diff options
Diffstat (limited to 'arch')
71 files changed, 580 insertions, 416 deletions
diff --git a/arch/alpha/mm/init.c b/arch/alpha/mm/init.c index 486d7945583..544ac5dc09e 100644 --- a/arch/alpha/mm/init.c +++ b/arch/alpha/mm/init.c @@ -357,7 +357,7 @@ free_reserved_mem(void *start, void *end) void *__start = start; for (; __start < end; __start += PAGE_SIZE) { ClearPageReserved(virt_to_page(__start)); - set_page_count(virt_to_page(__start), 1); + init_page_count(virt_to_page(__start)); free_page((long)__start); totalram_pages++; } diff --git a/arch/arm/mm/consistent.c b/arch/arm/mm/consistent.c index c2ee18d2075..8a1bfcd5008 100644 --- a/arch/arm/mm/consistent.c +++ b/arch/arm/mm/consistent.c @@ -223,6 +223,8 @@ __dma_alloc(struct device *dev, size_t size, dma_addr_t *handle, gfp_t gfp, pte = consistent_pte[idx] + off; c->vm_pages = page; + split_page(page, order); + /* * Set the "dma handle" */ @@ -231,7 +233,6 @@ __dma_alloc(struct device *dev, size_t size, dma_addr_t *handle, gfp_t gfp, do { BUG_ON(!pte_none(*pte)); - set_page_count(page, 1); /* * x86 does not mark the pages reserved... */ @@ -250,7 +251,6 @@ __dma_alloc(struct device *dev, size_t size, dma_addr_t *handle, gfp_t gfp, * Free the otherwise unused pages. */ while (page < end) { - set_page_count(page, 1); __free_page(page); page++; } diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c index 8b276ee38ac..b0321e943b7 100644 --- a/arch/arm/mm/init.c +++ b/arch/arm/mm/init.c @@ -531,7 +531,7 @@ static inline void free_area(unsigned long addr, unsigned long end, char *s) for (; addr < end; addr += PAGE_SIZE) { struct page *page = virt_to_page(addr); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); free_page(addr); totalram_pages++; } diff --git a/arch/arm26/mm/init.c b/arch/arm26/mm/init.c index 1f09a9d0fb8..e3ecaa45374 100644 --- a/arch/arm26/mm/init.c +++ b/arch/arm26/mm/init.c @@ -324,7 +324,7 @@ static inline void free_area(unsigned long addr, unsigned long end, char *s) for (; addr < end; addr += PAGE_SIZE) { struct page *page = virt_to_page(addr); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); free_page(addr); totalram_pages++; } diff --git a/arch/cris/mm/init.c b/arch/cris/mm/init.c index 31a0018b525..b7842ff213a 100644 --- a/arch/cris/mm/init.c +++ b/arch/cris/mm/init.c @@ -216,7 +216,7 @@ free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } diff --git a/arch/frv/kernel/frv_ksyms.c b/arch/frv/kernel/frv_ksyms.c index 0f1c6cbc4f5..aa6b7d0a210 100644 --- a/arch/frv/kernel/frv_ksyms.c +++ b/arch/frv/kernel/frv_ksyms.c @@ -27,6 +27,7 @@ EXPORT_SYMBOL(__ioremap); EXPORT_SYMBOL(iounmap); EXPORT_SYMBOL(strnlen); +EXPORT_SYMBOL(strpbrk); EXPORT_SYMBOL(strrchr); EXPORT_SYMBOL(strstr); EXPORT_SYMBOL(strchr); diff --git a/arch/frv/mm/dma-alloc.c b/arch/frv/mm/dma-alloc.c index 342823aad75..636b2f8b5d9 100644 --- a/arch/frv/mm/dma-alloc.c +++ b/arch/frv/mm/dma-alloc.c @@ -115,9 +115,7 @@ void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *dma_handle) */ if (order > 0) { struct page *rpage = virt_to_page(page); - - for (i = 1; i < (1 << order); i++) - set_page_count(rpage + i, 1); + split_page(rpage, order); } err = 0; diff --git a/arch/frv/mm/init.c b/arch/frv/mm/init.c index 765088ea8a5..8899aa1a4f0 100644 --- a/arch/frv/mm/init.c +++ b/arch/frv/mm/init.c @@ -169,7 +169,7 @@ void __init mem_init(void) struct page *page = &mem_map[pfn]; ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalram_pages++; } @@ -210,7 +210,7 @@ void __init free_initmem(void) /* next to check that the page we free is not a partial page */ for (addr = start; addr < end; addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -230,7 +230,7 @@ void __init free_initrd_mem(unsigned long start, unsigned long end) int pages = 0; for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; pages++; diff --git a/arch/h8300/kernel/h8300_ksyms.c b/arch/h8300/kernel/h8300_ksyms.c index 5cc76efaf7a..69d6ad32d56 100644 --- a/arch/h8300/kernel/h8300_ksyms.c +++ b/arch/h8300/kernel/h8300_ksyms.c @@ -25,6 +25,7 @@ extern char h8300_debug_device[]; /* platform dependent support */ EXPORT_SYMBOL(strnlen); +EXPORT_SYMBOL(strpbrk); EXPORT_SYMBOL(strrchr); EXPORT_SYMBOL(strstr); EXPORT_SYMBOL(strchr); diff --git a/arch/h8300/mm/init.c b/arch/h8300/mm/init.c index 1e0929ddc8c..09efc4b1f03 100644 --- a/arch/h8300/mm/init.c +++ b/arch/h8300/mm/init.c @@ -196,7 +196,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) int pages = 0; for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; pages++; @@ -219,7 +219,7 @@ free_initmem() /* next to check that the page we free is not a partial page */ for (; addr + PAGE_SIZE < (unsigned long)(&__init_end); addr +=PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } diff --git a/arch/i386/kernel/efi.c b/arch/i386/kernel/efi.c index c9cad7ba0d2..aeabb419686 100644 --- a/arch/i386/kernel/efi.c +++ b/arch/i386/kernel/efi.c @@ -115,7 +115,7 @@ static void efi_call_phys_epilog(void) unsigned long cr4; struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, 0); - cpu_gdt_descr->address = __va(cpu_gdt_descr->address); + cpu_gdt_descr->address = (unsigned long)__va(cpu_gdt_descr->address); load_gdt(cpu_gdt_descr); cr4 = read_cr4(); diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c index 218d725a5a1..d134e9643a5 100644 --- a/arch/i386/kernel/smp.c +++ b/arch/i386/kernel/smp.c @@ -504,27 +504,23 @@ void unlock_ipi_call_lock(void) spin_unlock_irq(&call_lock); } -static struct call_data_struct * call_data; - -/* - * this function sends a 'generic call function' IPI to all other CPUs - * in the system. - */ - -int smp_call_function (void (*func) (void *info), void *info, int nonatomic, - int wait) -/* - * [SUMMARY] Run a function on all other CPUs. - * <func> The function to run. This must be fast and non-blocking. - * <info> An arbitrary pointer to pass to the function. - * <nonatomic> currently unused. - * <wait> If true, wait (atomically) until function has completed on other CPUs. - * [RETURNS] 0 on success, else a negative status code. Does not return until +static struct call_data_struct *call_data; + +/** + * smp_call_function(): Run a function on all other CPUs. + * @func: The function to run. This must be fast and non-blocking. + * @info: An arbitrary pointer to pass to the function. + * @nonatomic: currently unused. + * @wait: If true, wait (atomically) until function has completed on other CPUs. + * + * Returns 0 on success, else a negative status code. Does not return until * remote CPUs are nearly ready to execute <<func>> or are or have executed. * * You must not call this function with disabled interrupts or from a * hardware interrupt handler or from a bottom half handler. */ +int smp_call_function (void (*func) (void *info), void *info, int nonatomic, + int wait) { struct call_data_struct data; int cpus; diff --git a/arch/i386/kernel/sys_i386.c b/arch/i386/kernel/sys_i386.c index a4a61976ecb..8fdb1fb17a5 100644 --- a/arch/i386/kernel/sys_i386.c +++ b/arch/i386/kernel/sys_i386.c @@ -40,14 +40,13 @@ asmlinkage int sys_pipe(unsigned long __user * fildes) return error; } -/* common code for old and new mmaps */ -static inline long do_mmap2( - unsigned long addr, unsigned long len, - unsigned long prot, unsigned long flags, - unsigned long fd, unsigned long pgoff) +asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long pgoff) { int error = -EBADF; - struct file * file = NULL; + struct file *file = NULL; + struct mm_struct *mm = current->mm; flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); if (!(flags & MAP_ANONYMOUS)) { @@ -56,9 +55,9 @@ static inline long do_mmap2( goto out; } - down_write(¤t->mm->mmap_sem); + down_write(&mm->mmap_sem); error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff); - up_write(¤t->mm->mmap_sem); + up_write(&mm->mmap_sem); if (file) fput(file); @@ -66,13 +65,6 @@ out: return error; } -asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, - unsigned long prot, unsigned long flags, - unsigned long fd, unsigned long pgoff) -{ - return do_mmap2(addr, len, prot, flags, fd, pgoff); -} - /* * Perform the select(nd, in, out, ex, tv) and mmap() system * calls. Linux/i386 didn't use to be able to handle more than @@ -101,7 +93,8 @@ asmlinkage int old_mmap(struct mmap_arg_struct __user *arg) if (a.offset & ~PAGE_MASK) goto out; - err = do_mmap2(a.addr, a.len, a.prot, a.flags, a.fd, a.offset >> PAGE_SHIFT); + err = sys_mmap2(a.addr, a.len, a.prot, a.flags, + a.fd, a.offset >> PAGE_SHIFT); out: return err; } diff --git a/arch/i386/kernel/timers/timer_hpet.c b/arch/i386/kernel/timers/timer_hpet.c index be242723c33..17a6fe7166e 100644 --- a/arch/i386/kernel/timers/timer_hpet.c +++ b/arch/i386/kernel/timers/timer_hpet.c @@ -46,7 +46,7 @@ static seqlock_t monotonic_lock = SEQLOCK_UNLOCKED; * * -johnstul@us.ibm.com "math is hard, lets go shopping!" */ -static unsigned long cyc2ns_scale; +static unsigned long cyc2ns_scale __read_mostly; #define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ static inline void set_cyc2ns_scale(unsigned long cpu_khz) diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c index a7f5a2aceba..5e41ee29c8c 100644 --- a/arch/i386/kernel/timers/timer_tsc.c +++ b/arch/i386/kernel/timers/timer_tsc.c @@ -74,7 +74,7 @@ late_initcall(start_lost_tick_compensation); * * -johnstul@us.ibm.com "math is hard, lets go shopping!" */ -static unsigned long cyc2ns_scale; +static unsigned long cyc2ns_scale __read_mostly; #define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ static inline void set_cyc2ns_scale(unsigned long cpu_khz) diff --git a/arch/i386/mm/hugetlbpage.c b/arch/i386/mm/hugetlbpage.c index d524127c9af..a7d89158541 100644 --- a/arch/i386/mm/hugetlbpage.c +++ b/arch/i386/mm/hugetlbpage.c @@ -48,18 +48,6 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return (pte_t *) pmd; } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - return 0; -} - #if 0 /* This is just for testing */ struct page * follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c index 2700f01994b..7ba55a6e2db 100644 --- a/arch/i386/mm/init.c +++ b/arch/i386/mm/init.c @@ -270,7 +270,7 @@ static void __init permanent_kmaps_init(pgd_t *pgd_base) static void __meminit free_new_highpage(struct page *page) { - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } @@ -727,7 +727,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); memset((void *)addr, 0xcc, PAGE_SIZE); free_page(addr); totalram_pages++; @@ -766,7 +766,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk (KERN_INFO "Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/i386/mm/pageattr.c b/arch/i386/mm/pageattr.c index d0cadb33b54..92c3d9f0e73 100644 --- a/arch/i386/mm/pageattr.c +++ b/arch/i386/mm/pageattr.c @@ -51,6 +51,13 @@ static struct page *split_large_page(unsigned long address, pgprot_t prot, if (!base) return NULL; + /* + * page_private is used to track the number of entries in + * the page table page that have non standard attributes. + */ + SetPagePrivate(base); + page_private(base) = 0; + address = __pa(address); addr = address & LARGE_PAGE_MASK; pbase = (pte_t *)page_address(base); @@ -143,11 +150,12 @@ __change_page_attr(struct page *page, pgprot_t prot) return -ENOMEM; set_pmd_pte(kpte,address,mk_pte(split, ref_prot)); kpte_page = split; - } - get_page(kpte_page); + } + page_private(kpte_page)++; } else if ((pte_val(*kpte) & _PAGE_PSE) == 0) { set_pte_atomic(kpte, mk_pte(page, PAGE_KERNEL)); - __put_page(kpte_page); + BUG_ON(page_private(kpte_page) == 0); + page_private(kpte_page)--; } else BUG(); @@ -157,10 +165,8 @@ __change_page_attr(struct page *page, pgprot_t prot) * replace it with a largepage. */ if (!PageReserved(kpte_page)) { - /* memleak and potential failed 2M page regeneration */ - BUG_ON(!page_count(kpte_page)); - - if (cpu_has_pse && (page_count(kpte_page) == 1)) { + if (cpu_has_pse && (page_private(kpte_page) == 0)) { + ClearPagePrivate(kpte_page); list_add(&kpte_page->lru, &df_list); revert_page(kpte_page, address); } diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index 2d13889d0a9..9dbc7dadd16 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -68,9 +68,10 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr) #define mk_pte_huge(entry) { pte_val(entry) |= _PAGE_P; } /* - * This function checks for proper alignment of input addr and len parameters. + * Don't actually need to do any preparation, but need to make sure + * the address is in the right region. */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(unsigned long addr, unsigned long len) { if (len & ~HPAGE_MASK) return -EINVAL; diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index b38b6d213c1..08d94e6bfa1 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -197,7 +197,7 @@ free_initmem (void) eaddr = (unsigned long) ia64_imva(__init_end); while (addr < eaddr) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); ++totalram_pages; addr += PAGE_SIZE; @@ -252,7 +252,7 @@ free_initrd_mem (unsigned long start, unsigned long end) continue; page = virt_to_page(start); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); free_page(start); ++totalram_pages; } @@ -640,7 +640,7 @@ mem_init (void) void online_page(struct page *page) { ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalram_pages++; num_physpages++; diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c index 6facf15b04f..c9e7dad860b 100644 --- a/arch/m32r/mm/init.c +++ b/arch/m32r/mm/init.c @@ -226,7 +226,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -244,7 +244,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) unsigned long p; for (p = start; p < end; p += PAGE_SIZE) { ClearPageReserved(virt_to_page(p)); - set_page_count(virt_to_page(p), 1); + init_page_count(virt_to_page(p)); free_page(p); totalram_pages++; } diff --git a/arch/m68k/mm/init.c b/arch/m68k/mm/init.c index c45beb95594..a190e39c907 100644 --- a/arch/m68k/mm/init.c +++ b/arch/m68k/mm/init.c @@ -137,7 +137,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) int pages = 0; for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; pages++; diff --git a/arch/m68k/mm/memory.c b/arch/m68k/mm/memory.c index 559942ce0e1..d6d582a5abb 100644 --- a/arch/m68k/mm/memory.c +++ b/arch/m68k/mm/memory.c @@ -54,7 +54,7 @@ void __init init_pointer_table(unsigned long ptable) /* unreserve the page so it's possible to free that page */ PD_PAGE(dp)->flags &= ~(1 << PG_reserved); - set_page_count(PD_PAGE(dp), 1); + init_page_count(PD_PAGE(dp)); return; } diff --git a/arch/m68k/mm/motorola.c b/arch/m68k/mm/motorola.c index d855fec2631..afb57eeafdc 100644 --- a/arch/m68k/mm/motorola.c +++ b/arch/m68k/mm/motorola.c @@ -276,7 +276,7 @@ void free_initmem(void) addr = (unsigned long)&__init_begin; for (; addr < (unsigned long)&__init_end; addr += PAGE_SIZE) { virt_to_page(addr)->flags &= ~(1 << PG_reserved); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } diff --git a/arch/m68knommu/kernel/m68k_ksyms.c b/arch/m68knommu/kernel/m68k_ksyms.c index eddb8d3e130..d844c755945 100644 --- a/arch/m68knommu/kernel/m68k_ksyms.c +++ b/arch/m68knommu/kernel/m68k_ksyms.c @@ -26,6 +26,7 @@ EXPORT_SYMBOL(__ioremap); EXPORT_SYMBOL(iounmap); EXPORT_SYMBOL(dump_fpu); EXPORT_SYMBOL(strnlen); +EXPORT_SYMBOL(strpbrk); EXPORT_SYMBOL(strrchr); EXPORT_SYMBOL(strstr); EXPORT_SYMBOL(strchr); diff --git a/arch/m68knommu/mm/init.c b/arch/m68knommu/mm/init.c index 89f0b554ffb..d79503fe6e4 100644 --- a/arch/m68knommu/mm/init.c +++ b/arch/m68knommu/mm/init.c @@ -195,7 +195,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) int pages = 0; for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; pages++; @@ -218,7 +218,7 @@ free_initmem() /* next to check that the page we free is not a partial page */ for (; addr + PAGE_SIZE < (unsigned long)(&__init_end); addr +=PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } diff --git a/arch/mips/arc/memory.c b/arch/mips/arc/memory.c index 958d2eb7886..8a9ef58cc39 100644 --- a/arch/mips/arc/memory.c +++ b/arch/mips/arc/memory.c @@ -158,7 +158,7 @@ unsigned long __init prom_free_prom_memory(void) while (addr < boot_mem_map.map[i].addr + boot_mem_map.map[i].size) { ClearPageReserved(virt_to_page(__va(addr))); - set_page_count(virt_to_page(__va(addr)), 1); + init_page_count(virt_to_page(__va(addr))); free_page((unsigned long)__va(addr)); addr += PAGE_SIZE; freed += PAGE_SIZE; diff --git a/arch/mips/dec/prom/memory.c b/arch/mips/dec/prom/memory.c index 81cb5a76cfb..1edaf3074ee 100644 --- a/arch/mips/dec/prom/memory.c +++ b/arch/mips/dec/prom/memory.c @@ -118,7 +118,7 @@ unsigned long __init prom_free_prom_memory(void) addr = PAGE_SIZE; while (addr < end) { ClearPageReserved(virt_to_page(__va(addr))); - set_page_count(virt_to_page(__va(addr)), 1); + init_page_count(virt_to_page(__va(addr))); free_page((unsigned long)__va(addr)); addr += PAGE_SIZE; } diff --git a/arch/mips/mips-boards/generic/memory.c b/arch/mips/mips-boards/generic/memory.c index 2c8afd77a20..ee5e70c95cf 100644 --- a/arch/mips/mips-boards/generic/memory.c +++ b/arch/mips/mips-boards/generic/memory.c @@ -174,7 +174,7 @@ unsigned long __init prom_free_prom_memory(void) while (addr < boot_mem_map.map[i].addr + boot_mem_map.map[i].size) { ClearPageReserved(virt_to_page(__va(addr))); - set_page_count(virt_to_page(__va(addr)), 1); + init_page_count(virt_to_page(__va(addr))); free_page((unsigned long)__va(addr)); addr += PAGE_SIZE; freed += PAGE_SIZE; diff --git a/arch/mips/mips-boards/sim/sim_mem.c b/arch/mips/mips-boards/sim/sim_mem.c index 0dbd7435bb2..1ec4e75656b 100644 --- a/arch/mips/mips-boards/sim/sim_mem.c +++ b/arch/mips/mips-boards/sim/sim_mem.c @@ -117,7 +117,7 @@ unsigned long __init prom_free_prom_memory(void) while (addr < boot_mem_map.map[i].addr + boot_mem_map.map[i].size) { ClearPageReserved(virt_to_page(__va(addr))); - set_page_count(virt_to_page(__va(addr)), 1); + init_page_count(virt_to_page(__va(addr))); free_page((unsigned long)__va(addr)); addr += PAGE_SIZE; freed += PAGE_SIZE; diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index 0ff9a348b84..52f7d59fe61 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -54,7 +54,8 @@ unsigned long empty_zero_page, zero_page_mask; */ unsigned long setup_zero_pages(void) { - unsigned long order, size; + unsigned int order; + unsigned long size; struct page *page; if (cpu_has_vce) @@ -67,9 +68,9 @@ unsigned long setup_zero_pages(void) panic("Oh boy, that early out of memory?"); page = virt_to_page(empty_zero_page); + split_page(page, order); while (page < virt_to_page(empty_zero_page + (PAGE_SIZE << order))) { SetPageReserved(page); - set_page_count(page, 1); page++; } @@ -244,7 +245,7 @@ void __init mem_init(void) #ifdef CONFIG_LIMITED_DMA set_page_address(page, lowmem_page_address(page)); #endif - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } @@ -291,7 +292,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } @@ -314,7 +315,7 @@ void free_initmem(void) page = addr; #endif ClearPageReserved(virt_to_page(page)); - set_page_count(virt_to_page(page), 1); + init_page_count(virt_to_page(page)); free_page(page); totalram_pages++; freed += PAGE_SIZE; diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c index ed93a979295..e0d095daa5e 100644 --- a/arch/mips/sgi-ip27/ip27-memory.c +++ b/arch/mips/sgi-ip27/ip27-memory.c @@ -559,7 +559,7 @@ void __init mem_init(void) /* if (!page_is_ram(pgnr)) continue; */ /* commented out until page_is_ram works */ ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); totalram_pages++; } diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c index 7847ca13d6c..852eda3953d 100644 --- a/arch/parisc/mm/init.c +++ b/arch/parisc/mm/init.c @@ -398,7 +398,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); num_physpages++; totalram_pages++; @@ -1018,7 +1018,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk(KERN_INFO "Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); num_physpages++; totalram_pages++; diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index b51bb28c054..7370f9f33e2 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -133,21 +133,6 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, return __pte(old); } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - if (! (within_hugepage_low_range(addr, len) - || within_hugepage_high_range(addr, len)) ) - return -EINVAL; - return 0; -} - struct slb_flush_info { struct mm_struct *mm; u16 newareas; diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c index 7d0d75c1184..b57fb3a2b7b 100644 --- a/arch/powerpc/mm/init_32.c +++ b/arch/powerpc/mm/init_32.c @@ -216,7 +216,7 @@ static void free_sec(unsigned long start, unsigned long end, const char *name) while (start < end) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); cnt++; start += PAGE_SIZE; @@ -248,7 +248,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index 81cfb0c2ec5..bacb71c8981 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -140,7 +140,7 @@ void free_initmem(void) for (; addr < (unsigned long)__init_end; addr += PAGE_SIZE) { memset((void *)addr, 0xcc, PAGE_SIZE); ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -155,7 +155,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index 550517c2dd4..454cac01d8c 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -108,8 +108,8 @@ EXPORT_SYMBOL(phys_mem_access_prot); void online_page(struct page *page) { ClearPageReserved(page); - set_page_count(page, 0); - free_cold_page(page); + init_page_count(page); + __free_page(page); totalram_pages++; num_physpages++; } @@ -376,7 +376,7 @@ void __init mem_init(void) struct page *page = pfn_to_page(pfn); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } diff --git a/arch/powerpc/platforms/cell/setup.c b/arch/powerpc/platforms/cell/setup.c index b33a4443f5a..fec8e65b36e 100644 --- a/arch/powerpc/platforms/cell/setup.c +++ b/arch/powerpc/platforms/cell/setup.c @@ -115,7 +115,7 @@ static void __init cell_spuprop_present(struct device_node *spe, for (pfn = start_pfn; pfn < end_pfn; pfn++) { struct page *page = pfn_to_page(pfn); set_page_links(page, ZONE_DMA, node_id, pfn); - set_page_count(page, 1); + init_page_count(page); reset_page_mapcount(page); SetPageReserved(page); INIT_LIST_HEAD(&page->lru); diff --git a/arch/ppc/kernel/dma-mapping.c b/arch/ppc/kernel/dma-mapping.c index 685fd0defe2..61465ec88bc 100644 --- a/arch/ppc/kernel/dma-mapping.c +++ b/arch/ppc/kernel/dma-mapping.c @@ -223,6 +223,8 @@ __dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) pte_t *pte = consistent_pte + CONSISTENT_OFFSET(vaddr); struct page *end = page + (1 << order); + split_page(page, order); + /* * Set the "dma handle" */ @@ -231,7 +233,6 @@ __dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) do { BUG_ON(!pte_none(*pte)); - set_page_count(page, 1); SetPageReserved(page); set_pte_at(&init_mm, vaddr, pte, mk_pte(page, pgprot_noncached(PAGE_KERNEL))); @@ -244,7 +245,6 @@ __dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) * Free the otherwise unused pages. */ while (page < end) { - set_page_count(page, 1); __free_page(page); page++; } diff --git a/arch/ppc/mm/init.c b/arch/ppc/mm/init.c index 134db5c0420..cb1c294fb93 100644 --- a/arch/ppc/mm/init.c +++ b/arch/ppc/mm/init.c @@ -140,7 +140,7 @@ static void free_sec(unsigned long start, unsigned long end, const char *name) while (start < end) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); cnt++; start += PAGE_SIZE; @@ -172,7 +172,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } @@ -441,7 +441,7 @@ void __init mem_init(void) struct page *page = mem_map + pfn; ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index df953383724..a055894f3bd 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -292,7 +292,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -307,7 +307,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/sh/mm/consistent.c b/arch/sh/mm/consistent.c index df3a9e452cc..ee73e30263a 100644 --- a/arch/sh/mm/consistent.c +++ b/arch/sh/mm/consistent.c @@ -23,6 +23,7 @@ void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle) page = alloc_pages(gfp, order); if (!page) return NULL; + split_page(page, order); ret = page_address(page); *handle = virt_to_phys(ret); @@ -37,8 +38,6 @@ void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle) end = page + (1 << order); while (++page < end) { - set_page_count(page, 1); - /* Free any unused pages */ if (page >= free) { __free_page(page); diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c index 6b7a7688c98..a3568fd5150 100644 --- a/arch/sh/mm/hugetlbpage.c +++ b/arch/sh/mm/hugetlbpage.c @@ -84,18 +84,6 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, return entry; } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - return 0; -} - struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c index e342565f75f..77b4a838fe1 100644 --- a/arch/sh/mm/init.c +++ b/arch/sh/mm/init.c @@ -273,7 +273,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -286,7 +286,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) unsigned long p; for (p = start; p < end; p += PAGE_SIZE) { ClearPageReserved(virt_to_page(p)); - set_page_count(virt_to_page(p), 1); + init_page_count(virt_to_page(p)); free_page(p); totalram_pages++; } diff --git a/arch/sh64/mm/hugetlbpage.c b/arch/sh64/mm/hugetlbpage.c index ed6a505b3ee..3d89f2a6c78 100644 --- a/arch/sh64/mm/hugetlbpage.c +++ b/arch/sh64/mm/hugetlbpage.c @@ -84,18 +84,6 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, return entry; } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - return 0; -} - struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { diff --git a/arch/sh64/mm/init.c b/arch/sh64/mm/init.c index a65e8bb2c3c..1169757fb38 100644 --- a/arch/sh64/mm/init.c +++ b/arch/sh64/mm/init.c @@ -173,7 +173,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -186,7 +186,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) unsigned long p; for (p = start; p < end; p += PAGE_SIZE) { ClearPageReserved(virt_to_page(p)); - set_page_count(virt_to_page(p), 1); + init_page_count(virt_to_page(p)); free_page(p); totalram_pages++; } diff --git a/arch/sparc/kernel/sun4d_smp.c b/arch/sparc/kernel/sun4d_smp.c index 40d426cce82..4219dd2ce3a 100644 --- a/arch/sparc/kernel/sun4d_smp.c +++ b/arch/sparc/kernel/sun4d_smp.c @@ -266,19 +266,19 @@ void __init smp4d_boot_cpus(void) /* Free unneeded trap tables */ ClearPageReserved(virt_to_page(trapbase_cpu1)); - set_page_count(virt_to_page(trapbase_cpu1), 1); + init_page_count(virt_to_page(trapbase_cpu1)); free_page((unsigned long)trapbase_cpu1); totalram_pages++; num_physpages++; ClearPageReserved(virt_to_page(trapbase_cpu2)); - set_page_count(virt_to_page(trapbase_cpu2), 1); + init_page_count(virt_to_page(trapbase_cpu2)); free_page((unsigned long)trapbase_cpu2); totalram_pages++; num_physpages++; ClearPageReserved(virt_to_page(trapbase_cpu3)); - set_page_count(virt_to_page(trapbase_cpu3), 1); + init_page_count(virt_to_page(trapbase_cpu3)); free_page((unsigned long)trapbase_cpu3); totalram_pages++; num_physpages++; diff --git a/arch/sparc/kernel/sun4m_smp.c b/arch/sparc/kernel/sun4m_smp.c index a21f27d10e5..fbbd8a474c4 100644 --- a/arch/sparc/kernel/sun4m_smp.c +++ b/arch/sparc/kernel/sun4m_smp.c @@ -233,21 +233,21 @@ void __init smp4m_boot_cpus(void) /* Free unneeded trap tables */ if (!cpu_isset(i, cpu_present_map)) { ClearPageReserved(virt_to_page(trapbase_cpu1)); - set_page_count(virt_to_page(trapbase_cpu1), 1); + init_page_count(virt_to_page(trapbase_cpu1)); free_page((unsigned long)trapbase_cpu1); totalram_pages++; num_physpages++; } if (!cpu_isset(2, cpu_present_map)) { ClearPageReserved(virt_to_page(trapbase_cpu2)); - set_page_count(virt_to_page(trapbase_cpu2), 1); + init_page_count(virt_to_page(trapbase_cpu2)); free_page((unsigned long)trapbase_cpu2); totalram_pages++; num_physpages++; } if (!cpu_isset(3, cpu_present_map)) { ClearPageReserved(virt_to_page(trapbase_cpu3)); - set_page_count(virt_to_page(trapbase_cpu3), 1); + init_page_count(virt_to_page(trapbase_cpu3)); free_page((unsigned long)trapbase_cpu3); totalram_pages++; num_physpages++; diff --git a/arch/sparc/mm/generic.c b/arch/sparc/mm/generic.c index 2cb0728cee0..1ef7fa03fef 100644 --- a/arch/sparc/mm/generic.c +++ b/arch/sparc/mm/generic.c @@ -76,7 +76,6 @@ int io_remap_pfn_range(struct vm_area_struct *vma, unsigned long from, vma->vm_pgoff = (offset >> PAGE_SHIFT) | ((unsigned long)space << 28UL); - prot = __pgprot(pg_iobits); offset -= from; dir = pgd_offset(mm, from); flush_cache_range(vma, beg, end); diff --git a/arch/sparc/mm/init.c b/arch/sparc/mm/init.c index c03babaa049..89866973246 100644 --- a/arch/sparc/mm/init.c +++ b/arch/sparc/mm/init.c @@ -383,7 +383,7 @@ void map_high_region(unsigned long start_pfn, unsigned long end_pfn) struct page *page = pfn_to_page(tmp); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } @@ -480,7 +480,7 @@ void free_initmem (void) p = virt_to_page(addr); ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); totalram_pages++; num_physpages++; @@ -497,7 +497,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) struct page *p = virt_to_page(start); ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); num_physpages++; } diff --git a/arch/sparc/mm/loadmmu.c b/arch/sparc/mm/loadmmu.c index e9f9571601b..36b4d24988f 100644 --- a/arch/sparc/mm/loadmmu.c +++ b/arch/sparc/mm/loadmmu.c @@ -22,8 +22,6 @@ struct ctx_list *ctx_list_pool; struct ctx_list ctx_free; struct ctx_list ctx_used; -unsigned int pg_iobits; - extern void ld_mmu_sun4c(void); extern void ld_mmu_srmmu(void); diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c index c664b962987..27b0e0ba858 100644 --- a/arch/sparc/mm/srmmu.c +++ b/arch/sparc/mm/srmmu.c @@ -2130,6 +2130,13 @@ static unsigned long srmmu_pte_to_pgoff(pte_t pte) return pte_val(pte) >> SRMMU_PTE_FILE_SHIFT; } +static pgprot_t srmmu_pgprot_noncached(pgprot_t prot) +{ + prot &= ~__pgprot(SRMMU_CACHE); + + return prot; +} + /* Load up routines and constants for sun4m and sun4d mmu */ void __init ld_mmu_srmmu(void) { @@ -2150,9 +2157,9 @@ void __init ld_mmu_srmmu(void) BTFIXUPSET_INT(page_readonly, pgprot_val(SRMMU_PAGE_RDONLY)); BTFIXUPSET_INT(page_kernel, pgprot_val(SRMMU_PAGE_KERNEL)); page_kernel = pgprot_val(SRMMU_PAGE_KERNEL); - pg_iobits = SRMMU_VALID | SRMMU_WRITE | SRMMU_REF; /* Functions */ + BTFIXUPSET_CALL(pgprot_noncached, srmmu_pgprot_noncached, BTFIXUPCALL_NORM); #ifndef CONFIG_SMP BTFIXUPSET_CALL(___xchg32, ___xchg32_sun4md, BTFIXUPCALL_SWAPG1G2); #endif diff --git a/arch/sparc/mm/sun4c.c b/arch/sparc/mm/sun4c.c index 731f19603ca..49f28c1bdc6 100644 --- a/arch/sparc/mm/sun4c.c +++ b/arch/sparc/mm/sun4c.c @@ -1589,7 +1589,10 @@ static void sun4c_flush_tlb_page(struct vm_area_struct *vma, unsigned long page) static inline void sun4c_mapioaddr(unsigned long physaddr, unsigned long virt_addr) { - unsigned long page_entry; + unsigned long page_entry, pg_iobits; + + pg_iobits = _SUN4C_PAGE_PRESENT | _SUN4C_READABLE | _SUN4C_WRITEABLE | + _SUN4C_PAGE_IO | _SUN4C_PAGE_NOCACHE; page_entry = ((physaddr >> PAGE_SHIFT) & SUN4C_PFN_MASK); page_entry |= ((pg_iobits | _SUN4C_PAGE_PRIV) & ~(_SUN4C_PAGE_PRESENT)); @@ -2134,6 +2137,13 @@ void __init sun4c_paging_init(void) printk("SUN4C: %d mmu entries for the kernel\n", cnt); } +static pgprot_t sun4c_pgprot_noncached(pgprot_t prot) +{ + prot |= __pgprot(_SUN4C_PAGE_IO | _SUN4C_PAGE_NOCACHE); + + return prot; +} + /* Load up routines and constants for sun4c mmu */ void __init ld_mmu_sun4c(void) { @@ -2156,10 +2166,9 @@ void __init ld_mmu_sun4c(void) BTFIXUPSET_INT(page_readonly, pgprot_val(SUN4C_PAGE_READONLY)); BTFIXUPSET_INT(page_kernel, pgprot_val(SUN4C_PAGE_KERNEL)); page_kernel = pgprot_val(SUN4C_PAGE_KERNEL); - pg_iobits = _SUN4C_PAGE_PRESENT | _SUN4C_READABLE | _SUN4C_WRITEABLE | - _SUN4C_PAGE_IO | _SUN4C_PAGE_NOCACHE; /* Functions */ + BTFIXUPSET_CALL(pgprot_noncached, sun4c_pgprot_noncached, BTFIXUPCALL_NORM); BTFIXUPSET_CALL(___xchg32, ___xchg32_sun4c, BTFIXUPCALL_NORM); BTFIXUPSET_CALL(do_check_pgt_cache, sun4c_check_pgt_cache, BTFIXUPCALL_NORM); diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig index c3685b314d7..267afddf63c 100644 --- a/arch/sparc64/Kconfig +++ b/arch/sparc64/Kconfig @@ -175,11 +175,11 @@ config HUGETLB_PAGE_SIZE_4MB bool "4MB" config HUGETLB_PAGE_SIZE_512K - depends on !SPARC64_PAGE_SIZE_4MB + depends on !SPARC64_PAGE_SIZE_4MB && !SPARC64_PAGE_SIZE_512KB bool "512K" config HUGETLB_PAGE_SIZE_64K - depends on !SPARC64_PAGE_SIZE_4MB && !SPARC64_PAGE_SIZE_512KB + depends on !SPARC64_PAGE_SIZE_4MB && !SPARC64_PAGE_SIZE_512KB && !SPARC64_PAGE_SIZE_64K bool "64K" endchoice diff --git a/arch/sparc64/kernel/pci.c b/arch/sparc64/kernel/pci.c index 95ffa941862..dfccff29e18 100644 --- a/arch/sparc64/kernel/pci.c +++ b/arch/sparc64/kernel/pci.c @@ -656,6 +656,7 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma, __pci_mmap_set_flags(dev, vma, mmap_state); __pci_mmap_set_pgprot(dev, vma, mmap_state); + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); ret = io_remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff, vma->vm_end - vma->vm_start, @@ -663,7 +664,6 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma, if (ret) return ret; - vma->vm_flags |= VM_IO; return 0; } diff --git a/arch/sparc64/kernel/sun4v_tlb_miss.S b/arch/sparc64/kernel/sun4v_tlb_miss.S index ab23ddb7116..b731881224e 100644 --- a/arch/sparc64/kernel/sun4v_tlb_miss.S +++ b/arch/sparc64/kernel/sun4v_tlb_miss.S @@ -29,15 +29,15 @@ * * index_mask = (512 << (tsb_reg & 0x7UL)) - 1UL; * tsb_base = tsb_reg & ~0x7UL; - * tsb_index = ((vaddr >> PAGE_SHIFT) & tsb_mask); + * tsb_index = ((vaddr >> HASH_SHIFT) & tsb_mask); * tsb_ptr = tsb_base + (tsb_index * 16); */ -#define COMPUTE_TSB_PTR(TSB_PTR, VADDR, TMP1, TMP2) \ +#define COMPUTE_TSB_PTR(TSB_PTR, VADDR, HASH_SHIFT, TMP1, TMP2) \ and TSB_PTR, 0x7, TMP1; \ mov 512, TMP2; \ andn TSB_PTR, 0x7, TSB_PTR; \ sllx TMP2, TMP1, TMP2; \ - srlx VADDR, PAGE_SHIFT, TMP1; \ + srlx VADDR, HASH_SHIFT, TMP1; \ sub TMP2, 1, TMP2; \ and TMP1, TMP2, TMP1; \ sllx TMP1, 4, TMP1; \ @@ -53,7 +53,7 @@ sun4v_itlb_miss: LOAD_ITLB_INFO(%g2, %g4, %g5) COMPUTE_TAG_TARGET(%g6, %g4, %g5, kvmap_itlb_4v) - COMPUTE_TSB_PTR(%g1, %g4, %g3, %g7) + COMPUTE_TSB_PTR(%g1, %g4, PAGE_SHIFT, %g3, %g7) /* Load TSB tag/pte into %g2/%g3 and compare the tag. */ ldda [%g1] ASI_QUAD_LDD_PHYS_4V, %g2 @@ -99,7 +99,7 @@ sun4v_dtlb_miss: LOAD_DTLB_INFO(%g2, %g4, %g5) COMPUTE_TAG_TARGET(%g6, %g4, %g5, kvmap_dtlb_4v) - COMPUTE_TSB_PTR(%g1, %g4, %g3, %g7) + COMPUTE_TSB_PTR(%g1, %g4, PAGE_SHIFT, %g3, %g7) /* Load TSB tag/pte into %g2/%g3 and compare the tag. */ ldda [%g1] ASI_QUAD_LDD_PHYS_4V, %g2 @@ -171,21 +171,26 @@ sun4v_dtsb_miss: /* fallthrough */ - /* Create TSB pointer into %g1. This is something like: - * - * index_mask = (512 << (tsb_reg & 0x7UL)) - 1UL; - * tsb_base = tsb_reg & ~0x7UL; - * tsb_index = ((vaddr >> PAGE_SHIFT) & tsb_mask); - * tsb_ptr = tsb_base + (tsb_index * 16); - */ sun4v_tsb_miss_common: - COMPUTE_TSB_PTR(%g1, %g4, %g5, %g7) + COMPUTE_TSB_PTR(%g1, %g4, PAGE_SHIFT, %g5, %g7) - /* Branch directly to page table lookup. We have SCRATCHPAD_MMU_MISS - * still in %g2, so it's quite trivial to get at the PGD PHYS value - * so we can preload it into %g7. - */ sub %g2, TRAP_PER_CPU_FAULT_INFO, %g2 + +#ifdef CONFIG_HUGETLB_PAGE + mov SCRATCHPAD_UTSBREG2, %g5 + ldxa [%g5] ASI_SCRATCHPAD, %g5 + cmp %g5, -1 + be,pt %xcc, 80f + nop + COMPUTE_TSB_PTR(%g5, %g4, HPAGE_SHIFT, %g2, %g7) + + /* That clobbered %g2, reload it. */ + ldxa [%g0] ASI_SCRATCHPAD, %g2 + sub %g2, TRAP_PER_CPU_FAULT_INFO, %g2 + +80: stx %g5, [%g2 + TRAP_PER_CPU_TSB_HUGE_TEMP] +#endif + ba,pt %xcc, tsb_miss_page_table_walk_sun4v_fastpath ldx [%g2 + TRAP_PER_CPU_PGD_PADDR], %g7 diff --git a/arch/sparc64/kernel/traps.c b/arch/sparc64/kernel/traps.c index 7f7dba0ca96..df612e4f75f 100644 --- a/arch/sparc64/kernel/traps.c +++ b/arch/sparc64/kernel/traps.c @@ -2482,6 +2482,7 @@ void init_cur_cpu_trap(struct thread_info *t) extern void thread_info_offsets_are_bolixed_dave(void); extern void trap_per_cpu_offsets_are_bolixed_dave(void); +extern void tsb_config_offsets_are_bolixed_dave(void); /* Only invoked on boot processor. */ void __init trap_init(void) @@ -2535,9 +2536,27 @@ void __init trap_init(void) (TRAP_PER_CPU_CPU_MONDO_BLOCK_PA != offsetof(struct trap_per_cpu, cpu_mondo_block_pa)) || (TRAP_PER_CPU_CPU_LIST_PA != - offsetof(struct trap_per_cpu, cpu_list_pa))) + offsetof(struct trap_per_cpu, cpu_list_pa)) || + (TRAP_PER_CPU_TSB_HUGE != + offsetof(struct trap_per_cpu, tsb_huge)) || + (TRAP_PER_CPU_TSB_HUGE_TEMP != + offsetof(struct trap_per_cpu, tsb_huge_temp))) trap_per_cpu_offsets_are_bolixed_dave(); + if ((TSB_CONFIG_TSB != + offsetof(struct tsb_config, tsb)) || + (TSB_CONFIG_RSS_LIMIT != + offsetof(struct tsb_config, tsb_rss_limit)) || + (TSB_CONFIG_NENTRIES != + offsetof(struct tsb_config, tsb_nentries)) || + (TSB_CONFIG_REG_VAL != + offsetof(struct tsb_config, tsb_reg_val)) || + (TSB_CONFIG_MAP_VADDR != + offsetof(struct tsb_config, tsb_map_vaddr)) || + (TSB_CONFIG_MAP_PTE != + offsetof(struct tsb_config, tsb_map_pte))) + tsb_config_offsets_are_bolixed_dave(); + /* Attach to the address space of init_task. On SMP we * do this in smp.c:smp_callin for other cpus. */ diff --git a/arch/sparc64/kernel/tsb.S b/arch/sparc64/kernel/tsb.S index 118baea44f6..a0c8ba58920 100644 --- a/arch/sparc64/kernel/tsb.S +++ b/arch/sparc64/kernel/tsb.S @@ -3,8 +3,13 @@ * Copyright (C) 2006 David S. Miller <davem@davemloft.net> */ +#include <linux/config.h> + #include <asm/tsb.h> #include <asm/hypervisor.h> +#include <asm/page.h> +#include <asm/cpudata.h> +#include <asm/mmu.h> .text .align 32 @@ -34,34 +39,124 @@ tsb_miss_itlb: ldxa [%g4] ASI_IMMU, %g4 /* At this point we have: - * %g1 -- TSB entry address + * %g1 -- PAGE_SIZE TSB entry address * %g3 -- FAULT_CODE_{D,I}TLB * %g4 -- missing virtual address * %g6 -- TAG TARGET (vaddr >> 22) */ tsb_miss_page_table_walk: - TRAP_LOAD_PGD_PHYS(%g7, %g5) + TRAP_LOAD_TRAP_BLOCK(%g7, %g5) - /* And now we have the PGD base physical address in %g7. */ -tsb_miss_page_table_walk_sun4v_fastpath: - USER_PGTABLE_WALK_TL1(%g4, %g7, %g5, %g2, tsb_do_fault) + /* Before committing to a full page table walk, + * check the huge page TSB. + */ +#ifdef CONFIG_HUGETLB_PAGE + +661: ldx [%g7 + TRAP_PER_CPU_TSB_HUGE], %g5 + nop + .section .sun4v_2insn_patch, "ax" + .word 661b + mov SCRATCHPAD_UTSBREG2, %g5 + ldxa [%g5] ASI_SCRATCHPAD, %g5 + .previous + + cmp %g5, -1 + be,pt %xcc, 80f + nop + + /* We need an aligned pair of registers containing 2 values + * which can be easily rematerialized. %g6 and %g7 foot the + * bill just nicely. We'll save %g6 away into %g2 for the + * huge page TSB TAG comparison. + * + * Perform a huge page TSB lookup. + */ + mov %g6, %g2 + and %g5, 0x7, %g6 + mov 512, %g7 + andn %g5, 0x7, %g5 + sllx %g7, %g6, %g7 + srlx %g4, HPAGE_SHIFT, %g6 + sub %g7, 1, %g7 + and %g6, %g7, %g6 + sllx %g6, 4, %g6 + add %g5, %g6, %g5 + + TSB_LOAD_QUAD(%g5, %g6) + cmp %g6, %g2 + be,a,pt %xcc, tsb_tlb_reload + mov %g7, %g5 + + /* No match, remember the huge page TSB entry address, + * and restore %g6 and %g7. + */ + TRAP_LOAD_TRAP_BLOCK(%g7, %g6) + srlx %g4, 22, %g6 +80: stx %g5, [%g7 + TRAP_PER_CPU_TSB_HUGE_TEMP] + +#endif + + ldx [%g7 + TRAP_PER_CPU_PGD_PADDR], %g7 /* At this point we have: * %g1 -- TSB entry address * %g3 -- FAULT_CODE_{D,I}TLB - * %g5 -- physical address of PTE in Linux page tables + * %g4 -- missing virtual address * %g6 -- TAG TARGET (vaddr >> 22) + * %g7 -- page table physical address + * + * We know that both the base PAGE_SIZE TSB and the HPAGE_SIZE + * TSB both lack a matching entry. */ -tsb_reload: - TSB_LOCK_TAG(%g1, %g2, %g7) +tsb_miss_page_table_walk_sun4v_fastpath: + USER_PGTABLE_WALK_TL1(%g4, %g7, %g5, %g2, tsb_do_fault) /* Load and check PTE. */ ldxa [%g5] ASI_PHYS_USE_EC, %g5 - mov 1, %g7 - sllx %g7, TSB_TAG_INVALID_BIT, %g7 - brgez,a,pn %g5, tsb_do_fault - TSB_STORE(%g1, %g7) + brgez,pn %g5, tsb_do_fault + nop + +#ifdef CONFIG_HUGETLB_PAGE +661: sethi %uhi(_PAGE_SZALL_4U), %g7 + sllx %g7, 32, %g7 + .section .sun4v_2insn_patch, "ax" + .word 661b + mov _PAGE_SZALL_4V, %g7 + nop + .previous + + and %g5, %g7, %g2 + +661: sethi %uhi(_PAGE_SZHUGE_4U), %g7 + sllx %g7, 32, %g7 + .section .sun4v_2insn_patch, "ax" + .word 661b + mov _PAGE_SZHUGE_4V, %g7 + nop + .previous + + cmp %g2, %g7 + bne,pt %xcc, 60f + nop + + /* It is a huge page, use huge page TSB entry address we + * calculated above. + */ + TRAP_LOAD_TRAP_BLOCK(%g7, %g2) + ldx [%g7 + TRAP_PER_CPU_TSB_HUGE_TEMP], %g2 + cmp %g2, -1 + movne %xcc, %g2, %g1 +60: +#endif + /* At this point we have: + * %g1 -- TSB entry address + * %g3 -- FAULT_CODE_{D,I}TLB + * %g5 -- valid PTE + * %g6 -- TAG TARGET (vaddr >> 22) + */ +tsb_reload: + TSB_LOCK_TAG(%g1, %g2, %g7) TSB_WRITE(%g1, %g5, %g6) /* Finally, load TLB and return from trap. */ @@ -240,10 +335,9 @@ tsb_flush: * schedule() time. * * %o0: page table physical address - * %o1: TSB register value - * %o2: TSB virtual address - * %o3: TSB mapping locked PTE - * %o4: Hypervisor TSB descriptor physical address + * %o1: TSB base config pointer + * %o2: TSB huge config pointer, or NULL if none + * %o3: Hypervisor TSB descriptor physical address * * We have to run this whole thing with interrupts * disabled so that the current cpu doesn't change @@ -253,63 +347,79 @@ tsb_flush: .globl __tsb_context_switch .type __tsb_context_switch,#function __tsb_context_switch: - rdpr %pstate, %o5 - wrpr %o5, PSTATE_IE, %pstate + rdpr %pstate, %g1 + wrpr %g1, PSTATE_IE, %pstate + + TRAP_LOAD_TRAP_BLOCK(%g2, %g3) - ldub [%g6 + TI_CPU], %g1 - sethi %hi(trap_block), %g2 - sllx %g1, TRAP_BLOCK_SZ_SHIFT, %g1 - or %g2, %lo(trap_block), %g2 - add %g2, %g1, %g2 stx %o0, [%g2 + TRAP_PER_CPU_PGD_PADDR] - sethi %hi(tlb_type), %g1 - lduw [%g1 + %lo(tlb_type)], %g1 - cmp %g1, 3 - bne,pt %icc, 1f + ldx [%o1 + TSB_CONFIG_REG_VAL], %o0 + brz,pt %o2, 1f + mov -1, %g3 + + ldx [%o2 + TSB_CONFIG_REG_VAL], %g3 + +1: stx %g3, [%g2 + TRAP_PER_CPU_TSB_HUGE] + + sethi %hi(tlb_type), %g2 + lduw [%g2 + %lo(tlb_type)], %g2 + cmp %g2, 3 + bne,pt %icc, 50f nop /* Hypervisor TSB switch. */ - mov SCRATCHPAD_UTSBREG1, %g1 - stxa %o1, [%g1] ASI_SCRATCHPAD - mov -1, %g2 - mov SCRATCHPAD_UTSBREG2, %g1 - stxa %g2, [%g1] ASI_SCRATCHPAD - - /* Save away %o5's %pstate, we have to use %o5 for - * the hypervisor call. - */ - mov %o5, %g1 + mov SCRATCHPAD_UTSBREG1, %o5 + stxa %o0, [%o5] ASI_SCRATCHPAD + mov SCRATCHPAD_UTSBREG2, %o5 + stxa %g3, [%o5] ASI_SCRATCHPAD + + mov 2, %o0 + cmp %g3, -1 + move %xcc, 1, %o0 mov HV_FAST_MMU_TSB_CTXNON0, %o5 - mov 1, %o0 - mov %o4, %o1 + mov %o3, %o1 ta HV_FAST_TRAP - /* Finish up and restore %o5. */ + /* Finish up. */ ba,pt %xcc, 9f - mov %g1, %o5 + nop /* SUN4U TSB switch. */ -1: mov TSB_REG, %g1 - stxa %o1, [%g1] ASI_DMMU +50: mov TSB_REG, %o5 + stxa %o0, [%o5] ASI_DMMU membar #Sync - stxa %o1, [%g1] ASI_IMMU + stxa %o0, [%o5] ASI_IMMU membar #Sync -2: brz %o2, 9f - nop +2: ldx [%o1 + TSB_CONFIG_MAP_VADDR], %o4 + brz %o4, 9f + ldx [%o1 + TSB_CONFIG_MAP_PTE], %o5 sethi %hi(sparc64_highest_unlocked_tlb_ent), %g2 - mov TLB_TAG_ACCESS, %g1 + mov TLB_TAG_ACCESS, %g3 lduw [%g2 + %lo(sparc64_highest_unlocked_tlb_ent)], %g2 - stxa %o2, [%g1] ASI_DMMU + stxa %o4, [%g3] ASI_DMMU membar #Sync sllx %g2, 3, %g2 - stxa %o3, [%g2] ASI_DTLB_DATA_ACCESS + stxa %o5, [%g2] ASI_DTLB_DATA_ACCESS + membar #Sync + + brz,pt %o2, 9f + nop + + ldx [%o2 + TSB_CONFIG_MAP_VADDR], %o4 + ldx [%o2 + TSB_CONFIG_MAP_PTE], %o5 + mov TLB_TAG_ACCESS, %g3 + stxa %o4, [%g3] ASI_DMMU + membar #Sync + sub %g2, (1 << 3), %g2 + stxa %o5, [%g2] ASI_DTLB_DATA_ACCESS membar #Sync + 9: - wrpr %o5, %pstate + wrpr %g1, %pstate retl nop diff --git a/arch/sparc64/mm/fault.c b/arch/sparc64/mm/fault.c index 63b6cc0cd5d..d21ff3230c0 100644 --- a/arch/sparc64/mm/fault.c +++ b/arch/sparc64/mm/fault.c @@ -410,9 +410,18 @@ good_area: up_read(&mm->mmap_sem); mm_rss = get_mm_rss(mm); - if (unlikely(mm_rss >= mm->context.tsb_rss_limit)) - tsb_grow(mm, mm_rss); - +#ifdef CONFIG_HUGETLB_PAGE + mm_rss -= (mm->context.huge_pte_count * (HPAGE_SIZE / PAGE_SIZE)); +#endif + if (unlikely(mm_rss >= + mm->context.tsb_block[MM_TSB_BASE].tsb_rss_limit)) + tsb_grow(mm, MM_TSB_BASE, mm_rss); +#ifdef CONFIG_HUGETLB_PAGE + mm_rss = mm->context.huge_pte_count; + if (unlikely(mm_rss >= + mm->context.tsb_block[MM_TSB_HUGE].tsb_rss_limit)) + tsb_grow(mm, MM_TSB_HUGE, mm_rss); +#endif return; /* diff --git a/arch/sparc64/mm/generic.c b/arch/sparc64/mm/generic.c index 5fc5c579e35..8cb06205d26 100644 --- a/arch/sparc64/mm/generic.c +++ b/arch/sparc64/mm/generic.c @@ -140,7 +140,6 @@ int io_remap_pfn_range(struct vm_area_struct *vma, unsigned long from, vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP; vma->vm_pgoff = phys_base >> PAGE_SHIFT; - prot = __pgprot(pg_iobits); offset -= from; dir = pgd_offset(mm, from); flush_cache_range(vma, beg, end); diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c index a7a24869d04..074620d413d 100644 --- a/arch/sparc64/mm/hugetlbpage.c +++ b/arch/sparc64/mm/hugetlbpage.c @@ -199,13 +199,11 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) pte_t *pte = NULL; pgd = pgd_offset(mm, addr); - if (pgd) { - pud = pud_offset(pgd, addr); - if (pud) { - pmd = pmd_alloc(mm, pud, addr); - if (pmd) - pte = pte_alloc_map(mm, pmd, addr); - } + pud = pud_alloc(mm, pgd, addr); + if (pud) { + pmd = pmd_alloc(mm, pud, addr); + if (pmd) + pte = pte_alloc_map(mm, pmd, addr); } return pte; } @@ -231,13 +229,14 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return pte; } -#define mk_pte_huge(entry) do { pte_val(entry) |= _PAGE_SZHUGE; } while (0) - void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t entry) { int i; + if (!pte_present(*ptep) && pte_present(entry)) + mm->context.huge_pte_count++; + for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { set_pte_at(mm, addr, ptep, entry); ptep++; @@ -253,6 +252,8 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, int i; entry = *ptep; + if (pte_present(entry)) + mm->context.huge_pte_count--; for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { pte_clear(mm, addr, ptep); @@ -263,18 +264,6 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, return entry; } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - return 0; -} - struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { @@ -302,6 +291,15 @@ static void context_reload(void *__data) void hugetlb_prefault_arch_hook(struct mm_struct *mm) { + struct tsb_config *tp = &mm->context.tsb_block[MM_TSB_HUGE]; + + if (likely(tp->tsb != NULL)) + return; + + tsb_grow(mm, MM_TSB_HUGE, 0); + tsb_context_switch(mm); + smp_tsb_sync(mm); + /* On UltraSPARC-III+ and later, configure the second half of * the Data-TLB for huge pages. */ diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c index c2b556106fc..ded63ee9c4f 100644 --- a/arch/sparc64/mm/init.c +++ b/arch/sparc64/mm/init.c @@ -283,6 +283,7 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t p struct mm_struct *mm; struct tsb *tsb; unsigned long tag, flags; + unsigned long tsb_index, tsb_hash_shift; if (tlb_type != hypervisor) { unsigned long pfn = pte_pfn(pte); @@ -312,10 +313,26 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t p mm = vma->vm_mm; + tsb_index = MM_TSB_BASE; + tsb_hash_shift = PAGE_SHIFT; + spin_lock_irqsave(&mm->context.lock, flags); - tsb = &mm->context.tsb[(address >> PAGE_SHIFT) & - (mm->context.tsb_nentries - 1UL)]; +#ifdef CONFIG_HUGETLB_PAGE + if (mm->context.tsb_block[MM_TSB_HUGE].tsb != NULL) { + if ((tlb_type == hypervisor && + (pte_val(pte) & _PAGE_SZALL_4V) == _PAGE_SZHUGE_4V) || + (tlb_type != hypervisor && + (pte_val(pte) & _PAGE_SZALL_4U) == _PAGE_SZHUGE_4U)) { + tsb_index = MM_TSB_HUGE; + tsb_hash_shift = HPAGE_SHIFT; + } + } +#endif + + tsb = mm->context.tsb_block[tsb_index].tsb; + tsb += ((address >> tsb_hash_shift) & + (mm->context.tsb_block[tsb_index].tsb_nentries - 1UL)); tag = (address >> 22UL); tsb_insert(tsb, tag, pte_val(pte)); @@ -1461,7 +1478,7 @@ void free_initmem(void) p = virt_to_page(page); ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); num_physpages++; totalram_pages++; @@ -1477,7 +1494,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) struct page *p = virt_to_page(start); ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); num_physpages++; totalram_pages++; diff --git a/arch/sparc64/mm/tsb.c b/arch/sparc64/mm/tsb.c index b2064e2a44d..beaa02810f0 100644 --- a/arch/sparc64/mm/tsb.c +++ b/arch/sparc64/mm/tsb.c @@ -15,9 +15,9 @@ extern struct tsb swapper_tsb[KERNEL_TSB_NENTRIES]; -static inline unsigned long tsb_hash(unsigned long vaddr, unsigned long nentries) +static inline unsigned long tsb_hash(unsigned long vaddr, unsigned long hash_shift, unsigned long nentries) { - vaddr >>= PAGE_SHIFT; + vaddr >>= hash_shift; return vaddr & (nentries - 1); } @@ -36,7 +36,8 @@ void flush_tsb_kernel_range(unsigned long start, unsigned long end) unsigned long v; for (v = start; v < end; v += PAGE_SIZE) { - unsigned long hash = tsb_hash(v, KERNEL_TSB_NENTRIES); + unsigned long hash = tsb_hash(v, PAGE_SHIFT, + KERNEL_TSB_NENTRIES); struct tsb *ent = &swapper_tsb[hash]; if (tag_compare(ent->tag, v)) { @@ -46,49 +47,91 @@ void flush_tsb_kernel_range(unsigned long start, unsigned long end) } } -void flush_tsb_user(struct mmu_gather *mp) +static void __flush_tsb_one(struct mmu_gather *mp, unsigned long hash_shift, unsigned long tsb, unsigned long nentries) { - struct mm_struct *mm = mp->mm; - unsigned long nentries, base, flags; - struct tsb *tsb; - int i; - - spin_lock_irqsave(&mm->context.lock, flags); - - tsb = mm->context.tsb; - nentries = mm->context.tsb_nentries; + unsigned long i; - if (tlb_type == cheetah_plus || tlb_type == hypervisor) - base = __pa(tsb); - else - base = (unsigned long) tsb; - for (i = 0; i < mp->tlb_nr; i++) { unsigned long v = mp->vaddrs[i]; unsigned long tag, ent, hash; v &= ~0x1UL; - hash = tsb_hash(v, nentries); - ent = base + (hash * sizeof(struct tsb)); + hash = tsb_hash(v, hash_shift, nentries); + ent = tsb + (hash * sizeof(struct tsb)); tag = (v >> 22UL); tsb_flush(ent, tag); } +} + +void flush_tsb_user(struct mmu_gather *mp) +{ + struct mm_struct *mm = mp->mm; + unsigned long nentries, base, flags; + + spin_lock_irqsave(&mm->context.lock, flags); + base = (unsigned long) mm->context.tsb_block[MM_TSB_BASE].tsb; + nentries = mm->context.tsb_block[MM_TSB_BASE].tsb_nentries; + if (tlb_type == cheetah_plus || tlb_type == hypervisor) + base = __pa(base); + __flush_tsb_one(mp, PAGE_SHIFT, base, nentries); + +#ifdef CONFIG_HUGETLB_PAGE + if (mm->context.tsb_block[MM_TSB_HUGE].tsb) { + base = (unsigned long) mm->context.tsb_block[MM_TSB_HUGE].tsb; + nentries = mm->context.tsb_block[MM_TSB_HUGE].tsb_nentries; + if (tlb_type == cheetah_plus || tlb_type == hypervisor) + base = __pa(base); + __flush_tsb_one(mp, HPAGE_SHIFT, base, nentries); + } +#endif spin_unlock_irqrestore(&mm->context.lock, flags); } -static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_bytes) +#if defined(CONFIG_SPARC64_PAGE_SIZE_8KB) +#define HV_PGSZ_IDX_BASE HV_PGSZ_IDX_8K +#define HV_PGSZ_MASK_BASE HV_PGSZ_MASK_8K +#elif defined(CONFIG_SPARC64_PAGE_SIZE_64KB) +#define HV_PGSZ_IDX_BASE HV_PGSZ_IDX_64K +#define HV_PGSZ_MASK_BASE HV_PGSZ_MASK_64K +#elif defined(CONFIG_SPARC64_PAGE_SIZE_512KB) +#define HV_PGSZ_IDX_BASE HV_PGSZ_IDX_512K +#define HV_PGSZ_MASK_BASE HV_PGSZ_MASK_512K +#elif defined(CONFIG_SPARC64_PAGE_SIZE_4MB) +#define HV_PGSZ_IDX_BASE HV_PGSZ_IDX_4MB +#define HV_PGSZ_MASK_BASE HV_PGSZ_MASK_4MB +#else +#error Broken base page size setting... +#endif + +#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE_SIZE_64K) +#define HV_PGSZ_IDX_HUGE HV_PGSZ_IDX_64K +#define HV_PGSZ_MASK_HUGE HV_PGSZ_MASK_64K +#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512K) +#define HV_PGSZ_IDX_HUGE HV_PGSZ_IDX_512K +#define HV_PGSZ_MASK_HUGE HV_PGSZ_MASK_512K +#elif defined(CONFIG_HUGETLB_PAGE_SIZE_4MB) +#define HV_PGSZ_IDX_HUGE HV_PGSZ_IDX_4MB +#define HV_PGSZ_MASK_HUGE HV_PGSZ_MASK_4MB +#else +#error Broken huge page size setting... +#endif +#endif + +static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_idx, unsigned long tsb_bytes) { unsigned long tsb_reg, base, tsb_paddr; unsigned long page_sz, tte; - mm->context.tsb_nentries = tsb_bytes / sizeof(struct tsb); + mm->context.tsb_block[tsb_idx].tsb_nentries = + tsb_bytes / sizeof(struct tsb); base = TSBMAP_BASE; tte = pgprot_val(PAGE_KERNEL_LOCKED); - tsb_paddr = __pa(mm->context.tsb); + tsb_paddr = __pa(mm->context.tsb_block[tsb_idx].tsb); BUG_ON(tsb_paddr & (tsb_bytes - 1UL)); /* Use the smallest page size that can map the whole TSB @@ -147,61 +190,49 @@ static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_bytes) /* Physical mapping, no locked TLB entry for TSB. */ tsb_reg |= tsb_paddr; - mm->context.tsb_reg_val = tsb_reg; - mm->context.tsb_map_vaddr = 0; - mm->context.tsb_map_pte = 0; + mm->context.tsb_block[tsb_idx].tsb_reg_val = tsb_reg; + mm->context.tsb_block[tsb_idx].tsb_map_vaddr = 0; + mm->context.tsb_block[tsb_idx].tsb_map_pte = 0; } else { tsb_reg |= base; tsb_reg |= (tsb_paddr & (page_sz - 1UL)); tte |= (tsb_paddr & ~(page_sz - 1UL)); - mm->context.tsb_reg_val = tsb_reg; - mm->context.tsb_map_vaddr = base; - mm->context.tsb_map_pte = tte; + mm->context.tsb_block[tsb_idx].tsb_reg_val = tsb_reg; + mm->context.tsb_block[tsb_idx].tsb_map_vaddr = base; + mm->context.tsb_block[tsb_idx].tsb_map_pte = tte; } /* Setup the Hypervisor TSB descriptor. */ if (tlb_type == hypervisor) { - struct hv_tsb_descr *hp = &mm->context.tsb_descr; + struct hv_tsb_descr *hp = &mm->context.tsb_descr[tsb_idx]; - switch (PAGE_SIZE) { - case 8192: - default: - hp->pgsz_idx = HV_PGSZ_IDX_8K; + switch (tsb_idx) { + case MM_TSB_BASE: + hp->pgsz_idx = HV_PGSZ_IDX_BASE; break; - - case 64 * 1024: - hp->pgsz_idx = HV_PGSZ_IDX_64K; - break; - - case 512 * 1024: - hp->pgsz_idx = HV_PGSZ_IDX_512K; - break; - - case 4 * 1024 * 1024: - hp->pgsz_idx = HV_PGSZ_IDX_4MB; +#ifdef CONFIG_HUGETLB_PAGE + case MM_TSB_HUGE: + hp->pgsz_idx = HV_PGSZ_IDX_HUGE; break; +#endif + default: + BUG(); }; hp->assoc = 1; hp->num_ttes = tsb_bytes / 16; hp->ctx_idx = 0; - switch (PAGE_SIZE) { - case 8192: - default: - hp->pgsz_mask = HV_PGSZ_MASK_8K; - break; - - case 64 * 1024: - hp->pgsz_mask = HV_PGSZ_MASK_64K; - break; - - case 512 * 1024: - hp->pgsz_mask = HV_PGSZ_MASK_512K; + switch (tsb_idx) { + case MM_TSB_BASE: + hp->pgsz_mask = HV_PGSZ_MASK_BASE; break; - - case 4 * 1024 * 1024: - hp->pgsz_mask = HV_PGSZ_MASK_4MB; +#ifdef CONFIG_HUGETLB_PAGE + case MM_TSB_HUGE: + hp->pgsz_mask = HV_PGSZ_MASK_HUGE; break; +#endif + default: + BUG(); }; hp->tsb_base = tsb_paddr; hp->resv = 0; @@ -241,11 +272,11 @@ void __init tsb_cache_init(void) } } -/* When the RSS of an address space exceeds mm->context.tsb_rss_limit, - * do_sparc64_fault() invokes this routine to try and grow the TSB. +/* When the RSS of an address space exceeds tsb_rss_limit for a TSB, + * do_sparc64_fault() invokes this routine to try and grow it. * * When we reach the maximum TSB size supported, we stick ~0UL into - * mm->context.tsb_rss_limit so the grow checks in update_mmu_cache() + * tsb_rss_limit for that TSB so the grow checks in do_sparc64_fault() * will not trigger any longer. * * The TSB can be anywhere from 8K to 1MB in size, in increasing powers @@ -257,7 +288,7 @@ void __init tsb_cache_init(void) * the number of entries that the current TSB can hold at once. Currently, * we trigger when the RSS hits 3/4 of the TSB capacity. */ -void tsb_grow(struct mm_struct *mm, unsigned long rss) +void tsb_grow(struct mm_struct *mm, unsigned long tsb_index, unsigned long rss) { unsigned long max_tsb_size = 1 * 1024 * 1024; unsigned long new_size, old_size, flags; @@ -297,7 +328,8 @@ retry_tsb_alloc: * down to a 0-order allocation and force no TSB * growing for this address space. */ - if (mm->context.tsb == NULL && new_cache_index > 0) { + if (mm->context.tsb_block[tsb_index].tsb == NULL && + new_cache_index > 0) { new_cache_index = 0; new_size = 8192; new_rss_limit = ~0UL; @@ -307,8 +339,8 @@ retry_tsb_alloc: /* If we failed on a TSB grow, we are under serious * memory pressure so don't try to grow any more. */ - if (mm->context.tsb != NULL) - mm->context.tsb_rss_limit = ~0UL; + if (mm->context.tsb_block[tsb_index].tsb != NULL) + mm->context.tsb_block[tsb_index].tsb_rss_limit = ~0UL; return; } @@ -339,23 +371,26 @@ retry_tsb_alloc: */ spin_lock_irqsave(&mm->context.lock, flags); - old_tsb = mm->context.tsb; - old_cache_index = (mm->context.tsb_reg_val & 0x7UL); - old_size = mm->context.tsb_nentries * sizeof(struct tsb); + old_tsb = mm->context.tsb_block[tsb_index].tsb; + old_cache_index = + (mm->context.tsb_block[tsb_index].tsb_reg_val & 0x7UL); + old_size = (mm->context.tsb_block[tsb_index].tsb_nentries * + sizeof(struct tsb)); /* Handle multiple threads trying to grow the TSB at the same time. * One will get in here first, and bump the size and the RSS limit. * The others will get in here next and hit this check. */ - if (unlikely(old_tsb && (rss < mm->context.tsb_rss_limit))) { + if (unlikely(old_tsb && + (rss < mm->context.tsb_block[tsb_index].tsb_rss_limit))) { spin_unlock_irqrestore(&mm->context.lock, flags); kmem_cache_free(tsb_caches[new_cache_index], new_tsb); return; } - mm->context.tsb_rss_limit = new_rss_limit; + mm->context.tsb_block[tsb_index].tsb_rss_limit = new_rss_limit; if (old_tsb) { extern void copy_tsb(unsigned long old_tsb_base, @@ -372,8 +407,8 @@ retry_tsb_alloc: copy_tsb(old_tsb_base, old_size, new_tsb_base, new_size); } - mm->context.tsb = new_tsb; - setup_tsb_params(mm, new_size); + mm->context.tsb_block[tsb_index].tsb = new_tsb; + setup_tsb_params(mm, tsb_index, new_size); spin_unlock_irqrestore(&mm->context.lock, flags); @@ -394,40 +429,65 @@ retry_tsb_alloc: int init_new_context(struct task_struct *tsk, struct mm_struct *mm) { +#ifdef CONFIG_HUGETLB_PAGE + unsigned long huge_pte_count; +#endif + unsigned int i; + spin_lock_init(&mm->context.lock); mm->context.sparc64_ctx_val = 0UL; +#ifdef CONFIG_HUGETLB_PAGE + /* We reset it to zero because the fork() page copying + * will re-increment the counters as the parent PTEs are + * copied into the child address space. + */ + huge_pte_count = mm->context.huge_pte_count; + mm->context.huge_pte_count = 0; +#endif + /* copy_mm() copies over the parent's mm_struct before calling * us, so we need to zero out the TSB pointer or else tsb_grow() * will be confused and think there is an older TSB to free up. */ - mm->context.tsb = NULL; + for (i = 0; i < MM_NUM_TSBS; i++) + mm->context.tsb_block[i].tsb = NULL; /* If this is fork, inherit the parent's TSB size. We would * grow it to that size on the first page fault anyways. */ - tsb_grow(mm, get_mm_rss(mm)); + tsb_grow(mm, MM_TSB_BASE, get_mm_rss(mm)); - if (unlikely(!mm->context.tsb)) +#ifdef CONFIG_HUGETLB_PAGE + if (unlikely(huge_pte_count)) + tsb_grow(mm, MM_TSB_HUGE, huge_pte_count); +#endif + + if (unlikely(!mm->context.tsb_block[MM_TSB_BASE].tsb)) return -ENOMEM; return 0; } -void destroy_context(struct mm_struct *mm) +static void tsb_destroy_one(struct tsb_config *tp) { - unsigned long flags, cache_index; + unsigned long cache_index; - cache_index = (mm->context.tsb_reg_val & 0x7UL); - kmem_cache_free(tsb_caches[cache_index], mm->context.tsb); + if (!tp->tsb) + return; + cache_index = tp->tsb_reg_val & 0x7UL; + kmem_cache_free(tsb_caches[cache_index], tp->tsb); + tp->tsb = NULL; + tp->tsb_reg_val = 0UL; +} - /* We can remove these later, but for now it's useful - * to catch any bogus post-destroy_context() references - * to the TSB. - */ - mm->context.tsb = NULL; - mm->context.tsb_reg_val = 0UL; +void destroy_context(struct mm_struct *mm) +{ + unsigned long flags, i; + + for (i = 0; i < MM_NUM_TSBS; i++) + tsb_destroy_one(&mm->context.tsb_block[i]); spin_lock_irqsave(&ctx_alloc_lock, flags); diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index fa4f915be5c..92cce96b5e2 100644 --- a/arch/um/kernel/mem.c +++ b/arch/um/kernel/mem.c @@ -57,7 +57,7 @@ static void setup_highmem(unsigned long highmem_start, for(i = 0; i < highmem_len >> PAGE_SHIFT; i++){ page = &mem_map[highmem_pfn + i]; ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); } } @@ -296,7 +296,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c index 544665e0451..0e65340eee3 100644 --- a/arch/um/kernel/physmem.c +++ b/arch/um/kernel/physmem.c @@ -279,7 +279,7 @@ int init_maps(unsigned long physmem, unsigned long iomem, unsigned long highmem) for(i = 0; i < total_pages; i++){ p = &map[i]; - set_page_count(p, 0); + memset(p, 0, sizeof(struct page)); SetPageReserved(p); INIT_LIST_HEAD(&p->lru); } diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index 3080f84bf7b..ee5ce3d3cbc 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -477,7 +477,7 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) return IRQ_HANDLED; } -static unsigned int cyc2ns_scale; +static unsigned int cyc2ns_scale __read_mostly; #define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ static inline void set_cyc2ns_scale(unsigned long cpu_khz) diff --git a/arch/x86_64/kernel/x8664_ksyms.c b/arch/x86_64/kernel/x8664_ksyms.c index 3496abc8d37..c9dc7e46731 100644 --- a/arch/x86_64/kernel/x8664_ksyms.c +++ b/arch/x86_64/kernel/x8664_ksyms.c @@ -124,6 +124,7 @@ extern void * __memcpy(void *,const void *,__kernel_size_t); EXPORT_SYMBOL(memset); EXPORT_SYMBOL(strlen); +EXPORT_SYMBOL(strpbrk); EXPORT_SYMBOL(memmove); EXPORT_SYMBOL(memcpy); EXPORT_SYMBOL(__memcpy); diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index 7af1742aa95..40ed13d263c 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -486,7 +486,7 @@ void __init clear_kernel_mapping(unsigned long address, unsigned long size) void online_page(struct page *page) { ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalram_pages++; num_physpages++; @@ -592,7 +592,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); memset((void *)(addr & ~(PAGE_SIZE-1)), 0xcc, PAGE_SIZE); free_page(addr); totalram_pages++; @@ -632,7 +632,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/x86_64/mm/pageattr.c b/arch/x86_64/mm/pageattr.c index 35f1f1aab06..531ad21447b 100644 --- a/arch/x86_64/mm/pageattr.c +++ b/arch/x86_64/mm/pageattr.c @@ -45,6 +45,13 @@ static struct page *split_large_page(unsigned long address, pgprot_t prot, pte_t *pbase; if (!base) return NULL; + /* + * page_private is used to track the number of entries in + * the page table page have non standard attributes. + */ + SetPagePrivate(base); + page_private(base) = 0; + address = __pa(address); addr = address & LARGE_PAGE_MASK; pbase = (pte_t *)page_address(base); @@ -77,26 +84,12 @@ static inline void flush_map(unsigned long address) on_each_cpu(flush_kernel_map, (void *)address, 1, 1); } -struct deferred_page { - struct deferred_page *next; - struct page *fpage; - unsigned long address; -}; -static struct deferred_page *df_list; /* protected by init_mm.mmap_sem */ +static struct page *deferred_pages; /* protected by init_mm.mmap_sem */ -static inline void save_page(unsigned long address, struct page *fpage) +static inline void save_page(struct page *fpage) { - struct deferred_page *df; - df = kmalloc(sizeof(struct deferred_page), GFP_KERNEL); - if (!df) { - flush_map(address); - __free_page(fpage); - } else { - df->next = df_list; - df->fpage = fpage; - df->address = address; - df_list = df; - } + fpage->lru.next = (struct list_head *)deferred_pages; + deferred_pages = fpage; } /* @@ -138,8 +131,8 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot, set_pte(kpte, pfn_pte(pfn, prot)); } else { /* - * split_large_page will take the reference for this change_page_attr - * on the split page. + * split_large_page will take the reference for this + * change_page_attr on the split page. */ struct page *split; @@ -151,23 +144,20 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot, set_pte(kpte,mk_pte(split, ref_prot2)); kpte_page = split; } - get_page(kpte_page); + page_private(kpte_page)++; } else if ((kpte_flags & _PAGE_PSE) == 0) { set_pte(kpte, pfn_pte(pfn, ref_prot)); - __put_page(kpte_page); + BUG_ON(page_private(kpte_page) == 0); + page_private(kpte_page)--; } else BUG(); /* on x86-64 the direct mapping set at boot is not using 4k pages */ BUG_ON(PageReserved(kpte_page)); - switch (page_count(kpte_page)) { - case 1: - save_page(address, kpte_page); + if (page_private(kpte_page) == 0) { + save_page(kpte_page); revert_page(address, ref_prot); - break; - case 0: - BUG(); /* memleak and failed 2M page regeneration */ } return 0; } @@ -220,17 +210,18 @@ int change_page_attr(struct page *page, int numpages, pgprot_t prot) void global_flush_tlb(void) { - struct deferred_page *df, *next_df; + struct page *dpage; down_read(&init_mm.mmap_sem); - df = xchg(&df_list, NULL); + dpage = xchg(&deferred_pages, NULL); up_read(&init_mm.mmap_sem); - flush_map((df && !df->next) ? df->address : 0); - for (; df; df = next_df) { - next_df = df->next; - if (df->fpage) - __free_page(df->fpage); - kfree(df); + + flush_map((dpage && !dpage->lru.next) ? (unsigned long)page_address(dpage) : 0); + while (dpage) { + struct page *tmp = dpage; + dpage = (struct page *)dpage->lru.next; + ClearPagePrivate(tmp); + __free_page(tmp); } } diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c index 5a91d6c9e66..e1be4235f36 100644 --- a/arch/xtensa/mm/init.c +++ b/arch/xtensa/mm/init.c @@ -272,7 +272,7 @@ free_reserved_mem(void *start, void *end) { for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page((unsigned long)start); totalram_pages++; } diff --git a/arch/xtensa/mm/pgtable.c b/arch/xtensa/mm/pgtable.c index e5e119c820e..7d28914d11c 100644 --- a/arch/xtensa/mm/pgtable.c +++ b/arch/xtensa/mm/pgtable.c @@ -14,25 +14,21 @@ pte_t* pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { - pte_t *pte, p; + pte_t *pte = NULL, *p; int color = ADDR_COLOR(address); int i; p = (pte_t*) __get_free_pages(GFP_KERNEL|__GFP_REPEAT, COLOR_ORDER); if (likely(p)) { - struct page *page; - - for (i = 0; i < COLOR_SIZE; i++, p++) { - page = virt_to_page(pte); - - set_page_count(page, 1); - ClearPageCompound(page); + split_page(virt_to_page(p), COLOR_ORDER); + for (i = 0; i < COLOR_SIZE; i++) { if (ADDR_COLOR(p) == color) pte = p; else free_page(p); + p += PTRS_PER_PTE; } clear_page(pte); } @@ -49,20 +45,20 @@ int flush; struct page* pte_alloc_one(struct mm_struct *mm, unsigned long address) { - struct page *page, p; + struct page *page = NULL, *p; int color = ADDR_COLOR(address); p = alloc_pages(GFP_KERNEL | __GFP_REPEAT, PTE_ORDER); if (likely(p)) { - for (i = 0; i < PAGE_ORDER; i++) { - set_page_count(p, 1); - ClearPageCompound(p); + split_page(p, COLOR_ORDER); - if (PADDR_COLOR(page_address(pg)) == color) + for (i = 0; i < PAGE_ORDER; i++) { + if (PADDR_COLOR(page_address(p)) == color) page = p; else - free_page(p); + __free_page(p); + p++; } clear_highpage(page); } |