linux-ext4 - [PATCH v5 00/12] ext4: port direct I/O to iomap infrastructure

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 21 Oct 2019 20:17:18 +1100
From:   Matthew Bobrowski <mbobrowski@...browski.org>
To:     tytso@....edu, jack@...e.cz, adilger.kernel@...ger.ca
Cc:     linux-ext4@...r.kernel.org, linux-fsdevel@...r.kernel.org,
        hch@...radead.org, david@...morbit.com, darrick.wong@...cle.com
Subject: [PATCH v5 00/12] ext4: port direct I/O to iomap infrastructure

Hello!

I managed to get some time over the weekend to fix up the stuff that
came out from the last review. Please feel free to review and provide
any necessary constructive criticism.

This patch series ports the ext4 direct I/O paths over to the iomap
infrastructure. The legacy buffer_head based direct I/O implementation
has subsequently been removed as it's no longer in use. The result of
this change is that ext4 now uses the newer iomap framework for direct
I/O read/write operations. Overall, this results in a much cleaner
direct I/O implementation and keeps this code isolated from the
buffer_head internals. In addition to this, a slight performance boost
may be expected while using O_SYNC | O_DIRECT.

The changes within this patch series have been tested via xfstests in
both DAX and non-DAX modes using the various filesystem configuration
options i.e. 4k, dioread_nolock, etc.

Changes since v4:

- Removed the ext4_handle_inode_extension() function out from the
  ->end_io() callback so that we can realiably truncate any allocated
  blocks as we have all the information we need in
  ext4_dio_write_iter().

- Fixed a couple comment formatting issues in addition to spelling
  mistakes here and there.

- Incorporated the fix for the issue that Dave Chinner found around
  writes extending beyond and not marking the iomap dirty.

Thank you all who took the time to review and provide feedback. :)

Jan Kara (2):
  iomap: Allow forcing of waiting for running DIO in iomap_dio_rw()
  xfs: Use iomap_dio_rw_wait()

Matthew Bobrowski (10):
  ext4: move set iomap routines into separate helper ext4_set_iomap()
  ext4: iomap that extends beyond EOF should be marked dirty
  ext4: split IOMAP_WRITE branch in ext4_iomap_begin() into helper
  ext4: introduce new callback for IOMAP_REPORT
  ext4: introduce direct I/O read using iomap infrastructure
  ext4: update direct I/O read to do trylock in IOCB_NOWAIT cases
  ext4: move inode extension/truncate code out from ->iomap_end()
    callback
  ext4: move inode extension check out from ext4_iomap_alloc()
  ext4: reorder map->m_flags checks in ext4_set_iomap()
  ext4: introduce direct I/O write using iomap infrastructure

 fs/ext4/ext4.h        |   4 +-
 fs/ext4/extents.c     |   4 +-
 fs/ext4/file.c        | 378 +++++++++++++++++-----
 fs/ext4/inode.c       | 716 +++++++++++-------------------------------
 fs/gfs2/file.c        |   6 +-
 fs/iomap/direct-io.c  |   7 +-
 fs/xfs/xfs_file.c     |  13 +-
 include/linux/iomap.h |   3 +-
 8 files changed, 498 insertions(+), 633 deletions(-)

-- 
2.20.1

--<M>--