Date: Thu, 24 Nov 2011 23:26:41 +0400 From: Lev Serebryakov <lev@FreeBSD.org> To: Ivan Voras <ivoras@freebsd.org> Cc: freebsd-fs@freebsd.org Subject: Re: Does UFS2 send BIO_FLUSH to GEOM when update metadata (with softupdates)? Message-ID: <113503620.20111124232641@serebryakov.spb.ru> In-Reply-To: <jals5e$96f$1@dough.gmane.org> References: <1957615267.20111123230026@serebryakov.spb.ru> <20111123194444.GE50300@deviant.kiev.zoral.com.ua> <jals5e$96f$1@dough.gmane.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Hello, Ivan. You wrote 24 =ED=EE=FF=E1=F0=FF 2011 =E3., 20:42:15: > Some time ago I've written a patch as an learning excercise which issues > a BIO_FLUSH when it thinks SU needs it - it may or may not be correct so > I didn't even try to commit it: > http://people.freebsd.org/~ivoras/diffs/fsync_flush.patch > Of course, reviews are welcome :) I'll give it a try. But adding BIO_SYNC flag will be better to avoid additional operations. I'm trying to measuer it impact now for very naive implementation (add BIO_SYNC flag to system, teach ata/scsi layer about it and set it for every BIO which is created from BUF without "ASYNC" flag). --=20 // Black Lion AKA Lev Serebryakov <lev@FreeBSD.org>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?113503620.20111124232641>