From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934042AbZHWUC3 (ORCPT ); Sun, 23 Aug 2009 16:02:29 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S933983AbZHWUC2 (ORCPT ); Sun, 23 Aug 2009 16:02:28 -0400 Received: from khc.piap.pl ([195.187.100.11]:59592 "EHLO khc.piap.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933948AbZHWUC2 (ORCPT ); Sun, 23 Aug 2009 16:02:28 -0400 To: Aviv Greenberg Cc: Walt Holman , David Miller , linux-kernel@vger.kernel.org, netdev@vger.kernel.org Subject: Re: Strange network timeouts w/ 2.6.30.5 References: <985009134.71250769263099.JavaMail.root@mail.holmansrus.com> <20090823191808.GA29972@linux-gec2.site> From: Krzysztof Halasa Date: Sun, 23 Aug 2009 22:02:27 +0200 In-Reply-To: <20090823191808.GA29972@linux-gec2.site> (Aviv Greenberg's message of "Sun\, 23 Aug 2009 22\:18\:08 +0300") Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Aviv Greenberg writes: > Why do you use PCI_DMA_BIDIRECTIONAL? From what i get, the device writes and the > cpu reads - so the direction should be PCI_DMA_FROMDEVICE. Maybe just maybe the > bi-directionality bites you in the swiotlb case. Yes, it's just been confirmed that FROMDEVICE at this point happens to work. -- Krzysztof Halasa