Messages in this thread | | | Date | Mon, 16 May 2011 18:25:14 +0200 (CEST) | From | Thomas Gleixner <> | Subject | Re: [PATCH v4 0/6] Micro-optimize vclock_gettime |
| |
On Mon, 16 May 2011, Andi Kleen wrote:
> > - Don't turn off frame pointers in vDSO code. > > Why? Note frame pointers are costly on some CPUs, like Atom.
That's the least worry on ATOM.
> Also unlike the primitive state of the kernel the user space debuggers > are usually advanced enough to support dwarf2 unwinding.
And unless you or someone else changes the primitive state of the kernel, framepointers are going to stay simply because removing them breaks profiling backtraces when the hit is inside vread().
Thanks,
tglx
| |