From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753947AbdASQr7 (ORCPT ); Thu, 19 Jan 2017 11:47:59 -0500 Received: from mga01.intel.com ([192.55.52.88]:14463 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753693AbdASQpy (ORCPT ); Thu, 19 Jan 2017 11:45:54 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.33,254,1477983600"; d="scan'208";a="810885983" From: Alexander Shishkin To: Peter Zijlstra Cc: Ingo Molnar , linux-kernel@vger.kernel.org, vince@deater.net, eranian@google.com, Arnaldo Carvalho de Melo , Alexander Shishkin Subject: [PATCH 0/2] perf: Scheduling optimizations Date: Thu, 19 Jan 2017 18:43:28 +0200 Message-Id: <20170119164330.22887-1-alexander.shishkin@linux.intel.com> X-Mailer: git-send-email 2.11.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Peter, These here two patches deal with rescheduling on active contexts. The common idea is that we shouldn't reschedule groups of higher priority than the ones that are being added or scheduled in. 1/1 takes on the scheduling path, where we don't want to touch cpu flexible groups if the scheduling context doesn't carry pinned groups. 1/2 deals with the enable/install cases, where we have to consider all four types of groups. Alexander Shishkin (2): perf: Don't re-schedule cpu flexible events needlessly perf: Optimize event rescheduling on active contexts kernel/events/core.c | 86 ++++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 74 insertions(+), 12 deletions(-) -- 2.11.0