The following patch still has not been tested but I'm reposting it anyway. I've sent a rebased patch on top of next-20121004 that includes this patch privately to Thierry for testing.
CMA requires that the PG_migrate_skip hint be skipped but it was only skipping it when isolating pages for migration, not for free. Ensure cc->isolate_skip_hint gets passed in both cases. This is a fix for mm-compaction-cache-if-a-pageblock-was-scanned-and-no-pages-were-isolated-fix.patch but is based on top of linux-next/akpm as of 20121004. It will cause minor conflicts when it is slotted into place but the resolutions should be straight-forward. Signed-off-by: Mel Gorman <[email protected]> --- mm/compaction.c | 16 ++++------------ mm/internal.h | 3 ++- mm/page_alloc.c | 43 ++++++++++++++++++++++--------------------- 3 files changed, 28 insertions(+), 34 deletions(-) diff --git a/mm/compaction.c b/mm/compaction.c index ed3b8f1..2c4ce17 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -373,22 +373,14 @@ static unsigned long isolate_freepages_block(struct compact_control *cc, * a free page). */ unsigned long -isolate_freepages_range(unsigned long start_pfn, unsigned long end_pfn) +isolate_freepages_range(struct compact_control *cc, + unsigned long start_pfn, unsigned long end_pfn) { unsigned long isolated, pfn, block_end_pfn; - struct zone *zone = NULL; LIST_HEAD(freelist); - /* cc needed for isolate_freepages_block to acquire zone->lock */ - struct compact_control cc = { - .sync = true, - }; - - if (pfn_valid(start_pfn)) - cc.zone = zone = page_zone(pfn_to_page(start_pfn)); - for (pfn = start_pfn; pfn < end_pfn; pfn += isolated) { - if (!pfn_valid(pfn) || zone != page_zone(pfn_to_page(pfn))) + if (!pfn_valid(pfn) || cc->zone != page_zone(pfn_to_page(pfn))) break; /* @@ -398,7 +390,7 @@ isolate_freepages_range(unsigned long start_pfn, unsigned long end_pfn) block_end_pfn = ALIGN(pfn + 1, pageblock_nr_pages); block_end_pfn = min(block_end_pfn, end_pfn); - isolated = isolate_freepages_block(&cc, pfn, block_end_pfn, + isolated = isolate_freepages_block(cc, pfn, block_end_pfn, &freelist, true); /* diff --git a/mm/internal.h b/mm/internal.h index 9d5d276..a3ce781 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -135,7 +135,8 @@ struct compact_control { }; unsigned long -isolate_freepages_range(unsigned long start_pfn, unsigned long end_pfn); +isolate_freepages_range(struct compact_control *cc, + unsigned long start_pfn, unsigned long end_pfn); unsigned long isolate_migratepages_range(struct zone *zone, struct compact_control *cc, unsigned long low_pfn, unsigned long end_pfn, bool unevictable); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8e1be1c..d66efcb 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5669,7 +5669,8 @@ static unsigned long pfn_max_align_up(unsigned long pfn) } /* [start, end) must belong to a single zone. */ -static int __alloc_contig_migrate_range(unsigned long start, unsigned long end) +static int __alloc_contig_migrate_range(struct compact_control *cc, + unsigned long start, unsigned long end) { /* This function is based on compact_zone() from compaction.c. */ unsigned long nr_reclaimed; @@ -5677,26 +5678,17 @@ static int __alloc_contig_migrate_range(unsigned long start, unsigned long end) unsigned int tries = 0; int ret = 0; - struct compact_control cc = { - .nr_migratepages = 0, - .order = -1, - .zone = page_zone(pfn_to_page(start)), - .sync = true, - .ignore_skip_hint = true, - }; - INIT_LIST_HEAD(&cc.migratepages); - migrate_prep_local(); - while (pfn < end || !list_empty(&cc.migratepages)) { + while (pfn < end || !list_empty(&cc->migratepages)) { if (fatal_signal_pending(current)) { ret = -EINTR; break; } - if (list_empty(&cc.migratepages)) { - cc.nr_migratepages = 0; - pfn = isolate_migratepages_range(cc.zone, &cc, + if (list_empty(&cc->migratepages)) { + cc->nr_migratepages = 0; + pfn = isolate_migratepages_range(cc->zone, cc, pfn, end, true); if (!pfn) { ret = -EINTR; @@ -5708,16 +5700,16 @@ static int __alloc_contig_migrate_range(unsigned long start, unsigned long end) break; } - nr_reclaimed = reclaim_clean_pages_from_list(cc.zone, - &cc.migratepages); - cc.nr_migratepages -= nr_reclaimed; + nr_reclaimed = reclaim_clean_pages_from_list(cc->zone, + &cc->migratepages); + cc->nr_migratepages -= nr_reclaimed; - ret = migrate_pages(&cc.migratepages, + ret = migrate_pages(&cc->migratepages, alloc_migrate_target, 0, false, MIGRATE_SYNC); } - putback_lru_pages(&cc.migratepages); + putback_lru_pages(&cc->migratepages); return ret > 0 ? 0 : ret; } @@ -5796,6 +5788,15 @@ int alloc_contig_range(unsigned long start, unsigned long end, unsigned long outer_start, outer_end; int ret = 0, order; + struct compact_control cc = { + .nr_migratepages = 0, + .order = -1, + .zone = page_zone(pfn_to_page(start)), + .sync = true, + .ignore_skip_hint = true, + }; + INIT_LIST_HEAD(&cc.migratepages); + /* * What we do here is we mark all pageblocks in range as * MIGRATE_ISOLATE. Because pageblock and max order pages may @@ -5825,7 +5826,7 @@ int alloc_contig_range(unsigned long start, unsigned long end, if (ret) goto done; - ret = __alloc_contig_migrate_range(start, end); + ret = __alloc_contig_migrate_range(&cc, start, end); if (ret) goto done; @@ -5874,7 +5875,7 @@ int alloc_contig_range(unsigned long start, unsigned long end, __reclaim_pages(zone, GFP_HIGHUSER_MOVABLE, end-start); /* Grab isolated pages from freelists. */ - outer_end = isolate_freepages_range(outer_start, end); + outer_end = isolate_freepages_range(&cc, outer_start, end); if (!outer_end) { ret = -EBUSY; goto done; -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
