On 10/24/2014, 05:52 AM, Hou Pengyang wrote: > hi gerg, > > Now on IvyBridge machines with 3.10 stable, perf could not get proper > dTLB-load-misses event when performing "perf stat -e dTLB-load-misses > sleep 1", which outputs: > > Performance counter stats for 'sleep 1': > 0 dTLB-load-misses > 1.000951106 seconds time elapsed > > 0 dTLB-load-misses means we fail to get the dTLB-load-misses event. > Upstream patch 1996388e explains the result and fix the bug. I have > already test the patch, by which I can get proper dTLB-load-misses on my > IvyBridge machine: > > 272 dTLB-load-misses > > So, I suggest cherry-picking it to 3.10-stable tree. Hi, and I picked it for 3.12 too. Thanks. > the commit is 1996388e9f4e3444db8273bc08d25164d2967c21 as follows: > > ------------------------------------------------------------------------ > commit 1996388e9f4e3444db8273bc08d25164d2967c21 > Author: Vince Weaver <vincent.weaver@xxxxxxxxx> > Date: Mon Jul 14 15:33:25 2014 -0400 > > perf/x86/intel: Use proper dTLB-load-misses event on IvyBridge > > This was discussed back in February: > > https://lkml.org/lkml/2014/2/18/956 > > But I never saw a patch come out of it. > > On IvyBridge we share the SandyBridge cache event tables, but the > dTLB-load-miss event is not compatible. Patch it up after > the fact to the proper DTLB_LOAD_MISSES.DEMAND_LD_MISS_CAUSES_A_WALK -- js suse labs -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html