Messages in this thread |  | | | Date | Tue, 25 Dec 2007 09:25:56 +0100 | | From | Ingo Molnar <> | | Subject | Re: volanoMark 24% regression in 2.6.24-rc6: why a simple patch makes it | |
* Zhang, Yanmin <yanmin_zhang@linux.intel.com> wrote:
> The solution is just to revert patch > fbdcf18df73758b2e187ab94678b30cd5f6ff9f9, because other 2 patches
> which fixed the same issue are already in 2.6.24-rc5.
Linus, please revert fbdcf18df73758, as requested by Yanmin. It was
noticed before (by Yinghai Lu) that this commit was not needed, but it
looked a harmless duplication and we incorrectly thought it's only a NOP
and wanted to fix it in v2.6.25 - but as Yanmin has analyzed it now, it
creates a sub-optimal sched-domains hierarchy (not setting up the domain
belonging to the core) when CONFIG_X86_HT=y.
Acked-by: Ingo Molnar <mingo@elte.hu>
nice work Yanmin!
Ingo
|  |