[0/2] fstests: fix a major test setup performance regression

Message ID	20220520012336.1542637-1-david@fromorbit.com (mailing list archive)
Headers	show Return-Path: <fstests-owner@kernel.org> From: Dave Chinner <david@fromorbit.com> To: fstests@vger.kernel.org Subject: [PATCH 0/2] fstests: fix a major test setup performance regression Date: Fri, 20 May 2022 11:23:34 +1000 Message-Id: <20220520012336.1542637-1-david@fromorbit.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Precedence: bulk
Series	fstests: fix a major test setup performance regression \| expand [0/2] fstests: fix a major test setup performance regression [1/2] fstests: tests should not source common/rc directly [2/2] common/rc: fix test init performance regression on XFS

Message ID

20220520012336.1542637-1-david@fromorbit.com (mailing list archive)

Headers

From: Dave Chinner <david@fromorbit.com>
To: fstests@vger.kernel.org
Subject: [PATCH 0/2] fstests: fix a major test setup performance regression
Date: Fri, 20 May 2022 11:23:34 +1000
Message-Id: <20220520012336.1542637-1-david@fromorbit.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Series

fstests: fix a major test setup performance regression | expand

Message

Dave Chinner May 20, 2022, 1:23 a.m. UTC

Hi folks,

I've been seeing symptoms of this issue described in patch 2 for
most of this merge cycle, and it's slowly been getting worse, to
the point where I had to fix it to be able to get any amount of
testing done. Turns out we're doing something that is not smart
with XFS test setup, and recent changes to mkfs.xfs caused that
to be exposed. These patches fix it.

The first patch is a cleanup - no test should be including common/rc
directly anymore - _being_fstest() does that now. The second patch
fixes the XFS issues in init_rc() which is called from _begin_fstest
when is sources common/rc....

Cheers,

Dave.