From: Chris Mason Subject: Re: [PATCH] jbd jbd2: fix dio write returning EIO whentry_to_release_page fails Date: Wed, 06 Aug 2008 09:25:13 -0400 Message-ID: <1218029114.15342.58.camel@think.oraclecorp.com> References: <6.0.0.20.2.20080804185338.03bcd488@172.19.0.2> <20080804145047.04794bf3.akpm@linux-foundation.org> <1217907353.7611.39.camel@think.oraclecorp.com> <6.0.0.20.2.20080805134429.044569a0@172.19.0.2> <1217953055.7899.11.camel@think.oraclecorp.com> <1217971027.7516.20.camel@mingming-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Hisashi Hifumi , Andrew Morton , jack@suse.cz, linux-ext4@vger.kernel.org, linux-fsdevel@vger.kernel.org To: Mingming Cao Return-path: Received: from rgminet01.oracle.com ([148.87.113.118]:24826 "EHLO rgminet01.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753680AbYHFNmT (ORCPT ); Wed, 6 Aug 2008 09:42:19 -0400 In-Reply-To: <1217971027.7516.20.camel@mingming-laptop> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Tue, 2008-08-05 at 14:17 -0700, Mingming Cao wrote: > =E5=9C=A8 2008-08-05=E4=BA=8C=E7=9A=84 12:17 -0400=EF=BC=8CChris Maso= n=E5=86=99=E9=81=93=EF=BC=9A > > On Tue, 2008-08-05 at 13:51 +0900, Hisashi Hifumi wrote: > > > >> >=20 > > > >> > diff -Nrup linux-2.6.27-rc1.org/fs/jbd/transaction.c=20 > > > >linux-2.6.27-rc1/fs/jbd/transaction.c > > > >> > --- linux-2.6.27-rc1.org/fs/jbd/transaction.c 2008-07-29=20 > > > >19:28:47.000000000 +0900 > > > >> > +++ linux-2.6.27-rc1/fs/jbd/transaction.c 2008-07-29 20:40:1= 2.000000000 +0900 > > > >> > @@ -1764,6 +1764,12 @@ int journal_try_to_free_buffers(journ= al_ > > > >> > */ > > > >> > if (ret =3D=3D 0 && (gfp_mask & __GFP_WAIT) && (gfp_mask &= __GFP_FS)) { > > > >> > journal_wait_for_transaction_sync_data(journal); > > > >> > + > > > >> > + bh =3D head; > > > >> > + do { > > > >> > + while (atomic_read(&bh->b_count)) > > > >> > + schedule(); > > > >> > + } while ((bh =3D bh->b_this_page) !=3D head); > > > >> > ret =3D try_to_free_buffers(page); > > > >> > } > > > >>=20 > > > >> The loop is problematic. If the scheduler decides to keep run= ning this > > > >> task then we have a busy loop. If this task has realtime poli= cy then > > > >> it might even lock up the kernel. > > > >>=20 > > > > > > > >ocfs2 calls journal_try_to_free_buffers too, looping on b_count = might > > > >not be the best idea there either. > > > > > > > >This code gets called from releasepage, which is used other plac= es than > > > >the O_DIRECT invalidation paths, I'd be worried about performanc= e > > > >problems here. > > > > > > >=20 > > > try_to_release_page has gfp_mask parameter. So when try_to_releas= epage > > > is called from performance sensitive part, gfp_mask should not be= set. > > > b_count check loop is inside of (gfp_mask & __GFP_WAIT) && (gfp_m= ask & __GFP_FS) check. > >=20 > > Looks like try_to_free_pages will go into releasepage with wait & f= s > > both set. This kind of change would make me very nervous. > >=20 >=20 > Hi Chris, >=20 > The gfp_mask try_to_free_pages() takes from it's caller will past it > down to try_to_release_page(). Based on the meaning of __GFP_WAIT an= d > GFP_FS, if the upper level caller set these two flags, I assume the > upper level caller expect delay and wait for fs to finish? >=20 >=20 > But I agree that using a loop in journal_try_to_free_buffers() to wai= t > for the busy bh release the counter is expensive... I rediscovered your old thread about trying to do this in a launder_pag= e call ;) Does it make more sense to fix do_launder_page to call into the FS on every page, and let the FS check for PageDirty on its own? That way invalidate_inode_pages2_range basically gets its own private call into the FS that says wait around until this page is really free. -chris -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html