This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: [Bug uprobes/13539] occasional oops, kernel SEGV, RHEL5, :uprobes:uprobe_free_process+0xba/0x131
- From: David Smith <dsmith at redhat dot com>
- To: Srikar Dronamraju <srikar at linux dot vnet dot ibm dot com>
- Cc: Frank Ch Eigler <fche at redhat dot com>, systemtap at sourceware dot org
- Date: Fri, 06 Jan 2012 11:00:47 -0600
- Subject: Re: [Bug uprobes/13539] occasional oops, kernel SEGV, RHEL5, :uprobes:uprobe_free_process+0xba/0x131
- References: <bug-13539-6586@http.sourceware.org/bugzilla/> <bug-13539-6586-afG7gI5RZb@http.sourceware.org/bugzilla/> <20120106122516.GA17178@linux.vnet.ibm.com>
On 01/06/2012 06:25 AM, Srikar Dronamraju wrote:
>
> While I still cannot see a reason how uprobe_{free,put}_process can
> race uprobe_report_{exit,exec}, I certainly think somebit of cleanup
> can be done. However I am dont think we need to do a utask or uproc lookup
> from the table. Especially in case of callbacks.
>
> Mostly similar to what Jim proposed.
> I havent tested this patch myself and I couldnt reproduce the problem.
With your patch, I got the same stapio hang as I mentioned in comment #5
on the bug, but happened on the 3rd run of the unprivileged_myproc.exp
test case (it has happened on every run with the pr13539 branch code for
me). So your code is a slight improvement.
That comment also describes my environment in duplicating the bug which
might help you.
--
David Smith
dsmith@redhat.com
Red Hat
http://www.redhat.com
256.217.0141 (direct)
256.837.0057 (fax)