From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx3-rdu2.redhat.com ([66.187.233.73]:43086 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1731577AbeGSOLt (ORCPT ); Thu, 19 Jul 2018 10:11:49 -0400 Date: Thu, 19 Jul 2018 15:28:34 +0200 From: Jiri Olsa To: Cong Wang Cc: LKML , Peter Zijlstra , Ingo Molnar , Linus Torvalds , Arnaldo Carvalho de Melo , Alexander Shishkin , Namhyung Kim , stable Subject: Re: [PATCH] perf/core: fix a possible deadlock scenario Message-ID: <20180719132834.GF18667@krava> References: <20180716215101.4713-1-xiyou.wangcong@gmail.com> <20180718081905.GA13520@krava> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: stable-owner@vger.kernel.org List-ID: On Wed, Jul 18, 2018 at 01:21:11PM -0700, Cong Wang wrote: SNIP > > hum, the swevent pmu does not triger NMI, so that timer > > will never be touched in NMI context > > Good to know! So I worry too much here. > > But still, given hrtimer interrupt is called with IRQ disabled, getting > stuck in a hrtimer callback could also block IPI handler, therefore > causing soft lockups. > > Let me know if you want me to adjust the changelog for this. yep, it's confusing, I think it should be removed from changelog jirka