From mboxrd@z Thu Jan 1 00:00:00 1970 From: Shaohua Li Subject: Re: [PATCH][RFC] A readahead complete notify approach to implement buffer aio Date: Fri, 04 Nov 2011 10:59:19 +0800 Message-ID: <1320375559.22361.199.camel@sli10-conroe> References: <1320138024-10837-1-git-send-email-gaoyang.zyh@taobao.com> <4EB2D8CF.2080108@coly.li> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Cc: "i@coly.li" , Zhu Yanhai , "linux-kernel@vger.kernel.org" , "bcrl@kvack.org" , "viro@zeniv.linux.org.uk" , "linux-aio@kvack.org" , "linux-fsdevel@vger.kernel.org" , "jaxboe@fusionio.com" , Zhu Yanhai To: Jeff Moyer Return-path: In-Reply-To: Sender: owner-linux-aio@kvack.org List-Id: linux-fsdevel.vger.kernel.org On Fri, 2011-11-04 at 02:01 +0800, Jeff Moyer wrote: > Coly Li writes: >=20 > > On 2011=E5=B9=B411=E6=9C=8801=E6=97=A5 17:00, Zhu Yanhai Wrote: > >> The current libaio/aio has to be Direct-IO, otherwise it falls back = into sync IO. > >> However, the aio core has already been asychronous naturally. This p= atch adds a complete > >> notify mechanism to implement buffer aio, the main idea is to readah= ead()-like in > >> io_submit(), counts the non-uptodated pages assocaiated with each io= cb, then put each ref > >> in the bio complete path just before unlock_page(), and hook them on= to the aio ring buffer > >> finally when the ref drops to zero. In io_getevents(), we call vfs_r= ead() as a safe net > >> since there is still little possibility that the pages had brought i= n were reclaimed > >> between io_submit() and io_getevents(). > >>=20 > >> I have tested this patch for a while, for the small size random io r= equest, its > >> performance is more or less the same with the traditional aio, for t= he big io request, > >> the overhead of one extra memory copy arises. > >>=20 > >> I think so far it has at least below obvious drawbacks, > >>=20 > >> * mpage_readpage() is a really narrow interface, I have no way to pa= ss down > >> the new control struct baiocb, so I just put it into struct task_str= uct and > >> refer it by current() as a workaround. > >>=20 > >> * the do_baio_read() routine is heavily similar with do_generic_file= _read(), but > >> the latter is really hard to modify. I think we may stuff these code= down into the > >> readahead path to reduce code reduplication. > >>=20 > >> Hopefully the explanations are clear enough and don't muddy the wate= r any worse. > >> I figure the code does need some better comments, and any suggestion= are welcome. > >>=20 > >> Signed-off-by: Zhu Yanhai > >>=20 > >> --- > >> fs/aio.c | 319 ++++++++++++++++++++++++++++++++= ++++++++++- > >> fs/buffer.c | 26 ++++- > >> fs/mpage.c | 28 ++++- > >> include/linux/aio.h | 9 ++ > >> include/linux/aio_abi.h | 1 + > >> include/linux/blk_types.h | 2 + > >> include/linux/buffer_head.h | 3 + > >> include/linux/page-flags.h | 2 + > >> include/linux/sched.h | 1 + > >> 9 files changed, 386 insertions(+), 5 deletions(-) > >>=20 > > > > Hmm, I don't see the usage from user space. Is it possible to post a = demo code in user space, so people are able to > > understand how to use/test your patch. >=20 > He added a new IOCB_CMD_blah for buffered aio reads. That is, > unfortunately, a really poor way to go about doing things. Please take > a look at the work Jens did on implementing buffered aio. It can be > found in his linux-block git tree. Interesting, I thought Jens's work is to make io_submit not blocking. That's good to know it supports async buffer io too. Last time I look at the code, it appears there still are threads to do the aio retry, so what's the difference just letting user space thread do a sync buffer io? -- To unsubscribe, send a message with 'unsubscribe linux-aio' in the body to majordomo@kvack.org. For more info on Linux AIO, see: http://www.kvack.org/aio/ Don't email: aart@kvack.org