Date: Tue, 17 Feb 2009 17:05:12 -0500 From: Mike Tancsa <mike@sentex.net> To: Luigi Rizzo <rizzo@iet.unipi.it>, Scott Long <scottl@samsco.org> Cc: freebsd-current@freebsd.org, freebsd-stable@freebsd.org Subject: Re: HEADS UP: Major CAM performance regression Message-ID: <200902172205.n1HM5IUf022092@pyroxene.sentex.ca> In-Reply-To: <20090217075742.GA69308@onelab2.iet.unipi.it> References: <499551B9.7050805@samsco.org> <gn3ssi$j1r$1@ger.gmane.org> <4995DFE5.1020205@samsco.org> <9bbcef730902131421r53efa13dq371658888747f387@mail.gmail.com> <4996D635.3000802@samsco.org> <20090217075742.GA69308@onelab2.iet.unipi.it>
next in thread | previous in thread | raw e-mail | index | archive | help
At 02:57 AM 2/17/2009, Luigi Rizzo wrote: >On Sat, Feb 14, 2009 at 07:33:25AM -0700, Scott Long wrote: >... > > >I'll try your suggestion if you have one. > > > > I don't have a magic universal testing suite in my back pocket, sorry. > > You need to look at your expected workload and develop tests to simulate > > it. When I do testing during driver development, I try a lot of > > different parallel, sequential, large i/o, and small i/o combinations. > >i just committed a port sysutils/fio that perhaps can help testing >IO performance with various patterns. Hi, Do you have any suggestions as to what tests to run with fio ? Am I right in assuming that the Areca is hit by this bug ? 0[releng7]# camcontrol tags da0 (pass0:arcmsr0:0:0:0): device openings: 1 0[releng7]# ---Mike
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200902172205.n1HM5IUf022092>