The patch titled bootmem: factor out the marking of a PFN range has been added to the -mm tree. Its filename is bootmem-factor-out-the-marking-of-a-pfn-range.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find out what to do about this The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: bootmem: factor out the marking of a PFN range From: Johannes Weiner <hannes@xxxxxxxxxxxx> Introduce new helpers that mark a range that resides completely on a node or node-agnostic ranges that might also span node boundaries. The free/reserve API functions will then directly use these helpers. Note that the free/reserve semantics become more strict: while the prior code took basically arbitrary range arguments and marked the PFNs that happen to fall into that range, the new code requires node-specific ranges to be completely on the node. The node-agnostic requests might span node boundaries as long as the nodes are contiguous. Passing ranges that do not satisfy these criteria is a bug. Signed-off-by: Johannes Weiner <hannes@xxxxxxxxxxxx> Cc: Ingo Molnar <mingo@xxxxxxx> Cc: Yinghai Lu <yhlu.kernel@xxxxxxxxx> Cc: Andi Kleen <andi@xxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- include/linux/bootmem.h | 2 mm/bootmem.c | 189 ++++++++++++++------------------------ 2 files changed, 71 insertions(+), 120 deletions(-) diff -puN include/linux/bootmem.h~bootmem-factor-out-the-marking-of-a-pfn-range include/linux/bootmem.h --- a/include/linux/bootmem.h~bootmem-factor-out-the-marking-of-a-pfn-range +++ a/include/linux/bootmem.h @@ -65,7 +65,7 @@ extern void free_bootmem(unsigned long a #define BOOTMEM_DEFAULT 0 #define BOOTMEM_EXCLUSIVE (1<<0) -extern void reserve_bootmem_node(pg_data_t *pgdat, +extern int reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size, int flags); diff -puN mm/bootmem.c~bootmem-factor-out-the-marking-of-a-pfn-range mm/bootmem.c --- a/mm/bootmem.c~bootmem-factor-out-the-marking-of-a-pfn-range +++ a/mm/bootmem.c @@ -234,6 +234,9 @@ static void __init __free(bootmem_data_t sidx + PFN_DOWN(bdata->node_boot_start), eidx + PFN_DOWN(bdata->node_boot_start)); + if (bdata->hint_idx > sidx) + bdata->hint_idx = sidx; + for (idx = sidx; idx < eidx; idx++) if (!test_and_clear_bit(idx, bdata->node_bootmem_map)) BUG(); @@ -263,40 +266,57 @@ static int __init __reserve(bootmem_data return 0; } -static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, - unsigned long size) +static int __init mark_bootmem_node(bootmem_data_t *bdata, + unsigned long start, unsigned long end, + int reserve, int flags) { unsigned long sidx, eidx; - unsigned long i; - BUG_ON(!size); + bdebug("nid=%d start=%lx end=%lx reserve=%d flags=%x\n", + bdata - bootmem_node_data, start, end, reserve, flags); - /* out range */ - if (addr + size < bdata->node_boot_start || - PFN_DOWN(addr) > bdata->node_low_pfn) - return; - /* - * round down end of usable mem, partially free pages are - * considered reserved. - */ + BUG_ON(start < PFN_DOWN(bdata->node_boot_start)); + BUG_ON(end > bdata->node_low_pfn); - if (addr >= bdata->node_boot_start && - PFN_DOWN(addr - bdata->node_boot_start) < bdata->hint_idx) - bdata->hint_idx = PFN_DOWN(addr - bdata->node_boot_start); + sidx = start - PFN_DOWN(bdata->node_boot_start); + eidx = end - PFN_DOWN(bdata->node_boot_start); - /* - * Round up to index to the range. - */ - if (PFN_UP(addr) > PFN_DOWN(bdata->node_boot_start)) - sidx = PFN_UP(addr) - PFN_DOWN(bdata->node_boot_start); + if (reserve) + return __reserve(bdata, sidx, eidx, flags); else - sidx = 0; + __free(bdata, sidx, eidx); + return 0; +} + +static int __init mark_bootmem(unsigned long start, unsigned long end, + int reserve, int flags) +{ + unsigned long pos; + bootmem_data_t *bdata; + + pos = start; + list_for_each_entry(bdata, &bdata_list, list) { + int err; + unsigned long max; - eidx = PFN_DOWN(addr + size - bdata->node_boot_start); - if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) - eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); + if (pos < PFN_DOWN(bdata->node_boot_start)) { + BUG_ON(pos != start); + continue; + } + + max = min(bdata->node_low_pfn, end); + + err = mark_bootmem_node(bdata, pos, max, reserve, flags); + if (reserve && err) { + mark_bootmem(start, pos, 0, 0); + return err; + } - __free(bdata, sidx, eidx); + if (max == end) + return 0; + pos = bdata->node_low_pfn; + } + BUG(); } /** @@ -307,12 +327,17 @@ static void __init free_bootmem_core(boo * * Partial pages will be considered reserved and left as they are. * - * Only physical pages that actually reside on @pgdat are marked. + * The range must reside completely on the specified node. */ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size) { - free_bootmem_core(pgdat->bdata, physaddr, size); + unsigned long start, end; + + start = PFN_UP(physaddr); + end = PFN_DOWN(physaddr + size); + + mark_bootmem_node(pgdat->bdata, start, end, 0, 0); } /** @@ -322,83 +347,16 @@ void __init free_bootmem_node(pg_data_t * * Partial pages will be considered reserved and left as they are. * - * All physical pages within the range are marked, no matter what - * node they reside on. + * The range must be contiguous but may span node boundaries. */ void __init free_bootmem(unsigned long addr, unsigned long size) { - bootmem_data_t *bdata; - list_for_each_entry(bdata, &bdata_list, list) - free_bootmem_core(bdata, addr, size); -} - -/* - * Marks a particular physical memory range as unallocatable. Usable RAM - * might be used for boot-time allocations - or it might get added - * to the free page pool later on. - */ -static int __init can_reserve_bootmem_core(bootmem_data_t *bdata, - unsigned long addr, unsigned long size, int flags) -{ - unsigned long sidx, eidx; - unsigned long i; - - BUG_ON(!size); - - /* out of range, don't hold other */ - if (addr + size < bdata->node_boot_start || - PFN_DOWN(addr) > bdata->node_low_pfn) - return 0; - - /* - * Round up to index to the range. - */ - if (addr > bdata->node_boot_start) - sidx= PFN_DOWN(addr - bdata->node_boot_start); - else - sidx = 0; - - eidx = PFN_UP(addr + size - bdata->node_boot_start); - if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) - eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); - - for (i = sidx; i < eidx; i++) { - if (test_bit(i, bdata->node_bootmem_map)) { - if (flags & BOOTMEM_EXCLUSIVE) - return -EBUSY; - } - } - - return 0; - -} - -static void __init reserve_bootmem_core(bootmem_data_t *bdata, - unsigned long addr, unsigned long size, int flags) -{ - unsigned long sidx, eidx; - unsigned long i; - - BUG_ON(!size); + unsigned long start, end; - /* out of range */ - if (addr + size < bdata->node_boot_start || - PFN_DOWN(addr) > bdata->node_low_pfn) - return; + start = PFN_UP(addr); + end = PFN_DOWN(addr + size); - /* - * Round up to index to the range. - */ - if (addr > bdata->node_boot_start) - sidx= PFN_DOWN(addr - bdata->node_boot_start); - else - sidx = 0; - - eidx = PFN_UP(addr + size - bdata->node_boot_start); - if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) - eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); - - return __reserve(bdata, sidx, eidx, flags); + mark_bootmem(start, end, 0, 0); } /** @@ -410,17 +368,17 @@ static void __init reserve_bootmem_core( * * Partial pages will be reserved. * - * Only physical pages that actually reside on @pgdat are marked. + * The range must reside completely on the specified node. */ -void __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, +int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size, int flags) { - int ret; + unsigned long start, end; - ret = can_reserve_bootmem_core(pgdat->bdata, physaddr, size, flags); - if (ret < 0) - return; - reserve_bootmem_core(pgdat->bdata, physaddr, size, flags); + start = PFN_DOWN(physaddr); + end = PFN_UP(physaddr + size); + + return mark_bootmem_node(pgdat->bdata, start, end, 1, flags); } #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE @@ -432,24 +390,17 @@ void __init reserve_bootmem_node(pg_data * * Partial pages will be reserved. * - * All physical pages within the range are marked, no matter what - * node they reside on. + * The range must be contiguous but may span node boundaries. */ int __init reserve_bootmem(unsigned long addr, unsigned long size, int flags) { - bootmem_data_t *bdata; - int ret; + unsigned long start, end; - list_for_each_entry(bdata, &bdata_list, list) { - ret = can_reserve_bootmem_core(bdata, addr, size, flags); - if (ret < 0) - return ret; - } - list_for_each_entry(bdata, &bdata_list, list) - reserve_bootmem_core(bdata, addr, size, flags); + start = PFN_DOWN(addr); + end = PFN_UP(addr + size); - return 0; + return mark_bootmem(start, end, 1, flags); } #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */ @@ -662,7 +613,7 @@ void * __init alloc_bootmem_section(unsi if (start_nr != section_nr || end_nr != section_nr) { printk(KERN_WARNING "alloc_bootmem failed on section %ld.\n", section_nr); - free_bootmem_core(pgdat->bdata, __pa(ptr), size); + free_bootmem_node(pgdat, __pa(ptr), size); ptr = NULL; } _ Patches currently in -mm which might be from hannes@xxxxxxxxxxxx are git-xtensa.patch mm-move-bootmem-descriptors-definition-to-a-single-place.patch mm-fix-free_all_bootmem_core-alignment-check.patch mm-normalize-internal-argument-passing-of-bootmem-data.patch mm-unexport-__alloc_bootmem_core.patch mm-drop-unneeded-pgdat-argument-from-free_area_init_node.patch hugetlb-guarantee-that-cow-faults-for-a-process-that-called-mmapmap_private-on-hugetlbfs-will-succeed-fix.patch huge-page-private-reservation-review-cleanups.patch huge-page-private-reservation-review-cleanups-fix.patch mm-record-map_noreserve-status-on-vmas-and-fix-small-page-mprotect-reservations.patch hugetlb-move-reservation-region-support-earlier.patch hugetlb-allow-huge-page-mappings-to-be-created-without-reservations.patch hugetlb-allow-huge-page-mappings-to-be-created-without-reservations-cleanups.patch bootmem-reorder-code-to-match-new-bootmem-structure.patch bootmem-clean-up-bootmemc-file-header.patch bootmem-add-documentation-to-api-functions.patch bootmem-add-debugging-framework.patch bootmem-revisit-bitmap-size-calculations.patch bootmem-revisit-bootmem-descriptor-list-handling.patch bootmem-clean-up-free_all_bootmem_core.patch bootmem-clean-up-alloc_bootmem_core.patch bootmem-free-reserve-helpers.patch bootmem-factor-out-the-marking-of-a-pfn-range.patch bootmem-respect-goal-more-likely.patch bootmem-make-__alloc_bootmem_low_node-fall-back-to-other-nodes.patch bootmem-revisit-alloc_bootmem_section.patch bootmem-replace-node_boot_start-in-struct-bootmem_data.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html