From: Theodore Ts'o Subject: Re: [PATCH v3] Add largedir feature Date: Sat, 24 Jun 2017 01:43:56 -0400 Message-ID: <20170624054356.43nui7l6o27dphrr@thunk.org> References: <1494094761-19108-1-git-send-email-artem.blagodarenko@seagate.com> <19F5F19B-23F5-466A-B934-EB15896E1B15@dilger.ca> <20170622011142.gu5qdyqsmuox7v4f@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Artem Blagodarenko , linux-ext4 , Alexey Lyashkov , Eric Biggers , Yang Sheng , Artem Blagodarenko To: Andreas Dilger Return-path: Received: from imap.thunk.org ([74.207.234.97]:39000 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750805AbdFXFoE (ORCPT ); Sat, 24 Jun 2017 01:44:04 -0400 Content-Disposition: inline In-Reply-To: <20170622011142.gu5qdyqsmuox7v4f@thunk.org> Sender: linux-ext4-owner@vger.kernel.org List-ID: It looks like there are at least two xfstests failures if the largedir feature is enabled. generic/021 fails, but I think this may be a test issue because it seems to be a mke2fs complaining about not having enough space to create the file system. generic/027, which is an ENOSPC hitter, is failing with largedir enabled, but apparently it hasn't been failing w/o this feature being enabled. The test description is "Run 8 processes writing 1k files to seperate files in seperate dirs to hit ENOSPC on small fs with little free space. Loop for 100 iterations." This was using the following configuration: export EXT_MKFS_OPTIONS="-I 2048 -O ^64bit,mmp,uninit_bg,^extents,large_dir,dir_nlink,quota,huge_file,flex_bg -E lazy_journal_init" - Ted