From a2bcd4731f77cb77ae4b5e4a3d7f5471cf346c33 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sun, 29 Mar 2009 23:47:48 +0200 Subject: x86/mm: further cleanups of fault.c's include file section Impact: cleanup Eliminate more than 20 unnecessary #include lines in fault.c Also fix include file dependency bug in asm/traps.h. (this was masked before, by implicit inclusion) Signed-off-by: Ingo Molnar LKML-Reference: Acked-by: H. Peter Anvin --- arch/x86/include/asm/traps.h | 1 + arch/x86/mm/fault.c | 44 ++++++++++---------------------------------- 2 files changed, 11 insertions(+), 34 deletions(-) diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h index 0d5342515b8..37fb07a9cda 100644 --- a/arch/x86/include/asm/traps.h +++ b/arch/x86/include/asm/traps.h @@ -2,6 +2,7 @@ #define _ASM_X86_TRAPS_H #include +#include /* TRAP_TRACE, ... */ #ifdef CONFIG_X86_32 #define dotraplinkage diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index a03b7279efa..24a36a6426a 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -3,40 +3,16 @@ * Copyright (C) 2001, 2002 Andi Kleen, SuSE Labs. * Copyright (C) 2008-2009, Red Hat Inc., Ingo Molnar */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include - -#include -#include -#include -#include -#include -#include -#include +#include /* STACK_END_MAGIC */ +#include /* test_thread_flag(), ... */ +#include /* oops_begin/end, ... */ +#include /* search_exception_table */ +#include /* max_low_pfn */ +#include /* __kprobes, ... */ +#include /* kmmio_handler, ... */ + +#include /* dotraplinkage, ... */ +#include /* pgd_*(), ... */ /* * Page fault error code bits: -- cgit v1.2.3 From 66aa230e437d89ca56224135f617e2d8e391a3ef Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Tue, 14 Apr 2009 12:54:29 +0530 Subject: x86: page_types.h unification of declarations Impact: unification of declarations, cleanup Unification of declarations: moved init_memory_mapping, initmem_init and free_initmem from page_XX_types.h to page_types.h Signed-off-by: Jaswinder Singh Rajput Acked-by: Pekka Enberg Cc: Andrew Morton LKML-Reference: <1239693869.3033.31.camel@ht.satnam> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/page_32_types.h | 4 ---- arch/x86/include/asm/page_64_types.h | 6 ------ arch/x86/include/asm/page_types.h | 6 ++++++ 3 files changed, 6 insertions(+), 10 deletions(-) diff --git a/arch/x86/include/asm/page_32_types.h b/arch/x86/include/asm/page_32_types.h index 0f915ae649a..6f1b7331313 100644 --- a/arch/x86/include/asm/page_32_types.h +++ b/arch/x86/include/asm/page_32_types.h @@ -54,10 +54,6 @@ extern unsigned int __VMALLOC_RESERVE; extern int sysctl_legacy_va_layout; 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); #endif /* !__ASSEMBLY__ */ diff --git a/arch/x86/include/asm/page_64_types.h b/arch/x86/include/asm/page_64_types.h index d38c91b7024..3f587188ae6 100644 --- a/arch/x86/include/asm/page_64_types.h +++ b/arch/x86/include/asm/page_64_types.h @@ -71,12 +71,6 @@ extern unsigned long __phys_addr(unsigned long); #define vmemmap ((struct page *)VMEMMAP_START) -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); diff --git a/arch/x86/include/asm/page_types.h b/arch/x86/include/asm/page_types.h index 826ad37006a..6473f5ccff8 100644 --- a/arch/x86/include/asm/page_types.h +++ b/arch/x86/include/asm/page_types.h @@ -46,6 +46,12 @@ extern int devmem_is_allowed(unsigned long pagenr); extern unsigned long max_low_pfn_mapped; extern unsigned long max_pfn_mapped; +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); + #endif /* !__ASSEMBLY__ */ #endif /* _ASM_X86_PAGE_DEFS_H */ -- cgit v1.2.3 From 6424fb38667fffbbb1b90be0ffd9a0c540db6a4b Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Mon, 13 Apr 2009 23:51:46 -0700 Subject: x86: remove (null) in /sys kernel_page_tables Impact: cleanup %p prints out 0x000000000000000 as (null) so use %lx instead. Signed-off-by: Yinghai Lu LKML-Reference: <49E43282.1090607@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/mm/dump_pagetables.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c index e7277cbcfb4..a725b7f760a 100644 --- a/arch/x86/mm/dump_pagetables.c +++ b/arch/x86/mm/dump_pagetables.c @@ -161,13 +161,14 @@ static void note_page(struct seq_file *m, struct pg_state *st, st->current_address >= st->marker[1].start_address) { const char *unit = units; unsigned long delta; + int width = sizeof(unsigned long) * 2; /* * Now print the actual finished series */ - seq_printf(m, "0x%p-0x%p ", - (void *)st->start_address, - (void *)st->current_address); + seq_printf(m, "0x%0*lx-0x%0*lx ", + width, st->start_address, + width, st->current_address); delta = (st->current_address - st->start_address) >> 10; while (!(delta & 1023) && unit[1]) { -- cgit v1.2.3 From 89388913f2c88a2cd15d24abab571b17a2596127 Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Tue, 21 Apr 2009 11:39:27 +0300 Subject: x86: unify noexec handling This patch unifies noexec handling on 32-bit and 64-bit. [ Impact: cleanup ] Signed-off-by: Pekka Enberg [ mingo@elte.hu: build fix ] LKML-Reference: <1240303167.771.69.camel@penberg-laptop> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/pgtable_types.h | 1 - arch/x86/mm/init.c | 67 +++++++++++++++++++++++++++++++++--- arch/x86/mm/init_32.c | 52 ---------------------------- arch/x86/mm/init_64.c | 33 ------------------ 4 files changed, 63 insertions(+), 90 deletions(-) diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h index b8238dc8786..4d258ad76a0 100644 --- a/arch/x86/include/asm/pgtable_types.h +++ b/arch/x86/include/asm/pgtable_types.h @@ -273,7 +273,6 @@ typedef struct page *pgtable_t; extern pteval_t __supported_pte_mask; extern int nx_enabled; -extern void set_nx(void); #define pgprot_writecombine pgprot_writecombine extern pgprot_t pgprot_writecombine(pgprot_t prot); diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index fd3da1dda1c..fedde5359a0 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -22,6 +22,69 @@ int direct_gbpages #endif ; +int nx_enabled; + +#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) +static int disable_nx __cpuinitdata; + +/* + * noexec = on|off + * + * Control non-executable mappings for processes. + * + * on Enable + * off Disable + */ +static int __init noexec_setup(char *str) +{ + if (!str) + return -EINVAL; + if (!strncmp(str, "on", 2)) { + __supported_pte_mask |= _PAGE_NX; + disable_nx = 0; + } else if (!strncmp(str, "off", 3)) { + disable_nx = 1; + __supported_pte_mask &= ~_PAGE_NX; + } + return 0; +} +early_param("noexec", noexec_setup); +#endif + +#ifdef CONFIG_X86_PAE +static void __init set_nx(void) +{ + unsigned int v[4], l, h; + + if (cpu_has_pae && (cpuid_eax(0x80000000) > 0x80000001)) { + cpuid(0x80000001, &v[0], &v[1], &v[2], &v[3]); + + if ((v[3] & (1 << 20)) && !disable_nx) { + rdmsr(MSR_EFER, l, h); + l |= EFER_NX; + wrmsr(MSR_EFER, l, h); + nx_enabled = 1; + __supported_pte_mask |= _PAGE_NX; + } + } +} +#else +static inline void set_nx(void) +{ +} +#endif + +#ifdef CONFIG_X86_64 +void __cpuinit check_efer(void) +{ + unsigned long efer; + + rdmsrl(MSR_EFER, efer); + if (!(efer & EFER_NX) || disable_nx) + __supported_pte_mask &= ~_PAGE_NX; +} +#endif + static void __init find_early_table_space(unsigned long end, int use_pse, int use_gbpages) { @@ -158,12 +221,9 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, use_gbpages = direct_gbpages; #endif -#ifdef CONFIG_X86_32 -#ifdef CONFIG_X86_PAE set_nx(); if (nx_enabled) printk(KERN_INFO "NX (Execute Disable) protection: active\n"); -#endif /* Enable PSE if available */ if (cpu_has_pse) @@ -174,7 +234,6 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, set_in_cr4(X86_CR4_PGE); __supported_pte_mask |= _PAGE_GLOBAL; } -#endif if (use_gbpages) page_size_mask |= 1 << PG_LEVEL_1G; diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 749559ed80f..2b27120665b 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -587,61 +587,9 @@ void zap_low_mappings(void) flush_tlb_all(); } -int nx_enabled; - pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL | _PAGE_IOMAP); EXPORT_SYMBOL_GPL(__supported_pte_mask); -#ifdef CONFIG_X86_PAE - -static int disable_nx __initdata; - -/* - * noexec = on|off - * - * Control non executable mappings. - * - * on Enable - * off Disable - */ -static int __init noexec_setup(char *str) -{ - if (!str || !strcmp(str, "on")) { - if (cpu_has_nx) { - __supported_pte_mask |= _PAGE_NX; - disable_nx = 0; - } - } else { - if (!strcmp(str, "off")) { - disable_nx = 1; - __supported_pte_mask &= ~_PAGE_NX; - } else { - return -EINVAL; - } - } - - return 0; -} -early_param("noexec", noexec_setup); - -void __init set_nx(void) -{ - unsigned int v[4], l, h; - - if (cpu_has_pae && (cpuid_eax(0x80000000) > 0x80000001)) { - cpuid(0x80000001, &v[0], &v[1], &v[2], &v[3]); - - if ((v[3] & (1 << 20)) && !disable_nx) { - rdmsr(MSR_EFER, l, h); - l |= EFER_NX; - wrmsr(MSR_EFER, l, h); - nx_enabled = 1; - __supported_pte_mask |= _PAGE_NX; - } - } -} -#endif - /* user-defined highmem size */ static unsigned int highmem_pages = -1; diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 1753e8020df..a4e7846efb1 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -85,39 +85,6 @@ early_param("gbpages", parse_direct_gbpages_on); pteval_t __supported_pte_mask __read_mostly = ~_PAGE_IOMAP; EXPORT_SYMBOL_GPL(__supported_pte_mask); -static int disable_nx __cpuinitdata; - -/* - * noexec=on|off - * Control non-executable mappings for 64-bit processes. - * - * on Enable (default) - * off Disable - */ -static int __init nonx_setup(char *str) -{ - if (!str) - return -EINVAL; - if (!strncmp(str, "on", 2)) { - __supported_pte_mask |= _PAGE_NX; - disable_nx = 0; - } else if (!strncmp(str, "off", 3)) { - disable_nx = 1; - __supported_pte_mask &= ~_PAGE_NX; - } - return 0; -} -early_param("noexec", nonx_setup); - -void __cpuinit check_efer(void) -{ - unsigned long efer; - - rdmsrl(MSR_EFER, efer); - if (!(efer & EFER_NX) || disable_nx) - __supported_pte_mask &= ~_PAGE_NX; -} - int force_personality32; /* -- cgit v1.2.3 From 2b72394e4089643f11669d9610907a1442fe044a Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Tue, 28 Apr 2009 16:00:49 +0300 Subject: x86: move max_pfn_mapped and max_low_pfn_mapped to setup.c This patch moves the max_pfn_mapped and max_low_pfn_mapped global variables to kernel/setup.c where they're initialized. [ Impact: cleanup ] Signed-off-by: Pekka Enberg LKML-Reference: <1240923649.1982.21.camel@penberg-laptop> Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 8 ++++++++ arch/x86/mm/init_32.c | 4 +--- arch/x86/mm/init_64.c | 8 -------- 3 files changed, 9 insertions(+), 11 deletions(-) diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index b4158439bf6..0d77e56e821 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -112,6 +112,14 @@ #define ARCH_SETUP #endif +/* + * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. + * The direct mapping extends to max_pfn_mapped, so that we can directly access + * apertures, ACPI and other tables without having to play with fixmaps. + */ +unsigned long max_low_pfn_mapped; +unsigned long max_pfn_mapped; + RESERVE_BRK(dmi_alloc, 65536); unsigned int boot_cpu_id __read_mostly; diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 2b27120665b..a640a7f0490 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -49,11 +49,9 @@ #include #include #include +#include #include -unsigned long max_low_pfn_mapped; -unsigned long max_pfn_mapped; - DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); unsigned long highstart_pfn, highend_pfn; diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index a4e7846efb1..1016ea01593 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -50,14 +50,6 @@ #include #include -/* - * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. - * The direct mapping extends to max_pfn_mapped, so that we can directly access - * apertures, ACPI and other tables without having to play with fixmaps. - */ -unsigned long max_low_pfn_mapped; -unsigned long max_pfn_mapped; - static unsigned long dma_reserve __initdata; DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); -- cgit v1.2.3 From 9518e0e4350a5ea8ca200ce320b28d6284a7b0ce Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Tue, 28 Apr 2009 16:00:50 +0300 Subject: x86: move per-cpu mmu_gathers to mm/init.c [ Impact: cleanup ] Signed-off-by: Pekka Enberg LKML-Reference: <1240923650.1982.22.camel@penberg-laptop> Signed-off-by: Ingo Molnar --- arch/x86/mm/init.c | 3 +++ arch/x86/mm/init_32.c | 1 - arch/x86/mm/init_64.c | 2 -- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index fedde5359a0..4d67c33a2e1 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -9,6 +9,9 @@ #include #include #include +#include + +DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); unsigned long __initdata e820_table_start; unsigned long __meminitdata e820_table_end; diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index a640a7f0490..fef1d90d4f1 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -52,7 +52,6 @@ #include #include -DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); unsigned long highstart_pfn, highend_pfn; static noinline int do_test_wp_bit(void); diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 1016ea01593..6a1a573e20f 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -52,8 +52,6 @@ static unsigned long dma_reserve __initdata; -DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); - static int __init parse_direct_gbpages_off(char *arg) { direct_gbpages = 0; -- cgit v1.2.3 From a454ab3110175d710f4f9a96226a26ce4d5d5de2 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sun, 3 May 2009 10:09:03 +0200 Subject: x86, mm: fault.c, use printk_once() in is_errata93() Andrew pointed out that the 'once' variable has a needlessly function-global scope. We can in fact eliminate it completely, via the use of printk_once(). [ Impact: cleanup ] Reported-by: Andrew Morton Cc: Linus Torvalds Signed-off-by: Ingo Molnar --- arch/x86/mm/fault.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 24a36a6426a..b9ca6d767db 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -514,8 +514,6 @@ bad: static int is_errata93(struct pt_regs *regs, unsigned long address) { #ifdef CONFIG_X86_64 - static int once; - if (address != regs->ip) return 0; @@ -525,10 +523,7 @@ static int is_errata93(struct pt_regs *regs, unsigned long address) address |= 0xffffffffUL << 32; if ((address >= (u64)_stext && address <= (u64)_etext) || (address >= MODULES_VADDR && address <= MODULES_END)) { - if (!once) { - printk(errata93_warning); - once = 1; - } + printk_once(errata93_warning); regs->ip = address; return 1; } -- cgit v1.2.3 From c898faf91b3ec6b0f6efa35831b3984fa3331db0 Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Tue, 5 May 2009 17:28:56 -0400 Subject: x86: 46 bit physical address support on 64 bits Extend the maximum addressable memory on x86-64 from 2^44 to 2^46 bytes. This requires some shuffling around of the vmalloc and virtual memmap memory areas, to keep them away from the direct mapping of up to 64TB of physical memory. This patch also introduces a guard hole between the vmalloc area and the virtual memory map space. There's really no good reason why we wouldn't have a guard hole there. [ Impact: future hardware enablement ] Signed-off-by: Rik van Riel LKML-Reference: <20090505172856.6820db22@cuia.bos.redhat.com> Signed-off-by: H. Peter Anvin --- Documentation/x86/x86_64/mm.txt | 9 +++++---- arch/x86/include/asm/page_64_types.h | 2 +- arch/x86/include/asm/pgtable_64_types.h | 8 ++++---- arch/x86/include/asm/sparsemem.h | 2 +- 4 files changed, 11 insertions(+), 10 deletions(-) diff --git a/Documentation/x86/x86_64/mm.txt b/Documentation/x86/x86_64/mm.txt index 29b52b14d0b..53941323584 100644 --- a/Documentation/x86/x86_64/mm.txt +++ b/Documentation/x86/x86_64/mm.txt @@ -6,10 +6,11 @@ Virtual memory map with 4 level page tables: 0000000000000000 - 00007fffffffffff (=47 bits) user space, different per mm hole caused by [48:63] sign extension ffff800000000000 - ffff80ffffffffff (=40 bits) guard hole -ffff880000000000 - ffffc0ffffffffff (=57 TB) direct mapping of all phys. memory -ffffc10000000000 - ffffc1ffffffffff (=40 bits) hole -ffffc20000000000 - ffffe1ffffffffff (=45 bits) vmalloc/ioremap space -ffffe20000000000 - ffffe2ffffffffff (=40 bits) virtual memory map (1TB) +ffff880000000000 - ffffc8ffffffffff (=64 TB) direct mapping of all phys. memory +ffffc80000000000 - ffffc8ffffffffff (=40 bits) hole +ffffc90000000000 - ffffe8ffffffffff (=45 bits) vmalloc/ioremap space +ffffe90000000000 - ffffe9ffffffffff (=40 bits) hole +ffffea0000000000 - ffffeaffffffffff (=40 bits) virtual memory map (1TB) ... unused hole ... ffffffff80000000 - ffffffffa0000000 (=512 MB) kernel text mapping, from phys 0 ffffffffa0000000 - fffffffffff00000 (=1536 MB) module mapping space diff --git a/arch/x86/include/asm/page_64_types.h b/arch/x86/include/asm/page_64_types.h index 3f587188ae6..6fadb020bd2 100644 --- a/arch/x86/include/asm/page_64_types.h +++ b/arch/x86/include/asm/page_64_types.h @@ -47,7 +47,7 @@ #define __START_KERNEL (__START_KERNEL_map + __PHYSICAL_START) #define __START_KERNEL_map _AC(0xffffffff80000000, UL) -/* See Documentation/x86_64/mm.txt for a description of the memory map. */ +/* See Documentation/x86/x86_64/mm.txt for a description of the memory map. */ #define __PHYSICAL_MASK_SHIFT 46 #define __VIRTUAL_MASK_SHIFT 48 diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h index fbf42b8e038..766ea16fbbb 100644 --- a/arch/x86/include/asm/pgtable_64_types.h +++ b/arch/x86/include/asm/pgtable_64_types.h @@ -51,11 +51,11 @@ typedef struct { pteval_t pte; } pte_t; #define PGDIR_SIZE (_AC(1, UL) << PGDIR_SHIFT) #define PGDIR_MASK (~(PGDIR_SIZE - 1)) - +/* See Documentation/x86/x86_64/mm.txt for a description of the memory map. */ #define MAXMEM _AC(__AC(1, UL) << MAX_PHYSMEM_BITS, UL) -#define VMALLOC_START _AC(0xffffc20000000000, UL) -#define VMALLOC_END _AC(0xffffe1ffffffffff, UL) -#define VMEMMAP_START _AC(0xffffe20000000000, UL) +#define VMALLOC_START _AC(0xffffc90000000000, UL) +#define VMALLOC_END _AC(0xffffe8ffffffffff, UL) +#define VMEMMAP_START _AC(0xffffea0000000000, UL) #define MODULES_VADDR _AC(0xffffffffa0000000, UL) #define MODULES_END _AC(0xffffffffff000000, UL) #define MODULES_LEN (MODULES_END - MODULES_VADDR) diff --git a/arch/x86/include/asm/sparsemem.h b/arch/x86/include/asm/sparsemem.h index e3cc3c063ec..4517d6b9318 100644 --- a/arch/x86/include/asm/sparsemem.h +++ b/arch/x86/include/asm/sparsemem.h @@ -27,7 +27,7 @@ #else /* CONFIG_X86_32 */ # define SECTION_SIZE_BITS 27 /* matt - 128 is convenient right now */ # define MAX_PHYSADDR_BITS 44 -# define MAX_PHYSMEM_BITS 44 /* Can be max 45 bits */ +# define MAX_PHYSMEM_BITS 46 #endif #endif /* CONFIG_SPARSEMEM */ -- cgit v1.2.3 From 2feceeff1e771850e49f9074307f071964fd9e3e Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Tue, 5 May 2009 19:07:07 -0700 Subject: x86: fix typo in address space documentation Fix a trivial typo in Documentation/x86/x86_64/mm.txt. [ Impact: documentation only ] Signed-off-by: H. Peter Anvin Cc: Rik van Riel --- Documentation/x86/x86_64/mm.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/x86/x86_64/mm.txt b/Documentation/x86/x86_64/mm.txt index 53941323584..d6498e3cd71 100644 --- a/Documentation/x86/x86_64/mm.txt +++ b/Documentation/x86/x86_64/mm.txt @@ -6,7 +6,7 @@ Virtual memory map with 4 level page tables: 0000000000000000 - 00007fffffffffff (=47 bits) user space, different per mm hole caused by [48:63] sign extension ffff800000000000 - ffff80ffffffffff (=40 bits) guard hole -ffff880000000000 - ffffc8ffffffffff (=64 TB) direct mapping of all phys. memory +ffff880000000000 - ffffc7ffffffffff (=64 TB) direct mapping of all phys. memory ffffc80000000000 - ffffc8ffffffffff (=40 bits) hole ffffc90000000000 - ffffe8ffffffffff (=45 bits) vmalloc/ioremap space ffffe90000000000 - ffffe9ffffffffff (=40 bits) hole -- cgit v1.2.3 From 45fbe3ee01b8e463b28c2751b5dcc0cbdc142d90 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 6 May 2009 08:06:44 -0700 Subject: x86, e820, pci: reserve extra free space near end of RAM The point is to take all RAM resources we have, and _after_ we've added all the resources we've seen in the E820 tree, we then _also_ try to add fake reserved entries for any "round up to X" at the end of the RAM resources. [ Impact: improve PCI mem-resource allocation robustness, protect "stolen RAM" ] Reported-by: Yannick Roehlly Acked-by: Jesse Barnes Signed-off-by: Yinghai Lu Cc: Ivan Kokshaysky Cc: Andrew Morton Cc: yannick.roehlly@free.fr LKML-Reference: <4A01A784.2050407@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/e820.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 00628130292..a2335d9de05 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -1371,6 +1371,23 @@ void __init e820_reserve_resources(void) } } +/* How much should we pad RAM ending depending on where it is? */ +static unsigned long ram_alignment(resource_size_t pos) +{ + unsigned long mb = pos >> 20; + + /* To 64kB in the first megabyte */ + if (!mb) + return 64*1024; + + /* To 1MB in the first 16MB */ + if (mb < 16) + return 1024*1024; + + /* To 32MB for anything above that */ + return 32*1024*1024; +} + void __init e820_reserve_resources_late(void) { int i; @@ -1382,6 +1399,24 @@ void __init e820_reserve_resources_late(void) insert_resource_expand_to_fit(&iomem_resource, res); res++; } + + /* + * Try to bump up RAM regions to reasonable boundaries to + * avoid stolen RAM: + */ + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *entry = &e820_saved.map[i]; + resource_size_t start, end; + + if (entry->type != E820_RAM) + continue; + start = entry->addr + entry->size; + end = round_up(start, ram_alignment(start)); + if (start == end) + continue; + reserve_region_with_split(&iomem_resource, start, + end - 1, "RAM buffer"); + } } char *__init default_machine_specific_memory_setup(void) -- cgit v1.2.3 From 5d423ccd7ba4285f1084e91b26805e1d0ae978ed Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 6 May 2009 08:07:52 -0700 Subject: x86/pci: remove rounding quirk from e820_setup_gap() Now that the e820 code explicitly reserves 'potentially dangerous' free physical memory address space to protect ACPI stolen RAM, there's no need for the rounding quirk in the PCI allocator anymore. Also, this quirk was open-ended iteration that could end up reserving a lot of free space and potentially breaking drivers - such as the one reported by Yannick Roehlly where there's a PCI device with a large memory resource. So remove it. [ Impact: make more of the PCI hole available for assigning pci devices ] Reported-by: Yannick Roehlly Signed-off-by: Yinghai Lu Acked-by: Jesse Barnes Cc: Ivan Kokshaysky Cc: Linus Torvalds Cc: Andrew Morton LKML-Reference: <4A01A7C8.5090701@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/e820.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index a2335d9de05..7271fa33d79 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -617,7 +617,7 @@ __init int e820_search_gap(unsigned long *gapstart, unsigned long *gapsize, */ __init void e820_setup_gap(void) { - unsigned long gapstart, gapsize, round; + unsigned long gapstart, gapsize; int found; gapstart = 0x10000000; @@ -635,14 +635,9 @@ __init void e820_setup_gap(void) #endif /* - * See how much we want to round up: start off with - * rounding to the next 1MB area. + * e820_reserve_resources_late protect stolen RAM already */ - round = 0x100000; - while ((gapsize >> 4) > round) - round += round; - /* Fun with two's complement */ - pci_mem_start = (gapstart + round) & -round; + pci_mem_start = gapstart; printk(KERN_INFO "Allocating PCI resources starting at %lx (gap: %lx:%lx)\n", -- cgit v1.2.3 From 80989ce0643c1034822f3e339ed8d790b649abe1 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sat, 9 May 2009 23:47:42 -0700 Subject: x86: clean up and and print out initial max_pfn_mapped Do this so we can check the range that is mapped before init_memory_mapping(). To be able to print out meaningful info, we first have to fix 64-bit to have max_pfn_mapped assigned before that call. This also unifies the code-path a bit. [ Impact: print more debug info, cleanup ] Signed-off-by: Yinghai Lu LKML-Reference: <49BF0978.40605@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 4 ++++ arch/x86/mm/init.c | 7 +++---- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 0d77e56e821..4031d6cb3ff 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -862,12 +862,16 @@ void __init setup_arch(char **cmdline_p) max_low_pfn = max_pfn; high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1; + max_pfn_mapped = KERNEL_IMAGE_SIZE >> PAGE_SHIFT; #endif #ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION setup_bios_corruption_check(); #endif + printk(KERN_DEBUG "initial memory mapped : 0 - %08lx\n", + max_pfn_mapped< Date: Fri, 8 May 2009 00:36:44 -0700 Subject: x86: Sanity check the e820 against the SRAT table using e820 map only node_cover_memory() sanity checks the SRAT table by ensuring that all PXMs cover the memory reported in the e820. However, when calculating the size of the holes in the e820, it uses the early_node_map[] which contains information taken from both SRAT and e820. If the SRAT is missing an entry, then it is not detected that the SRAT table is incorrect and missing entries. This patch uses the e820 map to calculate the holes instead of early_node_map[]. comment is from Mel. [ Impact: reject incorrect SRAT tables ] Signed-off-by: Yinghai Lu Acked-by: Mel Gorman Cc: Andrew Morton LKML-Reference: <4A03E10C.60906@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/mm/srat_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c index 01765955baa..c7a18aaccf8 100644 --- a/arch/x86/mm/srat_64.c +++ b/arch/x86/mm/srat_64.c @@ -345,7 +345,7 @@ static int __init nodes_cover_memory(const struct bootnode *nodes) pxmram = 0; } - e820ram = max_pfn - absent_pages_in_range(0, max_pfn); + e820ram = max_pfn - (e820_hole_size(0, max_pfn<>PAGE_SHIFT); /* We seem to lose 3 pages somewhere. Allow a bit of slack. */ if ((long)(e820ram - pxmram) >= 1*1024*1024) { printk(KERN_ERR -- cgit v1.2.3 From 0964b0562bb9c93194e852b47bab2397b9e11c18 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 8 May 2009 00:37:34 -0700 Subject: x86: Allow 1MB of slack between the e820 map and SRAT, not 4GB It is expected that there might be slight differences between the e820 map and the SRAT table and the intention was that 1MB of slack be allowed. The calculation comparing e820ram and pxmram assumes the units are bytes, when they are in fact pages. This means 4GB of slack is being allowed, not 1MB. This patch makes the correct comparison. comment is from Mel. [ Impact: don't accept buggy SRATs that could dump up to 4G of RAM ] Signed-off-by: Yinghai Lu Acked-by: Mel Gorman Cc: Andrew Morton LKML-Reference: <4A03E13E.6050107@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/mm/srat_64.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c index c7a18aaccf8..87b45bff250 100644 --- a/arch/x86/mm/srat_64.c +++ b/arch/x86/mm/srat_64.c @@ -346,8 +346,8 @@ static int __init nodes_cover_memory(const struct bootnode *nodes) } e820ram = max_pfn - (e820_hole_size(0, max_pfn<>PAGE_SHIFT); - /* We seem to lose 3 pages somewhere. Allow a bit of slack. */ - if ((long)(e820ram - pxmram) >= 1*1024*1024) { + /* We seem to lose 3 pages somewhere. Allow 1M of slack. */ + if ((long)(e820ram - pxmram) >= (1<<(20 - PAGE_SHIFT))) { printk(KERN_ERR "SRAT: PXMs only cover %luMB of your %luMB e820 RAM. Not used.\n", (pxmram << PAGE_SHIFT) >> 20, -- cgit v1.2.3 From 3551f88f6439cf4da3f5a3747b320280e30500de Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Thu, 7 May 2009 15:35:41 +0300 Subject: x86: unify 64-bit UMA and NUMA paging_init() 64-bit UMA and NUMA versions of paging_init() are almost identical. Therefore, merge the copy in mm/numa_64.c to mm/init_64.c to remove duplicate code. [ Impact: cleanup ] Signed-off-by: Pekka Enberg LKML-Reference: <1241699741.17846.30.camel@penberg-laptop> Signed-off-by: Ingo Molnar --- arch/x86/mm/init_64.c | 6 +++++- arch/x86/mm/numa_64.c | 15 --------------- 2 files changed, 5 insertions(+), 16 deletions(-) diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 6a1a573e20f..be7e1279178 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -585,6 +585,7 @@ void __init initmem_init(unsigned long start_pfn, unsigned long end_pfn) early_res_to_bootmem(0, end_pfn< Date: Thu, 7 May 2009 15:35:42 +0300 Subject: x86: use sparse_memory_present_with_active_regions() on UMA There's no need to use call memory_present() manually on UMA because initmem_init() sets up early_node_map by calling e820_register_active_regions(). [ Impact: cleanup ] Signed-off-by: Pekka Enberg LKML-Reference: <1241699742.17846.31.camel@penberg-laptop> Signed-off-by: Ingo Molnar --- arch/x86/mm/init_64.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index be7e1279178..52bb9519bb8 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -596,11 +596,7 @@ void __init paging_init(void) max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN; max_zone_pfns[ZONE_NORMAL] = max_pfn; -#ifdef CONFIG_NUMA sparse_memory_present_with_active_regions(MAX_NUMNODES); -#else - memory_present(0, 0, max_pfn); -#endif sparse_init(); free_area_init_nodes(max_zone_pfns); } -- cgit v1.2.3 From 9d62dcdfa6f6fc843f7d9b494bcd48f00b94f883 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Mon, 11 May 2009 22:05:28 -0400 Subject: x86: merge process.c a bit Merge arch_align_stack() and arch_randomize_brk(), since they are the same. Tested on x86_64. [ Impact: cleanup ] Signed-off-by: Amerigo Wang Signed-off-by: Ingo Molnar --- arch/x86/kernel/process.c | 14 ++++++++++++++ arch/x86/kernel/process_32.c | 13 ------------- arch/x86/kernel/process_64.c | 13 ------------- 3 files changed, 14 insertions(+), 26 deletions(-) diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index ca989158e84..2b9a8d0fb47 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -613,3 +614,16 @@ static int __init idle_setup(char *str) } early_param("idle", idle_setup); +unsigned long arch_align_stack(unsigned long sp) +{ + if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) + sp -= get_random_int() % 8192; + return sp & ~0xf; +} + +unsigned long arch_randomize_brk(struct mm_struct *mm) +{ + unsigned long range_end = mm->brk + 0x02000000; + return randomize_range(mm->brk, range_end, 0) ? : mm->brk; +} + diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 76f8f84043a..a3bb049ad08 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -33,7 +33,6 @@ #include #include #include -#include #include #include #include @@ -497,15 +496,3 @@ unsigned long get_wchan(struct task_struct *p) return 0; } -unsigned long arch_align_stack(unsigned long sp) -{ - if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) - sp -= get_random_int() % 8192; - return sp & ~0xf; -} - -unsigned long arch_randomize_brk(struct mm_struct *mm) -{ - unsigned long range_end = mm->brk + 0x02000000; - return randomize_range(mm->brk, range_end, 0) ? : mm->brk; -} diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index b751a41392b..34386f72bdc 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -32,7 +32,6 @@ #include #include #include -#include #include #include #include @@ -660,15 +659,3 @@ long sys_arch_prctl(int code, unsigned long addr) return do_arch_prctl(current, code, addr); } -unsigned long arch_align_stack(unsigned long sp) -{ - if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) - sp -= get_random_int() % 8192; - return sp & ~0xf; -} - -unsigned long arch_randomize_brk(struct mm_struct *mm) -{ - unsigned long range_end = mm->brk + 0x02000000; - return randomize_range(mm->brk, range_end, 0) ? : mm->brk; -} -- cgit v1.2.3 From bf78ad69cd351798b9447a269c6bd41ce1f111f4 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Mon, 11 May 2009 23:29:09 -0400 Subject: x86: process.c, remove useless headers is not needed by these files, remove them. [ Impact: cleanup ] Signed-off-by: WANG Cong Cc: akpm@linux-foundation.org LKML-Reference: <20090512032956.5040.77055.sendpatchset@localhost.localdomain> Signed-off-by: Ingo Molnar --- arch/x86/kernel/process_32.c | 2 -- arch/x86/kernel/process_64.c | 2 -- 2 files changed, 4 deletions(-) diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index a3bb049ad08..56d50b7d71d 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -9,8 +9,6 @@ * This file handles the architecture-dependent parts of process handling.. */ -#include - #include #include #include diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 34386f72bdc..9d6b20e6cd8 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -14,8 +14,6 @@ * This file handles the architecture-dependent parts of process handling.. */ -#include - #include #include #include -- cgit v1.2.3 From ed077b58f6146684069975122b1728a9d248a501 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 12 May 2009 16:40:00 +0800 Subject: x86: make sparse mem work in non-NUMA mode With sparse memory, holes should not be marked present for memmap. This patch makes sure sparsemem really works on SMP mode (!NUMA). [ Impact: use less memory to map fragmented RAM, avoid boot-OOM/crash ] Signed-off-by: Shaohua Li Signed-off-by: Sheng Yang LKML-Reference: <1242117600.22431.0.camel@sli10-desk.sh.intel.com> Signed-off-by: Ingo Molnar --- arch/x86/mm/init_32.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index fef1d90d4f1..949708d7a48 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -706,15 +706,15 @@ void __init initmem_init(unsigned long start_pfn, highstart_pfn = highend_pfn = max_pfn; if (max_pfn > max_low_pfn) highstart_pfn = max_low_pfn; - memory_present(0, 0, highend_pfn); e820_register_active_regions(0, 0, highend_pfn); + sparse_memory_present_with_active_regions(0); printk(KERN_NOTICE "%ldMB HIGHMEM available.\n", pages_to_mb(highend_pfn - highstart_pfn)); num_physpages = highend_pfn; high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1; #else - memory_present(0, 0, max_low_pfn); e820_register_active_regions(0, 0, max_low_pfn); + sparse_memory_present_with_active_regions(0); num_physpages = max_low_pfn; high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1; #endif -- cgit v1.2.3 From 888a589f6be07d624e21e2174d98375e9f95911b Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 15 May 2009 13:59:37 -0700 Subject: mm, x86: remove MEMORY_HOTPLUG_RESERVE related code after: | commit b263295dbffd33b0fbff670720fa178c30e3392a | Author: Christoph Lameter | Date: Wed Jan 30 13:30:47 2008 +0100 | | x86: 64-bit, make sparsemem vmemmap the only memory model we don't have MEMORY_HOTPLUG_RESERVE anymore. Historically, x86-64 had an architecture-specific method for memory hotplug whereby it scanned the SRAT for physical memory ranges that could be potentially used for memory hot-add later. By reserving those ranges without physical memory, the memmap would be allocated and left dormant until needed. This depended on the DISCONTIG memory model which has been removed so the code implementing HOTPLUG_RESERVE is now dead. This patch removes the dead code used by MEMORY_HOTPLUG_RESERVE. (Changelog authored by Mel.) v2: updated changelog, and remove hotadd= in doc [ Impact: remove dead code ] Signed-off-by: Yinghai Lu Reviewed-by: Christoph Lameter Reviewed-by: Mel Gorman Workflow-found-OK-by: Andrew Morton LKML-Reference: <4A0C4910.7090508@kernel.org> Signed-off-by: Ingo Molnar --- Documentation/x86/x86_64/boot-options.txt | 5 --- arch/x86/include/asm/numa_64.h | 3 -- arch/x86/mm/numa_64.c | 5 --- arch/x86/mm/srat_64.c | 63 ++++++---------------------- include/linux/mm.h | 2 - mm/page_alloc.c | 69 ------------------------------- 6 files changed, 12 insertions(+), 135 deletions(-) diff --git a/Documentation/x86/x86_64/boot-options.txt b/Documentation/x86/x86_64/boot-options.txt index 34c13040a71..2db5893d6c9 100644 --- a/Documentation/x86/x86_64/boot-options.txt +++ b/Documentation/x86/x86_64/boot-options.txt @@ -150,11 +150,6 @@ NUMA Otherwise, the remaining system RAM is allocated to an additional node. - numa=hotadd=percent - Only allow hotadd memory to preallocate page structures upto - percent of already available memory. - numa=hotadd=0 will disable hotadd memory. - ACPI acpi=off Don't enable ACPI diff --git a/arch/x86/include/asm/numa_64.h b/arch/x86/include/asm/numa_64.h index 064ed6df4cb..7feff0648d7 100644 --- a/arch/x86/include/asm/numa_64.h +++ b/arch/x86/include/asm/numa_64.h @@ -17,9 +17,6 @@ extern int compute_hash_shift(struct bootnode *nodes, int numblks, extern void numa_init_array(void); extern int numa_off; -extern void srat_reserve_add_area(int nodeid); -extern int hotadd_percent; - extern s16 apicid_to_node[MAX_LOCAL_APIC]; extern unsigned long numa_free_all_bootmem(void); diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index fb61d81a656..a6a93c39523 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -272,9 +272,6 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, reserve_bootmem_node(NODE_DATA(nodeid), bootmap_start, bootmap_pages<> PAGE_SHIFT; unsigned long e_pfn = end >> PAGE_SHIFT; - int ret = 0, changed = 0; + int changed = 0; struct bootnode *nd = &nodes_add[node]; /* I had some trouble with strange memory hotadd regions breaking @@ -210,7 +201,7 @@ reserve_hotadd(int node, unsigned long start, unsigned long end) mistakes */ if ((signed long)(end - start) < NODE_MIN_SIZE) { printk(KERN_ERR "SRAT: Hotplug area too small\n"); - return -1; + return; } /* This check might be a bit too strict, but I'm keeping it for now. */ @@ -218,12 +209,7 @@ reserve_hotadd(int node, unsigned long start, unsigned long end) printk(KERN_ERR "SRAT: Hotplug area %lu -> %lu has existing memory\n", s_pfn, e_pfn); - return -1; - } - - if (!hotadd_enough_memory(&nodes_add[node])) { - printk(KERN_ERR "SRAT: Hotplug area too large\n"); - return -1; + return; } /* Looks good */ @@ -245,11 +231,9 @@ reserve_hotadd(int node, unsigned long start, unsigned long end) printk(KERN_ERR "SRAT: Hotplug zone not continuous. Partly ignored\n"); } - ret = update_end_of_memory(nd->end); - if (changed) - printk(KERN_INFO "SRAT: hot plug zone found %Lx - %Lx\n", nd->start, nd->end); - return ret; + printk(KERN_INFO "SRAT: hot plug zone found %Lx - %Lx\n", + nd->start, nd->end); } /* Callback for parsing of the Proximity Domain <-> Memory Area mappings */ @@ -310,13 +294,10 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) start, end); e820_register_active_regions(node, start >> PAGE_SHIFT, end >> PAGE_SHIFT); - push_node_boundaries(node, nd->start >> PAGE_SHIFT, - nd->end >> PAGE_SHIFT); - if ((ma->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE) && - (reserve_hotadd(node, start, end) < 0)) { - /* Ignore hotadd region. Undo damage */ - printk(KERN_NOTICE "SRAT: Hotplug region ignored\n"); + if (ma->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE) { + update_nodes_add(node, start, end); + /* restore nodes[node] */ *nd = oldnode; if ((nd->start | nd->end) == 0) node_clear(node, nodes_parsed); @@ -510,26 +491,6 @@ static int null_slit_node_compare(int a, int b) } #endif /* CONFIG_NUMA_EMU */ -void __init srat_reserve_add_area(int nodeid) -{ - if (found_add_area && nodes_add[nodeid].end) { - u64 total_mb; - - printk(KERN_INFO "SRAT: Reserving hot-add memory space " - "for node %d at %Lx-%Lx\n", - nodeid, nodes_add[nodeid].start, nodes_add[nodeid].end); - total_mb = (nodes_add[nodeid].end - nodes_add[nodeid].start) - >> PAGE_SHIFT; - total_mb *= sizeof(struct page); - total_mb >>= 20; - printk(KERN_INFO "SRAT: This will cost you %Lu MB of " - "pre-allocated memory.\n", (unsigned long long)total_mb); - reserve_bootmem_node(NODE_DATA(nodeid), nodes_add[nodeid].start, - nodes_add[nodeid].end - nodes_add[nodeid].start, - BOOTMEM_DEFAULT); - } -} - int __node_distance(int a, int b) { int index; diff --git a/include/linux/mm.h b/include/linux/mm.h index bff1f0d475c..511b0986709 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1031,8 +1031,6 @@ extern void add_active_range(unsigned int nid, unsigned long start_pfn, unsigned long end_pfn); extern void remove_active_range(unsigned int nid, unsigned long start_pfn, unsigned long end_pfn); -extern void push_node_boundaries(unsigned int nid, unsigned long start_pfn, - unsigned long end_pfn); extern void remove_all_active_ranges(void); extern unsigned long absent_pages_in_range(unsigned long start_pfn, unsigned long end_pfn); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index fe753ecf2aa..474c7e9dd51 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -149,10 +149,6 @@ static unsigned long __meminitdata dma_reserve; static int __meminitdata nr_nodemap_entries; static unsigned long __meminitdata arch_zone_lowest_possible_pfn[MAX_NR_ZONES]; static unsigned long __meminitdata arch_zone_highest_possible_pfn[MAX_NR_ZONES]; -#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE - static unsigned long __meminitdata node_boundary_start_pfn[MAX_NUMNODES]; - static unsigned long __meminitdata node_boundary_end_pfn[MAX_NUMNODES]; -#endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */ static unsigned long __initdata required_kernelcore; static unsigned long __initdata required_movablecore; static unsigned long __meminitdata zone_movable_pfn[MAX_NUMNODES]; @@ -3102,64 +3098,6 @@ void __init sparse_memory_present_with_active_regions(int nid) early_node_map[i].end_pfn); } -/** - * push_node_boundaries - Push node boundaries to at least the requested boundary - * @nid: The nid of the node to push the boundary for - * @start_pfn: The start pfn of the node - * @end_pfn: The end pfn of the node - * - * In reserve-based hot-add, mem_map is allocated that is unused until hotadd - * time. Specifically, on x86_64, SRAT will report ranges that can potentially - * be hotplugged even though no physical memory exists. This function allows - * an arch to push out the node boundaries so mem_map is allocated that can - * be used later. - */ -#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE -void __init push_node_boundaries(unsigned int nid, - unsigned long start_pfn, unsigned long end_pfn) -{ - mminit_dprintk(MMINIT_TRACE, "zoneboundary", - "Entering push_node_boundaries(%u, %lu, %lu)\n", - nid, start_pfn, end_pfn); - - /* Initialise the boundary for this node if necessary */ - if (node_boundary_end_pfn[nid] == 0) - node_boundary_start_pfn[nid] = -1UL; - - /* Update the boundaries */ - if (node_boundary_start_pfn[nid] > start_pfn) - node_boundary_start_pfn[nid] = start_pfn; - if (node_boundary_end_pfn[nid] < end_pfn) - node_boundary_end_pfn[nid] = end_pfn; -} - -/* If necessary, push the node boundary out for reserve hotadd */ -static void __meminit account_node_boundary(unsigned int nid, - unsigned long *start_pfn, unsigned long *end_pfn) -{ - mminit_dprintk(MMINIT_TRACE, "zoneboundary", - "Entering account_node_boundary(%u, %lu, %lu)\n", - nid, *start_pfn, *end_pfn); - - /* Return if boundary information has not been provided */ - if (node_boundary_end_pfn[nid] == 0) - return; - - /* Check the boundaries and update if necessary */ - if (node_boundary_start_pfn[nid] < *start_pfn) - *start_pfn = node_boundary_start_pfn[nid]; - if (node_boundary_end_pfn[nid] > *end_pfn) - *end_pfn = node_boundary_end_pfn[nid]; -} -#else -void __init push_node_boundaries(unsigned int nid, - unsigned long start_pfn, unsigned long end_pfn) {} - -static void __meminit account_node_boundary(unsigned int nid, - unsigned long *start_pfn, unsigned long *end_pfn) {} -#endif - - /** * get_pfn_range_for_nid - Return the start and end page frames for a node * @nid: The nid to return the range for. If MAX_NUMNODES, the min and max PFN are returned. @@ -3185,9 +3123,6 @@ void __meminit get_pfn_range_for_nid(unsigned int nid, if (*start_pfn == -1UL) *start_pfn = 0; - - /* Push the node boundaries out if requested */ - account_node_boundary(nid, start_pfn, end_pfn); } /* @@ -3793,10 +3728,6 @@ void __init remove_all_active_ranges(void) { memset(early_node_map, 0, sizeof(early_node_map)); nr_nodemap_entries = 0; -#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE - memset(node_boundary_start_pfn, 0, sizeof(node_boundary_start_pfn)); - memset(node_boundary_end_pfn, 0, sizeof(node_boundary_end_pfn)); -#endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */ } /* Compare two active node_active_regions */ -- cgit v1.2.3 From 7c43769a9776141ec23ca81a1bdd5a9c0512f165 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 15 May 2009 13:59:37 -0700 Subject: x86, mm: Fix node_possible_map logic Recently there were some changes to the meaning of node_possible_map, and it is quite strange: - the node without memory would be set in node_possible_map - but some node with less NODE_MIN_SIZE will be kicked out of node_possible_map. fix it by adding strict_setup_node_bootmem(). Also, remove unparse_node(). so result will be: 1. cpu_to_node() will return online node only (nearest one) 2. apicid_to_node() still returns the node that could be not online but is set in node_possible_map. 3. node_possible_map will include nodes that mem on it are less NODE_MIN_SIZE v2: after move_cpus_to_node change. [ Impact: get node_possible_map right ] Signed-off-by: Yinghai Lu Tested-by: Jack Steiner LKML-Reference: <4A0C49BE.6080800@kernel.org> [ v3: various small cleanups and comment clarifications ] Signed-off-by: Ingo Molnar --- arch/x86/include/asm/numa_64.h | 7 +++++++ arch/x86/mm/numa_64.c | 13 ++++++++++--- arch/x86/mm/srat_64.c | 29 ++--------------------------- 3 files changed, 19 insertions(+), 30 deletions(-) diff --git a/arch/x86/include/asm/numa_64.h b/arch/x86/include/asm/numa_64.h index 7feff0648d7..c4ae822e415 100644 --- a/arch/x86/include/asm/numa_64.h +++ b/arch/x86/include/asm/numa_64.h @@ -24,6 +24,13 @@ extern void setup_node_bootmem(int nodeid, unsigned long start, unsigned long end); #ifdef CONFIG_NUMA +/* + * Too small node sizes may confuse the VM badly. Usually they + * result from BIOS bugs. So dont recognize nodes as standalone + * NUMA entities that have less than this amount of RAM listed: + */ +#define NODE_MIN_SIZE (4*1024*1024) + extern void __init init_cpu_to_node(void); extern void __cpuinit numa_set_node(int cpu, int node); extern void __cpuinit numa_clear_node(int cpu); diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index a6a93c39523..459913beac7 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -179,18 +179,25 @@ static void * __init early_node_mem(int nodeid, unsigned long start, } /* Initialize bootmem allocator for a node */ -void __init setup_node_bootmem(int nodeid, unsigned long start, - unsigned long end) +void __init +setup_node_bootmem(int nodeid, unsigned long start, unsigned long end) { unsigned long start_pfn, last_pfn, bootmap_pages, bootmap_size; + const int pgdat_size = roundup(sizeof(pg_data_t), PAGE_SIZE); unsigned long bootmap_start, nodedata_phys; void *bootmap; - const int pgdat_size = roundup(sizeof(pg_data_t), PAGE_SIZE); int nid; if (!end) return; + /* + * Don't confuse VM with a node that doesn't have the + * minimum amount of memory: + */ + if (end && (end - start) < NODE_MIN_SIZE) + return; + start = roundup(start, ZONE_ALIGN); printk(KERN_INFO "Bootmem setup node %d %016lx-%016lx\n", nodeid, diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c index b0dbbd48e58..2dfcbf9df2a 100644 --- a/arch/x86/mm/srat_64.c +++ b/arch/x86/mm/srat_64.c @@ -36,10 +36,6 @@ static int num_node_memblks __initdata; static struct bootnode node_memblk_range[NR_NODE_MEMBLKS] __initdata; static int memblk_nodeid[NR_NODE_MEMBLKS] __initdata; -/* Too small nodes confuse the VM badly. Usually they result - from BIOS bugs. */ -#define NODE_MIN_SIZE (4*1024*1024) - static __init int setup_node(int pxm) { return acpi_map_pxm_to_node(pxm); @@ -338,17 +334,6 @@ static int __init nodes_cover_memory(const struct bootnode *nodes) return 1; } -static void __init unparse_node(int node) -{ - int i; - node_clear(node, nodes_parsed); - node_clear(node, cpu_nodes_parsed); - for (i = 0; i < MAX_LOCAL_APIC; i++) { - if (apicid_to_node[i] == node) - apicid_to_node[i] = NUMA_NO_NODE; - } -} - void __init acpi_numa_arch_fixup(void) {} /* Use the information discovered above to actually set up the nodes. */ @@ -360,18 +345,8 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end) return -1; /* First clean up the node list */ - for (i = 0; i < MAX_NUMNODES; i++) { + for (i = 0; i < MAX_NUMNODES; i++) cutoff_node(i, start, end); - /* - * don't confuse VM with a node that doesn't have the - * minimum memory. - */ - if (nodes[i].end && - (nodes[i].end - nodes[i].start) < NODE_MIN_SIZE) { - unparse_node(i); - node_set_offline(i); - } - } if (!nodes_cover_memory(nodes)) { bad_srat(); @@ -404,7 +379,7 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end) if (node == NUMA_NO_NODE) continue; - if (!node_isset(node, node_possible_map)) + if (!node_online(node)) numa_clear_node(i); } numa_init_array(); -- cgit v1.2.3 From 35d5a9a61490bf39d2e48d7f499c8c801a39ebe9 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 15 May 2009 13:59:37 -0700 Subject: x86: fix system without memory on node0 Jack found a boot crash on a system which doesn't have memory on node0. It turns out with recent per_cpu changes, node_number for BSP will always be 0, and it is not consistent to cpu_to_node() that might set it to a different (nearer) node already. aka when numa_set_node() for node0 is called early before per_cpu area is setup: two places touched that per_cpu(node_number,): 1. in cpu/common.c::cpu_init() and it is not for BP | #ifdef CONFIG_NUMA | if (cpu != 0 && percpu_read(node_number) == 0 && | cpu_to_node(cpu) != NUMA_NO_NODE) | percpu_write(node_number, cpu_to_node(cpu)); | #endif for BP: traps_init ==> cpu_init for AP: start_secondary ==> cpu_init 2. cpu/intel.c or amd.c::srat_detect_node via numa_set_node() for BP: check_bugs ==> identify_boot_cpu ==> identify_cpu() that is rather later before numa_node_id() is used for BP... for AP: start_secondary => smp_callin => smp_store_cpu_info() => => identify_secondary_cpu => identify_cpu() so try to set that for BP earlier in setup_per_cpu_areas(), and don't bother to set that for APs there (it will be updated later and will be used later) (and don't mess the 0 before the copying BP per_cpu data to APs) [ Impact: fix boot crash on memoryless node-0 ] Reported-and-tested-by: Jack Steiner Cc: Tejun Heo Signed-off-by: Yinghai Lu LKML-Reference: <4A0C4A02.7050401@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup_percpu.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 3a97a4cf187..3b5f3271e73 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -423,6 +423,14 @@ void __init setup_per_cpu_areas(void) early_per_cpu_ptr(x86_cpu_to_node_map) = NULL; #endif +#if defined(CONFIG_X86_64) && defined(CONFIG_NUMA) + /* + * make sure boot cpu node_number is right, when boot cpu is on the + * node that doesn't have mem installed + */ + per_cpu(node_number, boot_cpu_id) = cpu_to_node(boot_cpu_id); +#endif + /* Setup node to cpumask map */ setup_node_to_cpumask_map(); -- cgit v1.2.3