Content-Type: multipart/mixed; boundary="===============0343155925075324567==" MIME-Version: 1.0 From: Huang Ying To: lkp@lists.01.org Subject: Re: [mm] 721c21c17ab: +11.7% will-it-scale.per_thread_ops Date: Thu, 12 Feb 2015 12:56:31 +0800 Message-ID: <1423716991.5968.125.camel@intel.com> In-Reply-To: <1423114910.5968.9.camel@linux.intel.com> List-Id: --===============0343155925075324567== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Thu, 2015-02-05 at 13:41 +0800, Huang Ying wrote: > On Tue, 2015-02-03 at 10:27 +0000, Will Deacon wrote: > > On Tue, Feb 03, 2015 at 07:45:04AM +0000, Huang Ying wrote: > > > FYI, we noticed the below changes on > > > = > > > commit 721c21c17ab958abf19a8fc611c3bd4743680e38 ("mm: mmu_gather: use > > > tlb->end !=3D 0 only for TLB invalidation") > > = > > Is this delta relative to the previous commit, or something else? The > > commit in question restored some 3.18 behaviour on task exit (namely > > doing a fullmm TLB flush), so it would be interesting to know if you > > see the same change against 3.18. > = > Sorry for late. The delta is relative to the previous commit. I will > queue some test for 3.18 to check. The test result for v3.18 compared with the commit 721c21c is as below. You can find that the difference between v3.18 and 721c21c should be less that between v3.18 and the previous commit (parent) for will-it-scale.per_thread_ops. I think this is expected. testbox/testcase/testparams: nhm4/will-it-scale/performance-pread2 v3.18 721c21c17ab958abf19a8fc611 ---------------- -------------------------- %stddev %change %stddev \ | \ = 31541022 =C2=B1 0% -12.2% 27698665 =C2=B1 0% will-it-scale.time.= voluntary_context_switches 1079302 =C2=B1 0% -6.3% 1011350 =C2=B1 2% will-it-scale.per_p= rocess_ops 0.04 =C2=B1 0% -4.5% 0.03 =C2=B1 1% will-it-scale.scala= bility 1038581 =C2=B1 0% -3.2% 1005724 =C2=B1 0% will-it-scale.per_t= hread_ops Best Regards, Huang, Ying --===============0343155925075324567==--