diff options
Diffstat (limited to 'arch/ia64/mm')
| -rw-r--r-- | arch/ia64/mm/contig.c | 245 | ||||
| -rw-r--r-- | arch/ia64/mm/discontig.c | 307 | ||||
| -rw-r--r-- | arch/ia64/mm/extable.c | 28 | ||||
| -rw-r--r-- | arch/ia64/mm/fault.c | 148 | ||||
| -rw-r--r-- | arch/ia64/mm/hugetlbpage.c | 61 | ||||
| -rw-r--r-- | arch/ia64/mm/init.c | 360 | ||||
| -rw-r--r-- | arch/ia64/mm/ioremap.c | 93 | ||||
| -rw-r--r-- | arch/ia64/mm/numa.c | 46 | ||||
| -rw-r--r-- | arch/ia64/mm/tlb.c | 418 |
9 files changed, 1129 insertions, 577 deletions
diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c index daf977ff292..52715a71aed 100644 --- a/arch/ia64/mm/contig.c +++ b/arch/ia64/mm/contig.c @@ -16,7 +16,9 @@ */ #include <linux/bootmem.h> #include <linux/efi.h> +#include <linux/memblock.h> #include <linux/mm.h> +#include <linux/nmi.h> #include <linux/swap.h> #include <asm/meminit.h> @@ -29,72 +31,10 @@ static unsigned long max_gap; #endif -/** - * show_mem - display a memory statistics summary - * - * Just walks the pages in the system and describes where they're allocated. - */ -void -show_mem (void) -{ - int i, total = 0, reserved = 0; - int shared = 0, cached = 0; - - printk(KERN_INFO "Mem-info:\n"); - show_free_areas(); - - printk(KERN_INFO "Free swap: %6ldkB\n", - nr_swap_pages<<(PAGE_SHIFT-10)); - i = max_mapnr; - for (i = 0; i < max_mapnr; i++) { - if (!pfn_valid(i)) { -#ifdef CONFIG_VIRTUAL_MEM_MAP - if (max_gap < LARGE_GAP) - continue; - i = vmemmap_find_next_valid_pfn(0, i) - 1; -#endif - continue; - } - total++; - if (PageReserved(mem_map+i)) - reserved++; - else if (PageSwapCache(mem_map+i)) - cached++; - else if (page_count(mem_map + i)) - shared += page_count(mem_map + i) - 1; - } - printk(KERN_INFO "%d pages of RAM\n", total); - printk(KERN_INFO "%d reserved pages\n", reserved); - printk(KERN_INFO "%d pages shared\n", shared); - printk(KERN_INFO "%d pages swap cached\n", cached); - printk(KERN_INFO "%ld pages in page table cache\n", - pgtable_quicklist_total_size()); -} - /* physical address where the bootmem map is located */ unsigned long bootmap_start; /** - * find_max_pfn - adjust the maximum page number callback - * @start: start of range - * @end: end of range - * @arg: address of pointer to global max_pfn variable - * - * Passed as a callback function to efi_memmap_walk() to determine the highest - * available page frame number in the system. - */ -int -find_max_pfn (unsigned long start, unsigned long end, void *arg) -{ - unsigned long *max_pfnp = arg, pfn; - - pfn = (PAGE_ALIGN(end - 1) - PAGE_OFFSET) >> PAGE_SHIFT; - if (pfn > *max_pfnp) - *max_pfnp = pfn; - return 0; -} - -/** * find_bootmap_location - callback to find a memory area for the bootmap * @start: start of region * @end: end of region @@ -104,10 +44,10 @@ find_max_pfn (unsigned long start, unsigned long end, void *arg) * bootmap_start. This address must be page-aligned. */ static int __init -find_bootmap_location (unsigned long start, unsigned long end, void *arg) +find_bootmap_location (u64 start, u64 end, void *arg) { - unsigned long needed = *(unsigned long *)arg; - unsigned long range_start, range_end, free_start; + u64 needed = *(unsigned long *)arg; + u64 range_start, range_end, free_start; int i; #if IGNORE_PFN0 @@ -141,6 +81,112 @@ find_bootmap_location (unsigned long start, unsigned long end, void *arg) return 0; } +#ifdef CONFIG_SMP +static void *cpu_data; +/** + * per_cpu_init - setup per-cpu variables + * + * Allocate and setup per-cpu data areas. + */ +void *per_cpu_init(void) +{ + static bool first_time = true; + void *cpu0_data = __cpu0_per_cpu; + unsigned int cpu; + + if (!first_time) + goto skip; + first_time = false; + + /* + * get_free_pages() cannot be used before cpu_init() done. + * BSP allocates PERCPU_PAGE_SIZE bytes for all possible CPUs + * to avoid that AP calls get_zeroed_page(). + */ + for_each_possible_cpu(cpu) { + void *src = cpu == 0 ? cpu0_data : __phys_per_cpu_start; + + memcpy(cpu_data, src, __per_cpu_end - __per_cpu_start); + __per_cpu_offset[cpu] = (char *)cpu_data - __per_cpu_start; + per_cpu(local_per_cpu_offset, cpu) = __per_cpu_offset[cpu]; + + /* + * percpu area for cpu0 is moved from the __init area + * which is setup by head.S and used till this point. + * Update ar.k3. This move is ensures that percpu + * area for cpu0 is on the correct node and its + * virtual address isn't insanely far from other + * percpu areas which is important for congruent + * percpu allocator. + */ + if (cpu == 0) + ia64_set_kr(IA64_KR_PER_CPU_DATA, __pa(cpu_data) - + (unsigned long)__per_cpu_start); + + cpu_data += PERCPU_PAGE_SIZE; + } +skip: + return __per_cpu_start + __per_cpu_offset[smp_processor_id()]; +} + +static inline void +alloc_per_cpu_data(void) +{ + cpu_data = __alloc_bootmem(PERCPU_PAGE_SIZE * num_possible_cpus(), + PERCPU_PAGE_SIZE, __pa(MAX_DMA_ADDRESS)); +} + +/** + * setup_per_cpu_areas - setup percpu areas + * + * Arch code has already allocated and initialized percpu areas. All + * this function has to do is to teach the determined layout to the + * dynamic percpu allocator, which happens to be more complex than + * creating whole new ones using helpers. + */ +void __init +setup_per_cpu_areas(void) +{ + struct pcpu_alloc_info *ai; + struct pcpu_group_info *gi; + unsigned int cpu; + ssize_t static_size, reserved_size, dyn_size; + int rc; + + ai = pcpu_alloc_alloc_info(1, num_possible_cpus()); + if (!ai) + panic("failed to allocate pcpu_alloc_info"); + gi = &ai->groups[0]; + + /* units are assigned consecutively to possible cpus */ + for_each_possible_cpu(cpu) + gi->cpu_map[gi->nr_units++] = cpu; + + /* set parameters */ + static_size = __per_cpu_end - __per_cpu_start; + reserved_size = PERCPU_MODULE_RESERVE; + dyn_size = PERCPU_PAGE_SIZE - static_size - reserved_size; + if (dyn_size < 0) + panic("percpu area overflow static=%zd reserved=%zd\n", + static_size, reserved_size); + + ai->static_size = static_size; + ai->reserved_size = reserved_size; + ai->dyn_size = dyn_size; + ai->unit_size = PERCPU_PAGE_SIZE; + ai->atom_size = PAGE_SIZE; + ai->alloc_size = PERCPU_PAGE_SIZE; + + rc = pcpu_setup_first_chunk(ai, __per_cpu_start + __per_cpu_offset[0]); + if (rc) + panic("failed to setup percpu area (err=%d)", rc); + + pcpu_free_alloc_info(ai); +} +#else +#define alloc_per_cpu_data() do { } while (0) +#endif /* CONFIG_SMP */ + /** * find_memory - setup memory map * @@ -155,9 +201,10 @@ find_memory (void) reserve_memory(); /* first find highest page frame number */ - max_pfn = 0; - efi_memmap_walk(find_max_pfn, &max_pfn); - + min_low_pfn = ~0UL; + max_low_pfn = 0; + efi_memmap_walk(find_max_min_low_pfn, NULL); + max_pfn = max_low_pfn; /* how many bytes to cover all the pages */ bootmap_size = bootmem_bootmap_pages(max_pfn) << PAGE_SHIFT; @@ -167,55 +214,16 @@ find_memory (void) if (bootmap_start == ~0UL) panic("Cannot find %ld bytes for bootmap\n", bootmap_size); - bootmap_size = init_bootmem(bootmap_start >> PAGE_SHIFT, max_pfn); + bootmap_size = init_bootmem_node(NODE_DATA(0), + (bootmap_start >> PAGE_SHIFT), 0, max_pfn); /* Free all available memory, then mark bootmem-map as being in use. */ efi_memmap_walk(filter_rsvd_memory, free_bootmem); - reserve_bootmem(bootmap_start, bootmap_size); + reserve_bootmem(bootmap_start, bootmap_size, BOOTMEM_DEFAULT); find_initrd(); -} -#ifdef CONFIG_SMP -/** - * per_cpu_init - setup per-cpu variables - * - * Allocate and setup per-cpu data areas. - */ -void * __cpuinit -per_cpu_init (void) -{ - void *cpu_data; - int cpu; - static int first_time=1; - - /* - * get_free_pages() cannot be used before cpu_init() done. BSP - * allocates "NR_CPUS" pages for all CPUs to avoid that AP calls - * get_zeroed_page(). - */ - if (first_time) { - first_time=0; - cpu_data = __alloc_bootmem(PERCPU_PAGE_SIZE * NR_CPUS, - PERCPU_PAGE_SIZE, __pa(MAX_DMA_ADDRESS)); - for (cpu = 0; cpu < NR_CPUS; cpu++) { - memcpy(cpu_data, __phys_per_cpu_start, __per_cpu_end - __per_cpu_start); - __per_cpu_offset[cpu] = (char *) cpu_data - __per_cpu_start; - cpu_data += PERCPU_PAGE_SIZE; - per_cpu(local_per_cpu_offset, cpu) = __per_cpu_offset[cpu]; - } - } - return __per_cpu_start + __per_cpu_offset[smp_processor_id()]; -} -#endif /* CONFIG_SMP */ - -static int -count_pages (u64 start, u64 end, void *arg) -{ - unsigned long *count = arg; - - *count += (end - start) >> PAGE_SHIFT; - return 0; + alloc_per_cpu_data(); } /* @@ -226,18 +234,17 @@ void __init paging_init (void) { unsigned long max_dma; - unsigned long nid = 0; unsigned long max_zone_pfns[MAX_NR_ZONES]; - num_physpages = 0; - efi_memmap_walk(count_pages, &num_physpages); - + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); +#ifdef CONFIG_ZONE_DMA max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT; max_zone_pfns[ZONE_DMA] = max_dma; +#endif max_zone_pfns[ZONE_NORMAL] = max_low_pfn; #ifdef CONFIG_VIRTUAL_MEM_MAP - efi_memmap_walk(register_active_ranges, &nid); + efi_memmap_walk(filter_memory, register_active_ranges); efi_memmap_walk(find_largest_hole, (u64 *)&max_gap); if (max_gap < LARGE_GAP) { vmem_map = (struct page *) 0; @@ -249,8 +256,8 @@ paging_init (void) map_size = PAGE_ALIGN(ALIGN(max_low_pfn, MAX_ORDER_NR_PAGES) * sizeof(struct page)); - vmalloc_end -= map_size; - vmem_map = (struct page *) vmalloc_end; + VMALLOC_END -= map_size; + vmem_map = (struct page *) VMALLOC_END; efi_memmap_walk(create_mem_map_page_table, NULL); /* @@ -264,7 +271,7 @@ paging_init (void) printk("Virtual mem_map starts at 0x%p\n", mem_map); } #else /* !CONFIG_VIRTUAL_MEM_MAP */ - add_active_range(0, 0, max_low_pfn); + memblock_add_node(0, PFN_PHYS(max_low_pfn), 0); free_area_init_nodes(max_zone_pfns); #endif /* !CONFIG_VIRTUAL_MEM_MAP */ zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page)); diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index d497b6b0f5b..87862680536 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -16,11 +16,13 @@ #include <linux/kernel.h> #include <linux/mm.h> +#include <linux/nmi.h> #include <linux/swap.h> #include <linux/bootmem.h> #include <linux/acpi.h> #include <linux/efi.h> #include <linux/nodemask.h> +#include <linux/slab.h> #include <asm/pgalloc.h> #include <asm/tlb.h> #include <asm/meminit.h> @@ -35,9 +37,9 @@ struct early_node_data { struct ia64_node_data *node_data; unsigned long pernode_addr; unsigned long pernode_size; - struct bootmem_data bootmem_data; - unsigned long num_physpages; +#ifdef CONFIG_ZONE_DMA unsigned long num_dma_physpages; +#endif unsigned long min_pfn; unsigned long max_pfn; }; @@ -45,7 +47,7 @@ struct early_node_data { static struct early_node_data mem_data[MAX_NUMNODES] __initdata; static nodemask_t memory_less_mask __initdata; -static pg_data_t *pgdat_list[MAX_NUMNODES]; +pg_data_t *pgdat_list[MAX_NUMNODES]; /* * To prevent cache aliasing effects, align per-node structures so that they @@ -72,23 +74,20 @@ static pg_data_t *pgdat_list[MAX_NUMNODES]; static int __init build_node_maps(unsigned long start, unsigned long len, int node) { - unsigned long cstart, epfn, end = start + len; - struct bootmem_data *bdp = &mem_data[node].bootmem_data; + unsigned long spfn, epfn, end = start + len; + struct bootmem_data *bdp = &bootmem_node_data[node]; epfn = GRANULEROUNDUP(end) >> PAGE_SHIFT; - cstart = GRANULEROUNDDOWN(start); + spfn = GRANULEROUNDDOWN(start) >> PAGE_SHIFT; if (!bdp->node_low_pfn) { - bdp->node_boot_start = cstart; + bdp->node_min_pfn = spfn; bdp->node_low_pfn = epfn; } else { - bdp->node_boot_start = min(cstart, bdp->node_boot_start); + bdp->node_min_pfn = min(spfn, bdp->node_min_pfn); bdp->node_low_pfn = max(epfn, bdp->node_low_pfn); } - min_low_pfn = min(min_low_pfn, bdp->node_boot_start>>PAGE_SHIFT); - max_low_pfn = max(max_low_pfn, bdp->node_low_pfn); - return 0; } @@ -104,7 +103,7 @@ static int __meminit early_nr_cpus_node(int node) { int cpu, n = 0; - for (cpu = 0; cpu < NR_CPUS; cpu++) + for_each_possible_early_cpu(cpu) if (node == node_cpuid[cpu].nid) n++; @@ -124,6 +123,7 @@ static unsigned long __meminit compute_pernodesize(int node) pernodesize += node * L1_CACHE_BYTES; pernodesize += L1_CACHE_ALIGN(sizeof(pg_data_t)); pernodesize += L1_CACHE_ALIGN(sizeof(struct ia64_node_data)); + pernodesize += L1_CACHE_ALIGN(sizeof(pg_data_t)); pernodesize = PAGE_ALIGN(pernodesize); return pernodesize; } @@ -142,19 +142,121 @@ static void *per_cpu_node_setup(void *cpu_data, int node) #ifdef CONFIG_SMP int cpu; - for (cpu = 0; cpu < NR_CPUS; cpu++) { - if (node == node_cpuid[cpu].nid) { - memcpy(__va(cpu_data), __phys_per_cpu_start, - __per_cpu_end - __per_cpu_start); - __per_cpu_offset[cpu] = (char*)__va(cpu_data) - - __per_cpu_start; - cpu_data += PERCPU_PAGE_SIZE; - } + for_each_possible_early_cpu(cpu) { + void *src = cpu == 0 ? __cpu0_per_cpu : __phys_per_cpu_start; + + if (node != node_cpuid[cpu].nid) + continue; + + memcpy(__va(cpu_data), src, __per_cpu_end - __per_cpu_start); + __per_cpu_offset[cpu] = (char *)__va(cpu_data) - + __per_cpu_start; + + /* + * percpu area for cpu0 is moved from the __init area + * which is setup by head.S and used till this point. + * Update ar.k3. This move is ensures that percpu + * area for cpu0 is on the correct node and its + * virtual address isn't insanely far from other + * percpu areas which is important for congruent + * percpu allocator. + */ + if (cpu == 0) + ia64_set_kr(IA64_KR_PER_CPU_DATA, + (unsigned long)cpu_data - + (unsigned long)__per_cpu_start); + + cpu_data += PERCPU_PAGE_SIZE; } #endif return cpu_data; } +#ifdef CONFIG_SMP +/** + * setup_per_cpu_areas - setup percpu areas + * + * Arch code has already allocated and initialized percpu areas. All + * this function has to do is to teach the determined layout to the + * dynamic percpu allocator, which happens to be more complex than + * creating whole new ones using helpers. + */ +void __init setup_per_cpu_areas(void) +{ + struct pcpu_alloc_info *ai; + struct pcpu_group_info *uninitialized_var(gi); + unsigned int *cpu_map; + void *base; + unsigned long base_offset; + unsigned int cpu; + ssize_t static_size, reserved_size, dyn_size; + int node, prev_node, unit, nr_units, rc; + + ai = pcpu_alloc_alloc_info(MAX_NUMNODES, nr_cpu_ids); + if (!ai) + panic("failed to allocate pcpu_alloc_info"); + cpu_map = ai->groups[0].cpu_map; + + /* determine base */ + base = (void *)ULONG_MAX; + for_each_possible_cpu(cpu) + base = min(base, + (void *)(__per_cpu_offset[cpu] + __per_cpu_start)); + base_offset = (void *)__per_cpu_start - base; + + /* build cpu_map, units are grouped by node */ + unit = 0; + for_each_node(node) + for_each_possible_cpu(cpu) + if (node == node_cpuid[cpu].nid) + cpu_map[unit++] = cpu; + nr_units = unit; + + /* set basic parameters */ + static_size = __per_cpu_end - __per_cpu_start; + reserved_size = PERCPU_MODULE_RESERVE; + dyn_size = PERCPU_PAGE_SIZE - static_size - reserved_size; + if (dyn_size < 0) + panic("percpu area overflow static=%zd reserved=%zd\n", + static_size, reserved_size); + + ai->static_size = static_size; + ai->reserved_size = reserved_size; + ai->dyn_size = dyn_size; + ai->unit_size = PERCPU_PAGE_SIZE; + ai->atom_size = PAGE_SIZE; + ai->alloc_size = PERCPU_PAGE_SIZE; + + /* + * CPUs are put into groups according to node. Walk cpu_map + * and create new groups at node boundaries. + */ + prev_node = -1; + ai->nr_groups = 0; + for (unit = 0; unit < nr_units; unit++) { + cpu = cpu_map[unit]; + node = node_cpuid[cpu].nid; + + if (node == prev_node) { + gi->nr_units++; + continue; + } + prev_node = node; + + gi = &ai->groups[ai->nr_groups++]; + gi->nr_units = 1; + gi->base_offset = __per_cpu_offset[cpu] + base_offset; + gi->cpu_map = &cpu_map[unit]; + } + + rc = pcpu_setup_first_chunk(ai, base); + if (rc) + panic("failed to setup percpu area (err=%d)", rc); + + pcpu_free_alloc_info(ai); +} +#endif + /** * fill_pernode - initialize pernode data. * @node: the node id. @@ -166,7 +268,7 @@ static void __init fill_pernode(int node, unsigned long pernode, { void *cpu_data; int cpus = early_nr_cpus_node(node); - struct bootmem_data *bdp = &mem_data[node].bootmem_data; + struct bootmem_data *bdp = &bootmem_node_data[node]; mem_data[node].pernode_addr = pernode; mem_data[node].pernode_size = pernodesize; @@ -221,20 +323,21 @@ static void __init fill_pernode(int node, unsigned long pernode, static int __init find_pernode_space(unsigned long start, unsigned long len, int node) { - unsigned long epfn; + unsigned long spfn, epfn; unsigned long pernodesize = 0, pernode, pages, mapsize; - struct bootmem_data *bdp = &mem_data[node].bootmem_data; + struct bootmem_data *bdp = &bootmem_node_data[node]; + spfn = start >> PAGE_SHIFT; epfn = (start + len) >> PAGE_SHIFT; - pages = bdp->node_low_pfn - (bdp->node_boot_start >> PAGE_SHIFT); + pages = bdp->node_low_pfn - bdp->node_min_pfn; mapsize = bootmem_bootmap_pages(pages) << PAGE_SHIFT; /* * Make sure this memory falls within this node's usable memory * since we may have thrown some away in build_maps(). */ - if (start < bdp->node_boot_start || epfn > bdp->node_low_pfn) + if (spfn < bdp->node_min_pfn || epfn > bdp->node_low_pfn) return 0; /* Don't setup this node's local space twice... */ @@ -296,15 +399,15 @@ static void __init reserve_pernode_space(void) bdp = pdp->bdata; /* First the bootmem_map itself */ - pages = bdp->node_low_pfn - (bdp->node_boot_start>>PAGE_SHIFT); + pages = bdp->node_low_pfn - bdp->node_min_pfn; size = bootmem_bootmap_pages(pages) << PAGE_SHIFT; base = __pa(bdp->node_bootmem_map); - reserve_bootmem_node(pdp, base, size); + reserve_bootmem_node(pdp, base, size, BOOTMEM_DEFAULT); /* Now the per-node space */ size = mem_data[node].pernode_size; base = __pa(mem_data[node].pernode_addr); - reserve_bootmem_node(pdp, base, size); + reserve_bootmem_node(pdp, base, size, BOOTMEM_DEFAULT); } } @@ -318,7 +421,7 @@ static void __meminit scatter_node_data(void) * node_online_map is not set for hot-added nodes at this time, * because we are halfway through initialization of the new node's * structures. If for_each_online_node() is used, a new node's - * pg_data_ptrs will be not initialized. Insted of using it, + * pg_data_ptrs will be not initialized. Instead of using it, * pgdat_list[] is checked. */ for_each_node(node) { @@ -345,9 +448,10 @@ static void __init initialize_pernode_data(void) #ifdef CONFIG_SMP /* Set the node_data pointer for each per-cpu struct */ - for (cpu = 0; cpu < NR_CPUS; cpu++) { + for_each_possible_early_cpu(cpu) { node = node_cpuid[cpu].nid; - per_cpu(cpu_info, cpu).node_data = mem_data[node].node_data; + per_cpu(ia64_cpu_info, cpu).node_data = + mem_data[node].node_data; } #else { @@ -355,7 +459,7 @@ static void __init initialize_pernode_data(void) cpu = 0; node = node_cpuid[cpu].nid; cpu0_cpu_info = (struct cpuinfo_ia64 *)(__phys_per_cpu_start + - ((char *)&per_cpu__cpu_info - __per_cpu_start)); + ((char *)&ia64_cpu_info - __per_cpu_start)); cpu0_cpu_info->node_data = mem_data[node].node_data; } #endif /* CONFIG_SMP */ @@ -412,37 +516,6 @@ static void __init memory_less_nodes(void) return; } -#ifdef CONFIG_SPARSEMEM -/** - * register_sparse_mem - notify SPARSEMEM that this memory range exists. - * @start: physical start of range - * @end: physical end of range - * @arg: unused - * - * Simply calls SPARSEMEM to register memory section(s). - */ -static int __init register_sparse_mem(unsigned long start, unsigned long end, - void *arg) -{ - int nid; - - start = __pa(start) >> PAGE_SHIFT; - end = __pa(end) >> PAGE_SHIFT; - nid = early_pfn_to_nid(start); - memory_present(nid, start, end); - - return 0; -} - -static void __init arch_sparse_init(void) -{ - efi_memmap_walk(register_sparse_mem, NULL); - sparse_init(); -} -#else -#define arch_sparse_init() do {} while (0) -#endif - /** * find_memory - walk the EFI memory map and setup the bootmem allocator * @@ -467,12 +540,16 @@ void __init find_memory(void) /* These actually end up getting called by call_pernode_memory() */ efi_memmap_walk(filter_rsvd_memory, build_node_maps); efi_memmap_walk(filter_rsvd_memory, find_pernode_space); + efi_memmap_walk(find_max_min_low_pfn, NULL); for_each_online_node(node) - if (mem_data[node].bootmem_data.node_low_pfn) { + if (bootmem_node_data[node].node_low_pfn) { node_clear(node, memory_less_mask); mem_data[node].min_pfn = ~0UL; } + + efi_memmap_walk(filter_memory, register_active_ranges); + /* * Initialize the boot memory maps in reverse order since that's * what the bootmem allocator expects @@ -486,14 +563,14 @@ void __init find_memory(void) else if (node_isset(node, memory_less_mask)) continue; - bdp = &mem_data[node].bootmem_data; + bdp = &bootmem_node_data[node]; pernode = mem_data[node].pernode_addr; pernodesize = mem_data[node].pernode_size; map = pernode + pernodesize; init_bootmem_node(pgdat_list[node], map>>PAGE_SHIFT, - bdp->node_boot_start>>PAGE_SHIFT, + bdp->node_min_pfn, bdp->node_low_pfn); } @@ -515,18 +592,14 @@ void __init find_memory(void) * find_pernode_space() does most of this already, we just need to set * local_per_cpu_offset */ -void __cpuinit *per_cpu_init(void) +void *per_cpu_init(void) { int cpu; static int first_time = 1; - - if (smp_processor_id() != 0) - return __per_cpu_start + __per_cpu_offset[smp_processor_id()]; - if (first_time) { first_time = 0; - for (cpu = 0; cpu < NR_CPUS; cpu++) + for_each_possible_early_cpu(cpu) per_cpu(local_per_cpu_offset, cpu) = __per_cpu_offset[cpu]; } @@ -535,65 +608,6 @@ void __cpuinit *per_cpu_init(void) #endif /* CONFIG_SMP */ /** - * show_mem - give short summary of memory stats - * - * Shows a simple page count of reserved and used pages in the system. - * For discontig machines, it does this on a per-pgdat basis. - */ -void show_mem(void) -{ - int i, total_reserved = 0; - int total_shared = 0, total_cached = 0; - unsigned long total_present = 0; - pg_data_t *pgdat; - - printk(KERN_INFO "Mem-info:\n"); - show_free_areas(); - printk(KERN_INFO "Free swap: %6ldkB\n", - nr_swap_pages<<(PAGE_SHIFT-10)); - printk(KERN_INFO "Node memory in pages:\n"); - for_each_online_pgdat(pgdat) { - unsigned long present; - unsigned long flags; - int shared = 0, cached = 0, reserved = 0; - - pgdat_resize_lock(pgdat, &flags); - present = pgdat->node_present_pages; - for(i = 0; i < pgdat->node_spanned_pages; i++) { - struct page *page; - if (pfn_valid(pgdat->node_start_pfn + i)) - page = pfn_to_page(pgdat->node_start_pfn + i); - else { - i = vmemmap_find_next_valid_pfn(pgdat->node_id, - i) - 1; - continue; - } - if (PageReserved(page)) - reserved++; - else if (PageSwapCache(page)) - cached++; - else if (page_count(page)) - shared += page_count(page)-1; - } - pgdat_resize_unlock(pgdat, &flags); - total_present += present; - total_reserved += reserved; - total_cached += cached; - total_shared += shared; - printk(KERN_INFO "Node %4d: RAM: %11ld, rsvd: %8d, " - "shrd: %10d, swpd: %10d\n", pgdat->node_id, - present, reserved, shared, cached); - } - printk(KERN_INFO "%ld pages of RAM\n", total_present); - printk(KERN_INFO "%d reserved pages\n", total_reserved); - printk(KERN_INFO "%d pages shared\n", total_shared); - printk(KERN_INFO "%d pages swap cached\n", total_cached); - printk(KERN_INFO "Total of %ld pages in page table cache\n", - pgtable_quicklist_total_size()); - printk(KERN_INFO "%d free buffer pages\n", nr_free_buffer_pages()); -} - -/** * call_pernode_memory - use SRAT to call callback functions with node info * @start: physical start of range * @len: length of range @@ -654,13 +668,12 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n { unsigned long end = start + len; - add_active_range(node, start >> PAGE_SHIFT, end >> PAGE_SHIFT); - mem_data[node].num_physpages += len >> PAGE_SHIFT; +#ifdef CONFIG_ZONE_DMA if (start <= __pa(MAX_DMA_ADDRESS)) mem_data[node].num_dma_physpages += (min(end, __pa(MAX_DMA_ADDRESS)) - start) >>PAGE_SHIFT; +#endif start = GRANULEROUNDDOWN(start); - start = ORDERROUNDDOWN(start); end = GRANULEROUNDUP(end); mem_data[node].max_pfn = max(mem_data[node].max_pfn, end >> PAGE_SHIFT); @@ -686,20 +699,20 @@ void __init paging_init(void) max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT; - arch_sparse_init(); - efi_memmap_walk(filter_rsvd_memory, count_node_pages); + sparse_memory_present_with_active_regions(MAX_NUMNODES); + sparse_init(); + #ifdef CONFIG_VIRTUAL_MEM_MAP - vmalloc_end -= PAGE_ALIGN(ALIGN(max_low_pfn, MAX_ORDER_NR_PAGES) * + VMALLOC_END -= PAGE_ALIGN(ALIGN(max_low_pfn, MAX_ORDER_NR_PAGES) * sizeof(struct page)); - vmem_map = (struct page *) vmalloc_end; + vmem_map = (struct page *) VMALLOC_END; efi_memmap_walk(create_mem_map_page_table, NULL); printk("Virtual mem_map starts at 0x%p\n", vmem_map); #endif for_each_online_node(node) { - num_physpages += mem_data[node].num_physpages; pfn_offset = mem_data[node].min_pfn; #ifdef CONFIG_VIRTUAL_MEM_MAP @@ -709,13 +722,17 @@ void __init paging_init(void) max_pfn = mem_data[node].max_pfn; } + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); +#ifdef CONFIG_ZONE_DMA max_zone_pfns[ZONE_DMA] = max_dma; +#endif max_zone_pfns[ZONE_NORMAL] = max_pfn; free_area_init_nodes(max_zone_pfns); zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page)); } +#ifdef CONFIG_MEMORY_HOTPLUG pg_data_t *arch_alloc_nodedata(int nid) { unsigned long size = compute_pernodesize(nid); @@ -733,3 +750,15 @@ void arch_refresh_nodedata(int update_node, pg_data_t *update_pgdat) pgdat_list[update_node] = update_pgdat; scatter_node_data(); } +#endif + +#ifdef CONFIG_SPARSEMEM_VMEMMAP +int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node) +{ + return vmemmap_populate_basepages(start, end, node); +} + +void vmemmap_free(unsigned long start, unsigned long end) +{ +} +#endif diff --git a/arch/ia64/mm/extable.c b/arch/ia64/mm/extable.c index 71c50dd8f87..c99a41e29fe 100644 --- a/arch/ia64/mm/extable.c +++ b/arch/ia64/mm/extable.c @@ -8,7 +8,7 @@ #include <linux/sort.h> #include <asm/uaccess.h> -#include <asm/module.h> +#include <linux/module.h> static int cmp_ex(const void *a, const void *b) { @@ -53,6 +53,32 @@ void sort_extable (struct exception_table_entry *start, cmp_ex, swap_ex); } +static inline unsigned long ex_to_addr(const struct exception_table_entry *x) +{ + return (unsigned long)&x->addr + x->addr; +} + +#ifdef CONFIG_MODULES +/* + * Any entry referring to the module init will be at the beginning or + * the end. + */ +void trim_init_extable(struct module *m) +{ + /*trim the beginning*/ + while (m->num_exentries && + within_module_init(ex_to_addr(&m->extable[0]), m)) { + m->extable++; + m->num_exentries--; + } + /*trim the end*/ + while (m->num_exentries && + within_module_init(ex_to_addr(&m->extable[m->num_exentries-1]), + m)) + m->num_exentries--; +} +#endif /* CONFIG_MODULES */ + const struct exception_table_entry * search_extable (const struct exception_table_entry *first, const struct exception_table_entry *last, diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c index 14ef7cceb20..7225dad8709 100644 --- a/arch/ia64/mm/fault.c +++ b/arch/ia64/mm/fault.c @@ -7,49 +7,36 @@ #include <linux/sched.h> #include <linux/kernel.h> #include <linux/mm.h> -#include <linux/smp_lock.h> #include <linux/interrupt.h> #include <linux/kprobes.h> +#include <linux/kdebug.h> +#include <linux/prefetch.h> #include <asm/pgtable.h> #include <asm/processor.h> -#include <asm/system.h> #include <asm/uaccess.h> -#include <asm/kdebug.h> -extern void die (char *, struct pt_regs *, long); +extern int die(char *, struct pt_regs *, long); #ifdef CONFIG_KPROBES -ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain); - -/* Hook to register for page fault notifications */ -int register_page_fault_notifier(struct notifier_block *nb) -{ - return atomic_notifier_chain_register(¬ify_page_fault_chain, nb); -} - -int unregister_page_fault_notifier(struct notifier_block *nb) +static inline int notify_page_fault(struct pt_regs *regs, int trap) { - return atomic_notifier_chain_unregister(¬ify_page_fault_chain, nb); -} + int ret = 0; + + if (!user_mode(regs)) { + /* kprobe_running() needs smp_processor_id() */ + preempt_disable(); + if (kprobe_running() && kprobe_fault_handler(regs, trap)) + ret = 1; + preempt_enable(); + } -static inline int notify_page_fault(enum die_val val, const char *str, - struct pt_regs *regs, long err, int trap, int sig) -{ - struct die_args args = { - .regs = regs, - .str = str, - .err = err, - .trapnr = trap, - .signr = sig - }; - return atomic_notifier_call_chain(¬ify_page_fault_chain, val, &args); + return ret; } #else -static inline int notify_page_fault(enum die_val val, const char *str, - struct pt_regs *regs, long err, int trap, int sig) +static inline int notify_page_fault(struct pt_regs *regs, int trap) { - return NOTIFY_DONE; + return 0; } #endif @@ -85,6 +72,10 @@ mapped_kernel_page_is_present (unsigned long address) return pte_present(pte); } +# define VM_READ_BIT 0 +# define VM_WRITE_BIT 1 +# define VM_EXEC_BIT 2 + void __kprobes ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *regs) { @@ -93,6 +84,11 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re struct mm_struct *mm = current->mm; struct siginfo si; unsigned long mask; + int fault; + unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE; + + mask = ((((isr >> IA64_ISR_X_BIT) & 1UL) << VM_EXEC_BIT) + | (((isr >> IA64_ISR_W_BIT) & 1UL) << VM_WRITE_BIT)); /* mmap_sem is performance critical.... */ prefetchw(&mm->mmap_sem); @@ -118,18 +114,28 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re /* * This is to handle the kprobes on user space access instructions */ - if (notify_page_fault(DIE_PAGE_FAULT, "page fault", regs, code, TRAP_BRKPT, - SIGSEGV) == NOTIFY_STOP) + if (notify_page_fault(regs, TRAP_BRKPT)) return; + if (user_mode(regs)) + flags |= FAULT_FLAG_USER; + if (mask & VM_WRITE) + flags |= FAULT_FLAG_WRITE; +retry: down_read(&mm->mmap_sem); vma = find_vma_prev(mm, address, &prev_vma); - if (!vma) + if (!vma && !prev_vma ) goto bad_area; - /* find_vma_prev() returns vma such that address < vma->vm_end or NULL */ - if (address < vma->vm_start) + /* + * find_vma_prev() returns vma such that address < vma->vm_end or NULL + * + * May find no vma, but could be that the last vm area is the + * register backing store that needs to expand upwards, in + * this case vma will be null, but prev_vma will ne non-null + */ + if (( !vma && prev_vma ) || (address < vma->vm_start) ) goto check_expansion; good_area: @@ -137,53 +143,67 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re /* OK, we've got a good vm_area for this memory area. Check the access permissions: */ -# define VM_READ_BIT 0 -# define VM_WRITE_BIT 1 -# define VM_EXEC_BIT 2 - # if (((1 << VM_READ_BIT) != VM_READ || (1 << VM_WRITE_BIT) != VM_WRITE) \ || (1 << VM_EXEC_BIT) != VM_EXEC) # error File is out of sync with <linux/mm.h>. Please update. # endif - mask = ( (((isr >> IA64_ISR_X_BIT) & 1UL) << VM_EXEC_BIT) - | (((isr >> IA64_ISR_W_BIT) & 1UL) << VM_WRITE_BIT) - | (((isr >> IA64_ISR_R_BIT) & 1UL) << VM_READ_BIT)); + if (((isr >> IA64_ISR_R_BIT) & 1UL) && (!(vma->vm_flags & (VM_READ | VM_WRITE)))) + goto bad_area; if ((vma->vm_flags & mask) != mask) goto bad_area; - survive: /* * If for any reason at all we couldn't handle the fault, make * sure we exit gracefully rather than endlessly redo the * fault. */ - switch (handle_mm_fault(mm, vma, address, (mask & VM_WRITE) != 0)) { - case VM_FAULT_MINOR: - ++current->min_flt; - break; - case VM_FAULT_MAJOR: - ++current->maj_flt; - break; - case VM_FAULT_SIGBUS: + fault = handle_mm_fault(mm, vma, address, flags); + + if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) + return; + + if (unlikely(fault & VM_FAULT_ERROR)) { /* * We ran out of memory, or some other thing happened * to us that made us unable to handle the page fault * gracefully. */ - signal = SIGBUS; - goto bad_area; - case VM_FAULT_OOM: - goto out_of_memory; - default: + if (fault & VM_FAULT_OOM) { + goto out_of_memory; + } else if (fault & VM_FAULT_SIGBUS) { + signal = SIGBUS; + goto bad_area; + } BUG(); } + + if (flags & FAULT_FLAG_ALLOW_RETRY) { + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; + if (fault & VM_FAULT_RETRY) { + flags &= ~FAULT_FLAG_ALLOW_RETRY; + flags |= FAULT_FLAG_TRIED; + + /* No need to up_read(&mm->mmap_sem) as we would + * have already released it in __lock_page_or_retry + * in mm/filemap.c. + */ + + goto retry; + } + } + up_read(&mm->mmap_sem); return; check_expansion: if (!(prev_vma && (prev_vma->vm_flags & VM_GROWSUP) && (address == prev_vma->vm_end))) { + if (!vma) + goto bad_area; if (!(vma->vm_flags & VM_GROWSDOWN)) goto bad_area; if (REGION_NUMBER(address) != REGION_NUMBER(vma->vm_start) @@ -271,20 +291,16 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re else printk(KERN_ALERT "Unable to handle kernel paging request at " "virtual address %016lx\n", address); - die("Oops", regs, isr); + if (die("Oops", regs, isr)) + regs = NULL; bust_spinlocks(0); - do_exit(SIGKILL); + if (regs) + do_exit(SIGKILL); return; out_of_memory: up_read(&mm->mmap_sem); - if (current->pid == 1) { - yield(); - down_read(&mm->mmap_sem); - goto survive; - } - printk(KERN_CRIT "VM: killing process %s\n", current->comm); - if (user_mode(regs)) - do_exit(SIGKILL); - goto no_context; + if (!user_mode(regs)) + goto no_context; + pagefault_out_of_memory(); } diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index eee5c1cfbe3..76069c18ee4 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -13,18 +13,19 @@ #include <linux/mm.h> #include <linux/hugetlb.h> #include <linux/pagemap.h> -#include <linux/smp_lock.h> -#include <linux/slab.h> +#include <linux/module.h> #include <linux/sysctl.h> +#include <linux/log2.h> #include <asm/mman.h> #include <asm/pgalloc.h> #include <asm/tlb.h> #include <asm/tlbflush.h> -unsigned int hpage_shift=HPAGE_SHIFT_DEFAULT; +unsigned int hpage_shift = HPAGE_SHIFT_DEFAULT; +EXPORT_SYMBOL(hpage_shift); pte_t * -huge_pte_alloc (struct mm_struct *mm, unsigned long addr) +huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz) { unsigned long taddr = htlbpage_to_page(addr); pgd_t *pgd; @@ -37,7 +38,7 @@ huge_pte_alloc (struct mm_struct *mm, unsigned long addr) if (pud) { pmd = pmd_alloc(mm, pud, taddr); if (pmd) - pte = pte_alloc_map(mm, pmd, taddr); + pte = pte_alloc_map(mm, NULL, pmd, taddr); } return pte; } @@ -64,13 +65,19 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr) return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + #define mk_pte_huge(entry) { pte_val(entry) |= _PAGE_P; } /* * Don't actually need to do any preparation, but need to make sure * the address is in the right region. */ -int prepare_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(struct file *file, + unsigned long addr, unsigned long len) { if (len & ~HPAGE_MASK) return -EINVAL; @@ -101,13 +108,19 @@ int pmd_huge(pmd_t pmd) { return 0; } + +int pud_huge(pud_t pud) +{ + return 0; +} + struct page * follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write) { return NULL; } -void hugetlb_free_pgd_range(struct mmu_gather **tlb, +void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) { @@ -135,25 +148,31 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { - struct vm_area_struct *vmm; + struct vm_unmapped_area_info info; if (len > RGN_MAP_LIMIT) return -ENOMEM; if (len & ~HPAGE_MASK) return -EINVAL; + + /* Handle MAP_FIXED */ + if (flags & MAP_FIXED) { + if (prepare_hugepage_range(file, addr, len)) + return -EINVAL; + return addr; + } + /* This code assumes that RGN_HPAGE != 0. */ if ((REGION_NUMBER(addr) != RGN_HPAGE) || (addr & (HPAGE_SIZE - 1))) addr = HPAGE_REGION_BASE; - else - addr = ALIGN(addr, HPAGE_SIZE); - for (vmm = find_vma(current->mm, addr); ; vmm = vmm->vm_next) { - /* At this point: (!vmm || addr < vmm->vm_end). */ - if (REGION_OFFSET(addr) + len > RGN_MAP_LIMIT) - return -ENOMEM; - if (!vmm || (addr + len) <= vmm->vm_start) - return addr; - addr = ALIGN(vmm->vm_end, HPAGE_SIZE); - } + + info.flags = 0; + info.length = len; + info.low_limit = addr; + info.high_limit = HPAGE_REGION_BASE + RGN_MAP_LIMIT; + info.align_mask = PAGE_MASK & (HPAGE_SIZE - 1); + info.align_offset = 0; + return vm_unmapped_area(&info); } static int __init hugetlb_setup_sz(char *str) @@ -168,7 +187,7 @@ static int __init hugetlb_setup_sz(char *str) tr_pages = 0x15557000UL; size = memparse(str, &str); - if (*str || (size & (size-1)) || !(tr_pages & size) || + if (*str || !is_power_of_2(size) || !(tr_pages & size) || size <= PAGE_SIZE || size >= (1UL << PAGE_SHIFT << MAX_ORDER)) { printk(KERN_WARNING "Invalid huge page size specified\n"); @@ -181,6 +200,6 @@ static int __init hugetlb_setup_sz(char *str) * override here with new page shift. */ ia64_set_rr(HPAGE_REGION_BASE, hpage_shift << 2); - return 1; + return 0; } -__setup("hugepagesz=", hugetlb_setup_sz); +early_param("hugepagesz", hugetlb_setup_sz); diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index ff87a5cba39..25c350264a4 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -10,6 +10,7 @@ #include <linux/bootmem.h> #include <linux/efi.h> #include <linux/elf.h> +#include <linux/memblock.h> #include <linux/mm.h> #include <linux/mmzone.h> #include <linux/module.h> @@ -19,10 +20,9 @@ #include <linux/swap.h> #include <linux/proc_fs.h> #include <linux/bitops.h> +#include <linux/kexec.h> -#include <asm/a.out.h> #include <asm/dma.h> -#include <asm/ia32.h> #include <asm/io.h> #include <asm/machvec.h> #include <asm/numa.h> @@ -30,24 +30,19 @@ #include <asm/pgalloc.h> #include <asm/sal.h> #include <asm/sections.h> -#include <asm/system.h> #include <asm/tlb.h> #include <asm/uaccess.h> #include <asm/unistd.h> #include <asm/mca.h> - -DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); - -DEFINE_PER_CPU(unsigned long *, __pgtable_quicklist); -DEFINE_PER_CPU(long, __pgtable_quicklist_size); +#include <asm/paravirt.h> extern void ia64_tlb_init (void); unsigned long MAX_DMA_ADDRESS = PAGE_OFFSET + 0x100000000UL; #ifdef CONFIG_VIRTUAL_MEM_MAP -unsigned long vmalloc_end = VMALLOC_END_INIT; -EXPORT_SYMBOL(vmalloc_end); +unsigned long VMALLOC_END = VMALLOC_END_INIT; +EXPORT_SYMBOL(VMALLOC_END); struct page *vmem_map; EXPORT_SYMBOL(vmem_map); #endif @@ -55,63 +50,11 @@ EXPORT_SYMBOL(vmem_map); struct page *zero_page_memmap_ptr; /* map entry for zero page */ EXPORT_SYMBOL(zero_page_memmap_ptr); -#define MIN_PGT_PAGES 25UL -#define MAX_PGT_FREES_PER_PASS 16L -#define PGT_FRACTION_OF_NODE_MEM 16 - -static inline long -max_pgt_pages(void) -{ - u64 node_free_pages, max_pgt_pages; - -#ifndef CONFIG_NUMA - node_free_pages = nr_free_pages(); -#else - node_free_pages = nr_free_pages_pgdat(NODE_DATA(numa_node_id())); -#endif - max_pgt_pages = node_free_pages / PGT_FRACTION_OF_NODE_MEM; - max_pgt_pages = max(max_pgt_pages, MIN_PGT_PAGES); - return max_pgt_pages; -} - -static inline long -min_pages_to_free(void) -{ - long pages_to_free; - - pages_to_free = pgtable_quicklist_size - max_pgt_pages(); - pages_to_free = min(pages_to_free, MAX_PGT_FREES_PER_PASS); - return pages_to_free; -} - void -check_pgt_cache(void) -{ - long pages_to_free; - - if (unlikely(pgtable_quicklist_size <= MIN_PGT_PAGES)) - return; - - preempt_disable(); - while (unlikely((pages_to_free = min_pages_to_free()) > 0)) { - while (pages_to_free--) { - free_page((unsigned long)pgtable_quicklist_alloc()); - } - preempt_enable(); - preempt_disable(); - } - preempt_enable(); -} - -void -lazy_mmu_prot_update (pte_t pte) +__ia64_sync_icache_dcache (pte_t pte) { unsigned long addr; struct page *page; - unsigned long order; - - if (!pte_exec(pte)) - return; /* not an executable page... */ page = pte_page(pte); addr = (unsigned long) page_address(page); @@ -119,23 +62,37 @@ lazy_mmu_prot_update (pte_t pte) if (test_bit(PG_arch_1, &page->flags)) return; /* i-cache is already coherent with d-cache */ - if (PageCompound(page)) { - order = (unsigned long) (page[1].lru.prev); - flush_icache_range(addr, addr + (1UL << order << PAGE_SHIFT)); - } - else - flush_icache_range(addr, addr + PAGE_SIZE); + flush_icache_range(addr, addr + (PAGE_SIZE << compound_order(page))); set_bit(PG_arch_1, &page->flags); /* mark page as clean */ } +/* + * Since DMA is i-cache coherent, any (complete) pages that were written via + * DMA can be marked as "clean" so that lazy_mmu_prot_update() doesn't have to + * flush them when they get mapped into an executable vm-area. + */ +void +dma_mark_clean(void *addr, size_t size) +{ + unsigned long pg_addr, end; + + pg_addr = PAGE_ALIGN((unsigned long) addr); + end = (unsigned long) addr + size; + while (pg_addr + PAGE_SIZE <= end) { + struct page *page = virt_to_page(pg_addr); + set_bit(PG_arch_1, &page->flags); + pg_addr += PAGE_SIZE; + } +} + inline void ia64_set_rbs_bot (void) { - unsigned long stack_size = current->signal->rlim[RLIMIT_STACK].rlim_max & -16; + unsigned long stack_size = rlimit_max(RLIMIT_STACK) & -16; if (stack_size > MAX_USER_STACK_SIZE) stack_size = MAX_USER_STACK_SIZE; - current->thread.rbs_bot = STACK_TOP - stack_size; + current->thread.rbs_bot = PAGE_ALIGN(current->mm->start_stack - stack_size); } /* @@ -156,14 +113,14 @@ ia64_init_addr_space (void) * the problem. When the process attempts to write to the register backing store * for the first time, it will get a SEGFAULT in this case. */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (vma) { - memset(vma, 0, sizeof(*vma)); + INIT_LIST_HEAD(&vma->anon_vma_chain); vma->vm_mm = current->mm; vma->vm_start = current->thread.rbs_bot & PAGE_MASK; vma->vm_end = vma->vm_start + PAGE_SIZE; - vma->vm_page_prot = protection_map[VM_DATA_DEFAULT_FLAGS & 0x7]; vma->vm_flags = VM_DATA_DEFAULT_FLAGS|VM_GROWSUP|VM_ACCOUNT; + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); down_write(¤t->mm->mmap_sem); if (insert_vm_struct(current->mm, vma)) { up_write(¤t->mm->mmap_sem); @@ -175,13 +132,14 @@ ia64_init_addr_space (void) /* map NaT-page at address zero to speed up speculative dereferencing of NULL: */ if (!(current->personality & MMAP_PAGE_ZERO)) { - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (vma) { - memset(vma, 0, sizeof(*vma)); + INIT_LIST_HEAD(&vma->anon_vma_chain); vma->vm_mm = current->mm; vma->vm_end = PAGE_SIZE; vma->vm_page_prot = __pgprot(pgprot_val(PAGE_READONLY) | _PAGE_MA_NAT); - vma->vm_flags = VM_READ | VM_MAYREAD | VM_IO | VM_RESERVED; + vma->vm_flags = VM_READ | VM_MAYREAD | VM_IO | + VM_DONTEXPAND | VM_DONTDUMP; down_write(¤t->mm->mmap_sem); if (insert_vm_struct(current->mm, vma)) { up_write(¤t->mm->mmap_sem); @@ -196,25 +154,13 @@ ia64_init_addr_space (void) void free_initmem (void) { - unsigned long addr, eaddr; - - addr = (unsigned long) ia64_imva(__init_begin); - eaddr = (unsigned long) ia64_imva(__init_end); - while (addr < eaddr) { - ClearPageReserved(virt_to_page(addr)); - init_page_count(virt_to_page(addr)); - free_page(addr); - ++totalram_pages; - addr += PAGE_SIZE; - } - printk(KERN_INFO "Freeing unused kernel memory: %ldkB freed\n", - (__init_end - __init_begin) >> 10); + free_reserved_area(ia64_imva(__init_begin), ia64_imva(__init_end), + -1, "unused kernel"); } void __init free_initrd_mem (unsigned long start, unsigned long end) { - struct page *page; /* * EFI uses 4KB pages while the kernel can use 4KB or bigger. * Thus EFI and the kernel may have different page sizes. It is @@ -255,11 +201,7 @@ free_initrd_mem (unsigned long start, unsigned long end) for (; start < end; start += PAGE_SIZE) { if (!virt_addr_valid(start)) continue; - page = virt_to_page(start); - ClearPageReserved(page); - init_page_count(page); - free_page(start); - ++totalram_pages; + free_reserved_page(virt_to_page(start)); } } @@ -302,6 +244,7 @@ put_kernel_page (struct page *page, unsigned long address, pgprot_t pgprot) static void __init setup_gate (void) { + void *gate_section; struct page *page; /* @@ -309,10 +252,11 @@ setup_gate (void) * headers etc. and once execute-only page to enable * privilege-promotion via "epc": */ - page = virt_to_page(ia64_imva(__start_gate_section)); + gate_section = paravirt_get_gate_section(); + page = virt_to_page(ia64_imva(gate_section)); put_kernel_page(page, GATE_ADDR, PAGE_READONLY); #ifdef HAVE_BUGGY_SEGREL - page = virt_to_page(ia64_imva(__start_gate_section + PAGE_SIZE)); + page = virt_to_page(ia64_imva(gate_section + PAGE_SIZE)); put_kernel_page(page, GATE_ADDR + PAGE_SIZE, PAGE_GATE); #else put_kernel_page(page, GATE_ADDR + PERCPU_PAGE_SIZE, PAGE_GATE); @@ -334,11 +278,10 @@ setup_gate (void) ia64_patch_gate(); } -void __devinit -ia64_mmu_init (void *my_cpu_data) +void ia64_mmu_init(void *my_cpu_data) { - unsigned long psr, pta, impl_va_bits; - extern void __devinit tlb_init (void); + unsigned long pta, impl_va_bits; + extern void tlb_init(void); #ifdef CONFIG_DISABLE_VHPT # define VHPT_ENABLE_BIT 0 @@ -346,15 +289,6 @@ ia64_mmu_init (void *my_cpu_data) # define VHPT_ENABLE_BIT 1 #endif - /* Pin mapping for percpu area into TLB */ - psr = ia64_clear_ic(); - ia64_itr(0x2, IA64_TR_PERCPU_DATA, PERCPU_ADDR, - pte_val(pfn_pte(__pa(my_cpu_data) >> PAGE_SHIFT, PAGE_KERNEL)), - PERCPU_PAGE_SHIFT); - - ia64_set_psr(psr); - ia64_srlz_i(); - /* * Check if the virtually mapped linear page table (VMLPT) overlaps with a mapped * address space. The IA-64 architecture guarantees that at least 50 bits of @@ -423,9 +357,7 @@ int vmemmap_find_next_valid_pfn(int node, int i) end_address = (unsigned long) &vmem_map[pgdat->node_start_pfn + i]; end_address = PAGE_ALIGN(end_address); - - stop_address = (unsigned long) &vmem_map[ - pgdat->node_start_pfn + pgdat->node_spanned_pages]; + stop_address = (unsigned long) &vmem_map[pgdat_end_pfn(pgdat)]; do { pgd_t *pgd; @@ -471,8 +403,7 @@ retry_pte: return hole_next_pfn - pgdat->node_start_pfn; } -int __init -create_mem_map_page_table (u64 start, u64 end, void *arg) +int __init create_mem_map_page_table(u64 start, u64 end, void *arg) { unsigned long address, start_page, end_page; struct page *map_start, *map_end; @@ -517,8 +448,8 @@ struct memmap_init_callback_data { unsigned long zone; }; -static int -virtual_memmap_init (u64 start, u64 end, void *arg) +static int __meminit +virtual_memmap_init(u64 start, u64 end, void *arg) { struct memmap_init_callback_data *args; struct page *map_start, *map_end; @@ -543,16 +474,17 @@ virtual_memmap_init (u64 start, u64 end, void *arg) if (map_start < map_end) memmap_init_zone((unsigned long)(map_end - map_start), - args->nid, args->zone, page_to_pfn(map_start)); + args->nid, args->zone, page_to_pfn(map_start), + MEMMAP_EARLY); return 0; } -void +void __meminit memmap_init (unsigned long size, int nid, unsigned long zone, unsigned long start_pfn) { if (!vmem_map) - memmap_init_zone(size, nid, zone, start_pfn); + memmap_init_zone(size, nid, zone, start_pfn, MEMMAP_EARLY); else { struct page *start; struct memmap_init_callback_data args; @@ -579,8 +511,7 @@ ia64_pfn_valid (unsigned long pfn) } EXPORT_SYMBOL(ia64_pfn_valid); -int __init -find_largest_hole (u64 start, u64 end, void *arg) +int __init find_largest_hole(u64 start, u64 end, void *arg) { u64 *max_gap = arg; @@ -594,29 +525,37 @@ find_largest_hole (u64 start, u64 end, void *arg) return 0; } -int __init -register_active_ranges(u64 start, u64 end, void *nid) +#endif /* CONFIG_VIRTUAL_MEM_MAP */ + +int __init register_active_ranges(u64 start, u64 len, int nid) { - BUG_ON(nid == NULL); - BUG_ON(*(unsigned long *)nid >= MAX_NUMNODES); + u64 end = start + len; - add_active_range(*(unsigned long *)nid, - __pa(start) >> PAGE_SHIFT, - __pa(end) >> PAGE_SHIFT); +#ifdef CONFIG_KEXEC + if (start > crashk_res.start && start < crashk_res.end) + start = crashk_res.end; + if (end > crashk_res.start && end < crashk_res.end) + end = crashk_res.start; +#endif + + if (start < end) + memblock_add_node(__pa(start), end - start, nid); return 0; } -#endif /* CONFIG_VIRTUAL_MEM_MAP */ -static int __init -count_reserved_pages (u64 start, u64 end, void *arg) +int +find_max_min_low_pfn (u64 start, u64 end, void *arg) { - unsigned long num_reserved = 0; - unsigned long *count = arg; - - for (; start < end; start += PAGE_SIZE) - if (PageReserved(virt_to_page(start))) - ++num_reserved; - *count += num_reserved; + unsigned long pfn_start, pfn_end; +#ifdef CONFIG_FLATMEM + pfn_start = (PAGE_ALIGN(__pa(start))) >> PAGE_SHIFT; + pfn_end = (PAGE_ALIGN(__pa(end - 1))) >> PAGE_SHIFT; +#else + pfn_start = GRANULEROUNDDOWN(__pa(start)) >> PAGE_SHIFT; + pfn_end = GRANULEROUNDUP(__pa(end - 1)) >> PAGE_SHIFT; +#endif + min_low_pfn = min(min_low_pfn, pfn_start); + max_low_pfn = max(max_low_pfn, pfn_end); return 0; } @@ -642,10 +581,7 @@ __setup("nolwsys", nolwsys_setup); void __init mem_init (void) { - long reserved_pages, codesize, datasize, initsize; - pg_data_t *pgdat; int i; - static struct kcore_list kcore_mem, kcore_vmem, kcore_kernel; BUG_ON(PTRS_PER_PGD * sizeof(pgd_t) != PAGE_SIZE); BUG_ON(PTRS_PER_PMD * sizeof(pmd_t) != PAGE_SIZE); @@ -661,33 +597,13 @@ mem_init (void) #endif #ifdef CONFIG_FLATMEM - if (!mem_map) - BUG(); - max_mapnr = max_low_pfn; + BUG_ON(!mem_map); #endif + set_max_mapnr(max_low_pfn); high_memory = __va(max_low_pfn * PAGE_SIZE); - - kclist_add(&kcore_mem, __va(0), max_low_pfn * PAGE_SIZE); - kclist_add(&kcore_vmem, (void *)VMALLOC_START, VMALLOC_END-VMALLOC_START); - kclist_add(&kcore_kernel, _stext, _end - _stext); - - for_each_online_pgdat(pgdat) - if (pgdat->bdata->node_bootmem_map) - totalram_pages += free_all_bootmem_node(pgdat); - - reserved_pages = 0; - efi_memmap_walk(count_reserved_pages, &reserved_pages); - - codesize = (unsigned long) _etext - (unsigned long) _stext; - datasize = (unsigned long) _edata - (unsigned long) _etext; - initsize = (unsigned long) __init_end - (unsigned long) __init_begin; - - printk(KERN_INFO "Memory: %luk/%luk available (%luk code, %luk reserved, " - "%luk data, %luk init)\n", (unsigned long) nr_free_pages() << (PAGE_SHIFT - 10), - num_physpages << (PAGE_SHIFT - 10), codesize >> 10, - reserved_pages << (PAGE_SHIFT - 10), datasize >> 10, initsize >> 10); - + free_all_bootmem(); + mem_init_print_info(NULL); /* * For fsyscall entrpoints with no light-weight handler, use the ordinary @@ -695,29 +611,16 @@ mem_init (void) * code can tell them apart. */ for (i = 0; i < NR_syscalls; ++i) { - extern unsigned long fsyscall_table[NR_syscalls]; extern unsigned long sys_call_table[NR_syscalls]; + unsigned long *fsyscall_table = paravirt_get_fsyscall_table(); if (!fsyscall_table[i] || nolwsys) fsyscall_table[i] = sys_call_table[i] | 1; } setup_gate(); - -#ifdef CONFIG_IA32_SUPPORT - ia32_mem_init(); -#endif } #ifdef CONFIG_MEMORY_HOTPLUG -void online_page(struct page *page) -{ - ClearPageReserved(page); - init_page_count(page); - __free_page(page); - totalram_pages++; - num_physpages++; -} - int arch_add_memory(int nid, u64 start, u64 size) { pg_data_t *pgdat; @@ -729,18 +632,103 @@ int arch_add_memory(int nid, u64 start, u64 size) pgdat = NODE_DATA(nid); zone = pgdat->node_zones + ZONE_NORMAL; - ret = __add_pages(zone, start_pfn, nr_pages); + ret = __add_pages(nid, zone, start_pfn, nr_pages); if (ret) printk("%s: Problem encountered in __add_pages() as ret=%d\n", - __FUNCTION__, ret); + __func__, ret); return ret; } -int remove_memory(u64 start, u64 size) +#ifdef CONFIG_MEMORY_HOTREMOVE +int arch_remove_memory(u64 start, u64 size) { - return -EINVAL; + unsigned long start_pfn = start >> PAGE_SHIFT; + unsigned long nr_pages = size >> PAGE_SHIFT; + struct zone *zone; + int ret; + + zone = page_zone(pfn_to_page(start_pfn)); + ret = __remove_pages(zone, start_pfn, nr_pages); + if (ret) + pr_warn("%s: Problem encountered in __remove_pages() as" + " ret=%d\n", __func__, ret); + + return ret; } -EXPORT_SYMBOL_GPL(remove_memory); #endif +#endif + +/* + * Even when CONFIG_IA32_SUPPORT is not enabled it is + * useful to have the Linux/x86 domain registered to + * avoid an attempted module load when emulators call + * personality(PER_LINUX32). This saves several milliseconds + * on each such call. + */ +static struct exec_domain ia32_exec_domain; + +static int __init +per_linux32_init(void) +{ + ia32_exec_domain.name = "Linux/x86"; + ia32_exec_domain.handler = NULL; + ia32_exec_domain.pers_low = PER_LINUX32; + ia32_exec_domain.pers_high = PER_LINUX32; + ia32_exec_domain.signal_map = default_exec_domain.signal_map; + ia32_exec_domain.signal_invmap = default_exec_domain.signal_invmap; + register_exec_domain(&ia32_exec_domain); + + return 0; +} + +__initcall(per_linux32_init); + +/** + * show_mem - give short summary of memory stats + * + * Shows a simple page count of reserved and used pages in the system. + * For discontig machines, it does this on a per-pgdat basis. + */ +void show_mem(unsigned int filter) +{ + int total_reserved = 0; + unsigned long total_present = 0; + pg_data_t *pgdat; + + printk(KERN_INFO "Mem-info:\n"); + show_free_areas(filter); + printk(KERN_INFO "Node memory in pages:\n"); + for_each_online_pgdat(pgdat) { + unsigned long present; + unsigned long flags; + int reserved = 0; + int nid = pgdat->node_id; + int zoneid; + + if (skip_free_areas_node(filter, nid)) + continue; + pgdat_resize_lock(pgdat, &flags); + + for (zoneid = 0; zoneid < MAX_NR_ZONES; zoneid++) { + struct zone *zone = &pgdat->node_zones[zoneid]; + if (!populated_zone(zone)) + continue; + + reserved += zone->present_pages - zone->managed_pages; + } + present = pgdat->node_present_pages; + + pgdat_resize_unlock(pgdat, &flags); + total_present += present; + total_reserved += reserved; + printk(KERN_INFO "Node %4d: RAM: %11ld, rsvd: %8d, ", + nid, present, reserved); + } + printk(KERN_INFO "%ld pages of RAM\n", total_present); + printk(KERN_INFO "%d reserved pages\n", total_reserved); + printk(KERN_INFO "Total of %ld pages in page table cache\n", + quicklist_total_size()); + printk(KERN_INFO "%ld free buffer pages\n", nr_free_buffer_pages()); +} diff --git a/arch/ia64/mm/ioremap.c b/arch/ia64/mm/ioremap.c index 4280c074d64..43964cde621 100644 --- a/arch/ia64/mm/ioremap.c +++ b/arch/ia64/mm/ioremap.c @@ -1,5 +1,5 @@ /* - * (c) Copyright 2006 Hewlett-Packard Development Company, L.P. + * (c) Copyright 2006, 2007 Hewlett-Packard Development Company, L.P. * Bjorn Helgaas <bjorn.helgaas@hp.com> * * This program is free software; you can redistribute it and/or modify @@ -10,51 +10,116 @@ #include <linux/compiler.h> #include <linux/module.h> #include <linux/efi.h> +#include <linux/io.h> +#include <linux/vmalloc.h> #include <asm/io.h> #include <asm/meminit.h> static inline void __iomem * -__ioremap (unsigned long offset, unsigned long size) +__ioremap_uc(unsigned long phys_addr) { - return (void __iomem *) (__IA64_UNCACHED_OFFSET | offset); + return (void __iomem *) (__IA64_UNCACHED_OFFSET | phys_addr); } void __iomem * -ioremap (unsigned long offset, unsigned long size) +early_ioremap (unsigned long phys_addr, unsigned long size) { u64 attr; + attr = kern_mem_attribute(phys_addr, size); + if (attr & EFI_MEMORY_WB) + return (void __iomem *) phys_to_virt(phys_addr); + return __ioremap_uc(phys_addr); +} + +void __iomem * +ioremap (unsigned long phys_addr, unsigned long size) +{ + void __iomem *addr; + struct vm_struct *area; + unsigned long offset; + pgprot_t prot; + u64 attr; unsigned long gran_base, gran_size; + unsigned long page_base; /* * For things in kern_memmap, we must use the same attribute * as the rest of the kernel. For more details, see * Documentation/ia64/aliasing.txt. */ - attr = kern_mem_attribute(offset, size); + attr = kern_mem_attribute(phys_addr, size); if (attr & EFI_MEMORY_WB) - return (void __iomem *) phys_to_virt(offset); + return (void __iomem *) phys_to_virt(phys_addr); else if (attr & EFI_MEMORY_UC) - return __ioremap(offset, size); + return __ioremap_uc(phys_addr); /* * Some chipsets don't support UC access to memory. If * WB is supported for the whole granule, we prefer that. */ - gran_base = GRANULEROUNDDOWN(offset); - gran_size = GRANULEROUNDUP(offset + size) - gran_base; + gran_base = GRANULEROUNDDOWN(phys_addr); + gran_size = GRANULEROUNDUP(phys_addr + size) - gran_base; if (efi_mem_attribute(gran_base, gran_size) & EFI_MEMORY_WB) - return (void __iomem *) phys_to_virt(offset); + return (void __iomem *) phys_to_virt(phys_addr); + + /* + * WB is not supported for the whole granule, so we can't use + * the region 7 identity mapping. If we can safely cover the + * area with kernel page table mappings, we can use those + * instead. + */ + page_base = phys_addr & PAGE_MASK; + size = PAGE_ALIGN(phys_addr + size) - page_base; + if (efi_mem_attribute(page_base, size) & EFI_MEMORY_WB) { + prot = PAGE_KERNEL; + + /* + * Mappings have to be page-aligned + */ + offset = phys_addr & ~PAGE_MASK; + phys_addr &= PAGE_MASK; - return __ioremap(offset, size); + /* + * Ok, go for it.. + */ + area = get_vm_area(size, VM_IOREMAP); + if (!area) + return NULL; + + area->phys_addr = phys_addr; + addr = (void __iomem *) area->addr; + if (ioremap_page_range((unsigned long) addr, + (unsigned long) addr + size, phys_addr, prot)) { + vunmap((void __force *) addr); + return NULL; + } + + return (void __iomem *) (offset + (char __iomem *)addr); + } + + return __ioremap_uc(phys_addr); } EXPORT_SYMBOL(ioremap); void __iomem * -ioremap_nocache (unsigned long offset, unsigned long size) +ioremap_nocache (unsigned long phys_addr, unsigned long size) { - if (kern_mem_attribute(offset, size) & EFI_MEMORY_WB) + if (kern_mem_attribute(phys_addr, size) & EFI_MEMORY_WB) return NULL; - return __ioremap(offset, size); + return __ioremap_uc(phys_addr); } EXPORT_SYMBOL(ioremap_nocache); + +void +early_iounmap (volatile void __iomem *addr, unsigned long size) +{ +} + +void +iounmap (volatile void __iomem *addr) +{ + if (REGION_NUMBER(addr) == RGN_GATE) + vunmap((void *) ((unsigned long) addr & PAGE_MASK)); +} +EXPORT_SYMBOL(iounmap); diff --git a/arch/ia64/mm/numa.c b/arch/ia64/mm/numa.c index 64e4c21f311..ea21d4cad54 100644 --- a/arch/ia64/mm/numa.c +++ b/arch/ia64/mm/numa.c @@ -16,6 +16,7 @@ #include <linux/node.h> #include <linux/init.h> #include <linux/bootmem.h> +#include <linux/module.h> #include <asm/mmzone.h> #include <asm/numa.h> @@ -26,7 +27,9 @@ */ int num_node_memblks; struct node_memblk_s node_memblk[NR_NODE_MEMBLKS]; -struct node_cpuid_s node_cpuid[NR_CPUS]; +struct node_cpuid_s node_cpuid[NR_CPUS] = + { [0 ... NR_CPUS-1] = { .phys_id = 0, .nid = NUMA_NO_NODE } }; + /* * This is a matrix with "distances" between nodes, they should be * proportional to the memory access latency ratios. @@ -55,18 +58,53 @@ paddr_to_nid(unsigned long paddr) * SPARSEMEM to allocate the SPARSEMEM sectionmap on the NUMA node where * the section resides. */ -int early_pfn_to_nid(unsigned long pfn) +int __meminit __early_pfn_to_nid(unsigned long pfn) { int i, section = pfn >> PFN_SECTION_SHIFT, ssec, esec; + /* + * NOTE: The following SMP-unsafe globals are only used early in boot + * when the kernel is running single-threaded. + */ + static int __meminitdata last_ssec, last_esec; + static int __meminitdata last_nid; + + if (section >= last_ssec && section < last_esec) + return last_nid; for (i = 0; i < num_node_memblks; i++) { ssec = node_memblk[i].start_paddr >> PA_SECTION_SHIFT; esec = (node_memblk[i].start_paddr + node_memblk[i].size + ((1L << PA_SECTION_SHIFT) - 1)) >> PA_SECTION_SHIFT; - if (section >= ssec && section < esec) + if (section >= ssec && section < esec) { + last_ssec = ssec; + last_esec = esec; + last_nid = node_memblk[i].nid; return node_memblk[i].nid; + } } - return 0; + return -1; } + +void numa_clear_node(int cpu) +{ + unmap_cpu_from_node(cpu, NUMA_NO_NODE); +} + +#ifdef CONFIG_MEMORY_HOTPLUG +/* + * SRAT information is stored in node_memblk[], then we can use SRAT + * information at memory-hot-add if necessary. + */ + +int memory_add_physaddr_to_nid(u64 addr) +{ + int nid = paddr_to_nid(addr); + if (nid < 0) + return 0; + return nid; +} + +EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid); +#endif #endif diff --git a/arch/ia64/mm/tlb.c b/arch/ia64/mm/tlb.c index ffad7624436..ed612976868 100644 --- a/arch/ia64/mm/tlb.c +++ b/arch/ia64/mm/tlb.c @@ -10,6 +10,10 @@ * IPI based ptc implementation and A-step IPI implementation. * Rohit Seth <rohit.seth@intel.com> * Ken Chen <kenneth.w.chen@intel.com> + * Christophe de Dinechin <ddd@hp.com>: Avoid ptc.e on memory allocation + * Copyright (C) 2007 Intel Corp + * Fenghua Yu <fenghua.yu@intel.com> + * Add multiple ptc.g/ptc.ga instruction support in global tlb purge. */ #include <linux/module.h> #include <linux/init.h> @@ -18,6 +22,7 @@ #include <linux/smp.h> #include <linux/mm.h> #include <linux/bootmem.h> +#include <linux/slab.h> #include <asm/delay.h> #include <asm/mmu_context.h> @@ -25,19 +30,26 @@ #include <asm/pal.h> #include <asm/tlbflush.h> #include <asm/dma.h> +#include <asm/processor.h> +#include <asm/sal.h> +#include <asm/tlb.h> static struct { - unsigned long mask; /* mask of supported purge page-sizes */ + u64 mask; /* mask of supported purge page-sizes */ unsigned long max_bits; /* log2 of largest supported purge page-size */ } purge; struct ia64_ctx ia64_ctx = { - .lock = SPIN_LOCK_UNLOCKED, - .next = 1, - .max_ctx = ~0U + .lock = __SPIN_LOCK_UNLOCKED(ia64_ctx.lock), + .next = 1, + .max_ctx = ~0U }; DEFINE_PER_CPU(u8, ia64_need_tlb_flush); +DEFINE_PER_CPU(u8, ia64_tr_num); /*Number of TR slots in current processor*/ +DEFINE_PER_CPU(u8, ia64_tr_used); /*Max Slot number used by kernel*/ + +struct ia64_tr_entry *ia64_idtrs[NR_CPUS]; /* * Initializes the ia64_ctx.bitmap array based on max_ctx+1. @@ -83,30 +95,183 @@ wrap_mmu_context (struct mm_struct *mm) local_flush_tlb_all(); } +/* + * Implement "spinaphores" ... like counting semaphores, but they + * spin instead of sleeping. If there are ever any other users for + * this primitive it can be moved up to a spinaphore.h header. + */ +struct spinaphore { + unsigned long ticket; + unsigned long serve; +}; + +static inline void spinaphore_init(struct spinaphore *ss, int val) +{ + ss->ticket = 0; + ss->serve = val; +} + +static inline void down_spin(struct spinaphore *ss) +{ + unsigned long t = ia64_fetchadd(1, &ss->ticket, acq), serve; + + if (time_before(t, ss->serve)) + return; + + ia64_invala(); + + for (;;) { + asm volatile ("ld8.c.nc %0=[%1]" : "=r"(serve) : "r"(&ss->serve) : "memory"); + if (time_before(t, serve)) + return; + cpu_relax(); + } +} + +static inline void up_spin(struct spinaphore *ss) +{ + ia64_fetchadd(1, &ss->serve, rel); +} + +static struct spinaphore ptcg_sem; +static u16 nptcg = 1; +static int need_ptcg_sem = 1; +static int toolatetochangeptcgsem = 0; + +/* + * Kernel parameter "nptcg=" overrides max number of concurrent global TLB + * purges which is reported from either PAL or SAL PALO. + * + * We don't have sanity checking for nptcg value. It's the user's responsibility + * for valid nptcg value on the platform. Otherwise, kernel may hang in some + * cases. + */ +static int __init +set_nptcg(char *str) +{ + int value = 0; + + get_option(&str, &value); + setup_ptcg_sem(value, NPTCG_FROM_KERNEL_PARAMETER); + + return 1; +} + +__setup("nptcg=", set_nptcg); + +/* + * Maximum number of simultaneous ptc.g purges in the system can + * be defined by PAL_VM_SUMMARY (in which case we should take + * the smallest value for any cpu in the system) or by the PAL + * override table (in which case we should ignore the value from + * PAL_VM_SUMMARY). + * + * Kernel parameter "nptcg=" overrides maximum number of simultanesous ptc.g + * purges defined in either PAL_VM_SUMMARY or PAL override table. In this case, + * we should ignore the value from either PAL_VM_SUMMARY or PAL override table. + * + * Complicating the logic here is the fact that num_possible_cpus() + * isn't fully setup until we start bringing cpus online. + */ void -ia64_global_tlb_purge (struct mm_struct *mm, unsigned long start, - unsigned long end, unsigned long nbits) +setup_ptcg_sem(int max_purges, int nptcg_from) { - static DEFINE_SPINLOCK(ptcg_lock); + static int kp_override; + static int palo_override; + static int firstcpu = 1; - if (mm != current->active_mm || !current->mm) { - flush_tlb_all(); + if (toolatetochangeptcgsem) { + if (nptcg_from == NPTCG_FROM_PAL && max_purges == 0) + BUG_ON(1 < nptcg); + else + BUG_ON(max_purges < nptcg); return; } - /* HW requires global serialization of ptc.ga. */ - spin_lock(&ptcg_lock); - { - do { - /* - * Flush ALAT entries also. - */ - ia64_ptcga(start, (nbits<<2)); - ia64_srlz_i(); - start += (1UL << nbits); - } while (start < end); + if (nptcg_from == NPTCG_FROM_KERNEL_PARAMETER) { + kp_override = 1; + nptcg = max_purges; + goto resetsema; + } + if (kp_override) { + need_ptcg_sem = num_possible_cpus() > nptcg; + return; + } + + if (nptcg_from == NPTCG_FROM_PALO) { + palo_override = 1; + + /* In PALO max_purges == 0 really means it! */ + if (max_purges == 0) + panic("Whoa! Platform does not support global TLB purges.\n"); + nptcg = max_purges; + if (nptcg == PALO_MAX_TLB_PURGES) { + need_ptcg_sem = 0; + return; + } + goto resetsema; + } + if (palo_override) { + if (nptcg != PALO_MAX_TLB_PURGES) + need_ptcg_sem = (num_possible_cpus() > nptcg); + return; + } + + /* In PAL_VM_SUMMARY max_purges == 0 actually means 1 */ + if (max_purges == 0) max_purges = 1; + + if (firstcpu) { + nptcg = max_purges; + firstcpu = 0; } - spin_unlock(&ptcg_lock); + if (max_purges < nptcg) + nptcg = max_purges; + if (nptcg == PAL_MAX_PURGES) { + need_ptcg_sem = 0; + return; + } else + need_ptcg_sem = (num_possible_cpus() > nptcg); + +resetsema: + spinaphore_init(&ptcg_sem, max_purges); +} + +void +ia64_global_tlb_purge (struct mm_struct *mm, unsigned long start, + unsigned long end, unsigned long nbits) +{ + struct mm_struct *active_mm = current->active_mm; + + toolatetochangeptcgsem = 1; + + if (mm != active_mm) { + /* Restore region IDs for mm */ + if (mm && active_mm) { + activate_context(mm); + } else { + flush_tlb_all(); + return; + } + } + + if (need_ptcg_sem) + down_spin(&ptcg_sem); + + do { + /* + * Flush ALAT entries also. + */ + ia64_ptcga(start, (nbits << 2)); + ia64_srlz_i(); + start += (1UL << nbits); + } while (start < end); + + if (need_ptcg_sem) + up_spin(&ptcg_sem); + + if (mm != active_mm) { + activate_context(active_mm); + } } void @@ -157,7 +322,7 @@ flush_tlb_range (struct vm_area_struct *vma, unsigned long start, preempt_disable(); #ifdef CONFIG_SMP - if (mm != current->active_mm || cpus_weight(mm->cpu_vm_mask) != 1) { + if (mm != current->active_mm || cpumask_weight(mm_cpumask(mm)) != 1) { platform_global_tlb_purge(mm, start, end, nbits); preempt_enable(); return; @@ -172,15 +337,17 @@ flush_tlb_range (struct vm_area_struct *vma, unsigned long start, } EXPORT_SYMBOL(flush_tlb_range); -void __devinit -ia64_tlb_init (void) +void ia64_tlb_init(void) { - ia64_ptce_info_t ptce_info; - unsigned long tr_pgbits; + ia64_ptce_info_t uninitialized_var(ptce_info); /* GCC be quiet */ + u64 tr_pgbits; long status; + pal_vm_info_1_u_t vm_info_1; + pal_vm_info_2_u_t vm_info_2; + int cpu = smp_processor_id(); if ((status = ia64_pal_vm_page_size(&tr_pgbits, &purge.mask)) != 0) { - printk(KERN_ERR "PAL_VM_PAGE_SIZE failed with status=%ld;" + printk(KERN_ERR "PAL_VM_PAGE_SIZE failed with status=%ld; " "defaulting to architected purge page-sizes.\n", status); purge.mask = 0x115557000UL; } @@ -194,4 +361,201 @@ ia64_tlb_init (void) local_cpu_data->ptce_stride[1] = ptce_info.stride[1]; local_flush_tlb_all(); /* nuke left overs from bootstrapping... */ + status = ia64_pal_vm_summary(&vm_info_1, &vm_info_2); + + if (status) { + printk(KERN_ERR "ia64_pal_vm_summary=%ld\n", status); + per_cpu(ia64_tr_num, cpu) = 8; + return; + } + per_cpu(ia64_tr_num, cpu) = vm_info_1.pal_vm_info_1_s.max_itr_entry+1; + if (per_cpu(ia64_tr_num, cpu) > + (vm_info_1.pal_vm_info_1_s.max_dtr_entry+1)) + per_cpu(ia64_tr_num, cpu) = + vm_info_1.pal_vm_info_1_s.max_dtr_entry+1; + if (per_cpu(ia64_tr_num, cpu) > IA64_TR_ALLOC_MAX) { + static int justonce = 1; + per_cpu(ia64_tr_num, cpu) = IA64_TR_ALLOC_MAX; + if (justonce) { + justonce = 0; + printk(KERN_DEBUG "TR register number exceeds " + "IA64_TR_ALLOC_MAX!\n"); + } + } +} + +/* + * is_tr_overlap + * + * Check overlap with inserted TRs. + */ +static int is_tr_overlap(struct ia64_tr_entry *p, u64 va, u64 log_size) +{ + u64 tr_log_size; + u64 tr_end; + u64 va_rr = ia64_get_rr(va); + u64 va_rid = RR_TO_RID(va_rr); + u64 va_end = va + (1<<log_size) - 1; + + if (va_rid != RR_TO_RID(p->rr)) + return 0; + tr_log_size = (p->itir & 0xff) >> 2; + tr_end = p->ifa + (1<<tr_log_size) - 1; + + if (va > tr_end || p->ifa > va_end) + return 0; + return 1; + +} + +/* + * ia64_insert_tr in virtual mode. Allocate a TR slot + * + * target_mask : 0x1 : itr, 0x2 : dtr, 0x3 : idtr + * + * va : virtual address. + * pte : pte entries inserted. + * log_size: range to be covered. + * + * Return value: <0 : error No. + * + * >=0 : slot number allocated for TR. + * Must be called with preemption disabled. + */ +int ia64_itr_entry(u64 target_mask, u64 va, u64 pte, u64 log_size) +{ + int i, r; + unsigned long psr; + struct ia64_tr_entry *p; + int cpu = smp_processor_id(); + + if (!ia64_idtrs[cpu]) { + ia64_idtrs[cpu] = kmalloc(2 * IA64_TR_ALLOC_MAX * + sizeof (struct ia64_tr_entry), GFP_KERNEL); + if (!ia64_idtrs[cpu]) + return -ENOMEM; + } + r = -EINVAL; + /*Check overlap with existing TR entries*/ + if (target_mask & 0x1) { + p = ia64_idtrs[cpu]; + for (i = IA64_TR_ALLOC_BASE; i <= per_cpu(ia64_tr_used, cpu); + i++, p++) { + if (p->pte & 0x1) + if (is_tr_overlap(p, va, log_size)) { + printk(KERN_DEBUG "Overlapped Entry" + "Inserted for TR Reigster!!\n"); + goto out; + } + } + } + if (target_mask & 0x2) { + p = ia64_idtrs[cpu] + IA64_TR_ALLOC_MAX; + for (i = IA64_TR_ALLOC_BASE; i <= per_cpu(ia64_tr_used, cpu); + i++, p++) { + if (p->pte & 0x1) + if (is_tr_overlap(p, va, log_size)) { + printk(KERN_DEBUG "Overlapped Entry" + "Inserted for TR Reigster!!\n"); + goto out; + } + } + } + + for (i = IA64_TR_ALLOC_BASE; i < per_cpu(ia64_tr_num, cpu); i++) { + switch (target_mask & 0x3) { + case 1: + if (!((ia64_idtrs[cpu] + i)->pte & 0x1)) + goto found; + continue; + case 2: + if (!((ia64_idtrs[cpu] + IA64_TR_ALLOC_MAX + i)->pte & 0x1)) + goto found; + continue; + case 3: + if (!((ia64_idtrs[cpu] + i)->pte & 0x1) && + !((ia64_idtrs[cpu] + IA64_TR_ALLOC_MAX + i)->pte & 0x1)) + goto found; + continue; + default: + r = -EINVAL; + goto out; + } + } +found: + if (i >= per_cpu(ia64_tr_num, cpu)) + return -EBUSY; + + /*Record tr info for mca hander use!*/ + if (i > per_cpu(ia64_tr_used, cpu)) + per_cpu(ia64_tr_used, cpu) = i; + + psr = ia64_clear_ic(); + if (target_mask & 0x1) { + ia64_itr(0x1, i, va, pte, log_size); + ia64_srlz_i(); + p = ia64_idtrs[cpu] + i; + p->ifa = va; + p->pte = pte; + p->itir = log_size << 2; + p->rr = ia64_get_rr(va); + } + if (target_mask & 0x2) { + ia64_itr(0x2, i, va, pte, log_size); + ia64_srlz_i(); + p = ia64_idtrs[cpu] + IA64_TR_ALLOC_MAX + i; + p->ifa = va; + p->pte = pte; + p->itir = log_size << 2; + p->rr = ia64_get_rr(va); + } + ia64_set_psr(psr); + r = i; +out: + return r; +} +EXPORT_SYMBOL_GPL(ia64_itr_entry); + +/* + * ia64_purge_tr + * + * target_mask: 0x1: purge itr, 0x2 : purge dtr, 0x3 purge idtr. + * slot: slot number to be freed. + * + * Must be called with preemption disabled. + */ +void ia64_ptr_entry(u64 target_mask, int slot) +{ + int cpu = smp_processor_id(); + int i; + struct ia64_tr_entry *p; + + if (slot < IA64_TR_ALLOC_BASE || slot >= per_cpu(ia64_tr_num, cpu)) + return; + + if (target_mask & 0x1) { + p = ia64_idtrs[cpu] + slot; + if ((p->pte&0x1) && is_tr_overlap(p, p->ifa, p->itir>>2)) { + p->pte = 0; + ia64_ptr(0x1, p->ifa, p->itir>>2); + ia64_srlz_i(); + } + } + + if (target_mask & 0x2) { + p = ia64_idtrs[cpu] + IA64_TR_ALLOC_MAX + slot; + if ((p->pte & 0x1) && is_tr_overlap(p, p->ifa, p->itir>>2)) { + p->pte = 0; + ia64_ptr(0x2, p->ifa, p->itir>>2); + ia64_srlz_i(); + } + } + + for (i = per_cpu(ia64_tr_used, cpu); i >= IA64_TR_ALLOC_BASE; i--) { + if (((ia64_idtrs[cpu] + i)->pte & 0x1) || + ((ia64_idtrs[cpu] + IA64_TR_ALLOC_MAX + i)->pte & 0x1)) + break; + } + per_cpu(ia64_tr_used, cpu) = i; } +EXPORT_SYMBOL_GPL(ia64_ptr_entry); |