RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-02-21 01:47:25 +00:00
Code Issues Packages Projects Releases Wiki Activity
98,258 Commits 216 Branches 395 Tags 1.3 GiB
a786cfd0ab
Commit Graph

98258 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Jason Dillaman
a786cfd0ab librbd: avoid bl manipulation on single object read/writes 
If a read or write only spans a single object and buffer extent,
there isn't any need to build a new bufferlist from a full sub-bl.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2019-05-13 13:30:09 -04:00
Jason Dillaman
cac640c564 librbd: switch to lightweight striper for hot IO path 
Optimize non-complex IO requests using the new lightweight object
striper that avoids heap allocations for small requests.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2019-05-13 13:30:09 -04:00
Jason Dillaman
f5a93d6802 librbd: drop data oid parameter from IO path 
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2019-05-13 13:30:09 -04:00
Jason Dillaman
10bebfcac0 librbd: helper method to generate data object OIDs on-demand 
The new generator avoids multiple string length and copy operations
that are currently required in the Striper.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2019-05-13 13:07:04 -04:00
Jason Dillaman
58bbb87c5c osdc/Striper: implemented a lightweight version of StripedReadResult 
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2019-05-13 13:07:04 -04:00
Jason Dillaman
a3448eda33 osdc/Striper: implemented a lightweight version of file_to_extents 
The previous versions of file_to_extents resulted in numerous heap
allocations for maps, vectors, and strings -- in addition to using
a CPU inefficient oid name formater.

librbd doesn't require the use of object_t and it would be better to
avoid any heap allocations on the IO path for the common case.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2019-05-13 12:38:13 -04:00
Jason Dillaman
cc7ec3e18d common/Finisher: avoid memory re-allocations for finisher queue 
Since there is only a single thread associated with the Finisher,
swap between two queues to avoid the need to re-allocate a new
vector for each iteration through the loop. Also replace the
condition broadcast with a signal since there is only a single
thread to wake up.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2019-05-13 12:38:12 -04:00
Jason Dillaman
99826fbe75
Merge pull request #27999 from trociny/wip-rbd-sparse_copyup 
librbd: clone copy-on-write operations should preserve sparseness

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2019-05-13 12:15:51 -04:00
Kefu Chai
5a79b4c822
Merge pull request #28071 from tchaikov/wip-crimson-perf-regres 
crimson,osd: performance fixes

Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
 2019-05-14 00:00:00 +08:00
Casey Bodley
782be351c8
Merge pull request #27270 from theanalyst/beast-fe-ipv6 
rgw: normalize v6 endpoint behaviour for the beast frontend

Reviewed-by: Matt Benjamin <mbenjamin@redhat.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2019-05-13 10:02:21 -04:00
Nathan Cutler
cfa2c5fb4d
Merge pull request #26281 from smithfarm/wip-git-describe-long 
make-dist: set version number only once

Reviewed-by: Tim Serong <tserong@suse.com>
 2019-05-13 15:10:20 +02:00
Casey Bodley
d39f6a5008
Merge pull request #28045 from cbodley/wip-39635 
cls/user: cls_user_set_buckets_info overwrites creation_time

Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
 2019-05-13 09:09:17 -04:00
Casey Bodley
f1f223d103
Merge pull request #27980 from cbodley/wip-24991 
rgw: delete_obj_index() takes mtime for bilog

Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>
 2019-05-13 09:07:04 -04:00
Casey Bodley
cd236e7a42
Merge pull request #28031 from ivancich/wip-rgw-admin-unordered 
rgw: allow radosgw-admin to list bucket w --allow-unordered

Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
 2019-05-13 09:05:29 -04:00
Casey Bodley
b59fcb0c95
Merge pull request #27676 from mkogan1/wip-rgw-39380 
rgw: fix default_placement containing "/" when storage_class is standard

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2019-05-13 08:53:58 -04:00
Casey Bodley
1ad2cb777d
Merge pull request #27984 from cbodley/wip-test-rgw-multisite-sh 
test/rgw: fix test-rgw-multisite.sh script for creating multisite clusters

Reviewed-by: Shilpa Jagannath <smanjara@redhat.com>
 2019-05-13 08:51:47 -04:00
Kefu Chai
e8da9e1e13
Merge pull request #27807 from cyb70289/utf8-encode 
common: optimize encode_utf8

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2019-05-13 20:42:24 +08:00
Mykola Golub
76e4f2ce24
Merge pull request #28051 from dillaman/wip-39521 
librbd: fix issues with object-map/fast-diff feature interlock

Reviewed-by: Mykola Golub <mgolub@suse.com>
 2019-05-13 15:25:09 +03:00
Mykola Golub
b2c799e7d0
Merge pull request #28063 from dillaman/wip-39656 
qa/workunits/rbd: use https protocol for devstack git operations

Reviewed-by: Mykola Golub <mgolub@suse.com>
 2019-05-13 15:24:39 +03:00
Kefu Chai
a09a4f342c cmake: define BOOST_SP_DISABLE_THREADS for crimson 
because we are single-threaded

Signed-off-by: Kefu Chai <kchai@redhat.com>
 2019-05-12 22:41:02 +08:00
Radoslaw Zarzynski
0291178ede osd/osd_internal_types: bring moving semantics to ObjectState. 
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
 2019-05-12 22:40:43 +08:00
Radoslaw Zarzynski
d2b49dae15 crimson/osd: simplify write handling in PG::do_osd_ops. 
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
 2019-05-12 22:40:35 +08:00
Mykola Golub
750e61ac91 librbd: clone copy-on-write operations should preserve sparseness 
Signed-off-by: Mykola Golub <mgolub@suse.com>
 2019-05-12 09:03:25 +01:00
J. Eric Ivancich
ce1df3ceef
Merge pull request #27870 from theanalyst/rgw-objexp-fixes-cli 
rgw: object expirer fixes

Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>
 2019-05-11 09:10:54 -04:00
Kefu Chai
93ba1a31df
Merge pull request #27926 from tchaikov/wip-38627 
pybind/mgr: convert str to int using "int()"

Reviewed-by: Noah Watkins <noahwatkins@gmail.com>
 2019-05-11 09:37:39 +08:00
David Zafman
624d1172c3
Merge pull request #27914 from dzafman/wip-39304 
osd: Include dups in copy_after() and copy_up_to()

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Neha Ojha <nojha@redhat.com>
 2019-05-10 17:41:31 -07:00
David Zafman
09fb7a5a12
Merge pull request #27961 from dzafman/wip-39582 
osd: Output Base64 encoding of CRC header if binary data present

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2019-05-10 15:46:16 -07:00
David Zafman
f526c6d55c osd: Fix merge_log_dups() not remove dup corresponding to log.tail 
The log tail is not inclusive of the log, so not an overlap
Fix unittest_pglog to verify proper operation

Signed-off-by: David Zafman <dzafman@redhat.com>
 2019-05-10 15:31:51 -07:00
David Zafman
d0d244d222 test: unittest_pglog: Additional test cases for trim, copy_up_to and copy_after 
Signed-off-by: David Zafman <dzafman@redhat.com>
 2019-05-10 15:31:51 -07:00
David Zafman
a008bd8c18 test: unittest_pglog: Remove irrelevant config values 
osd_min_pg_log_entries and osd_max_pg_log_entries are not involved
in the log handling code being tested.

Signed-off-by: David Zafman <dzafman@redhat.com>
 2019-05-10 15:31:51 -07:00
David Zafman
7959159e83 test: Adding standalone test of log copy handling 
Signed-off-by: David Zafman <dzafman@redhat.com>
 2019-05-10 15:31:51 -07:00
Sage Weil
62e241e4da Merge PR #27782 into master 
* refs/pull/27782/head:
	os/bluestore: protect BlueFS::FileReader::buf
	os/bluestore: log 'buffered' in KernelDevice::read_random
	os/bluestore: support RocksDB prefetching at BlueFS
	os/bluestore: call BlueFS::_init_logger earlier.
	tools/ceph-objectstore-tool: dump perf counters on competion.
	os/bluestore: optimize mapping offset to extent in BlueFS.

Reviewed-by: Sage Weil <sage@redhat.com>
 2019-05-10 16:29:35 -05:00
Sage Weil
5229e1ec7f Merge PR #27928 into master 
* refs/pull/27928/head:
	mon/OSDMonitor: require nautilus to set pg_autoscale_mode
	mon/OSDMonitor: allow pg_num to increase when require_osd_release < N

Reviewed-by: Neha Ojha <nojha@redhat.com>
 2019-05-10 16:22:40 -05:00
Sage Weil
4982bd58fc Merge PR #27877 into master 
* refs/pull/27877/head:
	mon: paxos: introduce new reset_pending_committing_finishers for safety

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2019-05-10 16:21:58 -05:00
Neha Ojha
de5c5d8468
Merge pull request #21722 from mslovy/wip-parial-recovery 
osd: partial recovery strategy based on PGLog

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Neha Ojha <nojha@redhat.com>
 2019-05-10 11:48:23 -07:00
Alfredo Deza
d362301306
Merge pull request #28060 from ceph/wip-bz1666822 
ceph-volume: use the Device.rotational property instead of sys_api

Reviewed-by: Alfredo Deza <adeza@redhat.com>
 2019-05-10 14:37:35 -04:00
Andrew Schoen
f1db0a8648 ceph-volume: skip missing interpreters when running tox tests 
This will skip tests for versions of python that don't exist on the node
that's running the tests.

Signed-off-by: Andrew Schoen <aschoen@redhat.com>
 2019-05-10 11:04:35 -05:00
Andrew Schoen
8ebff475df ceph-volume: use the Device.rotational property instead of sys_api 
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1666822

Signed-off-by: Andrew Schoen <aschoen@redhat.com>
 2019-05-10 10:41:49 -05:00
Jason Dillaman
fb4f9a8a08 qa/workunits/rbd: use https protocol for devstack git operations 
Fixes: http://tracker.ceph.com/issues/39656
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2019-05-10 11:09:54 -04:00
Sage Weil
d2c591a142 mon/OSDMonitor: require nautilus to set pg_autoscale_mode 
Signed-off-by: Sage Weil <sage@redhat.com>
 2019-05-10 09:31:03 -05:00
Sage Weil
7e7134aa93 mon/OSDMonitor: allow pg_num to increase when require_osd_release < N 
Setting pg_num_target has no effect when we encode our OSDMap with mimic
or luminous features.  Fall back to the pre-nautilus behavior of directly
setting pg_num.

Fixes: http://tracker.ceph.com/issues/39570
Signed-off-by: Sage Weil <sage@redhat.com>
 2019-05-10 09:31:03 -05:00
Casey Bodley
8107bc6aef
Merge pull request #27795 from cbodley/wip-rgw-beast-async-body 
rgw: beast frontend uses yield_context to read/write body

Reviewed-by: Adam C. Emerson <aemerson@redhat.com>
Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
 2019-05-10 09:05:32 -04:00
Lenz Grimmer
e7ad5ed68d
mgr/dashboard: Added Object Gateway menu tab tests (#27565) 
mgr/dashboard: Added Object Gateway menu tab tests

Reviewed-by: Laura Paduano <lpaduano@suse.com>
Reviewed-by: Tatjana Dehler <tdehler@suse.com>
 2019-05-10 12:41:09 +02:00
Nathan Cutler
ae4f94c868
Merge pull request #28042 from smithfarm/wip-doc-ceph-backport-sh 
doc: improve ceph-backport.sh comment block

Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
 2019-05-10 11:27:30 +02:00
Nathan Cutler
0b6c121805 doc: improve ceph-backport.sh comment block 
Signed-off-by: Nathan Cutler <ncutler@suse.com>
 2019-05-10 10:22:50 +02:00
Samuel Just
d20076b7b4
Merge pull request #28048 from athanatos/sjust/wip-peering-state-cleanup 
More PeeringState and related cleanups to ease use in crimson

Reviewed-by: Neha Ojha <nojha@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2019-05-09 20:34:06 -07:00
Sage Weil
5cca06aab8 Merge PR #27811 into master 
* refs/pull/27811/head:
	commom: make cluster_network work.

Reviewed-by: Sage Weil <sage@redhat.com>
 2019-05-09 20:42:16 -05:00
David Zafman
00f279db37 osd: Include dups in copy_after() and copy_up_to() 
Client saw out of order results in a test with a very small
pg_log because a backfill/recovery sent a small number of
log entries for duplicate checks.

Since these copy operations are equivalent to both a copy
and trim (an argument controls how many log entries
transfer), we need to include any new dups with enough
existing dups up to the configured maximum.

Fixes: http://tracker.ceph.com/issues/39304

Signed-off-by: David Zafman <dzafman@redhat.com>
 2019-05-09 17:50:27 -07:00
Xie Xingguo
a14545d592
Merge pull request #28050 from dillaman/wip-39448 
msg/async: reduce verbosity of connection timeout failures

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
 2019-05-10 08:39:15 +08:00
sjust@redhat.com
0d7115584e osd/: Move PG::_prepare_write_info into osd_types prepare_info_keymap 
Signed-off-by: Samuel Just <sjust@redhat.com>
 2019-05-09 17:22:34 -07:00