From: Jeff Garzik Subject: Re: [GIT PULL] Ext3 latency fixes Date: Sun, 05 Apr 2009 16:57:21 -0400 Message-ID: <49D91B31.90300@garzik.org> References: <20090404135719.GA9812@mit.edu> <20090404151649.GE5178@kernel.dk> <20090404173412.GF5178@kernel.dk> <20090404180108.GH5178@kernel.dk> <20090404232222.GA7480@mit.edu> <20090404163349.20df1208@infradead.org> <20090405001005.GA7553@mit.edu> <49D8E71F.6000703@rtr.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: Linus Torvalds , Theodore Tso , Arjan van de Ven , Jens Axboe , Linux Kernel Developers List , Ext4 Developers List To: Mark Lord Return-path: Received: from srv5.dvmed.net ([207.36.208.214]:42346 "EHLO mail.dvmed.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751972AbZDEU6G (ORCPT ); Sun, 5 Apr 2009 16:58:06 -0400 In-Reply-To: <49D8E71F.6000703@rtr.ca> Sender: linux-ext4-owner@vger.kernel.org List-ID: Mark Lord wrote: > Linus Torvalds wrote: >> >> Non-rotational media would tend to want the unplugging, but the block >> layer already does that (ie in __make_request() we already do: >> >> if (unplug || blk_queue_nonrot(q)) >> __generic_unplug_device(q); >> >> so non-rotational devices get unplugged whether the request was a >> unplugging request or not). > .. > > Mmm.. except that Jeff's SSD doesn't appear to identify itself > as a non-rotating media. I suppose there's a way to "quirk" that > somewhere, though. We set it in libata-scsi.c:ata_scsi_dev_config() based on ata_id_is_ssd() That hueristic probably assumes Intel SSDs or something :/ Jeff