From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Lord Subject: Re: libata sata_via unusual behaviour Date: Thu, 01 May 2008 22:40:23 -0400 Message-ID: <481A7F17.6010103@rtr.ca> References: <3bbeafe90805011018o28f46794h369242ccb02b7ced@mail.gmail.com> <3bbeafe90805011158n5b45283awd43a7a534ba11456@mail.gmail.com> <20080501195823.79786815@core> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from rtr.ca ([76.10.145.34]:1902 "EHLO mail.rtr.ca" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755921AbYEBCk0 (ORCPT ); Thu, 1 May 2008 22:40:26 -0400 In-Reply-To: <20080501195823.79786815@core> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Paolo Cc: Alan Cox , jgarzik@pobox.com, linux-ide@vger.kernel.org Alan Cox wrote: > On Thu, 1 May 2008 20:58:37 +0200 > Paolo wrote: > >> Hello! >> >> It seems that me and many others are experiencing poor sata performances >> in the last months, probably with 2.6.20+ kernels. >> >> an example of these problems is here: >> >> http://ubuntuforums.org/showthread.php?p=4845777 >> >> unfortunately it seems that sata subsystem on linux is little to none >> tweakable > > It doesn't normally need tweaking, and the hdparm numbers show that it is > working correctly in the example you give. There may be problems higher > up the stack but the ATA layer appears to be doing just fine. .. Yup. The drive seems to be doing fine, according to the info you've provided. But the "-T" (big T) number looks *very* slow: 220.02 MB/sec That number has nothing to do with the drive. Rather, it's an indication of how fast the CPU and memory are together, and that's about 1/20 to 1/10 of a modern system. Note that it's not a useful number in any absolute sense, but only when compared with a "-T" value from some other system or Linux distro. Strange.