From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753387AbbAEI5A (ORCPT ); Mon, 5 Jan 2015 03:57:00 -0500 Received: from cantor2.suse.de ([195.135.220.15]:44788 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753204AbbAEI46 (ORCPT ); Mon, 5 Jan 2015 03:56:58 -0500 From: Vlastimil Babka To: Andrew Morton , linux-mm@kvack.org Cc: linux-kernel@vger.kernel.org, Vlastimil Babka , Mel Gorman , Johannes Weiner , Michal Hocko , Vladimir Davydov , Rik van Riel , Subject: [PATCH V3 2/2] mm, vmscan: wake up all pfmemalloc-throttled processes at once Date: Mon, 5 Jan 2015 09:56:43 +0100 Message-Id: <1420448203-30212-2-git-send-email-vbabka@suse.cz> X-Mailer: git-send-email 2.1.2 In-Reply-To: <1420448203-30212-1-git-send-email-vbabka@suse.cz> References: <1420448203-30212-1-git-send-email-vbabka@suse.cz> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Kswapd in balance_pgdate() currently uses wake_up() on processes waiting in throttle_direct_reclaim(), which only wakes up a single process. This might leave processes waiting for longer than necessary, until the check is reached in the next loop iteration. Processes might also be left waiting if zone was fully balanced in single iteration. Note that the comment in balance_pgdat() also says "Wake them", so waking up a single process does not seem intentional. Thus, replace wake_up() with wake_up_all(). Signed-off-by: Vlastimil Babka Cc: Mel Gorman Cc: Johannes Weiner Cc: Michal Hocko Cc: Vladimir Davydov Cc: Rik van Riel --- mm/vmscan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index ab2505c..f73afa3 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -3175,7 +3175,7 @@ static unsigned long balance_pgdat(pg_data_t *pgdat, int order, */ if (waitqueue_active(&pgdat->pfmemalloc_wait) && pfmemalloc_watermark_ok(pgdat)) - wake_up(&pgdat->pfmemalloc_wait); + wake_up_all(&pgdat->pfmemalloc_wait); /* * Fragmentation may mean that the system cannot be rebalanced -- 2.1.2