antirez
5f0fef5eb9
Regression test for issue #2813 .
2015-10-15 11:23:15 +02:00
antirez
ed6228851c
PR 2813 fix ported to unstable.
2015-10-15 10:20:09 +02:00
antirez
35a0c772b5
DEBUG RESTART/CRASH-AND-RECOVER [delay] implemented.
2015-10-13 11:12:25 +02:00
antirez
73427462ed
Server: restartServer() API.
...
This new function is able to restart the server "in place". The current
Redis process executes the same executable it was executed with, using
the same arguments and configuration file.
2015-10-13 11:02:35 +02:00
antirez
c372a59520
Cluster: redis-trib fix, coverage for migrating=1 case.
...
Kinda related to #2770 .
2015-10-09 16:15:58 +02:00
antirez
6ddcba6ec9
Test: basic lazyfree unit test.
2015-10-09 09:47:17 +02:00
antirez
363c0f67b9
Test: fix attach_to_replication_stream to handle newlines.
2015-10-07 22:32:24 +02:00
antirez
f29e384853
Fix extractLongLatOrReply() sanity check conditionals.
...
the check for lat/long valid ranges were performed inside the for loop,
two times instead of one, and the first time when the second element of
the array, xy[1], was yet not populated. This resulted into issue #2799 .
Close issue #2799 .
2015-10-07 22:27:18 +02:00
antirez
3f38b51ad7
Jemalloc configure script fixed to work nested.
...
Now way to make unmodified Jemalloc configure to work when the jemalloc
source tree is inside a subdirectory of a different git repository.
Problem signaled here:
http://www.canonware.com/pipermail/jemalloc-discuss/2015-October/001174.html
2015-10-07 09:17:06 +02:00
antirez
a9951b1b6a
Jemalloc updated to 4.0.3.
2015-10-06 16:55:37 +02:00
antirez
e3ded0273c
Added a README into deps on dependencies and how to upgrade.
2015-10-06 10:56:16 +02:00
antirez
ceaf58dfd5
Regression test for GEORADIUS COUNT arity check.
2015-10-06 09:27:29 +02:00
antirez
319d180136
Fix GEORADIUS COUNT option arity checks.
2015-10-06 09:25:28 +02:00
antirez
19f6ab50c3
Lazyfree options documented in the example redis.conf.
2015-10-05 12:24:16 +02:00
antirez
880c606a7c
Lazyfree options implemented in the configuration.
2015-10-05 12:11:27 +02:00
antirez
252cfa0a39
Lazyfree: cond vars to enabled/disable it based on DEL context.
2015-10-02 15:27:57 +02:00
antirez
5359696796
Fixed a bug in the emptyDb() new implementation.
2015-10-01 13:02:26 +02:00
antirez
ecdbc333a3
FLUSHDB and FLUSHALL ASYNC option implemented.
2015-10-01 13:02:26 +02:00
antirez
1f26a9468f
Lazyfree: pending objects count in INFO output.
2015-10-01 13:02:26 +02:00
antirez
c69c6c80fb
Lazyfree: ability to free whole DBs in background.
2015-10-01 13:02:26 +02:00
antirez
b08c36c5f2
Lazyfree: keep count of objects to free.
2015-10-01 13:02:25 +02:00
antirez
c7b46a4719
zmalloc.c converted to use atomicvar.h.
2015-10-01 13:02:25 +02:00
antirez
7e5d69019e
Atomic vars implemented in a more general way.
...
We have them into zmalloc.c, but this is going to replace that
implementation, so that it's possible to use the same idea everywhere
inside the code base.
2015-10-01 13:02:25 +02:00
antirez
7af4eeb745
Lazyfree: incremental removed, only threaded survived.
2015-10-01 13:02:25 +02:00
antirez
9253d85073
Threaded lazyfree WIP #1 .
2015-10-01 13:02:25 +02:00
antirez
4d50d691e3
bio.c: new API bioWaitStepOfType().
2015-10-01 13:02:25 +02:00
antirez
5b850d7a09
Test: stack_logging var should be initialized to 0.
2015-10-01 13:02:25 +02:00
antirez
1dab60df81
Hash new implementation memleaks fixed.
2015-10-01 13:02:25 +02:00
antirez
97ba4e3886
Lazyfree: Hash converted to use plain SDS WIP 5.
2015-10-01 13:02:25 +02:00
antirez
36be34bb87
Test: support for stack logging for OSX malloc/leaks.
2015-10-01 13:02:25 +02:00
antirez
974514b936
Lazyfree: Hash converted to use plain SDS WIP 4.
2015-10-01 13:02:25 +02:00
antirez
4a18352877
Lazyfree: Hash converted to use plain SDS WIP 3.
2015-10-01 13:02:24 +02:00
antirez
777396aeba
Lazyfree: Hash converted to use plain SDS WIP 2.
2015-10-01 13:02:24 +02:00
antirez
1c247556c6
Lazyfree: Hash converted to use plain SDS WIP 1.
2015-10-01 13:02:24 +02:00
antirez
afc4b9241c
DEBUG DIGEST Set type memory leak fixed.
2015-10-01 13:02:24 +02:00
antirez
34e489cb8c
SORT memory leak fixed.
2015-10-01 13:02:24 +02:00
antirez
a7c5be18a8
Lazyfree: Sorted sets convereted to plain SDS. (several commits squashed)
2015-10-01 13:02:24 +02:00
antirez
86d48efbfd
Lazyfree: Convert Sets to use plains SDS (several commits squashed).
2015-10-01 13:02:24 +02:00
antirez
4ff3c17a20
Lazyfree: client output buffers no longer use Redis Objects.
2015-10-01 13:02:24 +02:00
antirez
0c05436cef
Lazyfree: a first implementation of non blocking DEL.
2015-10-01 13:00:19 +02:00
antirez
712ea7296d
Call writeToClient() directly instead of the write handler.
2015-09-30 17:41:52 +02:00
antirez
01c08b5089
Fix processEventsWhileBlocked() to handle PENDING_WRITE clients.
...
After the introduction of the list with clients with pending writes, to
process clients incrementally outside of the event loop we also need to
process the pending writes list.
2015-09-30 17:23:44 +02:00
antirez
1e7153831d
Refactoring: unlinkClient() added to lower freeClient() complexity.
2015-09-30 17:10:03 +02:00
antirez
fdb3be939e
Refactoring: new function to test if client has pending output.
2015-09-30 16:41:48 +02:00
antirez
825f65d2bd
Reverse list of clients with pending writes.
...
May potentially improve locality... not exactly clear if this makes a
difference or not. But for sure is harmless.
2015-09-30 16:29:42 +02:00
antirez
063ecbd5e5
writeToClient(): don't remove write handler if not needed.
2015-09-30 16:29:42 +02:00
antirez
b741a90ce9
handleClientsWithPendingWrites(): detect dead clients.
2015-09-30 16:29:42 +02:00
antirez
481a0db315
Move handleClientsWithPendingWrites() in networking.c.
2015-09-30 16:29:42 +02:00
antirez
1c7d87df0c
Avoid installing the client write handler when possible.
2015-09-30 16:29:41 +02:00
antirez
d1b6a17d1e
redis-cli pipe mode: don't stay in the write loop forever.
...
The code was broken and resulted in redis-cli --pipe to, most of the
times, writing everything received in the standard input to the Redis
connection socket without ever reading back the replies, until all the
content to write was written.
This means that Redis had to accumulate all the output in the output
buffers of the client, consuming a lot of memory.
Fixed thanks to the original report of anomalies in the behavior
provided by Twitter user @fsaintjacques.
2015-09-30 16:24:21 +02:00
