* [GIT PULL] MD update for 4.11
@ 2017-02-24 15:52 Shaohua Li
2017-02-24 18:36 ` Shaohua Li
0 siblings, 1 reply; 2+ messages in thread
From: Shaohua Li @ 2017-02-24 15:52 UTC (permalink / raw)
To: torvalds; +Cc: linux-kernel, linux-raid, neilb
Hi,
please pull MD update for 4.11, this pull mainly fixes bugs and improves
performance:
- Improve scalability for raid1 from Coly
- Improve raid5-cache read performance, disk efficiency and IO pattern from
Song and me
- Fix a race condition of disk hotplug for linear from Coly
- A few cleanup patches from Ming and Byungchul
- Fix a memory leak from Neil
- Fix WRITE SAME IO failure from me
- Add doc for raid5-cache from me
Thanks,
Shaohua
The following changes since commit 7089db84e356562f8ba737c29e472cc42d530dbc:
Linux 4.10-rc8 (2017-02-12 13:03:20 -0800)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/shli/md.git for-next
for you to fetch changes up to 1ec492232ed659acde8cc00b9ecc7529778e03e1:
md/raid1: fix write behind issues introduced by bio_clone_bioset_partial (2017-02-23 11:59:44 -0800)
----------------------------------------------------------------
Byungchul Park (1):
md/raid5: Don't reinvent the wheel but use existing llist API
Ming Lei (5):
block: introduce bio_clone_bioset_partial()
md: fail if mddev->bio_set can't be created
md/raid1: use bio_clone_bioset_partial() in case of write behind
md: remove unnecessary check on mddev
md: fast clone bio in bio_clone_mddev()
NeilBrown (1):
md: ensure md devices are freed before module is unloaded.
Shaohua Li (10):
raid5: only dispatch IO from raid5d for harddisk raid
Documentation: move MD related doc into a separate dir
MD: add doc for raid5-cache
md/raid5-cache: stripe reclaim only counts valid stripes
md/raid5-cache: exclude reclaiming stripes in reclaim check
md: disable WRITE SAME if it fails in underlayer disks
md/raid1: fix a use-after-free bug
md/linear: shutup lockdep warnning
md/raid1: handle flush request correctly
md/raid1: fix write behind issues introduced by bio_clone_bioset_partial
Song Liu (3):
EXPORT_SYMBOL radix_tree_replace_slot
md/r5cache: enable chunk_aligned_read with write back cache
md/r5cache: improve journal device efficiency
colyli@suse.de (3):
md linear: fix a race between linear_add() and linear_congested()
RAID1: a new I/O barrier implementation to remove resync window
RAID1: avoid unnecessary spin locks in I/O barrier code
Documentation/00-INDEX | 4 +-
Documentation/admin-guide/md.rst | 5 +
Documentation/{ => md}/md-cluster.txt | 0
Documentation/md/raid5-cache.txt | 109 +++++++
block/bio.c | 61 +++-
drivers/md/faulty.c | 2 +-
drivers/md/linear.c | 41 ++-
drivers/md/linear.h | 1 +
drivers/md/md.c | 22 +-
drivers/md/md.h | 9 +-
drivers/md/multipath.c | 1 +
drivers/md/raid0.c | 1 +
drivers/md/raid1.c | 598 +++++++++++++++++++++-------------
drivers/md/raid1.h | 58 ++--
drivers/md/raid10.c | 11 +-
drivers/md/raid5-cache.c | 225 +++++++++++--
drivers/md/raid5.c | 129 ++++++--
drivers/md/raid5.h | 7 +
include/linux/bio.h | 11 +-
lib/radix-tree.c | 1 +
20 files changed, 943 insertions(+), 353 deletions(-)
rename Documentation/{ => md}/md-cluster.txt (100%)
create mode 100644 Documentation/md/raid5-cache.txt
^ permalink raw reply [flat|nested] 2+ messages in thread* Re: [GIT PULL] MD update for 4.11
2017-02-24 15:52 [GIT PULL] MD update for 4.11 Shaohua Li
@ 2017-02-24 18:36 ` Shaohua Li
0 siblings, 0 replies; 2+ messages in thread
From: Shaohua Li @ 2017-02-24 18:36 UTC (permalink / raw)
To: torvalds; +Cc: linux-kernel, linux-raid, neilb
On Fri, Feb 24, 2017 at 07:52:06AM -0800, Shaohua Li wrote:
> Hi,
>
> please pull MD update for 4.11, this pull mainly fixes bugs and improves
> performance:
> - Improve scalability for raid1 from Coly
> - Improve raid5-cache read performance, disk efficiency and IO pattern from
> Song and me
> - Fix a race condition of disk hotplug for linear from Coly
> - A few cleanup patches from Ming and Byungchul
> - Fix a memory leak from Neil
> - Fix WRITE SAME IO failure from me
> - Add doc for raid5-cache from me
I forgot to mention there is a merge conflict because of a last minute fix.
Below is the fix. Sorry about this.
Thanks,
Shaohua
diff --cc drivers/md/raid1.c
index 830ff2b,d4e8796..0000000
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@@ -1385,8 -1515,7 +1512,8 @@@ static void raid1_write_request(struct
conf->mirrors[i].rdev->data_offset);
mbio->bi_bdev = conf->mirrors[i].rdev->bdev;
mbio->bi_end_io = raid1_end_write_request;
- bio_set_op_attrs(mbio, op, do_fua | do_sync);
+ mbio->bi_opf = bio_op(bio) |
- (bio->bi_opf & (REQ_SYNC | REQ_PREFLUSH | REQ_FUA));
++ (bio->bi_opf & (REQ_SYNC | REQ_FUA));
if (test_bit(FailFast, &conf->mirrors[i].rdev->flags) &&
!test_bit(WriteMostly, &conf->mirrors[i].rdev->flags) &&
conf->raid_disks - mddev->degraded > 1)
>
> Thanks,
> Shaohua
>
> The following changes since commit 7089db84e356562f8ba737c29e472cc42d530dbc:
>
> Linux 4.10-rc8 (2017-02-12 13:03:20 -0800)
>
> are available in the git repository at:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/shli/md.git for-next
>
> for you to fetch changes up to 1ec492232ed659acde8cc00b9ecc7529778e03e1:
>
> md/raid1: fix write behind issues introduced by bio_clone_bioset_partial (2017-02-23 11:59:44 -0800)
>
> ----------------------------------------------------------------
> Byungchul Park (1):
> md/raid5: Don't reinvent the wheel but use existing llist API
>
> Ming Lei (5):
> block: introduce bio_clone_bioset_partial()
> md: fail if mddev->bio_set can't be created
> md/raid1: use bio_clone_bioset_partial() in case of write behind
> md: remove unnecessary check on mddev
> md: fast clone bio in bio_clone_mddev()
>
> NeilBrown (1):
> md: ensure md devices are freed before module is unloaded.
>
> Shaohua Li (10):
> raid5: only dispatch IO from raid5d for harddisk raid
> Documentation: move MD related doc into a separate dir
> MD: add doc for raid5-cache
> md/raid5-cache: stripe reclaim only counts valid stripes
> md/raid5-cache: exclude reclaiming stripes in reclaim check
> md: disable WRITE SAME if it fails in underlayer disks
> md/raid1: fix a use-after-free bug
> md/linear: shutup lockdep warnning
> md/raid1: handle flush request correctly
> md/raid1: fix write behind issues introduced by bio_clone_bioset_partial
>
> Song Liu (3):
> EXPORT_SYMBOL radix_tree_replace_slot
> md/r5cache: enable chunk_aligned_read with write back cache
> md/r5cache: improve journal device efficiency
>
> colyli@suse.de (3):
> md linear: fix a race between linear_add() and linear_congested()
> RAID1: a new I/O barrier implementation to remove resync window
> RAID1: avoid unnecessary spin locks in I/O barrier code
>
> Documentation/00-INDEX | 4 +-
> Documentation/admin-guide/md.rst | 5 +
> Documentation/{ => md}/md-cluster.txt | 0
> Documentation/md/raid5-cache.txt | 109 +++++++
> block/bio.c | 61 +++-
> drivers/md/faulty.c | 2 +-
> drivers/md/linear.c | 41 ++-
> drivers/md/linear.h | 1 +
> drivers/md/md.c | 22 +-
> drivers/md/md.h | 9 +-
> drivers/md/multipath.c | 1 +
> drivers/md/raid0.c | 1 +
> drivers/md/raid1.c | 598 +++++++++++++++++++++-------------
> drivers/md/raid1.h | 58 ++--
> drivers/md/raid10.c | 11 +-
> drivers/md/raid5-cache.c | 225 +++++++++++--
> drivers/md/raid5.c | 129 ++++++--
> drivers/md/raid5.h | 7 +
> include/linux/bio.h | 11 +-
> lib/radix-tree.c | 1 +
> 20 files changed, 943 insertions(+), 353 deletions(-)
> rename Documentation/{ => md}/md-cluster.txt (100%)
> create mode 100644 Documentation/md/raid5-cache.txt
> --
> To unsubscribe from this list: send the line "unsubscribe linux-raid" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2017-02-24 18:36 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-02-24 15:52 [GIT PULL] MD update for 4.11 Shaohua Li
2017-02-24 18:36 ` Shaohua Li
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).