Re: MM kernels 2.6.24-rc*-mm*, 2.6.24-mm1: gnome-terminal stuck intty_poll

[Jay Cliburn]

On Thu, 07 Feb 2008 21:24:47 +0100
Peter Zijlstra <a.p.zijlstra@xxxxxxxxx> wrote:

> 
> On Thu, 2008-02-07 at 20:49 +0100, Peter Zijlstra wrote:
> > On Wed, 2008-02-06 at 18:23 -0800, Andrew Morton wrote:
> > > On Wed, 06 Feb 2008 20:10:50 -0600 "J. K. Cliburn"
> > > <jacliburn@xxxxxxxxxxxxx> wrote:
> > > 
> > > > Zan Lynx wrote:
> > > > 
> > > > > gnome-terminal gets stuck.
> > > > 
> > > > I began seeing this very thing around 2.6.24 time.  (Fedora 8,
> > > > vanilla kernel.)  I could usually cause the gnome terminal to
> > > > hang if I rapidly resized the window while executing make
> > > > check-headers.
> 
> Weird, .24 proper doesn't have that patch. 

Yeah, my reference to "around 2.6.24 time" was simply a gross
demarcation along the passage of time rather than an indictment of
2.6.24 itself.

I began noticing gnome-terminal hangs during kernel builds a couple of
weeks ago, but I ignored them, thinking it'd be fixed with a Fedora
package update.  When that didn't happen, I began looking for the
culprit in the kernel in earnest this past Saturday.

> What exact fedora kernel was that (so I can look at it).

I didn't use a Fedora kernel; I bisected Linus' git current as of
Saturday Feb 2, using 2.6.24 as the "good" side of the bisect.  It took
the better part of two days to whittle down the throng, but by Sunday
night I'd settled on the 37bb6cb4 hrtimer commit.  My last act before
going to bed that night was to reset the bisect, revert the commit, and
rebuild.  I couldn't get gnome-terminal to hang using that rebuilt
kernel.

The next morning I went out of town for a couple of days, only to return
to find my workstation dead after some weather-related power outages.
I just got it back online tonight.

> Which is even weirder, because the provided trace indicates
> schedule_timeout()

The trace isn't from me; it's from Zan.  He's running -mm, I'm not, if
that makes a difference.

>         
>         Call Trace:
>          [schedule_timeout+149/208] schedule_timeout+0x95/0xd0
>          [<ffffffff8057dde5>] schedule_timeout+0x95/0xd0
>          [tty_poll+145/160] tty_poll+0x91/0xa0
>          [<ffffffff80430f11>] tty_poll+0x91/0xa0
>          [do_sys_poll+617/880] do_sys_poll+0x269/0x370
>          [<ffffffff802bbe69>] do_sys_poll+0x269/0x370
>          [__pollwait+0/304] __pollwait+0x0/0x130
>          [<ffffffff802bcb30>] __pollwait+0x0/0x130

Now that my computer is back on the air again, I'll be happy to help
track this down.  Just tell me what to do.

Jay
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/