From mboxrd@z Thu Jan 1 00:00:00 1970 From: "changtao381" Subject: =?UTF-8?Q?=E7=AD=94=E5=A4=8D:_journal_alignment?= Date: Fri, 20 Nov 2015 17:12:25 +0800 Message-ID: <023e01d12373$941f1050$bc5d30f0$@com> References: <019801d1236e$206f12e0$614d38a0$@com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from m12-13.163.com ([220.181.12.13]:56644 "EHLO m12-13.163.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933930AbbKTJMl convert rfc822-to-8bit (ORCPT ); Fri, 20 Nov 2015 04:12:41 -0500 In-Reply-To: Content-Language: zh-cn Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Piotr.Dalek@ts.fujitsu.com, haomaiwang@gmail.com, ceph-devel@vger.kernel.org Hi All, Thanks for you apply! If directioIO + async IO requirement that alignment, it shouldn't align= ed by PAGE for each journal entry. =46or it may write many entries of journal once time So it should do alignment after [prepare_multi_write] rather than do it= each in [prepare_single_write] ? Am I right? -----=E9=82=AE=E4=BB=B6=E5=8E=9F=E4=BB=B6----- =E5=8F=91=E4=BB=B6=E4=BA=BA: Piotr.Dalek@ts.fujitsu.com [mailto:Piotr.D= alek@ts.fujitsu.com]=20 =E5=8F=91=E9=80=81=E6=97=B6=E9=97=B4: 2015=E5=B9=B411=E6=9C=8820=E6=97=A5= 16:42 =E6=94=B6=E4=BB=B6=E4=BA=BA: changtao381; ceph-devel@vger.kernel.org =E4=B8=BB=E9=A2=98: RE: journal alignment > -----Original Message----- > From: ceph-devel-owner@vger.kernel.org [mailto:ceph-devel- > owner@vger.kernel.org] On Behalf Of changtao381 > Sent: Friday, November 20, 2015 9:33 AM >=20 > HI All, >=20 > Why it is needed an entry of journal t is aligned by CEPH_PAGE_MASK= ? For > it causes the data of journal write are amplified by 2X for small io >=20 > For example write io size 4096 bytes, it may write 8192 bytes >=20 > prepare_single_write 2 will write 98304 : seq 24 len 4324 -> 8192 (he= ad 40 > pre_pad 0 ebl 4324 post_pad 3788 tail 40) (ebl alignment -1) >=20 > Thanks! It is a direct-IO+async IO requirement (writing on page-size boundary w= hich is incidentally also a multiply of sector size). With best regards / Pozdrawiam Piotr Da=C5=82ek -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html