Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1D0D4C433F5 for ; Fri, 26 Nov 2021 18:35:44 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240326AbhKZSi4 (ORCPT ); Fri, 26 Nov 2021 13:38:56 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:60174 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S238037AbhKZSg4 (ORCPT ); Fri, 26 Nov 2021 13:36:56 -0500 Received: from mail-io1-xd2a.google.com (mail-io1-xd2a.google.com [IPv6:2607:f8b0:4864:20::d2a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id C1D50C07E5F4 for ; Fri, 26 Nov 2021 10:10:11 -0800 (PST) Received: by mail-io1-xd2a.google.com with SMTP id z26so12383235iod.10 for ; Fri, 26 Nov 2021 10:10:11 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20210112.gappssmtp.com; s=20210112; h=subject:to:cc:references:from:message-id:date:user-agent :mime-version:in-reply-to:content-language:content-transfer-encoding; bh=w7CM3ybXkX6FdnmkDgtC04LvePrjVXFwUyZ4uxrkvxw=; b=P8Ix0b+8pqLzJ0JCQMU2IsK8LUujjsUrb3I7IY1MpjfctmRfdlJ4p5tQIRJzkm1aOz NEkeOe/Fpw0bltwzcwIBsjrvsMFs+atSgWqaJBoi1ve+1El4o+XCYCZk7ONPmcUPhr0Q VY2Y363OEpPx+MVKPTIo7svhG2nMbqSLtYfjGO7290AYp504mY/rhBkFJqwLmxNGmdCC IR6hV7/8/Ng6R2bTvHJ81I08OFXyRyKS6gBHjJKWfVSk5khiZDKwJlCNGwfwUBkXrYO0 yr6CtDeMftUyZMd+0437LW51xaN8/Ora/MppBtTIvenEiWXu06d1T0X1om2tC4k/M9DP NNCQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:subject:to:cc:references:from:message-id:date :user-agent:mime-version:in-reply-to:content-language :content-transfer-encoding; bh=w7CM3ybXkX6FdnmkDgtC04LvePrjVXFwUyZ4uxrkvxw=; b=aCkgUlS/nUVNvjo2GcWi6dN391Qo8Wm+walgtcwZiw4Evjm5NofMb5fRebGXX38rlv UgvjUjYnldq+pi6rdWE1Rzh+7hiqdYHVmHHPfhLUpSNiO3DRN3q69alTRsyeRz3LwHEP oZ9lq4FXZzqfUhZMGFjtxKTmI5EhijCYRNUWFOu//hnGLzoh8F5wqbeinRq4Av9c6mRH ZBmYzUu4qgTg7nU0XQINbDASHuK6370TqyhuOhXzcFmaZc+vWPUEuzbGsqwHrer+7fl7 MmbXD73jWVJIVZQkO9VrWse/StHIDR10JGHv8Awr2pAzJmcmmbODSLGFcB/i65M/2mU8 fPUg== X-Gm-Message-State: AOAM532DUPrSD4VHcNTOy/C9kJFdTzKWEvPPDRvufpF6TEnfRUUWxBpF d8/csuWZikFR5R6lveFDnjNXMg== X-Google-Smtp-Source: ABdhPJxGCB354MvIlJjaIRBZzlJPUUDQyfSJlCoz2hhDCkTQKLQVSnsDHVTgVFm+cuSMnozo2m/8Xw== X-Received: by 2002:a05:6638:3014:: with SMTP id r20mr20589348jak.146.1637950211126; Fri, 26 Nov 2021 10:10:11 -0800 (PST) Received: from [192.168.1.116] ([66.219.217.159]) by smtp.gmail.com with ESMTPSA id d137sm3510617iof.16.2021.11.26.10.10.10 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Fri, 26 Nov 2021 10:10:10 -0800 (PST) Subject: Re: Write I/O queue hangup at random on recent Linus' kernels To: "Kenneth R. Crudup" Cc: linux-bcache@vger.kernel.org, linux-block@vger.kernel.org, linux-ext4@vger.kernel.org, linux-nvme@lists.infradead.org, Shinichiro Kawasaki References: <17206ea6-506d-b1de-09e8-c935ff308bd6@kernel.dk> <903be817-4118-f34e-1b35-a0108045590f@kernel.dk> <986e942b-d430-783b-5b1c-4525d4a94e48@panix.com> From: Jens Axboe Message-ID: Date: Fri, 26 Nov 2021 11:10:09 -0700 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.10.0 MIME-Version: 1.0 In-Reply-To: <986e942b-d430-783b-5b1c-4525d4a94e48@panix.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Precedence: bulk List-ID: X-Mailing-List: linux-ext4@vger.kernel.org On 11/26/21 10:58 AM, Kenneth R. Crudup wrote: > > On Fri, 26 Nov 2021, Jens Axboe wrote: > >> Can you apply this on top of 5.16-rc2 or current -git and see if it fixes >> it for you? > > Since it's related to writeback throttling, is there anything I can try to > taunt it? I was thinking "iozone" or "bonnie++", but if there's something > better, let me know. I'd just do what you usually do, that's usually the best way to gain confidence in the fix. But it's related to writeback throttling _and_ racing with eg iostats grabbing a reference to the request, so it would be more likely to trigger with a heavier combination of those two. That said, I'm pretty confident in the fix, so if you'd just use it for a day or two, then with the previous amount of hangs you saw on -rc2 it should provide a good level of confidence in it. -- Jens Axboe