[V2,0/5] xfs: speed up large directory modifications

Message ID	20190829063042.22902-1-david@fromorbit.com (mailing list archive)
Headers	show Return-Path: <SRS0=CuST=WZ=vger.kernel.org=linux-xfs-owner@kernel.org> From: Dave Chinner <david@fromorbit.com> To: linux-xfs@vger.kernel.org Subject: [PATCH V2 0/5] xfs: speed up large directory modifications Date: Thu, 29 Aug 2019 16:30:37 +1000 Message-Id: <20190829063042.22902-1-david@fromorbit.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Sender: linux-xfs-owner@vger.kernel.org Precedence: bulk
Series	xfs: speed up large directory modifications \| expand [V2,0/5] xfs: speed up large directory modifications [1/5] xfs: move xfs_dir2_addname() [2/5] xfs: factor data block addition from xfs_dir2_node_addname_int() [3/5] xfs: factor free block index lookup from xfs_dir2_node_addname_int() [4/5] xfs: speed up directory bestfree block scanning [5/5] xfs: reverse search directory freespace indexes

Message ID

20190829063042.22902-1-david@fromorbit.com (mailing list archive)

Headers

From: Dave Chinner <david@fromorbit.com>
To: linux-xfs@vger.kernel.org
Subject: [PATCH V2 0/5] xfs: speed up large directory modifications
Date: Thu, 29 Aug 2019 16:30:37 +1000
Message-Id: <20190829063042.22902-1-david@fromorbit.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Sender: linux-xfs-owner@vger.kernel.org
Precedence: bulk

Series

xfs: speed up large directory modifications | expand

Message

Dave Chinner Aug. 29, 2019, 6:30 a.m. UTC

Hi folks,

After a long time I've managed to get back to these directory
speedup patches, originally posted here:

https://lore.kernel.org/linux-xfs/20181024225716.19459-1-david@fromorbit.com/

I've addressed all of Christoph's original issues, incorporated his
suggestions, updated the benchmark results (same/slightly better
improvement) and done more testing on it. The series has been in my
test tree since I posted it ~9 months ago and has been in all my
benchmarking work over that time. I haven't seen any performance
regression as a result of the change of algorithm, but there are a
few that go a lot faster....

Comments welcome.

-Dave.