static int no_timer_check; @@ -965,7 +965,8 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, ioapic_write_entry(apic, pin, entry); } -void __apicdebuginit print_IO_APIC(void) + +__apicdebuginit(void) print_IO_APIC(void) { int apic, i; union IO_APIC_reg_00 reg_00; @@ -1059,9 +1060,7 @@ void __apicdebuginit print_IO_APIC(void) return; } -#if 0 - -static __apicdebuginit void print_APIC_bitfield (int base) +__apicdebuginit(void) print_APIC_bitfield(int base) { unsigned int v; int i, j; @@ -1082,7 +1081,7 @@ static __apicdebuginit void print_APIC_bitfield (int base) } } -void __apicdebuginit print_local_APIC(void * dummy) +__apicdebuginit(void) print_local_APIC(void *dummy) { unsigned int v, ver, maxlvt; @@ -1159,12 +1158,12 @@ void __apicdebuginit print_local_APIC(void * dummy) printk("\n"); } -void print_all_local_APICs (void) +__apicdebuginit(void) print_all_local_APICs(void) { on_each_cpu(print_local_APIC, NULL, 1); } -void __apicdebuginit print_PIC(void) +__apicdebuginit(void) print_PIC(void) { unsigned int v; unsigned long flags; @@ -1196,7 +1195,17 @@ void __apicdebuginit print_PIC(void) printk(KERN_DEBUG "... PIC ELCR: %04x\n", v); } -#endif /* 0 */ +__apicdebuginit(int) print_all_ICs(void) +{ + print_PIC(); + print_all_local_APICs(); + print_IO_APIC(); + + return 0; +} + +fs_initcall(print_all_ICs); + void __init enable_IO_APIC(void) { @@ -1849,8 +1858,6 @@ void __init setup_IO_APIC(void) setup_IO_APIC_irqs(); init_IO_APIC_traps(); check_timer(); - if (!acpi_ioapic) - print_IO_APIC(); } struct sysfs_ioapic_data { diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c index 19af06927fb..1d88d2b3977 100644 --- a/arch/x86/pci/acpi.c +++ b/arch/x86/pci/acpi.c @@ -250,10 +250,5 @@ int __init pci_acpi_init(void) acpi_pci_irq_enable(dev); } -#ifdef CONFIG_X86_IO_APIC - if (acpi_ioapic) - print_IO_APIC(); -#endif - return 0; } diff --git a/include/asm-x86/hw_irq.h b/include/asm-x86/hw_irq.h index 77ba51df566..83e0048dca7 100644 --- a/include/asm-x86/hw_irq.h +++ b/include/asm-x86/hw_irq.h @@ -64,7 +64,6 @@ extern unsigned long io_apic_irqs; extern void init_VISWS_APIC_irqs(void); extern void setup_IO_APIC(void); extern void disable_IO_APIC(void); -extern void print_IO_APIC(void); extern int IO_APIC_get_PCI_irq_vector(int bus, int slot, int fn); extern void setup_ioapic_dest(void); -- cgit v1.2.3-70-g09d2 From 0791e13fbb1ea4e1808d055922c3f116b924bdc9 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Mon, 21 Jul 2008 01:28:43 +0100 Subject: x86: fix up a comment in ack_APIC_irq() Adjust a comment in ack_APIC_irq() according to the recent removal of CONFIG_X86_GOOD_APIC. Signed-off-by: Maciej W. Rozycki Signed-off-by: Ingo Molnar --- include/asm-x86/apic.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h index 133c998161c..e9e09b2ee7c 100644 --- a/include/asm-x86/apic.h +++ b/include/asm-x86/apic.h @@ -76,9 +76,7 @@ extern int get_physical_broadcast(void); static inline void ack_APIC_irq(void) { /* - * ack_APIC_irq() actually gets compiled as a single instruction: - * - a single rmw on Pentium/82489DX - * - a single write on P6+ cores (CONFIG_X86_GOOD_APIC) + * ack_APIC_irq() actually gets compiled as a single instruction * ... yummie. */ -- cgit v1.2.3-70-g09d2 From 461d159694d683c9e43ead07720a3a0f17f6f060 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Fri, 25 Jul 2008 09:19:05 +0530 Subject: x86_64: Declare new_utsname in asm-x86/syscalls.h Signed-off-by: Jaswinder Singh --- include/asm-x86/syscalls.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/syscalls.h b/include/asm-x86/syscalls.h index 3d6b159d348..87803da4401 100644 --- a/include/asm-x86/syscalls.h +++ b/include/asm-x86/syscalls.h @@ -5,7 +5,6 @@ * * This file is released under the GPLv2. * See the file COPYING for more details. - * */ #ifndef _ASM_X86_SYSCALLS_H @@ -87,6 +86,7 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *); /* kernel/sys_x86_64.c */ asmlinkage long sys_mmap(unsigned long, unsigned long, unsigned long, unsigned long, unsigned long, unsigned long); +struct new_utsname; asmlinkage long sys_uname(struct new_utsname __user *); #endif /* CONFIG_X86_32 */ -- cgit v1.2.3-70-g09d2 From e7f08dfdaa82def3d685d16fdb99203cbb67ec95 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Fri, 25 Jul 2008 10:42:26 +0530 Subject: X86_SMP: smp.c declare functions before they get used declared following smp interrupts in asm-x86/hw_irq.h: smp_reschedule_interrupt, smp_call_function_interrupt, smp_call_function_single_interrupt Signed-off-by: Jaswinder Singh --- include/asm-x86/hw_irq.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/hw_irq.h b/include/asm-x86/hw_irq.h index 40b941ac81c..08900eb90a9 100644 --- a/include/asm-x86/hw_irq.h +++ b/include/asm-x86/hw_irq.h @@ -102,6 +102,11 @@ extern void smp_error_interrupt(struct pt_regs *); extern asmlinkage void smp_spurious_interrupt(void); extern asmlinkage void smp_error_interrupt(void); #endif +#ifdef CONFIG_X86_SMP +extern void smp_reschedule_interrupt(struct pt_regs *); +extern void smp_call_function_interrupt(struct pt_regs *); +extern void smp_call_function_single_interrupt(struct pt_regs *); +#endif #ifdef CONFIG_X86_32 extern void (*const interrupt[NR_IRQS])(void); -- cgit v1.2.3-70-g09d2 From b2139aa0eec330c711c5a279db361e5ef1178e78 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Date: Fri, 25 Jul 2008 11:32:38 +0530 Subject: X86_SMP: tlb_XX.c declare smp_invalidate_interrupt before they get used declare smp_invalidate_interrupt in asm-x86/hw_irq.h for X86_32 and X86_64 Signed-off-by: Jaswinder Singh --- include/asm-x86/hw_irq.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/hw_irq.h b/include/asm-x86/hw_irq.h index 08900eb90a9..67685f1d318 100644 --- a/include/asm-x86/hw_irq.h +++ b/include/asm-x86/hw_irq.h @@ -106,6 +106,11 @@ extern asmlinkage void smp_error_interrupt(void); extern void smp_reschedule_interrupt(struct pt_regs *); extern void smp_call_function_interrupt(struct pt_regs *); extern void smp_call_function_single_interrupt(struct pt_regs *); +#ifdef CONFIG_X86_32 +extern void smp_invalidate_interrupt(struct pt_regs *); +#else +extern asmlinkage void smp_invalidate_interrupt(struct pt_regs *); +#endif #endif #ifdef CONFIG_X86_32 -- cgit v1.2.3-70-g09d2 From a4dbc34d181e87a0d724dee365921e9251f831d4 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 25 Jul 2008 02:14:28 -0700 Subject: x86: add setup_ioapic_ids for numaq in x86_quirks Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/io_apic_32.c | 5 ++--- arch/x86/kernel/numaq_32.c | 7 +++++++ include/asm-x86/setup.h | 1 + 3 files changed, 10 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c index 98e4db5373f..72ba06314c7 100644 --- a/arch/x86/kernel/io_apic_32.c +++ b/arch/x86/kernel/io_apic_32.c @@ -46,6 +46,7 @@ #include #include #include +#include #include #include @@ -1728,10 +1729,8 @@ static void __init setup_ioapic_ids_from_mpc(void) unsigned char old_id; unsigned long flags; -#ifdef CONFIG_X86_NUMAQ - if (found_numaq) + if (x86_quirks->setup_ioapic_ids && x86_quirks->setup_ioapic_ids()) return; -#endif /* * Don't check I/O APIC IDs for xAPIC systems. They have diff --git a/arch/x86/kernel/numaq_32.c b/arch/x86/kernel/numaq_32.c index b8c45610b20..2434467ddf7 100644 --- a/arch/x86/kernel/numaq_32.c +++ b/arch/x86/kernel/numaq_32.c @@ -229,6 +229,12 @@ static void __init smp_read_mpc_oem(struct mp_config_oemtable *oemtable, } } +static int __init numaq_setup_ioapic_ids(void) +{ + /* so can skip it */ + return 1; +} + static struct x86_quirks numaq_x86_quirks __initdata = { .arch_pre_time_init = numaq_pre_time_init, .arch_time_init = NULL, @@ -243,6 +249,7 @@ static struct x86_quirks numaq_x86_quirks __initdata = { .mpc_oem_bus_info = mpc_oem_bus_info, .mpc_oem_pci_bus = mpc_oem_pci_bus, .smp_read_mpc_oem = smp_read_mpc_oem, + .setup_ioapic_ids = numaq_setup_ioapic_ids, }; void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem, diff --git a/include/asm-x86/setup.h b/include/asm-x86/setup.h index b7aeba94e43..80fcc0de80b 100644 --- a/include/asm-x86/setup.h +++ b/include/asm-x86/setup.h @@ -38,6 +38,7 @@ struct x86_quirks { void (*mpc_oem_pci_bus)(struct mpc_config_bus *m); void (*smp_read_mpc_oem)(struct mp_config_oemtable *oemtable, unsigned short oemsize); + int (*setup_ioapic_ids)(void); }; extern struct x86_quirks *x86_quirks; -- cgit v1.2.3-70-g09d2 From 1176fa919292887aa738d317d61fe2bba4d764f2 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 25 Jul 2008 02:17:21 -0700 Subject: x86: mach-bigsmp to bigsmp Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/mach-generic/bigsmp.c | 6 +- include/asm-x86/bigsmp/apic.h | 144 +++++++++++++++++++++++++++++ include/asm-x86/bigsmp/apicdef.h | 13 +++ include/asm-x86/bigsmp/ipi.h | 25 +++++ include/asm-x86/mach-bigsmp/mach_apic.h | 144 ----------------------------- include/asm-x86/mach-bigsmp/mach_apicdef.h | 13 --- include/asm-x86/mach-bigsmp/mach_ipi.h | 25 ----- 7 files changed, 185 insertions(+), 185 deletions(-) create mode 100644 include/asm-x86/bigsmp/apic.h create mode 100644 include/asm-x86/bigsmp/apicdef.h create mode 100644 include/asm-x86/bigsmp/ipi.h delete mode 100644 include/asm-x86/mach-bigsmp/mach_apic.h delete mode 100644 include/asm-x86/mach-bigsmp/mach_apicdef.h delete mode 100644 include/asm-x86/mach-bigsmp/mach_ipi.h (limited to 'include/asm-x86') diff --git a/arch/x86/mach-generic/bigsmp.c b/arch/x86/mach-generic/bigsmp.c index b31f2800638..df37fc9d6a2 100644 --- a/arch/x86/mach-generic/bigsmp.c +++ b/arch/x86/mach-generic/bigsmp.c @@ -12,10 +12,10 @@ #include #include #include -#include +#include #include -#include -#include +#include +#include #include static int dmi_bigsmp; /* can be set by dmi scanners */ diff --git a/include/asm-x86/bigsmp/apic.h b/include/asm-x86/bigsmp/apic.h new file mode 100644 index 00000000000..0a9cd7c5ca0 --- /dev/null +++ b/include/asm-x86/bigsmp/apic.h @@ -0,0 +1,144 @@ +#ifndef __ASM_MACH_APIC_H +#define __ASM_MACH_APIC_H + +#define xapic_phys_to_log_apicid(cpu) (per_cpu(x86_bios_cpu_apicid, cpu)) +#define esr_disable (1) + +static inline int apic_id_registered(void) +{ + return (1); +} + +/* Round robin the irqs amoung the online cpus */ +static inline cpumask_t target_cpus(void) +{ + static unsigned long cpu = NR_CPUS; + do { + if (cpu >= NR_CPUS) + cpu = first_cpu(cpu_online_map); + else + cpu = next_cpu(cpu, cpu_online_map); + } while (cpu >= NR_CPUS); + return cpumask_of_cpu(cpu); +} + +#undef APIC_DEST_LOGICAL +#define APIC_DEST_LOGICAL 0 +#define TARGET_CPUS (target_cpus()) +#define APIC_DFR_VALUE (APIC_DFR_FLAT) +#define INT_DELIVERY_MODE (dest_Fixed) +#define INT_DEST_MODE (0) /* phys delivery to target proc */ +#define NO_BALANCE_IRQ (0) +#define WAKE_SECONDARY_VIA_INIT + + +static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid) +{ + return (0); +} + +static inline unsigned long check_apicid_present(int bit) +{ + return (1); +} + +static inline unsigned long calculate_ldr(int cpu) +{ + unsigned long val, id; + val = apic_read(APIC_LDR) & ~APIC_LDR_MASK; + id = xapic_phys_to_log_apicid(cpu); + val |= SET_APIC_LOGICAL_ID(id); + return val; +} + +/* + * Set up the logical destination ID. + * + * Intel recommends to set DFR, LDR and TPR before enabling + * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel + * document number 292116). So here it goes... + */ +static inline void init_apic_ldr(void) +{ + unsigned long val; + int cpu = smp_processor_id(); + + apic_write(APIC_DFR, APIC_DFR_VALUE); + val = calculate_ldr(cpu); + apic_write(APIC_LDR, val); +} + +static inline void setup_apic_routing(void) +{ + printk("Enabling APIC mode: %s. Using %d I/O APICs\n", + "Physflat", nr_ioapics); +} + +static inline int multi_timer_check(int apic, int irq) +{ + return (0); +} + +static inline int apicid_to_node(int logical_apicid) +{ + return apicid_2_node[hard_smp_processor_id()]; +} + +static inline int cpu_present_to_apicid(int mps_cpu) +{ + if (mps_cpu < NR_CPUS) + return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu); + + return BAD_APICID; +} + +static inline physid_mask_t apicid_to_cpu_present(int phys_apicid) +{ + return physid_mask_of_physid(phys_apicid); +} + +extern u8 cpu_2_logical_apicid[]; +/* Mapping from cpu number to logical apicid */ +static inline int cpu_to_logical_apicid(int cpu) +{ + if (cpu >= NR_CPUS) + return BAD_APICID; + return cpu_physical_id(cpu); +} + +static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map) +{ + /* For clustered we don't have a good way to do this yet - hack */ + return physids_promote(0xFFL); +} + +static inline void setup_portio_remap(void) +{ +} + +static inline void enable_apic_mode(void) +{ +} + +static inline int check_phys_apicid_present(int boot_cpu_physical_apicid) +{ + return (1); +} + +/* As we are using single CPU as destination, pick only one CPU here */ +static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) +{ + int cpu; + int apicid; + + cpu = first_cpu(cpumask); + apicid = cpu_to_logical_apicid(cpu); + return apicid; +} + +static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) +{ + return cpuid_apic >> index_msb; +} + +#endif /* __ASM_MACH_APIC_H */ diff --git a/include/asm-x86/bigsmp/apicdef.h b/include/asm-x86/bigsmp/apicdef.h new file mode 100644 index 00000000000..392c3f5ef2f --- /dev/null +++ b/include/asm-x86/bigsmp/apicdef.h @@ -0,0 +1,13 @@ +#ifndef __ASM_MACH_APICDEF_H +#define __ASM_MACH_APICDEF_H + +#define APIC_ID_MASK (0xFF<<24) + +static inline unsigned get_apic_id(unsigned long x) +{ + return (((x)>>24)&0xFF); +} + +#define GET_APIC_ID(x) get_apic_id(x) + +#endif diff --git a/include/asm-x86/bigsmp/ipi.h b/include/asm-x86/bigsmp/ipi.h new file mode 100644 index 00000000000..9404c535b7e --- /dev/null +++ b/include/asm-x86/bigsmp/ipi.h @@ -0,0 +1,25 @@ +#ifndef __ASM_MACH_IPI_H +#define __ASM_MACH_IPI_H + +void send_IPI_mask_sequence(cpumask_t mask, int vector); + +static inline void send_IPI_mask(cpumask_t mask, int vector) +{ + send_IPI_mask_sequence(mask, vector); +} + +static inline void send_IPI_allbutself(int vector) +{ + cpumask_t mask = cpu_online_map; + cpu_clear(smp_processor_id(), mask); + + if (!cpus_empty(mask)) + send_IPI_mask(mask, vector); +} + +static inline void send_IPI_all(int vector) +{ + send_IPI_mask(cpu_online_map, vector); +} + +#endif /* __ASM_MACH_IPI_H */ diff --git a/include/asm-x86/mach-bigsmp/mach_apic.h b/include/asm-x86/mach-bigsmp/mach_apic.h deleted file mode 100644 index 05362d44a3e..00000000000 --- a/include/asm-x86/mach-bigsmp/mach_apic.h +++ /dev/null @@ -1,144 +0,0 @@ -#ifndef ASM_X86__MACH_BIGSMP__MACH_APIC_H -#define ASM_X86__MACH_BIGSMP__MACH_APIC_H - -#define xapic_phys_to_log_apicid(cpu) (per_cpu(x86_bios_cpu_apicid, cpu)) -#define esr_disable (1) - -static inline int apic_id_registered(void) -{ - return (1); -} - -/* Round robin the irqs amoung the online cpus */ -static inline cpumask_t target_cpus(void) -{ - static unsigned long cpu = NR_CPUS; - do { - if (cpu >= NR_CPUS) - cpu = first_cpu(cpu_online_map); - else - cpu = next_cpu(cpu, cpu_online_map); - } while (cpu >= NR_CPUS); - return cpumask_of_cpu(cpu); -} - -#undef APIC_DEST_LOGICAL -#define APIC_DEST_LOGICAL 0 -#define TARGET_CPUS (target_cpus()) -#define APIC_DFR_VALUE (APIC_DFR_FLAT) -#define INT_DELIVERY_MODE (dest_Fixed) -#define INT_DEST_MODE (0) /* phys delivery to target proc */ -#define NO_BALANCE_IRQ (0) -#define WAKE_SECONDARY_VIA_INIT - - -static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid) -{ - return (0); -} - -static inline unsigned long check_apicid_present(int bit) -{ - return (1); -} - -static inline unsigned long calculate_ldr(int cpu) -{ - unsigned long val, id; - val = apic_read(APIC_LDR) & ~APIC_LDR_MASK; - id = xapic_phys_to_log_apicid(cpu); - val |= SET_APIC_LOGICAL_ID(id); - return val; -} - -/* - * Set up the logical destination ID. - * - * Intel recommends to set DFR, LDR and TPR before enabling - * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel - * document number 292116). So here it goes... - */ -static inline void init_apic_ldr(void) -{ - unsigned long val; - int cpu = smp_processor_id(); - - apic_write(APIC_DFR, APIC_DFR_VALUE); - val = calculate_ldr(cpu); - apic_write(APIC_LDR, val); -} - -static inline void setup_apic_routing(void) -{ - printk("Enabling APIC mode: %s. Using %d I/O APICs\n", - "Physflat", nr_ioapics); -} - -static inline int multi_timer_check(int apic, int irq) -{ - return (0); -} - -static inline int apicid_to_node(int logical_apicid) -{ - return apicid_2_node[hard_smp_processor_id()]; -} - -static inline int cpu_present_to_apicid(int mps_cpu) -{ - if (mps_cpu < NR_CPUS) - return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu); - - return BAD_APICID; -} - -static inline physid_mask_t apicid_to_cpu_present(int phys_apicid) -{ - return physid_mask_of_physid(phys_apicid); -} - -extern u8 cpu_2_logical_apicid[]; -/* Mapping from cpu number to logical apicid */ -static inline int cpu_to_logical_apicid(int cpu) -{ - if (cpu >= NR_CPUS) - return BAD_APICID; - return cpu_physical_id(cpu); -} - -static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map) -{ - /* For clustered we don't have a good way to do this yet - hack */ - return physids_promote(0xFFL); -} - -static inline void setup_portio_remap(void) -{ -} - -static inline void enable_apic_mode(void) -{ -} - -static inline int check_phys_apicid_present(int boot_cpu_physical_apicid) -{ - return (1); -} - -/* As we are using single CPU as destination, pick only one CPU here */ -static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) -{ - int cpu; - int apicid; - - cpu = first_cpu(cpumask); - apicid = cpu_to_logical_apicid(cpu); - return apicid; -} - -static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) -{ - return cpuid_apic >> index_msb; -} - -#endif /* ASM_X86__MACH_BIGSMP__MACH_APIC_H */ diff --git a/include/asm-x86/mach-bigsmp/mach_apicdef.h b/include/asm-x86/mach-bigsmp/mach_apicdef.h deleted file mode 100644 index 811935d9d49..00000000000 --- a/include/asm-x86/mach-bigsmp/mach_apicdef.h +++ /dev/null @@ -1,13 +0,0 @@ -#ifndef ASM_X86__MACH_BIGSMP__MACH_APICDEF_H -#define ASM_X86__MACH_BIGSMP__MACH_APICDEF_H - -#define APIC_ID_MASK (0xFF<<24) - -static inline unsigned get_apic_id(unsigned long x) -{ - return (((x)>>24)&0xFF); -} - -#define GET_APIC_ID(x) get_apic_id(x) - -#endif /* ASM_X86__MACH_BIGSMP__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-bigsmp/mach_ipi.h b/include/asm-x86/mach-bigsmp/mach_ipi.h deleted file mode 100644 index b1b0f966a00..00000000000 --- a/include/asm-x86/mach-bigsmp/mach_ipi.h +++ /dev/null @@ -1,25 +0,0 @@ -#ifndef ASM_X86__MACH_BIGSMP__MACH_IPI_H -#define ASM_X86__MACH_BIGSMP__MACH_IPI_H - -void send_IPI_mask_sequence(cpumask_t mask, int vector); - -static inline void send_IPI_mask(cpumask_t mask, int vector) -{ - send_IPI_mask_sequence(mask, vector); -} - -static inline void send_IPI_allbutself(int vector) -{ - cpumask_t mask = cpu_online_map; - cpu_clear(smp_processor_id(), mask); - - if (!cpus_empty(mask)) - send_IPI_mask(mask, vector); -} - -static inline void send_IPI_all(int vector) -{ - send_IPI_mask(cpu_online_map, vector); -} - -#endif /* ASM_X86__MACH_BIGSMP__MACH_IPI_H */ -- cgit v1.2.3-70-g09d2 From c7e7964c9828083aeb41c2664cbf5a32acbfa9d1 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 25 Jul 2008 02:17:33 -0700 Subject: x86: mach_es7000 to es7000 Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/es7000/Makefile | 5 + arch/x86/es7000/es7000.h | 114 +++++++++++++ arch/x86/es7000/es7000plat.c | 262 +++++++++++++++++++++++++++++ arch/x86/mach-es7000/Makefile | 5 - arch/x86/mach-es7000/es7000.h | 114 ------------- arch/x86/mach-es7000/es7000plat.c | 262 ----------------------------- arch/x86/mach-generic/Makefile | 2 +- arch/x86/mach-generic/es7000.c | 10 +- include/asm-x86/es7000/apic.h | 194 +++++++++++++++++++++ include/asm-x86/es7000/apicdef.h | 13 ++ include/asm-x86/es7000/ipi.h | 24 +++ include/asm-x86/es7000/mpparse.h | 29 ++++ include/asm-x86/es7000/wakecpu.h | 59 +++++++ include/asm-x86/mach-es7000/mach_apic.h | 194 --------------------- include/asm-x86/mach-es7000/mach_apicdef.h | 13 -- include/asm-x86/mach-es7000/mach_ipi.h | 24 --- include/asm-x86/mach-es7000/mach_mpparse.h | 29 ---- include/asm-x86/mach-es7000/mach_wakecpu.h | 59 ------- 18 files changed, 706 insertions(+), 706 deletions(-) create mode 100644 arch/x86/es7000/Makefile create mode 100644 arch/x86/es7000/es7000.h create mode 100644 arch/x86/es7000/es7000plat.c delete mode 100644 arch/x86/mach-es7000/Makefile delete mode 100644 arch/x86/mach-es7000/es7000.h delete mode 100644 arch/x86/mach-es7000/es7000plat.c create mode 100644 include/asm-x86/es7000/apic.h create mode 100644 include/asm-x86/es7000/apicdef.h create mode 100644 include/asm-x86/es7000/ipi.h create mode 100644 include/asm-x86/es7000/mpparse.h create mode 100644 include/asm-x86/es7000/wakecpu.h delete mode 100644 include/asm-x86/mach-es7000/mach_apic.h delete mode 100644 include/asm-x86/mach-es7000/mach_apicdef.h delete mode 100644 include/asm-x86/mach-es7000/mach_ipi.h delete mode 100644 include/asm-x86/mach-es7000/mach_mpparse.h delete mode 100644 include/asm-x86/mach-es7000/mach_wakecpu.h (limited to 'include/asm-x86') diff --git a/arch/x86/es7000/Makefile b/arch/x86/es7000/Makefile new file mode 100644 index 00000000000..3ef8b43b62f --- /dev/null +++ b/arch/x86/es7000/Makefile @@ -0,0 +1,5 @@ +# +# Makefile for the linux kernel. +# + +obj-$(CONFIG_X86_ES7000) := es7000plat.o diff --git a/arch/x86/es7000/es7000.h b/arch/x86/es7000/es7000.h new file mode 100644 index 00000000000..4e62f6fa95b --- /dev/null +++ b/arch/x86/es7000/es7000.h @@ -0,0 +1,114 @@ +/* + * Written by: Garry Forsgren, Unisys Corporation + * Natalie Protasevich, Unisys Corporation + * This file contains the code to configure and interface + * with Unisys ES7000 series hardware system manager. + * + * Copyright (c) 2003 Unisys Corporation. All Rights Reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write the Free Software Foundation, Inc., 59 + * Temple Place - Suite 330, Boston MA 02111-1307, USA. + * + * Contact information: Unisys Corporation, Township Line & Union Meeting + * Roads-A, Unisys Way, Blue Bell, Pennsylvania, 19424, or: + * + * http://www.unisys.com + */ + +/* + * ES7000 chipsets + */ + +#define NON_UNISYS 0 +#define ES7000_CLASSIC 1 +#define ES7000_ZORRO 2 + + +#define MIP_REG 1 +#define MIP_PSAI_REG 4 + +#define MIP_BUSY 1 +#define MIP_SPIN 0xf0000 +#define MIP_VALID 0x0100000000000000ULL +#define MIP_PORT(VALUE) ((VALUE >> 32) & 0xffff) + +#define MIP_RD_LO(VALUE) (VALUE & 0xffffffff) + +struct mip_reg_info { + unsigned long long mip_info; + unsigned long long delivery_info; + unsigned long long host_reg; + unsigned long long mip_reg; +}; + +struct part_info { + unsigned char type; + unsigned char length; + unsigned char part_id; + unsigned char apic_mode; + unsigned long snum; + char ptype[16]; + char sname[64]; + char pname[64]; +}; + +struct psai { + unsigned long long entry_type; + unsigned long long addr; + unsigned long long bep_addr; +}; + +struct es7000_mem_info { + unsigned char type; + unsigned char length; + unsigned char resv[6]; + unsigned long long start; + unsigned long long size; +}; + +struct es7000_oem_table { + unsigned long long hdr; + struct mip_reg_info mip; + struct part_info pif; + struct es7000_mem_info shm; + struct psai psai; +}; + +#ifdef CONFIG_ACPI + +struct oem_table { + struct acpi_table_header Header; + u32 OEMTableAddr; + u32 OEMTableSize; +}; + +extern int find_unisys_acpi_oem_table(unsigned long *oem_addr); +#endif + +struct mip_reg { + unsigned long long off_0; + unsigned long long off_8; + unsigned long long off_10; + unsigned long long off_18; + unsigned long long off_20; + unsigned long long off_28; + unsigned long long off_30; + unsigned long long off_38; +}; + +#define MIP_SW_APIC 0x1020b +#define MIP_FUNC(VALUE) (VALUE & 0xff) + +extern int parse_unisys_oem (char *oemptr); +extern void setup_unisys(void); +extern int es7000_start_cpu(int cpu, unsigned long eip); +extern void es7000_sw_apic(void); diff --git a/arch/x86/es7000/es7000plat.c b/arch/x86/es7000/es7000plat.c new file mode 100644 index 00000000000..7789fde13c3 --- /dev/null +++ b/arch/x86/es7000/es7000plat.c @@ -0,0 +1,262 @@ +/* + * Written by: Garry Forsgren, Unisys Corporation + * Natalie Protasevich, Unisys Corporation + * This file contains the code to configure and interface + * with Unisys ES7000 series hardware system manager. + * + * Copyright (c) 2003 Unisys Corporation. All Rights Reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write the Free Software Foundation, Inc., 59 + * Temple Place - Suite 330, Boston MA 02111-1307, USA. + * + * Contact information: Unisys Corporation, Township Line & Union Meeting + * Roads-A, Unisys Way, Blue Bell, Pennsylvania, 19424, or: + * + * http://www.unisys.com + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "es7000.h" +#include + +/* + * ES7000 Globals + */ + +static volatile unsigned long *psai = NULL; +static struct mip_reg *mip_reg; +static struct mip_reg *host_reg; +static int mip_port; +static unsigned long mip_addr, host_addr; + +int es7000_plat; + +/* + * GSI override for ES7000 platforms. + */ + +static unsigned int base; + +static int +es7000_rename_gsi(int ioapic, int gsi) +{ + if (es7000_plat == ES7000_ZORRO) + return gsi; + + if (!base) { + int i; + for (i = 0; i < nr_ioapics; i++) + base += nr_ioapic_registers[i]; + } + + if (!ioapic && (gsi < 16)) + gsi += base; + return gsi; +} + +void __init +setup_unisys(void) +{ + /* + * Determine the generation of the ES7000 currently running. + * + * es7000_plat = 1 if the machine is a 5xx ES7000 box + * es7000_plat = 2 if the machine is a x86_64 ES7000 box + * + */ + if (!(boot_cpu_data.x86 <= 15 && boot_cpu_data.x86_model <= 2)) + es7000_plat = ES7000_ZORRO; + else + es7000_plat = ES7000_CLASSIC; + ioapic_renumber_irq = es7000_rename_gsi; +} + +/* + * Parse the OEM Table + */ + +int __init +parse_unisys_oem (char *oemptr) +{ + int i; + int success = 0; + unsigned char type, size; + unsigned long val; + char *tp = NULL; + struct psai *psaip = NULL; + struct mip_reg_info *mi; + struct mip_reg *host, *mip; + + tp = oemptr; + + tp += 8; + + for (i=0; i <= 6; i++) { + type = *tp++; + size = *tp++; + tp -= 2; + switch (type) { + case MIP_REG: + mi = (struct mip_reg_info *)tp; + val = MIP_RD_LO(mi->host_reg); + host_addr = val; + host = (struct mip_reg *)val; + host_reg = __va(host); + val = MIP_RD_LO(mi->mip_reg); + mip_port = MIP_PORT(mi->mip_info); + mip_addr = val; + mip = (struct mip_reg *)val; + mip_reg = __va(mip); + pr_debug("es7000_mipcfg: host_reg = 0x%lx \n", + (unsigned long)host_reg); + pr_debug("es7000_mipcfg: mip_reg = 0x%lx \n", + (unsigned long)mip_reg); + success++; + break; + case MIP_PSAI_REG: + psaip = (struct psai *)tp; + if (tp != NULL) { + if (psaip->addr) + psai = __va(psaip->addr); + else + psai = NULL; + success++; + } + break; + default: + break; + } + tp += size; + } + + if (success < 2) { + es7000_plat = NON_UNISYS; + } else + setup_unisys(); + return es7000_plat; +} + +#ifdef CONFIG_ACPI +int __init +find_unisys_acpi_oem_table(unsigned long *oem_addr) +{ + struct acpi_table_header *header = NULL; + int i = 0; + while (ACPI_SUCCESS(acpi_get_table("OEM1", i++, &header))) { + if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) { + struct oem_table *t = (struct oem_table *)header; + *oem_addr = (unsigned long)__acpi_map_table(t->OEMTableAddr, + t->OEMTableSize); + return 0; + } + } + return -1; +} +#endif + +static void +es7000_spin(int n) +{ + int i = 0; + + while (i++ < n) + rep_nop(); +} + +static int __init +es7000_mip_write(struct mip_reg *mip_reg) +{ + int status = 0; + int spin; + + spin = MIP_SPIN; + while (((unsigned long long)host_reg->off_38 & + (unsigned long long)MIP_VALID) != 0) { + if (--spin <= 0) { + printk("es7000_mip_write: Timeout waiting for Host Valid Flag"); + return -1; + } + es7000_spin(MIP_SPIN); + } + + memcpy(host_reg, mip_reg, sizeof(struct mip_reg)); + outb(1, mip_port); + + spin = MIP_SPIN; + + while (((unsigned long long)mip_reg->off_38 & + (unsigned long long)MIP_VALID) == 0) { + if (--spin <= 0) { + printk("es7000_mip_write: Timeout waiting for MIP Valid Flag"); + return -1; + } + es7000_spin(MIP_SPIN); + } + + status = ((unsigned long long)mip_reg->off_0 & + (unsigned long long)0xffff0000000000ULL) >> 48; + mip_reg->off_38 = ((unsigned long long)mip_reg->off_38 & + (unsigned long long)~MIP_VALID); + return status; +} + +int +es7000_start_cpu(int cpu, unsigned long eip) +{ + unsigned long vect = 0, psaival = 0; + + if (psai == NULL) + return -1; + + vect = ((unsigned long)__pa(eip)/0x1000) << 16; + psaival = (0x1000000 | vect | cpu); + + while (*psai & 0x1000000) + ; + + *psai = psaival; + + return 0; + +} + +void __init +es7000_sw_apic(void) +{ + if (es7000_plat) { + int mip_status; + struct mip_reg es7000_mip_reg; + + printk("ES7000: Enabling APIC mode.\n"); + memset(&es7000_mip_reg, 0, sizeof(struct mip_reg)); + es7000_mip_reg.off_0 = MIP_SW_APIC; + es7000_mip_reg.off_38 = (MIP_VALID); + while ((mip_status = es7000_mip_write(&es7000_mip_reg)) != 0) + printk("es7000_sw_apic: command failed, status = %x\n", + mip_status); + return; + } +} diff --git a/arch/x86/mach-es7000/Makefile b/arch/x86/mach-es7000/Makefile deleted file mode 100644 index 3ef8b43b62f..00000000000 --- a/arch/x86/mach-es7000/Makefile +++ /dev/null @@ -1,5 +0,0 @@ -# -# Makefile for the linux kernel. -# - -obj-$(CONFIG_X86_ES7000) := es7000plat.o diff --git a/arch/x86/mach-es7000/es7000.h b/arch/x86/mach-es7000/es7000.h deleted file mode 100644 index c8d5aa132fa..00000000000 --- a/arch/x86/mach-es7000/es7000.h +++ /dev/null @@ -1,114 +0,0 @@ -/* - * Written by: Garry Forsgren, Unisys Corporation - * Natalie Protasevich, Unisys Corporation - * This file contains the code to configure and interface - * with Unisys ES7000 series hardware system manager. - * - * Copyright (c) 2003 Unisys Corporation. All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Unisys Corporation, Township Line & Union Meeting - * Roads-A, Unisys Way, Blue Bell, Pennsylvania, 19424, or: - * - * http://www.unisys.com - */ - -/* - * ES7000 chipsets - */ - -#define NON_UNISYS 0 -#define ES7000_CLASSIC 1 -#define ES7000_ZORRO 2 - - -#define MIP_REG 1 -#define MIP_PSAI_REG 4 - -#define MIP_BUSY 1 -#define MIP_SPIN 0xf0000 -#define MIP_VALID 0x0100000000000000ULL -#define MIP_PORT(VALUE) ((VALUE >> 32) & 0xffff) - -#define MIP_RD_LO(VALUE) (VALUE & 0xffffffff) - -struct mip_reg_info { - unsigned long long mip_info; - unsigned long long delivery_info; - unsigned long long host_reg; - unsigned long long mip_reg; -}; - -struct part_info { - unsigned char type; - unsigned char length; - unsigned char part_id; - unsigned char apic_mode; - unsigned long snum; - char ptype[16]; - char sname[64]; - char pname[64]; -}; - -struct psai { - unsigned long long entry_type; - unsigned long long addr; - unsigned long long bep_addr; -}; - -struct es7000_mem_info { - unsigned char type; - unsigned char length; - unsigned char resv[6]; - unsigned long long start; - unsigned long long size; -}; - -struct es7000_oem_table { - unsigned long long hdr; - struct mip_reg_info mip; - struct part_info pif; - struct es7000_mem_info shm; - struct psai psai; -}; - -#ifdef CONFIG_ACPI - -struct oem_table { - struct acpi_table_header Header; - u32 OEMTableAddr; - u32 OEMTableSize; -}; - -extern int find_unisys_acpi_oem_table(unsigned long *oem_addr); -#endif - -struct mip_reg { - unsigned long long off_0; - unsigned long long off_8; - unsigned long long off_10; - unsigned long long off_18; - unsigned long long off_20; - unsigned long long off_28; - unsigned long long off_30; - unsigned long long off_38; -}; - -#define MIP_SW_APIC 0x1020b -#define MIP_FUNC(VALUE) (VALUE & 0xff) - -extern int parse_unisys_oem (char *oemptr); -extern void setup_unisys(void); -extern int es7000_start_cpu(int cpu, unsigned long eip); -extern void es7000_sw_apic(void); diff --git a/arch/x86/mach-es7000/es7000plat.c b/arch/x86/mach-es7000/es7000plat.c deleted file mode 100644 index 50189af14b8..00000000000 --- a/arch/x86/mach-es7000/es7000plat.c +++ /dev/null @@ -1,262 +0,0 @@ -/* - * Written by: Garry Forsgren, Unisys Corporation - * Natalie Protasevich, Unisys Corporation - * This file contains the code to configure and interface - * with Unisys ES7000 series hardware system manager. - * - * Copyright (c) 2003 Unisys Corporation. All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Unisys Corporation, Township Line & Union Meeting - * Roads-A, Unisys Way, Blue Bell, Pennsylvania, 19424, or: - * - * http://www.unisys.com - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "es7000.h" -#include - -/* - * ES7000 Globals - */ - -static volatile unsigned long *psai = NULL; -static struct mip_reg *mip_reg; -static struct mip_reg *host_reg; -static int mip_port; -static unsigned long mip_addr, host_addr; - -int es7000_plat; - -/* - * GSI override for ES7000 platforms. - */ - -static unsigned int base; - -static int -es7000_rename_gsi(int ioapic, int gsi) -{ - if (es7000_plat == ES7000_ZORRO) - return gsi; - - if (!base) { - int i; - for (i = 0; i < nr_ioapics; i++) - base += nr_ioapic_registers[i]; - } - - if (!ioapic && (gsi < 16)) - gsi += base; - return gsi; -} - -void __init -setup_unisys(void) -{ - /* - * Determine the generation of the ES7000 currently running. - * - * es7000_plat = 1 if the machine is a 5xx ES7000 box - * es7000_plat = 2 if the machine is a x86_64 ES7000 box - * - */ - if (!(boot_cpu_data.x86 <= 15 && boot_cpu_data.x86_model <= 2)) - es7000_plat = ES7000_ZORRO; - else - es7000_plat = ES7000_CLASSIC; - ioapic_renumber_irq = es7000_rename_gsi; -} - -/* - * Parse the OEM Table - */ - -int __init -parse_unisys_oem (char *oemptr) -{ - int i; - int success = 0; - unsigned char type, size; - unsigned long val; - char *tp = NULL; - struct psai *psaip = NULL; - struct mip_reg_info *mi; - struct mip_reg *host, *mip; - - tp = oemptr; - - tp += 8; - - for (i=0; i <= 6; i++) { - type = *tp++; - size = *tp++; - tp -= 2; - switch (type) { - case MIP_REG: - mi = (struct mip_reg_info *)tp; - val = MIP_RD_LO(mi->host_reg); - host_addr = val; - host = (struct mip_reg *)val; - host_reg = __va(host); - val = MIP_RD_LO(mi->mip_reg); - mip_port = MIP_PORT(mi->mip_info); - mip_addr = val; - mip = (struct mip_reg *)val; - mip_reg = __va(mip); - pr_debug("es7000_mipcfg: host_reg = 0x%lx \n", - (unsigned long)host_reg); - pr_debug("es7000_mipcfg: mip_reg = 0x%lx \n", - (unsigned long)mip_reg); - success++; - break; - case MIP_PSAI_REG: - psaip = (struct psai *)tp; - if (tp != NULL) { - if (psaip->addr) - psai = __va(psaip->addr); - else - psai = NULL; - success++; - } - break; - default: - break; - } - tp += size; - } - - if (success < 2) { - es7000_plat = NON_UNISYS; - } else - setup_unisys(); - return es7000_plat; -} - -#ifdef CONFIG_ACPI -int __init -find_unisys_acpi_oem_table(unsigned long *oem_addr) -{ - struct acpi_table_header *header = NULL; - int i = 0; - while (ACPI_SUCCESS(acpi_get_table("OEM1", i++, &header))) { - if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) { - struct oem_table *t = (struct oem_table *)header; - *oem_addr = (unsigned long)__acpi_map_table(t->OEMTableAddr, - t->OEMTableSize); - return 0; - } - } - return -1; -} -#endif - -static void -es7000_spin(int n) -{ - int i = 0; - - while (i++ < n) - rep_nop(); -} - -static int __init -es7000_mip_write(struct mip_reg *mip_reg) -{ - int status = 0; - int spin; - - spin = MIP_SPIN; - while (((unsigned long long)host_reg->off_38 & - (unsigned long long)MIP_VALID) != 0) { - if (--spin <= 0) { - printk("es7000_mip_write: Timeout waiting for Host Valid Flag"); - return -1; - } - es7000_spin(MIP_SPIN); - } - - memcpy(host_reg, mip_reg, sizeof(struct mip_reg)); - outb(1, mip_port); - - spin = MIP_SPIN; - - while (((unsigned long long)mip_reg->off_38 & - (unsigned long long)MIP_VALID) == 0) { - if (--spin <= 0) { - printk("es7000_mip_write: Timeout waiting for MIP Valid Flag"); - return -1; - } - es7000_spin(MIP_SPIN); - } - - status = ((unsigned long long)mip_reg->off_0 & - (unsigned long long)0xffff0000000000ULL) >> 48; - mip_reg->off_38 = ((unsigned long long)mip_reg->off_38 & - (unsigned long long)~MIP_VALID); - return status; -} - -int -es7000_start_cpu(int cpu, unsigned long eip) -{ - unsigned long vect = 0, psaival = 0; - - if (psai == NULL) - return -1; - - vect = ((unsigned long)__pa(eip)/0x1000) << 16; - psaival = (0x1000000 | vect | cpu); - - while (*psai & 0x1000000) - ; - - *psai = psaival; - - return 0; - -} - -void __init -es7000_sw_apic(void) -{ - if (es7000_plat) { - int mip_status; - struct mip_reg es7000_mip_reg; - - printk("ES7000: Enabling APIC mode.\n"); - memset(&es7000_mip_reg, 0, sizeof(struct mip_reg)); - es7000_mip_reg.off_0 = MIP_SW_APIC; - es7000_mip_reg.off_38 = (MIP_VALID); - while ((mip_status = es7000_mip_write(&es7000_mip_reg)) != 0) - printk("es7000_sw_apic: command failed, status = %x\n", - mip_status); - return; - } -} diff --git a/arch/x86/mach-generic/Makefile b/arch/x86/mach-generic/Makefile index 0dbd7803a1d..4706de7676b 100644 --- a/arch/x86/mach-generic/Makefile +++ b/arch/x86/mach-generic/Makefile @@ -9,4 +9,4 @@ obj-$(CONFIG_X86_NUMAQ) += numaq.o obj-$(CONFIG_X86_SUMMIT) += summit.o obj-$(CONFIG_X86_BIGSMP) += bigsmp.o obj-$(CONFIG_X86_ES7000) += es7000.o -obj-$(CONFIG_X86_ES7000) += ../../x86/mach-es7000/ +obj-$(CONFIG_X86_ES7000) += ../../x86/es7000/ diff --git a/arch/x86/mach-generic/es7000.c b/arch/x86/mach-generic/es7000.c index 9b30547d746..520cca0ee04 100644 --- a/arch/x86/mach-generic/es7000.c +++ b/arch/x86/mach-generic/es7000.c @@ -11,12 +11,12 @@ #include #include #include -#include +#include #include -#include -#include -#include -#include +#include +#include +#include +#include static int probe_es7000(void) { diff --git a/include/asm-x86/es7000/apic.h b/include/asm-x86/es7000/apic.h new file mode 100644 index 00000000000..bd2c44d1f7a --- /dev/null +++ b/include/asm-x86/es7000/apic.h @@ -0,0 +1,194 @@ +#ifndef __ASM_ES7000_APIC_H +#define __ASM_ES7000_APIC_H + +#define xapic_phys_to_log_apicid(cpu) per_cpu(x86_bios_cpu_apicid, cpu) +#define esr_disable (1) + +static inline int apic_id_registered(void) +{ + return (1); +} + +static inline cpumask_t target_cpus(void) +{ +#if defined CONFIG_ES7000_CLUSTERED_APIC + return CPU_MASK_ALL; +#else + return cpumask_of_cpu(smp_processor_id()); +#endif +} +#define TARGET_CPUS (target_cpus()) + +#if defined CONFIG_ES7000_CLUSTERED_APIC +#define APIC_DFR_VALUE (APIC_DFR_CLUSTER) +#define INT_DELIVERY_MODE (dest_LowestPrio) +#define INT_DEST_MODE (1) /* logical delivery broadcast to all procs */ +#define NO_BALANCE_IRQ (1) +#undef WAKE_SECONDARY_VIA_INIT +#define WAKE_SECONDARY_VIA_MIP +#else +#define APIC_DFR_VALUE (APIC_DFR_FLAT) +#define INT_DELIVERY_MODE (dest_Fixed) +#define INT_DEST_MODE (0) /* phys delivery to target procs */ +#define NO_BALANCE_IRQ (0) +#undef APIC_DEST_LOGICAL +#define APIC_DEST_LOGICAL 0x0 +#define WAKE_SECONDARY_VIA_INIT +#endif + +static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid) +{ + return 0; +} +static inline unsigned long check_apicid_present(int bit) +{ + return physid_isset(bit, phys_cpu_present_map); +} + +#define apicid_cluster(apicid) (apicid & 0xF0) + +static inline unsigned long calculate_ldr(int cpu) +{ + unsigned long id; + id = xapic_phys_to_log_apicid(cpu); + return (SET_APIC_LOGICAL_ID(id)); +} + +/* + * Set up the logical destination ID. + * + * Intel recommends to set DFR, LdR and TPR before enabling + * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel + * document number 292116). So here it goes... + */ +static inline void init_apic_ldr(void) +{ + unsigned long val; + int cpu = smp_processor_id(); + + apic_write(APIC_DFR, APIC_DFR_VALUE); + val = calculate_ldr(cpu); + apic_write(APIC_LDR, val); +} + +#ifndef CONFIG_X86_GENERICARCH +extern void enable_apic_mode(void); +#endif + +extern int apic_version [MAX_APICS]; +static inline void setup_apic_routing(void) +{ + int apic = per_cpu(x86_bios_cpu_apicid, smp_processor_id()); + printk("Enabling APIC mode: %s. Using %d I/O APICs, target cpus %lx\n", + (apic_version[apic] == 0x14) ? + "Physical Cluster" : "Logical Cluster", nr_ioapics, cpus_addr(TARGET_CPUS)[0]); +} + +static inline int multi_timer_check(int apic, int irq) +{ + return 0; +} + +static inline int apicid_to_node(int logical_apicid) +{ + return 0; +} + + +static inline int cpu_present_to_apicid(int mps_cpu) +{ + if (!mps_cpu) + return boot_cpu_physical_apicid; + else if (mps_cpu < NR_CPUS) + return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu); + else + return BAD_APICID; +} + +static inline physid_mask_t apicid_to_cpu_present(int phys_apicid) +{ + static int id = 0; + physid_mask_t mask; + mask = physid_mask_of_physid(id); + ++id; + return mask; +} + +extern u8 cpu_2_logical_apicid[]; +/* Mapping from cpu number to logical apicid */ +static inline int cpu_to_logical_apicid(int cpu) +{ +#ifdef CONFIG_SMP + if (cpu >= NR_CPUS) + return BAD_APICID; + return (int)cpu_2_logical_apicid[cpu]; +#else + return logical_smp_processor_id(); +#endif +} + +static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map) +{ + /* For clustered we don't have a good way to do this yet - hack */ + return physids_promote(0xff); +} + + +static inline void setup_portio_remap(void) +{ +} + +extern unsigned int boot_cpu_physical_apicid; +static inline int check_phys_apicid_present(int cpu_physical_apicid) +{ + boot_cpu_physical_apicid = read_apic_id(); + return (1); +} + +static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) +{ + int num_bits_set; + int cpus_found = 0; + int cpu; + int apicid; + + num_bits_set = cpus_weight(cpumask); + /* Return id to all */ + if (num_bits_set == NR_CPUS) +#if defined CONFIG_ES7000_CLUSTERED_APIC + return 0xFF; +#else + return cpu_to_logical_apicid(0); +#endif + /* + * The cpus in the mask must all be on the apic cluster. If are not + * on the same apicid cluster return default value of TARGET_CPUS. + */ + cpu = first_cpu(cpumask); + apicid = cpu_to_logical_apicid(cpu); + while (cpus_found < num_bits_set) { + if (cpu_isset(cpu, cpumask)) { + int new_apicid = cpu_to_logical_apicid(cpu); + if (apicid_cluster(apicid) != + apicid_cluster(new_apicid)){ + printk ("%s: Not a valid mask!\n",__FUNCTION__); +#if defined CONFIG_ES7000_CLUSTERED_APIC + return 0xFF; +#else + return cpu_to_logical_apicid(0); +#endif + } + apicid = new_apicid; + cpus_found++; + } + cpu++; + } + return apicid; +} + +static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) +{ + return cpuid_apic >> index_msb; +} + +#endif /* __ASM_ES7000_APIC_H */ diff --git a/include/asm-x86/es7000/apicdef.h b/include/asm-x86/es7000/apicdef.h new file mode 100644 index 00000000000..8b234a3cb85 --- /dev/null +++ b/include/asm-x86/es7000/apicdef.h @@ -0,0 +1,13 @@ +#ifndef __ASM_ES7000_APICDEF_H +#define __ASM_ES7000_APICDEF_H + +#define APIC_ID_MASK (0xFF<<24) + +static inline unsigned get_apic_id(unsigned long x) +{ + return (((x)>>24)&0xFF); +} + +#define GET_APIC_ID(x) get_apic_id(x) + +#endif diff --git a/include/asm-x86/es7000/ipi.h b/include/asm-x86/es7000/ipi.h new file mode 100644 index 00000000000..632a955fcc0 --- /dev/null +++ b/include/asm-x86/es7000/ipi.h @@ -0,0 +1,24 @@ +#ifndef __ASM_ES7000_IPI_H +#define __ASM_ES7000_IPI_H + +void send_IPI_mask_sequence(cpumask_t mask, int vector); + +static inline void send_IPI_mask(cpumask_t mask, int vector) +{ + send_IPI_mask_sequence(mask, vector); +} + +static inline void send_IPI_allbutself(int vector) +{ + cpumask_t mask = cpu_online_map; + cpu_clear(smp_processor_id(), mask); + if (!cpus_empty(mask)) + send_IPI_mask(mask, vector); +} + +static inline void send_IPI_all(int vector) +{ + send_IPI_mask(cpu_online_map, vector); +} + +#endif /* __ASM_ES7000_IPI_H */ diff --git a/include/asm-x86/es7000/mpparse.h b/include/asm-x86/es7000/mpparse.h new file mode 100644 index 00000000000..7b5c889d8e7 --- /dev/null +++ b/include/asm-x86/es7000/mpparse.h @@ -0,0 +1,29 @@ +#ifndef __ASM_ES7000_MPPARSE_H +#define __ASM_ES7000_MPPARSE_H + +#include + +extern int parse_unisys_oem (char *oemptr); +extern int find_unisys_acpi_oem_table(unsigned long *oem_addr); +extern void setup_unisys(void); + +#ifndef CONFIG_X86_GENERICARCH +extern int acpi_madt_oem_check(char *oem_id, char *oem_table_id); +extern int mps_oem_check(struct mp_config_table *mpc, char *oem, + char *productid); +#endif + +#ifdef CONFIG_ACPI + +static inline int es7000_check_dsdt(void) +{ + struct acpi_table_header header; + + if (ACPI_SUCCESS(acpi_get_table_header(ACPI_SIG_DSDT, 0, &header)) && + !strncmp(header.oem_id, "UNISYS", 6)) + return 1; + return 0; +} +#endif + +#endif /* __ASM_MACH_MPPARSE_H */ diff --git a/include/asm-x86/es7000/wakecpu.h b/include/asm-x86/es7000/wakecpu.h new file mode 100644 index 00000000000..3ffc5a7bf66 --- /dev/null +++ b/include/asm-x86/es7000/wakecpu.h @@ -0,0 +1,59 @@ +#ifndef __ASM_ES7000_WAKECPU_H +#define __ASM_ES7000_WAKECPU_H + +/* + * This file copes with machines that wakeup secondary CPUs by the + * INIT, INIT, STARTUP sequence. + */ + +#ifdef CONFIG_ES7000_CLUSTERED_APIC +#define WAKE_SECONDARY_VIA_MIP +#else +#define WAKE_SECONDARY_VIA_INIT +#endif + +#ifdef WAKE_SECONDARY_VIA_MIP +extern int es7000_start_cpu(int cpu, unsigned long eip); +static inline int +wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) +{ + int boot_error = 0; + boot_error = es7000_start_cpu(phys_apicid, start_eip); + return boot_error; +} +#endif + +#define TRAMPOLINE_LOW phys_to_virt(0x467) +#define TRAMPOLINE_HIGH phys_to_virt(0x469) + +#define boot_cpu_apicid boot_cpu_physical_apicid + +static inline void wait_for_init_deassert(atomic_t *deassert) +{ +#ifdef WAKE_SECONDARY_VIA_INIT + while (!atomic_read(deassert)) + cpu_relax(); +#endif + return; +} + +/* Nothing to do for most platforms, since cleared by the INIT cycle */ +static inline void smp_callin_clear_local_apic(void) +{ +} + +static inline void store_NMI_vector(unsigned short *high, unsigned short *low) +{ +} + +static inline void restore_NMI_vector(unsigned short *high, unsigned short *low) +{ +} + +#if APIC_DEBUG + #define inquire_remote_apic(apicid) __inquire_remote_apic(apicid) +#else + #define inquire_remote_apic(apicid) {} +#endif + +#endif /* __ASM_MACH_WAKECPU_H */ diff --git a/include/asm-x86/mach-es7000/mach_apic.h b/include/asm-x86/mach-es7000/mach_apic.h deleted file mode 100644 index ae877ec267f..00000000000 --- a/include/asm-x86/mach-es7000/mach_apic.h +++ /dev/null @@ -1,194 +0,0 @@ -#ifndef ASM_X86__MACH_ES7000__MACH_APIC_H -#define ASM_X86__MACH_ES7000__MACH_APIC_H - -#define xapic_phys_to_log_apicid(cpu) per_cpu(x86_bios_cpu_apicid, cpu) -#define esr_disable (1) - -static inline int apic_id_registered(void) -{ - return (1); -} - -static inline cpumask_t target_cpus(void) -{ -#if defined CONFIG_ES7000_CLUSTERED_APIC - return CPU_MASK_ALL; -#else - return cpumask_of_cpu(smp_processor_id()); -#endif -} -#define TARGET_CPUS (target_cpus()) - -#if defined CONFIG_ES7000_CLUSTERED_APIC -#define APIC_DFR_VALUE (APIC_DFR_CLUSTER) -#define INT_DELIVERY_MODE (dest_LowestPrio) -#define INT_DEST_MODE (1) /* logical delivery broadcast to all procs */ -#define NO_BALANCE_IRQ (1) -#undef WAKE_SECONDARY_VIA_INIT -#define WAKE_SECONDARY_VIA_MIP -#else -#define APIC_DFR_VALUE (APIC_DFR_FLAT) -#define INT_DELIVERY_MODE (dest_Fixed) -#define INT_DEST_MODE (0) /* phys delivery to target procs */ -#define NO_BALANCE_IRQ (0) -#undef APIC_DEST_LOGICAL -#define APIC_DEST_LOGICAL 0x0 -#define WAKE_SECONDARY_VIA_INIT -#endif - -static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid) -{ - return 0; -} -static inline unsigned long check_apicid_present(int bit) -{ - return physid_isset(bit, phys_cpu_present_map); -} - -#define apicid_cluster(apicid) (apicid & 0xF0) - -static inline unsigned long calculate_ldr(int cpu) -{ - unsigned long id; - id = xapic_phys_to_log_apicid(cpu); - return (SET_APIC_LOGICAL_ID(id)); -} - -/* - * Set up the logical destination ID. - * - * Intel recommends to set DFR, LdR and TPR before enabling - * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel - * document number 292116). So here it goes... - */ -static inline void init_apic_ldr(void) -{ - unsigned long val; - int cpu = smp_processor_id(); - - apic_write(APIC_DFR, APIC_DFR_VALUE); - val = calculate_ldr(cpu); - apic_write(APIC_LDR, val); -} - -#ifndef CONFIG_X86_GENERICARCH -extern void enable_apic_mode(void); -#endif - -extern int apic_version [MAX_APICS]; -static inline void setup_apic_routing(void) -{ - int apic = per_cpu(x86_bios_cpu_apicid, smp_processor_id()); - printk("Enabling APIC mode: %s. Using %d I/O APICs, target cpus %lx\n", - (apic_version[apic] == 0x14) ? - "Physical Cluster" : "Logical Cluster", nr_ioapics, cpus_addr(TARGET_CPUS)[0]); -} - -static inline int multi_timer_check(int apic, int irq) -{ - return 0; -} - -static inline int apicid_to_node(int logical_apicid) -{ - return 0; -} - - -static inline int cpu_present_to_apicid(int mps_cpu) -{ - if (!mps_cpu) - return boot_cpu_physical_apicid; - else if (mps_cpu < NR_CPUS) - return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu); - else - return BAD_APICID; -} - -static inline physid_mask_t apicid_to_cpu_present(int phys_apicid) -{ - static int id = 0; - physid_mask_t mask; - mask = physid_mask_of_physid(id); - ++id; - return mask; -} - -extern u8 cpu_2_logical_apicid[]; -/* Mapping from cpu number to logical apicid */ -static inline int cpu_to_logical_apicid(int cpu) -{ -#ifdef CONFIG_SMP - if (cpu >= NR_CPUS) - return BAD_APICID; - return (int)cpu_2_logical_apicid[cpu]; -#else - return logical_smp_processor_id(); -#endif -} - -static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map) -{ - /* For clustered we don't have a good way to do this yet - hack */ - return physids_promote(0xff); -} - - -static inline void setup_portio_remap(void) -{ -} - -extern unsigned int boot_cpu_physical_apicid; -static inline int check_phys_apicid_present(int cpu_physical_apicid) -{ - boot_cpu_physical_apicid = read_apic_id(); - return (1); -} - -static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) -{ - int num_bits_set; - int cpus_found = 0; - int cpu; - int apicid; - - num_bits_set = cpus_weight(cpumask); - /* Return id to all */ - if (num_bits_set == NR_CPUS) -#if defined CONFIG_ES7000_CLUSTERED_APIC - return 0xFF; -#else - return cpu_to_logical_apicid(0); -#endif - /* - * The cpus in the mask must all be on the apic cluster. If are not - * on the same apicid cluster return default value of TARGET_CPUS. - */ - cpu = first_cpu(cpumask); - apicid = cpu_to_logical_apicid(cpu); - while (cpus_found < num_bits_set) { - if (cpu_isset(cpu, cpumask)) { - int new_apicid = cpu_to_logical_apicid(cpu); - if (apicid_cluster(apicid) != - apicid_cluster(new_apicid)){ - printk ("%s: Not a valid mask!\n",__FUNCTION__); -#if defined CONFIG_ES7000_CLUSTERED_APIC - return 0xFF; -#else - return cpu_to_logical_apicid(0); -#endif - } - apicid = new_apicid; - cpus_found++; - } - cpu++; - } - return apicid; -} - -static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) -{ - return cpuid_apic >> index_msb; -} - -#endif /* ASM_X86__MACH_ES7000__MACH_APIC_H */ diff --git a/include/asm-x86/mach-es7000/mach_apicdef.h b/include/asm-x86/mach-es7000/mach_apicdef.h deleted file mode 100644 index a07e5674402..00000000000 --- a/include/asm-x86/mach-es7000/mach_apicdef.h +++ /dev/null @@ -1,13 +0,0 @@ -#ifndef ASM_X86__MACH_ES7000__MACH_APICDEF_H -#define ASM_X86__MACH_ES7000__MACH_APICDEF_H - -#define APIC_ID_MASK (0xFF<<24) - -static inline unsigned get_apic_id(unsigned long x) -{ - return (((x)>>24)&0xFF); -} - -#define GET_APIC_ID(x) get_apic_id(x) - -#endif /* ASM_X86__MACH_ES7000__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-es7000/mach_ipi.h b/include/asm-x86/mach-es7000/mach_ipi.h deleted file mode 100644 index 3a21240e03d..00000000000 --- a/include/asm-x86/mach-es7000/mach_ipi.h +++ /dev/null @@ -1,24 +0,0 @@ -#ifndef ASM_X86__MACH_ES7000__MACH_IPI_H -#define ASM_X86__MACH_ES7000__MACH_IPI_H - -void send_IPI_mask_sequence(cpumask_t mask, int vector); - -static inline void send_IPI_mask(cpumask_t mask, int vector) -{ - send_IPI_mask_sequence(mask, vector); -} - -static inline void send_IPI_allbutself(int vector) -{ - cpumask_t mask = cpu_online_map; - cpu_clear(smp_processor_id(), mask); - if (!cpus_empty(mask)) - send_IPI_mask(mask, vector); -} - -static inline void send_IPI_all(int vector) -{ - send_IPI_mask(cpu_online_map, vector); -} - -#endif /* ASM_X86__MACH_ES7000__MACH_IPI_H */ diff --git a/include/asm-x86/mach-es7000/mach_mpparse.h b/include/asm-x86/mach-es7000/mach_mpparse.h deleted file mode 100644 index befde24705b..00000000000 --- a/include/asm-x86/mach-es7000/mach_mpparse.h +++ /dev/null @@ -1,29 +0,0 @@ -#ifndef ASM_X86__MACH_ES7000__MACH_MPPARSE_H -#define ASM_X86__MACH_ES7000__MACH_MPPARSE_H - -#include - -extern int parse_unisys_oem (char *oemptr); -extern int find_unisys_acpi_oem_table(unsigned long *oem_addr); -extern void setup_unisys(void); - -#ifndef CONFIG_X86_GENERICARCH -extern int acpi_madt_oem_check(char *oem_id, char *oem_table_id); -extern int mps_oem_check(struct mp_config_table *mpc, char *oem, - char *productid); -#endif - -#ifdef CONFIG_ACPI - -static inline int es7000_check_dsdt(void) -{ - struct acpi_table_header header; - - if (ACPI_SUCCESS(acpi_get_table_header(ACPI_SIG_DSDT, 0, &header)) && - !strncmp(header.oem_id, "UNISYS", 6)) - return 1; - return 0; -} -#endif - -#endif /* ASM_X86__MACH_ES7000__MACH_MPPARSE_H */ diff --git a/include/asm-x86/mach-es7000/mach_wakecpu.h b/include/asm-x86/mach-es7000/mach_wakecpu.h deleted file mode 100644 index 97c776ce13f..00000000000 --- a/include/asm-x86/mach-es7000/mach_wakecpu.h +++ /dev/null @@ -1,59 +0,0 @@ -#ifndef ASM_X86__MACH_ES7000__MACH_WAKECPU_H -#define ASM_X86__MACH_ES7000__MACH_WAKECPU_H - -/* - * This file copes with machines that wakeup secondary CPUs by the - * INIT, INIT, STARTUP sequence. - */ - -#ifdef CONFIG_ES7000_CLUSTERED_APIC -#define WAKE_SECONDARY_VIA_MIP -#else -#define WAKE_SECONDARY_VIA_INIT -#endif - -#ifdef WAKE_SECONDARY_VIA_MIP -extern int es7000_start_cpu(int cpu, unsigned long eip); -static inline int -wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip) -{ - int boot_error = 0; - boot_error = es7000_start_cpu(phys_apicid, start_eip); - return boot_error; -} -#endif - -#define TRAMPOLINE_LOW phys_to_virt(0x467) -#define TRAMPOLINE_HIGH phys_to_virt(0x469) - -#define boot_cpu_apicid boot_cpu_physical_apicid - -static inline void wait_for_init_deassert(atomic_t *deassert) -{ -#ifdef WAKE_SECONDARY_VIA_INIT - while (!atomic_read(deassert)) - cpu_relax(); -#endif - return; -} - -/* Nothing to do for most platforms, since cleared by the INIT cycle */ -static inline void smp_callin_clear_local_apic(void) -{ -} - -static inline void store_NMI_vector(unsigned short *high, unsigned short *low) -{ -} - -static inline void restore_NMI_vector(unsigned short *high, unsigned short *low) -{ -} - -#if APIC_DEBUG - #define inquire_remote_apic(apicid) __inquire_remote_apic(apicid) -#else - #define inquire_remote_apic(apicid) {} -#endif - -#endif /* ASM_X86__MACH_ES7000__MACH_WAKECPU_H */ -- cgit v1.2.3-70-g09d2 From e8c48efdb971cfb1b043076cf68be535d461a20b Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 25 Jul 2008 02:17:44 -0700 Subject: x86: mach_summit to summit Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/summit_32.c | 2 +- arch/x86/mach-generic/summit.c | 8 +- include/asm-x86/mach-summit/irq_vectors_limits.h | 14 -- include/asm-x86/mach-summit/mach_apic.h | 185 ----------------------- include/asm-x86/mach-summit/mach_apicdef.h | 13 -- include/asm-x86/mach-summit/mach_ipi.h | 25 --- include/asm-x86/mach-summit/mach_mpparse.h | 110 -------------- include/asm-x86/summit/apic.h | 185 +++++++++++++++++++++++ include/asm-x86/summit/apicdef.h | 13 ++ include/asm-x86/summit/ipi.h | 25 +++ include/asm-x86/summit/irq_vectors_limits.h | 14 ++ include/asm-x86/summit/mpparse.h | 109 +++++++++++++ 12 files changed, 351 insertions(+), 352 deletions(-) delete mode 100644 include/asm-x86/mach-summit/irq_vectors_limits.h delete mode 100644 include/asm-x86/mach-summit/mach_apic.h delete mode 100644 include/asm-x86/mach-summit/mach_apicdef.h delete mode 100644 include/asm-x86/mach-summit/mach_ipi.h delete mode 100644 include/asm-x86/mach-summit/mach_mpparse.h create mode 100644 include/asm-x86/summit/apic.h create mode 100644 include/asm-x86/summit/apicdef.h create mode 100644 include/asm-x86/summit/ipi.h create mode 100644 include/asm-x86/summit/irq_vectors_limits.h create mode 100644 include/asm-x86/summit/mpparse.h (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/summit_32.c b/arch/x86/kernel/summit_32.c index d67ce5f044b..7b987852e87 100644 --- a/arch/x86/kernel/summit_32.c +++ b/arch/x86/kernel/summit_32.c @@ -30,7 +30,7 @@ #include #include #include -#include +#include static struct rio_table_hdr *rio_table_hdr __initdata; static struct scal_detail *scal_devs[MAX_NUMNODES] __initdata; diff --git a/arch/x86/mach-generic/summit.c b/arch/x86/mach-generic/summit.c index 752edd96b1b..6ad6b67a723 100644 --- a/arch/x86/mach-generic/summit.c +++ b/arch/x86/mach-generic/summit.c @@ -11,11 +11,11 @@ #include #include #include -#include +#include #include -#include -#include -#include +#include +#include +#include static int probe_summit(void) { diff --git a/include/asm-x86/mach-summit/irq_vectors_limits.h b/include/asm-x86/mach-summit/irq_vectors_limits.h deleted file mode 100644 index 22f376ad68e..00000000000 --- a/include/asm-x86/mach-summit/irq_vectors_limits.h +++ /dev/null @@ -1,14 +0,0 @@ -#ifndef ASM_X86__MACH_SUMMIT__IRQ_VECTORS_LIMITS_H -#define ASM_X86__MACH_SUMMIT__IRQ_VECTORS_LIMITS_H - -/* - * For Summit or generic (i.e. installer) kernels, we have lots of I/O APICs, - * even with uni-proc kernels, so use a big array. - * - * This value should be the same in both the generic and summit subarches. - * Change one, change 'em both. - */ -#define NR_IRQS 224 -#define NR_IRQ_VECTORS 1024 - -#endif /* ASM_X86__MACH_SUMMIT__IRQ_VECTORS_LIMITS_H */ diff --git a/include/asm-x86/mach-summit/mach_apic.h b/include/asm-x86/mach-summit/mach_apic.h deleted file mode 100644 index 7a66758d701..00000000000 --- a/include/asm-x86/mach-summit/mach_apic.h +++ /dev/null @@ -1,185 +0,0 @@ -#ifndef ASM_X86__MACH_SUMMIT__MACH_APIC_H -#define ASM_X86__MACH_SUMMIT__MACH_APIC_H - -#include - -#define esr_disable (1) -#define NO_BALANCE_IRQ (0) - -/* In clustered mode, the high nibble of APIC ID is a cluster number. - * The low nibble is a 4-bit bitmap. */ -#define XAPIC_DEST_CPUS_SHIFT 4 -#define XAPIC_DEST_CPUS_MASK ((1u << XAPIC_DEST_CPUS_SHIFT) - 1) -#define XAPIC_DEST_CLUSTER_MASK (XAPIC_DEST_CPUS_MASK << XAPIC_DEST_CPUS_SHIFT) - -#define APIC_DFR_VALUE (APIC_DFR_CLUSTER) - -static inline cpumask_t target_cpus(void) -{ - /* CPU_MASK_ALL (0xff) has undefined behaviour with - * dest_LowestPrio mode logical clustered apic interrupt routing - * Just start on cpu 0. IRQ balancing will spread load - */ - return cpumask_of_cpu(0); -} -#define TARGET_CPUS (target_cpus()) - -#define INT_DELIVERY_MODE (dest_LowestPrio) -#define INT_DEST_MODE 1 /* logical delivery broadcast to all procs */ - -static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid) -{ - return 0; -} - -/* we don't use the phys_cpu_present_map to indicate apicid presence */ -static inline unsigned long check_apicid_present(int bit) -{ - return 1; -} - -#define apicid_cluster(apicid) ((apicid) & XAPIC_DEST_CLUSTER_MASK) - -extern u8 cpu_2_logical_apicid[]; - -static inline void init_apic_ldr(void) -{ - unsigned long val, id; - int count = 0; - u8 my_id = (u8)hard_smp_processor_id(); - u8 my_cluster = (u8)apicid_cluster(my_id); -#ifdef CONFIG_SMP - u8 lid; - int i; - - /* Create logical APIC IDs by counting CPUs already in cluster. */ - for (count = 0, i = NR_CPUS; --i >= 0; ) { - lid = cpu_2_logical_apicid[i]; - if (lid != BAD_APICID && apicid_cluster(lid) == my_cluster) - ++count; - } -#endif - /* We only have a 4 wide bitmap in cluster mode. If a deranged - * BIOS puts 5 CPUs in one APIC cluster, we're hosed. */ - BUG_ON(count >= XAPIC_DEST_CPUS_SHIFT); - id = my_cluster | (1UL << count); - apic_write(APIC_DFR, APIC_DFR_VALUE); - val = apic_read(APIC_LDR) & ~APIC_LDR_MASK; - val |= SET_APIC_LOGICAL_ID(id); - apic_write(APIC_LDR, val); -} - -static inline int multi_timer_check(int apic, int irq) -{ - return 0; -} - -static inline int apic_id_registered(void) -{ - return 1; -} - -static inline void setup_apic_routing(void) -{ - printk("Enabling APIC mode: Summit. Using %d I/O APICs\n", - nr_ioapics); -} - -static inline int apicid_to_node(int logical_apicid) -{ -#ifdef CONFIG_SMP - return apicid_2_node[hard_smp_processor_id()]; -#else - return 0; -#endif -} - -/* Mapping from cpu number to logical apicid */ -static inline int cpu_to_logical_apicid(int cpu) -{ -#ifdef CONFIG_SMP - if (cpu >= NR_CPUS) - return BAD_APICID; - return (int)cpu_2_logical_apicid[cpu]; -#else - return logical_smp_processor_id(); -#endif -} - -static inline int cpu_present_to_apicid(int mps_cpu) -{ - if (mps_cpu < NR_CPUS) - return (int)per_cpu(x86_bios_cpu_apicid, mps_cpu); - else - return BAD_APICID; -} - -static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_id_map) -{ - /* For clustered we don't have a good way to do this yet - hack */ - return physids_promote(0x0F); -} - -static inline physid_mask_t apicid_to_cpu_present(int apicid) -{ - return physid_mask_of_physid(apicid); -} - -static inline void setup_portio_remap(void) -{ -} - -static inline int check_phys_apicid_present(int boot_cpu_physical_apicid) -{ - return 1; -} - -static inline void enable_apic_mode(void) -{ -} - -static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) -{ - int num_bits_set; - int cpus_found = 0; - int cpu; - int apicid; - - num_bits_set = cpus_weight(cpumask); - /* Return id to all */ - if (num_bits_set == NR_CPUS) - return (int) 0xFF; - /* - * The cpus in the mask must all be on the apic cluster. If are not - * on the same apicid cluster return default value of TARGET_CPUS. - */ - cpu = first_cpu(cpumask); - apicid = cpu_to_logical_apicid(cpu); - while (cpus_found < num_bits_set) { - if (cpu_isset(cpu, cpumask)) { - int new_apicid = cpu_to_logical_apicid(cpu); - if (apicid_cluster(apicid) != - apicid_cluster(new_apicid)){ - printk ("%s: Not a valid mask!\n",__FUNCTION__); - return 0xFF; - } - apicid = apicid | new_apicid; - cpus_found++; - } - cpu++; - } - return apicid; -} - -/* cpuid returns the value latched in the HW at reset, not the APIC ID - * register's value. For any box whose BIOS changes APIC IDs, like - * clustered APIC systems, we must use hard_smp_processor_id. - * - * See Intel's IA-32 SW Dev's Manual Vol2 under CPUID. - */ -static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) -{ - return hard_smp_processor_id() >> index_msb; -} - -#endif /* ASM_X86__MACH_SUMMIT__MACH_APIC_H */ diff --git a/include/asm-x86/mach-summit/mach_apicdef.h b/include/asm-x86/mach-summit/mach_apicdef.h deleted file mode 100644 index d4bc8590c4f..00000000000 --- a/include/asm-x86/mach-summit/mach_apicdef.h +++ /dev/null @@ -1,13 +0,0 @@ -#ifndef ASM_X86__MACH_SUMMIT__MACH_APICDEF_H -#define ASM_X86__MACH_SUMMIT__MACH_APICDEF_H - -#define APIC_ID_MASK (0xFF<<24) - -static inline unsigned get_apic_id(unsigned long x) -{ - return (((x)>>24)&0xFF); -} - -#define GET_APIC_ID(x) get_apic_id(x) - -#endif /* ASM_X86__MACH_SUMMIT__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-summit/mach_ipi.h b/include/asm-x86/mach-summit/mach_ipi.h deleted file mode 100644 index a3b31c528d9..00000000000 --- a/include/asm-x86/mach-summit/mach_ipi.h +++ /dev/null @@ -1,25 +0,0 @@ -#ifndef ASM_X86__MACH_SUMMIT__MACH_IPI_H -#define ASM_X86__MACH_SUMMIT__MACH_IPI_H - -void send_IPI_mask_sequence(cpumask_t mask, int vector); - -static inline void send_IPI_mask(cpumask_t mask, int vector) -{ - send_IPI_mask_sequence(mask, vector); -} - -static inline void send_IPI_allbutself(int vector) -{ - cpumask_t mask = cpu_online_map; - cpu_clear(smp_processor_id(), mask); - - if (!cpus_empty(mask)) - send_IPI_mask(mask, vector); -} - -static inline void send_IPI_all(int vector) -{ - send_IPI_mask(cpu_online_map, vector); -} - -#endif /* ASM_X86__MACH_SUMMIT__MACH_IPI_H */ diff --git a/include/asm-x86/mach-summit/mach_mpparse.h b/include/asm-x86/mach-summit/mach_mpparse.h deleted file mode 100644 index 92396f28772..00000000000 --- a/include/asm-x86/mach-summit/mach_mpparse.h +++ /dev/null @@ -1,110 +0,0 @@ -#ifndef ASM_X86__MACH_SUMMIT__MACH_MPPARSE_H -#define ASM_X86__MACH_SUMMIT__MACH_MPPARSE_H - -#include -#include - -extern int use_cyclone; - -#ifdef CONFIG_X86_SUMMIT_NUMA -extern void setup_summit(void); -#else -#define setup_summit() {} -#endif - -static inline int mps_oem_check(struct mp_config_table *mpc, char *oem, - char *productid) -{ - if (!strncmp(oem, "IBM ENSW", 8) && - (!strncmp(productid, "VIGIL SMP", 9) - || !strncmp(productid, "EXA", 3) - || !strncmp(productid, "RUTHLESS SMP", 12))){ - mark_tsc_unstable("Summit based system"); - use_cyclone = 1; /*enable cyclone-timer*/ - setup_summit(); - return 1; - } - return 0; -} - -/* Hook from generic ACPI tables.c */ -static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) -{ - if (!strncmp(oem_id, "IBM", 3) && - (!strncmp(oem_table_id, "SERVIGIL", 8) - || !strncmp(oem_table_id, "EXA", 3))){ - mark_tsc_unstable("Summit based system"); - use_cyclone = 1; /*enable cyclone-timer*/ - setup_summit(); - return 1; - } - return 0; -} - -struct rio_table_hdr { - unsigned char version; /* Version number of this data structure */ - /* Version 3 adds chassis_num & WP_index */ - unsigned char num_scal_dev; /* # of Scalability devices (Twisters for Vigil) */ - unsigned char num_rio_dev; /* # of RIO I/O devices (Cyclones and Winnipegs) */ -} __attribute__((packed)); - -struct scal_detail { - unsigned char node_id; /* Scalability Node ID */ - unsigned long CBAR; /* Address of 1MB register space */ - unsigned char port0node; /* Node ID port connected to: 0xFF=None */ - unsigned char port0port; /* Port num port connected to: 0,1,2, or 0xFF=None */ - unsigned char port1node; /* Node ID port connected to: 0xFF = None */ - unsigned char port1port; /* Port num port connected to: 0,1,2, or 0xFF=None */ - unsigned char port2node; /* Node ID port connected to: 0xFF = None */ - unsigned char port2port; /* Port num port connected to: 0,1,2, or 0xFF=None */ - unsigned char chassis_num; /* 1 based Chassis number (1 = boot node) */ -} __attribute__((packed)); - -struct rio_detail { - unsigned char node_id; /* RIO Node ID */ - unsigned long BBAR; /* Address of 1MB register space */ - unsigned char type; /* Type of device */ - unsigned char owner_id; /* For WPEG: Node ID of Cyclone that owns this WPEG*/ - /* For CYC: Node ID of Twister that owns this CYC */ - unsigned char port0node; /* Node ID port connected to: 0xFF=None */ - unsigned char port0port; /* Port num port connected to: 0,1,2, or 0xFF=None */ - unsigned char port1node; /* Node ID port connected to: 0xFF=None */ - unsigned char port1port; /* Port num port connected to: 0,1,2, or 0xFF=None */ - unsigned char first_slot; /* For WPEG: Lowest slot number below this WPEG */ - /* For CYC: 0 */ - unsigned char status; /* For WPEG: Bit 0 = 1 : the XAPIC is used */ - /* = 0 : the XAPIC is not used, ie:*/ - /* ints fwded to another XAPIC */ - /* Bits1:7 Reserved */ - /* For CYC: Bits0:7 Reserved */ - unsigned char WP_index; /* For WPEG: WPEG instance index - lower ones have */ - /* lower slot numbers/PCI bus numbers */ - /* For CYC: No meaning */ - unsigned char chassis_num; /* 1 based Chassis number */ - /* For LookOut WPEGs this field indicates the */ - /* Expansion Chassis #, enumerated from Boot */ - /* Node WPEG external port, then Boot Node CYC */ - /* external port, then Next Vigil chassis WPEG */ - /* external port, etc. */ - /* Shared Lookouts have only 1 chassis number (the */ - /* first one assigned) */ -} __attribute__((packed)); - - -typedef enum { - CompatTwister = 0, /* Compatibility Twister */ - AltTwister = 1, /* Alternate Twister of internal 8-way */ - CompatCyclone = 2, /* Compatibility Cyclone */ - AltCyclone = 3, /* Alternate Cyclone of internal 8-way */ - CompatWPEG = 4, /* Compatibility WPEG */ - AltWPEG = 5, /* Second Planar WPEG */ - LookOutAWPEG = 6, /* LookOut WPEG */ - LookOutBWPEG = 7, /* LookOut WPEG */ -} node_type; - -static inline int is_WPEG(struct rio_detail *rio){ - return (rio->type == CompatWPEG || rio->type == AltWPEG || - rio->type == LookOutAWPEG || rio->type == LookOutBWPEG); -} - -#endif /* ASM_X86__MACH_SUMMIT__MACH_MPPARSE_H */ diff --git a/include/asm-x86/summit/apic.h b/include/asm-x86/summit/apic.h new file mode 100644 index 00000000000..c5b2e4b1035 --- /dev/null +++ b/include/asm-x86/summit/apic.h @@ -0,0 +1,185 @@ +#ifndef __ASM_SUMMIT_APIC_H +#define __ASM_SUMMIT_APIC_H + +#include + +#define esr_disable (1) +#define NO_BALANCE_IRQ (0) + +/* In clustered mode, the high nibble of APIC ID is a cluster number. + * The low nibble is a 4-bit bitmap. */ +#define XAPIC_DEST_CPUS_SHIFT 4 +#define XAPIC_DEST_CPUS_MASK ((1u << XAPIC_DEST_CPUS_SHIFT) - 1) +#define XAPIC_DEST_CLUSTER_MASK (XAPIC_DEST_CPUS_MASK << XAPIC_DEST_CPUS_SHIFT) + +#define APIC_DFR_VALUE (APIC_DFR_CLUSTER) + +static inline cpumask_t target_cpus(void) +{ + /* CPU_MASK_ALL (0xff) has undefined behaviour with + * dest_LowestPrio mode logical clustered apic interrupt routing + * Just start on cpu 0. IRQ balancing will spread load + */ + return cpumask_of_cpu(0); +} +#define TARGET_CPUS (target_cpus()) + +#define INT_DELIVERY_MODE (dest_LowestPrio) +#define INT_DEST_MODE 1 /* logical delivery broadcast to all procs */ + +static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid) +{ + return 0; +} + +/* we don't use the phys_cpu_present_map to indicate apicid presence */ +static inline unsigned long check_apicid_present(int bit) +{ + return 1; +} + +#define apicid_cluster(apicid) ((apicid) & XAPIC_DEST_CLUSTER_MASK) + +extern u8 cpu_2_logical_apicid[]; + +static inline void init_apic_ldr(void) +{ + unsigned long val, id; + int count = 0; + u8 my_id = (u8)hard_smp_processor_id(); + u8 my_cluster = (u8)apicid_cluster(my_id); +#ifdef CONFIG_SMP + u8 lid; + int i; + + /* Create logical APIC IDs by counting CPUs already in cluster. */ + for (count = 0, i = NR_CPUS; --i >= 0; ) { + lid = cpu_2_logical_apicid[i]; + if (lid != BAD_APICID && apicid_cluster(lid) == my_cluster) + ++count; + } +#endif + /* We only have a 4 wide bitmap in cluster mode. If a deranged + * BIOS puts 5 CPUs in one APIC cluster, we're hosed. */ + BUG_ON(count >= XAPIC_DEST_CPUS_SHIFT); + id = my_cluster | (1UL << count); + apic_write(APIC_DFR, APIC_DFR_VALUE); + val = apic_read(APIC_LDR) & ~APIC_LDR_MASK; + val |= SET_APIC_LOGICAL_ID(id); + apic_write(APIC_LDR, val); +} + +static inline int multi_timer_check(int apic, int irq) +{ + return 0; +} + +static inline int apic_id_registered(void) +{ + return 1; +} + +static inline void setup_apic_routing(void) +{ + printk("Enabling APIC mode: Summit. Using %d I/O APICs\n", + nr_ioapics); +} + +static inline int apicid_to_node(int logical_apicid) +{ +#ifdef CONFIG_SMP + return apicid_2_node[hard_smp_processor_id()]; +#else + return 0; +#endif +} + +/* Mapping from cpu number to logical apicid */ +static inline int cpu_to_logical_apicid(int cpu) +{ +#ifdef CONFIG_SMP + if (cpu >= NR_CPUS) + return BAD_APICID; + return (int)cpu_2_logical_apicid[cpu]; +#else + return logical_smp_processor_id(); +#endif +} + +static inline int cpu_present_to_apicid(int mps_cpu) +{ + if (mps_cpu < NR_CPUS) + return (int)per_cpu(x86_bios_cpu_apicid, mps_cpu); + else + return BAD_APICID; +} + +static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_id_map) +{ + /* For clustered we don't have a good way to do this yet - hack */ + return physids_promote(0x0F); +} + +static inline physid_mask_t apicid_to_cpu_present(int apicid) +{ + return physid_mask_of_physid(0); +} + +static inline void setup_portio_remap(void) +{ +} + +static inline int check_phys_apicid_present(int boot_cpu_physical_apicid) +{ + return 1; +} + +static inline void enable_apic_mode(void) +{ +} + +static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) +{ + int num_bits_set; + int cpus_found = 0; + int cpu; + int apicid; + + num_bits_set = cpus_weight(cpumask); + /* Return id to all */ + if (num_bits_set == NR_CPUS) + return (int) 0xFF; + /* + * The cpus in the mask must all be on the apic cluster. If are not + * on the same apicid cluster return default value of TARGET_CPUS. + */ + cpu = first_cpu(cpumask); + apicid = cpu_to_logical_apicid(cpu); + while (cpus_found < num_bits_set) { + if (cpu_isset(cpu, cpumask)) { + int new_apicid = cpu_to_logical_apicid(cpu); + if (apicid_cluster(apicid) != + apicid_cluster(new_apicid)){ + printk ("%s: Not a valid mask!\n",__FUNCTION__); + return 0xFF; + } + apicid = apicid | new_apicid; + cpus_found++; + } + cpu++; + } + return apicid; +} + +/* cpuid returns the value latched in the HW at reset, not the APIC ID + * register's value. For any box whose BIOS changes APIC IDs, like + * clustered APIC systems, we must use hard_smp_processor_id. + * + * See Intel's IA-32 SW Dev's Manual Vol2 under CPUID. + */ +static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) +{ + return hard_smp_processor_id() >> index_msb; +} + +#endif /* __ASM_SUMMIT_APIC_H */ diff --git a/include/asm-x86/summit/apicdef.h b/include/asm-x86/summit/apicdef.h new file mode 100644 index 00000000000..f3fbca1f61c --- /dev/null +++ b/include/asm-x86/summit/apicdef.h @@ -0,0 +1,13 @@ +#ifndef __ASM_SUMMIT_APICDEF_H +#define __ASM_SUMMIT_APICDEF_H + +#define APIC_ID_MASK (0xFF<<24) + +static inline unsigned get_apic_id(unsigned long x) +{ + return (x>>24)&0xFF; +} + +#define GET_APIC_ID(x) get_apic_id(x) + +#endif diff --git a/include/asm-x86/summit/ipi.h b/include/asm-x86/summit/ipi.h new file mode 100644 index 00000000000..53bd1e7bd7b --- /dev/null +++ b/include/asm-x86/summit/ipi.h @@ -0,0 +1,25 @@ +#ifndef __ASM_SUMMIT_IPI_H +#define __ASM_SUMMIT_IPI_H + +void send_IPI_mask_sequence(cpumask_t mask, int vector); + +static inline void send_IPI_mask(cpumask_t mask, int vector) +{ + send_IPI_mask_sequence(mask, vector); +} + +static inline void send_IPI_allbutself(int vector) +{ + cpumask_t mask = cpu_online_map; + cpu_clear(smp_processor_id(), mask); + + if (!cpus_empty(mask)) + send_IPI_mask(mask, vector); +} + +static inline void send_IPI_all(int vector) +{ + send_IPI_mask(cpu_online_map, vector); +} + +#endif /* __ASM_SUMMIT_IPI_H */ diff --git a/include/asm-x86/summit/irq_vectors_limits.h b/include/asm-x86/summit/irq_vectors_limits.h new file mode 100644 index 00000000000..890ce3f5e09 --- /dev/null +++ b/include/asm-x86/summit/irq_vectors_limits.h @@ -0,0 +1,14 @@ +#ifndef _ASM_IRQ_VECTORS_LIMITS_H +#define _ASM_IRQ_VECTORS_LIMITS_H + +/* + * For Summit or generic (i.e. installer) kernels, we have lots of I/O APICs, + * even with uni-proc kernels, so use a big array. + * + * This value should be the same in both the generic and summit subarches. + * Change one, change 'em both. + */ +#define NR_IRQS 224 +#define NR_IRQ_VECTORS 1024 + +#endif /* _ASM_IRQ_VECTORS_LIMITS_H */ diff --git a/include/asm-x86/summit/mpparse.h b/include/asm-x86/summit/mpparse.h new file mode 100644 index 00000000000..013ce6fab2d --- /dev/null +++ b/include/asm-x86/summit/mpparse.h @@ -0,0 +1,109 @@ +#ifndef __ASM_SUMMIT_MPPARSE_H +#define __ASM_SUMMIT_MPPARSE_H + +#include + +extern int use_cyclone; + +#ifdef CONFIG_X86_SUMMIT_NUMA +extern void setup_summit(void); +#else +#define setup_summit() {} +#endif + +static inline int mps_oem_check(struct mp_config_table *mpc, char *oem, + char *productid) +{ + if (!strncmp(oem, "IBM ENSW", 8) && + (!strncmp(productid, "VIGIL SMP", 9) + || !strncmp(productid, "EXA", 3) + || !strncmp(productid, "RUTHLESS SMP", 12))){ + mark_tsc_unstable("Summit based system"); + use_cyclone = 1; /*enable cyclone-timer*/ + setup_summit(); + return 1; + } + return 0; +} + +/* Hook from generic ACPI tables.c */ +static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + if (!strncmp(oem_id, "IBM", 3) && + (!strncmp(oem_table_id, "SERVIGIL", 8) + || !strncmp(oem_table_id, "EXA", 3))){ + mark_tsc_unstable("Summit based system"); + use_cyclone = 1; /*enable cyclone-timer*/ + setup_summit(); + return 1; + } + return 0; +} + +struct rio_table_hdr { + unsigned char version; /* Version number of this data structure */ + /* Version 3 adds chassis_num & WP_index */ + unsigned char num_scal_dev; /* # of Scalability devices (Twisters for Vigil) */ + unsigned char num_rio_dev; /* # of RIO I/O devices (Cyclones and Winnipegs) */ +} __attribute__((packed)); + +struct scal_detail { + unsigned char node_id; /* Scalability Node ID */ + unsigned long CBAR; /* Address of 1MB register space */ + unsigned char port0node; /* Node ID port connected to: 0xFF=None */ + unsigned char port0port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char port1node; /* Node ID port connected to: 0xFF = None */ + unsigned char port1port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char port2node; /* Node ID port connected to: 0xFF = None */ + unsigned char port2port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char chassis_num; /* 1 based Chassis number (1 = boot node) */ +} __attribute__((packed)); + +struct rio_detail { + unsigned char node_id; /* RIO Node ID */ + unsigned long BBAR; /* Address of 1MB register space */ + unsigned char type; /* Type of device */ + unsigned char owner_id; /* For WPEG: Node ID of Cyclone that owns this WPEG*/ + /* For CYC: Node ID of Twister that owns this CYC */ + unsigned char port0node; /* Node ID port connected to: 0xFF=None */ + unsigned char port0port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char port1node; /* Node ID port connected to: 0xFF=None */ + unsigned char port1port; /* Port num port connected to: 0,1,2, or 0xFF=None */ + unsigned char first_slot; /* For WPEG: Lowest slot number below this WPEG */ + /* For CYC: 0 */ + unsigned char status; /* For WPEG: Bit 0 = 1 : the XAPIC is used */ + /* = 0 : the XAPIC is not used, ie:*/ + /* ints fwded to another XAPIC */ + /* Bits1:7 Reserved */ + /* For CYC: Bits0:7 Reserved */ + unsigned char WP_index; /* For WPEG: WPEG instance index - lower ones have */ + /* lower slot numbers/PCI bus numbers */ + /* For CYC: No meaning */ + unsigned char chassis_num; /* 1 based Chassis number */ + /* For LookOut WPEGs this field indicates the */ + /* Expansion Chassis #, enumerated from Boot */ + /* Node WPEG external port, then Boot Node CYC */ + /* external port, then Next Vigil chassis WPEG */ + /* external port, etc. */ + /* Shared Lookouts have only 1 chassis number (the */ + /* first one assigned) */ +} __attribute__((packed)); + + +typedef enum { + CompatTwister = 0, /* Compatibility Twister */ + AltTwister = 1, /* Alternate Twister of internal 8-way */ + CompatCyclone = 2, /* Compatibility Cyclone */ + AltCyclone = 3, /* Alternate Cyclone of internal 8-way */ + CompatWPEG = 4, /* Compatibility WPEG */ + AltWPEG = 5, /* Second Planar WPEG */ + LookOutAWPEG = 6, /* LookOut WPEG */ + LookOutBWPEG = 7, /* LookOut WPEG */ +} node_type; + +static inline int is_WPEG(struct rio_detail *rio){ + return (rio->type == CompatWPEG || rio->type == AltWPEG || + rio->type == LookOutAWPEG || rio->type == LookOutBWPEG); +} + +#endif /* __ASM_SUMMIT_MPPARSE_H */ -- cgit v1.2.3-70-g09d2 From edb181ac4b0c7c1240503da46349a3fb69af9ef6 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 25 Jul 2008 02:17:55 -0700 Subject: x86: mach-numaq to numaq Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/mach-generic/numaq.c | 10 +-- include/asm-x86/mach-numaq/mach_apic.h | 138 ------------------------------ include/asm-x86/mach-numaq/mach_apicdef.h | 14 --- include/asm-x86/mach-numaq/mach_ipi.h | 25 ------ include/asm-x86/mach-numaq/mach_mpparse.h | 7 -- include/asm-x86/mach-numaq/mach_wakecpu.h | 43 ---------- include/asm-x86/numaq/apic.h | 138 ++++++++++++++++++++++++++++++ include/asm-x86/numaq/apicdef.h | 14 +++ include/asm-x86/numaq/ipi.h | 25 ++++++ include/asm-x86/numaq/mpparse.h | 7 ++ include/asm-x86/numaq/wakecpu.h | 43 ++++++++++ 11 files changed, 232 insertions(+), 232 deletions(-) delete mode 100644 include/asm-x86/mach-numaq/mach_apic.h delete mode 100644 include/asm-x86/mach-numaq/mach_apicdef.h delete mode 100644 include/asm-x86/mach-numaq/mach_ipi.h delete mode 100644 include/asm-x86/mach-numaq/mach_mpparse.h delete mode 100644 include/asm-x86/mach-numaq/mach_wakecpu.h create mode 100644 include/asm-x86/numaq/apic.h create mode 100644 include/asm-x86/numaq/apicdef.h create mode 100644 include/asm-x86/numaq/ipi.h create mode 100644 include/asm-x86/numaq/mpparse.h create mode 100644 include/asm-x86/numaq/wakecpu.h (limited to 'include/asm-x86') diff --git a/arch/x86/mach-generic/numaq.c b/arch/x86/mach-generic/numaq.c index 95c07efff6b..8cf58394975 100644 --- a/arch/x86/mach-generic/numaq.c +++ b/arch/x86/mach-generic/numaq.c @@ -11,12 +11,12 @@ #include #include #include -#include +#include #include -#include -#include -#include -#include +#include +#include +#include +#include #include static int mps_oem_check(struct mp_config_table *mpc, char *oem, diff --git a/include/asm-x86/mach-numaq/mach_apic.h b/include/asm-x86/mach-numaq/mach_apic.h deleted file mode 100644 index 7a0d39edfcf..00000000000 --- a/include/asm-x86/mach-numaq/mach_apic.h +++ /dev/null @@ -1,138 +0,0 @@ -#ifndef ASM_X86__MACH_NUMAQ__MACH_APIC_H -#define ASM_X86__MACH_NUMAQ__MACH_APIC_H - -#include -#include -#include - -#define APIC_DFR_VALUE (APIC_DFR_CLUSTER) - -static inline cpumask_t target_cpus(void) -{ - return CPU_MASK_ALL; -} - -#define TARGET_CPUS (target_cpus()) - -#define NO_BALANCE_IRQ (1) -#define esr_disable (1) - -#define INT_DELIVERY_MODE dest_LowestPrio -#define INT_DEST_MODE 0 /* physical delivery on LOCAL quad */ - -static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid) -{ - return physid_isset(apicid, bitmap); -} -static inline unsigned long check_apicid_present(int bit) -{ - return physid_isset(bit, phys_cpu_present_map); -} -#define apicid_cluster(apicid) (apicid & 0xF0) - -static inline int apic_id_registered(void) -{ - return 1; -} - -static inline void init_apic_ldr(void) -{ - /* Already done in NUMA-Q firmware */ -} - -static inline void setup_apic_routing(void) -{ - printk("Enabling APIC mode: %s. Using %d I/O APICs\n", - "NUMA-Q", nr_ioapics); -} - -/* - * Skip adding the timer int on secondary nodes, which causes - * a small but painful rift in the time-space continuum. - */ -static inline int multi_timer_check(int apic, int irq) -{ - return apic != 0 && irq == 0; -} - -static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map) -{ - /* We don't have a good way to do this yet - hack */ - return physids_promote(0xFUL); -} - -/* Mapping from cpu number to logical apicid */ -extern u8 cpu_2_logical_apicid[]; -static inline int cpu_to_logical_apicid(int cpu) -{ - if (cpu >= NR_CPUS) - return BAD_APICID; - return (int)cpu_2_logical_apicid[cpu]; -} - -/* - * Supporting over 60 cpus on NUMA-Q requires a locality-dependent - * cpu to APIC ID relation to properly interact with the intelligent - * mode of the cluster controller. - */ -static inline int cpu_present_to_apicid(int mps_cpu) -{ - if (mps_cpu < 60) - return ((mps_cpu >> 2) << 4) | (1 << (mps_cpu & 0x3)); - else - return BAD_APICID; -} - -static inline int apicid_to_node(int logical_apicid) -{ - return logical_apicid >> 4; -} - -static inline physid_mask_t apicid_to_cpu_present(int logical_apicid) -{ - int node = apicid_to_node(logical_apicid); - int cpu = __ffs(logical_apicid & 0xf); - - return physid_mask_of_physid(cpu + 4*node); -} - -extern void *xquad_portio; - -static inline void setup_portio_remap(void) -{ - int num_quads = num_online_nodes(); - - if (num_quads <= 1) - return; - - printk("Remapping cross-quad port I/O for %d quads\n", num_quads); - xquad_portio = ioremap(XQUAD_PORTIO_BASE, num_quads*XQUAD_PORTIO_QUAD); - printk("xquad_portio vaddr 0x%08lx, len %08lx\n", - (u_long) xquad_portio, (u_long) num_quads*XQUAD_PORTIO_QUAD); -} - -static inline int check_phys_apicid_present(int boot_cpu_physical_apicid) -{ - return (1); -} - -static inline void enable_apic_mode(void) -{ -} - -/* - * We use physical apicids here, not logical, so just return the default - * physical broadcast to stop people from breaking us - */ -static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) -{ - return (int) 0xF; -} - -/* No NUMA-Q box has a HT CPU, but it can't hurt to use the default code. */ -static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) -{ - return cpuid_apic >> index_msb; -} - -#endif /* ASM_X86__MACH_NUMAQ__MACH_APIC_H */ diff --git a/include/asm-x86/mach-numaq/mach_apicdef.h b/include/asm-x86/mach-numaq/mach_apicdef.h deleted file mode 100644 index f870ec5f778..00000000000 --- a/include/asm-x86/mach-numaq/mach_apicdef.h +++ /dev/null @@ -1,14 +0,0 @@ -#ifndef ASM_X86__MACH_NUMAQ__MACH_APICDEF_H -#define ASM_X86__MACH_NUMAQ__MACH_APICDEF_H - - -#define APIC_ID_MASK (0xF<<24) - -static inline unsigned get_apic_id(unsigned long x) -{ - return (((x)>>24)&0x0F); -} - -#define GET_APIC_ID(x) get_apic_id(x) - -#endif /* ASM_X86__MACH_NUMAQ__MACH_APICDEF_H */ diff --git a/include/asm-x86/mach-numaq/mach_ipi.h b/include/asm-x86/mach-numaq/mach_ipi.h deleted file mode 100644 index 1e835823f4b..00000000000 --- a/include/asm-x86/mach-numaq/mach_ipi.h +++ /dev/null @@ -1,25 +0,0 @@ -#ifndef ASM_X86__MACH_NUMAQ__MACH_IPI_H -#define ASM_X86__MACH_NUMAQ__MACH_IPI_H - -void send_IPI_mask_sequence(cpumask_t, int vector); - -static inline void send_IPI_mask(cpumask_t mask, int vector) -{ - send_IPI_mask_sequence(mask, vector); -} - -static inline void send_IPI_allbutself(int vector) -{ - cpumask_t mask = cpu_online_map; - cpu_clear(smp_processor_id(), mask); - - if (!cpus_empty(mask)) - send_IPI_mask(mask, vector); -} - -static inline void send_IPI_all(int vector) -{ - send_IPI_mask(cpu_online_map, vector); -} - -#endif /* ASM_X86__MACH_NUMAQ__MACH_IPI_H */ diff --git a/include/asm-x86/mach-numaq/mach_mpparse.h b/include/asm-x86/mach-numaq/mach_mpparse.h deleted file mode 100644 index 74ade184920..00000000000 --- a/include/asm-x86/mach-numaq/mach_mpparse.h +++ /dev/null @@ -1,7 +0,0 @@ -#ifndef ASM_X86__MACH_NUMAQ__MACH_MPPARSE_H -#define ASM_X86__MACH_NUMAQ__MACH_MPPARSE_H - -extern void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem, - char *productid); - -#endif /* ASM_X86__MACH_NUMAQ__MACH_MPPARSE_H */ diff --git a/include/asm-x86/mach-numaq/mach_wakecpu.h b/include/asm-x86/mach-numaq/mach_wakecpu.h deleted file mode 100644 index 0db8cea643c..00000000000 --- a/include/asm-x86/mach-numaq/mach_wakecpu.h +++ /dev/null @@ -1,43 +0,0 @@ -#ifndef ASM_X86__MACH_NUMAQ__MACH_WAKECPU_H -#define ASM_X86__MACH_NUMAQ__MACH_WAKECPU_H - -/* This file copes with machines that wakeup secondary CPUs by NMIs */ - -#define WAKE_SECONDARY_VIA_NMI - -#define TRAMPOLINE_LOW phys_to_virt(0x8) -#define TRAMPOLINE_HIGH phys_to_virt(0xa) - -#define boot_cpu_apicid boot_cpu_logical_apicid - -/* We don't do anything here because we use NMI's to boot instead */ -static inline void wait_for_init_deassert(atomic_t *deassert) -{ -} - -/* - * Because we use NMIs rather than the INIT-STARTUP sequence to - * bootstrap the CPUs, the APIC may be in a weird state. Kick it. - */ -static inline void smp_callin_clear_local_apic(void) -{ - clear_local_APIC(); -} - -static inline void store_NMI_vector(unsigned short *high, unsigned short *low) -{ - printk("Storing NMI vector\n"); - *high = *((volatile unsigned short *) TRAMPOLINE_HIGH); - *low = *((volatile unsigned short *) TRAMPOLINE_LOW); -} - -static inline void restore_NMI_vector(unsigned short *high, unsigned short *low) -{ - printk("Restoring NMI vector\n"); - *((volatile unsigned short *) TRAMPOLINE_HIGH) = *high; - *((volatile unsigned short *) TRAMPOLINE_LOW) = *low; -} - -#define inquire_remote_apic(apicid) {} - -#endif /* ASM_X86__MACH_NUMAQ__MACH_WAKECPU_H */ diff --git a/include/asm-x86/numaq/apic.h b/include/asm-x86/numaq/apic.h new file mode 100644 index 00000000000..a8344ba6ea1 --- /dev/null +++ b/include/asm-x86/numaq/apic.h @@ -0,0 +1,138 @@ +#ifndef __ASM_NUMAQ_APIC_H +#define __ASM_NUMAQ_APIC_H + +#include +#include +#include + +#define APIC_DFR_VALUE (APIC_DFR_CLUSTER) + +static inline cpumask_t target_cpus(void) +{ + return CPU_MASK_ALL; +} + +#define TARGET_CPUS (target_cpus()) + +#define NO_BALANCE_IRQ (1) +#define esr_disable (1) + +#define INT_DELIVERY_MODE dest_LowestPrio +#define INT_DEST_MODE 0 /* physical delivery on LOCAL quad */ + +static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid) +{ + return physid_isset(apicid, bitmap); +} +static inline unsigned long check_apicid_present(int bit) +{ + return physid_isset(bit, phys_cpu_present_map); +} +#define apicid_cluster(apicid) (apicid & 0xF0) + +static inline int apic_id_registered(void) +{ + return 1; +} + +static inline void init_apic_ldr(void) +{ + /* Already done in NUMA-Q firmware */ +} + +static inline void setup_apic_routing(void) +{ + printk("Enabling APIC mode: %s. Using %d I/O APICs\n", + "NUMA-Q", nr_ioapics); +} + +/* + * Skip adding the timer int on secondary nodes, which causes + * a small but painful rift in the time-space continuum. + */ +static inline int multi_timer_check(int apic, int irq) +{ + return apic != 0 && irq == 0; +} + +static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map) +{ + /* We don't have a good way to do this yet - hack */ + return physids_promote(0xFUL); +} + +/* Mapping from cpu number to logical apicid */ +extern u8 cpu_2_logical_apicid[]; +static inline int cpu_to_logical_apicid(int cpu) +{ + if (cpu >= NR_CPUS) + return BAD_APICID; + return (int)cpu_2_logical_apicid[cpu]; +} + +/* + * Supporting over 60 cpus on NUMA-Q requires a locality-dependent + * cpu to APIC ID relation to properly interact with the intelligent + * mode of the cluster controller. + */ +static inline int cpu_present_to_apicid(int mps_cpu) +{ + if (mps_cpu < 60) + return ((mps_cpu >> 2) << 4) | (1 << (mps_cpu & 0x3)); + else + return BAD_APICID; +} + +static inline int apicid_to_node(int logical_apicid) +{ + return logical_apicid >> 4; +} + +static inline physid_mask_t apicid_to_cpu_present(int logical_apicid) +{ + int node = apicid_to_node(logical_apicid); + int cpu = __ffs(logical_apicid & 0xf); + + return physid_mask_of_physid(cpu + 4*node); +} + +extern void *xquad_portio; + +static inline void setup_portio_remap(void) +{ + int num_quads = num_online_nodes(); + + if (num_quads <= 1) + return; + + printk("Remapping cross-quad port I/O for %d quads\n", num_quads); + xquad_portio = ioremap(XQUAD_PORTIO_BASE, num_quads*XQUAD_PORTIO_QUAD); + printk("xquad_portio vaddr 0x%08lx, len %08lx\n", + (u_long) xquad_portio, (u_long) num_quads*XQUAD_PORTIO_QUAD); +} + +static inline int check_phys_apicid_present(int boot_cpu_physical_apicid) +{ + return (1); +} + +static inline void enable_apic_mode(void) +{ +} + +/* + * We use physical apicids here, not logical, so just return the default + * physical broadcast to stop people from breaking us + */ +static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) +{ + return (int) 0xF; +} + +/* No NUMA-Q box has a HT CPU, but it can't hurt to use the default code. */ +static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb) +{ + return cpuid_apic >> index_msb; +} + +#endif /* __ASM_NUMAQ_APIC_H */ diff --git a/include/asm-x86/numaq/apicdef.h b/include/asm-x86/numaq/apicdef.h new file mode 100644 index 00000000000..e012a46cc22 --- /dev/null +++ b/include/asm-x86/numaq/apicdef.h @@ -0,0 +1,14 @@ +#ifndef __ASM_NUMAQ_APICDEF_H +#define __ASM_NUMAQ_APICDEF_H + + +#define APIC_ID_MASK (0xF<<24) + +static inline unsigned get_apic_id(unsigned long x) +{ + return (((x)>>24)&0x0F); +} + +#define GET_APIC_ID(x) get_apic_id(x) + +#endif diff --git a/include/asm-x86/numaq/ipi.h b/include/asm-x86/numaq/ipi.h new file mode 100644 index 00000000000..935588d286c --- /dev/null +++ b/include/asm-x86/numaq/ipi.h @@ -0,0 +1,25 @@ +#ifndef __ASM_NUMAQ_IPI_H +#define __ASM_NUMAQ_IPI_H + +void send_IPI_mask_sequence(cpumask_t, int vector); + +static inline void send_IPI_mask(cpumask_t mask, int vector) +{ + send_IPI_mask_sequence(mask, vector); +} + +static inline void send_IPI_allbutself(int vector) +{ + cpumask_t mask = cpu_online_map; + cpu_clear(smp_processor_id(), mask); + + if (!cpus_empty(mask)) + send_IPI_mask(mask, vector); +} + +static inline void send_IPI_all(int vector) +{ + send_IPI_mask(cpu_online_map, vector); +} + +#endif /* __ASM_NUMAQ_IPI_H */ diff --git a/include/asm-x86/numaq/mpparse.h b/include/asm-x86/numaq/mpparse.h new file mode 100644 index 00000000000..252292e077b --- /dev/null +++ b/include/asm-x86/numaq/mpparse.h @@ -0,0 +1,7 @@ +#ifndef __ASM_NUMAQ_MPPARSE_H +#define __ASM_NUMAQ_MPPARSE_H + +extern void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem, + char *productid); + +#endif /* __ASM_NUMAQ_MPPARSE_H */ diff --git a/include/asm-x86/numaq/wakecpu.h b/include/asm-x86/numaq/wakecpu.h new file mode 100644 index 00000000000..c577bda5b1c --- /dev/null +++ b/include/asm-x86/numaq/wakecpu.h @@ -0,0 +1,43 @@ +#ifndef __ASM_NUMAQ_WAKECPU_H +#define __ASM_NUMAQ_WAKECPU_H + +/* This file copes with machines that wakeup secondary CPUs by NMIs */ + +#define WAKE_SECONDARY_VIA_NMI + +#define TRAMPOLINE_LOW phys_to_virt(0x8) +#define TRAMPOLINE_HIGH phys_to_virt(0xa) + +#define boot_cpu_apicid boot_cpu_logical_apicid + +/* We don't do anything here because we use NMI's to boot instead */ +static inline void wait_for_init_deassert(atomic_t *deassert) +{ +} + +/* + * Because we use NMIs rather than the INIT-STARTUP sequence to + * bootstrap the CPUs, the APIC may be in a weird state. Kick it. + */ +static inline void smp_callin_clear_local_apic(void) +{ + clear_local_APIC(); +} + +static inline void store_NMI_vector(unsigned short *high, unsigned short *low) +{ + printk("Storing NMI vector\n"); + *high = *((volatile unsigned short *) TRAMPOLINE_HIGH); + *low = *((volatile unsigned short *) TRAMPOLINE_LOW); +} + +static inline void restore_NMI_vector(unsigned short *high, unsigned short *low) +{ + printk("Restoring NMI vector\n"); + *((volatile unsigned short *) TRAMPOLINE_HIGH) = *high; + *((volatile unsigned short *) TRAMPOLINE_LOW) = *low; +} + +#define inquire_remote_apic(apicid) {} + +#endif /* __ASM_NUMAQ_WAKECPU_H */ -- cgit v1.2.3-70-g09d2 From 68bd0f4ef7750fc277e1268bf40f443898382409 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 18 Apr 2008 17:08:44 -0700 Subject: x86: tracehook: asm/syscall.h Add asm/syscall.h for x86 with all the required entry points. This will allow arch-independent tracing code for system calls. Signed-off-by: Roland McGrath --- include/asm-x86/ptrace.h | 5 ++ include/asm-x86/syscall.h | 210 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 215 insertions(+) create mode 100644 include/asm-x86/syscall.h (limited to 'include/asm-x86') diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h index 8a71db803da..91a77f5c467 100644 --- a/include/asm-x86/ptrace.h +++ b/include/asm-x86/ptrace.h @@ -213,6 +213,11 @@ static inline unsigned long frame_pointer(struct pt_regs *regs) return regs->bp; } +static inline unsigned long user_stack_pointer(struct pt_regs *regs) +{ + return regs->sp; +} + /* * These are defined as per linux/ptrace.h, which see. */ diff --git a/include/asm-x86/syscall.h b/include/asm-x86/syscall.h new file mode 100644 index 00000000000..6f293892895 --- /dev/null +++ b/include/asm-x86/syscall.h @@ -0,0 +1,210 @@ +/* + * Access to user system call parameters and results + * + * Copyright (C) 2008 Red Hat, Inc. All rights reserved. + * + * This copyrighted material is made available to anyone wishing to use, + * modify, copy, or redistribute it subject to the terms and conditions + * of the GNU General Public License v.2. + * + * See asm-generic/syscall.h for descriptions of what we must do here. + */ + +#ifndef _ASM_SYSCALL_H +#define _ASM_SYSCALL_H 1 + +#include + +static inline long syscall_get_nr(struct task_struct *task, + struct pt_regs *regs) +{ + /* + * We always sign-extend a -1 value being set here, + * so this is always either -1L or a syscall number. + */ + return regs->orig_ax; +} + +static inline void syscall_rollback(struct task_struct *task, + struct pt_regs *regs) +{ + regs->ax = regs->orig_ax; +} + +static inline long syscall_get_error(struct task_struct *task, + struct pt_regs *regs) +{ + unsigned long error = regs->ax; +#ifdef CONFIG_IA32_EMULATION + /* + * TS_COMPAT is set for 32-bit syscall entries and then + * remains set until we return to user mode. + */ + if (task_thread_info(task)->status & TS_COMPAT) + /* + * Sign-extend the value so (int)-EFOO becomes (long)-EFOO + * and will match correctly in comparisons. + */ + error = (long) (int) error; +#endif + return error >= -4095L ? error : 0; +} + +static inline long syscall_get_return_value(struct task_struct *task, + struct pt_regs *regs) +{ + return regs->ax; +} + +static inline void syscall_set_return_value(struct task_struct *task, + struct pt_regs *regs, + int error, long val) +{ + regs->ax = (long) error ?: val; +} + +#ifdef CONFIG_X86_32 + +static inline void syscall_get_arguments(struct task_struct *task, + struct pt_regs *regs, + unsigned int i, unsigned int n, + unsigned long *args) +{ + BUG_ON(i + n > 6); + memcpy(args, ®s->bx + i, n * sizeof(args[0])); +} + +static inline void syscall_set_arguments(struct task_struct *task, + struct pt_regs *regs, + unsigned int i, unsigned int n, + const unsigned long *args) +{ + BUG_ON(i + n > 6); + memcpy(®s->bx + i, args, n * sizeof(args[0])); +} + +#else /* CONFIG_X86_64 */ + +static inline void syscall_get_arguments(struct task_struct *task, + struct pt_regs *regs, + unsigned int i, unsigned int n, + unsigned long *args) +{ +# ifdef CONFIG_IA32_EMULATION + if (task_thread_info(task)->status & TS_COMPAT) + switch (i + n) { + case 6: + if (!n--) break; + *args++ = regs->bp; + case 5: + if (!n--) break; + *args++ = regs->di; + case 4: + if (!n--) break; + *args++ = regs->si; + case 3: + if (!n--) break; + *args++ = regs->dx; + case 2: + if (!n--) break; + *args++ = regs->cx; + case 1: + if (!n--) break; + *args++ = regs->bx; + case 0: + if (!n--) break; + default: + BUG(); + break; + } + else +# endif + switch (i + n) { + case 6: + if (!n--) break; + *args++ = regs->r9; + case 5: + if (!n--) break; + *args++ = regs->r8; + case 4: + if (!n--) break; + *args++ = regs->r10; + case 3: + if (!n--) break; + *args++ = regs->dx; + case 2: + if (!n--) break; + *args++ = regs->si; + case 1: + if (!n--) break; + *args++ = regs->di; + case 0: + if (!n--) break; + default: + BUG(); + break; + } +} + +static inline void syscall_set_arguments(struct task_struct *task, + struct pt_regs *regs, + unsigned int i, unsigned int n, + const unsigned long *args) +{ +# ifdef CONFIG_IA32_EMULATION + if (task_thread_info(task)->status & TS_COMPAT) + switch (i + n) { + case 6: + if (!n--) break; + regs->bp = *args++; + case 5: + if (!n--) break; + regs->di = *args++; + case 4: + if (!n--) break; + regs->si = *args++; + case 3: + if (!n--) break; + regs->dx = *args++; + case 2: + if (!n--) break; + regs->cx = *args++; + case 1: + if (!n--) break; + regs->bx = *args++; + case 0: + if (!n--) break; + default: + BUG(); + } + else +# endif + switch (i + n) { + case 6: + if (!n--) break; + regs->r9 = *args++; + case 5: + if (!n--) break; + regs->r8 = *args++; + case 4: + if (!n--) break; + regs->r10 = *args++; + case 3: + if (!n--) break; + regs->dx = *args++; + case 2: + if (!n--) break; + regs->si = *args++; + case 1: + if (!n--) break; + regs->di = *args++; + case 0: + if (!n--) break; + default: + BUG(); + } +} + +#endif /* CONFIG_X86_32 */ + +#endif /* _ASM_SYSCALL_H */ -- cgit v1.2.3-70-g09d2 From 59e52130f04537d2c80ea44bb007cadd1ad29543 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Sat, 19 Apr 2008 19:10:57 -0700 Subject: x86: tracehook: TIF_NOTIFY_RESUME This adds TIF_NOTIFY_RESUME support for x86, both 64-bit and 32-bit. When set, we call tracehook_notify_resume() on the way to user mode. Signed-off-by: Roland McGrath --- arch/x86/kernel/signal_32.c | 5 +++++ arch/x86/kernel/signal_64.c | 5 +++++ include/asm-x86/thread_info.h | 4 +++- 3 files changed, 13 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c index 22aae1683c1..4445d26efd4 100644 --- a/arch/x86/kernel/signal_32.c +++ b/arch/x86/kernel/signal_32.c @@ -663,5 +663,10 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) if (thread_info_flags & _TIF_SIGPENDING) do_signal(regs); + if (thread_info_flags & _TIF_NOTIFY_RESUME) { + clear_thread_flag(TIF_NOTIFY_RESUME); + tracehook_notify_resume(regs); + } + clear_thread_flag(TIF_IRET); } diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c index cb7cf0216ab..d01e3f6ef26 100644 --- a/arch/x86/kernel/signal_64.c +++ b/arch/x86/kernel/signal_64.c @@ -523,6 +523,11 @@ void do_notify_resume(struct pt_regs *regs, void *unused, /* deal with pending signal delivery */ if (thread_info_flags & _TIF_SIGPENDING) do_signal(regs); + + if (thread_info_flags & _TIF_NOTIFY_RESUME) { + clear_thread_flag(TIF_NOTIFY_RESUME); + tracehook_notify_resume(regs); + } } void signal_fault(struct pt_regs *regs, void __user *frame, char *where) diff --git a/include/asm-x86/thread_info.h b/include/asm-x86/thread_info.h index da0a675adf9..4cd5b7bdc8c 100644 --- a/include/asm-x86/thread_info.h +++ b/include/asm-x86/thread_info.h @@ -71,6 +71,7 @@ struct thread_info { * Warning: layout of LSW is hardcoded in entry.S */ #define TIF_SYSCALL_TRACE 0 /* syscall trace active */ +#define TIF_NOTIFY_RESUME 1 /* callback before returning to user */ #define TIF_SIGPENDING 2 /* signal pending */ #define TIF_NEED_RESCHED 3 /* rescheduling necessary */ #define TIF_SINGLESTEP 4 /* reenable singlestep on user return*/ @@ -93,6 +94,7 @@ struct thread_info { #define TIF_BTS_TRACE_TS 27 /* record scheduling event timestamps */ #define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE) +#define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME) #define _TIF_SIGPENDING (1 << TIF_SIGPENDING) #define _TIF_SINGLESTEP (1 << TIF_SINGLESTEP) #define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED) @@ -133,7 +135,7 @@ struct thread_info { /* Only used for 64 bit */ #define _TIF_DO_NOTIFY_MASK \ - (_TIF_SIGPENDING|_TIF_MCE_NOTIFY) + (_TIF_SIGPENDING|_TIF_MCE_NOTIFY|_TIF_NOTIFY_RESUME) /* flags to check in __switch_to() */ #define _TIF_WORK_CTXSW \ -- cgit v1.2.3-70-g09d2 From 64f53a0492b4bc11868307990bb8f7c1e0764f89 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Sun, 27 Jul 2008 08:42:32 -0700 Subject: x86: fix initialization of 'l' bit in ldt descriptors Make sure that fill_ldt() initializes the 'l' bit in the descriptor. It always sets it to 0, ignoring 'lm' in user_desc, preserving original x86_64 behaviour. Previously it was leaving 'l' uninitialized. Signed-off-by: Jeremy Fitzhardinge Cc: Glauber de Oliveira Costa Signed-off-by: Ingo Molnar Cc: Glauber de Oliveira Costa --- include/asm-x86/desc.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h index 24a524f5e1a..06f786f4b4f 100644 --- a/include/asm-x86/desc.h +++ b/include/asm-x86/desc.h @@ -24,6 +24,11 @@ static inline void fill_ldt(struct desc_struct *desc, desc->d = info->seg_32bit; desc->g = info->limit_in_pages; desc->base2 = (info->base_addr & 0xff000000) >> 24; + /* + * Don't allow setting of the lm bit. It is useless anyway + * because 64bit system calls require __USER_CS: + */ + desc->l = 0; } extern struct desc_ptr idt_descr; -- cgit v1.2.3-70-g09d2 From 9a56a0f80b52cb41c5e0add47c7ce0bb2ef25eb0 Mon Sep 17 00:00:00 2001 From: Peter Oruba Date: Mon, 28 Jul 2008 18:44:13 +0200 Subject: x86: moved Intel microcode patch loader declarations to seperate header file Intel specific microcode declarations have been moved to a seperate header file. There are no code changes to the code itself and no side effects to other parts. Signed-off-by: Peter Oruba Cc: Tigran Aivazian Signed-off-by: Ingo Molnar --- arch/x86/kernel/microcode.c | 1 + include/asm-x86/microcode.h | 34 ++++++++++++++++++++++++++++++++++ include/asm-x86/processor.h | 35 ----------------------------------- 3 files changed, 35 insertions(+), 35 deletions(-) create mode 100644 include/asm-x86/microcode.h (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c index 6994c751590..0d654bd3292 100644 --- a/arch/x86/kernel/microcode.c +++ b/arch/x86/kernel/microcode.c @@ -93,6 +93,7 @@ #include #include #include +#include MODULE_DESCRIPTION("Intel CPU (IA-32) Microcode Update Driver"); MODULE_AUTHOR("Tigran Aivazian "); diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h new file mode 100644 index 00000000000..5a055685515 --- /dev/null +++ b/include/asm-x86/microcode.h @@ -0,0 +1,34 @@ +struct microcode_header { + unsigned int hdrver; + unsigned int rev; + unsigned int date; + unsigned int sig; + unsigned int cksum; + unsigned int ldrver; + unsigned int pf; + unsigned int datasize; + unsigned int totalsize; + unsigned int reserved[3]; +}; + +struct microcode { + struct microcode_header hdr; + unsigned int bits[0]; +}; + +typedef struct microcode microcode_t; +typedef struct microcode_header microcode_header_t; + +/* microcode format is extended from prescott processors */ +struct extended_signature { + unsigned int sig; + unsigned int pf; + unsigned int cksum; +}; + +struct extended_sigtable { + unsigned int count; + unsigned int cksum; + unsigned int reserved[3]; + struct extended_signature sigs[0]; +}; diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index 5f58da401b4..58a76f69ee3 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -561,41 +561,6 @@ static inline void clear_in_cr4(unsigned long mask) write_cr4(cr4); } -struct microcode_header { - unsigned int hdrver; - unsigned int rev; - unsigned int date; - unsigned int sig; - unsigned int cksum; - unsigned int ldrver; - unsigned int pf; - unsigned int datasize; - unsigned int totalsize; - unsigned int reserved[3]; -}; - -struct microcode { - struct microcode_header hdr; - unsigned int bits[0]; -}; - -typedef struct microcode microcode_t; -typedef struct microcode_header microcode_header_t; - -/* microcode format is extended from prescott processors */ -struct extended_signature { - unsigned int sig; - unsigned int pf; - unsigned int cksum; -}; - -struct extended_sigtable { - unsigned int count; - unsigned int cksum; - unsigned int reserved[3]; - struct extended_signature sigs[0]; -}; - typedef struct { unsigned long seg; } mm_segment_t; -- cgit v1.2.3-70-g09d2 From 8e61028dfdc6b8ca996abfe8f9baef6792ea2904 Mon Sep 17 00:00:00 2001 From: Peter Oruba Date: Mon, 28 Jul 2008 18:44:14 +0200 Subject: x86: typedef removal Removed typedefs. No functional changes to the code. Signed-off-by: Peter Oruba Cc: Tigran Aivazian Signed-off-by: Ingo Molnar --- arch/x86/kernel/microcode.c | 24 ++++++++++++------------ include/asm-x86/microcode.h | 3 --- 2 files changed, 12 insertions(+), 15 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c index 0d654bd3292..74e6a77bf19 100644 --- a/arch/x86/kernel/microcode.c +++ b/arch/x86/kernel/microcode.c @@ -102,17 +102,17 @@ MODULE_LICENSE("GPL"); #define MICROCODE_VERSION "1.14a" #define DEFAULT_UCODE_DATASIZE (2000) /* 2000 bytes */ -#define MC_HEADER_SIZE (sizeof (microcode_header_t)) /* 48 bytes */ +#define MC_HEADER_SIZE (sizeof (struct microcode_header)) /* 48 bytes */ #define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE) /* 2048 bytes */ #define EXT_HEADER_SIZE (sizeof (struct extended_sigtable)) /* 20 bytes */ #define EXT_SIGNATURE_SIZE (sizeof (struct extended_signature)) /* 12 bytes */ #define DWSIZE (sizeof (u32)) #define get_totalsize(mc) \ - (((microcode_t *)mc)->hdr.totalsize ? \ - ((microcode_t *)mc)->hdr.totalsize : DEFAULT_UCODE_TOTALSIZE) + (((struct microcode *)mc)->hdr.totalsize ? \ + ((struct microcode *)mc)->hdr.totalsize : DEFAULT_UCODE_TOTALSIZE) #define get_datasize(mc) \ - (((microcode_t *)mc)->hdr.datasize ? \ - ((microcode_t *)mc)->hdr.datasize : DEFAULT_UCODE_DATASIZE) + (((struct microcode *)mc)->hdr.datasize ? \ + ((struct microcode *)mc)->hdr.datasize : DEFAULT_UCODE_DATASIZE) #define sigmatch(s1, s2, p1, p2) \ (((s1) == (s2)) && (((p1) & (p2)) || (((p1) == 0) && ((p2) == 0)))) @@ -130,7 +130,7 @@ static struct ucode_cpu_info { unsigned int sig; unsigned int pf; unsigned int rev; - microcode_t *mc; + struct microcode *mc; } ucode_cpu_info[NR_CPUS]; static void collect_cpu_info(int cpu_num) @@ -171,7 +171,7 @@ static void collect_cpu_info(int cpu_num) } static inline int microcode_update_match(int cpu_num, - microcode_header_t *mc_header, int sig, int pf) + struct microcode_header *mc_header, int sig, int pf) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; @@ -183,7 +183,7 @@ static inline int microcode_update_match(int cpu_num, static int microcode_sanity_check(void *mc) { - microcode_header_t *mc_header = mc; + struct microcode_header *mc_header = mc; struct extended_sigtable *ext_header = NULL; struct extended_signature *ext_sig; unsigned long total_size, data_size, ext_table_size; @@ -268,7 +268,7 @@ static int microcode_sanity_check(void *mc) static int get_maching_microcode(void *mc, int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - microcode_header_t *mc_header = mc; + struct microcode_header *mc_header = mc; struct extended_sigtable *ext_header; unsigned long total_size = get_totalsize(mc_header); int ext_sigcount, i; @@ -355,7 +355,7 @@ static unsigned int user_buffer_size; /* it's size */ static long get_next_ucode(void **mc, long offset) { - microcode_header_t mc_header; + struct microcode_header mc_header; unsigned long total_size; /* No more data */ @@ -497,13 +497,13 @@ MODULE_ALIAS_MISCDEV(MICROCODE_MINOR); static long get_next_ucode_from_buffer(void **mc, const u8 *buf, unsigned long size, long offset) { - microcode_header_t *mc_header; + struct microcode_header *mc_header; unsigned long total_size; /* No more data */ if (offset >= size) return 0; - mc_header = (microcode_header_t *)(buf + offset); + mc_header = (struct microcode_header *)(buf + offset); total_size = get_totalsize(mc_header); if (offset + total_size > size) { diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index 5a055685515..1519ef0674b 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -16,9 +16,6 @@ struct microcode { unsigned int bits[0]; }; -typedef struct microcode microcode_t; -typedef struct microcode_header microcode_header_t; - /* microcode format is extended from prescott processors */ struct extended_signature { unsigned int sig; -- cgit v1.2.3-70-g09d2 From c3b71bcec0380836caac9b524fa1585b469b7456 Mon Sep 17 00:00:00 2001 From: Peter Oruba Date: Mon, 28 Jul 2008 18:44:15 +0200 Subject: x86: move per CPU microcode structure declaration to header file This structure will be later used by other modules as well and needs therfore to be moved out to a header file. Signed-off-by: Peter Oruba Cc: Tigran Aivazian Signed-off-by: Ingo Molnar --- arch/x86/kernel/microcode.c | 8 +------- include/asm-x86/microcode.h | 8 ++++++++ 2 files changed, 9 insertions(+), 7 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c index 74e6a77bf19..4e7b2f65fed 100644 --- a/arch/x86/kernel/microcode.c +++ b/arch/x86/kernel/microcode.c @@ -125,13 +125,7 @@ static DEFINE_SPINLOCK(microcode_update_lock); /* no concurrent ->write()s are allowed on /dev/cpu/microcode */ static DEFINE_MUTEX(microcode_mutex); -static struct ucode_cpu_info { - int valid; - unsigned int sig; - unsigned int pf; - unsigned int rev; - struct microcode *mc; -} ucode_cpu_info[NR_CPUS]; +static struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; static void collect_cpu_info(int cpu_num) { diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index 1519ef0674b..d34a1fc1b17 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -29,3 +29,11 @@ struct extended_sigtable { unsigned int reserved[3]; struct extended_signature sigs[0]; }; + +struct ucode_cpu_info { + int valid; + unsigned int sig; + unsigned int pf; + unsigned int rev; + struct microcode *mc; +}; -- cgit v1.2.3-70-g09d2 From d4ee36686853d5714437c4409f17ad42bfaf4211 Mon Sep 17 00:00:00 2001 From: Peter Oruba Date: Mon, 28 Jul 2008 18:44:18 +0200 Subject: x86: structure declaration renaming Renamed common structures to vendor specific naming scheme so other vendors will be able to use the same naming convention. Signed-off-by: Peter Oruba Cc: Tigran Aivazian Signed-off-by: Ingo Molnar --- arch/x86/kernel/microcode_intel.c | 46 ++++++++++++++++++++------------------- include/asm-x86/microcode.h | 10 +++++---- 2 files changed, 30 insertions(+), 26 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index eded0a154ea..ca9861bf067 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -100,17 +100,19 @@ MODULE_AUTHOR("Tigran Aivazian "); MODULE_LICENSE("GPL"); #define DEFAULT_UCODE_DATASIZE (2000) /* 2000 bytes */ -#define MC_HEADER_SIZE (sizeof(struct microcode_header)) /* 48 bytes */ +#define MC_HEADER_SIZE (sizeof(struct microcode_header_intel)) /* 48 bytes */ #define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE) /* 2048 bytes */ #define EXT_HEADER_SIZE (sizeof(struct extended_sigtable)) /* 20 bytes */ #define EXT_SIGNATURE_SIZE (sizeof(struct extended_signature)) /* 12 bytes */ #define DWSIZE (sizeof(u32)) #define get_totalsize(mc) \ - (((struct microcode *)mc)->hdr.totalsize ? \ - ((struct microcode *)mc)->hdr.totalsize : DEFAULT_UCODE_TOTALSIZE) + (((struct microcode_intel *)mc)->hdr.totalsize ? \ + ((struct microcode_intel *)mc)->hdr.totalsize : \ + DEFAULT_UCODE_TOTALSIZE) + #define get_datasize(mc) \ - (((struct microcode *)mc)->hdr.datasize ? \ - ((struct microcode *)mc)->hdr.datasize : DEFAULT_UCODE_DATASIZE) + (((struct microcode_intel *)mc)->hdr.datasize ? \ + ((struct microcode_intel *)mc)->hdr.datasize : DEFAULT_UCODE_DATASIZE) #define sigmatch(s1, s2, p1, p2) \ (((s1) == (s2)) && (((p1) & (p2)) || (((p1) == 0) && ((p2) == 0)))) @@ -134,7 +136,7 @@ void collect_cpu_info(int cpu_num) /* We should bind the task to the CPU */ BUG_ON(raw_smp_processor_id() != cpu_num); uci->pf = uci->rev = 0; - uci->mc = NULL; + uci->mc.mc_intel = NULL; uci->valid = 1; if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 || @@ -163,7 +165,7 @@ void collect_cpu_info(int cpu_num) } static inline int microcode_update_match(int cpu_num, - struct microcode_header *mc_header, int sig, int pf) + struct microcode_header_intel *mc_header, int sig, int pf) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; @@ -175,7 +177,7 @@ static inline int microcode_update_match(int cpu_num, int microcode_sanity_check(void *mc) { - struct microcode_header *mc_header = mc; + struct microcode_header_intel *mc_header = mc; struct extended_sigtable *ext_header = NULL; struct extended_signature *ext_sig; unsigned long total_size, data_size, ext_table_size; @@ -260,7 +262,7 @@ int microcode_sanity_check(void *mc) int get_matching_microcode(void *mc, int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - struct microcode_header *mc_header = mc; + struct microcode_header_intel *mc_header = mc; struct extended_sigtable *ext_header; unsigned long total_size = get_totalsize(mc_header); int ext_sigcount, i; @@ -294,10 +296,10 @@ find: } /* free previous update file */ - vfree(uci->mc); + vfree(uci->mc.mc_intel); memcpy(new_mc, mc, total_size); - uci->mc = new_mc; + uci->mc.mc_intel = new_mc; return 1; } @@ -311,7 +313,7 @@ void apply_microcode(int cpu) /* We should bind the task to the CPU */ BUG_ON(cpu_num != cpu); - if (uci->mc == NULL) + if (uci->mc.mc_intel == NULL) return; /* serialize access to the physical write to MSR 0x79 */ @@ -319,8 +321,8 @@ void apply_microcode(int cpu) /* write microcode via MSR 0x79 */ wrmsr(MSR_IA32_UCODE_WRITE, - (unsigned long) uci->mc->bits, - (unsigned long) uci->mc->bits >> 16 >> 16); + (unsigned long) uci->mc.mc_intel->bits, + (unsigned long) uci->mc.mc_intel->bits >> 16 >> 16); wrmsr(MSR_IA32_UCODE_REV, 0, 0); /* see notes above for revision 1.07. Apparent chip bug */ @@ -330,14 +332,14 @@ void apply_microcode(int cpu) rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]); spin_unlock_irqrestore(µcode_update_lock, flags); - if (val[1] != uci->mc->hdr.rev) { + if (val[1] != uci->mc.mc_intel->hdr.rev) { printk(KERN_ERR "microcode: CPU%d update from revision " "0x%x to 0x%x failed\n", cpu_num, uci->rev, val[1]); return; } printk(KERN_INFO "microcode: CPU%d updated from revision " "0x%x to 0x%x, date = %08x \n", - cpu_num, uci->rev, val[1], uci->mc->hdr.date); + cpu_num, uci->rev, val[1], uci->mc.mc_intel->hdr.date); uci->rev = val[1]; } @@ -347,7 +349,7 @@ extern unsigned int user_buffer_size; /* it's size */ long get_next_ucode(void **mc, long offset) { - struct microcode_header mc_header; + struct microcode_header_intel mc_header; unsigned long total_size; /* No more data */ @@ -378,13 +380,13 @@ long get_next_ucode(void **mc, long offset) static long get_next_ucode_from_buffer(void **mc, const u8 *buf, unsigned long size, long offset) { - struct microcode_header *mc_header; + struct microcode_header_intel *mc_header; unsigned long total_size; /* No more data */ if (offset >= size) return 0; - mc_header = (struct microcode_header *)(buf + offset); + mc_header = (struct microcode_header_intel *)(buf + offset); total_size = get_totalsize(mc_header); if (offset + total_size > size) { @@ -463,7 +465,7 @@ int apply_microcode_check_cpu(int cpu) int err = 0; /* Check if the microcode is available */ - if (!uci->mc) + if (!uci->mc.mc_intel) return 0; old = current->cpus_allowed; @@ -508,7 +510,7 @@ void microcode_fini_cpu(int cpu) mutex_lock(µcode_mutex); uci->valid = 0; - kfree(uci->mc); - uci->mc = NULL; + kfree(uci->mc.mc_intel); + uci->mc.mc_intel = NULL; mutex_unlock(µcode_mutex); } diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index d34a1fc1b17..ef77c6f438b 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -1,4 +1,4 @@ -struct microcode_header { +struct microcode_header_intel { unsigned int hdrver; unsigned int rev; unsigned int date; @@ -11,8 +11,8 @@ struct microcode_header { unsigned int reserved[3]; }; -struct microcode { - struct microcode_header hdr; +struct microcode_intel { + struct microcode_header_intel hdr; unsigned int bits[0]; }; @@ -35,5 +35,7 @@ struct ucode_cpu_info { unsigned int sig; unsigned int pf; unsigned int rev; - struct microcode *mc; + union { + struct microcode_intel *mc_intel; + } mc; }; -- cgit v1.2.3-70-g09d2 From 9835fd4ad9ee5fc6b909df72aa3e3dba04415f4b Mon Sep 17 00:00:00 2001 From: Peter Oruba Date: Mon, 28 Jul 2008 18:44:19 +0200 Subject: x86: add AMD specific declarations Added AMD specific declarations to header file. Signed-off-by: Peter Oruba Cc: Tigran Aivazian Signed-off-by: Ingo Molnar --- include/asm-x86/microcode.h | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index ef77c6f438b..4e941721c0d 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -30,6 +30,35 @@ struct extended_sigtable { struct extended_signature sigs[0]; }; +struct equiv_cpu_entry { + unsigned int installed_cpu; + unsigned int fixed_errata_mask; + unsigned int fixed_errata_compare; + unsigned int equiv_cpu; +}; + +struct microcode_header_amd { + unsigned int data_code; + unsigned int patch_id; + unsigned char mc_patch_data_id[2]; + unsigned char mc_patch_data_len; + unsigned char init_flag; + unsigned int mc_patch_data_checksum; + unsigned int nb_dev_id; + unsigned int sb_dev_id; + unsigned char processor_rev_id[2]; + unsigned char nb_rev_id; + unsigned char sb_rev_id; + unsigned char bios_api_rev; + unsigned char reserved1[3]; + unsigned int match_reg[8]; +}; + +struct microcode_amd { + struct microcode_header_amd hdr; + unsigned int mpb[0]; +}; + struct ucode_cpu_info { int valid; unsigned int sig; @@ -37,5 +66,6 @@ struct ucode_cpu_info { unsigned int rev; union { struct microcode_intel *mc_intel; + struct microcode_amd *mc_amd; } mc; }; -- cgit v1.2.3-70-g09d2 From 26bf7a48c33906cc3415a4492aa9ead7a75f1353 Mon Sep 17 00:00:00 2001 From: Peter Oruba Date: Mon, 28 Jul 2008 18:44:20 +0200 Subject: x86: first step of refactoring, introducing microcode_ops Refactoring with the goal of having one general module and separate vendor specific modules that hook into the general one. Microcode_ops is a function pointer structure in which vendor specific modules will enter all functions that differ between vendors and that need to be accessed from the general module. Signed-off-by: Peter Oruba Cc: Tigran Aivazian Signed-off-by: Ingo Molnar --- include/asm-x86/microcode.h | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index 4e941721c0d..9231c876e37 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -1,3 +1,16 @@ +struct microcode_ops { + long (*get_next_ucode)(void **mc, long offset); + long (*microcode_get_next_ucode)(void **mc, long offset); + int (*get_matching_microcode)(void *mc, int cpu); + int (*apply_microcode_check_cpu)(int cpu); + int (*microcode_sanity_check)(void *mc); + int (*cpu_request_microcode)(int cpu); + void (*collect_cpu_info)(int cpu_num); + void (*apply_microcode)(int cpu); + void (*microcode_fini_cpu)(int cpu); + void (*clear_patch)(void *data); +}; + struct microcode_header_intel { unsigned int hdrver; unsigned int rev; -- cgit v1.2.3-70-g09d2 From 8d86f390d9bb5b39f0a315838d1616de6363e1b9 Mon Sep 17 00:00:00 2001 From: Peter Oruba Date: Mon, 28 Jul 2008 18:44:21 +0200 Subject: x86: major refactoring Refactored code by introducing a two-module solution. There is one general module in which vendor specific modules can hook into. However, that is exclusive, there is only one vendor specific module allowed at a time. A CPU vendor check makes sure only the correct module for the underlying system gets called. Functinally in terms of patch loading itself there are no changes. This refactoring provides a basis for future implementations of other vendors' patch loaders. Signed-off-by: Peter Oruba Cc: Tigran Aivazian Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 25 +++++++++--- arch/x86/kernel/Makefile | 4 +- arch/x86/kernel/microcode.c | 80 +++++++++++++++++++++++---------------- arch/x86/kernel/microcode_intel.c | 50 +++++++++++++++++++----- include/asm-x86/microcode.h | 3 ++ 5 files changed, 112 insertions(+), 50 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index b6fa2877b17..6b0b885cf47 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -782,7 +782,7 @@ config X86_REBOOTFIXUPS Say N otherwise. config MICROCODE - tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support" + tristate "/dev/cpu/microcode - microcode support" select FW_LOADER ---help--- If you say Y here, you will be able to update the microcode on @@ -791,14 +791,29 @@ config MICROCODE actual microcode binary data itself which is not shipped with the Linux kernel. - For latest news and information on obtaining all the required - ingredients for this driver, check: - . + This option selects the general module only, you need to select + at least one vendor specific module as well. To compile this driver as a module, choose M here: the module will be called microcode. -config MICROCODE_OLD_INTERFACE +config MICROCODE_INTEL + tristate "Intel microcode patch loading support" + depends on MICROCODE + default MICROCODE + select FW_LOADER + --help--- + This options enables microcode patch loading support for Intel + processors. + + For latest news and information on obtaining all the required + Intel ingredients for this driver, check: + . + + This driver is only available as a module: the module + will be called microcode_intel. + + config MICROCODE_OLD_INTERFACE def_bool y depends on MICROCODE diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index abb32aed7f4..f2f9f6da8c1 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -51,8 +51,8 @@ obj-$(CONFIG_X86_BIOS_REBOOT) += reboot.o obj-$(CONFIG_MCA) += mca_32.o obj-$(CONFIG_X86_MSR) += msr.o obj-$(CONFIG_X86_CPUID) += cpuid.o -obj-$(CONFIG_MICROCODE) += ucode.o -ucode-objs := microcode.o microcode_intel.o +obj-$(CONFIG_MICROCODE) += microcode.o +obj-$(CONFIG_MICROCODE_INTEL) += microcode_intel.o obj-$(CONFIG_PCI) += early-quirks.o apm-y := apm_32.o obj-$(CONFIG_APM) += apm.o diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c index c1047d7f7ed..1e42e79ca69 100644 --- a/arch/x86/kernel/microcode.c +++ b/arch/x86/kernel/microcode.c @@ -99,25 +99,22 @@ MODULE_DESCRIPTION("Microcode Update Driver"); MODULE_AUTHOR("Tigran Aivazian "); MODULE_LICENSE("GPL"); -#define MICROCODE_VERSION "1.14a" +#define MICROCODE_VERSION "2.00" -/* no concurrent ->write()s are allowed on /dev/cpu/microcode */ -DEFINE_MUTEX(microcode_mutex); +struct microcode_ops *microcode_ops; -struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; +/* no concurrent ->write()s are allowed on /dev/cpu/microcode */ +static DEFINE_MUTEX(microcode_mutex); +EXPORT_SYMBOL_GPL(microcode_mutex); -extern long get_next_ucode(void **mc, long offset); -extern int microcode_sanity_check(void *mc); -extern int get_matching_microcode(void *mc, int cpu); -extern void collect_cpu_info(int cpu_num); -extern int cpu_request_microcode(int cpu); -extern void microcode_fini_cpu(int cpu); -extern void apply_microcode(int cpu); -extern int apply_microcode_check_cpu(int cpu); +static struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; +EXPORT_SYMBOL_GPL(ucode_cpu_info); #ifdef CONFIG_MICROCODE_OLD_INTERFACE -void __user *user_buffer; /* user area microcode data buffer */ -unsigned int user_buffer_size; /* it's size */ +static void __user *user_buffer; /* user area microcode data buffer */ +EXPORT_SYMBOL_GPL(user_buffer); +static unsigned int user_buffer_size; /* it's size */ +EXPORT_SYMBOL_GPL(user_buffer_size); static int do_microcode_update (void) { @@ -130,8 +127,8 @@ static int do_microcode_update (void) old = current->cpus_allowed; - while ((cursor = get_next_ucode(&new_mc, cursor)) > 0) { - error = microcode_sanity_check(new_mc); + while ((cursor = microcode_ops->get_next_ucode(&new_mc, cursor)) > 0) { + error = microcode_ops->microcode_sanity_check(new_mc); if (error) goto out; /* @@ -145,11 +142,12 @@ static int do_microcode_update (void) continue; cpumask_of_cpu_ptr_next(newmask, cpu); set_cpus_allowed_ptr(current, newmask); - error = get_maching_microcode(new_mc, cpu); + error = microcode_ops->get_matching_microcode(new_mc, + cpu); if (error < 0) goto out; if (error == 1) - apply_microcode(cpu); + microcode_ops->apply_microcode(cpu); } vfree(new_mc); } @@ -232,7 +230,8 @@ MODULE_ALIAS_MISCDEV(MICROCODE_MINOR); #endif /* fake device for request_firmware */ -struct platform_device *microcode_pdev; +static struct platform_device *microcode_pdev; +EXPORT_SYMBOL_GPL(microcode_pdev); static void microcode_init_cpu(int cpu, int resume) { @@ -244,9 +243,9 @@ static void microcode_init_cpu(int cpu, int resume) set_cpus_allowed_ptr(current, newmask); mutex_lock(µcode_mutex); - collect_cpu_info(cpu); + microcode_ops->collect_cpu_info(cpu); if (uci->valid && system_state == SYSTEM_RUNNING && !resume) - cpu_request_microcode(cpu); + microcode_ops->cpu_request_microcode(cpu); mutex_unlock(µcode_mutex); set_cpus_allowed_ptr(current, &old); } @@ -274,7 +273,7 @@ static ssize_t reload_store(struct sys_device *dev, mutex_lock(µcode_mutex); if (uci->valid) - err = cpu_request_microcode(cpu); + err = microcode_ops->cpu_request_microcode(cpu); mutex_unlock(µcode_mutex); put_online_cpus(); set_cpus_allowed_ptr(current, &old); @@ -349,7 +348,7 @@ static int mc_sysdev_remove(struct sys_device *sys_dev) return 0; pr_debug("microcode: CPU%d removed\n", cpu); - microcode_fini_cpu(cpu); + microcode_ops->microcode_fini_cpu(cpu); sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); return 0; } @@ -362,7 +361,7 @@ static int mc_sysdev_resume(struct sys_device *dev) return 0; pr_debug("microcode: CPU%d resumed\n", cpu); /* only CPU 0 will apply ucode here */ - apply_microcode(0); + microcode_ops->apply_microcode(0); return 0; } @@ -382,7 +381,7 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) switch (action) { case CPU_UP_CANCELED_FROZEN: /* The CPU refused to come up during a system resume */ - microcode_fini_cpu(cpu); + microcode_ops->microcode_fini_cpu(cpu); break; case CPU_ONLINE: case CPU_DOWN_FAILED: @@ -390,9 +389,9 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) break; case CPU_ONLINE_FROZEN: /* System-wide resume is in progress, try to apply microcode */ - if (apply_microcode_check_cpu(cpu)) { + if (microcode_ops->apply_microcode_check_cpu(cpu)) { /* The application of microcode failed */ - microcode_fini_cpu(cpu); + microcode_ops->microcode_fini_cpu(cpu); __mc_sysdev_add(sys_dev, 1); break; } @@ -416,12 +415,17 @@ static struct notifier_block __refdata mc_cpu_notifier = { .notifier_call = mc_cpu_callback, }; -static int __init microcode_init (void) +static int microcode_init(void *opaque, struct module *module) { + struct microcode_ops *ops = (struct microcode_ops *)opaque; int error; - printk(KERN_INFO - "IA-32 Microcode Update Driver: v" MICROCODE_VERSION " \n"); + if (microcode_ops) { + printk(KERN_ERR "microcode: already loaded the other module\n"); + return -EEXIST; + } + + microcode_ops = ops; error = microcode_dev_init(); if (error) @@ -443,8 +447,15 @@ static int __init microcode_init (void) } register_hotcpu_notifier(&mc_cpu_notifier); + + printk(KERN_INFO + "Microcode Update Driver: v" MICROCODE_VERSION + " " + " \n"); + return 0; } +EXPORT_SYMBOL_GPL(microcode_init); static void __exit microcode_exit (void) { @@ -457,7 +468,10 @@ static void __exit microcode_exit (void) put_online_cpus(); platform_device_unregister(microcode_pdev); -} -module_init(microcode_init) -module_exit(microcode_exit) + microcode_ops = NULL; + + printk(KERN_INFO + "Microcode Update Driver: v" MICROCODE_VERSION " removed.\n"); +} +EXPORT_SYMBOL_GPL(microcode_exit); diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index ca9861bf067..831db5363db 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -127,7 +127,7 @@ extern struct mutex microcode_mutex; extern struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; -void collect_cpu_info(int cpu_num) +static void collect_cpu_info(int cpu_num) { struct cpuinfo_x86 *c = &cpu_data(cpu_num); struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; @@ -175,7 +175,7 @@ static inline int microcode_update_match(int cpu_num, return 1; } -int microcode_sanity_check(void *mc) +static int microcode_sanity_check(void *mc) { struct microcode_header_intel *mc_header = mc; struct extended_sigtable *ext_header = NULL; @@ -259,7 +259,7 @@ int microcode_sanity_check(void *mc) * return 1 - found update * return < 0 - error */ -int get_matching_microcode(void *mc, int cpu) +static int get_matching_microcode(void *mc, int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; struct microcode_header_intel *mc_header = mc; @@ -288,7 +288,8 @@ int get_matching_microcode(void *mc, int cpu) return 0; find: pr_debug("microcode: CPU%d found a matching microcode update with" - " version 0x%x (current=0x%x)\n", cpu, mc_header->rev, uci->rev); + " version 0x%x (current=0x%x)\n", + cpu, mc_header->rev, uci->rev); new_mc = vmalloc(total_size); if (!new_mc) { printk(KERN_ERR "microcode: error! Can not allocate memory\n"); @@ -303,7 +304,7 @@ find: return 1; } -void apply_microcode(int cpu) +static void apply_microcode(int cpu) { unsigned long flags; unsigned int val[2]; @@ -347,7 +348,7 @@ void apply_microcode(int cpu) extern void __user *user_buffer; /* user area microcode data buffer */ extern unsigned int user_buffer_size; /* it's size */ -long get_next_ucode(void **mc, long offset) +static long get_next_ucode(void **mc, long offset) { struct microcode_header_intel mc_header; unsigned long total_size; @@ -406,7 +407,7 @@ static long get_next_ucode_from_buffer(void **mc, const u8 *buf, /* fake device for request_firmware */ extern struct platform_device *microcode_pdev; -int cpu_request_microcode(int cpu) +static int cpu_request_microcode(int cpu) { char name[30]; struct cpuinfo_x86 *c = &cpu_data(cpu); @@ -455,7 +456,7 @@ int cpu_request_microcode(int cpu) return error; } -int apply_microcode_check_cpu(int cpu) +static int apply_microcode_check_cpu(int cpu) { struct cpuinfo_x86 *c = &cpu_data(cpu); struct ucode_cpu_info *uci = ucode_cpu_info + cpu; @@ -504,13 +505,42 @@ int apply_microcode_check_cpu(int cpu) return err; } -void microcode_fini_cpu(int cpu) +static void microcode_fini_cpu(int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; mutex_lock(µcode_mutex); uci->valid = 0; - kfree(uci->mc.mc_intel); + vfree(uci->mc.mc_intel); uci->mc.mc_intel = NULL; mutex_unlock(µcode_mutex); } + +static struct microcode_ops microcode_intel_ops = { + .get_next_ucode = get_next_ucode, + .get_matching_microcode = get_matching_microcode, + .microcode_sanity_check = microcode_sanity_check, + .apply_microcode_check_cpu = apply_microcode_check_cpu, + .cpu_request_microcode = cpu_request_microcode, + .collect_cpu_info = collect_cpu_info, + .apply_microcode = apply_microcode, + .microcode_fini_cpu = microcode_fini_cpu, +}; + +static int __init microcode_intel_module_init(void) +{ + struct cpuinfo_x86 *c = &cpu_data(get_cpu()); + + if (c->x86_vendor == X86_VENDOR_INTEL) + return microcode_init(µcode_intel_ops, THIS_MODULE); + else + return -ENODEV; +} + +static void __exit microcode_intel_module_exit(void) +{ + microcode_exit(); +} + +module_init(microcode_intel_module_init) +module_exit(microcode_intel_module_exit) diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index 9231c876e37..18b2aeec2ad 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -1,3 +1,6 @@ +extern int microcode_init(void *opaque, struct module *module); +extern void microcode_exit(void); + struct microcode_ops { long (*get_next_ucode)(void **mc, long offset); long (*microcode_get_next_ucode)(void **mc, long offset); -- cgit v1.2.3-70-g09d2 From a648bf4632628c787abb0514277f2a231fca39ca Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 29 Jul 2008 10:29:18 -0700 Subject: x86, xsave: xsave cpuid feature bits Add xsave CPU feature bits. Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/feature_names.c | 2 +- include/asm-x86/cpufeature.h | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/feature_names.c b/arch/x86/kernel/cpu/feature_names.c index 0bf4d37a048..74154722565 100644 --- a/arch/x86/kernel/cpu/feature_names.c +++ b/arch/x86/kernel/cpu/feature_names.c @@ -46,7 +46,7 @@ const char * const x86_cap_flags[NCAPINTS*32] = { "pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est", "tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL, NULL, NULL, "dca", "sse4_1", "sse4_2", "x2apic", NULL, "popcnt", - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + NULL, NULL, "xsave", NULL, NULL, NULL, NULL, NULL, /* VIA/Cyrix/Centaur-defined */ NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en", diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 42afe9ca3a3..c76b3f67cb3 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -92,6 +92,7 @@ #define X86_FEATURE_XTPR (4*32+14) /* Send Task Priority Messages */ #define X86_FEATURE_DCA (4*32+18) /* Direct Cache Access */ #define X86_FEATURE_X2APIC (4*32+21) /* x2APIC */ +#define X86_FEATURE_XSAVE (4*32+26) /* XSAVE */ /* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */ #define X86_FEATURE_XSTORE (5*32+ 2) /* on-CPU RNG present (xstore insn) */ @@ -191,6 +192,7 @@ extern const char * const x86_power_flags[32]; #define cpu_has_arch_perfmon boot_cpu_has(X86_FEATURE_ARCH_PERFMON) #define cpu_has_pat boot_cpu_has(X86_FEATURE_PAT) #define cpu_has_x2apic boot_cpu_has(X86_FEATURE_X2APIC) +#define cpu_has_xsave boot_cpu_has(X86_FEATURE_XSAVE) #if defined(CONFIG_X86_INVLPG) || defined(CONFIG_X86_64) # define cpu_has_invlpg 1 -- cgit v1.2.3-70-g09d2 From dc1e35c6e95e8923cf1d3510438b63c600fee1e2 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 29 Jul 2008 10:29:19 -0700 Subject: x86, xsave: enable xsave/xrstor on cpus with xsave support Enables xsave/xrstor by turning on cr4.osxsave on cpu's which have the xsave support. For now, features that OS supports/enabled are FP and SSE. Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- arch/x86/kernel/Makefile | 2 +- arch/x86/kernel/cpu/common.c | 8 ++++ arch/x86/kernel/i387.c | 12 ++++++ arch/x86/kernel/traps_32.c | 1 - arch/x86/kernel/traps_64.c | 4 -- arch/x86/kernel/xsave.c | 87 +++++++++++++++++++++++++++++++++++++++ include/asm-x86/i387.h | 1 + include/asm-x86/processor-flags.h | 1 + include/asm-x86/processor.h | 12 ++++++ include/asm-x86/xsave.h | 26 ++++++++++++ 10 files changed, 148 insertions(+), 6 deletions(-) create mode 100644 arch/x86/kernel/xsave.c create mode 100644 include/asm-x86/xsave.h (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index a07ec14f331..d6ea91abaeb 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -38,7 +38,7 @@ obj-y += tsc.o io_delay.o rtc.o obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o obj-y += process.o -obj-y += i387.o +obj-y += i387.o xsave.o obj-y += ptrace.o obj-y += ds.o obj-$(CONFIG_X86_32) += tls.o diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 80ab20d4fa3..fabbcb7020f 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -712,6 +712,14 @@ void __cpuinit cpu_init(void) current_thread_info()->status = 0; clear_used_math(); mxcsr_feature_mask_init(); + + /* + * Boot processor to setup the FP and extended state context info. + */ + if (!smp_processor_id()) + init_thread_xstate(); + + xsave_init(); } #ifdef CONFIG_HOTPLUG_CPU diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index eb9ddd8efb8..e22a9a9dce8 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -61,6 +61,11 @@ void __init init_thread_xstate(void) return; } + if (cpu_has_xsave) { + xsave_cntxt_init(); + return; + } + if (cpu_has_fxsr) xstate_size = sizeof(struct i387_fxsave_struct); #ifdef CONFIG_X86_32 @@ -83,6 +88,13 @@ void __cpuinit fpu_init(void) write_cr0(oldcr0 & ~(X86_CR0_TS|X86_CR0_EM)); /* clear TS and EM */ + /* + * Boot processor to setup the FP and extended state context info. + */ + if (!smp_processor_id()) + init_thread_xstate(); + xsave_init(); + mxcsr_feature_mask_init(); /* clean state in init */ current_thread_info()->status = 0; diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index 03df8e45e5a..da5a5964fcc 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -1228,7 +1228,6 @@ void __init trap_init(void) set_bit(SYSCALL_VECTOR, used_vectors); - init_thread_xstate(); /* * Should be a barrier for any external CPU state: */ diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index 513caaca711..3580a7938a2 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -1172,10 +1172,6 @@ void __init trap_init(void) #ifdef CONFIG_IA32_EMULATION set_system_gate(IA32_SYSCALL_VECTOR, ia32_syscall); #endif - /* - * initialize the per thread extended state: - */ - init_thread_xstate(); /* * Should be a barrier for any external CPU state: */ diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c new file mode 100644 index 00000000000..c68b7c4ca24 --- /dev/null +++ b/arch/x86/kernel/xsave.c @@ -0,0 +1,87 @@ +/* + * xsave/xrstor support. + * + * Author: Suresh Siddha + */ +#include +#include +#include + +/* + * Supported feature mask by the CPU and the kernel. + */ +unsigned int pcntxt_hmask, pcntxt_lmask; + +/* + * Represents init state for the supported extended state. + */ +struct xsave_struct *init_xstate_buf; + +/* + * Enable the extended processor state save/restore feature + */ +void __cpuinit xsave_init(void) +{ + if (!cpu_has_xsave) + return; + + set_in_cr4(X86_CR4_OSXSAVE); + + /* + * Enable all the features that the HW is capable of + * and the Linux kernel is aware of. + * + * xsetbv(); + */ + asm volatile(".byte 0x0f,0x01,0xd1" : : "c" (0), + "a" (pcntxt_lmask), "d" (pcntxt_hmask)); +} + +/* + * setup the xstate image representing the init state + */ +void setup_xstate_init(void) +{ + init_xstate_buf = alloc_bootmem(xstate_size); + init_xstate_buf->i387.mxcsr = MXCSR_DEFAULT; +} + +/* + * Enable and initialize the xsave feature. + */ +void __init xsave_cntxt_init(void) +{ + unsigned int eax, ebx, ecx, edx; + + cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx); + + pcntxt_lmask = eax; + pcntxt_hmask = edx; + + if ((pcntxt_lmask & XSTATE_FPSSE) != XSTATE_FPSSE) { + printk(KERN_ERR "FP/SSE not shown under xsave features %x\n", + pcntxt_lmask); + BUG(); + } + + /* + * for now OS knows only about FP/SSE + */ + pcntxt_lmask = pcntxt_lmask & XCNTXT_LMASK; + pcntxt_hmask = pcntxt_hmask & XCNTXT_HMASK; + + xsave_init(); + + /* + * Recompute the context size for enabled features + */ + cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx); + + xstate_size = ebx; + + setup_xstate_init(); + + printk(KERN_INFO "xsave/xrstor: enabled xstate_bv 0x%Lx, " + "cntxt size 0x%x\n", + (pcntxt_lmask | ((u64) pcntxt_hmask << 32)), xstate_size); +} diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h index 3958de6aad0..6a664789667 100644 --- a/include/asm-x86/i387.h +++ b/include/asm-x86/i387.h @@ -18,6 +18,7 @@ #include #include #include +#include extern void fpu_init(void); extern void mxcsr_feature_mask_init(void); diff --git a/include/asm-x86/processor-flags.h b/include/asm-x86/processor-flags.h index 5dd79774f69..dc5f0712f9f 100644 --- a/include/asm-x86/processor-flags.h +++ b/include/asm-x86/processor-flags.h @@ -59,6 +59,7 @@ #define X86_CR4_OSFXSR 0x00000200 /* enable fast FPU save and restore */ #define X86_CR4_OSXMMEXCPT 0x00000400 /* enable unmasked SSE exceptions */ #define X86_CR4_VMXE 0x00002000 /* enable VMX virtualization */ +#define X86_CR4_OSXSAVE 0x00040000 /* enable xsave and xrestore */ /* * x86-64 Task Priority Register, CR8 diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index d60b4d81feb..d7c0221c027 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -346,6 +346,18 @@ struct i387_soft_struct { u32 entry_eip; }; +struct xsave_hdr_struct { + u64 xstate_bv; + u64 reserved1[2]; + u64 reserved2[5]; +} __attribute__((packed)); + +struct xsave_struct { + struct i387_fxsave_struct i387; + struct xsave_hdr_struct xsave_hdr; + /* new processor state extensions will go here */ +} __attribute__ ((packed, aligned (64))); + union thread_xstate { struct i387_fsave_struct fsave; struct i387_fxsave_struct fxsave; diff --git a/include/asm-x86/xsave.h b/include/asm-x86/xsave.h new file mode 100644 index 00000000000..6d70e62c6bd --- /dev/null +++ b/include/asm-x86/xsave.h @@ -0,0 +1,26 @@ +#ifndef __ASM_X86_XSAVE_H +#define __ASM_X86_XSAVE_H + +#include +#include + +#define XSTATE_FP 0x1 +#define XSTATE_SSE 0x2 + +#define XSTATE_FPSSE (XSTATE_FP | XSTATE_SSE) + +#define FXSAVE_SIZE 512 + +/* + * These are the features that the OS can handle currently. + */ +#define XCNTXT_LMASK (XSTATE_FP | XSTATE_SSE) +#define XCNTXT_HMASK 0x0 + +extern unsigned int xstate_size, pcntxt_hmask, pcntxt_lmask; +extern struct xsave_struct *init_xstate_buf; + +extern void xsave_cntxt_init(void); +extern void xsave_init(void); + +#endif -- cgit v1.2.3-70-g09d2 From b359e8a434cc3d09847010fc4aeccf48d69740e4 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 29 Jul 2008 10:29:20 -0700 Subject: x86, xsave: context switch support using xsave/xrstor Uses xsave/xrstor (instead of traditional fxsave/fxrstor) in context switch when available. Introduces TS_XSAVE flag, which determine the need to use xsave/xrstor instructions during context switch instead of the legacy fxsave/fxrstor instructions. Thread-synchronous status word is already in L1 cache during this code patch and thus minimizes the performance penality compared to (cpu_has_xsave) checks. Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/common.c | 5 +++- arch/x86/kernel/i387.c | 5 +++- arch/x86/kernel/traps_64.c | 2 +- include/asm-x86/i387.h | 64 +++++++++++++++++++++++++++++++++++++++---- include/asm-x86/processor.h | 1 + include/asm-x86/thread_info.h | 1 + include/asm-x86/xsave.h | 35 ++++++++++++++++++++++- 7 files changed, 104 insertions(+), 9 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index fabbcb7020f..6c2b9e756db 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -709,7 +709,10 @@ void __cpuinit cpu_init(void) /* * Force FPU initialization: */ - current_thread_info()->status = 0; + if (cpu_has_xsave) + current_thread_info()->status = TS_XSAVE; + else + current_thread_info()->status = 0; clear_used_math(); mxcsr_feature_mask_init(); diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index e22a9a9dce8..b778e17e4b0 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -97,7 +97,10 @@ void __cpuinit fpu_init(void) mxcsr_feature_mask_init(); /* clean state in init */ - current_thread_info()->status = 0; + if (cpu_has_xsave) + current_thread_info()->status = TS_XSAVE; + else + current_thread_info()->status = 0; clear_used_math(); } #endif /* CONFIG_X86_64 */ diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index 3580a7938a2..38eb76156a4 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -1134,7 +1134,7 @@ asmlinkage void math_state_restore(void) /* * Paranoid restore. send a SIGSEGV if we fail to restore the state. */ - if (unlikely(restore_fpu_checking(&me->thread.xstate->fxsave))) { + if (unlikely(restore_fpu_checking(me))) { stts(); force_sig(SIGSEGV, me); return; diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h index 6a664789667..a6d256f4ac8 100644 --- a/include/asm-x86/i387.h +++ b/include/asm-x86/i387.h @@ -36,6 +36,8 @@ extern int save_i387_ia32(struct _fpstate_ia32 __user *buf); extern int restore_i387_ia32(struct _fpstate_ia32 __user *buf); #endif +#define X87_FSW_ES (1 << 7) /* Exception Summary */ + #ifdef CONFIG_X86_64 /* Ignore delayed exceptions from user space */ @@ -46,7 +48,7 @@ static inline void tolerant_fwait(void) _ASM_EXTABLE(1b, 2b)); } -static inline int restore_fpu_checking(struct i387_fxsave_struct *fx) +static inline int fxrstor_checking(struct i387_fxsave_struct *fx) { int err; @@ -66,15 +68,31 @@ static inline int restore_fpu_checking(struct i387_fxsave_struct *fx) return err; } -#define X87_FSW_ES (1 << 7) /* Exception Summary */ +static inline int restore_fpu_checking(struct task_struct *tsk) +{ + if (task_thread_info(tsk)->status & TS_XSAVE) + return xrstor_checking(&tsk->thread.xstate->xsave); + else + return fxrstor_checking(&tsk->thread.xstate->fxsave); +} /* AMD CPUs don't save/restore FDP/FIP/FOP unless an exception is pending. Clear the x87 state here by setting it to fixed values. The kernel data segment can be sometimes 0 and sometimes new user value. Both should be ok. Use the PDA as safe address because it should be already in L1. */ -static inline void clear_fpu_state(struct i387_fxsave_struct *fx) +static inline void clear_fpu_state(struct task_struct *tsk) { + struct xsave_struct *xstate = &tsk->thread.xstate->xsave; + struct i387_fxsave_struct *fx = &tsk->thread.xstate->fxsave; + + /* + * xsave header may indicate the init state of the FP. + */ + if ((task_thread_info(tsk)->status & TS_XSAVE) && + !(xstate->xsave_hdr.xstate_bv & XSTATE_FP)) + return; + if (unlikely(fx->swd & X87_FSW_ES)) asm volatile("fnclex"); alternative_input(ASM_NOP8 ASM_NOP2, @@ -107,7 +125,7 @@ static inline int save_i387_checking(struct i387_fxsave_struct __user *fx) return err; } -static inline void __save_init_fpu(struct task_struct *tsk) +static inline void fxsave(struct task_struct *tsk) { /* Using "rex64; fxsave %0" is broken because, if the memory operand uses any extended registers for addressing, a second REX prefix @@ -132,7 +150,16 @@ static inline void __save_init_fpu(struct task_struct *tsk) : "=m" (tsk->thread.xstate->fxsave) : "cdaSDb" (&tsk->thread.xstate->fxsave)); #endif - clear_fpu_state(&tsk->thread.xstate->fxsave); +} + +static inline void __save_init_fpu(struct task_struct *tsk) +{ + if (task_thread_info(tsk)->status & TS_XSAVE) + xsave(tsk); + else + fxsave(tsk); + + clear_fpu_state(tsk); task_thread_info(tsk)->status &= ~TS_USEDFPU; } @@ -147,6 +174,10 @@ static inline void tolerant_fwait(void) static inline void restore_fpu(struct task_struct *tsk) { + if (task_thread_info(tsk)->status & TS_XSAVE) { + xrstor_checking(&tsk->thread.xstate->xsave); + return; + } /* * The "nop" is needed to make the instructions the same * length. @@ -172,6 +203,27 @@ static inline void restore_fpu(struct task_struct *tsk) */ static inline void __save_init_fpu(struct task_struct *tsk) { + if (task_thread_info(tsk)->status & TS_XSAVE) { + struct xsave_struct *xstate = &tsk->thread.xstate->xsave; + struct i387_fxsave_struct *fx = &tsk->thread.xstate->fxsave; + + xsave(tsk); + + /* + * xsave header may indicate the init state of the FP. + */ + if (!(xstate->xsave_hdr.xstate_bv & XSTATE_FP)) + goto end; + + if (unlikely(fx->swd & X87_FSW_ES)) + asm volatile("fnclex"); + + /* + * we can do a simple return here or be paranoid :) + */ + goto clear_state; + } + /* Use more nops than strictly needed in case the compiler varies code */ alternative_input( @@ -181,6 +233,7 @@ static inline void __save_init_fpu(struct task_struct *tsk) X86_FEATURE_FXSR, [fx] "m" (tsk->thread.xstate->fxsave), [fsw] "m" (tsk->thread.xstate->fxsave.swd) : "memory"); +clear_state: /* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception is pending. Clear the x87 state here by setting it to fixed values. safe_address is a random variable that should be in L1 */ @@ -190,6 +243,7 @@ static inline void __save_init_fpu(struct task_struct *tsk) "fildl %[addr]", /* set F?P to defined value */ X86_FEATURE_FXSAVE_LEAK, [addr] "m" (safe_address)); +end: task_thread_info(tsk)->status &= ~TS_USEDFPU; } diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index d7c0221c027..77b7af6b573 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -362,6 +362,7 @@ union thread_xstate { struct i387_fsave_struct fsave; struct i387_fxsave_struct fxsave; struct i387_soft_struct soft; + struct xsave_struct xsave; }; #ifdef CONFIG_X86_64 diff --git a/include/asm-x86/thread_info.h b/include/asm-x86/thread_info.h index e64be8863b7..30586f2ee55 100644 --- a/include/asm-x86/thread_info.h +++ b/include/asm-x86/thread_info.h @@ -239,6 +239,7 @@ static inline struct thread_info *stack_thread_info(void) #define TS_POLLING 0x0004 /* true if in idle loop and not sleeping */ #define TS_RESTORE_SIGMASK 0x0008 /* restore signal mask in do_signal() */ +#define TS_XSAVE 0x0010 /* Use xsave/xrstor */ #define tsk_is_polling(t) (task_thread_info(t)->status & TS_POLLING) diff --git a/include/asm-x86/xsave.h b/include/asm-x86/xsave.h index 6d70e62c6bd..e835a917ee1 100644 --- a/include/asm-x86/xsave.h +++ b/include/asm-x86/xsave.h @@ -17,10 +17,43 @@ #define XCNTXT_LMASK (XSTATE_FP | XSTATE_SSE) #define XCNTXT_HMASK 0x0 +#ifdef CONFIG_X86_64 +#define REX_PREFIX "0x48, " +#else +#define REX_PREFIX +#endif + extern unsigned int xstate_size, pcntxt_hmask, pcntxt_lmask; extern struct xsave_struct *init_xstate_buf; extern void xsave_cntxt_init(void); extern void xsave_init(void); - +extern int init_fpu(struct task_struct *child); + +static inline int xrstor_checking(struct xsave_struct *fx) +{ + int err; + + asm volatile("1: .byte " REX_PREFIX "0x0f,0xae,0x2f\n\t" + "2:\n" + ".section .fixup,\"ax\"\n" + "3: movl $-1,%[err]\n" + " jmp 2b\n" + ".previous\n" + _ASM_EXTABLE(1b, 3b) + : [err] "=r" (err) + : "D" (fx), "m" (*fx), "a" (-1), "d" (-1), "0" (0) + : "memory"); + + return err; +} + +static inline void xsave(struct task_struct *tsk) +{ + /* This, however, we can work around by forcing the compiler to select + an addressing mode that doesn't require extended registers. */ + __asm__ __volatile__(".byte " REX_PREFIX "0x0f,0xae,0x27" + : : "D" (&(tsk->thread.xstate->xsave)), + "a" (-1), "d"(-1) : "memory"); +} #endif -- cgit v1.2.3-70-g09d2 From 3c1c7f101426cb2ecc79d817a8a65928965fc860 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 29 Jul 2008 10:29:21 -0700 Subject: x86, xsave: dynamically allocate sigframes fpstate instead of static allocation dynamically allocate fpstate on the stack, instead of static allocation in the current sigframe layout on the user stack. This will allow the fpstate structure to grow in the future, which includes extended state information supporting xsave/xrstor. signal handlers will be able to access the fpstate pointer from the sigcontext structure asusual, with no change. For the non RT sigframe's (which are supported only for 32bit apps), current static fpstate layout in the sigframe will be unused(so that we don't change the extramask[] offset in the sigframe and thus prevent breaking app's which modify extramask[]). Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- arch/x86/ia32/ia32_signal.c | 24 ++++++++++++++++-------- arch/x86/kernel/i387.c | 2 ++ arch/x86/kernel/sigframe.h | 14 ++++++++++++-- arch/x86/kernel/signal_32.c | 18 +++++++++++++----- arch/x86/kernel/signal_64.c | 2 +- arch/x86/kernel/xsave.c | 4 ++++ include/asm-x86/i387.h | 2 ++ 7 files changed, 50 insertions(+), 16 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c index 20af4c79579..a05bf0fb741 100644 --- a/arch/x86/ia32/ia32_signal.c +++ b/arch/x86/ia32/ia32_signal.c @@ -179,9 +179,10 @@ struct sigframe u32 pretcode; int sig; struct sigcontext_ia32 sc; - struct _fpstate_ia32 fpstate; + struct _fpstate_ia32 fpstate_unused; /* look at kernel/sigframe.h */ unsigned int extramask[_COMPAT_NSIG_WORDS-1]; char retcode[8]; + /* fp state follows here */ }; struct rt_sigframe @@ -192,8 +193,8 @@ struct rt_sigframe u32 puc; compat_siginfo_t info; struct ucontext_ia32 uc; - struct _fpstate_ia32 fpstate; char retcode[8]; + /* fp state follows here */ }; #define COPY(x) { \ @@ -402,7 +403,8 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc, * Determine which stack to use.. */ static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, - size_t frame_size) + size_t frame_size, + struct _fpstate_ia32 **fpstate) { unsigned long sp; @@ -421,6 +423,11 @@ static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, ka->sa.sa_restorer) sp = (unsigned long) ka->sa.sa_restorer; + if (used_math()) { + sp = sp - sig_xstate_ia32_size; + *fpstate = (struct _fpstate_ia32 *) sp; + } + sp -= frame_size; /* Align the stack pointer according to the i386 ABI, * i.e. so that on function entry ((sp + 4) & 15) == 0. */ @@ -434,6 +441,7 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka, struct sigframe __user *frame; void __user *restorer; int err = 0; + struct _fpstate_ia32 __user *fpstate = NULL; /* copy_to_user optimizes that into a single 8 byte store */ static const struct { @@ -448,7 +456,7 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka, 0, }; - frame = get_sigframe(ka, regs, sizeof(*frame)); + frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate); if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) goto give_sigsegv; @@ -457,8 +465,7 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka, if (err) goto give_sigsegv; - err |= ia32_setup_sigcontext(&frame->sc, &frame->fpstate, regs, - set->sig[0]); + err |= ia32_setup_sigcontext(&frame->sc, fpstate, regs, set->sig[0]); if (err) goto give_sigsegv; @@ -522,6 +529,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, struct rt_sigframe __user *frame; void __user *restorer; int err = 0; + struct _fpstate_ia32 __user *fpstate = NULL; /* __copy_to_user optimizes that into a single 8 byte store */ static const struct { @@ -537,7 +545,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, 0, }; - frame = get_sigframe(ka, regs, sizeof(*frame)); + frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate); if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) goto give_sigsegv; @@ -556,7 +564,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, err |= __put_user(sas_ss_flags(regs->sp), &frame->uc.uc_stack.ss_flags); err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size); - err |= ia32_setup_sigcontext(&frame->uc.uc_mcontext, &frame->fpstate, + err |= ia32_setup_sigcontext(&frame->uc.uc_mcontext, fpstate, regs, set->sig[0]); err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); if (err) diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index b778e17e4b0..51fb288a2c9 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -24,6 +24,7 @@ # define save_i387_ia32 save_i387 # define restore_i387_ia32 restore_i387 # define _fpstate_ia32 _fpstate +# define sig_xstate_ia32_size sig_xstate_size # define user_i387_ia32_struct user_i387_struct # define user32_fxsr_struct user_fxsr_struct #endif @@ -36,6 +37,7 @@ static unsigned int mxcsr_feature_mask __read_mostly = 0xffffffffu; unsigned int xstate_size; +unsigned int sig_xstate_ia32_size = sizeof(struct _fpstate_ia32); static struct i387_fxsave_struct fx_scratch __cpuinitdata; void __cpuinit mxcsr_feature_mask_init(void) diff --git a/arch/x86/kernel/sigframe.h b/arch/x86/kernel/sigframe.h index 72bbb519d2d..6dd7e2b70a4 100644 --- a/arch/x86/kernel/sigframe.h +++ b/arch/x86/kernel/sigframe.h @@ -3,9 +3,18 @@ struct sigframe { char __user *pretcode; int sig; struct sigcontext sc; - struct _fpstate fpstate; + /* + * fpstate is unused. fpstate is moved/allocated after + * retcode[] below. This movement allows to have the FP state and the + * future state extensions (xsave) stay together. + * And at the same time retaining the unused fpstate, prevents changing + * the offset of extramask[] in the sigframe and thus prevent any + * legacy application accessing/modifying it. + */ + struct _fpstate fpstate_unused; unsigned long extramask[_NSIG_WORDS-1]; char retcode[8]; + /* fp state follows here */ }; struct rt_sigframe { @@ -15,13 +24,14 @@ struct rt_sigframe { void __user *puc; struct siginfo info; struct ucontext uc; - struct _fpstate fpstate; char retcode[8]; + /* fp state follows here */ }; #else struct rt_sigframe { char __user *pretcode; struct ucontext uc; struct siginfo info; + /* fp state follows here */ }; #endif diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c index 6fb5bcdd893..19a7a5669b5 100644 --- a/arch/x86/kernel/signal_32.c +++ b/arch/x86/kernel/signal_32.c @@ -306,7 +306,8 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate, * Determine which stack to use.. */ static inline void __user * -get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size) +get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size, + struct _fpstate **fpstate) { unsigned long sp; @@ -332,6 +333,11 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size) sp = (unsigned long) ka->sa.sa_restorer; } + if (used_math()) { + sp = sp - sig_xstate_size; + *fpstate = (struct _fpstate *) sp; + } + sp -= frame_size; /* * Align the stack pointer according to the i386 ABI, @@ -350,8 +356,9 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set, void __user *restorer; int err = 0; int usig; + struct _fpstate __user *fpstate = NULL; - frame = get_sigframe(ka, regs, sizeof(*frame)); + frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate); if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) goto give_sigsegv; @@ -366,7 +373,7 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set, if (err) goto give_sigsegv; - err = setup_sigcontext(&frame->sc, &frame->fpstate, regs, set->sig[0]); + err = setup_sigcontext(&frame->sc, fpstate, regs, set->sig[0]); if (err) goto give_sigsegv; @@ -427,8 +434,9 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, void __user *restorer; int err = 0; int usig; + struct _fpstate __user *fpstate = NULL; - frame = get_sigframe(ka, regs, sizeof(*frame)); + frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate); if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) goto give_sigsegv; @@ -453,7 +461,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, err |= __put_user(sas_ss_flags(regs->sp), &frame->uc.uc_stack.ss_flags); err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size); - err |= setup_sigcontext(&frame->uc.uc_mcontext, &frame->fpstate, + err |= setup_sigcontext(&frame->uc.uc_mcontext, fpstate, regs, set->sig[0]); err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); if (err) diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c index ca316b5b742..0deab8eff33 100644 --- a/arch/x86/kernel/signal_64.c +++ b/arch/x86/kernel/signal_64.c @@ -281,7 +281,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, struct task_struct *me = current; if (used_math()) { - fp = get_stack(ka, regs, sizeof(struct _fpstate)); + fp = get_stack(ka, regs, sig_xstate_size); frame = (void __user *)round_down( (unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8; diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c index c68b7c4ca24..7ad169e3352 100644 --- a/arch/x86/kernel/xsave.c +++ b/arch/x86/kernel/xsave.c @@ -17,6 +17,10 @@ unsigned int pcntxt_hmask, pcntxt_lmask; */ struct xsave_struct *init_xstate_buf; +#ifdef CONFIG_X86_64 +unsigned int sig_xstate_size = sizeof(struct _fpstate); +#endif + /* * Enable the extended processor state save/restore feature */ diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h index a6d256f4ac8..36dca8db166 100644 --- a/include/asm-x86/i387.h +++ b/include/asm-x86/i387.h @@ -20,6 +20,7 @@ #include #include +extern unsigned int sig_xstate_size; extern void fpu_init(void); extern void mxcsr_feature_mask_init(void); extern int init_fpu(struct task_struct *child); @@ -31,6 +32,7 @@ extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get; extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set; #ifdef CONFIG_IA32_EMULATION +extern unsigned int sig_xstate_ia32_size; struct _fpstate_ia32; extern int save_i387_ia32(struct _fpstate_ia32 __user *buf); extern int restore_i387_ia32(struct _fpstate_ia32 __user *buf); -- cgit v1.2.3-70-g09d2 From ab5137015fed9b948fe835a2d99a4cfbd50a0c40 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 29 Jul 2008 10:29:22 -0700 Subject: x86, xsave: reorganization of signal save/restore fpstate code layout move 64bit routines that saves/restores fpstate in/from user stack from signal_64.c to xsave.c restore_i387_xstate() now handles the condition when user passes NULL fpstate. Other misc changes for prepartion of xsave/xrstor sigcontext support. Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- arch/x86/ia32/ia32_signal.c | 28 ++++----------- arch/x86/kernel/i387.c | 44 +++++++++++++++++------- arch/x86/kernel/signal_32.c | 28 ++++----------- arch/x86/kernel/signal_64.c | 83 ++------------------------------------------- arch/x86/kernel/xsave.c | 79 ++++++++++++++++++++++++++++++++++++++++++ include/asm-x86/i387.h | 13 +++---- 6 files changed, 134 insertions(+), 141 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c index a05bf0fb741..c596eabbe98 100644 --- a/arch/x86/ia32/ia32_signal.c +++ b/arch/x86/ia32/ia32_signal.c @@ -216,7 +216,7 @@ static int ia32_restore_sigcontext(struct pt_regs *regs, unsigned int *peax) { unsigned int tmpflags, gs, oldgs, err = 0; - struct _fpstate_ia32 __user *buf; + void __user *buf; u32 tmp; /* Always make any pending restarted system calls return -EINTR */ @@ -260,26 +260,12 @@ static int ia32_restore_sigcontext(struct pt_regs *regs, err |= __get_user(tmp, &sc->fpstate); buf = compat_ptr(tmp); - if (buf) { - if (!access_ok(VERIFY_READ, buf, sizeof(*buf))) - goto badframe; - err |= restore_i387_ia32(buf); - } else { - struct task_struct *me = current; - - if (used_math()) { - clear_fpu(me); - clear_used_math(); - } - } + err |= restore_i387_xstate_ia32(buf); err |= __get_user(tmp, &sc->ax); *peax = tmp; return err; - -badframe: - return 1; } asmlinkage long sys32_sigreturn(struct pt_regs *regs) @@ -351,7 +337,7 @@ badframe: */ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc, - struct _fpstate_ia32 __user *fpstate, + void __user *fpstate, struct pt_regs *regs, unsigned int mask) { int tmp, err = 0; @@ -382,7 +368,7 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc, err |= __put_user((u32)regs->flags, &sc->flags); err |= __put_user((u32)regs->sp, &sc->sp_at_signal); - tmp = save_i387_ia32(fpstate); + tmp = save_i387_xstate_ia32(fpstate); if (tmp < 0) err = -EFAULT; else { @@ -404,7 +390,7 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc, */ static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size, - struct _fpstate_ia32 **fpstate) + void **fpstate) { unsigned long sp; @@ -441,7 +427,7 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka, struct sigframe __user *frame; void __user *restorer; int err = 0; - struct _fpstate_ia32 __user *fpstate = NULL; + void __user *fpstate = NULL; /* copy_to_user optimizes that into a single 8 byte store */ static const struct { @@ -529,7 +515,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, struct rt_sigframe __user *frame; void __user *restorer; int err = 0; - struct _fpstate_ia32 __user *fpstate = NULL; + void __user *fpstate = NULL; /* __copy_to_user optimizes that into a single 8 byte store */ static const struct { diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 51fb288a2c9..7daf3a011dd 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -21,9 +21,10 @@ # include # include #else -# define save_i387_ia32 save_i387 -# define restore_i387_ia32 restore_i387 +# define save_i387_xstate_ia32 save_i387_xstate +# define restore_i387_xstate_ia32 restore_i387_xstate # define _fpstate_ia32 _fpstate +# define _xstate_ia32 _xstate # define sig_xstate_ia32_size sig_xstate_size # define user_i387_ia32_struct user_i387_struct # define user32_fxsr_struct user_fxsr_struct @@ -424,7 +425,6 @@ static inline int save_i387_fsave(struct _fpstate_ia32 __user *buf) struct task_struct *tsk = current; struct i387_fsave_struct *fp = &tsk->thread.xstate->fsave; - unlazy_fpu(tsk); fp->status = fp->swd; if (__copy_to_user(buf, fp, sizeof(struct i387_fsave_struct))) return -1; @@ -438,8 +438,6 @@ static int save_i387_fxsave(struct _fpstate_ia32 __user *buf) struct user_i387_ia32_struct env; int err = 0; - unlazy_fpu(tsk); - convert_from_fxsr(&env, tsk); if (__copy_to_user(buf, &env, sizeof(env))) return -1; @@ -455,10 +453,16 @@ static int save_i387_fxsave(struct _fpstate_ia32 __user *buf) return 1; } -int save_i387_ia32(struct _fpstate_ia32 __user *buf) +int save_i387_xstate_ia32(void __user *buf) { + struct _fpstate_ia32 __user *fp = (struct _fpstate_ia32 __user *) buf; + struct task_struct *tsk = current; + if (!used_math()) return 0; + + if (!access_ok(VERIFY_WRITE, buf, sig_xstate_ia32_size)) + return -EACCES; /* * This will cause a "finit" to be triggered by the next * attempted FPU operation by the 'current' process. @@ -468,13 +472,15 @@ int save_i387_ia32(struct _fpstate_ia32 __user *buf) if (!HAVE_HWFP) { return fpregs_soft_get(current, NULL, 0, sizeof(struct user_i387_ia32_struct), - NULL, buf) ? -1 : 1; + NULL, fp) ? -1 : 1; } + unlazy_fpu(tsk); + if (cpu_has_fxsr) - return save_i387_fxsave(buf); + return save_i387_fxsave(fp); else - return save_i387_fsave(buf); + return save_i387_fsave(fp); } static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf) @@ -502,14 +508,26 @@ static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf) return 0; } -int restore_i387_ia32(struct _fpstate_ia32 __user *buf) +int restore_i387_xstate_ia32(void __user *buf) { int err; struct task_struct *tsk = current; + struct _fpstate_ia32 __user *fp = (struct _fpstate_ia32 __user *) buf; if (HAVE_HWFP) clear_fpu(tsk); + if (!buf) { + if (used_math()) { + clear_fpu(tsk); + clear_used_math(); + } + + return 0; + } else + if (!access_ok(VERIFY_READ, buf, sig_xstate_ia32_size)) + return -EACCES; + if (!used_math()) { err = init_fpu(tsk); if (err) @@ -518,13 +536,13 @@ int restore_i387_ia32(struct _fpstate_ia32 __user *buf) if (HAVE_HWFP) { if (cpu_has_fxsr) - err = restore_i387_fxsave(buf); + err = restore_i387_fxsave(fp); else - err = restore_i387_fsave(buf); + err = restore_i387_fsave(fp); } else { err = fpregs_soft_set(current, NULL, 0, sizeof(struct user_i387_ia32_struct), - NULL, buf) != 0; + NULL, fp) != 0; } set_used_math(); diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c index 19a7a5669b5..690cc616ac0 100644 --- a/arch/x86/kernel/signal_32.c +++ b/arch/x86/kernel/signal_32.c @@ -159,28 +159,14 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, } { - struct _fpstate __user *buf; + void __user *buf; err |= __get_user(buf, &sc->fpstate); - if (buf) { - if (!access_ok(VERIFY_READ, buf, sizeof(*buf))) - goto badframe; - err |= restore_i387(buf); - } else { - struct task_struct *me = current; - - if (used_math()) { - clear_fpu(me); - clear_used_math(); - } - } + err |= restore_i387_xstate(buf); } err |= __get_user(*pax, &sc->ax); return err; - -badframe: - return 1; } asmlinkage unsigned long sys_sigreturn(unsigned long __unused) @@ -262,7 +248,7 @@ badframe: * Set up a signal frame. */ static int -setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate, +setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate, struct pt_regs *regs, unsigned long mask) { int tmp, err = 0; @@ -289,7 +275,7 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate, err |= __put_user(regs->sp, &sc->sp_at_signal); err |= __put_user(regs->ss, (unsigned int __user *)&sc->ss); - tmp = save_i387(fpstate); + tmp = save_i387_xstate(fpstate); if (tmp < 0) err = 1; else @@ -307,7 +293,7 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate, */ static inline void __user * get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size, - struct _fpstate **fpstate) + void **fpstate) { unsigned long sp; @@ -356,7 +342,7 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set, void __user *restorer; int err = 0; int usig; - struct _fpstate __user *fpstate = NULL; + void __user *fpstate = NULL; frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate); @@ -434,7 +420,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, void __user *restorer; int err = 0; int usig; - struct _fpstate __user *fpstate = NULL; + void __user *fpstate = NULL; frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate); diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c index 0deab8eff33..ddf6123a55c 100644 --- a/arch/x86/kernel/signal_64.c +++ b/arch/x86/kernel/signal_64.c @@ -53,69 +53,6 @@ sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, return do_sigaltstack(uss, uoss, regs->sp); } -/* - * Signal frame handlers. - */ - -static inline int save_i387(struct _fpstate __user *buf) -{ - struct task_struct *tsk = current; - int err = 0; - - BUILD_BUG_ON(sizeof(struct user_i387_struct) != - sizeof(tsk->thread.xstate->fxsave)); - - if ((unsigned long)buf % 16) - printk("save_i387: bad fpstate %p\n", buf); - - if (!used_math()) - return 0; - clear_used_math(); /* trigger finit */ - if (task_thread_info(tsk)->status & TS_USEDFPU) { - err = save_i387_checking((struct i387_fxsave_struct __user *) - buf); - if (err) - return err; - task_thread_info(tsk)->status &= ~TS_USEDFPU; - stts(); - } else { - if (__copy_to_user(buf, &tsk->thread.xstate->fxsave, - sizeof(struct i387_fxsave_struct))) - return -1; - } - return 1; -} - -/* - * This restores directly out of user space. Exceptions are handled. - */ -static inline int restore_i387(struct _fpstate __user *buf) -{ - struct task_struct *tsk = current; - int err; - - if (!used_math()) { - err = init_fpu(tsk); - if (err) - return err; - } - - if (!(task_thread_info(current)->status & TS_USEDFPU)) { - clts(); - task_thread_info(current)->status |= TS_USEDFPU; - } - err = restore_fpu_checking((__force struct i387_fxsave_struct *)buf); - if (unlikely(err)) { - /* - * Encountered an error while doing the restore from the - * user buffer, clear the fpu state. - */ - clear_fpu(tsk); - clear_used_math(); - } - return err; -} - /* * Do a signal return; undo the signal stack. */ @@ -160,25 +97,11 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, { struct _fpstate __user * buf; err |= __get_user(buf, &sc->fpstate); - - if (buf) { - if (!access_ok(VERIFY_READ, buf, sizeof(*buf))) - goto badframe; - err |= restore_i387(buf); - } else { - struct task_struct *me = current; - if (used_math()) { - clear_fpu(me); - clear_used_math(); - } - } + err |= restore_i387_xstate(buf); } err |= __get_user(*pax, &sc->ax); return err; - -badframe: - return 1; } asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) @@ -276,7 +199,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, struct pt_regs * regs) { struct rt_sigframe __user *frame; - struct _fpstate __user *fp = NULL; + void __user *fp = NULL; int err = 0; struct task_struct *me = current; @@ -288,7 +211,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, if (!access_ok(VERIFY_WRITE, fp, sizeof(struct _fpstate))) goto give_sigsegv; - if (save_i387(fp) < 0) + if (save_i387_xstate(fp) < 0) err |= -1; } else frame = get_stack(ka, regs, sizeof(struct rt_sigframe)) - 8; diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c index 7ad169e3352..608e72d7ca6 100644 --- a/arch/x86/kernel/xsave.c +++ b/arch/x86/kernel/xsave.c @@ -12,6 +12,85 @@ */ unsigned int pcntxt_hmask, pcntxt_lmask; +#ifdef CONFIG_X86_64 +/* + * Signal frame handlers. + */ + +int save_i387_xstate(void __user *buf) +{ + struct task_struct *tsk = current; + int err = 0; + + if (!access_ok(VERIFY_WRITE, buf, sig_xstate_size)) + return -EACCES; + + BUILD_BUG_ON(sizeof(struct user_i387_struct) != + sizeof(tsk->thread.xstate->fxsave)); + + if ((unsigned long)buf % 16) + printk("save_i387_xstate: bad fpstate %p\n", buf); + + if (!used_math()) + return 0; + clear_used_math(); /* trigger finit */ + if (task_thread_info(tsk)->status & TS_USEDFPU) { + err = save_i387_checking((struct i387_fxsave_struct __user *) + buf); + if (err) + return err; + task_thread_info(tsk)->status &= ~TS_USEDFPU; + stts(); + } else { + if (__copy_to_user(buf, &tsk->thread.xstate->fxsave, + xstate_size)) + return -1; + } + return 1; +} + +/* + * This restores directly out of user space. Exceptions are handled. + */ +int restore_i387_xstate(void __user *buf) +{ + struct task_struct *tsk = current; + int err; + + if (!buf) { + if (used_math()) { + clear_fpu(tsk); + clear_used_math(); + } + + return 0; + } else + if (!access_ok(VERIFY_READ, buf, sig_xstate_size)) + return -EACCES; + + if (!used_math()) { + err = init_fpu(tsk); + if (err) + return err; + } + + if (!(task_thread_info(current)->status & TS_USEDFPU)) { + clts(); + task_thread_info(current)->status |= TS_USEDFPU; + } + err = fxrstor_checking((__force struct i387_fxsave_struct *)buf); + if (unlikely(err)) { + /* + * Encountered an error while doing the restore from the + * user buffer, clear the fpu state. + */ + clear_fpu(tsk); + clear_used_math(); + } + return err; +} +#endif + /* * Represents init state for the supported extended state. */ diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h index 36dca8db166..dc3745e8040 100644 --- a/include/asm-x86/i387.h +++ b/include/asm-x86/i387.h @@ -34,8 +34,9 @@ extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set; #ifdef CONFIG_IA32_EMULATION extern unsigned int sig_xstate_ia32_size; struct _fpstate_ia32; -extern int save_i387_ia32(struct _fpstate_ia32 __user *buf); -extern int restore_i387_ia32(struct _fpstate_ia32 __user *buf); +struct _xstate_ia32; +extern int save_i387_xstate_ia32(void __user *buf); +extern int restore_i387_xstate_ia32(void __user *buf); #endif #define X87_FSW_ES (1 << 7) /* Exception Summary */ @@ -249,13 +250,13 @@ end: task_thread_info(tsk)->status &= ~TS_USEDFPU; } +#endif /* CONFIG_X86_64 */ + /* * Signal frame handlers... */ -extern int save_i387(struct _fpstate __user *buf); -extern int restore_i387(struct _fpstate __user *buf); - -#endif /* CONFIG_X86_64 */ +extern int save_i387_xstate(void __user *buf); +extern int restore_i387_xstate(void __user *buf); static inline void __unlazy_fpu(struct task_struct *tsk) { -- cgit v1.2.3-70-g09d2 From 9dc89c0f96a6ce6a1b7f9a47dd8bf6f17e2002c9 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 29 Jul 2008 10:29:23 -0700 Subject: x86, xsave: xsave/xrstor specific routines Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- include/asm-x86/xsave.h | 52 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/xsave.h b/include/asm-x86/xsave.h index e835a917ee1..b716511aede 100644 --- a/include/asm-x86/xsave.h +++ b/include/asm-x86/xsave.h @@ -48,6 +48,58 @@ static inline int xrstor_checking(struct xsave_struct *fx) return err; } +static inline int xsave_check(struct xsave_struct __user *buf) +{ + int err; + __asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x27\n" + "2:\n" + ".section .fixup,\"ax\"\n" + "3: movl $-1,%[err]\n" + " jmp 2b\n" + ".previous\n" + ".section __ex_table,\"a\"\n" + _ASM_ALIGN "\n" + _ASM_PTR "1b,3b\n" + ".previous" + : [err] "=r" (err) + : "D" (buf), "a" (-1), "d" (-1), "0" (0) + : "memory"); + if (unlikely(err) && __clear_user(buf, xstate_size)) + err = -EFAULT; + /* No need to clear here because the caller clears USED_MATH */ + return err; +} + +static inline int xrestore_user(struct xsave_struct __user *buf, + unsigned int lmask, + unsigned int hmask) +{ + int err; + struct xsave_struct *xstate = ((__force struct xsave_struct *)buf); + + __asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x2f\n" + "2:\n" + ".section .fixup,\"ax\"\n" + "3: movl $-1,%[err]\n" + " jmp 2b\n" + ".previous\n" + ".section __ex_table,\"a\"\n" + _ASM_ALIGN "\n" + _ASM_PTR "1b,3b\n" + ".previous" + : [err] "=r" (err) + : "D" (xstate), "a" (lmask), "d" (hmask), "0" (0) + : "memory"); /* memory required? */ + return err; +} + +static inline void xrstor_state(struct xsave_struct *fx, int lmask, int hmask) +{ + asm volatile(".byte " REX_PREFIX "0x0f,0xae,0x2f\n\t" + : : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask) + : "memory"); +} + static inline void xsave(struct task_struct *tsk) { /* This, however, we can work around by forcing the compiler to select -- cgit v1.2.3-70-g09d2 From bdd8caba5ed5bb7ee21c9f061597284ffeb280bf Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 29 Jul 2008 10:29:24 -0700 Subject: x86, xsave: struct _fpstate extensions to include extended state information Bytes 464..511 in the current 512byte layout of fxsave/fxrstor frame, are reserved for SW usage. On cpu's supporting xsave/xrstor, these bytes are used to extended the fpstate pointer in the sigcontext, which now includes the extended state information along with fpstate information. Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- include/asm-x86/processor.h | 7 +++- include/asm-x86/sigcontext.h | 87 ++++++++++++++++++++++++++++++++++++++++-- include/asm-x86/sigcontext32.h | 6 ++- 3 files changed, 94 insertions(+), 6 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index 77b7af6b573..eb4bd8c0773 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -322,7 +322,12 @@ struct i387_fxsave_struct { /* 16*16 bytes for each XMM-reg = 256 bytes: */ u32 xmm_space[64]; - u32 padding[24]; + u32 padding[12]; + + union { + u32 padding1[12]; + u32 sw_reserved[12]; + }; } __attribute__((aligned(16))); diff --git a/include/asm-x86/sigcontext.h b/include/asm-x86/sigcontext.h index 24879c85b29..899fe2f8abb 100644 --- a/include/asm-x86/sigcontext.h +++ b/include/asm-x86/sigcontext.h @@ -4,6 +4,40 @@ #include #include +#define FP_XSTATE_MAGIC1 0x46505853U +#define FP_XSTATE_MAGIC2 0x46505845U +#define FP_XSTATE_MAGIC2_SIZE sizeof(FP_XSTATE_MAGIC2) + +/* + * bytes 464..511 in the current 512byte layout of fxsave/fxrstor frame + * are reserved for SW usage. On cpu's supporting xsave/xrstor, these bytes + * are used to extended the fpstate pointer in the sigcontext, which now + * includes the extended state information along with fpstate information. + * + * Presence of FP_XSTATE_MAGIC1 at the beginning of this SW reserved + * area and FP_XSTATE_MAGIC2 at the end of memory layout + * (extended_size - FP_XSTATE_MAGIC2_SIZE) indicates the presence of the + * extended state information in the memory layout pointed by the fpstate + * pointer in sigcontext. + */ +struct _fpx_sw_bytes { + __u32 magic1; /* FP_XSTATE_MAGIC1 */ + __u32 extended_size; /* total size of the layout referred by + * fpstate pointer in the sigcontext. + */ + __u64 xstate_bv; + /* feature bit mask (including fp/sse/extended + * state) that is present in the memory + * layout. + */ + __u32 xstate_size; /* actual xsave state size, based on the + * features saved in the layout. + * 'extended_size' will be greater than + * 'xstate_size'. + */ + __u32 padding[7]; /* for future use. */ +}; + #ifdef __i386__ /* * As documented in the iBCS2 standard.. @@ -53,7 +87,13 @@ struct _fpstate { unsigned long reserved; struct _fpxreg _fxsr_st[8]; /* FXSR FPU reg data is ignored */ struct _xmmreg _xmm[8]; - unsigned long padding[56]; + unsigned long padding1[44]; + + union { + unsigned long padding2[12]; + struct _fpx_sw_bytes sw_reserved; /* represents the extended + * state info */ + }; }; #define X86_FXSR_MAGIC 0x0000 @@ -79,7 +119,15 @@ struct sigcontext { unsigned long flags; unsigned long sp_at_signal; unsigned short ss, __ssh; - struct _fpstate __user *fpstate; + + /* + * fpstate is really (struct _fpstate *) or (struct _xstate *) + * depending on the FP_XSTATE_MAGIC1 encoded in the SW reserved + * bytes of (struct _fpstate) and FP_XSTATE_MAGIC2 present at the end + * of extended memory layout. See comments at the defintion of + * (struct _fpx_sw_bytes) + */ + void __user *fpstate; /* zero when no FPU/extended context */ unsigned long oldmask; unsigned long cr2; }; @@ -130,7 +178,12 @@ struct _fpstate { __u32 mxcsr_mask; __u32 st_space[32]; /* 8*16 bytes for each FP-reg */ __u32 xmm_space[64]; /* 16*16 bytes for each XMM-reg */ - __u32 reserved2[24]; + __u32 reserved2[12]; + union { + __u32 reserved3[12]; + struct _fpx_sw_bytes sw_reserved; /* represents the extended + * state information */ + }; }; #ifdef __KERNEL__ @@ -161,7 +214,15 @@ struct sigcontext { unsigned long trapno; unsigned long oldmask; unsigned long cr2; - struct _fpstate __user *fpstate; /* zero when no FPU context */ + + /* + * fpstate is really (struct _fpstate *) or (struct _xstate *) + * depending on the FP_XSTATE_MAGIC1 encoded in the SW reserved + * bytes of (struct _fpstate) and FP_XSTATE_MAGIC2 present at the end + * of extended memory layout. See comments at the defintion of + * (struct _fpx_sw_bytes) + */ + void __user *fpstate; /* zero when no FPU/extended context */ unsigned long reserved1[8]; }; #else /* __KERNEL__ */ @@ -202,4 +263,22 @@ struct sigcontext { #endif /* !__i386__ */ +struct _xsave_hdr { + u64 xstate_bv; + u64 reserved1[2]; + u64 reserved2[5]; +}; + +/* + * Extended state pointed by the fpstate pointer in the sigcontext. + * In addition to the fpstate, information encoded in the xstate_hdr + * indicates the presence of other extended state information + * supported by the processor and OS. + */ +struct _xstate { + struct _fpstate fpstate; + struct _xsave_hdr xstate_hdr; + /* new processor state extensions go here */ +}; + #endif /* ASM_X86__SIGCONTEXT_H */ diff --git a/include/asm-x86/sigcontext32.h b/include/asm-x86/sigcontext32.h index 4e2ec732dd0..8c347032c2f 100644 --- a/include/asm-x86/sigcontext32.h +++ b/include/asm-x86/sigcontext32.h @@ -40,7 +40,11 @@ struct _fpstate_ia32 { __u32 reserved; struct _fpxreg _fxsr_st[8]; struct _xmmreg _xmm[8]; /* It's actually 16 */ - __u32 padding[56]; + __u32 padding[44]; + union { + __u32 padding2[12]; + struct _fpx_sw_bytes sw_reserved; + }; }; struct sigcontext_ia32 { -- cgit v1.2.3-70-g09d2 From c37b5efea43f9e500363f9973dd00e3d2cdcc685 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 29 Jul 2008 10:29:25 -0700 Subject: x86, xsave: save/restore the extended state context in sigframe On cpu's supporting xsave/xrstor, fpstate pointer in the sigcontext, will include the extended state information along with fpstate information. Presence of extended state information is indicated by the presence of FP_XSTATE_MAGIC1 at fpstate.sw_reserved.magic1 and FP_XSTATE_MAGIC2 at fpstate + (fpstate.sw_reserved.extended_size - FP_XSTATE_MAGIC2_SIZE). Extended feature bit mask that is saved in the memory layout is represented by the fpstate.sw_reserved.xstate_bv For RT signal frames, UC_FP_XSTATE in the uc_flags also indicate the presence of extended state information in the sigcontext's fpstate pointer. Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- arch/x86/ia32/ia32_signal.c | 5 +- arch/x86/kernel/i387.c | 82 +++++++++++++++++++-- arch/x86/kernel/signal_32.c | 5 +- arch/x86/kernel/signal_64.c | 7 +- arch/x86/kernel/xsave.c | 172 +++++++++++++++++++++++++++++++++++++++++--- include/asm-x86/i387.h | 4 +- include/asm-x86/ucontext.h | 6 ++ include/asm-x86/xsave.h | 5 +- 8 files changed, 264 insertions(+), 22 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c index c596eabbe98..f25a1012400 100644 --- a/arch/x86/ia32/ia32_signal.c +++ b/arch/x86/ia32/ia32_signal.c @@ -544,7 +544,10 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, goto give_sigsegv; /* Create the ucontext. */ - err |= __put_user(0, &frame->uc.uc_flags); + if (cpu_has_xsave) + err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags); + else + err |= __put_user(0, &frame->uc.uc_flags); err |= __put_user(0, &frame->uc.uc_link); err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp); err |= __put_user(sas_ss_flags(regs->sp), diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 7daf3a011dd..cbb9dc474a2 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -26,6 +26,7 @@ # define _fpstate_ia32 _fpstate # define _xstate_ia32 _xstate # define sig_xstate_ia32_size sig_xstate_size +# define fx_sw_reserved_ia32 fx_sw_reserved # define user_i387_ia32_struct user_i387_struct # define user32_fxsr_struct user_fxsr_struct #endif @@ -447,12 +448,30 @@ static int save_i387_fxsave(struct _fpstate_ia32 __user *buf) if (err) return -1; - if (__copy_to_user(&buf->_fxsr_env[0], fx, - sizeof(struct i387_fxsave_struct))) + if (__copy_to_user(&buf->_fxsr_env[0], fx, xstate_size)) return -1; return 1; } +static int save_i387_xsave(void __user *buf) +{ + struct _fpstate_ia32 __user *fx = buf; + int err = 0; + + if (save_i387_fxsave(fx) < 0) + return -1; + + err = __copy_to_user(&fx->sw_reserved, &fx_sw_reserved_ia32, + sizeof(struct _fpx_sw_bytes)); + err |= __put_user(FP_XSTATE_MAGIC2, + (__u32 __user *) (buf + sig_xstate_ia32_size + - FP_XSTATE_MAGIC2_SIZE)); + if (err) + return -1; + + return 1; +} + int save_i387_xstate_ia32(void __user *buf) { struct _fpstate_ia32 __user *fp = (struct _fpstate_ia32 __user *) buf; @@ -477,6 +496,8 @@ int save_i387_xstate_ia32(void __user *buf) unlazy_fpu(tsk); + if (cpu_has_xsave) + return save_i387_xsave(fp); if (cpu_has_fxsr) return save_i387_fxsave(fp); else @@ -491,14 +512,15 @@ static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf) sizeof(struct i387_fsave_struct)); } -static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf) +static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf, + unsigned int size) { struct task_struct *tsk = current; struct user_i387_ia32_struct env; int err; err = __copy_from_user(&tsk->thread.xstate->fxsave, &buf->_fxsr_env[0], - sizeof(struct i387_fxsave_struct)); + size); /* mxcsr reserved bits must be masked to zero for security reasons */ tsk->thread.xstate->fxsave.mxcsr &= mxcsr_feature_mask; if (err || __copy_from_user(&env, buf, sizeof(env))) @@ -508,6 +530,51 @@ static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf) return 0; } +static int restore_i387_xsave(void __user *buf) +{ + struct _fpx_sw_bytes fx_sw_user; + struct _fpstate_ia32 __user *fx_user = + ((struct _fpstate_ia32 __user *) buf); + struct i387_fxsave_struct __user *fx = + (struct i387_fxsave_struct __user *) &fx_user->_fxsr_env[0]; + struct xsave_hdr_struct *xsave_hdr = + ¤t->thread.xstate->xsave.xsave_hdr; + unsigned int lmask, hmask; + int err; + + if (check_for_xstate(fx, buf, &fx_sw_user)) + goto fx_only; + + lmask = fx_sw_user.xstate_bv; + hmask = fx_sw_user.xstate_bv >> 32; + + err = restore_i387_fxsave(buf, fx_sw_user.xstate_size); + + xsave_hdr->xstate_bv &= (pcntxt_lmask | (((u64) pcntxt_hmask) << 32)); + /* + * These bits must be zero. + */ + xsave_hdr->reserved1[0] = xsave_hdr->reserved1[1] = 0; + + /* + * Init the state that is not present in the memory layout + * and enabled by the OS. + */ + lmask = ~(pcntxt_lmask & ~lmask); + hmask = ~(pcntxt_hmask & ~hmask); + xsave_hdr->xstate_bv &= (lmask | (((u64) hmask) << 32)); + + return err; +fx_only: + /* + * Couldn't find the extended state information in the memory + * layout. Restore the FP/SSE and init the other extended state + * enabled by the OS. + */ + xsave_hdr->xstate_bv = XSTATE_FPSSE; + return restore_i387_fxsave(buf, sizeof(struct i387_fxsave_struct)); +} + int restore_i387_xstate_ia32(void __user *buf) { int err; @@ -535,8 +602,11 @@ int restore_i387_xstate_ia32(void __user *buf) } if (HAVE_HWFP) { - if (cpu_has_fxsr) - err = restore_i387_fxsave(fp); + if (cpu_has_xsave) + err = restore_i387_xsave(buf); + else if (cpu_has_fxsr) + err = restore_i387_fxsave(fp, sizeof(struct + i387_fxsave_struct)); else err = restore_i387_fsave(fp); } else { diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c index 690cc616ac0..0f98d69fbdb 100644 --- a/arch/x86/kernel/signal_32.c +++ b/arch/x86/kernel/signal_32.c @@ -441,7 +441,10 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, goto give_sigsegv; /* Create the ucontext. */ - err |= __put_user(0, &frame->uc.uc_flags); + if (cpu_has_xsave) + err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags); + else + err |= __put_user(0, &frame->uc.uc_flags); err |= __put_user(0, &frame->uc.uc_link); err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp); err |= __put_user(sas_ss_flags(regs->sp), diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c index ddf6123a55c..2621b98f5bf 100644 --- a/arch/x86/kernel/signal_64.c +++ b/arch/x86/kernel/signal_64.c @@ -192,7 +192,7 @@ get_stack(struct k_sigaction *ka, struct pt_regs *regs, unsigned long size) sp = current->sas_ss_sp + current->sas_ss_size; } - return (void __user *)round_down(sp - size, 16); + return (void __user *)round_down(sp - size, 64); } static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, @@ -226,7 +226,10 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, } /* Create the ucontext. */ - err |= __put_user(0, &frame->uc.uc_flags); + if (cpu_has_xsave) + err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags); + else + err |= __put_user(0, &frame->uc.uc_flags); err |= __put_user(0, &frame->uc.uc_link); err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp); err |= __put_user(sas_ss_flags(regs->sp), diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c index 608e72d7ca6..dd66d0714c1 100644 --- a/arch/x86/kernel/xsave.c +++ b/arch/x86/kernel/xsave.c @@ -6,12 +6,68 @@ #include #include #include +#ifdef CONFIG_IA32_EMULATION +#include +#endif /* * Supported feature mask by the CPU and the kernel. */ unsigned int pcntxt_hmask, pcntxt_lmask; +struct _fpx_sw_bytes fx_sw_reserved; +#ifdef CONFIG_IA32_EMULATION +struct _fpx_sw_bytes fx_sw_reserved_ia32; +#endif + +/* + * Check for the presence of extended state information in the + * user fpstate pointer in the sigcontext. + */ +int check_for_xstate(struct i387_fxsave_struct __user *buf, + void __user *fpstate, + struct _fpx_sw_bytes *fx_sw_user) +{ + int min_xstate_size = sizeof(struct i387_fxsave_struct) + + sizeof(struct xsave_hdr_struct); + unsigned int magic2; + int err; + + err = __copy_from_user(fx_sw_user, &buf->sw_reserved[0], + sizeof(struct _fpx_sw_bytes)); + + if (err) + return err; + + /* + * First Magic check failed. + */ + if (fx_sw_user->magic1 != FP_XSTATE_MAGIC1) + return -1; + + /* + * Check for error scenarios. + */ + if (fx_sw_user->xstate_size < min_xstate_size || + fx_sw_user->xstate_size > xstate_size || + fx_sw_user->xstate_size > fx_sw_user->extended_size) + return -1; + + err = __get_user(magic2, (__u32 *) (((void *)fpstate) + + fx_sw_user->extended_size - + FP_XSTATE_MAGIC2_SIZE)); + /* + * Check for the presence of second magic word at the end of memory + * layout. This detects the case where the user just copied the legacy + * fpstate layout with out copying the extended state information + * in the memory layout. + */ + if (err || magic2 != FP_XSTATE_MAGIC2) + return -1; + + return 0; +} + #ifdef CONFIG_X86_64 /* * Signal frame handlers. @@ -28,15 +84,18 @@ int save_i387_xstate(void __user *buf) BUILD_BUG_ON(sizeof(struct user_i387_struct) != sizeof(tsk->thread.xstate->fxsave)); - if ((unsigned long)buf % 16) + if ((unsigned long)buf % 64) printk("save_i387_xstate: bad fpstate %p\n", buf); if (!used_math()) return 0; clear_used_math(); /* trigger finit */ if (task_thread_info(tsk)->status & TS_USEDFPU) { - err = save_i387_checking((struct i387_fxsave_struct __user *) - buf); + if (task_thread_info(tsk)->status & TS_XSAVE) + err = xsave_user(buf); + else + err = fxsave_user(buf); + if (err) return err; task_thread_info(tsk)->status &= ~TS_USEDFPU; @@ -46,23 +105,77 @@ int save_i387_xstate(void __user *buf) xstate_size)) return -1; } + + if (task_thread_info(tsk)->status & TS_XSAVE) { + struct _fpstate __user *fx = buf; + + err = __copy_to_user(&fx->sw_reserved, &fx_sw_reserved, + sizeof(struct _fpx_sw_bytes)); + + err |= __put_user(FP_XSTATE_MAGIC2, + (__u32 __user *) (buf + sig_xstate_size + - FP_XSTATE_MAGIC2_SIZE)); + } + return 1; } +/* + * Restore the extended state if present. Otherwise, restore the FP/SSE + * state. + */ +int restore_user_xstate(void __user *buf) +{ + struct _fpx_sw_bytes fx_sw_user; + unsigned int lmask, hmask; + int err; + + if (((unsigned long)buf % 64) || + check_for_xstate(buf, buf, &fx_sw_user)) + goto fx_only; + + lmask = fx_sw_user.xstate_bv; + hmask = fx_sw_user.xstate_bv >> 32; + + /* + * restore the state passed by the user. + */ + err = xrestore_user(buf, lmask, hmask); + if (err) + return err; + + /* + * init the state skipped by the user. + */ + lmask = pcntxt_lmask & ~lmask; + hmask = pcntxt_hmask & ~hmask; + + xrstor_state(init_xstate_buf, lmask, hmask); + + return 0; + +fx_only: + /* + * couldn't find the extended state information in the + * memory layout. Restore just the FP/SSE and init all + * the other extended state. + */ + xrstor_state(init_xstate_buf, pcntxt_lmask & ~XSTATE_FPSSE, + pcntxt_hmask); + return fxrstor_checking((__force struct i387_fxsave_struct *)buf); +} + /* * This restores directly out of user space. Exceptions are handled. */ int restore_i387_xstate(void __user *buf) { struct task_struct *tsk = current; - int err; + int err = 0; if (!buf) { - if (used_math()) { - clear_fpu(tsk); - clear_used_math(); - } - + if (used_math()) + goto clear; return 0; } else if (!access_ok(VERIFY_READ, buf, sig_xstate_size)) @@ -78,12 +191,17 @@ int restore_i387_xstate(void __user *buf) clts(); task_thread_info(current)->status |= TS_USEDFPU; } - err = fxrstor_checking((__force struct i387_fxsave_struct *)buf); + if (task_thread_info(tsk)->status & TS_XSAVE) + err = restore_user_xstate(buf); + else + err = fxrstor_checking((__force struct i387_fxsave_struct *) + buf); if (unlikely(err)) { /* * Encountered an error while doing the restore from the * user buffer, clear the fpu state. */ +clear: clear_fpu(tsk); clear_used_math(); } @@ -91,6 +209,38 @@ int restore_i387_xstate(void __user *buf) } #endif +/* + * Prepare the SW reserved portion of the fxsave memory layout, indicating + * the presence of the extended state information in the memory layout + * pointed by the fpstate pointer in the sigcontext. + * This will be saved when ever the FP and extended state context is + * saved on the user stack during the signal handler delivery to the user. + */ +void prepare_fx_sw_frame(void) +{ + int size_extended = (xstate_size - sizeof(struct i387_fxsave_struct)) + + FP_XSTATE_MAGIC2_SIZE; + + sig_xstate_size = sizeof(struct _fpstate) + size_extended; + +#ifdef CONFIG_IA32_EMULATION + sig_xstate_ia32_size = sizeof(struct _fpstate_ia32) + size_extended; +#endif + + memset(&fx_sw_reserved, 0, sizeof(fx_sw_reserved)); + + fx_sw_reserved.magic1 = FP_XSTATE_MAGIC1; + fx_sw_reserved.extended_size = sig_xstate_size; + fx_sw_reserved.xstate_bv = pcntxt_lmask | + (((u64) (pcntxt_hmask)) << 32); + fx_sw_reserved.xstate_size = xstate_size; +#ifdef CONFIG_IA32_EMULATION + memcpy(&fx_sw_reserved_ia32, &fx_sw_reserved, + sizeof(struct _fpx_sw_bytes)); + fx_sw_reserved_ia32.extended_size = sig_xstate_ia32_size; +#endif +} + /* * Represents init state for the supported extended state. */ @@ -162,6 +312,8 @@ void __init xsave_cntxt_init(void) xstate_size = ebx; + prepare_fx_sw_frame(); + setup_xstate_init(); printk(KERN_INFO "xsave/xrstor: enabled xstate_bv 0x%Lx, " diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h index dc3745e8040..d3dda716195 100644 --- a/include/asm-x86/i387.h +++ b/include/asm-x86/i387.h @@ -31,8 +31,10 @@ extern user_regset_active_fn fpregs_active, xfpregs_active; extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get; extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set; +extern struct _fpx_sw_bytes fx_sw_reserved; #ifdef CONFIG_IA32_EMULATION extern unsigned int sig_xstate_ia32_size; +extern struct _fpx_sw_bytes fx_sw_reserved_ia32; struct _fpstate_ia32; struct _xstate_ia32; extern int save_i387_xstate_ia32(void __user *buf); @@ -104,7 +106,7 @@ static inline void clear_fpu_state(struct task_struct *tsk) X86_FEATURE_FXSAVE_LEAK); } -static inline int save_i387_checking(struct i387_fxsave_struct __user *fx) +static inline int fxsave_user(struct i387_fxsave_struct __user *fx) { int err; diff --git a/include/asm-x86/ucontext.h b/include/asm-x86/ucontext.h index 9948dd32808..89eaa5456a7 100644 --- a/include/asm-x86/ucontext.h +++ b/include/asm-x86/ucontext.h @@ -1,6 +1,12 @@ #ifndef ASM_X86__UCONTEXT_H #define ASM_X86__UCONTEXT_H +#define UC_FP_XSTATE 0x1 /* indicates the presence of extended state + * information in the memory layout pointed + * by the fpstate pointer in the ucontext's + * sigcontext struct (uc_mcontext). + */ + struct ucontext { unsigned long uc_flags; struct ucontext *uc_link; diff --git a/include/asm-x86/xsave.h b/include/asm-x86/xsave.h index b716511aede..b7f64b9fcd9 100644 --- a/include/asm-x86/xsave.h +++ b/include/asm-x86/xsave.h @@ -29,6 +29,9 @@ extern struct xsave_struct *init_xstate_buf; extern void xsave_cntxt_init(void); extern void xsave_init(void); extern int init_fpu(struct task_struct *child); +extern int check_for_xstate(struct i387_fxsave_struct __user *buf, + void __user *fpstate, + struct _fpx_sw_bytes *sw); static inline int xrstor_checking(struct xsave_struct *fx) { @@ -48,7 +51,7 @@ static inline int xrstor_checking(struct xsave_struct *fx) return err; } -static inline int xsave_check(struct xsave_struct __user *buf) +static inline int xsave_user(struct xsave_struct __user *buf) { int err; __asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x27\n" -- cgit v1.2.3-70-g09d2 From b4a091a62c8e91d6077e575600363cff73fa02ef Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Tue, 29 Jul 2008 17:30:29 -0700 Subject: x86, xsave: add header file for XCR registers Add header file for the XCR registers and their access functions. Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- include/asm-x86/xcr.h | 49 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 49 insertions(+) create mode 100644 include/asm-x86/xcr.h (limited to 'include/asm-x86') diff --git a/include/asm-x86/xcr.h b/include/asm-x86/xcr.h new file mode 100644 index 00000000000..f2cba4e79a2 --- /dev/null +++ b/include/asm-x86/xcr.h @@ -0,0 +1,49 @@ +/* -*- linux-c -*- ------------------------------------------------------- * + * + * Copyright 2008 rPath, Inc. - All Rights Reserved + * + * This file is part of the Linux kernel, and is made available under + * the terms of the GNU General Public License version 2 or (at your + * option) any later version; incorporated herein by reference. + * + * ----------------------------------------------------------------------- */ + +/* + * asm-x86/xcr.h + * + * Definitions for the eXtended Control Register instructions + */ + +#ifndef _ASM_X86_XCR_H +#define _ASM_X86_XCR_H + +#define XCR_XFEATURE_ENABLED_MASK 0x00000000 + +#ifdef __KERNEL__ +# ifndef __ASSEMBLY__ + +#include + +static inline u64 xgetbv(u32 index) +{ + u32 eax, edx; + + asm volatile(".byte 0x0f,0x01,0xd0" /* xgetbv */ + : "=a" (eax), "=d" (edx) + : "c" (index)); + return eax + ((u64)edx << 32); +} + +static inline void xsetbv(u32 index, u64 value) +{ + u32 eax = value; + u32 edx = value >> 32; + + asm volatile(".byte 0x0f,0x01,0xd1" /* xsetbv */ + : : "a" (eax), "d" (edx), "c" (index)); +} + +# endif /* __ASSEMBLY__ */ +#endif /* __KERNEL__ */ + +#endif /* _ASM_X86_XCR_H */ -- cgit v1.2.3-70-g09d2 From 6152e4b1c99a3689fc318d092cd144597f7dbd14 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Tue, 29 Jul 2008 17:23:16 -0700 Subject: x86, xsave: keep the XSAVE feature mask as an u64 The XSAVE feature mask is a 64-bit number; keep it that way, in order to avoid the mistake done with rdmsr/wrmsr. Use the xsetbv() function provided in the previous patch. Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- arch/x86/kernel/i387.c | 12 +++++------- arch/x86/kernel/xsave.c | 45 +++++++++++++++++---------------------------- include/asm-x86/xsave.h | 18 +++++++++++------- 3 files changed, 33 insertions(+), 42 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index e0ed59f5c19..45723f1fe19 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -552,18 +552,17 @@ static int restore_i387_xsave(void __user *buf) (struct i387_fxsave_struct __user *) &fx_user->_fxsr_env[0]; struct xsave_hdr_struct *xsave_hdr = ¤t->thread.xstate->xsave.xsave_hdr; - unsigned int lmask, hmask; + u64 mask; int err; if (check_for_xstate(fx, buf, &fx_sw_user)) goto fx_only; - lmask = fx_sw_user.xstate_bv; - hmask = fx_sw_user.xstate_bv >> 32; + mask = fx_sw_user.xstate_bv; err = restore_i387_fxsave(buf, fx_sw_user.xstate_size); - xsave_hdr->xstate_bv &= (pcntxt_lmask | (((u64) pcntxt_hmask) << 32)); + xsave_hdr->xstate_bv &= pcntxt_mask; /* * These bits must be zero. */ @@ -573,9 +572,8 @@ static int restore_i387_xsave(void __user *buf) * Init the state that is not present in the memory layout * and enabled by the OS. */ - lmask = ~(pcntxt_lmask & ~lmask); - hmask = ~(pcntxt_hmask & ~hmask); - xsave_hdr->xstate_bv &= (lmask | (((u64) hmask) << 32)); + mask = ~(pcntxt_mask & ~mask); + xsave_hdr->xstate_bv &= mask; return err; fx_only: diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c index dd66d0714c1..7415f3e38a5 100644 --- a/arch/x86/kernel/xsave.c +++ b/arch/x86/kernel/xsave.c @@ -9,11 +9,12 @@ #ifdef CONFIG_IA32_EMULATION #include #endif +#include /* * Supported feature mask by the CPU and the kernel. */ -unsigned int pcntxt_hmask, pcntxt_lmask; +u64 pcntxt_mask; struct _fpx_sw_bytes fx_sw_reserved; #ifdef CONFIG_IA32_EMULATION @@ -127,30 +128,28 @@ int save_i387_xstate(void __user *buf) int restore_user_xstate(void __user *buf) { struct _fpx_sw_bytes fx_sw_user; - unsigned int lmask, hmask; + u64 mask; int err; if (((unsigned long)buf % 64) || check_for_xstate(buf, buf, &fx_sw_user)) goto fx_only; - lmask = fx_sw_user.xstate_bv; - hmask = fx_sw_user.xstate_bv >> 32; + mask = fx_sw_user.xstate_bv; /* * restore the state passed by the user. */ - err = xrestore_user(buf, lmask, hmask); + err = xrestore_user(buf, mask); if (err) return err; /* * init the state skipped by the user. */ - lmask = pcntxt_lmask & ~lmask; - hmask = pcntxt_hmask & ~hmask; + mask = pcntxt_mask & ~mask; - xrstor_state(init_xstate_buf, lmask, hmask); + xrstor_state(init_xstate_buf, mask); return 0; @@ -160,8 +159,7 @@ fx_only: * memory layout. Restore just the FP/SSE and init all * the other extended state. */ - xrstor_state(init_xstate_buf, pcntxt_lmask & ~XSTATE_FPSSE, - pcntxt_hmask); + xrstor_state(init_xstate_buf, pcntxt_mask & ~XSTATE_FPSSE); return fxrstor_checking((__force struct i387_fxsave_struct *)buf); } @@ -231,8 +229,7 @@ void prepare_fx_sw_frame(void) fx_sw_reserved.magic1 = FP_XSTATE_MAGIC1; fx_sw_reserved.extended_size = sig_xstate_size; - fx_sw_reserved.xstate_bv = pcntxt_lmask | - (((u64) (pcntxt_hmask)) << 32); + fx_sw_reserved.xstate_bv = pcntxt_mask; fx_sw_reserved.xstate_size = xstate_size; #ifdef CONFIG_IA32_EMULATION memcpy(&fx_sw_reserved_ia32, &fx_sw_reserved, @@ -263,11 +260,8 @@ void __cpuinit xsave_init(void) /* * Enable all the features that the HW is capable of * and the Linux kernel is aware of. - * - * xsetbv(); */ - asm volatile(".byte 0x0f,0x01,0xd1" : : "c" (0), - "a" (pcntxt_lmask), "d" (pcntxt_hmask)); + xsetbv(XCR_XFEATURE_ENABLED_MASK, pcntxt_mask); } /* @@ -287,36 +281,31 @@ void __init xsave_cntxt_init(void) unsigned int eax, ebx, ecx, edx; cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx); + pcntxt_mask = eax + ((u64)edx << 32); - pcntxt_lmask = eax; - pcntxt_hmask = edx; - - if ((pcntxt_lmask & XSTATE_FPSSE) != XSTATE_FPSSE) { - printk(KERN_ERR "FP/SSE not shown under xsave features %x\n", - pcntxt_lmask); + if ((pcntxt_mask & XSTATE_FPSSE) != XSTATE_FPSSE) { + printk(KERN_ERR "FP/SSE not shown under xsave features 0x%llx\n", + pcntxt_mask); BUG(); } /* * for now OS knows only about FP/SSE */ - pcntxt_lmask = pcntxt_lmask & XCNTXT_LMASK; - pcntxt_hmask = pcntxt_hmask & XCNTXT_HMASK; - + pcntxt_mask = pcntxt_mask & XCNTXT_MASK; xsave_init(); /* * Recompute the context size for enabled features */ cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx); - xstate_size = ebx; prepare_fx_sw_frame(); setup_xstate_init(); - printk(KERN_INFO "xsave/xrstor: enabled xstate_bv 0x%Lx, " + printk(KERN_INFO "xsave/xrstor: enabled xstate_bv 0x%llx, " "cntxt size 0x%x\n", - (pcntxt_lmask | ((u64) pcntxt_hmask << 32)), xstate_size); + pcntxt_mask, xstate_size); } diff --git a/include/asm-x86/xsave.h b/include/asm-x86/xsave.h index b7f64b9fcd9..08e9a1ac07a 100644 --- a/include/asm-x86/xsave.h +++ b/include/asm-x86/xsave.h @@ -1,6 +1,7 @@ #ifndef __ASM_X86_XSAVE_H #define __ASM_X86_XSAVE_H +#include #include #include @@ -14,8 +15,7 @@ /* * These are the features that the OS can handle currently. */ -#define XCNTXT_LMASK (XSTATE_FP | XSTATE_SSE) -#define XCNTXT_HMASK 0x0 +#define XCNTXT_MASK (XSTATE_FP | XSTATE_SSE) #ifdef CONFIG_X86_64 #define REX_PREFIX "0x48, " @@ -23,7 +23,8 @@ #define REX_PREFIX #endif -extern unsigned int xstate_size, pcntxt_hmask, pcntxt_lmask; +extern unsigned int xstate_size; +extern u64 pcntxt_mask; extern struct xsave_struct *init_xstate_buf; extern void xsave_cntxt_init(void); @@ -73,12 +74,12 @@ static inline int xsave_user(struct xsave_struct __user *buf) return err; } -static inline int xrestore_user(struct xsave_struct __user *buf, - unsigned int lmask, - unsigned int hmask) +static inline int xrestore_user(struct xsave_struct __user *buf, u64 mask) { int err; struct xsave_struct *xstate = ((__force struct xsave_struct *)buf); + u32 lmask = mask; + u32 hmask = mask >> 32; __asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x2f\n" "2:\n" @@ -96,8 +97,11 @@ static inline int xrestore_user(struct xsave_struct __user *buf, return err; } -static inline void xrstor_state(struct xsave_struct *fx, int lmask, int hmask) +static inline void xrstor_state(struct xsave_struct *fx, u64 mask) { + u32 lmask = mask; + u32 hmask = mask >> 32; + asm volatile(".byte " REX_PREFIX "0x0f,0xae,0x2f\n\t" : : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask) : "memory"); -- cgit v1.2.3-70-g09d2 From bafc1dae8215c862c2e6ae913ddadc20581e59b9 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Mon, 11 Aug 2008 00:11:13 +0200 Subject: x86: mmconf: fix section mismatch warning WARNING: arch/x86/kernel/built-in.o(.cpuinit.text+0x1591): Section mismatch in reference from the function init_amd() to the function .init.text:check_enable_amd_mmconf_dmi() The function __cpuinit init_amd() references a function __init check_enable_amd_mmconf_dmi(). If check_enable_amd_mmconf_dmi is only used by init_amd then annotate check_enable_amd_mmconf_dmi with a matching annotation. check_enable_amd_mmconf_dmi is only called from init_amd which is __cpuinit Signed-off-by: Marcin Slusarz Cc: Thomas Gleixner Cc: Ingo Molnar Cc: H. Peter Anvin Signed-off-by: H. Peter Anvin --- arch/x86/kernel/mmconf-fam10h_64.c | 2 +- include/asm-x86/mmconfig.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/mmconf-fam10h_64.c b/arch/x86/kernel/mmconf-fam10h_64.c index fdfdc550b36..efc2f361fe8 100644 --- a/arch/x86/kernel/mmconf-fam10h_64.c +++ b/arch/x86/kernel/mmconf-fam10h_64.c @@ -238,7 +238,7 @@ static struct dmi_system_id __devinitdata mmconf_dmi_table[] = { {} }; -void __init check_enable_amd_mmconf_dmi(void) +void __cpuinit check_enable_amd_mmconf_dmi(void) { dmi_check_system(mmconf_dmi_table); } diff --git a/include/asm-x86/mmconfig.h b/include/asm-x86/mmconfig.h index 95beda07c6f..e293ab81e85 100644 --- a/include/asm-x86/mmconfig.h +++ b/include/asm-x86/mmconfig.h @@ -3,7 +3,7 @@ #ifdef CONFIG_PCI_MMCONFIG extern void __cpuinit fam10h_check_enable_mmcfg(void); -extern void __init check_enable_amd_mmconf_dmi(void); +extern void __cpuinit check_enable_amd_mmconf_dmi(void); #else static inline void fam10h_check_enable_mmcfg(void) { } static inline void check_enable_amd_mmconf_dmi(void) { } -- cgit v1.2.3-70-g09d2 From 2ae111cdd8d83ebf9de72e36e68a8c84b6ebbeea Mon Sep 17 00:00:00 2001 From: Cyrill Gorcunov Date: Mon, 11 Aug 2008 18:34:08 +0400 Subject: x86: apic interrupts - move assignments to irqinit_32.c, v2 64bit mode APIC interrupt handlers are set within irqinit_64.c. Lets do tha same for 32bit mode which would help in furter code merging. Signed-off-by: Cyrill Gorcunov Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic_32.c | 48 ------------------------------------------ arch/x86/kernel/irqinit_32.c | 49 +++++++++++++++++++++++++++++++++++++++++++ arch/x86/mach-default/setup.c | 15 ------------- include/asm-x86/arch_hooks.h | 2 -- 4 files changed, 49 insertions(+), 65 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c index f93c18f5b79..9e341c9d941 100644 --- a/arch/x86/kernel/apic_32.c +++ b/arch/x86/kernel/apic_32.c @@ -1354,54 +1354,6 @@ void smp_error_interrupt(struct pt_regs *regs) irq_exit(); } -#ifdef CONFIG_SMP -void __init smp_intr_init(void) -{ - /* - * IRQ0 must be given a fixed assignment and initialized, - * because it's used before the IO-APIC is set up. - */ - set_intr_gate(FIRST_DEVICE_VECTOR, interrupt[0]); - - /* - * The reschedule interrupt is a CPU-to-CPU reschedule-helper - * IPI, driven by wakeup. - */ - alloc_intr_gate(RESCHEDULE_VECTOR, reschedule_interrupt); - - /* IPI for invalidation */ - alloc_intr_gate(INVALIDATE_TLB_VECTOR, invalidate_interrupt); - - /* IPI for generic function call */ - alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt); - - /* IPI for single call function */ - set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR, - call_function_single_interrupt); -} -#endif - -/* - * Initialize APIC interrupts - */ -void __init apic_intr_init(void) -{ -#ifdef CONFIG_SMP - smp_intr_init(); -#endif - /* self generated IPI for local APIC timer */ - alloc_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt); - - /* IPI vectors for APIC spurious and error interrupts */ - alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt); - alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt); - - /* thermal monitor LVT interrupt */ -#ifdef CONFIG_X86_MCE_P4THERMAL - alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt); -#endif -} - /** * connect_bsp_APIC - attach the APIC to the interrupt system */ diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c index d66914287ee..9200a1e2752 100644 --- a/arch/x86/kernel/irqinit_32.c +++ b/arch/x86/kernel/irqinit_32.c @@ -74,6 +74,15 @@ void __init init_ISA_irqs (void) } } +/* + * IRQ2 is cascade interrupt to second interrupt controller + */ +static struct irqaction irq2 = { + .handler = no_action, + .mask = CPU_MASK_NONE, + .name = "cascade", +}; + /* Overridden in paravirt.c */ void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ"))); @@ -98,6 +107,46 @@ void __init native_init_IRQ(void) set_intr_gate(vector, interrupt[i]); } +#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_SMP) + /* + * IRQ0 must be given a fixed assignment and initialized, + * because it's used before the IO-APIC is set up. + */ + set_intr_gate(FIRST_DEVICE_VECTOR, interrupt[0]); + + /* + * The reschedule interrupt is a CPU-to-CPU reschedule-helper + * IPI, driven by wakeup. + */ + alloc_intr_gate(RESCHEDULE_VECTOR, reschedule_interrupt); + + /* IPI for invalidation */ + alloc_intr_gate(INVALIDATE_TLB_VECTOR, invalidate_interrupt); + + /* IPI for generic function call */ + alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt); + + /* IPI for single call function */ + set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR, call_function_single_interrupt); +#endif + +#ifdef CONFIG_X86_LOCAL_APIC + /* self generated IPI for local APIC timer */ + alloc_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt); + + /* IPI vectors for APIC spurious and error interrupts */ + alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt); + alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt); +#endif + +#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86_MCE_P4THERMAL) + /* thermal monitor LVT interrupt */ + alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt); +#endif + + if (!acpi_ioapic) + setup_irq(2, &irq2); + /* setup after call gates are initialised (usually add in * the architecture specific gates) */ diff --git a/arch/x86/mach-default/setup.c b/arch/x86/mach-default/setup.c index 3d317836be9..b00f5ad10ca 100644 --- a/arch/x86/mach-default/setup.c +++ b/arch/x86/mach-default/setup.c @@ -36,15 +36,6 @@ void __init pre_intr_init_hook(void) init_ISA_irqs(); } -/* - * IRQ2 is cascade interrupt to second interrupt controller - */ -static struct irqaction irq2 = { - .handler = no_action, - .mask = CPU_MASK_NONE, - .name = "cascade", -}; - /** * intr_init_hook - post gate setup interrupt initialisation * @@ -60,12 +51,6 @@ void __init intr_init_hook(void) if (x86_quirks->arch_intr_init()) return; } -#ifdef CONFIG_X86_LOCAL_APIC - apic_intr_init(); -#endif - - if (!acpi_ioapic) - setup_irq(2, &irq2); } /** diff --git a/include/asm-x86/arch_hooks.h b/include/asm-x86/arch_hooks.h index 72adc3a109c..de4596b24c2 100644 --- a/include/asm-x86/arch_hooks.h +++ b/include/asm-x86/arch_hooks.h @@ -12,8 +12,6 @@ /* these aren't arch hooks, they are generic routines * that can be used by the hooks */ extern void init_ISA_irqs(void); -extern void apic_intr_init(void); -extern void smp_intr_init(void); extern irqreturn_t timer_interrupt(int irq, void *dev_id); /* these are the defined hooks */ -- cgit v1.2.3-70-g09d2 From 26d809af6397ce5c37f5c44d89734d19cce1ad25 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 13 Aug 2008 12:46:26 +0200 Subject: x86: fix xsave build error MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit fix this build failure with certain glibc versions: In file included from /usr/include/bits/sigcontext.h:28, from /usr/include/signal.h:333, from Documentation/accounting/getdelays.c:24: /home/mingo/tip/usr/include/asm/sigcontext.h:191: error: expected specifier-qualifier-list before ‘u64’ Signed-off-by: Ingo Molnar --- include/asm-x86/sigcontext.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/sigcontext.h b/include/asm-x86/sigcontext.h index 899fe2f8abb..ee813f4fe5d 100644 --- a/include/asm-x86/sigcontext.h +++ b/include/asm-x86/sigcontext.h @@ -264,9 +264,9 @@ struct sigcontext { #endif /* !__i386__ */ struct _xsave_hdr { - u64 xstate_bv; - u64 reserved1[2]; - u64 reserved2[5]; + __u64 xstate_bv; + __u64 reserved1[2]; + __u64 reserved2[5]; }; /* -- cgit v1.2.3-70-g09d2 From 1ac2f7d55b7ee1613c90631e87fea22ec06781e5 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Mon, 4 Aug 2008 14:51:24 +0800 Subject: introduce two APIs for page attribute Introduce two APIs for page attribute. flushing tlb/cache in every page attribute is expensive. AGP gart usually will do a lot of operations to change a page to uc, new APIs can reduce flush. Signed-off-by: Shaohua Li Cc: airlied@linux.ie Cc: Andrew Morton Cc: Arjan van de Ven Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr.c | 58 ++++++++++++++++++++++++++++++++++++++------ include/asm-x86/cacheflush.h | 3 +++ 2 files changed, 53 insertions(+), 8 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 65c6e46bf05..2c5c18c2464 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -752,12 +752,12 @@ static inline int cache_attr(pgprot_t attr) (_PAGE_PAT | _PAGE_PAT_LARGE | _PAGE_PWT | _PAGE_PCD); } -static int change_page_attr_set_clr(unsigned long addr, int numpages, +static int do_change_page_attr_set_clr(unsigned long addr, int numpages, pgprot_t mask_set, pgprot_t mask_clr, - int force_split) + int force_split, int *tlb_flush) { struct cpa_data cpa; - int ret, cache, checkalias; + int ret, checkalias; /* * Check, if we are requested to change a not supported @@ -792,9 +792,22 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, /* * Check whether we really changed something: */ - if (!cpa.flushtlb) - goto out; + *tlb_flush = cpa.flushtlb; + cpa_fill_pool(NULL); + + return ret; +} + +static int change_page_attr_set_clr(unsigned long addr, int numpages, + pgprot_t mask_set, pgprot_t mask_clr, + int force_split) +{ + int cache, flush_cache = 0, ret; + ret = do_change_page_attr_set_clr(addr, numpages, mask_set, mask_clr, + force_split, &flush_cache); + if (!flush_cache) + goto out; /* * No need to flush, when we did not set any of the caching * attributes: @@ -811,10 +824,7 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, cpa_flush_range(addr, numpages, cache); else cpa_flush_all(cache); - out: - cpa_fill_pool(NULL); - return ret; } @@ -852,6 +862,30 @@ int set_memory_uc(unsigned long addr, int numpages) } EXPORT_SYMBOL(set_memory_uc); +int set_memory_uc_noflush(unsigned long addr, int numpages) +{ + int flush; + /* + * for now UC MINUS. see comments in ioremap_nocache() + */ + if (reserve_memtype(addr, addr + numpages * PAGE_SIZE, + _PAGE_CACHE_UC_MINUS, NULL)) + return -EINVAL; + /* + * for now UC MINUS. see comments in ioremap_nocache() + */ + return do_change_page_attr_set_clr(addr, numpages, + __pgprot(_PAGE_CACHE_UC_MINUS), + __pgprot(0), 0, &flush); +} +EXPORT_SYMBOL(set_memory_uc_noflush); + +void set_memory_flush_all(void) +{ + cpa_flush_all(1); +} +EXPORT_SYMBOL(set_memory_flush_all); + int _set_memory_wc(unsigned long addr, int numpages) { return change_page_attr_set(addr, numpages, @@ -926,6 +960,14 @@ int set_pages_uc(struct page *page, int numpages) } EXPORT_SYMBOL(set_pages_uc); +int set_pages_uc_noflush(struct page *page, int numpages) +{ + unsigned long addr = (unsigned long)page_address(page); + + return set_memory_uc_noflush(addr, numpages); +} +EXPORT_SYMBOL(set_pages_uc_noflush); + int set_pages_wb(struct page *page, int numpages) { unsigned long addr = (unsigned long)page_address(page); diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h index f4c0ab50d2c..57bac7b68c4 100644 --- a/include/asm-x86/cacheflush.h +++ b/include/asm-x86/cacheflush.h @@ -57,6 +57,8 @@ int _set_memory_uc(unsigned long addr, int numpages); int _set_memory_wc(unsigned long addr, int numpages); int _set_memory_wb(unsigned long addr, int numpages); int set_memory_uc(unsigned long addr, int numpages); +int set_memory_uc_noflush(unsigned long addr, int numpages); +void set_memory_flush_all(void); int set_memory_wc(unsigned long addr, int numpages); int set_memory_wb(unsigned long addr, int numpages); int set_memory_x(unsigned long addr, int numpages); @@ -87,6 +89,7 @@ int set_memory_4k(unsigned long addr, int numpages); */ int set_pages_uc(struct page *page, int numpages); +int set_pages_uc_noflush(struct page *page, int numpages); int set_pages_wb(struct page *page, int numpages); int set_pages_x(struct page *page, int numpages); int set_pages_nx(struct page *page, int numpages); -- cgit v1.2.3-70-g09d2 From 466ae837424dcc538b1af2a0eaf53be32edcdbe7 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Mon, 4 Aug 2008 14:51:30 +0800 Subject: reduce tlb/cache flush times of agpgart memory allocation To reduce tlb/cache flush, makes agp memory allocation do one flush after all pages in a region are changed to uc. All agp drivers except agp-sgi uses agp_generic_alloc_page() for .agp_alloc_page, so the patch should work for them. agp-sgi is only for ia64, so not a problem too. Signed-off-by: Shaohua Li Cc: airlied@linux.ie Cc: Andrew Morton Cc: Arjan van de Ven Signed-off-by: Ingo Molnar --- drivers/char/agp/agp.h | 4 ++++ drivers/char/agp/generic.c | 4 +++- include/asm-x86/agp.h | 3 +++ 3 files changed, 10 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/drivers/char/agp/agp.h b/drivers/char/agp/agp.h index 81e14bea54b..395168fb17e 100644 --- a/drivers/char/agp/agp.h +++ b/drivers/char/agp/agp.h @@ -30,6 +30,10 @@ #define _AGP_BACKEND_PRIV_H 1 #include /* for flush_agp_cache() */ +#ifndef map_page_into_agp_noflush +#define map_page_into_agp_noflush(page) map_page_into_agp(page) +#define map_page_into_agp_global_flush() +#endif #define PFX "agpgart: " diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c index eaa1a355bb3..bf239b8ecac 100644 --- a/drivers/char/agp/generic.c +++ b/drivers/char/agp/generic.c @@ -274,6 +274,7 @@ struct agp_memory *agp_allocate_memory(struct agp_bridge_data *bridge, new->memory[i] = virt_to_gart(addr); new->page_count++; } + map_page_into_agp_global_flush(); new->bridge = bridge; return new; @@ -1186,7 +1187,8 @@ void *agp_generic_alloc_page(struct agp_bridge_data *bridge) if (page == NULL) return NULL; - map_page_into_agp(page); + /* agp_allocate_memory will do flush */ + map_page_into_agp_noflush(page); get_page(page); atomic_inc(&agp_bridge->current_memory_agp); diff --git a/include/asm-x86/agp.h b/include/asm-x86/agp.h index e4004a9f6a9..181b9e984b3 100644 --- a/include/asm-x86/agp.h +++ b/include/asm-x86/agp.h @@ -15,6 +15,9 @@ #define map_page_into_agp(page) set_pages_uc(page, 1) #define unmap_page_from_agp(page) set_pages_wb(page, 1) +#define map_page_into_agp_noflush(page) set_pages_uc_noflush(page, 1) +#define map_page_into_agp_global_flush() set_memory_flush_all() + /* * Could use CLFLUSH here if the cpu supports it. But then it would * need to be called for each cacheline of the whole page so it may -- cgit v1.2.3-70-g09d2 From 1f49a2c2aeb22d5abc6d4ea574ff63d37ca55fbe Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Fri, 15 Aug 2008 12:45:09 -0400 Subject: x86: revert replace LOCK_PREFIX in futex.h Since we now use DS prefixes instead of NOP to remove LOCK prefixes, there are no longer any issues with instruction boundaries moving around. Depends on : x86 alternatives : fix LOCK_PREFIX race with preemptible kernel and CPU hotplug On Thu, 14 Aug 2008, Mathieu Desnoyers wrote: > > Changing the 0x90 (single-byte nop) currently used into a 0x3E DS segment > override prefix should fix this issue. Since the default of the atomic > instructions is to use the DS segment anyway, it should not affect the > behavior. Ok, so I think this is an _excellent_ patch, but I'd like to also then use LOCK_PREFIX in include/asm-x86/futex.h. See commit 9d55b9923a1b7ea8193b8875c57ec940dc2ff027. Linus Applies to 2.6.27-rc2 (and -rc3 unless hell broke loose in futex.h between rc2 and rc3). Signed-off-by: Mathieu Desnoyers CC: Linus Torvalds CC: H. Peter Anvin CC: Jeremy Fitzhardinge CC: Roland McGrath CC: Ingo Molnar Cc: Steven Rostedt CC: Steven Rostedt CC: Thomas Gleixner CC: Peter Zijlstra CC: Andrew Morton CC: David Miller CC: Ulrich Drepper CC: Rusty Russell CC: Gregory Haskins CC: Arnaldo Carvalho de Melo CC: "Luis Claudio R. Goncalves" CC: Clark Williams CC: Christoph Lameter CC: Andi Kleen CC: Harvey Harrison Signed-off-by: H. Peter Anvin --- include/asm-x86/futex.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/futex.h b/include/asm-x86/futex.h index e7a76b37b33..d1b988ce080 100644 --- a/include/asm-x86/futex.h +++ b/include/asm-x86/futex.h @@ -25,7 +25,7 @@ asm volatile("1:\tmovl %2, %0\n" \ "\tmovl\t%0, %3\n" \ "\t" insn "\n" \ - "2:\tlock; cmpxchgl %3, %2\n" \ + "2:\t" LOCK_PREFIX "cmpxchgl %3, %2\n" \ "\tjnz\t1b\n" \ "3:\t.section .fixup,\"ax\"\n" \ "4:\tmov\t%5, %1\n" \ @@ -64,7 +64,7 @@ static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) __futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg); break; case FUTEX_OP_ADD: - __futex_atomic_op1("lock; xaddl %0, %2", ret, oldval, + __futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret, oldval, uaddr, oparg); break; case FUTEX_OP_OR: @@ -122,7 +122,7 @@ static inline int futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - asm volatile("1:\tlock; cmpxchgl %3, %1\n" + asm volatile("1:\t" LOCK_PREFIX "cmpxchgl %3, %1\n" "2:\t.section .fixup, \"ax\"\n" "3:\tmov %2, %0\n" "\tjmp 2b\n" -- cgit v1.2.3-70-g09d2 From 5bbd4c3724008c93cf3efdfc38a3402e245ab506 Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Fri, 15 Aug 2008 12:56:59 -0400 Subject: x86: spinlock use LOCK_PREFIX Since we are now using DS prefixes instead of NOP to remove LOCK prefixes, there is no longer any problems with instruction boundaries moving around. * Linus Torvalds (torvalds@linux-foundation.org) wrote: > > > On Thu, 14 Aug 2008, Mathieu Desnoyers wrote: > > > > Changing the 0x90 (single-byte nop) currently used into a 0x3E DS segment > > override prefix should fix this issue. Since the default of the atomic > > instructions is to use the DS segment anyway, it should not affect the > > behavior. > > Ok, so I think this is an _excellent_ patch, but I'd like to also then use > LOCK_PREFIX in include/asm-x86/futex.h. > > See commit 9d55b9923a1b7ea8193b8875c57ec940dc2ff027. > > Linus Unless there a rationale for this, I think these be changed to LOCK_PREFIX too. grep "lock ;" include/asm-x86/spinlock.h "lock ; cmpxchgw %w1,%2\n\t" asm volatile("lock ; xaddl %0, %1\n" "lock ; cmpxchgl %1,%2\n\t" Applies to 2.6.27-rc2. Signed-off-by: Mathieu Desnoyers Acked-by: Linus Torvalds CC: Linus Torvalds CC: H. Peter Anvin CC: Jeremy Fitzhardinge CC: Roland McGrath CC: Ingo Molnar Cc: Steven Rostedt CC: Steven Rostedt CC: Thomas Gleixner CC: Peter Zijlstra CC: Andrew Morton CC: David Miller CC: Ulrich Drepper CC: Rusty Russell CC: Gregory Haskins CC: Arnaldo Carvalho de Melo CC: "Luis Claudio R. Goncalves" CC: Clark Williams CC: Christoph Lameter CC: Andi Kleen CC: Harvey Harrison Signed-off-by: H. Peter Anvin --- include/asm-x86/spinlock.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h index 4f9a9861799..0b4d59a93d2 100644 --- a/include/asm-x86/spinlock.h +++ b/include/asm-x86/spinlock.h @@ -97,7 +97,7 @@ static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) "jne 1f\n\t" "movw %w0,%w1\n\t" "incb %h1\n\t" - "lock ; cmpxchgw %w1,%2\n\t" + LOCK_PREFIX "cmpxchgw %w1,%2\n\t" "1:" "sete %b1\n\t" "movzbl %b1,%0\n\t" @@ -135,7 +135,7 @@ static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) int inc = 0x00010000; int tmp; - asm volatile("lock ; xaddl %0, %1\n" + asm volatile(LOCK_PREFIX "xaddl %0, %1\n" "movzwl %w0, %2\n\t" "shrl $16, %0\n\t" "1:\t" @@ -162,7 +162,7 @@ static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) "cmpl %0,%1\n\t" "jne 1f\n\t" "addl $0x00010000, %1\n\t" - "lock ; cmpxchgl %1,%2\n\t" + LOCK_PREFIX "cmpxchgl %1,%2\n\t" "1:" "sete %b1\n\t" "movzbl %b1,%0\n\t" -- cgit v1.2.3-70-g09d2 From 8d02c2110b3fb8e2700b31596a582a2989fd72ba Mon Sep 17 00:00:00 2001 From: Thomas Petazzoni Date: Tue, 5 Aug 2008 11:45:19 +0200 Subject: x86: configuration options to compile out x86 CPU support code This patch adds some configuration options that allow to compile out CPU vendor-specific code in x86 kernels (in arch/x86/kernel/cpu). The new configuration options are only visible when CONFIG_EMBEDDED is selected, as they are mostly interesting for space savings reasons. An example of size saving, on x86 with only Intel CPU support: text data bss dec hex filename 1125479 118760 212992 1457231 163c4f vmlinux.old 1121355 116536 212992 1450883 162383 vmlinux -4124 -2224 0 -6348 -18CC +/- However, I'm not exactly sure that the Kconfig wording is correct with regard to !64BIT / 64BIT. [ mingo@elte.hu: convert macro to inline ] Signed-off-by: Thomas Petazzoni Signed-off-by: Ingo Molnar --- arch/x86/Kconfig.cpu | 70 ++++++++++++++++++++++++++++++++++++++++++++ arch/x86/kernel/cpu/Makefile | 19 ++++++------ include/asm-x86/bugs.h | 5 ++++ 3 files changed, 85 insertions(+), 9 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index 2c518fbc52e..6156ac25ff8 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -415,3 +415,73 @@ config X86_MINIMUM_CPU_FAMILY config X86_DEBUGCTLMSR def_bool y depends on !(MK6 || MWINCHIPC6 || MWINCHIP2 || MWINCHIP3D || MCYRIXIII || M586MMX || M586TSC || M586 || M486 || M386) + +menuconfig PROCESSOR_SELECT + default y + bool "Supported processor vendors" if EMBEDDED + help + This lets you choose what x86 vendor support code your kernel + will include. + +config CPU_SUP_INTEL_32 + default y + bool "Support Intel processors" if PROCESSOR_SELECT + depends on !64BIT + help + This enables extended support for Intel processors + +config CPU_SUP_INTEL_64 + default y + bool "Support Intel processors" if PROCESSOR_SELECT + depends on 64BIT + help + This enables extended support for Intel processors + +config CPU_SUP_CYRIX_32 + default y + bool "Support Cyrix processors" if PROCESSOR_SELECT + depends on !64BIT + help + This enables extended support for Cyrix processors + +config CPU_SUP_AMD_32 + default y + bool "Support AMD processors" if PROCESSOR_SELECT + depends on !64BIT + help + This enables extended support for AMD processors + +config CPU_SUP_AMD_64 + default y + bool "Support AMD processors" if PROCESSOR_SELECT + depends on 64BIT + help + This enables extended support for AMD processors + +config CPU_SUP_CENTAUR_32 + default y + bool "Support Centaur processors" if PROCESSOR_SELECT + depends on !64BIT + help + This enables extended support for Centaur processors + +config CPU_SUP_CENTAUR_64 + default y + bool "Support Centaur processors" if PROCESSOR_SELECT + depends on 64BIT + help + This enables extended support for Centaur processors + +config CPU_SUP_TRANSMETA_32 + default y + bool "Support Transmeta processors" if PROCESSOR_SELECT + depends on !64BIT + help + This enables extended support for Transmeta processors + +config CPU_SUP_UMC_32 + default y + bool "Support UMC processors" if PROCESSOR_SELECT + depends on !64BIT + help + This enables extended support for UMC processors diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 25b4c063fbf..a0fc6c14438 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -7,15 +7,16 @@ obj-y += proc.o feature_names.o obj-$(CONFIG_X86_32) += common.o bugs.o cmpxchg.o obj-$(CONFIG_X86_64) += common_64.o bugs_64.o -obj-$(CONFIG_X86_32) += amd.o -obj-$(CONFIG_X86_64) += amd_64.o -obj-$(CONFIG_X86_32) += cyrix.o -obj-$(CONFIG_X86_32) += centaur.o -obj-$(CONFIG_X86_64) += centaur_64.o -obj-$(CONFIG_X86_32) += transmeta.o -obj-$(CONFIG_X86_32) += intel.o -obj-$(CONFIG_X86_64) += intel_64.o -obj-$(CONFIG_X86_32) += umc.o + +obj-$(CONFIG_CPU_SUP_AMD_32) += amd.o +obj-$(CONFIG_CPU_SUP_AMD_64) += amd_64.o +obj-$(CONFIG_CPU_SUP_CYRIX_32) += cyrix.o +obj-$(CONFIG_CPU_SUP_CENTAUR_32) += centaur.o +obj-$(CONFIG_CPU_SUP_CENTAUR_64) += centaur_64.o +obj-$(CONFIG_CPU_SUP_TRANSMETA_32) += transmeta.o +obj-$(CONFIG_CPU_SUP_INTEL_32) += intel.o +obj-$(CONFIG_CPU_SUP_INTEL_64) += intel_64.o +obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o obj-$(CONFIG_X86_MCE) += mcheck/ obj-$(CONFIG_MTRR) += mtrr/ diff --git a/include/asm-x86/bugs.h b/include/asm-x86/bugs.h index 021cbdd5f25..00e4a0cd6f2 100644 --- a/include/asm-x86/bugs.h +++ b/include/asm-x86/bugs.h @@ -2,6 +2,11 @@ #define _ASM_X86_BUGS_H extern void check_bugs(void); + +#ifdef CONFIG_CPU_SUP_INTEL_32 int ppro_with_ram_bug(void); +#else +static inline int ppro_with_ram_bug(void) { return 0; } +#endif #endif /* _ASM_X86_BUGS_H */ -- cgit v1.2.3-70-g09d2 From 7e00df5818964298c9821365a6cb7a8304227c5c Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Mon, 18 Aug 2008 17:39:32 -0700 Subject: x86: add NOPL as a synthetic CPU feature bit The long noops ("NOPL") are supposed to be detected by family >= 6. Unfortunately, several non-Intel x86 implementations, both hardware and software, don't obey this dictum. Instead, probe for NOPL directly by executing a NOPL instruction and see if we get #UD. Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/common.c | 32 +++++++++++++++++++++++++++++++- arch/x86/kernel/cpu/common_64.c | 36 ++++++++++++++++++++++++++++++++++++ arch/x86/kernel/cpu/feature_names.c | 3 ++- include/asm-x86/cpufeature.h | 11 ++++++----- include/asm-x86/required-features.h | 8 +++++++- 5 files changed, 82 insertions(+), 8 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 80ab20d4fa3..0785b3c8d04 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -13,6 +13,7 @@ #include #include #include +#include #ifdef CONFIG_X86_LOCAL_APIC #include #include @@ -341,6 +342,35 @@ static void __init early_cpu_detect(void) early_get_cap(c); } +/* + * The NOPL instruction is supposed to exist on all CPUs with + * family >= 6, unfortunately, that's not true in practice because + * of early VIA chips and (more importantly) broken virtualizers that + * are not easy to detect. Hence, probe for it based on first + * principles. + */ +static void __cpuinit detect_nopl(struct cpuinfo_x86 *c) +{ + const u32 nopl_signature = 0x888c53b1; /* Random number */ + u32 has_nopl = nopl_signature; + + clear_cpu_cap(c, X86_FEATURE_NOPL); + if (c->x86 >= 6) { + asm volatile("\n" + "1: .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */ + "2:\n" + " .section .fixup,\"ax\"\n" + "3: xor %0,%0\n" + " jmp 2b\n" + " .previous\n" + _ASM_EXTABLE(1b,3b) + : "+a" (has_nopl)); + + if (has_nopl == nopl_signature) + set_cpu_cap(c, X86_FEATURE_NOPL); + } +} + static void __cpuinit generic_identify(struct cpuinfo_x86 *c) { u32 tfms, xlvl; @@ -395,8 +425,8 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 *c) } init_scattered_cpuid_features(c); + detect_nopl(c); } - } static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c) diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c index dd6e3f15017..c3afba5a81a 100644 --- a/arch/x86/kernel/cpu/common_64.c +++ b/arch/x86/kernel/cpu/common_64.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #ifdef CONFIG_X86_LOCAL_APIC #include @@ -215,6 +216,39 @@ static void __init early_cpu_support_print(void) } } +/* + * The NOPL instruction is supposed to exist on all CPUs with + * family >= 6, unfortunately, that's not true in practice because + * of early VIA chips and (more importantly) broken virtualizers that + * are not easy to detect. Hence, probe for it based on first + * principles. + * + * Note: no 64-bit chip is known to lack these, but put the code here + * for consistency with 32 bits, and to make it utterly trivial to + * diagnose the problem should it ever surface. + */ +static void __cpuinit detect_nopl(struct cpuinfo_x86 *c) +{ + const u32 nopl_signature = 0x888c53b1; /* Random number */ + u32 has_nopl = nopl_signature; + + clear_cpu_cap(c, X86_FEATURE_NOPL); + if (c->x86 >= 6) { + asm volatile("\n" + "1: .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */ + "2:\n" + " .section .fixup,\"ax\"\n" + "3: xor %0,%0\n" + " jmp 2b\n" + " .previous\n" + _ASM_EXTABLE(1b,3b) + : "+a" (has_nopl)); + + if (has_nopl == nopl_signature) + set_cpu_cap(c, X86_FEATURE_NOPL); + } +} + static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c); void __init early_cpu_init(void) @@ -313,6 +347,8 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) c->x86_phys_bits = eax & 0xff; } + detect_nopl(c); + if (c->x86_vendor != X86_VENDOR_UNKNOWN && cpu_devs[c->x86_vendor]->c_early_init) cpu_devs[c->x86_vendor]->c_early_init(c); diff --git a/arch/x86/kernel/cpu/feature_names.c b/arch/x86/kernel/cpu/feature_names.c index e43ad4ad4cb..c9017799497 100644 --- a/arch/x86/kernel/cpu/feature_names.c +++ b/arch/x86/kernel/cpu/feature_names.c @@ -39,7 +39,8 @@ const char * const x86_cap_flags[NCAPINTS*32] = { NULL, NULL, NULL, NULL, "constant_tsc", "up", NULL, "arch_perfmon", "pebs", "bts", NULL, NULL, - "rep_good", NULL, NULL, NULL, NULL, NULL, NULL, NULL, + "rep_good", NULL, NULL, NULL, + "nopl", NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, /* Intel-defined (#2) */ diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 2f5a792b0ac..be8b2ad5d41 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -72,14 +72,15 @@ #define X86_FEATURE_UP (3*32+ 9) /* smp kernel running on up */ #define X86_FEATURE_FXSAVE_LEAK (3*32+10) /* FXSAVE leaks FOP/FIP/FOP */ #define X86_FEATURE_ARCH_PERFMON (3*32+11) /* Intel Architectural PerfMon */ -#define X86_FEATURE_PEBS (3*32+12) /* Precise-Event Based Sampling */ -#define X86_FEATURE_BTS (3*32+13) /* Branch Trace Store */ -#define X86_FEATURE_SYSCALL32 (3*32+14) /* syscall in ia32 userspace */ -#define X86_FEATURE_SYSENTER32 (3*32+15) /* sysenter in ia32 userspace */ +#define X86_FEATURE_PEBS (3*32+12) /* Precise-Event Based Sampling */ +#define X86_FEATURE_BTS (3*32+13) /* Branch Trace Store */ +#define X86_FEATURE_SYSCALL32 (3*32+14) /* syscall in ia32 userspace */ +#define X86_FEATURE_SYSENTER32 (3*32+15) /* sysenter in ia32 userspace */ #define X86_FEATURE_REP_GOOD (3*32+16) /* rep microcode works well on this CPU */ #define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* Mfence synchronizes RDTSC */ #define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* Lfence synchronizes RDTSC */ -#define X86_FEATURE_11AP (3*32+19) /* Bad local APIC aka 11AP */ +#define X86_FEATURE_11AP (3*32+19) /* Bad local APIC aka 11AP */ +#define X86_FEATURE_NOPL (3*32+20) /* The NOPL (0F 1F) instructions */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* Streaming SIMD Extensions-3 */ diff --git a/include/asm-x86/required-features.h b/include/asm-x86/required-features.h index adec887dd7c..5c2ff4bc298 100644 --- a/include/asm-x86/required-features.h +++ b/include/asm-x86/required-features.h @@ -41,6 +41,12 @@ # define NEED_3DNOW 0 #endif +#if defined(CONFIG_X86_P6_NOP) || defined(CONFIG_X86_64) +# define NEED_NOPL (1<<(X86_FEATURE_NOPL & 31)) +#else +# define NEED_NOPL 0 +#endif + #ifdef CONFIG_X86_64 #define NEED_PSE 0 #define NEED_MSR (1<<(X86_FEATURE_MSR & 31)) @@ -67,7 +73,7 @@ #define REQUIRED_MASK1 (NEED_LM|NEED_3DNOW) #define REQUIRED_MASK2 0 -#define REQUIRED_MASK3 0 +#define REQUIRED_MASK3 (NEED_NOPL) #define REQUIRED_MASK4 0 #define REQUIRED_MASK5 0 #define REQUIRED_MASK6 0 -- cgit v1.2.3-70-g09d2 From 8df9676d6402563da91427e8d9f2da8a4598aede Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Mon, 18 Aug 2008 18:13:33 -0700 Subject: x86: consistency cleanups Rename _ASM_MOV_UL to _ASM_MOV for consistency with other _ASM_ instructions (_ASM_ADD, _ASM_SUB and so on.) Add ASM_SP, _ASM_BP, _ASM_SI, and _ASM_DI for consistency with _ASM_[ABCD]X. Signed-off-by: H. Peter Anvin --- include/asm-x86/asm.h | 7 ++++++- include/asm-x86/resume-trace.h | 2 +- 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/asm.h b/include/asm-x86/asm.h index 97220321f39..56be78f582f 100644 --- a/include/asm-x86/asm.h +++ b/include/asm-x86/asm.h @@ -20,17 +20,22 @@ #define _ASM_PTR __ASM_SEL(.long, .quad) #define _ASM_ALIGN __ASM_SEL(.balign 4, .balign 8) -#define _ASM_MOV_UL __ASM_SIZE(mov) +#define _ASM_MOV __ASM_SIZE(mov) #define _ASM_INC __ASM_SIZE(inc) #define _ASM_DEC __ASM_SIZE(dec) #define _ASM_ADD __ASM_SIZE(add) #define _ASM_SUB __ASM_SIZE(sub) #define _ASM_XADD __ASM_SIZE(xadd) + #define _ASM_AX __ASM_REG(ax) #define _ASM_BX __ASM_REG(bx) #define _ASM_CX __ASM_REG(cx) #define _ASM_DX __ASM_REG(dx) +#define _ASM_SP __ASM_REG(sp) +#define _ASM_BP __ASM_REG(bp) +#define _ASM_SI __ASM_REG(si) +#define _ASM_DI __ASM_REG(di) /* Exception table entry */ # define _ASM_EXTABLE(from,to) \ diff --git a/include/asm-x86/resume-trace.h b/include/asm-x86/resume-trace.h index 8d9f0b41ee8..e219b4c9119 100644 --- a/include/asm-x86/resume-trace.h +++ b/include/asm-x86/resume-trace.h @@ -7,7 +7,7 @@ do { \ if (pm_trace_enabled) { \ const void *tracedata; \ - asm volatile(_ASM_MOV_UL " $1f,%0\n" \ + asm volatile(_ASM_MOV " $1f,%0\n" \ ".section .tracedata,\"a\"\n" \ "1:\t.word %c1\n\t" \ _ASM_PTR " %c2\n" \ -- cgit v1.2.3-70-g09d2 From d45de40934897c6ee5b05141f7895bbb28512395 Mon Sep 17 00:00:00 2001 From: Dmitry Adamushko Date: Wed, 20 Aug 2008 00:22:26 +0200 Subject: x86-microcode: generic interface refactoring This is the 1st patch in the series. Here the aim was to avoid any significant changes, logically-wise. So it's mainly about generic interface refactoring: e.g. make microcode_{intel,amd}.c more about arch-specific details and less about policies like make-sure-we-run-on-a-target-cpu (no more set_cpus_allowed_ptr() here) and generic synchronization (no more microcode_mutex here). All in all, more line have been deleted than added. 4 files changed, 145 insertions(+), 198 deletions(-) Signed-off-by: Dmitry Adamushko Signed-off-by: Ingo Molnar --- arch/x86/kernel/microcode.c | 155 +++++++++++++++++++++++++------------- arch/x86/kernel/microcode_amd.c | 77 +++---------------- arch/x86/kernel/microcode_intel.c | 91 ++++------------------ include/asm-x86/microcode.h | 20 ++++- 4 files changed, 145 insertions(+), 198 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c index ad136ad99cb..b2f84ce5eed 100644 --- a/arch/x86/kernel/microcode.c +++ b/arch/x86/kernel/microcode.c @@ -71,7 +71,7 @@ * Thanks to Stuart Swales for pointing out this bug. */ -/*#define DEBUG pr_debug */ +/* #define DEBUG pr_debug */ #include #include #include @@ -104,8 +104,7 @@ MODULE_LICENSE("GPL"); struct microcode_ops *microcode_ops; /* no concurrent ->write()s are allowed on /dev/cpu/microcode */ -DEFINE_MUTEX(microcode_mutex); -EXPORT_SYMBOL_GPL(microcode_mutex); +static DEFINE_MUTEX(microcode_mutex); struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; EXPORT_SYMBOL_GPL(ucode_cpu_info); @@ -234,22 +233,6 @@ MODULE_ALIAS_MISCDEV(MICROCODE_MINOR); struct platform_device *microcode_pdev; EXPORT_SYMBOL_GPL(microcode_pdev); -static void microcode_init_cpu(int cpu, int resume) -{ - cpumask_t old; - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - - old = current->cpus_allowed; - - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - mutex_lock(µcode_mutex); - microcode_ops->collect_cpu_info(cpu); - if (uci->valid && system_state == SYSTEM_RUNNING && !resume) - microcode_ops->cpu_request_microcode(cpu); - mutex_unlock(µcode_mutex); - set_cpus_allowed_ptr(current, &old); -} - static ssize_t reload_store(struct sys_device *dev, struct sysdev_attribute *attr, const char *buf, size_t sz) @@ -266,14 +249,15 @@ static ssize_t reload_store(struct sys_device *dev, cpumask_t old = current->cpus_allowed; get_online_cpus(); - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - - mutex_lock(µcode_mutex); - if (uci->valid) - err = microcode_ops->cpu_request_microcode(cpu); - mutex_unlock(µcode_mutex); + if (cpu_online(cpu)) { + set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); + mutex_lock(µcode_mutex); + if (uci->valid) + err = microcode_ops->cpu_request_microcode(cpu); + mutex_unlock(µcode_mutex); + set_cpus_allowed_ptr(current, &old); + } put_online_cpus(); - set_cpus_allowed_ptr(current, &old); } if (err) return err; @@ -285,7 +269,7 @@ static ssize_t version_show(struct sys_device *dev, { struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; - return sprintf(buf, "0x%x\n", uci->rev); + return sprintf(buf, "0x%x\n", uci->cpu_sig.rev); } static ssize_t pf_show(struct sys_device *dev, @@ -293,7 +277,7 @@ static ssize_t pf_show(struct sys_device *dev, { struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; - return sprintf(buf, "0x%x\n", uci->pf); + return sprintf(buf, "0x%x\n", uci->cpu_sig.pf); } static SYSDEV_ATTR(reload, 0200, NULL, reload_store); @@ -312,7 +296,85 @@ static struct attribute_group mc_attr_group = { .name = "microcode", }; -static int __mc_sysdev_add(struct sys_device *sys_dev, int resume) +static void microcode_fini_cpu(int cpu) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + + mutex_lock(µcode_mutex); + microcode_ops->microcode_fini_cpu(cpu); + uci->valid = 0; + mutex_unlock(µcode_mutex); +} + +static void collect_cpu_info(int cpu) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + + memset(uci, 0, sizeof(*uci)); + if (!microcode_ops->collect_cpu_info(cpu, &uci->cpu_sig)) + uci->valid = 1; +} + +static void microcode_resume_cpu(int cpu) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + struct cpu_signature nsig; + + pr_debug("microcode: CPU%d resumed\n", cpu); + + if (!uci->mc.valid_mc) + return; + + /* + * Let's verify that the 'cached' ucode does belong + * to this cpu (a bit of paranoia): + */ + if (microcode_ops->collect_cpu_info(cpu, &nsig)) { + microcode_fini_cpu(cpu); + return; + } + + if (memcmp(&nsig, &uci->cpu_sig, sizeof(nsig))) { + microcode_fini_cpu(cpu); + /* Should we look for a new ucode here? */ + return; + } + + microcode_ops->apply_microcode(cpu); +} + +void microcode_update_cpu(int cpu) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + + /* We should bind the task to the CPU */ + BUG_ON(raw_smp_processor_id() != cpu); + + mutex_lock(µcode_mutex); + /* + * Check if the system resume is in progress (uci->valid != NULL), + * otherwise just request a firmware: + */ + if (uci->valid) { + microcode_resume_cpu(cpu); + } else { + collect_cpu_info(cpu); + if (uci->valid && system_state == SYSTEM_RUNNING) + microcode_ops->cpu_request_microcode(cpu); + } + mutex_unlock(µcode_mutex); +} + +static void microcode_init_cpu(int cpu) +{ + cpumask_t old = current->cpus_allowed; + + set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); + microcode_update_cpu(cpu); + set_cpus_allowed_ptr(current, &old); +} + +static int mc_sysdev_add(struct sys_device *sys_dev) { int err, cpu = sys_dev->id; struct ucode_cpu_info *uci = ucode_cpu_info + cpu; @@ -327,16 +389,10 @@ static int __mc_sysdev_add(struct sys_device *sys_dev, int resume) if (err) return err; - microcode_init_cpu(cpu, resume); - + microcode_init_cpu(cpu); return 0; } -static int mc_sysdev_add(struct sys_device *sys_dev) -{ - return __mc_sysdev_add(sys_dev, 0); -} - static int mc_sysdev_remove(struct sys_device *sys_dev) { int cpu = sys_dev->id; @@ -345,7 +401,7 @@ static int mc_sysdev_remove(struct sys_device *sys_dev) return 0; pr_debug("microcode: CPU%d removed\n", cpu); - microcode_ops->microcode_fini_cpu(cpu); + microcode_fini_cpu(cpu); sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); return 0; } @@ -376,33 +432,26 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) sys_dev = get_cpu_sysdev(cpu); switch (action) { - case CPU_UP_CANCELED_FROZEN: - /* The CPU refused to come up during a system resume */ - microcode_ops->microcode_fini_cpu(cpu); - break; case CPU_ONLINE: - case CPU_DOWN_FAILED: - mc_sysdev_add(sys_dev); - break; case CPU_ONLINE_FROZEN: - /* System-wide resume is in progress, try to apply microcode */ - if (microcode_ops->apply_microcode_check_cpu(cpu)) { - /* The application of microcode failed */ - microcode_ops->microcode_fini_cpu(cpu); - __mc_sysdev_add(sys_dev, 1); - break; - } + microcode_init_cpu(cpu); + case CPU_DOWN_FAILED: case CPU_DOWN_FAILED_FROZEN: + pr_debug("microcode: CPU%d added\n", cpu); if (sysfs_create_group(&sys_dev->kobj, &mc_attr_group)) printk(KERN_ERR "microcode: Failed to create the sysfs " "group for CPU%d\n", cpu); break; case CPU_DOWN_PREPARE: - mc_sysdev_remove(sys_dev); - break; case CPU_DOWN_PREPARE_FROZEN: /* Suspend is in progress, only remove the interface */ sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); + pr_debug("microcode: CPU%d removed\n", cpu); + break; + case CPU_DEAD: + case CPU_UP_CANCELED_FROZEN: + /* The CPU refused to come up during a system resume */ + microcode_fini_cpu(cpu); break; } return NOTIFY_OK; diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c index a6e76ccf815..4006e5e3adf 100644 --- a/arch/x86/kernel/microcode_amd.c +++ b/arch/x86/kernel/microcode_amd.c @@ -59,38 +59,28 @@ MODULE_LICENSE("GPL v2"); /* serialize access to the physical write */ static DEFINE_SPINLOCK(microcode_update_lock); -/* no concurrent ->write()s are allowed on /dev/cpu/microcode */ -extern struct mutex (microcode_mutex); - struct equiv_cpu_entry *equiv_cpu_table; -extern struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; - -static void collect_cpu_info_amd(int cpu) +static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig) { struct cpuinfo_x86 *c = &cpu_data(cpu); - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - /* We should bind the task to the CPU */ - BUG_ON(raw_smp_processor_id() != cpu); - uci->rev = 0; - uci->pf = 0; - uci->mc.mc_amd = NULL; - uci->valid = 1; + memset(csig, 0, sizeof(*csig)); if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) { printk(KERN_ERR "microcode: CPU%d not a capable AMD processor\n", cpu); - uci->valid = 0; - return; + return -1; } asm volatile("movl %1, %%ecx; rdmsr" - : "=a" (uci->rev) + : "=a" (csig->rev) : "i" (0x0000008B) : "ecx"); printk(KERN_INFO "microcode: collect_cpu_info_amd : patch_id=0x%x\n", - uci->rev); + csig->rev); + + return 0; } static int get_matching_microcode_amd(void *mc, int cpu) @@ -119,7 +109,7 @@ static int get_matching_microcode_amd(void *mc, int cpu) if (equiv_cpu_table == NULL) { printk(KERN_INFO "microcode: CPU%d microcode update with " "version 0x%x (current=0x%x)\n", - cpu, mc_header->patch_id, uci->rev); + cpu, mc_header->patch_id, uci->cpu_sig.rev); goto out; } @@ -185,12 +175,12 @@ static int get_matching_microcode_amd(void *mc, int cpu) pci_dev_put(sb_pci_dev); } - if (mc_header->patch_id <= uci->rev) + if (mc_header->patch_id <= uci->cpu_sig.rev) return 0; printk(KERN_INFO "microcode: CPU%d found a matching microcode " "update with version 0x%x (current=0x%x)\n", - cpu, mc_header->patch_id, uci->rev); + cpu, mc_header->patch_id, uci->cpu_sig.rev); out: new_mc = vmalloc(UCODE_MAX_SIZE); @@ -250,9 +240,9 @@ static void apply_microcode_amd(int cpu) printk(KERN_INFO "microcode: CPU%d updated from revision " "0x%x to 0x%x \n", - cpu_num, uci->rev, uci->mc.mc_amd->hdr.patch_id); + cpu_num, uci->cpu_sig.rev, uci->mc.mc_amd->hdr.patch_id); - uci->rev = rev; + uci->cpu_sig.rev = rev; } #ifdef CONFIG_MICROCODE_OLD_INTERFACE @@ -437,61 +427,18 @@ static int cpu_request_microcode_amd(int cpu) return error; } -static int apply_microcode_check_cpu_amd(int cpu) -{ - struct cpuinfo_x86 *c = &cpu_data(cpu); - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - unsigned int rev; - cpumask_t old; - int err = 0; - - /* Check if the microcode is available */ - if (!uci->mc.mc_amd) - return 0; - - old = current->cpus_allowed; - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - - /* Check if the microcode we have in memory matches the CPU */ - if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 16) - err = -EINVAL; - - if (!err) { - asm volatile("movl %1, %%ecx; rdmsr" - : "=a" (rev) - : "i" (0x0000008B) : "ecx"); - - if (uci->rev != rev) - err = -EINVAL; - } - - if (!err) - apply_microcode_amd(cpu); - else - printk(KERN_ERR "microcode: Could not apply microcode to CPU%d:" - " rev=0x%x\n", - cpu, uci->rev); - - set_cpus_allowed(current, old); - return err; -} - static void microcode_fini_cpu_amd(int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - mutex_lock(µcode_mutex); - uci->valid = 0; vfree(uci->mc.mc_amd); uci->mc.mc_amd = NULL; - mutex_unlock(µcode_mutex); } static struct microcode_ops microcode_amd_ops = { .get_next_ucode = get_next_ucode_amd, .get_matching_microcode = get_matching_microcode_amd, .microcode_sanity_check = NULL, - .apply_microcode_check_cpu = apply_microcode_check_cpu_amd, .cpu_request_microcode = cpu_request_microcode_amd, .collect_cpu_info = collect_cpu_info_amd, .apply_microcode = apply_microcode_amd, diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index 6dd8907ff22..c9b53202ba3 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -122,46 +122,37 @@ MODULE_LICENSE("GPL"); /* serialize access to the physical write to MSR 0x79 */ static DEFINE_SPINLOCK(microcode_update_lock); -/* no concurrent ->write()s are allowed on /dev/cpu/microcode */ -extern struct mutex microcode_mutex; - -extern struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; - -static void collect_cpu_info(int cpu_num) +static int collect_cpu_info(int cpu_num, struct cpu_signature *csig) { struct cpuinfo_x86 *c = &cpu_data(cpu_num); - struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; unsigned int val[2]; - /* We should bind the task to the CPU */ - BUG_ON(raw_smp_processor_id() != cpu_num); - uci->pf = uci->rev = 0; - uci->mc.mc_intel = NULL; - uci->valid = 1; + memset(csig, 0, sizeof(*csig)); if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 || cpu_has(c, X86_FEATURE_IA64)) { printk(KERN_ERR "microcode: CPU%d not a capable Intel " "processor\n", cpu_num); - uci->valid = 0; - return; + return -1; } - uci->sig = cpuid_eax(0x00000001); + csig->sig = cpuid_eax(0x00000001); if ((c->x86_model >= 5) || (c->x86 > 6)) { /* get processor flags from MSR 0x17 */ rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]); - uci->pf = 1 << ((val[1] >> 18) & 7); + csig->pf = 1 << ((val[1] >> 18) & 7); } wrmsr(MSR_IA32_UCODE_REV, 0, 0); /* see notes above for revision 1.07. Apparent chip bug */ sync_core(); /* get the current revision from MSR 0x8B */ - rdmsr(MSR_IA32_UCODE_REV, val[0], uci->rev); + rdmsr(MSR_IA32_UCODE_REV, val[0], csig->rev); pr_debug("microcode: collect_cpu_info : sig=0x%x, pf=0x%x, rev=0x%x\n", - uci->sig, uci->pf, uci->rev); + csig->sig, csig->pf, csig->rev); + + return 0; } static inline int microcode_update_match(int cpu_num, @@ -169,8 +160,8 @@ static inline int microcode_update_match(int cpu_num, { struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; - if (!sigmatch(sig, uci->sig, pf, uci->pf) - || mc_header->rev <= uci->rev) + if (!sigmatch(sig, uci->cpu_sig.sig, pf, uci->cpu_sig.pf) + || mc_header->rev <= uci->cpu_sig.rev) return 0; return 1; } @@ -289,7 +280,7 @@ static int get_matching_microcode(void *mc, int cpu) find: pr_debug("microcode: CPU%d found a matching microcode update with" " version 0x%x (current=0x%x)\n", - cpu, mc_header->rev, uci->rev); + cpu, mc_header->rev, uci->cpu_sig.rev); new_mc = vmalloc(total_size); if (!new_mc) { printk(KERN_ERR "microcode: error! Can not allocate memory\n"); @@ -335,16 +326,16 @@ static void apply_microcode(int cpu) spin_unlock_irqrestore(µcode_update_lock, flags); if (val[1] != uci->mc.mc_intel->hdr.rev) { printk(KERN_ERR "microcode: CPU%d update from revision " - "0x%x to 0x%x failed\n", cpu_num, uci->rev, val[1]); + "0x%x to 0x%x failed\n", cpu_num, uci->cpu_sig.rev, val[1]); return; } printk(KERN_INFO "microcode: CPU%d updated from revision " "0x%x to 0x%x, date = %04x-%02x-%02x \n", - cpu_num, uci->rev, val[1], + cpu_num, uci->cpu_sig.rev, val[1], uci->mc.mc_intel->hdr.date & 0xffff, uci->mc.mc_intel->hdr.date >> 24, (uci->mc.mc_intel->hdr.date >> 16) & 0xff); - uci->rev = val[1]; + uci->cpu_sig.rev = val[1]; } #ifdef CONFIG_MICROCODE_OLD_INTERFACE @@ -459,70 +450,18 @@ static int cpu_request_microcode(int cpu) return error; } -static int apply_microcode_check_cpu(int cpu) -{ - struct cpuinfo_x86 *c = &cpu_data(cpu); - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - cpumask_t old; - unsigned int val[2]; - int err = 0; - - /* Check if the microcode is available */ - if (!uci->mc.mc_intel) - return 0; - - old = current->cpus_allowed; - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - - /* Check if the microcode we have in memory matches the CPU */ - if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 || - cpu_has(c, X86_FEATURE_IA64) || uci->sig != cpuid_eax(0x00000001)) - err = -EINVAL; - - if (!err && ((c->x86_model >= 5) || (c->x86 > 6))) { - /* get processor flags from MSR 0x17 */ - rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]); - if (uci->pf != (1 << ((val[1] >> 18) & 7))) - err = -EINVAL; - } - - if (!err) { - wrmsr(MSR_IA32_UCODE_REV, 0, 0); - /* see notes above for revision 1.07. Apparent chip bug */ - sync_core(); - /* get the current revision from MSR 0x8B */ - rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]); - if (uci->rev != val[1]) - err = -EINVAL; - } - - if (!err) - apply_microcode(cpu); - else - printk(KERN_ERR "microcode: Could not apply microcode to CPU%d:" - " sig=0x%x, pf=0x%x, rev=0x%x\n", - cpu, uci->sig, uci->pf, uci->rev); - - set_cpus_allowed_ptr(current, &old); - return err; -} - static void microcode_fini_cpu(int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - mutex_lock(µcode_mutex); - uci->valid = 0; vfree(uci->mc.mc_intel); uci->mc.mc_intel = NULL; - mutex_unlock(µcode_mutex); } static struct microcode_ops microcode_intel_ops = { .get_next_ucode = get_next_ucode, .get_matching_microcode = get_matching_microcode, .microcode_sanity_check = microcode_sanity_check, - .apply_microcode_check_cpu = apply_microcode_check_cpu, .cpu_request_microcode = cpu_request_microcode, .collect_cpu_info = collect_cpu_info, .apply_microcode = apply_microcode, diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index 18b2aeec2ad..7ceff48fa65 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -1,14 +1,18 @@ +#ifndef ASM_X86__MICROCODE_H +#define ASM_X86__MICROCODE_H + extern int microcode_init(void *opaque, struct module *module); extern void microcode_exit(void); +struct cpu_signature; + struct microcode_ops { long (*get_next_ucode)(void **mc, long offset); long (*microcode_get_next_ucode)(void **mc, long offset); int (*get_matching_microcode)(void *mc, int cpu); - int (*apply_microcode_check_cpu)(int cpu); int (*microcode_sanity_check)(void *mc); int (*cpu_request_microcode)(int cpu); - void (*collect_cpu_info)(int cpu_num); + int (*collect_cpu_info)(int cpu_num, struct cpu_signature *csig); void (*apply_microcode)(int cpu); void (*microcode_fini_cpu)(int cpu); void (*clear_patch)(void *data); @@ -75,13 +79,21 @@ struct microcode_amd { unsigned int mpb[0]; }; -struct ucode_cpu_info { - int valid; +struct cpu_signature { unsigned int sig; unsigned int pf; unsigned int rev; +}; + +struct ucode_cpu_info { + struct cpu_signature cpu_sig; + int valid; union { struct microcode_intel *mc_intel; struct microcode_amd *mc_amd; + void *valid_mc; } mc; }; +extern struct ucode_cpu_info ucode_cpu_info[]; + +#endif /* ASM_X86__MICROCODE_H */ -- cgit v1.2.3-70-g09d2 From b6edbb1e045a7116d5571544dae25c6c37c94a48 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Tue, 19 Aug 2008 13:04:19 -0700 Subject: x86_64: use save/loadsegment in ia32 compat Use savesegment and loadsegment consistently in ia32 compat code. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/ia32/ia32_aout.c | 11 +++++++---- arch/x86/ia32/ia32_signal.c | 21 ++++++++++----------- include/asm-x86/elf.h | 5 +++-- 3 files changed, 20 insertions(+), 17 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c index a0e1dbe67dc..127ec3f0721 100644 --- a/arch/x86/ia32/ia32_aout.c +++ b/arch/x86/ia32/ia32_aout.c @@ -85,8 +85,10 @@ static void dump_thread32(struct pt_regs *regs, struct user32 *dump) dump->regs.ax = regs->ax; dump->regs.ds = current->thread.ds; dump->regs.es = current->thread.es; - asm("movl %%fs,%0" : "=r" (fs)); dump->regs.fs = fs; - asm("movl %%gs,%0" : "=r" (gs)); dump->regs.gs = gs; + savesegment(fs, fs); + dump->regs.fs = fs; + savesegment(gs, gs); + dump->regs.gs = gs; dump->regs.orig_ax = regs->orig_ax; dump->regs.ip = regs->ip; dump->regs.cs = regs->cs; @@ -430,8 +432,9 @@ beyond_if: current->mm->start_stack = (unsigned long)create_aout_tables((char __user *)bprm->p, bprm); /* start thread */ - asm volatile("movl %0,%%fs" :: "r" (0)); \ - asm volatile("movl %0,%%es; movl %0,%%ds": :"r" (__USER32_DS)); + loadsegment(fs, 0); + loadsegment(ds, __USER32_DS); + loadsegment(es, __USER32_DS); load_gs_index(0); (regs)->ip = ex.a_entry; (regs)->sp = current->mm->start_stack; diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c index 20af4c79579..f1a2ac777fa 100644 --- a/arch/x86/ia32/ia32_signal.c +++ b/arch/x86/ia32/ia32_signal.c @@ -206,7 +206,7 @@ struct rt_sigframe { unsigned int cur; \ unsigned short pre; \ err |= __get_user(pre, &sc->seg); \ - asm volatile("movl %%" #seg ",%0" : "=r" (cur)); \ + savesegment(seg, cur); \ pre |= mask; \ if (pre != cur) loadsegment(seg, pre); } @@ -235,7 +235,7 @@ static int ia32_restore_sigcontext(struct pt_regs *regs, */ err |= __get_user(gs, &sc->gs); gs |= 3; - asm("movl %%gs,%0" : "=r" (oldgs)); + savesegment(gs, oldgs); if (gs != oldgs) load_gs_index(gs); @@ -355,14 +355,13 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc, { int tmp, err = 0; - tmp = 0; - __asm__("movl %%gs,%0" : "=r"(tmp): "0"(tmp)); + savesegment(gs, tmp); err |= __put_user(tmp, (unsigned int __user *)&sc->gs); - __asm__("movl %%fs,%0" : "=r"(tmp): "0"(tmp)); + savesegment(fs, tmp); err |= __put_user(tmp, (unsigned int __user *)&sc->fs); - __asm__("movl %%ds,%0" : "=r"(tmp): "0"(tmp)); + savesegment(ds, tmp); err |= __put_user(tmp, (unsigned int __user *)&sc->ds); - __asm__("movl %%es,%0" : "=r"(tmp): "0"(tmp)); + savesegment(es, tmp); err |= __put_user(tmp, (unsigned int __user *)&sc->es); err |= __put_user((u32)regs->di, &sc->di); @@ -498,8 +497,8 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka, regs->dx = 0; regs->cx = 0; - asm volatile("movl %0,%%ds" :: "r" (__USER32_DS)); - asm volatile("movl %0,%%es" :: "r" (__USER32_DS)); + loadsegment(ds, __USER32_DS); + loadsegment(es, __USER32_DS); regs->cs = __USER32_CS; regs->ss = __USER32_DS; @@ -591,8 +590,8 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, regs->dx = (unsigned long) &frame->info; regs->cx = (unsigned long) &frame->uc; - asm volatile("movl %0,%%ds" :: "r" (__USER32_DS)); - asm volatile("movl %0,%%es" :: "r" (__USER32_DS)); + loadsegment(ds, __USER32_DS); + loadsegment(es, __USER32_DS); regs->cs = __USER32_CS; regs->ss = __USER32_DS; diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h index 7be4733c793..acbf3451b96 100644 --- a/include/asm-x86/elf.h +++ b/include/asm-x86/elf.h @@ -148,8 +148,9 @@ do { \ static inline void start_ia32_thread(struct pt_regs *regs, u32 ip, u32 sp) { - asm volatile("movl %0,%%fs" :: "r" (0)); - asm volatile("movl %0,%%es; movl %0,%%ds" : : "r" (__USER32_DS)); + loadsegment(fs, 0); + loadsegment(ds, __USER32_DS); + loadsegment(es, __USER32_DS); load_gs_index(0); regs->ip = ip; regs->sp = sp; -- cgit v1.2.3-70-g09d2 From 6e833587e11ed0dbf12e647127f2650e2f80b26d Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Tue, 19 Aug 2008 13:16:17 -0700 Subject: xen: clean up domain mode predicates There are four operating modes Xen code may find itself running in: - native - hvm domain - pv dom0 - pv domU Clean up predicates for testing for these states to make them more consistent. Signed-off-by: Jeremy Fitzhardinge Cc: Xen-devel Signed-off-by: Ingo Molnar --- arch/x86/xen/enlighten.c | 9 +++++++-- drivers/block/xen-blkfront.c | 2 +- drivers/char/hvc_xen.c | 6 +++--- drivers/input/xen-kbdfront.c | 4 ++-- drivers/net/xen-netfront.c | 6 +++--- drivers/video/xen-fbfront.c | 4 ++-- drivers/xen/balloon.c | 2 +- drivers/xen/grant-table.c | 2 +- drivers/xen/xenbus/xenbus_probe.c | 8 ++++---- include/asm-x86/xen/hypervisor.h | 14 ++++++++++++-- 10 files changed, 36 insertions(+), 21 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 53afa14eb31..b106e825d26 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -56,6 +56,9 @@ EXPORT_SYMBOL_GPL(hypercall_page); DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu); DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info); +enum xen_domain_type xen_domain_type = XEN_NATIVE; +EXPORT_SYMBOL_GPL(xen_domain_type); + /* * Identity map, in addition to plain kernel map. This needs to be * large enough to allocate page table pages to allocate the rest. @@ -1613,6 +1616,8 @@ asmlinkage void __init xen_start_kernel(void) if (!xen_start_info) return; + xen_domain_type = XEN_PV_DOMAIN; + BUG_ON(memcmp(xen_start_info->magic, "xen-3", 5) != 0); xen_setup_features(); @@ -1650,7 +1655,7 @@ asmlinkage void __init xen_start_kernel(void) /* Prevent unwanted bits from being set in PTEs. */ __supported_pte_mask &= ~_PAGE_GLOBAL; - if (!is_initial_xendomain()) + if (!xen_initial_domain()) __supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD); /* Don't do the full vcpu_info placement stuff until we have a @@ -1685,7 +1690,7 @@ asmlinkage void __init xen_start_kernel(void) boot_params.hdr.ramdisk_size = xen_start_info->mod_len; boot_params.hdr.cmd_line_ptr = __pa(xen_start_info->cmd_line); - if (!is_initial_xendomain()) { + if (!xen_initial_domain()) { add_preferred_console("xenboot", 0, NULL); add_preferred_console("tty", 0, NULL); add_preferred_console("hvc", 0, NULL); diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 3ca643cafcc..d5e75325515 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -1032,7 +1032,7 @@ static struct xenbus_driver blkfront = { static int __init xlblk_init(void) { - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; if (register_blkdev(XENVBD_MAJOR, DEV_NAME)) { diff --git a/drivers/char/hvc_xen.c b/drivers/char/hvc_xen.c index 6b70aa66a58..538ceea5e7d 100644 --- a/drivers/char/hvc_xen.c +++ b/drivers/char/hvc_xen.c @@ -108,8 +108,8 @@ static int __init xen_init(void) { struct hvc_struct *hp; - if (!is_running_on_xen() || - is_initial_xendomain() || + if (!xen_pv_domain() || + xen_initial_domain() || !xen_start_info->console.domU.evtchn) return -ENODEV; @@ -142,7 +142,7 @@ static void __exit xen_fini(void) static int xen_cons_init(void) { - if (!is_running_on_xen()) + if (!xen_pv_domain()) return 0; hvc_instantiate(HVC_COOKIE, 0, &hvc_ops); diff --git a/drivers/input/xen-kbdfront.c b/drivers/input/xen-kbdfront.c index 9ce3b3baf3a..3ab6362f043 100644 --- a/drivers/input/xen-kbdfront.c +++ b/drivers/input/xen-kbdfront.c @@ -335,11 +335,11 @@ static struct xenbus_driver xenkbd = { static int __init xenkbd_init(void) { - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; /* Nothing to do if running in dom0. */ - if (is_initial_xendomain()) + if (xen_initial_domain()) return -ENODEV; return xenbus_register_frontend(&xenkbd); diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index c749bdba214..3c3dd403f5d 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -1794,10 +1794,10 @@ static struct xenbus_driver netfront = { static int __init netif_init(void) { - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; - if (is_initial_xendomain()) + if (xen_initial_domain()) return 0; printk(KERN_INFO "Initialising Xen virtual ethernet driver.\n"); @@ -1809,7 +1809,7 @@ module_init(netif_init); static void __exit netif_exit(void) { - if (is_initial_xendomain()) + if (xen_initial_domain()) return; xenbus_unregister_driver(&netfront); diff --git a/drivers/video/xen-fbfront.c b/drivers/video/xen-fbfront.c index 47ed39b52f9..a463b3dd837 100644 --- a/drivers/video/xen-fbfront.c +++ b/drivers/video/xen-fbfront.c @@ -680,11 +680,11 @@ static struct xenbus_driver xenfb = { static int __init xenfb_init(void) { - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; /* Nothing to do if running in dom0. */ - if (is_initial_xendomain()) + if (xen_initial_domain()) return -ENODEV; return xenbus_register_frontend(&xenfb); diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c index fff987b10e0..a51f3e17a5f 100644 --- a/drivers/xen/balloon.c +++ b/drivers/xen/balloon.c @@ -419,7 +419,7 @@ static int __init balloon_init(void) unsigned long pfn; struct page *page; - if (!is_running_on_xen()) + if (!xen_pv_domain()) return -ENODEV; pr_info("xen_balloon: Initialising balloon driver.\n"); diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index e9e11168616..06592b9da83 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -508,7 +508,7 @@ static int __devinit gnttab_init(void) unsigned int max_nr_glist_frames, nr_glist_frames; unsigned int nr_init_grefs; - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; nr_grant_frames = 1; diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c index 57ceb5346b7..7f24a98a446 100644 --- a/drivers/xen/xenbus/xenbus_probe.c +++ b/drivers/xen/xenbus/xenbus_probe.c @@ -814,7 +814,7 @@ static int __init xenbus_probe_init(void) DPRINTK(""); err = -ENODEV; - if (!is_running_on_xen()) + if (!xen_domain()) goto out_error; /* Register ourselves with the kernel bus subsystem */ @@ -829,7 +829,7 @@ static int __init xenbus_probe_init(void) /* * Domain0 doesn't have a store_evtchn or store_mfn yet. */ - if (is_initial_xendomain()) { + if (xen_initial_domain()) { /* dom0 not yet supported */ } else { xenstored_ready = 1; @@ -846,7 +846,7 @@ static int __init xenbus_probe_init(void) goto out_unreg_back; } - if (!is_initial_xendomain()) + if (!xen_initial_domain()) xenbus_probe(NULL); return 0; @@ -937,7 +937,7 @@ static void wait_for_devices(struct xenbus_driver *xendrv) unsigned long timeout = jiffies + 10*HZ; struct device_driver *drv = xendrv ? &xendrv->driver : NULL; - if (!ready_to_wait_for_devices || !is_running_on_xen()) + if (!ready_to_wait_for_devices || !xen_domain()) return; while (exists_disconnected_device(drv)) { diff --git a/include/asm-x86/xen/hypervisor.h b/include/asm-x86/xen/hypervisor.h index 8e15dd28c91..d9dd28caa50 100644 --- a/include/asm-x86/xen/hypervisor.h +++ b/include/asm-x86/xen/hypervisor.h @@ -55,7 +55,6 @@ /* arch/i386/kernel/setup.c */ extern struct shared_info *HYPERVISOR_shared_info; extern struct start_info *xen_start_info; -#define is_initial_xendomain() (xen_start_info->flags & SIF_INITDOMAIN) /* arch/i386/mach-xen/evtchn.c */ /* Force a proper event-channel callback from Xen. */ @@ -68,6 +67,17 @@ u64 jiffies_to_st(unsigned long jiffies); #define MULTI_UVMFLAGS_INDEX 3 #define MULTI_UVMDOMID_INDEX 4 -#define is_running_on_xen() (xen_start_info ? 1 : 0) +enum xen_domain_type { + XEN_NATIVE, + XEN_PV_DOMAIN, + XEN_HVM_DOMAIN, +}; + +extern enum xen_domain_type xen_domain_type; + +#define xen_domain() (xen_domain_type != XEN_NATIVE) +#define xen_pv_domain() (xen_domain_type == XEN_PV_DOMAIN) +#define xen_initial_domain() (xen_pv_domain() && xen_start_info->flags & SIF_INITDOMAIN) +#define xen_hvm_domain() (xen_domain_type == XEN_HVM_DOMAIN) #endif /* __HYPERVISOR_H__ */ -- cgit v1.2.3-70-g09d2 From 63d3a75d6f1fcf2f33e6abbe84e1f428c3586152 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Tue, 19 Aug 2008 13:19:36 -0700 Subject: x86/paravirt: add spin_lock_flags lock op It is useful for a pv_lock_ops backend to know whether interrupts are enabled or not in the context a spin_lock is being called. This allows it to enable interrupts while spinning, which could be particularly helpful when spinning becomes blocking. The default implementation just calls the normal spin_lock op, ignoring the flags. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/kernel/paravirt-spinlocks.c | 6 ++++++ include/asm-x86/paravirt.h | 7 +++++++ include/asm-x86/spinlock.h | 9 +++++++-- 3 files changed, 20 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c index 38d7f7f1dbc..206359a8e43 100644 --- a/arch/x86/kernel/paravirt-spinlocks.c +++ b/arch/x86/kernel/paravirt-spinlocks.c @@ -7,12 +7,18 @@ #include +static void default_spin_lock_flags(struct raw_spinlock *lock, unsigned long flags) +{ + __raw_spin_lock(lock); +} + struct pv_lock_ops pv_lock_ops = { #ifdef CONFIG_SMP .spin_is_locked = __ticket_spin_is_locked, .spin_is_contended = __ticket_spin_is_contended, .spin_lock = __ticket_spin_lock, + .spin_lock_flags = default_spin_lock_flags, .spin_trylock = __ticket_spin_trylock, .spin_unlock = __ticket_spin_unlock, #endif diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index db9b0647b34..8e9b1266898 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -333,6 +333,7 @@ struct pv_lock_ops { int (*spin_is_locked)(struct raw_spinlock *lock); int (*spin_is_contended)(struct raw_spinlock *lock); void (*spin_lock)(struct raw_spinlock *lock); + void (*spin_lock_flags)(struct raw_spinlock *lock, unsigned long flags); int (*spin_trylock)(struct raw_spinlock *lock); void (*spin_unlock)(struct raw_spinlock *lock); }; @@ -1414,6 +1415,12 @@ static __always_inline void __raw_spin_lock(struct raw_spinlock *lock) PVOP_VCALL1(pv_lock_ops.spin_lock, lock); } +static __always_inline void __raw_spin_lock_flags(struct raw_spinlock *lock, + unsigned long flags) +{ + PVOP_VCALL2(pv_lock_ops.spin_lock_flags, lock, flags); +} + static __always_inline int __raw_spin_trylock(struct raw_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_trylock, lock); diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h index e39c790dbfd..b755ea86367 100644 --- a/include/asm-x86/spinlock.h +++ b/include/asm-x86/spinlock.h @@ -182,8 +182,6 @@ static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) } #endif -#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) - #ifdef CONFIG_PARAVIRT /* * Define virtualization-friendly old-style lock byte lock, for use in @@ -272,6 +270,13 @@ static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock) { __ticket_spin_unlock(lock); } + +static __always_inline void __raw_spin_lock_flags(raw_spinlock_t *lock, + unsigned long flags) +{ + __raw_spin_lock(lock); +} + #endif /* CONFIG_PARAVIRT */ static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) -- cgit v1.2.3-70-g09d2 From e621bd18958ef5dbace3129ebe17a0a475e127d9 Mon Sep 17 00:00:00 2001 From: Dave Young Date: Wed, 20 Aug 2008 16:43:03 -0700 Subject: i386: vmalloc size fix Booting kernel with vmalloc=[any size<=16m] will oops on my pc (i386/1G memory). BUG_ON in arch/x86/mm/init_32.c triggered: BUG_ON((unsigned long)high_memory > VMALLOC_START); It's due to the vm area hole. In include/asm-x86/pgtable_32.h: #define VMALLOC_OFFSET (8 * 1024 * 1024) #define VMALLOC_START (((unsigned long)high_memory + 2 * VMALLOC_OFFSET - 1) \ & ~(VMALLOC_OFFSET - 1)) There's several related point: 1. MAXMEM : (-__PAGE_OFFSET - __VMALLOC_RESERVE). The space after VMALLOC_END is included as well, I set it to (VMALLOC_END - PAGE_OFFSET - __VMALLOC_RESERVE) 2. VMALLOC_OFFSET is not considered in __VMALLOC_RESERVE fixed by adding VMALLOC_OFFSET to it. 3. VMALLOC_START : (((unsigned long)high_memory + 2 * VMALLOC_OFFSET - 1) & ~(VMALLOC_OFFSET - 1)) So it's not always 8M, bigger than 8M possible. I set it to ((unsigned long)high_memory + VMALLOC_OFFSET) 4. the VMALLOC_RESERVE is an unused macro, so remove it here. Signed-off-by: Dave Young Cc: akpm@linux-foundation.org Cc: hidave.darkstar@gmail.com Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton --- arch/x86/mm/pgtable_32.c | 3 ++- include/asm-x86/page_32.h | 3 --- include/asm-x86/pgtable_32.h | 5 +++-- 3 files changed, 5 insertions(+), 6 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c index cab0abbd1eb..0951db9ee51 100644 --- a/arch/x86/mm/pgtable_32.c +++ b/arch/x86/mm/pgtable_32.c @@ -123,7 +123,8 @@ static int __init parse_vmalloc(char *arg) if (!arg) return -EINVAL; - __VMALLOC_RESERVE = memparse(arg, &arg); + /* Add VMALLOC_OFFSET to the parsed value due to vm area guard hole*/ + __VMALLOC_RESERVE = memparse(arg, &arg) + VMALLOC_OFFSET; return 0; } early_param("vmalloc", parse_vmalloc); diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index ab8528793f0..632feb156cc 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -89,9 +89,6 @@ extern int nx_enabled; extern unsigned int __VMALLOC_RESERVE; extern int sysctl_legacy_va_layout; -#define VMALLOC_RESERVE ((unsigned long)__VMALLOC_RESERVE) -#define MAXMEM (-__PAGE_OFFSET - __VMALLOC_RESERVE) - extern void find_low_pfn_range(void); extern unsigned long init_memory_mapping(unsigned long start, unsigned long end); diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h index 5c3b26567a9..9bb5269475c 100644 --- a/include/asm-x86/pgtable_32.h +++ b/include/asm-x86/pgtable_32.h @@ -56,8 +56,7 @@ void paging_init(void); * area for the same reason. ;) */ #define VMALLOC_OFFSET (8 * 1024 * 1024) -#define VMALLOC_START (((unsigned long)high_memory + 2 * VMALLOC_OFFSET - 1) \ - & ~(VMALLOC_OFFSET - 1)) +#define VMALLOC_START ((unsigned long)high_memory + VMALLOC_OFFSET) #ifdef CONFIG_X86_PAE #define LAST_PKMAP 512 #else @@ -73,6 +72,8 @@ void paging_init(void); # define VMALLOC_END (FIXADDR_START - 2 * PAGE_SIZE) #endif +#define MAXMEM (VMALLOC_END - PAGE_OFFSET - __VMALLOC_RESERVE) + /* * Define this if things work differently on an i386 and an i486: * it will (on an i486) warn about kernel memory accesses that are -- cgit v1.2.3-70-g09d2 From 11494547b1754c4f3bd7f707ab869e2adf54d52f Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 21 Aug 2008 01:01:19 -0700 Subject: x86: fix apic version warning after following patch, commit 1b313f4a6d7bee7b2c034b3f1e203bc360a71cca Author: Cyrill Gorcunov Date: Mon Aug 18 20:45:57 2008 +0400 x86: apic - generic_processor_info - use physid_set instead of phys_cpu and physids_or - set phys_cpu_present_map bit AFTER check for allowed number of processors - add checking for APIC valid version in 64bit mode (mostly not needed but added for merging purpose) - add apic_version definition for 64bit mode which is used now we are getting warning for acpi path on 64 bit system. make the 64-bit side fill in apic_version[] as well. [ mingo@elte.hu: build fix ] Signed-off-by: Ingo Molnar --- arch/x86/kernel/acpi/boot.c | 4 ---- include/asm-x86/mpspec.h | 3 ++- 2 files changed, 2 insertions(+), 5 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 12e260e8fb2..d9770a56511 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -239,10 +239,8 @@ static void __cpuinit acpi_register_lapic(int id, u8 enabled) return; } -#ifdef CONFIG_X86_32 if (boot_cpu_physical_apicid != -1U) ver = apic_version[boot_cpu_physical_apicid]; -#endif generic_processor_info(id, ver); } @@ -762,10 +760,8 @@ static void __init acpi_register_lapic_address(unsigned long address) set_fixmap_nocache(FIX_APIC_BASE, address); if (boot_cpu_physical_apicid == -1U) { boot_cpu_physical_apicid = read_apic_id(); -#ifdef CONFIG_X86_32 apic_version[boot_cpu_physical_apicid] = GET_APIC_VERSION(apic_read(APIC_LVR)); -#endif } } diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h index 118da365e37..be2241a818f 100644 --- a/include/asm-x86/mpspec.h +++ b/include/asm-x86/mpspec.h @@ -5,11 +5,12 @@ #include +extern int apic_version[MAX_APICS]; + #ifdef CONFIG_X86_32 #include extern unsigned int def_to_bigsmp; -extern int apic_version[MAX_APICS]; extern u8 apicid_2_node[]; extern int pic_mode; -- cgit v1.2.3-70-g09d2 From 671eef85a3e885dff4ce210d8774ad50a91d5967 Mon Sep 17 00:00:00 2001 From: "Cihula, Joseph" Date: Wed, 20 Aug 2008 16:43:07 -0700 Subject: x86, e820: add support for AddressRangeUnusuable ACPI memory type Add support for the E820_UNUSABLE memory type, which is defined in Revision 3.0b (Oct. 10, 2006) of the ACPI Specification on p. 394 Table 14-1: AddressRangeUnusuable This range of address contains memory in which errors have been detected. This range must not be used by the OSPM. Signed-off-by: Joseph Cihula Signed-off-by: Shane Wang Signed-off-by: Gang Wei Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar --- arch/x86/kernel/e820.c | 4 ++++ include/asm-x86/e820.h | 1 + 2 files changed, 5 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 9af89078f7b..291e6cd9f9c 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -148,6 +148,9 @@ void __init e820_print_map(char *who) case E820_NVS: printk(KERN_CONT "(ACPI NVS)\n"); break; + case E820_UNUSABLE: + printk("(unusable)\n"); + break; default: printk(KERN_CONT "type %u\n", e820.map[i].type); break; @@ -1260,6 +1263,7 @@ static inline const char *e820_type_to_string(int e820_type) case E820_RAM: return "System RAM"; case E820_ACPI: return "ACPI Tables"; case E820_NVS: return "ACPI Non-volatile Storage"; + case E820_UNUSABLE: return "Unusable memory"; default: return "reserved"; } } diff --git a/include/asm-x86/e820.h b/include/asm-x86/e820.h index f52daf176bc..ca433c36af7 100644 --- a/include/asm-x86/e820.h +++ b/include/asm-x86/e820.h @@ -43,6 +43,7 @@ #define E820_RESERVED 2 #define E820_ACPI 3 #define E820_NVS 4 +#define E820_UNUSABLE 5 /* reserved RAM used by kernel itself */ #define E820_RESERVED_KERN 128 -- cgit v1.2.3-70-g09d2 From 9326d61bf64c4293f834e86c11f52db5be9798d6 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 21 Aug 2008 13:46:25 +0200 Subject: Revert "reduce tlb/cache flush times of agpgart memory allocation" This reverts commit 466ae837424dcc538b1af2a0eaf53be32edcdbe7. --- drivers/char/agp/agp.h | 4 ---- drivers/char/agp/generic.c | 4 +--- include/asm-x86/agp.h | 3 --- 3 files changed, 1 insertion(+), 10 deletions(-) (limited to 'include/asm-x86') diff --git a/drivers/char/agp/agp.h b/drivers/char/agp/agp.h index 395168fb17e..81e14bea54b 100644 --- a/drivers/char/agp/agp.h +++ b/drivers/char/agp/agp.h @@ -30,10 +30,6 @@ #define _AGP_BACKEND_PRIV_H 1 #include /* for flush_agp_cache() */ -#ifndef map_page_into_agp_noflush -#define map_page_into_agp_noflush(page) map_page_into_agp(page) -#define map_page_into_agp_global_flush() -#endif #define PFX "agpgart: " diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c index bf239b8ecac..eaa1a355bb3 100644 --- a/drivers/char/agp/generic.c +++ b/drivers/char/agp/generic.c @@ -274,7 +274,6 @@ struct agp_memory *agp_allocate_memory(struct agp_bridge_data *bridge, new->memory[i] = virt_to_gart(addr); new->page_count++; } - map_page_into_agp_global_flush(); new->bridge = bridge; return new; @@ -1187,8 +1186,7 @@ void *agp_generic_alloc_page(struct agp_bridge_data *bridge) if (page == NULL) return NULL; - /* agp_allocate_memory will do flush */ - map_page_into_agp_noflush(page); + map_page_into_agp(page); get_page(page); atomic_inc(&agp_bridge->current_memory_agp); diff --git a/include/asm-x86/agp.h b/include/asm-x86/agp.h index 181b9e984b3..e4004a9f6a9 100644 --- a/include/asm-x86/agp.h +++ b/include/asm-x86/agp.h @@ -15,9 +15,6 @@ #define map_page_into_agp(page) set_pages_uc(page, 1) #define unmap_page_from_agp(page) set_pages_wb(page, 1) -#define map_page_into_agp_noflush(page) set_pages_uc_noflush(page, 1) -#define map_page_into_agp_global_flush() set_memory_flush_all() - /* * Could use CLFLUSH here if the cpu supports it. But then it would * need to be called for each cacheline of the whole page so it may -- cgit v1.2.3-70-g09d2 From cacf890694a36124ceddce44ff4c7b02d372ce7c Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 21 Aug 2008 13:46:33 +0200 Subject: Revert "introduce two APIs for page attribute" This reverts commit 1ac2f7d55b7ee1613c90631e87fea22ec06781e5. --- arch/x86/mm/pageattr.c | 58 ++++++-------------------------------------- include/asm-x86/cacheflush.h | 3 --- 2 files changed, 8 insertions(+), 53 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 4adb33628de..5c06469a065 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -752,12 +752,12 @@ static inline int cache_attr(pgprot_t attr) (_PAGE_PAT | _PAGE_PAT_LARGE | _PAGE_PWT | _PAGE_PCD); } -static int do_change_page_attr_set_clr(unsigned long addr, int numpages, +static int change_page_attr_set_clr(unsigned long addr, int numpages, pgprot_t mask_set, pgprot_t mask_clr, - int force_split, int *tlb_flush) + int force_split) { struct cpa_data cpa; - int ret, checkalias; + int ret, cache, checkalias; /* * Check, if we are requested to change a not supported @@ -795,22 +795,9 @@ static int do_change_page_attr_set_clr(unsigned long addr, int numpages, /* * Check whether we really changed something: */ - *tlb_flush = cpa.flushtlb; - cpa_fill_pool(NULL); - - return ret; -} - -static int change_page_attr_set_clr(unsigned long addr, int numpages, - pgprot_t mask_set, pgprot_t mask_clr, - int force_split) -{ - int cache, flush_cache = 0, ret; - - ret = do_change_page_attr_set_clr(addr, numpages, mask_set, mask_clr, - force_split, &flush_cache); - if (!flush_cache) + if (!cpa.flushtlb) goto out; + /* * No need to flush, when we did not set any of the caching * attributes: @@ -827,7 +814,10 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, cpa_flush_range(addr, numpages, cache); else cpa_flush_all(cache); + out: + cpa_fill_pool(NULL); + return ret; } @@ -865,30 +855,6 @@ int set_memory_uc(unsigned long addr, int numpages) } EXPORT_SYMBOL(set_memory_uc); -int set_memory_uc_noflush(unsigned long addr, int numpages) -{ - int flush; - /* - * for now UC MINUS. see comments in ioremap_nocache() - */ - if (reserve_memtype(addr, addr + numpages * PAGE_SIZE, - _PAGE_CACHE_UC_MINUS, NULL)) - return -EINVAL; - /* - * for now UC MINUS. see comments in ioremap_nocache() - */ - return do_change_page_attr_set_clr(addr, numpages, - __pgprot(_PAGE_CACHE_UC_MINUS), - __pgprot(0), 0, &flush); -} -EXPORT_SYMBOL(set_memory_uc_noflush); - -void set_memory_flush_all(void) -{ - cpa_flush_all(1); -} -EXPORT_SYMBOL(set_memory_flush_all); - int _set_memory_wc(unsigned long addr, int numpages) { return change_page_attr_set(addr, numpages, @@ -963,14 +929,6 @@ int set_pages_uc(struct page *page, int numpages) } EXPORT_SYMBOL(set_pages_uc); -int set_pages_uc_noflush(struct page *page, int numpages) -{ - unsigned long addr = (unsigned long)page_address(page); - - return set_memory_uc_noflush(addr, numpages); -} -EXPORT_SYMBOL(set_pages_uc_noflush); - int set_pages_wb(struct page *page, int numpages) { unsigned long addr = (unsigned long)page_address(page); diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h index 57bac7b68c4..f4c0ab50d2c 100644 --- a/include/asm-x86/cacheflush.h +++ b/include/asm-x86/cacheflush.h @@ -57,8 +57,6 @@ int _set_memory_uc(unsigned long addr, int numpages); int _set_memory_wc(unsigned long addr, int numpages); int _set_memory_wb(unsigned long addr, int numpages); int set_memory_uc(unsigned long addr, int numpages); -int set_memory_uc_noflush(unsigned long addr, int numpages); -void set_memory_flush_all(void); int set_memory_wc(unsigned long addr, int numpages); int set_memory_wb(unsigned long addr, int numpages); int set_memory_x(unsigned long addr, int numpages); @@ -89,7 +87,6 @@ int set_memory_4k(unsigned long addr, int numpages); */ int set_pages_uc(struct page *page, int numpages); -int set_pages_uc_noflush(struct page *page, int numpages); int set_pages_wb(struct page *page, int numpages); int set_pages_x(struct page *page, int numpages); int set_pages_nx(struct page *page, int numpages); -- cgit v1.2.3-70-g09d2 From d75586ad01e6c5a30e7337fb87d61e03556a1ecb Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Thu, 21 Aug 2008 10:46:06 +0800 Subject: x86, pageattr: introduce APIs to change pageattr of a page array Add array interface APIs of pageattr. page based cache flush is quite slow for a lot of pages. If pages are more than 1024 (4M), the patch will use a wbinvd(). We have a simple test here (run a 3d game - open arena), nearly all agp memory allocation are small (< 1M), so suppose this will not impact runtime performance. Signed-off-by: Dave Airlie Signed-off-by: Shaohua Li Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr.c | 216 +++++++++++++++++++++++++++++++++---------- include/asm-x86/cacheflush.h | 3 + 2 files changed, 169 insertions(+), 50 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 5c06469a065..041e81ef673 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -25,15 +25,19 @@ * The current flushing context - we pass it instead of 5 arguments: */ struct cpa_data { - unsigned long vaddr; + unsigned long *vaddr; pgprot_t mask_set; pgprot_t mask_clr; int numpages; - int flushtlb; + int flags; unsigned long pfn; unsigned force_split : 1; + int curpage; }; +#define CPA_FLUSHTLB 1 +#define CPA_ARRAY 2 + #ifdef CONFIG_PROC_FS static unsigned long direct_pages_count[PG_LEVEL_NUM]; @@ -184,6 +188,41 @@ static void cpa_flush_range(unsigned long start, int numpages, int cache) } } +static void cpa_flush_array(unsigned long *start, int numpages, int cache) +{ + unsigned int i, level; + unsigned long *addr; + + BUG_ON(irqs_disabled()); + + on_each_cpu(__cpa_flush_range, NULL, 1); + + if (!cache) + return; + + /* 4M threshold */ + if (numpages >= 1024) { + if (boot_cpu_data.x86_model >= 4) + wbinvd(); + return; + } + /* + * We only need to flush on one CPU, + * clflush is a MESI-coherent instruction that + * will cause all other CPUs to flush the same + * cachelines: + */ + for (i = 0, addr = start; i < numpages; i++, addr++) { + pte_t *pte = lookup_address(*addr, &level); + + /* + * Only flush present addresses: + */ + if (pte && (pte_val(*pte) & _PAGE_PRESENT)) + clflush_cache_range((void *) *addr, PAGE_SIZE); + } +} + /* * Certain areas of memory on x86 require very specific protection flags, * for example the BIOS area or kernel text. Callers don't always get this @@ -392,7 +431,7 @@ try_preserve_large_page(pte_t *kpte, unsigned long address, */ new_pte = pfn_pte(pte_pfn(old_pte), canon_pgprot(new_prot)); __set_pmd_pte(kpte, address, new_pte); - cpa->flushtlb = 1; + cpa->flags |= CPA_FLUSHTLB; do_split = 0; } @@ -578,11 +617,16 @@ out_unlock: static int __change_page_attr(struct cpa_data *cpa, int primary) { - unsigned long address = cpa->vaddr; + unsigned long address; int do_split, err; unsigned int level; pte_t *kpte, old_pte; + if (cpa->flags & CPA_ARRAY) + address = cpa->vaddr[cpa->curpage]; + else + address = *cpa->vaddr; + repeat: kpte = lookup_address(address, &level); if (!kpte) @@ -594,8 +638,8 @@ repeat: return 0; printk(KERN_WARNING "CPA: called for zero pte. " "vaddr = %lx cpa->vaddr = %lx\n", address, - cpa->vaddr); WARN_ON(1); + *cpa->vaddr); return -EINVAL; } @@ -621,7 +665,7 @@ repeat: */ if (pte_val(old_pte) != pte_val(new_pte)) { set_pte_atomic(kpte, new_pte); - cpa->flushtlb = 1; + cpa->flags |= CPA_FLUSHTLB; } cpa->numpages = 1; return 0; @@ -645,7 +689,7 @@ repeat: */ err = split_large_page(kpte, address); if (!err) { - cpa->flushtlb = 1; + cpa->flags |= CPA_FLUSHTLB; goto repeat; } @@ -658,6 +702,7 @@ static int cpa_process_alias(struct cpa_data *cpa) { struct cpa_data alias_cpa; int ret = 0; + unsigned long temp_cpa_vaddr, vaddr; if (cpa->pfn >= max_pfn_mapped) return 0; @@ -670,16 +715,24 @@ static int cpa_process_alias(struct cpa_data *cpa) * No need to redo, when the primary call touched the direct * mapping already: */ - if (!(within(cpa->vaddr, PAGE_OFFSET, + if (cpa->flags & CPA_ARRAY) + vaddr = cpa->vaddr[cpa->curpage]; + else + vaddr = *cpa->vaddr; + + if (!(within(vaddr, PAGE_OFFSET, PAGE_OFFSET + (max_low_pfn_mapped << PAGE_SHIFT)) #ifdef CONFIG_X86_64 - || within(cpa->vaddr, PAGE_OFFSET + (1UL<<32), + || within(vaddr, PAGE_OFFSET + (1UL<<32), PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT)) #endif )) { alias_cpa = *cpa; - alias_cpa.vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT); + temp_cpa_vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT); + alias_cpa.vaddr = &temp_cpa_vaddr; + alias_cpa.flags &= ~CPA_ARRAY; + ret = __change_page_attr_set_clr(&alias_cpa, 0); } @@ -691,7 +744,7 @@ static int cpa_process_alias(struct cpa_data *cpa) * No need to redo, when the primary call touched the high * mapping already: */ - if (within(cpa->vaddr, (unsigned long) _text, (unsigned long) _end)) + if (within(vaddr, (unsigned long) _text, (unsigned long) _end)) return 0; /* @@ -702,8 +755,9 @@ static int cpa_process_alias(struct cpa_data *cpa) return 0; alias_cpa = *cpa; - alias_cpa.vaddr = - (cpa->pfn << PAGE_SHIFT) + __START_KERNEL_map - phys_base; + temp_cpa_vaddr = (cpa->pfn << PAGE_SHIFT) + __START_KERNEL_map - phys_base; + alias_cpa.vaddr = &temp_cpa_vaddr; + alias_cpa.flags &= ~CPA_ARRAY; /* * The high mapping range is imprecise, so ignore the return value. @@ -723,6 +777,9 @@ static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias) * preservation check. */ cpa->numpages = numpages; + /* for array changes, we can't use large page */ + if (cpa->flags & CPA_ARRAY) + cpa->numpages = 1; ret = __change_page_attr(cpa, checkalias); if (ret) @@ -741,7 +798,11 @@ static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias) */ BUG_ON(cpa->numpages > numpages); numpages -= cpa->numpages; - cpa->vaddr += cpa->numpages * PAGE_SIZE; + if (cpa->flags & CPA_ARRAY) + cpa->curpage++; + else + *cpa->vaddr += cpa->numpages * PAGE_SIZE; + } return 0; } @@ -752,9 +813,9 @@ static inline int cache_attr(pgprot_t attr) (_PAGE_PAT | _PAGE_PAT_LARGE | _PAGE_PWT | _PAGE_PCD); } -static int change_page_attr_set_clr(unsigned long addr, int numpages, +static int change_page_attr_set_clr(unsigned long *addr, int numpages, pgprot_t mask_set, pgprot_t mask_clr, - int force_split) + int force_split, int array) { struct cpa_data cpa; int ret, cache, checkalias; @@ -769,12 +830,22 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, return 0; /* Ensure we are PAGE_SIZE aligned */ - if (addr & ~PAGE_MASK) { - addr &= PAGE_MASK; - /* - * People should not be passing in unaligned addresses: - */ - WARN_ON_ONCE(1); + if (!array) { + if (*addr & ~PAGE_MASK) { + *addr &= PAGE_MASK; + /* + * People should not be passing in unaligned addresses: + */ + WARN_ON_ONCE(1); + } + } else { + int i; + for (i = 0; i < numpages; i++) { + if (addr[i] & ~PAGE_MASK) { + addr[i] &= PAGE_MASK; + WARN_ON_ONCE(1); + } + } } /* Must avoid aliasing mappings in the highmem code */ @@ -784,9 +855,13 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, cpa.numpages = numpages; cpa.mask_set = mask_set; cpa.mask_clr = mask_clr; - cpa.flushtlb = 0; + cpa.flags = 0; + cpa.curpage = 0; cpa.force_split = force_split; + if (array) + cpa.flags |= CPA_ARRAY; + /* No alias checking for _NX bit modifications */ checkalias = (pgprot_val(mask_set) | pgprot_val(mask_clr)) != _PAGE_NX; @@ -795,7 +870,7 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, /* * Check whether we really changed something: */ - if (!cpa.flushtlb) + if (!(cpa.flags & CPA_FLUSHTLB)) goto out; /* @@ -810,9 +885,12 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages, * error case we fall back to cpa_flush_all (which uses * wbindv): */ - if (!ret && cpu_has_clflush) - cpa_flush_range(addr, numpages, cache); - else + if (!ret && cpu_has_clflush) { + if (cpa.flags & CPA_ARRAY) + cpa_flush_array(addr, numpages, cache); + else + cpa_flush_range(*addr, numpages, cache); + } else cpa_flush_all(cache); out: @@ -821,16 +899,18 @@ out: return ret; } -static inline int change_page_attr_set(unsigned long addr, int numpages, - pgprot_t mask) +static inline int change_page_attr_set(unsigned long *addr, int numpages, + pgprot_t mask, int array) { - return change_page_attr_set_clr(addr, numpages, mask, __pgprot(0), 0); + return change_page_attr_set_clr(addr, numpages, mask, __pgprot(0), 0, + array); } -static inline int change_page_attr_clear(unsigned long addr, int numpages, - pgprot_t mask) +static inline int change_page_attr_clear(unsigned long *addr, int numpages, + pgprot_t mask, int array) { - return change_page_attr_set_clr(addr, numpages, __pgprot(0), mask, 0); + return change_page_attr_set_clr(addr, numpages, __pgprot(0), mask, 0, + array); } int _set_memory_uc(unsigned long addr, int numpages) @@ -838,8 +918,8 @@ int _set_memory_uc(unsigned long addr, int numpages) /* * for now UC MINUS. see comments in ioremap_nocache() */ - return change_page_attr_set(addr, numpages, - __pgprot(_PAGE_CACHE_UC_MINUS)); + return change_page_attr_set(&addr, numpages, + __pgprot(_PAGE_CACHE_UC_MINUS), 0); } int set_memory_uc(unsigned long addr, int numpages) @@ -855,10 +935,31 @@ int set_memory_uc(unsigned long addr, int numpages) } EXPORT_SYMBOL(set_memory_uc); +int set_memory_array_uc(unsigned long *addr, int addrinarray) +{ + int i; + /* + * for now UC MINUS. see comments in ioremap_nocache() + */ + for (i = 0; i < addrinarray; i++) { + if (reserve_memtype(addr[i], addr[i] + PAGE_SIZE, + _PAGE_CACHE_UC_MINUS, NULL)) + goto out; + } + + return change_page_attr_set(addr, addrinarray, + __pgprot(_PAGE_CACHE_UC_MINUS), 1); +out: + while (--i >= 0) + free_memtype(addr[i], addr[i] + PAGE_SIZE); + return -EINVAL; +} +EXPORT_SYMBOL(set_memory_array_uc); + int _set_memory_wc(unsigned long addr, int numpages) { - return change_page_attr_set(addr, numpages, - __pgprot(_PAGE_CACHE_WC)); + return change_page_attr_set(&addr, numpages, + __pgprot(_PAGE_CACHE_WC), 0); } int set_memory_wc(unsigned long addr, int numpages) @@ -876,8 +977,8 @@ EXPORT_SYMBOL(set_memory_wc); int _set_memory_wb(unsigned long addr, int numpages) { - return change_page_attr_clear(addr, numpages, - __pgprot(_PAGE_CACHE_MASK)); + return change_page_attr_clear(&addr, numpages, + __pgprot(_PAGE_CACHE_MASK), 0); } int set_memory_wb(unsigned long addr, int numpages) @@ -888,37 +989,48 @@ int set_memory_wb(unsigned long addr, int numpages) } EXPORT_SYMBOL(set_memory_wb); +int set_memory_array_wb(unsigned long *addr, int addrinarray) +{ + int i; + for (i = 0; i < addrinarray; i++) + free_memtype(addr[i], addr[i] + PAGE_SIZE); + + return change_page_attr_clear(addr, addrinarray, + __pgprot(_PAGE_CACHE_MASK), 1); +} +EXPORT_SYMBOL(set_memory_array_wb); + int set_memory_x(unsigned long addr, int numpages) { - return change_page_attr_clear(addr, numpages, __pgprot(_PAGE_NX)); + return change_page_attr_clear(&addr, numpages, __pgprot(_PAGE_NX), 0); } EXPORT_SYMBOL(set_memory_x); int set_memory_nx(unsigned long addr, int numpages) { - return change_page_attr_set(addr, numpages, __pgprot(_PAGE_NX)); + return change_page_attr_set(&addr, numpages, __pgprot(_PAGE_NX), 0); } EXPORT_SYMBOL(set_memory_nx); int set_memory_ro(unsigned long addr, int numpages) { - return change_page_attr_clear(addr, numpages, __pgprot(_PAGE_RW)); + return change_page_attr_clear(&addr, numpages, __pgprot(_PAGE_RW), 0); } int set_memory_rw(unsigned long addr, int numpages) { - return change_page_attr_set(addr, numpages, __pgprot(_PAGE_RW)); + return change_page_attr_set(&addr, numpages, __pgprot(_PAGE_RW), 0); } int set_memory_np(unsigned long addr, int numpages) { - return change_page_attr_clear(addr, numpages, __pgprot(_PAGE_PRESENT)); + return change_page_attr_clear(&addr, numpages, __pgprot(_PAGE_PRESENT), 0); } int set_memory_4k(unsigned long addr, int numpages) { - return change_page_attr_set_clr(addr, numpages, __pgprot(0), - __pgprot(0), 1); + return change_page_attr_set_clr(&addr, numpages, __pgprot(0), + __pgprot(0), 1, 0); } int set_pages_uc(struct page *page, int numpages) @@ -971,20 +1083,24 @@ int set_pages_rw(struct page *page, int numpages) static int __set_pages_p(struct page *page, int numpages) { - struct cpa_data cpa = { .vaddr = (unsigned long) page_address(page), + unsigned long tempaddr = (unsigned long) page_address(page); + struct cpa_data cpa = { .vaddr = &tempaddr, .numpages = numpages, .mask_set = __pgprot(_PAGE_PRESENT | _PAGE_RW), - .mask_clr = __pgprot(0)}; + .mask_clr = __pgprot(0), + .flags = 0}; return __change_page_attr_set_clr(&cpa, 1); } static int __set_pages_np(struct page *page, int numpages) { - struct cpa_data cpa = { .vaddr = (unsigned long) page_address(page), + unsigned long tempaddr = (unsigned long) page_address(page); + struct cpa_data cpa = { .vaddr = &tempaddr, .numpages = numpages, .mask_set = __pgprot(0), - .mask_clr = __pgprot(_PAGE_PRESENT | _PAGE_RW)}; + .mask_clr = __pgprot(_PAGE_PRESENT | _PAGE_RW), + .flags = 0}; return __change_page_attr_set_clr(&cpa, 1); } diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h index f4c0ab50d2c..0a5f71817b3 100644 --- a/include/asm-x86/cacheflush.h +++ b/include/asm-x86/cacheflush.h @@ -66,6 +66,9 @@ int set_memory_rw(unsigned long addr, int numpages); int set_memory_np(unsigned long addr, int numpages); int set_memory_4k(unsigned long addr, int numpages); +int set_memory_array_uc(unsigned long *addr, int addrinarray); +int set_memory_array_wb(unsigned long *addr, int addrinarray); + /* * For legacy compatibility with the old APIs, a few functions * are provided that work on a "struct page". -- cgit v1.2.3-70-g09d2 From f86399396ce7a4f4069828b7dceac5aa5113dfb5 Mon Sep 17 00:00:00 2001 From: Eduardo Habkost Date: Wed, 30 Jul 2008 18:32:27 -0300 Subject: x86, paravirt_ops: use unsigned long instead of u32 for alloc_p*() pfn args This patch changes the pfn args from 'u32' to 'unsigned long' on alloc_p*() functions on paravirt_ops, and the corresponding implementations for Xen and VMI. The prototypes for CONFIG_PARAVIRT=n are already using unsigned long, so paravirt.h now matches the prototypes on asm-x86/pgalloc.h. It shouldn't result in any changes on generated code on 32-bit, with or without CONFIG_PARAVIRT. On both cases, 'codiff -f' didn't show any change after applying this patch. On 64-bit, there are (expected) binary changes only when CONFIG_PARAVIRT is enabled, as the patch is really supposed to change the size of the pfn args. [ v2: KVM_GUEST: use the right parameter type on kvm_release_pt() ] Signed-off-by: Eduardo Habkost Acked-by: Jeremy Fitzhardinge Acked-by: Zachary Amsden Signed-off-by: Ingo Molnar --- arch/x86/kernel/kvm.c | 2 +- arch/x86/kernel/vmi_32.c | 10 +++++----- arch/x86/xen/enlighten.c | 20 ++++++++++---------- include/asm-x86/paravirt.h | 30 +++++++++++++++--------------- 4 files changed, 31 insertions(+), 31 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index 8b7a3cf37d2..478bca986ec 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -178,7 +178,7 @@ static void kvm_flush_tlb(void) kvm_deferred_mmu_op(&ftlb, sizeof ftlb); } -static void kvm_release_pt(u32 pfn) +static void kvm_release_pt(unsigned long pfn) { struct kvm_mmu_op_release_pt rpt = { .header.op = KVM_MMU_OP_RELEASE_PT, diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index 0a1b1a9d922..340b03643b9 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -392,13 +392,13 @@ static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type) } #endif -static void vmi_allocate_pte(struct mm_struct *mm, u32 pfn) +static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn) { vmi_set_page_type(pfn, VMI_PAGE_L1); vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0); } -static void vmi_allocate_pmd(struct mm_struct *mm, u32 pfn) +static void vmi_allocate_pmd(struct mm_struct *mm, unsigned long pfn) { /* * This call comes in very early, before mem_map is setup. @@ -409,20 +409,20 @@ static void vmi_allocate_pmd(struct mm_struct *mm, u32 pfn) vmi_ops.allocate_page(pfn, VMI_PAGE_L2, 0, 0, 0); } -static void vmi_allocate_pmd_clone(u32 pfn, u32 clonepfn, u32 start, u32 count) +static void vmi_allocate_pmd_clone(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count) { vmi_set_page_type(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE); vmi_check_page_type(clonepfn, VMI_PAGE_L2); vmi_ops.allocate_page(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE, clonepfn, start, count); } -static void vmi_release_pte(u32 pfn) +static void vmi_release_pte(unsigned long pfn) { vmi_ops.release_page(pfn, VMI_PAGE_L1); vmi_set_page_type(pfn, VMI_PAGE_NORMAL); } -static void vmi_release_pmd(u32 pfn) +static void vmi_release_pmd(unsigned long pfn) { vmi_ops.release_page(pfn, VMI_PAGE_L2); vmi_set_page_type(pfn, VMI_PAGE_NORMAL); diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 9ff6e3cbf08..db970bdc5e3 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -812,7 +812,7 @@ static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high) /* Early in boot, while setting up the initial pagetable, assume everything is pinned. */ -static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn) +static __init void xen_alloc_pte_init(struct mm_struct *mm, unsigned long pfn) { #ifdef CONFIG_FLATMEM BUG_ON(mem_map); /* should only be used early */ @@ -822,7 +822,7 @@ static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn) /* Early release_pte assumes that all pts are pinned, since there's only init_mm and anything attached to that is pinned. */ -static void xen_release_pte_init(u32 pfn) +static void xen_release_pte_init(unsigned long pfn) { make_lowmem_page_readwrite(__va(PFN_PHYS(pfn))); } @@ -838,7 +838,7 @@ static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn) /* This needs to make sure the new pte page is pinned iff its being attached to a pinned pagetable. */ -static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level) +static void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, unsigned level) { struct page *page = pfn_to_page(pfn); @@ -856,12 +856,12 @@ static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level) } } -static void xen_alloc_pte(struct mm_struct *mm, u32 pfn) +static void xen_alloc_pte(struct mm_struct *mm, unsigned long pfn) { xen_alloc_ptpage(mm, pfn, PT_PTE); } -static void xen_alloc_pmd(struct mm_struct *mm, u32 pfn) +static void xen_alloc_pmd(struct mm_struct *mm, unsigned long pfn) { xen_alloc_ptpage(mm, pfn, PT_PMD); } @@ -909,7 +909,7 @@ static void xen_pgd_free(struct mm_struct *mm, pgd_t *pgd) } /* This should never happen until we're OK to use struct page */ -static void xen_release_ptpage(u32 pfn, unsigned level) +static void xen_release_ptpage(unsigned long pfn, unsigned level) { struct page *page = pfn_to_page(pfn); @@ -923,23 +923,23 @@ static void xen_release_ptpage(u32 pfn, unsigned level) } } -static void xen_release_pte(u32 pfn) +static void xen_release_pte(unsigned long pfn) { xen_release_ptpage(pfn, PT_PTE); } -static void xen_release_pmd(u32 pfn) +static void xen_release_pmd(unsigned long pfn) { xen_release_ptpage(pfn, PT_PMD); } #if PAGETABLE_LEVELS == 4 -static void xen_alloc_pud(struct mm_struct *mm, u32 pfn) +static void xen_alloc_pud(struct mm_struct *mm, unsigned long pfn) { xen_alloc_ptpage(mm, pfn, PT_PUD); } -static void xen_release_pud(u32 pfn) +static void xen_release_pud(unsigned long pfn) { xen_release_ptpage(pfn, PT_PUD); } diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index fbbde93f12d..497aea0f41a 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -257,13 +257,13 @@ struct pv_mmu_ops { * Hooks for allocating/releasing pagetable pages when they're * attached to a pagetable */ - void (*alloc_pte)(struct mm_struct *mm, u32 pfn); - void (*alloc_pmd)(struct mm_struct *mm, u32 pfn); - void (*alloc_pmd_clone)(u32 pfn, u32 clonepfn, u32 start, u32 count); - void (*alloc_pud)(struct mm_struct *mm, u32 pfn); - void (*release_pte)(u32 pfn); - void (*release_pmd)(u32 pfn); - void (*release_pud)(u32 pfn); + void (*alloc_pte)(struct mm_struct *mm, unsigned long pfn); + void (*alloc_pmd)(struct mm_struct *mm, unsigned long pfn); + void (*alloc_pmd_clone)(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count); + void (*alloc_pud)(struct mm_struct *mm, unsigned long pfn); + void (*release_pte)(unsigned long pfn); + void (*release_pmd)(unsigned long pfn); + void (*release_pud)(unsigned long pfn); /* Pagetable manipulation functions */ void (*set_pte)(pte_t *ptep, pte_t pteval); @@ -993,35 +993,35 @@ static inline void paravirt_pgd_free(struct mm_struct *mm, pgd_t *pgd) PVOP_VCALL2(pv_mmu_ops.pgd_free, mm, pgd); } -static inline void paravirt_alloc_pte(struct mm_struct *mm, unsigned pfn) +static inline void paravirt_alloc_pte(struct mm_struct *mm, unsigned long pfn) { PVOP_VCALL2(pv_mmu_ops.alloc_pte, mm, pfn); } -static inline void paravirt_release_pte(unsigned pfn) +static inline void paravirt_release_pte(unsigned long pfn) { PVOP_VCALL1(pv_mmu_ops.release_pte, pfn); } -static inline void paravirt_alloc_pmd(struct mm_struct *mm, unsigned pfn) +static inline void paravirt_alloc_pmd(struct mm_struct *mm, unsigned long pfn) { PVOP_VCALL2(pv_mmu_ops.alloc_pmd, mm, pfn); } -static inline void paravirt_alloc_pmd_clone(unsigned pfn, unsigned clonepfn, - unsigned start, unsigned count) +static inline void paravirt_alloc_pmd_clone(unsigned long pfn, unsigned long clonepfn, + unsigned long start, unsigned long count) { PVOP_VCALL4(pv_mmu_ops.alloc_pmd_clone, pfn, clonepfn, start, count); } -static inline void paravirt_release_pmd(unsigned pfn) +static inline void paravirt_release_pmd(unsigned long pfn) { PVOP_VCALL1(pv_mmu_ops.release_pmd, pfn); } -static inline void paravirt_alloc_pud(struct mm_struct *mm, unsigned pfn) +static inline void paravirt_alloc_pud(struct mm_struct *mm, unsigned long pfn) { PVOP_VCALL2(pv_mmu_ops.alloc_pud, mm, pfn); } -static inline void paravirt_release_pud(unsigned pfn) +static inline void paravirt_release_pud(unsigned long pfn) { PVOP_VCALL1(pv_mmu_ops.release_pud, pfn); } -- cgit v1.2.3-70-g09d2 From 6c505ce3930c6a6b455cda53fab3e88ae44f8221 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 19 Aug 2008 16:32:45 +0200 Subject: x86: move dma_*_coherent functions to include file All the x86 DMA-API functions are defined in asm/dma-mapping.h. This patch moves the dma_*_coherent functions also to this header file because they are now small enough to do so. This is done as a separate patch because it also includes some renaming and restructuring of the dma-mapping.h file. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/pci-dma.c | 49 +++---------------------------------------- arch/x86/kernel/pci-gart_64.c | 4 ++-- include/asm-x86/dma-mapping.h | 47 ++++++++++++++++++++++++++++++++--------- 3 files changed, 42 insertions(+), 58 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 613332b26e3..0a1408abcc6 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -41,11 +41,12 @@ EXPORT_SYMBOL(bad_dma_address); /* Dummy device used for NULL arguments (normally ISA). Better would be probably a smaller DMA mask, but this is bug-to-bug compatible to older i386. */ -struct device fallback_dev = { +struct device x86_dma_fallback_dev = { .bus_id = "fallback device", .coherent_dma_mask = DMA_32BIT_MASK, - .dma_mask = &fallback_dev.coherent_dma_mask, + .dma_mask = &x86_dma_fallback_dev.coherent_dma_mask, }; +EXPORT_SYMBOL(x86_dma_fallback_dev); int dma_set_mask(struct device *dev, u64 mask) { @@ -241,50 +242,6 @@ int dma_supported(struct device *dev, u64 mask) } EXPORT_SYMBOL(dma_supported); -/* - * Allocate memory for a coherent mapping. - */ - void * -dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, - gfp_t gfp) -{ - struct dma_mapping_ops *ops = get_dma_ops(dev); - void *memory; - - if (dma_alloc_from_coherent(dev, size, dma_handle, &memory)) - return memory; - - if (!dev) { - dev = &fallback_dev; - gfp |= GFP_DMA; - } - - if (ops->alloc_coherent) - return ops->alloc_coherent(dev, size, - dma_handle, gfp); - return NULL; -} -EXPORT_SYMBOL(dma_alloc_coherent); - -/* - * Unmap coherent memory. - * The caller must ensure that the device has finished accessing the mapping. - */ -void dma_free_coherent(struct device *dev, size_t size, - void *vaddr, dma_addr_t bus) -{ - struct dma_mapping_ops *ops = get_dma_ops(dev); - - WARN_ON(irqs_disabled()); /* for portability */ - - if (dma_release_from_coherent(dev, get_order(size), vaddr)) - return; - - if (ops->free_coherent) - ops->free_coherent(dev, size, vaddr, bus); -} -EXPORT_SYMBOL(dma_free_coherent); - static int __init pci_iommu_init(void) { calgary_iommu_init(); diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index ef753e23358..e81df25dc06 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -276,7 +276,7 @@ gart_map_single(struct device *dev, phys_addr_t paddr, size_t size, int dir) unsigned long bus; if (!dev) - dev = &fallback_dev; + dev = &x86_dma_fallback_dev; if (!need_iommu(dev, paddr, size)) return paddr; @@ -427,7 +427,7 @@ gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, int dir) return 0; if (!dev) - dev = &fallback_dev; + dev = &x86_dma_fallback_dev; out = 0; start = 0; diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index ad9cd6d49bf..8e16095d1fa 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -9,10 +9,11 @@ #include #include #include +#include extern dma_addr_t bad_dma_address; extern int iommu_merge; -extern struct device fallback_dev; +extern struct device x86_dma_fallback_dev; extern int panic_on_overflow; extern int force_iommu; @@ -87,13 +88,7 @@ static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f) #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) - -void *dma_alloc_coherent(struct device *dev, size_t size, - dma_addr_t *dma_handle, gfp_t flag); - -void dma_free_coherent(struct device *dev, size_t size, - void *vaddr, dma_addr_t dma_handle); - +#define dma_is_consistent(d, h) (1) extern int dma_supported(struct device *hwdev, u64 mask); extern int dma_set_mask(struct device *dev, u64 mask); @@ -247,7 +242,39 @@ static inline int dma_get_cache_alignment(void) return boot_cpu_data.x86_clflush_size; } -#define dma_is_consistent(d, h) (1) +static inline void * +dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, + gfp_t gfp) +{ + struct dma_mapping_ops *ops = get_dma_ops(dev); + void *memory; + + if (dma_alloc_from_coherent(dev, size, dma_handle, &memory)) + return memory; + + if (!dev) { + dev = &x86_dma_fallback_dev; + gfp |= GFP_DMA; + } + + if (ops->alloc_coherent) + return ops->alloc_coherent(dev, size, + dma_handle, gfp); + return NULL; +} + +static inline void dma_free_coherent(struct device *dev, size_t size, + void *vaddr, dma_addr_t bus) +{ + struct dma_mapping_ops *ops = get_dma_ops(dev); + + WARN_ON(irqs_disabled()); /* for portability */ + + if (dma_release_from_coherent(dev, get_order(size), vaddr)) + return; + + if (ops->free_coherent) + ops->free_coherent(dev, size, vaddr, bus); +} -#include #endif -- cgit v1.2.3-70-g09d2 From 766af9fa812f49feb4a3e62cf92f3d37f33c7fb6 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Fri, 22 Aug 2008 00:12:09 +0900 Subject: dma-mapping.h, x86: remove last user of dma_mapping_ops->map_simple pci-dma.c doesn't use map_simple hook any more so we can remove it from struct dma_mapping_ops now. Signed-off-by: FUJITA Tomonori Signed-off-by: Ingo Molnar --- arch/x86/kernel/pci-gart_64.c | 1 - include/asm-x86/dma-mapping.h | 3 --- 2 files changed, 4 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index e81df25dc06..cfd7ec25e37 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -720,7 +720,6 @@ extern int agp_amd64_init(void); static struct dma_mapping_ops gart_dma_ops = { .map_single = gart_map_single, - .map_simple = gart_map_simple, .unmap_single = gart_unmap_single, .sync_single_for_cpu = NULL, .sync_single_for_device = NULL, diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index 8e16095d1fa..3a9a6f5e681 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -26,9 +26,6 @@ struct dma_mapping_ops { void *vaddr, dma_addr_t dma_handle); dma_addr_t (*map_single)(struct device *hwdev, phys_addr_t ptr, size_t size, int direction); - /* like map_single, but doesn't check the device mask */ - dma_addr_t (*map_simple)(struct device *hwdev, phys_addr_t ptr, - size_t size, int direction); void (*unmap_single)(struct device *dev, dma_addr_t addr, size_t size, int direction); void (*sync_single_for_cpu)(struct device *hwdev, -- cgit v1.2.3-70-g09d2 From b05f78f5c713eda2c34e495d92495ee4f1c3b5e1 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 22 Aug 2008 01:32:50 -0700 Subject: x86_64: printout msr -v2 commandline show_msr=1 for bsp, show_msr=32 for all 32 cpus. [ mingo@elte.hu: added documentation ] Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- Documentation/kernel-parameters.txt | 6 +++++ arch/x86/kernel/cpu/common_64.c | 51 +++++++++++++++++++++++++++++++++++++ arch/x86/kernel/paravirt.c | 1 + include/asm-x86/msr.h | 23 +++++++++++++++++ include/asm-x86/paravirt.h | 12 +++++++++ 5 files changed, 93 insertions(+) (limited to 'include/asm-x86') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 47e7d8794fc..8679e80b9fc 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1852,6 +1852,12 @@ and is between 256 and 4096 characters. It is defined in the file shapers= [NET] Maximal number of shapers. + show_msr= [x86] show boot-time MSR settings + Format: { } + Show boot-time (BIOS-initialized) MSR settings. + The parameter means the number of CPUs to show, + for example 1 means boot CPU only. + sim710= [SCSI,HW] See header of drivers/scsi/sim710.c. diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c index dd6e3f15017..bca2d6980e8 100644 --- a/arch/x86/kernel/cpu/common_64.c +++ b/arch/x86/kernel/cpu/common_64.c @@ -394,6 +394,49 @@ static __init int setup_noclflush(char *arg) } __setup("noclflush", setup_noclflush); +struct msr_range { + unsigned min; + unsigned max; +}; + +static struct msr_range msr_range_array[] __cpuinitdata = { + { 0x00000000, 0x00000418}, + { 0xc0000000, 0xc000040b}, + { 0xc0010000, 0xc0010142}, + { 0xc0011000, 0xc001103b}, +}; + +static void __cpuinit print_cpu_msr(void) +{ + unsigned index; + u64 val; + int i; + unsigned index_min, index_max; + + for (i = 0; i < ARRAY_SIZE(msr_range_array); i++) { + index_min = msr_range_array[i].min; + index_max = msr_range_array[i].max; + for (index = index_min; index < index_max; index++) { + if (rdmsrl_amd_safe(index, &val)) + continue; + printk(KERN_INFO " MSR%08x: %016llx\n", index, val); + } + } +} + +static int show_msr __cpuinitdata; +static __init int setup_show_msr(char *arg) +{ + int num; + + get_option(&arg, &num); + + if (num > 0) + show_msr = num; + return 1; +} +__setup("show_msr=", setup_show_msr); + void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) { if (c->x86_model_id[0]) @@ -403,6 +446,14 @@ void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) printk(KERN_CONT " stepping %02x\n", c->x86_mask); else printk(KERN_CONT "\n"); + +#ifdef CONFIG_SMP + if (c->cpu_index < show_msr) + print_cpu_msr(); +#else + if (show_msr) + print_cpu_msr(); +#endif } static __init int setup_disablecpuid(char *arg) diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 94da4d52d79..c6044682e1e 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -330,6 +330,7 @@ struct pv_cpu_ops pv_cpu_ops = { #endif .wbinvd = native_wbinvd, .read_msr = native_read_msr_safe, + .read_msr_amd = native_read_msr_amd_safe, .write_msr = native_write_msr_safe, .read_tsc = native_read_tsc, .read_pmc = native_read_pmc, diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h index ca110ee73f0..a30e586df93 100644 --- a/include/asm-x86/msr.h +++ b/include/asm-x86/msr.h @@ -63,6 +63,22 @@ static inline unsigned long long native_read_msr_safe(unsigned int msr, return EAX_EDX_VAL(val, low, high); } +static inline unsigned long long native_read_msr_amd_safe(unsigned int msr, + int *err) +{ + DECLARE_ARGS(val, low, high); + + asm volatile("2: rdmsr ; xor %0,%0\n" + "1:\n\t" + ".section .fixup,\"ax\"\n\t" + "3: mov %3,%0 ; jmp 1b\n\t" + ".previous\n\t" + _ASM_EXTABLE(2b, 3b) + : "=r" (*err), EAX_EDX_RET(val, low, high) + : "c" (msr), "D" (0x9c5a203a), "i" (-EFAULT)); + return EAX_EDX_VAL(val, low, high); +} + static inline void native_write_msr(unsigned int msr, unsigned low, unsigned high) { @@ -158,6 +174,13 @@ static inline int rdmsrl_safe(unsigned msr, unsigned long long *p) *p = native_read_msr_safe(msr, &err); return err; } +static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p) +{ + int err; + + *p = native_read_msr_amd_safe(msr, &err); + return err; +} #define rdtscl(low) \ ((low) = (u32)native_read_tsc()) diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index fbbde93f12d..d5cfc5e3eb5 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -137,6 +137,7 @@ struct pv_cpu_ops { /* MSR, PMC and TSR operations. err = 0/-EFAULT. wrmsr returns 0/-EFAULT. */ + u64 (*read_msr_amd)(unsigned int msr, int *err); u64 (*read_msr)(unsigned int msr, int *err); int (*write_msr)(unsigned int msr, unsigned low, unsigned high); @@ -726,6 +727,10 @@ static inline u64 paravirt_read_msr(unsigned msr, int *err) { return PVOP_CALL2(u64, pv_cpu_ops.read_msr, msr, err); } +static inline u64 paravirt_read_msr_amd(unsigned msr, int *err) +{ + return PVOP_CALL2(u64, pv_cpu_ops.read_msr_amd, msr, err); +} static inline int paravirt_write_msr(unsigned msr, unsigned low, unsigned high) { return PVOP_CALL3(int, pv_cpu_ops.write_msr, msr, low, high); @@ -771,6 +776,13 @@ static inline int rdmsrl_safe(unsigned msr, unsigned long long *p) *p = paravirt_read_msr(msr, &err); return err; } +static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p) +{ + int err; + + *p = paravirt_read_msr_amd(msr, &err); + return err; +} static inline u64 paravirt_read_tsc(void) { -- cgit v1.2.3-70-g09d2 From bbb65d2d365efe9951290e61678dcf81ec60add4 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Sat, 23 Aug 2008 17:47:10 +0200 Subject: x86: use cpuid vector 0xb when available for detecting cpu topology cpuid leaf 0xb provides extended topology enumeration. This interface provides the 32-bit x2APIC id of the logical processor and it also provides a new mechanism to detect SMT and core siblings (which provides increased addressability). Signed-off-by: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/addon_cpuid_features.c | 86 ++++++++++++++++++++++++++++++ arch/x86/kernel/cpu/common_64.c | 3 ++ arch/x86/kernel/cpu/intel.c | 13 +++-- arch/x86/kernel/cpu/intel_64.c | 5 +- include/asm-x86/cpufeature.h | 1 + include/asm-x86/processor.h | 1 + 6 files changed, 105 insertions(+), 4 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c index 84a8220a607..aa9641ae670 100644 --- a/arch/x86/kernel/cpu/addon_cpuid_features.c +++ b/arch/x86/kernel/cpu/addon_cpuid_features.c @@ -7,6 +7,8 @@ #include #include +#include + struct cpuid_bit { u16 feature; u8 reg; @@ -48,6 +50,90 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) } } +/* leaf 0xb SMT level */ +#define SMT_LEVEL 0 + +/* leaf 0xb sub-leaf types */ +#define INVALID_TYPE 0 +#define SMT_TYPE 1 +#define CORE_TYPE 2 + +#define LEAFB_SUBTYPE(ecx) (((ecx) >> 8) & 0xff) +#define BITS_SHIFT_NEXT_LEVEL(eax) ((eax) & 0x1f) +#define LEVEL_MAX_SIBLINGS(ebx) ((ebx) & 0xffff) + +/* + * Check for extended topology enumeration cpuid leaf 0xb and if it + * exists, use it for populating initial_apicid and cpu topology + * detection. + */ +void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c) +{ + unsigned int eax, ebx, ecx, edx, sub_index; + unsigned int ht_mask_width, core_plus_mask_width; + unsigned int core_select_mask, core_level_siblings; + + if (c->cpuid_level < 0xb) + return; + + cpuid_count(0xb, SMT_LEVEL, &eax, &ebx, &ecx, &edx); + + /* + * check if the cpuid leaf 0xb is actually implemented. + */ + if (ebx == 0 || (LEAFB_SUBTYPE(ecx) != SMT_TYPE)) + return; + + set_cpu_cap(c, X86_FEATURE_XTOPOLOGY); + + /* + * initial apic id, which also represents 32-bit extended x2apic id. + */ + c->initial_apicid = edx; + + /* + * Populate HT related information from sub-leaf level 0. + */ + core_level_siblings = smp_num_siblings = LEVEL_MAX_SIBLINGS(ebx); + core_plus_mask_width = ht_mask_width = BITS_SHIFT_NEXT_LEVEL(eax); + + sub_index = 1; + do { + cpuid_count(0xb, sub_index, &eax, &ebx, &ecx, &edx); + + /* + * Check for the Core type in the implemented sub leaves. + */ + if (LEAFB_SUBTYPE(ecx) == CORE_TYPE) { + core_level_siblings = LEVEL_MAX_SIBLINGS(ebx); + core_plus_mask_width = BITS_SHIFT_NEXT_LEVEL(eax); + break; + } + + sub_index++; + } while (LEAFB_SUBTYPE(ecx) != INVALID_TYPE); + + core_select_mask = (~(-1 << core_plus_mask_width)) >> ht_mask_width; + +#ifdef CONFIG_X86_32 + c->cpu_core_id = phys_pkg_id(c->initial_apicid, ht_mask_width) + & core_select_mask; + c->phys_proc_id = phys_pkg_id(c->initial_apicid, core_plus_mask_width); +#else + c->cpu_core_id = phys_pkg_id(ht_mask_width) & core_select_mask; + c->phys_proc_id = phys_pkg_id(core_plus_mask_width); +#endif + c->x86_max_cores = (core_level_siblings / smp_num_siblings); + + + printk(KERN_INFO "CPU: Physical Processor ID: %d\n", + c->phys_proc_id); + if (c->x86_max_cores > 1) + printk(KERN_INFO "CPU: Processor Core ID: %d\n", + c->cpu_core_id); + return; +} + #ifdef CONFIG_X86_PAT void __cpuinit validate_pat_support(struct cpuinfo_x86 *c) { diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c index af569a964e7..a2888c7b56a 100644 --- a/arch/x86/kernel/cpu/common_64.c +++ b/arch/x86/kernel/cpu/common_64.c @@ -128,6 +128,9 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) u32 eax, ebx, ecx, edx; int index_msb, core_bits; + if (cpu_has(c, X86_FEATURE_XTOPOLOGY)) + return; + cpuid(1, &eax, &ebx, &ecx, &edx); diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 77618c717d7..58a6f1a0b29 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -176,9 +176,16 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) if (p) strcpy(c->x86_model_id, p); - c->x86_max_cores = num_cpu_cores(c); - - detect_ht(c); + detect_extended_topology(c); + + if (!cpu_has(c, X86_FEATURE_XTOPOLOGY)) { + /* + * let's use the legacy cpuid vector 0x1 and 0x4 for topology + * detection. + */ + c->x86_max_cores = num_cpu_cores(c); + detect_ht(c); + } /* Work around errata */ Intel_errata_workarounds(c); diff --git a/arch/x86/kernel/cpu/intel_64.c b/arch/x86/kernel/cpu/intel_64.c index 1019c58d39f..42d501a8c41 100644 --- a/arch/x86/kernel/cpu/intel_64.c +++ b/arch/x86/kernel/cpu/intel_64.c @@ -80,7 +80,10 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) if (c->x86 == 6) set_cpu_cap(c, X86_FEATURE_REP_GOOD); set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); - c->x86_max_cores = intel_num_cpu_cores(c); + + detect_extended_topology(c); + if (!cpu_has(c, X86_FEATURE_XTOPOLOGY)) + c->x86_max_cores = intel_num_cpu_cores(c); srat_detect_node(); } diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 5fc4d55906d..8d842af4cf7 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -81,6 +81,7 @@ #define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* Lfence synchronizes RDTSC */ #define X86_FEATURE_11AP (3*32+19) /* Bad local APIC aka 11AP */ #define X86_FEATURE_NOPL (3*32+20) /* The NOPL (0F 1F) instructions */ +#define X86_FEATURE_XTOPOLOGY (3*32+21) /* cpu topology enum extensions */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* Streaming SIMD Extensions-3 */ diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index 5f58da401b4..79338fe965d 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -161,6 +161,7 @@ extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c); extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c); extern unsigned short num_cache_leaves; +extern void detect_extended_topology(struct cpuinfo_x86 *c); #if defined(CONFIG_X86_HT) || defined(CONFIG_X86_64) extern void detect_ht(struct cpuinfo_x86 *c); #else -- cgit v1.2.3-70-g09d2 From 93be71b672f167b1e8c23725114f86305354f0ac Mon Sep 17 00:00:00 2001 From: Alex Nixon Date: Fri, 22 Aug 2008 11:52:11 +0100 Subject: x86: add cpu hotplug hooks into smp_ops Signed-off-by: Alex Nixon Acked-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/kernel/process_32.c | 4 ++-- arch/x86/kernel/process_64.c | 4 ++-- arch/x86/kernel/smp.c | 6 +++++- arch/x86/kernel/smpboot.c | 8 ++++---- include/asm-x86/smp.h | 28 ++++++++++++++++++++++++---- 5 files changed, 37 insertions(+), 13 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 3b7a1ddcc0b..e382fe0ccd6 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -91,7 +91,7 @@ static void cpu_exit_clear(void) } /* We don't actually take CPU down, just spin without interrupts. */ -static inline void play_dead(void) +void native_play_dead(void) { /* This must be done before dead CPU ack */ cpu_exit_clear(); @@ -107,7 +107,7 @@ static inline void play_dead(void) wbinvd_halt(); } #else -static inline void play_dead(void) +void native_play_dead(void) { BUG(); } diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 71553b664e2..dfd3f575208 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -90,7 +90,7 @@ DECLARE_PER_CPU(int, cpu_state); #include /* We halt the CPU with physical CPU hotplug */ -static inline void play_dead(void) +void native_play_dead(void) { idle_task_exit(); mb(); @@ -102,7 +102,7 @@ static inline void play_dead(void) wbinvd_halt(); } #else -static inline void play_dead(void) +void native_play_dead(void) { BUG(); } diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c index 361b7a4c640..18f9b19f5f8 100644 --- a/arch/x86/kernel/smp.c +++ b/arch/x86/kernel/smp.c @@ -214,12 +214,16 @@ void smp_call_function_single_interrupt(struct pt_regs *regs) struct smp_ops smp_ops = { .smp_prepare_boot_cpu = native_smp_prepare_boot_cpu, .smp_prepare_cpus = native_smp_prepare_cpus, - .cpu_up = native_cpu_up, .smp_cpus_done = native_smp_cpus_done, .smp_send_stop = native_smp_send_stop, .smp_send_reschedule = native_smp_send_reschedule, + .cpu_up = native_cpu_up, + .cpu_die = native_cpu_die, + .cpu_disable = native_cpu_disable, + .play_dead = native_play_dead, + .send_call_func_ipi = native_send_call_func_ipi, .send_call_func_single_ipi = native_send_call_func_single_ipi, }; diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 7985c5b3f91..c414cee296b 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1346,7 +1346,7 @@ static void __ref remove_cpu_from_maps(int cpu) numa_remove_cpu(cpu); } -int __cpu_disable(void) +int native_cpu_disable(void) { int cpu = smp_processor_id(); @@ -1385,7 +1385,7 @@ int __cpu_disable(void) return 0; } -void __cpu_die(unsigned int cpu) +void native_cpu_die(unsigned int cpu) { /* We don't do anything here: idle task is faking death itself. */ unsigned int i; @@ -1403,12 +1403,12 @@ void __cpu_die(unsigned int cpu) printk(KERN_ERR "CPU %u didn't die...\n", cpu); } #else /* ... !CONFIG_HOTPLUG_CPU */ -int __cpu_disable(void) +int native_cpu_disable(void) { return -ENOSYS; } -void __cpu_die(unsigned int cpu) +void native_cpu_die(unsigned int cpu) { /* We said "no" in __cpu_disable */ BUG(); diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 3c877f74f27..dbf4249e2a6 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -47,12 +47,16 @@ extern struct { struct smp_ops { void (*smp_prepare_boot_cpu)(void); void (*smp_prepare_cpus)(unsigned max_cpus); - int (*cpu_up)(unsigned cpu); void (*smp_cpus_done)(unsigned max_cpus); void (*smp_send_stop)(void); void (*smp_send_reschedule)(int cpu); + int (*cpu_up)(unsigned cpu); + int (*cpu_disable)(void); + void (*cpu_die)(unsigned int cpu); + void (*play_dead)(void); + void (*send_call_func_ipi)(cpumask_t mask); void (*send_call_func_single_ipi)(int cpu); }; @@ -91,6 +95,21 @@ static inline int __cpu_up(unsigned int cpu) return smp_ops.cpu_up(cpu); } +static inline int __cpu_disable(void) +{ + return smp_ops.cpu_disable(); +} + +static inline void __cpu_die(unsigned int cpu) +{ + smp_ops.cpu_die(cpu); +} + +static inline void play_dead(void) +{ + smp_ops.play_dead(); +} + static inline void smp_send_reschedule(int cpu) { smp_ops.smp_send_reschedule(cpu); @@ -110,12 +129,13 @@ void native_smp_prepare_boot_cpu(void); void native_smp_prepare_cpus(unsigned int max_cpus); void native_smp_cpus_done(unsigned int max_cpus); int native_cpu_up(unsigned int cpunum); +int native_cpu_disable(void); +void native_cpu_die(unsigned int cpu); +void native_play_dead(void); + void native_send_call_func_ipi(cpumask_t mask); void native_send_call_func_single_ipi(int cpu); -extern int __cpu_disable(void); -extern void __cpu_die(unsigned int cpu); - void smp_store_cpu_info(int id); #define cpu_physical_id(cpu) per_cpu(x86_cpu_to_apicid, cpu) -- cgit v1.2.3-70-g09d2 From 379002586368ae22916f668011c9118c8ce8189c Mon Sep 17 00:00:00 2001 From: Alex Nixon Date: Fri, 22 Aug 2008 11:52:12 +0100 Subject: x86_32: clean up play_dead The removal of the CPU from the various maps was redundant as it already happened in cpu_disable. After cleaning this up, cpu_uninit only resets the tlb state, so rename it and create a noop version for the X86_64 case (so the two play_deads can be unified later). Signed-off-by: Alex Nixon Acked-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/common.c | 6 +----- arch/x86/kernel/process_32.c | 17 ++++------------- include/asm-x86/smp.h | 7 ++++++- 3 files changed, 11 insertions(+), 19 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 80ab20d4fa3..18f5551b32d 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -714,14 +714,10 @@ void __cpuinit cpu_init(void) mxcsr_feature_mask_init(); } -#ifdef CONFIG_HOTPLUG_CPU -void __cpuinit cpu_uninit(void) +void reset_lazy_tlbstate(void) { int cpu = raw_smp_processor_id(); - cpu_clear(cpu, cpu_initialized); - /* lazy TLB state */ per_cpu(cpu_tlbstate, cpu).state = 0; per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm; } -#endif diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index e382fe0ccd6..d55eb49584b 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -75,26 +75,17 @@ unsigned long thread_saved_pc(struct task_struct *tsk) #ifdef CONFIG_HOTPLUG_CPU #include -static void cpu_exit_clear(void) +/* We don't actually take CPU down, just spin without interrupts. */ +void native_play_dead(void) { int cpu = raw_smp_processor_id(); idle_task_exit(); - cpu_uninit(); - irq_ctx_exit(cpu); + reset_lazy_tlbstate(); - cpu_clear(cpu, cpu_callout_map); - cpu_clear(cpu, cpu_callin_map); - - numa_remove_cpu(cpu); -} + irq_ctx_exit(cpu); -/* We don't actually take CPU down, just spin without interrupts. */ -void native_play_dead(void) -{ - /* This must be done before dead CPU ack */ - cpu_exit_clear(); mb(); /* Ack it */ __get_cpu_var(cpu_state) = CPU_DEAD; diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index dbf4249e2a6..3bec6b47821 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -221,7 +221,12 @@ static inline int hard_smp_processor_id(void) #endif /* CONFIG_X86_LOCAL_APIC */ #ifdef CONFIG_HOTPLUG_CPU -extern void cpu_uninit(void); +#ifdef CONFIG_X86_32 +extern void reset_lazy_tlbstate(void); +#else +static inline void reset_lazy_tlbstate(void) +{ } +#endif /* CONFIG_X86_32 */ #endif #endif /* __ASSEMBLY__ */ -- cgit v1.2.3-70-g09d2 From a21f5d88c17a40941f6239d1959d89e8493e8e01 Mon Sep 17 00:00:00 2001 From: Alex Nixon Date: Fri, 22 Aug 2008 11:52:13 +0100 Subject: x86: unify x86_32 and x86_64 play_dead into one function Add the new play_dead into smpboot.c, as it fits more cleanly in there alongside other CONFIG_HOTPLUG functions. Separate out the common code into its own function. Signed-off-by: Alex Nixon Acked-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/kernel/process_32.c | 32 -------------------------------- arch/x86/kernel/process_64.c | 23 ----------------------- arch/x86/kernel/smpboot.c | 29 +++++++++++++++++++++++++++++ include/asm-x86/smp.h | 1 + 4 files changed, 30 insertions(+), 55 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index d55eb49584b..633cf06578f 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -72,38 +72,6 @@ unsigned long thread_saved_pc(struct task_struct *tsk) return ((unsigned long *)tsk->thread.sp)[3]; } -#ifdef CONFIG_HOTPLUG_CPU -#include - -/* We don't actually take CPU down, just spin without interrupts. */ -void native_play_dead(void) -{ - int cpu = raw_smp_processor_id(); - - idle_task_exit(); - - reset_lazy_tlbstate(); - - irq_ctx_exit(cpu); - - mb(); - /* Ack it */ - __get_cpu_var(cpu_state) = CPU_DEAD; - - /* - * With physical CPU hotplug, we should halt the cpu - */ - local_irq_disable(); - /* mask all interrupts, flush any and all caches, and halt */ - wbinvd_halt(); -} -#else -void native_play_dead(void) -{ - BUG(); -} -#endif /* CONFIG_HOTPLUG_CPU */ - /* * The idle thread. There's no useful work to be * done, so just try to conserve power and have a diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index dfd3f575208..aa28ed01cdf 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -85,29 +85,6 @@ void exit_idle(void) __exit_idle(); } -#ifdef CONFIG_HOTPLUG_CPU -DECLARE_PER_CPU(int, cpu_state); - -#include -/* We halt the CPU with physical CPU hotplug */ -void native_play_dead(void) -{ - idle_task_exit(); - mb(); - /* Ack it */ - __get_cpu_var(cpu_state) = CPU_DEAD; - - local_irq_disable(); - /* mask all interrupts, flush any and all caches, and halt */ - wbinvd_halt(); -} -#else -void native_play_dead(void) -{ - BUG(); -} -#endif /* CONFIG_HOTPLUG_CPU */ - /* * The idle thread. There's no useful work to be * done, so just try to conserve power and have a diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index c414cee296b..28b4287296a 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1402,6 +1402,29 @@ void native_cpu_die(unsigned int cpu) } printk(KERN_ERR "CPU %u didn't die...\n", cpu); } + +void play_dead_common(void) +{ + idle_task_exit(); + reset_lazy_tlbstate(); + irq_ctx_exit(raw_smp_processor_id()); + + mb(); + /* Ack it */ + __get_cpu_var(cpu_state) = CPU_DEAD; + + /* + * With physical CPU hotplug, we should halt the cpu + */ + local_irq_disable(); +} + +void native_play_dead(void) +{ + play_dead_common(); + wbinvd_halt(); +} + #else /* ... !CONFIG_HOTPLUG_CPU */ int native_cpu_disable(void) { @@ -1413,4 +1436,10 @@ void native_cpu_die(unsigned int cpu) /* We said "no" in __cpu_disable */ BUG(); } + +void native_play_dead(void) +{ + BUG(); +} + #endif diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 3bec6b47821..17305e72f34 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -132,6 +132,7 @@ int native_cpu_up(unsigned int cpunum); int native_cpu_disable(void); void native_cpu_die(unsigned int cpu); void native_play_dead(void); +void play_dead_common(void); void native_send_call_func_ipi(cpumask_t mask); void native_send_call_func_single_ipi(int cpu); -- cgit v1.2.3-70-g09d2 From 8227dce7dc2cfdcc28ee0eadfb482a7ee77fba03 Mon Sep 17 00:00:00 2001 From: Alex Nixon Date: Fri, 22 Aug 2008 11:52:14 +0100 Subject: x86: separate generic cpu disabling code from APIC writes in cpu_disable It allows paravirt implementations of cpu_disable to share the cpu_disable_common code, without having to take on board APIC writes, which may not be appropriate. Signed-off-by: Alex Nixon Acked-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/kernel/smpboot.c | 40 +++++++++++++++++++++++----------------- include/asm-x86/smp.h | 1 + 2 files changed, 24 insertions(+), 17 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 28b4287296a..66b04e59881 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1346,25 +1346,9 @@ static void __ref remove_cpu_from_maps(int cpu) numa_remove_cpu(cpu); } -int native_cpu_disable(void) +void cpu_disable_common(void) { int cpu = smp_processor_id(); - - /* - * Perhaps use cpufreq to drop frequency, but that could go - * into generic code. - * - * We won't take down the boot processor on i386 due to some - * interrupts only being able to be serviced by the BSP. - * Especially so if we're not using an IOAPIC -zwane - */ - if (cpu == 0) - return -EBUSY; - - if (nmi_watchdog == NMI_LOCAL_APIC) - stop_apic_nmi_watchdog(NULL); - clear_local_APIC(); - /* * HACK: * Allow any queued timer interrupts to get serviced @@ -1382,6 +1366,28 @@ int native_cpu_disable(void) remove_cpu_from_maps(cpu); unlock_vector_lock(); fixup_irqs(cpu_online_map); +} + +int native_cpu_disable(void) +{ + int cpu = smp_processor_id(); + + /* + * Perhaps use cpufreq to drop frequency, but that could go + * into generic code. + * + * We won't take down the boot processor on i386 due to some + * interrupts only being able to be serviced by the BSP. + * Especially so if we're not using an IOAPIC -zwane + */ + if (cpu == 0) + return -EBUSY; + + if (nmi_watchdog == NMI_LOCAL_APIC) + stop_apic_nmi_watchdog(NULL); + clear_local_APIC(); + + cpu_disable_common(); return 0; } diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 17305e72f34..8bdaa4a25f0 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -125,6 +125,7 @@ static inline void arch_send_call_function_ipi(cpumask_t mask) smp_ops.send_call_func_ipi(mask); } +void cpu_disable_common(void); void native_smp_prepare_boot_cpu(void); void native_smp_prepare_cpus(unsigned int max_cpus); void native_smp_cpus_done(unsigned int max_cpus); -- cgit v1.2.3-70-g09d2 From 7414aa41a63348c3bc72d8c37b716024c29b6d50 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Wed, 27 Aug 2008 17:56:44 -0700 Subject: x86: generate names for /proc/cpuinfo from We have had a number of cases where (and its predecessors) have diverged substantially from the names list in /proc/cpuinfo. This patch generates the latter from the former. It retains the option for explicitly overriding the strings, but by making that require a separate action it should at least be less likely to happen. It would be good to do a future pass and rename strings that are gratuituously different in the kernel (/proc/cpuinfo is a userspace interface and must remain constant.) Signed-off-by: H. Peter Anvin --- arch/x86/boot/mkcpustr.c | 2 +- arch/x86/kernel/cpu/Makefile | 11 ++++- arch/x86/kernel/cpu/feature_names.c | 84 ------------------------------------ arch/x86/kernel/cpu/mkcapflags.pl | 32 ++++++++++++++ arch/x86/kernel/cpu/powerflags.c | 20 +++++++++ include/asm-x86/cpufeature.h | 86 +++++++++++++++++++++++-------------- 6 files changed, 117 insertions(+), 118 deletions(-) delete mode 100644 arch/x86/kernel/cpu/feature_names.c create mode 100644 arch/x86/kernel/cpu/mkcapflags.pl create mode 100644 arch/x86/kernel/cpu/powerflags.c (limited to 'include/asm-x86') diff --git a/arch/x86/boot/mkcpustr.c b/arch/x86/boot/mkcpustr.c index bbe76953bae..4589caa3e9d 100644 --- a/arch/x86/boot/mkcpustr.c +++ b/arch/x86/boot/mkcpustr.c @@ -15,7 +15,7 @@ #include -#include "../kernel/cpu/feature_names.c" +#include "../kernel/cpu/capflags.c" #if NCAPFLAGS > 8 # error "Need to adjust the boot code handling of CPUID strings" diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index a0fc6c14438..3ede19a4e0b 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -3,7 +3,7 @@ # obj-y := intel_cacheinfo.o addon_cpuid_features.o -obj-y += proc.o feature_names.o +obj-y += proc.o capflags.o powerflags.o obj-$(CONFIG_X86_32) += common.o bugs.o cmpxchg.o obj-$(CONFIG_X86_64) += common_64.o bugs_64.o @@ -23,3 +23,12 @@ obj-$(CONFIG_MTRR) += mtrr/ obj-$(CONFIG_CPU_FREQ) += cpufreq/ obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o + +quiet_cmd_mkcapflags = MKCAP $@ + cmd_mkcapflags = $(PERL) $(srctree)/$(src)/mkcapflags.pl $< $@ + +cpufeature = $(src)/../../../../include/asm-x86/cpufeature.h + +targets += capflags.c +$(obj)/capflags.c: $(cpufeature) $(src)/mkcapflags.pl FORCE + $(call if_changed,mkcapflags) diff --git a/arch/x86/kernel/cpu/feature_names.c b/arch/x86/kernel/cpu/feature_names.c deleted file mode 100644 index c9017799497..00000000000 --- a/arch/x86/kernel/cpu/feature_names.c +++ /dev/null @@ -1,84 +0,0 @@ -/* - * Strings for the various x86 capability flags. - * - * This file must not contain any executable code. - */ - -#include - -/* - * These flag bits must match the definitions in . - * NULL means this bit is undefined or reserved; either way it doesn't - * have meaning as far as Linux is concerned. Note that it's important - * to realize there is a difference between this table and CPUID -- if - * applications want to get the raw CPUID data, they should access - * /dev/cpu//cpuid instead. - */ -const char * const x86_cap_flags[NCAPINTS*32] = { - /* Intel-defined */ - "fpu", "vme", "de", "pse", "tsc", "msr", "pae", "mce", - "cx8", "apic", NULL, "sep", "mtrr", "pge", "mca", "cmov", - "pat", "pse36", "pn", "clflush", NULL, "dts", "acpi", "mmx", - "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", "pbe", - - /* AMD-defined */ - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL, - NULL, NULL, NULL, "mp", "nx", NULL, "mmxext", NULL, - NULL, "fxsr_opt", "pdpe1gb", "rdtscp", NULL, "lm", - "3dnowext", "3dnow", - - /* Transmeta-defined */ - "recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - - /* Other (Linux-defined) */ - "cxmmx", "k6_mtrr", "cyrix_arr", "centaur_mcr", - NULL, NULL, NULL, NULL, - "constant_tsc", "up", NULL, "arch_perfmon", - "pebs", "bts", NULL, NULL, - "rep_good", NULL, NULL, NULL, - "nopl", NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - - /* Intel-defined (#2) */ - "pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est", - "tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL, - NULL, NULL, "dca", "sse4_1", "sse4_2", NULL, NULL, "popcnt", - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - - /* VIA/Cyrix/Centaur-defined */ - NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en", - "ace2", "ace2_en", "phe", "phe_en", "pmm", "pmm_en", NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - - /* AMD-defined (#2) */ - "lahf_lm", "cmp_legacy", "svm", "extapic", - "cr8_legacy", "abm", "sse4a", "misalignsse", - "3dnowprefetch", "osvw", "ibs", "sse5", - "skinit", "wdt", NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - - /* Auxiliary (Linux-defined) */ - "ida", NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -}; - -const char *const x86_power_flags[32] = { - "ts", /* temperature sensor */ - "fid", /* frequency id control */ - "vid", /* voltage id control */ - "ttp", /* thermal trip */ - "tm", - "stc", - "100mhzsteps", - "hwpstate", - "", /* tsc invariant mapped to constant_tsc */ - /* nothing */ -}; diff --git a/arch/x86/kernel/cpu/mkcapflags.pl b/arch/x86/kernel/cpu/mkcapflags.pl new file mode 100644 index 00000000000..dfea390e160 --- /dev/null +++ b/arch/x86/kernel/cpu/mkcapflags.pl @@ -0,0 +1,32 @@ +#!/usr/bin/perl +# +# Generate the x86_cap_flags[] array from include/asm-x86/cpufeature.h +# + +($in, $out) = @ARGV; + +open(IN, "< $in\0") or die "$0: cannot open: $in: $!\n"; +open(OUT, "> $out\0") or die "$0: cannot create: $out: $!\n"; + +print OUT "#include \n\n"; +print OUT "const char * const x86_cap_flags[NCAPINTS*32] = {\n"; + +while (defined($line = )) { + if ($line =~ /^\s*\#\s*define\s+(X86_FEATURE_(\S+))\s+(.*)$/) { + $macro = $1; + $feature = $2; + $tail = $3; + if ($tail =~ /\/\*\s*\"([^"]*)\".*\*\//) { + $feature = $1; + } + + if ($feature ne '') { + printf OUT "\t%-32s = \"%s\",\n", + "[$macro]", "\L$feature"; + } + } +} +print OUT "};\n"; + +close(IN); +close(OUT); diff --git a/arch/x86/kernel/cpu/powerflags.c b/arch/x86/kernel/cpu/powerflags.c new file mode 100644 index 00000000000..5abbea297e0 --- /dev/null +++ b/arch/x86/kernel/cpu/powerflags.c @@ -0,0 +1,20 @@ +/* + * Strings for the various x86 power flags + * + * This file must not contain any executable code. + */ + +#include + +const char *const x86_power_flags[32] = { + "ts", /* temperature sensor */ + "fid", /* frequency id control */ + "vid", /* voltage id control */ + "ttp", /* thermal trip */ + "tm", + "stc", + "100mhzsteps", + "hwpstate", + "", /* tsc invariant mapped to constant_tsc */ + /* nothing */ +}; diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 9489283a4bc..611da2898b2 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -8,6 +8,12 @@ #define NCAPINTS 8 /* N 32-bit words worth of info */ +/* + * Note: If the comment begins with a quoted string, that string is used + * in /proc/cpuinfo instead of the macro name. If the string is "", + * this feature bit is not displayed in /proc/cpuinfo at all. + */ + /* Intel-defined CPU features, CPUID level 0x00000001 (edx), word 0 */ #define X86_FEATURE_FPU (0*32+ 0) /* Onboard FPU */ #define X86_FEATURE_VME (0*32+ 1) /* Virtual Mode Extensions */ @@ -27,18 +33,18 @@ #define X86_FEATURE_PAT (0*32+16) /* Page Attribute Table */ #define X86_FEATURE_PSE36 (0*32+17) /* 36-bit PSEs */ #define X86_FEATURE_PN (0*32+18) /* Processor serial number */ -#define X86_FEATURE_CLFLSH (0*32+19) /* Supports the CLFLUSH instruction */ -#define X86_FEATURE_DS (0*32+21) /* Debug Store */ +#define X86_FEATURE_CLFLSH (0*32+19) /* "clflush" Supports the CLFLUSH instruction */ +#define X86_FEATURE_DS (0*32+21) /* "dts" Debug Store */ #define X86_FEATURE_ACPI (0*32+22) /* ACPI via MSR */ #define X86_FEATURE_MMX (0*32+23) /* Multimedia Extensions */ -#define X86_FEATURE_FXSR (0*32+24) /* FXSAVE and FXRSTOR instructions (fast save and restore */ - /* of FPU context), and CR4.OSFXSR available */ -#define X86_FEATURE_XMM (0*32+25) /* Streaming SIMD Extensions */ -#define X86_FEATURE_XMM2 (0*32+26) /* Streaming SIMD Extensions-2 */ -#define X86_FEATURE_SELFSNOOP (0*32+27) /* CPU self snoop */ +#define X86_FEATURE_FXSR (0*32+24) /* FXSAVE/FXRSTOR, CR4.OSFXSR */ +#define X86_FEATURE_XMM (0*32+25) /* "sse" */ +#define X86_FEATURE_XMM2 (0*32+26) /* "sse2" */ +#define X86_FEATURE_SELFSNOOP (0*32+27) /* "ss" CPU self snoop */ #define X86_FEATURE_HT (0*32+28) /* Hyper-Threading */ -#define X86_FEATURE_ACC (0*32+29) /* Automatic clock control */ +#define X86_FEATURE_ACC (0*32+29) /* "tm" Automatic clock control */ #define X86_FEATURE_IA64 (0*32+30) /* IA-64 processor */ +#define X86_FEATURE_PBE (0*32+31) /* Pending Break Enable */ /* AMD-defined CPU features, CPUID level 0x80000001, word 1 */ /* Don't duplicate feature flags which are redundant with Intel! */ @@ -46,7 +52,8 @@ #define X86_FEATURE_MP (1*32+19) /* MP Capable. */ #define X86_FEATURE_NX (1*32+20) /* Execute Disable */ #define X86_FEATURE_MMXEXT (1*32+22) /* AMD MMX extensions */ -#define X86_FEATURE_GBPAGES (1*32+26) /* GB pages */ +#define X86_FEATURE_FXSR_OPT (1*32+25) /* FXSAVE/FXRSTOR optimizations */ +#define X86_FEATURE_GBPAGES (1*32+26) /* "pdpe1gb" GB pages */ #define X86_FEATURE_RDTSCP (1*32+27) /* RDTSCP */ #define X86_FEATURE_LM (1*32+29) /* Long Mode (x86-64) */ #define X86_FEATURE_3DNOWEXT (1*32+30) /* AMD 3DNow! extensions */ @@ -64,52 +71,67 @@ #define X86_FEATURE_CYRIX_ARR (3*32+ 2) /* Cyrix ARRs (= MTRRs) */ #define X86_FEATURE_CENTAUR_MCR (3*32+ 3) /* Centaur MCRs (= MTRRs) */ /* cpu types for specific tunings: */ -#define X86_FEATURE_K8 (3*32+ 4) /* Opteron, Athlon64 */ -#define X86_FEATURE_K7 (3*32+ 5) /* Athlon */ -#define X86_FEATURE_P3 (3*32+ 6) /* P3 */ -#define X86_FEATURE_P4 (3*32+ 7) /* P4 */ +#define X86_FEATURE_K8 (3*32+ 4) /* "" Opteron, Athlon64 */ +#define X86_FEATURE_K7 (3*32+ 5) /* "" Athlon */ +#define X86_FEATURE_P3 (3*32+ 6) /* "" P3 */ +#define X86_FEATURE_P4 (3*32+ 7) /* "" P4 */ #define X86_FEATURE_CONSTANT_TSC (3*32+ 8) /* TSC ticks at a constant rate */ #define X86_FEATURE_UP (3*32+ 9) /* smp kernel running on up */ -#define X86_FEATURE_FXSAVE_LEAK (3*32+10) /* FXSAVE leaks FOP/FIP/FOP */ +#define X86_FEATURE_FXSAVE_LEAK (3*32+10) /* "" FXSAVE leaks FOP/FIP/FOP */ #define X86_FEATURE_ARCH_PERFMON (3*32+11) /* Intel Architectural PerfMon */ #define X86_FEATURE_PEBS (3*32+12) /* Precise-Event Based Sampling */ #define X86_FEATURE_BTS (3*32+13) /* Branch Trace Store */ -#define X86_FEATURE_SYSCALL32 (3*32+14) /* syscall in ia32 userspace */ -#define X86_FEATURE_SYSENTER32 (3*32+15) /* sysenter in ia32 userspace */ +#define X86_FEATURE_SYSCALL32 (3*32+14) /* "" syscall in ia32 userspace */ +#define X86_FEATURE_SYSENTER32 (3*32+15) /* "" sysenter in ia32 userspace */ #define X86_FEATURE_REP_GOOD (3*32+16) /* rep microcode works well on this CPU */ -#define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* Mfence synchronizes RDTSC */ -#define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* Lfence synchronizes RDTSC */ -#define X86_FEATURE_11AP (3*32+19) /* Bad local APIC aka 11AP */ +#define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* "" Mfence synchronizes RDTSC */ +#define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* "" Lfence synchronizes RDTSC */ +#define X86_FEATURE_11AP (3*32+19) /* "" Bad local APIC aka 11AP */ #define X86_FEATURE_NOPL (3*32+20) /* The NOPL (0F 1F) instructions */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ -#define X86_FEATURE_XMM3 (4*32+ 0) /* Streaming SIMD Extensions-3 */ -#define X86_FEATURE_MWAIT (4*32+ 3) /* Monitor/Mwait support */ -#define X86_FEATURE_DSCPL (4*32+ 4) /* CPL Qualified Debug Store */ +#define X86_FEATURE_XMM3 (4*32+ 0) /* "pni" SSE-3 */ +#define X86_FEATURE_MWAIT (4*32+ 3) /* "monitor" Monitor/Mwait support */ +#define X86_FEATURE_DSCPL (4*32+ 4) /* "ds_cpl" CPL Qual. Debug Store */ +#define X86_FEATURE_VMX (4*32+ 5) /* Hardware virtualization */ +#define X86_FEATURE_SMX (4*32+ 6) /* "Safer" mode */ #define X86_FEATURE_EST (4*32+ 7) /* Enhanced SpeedStep */ #define X86_FEATURE_TM2 (4*32+ 8) /* Thermal Monitor 2 */ +#define X86_FEATURE_SSSE3 (4*32+ 9) /* Supplemental SSE-3 */ #define X86_FEATURE_CID (4*32+10) /* Context ID */ #define X86_FEATURE_CX16 (4*32+13) /* CMPXCHG16B */ #define X86_FEATURE_XTPR (4*32+14) /* Send Task Priority Messages */ #define X86_FEATURE_DCA (4*32+18) /* Direct Cache Access */ -#define X86_FEATURE_XMM4_2 (4*32+20) /* Streaming SIMD Extensions-4.2 */ +#define X86_FEATURE_XMM4_1 (4*32+19) /* "sse4_1" SSE-4.1 */ +#define X86_FEATURE_XMM4_2 (4*32+20) /* "sse4_2" SSE-4.2 */ /* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */ -#define X86_FEATURE_XSTORE (5*32+ 2) /* on-CPU RNG present (xstore insn) */ -#define X86_FEATURE_XSTORE_EN (5*32+ 3) /* on-CPU RNG enabled */ -#define X86_FEATURE_XCRYPT (5*32+ 6) /* on-CPU crypto (xcrypt insn) */ -#define X86_FEATURE_XCRYPT_EN (5*32+ 7) /* on-CPU crypto enabled */ +#define X86_FEATURE_XSTORE (5*32+ 2) /* "rng" RNG present (xstore) */ +#define X86_FEATURE_XSTORE_EN (5*32+ 3) /* "rng_en" RNG enabled */ +#define X86_FEATURE_XCRYPT (5*32+ 6) /* "ace" on-CPU crypto (xcrypt) */ +#define X86_FEATURE_XCRYPT_EN (5*32+ 7) /* "ace_en" on-CPU crypto enabled */ #define X86_FEATURE_ACE2 (5*32+ 8) /* Advanced Cryptography Engine v2 */ #define X86_FEATURE_ACE2_EN (5*32+ 9) /* ACE v2 enabled */ -#define X86_FEATURE_PHE (5*32+ 10) /* PadLock Hash Engine */ -#define X86_FEATURE_PHE_EN (5*32+ 11) /* PHE enabled */ -#define X86_FEATURE_PMM (5*32+ 12) /* PadLock Montgomery Multiplier */ -#define X86_FEATURE_PMM_EN (5*32+ 13) /* PMM enabled */ +#define X86_FEATURE_PHE (5*32+10) /* PadLock Hash Engine */ +#define X86_FEATURE_PHE_EN (5*32+11) /* PHE enabled */ +#define X86_FEATURE_PMM (5*32+12) /* PadLock Montgomery Multiplier */ +#define X86_FEATURE_PMM_EN (5*32+13) /* PMM enabled */ /* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */ #define X86_FEATURE_LAHF_LM (6*32+ 0) /* LAHF/SAHF in long mode */ #define X86_FEATURE_CMP_LEGACY (6*32+ 1) /* If yes HyperThreading not valid */ -#define X86_FEATURE_IBS (6*32+ 10) /* Instruction Based Sampling */ +#define X86_FEATURE_SVM (6*32+ 2) /* Secure virtual machine */ +#define X86_FEATURE_EXTAPIC (6*32+ 3) /* Extended APIC space */ +#define X86_FEATURE_CR8_LEGACY (6*32+ 4) /* CR8 in 32-bit mode */ +#define X86_FEATURE_ABM (6*32+ 5) /* Advanced bit manipulation */ +#define X86_FEATURE_SSE4A (6*32+ 6) /* SSE-4A */ +#define X86_FEATURE_MISALIGNSSE (6*32+ 7) /* Misaligned SSE mode */ +#define X86_FEATURE_3DNOWPREFETCH (6*32+ 8) /* 3DNow prefetch instructions */ +#define X86_FEATURE_OSVW (6*32+ 9) /* OS Visible Workaround */ +#define X86_FEATURE_IBS (6*32+10) /* Instruction Based Sampling */ +#define X86_FEATURE_SSE5 (6*32+11) /* SSE-5 */ +#define X86_FEATURE_SKINIT (6*32+12) /* SKINIT/STGI instructions */ +#define X86_FEATURE_WDT (6*32+13) /* Watchdog timer */ /* * Auxiliary flags: Linux defined - For features scattered in various -- cgit v1.2.3-70-g09d2 From f1240c002679a77990fd7c198991ed15a437d691 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Wed, 27 Aug 2008 18:53:07 -0700 Subject: x86: cpufeature: add Intel features from CPUID and AVX specs Add all Intel CPUID features currently documented in the CPUID spec (AP-485, 241618-032, Dec 2007) and the AVX Programming Reference (319433-003, Aug 2008). Signed-off-by: H. Peter Anvin --- include/asm-x86/cpufeature.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 611da2898b2..7bd98b724fd 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -91,6 +91,8 @@ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* "pni" SSE-3 */ +#define X86_FEATURE_PCLMULQDQ (4*32+ 1) /* PCLMULQDQ instruction */ +#define X86_FEATURE_DTES64 (4*32+ 2) /* 64-bit Debug Store */ #define X86_FEATURE_MWAIT (4*32+ 3) /* "monitor" Monitor/Mwait support */ #define X86_FEATURE_DSCPL (4*32+ 4) /* "ds_cpl" CPL Qual. Debug Store */ #define X86_FEATURE_VMX (4*32+ 5) /* Hardware virtualization */ @@ -99,11 +101,18 @@ #define X86_FEATURE_TM2 (4*32+ 8) /* Thermal Monitor 2 */ #define X86_FEATURE_SSSE3 (4*32+ 9) /* Supplemental SSE-3 */ #define X86_FEATURE_CID (4*32+10) /* Context ID */ +#define X86_FEATURE_FMA (4*32+12) /* Fused multiply-add */ #define X86_FEATURE_CX16 (4*32+13) /* CMPXCHG16B */ #define X86_FEATURE_XTPR (4*32+14) /* Send Task Priority Messages */ +#define X86_FEATURE_PDCM (4*32+15) /* Performance Capabilities */ #define X86_FEATURE_DCA (4*32+18) /* Direct Cache Access */ #define X86_FEATURE_XMM4_1 (4*32+19) /* "sse4_1" SSE-4.1 */ #define X86_FEATURE_XMM4_2 (4*32+20) /* "sse4_2" SSE-4.2 */ +#define X86_FEATURE_X2APIC (4*32+21) /* x2APIC */ +#define X86_FEATURE_AES (4*32+25) /* AES instructions */ +#define X86_FEATURE_XSAVE (4*32+26) /* XSAVE/XRSTOR/XSETBV/XGETBV */ +#define X86_FEATURE_OSXSAVE (4*32+27) /* "" XSAVE enabled in the OS */ +#define X86_FEATURE_AVX (4*32+28) /* Advanced Vector Extensions */ /* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */ #define X86_FEATURE_XSTORE (5*32+ 2) /* "rng" RNG present (xstore) */ @@ -213,7 +222,10 @@ extern const char * const x86_power_flags[32]; #define cpu_has_gbpages boot_cpu_has(X86_FEATURE_GBPAGES) #define cpu_has_arch_perfmon boot_cpu_has(X86_FEATURE_ARCH_PERFMON) #define cpu_has_pat boot_cpu_has(X86_FEATURE_PAT) +#define cpu_has_xmm4_1 boot_cpu_has(X86_FEATURE_XMM4_1) #define cpu_has_xmm4_2 boot_cpu_has(X86_FEATURE_XMM4_2) +#define cpu_has_x2apic boot_cpu_has(X86_FEATURE_X2APIC) +#define cpu_has_xsave boot_cpu_has(X86_FEATURE_XSAVE) #if defined(CONFIG_X86_INVLPG) || defined(CONFIG_X86_64) # define cpu_has_invlpg 1 -- cgit v1.2.3-70-g09d2 From 2798c63e65cb0f05cc12a060b9b0d56ac9523c4d Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Wed, 27 Aug 2008 21:20:07 -0700 Subject: x86: : clean up overlong lines, whitespace Clean up overlong lines and stealth whitespace in . Signed-off-by: H. Peter Anvin --- include/asm-x86/cpufeature.h | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 7bd98b724fd..7710686c11f 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -20,7 +20,7 @@ #define X86_FEATURE_DE (0*32+ 2) /* Debugging Extensions */ #define X86_FEATURE_PSE (0*32+ 3) /* Page Size Extensions */ #define X86_FEATURE_TSC (0*32+ 4) /* Time Stamp Counter */ -#define X86_FEATURE_MSR (0*32+ 5) /* Model-Specific Registers, RDMSR, WRMSR */ +#define X86_FEATURE_MSR (0*32+ 5) /* Model-Specific Registers */ #define X86_FEATURE_PAE (0*32+ 6) /* Physical Address Extensions */ #define X86_FEATURE_MCE (0*32+ 7) /* Machine Check Architecture */ #define X86_FEATURE_CX8 (0*32+ 8) /* CMPXCHG8 instruction */ @@ -29,11 +29,12 @@ #define X86_FEATURE_MTRR (0*32+12) /* Memory Type Range Registers */ #define X86_FEATURE_PGE (0*32+13) /* Page Global Enable */ #define X86_FEATURE_MCA (0*32+14) /* Machine Check Architecture */ -#define X86_FEATURE_CMOV (0*32+15) /* CMOV instruction (FCMOVCC and FCOMI too if FPU present) */ +#define X86_FEATURE_CMOV (0*32+15) /* CMOV instructions */ + /* (plus FCMOVcc, FCOMI with FPU) */ #define X86_FEATURE_PAT (0*32+16) /* Page Attribute Table */ #define X86_FEATURE_PSE36 (0*32+17) /* 36-bit PSEs */ #define X86_FEATURE_PN (0*32+18) /* Processor serial number */ -#define X86_FEATURE_CLFLSH (0*32+19) /* "clflush" Supports the CLFLUSH instruction */ +#define X86_FEATURE_CLFLSH (0*32+19) /* "clflush" CLFLUSH instruction */ #define X86_FEATURE_DS (0*32+21) /* "dts" Debug Store */ #define X86_FEATURE_ACPI (0*32+22) /* ACPI via MSR */ #define X86_FEATURE_MMX (0*32+23) /* Multimedia Extensions */ @@ -83,7 +84,7 @@ #define X86_FEATURE_BTS (3*32+13) /* Branch Trace Store */ #define X86_FEATURE_SYSCALL32 (3*32+14) /* "" syscall in ia32 userspace */ #define X86_FEATURE_SYSENTER32 (3*32+15) /* "" sysenter in ia32 userspace */ -#define X86_FEATURE_REP_GOOD (3*32+16) /* rep microcode works well on this CPU */ +#define X86_FEATURE_REP_GOOD (3*32+16) /* rep microcode works well */ #define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* "" Mfence synchronizes RDTSC */ #define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* "" Lfence synchronizes RDTSC */ #define X86_FEATURE_11AP (3*32+19) /* "" Bad local APIC aka 11AP */ @@ -181,7 +182,7 @@ extern const char * const x86_power_flags[32]; } while (0) #define setup_force_cpu_cap(bit) do { \ set_cpu_cap(&boot_cpu_data, bit); \ - clear_bit(bit, (unsigned long *)cleared_cpu_caps); \ + clear_bit(bit, (unsigned long *)cleared_cpu_caps); \ } while (0) #define cpu_has_fpu boot_cpu_has(X86_FEATURE_FPU) -- cgit v1.2.3-70-g09d2 From af2e1f276ff08f17192411ea3b71c13a758dfe12 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Wed, 27 Aug 2008 22:05:45 -0700 Subject: x86: cpufeature: fix SMX flag Impact: "smx" flags showed as "safer" in /proc/cpuinfo The SMX feature flag is the so-called "safer mode"... I had put that in quotes, but that flagged it as a cpuinfo flag name. Remove the quotes to correct the flag name in /proc/cpuinfo. Signed-off-by: H. Peter Anvin --- include/asm-x86/cpufeature.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 7710686c11f..24d99d65741 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -97,7 +97,7 @@ #define X86_FEATURE_MWAIT (4*32+ 3) /* "monitor" Monitor/Mwait support */ #define X86_FEATURE_DSCPL (4*32+ 4) /* "ds_cpl" CPL Qual. Debug Store */ #define X86_FEATURE_VMX (4*32+ 5) /* Hardware virtualization */ -#define X86_FEATURE_SMX (4*32+ 6) /* "Safer" mode */ +#define X86_FEATURE_SMX (4*32+ 6) /* Safer mode */ #define X86_FEATURE_EST (4*32+ 7) /* Enhanced SpeedStep */ #define X86_FEATURE_TM2 (4*32+ 8) /* Thermal Monitor 2 */ #define X86_FEATURE_SSSE3 (4*32+ 9) /* Supplemental SSE-3 */ -- cgit v1.2.3-70-g09d2 From 58f7c98850a226d3fb05b1095af9f7c4ea3507ba Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 28 Aug 2008 13:52:25 -0700 Subject: x86: split e820 reserved entries record to late v2 so could let BAR res register at first, or even pnp. v2: insert e820 reserve resources before pnp_system_init Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/e820.c | 20 +++++++++++-- arch/x86/pci/i386.c | 79 ++++++++++++++++++++++++++++++++++++++++++++++++++ include/asm-x86/e820.h | 1 + 3 files changed, 98 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 291e6cd9f9c..523d6c5605d 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -1271,13 +1271,15 @@ static inline const char *e820_type_to_string(int e820_type) /* * Mark e820 reserved areas as busy for the resource manager. */ +struct resource __initdata *e820_res; void __init e820_reserve_resources(void) { int i; - struct resource *res; u64 end; + struct resource *res; res = alloc_bootmem_low(sizeof(struct resource) * e820.nr_map); + e820_res = res; for (i = 0; i < e820.nr_map; i++) { end = e820.map[i].addr + e820.map[i].size - 1; #ifndef CONFIG_RESOURCES_64BIT @@ -1291,7 +1293,8 @@ void __init e820_reserve_resources(void) res->end = end; res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; - insert_resource(&iomem_resource, res); + if (e820.map[i].type != E820_RESERVED || res->start < (1ULL<<20)) + insert_resource(&iomem_resource, res); res++; } @@ -1303,6 +1306,19 @@ void __init e820_reserve_resources(void) } } +void __init e820_reserve_resources_late(void) +{ + int i; + struct resource *res; + + res = e820_res; + for (i = 0; i < e820.nr_map; i++) { + if (e820.map[i].type == E820_RESERVED && res->start >= (1ULL<<20)) + insert_resource(&iomem_resource, res); + res++; + } +} + char *__init default_machine_specific_memory_setup(void) { char *who = "BIOS-e820"; diff --git a/arch/x86/pci/i386.c b/arch/x86/pci/i386.c index 8791fc55e71..40811efaa25 100644 --- a/arch/x86/pci/i386.c +++ b/arch/x86/pci/i386.c @@ -31,8 +31,12 @@ #include #include #include +#include #include +#include +#include +#include #include "pci.h" @@ -77,6 +81,77 @@ pcibios_align_resource(void *data, struct resource *res, } EXPORT_SYMBOL(pcibios_align_resource); +static int check_res_with_valid(struct pci_dev *dev, struct resource *res) +{ + unsigned long base; + unsigned long size; + int i; + + base = res->start; + size = (res->start == 0 && res->end == res->start) ? 0 : + (res->end - res->start + 1); + + if (!base || !size) + return 0; + +#ifdef CONFIG_HPET_TIMER + /* for hpet */ + if (base == hpet_address && (res->flags & IORESOURCE_MEM)) { + dev_info(&dev->dev, "BAR has HPET at %08lx-%08lx\n", + base, base + size - 1); + return 1; + } +#endif + +#ifdef CONFIG_X86_IO_APIC + for (i = 0; i < nr_ioapics; i++) { + unsigned long ioapic_phys = mp_ioapics[i].mp_apicaddr; + + if (base == ioapic_phys && (res->flags & IORESOURCE_MEM)) { + dev_info(&dev->dev, "BAR has ioapic at %08lx-%08lx\n", + base, base + size - 1); + return 1; + } + } +#endif + +#ifdef CONFIG_PCI_MMCONFIG + for (i = 0; i < pci_mmcfg_config_num; i++) { + unsigned long addr; + + addr = pci_mmcfg_config[i].address; + if (base == addr && (res->flags & IORESOURCE_MEM)) { + dev_info(&dev->dev, "BAR has MMCONFIG at %08lx-%08lx\n", + base, base + size - 1); + return 1; + } + } +#endif + + return 0; +} + +static int check_platform(struct pci_dev *dev, struct resource *res) +{ + struct resource *root = NULL; + + /* + * forcibly insert it into the + * resource tree + */ + if (res->flags & IORESOURCE_MEM) + root = &iomem_resource; + else if (res->flags & IORESOURCE_IO) + root = &ioport_resource; + + if (root && check_res_with_valid(dev, res)) { + insert_resource(root, res); + + return 1; + } + + return 0; +} /* * Handle resources of PCI devices. If the world were perfect, we could * just allocate all the resource regions and do nothing more. It isn't. @@ -128,6 +203,8 @@ static void __init pcibios_allocate_bus_resources(struct list_head *bus_list) pr = pci_find_parent_resource(dev, r); if (!r->start || !pr || request_resource(pr, r) < 0) { + if (check_platform(dev, r)) + continue; dev_err(&dev->dev, "BAR %d: can't allocate resource\n", idx); /* * Something is wrong with the region. @@ -169,6 +246,8 @@ static void __init pcibios_allocate_resources(int pass) r->flags, disabled, pass); pr = pci_find_parent_resource(dev, r); if (!pr || request_resource(pr, r) < 0) { + if (check_platform(dev, r)) + continue; dev_err(&dev->dev, "BAR %d: can't allocate resource\n", idx); /* We'll assign a new address later */ r->end -= r->start; diff --git a/include/asm-x86/e820.h b/include/asm-x86/e820.h index ca433c36af7..5abbdec06bd 100644 --- a/include/asm-x86/e820.h +++ b/include/asm-x86/e820.h @@ -122,6 +122,7 @@ extern void e820_register_active_regions(int nid, unsigned long start_pfn, extern u64 e820_hole_size(u64 start, u64 end); extern void finish_e820_parsing(void); extern void e820_reserve_resources(void); +extern void e820_reserve_resources_late(void); extern void setup_memory_map(void); extern char *default_machine_specific_memory_setup(void); extern char *machine_specific_memory_setup(void); -- cgit v1.2.3-70-g09d2 From 3da99c97763703b23cbf2bd6e96252256d4e4617 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 4 Sep 2008 21:09:44 +0200 Subject: x86: make (early)_identify_cpu more the same between 32bit and 64 bit 1. add extended_cpuid_level for 32bit 2. add generic_identify for 64bit 3. add early_identify_cpu for 32bit 4. early_identify_cpu not be called by identify_cpu 5. remove early in get_cpu_vendor for 32bit 6. add get_cpu_cap 7. add cpu_detect for 64bit Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/common.c | 138 +++++++++++++++------------------------ arch/x86/kernel/cpu/common_64.c | 139 +++++++++++++++++++++++++--------------- include/asm-x86/processor.h | 2 +- 3 files changed, 142 insertions(+), 137 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 8aab8517642..96e1b8698d3 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -96,7 +96,7 @@ int __cpuinit get_model_name(struct cpuinfo_x86 *c) unsigned int *v; char *p, *q; - if (cpuid_eax(0x80000000) < 0x80000004) + if (c->extended_cpuid_level < 0x80000004) return 0; v = (unsigned int *) c->x86_model_id; @@ -125,7 +125,7 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) { unsigned int n, dummy, ecx, edx, l2size; - n = cpuid_eax(0x80000000); + n = c->extended_cpuid_level; if (n >= 0x80000005) { cpuid(0x80000005, &dummy, &dummy, &ecx, &edx); @@ -186,7 +186,7 @@ static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c) } -static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early) +static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) { char *v = c->x86_vendor_id; int i; @@ -198,8 +198,7 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early) (cpu_devs[i]->c_ident[1] && !strcmp(v, cpu_devs[i]->c_ident[1]))) { c->x86_vendor = i; - if (!early) - this_cpu = cpu_devs[i]; + this_cpu = cpu_devs[i]; return; } } @@ -284,34 +283,30 @@ void __init cpu_detect(struct cpuinfo_x86 *c) } } } -static void __cpuinit early_get_cap(struct cpuinfo_x86 *c) + +static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) { u32 tfms, xlvl; - unsigned int ebx; + u32 ebx; - memset(&c->x86_capability, 0, sizeof c->x86_capability); - if (have_cpuid_p()) { - /* Intel-defined flags: level 0x00000001 */ - if (c->cpuid_level >= 0x00000001) { - u32 capability, excap; - cpuid(0x00000001, &tfms, &ebx, &excap, &capability); - c->x86_capability[0] = capability; - c->x86_capability[4] = excap; - } + /* Intel-defined flags: level 0x00000001 */ + if (c->cpuid_level >= 0x00000001) { + u32 capability, excap; + cpuid(0x00000001, &tfms, &ebx, &excap, &capability); + c->x86_capability[0] = capability; + c->x86_capability[4] = excap; + } - /* AMD-defined flags: level 0x80000001 */ - xlvl = cpuid_eax(0x80000000); - if ((xlvl & 0xffff0000) == 0x80000000) { - if (xlvl >= 0x80000001) { - c->x86_capability[1] = cpuid_edx(0x80000001); - c->x86_capability[6] = cpuid_ecx(0x80000001); - } + /* AMD-defined flags: level 0x80000001 */ + xlvl = cpuid_eax(0x80000000); + c->extended_cpuid_level = xlvl; + if ((xlvl & 0xffff0000) == 0x80000000) { + if (xlvl >= 0x80000001) { + c->x86_capability[1] = cpuid_edx(0x80000001); + c->x86_capability[6] = cpuid_ecx(0x80000001); } - } - } - /* * Do minimum CPU detection early. * Fields really needed: vendor, cpuid_level, family, model, mask, @@ -321,25 +316,29 @@ static void __cpuinit early_get_cap(struct cpuinfo_x86 *c) * WARNING: this function is only called on the BP. Don't add code here * that is supposed to run on all CPUs. */ -static void __init early_cpu_detect(void) +static void __init early_identify_cpu(struct cpuinfo_x86 *c) { - struct cpuinfo_x86 *c = &boot_cpu_data; - c->x86_cache_alignment = 32; c->x86_clflush_size = 32; if (!have_cpuid_p()) return; + c->extended_cpuid_level = 0; + + memset(&c->x86_capability, 0, sizeof c->x86_capability); + cpu_detect(c); - get_cpu_vendor(c, 1); + get_cpu_vendor(c); - early_get_cap(c); + get_cpu_cap(c); if (c->x86_vendor != X86_VENDOR_UNKNOWN && cpu_devs[c->x86_vendor]->c_early_init) cpu_devs[c->x86_vendor]->c_early_init(c); + + validate_pat_support(c); } /* @@ -373,60 +372,32 @@ static void __cpuinit detect_nopl(struct cpuinfo_x86 *c) static void __cpuinit generic_identify(struct cpuinfo_x86 *c) { - u32 tfms, xlvl; - unsigned int ebx; - - if (have_cpuid_p()) { - /* Get vendor name */ - cpuid(0x00000000, (unsigned int *)&c->cpuid_level, - (unsigned int *)&c->x86_vendor_id[0], - (unsigned int *)&c->x86_vendor_id[8], - (unsigned int *)&c->x86_vendor_id[4]); - - get_cpu_vendor(c, 0); - /* Initialize the standard set of capabilities */ - /* Note that the vendor-specific code below might override */ - /* Intel-defined flags: level 0x00000001 */ - if (c->cpuid_level >= 0x00000001) { - u32 capability, excap; - cpuid(0x00000001, &tfms, &ebx, &excap, &capability); - c->x86_capability[0] = capability; - c->x86_capability[4] = excap; - c->x86 = (tfms >> 8) & 15; - c->x86_model = (tfms >> 4) & 15; - if (c->x86 == 0xf) - c->x86 += (tfms >> 20) & 0xff; - if (c->x86 >= 0x6) - c->x86_model += ((tfms >> 16) & 0xF) << 4; - c->x86_mask = tfms & 15; - c->initial_apicid = (ebx >> 24) & 0xFF; + if (!have_cpuid_p()) + return; + + c->extended_cpuid_level = 0; + + cpu_detect(c); + + get_cpu_vendor(c); + + get_cpu_cap(c); + + if (c->cpuid_level >= 0x00000001) { + c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xFF; #ifdef CONFIG_X86_HT - c->apicid = phys_pkg_id(c->initial_apicid, 0); - c->phys_proc_id = c->initial_apicid; + c->apicid = phys_pkg_id(c->initial_apicid, 0); + c->phys_proc_id = c->initial_apicid; #else - c->apicid = c->initial_apicid; + c->apicid = c->initial_apicid; #endif - if (test_cpu_cap(c, X86_FEATURE_CLFLSH)) - c->x86_clflush_size = ((ebx >> 8) & 0xff) * 8; - } else { - /* Have CPUID level 0 only - unheard of */ - c->x86 = 4; - } + } - /* AMD-defined flags: level 0x80000001 */ - xlvl = cpuid_eax(0x80000000); - if ((xlvl & 0xffff0000) == 0x80000000) { - if (xlvl >= 0x80000001) { - c->x86_capability[1] = cpuid_edx(0x80000001); - c->x86_capability[6] = cpuid_ecx(0x80000001); - } - if (xlvl >= 0x80000004) - get_model_name(c); /* Default name */ - } + if (c->extended_cpuid_level >= 0x80000004) + get_model_name(c); /* Default name */ - init_scattered_cpuid_features(c); - detect_nopl(c); - } + init_scattered_cpuid_features(c); + detect_nopl(c); } static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c) @@ -651,13 +622,10 @@ void __init early_cpu_init(void) { struct cpu_vendor_dev *cvdev; - for (cvdev = __x86cpuvendor_start ; - cvdev < __x86cpuvendor_end ; - cvdev++) + for (cvdev = __x86cpuvendor_start; cvdev < __x86cpuvendor_end; cvdev++) cpu_devs[cvdev->vendor] = cvdev->cpu_dev; - early_cpu_detect(); - validate_pat_support(&boot_cpu_data); + early_identify_cpu(&boot_cpu_data); } /* Make sure %fs is initialized properly in idle threads */ diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c index 699ecbfb63e..28719fe0794 100644 --- a/arch/x86/kernel/cpu/common_64.c +++ b/arch/x86/kernel/cpu/common_64.c @@ -195,6 +195,7 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) printk(KERN_ERR "CPU: Your system may be unstable.\n"); } c->x86_vendor = X86_VENDOR_UNKNOWN; + this_cpu = &default_cpu; } static void __init early_cpu_support_print(void) @@ -249,56 +250,18 @@ static void __cpuinit detect_nopl(struct cpuinfo_x86 *c) } } -static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c); - -void __init early_cpu_init(void) +void __cpuinit cpu_detect(struct cpuinfo_x86 *c) { - struct cpu_vendor_dev *cvdev; - - for (cvdev = __x86cpuvendor_start ; - cvdev < __x86cpuvendor_end ; - cvdev++) - cpu_devs[cvdev->vendor] = cvdev->cpu_dev; - early_cpu_support_print(); - early_identify_cpu(&boot_cpu_data); -} - -/* Do some early cpuid on the boot CPU to get some parameter that are - needed before check_bugs. Everything advanced is in identify_cpu - below. */ -static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) -{ - u32 tfms, xlvl; - - c->loops_per_jiffy = loops_per_jiffy; - c->x86_cache_size = -1; - c->x86_vendor = X86_VENDOR_UNKNOWN; - c->x86_model = c->x86_mask = 0; /* So far unknown... */ - c->x86_vendor_id[0] = '\0'; /* Unset */ - c->x86_model_id[0] = '\0'; /* Unset */ - c->x86_clflush_size = 64; - c->x86_cache_alignment = c->x86_clflush_size; - c->x86_max_cores = 1; - c->x86_coreid_bits = 0; - c->extended_cpuid_level = 0; - memset(&c->x86_capability, 0, sizeof c->x86_capability); - /* Get vendor name */ cpuid(0x00000000, (unsigned int *)&c->cpuid_level, (unsigned int *)&c->x86_vendor_id[0], (unsigned int *)&c->x86_vendor_id[8], (unsigned int *)&c->x86_vendor_id[4]); - get_cpu_vendor(c); - - /* Initialize the standard set of capabilities */ - /* Note that the vendor-specific code below might override */ - /* Intel-defined flags: level 0x00000001 */ if (c->cpuid_level >= 0x00000001) { - __u32 misc; - cpuid(0x00000001, &tfms, &misc, &c->x86_capability[4], - &c->x86_capability[0]); + u32 junk, tfms, cap0, misc; + cpuid(0x00000001, &tfms, &misc, &junk, &cap0); c->x86 = (tfms >> 8) & 0xf; c->x86_model = (tfms >> 4) & 0xf; c->x86_mask = tfms & 0xf; @@ -306,17 +269,32 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) c->x86 += (tfms >> 20) & 0xff; if (c->x86 >= 0x6) c->x86_model += ((tfms >> 16) & 0xF) << 4; - if (test_cpu_cap(c, X86_FEATURE_CLFLSH)) + if (cap0 & (1<<19)) c->x86_clflush_size = ((misc >> 8) & 0xff) * 8; } else { /* Have CPUID level 0 only - unheard of */ c->x86 = 4; } +} + + +static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) +{ + u32 tfms, xlvl; + u32 ebx; + + /* Initialize the standard set of capabilities */ + /* Note that the vendor-specific code below might override */ + + /* Intel-defined flags: level 0x00000001 */ + if (c->cpuid_level >= 0x00000001) { + u32 capability, excap; + + cpuid(0x00000001, &tfms, &ebx, &excap, &capability); + c->x86_capability[0] = capability; + c->x86_capability[4] = excap; + } - c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xff; -#ifdef CONFIG_SMP - c->phys_proc_id = c->initial_apicid; -#endif /* AMD-defined flags: level 0x80000001 */ xlvl = cpuid_eax(0x80000000); c->extended_cpuid_level = xlvl; @@ -325,8 +303,6 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) c->x86_capability[1] = cpuid_edx(0x80000001); c->x86_capability[6] = cpuid_ecx(0x80000001); } - if (xlvl >= 0x80000004) - get_model_name(c); /* Default name */ } /* Transmeta-defined flags: level 0x80860001 */ @@ -346,8 +322,26 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) c->x86_virt_bits = (eax >> 8) & 0xff; c->x86_phys_bits = eax & 0xff; } +} - detect_nopl(c); +/* Do some early cpuid on the boot CPU to get some parameter that are + needed before check_bugs. Everything advanced is in identify_cpu + below. */ +static void __init early_identify_cpu(struct cpuinfo_x86 *c) +{ + + c->x86_clflush_size = 64; + c->x86_cache_alignment = c->x86_clflush_size; + + memset(&c->x86_capability, 0, sizeof c->x86_capability); + + c->extended_cpuid_level = 0; + + cpu_detect(c); + + get_cpu_vendor(c); + + get_cpu_cap(c); if (c->x86_vendor != X86_VENDOR_UNKNOWN && cpu_devs[c->x86_vendor]->c_early_init) @@ -356,6 +350,39 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) validate_pat_support(c); } +void __init early_cpu_init(void) +{ + struct cpu_vendor_dev *cvdev; + + for (cvdev = __x86cpuvendor_start; cvdev < __x86cpuvendor_end; cvdev++) + cpu_devs[cvdev->vendor] = cvdev->cpu_dev; + + early_cpu_support_print(); + early_identify_cpu(&boot_cpu_data); +} + +static void __cpuinit generic_identify(struct cpuinfo_x86 *c) +{ + c->extended_cpuid_level = 0; + + cpu_detect(c); + + get_cpu_vendor(c); + + get_cpu_cap(c); + + c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xff; +#ifdef CONFIG_SMP + c->phys_proc_id = c->initial_apicid; +#endif + + if (c->extended_cpuid_level >= 0x80000004) + get_model_name(c); /* Default name */ + + init_scattered_cpuid_features(c); + detect_nopl(c); +} + /* * This does the hard work of actually picking apart the CPU stuff... */ @@ -363,9 +390,19 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) { int i; - early_identify_cpu(c); + c->loops_per_jiffy = loops_per_jiffy; + c->x86_cache_size = -1; + c->x86_vendor = X86_VENDOR_UNKNOWN; + c->x86_model = c->x86_mask = 0; /* So far unknown... */ + c->x86_vendor_id[0] = '\0'; /* Unset */ + c->x86_model_id[0] = '\0'; /* Unset */ + c->x86_clflush_size = 64; + c->x86_cache_alignment = c->x86_clflush_size; + c->x86_max_cores = 1; + c->x86_coreid_bits = 0; + memset(&c->x86_capability, 0, sizeof c->x86_capability); - init_scattered_cpuid_features(c); + generic_identify(c); c->apicid = phys_pkg_id(0); diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index 4df3e2f6fb5..8208cc1debc 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -77,9 +77,9 @@ struct cpuinfo_x86 { __u8 x86_phys_bits; /* CPUID returned core id bits: */ __u8 x86_coreid_bits; +#endif /* Max extended CPUID function supported: */ __u32 extended_cpuid_level; -#endif /* Maximum supported CPUID level, -1=no CPUID: */ int cpuid_level; __u32 x86_capability[NCAPINTS]; -- cgit v1.2.3-70-g09d2 From 97e4db7c8719f67c52793eeca5ec4df4c3407f2a Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 4 Sep 2008 20:08:59 -0700 Subject: x86: make detect_ht depend on CONFIG_X86_HT 64-bit has X86_HT set too, so use that instead of SMP. This also removes a include/asm-x86/processor.h ifdef. Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/common.c | 4 ++-- arch/x86/kernel/cpu/common_64.c | 2 +- include/asm-x86/processor.h | 4 ---- 3 files changed, 3 insertions(+), 7 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 7d5a07f0fd2..1f80ce07daa 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -263,9 +263,9 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) l2size, ecx & 0xFF); } -#ifdef CONFIG_X86_HT void __cpuinit detect_ht(struct cpuinfo_x86 *c) { +#ifdef CONFIG_X86_HT u32 eax, ebx, ecx, edx; int index_msb, core_bits; @@ -311,8 +311,8 @@ out: printk(KERN_INFO "CPU: Processor Core ID: %d\n", c->cpu_core_id); } -} #endif +} static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) { diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c index bcb48ce05d2..8e630734e1f 100644 --- a/arch/x86/kernel/cpu/common_64.c +++ b/arch/x86/kernel/cpu/common_64.c @@ -141,7 +141,7 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) void __cpuinit detect_ht(struct cpuinfo_x86 *c) { -#ifdef CONFIG_SMP +#ifdef CONFIG_X86_HT u32 eax, ebx, ecx, edx; int index_msb, core_bits; diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index bbbbe1fc5ce..fc5e961c5b6 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -166,11 +166,7 @@ extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c); extern unsigned short num_cache_leaves; extern void detect_extended_topology(struct cpuinfo_x86 *c); -#if defined(CONFIG_X86_HT) || defined(CONFIG_X86_64) extern void detect_ht(struct cpuinfo_x86 *c); -#else -static inline void detect_ht(struct cpuinfo_x86 *c) {} -#endif static inline void native_cpuid(unsigned int *eax, unsigned int *ebx, unsigned int *ecx, unsigned int *edx) -- cgit v1.2.3-70-g09d2 From ef1f3413284b9270266cb04a944647e59735f0f1 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Fri, 5 Sep 2008 13:26:39 +0100 Subject: x86: ticket spin locks: fix asm constraints In addition to these changes I doubt the 'volatile' on all the ticket lock asm()-s are really necessary. Signed-off-by: Jan Beulich Signed-off-by: Ingo Molnar --- include/asm-x86/spinlock.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h index 93adae338ac..acd9bdda55c 100644 --- a/include/asm-x86/spinlock.h +++ b/include/asm-x86/spinlock.h @@ -101,7 +101,7 @@ static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) "1:" "sete %b1\n\t" "movzbl %b1,%0\n\t" - : "=&a" (tmp), "=Q" (new), "+m" (lock->slock) + : "=&a" (tmp), "=&Q" (new), "+m" (lock->slock) : : "memory", "cc"); @@ -146,7 +146,7 @@ static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) /* don't need lfence here, because loads are in-order */ "jmp 1b\n" "2:" - : "+Q" (inc), "+m" (lock->slock), "=r" (tmp) + : "+r" (inc), "+m" (lock->slock), "=&r" (tmp) : : "memory", "cc"); } @@ -166,7 +166,7 @@ static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) "1:" "sete %b1\n\t" "movzbl %b1,%0\n\t" - : "=&a" (tmp), "=r" (new), "+m" (lock->slock) + : "=&a" (tmp), "=&q" (new), "+m" (lock->slock) : : "memory", "cc"); -- cgit v1.2.3-70-g09d2 From 08f5fcbe6e0ea029c7e9b1b1c338700ab7809daf Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Fri, 5 Sep 2008 13:26:39 +0100 Subject: x86: ticket spin locks: factor out more common code Signed-off-by: Jan Beulich Signed-off-by: Ingo Molnar --- include/asm-x86/spinlock.h | 42 ++++++++++++++++-------------------------- 1 file changed, 16 insertions(+), 26 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h index acd9bdda55c..63d3b610a5e 100644 --- a/include/asm-x86/spinlock.h +++ b/include/asm-x86/spinlock.h @@ -54,19 +54,7 @@ * much between them in performance though, especially as locks are out of line. */ #if (NR_CPUS < 256) -static inline int __ticket_spin_is_locked(raw_spinlock_t *lock) -{ - int tmp = ACCESS_ONCE(lock->slock); - - return (((tmp >> 8) & 0xff) != (tmp & 0xff)); -} - -static inline int __ticket_spin_is_contended(raw_spinlock_t *lock) -{ - int tmp = ACCESS_ONCE(lock->slock); - - return (((tmp >> 8) - tmp) & 0xff) > 1; -} +#define TICKET_SHIFT 8 static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) { @@ -116,19 +104,7 @@ static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) : "memory", "cc"); } #else -static inline int __ticket_spin_is_locked(raw_spinlock_t *lock) -{ - int tmp = ACCESS_ONCE(lock->slock); - - return (((tmp >> 16) & 0xffff) != (tmp & 0xffff)); -} - -static inline int __ticket_spin_is_contended(raw_spinlock_t *lock) -{ - int tmp = ACCESS_ONCE(lock->slock); - - return (((tmp >> 16) - tmp) & 0xffff) > 1; -} +#define TICKET_SHIFT 16 static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) { @@ -182,6 +158,20 @@ static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) } #endif +static inline int __ticket_spin_is_locked(raw_spinlock_t *lock) +{ + int tmp = ACCESS_ONCE(lock->slock); + + return !!(((tmp >> TICKET_SHIFT) ^ tmp) & ((1 << TICKET_SHIFT) - 1)); +} + +static inline int __ticket_spin_is_contended(raw_spinlock_t *lock) +{ + int tmp = ACCESS_ONCE(lock->slock); + + return (((tmp >> TICKET_SHIFT) - tmp) & ((1 << TICKET_SHIFT) - 1)) > 1; +} + #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) #ifdef CONFIG_PARAVIRT -- cgit v1.2.3-70-g09d2 From 74e91604b2452c15bbe72d77b37cf47ed0310d13 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Fri, 5 Sep 2008 13:27:45 +0100 Subject: x86: ticket spin locks: reduce instruction dependencies Reduce the amount of partial register accesses in the NR_CPUS < 256 case, and slightly weaken resource dependencies in the other case. Signed-off-by: Jan Beulich Signed-off-by: Ingo Molnar --- include/asm-x86/spinlock.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h index 63d3b610a5e..b5a4551fd56 100644 --- a/include/asm-x86/spinlock.h +++ b/include/asm-x86/spinlock.h @@ -21,8 +21,10 @@ #ifdef CONFIG_X86_32 # define LOCK_PTR_REG "a" +# define REG_PTR_MODE "k" #else # define LOCK_PTR_REG "D" +# define REG_PTR_MODE "q" #endif #if defined(CONFIG_X86_32) && \ @@ -77,19 +79,17 @@ static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) { - int tmp; - short new; + int tmp, new; - asm volatile("movw %2,%w0\n\t" + asm volatile("movzwl %2, %0\n\t" "cmpb %h0,%b0\n\t" + "leal 0x100(%" REG_PTR_MODE "0), %1\n\t" "jne 1f\n\t" - "movw %w0,%w1\n\t" - "incb %h1\n\t" LOCK_PREFIX "cmpxchgw %w1,%2\n\t" "1:" "sete %b1\n\t" "movzbl %b1,%0\n\t" - : "=&a" (tmp), "=&Q" (new), "+m" (lock->slock) + : "=&a" (tmp), "=&q" (new), "+m" (lock->slock) : : "memory", "cc"); @@ -136,8 +136,8 @@ static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) "movl %0,%1\n\t" "roll $16, %0\n\t" "cmpl %0,%1\n\t" + "leal 0x00010000(%" REG_PTR_MODE "0), %1\n\t" "jne 1f\n\t" - "addl $0x00010000, %1\n\t" LOCK_PREFIX "cmpxchgl %1,%2\n\t" "1:" "sete %b1\n\t" -- cgit v1.2.3-70-g09d2 From 110e0358e7dfd9cc56d47077068f3680dae10b56 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 28 Aug 2008 13:58:39 -0700 Subject: x86: make sure the CPA test code's use of _PAGE_UNUSED1 is obvious The CPA test code uses _PAGE_UNUSED1, so make sure its obvious. Signed-off-by: Jeremy Fitzhardinge Cc: Nick Piggin Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr-test.c | 6 +++--- include/asm-x86/pgtable.h | 2 ++ 2 files changed, 5 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/pageattr-test.c b/arch/x86/mm/pageattr-test.c index 7c301728711..e1d10690921 100644 --- a/arch/x86/mm/pageattr-test.c +++ b/arch/x86/mm/pageattr-test.c @@ -32,7 +32,7 @@ enum { GPS = (1<<30) }; -#define PAGE_TESTBIT __pgprot(_PAGE_UNUSED1) +#define PAGE_CPA_TEST __pgprot(_PAGE_CPA_TEST) static int pte_testbit(pte_t pte) { @@ -174,7 +174,7 @@ static int pageattr_test(void) } test_addr = addr[i]; - err = change_page_attr_set(&test_addr, len[i], PAGE_TESTBIT, 0); + err = change_page_attr_set(&test_addr, len[i], PAGE_CPA_TEST, 0); if (err < 0) { printk(KERN_ERR "CPA %d failed %d\n", i, err); failed++; @@ -207,7 +207,7 @@ static int pageattr_test(void) continue; } test_addr = addr[i]; - err = change_page_attr_clear(&test_addr, len[i], PAGE_TESTBIT, 0); + err = change_page_attr_clear(&test_addr, len[i], PAGE_CPA_TEST, 0); if (err < 0) { printk(KERN_ERR "CPA reverting failed: %d\n", err); failed++; diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index 04caa2f544d..eccd52406ab 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -19,6 +19,7 @@ #define _PAGE_BIT_UNUSED3 11 #define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */ #define _PAGE_BIT_SPECIAL _PAGE_BIT_UNUSED1 +#define _PAGE_BIT_CPA_TEST _PAGE_BIT_UNUSED1 #define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */ #define _PAGE_PRESENT (_AT(pteval_t, 1) << _PAGE_BIT_PRESENT) @@ -36,6 +37,7 @@ #define _PAGE_PAT (_AT(pteval_t, 1) << _PAGE_BIT_PAT) #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE) #define _PAGE_SPECIAL (_AT(pteval_t, 1) << _PAGE_BIT_SPECIAL) +#define _PAGE_CPA_TEST (_AT(pteval_t, 1) << _PAGE_BIT_CPA_TEST) #define __HAVE_ARCH_PTE_SPECIAL #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) -- cgit v1.2.3-70-g09d2 From 913da64b54b2b3bb212a59aba2e6f2b8294ca1fa Mon Sep 17 00:00:00 2001 From: Alex Nixon Date: Wed, 3 Sep 2008 14:30:23 +0100 Subject: x86: build fix for !CONFIG_SMP Move reset_lazy_tlbstate into tlb_32.c, and define noop versions of play_dead() in process_{32,64}.c when !CONFIG_SMP. Signed-off-by: Alex Nixon Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/common.c | 7 ------- arch/x86/kernel/process_32.c | 7 +++++++ arch/x86/kernel/process_64.c | 7 +++++++ arch/x86/kernel/tlb_32.c | 8 ++++++++ include/asm-x86/smp.h | 9 --------- include/asm-x86/tlbflush.h | 10 ++++++++++ 6 files changed, 32 insertions(+), 16 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 18f5551b32d..76d10d5c2fa 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -714,10 +714,3 @@ void __cpuinit cpu_init(void) mxcsr_feature_mask_init(); } -void reset_lazy_tlbstate(void) -{ - int cpu = raw_smp_processor_id(); - - per_cpu(cpu_tlbstate, cpu).state = 0; - per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm; -} diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 633cf06578f..b76b38ff962 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -72,6 +72,13 @@ unsigned long thread_saved_pc(struct task_struct *tsk) return ((unsigned long *)tsk->thread.sp)[3]; } +#ifndef CONFIG_SMP +static inline void play_dead(void) +{ + BUG(); +} +#endif + /* * The idle thread. There's no useful work to be * done, so just try to conserve power and have a diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index aa28ed01cdf..ec27afa43d7 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -85,6 +85,13 @@ void exit_idle(void) __exit_idle(); } +#ifndef CONFIG_SMP +static inline void play_dead(void) +{ + BUG(); +} +#endif + /* * The idle thread. There's no useful work to be * done, so just try to conserve power and have a diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c index fec1ecedc9b..e00534b3353 100644 --- a/arch/x86/kernel/tlb_32.c +++ b/arch/x86/kernel/tlb_32.c @@ -241,3 +241,11 @@ void flush_tlb_all(void) on_each_cpu(do_flush_tlb_all, NULL, 1); } +void reset_lazy_tlbstate(void) +{ + int cpu = raw_smp_processor_id(); + + per_cpu(cpu_tlbstate, cpu).state = 0; + per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm; +} + diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 8bdaa4a25f0..30b5146cc43 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -222,14 +222,5 @@ static inline int hard_smp_processor_id(void) #endif /* CONFIG_X86_LOCAL_APIC */ -#ifdef CONFIG_HOTPLUG_CPU -#ifdef CONFIG_X86_32 -extern void reset_lazy_tlbstate(void); -#else -static inline void reset_lazy_tlbstate(void) -{ } -#endif /* CONFIG_X86_32 */ -#endif - #endif /* __ASSEMBLY__ */ #endif diff --git a/include/asm-x86/tlbflush.h b/include/asm-x86/tlbflush.h index 35c76ceb9f4..0e7bbb54911 100644 --- a/include/asm-x86/tlbflush.h +++ b/include/asm-x86/tlbflush.h @@ -119,6 +119,10 @@ static inline void native_flush_tlb_others(const cpumask_t *cpumask, { } +static inline void reset_lazy_tlbstate(void) +{ +} + #else /* SMP */ #include @@ -151,6 +155,12 @@ struct tlb_state { char __cacheline_padding[L1_CACHE_BYTES-8]; }; DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate); + +void reset_lazy_tlbstate(void); +#else +static inline void reset_lazy_tlbstate(void) +{ +} #endif #endif /* SMP */ -- cgit v1.2.3-70-g09d2 From 4ab4ba32aa16b012cb0faabf1a27952508fe67f2 Mon Sep 17 00:00:00 2001 From: Petr Tesarik Date: Wed, 3 Sep 2008 13:31:42 +0200 Subject: x86, tracehook: clean up implementation of syscall_get_error() The x86-tracehook code now contains this line in syscall_get_error(): return error >= -4095L ? error : 0; Hard-wiring a constant is not nice. Let's use the IS_ERR_VALUE macro from linux/err.h instead. Signed-off-by: Petr Tesarik Cc: utrace-devel@redhat.com Acked-by: Roland McGrath Signed-off-by: Ingo Molnar --- include/asm-x86/syscall.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/syscall.h b/include/asm-x86/syscall.h index 6f293892895..04c47dc5597 100644 --- a/include/asm-x86/syscall.h +++ b/include/asm-x86/syscall.h @@ -14,6 +14,7 @@ #define _ASM_SYSCALL_H 1 #include +#include static inline long syscall_get_nr(struct task_struct *task, struct pt_regs *regs) @@ -47,7 +48,7 @@ static inline long syscall_get_error(struct task_struct *task, */ error = (long) (int) error; #endif - return error >= -4095L ? error : 0; + return IS_ERR_VALUE(error) ? error : 0; } static inline long syscall_get_return_value(struct task_struct *task, -- cgit v1.2.3-70-g09d2 From e51af6630848406fc97adbd71443818cdcda297b Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Thu, 4 Sep 2008 09:54:37 +0100 Subject: x86: blacklist DMAR on Intel G31/G33 chipsets Some BIOSes (the Intel DG33BU, for example) wrongly claim to have DMAR when they don't. Avoid the resulting crashes when it doesn't work as expected. I'd still be grateful if someone could test it on a DG33BU with the old BIOS though, since I've killed mine. I tested the DMI version, but not this one. Signed-off-by: David Woodhouse Signed-off-by: Ingo Molnar --- arch/x86/kernel/early-quirks.c | 18 ++++++++++++++++++ drivers/pci/intel-iommu.c | 2 +- include/asm-x86/iommu.h | 1 + 3 files changed, 20 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c index 4353cf5e6fa..24bb5faf5ef 100644 --- a/arch/x86/kernel/early-quirks.c +++ b/arch/x86/kernel/early-quirks.c @@ -95,6 +95,20 @@ static void __init nvidia_bugs(int num, int slot, int func) } +#ifdef CONFIG_DMAR +static void __init intel_g33_dmar(int num, int slot, int func) +{ + struct acpi_table_header *dmar_tbl; + acpi_status status; + + status = acpi_get_table(ACPI_SIG_DMAR, 0, &dmar_tbl); + if (ACPI_SUCCESS(status)) { + printk(KERN_INFO "BIOS BUG: DMAR advertised on Intel G31/G33 chipset -- ignoring\n"); + dmar_disabled = 1; + } +} +#endif + #define QFLAG_APPLY_ONCE 0x1 #define QFLAG_APPLIED 0x2 #define QFLAG_DONE (QFLAG_APPLY_ONCE|QFLAG_APPLIED) @@ -114,6 +128,10 @@ static struct chipset early_qrk[] __initdata = { PCI_CLASS_BRIDGE_PCI, PCI_ANY_ID, QFLAG_APPLY_ONCE, via_bugs }, { PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB, PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, fix_hypertransport_config }, +#ifdef CONFIG_DMAR + { PCI_VENDOR_ID_INTEL, 0x29c0, + PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, intel_g33_dmar }, +#endif {} }; diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c index 8d0e60ac849..eaba6ecc2ad 100644 --- a/drivers/pci/intel-iommu.c +++ b/drivers/pci/intel-iommu.c @@ -80,7 +80,7 @@ static long list_size; static void domain_remove_dev_info(struct dmar_domain *domain); -static int dmar_disabled; +int dmar_disabled; static int __initdata dmar_map_gfx = 1; static int dmar_forcedac; static int intel_iommu_strict; diff --git a/include/asm-x86/iommu.h b/include/asm-x86/iommu.h index 5f888cc5be4..621a1af94c4 100644 --- a/include/asm-x86/iommu.h +++ b/include/asm-x86/iommu.h @@ -6,6 +6,7 @@ extern void no_iommu_init(void); extern struct dma_mapping_ops nommu_dma_ops; extern int force_iommu, no_iommu; extern int iommu_detected; +extern int dmar_disabled; extern unsigned long iommu_num_pages(unsigned long addr, unsigned long len); -- cgit v1.2.3-70-g09d2 From 72fa50f4ef9014f4212945b766af84ea94308903 Mon Sep 17 00:00:00 2001 From: Hiroshi Shimamoto Date: Fri, 5 Sep 2008 16:27:11 -0700 Subject: x86_32: signal: introduce signal_fault() implement signal_fault() for 32bit. Signed-off-by: Hiroshi Shimamoto Signed-off-by: Ingo Molnar --- arch/x86/kernel/signal_32.c | 18 +++++++++++++++++- include/asm-x86/ptrace.h | 4 ++-- 2 files changed, 19 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c index 3e4a688bb84..76d05d70384 100644 --- a/arch/x86/kernel/signal_32.c +++ b/arch/x86/kernel/signal_32.c @@ -243,7 +243,7 @@ asmlinkage int sys_rt_sigreturn(unsigned long __unused) return ax; badframe: - force_sig(SIGSEGV, current); + signal_fault(regs, frame, "rt sigreturn"); return 0; } @@ -669,3 +669,19 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) clear_thread_flag(TIF_IRET); } + +void signal_fault(struct pt_regs *regs, void __user *frame, char *where) +{ + struct task_struct *me = current; + + if (show_unhandled_signals && printk_ratelimit()) { + printk(KERN_INFO + "%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx", + me->comm, me->pid, where, frame, + regs->ip, regs->sp, regs->orig_ax); + print_vma_addr(" in ", regs->ip); + printk(KERN_CONT "\n"); + } + + force_sig(SIGSEGV, me); +} diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h index a33f027dbbe..fad80776991 100644 --- a/include/asm-x86/ptrace.h +++ b/include/asm-x86/ptrace.h @@ -144,10 +144,10 @@ convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs); #ifdef CONFIG_X86_32 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code); -#else -void signal_fault(struct pt_regs *regs, void __user *frame, char *where); #endif +void signal_fault(struct pt_regs *regs, void __user *frame, char *where); + extern long syscall_trace_enter(struct pt_regs *); extern void syscall_trace_leave(struct pt_regs *); -- cgit v1.2.3-70-g09d2 From afe73824f52d6767c77e9456f573a76075108279 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Fri, 29 Aug 2008 13:15:28 +0100 Subject: x86-64: eliminate dead code Signed-off-by: Jan Beulich Signed-off-by: Ingo Molnar --- include/asm-x86/mmu.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/mmu.h b/include/asm-x86/mmu.h index 00e88679e11..80a1dee5bea 100644 --- a/include/asm-x86/mmu.h +++ b/include/asm-x86/mmu.h @@ -7,14 +7,9 @@ /* * The x86 doesn't have a mmu context, but * we put the segment information here. - * - * cpu_vm_mask is used to optimize ldt flushing. */ typedef struct { void *ldt; -#ifdef CONFIG_X86_64 - rwlock_t ldtlock; -#endif int size; struct mutex lock; void *vdso; -- cgit v1.2.3-70-g09d2 From 96bf7adbaed24da0293dd87466e27b4151ae9da8 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Mon, 1 Sep 2008 14:10:44 +0100 Subject: x86_64: Use Signed-off-by: David Woodhouse Acked-by: Ingo Molnar --- include/asm-x86/statfs.h | 61 ++++-------------------------------------------- 1 file changed, 5 insertions(+), 56 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/statfs.h b/include/asm-x86/statfs.h index 7c651aa9725..4c935caaef2 100644 --- a/include/asm-x86/statfs.h +++ b/include/asm-x86/statfs.h @@ -1,63 +1,12 @@ #ifndef _ASM_X86_STATFS_H #define _ASM_X86_STATFS_H -#ifdef __i386__ -#include -#else - -#ifndef __KERNEL_STRICT_NAMES - -#include - -typedef __kernel_fsid_t fsid_t; - -#endif - /* - * This is ugly -- we're already 64-bit clean, so just duplicate the - * definitions. + * We need compat_statfs64 to be packed, because the i386 ABI won't + * add padding at the end to bring it to a multiple of 8 bytes, but + * the x86_64 ABI will. */ -struct statfs { - long f_type; - long f_bsize; - long f_blocks; - long f_bfree; - long f_bavail; - long f_files; - long f_ffree; - __kernel_fsid_t f_fsid; - long f_namelen; - long f_frsize; - long f_spare[5]; -}; - -struct statfs64 { - long f_type; - long f_bsize; - long f_blocks; - long f_bfree; - long f_bavail; - long f_files; - long f_ffree; - __kernel_fsid_t f_fsid; - long f_namelen; - long f_frsize; - long f_spare[5]; -}; +#define ARCH_PACK_COMPAT_STATFS64 __attribute__((packed,aligned(4))) -struct compat_statfs64 { - __u32 f_type; - __u32 f_bsize; - __u64 f_blocks; - __u64 f_bfree; - __u64 f_bavail; - __u64 f_files; - __u64 f_ffree; - __kernel_fsid_t f_fsid; - __u32 f_namelen; - __u32 f_frsize; - __u32 f_spare[5]; -} __attribute__((packed)); - -#endif /* !__i386__ */ +#include #endif -- cgit v1.2.3-70-g09d2 From 11fdd252bb5b461289fc5c21dc8fc87dc66f3284 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 7 Sep 2008 17:58:50 -0700 Subject: x86: cpu make amd.c more like amd_64.c v2 1. make 32bit have early_init_amd_mc and amd_detect_cmp 2. seperate init_amd_k5/k6/k7 ... v2: fix compiling for !CONFIG_SMP Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/amd.c | 396 ++++++++++++++++++++++++------------------- arch/x86/kernel/cpu/amd_64.c | 17 +- arch/x86/kernel/cpu/common.c | 2 +- include/asm-x86/processor.h | 2 +- 4 files changed, 236 insertions(+), 181 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index c3175da7bc6..a3a9e3cbdea 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -24,8 +24,200 @@ extern void vide(void); __asm__(".align 4\nvide: ret"); +static void __cpuinit init_amd_k5(struct cpuinfo_x86 *c) +{ +/* + * General Systems BIOSen alias the cpu frequency registers + * of the Elan at 0x000df000. Unfortuantly, one of the Linux + * drivers subsequently pokes it, and changes the CPU speed. + * Workaround : Remove the unneeded alias. + */ +#define CBAR (0xfffc) /* Configuration Base Address (32-bit) */ +#define CBAR_ENB (0x80000000) +#define CBAR_KEY (0X000000CB) + if (c->x86_model == 9 || c->x86_model == 10) { + if (inl (CBAR) & CBAR_ENB) + outl (0 | CBAR_KEY, CBAR); + } +} + + +static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c) +{ + u32 l, h; + int mbytes = num_physpages >> (20-PAGE_SHIFT); + + if (c->x86_model < 6) { + /* Based on AMD doc 20734R - June 2000 */ + if (c->x86_model == 0) { + clear_cpu_cap(c, X86_FEATURE_APIC); + set_cpu_cap(c, X86_FEATURE_PGE); + } + return; + } + + if (c->x86_model == 6 && c->x86_mask == 1) { + const int K6_BUG_LOOP = 1000000; + int n; + void (*f_vide)(void); + unsigned long d, d2; + + printk(KERN_INFO "AMD K6 stepping B detected - "); + + /* + * It looks like AMD fixed the 2.6.2 bug and improved indirect + * calls at the same time. + */ + + n = K6_BUG_LOOP; + f_vide = vide; + rdtscl(d); + while (n--) + f_vide(); + rdtscl(d2); + d = d2-d; + + if (d > 20*K6_BUG_LOOP) + printk("system stability may be impaired when more than 32 MB are used.\n"); + else + printk("probably OK (after B9730xxxx).\n"); + printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n"); + } + + /* K6 with old style WHCR */ + if (c->x86_model < 8 || + (c->x86_model == 8 && c->x86_mask < 8)) { + /* We can only write allocate on the low 508Mb */ + if (mbytes > 508) + mbytes = 508; + + rdmsr(MSR_K6_WHCR, l, h); + if ((l&0x0000FFFF) == 0) { + unsigned long flags; + l = (1<<0)|((mbytes/4)<<1); + local_irq_save(flags); + wbinvd(); + wrmsr(MSR_K6_WHCR, l, h); + local_irq_restore(flags); + printk(KERN_INFO "Enabling old style K6 write allocation for %d Mb\n", + mbytes); + } + return; + } + + if ((c->x86_model == 8 && c->x86_mask > 7) || + c->x86_model == 9 || c->x86_model == 13) { + /* The more serious chips .. */ + + if (mbytes > 4092) + mbytes = 4092; + + rdmsr(MSR_K6_WHCR, l, h); + if ((l&0xFFFF0000) == 0) { + unsigned long flags; + l = ((mbytes>>2)<<22)|(1<<16); + local_irq_save(flags); + wbinvd(); + wrmsr(MSR_K6_WHCR, l, h); + local_irq_restore(flags); + printk(KERN_INFO "Enabling new style K6 write allocation for %d Mb\n", + mbytes); + } + + return; + } + + if (c->x86_model == 10) { + /* AMD Geode LX is model 10 */ + /* placeholder for any needed mods */ + return; + } +} + +static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c) +{ + u32 l, h; + + /* + * Bit 15 of Athlon specific MSR 15, needs to be 0 + * to enable SSE on Palomino/Morgan/Barton CPU's. + * If the BIOS didn't enable it already, enable it here. + */ + if (c->x86_model >= 6 && c->x86_model <= 10) { + if (!cpu_has(c, X86_FEATURE_XMM)) { + printk(KERN_INFO "Enabling disabled K7/SSE Support.\n"); + rdmsr(MSR_K7_HWCR, l, h); + l &= ~0x00008000; + wrmsr(MSR_K7_HWCR, l, h); + set_cpu_cap(c, X86_FEATURE_XMM); + } + } + + /* + * It's been determined by AMD that Athlons since model 8 stepping 1 + * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx + * As per AMD technical note 27212 0.2 + */ + if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) { + rdmsr(MSR_K7_CLK_CTL, l, h); + if ((l & 0xfff00000) != 0x20000000) { + printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l, + ((l & 0x000fffff)|0x20000000)); + wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h); + } + } + + set_cpu_cap(c, X86_FEATURE_K7); +} + +/* + * On a AMD dual core setup the lower bits of the APIC id distingush the cores. + * Assumes number of cores is a power of two. + */ +static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c) +{ +#ifdef CONFIG_X86_HT + unsigned bits; + + bits = c->x86_coreid_bits; + + /* Low order bits define the core id (index of core in socket) */ + c->cpu_core_id = c->initial_apicid & ((1 << bits)-1); + /* Convert the initial APIC ID into the socket ID */ + c->phys_proc_id = c->initial_apicid >> bits; +#endif +} + +static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c) +{ +#ifdef CONFIG_X86_HT + unsigned bits, ecx; + + /* Multi core CPU? */ + if (c->extended_cpuid_level < 0x80000008) + return; + + ecx = cpuid_ecx(0x80000008); + + c->x86_max_cores = (ecx & 0xff) + 1; + + /* CPU telling us the core id bits shift? */ + bits = (ecx >> 12) & 0xF; + + /* Otherwise recompute */ + if (bits == 0) { + while ((1 << bits) < c->x86_max_cores) + bits++; + } + + c->x86_coreid_bits = bits; +#endif +} + static void __cpuinit early_init_amd(struct cpuinfo_x86 *c) { + early_init_amd_mc(c); + if (c->x86_power & (1<<8)) set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); @@ -37,9 +229,6 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c) static void __cpuinit init_amd(struct cpuinfo_x86 *c) { - u32 l, h; - int mbytes = num_physpages >> (20-PAGE_SHIFT); - #ifdef CONFIG_SMP unsigned long long value; @@ -50,7 +239,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) * Errata 63 for SH-B3 steppings * Errata 122 for all steppings (F+ have it disabled by default) */ - if (c->x86 == 15) { + if (c->x86 == 0xf) { rdmsrl(MSR_K7_HWCR, value); value |= 1 << 6; wrmsrl(MSR_K7_HWCR, value); @@ -73,192 +262,55 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) switch (c->x86) { case 4: - /* - * General Systems BIOSen alias the cpu frequency registers - * of the Elan at 0x000df000. Unfortuantly, one of the Linux - * drivers subsequently pokes it, and changes the CPU speed. - * Workaround : Remove the unneeded alias. - */ -#define CBAR (0xfffc) /* Configuration Base Address (32-bit) */ -#define CBAR_ENB (0x80000000) -#define CBAR_KEY (0X000000CB) - if (c->x86_model == 9 || c->x86_model == 10) { - if (inl (CBAR) & CBAR_ENB) - outl (0 | CBAR_KEY, CBAR); - } - break; + init_amd_k5(c); + break; case 5: - if (c->x86_model < 6) { - /* Based on AMD doc 20734R - June 2000 */ - if (c->x86_model == 0) { - clear_cpu_cap(c, X86_FEATURE_APIC); - set_cpu_cap(c, X86_FEATURE_PGE); - } - break; - } - - if (c->x86_model == 6 && c->x86_mask == 1) { - const int K6_BUG_LOOP = 1000000; - int n; - void (*f_vide)(void); - unsigned long d, d2; - - printk(KERN_INFO "AMD K6 stepping B detected - "); - - /* - * It looks like AMD fixed the 2.6.2 bug and improved indirect - * calls at the same time. - */ - - n = K6_BUG_LOOP; - f_vide = vide; - rdtscl(d); - while (n--) - f_vide(); - rdtscl(d2); - d = d2-d; - - if (d > 20*K6_BUG_LOOP) - printk("system stability may be impaired when more than 32 MB are used.\n"); - else - printk("probably OK (after B9730xxxx).\n"); - printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n"); - } - - /* K6 with old style WHCR */ - if (c->x86_model < 8 || - (c->x86_model == 8 && c->x86_mask < 8)) { - /* We can only write allocate on the low 508Mb */ - if (mbytes > 508) - mbytes = 508; - - rdmsr(MSR_K6_WHCR, l, h); - if ((l&0x0000FFFF) == 0) { - unsigned long flags; - l = (1<<0)|((mbytes/4)<<1); - local_irq_save(flags); - wbinvd(); - wrmsr(MSR_K6_WHCR, l, h); - local_irq_restore(flags); - printk(KERN_INFO "Enabling old style K6 write allocation for %d Mb\n", - mbytes); - } - break; - } - - if ((c->x86_model == 8 && c->x86_mask > 7) || - c->x86_model == 9 || c->x86_model == 13) { - /* The more serious chips .. */ - - if (mbytes > 4092) - mbytes = 4092; - - rdmsr(MSR_K6_WHCR, l, h); - if ((l&0xFFFF0000) == 0) { - unsigned long flags; - l = ((mbytes>>2)<<22)|(1<<16); - local_irq_save(flags); - wbinvd(); - wrmsr(MSR_K6_WHCR, l, h); - local_irq_restore(flags); - printk(KERN_INFO "Enabling new style K6 write allocation for %d Mb\n", - mbytes); - } - - break; - } - - if (c->x86_model == 10) { - /* AMD Geode LX is model 10 */ - /* placeholder for any needed mods */ - break; - } - break; - case 6: /* An Athlon/Duron */ - - /* - * Bit 15 of Athlon specific MSR 15, needs to be 0 - * to enable SSE on Palomino/Morgan/Barton CPU's. - * If the BIOS didn't enable it already, enable it here. - */ - if (c->x86_model >= 6 && c->x86_model <= 10) { - if (!cpu_has(c, X86_FEATURE_XMM)) { - printk(KERN_INFO "Enabling disabled K7/SSE Support.\n"); - rdmsr(MSR_K7_HWCR, l, h); - l &= ~0x00008000; - wrmsr(MSR_K7_HWCR, l, h); - set_cpu_cap(c, X86_FEATURE_XMM); - } - } - - /* - * It's been determined by AMD that Athlons since model 8 stepping 1 - * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx - * As per AMD technical note 27212 0.2 - */ - if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) { - rdmsr(MSR_K7_CLK_CTL, l, h); - if ((l & 0xfff00000) != 0x20000000) { - printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l, - ((l & 0x000fffff)|0x20000000)); - wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h); - } - } - break; - } - - switch (c->x86) { - case 15: - /* Use K8 tuning for Fam10h and Fam11h */ - case 0x10: - case 0x11: - set_cpu_cap(c, X86_FEATURE_K8); + init_amd_k6(c); break; - case 6: - set_cpu_cap(c, X86_FEATURE_K7); + case 6: /* An Athlon/Duron */ + init_amd_k7(c); break; } + + /* K6s reports MCEs but don't actually have all the MSRs */ + if (c->x86 < 6) + clear_cpu_cap(c, X86_FEATURE_MCE); + if (c->x86 >= 6) set_cpu_cap(c, X86_FEATURE_FXSAVE_LEAK); - display_cacheinfo(c); + if (!c->x86_model_id[0]) { + switch (c->x86) { + case 0xf: + /* Should distinguish Models here, but this is only + a fallback anyways. */ + strcpy(c->x86_model_id, "Hammer"); + break; + } + } - if (cpuid_eax(0x80000000) >= 0x80000008) - c->x86_max_cores = (cpuid_ecx(0x80000008) & 0xff) + 1; + display_cacheinfo(c); -#ifdef CONFIG_X86_HT - /* - * On a AMD multi core setup the lower bits of the APIC id - * distinguish the cores. - */ - if (c->x86_max_cores > 1) { - int cpu = smp_processor_id(); - unsigned bits = (cpuid_ecx(0x80000008) >> 12) & 0xf; + /* Multi core CPU? */ + if (c->extended_cpuid_level >= 0x80000008) + amd_detect_cmp(c); - if (bits == 0) { - while ((1 << bits) < c->x86_max_cores) - bits++; - } - c->cpu_core_id = c->phys_proc_id & ((1<phys_proc_id >>= bits; - printk(KERN_INFO "CPU %d(%d) -> Core %d\n", - cpu, c->x86_max_cores, c->cpu_core_id); - } -#endif + detect_ht(c); - if (cpuid_eax(0x80000000) >= 0x80000006) { - if ((c->x86 == 0x10) && (cpuid_edx(0x80000006) & 0xf000)) + if (c->extended_cpuid_level >= 0x80000006) { + if ((c->x86 >= 0x0f) && (cpuid_edx(0x80000006) & 0xf000)) num_cache_leaves = 4; else num_cache_leaves = 3; } - /* K6s reports MCEs but don't actually have all the MSRs */ - if (c->x86 < 6) - clear_cpu_cap(c, X86_FEATURE_MCE); + if (c->x86 >= 0xf && c->x86 <= 0x11) + set_cpu_cap(c, X86_FEATURE_K8); - if (cpu_has_xmm2) + if (cpu_has_xmm2) { + /* MFENCE stops RDTSC speculation */ set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC); + } } static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int size) diff --git a/arch/x86/kernel/cpu/amd_64.c b/arch/x86/kernel/cpu/amd_64.c index 8c2d07f06f1..7913e48f673 100644 --- a/arch/x86/kernel/cpu/amd_64.c +++ b/arch/x86/kernel/cpu/amd_64.c @@ -174,17 +174,20 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) if (c->extended_cpuid_level >= 0x80000008) amd_detect_cmp(c); - if (c->extended_cpuid_level >= 0x80000006 && - (cpuid_edx(0x80000006) & 0xf000)) - num_cache_leaves = 4; - else - num_cache_leaves = 3; + if (c->extended_cpuid_level >= 0x80000006) { + if ((c->x86 >= 0x0f) && (cpuid_edx(0x80000006) & 0xf000)) + num_cache_leaves = 4; + else + num_cache_leaves = 3; + } if (c->x86 >= 0xf && c->x86 <= 0x11) set_cpu_cap(c, X86_FEATURE_K8); - /* MFENCE stops RDTSC speculation */ - set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC); + if (cpu_has_xmm2) { + /* MFENCE stops RDTSC speculation */ + set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC); + } if (c->x86 == 0x10) { /* do this for boot cpu */ diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 5cde4b18448..9a57106c199 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -629,8 +629,8 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) c->x86_vendor_id[0] = '\0'; /* Unset */ c->x86_model_id[0] = '\0'; /* Unset */ c->x86_max_cores = 1; -#ifdef CONFIG_X86_64 c->x86_coreid_bits = 0; +#ifdef CONFIG_X86_64 c->x86_clflush_size = 64; #else c->cpuid_level = -1; /* CPUID not detected */ diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index fc5e961c5b6..62531ecbde1 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -75,9 +75,9 @@ struct cpuinfo_x86 { int x86_tlbsize; __u8 x86_virt_bits; __u8 x86_phys_bits; +#endif /* CPUID returned core id bits: */ __u8 x86_coreid_bits; -#endif /* Max extended CPUID function supported: */ __u32 extended_cpuid_level; /* Maximum supported CPUID level, -1=no CPUID: */ -- cgit v1.2.3-70-g09d2 From de9f521fb72dd091aa4989fe2e004ecf4785a850 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Mon, 8 Sep 2008 18:10:11 +0900 Subject: x86: move pci-nommu's dma_mask check to common code The check to see if dev->dma_mask is NULL in pci-nommu is more appropriate for dma_alloc_coherent(). Signed-off-by: FUJITA Tomonori Acked-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/pci-nommu.c | 3 --- include/asm-x86/dma-mapping.h | 3 +++ 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c index 73853d3fdca..0f51883cc6a 100644 --- a/arch/x86/kernel/pci-nommu.c +++ b/arch/x86/kernel/pci-nommu.c @@ -80,9 +80,6 @@ nommu_alloc_coherent(struct device *hwdev, size_t size, int node; struct page *page; - if (hwdev->dma_mask == NULL) - return NULL; - gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); gfp |= __GFP_ZERO; diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index 3a9a6f5e681..088c56814aa 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -254,6 +254,9 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp |= GFP_DMA; } + if (!dev->dma_mask) + return NULL; + if (ops->alloc_coherent) return ops->alloc_coherent(dev, size, dma_handle, gfp); -- cgit v1.2.3-70-g09d2 From 8a53ad675f86ee003482b557da944e070d3c4859 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Mon, 8 Sep 2008 18:10:12 +0900 Subject: x86: fix nommu_alloc_coherent allocation with NULL device argument We need to use __GFP_DMA for NULL device argument (fallback_dev) with pci-nommu. It's a hack for ISA (and some old code) so we need to use GFP_DMA. Signed-off-by: FUJITA Tomonori Acked-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/pci-nommu.c | 3 +-- include/asm-x86/dma-mapping.h | 2 ++ 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c index 0f51883cc6a..ada1c87cafc 100644 --- a/arch/x86/kernel/pci-nommu.c +++ b/arch/x86/kernel/pci-nommu.c @@ -80,7 +80,6 @@ nommu_alloc_coherent(struct device *hwdev, size_t size, int node; struct page *page; - gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); gfp |= __GFP_ZERO; dma_mask = hwdev->coherent_dma_mask; @@ -93,7 +92,7 @@ nommu_alloc_coherent(struct device *hwdev, size_t size, node = dev_to_node(hwdev); #ifdef CONFIG_X86_64 - if (dma_mask <= DMA_32BIT_MASK) + if (dma_mask <= DMA_32BIT_MASK && !(gfp & GFP_DMA)) gfp |= GFP_DMA32; #endif diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index 088c56814aa..ad8b49032d1 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -246,6 +246,8 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, struct dma_mapping_ops *ops = get_dma_ops(dev); void *memory; + gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); + if (dma_alloc_from_coherent(dev, size, dma_handle, &memory)) return memory; -- cgit v1.2.3-70-g09d2 From 823e7e8c6ef12cd1943dc42fe7595ca74e8cc3d7 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Mon, 8 Sep 2008 18:10:13 +0900 Subject: x86: dma_alloc_coherent sets gfp flags properly Non real IOMMU implemenations (which doesn't do virtual mappings, e.g. swiotlb, pci-nommu, etc) need to use proper gfp flags and dma_mask to allocate pages in their own dma_alloc_coherent() (allocated page need to be suitable for device's coherent_dma_mask). This patch makes dma_alloc_coherent do this job so that IOMMUs don't need to take care of it any more. Real IOMMU implemenataions can simply ignore the gfp flags. Signed-off-by: FUJITA Tomonori Acked-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/pci-nommu.c | 19 ++----------------- include/asm-x86/dma-mapping.h | 32 ++++++++++++++++++++++++++++---- 2 files changed, 30 insertions(+), 21 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c index ada1c87cafc..8e398b56f50 100644 --- a/arch/x86/kernel/pci-nommu.c +++ b/arch/x86/kernel/pci-nommu.c @@ -80,26 +80,11 @@ nommu_alloc_coherent(struct device *hwdev, size_t size, int node; struct page *page; - gfp |= __GFP_ZERO; - - dma_mask = hwdev->coherent_dma_mask; - if (!dma_mask) - dma_mask = *(hwdev->dma_mask); + dma_mask = dma_alloc_coherent_mask(hwdev, gfp); - if (dma_mask < DMA_24BIT_MASK) - return NULL; + gfp |= __GFP_ZERO; node = dev_to_node(hwdev); - -#ifdef CONFIG_X86_64 - if (dma_mask <= DMA_32BIT_MASK && !(gfp & GFP_DMA)) - gfp |= GFP_DMA32; -#endif - - /* No alloc-free penalty for ISA devices */ - if (dma_mask == DMA_24BIT_MASK) - gfp |= GFP_DMA; - again: page = alloc_pages_node(node, gfp, get_order(size)); if (!page) diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index ad8b49032d1..0cc022b9a4a 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -239,6 +239,29 @@ static inline int dma_get_cache_alignment(void) return boot_cpu_data.x86_clflush_size; } +static inline unsigned long dma_alloc_coherent_mask(struct device *dev, + gfp_t gfp) +{ + unsigned long dma_mask = 0; + + dma_mask = dev->coherent_dma_mask; + if (!dma_mask) + dma_mask = (gfp & GFP_DMA) ? DMA_24BIT_MASK : DMA_32BIT_MASK; + + return dma_mask; +} + +static inline gfp_t dma_alloc_coherent_gfp_flags(struct device *dev, gfp_t gfp) +{ + unsigned long dma_mask = dma_alloc_coherent_mask(dev, gfp); + +#ifdef CONFIG_X86_64 + if (dma_mask <= DMA_32BIT_MASK && !(gfp & GFP_DMA)) + gfp |= GFP_DMA32; +#endif + return gfp; +} + static inline void * dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t gfp) @@ -259,10 +282,11 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, if (!dev->dma_mask) return NULL; - if (ops->alloc_coherent) - return ops->alloc_coherent(dev, size, - dma_handle, gfp); - return NULL; + if (!ops->alloc_coherent) + return NULL; + + return ops->alloc_coherent(dev, size, dma_handle, + dma_alloc_coherent_gfp_flags(dev, gfp)); } static inline void dma_free_coherent(struct device *dev, size_t size, -- cgit v1.2.3-70-g09d2 From 9fcaff0e660d886e9a766460adbe558dd25de31b Mon Sep 17 00:00:00 2001 From: Steven Noonan Date: Mon, 8 Sep 2008 16:19:11 -0700 Subject: x86: unused variable in dma_alloc_coherent_gfp_flags() Fixed a warning caused by a badly placed ifdef. Signed-off-by: Steven Noonan Signed-off-by: Ingo Molnar --- include/asm-x86/dma-mapping.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index 0cc022b9a4a..56075320b81 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -253,9 +253,9 @@ static inline unsigned long dma_alloc_coherent_mask(struct device *dev, static inline gfp_t dma_alloc_coherent_gfp_flags(struct device *dev, gfp_t gfp) { +#ifdef CONFIG_X86_64 unsigned long dma_mask = dma_alloc_coherent_mask(dev, gfp); -#ifdef CONFIG_X86_64 if (dma_mask <= DMA_32BIT_MASK && !(gfp & GFP_DMA)) gfp |= GFP_DMA32; #endif -- cgit v1.2.3-70-g09d2 From 879d792b66d633bbe466974f61d1acc9aa8c78eb Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Tue, 9 Sep 2008 16:40:37 -0700 Subject: x86: let intel 64-bit use intel.c now that arch/x86/kernel/cpu/intel_64.c and arch/x86/kernel/cpu/intel.c are equal, drop arch/x86/kernel/cpu/intel_64.c and fix up the glue. Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/Kconfig.cpu | 10 +- arch/x86/kernel/cpu/Makefile | 3 +- arch/x86/kernel/cpu/intel_64.c | 366 ----------------------------------------- include/asm-x86/bugs.h | 2 +- 4 files changed, 3 insertions(+), 378 deletions(-) delete mode 100644 arch/x86/kernel/cpu/intel_64.c (limited to 'include/asm-x86') diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index 1b29d6a8756..6761848329f 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -423,17 +423,9 @@ menuconfig PROCESSOR_SELECT This lets you choose what x86 vendor support code your kernel will include. -config CPU_SUP_INTEL_32 +config CPU_SUP_INTEL default y bool "Support Intel processors" if PROCESSOR_SELECT - depends on !64BIT - help - This enables extended support for Intel processors - -config CPU_SUP_INTEL_64 - default y - bool "Support Intel processors" if PROCESSOR_SELECT - depends on 64BIT help This enables extended support for Intel processors diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 510d1bcb058..7f0b45a5d78 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -8,8 +8,7 @@ obj-y += proc.o capflags.o powerflags.o common.o obj-$(CONFIG_X86_32) += bugs.o cmpxchg.o obj-$(CONFIG_X86_64) += bugs_64.o -obj-$(CONFIG_CPU_SUP_INTEL_32) += intel.o -obj-$(CONFIG_CPU_SUP_INTEL_64) += intel_64.o +obj-$(CONFIG_CPU_SUP_INTEL) += intel.o obj-$(CONFIG_CPU_SUP_AMD) += amd.o obj-$(CONFIG_CPU_SUP_CYRIX_32) += cyrix.o obj-$(CONFIG_CPU_SUP_CENTAUR_32) += centaur.o diff --git a/arch/x86/kernel/cpu/intel_64.c b/arch/x86/kernel/cpu/intel_64.c deleted file mode 100644 index 365a008080c..00000000000 --- a/arch/x86/kernel/cpu/intel_64.c +++ /dev/null @@ -1,366 +0,0 @@ -#include -#include - -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include - -#ifdef CONFIG_X86_64 -#include -#include -#endif - -#include "cpu.h" - -#ifdef CONFIG_X86_LOCAL_APIC -#include -#include -#include -#endif - -static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) -{ - if ((c->x86 == 0xf && c->x86_model >= 0x03) || - (c->x86 == 0x6 && c->x86_model >= 0x0e)) - set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); - -#ifdef CONFIG_X86_64 - set_cpu_cap(c, X86_FEATURE_SYSENTER32); -#else - /* Netburst reports 64 bytes clflush size, but does IO in 128 bytes */ - if (c->x86 == 15 && c->x86_cache_alignment == 64) - c->x86_cache_alignment = 128; -#endif -} - -#ifdef CONFIG_X86_32 -/* - * Early probe support logic for ppro memory erratum #50 - * - * This is called before we do cpu ident work - */ - -int __cpuinit ppro_with_ram_bug(void) -{ - /* Uses data from early_cpu_detect now */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && - boot_cpu_data.x86 == 6 && - boot_cpu_data.x86_model == 1 && - boot_cpu_data.x86_mask < 8) { - printk(KERN_INFO "Pentium Pro with Errata#50 detected. Taking evasive action.\n"); - return 1; - } - return 0; -} - - -/* - * P4 Xeon errata 037 workaround. - * Hardware prefetcher may cause stale data to be loaded into the cache. - */ -static void __cpuinit Intel_errata_workarounds(struct cpuinfo_x86 *c) -{ - unsigned long lo, hi; - - if ((c->x86 == 15) && (c->x86_model == 1) && (c->x86_mask == 1)) { - rdmsr(MSR_IA32_MISC_ENABLE, lo, hi); - if ((lo & (1<<9)) == 0) { - printk (KERN_INFO "CPU: C0 stepping P4 Xeon detected.\n"); - printk (KERN_INFO "CPU: Disabling hardware prefetching (Errata 037)\n"); - lo |= (1<<9); /* Disable hw prefetching */ - wrmsr (MSR_IA32_MISC_ENABLE, lo, hi); - } - } -} - - - -#ifdef CONFIG_X86_F00F_BUG -static void __cpuinit trap_init_f00f_bug(void) -{ - __set_fixmap(FIX_F00F_IDT, __pa(&idt_table), PAGE_KERNEL_RO); - - /* - * Update the IDT descriptor and reload the IDT so that - * it uses the read-only mapped virtual address. - */ - idt_descr.address = fix_to_virt(FIX_F00F_IDT); - load_idt(&idt_descr); -} -#endif -#endif - -static void __cpuinit srat_detect_node(void) -{ -#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64) - unsigned node; - int cpu = smp_processor_id(); - int apicid = hard_smp_processor_id(); - - /* Don't do the funky fallback heuristics the AMD version employs - for now. */ - node = apicid_to_node[apicid]; - if (node == NUMA_NO_NODE || !node_online(node)) - node = first_node(node_online_map); - numa_set_node(cpu, node); - - printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node); -#endif -} - -/* - * find out the number of processor cores on the die - */ -static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c) -{ - unsigned int eax, ebx, ecx, edx; - - if (c->cpuid_level < 4) - return 1; - - /* Intel has a non-standard dependency on %ecx for this CPUID level. */ - cpuid_count(4, 0, &eax, &ebx, &ecx, &edx); - if (eax & 0x1f) - return ((eax >> 26) + 1); - else - return 1; -} - -static void __cpuinit init_intel(struct cpuinfo_x86 *c) -{ - unsigned int l2 = 0; - char *p = NULL; - - early_init_intel(c); - -#ifdef CONFIG_X86_F00F_BUG - /* - * All current models of Pentium and Pentium with MMX technology CPUs - * have the F0 0F bug, which lets nonprivileged users lock up the system. - * Note that the workaround only should be initialized once... - */ - c->f00f_bug = 0; - if (!paravirt_enabled() && c->x86 == 5) { - static int f00f_workaround_enabled; - - c->f00f_bug = 1; - if (!f00f_workaround_enabled) { - trap_init_f00f_bug(); - printk(KERN_NOTICE "Intel Pentium with F0 0F bug - workaround enabled.\n"); - f00f_workaround_enabled = 1; - } - } -#endif - - l2 = init_intel_cacheinfo(c); - if (c->cpuid_level > 9) { - unsigned eax = cpuid_eax(10); - /* Check for version and the number of counters */ - if ((eax & 0xff) && (((eax>>8) & 0xff) > 1)) - set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON); - } - -#ifdef CONFIG_X86_32 - /* SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until model 3 mask 3 */ - if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633) - clear_cpu_cap(c, X86_FEATURE_SEP); - - /* - * Names for the Pentium II/Celeron processors - * detectable only by also checking the cache size. - * Dixon is NOT a Celeron. - */ - if (c->x86 == 6) { - switch (c->x86_model) { - case 5: - if (c->x86_mask == 0) { - if (l2 == 0) - p = "Celeron (Covington)"; - else if (l2 == 256) - p = "Mobile Pentium II (Dixon)"; - } - break; - - case 6: - if (l2 == 128) - p = "Celeron (Mendocino)"; - else if (c->x86_mask == 0 || c->x86_mask == 5) - p = "Celeron-A"; - break; - - case 8: - if (l2 == 128) - p = "Celeron (Coppermine)"; - break; - } - } - - if (p) - strcpy(c->x86_model_id, p); - - Intel_errata_workarounds(c); - -#ifdef CONFIG_X86_INTEL_USERCOPY - /* - * Set up the preferred alignment for movsl bulk memory moves - */ - switch (c->x86) { - case 4: /* 486: untested */ - break; - case 5: /* Old Pentia: untested */ - break; - case 6: /* PII/PIII only like movsl with 8-byte alignment */ - movsl_mask.mask = 7; - break; - case 15: /* P4 is OK down to 8-byte alignment */ - movsl_mask.mask = 7; - break; - } -#endif - -#endif - - if (cpu_has_xmm2) - set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); - if (cpu_has_ds) { - unsigned int l1; - rdmsr(MSR_IA32_MISC_ENABLE, l1, l2); - if (!(l1 & (1<<11))) - set_cpu_cap(c, X86_FEATURE_BTS); - if (!(l1 & (1<<12))) - set_cpu_cap(c, X86_FEATURE_PEBS); - ds_init_intel(c); - } - -#ifdef CONFIG_X86_64 - if (c->x86 == 15) - c->x86_cache_alignment = c->x86_clflush_size * 2; - if (c->x86 == 6) - set_cpu_cap(c, X86_FEATURE_REP_GOOD); -#else - if (c->x86 == 15) - set_cpu_cap(c, X86_FEATURE_P4); - if (c->x86 == 6) - set_cpu_cap(c, X86_FEATURE_P3); - - if (cpu_has_bts) - ptrace_bts_init_intel(c); - - /* - * See if we have a good local APIC by checking for buggy Pentia, - * i.e. all B steppings and the C2 stepping of P54C when using their - * integrated APIC (see 11AP erratum in "Pentium Processor - * Specification Update"). - */ - if (cpu_has_apic && (c->x86<<8 | c->x86_model<<4) == 0x520 && - (c->x86_mask < 0x6 || c->x86_mask == 0xb)) - set_cpu_cap(c, X86_FEATURE_11AP); - -#ifdef CONFIG_X86_NUMAQ - numaq_tsc_disable(); -#endif -#endif - - detect_extended_topology(c); - if (!cpu_has(c, X86_FEATURE_XTOPOLOGY)) { - /* - * let's use the legacy cpuid vector 0x1 and 0x4 for topology - * detection. - */ - c->x86_max_cores = intel_num_cpu_cores(c); -#ifdef CONFIG_X86_32 - detect_ht(c); -#endif - } - - /* Work around errata */ - srat_detect_node(); -} - -#ifdef CONFIG_X86_32 -static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 *c, unsigned int size) -{ - /* - * Intel PIII Tualatin. This comes in two flavours. - * One has 256kb of cache, the other 512. We have no way - * to determine which, so we use a boottime override - * for the 512kb model, and assume 256 otherwise. - */ - if ((c->x86 == 6) && (c->x86_model == 11) && (size == 0)) - size = 256; - return size; -} -#endif - -static struct cpu_dev intel_cpu_dev __cpuinitdata = { - .c_vendor = "Intel", - .c_ident = { "GenuineIntel" }, -#ifdef CONFIG_X86_32 - .c_models = { - { .vendor = X86_VENDOR_INTEL, .family = 4, .model_names = - { - [0] = "486 DX-25/33", - [1] = "486 DX-50", - [2] = "486 SX", - [3] = "486 DX/2", - [4] = "486 SL", - [5] = "486 SX/2", - [7] = "486 DX/2-WB", - [8] = "486 DX/4", - [9] = "486 DX/4-WB" - } - }, - { .vendor = X86_VENDOR_INTEL, .family = 5, .model_names = - { - [0] = "Pentium 60/66 A-step", - [1] = "Pentium 60/66", - [2] = "Pentium 75 - 200", - [3] = "OverDrive PODP5V83", - [4] = "Pentium MMX", - [7] = "Mobile Pentium 75 - 200", - [8] = "Mobile Pentium MMX" - } - }, - { .vendor = X86_VENDOR_INTEL, .family = 6, .model_names = - { - [0] = "Pentium Pro A-step", - [1] = "Pentium Pro", - [3] = "Pentium II (Klamath)", - [4] = "Pentium II (Deschutes)", - [5] = "Pentium II (Deschutes)", - [6] = "Mobile Pentium II", - [7] = "Pentium III (Katmai)", - [8] = "Pentium III (Coppermine)", - [10] = "Pentium III (Cascades)", - [11] = "Pentium III (Tualatin)", - } - }, - { .vendor = X86_VENDOR_INTEL, .family = 15, .model_names = - { - [0] = "Pentium 4 (Unknown)", - [1] = "Pentium 4 (Willamette)", - [2] = "Pentium 4 (Northwood)", - [4] = "Pentium 4 (Foster)", - [5] = "Pentium 4 (Foster)", - } - }, - }, - .c_size_cache = intel_size_cache, -#endif - .c_early_init = early_init_intel, - .c_init = init_intel, - .c_x86_vendor = X86_VENDOR_INTEL, -}; - -cpu_dev_register(intel_cpu_dev); - diff --git a/include/asm-x86/bugs.h b/include/asm-x86/bugs.h index ae514c76a96..dc604985f2a 100644 --- a/include/asm-x86/bugs.h +++ b/include/asm-x86/bugs.h @@ -3,7 +3,7 @@ extern void check_bugs(void); -#ifdef CONFIG_CPU_SUP_INTEL_32 +#if defined(CONFIG_CPU_SUP_INTEL) && defined(CONFIG_X86_32) int ppro_with_ram_bug(void); #else static inline int ppro_with_ram_bug(void) { return 0; } -- cgit v1.2.3-70-g09d2 From 91030ca1e739696812242c807b112ee3981a14be Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 9 Sep 2008 16:42:45 +0100 Subject: x86: unsigned long pte_pfn pte_pfn() has always been of type unsigned long, even on 32-bit PAE; but in the current tip/next/mm tree it works out to be unsigned long long on 64-bit, which gives an irritating warning if you try to printk a pfn with the usual %lx. Now use the same pte_pfn() function, moved from pgtable-3level.h to pgtable.h, for all models: as suggested by Jeremy Fitzhardinge. And pte_page() can well move along with it (remaining a macro to avoid dependence on mm_types.h). Signed-off-by: Hugh Dickins Acked-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- include/asm-x86/pgtable-2level.h | 2 -- include/asm-x86/pgtable-3level.h | 7 ------- include/asm-x86/pgtable.h | 7 +++++++ include/asm-x86/pgtable_64.h | 2 -- 4 files changed, 7 insertions(+), 11 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/pgtable-2level.h b/include/asm-x86/pgtable-2level.h index 46bc52c0eae..d8c55071ceb 100644 --- a/include/asm-x86/pgtable-2level.h +++ b/include/asm-x86/pgtable-2level.h @@ -53,9 +53,7 @@ static inline pte_t native_ptep_get_and_clear(pte_t *xp) #define native_ptep_get_and_clear(xp) native_local_ptep_get_and_clear(xp) #endif -#define pte_page(x) pfn_to_page(pte_pfn(x)) #define pte_none(x) (!(x).pte_low) -#define pte_pfn(x) (pte_val(x) >> PAGE_SHIFT) /* * Bits 0, 6 and 7 are taken, split up the 29 bits of offset diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h index 105057f3403..975da0dc142 100644 --- a/include/asm-x86/pgtable-3level.h +++ b/include/asm-x86/pgtable-3level.h @@ -151,18 +151,11 @@ static inline int pte_same(pte_t a, pte_t b) return a.pte_low == b.pte_low && a.pte_high == b.pte_high; } -#define pte_page(x) pfn_to_page(pte_pfn(x)) - static inline int pte_none(pte_t pte) { return !pte.pte_low && !pte.pte_high; } -static inline unsigned long pte_pfn(pte_t pte) -{ - return (pte_val(pte) & PTE_PFN_MASK) >> PAGE_SHIFT; -} - /* * Bits 0, 6 and 7 are taken in the low part of the pte, * put the 32 bits of offset into the high part. diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index 04caa2f544d..efc329b8a71 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -186,6 +186,13 @@ static inline int pte_special(pte_t pte) return pte_val(pte) & _PAGE_SPECIAL; } +static inline unsigned long pte_pfn(pte_t pte) +{ + return (pte_val(pte) & PTE_PFN_MASK) >> PAGE_SHIFT; +} + +#define pte_page(pte) pfn_to_page(pte_pfn(pte)) + static inline int pmd_large(pmd_t pte) { return (pmd_val(pte) & (_PAGE_PSE | _PAGE_PRESENT)) == diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h index 549144d03d9..e454e4ec016 100644 --- a/include/asm-x86/pgtable_64.h +++ b/include/asm-x86/pgtable_64.h @@ -175,8 +175,6 @@ static inline int pmd_bad(pmd_t pmd) #define pte_present(x) (pte_val((x)) & (_PAGE_PRESENT | _PAGE_PROTNONE)) #define pages_to_mb(x) ((x) >> (20 - PAGE_SHIFT)) /* FIXME: is this right? */ -#define pte_page(x) pfn_to_page(pte_pfn((x))) -#define pte_pfn(x) ((pte_val((x)) & __PHYSICAL_MASK) >> PAGE_SHIFT) /* * Macro to mark a page protection value as "uncacheable". -- cgit v1.2.3-70-g09d2 From 8a493d37f049b631e19584d1cb84cec88cbcf8fc Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Wed, 10 Sep 2008 00:49:47 +0900 Subject: x86: remove duplicated extern force_iommu Both iommu.h and dma-mapping.h have extern force_iommu. The latter doesn't need to do. Signed-off-by: FUJITA Tomonori Signed-off-by: Ingo Molnar --- include/asm-x86/dma-mapping.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index 56075320b81..f43420b8c57 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -15,7 +15,6 @@ extern dma_addr_t bad_dma_address; extern int iommu_merge; extern struct device x86_dma_fallback_dev; extern int panic_on_overflow; -extern int force_iommu; struct dma_mapping_ops { int (*mapping_error)(struct device *dev, -- cgit v1.2.3-70-g09d2 From 982162602b31041b426edec6480d327743abcbcc Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Wed, 10 Sep 2008 00:49:48 +0900 Subject: x86: convert dma_alloc_coherent to use is_device_dma_capable Signed-off-by: FUJITA Tomonori Signed-off-by: Ingo Molnar --- include/asm-x86/dma-mapping.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index f43420b8c57..f408e6dd177 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -278,7 +278,7 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp |= GFP_DMA; } - if (!dev->dma_mask) + if (!is_device_dma_capable(dev)) return NULL; if (!ops->alloc_coherent) -- cgit v1.2.3-70-g09d2 From 315a6558f30a264c88274fa70626615d1c7851c7 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Tue, 9 Sep 2008 14:54:53 +0800 Subject: x86: move VMX MSRs to msr-index.h They are hardware specific MSRs, and we would use them in virtualization feature detection later. Signed-off-by: Sheng Yang Signed-off-by: Ingo Molnar --- arch/x86/kvm/vmx.h | 15 --------------- include/asm-x86/msr-index.h | 16 ++++++++++++++++ 2 files changed, 16 insertions(+), 15 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/vmx.h b/arch/x86/kvm/vmx.h index 425a13436b3..b32d4e5b123 100644 --- a/arch/x86/kvm/vmx.h +++ b/arch/x86/kvm/vmx.h @@ -331,21 +331,6 @@ enum vmcs_field { #define AR_RESERVD_MASK 0xfffe0f00 -#define MSR_IA32_VMX_BASIC 0x480 -#define MSR_IA32_VMX_PINBASED_CTLS 0x481 -#define MSR_IA32_VMX_PROCBASED_CTLS 0x482 -#define MSR_IA32_VMX_EXIT_CTLS 0x483 -#define MSR_IA32_VMX_ENTRY_CTLS 0x484 -#define MSR_IA32_VMX_MISC 0x485 -#define MSR_IA32_VMX_CR0_FIXED0 0x486 -#define MSR_IA32_VMX_CR0_FIXED1 0x487 -#define MSR_IA32_VMX_CR4_FIXED0 0x488 -#define MSR_IA32_VMX_CR4_FIXED1 0x489 -#define MSR_IA32_VMX_VMCS_ENUM 0x48a -#define MSR_IA32_VMX_PROCBASED_CTLS2 0x48b -#define MSR_IA32_VMX_EPT_VPID_CAP 0x48c - -#define MSR_IA32_FEATURE_CONTROL 0x3a #define MSR_IA32_FEATURE_CONTROL_LOCKED 0x1 #define MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED 0x4 diff --git a/include/asm-x86/msr-index.h b/include/asm-x86/msr-index.h index 3052f058ab0..0bb43301a20 100644 --- a/include/asm-x86/msr-index.h +++ b/include/asm-x86/msr-index.h @@ -176,6 +176,7 @@ #define MSR_IA32_TSC 0x00000010 #define MSR_IA32_PLATFORM_ID 0x00000017 #define MSR_IA32_EBL_CR_POWERON 0x0000002a +#define MSR_IA32_FEATURE_CONTROL 0x0000003a #define MSR_IA32_APICBASE 0x0000001b #define MSR_IA32_APICBASE_BSP (1<<8) @@ -310,4 +311,19 @@ /* Geode defined MSRs */ #define MSR_GEODE_BUSCONT_CONF0 0x00001900 +/* Intel VT MSRs */ +#define MSR_IA32_VMX_BASIC 0x00000480 +#define MSR_IA32_VMX_PINBASED_CTLS 0x00000481 +#define MSR_IA32_VMX_PROCBASED_CTLS 0x00000482 +#define MSR_IA32_VMX_EXIT_CTLS 0x00000483 +#define MSR_IA32_VMX_ENTRY_CTLS 0x00000484 +#define MSR_IA32_VMX_MISC 0x00000485 +#define MSR_IA32_VMX_CR0_FIXED0 0x00000486 +#define MSR_IA32_VMX_CR0_FIXED1 0x00000487 +#define MSR_IA32_VMX_CR4_FIXED0 0x00000488 +#define MSR_IA32_VMX_CR4_FIXED1 0x00000489 +#define MSR_IA32_VMX_VMCS_ENUM 0x0000048a +#define MSR_IA32_VMX_PROCBASED_CTLS2 0x0000048b +#define MSR_IA32_VMX_EPT_VPID_CAP 0x0000048c + #endif /* ASM_X86__MSR_INDEX_H */ -- cgit v1.2.3-70-g09d2 From e38e05a85828dac23540cd007df5f20985388afc Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Wed, 10 Sep 2008 18:53:34 +0800 Subject: x86: extended "flags" to show virtualization HW feature in /proc/cpuinfo The hardware virtualization technology evolves very fast. But currently it's hard to tell if your CPU support a certain kind of HW technology without digging into the source code. The patch add a new catagory in "flags" under /proc/cpuinfo. Now "flags" can indicate the (important) HW virtulization features the CPU supported as well. Current implementation just cover Intel VMX side. Signed-off-by: Sheng Yang Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/intel.c | 41 +++++++++++++++++++++++++++++++++++++++++ include/asm-x86/cpufeature.h | 9 ++++++++- 2 files changed, 49 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 5f76bf139fd..99468dbd08d 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -196,6 +196,44 @@ static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c) return 1; } +static void __cpuinit detect_vmx_virtcap(struct cpuinfo_x86 *c) +{ + /* Intel VMX MSR indicated features */ +#define X86_VMX_FEATURE_PROC_CTLS_TPR_SHADOW 0x00200000 +#define X86_VMX_FEATURE_PROC_CTLS_VNMI 0x00400000 +#define X86_VMX_FEATURE_PROC_CTLS_2ND_CTLS 0x80000000 +#define X86_VMX_FEATURE_PROC_CTLS2_VIRT_APIC 0x00000001 +#define X86_VMX_FEATURE_PROC_CTLS2_EPT 0x00000002 +#define X86_VMX_FEATURE_PROC_CTLS2_VPID 0x00000020 + + u32 vmx_msr_low, vmx_msr_high, msr_ctl, msr_ctl2; + + clear_cpu_cap(c, X86_FEATURE_TPR_SHADOW); + clear_cpu_cap(c, X86_FEATURE_VNMI); + clear_cpu_cap(c, X86_FEATURE_FLEXPRIORITY); + clear_cpu_cap(c, X86_FEATURE_EPT); + clear_cpu_cap(c, X86_FEATURE_VPID); + + rdmsr(MSR_IA32_VMX_PROCBASED_CTLS, vmx_msr_low, vmx_msr_high); + msr_ctl = vmx_msr_high | vmx_msr_low; + if (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_TPR_SHADOW) + set_cpu_cap(c, X86_FEATURE_TPR_SHADOW); + if (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_VNMI) + set_cpu_cap(c, X86_FEATURE_VNMI); + if (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_2ND_CTLS) { + rdmsr(MSR_IA32_VMX_PROCBASED_CTLS2, + vmx_msr_low, vmx_msr_high); + msr_ctl2 = vmx_msr_high | vmx_msr_low; + if ((msr_ctl2 & X86_VMX_FEATURE_PROC_CTLS2_VIRT_APIC) && + (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_TPR_SHADOW)) + set_cpu_cap(c, X86_FEATURE_FLEXPRIORITY); + if (msr_ctl2 & X86_VMX_FEATURE_PROC_CTLS2_EPT) + set_cpu_cap(c, X86_FEATURE_EPT); + if (msr_ctl2 & X86_VMX_FEATURE_PROC_CTLS2_VPID) + set_cpu_cap(c, X86_FEATURE_VPID); + } +} + static void __cpuinit init_intel(struct cpuinfo_x86 *c) { unsigned int l2 = 0; @@ -289,6 +327,9 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) /* Work around errata */ srat_detect_node(); + + if (cpu_has(c, X86_FEATURE_VMX)) + detect_vmx_virtcap(c); } #ifdef CONFIG_X86_32 diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 7ac4d93d20e..8d45690bef5 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -6,7 +6,7 @@ #include -#define NCAPINTS 8 /* N 32-bit words worth of info */ +#define NCAPINTS 9 /* N 32-bit words worth of info */ /* * Note: If the comment begins with a quoted string, that string is used @@ -150,6 +150,13 @@ */ #define X86_FEATURE_IDA (7*32+ 0) /* Intel Dynamic Acceleration */ +/* Virtualization flags: Linux defined */ +#define X86_FEATURE_TPR_SHADOW (8*32+ 0) /* Intel TPR Shadow */ +#define X86_FEATURE_VNMI (8*32+ 1) /* Intel Virtual NMI */ +#define X86_FEATURE_FLEXPRIORITY (8*32+ 2) /* Intel FlexPriority */ +#define X86_FEATURE_EPT (8*32+ 3) /* Intel Extended Page Table */ +#define X86_FEATURE_VPID (8*32+ 4) /* Intel Virtual Processor ID */ + #if defined(__KERNEL__) && !defined(__ASSEMBLY__) #include -- cgit v1.2.3-70-g09d2 From a0a29b62a9cac6b7d83b7514679f2ed8d33d4372 Mon Sep 17 00:00:00 2001 From: Dmitry Adamushko Date: Thu, 11 Sep 2008 23:27:52 +0200 Subject: x86, microcode rework, v2 this is a rework of the microcode splitup in tip/x86/microcode (1) I think this new interface is cleaner (look at the changes in 'struct microcode_ops' in microcode.h); (2) it's -64 lines of code; Signed-off-by: Ingo Molnar --- arch/x86/kernel/microcode.c | 84 ++++------ arch/x86/kernel/microcode_amd.c | 329 ++++++++++++++++++-------------------- arch/x86/kernel/microcode_intel.c | 220 +++++++++++-------------- include/asm-x86/microcode.h | 17 +- 4 files changed, 293 insertions(+), 357 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c index b2f84ce5eed..902dada2eb6 100644 --- a/arch/x86/kernel/microcode.c +++ b/arch/x86/kernel/microcode.c @@ -110,50 +110,28 @@ struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; EXPORT_SYMBOL_GPL(ucode_cpu_info); #ifdef CONFIG_MICROCODE_OLD_INTERFACE -void __user *user_buffer; /* user area microcode data buffer */ -EXPORT_SYMBOL_GPL(user_buffer); -unsigned int user_buffer_size; /* it's size */ -EXPORT_SYMBOL_GPL(user_buffer_size); - -static int do_microcode_update(void) +static int do_microcode_update(const void __user *buf, size_t size) { - long cursor = 0; + cpumask_t old; int error = 0; - void *new_mc = NULL; int cpu; - cpumask_t old; old = current->cpus_allowed; - while ((cursor = microcode_ops->get_next_ucode(&new_mc, cursor)) > 0) { - if (microcode_ops->microcode_sanity_check != NULL) - error = microcode_ops->microcode_sanity_check(new_mc); - if (error) + for_each_online_cpu(cpu) { + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + + if (!uci->valid) + continue; + + set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); + error = microcode_ops->request_microcode_user(cpu, buf, size); + if (error < 0) goto out; - /* - * It's possible the data file has multiple matching ucode, - * lets keep searching till the latest version - */ - for_each_online_cpu(cpu) { - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - - if (!uci->valid) - continue; - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - error = microcode_ops->get_matching_microcode(new_mc, - cpu); - if (error < 0) - goto out; - if (error == 1) - microcode_ops->apply_microcode(cpu); - } - vfree(new_mc); + if (!error) + microcode_ops->apply_microcode(cpu); } out: - if (cursor > 0) - vfree(new_mc); - if (cursor < 0) - error = cursor; set_cpus_allowed_ptr(current, &old); return error; } @@ -178,10 +156,7 @@ static ssize_t microcode_write(struct file *file, const char __user *buf, get_online_cpus(); mutex_lock(µcode_mutex); - user_buffer = (void __user *) buf; - user_buffer_size = (int) len; - - ret = do_microcode_update(); + ret = do_microcode_update(buf, len); if (!ret) ret = (ssize_t)len; @@ -231,7 +206,6 @@ MODULE_ALIAS_MISCDEV(MICROCODE_MINOR); /* fake device for request_firmware */ struct platform_device *microcode_pdev; -EXPORT_SYMBOL_GPL(microcode_pdev); static ssize_t reload_store(struct sys_device *dev, struct sysdev_attribute *attr, @@ -252,8 +226,12 @@ static ssize_t reload_store(struct sys_device *dev, if (cpu_online(cpu)) { set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); mutex_lock(µcode_mutex); - if (uci->valid) - err = microcode_ops->cpu_request_microcode(cpu); + if (uci->valid) { + err = microcode_ops->request_microcode_fw(cpu, + µcode_pdev->dev); + if (!err) + microcode_ops->apply_microcode(cpu); + } mutex_unlock(µcode_mutex); set_cpus_allowed_ptr(current, &old); } @@ -315,7 +293,7 @@ static void collect_cpu_info(int cpu) uci->valid = 1; } -static void microcode_resume_cpu(int cpu) +static int microcode_resume_cpu(int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; struct cpu_signature nsig; @@ -323,7 +301,7 @@ static void microcode_resume_cpu(int cpu) pr_debug("microcode: CPU%d resumed\n", cpu); if (!uci->mc.valid_mc) - return; + return 1; /* * Let's verify that the 'cached' ucode does belong @@ -331,21 +309,22 @@ static void microcode_resume_cpu(int cpu) */ if (microcode_ops->collect_cpu_info(cpu, &nsig)) { microcode_fini_cpu(cpu); - return; + return -1; } if (memcmp(&nsig, &uci->cpu_sig, sizeof(nsig))) { microcode_fini_cpu(cpu); /* Should we look for a new ucode here? */ - return; + return 1; } - microcode_ops->apply_microcode(cpu); + return 0; } void microcode_update_cpu(int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + int err = 0; /* We should bind the task to the CPU */ BUG_ON(raw_smp_processor_id() != cpu); @@ -356,12 +335,17 @@ void microcode_update_cpu(int cpu) * otherwise just request a firmware: */ if (uci->valid) { - microcode_resume_cpu(cpu); + err = microcode_resume_cpu(cpu); } else { collect_cpu_info(cpu); if (uci->valid && system_state == SYSTEM_RUNNING) - microcode_ops->cpu_request_microcode(cpu); + err = microcode_ops->request_microcode_fw(cpu, + µcode_pdev->dev); } + + if (!err) + microcode_ops->apply_microcode(cpu); + mutex_unlock(µcode_mutex); } @@ -414,7 +398,7 @@ static int mc_sysdev_resume(struct sys_device *dev) return 0; pr_debug("microcode: CPU%d resumed\n", cpu); /* only CPU 0 will apply ucode here */ - microcode_ops->apply_microcode(0); + microcode_update_cpu(0); return 0; } diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c index d606a05545c..6815837a775 100644 --- a/arch/x86/kernel/microcode_amd.c +++ b/arch/x86/kernel/microcode_amd.c @@ -59,7 +59,7 @@ MODULE_LICENSE("GPL v2"); /* serialize access to the physical write */ static DEFINE_SPINLOCK(microcode_update_lock); -struct equiv_cpu_entry *equiv_cpu_table; +static struct equiv_cpu_entry *equiv_cpu_table; static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig) { @@ -83,36 +83,37 @@ static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig) return 0; } -static int get_matching_microcode_amd(void *mc, int cpu) +static int get_matching_microcode(int cpu, void *mc, int rev) { - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; struct microcode_header_amd *mc_header = mc; - unsigned long total_size = get_totalsize(mc_header); - void *new_mc; struct pci_dev *nb_pci_dev, *sb_pci_dev; unsigned int current_cpu_id; unsigned int equiv_cpu_id = 0x00; unsigned int i = 0; - /* We should bind the task to the CPU */ - BUG_ON(cpu != raw_smp_processor_id()); - - /* This is a tricky part. We might be called from a write operation */ - /* to the device file instead of the usual process of firmware */ - /* loading. This routine needs to be able to distinguish both */ -/* cases. This is done by checking if there alread is a equivalent */ - /* CPU table installed. If not, we're written through */ - /* /dev/cpu/microcode. */ -/* Since we ignore all checks. The error case in which going through */ -/* firmware loading and that table is not loaded has already been */ - /* checked earlier. */ + /* + * dimm: do we need this? Why an update via /dev/... is different + * from the one via firmware? + * + * This is a tricky part. We might be called from a write operation + * to the device file instead of the usual process of firmware + * loading. This routine needs to be able to distinguish both + * cases. This is done by checking if there alread is a equivalent + * CPU table installed. If not, we're written through + * /dev/cpu/microcode. + * Since we ignore all checks. The error case in which going through + * firmware loading and that table is not loaded has already been + * checked earlier. + */ + BUG_ON(equiv_cpu_table == NULL); +#if 0 if (equiv_cpu_table == NULL) { printk(KERN_INFO "microcode: CPU%d microcode update with " "version 0x%x (current=0x%x)\n", cpu, mc_header->patch_id, uci->cpu_sig.rev); goto out; } - +#endif current_cpu_id = cpuid_eax(0x00000001); while (equiv_cpu_table[i].installed_cpu != 0) { @@ -175,27 +176,9 @@ static int get_matching_microcode_amd(void *mc, int cpu) pci_dev_put(sb_pci_dev); } - if (mc_header->patch_id <= uci->cpu_sig.rev) + if (mc_header->patch_id <= rev) return 0; - printk(KERN_INFO "microcode: CPU%d found a matching microcode " - "update with version 0x%x (current=0x%x)\n", - cpu, mc_header->patch_id, uci->cpu_sig.rev); - -out: - new_mc = vmalloc(UCODE_MAX_SIZE); - if (!new_mc) { - printk(KERN_ERR "microcode: error, can't allocate memory\n"); - return -ENOMEM; - } - memset(new_mc, 0, UCODE_MAX_SIZE); - - /* free previous update file */ - vfree(uci->mc.mc_amd); - - memcpy(new_mc, mc, total_size); - - uci->mc.mc_amd = new_mc; return 1; } @@ -245,104 +228,65 @@ static void apply_microcode_amd(int cpu) uci->cpu_sig.rev = rev; } -#ifdef CONFIG_MICROCODE_OLD_INTERFACE -extern void __user *user_buffer; /* user area microcode data buffer */ -extern unsigned int user_buffer_size; /* it's size */ - -static long get_next_ucode_amd(void **mc, long offset) -{ - struct microcode_header_amd mc_header; - unsigned long total_size; - - /* No more data */ - if (offset >= user_buffer_size) - return 0; - if (copy_from_user(&mc_header, user_buffer + offset, MC_HEADER_SIZE)) { - printk(KERN_ERR "microcode: error! Can not read user data\n"); - return -EFAULT; - } - total_size = get_totalsize(&mc_header); - if (offset + total_size > user_buffer_size) { - printk(KERN_ERR "microcode: error! Bad total size in microcode " - "data file\n"); - return -EINVAL; - } - *mc = vmalloc(UCODE_MAX_SIZE); - if (!*mc) - return -ENOMEM; - memset(*mc, 0, UCODE_MAX_SIZE); - - if (copy_from_user(*mc, user_buffer + offset, total_size)) { - printk(KERN_ERR "microcode: error! Can not read user data\n"); - vfree(*mc); - return -EFAULT; - } - return offset + total_size; -} -#else -#define get_next_ucode_amd() NULL -#endif - -static long get_next_ucode_from_buffer_amd(void **mc, void *buf, - unsigned long size, long offset) +static void * get_next_ucode(u8 *buf, unsigned int size, + int (*get_ucode_data)(void *, const void *, size_t), + unsigned int *mc_size) { - struct microcode_header_amd *mc_header; - unsigned long total_size; - unsigned char *buf_pos = buf; + unsigned int total_size; +#define UCODE_UNKNOWN_HDR 8 + u8 hdr[UCODE_UNKNOWN_HDR]; + void *mc; - /* No more data */ - if (offset >= size) - return 0; + if (get_ucode_data(hdr, buf, UCODE_UNKNOWN_HDR)) + return NULL; - if (buf_pos[offset] != UCODE_UCODE_TYPE) { + if (hdr[0] != UCODE_UCODE_TYPE) { printk(KERN_ERR "microcode: error! " "Wrong microcode payload type field\n"); - return -EINVAL; + return NULL; } - mc_header = (struct microcode_header_amd *)(&buf_pos[offset+8]); + /* Why not by means of get_totalsize(hdr)? */ + total_size = (unsigned long) (hdr[4] + (hdr[5] << 8)); - total_size = (unsigned long) (buf_pos[offset+4] + - (buf_pos[offset+5] << 8)); + printk(KERN_INFO "microcode: size %u, total_size %u\n", + size, total_size); - printk(KERN_INFO "microcode: size %lu, total_size %lu, offset %ld\n", - size, total_size, offset); - - if (offset + total_size > size) { + if (total_size > size || total_size > UCODE_MAX_SIZE) { printk(KERN_ERR "microcode: error! Bad data in microcode data file\n"); - return -EINVAL; + return NULL; } - *mc = vmalloc(UCODE_MAX_SIZE); - if (!*mc) { - printk(KERN_ERR "microcode: error! " - "Can not allocate memory for microcode patch\n"); - return -ENOMEM; + mc = vmalloc(UCODE_MAX_SIZE); + if (mc) { + memset(mc, 0, UCODE_MAX_SIZE); + if (get_ucode_data(mc, buf + UCODE_UNKNOWN_HDR, total_size)) { + vfree(mc); + mc = NULL; + } else + *mc_size = total_size + UCODE_UNKNOWN_HDR; } - - memset(*mc, 0, UCODE_MAX_SIZE); - memcpy(*mc, buf + offset + 8, total_size); - - return offset + total_size + 8; +#undef UCODE_UNKNOWN_HDR + return mc; } -static long install_equiv_cpu_table(void *buf, unsigned long size, long offset) + +static int install_equiv_cpu_table(u8 *buf, + int (*get_ucode_data)(void *, const void *, size_t)) { - unsigned int *buf_pos = buf; +#define UCODE_HEADER_SIZE 12 + u8 *hdr[UCODE_HEADER_SIZE]; + unsigned int *buf_pos = (unsigned int *)hdr; + unsigned long size; - /* No more data */ - if (offset >= size) + if (get_ucode_data(&hdr, buf, UCODE_HEADER_SIZE)) return 0; - if (buf_pos[1] != UCODE_EQUIV_CPU_TABLE_TYPE) { - printk(KERN_ERR "microcode: error! " - "Wrong microcode equivalnet cpu table type field\n"); - return 0; - } + size = buf_pos[2]; - if (size == 0) { + if (buf_pos[1] != UCODE_EQUIV_CPU_TABLE_TYPE || !size) { printk(KERN_ERR "microcode: error! " - "Wrong microcode equivalnet cpu table length\n"); + "Wrong microcode equivalnet cpu table\n"); return 0; } @@ -352,79 +296,118 @@ static long install_equiv_cpu_table(void *buf, unsigned long size, long offset) return 0; } - memset(equiv_cpu_table, 0, size); - memcpy(equiv_cpu_table, &buf_pos[3], size); + buf += UCODE_HEADER_SIZE; + if (get_ucode_data(equiv_cpu_table, buf, size)) { + vfree(equiv_cpu_table); + return 0; + } - return size + 12; /* add header length */ + return size + UCODE_HEADER_SIZE; /* add header length */ +#undef UCODE_HEADER_SIZE } -/* fake device for request_firmware */ -extern struct platform_device *microcode_pdev; - -static int cpu_request_microcode_amd(int cpu) +static void free_equiv_cpu_table(void) { - char name[30]; - const struct firmware *firmware; - void *buf; - unsigned int *buf_pos; - unsigned long size; - long offset = 0; - int error; - void *mc; - - /* We should bind the task to the CPU */ - BUG_ON(cpu != raw_smp_processor_id()); - - sprintf(name, "amd-ucode/microcode_amd.bin"); - error = request_firmware(&firmware, "amd-ucode/microcode_amd.bin", - µcode_pdev->dev); - if (error) { - printk(KERN_ERR "microcode: ucode data file %s load failed\n", - name); - return error; - } - - buf_pos = (unsigned int *)firmware->data; - buf = (void *)firmware->data; - size = firmware->size; - - if (buf_pos[0] != UCODE_MAGIC) { - printk(KERN_ERR "microcode: error! Wrong microcode patch file magic\n"); - return -EINVAL; + if (equiv_cpu_table) { + vfree(equiv_cpu_table); + equiv_cpu_table = NULL; } +} - offset = install_equiv_cpu_table(buf, buf_pos[2], offset); +static int generic_load_microcode(int cpu, void *data, size_t size, + int (*get_ucode_data)(void *, const void *, size_t)) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + u8 *ucode_ptr = data, *new_mc = NULL, *mc; + int new_rev = uci->cpu_sig.rev; + unsigned int leftover; + unsigned long offset; + offset = install_equiv_cpu_table(ucode_ptr, get_ucode_data); if (!offset) { printk(KERN_ERR "microcode: installing equivalent cpu table failed\n"); return -EINVAL; } - while ((offset = - get_next_ucode_from_buffer_amd(&mc, buf, size, offset)) > 0) { - error = get_matching_microcode_amd(mc, cpu); - if (error < 0) + ucode_ptr += offset; + leftover = size - offset; + + while (leftover) { + unsigned int mc_size; + struct microcode_header_amd *mc_header; + + mc = get_next_ucode(ucode_ptr, leftover, get_ucode_data, &mc_size); + if (!mc) break; - /* - * It's possible the data file has multiple matching ucode, - * lets keep searching till the latest version - */ - if (error == 1) { - apply_microcode_amd(cpu); - error = 0; - } - vfree(mc); + + mc_header = (struct microcode_header_amd *)mc; + if (get_matching_microcode(cpu, mc, new_rev)) { + new_rev = mc_header->patch_id; + new_mc = mc; + } else + vfree(mc); + + ucode_ptr += mc_size; + leftover -= mc_size; } - if (offset > 0) { - vfree(mc); - vfree(equiv_cpu_table); - equiv_cpu_table = NULL; + + if (new_mc) { + if (!leftover) { + if (uci->mc.mc_amd) + vfree(uci->mc.mc_amd); + uci->mc.mc_amd = (struct microcode_amd *)new_mc; + pr_debug("microcode: CPU%d found a matching microcode update with" + " version 0x%x (current=0x%x)\n", + cpu, uci->mc.mc_amd->hdr.patch_id, uci->cpu_sig.rev); + } else + vfree(new_mc); } - if (offset < 0) - error = offset; + + free_equiv_cpu_table(); + + return (int)leftover; +} + +static int get_ucode_fw(void *to, const void *from, size_t n) +{ + memcpy(to, from, n); + return 0; +} + +static int request_microcode_fw(int cpu, struct device *device) +{ + const char *fw_name = "amd-ucode/microcode_amd.bin"; + const struct firmware *firmware; + int ret; + + /* We should bind the task to the CPU */ + BUG_ON(cpu != raw_smp_processor_id()); + + ret = request_firmware(&firmware, fw_name, device); + if (ret) { + printk(KERN_ERR "microcode: ucode data file %s load failed\n", fw_name); + return ret; + } + + ret = generic_load_microcode(cpu, (void*)firmware->data, firmware->size, + &get_ucode_fw); + release_firmware(firmware); - return error; + return ret; +} + +static int get_ucode_user(void *to, const void *from, size_t n) +{ + return copy_from_user(to, from, n); +} + +static int request_microcode_user(int cpu, const void __user *buf, size_t size) +{ + /* We should bind the task to the CPU */ + BUG_ON(cpu != raw_smp_processor_id()); + + return generic_load_microcode(cpu, (void*)buf, size, &get_ucode_user); } static void microcode_fini_cpu_amd(int cpu) @@ -436,10 +419,8 @@ static void microcode_fini_cpu_amd(int cpu) } static struct microcode_ops microcode_amd_ops = { - .get_next_ucode = get_next_ucode_amd, - .get_matching_microcode = get_matching_microcode_amd, - .microcode_sanity_check = NULL, - .cpu_request_microcode = cpu_request_microcode_amd, + .request_microcode_user = request_microcode_user, + .request_microcode_fw = request_microcode_fw, .collect_cpu_info = collect_cpu_info_amd, .apply_microcode = apply_microcode_amd, .microcode_fini_cpu = microcode_fini_cpu_amd, diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index c9b53202ba3..f4930b55c6a 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -155,15 +155,15 @@ static int collect_cpu_info(int cpu_num, struct cpu_signature *csig) return 0; } -static inline int microcode_update_match(int cpu_num, - struct microcode_header_intel *mc_header, int sig, int pf) +static inline int update_match_cpu(struct cpu_signature *csig, int sig, int pf) { - struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; + return (!sigmatch(sig, csig->sig, pf, csig->pf)) ? 0 : 1; +} - if (!sigmatch(sig, uci->cpu_sig.sig, pf, uci->cpu_sig.pf) - || mc_header->rev <= uci->cpu_sig.rev) - return 0; - return 1; +static inline int +update_match_revision(struct microcode_header_intel *mc_header, int rev) +{ + return (mc_header->rev <= rev) ? 0 : 1; } static int microcode_sanity_check(void *mc) @@ -248,51 +248,36 @@ static int microcode_sanity_check(void *mc) /* * return 0 - no update found * return 1 - found update - * return < 0 - error */ -static int get_matching_microcode(void *mc, int cpu) +static int +get_matching_microcode(struct cpu_signature *cpu_sig, void *mc, int rev) { - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; struct microcode_header_intel *mc_header = mc; struct extended_sigtable *ext_header; unsigned long total_size = get_totalsize(mc_header); int ext_sigcount, i; struct extended_signature *ext_sig; - void *new_mc; - if (microcode_update_match(cpu, mc_header, - mc_header->sig, mc_header->pf)) - goto find; + if (!update_match_revision(mc_header, rev)) + return 0; + + if (update_match_cpu(cpu_sig, mc_header->sig, mc_header->pf)) + return 1; + /* Look for ext. headers: */ if (total_size <= get_datasize(mc_header) + MC_HEADER_SIZE) return 0; ext_header = mc + get_datasize(mc_header) + MC_HEADER_SIZE; ext_sigcount = ext_header->count; ext_sig = (void *)ext_header + EXT_HEADER_SIZE; + for (i = 0; i < ext_sigcount; i++) { - if (microcode_update_match(cpu, mc_header, - ext_sig->sig, ext_sig->pf)) - goto find; + if (update_match_cpu(cpu_sig, ext_sig->sig, ext_sig->pf)) + return 1; ext_sig++; } return 0; -find: - pr_debug("microcode: CPU%d found a matching microcode update with" - " version 0x%x (current=0x%x)\n", - cpu, mc_header->rev, uci->cpu_sig.rev); - new_mc = vmalloc(total_size); - if (!new_mc) { - printk(KERN_ERR "microcode: error! Can not allocate memory\n"); - return -ENOMEM; - } - - /* free previous update file */ - vfree(uci->mc.mc_intel); - - memcpy(new_mc, mc, total_size); - uci->mc.mc_intel = new_mc; - return 1; } static void apply_microcode(int cpu) @@ -300,7 +285,7 @@ static void apply_microcode(int cpu) unsigned long flags; unsigned int val[2]; int cpu_num = raw_smp_processor_id(); - struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; /* We should bind the task to the CPU */ BUG_ON(cpu_num != cpu); @@ -338,116 +323,105 @@ static void apply_microcode(int cpu) uci->cpu_sig.rev = val[1]; } -#ifdef CONFIG_MICROCODE_OLD_INTERFACE -extern void __user *user_buffer; /* user area microcode data buffer */ -extern unsigned int user_buffer_size; /* it's size */ - -static long get_next_ucode(void **mc, long offset) +static int generic_load_microcode(int cpu, void *data, size_t size, + int (*get_ucode_data)(void *, const void *, size_t)) { - struct microcode_header_intel mc_header; - unsigned long total_size; + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + u8 *ucode_ptr = data, *new_mc = NULL, *mc; + int new_rev = uci->cpu_sig.rev; + unsigned int leftover = size; - /* No more data */ - if (offset >= user_buffer_size) - return 0; - if (copy_from_user(&mc_header, user_buffer + offset, MC_HEADER_SIZE)) { - printk(KERN_ERR "microcode: error! Can not read user data\n"); - return -EFAULT; - } - total_size = get_totalsize(&mc_header); - if (offset + total_size > user_buffer_size) { - printk(KERN_ERR "microcode: error! Bad total size in microcode " - "data file\n"); - return -EINVAL; - } - *mc = vmalloc(total_size); - if (!*mc) - return -ENOMEM; - if (copy_from_user(*mc, user_buffer + offset, total_size)) { - printk(KERN_ERR "microcode: error! Can not read user data\n"); - vfree(*mc); - return -EFAULT; - } - return offset + total_size; -} -#endif + while (leftover) { + struct microcode_header_intel mc_header; + unsigned int mc_size; -static long get_next_ucode_from_buffer(void **mc, const u8 *buf, - unsigned long size, long offset) -{ - struct microcode_header_intel *mc_header; - unsigned long total_size; + if (get_ucode_data(&mc_header, ucode_ptr, sizeof(mc_header))) + break; - /* No more data */ - if (offset >= size) - return 0; - mc_header = (struct microcode_header_intel *)(buf + offset); - total_size = get_totalsize(mc_header); + mc_size = get_totalsize(&mc_header); + if (!mc_size || mc_size > leftover) { + printk(KERN_ERR "microcode: error!" + "Bad data in microcode data file\n"); + break; + } - if (offset + total_size > size) { - printk(KERN_ERR "microcode: error! Bad data in microcode data file\n"); - return -EINVAL; + mc = vmalloc(mc_size); + if (!mc) + break; + + if (get_ucode_data(mc, ucode_ptr, mc_size) || + microcode_sanity_check(mc) < 0) { + vfree(mc); + break; + } + + if (get_matching_microcode(&uci->cpu_sig, mc, new_rev)) { + new_rev = mc_header.rev; + new_mc = mc; + } else + vfree(mc); + + ucode_ptr += mc_size; + leftover -= mc_size; } - *mc = vmalloc(total_size); - if (!*mc) { - printk(KERN_ERR "microcode: error! Can not allocate memory\n"); - return -ENOMEM; + if (new_mc) { + if (!leftover) { + if (uci->mc.mc_intel) + vfree(uci->mc.mc_intel); + uci->mc.mc_intel = (struct microcode_intel *)new_mc; + pr_debug("microcode: CPU%d found a matching microcode update with" + " version 0x%x (current=0x%x)\n", + cpu, uci->mc.mc_intel->hdr.rev, uci->cpu_sig.rev); + } else + vfree(new_mc); } - memcpy(*mc, buf + offset, total_size); - return offset + total_size; + + return (int)leftover; } -/* fake device for request_firmware */ -extern struct platform_device *microcode_pdev; +static int get_ucode_fw(void *to, const void *from, size_t n) +{ + memcpy(to, from, n); + return 0; +} -static int cpu_request_microcode(int cpu) +static int request_microcode_fw(int cpu, struct device *device) { char name[30]; struct cpuinfo_x86 *c = &cpu_data(cpu); const struct firmware *firmware; - const u8 *buf; - unsigned long size; - long offset = 0; - int error; - void *mc; + int ret; /* We should bind the task to the CPU */ BUG_ON(cpu != raw_smp_processor_id()); sprintf(name, "intel-ucode/%02x-%02x-%02x", c->x86, c->x86_model, c->x86_mask); - error = request_firmware(&firmware, name, µcode_pdev->dev); - if (error) { + ret = request_firmware(&firmware, name, device); + if (ret) { pr_debug("microcode: data file %s load failed\n", name); - return error; - } - buf = firmware->data; - size = firmware->size; - while ((offset = get_next_ucode_from_buffer(&mc, buf, size, offset)) - > 0) { - error = microcode_sanity_check(mc); - if (error) - break; - error = get_matching_microcode(mc, cpu); - if (error < 0) - break; - /* - * It's possible the data file has multiple matching ucode, - * lets keep searching till the latest version - */ - if (error == 1) { - apply_microcode(cpu); - error = 0; - } - vfree(mc); + return ret; } - if (offset > 0) - vfree(mc); - if (offset < 0) - error = offset; + + ret = generic_load_microcode(cpu, (void*)firmware->data, firmware->size, + &get_ucode_fw); + release_firmware(firmware); - return error; + return ret; +} + +static int get_ucode_user(void *to, const void *from, size_t n) +{ + return copy_from_user(to, from, n); +} + +static int request_microcode_user(int cpu, const void __user *buf, size_t size) +{ + /* We should bind the task to the CPU */ + BUG_ON(cpu != raw_smp_processor_id()); + + return generic_load_microcode(cpu, (void*)buf, size, &get_ucode_user); } static void microcode_fini_cpu(int cpu) @@ -459,10 +433,8 @@ static void microcode_fini_cpu(int cpu) } static struct microcode_ops microcode_intel_ops = { - .get_next_ucode = get_next_ucode, - .get_matching_microcode = get_matching_microcode, - .microcode_sanity_check = microcode_sanity_check, - .cpu_request_microcode = cpu_request_microcode, + .request_microcode_user = request_microcode_user, + .request_microcode_fw = request_microcode_fw, .collect_cpu_info = collect_cpu_info, .apply_microcode = apply_microcode, .microcode_fini_cpu = microcode_fini_cpu, diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index 7ceff48fa65..e2887facdb4 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -5,17 +5,16 @@ extern int microcode_init(void *opaque, struct module *module); extern void microcode_exit(void); struct cpu_signature; +struct device; struct microcode_ops { - long (*get_next_ucode)(void **mc, long offset); - long (*microcode_get_next_ucode)(void **mc, long offset); - int (*get_matching_microcode)(void *mc, int cpu); - int (*microcode_sanity_check)(void *mc); - int (*cpu_request_microcode)(int cpu); - int (*collect_cpu_info)(int cpu_num, struct cpu_signature *csig); - void (*apply_microcode)(int cpu); - void (*microcode_fini_cpu)(int cpu); - void (*clear_patch)(void *data); + int (*request_microcode_user) (int cpu, const void __user *buf, size_t size); + int (*request_microcode_fw) (int cpu, struct device *device); + + void (*apply_microcode) (int cpu); + + int (*collect_cpu_info) (int cpu, struct cpu_signature *csig); + void (*microcode_fini_cpu) (int cpu); }; struct microcode_header_intel { -- cgit v1.2.3-70-g09d2 From 600715dcdf567c86f8b2c6173fcfb4b873e25a19 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 11 Sep 2008 01:31:45 -0700 Subject: generic: add phys_addr_t for holding physical addresses Add a kernel-wide "phys_addr_t" which is guaranteed to be able to hold any physical address. By default it equals the word size of the architecture, but a 32-bit architecture can set ARCH_PHYS_ADDR_T_64BIT if it needs a 64-bit phys_addr_t. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/powerpc/Kconfig | 3 +++ arch/powerpc/include/asm/types.h | 7 ------- arch/x86/Kconfig | 3 +++ include/asm-x86/page_32.h | 2 -- include/asm-x86/page_64.h | 1 - include/linux/types.h | 6 ++++++ mm/Kconfig | 3 +++ 7 files changed, 15 insertions(+), 10 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 587da5e0990..f5f83ee6041 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -22,6 +22,9 @@ config WORD_SIZE config PPC_MERGE def_bool y +config ARCH_PHYS_ADDR_T_64BIT + def_bool PPC64 || PHYS_64BIT + config MMU bool default y diff --git a/arch/powerpc/include/asm/types.h b/arch/powerpc/include/asm/types.h index d3374bc865b..c646f34c4e8 100644 --- a/arch/powerpc/include/asm/types.h +++ b/arch/powerpc/include/asm/types.h @@ -48,13 +48,6 @@ typedef struct { typedef __vector128 vector128; -/* Physical address used by some IO functions */ -#if defined(CONFIG_PPC64) || defined(CONFIG_PHYS_64BIT) -typedef u64 phys_addr_t; -#else -typedef u32 phys_addr_t; -#endif - #ifdef __powerpc64__ typedef u64 dma_addr_t; #else diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index ed92864d132..a0ffb5188c8 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -932,6 +932,9 @@ config X86_PAE has the cost of more pagetable lookup overhead, and also consumes more pagetable space per process. +config ARCH_PHYS_ADDR_T_64BIT + def_bool X86_64 || X86_PAE + # Common NUMA Features config NUMA bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)" diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index ab8528793f0..d0e58605a52 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -33,7 +33,6 @@ typedef u64 pmdval_t; typedef u64 pudval_t; typedef u64 pgdval_t; typedef u64 pgprotval_t; -typedef u64 phys_addr_t; typedef union { struct { @@ -54,7 +53,6 @@ typedef unsigned long pmdval_t; typedef unsigned long pudval_t; typedef unsigned long pgdval_t; typedef unsigned long pgprotval_t; -typedef unsigned long phys_addr_t; typedef union { pteval_t pte; diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h index c6916c83e6b..2456fbf2d7d 100644 --- a/include/asm-x86/page_64.h +++ b/include/asm-x86/page_64.h @@ -79,7 +79,6 @@ typedef unsigned long pmdval_t; typedef unsigned long pudval_t; typedef unsigned long pgdval_t; typedef unsigned long pgprotval_t; -typedef unsigned long phys_addr_t; typedef struct page *pgtable_t; diff --git a/include/linux/types.h b/include/linux/types.h index d4a9ce6e276..022c668496d 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -197,6 +197,12 @@ typedef u64 resource_size_t; typedef u32 resource_size_t; #endif +#ifdef CONFIG_PHYS_ADDR_T_64BIT +typedef u64 phys_addr_t; +#else +typedef u32 phys_addr_t; +#endif + struct ustat { __kernel_daddr_t f_tfree; __kernel_ino_t f_tinode; diff --git a/mm/Kconfig b/mm/Kconfig index 0bd9c2dbb2a..91ee3922510 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -187,6 +187,9 @@ config RESOURCES_64BIT help This option allows memory and IO resources to be 64 bit. +config PHYS_ADDR_T_64BIT + def_bool 64BIT || ARCH_PHYS_ADDR_T_64BIT + config ZONE_DMA_FLAG int default "0" if !ZONE_DMA -- cgit v1.2.3-70-g09d2 From 947d0496cf3e12ebfa70b3eaf561c25403247ce9 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 11 Sep 2008 01:31:48 -0700 Subject: generic: make PFN_PHYS explicitly return phys_addr_t PFN_PHYS, as its name suggests, turns a pfn into a physical address. However, it is a macro which just operates on its argument without modifying its type. pfns are typed unsigned long, but an unsigned long may not be long enough to hold a physical address (32-bit systems with more than 32 bits of physcial address). Make sure we cast to phys_addr_t to return a complete result. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/m32r/mm/discontig.c | 4 ++-- include/asm-x86/xen/page.h | 4 ++-- include/linux/pfn.h | 6 +++++- 3 files changed, 9 insertions(+), 5 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c index cbc3c4c5456..7daf897292c 100644 --- a/arch/m32r/mm/discontig.c +++ b/arch/m32r/mm/discontig.c @@ -111,9 +111,9 @@ unsigned long __init setup_memory(void) initrd_start, INITRD_SIZE); } else { printk("initrd extends beyond end of memory " - "(0x%08lx > 0x%08lx)\ndisabling initrd\n", + "(0x%08lx > 0x%08llx)\ndisabling initrd\n", INITRD_START + INITRD_SIZE, - PFN_PHYS(max_low_pfn)); + (unsigned long long)PFN_PHYS(max_low_pfn)); initrd_start = 0; } diff --git a/include/asm-x86/xen/page.h b/include/asm-x86/xen/page.h index 7b3835d3b77..de9170b537a 100644 --- a/include/asm-x86/xen/page.h +++ b/include/asm-x86/xen/page.h @@ -76,13 +76,13 @@ static inline unsigned long mfn_to_pfn(unsigned long mfn) static inline xmaddr_t phys_to_machine(xpaddr_t phys) { unsigned offset = phys.paddr & ~PAGE_MASK; - return XMADDR(PFN_PHYS((u64)pfn_to_mfn(PFN_DOWN(phys.paddr))) | offset); + return XMADDR(PFN_PHYS(pfn_to_mfn(PFN_DOWN(phys.paddr))) | offset); } static inline xpaddr_t machine_to_phys(xmaddr_t machine) { unsigned offset = machine.maddr & ~PAGE_MASK; - return XPADDR(PFN_PHYS((u64)mfn_to_pfn(PFN_DOWN(machine.maddr))) | offset); + return XPADDR(PFN_PHYS(mfn_to_pfn(PFN_DOWN(machine.maddr))) | offset); } /* diff --git a/include/linux/pfn.h b/include/linux/pfn.h index bb01f8b92b5..7646637221f 100644 --- a/include/linux/pfn.h +++ b/include/linux/pfn.h @@ -1,9 +1,13 @@ #ifndef _LINUX_PFN_H_ #define _LINUX_PFN_H_ +#ifndef __ASSEMBLY__ +#include +#endif + #define PFN_ALIGN(x) (((unsigned long)(x) + (PAGE_SIZE - 1)) & PAGE_MASK) #define PFN_UP(x) (((x) + PAGE_SIZE-1) >> PAGE_SHIFT) #define PFN_DOWN(x) ((x) >> PAGE_SHIFT) -#define PFN_PHYS(x) ((x) << PAGE_SHIFT) +#define PFN_PHYS(x) ((phys_addr_t)(x) << PAGE_SHIFT) #endif -- cgit v1.2.3-70-g09d2 From fbdbf709938d155c719c76b9894d28342632c797 Mon Sep 17 00:00:00 2001 From: Uwe Kleine-König Date: Mon, 15 Sep 2008 22:02:43 +0200 Subject: x86, debug: gpio_free might sleep MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit According to the documentation gpio_free should only be called from task context only. To make this more explicit add a might sleep to all implementations. This patch changes the gpio_free implementations for the x86 architecture. Signed-off-by: Uwe Kleine-König Signed-off-by: Ingo Molnar --- include/asm-x86/mach-rdc321x/gpio.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/mach-rdc321x/gpio.h b/include/asm-x86/mach-rdc321x/gpio.h index acce0b7d397..3639ece6485 100644 --- a/include/asm-x86/mach-rdc321x/gpio.h +++ b/include/asm-x86/mach-rdc321x/gpio.h @@ -1,6 +1,8 @@ #ifndef _RDC321X_GPIO_H #define _RDC321X_GPIO_H +#include + extern int rdc_gpio_get_value(unsigned gpio); extern void rdc_gpio_set_value(unsigned gpio, int value); extern int rdc_gpio_direction_input(unsigned gpio); @@ -18,6 +20,7 @@ static inline int gpio_request(unsigned gpio, const char *label) static inline void gpio_free(unsigned gpio) { + might_sleep(); rdc_gpio_free(gpio); } -- cgit v1.2.3-70-g09d2 From 2842e5bf3115193f05dc9dac20f940e7abf44c1a Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 18 Sep 2008 15:23:43 +0200 Subject: x86: move GART TLB flushing options to generic code The GART currently implements the iommu=[no]fullflush command line parameters which influence its IO/TLB flushing strategy. This patch makes these parameters generic so that they can be used by the AMD IOMMU too. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- Documentation/kernel-parameters.txt | 4 ++++ Documentation/x86/x86_64/boot-options.txt | 2 -- arch/x86/kernel/pci-dma.c | 13 +++++++++++++ arch/x86/kernel/pci-gart_64.c | 13 ------------- include/asm-x86/iommu.h | 1 + 5 files changed, 18 insertions(+), 15 deletions(-) (limited to 'include/asm-x86') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 1150444a21a..40066ceb48f 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -893,6 +893,10 @@ and is between 256 and 4096 characters. It is defined in the file nomerge forcesac soft + fullflush + Flush IO/TLB at every deallocation + nofullflush + Flush IO/TLB only when addresses are reused (default) intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option diff --git a/Documentation/x86/x86_64/boot-options.txt b/Documentation/x86/x86_64/boot-options.txt index b0c7b6c4abd..c83c8e4bc8e 100644 --- a/Documentation/x86/x86_64/boot-options.txt +++ b/Documentation/x86/x86_64/boot-options.txt @@ -233,8 +233,6 @@ IOMMU (input/output memory management unit) iommu options only relevant to the AMD GART hardware IOMMU: Set the size of the remapping area in bytes. allowed Overwrite iommu off workarounds for specific chipsets. - fullflush Flush IOMMU on each allocation (default). - nofullflush Don't use IOMMU fullflush. leak Turn on simple iommu leak tracing (only when CONFIG_IOMMU_LEAK is on). Default number of leak pages is 20. diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 0a1408abcc6..d2f2c0158dc 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -16,6 +16,15 @@ EXPORT_SYMBOL(dma_ops); static int iommu_sac_force __read_mostly; +/* + * If this is disabled the IOMMU will use an optimized flushing strategy + * of only flushing when an mapping is reused. With it true the GART is + * flushed for every mapping. Problem is that doing the lazy flush seems + * to trigger bugs with some popular PCI cards, in particular 3ware (but + * has been also also seen with Qlogic at least). + */ +int iommu_fullflush; + #ifdef CONFIG_IOMMU_DEBUG int panic_on_overflow __read_mostly = 1; int force_iommu __read_mostly = 1; @@ -171,6 +180,10 @@ static __init int iommu_setup(char *p) } if (!strncmp(p, "nomerge", 7)) iommu_merge = 0; + if (!strncmp(p, "fullflush", 8)) + iommu_fullflush = 1; + if (!strncmp(p, "nofullflush", 11)) + iommu_fullflush = 0; if (!strncmp(p, "forcesac", 8)) iommu_sac_force = 1; if (!strncmp(p, "allowdac", 8)) diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 9739d568209..508ef470b27 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -45,15 +45,6 @@ static unsigned long iommu_pages; /* .. and in pages */ static u32 *iommu_gatt_base; /* Remapping table */ -/* - * If this is disabled the IOMMU will use an optimized flushing strategy - * of only flushing when an mapping is reused. With it true the GART is - * flushed for every mapping. Problem is that doing the lazy flush seems - * to trigger bugs with some popular PCI cards, in particular 3ware (but - * has been also also seen with Qlogic at least). - */ -int iommu_fullflush = 1; - /* Allocation bitmap for the remapping area: */ static DEFINE_SPINLOCK(iommu_bitmap_lock); /* Guarded by iommu_bitmap_lock: */ @@ -901,10 +892,6 @@ void __init gart_parse_options(char *p) #endif if (isdigit(*p) && get_option(&p, &arg)) iommu_size = arg; - if (!strncmp(p, "fullflush", 8)) - iommu_fullflush = 1; - if (!strncmp(p, "nofullflush", 11)) - iommu_fullflush = 0; if (!strncmp(p, "noagp", 5)) no_agp = 1; if (!strncmp(p, "noaperture", 10)) diff --git a/include/asm-x86/iommu.h b/include/asm-x86/iommu.h index 621a1af94c4..67b2fd56c6d 100644 --- a/include/asm-x86/iommu.h +++ b/include/asm-x86/iommu.h @@ -7,6 +7,7 @@ extern struct dma_mapping_ops nommu_dma_ops; extern int force_iommu, no_iommu; extern int iommu_detected; extern int dmar_disabled; +extern int iommu_fullflush; extern unsigned long iommu_num_pages(unsigned long addr, unsigned long len); -- cgit v1.2.3-70-g09d2 From 1c65577398589bb44ab0980f9b9d30804b48a5db Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 4 Sep 2008 18:40:05 +0200 Subject: AMD IOMMU: implement lazy IO/TLB flushing The IO/TLB flushing on every unmaping operation is the most expensive part in AMD IOMMU code and not strictly necessary. It is sufficient to do the flush before any entries are reused. This is patch implements lazy IO/TLB flushing which does exactly this. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/amd_iommu.c | 26 ++++++++++++++++++++++---- arch/x86/kernel/amd_iommu_init.c | 7 ++++++- include/asm-x86/amd_iommu_types.h | 3 +++ 3 files changed, 31 insertions(+), 5 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 691e023695a..679f2a8e22e 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -203,6 +203,14 @@ static int iommu_flush_pages(struct amd_iommu *iommu, u16 domid, return 0; } +/* Flush the whole IO/TLB for a given protection domain */ +static void iommu_flush_tlb(struct amd_iommu *iommu, u16 domid) +{ + u64 address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS; + + iommu_queue_inv_iommu_pages(iommu, address, domid, 0, 1); +} + /**************************************************************************** * * The functions below are used the create the page table mappings for @@ -386,14 +394,18 @@ static unsigned long dma_ops_alloc_addresses(struct device *dev, PAGE_SIZE) >> PAGE_SHIFT; limit = limit < size ? limit : size; - if (dom->next_bit >= limit) + if (dom->next_bit >= limit) { dom->next_bit = 0; + dom->need_flush = true; + } address = iommu_area_alloc(dom->bitmap, limit, dom->next_bit, pages, 0 , boundary_size, 0); - if (address == -1) + if (address == -1) { address = iommu_area_alloc(dom->bitmap, limit, 0, pages, 0, boundary_size, 0); + dom->need_flush = true; + } if (likely(address != -1)) { dom->next_bit = address + pages; @@ -553,6 +565,8 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu, dma_dom->bitmap[0] = 1; dma_dom->next_bit = 0; + dma_dom->need_flush = false; + /* Intialize the exclusion range if necessary */ if (iommu->exclusion_start && iommu->exclusion_start < dma_dom->aperture_size) { @@ -795,7 +809,10 @@ static dma_addr_t __map_single(struct device *dev, } address += offset; - if (unlikely(iommu_has_npcache(iommu))) + if (unlikely(dma_dom->need_flush && !iommu_fullflush)) { + iommu_flush_tlb(iommu, dma_dom->domain.id); + dma_dom->need_flush = false; + } else if (unlikely(iommu_has_npcache(iommu))) iommu_flush_pages(iommu, dma_dom->domain.id, address, size); out: @@ -829,7 +846,8 @@ static void __unmap_single(struct amd_iommu *iommu, dma_ops_free_addresses(dma_dom, dma_addr, pages); - iommu_flush_pages(iommu, dma_dom->domain.id, dma_addr, size); + if (iommu_fullflush) + iommu_flush_pages(iommu, dma_dom->domain.id, dma_addr, size); } /* diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index a69cc0f5204..f2fa8dc81be 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -995,6 +995,11 @@ int __init amd_iommu_init(void) else printk("disabled\n"); + if (iommu_fullflush) + printk(KERN_INFO "AMD IOMMU: IO/TLB flush on unmap enabled\n"); + else + printk(KERN_INFO "AMD IOMMU: Lazy IO/TLB flushing enabled\n"); + out: return ret; @@ -1057,7 +1062,7 @@ void __init amd_iommu_detect(void) static int __init parse_amd_iommu_options(char *str) { for (; *str; ++str) { - if (strcmp(str, "isolate") == 0) + if (strncmp(str, "isolate", 7) == 0) amd_iommu_isolate = 1; } diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index dcc81206739..dcc472445ff 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -196,6 +196,9 @@ struct dma_ops_domain { * just calculate its address in constant time. */ u64 **pte_pages; + + /* This will be set to true when TLB needs to be flushed */ + bool need_flush; }; /* -- cgit v1.2.3-70-g09d2 From 335503e57b6b8de04cec5d27eb2c3d09ff98905b Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Fri, 5 Sep 2008 14:29:07 +0200 Subject: AMD IOMMU: add event buffer allocation This patch adds the allocation of a event buffer for each AMD IOMMU in the system. The hardware will log events like device page faults or other errors to this buffer once this is enabled. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/amd_iommu_init.c | 29 +++++++++++++++++++++++++++++ include/asm-x86/amd_iommu_types.h | 9 +++++++++ 2 files changed, 38 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index f2fa8dc81be..41ce8d5d626 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -417,6 +417,30 @@ static void __init free_command_buffer(struct amd_iommu *iommu) free_pages((unsigned long)iommu->cmd_buf, get_order(CMD_BUFFER_SIZE)); } +/* allocates the memory where the IOMMU will log its events to */ +static u8 * __init alloc_event_buffer(struct amd_iommu *iommu) +{ + u64 entry; + iommu->evt_buf = (u8 *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, + get_order(EVT_BUFFER_SIZE)); + + if (iommu->evt_buf == NULL) + return NULL; + + entry = (u64)virt_to_phys(iommu->evt_buf) | EVT_LEN_MASK; + memcpy_toio(iommu->mmio_base + MMIO_EVT_BUF_OFFSET, + &entry, sizeof(entry)); + + iommu->evt_buf_size = EVT_BUFFER_SIZE; + + return iommu->evt_buf; +} + +static void __init free_event_buffer(struct amd_iommu *iommu) +{ + free_pages((unsigned long)iommu->evt_buf, get_order(EVT_BUFFER_SIZE)); +} + /* sets a specific bit in the device table entry. */ static void set_dev_entry_bit(u16 devid, u8 bit) { @@ -622,6 +646,7 @@ static int __init init_iommu_devices(struct amd_iommu *iommu) static void __init free_iommu_one(struct amd_iommu *iommu) { free_command_buffer(iommu); + free_event_buffer(iommu); iommu_unmap_mmio_space(iommu); } @@ -661,6 +686,10 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h) if (!iommu->cmd_buf) return -ENOMEM; + iommu->evt_buf = alloc_event_buffer(iommu); + if (!iommu->evt_buf) + return -ENOMEM; + init_iommu_from_pci(iommu); init_iommu_from_acpi(iommu, h); init_iommu_devices(iommu); diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index dcc472445ff..8b8cd0c60b3 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -116,6 +116,10 @@ #define MMIO_CMD_SIZE_SHIFT 56 #define MMIO_CMD_SIZE_512 (0x9ULL << MMIO_CMD_SIZE_SHIFT) +/* constants for event buffer handling */ +#define EVT_BUFFER_SIZE 8192 /* 512 entries */ +#define EVT_LEN_MASK (0x9ULL << 56) + #define PAGE_MODE_1_LEVEL 0x01 #define PAGE_MODE_2_LEVEL 0x02 #define PAGE_MODE_3_LEVEL 0x03 @@ -243,6 +247,11 @@ struct amd_iommu { /* size of command buffer */ u32 cmd_buf_size; + /* event buffer virtual address */ + u8 *evt_buf; + /* size of event buffer */ + u32 evt_buf_size; + /* if one, we need to send a completion wait command */ int need_sync; -- cgit v1.2.3-70-g09d2 From ee893c24edb8ebab9a3fb66566855572579ad616 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 8 Sep 2008 14:48:04 +0200 Subject: AMD IOMMU: save pci segment from ACPI tables This patch adds the pci_seg field to the amd_iommu structure and fills it with the corresponding value from the ACPI table. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/amd_iommu_init.c | 1 + include/asm-x86/amd_iommu_types.h | 3 +++ 2 files changed, 4 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 41ce8d5d626..b50234ef91e 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -676,6 +676,7 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h) */ iommu->devid = h->devid; iommu->cap_ptr = h->cap_ptr; + iommu->pci_seg = h->pci_seg; iommu->mmio_phys = h->mmio_phys; iommu->mmio_base = iommu_map_mmio_space(h->mmio_phys); if (!iommu->mmio_base) diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index 8b8cd0c60b3..20814b85bbc 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -232,6 +232,9 @@ struct amd_iommu { /* capabilities of that IOMMU read from ACPI */ u32 cap; + /* pci domain of this IOMMU */ + u16 pci_seg; + /* first device this IOMMU handles. read from PCI */ u16 first_device; /* last device this IOMMU handles. read from PCI */ -- cgit v1.2.3-70-g09d2 From 3eaf28a1cd2686aaa185b54d5a5e18e91b41f7f2 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 8 Sep 2008 15:55:10 +0200 Subject: AMD IOMMU: save pci_dev instead of devid We need the pci_dev later anyways to enable MSI for the IOMMU hardware. So remove the devid pointing to the BDF and replace it with the pci_dev structure where the IOMMU is implemented. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/amd_iommu_init.c | 25 ++++++++++++++++--------- include/asm-x86/amd_iommu_types.h | 5 +++-- 2 files changed, 19 insertions(+), 11 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index b50234ef91e..a7eb89d8923 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -242,9 +242,12 @@ static void __init iommu_feature_disable(struct amd_iommu *iommu, u8 bit) /* Function to enable the hardware */ void __init iommu_enable(struct amd_iommu *iommu) { - printk(KERN_INFO "AMD IOMMU: Enabling IOMMU at "); - print_devid(iommu->devid, 0); - printk(" cap 0x%hx\n", iommu->cap_ptr); + printk(KERN_INFO "AMD IOMMU: Enabling IOMMU " + "at %02x:%02x.%x cap 0x%hx\n", + iommu->dev->bus->number, + PCI_SLOT(iommu->dev->devfn), + PCI_FUNC(iommu->dev->devfn), + iommu->cap_ptr); iommu_feature_enable(iommu, CONTROL_IOMMU_EN); } @@ -511,15 +514,14 @@ static void __init set_device_exclusion_range(u16 devid, struct ivmd_header *m) */ static void __init init_iommu_from_pci(struct amd_iommu *iommu) { - int bus = PCI_BUS(iommu->devid); - int dev = PCI_SLOT(iommu->devid); - int fn = PCI_FUNC(iommu->devid); int cap_ptr = iommu->cap_ptr; u32 range; - iommu->cap = read_pci_config(bus, dev, fn, cap_ptr+MMIO_CAP_HDR_OFFSET); + pci_read_config_dword(iommu->dev, cap_ptr + MMIO_CAP_HDR_OFFSET, + &iommu->cap); + pci_read_config_dword(iommu->dev, cap_ptr + MMIO_RANGE_OFFSET, + &range); - range = read_pci_config(bus, dev, fn, cap_ptr+MMIO_RANGE_OFFSET); iommu->first_device = calc_devid(MMIO_GET_BUS(range), MMIO_GET_FD(range)); iommu->last_device = calc_devid(MMIO_GET_BUS(range), @@ -674,7 +676,10 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h) /* * Copy data from ACPI table entry to the iommu struct */ - iommu->devid = h->devid; + iommu->dev = pci_get_bus_and_slot(PCI_BUS(h->devid), h->devid & 0xff); + if (!iommu->dev) + return 1; + iommu->cap_ptr = h->cap_ptr; iommu->pci_seg = h->pci_seg; iommu->mmio_phys = h->mmio_phys; @@ -695,6 +700,8 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h) init_iommu_from_acpi(iommu, h); init_iommu_devices(iommu); + pci_enable_device(iommu->dev); + return 0; } diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index 20814b85bbc..a5629a21557 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -215,8 +215,9 @@ struct amd_iommu { /* locks the accesses to the hardware */ spinlock_t lock; - /* device id of this IOMMU */ - u16 devid; + /* Pointer to PCI device of this IOMMU */ + struct pci_dev *dev; + /* * Capability pointer. There could be more than one IOMMU per PCI * device function if there are more than one AMD IOMMU capability -- cgit v1.2.3-70-g09d2 From a80dc3e0e0dc8393158de317d66ae0f345dc58f9 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 11 Sep 2008 16:51:41 +0200 Subject: AMD IOMMU: add MSI interrupt support The AMD IOMMU can generate interrupts for various reasons. This patch adds the basic interrupt enabling infrastructure to the driver. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 1 + arch/x86/kernel/amd_iommu.c | 11 +++++ arch/x86/kernel/amd_iommu_init.c | 99 ++++++++++++++++++++++++++++++++++++++- include/asm-x86/amd_iommu.h | 3 ++ include/asm-x86/amd_iommu_types.h | 7 +++ 5 files changed, 120 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index ed92864d132..39fd3f42696 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -553,6 +553,7 @@ config CALGARY_IOMMU_ENABLED_BY_DEFAULT config AMD_IOMMU bool "AMD IOMMU support" select SWIOTLB + select PCI_MSI depends on X86_64 && PCI && ACPI help With this option you can enable support for AMD IOMMU hardware in diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 15792ed082e..0e494b9d5f2 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -49,6 +49,17 @@ static int iommu_has_npcache(struct amd_iommu *iommu) return iommu->cap & IOMMU_CAP_NPCACHE; } +/**************************************************************************** + * + * Interrupt handling functions + * + ****************************************************************************/ + +irqreturn_t amd_iommu_int_handler(int irq, void *data) +{ + return IRQ_NONE; +} + /**************************************************************************** * * IOMMU command queuing functions diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index a7eb89d8923..14a06464a69 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -22,6 +22,8 @@ #include #include #include +#include +#include #include #include #include @@ -515,17 +517,20 @@ static void __init set_device_exclusion_range(u16 devid, struct ivmd_header *m) static void __init init_iommu_from_pci(struct amd_iommu *iommu) { int cap_ptr = iommu->cap_ptr; - u32 range; + u32 range, misc; pci_read_config_dword(iommu->dev, cap_ptr + MMIO_CAP_HDR_OFFSET, &iommu->cap); pci_read_config_dword(iommu->dev, cap_ptr + MMIO_RANGE_OFFSET, &range); + pci_read_config_dword(iommu->dev, cap_ptr + MMIO_MISC_OFFSET, + &misc); iommu->first_device = calc_devid(MMIO_GET_BUS(range), MMIO_GET_FD(range)); iommu->last_device = calc_devid(MMIO_GET_BUS(range), MMIO_GET_LD(range)); + iommu->evt_msi_num = MMIO_MSI_NUM(misc); } /* @@ -696,6 +701,8 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h) if (!iommu->evt_buf) return -ENOMEM; + iommu->int_enabled = false; + init_iommu_from_pci(iommu); init_iommu_from_acpi(iommu, h); init_iommu_devices(iommu); @@ -741,6 +748,95 @@ static int __init init_iommu_all(struct acpi_table_header *table) return 0; } +/**************************************************************************** + * + * The following functions initialize the MSI interrupts for all IOMMUs + * in the system. Its a bit challenging because there could be multiple + * IOMMUs per PCI BDF but we can call pci_enable_msi(x) only once per + * pci_dev. + * + ****************************************************************************/ + +static int __init iommu_setup_msix(struct amd_iommu *iommu) +{ + struct amd_iommu *curr; + struct msix_entry entries[32]; /* only 32 supported by AMD IOMMU */ + int nvec = 0, i; + + list_for_each_entry(curr, &amd_iommu_list, list) { + if (curr->dev == iommu->dev) { + entries[nvec].entry = curr->evt_msi_num; + entries[nvec].vector = 0; + curr->int_enabled = true; + nvec++; + } + } + + if (pci_enable_msix(iommu->dev, entries, nvec)) { + pci_disable_msix(iommu->dev); + return 1; + } + + for (i = 0; i < nvec; ++i) { + int r = request_irq(entries->vector, amd_iommu_int_handler, + IRQF_SAMPLE_RANDOM, + "AMD IOMMU", + NULL); + if (r) + goto out_free; + } + + return 0; + +out_free: + for (i -= 1; i >= 0; --i) + free_irq(entries->vector, NULL); + + pci_disable_msix(iommu->dev); + + return 1; +} + +static int __init iommu_setup_msi(struct amd_iommu *iommu) +{ + int r; + struct amd_iommu *curr; + + list_for_each_entry(curr, &amd_iommu_list, list) { + if (curr->dev == iommu->dev) + curr->int_enabled = true; + } + + + if (pci_enable_msi(iommu->dev)) + return 1; + + r = request_irq(iommu->dev->irq, amd_iommu_int_handler, + IRQF_SAMPLE_RANDOM, + "AMD IOMMU", + NULL); + + if (r) { + pci_disable_msi(iommu->dev); + return 1; + } + + return 0; +} + +static int __init iommu_init_msi(struct amd_iommu *iommu) +{ + if (iommu->int_enabled) + return 0; + + if (pci_find_capability(iommu->dev, PCI_CAP_ID_MSIX)) + return iommu_setup_msix(iommu); + else if (pci_find_capability(iommu->dev, PCI_CAP_ID_MSI)) + return iommu_setup_msi(iommu); + + return 1; +} + /**************************************************************************** * * The next functions belong to the third pass of parsing the ACPI @@ -862,6 +958,7 @@ static void __init enable_iommus(void) list_for_each_entry(iommu, &amd_iommu_list, list) { iommu_set_exclusion_range(iommu); + iommu_init_msi(iommu); iommu_enable(iommu); } } diff --git a/include/asm-x86/amd_iommu.h b/include/asm-x86/amd_iommu.h index 30a12049353..2fd97cb250c 100644 --- a/include/asm-x86/amd_iommu.h +++ b/include/asm-x86/amd_iommu.h @@ -20,10 +20,13 @@ #ifndef _ASM_X86_AMD_IOMMU_H #define _ASM_X86_AMD_IOMMU_H +#include + #ifdef CONFIG_AMD_IOMMU extern int amd_iommu_init(void); extern int amd_iommu_init_dma_ops(void); extern void amd_iommu_detect(void); +extern irqreturn_t amd_iommu_int_handler(int irq, void *data); #else static inline int amd_iommu_init(void) { return -ENODEV; } static inline void amd_iommu_detect(void) { } diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index a5629a21557..8533f09b34b 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -37,6 +37,7 @@ /* Capability offsets used by the driver */ #define MMIO_CAP_HDR_OFFSET 0x00 #define MMIO_RANGE_OFFSET 0x0c +#define MMIO_MISC_OFFSET 0x10 /* Masks, shifts and macros to parse the device range capability */ #define MMIO_RANGE_LD_MASK 0xff000000 @@ -48,6 +49,7 @@ #define MMIO_GET_LD(x) (((x) & MMIO_RANGE_LD_MASK) >> MMIO_RANGE_LD_SHIFT) #define MMIO_GET_FD(x) (((x) & MMIO_RANGE_FD_MASK) >> MMIO_RANGE_FD_SHIFT) #define MMIO_GET_BUS(x) (((x) & MMIO_RANGE_BUS_MASK) >> MMIO_RANGE_BUS_SHIFT) +#define MMIO_MSI_NUM(x) ((x) & 0x1f) /* Flag masks for the AMD IOMMU exclusion range */ #define MMIO_EXCL_ENABLE_MASK 0x01ULL @@ -255,10 +257,15 @@ struct amd_iommu { u8 *evt_buf; /* size of event buffer */ u32 evt_buf_size; + /* MSI number for event interrupt */ + u16 evt_msi_num; /* if one, we need to send a completion wait command */ int need_sync; + /* true if interrupts for this IOMMU are already enabled */ + bool int_enabled; + /* default dma_ops domain for that IOMMU */ struct dma_ops_domain *default_dom; }; -- cgit v1.2.3-70-g09d2 From 90008ee4b811c944455752dcb72b291a5ba81b53 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 9 Sep 2008 16:41:05 +0200 Subject: AMD IOMMU: add event handling code This patch adds code for polling and printing out events generated by the AMD IOMMU. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/amd_iommu.c | 87 ++++++++++++++++++++++++++++++++++++++- arch/x86/kernel/amd_iommu_init.c | 1 - include/asm-x86/amd_iommu_types.h | 22 ++++++++++ 3 files changed, 108 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 0e494b9d5f2..0cb8fd2359f 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -55,9 +55,94 @@ static int iommu_has_npcache(struct amd_iommu *iommu) * ****************************************************************************/ +static void iommu_print_event(void *__evt) +{ + u32 *event = __evt; + int type = (event[1] >> EVENT_TYPE_SHIFT) & EVENT_TYPE_MASK; + int devid = (event[0] >> EVENT_DEVID_SHIFT) & EVENT_DEVID_MASK; + int domid = (event[1] >> EVENT_DOMID_SHIFT) & EVENT_DOMID_MASK; + int flags = (event[1] >> EVENT_FLAGS_SHIFT) & EVENT_FLAGS_MASK; + u64 address = (u64)(((u64)event[3]) << 32) | event[2]; + + printk(KERN_ERR "AMD IOMMU: Event logged ["); + + switch (type) { + case EVENT_TYPE_ILL_DEV: + printk("ILLEGAL_DEV_TABLE_ENTRY device=%02x:%02x.%x " + "address=0x%016llx flags=0x%04x]\n", + PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid), + address, flags); + break; + case EVENT_TYPE_IO_FAULT: + printk("IO_PAGE_FAULT device=%02x:%02x.%x " + "domain=0x%04x address=0x%016llx flags=0x%04x]\n", + PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid), + domid, address, flags); + break; + case EVENT_TYPE_DEV_TAB_ERR: + printk("DEV_TAB_HARDWARE_ERROR device=%02x:%02x.%x " + "address=0x%016llx flags=0x%04x]\n", + PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid), + address, flags); + break; + case EVENT_TYPE_PAGE_TAB_ERR: + printk("PAGE_TAB_HARDWARE_ERROR device=%02x:%02x.%x " + "domain=0x%04x address=0x%016llx flags=0x%04x]\n", + PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid), + domid, address, flags); + break; + case EVENT_TYPE_ILL_CMD: + printk("ILLEGAL_COMMAND_ERROR address=0x%016llx]\n", address); + break; + case EVENT_TYPE_CMD_HARD_ERR: + printk("COMMAND_HARDWARE_ERROR address=0x%016llx " + "flags=0x%04x]\n", address, flags); + break; + case EVENT_TYPE_IOTLB_INV_TO: + printk("IOTLB_INV_TIMEOUT device=%02x:%02x.%x " + "address=0x%016llx]\n", + PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid), + address); + break; + case EVENT_TYPE_INV_DEV_REQ: + printk("INVALID_DEVICE_REQUEST device=%02x:%02x.%x " + "address=0x%016llx flags=0x%04x]\n", + PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid), + address, flags); + break; + default: + printk(KERN_ERR "UNKNOWN type=0x%02x]\n", type); + } +} + +static void iommu_poll_events(struct amd_iommu *iommu) +{ + u32 head, tail; + unsigned long flags; + + spin_lock_irqsave(&iommu->lock, flags); + + head = readl(iommu->mmio_base + MMIO_EVT_HEAD_OFFSET); + tail = readl(iommu->mmio_base + MMIO_EVT_TAIL_OFFSET); + + while (head != tail) { + iommu_print_event(iommu->evt_buf + head); + head = (head + EVENT_ENTRY_SIZE) % iommu->evt_buf_size; + } + + writel(head, iommu->mmio_base + MMIO_EVT_HEAD_OFFSET); + + spin_unlock_irqrestore(&iommu->lock, flags); +} + irqreturn_t amd_iommu_int_handler(int irq, void *data) { - return IRQ_NONE; + struct amd_iommu *iommu; + + list_for_each_entry(iommu, &amd_iommu_list, list) + iommu_poll_events(iommu); + + return IRQ_HANDLED; } /**************************************************************************** diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 14a06464a69..eed488892c0 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -32,7 +32,6 @@ /* * definitions for the ACPI scanning code */ -#define PCI_BUS(x) (((x) >> 8) & 0xff) #define IVRS_HEADER_LENGTH 48 #define ACPI_IVHD_TYPE 0x10 diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index 8533f09b34b..d8c5a6c6995 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -71,6 +71,25 @@ /* MMIO status bits */ #define MMIO_STATUS_COM_WAIT_INT_MASK 0x04 +/* event logging constants */ +#define EVENT_ENTRY_SIZE 0x10 +#define EVENT_TYPE_SHIFT 28 +#define EVENT_TYPE_MASK 0xf +#define EVENT_TYPE_ILL_DEV 0x1 +#define EVENT_TYPE_IO_FAULT 0x2 +#define EVENT_TYPE_DEV_TAB_ERR 0x3 +#define EVENT_TYPE_PAGE_TAB_ERR 0x4 +#define EVENT_TYPE_ILL_CMD 0x5 +#define EVENT_TYPE_CMD_HARD_ERR 0x6 +#define EVENT_TYPE_IOTLB_INV_TO 0x7 +#define EVENT_TYPE_INV_DEV_REQ 0x8 +#define EVENT_DEVID_MASK 0xffff +#define EVENT_DEVID_SHIFT 0 +#define EVENT_DOMID_MASK 0xffff +#define EVENT_DOMID_SHIFT 0 +#define EVENT_FLAGS_MASK 0xfff +#define EVENT_FLAGS_SHIFT 0x10 + /* feature control bits */ #define CONTROL_IOMMU_EN 0x00ULL #define CONTROL_HT_TUN_EN 0x01ULL @@ -165,6 +184,9 @@ #define MAX_DOMAIN_ID 65536 +/* FIXME: move this macro to */ +#define PCI_BUS(x) (((x) >> 8) & 0xff) + /* * This structure contains generic data for IOMMU protection domains * independent of their use. -- cgit v1.2.3-70-g09d2 From bd60b735c658e6e8c656e89771d281bcfcf51279 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 11 Sep 2008 10:24:48 +0200 Subject: AMD IOMMU: don't assign preallocated protection domains to devices In isolation mode the protection domains for the devices are preallocated and preassigned. This is bad if a device should be passed to a virtualization guest because the IOMMU code does not know if it is in use by a driver. This patch changes the code to assign the device to the preallocated domain only if there are dma mapping requests for it. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/amd_iommu.c | 43 ++++++++++++++++++++++++++++++++++----- include/asm-x86/amd_iommu_types.h | 6 ++++++ 2 files changed, 44 insertions(+), 5 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index a6a6f8ed1cf..7c179144745 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -33,6 +33,10 @@ static DEFINE_RWLOCK(amd_iommu_devtable_lock); +/* A list of preallocated protection domains */ +static LIST_HEAD(iommu_pd_list); +static DEFINE_SPINLOCK(iommu_pd_list_lock); + /* * general struct to manage commands send to an IOMMU */ @@ -663,6 +667,7 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu, dma_dom->next_bit = 0; dma_dom->need_flush = false; + dma_dom->target_dev = 0xffff; /* Intialize the exclusion range if necessary */ if (iommu->exclusion_start && @@ -768,6 +773,33 @@ static bool check_device(struct device *dev) return true; } +/* + * In this function the list of preallocated protection domains is traversed to + * find the domain for a specific device + */ +static struct dma_ops_domain *find_protection_domain(u16 devid) +{ + struct dma_ops_domain *entry, *ret = NULL; + unsigned long flags; + + if (list_empty(&iommu_pd_list)) + return NULL; + + spin_lock_irqsave(&iommu_pd_list_lock, flags); + + list_for_each_entry(entry, &iommu_pd_list, list) { + if (entry->target_dev == devid) { + ret = entry; + list_del(&ret->list); + break; + } + } + + spin_unlock_irqrestore(&iommu_pd_list_lock, flags); + + return ret; +} + /* * In the dma_ops path we only have the struct device. This function * finds the corresponding IOMMU, the protection domain and the @@ -803,9 +835,11 @@ static int get_device_resources(struct device *dev, *iommu = amd_iommu_rlookup_table[*bdf]; if (*iommu == NULL) return 0; - dma_dom = (*iommu)->default_dom; *domain = domain_for_device(*bdf); if (*domain == NULL) { + dma_dom = find_protection_domain(*bdf); + if (!dma_dom) + dma_dom = (*iommu)->default_dom; *domain = &dma_dom->domain; set_device_domain(*iommu, *domain, *bdf); printk(KERN_INFO "AMD IOMMU: Using protection domain %d for " @@ -1257,10 +1291,9 @@ void prealloc_protection_domains(void) if (!dma_dom) continue; init_unity_mappings_for_device(dma_dom, devid); - set_device_domain(iommu, &dma_dom->domain, devid); - printk(KERN_INFO "AMD IOMMU: Allocated domain %d for device ", - dma_dom->domain.id); - print_devid(devid, 1); + dma_dom->target_dev = devid; + + list_add_tail(&dma_dom->list, &iommu_pd_list); } } diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index d8c5a6c6995..9aa22ead22f 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -227,6 +227,12 @@ struct dma_ops_domain { /* This will be set to true when TLB needs to be flushed */ bool need_flush; + + /* + * if this is a preallocated domain, keep the device for which it was + * preallocated in this variable + */ + u16 target_dev; }; /* -- cgit v1.2.3-70-g09d2 From 38ddf41b198e21d3ecbe5752e875857b7ce7589e Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 11 Sep 2008 10:38:32 +0200 Subject: AMD IOMMU: some set_device_domain cleanups Remove some magic numbers and split the pte_root using standard functions. Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/amd_iommu.c | 9 +++++---- include/asm-x86/amd_iommu_types.h | 3 +++ 2 files changed, 8 insertions(+), 4 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 7c179144745..a34d8e915e3 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -739,12 +739,13 @@ static void set_device_domain(struct amd_iommu *iommu, u64 pte_root = virt_to_phys(domain->pt_root); - pte_root |= (domain->mode & 0x07) << 9; - pte_root |= IOMMU_PTE_IR | IOMMU_PTE_IW | IOMMU_PTE_P | 2; + pte_root |= (domain->mode & DEV_ENTRY_MODE_MASK) + << DEV_ENTRY_MODE_SHIFT; + pte_root |= IOMMU_PTE_IR | IOMMU_PTE_IW | IOMMU_PTE_P | IOMMU_PTE_TV; write_lock_irqsave(&amd_iommu_devtable_lock, flags); - amd_iommu_dev_table[devid].data[0] = pte_root; - amd_iommu_dev_table[devid].data[1] = pte_root >> 32; + amd_iommu_dev_table[devid].data[0] = lower_32_bits(pte_root); + amd_iommu_dev_table[devid].data[1] = upper_32_bits(pte_root); amd_iommu_dev_table[devid].data[2] = domain->id; amd_iommu_pd_table[devid] = domain; diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index 9aa22ead22f..f953309a636 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -130,6 +130,8 @@ #define DEV_ENTRY_NMI_PASS 0xba #define DEV_ENTRY_LINT0_PASS 0xbe #define DEV_ENTRY_LINT1_PASS 0xbf +#define DEV_ENTRY_MODE_MASK 0x07 +#define DEV_ENTRY_MODE_SHIFT 0x09 /* constants to configure the command buffer */ #define CMD_BUFFER_SIZE 8192 @@ -159,6 +161,7 @@ #define IOMMU_MAP_SIZE_L3 (1ULL << 39) #define IOMMU_PTE_P (1ULL << 0) +#define IOMMU_PTE_TV (1ULL << 1) #define IOMMU_PTE_U (1ULL << 59) #define IOMMU_PTE_FC (1ULL << 60) #define IOMMU_PTE_IR (1ULL << 61) -- cgit v1.2.3-70-g09d2 From b3e15bdef689641e7f1bb03efbe56112c3ee82e2 Mon Sep 17 00:00:00 2001 From: Aristeu Rozanski Date: Mon, 22 Sep 2008 13:13:59 -0400 Subject: x86, NMI watchdog: setup before enabling NMI watchdog There's a small window when NMI watchdog is being set up that if any NMIs are triggered, the NMI code will make make use of not initalized wd_ops elements: void setup_apic_nmi_watchdog(void *unused) { if (__get_cpu_var(wd_enabled)) return; /* cheap hack to support suspend/resume */ /* if cpu0 is not active neither should the other cpus */ if (smp_processor_id() != 0 && atomic_read(&nmi_active) <= 0) return; switch (nmi_watchdog) { case NMI_LOCAL_APIC: /* enable it before to avoid race with handler */ --> __get_cpu_var(wd_enabled) = 1; --> if (lapic_watchdog_init(nmi_hz) < 0) { (...) asmlinkage notrace __kprobes void default_do_nmi(struct pt_regs *regs) { (...) if (nmi_watchdog_tick(regs, reason)) return; (...) notrace __kprobes int nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) { (...) if (!__get_cpu_var(wd_enabled)) return rc; switch (nmi_watchdog) { case NMI_LOCAL_APIC: rc |= lapic_wd_event(nmi_hz); (...) int lapic_wd_event(unsigned nmi_hz) { struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); u64 ctr; --> rdmsrl(wd->perfctr_msr, ctr); and wd->*_msr will be initialized on each processor type specific setup, after enabling NMIs for PMIs. Since the counter was just set, the chances of an performance counter generated NMI is minimal, but any other unknown NMI would trigger the problem. This patch fixes the problem by setting everything up before enabling performance counter generated NMIs and will set wd_enabled using a callback function. Signed-off-by: Aristeu Rozanski Acked-by: Don Zickus Acked-by: Prarit Bhargava Acked-by: Vivek Goyal Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perfctr-watchdog.c | 45 +++++++++++++++++++++++++--------- arch/x86/kernel/nmi.c | 11 +++++++-- include/asm-x86/nmi.h | 1 + 3 files changed, 43 insertions(+), 14 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index 62c01006397..6bff382094f 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -295,13 +295,19 @@ static int setup_k7_watchdog(unsigned nmi_hz) /* setup the timer */ wrmsr(evntsel_msr, evntsel, 0); write_watchdog_counter(perfctr_msr, "K7_PERFCTR0",nmi_hz); - apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= K7_EVNTSEL_ENABLE; - wrmsr(evntsel_msr, evntsel, 0); + /* initialize the wd struct before enabling */ wd->perfctr_msr = perfctr_msr; wd->evntsel_msr = evntsel_msr; wd->cccr_msr = 0; /* unused */ + + /* ok, everything is initialized, announce that we're set */ + cpu_nmi_set_wd_enabled(); + + apic_write(APIC_LVTPC, APIC_DM_NMI); + evntsel |= K7_EVNTSEL_ENABLE; + wrmsr(evntsel_msr, evntsel, 0); + return 1; } @@ -379,13 +385,19 @@ static int setup_p6_watchdog(unsigned nmi_hz) wrmsr(evntsel_msr, evntsel, 0); nmi_hz = adjust_for_32bit_ctr(nmi_hz); write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0",nmi_hz); - apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= P6_EVNTSEL0_ENABLE; - wrmsr(evntsel_msr, evntsel, 0); + /* initialize the wd struct before enabling */ wd->perfctr_msr = perfctr_msr; wd->evntsel_msr = evntsel_msr; wd->cccr_msr = 0; /* unused */ + + /* ok, everything is initialized, announce that we're set */ + cpu_nmi_set_wd_enabled(); + + apic_write(APIC_LVTPC, APIC_DM_NMI); + evntsel |= P6_EVNTSEL0_ENABLE; + wrmsr(evntsel_msr, evntsel, 0); + return 1; } @@ -540,12 +552,17 @@ static int setup_p4_watchdog(unsigned nmi_hz) wrmsr(evntsel_msr, evntsel, 0); wrmsr(cccr_msr, cccr_val, 0); write_watchdog_counter(perfctr_msr, "P4_IQ_COUNTER0", nmi_hz); - apic_write(APIC_LVTPC, APIC_DM_NMI); - cccr_val |= P4_CCCR_ENABLE; - wrmsr(cccr_msr, cccr_val, 0); + wd->perfctr_msr = perfctr_msr; wd->evntsel_msr = evntsel_msr; wd->cccr_msr = cccr_msr; + + /* ok, everything is initialized, announce that we're set */ + cpu_nmi_set_wd_enabled(); + + apic_write(APIC_LVTPC, APIC_DM_NMI); + cccr_val |= P4_CCCR_ENABLE; + wrmsr(cccr_msr, cccr_val, 0); return 1; } @@ -661,13 +678,17 @@ static int setup_intel_arch_watchdog(unsigned nmi_hz) wrmsr(evntsel_msr, evntsel, 0); nmi_hz = adjust_for_32bit_ctr(nmi_hz); write_watchdog_counter32(perfctr_msr, "INTEL_ARCH_PERFCTR0", nmi_hz); - apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE; - wrmsr(evntsel_msr, evntsel, 0); wd->perfctr_msr = perfctr_msr; wd->evntsel_msr = evntsel_msr; wd->cccr_msr = 0; /* unused */ + + /* ok, everything is initialized, announce that we're set */ + cpu_nmi_set_wd_enabled(); + + apic_write(APIC_LVTPC, APIC_DM_NMI); + evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE; + wrmsr(evntsel_msr, evntsel, 0); intel_arch_wd_ops.checkbit = 1ULL << (eax.split.bit_width - 1); return 1; } diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c index abb78a2cc4a..2c97f07f1c2 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/nmi.c @@ -299,6 +299,15 @@ void acpi_nmi_disable(void) on_each_cpu(__acpi_nmi_disable, NULL, 1); } +/* + * This function is called as soon the LAPIC NMI watchdog driver has everything + * in place and it's ready to check if the NMIs belong to the NMI watchdog + */ +void cpu_nmi_set_wd_enabled(void) +{ + __get_cpu_var(wd_enabled) = 1; +} + void setup_apic_nmi_watchdog(void *unused) { if (__get_cpu_var(wd_enabled)) @@ -311,8 +320,6 @@ void setup_apic_nmi_watchdog(void *unused) switch (nmi_watchdog) { case NMI_LOCAL_APIC: - /* enable it before to avoid race with handler */ - __get_cpu_var(wd_enabled) = 1; if (lapic_watchdog_init(nmi_hz) < 0) { __get_cpu_var(wd_enabled) = 0; return; diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h index 21f8d0202a8..02bfc81cbd6 100644 --- a/include/asm-x86/nmi.h +++ b/include/asm-x86/nmi.h @@ -34,6 +34,7 @@ extern void stop_apic_nmi_watchdog(void *); extern void disable_timer_nmi_watchdog(void); extern void enable_timer_nmi_watchdog(void); extern int nmi_watchdog_tick(struct pt_regs *regs, unsigned reason); +extern void cpu_nmi_set_wd_enabled(void); extern atomic_t nmi_active; extern unsigned int nmi_watchdog; -- cgit v1.2.3-70-g09d2 From ed6dc4981368aa8ac89b0ea61535cfa2b03533cb Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Tue, 23 Sep 2008 01:15:10 +0900 Subject: x86: remove set_bit_string() "export iommu_area_reserve helper funciton" patch converted all the users of set_bit_string, GART, Calgary and AMD IOMMU drivers, to use iommu_area_reserve helper function. Now we can remove unused set_bit_string function. Signed-off-by: FUJITA Tomonori Signed-off-by: Ingo Molnar --- include/asm-x86/bitops.h | 10 ---------- 1 file changed, 10 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/bitops.h b/include/asm-x86/bitops.h index cfb2b64f76e..28ea8742b6e 100644 --- a/include/asm-x86/bitops.h +++ b/include/asm-x86/bitops.h @@ -424,16 +424,6 @@ static inline int fls(int x) #undef ADDR -static inline void set_bit_string(unsigned long *bitmap, - unsigned long i, int len) -{ - unsigned long end = i + len; - while (i < end) { - __set_bit(i, bitmap); - i++; - } -} - #ifdef __KERNEL__ #include -- cgit v1.2.3-70-g09d2 From afa9fdc2f5f8e4d98f3e77bfa204412cbc181346 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Sat, 20 Sep 2008 01:23:30 +0900 Subject: iommu: remove fullflush and nofullflush in IOMMU generic option This patch against tip/x86/iommu virtually reverts 2842e5bf3115193f05dc9dac20f940e7abf44c1a. But just reverting the commit breaks AMD IOMMU so this patch also includes some fixes. The above commit adds new two options to x86 IOMMU generic kernel boot options, fullflush and nofullflush. But such change that affects all the IOMMUs needs more discussion (all IOMMU parties need the chance to discuss it): http://lkml.org/lkml/2008/9/19/106 Signed-off-by: FUJITA Tomonori Acked-by: Joerg Roedel Signed-off-by: Ingo Molnar --- Documentation/kernel-parameters.txt | 9 +++++---- Documentation/x86/x86_64/boot-options.txt | 2 ++ arch/x86/kernel/amd_iommu.c | 4 ++-- arch/x86/kernel/amd_iommu_init.c | 5 ++++- arch/x86/kernel/pci-dma.c | 13 ------------- arch/x86/kernel/pci-gart_64.c | 13 +++++++++++++ include/asm-x86/amd_iommu_types.h | 6 ++++++ include/asm-x86/iommu.h | 1 - 8 files changed, 32 insertions(+), 21 deletions(-) (limited to 'include/asm-x86') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 40066ceb48f..040ce30632b 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -284,6 +284,11 @@ and is between 256 and 4096 characters. It is defined in the file isolate - enable device isolation (each device, as far as possible, will get its own protection domain) + fullflush - enable flushing of IO/TLB entries when + they are unmapped. Otherwise they are + flushed before they will be reused, which + is a lot of faster + amd_iommu_size= [HW,X86-64] Define the size of the aperture for the AMD IOMMU driver. Possible values are: @@ -893,10 +898,6 @@ and is between 256 and 4096 characters. It is defined in the file nomerge forcesac soft - fullflush - Flush IO/TLB at every deallocation - nofullflush - Flush IO/TLB only when addresses are reused (default) intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option diff --git a/Documentation/x86/x86_64/boot-options.txt b/Documentation/x86/x86_64/boot-options.txt index c83c8e4bc8e..b0c7b6c4abd 100644 --- a/Documentation/x86/x86_64/boot-options.txt +++ b/Documentation/x86/x86_64/boot-options.txt @@ -233,6 +233,8 @@ IOMMU (input/output memory management unit) iommu options only relevant to the AMD GART hardware IOMMU: Set the size of the remapping area in bytes. allowed Overwrite iommu off workarounds for specific chipsets. + fullflush Flush IOMMU on each allocation (default). + nofullflush Don't use IOMMU fullflush. leak Turn on simple iommu leak tracing (only when CONFIG_IOMMU_LEAK is on). Default number of leak pages is 20. diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 70537d117a9..c19212191c9 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -948,7 +948,7 @@ static dma_addr_t __map_single(struct device *dev, } address += offset; - if (unlikely(dma_dom->need_flush && !iommu_fullflush)) { + if (unlikely(dma_dom->need_flush && !amd_iommu_unmap_flush)) { iommu_flush_tlb(iommu, dma_dom->domain.id); dma_dom->need_flush = false; } else if (unlikely(iommu_has_npcache(iommu))) @@ -985,7 +985,7 @@ static void __unmap_single(struct amd_iommu *iommu, dma_ops_free_addresses(dma_dom, dma_addr, pages); - if (iommu_fullflush) + if (amd_iommu_unmap_flush) iommu_flush_pages(iommu, dma_dom->domain.id, dma_addr, size); } diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index db0c83af44d..148fcfe22f1 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -122,6 +122,7 @@ LIST_HEAD(amd_iommu_unity_map); /* a list of required unity mappings we find in ACPI */ unsigned amd_iommu_aperture_order = 26; /* size of aperture in power of 2 */ int amd_iommu_isolate; /* if 1, device isolation is enabled */ +bool amd_iommu_unmap_flush; /* if true, flush on every unmap */ LIST_HEAD(amd_iommu_list); /* list of all AMD IOMMUs in the system */ @@ -1144,7 +1145,7 @@ int __init amd_iommu_init(void) else printk("disabled\n"); - if (iommu_fullflush) + if (amd_iommu_unmap_flush) printk(KERN_INFO "AMD IOMMU: IO/TLB flush on unmap enabled\n"); else printk(KERN_INFO "AMD IOMMU: Lazy IO/TLB flushing enabled\n"); @@ -1214,6 +1215,8 @@ static int __init parse_amd_iommu_options(char *str) for (; *str; ++str) { if (strncmp(str, "isolate", 7) == 0) amd_iommu_isolate = 1; + if (strncmp(str, "fullflush", 11) == 0) + amd_iommu_unmap_flush = true; } return 1; diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index d2f2c0158dc..0a1408abcc6 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -16,15 +16,6 @@ EXPORT_SYMBOL(dma_ops); static int iommu_sac_force __read_mostly; -/* - * If this is disabled the IOMMU will use an optimized flushing strategy - * of only flushing when an mapping is reused. With it true the GART is - * flushed for every mapping. Problem is that doing the lazy flush seems - * to trigger bugs with some popular PCI cards, in particular 3ware (but - * has been also also seen with Qlogic at least). - */ -int iommu_fullflush; - #ifdef CONFIG_IOMMU_DEBUG int panic_on_overflow __read_mostly = 1; int force_iommu __read_mostly = 1; @@ -180,10 +171,6 @@ static __init int iommu_setup(char *p) } if (!strncmp(p, "nomerge", 7)) iommu_merge = 0; - if (!strncmp(p, "fullflush", 8)) - iommu_fullflush = 1; - if (!strncmp(p, "nofullflush", 11)) - iommu_fullflush = 0; if (!strncmp(p, "forcesac", 8)) iommu_sac_force = 1; if (!strncmp(p, "allowdac", 8)) diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 3dcb1ad86e3..9e390f1bd46 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -45,6 +45,15 @@ static unsigned long iommu_pages; /* .. and in pages */ static u32 *iommu_gatt_base; /* Remapping table */ +/* + * If this is disabled the IOMMU will use an optimized flushing strategy + * of only flushing when an mapping is reused. With it true the GART is + * flushed for every mapping. Problem is that doing the lazy flush seems + * to trigger bugs with some popular PCI cards, in particular 3ware (but + * has been also also seen with Qlogic at least). + */ +int iommu_fullflush = 1; + /* Allocation bitmap for the remapping area: */ static DEFINE_SPINLOCK(iommu_bitmap_lock); /* Guarded by iommu_bitmap_lock: */ @@ -892,6 +901,10 @@ void __init gart_parse_options(char *p) #endif if (isdigit(*p) && get_option(&p, &arg)) iommu_size = arg; + if (!strncmp(p, "fullflush", 8)) + iommu_fullflush = 1; + if (!strncmp(p, "nofullflush", 11)) + iommu_fullflush = 0; if (!strncmp(p, "noagp", 5)) no_agp = 1; if (!strncmp(p, "noaperture", 10)) diff --git a/include/asm-x86/amd_iommu_types.h b/include/asm-x86/amd_iommu_types.h index f953309a636..4ff892f3b0a 100644 --- a/include/asm-x86/amd_iommu_types.h +++ b/include/asm-x86/amd_iommu_types.h @@ -376,6 +376,12 @@ extern unsigned long *amd_iommu_pd_alloc_bitmap; /* will be 1 if device isolation is enabled */ extern int amd_iommu_isolate; +/* + * If true, the addresses will be flushed on unmap time, not when + * they are reused + */ +extern bool amd_iommu_unmap_flush; + /* takes a PCI device id and prints it out in a readable form */ static inline void print_devid(u16 devid, int nl) { diff --git a/include/asm-x86/iommu.h b/include/asm-x86/iommu.h index 67b2fd56c6d..621a1af94c4 100644 --- a/include/asm-x86/iommu.h +++ b/include/asm-x86/iommu.h @@ -7,7 +7,6 @@ extern struct dma_mapping_ops nommu_dma_ops; extern int force_iommu, no_iommu; extern int iommu_detected; extern int dmar_disabled; -extern int iommu_fullflush; extern unsigned long iommu_num_pages(unsigned long addr, unsigned long len); -- cgit v1.2.3-70-g09d2 From 4faac97d44ac27bdbb010a9c3597401a8f89341f Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 22 Sep 2008 18:54:29 +0200 Subject: x86: prevent stale state of c1e_mask across CPU offline/online Impact: hang which happens across CPU offline/online on AMD C1E systems. When a CPU goes offline then the corresponding bit in the broadcast mask is cleared. For AMD C1E enabled CPUs we do not reenable the broadcast when the CPU comes online again as we do not clear the corresponding bit in the c1e_mask, which keeps track which CPUs have been switched to broadcast already. So on those !$@#& machines we never switch back to broadcasting after a CPU offline/online cycle. Clear the bit when the CPU plays dead. Signed-off-by: Thomas Gleixner --- arch/x86/kernel/process.c | 11 ++++++++--- arch/x86/kernel/process_32.c | 1 + arch/x86/kernel/process_64.c | 2 ++ include/asm-x86/idle.h | 2 ++ 4 files changed, 13 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 7fc4d5b0a6a..2e2247117f6 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -246,6 +246,14 @@ static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c) return 1; } +static cpumask_t c1e_mask = CPU_MASK_NONE; +static int c1e_detected; + +void c1e_remove_cpu(int cpu) +{ + cpu_clear(cpu, c1e_mask); +} + /* * C1E aware idle routine. We check for C1E active in the interrupt * pending message MSR. If we detect C1E, then we handle it the same @@ -253,9 +261,6 @@ static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c) */ static void c1e_idle(void) { - static cpumask_t c1e_mask = CPU_MASK_NONE; - static int c1e_detected; - if (need_resched()) return; diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 3b7a1ddcc0b..4b3cfdf5421 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -88,6 +88,7 @@ static void cpu_exit_clear(void) cpu_clear(cpu, cpu_callin_map); numa_remove_cpu(cpu); + c1e_remove_cpu(cpu); } /* We don't actually take CPU down, just spin without interrupts. */ diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 71553b664e2..e12e0e4dd25 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -93,6 +93,8 @@ DECLARE_PER_CPU(int, cpu_state); static inline void play_dead(void) { idle_task_exit(); + c1e_remove_cpu(raw_smp_processor_id()); + mb(); /* Ack it */ __get_cpu_var(cpu_state) = CPU_DEAD; diff --git a/include/asm-x86/idle.h b/include/asm-x86/idle.h index d240e5b30a4..cbb64912361 100644 --- a/include/asm-x86/idle.h +++ b/include/asm-x86/idle.h @@ -10,4 +10,6 @@ void idle_notifier_register(struct notifier_block *n); void enter_idle(void); void exit_idle(void); +void c1e_remove_cpu(int cpu); + #endif -- cgit v1.2.3-70-g09d2 From a8d6829044901a67732904be5f1eacdf8539604f Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 22 Sep 2008 19:02:25 +0200 Subject: x86: prevent C-states hang on AMD C1E enabled machines Impact: System hang when AMD C1E machines switch into C2/C3 AMD C1E enabled systems do not work with normal ACPI C-states even if the BIOS is advertising them. Limit the C-states to C1 for the ACPI processor idle code. Signed-off-by: Thomas Gleixner --- arch/x86/kernel/process.c | 1 + include/asm-x86/acpi.h | 2 ++ include/asm-x86/cpufeature.h | 1 + 3 files changed, 4 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 2e2247117f6..d8c2a299bfe 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -272,6 +272,7 @@ static void c1e_idle(void) c1e_detected = 1; mark_tsc_unstable("TSC halt in C1E"); printk(KERN_INFO "System has C1E enabled\n"); + set_cpu_cap(&boot_cpu_data, X86_FEATURE_AMDC1E); } } diff --git a/include/asm-x86/acpi.h b/include/asm-x86/acpi.h index 635d764dc13..35d1743b57a 100644 --- a/include/asm-x86/acpi.h +++ b/include/asm-x86/acpi.h @@ -140,6 +140,8 @@ static inline unsigned int acpi_processor_cstate_check(unsigned int max_cstate) boot_cpu_data.x86_model <= 0x05 && boot_cpu_data.x86_mask < 0x0A) return 1; + else if (boot_cpu_has(X86_FEATURE_AMDC1E)) + return 1; else return max_cstate; } diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h index 9489283a4bc..cfcfb0a806b 100644 --- a/include/asm-x86/cpufeature.h +++ b/include/asm-x86/cpufeature.h @@ -81,6 +81,7 @@ #define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* Lfence synchronizes RDTSC */ #define X86_FEATURE_11AP (3*32+19) /* Bad local APIC aka 11AP */ #define X86_FEATURE_NOPL (3*32+20) /* The NOPL (0F 1F) instructions */ +#define X86_FEATURE_AMDC1E (3*32+21) /* AMD C1E detected */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* Streaming SIMD Extensions-3 */ -- cgit v1.2.3-70-g09d2 From 18dbc9160507dc7df998e00cd1dcd7889557307b Mon Sep 17 00:00:00 2001 From: Dmitry Adamushko Date: Tue, 23 Sep 2008 12:08:44 +0200 Subject: x86: moved microcode.c to microcode_intel.c Combine both generic and arch-specific parts of microcode into a single module (arch-specific parts are config-dependent). Also while we are at it, move arch-specific parts from microcode.h into their respective arch-specific .c files. Signed-off-by: Dmitry Adamushko Cc: "Peter Oruba" Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 10 +- arch/x86/kernel/Makefile | 8 +- arch/x86/kernel/microcode.c | 505 ------------------------------------- arch/x86/kernel/microcode_amd.c | 72 +++--- arch/x86/kernel/microcode_core.c | 509 ++++++++++++++++++++++++++++++++++++++ arch/x86/kernel/microcode_intel.c | 79 +++--- include/asm-x86/microcode.h | 99 ++------ 7 files changed, 632 insertions(+), 650 deletions(-) delete mode 100644 arch/x86/kernel/microcode.c create mode 100644 arch/x86/kernel/microcode_core.c (limited to 'include/asm-x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 0e5bf1eddce..2e608095135 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -801,7 +801,7 @@ config MICROCODE module will be called microcode. config MICROCODE_INTEL - tristate "Intel microcode patch loading support" + bool "Intel microcode patch loading support" depends on MICROCODE default MICROCODE select FW_LOADER @@ -813,20 +813,14 @@ config MICROCODE_INTEL Intel ingredients for this driver, check: . - This driver is only available as a module: the module - will be called microcode_intel. - config MICROCODE_AMD - tristate "AMD microcode patch loading support" + bool "AMD microcode patch loading support" depends on MICROCODE select FW_LOADER --help--- If you select this option, microcode patch loading support for AMD processors will be enabled. - This driver is only available as a module: the module - will be called microcode_amd. - config MICROCODE_OLD_INTERFACE def_bool y depends on MICROCODE diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index be454f348c3..f891996f684 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -51,9 +51,6 @@ obj-$(CONFIG_X86_BIOS_REBOOT) += reboot.o obj-$(CONFIG_MCA) += mca_32.o obj-$(CONFIG_X86_MSR) += msr.o obj-$(CONFIG_X86_CPUID) += cpuid.o -obj-$(CONFIG_MICROCODE) += microcode.o -obj-$(CONFIG_MICROCODE_INTEL) += microcode_intel.o -obj-$(CONFIG_MICROCODE_AMD) += microcode_amd.o obj-$(CONFIG_PCI) += early-quirks.o apm-y := apm_32.o obj-$(CONFIG_APM) += apm.o @@ -101,6 +98,11 @@ scx200-y += scx200_32.o obj-$(CONFIG_OLPC) += olpc.o +microcode-y := microcode_core.o +microcode-$(CONFIG_MICROCODE_INTEL) += microcode_intel.o +microcode-$(CONFIG_MICROCODE_AMD) += microcode_amd.o +obj-$(CONFIG_MICROCODE) += microcode.o + ### # 64 bit specific files ifeq ($(CONFIG_X86_64),y) diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c deleted file mode 100644 index 0c2634f4fd7..00000000000 --- a/arch/x86/kernel/microcode.c +++ /dev/null @@ -1,505 +0,0 @@ -/* - * Intel CPU Microcode Update Driver for Linux - * - * Copyright (C) 2000-2006 Tigran Aivazian - * 2006 Shaohua Li - * - * This driver allows to upgrade microcode on Intel processors - * belonging to IA-32 family - PentiumPro, Pentium II, - * Pentium III, Xeon, Pentium 4, etc. - * - * Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture - * Software Developer's Manual - * Order Number 253668 or free download from: - * - * http://developer.intel.com/design/pentium4/manuals/253668.htm - * - * For more information, go to http://www.urbanmyth.org/microcode - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * 1.0 16 Feb 2000, Tigran Aivazian - * Initial release. - * 1.01 18 Feb 2000, Tigran Aivazian - * Added read() support + cleanups. - * 1.02 21 Feb 2000, Tigran Aivazian - * Added 'device trimming' support. open(O_WRONLY) zeroes - * and frees the saved copy of applied microcode. - * 1.03 29 Feb 2000, Tigran Aivazian - * Made to use devfs (/dev/cpu/microcode) + cleanups. - * 1.04 06 Jun 2000, Simon Trimmer - * Added misc device support (now uses both devfs and misc). - * Added MICROCODE_IOCFREE ioctl to clear memory. - * 1.05 09 Jun 2000, Simon Trimmer - * Messages for error cases (non Intel & no suitable microcode). - * 1.06 03 Aug 2000, Tigran Aivazian - * Removed ->release(). Removed exclusive open and status bitmap. - * Added microcode_rwsem to serialize read()/write()/ioctl(). - * Removed global kernel lock usage. - * 1.07 07 Sep 2000, Tigran Aivazian - * Write 0 to 0x8B msr and then cpuid before reading revision, - * so that it works even if there were no update done by the - * BIOS. Otherwise, reading from 0x8B gives junk (which happened - * to be 0 on my machine which is why it worked even when I - * disabled update by the BIOS) - * Thanks to Eric W. Biederman for the fix. - * 1.08 11 Dec 2000, Richard Schaal and - * Tigran Aivazian - * Intel Pentium 4 processor support and bugfixes. - * 1.09 30 Oct 2001, Tigran Aivazian - * Bugfix for HT (Hyper-Threading) enabled processors - * whereby processor resources are shared by all logical processors - * in a single CPU package. - * 1.10 28 Feb 2002 Asit K Mallick and - * Tigran Aivazian , - * Serialize updates as required on HT processors due to - * speculative nature of implementation. - * 1.11 22 Mar 2002 Tigran Aivazian - * Fix the panic when writing zero-length microcode chunk. - * 1.12 29 Sep 2003 Nitin Kamble , - * Jun Nakajima - * Support for the microcode updates in the new format. - * 1.13 10 Oct 2003 Tigran Aivazian - * Removed ->read() method and obsoleted MICROCODE_IOCFREE ioctl - * because we no longer hold a copy of applied microcode - * in kernel memory. - * 1.14 25 Jun 2004 Tigran Aivazian - * Fix sigmatch() macro to handle old CPUs with pf == 0. - * Thanks to Stuart Swales for pointing out this bug. - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include - -MODULE_DESCRIPTION("Microcode Update Driver"); -MODULE_AUTHOR("Tigran Aivazian "); -MODULE_LICENSE("GPL"); - -#define MICROCODE_VERSION "2.00" - -struct microcode_ops *microcode_ops; - -/* no concurrent ->write()s are allowed on /dev/cpu/microcode */ -static DEFINE_MUTEX(microcode_mutex); - -struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; -EXPORT_SYMBOL_GPL(ucode_cpu_info); - -#ifdef CONFIG_MICROCODE_OLD_INTERFACE -static int do_microcode_update(const void __user *buf, size_t size) -{ - cpumask_t old; - int error = 0; - int cpu; - - old = current->cpus_allowed; - - for_each_online_cpu(cpu) { - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - - if (!uci->valid) - continue; - - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - error = microcode_ops->request_microcode_user(cpu, buf, size); - if (error < 0) - goto out; - if (!error) - microcode_ops->apply_microcode(cpu); - } -out: - set_cpus_allowed_ptr(current, &old); - return error; -} - -static int microcode_open(struct inode *unused1, struct file *unused2) -{ - cycle_kernel_lock(); - return capable(CAP_SYS_RAWIO) ? 0 : -EPERM; -} - -static ssize_t microcode_write(struct file *file, const char __user *buf, - size_t len, loff_t *ppos) -{ - ssize_t ret; - - if ((len >> PAGE_SHIFT) > num_physpages) { - printk(KERN_ERR "microcode: too much data (max %ld pages)\n", - num_physpages); - return -EINVAL; - } - - get_online_cpus(); - mutex_lock(µcode_mutex); - - ret = do_microcode_update(buf, len); - if (!ret) - ret = (ssize_t)len; - - mutex_unlock(µcode_mutex); - put_online_cpus(); - - return ret; -} - -static const struct file_operations microcode_fops = { - .owner = THIS_MODULE, - .write = microcode_write, - .open = microcode_open, -}; - -static struct miscdevice microcode_dev = { - .minor = MICROCODE_MINOR, - .name = "microcode", - .fops = µcode_fops, -}; - -static int __init microcode_dev_init(void) -{ - int error; - - error = misc_register(µcode_dev); - if (error) { - printk(KERN_ERR - "microcode: can't misc_register on minor=%d\n", - MICROCODE_MINOR); - return error; - } - - return 0; -} - -static void microcode_dev_exit(void) -{ - misc_deregister(µcode_dev); -} - -MODULE_ALIAS_MISCDEV(MICROCODE_MINOR); -#else -#define microcode_dev_init() 0 -#define microcode_dev_exit() do { } while (0) -#endif - -/* fake device for request_firmware */ -struct platform_device *microcode_pdev; - -static ssize_t reload_store(struct sys_device *dev, - struct sysdev_attribute *attr, - const char *buf, size_t sz) -{ - struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; - char *end; - unsigned long val = simple_strtoul(buf, &end, 0); - int err = 0; - int cpu = dev->id; - - if (end == buf) - return -EINVAL; - if (val == 1) { - cpumask_t old = current->cpus_allowed; - - get_online_cpus(); - if (cpu_online(cpu)) { - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - mutex_lock(µcode_mutex); - if (uci->valid) { - err = microcode_ops->request_microcode_fw(cpu, - µcode_pdev->dev); - if (!err) - microcode_ops->apply_microcode(cpu); - } - mutex_unlock(µcode_mutex); - set_cpus_allowed_ptr(current, &old); - } - put_online_cpus(); - } - if (err) - return err; - return sz; -} - -static ssize_t version_show(struct sys_device *dev, - struct sysdev_attribute *attr, char *buf) -{ - struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; - - return sprintf(buf, "0x%x\n", uci->cpu_sig.rev); -} - -static ssize_t pf_show(struct sys_device *dev, - struct sysdev_attribute *attr, char *buf) -{ - struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; - - return sprintf(buf, "0x%x\n", uci->cpu_sig.pf); -} - -static SYSDEV_ATTR(reload, 0200, NULL, reload_store); -static SYSDEV_ATTR(version, 0400, version_show, NULL); -static SYSDEV_ATTR(processor_flags, 0400, pf_show, NULL); - -static struct attribute *mc_default_attrs[] = { - &attr_reload.attr, - &attr_version.attr, - &attr_processor_flags.attr, - NULL -}; - -static struct attribute_group mc_attr_group = { - .attrs = mc_default_attrs, - .name = "microcode", -}; - -static void microcode_fini_cpu(int cpu) -{ - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - - mutex_lock(µcode_mutex); - microcode_ops->microcode_fini_cpu(cpu); - uci->valid = 0; - mutex_unlock(µcode_mutex); -} - -static void collect_cpu_info(int cpu) -{ - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - - memset(uci, 0, sizeof(*uci)); - if (!microcode_ops->collect_cpu_info(cpu, &uci->cpu_sig)) - uci->valid = 1; -} - -static int microcode_resume_cpu(int cpu) -{ - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - struct cpu_signature nsig; - - pr_debug("microcode: CPU%d resumed\n", cpu); - - if (!uci->mc.valid_mc) - return 1; - - /* - * Let's verify that the 'cached' ucode does belong - * to this cpu (a bit of paranoia): - */ - if (microcode_ops->collect_cpu_info(cpu, &nsig)) { - microcode_fini_cpu(cpu); - return -1; - } - - if (memcmp(&nsig, &uci->cpu_sig, sizeof(nsig))) { - microcode_fini_cpu(cpu); - /* Should we look for a new ucode here? */ - return 1; - } - - return 0; -} - -void microcode_update_cpu(int cpu) -{ - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - int err = 0; - - /* We should bind the task to the CPU */ - BUG_ON(raw_smp_processor_id() != cpu); - - mutex_lock(µcode_mutex); - /* - * Check if the system resume is in progress (uci->valid != NULL), - * otherwise just request a firmware: - */ - if (uci->valid) { - err = microcode_resume_cpu(cpu); - } else { - collect_cpu_info(cpu); - if (uci->valid && system_state == SYSTEM_RUNNING) - err = microcode_ops->request_microcode_fw(cpu, - µcode_pdev->dev); - } - - if (!err) - microcode_ops->apply_microcode(cpu); - - mutex_unlock(µcode_mutex); -} - -static void microcode_init_cpu(int cpu) -{ - cpumask_t old = current->cpus_allowed; - - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - microcode_update_cpu(cpu); - set_cpus_allowed_ptr(current, &old); -} - -static int mc_sysdev_add(struct sys_device *sys_dev) -{ - int err, cpu = sys_dev->id; - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - - if (!cpu_online(cpu)) - return 0; - - pr_debug("microcode: CPU%d added\n", cpu); - memset(uci, 0, sizeof(*uci)); - - err = sysfs_create_group(&sys_dev->kobj, &mc_attr_group); - if (err) - return err; - - microcode_init_cpu(cpu); - return 0; -} - -static int mc_sysdev_remove(struct sys_device *sys_dev) -{ - int cpu = sys_dev->id; - - if (!cpu_online(cpu)) - return 0; - - pr_debug("microcode: CPU%d removed\n", cpu); - microcode_fini_cpu(cpu); - sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); - return 0; -} - -static int mc_sysdev_resume(struct sys_device *dev) -{ - int cpu = dev->id; - - if (!cpu_online(cpu)) - return 0; - - /* only CPU 0 will apply ucode here */ - microcode_update_cpu(0); - return 0; -} - -static struct sysdev_driver mc_sysdev_driver = { - .add = mc_sysdev_add, - .remove = mc_sysdev_remove, - .resume = mc_sysdev_resume, -}; - -static __cpuinit int -mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) -{ - unsigned int cpu = (unsigned long)hcpu; - struct sys_device *sys_dev; - - sys_dev = get_cpu_sysdev(cpu); - switch (action) { - case CPU_ONLINE: - case CPU_ONLINE_FROZEN: - microcode_init_cpu(cpu); - case CPU_DOWN_FAILED: - case CPU_DOWN_FAILED_FROZEN: - pr_debug("microcode: CPU%d added\n", cpu); - if (sysfs_create_group(&sys_dev->kobj, &mc_attr_group)) - printk(KERN_ERR "microcode: Failed to create the sysfs " - "group for CPU%d\n", cpu); - break; - case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: - /* Suspend is in progress, only remove the interface */ - sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); - pr_debug("microcode: CPU%d removed\n", cpu); - break; - case CPU_DEAD: - case CPU_UP_CANCELED_FROZEN: - /* The CPU refused to come up during a system resume */ - microcode_fini_cpu(cpu); - break; - } - return NOTIFY_OK; -} - -static struct notifier_block __refdata mc_cpu_notifier = { - .notifier_call = mc_cpu_callback, -}; - -int microcode_init(void *opaque, struct module *module) -{ - struct microcode_ops *ops = (struct microcode_ops *)opaque; - int error; - - if (microcode_ops) { - printk(KERN_ERR "microcode: already loaded the other module\n"); - return -EEXIST; - } - - microcode_ops = ops; - - error = microcode_dev_init(); - if (error) - return error; - microcode_pdev = platform_device_register_simple("microcode", -1, - NULL, 0); - if (IS_ERR(microcode_pdev)) { - microcode_dev_exit(); - return PTR_ERR(microcode_pdev); - } - - get_online_cpus(); - error = sysdev_driver_register(&cpu_sysdev_class, &mc_sysdev_driver); - put_online_cpus(); - if (error) { - microcode_dev_exit(); - platform_device_unregister(microcode_pdev); - return error; - } - - register_hotcpu_notifier(&mc_cpu_notifier); - - printk(KERN_INFO - "Microcode Update Driver: v" MICROCODE_VERSION - " " - " \n"); - - return 0; -} -EXPORT_SYMBOL_GPL(microcode_init); - -void __exit microcode_exit(void) -{ - microcode_dev_exit(); - - unregister_hotcpu_notifier(&mc_cpu_notifier); - - get_online_cpus(); - sysdev_driver_unregister(&cpu_sysdev_class, &mc_sysdev_driver); - put_online_cpus(); - - platform_device_unregister(microcode_pdev); - - microcode_ops = NULL; - - printk(KERN_INFO - "Microcode Update Driver: v" MICROCODE_VERSION " removed.\n"); -} -EXPORT_SYMBOL_GPL(microcode_exit); diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c index 48aec9f48e4..03ea4e52e87 100644 --- a/arch/x86/kernel/microcode_amd.c +++ b/arch/x86/kernel/microcode_amd.c @@ -46,6 +46,35 @@ MODULE_LICENSE("GPL v2"); #define UCODE_EQUIV_CPU_TABLE_TYPE 0x00000000 #define UCODE_UCODE_TYPE 0x00000001 +struct equiv_cpu_entry { + unsigned int installed_cpu; + unsigned int fixed_errata_mask; + unsigned int fixed_errata_compare; + unsigned int equiv_cpu; +}; + +struct microcode_header_amd { + unsigned int data_code; + unsigned int patch_id; + unsigned char mc_patch_data_id[2]; + unsigned char mc_patch_data_len; + unsigned char init_flag; + unsigned int mc_patch_data_checksum; + unsigned int nb_dev_id; + unsigned int sb_dev_id; + unsigned char processor_rev_id[2]; + unsigned char nb_rev_id; + unsigned char sb_rev_id; + unsigned char bios_api_rev; + unsigned char reserved1[3]; + unsigned int match_reg[8]; +}; + +struct microcode_amd { + struct microcode_header_amd hdr; + unsigned int mpb[0]; +}; + #define UCODE_MAX_SIZE (2048) #define DEFAULT_UCODE_DATASIZE (896) #define MC_HEADER_SIZE (sizeof(struct microcode_header_amd)) @@ -189,17 +218,18 @@ static void apply_microcode_amd(int cpu) unsigned int rev; int cpu_num = raw_smp_processor_id(); struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num; + struct microcode_amd *mc_amd = uci->mc; unsigned long addr; /* We should bind the task to the CPU */ BUG_ON(cpu_num != cpu); - if (uci->mc.mc_amd == NULL) + if (mc_amd == NULL) return; spin_lock_irqsave(µcode_update_lock, flags); - addr = (unsigned long)&uci->mc.mc_amd->hdr.data_code; + addr = (unsigned long)&mc_amd->hdr.data_code; edx = (unsigned int)(((unsigned long)upper_32_bits(addr))); eax = (unsigned int)(((unsigned long)lower_32_bits(addr))); @@ -214,16 +244,16 @@ static void apply_microcode_amd(int cpu) spin_unlock_irqrestore(µcode_update_lock, flags); /* check current patch id and patch's id for match */ - if (rev != uci->mc.mc_amd->hdr.patch_id) { + if (rev != mc_amd->hdr.patch_id) { printk(KERN_ERR "microcode: CPU%d update from revision " "0x%x to 0x%x failed\n", cpu_num, - uci->mc.mc_amd->hdr.patch_id, rev); + mc_amd->hdr.patch_id, rev); return; } printk(KERN_INFO "microcode: CPU%d updated from revision " "0x%x to 0x%x \n", - cpu_num, uci->cpu_sig.rev, uci->mc.mc_amd->hdr.patch_id); + cpu_num, uci->cpu_sig.rev, mc_amd->hdr.patch_id); uci->cpu_sig.rev = rev; } @@ -355,12 +385,12 @@ static int generic_load_microcode(int cpu, void *data, size_t size, if (new_mc) { if (!leftover) { - if (uci->mc.mc_amd) - vfree(uci->mc.mc_amd); - uci->mc.mc_amd = (struct microcode_amd *)new_mc; + if (uci->mc) + vfree(uci->mc); + uci->mc = new_mc; pr_debug("microcode: CPU%d found a matching microcode update with" " version 0x%x (current=0x%x)\n", - cpu, uci->mc.mc_amd->hdr.patch_id, uci->cpu_sig.rev); + cpu, new_rev, uci->cpu_sig.rev); } else vfree(new_mc); } @@ -416,8 +446,8 @@ static void microcode_fini_cpu_amd(int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - vfree(uci->mc.mc_amd); - uci->mc.mc_amd = NULL; + vfree(uci->mc); + uci->mc = NULL; } static struct microcode_ops microcode_amd_ops = { @@ -428,23 +458,7 @@ static struct microcode_ops microcode_amd_ops = { .microcode_fini_cpu = microcode_fini_cpu_amd, }; -static int __init microcode_amd_module_init(void) +struct microcode_ops * __init init_amd_microcode(void) { - struct cpuinfo_x86 *c = &cpu_data(0); - - equiv_cpu_table = NULL; - if (c->x86_vendor != X86_VENDOR_AMD) { - printk(KERN_ERR "microcode: CPU platform is not AMD-capable\n"); - return -ENODEV; - } - - return microcode_init(µcode_amd_ops, THIS_MODULE); -} - -static void __exit microcode_amd_module_exit(void) -{ - microcode_exit(); + return µcode_amd_ops; } - -module_init(microcode_amd_module_init) -module_exit(microcode_amd_module_exit) diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c new file mode 100644 index 00000000000..ff031dbccdf --- /dev/null +++ b/arch/x86/kernel/microcode_core.c @@ -0,0 +1,509 @@ +/* + * Intel CPU Microcode Update Driver for Linux + * + * Copyright (C) 2000-2006 Tigran Aivazian + * 2006 Shaohua Li + * + * This driver allows to upgrade microcode on Intel processors + * belonging to IA-32 family - PentiumPro, Pentium II, + * Pentium III, Xeon, Pentium 4, etc. + * + * Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture + * Software Developer's Manual + * Order Number 253668 or free download from: + * + * http://developer.intel.com/design/pentium4/manuals/253668.htm + * + * For more information, go to http://www.urbanmyth.org/microcode + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * 1.0 16 Feb 2000, Tigran Aivazian + * Initial release. + * 1.01 18 Feb 2000, Tigran Aivazian + * Added read() support + cleanups. + * 1.02 21 Feb 2000, Tigran Aivazian + * Added 'device trimming' support. open(O_WRONLY) zeroes + * and frees the saved copy of applied microcode. + * 1.03 29 Feb 2000, Tigran Aivazian + * Made to use devfs (/dev/cpu/microcode) + cleanups. + * 1.04 06 Jun 2000, Simon Trimmer + * Added misc device support (now uses both devfs and misc). + * Added MICROCODE_IOCFREE ioctl to clear memory. + * 1.05 09 Jun 2000, Simon Trimmer + * Messages for error cases (non Intel & no suitable microcode). + * 1.06 03 Aug 2000, Tigran Aivazian + * Removed ->release(). Removed exclusive open and status bitmap. + * Added microcode_rwsem to serialize read()/write()/ioctl(). + * Removed global kernel lock usage. + * 1.07 07 Sep 2000, Tigran Aivazian + * Write 0 to 0x8B msr and then cpuid before reading revision, + * so that it works even if there were no update done by the + * BIOS. Otherwise, reading from 0x8B gives junk (which happened + * to be 0 on my machine which is why it worked even when I + * disabled update by the BIOS) + * Thanks to Eric W. Biederman for the fix. + * 1.08 11 Dec 2000, Richard Schaal and + * Tigran Aivazian + * Intel Pentium 4 processor support and bugfixes. + * 1.09 30 Oct 2001, Tigran Aivazian + * Bugfix for HT (Hyper-Threading) enabled processors + * whereby processor resources are shared by all logical processors + * in a single CPU package. + * 1.10 28 Feb 2002 Asit K Mallick and + * Tigran Aivazian , + * Serialize updates as required on HT processors due to + * speculative nature of implementation. + * 1.11 22 Mar 2002 Tigran Aivazian + * Fix the panic when writing zero-length microcode chunk. + * 1.12 29 Sep 2003 Nitin Kamble , + * Jun Nakajima + * Support for the microcode updates in the new format. + * 1.13 10 Oct 2003 Tigran Aivazian + * Removed ->read() method and obsoleted MICROCODE_IOCFREE ioctl + * because we no longer hold a copy of applied microcode + * in kernel memory. + * 1.14 25 Jun 2004 Tigran Aivazian + * Fix sigmatch() macro to handle old CPUs with pf == 0. + * Thanks to Stuart Swales for pointing out this bug. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +MODULE_DESCRIPTION("Microcode Update Driver"); +MODULE_AUTHOR("Tigran Aivazian "); +MODULE_LICENSE("GPL"); + +#define MICROCODE_VERSION "2.00" + +struct microcode_ops *microcode_ops; + +/* no concurrent ->write()s are allowed on /dev/cpu/microcode */ +static DEFINE_MUTEX(microcode_mutex); + +struct ucode_cpu_info ucode_cpu_info[NR_CPUS]; +EXPORT_SYMBOL_GPL(ucode_cpu_info); + +#ifdef CONFIG_MICROCODE_OLD_INTERFACE +static int do_microcode_update(const void __user *buf, size_t size) +{ + cpumask_t old; + int error = 0; + int cpu; + + old = current->cpus_allowed; + + for_each_online_cpu(cpu) { + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + + if (!uci->valid) + continue; + + set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); + error = microcode_ops->request_microcode_user(cpu, buf, size); + if (error < 0) + goto out; + if (!error) + microcode_ops->apply_microcode(cpu); + } +out: + set_cpus_allowed_ptr(current, &old); + return error; +} + +static int microcode_open(struct inode *unused1, struct file *unused2) +{ + cycle_kernel_lock(); + return capable(CAP_SYS_RAWIO) ? 0 : -EPERM; +} + +static ssize_t microcode_write(struct file *file, const char __user *buf, + size_t len, loff_t *ppos) +{ + ssize_t ret; + + if ((len >> PAGE_SHIFT) > num_physpages) { + printk(KERN_ERR "microcode: too much data (max %ld pages)\n", + num_physpages); + return -EINVAL; + } + + get_online_cpus(); + mutex_lock(µcode_mutex); + + ret = do_microcode_update(buf, len); + if (!ret) + ret = (ssize_t)len; + + mutex_unlock(µcode_mutex); + put_online_cpus(); + + return ret; +} + +static const struct file_operations microcode_fops = { + .owner = THIS_MODULE, + .write = microcode_write, + .open = microcode_open, +}; + +static struct miscdevice microcode_dev = { + .minor = MICROCODE_MINOR, + .name = "microcode", + .fops = µcode_fops, +}; + +static int __init microcode_dev_init(void) +{ + int error; + + error = misc_register(µcode_dev); + if (error) { + printk(KERN_ERR + "microcode: can't misc_register on minor=%d\n", + MICROCODE_MINOR); + return error; + } + + return 0; +} + +static void microcode_dev_exit(void) +{ + misc_deregister(µcode_dev); +} + +MODULE_ALIAS_MISCDEV(MICROCODE_MINOR); +#else +#define microcode_dev_init() 0 +#define microcode_dev_exit() do { } while (0) +#endif + +/* fake device for request_firmware */ +struct platform_device *microcode_pdev; + +static ssize_t reload_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t sz) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; + char *end; + unsigned long val = simple_strtoul(buf, &end, 0); + int err = 0; + int cpu = dev->id; + + if (end == buf) + return -EINVAL; + if (val == 1) { + cpumask_t old = current->cpus_allowed; + + get_online_cpus(); + if (cpu_online(cpu)) { + set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); + mutex_lock(µcode_mutex); + if (uci->valid) { + err = microcode_ops->request_microcode_fw(cpu, + µcode_pdev->dev); + if (!err) + microcode_ops->apply_microcode(cpu); + } + mutex_unlock(µcode_mutex); + set_cpus_allowed_ptr(current, &old); + } + put_online_cpus(); + } + if (err) + return err; + return sz; +} + +static ssize_t version_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; + + return sprintf(buf, "0x%x\n", uci->cpu_sig.rev); +} + +static ssize_t pf_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; + + return sprintf(buf, "0x%x\n", uci->cpu_sig.pf); +} + +static SYSDEV_ATTR(reload, 0200, NULL, reload_store); +static SYSDEV_ATTR(version, 0400, version_show, NULL); +static SYSDEV_ATTR(processor_flags, 0400, pf_show, NULL); + +static struct attribute *mc_default_attrs[] = { + &attr_reload.attr, + &attr_version.attr, + &attr_processor_flags.attr, + NULL +}; + +static struct attribute_group mc_attr_group = { + .attrs = mc_default_attrs, + .name = "microcode", +}; + +static void microcode_fini_cpu(int cpu) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + + mutex_lock(µcode_mutex); + microcode_ops->microcode_fini_cpu(cpu); + uci->valid = 0; + mutex_unlock(µcode_mutex); +} + +static void collect_cpu_info(int cpu) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + + memset(uci, 0, sizeof(*uci)); + if (!microcode_ops->collect_cpu_info(cpu, &uci->cpu_sig)) + uci->valid = 1; +} + +static int microcode_resume_cpu(int cpu) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + struct cpu_signature nsig; + + pr_debug("microcode: CPU%d resumed\n", cpu); + + if (!uci->mc) + return 1; + + /* + * Let's verify that the 'cached' ucode does belong + * to this cpu (a bit of paranoia): + */ + if (microcode_ops->collect_cpu_info(cpu, &nsig)) { + microcode_fini_cpu(cpu); + return -1; + } + + if (memcmp(&nsig, &uci->cpu_sig, sizeof(nsig))) { + microcode_fini_cpu(cpu); + /* Should we look for a new ucode here? */ + return 1; + } + + return 0; +} + +void microcode_update_cpu(int cpu) +{ + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + int err = 0; + + /* We should bind the task to the CPU */ + BUG_ON(raw_smp_processor_id() != cpu); + + mutex_lock(µcode_mutex); + /* + * Check if the system resume is in progress (uci->valid != NULL), + * otherwise just request a firmware: + */ + if (uci->valid) { + err = microcode_resume_cpu(cpu); + } else { + collect_cpu_info(cpu); + if (uci->valid && system_state == SYSTEM_RUNNING) + err = microcode_ops->request_microcode_fw(cpu, + µcode_pdev->dev); + } + + if (!err) + microcode_ops->apply_microcode(cpu); + + mutex_unlock(µcode_mutex); +} + +static void microcode_init_cpu(int cpu) +{ + cpumask_t old = current->cpus_allowed; + + set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); + microcode_update_cpu(cpu); + set_cpus_allowed_ptr(current, &old); +} + +static int mc_sysdev_add(struct sys_device *sys_dev) +{ + int err, cpu = sys_dev->id; + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + + if (!cpu_online(cpu)) + return 0; + + pr_debug("microcode: CPU%d added\n", cpu); + memset(uci, 0, sizeof(*uci)); + + err = sysfs_create_group(&sys_dev->kobj, &mc_attr_group); + if (err) + return err; + + microcode_init_cpu(cpu); + return 0; +} + +static int mc_sysdev_remove(struct sys_device *sys_dev) +{ + int cpu = sys_dev->id; + + if (!cpu_online(cpu)) + return 0; + + pr_debug("microcode: CPU%d removed\n", cpu); + microcode_fini_cpu(cpu); + sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); + return 0; +} + +static int mc_sysdev_resume(struct sys_device *dev) +{ + int cpu = dev->id; + + if (!cpu_online(cpu)) + return 0; + + /* only CPU 0 will apply ucode here */ + microcode_update_cpu(0); + return 0; +} + +static struct sysdev_driver mc_sysdev_driver = { + .add = mc_sysdev_add, + .remove = mc_sysdev_remove, + .resume = mc_sysdev_resume, +}; + +static __cpuinit int +mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) +{ + unsigned int cpu = (unsigned long)hcpu; + struct sys_device *sys_dev; + + sys_dev = get_cpu_sysdev(cpu); + switch (action) { + case CPU_ONLINE: + case CPU_ONLINE_FROZEN: + microcode_init_cpu(cpu); + case CPU_DOWN_FAILED: + case CPU_DOWN_FAILED_FROZEN: + pr_debug("microcode: CPU%d added\n", cpu); + if (sysfs_create_group(&sys_dev->kobj, &mc_attr_group)) + printk(KERN_ERR "microcode: Failed to create the sysfs " + "group for CPU%d\n", cpu); + break; + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + /* Suspend is in progress, only remove the interface */ + sysfs_remove_group(&sys_dev->kobj, &mc_attr_group); + pr_debug("microcode: CPU%d removed\n", cpu); + break; + case CPU_DEAD: + case CPU_UP_CANCELED_FROZEN: + /* The CPU refused to come up during a system resume */ + microcode_fini_cpu(cpu); + break; + } + return NOTIFY_OK; +} + +static struct notifier_block __refdata mc_cpu_notifier = { + .notifier_call = mc_cpu_callback, +}; + +static int __init microcode_init(void) +{ + struct cpuinfo_x86 *c = &cpu_data(0); + int error; + + if (c->x86_vendor == X86_VENDOR_INTEL) + microcode_ops = init_intel_microcode(); + else if (c->x86_vendor != X86_VENDOR_AMD) + microcode_ops = init_amd_microcode(); + + if (!microcode_ops) { + printk(KERN_ERR "microcode: no support for this CPU vendor\n"); + return -ENODEV; + } + + error = microcode_dev_init(); + if (error) + return error; + microcode_pdev = platform_device_register_simple("microcode", -1, + NULL, 0); + if (IS_ERR(microcode_pdev)) { + microcode_dev_exit(); + return PTR_ERR(microcode_pdev); + } + + get_online_cpus(); + error = sysdev_driver_register(&cpu_sysdev_class, &mc_sysdev_driver); + put_online_cpus(); + if (error) { + microcode_dev_exit(); + platform_device_unregister(microcode_pdev); + return error; + } + + register_hotcpu_notifier(&mc_cpu_notifier); + + printk(KERN_INFO + "Microcode Update Driver: v" MICROCODE_VERSION + " " + " \n"); + + return 0; +} + +static void __exit microcode_exit(void) +{ + microcode_dev_exit(); + + unregister_hotcpu_notifier(&mc_cpu_notifier); + + get_online_cpus(); + sysdev_driver_unregister(&cpu_sysdev_class, &mc_sysdev_driver); + put_online_cpus(); + + platform_device_unregister(microcode_pdev); + + microcode_ops = NULL; + + printk(KERN_INFO + "Microcode Update Driver: v" MICROCODE_VERSION " removed.\n"); +} + +module_init(microcode_init); +module_exit(microcode_exit); diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index 48ed3cef58c..622dc4a2178 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -97,6 +97,38 @@ MODULE_DESCRIPTION("Microcode Update Driver"); MODULE_AUTHOR("Tigran Aivazian "); MODULE_LICENSE("GPL"); +struct microcode_header_intel { + unsigned int hdrver; + unsigned int rev; + unsigned int date; + unsigned int sig; + unsigned int cksum; + unsigned int ldrver; + unsigned int pf; + unsigned int datasize; + unsigned int totalsize; + unsigned int reserved[3]; +}; + +struct microcode_intel { + struct microcode_header_intel hdr; + unsigned int bits[0]; +}; + +/* microcode format is extended from prescott processors */ +struct extended_signature { + unsigned int sig; + unsigned int pf; + unsigned int cksum; +}; + +struct extended_sigtable { + unsigned int count; + unsigned int cksum; + unsigned int reserved[3]; + struct extended_signature sigs[0]; +}; + #define DEFAULT_UCODE_DATASIZE (2000) #define MC_HEADER_SIZE (sizeof(struct microcode_header_intel)) #define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE) @@ -284,11 +316,12 @@ static void apply_microcode(int cpu) unsigned int val[2]; int cpu_num = raw_smp_processor_id(); struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + struct microcode_intel *mc_intel = uci->mc; /* We should bind the task to the CPU */ BUG_ON(cpu_num != cpu); - if (uci->mc.mc_intel == NULL) + if (mc_intel == NULL) return; /* serialize access to the physical write to MSR 0x79 */ @@ -296,8 +329,8 @@ static void apply_microcode(int cpu) /* write microcode via MSR 0x79 */ wrmsr(MSR_IA32_UCODE_WRITE, - (unsigned long) uci->mc.mc_intel->bits, - (unsigned long) uci->mc.mc_intel->bits >> 16 >> 16); + (unsigned long) mc_intel->bits, + (unsigned long) mc_intel->bits >> 16 >> 16); wrmsr(MSR_IA32_UCODE_REV, 0, 0); /* see notes above for revision 1.07. Apparent chip bug */ @@ -307,7 +340,7 @@ static void apply_microcode(int cpu) rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]); spin_unlock_irqrestore(µcode_update_lock, flags); - if (val[1] != uci->mc.mc_intel->hdr.rev) { + if (val[1] != mc_intel->hdr.rev) { printk(KERN_ERR "microcode: CPU%d update from revision " "0x%x to 0x%x failed\n", cpu_num, uci->cpu_sig.rev, val[1]); return; @@ -315,9 +348,9 @@ static void apply_microcode(int cpu) printk(KERN_INFO "microcode: CPU%d updated from revision " "0x%x to 0x%x, date = %04x-%02x-%02x \n", cpu_num, uci->cpu_sig.rev, val[1], - uci->mc.mc_intel->hdr.date & 0xffff, - uci->mc.mc_intel->hdr.date >> 24, - (uci->mc.mc_intel->hdr.date >> 16) & 0xff); + mc_intel->hdr.date & 0xffff, + mc_intel->hdr.date >> 24, + (mc_intel->hdr.date >> 16) & 0xff); uci->cpu_sig.rev = val[1]; } @@ -367,12 +400,12 @@ static int generic_load_microcode(int cpu, void *data, size_t size, if (new_mc) { if (!leftover) { - if (uci->mc.mc_intel) - vfree(uci->mc.mc_intel); - uci->mc.mc_intel = (struct microcode_intel *)new_mc; + if (uci->mc) + vfree(uci->mc); + uci->mc = (struct microcode_intel *)new_mc; pr_debug("microcode: CPU%d found a matching microcode update with" " version 0x%x (current=0x%x)\n", - cpu, uci->mc.mc_intel->hdr.rev, uci->cpu_sig.rev); + cpu, new_rev, uci->cpu_sig.rev); } else vfree(new_mc); } @@ -428,11 +461,11 @@ static void microcode_fini_cpu(int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - vfree(uci->mc.mc_intel); - uci->mc.mc_intel = NULL; + vfree(uci->mc); + uci->mc = NULL; } -static struct microcode_ops microcode_intel_ops = { +struct microcode_ops microcode_intel_ops = { .request_microcode_user = request_microcode_user, .request_microcode_fw = request_microcode_fw, .collect_cpu_info = collect_cpu_info, @@ -440,22 +473,8 @@ static struct microcode_ops microcode_intel_ops = { .microcode_fini_cpu = microcode_fini_cpu, }; -static int __init microcode_intel_module_init(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - - if (c->x86_vendor != X86_VENDOR_INTEL) { - printk(KERN_ERR "microcode: CPU platform is not Intel-capable\n"); - return -ENODEV; - } - - return microcode_init(µcode_intel_ops, THIS_MODULE); -} - -static void __exit microcode_intel_module_exit(void) +struct microcode_ops * __init init_intel_microcode(void) { - microcode_exit(); + return µcode_intel_ops; } -module_init(microcode_intel_module_init) -module_exit(microcode_intel_module_exit) diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index e2887facdb4..62c793bb70c 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -1,10 +1,12 @@ #ifndef ASM_X86__MICROCODE_H #define ASM_X86__MICROCODE_H -extern int microcode_init(void *opaque, struct module *module); -extern void microcode_exit(void); +struct cpu_signature { + unsigned int sig; + unsigned int pf; + unsigned int rev; +}; -struct cpu_signature; struct device; struct microcode_ops { @@ -17,82 +19,29 @@ struct microcode_ops { void (*microcode_fini_cpu) (int cpu); }; -struct microcode_header_intel { - unsigned int hdrver; - unsigned int rev; - unsigned int date; - unsigned int sig; - unsigned int cksum; - unsigned int ldrver; - unsigned int pf; - unsigned int datasize; - unsigned int totalsize; - unsigned int reserved[3]; -}; - -struct microcode_intel { - struct microcode_header_intel hdr; - unsigned int bits[0]; -}; - -/* microcode format is extended from prescott processors */ -struct extended_signature { - unsigned int sig; - unsigned int pf; - unsigned int cksum; -}; - -struct extended_sigtable { - unsigned int count; - unsigned int cksum; - unsigned int reserved[3]; - struct extended_signature sigs[0]; -}; - -struct equiv_cpu_entry { - unsigned int installed_cpu; - unsigned int fixed_errata_mask; - unsigned int fixed_errata_compare; - unsigned int equiv_cpu; -}; - -struct microcode_header_amd { - unsigned int data_code; - unsigned int patch_id; - unsigned char mc_patch_data_id[2]; - unsigned char mc_patch_data_len; - unsigned char init_flag; - unsigned int mc_patch_data_checksum; - unsigned int nb_dev_id; - unsigned int sb_dev_id; - unsigned char processor_rev_id[2]; - unsigned char nb_rev_id; - unsigned char sb_rev_id; - unsigned char bios_api_rev; - unsigned char reserved1[3]; - unsigned int match_reg[8]; -}; - -struct microcode_amd { - struct microcode_header_amd hdr; - unsigned int mpb[0]; -}; - -struct cpu_signature { - unsigned int sig; - unsigned int pf; - unsigned int rev; -}; - struct ucode_cpu_info { struct cpu_signature cpu_sig; int valid; - union { - struct microcode_intel *mc_intel; - struct microcode_amd *mc_amd; - void *valid_mc; - } mc; + void *mc; }; extern struct ucode_cpu_info ucode_cpu_info[]; +#ifdef CONFIG_MICROCODE_INTEL +extern struct microcode_ops * __init init_intel_microcode(void); +#else +static inline struct microcode_ops * __init init_intel_microcode(void) +{ + return NULL; +} +#endif /* CONFIG_MICROCODE_INTEL */ + +#ifdef CONFIG_MICROCODE_AMD +extern struct microcode_ops * __init init_amd_microcode(void); +#else +static inline struct microcode_ops * __init init_amd_microcode(void) +{ + return NULL; +} +#endif + #endif /* ASM_X86__MICROCODE_H */ -- cgit v1.2.3-70-g09d2 From da654b74bda14c45a7d98c731bf3c1a43b6b74e2 Mon Sep 17 00:00:00 2001 From: Srinivasa Ds Date: Tue, 23 Sep 2008 15:23:52 +0530 Subject: signals: demultiplexing SIGTRAP signal Currently a SIGTRAP can denote any one of below reasons. - Breakpoint hit - H/W debug register hit - Single step - Signal sent through kill() or rasie() Architectures like powerpc/parisc provides infrastructure to demultiplex SIGTRAP signal by passing down the information for receiving SIGTRAP through si_code of siginfot_t structure. Here is an attempt is generalise this infrastructure by extending it to x86 and x86_64 archs. Signed-off-by: Srinivasa DS Cc: Roland McGrath Cc: akpm@linux-foundation.org Cc: paulus@samba.org Cc: linuxppc-dev@ozlabs.org Signed-off-by: Ingo Molnar --- arch/ia64/include/asm/siginfo.h | 5 ----- arch/powerpc/include/asm/siginfo.h | 5 ----- arch/x86/kernel/ptrace.c | 7 ++++--- arch/x86/kernel/traps_32.c | 4 +++- arch/x86/kernel/traps_64.c | 2 +- include/asm-generic/siginfo.h | 2 ++ include/asm-parisc/siginfo.h | 5 ----- include/asm-x86/ptrace.h | 2 +- include/asm-x86/traps.h | 10 ++++++++++ 9 files changed, 21 insertions(+), 21 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/ia64/include/asm/siginfo.h b/arch/ia64/include/asm/siginfo.h index 9294e4b0c8b..118d4297900 100644 --- a/arch/ia64/include/asm/siginfo.h +++ b/arch/ia64/include/asm/siginfo.h @@ -113,11 +113,6 @@ typedef struct siginfo { #undef NSIGSEGV #define NSIGSEGV 3 -/* - * SIGTRAP si_codes - */ -#define TRAP_BRANCH (__SI_FAULT|3) /* process taken branch trap */ -#define TRAP_HWBKPT (__SI_FAULT|4) /* hardware breakpoint or watchpoint */ #undef NSIGTRAP #define NSIGTRAP 4 diff --git a/arch/powerpc/include/asm/siginfo.h b/arch/powerpc/include/asm/siginfo.h index 12f1bce037b..49495b0534e 100644 --- a/arch/powerpc/include/asm/siginfo.h +++ b/arch/powerpc/include/asm/siginfo.h @@ -15,11 +15,6 @@ #include -/* - * SIGTRAP si_codes - */ -#define TRAP_BRANCH (__SI_FAULT|3) /* process taken branch trap */ -#define TRAP_HWBKPT (__SI_FAULT|4) /* hardware breakpoint or watchpoint */ #undef NSIGTRAP #define NSIGTRAP 4 diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 9e43a48ad6e..bf45cdf1aac 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -1358,7 +1358,8 @@ const struct user_regset_view *task_user_regset_view(struct task_struct *task) #endif } -void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code) +void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, + int error_code, int si_code) { struct siginfo info; @@ -1367,7 +1368,7 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code) memset(&info, 0, sizeof(info)); info.si_signo = SIGTRAP; - info.si_code = TRAP_BRKPT; + info.si_code = si_code; /* User-mode ip? */ info.si_addr = user_mode_vm(regs) ? (void __user *) regs->ip : NULL; @@ -1454,5 +1455,5 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs) */ if (test_thread_flag(TIF_SINGLESTEP) && tracehook_consider_fatal_signal(current, SIGTRAP, SIG_DFL)) - send_sigtrap(current, regs, 0); + send_sigtrap(current, regs, 0, TRAP_BRKPT); } diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index da5a5964fcc..0429c5de5ea 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -891,6 +891,7 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code) { struct task_struct *tsk = current; unsigned int condition; + int si_code; trace_hardirqs_fixup(); @@ -935,8 +936,9 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code) goto clear_TF_reenable; } + si_code = get_si_code((unsigned long)condition); /* Ok, finally something we can handle */ - send_sigtrap(tsk, regs, error_code); + send_sigtrap(tsk, regs, error_code, si_code); /* * Disable additional traps. They'll be re-enabled when diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index 56d6f114778..011d8e1fac6 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -941,7 +941,7 @@ asmlinkage void __kprobes do_debug(struct pt_regs *regs, tsk->thread.error_code = error_code; info.si_signo = SIGTRAP; info.si_errno = 0; - info.si_code = TRAP_BRKPT; + info.si_code = get_si_code(condition); info.si_addr = user_mode(regs) ? (void __user *)regs->ip : NULL; force_sig_info(SIGTRAP, &info, tsk); diff --git a/include/asm-generic/siginfo.h b/include/asm-generic/siginfo.h index 8786e01e0db..969570167e9 100644 --- a/include/asm-generic/siginfo.h +++ b/include/asm-generic/siginfo.h @@ -199,6 +199,8 @@ typedef struct siginfo { */ #define TRAP_BRKPT (__SI_FAULT|1) /* process breakpoint */ #define TRAP_TRACE (__SI_FAULT|2) /* process trace trap */ +#define TRAP_BRANCH (__SI_FAULT|3) /* process taken branch trap */ +#define TRAP_HWBKPT (__SI_FAULT|4) /* hardware breakpoint/watchpoint */ #define NSIGTRAP 2 /* diff --git a/include/asm-parisc/siginfo.h b/include/asm-parisc/siginfo.h index d4909f55fe3..d7034728f37 100644 --- a/include/asm-parisc/siginfo.h +++ b/include/asm-parisc/siginfo.h @@ -3,11 +3,6 @@ #include -/* - * SIGTRAP si_codes - */ -#define TRAP_BRANCH (__SI_FAULT|3) /* process taken branch trap */ -#define TRAP_HWBKPT (__SI_FAULT|4) /* hardware breakpoint or watchpoint */ #undef NSIGTRAP #define NSIGTRAP 4 diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h index fad80776991..c2f36827307 100644 --- a/include/asm-x86/ptrace.h +++ b/include/asm-x86/ptrace.h @@ -143,7 +143,7 @@ convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs); #ifdef CONFIG_X86_32 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, - int error_code); + int error_code, int si_code); #endif void signal_fault(struct pt_regs *regs, void __user *frame, char *where); diff --git a/include/asm-x86/traps.h b/include/asm-x86/traps.h index 2ccebc6fb0b..4b1e9040925 100644 --- a/include/asm-x86/traps.h +++ b/include/asm-x86/traps.h @@ -36,6 +36,16 @@ void do_invalid_op(struct pt_regs *, long); void do_general_protection(struct pt_regs *, long); void do_nmi(struct pt_regs *, long); +static inline int get_si_code(unsigned long condition) +{ + if (condition & DR_STEP) + return TRAP_TRACE; + else if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) + return TRAP_HWBKPT; + else + return TRAP_BRKPT; +} + extern int panic_on_unrecovered_nmi; extern int kstack_depth_to_print; -- cgit v1.2.3-70-g09d2 From e8d3f455de4f42d4bab2f6f1aeb2cf3bd18eb508 Mon Sep 17 00:00:00 2001 From: Srinivasa Ds Date: Tue, 23 Sep 2008 15:23:52 +0530 Subject: signals: demultiplexing SIGTRAP signal, fix fix build breakage, missing header file. Signed-off-by: Srinivasa DS Signed-off-by: Ingo Molnar --- include/asm-x86/traps.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/traps.h b/include/asm-x86/traps.h index 4b1e9040925..7a692baa51a 100644 --- a/include/asm-x86/traps.h +++ b/include/asm-x86/traps.h @@ -1,6 +1,8 @@ #ifndef ASM_X86__TRAPS_H #define ASM_X86__TRAPS_H +#include + /* Common in X86_32 and X86_64 */ asmlinkage void divide_error(void); asmlinkage void debug(void); -- cgit v1.2.3-70-g09d2 From 9f6ac57729724b58df81ca5dc005326759a806fe Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Wed, 24 Sep 2008 20:48:35 +0900 Subject: x86: export pci-nommu's alloc_coherent This patch exports nommu_alloc_coherent (renamed dma_generic_alloc_coherent). GART needs this function. Signed-off-by: FUJITA Tomonori Signed-off-by: Ingo Molnar --- arch/x86/kernel/pci-dma.c | 31 +++++++++++++++++++++++++++++++ arch/x86/kernel/pci-nommu.c | 39 +-------------------------------------- include/asm-x86/dma-mapping.h | 3 +++ 3 files changed, 35 insertions(+), 38 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 0a1408abcc6..4e612d20170 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -134,6 +134,37 @@ unsigned long iommu_num_pages(unsigned long addr, unsigned long len) EXPORT_SYMBOL(iommu_num_pages); #endif +void *dma_generic_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *dma_addr, gfp_t flag) +{ + unsigned long dma_mask; + struct page *page; + dma_addr_t addr; + + dma_mask = dma_alloc_coherent_mask(dev, flag); + + flag |= __GFP_ZERO; +again: + page = alloc_pages_node(dev_to_node(dev), flag, get_order(size)); + if (!page) + return NULL; + + addr = page_to_phys(page); + if (!is_buffer_dma_capable(dma_mask, addr, size)) { + __free_pages(page, get_order(size)); + + if (dma_mask < DMA_32BIT_MASK && !(flag & GFP_DMA)) { + flag = (flag & ~GFP_DMA32) | GFP_DMA; + goto again; + } + + return NULL; + } + + *dma_addr = addr; + return page_address(page); +} + /* * See for the iommu kernel parameter * documentation. diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c index 1c1c98a31d5..c70ab5a5d4c 100644 --- a/arch/x86/kernel/pci-nommu.c +++ b/arch/x86/kernel/pci-nommu.c @@ -72,43 +72,6 @@ static int nommu_map_sg(struct device *hwdev, struct scatterlist *sg, return nents; } -static void * -nommu_alloc_coherent(struct device *hwdev, size_t size, - dma_addr_t *dma_addr, gfp_t gfp) -{ - unsigned long dma_mask; - int node; - struct page *page; - dma_addr_t addr; - - dma_mask = dma_alloc_coherent_mask(hwdev, gfp); - - gfp |= __GFP_ZERO; - - node = dev_to_node(hwdev); -again: - page = alloc_pages_node(node, gfp, get_order(size)); - if (!page) - return NULL; - - addr = page_to_phys(page); - if (!is_buffer_dma_capable(dma_mask, addr, size) && !(gfp & GFP_DMA)) { - free_pages((unsigned long)page_address(page), get_order(size)); - gfp |= GFP_DMA; - goto again; - } - - if (check_addr("alloc_coherent", hwdev, addr, size)) { - *dma_addr = addr; - flush_write_buffers(); - return page_address(page); - } - - free_pages((unsigned long)page_address(page), get_order(size)); - - return NULL; -} - static void nommu_free_coherent(struct device *dev, size_t size, void *vaddr, dma_addr_t dma_addr) { @@ -116,7 +79,7 @@ static void nommu_free_coherent(struct device *dev, size_t size, void *vaddr, } struct dma_mapping_ops nommu_dma_ops = { - .alloc_coherent = nommu_alloc_coherent, + .alloc_coherent = dma_generic_alloc_coherent, .free_coherent = nommu_free_coherent, .map_single = nommu_map_single, .map_sg = nommu_map_sg, diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h index f408e6dd177..3b808e9bb72 100644 --- a/include/asm-x86/dma-mapping.h +++ b/include/asm-x86/dma-mapping.h @@ -89,6 +89,9 @@ static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) extern int dma_supported(struct device *hwdev, u64 mask); extern int dma_set_mask(struct device *dev, u64 mask); +extern void *dma_generic_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *dma_addr, gfp_t flag); + static inline dma_addr_t dma_map_single(struct device *hwdev, void *ptr, size_t size, int direction) -- cgit v1.2.3-70-g09d2 From 95dbf1dbe39ed336a3e72116c95cfa98dd3457e6 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Fri, 26 Sep 2008 10:36:42 -0500 Subject: kgdb, x86_64: gdb serial has BX and DX reversed The BX and DX registers in the gdb serial register packet need to be flipped for gdb to receive the correct data. Signed-off-by: Jason Wessel --- include/asm-x86/kgdb.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/kgdb.h b/include/asm-x86/kgdb.h index 484c47554f3..e86b3060bdc 100644 --- a/include/asm-x86/kgdb.h +++ b/include/asm-x86/kgdb.h @@ -42,9 +42,9 @@ enum regnames { #else /* ! CONFIG_X86_32 */ enum regnames { GDB_AX, /* 0 */ - GDB_DX, /* 1 */ + GDB_BX, /* 1 */ GDB_CX, /* 2 */ - GDB_BX, /* 3 */ + GDB_DX, /* 3 */ GDB_SI, /* 4 */ GDB_DI, /* 5 */ GDB_BP, /* 6 */ -- cgit v1.2.3-70-g09d2 From 703a1edcd1534468fc18f733c03bd91a65c8c6f0 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Fri, 26 Sep 2008 10:36:42 -0500 Subject: kgdb, x86_64: fix PS CS SS registers in gdb serial On x86_64 the gdb serial register structure defines the PS (also known as eflags), CS and SS registers as 4 bytes entities. This patch splits the x86_64 regnames enum into a 32 and 64 version to account for the 32 bit entities in the gdb serial packets. Also the program counter is properly filled in for the sleeping threads. Signed-off-by: Jason Wessel --- arch/x86/kernel/kgdb.c | 25 +++++++++++++++++++++---- include/asm-x86/kgdb.h | 20 +++++++++----------- 2 files changed, 30 insertions(+), 15 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index 00f7896c9a1..8282a213968 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -69,6 +69,9 @@ static int gdb_x86vector = -1; */ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) { +#ifndef CONFIG_X86_32 + u32 *gdb_regs32 = (u32 *)gdb_regs; +#endif gdb_regs[GDB_AX] = regs->ax; gdb_regs[GDB_BX] = regs->bx; gdb_regs[GDB_CX] = regs->cx; @@ -76,9 +79,9 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) gdb_regs[GDB_SI] = regs->si; gdb_regs[GDB_DI] = regs->di; gdb_regs[GDB_BP] = regs->bp; - gdb_regs[GDB_PS] = regs->flags; gdb_regs[GDB_PC] = regs->ip; #ifdef CONFIG_X86_32 + gdb_regs[GDB_PS] = regs->flags; gdb_regs[GDB_DS] = regs->ds; gdb_regs[GDB_ES] = regs->es; gdb_regs[GDB_CS] = regs->cs; @@ -94,6 +97,9 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) gdb_regs[GDB_R13] = regs->r13; gdb_regs[GDB_R14] = regs->r14; gdb_regs[GDB_R15] = regs->r15; + gdb_regs32[GDB_PS] = regs->flags; + gdb_regs32[GDB_CS] = regs->cs; + gdb_regs32[GDB_SS] = regs->ss; #endif gdb_regs[GDB_SP] = regs->sp; } @@ -112,6 +118,9 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) */ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) { +#ifndef CONFIG_X86_32 + u32 *gdb_regs32 = (u32 *)gdb_regs; +#endif gdb_regs[GDB_AX] = 0; gdb_regs[GDB_BX] = 0; gdb_regs[GDB_CX] = 0; @@ -129,8 +138,10 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) gdb_regs[GDB_FS] = 0xFFFF; gdb_regs[GDB_GS] = 0xFFFF; #else - gdb_regs[GDB_PS] = *(unsigned long *)(p->thread.sp + 8); - gdb_regs[GDB_PC] = 0; + gdb_regs32[GDB_PS] = *(unsigned long *)(p->thread.sp + 8); + gdb_regs32[GDB_CS] = __KERNEL_CS; + gdb_regs32[GDB_SS] = __KERNEL_DS; + gdb_regs[GDB_PC] = p->thread.ip; gdb_regs[GDB_R8] = 0; gdb_regs[GDB_R9] = 0; gdb_regs[GDB_R10] = 0; @@ -153,6 +164,9 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) */ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) { +#ifndef CONFIG_X86_32 + u32 *gdb_regs32 = (u32 *)gdb_regs; +#endif regs->ax = gdb_regs[GDB_AX]; regs->bx = gdb_regs[GDB_BX]; regs->cx = gdb_regs[GDB_CX]; @@ -160,9 +174,9 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) regs->si = gdb_regs[GDB_SI]; regs->di = gdb_regs[GDB_DI]; regs->bp = gdb_regs[GDB_BP]; - regs->flags = gdb_regs[GDB_PS]; regs->ip = gdb_regs[GDB_PC]; #ifdef CONFIG_X86_32 + regs->flags = gdb_regs[GDB_PS]; regs->ds = gdb_regs[GDB_DS]; regs->es = gdb_regs[GDB_ES]; regs->cs = gdb_regs[GDB_CS]; @@ -175,6 +189,9 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) regs->r13 = gdb_regs[GDB_R13]; regs->r14 = gdb_regs[GDB_R14]; regs->r15 = gdb_regs[GDB_R15]; + regs->flags = gdb_regs32[GDB_PS]; + regs->cs = gdb_regs32[GDB_CS]; + regs->ss = gdb_regs32[GDB_SS]; #endif } diff --git a/include/asm-x86/kgdb.h b/include/asm-x86/kgdb.h index e86b3060bdc..94d63db1036 100644 --- a/include/asm-x86/kgdb.h +++ b/include/asm-x86/kgdb.h @@ -39,8 +39,9 @@ enum regnames { GDB_FS, /* 14 */ GDB_GS, /* 15 */ }; +#define NUMREGBYTES ((GDB_GS+1)*4) #else /* ! CONFIG_X86_32 */ -enum regnames { +enum regnames64 { GDB_AX, /* 0 */ GDB_BX, /* 1 */ GDB_CX, /* 2 */ @@ -58,18 +59,15 @@ enum regnames { GDB_R14, /* 14 */ GDB_R15, /* 15 */ GDB_PC, /* 16 */ - GDB_PS, /* 17 */ }; -#endif /* CONFIG_X86_32 */ -/* - * Number of bytes of registers: - */ -#ifdef CONFIG_X86_32 -# define NUMREGBYTES 64 -#else -# define NUMREGBYTES ((GDB_PS+1)*8) -#endif +enum regnames32 { + GDB_PS = 34, + GDB_CS, + GDB_SS, +}; +#define NUMREGBYTES ((GDB_SS+1)*4) +#endif /* CONFIG_X86_32 */ static inline void arch_kgdb_breakpoint(void) { -- cgit v1.2.3-70-g09d2 From 237a62247c2879331986a300d6ab36ad21264c68 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 25 Sep 2008 12:13:53 +0200 Subject: x86/iommu: make GART driver checkpatch clean Signed-off-by: Joerg Roedel Signed-off-by: Ingo Molnar --- arch/x86/kernel/pci-gart_64.c | 17 +++++++++-------- include/asm-x86/gart.h | 2 ++ 2 files changed, 11 insertions(+), 8 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index b85a2f9bb34..aa569db73a2 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -27,8 +27,8 @@ #include #include #include +#include #include -#include #include #include #include @@ -175,7 +175,8 @@ static void dump_leak(void) iommu_leak_pages); for (i = 0; i < iommu_leak_pages; i += 2) { printk(KERN_DEBUG "%lu: ", iommu_pages-i); - printk_address((unsigned long) iommu_leak_tab[iommu_pages-i], 0); + printk_address((unsigned long) iommu_leak_tab[iommu_pages-i], + 0); printk(KERN_CONT "%c", (i+1)%2 == 0 ? '\n' : ' '); } printk(KERN_DEBUG "\n"); @@ -688,7 +689,8 @@ static __init int init_k8_gatt(struct agp_kern_info *info) if (!error) error = sysdev_register(&device_gart); if (error) - panic("Could not register gart_sysdev -- would corrupt data on next suspend"); + panic("Could not register gart_sysdev -- " + "would corrupt data on next suspend"); flush_gart(); @@ -710,8 +712,6 @@ static __init int init_k8_gatt(struct agp_kern_info *info) return -1; } -extern int agp_amd64_init(void); - static struct dma_mapping_ops gart_dma_ops = { .map_single = gart_map_single, .unmap_single = gart_unmap_single, @@ -777,8 +777,8 @@ void __init gart_iommu_init(void) (no_agp && init_k8_gatt(&info) < 0)) { if (max_pfn > MAX_DMA32_PFN) { printk(KERN_WARNING "More than 4GB of memory " - "but GART IOMMU not available.\n" - KERN_WARNING "falling back to iommu=soft.\n"); + "but GART IOMMU not available.\n"); + printk(KERN_WARNING "falling back to iommu=soft.\n"); } return; } @@ -868,7 +868,8 @@ void __init gart_parse_options(char *p) if (!strncmp(p, "leak", 4)) { leak_trace = 1; p += 4; - if (*p == '=') ++p; + if (*p == '=') + ++p; if (isdigit(*p) && get_option(&p, &arg)) iommu_leak_pages = arg; } diff --git a/include/asm-x86/gart.h b/include/asm-x86/gart.h index 3f62a83887f..cdf4f78e081 100644 --- a/include/asm-x86/gart.h +++ b/include/asm-x86/gart.h @@ -29,6 +29,8 @@ extern int fix_aperture; #define AMD64_GARTCACHECTL 0x9c #define AMD64_GARTEN (1<<0) +extern int agp_amd64_init(void); + static inline void enable_gart_translation(struct pci_dev *dev, u64 addr) { u32 tmp, ctl; -- cgit v1.2.3-70-g09d2 From 9b1568458a3ef006361710dc12848aec891883b5 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Mon, 29 Sep 2008 14:52:03 -0400 Subject: x86, debug printouts: IOMMU setup failures should not be KERN_ERR The number of BIOSes that have an option to enable the IOMMU, or fix anything about its configuration, is vanishingly small. There's no good reason to punish quiet boot for this. Signed-off-by: Adam Jackson Signed-off-by: Ingo Molnar --- arch/x86/kernel/aperture_64.c | 6 +++--- include/asm-x86/gart.h | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 44e21826db1..9a32b37ee2e 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -455,11 +455,11 @@ out: force_iommu || valid_agp || fallback_aper_force) { - printk(KERN_ERR + printk(KERN_INFO "Your BIOS doesn't leave a aperture memory hole\n"); - printk(KERN_ERR + printk(KERN_INFO "Please enable the IOMMU option in the BIOS setup\n"); - printk(KERN_ERR + printk(KERN_INFO "This costs you %d MB of RAM\n", 32 << fallback_aper_order); diff --git a/include/asm-x86/gart.h b/include/asm-x86/gart.h index 3f62a83887f..583031cf45f 100644 --- a/include/asm-x86/gart.h +++ b/include/asm-x86/gart.h @@ -52,15 +52,15 @@ static inline int aperture_valid(u64 aper_base, u32 aper_size, u32 min_size) return 0; if (aper_base + aper_size > 0x100000000ULL) { - printk(KERN_ERR "Aperture beyond 4GB. Ignoring.\n"); + printk(KERN_INFO "Aperture beyond 4GB. Ignoring.\n"); return 0; } if (e820_any_mapped(aper_base, aper_base + aper_size, E820_RAM)) { - printk(KERN_ERR "Aperture pointing to e820 RAM. Ignoring.\n"); + printk(KERN_INFO "Aperture pointing to e820 RAM. Ignoring.\n"); return 0; } if (aper_size < min_size) { - printk(KERN_ERR "Aperture too small (%d MB) than (%d MB)\n", + printk(KERN_INFO "Aperture too small (%d MB) than (%d MB)\n", aper_size>>20, min_size>>20); return 0; } -- cgit v1.2.3-70-g09d2 From 16dbc6c9616363fe53811abcbd935336dc0a0f01 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Thu, 2 Oct 2008 14:50:12 -0700 Subject: inotify: fix lock ordering wrt do_page_fault's mmap_sem Fix inotify lock order reversal with mmap_sem due to holding locks over copy_to_user. Signed-off-by: Nick Piggin Reported-by: "Daniel J Blueman" Tested-by: "Daniel J Blueman" Cc: Ingo Molnar Cc: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/inotify_user.c | 27 ++++++++++++++++++++------- include/asm-x86/uaccess_64.h | 1 + 2 files changed, 21 insertions(+), 7 deletions(-) (limited to 'include/asm-x86') diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 60249429a25..d85c7d931cd 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -323,7 +323,7 @@ out: } /* - * remove_kevent - cleans up and ultimately frees the given kevent + * remove_kevent - cleans up the given kevent * * Caller must hold dev->ev_mutex. */ @@ -334,7 +334,13 @@ static void remove_kevent(struct inotify_device *dev, dev->event_count--; dev->queue_size -= sizeof(struct inotify_event) + kevent->event.len; +} +/* + * free_kevent - frees the given kevent. + */ +static void free_kevent(struct inotify_kernel_event *kevent) +{ kfree(kevent->name); kmem_cache_free(event_cachep, kevent); } @@ -350,6 +356,7 @@ static void inotify_dev_event_dequeue(struct inotify_device *dev) struct inotify_kernel_event *kevent; kevent = inotify_dev_get_event(dev); remove_kevent(dev, kevent); + free_kevent(kevent); } } @@ -433,17 +440,15 @@ static ssize_t inotify_read(struct file *file, char __user *buf, dev = file->private_data; while (1) { - int events; prepare_to_wait(&dev->wq, &wait, TASK_INTERRUPTIBLE); mutex_lock(&dev->ev_mutex); - events = !list_empty(&dev->events); - mutex_unlock(&dev->ev_mutex); - if (events) { + if (!list_empty(&dev->events)) { ret = 0; break; } + mutex_unlock(&dev->ev_mutex); if (file->f_flags & O_NONBLOCK) { ret = -EAGAIN; @@ -462,7 +467,6 @@ static ssize_t inotify_read(struct file *file, char __user *buf, if (ret) return ret; - mutex_lock(&dev->ev_mutex); while (1) { struct inotify_kernel_event *kevent; @@ -481,6 +485,13 @@ static ssize_t inotify_read(struct file *file, char __user *buf, } break; } + remove_kevent(dev, kevent); + + /* + * Must perform the copy_to_user outside the mutex in order + * to avoid a lock order reversal with mmap_sem. + */ + mutex_unlock(&dev->ev_mutex); if (copy_to_user(buf, &kevent->event, event_size)) { ret = -EFAULT; @@ -498,7 +509,9 @@ static ssize_t inotify_read(struct file *file, char __user *buf, count -= kevent->event.len; } - remove_kevent(dev, kevent); + free_kevent(kevent); + + mutex_lock(&dev->ev_mutex); } mutex_unlock(&dev->ev_mutex); diff --git a/include/asm-x86/uaccess_64.h b/include/asm-x86/uaccess_64.h index 515d4dce96b..45806d60bcb 100644 --- a/include/asm-x86/uaccess_64.h +++ b/include/asm-x86/uaccess_64.h @@ -7,6 +7,7 @@ #include #include #include +#include #include /* -- cgit v1.2.3-70-g09d2 From ba31a5f88b6f907e715ff43db06403e12465b703 Mon Sep 17 00:00:00 2001 From: Paul Bolle Date: Sat, 4 Oct 2008 21:21:44 +0200 Subject: x86 setup: remove DEF_INITSEG and DEF_SETUPSEG Since v.2.6.23 DEF_INITSEG and DEF_SETUPSEG are unused. Commit c397368 ("Remove old i386 setup code") dropped their usage for i386. They did not return in the x86 tree. (Something similar must have happened for x86_64.) Remove these. Signed-off-by: Paul Bolle Signed-off-by: H. Peter Anvin --- include/asm-x86/boot.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/boot.h b/include/asm-x86/boot.h index 2faed7ecb09..7b287df4ab5 100644 --- a/include/asm-x86/boot.h +++ b/include/asm-x86/boot.h @@ -2,9 +2,7 @@ #define _ASM_BOOT_H /* Don't touch these, unless you really know what you're doing. */ -#define DEF_INITSEG 0x9000 #define DEF_SYSSEG 0x1000 -#define DEF_SETUPSEG 0x9020 #define DEF_SYSSIZE 0x7F00 /* Internal svga startup constants */ -- cgit v1.2.3-70-g09d2 From b2bc27314664c4d1a2f02e6f4cd0c32e4681d61e Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 23 Sep 2008 14:00:36 -0700 Subject: x86, cpa: rename PTE attribute macros for kernel direct mapping in early boot Signed-off-by: Suresh Siddha Cc: Suresh Siddha Cc: arjan@linux.intel.com Cc: venkatesh.pallipadi@intel.com Cc: jeremy@goop.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/head_32.S | 34 +++++++++++++++------------------- arch/x86/kernel/head_64.S | 4 ++-- include/asm-x86/pgtable.h | 11 +++++++++++ 3 files changed, 28 insertions(+), 21 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index a7010c3a377..e835b4eea70 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -172,10 +172,6 @@ num_subarch_entries = (. - subarch_entries) / 4 * * Note that the stack is not yet set up! */ -#define PTE_ATTR 0x007 /* PRESENT+RW+USER */ -#define PDE_ATTR 0x067 /* PRESENT+RW+USER+DIRTY+ACCESSED */ -#define PGD_ATTR 0x001 /* PRESENT (no other attributes) */ - default_entry: #ifdef CONFIG_X86_PAE @@ -196,9 +192,9 @@ default_entry: movl $pa(pg0), %edi movl %edi, pa(init_pg_tables_start) movl $pa(swapper_pg_pmd), %edx - movl $PTE_ATTR, %eax + movl $PTE_IDENT_ATTR, %eax 10: - leal PDE_ATTR(%edi),%ecx /* Create PMD entry */ + leal PDE_IDENT_ATTR(%edi),%ecx /* Create PMD entry */ movl %ecx,(%edx) /* Store PMD entry */ /* Upper half already zero */ addl $8,%edx @@ -215,7 +211,7 @@ default_entry: * End condition: we must map up to and including INIT_MAP_BEYOND_END * bytes beyond the end of our own page tables. */ - leal (INIT_MAP_BEYOND_END+PTE_ATTR)(%edi),%ebp + leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp cmpl %ebp,%eax jb 10b 1: @@ -224,7 +220,7 @@ default_entry: movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_ATTR,%eax + movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax movl %eax,pa(swapper_pg_pmd+0x1000*KPMDS-8) #else /* Not PAE */ @@ -233,9 +229,9 @@ page_pde_offset = (__PAGE_OFFSET >> 20); movl $pa(pg0), %edi movl %edi, pa(init_pg_tables_start) movl $pa(swapper_pg_dir), %edx - movl $PTE_ATTR, %eax + movl $PTE_IDENT_ATTR, %eax 10: - leal PDE_ATTR(%edi),%ecx /* Create PDE entry */ + leal PDE_IDENT_ATTR(%edi),%ecx /* Create PDE entry */ movl %ecx,(%edx) /* Store identity PDE entry */ movl %ecx,page_pde_offset(%edx) /* Store kernel PDE entry */ addl $4,%edx @@ -249,7 +245,7 @@ page_pde_offset = (__PAGE_OFFSET >> 20); * bytes beyond the end of our own page tables; the +0x007 is * the attribute bits */ - leal (INIT_MAP_BEYOND_END+PTE_ATTR)(%edi),%ebp + leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp cmpl %ebp,%eax jb 10b movl %edi,pa(init_pg_tables_end) @@ -257,7 +253,7 @@ page_pde_offset = (__PAGE_OFFSET >> 20); movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_ATTR,%eax + movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax movl %eax,pa(swapper_pg_dir+0xffc) #endif jmp 3f @@ -634,19 +630,19 @@ ENTRY(empty_zero_page) /* Page-aligned for the benefit of paravirt? */ .align PAGE_SIZE_asm ENTRY(swapper_pg_dir) - .long pa(swapper_pg_pmd+PGD_ATTR),0 /* low identity map */ + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 /* low identity map */ # if KPMDS == 3 - .long pa(swapper_pg_pmd+PGD_ATTR),0 - .long pa(swapper_pg_pmd+PGD_ATTR+0x1000),0 - .long pa(swapper_pg_pmd+PGD_ATTR+0x2000),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x2000),0 # elif KPMDS == 2 .long 0,0 - .long pa(swapper_pg_pmd+PGD_ATTR),0 - .long pa(swapper_pg_pmd+PGD_ATTR+0x1000),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0 # elif KPMDS == 1 .long 0,0 .long 0,0 - .long pa(swapper_pg_pmd+PGD_ATTR),0 + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 # else # error "Kernel PMDs should be 1, 2 or 3" # endif diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index db3280afe88..26cfdc1d7c7 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -110,7 +110,7 @@ startup_64: movq %rdi, %rax shrq $PMD_SHIFT, %rax andq $(PTRS_PER_PMD - 1), %rax - leaq __PAGE_KERNEL_LARGE_EXEC(%rdi), %rdx + leaq __PAGE_KERNEL_IDENT_LARGE_EXEC(%rdi), %rdx leaq level2_spare_pgt(%rip), %rbx movq %rdx, 0(%rbx, %rax, 8) ident_complete: @@ -374,7 +374,7 @@ NEXT_PAGE(level2_ident_pgt) /* Since I easily can, map the first 1G. * Don't set NX because code runs from these pages. */ - PMDS(0, __PAGE_KERNEL_LARGE_EXEC, PTRS_PER_PMD) + PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD) NEXT_PAGE(level2_kernel_pgt) /* diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index eccd52406ab..0ff73e7737a 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -132,6 +132,17 @@ #define __S110 PAGE_SHARED_EXEC #define __S111 PAGE_SHARED_EXEC +/* + * early identity mapping pte attrib macros. + */ +#ifdef CONFIG_X86_64 +#define __PAGE_KERNEL_IDENT_LARGE_EXEC __PAGE_KERNEL_LARGE_EXEC +#else +#define PTE_IDENT_ATTR 0x007 /* PRESENT+RW+USER */ +#define PDE_IDENT_ATTR 0x067 /* PRESENT+RW+USER+DIRTY+ACCESSED */ +#define PGD_IDENT_ATTR 0x001 /* PRESENT (no other attributes) */ +#endif + #ifndef __ASSEMBLY__ /* -- cgit v1.2.3-70-g09d2 From 3a85e770aa77e4f1a4096275c97b64c10cd7323e Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 23 Sep 2008 14:00:37 -0700 Subject: x86, cpa: remove USER permission from the very early identity mapping attribute remove USER from the PTE/PDE attributes for the very early identity mapping. We overwrite these mappings with KERNEL attribute later in the boot. Just being paranoid here as there is no need for USER bit to be set. If this breaks something(don't know the history), then we can simply drop this change. Signed-off-by: Suresh Siddha Cc: Suresh Siddha Cc: arjan@linux.intel.com Cc: venkatesh.pallipadi@intel.com Cc: jeremy@goop.org Signed-off-by: Ingo Molnar --- include/asm-x86/pgtable.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index 0ff73e7737a..bbf0f591d1b 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -138,8 +138,8 @@ #ifdef CONFIG_X86_64 #define __PAGE_KERNEL_IDENT_LARGE_EXEC __PAGE_KERNEL_LARGE_EXEC #else -#define PTE_IDENT_ATTR 0x007 /* PRESENT+RW+USER */ -#define PDE_IDENT_ATTR 0x067 /* PRESENT+RW+USER+DIRTY+ACCESSED */ +#define PTE_IDENT_ATTR 0x003 /* PRESENT+RW */ +#define PDE_IDENT_ATTR 0x063 /* PRESENT+RW+DIRTY+ACCESSED */ #define PGD_IDENT_ATTR 0x001 /* PRESENT (no other attributes) */ #endif -- cgit v1.2.3-70-g09d2 From 8311eb84bf842d345f543f4c62ca2b6ea26f638c Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 23 Sep 2008 14:00:41 -0700 Subject: x86, cpa: remove cpa pool code Interrupt context no longer splits large page in cpa(). So we can do away with cpa memory pool code. Signed-off-by: Suresh Siddha Cc: Suresh Siddha Cc: arjan@linux.intel.com Cc: venkatesh.pallipadi@intel.com Cc: jeremy@goop.org Signed-off-by: Ingo Molnar --- arch/x86/mm/init_32.c | 1 - arch/x86/mm/init_64.c | 2 - arch/x86/mm/pageattr.c | 157 ++----------------------------------------- include/asm-x86/cacheflush.h | 2 - 4 files changed, 5 insertions(+), 157 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 44ccb028c35..74780800e7e 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -1051,7 +1051,6 @@ void __init mem_init(void) if (boot_cpu_data.wp_works_ok < 0) test_wp_bit(); - cpa_init(); save_pg_dir(); zap_low_mappings(); } diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 9d7587ac1eb..f54a4d97530 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -889,8 +889,6 @@ void __init mem_init(void) reservedpages << (PAGE_SHIFT-10), datasize >> 10, initsize >> 10); - - cpa_init(); } void free_init_pages(char *what, unsigned long begin, unsigned long end) diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 162812b05d2..f5e8663c0f7 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -447,114 +447,17 @@ out_unlock: return do_split; } -static LIST_HEAD(page_pool); -static unsigned long pool_size, pool_pages, pool_low; -static unsigned long pool_used, pool_failed; - -static void cpa_fill_pool(struct page **ret) -{ - gfp_t gfp = GFP_KERNEL; - unsigned long flags; - struct page *p; - - /* - * Avoid recursion (on debug-pagealloc) and also signal - * our priority to get to these pagetables: - */ - if (current->flags & PF_MEMALLOC) - return; - current->flags |= PF_MEMALLOC; - - /* - * Allocate atomically from atomic contexts: - */ - if (in_atomic() || irqs_disabled() || debug_pagealloc) - gfp = GFP_ATOMIC | __GFP_NORETRY | __GFP_NOWARN; - - while (pool_pages < pool_size || (ret && !*ret)) { - p = alloc_pages(gfp, 0); - if (!p) { - pool_failed++; - break; - } - /* - * If the call site needs a page right now, provide it: - */ - if (ret && !*ret) { - *ret = p; - continue; - } - spin_lock_irqsave(&pgd_lock, flags); - list_add(&p->lru, &page_pool); - pool_pages++; - spin_unlock_irqrestore(&pgd_lock, flags); - } - - current->flags &= ~PF_MEMALLOC; -} - -#define SHIFT_MB (20 - PAGE_SHIFT) -#define ROUND_MB_GB ((1 << 10) - 1) -#define SHIFT_MB_GB 10 -#define POOL_PAGES_PER_GB 16 - -void __init cpa_init(void) -{ - struct sysinfo si; - unsigned long gb; - - si_meminfo(&si); - /* - * Calculate the number of pool pages: - * - * Convert totalram (nr of pages) to MiB and round to the next - * GiB. Shift MiB to Gib and multiply the result by - * POOL_PAGES_PER_GB: - */ - if (debug_pagealloc) { - gb = ((si.totalram >> SHIFT_MB) + ROUND_MB_GB) >> SHIFT_MB_GB; - pool_size = POOL_PAGES_PER_GB * gb; - } else { - pool_size = 1; - } - pool_low = pool_size; - - cpa_fill_pool(NULL); - printk(KERN_DEBUG - "CPA: page pool initialized %lu of %lu pages preallocated\n", - pool_pages, pool_size); -} - static int split_large_page(pte_t *kpte, unsigned long address) { unsigned long flags, pfn, pfninc = 1; unsigned int i, level; pte_t *pbase, *tmp; pgprot_t ref_prot; - struct page *base; + struct page *base = alloc_pages(GFP_KERNEL, 0); + if (!base) + return -ENOMEM; - /* - * Get a page from the pool. The pool list is protected by the - * pgd_lock, which we have to take anyway for the split - * operation: - */ spin_lock_irqsave(&pgd_lock, flags); - if (list_empty(&page_pool)) { - spin_unlock_irqrestore(&pgd_lock, flags); - base = NULL; - cpa_fill_pool(&base); - if (!base) - return -ENOMEM; - spin_lock_irqsave(&pgd_lock, flags); - } else { - base = list_first_entry(&page_pool, struct page, lru); - list_del(&base->lru); - pool_pages--; - - if (pool_pages < pool_low) - pool_low = pool_pages; - } - /* * Check for races, another CPU might have split this page * up for us already: @@ -611,11 +514,8 @@ out_unlock: * If we dropped out via the lookup_address check under * pgd_lock then stick the page back into the pool: */ - if (base) { - list_add(&base->lru, &page_pool); - pool_pages++; - } else - pool_used++; + if (base) + __free_page(base); spin_unlock_irqrestore(&pgd_lock, flags); return 0; @@ -899,8 +799,6 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, cpa_flush_all(cache); out: - cpa_fill_pool(NULL); - return ret; } @@ -1178,53 +1076,8 @@ void kernel_map_pages(struct page *page, int numpages, int enable) * but that can deadlock->flush only current cpu: */ __flush_tlb_all(); - - /* - * Try to refill the page pool here. We can do this only after - * the tlb flush. - */ - cpa_fill_pool(NULL); -} - -#ifdef CONFIG_DEBUG_FS -static int dpa_show(struct seq_file *m, void *v) -{ - seq_puts(m, "DEBUG_PAGEALLOC\n"); - seq_printf(m, "pool_size : %lu\n", pool_size); - seq_printf(m, "pool_pages : %lu\n", pool_pages); - seq_printf(m, "pool_low : %lu\n", pool_low); - seq_printf(m, "pool_used : %lu\n", pool_used); - seq_printf(m, "pool_failed : %lu\n", pool_failed); - - return 0; -} - -static int dpa_open(struct inode *inode, struct file *filp) -{ - return single_open(filp, dpa_show, NULL); } -static const struct file_operations dpa_fops = { - .open = dpa_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static int __init debug_pagealloc_proc_init(void) -{ - struct dentry *de; - - de = debugfs_create_file("debug_pagealloc", 0600, NULL, NULL, - &dpa_fops); - if (!de) - return -ENOMEM; - - return 0; -} -__initcall(debug_pagealloc_proc_init); -#endif - #ifdef CONFIG_HIBERNATION bool kernel_page_present(struct page *page) diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h index 0a5f71817b3..8e205a13125 100644 --- a/include/asm-x86/cacheflush.h +++ b/include/asm-x86/cacheflush.h @@ -99,8 +99,6 @@ int set_pages_rw(struct page *page, int numpages); void clflush_cache_range(void *addr, unsigned int size); -void cpa_init(void); - #ifdef CONFIG_DEBUG_RODATA void mark_rodata_ro(void); extern const int rodata_test_data; -- cgit v1.2.3-70-g09d2 From 9542ada803198e6eba29d3289abb39ea82047b92 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Wed, 24 Sep 2008 08:53:33 -0700 Subject: x86: track memtype for RAM in page struct Track the memtype for RAM pages in page struct instead of using the memtype list. This avoids the explosion in the number of entries in memtype list (of the order of 20,000 with AGP) and makes the PAT tracking simpler. We are using PG_arch_1 bit in page->flags. We still use the memtype list for non RAM pages. Signed-off-by: Suresh Siddha Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- arch/x86/mm/ioremap.c | 19 ++++++++++ arch/x86/mm/pat.c | 83 ++++++++++++++++++++++++++++++++++++++++++++ include/asm-x86/cacheflush.h | 2 ++ include/asm-x86/page.h | 1 + 4 files changed, 105 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index d4b6e6a29ae..d03c461e045 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -83,6 +83,25 @@ int page_is_ram(unsigned long pagenr) return 0; } +int pagerange_is_ram(unsigned long start, unsigned long end) +{ + int ram_page = 0, not_rampage = 0; + unsigned long page_nr; + + for (page_nr = (start >> PAGE_SHIFT); page_nr < (end >> PAGE_SHIFT); + ++page_nr) { + if (page_is_ram(page_nr)) + ram_page = 1; + else + not_rampage = 1; + + if (ram_page == not_rampage) + return -1; + } + + return ram_page; +} + /* * Fix up the linear direct mapping of the kernel to avoid cache attribute * conflicts. diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index f049b1d6ebd..aceb6c7c6db 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -210,6 +210,75 @@ static int chk_conflict(struct memtype *new, struct memtype *entry, static struct memtype *cached_entry; static u64 cached_start; +/* + * RED-PEN: TODO: Add PageReserved() check as well here, + * once we add SetPageReserved() to all the drivers using + * set_memory_* or set_pages_*. + * + * This will help prevent accidentally freeing pages + * before setting the attribute back to WB. + */ + +/* + * For RAM pages, mark the pages as non WB memory type using + * PageNonWB (PG_arch_1). We allow only one set_memory_uc() or + * set_memory_wc() on a RAM page at a time before marking it as WB again. + * This is ok, because only one driver will be owning the page and + * doing set_memory_*() calls. + * + * For now, we use PageNonWB to track that the RAM page is being mapped + * as non WB. In future, we will have to use one more flag + * (or some other mechanism in page_struct) to distinguish between + * UC and WC mapping. + */ +static int reserve_ram_pages_type(u64 start, u64 end, unsigned long req_type, + unsigned long *new_type) +{ + struct page *page; + u64 pfn, end_pfn; + + for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) { + page = pfn_to_page(pfn); + if (page_mapped(page) || PageNonWB(page)) + goto out; + + SetPageNonWB(page); + } + return 0; + +out: + end_pfn = pfn; + for (pfn = (start >> PAGE_SHIFT); pfn < end_pfn; ++pfn) { + page = pfn_to_page(pfn); + ClearPageNonWB(page); + } + + return -EINVAL; +} + +static int free_ram_pages_type(u64 start, u64 end) +{ + struct page *page; + u64 pfn, end_pfn; + + for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) { + page = pfn_to_page(pfn); + if (page_mapped(page) || !PageNonWB(page)) + goto out; + + ClearPageNonWB(page); + } + return 0; + +out: + end_pfn = pfn; + for (pfn = (start >> PAGE_SHIFT); pfn < end_pfn; ++pfn) { + page = pfn_to_page(pfn); + SetPageNonWB(page); + } + return -EINVAL; +} + /* * req_type typically has one of the: * - _PAGE_CACHE_WB @@ -232,6 +301,7 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, unsigned long actual_type; struct list_head *where; int err = 0; + int is_range_ram; BUG_ON(start >= end); /* end is exclusive */ @@ -270,6 +340,12 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, actual_type = pat_x_mtrr_type(start, end, req_type & _PAGE_CACHE_MASK); + is_range_ram = pagerange_is_ram(start, end); + if (is_range_ram == 1) + return reserve_ram_pages_type(start, end, req_type, new_type); + else if (is_range_ram < 0) + return -EINVAL; + new = kmalloc(sizeof(struct memtype), GFP_KERNEL); if (!new) return -ENOMEM; @@ -358,6 +434,7 @@ int free_memtype(u64 start, u64 end) { struct memtype *entry; int err = -EINVAL; + int is_range_ram; if (!pat_enabled) return 0; @@ -366,6 +443,12 @@ int free_memtype(u64 start, u64 end) if (is_ISA_range(start, end - 1)) return 0; + is_range_ram = pagerange_is_ram(start, end); + if (is_range_ram == 1) + return free_ram_pages_type(start, end); + else if (is_range_ram < 0) + return -EINVAL; + spin_lock(&memtype_lock); list_for_each_entry(entry, &memtype_list, nd) { if (entry->start == start && entry->end == end) { diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h index 8e205a13125..092b9b4eb00 100644 --- a/include/asm-x86/cacheflush.h +++ b/include/asm-x86/cacheflush.h @@ -24,6 +24,8 @@ #define copy_from_user_page(vma, page, vaddr, dst, src, len) \ memcpy((dst), (src), (len)) +#define PG_non_WB PG_arch_1 +PAGEFLAG(NonWB, non_WB) /* * The set_memory_* API can be used to change various attributes of a virtual diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h index 49982110e4d..3407ac12ba3 100644 --- a/include/asm-x86/page.h +++ b/include/asm-x86/page.h @@ -57,6 +57,7 @@ typedef struct { pgdval_t pgd; } pgd_t; typedef struct { pgprotval_t pgprot; } pgprot_t; extern int page_is_ram(unsigned long pagenr); +extern int pagerange_is_ram(unsigned long start, unsigned long end); extern int devmem_is_allowed(unsigned long pagenr); extern void map_devmem(unsigned long pfn, unsigned long size, pgprot_t vma_prot); -- cgit v1.2.3-70-g09d2 From 46eaa6702016e3ac9a188172a2c309d6ca1be1cd Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sun, 12 Oct 2008 15:06:29 +0200 Subject: x86: memory corruption check - cleanup Move the prototypes from the generic kernel.h header to the more appropriate include/asm-x86/bios_ebda.h header file. Also, remove the check from the power management code - this is a pure x86 matter for now. Signed-off-by: Ingo Molnar --- arch/x86/mm/init_32.c | 1 + arch/x86/mm/init_64.c | 1 + drivers/base/power/main.c | 1 - include/asm-x86/bios_ebda.h | 17 +++++++++++++++++ include/linux/kernel.h | 17 ----------------- 5 files changed, 19 insertions(+), 18 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 7e05462ffb1..bbe044dbe01 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -31,6 +31,7 @@ #include #include +#include #include #include #include diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index d84d3e91d34..3e10054c573 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -31,6 +31,7 @@ #include #include +#include #include #include #include diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index bf6d3554e50..273a944d404 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -254,7 +254,6 @@ static char *pm_verb(int event) static void pm_dev_dbg(struct device *dev, pm_message_t state, char *info) { - check_for_bios_corruption(); dev_dbg(dev, "%s%s%s\n", info, pm_verb(state.event), ((state.event & PM_EVENT_SLEEP) && device_may_wakeup(dev)) ? ", may wakeup" : ""); diff --git a/include/asm-x86/bios_ebda.h b/include/asm-x86/bios_ebda.h index ec42ed87459..79b4b88505d 100644 --- a/include/asm-x86/bios_ebda.h +++ b/include/asm-x86/bios_ebda.h @@ -16,4 +16,21 @@ static inline unsigned int get_bios_ebda(void) void reserve_ebda_region(void); +#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION +/* + * This is obviously not a great place for this, but we want to be + * able to scatter it around anywhere in the kernel. + */ +void check_for_bios_corruption(void); +void start_periodic_check_for_corruption(void); +#else +static inline void check_for_bios_corruption(void) +{ +} + +static inline void start_periodic_check_for_corruption(void) +{ +} +#endif + #endif /* ASM_X86__BIOS_EBDA_H */ diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 50873b21178..2651f805ba6 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -240,23 +240,6 @@ extern const char *print_tainted(void); extern void add_taint(unsigned); extern int root_mountflags; -#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION -/* - * This is obviously not a great place for this, but we want to be - * able to scatter it around anywhere in the kernel. - */ -void check_for_bios_corruption(void); -void start_periodic_check_for_corruption(void); -#else -static inline void check_for_bios_corruption(void) -{ -} - -static inline void start_periodic_check_for_corruption(void) -{ -} -#endif - /* Values used for system_state */ extern enum system_states { SYSTEM_BOOTING, -- cgit v1.2.3-70-g09d2 From 9f8305fe164077bcef42034c56275c8208042e88 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Fri, 29 Aug 2008 13:14:30 +0100 Subject: x86-64: reduce boot fixmap space Just like for 32-bit - as 256 entries are needed, aligning to a 256-entry boundary is sufficient and still guarantees the single pte table requirement. Likewise move up __end_of_permanent_fixed_addresses, to match 32-bit. Signed-off-by: Jan Beulich Signed-off-by: Ingo Molnar --- include/asm-x86/fixmap_64.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/fixmap_64.h b/include/asm-x86/fixmap_64.h index dafb24bc042..33df64fd0e1 100644 --- a/include/asm-x86/fixmap_64.h +++ b/include/asm-x86/fixmap_64.h @@ -49,6 +49,7 @@ enum fixed_addresses { #ifdef CONFIG_PARAVIRT FIX_PARAVIRT_BOOTMAP, #endif + __end_of_permanent_fixed_addresses, #ifdef CONFIG_ACPI FIX_ACPI_BEGIN, FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1, @@ -56,18 +57,17 @@ enum fixed_addresses { #ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT FIX_OHCI1394_BASE, #endif - __end_of_permanent_fixed_addresses, /* * 256 temporary boot-time mappings, used by early_ioremap(), * before ioremap() is functional. * - * We round it up to the next 512 pages boundary so that we + * We round it up to the next 256 pages boundary so that we * can have a single pgd entry and a single pte table: */ #define NR_FIX_BTMAPS 64 #define FIX_BTMAPS_NESTING 4 - FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 512 - - (__end_of_permanent_fixed_addresses & 511), + FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 - + (__end_of_permanent_fixed_addresses & 255), FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1, __end_of_fixed_addresses }; -- cgit v1.2.3-70-g09d2 From 762db4347060c5d23e9675846e02f7d95d7f944f Mon Sep 17 00:00:00 2001 From: Alexander van Heukelum Date: Tue, 9 Sep 2008 21:55:55 +0200 Subject: i386: remove kprobes' restore_interrupts in favour of conditional_sti x86_64 uses a helper function conditional_sti in traps_64.c which is equal to restore_interrupts in kprobes.h. The only user of restore_interrupts is in traps_32.c. Introduce conditional_sti for i386 and remove restore_interrupts. Signed-off-by: Alexander van Heukelum Signed-off-by: Ingo Molnar --- arch/x86/kernel/traps_32.c | 8 +++++++- include/asm-x86/kprobes.h | 9 --------- 2 files changed, 7 insertions(+), 10 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index 0429c5de5ea..203b863ac2a 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -84,6 +84,12 @@ static unsigned int code_bytes = 64; static int ignore_nmis; static int die_counter; +static inline void conditional_sti(struct pt_regs *regs) +{ + if (regs->flags & X86_EFLAGS_IF) + local_irq_enable(); +} + void printk_address(unsigned long address, int reliable) { #ifdef CONFIG_KALLSYMS @@ -859,7 +865,7 @@ void __kprobes do_int3(struct pt_regs *regs, long error_code) * This is an interrupt gate, because kprobes wants interrupts * disabled. Normal trap handlers don't. */ - restore_interrupts(regs); + conditional_sti(regs); do_trap(3, SIGTRAP, "int3", 1, regs, error_code, NULL); } diff --git a/include/asm-x86/kprobes.h b/include/asm-x86/kprobes.h index bd8407863c1..8a0748d0103 100644 --- a/include/asm-x86/kprobes.h +++ b/include/asm-x86/kprobes.h @@ -82,15 +82,6 @@ struct kprobe_ctlblk { struct prev_kprobe prev_kprobe; }; -/* trap3/1 are intr gates for kprobes. So, restore the status of IF, - * if necessary, before executing the original int3/1 (trap) handler. - */ -static inline void restore_interrupts(struct pt_regs *regs) -{ - if (regs->flags & X86_EFLAGS_IF) - local_irq_enable(); -} - extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr); extern int kprobe_exceptions_notify(struct notifier_block *self, unsigned long val, void *data); -- cgit v1.2.3-70-g09d2 From be43d72835ba610e4af274f2d123b26f66f4f7ed Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Sun, 7 Sep 2008 15:21:13 -0700 Subject: x86: add _PAGE_IOMAP pte flag for IO mappings Use one of the software-defined PTE bits to indicate that a mapping is intended for an IO address. On native hardware this is irrelevent, since a physical address is a physical address. But in a virtual environment, physical addresses are also virtualized, so there needs to be some way to distinguish between pseudo-physical addresses and actual hardware addresses; _PAGE_IOMAP indicates this intent. By default, __supported_pte_mask masks out _PAGE_IOMAP, so it doesn't even appear in the final pagetable. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/mm/init_32.c | 2 +- arch/x86/mm/init_64.c | 2 +- arch/x86/mm/ioremap.c | 8 ++++---- include/asm-x86/pgtable.h | 14 ++++++++++++-- 4 files changed, 18 insertions(+), 8 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index bbe044dbe01..8396868e82c 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -558,7 +558,7 @@ void zap_low_mappings(void) int nx_enabled; -pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL); +pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL | _PAGE_IOMAP); EXPORT_SYMBOL_GPL(__supported_pte_mask); #ifdef CONFIG_X86_PAE diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 3e10054c573..dec5c775e92 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -89,7 +89,7 @@ early_param("gbpages", parse_direct_gbpages_on); int after_bootmem; -unsigned long __supported_pte_mask __read_mostly = ~0UL; +pteval_t __supported_pte_mask __read_mostly = ~_PAGE_IOMAP; EXPORT_SYMBOL_GPL(__supported_pte_mask); static int do_not_nx __cpuinitdata; diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 8cbeda15cd2..43c3b6896cd 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -242,16 +242,16 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr, switch (prot_val) { case _PAGE_CACHE_UC: default: - prot = PAGE_KERNEL_NOCACHE; + prot = PAGE_KERNEL_IO_NOCACHE; break; case _PAGE_CACHE_UC_MINUS: - prot = PAGE_KERNEL_UC_MINUS; + prot = PAGE_KERNEL_IO_UC_MINUS; break; case _PAGE_CACHE_WC: - prot = PAGE_KERNEL_WC; + prot = PAGE_KERNEL_IO_WC; break; case _PAGE_CACHE_WB: - prot = PAGE_KERNEL; + prot = PAGE_KERNEL_IO; break; } diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index ed932453ef2..81805403b64 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -15,7 +15,7 @@ #define _PAGE_BIT_PAT 7 /* on 4KB pages */ #define _PAGE_BIT_GLOBAL 8 /* Global TLB entry PPro+ */ #define _PAGE_BIT_UNUSED1 9 /* available for programmer */ -#define _PAGE_BIT_UNUSED2 10 +#define _PAGE_BIT_IOMAP 10 /* flag used to indicate IO mapping */ #define _PAGE_BIT_UNUSED3 11 #define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */ #define _PAGE_BIT_SPECIAL _PAGE_BIT_UNUSED1 @@ -32,7 +32,7 @@ #define _PAGE_PSE (_AT(pteval_t, 1) << _PAGE_BIT_PSE) #define _PAGE_GLOBAL (_AT(pteval_t, 1) << _PAGE_BIT_GLOBAL) #define _PAGE_UNUSED1 (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED1) -#define _PAGE_UNUSED2 (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED2) +#define _PAGE_IOMAP (_AT(pteval_t, 1) << _PAGE_BIT_IOMAP) #define _PAGE_UNUSED3 (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED3) #define _PAGE_PAT (_AT(pteval_t, 1) << _PAGE_BIT_PAT) #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE) @@ -99,6 +99,11 @@ #define __PAGE_KERNEL_LARGE_NOCACHE (__PAGE_KERNEL | _PAGE_CACHE_UC | _PAGE_PSE) #define __PAGE_KERNEL_LARGE_EXEC (__PAGE_KERNEL_EXEC | _PAGE_PSE) +#define __PAGE_KERNEL_IO (__PAGE_KERNEL | _PAGE_IOMAP) +#define __PAGE_KERNEL_IO_NOCACHE (__PAGE_KERNEL_NOCACHE | _PAGE_IOMAP) +#define __PAGE_KERNEL_IO_UC_MINUS (__PAGE_KERNEL_UC_MINUS | _PAGE_IOMAP) +#define __PAGE_KERNEL_IO_WC (__PAGE_KERNEL_WC | _PAGE_IOMAP) + #define PAGE_KERNEL __pgprot(__PAGE_KERNEL) #define PAGE_KERNEL_RO __pgprot(__PAGE_KERNEL_RO) #define PAGE_KERNEL_EXEC __pgprot(__PAGE_KERNEL_EXEC) @@ -113,6 +118,11 @@ #define PAGE_KERNEL_VSYSCALL __pgprot(__PAGE_KERNEL_VSYSCALL) #define PAGE_KERNEL_VSYSCALL_NOCACHE __pgprot(__PAGE_KERNEL_VSYSCALL_NOCACHE) +#define PAGE_KERNEL_IO __pgprot(__PAGE_KERNEL_IO) +#define PAGE_KERNEL_IO_NOCACHE __pgprot(__PAGE_KERNEL_IO_NOCACHE) +#define PAGE_KERNEL_IO_UC_MINUS __pgprot(__PAGE_KERNEL_IO_UC_MINUS) +#define PAGE_KERNEL_IO_WC __pgprot(__PAGE_KERNEL_IO_WC) + /* xwr */ #define __P000 PAGE_NONE #define __P001 PAGE_READONLY -- cgit v1.2.3-70-g09d2 From efc9eb20b2f5125642fc37a1dbabadc3ce5d321c Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Sun, 7 Sep 2008 15:21:14 -0700 Subject: x86: remove duplicate early_ioremap declarations early_ioremap() is redeclared in several places; remove them. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- include/asm-x86/io.h | 14 -------------- include/asm-x86/io_64.h | 3 --- 2 files changed, 17 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/io.h b/include/asm-x86/io.h index 72b7719523b..e091f3949ec 100644 --- a/include/asm-x86/io.h +++ b/include/asm-x86/io.h @@ -5,20 +5,6 @@ #include -/* - * early_ioremap() and early_iounmap() are for temporary early boot-time - * mappings, before the real ioremap() is functional. - * A boot-time mapping is currently limited to at most 16 pages. - */ -#ifndef __ASSEMBLY__ -extern void early_ioremap_init(void); -extern void early_ioremap_clear(void); -extern void early_ioremap_reset(void); -extern void *early_ioremap(unsigned long offset, unsigned long size); -extern void early_iounmap(void *addr, unsigned long size); -extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys); -#endif - #define build_mmio_read(name, size, type, reg, barrier) \ static inline type name(const volatile void __iomem *addr) \ { type ret; asm volatile("mov" size " %1,%0":reg (ret) \ diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h index 64429e9431a..ee6e086b7df 100644 --- a/include/asm-x86/io_64.h +++ b/include/asm-x86/io_64.h @@ -165,9 +165,6 @@ static inline void *phys_to_virt(unsigned long address) #include -extern void *early_ioremap(unsigned long addr, unsigned long size); -extern void early_iounmap(void *addr, unsigned long size); - /* * This one maps high address device memory and turns off caching for that area. * it's useful if some control registers are in such an area and write combining -- cgit v1.2.3-70-g09d2 From 1494177942b23b7094ca291d37e6f6263fa60fdd Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Sun, 7 Sep 2008 15:21:15 -0700 Subject: x86: add early_memremap() early_ioremap() is also used to map normal memory when constructing the linear memory mapping. However, since we sometimes need to be able to distinguish between actual IO mappings and normal memory mappings, add a early_memremap() call, which maps with PAGE_KERNEL (as opposed to PAGE_KERNEL_IO for early_ioremap()), and use it when constructing pagetables. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/mm/init_64.c | 2 +- arch/x86/mm/ioremap.c | 22 +++++++++++++++++----- include/asm-x86/io.h | 1 + 3 files changed, 19 insertions(+), 6 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index dec5c775e92..5beb8968345 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -313,7 +313,7 @@ static __ref void *alloc_low_page(unsigned long *phys) if (pfn >= table_top) panic("alloc_low_page: ran out of memory"); - adr = early_ioremap(pfn * PAGE_SIZE, PAGE_SIZE); + adr = early_memremap(pfn * PAGE_SIZE, PAGE_SIZE); memset(adr, 0, PAGE_SIZE); *phys = pfn * PAGE_SIZE; return adr; diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 43c3b6896cd..7fb737c6b54 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -568,12 +568,12 @@ static void __init __early_set_fixmap(enum fixed_addresses idx, } static inline void __init early_set_fixmap(enum fixed_addresses idx, - unsigned long phys) + unsigned long phys, pgprot_t prot) { if (after_paging_init) - set_fixmap(idx, phys); + __set_fixmap(idx, phys, prot); else - __early_set_fixmap(idx, phys, PAGE_KERNEL); + __early_set_fixmap(idx, phys, prot); } static inline void __init early_clear_fixmap(enum fixed_addresses idx) @@ -601,7 +601,7 @@ static int __init check_early_ioremap_leak(void) } late_initcall(check_early_ioremap_leak); -void __init *early_ioremap(unsigned long phys_addr, unsigned long size) +static void __init *__early_ioremap(unsigned long phys_addr, unsigned long size, pgprot_t prot) { unsigned long offset, last_addr; unsigned int nrpages, nesting; @@ -650,7 +650,7 @@ void __init *early_ioremap(unsigned long phys_addr, unsigned long size) idx0 = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*nesting; idx = idx0; while (nrpages > 0) { - early_set_fixmap(idx, phys_addr); + early_set_fixmap(idx, phys_addr, prot); phys_addr += PAGE_SIZE; --idx; --nrpages; @@ -661,6 +661,18 @@ void __init *early_ioremap(unsigned long phys_addr, unsigned long size) return (void *) (offset + fix_to_virt(idx0)); } +/* Remap an IO device */ +void __init *early_ioremap(unsigned long phys_addr, unsigned long size) +{ + return __early_ioremap(phys_addr, size, PAGE_KERNEL_IO); +} + +/* Remap memory */ +void __init *early_memremap(unsigned long phys_addr, unsigned long size) +{ + return __early_ioremap(phys_addr, size, PAGE_KERNEL); +} + void __init early_iounmap(void *addr, unsigned long size) { unsigned long virt_addr; diff --git a/include/asm-x86/io.h b/include/asm-x86/io.h index e091f3949ec..a233f835e0b 100644 --- a/include/asm-x86/io.h +++ b/include/asm-x86/io.h @@ -83,6 +83,7 @@ extern void early_ioremap_init(void); extern void early_ioremap_clear(void); extern void early_ioremap_reset(void); extern void *early_ioremap(unsigned long offset, unsigned long size); +extern void *early_memremap(unsigned long offset, unsigned long size); extern void early_iounmap(void *addr, unsigned long size); extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys); -- cgit v1.2.3-70-g09d2 From a73aaedd95703bd49f4c3f9df06fb7b7373ba905 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 14 Sep 2008 02:33:14 -0700 Subject: x86: check dsdt before find oem table for es7000, v2 v2: use __acpi_unmap_table() Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/es7000_32.c | 28 +++++++++++++++++++++++----- arch/x86/mach-generic/es7000.c | 20 +++++++++++++++----- include/asm-x86/es7000/mpparse.h | 1 + 3 files changed, 39 insertions(+), 10 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/es7000_32.c b/arch/x86/kernel/es7000_32.c index 849e5cd485b..f454c78fcef 100644 --- a/arch/x86/kernel/es7000_32.c +++ b/arch/x86/kernel/es7000_32.c @@ -109,6 +109,7 @@ struct oem_table { }; extern int find_unisys_acpi_oem_table(unsigned long *oem_addr); +extern void unmap_unisys_acpi_oem_table(unsigned long oem_addr); #endif struct mip_reg { @@ -243,21 +244,38 @@ parse_unisys_oem (char *oemptr) } #ifdef CONFIG_ACPI -int __init -find_unisys_acpi_oem_table(unsigned long *oem_addr) +static unsigned long oem_addrX; +static unsigned long oem_size; +int __init find_unisys_acpi_oem_table(unsigned long *oem_addr) { struct acpi_table_header *header = NULL; int i = 0; - while (ACPI_SUCCESS(acpi_get_table("OEM1", i++, &header))) { + acpi_size tbl_size; + + while (ACPI_SUCCESS(acpi_get_table_with_size("OEM1", i++, &header, &tbl_size))) { if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) { struct oem_table *t = (struct oem_table *)header; - *oem_addr = (unsigned long)__acpi_map_table(t->OEMTableAddr, - t->OEMTableSize); + + oem_addrX = t->OEMTableAddr; + oem_size = t->OEMTableSize; + early_acpi_os_unmap_memory(header, tbl_size); + + *oem_addr = (unsigned long)__acpi_map_table(oem_addrX, + oem_size); return 0; } + early_acpi_os_unmap_memory(header, tbl_size); } return -1; } + +void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr) +{ + if (!oem_addr) + return; + + __acpi_unmap_table((char *)oem_addr, oem_size); +} #endif static void diff --git a/arch/x86/mach-generic/es7000.c b/arch/x86/mach-generic/es7000.c index 520cca0ee04..6513d41ea21 100644 --- a/arch/x86/mach-generic/es7000.c +++ b/arch/x86/mach-generic/es7000.c @@ -47,16 +47,26 @@ static __init int mps_oem_check(struct mp_config_table *mpc, char *oem, /* Hook from generic ACPI tables.c */ static int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) { - unsigned long oem_addr; + unsigned long oem_addr = 0; + int check_dsdt; + int ret = 0; + + /* check dsdt at first to avoid clear fix_map for oem_addr */ + check_dsdt = es7000_check_dsdt(); + if (!find_unisys_acpi_oem_table(&oem_addr)) { - if (es7000_check_dsdt()) - return parse_unisys_oem((char *)oem_addr); + if (check_dsdt) + ret = parse_unisys_oem((char *)oem_addr); else { setup_unisys(); - return 1; + ret = 1; } + /* + * we need to unmap it + */ + unmap_unisys_acpi_oem_table(oem_addr); } - return 0; + return ret; } #else static int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) diff --git a/include/asm-x86/es7000/mpparse.h b/include/asm-x86/es7000/mpparse.h index 7b5c889d8e7..ed5a3caae14 100644 --- a/include/asm-x86/es7000/mpparse.h +++ b/include/asm-x86/es7000/mpparse.h @@ -5,6 +5,7 @@ extern int parse_unisys_oem (char *oemptr); extern int find_unisys_acpi_oem_table(unsigned long *oem_addr); +extern void unmap_unisys_acpi_oem_table(unsigned long oem_addr); extern void setup_unisys(void); #ifndef CONFIG_X86_GENERICARCH -- cgit v1.2.3-70-g09d2 From 33f8c40a30ae99e971d068c9ec6088e713c46f5f Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Sun, 14 Sep 2008 19:03:53 +0200 Subject: x86: add memory clobber in switch_to() Segment registers are reloaded, so we should add a memory clobber. The generated assembly code is identical in my tests, but this doesn't mean it is necessarily true for all configurations/compilers. x86_64 already has the memory clobber. Signed-off-by: Vegard Nossum Signed-off-by: Ingo Molnar --- include/asm-x86/system.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h index 34505dd7b24..b20c894660f 100644 --- a/include/asm-x86/system.h +++ b/include/asm-x86/system.h @@ -64,7 +64,10 @@ do { \ \ /* regparm parameters for __switch_to(): */ \ [prev] "a" (prev), \ - [next] "d" (next)); \ + [next] "d" (next) \ + \ + : /* reloaded segment registers */ \ + "memory"); \ } while (0) /* -- cgit v1.2.3-70-g09d2 From 606ee44dbb72fd48beb47f171d7b9cecf6ade6dd Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Wed, 17 Sep 2008 16:48:17 +0100 Subject: x86: make mm/gup.c more virtualization friendly Since pte_flags() is much cheaper than pte_val() in some virtualized environments (namely, Xen), use the former whereever possible. Signed-off-by: Jan Beulich Cc: "Nick Piggin" Signed-off-by: Ingo Molnar --- arch/x86/mm/gup.c | 10 +++++----- include/asm-x86/pgtable.h | 2 +- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c index 007bb06c750..4ba373c5b8c 100644 --- a/arch/x86/mm/gup.c +++ b/arch/x86/mm/gup.c @@ -82,7 +82,7 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr, pte_t pte = gup_get_pte(ptep); struct page *page; - if ((pte_val(pte) & (mask | _PAGE_SPECIAL)) != mask) { + if ((pte_flags(pte) & (mask | _PAGE_SPECIAL)) != mask) { pte_unmap(ptep); return 0; } @@ -116,10 +116,10 @@ static noinline int gup_huge_pmd(pmd_t pmd, unsigned long addr, mask = _PAGE_PRESENT|_PAGE_USER; if (write) mask |= _PAGE_RW; - if ((pte_val(pte) & mask) != mask) + if ((pte_flags(pte) & mask) != mask) return 0; /* hugepages are never "special" */ - VM_BUG_ON(pte_val(pte) & _PAGE_SPECIAL); + VM_BUG_ON(pte_flags(pte) & _PAGE_SPECIAL); VM_BUG_ON(!pfn_valid(pte_pfn(pte))); refs = 0; @@ -173,10 +173,10 @@ static noinline int gup_huge_pud(pud_t pud, unsigned long addr, mask = _PAGE_PRESENT|_PAGE_USER; if (write) mask |= _PAGE_RW; - if ((pte_val(pte) & mask) != mask) + if ((pte_flags(pte) & mask) != mask) return 0; /* hugepages are never "special" */ - VM_BUG_ON(pte_val(pte) & _PAGE_SPECIAL); + VM_BUG_ON(pte_flags(pte) & _PAGE_SPECIAL); VM_BUG_ON(!pfn_valid(pte_pfn(pte))); refs = 0; diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index 81805403b64..182f9d4c570 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -206,7 +206,7 @@ static inline int pte_exec(pte_t pte) static inline int pte_special(pte_t pte) { - return pte_val(pte) & _PAGE_SPECIAL; + return pte_flags(pte) & _PAGE_SPECIAL; } static inline unsigned long pte_pfn(pte_t pte) -- cgit v1.2.3-70-g09d2 From cf4cfb225ab2b48611cb4f9e8db23c87486416d6 Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Fri, 11 Jul 2008 13:45:46 -0300 Subject: x86: use user_mode macro Instead of using SEGMENT_IS_KERNEL_CODE, use the "user_mode" macro, which can play the same role. Delete the former, since it now lacks any user. Signed-off-by: Glauber Costa Signed-off-by: Ingo Molnar --- arch/x86/kernel/time_32.c | 2 +- include/asm-x86/segment.h | 3 --- 2 files changed, 1 insertion(+), 4 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c index bbecf8b6bf9..8abcb7b0869 100644 --- a/arch/x86/kernel/time_32.c +++ b/arch/x86/kernel/time_32.c @@ -47,7 +47,7 @@ unsigned long profile_pc(struct pt_regs *regs) unsigned long pc = instruction_pointer(regs); #ifdef CONFIG_SMP - if (!v8086_mode(regs) && SEGMENT_IS_KERNEL_CODE(regs->cs) && + if (!v8086_mode(regs) && !user_mode(regs) && in_lock_functions(pc)) { #ifdef CONFIG_FRAME_POINTER return *(unsigned long *)(regs->bp + 4); diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h index ea5f0a8686f..4dcd8c6dd22 100644 --- a/include/asm-x86/segment.h +++ b/include/asm-x86/segment.h @@ -131,9 +131,6 @@ * Matching rules for certain types of segments. */ -/* Matches only __KERNEL_CS, ignoring PnP / USER / APM segments */ -#define SEGMENT_IS_KERNEL_CODE(x) (((x) & 0xfc) == GDT_ENTRY_KERNEL_CS * 8) - /* Matches __KERNEL_CS and __USER_CS (they must be 2 entries apart) */ #define SEGMENT_IS_FLAT_CODE(x) (((x) & 0xec) == GDT_ENTRY_KERNEL_CS * 8) -- cgit v1.2.3-70-g09d2 From efa323abd424a2450f810b96203c1fbf138998b9 Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Fri, 11 Jul 2008 14:00:14 -0300 Subject: x86: remove SEGMENT_IS_FLAT_CODE There are no users in the kernel. Signed-off-by: Glauber Costa Signed-off-by: Ingo Molnar --- include/asm-x86/segment.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h index 4dcd8c6dd22..5d6e6945489 100644 --- a/include/asm-x86/segment.h +++ b/include/asm-x86/segment.h @@ -131,9 +131,6 @@ * Matching rules for certain types of segments. */ -/* Matches __KERNEL_CS and __USER_CS (they must be 2 entries apart) */ -#define SEGMENT_IS_FLAT_CODE(x) (((x) & 0xec) == GDT_ENTRY_KERNEL_CS * 8) - /* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */ #define SEGMENT_IS_PNP_CODE(x) (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8) -- cgit v1.2.3-70-g09d2 From 69c89b5bf7f253756f3056e84b8603abe1c50f5b Mon Sep 17 00:00:00 2001 From: Alexander van Heukelum Date: Fri, 26 Sep 2008 14:03:07 +0200 Subject: traps: x86: remove trace_hardirqs_fixup from pagefault handler The last use of trace_hardirqs_fixup is unnecessary, because the trap is taken with interrupt off on i386 as well as x86_64, and the irq-tracer is notified of this from the assembly code. trace_hardirqs_fixup and trace_hardirqs_fixup_flags are removed from include/asm-x86/irqflags.h as they are no longer used. Signed-off-by: Alexander van Heukelum Signed-off-by: Ingo Molnar --- arch/x86/mm/fault.c | 5 ----- include/asm-x86/irqflags.h | 21 --------------------- 2 files changed, 26 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index a742d753d5b..3f2b8962cbd 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -592,11 +592,6 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) unsigned long flags; #endif - /* - * We can fault from pretty much anywhere, with unknown IRQ state. - */ - trace_hardirqs_fixup(); - tsk = current; mm = tsk->mm; prefetchw(&mm->mmap_sem); diff --git a/include/asm-x86/irqflags.h b/include/asm-x86/irqflags.h index 424acb48cd6..2bdab21f089 100644 --- a/include/asm-x86/irqflags.h +++ b/include/asm-x86/irqflags.h @@ -166,27 +166,6 @@ static inline int raw_irqs_disabled(void) return raw_irqs_disabled_flags(flags); } -/* - * makes the traced hardirq state match with the machine state - * - * should be a rarely used function, only in places where its - * otherwise impossible to know the irq state, like in traps. - */ -static inline void trace_hardirqs_fixup_flags(unsigned long flags) -{ - if (raw_irqs_disabled_flags(flags)) - trace_hardirqs_off(); - else - trace_hardirqs_on(); -} - -static inline void trace_hardirqs_fixup(void) -{ - unsigned long flags = __raw_local_save_flags(); - - trace_hardirqs_fixup_flags(flags); -} - #else #ifdef CONFIG_X86_64 -- cgit v1.2.3-70-g09d2 From 69d45dd1c3bb512a9f5f9c464ac625eb707669ec Mon Sep 17 00:00:00 2001 From: Krzysztof Helt Date: Sun, 28 Sep 2008 21:28:15 +0200 Subject: x86: merge winchip-2 and winchip-2a cpu choices The Winchip-2 and Winchip-2A cpu choices select the same options for kernel and compiler. Merge them to save few bytes and reduce confusion. Signed-off-by: Krzysztof Helt Acked-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- arch/x86/Kconfig.cpu | 27 +++++++++------------------ arch/x86/Makefile_32.cpu | 1 - arch/x86/configs/i386_defconfig | 1 - arch/x86/configs/x86_64_defconfig | 1 - include/asm-x86/module.h | 2 -- 5 files changed, 9 insertions(+), 23 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index c5f10136052..f4bdc0492d3 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -38,8 +38,7 @@ config M386 - "Crusoe" for the Transmeta Crusoe series. - "Efficeon" for the Transmeta Efficeon series. - "Winchip-C6" for original IDT Winchip. - - "Winchip-2" for IDT Winchip 2. - - "Winchip-2A" for IDT Winchips with 3dNow! capabilities. + - "Winchip-2" for IDT Winchips with 3dNow! capabilities. - "GeodeGX1" for Geode GX1 (Cyrix MediaGX). - "Geode GX/LX" For AMD Geode GX and LX processors. - "CyrixIII/VIA C3" for VIA Cyrix III or VIA C3. @@ -194,19 +193,11 @@ config MWINCHIPC6 treat this chip as a 586TSC with some extended instructions and alignment requirements. -config MWINCHIP2 - bool "Winchip-2" - depends on X86_32 - help - Select this for an IDT Winchip-2. Linux and GCC - treat this chip as a 586TSC with some extended instructions - and alignment requirements. - config MWINCHIP3D - bool "Winchip-2A/Winchip-3" + bool "Winchip-2/Winchip-2A/Winchip-3" depends on X86_32 help - Select this for an IDT Winchip-2A or 3. Linux and GCC + Select this for an IDT Winchip-2, 2A or 3. Linux and GCC treat this chip as a 586TSC with some extended instructions and alignment requirements. Also enable out of order memory stores for this CPU, which can increase performance of some @@ -318,7 +309,7 @@ config X86_L1_CACHE_SHIFT int default "7" if MPENTIUM4 || X86_GENERIC || GENERIC_CPU || MPSC default "4" if X86_ELAN || M486 || M386 || MGEODEGX1 - default "5" if MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE_LX + default "5" if MWINCHIP3D || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE_LX default "6" if MK7 || MK8 || MPENTIUMM || MCORE2 || MVIAC7 config X86_XADD @@ -360,7 +351,7 @@ config X86_POPAD_OK config X86_ALIGNMENT_16 def_bool y - depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 + depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 config X86_INTEL_USERCOPY def_bool y @@ -368,7 +359,7 @@ config X86_INTEL_USERCOPY config X86_USE_PPRO_CHECKSUM def_bool y - depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2 + depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2 config X86_USE_3DNOW def_bool y @@ -376,7 +367,7 @@ config X86_USE_3DNOW config X86_OOSTORE def_bool y - depends on (MWINCHIP3D || MWINCHIP2 || MWINCHIPC6) && MTRR + depends on (MWINCHIP3D || MWINCHIPC6) && MTRR # # P6_NOPs are a relatively minor optimization that require a family >= @@ -396,7 +387,7 @@ config X86_P6_NOP config X86_TSC def_bool y - depends on ((MWINCHIP3D || MWINCHIP2 || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MVIAC7 || MGEODEGX1 || MGEODE_LX || MCORE2) && !X86_NUMAQ) || X86_64 + depends on ((MWINCHIP3D || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MVIAC7 || MGEODEGX1 || MGEODE_LX || MCORE2) && !X86_NUMAQ) || X86_64 config X86_CMPXCHG64 def_bool y @@ -417,7 +408,7 @@ config X86_MINIMUM_CPU_FAMILY config X86_DEBUGCTLMSR def_bool y - depends on !(MK6 || MWINCHIPC6 || MWINCHIP2 || MWINCHIP3D || MCYRIXIII || M586MMX || M586TSC || M586 || M486 || M386) + depends on !(MK6 || MWINCHIPC6 || MWINCHIP3D || MCYRIXIII || M586MMX || M586TSC || M586 || M486 || M386) menuconfig PROCESSOR_SELECT bool "Supported processor vendors" if EMBEDDED diff --git a/arch/x86/Makefile_32.cpu b/arch/x86/Makefile_32.cpu index b72b4f75311..80177ec052f 100644 --- a/arch/x86/Makefile_32.cpu +++ b/arch/x86/Makefile_32.cpu @@ -28,7 +28,6 @@ cflags-$(CONFIG_MK8) += $(call cc-option,-march=k8,-march=athlon) cflags-$(CONFIG_MCRUSOE) += -march=i686 $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0 cflags-$(CONFIG_MEFFICEON) += -march=i686 $(call tune,pentium3) $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0 cflags-$(CONFIG_MWINCHIPC6) += $(call cc-option,-march=winchip-c6,-march=i586) -cflags-$(CONFIG_MWINCHIP2) += $(call cc-option,-march=winchip2,-march=i586) cflags-$(CONFIG_MWINCHIP3D) += $(call cc-option,-march=winchip2,-march=i586) cflags-$(CONFIG_MCYRIXIII) += $(call cc-option,-march=c3,-march=i486) $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0 cflags-$(CONFIG_MVIAC3_2) += $(call cc-option,-march=c3-2,-march=i686) diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig index ca226ca3128..52d0359719d 100644 --- a/arch/x86/configs/i386_defconfig +++ b/arch/x86/configs/i386_defconfig @@ -213,7 +213,6 @@ CONFIG_M686=y # CONFIG_MCRUSOE is not set # CONFIG_MEFFICEON is not set # CONFIG_MWINCHIPC6 is not set -# CONFIG_MWINCHIP2 is not set # CONFIG_MWINCHIP3D is not set # CONFIG_MGEODEGX1 is not set # CONFIG_MGEODE_LX is not set diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig index 2c4b1c771e2..f0a03d7a7d6 100644 --- a/arch/x86/configs/x86_64_defconfig +++ b/arch/x86/configs/x86_64_defconfig @@ -210,7 +210,6 @@ CONFIG_X86_PC=y # CONFIG_MCRUSOE is not set # CONFIG_MEFFICEON is not set # CONFIG_MWINCHIPC6 is not set -# CONFIG_MWINCHIP2 is not set # CONFIG_MWINCHIP3D is not set # CONFIG_MGEODEGX1 is not set # CONFIG_MGEODE_LX is not set diff --git a/include/asm-x86/module.h b/include/asm-x86/module.h index 48dc3e0c07d..864f2005fc1 100644 --- a/include/asm-x86/module.h +++ b/include/asm-x86/module.h @@ -52,8 +52,6 @@ struct mod_arch_specific {}; #define MODULE_PROC_FAMILY "EFFICEON " #elif defined CONFIG_MWINCHIPC6 #define MODULE_PROC_FAMILY "WINCHIPC6 " -#elif defined CONFIG_MWINCHIP2 -#define MODULE_PROC_FAMILY "WINCHIP2 " #elif defined CONFIG_MWINCHIP3D #define MODULE_PROC_FAMILY "WINCHIP3D " #elif defined CONFIG_MCYRIXIII -- cgit v1.2.3-70-g09d2 From 14adf855baefad5ac3b545be23a64e6b61d6b74a Mon Sep 17 00:00:00 2001 From: Chuck Ebbert Date: Mon, 29 Sep 2008 18:29:42 -0400 Subject: x86: move prefill_possible_map calling early, fix, V2 Commit 4a701737 ("x86: move prefill_possible_map calling early, fix") is the wrong fix: prefill_possible_map() needs to be available even when CONFIG_HOTPLUG_CPU is not set. A followon patch will do that. Fix this correctly by making prefill_possible_map() available even when CONFIG_HOTPLUG_CPU is not set. The function is needed so that the number of possible CPUs can be determined. Tested on uniprocessor machine with CPU hotplug disabled. From boot log: Before: NR_CPUS: 512, nr_cpu_ids: 512, nr_node_ids 1 After: NR_CPUS: 512, nr_cpu_ids: 1, nr_node_ids 1 Signed-off-by: Chuck Ebbert Signed-off-by: Ingo Molnar --- arch/x86/kernel/smpboot.c | 62 +++++++++++++++++++++++------------------------ include/asm-x86/smp.h | 8 +++--- 2 files changed, 34 insertions(+), 36 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index a778e221ccf..23913785c26 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1261,39 +1261,8 @@ void __init native_smp_cpus_done(unsigned int max_cpus) check_nmi_watchdog(); } -#ifdef CONFIG_HOTPLUG_CPU - -static void remove_siblinginfo(int cpu) -{ - int sibling; - struct cpuinfo_x86 *c = &cpu_data(cpu); - - for_each_cpu_mask_nr(sibling, per_cpu(cpu_core_map, cpu)) { - cpu_clear(cpu, per_cpu(cpu_core_map, sibling)); - /*/ - * last thread sibling in this cpu core going down - */ - if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1) - cpu_data(sibling).booted_cores--; - } - - for_each_cpu_mask_nr(sibling, per_cpu(cpu_sibling_map, cpu)) - cpu_clear(cpu, per_cpu(cpu_sibling_map, sibling)); - cpus_clear(per_cpu(cpu_sibling_map, cpu)); - cpus_clear(per_cpu(cpu_core_map, cpu)); - c->phys_proc_id = 0; - c->cpu_core_id = 0; - cpu_clear(cpu, cpu_sibling_setup_map); -} - static int additional_cpus __initdata = -1; -static __init int setup_additional_cpus(char *s) -{ - return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL; -} -early_param("additional_cpus", setup_additional_cpus); - /* * cpu_possible_map should be static, it cannot change as cpu's * are onlined, or offlined. The reason is per-cpu data-structures @@ -1340,6 +1309,37 @@ __init void prefill_possible_map(void) nr_cpu_ids = possible; } +#ifdef CONFIG_HOTPLUG_CPU + +static void remove_siblinginfo(int cpu) +{ + int sibling; + struct cpuinfo_x86 *c = &cpu_data(cpu); + + for_each_cpu_mask_nr(sibling, per_cpu(cpu_core_map, cpu)) { + cpu_clear(cpu, per_cpu(cpu_core_map, sibling)); + /*/ + * last thread sibling in this cpu core going down + */ + if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1) + cpu_data(sibling).booted_cores--; + } + + for_each_cpu_mask_nr(sibling, per_cpu(cpu_sibling_map, cpu)) + cpu_clear(cpu, per_cpu(cpu_sibling_map, sibling)); + cpus_clear(per_cpu(cpu_sibling_map, cpu)); + cpus_clear(per_cpu(cpu_core_map, cpu)); + c->phys_proc_id = 0; + c->cpu_core_id = 0; + cpu_clear(cpu, cpu_sibling_setup_map); +} + +static __init int setup_additional_cpus(char *s) +{ + return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL; +} +early_param("additional_cpus", setup_additional_cpus); + static void __ref remove_cpu_from_maps(int cpu) { cpu_clear(cpu, cpu_online_map); diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 6df2615f913..a6afc29f2dd 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -141,6 +141,8 @@ void play_dead_common(void); void native_send_call_func_ipi(cpumask_t mask); void native_send_call_func_single_ipi(int cpu); +extern void prefill_possible_map(void); + void smp_store_cpu_info(int id); #define cpu_physical_id(cpu) per_cpu(x86_cpu_to_apicid, cpu) @@ -149,15 +151,11 @@ static inline int num_booting_cpus(void) { return cpus_weight(cpu_callout_map); } -#endif /* CONFIG_SMP */ - -#if defined(CONFIG_SMP) && defined(CONFIG_HOTPLUG_CPU) -extern void prefill_possible_map(void); #else static inline void prefill_possible_map(void) { } -#endif +#endif /* CONFIG_SMP */ extern unsigned disabled_cpus __cpuinitdata; -- cgit v1.2.3-70-g09d2 From e2ce07c8042975e52df4cec1f41faf15b83f2e42 Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Thu, 3 Apr 2008 16:40:48 +0300 Subject: x86: __show_registers() and __show_regs() API unification Currently the low-level function to dump user-passed registers on i386 is called __show_registers() whereas on x86-64 it's called __show_regs(). Unify the API to simplify porting of kmemcheck to x86-64. Signed-off-by: Pekka Enberg Acked-by: Vegard Nossum Signed-off-by: Ingo Molnar --- arch/x86/kernel/process_32.c | 4 ++-- arch/x86/kernel/process_64.c | 7 +++++-- arch/x86/kernel/traps_32.c | 2 +- arch/x86/kernel/traps_64.c | 2 +- include/asm-x86/kdebug.h | 3 +-- 5 files changed, 10 insertions(+), 8 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 922c14058f9..0a1302fe6d4 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -123,7 +123,7 @@ void cpu_idle(void) } } -void __show_registers(struct pt_regs *regs, int all) +void __show_regs(struct pt_regs *regs, int all) { unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L; unsigned long d0, d1, d2, d3, d6, d7; @@ -189,7 +189,7 @@ void __show_registers(struct pt_regs *regs, int all) void show_regs(struct pt_regs *regs) { - __show_registers(regs, 1); + __show_regs(regs, 1); show_trace(NULL, regs, ®s->sp, regs->bp); } diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index ca80394ef5b..cd8c0ed02b7 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -136,7 +136,7 @@ void cpu_idle(void) } /* Prints also some state that isn't saved in the pt_regs */ -void __show_regs(struct pt_regs *regs) +void __show_regs(struct pt_regs *regs, int all) { unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L, fs, gs, shadowgs; unsigned long d0, d1, d2, d3, d6, d7; @@ -175,6 +175,9 @@ void __show_regs(struct pt_regs *regs) rdmsrl(MSR_GS_BASE, gs); rdmsrl(MSR_KERNEL_GS_BASE, shadowgs); + if (!all) + return; + cr0 = read_cr0(); cr2 = read_cr2(); cr3 = read_cr3(); @@ -200,7 +203,7 @@ void __show_regs(struct pt_regs *regs) void show_regs(struct pt_regs *regs) { printk(KERN_INFO "CPU %d:", smp_processor_id()); - __show_regs(regs); + __show_regs(regs, 1); show_trace(NULL, regs, (void *)(regs + 1), regs->bp); } diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index bf4d71231d4..4f0831593e3 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -319,7 +319,7 @@ void show_registers(struct pt_regs *regs) int i; print_modules(); - __show_registers(regs, 0); + __show_regs(regs, 0); printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)", TASK_COMM_LEN, current->comm, task_pid_nr(current), diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index 1efd1ea6466..729157ee4c1 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -430,7 +430,7 @@ void show_registers(struct pt_regs *regs) sp = regs->sp; printk("CPU %d ", cpu); - __show_regs(regs); + __show_regs(regs, 1); printk("Process %s (pid: %d, threadinfo %p, task %p)\n", cur->comm, cur->pid, task_thread_info(cur), cur); diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h index 5ec3ad3e825..fbbab66ee9d 100644 --- a/include/asm-x86/kdebug.h +++ b/include/asm-x86/kdebug.h @@ -27,10 +27,9 @@ extern void printk_address(unsigned long address, int reliable); extern void die(const char *, struct pt_regs *,long); extern int __must_check __die(const char *, struct pt_regs *, long); extern void show_registers(struct pt_regs *regs); -extern void __show_registers(struct pt_regs *, int all); extern void show_trace(struct task_struct *t, struct pt_regs *regs, unsigned long *sp, unsigned long bp); -extern void __show_regs(struct pt_regs *regs); +extern void __show_regs(struct pt_regs *regs, int all); extern void show_regs(struct pt_regs *regs); extern unsigned long oops_begin(void); extern void oops_end(unsigned long, struct pt_regs *, int signr); -- cgit v1.2.3-70-g09d2 From af5c2bd16ac2e5688c3bf46ea1f95112d696d294 Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Fri, 3 Oct 2008 17:54:25 +0200 Subject: x86: fix virt_addr_valid() with CONFIG_DEBUG_VIRTUAL=y, v2 virt_addr_valid() calls __pa(), which calls __phys_addr(). With CONFIG_DEBUG_VIRTUAL=y, __phys_addr() will kill the kernel if the address *isn't* valid. That's clearly wrong for virt_addr_valid(). We also incorporate the debugging checks into virt_addr_valid(). Signed-off-by: Vegard Nossum Acked-by: Jiri Slaby Signed-off-by: Ingo Molnar --- arch/x86/kernel/doublefault_32.c | 2 +- arch/x86/mm/ioremap.c | 36 ++++++++++++++++++++++++++++++++++-- include/asm-x86/page.h | 8 +++++++- include/asm-x86/page_32.h | 4 ++-- 4 files changed, 44 insertions(+), 6 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c index 395acb12b0d..b4f14c6c09d 100644 --- a/arch/x86/kernel/doublefault_32.c +++ b/arch/x86/kernel/doublefault_32.c @@ -66,6 +66,6 @@ struct tss_struct doublefault_tss __cacheline_aligned = { .ds = __USER_DS, .fs = __KERNEL_PERCPU, - .__cr3 = __phys_addr_const((unsigned long)swapper_pg_dir) + .__cr3 = __pa_nodebug(swapper_pg_dir), } }; diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 7fb737c6b54..8876220d906 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -45,6 +45,27 @@ unsigned long __phys_addr(unsigned long x) } EXPORT_SYMBOL(__phys_addr); +bool __virt_addr_valid(unsigned long x) +{ + if (x >= __START_KERNEL_map) { + x -= __START_KERNEL_map; + if (x >= KERNEL_IMAGE_SIZE) + return false; + x += phys_base; + } else { + if (x < PAGE_OFFSET) + return false; + x -= PAGE_OFFSET; + if (system_state == SYSTEM_BOOTING ? + x > MAXMEM : !phys_addr_valid(x)) { + return false; + } + } + + return pfn_valid(x >> PAGE_SHIFT); +} +EXPORT_SYMBOL(__virt_addr_valid); + #else static inline int phys_addr_valid(unsigned long addr) @@ -56,13 +77,24 @@ static inline int phys_addr_valid(unsigned long addr) unsigned long __phys_addr(unsigned long x) { /* VMALLOC_* aren't constants; not available at the boot time */ - VIRTUAL_BUG_ON(x < PAGE_OFFSET || (system_state != SYSTEM_BOOTING && - is_vmalloc_addr((void *)x))); + VIRTUAL_BUG_ON(x < PAGE_OFFSET); + VIRTUAL_BUG_ON(system_state != SYSTEM_BOOTING && + is_vmalloc_addr((void *) x)); return x - PAGE_OFFSET; } EXPORT_SYMBOL(__phys_addr); #endif +bool __virt_addr_valid(unsigned long x) +{ + if (x < PAGE_OFFSET) + return false; + if (system_state != SYSTEM_BOOTING && is_vmalloc_addr((void *) x)) + return false; + return pfn_valid((x - PAGE_OFFSET) >> PAGE_SHIFT); +} +EXPORT_SYMBOL(__virt_addr_valid); + #endif int page_is_ram(unsigned long pagenr) diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h index c9157477675..d4f1d5791fc 100644 --- a/include/asm-x86/page.h +++ b/include/asm-x86/page.h @@ -179,6 +179,7 @@ static inline pteval_t native_pte_flags(pte_t pte) #endif /* CONFIG_PARAVIRT */ #define __pa(x) __phys_addr((unsigned long)(x)) +#define __pa_nodebug(x) __phys_addr_nodebug((unsigned long)(x)) /* __pa_symbol should be used for C visible symbols. This seems to be the official gcc blessed way to do such arithmetic. */ #define __pa_symbol(x) __pa(__phys_reloc_hide((unsigned long)(x))) @@ -188,9 +189,14 @@ static inline pteval_t native_pte_flags(pte_t pte) #define __boot_va(x) __va(x) #define __boot_pa(x) __pa(x) +/* + * virt_to_page(kaddr) returns a valid pointer if and only if + * virt_addr_valid(kaddr) returns true. + */ #define virt_to_page(kaddr) pfn_to_page(__pa(kaddr) >> PAGE_SHIFT) #define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT) -#define virt_addr_valid(kaddr) pfn_valid(__pa(kaddr) >> PAGE_SHIFT) +extern bool __virt_addr_valid(unsigned long kaddr); +#define virt_addr_valid(kaddr) __virt_addr_valid((unsigned long) (kaddr)) #endif /* __ASSEMBLY__ */ diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index 9c5a737a9af..5d6a68a1067 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -73,11 +73,11 @@ typedef struct page *pgtable_t; #endif #ifndef __ASSEMBLY__ -#define __phys_addr_const(x) ((x) - PAGE_OFFSET) +#define __phys_addr_nodebug(x) ((x) - PAGE_OFFSET) #ifdef CONFIG_DEBUG_VIRTUAL extern unsigned long __phys_addr(unsigned long); #else -#define __phys_addr(x) ((x) - PAGE_OFFSET) +#define __phys_addr(x) __phys_addr_nodebug(x) #endif #define __phys_reloc_hide(x) RELOC_HIDE((x), 0) -- cgit v1.2.3-70-g09d2 From a28680b4b821a262fd3b5e57a28c148b5f9e662a Mon Sep 17 00:00:00 2001 From: Alexander van Heukelum Date: Tue, 30 Sep 2008 18:41:34 +0200 Subject: x86, traps: split out math_error and simd_math_error Split out math_error from do_coprocessor_error and simd_math_error from do_simd_coprocessor_error, like on i386. While at it, add the "error_code" parameter to do_coprocessor_error, do_simd_coprocessor_error and do_spurious_interrupt_bug. This does not change the generated code, but brings the declarations in line with all the other trap handlers. Signed-off-by: Alexander van Heukelum Signed-off-by: Ingo Molnar --- arch/x86/kernel/traps_64.c | 36 +++++++++++++++++++++--------------- include/asm-x86/traps.h | 6 +++--- 2 files changed, 24 insertions(+), 18 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index 1cd61ddd90b..b295ebf0cb3 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -452,18 +452,12 @@ static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr) * the correct behaviour even in the presence of the asynchronous * IRQ13 behaviour */ -asmlinkage void do_coprocessor_error(struct pt_regs *regs) +void math_error(void __user *ip) { - void __user *ip = (void __user *)(regs->ip); struct task_struct *task; siginfo_t info; unsigned short cwd, swd; - conditional_sti(regs); - if (!user_mode(regs) && - kernel_math_error(regs, "kernel x87 math error", 16)) - return; - /* * Save the info for the exception handler and clear the error. */ @@ -516,23 +510,26 @@ asmlinkage void do_coprocessor_error(struct pt_regs *regs) force_sig_info(SIGFPE, &info, task); } +asmlinkage void do_coprocessor_error(struct pt_regs *regs, long error_code) +{ + conditional_sti(regs); + if (!user_mode(regs) && + kernel_math_error(regs, "kernel x87 math error", 16)) + return; + math_error((void __user *)regs->ip); +} + asmlinkage void bad_intr(void) { printk("bad interrupt"); } -asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs) +static void simd_math_error(void __user *ip) { - void __user *ip = (void __user *)(regs->ip); struct task_struct *task; siginfo_t info; unsigned short mxcsr; - conditional_sti(regs); - if (!user_mode(regs) && - kernel_math_error(regs, "kernel simd math error", 19)) - return; - /* * Save the info for the exception handler and clear the error. */ @@ -575,7 +572,16 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs) force_sig_info(SIGFPE, &info, task); } -asmlinkage void do_spurious_interrupt_bug(struct pt_regs *regs) +asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs, long error_code) +{ + conditional_sti(regs); + if (!user_mode(regs) && + kernel_math_error(regs, "kernel simd math error", 19)) + return; + simd_math_error((void __user *)regs->ip); +} + +asmlinkage void do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) { } diff --git a/include/asm-x86/traps.h b/include/asm-x86/traps.h index 7a692baa51a..c82c39c7b5e 100644 --- a/include/asm-x86/traps.h +++ b/include/asm-x86/traps.h @@ -72,9 +72,9 @@ asmlinkage void double_fault(void); asmlinkage void do_int3(struct pt_regs *, long); asmlinkage void do_stack_segment(struct pt_regs *, long); asmlinkage void do_debug(struct pt_regs *, unsigned long); -asmlinkage void do_coprocessor_error(struct pt_regs *); -asmlinkage void do_simd_coprocessor_error(struct pt_regs *); -asmlinkage void do_spurious_interrupt_bug(struct pt_regs *); +asmlinkage void do_coprocessor_error(struct pt_regs *, long); +asmlinkage void do_simd_coprocessor_error(struct pt_regs *, long); +asmlinkage void do_spurious_interrupt_bug(struct pt_regs *, long); asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code); -- cgit v1.2.3-70-g09d2 From e407d62088b7f61f38e1086062650c75a4f2757a Mon Sep 17 00:00:00 2001 From: Alexander van Heukelum Date: Tue, 30 Sep 2008 18:41:36 +0200 Subject: x86, traps: introduce dotraplinkage Mark the exception handlers with "dotraplinkage" to hide the calling convention differences between i386 and x86_64. Signed-off-by: Alexander van Heukelum Signed-off-by: Ingo Molnar --- arch/x86/kernel/entry_64.S | 2 +- arch/x86/kernel/traps_32.c | 28 ++++++++++-------- arch/x86/kernel/traps_64.c | 31 ++++++++++++-------- include/asm-x86/traps.h | 73 +++++++++++++++++++++++----------------------- 4 files changed, 72 insertions(+), 62 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 977c8ea6602..1db6ce4314e 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -1243,7 +1243,7 @@ ENTRY(simd_coprocessor_error) END(simd_coprocessor_error) ENTRY(device_not_available) - zeroentry math_state_restore + zeroentry do_device_not_available END(device_not_available) /* runs on exception stack */ diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index 0206c915748..7eb1203c909 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -190,7 +190,7 @@ vm86_trap: } #define DO_ERROR(trapnr, signr, str, name) \ -void do_##name(struct pt_regs *regs, long error_code) \ +dotraplinkage void do_##name(struct pt_regs *regs, long error_code) \ { \ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ == NOTIFY_STOP) \ @@ -200,7 +200,7 @@ void do_##name(struct pt_regs *regs, long error_code) \ } #define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ -void do_##name(struct pt_regs *regs, long error_code) \ +dotraplinkage void do_##name(struct pt_regs *regs, long error_code) \ { \ siginfo_t info; \ info.si_signo = signr; \ @@ -224,7 +224,7 @@ DO_ERROR(11, SIGBUS, "segment not present", segment_not_present) DO_ERROR(12, SIGBUS, "stack segment", stack_segment) DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0) -void __kprobes +dotraplinkage void __kprobes do_general_protection(struct pt_regs *regs, long error_code) { struct task_struct *tsk; @@ -428,7 +428,8 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) reassert_nmi(); } -notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code) +dotraplinkage notrace __kprobes void +do_nmi(struct pt_regs *regs, long error_code) { int cpu; @@ -456,7 +457,7 @@ void restart_nmi(void) acpi_nmi_enable(); } -void __kprobes do_int3(struct pt_regs *regs, long error_code) +dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) { #ifdef CONFIG_KPROBES if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) @@ -494,7 +495,7 @@ void __kprobes do_int3(struct pt_regs *regs, long error_code) * find every occurrence of the TF bit that could be saved away even * by user code) */ -void __kprobes do_debug(struct pt_regs *regs, long error_code) +dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) { struct task_struct *tsk = current; unsigned int condition; @@ -627,7 +628,7 @@ void math_error(void __user *ip) force_sig_info(SIGFPE, &info, task); } -void do_coprocessor_error(struct pt_regs *regs, long error_code) +dotraplinkage void do_coprocessor_error(struct pt_regs *regs, long error_code) { conditional_sti(regs); ignore_fpu_irq = 1; @@ -682,7 +683,8 @@ static void simd_math_error(void __user *ip) force_sig_info(SIGFPE, &info, task); } -void do_simd_coprocessor_error(struct pt_regs *regs, long error_code) +dotraplinkage void +do_simd_coprocessor_error(struct pt_regs *regs, long error_code) { conditional_sti(regs); @@ -706,7 +708,8 @@ void do_simd_coprocessor_error(struct pt_regs *regs, long error_code) force_sig(SIGSEGV, current); } -void do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) +dotraplinkage void +do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) { conditional_sti(regs); #if 0 @@ -784,7 +787,8 @@ asmlinkage void math_emulate(long arg) #endif /* CONFIG_MATH_EMULATION */ -void __kprobes do_device_not_available(struct pt_regs *regs, long error) +dotraplinkage void __kprobes +do_device_not_available(struct pt_regs *regs, long error) { if (read_cr0() & X86_CR0_EM) { conditional_sti(regs); @@ -796,14 +800,14 @@ void __kprobes do_device_not_available(struct pt_regs *regs, long error) } #ifdef CONFIG_X86_MCE -void __kprobes do_machine_check(struct pt_regs *regs, long error) +dotraplinkage void __kprobes do_machine_check(struct pt_regs *regs, long error) { conditional_sti(regs); machine_check_vector(regs, error); } #endif -void do_iret_error(struct pt_regs *regs, long error_code) +dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code) { siginfo_t info; local_irq_enable(); diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index b295ebf0cb3..be73323ca95 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -125,7 +125,7 @@ kernel_trap: } #define DO_ERROR(trapnr, signr, str, name) \ -asmlinkage void do_##name(struct pt_regs *regs, long error_code) \ +dotraplinkage void do_##name(struct pt_regs *regs, long error_code) \ { \ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ == NOTIFY_STOP) \ @@ -135,7 +135,7 @@ asmlinkage void do_##name(struct pt_regs *regs, long error_code) \ } #define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ -asmlinkage void do_##name(struct pt_regs *regs, long error_code) \ +dotraplinkage void do_##name(struct pt_regs *regs, long error_code) \ { \ siginfo_t info; \ info.si_signo = signr; \ @@ -159,7 +159,7 @@ DO_ERROR(11, SIGBUS, "segment not present", segment_not_present) DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0) /* Runs on IST stack */ -asmlinkage void do_stack_segment(struct pt_regs *regs, long error_code) +dotraplinkage void do_stack_segment(struct pt_regs *regs, long error_code) { if (notify_die(DIE_TRAP, "stack segment", regs, error_code, 12, SIGBUS) == NOTIFY_STOP) @@ -169,7 +169,7 @@ asmlinkage void do_stack_segment(struct pt_regs *regs, long error_code) preempt_conditional_cli(regs); } -asmlinkage void do_double_fault(struct pt_regs *regs, long error_code) +dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code) { static const char str[] = "double fault"; struct task_struct *tsk = current; @@ -186,7 +186,7 @@ asmlinkage void do_double_fault(struct pt_regs *regs, long error_code) die(str, regs, error_code); } -asmlinkage void __kprobes +dotraplinkage void __kprobes do_general_protection(struct pt_regs *regs, long error_code) { struct task_struct *tsk; @@ -317,7 +317,7 @@ asmlinkage notrace __kprobes void default_do_nmi(struct pt_regs *regs) io_check_error(reason, regs); } -asmlinkage notrace __kprobes void +dotraplinkage notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code) { nmi_enter(); @@ -343,7 +343,7 @@ void restart_nmi(void) } /* runs on IST stack. */ -asmlinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) +dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) { if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) @@ -376,8 +376,7 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) } /* runs on IST stack. */ -asmlinkage void __kprobes do_debug(struct pt_regs *regs, - unsigned long error_code) +dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) { struct task_struct *tsk = current; unsigned long condition; @@ -510,7 +509,7 @@ void math_error(void __user *ip) force_sig_info(SIGFPE, &info, task); } -asmlinkage void do_coprocessor_error(struct pt_regs *regs, long error_code) +dotraplinkage void do_coprocessor_error(struct pt_regs *regs, long error_code) { conditional_sti(regs); if (!user_mode(regs) && @@ -572,7 +571,8 @@ static void simd_math_error(void __user *ip) force_sig_info(SIGFPE, &info, task); } -asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs, long error_code) +dotraplinkage void +do_simd_coprocessor_error(struct pt_regs *regs, long error_code) { conditional_sti(regs); if (!user_mode(regs) && @@ -581,7 +581,8 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs, long error_code) simd_math_error((void __user *)regs->ip); } -asmlinkage void do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) +dotraplinkage void +do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) { } @@ -633,6 +634,12 @@ asmlinkage void math_state_restore(void) } EXPORT_SYMBOL_GPL(math_state_restore); +dotraplinkage void __kprobes +do_device_not_available(struct pt_regs *regs, long error) +{ + math_state_restore(); +} + void __init trap_init(void) { set_intr_gate(0, ÷_error); diff --git a/include/asm-x86/traps.h b/include/asm-x86/traps.h index c82c39c7b5e..6c3dc2c6575 100644 --- a/include/asm-x86/traps.h +++ b/include/asm-x86/traps.h @@ -3,7 +3,12 @@ #include -/* Common in X86_32 and X86_64 */ +#ifdef CONFIG_X86_32 +#define dotraplinkage +#else +#define dotraplinkage asmlinkage +#endif + asmlinkage void divide_error(void); asmlinkage void debug(void); asmlinkage void nmi(void); @@ -12,31 +17,47 @@ asmlinkage void overflow(void); asmlinkage void bounds(void); asmlinkage void invalid_op(void); asmlinkage void device_not_available(void); +#ifdef CONFIG_X86_64 +asmlinkage void double_fault(void); +#endif asmlinkage void coprocessor_segment_overrun(void); asmlinkage void invalid_TSS(void); asmlinkage void segment_not_present(void); asmlinkage void stack_segment(void); asmlinkage void general_protection(void); asmlinkage void page_fault(void); +asmlinkage void spurious_interrupt_bug(void); asmlinkage void coprocessor_error(void); -asmlinkage void simd_coprocessor_error(void); asmlinkage void alignment_check(void); -asmlinkage void spurious_interrupt_bug(void); #ifdef CONFIG_X86_MCE asmlinkage void machine_check(void); #endif /* CONFIG_X86_MCE */ +asmlinkage void simd_coprocessor_error(void); -void do_divide_error(struct pt_regs *, long); -void do_overflow(struct pt_regs *, long); -void do_bounds(struct pt_regs *, long); -void do_coprocessor_segment_overrun(struct pt_regs *, long); -void do_invalid_TSS(struct pt_regs *, long); -void do_segment_not_present(struct pt_regs *, long); -void do_stack_segment(struct pt_regs *, long); -void do_alignment_check(struct pt_regs *, long); -void do_invalid_op(struct pt_regs *, long); -void do_general_protection(struct pt_regs *, long); -void do_nmi(struct pt_regs *, long); +dotraplinkage void do_divide_error(struct pt_regs *, long); +dotraplinkage void do_debug(struct pt_regs *, long); +dotraplinkage void do_nmi(struct pt_regs *, long); +dotraplinkage void do_int3(struct pt_regs *, long); +dotraplinkage void do_overflow(struct pt_regs *, long); +dotraplinkage void do_bounds(struct pt_regs *, long); +dotraplinkage void do_invalid_op(struct pt_regs *, long); +dotraplinkage void do_device_not_available(struct pt_regs *, long); +dotraplinkage void do_coprocessor_segment_overrun(struct pt_regs *, long); +dotraplinkage void do_invalid_TSS(struct pt_regs *, long); +dotraplinkage void do_segment_not_present(struct pt_regs *, long); +dotraplinkage void do_stack_segment(struct pt_regs *, long); +dotraplinkage void do_general_protection(struct pt_regs *, long); +dotraplinkage void do_page_fault(struct pt_regs *, unsigned long); +dotraplinkage void do_spurious_interrupt_bug(struct pt_regs *, long); +dotraplinkage void do_coprocessor_error(struct pt_regs *, long); +dotraplinkage void do_alignment_check(struct pt_regs *, long); +#ifdef CONFIG_X86_MCE +dotraplinkage void do_machine_check(struct pt_regs *, long); +#endif +dotraplinkage void do_simd_coprocessor_error(struct pt_regs *, long); +#ifdef CONFIG_X86_32 +dotraplinkage void do_iret_error(struct pt_regs *, long); +#endif static inline int get_si_code(unsigned long condition) { @@ -52,31 +73,9 @@ extern int panic_on_unrecovered_nmi; extern int kstack_depth_to_print; #ifdef CONFIG_X86_32 - -void do_iret_error(struct pt_regs *, long); -void do_int3(struct pt_regs *, long); -void do_debug(struct pt_regs *, long); void math_error(void __user *); -void do_coprocessor_error(struct pt_regs *, long); -void do_simd_coprocessor_error(struct pt_regs *, long); -void do_spurious_interrupt_bug(struct pt_regs *, long); unsigned long patch_espfix_desc(unsigned long, unsigned long); asmlinkage void math_emulate(long); +#endif -void do_page_fault(struct pt_regs *regs, unsigned long error_code); - -#else /* CONFIG_X86_32 */ - -asmlinkage void double_fault(void); - -asmlinkage void do_int3(struct pt_regs *, long); -asmlinkage void do_stack_segment(struct pt_regs *, long); -asmlinkage void do_debug(struct pt_regs *, unsigned long); -asmlinkage void do_coprocessor_error(struct pt_regs *, long); -asmlinkage void do_simd_coprocessor_error(struct pt_regs *, long); -asmlinkage void do_spurious_interrupt_bug(struct pt_regs *, long); - -asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code); - -#endif /* CONFIG_X86_32 */ #endif /* ASM_X86__TRAPS_H */ -- cgit v1.2.3-70-g09d2 From 3d2a71a596bd9c761c8487a2178e95f8a61da083 Mon Sep 17 00:00:00 2001 From: Alexander van Heukelum Date: Tue, 30 Sep 2008 18:41:37 +0200 Subject: x86, traps: converge do_debug handlers Make the x86_64-version and the i386-version of do_debug more similar. - introduce preempt_conditional_sti/cli to i386. The preempt-count is now elevated during the trap handler, like on x86_64. It does not run on a separate stack, however. - replace an open-coded "send_sigtrap" - copy some comments Signed-off-by: Alexander van Heukelum Signed-off-by: Ingo Molnar --- arch/x86/kernel/traps_32.c | 30 +++++++++++++++++++++--------- arch/x86/kernel/traps_64.c | 17 +++++++++-------- include/asm-x86/ptrace.h | 4 ---- 3 files changed, 30 insertions(+), 21 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index 7eb1203c909..953172af6e5 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -88,6 +88,20 @@ static inline void conditional_sti(struct pt_regs *regs) local_irq_enable(); } +static inline void preempt_conditional_sti(struct pt_regs *regs) +{ + inc_preempt_count(); + if (regs->flags & X86_EFLAGS_IF) + local_irq_enable(); +} + +static inline void preempt_conditional_cli(struct pt_regs *regs) +{ + if (regs->flags & X86_EFLAGS_IF) + local_irq_disable(); + dec_preempt_count(); +} + static inline void die_if_kernel(const char *str, struct pt_regs *regs, long err) { @@ -498,7 +512,7 @@ dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) { struct task_struct *tsk = current; - unsigned int condition; + unsigned long condition; int si_code; get_debugreg(condition, 6); @@ -512,9 +526,9 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code, SIGTRAP) == NOTIFY_STOP) return; + /* It's safe to allow irq's after DR6 has been saved */ - if (regs->flags & X86_EFLAGS_IF) - local_irq_enable(); + preempt_conditional_sti(regs); /* Mask out spurious debug traps due to lazy DR7 setting */ if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) { @@ -533,16 +547,11 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) * kernel space (but re-enable TF when returning to user mode). */ if (condition & DR_STEP) { - /* - * We already checked v86 mode above, so we can - * check for kernel mode by just checking the CPL - * of CS. - */ if (!user_mode(regs)) goto clear_TF_reenable; } - si_code = get_si_code((unsigned long)condition); + si_code = get_si_code(condition); /* Ok, finally something we can handle */ send_sigtrap(tsk, regs, error_code, si_code); @@ -552,15 +561,18 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) */ clear_dr7: set_debugreg(0, 7); + preempt_conditional_cli(regs); return; debug_vm86: handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1); + preempt_conditional_cli(regs); return; clear_TF_reenable: set_tsk_thread_flag(tsk, TIF_SINGLESTEP); regs->flags &= ~X86_EFLAGS_TF; + preempt_conditional_cli(regs); return; } diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index be73323ca95..a851eca6d04 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -380,7 +380,7 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) { struct task_struct *tsk = current; unsigned long condition; - siginfo_t info; + int si_code; get_debugreg(condition, 6); @@ -394,6 +394,7 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) SIGTRAP) == NOTIFY_STOP) return; + /* It's safe to allow irq's after DR6 has been saved */ preempt_conditional_sti(regs); /* Mask out spurious debug traps due to lazy DR7 setting */ @@ -402,6 +403,7 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) goto clear_dr7; } + /* Save debug status register where ptrace can see it */ tsk->thread.debugreg6 = condition; /* @@ -413,15 +415,14 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) goto clear_TF_reenable; } + si_code = get_si_code(condition); /* Ok, finally something we can handle */ - tsk->thread.trap_no = 1; - tsk->thread.error_code = error_code; - info.si_signo = SIGTRAP; - info.si_errno = 0; - info.si_code = get_si_code(condition); - info.si_addr = user_mode(regs) ? (void __user *)regs->ip : NULL; - force_sig_info(SIGTRAP, &info, tsk); + send_sigtrap(tsk, regs, error_code, si_code); + /* + * Disable additional traps. They'll be re-enabled when + * the signal is delivered. + */ clear_dr7: set_debugreg(0, 7); preempt_conditional_cli(regs); diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h index ac578f11c1c..a2025525a15 100644 --- a/include/asm-x86/ptrace.h +++ b/include/asm-x86/ptrace.h @@ -174,12 +174,8 @@ extern unsigned long profile_pc(struct pt_regs *regs); extern unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs); - -#ifdef CONFIG_X86_32 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code, int si_code); -#endif - void signal_fault(struct pt_regs *regs, void __user *frame, char *where); extern long syscall_trace_enter(struct pt_regs *); -- cgit v1.2.3-70-g09d2 From 699d2937d45d9dabc1772d0d07501ccc43885c23 Mon Sep 17 00:00:00 2001 From: Alexander van Heukelum Date: Fri, 3 Oct 2008 22:00:32 +0200 Subject: traps: x86: converge trap_init functions - set_system_gate on i386 is really set_system_trap_gate - set_system_gate on x86_64 is really set_system_intr_gate - ist=0 means no special stack switch is done: - introduce STACKFAULT_STACK, DOUBLEFAULT_STACK, NMI_STACK, DEBUG_STACK and MCE_STACK as on x86_64. - use the _ist variants with XXX_STACK set to zero - remove set_system_gate Signed-off-by: Alexander van Heukelum Signed-off-by: Ingo Molnar traps: x86: correct copy/paste bug: a trap is a GATE_TRAP Fix copy/paste/forgot-to-edit bug in desc.h. Signed-off-by: Alexander van Heukelum Signed-off-by: Ingo Molnar --- arch/x86/kernel/traps_32.c | 16 +++++++++------- arch/x86/kernel/traps_64.c | 6 +++--- include/asm-x86/desc.h | 14 +++++--------- include/asm-x86/page_32.h | 6 ++++++ 4 files changed, 23 insertions(+), 19 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index 953172af6e5..2c7ea382771 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -847,10 +847,12 @@ void __init trap_init(void) #endif set_intr_gate(0, ÷_error); - set_intr_gate(1, &debug); - set_intr_gate(2, &nmi); - set_system_intr_gate(3, &int3); /* int3 can be called from all */ - set_system_intr_gate(4, &overflow); /* int4 can be called from all */ + set_intr_gate_ist(1, &debug, DEBUG_STACK); + set_intr_gate_ist(2, &nmi, NMI_STACK); + /* int3 can be called from all */ + set_system_intr_gate_ist(3, &int3, DEBUG_STACK); + /* int4 can be called from all */ + set_system_intr_gate(4, &overflow); set_intr_gate(5, &bounds); set_intr_gate(6, &invalid_op); set_intr_gate(7, &device_not_available); @@ -858,14 +860,14 @@ void __init trap_init(void) set_intr_gate(9, &coprocessor_segment_overrun); set_intr_gate(10, &invalid_TSS); set_intr_gate(11, &segment_not_present); - set_intr_gate(12, &stack_segment); + set_intr_gate_ist(12, &stack_segment, STACKFAULT_STACK); set_intr_gate(13, &general_protection); set_intr_gate(14, &page_fault); set_intr_gate(15, &spurious_interrupt_bug); set_intr_gate(16, &coprocessor_error); set_intr_gate(17, &alignment_check); #ifdef CONFIG_X86_MCE - set_intr_gate(18, &machine_check); + set_intr_gate_ist(18, &machine_check, MCE_STACK); #endif set_intr_gate(19, &simd_coprocessor_error); @@ -881,7 +883,7 @@ void __init trap_init(void) printk("done.\n"); } - set_system_gate(SYSCALL_VECTOR, &system_call); + set_system_trap_gate(SYSCALL_VECTOR, &system_call); /* Reserve all the builtin and the syscall vector: */ for (i = 0; i < FIRST_EXTERNAL_VECTOR; i++) diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index a851eca6d04..ea091dfe0cd 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -647,9 +647,9 @@ void __init trap_init(void) set_intr_gate_ist(1, &debug, DEBUG_STACK); set_intr_gate_ist(2, &nmi, NMI_STACK); /* int3 can be called from all */ - set_system_gate_ist(3, &int3, DEBUG_STACK); + set_system_intr_gate_ist(3, &int3, DEBUG_STACK); /* int4 can be called from all */ - set_system_gate(4, &overflow); + set_system_intr_gate(4, &overflow); set_intr_gate(5, &bounds); set_intr_gate(6, &invalid_op); set_intr_gate(7, &device_not_available); @@ -669,7 +669,7 @@ void __init trap_init(void) set_intr_gate(19, &simd_coprocessor_error); #ifdef CONFIG_IA32_EMULATION - set_system_gate(IA32_SYSCALL_VECTOR, ia32_syscall); + set_system_intr_gate(IA32_SYSCALL_VECTOR, ia32_syscall); #endif /* * Should be a barrier for any external CPU state: diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h index ebc307817e9..f06adac7938 100644 --- a/include/asm-x86/desc.h +++ b/include/asm-x86/desc.h @@ -351,20 +351,16 @@ static inline void set_system_intr_gate(unsigned int n, void *addr) _set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS); } -static inline void set_trap_gate(unsigned int n, void *addr) +static inline void set_system_trap_gate(unsigned int n, void *addr) { BUG_ON((unsigned)n > 0xFF); - _set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS); + _set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS); } -static inline void set_system_gate(unsigned int n, void *addr) +static inline void set_trap_gate(unsigned int n, void *addr) { BUG_ON((unsigned)n > 0xFF); -#ifdef CONFIG_X86_32 - _set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS); -#else - _set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS); -#endif + _set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS); } static inline void set_task_gate(unsigned int n, unsigned int gdt_entry) @@ -379,7 +375,7 @@ static inline void set_intr_gate_ist(int n, void *addr, unsigned ist) _set_gate(n, GATE_INTERRUPT, addr, 0, ist, __KERNEL_CS); } -static inline void set_system_gate_ist(int n, void *addr, unsigned ist) +static inline void set_system_intr_gate_ist(int n, void *addr, unsigned ist) { BUG_ON((unsigned)n > 0xFF); _set_gate(n, GATE_INTERRUPT, addr, 0x3, ist, __KERNEL_CS); diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index 5d6a68a1067..e8d80d1de23 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -20,6 +20,12 @@ #endif #define THREAD_SIZE (PAGE_SIZE << THREAD_ORDER) +#define STACKFAULT_STACK 0 +#define DOUBLEFAULT_STACK 1 +#define NMI_STACK 0 +#define DEBUG_STACK 0 +#define MCE_STACK 0 +#define N_EXCEPTION_STACKS 1 #ifdef CONFIG_X86_PAE /* 44=32+12, the limit we can fit into an unsigned long pfn */ -- cgit v1.2.3-70-g09d2 From 7970479c4881e156a0c07c1a7fdc564c8e3b2bfc Mon Sep 17 00:00:00 2001 From: Alexander van Heukelum Date: Fri, 3 Oct 2008 22:00:36 +0200 Subject: traps: i386: expand clear_mem_error and remove from mach_traps.h This is the last user of clear_mem_error, which is defined only on i386. Expand the inline function and remove it from include/asm-x86/mach-default/mach_traps.h Signed-off-by: Alexander van Heukelum Signed-off-by: Ingo Molnar --- arch/x86/kernel/traps_32.c | 3 ++- include/asm-x86/mach-default/mach_traps.h | 6 ------ 2 files changed, 2 insertions(+), 7 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index 67953bbe193..01e7ca8c766 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -313,7 +313,8 @@ mem_parity_error(unsigned char reason, struct pt_regs *regs) printk(KERN_EMERG "Dazed and confused, but trying to continue\n"); /* Clear and disable the memory parity error line. */ - clear_mem_error(reason); + reason = (reason & 0xf) | 4; + outb(reason, 0x61); } static notrace __kprobes void diff --git a/include/asm-x86/mach-default/mach_traps.h b/include/asm-x86/mach-default/mach_traps.h index de9ac3f5c4c..ff8778f26b8 100644 --- a/include/asm-x86/mach-default/mach_traps.h +++ b/include/asm-x86/mach-default/mach_traps.h @@ -7,12 +7,6 @@ #include -static inline void clear_mem_error(unsigned char reason) -{ - reason = (reason & 0xf) | 4; - outb(reason, 0x61); -} - static inline unsigned char get_nmi_reason(void) { return inb(0x61); -- cgit v1.2.3-70-g09d2 From c1d518c8422ff7d3f377958771b265753028579c Mon Sep 17 00:00:00 2001 From: Alexander van Heukelum Date: Fri, 3 Oct 2008 23:17:11 +0200 Subject: traps: x86: various noop-changes preparing for unification of traps_xx.c - reordering include files - whitespace changes - comment changes - removed unused bad_intr() - make default_do_nmi static Signed-off-by: Alexander van Heukelum Signed-off-by: Ingo Molnar --- arch/x86/kernel/traps_32.c | 28 +++++++++++--------- arch/x86/kernel/traps_64.c | 65 ++++++++++++++++++++++++++++++---------------- include/asm-x86/nmi.h | 4 --- 3 files changed, 58 insertions(+), 39 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index 01e7ca8c766..076739863d2 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -7,13 +7,11 @@ */ /* - * 'Traps.c' handles hardware traps and faults after we have saved some - * state in 'asm.s'. + * Handle hardware traps and faults. */ #include #include #include -#include #include #include #include @@ -32,6 +30,8 @@ #include #include #include +#include +#include #ifdef CONFIG_EISA #include @@ -46,22 +46,26 @@ #include #endif -#include -#include #include #include +#include #include #include #include #include +#include #include #include + +#include + +#include +#include #include #include #include #include -#include "mach_traps.h" #include "cpu/mcheck/mce.h" DECLARE_BITMAP(used_vectors, NR_VECTORS); @@ -340,7 +344,8 @@ io_check_error(unsigned char reason, struct pt_regs *regs) static notrace __kprobes void unknown_nmi_error(unsigned char reason, struct pt_regs *regs) { - if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) + if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == + NOTIFY_STOP) return; #ifdef CONFIG_MCA /* @@ -446,13 +451,9 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) dotraplinkage notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code) { - int cpu; - nmi_enter(); - cpu = smp_processor_id(); - - ++nmi_count(cpu); + { int cpu; cpu = smp_processor_id(); ++nmi_count(cpu); } if (!ignore_nmis) default_do_nmi(regs); @@ -472,6 +473,7 @@ void restart_nmi(void) acpi_nmi_enable(); } +/* May run on IST stack. */ dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) { #ifdef CONFIG_KPROBES @@ -510,6 +512,8 @@ dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) * about restoring all the debug state, and ptrace doesn't have to * find every occurrence of the TF bit that could be saved away even * by user code) + * + * May run on IST stack. */ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) { diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index 71c2629997b..22fe62a24ed 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -7,10 +7,8 @@ */ /* - * 'Traps.c' handles hardware traps and faults after we have saved some - * state in 'entry.S'. + * Handle hardware traps and faults. */ -#include #include #include #include @@ -41,18 +39,21 @@ #include #include +#include #include #include #include #include +#include #include #include + +#include + #include #include #include -#include -#include static int ignore_nmis; @@ -73,8 +74,6 @@ static inline void preempt_conditional_cli(struct pt_regs *regs) { if (regs->flags & X86_EFLAGS_IF) local_irq_disable(); - /* Make sure to not schedule here because we could be running - on an exception stack. */ dec_preempt_count(); } @@ -228,9 +227,12 @@ gp_in_kernel: static notrace __kprobes void mem_parity_error(unsigned char reason, struct pt_regs *regs) { - printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n", - reason); - printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n"); + printk(KERN_EMERG + "Uhhuh. NMI received for unknown reason %02x on CPU %d.\n", + reason, smp_processor_id()); + + printk(KERN_EMERG + "You have some hardware problem, likely on the PCI bus.\n"); #if defined(CONFIG_EDAC) if (edac_handler_set()) { @@ -275,19 +277,18 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs) if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) return; - printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n", - reason); - printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n"); + printk(KERN_EMERG + "Uhhuh. NMI received for unknown reason %02x on CPU %d.\n", + reason, smp_processor_id()); + printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n"); if (panic_on_unrecovered_nmi) panic("NMI: Not continuing"); printk(KERN_EMERG "Dazed and confused, but trying to continue\n"); } -/* Runs on IST stack. This code must keep interrupts off all the time. - Nested NMIs are prevented by the CPU. */ -asmlinkage notrace __kprobes void default_do_nmi(struct pt_regs *regs) +static notrace __kprobes void default_do_nmi(struct pt_regs *regs) { unsigned char reason = 0; int cpu; @@ -348,7 +349,7 @@ void restart_nmi(void) acpi_nmi_enable(); } -/* runs on IST stack. */ +/* May run on IST stack. */ dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) { if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) @@ -381,7 +382,30 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) return regs; } -/* runs on IST stack. */ +/* + * Our handling of the processor debug registers is non-trivial. + * We do not clear them on entry and exit from the kernel. Therefore + * it is possible to get a watchpoint trap here from inside the kernel. + * However, the code in ./ptrace.c has ensured that the user can + * only set watchpoints on userspace addresses. Therefore the in-kernel + * watchpoint trap can only occur in code which is reading/writing + * from user space. Such code must not hold kernel locks (since it + * can equally take a page fault), therefore it is safe to call + * force_sig_info even though that claims and releases locks. + * + * Code in ./signal.c ensures that the debug control register + * is restored before we deliver any signal, and therefore that + * user code runs with the correct debug control register even though + * we clear it here. + * + * Being careful here means that we don't have to be as careful in a + * lot of more complicated places (task switching can be a bit lazy + * about restoring all the debug state, and ptrace doesn't have to + * find every occurrence of the TF bit that could be saved away even + * by user code) + * + * May run on IST stack. + */ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) { struct task_struct *tsk = current; @@ -525,11 +549,6 @@ dotraplinkage void do_coprocessor_error(struct pt_regs *regs, long error_code) math_error((void __user *)regs->ip); } -asmlinkage void bad_intr(void) -{ - printk("bad interrupt"); -} - static void simd_math_error(void __user *ip) { struct task_struct *task; diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h index d5e715f024d..a53f829a97c 100644 --- a/include/asm-x86/nmi.h +++ b/include/asm-x86/nmi.h @@ -15,10 +15,6 @@ */ int do_nmi_callback(struct pt_regs *regs, int cpu); -#ifdef CONFIG_X86_64 -extern void default_do_nmi(struct pt_regs *); -#endif - extern void die_nmi(char *str, struct pt_regs *regs, int do_panic); extern int check_nmi_watchdog(void); extern int nmi_watchdog_enabled; -- cgit v1.2.3-70-g09d2 From c1a2f4b10852ce68e70f7e4c53600c36cc63ea45 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 14 Sep 2008 02:33:12 -0700 Subject: x86: change early_ioremap to use slots instead of nesting so we could remove the requirement that one needs to call early_iounmap() in exactly reverse order of early_ioremap(). Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/mm/ioremap.c | 77 +++++++++++++++++++++++++++++++++------------ include/asm-x86/fixmap_32.h | 4 +-- include/asm-x86/fixmap_64.h | 4 +-- 3 files changed, 61 insertions(+), 24 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 8876220d906..e4c43ec71b2 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -616,16 +616,22 @@ static inline void __init early_clear_fixmap(enum fixed_addresses idx) __early_set_fixmap(idx, 0, __pgprot(0)); } - -static int __initdata early_ioremap_nested; - +static void *prev_map[FIX_BTMAPS_SLOTS] __initdata; +static unsigned long prev_size[FIX_BTMAPS_SLOTS] __initdata; static int __init check_early_ioremap_leak(void) { - if (!early_ioremap_nested) + int count = 0; + int i; + + for (i = 0; i < FIX_BTMAPS_SLOTS; i++) + if (prev_map[i]) + count++; + + if (!count) return 0; WARN(1, KERN_WARNING "Debug warning: early ioremap leak of %d areas detected.\n", - early_ioremap_nested); + count); printk(KERN_WARNING "please boot with early_ioremap_debug and report the dmesg.\n"); @@ -636,15 +642,30 @@ late_initcall(check_early_ioremap_leak); static void __init *__early_ioremap(unsigned long phys_addr, unsigned long size, pgprot_t prot) { unsigned long offset, last_addr; - unsigned int nrpages, nesting; + unsigned int nrpages; enum fixed_addresses idx0, idx; + int i, slot; WARN_ON(system_state != SYSTEM_BOOTING); - nesting = early_ioremap_nested; + slot = -1; + for (i = 0; i < FIX_BTMAPS_SLOTS; i++) { + if (!prev_map[i]) { + slot = i; + break; + } + } + + if (slot < 0) { + printk(KERN_INFO "early_iomap(%08lx, %08lx) not found slot\n", + phys_addr, size); + WARN_ON(1); + return NULL; + } + if (early_ioremap_debug) { printk(KERN_INFO "early_ioremap(%08lx, %08lx) [%d] => ", - phys_addr, size, nesting); + phys_addr, size, slot); dump_stack(); } @@ -655,11 +676,7 @@ static void __init *__early_ioremap(unsigned long phys_addr, unsigned long size, return NULL; } - if (nesting >= FIX_BTMAPS_NESTING) { - WARN_ON(1); - return NULL; - } - early_ioremap_nested++; + prev_size[slot] = size; /* * Mappings have to be page-aligned */ @@ -679,7 +696,7 @@ static void __init *__early_ioremap(unsigned long phys_addr, unsigned long size, /* * Ok, go for it.. */ - idx0 = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*nesting; + idx0 = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*slot; idx = idx0; while (nrpages > 0) { early_set_fixmap(idx, phys_addr, prot); @@ -690,7 +707,8 @@ static void __init *__early_ioremap(unsigned long phys_addr, unsigned long size, if (early_ioremap_debug) printk(KERN_CONT "%08lx + %08lx\n", offset, fix_to_virt(idx0)); - return (void *) (offset + fix_to_virt(idx0)); + prev_map[slot] = (void *) (offset + fix_to_virt(idx0)); + return prev_map[slot]; } /* Remap an IO device */ @@ -711,15 +729,33 @@ void __init early_iounmap(void *addr, unsigned long size) unsigned long offset; unsigned int nrpages; enum fixed_addresses idx; - int nesting; + int i, slot; + + slot = -1; + for (i = 0; i < FIX_BTMAPS_SLOTS; i++) { + if (prev_map[i] == addr) { + slot = i; + break; + } + } - nesting = --early_ioremap_nested; - if (WARN_ON(nesting < 0)) + if (slot < 0) { + printk(KERN_INFO "early_iounmap(%p, %08lx) not found slot\n", + addr, size); + WARN_ON(1); + return; + } + + if (prev_size[slot] != size) { + printk(KERN_INFO "early_iounmap(%p, %08lx) [%d] size not consistent %08lx\n", + addr, size, slot, prev_size[slot]); + WARN_ON(1); return; + } if (early_ioremap_debug) { printk(KERN_INFO "early_iounmap(%p, %08lx) [%d]\n", addr, - size, nesting); + size, slot); dump_stack(); } @@ -731,12 +767,13 @@ void __init early_iounmap(void *addr, unsigned long size) offset = virt_addr & ~PAGE_MASK; nrpages = PAGE_ALIGN(offset + size - 1) >> PAGE_SHIFT; - idx = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*nesting; + idx = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*slot; while (nrpages > 0) { early_clear_fixmap(idx); --idx; --nrpages; } + prev_map[slot] = 0; } void __this_fixmap_does_not_exist(void) diff --git a/include/asm-x86/fixmap_32.h b/include/asm-x86/fixmap_32.h index 784e3e75986..8844002da0e 100644 --- a/include/asm-x86/fixmap_32.h +++ b/include/asm-x86/fixmap_32.h @@ -94,10 +94,10 @@ enum fixed_addresses { * can have a single pgd entry and a single pte table: */ #define NR_FIX_BTMAPS 64 -#define FIX_BTMAPS_NESTING 4 +#define FIX_BTMAPS_SLOTS 4 FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 - (__end_of_permanent_fixed_addresses & 255), - FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1, + FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1, FIX_WP_TEST, #ifdef CONFIG_ACPI FIX_ACPI_BEGIN, diff --git a/include/asm-x86/fixmap_64.h b/include/asm-x86/fixmap_64.h index 33df64fd0e1..dab4751d130 100644 --- a/include/asm-x86/fixmap_64.h +++ b/include/asm-x86/fixmap_64.h @@ -65,10 +65,10 @@ enum fixed_addresses { * can have a single pgd entry and a single pte table: */ #define NR_FIX_BTMAPS 64 -#define FIX_BTMAPS_NESTING 4 +#define FIX_BTMAPS_SLOTS 4 FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 - (__end_of_permanent_fixed_addresses & 255), - FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1, + FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1, __end_of_fixed_addresses }; -- cgit v1.2.3-70-g09d2 From c26c56c0f40e200e61d1390629c806f6adaffbcc Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 13 Oct 2008 10:37:48 +0100 Subject: tty: Cris has a nice RS485 ioctl so we should steal it JP Tosoni observed: "About a RS485 ioctl: could you consider the attached files which are already in the Linux kernel (in include/asm-cris). They define a TIOCSERSETRS485 (ioctl.h), and the data structure (rs485.h) with allows to specify timings. Sounds just like what we want ?" and he's right: sort of. Rework the structure to use flag bits and make the time delay a fixed sized field so we don't get 32/64bit problems. Add the ioctls to x86 so that people know what to add to their platform of choice. Signed-off-by: Alan Cox Signed-off-by: Linus Torvalds --- include/asm-x86/ioctls.h | 2 ++ include/linux/serial.h | 16 ++++++++++++++++ 2 files changed, 18 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/ioctls.h b/include/asm-x86/ioctls.h index 33660351239..1a8ac45b28b 100644 --- a/include/asm-x86/ioctls.h +++ b/include/asm-x86/ioctls.h @@ -51,6 +51,8 @@ #define TCSETS2 _IOW('T', 0x2B, struct termios2) #define TCSETSW2 _IOW('T', 0x2C, struct termios2) #define TCSETSF2 _IOW('T', 0x2D, struct termios2) +#define TIOCGRS485 0x542E +#define TIOCSRS485 0x542F #define TIOCGPTN _IOR('T', 0x30, unsigned int) /* Get Pty Number (of pty-mux device) */ #define TIOCSPTLCK _IOW('T', 0x31, int) /* Lock/unlock Pty */ diff --git a/include/linux/serial.h b/include/linux/serial.h index deb714314fb..1ea8d9265bf 100644 --- a/include/linux/serial.h +++ b/include/linux/serial.h @@ -173,6 +173,22 @@ struct serial_icounter_struct { int reserved[9]; }; +/* + * Serial interface for controlling RS485 settings on chips with suitable + * support. Set with TIOCSRS485 and get with TIOCGRS485 if supported by your + * platform. The set function returns the new state, with any unsupported bits + * reverted appropriately. + */ + +struct serial_rs485 { + __u32 flags; /* RS485 feature flags */ +#define SER_RS485_ENABLED (1 << 0) +#define SER_RS485_RTS_ON_SEND (1 << 1) +#define SER_RS485_RTS_AFTER_SEND (1 << 2) + __u32 delay_rts_before_send; /* Milliseconds */ + __u32 padding[6]; /* Memory is cheap, new structs + are a royal PITA .. */ +}; #ifdef __KERNEL__ #include -- cgit v1.2.3-70-g09d2 From 1d65b4a088de407e99714fdc27862449db04fb5c Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 13 Oct 2008 10:38:18 +0100 Subject: tty: Add termiox We need a way to describe the various additional modes and flow control features that random weird hardware shows up and software such as wine wants to emulate as Windows supports them. TCGETX/TCSETX and the termiox ioctl are a SYS5 extension that we might as well adopt. This patches adds the structures and the basic ioctl interfaces when the TCGETX etc defines are added for an architecture. Drivers wishing to use this stuff need to add new methods. Signed-off-by: Alan Cox Signed-off-by: Linus Torvalds --- drivers/char/tty_ioctl.c | 58 ++++++++++++++++++++++++++++++++++++++++++++++ include/asm-x86/ioctls.h | 4 ++++ include/linux/termios.h | 15 ++++++++++++ include/linux/tty.h | 1 + include/linux/tty_driver.h | 9 +++++++ 5 files changed, 87 insertions(+) (limited to 'include/asm-x86') diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c index bf34e459742..93bfa1d6cc9 100644 --- a/drivers/char/tty_ioctl.c +++ b/drivers/char/tty_ioctl.c @@ -579,6 +579,50 @@ static int get_termio(struct tty_struct *tty, struct termio __user *termio) return 0; } + +#ifdef TCGETX + +/** + * set_termiox - set termiox fields if possible + * @tty: terminal + * @arg: termiox structure from user + * @opt: option flags for ioctl type + * + * Implement the device calling points for the SYS5 termiox ioctl + * interface in Linux + */ + +static int set_termiox(struct tty_struct *tty, void __user *arg, int opt) +{ + struct termiox tnew; + struct tty_ldisc *ld; + + if (tty->termiox == NULL) + return -EINVAL; + if (copy_from_user(&tnew, arg, sizeof(struct termiox))) + return -EFAULT; + + ld = tty_ldisc_ref(tty); + if (ld != NULL) { + if ((opt & TERMIOS_FLUSH) && ld->ops->flush_buffer) + ld->ops->flush_buffer(tty); + tty_ldisc_deref(ld); + } + if (opt & TERMIOS_WAIT) { + tty_wait_until_sent(tty, 0); + if (signal_pending(current)) + return -EINTR; + } + + mutex_lock(&tty->termios_mutex); + if (tty->ops->set_termiox) + tty->ops->set_termiox(tty, &tnew); + mutex_unlock(&tty->termios_mutex); + return 0; +} + +#endif + static unsigned long inq_canon(struct tty_struct *tty) { int nr, head, tail; @@ -936,6 +980,20 @@ int tty_mode_ioctl(struct tty_struct *tty, struct file *file, return -EFAULT; return 0; #endif +#ifdef TCGETX + case TCGETX: + if (real_tty->termiox == NULL) + return -EINVAL; + if (copy_to_user(p, real_tty->termiox, sizeof(struct termiox))) + return -EFAULT; + return 0; + case TCSETX: + return set_termiox(real_tty, p, 0); + case TCSETXW: + return set_termiox(real_tty, p, TERMIOS_WAIT); + case TCSETXF: + return set_termiox(real_tty, p, TERMIOS_FLUSH); +#endif case TIOCGSOFTCAR: /* FIXME: for correctness we may need to take the termios lock here - review */ diff --git a/include/asm-x86/ioctls.h b/include/asm-x86/ioctls.h index 1a8ac45b28b..06752a64904 100644 --- a/include/asm-x86/ioctls.h +++ b/include/asm-x86/ioctls.h @@ -56,6 +56,10 @@ #define TIOCGPTN _IOR('T', 0x30, unsigned int) /* Get Pty Number (of pty-mux device) */ #define TIOCSPTLCK _IOW('T', 0x31, int) /* Lock/unlock Pty */ +#define TCGETX 0x5432 /* SYS5 TCGETX compatibility */ +#define TCSETX 0x5433 +#define TCSETXF 0x5434 +#define TCSETXW 0x5435 #define FIONCLEX 0x5450 #define FIOCLEX 0x5451 diff --git a/include/linux/termios.h b/include/linux/termios.h index 478662889f4..2acd0c1f8a2 100644 --- a/include/linux/termios.h +++ b/include/linux/termios.h @@ -4,4 +4,19 @@ #include #include +#define NFF 5 + +struct termiox +{ + __u16 x_hflag; + __u16 x_cflag; + __u16 x_rflag[NFF]; + __u16 x_sflag; +}; + +#define RTSXOFF 0x0001 /* RTS flow control on input */ +#define CTSXON 0x0002 /* CTS flow control on output */ +#define DTRXOFF 0x0004 /* DTR flow control on input */ +#define DSRXON 0x0008 /* DCD flow control on output */ + #endif diff --git a/include/linux/tty.h b/include/linux/tty.h index b6e6c26883e..b64d10b6654 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -219,6 +219,7 @@ struct tty_struct { spinlock_t ctrl_lock; /* Termios values are protected by the termios mutex */ struct ktermios *termios, *termios_locked; + struct termiox *termiox; /* May be NULL for unsupported */ char name[64]; struct pid *pgrp; /* Protected by ctrl lock */ struct pid *session; diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h index 16d27944c32..ac6e58e26b7 100644 --- a/include/linux/tty_driver.h +++ b/include/linux/tty_driver.h @@ -180,6 +180,14 @@ * not force errors here if they are not resizable objects (eg a serial * line). See tty_do_resize() if you need to wrap the standard method * in your own logic - the usual case. + * + * void (*set_termiox)(struct tty_struct *tty, struct termiox *new); + * + * Called when the device receives a termiox based ioctl. Passes down + * the requested data from user space. This method will not be invoked + * unless the tty also has a valid tty->termiox pointer. + * + * Optional: Called under the termios lock */ #include @@ -220,6 +228,7 @@ struct tty_operations { unsigned int set, unsigned int clear); int (*resize)(struct tty_struct *tty, struct tty_struct *real_tty, struct winsize *ws); + int (*set_termiox)(struct tty_struct *tty, struct termiox *tnew); #ifdef CONFIG_CONSOLE_POLL int (*poll_init)(struct tty_driver *driver, int line, char *options); int (*poll_get_char)(struct tty_driver *driver, int line); -- cgit v1.2.3-70-g09d2 From 5fdbf9765b7ba6a45100851154768de703d51e76 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 27 Jun 2008 14:58:02 -0300 Subject: KVM: x86: accessors for guest registers As suggested by Avi, introduce accessors to read/write guest registers. This simplifies the ->cache_regs/->decache_regs interface, and improves register caching which is important for VMX, where the cost of vmcs_read/vmcs_write is significant. [avi: fix warnings] Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/kvm_cache_regs.h | 32 +++++ arch/x86/kvm/lapic.c | 4 +- arch/x86/kvm/svm.c | 56 ++++----- arch/x86/kvm/vmx.c | 103 ++++++++-------- arch/x86/kvm/x86.c | 268 ++++++++++++++++++++++-------------------- arch/x86/kvm/x86_emulate.c | 19 +-- include/asm-x86/kvm_host.h | 15 ++- 7 files changed, 264 insertions(+), 233 deletions(-) create mode 100644 arch/x86/kvm/kvm_cache_regs.h (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/kvm_cache_regs.h b/arch/x86/kvm/kvm_cache_regs.h new file mode 100644 index 00000000000..1ff819dce7d --- /dev/null +++ b/arch/x86/kvm/kvm_cache_regs.h @@ -0,0 +1,32 @@ +#ifndef ASM_KVM_CACHE_REGS_H +#define ASM_KVM_CACHE_REGS_H + +static inline unsigned long kvm_register_read(struct kvm_vcpu *vcpu, + enum kvm_reg reg) +{ + if (!test_bit(reg, (unsigned long *)&vcpu->arch.regs_avail)) + kvm_x86_ops->cache_reg(vcpu, reg); + + return vcpu->arch.regs[reg]; +} + +static inline void kvm_register_write(struct kvm_vcpu *vcpu, + enum kvm_reg reg, + unsigned long val) +{ + vcpu->arch.regs[reg] = val; + __set_bit(reg, (unsigned long *)&vcpu->arch.regs_dirty); + __set_bit(reg, (unsigned long *)&vcpu->arch.regs_avail); +} + +static inline unsigned long kvm_rip_read(struct kvm_vcpu *vcpu) +{ + return kvm_register_read(vcpu, VCPU_REGS_RIP); +} + +static inline void kvm_rip_write(struct kvm_vcpu *vcpu, unsigned long val) +{ + kvm_register_write(vcpu, VCPU_REGS_RIP, val); +} + +#endif diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 73f43de69f6..9fde0ac2426 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -32,6 +32,7 @@ #include #include #include +#include "kvm_cache_regs.h" #include "irq.h" #define PRId64 "d" @@ -558,8 +559,7 @@ static void __report_tpr_access(struct kvm_lapic *apic, bool write) struct kvm_run *run = vcpu->run; set_bit(KVM_REQ_REPORT_TPR_ACCESS, &vcpu->requests); - kvm_x86_ops->cache_regs(vcpu); - run->tpr_access.rip = vcpu->arch.rip; + run->tpr_access.rip = kvm_rip_read(vcpu); run->tpr_access.is_write = write; } diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 8233b86c778..54b0bf33e21 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -18,6 +18,7 @@ #include "kvm_svm.h" #include "irq.h" #include "mmu.h" +#include "kvm_cache_regs.h" #include #include @@ -236,13 +237,11 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu) printk(KERN_DEBUG "%s: NOP\n", __func__); return; } - if (svm->next_rip - svm->vmcb->save.rip > MAX_INST_SIZE) - printk(KERN_ERR "%s: ip 0x%llx next 0x%llx\n", - __func__, - svm->vmcb->save.rip, - svm->next_rip); + if (svm->next_rip - kvm_rip_read(vcpu) > MAX_INST_SIZE) + printk(KERN_ERR "%s: ip 0x%lx next 0x%llx\n", + __func__, kvm_rip_read(vcpu), svm->next_rip); - vcpu->arch.rip = svm->vmcb->save.rip = svm->next_rip; + kvm_rip_write(vcpu, svm->next_rip); svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK; vcpu->arch.interrupt_window_open = 1; @@ -581,6 +580,7 @@ static void init_vmcb(struct vcpu_svm *svm) save->dr7 = 0x400; save->rflags = 2; save->rip = 0x0000fff0; + svm->vcpu.arch.regs[VCPU_REGS_RIP] = save->rip; /* * cr0 val on cpu init should be 0x60000010, we enable cpu @@ -615,10 +615,12 @@ static int svm_vcpu_reset(struct kvm_vcpu *vcpu) init_vmcb(svm); if (vcpu->vcpu_id != 0) { - svm->vmcb->save.rip = 0; + kvm_rip_write(vcpu, 0); svm->vmcb->save.cs.base = svm->vcpu.arch.sipi_vector << 12; svm->vmcb->save.cs.selector = svm->vcpu.arch.sipi_vector << 8; } + vcpu->arch.regs_avail = ~0; + vcpu->arch.regs_dirty = ~0; return 0; } @@ -721,23 +723,6 @@ static void svm_vcpu_put(struct kvm_vcpu *vcpu) rdtscll(vcpu->arch.host_tsc); } -static void svm_cache_regs(struct kvm_vcpu *vcpu) -{ - struct vcpu_svm *svm = to_svm(vcpu); - - vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax; - vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; - vcpu->arch.rip = svm->vmcb->save.rip; -} - -static void svm_decache_regs(struct kvm_vcpu *vcpu) -{ - struct vcpu_svm *svm = to_svm(vcpu); - svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; - svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; - svm->vmcb->save.rip = vcpu->arch.rip; -} - static unsigned long svm_get_rflags(struct kvm_vcpu *vcpu) { return to_svm(vcpu)->vmcb->save.rflags; @@ -1139,14 +1124,14 @@ static int nop_on_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) static int halt_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { - svm->next_rip = svm->vmcb->save.rip + 1; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 1; skip_emulated_instruction(&svm->vcpu); return kvm_emulate_halt(&svm->vcpu); } static int vmmcall_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { - svm->next_rip = svm->vmcb->save.rip + 3; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; skip_emulated_instruction(&svm->vcpu); kvm_emulate_hypercall(&svm->vcpu); return 1; @@ -1178,7 +1163,7 @@ static int task_switch_interception(struct vcpu_svm *svm, static int cpuid_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { - svm->next_rip = svm->vmcb->save.rip + 2; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 2; kvm_emulate_cpuid(&svm->vcpu); return 1; } @@ -1273,9 +1258,9 @@ static int rdmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) KVMTRACE_3D(MSR_READ, &svm->vcpu, ecx, (u32)data, (u32)(data >> 32), handler); - svm->vmcb->save.rax = data & 0xffffffff; + svm->vcpu.arch.regs[VCPU_REGS_RAX] = data & 0xffffffff; svm->vcpu.arch.regs[VCPU_REGS_RDX] = data >> 32; - svm->next_rip = svm->vmcb->save.rip + 2; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 2; skip_emulated_instruction(&svm->vcpu); } return 1; @@ -1359,13 +1344,13 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data) static int wrmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX]; - u64 data = (svm->vmcb->save.rax & -1u) + u64 data = (svm->vcpu.arch.regs[VCPU_REGS_RAX] & -1u) | ((u64)(svm->vcpu.arch.regs[VCPU_REGS_RDX] & -1u) << 32); KVMTRACE_3D(MSR_WRITE, &svm->vcpu, ecx, (u32)data, (u32)(data >> 32), handler); - svm->next_rip = svm->vmcb->save.rip + 2; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 2; if (svm_set_msr(&svm->vcpu, ecx, data)) kvm_inject_gp(&svm->vcpu, 0); else @@ -1723,6 +1708,10 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) u16 gs_selector; u16 ldt_selector; + svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; + svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; + svm->vmcb->save.rip = vcpu->arch.regs[VCPU_REGS_RIP]; + pre_svm_run(svm); sync_lapic_to_cr8(vcpu); @@ -1858,6 +1847,9 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) load_db_regs(svm->host_db_regs); vcpu->arch.cr2 = svm->vmcb->save.cr2; + vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax; + vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; + vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip; write_dr6(svm->host_dr6); write_dr7(svm->host_dr7); @@ -1977,8 +1969,6 @@ static struct kvm_x86_ops svm_x86_ops = { .set_gdt = svm_set_gdt, .get_dr = svm_get_dr, .set_dr = svm_set_dr, - .cache_regs = svm_cache_regs, - .decache_regs = svm_decache_regs, .get_rflags = svm_get_rflags, .set_rflags = svm_set_rflags, diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 81dac721ec3..5a3a0326c27 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -26,6 +26,7 @@ #include #include #include +#include "kvm_cache_regs.h" #include #include @@ -715,9 +716,9 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu) unsigned long rip; u32 interruptibility; - rip = vmcs_readl(GUEST_RIP); + rip = kvm_rip_read(vcpu); rip += vmcs_read32(VM_EXIT_INSTRUCTION_LEN); - vmcs_writel(GUEST_RIP, rip); + kvm_rip_write(vcpu, rip); /* * We emulated an instruction, so temporary interrupt blocking @@ -947,24 +948,19 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) return ret; } -/* - * Sync the rsp and rip registers into the vcpu structure. This allows - * registers to be accessed by indexing vcpu->arch.regs. - */ -static void vcpu_load_rsp_rip(struct kvm_vcpu *vcpu) -{ - vcpu->arch.regs[VCPU_REGS_RSP] = vmcs_readl(GUEST_RSP); - vcpu->arch.rip = vmcs_readl(GUEST_RIP); -} - -/* - * Syncs rsp and rip back into the vmcs. Should be called after possible - * modification. - */ -static void vcpu_put_rsp_rip(struct kvm_vcpu *vcpu) +static void vmx_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg) { - vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]); - vmcs_writel(GUEST_RIP, vcpu->arch.rip); + __set_bit(reg, (unsigned long *)&vcpu->arch.regs_avail); + switch (reg) { + case VCPU_REGS_RSP: + vcpu->arch.regs[VCPU_REGS_RSP] = vmcs_readl(GUEST_RSP); + break; + case VCPU_REGS_RIP: + vcpu->arch.regs[VCPU_REGS_RIP] = vmcs_readl(GUEST_RIP); + break; + default: + break; + } } static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg) @@ -2019,6 +2015,7 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) u64 msr; int ret; + vcpu->arch.regs_avail = ~((1 << VCPU_REGS_RIP) | (1 << VCPU_REGS_RSP)); down_read(&vcpu->kvm->slots_lock); if (!init_rmode(vmx->vcpu.kvm)) { ret = -ENOMEM; @@ -2072,10 +2069,10 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) vmcs_writel(GUEST_RFLAGS, 0x02); if (vmx->vcpu.vcpu_id == 0) - vmcs_writel(GUEST_RIP, 0xfff0); + kvm_rip_write(vcpu, 0xfff0); else - vmcs_writel(GUEST_RIP, 0); - vmcs_writel(GUEST_RSP, 0); + kvm_rip_write(vcpu, 0); + kvm_register_write(vcpu, VCPU_REGS_RSP, 0); /* todo: dr0 = dr1 = dr2 = dr3 = 0; dr6 = 0xffff0ff0 */ vmcs_writel(GUEST_DR7, 0x400); @@ -2139,11 +2136,11 @@ static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq) if (vcpu->arch.rmode.active) { vmx->rmode.irq.pending = true; vmx->rmode.irq.vector = irq; - vmx->rmode.irq.rip = vmcs_readl(GUEST_RIP); + vmx->rmode.irq.rip = kvm_rip_read(vcpu); vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, irq | INTR_TYPE_SOFT_INTR | INTR_INFO_VALID_MASK); vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1); - vmcs_writel(GUEST_RIP, vmx->rmode.irq.rip - 1); + kvm_rip_write(vcpu, vmx->rmode.irq.rip - 1); return; } vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, @@ -2288,7 +2285,7 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } error_code = 0; - rip = vmcs_readl(GUEST_RIP); + rip = kvm_rip_read(vcpu); if (intr_info & INTR_INFO_DELIVER_CODE_MASK) error_code = vmcs_read32(VM_EXIT_INTR_ERROR_CODE); if (is_page_fault(intr_info)) { @@ -2386,27 +2383,25 @@ static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) reg = (exit_qualification >> 8) & 15; switch ((exit_qualification >> 4) & 3) { case 0: /* mov to cr */ - KVMTRACE_3D(CR_WRITE, vcpu, (u32)cr, (u32)vcpu->arch.regs[reg], - (u32)((u64)vcpu->arch.regs[reg] >> 32), handler); + KVMTRACE_3D(CR_WRITE, vcpu, (u32)cr, + (u32)kvm_register_read(vcpu, reg), + (u32)((u64)kvm_register_read(vcpu, reg) >> 32), + handler); switch (cr) { case 0: - vcpu_load_rsp_rip(vcpu); - kvm_set_cr0(vcpu, vcpu->arch.regs[reg]); + kvm_set_cr0(vcpu, kvm_register_read(vcpu, reg)); skip_emulated_instruction(vcpu); return 1; case 3: - vcpu_load_rsp_rip(vcpu); - kvm_set_cr3(vcpu, vcpu->arch.regs[reg]); + kvm_set_cr3(vcpu, kvm_register_read(vcpu, reg)); skip_emulated_instruction(vcpu); return 1; case 4: - vcpu_load_rsp_rip(vcpu); - kvm_set_cr4(vcpu, vcpu->arch.regs[reg]); + kvm_set_cr4(vcpu, kvm_register_read(vcpu, reg)); skip_emulated_instruction(vcpu); return 1; case 8: - vcpu_load_rsp_rip(vcpu); - kvm_set_cr8(vcpu, vcpu->arch.regs[reg]); + kvm_set_cr8(vcpu, kvm_register_read(vcpu, reg)); skip_emulated_instruction(vcpu); if (irqchip_in_kernel(vcpu->kvm)) return 1; @@ -2415,7 +2410,6 @@ static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) }; break; case 2: /* clts */ - vcpu_load_rsp_rip(vcpu); vmx_fpu_deactivate(vcpu); vcpu->arch.cr0 &= ~X86_CR0_TS; vmcs_writel(CR0_READ_SHADOW, vcpu->arch.cr0); @@ -2426,21 +2420,17 @@ static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) case 1: /*mov from cr*/ switch (cr) { case 3: - vcpu_load_rsp_rip(vcpu); - vcpu->arch.regs[reg] = vcpu->arch.cr3; - vcpu_put_rsp_rip(vcpu); + kvm_register_write(vcpu, reg, vcpu->arch.cr3); KVMTRACE_3D(CR_READ, vcpu, (u32)cr, - (u32)vcpu->arch.regs[reg], - (u32)((u64)vcpu->arch.regs[reg] >> 32), + (u32)kvm_register_read(vcpu, reg), + (u32)((u64)kvm_register_read(vcpu, reg) >> 32), handler); skip_emulated_instruction(vcpu); return 1; case 8: - vcpu_load_rsp_rip(vcpu); - vcpu->arch.regs[reg] = kvm_get_cr8(vcpu); - vcpu_put_rsp_rip(vcpu); + kvm_register_write(vcpu, reg, kvm_get_cr8(vcpu)); KVMTRACE_2D(CR_READ, vcpu, (u32)cr, - (u32)vcpu->arch.regs[reg], handler); + (u32)kvm_register_read(vcpu, reg), handler); skip_emulated_instruction(vcpu); return 1; } @@ -2472,7 +2462,6 @@ static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) exit_qualification = vmcs_readl(EXIT_QUALIFICATION); dr = exit_qualification & 7; reg = (exit_qualification >> 8) & 15; - vcpu_load_rsp_rip(vcpu); if (exit_qualification & 16) { /* mov from dr */ switch (dr) { @@ -2485,12 +2474,11 @@ static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) default: val = 0; } - vcpu->arch.regs[reg] = val; + kvm_register_write(vcpu, reg, val); KVMTRACE_2D(DR_READ, vcpu, (u32)dr, (u32)val, handler); } else { /* mov to dr */ } - vcpu_put_rsp_rip(vcpu); skip_emulated_instruction(vcpu); return 1; } @@ -2735,8 +2723,8 @@ static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) struct vcpu_vmx *vmx = to_vmx(vcpu); u32 vectoring_info = vmx->idt_vectoring_info; - KVMTRACE_3D(VMEXIT, vcpu, exit_reason, (u32)vmcs_readl(GUEST_RIP), - (u32)((u64)vmcs_readl(GUEST_RIP) >> 32), entryexit); + KVMTRACE_3D(VMEXIT, vcpu, exit_reason, (u32)kvm_rip_read(vcpu), + (u32)((u64)kvm_rip_read(vcpu) >> 32), entryexit); /* Access CR3 don't cause VMExit in paging mode, so we need * to sync with guest real CR3. */ @@ -2922,9 +2910,9 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) static void fixup_rmode_irq(struct vcpu_vmx *vmx) { vmx->rmode.irq.pending = 0; - if (vmcs_readl(GUEST_RIP) + 1 != vmx->rmode.irq.rip) + if (kvm_rip_read(&vmx->vcpu) + 1 != vmx->rmode.irq.rip) return; - vmcs_writel(GUEST_RIP, vmx->rmode.irq.rip); + kvm_rip_write(&vmx->vcpu, vmx->rmode.irq.rip); if (vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK) { vmx->idt_vectoring_info &= ~VECTORING_INFO_TYPE_MASK; vmx->idt_vectoring_info |= INTR_TYPE_EXT_INTR; @@ -2941,6 +2929,11 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) struct vcpu_vmx *vmx = to_vmx(vcpu); u32 intr_info; + if (test_bit(VCPU_REGS_RSP, (unsigned long *)&vcpu->arch.regs_dirty)) + vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]); + if (test_bit(VCPU_REGS_RIP, (unsigned long *)&vcpu->arch.regs_dirty)) + vmcs_writel(GUEST_RIP, vcpu->arch.regs[VCPU_REGS_RIP]); + /* * Loading guest fpu may have cleared host cr0.ts */ @@ -3061,6 +3054,9 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) #endif ); + vcpu->arch.regs_avail = ~((1 << VCPU_REGS_RIP) | (1 << VCPU_REGS_RSP)); + vcpu->arch.regs_dirty = 0; + vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD); if (vmx->rmode.irq.pending) fixup_rmode_irq(vmx); @@ -3224,8 +3220,7 @@ static struct kvm_x86_ops vmx_x86_ops = { .set_idt = vmx_set_idt, .get_gdt = vmx_get_gdt, .set_gdt = vmx_set_gdt, - .cache_regs = vcpu_load_rsp_rip, - .decache_regs = vcpu_put_rsp_rip, + .cache_reg = vmx_cache_reg, .get_rflags = vmx_get_rflags, .set_rflags = vmx_set_rflags, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 0d682fc6aeb..2f0696bc7d2 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -19,6 +19,7 @@ #include "mmu.h" #include "i8254.h" #include "tss.h" +#include "kvm_cache_regs.h" #include #include @@ -61,6 +62,7 @@ static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, struct kvm_cpuid_entry2 __user *entries); struct kvm_x86_ops *kvm_x86_ops; +EXPORT_SYMBOL_GPL(kvm_x86_ops); struct kvm_stats_debugfs_item debugfs_entries[] = { { "pf_fixed", VCPU_STAT(pf_fixed) }, @@ -2080,7 +2082,7 @@ int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value) void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context) { u8 opcodes[4]; - unsigned long rip = vcpu->arch.rip; + unsigned long rip = kvm_rip_read(vcpu); unsigned long rip_linear; if (!printk_ratelimit()) @@ -2102,6 +2104,14 @@ static struct x86_emulate_ops emulate_ops = { .cmpxchg_emulated = emulator_cmpxchg_emulated, }; +static void cache_all_regs(struct kvm_vcpu *vcpu) +{ + kvm_register_read(vcpu, VCPU_REGS_RAX); + kvm_register_read(vcpu, VCPU_REGS_RSP); + kvm_register_read(vcpu, VCPU_REGS_RIP); + vcpu->arch.regs_dirty = ~0; +} + int emulate_instruction(struct kvm_vcpu *vcpu, struct kvm_run *run, unsigned long cr2, @@ -2112,7 +2122,13 @@ int emulate_instruction(struct kvm_vcpu *vcpu, struct decode_cache *c; vcpu->arch.mmio_fault_cr2 = cr2; - kvm_x86_ops->cache_regs(vcpu); + /* + * TODO: fix x86_emulate.c to use guest_read/write_register + * instead of direct ->regs accesses, can save hundred cycles + * on Intel for instructions that don't read/change RSP, for + * for example. + */ + cache_all_regs(vcpu); vcpu->mmio_is_write = 0; vcpu->arch.pio.string = 0; @@ -2172,7 +2188,6 @@ int emulate_instruction(struct kvm_vcpu *vcpu, return EMULATE_DO_MMIO; } - kvm_x86_ops->decache_regs(vcpu); kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags); if (vcpu->mmio_is_write) { @@ -2225,20 +2240,19 @@ int complete_pio(struct kvm_vcpu *vcpu) struct kvm_pio_request *io = &vcpu->arch.pio; long delta; int r; - - kvm_x86_ops->cache_regs(vcpu); + unsigned long val; if (!io->string) { - if (io->in) - memcpy(&vcpu->arch.regs[VCPU_REGS_RAX], vcpu->arch.pio_data, - io->size); + if (io->in) { + val = kvm_register_read(vcpu, VCPU_REGS_RAX); + memcpy(&val, vcpu->arch.pio_data, io->size); + kvm_register_write(vcpu, VCPU_REGS_RAX, val); + } } else { if (io->in) { r = pio_copy_data(vcpu); - if (r) { - kvm_x86_ops->cache_regs(vcpu); + if (r) return r; - } } delta = 1; @@ -2248,19 +2262,24 @@ int complete_pio(struct kvm_vcpu *vcpu) * The size of the register should really depend on * current address size. */ - vcpu->arch.regs[VCPU_REGS_RCX] -= delta; + val = kvm_register_read(vcpu, VCPU_REGS_RCX); + val -= delta; + kvm_register_write(vcpu, VCPU_REGS_RCX, val); } if (io->down) delta = -delta; delta *= io->size; - if (io->in) - vcpu->arch.regs[VCPU_REGS_RDI] += delta; - else - vcpu->arch.regs[VCPU_REGS_RSI] += delta; + if (io->in) { + val = kvm_register_read(vcpu, VCPU_REGS_RDI); + val += delta; + kvm_register_write(vcpu, VCPU_REGS_RDI, val); + } else { + val = kvm_register_read(vcpu, VCPU_REGS_RSI); + val += delta; + kvm_register_write(vcpu, VCPU_REGS_RSI, val); + } } - kvm_x86_ops->decache_regs(vcpu); - io->count -= io->cur_count; io->cur_count = 0; @@ -2313,6 +2332,7 @@ int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, int size, unsigned port) { struct kvm_io_device *pio_dev; + unsigned long val; vcpu->run->exit_reason = KVM_EXIT_IO; vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT; @@ -2333,8 +2353,8 @@ int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, KVMTRACE_2D(IO_WRITE, vcpu, vcpu->run->io.port, (u32)size, handler); - kvm_x86_ops->cache_regs(vcpu); - memcpy(vcpu->arch.pio_data, &vcpu->arch.regs[VCPU_REGS_RAX], 4); + val = kvm_register_read(vcpu, VCPU_REGS_RAX); + memcpy(vcpu->arch.pio_data, &val, 4); kvm_x86_ops->skip_emulated_instruction(vcpu); @@ -2519,13 +2539,11 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu) unsigned long nr, a0, a1, a2, a3, ret; int r = 1; - kvm_x86_ops->cache_regs(vcpu); - - nr = vcpu->arch.regs[VCPU_REGS_RAX]; - a0 = vcpu->arch.regs[VCPU_REGS_RBX]; - a1 = vcpu->arch.regs[VCPU_REGS_RCX]; - a2 = vcpu->arch.regs[VCPU_REGS_RDX]; - a3 = vcpu->arch.regs[VCPU_REGS_RSI]; + nr = kvm_register_read(vcpu, VCPU_REGS_RAX); + a0 = kvm_register_read(vcpu, VCPU_REGS_RBX); + a1 = kvm_register_read(vcpu, VCPU_REGS_RCX); + a2 = kvm_register_read(vcpu, VCPU_REGS_RDX); + a3 = kvm_register_read(vcpu, VCPU_REGS_RSI); KVMTRACE_1D(VMMCALL, vcpu, (u32)nr, handler); @@ -2548,8 +2566,7 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu) ret = -KVM_ENOSYS; break; } - vcpu->arch.regs[VCPU_REGS_RAX] = ret; - kvm_x86_ops->decache_regs(vcpu); + kvm_register_write(vcpu, VCPU_REGS_RAX, ret); ++vcpu->stat.hypercalls; return r; } @@ -2559,6 +2576,7 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu) { char instruction[3]; int ret = 0; + unsigned long rip = kvm_rip_read(vcpu); /* @@ -2568,9 +2586,8 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu) */ kvm_mmu_zap_all(vcpu->kvm); - kvm_x86_ops->cache_regs(vcpu); kvm_x86_ops->patch_hypercall(vcpu, instruction); - if (emulator_write_emulated(vcpu->arch.rip, instruction, 3, vcpu) + if (emulator_write_emulated(rip, instruction, 3, vcpu) != X86EMUL_CONTINUE) ret = -EFAULT; @@ -2700,13 +2717,12 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu) u32 function, index; struct kvm_cpuid_entry2 *e, *best; - kvm_x86_ops->cache_regs(vcpu); - function = vcpu->arch.regs[VCPU_REGS_RAX]; - index = vcpu->arch.regs[VCPU_REGS_RCX]; - vcpu->arch.regs[VCPU_REGS_RAX] = 0; - vcpu->arch.regs[VCPU_REGS_RBX] = 0; - vcpu->arch.regs[VCPU_REGS_RCX] = 0; - vcpu->arch.regs[VCPU_REGS_RDX] = 0; + function = kvm_register_read(vcpu, VCPU_REGS_RAX); + index = kvm_register_read(vcpu, VCPU_REGS_RCX); + kvm_register_write(vcpu, VCPU_REGS_RAX, 0); + kvm_register_write(vcpu, VCPU_REGS_RBX, 0); + kvm_register_write(vcpu, VCPU_REGS_RCX, 0); + kvm_register_write(vcpu, VCPU_REGS_RDX, 0); best = NULL; for (i = 0; i < vcpu->arch.cpuid_nent; ++i) { e = &vcpu->arch.cpuid_entries[i]; @@ -2724,18 +2740,17 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu) best = e; } if (best) { - vcpu->arch.regs[VCPU_REGS_RAX] = best->eax; - vcpu->arch.regs[VCPU_REGS_RBX] = best->ebx; - vcpu->arch.regs[VCPU_REGS_RCX] = best->ecx; - vcpu->arch.regs[VCPU_REGS_RDX] = best->edx; + kvm_register_write(vcpu, VCPU_REGS_RAX, best->eax); + kvm_register_write(vcpu, VCPU_REGS_RBX, best->ebx); + kvm_register_write(vcpu, VCPU_REGS_RCX, best->ecx); + kvm_register_write(vcpu, VCPU_REGS_RDX, best->edx); } - kvm_x86_ops->decache_regs(vcpu); kvm_x86_ops->skip_emulated_instruction(vcpu); KVMTRACE_5D(CPUID, vcpu, function, - (u32)vcpu->arch.regs[VCPU_REGS_RAX], - (u32)vcpu->arch.regs[VCPU_REGS_RBX], - (u32)vcpu->arch.regs[VCPU_REGS_RCX], - (u32)vcpu->arch.regs[VCPU_REGS_RDX], handler); + (u32)kvm_register_read(vcpu, VCPU_REGS_RAX), + (u32)kvm_register_read(vcpu, VCPU_REGS_RBX), + (u32)kvm_register_read(vcpu, VCPU_REGS_RCX), + (u32)kvm_register_read(vcpu, VCPU_REGS_RDX), handler); } EXPORT_SYMBOL_GPL(kvm_emulate_cpuid); @@ -2917,8 +2932,8 @@ again: * Profile KVM exit RIPs: */ if (unlikely(prof_on == KVM_PROFILING)) { - kvm_x86_ops->cache_regs(vcpu); - profile_hit(KVM_PROFILING, (void *)vcpu->arch.rip); + unsigned long rip = kvm_rip_read(vcpu); + profile_hit(KVM_PROFILING, (void *)rip); } if (vcpu->arch.exception.pending && kvm_x86_ops->exception_injected(vcpu)) @@ -2999,11 +3014,9 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } } #endif - if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) { - kvm_x86_ops->cache_regs(vcpu); - vcpu->arch.regs[VCPU_REGS_RAX] = kvm_run->hypercall.ret; - kvm_x86_ops->decache_regs(vcpu); - } + if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) + kvm_register_write(vcpu, VCPU_REGS_RAX, + kvm_run->hypercall.ret); r = __vcpu_run(vcpu, kvm_run); @@ -3019,28 +3032,26 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) { vcpu_load(vcpu); - kvm_x86_ops->cache_regs(vcpu); - - regs->rax = vcpu->arch.regs[VCPU_REGS_RAX]; - regs->rbx = vcpu->arch.regs[VCPU_REGS_RBX]; - regs->rcx = vcpu->arch.regs[VCPU_REGS_RCX]; - regs->rdx = vcpu->arch.regs[VCPU_REGS_RDX]; - regs->rsi = vcpu->arch.regs[VCPU_REGS_RSI]; - regs->rdi = vcpu->arch.regs[VCPU_REGS_RDI]; - regs->rsp = vcpu->arch.regs[VCPU_REGS_RSP]; - regs->rbp = vcpu->arch.regs[VCPU_REGS_RBP]; + regs->rax = kvm_register_read(vcpu, VCPU_REGS_RAX); + regs->rbx = kvm_register_read(vcpu, VCPU_REGS_RBX); + regs->rcx = kvm_register_read(vcpu, VCPU_REGS_RCX); + regs->rdx = kvm_register_read(vcpu, VCPU_REGS_RDX); + regs->rsi = kvm_register_read(vcpu, VCPU_REGS_RSI); + regs->rdi = kvm_register_read(vcpu, VCPU_REGS_RDI); + regs->rsp = kvm_register_read(vcpu, VCPU_REGS_RSP); + regs->rbp = kvm_register_read(vcpu, VCPU_REGS_RBP); #ifdef CONFIG_X86_64 - regs->r8 = vcpu->arch.regs[VCPU_REGS_R8]; - regs->r9 = vcpu->arch.regs[VCPU_REGS_R9]; - regs->r10 = vcpu->arch.regs[VCPU_REGS_R10]; - regs->r11 = vcpu->arch.regs[VCPU_REGS_R11]; - regs->r12 = vcpu->arch.regs[VCPU_REGS_R12]; - regs->r13 = vcpu->arch.regs[VCPU_REGS_R13]; - regs->r14 = vcpu->arch.regs[VCPU_REGS_R14]; - regs->r15 = vcpu->arch.regs[VCPU_REGS_R15]; + regs->r8 = kvm_register_read(vcpu, VCPU_REGS_R8); + regs->r9 = kvm_register_read(vcpu, VCPU_REGS_R9); + regs->r10 = kvm_register_read(vcpu, VCPU_REGS_R10); + regs->r11 = kvm_register_read(vcpu, VCPU_REGS_R11); + regs->r12 = kvm_register_read(vcpu, VCPU_REGS_R12); + regs->r13 = kvm_register_read(vcpu, VCPU_REGS_R13); + regs->r14 = kvm_register_read(vcpu, VCPU_REGS_R14); + regs->r15 = kvm_register_read(vcpu, VCPU_REGS_R15); #endif - regs->rip = vcpu->arch.rip; + regs->rip = kvm_rip_read(vcpu); regs->rflags = kvm_x86_ops->get_rflags(vcpu); /* @@ -3058,29 +3069,29 @@ int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) { vcpu_load(vcpu); - vcpu->arch.regs[VCPU_REGS_RAX] = regs->rax; - vcpu->arch.regs[VCPU_REGS_RBX] = regs->rbx; - vcpu->arch.regs[VCPU_REGS_RCX] = regs->rcx; - vcpu->arch.regs[VCPU_REGS_RDX] = regs->rdx; - vcpu->arch.regs[VCPU_REGS_RSI] = regs->rsi; - vcpu->arch.regs[VCPU_REGS_RDI] = regs->rdi; - vcpu->arch.regs[VCPU_REGS_RSP] = regs->rsp; - vcpu->arch.regs[VCPU_REGS_RBP] = regs->rbp; + kvm_register_write(vcpu, VCPU_REGS_RAX, regs->rax); + kvm_register_write(vcpu, VCPU_REGS_RBX, regs->rbx); + kvm_register_write(vcpu, VCPU_REGS_RCX, regs->rcx); + kvm_register_write(vcpu, VCPU_REGS_RDX, regs->rdx); + kvm_register_write(vcpu, VCPU_REGS_RSI, regs->rsi); + kvm_register_write(vcpu, VCPU_REGS_RDI, regs->rdi); + kvm_register_write(vcpu, VCPU_REGS_RSP, regs->rsp); + kvm_register_write(vcpu, VCPU_REGS_RBP, regs->rbp); #ifdef CONFIG_X86_64 - vcpu->arch.regs[VCPU_REGS_R8] = regs->r8; - vcpu->arch.regs[VCPU_REGS_R9] = regs->r9; - vcpu->arch.regs[VCPU_REGS_R10] = regs->r10; - vcpu->arch.regs[VCPU_REGS_R11] = regs->r11; - vcpu->arch.regs[VCPU_REGS_R12] = regs->r12; - vcpu->arch.regs[VCPU_REGS_R13] = regs->r13; - vcpu->arch.regs[VCPU_REGS_R14] = regs->r14; - vcpu->arch.regs[VCPU_REGS_R15] = regs->r15; + kvm_register_write(vcpu, VCPU_REGS_R8, regs->r8); + kvm_register_write(vcpu, VCPU_REGS_R9, regs->r9); + kvm_register_write(vcpu, VCPU_REGS_R10, regs->r10); + kvm_register_write(vcpu, VCPU_REGS_R11, regs->r11); + kvm_register_write(vcpu, VCPU_REGS_R12, regs->r12); + kvm_register_write(vcpu, VCPU_REGS_R13, regs->r13); + kvm_register_write(vcpu, VCPU_REGS_R14, regs->r14); + kvm_register_write(vcpu, VCPU_REGS_R15, regs->r15); + #endif - vcpu->arch.rip = regs->rip; + kvm_rip_write(vcpu, regs->rip); kvm_x86_ops->set_rflags(vcpu, regs->rflags); - kvm_x86_ops->decache_regs(vcpu); vcpu->arch.exception.pending = false; @@ -3316,17 +3327,16 @@ static void save_state_to_tss32(struct kvm_vcpu *vcpu, struct tss_segment_32 *tss) { tss->cr3 = vcpu->arch.cr3; - tss->eip = vcpu->arch.rip; + tss->eip = kvm_rip_read(vcpu); tss->eflags = kvm_x86_ops->get_rflags(vcpu); - tss->eax = vcpu->arch.regs[VCPU_REGS_RAX]; - tss->ecx = vcpu->arch.regs[VCPU_REGS_RCX]; - tss->edx = vcpu->arch.regs[VCPU_REGS_RDX]; - tss->ebx = vcpu->arch.regs[VCPU_REGS_RBX]; - tss->esp = vcpu->arch.regs[VCPU_REGS_RSP]; - tss->ebp = vcpu->arch.regs[VCPU_REGS_RBP]; - tss->esi = vcpu->arch.regs[VCPU_REGS_RSI]; - tss->edi = vcpu->arch.regs[VCPU_REGS_RDI]; - + tss->eax = kvm_register_read(vcpu, VCPU_REGS_RAX); + tss->ecx = kvm_register_read(vcpu, VCPU_REGS_RCX); + tss->edx = kvm_register_read(vcpu, VCPU_REGS_RDX); + tss->ebx = kvm_register_read(vcpu, VCPU_REGS_RBX); + tss->esp = kvm_register_read(vcpu, VCPU_REGS_RSP); + tss->ebp = kvm_register_read(vcpu, VCPU_REGS_RBP); + tss->esi = kvm_register_read(vcpu, VCPU_REGS_RSI); + tss->edi = kvm_register_read(vcpu, VCPU_REGS_RDI); tss->es = get_segment_selector(vcpu, VCPU_SREG_ES); tss->cs = get_segment_selector(vcpu, VCPU_SREG_CS); tss->ss = get_segment_selector(vcpu, VCPU_SREG_SS); @@ -3342,17 +3352,17 @@ static int load_state_from_tss32(struct kvm_vcpu *vcpu, { kvm_set_cr3(vcpu, tss->cr3); - vcpu->arch.rip = tss->eip; + kvm_rip_write(vcpu, tss->eip); kvm_x86_ops->set_rflags(vcpu, tss->eflags | 2); - vcpu->arch.regs[VCPU_REGS_RAX] = tss->eax; - vcpu->arch.regs[VCPU_REGS_RCX] = tss->ecx; - vcpu->arch.regs[VCPU_REGS_RDX] = tss->edx; - vcpu->arch.regs[VCPU_REGS_RBX] = tss->ebx; - vcpu->arch.regs[VCPU_REGS_RSP] = tss->esp; - vcpu->arch.regs[VCPU_REGS_RBP] = tss->ebp; - vcpu->arch.regs[VCPU_REGS_RSI] = tss->esi; - vcpu->arch.regs[VCPU_REGS_RDI] = tss->edi; + kvm_register_write(vcpu, VCPU_REGS_RAX, tss->eax); + kvm_register_write(vcpu, VCPU_REGS_RCX, tss->ecx); + kvm_register_write(vcpu, VCPU_REGS_RDX, tss->edx); + kvm_register_write(vcpu, VCPU_REGS_RBX, tss->ebx); + kvm_register_write(vcpu, VCPU_REGS_RSP, tss->esp); + kvm_register_write(vcpu, VCPU_REGS_RBP, tss->ebp); + kvm_register_write(vcpu, VCPU_REGS_RSI, tss->esi); + kvm_register_write(vcpu, VCPU_REGS_RDI, tss->edi); if (kvm_load_segment_descriptor(vcpu, tss->ldt_selector, 0, VCPU_SREG_LDTR)) return 1; @@ -3380,16 +3390,16 @@ static int load_state_from_tss32(struct kvm_vcpu *vcpu, static void save_state_to_tss16(struct kvm_vcpu *vcpu, struct tss_segment_16 *tss) { - tss->ip = vcpu->arch.rip; + tss->ip = kvm_rip_read(vcpu); tss->flag = kvm_x86_ops->get_rflags(vcpu); - tss->ax = vcpu->arch.regs[VCPU_REGS_RAX]; - tss->cx = vcpu->arch.regs[VCPU_REGS_RCX]; - tss->dx = vcpu->arch.regs[VCPU_REGS_RDX]; - tss->bx = vcpu->arch.regs[VCPU_REGS_RBX]; - tss->sp = vcpu->arch.regs[VCPU_REGS_RSP]; - tss->bp = vcpu->arch.regs[VCPU_REGS_RBP]; - tss->si = vcpu->arch.regs[VCPU_REGS_RSI]; - tss->di = vcpu->arch.regs[VCPU_REGS_RDI]; + tss->ax = kvm_register_read(vcpu, VCPU_REGS_RAX); + tss->cx = kvm_register_read(vcpu, VCPU_REGS_RCX); + tss->dx = kvm_register_read(vcpu, VCPU_REGS_RDX); + tss->bx = kvm_register_read(vcpu, VCPU_REGS_RBX); + tss->sp = kvm_register_read(vcpu, VCPU_REGS_RSP); + tss->bp = kvm_register_read(vcpu, VCPU_REGS_RBP); + tss->si = kvm_register_read(vcpu, VCPU_REGS_RSI); + tss->di = kvm_register_read(vcpu, VCPU_REGS_RDI); tss->es = get_segment_selector(vcpu, VCPU_SREG_ES); tss->cs = get_segment_selector(vcpu, VCPU_SREG_CS); @@ -3402,16 +3412,16 @@ static void save_state_to_tss16(struct kvm_vcpu *vcpu, static int load_state_from_tss16(struct kvm_vcpu *vcpu, struct tss_segment_16 *tss) { - vcpu->arch.rip = tss->ip; + kvm_rip_write(vcpu, tss->ip); kvm_x86_ops->set_rflags(vcpu, tss->flag | 2); - vcpu->arch.regs[VCPU_REGS_RAX] = tss->ax; - vcpu->arch.regs[VCPU_REGS_RCX] = tss->cx; - vcpu->arch.regs[VCPU_REGS_RDX] = tss->dx; - vcpu->arch.regs[VCPU_REGS_RBX] = tss->bx; - vcpu->arch.regs[VCPU_REGS_RSP] = tss->sp; - vcpu->arch.regs[VCPU_REGS_RBP] = tss->bp; - vcpu->arch.regs[VCPU_REGS_RSI] = tss->si; - vcpu->arch.regs[VCPU_REGS_RDI] = tss->di; + kvm_register_write(vcpu, VCPU_REGS_RAX, tss->ax); + kvm_register_write(vcpu, VCPU_REGS_RCX, tss->cx); + kvm_register_write(vcpu, VCPU_REGS_RDX, tss->dx); + kvm_register_write(vcpu, VCPU_REGS_RBX, tss->bx); + kvm_register_write(vcpu, VCPU_REGS_RSP, tss->sp); + kvm_register_write(vcpu, VCPU_REGS_RBP, tss->bp); + kvm_register_write(vcpu, VCPU_REGS_RSI, tss->si); + kvm_register_write(vcpu, VCPU_REGS_RDI, tss->di); if (kvm_load_segment_descriptor(vcpu, tss->ldt, 0, VCPU_SREG_LDTR)) return 1; @@ -3534,7 +3544,6 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason) } kvm_x86_ops->skip_emulated_instruction(vcpu); - kvm_x86_ops->cache_regs(vcpu); if (nseg_desc.type & 8) ret = kvm_task_switch_32(vcpu, tss_selector, old_tss_base, @@ -3559,7 +3568,6 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason) tr_seg.type = 11; kvm_set_segment(vcpu, &tr_seg, VCPU_SREG_TR); out: - kvm_x86_ops->decache_regs(vcpu); return ret; } EXPORT_SYMBOL_GPL(kvm_task_switch); diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c index f2f90468f8b..d5da7f14d53 100644 --- a/arch/x86/kvm/x86_emulate.c +++ b/arch/x86/kvm/x86_emulate.c @@ -26,6 +26,7 @@ #define DPRINTF(_f, _a ...) printf(_f , ## _a) #else #include +#include "kvm_cache_regs.h" #define DPRINTF(x...) do {} while (0) #endif #include @@ -839,7 +840,7 @@ x86_decode_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) /* Shadow copy of register state. Committed on successful emulation. */ memset(c, 0, sizeof(struct decode_cache)); - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); ctxt->cs_base = seg_base(ctxt, VCPU_SREG_CS); memcpy(c->regs, ctxt->vcpu->arch.regs, sizeof c->regs); @@ -1267,7 +1268,7 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) if (c->rep_prefix && (c->d & String)) { /* All REP prefixes have the same first termination condition */ if (c->regs[VCPU_REGS_RCX] == 0) { - ctxt->vcpu->arch.rip = c->eip; + kvm_rip_write(ctxt->vcpu, c->eip); goto done; } /* The second termination condition only applies for REPE @@ -1281,17 +1282,17 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) (c->b == 0xae) || (c->b == 0xaf)) { if ((c->rep_prefix == REPE_PREFIX) && ((ctxt->eflags & EFLG_ZF) == 0)) { - ctxt->vcpu->arch.rip = c->eip; + kvm_rip_write(ctxt->vcpu, c->eip); goto done; } if ((c->rep_prefix == REPNE_PREFIX) && ((ctxt->eflags & EFLG_ZF) == EFLG_ZF)) { - ctxt->vcpu->arch.rip = c->eip; + kvm_rip_write(ctxt->vcpu, c->eip); goto done; } } c->regs[VCPU_REGS_RCX]--; - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); } if (c->src.type == OP_MEM) { @@ -1768,7 +1769,7 @@ writeback: /* Commit shadow register state. */ memcpy(ctxt->vcpu->arch.regs, c->regs, sizeof c->regs); - ctxt->vcpu->arch.rip = c->eip; + kvm_rip_write(ctxt->vcpu, c->eip); done: if (rc == X86EMUL_UNHANDLEABLE) { @@ -1793,7 +1794,7 @@ twobyte_insn: goto done; /* Let the processor re-execute the fixed hypercall */ - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); /* Disable writeback. */ c->dst.type = OP_NONE; break; @@ -1889,7 +1890,7 @@ twobyte_insn: rc = kvm_set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data); if (rc) { kvm_inject_gp(ctxt->vcpu, 0); - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); } rc = X86EMUL_CONTINUE; c->dst.type = OP_NONE; @@ -1899,7 +1900,7 @@ twobyte_insn: rc = kvm_get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data); if (rc) { kvm_inject_gp(ctxt->vcpu, 0); - c->eip = ctxt->vcpu->arch.rip; + c->eip = kvm_rip_read(ctxt->vcpu); } else { c->regs[VCPU_REGS_RAX] = (u32)msr_data; c->regs[VCPU_REGS_RDX] = msr_data >> 32; diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 69794547f51..8c886be0103 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -89,7 +89,7 @@ extern struct list_head vm_list; struct kvm_vcpu; struct kvm; -enum { +enum kvm_reg { VCPU_REGS_RAX = 0, VCPU_REGS_RCX = 1, VCPU_REGS_RDX = 2, @@ -108,6 +108,7 @@ enum { VCPU_REGS_R14 = 14, VCPU_REGS_R15 = 15, #endif + VCPU_REGS_RIP, NR_VCPU_REGS }; @@ -219,8 +220,13 @@ struct kvm_vcpu_arch { int interrupt_window_open; unsigned long irq_summary; /* bit vector: 1 per word in irq_pending */ DECLARE_BITMAP(irq_pending, KVM_NR_INTERRUPTS); - unsigned long regs[NR_VCPU_REGS]; /* for rsp: vcpu_load_rsp_rip() */ - unsigned long rip; /* needs vcpu_load_rsp_rip() */ + /* + * rip and regs accesses must go through + * kvm_{register,rip}_{read,write} functions. + */ + unsigned long regs[NR_VCPU_REGS]; + u32 regs_avail; + u32 regs_dirty; unsigned long cr0; unsigned long cr2; @@ -414,8 +420,7 @@ struct kvm_x86_ops { unsigned long (*get_dr)(struct kvm_vcpu *vcpu, int dr); void (*set_dr)(struct kvm_vcpu *vcpu, int dr, unsigned long value, int *exception); - void (*cache_regs)(struct kvm_vcpu *vcpu); - void (*decache_regs)(struct kvm_vcpu *vcpu); + void (*cache_reg)(struct kvm_vcpu *vcpu, enum kvm_reg reg); unsigned long (*get_rflags)(struct kvm_vcpu *vcpu); void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags); -- cgit v1.2.3-70-g09d2 From d98e6346350ac909f095768beb28b82368bd126f Mon Sep 17 00:00:00 2001 From: Hollis Blanchard Date: Tue, 1 Jul 2008 16:23:49 -0500 Subject: KVM: Move KVM TRACE DEFINITIONS to common header Move KVM trace definitions from x86 specific kvm headers to common kvm headers to create a cross-architecture numbering scheme for trace events. This means the kvmtrace_format userspace tool won't need to know which architecture produced the log file being processed. Signed-off-by: Jerone Young Signed-off-by: Hollis Blanchard Signed-off-by: Avi Kivity --- include/asm-x86/kvm.h | 22 ---------------------- include/asm-x86/kvm_host.h | 18 ------------------ include/linux/kvm.h | 21 +++++++++++++++++++++ include/linux/kvm_host.h | 19 +++++++++++++++++++ 4 files changed, 40 insertions(+), 40 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h index 78e954db1e7..ba0dd791fad 100644 --- a/include/asm-x86/kvm.h +++ b/include/asm-x86/kvm.h @@ -208,26 +208,4 @@ struct kvm_pit_channel_state { struct kvm_pit_state { struct kvm_pit_channel_state channels[3]; }; - -#define KVM_TRC_INJ_VIRQ (KVM_TRC_HANDLER + 0x02) -#define KVM_TRC_REDELIVER_EVT (KVM_TRC_HANDLER + 0x03) -#define KVM_TRC_PEND_INTR (KVM_TRC_HANDLER + 0x04) -#define KVM_TRC_IO_READ (KVM_TRC_HANDLER + 0x05) -#define KVM_TRC_IO_WRITE (KVM_TRC_HANDLER + 0x06) -#define KVM_TRC_CR_READ (KVM_TRC_HANDLER + 0x07) -#define KVM_TRC_CR_WRITE (KVM_TRC_HANDLER + 0x08) -#define KVM_TRC_DR_READ (KVM_TRC_HANDLER + 0x09) -#define KVM_TRC_DR_WRITE (KVM_TRC_HANDLER + 0x0A) -#define KVM_TRC_MSR_READ (KVM_TRC_HANDLER + 0x0B) -#define KVM_TRC_MSR_WRITE (KVM_TRC_HANDLER + 0x0C) -#define KVM_TRC_CPUID (KVM_TRC_HANDLER + 0x0D) -#define KVM_TRC_INTR (KVM_TRC_HANDLER + 0x0E) -#define KVM_TRC_NMI (KVM_TRC_HANDLER + 0x0F) -#define KVM_TRC_VMMCALL (KVM_TRC_HANDLER + 0x10) -#define KVM_TRC_HLT (KVM_TRC_HANDLER + 0x11) -#define KVM_TRC_CLTS (KVM_TRC_HANDLER + 0x12) -#define KVM_TRC_LMSW (KVM_TRC_HANDLER + 0x13) -#define KVM_TRC_APIC_ACCESS (KVM_TRC_HANDLER + 0x14) -#define KVM_TRC_TDP_FAULT (KVM_TRC_HANDLER + 0x15) - #endif /* ASM_X86__KVM_H */ diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 8c886be0103..4f2bd884fd3 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -691,24 +691,6 @@ enum { TASK_SWITCH_GATE = 3, }; -#define KVMTRACE_5D(evt, vcpu, d1, d2, d3, d4, d5, name) \ - trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ - vcpu, 5, d1, d2, d3, d4, d5) -#define KVMTRACE_4D(evt, vcpu, d1, d2, d3, d4, name) \ - trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ - vcpu, 4, d1, d2, d3, d4, 0) -#define KVMTRACE_3D(evt, vcpu, d1, d2, d3, name) \ - trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ - vcpu, 3, d1, d2, d3, 0, 0) -#define KVMTRACE_2D(evt, vcpu, d1, d2, name) \ - trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ - vcpu, 2, d1, d2, 0, 0, 0) -#define KVMTRACE_1D(evt, vcpu, d1, name) \ - trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ - vcpu, 1, d1, 0, 0, 0, 0) -#define KVMTRACE_0D(evt, vcpu, name) \ - trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ - vcpu, 0, 0, 0, 0, 0, 0) #ifdef CONFIG_64BIT # define KVM_EX_ENTRY ".quad" diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 70a30651cd1..8a3ceadb136 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -440,4 +440,25 @@ struct kvm_trace_rec { #define KVM_GET_MP_STATE _IOR(KVMIO, 0x98, struct kvm_mp_state) #define KVM_SET_MP_STATE _IOW(KVMIO, 0x99, struct kvm_mp_state) +#define KVM_TRC_INJ_VIRQ (KVM_TRC_HANDLER + 0x02) +#define KVM_TRC_REDELIVER_EVT (KVM_TRC_HANDLER + 0x03) +#define KVM_TRC_PEND_INTR (KVM_TRC_HANDLER + 0x04) +#define KVM_TRC_IO_READ (KVM_TRC_HANDLER + 0x05) +#define KVM_TRC_IO_WRITE (KVM_TRC_HANDLER + 0x06) +#define KVM_TRC_CR_READ (KVM_TRC_HANDLER + 0x07) +#define KVM_TRC_CR_WRITE (KVM_TRC_HANDLER + 0x08) +#define KVM_TRC_DR_READ (KVM_TRC_HANDLER + 0x09) +#define KVM_TRC_DR_WRITE (KVM_TRC_HANDLER + 0x0A) +#define KVM_TRC_MSR_READ (KVM_TRC_HANDLER + 0x0B) +#define KVM_TRC_MSR_WRITE (KVM_TRC_HANDLER + 0x0C) +#define KVM_TRC_CPUID (KVM_TRC_HANDLER + 0x0D) +#define KVM_TRC_INTR (KVM_TRC_HANDLER + 0x0E) +#define KVM_TRC_NMI (KVM_TRC_HANDLER + 0x0F) +#define KVM_TRC_VMMCALL (KVM_TRC_HANDLER + 0x10) +#define KVM_TRC_HLT (KVM_TRC_HANDLER + 0x11) +#define KVM_TRC_CLTS (KVM_TRC_HANDLER + 0x12) +#define KVM_TRC_LMSW (KVM_TRC_HANDLER + 0x13) +#define KVM_TRC_APIC_ACCESS (KVM_TRC_HANDLER + 0x14) +#define KVM_TRC_TDP_FAULT (KVM_TRC_HANDLER + 0x15) + #endif diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 8525afc5310..a18aaad2ab7 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -326,6 +326,25 @@ struct kvm_stats_debugfs_item { extern struct kvm_stats_debugfs_item debugfs_entries[]; extern struct dentry *kvm_debugfs_dir; +#define KVMTRACE_5D(evt, vcpu, d1, d2, d3, d4, d5, name) \ + trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ + vcpu, 5, d1, d2, d3, d4, d5) +#define KVMTRACE_4D(evt, vcpu, d1, d2, d3, d4, name) \ + trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ + vcpu, 4, d1, d2, d3, d4, 0) +#define KVMTRACE_3D(evt, vcpu, d1, d2, d3, name) \ + trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ + vcpu, 3, d1, d2, d3, 0, 0) +#define KVMTRACE_2D(evt, vcpu, d1, d2, name) \ + trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ + vcpu, 2, d1, d2, 0, 0, 0) +#define KVMTRACE_1D(evt, vcpu, d1, name) \ + trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ + vcpu, 1, d1, 0, 0, 0, 0) +#define KVMTRACE_0D(evt, vcpu, name) \ + trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ + vcpu, 0, 0, 0, 0, 0, 0) + #ifdef CONFIG_KVM_TRACE int kvm_trace_ioctl(unsigned int ioctl, unsigned long arg); void kvm_trace_cleanup(void); -- cgit v1.2.3-70-g09d2 From 668f612fa0d8d4120ec5dc0725d7e1ca3152a954 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Wed, 2 Jul 2008 09:28:55 +0300 Subject: KVM: VMX: Move nmi injection failure processing to vm exit path Instead of processing nmi injection failure in the vm entry path, move it to the vm exit path (vm_complete_interrupts()). This separates nmi injection from nmi post-processing, and moves the nmi state from the VT state into vcpu state (new variable nmi_injected specifying an injection in progress). Signed-off-by: Avi Kivity --- arch/x86/kvm/vmx.c | 49 ++++++++++++++++++++++++++++++---------------- include/asm-x86/kvm_host.h | 1 + 2 files changed, 33 insertions(+), 17 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 2adfacb0033..ce13b53d21c 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2151,7 +2151,6 @@ static void vmx_inject_nmi(struct kvm_vcpu *vcpu) { vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, INTR_TYPE_NMI_INTR | INTR_INFO_VALID_MASK | NMI_VECTOR); - vcpu->arch.nmi_pending = 0; } static void kvm_do_inject_irq(struct kvm_vcpu *vcpu) @@ -2820,8 +2819,11 @@ static void enable_intr_window(struct kvm_vcpu *vcpu) static void vmx_complete_interrupts(struct vcpu_vmx *vmx) { u32 exit_intr_info; + u32 idt_vectoring_info; bool unblock_nmi; u8 vector; + int type; + bool idtv_info_valid; exit_intr_info = vmcs_read32(VM_EXIT_INTR_INFO); if (cpu_has_virtual_nmis()) { @@ -2836,18 +2838,34 @@ static void vmx_complete_interrupts(struct vcpu_vmx *vmx) vmcs_set_bits(GUEST_INTERRUPTIBILITY_INFO, GUEST_INTR_STATE_NMI); } + + idt_vectoring_info = vmx->idt_vectoring_info; + idtv_info_valid = idt_vectoring_info & VECTORING_INFO_VALID_MASK; + vector = idt_vectoring_info & VECTORING_INFO_VECTOR_MASK; + type = idt_vectoring_info & VECTORING_INFO_TYPE_MASK; + if (vmx->vcpu.arch.nmi_injected) { + /* + * SDM 3: 25.7.1.2 + * Clear bit "block by NMI" before VM entry if a NMI delivery + * faulted. + */ + if (idtv_info_valid && type == INTR_TYPE_NMI_INTR) + vmcs_clear_bits(GUEST_INTERRUPTIBILITY_INFO, + GUEST_INTR_STATE_NMI); + else + vmx->vcpu.arch.nmi_injected = false; + } } static void vmx_intr_assist(struct kvm_vcpu *vcpu) { struct vcpu_vmx *vmx = to_vmx(vcpu); - u32 idtv_info_field, intr_info_field, exit_intr_info_field; + u32 idtv_info_field, intr_info_field; int vector; update_tpr_threshold(vcpu); intr_info_field = vmcs_read32(VM_ENTRY_INTR_INFO_FIELD); - exit_intr_info_field = vmcs_read32(VM_EXIT_INTR_INFO); idtv_info_field = vmx->idt_vectoring_info; if (intr_info_field & INTR_INFO_VALID_MASK) { if (idtv_info_field & INTR_INFO_VALID_MASK) { @@ -2871,17 +2889,6 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) KVMTRACE_1D(REDELIVER_EVT, vcpu, idtv_info_field, handler); - /* - * SDM 3: 25.7.1.2 - * Clear bit "block by NMI" before VM entry if a NMI delivery - * faulted. - */ - if ((idtv_info_field & VECTORING_INFO_TYPE_MASK) - == INTR_TYPE_NMI_INTR && cpu_has_virtual_nmis()) - vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, - vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & - ~GUEST_INTR_STATE_NMI); - vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field & ~INTR_INFO_RESVD_BITS_MASK); vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, @@ -2894,9 +2901,17 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) return; } if (cpu_has_virtual_nmis()) { - if (vcpu->arch.nmi_pending) { - if (vmx_nmi_enabled(vcpu)) - vmx_inject_nmi(vcpu); + if (vcpu->arch.nmi_pending && !vcpu->arch.nmi_injected) { + if (vmx_nmi_enabled(vcpu)) { + vcpu->arch.nmi_pending = false; + vcpu->arch.nmi_injected = true; + } else { + enable_intr_window(vcpu); + return; + } + } + if (vcpu->arch.nmi_injected) { + vmx_inject_nmi(vcpu); enable_intr_window(vcpu); return; } diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 4f2bd884fd3..7cf69fd1dcf 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -300,6 +300,7 @@ struct kvm_vcpu_arch { struct page *time_page; bool nmi_pending; + bool nmi_injected; u64 mtrr[0x100]; }; -- cgit v1.2.3-70-g09d2 From 937a7eaef9f08342958d17055a350982b7bd92cb Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 3 Jul 2008 15:17:01 +0300 Subject: KVM: Add a pending interrupt queue Similar to the exception queue, this hold interrupts that have been accepted by the virtual processor core but not yet injected. Not yet used. Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.h | 11 +++++++++++ include/asm-x86/kvm_host.h | 5 +++++ 2 files changed, 16 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h index c666649c4bb..6a4be78a738 100644 --- a/arch/x86/kvm/x86.h +++ b/arch/x86/kvm/x86.h @@ -8,4 +8,15 @@ static inline void kvm_clear_exception_queue(struct kvm_vcpu *vcpu) vcpu->arch.exception.pending = false; } +static inline void kvm_queue_interrupt(struct kvm_vcpu *vcpu, u8 vector) +{ + vcpu->arch.interrupt.pending = true; + vcpu->arch.interrupt.nr = vector; +} + +static inline void kvm_clear_interrupt_queue(struct kvm_vcpu *vcpu) +{ + vcpu->arch.interrupt.pending = false; +} + #endif diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 7cf69fd1dcf..65c6a0e5b73 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -275,6 +275,11 @@ struct kvm_vcpu_arch { u32 error_code; } exception; + struct kvm_queued_interrupt { + bool pending; + u8 nr; + } interrupt; + struct { int active; u8 save_iopl; -- cgit v1.2.3-70-g09d2 From 19bd8afdc4e6fbb47e4841f8d771f8cb29916d9f Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Sun, 13 Jul 2008 13:40:55 +0200 Subject: KVM: Consolidate XX_VECTOR defines Signed-off-by: Jan Kiszka Signed-off-by: Avi Kivity --- arch/x86/kvm/svm.c | 4 ---- arch/x86/kvm/vmx.c | 2 +- include/asm-x86/kvm_host.h | 1 + 3 files changed, 2 insertions(+), 5 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 54b0bf33e21..743c98d135c 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -36,10 +36,6 @@ MODULE_LICENSE("GPL"); #define IOPM_ALLOC_ORDER 2 #define MSRPM_ALLOC_ORDER 1 -#define DB_VECTOR 1 -#define UD_VECTOR 6 -#define GP_VECTOR 13 - #define DR7_GD_MASK (1 << 13) #define DR6_BD_MASK (1 << 13) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 9591ca73191..5c9dac567a7 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -470,7 +470,7 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu) if (!vcpu->fpu_active) eb |= 1u << NM_VECTOR; if (vcpu->guest_debug.enabled) - eb |= 1u << 1; + eb |= 1u << DB_VECTOR; if (vcpu->arch.rmode.active) eb = ~0; if (vm_need_ept()) diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 65c6a0e5b73..da3c1979b73 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -57,6 +57,7 @@ #define KVM_PAGES_PER_HPAGE (KVM_HPAGE_SIZE / PAGE_SIZE) #define DE_VECTOR 0 +#define DB_VECTOR 1 #define UD_VECTOR 6 #define NM_VECTOR 7 #define DF_VECTOR 8 -- cgit v1.2.3-70-g09d2 From 77ab6db0a1c403387b403e9351ab3f5ae1df83e6 Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Mon, 14 Jul 2008 12:28:51 +0200 Subject: KVM: VMX: Reinject real mode exception As we execute real mode guests in VM86 mode, exception have to be reinjected appropriately when the guest triggered them. For this purpose the patch adopts the real-mode injection pattern used in vmx_inject_irq to vmx_queue_exception, additionally taking care that the IP is set correctly for #BP exceptions. Furthermore it extends handle_rmode_exception to reinject all those exceptions that can be raised in real mode. This fixes the execution of himem.exe from FreeDOS and also makes its debug.com work properly. Note that guest debugging in real mode is broken now. This has to be fixed by the scheduled debugging infrastructure rework (will be done once base patches for QEMU have been accepted). Signed-off-by: Jan Kiszka Signed-off-by: Avi Kivity --- arch/x86/kvm/vmx.c | 41 +++++++++++++++++++++++++++++++++++++++-- include/asm-x86/kvm_host.h | 4 ++++ 2 files changed, 43 insertions(+), 2 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 5c9dac567a7..2879880076d 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -735,12 +735,30 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu) static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr, bool has_error_code, u32 error_code) { + struct vcpu_vmx *vmx = to_vmx(vcpu); + + if (has_error_code) + vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code); + + if (vcpu->arch.rmode.active) { + vmx->rmode.irq.pending = true; + vmx->rmode.irq.vector = nr; + vmx->rmode.irq.rip = kvm_rip_read(vcpu); + if (nr == BP_VECTOR) + vmx->rmode.irq.rip++; + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, + nr | INTR_TYPE_SOFT_INTR + | (has_error_code ? INTR_INFO_DELIVER_CODE_MASK : 0) + | INTR_INFO_VALID_MASK); + vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1); + kvm_rip_write(vcpu, vmx->rmode.irq.rip - 1); + return; + } + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, nr | INTR_TYPE_EXCEPTION | (has_error_code ? INTR_INFO_DELIVER_CODE_MASK : 0) | INTR_INFO_VALID_MASK); - if (has_error_code) - vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code); } static bool vmx_exception_injected(struct kvm_vcpu *vcpu) @@ -2231,6 +2249,25 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu, if (((vec == GP_VECTOR) || (vec == SS_VECTOR)) && err_code == 0) if (emulate_instruction(vcpu, NULL, 0, 0, 0) == EMULATE_DONE) return 1; + /* + * Forward all other exceptions that are valid in real mode. + * FIXME: Breaks guest debugging in real mode, needs to be fixed with + * the required debugging infrastructure rework. + */ + switch (vec) { + case DE_VECTOR: + case DB_VECTOR: + case BP_VECTOR: + case OF_VECTOR: + case BR_VECTOR: + case UD_VECTOR: + case DF_VECTOR: + case SS_VECTOR: + case GP_VECTOR: + case MF_VECTOR: + kvm_queue_exception(vcpu, vec); + return 1; + } return 0; } diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index da3c1979b73..83afa10c77f 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -58,6 +58,9 @@ #define DE_VECTOR 0 #define DB_VECTOR 1 +#define BP_VECTOR 3 +#define OF_VECTOR 4 +#define BR_VECTOR 5 #define UD_VECTOR 6 #define NM_VECTOR 7 #define DF_VECTOR 8 @@ -66,6 +69,7 @@ #define SS_VECTOR 12 #define GP_VECTOR 13 #define PF_VECTOR 14 +#define MF_VECTOR 16 #define MC_VECTOR 18 #define SELECTOR_TI_MASK (1 << 2) -- cgit v1.2.3-70-g09d2 From 564f15378f04921d5749f27ec53d5e68a6d1d446 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sat, 26 Jul 2008 17:00:59 -0300 Subject: KVM: Add irq ack notifier list This can be used by kvm subsystems that are interested in when interrupts are acked, for example time drift compensation. Signed-off-by: Avi Kivity --- arch/x86/kvm/irq.c | 22 ++++++++++++++++++++++ arch/x86/kvm/irq.h | 5 +++++ include/asm-x86/kvm_host.h | 7 +++++++ 3 files changed, 34 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c index 0d9e55275af..90911958d85 100644 --- a/arch/x86/kvm/irq.c +++ b/arch/x86/kvm/irq.c @@ -111,3 +111,25 @@ void kvm_set_irq(struct kvm *kvm, int irq, int level) kvm_ioapic_set_irq(kvm->arch.vioapic, irq, level); kvm_pic_set_irq(pic_irqchip(kvm), irq, level); } + +void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi) +{ + struct kvm_irq_ack_notifier *kian; + struct hlist_node *n; + + hlist_for_each_entry(kian, n, &kvm->arch.irq_ack_notifier_list, link) + if (kian->gsi == gsi) + kian->irq_acked(kian); +} + +void kvm_register_irq_ack_notifier(struct kvm *kvm, + struct kvm_irq_ack_notifier *kian) +{ + hlist_add_head(&kian->link, &kvm->arch.irq_ack_notifier_list); +} + +void kvm_unregister_irq_ack_notifier(struct kvm *kvm, + struct kvm_irq_ack_notifier *kian) +{ + hlist_del(&kian->link); +} diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index 07ff2aef0c1..95fe718e3ab 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -83,6 +83,11 @@ static inline int irqchip_in_kernel(struct kvm *kvm) void kvm_pic_reset(struct kvm_kpic_state *s); void kvm_set_irq(struct kvm *kvm, int irq, int level); +void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi); +void kvm_register_irq_ack_notifier(struct kvm *kvm, + struct kvm_irq_ack_notifier *kian); +void kvm_unregister_irq_ack_notifier(struct kvm *kvm, + struct kvm_irq_ack_notifier *kian); void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec); void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu); diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 83afa10c77f..d451928fc84 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -321,6 +321,12 @@ struct kvm_mem_alias { gfn_t target_gfn; }; +struct kvm_irq_ack_notifier { + struct hlist_node link; + unsigned gsi; + void (*irq_acked)(struct kvm_irq_ack_notifier *kian); +}; + struct kvm_arch{ int naliases; struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS]; @@ -336,6 +342,7 @@ struct kvm_arch{ struct kvm_pic *vpic; struct kvm_ioapic *vioapic; struct kvm_pit *vpit; + struct hlist_head irq_ack_notifier_list; int round_robin_prev_vcpu; unsigned int tss_addr; -- cgit v1.2.3-70-g09d2 From 3807f345b2c610336c17c7624a0d496a38df75a0 Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Mon, 28 Jul 2008 11:47:52 -0300 Subject: x86: paravirt: factor out cpu_khz to common code KVM intends to use paravirt code to calibrate khz. Xen current code will do just fine. So as a first step, factor out code to pvclock.c. Signed-off-by: Glauber Costa Signed-off-by: Avi Kivity --- arch/x86/kernel/pvclock.c | 12 ++++++++++++ arch/x86/xen/time.c | 11 ++--------- include/asm-x86/pvclock.h | 1 + 3 files changed, 15 insertions(+), 9 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index 05fbe9a0325..1c54b5fb7ae 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -97,6 +97,18 @@ static unsigned pvclock_get_time_values(struct pvclock_shadow_time *dst, return dst->version; } +unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src) +{ + u64 tsc_khz = 1000000ULL << 32; + + do_div(tsc_khz, src->tsc_to_system_mul); + if (src->tsc_shift < 0) + tsc_khz <<= -src->tsc_shift; + else + tsc_khz >>= src->tsc_shift; + return tsc_khz; +} + cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src) { struct pvclock_shadow_time shadow; diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index 004ba86326a..c9f7cda48ed 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -198,17 +198,10 @@ unsigned long long xen_sched_clock(void) /* Get the TSC speed from Xen */ unsigned long xen_tsc_khz(void) { - u64 xen_khz = 1000000ULL << 32; - const struct pvclock_vcpu_time_info *info = + struct pvclock_vcpu_time_info *info = &HYPERVISOR_shared_info->vcpu_info[0].time; - do_div(xen_khz, info->tsc_to_system_mul); - if (info->tsc_shift < 0) - xen_khz <<= -info->tsc_shift; - else - xen_khz >>= info->tsc_shift; - - return xen_khz; + return pvclock_tsc_khz(info); } cycle_t xen_clocksource_read(void) diff --git a/include/asm-x86/pvclock.h b/include/asm-x86/pvclock.h index 1a38f683480..ad29e277fd6 100644 --- a/include/asm-x86/pvclock.h +++ b/include/asm-x86/pvclock.h @@ -6,6 +6,7 @@ /* some helper functions for xen and kvm pv clock sources */ cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src); +unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src); void pvclock_read_wallclock(struct pvclock_wall_clock *wall, struct pvclock_vcpu_time_info *vcpu, struct timespec *ts); -- cgit v1.2.3-70-g09d2 From 4d5c5d0fe89c921336b95f5e7e4f529a9df92f53 Mon Sep 17 00:00:00 2001 From: Ben-Ami Yassour Date: Mon, 28 Jul 2008 19:26:26 +0300 Subject: KVM: pci device assignment Based on a patch from: Amit Shah This patch adds support for handling PCI devices that are assigned to the guest. The device to be assigned to the guest is registered in the host kernel and interrupt delivery is handled. If a device is already assigned, or the device driver for it is still loaded on the host, the device assignment is failed by conveying a -EBUSY reply to the userspace. Devices that share their interrupt line are not supported at the moment. By itself, this patch will not make devices work within the guest. The VT-d extension is required to enable the device to perform DMA. Another alternative is PVDMA. Signed-off-by: Amit Shah Signed-off-by: Ben-Ami Yassour Signed-off-by: Weidong Han Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 243 +++++++++++++++++++++++++++++++++++++++++++++ include/asm-x86/kvm_host.h | 16 +++ include/linux/kvm.h | 19 ++++ 3 files changed, 278 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 94a216562f1..a97157cc42a 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4,10 +4,14 @@ * derived from drivers/kvm/kvm_main.c * * Copyright (C) 2006 Qumranet, Inc. + * Copyright (C) 2008 Qumranet, Inc. + * Copyright IBM Corporation, 2008 * * Authors: * Avi Kivity * Yaniv Kamay + * Amit Shah + * Ben-Ami Yassour * * This work is licensed under the terms of the GNU GPL, version 2. See * the COPYING file in the top-level directory. @@ -23,8 +27,10 @@ #include "x86.h" #include +#include #include #include +#include #include #include #include @@ -98,6 +104,219 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { NULL } }; +struct kvm_assigned_dev_kernel *kvm_find_assigned_dev(struct list_head *head, + int assigned_dev_id) +{ + struct list_head *ptr; + struct kvm_assigned_dev_kernel *match; + + list_for_each(ptr, head) { + match = list_entry(ptr, struct kvm_assigned_dev_kernel, list); + if (match->assigned_dev_id == assigned_dev_id) + return match; + } + return NULL; +} + +static void kvm_assigned_dev_interrupt_work_handler(struct work_struct *work) +{ + struct kvm_assigned_dev_kernel *assigned_dev; + + assigned_dev = container_of(work, struct kvm_assigned_dev_kernel, + interrupt_work); + + /* This is taken to safely inject irq inside the guest. When + * the interrupt injection (or the ioapic code) uses a + * finer-grained lock, update this + */ + mutex_lock(&assigned_dev->kvm->lock); + kvm_set_irq(assigned_dev->kvm, + assigned_dev->guest_irq, 1); + mutex_unlock(&assigned_dev->kvm->lock); + kvm_put_kvm(assigned_dev->kvm); +} + +/* FIXME: Implement the OR logic needed to make shared interrupts on + * this line behave properly + */ +static irqreturn_t kvm_assigned_dev_intr(int irq, void *dev_id) +{ + struct kvm_assigned_dev_kernel *assigned_dev = + (struct kvm_assigned_dev_kernel *) dev_id; + + kvm_get_kvm(assigned_dev->kvm); + schedule_work(&assigned_dev->interrupt_work); + disable_irq_nosync(irq); + return IRQ_HANDLED; +} + +/* Ack the irq line for an assigned device */ +static void kvm_assigned_dev_ack_irq(struct kvm_irq_ack_notifier *kian) +{ + struct kvm_assigned_dev_kernel *dev; + + if (kian->gsi == -1) + return; + + dev = container_of(kian, struct kvm_assigned_dev_kernel, + ack_notifier); + kvm_set_irq(dev->kvm, dev->guest_irq, 0); + enable_irq(dev->host_irq); +} + +static int kvm_vm_ioctl_assign_irq(struct kvm *kvm, + struct kvm_assigned_irq + *assigned_irq) +{ + int r = 0; + struct kvm_assigned_dev_kernel *match; + + mutex_lock(&kvm->lock); + + match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head, + assigned_irq->assigned_dev_id); + if (!match) { + mutex_unlock(&kvm->lock); + return -EINVAL; + } + + if (match->irq_requested) { + match->guest_irq = assigned_irq->guest_irq; + match->ack_notifier.gsi = assigned_irq->guest_irq; + mutex_unlock(&kvm->lock); + return 0; + } + + INIT_WORK(&match->interrupt_work, + kvm_assigned_dev_interrupt_work_handler); + + if (irqchip_in_kernel(kvm)) { + if (assigned_irq->host_irq) + match->host_irq = assigned_irq->host_irq; + else + match->host_irq = match->dev->irq; + match->guest_irq = assigned_irq->guest_irq; + match->ack_notifier.gsi = assigned_irq->guest_irq; + match->ack_notifier.irq_acked = kvm_assigned_dev_ack_irq; + kvm_register_irq_ack_notifier(kvm, &match->ack_notifier); + + /* Even though this is PCI, we don't want to use shared + * interrupts. Sharing host devices with guest-assigned devices + * on the same interrupt line is not a happy situation: there + * are going to be long delays in accepting, acking, etc. + */ + if (request_irq(match->host_irq, kvm_assigned_dev_intr, 0, + "kvm_assigned_device", (void *)match)) { + printk(KERN_INFO "%s: couldn't allocate irq for pv " + "device\n", __func__); + r = -EIO; + goto out; + } + } + + match->irq_requested = true; +out: + mutex_unlock(&kvm->lock); + return r; +} + +static int kvm_vm_ioctl_assign_device(struct kvm *kvm, + struct kvm_assigned_pci_dev *assigned_dev) +{ + int r = 0; + struct kvm_assigned_dev_kernel *match; + struct pci_dev *dev; + + mutex_lock(&kvm->lock); + + match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head, + assigned_dev->assigned_dev_id); + if (match) { + /* device already assigned */ + r = -EINVAL; + goto out; + } + + match = kzalloc(sizeof(struct kvm_assigned_dev_kernel), GFP_KERNEL); + if (match == NULL) { + printk(KERN_INFO "%s: Couldn't allocate memory\n", + __func__); + r = -ENOMEM; + goto out; + } + dev = pci_get_bus_and_slot(assigned_dev->busnr, + assigned_dev->devfn); + if (!dev) { + printk(KERN_INFO "%s: host device not found\n", __func__); + r = -EINVAL; + goto out_free; + } + if (pci_enable_device(dev)) { + printk(KERN_INFO "%s: Could not enable PCI device\n", __func__); + r = -EBUSY; + goto out_put; + } + r = pci_request_regions(dev, "kvm_assigned_device"); + if (r) { + printk(KERN_INFO "%s: Could not get access to device regions\n", + __func__); + goto out_disable; + } + match->assigned_dev_id = assigned_dev->assigned_dev_id; + match->host_busnr = assigned_dev->busnr; + match->host_devfn = assigned_dev->devfn; + match->dev = dev; + + match->kvm = kvm; + + list_add(&match->list, &kvm->arch.assigned_dev_head); + +out: + mutex_unlock(&kvm->lock); + return r; +out_disable: + pci_disable_device(dev); +out_put: + pci_dev_put(dev); +out_free: + kfree(match); + mutex_unlock(&kvm->lock); + return r; +} + +static void kvm_free_assigned_devices(struct kvm *kvm) +{ + struct list_head *ptr, *ptr2; + struct kvm_assigned_dev_kernel *assigned_dev; + + list_for_each_safe(ptr, ptr2, &kvm->arch.assigned_dev_head) { + assigned_dev = list_entry(ptr, + struct kvm_assigned_dev_kernel, + list); + + if (irqchip_in_kernel(kvm) && assigned_dev->irq_requested) { + free_irq(assigned_dev->host_irq, + (void *)assigned_dev); + + kvm_unregister_irq_ack_notifier(kvm, + &assigned_dev-> + ack_notifier); + } + + if (cancel_work_sync(&assigned_dev->interrupt_work)) + /* We had pending work. That means we will have to take + * care of kvm_put_kvm. + */ + kvm_put_kvm(kvm); + + pci_release_regions(assigned_dev->dev); + pci_disable_device(assigned_dev->dev); + pci_dev_put(assigned_dev->dev); + + list_del(&assigned_dev->list); + kfree(assigned_dev); + } +} unsigned long segment_base(u16 selector) { @@ -1766,6 +1985,28 @@ long kvm_arch_vm_ioctl(struct file *filp, r = 0; break; } + case KVM_ASSIGN_PCI_DEVICE: { + struct kvm_assigned_pci_dev assigned_dev; + + r = -EFAULT; + if (copy_from_user(&assigned_dev, argp, sizeof assigned_dev)) + goto out; + r = kvm_vm_ioctl_assign_device(kvm, &assigned_dev); + if (r) + goto out; + break; + } + case KVM_ASSIGN_IRQ: { + struct kvm_assigned_irq assigned_irq; + + r = -EFAULT; + if (copy_from_user(&assigned_irq, argp, sizeof assigned_irq)) + goto out; + r = kvm_vm_ioctl_assign_irq(kvm, &assigned_irq); + if (r) + goto out; + break; + } case KVM_GET_PIT: { struct kvm_pit_state ps; r = -EFAULT; @@ -3945,6 +4186,7 @@ struct kvm *kvm_arch_create_vm(void) return ERR_PTR(-ENOMEM); INIT_LIST_HEAD(&kvm->arch.active_mmu_pages); + INIT_LIST_HEAD(&kvm->arch.assigned_dev_head); return kvm; } @@ -3977,6 +4219,7 @@ static void kvm_free_vcpus(struct kvm *kvm) void kvm_arch_destroy_vm(struct kvm *kvm) { + kvm_free_assigned_devices(kvm); kvm_free_pit(kvm); kfree(kvm->arch.vpic); kfree(kvm->arch.vioapic); diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index d451928fc84..99dddfcecf6 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -327,6 +327,21 @@ struct kvm_irq_ack_notifier { void (*irq_acked)(struct kvm_irq_ack_notifier *kian); }; +struct kvm_assigned_dev_kernel { + struct kvm_irq_ack_notifier ack_notifier; + struct work_struct interrupt_work; + struct list_head list; + struct kvm_assigned_pci_dev assigned_dev; + int assigned_dev_id; + int host_busnr; + int host_devfn; + int host_irq; + int guest_irq; + int irq_requested; + struct pci_dev *dev; + struct kvm *kvm; +}; + struct kvm_arch{ int naliases; struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS]; @@ -339,6 +354,7 @@ struct kvm_arch{ * Hash table of struct kvm_mmu_page. */ struct list_head active_mmu_pages; + struct list_head assigned_dev_head; struct kvm_pic *vpic; struct kvm_ioapic *vioapic; struct kvm_pit *vpit; diff --git a/include/linux/kvm.h b/include/linux/kvm.h index d29b6488144..ef4bc6f8977 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -383,6 +383,7 @@ struct kvm_trace_rec { #define KVM_CAP_MP_STATE 14 #define KVM_CAP_COALESCED_MMIO 15 #define KVM_CAP_SYNC_MMU 16 /* Changes to host mmap are reflected in guest */ +#define KVM_CAP_DEVICE_ASSIGNMENT 17 /* * ioctls for VM fds @@ -412,6 +413,10 @@ struct kvm_trace_rec { _IOW(KVMIO, 0x67, struct kvm_coalesced_mmio_zone) #define KVM_UNREGISTER_COALESCED_MMIO \ _IOW(KVMIO, 0x68, struct kvm_coalesced_mmio_zone) +#define KVM_ASSIGN_PCI_DEVICE _IOR(KVMIO, 0x69, \ + struct kvm_assigned_pci_dev) +#define KVM_ASSIGN_IRQ _IOR(KVMIO, 0x70, \ + struct kvm_assigned_irq) /* * ioctls for vcpu fds @@ -476,4 +481,18 @@ struct kvm_trace_rec { #define KVM_TRC_STLB_INVAL (KVM_TRC_HANDLER + 0x18) #define KVM_TRC_PPC_INSTR (KVM_TRC_HANDLER + 0x19) +struct kvm_assigned_pci_dev { + __u32 assigned_dev_id; + __u32 busnr; + __u32 devfn; + __u32 flags; +}; + +struct kvm_assigned_irq { + __u32 assigned_dev_id; + __u32 host_irq; + __u32 guest_irq; + __u32 flags; +}; + #endif -- cgit v1.2.3-70-g09d2 From 6ad18fba05228fb1d47cdbc0339fe8b3fca1ca26 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Mon, 11 Aug 2008 10:01:49 -0700 Subject: KVM: Reduce stack usage in kvm_pv_mmu_op() We're in a hot path. We can't use kmalloc() because it might impact performance. So, we just stick the buffer that we need into the kvm_vcpu_arch structure. This is used very often, so it is not really a waste. We also have to move the buffer structure's definition to the arch-specific x86 kvm header. Signed-off-by: Dave Hansen Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 23 ++++++++--------------- include/asm-x86/kvm_host.h | 10 ++++++++++ 2 files changed, 18 insertions(+), 15 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index c3afbfe6b0c..171bcea1be2 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -135,13 +135,6 @@ module_param(dbg, bool, 0644); #define ACC_USER_MASK PT_USER_MASK #define ACC_ALL (ACC_EXEC_MASK | ACC_WRITE_MASK | ACC_USER_MASK) -struct kvm_pv_mmu_op_buffer { - void *ptr; - unsigned len; - unsigned processed; - char buf[512] __aligned(sizeof(long)); -}; - struct kvm_rmap_desc { u64 *shadow_ptes[RMAP_EXT]; struct kvm_rmap_desc *more; @@ -2292,18 +2285,18 @@ int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes, gpa_t addr, unsigned long *ret) { int r; - struct kvm_pv_mmu_op_buffer buffer; + struct kvm_pv_mmu_op_buffer *buffer = &vcpu->arch.mmu_op_buffer; - buffer.ptr = buffer.buf; - buffer.len = min_t(unsigned long, bytes, sizeof buffer.buf); - buffer.processed = 0; + buffer->ptr = buffer->buf; + buffer->len = min_t(unsigned long, bytes, sizeof buffer->buf); + buffer->processed = 0; - r = kvm_read_guest(vcpu->kvm, addr, buffer.buf, buffer.len); + r = kvm_read_guest(vcpu->kvm, addr, buffer->buf, buffer->len); if (r) goto out; - while (buffer.len) { - r = kvm_pv_mmu_op_one(vcpu, &buffer); + while (buffer->len) { + r = kvm_pv_mmu_op_one(vcpu, buffer); if (r < 0) goto out; if (r == 0) @@ -2312,7 +2305,7 @@ int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes, r = 1; out: - *ret = buffer.processed; + *ret = buffer->processed; return r; } diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 99dddfcecf6..9cb4b4dae5c 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -201,6 +201,13 @@ struct kvm_mmu_page { }; }; +struct kvm_pv_mmu_op_buffer { + void *ptr; + unsigned len; + unsigned processed; + char buf[512] __aligned(sizeof(long)); +}; + /* * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level * 32-bit). The kvm_mmu structure abstracts the details of the current mmu @@ -248,6 +255,9 @@ struct kvm_vcpu_arch { bool tpr_access_reporting; struct kvm_mmu mmu; + /* only needed in kvm_pv_mmu_op() path, but it's hot so + * put it here to avoid allocation */ + struct kvm_pv_mmu_op_buffer mmu_op_buffer; struct kvm_mmu_memory_cache mmu_pte_chain_cache; struct kvm_mmu_memory_cache mmu_rmap_desc_cache; -- cgit v1.2.3-70-g09d2 From 8349b5cd816adf102c078553ec8ca63b3beb457c Mon Sep 17 00:00:00 2001 From: Ben-Ami Yassour Date: Tue, 5 Aug 2008 15:30:13 +0300 Subject: KVM: remove unused field from the assigned dev struct Remove unused field: struct kvm_assigned_pci_dev assigned_dev from struct: struct kvm_assigned_dev_kernel Signed-off-by: Ben-Ami Yassour Signed-off-by: Avi Kivity --- include/asm-x86/kvm_host.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 9cb4b4dae5c..225fdb835d4 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -341,7 +341,6 @@ struct kvm_assigned_dev_kernel { struct kvm_irq_ack_notifier ack_notifier; struct work_struct interrupt_work; struct list_head list; - struct kvm_assigned_pci_dev assigned_dev; int assigned_dev_id; int host_busnr; int host_devfn; -- cgit v1.2.3-70-g09d2 From 8ceed34744f81c4a33d68ab825fd9ad3dd5f5505 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 14 Aug 2008 21:25:47 +0300 Subject: KVM: Simplify exception entries by using __ASM_SIZE and _ASM_PTR Signed-off-by: Avi Kivity --- include/asm-x86/kvm_host.h | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 225fdb835d4..b6d26b80b75 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -734,15 +734,6 @@ enum { TASK_SWITCH_GATE = 3, }; - -#ifdef CONFIG_64BIT -# define KVM_EX_ENTRY ".quad" -# define KVM_EX_PUSH "pushq" -#else -# define KVM_EX_ENTRY ".long" -# define KVM_EX_PUSH "pushl" -#endif - /* * Hardware virtualization extension instructions may fault if a * reboot turns off virtualization while processes are running. @@ -754,11 +745,11 @@ asmlinkage void kvm_handle_fault_on_reboot(void); "666: " insn "\n\t" \ ".pushsection .fixup, \"ax\" \n" \ "667: \n\t" \ - KVM_EX_PUSH " $666b \n\t" \ + __ASM_SIZE(push) " $666b \n\t" \ "jmp kvm_handle_fault_on_reboot \n\t" \ ".popsection \n\t" \ ".pushsection __ex_table, \"a\" \n\t" \ - KVM_EX_ENTRY " 666b, 667b \n\t" \ + _ASM_PTR " 666b, 667b \n\t" \ ".popsection" #define KVM_ARCH_WANT_MMU_NOTIFIER -- cgit v1.2.3-70-g09d2 From fa89a81766e33343fa8f0fe0e371819bf94a17a1 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 1 Sep 2008 15:57:51 +0300 Subject: KVM: Add statistics for guest irq injections These can help show whether a guest is making progress or not. Signed-off-by: Avi Kivity --- arch/x86/kvm/svm.c | 1 + arch/x86/kvm/vmx.c | 1 + arch/x86/kvm/x86.c | 1 + include/asm-x86/kvm_host.h | 1 + 4 files changed, 4 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 60228888d1b..9b54550fa4d 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1519,6 +1519,7 @@ static inline void svm_inject_irq(struct vcpu_svm *svm, int irq) KVMTRACE_1D(INJ_VIRQ, &svm->vcpu, (u32)irq, handler); + ++svm->vcpu.stat.irq_injections; control = &svm->vmcb->control; control->int_vector = irq; control->int_ctl &= ~V_INTR_PRIO_MASK; diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 71e57ae1cab..e7e8c86f1b7 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2341,6 +2341,7 @@ static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq) KVMTRACE_1D(INJ_VIRQ, vcpu, (u32)irq, handler); + ++vcpu->stat.irq_injections; if (vcpu->arch.rmode.active) { vmx->rmode.irq.pending = true; vmx->rmode.irq.vector = irq; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index e3b89662cf6..3f3cb7107c0 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -92,6 +92,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "fpu_reload", VCPU_STAT(fpu_reload) }, { "insn_emulation", VCPU_STAT(insn_emulation) }, { "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) }, + { "irq_injections", VCPU_STAT(irq_injections) }, { "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) }, { "mmu_pte_write", VM_STAT(mmu_pte_write) }, { "mmu_pte_updated", VM_STAT(mmu_pte_updated) }, diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index b6d26b80b75..68a3ac13afc 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -413,6 +413,7 @@ struct kvm_vcpu_stat { u32 insn_emulation; u32 insn_emulation_fail; u32 hypercalls; + u32 irq_injections; }; struct descriptor_table { -- cgit v1.2.3-70-g09d2 From defed7ed926eca9f178a632df05baa866abe754e Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Thu, 11 Sep 2008 15:27:50 +0800 Subject: x86: Move FEATURE_CONTROL bits to msr-index.h For MSR_IA32_FEATURE_CONTROL is already there. Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/kvm/vmx.h | 3 --- include/asm-x86/msr-index.h | 3 +++ 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/vmx.h b/arch/x86/kvm/vmx.h index 44cfab70644..3e010d21fdd 100644 --- a/arch/x86/kvm/vmx.h +++ b/arch/x86/kvm/vmx.h @@ -331,9 +331,6 @@ enum vmcs_field { #define AR_RESERVD_MASK 0xfffe0f00 -#define FEATURE_CONTROL_LOCKED (1<<0) -#define FEATURE_CONTROL_VMXON_ENABLED (1<<2) - #define APIC_ACCESS_PAGE_PRIVATE_MEMSLOT 9 #define IDENTITY_PAGETABLE_PRIVATE_MEMSLOT 10 diff --git a/include/asm-x86/msr-index.h b/include/asm-x86/msr-index.h index 0bb43301a20..dabd10f0bbe 100644 --- a/include/asm-x86/msr-index.h +++ b/include/asm-x86/msr-index.h @@ -178,6 +178,9 @@ #define MSR_IA32_EBL_CR_POWERON 0x0000002a #define MSR_IA32_FEATURE_CONTROL 0x0000003a +#define FEATURE_CONTROL_LOCKED (1<<0) +#define FEATURE_CONTROL_VMXON_ENABLED (1<<2) + #define MSR_IA32_APICBASE 0x0000001b #define MSR_IA32_APICBASE_BSP (1<<8) #define MSR_IA32_APICBASE_ENABLE (1<<11) -- cgit v1.2.3-70-g09d2 From 62c476c7c7f25a5b245b9902a935636e6316e58c Mon Sep 17 00:00:00 2001 From: Ben-Ami Yassour Date: Sun, 14 Sep 2008 03:48:28 +0300 Subject: KVM: Device Assignment with VT-d Based on a patch by: Kay, Allen M This patch enables PCI device assignment based on VT-d support. When a device is assigned to the guest, the guest memory is pinned and the mapping is updated in the VT-d IOMMU. [Amit: Expose KVM_CAP_IOMMU so we can check if an IOMMU is present and also control enable/disable from userspace] Signed-off-by: Kay, Allen M Signed-off-by: Weidong Han Signed-off-by: Ben-Ami Yassour Signed-off-by: Amit Shah Acked-by: Mark Gross Signed-off-by: Avi Kivity --- arch/x86/kvm/Makefile | 3 + arch/x86/kvm/vtd.c | 198 +++++++++++++++++++++++++++++++++++++++++++++ arch/x86/kvm/x86.c | 14 ++++ include/asm-x86/kvm_host.h | 23 +----- include/linux/kvm.h | 3 + include/linux/kvm_host.h | 52 ++++++++++++ virt/kvm/kvm_main.c | 9 ++- 7 files changed, 281 insertions(+), 21 deletions(-) create mode 100644 arch/x86/kvm/vtd.c (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile index d0e940bb6f4..3072b17447a 100644 --- a/arch/x86/kvm/Makefile +++ b/arch/x86/kvm/Makefile @@ -12,6 +12,9 @@ EXTRA_CFLAGS += -Ivirt/kvm -Iarch/x86/kvm kvm-objs := $(common-objs) x86.o mmu.o x86_emulate.o i8259.o irq.o lapic.o \ i8254.o +ifeq ($(CONFIG_DMAR),y) +kvm-objs += vtd.o +endif obj-$(CONFIG_KVM) += kvm.o kvm-intel-objs = vmx.o obj-$(CONFIG_KVM_INTEL) += kvm-intel.o diff --git a/arch/x86/kvm/vtd.c b/arch/x86/kvm/vtd.c new file mode 100644 index 00000000000..667bf3fb64b --- /dev/null +++ b/arch/x86/kvm/vtd.c @@ -0,0 +1,198 @@ +/* + * Copyright (c) 2006, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple + * Place - Suite 330, Boston, MA 02111-1307 USA. + * + * Copyright (C) 2006-2008 Intel Corporation + * Copyright IBM Corporation, 2008 + * Author: Allen M. Kay + * Author: Weidong Han + * Author: Ben-Ami Yassour + */ + +#include +#include +#include +#include +#include + +static int kvm_iommu_unmap_memslots(struct kvm *kvm); +static void kvm_iommu_put_pages(struct kvm *kvm, + gfn_t base_gfn, unsigned long npages); + +int kvm_iommu_map_pages(struct kvm *kvm, + gfn_t base_gfn, unsigned long npages) +{ + gfn_t gfn = base_gfn; + pfn_t pfn; + int i, r; + struct dmar_domain *domain = kvm->arch.intel_iommu_domain; + + /* check if iommu exists and in use */ + if (!domain) + return 0; + + r = -EINVAL; + for (i = 0; i < npages; i++) { + /* check if already mapped */ + pfn = (pfn_t)intel_iommu_iova_to_pfn(domain, + gfn_to_gpa(gfn)); + if (pfn && !is_mmio_pfn(pfn)) + continue; + + pfn = gfn_to_pfn(kvm, gfn); + if (!is_mmio_pfn(pfn)) { + r = intel_iommu_page_mapping(domain, + gfn_to_gpa(gfn), + pfn_to_hpa(pfn), + PAGE_SIZE, + DMA_PTE_READ | + DMA_PTE_WRITE); + if (r) { + printk(KERN_DEBUG "kvm_iommu_map_pages:" + "iommu failed to map pfn=%lx\n", pfn); + goto unmap_pages; + } + } else { + printk(KERN_DEBUG "kvm_iommu_map_page:" + "invalid pfn=%lx\n", pfn); + goto unmap_pages; + } + gfn++; + } + return 0; + +unmap_pages: + kvm_iommu_put_pages(kvm, base_gfn, i); + return r; +} + +static int kvm_iommu_map_memslots(struct kvm *kvm) +{ + int i, r; + + down_read(&kvm->slots_lock); + for (i = 0; i < kvm->nmemslots; i++) { + r = kvm_iommu_map_pages(kvm, kvm->memslots[i].base_gfn, + kvm->memslots[i].npages); + if (r) + break; + } + up_read(&kvm->slots_lock); + return r; +} + +int kvm_iommu_map_guest(struct kvm *kvm, + struct kvm_assigned_dev_kernel *assigned_dev) +{ + struct pci_dev *pdev = NULL; + int r; + + if (!intel_iommu_found()) { + printk(KERN_ERR "%s: intel iommu not found\n", __func__); + return -ENODEV; + } + + printk(KERN_DEBUG "VT-d direct map: host bdf = %x:%x:%x\n", + assigned_dev->host_busnr, + PCI_SLOT(assigned_dev->host_devfn), + PCI_FUNC(assigned_dev->host_devfn)); + + pdev = assigned_dev->dev; + + if (pdev == NULL) { + if (kvm->arch.intel_iommu_domain) { + intel_iommu_domain_exit(kvm->arch.intel_iommu_domain); + kvm->arch.intel_iommu_domain = NULL; + } + return -ENODEV; + } + + kvm->arch.intel_iommu_domain = intel_iommu_domain_alloc(pdev); + if (!kvm->arch.intel_iommu_domain) + return -ENODEV; + + r = kvm_iommu_map_memslots(kvm); + if (r) + goto out_unmap; + + intel_iommu_detach_dev(kvm->arch.intel_iommu_domain, + pdev->bus->number, pdev->devfn); + + r = intel_iommu_context_mapping(kvm->arch.intel_iommu_domain, + pdev); + if (r) { + printk(KERN_ERR "Domain context map for %s failed", + pci_name(pdev)); + goto out_unmap; + } + return 0; + +out_unmap: + kvm_iommu_unmap_memslots(kvm); + return r; +} + +static void kvm_iommu_put_pages(struct kvm *kvm, + gfn_t base_gfn, unsigned long npages) +{ + gfn_t gfn = base_gfn; + pfn_t pfn; + struct dmar_domain *domain = kvm->arch.intel_iommu_domain; + int i; + + for (i = 0; i < npages; i++) { + pfn = (pfn_t)intel_iommu_iova_to_pfn(domain, + gfn_to_gpa(gfn)); + kvm_release_pfn_clean(pfn); + gfn++; + } +} + +static int kvm_iommu_unmap_memslots(struct kvm *kvm) +{ + int i; + down_read(&kvm->slots_lock); + for (i = 0; i < kvm->nmemslots; i++) { + kvm_iommu_put_pages(kvm, kvm->memslots[i].base_gfn, + kvm->memslots[i].npages); + } + up_read(&kvm->slots_lock); + + return 0; +} + +int kvm_iommu_unmap_guest(struct kvm *kvm) +{ + struct kvm_assigned_dev_kernel *entry; + struct dmar_domain *domain = kvm->arch.intel_iommu_domain; + + /* check if iommu exists and in use */ + if (!domain) + return 0; + + list_for_each_entry(entry, &kvm->arch.assigned_dev_head, list) { + printk(KERN_DEBUG "VT-d unmap: host bdf = %x:%x:%x\n", + entry->host_busnr, + PCI_SLOT(entry->host_devfn), + PCI_FUNC(entry->host_devfn)); + + /* detach kvm dmar domain */ + intel_iommu_detach_dev(domain, entry->host_busnr, + entry->host_devfn); + } + kvm_iommu_unmap_memslots(kvm); + intel_iommu_domain_exit(domain); + return 0; +} diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 2134f3e0a51..c8a2793626e 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include @@ -277,9 +278,18 @@ static int kvm_vm_ioctl_assign_device(struct kvm *kvm, list_add(&match->list, &kvm->arch.assigned_dev_head); + if (assigned_dev->flags & KVM_DEV_ASSIGN_ENABLE_IOMMU) { + r = kvm_iommu_map_guest(kvm, match); + if (r) + goto out_list_del; + } + out: mutex_unlock(&kvm->lock); return r; +out_list_del: + list_del(&match->list); + pci_release_regions(dev); out_disable: pci_disable_device(dev); out_put: @@ -1147,6 +1157,9 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_PV_MMU: r = !tdp_enabled; break; + case KVM_CAP_IOMMU: + r = intel_iommu_found(); + break; default: r = 0; break; @@ -4282,6 +4295,7 @@ static void kvm_free_vcpus(struct kvm *kvm) void kvm_arch_destroy_vm(struct kvm *kvm) { + kvm_iommu_unmap_guest(kvm); kvm_free_assigned_devices(kvm); kvm_free_pit(kvm); kfree(kvm->arch.vpic); diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 68a3ac13afc..805629c0f15 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -331,26 +331,6 @@ struct kvm_mem_alias { gfn_t target_gfn; }; -struct kvm_irq_ack_notifier { - struct hlist_node link; - unsigned gsi; - void (*irq_acked)(struct kvm_irq_ack_notifier *kian); -}; - -struct kvm_assigned_dev_kernel { - struct kvm_irq_ack_notifier ack_notifier; - struct work_struct interrupt_work; - struct list_head list; - int assigned_dev_id; - int host_busnr; - int host_devfn; - int host_irq; - int guest_irq; - int irq_requested; - struct pci_dev *dev; - struct kvm *kvm; -}; - struct kvm_arch{ int naliases; struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS]; @@ -364,6 +344,7 @@ struct kvm_arch{ */ struct list_head active_mmu_pages; struct list_head assigned_dev_head; + struct dmar_domain *intel_iommu_domain; struct kvm_pic *vpic; struct kvm_ioapic *vioapic; struct kvm_pit *vpit; @@ -514,6 +495,8 @@ int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa, int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes, gpa_t addr, unsigned long *ret); +int is_mmio_pfn(pfn_t pfn); + extern bool tdp_enabled; enum emulation_result { diff --git a/include/linux/kvm.h b/include/linux/kvm.h index ef4bc6f8977..4269be171fa 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -384,6 +384,7 @@ struct kvm_trace_rec { #define KVM_CAP_COALESCED_MMIO 15 #define KVM_CAP_SYNC_MMU 16 /* Changes to host mmap are reflected in guest */ #define KVM_CAP_DEVICE_ASSIGNMENT 17 +#define KVM_CAP_IOMMU 18 /* * ioctls for VM fds @@ -495,4 +496,6 @@ struct kvm_assigned_irq { __u32 flags; }; +#define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0) + #endif diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 4b036430ea2..6252802c3cc 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -286,6 +286,53 @@ int kvm_cpu_has_interrupt(struct kvm_vcpu *v); int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu); void kvm_vcpu_kick(struct kvm_vcpu *vcpu); +struct kvm_irq_ack_notifier { + struct hlist_node link; + unsigned gsi; + void (*irq_acked)(struct kvm_irq_ack_notifier *kian); +}; + +struct kvm_assigned_dev_kernel { + struct kvm_irq_ack_notifier ack_notifier; + struct work_struct interrupt_work; + struct list_head list; + int assigned_dev_id; + int host_busnr; + int host_devfn; + int host_irq; + int guest_irq; + int irq_requested; + struct pci_dev *dev; + struct kvm *kvm; +}; + +#ifdef CONFIG_DMAR +int kvm_iommu_map_pages(struct kvm *kvm, gfn_t base_gfn, + unsigned long npages); +int kvm_iommu_map_guest(struct kvm *kvm, + struct kvm_assigned_dev_kernel *assigned_dev); +int kvm_iommu_unmap_guest(struct kvm *kvm); +#else /* CONFIG_DMAR */ +static inline int kvm_iommu_map_pages(struct kvm *kvm, + gfn_t base_gfn, + unsigned long npages) +{ + return 0; +} + +static inline int kvm_iommu_map_guest(struct kvm *kvm, + struct kvm_assigned_dev_kernel + *assigned_dev) +{ + return -ENODEV; +} + +static inline int kvm_iommu_unmap_guest(struct kvm *kvm) +{ + return 0; +} +#endif /* CONFIG_DMAR */ + static inline void kvm_guest_enter(void) { account_system_vtime(current); @@ -308,6 +355,11 @@ static inline gpa_t gfn_to_gpa(gfn_t gfn) return (gpa_t)gfn << PAGE_SHIFT; } +static inline hpa_t pfn_to_hpa(pfn_t pfn) +{ + return (hpa_t)pfn << PAGE_SHIFT; +} + static inline void kvm_migrate_timers(struct kvm_vcpu *vcpu) { set_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 63e661be040..f42d5c2a396 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -41,6 +41,7 @@ #include #include #include +#include #include #include @@ -76,7 +77,7 @@ static inline int valid_vcpu(int n) return likely(n >= 0 && n < KVM_MAX_VCPUS); } -static inline int is_mmio_pfn(pfn_t pfn) +inline int is_mmio_pfn(pfn_t pfn) { if (pfn_valid(pfn)) return PageReserved(pfn_to_page(pfn)); @@ -578,6 +579,12 @@ int __kvm_set_memory_region(struct kvm *kvm, } kvm_free_physmem_slot(&old, &new); + + /* map the pages in iommu page table */ + r = kvm_iommu_map_pages(kvm, base_gfn, npages); + if (r) + goto out; + return 0; out_free: -- cgit v1.2.3-70-g09d2 From e8bc217aef67d41d767ede6e7a7eb10f1d47c86c Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 23 Sep 2008 13:18:33 -0300 Subject: KVM: MMU: mode specific sync_page Examine guest pagetable and bring the shadow back in sync. Caller is responsible for local TLB flush before re-entering guest mode. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 10 +++++++++ arch/x86/kvm/paging_tmpl.h | 54 ++++++++++++++++++++++++++++++++++++++++++++++ include/asm-x86/kvm_host.h | 2 ++ 3 files changed, 66 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 731e6fe9cb0..90f01169c8f 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -871,6 +871,12 @@ static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu, sp->spt[i] = shadow_trap_nonpresent_pte; } +static int nonpaging_sync_page(struct kvm_vcpu *vcpu, + struct kvm_mmu_page *sp) +{ + return 1; +} + static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) { unsigned index; @@ -1547,6 +1553,7 @@ static int nonpaging_init_context(struct kvm_vcpu *vcpu) context->gva_to_gpa = nonpaging_gva_to_gpa; context->free = nonpaging_free; context->prefetch_page = nonpaging_prefetch_page; + context->sync_page = nonpaging_sync_page; context->root_level = 0; context->shadow_root_level = PT32E_ROOT_LEVEL; context->root_hpa = INVALID_PAGE; @@ -1594,6 +1601,7 @@ static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level) context->page_fault = paging64_page_fault; context->gva_to_gpa = paging64_gva_to_gpa; context->prefetch_page = paging64_prefetch_page; + context->sync_page = paging64_sync_page; context->free = paging_free; context->root_level = level; context->shadow_root_level = level; @@ -1615,6 +1623,7 @@ static int paging32_init_context(struct kvm_vcpu *vcpu) context->gva_to_gpa = paging32_gva_to_gpa; context->free = paging_free; context->prefetch_page = paging32_prefetch_page; + context->sync_page = paging32_sync_page; context->root_level = PT32_ROOT_LEVEL; context->shadow_root_level = PT32E_ROOT_LEVEL; context->root_hpa = INVALID_PAGE; @@ -1634,6 +1643,7 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) context->page_fault = tdp_page_fault; context->free = nonpaging_free; context->prefetch_page = nonpaging_prefetch_page; + context->sync_page = nonpaging_sync_page; context->shadow_root_level = kvm_x86_ops->get_tdp_level(); context->root_hpa = INVALID_PAGE; diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index e9fbaa44d44..776fb6d2fd8 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -507,6 +507,60 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu, } } +/* + * Using the cached information from sp->gfns is safe because: + * - The spte has a reference to the struct page, so the pfn for a given gfn + * can't change unless all sptes pointing to it are nuked first. + * - Alias changes zap the entire shadow cache. + */ +static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + int i, offset, nr_present; + + offset = nr_present = 0; + + if (PTTYPE == 32) + offset = sp->role.quadrant << PT64_LEVEL_BITS; + + for (i = 0; i < PT64_ENT_PER_PAGE; i++) { + unsigned pte_access; + pt_element_t gpte; + gpa_t pte_gpa; + gfn_t gfn = sp->gfns[i]; + + if (!is_shadow_present_pte(sp->spt[i])) + continue; + + pte_gpa = gfn_to_gpa(sp->gfn); + pte_gpa += (i+offset) * sizeof(pt_element_t); + + if (kvm_read_guest_atomic(vcpu->kvm, pte_gpa, &gpte, + sizeof(pt_element_t))) + return -EINVAL; + + if (gpte_to_gfn(gpte) != gfn || !is_present_pte(gpte) || + !(gpte & PT_ACCESSED_MASK)) { + u64 nonpresent; + + rmap_remove(vcpu->kvm, &sp->spt[i]); + if (is_present_pte(gpte)) + nonpresent = shadow_trap_nonpresent_pte; + else + nonpresent = shadow_notrap_nonpresent_pte; + set_shadow_pte(&sp->spt[i], nonpresent); + continue; + } + + nr_present++; + pte_access = sp->role.access & FNAME(gpte_access)(vcpu, gpte); + set_spte(vcpu, &sp->spt[i], pte_access, 0, 0, + is_dirty_pte(gpte), 0, gfn, + spte_to_pfn(sp->spt[i]), true); + } + + return !nr_present; +} + #undef pt_element_t #undef guest_walker #undef shadow_walker diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 805629c0f15..8bad9bd9b37 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -220,6 +220,8 @@ struct kvm_mmu { gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva); void (*prefetch_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page); + int (*sync_page)(struct kvm_vcpu *vcpu, + struct kvm_mmu_page *sp); hpa_t root_hpa; int root_level; int shadow_root_level; -- cgit v1.2.3-70-g09d2 From 0ba73cdadb8ac172f396df7e23c4a9cebd59b550 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 23 Sep 2008 13:18:34 -0300 Subject: KVM: MMU: sync roots on mmu reload Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 36 ++++++++++++++++++++++++++++++++++++ arch/x86/kvm/x86.c | 1 + include/asm-x86/kvm_host.h | 1 + 3 files changed, 38 insertions(+) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 90f01169c8f..9d8c4bb68a8 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -1471,6 +1471,41 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu) vcpu->arch.mmu.root_hpa = __pa(vcpu->arch.mmu.pae_root); } +static void mmu_sync_children(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ +} + +static void mmu_sync_roots(struct kvm_vcpu *vcpu) +{ + int i; + struct kvm_mmu_page *sp; + + if (!VALID_PAGE(vcpu->arch.mmu.root_hpa)) + return; + if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) { + hpa_t root = vcpu->arch.mmu.root_hpa; + sp = page_header(root); + mmu_sync_children(vcpu, sp); + return; + } + for (i = 0; i < 4; ++i) { + hpa_t root = vcpu->arch.mmu.pae_root[i]; + + if (root) { + root &= PT64_BASE_ADDR_MASK; + sp = page_header(root); + mmu_sync_children(vcpu, sp); + } + } +} + +void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu) +{ + spin_lock(&vcpu->kvm->mmu_lock); + mmu_sync_roots(vcpu); + spin_unlock(&vcpu->kvm->mmu_lock); +} + static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t vaddr) { return vaddr; @@ -1715,6 +1750,7 @@ int kvm_mmu_load(struct kvm_vcpu *vcpu) spin_lock(&vcpu->kvm->mmu_lock); kvm_mmu_free_some_pages(vcpu); mmu_alloc_roots(vcpu); + mmu_sync_roots(vcpu); spin_unlock(&vcpu->kvm->mmu_lock); kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa); kvm_mmu_flush_tlb(vcpu); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 08edeabf15e..88e6d9abbd2 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -594,6 +594,7 @@ EXPORT_SYMBOL_GPL(kvm_set_cr4); void kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) { if (cr3 == vcpu->arch.cr3 && !pdptrs_changed(vcpu)) { + kvm_mmu_sync_roots(vcpu); kvm_mmu_flush_tlb(vcpu); return; } diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 8bad9bd9b37..475d8ab83bf 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -584,6 +584,7 @@ int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva); void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu); int kvm_mmu_load(struct kvm_vcpu *vcpu); void kvm_mmu_unload(struct kvm_vcpu *vcpu); +void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu); int kvm_emulate_hypercall(struct kvm_vcpu *vcpu); -- cgit v1.2.3-70-g09d2 From a7052897b3bcd568a9f5bfaa558957039e7e7ec0 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 23 Sep 2008 13:18:35 -0300 Subject: KVM: x86: trap invlpg With pages out of sync invlpg needs to be trapped. For now simply nuke the entry. Untested on AMD. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 18 ++++++++++++++++++ arch/x86/kvm/paging_tmpl.h | 25 +++++++++++++++++++++++++ arch/x86/kvm/svm.c | 13 +++++++++++-- arch/x86/kvm/vmx.c | 19 ++++++++++++++++--- arch/x86/kvm/x86.c | 1 + include/asm-x86/kvm_host.h | 2 ++ 6 files changed, 73 insertions(+), 5 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 9d8c4bb68a8..e89af1df4fc 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -877,6 +877,10 @@ static int nonpaging_sync_page(struct kvm_vcpu *vcpu, return 1; } +static void nonpaging_invlpg(struct kvm_vcpu *vcpu, gva_t gva) +{ +} + static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) { unsigned index; @@ -1589,6 +1593,7 @@ static int nonpaging_init_context(struct kvm_vcpu *vcpu) context->free = nonpaging_free; context->prefetch_page = nonpaging_prefetch_page; context->sync_page = nonpaging_sync_page; + context->invlpg = nonpaging_invlpg; context->root_level = 0; context->shadow_root_level = PT32E_ROOT_LEVEL; context->root_hpa = INVALID_PAGE; @@ -1637,6 +1642,7 @@ static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level) context->gva_to_gpa = paging64_gva_to_gpa; context->prefetch_page = paging64_prefetch_page; context->sync_page = paging64_sync_page; + context->invlpg = paging64_invlpg; context->free = paging_free; context->root_level = level; context->shadow_root_level = level; @@ -1659,6 +1665,7 @@ static int paging32_init_context(struct kvm_vcpu *vcpu) context->free = paging_free; context->prefetch_page = paging32_prefetch_page; context->sync_page = paging32_sync_page; + context->invlpg = paging32_invlpg; context->root_level = PT32_ROOT_LEVEL; context->shadow_root_level = PT32E_ROOT_LEVEL; context->root_hpa = INVALID_PAGE; @@ -1679,6 +1686,7 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) context->free = nonpaging_free; context->prefetch_page = nonpaging_prefetch_page; context->sync_page = nonpaging_sync_page; + context->invlpg = nonpaging_invlpg; context->shadow_root_level = kvm_x86_ops->get_tdp_level(); context->root_hpa = INVALID_PAGE; @@ -2071,6 +2079,16 @@ out: } EXPORT_SYMBOL_GPL(kvm_mmu_page_fault); +void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva) +{ + spin_lock(&vcpu->kvm->mmu_lock); + vcpu->arch.mmu.invlpg(vcpu, gva); + spin_unlock(&vcpu->kvm->mmu_lock); + kvm_mmu_flush_tlb(vcpu); + ++vcpu->stat.invlpg; +} +EXPORT_SYMBOL_GPL(kvm_mmu_invlpg); + void kvm_enable_tdp(void) { tdp_enabled = true; diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index 776fb6d2fd8..dc169e8148b 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -461,6 +461,31 @@ out_unlock: return 0; } +static int FNAME(shadow_invlpg_entry)(struct kvm_shadow_walk *_sw, + struct kvm_vcpu *vcpu, u64 addr, + u64 *sptep, int level) +{ + + if (level == PT_PAGE_TABLE_LEVEL) { + if (is_shadow_present_pte(*sptep)) + rmap_remove(vcpu->kvm, sptep); + set_shadow_pte(sptep, shadow_trap_nonpresent_pte); + return 1; + } + if (!is_shadow_present_pte(*sptep)) + return 1; + return 0; +} + +static void FNAME(invlpg)(struct kvm_vcpu *vcpu, gva_t gva) +{ + struct shadow_walker walker = { + .walker = { .entry = FNAME(shadow_invlpg_entry), }, + }; + + walk_shadow(&walker.walker, vcpu, gva); +} + static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr) { struct guest_walker walker; diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 9b54550fa4d..9c4ce657d96 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -525,6 +525,7 @@ static void init_vmcb(struct vcpu_svm *svm) (1ULL << INTERCEPT_CPUID) | (1ULL << INTERCEPT_INVD) | (1ULL << INTERCEPT_HLT) | + (1ULL << INTERCEPT_INVLPG) | (1ULL << INTERCEPT_INVLPGA) | (1ULL << INTERCEPT_IOIO_PROT) | (1ULL << INTERCEPT_MSR_PROT) | @@ -589,7 +590,8 @@ static void init_vmcb(struct vcpu_svm *svm) if (npt_enabled) { /* Setup VMCB for Nested Paging */ control->nested_ctl = 1; - control->intercept &= ~(1ULL << INTERCEPT_TASK_SWITCH); + control->intercept &= ~((1ULL << INTERCEPT_TASK_SWITCH) | + (1ULL << INTERCEPT_INVLPG)); control->intercept_exceptions &= ~(1 << PF_VECTOR); control->intercept_cr_read &= ~(INTERCEPT_CR0_MASK| INTERCEPT_CR3_MASK); @@ -1164,6 +1166,13 @@ static int cpuid_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) return 1; } +static int invlpg_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) +{ + if (emulate_instruction(&svm->vcpu, kvm_run, 0, 0, 0) != EMULATE_DONE) + pr_unimpl(&svm->vcpu, "%s: failed\n", __func__); + return 1; +} + static int emulate_on_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { @@ -1417,7 +1426,7 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm, [SVM_EXIT_CPUID] = cpuid_interception, [SVM_EXIT_INVD] = emulate_on_interception, [SVM_EXIT_HLT] = halt_interception, - [SVM_EXIT_INVLPG] = emulate_on_interception, + [SVM_EXIT_INVLPG] = invlpg_interception, [SVM_EXIT_INVLPGA] = invalid_op_interception, [SVM_EXIT_IOIO] = io_interception, [SVM_EXIT_MSR] = msr_interception, diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 025bf4011ab..4556cc3715b 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -1130,7 +1130,8 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) CPU_BASED_CR3_STORE_EXITING | CPU_BASED_USE_IO_BITMAPS | CPU_BASED_MOV_DR_EXITING | - CPU_BASED_USE_TSC_OFFSETING; + CPU_BASED_USE_TSC_OFFSETING | + CPU_BASED_INVLPG_EXITING; opt = CPU_BASED_TPR_SHADOW | CPU_BASED_USE_MSR_BITMAPS | CPU_BASED_ACTIVATE_SECONDARY_CONTROLS; @@ -1159,9 +1160,11 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) _cpu_based_exec_control &= ~CPU_BASED_TPR_SHADOW; #endif if (_cpu_based_2nd_exec_control & SECONDARY_EXEC_ENABLE_EPT) { - /* CR3 accesses don't need to cause VM Exits when EPT enabled */ + /* CR3 accesses and invlpg don't need to cause VM Exits when EPT + enabled */ min &= ~(CPU_BASED_CR3_LOAD_EXITING | - CPU_BASED_CR3_STORE_EXITING); + CPU_BASED_CR3_STORE_EXITING | + CPU_BASED_INVLPG_EXITING); if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PROCBASED_CTLS, &_cpu_based_exec_control) < 0) return -EIO; @@ -2790,6 +2793,15 @@ static int handle_vmcall(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return 1; } +static int handle_invlpg(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u64 exit_qualification = vmcs_read64(EXIT_QUALIFICATION); + + kvm_mmu_invlpg(vcpu, exit_qualification); + skip_emulated_instruction(vcpu); + return 1; +} + static int handle_wbinvd(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { skip_emulated_instruction(vcpu); @@ -2958,6 +2970,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu, [EXIT_REASON_MSR_WRITE] = handle_wrmsr, [EXIT_REASON_PENDING_INTERRUPT] = handle_interrupt_window, [EXIT_REASON_HLT] = handle_halt, + [EXIT_REASON_INVLPG] = handle_invlpg, [EXIT_REASON_VMCALL] = handle_vmcall, [EXIT_REASON_TPR_BELOW_THRESHOLD] = handle_tpr_below_threshold, [EXIT_REASON_APIC_ACCESS] = handle_apic_access, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 88e6d9abbd2..efee85ba07e 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2341,6 +2341,7 @@ static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg) int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address) { + kvm_mmu_invlpg(vcpu, address); return X86EMUL_CONTINUE; } diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 475d8ab83bf..8b935cc4c14 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -222,6 +222,7 @@ struct kvm_mmu { struct kvm_mmu_page *page); int (*sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp); + void (*invlpg)(struct kvm_vcpu *vcpu, gva_t gva); hpa_t root_hpa; int root_level; int shadow_root_level; @@ -591,6 +592,7 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu); int kvm_fix_hypercall(struct kvm_vcpu *vcpu); int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code); +void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva); void kvm_enable_tdp(void); void kvm_disable_tdp(void); -- cgit v1.2.3-70-g09d2 From 4731d4c7a07769cf2926c327177b97bb8c68cafc Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 23 Sep 2008 13:18:39 -0300 Subject: KVM: MMU: out of sync shadow core Allow guest pagetables to go out of sync. Instead of emulating write accesses to guest pagetables, or unshadowing them, we un-write-protect the page table and allow the guest to modify it at will. We rely on invlpg executions to synchronize individual ptes, and will synchronize the entire pagetable on tlb flushes. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 210 +++++++++++++++++++++++++++++++++++++++++---- arch/x86/kvm/paging_tmpl.h | 2 +- arch/x86/kvm/x86.c | 3 + include/asm-x86/kvm_host.h | 3 + include/linux/kvm_host.h | 1 + 5 files changed, 201 insertions(+), 18 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 57c7580e7f9..d88659ae777 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -147,6 +147,10 @@ struct kvm_shadow_walk { u64 addr, u64 *spte, int level); }; +struct kvm_unsync_walk { + int (*entry) (struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk); +}; + typedef int (*mmu_parent_walk_fn) (struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp); static struct kmem_cache *pte_chain_cache; @@ -654,8 +658,6 @@ static void rmap_write_protect(struct kvm *kvm, u64 gfn) if (write_protected) kvm_flush_remote_tlbs(kvm); - - account_shadowed(kvm, gfn); } static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp) @@ -908,6 +910,41 @@ static void nonpaging_invlpg(struct kvm_vcpu *vcpu, gva_t gva) { } +static int mmu_unsync_walk(struct kvm_mmu_page *sp, + struct kvm_unsync_walk *walker) +{ + int i, ret; + + if (!sp->unsync_children) + return 0; + + for (i = 0; i < PT64_ENT_PER_PAGE; ++i) { + u64 ent = sp->spt[i]; + + if (is_shadow_present_pte(ent)) { + struct kvm_mmu_page *child; + child = page_header(ent & PT64_BASE_ADDR_MASK); + + if (child->unsync_children) { + ret = mmu_unsync_walk(child, walker); + if (ret) + return ret; + } + + if (child->unsync) { + ret = walker->entry(child, walker); + if (ret) + return ret; + } + } + } + + if (i == PT64_ENT_PER_PAGE) + sp->unsync_children = 0; + + return 0; +} + static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) { unsigned index; @@ -928,6 +965,59 @@ static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) return NULL; } +static void kvm_unlink_unsync_page(struct kvm *kvm, struct kvm_mmu_page *sp) +{ + WARN_ON(!sp->unsync); + sp->unsync = 0; + --kvm->stat.mmu_unsync; +} + +static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp); + +static int kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + if (sp->role.glevels != vcpu->arch.mmu.root_level) { + kvm_mmu_zap_page(vcpu->kvm, sp); + return 1; + } + + rmap_write_protect(vcpu->kvm, sp->gfn); + if (vcpu->arch.mmu.sync_page(vcpu, sp)) { + kvm_mmu_zap_page(vcpu->kvm, sp); + return 1; + } + + kvm_mmu_flush_tlb(vcpu); + kvm_unlink_unsync_page(vcpu->kvm, sp); + return 0; +} + +struct sync_walker { + struct kvm_vcpu *vcpu; + struct kvm_unsync_walk walker; +}; + +static int mmu_sync_fn(struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk) +{ + struct sync_walker *sync_walk = container_of(walk, struct sync_walker, + walker); + struct kvm_vcpu *vcpu = sync_walk->vcpu; + + kvm_sync_page(vcpu, sp); + return (need_resched() || spin_needbreak(&vcpu->kvm->mmu_lock)); +} + +static void mmu_sync_children(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + struct sync_walker walker = { + .walker = { .entry = mmu_sync_fn, }, + .vcpu = vcpu, + }; + + while (mmu_unsync_walk(sp, &walker.walker)) + cond_resched_lock(&vcpu->kvm->mmu_lock); +} + static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gaddr, @@ -941,7 +1031,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, unsigned quadrant; struct hlist_head *bucket; struct kvm_mmu_page *sp; - struct hlist_node *node; + struct hlist_node *node, *tmp; role.word = 0; role.glevels = vcpu->arch.mmu.root_level; @@ -957,8 +1047,18 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, gfn, role.word); index = kvm_page_table_hashfn(gfn); bucket = &vcpu->kvm->arch.mmu_page_hash[index]; - hlist_for_each_entry(sp, node, bucket, hash_link) - if (sp->gfn == gfn && sp->role.word == role.word) { + hlist_for_each_entry_safe(sp, node, tmp, bucket, hash_link) + if (sp->gfn == gfn) { + if (sp->unsync) + if (kvm_sync_page(vcpu, sp)) + continue; + + if (sp->role.word != role.word) + continue; + + if (sp->unsync_children) + set_bit(KVM_REQ_MMU_SYNC, &vcpu->requests); + mmu_page_add_parent_pte(vcpu, sp, parent_pte); pgprintk("%s: found\n", __func__); return sp; @@ -971,8 +1071,10 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, sp->gfn = gfn; sp->role = role; hlist_add_head(&sp->hash_link, bucket); - if (!metaphysical) + if (!metaphysical) { rmap_write_protect(vcpu->kvm, gfn); + account_shadowed(vcpu->kvm, gfn); + } if (shadow_trap_nonpresent_pte != shadow_notrap_nonpresent_pte) vcpu->arch.mmu.prefetch_page(vcpu, sp); else @@ -1078,14 +1180,47 @@ static void kvm_mmu_unlink_parents(struct kvm *kvm, struct kvm_mmu_page *sp) } } +struct zap_walker { + struct kvm_unsync_walk walker; + struct kvm *kvm; + int zapped; +}; + +static int mmu_zap_fn(struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk) +{ + struct zap_walker *zap_walk = container_of(walk, struct zap_walker, + walker); + kvm_mmu_zap_page(zap_walk->kvm, sp); + zap_walk->zapped = 1; + return 0; +} + +static int mmu_zap_unsync_children(struct kvm *kvm, struct kvm_mmu_page *sp) +{ + struct zap_walker walker = { + .walker = { .entry = mmu_zap_fn, }, + .kvm = kvm, + .zapped = 0, + }; + + if (sp->role.level == PT_PAGE_TABLE_LEVEL) + return 0; + mmu_unsync_walk(sp, &walker.walker); + return walker.zapped; +} + static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) { + int ret; ++kvm->stat.mmu_shadow_zapped; + ret = mmu_zap_unsync_children(kvm, sp); kvm_mmu_page_unlink_children(kvm, sp); kvm_mmu_unlink_parents(kvm, sp); kvm_flush_remote_tlbs(kvm); if (!sp->role.invalid && !sp->role.metaphysical) unaccount_shadowed(kvm, sp->gfn); + if (sp->unsync) + kvm_unlink_unsync_page(kvm, sp); if (!sp->root_count) { hlist_del(&sp->hash_link); kvm_mmu_free_page(kvm, sp); @@ -1095,7 +1230,7 @@ static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) kvm_reload_remote_mmus(kvm); } kvm_mmu_reset_last_pte_updated(kvm); - return 0; + return ret; } /* @@ -1201,10 +1336,58 @@ struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva) return page; } +static int unsync_walk_fn(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + sp->unsync_children = 1; + return 1; +} + +static int kvm_unsync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + unsigned index; + struct hlist_head *bucket; + struct kvm_mmu_page *s; + struct hlist_node *node, *n; + + index = kvm_page_table_hashfn(sp->gfn); + bucket = &vcpu->kvm->arch.mmu_page_hash[index]; + /* don't unsync if pagetable is shadowed with multiple roles */ + hlist_for_each_entry_safe(s, node, n, bucket, hash_link) { + if (s->gfn != sp->gfn || s->role.metaphysical) + continue; + if (s->role.word != sp->role.word) + return 1; + } + mmu_parent_walk(vcpu, sp, unsync_walk_fn); + ++vcpu->kvm->stat.mmu_unsync; + sp->unsync = 1; + mmu_convert_notrap(sp); + return 0; +} + +static int mmu_need_write_protect(struct kvm_vcpu *vcpu, gfn_t gfn, + bool can_unsync) +{ + struct kvm_mmu_page *shadow; + + shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn); + if (shadow) { + if (shadow->role.level != PT_PAGE_TABLE_LEVEL) + return 1; + if (shadow->unsync) + return 0; + if (can_unsync) + return kvm_unsync_page(vcpu, shadow); + return 1; + } + return 0; +} + static int set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, unsigned pte_access, int user_fault, int write_fault, int dirty, int largepage, - gfn_t gfn, pfn_t pfn, bool speculative) + gfn_t gfn, pfn_t pfn, bool speculative, + bool can_unsync) { u64 spte; int ret = 0; @@ -1231,7 +1414,6 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, if ((pte_access & ACC_WRITE_MASK) || (write_fault && !is_write_protection(vcpu) && !user_fault)) { - struct kvm_mmu_page *shadow; if (largepage && has_wrprotected_page(vcpu->kvm, gfn)) { ret = 1; @@ -1241,8 +1423,7 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, spte |= PT_WRITABLE_MASK; - shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn); - if (shadow) { + if (mmu_need_write_protect(vcpu, gfn, can_unsync)) { pgprintk("%s: found shadow page for %lx, marking ro\n", __func__, gfn); ret = 1; @@ -1260,7 +1441,6 @@ set_pte: return ret; } - static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, unsigned pt_access, unsigned pte_access, int user_fault, int write_fault, int dirty, @@ -1298,7 +1478,7 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, } } if (set_spte(vcpu, shadow_pte, pte_access, user_fault, write_fault, - dirty, largepage, gfn, pfn, speculative)) { + dirty, largepage, gfn, pfn, speculative, true)) { if (write_fault) *ptwrite = 1; kvm_x86_ops->tlb_flush(vcpu); @@ -1518,10 +1698,6 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu) vcpu->arch.mmu.root_hpa = __pa(vcpu->arch.mmu.pae_root); } -static void mmu_sync_children(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) -{ -} - static void mmu_sync_roots(struct kvm_vcpu *vcpu) { int i; diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index dc169e8148b..613ec9aa674 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -580,7 +580,7 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) pte_access = sp->role.access & FNAME(gpte_access)(vcpu, gpte); set_spte(vcpu, &sp->spt[i], pte_access, 0, 0, is_dirty_pte(gpte), 0, gfn, - spte_to_pfn(sp->spt[i]), true); + spte_to_pfn(sp->spt[i]), true, false); } return !nr_present; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index efee85ba07e..1c5864ac083 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -101,6 +101,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "mmu_flooded", VM_STAT(mmu_flooded) }, { "mmu_recycled", VM_STAT(mmu_recycled) }, { "mmu_cache_miss", VM_STAT(mmu_cache_miss) }, + { "mmu_unsync", VM_STAT(mmu_unsync) }, { "remote_tlb_flush", VM_STAT(remote_tlb_flush) }, { "largepages", VM_STAT(lpages) }, { NULL } @@ -3120,6 +3121,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) if (vcpu->requests) { if (test_and_clear_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests)) __kvm_migrate_timers(vcpu); + if (test_and_clear_bit(KVM_REQ_MMU_SYNC, &vcpu->requests)) + kvm_mmu_sync_roots(vcpu); if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests)) kvm_x86_ops->tlb_flush(vcpu); if (test_and_clear_bit(KVM_REQ_REPORT_TPR_ACCESS, diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 8b935cc4c14..7d36fcc0281 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -195,6 +195,8 @@ struct kvm_mmu_page { */ int multimapped; /* More than one parent_pte? */ int root_count; /* Currently serving as active root */ + bool unsync; + bool unsync_children; union { u64 *parent_pte; /* !multimapped */ struct hlist_head parent_ptes; /* multimapped, kvm_pte_chain */ @@ -371,6 +373,7 @@ struct kvm_vm_stat { u32 mmu_flooded; u32 mmu_recycled; u32 mmu_cache_miss; + u32 mmu_unsync; u32 remote_tlb_flush; u32 lpages; }; diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 6252802c3cc..73b7c52b949 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -35,6 +35,7 @@ #define KVM_REQ_TRIPLE_FAULT 4 #define KVM_REQ_PENDING_TIMER 5 #define KVM_REQ_UNHALT 6 +#define KVM_REQ_MMU_SYNC 7 struct kvm_vcpu; extern struct kmem_cache *kvm_vcpu_cache; -- cgit v1.2.3-70-g09d2 From 0074ff63ebc195701062ca46e0d82fcea0fa3a0a Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 23 Sep 2008 13:18:40 -0300 Subject: KVM: MMU: speed up mmu_unsync_walk Cache the unsynced children information in a per-page bitmap. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 72 ++++++++++++++++++++++++++++++++++++++-------- include/asm-x86/kvm_host.h | 1 + 2 files changed, 61 insertions(+), 12 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index d88659ae777..cb391d629af 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -891,6 +891,52 @@ static void mmu_parent_walk(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, } } +static void kvm_mmu_update_unsync_bitmap(u64 *spte) +{ + unsigned int index; + struct kvm_mmu_page *sp = page_header(__pa(spte)); + + index = spte - sp->spt; + __set_bit(index, sp->unsync_child_bitmap); + sp->unsync_children = 1; +} + +static void kvm_mmu_update_parents_unsync(struct kvm_mmu_page *sp) +{ + struct kvm_pte_chain *pte_chain; + struct hlist_node *node; + int i; + + if (!sp->parent_pte) + return; + + if (!sp->multimapped) { + kvm_mmu_update_unsync_bitmap(sp->parent_pte); + return; + } + + hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link) + for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i) { + if (!pte_chain->parent_ptes[i]) + break; + kvm_mmu_update_unsync_bitmap(pte_chain->parent_ptes[i]); + } +} + +static int unsync_walk_fn(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + sp->unsync_children = 1; + kvm_mmu_update_parents_unsync(sp); + return 1; +} + +static void kvm_mmu_mark_parents_unsync(struct kvm_vcpu *vcpu, + struct kvm_mmu_page *sp) +{ + mmu_parent_walk(vcpu, sp, unsync_walk_fn); + kvm_mmu_update_parents_unsync(sp); +} + static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) { @@ -910,6 +956,11 @@ static void nonpaging_invlpg(struct kvm_vcpu *vcpu, gva_t gva) { } +#define for_each_unsync_children(bitmap, idx) \ + for (idx = find_first_bit(bitmap, 512); \ + idx < 512; \ + idx = find_next_bit(bitmap, 512, idx+1)) + static int mmu_unsync_walk(struct kvm_mmu_page *sp, struct kvm_unsync_walk *walker) { @@ -918,7 +969,7 @@ static int mmu_unsync_walk(struct kvm_mmu_page *sp, if (!sp->unsync_children) return 0; - for (i = 0; i < PT64_ENT_PER_PAGE; ++i) { + for_each_unsync_children(sp->unsync_child_bitmap, i) { u64 ent = sp->spt[i]; if (is_shadow_present_pte(ent)) { @@ -929,17 +980,19 @@ static int mmu_unsync_walk(struct kvm_mmu_page *sp, ret = mmu_unsync_walk(child, walker); if (ret) return ret; + __clear_bit(i, sp->unsync_child_bitmap); } if (child->unsync) { ret = walker->entry(child, walker); + __clear_bit(i, sp->unsync_child_bitmap); if (ret) return ret; } } } - if (i == PT64_ENT_PER_PAGE) + if (find_first_bit(sp->unsync_child_bitmap, 512) == 512) sp->unsync_children = 0; return 0; @@ -1056,10 +1109,11 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, if (sp->role.word != role.word) continue; - if (sp->unsync_children) - set_bit(KVM_REQ_MMU_SYNC, &vcpu->requests); - mmu_page_add_parent_pte(vcpu, sp, parent_pte); + if (sp->unsync_children) { + set_bit(KVM_REQ_MMU_SYNC, &vcpu->requests); + kvm_mmu_mark_parents_unsync(vcpu, sp); + } pgprintk("%s: found\n", __func__); return sp; } @@ -1336,12 +1390,6 @@ struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva) return page; } -static int unsync_walk_fn(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) -{ - sp->unsync_children = 1; - return 1; -} - static int kvm_unsync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) { unsigned index; @@ -1358,7 +1406,7 @@ static int kvm_unsync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) if (s->role.word != sp->role.word) return 1; } - mmu_parent_walk(vcpu, sp, unsync_walk_fn); + kvm_mmu_mark_parents_unsync(vcpu, sp); ++vcpu->kvm->stat.mmu_unsync; sp->unsync = 1; mmu_convert_notrap(sp); diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 7d36fcc0281..0992d721c5f 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -201,6 +201,7 @@ struct kvm_mmu_page { u64 *parent_pte; /* !multimapped */ struct hlist_head parent_ptes; /* multimapped, kvm_pte_chain */ }; + DECLARE_BITMAP(unsync_child_bitmap, 512); }; struct kvm_pv_mmu_op_buffer { -- cgit v1.2.3-70-g09d2 From e5fcfc821a467bd0827635db8fd39ab1213987e5 Mon Sep 17 00:00:00 2001 From: Weidong Han Date: Thu, 25 Sep 2008 23:32:02 +0800 Subject: KVM: Device Assignment: Map mmio pages into VT-d page table Assigned device could DMA to mmio pages, so also need to map mmio pages into VT-d page table. Signed-off-by: Weidong Han Signed-off-by: Avi Kivity --- arch/x86/kvm/vtd.c | 29 +++++++++++------------------ include/asm-x86/kvm_host.h | 2 -- virt/kvm/kvm_main.c | 2 +- 3 files changed, 12 insertions(+), 21 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kvm/vtd.c b/arch/x86/kvm/vtd.c index 667bf3fb64b..a770874f3a3 100644 --- a/arch/x86/kvm/vtd.c +++ b/arch/x86/kvm/vtd.c @@ -36,37 +36,30 @@ int kvm_iommu_map_pages(struct kvm *kvm, { gfn_t gfn = base_gfn; pfn_t pfn; - int i, r; + int i, r = 0; struct dmar_domain *domain = kvm->arch.intel_iommu_domain; /* check if iommu exists and in use */ if (!domain) return 0; - r = -EINVAL; for (i = 0; i < npages; i++) { /* check if already mapped */ pfn = (pfn_t)intel_iommu_iova_to_pfn(domain, gfn_to_gpa(gfn)); - if (pfn && !is_mmio_pfn(pfn)) + if (pfn) continue; pfn = gfn_to_pfn(kvm, gfn); - if (!is_mmio_pfn(pfn)) { - r = intel_iommu_page_mapping(domain, - gfn_to_gpa(gfn), - pfn_to_hpa(pfn), - PAGE_SIZE, - DMA_PTE_READ | - DMA_PTE_WRITE); - if (r) { - printk(KERN_DEBUG "kvm_iommu_map_pages:" - "iommu failed to map pfn=%lx\n", pfn); - goto unmap_pages; - } - } else { - printk(KERN_DEBUG "kvm_iommu_map_page:" - "invalid pfn=%lx\n", pfn); + r = intel_iommu_page_mapping(domain, + gfn_to_gpa(gfn), + pfn_to_hpa(pfn), + PAGE_SIZE, + DMA_PTE_READ | + DMA_PTE_WRITE); + if (r) { + printk(KERN_ERR "kvm_iommu_map_pages:" + "iommu failed to map pfn=%lx\n", pfn); goto unmap_pages; } gfn++; diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 0992d721c5f..ca6bbc0bd97 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -502,8 +502,6 @@ int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa, int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes, gpa_t addr, unsigned long *ret); -int is_mmio_pfn(pfn_t pfn); - extern bool tdp_enabled; enum emulation_result { diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 6cf042789ad..98cd916448a 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -76,7 +76,7 @@ static inline int valid_vcpu(int n) return likely(n >= 0 && n < KVM_MAX_VCPUS); } -inline int is_mmio_pfn(pfn_t pfn) +static inline int is_mmio_pfn(pfn_t pfn) { if (pfn_valid(pfn)) return PageReserved(pfn_to_page(pfn)); -- cgit v1.2.3-70-g09d2 From 3de42dc094ecd313dc7d551e007a134b52f8663d Mon Sep 17 00:00:00 2001 From: Xiantao Zhang Date: Mon, 6 Oct 2008 13:48:45 +0800 Subject: KVM: Separate irq ack notification out of arch/x86/kvm/irq.c Moving irq ack notification logic as common, and make it shared with ia64 side. Signed-off-by: Xiantao Zhang Signed-off-by: Avi Kivity --- arch/ia64/include/asm/kvm_host.h | 4 +++ arch/ia64/kvm/Makefile | 2 +- arch/ia64/kvm/irq.h | 5 ---- arch/x86/kvm/Makefile | 2 +- arch/x86/kvm/irq.c | 33 ---------------------- arch/x86/kvm/irq.h | 8 ------ include/asm-x86/kvm_host.h | 2 ++ include/linux/kvm_host.h | 6 ++++ virt/kvm/irq_comm.c | 60 ++++++++++++++++++++++++++++++++++++++++ 9 files changed, 74 insertions(+), 48 deletions(-) create mode 100644 virt/kvm/irq_comm.c (limited to 'include/asm-x86') diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h index 1efe513a994..da579a33db1 100644 --- a/arch/ia64/include/asm/kvm_host.h +++ b/arch/ia64/include/asm/kvm_host.h @@ -413,6 +413,10 @@ struct kvm_arch { struct kvm_ioapic *vioapic; struct kvm_vm_stat stat; struct kvm_sal_data rdv_sal_data; + + struct list_head assigned_dev_head; + struct dmar_domain *intel_iommu_domain; + struct hlist_head irq_ack_notifier_list; }; union cpuid3_t { diff --git a/arch/ia64/kvm/Makefile b/arch/ia64/kvm/Makefile index bf22fb9e6dc..3b1a1c156ef 100644 --- a/arch/ia64/kvm/Makefile +++ b/arch/ia64/kvm/Makefile @@ -44,7 +44,7 @@ EXTRA_CFLAGS += -Ivirt/kvm -Iarch/ia64/kvm/ EXTRA_AFLAGS += -Ivirt/kvm -Iarch/ia64/kvm/ common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \ - coalesced_mmio.o) + coalesced_mmio.o irq_comm.o) kvm-objs := $(common-objs) kvm-ia64.o kvm_fw.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/ia64/kvm/irq.h b/arch/ia64/kvm/irq.h index f2e6545debf..604329ac3c9 100644 --- a/arch/ia64/kvm/irq.h +++ b/arch/ia64/kvm/irq.h @@ -23,10 +23,5 @@ #ifndef __IRQ_H #define __IRQ_H -struct kvm; - -static inline void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi) -{ -} #endif diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile index 7dce593b9c6..c02343594b4 100644 --- a/arch/x86/kvm/Makefile +++ b/arch/x86/kvm/Makefile @@ -3,7 +3,7 @@ # common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \ - coalesced_mmio.o) + coalesced_mmio.o irq_comm.o) ifeq ($(CONFIG_KVM_TRACE),y) common-objs += $(addprefix ../../../virt/kvm/, kvm_trace.o) endif diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c index 8c1b9c5def7..c019b8edcdb 100644 --- a/arch/x86/kvm/irq.c +++ b/arch/x86/kvm/irq.c @@ -99,36 +99,3 @@ void __kvm_migrate_timers(struct kvm_vcpu *vcpu) __kvm_migrate_apic_timer(vcpu); __kvm_migrate_pit_timer(vcpu); } - -/* This should be called with the kvm->lock mutex held */ -void kvm_set_irq(struct kvm *kvm, int irq, int level) -{ - /* Not possible to detect if the guest uses the PIC or the - * IOAPIC. So set the bit in both. The guest will ignore - * writes to the unused one. - */ - kvm_ioapic_set_irq(kvm->arch.vioapic, irq, level); - kvm_pic_set_irq(pic_irqchip(kvm), irq, level); -} - -void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi) -{ - struct kvm_irq_ack_notifier *kian; - struct hlist_node *n; - - hlist_for_each_entry(kian, n, &kvm->arch.irq_ack_notifier_list, link) - if (kian->gsi == gsi) - kian->irq_acked(kian); -} - -void kvm_register_irq_ack_notifier(struct kvm *kvm, - struct kvm_irq_ack_notifier *kian) -{ - hlist_add_head(&kian->link, &kvm->arch.irq_ack_notifier_list); -} - -void kvm_unregister_irq_ack_notifier(struct kvm *kvm, - struct kvm_irq_ack_notifier *kian) -{ - hlist_del(&kian->link); -} diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index 4748532fd1d..f17c8f5bbf3 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -68,7 +68,6 @@ struct kvm_pic { }; struct kvm_pic *kvm_create_pic(struct kvm *kvm); -void kvm_pic_set_irq(void *opaque, int irq, int level); int kvm_pic_read_irq(struct kvm *kvm); void kvm_pic_update_irq(struct kvm_pic *s); void kvm_pic_clear_isr_ack(struct kvm *kvm); @@ -85,13 +84,6 @@ static inline int irqchip_in_kernel(struct kvm *kvm) void kvm_pic_reset(struct kvm_kpic_state *s); -void kvm_set_irq(struct kvm *kvm, int irq, int level); -void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi); -void kvm_register_irq_ack_notifier(struct kvm *kvm, - struct kvm_irq_ack_notifier *kian); -void kvm_unregister_irq_ack_notifier(struct kvm *kvm, - struct kvm_irq_ack_notifier *kian); - void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec); void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu); void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu); diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index ca6bbc0bd97..411fb8cfb24 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -565,6 +565,8 @@ void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code); void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long cr2, u32 error_code); +void kvm_pic_set_irq(void *opaque, int irq, int level); + void kvm_inject_nmi(struct kvm_vcpu *vcpu); void fx_init(struct kvm_vcpu *vcpu); diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index b3b7598b4d9..3833c48fae3 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -309,6 +309,12 @@ struct kvm_assigned_dev_kernel { struct pci_dev *dev; struct kvm *kvm; }; +void kvm_set_irq(struct kvm *kvm, int irq, int level); +void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi); +void kvm_register_irq_ack_notifier(struct kvm *kvm, + struct kvm_irq_ack_notifier *kian); +void kvm_unregister_irq_ack_notifier(struct kvm *kvm, + struct kvm_irq_ack_notifier *kian); #ifdef CONFIG_DMAR int kvm_iommu_map_pages(struct kvm *kvm, gfn_t base_gfn, diff --git a/virt/kvm/irq_comm.c b/virt/kvm/irq_comm.c new file mode 100644 index 00000000000..d0169f5e604 --- /dev/null +++ b/virt/kvm/irq_comm.c @@ -0,0 +1,60 @@ +/* + * irq_comm.c: Common API for in kernel interrupt controller + * Copyright (c) 2007, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple + * Place - Suite 330, Boston, MA 02111-1307 USA. + * Authors: + * Yaozu (Eddie) Dong + * + */ + +#include +#include "irq.h" + +#include "ioapic.h" + +/* This should be called with the kvm->lock mutex held */ +void kvm_set_irq(struct kvm *kvm, int irq, int level) +{ + /* Not possible to detect if the guest uses the PIC or the + * IOAPIC. So set the bit in both. The guest will ignore + * writes to the unused one. + */ + kvm_ioapic_set_irq(kvm->arch.vioapic, irq, level); +#ifdef CONFIG_X86 + kvm_pic_set_irq(pic_irqchip(kvm), irq, level); +#endif +} + +void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi) +{ + struct kvm_irq_ack_notifier *kian; + struct hlist_node *n; + + hlist_for_each_entry(kian, n, &kvm->arch.irq_ack_notifier_list, link) + if (kian->gsi == gsi) + kian->irq_acked(kian); +} + +void kvm_register_irq_ack_notifier(struct kvm *kvm, + struct kvm_irq_ack_notifier *kian) +{ + hlist_add_head(&kian->link, &kvm->arch.irq_ack_notifier_list); +} + +void kvm_unregister_irq_ack_notifier(struct kvm *kvm, + struct kvm_irq_ack_notifier *kian) +{ + hlist_del(&kian->link); +} -- cgit v1.2.3-70-g09d2 From f61f1b5717dec9968efd6c8d61bd73b7a61b72a0 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Wed, 15 Oct 2008 07:46:27 -0700 Subject: x86: fix compat-vdso Jeff Chua reported that this patch: > -#define PTE_IDENT_ATTR 0x007 /* PRESENT+RW+USER */ > -#define PDE_IDENT_ATTR 0x067 /* PRESENT+RW+USER+DIRTY+ACCESSED */ > +#define PTE_IDENT_ATTR 0x003 /* PRESENT+RW */ > +#define PDE_IDENT_ATTR 0x063 /* PRESENT+RW+DIRTY+ACCESSED */ broke kernels with CONFIG_COMPAT_VDSO set with this init segfault: init[1]: segfault at ffffe01c up b7f0dc28 sp bfc26628 error 5 in ld-2.7.90.so[b7f0b000+1c000] Include USER bit in the PDE_IDENT_ATTR only, as the protection bits are combined from the PDE and PTE entries. This will allow the high mapped VDSO page in the case of CONFIG_COMPAT_VDSO to be user readable. Reported-by: Jeff Chua Signed-off-by: Suresh Siddha Tested-by: Jeff Chua Signed-off-by: Ingo Molnar --- include/asm-x86/pgtable.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index ed932453ef2..e9f10af50f4 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -138,8 +138,13 @@ #ifdef CONFIG_X86_64 #define __PAGE_KERNEL_IDENT_LARGE_EXEC __PAGE_KERNEL_LARGE_EXEC #else +/* + * For PDE_IDENT_ATTR include USER bit. As the PDE and PTE protection + * bits are combined, this will alow user to access the high address mapped + * VDSO in the presence of CONFIG_COMPAT_VDSO + */ #define PTE_IDENT_ATTR 0x003 /* PRESENT+RW */ -#define PDE_IDENT_ATTR 0x063 /* PRESENT+RW+DIRTY+ACCESSED */ +#define PDE_IDENT_ATTR 0x067 /* PRESENT+RW+USER+DIRTY+ACCESSED */ #define PGD_IDENT_ATTR 0x001 /* PRESENT (no other attributes) */ #endif -- cgit v1.2.3-70-g09d2 From 0b59268285ca6cdc46191f2995bf632088e3e277 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 16 Oct 2008 15:39:57 +0200 Subject: [PATCH] remove unused ibcs2/PER_SVR4 in SET_PERSONALITY The SET_PERSONALITY macro is always called with a second argument of 0. Remove the ibcs argument and the various tests to set the PER_SVR4 personality. Signed-off-by: Martin Schwidefsky --- arch/alpha/include/asm/elf.h | 4 ++-- arch/arm/include/asm/elf.h | 2 +- arch/avr32/include/asm/elf.h | 2 +- arch/blackfin/include/asm/elf.h | 2 +- arch/h8300/include/asm/elf.h | 2 +- arch/ia64/ia32/binfmt_elf32.c | 2 +- arch/ia64/ia32/ia32priv.h | 4 ++-- arch/ia64/include/asm/elf.h | 2 +- arch/m68knommu/include/asm/elf.h | 2 +- arch/mips/include/asm/elf.h | 10 +++------- arch/parisc/kernel/binfmt_elf32.c | 2 +- arch/powerpc/include/asm/elf.h | 4 ++-- arch/s390/include/asm/elf.h | 8 +++----- arch/sh/include/asm/elf.h | 2 +- arch/sparc/include/asm/elf_32.h | 2 +- arch/sparc/include/asm/elf_64.h | 6 ++---- fs/binfmt_elf.c | 6 +++--- include/asm-cris/elf.h | 2 +- include/asm-frv/elf.h | 2 +- include/asm-m32r/elf.h | 2 +- include/asm-m68k/elf.h | 2 +- include/asm-mn10300/elf.h | 2 +- include/asm-parisc/elf.h | 2 +- include/asm-um/elf-i386.h | 2 +- include/asm-um/elf-ppc.h | 2 +- include/asm-um/elf-x86_64.h | 2 +- include/asm-x86/elf.h | 4 ++-- include/asm-xtensa/elf.h | 2 +- 28 files changed, 39 insertions(+), 47 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/alpha/include/asm/elf.h b/arch/alpha/include/asm/elf.h index fc1002ea1e0..5c75c1b2352 100644 --- a/arch/alpha/include/asm/elf.h +++ b/arch/alpha/include/asm/elf.h @@ -144,9 +144,9 @@ extern int dump_elf_task_fp(elf_fpreg_t *dest, struct task_struct *task); : amask (AMASK_CIX) ? "ev6" : "ev67"); \ }) -#define SET_PERSONALITY(EX, IBCS2) \ +#define SET_PERSONALITY(EX) \ set_personality(((EX).e_flags & EF_ALPHA_32BIT) \ - ? PER_LINUX_32BIT : (IBCS2) ? PER_SVR4 : PER_LINUX) + ? PER_LINUX_32BIT : PER_LINUX) extern int alpha_l1i_cacheshape; extern int alpha_l1d_cacheshape; diff --git a/arch/arm/include/asm/elf.h b/arch/arm/include/asm/elf.h index 5be016980c1..a58378c343b 100644 --- a/arch/arm/include/asm/elf.h +++ b/arch/arm/include/asm/elf.h @@ -107,6 +107,6 @@ extern int arm_elf_read_implies_exec(const struct elf32_hdr *, int); #define ELF_PLAT_INIT(_r, load_addr) (_r)->ARM_r0 = 0 extern void elf_set_personality(const struct elf32_hdr *); -#define SET_PERSONALITY(ex, ibcs2) elf_set_personality(&(ex)) +#define SET_PERSONALITY(ex) elf_set_personality(&(ex)) #endif diff --git a/arch/avr32/include/asm/elf.h b/arch/avr32/include/asm/elf.h index 64ce40ee1d5..d5d1d41c600 100644 --- a/arch/avr32/include/asm/elf.h +++ b/arch/avr32/include/asm/elf.h @@ -103,6 +103,6 @@ typedef struct user_fpu_struct elf_fpregset_t; #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX_32BIT) #endif /* __ASM_AVR32_ELF_H */ diff --git a/arch/blackfin/include/asm/elf.h b/arch/blackfin/include/asm/elf.h index 67a03a8a353..cdbfcfc30f6 100644 --- a/arch/blackfin/include/asm/elf.h +++ b/arch/blackfin/include/asm/elf.h @@ -122,6 +122,6 @@ do { \ #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #endif diff --git a/arch/h8300/include/asm/elf.h b/arch/h8300/include/asm/elf.h index a8b57d1f412..94e2284c881 100644 --- a/arch/h8300/include/asm/elf.h +++ b/arch/h8300/include/asm/elf.h @@ -55,7 +55,7 @@ typedef unsigned long elf_fpregset_t; #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #define R_H8_NONE 0 #define R_H8_DIR32 1 diff --git a/arch/ia64/ia32/binfmt_elf32.c b/arch/ia64/ia32/binfmt_elf32.c index 4f0c30c38e9..f92bdaac897 100644 --- a/arch/ia64/ia32/binfmt_elf32.c +++ b/arch/ia64/ia32/binfmt_elf32.c @@ -41,7 +41,7 @@ randomize_stack_top(unsigned long stack_top); #define elf_map elf32_map #undef SET_PERSONALITY -#define SET_PERSONALITY(ex, ibcs2) elf32_set_personality() +#define SET_PERSONALITY(ex) elf32_set_personality() #define elf_read_implies_exec(ex, have_pt_gnu_stack) (!(have_pt_gnu_stack)) diff --git a/arch/ia64/ia32/ia32priv.h b/arch/ia64/ia32/ia32priv.h index dd0c53687a9..0f15349c3c6 100644 --- a/arch/ia64/ia32/ia32priv.h +++ b/arch/ia64/ia32/ia32priv.h @@ -332,8 +332,8 @@ void ia64_elf32_init(struct pt_regs *regs); #define ELF_PLATFORM NULL #ifdef __KERNEL__ -# define SET_PERSONALITY(EX,IBCS2) \ - (current->personality = (IBCS2) ? PER_SVR4 : PER_LINUX) +# define SET_PERSONALITY(EX) \ + (current->personality = PER_LINUX) #endif #define IA32_EFLAG 0x200 diff --git a/arch/ia64/include/asm/elf.h b/arch/ia64/include/asm/elf.h index 2acb6b6543c..86eddee029c 100644 --- a/arch/ia64/include/asm/elf.h +++ b/arch/ia64/include/asm/elf.h @@ -202,7 +202,7 @@ extern void ia64_elf_core_copy_regs (struct pt_regs *src, elf_gregset_t dst); relevant until we have real hardware to play with... */ #define ELF_PLATFORM NULL -#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #define elf_read_implies_exec(ex, executable_stack) \ ((executable_stack!=EXSTACK_DISABLE_X) && ((ex).e_flags & EF_IA_64_LINUX_EXECUTABLE_STACK) != 0) diff --git a/arch/m68knommu/include/asm/elf.h b/arch/m68knommu/include/asm/elf.h index 27f0ec70fba..b8046837f38 100644 --- a/arch/m68knommu/include/asm/elf.h +++ b/arch/m68knommu/include/asm/elf.h @@ -105,6 +105,6 @@ typedef struct user_m68kfp_struct elf_fpregset_t; #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #endif diff --git a/arch/mips/include/asm/elf.h b/arch/mips/include/asm/elf.h index f69f7acba63..a8eac1697b3 100644 --- a/arch/mips/include/asm/elf.h +++ b/arch/mips/include/asm/elf.h @@ -247,10 +247,8 @@ extern struct mips_abi mips_abi_n32; #ifdef CONFIG_32BIT -#define SET_PERSONALITY(ex, ibcs2) \ +#define SET_PERSONALITY(ex) \ do { \ - if (ibcs2) \ - set_personality(PER_SVR4); \ set_personality(PER_LINUX); \ \ current->thread.abi = &mips_abi; \ @@ -296,7 +294,7 @@ do { \ #define __SET_PERSONALITY32(ex) do { } while (0) #endif -#define SET_PERSONALITY(ex, ibcs2) \ +#define SET_PERSONALITY(ex) \ do { \ clear_thread_flag(TIF_32BIT_REGS); \ clear_thread_flag(TIF_32BIT_ADDR); \ @@ -306,9 +304,7 @@ do { \ else \ current->thread.abi = &mips_abi; \ \ - if (ibcs2) \ - set_personality(PER_SVR4); \ - else if (current->personality != PER_LINUX32) \ + if (current->personality != PER_LINUX32) \ set_personality(PER_LINUX); \ } while (0) diff --git a/arch/parisc/kernel/binfmt_elf32.c b/arch/parisc/kernel/binfmt_elf32.c index ecb10a4f63c..f61692d2b55 100644 --- a/arch/parisc/kernel/binfmt_elf32.c +++ b/arch/parisc/kernel/binfmt_elf32.c @@ -85,7 +85,7 @@ struct elf_prpsinfo32 * could set a processor dependent flag in the thread_struct. */ -#define SET_PERSONALITY(ex, ibcs2) \ +#define SET_PERSONALITY(ex) \ set_thread_flag(TIF_32BIT); \ current->thread.map_base = DEFAULT_MAP_BASE32; \ current->thread.task_size = DEFAULT_TASK_SIZE32 \ diff --git a/arch/powerpc/include/asm/elf.h b/arch/powerpc/include/asm/elf.h index 64c6ee22eef..d812929390e 100644 --- a/arch/powerpc/include/asm/elf.h +++ b/arch/powerpc/include/asm/elf.h @@ -232,7 +232,7 @@ typedef elf_vrregset_t elf_fpxregset_t; #endif /* __powerpc64__ */ #ifdef __powerpc64__ -# define SET_PERSONALITY(ex, ibcs2) \ +# define SET_PERSONALITY(ex) \ do { \ unsigned long new_flags = 0; \ if ((ex).e_ident[EI_CLASS] == ELFCLASS32) \ @@ -256,7 +256,7 @@ do { \ # define elf_read_implies_exec(ex, exec_stk) (test_thread_flag(TIF_32BIT) ? \ (exec_stk != EXSTACK_DISABLE_X) : 0) #else -# define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX) +# define SET_PERSONALITY(ex) set_personality(PER_LINUX) #endif /* __powerpc64__ */ extern int dcache_bsize; diff --git a/arch/s390/include/asm/elf.h b/arch/s390/include/asm/elf.h index 3cad5692381..261785ab5b2 100644 --- a/arch/s390/include/asm/elf.h +++ b/arch/s390/include/asm/elf.h @@ -166,13 +166,11 @@ extern char elf_platform[]; #define ELF_PLATFORM (elf_platform) #ifndef __s390x__ -#define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #else /* __s390x__ */ -#define SET_PERSONALITY(ex, ibcs2) \ +#define SET_PERSONALITY(ex) \ do { \ - if (ibcs2) \ - set_personality(PER_SVR4); \ - else if (current->personality != PER_LINUX32) \ + if (current->personality != PER_LINUX32) \ set_personality(PER_LINUX); \ if ((ex).e_ident[EI_CLASS] == ELFCLASS32) \ set_thread_flag(TIF_31BIT); \ diff --git a/arch/sh/include/asm/elf.h b/arch/sh/include/asm/elf.h index f01449a8d37..ee02db110f0 100644 --- a/arch/sh/include/asm/elf.h +++ b/arch/sh/include/asm/elf.h @@ -189,7 +189,7 @@ do { \ } while (0) #endif -#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX_32BIT) struct task_struct; extern int dump_task_regs (struct task_struct *, elf_gregset_t *); extern int dump_task_fpu (struct task_struct *, elf_fpregset_t *); diff --git a/arch/sparc/include/asm/elf_32.h b/arch/sparc/include/asm/elf_32.h index b7ab6054782..381a1b5256d 100644 --- a/arch/sparc/include/asm/elf_32.h +++ b/arch/sparc/include/asm/elf_32.h @@ -137,6 +137,6 @@ typedef struct { #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #endif /* !(__ASMSPARC_ELF_H) */ diff --git a/arch/sparc/include/asm/elf_64.h b/arch/sparc/include/asm/elf_64.h index 0818a1308f4..425c2f9be6d 100644 --- a/arch/sparc/include/asm/elf_64.h +++ b/arch/sparc/include/asm/elf_64.h @@ -195,7 +195,7 @@ static inline unsigned int sparc64_elf_hwcap(void) #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) \ +#define SET_PERSONALITY(ex) \ do { unsigned long new_flags = current_thread_info()->flags; \ new_flags &= _TIF_32BIT; \ if ((ex).e_ident[EI_CLASS] == ELFCLASS32) \ @@ -208,9 +208,7 @@ do { unsigned long new_flags = current_thread_info()->flags; \ else \ clear_thread_flag(TIF_ABI_PENDING); \ /* flush_thread will update pgd cache */ \ - if (ibcs2) \ - set_personality(PER_SVR4); \ - else if (current->personality != PER_LINUX32) \ + if (current->personality != PER_LINUX32) \ set_personality(PER_LINUX); \ } while (0) diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 655ed8d30a8..c76afa26edf 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -683,7 +683,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) * switch really is going to happen - do this in * flush_thread(). - akpm */ - SET_PERSONALITY(loc->elf_ex, 0); + SET_PERSONALITY(loc->elf_ex); interpreter = open_exec(elf_interpreter); retval = PTR_ERR(interpreter); @@ -734,7 +734,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) goto out_free_dentry; } else { /* Executables without an interpreter also need a personality */ - SET_PERSONALITY(loc->elf_ex, 0); + SET_PERSONALITY(loc->elf_ex); } /* Flush all traces of the currently running executable */ @@ -748,7 +748,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) /* Do this immediately, since STACK_TOP as used in setup_arg_pages may depend on the personality. */ - SET_PERSONALITY(loc->elf_ex, 0); + SET_PERSONALITY(loc->elf_ex); if (elf_read_implies_exec(loc->elf_ex, executable_stack)) current->personality |= READ_IMPLIES_EXEC; diff --git a/include/asm-cris/elf.h b/include/asm-cris/elf.h index 001f64ad11e..f0d17fbc81b 100644 --- a/include/asm-cris/elf.h +++ b/include/asm-cris/elf.h @@ -88,6 +88,6 @@ typedef unsigned long elf_fpregset_t; #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #endif diff --git a/include/asm-frv/elf.h b/include/asm-frv/elf.h index 9fb946bb7dc..7279ec07d62 100644 --- a/include/asm-frv/elf.h +++ b/include/asm-frv/elf.h @@ -137,6 +137,6 @@ do { \ #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #endif diff --git a/include/asm-m32r/elf.h b/include/asm-m32r/elf.h index 67bcd77494a..0cc34c94bf2 100644 --- a/include/asm-m32r/elf.h +++ b/include/asm-m32r/elf.h @@ -129,6 +129,6 @@ typedef elf_fpreg_t elf_fpregset_t; intent than poking at uname or /proc/cpuinfo. */ #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #endif /* _ASM_M32R__ELF_H */ diff --git a/include/asm-m68k/elf.h b/include/asm-m68k/elf.h index 14ea42152b9..0b0f49eb876 100644 --- a/include/asm-m68k/elf.h +++ b/include/asm-m68k/elf.h @@ -114,6 +114,6 @@ typedef struct user_m68kfp_struct elf_fpregset_t; #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #endif diff --git a/include/asm-mn10300/elf.h b/include/asm-mn10300/elf.h index 256a70466ca..bf09f8bb392 100644 --- a/include/asm-mn10300/elf.h +++ b/include/asm-mn10300/elf.h @@ -141,7 +141,7 @@ do { \ #define ELF_PLATFORM (NULL) #ifdef __KERNEL__ -#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) #endif #endif /* _ASM_ELF_H */ diff --git a/include/asm-parisc/elf.h b/include/asm-parisc/elf.h index d0a4a826281..7fa675799e6 100644 --- a/include/asm-parisc/elf.h +++ b/include/asm-parisc/elf.h @@ -236,7 +236,7 @@ typedef unsigned long elf_greg_t; #define ELF_PLATFORM ("PARISC\0") -#define SET_PERSONALITY(ex, ibcs2) \ +#define SET_PERSONALITY(ex) \ current->personality = PER_LINUX; \ current->thread.map_base = DEFAULT_MAP_BASE; \ current->thread.task_size = DEFAULT_TASK_SIZE \ diff --git a/include/asm-um/elf-i386.h b/include/asm-um/elf-i386.h index 23d6893e861..d0da9d7c537 100644 --- a/include/asm-um/elf-i386.h +++ b/include/asm-um/elf-i386.h @@ -86,7 +86,7 @@ extern long elf_aux_hwcap; extern char * elf_aux_platform; #define ELF_PLATFORM (elf_aux_platform) -#define SET_PERSONALITY(ex, ibcs2) do { } while (0) +#define SET_PERSONALITY(ex) do { } while (0) extern unsigned long vsyscall_ehdr; extern unsigned long vsyscall_end; diff --git a/include/asm-um/elf-ppc.h b/include/asm-um/elf-ppc.h index d3b90b7ac3e..af9463cd8ce 100644 --- a/include/asm-um/elf-ppc.h +++ b/include/asm-um/elf-ppc.h @@ -5,7 +5,7 @@ extern long elf_aux_hwcap; #define ELF_HWCAP (elf_aux_hwcap) -#define SET_PERSONALITY(ex, ibcs2) do ; while(0) +#define SET_PERSONALITY(ex) do ; while(0) #define ELF_EXEC_PAGESIZE 4096 diff --git a/include/asm-um/elf-x86_64.h b/include/asm-um/elf-x86_64.h index 3b2d5224a7e..6e8a9195e95 100644 --- a/include/asm-um/elf-x86_64.h +++ b/include/asm-um/elf-x86_64.h @@ -114,6 +114,6 @@ extern long elf_aux_hwcap; #define ELF_PLATFORM "x86_64" -#define SET_PERSONALITY(ex, ibcs2) do ; while(0) +#define SET_PERSONALITY(ex) do ; while(0) #endif diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h index 5c4745bec90..26bc15f01e7 100644 --- a/include/asm-x86/elf.h +++ b/include/asm-x86/elf.h @@ -186,7 +186,7 @@ do { \ set_fs(USER_DS); \ } while (0) -#define COMPAT_SET_PERSONALITY(ex, ibcs2) \ +#define COMPAT_SET_PERSONALITY(ex) \ do { \ if (test_thread_flag(TIF_IA32)) \ clear_thread_flag(TIF_ABI_PENDING); \ @@ -267,7 +267,7 @@ extern int force_personality32; For the moment, we have only optimizations for the Intel generations, but that could change... */ -#define SET_PERSONALITY(ex, ibcs2) set_personality_64bit() +#define SET_PERSONALITY(ex) set_personality_64bit() /* * An executable for which elf_read_implies_exec() returns TRUE will diff --git a/include/asm-xtensa/elf.h b/include/asm-xtensa/elf.h index ca6e5101a2c..c3f53e755ca 100644 --- a/include/asm-xtensa/elf.h +++ b/include/asm-xtensa/elf.h @@ -189,7 +189,7 @@ typedef struct { #endif } elf_xtregs_t; -#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT) +#define SET_PERSONALITY(ex) set_personality(PER_LINUX_32BIT) struct task_struct; -- cgit v1.2.3-70-g09d2 From d5c003b4d1690e666dbab02bc8e705947baa848c Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Wed, 15 Oct 2008 22:01:24 -0700 Subject: include: replace __FUNCTION__ with __func__ __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/acpi/acmacros.h | 6 +++--- include/acpi/platform/acgcc.h | 2 +- include/asm-generic/bug.h | 2 +- include/asm-x86/es7000/apic.h | 2 +- include/asm-x86/summit/apic.h | 2 +- include/linux/ext2_fs.h | 2 +- include/linux/ext3_fs.h | 4 ++-- include/linux/ext3_jbd.h | 14 +++++++------- include/linux/jbd.h | 4 ++-- include/linux/jbd2.h | 4 ++-- include/linux/pm.h | 2 +- include/linux/reiserfs_fs.h | 2 +- include/linux/rtmutex.h | 2 +- include/media/saa7146.h | 2 +- include/net/9p/9p.h | 4 ++-- include/net/bluetooth/bluetooth.h | 4 ++-- include/net/ieee80211.h | 2 +- include/net/ip_vs.h | 4 ++-- include/net/irda/irda.h | 2 +- include/net/sctp/sctp.h | 2 +- include/video/cyblafb.h | 2 +- 21 files changed, 35 insertions(+), 35 deletions(-) (limited to 'include/asm-x86') diff --git a/include/acpi/acmacros.h b/include/acpi/acmacros.h index 57ab9e9d759..74a9617776a 100644 --- a/include/acpi/acmacros.h +++ b/include/acpi/acmacros.h @@ -467,7 +467,7 @@ struct acpi_integer_overlay { /* * If ACPI_GET_FUNCTION_NAME was not defined in the compiler-dependent header, * define it now. This is the case where there the compiler does not support - * a __FUNCTION__ macro or equivalent. + * a __func__ macro or equivalent. */ #ifndef ACPI_GET_FUNCTION_NAME #define ACPI_GET_FUNCTION_NAME _acpi_function_name @@ -475,12 +475,12 @@ struct acpi_integer_overlay { * The Name parameter should be the procedure name as a quoted string. * The function name is also used by the function exit macros below. * Note: (const char) is used to be compatible with the debug interfaces - * and macros such as __FUNCTION__. + * and macros such as __func__. */ #define ACPI_FUNCTION_NAME(name) static const char _acpi_function_name[] = #name; #else -/* Compiler supports __FUNCTION__ (or equivalent) -- Ignore this macro */ +/* Compiler supports __func__ (or equivalent) -- Ignore this macro */ #define ACPI_FUNCTION_NAME(name) #endif diff --git a/include/acpi/platform/acgcc.h b/include/acpi/platform/acgcc.h index 8996dba90cd..8e2cdc57b19 100644 --- a/include/acpi/platform/acgcc.h +++ b/include/acpi/platform/acgcc.h @@ -46,7 +46,7 @@ /* Function name is used for debug output. Non-ANSI, compiler-dependent */ -#define ACPI_GET_FUNCTION_NAME __FUNCTION__ +#define ACPI_GET_FUNCTION_NAME __func__ /* * This macro is used to tag functions as "printf-like" because diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h index edc6ba82e09..0f6dabd4b51 100644 --- a/include/asm-generic/bug.h +++ b/include/asm-generic/bug.h @@ -22,7 +22,7 @@ struct bug_entry { #ifndef HAVE_ARCH_BUG #define BUG() do { \ - printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __FUNCTION__); \ + printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __func__); \ panic("BUG!"); \ } while (0) #endif diff --git a/include/asm-x86/es7000/apic.h b/include/asm-x86/es7000/apic.h index bd2c44d1f7a..aae50c2fb30 100644 --- a/include/asm-x86/es7000/apic.h +++ b/include/asm-x86/es7000/apic.h @@ -171,7 +171,7 @@ static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) int new_apicid = cpu_to_logical_apicid(cpu); if (apicid_cluster(apicid) != apicid_cluster(new_apicid)){ - printk ("%s: Not a valid mask!\n",__FUNCTION__); + printk ("%s: Not a valid mask!\n", __func__); #if defined CONFIG_ES7000_CLUSTERED_APIC return 0xFF; #else diff --git a/include/asm-x86/summit/apic.h b/include/asm-x86/summit/apic.h index c5b2e4b1035..394b00bb5e7 100644 --- a/include/asm-x86/summit/apic.h +++ b/include/asm-x86/summit/apic.h @@ -160,7 +160,7 @@ static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask) int new_apicid = cpu_to_logical_apicid(cpu); if (apicid_cluster(apicid) != apicid_cluster(new_apicid)){ - printk ("%s: Not a valid mask!\n",__FUNCTION__); + printk ("%s: Not a valid mask!\n", __func__); return 0xFF; } apicid = apicid | new_apicid; diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h index 2efe7b863cf..78c775a83f7 100644 --- a/include/linux/ext2_fs.h +++ b/include/linux/ext2_fs.h @@ -47,7 +47,7 @@ #ifdef EXT2FS_DEBUG # define ext2_debug(f, a...) { \ printk ("EXT2-fs DEBUG (%s, %d): %s:", \ - __FILE__, __LINE__, __FUNCTION__); \ + __FILE__, __LINE__, __func__); \ printk (f, ## a); \ } #else diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h index 8120fa1bc23..159d9b476cd 100644 --- a/include/linux/ext3_fs.h +++ b/include/linux/ext3_fs.h @@ -43,7 +43,7 @@ #define ext3_debug(f, a...) \ do { \ printk (KERN_DEBUG "EXT3-fs DEBUG (%s, %d): %s:", \ - __FILE__, __LINE__, __FUNCTION__); \ + __FILE__, __LINE__, __func__); \ printk (KERN_DEBUG f, ## a); \ } while (0) #else @@ -871,7 +871,7 @@ extern void ext3_update_dynamic_rev (struct super_block *sb); #define ext3_std_error(sb, errno) \ do { \ if ((errno)) \ - __ext3_std_error((sb), __FUNCTION__, (errno)); \ + __ext3_std_error((sb), __func__, (errno)); \ } while (0) /* diff --git a/include/linux/ext3_jbd.h b/include/linux/ext3_jbd.h index 8c43b13a02f..cf82d519be4 100644 --- a/include/linux/ext3_jbd.h +++ b/include/linux/ext3_jbd.h @@ -137,17 +137,17 @@ int __ext3_journal_dirty_metadata(const char *where, handle_t *handle, struct buffer_head *bh); #define ext3_journal_get_undo_access(handle, bh) \ - __ext3_journal_get_undo_access(__FUNCTION__, (handle), (bh)) + __ext3_journal_get_undo_access(__func__, (handle), (bh)) #define ext3_journal_get_write_access(handle, bh) \ - __ext3_journal_get_write_access(__FUNCTION__, (handle), (bh)) + __ext3_journal_get_write_access(__func__, (handle), (bh)) #define ext3_journal_revoke(handle, blocknr, bh) \ - __ext3_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh)) + __ext3_journal_revoke(__func__, (handle), (blocknr), (bh)) #define ext3_journal_get_create_access(handle, bh) \ - __ext3_journal_get_create_access(__FUNCTION__, (handle), (bh)) + __ext3_journal_get_create_access(__func__, (handle), (bh)) #define ext3_journal_dirty_metadata(handle, bh) \ - __ext3_journal_dirty_metadata(__FUNCTION__, (handle), (bh)) + __ext3_journal_dirty_metadata(__func__, (handle), (bh)) #define ext3_journal_forget(handle, bh) \ - __ext3_journal_forget(__FUNCTION__, (handle), (bh)) + __ext3_journal_forget(__func__, (handle), (bh)) int ext3_journal_dirty_data(handle_t *handle, struct buffer_head *bh); @@ -160,7 +160,7 @@ static inline handle_t *ext3_journal_start(struct inode *inode, int nblocks) } #define ext3_journal_stop(handle) \ - __ext3_journal_stop(__FUNCTION__, (handle)) + __ext3_journal_stop(__func__, (handle)) static inline handle_t *ext3_journal_current_handle(void) { diff --git a/include/linux/jbd.h b/include/linux/jbd.h index 07a9b52a265..7ebbcb1c9ba 100644 --- a/include/linux/jbd.h +++ b/include/linux/jbd.h @@ -61,7 +61,7 @@ extern u8 journal_enable_debug; do { \ if ((n) <= journal_enable_debug) { \ printk (KERN_DEBUG "(%s, %d): %s: ", \ - __FILE__, __LINE__, __FUNCTION__); \ + __FILE__, __LINE__, __func__); \ printk (f, ## a); \ } \ } while (0) @@ -984,7 +984,7 @@ extern int cleanup_journal_tail(journal_t *); #define jbd_ENOSYS() \ do { \ - printk (KERN_ERR "JBD unimplemented function %s\n", __FUNCTION__); \ + printk (KERN_ERR "JBD unimplemented function %s\n", __func__); \ current->state = TASK_UNINTERRUPTIBLE; \ schedule(); \ } while (1) diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index d2e91ea998f..463d6f10b64 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -61,7 +61,7 @@ extern u8 jbd2_journal_enable_debug; do { \ if ((n) <= jbd2_journal_enable_debug) { \ printk (KERN_DEBUG "(%s, %d): %s: ", \ - __FILE__, __LINE__, __FUNCTION__); \ + __FILE__, __LINE__, __func__); \ printk (f, ## a); \ } \ } while (0) @@ -1143,7 +1143,7 @@ extern int jbd2_cleanup_journal_tail(journal_t *); #define jbd_ENOSYS() \ do { \ - printk (KERN_ERR "JBD unimplemented function %s\n", __FUNCTION__); \ + printk (KERN_ERR "JBD unimplemented function %s\n", __func__); \ current->state = TASK_UNINTERRUPTIBLE; \ schedule(); \ } while (1) diff --git a/include/linux/pm.h b/include/linux/pm.h index 4dcce54b6d7..42de4003c4e 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -419,7 +419,7 @@ extern void __suspend_report_result(const char *function, void *fn, int ret); #define suspend_report_result(fn, ret) \ do { \ - __suspend_report_result(__FUNCTION__, fn, ret); \ + __suspend_report_result(__func__, fn, ret); \ } while (0) #else /* !CONFIG_PM_SLEEP */ diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index e9963af16cd..bc5114d35e9 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -87,7 +87,7 @@ void reiserfs_warning(struct super_block *s, const char *fmt, ...); if( !( cond ) ) \ reiserfs_panic( NULL, "reiserfs[%i]: assertion " scond " failed at " \ __FILE__ ":%i:%s: " format "\n", \ - in_interrupt() ? -1 : task_pid_nr(current), __LINE__ , __FUNCTION__ , ##args ) + in_interrupt() ? -1 : task_pid_nr(current), __LINE__ , __func__ , ##args ) #define RASSERT(cond, format, args...) __RASSERT(cond, #cond, format, ##args) diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h index 382bb795116..f19b00b7d53 100644 --- a/include/linux/rtmutex.h +++ b/include/linux/rtmutex.h @@ -54,7 +54,7 @@ struct hrtimer_sleeper; #ifdef CONFIG_DEBUG_RT_MUTEXES # define __DEBUG_RT_MUTEX_INITIALIZER(mutexname) \ , .name = #mutexname, .file = __FILE__, .line = __LINE__ -# define rt_mutex_init(mutex) __rt_mutex_init(mutex, __FUNCTION__) +# define rt_mutex_init(mutex) __rt_mutex_init(mutex, __func__) extern void rt_mutex_debug_task_free(struct task_struct *tsk); #else # define __DEBUG_RT_MUTEX_INITIALIZER(mutexname) diff --git a/include/media/saa7146.h b/include/media/saa7146.h index 64a2ec746a3..c5a6e22a4b3 100644 --- a/include/media/saa7146.h +++ b/include/media/saa7146.h @@ -24,7 +24,7 @@ extern unsigned int saa7146_debug; -//#define DEBUG_PROLOG printk("(0x%08x)(0x%08x) %s: %s(): ",(dev==0?-1:(dev->mem==0?-1:saa7146_read(dev,RPS_ADDR0))),(dev==0?-1:(dev->mem==0?-1:saa7146_read(dev,IER))),KBUILD_MODNAME,__FUNCTION__) +//#define DEBUG_PROLOG printk("(0x%08x)(0x%08x) %s: %s(): ",(dev==0?-1:(dev->mem==0?-1:saa7146_read(dev,RPS_ADDR0))),(dev==0?-1:(dev->mem==0?-1:saa7146_read(dev,IER))),KBUILD_MODNAME,__func__) #ifndef DEBUG_VARIABLE #define DEBUG_VARIABLE saa7146_debug diff --git a/include/net/9p/9p.h b/include/net/9p/9p.h index c3626c0ba9d..fb163e2e0de 100644 --- a/include/net/9p/9p.h +++ b/include/net/9p/9p.h @@ -61,7 +61,7 @@ extern unsigned int p9_debug_level; do { \ if ((p9_debug_level & level) == level) \ printk(KERN_NOTICE "-- %s (%d): " \ - format , __FUNCTION__, task_pid_nr(current) , ## arg); \ + format , __func__, task_pid_nr(current) , ## arg); \ } while (0) #define PRINT_FCALL_ERROR(s, fcall) P9_DPRINTK(P9_DEBUG_ERROR, \ @@ -76,7 +76,7 @@ do { \ #define P9_EPRINTK(level, format, arg...) \ do { \ printk(level "9p: %s (%d): " \ - format , __FUNCTION__, task_pid_nr(current), ## arg); \ + format , __func__, task_pid_nr(current), ## arg); \ } while (0) /** diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h index 6f8418bf424..996d12df759 100644 --- a/include/net/bluetooth/bluetooth.h +++ b/include/net/bluetooth/bluetooth.h @@ -54,8 +54,8 @@ #define SOL_RFCOMM 18 #define BT_INFO(fmt, arg...) printk(KERN_INFO "Bluetooth: " fmt "\n" , ## arg) -#define BT_DBG(fmt, arg...) printk(KERN_INFO "%s: " fmt "\n" , __FUNCTION__ , ## arg) -#define BT_ERR(fmt, arg...) printk(KERN_ERR "%s: " fmt "\n" , __FUNCTION__ , ## arg) +#define BT_DBG(fmt, arg...) printk(KERN_INFO "%s: " fmt "\n" , __func__ , ## arg) +#define BT_ERR(fmt, arg...) printk(KERN_ERR "%s: " fmt "\n" , __func__ , ## arg) /* Connection and socket states */ enum { diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h index 6048579d0b2..93a56de3594 100644 --- a/include/net/ieee80211.h +++ b/include/net/ieee80211.h @@ -114,7 +114,7 @@ extern u32 ieee80211_debug_level; #define IEEE80211_DEBUG(level, fmt, args...) \ do { if (ieee80211_debug_level & (level)) \ printk(KERN_DEBUG "ieee80211: %c %s " fmt, \ - in_interrupt() ? 'I' : 'U', __FUNCTION__ , ## args); } while (0) + in_interrupt() ? 'I' : 'U', __func__ , ## args); } while (0) static inline bool ieee80211_ratelimit_debug(u32 level) { return (ieee80211_debug_level & level) && net_ratelimit(); diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index 0b2071d9326..fe9fcf73c85 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -165,13 +165,13 @@ static inline const char *ip_vs_dbg_addr(int af, char *buf, size_t buf_len, do { \ if (level <= ip_vs_get_debug_level()) \ printk(KERN_DEBUG "Enter: %s, %s line %i\n", \ - __FUNCTION__, __FILE__, __LINE__); \ + __func__, __FILE__, __LINE__); \ } while (0) #define LeaveFunction(level) \ do { \ if (level <= ip_vs_get_debug_level()) \ printk(KERN_DEBUG "Leave: %s, %s line %i\n", \ - __FUNCTION__, __FILE__, __LINE__); \ + __func__, __FILE__, __LINE__); \ } while (0) #else #define EnterFunction(level) do {} while (0) diff --git a/include/net/irda/irda.h b/include/net/irda/irda.h index 08387553b57..7e582061b23 100644 --- a/include/net/irda/irda.h +++ b/include/net/irda/irda.h @@ -72,7 +72,7 @@ do { if (irda_debug >= (n)) \ #define IRDA_ASSERT(expr, func) \ do { if(!(expr)) { \ printk( "Assertion failed! %s:%s:%d %s\n", \ - __FILE__,__FUNCTION__,__LINE__,(#expr) ); \ + __FILE__,__func__,__LINE__,(#expr) ); \ func } } while (0) #define IRDA_ASSERT_LABEL(label) label #else diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 703305d0036..ed71b110edf 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -303,7 +303,7 @@ extern int sctp_debug_flag; #define SCTP_ASSERT(expr, str, func) \ if (!(expr)) { \ SCTP_DEBUG_PRINTK("Assertion Failed: %s(%s) at %s:%s:%d\n", \ - str, (#expr), __FILE__, __FUNCTION__, __LINE__); \ + str, (#expr), __FILE__, __func__, __LINE__); \ func; \ } diff --git a/include/video/cyblafb.h b/include/video/cyblafb.h index 71744057538..d3c1d4e2c8e 100644 --- a/include/video/cyblafb.h +++ b/include/video/cyblafb.h @@ -4,7 +4,7 @@ #endif #if CYBLAFB_DEBUG -#define debug(f,a...) printk("%s:" f, __FUNCTION__ , ## a); +#define debug(f,a...) printk("%s:" f, __func__ , ## a); #else #define debug(f,a...) #endif -- cgit v1.2.3-70-g09d2 From bdab0ba3d9ad8de257ee6236daf314723748fde6 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Wed, 15 Oct 2008 22:02:07 -0700 Subject: x86: rename iommu_num_pages function to iommu_nr_pages This series of patches re-introduces the iommu_num_pages function so that it can be used by each architecture specific IOMMU implementations. The series also changes IOMMU implementations for X86, Alpha, PowerPC and UltraSparc. The other implementations are not yet changed because the modifications required are not obvious and I can't test them on real hardware. This patch: This is a preparation patch for introducing a generic iommu_num_pages function. Signed-off-by: Joerg Roedel Cc: "David S. Miller" Cc: Richard Henderson Cc: Ivan Kokshaysky Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Ingo Molnar Cc: Thomas Gleixner Cc: FUJITA Tomonori Cc: Muli Ben-Yehuda Cc: Dave Airlie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/kernel/amd_iommu.c | 8 ++++---- arch/x86/kernel/pci-dma.c | 4 ++-- arch/x86/kernel/pci-gart_64.c | 8 ++++---- include/asm-x86/iommu.h | 2 +- 4 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include/asm-x86') diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 34e4d112b1e..10646acba9b 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -295,7 +295,7 @@ static int iommu_flush_pages(struct amd_iommu *iommu, u16 domid, u64 address, size_t size) { int s = 0; - unsigned pages = iommu_num_pages(address, size); + unsigned pages = iommu_nr_pages(address, size); address &= PAGE_MASK; @@ -679,7 +679,7 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu, if (iommu->exclusion_start && iommu->exclusion_start < dma_dom->aperture_size) { unsigned long startpage = iommu->exclusion_start >> PAGE_SHIFT; - int pages = iommu_num_pages(iommu->exclusion_start, + int pages = iommu_nr_pages(iommu->exclusion_start, iommu->exclusion_length); dma_ops_reserve_addresses(dma_dom, startpage, pages); } @@ -935,7 +935,7 @@ static dma_addr_t __map_single(struct device *dev, unsigned long align_mask = 0; int i; - pages = iommu_num_pages(paddr, size); + pages = iommu_nr_pages(paddr, size); paddr &= PAGE_MASK; if (align) @@ -980,7 +980,7 @@ static void __unmap_single(struct amd_iommu *iommu, if ((dma_addr == 0) || (dma_addr + size > dma_dom->aperture_size)) return; - pages = iommu_num_pages(dma_addr, size); + pages = iommu_nr_pages(dma_addr, size); dma_addr &= PAGE_MASK; start = dma_addr; diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 0a3824e837b..19262482021 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -125,13 +125,13 @@ void __init pci_iommu_alloc(void) pci_swiotlb_init(); } -unsigned long iommu_num_pages(unsigned long addr, unsigned long len) +unsigned long iommu_nr_pages(unsigned long addr, unsigned long len) { unsigned long size = roundup((addr & ~PAGE_MASK) + len, PAGE_SIZE); return size >> PAGE_SHIFT; } -EXPORT_SYMBOL(iommu_num_pages); +EXPORT_SYMBOL(iommu_nr_pages); #endif void *dma_generic_alloc_coherent(struct device *dev, size_t size, diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 145f1c83369..14f1b41348f 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -231,7 +231,7 @@ nonforced_iommu(struct device *dev, unsigned long addr, size_t size) static dma_addr_t dma_map_area(struct device *dev, dma_addr_t phys_mem, size_t size, int dir, unsigned long align_mask) { - unsigned long npages = iommu_num_pages(phys_mem, size); + unsigned long npages = iommu_nr_pages(phys_mem, size); unsigned long iommu_page = alloc_iommu(dev, npages, align_mask); int i; @@ -285,7 +285,7 @@ static void gart_unmap_single(struct device *dev, dma_addr_t dma_addr, return; iommu_page = (dma_addr - iommu_bus_base)>>PAGE_SHIFT; - npages = iommu_num_pages(dma_addr, size); + npages = iommu_nr_pages(dma_addr, size); for (i = 0; i < npages; i++) { iommu_gatt_base[iommu_page + i] = gart_unmapped_entry; CLEAR_LEAK(iommu_page + i); @@ -368,7 +368,7 @@ static int __dma_map_cont(struct device *dev, struct scatterlist *start, } addr = phys_addr; - pages = iommu_num_pages(s->offset, s->length); + pages = iommu_nr_pages(s->offset, s->length); while (pages--) { iommu_gatt_base[iommu_page] = GPTE_ENCODE(addr); SET_LEAK(iommu_page); @@ -451,7 +451,7 @@ gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, int dir) seg_size += s->length; need = nextneed; - pages += iommu_num_pages(s->offset, s->length); + pages += iommu_nr_pages(s->offset, s->length); ps = s; } if (dma_map_cont(dev, start_sg, i - start, sgmap, pages, need) < 0) diff --git a/include/asm-x86/iommu.h b/include/asm-x86/iommu.h index 546ad3110fe..961e746da97 100644 --- a/include/asm-x86/iommu.h +++ b/include/asm-x86/iommu.h @@ -8,7 +8,7 @@ extern int force_iommu, no_iommu; extern int iommu_detected; extern int dmar_disabled; -extern unsigned long iommu_num_pages(unsigned long addr, unsigned long len); +extern unsigned long iommu_nr_pages(unsigned long addr, unsigned long len); #ifdef CONFIG_GART_IOMMU extern int gart_iommu_aperture; -- cgit v1.2.3-70-g09d2