From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: [PATCH resend] bsg: Add support for submitting requests at tail of queue Date: Tue, 24 Mar 2009 13:26:09 +0100 Message-ID: <20090324122608.GR27476@kernel.dk> References: <49C89CBC.7050306@panasas.com> <20090324210440Z.fujita.tomonori@lab.ntt.co.jp> <20090324121438.GQ27476@kernel.dk> <20090324212248A.fujita.tomonori@lab.ntt.co.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from brick.kernel.dk ([93.163.65.50]:52972 "EHLO kernel.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757274AbZCXM0L (ORCPT ); Tue, 24 Mar 2009 08:26:11 -0400 Content-Disposition: inline In-Reply-To: <20090324212248A.fujita.tomonori@lab.ntt.co.jp> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: FUJITA Tomonori Cc: bharrosh@panasas.com, linux-scsi@vger.kernel.org, osd-dev@open-osd.org, dgilbert@interlog.com On Tue, Mar 24 2009, FUJITA Tomonori wrote: > On Tue, 24 Mar 2009 13:14:38 +0100 > Jens Axboe wrote: > > > On Tue, Mar 24 2009, FUJITA Tomonori wrote: > > > On Tue, 24 Mar 2009 10:41:32 +0200 > > > Boaz Harrosh wrote: > > > > > > > Boaz Harrosh wrote: > > > > > Hi Tomo Jens > > > > > > > > > > Tomo you never ack-by on this patch. I absolutely needs this for the > > > > > user-mode API of osd-initiator. Which is needed with up-coming exofs > > > > > utilities. > > > > > > > > > > What do you want to do? > > > > > > > > > > Thanks > > > > > > > > > > > > > Hi Jens. > > > > > > > > I absolutely need this patch for 2.6.30 merge window. It is totally > > > > un-dangerous since defaults are left unchanged. > > > > > > The question is we really need this feature or not. Though I guess > > > that we need to address this starvation issue. > > > > I'd argue that tail insertion should be the most used way to queue > > commands, head insertion should only be for the cases where you want > > immediate access (like error handling). > > As I wrote, I think that we need to address this starvation issue. > > But nobody has complained about this (with sg and bsg) for years > because we have not seen applications sending scsi commands that are > many enough to make a queue full. So it's worth knowing what unknown > applications do for what, I think. Not sure it exists, just saying that if I were to write an app that uses bsg for doing "normal" IO, I would certainly use tail insertion. -- Jens Axboe