From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758830Ab2JXVNf (ORCPT ); Wed, 24 Oct 2012 17:13:35 -0400 Received: from condef004-v.nifty.com ([210.131.4.241]:49456 "EHLO condef004-v.nifty.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751447Ab2JXVNe (ORCPT ); Wed, 24 Oct 2012 17:13:34 -0400 X-Nifty-SrcIP: [124.34.79.27] Message-Id: <201210242111.q9OLBL28027212@msag502.nifty.com> MIME-Version: 1.0 Date: Thu, 25 Oct 2012 06:11:17 +0900 X-Mailer: Denshin 8 Go V32.1.6.0 070 From: MITSUNARI Shigeo To: Jeff Moyer Cc: linux-fsdevel@vger.kernel.org, Alexander Viro , linux-kernel@vger.kernel.org, Jens Axboe In-Reply-To: (Your message of "Wed, 24 Oct 2012 09:44:23 -0400") References: Subject: Re: fs: page cache wrongly left invalidated after revalidate_disk Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Jeff Moyer writes: > rescan_partitions (called in the bdev open path) should reset > bd_invalidated. I think you need to dig into why that isn't happening. rescan_partitions() and invalidate_partitions() reset bd_invalidated after calling check_disk_size_change(). But revalidate_disk() does not it. So, I think bdev->bd_invalidated = 0 is necessary for revalidate_disk(), or check_disk_size_change() should be it more simply. Yours, Shigeo