From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-it1-f194.google.com ([209.85.166.194]:36549 "EHLO mail-it1-f194.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726649AbeJLC77 (ORCPT ); Thu, 11 Oct 2018 22:59:59 -0400 Received: by mail-it1-f194.google.com with SMTP id c85-v6so14968037itd.1 for ; Thu, 11 Oct 2018 12:31:21 -0700 (PDT) Subject: Re: [PATCH] blk-wbt: wake up all when we scale up, not down To: Josef Bacik , linux-block@vger.kernel.org, kernel-team@fb.com Cc: stable@vger.kernel.org References: <20181011192930.2844-1-josef@toxicpanda.com> From: Jens Axboe Message-ID: <729350ca-183a-ad1f-9d4f-3e90913a40f8@kernel.dk> Date: Thu, 11 Oct 2018 13:31:17 -0600 MIME-Version: 1.0 In-Reply-To: <20181011192930.2844-1-josef@toxicpanda.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: stable-owner@vger.kernel.org List-ID: On 10/11/18 1:29 PM, Josef Bacik wrote: > Tetsuo brought to my attention that I screwed up the scale_up/scale_down > helpers when I factored out the rq-qos code. We need to wake up all the > waiters when we add slots for requests to make, not when we shrink the > slots. Otherwise we'll end up things waiting forever. This was a > mistake and simply puts everything back the way it was. I'll apply this for 4.19. I haven't hit this in testing, I'm guessing the jitter when going up and down hides the issue. -- Jens Axboe