summaryrefslogtreecommitdiffstats
path: root/Documentation
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2013-12-13 16:22:22 -0500
committerLinus Torvalds <torvalds@linux-foundation.org>2013-12-13 16:22:22 -0500
commit93e1585e2c909662e97c59fe0204f93fcf985f04 (patch)
tree8c4685fa040f5a2b474e9981791b1f4586918ed5 /Documentation
parent1008ebb61e01e3152901b4c5f58bd01a60ed115b (diff)
parented9571f0cf1fe09d3506302610f3ccdfa1d22c4a (diff)
Merge tag 'dm-3.13-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm
Pull device mapper fixes from Mike Snitzer: "A set of device-mapper fixes for 3.13. A fix for possible memory corruption during DM table load, fix a possible leak of snapshot space in case of a crash, fix a possible deadlock due to a shared workqueue in the delay target, fix to initialize read-only module parameters that are used to export metrics for dm stats and dm bufio. Quite a few stable fixes were identified for both the thin- provisioning and caching targets as a result of increased regression testing using the device-mapper-test-suite (dmts). The most notable of these are the reference counting fixes for the space map btree that is used by the dm-array interface -- without these the dm-cache metadata will leak, resulting in dm-cache devices running out of metadata blocks. Also, some important fixes related to the thin-provisioning target's transition to read-only mode on error" * tag 'dm-3.13-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: dm array: fix a reference counting bug in shadow_ablock dm space map: disallow decrementing a reference count below zero dm stats: initialize read-only module parameter dm bufio: initialize read-only module parameters dm cache: actually resize cache dm cache: update Documentation for invalidate_cblocks's range syntax dm cache policy mq: fix promotions to occur as expected dm thin: allow pool in read-only mode to transition to read-write mode dm thin: re-establish read-only state when switching to fail mode dm thin: always fallback the pool mode if commit fails dm thin: switch to read-only mode if metadata space is exhausted dm thin: switch to read only mode if a mapping insert fails dm space map metadata: return on failure in sm_metadata_new_block dm table: fail dm_table_create on dm_round_up overflow dm snapshot: avoid snapshot space leak on crash dm delay: fix a possible deadlock due to shared workqueue
Diffstat (limited to 'Documentation')
-rw-r--r--Documentation/device-mapper/cache.txt10
1 files changed, 6 insertions, 4 deletions
diff --git a/Documentation/device-mapper/cache.txt b/Documentation/device-mapper/cache.txt
index 274752f8bdf9..719320b5ed3f 100644
--- a/Documentation/device-mapper/cache.txt
+++ b/Documentation/device-mapper/cache.txt
@@ -266,10 +266,12 @@ E.g.
266Invalidation is removing an entry from the cache without writing it 266Invalidation is removing an entry from the cache without writing it
267back. Cache blocks can be invalidated via the invalidate_cblocks 267back. Cache blocks can be invalidated via the invalidate_cblocks
268message, which takes an arbitrary number of cblock ranges. Each cblock 268message, which takes an arbitrary number of cblock ranges. Each cblock
269must be expressed as a decimal value, in the future a variant message 269range's end value is "one past the end", meaning 5-10 expresses a range
270that takes cblock ranges expressed in hexidecimal may be needed to 270of values from 5 to 9. Each cblock must be expressed as a decimal
271better support efficient invalidation of larger caches. The cache must 271value, in the future a variant message that takes cblock ranges
272be in passthrough mode when invalidate_cblocks is used. 272expressed in hexidecimal may be needed to better support efficient
273invalidation of larger caches. The cache must be in passthrough mode
274when invalidate_cblocks is used.
273 275
274 invalidate_cblocks [<cblock>|<cblock begin>-<cblock end>]* 276 invalidate_cblocks [<cblock>|<cblock begin>-<cblock end>]*
275 277