From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1422742AbdAIRpV (ORCPT ); Mon, 9 Jan 2017 12:45:21 -0500 Received: from mx2.suse.de ([195.135.220.15]:53329 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932792AbdAIRpP (ORCPT ); Mon, 9 Jan 2017 12:45:15 -0500 Date: Mon, 9 Jan 2017 18:45:11 +0100 From: Michal Hocko To: Eric Dumazet Cc: linux-mm@kvack.org, LKML Subject: Re: __GFP_REPEAT usage in fq_alloc_node Message-ID: <20170109174511.GA8306@dhcp22.suse.cz> References: <20170106152052.GS5556@dhcp22.suse.cz> <20170106160743.GU5556@dhcp22.suse.cz> <20170106161944.GW5556@dhcp22.suse.cz> <20170109102219.GF7495@dhcp22.suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.6.0 (2016-04-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon 09-01-17 08:00:16, Eric Dumazet wrote: > On Mon, Jan 9, 2017 at 2:22 AM, Michal Hocko wrote: > > > > the changelog doesn't mention it but this, unlike other kvmalloc > > conversions is not without functional changes. The kmalloc part > > will be weaker than it is with the original code for !costly (<64kB) > > requests, because we are enforcing __GFP_NORETRY to break out from the > > page allocator which doesn't really fail such a small requests. > > > > Now the question is what those code paths really prefer. Do they really > > want to potentially loop in the page allocator and invoke the OOM killer > > when the memory is short/fragmeted? I mean we can get into a situation > > when no order-3 pages can be compacted and shooting the system down just > > for that reason sounds quite dangerous to me. > > > > So the main question is how hard should we try before falling back to > > vmalloc here? > > This patch is fine : > > 1) Default hash size is 1024 slots, 8192 bytes on 64bit arches. What about those non-default configurations. Do they really want to invoke the OOM killer rather than fallback to the vmalloc? -- Michal Hocko SUSE Labs