RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2024-12-27 14:03:25 +00:00
Code Issues Packages Projects Releases Wiki Activity
55,076 Commits 211 Branches 394 Tags 1.4 GiB
86944f0ea1
Commit Graph

55076 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
xie xingguo
86944f0ea1 os/bluestore: add sanity check for underflow 
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-23 20:09:01 +08:00
xie xingguo
319a14e201 os/bluestore: use hex version for offset/length 
We update bluestore this way a couple of days ago.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-23 20:07:25 +08:00
xie xingguo
ed71c3ed00 os/bluestore: kill dead code 
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-23 20:07:25 +08:00
xie xingguo
9e35ac192e os/bluestore: shutdown allocator of bluefs more gracefully 
The new BitAllocator requires we call shutdown() explicitly,
thus we shall fix it here before we apply to the new BitAllocator.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-23 20:07:25 +08:00
Jason Dillaman
48d18030fd Merge pull request #9207 from rjfd/wip-15670 
rbd-mirror: image resync

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-22 18:03:22 -04:00
Sage Weil
c83374ff71 Merge pull request #9626 from liewegas/wip-bluestore-2q 
os/bluestore: implement 2Q buffer cache policy
 2016-06-22 15:09:23 -04:00
Sage Weil
b6a2492b1c ceph_test_objectstore: define matrixes over config options and write sizes 
These replace a few of the ad hoc synthetic tests.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 15:08:36 -04:00
Sage Weil
3064c819fd os/bluestore: handle bnode removal race 
One thread might try to drop a bnode while another does a lookup and
retakes a ref:

 1- drop bnode ref (-> 0)
 2- take bnode_set lock, do lookup
 2- take bnode ref (-> 1)
 2- drop lock
 1- take bnode_set lock
 1- remove from set
 1- drop lock

Fix this race by rechecking the ref count while we are holding the lock.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 15:08:36 -04:00
Samuel Just
73e67eb6f6 Revert "msg: default ms type = async" 
Not ready quite yet, causing too many test failures.

This reverts commit 928887a09c.
Reviewed-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:21:24 -07:00
Jason Dillaman
070bb07874 Merge pull request #9451 from s09816/master 
librbd: enable/disable of features is not allowed when already enabled/disabled

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-22 14:02:42 -04:00
Jason Dillaman
2b7f28c234 Merge pull request #9715 from trociny/wip-16321 
librbd: re-register watch on old format image rename

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-22 14:02:01 -04:00
Samuel Just
8833e8e008 Merge pull request #9874 from athanatos/wip-16185 
OSDMonitor::prepare_pgtemp: only update up_thru if newer

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-06-22 10:26:49 -07:00
Samuel Just
55d5392aa3 Merge pull request #9875 from athanatos/wip-16127 
OSDMonitor: drop pg temp from sources other than the current primary

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-06-22 10:26:35 -07:00
Mykola Golub
74c5da8b24 Merge pull request #9850 from dillaman/wip-16404 
librbd: flag image as updated after proxying maintenance op

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
 2016-06-22 20:12:59 +03:00
Samuel Just
d68b3a12ad Merge pull request #8370 from jack-changtao/pg_del 
osd/PG: cleanup: delete find_best_info again

Reviewed-by: Samuel Just <sjust@redhat.com>
 2016-06-22 09:55:23 -07:00
Samuel Just
cab168e5fc Merge pull request #9802 from yuyuyu101/wip-16378 
msg/async: fix test stuck into STATE_WAIT

Reviewed-by: Samuel Just <sjust@redhat.com>
 2016-06-22 09:51:01 -07:00
Kefu Chai
b20af78f06 Merge pull request #8195 from tchaikov/wip-cmake-visibility 
cmake changes

Reviewed-by: Ali Maredia <amaredia@redhat.com>
 2016-06-22 23:37:38 +08:00
Sage Weil
7c6bdb0e21 os/bluestore: drop unneeded conditional in LRUCache trim 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:43 -04:00
Sage Weil
2215e227a9 os/bluestore: protect BnodeSet with mutex 
The onodes (and thus bnodes) may not get destroyed with the collection
lock held, which means we cannot rely on that lock to protected the
bnodeset structure.  Give BnodeSet it's own mutex to protect itself.

While we are at it, make dummy a member so we can avoid constructing it
each time.  This is now safe since we are under the new lock.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:42 -04:00
Sage Weil
c28188ea3c os/bluestore: remove bluestore_cache_tails 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:42 -04:00
Sage Weil
9ebbca0375 os/bluestore: fix 8-bit csum 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:42 -04:00
Sage Weil
d2023cf3b4 os/bluestore: default to 512MB cache, buffered reads 
Test with small cache to exercise trimming.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:42 -04:00
Sage Weil
f1d7a70614 os/bluestore: 2q by default 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:42 -04:00
Sage Weil
b03d55224a os/bluestore: implement 2Q buffer cache policy 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:42 -04:00
Sage Weil
fb03a64358 os/bluestore: move merge operator 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:41 -04:00
Sage Weil
98748b3623 os/bluestore: buffer cache_private field, returned by discard 
2Q (and likly MQ?) need to know when we re-add a block that was previously
evicted.  We track these as STATE_EMPTY buffers, but we throw them out
with _discard before we add the (re)read data into the BufferSpace. Add
a Buffer field called cache_private that is returned by discard (actually,
a max over everything that is discarded) that a cache implementation can
use to discover it has re-read a recently evicted item.

The Cache implemention needs to set cache_private values such that the MAX
over them is what it actually wants...

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:41 -04:00
Sage Weil
fc6e11fedf os/bluestore: include buffer * in operator<< 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:41 -04:00
Sage Weil
fda7a08106 os/bluestore: cache: simplify lru trim 
Everything in the LRU is clean.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:41 -04:00
Sage Weil
ecc168fca3 os/bluestore: make Cache implementation configurable 
For now we only do "lru" (LRUCache).

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:41 -04:00
Sage Weil
ba0a7c4027 os/bluestore: cache: define 'empty' buffer 
This is a buffer with no data.  It's a placeholder that is there to be used
by cache replacement algorithms like 2Q and MQ that needs history for
trimmed buffers.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:40 -04:00
Sage Weil
7d9629c07e os/bluestore: cache: keep writing buffers out of LRU 
Don't bother putting a writing buffer in the LRU until it is clean (and
trimmable).  Otherwise, we end up with writing items near the end of the
LRU that we have to iterate over on every trim() invocation.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:40 -04:00
Sage Weil
b55255d23e os/bluestore: clean up Cache::trim 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:40 -04:00
Sage Weil
94648e2bd5 os/bluestore: cache: control insert point of new onodes 
Not used yet.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:40 -04:00
Sage Weil
f45e4a7d6d os/bluestore: cache: control insertion point of new buffers 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:40 -04:00
Sage Weil
3c3dcc63d8 os/bluestore: make buffer lru private 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:40 -04:00
Sage Weil
5f0bb6de23 os/bluestore: make onode lru private 
Add cache->onode_lru.push_back(*po->second).  Make LRU private.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:40 -04:00
Sage Weil
2f6546dd43 os/bluestore: fix locking for _dump_blob_map 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 11:28:40 -04:00
Kefu Chai
4d3de44385 cmake: use ${CEPH_SHARED} when adding libradosstriper 
Signed-off-by: Kefu Chai <kchai@redhat.com>
 2016-06-22 23:21:46 +08:00
Sage Weil
db70056844 Merge pull request #9827 from liewegas/wip-bluestore-invalidate 
os/bluestore: discard unreferenced buffers as they are overwritten

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
 2016-06-22 10:56:02 -04:00
Sage Weil
90ea1a8d13 Merge pull request #9868 from liewegas/wip-bluestore-lba 
include/small_encoding: fix lba encoding with many low zeros

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
 2016-06-22 10:55:19 -04:00
Jason Dillaman
dcf26b00db Merge pull request #9863 from trociny/fixup-comp-warn 
test: fix compilation warnings

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-22 10:43:41 -04:00
Kefu Chai
4e911c8c8b Merge pull request #9821 from dx9/wip-ac-alpine 
configure.ac: add --with-reentrant-strsignal and --with-thread-safe-r…

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2016-06-22 22:42:46 +08:00
Sage Weil
73796fe0a9 os/bluestore: discard unreferenced buffers as they are overwritten 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 10:30:27 -04:00
Sage Weil
3c8c882c31 include/small_encoding: fix lba encoding with many low zeros 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 10:28:51 -04:00
Haomai Wang
86f76f96df msg/async: recover lock to continue later 
Signed-off-by: Haomai Wang <haomai@xsky.com>
 2016-06-22 22:26:46 +08:00
Sage Weil
6dd0a8dcd9 Merge pull request #9865 from liewegas/wip-bluestore-encoding 
os/bluestore: fix blob_t encoding test cases, and clean up initialization

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
 2016-06-22 10:19:18 -04:00
John Coyle
6b8ff0c61c configure.ac: add --with-reentrant-strsignal and --with-thread-safe-res-query options for Alpine support 
Signed-off-by: John Coyle <dx9err@gmail.com>
 2016-06-22 10:15:57 -04:00
Mykola Golub
64b9246b7e test: fix compilation warnings 
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
 2016-06-22 16:41:06 +03:00
Sage Weil
a7749eda96 os/bluestore/bluestore_types: fix blob initializers, test instances 
- initialize members in c++11 style
- drop pextent ctor
- fix up test instances
- make csum_order default to 0, so it matches the
  CSUM_NONE decoding logic.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 09:23:31 -04:00
Sage Weil
0e4008f72e os/bluestore/bluestore_types: drop unused var 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-06-22 08:41:05 -04:00