From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Thu, 20 Dec 2018 16:41:36 +0100 From: Greg Kroah-Hartman To: Sudip Mukherjee Cc: linux-kernel , Stable , Will Deacon , "Peter Zijlstra (Intel)" , Linus Torvalds , Thomas Gleixner , Ingo Molnar , Sebastian Andrzej Siewior , Sasha Levin Subject: Re: [PATCH 4.14 21/72] locking/qspinlock: Ensure node is initialised before updating prev->next Message-ID: <20181220154135.GB9962@kroah.com> References: <20181220085922.332225035@linuxfoundation.org> <20181220085923.169451350@linuxfoundation.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: List-ID: On Thu, Dec 20, 2018 at 12:19:04PM +0000, Sudip Mukherjee wrote: > Hi Greg, > > On Thu, Dec 20, 2018 at 9:27 AM Greg Kroah-Hartman > wrote: > > > > 4.14-stable review patch. If anyone has any objections, please let me know. > > > > ------------------ > > > > commit 95bcade33a8af38755c9b0636e36a36ad3789fe6 upstream. > > It seems 9d4646d14d51 ("locking/qspinlock: Elide back-to-back RELEASE > operations with smp_wmb()") is improving the fix. Is it worth to take > this also? If someone could test that this really does help things out, I would be glad to take it :) thanks, greg k-h