From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.wl.linuxfoundation.org ([198.145.29.98]:35286 "EHLO mail.wl.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726548AbfA3KuN (ORCPT ); Wed, 30 Jan 2019 05:50:13 -0500 Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 4D3132E762 for ; Wed, 30 Jan 2019 10:50:13 +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: Wed, 30 Jan 2019 10:50:12 +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 #24 from Roger (rogan6710@gmail.com) --- A few moments ago, a piece of the puzzle fell into place. The problem persists. All this revert did, was to make the issue granularity dependent from what I gather. While redoing the test with all those source trees I used to compile the kernels, the same exact problem came back. The earlier tests were done using mostly multi GB files. After some more testing I found out that rc4 is completely problem free in this regard, but rc5 is definately not, so the real culprit was probably introduced in rc5. After applying our inode.c revert on 4.19.18, it shows the same behaviour as rc5. I can test on intermediate versions as well if necessary, as I still have them all compiled and ready. -- You are receiving this mail because: You are watching the assignee of the bug.