From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx2.suse.de (cantor2.suse.de [195.135.220.15]) (using TLSv1 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 818901A0008 for ; Fri, 6 Mar 2015 10:54:59 +1100 (AEDT) From: Mel Gorman To: Dave Chinner Subject: [RFC PATCH 0/2] Automatic NUMA balancing and PROT_NONE handling followup Date: Thu, 5 Mar 2015 23:54:50 +0000 Message-Id: <1425599692-32445-1-git-send-email-mgorman@suse.de> Cc: linuxppc-dev@lists.ozlabs.org, Linux Kernel Mailing List , xfs@oss.sgi.com, Linux-MM , Aneesh Kumar , Andrew Morton , Linus Torvalds , Ingo Molnar , Mel Gorman List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Dave Chinner reported a problem due to excessive NUMA balancing activity and bisected it. These are two patches that address two major issues with that series. The first patch is almost certainly unrelated to what he saw due to fact his vmstats showed no huge page activity but the fix is important. The second patch restores performance of one benchmark to similar levels to 3.19-vanilla but it still has to be tested on his workload. While I have a test configuration for his workload, I don't have either the KVM setup or suitable storage to test against. It also needs to be reviewed and tested on ppc64. arch/powerpc/include/asm/pgtable-ppc64.h | 16 ++++++++++++++++ arch/x86/include/asm/pgtable.h | 14 ++++++++++++++ include/asm-generic/pgtable.h | 19 +++++++++++++++++++ mm/huge_memory.c | 23 ++++++++++++++++++----- mm/mprotect.c | 5 +++++ 5 files changed, 72 insertions(+), 5 deletions(-) -- 2.1.2