diff options
author | Ingo Molnar <mingo@elte.hu> | 2008-07-08 11:14:58 +0200 |
---|---|---|
committer | Ingo Molnar <mingo@elte.hu> | 2008-07-08 11:14:58 +0200 |
commit | 3de352bbd86f890dd0c5e1c09a6a1b0b29e0f8ce (patch) | |
tree | d4c5eba8cd2abefd7c9f16d089393f0f5999cf63 /arch/x86/mm | |
parent | 1b8ba39a3fad9c58532f6dad12c94d6e675be656 (diff) | |
parent | 9340e1ccdf7b9b22a2be7f51cd74e8b5e11961bf (diff) | |
download | op-kernel-dev-3de352bbd86f890dd0c5e1c09a6a1b0b29e0f8ce.zip op-kernel-dev-3de352bbd86f890dd0c5e1c09a6a1b0b29e0f8ce.tar.gz |
Merge branch 'x86/mpparse' into x86/devel
Conflicts:
arch/x86/Kconfig
arch/x86/kernel/io_apic_32.c
arch/x86/kernel/setup_64.c
arch/x86/mm/init_32.c
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Diffstat (limited to 'arch/x86/mm')
-rw-r--r-- | arch/x86/mm/discontig_32.c | 186 | ||||
-rw-r--r-- | arch/x86/mm/init_32.c | 86 | ||||
-rw-r--r-- | arch/x86/mm/init_64.c | 30 | ||||
-rw-r--r-- | arch/x86/mm/k8topology_64.c | 4 | ||||
-rw-r--r-- | arch/x86/mm/numa_64.c | 2 |
5 files changed, 182 insertions, 126 deletions
diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c index 8b4eac0..a2f73ba 100644 --- a/arch/x86/mm/discontig_32.c +++ b/arch/x86/mm/discontig_32.c @@ -38,6 +38,7 @@ #include <asm/setup.h> #include <asm/mmzone.h> #include <asm/bios_ebda.h> +#include <asm/proto.h> struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; EXPORT_SYMBOL(node_data); @@ -59,14 +60,14 @@ unsigned long node_end_pfn[MAX_NUMNODES] __read_mostly; /* * 4) physnode_map - the mapping between a pfn and owning node * physnode_map keeps track of the physical memory layout of a generic - * numa node on a 256Mb break (each element of the array will - * represent 256Mb of memory and will be marked by the node id. so, + * numa node on a 64Mb break (each element of the array will + * represent 64Mb of memory and will be marked by the node id. so, * if the first gig is on node 0, and the second gig is on node 1 * physnode_map will contain: * - * physnode_map[0-3] = 0; - * physnode_map[4-7] = 1; - * physnode_map[8- ] = -1; + * physnode_map[0-15] = 0; + * physnode_map[16-31] = 1; + * physnode_map[32- ] = -1; */ s8 physnode_map[MAX_ELEMENTS] __read_mostly = { [0 ... (MAX_ELEMENTS - 1)] = -1}; EXPORT_SYMBOL(physnode_map); @@ -81,9 +82,9 @@ void memory_present(int nid, unsigned long start, unsigned long end) printk(KERN_DEBUG " "); for (pfn = start; pfn < end; pfn += PAGES_PER_ELEMENT) { physnode_map[pfn / PAGES_PER_ELEMENT] = nid; - printk("%ld ", pfn); + printk(KERN_CONT "%ld ", pfn); } - printk("\n"); + printk(KERN_CONT "\n"); } unsigned long node_memmap_size_bytes(int nid, unsigned long start_pfn, @@ -99,7 +100,6 @@ unsigned long node_memmap_size_bytes(int nid, unsigned long start_pfn, #endif extern unsigned long find_max_low_pfn(void); -extern void add_one_highpage_init(struct page *, int, int); extern unsigned long highend_pfn, highstart_pfn; #define LARGE_PAGE_BYTES (PTRS_PER_PTE * PAGE_SIZE) @@ -119,11 +119,11 @@ int __init get_memcfg_numa_flat(void) { printk("NUMA - single node, flat memory mode\n"); - /* Run the memory configuration and find the top of memory. */ - propagate_e820_map(); node_start_pfn[0] = 0; node_end_pfn[0] = max_pfn; + e820_register_active_regions(0, 0, max_pfn); memory_present(0, 0, max_pfn); + node_remap_size[0] = node_memmap_size_bytes(0, 0, max_pfn); /* Indicate there is one node available. */ nodes_clear(node_online_map); @@ -159,9 +159,17 @@ static void __init allocate_pgdat(int nid) if (nid && node_has_online_mem(nid) && node_remap_start_vaddr[nid]) NODE_DATA(nid) = (pg_data_t *)node_remap_start_vaddr[nid]; else { - NODE_DATA(nid) = (pg_data_t *)(pfn_to_kaddr(min_low_pfn)); - min_low_pfn += PFN_UP(sizeof(pg_data_t)); + unsigned long pgdat_phys; + pgdat_phys = find_e820_area(min_low_pfn<<PAGE_SHIFT, + (nid ? max_low_pfn:max_pfn_mapped)<<PAGE_SHIFT, + sizeof(pg_data_t), + PAGE_SIZE); + NODE_DATA(nid) = (pg_data_t *)(pfn_to_kaddr(pgdat_phys>>PAGE_SHIFT)); + reserve_early(pgdat_phys, pgdat_phys + sizeof(pg_data_t), + "NODE_DATA"); } + printk(KERN_DEBUG "allocate_pgdat: node %d NODE_DATA %08lx\n", + nid, (unsigned long)NODE_DATA(nid)); } /* @@ -199,8 +207,12 @@ void __init remap_numa_kva(void) int node; for_each_online_node(node) { + printk(KERN_DEBUG "remap_numa_kva: node %d\n", node); for (pfn=0; pfn < node_remap_size[node]; pfn += PTRS_PER_PTE) { vaddr = node_remap_start_vaddr[node]+(pfn<<PAGE_SHIFT); + printk(KERN_DEBUG "remap_numa_kva: %08lx to pfn %08lx\n", + (unsigned long)vaddr, + node_remap_start_pfn[node] + pfn); set_pmd_pfn((ulong) vaddr, node_remap_start_pfn[node] + pfn, PAGE_KERNEL_LARGE); @@ -212,17 +224,21 @@ static unsigned long calculate_numa_remap_pages(void) { int nid; unsigned long size, reserve_pages = 0; - unsigned long pfn; for_each_online_node(nid) { - unsigned old_end_pfn = node_end_pfn[nid]; + u64 node_kva_target; + u64 node_kva_final; /* * The acpi/srat node info can show hot-add memroy zones * where memory could be added but not currently present. */ + printk("node %d pfn: [%lx - %lx]\n", + nid, node_start_pfn[nid], node_end_pfn[nid]); if (node_start_pfn[nid] > max_pfn) continue; + if (!node_end_pfn[nid]) + continue; if (node_end_pfn[nid] > max_pfn) node_end_pfn[nid] = max_pfn; @@ -234,39 +250,45 @@ static unsigned long calculate_numa_remap_pages(void) /* now the roundup is correct, convert to PAGE_SIZE pages */ size = size * PTRS_PER_PTE; - /* - * Validate the region we are allocating only contains valid - * pages. - */ - for (pfn = node_end_pfn[nid] - size; - pfn < node_end_pfn[nid]; pfn++) - if (!page_is_ram(pfn)) - break; + node_kva_target = round_down(node_end_pfn[nid] - size, + PTRS_PER_PTE); + node_kva_target <<= PAGE_SHIFT; + do { + node_kva_final = find_e820_area(node_kva_target, + ((u64)node_end_pfn[nid])<<PAGE_SHIFT, + ((u64)size)<<PAGE_SHIFT, + LARGE_PAGE_BYTES); + node_kva_target -= LARGE_PAGE_BYTES; + } while (node_kva_final == -1ULL && + (node_kva_target>>PAGE_SHIFT) > (node_start_pfn[nid])); + + if (node_kva_final == -1ULL) + panic("Can not get kva ram\n"); - if (pfn != node_end_pfn[nid]) - size = 0; - - printk("Reserving %ld pages of KVA for lmem_map of node %d\n", - size, nid); node_remap_size[nid] = size; node_remap_offset[nid] = reserve_pages; reserve_pages += size; - printk("Shrinking node %d from %ld pages to %ld pages\n", - nid, node_end_pfn[nid], node_end_pfn[nid] - size); - - if (node_end_pfn[nid] & (PTRS_PER_PTE-1)) { - /* - * Align node_end_pfn[] and node_remap_start_pfn[] to - * pmd boundary. remap_numa_kva will barf otherwise. - */ - printk("Shrinking node %d further by %ld pages for proper alignment\n", - nid, node_end_pfn[nid] & (PTRS_PER_PTE-1)); - size += node_end_pfn[nid] & (PTRS_PER_PTE-1); - } + printk("Reserving %ld pages of KVA for lmem_map of node %d at %llx\n", + size, nid, node_kva_final>>PAGE_SHIFT); + + /* + * prevent kva address below max_low_pfn want it on system + * with less memory later. + * layout will be: KVA address , KVA RAM + * + * we are supposed to only record the one less then max_low_pfn + * but we could have some hole in high memory, and it will only + * check page_is_ram(pfn) && !page_is_reserved_early(pfn) to decide + * to use it as free. + * So reserve_early here, hope we don't run out of that array + */ + reserve_early(node_kva_final, + node_kva_final+(((u64)size)<<PAGE_SHIFT), + "KVA RAM"); - node_end_pfn[nid] -= size; - node_remap_start_pfn[nid] = node_end_pfn[nid]; - shrink_active_range(nid, old_end_pfn, node_end_pfn[nid]); + node_remap_start_pfn[nid] = node_kva_final>>PAGE_SHIFT; + remove_active_range(nid, node_remap_start_pfn[nid], + node_remap_start_pfn[nid] + size); } printk("Reserving total of %ld pages for numa KVA remap\n", reserve_pages); @@ -284,8 +306,7 @@ static void init_remap_allocator(int nid) printk ("node %d will remap to vaddr %08lx - %08lx\n", nid, (ulong) node_remap_start_vaddr[nid], - (ulong) pfn_to_kaddr(highstart_pfn - + node_remap_offset[nid] + node_remap_size[nid])); + (ulong) node_remap_end_vaddr[nid]); } extern void setup_bootmem_allocator(void); @@ -293,7 +314,7 @@ unsigned long __init setup_memory(void) { int nid; unsigned long system_start_pfn, system_max_low_pfn; - unsigned long wasted_pages; + long kva_target_pfn; /* * When mapping a NUMA machine we allocate the node_mem_map arrays @@ -302,34 +323,38 @@ unsigned long __init setup_memory(void) * this space and use it to adjust the boundary between ZONE_NORMAL * and ZONE_HIGHMEM. */ + + /* call find_max_low_pfn at first, it could update max_pfn */ + system_max_low_pfn = max_low_pfn = find_max_low_pfn(); + + remove_all_active_ranges(); get_memcfg_numa(); - kva_pages = calculate_numa_remap_pages(); + kva_pages = round_up(calculate_numa_remap_pages(), PTRS_PER_PTE); /* partially used pages are not usable - thus round upwards */ system_start_pfn = min_low_pfn = PFN_UP(init_pg_tables_end); - kva_start_pfn = find_max_low_pfn() - kva_pages; - -#ifdef CONFIG_BLK_DEV_INITRD - /* Numa kva area is below the initrd */ - if (initrd_start) - kva_start_pfn = PFN_DOWN(initrd_start - PAGE_OFFSET) - - kva_pages; -#endif + kva_target_pfn = round_down(max_low_pfn - kva_pages, PTRS_PER_PTE); + do { + kva_start_pfn = find_e820_area(kva_target_pfn<<PAGE_SHIFT, + max_low_pfn<<PAGE_SHIFT, + kva_pages<<PAGE_SHIFT, + PTRS_PER_PTE<<PAGE_SHIFT) >> PAGE_SHIFT; + kva_target_pfn -= PTRS_PER_PTE; + } while (kva_start_pfn == -1UL && kva_target_pfn > min_low_pfn); - /* - * We waste pages past at the end of the KVA for no good reason other - * than how it is located. This is bad. - */ - wasted_pages = kva_start_pfn & (PTRS_PER_PTE-1); - kva_start_pfn -= wasted_pages; - kva_pages += wasted_pages; + if (kva_start_pfn == -1UL) + panic("Can not get kva space\n"); - system_max_low_pfn = max_low_pfn = find_max_low_pfn(); printk("kva_start_pfn ~ %ld find_max_low_pfn() ~ %ld\n", kva_start_pfn, max_low_pfn); printk("max_pfn = %ld\n", max_pfn); + + /* avoid clash with initrd */ + reserve_early(kva_start_pfn<<PAGE_SHIFT, + (kva_start_pfn + kva_pages)<<PAGE_SHIFT, + "KVA PG"); #ifdef CONFIG_HIGHMEM highstart_pfn = highend_pfn = max_pfn; if (max_pfn > system_max_low_pfn) @@ -365,16 +390,8 @@ unsigned long __init setup_memory(void) return max_low_pfn; } -void __init numa_kva_reserve(void) -{ - if (kva_pages) - reserve_bootmem(PFN_PHYS(kva_start_pfn), PFN_PHYS(kva_pages), - BOOTMEM_DEFAULT); -} - void __init zone_sizes_init(void) { - int nid; unsigned long max_zone_pfns[MAX_NR_ZONES]; memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); max_zone_pfns[ZONE_DMA] = @@ -384,27 +401,18 @@ void __init zone_sizes_init(void) max_zone_pfns[ZONE_HIGHMEM] = highend_pfn; #endif - /* If SRAT has not registered memory, register it now */ - if (find_max_pfn_with_active_regions() == 0) { - for_each_online_node(nid) { - if (node_has_online_mem(nid)) - add_active_range(nid, node_start_pfn[nid], - node_end_pfn[nid]); - } - } - free_area_init_nodes(max_zone_pfns); return; } -void __init set_highmem_pages_init(int bad_ppro) +void __init set_highmem_pages_init(void) { #ifdef CONFIG_HIGHMEM struct zone *zone; - struct page *page; + int nid; for_each_zone(zone) { - unsigned long node_pfn, zone_start_pfn, zone_end_pfn; + unsigned long zone_start_pfn, zone_end_pfn; if (!is_highmem(zone)) continue; @@ -412,16 +420,12 @@ void __init set_highmem_pages_init(int bad_ppro) zone_start_pfn = zone->zone_start_pfn; zone_end_pfn = zone_start_pfn + zone->spanned_pages; + nid = zone_to_nid(zone); printk("Initializing %s for node %d (%08lx:%08lx)\n", - zone->name, zone_to_nid(zone), - zone_start_pfn, zone_end_pfn); - - for (node_pfn = zone_start_pfn; node_pfn < zone_end_pfn; node_pfn++) { - if (!pfn_valid(node_pfn)) - continue; - page = pfn_to_page(node_pfn); - add_one_highpage_init(page, node_pfn, bad_ppro); - } + zone->name, nid, zone_start_pfn, zone_end_pfn); + + add_highpages_with_active_regions(nid, zone_start_pfn, + zone_end_pfn); } totalram_pages += totalhigh_pages; #endif diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index d71be0e..65d55056 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -225,13 +225,6 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base) update_page_count(PG_LEVEL_4K, pages_4k); } -static inline int page_kills_ppro(unsigned long pagenr) -{ - if (pagenr >= 0x70000 && pagenr <= 0x7003F) - return 1; - return 0; -} - /* * devmem_is_allowed() checks to see if /dev/mem access to a certain address * is valid. The argument is a physical page number. @@ -292,29 +285,60 @@ static void __init permanent_kmaps_init(pgd_t *pgd_base) pkmap_page_table = pte; } -void __init add_one_highpage_init(struct page *page, int pfn, int bad_ppro) +static void __init add_one_highpage_init(struct page *page, int pfn) { - if (page_is_ram(pfn) && !(bad_ppro && page_kills_ppro(pfn))) { - ClearPageReserved(page); - init_page_count(page); - __free_page(page); - totalhigh_pages++; - } else - SetPageReserved(page); + ClearPageReserved(page); + init_page_count(page); + __free_page(page); + totalhigh_pages++; } -#ifndef CONFIG_NUMA -static void __init set_highmem_pages_init(int bad_ppro) +struct add_highpages_data { + unsigned long start_pfn; + unsigned long end_pfn; +}; + +static void __init add_highpages_work_fn(unsigned long start_pfn, + unsigned long end_pfn, void *datax) { - int pfn; + int node_pfn; + struct page *page; + unsigned long final_start_pfn, final_end_pfn; + struct add_highpages_data *data; - for (pfn = highstart_pfn; pfn < highend_pfn; pfn++) { - /* - * Holes under sparsemem might not have no mem_map[]: - */ - if (pfn_valid(pfn)) - add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro); + data = (struct add_highpages_data *)datax; + + final_start_pfn = max(start_pfn, data->start_pfn); + final_end_pfn = min(end_pfn, data->end_pfn); + if (final_start_pfn >= final_end_pfn) + return; + + for (node_pfn = final_start_pfn; node_pfn < final_end_pfn; + node_pfn++) { + if (!pfn_valid(node_pfn)) + continue; + page = pfn_to_page(node_pfn); + add_one_highpage_init(page, node_pfn); } + +} + +void __init add_highpages_with_active_regions(int nid, unsigned long start_pfn, + unsigned long end_pfn) +{ + struct add_highpages_data data; + + data.start_pfn = start_pfn; + data.end_pfn = end_pfn; + + work_with_active_regions(nid, add_highpages_work_fn, &data); +} + +#ifndef CONFIG_NUMA +static void __init set_highmem_pages_init(void) +{ + add_highpages_with_active_regions(0, highstart_pfn, highend_pfn); + totalram_pages += totalhigh_pages; } #endif /* !CONFIG_NUMA */ @@ -322,7 +346,7 @@ static void __init set_highmem_pages_init(int bad_ppro) #else # define kmap_init() do { } while (0) # define permanent_kmaps_init(pgd_base) do { } while (0) -# define set_highmem_pages_init(bad_ppro) do { } while (0) +# define set_highmem_pages_init() do { } while (0) #endif /* CONFIG_HIGHMEM */ pteval_t __PAGE_KERNEL = _PAGE_KERNEL; @@ -569,13 +593,11 @@ static struct kcore_list kcore_mem, kcore_vmalloc; void __init mem_init(void) { int codesize, reservedpages, datasize, initsize; - int tmp, bad_ppro; + int tmp; #ifdef CONFIG_FLATMEM BUG_ON(!mem_map); #endif - bad_ppro = ppro_with_ram_bug(); - /* this will put all low memory onto the freelists */ totalram_pages += free_all_bootmem(); @@ -587,7 +609,7 @@ void __init mem_init(void) if (page_is_ram(tmp) && PageReserved(pfn_to_page(tmp))) reservedpages++; - set_highmem_pages_init(bad_ppro); + set_highmem_pages_init(); codesize = (unsigned long) &_etext - (unsigned long) &_text; datasize = (unsigned long) &_edata - (unsigned long) &_etext; @@ -776,3 +798,9 @@ void free_initrd_mem(unsigned long start, unsigned long end) free_init_pages("initrd memory", start, end); } #endif + +int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, + int flags) +{ + return reserve_bootmem(phys, len, flags); +} diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 48623ae..18c6a00 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -48,6 +48,18 @@ #include <asm/numa.h> #include <asm/cacheflush.h> +/* + * PFN of last memory page. + */ +unsigned long end_pfn; + +/* + * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. + * The direct mapping extends to max_pfn_mapped, so that we can directly access + * apertures, ACPI and other tables without having to play with fixmaps. + */ +unsigned long max_pfn_mapped; + static unsigned long dma_reserve __initdata; DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); @@ -808,12 +820,14 @@ void free_initrd_mem(unsigned long start, unsigned long end) } #endif -void __init reserve_bootmem_generic(unsigned long phys, unsigned len) +int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, + int flags) { #ifdef CONFIG_NUMA int nid, next_nid; #endif unsigned long pfn = phys >> PAGE_SHIFT; + int ret; if (pfn >= end_pfn) { /* @@ -821,11 +835,11 @@ void __init reserve_bootmem_generic(unsigned long phys, unsigned len) * firmware tables: */ if (pfn < max_pfn_mapped) - return; + return -EFAULT; printk(KERN_ERR "reserve_bootmem: illegal reserve %lx %u\n", phys, len); - return; + return -EFAULT; } /* Should check here against the e820 map to avoid double free */ @@ -833,9 +847,13 @@ void __init reserve_bootmem_generic(unsigned long phys, unsigned len) nid = phys_to_nid(phys); next_nid = phys_to_nid(phys + len - 1); if (nid == next_nid) - reserve_bootmem_node(NODE_DATA(nid), phys, len, BOOTMEM_DEFAULT); + ret = reserve_bootmem_node(NODE_DATA(nid), phys, len, flags); else - reserve_bootmem(phys, len, BOOTMEM_DEFAULT); + ret = reserve_bootmem(phys, len, flags); + + if (ret != 0) + return ret; + #else reserve_bootmem(phys, len, BOOTMEM_DEFAULT); #endif @@ -844,6 +862,8 @@ void __init reserve_bootmem_generic(unsigned long phys, unsigned len) dma_reserve += len / PAGE_SIZE; set_dma_reserve(dma_reserve); } + + return 0; } int kern_addr_valid(unsigned long addr) diff --git a/arch/x86/mm/k8topology_64.c b/arch/x86/mm/k8topology_64.c index 0ea66b5..317573e 100644 --- a/arch/x86/mm/k8topology_64.c +++ b/arch/x86/mm/k8topology_64.c @@ -57,18 +57,22 @@ static __init void early_get_boot_cpu_id(void) /* * Find possible boot-time SMP configuration: */ +#ifdef CONFIG_X86_MPPARSE early_find_smp_config(); +#endif #ifdef CONFIG_ACPI /* * Read APIC information from ACPI tables. */ early_acpi_boot_init(); #endif +#ifdef CONFIG_X86_MPPARSE /* * get boot-time SMP configuration: */ if (smp_found_config) early_get_smp_config(); +#endif early_init_lapic_mapping(); } diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index c5066d5..afb07ff 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -233,7 +233,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, else bootmap_start = round_up(start, PAGE_SIZE); /* - * SMP_CAHCE_BYTES could be enough, but init_bootmem_node like + * SMP_CACHE_BYTES could be enough, but init_bootmem_node like * to use that to align to PAGE_SIZE */ bootmap = early_node_mem(nodeid, bootmap_start, end, |