From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752039AbdKFKjq (ORCPT ); Mon, 6 Nov 2017 05:39:46 -0500 Received: from mx2.suse.de ([195.135.220.15]:51472 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751754AbdKFKjp (ORCPT ); Mon, 6 Nov 2017 05:39:45 -0500 Date: Mon, 06 Nov 2017 11:39:41 +0100 Message-ID: From: Takashi Iwai To: "Kirill A. Shutemov" Cc: Dmitry Vyukov , syzbot , aaron.lu@intel.com, alsa-devel@alsa-project.org, Michal Hocko , Jan Kara , Minchan Kim , Peter Zijlstra , ying.huang@intel.com, syzkaller-bugs@googlegroups.com, LKML , linux-mm@kvack.org, "Aneesh Kumar K.V" , shli@fb.com, David Rientjes , Stephen Rothwell , Andrew Morton , zi.yan@cs.rutgers.edu, Matthew Wilcox , "Kirill A. Shutemov" Subject: Re: [alsa-devel] BUG: soft lockup In-Reply-To: <20171102090951.drjf7wc2urcmtla5@node.shutemov.name> References: <94eb2c19df188b1926055cf13c21@google.com> <20171102090951.drjf7wc2urcmtla5@node.shutemov.name> User-Agent: Wanderlust/2.15.9 (Almost Unreal) SEMI/1.14.6 (Maruoka) FLIM/1.14.9 (=?UTF-8?B?R29qxY0=?=) APEL/10.8 Emacs/25.3 (x86_64-suse-linux-gnu) MULE/6.0 (HANACHIRUSATO) MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 02 Nov 2017 10:09:51 +0100, Kirill A. Shutemov wrote: > > On Thu, Nov 02, 2017 at 09:23:58AM +0100, Takashi Iwai wrote: > > > > Currently the least ALSA timer interrupt period is limited to 1ms. > > Does it still too much? > > > > Can the reproducer triggers it reliably? If yes, could you forward > > it, too (and config as well), so that I'll try to dig down more exact > > code paths? > > All of that is part of original report: > > http://lkml.kernel.org/r/94eb2c19df188b1926055cf13c21@google.com > > marc.info hasn't stored repro.c for some reasone. Attached. > > I've just check it reproduces reliably for me in KVM. > > I also checked that it's not specific to THP -- still trigirable with huge > pages disabled. I guess this is the same issue Jerome forwarded recently, and it was fixed by limiting the amount of ALSA timer instances. I queued the fix in sound git tree for-linus branch, commit 9b7d869ee5a77ed4a462372bb89af622e705bfb8 ALSA: timer: Limit max instances per timer It'll be likely included in 4.14 final. thanks, Takashi