From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: "Fix ATAPI transfer lengths" causes CD writing regression Date: Mon, 05 Nov 2007 09:05:48 +0900 Message-ID: <472E5E5C.8040308@gmail.com> References: <47274A5F.6070409@gentoo.org> <20071030153417.59b9182c@the-village.bc.nu> <47276DCA.1000808@gentoo.org> <20071030190153.373c9347@the-village.bc.nu> <47278439.4030801@gentoo.org> <20071031114958.210bd7cc@the-village.bc.nu> <20071031115754.GK5059@kernel.dk> <4729A0DF.20800@garzik.org> <20071101105335.1f20bab3@the-village.bc.nu> <4729B3EA.6040707@garzik.org> <20071101141501.3746cec2@the-village.bc.nu> <4729F1BB.20306@gentoo.org> <4729F8F1.4040103@gmail.com> <472B946F.4030004@gentoo.org> <472BCC18.7070503@gmail.com> <472CD3F3.7050701@gentoo.org> <472D0D4A.70404@gmail.com> <472D445B.1080005@tw.ibm.com> <20071104234205.7627d007@the-village.bc.nu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from wa-out-1112.google.com ([209.85.146.176]:49011 "EHLO wa-out-1112.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752953AbXKEAF7 (ORCPT ); Sun, 4 Nov 2007 19:05:59 -0500 Received: by wa-out-1112.google.com with SMTP id v27so1772455wah for ; Sun, 04 Nov 2007 16:05:57 -0800 (PST) In-Reply-To: <20071104234205.7627d007@the-village.bc.nu> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Alan Cox Cc: albertl@mail.com, albertcc@tw.ibm.com, Daniel Drake , Jeff Garzik , Jens Axboe , linux list , linux-ide@vger.kernel.org Alan Cox wrote: >> Maybe we could set a limit here. If the ATAPI device keeps DRQ=1 and >> exceeds the limit, we consider it as HSM violation and have EH handle it. > > On a DMA transfer its basically out of our control (and a PIO drain will > lock some controllers solid until power cycle), Do such controllers lock up on PIO draining after PIO transfers too? Can you tell which are those controllers? > but on PIO we know we > never set a chunk size over 64K, so if we exceed 64K its time to apply a > larger hammer Draining is related to the amount of data the drive responds not to the chunk size. I agree 64k should be enough for most cases but I think there can be corner cases where this doesn't hold. Thanks. -- tejun