Date: Fri, 5 Jan 2007 16:59:43 -0800 From: Paul Allen <nospam@nospam.com> To: Luigi Rizzo <rizzo@icir.org> Cc: freebsd-current@freebsd.org, lulf@stud.ntnu.no, Ivan Voras <ivoras@fer.hr>, freebsd-geom@freebsd.org Subject: Re: Pluggable Disk Schedulers in GEOM Message-ID: <20070106005943.GB8574@heave.ugcs.caltech.edu> In-Reply-To: <20070105145915.B94175@xorpc.icir.org> References: <20070105123127.gnk0v58p44488g48@webmail.ntnu.no> <4085.1167997049@critter.freebsd.dk> <enlelj$63g$1@sea.gmane.org> <20070105184621.dh8kgoy7ko4gk4gc@webmail.ntnu.no> <20070105102905.A91349@xorpc.icir.org> <20070105212543.GA8574@heave.ugcs.caltech.edu> <20070105145915.B94175@xorpc.icir.org>
next in thread | previous in thread | raw e-mail | index | archive | help
>From Luigi Rizzo <rizzo@icir.org>, Fri, Jan 05, 2007 at 02:59:15PM -0800: > that does not mean that a scheduler is useless in general. > in fact, what i find questionable is hardwiring unnecessary > (in the sense that they are done only for performance reasons) > ordering constraints in the layers above. I cannot comment for I agree in so much as my point was strictly focused on ordering constraints imposed for consistency reasons. softupdates is not a performance mechanism per se; it is a mechanism to allow async writes of fs data occur in a consistent fashion. Anyways, you can easily resolve this issue by mounting async with softupdates disabled and then repeating the test.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070106005943.GB8574>