From: Azat Khuzhin Subject: Re: [PATCH] ext4: initialize multi-block allocator before checking block descriptors Date: Sun, 16 Mar 2014 03:54:34 +0400 Message-ID: References: <20140315214029.GA18873@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Cc: "open list:EXT4 FILE SYSTEM" To: "Theodore Ts'o" Return-path: Received: from mail-qa0-f41.google.com ([209.85.216.41]:46191 "EHLO mail-qa0-f41.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756590AbaCOXye (ORCPT ); Sat, 15 Mar 2014 19:54:34 -0400 Received: by mail-qa0-f41.google.com with SMTP id j5so4025865qaq.14 for ; Sat, 15 Mar 2014 16:54:34 -0700 (PDT) In-Reply-To: <20140315214029.GA18873@thunk.org> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Sun, Mar 16, 2014 at 1:40 AM, wrote: >>With EXT4FS_DEBUG ext4_count_free_clusters() will call >>ext4_read_block_bitmap() without s_group_info initialized, so we need to >>initialize multi-block allocator before. >> >>And we can't initialize multi-block allocator without group descriptors, >>since it use them. >>Also we need to install s_op before initializing multi-block allocator, >>because in ext4_mb_init_backend() new inode is created. > > I've had to drop this patch because it's causing a series of bigalloc > failures. I'll take a closer look at this later, but I wantd to give > you a heads up. Thanks Ted, I'll take a look. Are you talking about xfs-tests failures? > > - Ted > > > -- Respectfully Azat Khuzhin