os/bluestore: _txc_release_alloc when do wal cleaning

Signed-off-by: Xinze Chi <xinze@xsky.com>

Merge pull request #12336 from kylinstorage/wip-cleanup-use-string-ref

osd: cleanup: use string & to avoid unnecessary copy

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12569 from liewegas/wip-bluestore-raw-unused

os/bluestore/bluestore_types: drop std::bitset for blob unused

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
Reviewed-by: Varada Kari <varada.kari@sandisk.com>

Merge pull request #12612 from fangyuxiangGL/master

osd/ReplicatedBackend: do not set omap header if it is empty

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Haomai Wang <haomai@xsky.com>

Merge pull request #12668 from hjwsm1989/pg-unfound

osd/PG: add new have_unfound() function in MissingLoc

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12674 from Liuchang0812/tools-rados-offset-support

tools/rados: add a parameter "--offset" to rados put command

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12675 from Liuchang0812/fix-test-case-in-rados

test: replace hard-code binary names with varibles

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12655 from ceph/wip-fix-fio

test/fio_ceph_objectstore: fix fio plugin build failure caused by rec…

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12682 from renhwztetecs/renhw-wip-mon-osdmonitor-sortbitwise

mon/OSDMonitor: fixup sortbitwise flag warning

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12654 from yuyuyu101/wip-fix-async-subop

messages/MOSDSubOp: Make encode_payload can be reentrant

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12662 from Liuchang0812/doc-fix

doc: fix document about rados mon

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12198 from tchaikov/wip-pybind-freebsd

pybind: conditional compile the linux specific constant

Reviewed-by: Willem Jan Withagen <wjw@digiware.nl>

mon/OSDMonitor: fixup sortbitwise flag warning

"ceph -s" does not report warning when using
command "ceph osd unset sortbitwise" to drop
sortbitwise flag.
we should use "osdmap.get_up_osd_features() &
CEPH_FEATURE_OSD_BITWISE_HOBJ_SORT"
instead of "(osdmap.get_features(CEPH_ENTITY_TYPE_OSD, NULL) &
CEPH_FEATURE_OSD_BITWISE_HOBJ_SORT)",
because osdmap.get_features only get local "features"

Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>

ReplicatedBackend: do not set omap header if it is empty

In replicatedpg backfill, objects have no omap header or keys previously, but an empty omap header will be inserted when they are pushed to replica, which leads leveldb performance degrade seriously when cluster has huge number of objects.
so my opinion is that we does't insert omap header if is size is zero when handle PushOp in recovery.

Signed-off-by: fang yuxiang <fang.yuxiang@eisoo.com>

tools/rados: add parameter offset for put cmd and test case for it

add a parameter "--offset" for put command
add a test case for rados put command

Signed-off-by: liuchang0812 <liuchang0812@gmail.com>

doc: fix document about rados mon

Signed-off-by: liuchang0812 <liuchang0812@gmail.com>

Merge pull request #12663 from badone/wip-ceph_test_rados_api_c_write_operations-valgrind-fixes

c_write_operations.cc: Fix trivial memory leak

Reviewed-by: Kefu Chai <kchai@redhat.com>

test: replace hard-code binary names with varibles

This test case couldn't run under customize env CEPH_BIN.

Signed-off-by: liuchang0812 <liuchang0812@gmail.com>

Merge pull request #12672 from liupan1111/wip-fix-nvme-driver-compile-issue

bluestore: fixed compilation error when enable spdk

Reviewed-by: Haomai Wang <haomai@xsky.com>

bluestore: fixed compilation error when enable spdk

Signed-off-by: Pan Liu <pan.liu@istuary.com>

Merge pull request #12659 from kylinstorage/wip-remove-obsolete-configuration-option

common/config_opts.h: remove obsolete configuration option

Reviewed-by: Sage Weil <sage@redhat.com>

  osd/PG: add new have_unfound() function in MissingLoc
          No need to lookup all objects in needs_recovery_map if only want to
          know whether if there are unfound objects.

Signed-off-by: huangjun <hjwsm1989@gmail.com>

Merge pull request #12648 from Adirl/fix_RoCEv2_deafult_value

msg/async/rdma: fix RoCE v2 deafult value

Reviewed-by: Haomai Wang <haomai@xsky.com>

c_write_operations.cc: Fix trivial memory leak

Signed-off-by: Brad Hubbard <bhubbard@redhat.com>

Merge pull request #12657 from badone/wip-c_read_operations-valgrind-fixes

test/librados/c_read_operations.cc: Fix valgrind errors

Reviewed-by: liuchang0812 <liuchang0812@gmail.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12660 from liupan1111/wip-fix-dpdk-compile

async: Fixed compilation error when enable -DWITH_DPDK

Reviewed-by: Haomai Wang <haomai@xsky.com>

async: Fixed compilation error when enable -DWITH_DPDK

Signed-off-by: Pan Liu <pan.liu@istuary.com>

test/librados/c_read_operations.cc: Fix valgrind errors

Fixes: http://tracker.ceph.com/issues/18354
Signed-off-by: Brad Hubbard <bhubbard@redhat.com>

common/config_opts.h: remove obsolete configuration option

Signed-off-by: Li Wang <li.wang@kylin-cloud.com>

Merge pull request #12474 from cbodley/wip-rgw-test-versioned

test/rgw: add test for versioned object sync
Reviewed-by: Orit wasserman <owasserm@redhat.com>

Merge pull request #12597 from badone/wip-osdmap-nocopy

osd/OSDMap: Uncomment code to enable private default constructors

Reviewed-by: Sage Weil <sage@redhat.com>

test/fio_ceph_objectstore: fix fio plugin build failure caused by recent g_ceph_context removal

Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>

Merge pull request #12642 from theanalyst/doc/release-notes/11.1.1

doc: release notes for v11.1.1

Merge pull request #12582 from Liuchang0812/fix-store-version-bug

os/filestore: version will be uninitialized varible if store_version doesn't exist

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12581 from Liuchang0812/fix-clang-warn-filestore

os/filestore: fix clang static check warn "use-after-free“

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12517 from hjwsm1989/wip-18266

messages/MOSDPGTrim: add the missed HEAD_VERSION AND COMPAT_VERSION

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12518 from kylinstorage/wip-safe-read-cmd

common/admin-socket: fix potential buffer overflow

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12653 from wjwithagen/wjw-wip-denc-raw

src/test/test_denc.cc: Fix errors in buffer overflow

Reviewed-by: Brad Hubbard <bhubbard@redhat.com>

Merge pull request #12469 from bassam/pr-jemalloc-mallinfo

cmake: disable mallinfo for jemalloc

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12468 from bassam/pr-simplify-jemalloc

cmake: simplify find_package jemalloc

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12519 from hjwsm1989/pgbackfill

osd/PG: simplify the logic of backfill_targets checking

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12585 from liewegas/wip-notify-ack

osd/osd_types: print notify-ack op properly

Merge pull request #12563 from xiexingguo/xxg-wip-osd-1217

common: make attempts of auth rotating configurable

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #11065 from LiumxNL/wip-fix-osdmap-update

mon/OSDMonitor: should propose osdmap update when cluster addr changed

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12634 from liewegas/wip-bluestore-shard-keys

os/bluestore: do not cache shard keys

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
Reviewed-by: Allen Samuels <allen.samuels@sandisk.com>

Merge pull request #12635 from liewegas/wip-bluestore-cache

os/bluestore: set cache meta ratio to .9

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>

Merge pull request #12644 from xiexingguo/xxg-wip-bluestore-keyreserve

os/bluestore: preallocate object[extent_shard] key to avoid reallocate

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>

messages/MOSDSubOp: Make encode_payload can be reentrant

Otherwise, AsyncConnection will use uninitialized feature to encode
message. The caller side will set a old version header but encode with newest
feature since passing feature is ok. It will let receiver side got a old
header version but newest payload decode result.

Signed-off-by: Haomai Wang <haomai@xsky.com>

src/test/test_denc.cc: Fix errors in buffer overflow

Signed-off-by: Willem Jan Withagen <wjw@digiware.nl>

os/bluestore: do not precalculate/cache shard keys

Note that we are generating it on demand now.  We can
probably do better, especially when in a loop.

Signed-off-by: Sage Weil <sage@redhat.com>

Merge pull request #12149 from adamemerson/wip-g_ceph_context-exterminate

global: start removing g_ceph_context

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge remote-tracking branch 'gh/kraken'

Merge pull request #12630 from liewegas/wip-workunit-retry

qa/tasks/workunit: clear clone dir before retrying checkout

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

Merge pull request #12614 from liewegas/wip-task-peer

qa/tasks/peer: update task based on current peering behavior

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

Merge pull request #12559 from Adirl/verify_transport_type

msg/async: assert if compiled code doesn't support the configured ms_…

Reviewed-by: Haomai Wang <haomai@xsky.com>

msg/async: assert if compiled code doesn't support the configured ms_async_transport_type

for example, currently there is a crash when using the async stack (access of null pointer)
in case the user configured rdma for a code that was not compiled with RDMA

issue: 913397

Change-Id: I4e3d88006312a713b8d3aaefbdcd0596039270e1
Signed-off-by: Avner BenHanoch <avnerb@mellanox.com>

Merge pull request #12643 from xiexingguo/xxg-wip-bluestore-1123

os/bluestore: remove intermediate key var to avoid string copy

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
Reviewed-by: whu liuchang <liuchang0812@gmail.com>

msg/async/rdma: fix RoCE v2 deafult value

For backward compatibility, if no GID given or given GID is malformed,
use GID index 0.

Change-Id: I8c8ecbcfc0261e58c5a226ef6bc91297ec756182
Signed-off-by: Oren Duer <oren@mellanox.com> Adir Lev <adirl@mellanox.com>
Signed-off-by: Adir Lev <adirl@mellanox.com>

os/bluestore: preallocate object[extent_shard] key to avoid reallocate

which is less efficient.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>

os/bluestore: remove intermediate key var to avoid string copy

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>

doc: release notes for v11.1.1

Signed-off-by: Abhishek Lekshmanan <alekshmanan@suse.com>

Merge pull request #12636 from badone/wip-doc-pg-states

doc: Remove "splitting" state

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>

Merge pull request #12543 from hjwsm1989/cleanup2

  osd/ECBackend: remove unused variable "ReadCB"

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>

doc: Remove "splitting" state

There is no splitting state so remove it from pg-states.rst

Signed-off-by: Brad Hubbard <bhubbard@redhat.com>

Merge pull request #12632 from liewegas/wip-centos-73

qa/suites: switch from centos 7.2 to 7.x

Reviewed-by: Yuri Weinstein <yweins@redhat.com>

Merge remote-tracking branch 'gh/kraken'

# Conflicts:
# src/os/bluestore/BlueStore.cc

os/bluestore: set cache meta ratio to .9

More metadata, less data.

Signed-off-by: Sage Weil <sage@redhat.com>

qa/distros: centos_7.yaml -> centos.yaml

Signed-off-by: Sage Weil <sage@redhat.com>

qa/suites: centos_7.2.yaml -> centos_7.yaml

Signed-off-by: Sage Weil <sage@redhat.com>

qa/distros: add centos 7.3

Signed-off-by: Sage Weil <sage@redhat.com>

qa/distros: add centos 7 yaml; use that instead

No need to be picky about the point release here.

Signed-off-by: Sage Weil <sage@redhat.com>

msg: Remove mention of g_ceph_context

Since it won't exist any more there's no point in saying it can't be
relied on, now is there?

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

objclass: Utterly destroy g_ceph_context

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

os: Complete removal of g_ceph_context and g_conf

And so it continues.

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

os: Continuing the removal of g_ceph_context and g_conf

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

os/filestore: Destroy g_ceph_context and g_conf

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

os/bluestore: Remove g_ceph_context and g_conf

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

os: Add CephContext to ObjectStore and ObjectMap bases

Preparatory to removing g_ceph_context and g_conf uses

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

osd: Remove g_conf and g_ceph_context

So begins the ruthless annihilation of g_ceph_context.

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

dout: Use dout_context

As a transition for g_ceph_context removal, make all dout macros depends
on a local macro.

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

common: Remove now unused clock_offset option

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

common: Unskew clock

In preparation to deglobalizing CephContext, remove the CephContext*
parameter to ceph_clock_now() and ceph::real_clock::now() that carries
a configurable offset.

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>

Merge pull request #12629 from liewegas/wip-bluestore-shard-extents

os/bluestore: remove 'extents' from shard_info

Reviewed-by: Somnath Roy <somnath.roy@sandisk.com>

Merge pull request #12616 from atheism/rgw-ldap-fix-simple-bind

rgw: ldap: simple_bind() should set ldap version option on tldap

qa/tasks/workunit: clear clone dir before retrying checkout

If we checkout ceph-ci.git, and don't find a branch,
we'll try again from ceph.git. But the checkout will
already exist and the clone will fail, so we'll still
fail to find the branch.

The same can happen if a previous workunit task already
checked out the repo.

Fix by removing the repo before checkout (the first and
second times).  Note that this may break if there are
multiple workunit tasks running in parallel on the same
role.  That is already racy, so if it's happening, we'll
want to switch to using a truly unique clonedir for each
instantiation.

Fixes: http://tracker.ceph.com/issues/18336
Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: remove 'extents' from shard_info

We don't use this field; no need to consume the space
on disk.

Signed-off-by: Sage Weil <sage@redhat.com>

Merge pull request #12394 from ifed01/wip-bluestore-fix-reshard

os/bluestore: refactor ExtentMap::update to avoid preceeding db updat…

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12464 from ifed01/wip-bluestore-turnoff-block-write

os/bluestore: add a debug option to bypass block device writes for bl…

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12526 from hjwsm1989/ec-cleanup

osd/ECBackend: only need check missing_loc when doing recovery

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12527 from hjwsm1989/cleanup

osd/PrimaryLogPG: unify the access to primary pg

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12564 from kylinstorage/wip-cleanup-pg-stat-t

osd/ECBackend: cleanup for unnecessary copy with pg_stat_t

Reviewed-by: Samuel Just <sjust@redhat.com>

Merge pull request #12576 from kylinstorage/wip-cleanup-bluestore-coll-ref

bluestore: avoid unnecessary copy with coll_t

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12579 from hjwsm1989/smallfix

osd/PG: check the connection first in fulfill_log

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12583 from liupan1111/wip-add-throttle-bluestore

os/bluestore: add perf variable for throttle info in bluestore

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Reviewed-by: Sage Weil <sage@redhat.com>

OSDMonitor: should propose osdmap update when OSD cluster_addr was changed

Signed-off-by: Mingxin Liu <mingxin@xsky.com>

os/bluestore/bluefs: make bluefs_preextend_wal_files default is false.

This make rocksdb log corruption. Before fix, disable this function.

Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>

Merge remote-tracking branch 'gh/kraken'

mon: MonMap: clear 'mon_info' during decode

It so happens that it's not safe to assume the monmap will be in an
empty state upon decoding.

Turns out the MonClient will reuse the MonMap instance when decoding
the just received map from the monitors. Should the monitors be on an
older version that do not support 'mon_info', this field will not be
decoded (after all, there's no field to decode from); but by this time,
the MonClient would already have a built monmap, which could have
populated 'mon_info' with temporary mon names from 'mon initial
members'.

Given the existing entries in 'mon_info', and the conflicting entries in
'mon_addr', we would end up asserting in 'sanitize_mons()'. This becomes
a non-issue if 'mon_info' is empty, as was unfortunately presumed.

Fixes: http://tracker.ceph.com/issues/18265
Signed-off-by: Joao Eduardo Luis <joao@suse.de>

Merge pull request #12604 from ommoreno/wip-16966

os/bluestore: allow multiple SPDK BlueStore OSD instances

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>

Merge pull request #12618 from Liuchang0812/fix-blue-store-unused-var

os/bluestore: remove a never read value