Re: please (re) test if_ath in -HEAD

Woo. That's great news. On a non-11n card too. Excellent!



adrian

On 9 March 2011 17:12, Urankar Mikael <mikael.urankar_at_ujf-grenoble.fr>wrote:

> Yes !
>
> On Wed 09 March 2011 at 04:58:50PM +0800, Adrian Chadd wrote:
> > So you're saying -head is better, right?
> >
> > On 9 March 2011 15:35, Urankar Mikael <mikael.urankar_at_ujf-grenoble.fr
> >wrote:
> >
> > > Here are some dumb benchmarks through ftp :
> > >
> > > with 8.2R RX rate is 2400kBps and TX rate is 810KBps
> > > on a 9-CURRENT RX rate is 2700kBps and TX rate is 2700kBps
> > >
> > > More info on my system :
> > > ath0_at_pci0:4:0:0:        class=0x020000 card=0x5b001385 chip=0x0013168c
> > > rev=0x01 hdr=0x00
> > >    vendor     = 'Atheros Communications Inc.'
> > >     device     = '802.11a/b/g Wireless Adapter (AR2312)'
> > >
> > > ath0: <Atheros 5212> mem 0x88000000-0x8800ffff irq 19 at device 0.0 on
> > > cardbus0
> > > ath0: AR2413 mac 7.9 RF2413 phy 4.5
> > >
> > > FreeBSD iut-mir-o.ujf-grenoble.fr 9.0-CURRENT FreeBSD 9.0-CURRENT #39
> > > r219368M: Mon Mar  7 12:15:50 CET 2011
> > >
> > > $FreeBSD: head/sys/dev/ath/ah_osdep.c 219315 2011-03-05 21:20:18Z
> adrian $
> > > $FreeBSD: head/sys/dev/ath/if_ath.c 219185 2011-03-02 17:19:54Z adrian
> $
> > > $FreeBSD: head/sys/dev/ath/if_ath_debug.c 218058 2011-01-29 05:08:21Z
> > > adrian $
> > > $FreeBSD: head/sys/dev/ath/if_ath_keycache.c 219185 2011-03-02
> 17:19:54Z
> > > adrian $
> > > $FreeBSD: head/sys/dev/ath/if_ath_pci.c 192147 2009-05-15 17:02:11Z imp
> $
> > > $FreeBSD: head/sys/dev/ath/if_ath_sysctl.c 219180 2011-03-02 16:03:19Z
> > > adrian $
> > > $FreeBSD: head/sys/dev/ath/if_ath_tx.c 218932 2011-02-22 00:37:53Z
> adrian $
> > > $FreeBSD: head/sys/dev/ath/if_ath_tx_ht.c 219214 2011-03-03 03:02:06Z
> > > adrian $
> > >
> > >
> > > On Sun 06 March 2011 at 05:02:38AM +0800, Adrian Chadd wrote:
> > > > What was the previous performance? And which chipsets?
> > > >
> > > >
> > > >
> > > > Adrian
> > > >
> > > > On 6 March 2011 00:30, Urankar Mikael <
> mikael.urankar_at_ujf-grenoble.fr
> > > >wrote:
> > > >
> > > > > On Sat 05 March 2011 at 03:27:48AM -0800, Adrian Chadd wrote:
> > > > > > Absolutely! Let me know if I've broken anything!
> > > > > >
> > > > > > Adrian
> > > > > >
> > > > > > On 5 March 2011 00:45, Urankar Mikael <
> > > mikael.urankar_at_ujf-grenoble.fr
> > > > > >wrote:
> > > > > >
> > > > > > > On Thu 03 March 2011 at 02:31:05PM -0800, Adrian Chadd wrote:
> > > > > > > > Hi all,
> > > > > > > >
> > > > > > > > For those of you who are testing out my if_ath changes, I'd
> > > really
> > > > > > > > appreciate it if you'd update to -HEAD and re-test.
> > > > > > > >
> > > > > > > > I've done a variety of changes to the radio setup and
> found/fixed
> > > a
> > > > > few
> > > > > > > bugs
> > > > > > > > in the TX path. It's quite possible these have introduced
> > > > > regressions.
> > > > > > > I'd
> > > > > > > > like to make sure that I haven't broken legacy (11abg)
> support in
> > > > > > > > weird/wonderful ways. I'd also like to make sure that I
> haven't
> > > > > > > > broken/changed the behaviour or performance of the NICs in
> any
> > > way.
> > > > > > > >
> > > > > > > > Please give things a good thrashing and let me know the
> results.
> > > > > > > >
> > > > > > > > I'm still working towards debugging and enabling basic 11n
> > > support,
> > > > > but I
> > > > > > > > need to first make sure that I haven't broken legacy
> operation in
> > > any
> > > > > > > way.
> > > > > > > >
> > > > > > >
> > > > > > > Are you interested by tests on an old atheros card (Atheros
> 5212,
> > > > > AR2413
> > > > > > > mac 7.9 RF2413 phy 4.5) ?
> > > > > > >
> > > > >
> > > > > Everything seems to work fine, performances are as good as before.
> > > > > I'm only doing basic stuff with this card (mostly web surfing), I'm
> > > > > connected to an AP with WPA encryption. If you want me to test
> other
> > > > > encryption or setup an AP with this card or whatever, feel free to
> ask.
> > > > >
> > > > > And thanks for you work !
>