Messages in this thread | | | Date | Thu, 11 Nov 2010 14:15:45 +0100 | From | Jiri Olsa <> | Subject | Re: [PATCHv2 2/2] tracing,mm - add kernel pagefault tracepoint for x86 & x86_64 |
| |
On Thu, Nov 11, 2010 at 07:51:57AM -0500, Christoph Hellwig wrote: > Can you explain again who this is supposed to be used? It's missing > a lot of essential information about the page fault, but on the other > hand traces totally useless information like the task struct pointer. > > What's your motivation for posting this given that you obviously > haven't actually tried to use it in practice? yep, kind of doing this for Larry... hoping he'd fill in to the discussion ;)
> > Btw, this doesn't mean I'm against a page faul trace point - quite > contrary. It just seems you're not actually pushing something that's > been tested in real life or is generally useful. What would be your idea of the page fault tracepoint?
thanks, jirka
| |