From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965255AbbLONjz (ORCPT ); Tue, 15 Dec 2015 08:39:55 -0500 Received: from casper.infradead.org ([85.118.1.10]:41304 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S965187AbbLONjw (ORCPT ); Tue, 15 Dec 2015 08:39:52 -0500 Date: Tue, 15 Dec 2015 14:39:47 +0100 From: Peter Zijlstra To: Dmitry Vyukov Cc: Ingo Molnar , Arnaldo Carvalho de Melo , LKML , syzkaller , Kostya Serebryany , Alexander Potapenko , Sasha Levin , Eric Dumazet Subject: Re: perf: stalls in perf_install_in_context/perf_remove_from_context Message-ID: <20151215133947.GG6357@twins.programming.kicks-ass.net> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Dec 15, 2015 at 02:01:32PM +0100, Dmitry Vyukov wrote: > Call Trace: > [] sched_show_task+0x148/0x250 kernel/sched/core.c:4934 > [] apic_timer_interrupt+0x8c/0xa0 arch/x86/entry/entry_64.S:520 > > [< inline >] ? arch_local_irq_enable ./arch/x86/include/asm/paravirt.h:816 > [< inline >] ? __raw_spin_unlock_irq include/linux/spinlock_api_smp.h:170 > [] ? _raw_spin_unlock_irq+0x2e/0x40 kernel/locking/spinlock.c:199 > [< inline >] ? __raw_spin_unlock_irq include/linux/spinlock_api_smp.h:170 > [] ? _raw_spin_unlock_irq+0x27/0x40 kernel/locking/spinlock.c:199 > [] perf_install_in_context+0x6b/0xf0 kernel/events/core.c:2173 > [] SYSC_perf_event_open+0x63c/0xe40 kernel/events/core.c:8565 > [] SyS_perf_event_open+0x9/0x10 kernel/events/core.c:8261 > [] entry_SYSCALL_64_fastpath+0x16/0x7a arch/x86/entry/entry_64.S:185 So this looks like its stuck in the retry loop in perf_install_in_context(). We do the raw_spin_unlock_irq(), which allows the above watchdog interrupt to happen, but make no progress. This means the task_function_call() thing keeps failing, which is extremely unlikely. Let me have a play with your test-case to see if it reproduces.