From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932981AbcA2WkJ (ORCPT ); Fri, 29 Jan 2016 17:40:09 -0500 Received: from shelob.surriel.com ([74.92.59.67]:47528 "EHLO shelob.surriel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755777AbcA2WkG (ORCPT ); Fri, 29 Jan 2016 17:40:06 -0500 From: riel@redhat.com To: linux-kernel@vger.kernel.org Cc: tglx@linutronix.de, mingo@kernel.org, peterz@infradead.org, luto@amacapital.net, fweisbec@gmail.com Subject: [PATCH 0/2] sched,time: reduce nohz_full syscall overhead 40% Date: Fri, 29 Jan 2016 17:22:58 -0500 Message-Id: <1454106180-20918-1-git-send-email-riel@redhat.com> X-Mailer: git-send-email 2.5.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Running with nohz_full introduces a fair amount of overhead. Specifically, various things that are usually done from the timer interrupt are now done at syscall, irq, and guest entry and exit times. However, some of the code that is called every single time has only ever worked at jiffy resolution. The code in __acct_update_integrals was also doing some unnecessary calculations. Getting rid of the unnecessary calculations, without changing any of the functionality in __acct_update_integrals gets us about a 10% win. Not calling __acct_update_integrals and related code unless jiffies changed (__acct_update_integrals does not do anything with smaller time intervals anyway) shaves off a further 30%. Run times for the microbenchmark: 4.4 3.8 seconds 4.5-rc1 3.7 seconds 4.5-rc1 + first patch 3.3 seconds 4.5-rc1 + both patches 2.3 seconds