From: David Rientjes Subject: Re: [patch v2 1/5] mm: add nofail variants of kmalloc kcalloc and kzalloc Date: Sun, 5 Sep 2010 16:03:52 -0700 (PDT) Message-ID: References: <4C7F5951.6040809@gmail.com> <20100902145141.GA3273@quack.suse.cz> <20100903071506.6e6b4d63@notabene> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Cc: Jan Kara , Jiri Slaby , Andrew Morton , Alasdair G Kergon , Chris Mason , Steven Whitehouse , Jens Axboe , Frederic Weisbecker , linux-raid@vger.kernel.org, linux-btrfs@vger.kernel.org, cluster-devel@redhat.com, linux-ext4@vger.kernel.org, reiserfs-devel@vger.kernel.org, linux-kernel@vger.kernel.org To: Neil Brown Return-path: Received: from smtp-out.google.com ([216.239.44.51]:18830 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754503Ab0IEXEF (ORCPT ); Sun, 5 Sep 2010 19:04:05 -0400 In-Reply-To: <20100903071506.6e6b4d63@notabene> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Fri, 3 Sep 2010, Neil Brown wrote: > I'm actually a bit confused about this too. > I thought David said he was removing a branch on the *slow* path - which make > sense as you wouldn't even test NOFAIL until you had a failure. > Why are branches on the slow-path an issue?? They aren't necessarily an issue in the performance sense, this is a cleanup series since all converted callers to using these new functions (and the eventual removal of __GFP_NOFAIL entirely) are using the bit unnecessarily since they all have orders that implicitly loop in the page allocator forever already, with or without the flag.