Messages in this thread | | | Date | Thu, 3 Jul 2008 11:20:52 +0200 | From | Ingo Molnar <> | Subject | Re: [PATCH 1/3] AMD Family10h+ IBS support for oProfile driver: Setup routines |
| |
* Barry Kasindorf <barry.kasindorf@amd.com> wrote:
> This patchset supports the new profiling hardware available in the > latest AMD CPUs in the oProfile driver.
Looks interesting! There are some patch logistics issues:
1) please rebase your patches ontop of -mm, which carries the oprofile multiplexing cleanups and enhancements already which heavily interact with your patchset.
2) -mm is also based on linux-next which has a few other changes in drivers/oprofile as well: i did a test-merge of your patches and 3/3 interacted heavily with API changes in linux-next. Part of the problem was that you mixed whitespace changes into your patch which randomly iteracted with other stuff. Cleanups are nice and welcome, but try to keep them in a separate patch. (multiple patches if the situation requires it)
3) please Cc: the x86 maintainers too because these patches change quite a few things in the x86 architecture code.
also:
> case 0x10: > model = &op_athlon_spec; > - cpu_type = "x86-64/family10"; > + cpu_type = "x86-64/family10h"; > + break;
please just keep the ABI string at "x86-64/family10" - we can live with this small quirk forever. The mess to remove that quirk is much larger than the benefit it brings (which is small to non-existent).
[ you really dont want such a small issue block your patches.]
the next one is OK:
> + case 0x11: > + model = &op_athlon_spec; > + cpu_type = "x86-64/family11h";
as it's really 0x11.
Ingo
| |