From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ondrej Zary Subject: Re: sata_via: write errors on PATA drive connected to VT6421 Date: Tue, 21 Aug 2007 19:09:30 +0200 Message-ID: <200708211909.30652.linux@rainbow-software.org> References: <200708202242.07290.linux@rainbow-software.org> <20070821132731.6bf10b50@the-village.bc.nu> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: Received: from mail.atlantis.sk ([80.94.52.35]:37631 "EHLO mail.atlantis.sk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754576AbXHURJU (ORCPT ); Tue, 21 Aug 2007 13:09:20 -0400 In-Reply-To: <20070821132731.6bf10b50@the-village.bc.nu> Content-Disposition: inline Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Alan Cox Cc: jgarzik@pobox.com, linux-ide@vger.kernel.org, linux-kernel@vger.kernel.org On Tuesday 21 August 2007 14:27:31 Alan Cox wrote: > > The transfer mode decreases from UDMA/100 to UDMA/66 and then UDMA/33 but > > it doesn't help at all. The drive itself is fine, the problems also > > appear with > > Does it settle at UDMA33 or keep going down lower ? It does not go below UDMA33. dd if=/dev/zero of=some.file bs=131072 count=1024 takes 526 seconds (amazing transfer rate 255KB/s) with system being unresponsive most of the time. Commenting out the UDMA support in sata_via.c (.udma_mask and .set_dmamode for PATA port) causes the driver to run in PIO4 mode which seems to work without errors but slowly (3MB/s with 100% CPU load). -- Ondrej Zary