Re: 2.5.38-mm1 dbench 512 might sleep backtrace emitted

William Lee Irwin III (wli@holomorphy.com)
Sun, 22 Sep 2002 22:33:54 -0700


William Lee Irwin III wrote:
>> Trace; c01175f7 <__might_sleep+27/2b>

On Sun, Sep 22, 2002 at 10:31:04PM -0700, Andrew Morton wrote:
> Well I can't immediately see any held locks on that path, can
> you? Odd.
> Might be best to put a breakpoint on the printk in __might_sleep(),
> get some more info if it bites again.
> If there _are_ no locks held in that chain then there is
> something wrong with in_atomic(). So check the current
> task state with `task25' and `thread25' from my .gdbinit.

I squirted it out just in case someone (i.e. you) recognized it off the
bat. I'm looking into it myself also with methods similar to what's
suggested.

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