Date: Mon, 26 Mar 2007 18:38:03 -0400 From: Mikhail Teterin <mi+kde@aldan.algebra.com> To: =?iso-8859-1?q?S=F8ren_Schmidt?= <sos@deepcore.dk> Cc: Dmitry Pryanishnikov <dmitry@atlantis.dp.ua>, stable@freebsd.org, Derek Ragona <derek@computinginnovations.com>, Brian Candler <B.Candler@pobox.com> Subject: Re: pitiful performance of an SATA150 drive Message-ID: <200703261838.04131@aldan> In-Reply-To: <46083CBB.9050404@deepcore.dk> References: <200603010505.k2155HfQ003205@aldan.algebra.com> <200703261632.01380@aldan> <46083CBB.9050404@deepcore.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 26 March 2007 17:35, Søren Schmidt wrote: = Nopes its the stock AMD and a SiI chip.. Yes, that's correct. Sorry for the confusion. = Anyhow there has been some changes in that area that actually might fix = an interrupt routing bogon. Please try the attached patch against an up = to date 6-stable source and let me know if that helps... Ok, I'm certainly seeing improvement. The amount of "ehci" interrupts is down to hundreds (although still very high, considering, there is not USB activity). The disk's write throughput increased a little from 7.5Mb/s, and even spikes to 10Mb/s occasionally now, while averaging at about 8.3Mb/s. Curiously, the bandwidth appears BETTER (9Mb/s), when boinc (setiathome) IS RUNNING on all four cores... I'm running a compressed dump now to see, if the data corruption is still here. That said, it still sucks... The drive can read at the healthy 60Mb/s (and higher) -- I'd expect writing to average at least 20Mb/s, when recording a single stream. BTW, when I just got the drive 15 months ago, reading sucked too. But it improved over the period -- not sure, with which revision... Thanks! -mi
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200703261838.04131>