From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: [PATCH] tmscsim: implement change_queue_depth Date: Mon, 27 Dec 2004 12:31:38 -0600 Message-ID: <1104172298.7447.10.camel@mulgrave> References: <1103826862.5290.39.camel@mulgrave> <1104076166.5268.22.camel@mulgrave> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from stat16.steeleye.com ([209.192.50.48]:3045 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S261940AbUL0Sbr (ORCPT ); Mon, 27 Dec 2004 13:31:47 -0500 In-Reply-To: Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Guennadi Liakhovetski Cc: SCSI Mailing List On Sun, 2004-12-26 at 22:49 +0100, Guennadi Liakhovetski wrote: > Ok. Would it make sense / be accepted to change this sitation by either 1) > adding a new function __scsi_adjust_queue_depth() which would pass the > return code of the block-layer, or 2) modifying the existing function to > return an error? 1) has disadvantages, that it's a) confusing, b) even > more confusing, because that other function also has to be exported. 2) > has the disadvantage of changing an exported API in the "stable" series... > I would go for 2) especially with the current extended understanding of > the "stable" kernel series. I could roll a patch for either of these 2 > possibilities, if the latter is chosen I could try to fix the current > users to check the return code. To be honest, unless you can find a driver that cares about this failure, I don't think so. James