Re: [BUG] lockup with the latest kernel

[Steven Rostedt <rostedt@goodmis.org>]

On Fri, 2009-08-28 at 14:33 -0400, Steven Rostedt wrote:
> On Fri, 28 Aug 2009, Steven Rostedt wrote:
> > On Thu, 27 Aug 2009, Linus Torvalds wrote:
> > > > ...
> > > > >> [13288.222084] EIP: 0060:[<c0110821>] EFLAGS: 00000002 CPU: 0
> > > > >> [13288.222084] EIP is at default_send_IPI_mask_logical+0x53/0x92
> > > 
> > > Is this one perhaps fixed by b04e6373d694 ("x86: don't call 
> > > '->send_IPI_mask()' with an empty mask")
> > > 
> > > It sounds a _lot_ like that bug. Older dual-cpu x86 box, and APIC getting 
> > > confused by the occasional empty CPU mask, and then subsequent IPI's will 
> > > hang.
> > 
> > I'll update my box to that kernel and see what happens and let you know.
> 
> I checked out the commit at b04e6373d694, compiled and booted that kernel. 
> So far, so good. It survived logging into a gnome session (sometimes locks 
> up there), and a full kernel compile (usually locks up there too). But 
> there's been rare times that it would take days before hitting the bug.
> 
> I'll keep it in this kernel for a while and if it lasts the week, I'll 
> consider it fixed. This box acts as my email, dhcp and web server, as well 
> as one of the boxes in my distcc farm, thus it gets a bit of use. It will 
> not be idle during this time period.

Just an update.

The box currently has an uptime of 11 days. I have not hit the bug at
all, and it has survived numerous compiles (both locally and as a distcc
server). It has been serving up my email and web pages without a hitch.

I'd say that commit fixed the bug. (knock on wood).

-- Steve

(or it could have been the penguins:

  http://rostedt.homelinux.com/gallery2/main.php?g2_itemId=7860 )