Messages in this thread | | | From | Andi Kleen <> | Subject | Re: [PATCH] Revert "x86: optimize page faults like all other achitectures and kill notifier cruft" | Date | Wed, 9 Jan 2008 20:07:13 +0100 |
| |
> Probing vmalloc faults is _really_ tricky : it also implies that the > handler (let's call it probe) connected to the probe point (marker or > kprobe) should _never_ cause a vmalloc page fault,
That is why vmalloc_sync_all() was invented. It might make sense to just call that on kprobe registration.
But I agree the other problems makes it a bad idea.
I think the better solution is to keep the notifier, but make it cheaper (e.g. by using constant patching ...)
-Andi
| |