From: Stanislaw Gruszka <sgruszka@redhat.com>
To: Jean-Pierre TOSONI <jp.tosoni@acksys.fr>,
"John W. Linville" <linville@tuxdriver.com>
Cc: linux-wireless@vger.kernel.org
Subject: Re: mac80211/scan.c: scanning ceases after a double rescheduling
Date: Fri, 1 Jun 2012 11:12:51 +0200 [thread overview]
Message-ID: <20120601091250.GA30893@redhat.com> (raw)
In-Reply-To: <773DB8A82AB6A046AE0195C68612A319012CA1B7@sbs2003.acksys.local>
On Thu, May 31, 2012 at 06:39:51PM +0200, Jean-Pierre TOSONI wrote:
> Hello all,
>
> I am using compat-wireless-2012-04-17 in a OpenWrt box, client mode,
> kernel 2.6.33.
> I arrange to run an active background scan every second under some load
> (iperf UDP 40 Mbps).
>
> I observed that the probes can be delayed due to two events flagged as
> IEEE80211_STA_BEACON_POLL and IEEE80211_STA_CONNECTION_POLL. They make
> __ieee80211_start_scan() return early in mac80211/scan.c when it is
> called by ieee80211_request_scan().
>
> When one of these events occurs, the function returns and
> __ieee80211_start_scan() is reentered later, called this time by
> ieee80211_scan_work(). Now if it finds the other event flagged, it
> returns early but will never be called again. As a result later scan
> requests result in a "EBUSY" error.
>
> I made a workaround by always returning an error, but I am pretty sure
> this is a wrong solution. I wonder if scan_work should not be
> rescheduled in this case.
I posted fix here:
http://marc.info/?l=linux-wireless&m=133717501822158&w=2
It is needed for 3.5 and it is still not appied in wireless-testing ...
Stanislaw
prev parent reply other threads:[~2012-06-01 9:13 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-05-31 16:39 mac80211/scan.c: scanning ceases after a double rescheduling Jean-Pierre TOSONI
2012-06-01 9:12 ` Stanislaw Gruszka [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20120601091250.GA30893@redhat.com \
--to=sgruszka@redhat.com \
--cc=jp.tosoni@acksys.fr \
--cc=linux-wireless@vger.kernel.org \
--cc=linville@tuxdriver.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).