Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 19 Feb 2012 18:47:30 -0800
From:      Adrian Chadd <adrian@freebsd.org>
To:        freebsd-wireless@freebsd.org
Subject:   Re: kern/165306: [ath] race conditions between scanning and beacon timeout programming
Message-ID:  <CAJ-Vmon%2BC=ahWjfUeCcKT3UqJzbpqBdSX=c5%2BbU0ocnqxQ=e0g@mail.gmail.com>
In-Reply-To: <201202200243.q1K2h7Zi081981@freefall.freebsd.org>
References:  <201202200243.q1K2h7Zi081981@freefall.freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On 19 February 2012 18:43,  <adrian@freebsd.org> wrote:
> Synopsis: [ath] race conditions between scanning and beacon timeout programming

This bug is ridiculously subtle and took me a while to figure out what
was going on. Grr.

I'm very tempted to just always flip on the beacon interrupt
programming by writing the last known TSF if there was one, then
updating it the next time we see a beacon. That way we don't end up in
this situation where we never register that the AP has disappeared.

Comments?



Adrian



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-Vmon%2BC=ahWjfUeCcKT3UqJzbpqBdSX=c5%2BbU0ocnqxQ=e0g>