From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.wl.linuxfoundation.org ([198.145.29.98]:46976 "EHLO mail.wl.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726116AbfBAWAB (ORCPT ); Fri, 1 Feb 2019 17:00:01 -0500 Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 8B28632F5A for ; Fri, 1 Feb 2019 22:00:00 +0000 (UTC) From: bugzilla-daemon@bugzilla.kernel.org Subject: [Bug 202441] Possibly vfs cache related replicable xfs regression since 4.19.0 on sata hdd:s Date: Fri, 01 Feb 2019 21:59:59 +0000 Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8BIT MIME-Version: 1.0 Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: linux-xfs@vger.kernel.org https://bugzilla.kernel.org/show_bug.cgi?id=202441 --- Comment #26 from Roger (rogan6710@gmail.com) --- I have tried both your suggested reverts on a customized non-modular server oriented 5.0.0-rc3, and subjected the poor machine to a brutal test: 8 "simultaneous" defonfig compiles, a sort of load balancing test I do to evaluate io and cpu, on a xfs formatted hdd, while copying GB sized files to a btrfs formatted hdd ditto and my, by now, significant collection of kernel source trees to yet another ext4 formatted hdd. All this while watching a (very boring) youtube HD video and writing this. All working perfectly with a very responsive ui, while buff/cache is kept at 20-23 Gb, and almost no cpu idle. I have also done the earlier tests with very good results :) 5.5.0-rc3 locked up completely, and I had do do a "Norwegian reboot", when I tried loading it like that, earlier, without the reverts. >>From my point of view, they entirely solve this particular problem. I'll continue testing the reverts on other versions as well. All kernel compile processes finished with a delta t < 5o seconds, good :) (now try that on jfs ;)) -- You are receiving this mail because: You are watching the assignee of the bug.