From 93dbda7cbcd70a0bd1a99f39f44a9ccde8ab9040 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 26 Feb 2009 17:35:44 -0800 Subject: x86: add brk allocation for very, very early allocations Impact: new interface Add a brk()-like allocator which effectively extends the bss in order to allow very early code to do dynamic allocations. This is better than using statically allocated arrays for data in subsystems which may never get used. The space for brk allocations is in the bss ELF segment, so that the space is mapped properly by the code which maps the kernel, and so that bootloaders keep the space free rather than putting a ramdisk or something into it. The bss itself, delimited by __bss_stop, ends before the brk area (__brk_base to __brk_limit). The kernel text, data and bss is reserved up to __bss_stop. Any brk-allocated data is reserved separately just before the kernel pagetable is built, as that code allocates from unreserved spaces in the e820 map, potentially allocating from any unused brk memory. Ultimately any unused memory in the brk area is used in the general kernel memory pool. Initially the brk space is set to 1MB, which is probably much larger than any user needs (the largest current user is i386 head_32.S's code to build the pagetables to map the kernel, which can get fairly large with a big kernel image and no PSE support). So long as the system has sufficient memory for the bootloader to reserve the kernel+1MB brk, there are no bad effects resulting from an over-large brk. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/mm/pageattr.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/x86/mm/pageattr.c') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 9c4294986af..1280565670e 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -95,7 +96,7 @@ static inline unsigned long highmap_start_pfn(void) static inline unsigned long highmap_end_pfn(void) { - return __pa(roundup((unsigned long)_end, PMD_SIZE)) >> PAGE_SHIFT; + return __pa(roundup(_brk_end, PMD_SIZE)) >> PAGE_SHIFT; } #endif @@ -711,7 +712,7 @@ static int cpa_process_alias(struct cpa_data *cpa) * No need to redo, when the primary call touched the high * mapping already: */ - if (within(vaddr, (unsigned long) _text, (unsigned long) _end)) + if (within(vaddr, (unsigned long) _text, _brk_end)) return 0; /* -- cgit v1.2.3 From 728c9518873de0bbb92b66daa1943b12e5b9f80f Mon Sep 17 00:00:00 2001 From: "venkatesh.pallipadi@intel.com" Date: Thu, 19 Mar 2009 14:51:13 -0700 Subject: x86, CPA: Add a flag parameter to cpa set_clr() Change change_page_attr_set_clr() array parameter to a flag. This helps following patches which adds an interface to change attr to uc/wb over a set of pages referred by struct page. Signed-off-by: Venkatesh Pallipadi Cc: arjan@infradead.org Cc: eric@anholt.net Cc: Venkatesh Pallipadi Cc: airlied@redhat.com LKML-Reference: <20090319215358.611346000@intel.com> Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/x86/mm/pageattr.c') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 1280565670e..69009afa98c 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -787,7 +787,7 @@ static inline int cache_attr(pgprot_t attr) static int change_page_attr_set_clr(unsigned long *addr, int numpages, pgprot_t mask_set, pgprot_t mask_clr, - int force_split, int array) + int force_split, int in_flag) { struct cpa_data cpa; int ret, cache, checkalias; @@ -802,7 +802,7 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, return 0; /* Ensure we are PAGE_SIZE aligned */ - if (!array) { + if (!(in_flag & CPA_ARRAY)) { if (*addr & ~PAGE_MASK) { *addr &= PAGE_MASK; /* @@ -840,7 +840,7 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, cpa.curpage = 0; cpa.force_split = force_split; - if (array) + if (in_flag & CPA_ARRAY) cpa.flags |= CPA_ARRAY; /* No alias checking for _NX bit modifications */ @@ -889,14 +889,14 @@ 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, - array); + (array ? CPA_ARRAY : 0)); } 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, - array); + (array ? CPA_ARRAY : 0)); } int _set_memory_uc(unsigned long addr, int numpages) -- cgit v1.2.3 From 9ae2847591c857bed44bc094b908b412bfa1b244 Mon Sep 17 00:00:00 2001 From: "venkatesh.pallipadi@intel.com" Date: Thu, 19 Mar 2009 14:51:14 -0700 Subject: x86, PAT: Add support for struct page pointer array in cpa set_clr Add struct page array pointer to cpa struct and CPA_PAGES_ARRAY. With that we can add change_page_attr_set_clr() a parameter to pass struct page array pointer and that can be handled by the underlying cpa code. cpa_flush_array() is also changed to support both addr array or struct page pointer array, depending on the flag. Signed-off-by: Venkatesh Pallipadi Cc: arjan@infradead.org Cc: eric@anholt.net Cc: airlied@redhat.com LKML-Reference: <20090319215358.758513000@intel.com> Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr.c | 79 ++++++++++++++++++++++++++++++++------------------ 1 file changed, 50 insertions(+), 29 deletions(-) (limited to 'arch/x86/mm/pageattr.c') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 69009afa98c..e5c257fb41e 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -34,6 +34,7 @@ struct cpa_data { unsigned long pfn; unsigned force_split : 1; int curpage; + struct page **pages; }; /* @@ -46,6 +47,7 @@ static DEFINE_SPINLOCK(cpa_lock); #define CPA_FLUSHTLB 1 #define CPA_ARRAY 2 +#define CPA_PAGES_ARRAY 4 #ifdef CONFIG_PROC_FS static unsigned long direct_pages_count[PG_LEVEL_NUM]; @@ -202,10 +204,10 @@ static void cpa_flush_range(unsigned long start, int numpages, int cache) } } -static void cpa_flush_array(unsigned long *start, int numpages, int cache) +static void cpa_flush_array(unsigned long *start, int numpages, int cache, + int in_flags, struct page **pages) { unsigned int i, level; - unsigned long *addr; BUG_ON(irqs_disabled()); @@ -226,14 +228,22 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache) * 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); + for (i = 0; i < numpages; i++) { + unsigned long addr; + pte_t *pte; + + if (in_flags & CPA_PAGES_ARRAY) + addr = (unsigned long)page_address(pages[i]); + else + addr = start[i]; + + pte = lookup_address(addr, &level); /* * Only flush present addresses: */ if (pte && (pte_val(*pte) & _PAGE_PRESENT)) - clflush_cache_range((void *) *addr, PAGE_SIZE); + clflush_cache_range((void *)addr, PAGE_SIZE); } } @@ -585,7 +595,9 @@ static int __change_page_attr(struct cpa_data *cpa, int primary) unsigned int level; pte_t *kpte, old_pte; - if (cpa->flags & CPA_ARRAY) + if (cpa->flags & CPA_PAGES_ARRAY) + address = (unsigned long)page_address(cpa->pages[cpa->curpage]); + else if (cpa->flags & CPA_ARRAY) address = cpa->vaddr[cpa->curpage]; else address = *cpa->vaddr; @@ -688,7 +700,9 @@ static int cpa_process_alias(struct cpa_data *cpa) * No need to redo, when the primary call touched the direct * mapping already: */ - if (cpa->flags & CPA_ARRAY) + if (cpa->flags & CPA_PAGES_ARRAY) + vaddr = (unsigned long)page_address(cpa->pages[cpa->curpage]); + else if (cpa->flags & CPA_ARRAY) vaddr = cpa->vaddr[cpa->curpage]; else vaddr = *cpa->vaddr; @@ -699,7 +713,7 @@ static int cpa_process_alias(struct cpa_data *cpa) alias_cpa = *cpa; temp_cpa_vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT); alias_cpa.vaddr = &temp_cpa_vaddr; - alias_cpa.flags &= ~CPA_ARRAY; + alias_cpa.flags &= ~(CPA_PAGES_ARRAY | CPA_ARRAY); ret = __change_page_attr_set_clr(&alias_cpa, 0); @@ -725,7 +739,7 @@ static int cpa_process_alias(struct cpa_data *cpa) alias_cpa = *cpa; temp_cpa_vaddr = (cpa->pfn << PAGE_SHIFT) + __START_KERNEL_map - phys_base; alias_cpa.vaddr = &temp_cpa_vaddr; - alias_cpa.flags &= ~CPA_ARRAY; + alias_cpa.flags &= ~(CPA_PAGES_ARRAY | CPA_ARRAY); /* * The high mapping range is imprecise, so ignore the return value. @@ -746,7 +760,7 @@ static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias) */ cpa->numpages = numpages; /* for array changes, we can't use large page */ - if (cpa->flags & CPA_ARRAY) + if (cpa->flags & (CPA_ARRAY | CPA_PAGES_ARRAY)) cpa->numpages = 1; if (!debug_pagealloc) @@ -770,7 +784,7 @@ static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias) */ BUG_ON(cpa->numpages > numpages); numpages -= cpa->numpages; - if (cpa->flags & CPA_ARRAY) + if (cpa->flags & (CPA_PAGES_ARRAY | CPA_ARRAY)) cpa->curpage++; else *cpa->vaddr += cpa->numpages * PAGE_SIZE; @@ -787,7 +801,8 @@ static inline int cache_attr(pgprot_t attr) static int change_page_attr_set_clr(unsigned long *addr, int numpages, pgprot_t mask_set, pgprot_t mask_clr, - int force_split, int in_flag) + int force_split, int in_flag, + struct page **pages) { struct cpa_data cpa; int ret, cache, checkalias; @@ -802,15 +817,7 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, return 0; /* Ensure we are PAGE_SIZE aligned */ - if (!(in_flag & CPA_ARRAY)) { - if (*addr & ~PAGE_MASK) { - *addr &= PAGE_MASK; - /* - * People should not be passing in unaligned addresses: - */ - WARN_ON_ONCE(1); - } - } else { + if (in_flag & CPA_ARRAY) { int i; for (i = 0; i < numpages; i++) { if (addr[i] & ~PAGE_MASK) { @@ -818,6 +825,18 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, WARN_ON_ONCE(1); } } + } else if (!(in_flag & CPA_PAGES_ARRAY)) { + /* + * in_flag of CPA_PAGES_ARRAY implies it is aligned. + * No need to cehck in that case + */ + if (*addr & ~PAGE_MASK) { + *addr &= PAGE_MASK; + /* + * People should not be passing in unaligned addresses: + */ + WARN_ON_ONCE(1); + } } /* Must avoid aliasing mappings in the highmem code */ @@ -833,6 +852,7 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, arch_flush_lazy_mmu_mode(); cpa.vaddr = addr; + cpa.pages = pages; cpa.numpages = numpages; cpa.mask_set = mask_set; cpa.mask_clr = mask_clr; @@ -840,8 +860,8 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, cpa.curpage = 0; cpa.force_split = force_split; - if (in_flag & CPA_ARRAY) - cpa.flags |= CPA_ARRAY; + if (in_flag & (CPA_ARRAY | CPA_PAGES_ARRAY)) + cpa.flags |= in_flag; /* No alias checking for _NX bit modifications */ checkalias = (pgprot_val(mask_set) | pgprot_val(mask_clr)) != _PAGE_NX; @@ -867,9 +887,10 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, * wbindv): */ if (!ret && cpu_has_clflush) { - if (cpa.flags & CPA_ARRAY) - cpa_flush_array(addr, numpages, cache); - else + if (cpa.flags & (CPA_PAGES_ARRAY | CPA_ARRAY)) { + cpa_flush_array(addr, numpages, cache, + cpa.flags, pages); + } else cpa_flush_range(*addr, numpages, cache); } else cpa_flush_all(cache); @@ -889,14 +910,14 @@ 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, - (array ? CPA_ARRAY : 0)); + (array ? CPA_ARRAY : 0), NULL); } 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, - (array ? CPA_ARRAY : 0)); + (array ? CPA_ARRAY : 0), NULL); } int _set_memory_uc(unsigned long addr, int numpages) @@ -1044,7 +1065,7 @@ int set_memory_np(unsigned long addr, int numpages) int set_memory_4k(unsigned long addr, int numpages) { return change_page_attr_set_clr(&addr, numpages, __pgprot(0), - __pgprot(0), 1, 0); + __pgprot(0), 1, 0, NULL); } int set_pages_uc(struct page *page, int numpages) -- cgit v1.2.3 From 0f3507555f6fa4acbc85a646d6e8766230db38fc Mon Sep 17 00:00:00 2001 From: "venkatesh.pallipadi@intel.com" Date: Thu, 19 Mar 2009 14:51:15 -0700 Subject: x86, CPA: Add set_pages_arrayuc and set_pages_array_wb Add new interfaces: set_pages_array_uc() set_pages_array_wb() that can be used change the page attribute for a bunch of pages with flush etc done once at the end of all the changes. These interfaces are similar to existing set_memory_array_uc() and set_memory_array_wc(). Signed-off-by: Venkatesh Pallipadi Cc: arjan@infradead.org Cc: eric@anholt.net Cc: airlied@redhat.com LKML-Reference: <20090319215358.901545000@intel.com> Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr.c | 63 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 63 insertions(+) (limited to 'arch/x86/mm/pageattr.c') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index e5c257fb41e..d71e1b636ce 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -920,6 +920,20 @@ static inline int change_page_attr_clear(unsigned long *addr, int numpages, (array ? CPA_ARRAY : 0), NULL); } +static inline int cpa_set_pages_array(struct page **pages, int numpages, + pgprot_t mask) +{ + return change_page_attr_set_clr(NULL, numpages, mask, __pgprot(0), 0, + CPA_PAGES_ARRAY, pages); +} + +static inline int cpa_clear_pages_array(struct page **pages, int numpages, + pgprot_t mask) +{ + return change_page_attr_set_clr(NULL, numpages, __pgprot(0), mask, 0, + CPA_PAGES_ARRAY, pages); +} + int _set_memory_uc(unsigned long addr, int numpages) { /* @@ -1076,6 +1090,35 @@ int set_pages_uc(struct page *page, int numpages) } EXPORT_SYMBOL(set_pages_uc); +int set_pages_array_uc(struct page **pages, int addrinarray) +{ + unsigned long start; + unsigned long end; + int i; + int free_idx; + + for (i = 0; i < addrinarray; i++) { + start = (unsigned long)page_address(pages[i]); + end = start + PAGE_SIZE; + if (reserve_memtype(start, end, _PAGE_CACHE_UC_MINUS, NULL)) + goto err_out; + } + + if (cpa_set_pages_array(pages, addrinarray, + __pgprot(_PAGE_CACHE_UC_MINUS)) == 0) { + return 0; /* Success */ + } +err_out: + free_idx = i; + for (i = 0; i < free_idx; i++) { + start = (unsigned long)page_address(pages[i]); + end = start + PAGE_SIZE; + free_memtype(start, end); + } + return -EINVAL; +} +EXPORT_SYMBOL(set_pages_array_uc); + int set_pages_wb(struct page *page, int numpages) { unsigned long addr = (unsigned long)page_address(page); @@ -1084,6 +1127,26 @@ int set_pages_wb(struct page *page, int numpages) } EXPORT_SYMBOL(set_pages_wb); +int set_pages_array_wb(struct page **pages, int addrinarray) +{ + int retval; + unsigned long start; + unsigned long end; + int i; + + retval = cpa_clear_pages_array(pages, addrinarray, + __pgprot(_PAGE_CACHE_MASK)); + + for (i = 0; i < addrinarray; i++) { + start = (unsigned long)page_address(pages[i]); + end = start + PAGE_SIZE; + free_memtype(start, end); + } + + return retval; +} +EXPORT_SYMBOL(set_pages_array_wb); + int set_pages_x(struct page *page, int numpages) { unsigned long addr = (unsigned long)page_address(page); -- cgit v1.2.3