Date: Tue, 15 Jan 2008 15:25:30 +0100 From: Laurent Frigault <lfrigault@agneau.org> To: Sepherosa Ziehau <sepherosa@gmail.com> Cc: freebsd-net@freebsd.org, remko@freebsd.org Subject: Re: kern/119361: [bge] bge(4) transmit performance problem Message-ID: <20080115142530.GA25951@obelix.bergerie.agneau.org> In-Reply-To: <ea7b9c170801141656j5e8371fcqf1f2e5fc895df638@mail.gmail.com> References: <200801071728.m07HSsFl030966@freefall.freebsd.org> <ea7b9c170801110602q4abb010ayda7947205b47fe4e@mail.gmail.com> <20080114160756.GA7716@obelix.bergerie.agneau.org> <ea7b9c170801141656j5e8371fcqf1f2e5fc895df638@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jan 15, 2008 at 08:56:36AM +0800, Sepherosa Ziehau wrote: > > > > http://www.freebsd.org/cgi/query-pr.cgi?pr=119361 > > > > > > Have you tested polling(4)? If polling could give you better TX > > > performance, you should consider raise: > > > in bge_attach() > > > sc->bge_tx_max_coal_bds = 10; > > > to 64 or more > > > > I tested with polling and did not see any changes for the transmit > > performance problem. > > > > I just re-test with polling AND bge_tx_max_coal_bds raised to 64 => no > > changes. > > Please test following patch: > http://people.freebsd.org/~sephe/if_bge.c.6.diff I apply your patch and re-run the test (without and with polling) => no changes. Best Regards, -- Laurent Frigault | <url:http://www.agneau.org/>; UNIX _IS_ user friendly. It's just selective about who its friends are.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080115142530.GA25951>