From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=41287 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Oqkpf-0004qO-Lm for qemu-devel@nongnu.org; Wed, 01 Sep 2010 06:46:20 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1Oqkpe-0004lk-KC for qemu-devel@nongnu.org; Wed, 01 Sep 2010 06:46:19 -0400 Received: from mx1.redhat.com ([209.132.183.28]:57410) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1Oqkpe-0004lZ-7N for qemu-devel@nongnu.org; Wed, 01 Sep 2010 06:46:18 -0400 Date: Wed, 1 Sep 2010 13:40:25 +0300 From: "Michael S. Tsirkin" Message-ID: <20100901104025.GC7513@redhat.com> References: <20100827223659.2696.3589.stgit@s20.home> <20100831202705.GE31570@redhat.com> <4C7D751B.8070705@codemonkey.ws> <1283293567.5826.147.camel@x201> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1283293567.5826.147.camel@x201> Subject: [Qemu-devel] Re: [PATCH 0/5] virtio-net: More configurability and bh handling for tx List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alex Williamson Cc: jes.sorensen@redhat.com, kraxel@redhat.com, qemu-devel@nongnu.org, kvm@vger.kernel.org On Tue, Aug 31, 2010 at 04:26:07PM -0600, Alex Williamson wrote: > On Tue, 2010-08-31 at 16:33 -0500, Anthony Liguori wrote: > > On 08/31/2010 03:27 PM, Michael S. Tsirkin wrote: > > > On Fri, Aug 27, 2010 at 04:36:59PM -0600, Alex Williamson wrote: > > > > > >> Add the ability to configure the tx_timer timeout and add a bottom > > >> half tx handler that typically shows a nice perf boost over the > > >> time based approach. See last patch for perf details. Make this > > >> the new default when the iothread is enabled. Thanks, > > >> > > >> Alex > > >> > > > As a further thought, maybe it would help to have a > > > separate namespace for unsupported, developer-only parameters. > > > Thoughts? > > > > > > > We already have an undocumented one, just prefix with 'x-' to indicate > > that it's experimental. > > Any objection then to x-txburst and x-txtimer then? Easy enough to > rename. Thanks, > > Alex No objection. Let's put the 256 constant in a header, with a comment explaining where it came from? -- MST