Thanks John. I'll give profile-0.11-2.4.16-sn1.patch a try.
We are trying to find best tool(s) to help identify regions
of code in the ia64 kernel that would benefit from improved
compiler code optimization. Some other folks in my group are
working on ia64 specific improvements to GCC. Kernprof looks
like an obvious candidate for our purposes.
If we are unable to locate a better tool (suggestions welcome!)
to help identify areas of code that could benefit from better
ia64 code optimization we may try to convince you that there
really is some "demand" out here for Kernprof on ia64 :-)
For our work, ia64 Kernprof patches that might break non-ia64
architectures would *not* be a problem.
Gary
On Thu, Feb 21, 2002 at 04:32:26PM -0800, John Hawkes wrote:
> From: "Gary Hade" <garyhade@xxxxxxxxxx>
> ...
> > This patch didn't seem to apply cleanly to IA-64 2.4.17
> > kernel source.
> ...
> > Are there any plans to provide patches for IA-64 kernels
> > later than 2.4.10?
>
> You might see how far you get with profile-0.11-2.4.16-sn1.patch. "sn1"
> is an offshoot of 2.4.16+ia64, and 2.4.17 might be fairly close to
> 2.4.16 in this area, I think.
>
> I stopped doing <kernel.org>+<ia64> patches because I didn't get the
> sense that anyone was using them. If there's a demand for them, then
> I'll start doing them again.
>
> My preference is to have a kernprof patch that patches cleanly against
> the kernel.org kernel, and to have an ia64 patch that patches cleanly
> against <kernel.org>+<ia64> and *only* works for ia64 (e.g., you can't
> use this patched kernel to also build i386 or sparc64).
>
> John Hawkes
--
Gary Hade
IBM Linux Technology Center
503-578-4503 IBM T/L: 775-4503
garyhade@xxxxxxxxxx
http://www.ibm.com/linux/ltc
|