gvsafronov/futriix
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
27,547 Commits 1 Branch 0 Tags
Commit Graph

21057 Commits

Author SHA1 Message Date
John Sully
89a44bd7e5 Remove an unnecessary lock that eats performance 
Former-commit-id: 8eb4e355b934e394608b082c7d1973152238fa1f
 2019-06-17 23:00:12 -04:00
John Sully
049ca9bf58 Even the next up thread should sleep in the futex 
Former-commit-id: 8b76b01bfe710603bcdc101da6eb27afcee7e1b1
 2019-06-17 23:00:12 -04:00
John Sully
0cd9bc48fc Even the next up thread should sleep in the futex 
Former-commit-id: 8b76b01bfe710603bcdc101da6eb27afcee7e1b1
 2019-06-17 23:00:12 -04:00
John Sully
dd7cbbe730 Fallback to a futex if we spin for a long time 
Former-commit-id: ec57b4b0248bba671e388a2257b1bd65ed8d0f44
 2019-06-17 23:00:12 -04:00
John Sully
5573f021c5 Fallback to a futex if we spin for a long time 
Former-commit-id: ec57b4b0248bba671e388a2257b1bd65ed8d0f44
 2019-06-17 23:00:12 -04:00
Madelyn Olson
bf7eb02d3e Refactored yesno configs so there was less duplication 2019-06-13 12:18:25 -07:00
Madelyn Olson
dd51fc5a4a Refactored yesno configs so there was less duplication 2019-06-13 12:18:25 -07:00
Angus Pearson
49edda41b9 Spelling cannonical -> canonical 2019-06-13 17:49:33 +01:00
Angus Pearson
38cd5fd9f6 Spelling cannonical -> canonical 2019-06-13 17:49:33 +01:00
swilly22
7077d14afe Extend REDISMODULE_CTX_FLAGS to indicate if redis is currently loading from either RDB or AOF 2019-06-12 15:37:19 +03:00
swilly22
f54d95e803 Extend REDISMODULE_CTX_FLAGS to indicate if redis is currently loading from either RDB or AOF 2019-06-12 15:37:19 +03:00
John Sully
7fc6f7f607 Enable load balancing of UNIX sockets across threads 
Former-commit-id: 46d6caa47c57fa13055615dfc2376b0962468b81
 2019-06-11 01:34:36 -04:00
John Sully
c64b6127ea Enable load balancing of UNIX sockets across threads 
Former-commit-id: 46d6caa47c57fa13055615dfc2376b0962468b81
 2019-06-11 01:34:36 -04:00
Angus Pearson
5ca48db2e2 Add char* typeNameCanonicalize(robj*) to remove duplicate code between SCAN and TYPE commands, 
and to keep OBJ_* enum to string canonicalization in one place.
 2019-06-10 17:41:44 +01:00
Angus Pearson
e2adea2188 Add char* typeNameCanonicalize(robj*) to remove duplicate code between SCAN and TYPE commands, 
and to keep OBJ_* enum to string canonicalization in one place.
 2019-06-10 17:41:44 +01:00
Salvatore Sanfilippo
2af8ebe147 Merge pull request #6149 from madolson/acl-spelling 
Fixed some spelling issues in ACL codepath including a user facing error
 2019-06-07 22:54:52 +02:00
Salvatore Sanfilippo
0cabe0cfa7
Merge pull request #6149 from madolson/acl-spelling 
Fixed some spelling issues in ACL codepath including a user facing error
 2019-06-07 22:54:52 +02:00
Madelyn Olson
67a4bcac1b Fixed some spelling issues in ACL codepath including user facing error 2019-06-07 13:25:22 -07:00
Madelyn Olson
dd5f437871 Fixed some spelling issues in ACL codepath including user facing error 2019-06-07 13:25:22 -07:00
artix
ec45f5b395 Redis Benchmark: prevent CONFIG failure from exiting program 2019-06-05 16:34:55 +02:00
artix
d3ed53f11b Redis Benchmark: prevent CONFIG failure from exiting program 2019-06-05 16:34:55 +02:00
John Sully
193522b54e Fix crash adding duplicate replica. Just send an error instead 
Former-commit-id: 97b65ce19c09188acd0d662aad8d774a55531437
 2019-06-02 15:50:50 -04:00
John Sully
88b5a4e25f Fix crash adding duplicate replica. Just send an error instead 
Former-commit-id: 97b65ce19c09188acd0d662aad8d774a55531437
 2019-06-02 15:50:50 -04:00
Oran Agra
f7833d560d make redis purge jemalloc after flush, and enable background purging thread 
jemalloc 5 doesn't immediately release memory back to the OS, instead there's a decaying
mechanism, which doesn't work when there's no traffic (no allocations).
this is most evident if there's no traffic after flushdb, the RSS will remain high.

1) enable jemalloc background purging
2) explicitly purge in flushdb
 2019-06-02 15:33:14 +03:00
Oran Agra
09f99c2a92 make redis purge jemalloc after flush, and enable background purging thread 
jemalloc 5 doesn't immediately release memory back to the OS, instead there's a decaying
mechanism, which doesn't work when there's no traffic (no allocations).
this is most evident if there's no traffic after flushdb, the RSS will remain high.

1) enable jemalloc background purging
2) explicitly purge in flushdb
 2019-06-02 15:33:14 +03:00
Yuan Zhou
4620bfb47a aof: fix assignment for aof_fsync_offset 
Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
 2019-05-29 14:48:46 +08:00
Yuan Zhou
412eb67d21 aof: fix assignment for aof_fsync_offset 
Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
 2019-05-29 14:48:46 +08:00
Angus Pearson
4bbaf621a1 Implement SCAN cursor [TYPE type] modifier suggested in issue #6107. 
Add tests to check basic functionality of this optional keyword, and also tested with
a module (redisgraph). Checked quickly with valgrind, no issues.

Copies name the type name canonicalisation code from `typeCommand`, perhaps this would
be better factored out to prevent the two diverging and both needing to be edited to
add new `OBJ_*` types, but this is a little fiddly with C strings.

The [redis-doc](https://github.com/antirez/redis-doc/blob/master/commands.json) repo
will need to be updated with this new arg if accepted.

A quirk to be aware of here is that the GEO commands are backed by zsets not their own
type, so they're not distinguishable from other zsets.

Additionally, for sparse types this has the same behaviour as `MATCH` in that it may
return many empty results before giving something, even for large `COUNT`s.
 2019-05-22 16:39:04 +01:00
Angus Pearson
bf963253ec Implement SCAN cursor [TYPE type] modifier suggested in issue #6107. 
Add tests to check basic functionality of this optional keyword, and also tested with
a module (redisgraph). Checked quickly with valgrind, no issues.

Copies name the type name canonicalisation code from `typeCommand`, perhaps this would
be better factored out to prevent the two diverging and both needing to be edited to
add new `OBJ_*` types, but this is a little fiddly with C strings.

The [redis-doc](https://github.com/antirez/redis-doc/blob/master/commands.json) repo
will need to be updated with this new arg if accepted.

A quirk to be aware of here is that the GEO commands are backed by zsets not their own
type, so they're not distinguishable from other zsets.

Additionally, for sparse types this has the same behaviour as `MATCH` in that it may
return many empty results before giving something, even for large `COUNT`s.
 2019-05-22 16:39:04 +01:00
John Sully
2679450489 Implement MOTD feature in keydb-cli 
Former-commit-id: a6c56416fc1bf09f1ecbae45195290209aa93c89
 2019-05-21 15:12:12 -04:00
John Sully
20d8346846 Implement MOTD feature in keydb-cli 
Former-commit-id: a6c56416fc1bf09f1ecbae45195290209aa93c89
 2019-05-21 15:12:12 -04:00
zhaozhao.zz
4517ac981f Threaded IO: use main thread to handle read work 2019-05-21 11:42:10 +08:00
zhaozhao.zz
1398fac3f1 Threaded IO: use main thread to handle read work 2019-05-21 11:42:10 +08:00
John Sully
dd5152f836 Fix regressions from RESP3 changes in commit c8304b099d82dba4da35f4cfaa99595bcdb46d1a 
Former-commit-id: c5cbb77fc7e3b2b5124b9432e144d8c74190c755
 2019-05-20 23:39:44 -04:00
John Sully
a7afb4c3f7 Fix regressions from RESP3 changes in commit 317f8b9d383f1b7f171aef7ea29f9e05abf0ba83 
Former-commit-id: c5cbb77fc7e3b2b5124b9432e144d8c74190c755
 2019-05-20 23:39:44 -04:00
zhaozhao.zz
27dc71f8dc Threaded IO: use main thread to handle write work 2019-05-21 11:37:13 +08:00
zhaozhao.zz
8b33975944 Threaded IO: use main thread to handle write work 2019-05-21 11:37:13 +08:00
John Sully
b7fff89e3e The unstable branch doesn't have a version # 
Former-commit-id: d7f21f02b30c3f9092dbf9aec75481baf45c3311
 2019-05-20 16:12:26 -04:00
John Sully
7ec3b111af The unstable branch doesn't have a version # 
Former-commit-id: d7f21f02b30c3f9092dbf9aec75481baf45c3311
 2019-05-20 16:12:26 -04:00
John Sully
7d40b9b7d5 Fix double unlock 
Former-commit-id: 87dc8ce9b3e376e4a6264aee1d2787b9f8d6437f
 2019-05-20 15:52:04 -04:00
John Sully
be60fc72a5 Fix double unlock 
Former-commit-id: 87dc8ce9b3e376e4a6264aee1d2787b9f8d6437f
 2019-05-20 15:52:04 -04:00
John Sully
fa828d8c92 Merge branch 'unstable' of https://github.com/JohnSully/redis2 into unstable 
Former-commit-id: 2e22aa1a9cd9247a51c03afc161cb84a98139438
 2019-05-20 10:57:59 -04:00
John Sully
680b49045f Merge branch 'unstable' of https://github.com/JohnSully/redis2 into unstable 
Former-commit-id: 2e22aa1a9cd9247a51c03afc161cb84a98139438
 2019-05-20 10:57:59 -04:00
John Sully
a3706352d0 Protocol regression from RESP3 work 
Former-commit-id: 8c56dbf971aced9e18d170dfffb0c074b545f718
 2019-05-20 10:57:37 -04:00
John Sully
e9b8d4e73a Protocol regression from RESP3 work 
Former-commit-id: 8c56dbf971aced9e18d170dfffb0c074b545f718
 2019-05-20 10:57:37 -04:00
Qix
36ad4c5907 cast void data pointer to const char * to fix gcc 
Former-commit-id: a78169e9e043aea83c3f8b8fbb14e605575f8575
 2019-05-19 16:24:39 -04:00
Qix
9b2adb19dd cast void data pointer to const char * to fix gcc 
Former-commit-id: a78169e9e043aea83c3f8b8fbb14e605575f8575
 2019-05-19 16:24:39 -04:00
John Sully
2783649378 Fix build breaks on some older compilers 
Former-commit-id: 39a60c8120ac4a7c2aad83df00a4fbea07a801cc
 2019-05-18 20:58:48 -04:00
John Sully
e2b3627360 Fix build breaks on some older compilers 
Former-commit-id: 39a60c8120ac4a7c2aad83df00a4fbea07a801cc
 2019-05-18 20:58:48 -04:00
John Sully
7074a7615e Fix build break on some compilers 
Former-commit-id: c57d8760bd9f824555366d1685b12e959ef36f4c
 2019-05-18 20:41:05 -04:00