Messages in this thread | | | Date | Sat, 9 Aug 2008 03:55:39 +0200 | From | Andi Kleen <> | Subject | Re: [PATCH 0/5] ftrace: to kill a daemon |
| |
> Funny, CONFIG_FTRACE happens to select that. Now the question is, would > mcount work without it?
Not without fixing gcc first. It would work if gcc always called mcount the first thing before setting up the stack frame. Not sure why it doesn't do that.
Still do a benchmark of frame pointer vs no frame pointer kernel and you'll see, especially on a older CPUs without special hardware to avoid stack stalls (e.g. not Core2)
BTW always forcing frame pointers also means that ftrace is far from near zero overhead even when disabled. That is unless you find a way to nop the frame pointers too, but that would be likely very difficult because the code will actually use it.
-Andi
| |