linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCHSET v3 0/4] Submit ->readpages() IO as read-ahead
@ 2018-05-30 14:42 Jens Axboe
  2018-05-30 14:42 ` [PATCH 1/4] mpage: add argument structure for do_mpage_readpage() Jens Axboe
                   ` (4 more replies)
  0 siblings, 5 replies; 18+ messages in thread
From: Jens Axboe @ 2018-05-30 14:42 UTC (permalink / raw)
  To: linux-fsdevel; +Cc: viro, akpm

The only caller of ->readpages() is from read-ahead, yet we don't
submit IO flagged with REQ_RAHEAD. This means we don't see it in
blktrace, for instance, which is a shame. We already make assumptions
about ->readpages() just being for read-ahead in the mpage
implementation, using readahead_gfp_mask(mapping) as out GFP mask of
choice.

This small series fixes up mpage_readpages() to submit with
REQ_RAHEAD, which takes care of file systems using mpage_readpages().
The first patch is a prep patch, that makes do_mpage_readpage() take
an argument structure.

Changes since v2:

- Get rid of 'gfp' passing once we have is_readahead
- Pack struct better, makes it 8 bytes smaller.

Changes since v1:

- Fix ext4_mpage_readpages() also being used for regular reads
- Add prep patch with struct arguments for do_mpage_readpage()

 fs/btrfs/extent_io.c |   2 +-
 fs/ext4/ext4.h       |   2 +-
 fs/ext4/inode.c      |   5 +-
 fs/ext4/readpage.c   |   5 +-
 fs/mpage.c           | 115 ++++++++++++++++++++++++-------------------
 5 files changed, 71 insertions(+), 58 deletions(-)

^ permalink raw reply	[flat|nested] 18+ messages in thread
* [PATCH v4 0/4] Submit ->readpages() IO as read-ahead
@ 2018-06-21  1:07 Jens Axboe
  2018-06-21  1:07 ` [PATCH 3/4] btrfs: readpages() should submit " Jens Axboe
  0 siblings, 1 reply; 18+ messages in thread
From: Jens Axboe @ 2018-06-21  1:07 UTC (permalink / raw)
  To: linux-fsdevel; +Cc: viro, akpm

The only caller of ->readpages() is from read-ahead, yet we don't
submit IO flagged with REQ_RAHEAD. This means we don't see it in
blktrace, for instance, which is a shame. Additionally, it's
preventing further functional changes in the block layer for
deadling with read-ahead more intelligently. We already make assumptions
about ->readpages() just being for read-ahead in the mpage
implementation, using readahead_gfp_mask(mapping) as out GFP mask of
choice.

This small series fixes up mpage_readpages() to submit with REQ_RAHEAD,
which takes care of file systems using mpage_readpages().  The first
patch is a prep patch, that makes do_mpage_readpage() take an argument
structure.

Changes since v3:

- Add comments on ->readpages() purely being read-ahead
- Rebase on current -git

Changes since v2:

- Get rid of 'gfp' passing once we have is_readahead
- Pack struct better, makes it 8 bytes smaller.

Changes since v1:

- Fix ext4_mpage_readpages() also being used for regular reads
- Add prep patch with struct arguments for do_mpage_readpage()

 fs/btrfs/extent_io.c |   2 +-
 fs/ext4/ext4.h       |   2 +-
 fs/ext4/inode.c      |   5 +-
 fs/ext4/readpage.c   |   5 +-
 fs/f2fs/data.c       |   5 ++
 fs/mpage.c           | 115 ++++++++++++++++++++++++-------------------
 include/linux/fs.h   |   4 ++
 7 files changed, 80 insertions(+), 58 deletions(-)

-- 
Jens Axboe

^ permalink raw reply	[flat|nested] 18+ messages in thread
* [PATCHSET v2 0/4] Submit ->readpages() IO as read-ahead
@ 2018-05-29 22:17 Jens Axboe
  2018-05-29 22:17 ` [PATCH 3/4] btrfs: readpages() should submit " Jens Axboe
  0 siblings, 1 reply; 18+ messages in thread
From: Jens Axboe @ 2018-05-29 22:17 UTC (permalink / raw)
  To: linux-fsdevel; +Cc: viro, akpm

The only caller of ->readpages() is from read-ahead, yet we don't
submit IO flagged with REQ_RAHEAD. This means we don't see it in
blktrace, for instance, which is a shame. We already make assumptions
about ->readpages() just being for read-ahead in the mpage
implementation, using readahead_gfp_mask(mapping) as out GFP mask of
choice.

This small series fixes up mpage_readpages() to submit with
REQ_RAHEAD, which takes care of file systems using mpage_readpages().
The first patch is a prep patch, that makes do_mpage_readpage() take
an argument structure.

Changes since v1:

- Fix ext4_mpage_readpages() also being used for regular reads
- Add prep patch with struct arguments for do_mpage_readpage()

 fs/btrfs/extent_io.c |   2 +-
 fs/ext4/ext4.h       |   2 +-
 fs/ext4/inode.c      |   5 +-
 fs/ext4/readpage.c   |   5 +-
 fs/mpage.c           | 109 ++++++++++++++++++++++---------------------
 5 files changed, 65 insertions(+), 58 deletions(-)

-- 
Jens Axboe

^ permalink raw reply	[flat|nested] 18+ messages in thread

end of thread, other threads:[~2018-06-21 13:52 UTC | newest]

Thread overview: 18+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-05-30 14:42 [PATCHSET v3 0/4] Submit ->readpages() IO as read-ahead Jens Axboe
2018-05-30 14:42 ` [PATCH 1/4] mpage: add argument structure for do_mpage_readpage() Jens Axboe
2018-05-30 14:42 ` [PATCH 2/4] mpage: mpage_readpages() should submit IO as read-ahead Jens Axboe
2018-05-30 14:42 ` [PATCH 3/4] btrfs: readpages() " Jens Axboe
2018-05-30 14:42 ` [PATCH 4/4] ext4: " Jens Axboe
2018-06-19 23:56 ` [PATCHSET v3 0/4] Submit ->readpages() " Andrew Morton
2018-06-20 14:07   ` Jens Axboe
2018-06-20 19:58     ` Andrew Morton
2018-06-20 20:15       ` Chris Mason
2018-06-20 22:28       ` Jens Axboe
2018-06-20 23:23         ` Andrew Morton
2018-06-20 23:33           ` Jens Axboe
2018-06-20 23:45             ` Andrew Morton
2018-06-21  0:06               ` Jens Axboe
2018-06-21 12:21         ` Christoph Hellwig
2018-06-21 13:52           ` Jens Axboe
  -- strict thread matches above, loose matches on Subject: below --
2018-06-21  1:07 [PATCH v4 " Jens Axboe
2018-06-21  1:07 ` [PATCH 3/4] btrfs: readpages() should submit " Jens Axboe
2018-05-29 22:17 [PATCHSET v2 0/4] Submit ->readpages() " Jens Axboe
2018-05-29 22:17 ` [PATCH 3/4] btrfs: readpages() should submit " Jens Axboe

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).