Is there a bug in 2.6.29-rc8 support of Ad-Hoc mode?

Is there a bug in 2.6.29-rc8 support of Ad-Hoc mode?

[Charles Gordon]

I am trying to get a driver I'm developing to work with Ad-Hoc mode on
Linux 2.6.29-rc8.  I am running into a problem.  I have traced it down
to the routine ieee80211_scan_work.  I use the following commands to
start Ad-Hoc mode:

ifconfig wlan0 down
iwconfig wlan0 mode Ad-Hoc essid "AdHocID"
ifconfig wlan0 up

What happens is that the iwconfig command starts a scan for other PCs
using the AdHocID before the second ifconfig command is executed
(while the interface is still down).  When ieee80211_scan_work is
executed, it bails out at this statement:

	/*
	 * Avoid re-scheduling when the sdata is going away.
	 */
	if (!netif_running(sdata->dev))
		return;

This aborts the scan, but leaves local->sw_scanning set.  This
prevents future scans from running.  When ieee80211_start_scan
executes it bails out at this statement:

	if (local->sw_scanning || local->hw_scanning) {
		if (local->scan_sdata == scan_sdata)
			return 0;
		return -EBUSY;
	}

Can someone either confirm that this is a known problem with
2.6.28-rc8, or tell me what my driver could be doing to cause this
problem?

Thanks for any help you can provide.

Is there a bug in 2.6.29-rc8 support of Ad-Hoc mode?

[Gordon, Charles]

I am trying to get a driver I'm developing to work with Ad-Hoc mode on
Linux 2.6.29-rc8.  I am running into a problem.  I have traced it down
to the routine ieee80211_scan_work.  I use the following commands to
start Ad-Hoc mode:

ifconfig wlan0 down
iwconfig wlan0 mode Ad-Hoc essid "AdHocID"
ifconfig wlan0 up

What happens is that the iwconfig command starts a scan for other PCs
using the AdHocID before the second ifconfig command is executed
(while the interface is still down).  When ieee80211_scan_work is
executed, it bails out at this statement:

       /*
        * Avoid re-scheduling when the sdata is going away.
        */
       if (!netif_running(sdata->dev))
               return;

This aborts the scan, but leaves local->sw_scanning set.  This
prevents future scans from running.  When ieee80211_start_scan
executes it bails out at this statement:

       if (local->sw_scanning || local->hw_scanning) {
               if (local->scan_sdata == scan_sdata)
                       return 0;
               return -EBUSY;
       }

Can someone either confirm that this is a known problem with
2.6.28-rc8, or tell me what my driver could be doing to cause this
problem?

Thanks for any help you can provide.

Re: Is there a bug in 2.6.29-rc8 support of Ad-Hoc mode?

[John W. Linville]

On Wed, May 26, 2010 at 06:57:31AM -0500, Gordon, Charles wrote:
> I am trying to get a driver I'm developing to work with Ad-Hoc mode on
> Linux 2.6.29-rc8.  I am running into a problem.  I have traced it down
> to the routine ieee80211_scan_work.  I use the following commands to
> start Ad-Hoc mode:
> 
> ifconfig wlan0 down
> iwconfig wlan0 mode Ad-Hoc essid "AdHocID"
> ifconfig wlan0 up

FWIW, this should probably look more like the following:

	ifconfig wlan0 down
	iwconfig wlan0 mode Ad-Hoc
	ifconfig wlan0 up
	iwconfig wlan0 essid "AdHocID"

If that doesn't work, you may be experiencing some bug.  But...

> Can someone either confirm that this is a known problem with
> 2.6.28-rc8, or tell me what my driver could be doing to cause this
> problem?

This kernel is ancient, even if you meant 2.6.29-rc8 like you
said originally.  And it is not even an actual release version
(e.g. 2.6.29).  I could be wrong, but I doubt there will be much
interest among the people here in working on the internals of such
an old kernel.  Is there some reason you can't try something newer?
2.6.34 was just released...

Good luck!

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.

RE: Is there a bug in 2.6.29-rc8 support of Ad-Hoc mode?

[Gordon, Charles]

Hi John,

Switching to your sequence of commands:

>	ifconfig wlan0 down
>	iwconfig wlan0 mode Ad-Hoc
>	ifconfig wlan0 up
>	iwconfig wlan0 essid "AdHocID"

solves the problem.  

Thanks for the help!



-----Original Message-----
From: John W. Linville [mailto:linville@tuxdriver.com] 
Sent: Wednesday, May 26, 2010 9:28 AM
To: Gordon, Charles
Cc: linux-wireless@vger.kernel.org
Subject: Re: Is there a bug in 2.6.29-rc8 support of Ad-Hoc mode?

On Wed, May 26, 2010 at 06:57:31AM -0500, Gordon, Charles wrote:
> I am trying to get a driver I'm developing to work with Ad-Hoc mode on
> Linux 2.6.29-rc8.  I am running into a problem.  I have traced it down
> to the routine ieee80211_scan_work.  I use the following commands to
> start Ad-Hoc mode:
> 
> ifconfig wlan0 down
> iwconfig wlan0 mode Ad-Hoc essid "AdHocID"
> ifconfig wlan0 up

FWIW, this should probably look more like the following:

	ifconfig wlan0 down
	iwconfig wlan0 mode Ad-Hoc
	ifconfig wlan0 up
	iwconfig wlan0 essid "AdHocID"

If that doesn't work, you may be experiencing some bug.  But...

> Can someone either confirm that this is a known problem with
> 2.6.28-rc8, or tell me what my driver could be doing to cause this
> problem?

This kernel is ancient, even if you meant 2.6.29-rc8 like you
said originally.  And it is not even an actual release version
(e.g. 2.6.29).  I could be wrong, but I doubt there will be much
interest among the people here in working on the internals of such
an old kernel.  Is there some reason you can't try something newer?
2.6.34 was just released...

Good luck!

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.