From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeremy Sanders Subject: Re: Sil 3124: "softreset failed (SRST command error)" Date: Thu, 14 Oct 2010 14:19:07 +0100 Message-ID: References: <4CB4665F.5030207@kernel.org> <4CB481E0.4040705@seoss.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7Bit Return-path: Received: from lo.gmane.org ([80.91.229.12]:46164 "EHLO lo.gmane.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754157Ab0JNNTT (ORCPT ); Thu, 14 Oct 2010 09:19:19 -0400 Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1P6NiG-0002id-94 for linux-ide@vger.kernel.org; Thu, 14 Oct 2010 15:19:16 +0200 Received: from xpc1.ast.cam.ac.uk ([131.111.68.219]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Thu, 14 Oct 2010 15:19:16 +0200 Received: from jeremy by xpc1.ast.cam.ac.uk with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Thu, 14 Oct 2010 15:19:16 +0200 Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: linux-ide@vger.kernel.org Jeremy Sanders wrote: > Tim Small wrote: > >> Bad PCI express to PCI-X bridge chip maybe, or maybe the Linux bug lies >> there? I have seen cards with 2 3132s on instead (and a PCI express >> bridge), perhaps it'd be worth trying one of those? > > It looks like the problem was an incompatibility with the Intel > motherboard. > > It seems to work fine in an old foxconn Athlon 64 motherboard (in the > PCIex graphics slot when using a PCI card for video). Unfortunately it doesn't look like it works that well :-( Read speeds to a standard drive are great - 70 MB/s. Write speeds are awful - they range on one system at 17 MB/s down to a more modern system at 1.7 MB/s! There is no CPU load. I assume this is a problem with the PCIe bridge. There are no errors or warnings given. I couldn't get MSI interrupts to work on this system. Time to throw the card away I think! Jeremy