4386 Commits

Author SHA1 Message Date
antirez
c7ce8c0624 Hash: HSTRLEN crash fixed when getting len of int-encoded value 2015-02-27 15:37:04 +01:00
antirez
d4a13bd2f4 Hash: HSTRLEN (was HVSTRLEN) improved.
1. HVSTRLEN -> HSTRLEN. It's unlikely one needs the length of the key,
   not clear how the API would work (by value does not make sense) and
   there will be better names anyway.
2. Default is to return 0 when field is missing.
3. Default is to return 0 when key is missing.
4. The implementation was slower than needed, and produced unnecessary COW.

Related issue #2415.
2015-02-27 15:31:55 +01:00
antirez
b358007d7c Merge branch 'unstable' of github.com:/antirez/redis into unstable 2015-02-27 15:24:25 +01:00
Salvatore Sanfilippo
261b9d1b0f Merge pull request #2415 from landmime/unstable
added a new hvstrlen command
2015-02-27 15:24:04 +01:00
antirez
4aafbcb86b Hash: API to get value string len by field name. 2015-02-27 15:22:49 +01:00
antirez
b3d03d79d6 Utils: added function to get radix 10 string length of signed integer. 2015-02-27 15:22:10 +01:00
antirez
e726eca8e2 server.current_client fix and minor refactoring.
Thanks to @codeslinger (Toby DiPasquale) for identifying the issue.

Related to issue #2409.
2015-02-27 14:17:46 +01:00
antirez
7e11b8bb05 Improvements to PR #2425
1. Remove useless "cs" initialization.
2. Add a "select" var to capture a condition checked multiple times.
3. Avoid duplication of the same if (!copy) conditional.
4. Don't increment dirty if copy is given (no deletion is performed),
   otherwise we propagate MIGRATE when not needed.
2015-02-26 10:27:56 +01:00
Tommy Wang
e06d62c2aa Add last_dbid to migrateCachedSocket to avoid redundant SELECT
Avoid redundant SELECT calls when continuously migrating keys to
the same dbid within a target Redis instance.
2015-02-26 10:18:43 +01:00
antirez
84e1096f80 Cast sentlen to int before comparison wit bufpos.
This is safe since bufpos is small, inside the range of the local
client buffer.
2015-02-25 10:33:37 +01:00
Salvatore Sanfilippo
31e8dbbfdc Merge pull request #2050 from mattsta/bitops-no-overalloc
Bitops: Stop overallocating storage space on set
2015-02-25 10:18:07 +01:00
Salvatore Sanfilippo
b0eb128ac7 Merge pull request #2054 from mattsta/fix-set-sentinel-quorum
Sentinel: Add initial quorum bounds check
2015-02-25 10:09:40 +01:00
Matt Stancliff
92f3e19db7 Fix types broken by previous type cleanup
Revert some size_t back to off_t
Set reply_bytes needs to 64 bits everywhere
Revert bufpos to int since it's a max of 16k into buf[]
2015-02-24 17:39:59 +01:00
Salvatore Sanfilippo
de4eacd132 Merge pull request #2301 from mattsta/fix/lengths
Improve type correctness
2015-02-24 17:22:53 +01:00
Salvatore Sanfilippo
b2ffd67c91 Merge pull request #1966 from mattsta/fix-sentinel-info
Sentinel: Improve INFO command behavior
2015-02-24 17:20:09 +01:00
Salvatore Sanfilippo
813c1d300a Merge pull request #2273 from mattsta/improve/consistency/INFO/memory
Improve consistency of INFO MEMORY
2015-02-24 17:19:17 +01:00
Salvatore Sanfilippo
37238841e5 Merge pull request #2401 from mattsta/fix/quicklist/pop
Fix quicklist Pop()
2015-02-24 17:13:39 +01:00
antirez
fa8fcff3f0 Change RENAME behavior when src and dst keys are the same.
Fixes issue #2392.
2015-02-23 11:24:24 +01:00
Jason Roth
6cc1d51719 added a new hvstrlen command
the hvstrlen command returns the length of a hash field value
2015-02-21 15:01:18 +00:00
Matt Stancliff
58061fe49b Fix quicklist tests for Pop()
Now the tests actually compare return values instead of just
verifying _something_ got returned.
2015-02-17 11:33:08 -05:00
John Doe
62a6895054 Fix quicklist Pop() result
Closes #2398
2015-02-17 11:31:08 -05:00
antirez
7009630914 Version bumped to 3.1.999 (3.2 alpha versions). 2015-02-14 17:19:17 +01:00
antirez
40f50b38e0 redis-cli --latency-dist: one gray more, and --mono support. 2015-02-13 14:38:21 +01:00
antirez
b0d3557307 redis-cli --latency-dist, hopefully better palette.
Less grays: more readable palette since usually we have a non linear
distribution of percentages and very near gray tones are hard to take
apart. Final part of the palette is gradient from yellow to red. The red
part is hardly reached because of usual distribution of latencies, but
shows up mainly when latencies are very high because of the logarithmic
scale, this is coherent to what people expect: red = bad.
2015-02-13 12:52:46 +01:00
antirez
75a2dbffa8 Extend memory unit support in CONFIG SET.
Related to PR #2357.
2015-02-12 16:51:20 +01:00
Salvatore Sanfilippo
a4e2db7451 Merge pull request #2357 from lamby/config-set-maxmemory-units
Support "1G" etc. units in CONFIG SET maxmemory
2015-02-12 16:42:09 +01:00
antirez
07d4c76179 Better memtoll() error checking.
Related to PR #2357.
2015-02-12 16:40:46 +01:00
antirez
38000e913a SPOP with count: fix replication for code path #3. 2015-02-11 10:52:28 +01:00
antirez
87a6696d89 SPOP: reimplemented for speed and better distribution.
The old version of SPOP with "count" argument used an API call of dict.c
which was actually designed for a different goal, and was not capable of
good distribution. We follow a different three-cases approach optimized
for different ratiion between sets and requested number of elements.

The implementation is simpler and allowed the removal of a large amount
of code.
2015-02-11 10:52:28 +01:00
antirez
38afac7752 alsoPropagate: handle REDIS_CALL_PROPAGATE and AOF loading. 2015-02-11 10:52:28 +01:00
antirez
9bffc3f0c6 Change alsoPropagate() behavior to make it more usable.
Now the API automatically creates its argv copy and increment ref count
of passed objects.
2015-02-11 10:52:27 +01:00
antirez
54def17f37 SPOP with count: initial fixes to the implementation.
Severan problems are addressed but still a few missing.
Since replication of this command was more complex than others since it
needs to replicate multiple SREM commands, an old API able to do this
was reused (it was taken inside the implementation since it was pretty
obvious soon or later that would be useful). The API was improved a bit
so that now a command may opt-out for the standard command replication
when the server.dirty counter is incremented, in order to "manually"
replicate what it wants.
2015-02-11 10:52:27 +01:00
antirez
c752dc5adb Separate latency monitoring of eviction loop and eviction DELs. 2015-02-11 10:52:27 +01:00
antirez
e1da371706 dict.c: reset emptylen when bucket is not empty.
Fixed by @oranagra, thank you.
2015-02-11 10:52:27 +01:00
antirez
33799b02ab redis-cli --stat: show LOAD when loading. 2015-02-11 10:52:27 +01:00
antirez
b139745a35 Remove optional single-key path from evictionPoolPopulate(). 2015-02-11 10:52:27 +01:00
antirez
2bb647770d dict.c: add dictGetSomeKeys(), specialized for eviction. 2015-02-11 10:52:27 +01:00
antirez
6086a3656a dict.c: avoid code repetition in dictRehash().
Avoid code repetition introduced with PR #2367, also fixes the return
value to always return 0 if there is nothing more to rehash.
2015-02-11 10:52:27 +01:00
Sun He
9c047bbfec dict.c/dictRehash: check again to update 2015-02-11 10:52:26 +01:00
antirez
15940c1988 dict.c: don't try buckets that are empty for sure in dictGetRandomKey().
This is very similar to the optimization applied to dictGetRandomKeys,
but applied to the single key variant.

Related to issue #2306.
2015-02-11 10:52:26 +01:00
antirez
5d106de0c9 dict.c: dictGetRandomKeys() optimization for big->small table case.
Related to issue #2306.
2015-02-11 10:52:26 +01:00
antirez
71b990a7ab dict.c: dictGetRandomKeys() visit pattern optimization.
We use the invariant that the original table ht[0] is never populated up
to the index before the current rehashing index.

Related to issue #2306.
2015-02-11 10:52:26 +01:00
antirez
19d51b181b dict.c: put a bound to max work dictRehash() call can do.
Related to issue #2306.
2015-02-11 10:52:26 +01:00
antirez
1115984e9f dict.c: prevent useless resize to same size.
Related to issue #2306.
2015-02-11 10:52:26 +01:00
antirez
4c149c5ab9 Less blocking dictGetRandomKeys().
Related to issue #2306.
2015-02-11 10:52:26 +01:00
antirez
b3c69f291e HAVE_SYNC_FILE_RANGE should be protected by ifdef __linux__.
Related to issue #2372.
2015-02-10 09:29:14 +01:00
Salvatore Sanfilippo
0dcb6b144f Merge pull request #2372 from mariano-perez-rodriguez/patch-2
Fix HAVE_BACKTRACE on Linux
2015-02-10 09:27:46 +01:00
antirez
77494007b5 redis-cli --lru-test implemented (cache workload simulator). 2015-02-09 11:06:55 +01:00
antirez
4847cdfb4f redis-cli: interactive reconnection for latency modes.
--stat mode already used to reconnect automatically if the server is no
longer available. This is useful since this is an interactive mode used
for debugging, however the same applies to --latency and --latency-dist
modes, so now both use the reconnecting command execution as well.

The reconnection code was modified to use basic VT100 escape sequences
in order to play better with different kinds of output on the screen
when the reconnection happens, and to hide the reconnection attempt
output when finally the reconnection happens.
2015-02-08 23:28:27 +01:00
antirez
d1d81024d0 redis-cli --latecy-dist reverted to gray scale.
So far not able to find a color palette within the 256 colors which is
not confusing. However I believe it is a possible task, so will try
better later.
2015-02-08 17:39:42 +01:00