Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751458Ab1BZGoT (ORCPT ); Sat, 26 Feb 2011 01:44:19 -0500 Received: from mx1.redhat.com ([209.132.183.28]:27863 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751376Ab1BZGoR (ORCPT ); Sat, 26 Feb 2011 01:44:17 -0500 Date: Sat, 26 Feb 2011 07:43:39 +0100 From: Andrea Arcangeli To: Arthur Marsh Cc: Clemens Ladisch , alsa-user@lists.sourceforge.net, linux-kernel@vger.kernel.org, Mel Gorman Subject: Re: [Alsa-user] new source of MIDI playback slow-down identified - 5a03b051ed87e72b959f32a86054e1142ac4cf55 thp: use compaction in kswapd for GFP_ATOMIC order > 0 Message-ID: <20110226064339.GC19630@random.random> References: <20110222161513.GC13092@random.random> <4D63F6C0.7060204@internode.on.net> <20110223162432.GL31195@random.random> <20110223165550.GP31195@random.random> <4D65691E.9080600@internode.on.net> <20110223212541.GV31195@random.random> <4D65825F.5080403@internode.on.net> <20110223235946.GW31195@random.random> <4D65B71D.5030109@internode.on.net> <20110224015405.GD31195@random.random> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20110224015405.GD31195@random.random> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5670 Lines: 170 On Thu, Feb 24, 2011 at 02:54:05AM +0100, Andrea Arcangeli wrote: > Ok so tomorrow I'll get all results on these 3 kernels ( > compaction-kswapd-3+compaction_alloc_lowlat-2 vs > compaction-no-kswapd-3+compaction_alloc_lowlat-2 vs > compaction_alloc_lowlat2) on network server load, where throughout is > measured in addition to latency. Then we'll have a better picture to Latency is still lowest with compaction-no-kswapd-3. compaction_alloc still is at the top of the profiling with compaction-kswapd-3+compaction_alloc_lowlat-2. However compaction-kswapd-3 reduced the overhead somewhat but not enough to be as fast as with compaction-no-kswapd-3 (even if much better than before). So we should apply compaction-no-kswapd-3 to 2.6.38 I think. ==== Subject: compaction: fix high compaction latencies and remove compaction-kswapd From: Andrea Arcangeli It's safer to stop calling compaction from kswapd as that creates too high load during memory pressure that can't be offseted by the improved performance of compound allocations. NOTE: this is not related to THP (THP allocations uses __GFP_NO_KSWAPD), this is only related to frequent and small order allocations that make kswapd go wild with compaction. Signed-off-by: Andrea Arcangeli --- --- a/mm/compaction.c +++ b/mm/compaction.c @@ -405,10 +423,7 @@ static int compact_finished(struct zone return COMPACT_COMPLETE; /* Compaction run is not finished if the watermark is not met */ - if (cc->compact_mode != COMPACT_MODE_KSWAPD) - watermark = low_wmark_pages(zone); - else - watermark = high_wmark_pages(zone); + watermark = low_wmark_pages(zone); watermark += (1 << cc->order); if (!zone_watermark_ok(zone, cc->order, watermark, 0, 0)) @@ -421,15 +436,6 @@ static int compact_finished(struct zone if (cc->order == -1) return COMPACT_CONTINUE; - /* - * Generating only one page of the right order is not enough - * for kswapd, we must continue until we're above the high - * watermark as a pool for high order GFP_ATOMIC allocations - * too. - */ - if (cc->compact_mode == COMPACT_MODE_KSWAPD) - return COMPACT_CONTINUE; - /* Direct compactor: Is a suitable page free? */ for (order = cc->order; order < MAX_ORDER; order++) { /* Job done if page is free of the right migratetype */ @@ -551,8 +557,7 @@ static int compact_zone(struct zone *zon unsigned long compact_zone_order(struct zone *zone, int order, gfp_t gfp_mask, - bool sync, - int compact_mode) + bool sync) { struct compact_control cc = { .nr_freepages = 0, @@ -561,7 +566,6 @@ unsigned long compact_zone_order(struct .migratetype = allocflags_to_migratetype(gfp_mask), .zone = zone, .sync = sync, - .compact_mode = compact_mode, }; INIT_LIST_HEAD(&cc.freepages); INIT_LIST_HEAD(&cc.migratepages); @@ -607,8 +611,7 @@ unsigned long try_to_compact_pages(struc nodemask) { int status; - status = compact_zone_order(zone, order, gfp_mask, sync, - COMPACT_MODE_DIRECT_RECLAIM); + status = compact_zone_order(zone, order, gfp_mask, sync); rc = max(status, rc); /* If a normal allocation would succeed, stop compacting */ @@ -639,7 +642,6 @@ static int compact_node(int nid) .nr_freepages = 0, .nr_migratepages = 0, .order = -1, - .compact_mode = COMPACT_MODE_DIRECT_RECLAIM, }; zone = &pgdat->node_zones[zoneid]; --- a/include/linux/compaction.h +++ b/include/linux/compaction.h @@ -11,9 +11,6 @@ /* The full zone was compacted */ #define COMPACT_COMPLETE 3 -#define COMPACT_MODE_DIRECT_RECLAIM 0 -#define COMPACT_MODE_KSWAPD 1 - #ifdef CONFIG_COMPACTION extern int sysctl_compact_memory; extern int sysctl_compaction_handler(struct ctl_table *table, int write, @@ -28,8 +25,7 @@ extern unsigned long try_to_compact_page bool sync); extern unsigned long compaction_suitable(struct zone *zone, int order); extern unsigned long compact_zone_order(struct zone *zone, int order, - gfp_t gfp_mask, bool sync, - int compact_mode); + gfp_t gfp_mask, bool sync); /* Do not skip compaction more than 64 times */ #define COMPACT_MAX_DEFER_SHIFT 6 @@ -74,8 +70,7 @@ static inline unsigned long compaction_s } static inline unsigned long compact_zone_order(struct zone *zone, int order, - gfp_t gfp_mask, bool sync, - int compact_mode) + gfp_t gfp_mask, bool sync) { return COMPACT_CONTINUE; } --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2385,7 +2385,6 @@ loop_again: * cause too much scanning of the lower zones. */ for (i = 0; i <= end_zone; i++) { - int compaction; struct zone *zone = pgdat->node_zones + i; int nr_slab; unsigned long balance_gap; @@ -2426,24 +2425,9 @@ loop_again: sc.nr_reclaimed += reclaim_state->reclaimed_slab; total_scanned += sc.nr_scanned; - compaction = 0; - if (order && - zone_watermark_ok(zone, 0, - high_wmark_pages(zone), - end_zone, 0) && - !zone_watermark_ok(zone, order, - high_wmark_pages(zone), - end_zone, 0)) { - compact_zone_order(zone, - order, - sc.gfp_mask, false, - COMPACT_MODE_KSWAPD); - compaction = 1; - } - if (zone->all_unreclaimable) continue; - if (!compaction && nr_slab == 0 && + if (nr_slab == 0 && !zone_reclaimable(zone)) zone->all_unreclaimable = 1; /* -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/