From: Tejun Heo Subject: Re: [PATCHSET block/for-next] IO cost model based work-conserving porportional controller Date: Thu, 5 Sep 2019 09:55:40 -0700 Message-ID: <20190905165540.GJ2263813@devbig004.ftw2.facebook.com> References: <20190614015620.1587672-1-tj@kernel.org> <20190614175642.GA657710@devbig004.ftw2.facebook.com> <5A63F937-F7B5-4D09-9DB4-C73D6F571D50@linaro.org> <20190820151903.GH2263813@devbig004.ftw2.facebook.com> <9EB760CE-0028-4766-AE9D-6E90028D8579@linaro.org> <20190831065358.GF2263813@devbig004.ftw2.facebook.com> <88C7DC68-680E-49BB-9699-509B9B0B12A0@linaro.org> <20190902155652.GH2263813@devbig004.ftw2.facebook.com> Mime-Version: 1.0 Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=sender:date:from:to:cc:subject:message-id:references:mime-version :content-disposition:in-reply-to:user-agent; bh=U86ZblmgaF4eBV4iyqRJf/CbV9WbHkWYMeJ0zFkqIu0=; b=TyUOTJJGoo8GODrCt/6YB1x/8aj4n9IE8jT/v+K7tCsOLINcESeyD/Ij4NXtMIB6Sy 17L6qDvWQyCyeC7XEmXxoOnrWrvCkDJaDG0IEmRd9h3bbbHGAAp7YN0Z5zIHZIZQLjXo e7R4fvWV6g6XW9aD+hLJiCg7Lmv5aeBV8tuo+o7VsPjG3fXt7UmiwBwDqQNJlFXWk3TH 5zwtfxF6mzHV0xMeRObFu3TfF099rR2GJ8US0yfbmsALUGRng05iBs5g6jzYHGRvKk8h FRvrTCHS5oUC5grWyibA9V7qghdaMk9oAmt6exGDzNOXAJpBRPdCLJ0cj/yBZJkPaFfd sbuQ== Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Paolo Valente Cc: Jens Axboe , newella@fb.com, clm@fb.com, Josef Bacik , dennisz@fb.com, Li Zefan , Johannes Weiner , linux-kernel , linux-block , kernel-team@fb.com, cgroups@vger.kernel.org, ast@kernel.org, daniel@iogearbox.net, kafai@fb.com, songliubraving@fb.com, yhs@fb.com, bpf@vger.kernel.org Hello, Paolo. So, I'm currently verifying iocost in the FB fleet. Around three thousand machines running v5.2 (+ some backports) with btrfs on a handful of different models of consumer grade SSDs. I haven't seen complete loss of control as you're reporting. Given that you're reporting the same thing on io.latency, which is deployed on multiple orders of magnitude more machines at this point, it's likely that there's something common affecting your test setup. Can you please describe your test configuration and if you aren't already try testing on btrfs? Thanks. -- tejun