Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933088AbZGPTOZ (ORCPT ); Thu, 16 Jul 2009 15:14:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S933069AbZGPTOX (ORCPT ); Thu, 16 Jul 2009 15:14:23 -0400 Received: from smtp-out.google.com ([216.239.45.13]:6768 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932439AbZGPTOW (ORCPT ); Thu, 16 Jul 2009 15:14:22 -0400 DomainKey-Signature: a=rsa-sha1; s=beta; d=google.com; c=nofws; q=dns; h=date:from:x-x-sender:to:cc:subject:in-reply-to:message-id: references:user-agent:mime-version:content-type:x-system-of-record; b=NRqLVnZWdm/jZ13v1ODS6CNEcR9QIobH7UD7p7pHTQNKGPIGnWAYIwdfz2wY5LAcD k1Up7nzD+iWp2Avq9enXw== Date: Thu, 16 Jul 2009 12:14:13 -0700 (PDT) From: David Rientjes X-X-Sender: rientjes@chino.kir.corp.google.com To: Mel Gorman cc: Andrew Morton , Nick Piggin , linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH] page-allocator: Ensure that processes that have been OOM killed exit the page allocator (resend) In-Reply-To: <20090716110328.GB22499@csn.ul.ie> Message-ID: References: <20090715104944.GC9267@csn.ul.ie> <20090716110328.GB22499@csn.ul.ie> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-System-Of-Record: true Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2104 Lines: 58 On Thu, 16 Jul 2009, Mel Gorman wrote: > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > index 4b8552e..b381a6b 100644 > --- a/mm/page_alloc.c > +++ b/mm/page_alloc.c > @@ -1830,8 +1830,6 @@ rebalance: > if (order > PAGE_ALLOC_COSTLY_ORDER && > !(gfp_mask & __GFP_NOFAIL)) > goto nopage; > - > - goto restart; > } > } > > This isn't right (and not only because it'll add a compiler warning because `restart' is now unused). This would immediately fail any allocation that triggered the oom killer and ended up being selected that isn't __GFP_NOFAIL, even if it would have succeeded without even killing any task simply because it allocates without watermarks. It will also, coupled with your earlier patch, inappropriately warn about an infinite loop with __GFP_NOFAIL even though it hasn't even attempted to loop once since that decision is now handled by should_alloc_retry(). The liklihood of such an infinite loop, considering only one thread per system (or cpuset) can be TIF_MEMDIE at a time, is very low. I've never seen memory reserves completely depleted such that the next high-priority allocation wouldn't succeed so that current could handle its pending SIGKILL. You get the same behavior with my patch, but are allowed to try the high priority allocation again for the attempt that triggered the oom killer (and not only subsequent ones). --- diff --git a/mm/page_alloc.c b/mm/page_alloc.c --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1789,6 +1789,10 @@ rebalance: if (p->flags & PF_MEMALLOC) goto nopage; + /* Avoid allocations with no watermarks from looping endlessly */ + if (test_thread_flag(TIF_MEMDIE) && !(gfp_mask & __GFP_NOFAIL)) + goto nopage; + /* Try direct reclaim and then allocating */ page = __alloc_pages_direct_reclaim(gfp_mask, order, zonelist, high_zoneidx, -- 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/