From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mx1.redhat.com ([209.132.183.28]:41193 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750707AbcBLF3q (ORCPT ); Fri, 12 Feb 2016 00:29:46 -0500 Date: Fri, 12 Feb 2016 13:29:37 +0800 From: Eryu Guan Subject: Re: [PATCH 04/13] generic: remove the generic/125 test Message-ID: <20160212052937.GC11419@eguan.usersys.redhat.com> References: <1455170182-32587-1-git-send-email-tytso@mit.edu> <1455170182-32587-5-git-send-email-tytso@mit.edu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1455170182-32587-5-git-send-email-tytso@mit.edu> Sender: fstests-owner@vger.kernel.org To: Theodore Ts'o Cc: fstests@vger.kernel.org, hughd@google.com List-ID: On Thu, Feb 11, 2016 at 12:56:13AM -0500, Theodore Ts'o wrote: > Dave Chinner has pointed out this test is largely pointless: > > Perhaps one should question whether generic/125 is actually testing > anything useful in the first place. > > i.e. it's writing 16k, then truncating it back to 1000 bytes, then > reading it with direct IO for 60s to see if it returns the correct > data for that entire time. > > What, exactly, is going to cause that test to fail? > > Keep in mind this was ported from a CXFS test, where the metadata > server did the truncation operation (including the data zeroing), > but the reads are being done from the client. i.e. it was designed > to test whether a remote machine is doing a truncate correctly on a > access shared disk. > > IMO, I think we just remove the test, src/trunc.c and src/ftrunc.c > because it's 60s of testing that doesn't actually provide any value > to us. > > Signed-off-by: Theodore Ts'o > --- > tests/generic/125 | 68 ----------------------------------------------------- > tests/generic/group | 1 - > 2 files changed, 69 deletions(-) > delete mode 100755 tests/generic/125 Also remove 125.out ? Thanks, Eryu