HCI crash after tearing down ppp connection

Previous thread: by FedEx Courier Servic on Tuesday, January 27, 2009 - 11:59 am. (1 message)

Next thread: Re: [Bugme-new] [Bug 12515] New: possible circular locking #0: (sk_lock-AF_PACKET){--..}, at: [<c1279838>] sock_setsockopt+0x12b/0x4a4 by Andrew Morton on Tuesday, January 27, 2009 - 2:43 pm. (9 messages)
From: Johannes Berg
Date: Tuesday, January 27, 2009 - 12:01 pm

I use /dev/rfcomm0 to connect to my cell phone, and then to the internet
with ppp. Occasionally (two out of the three times I used it today)
after I poff the connection the machine locks up. The second time I got
a log, but because my display resolution is bad it mostly scrolled off,
so I only got to see that hci was involved somehow :/

Anyone aware of any such issue? This is on
2.6.29-rc2-wl-11979-gc32422d-dirty (-rc2 + wireless bits).

johannes
From: Marcel Holtmann
Date: Tuesday, January 27, 2009 - 11:35 pm

I was assuming we fixed these, but we might ended with a new one since
the TTY is changing a little bit and the Bluetooth TTYs are special
since they are pure virtual TTYs. Can you reproduce this crash every
time or is it happening only now and then?

During the last weeks, I am chasing an USB issue with the USB anchors
and how we use it. You might be hitting the one.

Regards

Marcel


--

From: Johannes Berg
Date: Wednesday, January 28, 2009 - 2:53 am

Yeah, I wish I could capture more information here. Can't seem to figure
out how to get more VT resolution. I'll try again. :/

johannes
Previous thread: by FedEx Courier Servic on Tuesday, January 27, 2009 - 11:59 am. (1 message)

Next thread: Re: [Bugme-new] [Bug 12515] New: possible circular locking #0: (sk_lock-AF_PACKET){--..}, at: [<c1279838>] sock_setsockopt+0x12b/0x4a4 by Andrew Morton on Tuesday, January 27, 2009 - 2:43 pm. (9 messages)