Dumb kernel stacksize question

[Anton Blanchard]

> It only catches you whenever you get an _interrupt_ and are deep into
> the stack.  It's all a question of whether the interrupt comes in when
> you have your pants down.  This will catch a lot of problems, but it
> isn't very deterministic.

Yep but its simple :)

> There was an x86 patch from Ben La Haise that used gcc's mcount feature
> to check stack depth at every function call, but it was never ported to
> ppc and will probably impair performance.  How is ppc mcount support
> these days?

It should be fine, some of the performance guys were using it a while
ago I think.

Anton

** Sent via the linuxppc64-dev mail list. See http://lists.linuxppc.org/