From mboxrd@z Thu Jan 1 00:00:00 1970 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 E884AEB64DA for ; Wed, 19 Jul 2023 15:25:54 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231714AbjGSPZx (ORCPT ); Wed, 19 Jul 2023 11:25:53 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35306 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231697AbjGSPZw (ORCPT ); Wed, 19 Jul 2023 11:25:52 -0400 Received: from verein.lst.de (verein.lst.de [213.95.11.211]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 80F931FD3; Wed, 19 Jul 2023 08:25:33 -0700 (PDT) Received: by verein.lst.de (Postfix, from userid 2407) id 805CE67373; Wed, 19 Jul 2023 17:25:29 +0200 (CEST) Date: Wed, 19 Jul 2023 17:25:29 +0200 From: Christoph Hellwig To: Josef Bacik Cc: Christoph Hellwig , Chris Mason , David Sterba , linux-btrfs@vger.kernel.org, linux-fsdevel@vger.kernel.org Subject: Re: small writeback fixes Message-ID: <20230719152529.GA434@lst.de> References: <20230713130431.4798-1-hch@lst.de> <20230718171744.GA843162@perftesting> <20230719053901.GA3241@lst.de> <20230719115010.GA15617@lst.de> <20230719143032.GA856129@perftesting> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20230719143032.GA856129@perftesting> User-Agent: Mutt/1.5.17 (2007-11-01) Precedence: bulk List-ID: X-Mailing-List: linux-btrfs@vger.kernel.org On Wed, Jul 19, 2023 at 10:30:32AM -0400, Josef Bacik wrote: > I backed your patches out and re-ran and I hit hangs with generic/475 still, so > I think you're clear. There's something awkward going on here, the below hang > just looks like we're waiting for IO. The caching thread is blocking the > transaction commit because it's trying to read some old blocks, and it's been > waiting for them to come back for 2 minutes. That's holding everybody else up. > I'll dig into all of this, misc-next is definitely fucked somehow, your stuff > may just be a victim. Thanks, Yes. I think this has shown up in misc-next as so far I've not been able to reproduce anything on 6.5-rc2.