Re: [PATCH][5/8] Arch agnostic completely out of line locks / ppc64

[William Lee Irwin III]

William Lee Irwin III writes:
>> The semantics of profile_pc() have never included backtracing through
>> scheduling primitives, so I'd say just report __preempt_spin_lock().

On Fri, Sep 10, 2004 at 09:36:52AM +1000, Paul Mackerras wrote:
> I disagree that __preempt_spin_lock is a scheduling primitive, or at
> least I disagree that it is primarily a scheduling primitive.  We
> don't spend vast amounts of time spinning in any of the other
> scheduling primitives; if we have to wait we call schedule().

Unfortunately the alternative appears to be stack unwinding in
profile_pc(), which was why I hoped we could punt. Any other ideas?


-- wli
-
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/