From mboxrd@z Thu Jan 1 00:00:00 1970 From: Herbert Xu Subject: Re: [PATCH 2/4] virtio_net: return NETDEV_TX_BUSY instead of queueing an extra skb. Date: Tue, 23 Jun 2009 10:54:59 +0800 Message-ID: <20090623025459.GA5991@gondor.apana.org.au> References: <200905292346.04815.rusty@rustcorp.com.au> <200906181647.51045.rusty@rustcorp.com.au> <20090618073422.GA4087@gondor.apana.org.au> <200906191307.20360.rusty@rustcorp.com.au> <20090619043613.GA15163@gondor.apana.org.au> <20090622073417.GA21698@gondor.apana.org.au> <20090622182529.GA4990@xw6200.broadcom.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Krishna Kumar2 , David Miller , "netdev@vger.kernel.org" , Rusty Russell , "virtualization@lists.linux-foundation.org" To: Matt Carlson Return-path: Received: from rhun.apana.org.au ([64.62.148.172]:58285 "EHLO arnor.apana.org.au" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751576AbZFWCzG (ORCPT ); Mon, 22 Jun 2009 22:55:06 -0400 Content-Disposition: inline In-Reply-To: <20090622182529.GA4990@xw6200.broadcom.net> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, Jun 22, 2009 at 11:25:29AM -0700, Matt Carlson wrote: > > As was said elsewhere, from the driver writer's perspective every packet > that has already been submitted (queued) to the hardware cannot be > preempted. Slightly extending that logic doesn't seem like that much of > a problem, especially if it saves the troublesome requeuing logic higher up. I think this is pretty much the same as returning TX_BUSY :) Do you have access to one of these buggy cards? Can you run some tests to see what the difference between TSO and GSO is on them? Thanks, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt