Messages in this thread | | | Date | Sat, 27 Jul 2013 02:22:53 -0400 | Subject | Re: RFC: revert request for cpuidle patches e11538d1 and 69a37bea | From | Len Brown <> |
| |
>> OK, I'll queue up the reverts as fixes for 3.11-rc4. > > So, the reverts are on the fixes-next branch of the linux-pm.git tree that you > can access at > > http://git.kernel.org/cgit/linux/kernel/git/rafael/linux-pm.git/log/?h=fixes-next > > However, they are not simple reverts as we've had some non-trivial changes on > top of those commits already, so I'd appreciate it a lot if somebody could > double check if I didn't break anything in them.
I've verified that the reverts improve netperf TCP_RR performance.
Here I've got two Xeon's, slightly different, so I run in both directions. Also I run two ways -- Out-of-the-box, plus with cpufreq set to max frequency. The reverts improve all 4 cases:
JKT → IVT IVT → JKT
3.11.0-rc2 baseline w/o revert Out of Box
20420 19963 20658 19915 20298 20320
3.11.0-rc2 baseline w/o revert max freq
59658 51427 59663 51503 59416 51343
3.11.0-rc2-00002-g74bce39 Linux-pm “fixes-next” branch
3.11.0-rc2-00002-g74bce39 Out of box 23227 22056 23306 22125 23387 40226 40k result saw some 2.2 ghz vs 1.2 ghz in other runs 21991
3.11.0-rc2-00002-g74bce39 Max-freq 67240 57645 64880 56764 65924 57435
Tested-by: Len Brown <len.brown@intel.com>
thanks, Len Brown, Intel Open Source Technology Center -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |