Return-path: Received: from extu-mxob-1.symantec.com ([216.10.194.28]:45263 "EHLO extu-mxob-1.symantec.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753594AbZAIOLI (ORCPT ); Fri, 9 Jan 2009 09:11:08 -0500 Received: from [172.20.26.116]([172.20.26.116]) (2475 bytes) by megami.veritas.com via sendmail with P:esmtp/R:smart_host/T:smtp (sender: ) id for ; Fri, 9 Jan 2009 06:10:41 -0800 (PST) (Smail-3.2.0.101 1997-Dec-17 #15 built 2001-Aug-30) Date: Fri, 9 Jan 2009 14:10:50 +0000 (GMT) From: Hugh Dickins To: Bob Copeland cc: Maxim Levitsky , linux-wireless@vger.kernel.org, Jiri Slaby , ath5k-devel@venema.h4ckr.net Subject: Re: [ath5k-devel] ath5k_tasklet_rx BUG_ON(bf->skb == NULL) In-Reply-To: Message-ID: (sfid-20090109_151113_476253_EB3ECDD9) References: <1231426017.922.2.camel@maxim-laptop> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-wireless-owner@vger.kernel.org List-ID: On Fri, 9 Jan 2009, Bob Copeland wrote: > On Thu, Jan 8, 2009 at 1:41 PM, Bob Copeland wrote: > > On Thu, Jan 8, 2009 at 12:55 PM, Hugh Dickins wrote: > >> Let me see if I can reproduce the ath5k BUG with a straightforward > >> memhog, repeatedly dirtying more anon memory than RAM can provide. > >> Is there something suitable I could run to exercise that wireless > >> path concurrently? It was just idling when I hit the BUGs before. > > > > Receiving any packets will do it, in your case it was probably > > beacons from nearby APs if you weren't using the wifi. So iperf > > or similar will work. > > Actually it's probably enough to just intentionally fail one > of the allocations in the driver. I'll do that and let you know > if I can easily break it. Good, that should be a lot quicker. My overnight try with simple memhog, and iperf -s receiving from another machine doing iperf -c at intervals, didn't hit it (but my more involved test wouldn't necessarily have hit it in that time either). Currently trying just a 2.6.28 defconfig kernel build in tmpfs (other filesystems should get their dirty pages pdflushed sooner) along with the iperf, but haven't got the sizes suitably tuned yet. Hugh