Skip to content

dm thin: address a few fundamental problems

Mike Snitzer dm-devel 1 version NAK

Status from this message: NACK

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest8 patches 14 repliesNAK

Patches in v1

#Subject and filesSizeRepliesApplied
1/8dm thin: synchronize the pool mode during suspenddrivers/md/dm-thin.c+6 -01 reply
2/8dm thin: set flag when over the metadata low watermark thresholddrivers/md/dm-thin.c drivers/md/persistent-data/dm-space-map-metadata.c+11 -193 replies
3/8dm thin: set flag if metadata is out of spacedrivers/md/dm-thin-metadata.c drivers/md/dm-thin-metadata.h drivers/md/dm-thin.c+42 -63 replies
4/8dm thin: error out I/O if inappropriate for it to be retriedDocumentation/device-mapper/cache.txt Documentation/device-mapper/thin-provisioning.txt drivers/md/dm-thin.c+41 -72 replies
5/8dm thin: fix the error path for the thin device constructordrivers/md/dm-thin.c+4 -10 replies
6/8dm thin: fix pool_preresume resize with heavy IO racesdrivers/md/dm-thin-metadata.c drivers/md/dm-thin-metadata.h drivers/md/dm-thin.c+112 -133 replies
7/8dm thin: ensure user takes action to validate data and metadata consistencyDocumentation/device-mapper/thin-provisioning.txt drivers/md/dm-thin-metadata.c drivers/md/dm-thin-metadata.h +1 more+102 -291 reply
8/8dm thin: allow metadata space larger than supported to go unuseddrivers/md/dm-cache-metadata.c drivers/md/dm-thin-metadata.c drivers/md/dm-thin-metadata.h +3 more+47 -311 reply

Reviews trailers per patch

1 of 8 patches has a Reviewed-by or Acked-by.

PatchJoe Thornber
1/8
2/8
3/8
4/8
5/8A-b
6/8
7/8
8/8

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v1 thread

Cover letter [PATCH for-dm-3.14-fixes 0/8] dm thin: address a few fundamental problems

This patchset fixes various thinp issues. Aside from fixing some bugs in earlier 3.14-rc1 code (as noted with commit references): the most significant issue addressed is that until now we didn't have the awareness or ability to force the user to perform consistency checks on the thin pool's metadata and data if warranted.

While there is a fair amount of change here I feel pretty strongly that these fixes should get upstream ASAP. Posting for review with the hope that I'll get timely feedback/acks so that I can send to Linus early next week.

(Alasdair, I went with "needs check" for the flag used in patch 7 even though you initially disagreed with the name.. it is short and to the point, and really: it is internal to the thin implementation. Userspace can spin it however you'd like).

Mike Snitzer (8):
dm thin: synchronize the pool mode during suspend
dm thin: set flag when over the metadata low watermark threshold dm thin: set flag if metadata is out of space
dm thin: error out I/O if inappropriate for it to be retried
dm thin: fix the error path for the thin device constructor
dm thin: fix pool_preresume resize with heavy IO races
dm thin: ensure user takes action to validate data and metadata consistency dm thin: allow metadata space larger than supported to go unused

 Documentation/device-mapper/cache.txt              |  11 +-
 Documentation/device-mapper/thin-provisioning.txt  |  28 +++
 drivers/md/dm-cache-metadata.c                     |   2 +-
 drivers/md/dm-thin-metadata.c                      | 118 ++++++++++-
 drivers/md/dm-thin-metadata.h                      |  23 ++-
 drivers/md/dm-thin.c                               | 225 ++++++++++++++++-----
 drivers/md/persistent-data/dm-space-map-metadata.c |  19 +-
 .../md/persistent-data/dm-transaction-manager.c    |  13 +-
 .../md/persistent-data/dm-transaction-manager.h    |   2 +-
 9 files changed, 350 insertions(+), 91 deletions(-)
Signature · 1 line
1.8.3.1