Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965350AbXBLTuA (ORCPT ); Mon, 12 Feb 2007 14:50:00 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S965352AbXBLTuA (ORCPT ); Mon, 12 Feb 2007 14:50:00 -0500 Received: from e2.ny.us.ibm.com ([32.97.182.142]:37638 "EHLO e2.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S965350AbXBLTt6 (ORCPT ); Mon, 12 Feb 2007 14:49:58 -0500 Subject: Re: [PATCH -mm] readahead: partial sendfile fix From: Ram Pai To: Fengguang Wu Cc: Andrew Morton , Suparna Bhattacharya , John Tran , Mike Sullivan , Chris Mason , Steven Pratt , Badari Pulavarty , Mingming Cao , Linux Kernel In-Reply-To: <20070210014049.GA11269@mail.ustc.edu.cn> References: <20070210014049.GA11269@mail.ustc.edu.cn> Content-Type: text/plain Date: Mon, 12 Feb 2007 11:49:10 -0800 Message-Id: <1171309751.2891.70.camel@ram.us.ibm.com> Mime-Version: 1.0 X-Mailer: Evolution 2.8.0 (2.8.0-7.fc6) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3718 Lines: 110 On Sat, 2007-02-10 at 09:40 +0800, Fengguang Wu wrote: > Enable readahead to handle partially done read requests, e.g. > > sendfile(188, 1921, [1478592], 19553028) = 37440 > sendfile(188, 1921, [1516032], 19515588) = 28800 > sendfile(188, 1921, [1544832], 19486788) = 37440 > sendfile(188, 1921, [1582272], 19449348) = 14400 > sendfile(188, 1921, [1596672], 19434948) = 37440 > sendfile(188, 1921, [1634112], 19397508) = 37440 > > In the above strace log, > - some lighttpd is doing _sequential_ reading > - every sendfile() returns with only _partial_ work done > > page_cache_readahead() expects that if it returns @next_index, it will > be > called exactly at @next_index next time. That's not true here. So the > pattern > will be falsely recognized as a random read trace. > > Also documented in "Linux AIO Performance and Robustness for > Enterprise > Workloads" section 3.5: > > sendfile(fd, 0, 2GB, fd2) = 8192, > tells readahead about up to 128KB of the read > sendfile(fd, 8192, 2GB - 8192, fd2) = 8192, > tells readahead about 8KB - 132KB of the read > sendfile(fd, 16384, 2GB - 16384, fd2) = 8192, > tells readahead about 16KB-140KB of the read > ... > This confuses the readahead logic about the I/O pattern which > appears > to be 0-128K, 8K-132K, 16K-140K instead of clear sequentiality > from > 0-2GB that is really appropriate. > > Retry based AIO shares the same read pattern and readahead problem. > In this case, simply disabling readahead on restarted aio is not a > good option: > we still need to call into readahead in the rare case of (req_size > > ra_max). The solution you proposed seems kludgy to me. If you determine that the its a restarted aio, then start reading from where readahead had left reading from earlier. To me a simple fix is: - if (unlikely(aio_restarted())) - next_index = last_index; /* Avoid repeat readahead */ + if (unlikely(aio_restarted())) + next_index = min(prev_index+1, last_index); No? RP > > Signed-off-by: Fengguang Wu > --- > mm/filemap.c | 3 --- > mm/readahead.c | 9 +++++++++ > 2 files changed, 9 insertions(+), 3 deletions(-) > > --- linux-2.6.20-rc6-mm3.orig/mm/readahead.c > +++ linux-2.6.20-rc6-mm3/mm/readahead.c > @@ -581,6 +581,15 @@ page_cache_readahead(struct address_spac > int sequential; > > /* > + * A previous read request is partially completed, > + * causing the retried/continued read calls into us > prematurely. > + */ > + if (ra->start < offset && > + offset < ra->prev_page && > + ra->prev_page < ra->ahead_start + > ra->ahead_size) > + goto out; > + > + /* > * We avoid doing extra work and bogusly perturbing the > readahead > * window expansion logic. > */ > --- linux-2.6.20-rc6-mm3.orig/mm/filemap.c > +++ linux-2.6.20-rc6-mm3/mm/filemap.c > @@ -915,9 +915,6 @@ void do_generic_mapping_read(struct addr > if (!isize) > goto out; > > - if (unlikely(aio_restarted())) > - next_index = last_index; /* Avoid repeat readahead */ > - > end_index = (isize - 1) >> PAGE_CACHE_SHIFT; > for (;;) { > struct page *page; > - 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/