Date: Sun, 13 Apr 2003 22:51:42 -0600 (MDT) From: "M. Warner Losh" <imp@bsdimp.com> To: allbery@ece.cmu.edu Cc: freebsd-stable@freebsd.org Subject: Re: PATCH: Forcible delaying of UFS (soft)updates Message-ID: <20030413.225142.39143716.imp@bsdimp.com> In-Reply-To: <1050134860.7300.0.camel@rushlight.kf8nh.apk.net> References: <3E976EBD.C3E66EF8@tel.fer.hr> <20030412033307.GR30960@elvis.mu.org> <1050134860.7300.0.camel@rushlight.kf8nh.apk.net>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <1050134860.7300.0.camel@rushlight.kf8nh.apk.net>
"Brandon S. Allbery " KF8NH <allbery@ece.cmu.edu> writes:
: On Fri, 2003-04-11 at 23:33, Alfred Perlstein wrote:
: > * Marko Zec <zec@tel.fer.hr> [030411 19:01] wrote:
: > > - fsync() no longer flushes the buffers to disk, but returns immediately
: > > instead;
: >
: > This is really the only bad thing I can see here, what about introducing
: > a slight delay and seeing if one can coalesce the writes? Is this
: > part really needed? Making fsync() not work is a good way to make
: > any sort of userland based transactional system break badly.
:
: If you're running that kind of thing you really don't want to be using
: extended delays anyway, I'd think.
Trouble is that lots of things uses fsync behind the scenes...
Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030413.225142.39143716.imp>