The patch titled Subject: mm, compaction: pass classzone_idx and alloc_flags to watermark checking has been removed from the -mm tree. Its filename was mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking.patch This patch was dropped because it was merged into mainline or a subsystem tree ------------------------------------------------------ From: Vlastimil Babka <vbabka@xxxxxxx> Subject: mm, compaction: pass classzone_idx and alloc_flags to watermark checking Compaction relies on zone watermark checks for decisions such as if it's worth to start compacting in compaction_suitable() or whether compaction should stop in compact_finished(). The watermark checks take classzone_idx and alloc_flags parameters, which are related to the memory allocation request. But from the context of compaction they are currently passed as 0, including the direct compaction which is invoked to satisfy the allocation request, and could therefore know the proper values. The lack of proper values can lead to mismatch between decisions taken during compaction and decisions related to the allocation request. Lack of proper classzone_idx value means that lowmem_reserve is not taken into account. This has manifested (during recent changes to deferred compaction) when DMA zone was used as fallback for preferred Normal zone. compaction_suitable() without proper classzone_idx would think that the watermarks are already satisfied, but watermark check in get_page_from_freelist() would fail. Because of this problem, deferring compaction has extra complexity that can be removed in the following patch. The issue (not confirmed in practice) with missing alloc_flags is opposite in nature. For allocations that include ALLOC_HIGH, ALLOC_HIGHER or ALLOC_CMA in alloc_flags (the last includes all MOVABLE allocations on CMA-enabled systems) the watermark checking in compaction with 0 passed will be stricter than in get_page_from_freelist(). In these cases compaction might be running for a longer time than is really needed. Another issue compaction_suitable() is that the check for "does the zone need compaction at all?" comes only after the check "does the zone have enough free free pages to succeed compaction". The latter considers extra pages for migration and can therefore in some situations fail and return COMPACT_SKIPPED, although the high-order allocation would succeed and we should return COMPACT_PARTIAL. This patch fixes these problems by adding alloc_flags and classzone_idx to struct compact_control and related functions involved in direct compaction and watermark checking. Where possible, all other callers of compaction_suitable() pass proper values where those are known. This is currently limited to classzone_idx, which is sometimes known in kswapd context. However, the direct reclaim callers should_continue_reclaim() and compaction_ready() do not currently know the proper values, so the coordination between reclaim and compaction may still not be as accurate as it could. This can be fixed later, if it's shown to be an issue. Additionaly the checks in compact_suitable() are reordered to address the second issue described above. The effect of this patch should be slightly better high-order allocation success rates and/or less compaction overhead, depending on the type of allocations and presence of CMA. It allows simplifying deferred compaction code in a followup patch. When testing with stress-highalloc, there was some slight improvement (which might be just due to variance) in success rates of non-THP-like allocations. Signed-off-by: Vlastimil Babka <vbabka@xxxxxxx> Cc: Minchan Kim <minchan@xxxxxxxxxx> Cc: Mel Gorman <mgorman@xxxxxxx> Cc: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx> Cc: Michal Nazarewicz <mina86@xxxxxxxxxx> Cc: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx> Cc: Christoph Lameter <cl@xxxxxxxxx> Acked-by: Rik van Riel <riel@xxxxxxxxxx> Cc: David Rientjes <rientjes@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- include/linux/compaction.h | 8 ++++- mm/compaction.c | 48 +++++++++++++++++++---------------- mm/internal.h | 2 + mm/page_alloc.c | 1 mm/vmscan.c | 12 ++++---- 5 files changed, 42 insertions(+), 29 deletions(-) diff -puN include/linux/compaction.h~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking include/linux/compaction.h --- a/include/linux/compaction.h~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking +++ a/include/linux/compaction.h @@ -33,10 +33,12 @@ extern int fragmentation_index(struct zo extern unsigned long try_to_compact_pages(struct zonelist *zonelist, int order, gfp_t gfp_mask, nodemask_t *mask, enum migrate_mode mode, int *contended, + int alloc_flags, int classzone_idx, struct zone **candidate_zone); extern void compact_pgdat(pg_data_t *pgdat, int order); extern void reset_isolation_suitable(pg_data_t *pgdat); -extern unsigned long compaction_suitable(struct zone *zone, int order); +extern unsigned long compaction_suitable(struct zone *zone, int order, + int alloc_flags, int classzone_idx); /* Do not skip compaction more than 64 times */ #define COMPACT_MAX_DEFER_SHIFT 6 @@ -103,6 +105,7 @@ static inline bool compaction_restarting static inline unsigned long try_to_compact_pages(struct zonelist *zonelist, int order, gfp_t gfp_mask, nodemask_t *nodemask, enum migrate_mode mode, int *contended, + int alloc_flags, int classzone_idx, struct zone **candidate_zone) { return COMPACT_CONTINUE; @@ -116,7 +119,8 @@ static inline void reset_isolation_suita { } -static inline unsigned long compaction_suitable(struct zone *zone, int order) +static inline unsigned long compaction_suitable(struct zone *zone, int order, + int alloc_flags, int classzone_idx) { return COMPACT_SKIPPED; } diff -puN mm/compaction.c~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking mm/compaction.c --- a/mm/compaction.c~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking +++ a/mm/compaction.c @@ -1086,9 +1086,9 @@ static int compact_finished(struct zone /* Compaction run is not finished if the watermark is not met */ watermark = low_wmark_pages(zone); - watermark += (1 << cc->order); - if (!zone_watermark_ok(zone, cc->order, watermark, 0, 0)) + if (!zone_watermark_ok(zone, cc->order, watermark, cc->classzone_idx, + cc->alloc_flags)) return COMPACT_CONTINUE; /* Direct compactor: Is a suitable page free? */ @@ -1114,7 +1114,8 @@ static int compact_finished(struct zone * COMPACT_PARTIAL - If the allocation would succeed without compaction * COMPACT_CONTINUE - If compaction should run now */ -unsigned long compaction_suitable(struct zone *zone, int order) +unsigned long compaction_suitable(struct zone *zone, int order, + int alloc_flags, int classzone_idx) { int fragindex; unsigned long watermark; @@ -1126,21 +1127,30 @@ unsigned long compaction_suitable(struct if (order == -1) return COMPACT_CONTINUE; + watermark = low_wmark_pages(zone); + /* + * If watermarks for high-order allocation are already met, there + * should be no need for compaction at all. + */ + if (zone_watermark_ok(zone, order, watermark, classzone_idx, + alloc_flags)) + return COMPACT_PARTIAL; + /* * Watermarks for order-0 must be met for compaction. Note the 2UL. * This is because during migration, copies of pages need to be * allocated and for a short time, the footprint is higher */ - watermark = low_wmark_pages(zone) + (2UL << order); - if (!zone_watermark_ok(zone, 0, watermark, 0, 0)) + watermark += (2UL << order); + if (!zone_watermark_ok(zone, 0, watermark, classzone_idx, alloc_flags)) return COMPACT_SKIPPED; /* * fragmentation index determines if allocation failures are due to * low memory or external fragmentation * - * index of -1000 implies allocations might succeed depending on - * watermarks + * index of -1000 would imply allocations might succeed depending on + * watermarks, but we already failed the high-order watermark check * index towards 0 implies failure is due to lack of memory * index towards 1000 implies failure is due to fragmentation * @@ -1150,10 +1160,6 @@ unsigned long compaction_suitable(struct if (fragindex >= 0 && fragindex <= sysctl_extfrag_threshold) return COMPACT_SKIPPED; - if (fragindex == -1000 && zone_watermark_ok(zone, order, watermark, - 0, 0)) - return COMPACT_PARTIAL; - return COMPACT_CONTINUE; } @@ -1165,7 +1171,8 @@ static int compact_zone(struct zone *zon const int migratetype = gfpflags_to_migratetype(cc->gfp_mask); const bool sync = cc->mode != MIGRATE_ASYNC; - ret = compaction_suitable(zone, cc->order); + ret = compaction_suitable(zone, cc->order, cc->alloc_flags, + cc->classzone_idx); switch (ret) { case COMPACT_PARTIAL: case COMPACT_SKIPPED: @@ -1254,7 +1261,8 @@ out: } static unsigned long compact_zone_order(struct zone *zone, int order, - gfp_t gfp_mask, enum migrate_mode mode, int *contended) + gfp_t gfp_mask, enum migrate_mode mode, int *contended, + int alloc_flags, int classzone_idx) { unsigned long ret; struct compact_control cc = { @@ -1264,6 +1272,8 @@ static unsigned long compact_zone_order( .gfp_mask = gfp_mask, .zone = zone, .mode = mode, + .alloc_flags = alloc_flags, + .classzone_idx = classzone_idx, }; INIT_LIST_HEAD(&cc.freepages); INIT_LIST_HEAD(&cc.migratepages); @@ -1295,6 +1305,7 @@ int sysctl_extfrag_threshold = 500; unsigned long try_to_compact_pages(struct zonelist *zonelist, int order, gfp_t gfp_mask, nodemask_t *nodemask, enum migrate_mode mode, int *contended, + int alloc_flags, int classzone_idx, struct zone **candidate_zone) { enum zone_type high_zoneidx = gfp_zone(gfp_mask); @@ -1303,7 +1314,6 @@ unsigned long try_to_compact_pages(struc struct zoneref *z; struct zone *zone; int rc = COMPACT_DEFERRED; - int alloc_flags = 0; int all_zones_contended = COMPACT_CONTENDED_LOCK; /* init for &= op */ *contended = COMPACT_CONTENDED_NONE; @@ -1312,10 +1322,6 @@ unsigned long try_to_compact_pages(struc if (!order || !may_enter_fs || !may_perform_io) return COMPACT_SKIPPED; -#ifdef CONFIG_CMA - if (gfpflags_to_migratetype(gfp_mask) == MIGRATE_MOVABLE) - alloc_flags |= ALLOC_CMA; -#endif /* Compact each zone in the list */ for_each_zone_zonelist_nodemask(zone, z, zonelist, high_zoneidx, nodemask) { @@ -1326,7 +1332,7 @@ unsigned long try_to_compact_pages(struc continue; status = compact_zone_order(zone, order, gfp_mask, mode, - &zone_contended); + &zone_contended, alloc_flags, classzone_idx); rc = max(status, rc); /* * It takes at least one zone that wasn't lock contended @@ -1335,8 +1341,8 @@ unsigned long try_to_compact_pages(struc all_zones_contended &= zone_contended; /* If a normal allocation would succeed, stop compacting */ - if (zone_watermark_ok(zone, order, low_wmark_pages(zone), 0, - alloc_flags)) { + if (zone_watermark_ok(zone, order, low_wmark_pages(zone), + classzone_idx, alloc_flags)) { *candidate_zone = zone; /* * We think the allocation will succeed in this zone, diff -puN mm/internal.h~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking mm/internal.h --- a/mm/internal.h~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking +++ a/mm/internal.h @@ -168,6 +168,8 @@ struct compact_control { int order; /* order a direct compactor needs */ const gfp_t gfp_mask; /* gfp mask of a direct compactor */ + const int alloc_flags; /* alloc flags of a direct compactor */ + const int classzone_idx; /* zone index of a direct compactor */ struct zone *zone; int contended; /* Signal need_sched() or lock * contention detected during diff -puN mm/page_alloc.c~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking mm/page_alloc.c --- a/mm/page_alloc.c~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking +++ a/mm/page_alloc.c @@ -2341,6 +2341,7 @@ __alloc_pages_direct_compact(gfp_t gfp_m compact_result = try_to_compact_pages(zonelist, order, gfp_mask, nodemask, mode, contended_compaction, + alloc_flags, classzone_idx, &last_compact_zone); current->flags &= ~PF_MEMALLOC; diff -puN mm/vmscan.c~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking mm/vmscan.c --- a/mm/vmscan.c~mm-compaction-pass-classzone_idx-and-alloc_flags-to-watermark-checking +++ a/mm/vmscan.c @@ -2249,7 +2249,7 @@ static inline bool should_continue_recla return true; /* If compaction would go ahead or the allocation would succeed, stop */ - switch (compaction_suitable(zone, sc->order)) { + switch (compaction_suitable(zone, sc->order, 0, 0)) { case COMPACT_PARTIAL: case COMPACT_CONTINUE: return false; @@ -2346,7 +2346,7 @@ static inline bool compaction_ready(stru * If compaction is not ready to start and allocation is not likely * to succeed without it, then keep reclaiming. */ - if (compaction_suitable(zone, order) == COMPACT_SKIPPED) + if (compaction_suitable(zone, order, 0, 0) == COMPACT_SKIPPED) return false; return watermark_ok; @@ -2824,8 +2824,8 @@ static bool zone_balanced(struct zone *z balance_gap, classzone_idx, 0)) return false; - if (IS_ENABLED(CONFIG_COMPACTION) && order && - compaction_suitable(zone, order) == COMPACT_SKIPPED) + if (IS_ENABLED(CONFIG_COMPACTION) && order && compaction_suitable(zone, + order, 0, classzone_idx) == COMPACT_SKIPPED) return false; return true; @@ -2952,8 +2952,8 @@ static bool kswapd_shrink_zone(struct zo * from memory. Do not reclaim more than needed for compaction. */ if (IS_ENABLED(CONFIG_COMPACTION) && sc->order && - compaction_suitable(zone, sc->order) != - COMPACT_SKIPPED) + compaction_suitable(zone, sc->order, 0, classzone_idx) + != COMPACT_SKIPPED) testorder = 0; /* _ Patches currently in -mm which might be from vbabka@xxxxxxx are origin.patch mm-debug-pagealloc-cleanup-page-guard-code.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