Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 7 Aug 2012 09:29:17 -0700
From:      Adrian Chadd <adrian@freebsd.org>
To:        Vincent Hoffman <vince@unsane.co.uk>
Cc:        freebsd-wireless@freebsd.org
Subject:   Re: (ANother) stall fixed, please update to HEAD
Message-ID:  <CAJ-Vmo=JOZGFqd%2BugJ1z%2BXcjDshWsC_%2Bfr5WGSHsE-ahc9DAxQ@mail.gmail.com>
In-Reply-To: <5021416D.6050002@unsane.co.uk>
References:  <CAJ-VmokU89Xo0UFCJynuhJ6tm-A6NaKS7xLc0964FtCfKDrtBg@mail.gmail.com> <5021416D.6050002@unsane.co.uk>

next in thread | previous in thread | raw e-mail | index | archive | help
On 7 August 2012 09:25, Vincent Hoffman <vince@unsane.co.uk> wrote:

> Aug  7 11:06:04 ostracod kernel: FreeBSD 10.0-CURRENT #12 r239123: Tue
> Aug  7 08:57:47 BST 2012
>
> Then while ago got these
> Aug  7 13:58:35 ostracod kernel: ath0: ath_rx_pkt: rs_antenna > 7 (9560157)
> Aug  7 13:58:35 ostracod kernel: R[ 0] (DS.V:0xffffff80e7e868a0
> DS.P:0x2b168a0) L:02b16900 D:0e66e000 *
> Aug  7 13:58:35 ostracod kernel: 3879e2f0 00000800 00030902 0030400c
> Aug  7 13:58:35 ostracod kernel: 07232400 91e05dfb b95b48f5 0d4b3360
> baa0f8e1 00a00000 20005e4b
> Aug  7 13:58:35 ostracod kernel: ath0: ath_rx_pkt: rs_antenna > 7 (1709845)
> Aug  7 13:58:35 ostracod kernel: R[ 0] (DS.V:0xffffff80e7e83180
> DS.P:0x2b13180) L:02b131e0 D:2c220800 *
> Aug  7 13:58:35 ostracod kernel: 2332001c 00000800 9b8cc2a4 00000010
> Aug  7 13:58:35 ostracod kernel: 2f808080 1a17151c 00000000 00000000
> 00020902 002c4288 07232401

I do occasionally see these. There's nothing about the RX descriptors
that I've seen that indicate what's going on. Sometimes I see a
corrupted RX descriptor. I've no idea why at the present.

> Which seemed to break my connection (could no longer ping or ssh in.)
> Sadly since this mail goes via that box I had to reboot it, any
> suggestions on debugging I can give you if it happens again?

When it happens, check the output of 'sysctl dev.ath.X.txagg=1' (check
dmesg) where X is the physical NIC. In your instance it should be
'dev.ath.0.txagg' for ath0.

There are still some issues with TX buffer exhaustion and lockups, as
well as some as of yet unknown hangs due to BAR TX handling.


Adrian



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-Vmo=JOZGFqd%2BugJ1z%2BXcjDshWsC_%2Bfr5WGSHsE-ahc9DAxQ>