From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: RE: [PATCH]: Flexible timeout infrastructure Date: 16 Jun 2004 12:38:20 -0500 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <1087407501.2067.37.camel@mulgrave> References: <3356669BBE90C448AD4645C843E2BF28034F94F3@xbl.ma.emulex.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from stat1.steeleye.com ([65.114.3.130]:12717 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S264288AbUFPRie (ORCPT ); Wed, 16 Jun 2004 13:38:34 -0400 In-Reply-To: <3356669BBE90C448AD4645C843E2BF28034F94F3@xbl.ma.emulex.com> List-Id: linux-scsi@vger.kernel.org To: "Smart, James" Cc: Luben Tuikov , Mike Anderson , SCSI Mailing List On Wed, 2004-06-16 at 12:33, Smart, James wrote: > I'll try to restate. The patch proposed to allow only 1 restart of the timer > for a command. My comment was - why not 2, or 3? I would think - the number > of restarts needed is a function of how long the timeout value is vs when > the LLDD event occurs and what the duration of the LLDD event is. I can see > capping the number of timer restarts, but am not sure 1 is the best choice. It does. But the number of allowed restarts is linked to the command retries (actually allowed retries + 1). James