Date: Tue, 7 Mar 2017 12:29:22 +0300 From: Slawa Olhovchenkov <slw@zxy.spb.ru> To: Andriy Gapon <avg@FreeBSD.org> Cc: src-committers@FreeBSD.org, svn-src-all@FreeBSD.org, svn-src-head@FreeBSD.org Subject: Re: svn commit: r314274 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs Message-ID: <20170307092922.GZ70430@zxy.spb.ru> In-Reply-To: <5e1fa060-719e-4677-164a-856e3f6c779f@FreeBSD.org> References: <201702251703.v1PH3nQd044474@repo.freebsd.org> <20170307035308.GN15630@zxy.spb.ru> <e7adf535-2c30-dba3-478f-6c1289af817e@FreeBSD.org> <20170307091818.GY70430@zxy.spb.ru> <5e1fa060-719e-4677-164a-856e3f6c779f@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Mar 07, 2017 at 11:24:42AM +0200, Andriy Gapon wrote: > On 07/03/2017 11:18, Slawa Olhovchenkov wrote: > > On Tue, Mar 07, 2017 at 11:02:46AM +0200, Andriy Gapon wrote: > > > >> On 07/03/2017 05:53, Slawa Olhovchenkov wrote: > >>> On Sat, Feb 25, 2017 at 05:03:49PM +0000, Andriy Gapon wrote: > >>> > >>>> Author: avg > >>>> Date: Sat Feb 25 17:03:48 2017 > >>>> New Revision: 314274 > >>>> URL: https://svnweb.freebsd.org/changeset/base/314274 > >>>> > >>>> Log: > >>>> l2arc: try to fix write size calculation broken by Compressed ARC commit > >>>> > >>>> While there, make a change to not evict a first buffer outside the > >>>> requested eviciton range. > >>>> > >>>> To do: > >>>> - give more consistent names to the size variables > >>>> - upstream to OpenZFS > >>> > >>> Mem: 6647M Active, 8399M Inact, 87G Wired, 49K Buf, 23G Free > >>> ^^^^^^^^^ > >>> ARC: 130G Total, 255M MFU, 80G MRU, 267M Anon, 72M Header, 41M Other > >>> ^^^^^ > >>> > >>> 128GB phys mem > >>> > >> > >> Why this commit? > > > > revert this commit and r307265 resolve this. > > i.e. may be r307265, but r307265 related to this. > > So, why not comment on r307265 or bisect first? You are kidding?! How to work this commit w/o r307265? r307265 is l2arc compression suport (buggly). this commit fix l2arc part. arc part still buggly. hmm, may be I am not clean, I am not rever ALL to r307265. I am rever exact r307265 and 314274. This is like bisect by one try.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20170307092922.GZ70430>